The URL can be used to link to this page

Home My WebLink AboutAPA4003DRAFT/PAGE 2,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 THE RELATIVE ABUNDANCE,DISTRIBUTION,AND INSTREAM FLOW RELATIONSHIPS OF JUVENILE SALMON - .... DRAFT/PAGE 1 5/1/85,4/25/85 NUM4/Title Page ALASKA DEPARTMENT OF FISH AND GAME SUSITNA AQUATIC STUDIES PROGRAM, UNIV;::.r ,-,,'ALASKA ARCTIC ENVlRONMENTi\L INFORMATiON "''''''l'''",r","'''<'~-''RANDU:.:\\I t:;,' 707 A STREET ANCHORAGE,AK 99501 REPORT NO.7 RESIDENT AND JUVENILE ANADROMOUS FISH INVESTIGATIONS (MAY -OCTOBER 1984) T\< 1L{1-S .SB A~B 'no,4,003, ARLIS Alaska Resources Library &Information Services Ancld~l£'~,~.iSka Project Leader:Dana C.Schmidt Acting Project Leader:Stephen S.Hale Editors:Drew L.Crawford Stephen S.Hale,and Dana C.Schmidt !r~. DRAFT/PAGE 1 5/1/85,4/25/85 NUM4/Title Page ALASKA DEPARTMENT OF FISH AND GAME SUSITNA AQUATIC STUDIES PROGRAM . REPORT NO.7 RESIDENT AND JUVENILE ANADROMOUS FISH INVESTIGATIONS (MAY -OCTOBER 1984) T\< 1'11-1? .SB Ab5 no.'-\.003" ~,ARLIS Alaska Resources Library &Information.Services Ancl.<J~3£::,~:iSka Project Leader:Dana C.Schmidt Acting Project Leader:Stephen S.Hale Editors:Drew L.Crawford Stephen S.Hale,and Dana C.Schmidt ,.,Prepared for: 'Alaska Power Authority 334 W.Fifth Avenue,Second Floor Anchorage,Alaska 99501 DATE DUE Demeo, ;:;; - - ~- - ..... - - - -- Inc.38-293 -- ARLIS Alaska Resources Library &Information SerVices Anct .C\;:,.~~ska - "-.$$ 55 .- -. - DRAFT/PAGE 1,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 PART 2 The Relative Abundance,Distribution,and Instream Flow Relationships of Juvenile Salmon in the Lower Susitna River. PREFACE CriAFT DRAFT/PAGE 1 5/l/85~4/25/85 NUM4/Pr,eface - ,~ LO <.0r-- <.0 ~ ooo LO LO ,.....i ('t) ('t) This report is one of a series of reports prepared for the Alaska Power Authority (APA)by the Al aska Department of Fi sh and Game (ADF&G)to provide information to be used in evaluating the feasibility of the proposed Sus.itna Hydroelectric Project.The ADF&G Susitna Hydro Aquatic Studies program was initiated in November 1980. The report covers studies of juvenile salmon and resident fish species of the Susitna River conducted from May through October 1984.In addition,some information on overwintering of resident fish radio- tagged in 1983 is included.The majority of the effort during the 1984 open-water season was on the lower river (from the mouth to the Chulitna River confluence).No studies were conducted this year in the area above Devil Canyon.This volume consists of four parts. Part 1 (RSA Tasks 16A and 168)covers the migration and growth of juvenile salmon.Coded wire tagging of chum and sockeye fry in the middle river (Chulitna River confluence to Devil Canyon)and collecting of all species of outmigrating fry at Talkeetna Station were similar to 1983 studies.In addition,a mark-and-recapture cold branding study was conducted in tributaries,sloughs~and side channels of the middle river to obta in an index of ch i nook and coho juveni 1e sa Tmon abundance and residence time in these rearing areas.This study c9mplements the coded wire tagging studies of chum and sockeye fry in the middle river.Also, outmigrant traps were operated at Flathorn Station (River Nile 22.4) CriAFT __u -. DRAFT/PAGE 2 5/1/85,4/25/85 NUM4/Pr-eface ~.J near the mouth of the river to obtain a timing index of outmigration from the lower river. Studies of the distribution and relative abundance of juvenile salmon and modelling of rearing habitat in the lower river are discussed in Part 2 (RSA Tasks 14 and 36).These studies were similar to those conducted in the middle river in 1983.Habitat suitabi1 ity criteria developed for the middle river were used for the lower river unless evidence of different conditions in the lower -river necessitated modifications.Habitat modelling results from 14 RJHAB model sites and 6 IFIM model sites are presented.The RJHAB and IFIM model s were compared by using both at two sites. Part 3 (RSA Task 14)contains the results of resident fish studies in both the middle and lower river.Monitoring of fish movement through use of radio tags was continued and index sites in the middle river were sampled as part of the long term monitoring effort.Population esti- mates for some species were made from multiple year mark-recapture data. Part 4 (RSA Task 16A)is a statistical time series analysis of 1983 and 1984 discharge,turbidity,and juvenile salmon outmigration data in the middle river.This part represents the beginning of an effort to analyze,integrate,and summarize the five years of data collected by the Susitna Aquatic Studies Program.The final report on this five year summary will be completed a year from now. - -! - CtiAFT TITLES IN THIS SERIES DRAFT/PAGE 4 5/1/85,4/25/85 NUM4/Preface -~ Report Publication Number Title Date ~:~~-1 Adult Anadromous Fish Investigations:April 1984 May -October 1983 ~2 Resident and Juvenile Anadromous Fish July 1984 Investigations:May -October 1983 3 Aquatic Habitat and Instream Flow September 1984 Investigations:May -October 1983 i~4 Access and Transmission Corridor Aquatic September 1984 Investigations:May -October 1983 5 Winter Aquatic Investigations:March 1985 September 1983 to May 1984 .6 Adult Anadromous Fish Investigations:1985 ,~May -October 1984 7 Resident and Juvenile Anadromous Fish 1985 .-Investigations:May -October 1984 I~ ,...., ,.... , Questions concerning this report should be directed to: Alaska Power Authority 334 West 5th Avenue Anchorage,Alaska 99501 Telephone:(907)276-0001 -----------------------_. C~AFT CONTENTS OF REPORT NO.7 DRAFT/PAGE 5 5/1/85,4/25/85 NUM4/Preface Part 1. Part 2. Pa rt 3. Part 4. The Migration and Growth of Juvenile Salmon in the Susitna River. The Relative Abundance,Distribution,and Instream Flow Relationships of Juvenile ·Salmon in the Lower Susitna River. Resident Fish Distribution and Population Dynamics in the Susitna River below Devil Canyon. Time Series Analysis of Juvenile Salmon Outmigration, ,Discharge,and Turbidity in the Susitna River,Alaska. - - - - DRAFT/PAGE 2,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 THE RELATIVE ABUNDANCE,DISTRIBUTION,AND INSTREAM FLOW RELATIONSHIPS OF JUVENILE SALMON IN THE LOWER SUSITNA RIVER Report No.7,Part 2 by Paul M.Suchanek,Karl J.Kuntz,and John P.McDonell Alaska Department of Fish and Game Susitna Aquatic Studies 620 East 10th Avenue,Suite 302 Anchorage,Alaska 99501 ABSTRACT Juvenile salmon abundance and distribution were studied in the lower Susitna River and juvenile salmon habitat was modelled at 20 sites within the reach.Chinook,chum,and sockeye .salmon juveniles made use of side channels,however,high turbidity limited use of side channels located in the Chulitna River plume.Coho salmon juveniles were found primarily in tributary mouths;sockeye,chinook,and chum salmon also used these areas.Sloughs were limited in occurrence and were not used heavily by any of the salmon species. Both tributary mouths and side channel/slough sites were modelled using one of two habitat models.At tributary mouths,increases in weighted usable area with increases in mainstem discharge were due to the forma- tion of backwaters which led to lower velocities and in.crease~incQy~r and area.At side channels,chinook weighted usable--area'TliCre-ase'd after overtopping due---to_jncreases in cover suitability (tLJr:bjdJ~Y)a"."~,,.'.'/X~lo~.it~an.d .~rea....J The 'weighted u.sable area respo.nse to(~c~ange.~)in JI,vCK'~ ( Ymal11stem di scnarge for sockeye and chum salmon juveni 1es at "'s'ra~r"chan--~~ls w~s ~l2Lusually positive~--Ffabitat indices at side channels for ./ c:nfi1OOl<;chum,aff~~j-t1Veniles at mainstem discharges and side I channel flows above the overtopping discharge declined as velocities I became unsuitably high.Weighted usable area for these species some-\ times did not decl ine at hi gh di scharges,however,because the total ) area of the site was also increasing._4~~ ,",<,..,-;''''~ ----,-,---------- DRAFT/PAGE 3,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 TABLE OF CONTENTS ABSTRACT LIST OF FIGURES LIST OF TABLES LIST OF APPEND'IX FIGURES . LIST OF APPENDIX TABLES 1.0 INTRODUCTION 2.0 METHODS 2.1 Field Sampling Design 2.1.1 Study locations and selection criteria 2.1.2 Field data collection 2.1.2.1 Resident Juvenile Habitat (RJHAB)model sites 2.1.2.2 Instream Flow Incremental Methodology (IFIM)sites 2.1.2.3 Opportunistic sites 2.1.3 Schedule of activities and frequency of sampling 2.2 Data Analysis 2.2.1 Physical data 2.2.2 Abundance and distribution 2.2.3 Habitat modelling of rearing salmon 2.2.3.1 Suitability criteria development 2.2.3.2 Instream Flow Incremental Methodology (IFIM)models 2.2.3.3 Resident Juvenile Habitat (RJHAB)models 2.2.3.4 Model verification 3.0 RESULTS 3.1 Seasonal,Spatial,and Discharge Related Variations in Habitat 3.1.1 Macrohabitat type classifications of study sites 3.1.2 Chulitna and Talkeetna River plume influences on turbidity of side channels 3.1.3 Physical responses of sampling sites to mainstem discharge variations 3.1.3.1 Area 3.1.3.2 Cover J I - DRAFT/PAGE 4,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 jlf!!$i1il!i11 -- TABLE OF CONTENTS (Continued) 3.2 Distribution and Abundance of Juvenile Salmon 3.2.1 Chinook salmon 3.2.2 Coho salmon 3.2.3 Chumsalmon 3.2.4 Sockeye salmon 3.3 Habitat Modelling of Rearing Juvenile Salmon 3.3.1 Chinook salmon 3.3.2 Coho salmon 3.3.3 Chum salmon 3.3.4 Sockeye salmon 4.0 DISCUSSION 4.1 Chinook Salmon .4.2 Coho Salmon 4.3 Chum Salmon 4.4 Sockeye Salmon 5.0 CONTRIBUTORS 6.0 ACKNOWLEDGEMENTS 7.0 LITERATURE CITED .8.0 APPENDICES Appendix A Lower Susitna River Juvenile Salmon Rearing Suitability Criteria Appendix B Modelled Site Turbidities, Juvenile Salmon Catches,Areas, Weighted Usable Areas,and Habitat Indices Appendix C Comparison of the IFIM and RJHAB Modell ing Techniques at Two Sel ected Sites Appendix D Hydraulic Models for Use in Assessing the Rearing Habitat of Juvenile Salmon in Six Side Channels of the Lower Susitna River I i 2 DRAFT/PAGE 5,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 LIST OF FIGURES Figure 1 Location of study sites on the lower Susitna River at which juvenile salmon habitat was modelled,June through October 1984. Arrangement of transects and samp1 i ng cell s within a grid at a hypothetical RJHAB mode1- 1i ng site. 3 Turbidities at modelled side channel sand tributary mouths on the lower Susitna River, June through October 1984. 4 Comparison of turbidities in the lower Susitna River below the Chulitna and Tal- keetna River confluences on July 19 and August 16,1984. 5 Area within modelled tributary mouths as a function of mainstem discharge at the USGS Sunshine gaging station,1984. 6 Area within modelled sloughs and side chan- nels as a function of mainstem discharge at the USGS Sunshine gaging station,1984. 7 Instream cover response at Beaver Dam Slough, Rolly Creek and Caswell Creek mouths as a function of mainstem discharge at the USGS Sunshine gaging station,1984. 8 Seasonal distribution and relative abundance of juvenile chinook salmon on the lower Susitna River,June through mid-October 1984. 9 Juvenile chinook salmon mean catch per cell at side channels and tributary mouths on the lower Susitna River by sampling period,June through mid-October 1984. 10 Juvenile chinook salmon mean catch per cell at modelled side channels on the lower Susitna River by turbidity increment,June through mid-October 1984. 11 Seasonal distribution and relative abundance of juvenile coho salmon on the lower Susitna River,June through mid-October 1984. 1'1 15 DRAFT/PAGE 6,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 LIST OF FIGURES (Continued) 12 Juvenile"coho salmon mean catch per cell at four tributary mouths on the lower Susitna River by sampling period,June through mid-October 1984. 13 Seasonal distribution and relative abundance of juvenile c~um salmon on the lower Susitna River,June through mid-October 1984. 14 Juvenile chum salmon mean catch per &ell at modelled side channels and tributary mouths on the lower Susitna River by sampling period,June through mid-October 1984. Juvenile chum salmon mean catch per cell at modelled side channels on the 10wer"Susitna River by turbidity increment,June through mid-July 1984. 16 Seasonal distribution and relative abundance of juvenile sockeye salmon on the lower Susitna River,June through mid-October 1984. 17 Juvenile sockeye salmon mean catch per cell at side channel and tributary mouths on the lower Susitna River by sampling period,June through mid-October 1984. -18 Juvenile sockeye salmon mean catch per cell at modelled side channels on the lower SusitnaRiver by turbidity increment (with and wi thout Beaver Dam Si de Channel),June through mid-October 1984. 20 19 Juvenile sockeye salmon mean catch per cell at Beaver Dam Slough,Beaver Dam Side Chan- nel,and Rolly Creek Mouth by sampling period,June through mid-October 1984. "Weighted usable area for juvenile chinook salmon at the Rolly Creek Mouth,Kroto Slough Head and Sucker Side Channel study sites as a function of main stem discharge,1984. 21 Weighted usable area and habitat indices for juvenile chinook salmon at tributary mouth study sites as a function of mainstem dis- charge,1984. v -"._m ~_~__m;m'Wii'~--_,--"------------------,- 22 DRAFT/PAGE 7,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part2,4/14/85 LIST OF FIGURES (Continued) Weighted usable area and habitat indices for juvenile chinook salman at side channell slough study sites as a function of mainstem discharge,1984. 23 Adjusted weighted usable area and habitat indices for juvenile chinook salmon at side channe 1/slough study sites as a funct i on of mainstem discharge,1984. 24 Juvenile chinook salmon mean catch per cell versus seasonal mean habitat indices at side channel and tributary mouth modelling sites on the lower Susitna River,1984. 25 Weighted usable area for juvenile coho salmon at the Caswell Creek,Rolly Creek and Beaver Dam Slough tributary study sites as a func- tion of mainstem discharge,1984. 26 Weighted usable area and habitat indices for juvenile coho salmon at tributary mouth study sites exc 1udi ng Bi rch C.reek as a funct i on of mainstem discharge,1984. 27 Juvenile coho salmon mean catch per cell versus seasonal mean habitat indices at tributary mouth modelling sites on the lower Susitna River,1984. 28 Weighted usable area for juvenile chum salmon at the Rustic Wilderness and Last Chance Side Channel study sites as a function of mainstem discharge,1984. 29 Weighted usable area for juvenile chum salmon at the Trapper Creek and Sunset Side Channel study sites as a function of mainstem dis- charge,1984. 30 Weighted usable area and habitat indices for juvenile chum salmon at side channel/slough study sites as a function of mainstem dis- charge,1984. 31 Adjusted weighted usable area and habitat indices for juvenile chum salmon at side channel/slough study sites as a function of mainstem discharge,1984. ",I I - DRAFT/PAGE 8,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 LIST OF FIGURES (Continued) -- 32 Juvenile chum salmon mean catch per cell versus seasonal mean habitat indices at side channel and slough modell i ng sites on the lower Susitna River,1984. 36 35- - 33 Weighted usable area for juvenile sockeye salmon at the Rolly Creek Mouth and Sucker Side Channel study sites as a function of mainstem discharge,1984. 34 Weighted usable areas for juvenile sockeye salmon at the Beaver Dam and Sunrise Side Channel study sites as a function of mainstem discharge,1984. Weighted usabl e area for juvenil e s_ockeye salmon at the Sunset Side Channel study site as a function of mainstem discharge,1984. Weighted usable area and habitat indices for juvenile sockeye salmon at tributary mouth study sites on the 10wer.Susitna River as a function of mainstem discharge,1984. 37 Weighted usab1e area and habitat indices for juvenile sockeye salmon at side channel and slough study sites on the lower Susitna River as a function of mainstem discharge,1ge84. 38 Juvenile sockeye salmon mean catch per cell versus seasonal habitat indices at side channel and tributary mouth modell i ng sites on the lower Susitna River,1984. ,i DRAFT/PAGE 9,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 LI ST OF TABLES Table 1 List of study sites on the lower Susitna River at which juvenile salmon habitat.was modelled,June through October 1984.-, 2 Percent cover and cover type categories. 3 Partitioning of wetted channel width into stream cells. 4 Cl assificati ons and habitat characteri sti cs of study sites on the lower Susitna River at which juvenile salmon habitat was modelled, June through October 1984. 5 Percentages of lower river habitat modelling sites associated with nine cover-type cat- egories. ""'" 7 8 6 Evaluation of RaHAB model quality for extrap- olating WUAs over the range of 12,000 to 75,000 cfs as measured at Sunshine gaging station,1984. Discharge ranges of IFIM models at lower Susitna River sites for which hydraulics are rated acceptable,1984. Preliminary juvenile chinook salmon turbidity criteria derived from lower Susitna River distribution data,1984. 9 Weighting factor for turbidity by site for analysis of juvenile chinook salmon habitat use,1984. 10 Weighting factors for turbidity by site for analysis of juvenile chum salmon habitat use, 1984. >;j \l '1 !'../~' - - -~I A-I A-2 - DRAFT/PAGE 10,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 LIST OF APPENDIX FIGURES Figure Mean catch of juvenile chinook salmon per cell by percent cover category (bars)in clear water of the lower Susitna River,1984 and comparison of fitted suitability indices (lines)calculated in 1984 and for the middle Susitna River,1983. Mean catch of juvenile chinook salmon per cell by velocity intervals (bars)in clear water of the lower Susitna River,1984 and fitted suitability index (line)developed for the middle Susitna River,1983. A-3 Mean catch of juvenile chinook salmon per cell by velocity intervals (bars)in clear water of the lower Susitna River,1984 and fitted suitability index (line)developed for turbid water in the middle Susitna River, 1983. A-4 A-5 Comparison of cover type suitability indices for juveni 1echi nook salmon in clear water calculated from 1984 lower Susitna River distribution data and 1983 middle Susitna River distribution data. Cover type suitabil ity i ndi ces for juvenil e chinook salmon in clear water calculated from 1984 lower Susitna River distribution data. A-6 Mean catch of juvenile chinook salmon per cell by depth intervals (bars)in clear water of the lowerSusitna River,1984 and fitted suitability index (line)developed for the middle Susitna River,1983. A-7 Mean catch of juvenil e chinook salmon per cell by depth intervals (bars)in clear water of the lower Susitna River,1984.Suita- bility index (line)fitted by hand using professional judgement. A-8 Mean catch of juvenile chinook salmon per cell by velocity intervals (bars)in turbid water of the lower Susitna River,1984 and fitted suitabil ity index (1 ine)developed for the middle Susitna River,1983. DRAFT/PAGE 11,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 LIST OF APPENDIX FIGURES (Continued) - A-9 Mean catch of juvenile chinook salmon per cell by percent cover category (bars)in turbid water of the lower Susitna River,1984 and fitted suitability index (line)calcu- lated for the middle Susitna River,1983. A-10 A-ll Cover type suitability indices for juvenile chinook salmon in turbid water developed from 1984 lower Susitna River chinook clear water distribution data. Mean catch of juven"ile ch"inook salmon per cell by depth intervals (bars)in turbid water of the lower Susitna River,1984 and fitted suitability index (line)developed for the middle Susitna River,1983. - A-12 Mean catch of juvenile chinook salmon per cell by depth intervals (bars)in turbid water of the lower Susitna River,1984. Suitability index (line)fitted by hand using professional judgement. A-13 Mean catch of juvenile coho salmon per cell by velocity intervals (bars)in the lower Susitna River,1984 and fitted suitabil ity index (line)developed for the middle Susitna River,1983. - A-14 A-15 Mean catch of juvenile coho salmon per cell by percent cover category (bars)in the lower Susitna River,1984 and comparison of fitted suitability indices (lines)calculated in 1984 and for the middle Susitna River,1983. Comparison of cover type suitability indices for juvenile coho salmon calculated from 1984 lower Susitna River distribution data and 1983 middle Susitna River distribution data. A-16 Cover type suitabi 1i ty indices for juvenile coho salmon calculated from 1984 lower Susitna River distribution data. A-I?Mean catch of juvenile coho salmon per cell by depth intervals (bars)in clear water of the lower Susitna River,1984 and fitted suitability index (line)developed-for the middle Susitna River,1983. x """" DRAFT/PAGE 12,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 LIST OF APPENDIX FIGURES (Continued) Proportion of cells with juvenile sockeye salmon present by velocity intervals (bars) in the lower Susitna River,1984 and fitted suitability index (line)developed for the middle Susitna River,1983 and revised in 1984 for the lower river using professional judgement. A-19 Proportion of cells with juvenile sockeye salmon by percent cover category (bars)in the lower Susitna River,1984 and comparison of fitted suitability indices (lines)cal- culated in 1984 and for the middle Susitna River,1983. A-20 Comparison of cover type suitabil ity indices for juvenile sockeye salmon calculated from 1984 lower Susitna River distribution data and 1983 middle Susitna River distribution data. A-22 A-24 \~ - ,.... .~ A-21 Proportion of cells with juvenile sockeye salmon present by depth intervals (bars)in the lower Susitna River,1984 and fitted suitability index (line)developed for the middle Susitna River,1983. Proportion of cells with juvenile chum salmon present by velocity intervals (bars)in the lower Susitna River,1984 and fitted suita- bility index (line)developed.for the middle Susitna River,1983. A-23 Proportion of cells with juvenile chum salmon present by percent cover category (bars)in the lower Susitna River,1984 and fitted suitability index (line)calculated for the middle Susitna River,1983. Proportion of cells with juvenile chum salmon present by cover types (bars)in the lower Susitna River,1984. A-25 Proportion of cells with juvenile chum salmon present by depth intervals (bars)in the lower Susitna River,1984 and fitted suita- bility index (line)developed for the middle Susftna River,1983 . -------------------------------------- B-1 B-5 DRAFT/PAGE 13,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 LIST OF APPENDIX FIGURES (Continued) Weighted usable area for juvenile chinook salmon at the Caswell Creek and BeaverDam tributary study sites as a function of mainstem discharge. B-2 Weighted usable area for juvenile chinook salmon at the Hooligan and Bearbait Side Channel study sites as a function of mainstem discharge. B-3 Weighted usable area for juvenile chinook salmon at the Last Chance and Rustic Wilder- ness Side Channel study sites as a function of mainstem discharge. B-4 Weighted usable area for juvenile chinook salmon at the Island Side Channel and Mainstem West Bank study sites as a function of mainstem discharge. Weighted usable area for juvenile chinook salmon at the Goose 2 and Circular Side Channel study sites as a function of mainstem discharge. B-6 Weighted usable area for juvenile chinook salman at the Sauna and Beaver Dam Side Channel study sites as a function of mainstem discharge. B-7 Weighted usable area for juvenile chinook salmon at the Sunset and Sunrise Side Channel study sites as function of mainstem dis- charge. B-8 Weighted usable area for juvenile chinook salmon at the Trapper Creek Side Channel study site as a function of mainstem dis- cha rge. B-9 Weighted usable area for juvenile chum salmon at the Hooligan Side Channel and Kroto Slough Head study sites ad a function of mainstem di scharge. B-I0 Weighted usable area for juvenile chum salmon at the Bearbait and Island Side Channel study sites as a function of mainstem discharge. 'X \l .- - DRAFT/PAGE 14,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 LIST OF APPENDIX FIGURES (Continued) B-ll Weighted usable area for juvenile chum salmon at the Mainstem West Bank and Goose 2 Si de Channel study sites as a function of mainstem discharge. B-12 Weighted usable area for juvenile chum salmon at the Circular and Sauna Side Channel study sites as a function of mainstem discharge. - B-13 B-14 Weighted usable area for juvenile chum salmon at the Sucker and Beaver Dam Si de Channel study sites as a function of mainstem dis- charge. Weighted usable area for juvenile chum salmon at the Sunrise Side Channel study site as a function of mainstem discharge. B-15 Weighted usable area for juvenile sockeye salmon at the Caswell Creek and Beaver Dam tributary study sites as a function of mainstem discharge. Comparison of site areas calculated with the RJHAB and IFIM modelling techniques for the Trapper Creek and Island side channel study sites. C-3 C-4 C-5 Comparison of weighted usable areas calcu- lated with the RJHAB and IFIM modelling techniques for juvenile chinook and chum salmon at Trapper Creek Side Channel,1984. Comparison of habitat indices calculated with the RJHAB and IFIM model 1 ing techniques for juvenile chinook and chum salmon at Trapper Creek Side .Channel,1984. Comparison of weighted usable areas calculat- ed with the RaHAB and IFIM mode"ing tech- niques for juvenile chinook and chum salmon at Island Side Cha,nnel,1984 Comparison of habitat indices calculated with the RJHAB and IFIM modelling techniques for juvenile chinook and chum salmon at Island Side Channel,1984. ,,, X i I J ~--------------- DRAFT/PAGE 15,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 - LIST OF APPENDIX TABLES Table - A-2 Kendall correlation coefficients between habitat variables and chinook catch by cell (N=744)for all gear types,in turbid water. A-I Percent cover and cover type categories.- A-3 Kendall correlation coefficients .between habitat variables and chinook catch by cell (N=396)for all gear types,in clear water. A-4 Calculations of turbidity factors for 1984 lower river data.- A-5 Kendall correlation coefficients between habitat variables and coho catch by cell (N=345)in clear water. - - Proportional presence of chum salmon fry associ ated wi th several composite wei ghti ng factors. Kendall correlation coefficients between habitat variables and sockeye catch by cell (N.c:922)• A-9 A-7 Proportional presence of sockeye salmon associ ated wi th the composite wei ght i ng factor calculated by multiplying velocity and cover suitabilities together. A-8 Kendall correlation coefficients between habitat variables and chum catch by cell (N=249)for all gear types,turbidity below 200 NTU. A-6 A-I0 A-ll B-1 Summary of revisions of 1983 middle river juvenile salmon criteria for use in the lower Susitna River,1984. Suitabil ity indices for juvenile salmon for velocity,depth,and cover in the lower Susitna River,1984. Turbidities within modelled side channels of the lower Susitna River,June through August 1984.- - DRAFT/PAGE 16,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 LIST OF APPENDIX TABLES (Continued) B-2 ".... B-3 8-4 B-5 F 8-6 Catch and catch per cell (CPUE)of juvenile salmon within lower Susitna River sampling sites,1984 Lengths of RJHA8 model sites in the lower Susitna River,1984. Side channel flows at the 15 modelled side channels as a function of mainstem discharge. Weighted usable areas and habitat indices for juvenile chinook salmon in lower Susitna River model sites,1984. Weighted usable areas and habitat indices for juvenile coho salmon 'in lower Susitna River model sites,1984.~ B-7 Weighted usable area and habitat indices for juvenile chum salmon in the lower Susitna River model sites,1984. 8-8 Weighted usable areas and habitat indices for juvenile sockeye salmon in lower Susitmi River model sites,1984 • ..... ><V -----~----------------------~-------------- - - DRAFT/PAGE 17,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 1.0 INTRODUCTION - The SU-Hydro juvenile anadromous distribution and abundance studies initiated during 1981 and 1982 outl ined the general distribution pat- terns of juvenile salmon and their habitat utilization (ADF&G 1981a, 1981b;1983a,1983b).The 1982 studies also investigated the response of selected areas to mainstem discharge changes and demonstrated species differences in the use of "hydraul ic zones"(ADF&G 1983c).These zones were subsections of slough and tributary mouth areas.Some zones were affected by rna i nstem backwater,other zones were above the backwater, and other zones included mixing areas of the mainstem with slough or tributary flow.The relative use of the hydraulic zones by each species of juvenile salmon was analyzed to provide an incremental index of habitat availability at each site for each species.This analysis provided evidence that the relative use by Juvenile salmon of these sites was affected by changes in mainstem discharge.Also the distri- bution of juvenile salmon suggested certain microhabitat factors within the zone such as turbidity and instream cover responded to discharge changes at a higher rate than did zone surface area. Studies conducted during the 1983 open-water season concentrated on the instream flow relationships of juvenile salmon in the middle reach of the Susitna River between the Chulitna River confluence and Devil Canyon (Schmidt et al.1984).Suitability criteria for juvenile salmon were developed and these were used in two types of habitat models to model the site-specific response of juvenile salmon habitat to variations in DRAFT/PAGE 18,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 mai nstem di scharge.Additi onal i nformati on was al so gathered on juve- nile salmon abundance and distribution in the middle reach. The 1983 studies suggested that juvenile chinook salmon made heavy use of mainstem side channels and used the turbid water in these areas as cover.Juvenile coho,chum,and sockeye salmon tended to occupy areas that were less influenced by mainstem flow. In the Susitna River below the Chulitna River confluence (lower river), the braided nature of the river and lower gradient provides large amounts of potential side channel habitat for juvenile salmon.A study plan was formulated,therefore,to examine juvenile salmon distribution and the habitat availability of different morphological components of the lower Susitna River for juvenile salmon during the 1984 open-water season.The resul ts of these studi es,whi ch i ncl ude the responses of rearing juvenile salmon and their habitat within these morphologica1 components to variations in mainstem discharge,are detailed in this paper.These results will be integrated with responses of side channel and slough complex wetted surface areas to variations in mainstem discharge in order to estimate the response of juvenile salmon habitat in the lower river to flow regulation. Large scale aerial mapping of side channel and slough complex changes in area with variations in mainstem discharge is currently being performed by R &M Consultants,Inc.and E.Woody Trihey and Associates in asso- ciation with these studies.Habitat types identified in the mapping 2. - - - DRAFT/PAGE 19,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 include tributaries,tributary mouths,upland sloughs,side sloughs, primary side channels,secondary side channels,and turbid backwaters. Tributaries,tributary mouths,upland sloughs,and side sloughs are defined as in the upper river (Klinger and Trihey,1984).Primary side ~t ItS"channels have characteristics similar to the mainstem in the upper river V and therefore offer little potential habitat for juvenile salmon and are not discussed in this report.Turbid backwaters are unbreached channels which contain turbid water from being breached at higher mainstem discharges and therefore are transitory in nature.Turbid backwaters are not addressed in this report but their habitat values are probably similar to barely breached side channels. The major emphasis of this report is the evaluation of juvenile salmon use and related habitat values of secondary side channels.Some of the 1arger secondary si de channel s are consi dered primary si de channel s at higher mainstem discharges. Tributary mouths and side sloughs were also evaluated.Due to their limited occurrence in the reach,upland sloughs were not sampled.The macrohabitat evaluation data presented here will be integrated with the aerial mapping data in later reports to formulate the reach-wide re- sponse of juvenile salmon habitat to discharge variations. 2.0 METHODS 2.1 Field Sampling Design '--....,f'FTl,riA DRAFT/PAGE 20,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 - - Three Juvenile Anadromous Habitat Study (JAHS)field crews,of two biologists each,examined rearing habitats used by juveni1e salmon at selected side channels,triblJtary mouths,sloughs,and mainstem sites of the Sus itna Ri ver between the Yentna Ri ver confl uence (RM 28.5)and Chulitna River confluence (RM 98.5).JAHS sampling was conducted from river boats during the open-water season,with helicopter support enlisted as needed.The crews operated out of camps located on the Susitna River at the Deshka River (RM 40.6),Sunshine Station (RM 79.0), and Talkeetna (RM 97.5). The JAHS field crews sampled three categories of sampling sites.Most of the sampling occurred at Resident Juvenile Habitat (RJHAB)model sites where the response of the site to changes in mainstem discharge was evaluated along with juvenile salmon"use of the site.Crews also sampled Instream Flow Incremental Methodology (IFIM)model sites for fish distribution and abundance at which hydraulic habitat models were developed.The third category of sites,at which further data on fish di stributi on and habi tat were gathered,were known as lIopportuni sti ell sites.Further details on specific sampling techniques and methods used in the JAHS studies are given in ADF&G (1984a,1984b). - - 2.1.1 Study locations and selection criteria DRAFT/PAGE 21,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 The sampling sites modelled were chosen from side channels,tributary mouths,and side sloughs,which met the following basic criteria: ,~ A.The effects of mainstem discharge (stage and flow)on the ~sites are measurable. B.The sites are documented or thought to contain potential habitat for rearing juvenile salmon. C.The sites are accessible by boat at normal mainstem discharges during the open-water season. The sites modelled with RJHAB and IFIM models are listed in Table 1 and - -\ \\ their distribution is shown in Figure 1.Fourteen of the sites were modelled only with the RJHAB model,four with only IFIM models,and two with both RJHAB and IFIM models.Eight of the sites are located within slough or side channel complexes which were picked by R&M Consultants and E.W.Trihey and Associates as representative of lower Susitna River side channel complexes.Four of the sites are normally clear-water sloughs or tributary mouths while the other sites are turbid secondary side channels at normal summer flows.Secondary side channels selected ranged greatly in size,shape,and overtopping discharge.The majority of the habitat model sites are secondary side channels because the majority of the potential available habitat for juvenile fish in lower Tab 1e 1. DRAFT/PAGE 1 4/28/85 NUM2B/Table 1 Juvenile Anadromous Habitat Study (JAHS)modelling sites on the Susitna River between the Yentna River and Talkeetna River confluences,1984.- Site *Hooligan Side Channel *Eagles Nest Side Channel Kroto Slough Head Rolly Creek Mouth Bear Bait Side Channel Last Chance Side Channel Rustic Wilderness Side Channel Caswell Creek Mouth Island Side Channel Mainstem West Bank Goose 2 Side Channel Circular Side Channel Sauna Side Channel *Sucker Side Channel *Beaver Dam Slough *Beaver Dam Side Channel *Sunset Side Channel *Sunrise Side Channel *Birch Creek Slough Trapper Creek Side Channel River ~'1ile 35.2 36.2 36.3 39.0 42.9 44.4 59.5 63.0 63.2 74.4 74.8 75.3 79.8 84.5 86.3 86.3 86.9 87.0 88.4 91.6 RJHAB x X X X X ; X X X X X X X X X X X IFIM X X X X X X -*Located within side channel or slough complexes picked by R&M Consul- tants,Inc.and E.Woody Trihey and Associates as representative of lower Susitna River slough or side channel complexes. !o - ,- Tropper Creek S.C. Birch Slough Sunrise S.C. Sunset S.C. Beaver Dam Slough Beaver Dam S.C. Sucker S.C. Sauna S.C. Ci rculor S.C. Goose 2 S.C. Mainstem West Bonk Island S.C. Caswell Creek Mouth Rustic Wilderness S.C. Last Chance S.C. Bear .Bait S.C. Roll'1 Creek Mouth Kroto Slough Head EogIes Nest S.C. Hooli~on S.C. COOK InlfJl Figure 1.Location of study sites on the lower Sus1tna River at which juvenile salmon habitat was modelled.JunethroughOctOber1984. DRAFT/PAGE 22,"4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 Susitna River mainstem affected areas is composed of secondary side channel s. Opportunistic sampling sites were selected by the sampling crews as potential habitat which upon sampling might pro~;de for a better analy- sis of fish abundance and distribution.Sites sampled were more diverse than the RJHAB and IFIM sites and included areas within alluvial island complexes. 2.1.2 Field data collection 2.1.2.1 Resident Juvenile Habitat (RJHAB)model sites Two types of data were collected at the RJHAB model sites.Habitat data were collected for the purpose of modell ing the response of the site to changes in mainstem di scharge.Fi sh di stributi on data were coll ected for use in verifying the habitat model data,documenting abundance and distribution,and modifying suitability criteria,if necessary.A discussion of the techniques used in the collection of habitat modelling data will be followed by a discussion of methodology used in the col- lection of fish sampling data. Each of the RJHAB sites was sampled within a grid consisting of a series of transects with associated sampling cells which intersect the channel of the study site at right angles (Figure 2).Grids were located so that water quality within them was uniform and so that they encompassed i I ~,... &~ ;,R~NSEC' TRANSECT 2 TRANSECT I Cell Unit Area Sampled Figure 2.Arrangement of transects and sampling cells within a grid at a hypothetical RJHAB modelling site. Q ! DRAFT/PAGE 23,4/26/85 1/9185,3/28/85,5/1/85 NUM2/Part 2,4/14/85 a variety of habitat types.Survey stakes and orange flagging were used to mark each transect within a grid.Initial measurements within each grid included distances and angles between transect bench marks. Transects were spaced from 50 to 300 feet apart in order to encompass a vaTiety of hapitat types within each grid.Aerial photos of all the RJHAB sites showing placement of all transects·within each site are presented in Quane et al.(1985). Up to four 6-by-50 foot rectangular sampling cells extending upstream from every transect within each grid were characterized by habitat measurements (Figure 2).If the top width of the wetted channel was greater than 42 feet,two of the four cells paralleled both edges of the channel and the third and fourth cells were located parallel to the shoreline cells so as to split the channel into thirds.If the channel measured 30 to 41 feet in width at the transect,there was a cell on each shorel ine of the channel and one cell located approximately mid channel.If the wetted edge was 18 to 29 feet in width,there was one cellon each side of the channel parallel with the bank.If the channel was less than 18 feet in width,there was only one cell. Transects were numbered consecutively beginning with the transect furthest downstream within the site.Cells were also numbered consecu- tively from right to left looking upriver.If there were less than four cells within a transect,cells were numbered as if the missing cells were present. -I - """ - - - .- r - DRAFT/PAGE 24,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 One or more staff gages were installed by Aquatic Habitat and Instream Flow Project (AH)personnel at each site to document changes in the stage at each site with changes in mainstem discharge.These gages provided an index to the changes in habitat and hydraulic conditions at the site between sampling occasions.AH staff also developed mainstem stage and site flow relationships and mapped the thalweg at selected sites. Habitat modelling data were collected over a broad range of mainstem discharges.Emphasis was placed on data collection at mainstem dis- charges of 30,000 to 60,000 cfs as measured at the Sunshine USGS gaging station.When staff gage readings and observations indicated that the habitat at the site had not changed from a previous sampling occasion, no habitat data were taken. Habitat data taken at each grid on a modelling occasion included the following.At each transect,the distance between the left and right edge of water and the left bank transect marker was measured.If the water quality within the grid or grids was uniform,one measurement of water pH,temperature,conductivity,and dissolved oxygen was taken.A turbidity sample was collected in a 250 ml plastic bottle and stored in a cool dark location prior to analysis.If the water quality within the grid appeared to vary because of mixed water sources,additional water quality and turbidity measurements were taken as necessary to describe these within grid variations. , I______'_l_.........--~-_ DRAFT/PAGE 25,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 In addition to the above measurements,each sampling cell within the grid was characterized by several habitat measurements.A representa- tive depth and velocity were measured by taking one or more point measurements along the midline of each cell.The entire cell was walked so measurements taken were representati ve.A velocity measurement was taken at 0.6 of the distance from the top of the water column at one representative location for the entire cell. Additionally,cover type and amount were estimated in each cell and coded into categories (Table 2).Initially,the total amount of cover of all types was estimated for the entire cell.Next,the primary and ~ I '\_..-J~.._secondary cover type was recorded along with a percentage of ~.;.-,1/for each.Cover was defined as hiding or ~escape c~for I!Y ~,ill"")1 ---~d~•than or equal to 100 mm in total length. ,'1\ Table 2.Percent cover and cover type categories. the total fish less - Group # 1 2 3 4 5 6 %Cover 0-5% 6-25% 26-50% 51-75% 76-96% 96-100% Group # 1 2 3 4 5 6 7 8 9 Cover Type No object cover Emergent vegetation Aquatic vegetation Debris or deadfall Overhanging riparian vegetation Undercut banks Gravel (lllto 311 diameter) Rubble (3 11 to 511 diameter) Cobble (larger than 511 diameter) In September,when the water levels in the Susitna River were low,the cover on all the transects within each site was systematically recorded. ~ I i - - ..- DRAFT/PAGE 26,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Pa,rt 2,4/14/85 One person di d most of the recordi ng so that observer bi as was mi ni- mized.The cover was recorded by distance from the left bank transect marker'along the transect line. Fish distribution data were normally collected from a minimum of seven cells within each RJHAB site during each sampling occasion.Cells to be sampled were selected randomly by using a random numbers table (ADF&G 1985).If a cell was miss"ing or could not be sampled,an additional cell was randomly chosen for sampling.Some cells could not be sampled due to high velocities or deep depths and~therefore,the sampling was not completely random.Each cell selected was then sampled for fish with one pass through the entire cell with a backpack electroshocker or beach seine. The gear used was that thought most efficient for sampl ing the area, normally beach seines are more efficient in turbid water while electro- fishing gear is most efficient in clear water (Dugan et al.1984).The area of the cell sampled for fish was recorded so that catches in cells with areas different than 300 ft 2 could be adjusted to this standard cell size. Additional selected cells were occasionally fished at the site if sampling of the random cells failed to capture many fish.In this case, the sampling crew fished areas which were thought to be flgood H habitat. Areas fished were not limited to cells on the transects . ._--------~--------------------_.........---- DRAFT/PAGE 27 5 4/26/85 1/9/85 5 3/28/85,5/1/85 NUM2/Part 2,4/14/85 After each cell was sampled,juvenile salmon captured were identified to species and then released.The total length of each of the first 50 fish of each species in each size class was measured in millimeters. If staff gage readings indicated the habitat at the site had not changed from a previous sampling period only limited habitat measurements were taken.These included water chemistry data and a turbidity sample. Fish distribution data were taken during each visit to the site 5 how- ever.Each cell sampled for fish was also characterized by a represen- tative velocity,depth,and estimate of cover type and abundance. 2.1.2.2 Instream Flow Incremental Methodology (IFIM)sites In addition to the RJHAB model sites,there were also six sites modelled for juvenile fish using the "instream flow incremental methodology" (IFIM)(Bovee 1982).A summary of this methodology and specific data collection and modelling techniques are presented in Appendix D of this report.All habitat data used in the IFIM models were collected and analyzed by Aquatic Habitat (AH)personnel.Two of the IFIM sites were also modelled with RJHAB models using the same transects in order to compare output from the two modelling methods.At these two sites,RJ ~. personnel collected the RJHAB and fish distribution data and AH person- nel collected the IFIM data,so the two models were independent. Fish abundance and distribution data were also collected at the other four IFIM model sites.Sampling effort at these sites was secondary in /4 DRAFT/PAGE 28,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/P~rt 2,4/14/85 importance to the sampling of the RJHAB sites.Cells were sampled for fish using the transects placed for the IFIM models.Cells were ran- domly selected and then sampled with the same procedures used at RJHAB sites.Cell numbering was the same as that used in the RJHAB studies. The di stance from the transect end markers to the cell edge was mea- sured,however,so that the location of the cell on the transect was specified.Other data collected at each cell fished included amount and type of cover,water depth,and water velocity.Water chemistry mea- surementsand a turbidity sample were also taken at a selected location within the site. 2.1.2.3 Opportunistic sites In addition to theRJHAB and IFIM sites,other sites were sampled for fish opportunistically as time permitted.The purpose of this sampling was to gather juvenile abundance and distribution information at a wider variety of sites and to gather data for further analysis of juvenile suitability criteria.No permanent grids or transects were marked at opportunistic sites.Selected 6-by-50 foot cells were sampled for juvenile salmon at the opportunistic sites.Water chemistry was measured at mid-site.Each cell sampled for fish was characterized to amount and type of cover,water depth,and water velocity as were cells sampled at RJHAB and IFIM sites,if time permitted. Early in the sampling season,large differences in turbidity were noted between sites located on the east and west banks of the Susitna River is DRAFT/PAGE 29,4/26/85 1/9/85,3/28/85.5/1/85 NUM2/Part 2,4/14/85 mainstem below the Chulitna River confluence.In order to better understand the reason for these di fferences,turbidi ti es were taken within the Talkeetna and Chulitna rivers just above their respective confluences with the Susitna and also in the middle Susitna River above its confluence with the Chulitna River.The turbidity measurements were then repeated in the lower Susitna River below the Chulitna River on the left (west)bank channel,center channel,and right (east)bank channel at intervals from RM 92.7 downstream to RM 60.6.Two sets of measure- -I ments were taken.one on July 19 and the other on August 16. measurements were recorded within a four hour period on each date. 2.1.3 Schedule of activities and frequency of sampling The Fi el d sampl ing tri ps,1asting approximately 7-10 days,were conducted bimonthly from June through mid-October.Each RJHAB site was sampled ~ for fish on each sampling occasion if fish habitat was present.Habitat data were collected on at least three occasions when staff gage readings or observations suggested a change in the habitat within a site.The coll ection of habitat data was therefore very dependent upon ma i nstem di scha rge. The IFIM sites were sampled at least once a month during the open-water season.Opportunistic sites were sampled as toime permitted and some were only sampled once.Opportunistic sites were sampled mainly in September and early October when many of the RJHAB and IFIM sites were dewatered. DRAFT/PAGE 30,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85- 2.2 Data Analysis ,.... !, A11 fi el d data were recorded on the appropriate data forms and trans- mitted to the office where the fish distribution data and much of the habitat data were entered into a mainframe computer data base.Data sorts,summary retrievals,and selected computer files were extracted ,...,from this data base as needed.Other habitat data were entered directly into basic programs or commercial software on a personal computer. 2.2.1 Physical data Overtoppi ng flows at the study sites were observed or estimated from staff gage measurements and flow observations.Data were grouped into nine half-month sampling periods from early June (June 1 -June 15)to early October (October 1 -October 15).Due to logistical constraints, the actual sampling periods did not always run from the 1st to the 15th and 16th through the end of the month. An index to the amount and type of cover within the RJHAB and IFIM model sites was calculated by totalling the linear feet of all the cover types along the transects at a mainstem discharge within the range of 49,000 to 57,000 cfs .At Rolly Creek mouth,Caswell Creek mouth,and Beaver Dam Slough,the response of physical cover to changes in mainstem discharge was also plotted by totalling the cover along the transects at all measured discharges. 17 ---,-,-------------------_._------------------------ DRAFT/PAGE 31,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 The response of RJHAB site wetted areas to mainstem discharge was also plotted using a BASIC language geometry program to calculate wetted area at each transect within a site on each modell ing occasion.After fitting these points by hand using professional judgement,site areas at 3000 cfs increments were measured on the graphs with a digitizer.The .~ IFG HABTAT program calculated wetted areas at the six IFIM sites as a function of side channel flow,and these were also plotted using a mainstem discharge-side channel flow relationship. 2.2.2 Abundance and distribution The classification of macrohabitats used to examine differences in fish distribution among the sites was that discussed in Dugan et al.(1984). The sites were classified as tributary mouths,side sloughs,and side channels.Tributary mouths are sites which are influenced only by .-., backwater effects from the mainstem as their heads do not overtop.Side channels are channels whose heads are overtopped by the mainstem while side sloughs are side channels whose heads are not currently overtopped. Birch Creek slough was classified as a tributary mouth in 1984 because road building activities in the upper part of the slough have closed the head off from the mainstem.Beaver Dam Slough was also classified as a tributary mouth because it only overtops at discharges greater than 80,000 cfs and normally runs clear.Beaver Dam Slough is much more similar to Rolly Creek mouth than to any of the other side sloughs in the lower reach. .- ! '.- DRAFT/PAGE 32,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 Catches within cells with areas other than the standard 300 ft 2 were adjusted to correspond to this standard cell area.The analysis was· then based on the adjusted mean catch per cell. 2.2.3 Habitat modelling of rearing salmon 2.2.3.1 Suitability criteria development SUitability criteria used in modelling the response of juvenile salmon habitat to variations in mainstem discharge for the middle reach of the Susitna River have been developed in Suchanek et al.(1984).Since habitat data collection techniques used in 1984 were the same as those used during the 1983 field season,the middle river suitability criteria were exami ned and modifi ed,if necessa ry,in Appendi x A by exami ni ng lower river distribution data.The suitability criteria developed in Appendix A are used in all subsequent habitat modelling for the lower river. 2.2.3.2 Instream Flow Incremental Methodology (IFIM)models The IFIM PHABSIM system of computer programs was developed by the U.S. Fish and Wildlife Service as a means of describing the mosaic of phys- ical features of a stream which includes hydraulic variables such as depth and velocity and other features such as substrate or cover (Bovee 1982).A hydraulic model is first calibrated which describes the response of hydraulic variables such as depth and velocity to stream flow (Milhous et ala 1981).The HABTAT program is then used to incorpo- 11 ___,_vm~....._........,------_ DRAFT/PAGE 33,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 rate output from the hydraulic model and substrate data with the suita- bil i ty criteri a to produce estimates of the habitat potential (wei ghted usable area)for a given life stage of a species.Weighted usable area (WUA)is calculated as follows (Bovee 1982): C.,,s WUA =C.X A,_, ,s where:=the composite weighting factor (sometimes called the joint preference factor)for cover,velocity, and depth of the cell (i)for the species and life stage (s) =the surface area of the cell Each cell is a small section of the study channel which ;s bounded by other cells or the shorel ine and extends midway between transects.The WUA for the study site at a given discharge was calculated by totalling all the individual cell WUA's.The composite weighting factor was calculated by multiplying the suitability indices for cover,velocity, and depth of the cell together.WUA's at each study site were calculat- ed at flows which corresponded to 3,000 cfs increments of mainstem discharge as measured at Sunshine gaging station. - Much more detail ed descri pti ons of the IFIM data ana lysi s methods and hydraulic simulation results are presented in Appendix D.Only selected WUA results as a function of mainstem discharge are presented here.All species and site combinations were run and are available on request but ~ space limitations prevent presentation here.Site/species combinations presented were selected on the basis of fish catches at the site. 20 -, - - ,., - r- i DRAFT/PAGE 34 s 4/26/85 1/9/85 s 3/28/85 s 5/1/85 NUM2/Part 2 s 4/14/85 2.2.3.3 Resident Juvenile Habitat (RJHAB)models The original RJHAB model was designed to calculate weighted usable areas for the habitat within a site without using hydraulic models (Marshall et al.1984).The model divided the site into shoreline and mid-channel sections,and calculated weighting factors for cover and velocity for each section which were then multiplied together with area to produce a weighted usable area estimate at each of the discharges measured. The original RJHAB model was greatly modified for the 1984 analyses. These changes were made so that the RJHAB model calculates weighted usable areas similarly to the HABTAT program described by Milhous et al. (1981)that is used in IFIM analysis.Also the cover coding has been standardized for input so that observer variations in rating cover at different discharges do not lead to variations in cover estimates unrelated to changes in wetted area. The current~model is a spreadsheet developed on commercial soft- ware.Though no hydraulic model is developed,it closely resembles the HABTAT model in its procedures for calculating weighted usable areas within a site.Instead of calculating weighting factors for cover and velocity in shorel ine and mid-channel sections on a given sampl ing occasion as did the original RJHAB model,each site is partitioned into "s tream cell s"each with a uni que area,cover type s cover percentage, velocity,and depth.The site weighted usable area (WUA)is then the sum of the "stream cell"W.~.i.~Wi-lt~';l;_~'l&jp.l:fti.~"lI't "'~~__1I!*,~t~t>lf"ljMOgether. 21-_.......__.._,------------------------------------ DRAFT/PAGE 35,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 - The velocity and depth measurements of the 6'x 50'sampling cells are assumed to represent a much larger stream cell.The wetted surface area between transects was partitioned into one to four stream cells depen- dent upon wetted transect width (Table 3). How Area PartitionedWettedChannelWidth Table 3.Partitioning of wetted channel width into stream cells. No.of Stream Cell s >42 ft 4 Cellon each shoreline 6 ft in width,two center cells split the difference. 30-42 ft 18-29 ft 18 ft 3 2 1 Cellon each shoreline 6 ft in width,middle cell is the rest. Each cell with half the width. Entire width. """'1 J Occasionally,islands prevented a simple partitioning of the site but in each case,areas were partitioned so that sampling cells best repre- sented a given stream cell.Once the wetted width of stream cells was partitioned,a computer program written in BASIC was used to calculate the surface area of each stream cellon each sampling occasion.The areas of islands were estimated from the observations and sketch maps and then subtracted from the area of each stream cell. ~I ~ I i Cover suitabilities for each stream cell were calculated with a BASIC program which integrated the standard cover data taken on each transect with the partitioned wetted width of each stream cell.The cover suitabil ity of each cover type on the stream cell wetted wi dth was 22. - - - - ---~ DRAFT/PAGE 36,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 averaged with the other cover suitabilities present (proporti anal to their occurrence)to give an average cover suitability.For example,if the stream cell was 15 ft in width and ten ft of the width was a cover type with a suitability of 0.5 and the other five feet was a cover type with a'suitability of 1.0,the average cover suitability for the cell would be :[(10 x 0.5)+(5 x 1.0)]/15 =0.67. The RJHAB spreadsheet then took.the stream cell areas and cover suit- abilities,and multiplied these with the depth and velocity suitabil- ities which it assigned to the sampling cell depth and velocity measure- ments.The products of these calculations (stream cell WUA's)are then totalled to calculate site WUA1s for each sampling occasion.Weighted usable areas for chinook salmon in turbid and clear water and chum, coho,and sockeye salmon were all calculated concurrentl~. Weighted usable areas were plotted over the range of mainstem discharges sampled.Since initial overtopping flows were estimated for each side channel,WUA response was extrapolated in the range around breaching using this information.Habitat indices were calculated by dividing the WUA of the site at a gi ven di scharge by the site area at the same discharge and these were also plotted.Only selected site and species combinations are presented here,all other WUA calculations are avail- able upon request.Individual sampling cell measurements are also available upon request. In order to compare output from the RJHAB model with that of the IFIM methodology,two sites (Isl and and Trapper Creek si de channel s)were ---_._,-----_.........-~--~--........._----~----------------- DRAFT/PAGE 37,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 modelled with both techniques.Output from both techniques were graphed as a function of mainstem discharge and then correlated with each other at the measured RJHAB discharges. 2.2.3.4 Model verification Fish abundance data were collected at all of the IFIM and RJHAB sites. High mean catches per cell (CPUE's)should reflect high densities of fish within the site.Since WUA's partially reflect the size of a site, they do not by themselves reflect the habitat qual ity of a site.The ,,~~calculated by dividing WUA by site area (at any given v discharge),however,does reflect site quality,independently of site "'----~-."'-_._.,----"'-_._---------._--~ area. Mean seasonal habitat indices for each site were calculated for each species "lith the following procedure.Mean daily discharges for each day between May 15 and October 15 were rounded to the nearest 3,000 cfs increment in the range from 12,000 to 75,000 cfs.The season for chum salmon ran from May 15 to July 15.If the discharge was greater than 75,000 cfs,the discharge was assumed to be 75,000 cfs because WUA's were cal cul ated only up to 75,000 cfs.Corresponding WUA's and si te areas corresponding to these di scharges were then totall ed to fi nd the total WUA and site area for the season.The mean seasonal habitat index -I - was then calculated by dividing the total WUA by the total site area . •ft~{~i~.~iif«..t~1JIlj'~6y~~ze~~§ktJor ~ The turbidity factor was calculated by fitting a suitability index from °to 1.0 on the dis- - /..... - -- DRAFT/PAGE 38,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 tribution of mean chum and chinook juvenile salmon catch by 50 NTU l turbidity incremen2.Site mean CPUE's were regressed against site t\ habitat indices at each site. ~ I ,.,-L~ DRAFT/PAGE 38.1,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 3.0 RESULTS - 3.1 Seasonal,Spatial,and Discharge Related Variations in Habitat 3.1.1 Macrohabitat type classifications of study sites All the study sites were classified into one of three macrohabitat types:tributary mouths,side channels,or side sloughs.Classifica- tion and habitat characteristics of the twenty modelled study sites are given in Table 4.Initial overtopping discharges for the side channels ranged from approximately 8,000 to 46,500 cfs with flows controlled by the mainstem at least 50%of the time while the tributary mouth sites were never overtopped at flows less than 53,000 cfs and site flows were controlled by the mainstem less than 5%of the time.Backwater effects were the only effects attributable to mainstem discharge at the tribu- tary mouths on all sampling occasions except at Beaver Dam Slough where di scharges greater than 75,000 cfs caused the head to overtop and flow _ to increase through the site.Even at discharges greater than 75,000 cfs however,the major effect of mainstem discharge on Beaver Dam Slough was a backwater response. The side slough macrohabitat type was not represented by any of the sites when mainstem discharges were highest during the period from late June through early August.Side slough habitat increased with decreases in mainstem discharges. - ]}-.1 1 )}0))-I 1 DJAFT /PAGJ 1 ] 4/11/85,4/29/85 NUM2!Table 4 Tabl e 4.Classifications and habitat characteristics of study sites on the lower Susitna River at which juvenile salmon habitat was modelled,June through October 1984. Site Ri ver Mile Initial Overtopping Discharge (cfs) Percent of Time Flow Controlled by 1 Mainstem in 1984 Non-mainstem Water Sources 35.2 22,000 80 Pools only 36.2 8,000 94 Unknown 36.3 33,500 62 Minor upwelling 42.9 35,000 64 Pools only 44.4 25,000 79 Pools only 59.5 19,000 86 Pools only 63.2 33,000 64 Major upwelling 74.4 20,000 86 Major upwell i ng 74.8 28,000 68 Minor upwelling 75.3 35,000 64 Major upwelling 79.8 32,000 62 Mi nor upwell i ng 84.5 28,000 71 Minor upwelling 86.3 46,500 50 Unnamed tributary 86.9 27,000 68 Major upwell i ng 87.0 35,000 64 None 91.6 35,000 57 Cache Creek 39.0 >100,000 0 Rolly Creek 63.0 >100,000 0 Caswell Creek 86.3 >75,000 <5 Unnamed tributary 88.4 53,000 <5 Birch Creek N --) Side Channels (head open)/ Sloughs (head closed) Hooligan Side Channel Eagles Nest Side Channel Kroto Slough Head Bear Bait Side Channel Last Chance Side Channel Rustic Wilderness Side Channel Island Side Channel Mainstem West Bank Goose 2 Side Channel Circular Side Channel Sauna Side Channel Sucker Side Channel Beaver Dam Side Channel Sunset Side Channel Sunrise Side Channel Trapper Creek Side Channel Tributary Moutps Rolly Creek Mouth Caswell Creek Mouth Beaver Dam Slough 2 Birch Creek Slough These percentages based on controlling breaching discharges presented in Quane et al.(1985)for the period from May 15 to October 15,1984. 2 A culvert at the head of this slough is frequently blocked and therefore little mainstem water flows into the slough,even if the· slough head is overtopped.The effect of mainstem discharge on this site is minimal for this reason. i! I.!l;. DRAFT/PAGE 39,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 Major object cover differences among the mode"ing sites were differ- entiated by macrohabitat type.An index of cover for each site at a discharge of approximately 52,000 cfs (range 45,500 to 58,800 cfs)was calculated for between-site comparisons of cover (Table 5).The per- centage of the site with the primary cover type,aquatic vegetation, varied from 8.5%to 68.5%for the tributary mouths,while none of the side channel/sloughs had anyoaquatic vegetation.Substrate in the form of large gravel (1-3"diameter)and rubble (3_5"diameter)was the primary cover type for an average of 62%of the area of side channels, while these two cover types only covered an average of 14%of the area of tributary mouth sites.The density of cover at tributary mouths was almost three times that of side channels also.Side sloughs,which by definition are dewatered side channels,often were even more cover poor than side channels. Cover,in the form of turbidity was much more frequent within side channels than at tributary mouths.Turbidities were consistently much higher in the side channels than in the tributary mouths during the entire open-water season (Figure 3).A few turbidities of 100 to 150 NTH were recorded at Rolly Creek mouth and Beaver Dam Slough due to rapid increases in mainstem discharge which caused mainstem backwater to intrude into the sites,or in the case of Beaver Dam Slough,by a slight overtopping of the channel head by mainstem water.Turbidities within the side sloughs ranged from 1 to 19 NTU with a mean of 5.2 NTU. - - r);'} ~61 /, :~;',,'Ill!!_a -_ })))r }1 1 1 1 }1 )"1l )J J DRAFT /PAGE 1 4/11/65,4/29/65 NUM2ITable 5 Table 5.Percentage&of ·Iower river habitat mo<lelling &Ite&auociate<l \'WIth nine cover~type categorle&.Percentages are base<l on the wi <lth.of transect \'With each COver type. Cover Index calculated by dividing total cover by total area of &Ite. Percentage of Site With Primary Cover Type Overhang. River Discharge No Emergent Aquatic Large Riparian U.C.Cover Side Channeh/Slough&Hile Date (ch)Cover Vega Vega Gravel Rubble Cobble Debris Vega Banks Total Index Hooligan Side Channel 35.2 7/14 52400 18.9 0.0 0.0 72.0 0.0 0.0 B.5 0.6 0.0 100.0 13.7 Kroto Slough Head 36.3 7/17 49600 56.4 0.0 0.0 8.6 0.0 0.0 33.5 1,6 0.0 100.1 1.6 Bear Bait Side Channel 42.9 7/13 52400 0.0 0.0 0.0 66.6 0.0 0.0 26.1 3.7 1.4 100.0 11.5 Lut Chance Side Channel 44.4 7/12 54100 23.5 0.0 0.0 63.5 0.0 0.0 12.3 0.8 0.0 100.1 5.9 Rustle Wllderneu Side Channel 59.5 6/12 52900 0.0 0.0 0.0 60.9 30.0 0.0 7.6 0.8 0.5 100.0 13.7 1$land Side Channel 63.2 7/19 51600 13.4 0.0 0.0 62.0 21.6 0.0 0.0 1.4 1.6 100.0 10.5 Maln&tem West Bank 74.4 Extrapolated 1.0 0.4 0.0 43.4 49.3 0.0 2.2 3.4 0.4 100.1 22.7 Goose 2 Side Channel 74.8 7/20 52600 2.0 0.9 0.0 24.3 51.8 13.7 3.5 3.5 0.2 99.9 22.5 Circular Side Channel 75;3 7/24 56600 20.4 0.0 0.0 48.4 21.3 0.0 5.3 4.6 0.1 100.1 9.3 Sauna Si de Channel 79.8 7/23 56600 93.4 0.0 0.0 0.0 0.0 0.0 4.3 2.4 0.0 100.1 0.5 Sucker Si de Channel 84.5 7/09 55400 80.2 8.4 0.0 6.6 0.0 0.0 3.9 1.0 0.0 100.1 1 .1 Beaver Dam Side Channel 86.3 7/08 57100 55.9 0.9 0.0 18.6 5.9 0.0 18.6 0.0 0.0 99.9 1.9 Sun&et Side Channel 86.9 7/22 57800 15.0 0.0 0.0 66.8 9.7 0.0 7.7 0.5 0.3 100.0 4.8 Sunrise Side Channel 87.0 7/07 58800 4.0 0.0 0.0 51.4 44.6 0.0 0.0 0.0 0.0 100.0 10.0 Trapper Creek Side Channel 91.6 8/19 57200 2.2 0.0 0.0 39.1 58.8 0.0 0.0 0.0 0.0 100.1 12.3t-,MEAN 25.8 0.7 0.0 42.2 19.5 0.9 9.0 1.6 0.3 100.0 9.5 ...U Tributary Mouths Rolly Creek Mouth 39.0 7/11 55100 6.9 25.2 46.2 0.0 0.0 0.0 21.5 0.1 0.0 99.9 24.2 Caswe 11 Creek Mouth 63.0 8/18 45400 2.9 5.3 46.2 17.6 0.0 0.0 18.4 1.6 6.1 100.1 19.0 Beaver Dam Slough 86.3 7/08 57100 6.B 9.9 68.5 0.0 0.0 0.0 11.1 3.1 0.6 100.0 57.8 51 rch Creek Slough B8.4 7/20 52600 36.8 0.5 8.5 29.2 9.0 0.0 13.6 2.2 0.3 100.1 6.3 MEAN 13.4 10.2 42.9 11.7 2.3 0.0 16.2 1.8 1.8 100.0 26.8 1 -~t t f I I I I I 61 62 71 72 81 82 91 92 101 SAMPLING PERIOD 150 140 - 130 120 110 - ~100 -....z 90 - >-80 -.... 0 70 - CO 60 -a: ~ 50 -... 40 - 30 - 20 - 10 0 TRIBUTARY MOUTH TURBIDITIES (MODELLED SITES ONLY) •MEAN IRANGE - $u.s 4 Figure 3.Turbidities of modelled side channels and tributary mouths on the lower Susitna River,June through October 1984. 3D - """ - - ..- - .DRAFT/PAGE 40,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 3.1.2 Chulitna and Talkeetna River plume influences on turbidity of side channels Turbidity measurements of the lower Susitna River taken in west bank, mid-channel,and east bank portions of the mainstem indicate that)llume influences of the Chulitna and Talkeetna Rivers extend at least 20 to 30 --------'Q -'-",--......-..._..........--~--......_"---- .miles downriver "(Figure 4).On September 2,turbidities at RM 83.8 <t'"--,~.--,,,;,,,,,,,-,,,,,,,,",",.,_~~- ranged from 60 NTU on the east bank,to 77 NTU in mid-channel and 88 NTU on the west bank.West bank turbidities are much higher than on the east bank,because the Chulitna River is three or more times as turbid as the Talkeetna River and middle reach of the Susitna River. A comparison of turbidities at the modelled side channels located above RM 70 also suggests that the plumes have major effects on turbidities downst ream.Mean turbi di ty at side channel s located on the west bank (Mainstem West Bank,Sauna S.C.,and Trapper Creek S.C.)during June through late August was 377 NTU.During the same time period,side channels located on the east bank (Goose S.C.,Sunset S.C.,and Beaver Dam,S.C.)had a much lower mean turbidity of 158 NTU.Mean turbidities for all the side channels modelled with the exception of Eagle's Nest Side Channel have been calculated in Appendix Table B-1 • 31 - -'( - 100 10C 90 90 "MIDgLI 8U8ITNA \\ '-.......:\ '----l>---~ TALKU-;;'-;'-~-". ,------ SO RIVER WILE 80 RIVER MIlE "\ -8"----'...--£01---e-----a---_._---g------ ----+----..//<;::::---¥'" .--_._-~--_.~,. 70 70 ----,-----r-----,-----i-----,---r- "i _ *\,.J -e - CHULITNA 60 60 j __0,,,•••_"_-----------------._-,,_.-._-- oIULY 1 ....,••4 -Cl WEST BANK +MID •CHANNEL -Q EAST BANK o 100 1300 ,------------------ 1200 _AUaUIT 18.1'U Figure 4.Comparison of turbidities in the lower Susitna River below the Chulitna and Talkeetna river confluence on July 19 and August 16~1984. 1100 1000 900 - ---:>800....z--700~ t::6009 III lX._500 -:>.... 400 - 300 200 100 - 0 1300 1200 1100 1000 900 ---j;:800 z 700 ~ t::6009 III lX._SOD:~~ ~ 400 300 200 - DRAFT/PAGE 41,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 3.1.3 Physical responses of sampling sites to mainstem discharge variations Variations "in mainstem discharge cause the heads of side channels to open and close,thereby affecti ng macrohabitat cl assifi cati ons due to the subsequent changes in water quality.Increases in side channel flows,areas,and the amount of cover al so occur with increases in mainstem discharge after head overtopping.The relationships between side channel flows and mainstem discharge are presented in Quane et al. (1985). Changes in area of the sites due to increases in mainstem discharge are important because increases may directly increase habitat.Area in- creases measured from aerial photos are being compiled for selected side channel and slough complexes by R&M Consultants Inc.and E.Woody Trihey and Associates.Also cover responses at tributary mouths caused by mainstem backwater effects are significant because object cover is an important component of these sites for juvenile salmon.Discharge related responses of site area for all sites pooled and cover for Z selected tributary mouths will be presented in the next two sections. 3.1.3.1 Area The areas of the RJHAB study sites were calculated geometrically at modelled discharges,and then plotted against mainstem discharge.These points were then fit by hand uS'ing professional judgement.Area mea- surements at 3,000 cfs increments were then taken from these graphs in r".--,....,;-.-i DRAFT/PAGE 42,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 the range from 12,000 to 75,000 cfs.Since Eagles Nest Side Channel was modelled only at discharges less than 20,000 cfs,we did not try to extrapolate values over this range for this site.Similarly,area response at the six IFIM sites were calculated by the IFG program at side channel flows which corresponded to inc"rements of 3000 cfs within the 12,000 to 75,000 cfs mainstem discharge range. Individual area responses for all the modelling sites have been tabu- lated in Appendix Table 8-4 at 3,000 cfs discharge increments.Also, side channel flows associated with these increments have been tabulated. By summing areas of the sites by macrohabitat type,the response of the area of all the sites pooled can be illustrated.The combined area of three tributary mouths increased greatly at discharges greater than 27,000 cfs (Figure 5).Since sloughs become side channels at greater di scharges,slough habitat decreased with di scharge whi 1e si de channel habitat steadily increased (Figure 6).The slough habitat is broken into the two categories,total and accessible.The total category includes ponded water with no access from the mainstem while the acces- sible sloughs are those with potential access from the mainstem. 3.1.3.2 Cover Since instream cover is an important component of fish habitat,the response of available cover to mainstem discharge at individual sites is of interest.Increases in instream cover at si de channel s are often accompanied by large increases in flows and related water column veloc- ities.Therefore,increases in cover at side channels are often offset - - T F~:1E3 U TA F~::(~,,/1 ()UT H :; (8IRCH SLOUGH EXCLUDED) 70 i I '-'-~---'-----'------1 30 50 (Thousands) tdAINSTEM DISCHARGE AT SUNSHINE (ds) 22D - 280 - 270 - 260 - 210 //lJ'/ /fZJ 111 / ,l 200 / 190 -,¢ 180 I 170 -"~:~-.l2I//ri/ 140 ' -//1.30 A::J /' 120 .....n' 1 10 +_=8-_'-I3--G--="--'~-r~=-~-'---rI --r__ 10 250 - 240 - 230 300 -- 290 - -- Figure 5.Area within modelled tributary mouths as a function of mainstem discharge at the USGS Sunshine gaging station.1984. -- 35 SLOUGHS + o TOT A L----------j ACCESSIBLE .30 ~o 70 (Thousands)~INSTEt.A DISC'HARGE AT !;UNSHINE (cfs) SIDE CHA!'JNEL~3 (EAGL.ES NEST S.C.EXCL.UDED) 2 .,-------~----------.....:...---..a--&- 1.92"'- 1.a _..,// 1.7 ,"" 1 .6 GIf'/s 1.5 /""/ 1.4 1.3 1.2 1.1 1 0.9 o.a 0.7 0.6 0.5 0.4 0 ..3 0.2 0.1 /;!/ o --I-is-e--a:-r-----r---....,..----,r--------,..----,--.-----l 10 Figure 6.Area within modelled sloughs and side channels as a function of mainstem discharge at the USGS Sunshine gaging station,1984. 3G - - - DRAFT/PAGE 43,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 by increases in velocities which make the site unsuitable.Turbid water in side channels may also provide cover for juvenile salmon and there- fore,instream object cover may be 1ess necessary under turbid con- ditions (Suchanek et a1.1984). At tri buta ry mouths,on the other hand,flows are independent of rna i n- stem discharge,the water is often clear,and the primary effect of mainstem discharge is the formation of a backwater zone.Increases in mainstem stage typically decrease water velocities at tributary mouths while cover increases.Increases in cover at tributary mouths with increases in discharge may be of more importance than at side channels because the clear water provides little cover,and changes in velocities are relatively less.In the habitat modelling section,cover responses are integrated with velocity and depth changes for all the sites,but here only backwater instream cover responses at tributary mouths wi 11 be presented. Cover responses to mainstem discharge at the four tributary mouths va ri ed greatly.At Bi rch Creek Slough"there was no response of cover to changes in mainstem stage during the 1984 open-water season.This occurred because the sampling site was located so far up the channel that it was not influenced by mainstem stage during the 1984 sampling trips.At the other three sites,however,mainstem backwater effects were measurable. At Beaver Dam Slough,only limited increases in total'cover were caused by increases in mainstem discharge because most of the cover was aquatic --------,-_.......-------'-------------------------~ DRAFT/PAGE 44,4/26/85 1/9/85,3/28/85,5/1/85 NUM2/Part 2,4/14/85 vegetation (Figure 7).At Rolly Creek and Caswell Creek mouths,how- ever,the amount of cover at the sites increased rapidly within the sites at discharges larger than 45,000 cfs.Increases in total cover at Rolly Creek mouth were caused primari ly by increases in emergent vege- tati on whi 1e increases in both emergent and overhangi ng riparian vege- . tation were responsible for the large increase in total cover at Caswell Creek mouth. - ,.... BEAVER DAM SLOUGH COYER RtSf'ONS E ~0.' ~- 0.11 0.7 0 TOTAL.....+AooAT 'lEG"z 0.60...0 tWEll'vEt:..w 0.5II< II<...0.6~>0 0 0.3 0.% D.'Q ~-_'<I 0 ZO 3D "0'50 eo 70_('rholloando) ROLLY CREEK MOUTH ..",-0.'TOTAL /'A / D.a +AooAT YEt: " ,/ c EWER YEt: 0.7 j]..0 GEHRIS //,..z 0.80......0.5II<.. W o.~>-0 J Q 0.3 0.% ,...., 0.1 D %0 60 eo 110tn..*\ ~ CASWELL CREEK MOUTH 0.\1 ..TOTA~/+AooAT 0.8 EWER0 0.7 0 OYERRIP../.. 0.6z 0.... 0.5.. II<....0.6>0 0 ,~:0.3 0.2 ... "-0.\ 0 10 30 so 70\Thoua_cbl WAlNSTtw OISCHARGt A1'_UNSHINE (cl.) Figure 7.lnstream cover response at Beaver Dam Slough,Rolly Creek and Caswell Creek mouths as a function of mainstem discharge at the USGS Sunshine gaging station,1984. 3j DRAFT/PAGE 1 3/7 /85,4/15/85 NUM2/Results,4/28/85 -3.2 Distribution and Abundance of Juvenile Salmon Chinook,coho,chum,and sockeye salmon juveniles were captured at the twenty habitat model sites,but only one pink salmon fry was captured. Pink salmon outmigrate early and our methods are not effective at capturing them.A summary of the juvenile chinook,coho,chum and sockeye salmon catch and catch per cell (CPUE)data by site is given in Appendix .Table B-2. 3.2.1 Chinook salmon - A total of 1,458 juvenile chinook salmon were collected in the lower II!!01!i reach of the Susitna River from June through early October.About 83% (1,209)of these fish were captu'red at the 20 habitat model sites.Age 0+fry accounted for 93%of the chinook salmon juveniles captured.The percentage of 0+fry increased from 66%in late June to 99%in early August.All chinook fry captured after early August were 0+fish, indicating that 1+chinooks had outmigrated from the study reach prior to August 15. -, Chinook fry were widely distributed at the modelling sites from early June through late August (Figure 8).Last Chance Side Channel was the only site where no chinook juveniles were captured.Chinook juveniles were captured at 80%or more of the sites sampled in early June and late August.In September and early October,the proportion of sites where chinook salmon were captured decreased.vl.J.2 ..f (.J-{~w'1- -~c'I - ,- ,- / ~~'~•~".I/CHINOOK SALMON ':0 a'~ -.'f-ll .E 0~,.~o\.SAMPLING P RI 0 -~I JUN .JUL AUG SEP OCT ~b.(Site I.II I II I ll:I ::n:I ,L-...Trapper Creek S.C..'0 0 IZ1 0 ~~0 0 ..-.~'Birch Slough I2l /:;iiI 0 0 IZJ Gil 0 0 0 \~\."'~sunrise s.c.0 -!A [;iii •~ \~~""Sunset s.c.--.•~l::ii I2l -I Beaver Dam Slough 0 l::ii 0 0 0 [;jI 0 0 -~ lq Beaver Dam S.C.1ZI •[;iii •[;iii ~Ciii 0 - ~Sucker S.C.Ci IA lZJ IA 0 lZJ 121 - -~(SQuna S.C.(2]-10 0 1211121 1 []-- ~Circular S.C.lA 0 IZ1 lZl [;j Gil 0 - -~Goose 2 S.C.[;;j [;iii ~0 ~[;iii 0 - -::)~Mainstem West Bank - - -0 0 [2]f:jji 0t~:J Island S.C.IZl CiiI [:;;iiiI IA [;;iiI fZI 0 0 -,(,J lr~",-\.:c:':'a:':sw'::"e:"II':'C:';:r~e8-k-·M-o-u-th--·-+:.~~+=O=+:.:=i-=.=+.=r.0=;-t;::0:;-'r.O=:;i ~9 .,)~(\.:R:';:U~st;;':iC~w::-ild--e;"'r-ne;;':s-s-:s:-.c:-·--HI:;;iI~·H;:Cj:;.+=.~IA~-;:fA:;t;I2]~rrCi=it~O~':"'~ ~~~~~~Last Chance S.C.- -0 0 0 0 - 0 - Q ~---...........Bear Bait S.C.-0 IZ1 0 lZl 0 0 - -~'i..:~R~O~II-y~...='.:c~re;e~k~M~ou;t=h=====~f2]~~:I21:':f2l:.'~~IZI==H:+~~~~[;.j~~~[;jI~~';:::IA=,=-t;rlZl~~ ""P"OJ ..........:K.::..ro~t..:..O...:S;..,IO_U.:.9h_...:H_ea_d __-r0~.:::O=+0::::=-·rD=:.p=D=--rlZJ=-~O:::::;·;t;:::::;t;=;-1 Eagles Nest S.C.- - - - - -0 0 r:;;jjI Hooligan S.C.CiiI 0 (2'1 Cij l::ii 0 0 -10 MEAN CATCH PER CELL RELATIVE ABUNDANCE KEYo0.00 ~0.25-2.50 (2]0.0/"0.25 •>2.5,0 -No sample Cook Inlsl l - - Figure 8.Seasonal distribution and relative abundance of juvenileA chinook salmon on the lower Susitna River,June through '\. mid-October 1984. ,} .....-.a • of chinook salmon was -0.63 (p<:0.05). T4'+~~rrL V--~~/~-. :- () DRAFT/PAGE 2 3/7/85.4/15/85 NUM2/Results,4/28/85 Mean juvenile chinook CPUE was highest at tributary mouths,where 1.5 fish per cell (fpc)were captured,although a side channel CPUE of 0.8 fpc was also substantial.Slough catch rates were consistently low (0.1 fpc).Mean catch rates at side channels remained generally constant throughout the season,while tributary mouth CPUE's peaked in August (Figure 9).The peak CPUE for tributary mouths occurred in late August at Caswell Creek mouth (20.2 fpc).The peak CPUE in a modelled side channel (4.4 fpc)occurred at Sunset Side Channel.CPUE's within the side channels peaked at turbidities of 100 to 150 NTU (Figure 10).The ~r'correl ation between the mean turbidi ty for the model 1ed si de channel s/~------'-_._~--- ~and mean catch per cell ~cr--tv-{"a I Catches at Trapper Creek Side Channel appeared to reflect the effect of turbidity upon chinook fry use.This west bank site was below the Chulitna River,the major turbidity source in the lower river.The site had a high CPUE for early June (2.7 fpc),then zero catches of chinook in 1ate June and early July when turbi dity 1evels were above 550 NTU.~ Chinook fry were captured at low levels on subsequent trips when the turbidities again decreased. 3.2.2 Coho Salmon A total of 442 juvenile coho salmon were captured within the lower Susitna River study area.All but five of the juvenile coho salmon were captured within the habitat model sites.Three age classes of juvenile coho salmon were captured.Eighty-six percent of the juvenile coho - i !,-,iL -l.S -SIDE CHANNELS 4 "lSSJ TRIBUTARY wc>UT-HS~e- O 3.5z i 0 3 ......2.5... 0 Q<...2~ xu...1.5<{ t> Z <{: w :::E 0.5 0 11 62 71 72 81 82 el ~2 Hil SANPUNG PERIOD Figure 9.Juvenile chinook salmon mean catch per cell at side channels an~tributary mouths on the lower Susitna River by sampling period.June through mid-October 1984. Juvenile chinook salmon mean catch per cell at modelled side channels on the lower Susitna River by turbidity increment. June through mid-October 1984. :>400575125"'75 225 275 TURBIDITY (NTU) SIDE CHANNEL CHINOOKS (Iolod.n.d .,t..only) 75 1.9 ~1.6, ,.7 ,.6 1.5 ::l 1.4 ...'.5 <..'.~ 0< I.";' '.oJ ,.1"- :c~C.9 -ell:'J.B'-' ~C.7 :::E 0.6 0.5 0 .... 0.5 :l.~ a.l 0 25 Figure 1O, - UM DRAFT/PAGE 3 3/7/85,4/15/85 NUM2/Results,4/28/85 captures were from the 1983 brood year (0+),14%were from the 1982 brood year (1+)and one 1981 brood year (2+)juvenile was captured.The percentage of 1+fry captured decreased from a high of approximately 50% in early June to approximately 2%in early October. Juvenile coho salmon were unevenly distributed throughout the study area (Figure 11).Coho were captu.red at only 50%of the 20 modelled sites and only single coho captures were made at four of these sites.Juve- nile coho CPUE's,in most instances,tended to be somewhat higher in late summer. Di stribution of juvenile coho salmon was extremely di sproportionate among the three macrohabitat types.The tributary mouths had a mean juvenile coho CPUE of 1.2 fpc while sloughs and side channels had CPUE's of 0.02 and 0.01 fpc,respectively.Juvenile coho were captured at all four tributary mouths,five of the 16 side channels (31%)and two of the 14 sloughs (14%)sampled.Caswell Creek was the primary capture site, contributing over half of the total juvenile coho CPUE,with most captured from mid to late August. The juvenile coho catch rate at tributary mouths ranged from near ten juveniles per cell at Caswell Creek in late August to zero fish per cell at several sites during various sampling periods (Figure 12).With the excepti on of Bi rch Creek 51 ough,coho CPUE I S were hi gher duri ng 1ate summer and fall than during early summer sampling periods. ~, / ~ :)0 /~~f/~\,i COHO SALMON ~I) ":1:1 e\t- _....'--;.".\/~6\SAMPLING PERIOD "~JUN JUL AUG SEP OCT:.::d Site I II I ]I I II:I :n:I~~Trapper Creek S.C.0 0 DOD DOD 0 .<l ~"""Birch Slough •IZl l:;;j 0 I:;;iiI 0 I:;;iiI I:;;iiI [2J~~sunrise S.C.0 - 0 0 0 0 - - - \ Sunset s.c.- - -0 0 DOD - ~I Beaver Dam Slough 0 ~f21 I:A f2I GiiI lZl • - ~~Beaver Dam S.C.0 0 IZI IZ1 [211171 0 In - ~~Sucker S;C.DID 0 0 0.0 0 - - Sauna s.c.0 - 0 010 0 0 - - Last Chance S.C.- -0 0 0 0 - 0 - MEAN CATCH PER CELL RELATIVE ABUNDANCE KEYo0.00 [;;iiI 0,25-2.50 (2]0.01 -0.25 •>2.50 -No sample ~o~..:A ~Bear Bait S.C.-DOD 0 0 0 - - ~[J;Rolly Creek Mouth 0 D·0 0 (;iI I:ZI (;ilIA [;iii .~...!P'"J<roto Slough Head 0 0 0 0 0 0 0 - -r::rr Eagles Nest S.C.- - - - - -0 0 0,f Hooligan S.C.0 0 0 0 0 0 0 -0 ~o - "... , , \ .L.• - Cook Inltl' Figure 11.Seasonal distribution and relative abundance of juvenile I coho salmon on the lower Susitna River.June through mid-October 1984. ,-i~ 10 • C"SWIU'&'C"I:I:~MOUlH .AV[fI o...~,U)UCH i •I •I 3.1I I II ~7 3 I 8 I II ::I 2.S II<0 11 r II W II<0 1 r 2 W %..J ()..a.1i ..J 5 :iE 1.S a. 3 «5 :iE i C/)I ~-<C,C/) 2 I-~0 I- O.S 0 Z 1'77?'Z "0 II'112 7'n II'112 "'2 ''''112 71 72 .,112 .,1I2 '01"-.t SA""'UNO HItIOOS SAIJ"'UWO ""ItIOOS (1- lI'ItCH CItl[I:~!!LOUOH ItOLLY CIOtUK ~IOUlH '0 .. 1 S.li ~II I 3 7 I I ::I II ::I 2.1I 11 lj "'r II "'2 :r r §..~ 5 t •• a i 2 r-T'7A I ~~ 0.11 I f7?Z'fZ?J1ZpI •I i °-----.-- II'112 7t T.l II'112 "12 10' .n"112 71 .,112 II .2 101 SA.."'UNO HItIODS SAM"'UNO HItIOOS Figure 12.Juvenile coho salmon mean catch per cell at four tributary mouths on the lower Susitna River by sampling period,June through mid-October 1984. I I -.1 J J ..1 I ,I I ,~ -. - DRAFT/PAGE 4 3/7 /85,4/15/85 NUM2/Results ,4/28/85 3.2.3 Chum salmon A total of 608 juvenile chum salmon were collected in the lower Susitna River.Only ten of these juvenile chum salmon were captured at oppor- tunistic sites. In early June,chum fry were captured at 13 of the 15 (87%)modelling sites sampled (Figure 13).By late July,chum were only captured at six of the 19(32%)sites sampled.Over 99%of the total catch was made prior to August and no chum salmon fry were captured after August 15. The majority of sites with high CPUE's were located in the reach from Island Side Channel (RM 63.2)to Sucker Side Channel (RM 84.5). Chum fry CPUE's declined steadily from early June to mid-August (Figure 14),reflecting outmigration of juvenile chum salmon from the Susitna system.In a pre-study trip in late May,chum fry were also collected at a number of lower river sites and appeared widely distributed in the river. Juvenile chum CPUE1s were highest in side channels (0.6 fpc)and tributary mouths (0.1 fpc).Slough CPUE's of juvenile chum were extremely low (0.01 fpc),however,sampling effort within sloughs was limited during the period from early June through early July.Tributary mouth densities are unequally distributed by a single site catch of 39 fry at Birch Creek Slough in late June.Chum catches within the side channels were affected greatly by turbidity as peak catches were made in side channels with a turbidity of less than 50 NTU (Figure 15). 1,".,..,-... I I - CHUM SALMON Hooligan S.C.•~0 12]0 DID -10 MEAN CATCH PER CELL REl,ATIVE ABUNDANCE KEYo0.00 [;jjjI 0.25-2.501210.01-0.25 •>2.50 -No sample / C') ~/~(e(~~&\~~~\04 _.'f-i •~'Ii,~6\SAMPLING PERIOD...,,)I JUN JUL AUG SEP OCT:.Au =Si:.:..te:........--r:::I:-r:n::-r-:::I=..=JI:-r.::l=:-..l[::::::,_;::I:::-r:JI~_;::I:::_, ~~Trapper Creek S.C./::iiI 0 0 ~0 DOD 0 ~~"'-Birch SlouQh l:A • 0 0 0 ClOD 0 ,~Sunrise S.C.~-~IZl 0 1 £1 - - - ,Sunset s.c.- - -0 0 0 0 0 - 00 [J Ol[]000 - ~Beaver Dam S.C.IZl Ci 0 0 lZl 0 0 0 -~I ,Sucker S.C..-.•[ZJ 0 0 0 0 - -Cl:'UtI~• \N Sauna S.C.'• -171 0 0 0 0 - - ~Circular S.C.•[;iii lZ IZl 0 0 0 - - ~A Goose 2 S.C.~Gil C lZl DOD - - .~p.Moinstem West Bonk - - -I[J 0 0 0 0 - r~~r:\.:.1.=5I.:::.on::..d-=S.:..:'C=...'----1~.~.!!!!.1_![;;ii~~0~IZI~D:::=-¥.0=-r.D==-+;;-=-f fJ'1fI --Caswell Creek Mouth .0 -DOD 0 0 0 0 .)~[dL 'RustiC Wilderness S.C.l:A 0'[2]0 0 0 0 0 -$Last,Chance S.C.- -£2r 0 0 0 - 0 -~~Bear Bait S.C.l2l f2]ClOD 0[l RollyCreelc Mouth [ZJ 0 0 0 0 D.[ ] 0 0 ~~.......:.K::.ro:.:.t.:..o..:S.:.::lo.:.u9:.:,h:....;H:.:.,:e_o_d__-+==0:::,'~0=-t-=D~0==-l.-=O=+D=-·~C:sI:t=:+::-;" Eagles Nest S.C.- - - - --Il J 0 0 Cook ''''el Figure 13.Seasonal"distribution and relative abundance of juvenile chum salmon on the lower Susitna River,June through mid-October 1984. - $4 ~;:, :I<.> I ...I ~... <.> 0:..... 8. :I (.) ':t <.> z•... :t 61 _5/DE CH4WNE:LS ts::sI TRIBUTAR'l'I.lOUTHS i-='-l....I~-.......I....,IIIIIt'--...,..---,---~--_---., 62 71 72 81 82 91 92 101 SANPUNG PERIOD Figure 14.Juvenile chum salmon mean catch per cell at modelled side channels and tributary mouths on the lower Susitna River by sampling period,June through mid-October 1984. 4-00375325 / / 175 225 275 TU RBIOIT'l'(NTU) 3 6..,-----'---.....:..---------1.-..;..--=-...,.1.------, Figure 15.nile chum salmon mean catch per cell at modelled side ~~~·~ls on the lower Susitna River by turbidity increment. June hrough mid-October 1984. DRAFT/PAGE 5 3/7/85,4/15/85 NUM2/Results,4/28/85 3.2.4 Sockeye salmon A total of 412 juvenile sockeye salmon were captured in the lower Susitna River study reach from early June through early October.Ninety percent (369)of the juvenile sockeye salmon were captured at the habitat model sites.Age 0+sockeye fry comprised 99%of the total season catch at modelled Jft.HS sites.No age 1+sockeye were captured after June.Age 1+sockeye were captured in early June at Hooligan Side Channel,a site which produced no further sockeye juveniles all season, and in 1ate June at Beaver Dam Slough.Sockeye juveni 1es were most widely distributed within modelled sites upstream of Goose 2 Side Channel (Figure 16). Tributary mouths exhibited the greatest densities of juvenile sockeye salmon with a mean catch of 0.7 fpc,and the highest CPUE at Beaver Dam Slough (1.2 fpc).Si de channels had a mean sockeye CPUE of 0.1 fpc. Beaver Dam Side Channel had the highest CPUE for a side channel of 0.7 fpc.Side slough CPUEs of sockeye juveniles were minimal (0.03 fpc). Side channel CPUE's remained fairly constant at low levels through August in comparison to tributary mouth CPUE's which varied greatly (Figure 17).No sockeye juveniles were captured in side channels after August,however,sampling was limited. The greatest mean CPUE for si de channel sockeye fry was found at tur- bidities between 100 and 150 NTU (Figure 18).The numbers of sockeye juveniles captured in Beaver Dam Side Channel,immediately below and """ /I;;-(SOCKEYE SALMON-.0~6-~~\Cl G ':0 I'~-.~, SAMPLING PERIOD~'Ii.~O\-.rJl JUN JUL AUG SEP OCT~Sit.I II I n:I rr I l[I~T"PO"C"ek s.c.I2l [J [2j 0 0 0 0 0 0 .li Birch Slough I2l ~I2l 0 I2l I2l []0 0.~Sunrise S.C.IZI -!Zl fZ]IZI 1ZI ---1 Sunset s.c.---!A [2]r2l 0 0 - Beaver Dam Slough 0 •c;j '[]0 (;.iiI ~[;i -~0 c;j CiiiI CiiiI CiiiI [;iii IZI Dl«,Beaver Dam S.C. ~Sucker S.C.IZI IA 0 [;ii 0 IZ1 0 --~Sauna S.C.[2]-0 0 0 c;j 0 -- ~Circular S.C.~0 0 IZI 0 0 0 --~Goose 2 S.C.IZI 0 0 0 0 0 0 - ~~fI)MainstemWest Bank --0 0 [ZJ []0 - hfl),~:J I"slandS.C.lZl 0 0 '0 0 0 0 0 -If Caswell Creek Mouth 0 -0 [;iii [;I [;I lZl 0 0~Rustic Wilderness S.C.0 0 0 0 0 0 0 0 - -:>~....,Chanoe S.C.--0 0 0 0 -0 ---:> 0 0 0 0 0 00BearBaitS.C.---~.~R.n,C,__••lZl 121 0 [;iiiI 0 l:;;iI ~r:;j [;;jjjI......, rp-'o,-Kroto Slough Head lZl 0 0 0 0 0 0(2]-.Eagles Nest S.C.------0 0 10~ 0 Hooligan S.C.~0 0 0 0 0 0 -0 I>MEAN CATCH PER CELL RELATIVE ABUNDANCE KEY 00.00 Gil 0.25 -2.50 ~u [2]om -0.25 •>2.50 -No sample \ ..A.... Cook Inlet Figure 16.Seasonal distribution and relative abundance of juvenile sockeye salmon on the lower Susitna River t June through mid-October 1984. _SlOE CHANNELS !SSl TRISUTNlY WOUTHS " 2 ..8 2.6 w ~2~4w "0 ::.20 tit =: ~1.8 ~w 1.60 It 1..4w L X 1 ..2 u !Cu 0.8z <C .0.6... :l o,,~ 0.2 0 .1 .2 11 72 81 82 SAWPUNG PERIOO .1 101 - Figure 17.Juvenile sockeye salmon mean catch per cell at side channels and tributary mouths on the lower Susitna River by sampling period,June through mid-October 1984. ------.......ci 325 I --,.-.-..--........ 375 ".CO lSSI WITH BEAVER DAM $.C _LESS BEAVER DAM S.C 175 Z25 275 TURBIDITY (NTU) 125 SIDE CHANNEL SOCKEYE (lolODE:LLED SITES ONLY)---- 75 0.45 c .... 0.35 ~ ~...0.3u Itw 0.25A- x 0 ~0.:2:.. 0 Z <C 0.15... :l 0.1 0.05 0 25 Figure 18.Juvenile sockeye salmon mean catch per cell at modelled side channels on the lower Susitna River by turbidity increment (with and without Beaver Dam Side Channel),June through mid-October 1984.- 5?- DRAFT/PAGE 6 3/7 /85,4/15/85 NUM2/Results,4/28/85 conti guous wi th Beaver Dam Slough,may have been enhanced by site to site movement.With Beaver Dam Side Channel captures excluded,the peak CPUE occurred at turbidities between 50 and 100 NTU. Catches at Beaver Dam Slough and Beaver Dam Side Channel also showed the effects of turbidity as related to cover on the distribution of sockeye juveniles (Figure 19).During late June through August,Beaver Dam Side Channel was breached by the mainstem and turbid,and sockeye CPUE's were high.However,in early June and September,the site was much cl earer and few sockeye juveni 1es were caught in the now covettt poor envi ronment.In Beaver Dam Slough,however,whi ch has abundant aquati c vegetation cover,CPUE's of sockeye juveniles in late August and September were quite high.Catches at Rolly Creek also increased in late August and remained fairly high through early October (Figure 19). 3.3 Habitat Modelling of Rearing Juvenile Salmon Two types of habitat modelling techniques were used to model the re- sponse of juvenile salmon habitat at the study sites to variations in mainstem discharge.The two methods are:(1)the RJHAB model developed in Marshall et al.(1984)and (2)the IFIM hydraulic models discussed by Bovee (1982).Suitability criteria for important microhabitat variables are necessary as inputs to both models and criteria specific to the lower reach of the Susitna River for juvenile chinook,coho,chum,and sockeye salmon have been developed in Appendix A. - BEAVER OAW S~OUOH ..•~h >-.. '"u S0.. I..4~II""!!... ::i cL5 %W........ !(~...Z ~ z -<..III.. :0 ~ 0 Z 0 .,n 1:1 JZ .,n .,IZ 101 IAMPUIlO PIII_ M!l, etjl.VDt Dol ..SlOE CHjl.HNE~ 7 ..•>.. '"~S.. ~ i;l 4 U•.. L 5 CzWu.....~..~..Z 2z-<..III.. :0 ~ 0 Z 0 .,n 71 7Z ".Z II IZ 10' $jl.lofPUIIG P_ ROLLY ClItEIC MOUTH ..>-.. '"u s5l ~4 u •.. L 5 %... !(...Z z.... ::I II IZ 101 Figure 19.Juvenile sockeye salmon mean catch per cell at Beaver Dam Slough, Beaver Dam Side Channel,and Rolly Creek Mouth by sampl ing period, June through mid-October 1984. ,- DRAFT/PAGE 7 3/7/85,4/15/85 NUM2/Results,4/28/85 In the following discussion,results are presented by individual spe- cies.Within the presentations of results for each species,modelling results from selected sites using the RJHAB'or IFIM models are presented,discharge effects upon juvenile salmon habitat for the pooled sites are presented,and models are tested for verification. No results from the Birch Creek.Slough and Eagles Nest Side Channel modelling sites are presented here.At Birch Creek Slough,there was no measurable effect of mainstem discharge upon the site as the mainstem backwater at discharges less than 75,000 cfs did not extend to the site and a blocked culvert at the head of the slough stopped mainstem water from flowing through the site from the head.The Eagles Nest Side Channel site was modelled only twice at mainstem flows of 14,900 and 20~400 cfs and therefore could not be readily extrapolated to discharges of 75,000 cfs.All of the other sites were modelled at three or more discharges and results were extrapolated to discharges over the range of 12,000 to 75,000 cfs.The WUAs and site areas at the RJHAB sites were not adjusted to a reach 1ength of 1,000 ft as were the IF IM WUAs. Lengths of all the RJHAB sites are listed in Appendix Table B-3,so that the WUAs could be adjusted if desired. The instream flow results have been generated only to discharges of 75,000 cfs because it is very difficult to collect data at these discharges.Also,most of the side channel sites have very large flows at 75,000 cfs and are poor habitat for juvenile fish.At higher discharges,the entire flood plain becomes full and the flows are barely DRAFT /PAGE 8 3/7/85,4/15/85 NUM2/Results,4/28/85 constrained within the side channels.Refuge for the juvenile fish at these times presumably incl ude 1arge backwater areas and small si de channels which are very infrequently flooded. At two of the model sites,Island and Trapper Creek side channels,both the RJHAB and IFIM models were run on the same transects.Comparative - resul ts for these two figures presented here two side channels. models are given in Appendix C.The summary incorporatl data from the RJHAB model at these A The ability of the RJHAB models to extrapolate WUA between discharges of 12,000 and 75,000 cfs was evaluated subjectively by rating them from unacceptable to good (Table 6).Some models were rated fair because there were no habitat measurements taken at di scharges just above overtopping of the side channel.Eagle's Nest Side Channel was rated unacceptab 1e because measurements were taken on only two occasions at discharges less than 21,000 cfs. The IFIM models were evaluated according to hydraulic criteria on the basis of excellent to acceptable (Appendix D).Acceptable ranges of the models usually extend to over 60,000 cfs (Table 7).The models were run and WUAs generated at side channel flows which corresponded to discharges ranging to 75,000 cfs,so reliability at these flows is unknown.At discharges below overtopping,the WUAs at a site flow of 5 cfs were used as a minimum,except at Trapper Creek Side Channel where the non_overtopped flow was assumed to be 14 cfs. '\ 5G - Table 6. oRft,FT /PAGE 1 4/15/85,4/29/85 NUM2B/Tab 1e X-I Evaluation of RJHAB model quality for extrapolating WUAs over the range of 12,000 to 75,000 cfs as measured at Sunshine gaging station,1984. Site Number of Habitat Measurements Model Quality - Hooligan Side Channel Eagle Side Channel Kroto Slough Head Rolly Creek Mouth Bear Bait Side Channel Last Chance Side Channel Rustic Wilderness Side Channel Caswell Creek Mouth Island Side Channel Goose 2 Side Channel Sucker Side Channel Beaver Dam Slough Beaver Dam Side Channel Sunrise Side Channel Birch Slough Trapper Creek Side Channel 5, 5 2 4 4 4 5 5 3 5 4 4 4 3 4 3 4 Good Unacceptable Fair Good Fair Fair Good Fai r Good Fair Good Good Good Fair Good Good Table 7. DRAFT/PAGE 1 4/15/85,4/29/85 NUM2BjTable X-2 ~ Discharge ranges of IFrr~models at lower Susitna River sites for which hydraulics are rated acceptable,1984. Data taken from Appendix D. Site ~cceptable Range Island Side Channel 35,000 to 70,000 cfs "1MainstemWestBank18,000 to 48,000 cfs Circular Side Channel 36,000 to 63,000 cfs Sauna Side Channel 44,000 to 63,000 cfs Sunset Side Channel 32,000 to 67,000 cfs Trapper Creek Side Channel 20,000 to 66,000 cfs ~ - 58 - - DRAFT/PAGE 9 3/7 /85,4/15/85 NUM2/Results,4/28/85 Since suitability criteria for chinook salmon juveniles have been developed for both turbid J>30 NTU)and clear «30 NTU)conditions, several assumptions have been made.The tributary mouth sites have been assumed to be clear «30 NTU)at all discharges less that 75,000 cfs. This is not always the case as occasionally turbid mainstem water may back up into the site with a rapid increase in mainstem stage.Also spring runoff or large storms might increase turbidities over 30 NTU. Available data,however,have indicated turbidities are normally less that 30 NTU (Figure 3). At side channel/slough sites,turbidities were assumed to be always greater than 30 NTU when the site was breached and less than 30 NTU when the site was not breached.In early June,.September,and early October, turbidities in side channels were sometimes less than 30 NTU (Figure 3). Many of the model sites were not overtopped during these periods with low discharges.Turbidities in sloughs were usually much less than 30 NTU. 3.3.1 Chinook Salmon Chinook salmon juveniles were captured at all of the study sites with the exception of Last Chance Side Channel (Figure 8).Since chinook juveniles were widely distributed,~ults from all sites modelled with RJHAB and IFIM techniques will be presented. Graphs of the weighted usable area responses to mainstem discharges for all sites not presented here are included in Appendix B.Appendix B 59 DRAFT/PAGE 10 3/7/85,4/15/85 NUM2/Results,4/28/85 also contains the tabulated values of weighted usable areas at 3000 cfs increments as digitized from these graphs (including sjte graphs pre-- sented here).Also tabulated are habitat indices which were calculated by dividing the weighted usable area at a given discharge by the site area at the same discharge. At the Rolly Creek,Caswell Creek,and Beaver Dam Slough tributary mouth sites,the responses of weighted usable area to mainstem discharge were very similar.The Rolly Creek mouth weighted usable area response to discharge is presented here as an example (Figure 20).The great increase in weighted usable area with discharge is due to the effect of mainstem backwater causing large increases in area,depth,and amount of cover. At side channel/slough sites,the responses of weighted usable areas to mai nstem di scharge vari ed somewhat.Normally,the wei ghted usable area increased greatly after overtopping and then decreased wi th further increases in mainstem discharge as at Kroto Slough Head (Figure 20). The increase ;n weighted usable area right after overtopping is due to increases in area and also increases in cover suitability as the tur- bidi ty provides cover in otherwi se cover poor habitat.As di scha rge increases along with site flow,velocities initially become more suitable,but then as site flows increase,velocities became unsuitable and theWUA decreases. At Sucker Si de Channel,backwater effects buffer the vel oci ti es from becoming too high and so weighted usable area gradually increases after - - - CHINOOK WUA ,"'_O_L_L_y_C_"'_E_E_K_M_o_u_n/., .-'.~ ~O _JTh"':!:!Gn!!·t 30 +-----..---,...-----",...---,---.-,----.,---- 70 K"'OTO SLOUGH HE:AO-------------------·----------'1 SUCKE'"SIDE CHANN£:L 2 • '?'Mdted t ............ 7030SO (Thou»ond»} loIAlNSTEloil OISC'HARGE:AT $UNSHINE:(cia) o -----~---J.....,..---..._---r_-- 10 - Figure 20.Weighted usable area for juvenile chinook salmon at the Rolly Creek Mouth,Kroto Slough Head.and Sucker Side Channel study sites as a function of mainstem discharge.1984. - DRAFT/PAGE 11 3/7 /85,4/15/85 NUM2/Results,4/28/85 overtopping (Figure 20).At 70,000 cfs,WUA's begin to decline at this site,however,as velocities and depths become unsuitable.At other sites,WUA held quite constant after overtopping or slowly increased (see Appendix B). When WUA 's from three tributary mouths are pooled there is no large change in WUA until approximately 45,000 cfs when the WUA increases greatly with discharge (Figure 21).By dividing the WUA at 3000 cfs increments by pooled area for the three sites and plotting the habitat index,it becomes apparent that the change in WUA is not simply due to increases in site area.Increases in habitat indices are due to increases in the amount of instream cover,more suitable velocities,and deeper water which may also provide cover. When WUA's from the modelled side channels/sloughs are pooled,WUA's _ increase greatly to approximately 40,000 cfs and then very gradually decline (Figure 22).Habitat indices for the pooled side channels show a similar rise to a peak at 40,000 cfs but then a rapid decrease to approximately 60,000 cfs when the habitat index levels off.The rela- tively more rapid decrease in the habitat index is due primarily to ~ velocities and depths becoming very unsuitable at the higher discharges. ~ '1- Turbidity has been shown to be an important determinant of chinook distribution (Figure 10)and varies from east to west downstream from the Chulitna and Talkeetna river confluences (Figure 4).In formulating the pooled side channel/slough response of juvenile salmon habitat,it TR IB UT.i\r~:'·{~\/1 ()UTH ~:, CHINOOK SALMON 45 --"' u- 30 50 {Thousands) MAINSTEM DISC'HARGE AT SUNSHINE (cfs) ",- )1 ";121 )<1/ /l f 1,l1't / _w /I/I-a-~-el I _~~,Ii13-----E1-'---IO~-fO~-fO--f'D+---e--+----..-----,-----,------,-------.--------I--------------j 70 ~-,.. :=4-0 0- ~') 35 <L1.0.1,-. ::l:..::2«'-'30 - w Ci ...J '., ill .J«c·25 -V)_c -,I--"'-' Cl 20w l- I. e-'w 15 -s- 10 - 5 10 ---------------------------------------~rT-----u-- P j-l/ / yf / Ill' Figure 21.Weighted usable area and habitat indices for juvenile chinook salmon at tributary mouth study sites as a function of mainstem discharge, 1984, I 1 - 30 MAINSTEM I I j -r-----r------------r-----------------j 50 70 (Thousands) DISCHARGE AT SUNSHINE (cfs) (~IDE.~ I I I I I I i -,---------~ :50 50 70 (Thousandsi MAINSTEM Dt·SCHARGE AT :;>UNSHfNE (cfs) l3--G-_r /--B~--s-- / -Q~,s~ 's-_ "H--g.-n··-fl.f ---tJ / I f j CHINOOK SALMON ---------------------------.-------------------------- l70- 60 - - v 50"~ « uV-",u:.u.4.0<{~ w "..J ",. lDJ«0 .30VIc.::- ~!.:.. a w t-201:- {') w 3: 10 13 0 0 1 0 0.07 ----------------------------------.----------.-- ~wz. ill -4: T 0.06 0.05 0.04 .. 0.03 0.02 .. 10 ------,-------,------- 30 ~,o 70 ..(Thousands) MAINSTEM DISCHARGE AT SUNSHINE (ch) Figure 22.Weighted usabl e area and habitat ind ices for juvenil e ch inook sal roon at side channel/slough study sites as a function of mainstem discharge. 1984. - DRAFT/PAGE 12 3/7/85,4/15/85 NUM2/Results,4/28/85 may be desirable to weight JjU~.tQ.!:Lsuch as turbidity which vary from site to site.t-J v c-J 0 .J.1~·'~\riJ,.".v~-4.. Although turbidity data for the model sites are 1 imited,an average turbidity for the side channels modelled was calculated in Appendix Table B-1.A prel iminary suitabil ity index for turbidity can al so be fit to the data in Figure 10 (Table 8).When these data are tied together we can weight WUA estimates for the sites differently (Table 9). When the WUA estimates for each site are adjusted by these factors and then WUA's are again totalled,the WUA and habitat index response adjusted for turbidity for the side channels combined can again be examined (Figure 23).There is very little change from the previous unadjusted graph in the shape of theWUA response curve.Similarly,the shape of the habitat index responses curve has al so been changed very little by these adjustments. The mean seasonal chinook salmon habitat index for the 15 side channels and four tributary mouths were calculated and compared with mean chinook catch (Figure 24).The positive relationship was statistically signif- icant (p<O.OOl)but not very strong.Most of the correlation was due to the large catch (5.16 fpc)and habitat index (0.19)at Caswell Creek mouth.Another outlier is Beaver Dam Slough with a habitat index of 0.17 and a mean catch of 0.17 chinook per cell. Table 8. Table 9. DRJ\FT/PAGE 1 4/11/85,4/29/85 NUM2/Table E-1 Preliminary juvenile chinook salmon turbidity criteria derived from lower Susitna River distribution data,1984. Turbidity (NTU)Suitabil ity <200 1.00 201 -250 0.65 251 -300 0.55 301 -350 0.40 >350 0.15 Weighting factors for turbidity by site for analysis of juvenile chinook salmon habitat use,1984. Site Hooligan Side Channel Kroto Slough Head Bear Bait Side Channel Last Chance Side Channel Rustic Wilderness Side Channel Island Side Channel Mainstem West Bank Goose 2 Side Channel Circular Side Channel Sauna Side Channel Sucker Side Channel Beaver Dam Side Channel Sunset Side Channel Sunrise Side Channel Trapper Creek Side Channel Mean Turbidity (NTU) 377 388 254 365 118 215 279 194 241 266 140 139 152 121 499 Turbidity ~Jei ght i ng Factor 0.15 0.15 0.55 0.15 1.00 0.65 0.55 1.00 0.65 0.55 1.00 1.00 1.00 1.00 0.15 - ':::~I [·····)1 I=- ~--'--~- 40 ..-------..-.-.. ADJUSTED CHINOOK WUA ----_._-_._-------_._-------._.... ',., ,' - ,"';:'" .- 3~,.. -'- 30 .. -.";;. c.{25 -:.c; It.~, "-\T.:i w ,:} 20,>'-a..::;; «;-,) I'.c:::::).r- 1 5w tJJ'..- l c:~1 0 W ~ 5 - o -·--·------·r---·---·---r--------·T-·----·-··------l-..---'------'---T---'.-.' 10 30 50 70 (Th.)usands) MAJNSTEM DISCHARGE AT SUNSHiNE (d~) i~, - Co ., ~ <l I 0.036 .0034 - 0.032 0.03 -- (].026 -- O.Q·24· 0_022 0.02 0_0 18 -. 0016 0.01 d. J.Ol C .008 - 0.006 v 004' 1 0 r.. I I i llL ,/ I 'u I-"-B'-"~'{ Il3-B--~ r----,------1'------,-------·-1-----..-._..; 30 ,5c.?7 C lThQusandsl MAINSTEM DISc:'HAPGE AT SUNSH INC (d") , i Figure 23.Adjusted weighted usable area and habitat indices for juvenile chinook salmon at side channel/slough study sites as a function of mainstem di scha rge,1984. bl C H !I,j ()(J f<,h/1 DEL V [F:I F':(~--'~,.D,T I C)["1 (SIDE CHANNELS AND TRI8UTARY MOUTHS)6 ...._._.____.---.--..---------------.----.--------- -I - - 5 _. ..J ....l W 4 -1c.. i:c. W 0.. I .3 .. 0 ~ <f L' Z«w :E y =0.15 +15.04x p 0.001 r 2 =0.39 r----1 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 SEASONAL MEAN HAS!TAT INDE:< .:£k " Fi gu re 24.Juvenil e chi nook sa 1man mean catch per ce 11 versu s seasonal mean habitat indices at side channel and tributary mouth modelling sites on the lower Susitna River,1984 . DRAFT/PAGE 13 3/7 /85,4/15/85 NUM2/Results,4/28/85 3.3.2 Coho salmon - ..... l"- I Since coho salmon were captured ~Onlyat the tributary mouth sites,only results from these sites will be presented here.In Appen- dix B,values of WUAls and habitat indices at 3000 cfs increments for these areas are presented. The response of WUA to mainstem discharge at the three tributary mouths varied (Figure 25).At Caswell Creek mouth,WUA increased as discharge increased.This increase was due to increases in area and the amount of preferred cover.At Rolly Creek mouth,the WUA fi rst decreased with discharge and then began to increase greatly.The initial decrease was due to the forma ti on of zero vel DCity backwater from a free fl owi ng state without major increases in cover or area,but then the WUA increased due to increases in area and usable cover.At Beaver Dam Slough,these effects of back water formation and increases in cover offset one another so that there was little change in WUA with discharge. When the WUAls from all three sites are summed (Figure 26),there is little change in WUA until approximately 50,000 cfs when the WUA begins to increase greatly with discharge.When the effect of change in area is taken out by cal cul ati ng a habi tat index J site quality decreases initially as the backwater is formed and then begins to increases as cover increases due to the backwater. ?0OJ 4 Figure 25.Weighted usable area for juvenile coho salmon at the Caswell Creek.Rolly Creek and Beaver Dam Slough tributary study sites as a function of mainstem discharge.1984. 10 - TR 18 U Tl\.F~-Y ~·/1 C:>UT H S (BIRCH CRE:EK 8LOUGH EXCLUDED i l I i I I \ I I I .D 70 I -,-------r----I /~/ I I r/ / I 11 -j (~--€---G-e-H-fr-B-_e----s.-../ 1-S-'E!--_S../ri 10 -t-------,I T----,-r-- 10 50 50 (Thousands) MAINSTEM DISc'HARGE AT SUNSHINE (cfs) 14 13 1 Z .. 1 ~I 17 16 .. 19 .. 18 .. 20 .. ow l- I Q W ~ <[ w"'-'"a:.If) <[~ wO -I", W:J <[0v..::=.:J.!:.- p=s=il: 705050 (Thousands) MAINSTe:M DISC'HARGE AT SUNSHINE (cfs) 0.1 G-O 0 0.09 - 0.08 0.07 :- w 0.06 '0 Z ~0.05« !::: I1l 0.04 ..<[ I 0.03 .. 0.02 .. 0.01 0 10 - .- .~ Figure 26.Weighted usable area and habitat indices for juvenile coho salmon at tributary mouth study sites excluding Birch Creek Slough as a function of mainstem discharge.1984. '7f, DRAFT/PAGE 14 3/7 /85,4/15/85 NUM2/Resu1ts,4/28/85 The mean habitat index for the season (May 15 to October 15)was cal- cu1 ated for the four tributary mouths.Si nce Bi r.ch Creek Slough was a IlIII'!!l\ natal area,only catches from mid-July through mid-October were used in calculating the mean site catch.The mean catch per cell of coho ~uveni1es in~reased with the mean habitat index but a linear regression was not statistically significant at the 0.05 level (Figure 27).None of the side channels had mean seasonal habitat indices greater than 0.05 and most were 0.03 or less,primarily due to the lack of suitable cover types. 3.3.3 Chum salmon Chum salmon were widely distributed at all of the side channel sites sampled duri ngearly June through July 15 (Figure 13 ).Therefore, graphs of the WUA response as a function of mainstem discharge for all the side channell slough sites not presented here are included in Appendix B.Also tabulated in Appendix B are values of WUAls and habitat indices at 3000 cfs increments as digitized from the graphs. Responses of WUA's at the sites to increases in mainstem discharge were variable.At Rustic Wilderness Side Channel,WUA greatly increased after overtopping and then declined with further increases in discharge as velocities and depths became unsuitable (Figure 28).At other sites such as Last Chance Side Channel,the increase in WUA after overtopping was much less great while at Trapper Creek Side Channel (Figure 29), WUA's decreased after overtopping.At Sunset Side Channel,WUA r-'!.., IL - - - -" - - -'....J W..... CASWELL [J 3- HABITAT IND£;< BEAVER DAM o ROLLY oBIRCH o I :-j I 1 _I Io-r---.T-·~_,_~,_-_._--,___.-__r-_,__-_r_-.,..I --------r~-.----r_,_-··-I C 0.02 0.04 0.06 0.08 O.i 0.1 Z C.;4- ..... Figure 27.Juvenile coho salmon mean catch per cell versus seasonal mean habitat indices at tributary mouth modelling sites on the lower Susitna River,1984. ;~ CHUM WUA RUSTIC:WILDERNESS SIDE CHANNEL 34 32 Breached t30 ..--.28-260- '"Z4'-'...Z2.w- ct.'"...":0c: ...0 18 -oJ'"ID~ ...0 16If)..L: ::::>1-'-'14a !oJ 12l- I. e>10... ~8 6 4 2 10 30 50 (Thauacmds) 70 - LAST CHANCE SIDE CHANNEL. ..--.1 B 20 - - I I J Ir-·----~ 90505070 {Thousands) "AI NSTEt.4 DISC'HAElGE AT SUNSH IN E (cis) Breached t 6 - 10 ,4- 10 - 12 16 - a... l- I. e> lY ~ CT '"--- Figure 28.Weighted usable area for juvenile chum salmon at the Rustic Wilderness and Last Chance Side Channel study sites as a function of mainstem discharge,1984. - -. CHUM \NUI-':... TRAPPER CREEK SIDE CHAN NEL 50 -- 45 - -40 -Bre.ched ~ ;or 'n .35 ........,- Q:.JI..~30 ... ...5 ..J:J) G1j ..0 :5iJ'~",l- e.20... ~ I e' W 15 ... ;; 10 •. S 10 30 50 70 (ThQu'Qnds) .- «...v-- Q:.'".."='...5 ..J:J) Wj«0 11).c.Jt', o... l- Io... it 52 ... 50 .. 48 - 46 44 - 42 .. 40 .. 38 34 30 .. 28 .. SUNSET SIDE CHANNEL .."-----'------l ! 26 -.....------r----~--.-~-.-.---__r.-----_r---..-....r ....-·--_.... 10 30 50 70 (Thousands) MAINSTEM OISC'HARGE AT SUNSHINE (cf,,) Figure 29.Weighted usable area for juvenile chum salmon at the Trapper Creek and Sunset Side Channel study sites as a function of mainstem discharge,1984. DRAFT/PAGE 15 3/7 /85,4/15/85 NUM2jResul ts,4/28/85 -increased after overtopping until about 53,000 cfs when WUA quickly declined.The other sites also showed variations of these response -.. curves (see Appendix B figures). When WUA'S from all the modelled side channel/slough sites are pooled, the peak in WUA's for the sites occurs at a discharge of 40,000 to 52,000 cfs (Figure 30).Above this discharge range,WUA's decrease rapidly due to unsuitable high velocities and deep depths.,Habitat indices for the same pooled sites are constant through about 24,000 cfs and then decrease steadily. Chum salmon use of side channels is affected by turbidity (Figure 15) and since turbidity varies from site to site,WUA's for each site should be adjusted for turbidity.Since chum salmon outmigration is mostly completed by July 15,turbidity data contained in Appendix Table B-1 through July 15 were examined.Since turbidities greater than 200 NTU appear to affect use greatly (Figure 15),site WUA1s were adjusted for periods when the turbidity exceeded 200 NTU.Adjustment factors for the sites ranged from 0.50 to 1.0 (Table 10). When the chum salmon WUA's were adjusted for turbidity and again totalled,very little changes were noted in the shape of the WUA or habitat index response curves although of course both WUAI S and habitat indices decreased (Figure 31). Mean chum salmon adjusted habitat indices were calculated for the period from May 15 through July 15 and compared with mean chum catch during the - - (I--J i\r\I r''I l -L (~. .,j I 0....."L..'_,.I ."1 ..(,)I 41 0 '-·--....------. CHUM SALMON '~. 400 - 390 -'380~. r-.-370 '''-:' -.;,360 w"'-"U'c ,1) 350 -<.(~ W ::j. 340.-":,<", '.L:;,.. 330</) .... -.'t- ,.:.320 - ~.'J 1.3 1 0 c> i.&.!300~~ 290 - 280 - \', \ £r:;l '; \" '~1---r----.-----,-..---.,..,--,. 30 50 70 (Thousands) MAINSTEM DISC\iARGE AT SUNSHINE (ch) 270 .------r..----r-- 10 07 .-_.~...~---..---------...-.------.,-----:.....---.--,..~~_.-.---._-'.--..-.;.......~-~·__···__r·._···_..._ - ,: 0,6 .. 0.3 n.2 fl. t.J o -.----..,-----T-'T-·----..-T-......·....'--.'.i- 10 30 50 70 l'Thou-sands) MAINSTEM DISC'HARGE AT SUNSHINE (crs) Figure 30.Weighted usable area and habitat indices for juvenile chum salmon at side channel/slough study sites as a function of mainstem discharge,1984. DRAFT/PAGE 1 4/11/85,4/29/85 NUM2/Table G-1 Table 10.Heighting factors for turbidity by site for analysis of juvenile chum salmon habitat use. ] J Site Hooligan Side Channel Kroto Slough Head Bear Bait Side Channel Last Chance Side Channel Rustic Wilderness Side Channel Island Side Channel Mainstem West Bank Goose 2 Side Channel Circular Side Channel Sauna Side Channel Sucker Side Channel Beaver Dam Side Channel Sunset Side Channel Sunrise Side Channel Trapper Creek Side Channel Sampling Period When Turbi dity Exceeds 200 NTU June 16-30 June 16-30 June 16-30 June 16-30 July 16-30 July 1-15 June 16-30 July 1-15 July 1-15 June 16-30 July 1-15 July 1-15 July 1...;15 July 1-15 June 16-30 Turbi di ty Weighting Factor 0.50 0.50 0.50 0.50 1.00 0.75 0.50 0.75 0.75 0.50 0.75 0.75 0.75 0.75 0.50 r - - '.! ADJUSTED CHUM :SALMON WUA 270 250 260 -- ~, ;:0:. l.1J-'-.- '~.~ ...;...:-'_....:) ..L.::: If}..:.:::. .',.- 240 230 220 - 210 - 200- 190 - 180 .. 170 . 1 0 -,--r------l------r---·---,----··-----r-..-----'-,-..,r··· 30 50 70 (Thousands) MAINSTEM DISCHARGE AT SUNSHINE (ds) ,- .-. 0,,42 - OA _. 0.38 0.36 -- 0.34 {).32 -, )(0.3 .- w 0.28c - J.," 0.26 - 0].'0.24 - -- ~.L 0.22 -..~ f 0.2 O.1 8 o~1 6 - O.1 4- O.1 2 .- 0.1 - 0.08 - .......... N,,_ ....[1 --·---.,---------------,-----·-·-----T----·--r---·-----r-------------- 1.0 30 50 (Thousands) lYiAINSTEM DISCHARGE AT SUNSHINE «;[3) 70 Figure 31.Adjusted weighted usable area and habitat indices for juvenile chum salmon at side channel/slough study sites as a function of mainstem discharge,1984. ...., DRAFT /PAGE 16 3/7/85,4/15/85 NUM2/Results,4/28/85 -same time period (Figure 32).There was no sampling effort at two of the side channels,Mainstem West Bank and Sunset Side Channel,during ~ this time so they are not included in this graph.There was no significant (p;>O.05)correlation between the seasonal habitat index and chum catch although the correlation (0.53)did suggest a relationship. 3.3.4 Sockeye Salmon Sockeye salmon were most numerous at the tributary mouth sites with most side,channels having some use (Figure 16).Presented here or in Appendix B are graphs of the WUA responses to discharge of the three tributary mouths and the four side channels (Beaver Dam,Sucker,Sunrise ~ and Sunset)which were found to have sockeye salmon present more than half the times sampled. The typical response of WUj\.at the tributary mouths to increases in discharge was a steady increase as shown here by the modelling results .... from Rolly Creek (Figure 33).The WUA increased as the backwater zone increased because sockeye find zero velocity water most suitable and because site area and cover also increased greatly with discharge.The vJUA response at Sucker Side Channel was similar to that of the tributary mouths as ~~UA generally increased with discharge after overtopping. This site is influenced greatly by backwater effects from the side channel at its mouth.At Beaver Dam and Sunri se Si de Channel s,WUA increased after overtopping and then declined somewhat (Figure 34).At Sunset Side Channel,WUA fluctuated up and down with discharge in no real pattern (Figure 35). C:H I,J ~·./1 f\~()DEL \/E F~:I F--I C,D.,TI (J I',J (SIDE CHANNELS/SLOUGHS ONLY) ,..... .4..5 4--iJ 0 3.5 _. 0 J I.J LoJ 3 ..0<-> Il:. U.I ll..::.5 ,. :...; I U ~.., ..:(.... Co Z 0 <{,.5 -......:a C] rtr I 0.3 C ..>'-1-0.260.18 j I, I.......--'-o'---'--,I--,----r--..,----r-,.------..·-, 0.220.14 0.5 0 0 n 00wI 0.06 0.1 SEASONAL MEAN HA81TAT INDEl Figure 32.Juvenile chum salmon mean catch per cell versus seasonal mean habitat indices at side channel and slough modelling sites on the lower Susitna River,1984. 6'1 SOCKEYE WUA, ROLLY CREEl<MOUTH ----- 30 50 70 (Thousands) WAINSTEt.l DISCH~RGE AT ~UNSHINE (cts) -..--------,----,----,...--_._,---..,..-- 1 10 100 -90--0-80..---c 70w- 0:."«'0 60c:we ...JOlt m::J 50CO li)L; :>........,. -1-0 -0w.... :I:.30C> W ~20 10 ------- 0 ----- 10 SUCKER SIDE CHANNEL 30 _50 70 (Thousands) MAINSTEM DISCHARGE AT ~UNSHINE (cfs) I I 4----r------<---.------.-----.-----r---.----J 6 5- 0- "---4 C................ 0:.'It c'Oc:wO 3...JUt m~ CO V)L; :>........,. 0 2 -w.... .I C> W ~1 - 0 10 Figure 33.Weighted usable area for juvenile sockeye salmon at the Rolly Creek Mouth and Sucker Side Channel study sites as a function of mainstem discharge,1984. so C r<E·(E \1\,/)J/~\ SEAVER DAM SiDE CHANN EL 4.5 - 4- " ":5 - - - - - --- - - - - --- - - --s .-J 3.5 Breached + 1 - 2 1.5 2.5 1f!\!"I'I'!:\.,-.. :;: I'T (/I '-' « I.U'"""o::.'h«'t.1c-.....0 -I '"m :I'«0 If'.J::':~C· 0..... l- I 0 ..... ~ 0.5 - ,----------- 70 --~-- 50 (ThQusands) .----r----..,--. 30 o - 10 SUNRISE SIDE CHANNEL 70 I ----l':r------..:;- --- - --i Breached t -----,------r-----_-----~ 30 50 (Th<J u sands) MAINSTEM DISCHARGE AT SUNSHINE (cfs) 7 6 '"""-- t7 5 - (/I -...; « I.U"'--' 0::(/1 4«v... wO -101 1II :::l«0 ;n.J::3 - -.t- -~..../ ar..., to-">1 .. c· w ~ 1 - 0 - 1 0 - Figure 34.Weighted usable area for juvenile sockeye salmon at the Beaver Dam and Sunrise Side Channel study sites as a function of mainstem discharge, 1984. - e-o "-'I-'-E-'/E \11'1 J '\,::.'J \j r\-I:.,I J ~1-, SUNSET SIDE CHANNEL 10 -r-------------------------------------.- 9 <l:....,,-..o::.fh«"g ....0 ...J(/) III:J-«0 1/).J::. Jr- '-" o....r- Io.... ~ 8 7 6 Breached t 703050 (ThQusands) MAINSTEM DISC'HARGE AT SUNSHINE Cds) 5 -1-----.-------,------r-----,-------,.---------rj-----. 10 Figure 35.Weighted usable area for juvenile sockeye salmon at the Sunset Side Channel study site as a function of mainstem discharge,1984. DRAFT/PAGE 17 3/7/85,4/15/85 NUM2/Results,4/28/85 At the combined tributary mouth sites,both WUA and habitat indices increased above discharges of approximately 30,000 cfs (Figure 36).At the pooled side channel/sloughs,on the other hand,WUA's also increased after approximately 30,000 cfs while habitat indices generally declined from the peak at 12,000 to 24,000 cfs (Figure 37).The d·ecrease in the habitat index is due to the steadily increasing velocities in the side channels with increases in flow.No adjustments in turbidity are necessary for the four side channel/slough sites as these have very similar turbidity regimes,being located on the same general location on the river.Use of many of the other side channels is probably limited by turbidity. The mean seasonal habitat index for sockeye salmon at the four tributary mouths and four side channel sites was calculated for the period from May 15 to October 15,1984.The mean catch of sockeye salmon juveniles was positively related to the mean habitat index (Figure 38).High turbidities and velocities within the other side channels presumably limited use by sockeye salmon juveniles. ------------~----........----------------------- T RIB U T.A.R'{~/1 ()UT HS (BIRCH SLOUGH EXCLUDED) 1 30 ---.--.-------------------.-.---------.-.. 120 110--100 90 «w...-...a:.f}) ~--g we; .-J 111 11I:J «0 ii'.s:J.!:.. a lJJ ~ I (~) 80 70 - 60 50 40 30 20 I /;;1 13'o~o -EDl-~-"/ 1a -l----~---__,----.,__~--_.._--'---__,_---_-.-.,.-----.-- 10 30 50 70 (Thousands) MAINSTEM DISCHARGE AT SUNSHINE (cfs) -I--l'I=~=li:l=rEF::::B::=:::!2-r__---~---_._-----r___-·--r---- 30 50 70 (Thousands) MAINSTEM DISCHARGE AT SUNSHINE (cfs) 0.42 - 0.4 - 0.38 0.36 0.34 w 0.32. a...0.3 ~ ~0.28!::: III ~0.26T 0.24 O~22 0.2 0.18 0.16 10 Figure 36.Weighted usable area and habitat indices for juvenile sockeye salmon at tributary mouth study sites on the lower Susitna River as a function of mainstem discharge.1984. i:i\\,---------------------------------------------------'" SIDE CHA~JI\',IELS /SL_C'I,J (ONLY FOUR INCLUDED) _I-I ...,--------------------------------------------- -r---------,-- .30 50 70 (Thousands) MAINSTEM DISCHARGE AT SUNSHtNE (cfs) 20 19 ,-, 18- 0-17<II '-'«......-...16IXr{)«-g wO 15..J.f> w::J ~c- if'J::,I-14.-....- a 1LI t-13r c· w 12~ 1.1 10 10 02-r------------'--~-------- 70 n--H-FJ .30 50 (Thousands) MAINSTEM DISCHARGE AT SUNSHINE (cfs) 0.19 0.18 0.17 0.16 0.15 0.14- 0.13 0.12 0.11 0.1 0.09 0;08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 o +------,----,------...----.-----.-----------"-_.---------- 10 ,. waz Figure 37.Weighted usable area and habitat indices for juvenile sockeye salmon at side channel and slough study sites on the lower Susitna River as a function of mainstem discharge.1984. SC)C ~<EYE ~/1 ()O EL \/ER:1 FI C)\TI ()I'~ (SIDE CHAN NELS AND TRI8 MOUTHS) o +------.,------,-------,-- 1 .3 ,... ~-~ .1 ..J J 0.9 -..... 0 Q:0.8 I.&.t Q..0.7 - I 0 0.6I- 0:{ "() ()5z 0:{.....n 4-~ 0.3 - 0.2 o y =0.10 +2.21x P =0.001 r 2 0 =0.78 o o 0.2 I 1 I -_..---_---j 0.4 SEASONAL MEAN HABITAT INDr;::< Figure 38.Juvenile sockeye salmon mean catch per cell versus seasonal mean habitat indices at side channel and tributary mouthmodel1ing sites on the lower Susitna River,1984. - - I~ DRAFT/PAGE 1 3/7 /85,4/15/85 NUM2/Discussion,4/28/85 4.0 DISCUSSION 4.1 Chinook Salmon Chinook salmon were widely distributed throughout tributary mouths and side channels of the lower Susitna River.Densities of juvenile chinook were highest within tributary mouths.This distribution of chinook fry substantiates earlier observations (ADF&G 1981c)that densities of chinook are generally highest at tributary mouths.Middle Susitna River data from 1983 also showed the highest densities within tributaries (Dugan et al.1983).Caswell Creek mouth had the highest CPUE of juvenile chinook salmon and appears to be a major rearing or holding area. Chinook salmon juveniles used side channels for rearing in both the middle and lower Susitna River after moving from the tributary natal areas.The redistribution of chinook fry from natal areas to lower density rearing areas has been observed in the Deshka River (Delaney et ale 1981)and Montana Creek (Riis and Freise 1978).This'phenomenon reflects a downstream movement or dispersal of the 0+age fish (ADF&G 1981c).Most of the 1+chinook juveniles have outmigrated by August 1. Use of tributary mouths is limited by the amount of instream cover and suitable velocities.Also depth may be important to chinook juveniles in tributaries because it probably provides cover in slightly turbid water (10 to 20 NTU)(Appendix A).At Caswell Creek mouth,catches of juveni 1e chi nook were low in September as the rna instem water stage DRAFT/PAGE 2 3/7 /85,4/15/85 NUM2/Discussion,4/28/85 "'"dropped and the site became much shallower,with higher velocities,and greatly reduced cover. Use of side channels by chinook juveniles for rearing is widespread; however,use is limited by turbidity.Side channels located in the Talkeetna River plume had much higher use than those located in the more turbi d Chul i tna River pl ume or those located further downstream where the water of these two tributaries are mixed.Side channel catch rates of juvenile chinook (in similar habitat)in the middle Susitna River in 1983 were approximately four times higher than those in the lower river in 1984 (Dugan et ale 1984). Since lower __Ei\,l~r'__~j<!~channels are used less by ctlirtoo_k thanmiddle ....--..---.----- Susitna River side channels,it is not surprising thatsloughsarealso f""'-~---"---~~----_.--"-"~'-"~.-"~~~.""-~»"" used less in the lower reach.As water levels decreased in the fall and side channel heads dewatered,there were very few chinook fry at slough sites in the lower river to take advantage of the lowered turbidity. Also the side sloughs are normally very cover poor.Upwelling in Instream flow effects upon juvenile chinook salmon are related to backwater effects at the tributary mouths and to breaching and side channel flows as well as backwater at the side channel/slough sites. When a side slough is not overtopped by the mainstem,cover and access into the site are usually poor . • .- - .- DRAFT/PAGE 3 3/7/85,4/15/85 NUM2/Discussion,4/28/85 At tributa ry mouths,backwater effects increase chinook use si gnifi- cantly because of increases in instream cover and d~th and decreases in water velocity.Also turbid backwater from the mainstem sometimes intrudes into the sites with rapid rises in mainstem stage.Pooled data from three tributary mouths showed major increases in v.fUA at mainstem discharges greater than 45,000 cfs. If the study sites would have been chosen further upstream in the tributary mouths,WUAs wou1<f have begun to increase at a higher dis- charge so the 45,000 cfs figure is not absolute.At Birch Creek Slough, for example,there were no measurable effects of backwater to mainstem discharges of 72,000 cfs.In general,increases in mainstem discharge increase the amount of juvenil echi nook salmon habitat at tributary mouths.Also,these backwaters may increase access into tributaries where rearing could occur by decreasing water velocities at the mouth. Within side channel/slough sites,mainstem discharge is very important. When sloughs are breached the water becomes turbid and provides cover for the chi nook juveniles in otherwi se cover-poor habitat.Turbi di ty, however,may also limit use of the side channels by being too high (Figure 10).Turbidity varies seasonally.High turbidities generally occur from mid-June through September (especially during high dis- charges),and turbidities are low during the rest of the year. Turbidity also varies spatially within the river.Chul itna and Talkeetna river plume effects extend at least 20 miles downriver (Figure turbidity and higher juvenile chinook salmon use. 4).Sites located within the Talkeetna River plume have much lower ~~~ T !V\~~, ~Mainstem discharge DRAFT/PAGE 4 3/7 /8,5,4/15/85 NU~2/Discussion,4/28/8~~ initially increases chinook WUA .within a side chan- nel/slough after it overtops but with further incre~ses in flow,WUA usually remains constant or declines while the'pr()p·brtion of usable chinook habitat usually slowly decl ines .'as flows increase.The RJHAB model shows a decline in WUA with increasing discharge which is greater than that shown by the IFIM model (Appendix C). The results obtained by poolil1gWU1L.ffom all modelled sites should not /----_._--------~-~,_._---_.~".~-,-.-',._---"._._--_.-..~--- be directly extrapolated to represent the entire lower reach.The modelled side channels represented a wide range of sizes and shapes of channels with diverse breaching flows,but they are only a small frac- - tion of the side channels present within the lower river.The most=------_.- Juvenile coho salmon occurrence in the lower river was almost exclu- sively within tributary mouths.Tributaries and tributary mouths werer------_ 92.. 1:""'1 _ I-' DRAFT /PAGE 5 3/7/85,4/15/85 NUM2/Discussion,4/28/85 The heavy use of tributary mouths by juvenile coho may be due to their tendencies to favor waters with relatively low levels of """"=-, Sigler et al.(1984),for example,found that a larger juveni 1e coho salmon emi grated from experimental laboratory channels with turbidities of 25-50 NTU than from clear water channels. In another laboratory study,Bisson and B"ilby (1982)established that coho salmon avoided turbidities exceeding 70NTU.Turbidities in Susitna River side channels during June through August often greatly exceed 100 NTU. Since juvenile coho salmon are dependent upon visibility and background contrast in their food selection (Mundie 1969),turbidity may affect feeding.Juvenile coho salmon feeding effectiveness may be impaired by turbidity levels of 70-100 NTU (Alabaster 1972).Noggle (1978)found that juvenil e coho predation on caddi s 1a rvae decreases to zero at a sediment load of approximately 300 mg/l. The four tributary mouths sampled exhibited marked differences from one another in relative abundance and seasonal use.Rolly Creek and Beaver Dam Slough CPUE's,generally increased from early summer into late fall (Figure 12).This occurrence may be due to both the immigration of coho juveniles and a decrease in site area which increased coho densities. The area of Rolly Creek was reduced by approximately 63%from late June and July to September and early October,while the area of Beaver Dam Slough was reduced by approximately 33%during the same time period. DRAFT /PAGE 6 3/7/85,4/15/85 NUM2/0iscussion,4/28/85 In Birch Creek Slough,on the other hand,a relatively high CPUE occurred in early summer with much smaller values throughout the summer and fall.The relatively high CPUE's in early summer are probably due to a natal effect as Birch Creek has a spawning run of coho salmon (Barrett et al.1985). A comparison of juvenile coho catch rates between tributary mouths and the Talkeetna outmigrant trap (RM 103.0)suggests that a redistribution of juveniles into suitable rearing habitat peaks from late July to early August.The catch per hour of 0+coho at the Talkeetna outmigrant trap - - - -, increases during this time period while CPUE at various tributary sites undergo marked changes in rel ative abundance al so.Bi rch Creek 51 ough,- which habitat modelling indicates to be relatively poor coho tributary mouth rearing habitat ("Figure 27),shows a reduction in CPUE in late July,following natal emigration,while Caswell Creek,a site evaluated as having relatively good rearing habitat,has increasing CPUE's begin- ning at thi s time.A study conducted by Del aney and Wadman (1979)in the Little 5usitna River found emigration of emergent fry from natal areas after the end of June. Instream flow effects of the lower Susitna River upon juvenile coho salmon are 1imited to the backwater zone effects at tri butary mouths because coho juveniles make little use of the side channel/slough sites.... Initially backwater may decrease the amount of habitat slightly as the tributary mouths change from free flowing to a backwater zone but then ~JUA generally increases'as the amount of cover increases with further increases in stage.Overall the WUA generally increases with mainstem ..- DRAFT/PAGE 7 3/7 /85,4/15/85 NUM2/Discussion,4/28/85 discharge.Also,the backwater may improve access into small tribu- taries and beaver ponds where rearing and oven'/intering may occur. Studies of coho salmon distribution in 1982 by zone showed that the coho generally preferred free-flowing tributaries over backwater zones (ADF&G 1983},however,cover in the free-flowing tributaries was often better than in the backwater areas.For example,Si rch Creek Slough generally has poor cover whi 1e Si rch Creek itsel f has abundant emergent and aquatic vegetation in which coho were abundant. 4.3 Chum Salmon The use of minnow trapping during 1981 and 1982 juvenile anadromous studies makes comparisons of lower river catch and CPUE data with 1984 studies difficult because chum salmon are rarely captured in minnow traps.The necessity for very early sampling,almost concurrent with .ice-out,becomes important when studying chum salmon juveniles.Their early season movement and short time in the Susitna River system makes ~~nclusions difficult. Chum salmon fry CPUE I s by macrohabitat type contrast with the 1983 data for the middle reach (Dugan et al.1984),which indicated heaviest use of tributaries and side sloughs.The 1983 catch rates,however,reflect the prevalence of natal sloughs in the middle reach,while the lower reach contains few natal side channel/side sloughs and also few upland sloughs.Also,side channels were not extensively sampled until July in 1983. DRAFT/PAGE 8 3/7/85,4/15/85 NUM2/Discussion,4/28/85 .- Chum salmon spawning activity was observ >i 1984 in several side C::.yf""4UJt . channel sites where ..!lQ..-ne had ..e-v-ef'l6een observed (Barrett et al.1985).----~ The presence of adult spawning in these areas indicates that under certain conditions side channels provide suitable spawning habitat in the lower river.Chum salmon fry observed in some of the side channels may be rearing near their natal areas. The exact stimulus for the outmigration of chum salmon from the Susitna River is not known but probably reflects a combination of factors (Roth et al.1984).Increased turbidity and higher flows were noted as.r possible initiating cues to innate behavior toward increased movement, culminating in outmigration.Turbidity in most sites rose above the indicated preference range (<:200 NTU)by the middle of July.The sharp decline in CPUE,from early June (3+fpc)to late June (1+fpc)follows the peak monthly Susitna discharge of 17 June (USGS Sunshine provisional J:F-+~r-(,tJI/~~ata),and the mid-June peak of chum outmigrati on past the Talkeetna.+1't~'f""1 o4""~Jr~ J tk~~[5 CL,JvflU "trap •There was no difference in the average lengths of outmigrant chum 'f,~t~~/!salmon captured at Talkeetna and Flathorn stations (Part 1 of this (~.:~eport)which suggests .th_a_~_~!:,h_f!-OII1_j:.I1~_I11:i.<fd!-:.r-e~-,,_o~tmigra!e ~F~t1"-d:u-~~_!~:.~:_~~er reach without rearing.Sj!!-ce turbidity may trigger :f:::;1~~~,3~!!!!:,yr':t~~-tl,~!iiddl:,I,>ach,jt seems reasonable that chum sal mon ~~J ~fry would not rear in the lower reach where the turbidity is even ~J£4.higher, b Since chum salmon outmigration is mostly completed by mid-July,flow effects are limited to this time period for this species.Chum salmon made heavy use of the side channels during this time while use of the "'" - DRAFT/PAGE 9 3/7 /85,4/15/85 NUM2/Discussion,4/28/85 tributary mouths was much more 1imited.Apparently chum salmon do not move into the tributary mouths much,as presumably most of their Illove- ments are gradually downstream and out of the system.Most of the use of the side channels for rearing occurs before the turbidities become too high. i~ Use of the .side channels is limited by depth and velocity onlf)as instream cover seems unimportant (Appendix A).Chum fry were captured primarily in shallow sampling cells «1.0 ft)which had a relatively low velocity and low to moderate cover.This distribution suggests these fish are rearing as Hunter (1959)reported that chum salmon migrate in the center of the channel where water velocity is greatest. After breaching,side channel.s WUA's sometime§;may increase or decrease ~_.~~._'".....--""".." but the proporti on of the area that is suitable generally decreases as 'Z velocities and depths become unsuitable •.Turbidities also quickly----------------_.._---_.__.-.._---_.".~,--- increase seasonally so that some si de channel s become turbid more quickly than others dependent upon the turbidity regimes in the Chulitna,Talkeetna,and Susitna rivers. Since chum salmon side channel WUAls respond very similarly to those of chinook salmon at individual sites,it appears that an analysis of response to changes in mainstem discharge for chinook would also hold for chum salmon.A time series analysis of flow regimes,would only need to take place through mid-July for chum salmon,however,while chinook salmon fry occur throughout the season in side channels. DRAFT/PAGE 10 3/7/85,4/15/85 NUM2/Discussion,4/28/85 I!l!!!!l I 4.4 Sockeye Salmon Tributary mouths were the primary capture sites for sockeye salmon in the lower river while in the middle river sockeye salmon were captured primarily at side sloughs (Dugan et ale 1984).Side sloughs were the primary spawning areas in the middle river,while tributary/lake systems are the major spawning areas in the lower reach (Barrett et ale 1985). Since side sloughs were the primary spawning areas for sockeye salmon in the middle reach,large catches of juvenile sockeye in these side sloughs were due to natal effects. Few sockeye juveni 1es were captured in early June at modell ed JAHS' sites.This low incidence was probably due to lack of natal habitat in mainstem influenced areas of the lower river.Outmigrant trap catches - - at Talkeetna (RM 103.0)and Flathorn (RM 22.4)indicate that sockeye fry are redistributing,in the system by the middle of June (Roth et ale 1985).The greatest catch per cell of juveni 1e sockeye occurred at the ,- modelled sites during late June. The consistently low CPUE's in side channel sites suggest these areas are of l"imited value for juvenile sockeye rearing.Possibly these juvenile sockeye catches represent transient populations.An exception may be Beaver Dam Side Channel and other side channels located in the Sunshine side channel complex where lower turbidities may allow juve- niles to rear. - - - I~ DRAFT/PAGE 11 3/7/85,4/15/85 NUM2/0iscussion,4/28/85 Beaver Dam 510ugh probably had moderate numbers of sockeye present throughout the season.This site resembles a lake system as ~t has low velocities,high cover and relatively warm temperatures during the entire open-\'Jater season.This site might be used as overwintering habitat,but o it was not sampled after September so we have no data to support this hypothesis. Rolly Creek mouth produced only low CPUE I s of sockeye fry unti 1 early August.Emergent and aquatic vegetation were profuse at this site in mid season periods,however,which made sampling difficult.After late August,the numbers of chinook,coho,and sockeye juveniles increased. Although high numbers of these salmon fry were caught late in the season,the habitat at low water levels is less suitable for over- wintering than at Beaver Dam Slough.Extended late season sampling and sL!rveys in Apri 1 and May wou1 d be neces sa ry to determi ne use of these two sites for overwintering by sockeye salmon Instream flow effects upon sockeye salmon juveni1 es.-h.0~J~ AAJh ~,1ttf-c~ v'~L ~II....t,,-rW'v1 "2- reari ng occurl at both tlfle ' - - - ez () tributary mouths and in side channels.Occurrence of sockeye juveniles W\.-.-"L. in side channels appear~to be limited by _turbiditY~.tft~\0.n1y f~ur \,)J-'l.f-.fL-~ side channel sites wfl.e-l"e juvenile sockeye -we-re--captured more than half the times samPle~the Talkeetna River plume.Even at these four sites,numbers of sockeye fry captured were usually small. At tributary mouths,larger numbers of sockeye fry were found.In these sites,the formation of backwater zones probably has a major effect in DRAFT/PAGE 12 3/7/85,4/15/85 NU~2/Discussion,4/28/85-increasing WUA for sockeye salmon juveniles.The response o~the increase in WUA for sockeye is similar to that of chinook salmon._ Access into suitable rearing and overwintering areas may also occur with the increase in backwater.For example,access into potential rearing areas as Whitsol Lake may be inhibited if Kroto Slough Head is not overtopped.Also several other small tributaries along the Kroto Slough side channel may be inaccessible if flows are below those required for overtopping. Typically,WUAs for sockeye increase after overtopping of the side channels but then gradually decrease with further increases in discharge as side channel velocities became unsuitable.Sometimes backwater areas may form at the mouths of si de channels (for exampl e,Sucker Si de' Channel)and modify this relationsh'ip somewhat so that WUA may even increase with increases in discharge for much longer periods. Generally,the proportion of area that is usable within side channels decreases with flow as velocities become less suitable. 100 - ~ ! -. """" ;; )~----------------------------------------- DRAFT/PAGE 1 4/15/85,4/28/85 NUM2B/Contributors 5.0 CONTRIBUTORS Dana Schmidt provided the study design.David Sterritt,Robert - - - Marshall,Karl Kuntz,John McDonell Richard Sundet,and Stuart Pechek collected field data along with Dale Corzine and Mike Domeier.Pat Morrow,Isaac Quera l,Tommy Wi throw,Gl enn Freeman,Doug Sonnerholm, Sharie Methvin,and Chris Kent under the direction of Tim Quane installed staff gages,mapped the thalwegs,and helped with physical data analysis.Donna Buchholz keypunched the data and Chuck r~iller, Kathrin Zosel,Gail Heinemann,and Allen Bingham managed the mainframe computer data base.Cartography was done by Carol Hepler and Roxanne Peterson.Jim Anderson and Jeff Bigler ran the weighted usable areas for the IFIM models and also collected field data.Diane Hilliard helped choose study sites for the IFIM models.The typing was done by Skeers Word Processing. Jo f DRAFT/PAGE 1 4/15/85 NUM2B/Acknowledgements 6.0 ACKNOWLEDGEMENTS Funding for this study was provided by the State of Alaska,Alaska Power, Authority. 102 -i - Synop- habitat of Fish - .... - - DRAFT/PAGE 1 4/15/85,4/28/85 NUM2B/Literature Cited 7.0 LITERATURE CITED Alabaster,J.S.1972.Suspended solids and fisheries.Proceedings of the Royal Society of London.B.180:395-406. Alaska Department of Fish and Game.1981a.Phase 1 final draft report. Subtask 7.10.Juvenile anadromous fish study on the Lower Susitna River (November 1980-0ctober 1981).Alaska Department of Fish and Game Susitna Hydro Aquatic Studies.Anchorage,Alaska. ·1981b.Phase 1 final draft report.Volume 1 (December 1980- --=October 1981).Subtask 7.10.Aquatic habitat and instream flow project.Alaska Department of Fish and Game Susitna Hydro Aquatic Studies.Anchorage,Alaska. 1983a.Susitna Hydro aquatic studies phase II basic data report.Volume 3 (l of 2).Resident and juvenile anadromous fish studies on the Susitna River below Devil Canyon,1982.Alaska Department of Fish and Game Susitna Hydro Aquatic Studies. Anchorage,Alaska. 1983b.Susitna Hydro aquatic studies phase II basic data report.Volume 3 (2 of 2:Appendices A-H).Resident and juvenile anadromous fish studies on the Susitna River below Devil Canyon, 1982.Alaska Department of Fish and Game Susitna Hydro Aquatic Studies.Anchorage,Alaska. ___.1983c.Susitna Hydro aquatic studies phase II report. sis of the 1982 aquatic studies and analysis of fish and relationships (2 of 2:Appendices A-K).Alaska Department and Game Susitna Hydro Aquatic Studies.Anchorage,Alaska. ·1984a.Susitna Hydro aquatic studies (May 1983 -June 1984) ---procedures manual (l of 2).Alaska Department of Fish and Game Susitna Hydro Aquatic Studies.Anchorage,Alaska. ·1984b.Susitna Hydro aquatic studies (May 1983 -June 1984) ---procedures manual (2 of 2:Appendices).Alaska Department of Fish and Game Susitna Hydro Aquatic Studies.Anchorage,Alaska. ___.1985.(In Preparation).Susitna aquatic studies procedures manual (May 1984 -June 1985).Alaska Department of Fish and Game, SusitnaAquatic Studies.Anchorage,Alaska. Barrett,B.M.,F.M.Thompson,and S.N.Wick.1985.1984 salmon escape- ment studies in the Susitna River drainage (Draft)Report No.4. Alaska Department of Fish and Game Susitna Hydro Aquatic Studies. Anchorage,Alaska. Bisson,P.A.,and R.E.Bilby.1982.Avoidance of suspended sediment by juvenile cohQ salmon.North American Journal of Fisheries Manage- ment 4:371-374. DRAFT/PAGE 2 4/15/85,4/28/85 NUM2B/L i terature Ci ted .. Bovee,K.D.1982.A guide to stream habitat analysis using the in- stream flow incremental methodology.Instrea1ll Flow Information Paper.No.12.U.S.Fish and Wildlife Service.FWS/035-82/26. Delaney,K.J.,K.Hepler,and K.Roth.1981.Deshka River chinook and coho salmon study.Alaska Department of Fish and Game,Division of Sport Fish.Federal Aid in Fish Restoration,Project AFS-49,Vol. 22 • ....,...-__'and R.Wadman.1979.Little Susitna River juvenile chinook and coho study.Alaska Department of Fish and Game.Division of Sport Fish. Dugan,L.J.,D.A.Sterritt,and M.E.Stratton.1984.The distribution and relative abundance of juvenile salmon in the Susitna River drainage above the Chulitna River confluence.Part 2 in Schmidt, D.C.,S.S.Hale,D.L.Crawford,and P.M.Suchanek (ed~).1984. Resident and juvenile anadromous fish investigations (May -October 1983).Susitna Hydro Aquatic Studies.Report No.2.Alaska Department of Fish and Game.Anchorage,Alaska. Hunter,J.G.(l959).Survival and production of pink and chum salmon in a coastal stream.Journal of the Fisheries Research Board of Canada.16(6):835-886. Klinger,S.,and E.W.Trihey.1984.Response of aquatic habitat surface areas to mainstem discharge in the Talkeetna to Devil Canyon reach of the Susitna River,Alaska.E.Woody Trihey & Associates.Anchorage,Alaska. Marshall,R.P.,P.M.Suchanek,and D.C.Schmidt.1984.Juvenile salmon rearing habitat models.Part 4 in Schmidt,D.C.,S.S.Hale,D.L. Crawford,and P.M.Suchanek (ed5:l.1984.Resident and juvenile anadromous fish investigations (May -October 1983).Susitna Hydro Aquatic Studies.Report No.2.Alaska Department of Fish and Game.Anchorage,Alaska. Milhous,R.T.,D.L.Wegner,and T.Waddle.1981.User's guide to the physical habitat simulation system.United States Fish and Wild- life Service.Biological Services Program FWS/OBS-81/43. r~undie,J.H.1969.Ecological implications of the diet of juvenile coho in streams.p.135-152.In T.G.Northcote (ed.),symposium on salmon and trout in streams.H.R.MacMillan Lectures in Fisheries,Vniv.B.C.,Vancouver. Noggle,C.C.1978.Behavioral,physiological and lethal effects of suspended sediment on juvenile salmonids.Master's thesis,Univer- sity of Washington,Seattle,Washington,USA. - - ,~ DRAFT/PAGE 3 4/15/85,4/28/85 NUM2B/Literature Cited Quane,1.,P.Morrow,I.Queral,1.Keklak,and 1.Withrow.1985. Technical memorandum in support of Task 14 (Lower River Resident and Juveni 1e Anadromous Fi sh Studies).Al aska Department of Fi sh and Game,Susitna Aquatic Studies.Anchorage,Alaska. Riis,J.C.,and N.V.Friese.1978.Preliminary environmental assessment of hydroelectric development on the Susitna River. Alaska Department of Fish and Game.Div.of Sport Fish and Comm. Fish.,Anchorage,Alaska. Roth,K.J.,D.C.Gray,and D.C.Schmidt.1984.The outmigration of juvenile salmon from the Susitna River above the Chul itna River confluence.Part 1 in Schmidt,D.C.,S.S.Hale,D.L.Crawford and P.M.Suchanek (eds.)-.-1984.Resident and juvenile anadromous fish i nvesti gati ons (r1ay October 1983).Susi tna Hydro Aquati c Studies.Report No.2.Alaska Department of Fish and Game. Anchorage,Alaska. Sigler,J.W.,T.C.Bjornn,and F.H.Everest.1984.Effects of chronic turbi dity on dens i ty and growth of steel heads and coho salmon. Transactions of the American Fisheries Society.113:142-150. Suchanek,P.M.,R.P.Marshall,S.S.Hale,and D.C.Schmidt.1984. Juvenile salmon rearing suitability criteria.Part 3 in Schmidt, D.C.,S.S.Hale,D.L.Crawford and P.M.Suchanek (eds:-).1984. Resident and juvenile anadromous fish investigations (May -October 1983).Susitna Hydro Aquatic Studies.Report No.2.Alaska Department of Fish and Game.Anchorage,Alaska. 'c~lJ - - - - - - APPENDIX A LOWER SUSITNA RIVER JUVENILE SALMON REARING SUITABILITY CRITERIA DRAFT/PAGE 1,4/29/85 4/1/85,4/15/85 NUM2B/Appendix A - - - - - - - ..... TABLE OF CONTENTS I NTRODUCTI ON ~lETHODS RESULTS Chinook Salmon Cl ear Water Turbi d \~ater Coho Salmon Sockeye Salmon Chum Salmon Summary DISCUSSION Chinook Salmon Coho Salmon Sockeye Salmon Chum Salmon LITERATURE CITED DRAFT/PAGE 2,4/29/85 4/1/85,4/15/85 NUM2B/Appendix A -I -. DRAFT/PAGE 3,4/29/85 4/1/85,4/15/85 NUM2B/Appendix A """' ..... - INTRODUCTION Habitat suitability criteria are necessary for use in evaluating fish habitat using the instream flow incremental methodology (Bovee 1982). The criteria express the value of a habitat variable such as velocity on a zero (unusable)to one (optimum)basis for a given fish species and life stage.The suitability criteria are coupled with the habitat present within a study site to produce estimates of equivalent optimal habitat called weighted usable area. Juvenile salmon rearing suitability criteria have been used to model the response of jl,lvenile salmon habitat to variations in mainstem discharge of the middle reach (Chulitna River confluence to Devil Canyon)of the Susitna River (Hale et al.1984,Marshall et al.1984).The suitability criteria used in these studies were developed specifically for the middle Susitna River by Suchanek et ale (1984).EWT&A (1985)modified a few of the same suitability criteria for use .in impact analysis of chinook salmon rearing in the middle Susitna River. In 1984,some of the juvenile salmon habitat modeling effort was direct- ed toward evaluating responses of juvenile salmon habitat in the lower Susitna River (below the Chulitna River confluence)to discharge variations.Since habitat data collection techniques used in 1984 were similar to those used during the 1983 studies,suitability criteria specific to the lower reach can be developed.The purpose of this appendix is to verify the applicability of the suitability criteria _I I -DRAFT/PAGE 3.1,4/29/85 4/1/85,4/15/85 NUM2B/Appendix A developed in 1983 by Suchanek et al.(1984)for use in the lower river habitat studies.The general philosophy was to use the 1983 middle river criteria curves for the lower river unless the 1984 studies in the lower river provided evidence for modifications. ;\- -fl;-~ ""'" - - - -i - - "... - -- - - DRAFT/PAGE 5 t 4/29/85 4/1/85 t 4/15/85 NUM2B/Appendix A METHODS The field sampling methodology used is detailed in Section 2.1 of this report.·This methodology is very similar to that used during the 1983 studies (Suchanek et al.1984)and will only "be briefly s.ummarized here. Sampling sites included 20 habitat model sites normally sampled twice a month and 31 opportunistic sites which were usually only sampled once. At each site t 6 ft x 50 ft rectangular cells were sampled for fish and then habitat variables were measured in each.cell.Cells were selected randomly at model sites although sometimes additional selected cells with II goo dll habitat were sampled.At opportunistic sites t cells were selected to encompass a variety of habitat conditions within what was thought to be usable habitat.Habitat measurements taken at each cell ·sampled included a representative mean column velocity and depth and estimates of primary cover type and percent total cover (Appendix Table A-I). Appendix Table A-I.Percent cover and cover type categories. Group #%Cover Group #Cover Type 1 0-5%1 No object cover 2 6-25%2 Emergent vegetation 3 26-50%3 Aquatic vegetation 4 51-75%4 Debris or deadfall 5 76-96%5 Overhanging riparian vegetation 6 96-100%6 Undercut banks 7 Gravel (l1l to 311 di ameter) 8 Rubble (3 11 to 511 diameter) 9 Cobble (larger than 511 diameter) ,4 --3 DRAFT/PAGE 5.1,4/29/8: 4/1/85,4/15/85 NUM2B/Appendix A The data collected were examined for suitability criteria development by using the procedures described in Suchanek et al.(1984),with a few modifications. Suitability was represented by mean catch per cell for chinook and coho sa lmon and proporti onal presence (proporti on of cell s sampl ed in whi ch fi sh were captured)was used as the suitabil ity measure for chum and sockeye salmon.Data were pooled by species for analysis but some data were excluded from analysis by using resuits from the distribution and abundance analysis (Section 3.2)which indicated factors other than the microhabitat variables of velocity,depth,and cover were greatly affecting distribution.Cells excluded varied by species and are detailed in the results section.The beach seine ana electrofishing data were pooled for analysis because the sampling method used to sample a cell was thought to be most effective given the sampling conditions. - - Groupings of habitat variables were identical to those used in 1983. Percent object cover categories 76-95%and 96-100%were pool ed because of small sample sizes.Velocity and depth were pooled in groups identi- cal to those used in 1983 with the exception that cells with depths of ~ 0.1 feet were examined separately.In 1983,only two cells with a depth of 0.1 feet were sampled,and therefore insufficient data were available for examination of suitability of this depth. Compari sons of the 1983 data with the 1984 data were made by pl otti ng the suitabil ity criteria derived in 1983 on the same graph with com- -- - - - DRAFT/PAGE 6,4/29/85 4/1/85,4/15/85 NUM2B/Appendix A parable 1984 data.On the depth and velocity graphs this was done by normalizing the suitability to 1.0 for the 1984 depth or velocity increment with the highest suitability and then plotting the 1983 suitability criteria normalized to the same scale.The 1984 percent cover data were first regressed against catch per cell or proportional presence,and,if significant,the regression line was plotted and the suitability normalized to 1.0 for the highest cover category.The 1984 percent cover suitability line was then plotted on the same graph,by using the normalized 1.0 as the starting point.The suitability of cover type for each speci es was calculated with the 1984 data usi ng the methods described in Suchanek et al.(1984).The suitabilities cal- culated were then graphed against the cover type suitabilities calcu- lated in 1983. Variations in histogram distributions are to be expected on a univariate basis given that percent cover,cover type,velocity,and depth together affect suitabil ities of a cell.Therefore,composite weighting factors were calculated for each cell using the 1983 suitabil ity criteria and revised 1984 criteria and then these weighting factors were compared with catch.Composite weighting factors were calculated by multiplying suitability indices for cover type,percent cover,and velocity togeth- er.For chinook and coho salmon,Pearson correlation coefficients were calculated between composite w~ighting factors and catch per cell [transformed by natural log (X +l)J.Chi-square association tests were run between chum and sockeye proportional presence and composite weight- ing factor value intervals calculated us"ing the 1984 criteria data. ---,_.--------,.;,..-- DRAFT/PAGE 7,4/29/85 4/1/85,4/15/85 NUM2B/Appendix A Intervals of composite weighting factors were specified by dividing the data into four groups of approximately equal sizes by value of the ""'!'II composite weighting factor.Pearson correlation coefficients and results of the chi-square analysis were then compared with the same analyses done in 1983.,Most of th~statistical tests and data manipu- lations were done with the Statistical Package for the Social Sciences (SPSS)(Nie et al.1975). If the fit of the 1984 data to the 1983 suitability criteria did not· seem close upon visual inspection,the 1983 criteria were modified.One of the procedu~es for modification was as follows.If,for example,the 1984 velocity distribution data appeared to match closely the 1983 velocity criteria,the 1983 velocity criteria were input as suitabilities and averaged over each increment of a variable such as depth for which a modification of suitability was desired.These averages were then mul tiplied by the mean catch of fi sh per cell divided by the mean suitability.The actual mean catches per cell by depth increment were then divided by the adjusted mean velocity suitability. If this ratio was less than 1.0,this would indicate less use of a depth increment than expected given the average suitability for velocity.If the ratio was greater than 1.0,the use would be more than expected by adjusting for the effect of velocity.Sometimes this procedure would be effective in taking out variation caused by the other variable.If necessary,thi s .procedure was used to adjust for effects of two or more variables. - -- _. - - DRAFT/PAGE 8,4/29/85 4/1/85,4/15/85 NUt~2B/Appendi x A If the above procedure was not effective in discounting the extraneous variation,then the criteria were modified using professional judgement. Correlations or chi-square association tests were then calculated between mean catch and calculated composite weighting factors using the modified criteria. RESULTS Abundance artd distribution data (Section 3.2)have shown that the number of all four species of salmon was very small at sloughs in the lower reach.Even sampling cells at sloughs with good habitat failed to have any significant number of fish present in comparison with similar cells at the other macrohabitat types.Fi sh were therefore respondi ng to factors other than the availability of suitable mic'rohabitat in their use of sloughs.For this reason,data collected at sloughs were elim- inated from suitability criteria analyses to avoid comparing similar cells with large differences in mean catch. Chinook Salmon Chinook salmon suitability criteria were developed for both clear «30 NTU)and turbid (>30 NTU)water in 1983 as fish distribution was very different in the two categories (Suchanek et al.1984).The catch in cells without object cover was much greater in turbid water than in clear water.The data collected in the lower river have ShOrln that turbidity may limit the distribution of chinook salmon by being too high - ~DRAFT/PAGE 9,4/29/85 4/1/85,4/15/85 NUM2B/Appendix A (Figure 10).Since cells with good habitat were sampled when high turbidity was limiting use by chinook salmon fry,we decided to elimi-- nate sampled cells with turbidities greater than 350 NTU.- After eliminating cells in sloughs and cells with turbidities greater than 350 NTU,1155 cells were available for analysis of chinook dis- tribution.Of the 1155 cells,400 were sampled in water with a turbidity of 30 NTU or less.Mean adjusted catch (catch adjusted to a cell of 300 ft 2 )per cell of chinook fry in the clear water cells was 1.3,while mean adjusted catch per cell in the turbid cells was 1.1. A scatter plot of chinook salmon catch in cells without object cover - versus turbidities ranging to 100 NTU was examined.No notable inflections in catches of chinook salmon fry were noted over this range, although gradual increases in catches occurred across the range.It seemed reasonable,therefore,to keep the same 30 NTU breakpoint between high and low turbidity data for this year's analysis. Clear Water - Correl ations among the values of habitat attdbutes and cl ear water «30 NTU)chinook catch range to 0.32 in absolute value and a number of ~ the correlations are statistically significant (Appendix Table A-2).In addition to these data,partial habitat data were recorded for four additional clear water cells and these additional data are used in subsequent analyses. - Appendix Table A-2. DRAFT/PAGE 2 4/2/85,4/29/85 NUM2/Table A-2 Kendall correlation coefficients between habitat variables and chinook catch by cell (N=396)for all gear types,in clear water. Percent Cover Cover Type Velocity Depth Chinook Percent Cover 1.00- Cover Type 0.08*1.00 Velocity -0.32**0.04 1.00 Depth 0.03 --0.08*-0.04 1.00 Chinook 0.07 0.09*-0.09*0.21**1.00 ,....*Significantly different from 0 at p<0.05. **Significantly different from 0 at p<O.Ol. - ...... ;..-J I ~ DRAFT/PAGE 10,4/29/85 4/1/85,4/15/85 NUM2B/Appendix A Composite weighting factors for all cells sampled were calculated by using the 1983 suitability criteria and also with modification of the velocity criteria as proposed by EWT&A (1985)and then correlated with chinook catch transformed by natural log (x +1).In clear water,the correlation in 1983 was 0.43 but the correlation with the 1984 data was only 0.31 for the original criteria data and 0.26 with the change in velocity criteria proposed by EWT&A (1985).It was therefore deemed desirable to modify the criteria where large differences in individual criteria were found.- A Least squares regressions were run between chinook catch per cell and the percent cover categories in clear water.There was a significant ~ positive regression which is very similar to the suitability line developed in 1983 when the Y axis is normalized to a suitability of one (Appendix Figure A-I).The 1983 suitabil ity criteria was therefore retained as a good estimate of this relationship. The distribution of mean catch per cell of chinook fry by velocity interval in clear water in 1984 shows that peak catches were made in sampling cells with a velocity ranging from 0.1 to 0.3 fps (Appendix Figure A-2).By nonl1alizing this peak to a suitability of 1.0 and then overlaying the 1983 criteria,it appears that chinook use lower velocity water in the lower reach under clear conditions.It was noted that the 1984 clear water distribution of catch by velocity interval was very similar to the 1983 turbid water velocity suitability criteria and - - ..- -T.!.Standard Error .J... N =Number of Cells Sampled 1983 =-=1984 T , T ! i ~. ><.....cz:.... >-r-.... -I.... a:l<:r--=:> V) O.~ i 1 --J..-..<L...---'-_0..00 "'1:41 ",~'J'" (l'-·~)l51-'1s7oj 1/.0 I.D~ I [ I I~ - -PERCENT COVER CATEGORIES .-Appendix Figure A-I.Mean catch of juvenile chinook salmon per cell by percent cover category (bars)in clear water of the lower Susitna River,1984 and comparison of fitted suitability indices (lines)calculated in 1984 and for the middle Susitna River,1983. - ··l 1 T0.::'-°1 0 J +Standard Error 0z-N =Number of Cells Sampled:J::u 5.0 -' 1983 (turbid)100 -'u.J u 14.0c:r: LLI til Q.. ><:J::..... U 0~].0 \Zc:e \-u z l ~>-c:e I- l.LI -2::-''Lo ,~D a:lc:e !:; J ~,.0 AJ~53 Vl O.:U ,0.0 GOD 0 0.3 4'••'1 I.'Z,I.V '1.1 3.'t VELOCITY (ft/sec) Appendix Figure A-3.Mean catch of juvenile chinook salmon per cell by velocity intervals (bars)in clear water of the lower Susitna River,1984 and fitted suitability index (line)developed for turbid water in the middle Susitna River,1983. - - - -! ",... -- DRAFT/PAGE 11,4/29/85 4/1/85,4/15/85 NUM28/Appendix A therefore this criteria was plotted against the 1984 data (Appendix Figure A-3).Since it was so similar,it was taken as a good estimate of the lower river velocity suitability for chinooks in clear water. Cover type suitabil ities derived oin 1984 for chinook in clear water contrast sharply with those derived in the middle reach in 1983 (Appendix Figure A-4).Debris was used much less by chinook in the lower reach while emergent vegetation was more heavily used.The sample size of the cobbl e/boul der cover category was only one and therefore this cover type could not be evaluated.Catches in the cells without object cover were also relatively higher in 1984 than in 1983. Therefore,we believed that 1983 suitability for cover types would not apply in the lower reach.By adjusting for the effects of velocity and percent cover,better estimates of cover type suitability for the lower river were formul ated from the 1984 data (Appendix Figure A-5).Si nce cobble and boulder sample sizes were low,suitabilities for these cover types were kept proportional in suitability to large gravel as was the case in 1984.Since the lI no cover"catches were relatively large,we arbitrarily lowered the suitability for no cover cells to 0.10,the suitability found in 1983. A heavy use of deep,clear water by chinooks was found in 1984 while in 1983 the dat~suggested a peak in use of cell s 1.0 to 1.5 feet deep (Appendix Figure A-G).In 1983,an evaluation of depth found it had little effect on increasing the correlation of fish catch with composite weighting factors using it.Depth was used,however,in the 1983 -I, ~ QJ >ou oz -..., ~,., 1 j I-I • Ir .i' t ~- O'l <:: .~ 0..:;- <::." .".~ .<:~ ~." Q,lQ. >-~ 00:: co ~u .... .~."........ ."al ::l0'l O"al <C> ~1983 [j 1984 N =Number of Cells Sampled ·:< / (' i/ ( II>.... ~ .0 QJ Cl ~./. ><..... Cl Z- COVER TYPE Appendix Figure A-4.Comparison of cover type suitability indices for juvenile chinook salmon in clear water calculated from 1984 lower Susitna River distribution data and 1983 distribution data • .. ~- N Number of Cells Sampled a--I'~ - -•<. ~-II tI II '< ,- '"..... '-.0 QJ C QJ......c..c ::>cr: I::o.-u+' :;~QJ (i) ItS Q..J t:n-> ,::,~5-ta O"w "''-eX:>...J~ COVER TYPE I._.\..J 0> I::.-0>1:: 1::'"",.- -C'-'-'"we.>.-Ocr: n - - Appendix Figure A-5.Cover type suitability indices for juvenile chinook salmon in clear water calculated from 1984 lower Susitna River distribution data. A-IS - 12~I!~tandard Error J ""'" ; '10 N Number of I Cells Sampl ed 1983 ""'""goO -- :.c:j 0 I07.0 ---I Z :J-~ ::I: U .......... Ll.J ../ ,.0'0 -U ex:ILl.J Q..~• ::I: U 'f.0 --' x ....I Ll.J ~ <3 I Cl I ;z / -. z: ~ <C !PI >- Ll.J .... X -...... co ""'" .24 <C....-:::l V'l I·"1 o.b II:1"1.7 JJ:I". JJ ='3' C,O ~"" o.~o,ti Q.'~I~La IN ,"-1.1 %.0 w 4.5 ~ DEPTH (ft) Appendix Figure A-G.Mean catch of juvenile chinook salmon per cell by depth intervals (bars)in clear water of the lower Susitna River,1984 and fitted suitability index (line)developed for the middle Susitna River, 1983. - ,~ .-.. - -. - - DRAFT/PAGE 12,4/29/85 4/1/85,4/15/85 NUM2B/Appendix A modelling efforts as having no value if less than 0.14 ft and having a suitability of 1.0 if greater than 0.15 ft.In order to evaluate depth, suitability criteria were fit to the data using professional judgement after fi rst adjusti ng for mean velocity and percent cover suitabil ity (Appenqix Figure A-7). After the modifications to the cover suitability and depth criteria were made,we then correlated transformed chinook catch with the composite weighting factors calculated with the 1983<percent cover criteria and turbid water velocity criteria along with the 1984 lower river cover type and depth suitability criteria.The correlation was 0.61,substan- tially higher than the original 1983 criteria.If depth was eliminated from the calculations,the correlation dropped to 0.26 and if primary cover type was dropped the correlation dropped to 0.52.Therefore it seemed reasonable to keep the new modified cover type and depth criteria as inputs. Turbid Water Correlations between the values of habitat attributes and chinook catch in turbid water range to 0.39 in absolute value and a number are statis- tically significant (Appendix Table A-3).Partial habitat data were recorded for 11 additional turbid cells and these additional data were used in subsequent univariate histograms. Correlations between composite weighting factors calculated with the 1983 turbid water criteria and 1984 chinook catch was 0.31,while all ~ sa ::.£ 0 0 ....z...... :I: U '"-oJ -oJ...... U a:s,p ...... D- :I:u -1l-e:::u z<:?oi'...... ::E: '2» ~Cf I!Standard Error N =Number of Cell s Sampl ed 1984 -j ,.i DEPTH (ft) tJ,36 I I ~O-IP '].1.11.$ ~, - - - - .Wi Appendix Figure A-7.Mean catch of juvenile chinook salmon per cell by depth intervals (bars)in clear water of the lower Susitna River,1984.Suitability index (line)fitted by hand using professional judgement. I~-:..""-1r-\i,.\;~ - - DRAFT/PAGE 1 4/2/85,4/29/85 NUM2/Table A-2 Appendix Table A-3.Kendall correlation coefficients between habitat variables and chinook catch by cell (N=744)for all gear types,in turbid water. Percent Cover Cover Type Velocity Depth Chinook.... Percent Cover 1.00 Cover Type 0.39**1.00 Velocity 0.05*0.16**1.00 Depth 0.06*0.26**0.21**1.00 . .-Chinook -0.02 0.00 -0.17**-0.15**1.00 -*Significantly different from 0 at p<0.05. **Significantly different from 0 at p<O.01. -. DRAFT/PAGE 13,4/29/85 4/1/85,4/15/85 . NUM2B/Appendix A composite weighting factors calculated by incorporating the cover modifications proposed by EWT&A (1985)were correlated with an r-value ~ of 0.26.Comparable correlation with the 1983 data was 0.38.These data again suggested that some modifications could be made,especially given the changes already made in'the cl ear water cover type suitabilities.- A comparison 6f 1984 velocity distribution data and the 1983 velocity suitability criteria for chinook salmon showed few differences (Appendix Figure A-8),and therefore was accepted as the 1984 criteria curve. Least squares regressions were run between chinook catch per cell and the percent cover categories in turbid water.There was no significant relationship between catch per cell and percent cover category and mean catch per cell decreased with increases in cover (Appendix Figure A-9). By adjusting for velocity,a slight trend upward was noted over the first three categories.The percent cover criteria developed in 1983 was therefore accepted as reasonable,as intuitively,increases in the amount of object cover would seem more desirable for fish. In 1983,cover type for chinook in turbid water was not evaluated. EWT&A (1985)modified the turbid water criteria,however,so that they more closely reflected the clear water criteria developed in 1983.In 1984,mean catches of chinooks in turbid water were highest in the emergent vegetation,rubble,and debris-deadfall categories,but catches were only slightly higher than in the cover -category "no cover". A-?c - - r~" -....q ...0.i -I>' ~ i-0-~0 :5 i-o-~ ,...=: ;:; '" l-00" <C l- S '" :.......3 _OoS ;......,, I i 'l.i~I/..8'.:1 I !Sundard Error N •N.n:ber of Ce 11 s Samp 1ed __1983 /J.ZD"l ,J.1."1 AI •.go ! 250 -, i 2..uJ ~.oo ---! ,.~S '! "",.SO Jg z;: u I.U".... ~u :'"""',-, "-"""1 xu ~ <CU O;"'S~z: <C IlE!!l)oSllI oU1 .".. OD - - ,-- Appendix FigureA~8. VELOCITY (ft/sec l Mean catch of juvenile chinooK salmon per cell by veloci~y intervals (bars}in turbid waters of the lower Susitna Rwer, 1984 and fitted suitability index (line)developed for the middle Susitna River.1983. '"0 0:z:... ::I: .-U ...J ...J..... U c::.....a.. ::I:u l- <Cu :z: <C..... X I o.!iP; 1J~5J,,5 (0-57.) ][~St"d,"E"o, N =Number of Cells Sampled __1983 /.00 x .tJ.Jo:z:... >- l-... ...J....co <C l-... :::;) V"! PERCENT COVER CATEGORIES Appendix Figure A-9.Mean catch of juvenile chinook salmon per cell by percent cover category (bar~)in turbid water of the lower Susitna River.1984 and fitted suitability index (line)calculated for the middle Susitna River,1983. DRAFT/PAGE 14,4/29/85 4/1/85,4/15/85 NUM2B/Appendix A Cover type was evaluated in 1984 by using the method of EWT&A (1985)for calculating turbidity factors from the fitted regressions of percent cover in clear and turbid water and their associated chinook mean catches.Turbidity factors were calculated (Appendix Table A-4)and then applied to the revised lower river cover suitability data.These revised suitabilities were much too low for many categories given observed catches and therefore a suitabil ity of 0.15 was assigIJed as a minimum for cover type suitabil ity in turbid water based on observed mean catches.Using this method,none of the suitabilities for cover type in conjunction with percent cover in turbid water are greater than 0.40 (Appendix Figure A-10). Appendix Table A-4.Calculations of turbidity factors for 1984 lower river data. Number of Fish Per Cell (Fitted to a Line Percent Turbi dity Cover Clear Turbid Factor 0-5%0.5 1.1 2.2 6-25%1.5 1.3 0.9 25-50%2.5 1.5 0.6 51-75%3.5 1.7 0.5 76-100%4.5 1.9 0.4 In turbid water,peaks in chinook use were found in water less than 0.5 ft deep in both 1983 and 1984 (Appendix Figure A-ll).In 1983,since fitting the depth suitability line to the data did not increase the composite weighting factor much,the depth criteria used for clear water (0 if less than 0.14 ft,1.0 if greater than 0.15 ft)was used for modelling. - - -------------------------------------------' ,~ - LEGEND Percent Cover 0.1 0 - 5 0.2 6 -25 0.3 26 -50 0.4 51 -75 0.5 76 -100 0.10 -irTf I " c:'".=:c:....0 0 '"........u:;"'c:....-....u ......-...c:...c:...... ;:....~"O :;;............-e"5 .....>........,........0......."'>.......-.:>c:"J~.."''''....."l:>...C:..'"ICTQf·.....>-..'"0Q:::>cD ...'"'".cc>..'"0'".s::;c: Appendi~Figure A-10. PERCENT COVER BY COVER TYPE Cover type suitabil ity indi ces for juvenil e chinook sal lOOn in turbid water developed from 1984 lower Susitna River chinook clear water distribution data • .~ t.'11 \$I I.!.standard Error r-"" H a Number of Ce11 5 $amp 1ell 8 __1983 ;c: til'5 1.00 ..,-..i ""... U I '. '" I ---- >< &~f ......'""-/I'~ is ,,.. rl =~~ ==U '" z:II ~ co:...::> E 0$'" oo-J i 1 .....~5..N:'3.'3..,'.J.,U,,"~P:r.el !AJ.74 ...+-L-I I I I I (I I I I I ~ F- O.t ".f'~0.'I."JoI W'''''41 ~:u '1.1 DEPTH (ft) Appendix Figure A-ll.Mean catch of juvenile chinook salmon per cell by depth intervals (bars)in turbid water of the lower Susitna River,1984 and fitted suitability index (line)developed for the middle Susitna River,1983. A-23 DRAFT/PAGE 15,4/29/85 4/1/85,4/15/85 NUM2B/Appendix A In 1983 there was only one turbid cell sampled with a depth of 0.1 feet and therefore the value of cells with this depth could not be evaluated.~ For purposes of IFIM modelling,this depth was assigned a suitability of A ~, by The - 0,while in the RJHAB model data this depth did not occur.In turbid water,21 cells of 0.1 depth were fished in 1984 and the mean catch was 0.5 chinook juveniles per cell.These data suggest that under turbid conditions the value of these shallow cells is greater than O. suitability criteria line was fit to the 1984 turbid water depth data first adjusting for the effects of velocity (Appendix Figure A-12). optimum depth ranged from 0.3 to 1.5 fps. Once all the criteria were modified,correlations were calculated between catch transformed by natural log (x +1)and the composite weighting factor calculated by multiplying the suitabilities for velocity,cover,and depth together.The correlation was 0.33,and if depth were removed the correlation dropped to 0.28.If cover,on the other hand were removed from the composite weighting factor,the corre- lation increased to 0.36.Intuitively,since instream cover has value in turbid water,it seemed reasonable to keep velocity,cover,and depth in the modelling. Coho Salmon Juvenile coho salmon suitability criteria were developed only for clear water in 1983.Very few coho were captured in macrohabitat types other than tributary mouths in the lower reach and therefore only tributary - - - - - - ><UJ Clo..'a z I I.H1.1.Ib01'1. r- +Standard Error,...-- N =Number of Cells Sampled 1984--r----------------....~I I \ -I \ I .1 ~\ -L_\"'.-/\l- I \ l ,,:,_0. -1-----\0\--_. l/1 I - I I !T, I IV ~2."..N~':1.20 N<Ill.I /Vo (")N~/6 .1 1 "I I I I I I I I I I I I ·"'" ,:6 ,,9'-:><: 0 0z-~:t: (oJ .....,...... UJ (oJ ~a::J6LU Q. :t: (oJ r"""....<rJPu z<UJ:::;;;;.-.",15 - DEPTH (ft) Appendix Figure A-12.Mean catch of juvenile chinook salmon per cell by depth intervals (bars)in turbid water of the lower Susitna River,1984.Suitabil ity index (line)fitted by hand using professional judgement. A-25 DRAFT/PAGE 16,4/29/85 4/1/85,4/15/85 NUM2B/Appendix A mouth data were used in suitability criteria comparisons.Most of the turbidities in the tributary mouths were less than 30 NTU although on two occasions,turbidities were over 100 NTU. A total of 345 cells with complete habitat data were sampled in tribu- tary mouths and another 2 cells with partial habitat data were sampled. Mean adjusted catch in the cells sampled was 1.2 fpc.Correlations among the values of habitat attributes and coho catch ranged to 0.43 in absolute value (Appendix Table A-5).Cover type was most highly cor- related with coho catch. The distribution of mean coho catch per cell by velocity interval in 1984 matched qUlte closely with the suitability criteria derived in 1983 for the middle river (Appendix Figure A-13).The 1983 velocity criteria were therefore chosen as representative for the lower river. A regression of coho catch to percent cover category was significant (Appendix Figure A-14).When the 1983 and 1984 data were normalized to 1.0 on the V-axis for the 76-100%category,however,the 1983 suita- bi 1i ty 1i ne had a much greater slope,and suitabi 1i ty for 0-5 percent cover in 1983 was 0.12,while in 1984 it was 0.33.After adjusting for the effect of velocity,the distribution of catches by percent cover interval appeared to be more similar to the 1983 distribution and since the sample size in 1983 was larger,the 1983 percent cover suitability relationship was chosen for use in the lower river. /'t ...,,, ti-~'::;l - - - Appendix Table A-5. DRAFT/PAGE 3 4/2/85,4/29/85 NUM2/Table A-2 Kendall correlation coefficients between habitat variables and coho catch by cell (N=345)in clear water. Percent Cover Cover Type Velocity Depth Percent Cover 1.00 Cover Type 0.05 LOO Velocity -0.43**0.02 1.00 Depth 0.05 -0.09*-0.14**1.00 ~Coho 0.09*0.23**-0.01 0.05 *Significantly different from 0 at p<0.05.-**Significantly different from 0 at p<O.Ol. ~~ - -----'-,------------------------------_......_---- - ~~~I:.Standard Error,.~1~I ~~NUlllber of ')..SO Cells Sa~led __19B3 roo~~~......-0.\0...u >< a:1 j ... '"...:s...'\:r OoCoD <.>>-!C =<.>... z ~'"::5 ,"":E 1 =::> Cl'.~"" N~"'"."~'!i~\l~.Il::.'l 0.00OCt;). 0.0 0.3 Q"ClfI \-S-IS ~~ 1lIIIIFIil', VELOCITY 1ft/sec) Appendix Figure A-13.Mean catch of juvenile coho salmon per cell by velocity intervals (bars)in the lower Susitna River,1984 and fitted suitability index (line) developed for the middle Susitna River.1983. - - - I~Standard Error' N •Number of CellS Sa~led ;, '"'l,S;:r '"'1......!lp:>...u a:...... is \.SO ~ :5 ~~1 :>'S?1 :,-Q1D 0...0 .............-.=,.,...--'---'---="...---'---~"'_.-'-!'i~qo'-+w":'":tJ.-.il oro (Io-~){";l;;-~)l:sl-~"b)('/b.'oC:db) PERCENT COVER CATEGOR rES Appendix Figure A-14.Mean catch of juvenile coho salmon per cell by percent cover category (bars)in the lower Susitna River.1984 and comparison of fitted suitability indices (lines)calculated in 1984 and for the middle Susitna River.1983. - A-2~ - - ..... DRAFT/PAGE 17,4/29/85 4/1/85,4/15/85 NUM2B/Appendix A Initial calculations of the suitability of cover type for coho salmon indicated that suitabilities in the lower river were similar to those found in 1983 (Appendix Figure A-15).After adjusting for the effects of velocity and percent cover,these est"imates of cover suitabil ity for the other six cover types were revised and will be used in 1984 lower river calculations (Appendix Figure A-16).Since sample sizes for the three substrate cover types were small,the suitability of 0.10 calculated in 1983 for rubble and boulders was used for these three cat- egories. The distribution of CPUE 's for depth was very different from that found in 1983 (Appendix Figure A-17).By adjusting for the effects of velocity,percent cover,and cover type there still was no trend in depth suitabilities and therefore depth suitability was not changed from that used in 1983. Sockeye Salmon Juvenile sockeye salmon suitability criteria were developed by pooling data over gear type and turbidity level in 1983.Since abundance and distribution data have indicated that sockeye salmon use of lower river side channels is l"imited by high turbidities (Figure 18),cells with turbidities greater than 250 NTU were eliminated from suitability criteria development. After cells with turbidities greater than 250 NTU were eliminated,922 cells with complete habitat data were available for analysis.Sockeye A-29 ,---,-------------..,....--------------------~---- - ..... I -'iI -I COHO 1983 COHO 1984 fZ1 ~ N =Number of Cells Sampled / / "/,. ~ ,I 0.0 0 .L.JL.t..J.£II:t!-Ic...:...!-=l..L-l~.l--~~_--..t::..,,£.uo:..a....t~._..l.J.t:..t..l..oi:..J...J,'-"-lt:'.c.t:....---,_;Z;_ \.00 I I + 1 !~ I '.'I ~., 1/,, ·1 ''.-~" .../.: Q.'lS -V , .:/;..:/::.. ><!.oJ Cl Z... >- I-... ..s... a:I ~... :::I VI ~t ... ~ ..Q CU Cl ......1\.CUQl__-0 ..0..- ..o~00 U a:I co.... U+>....'"........ "'41~enCTCUce> ens::.... ens:: c:'"'".....&:I\. I\.'"cue.>....o a:: I\. QJ >ou oz COVER TYPE Appendix Figure A-IS.Comparison of cover type suitabil ity indices for juvenil e coho salmon calculated from 1984 lower Susitna River distribution data. A-30 ..- 1 - -N Number of cells sampledl)Q ="-.. "<-",-I'~ '-. - (f \11 ::i-N u.- II <. <:-'-- C tt'-.... iI Ij <-<.,, '-<.. r---r---....--0 \Jl 11 ~ n 0.'2JI 1,1>0 - X L<.J 0 ~.e. z:.-. >-....-...J-,~CDcet; :::) til 01 I::I::I:: 0 -0..-OIl::+>-s..::s ~u ..l::oU 1::..GJ III U evGJ ev -'"Qlr;.......GJo:I >-s..",~-a ::c ........::s..0\1->0s..Cll->t .L:i-",ev 01>s..",s..ev U..c "l:I1::..c::s ..c ::s Cl s..",eva.GJOIevc.",00 ::s crev ",s..>-Eev 0 0 =ua:::t IX ce>...J<.!'eo:::L<.J>z COVER TYPE Appendix Figure A-16.Cover type suitability indices for juvenile coho salmon calculated from 1984 lower Susitna River distribution data. A-3\__________________------_F"'"~._""_----------- - N =Number of Cells Sampled 1983 T 1.00/~,~ 1 I x I C>~UJ I Clz-.I ....·ll ,.-O!J;.....~ >-.... O.A(:)-I "...J-I 0:> I I <C 0.'-"0 ....~.J-......\I~C\N-;:<\a W::'.l.("1 ~ ,'d-=':;~\. VI I .00 ole ~\D I.').lA u.,I.'t,~o ~~45'- DEPTH (ft ) I~Standard Error 0.00 +--..J...-.---r---\.-r--~--"""-'--r---.-------->'-r o ;;l.g:> 0:r 0u I -1 ...Jw \50u ex:.... Cl.. :::ru 1.00.... <C U- Z <C....o.~_:::E: Appendix Figure A-17,Mean catch of juvenile coho salmon per cell by depth intervals (bars)in clear water of the lower Susitna River,1984 and fitted suitability index (line)developed for the middle Susitna River,1983. """", A-32... - ~- - "... ..... - DRAFT/PAGE 18,4/29/85 4/1/85,4/15/85 NUM28/Appendix A were captured in 117 (12.7%)of these cells.Correlations among the habitat variables ranged .to 0.65 in absolute value and velocity was most highly correlated with sockeye catch (Appendix Table A-6).In addition to these cells,partial habitat data were collected at six additional cells and these data are used in subsequent univariate histograms. The distribution of proportional presence by velocity interval was very similar to that found in 1983 (Appendix Figure A-18).There was no use of velocities greater than 1.2 fps,however,and in 1983 there also was no use of velocities greater than 1.2 fps although sample sizes were smaller.Since no use of these velocities has been found,the lower river velocity suitability criteria were modified so that velocities greater than 1.2 fps have 0 suitability (Appendix Figure A-18). Oi stributi on of proportional presence by percent cover categories was similar to that found in·1983 (Appendix Figure A-19).The 1983 suitability relationship was therefore selected for use in 1984 . The distribution of proportional presence by cover type categories was somewhat different than that found in 1983 (Appendix Figure A-20). Suitabilities for the cover types used in 1984 will be those developed in 1984 with the following two exceptions.S"ince sample sizes were small (l ess than 25)for the cover typecategori es,undercut banks and overhanging riparian vegetation,the suitabilities calculated "in 1983 were averaged with the 1984 suitabilities to give a value intermediate between the two. Appendix Table A-6.Kendall correlation coefficients between habitat variables and sockeye catch by cell (N =922). 'W J.....z ""V'l I.:Standard Error "",.."·s'" _.. ~C>- ""-\.<10>-""It a Nulltler of cells sampled><C.-:lC<->0 V'l 1983 o.9lI:z::-.. 1-1984 (revised)"":i ---'"o.l'S :': V'l r ClWJ J >-...::u II .......C.10 '"0 llliIO .., :z:Ji !::0;:,::>,'"'"oDs ' I0!9!o,:ao,C>-o -:1c< C>- l>I:"JAS H~t~t tl'ICI'.l. <>00 0.(;0 CO Q!l 0-"0.1 VaOCIn (ft/sec) Appendix Figure A-lB.Proportion of cell s with juvenil e sockeye sal mon present by veloeity intervals (bars)in the lower Susftna River,1984 and fitted suitabil ity index (line)developed for the middle Susitna River,1983 and revised in 1984 for the lower river using professional judgement. -, ,;J A -34- T PERCENT COVEQ CATEGORI ES Proportion of cells with juvenile sockeye salmon present by percent cover category (bars)in the lower Susitna River,1984 and comparison of fitted suitability indices (lines)calculated in 1984 and for the middleSusitna River,1983._. _. ..... - Appendix Figure A-19. \,bO;...t!- " J 7 ;;1 -,...I i i ~qsl ..: +~- 1'·:1 ""...;!~f-~....iCl:; >-OS)... ~ CI><... S '" G1::lS l!Standard Error 1 N =Number of Cells Sa~led 1983 :=1'84 -,U SOCKEYE 1983 77 ;;SOCKHE 1984 /~N·Number of /:Cells Sampled /) /.1 ~(I V~ j'; "".... Cl Z .. <: £>.. Cl ...,. '".........-g<: <:..:::><D ........... -'""QI£>~£>,. 00 UCI> <:e'"...~............,.0> <T"<:> COVER nn ..<; 0»............'" 0><: .,,<:<:.. "'~.r::...... ..Q.>~o a: <::o...-<:......."'..........'"e .......:> ....>8 ~ -Appendix Figure A-20.Comparison of cover type suitability indices for juvenile socke~e salmon calculated from 1984 lower Susitna River distr1bution data and 1983 middle Susitna River distribution data. I!!I!l!!ORAFT/PAGE 19,4/29/85 4/1/85,4/15/85 . NUM2B/Appendix A No trend was noted in the 1984 depth distribution data and therefore no suitability criteria were fit to these data (Appendix Figure A-21).Of ~ the 20 cells sampled with 0.1 ft depth,fish were sampled in 2 suggest- ing that this depth does have value.Therefore any depth will be assumed to have a suitability of 1.- Composite weighting factor intervals calculated by multiplying cover and velocity suitabilities together were associated with proportional presence of sockeye salmon (Appendix Table A-7). Appendix Table A-7.Proportional presence of sockeye salmon associated with the composite weighting factor calculated by multiplying velocity and cover suitabilities together. Composite Weighting Total Number Proportion rJith Factor Interval of Cell s Fish Present Chi -Square o -0.06 244 0.02 '.2 =55.3 0.07 -0.11 213 0.08 -0.12 -0.19 228 0.17 p<O.OOl 0.20 -1.00 241 0.23 Chum Salmon Juvenile chum salmon suitability criteria were developed by pooling data over gear type and turbidity in 1983.Abundance and distribution data .$t indicate that chum salmon use of lower river side channels is limited by high turbidities (Figure 15).Cell s with turbi diti esgreater than 200 NTU were eliminated from suitability criteria development.Also,since - - - - -!--~ N =Humber of Cells Sampled __1983;, 'W .j;OO' 0.'"x..... C Z-. oMI >-I-.... oJiD -l....co <C ~.I-.... ::::> V) _DEPTH (ft) Appendix Figure A-21.Proportion of cells with juvenile sockeye salmon present by depth intervals (bars)in the lower Susitna River,1984 and fitted' suitability index (line)developed for the middle Susitna River, 1983. - -DRAFT/PAGE 20,4/29/85 4/1/85,4/15/85 NUM2B/Appendix A most chum salmon outmigrate before July 16,only data collected before this date were retained for suitability criteria analysis. The number of cells available for analysis of chum distribution totaled 249 after elimination of the cells outlined above.Chum sa.lmon were captured in 98 (39.4%)of these cells.Correlations among the habitat variables and chum fry catch ranged to 0.32 in absolute value (Appendix Table A-8).Partial habitat data were collected at two additional cell s. The chum salmon distribution by velocity interval in 1984 was very similar to that found in 1983 (Appendix Figure A-22).Therefore,the suitability criteria for chum salmon developed in 1983 was selected for use in 1984. In 1983,the relationship of chum salmon use to both percent cover and cover type was the weakest of any of the four species.In 1984,the 0-5%cover category and the "no cover"type had the highest proportional presence within thei r respective di stri buti ons (Appendi x Figures A-23 and A-24).These data indicate that chum salmon fry do not orient to cover during rearing.Even if velocity suitability is adjusted for,no - qJ real trends in percent cover and cover type utilization were noted,I""'l although large gravel and rubble were used slightly more than was the "no cover"type.$i nce there were no trends,cover type and percent cover will not be used in the 1984 analysis of chum habitat use. - Appendix Table A-B.Kendall correlation coefficients between habitat variables and chum catch by cell (N=249)for all gear types.turbidity below 200 NTU. ~ Percent Cover Cover Type Velocity Depth Chum I"""Percent Cover 1.00 Cover Type 0.13**1.00 Velocity -0.25**0.15**1.00 Depth -0.05 -0.03 0.07 1.00 Chum -0.20**-0.07 -0.04 -0.32**1.00 *Significantly different from 0 at p<0.05. **Significantly different from 0 at p <0.01...... ..... ..... I f-w-I ~Standard Errorz LLJ ~V) LLJ r:>:: 0-0-'z:N=Number of Cells Sampled ::>::x:001 19B3...,-- ::x:,.tJD C oJ. 3:x V)0>-LLJ Q -.J Z -.J .............u >-....f- a 03 .... z ~ a <Xl.....~«....f-.... l5 ca.Vl ::> Q...,VI ae<:c-ao D./1O 0.0 o.~o.b 1ll.Cl I.a./oS ,.e 2.01 .... -VELOCITY (ftjsec) .-. Appendix Figure A-22.Proportion of cells with juvenile chum salmon present by velocity intervals (bars)in the lower Susitna River. 1984 and fitted suitability index (line)developed for the middle Susitna River.1983. A-3'1 I ~Standard Error - >-I-..... ~..... a:lc:e I-..... ::;) <Il ><.....cz.... r,OO 0.00 f</l~N-<:t (Sf -751.)-(7':J-fo) PERCENT COVER CATEGORIES N."",,).'3"7 (6 ::'7..sl.)(v;.·S.,1.) AJ.lIq (o-st.) r- ... I- -,.. I 1 l- I \.", (){it. ",S;) I-z: "...".1'V) "...Ic:::j 0-j :E ::;)O~1Xu X Il-I....D5.~:3 en ~ ~Iw U LL :j0 z 0.... t-o:: ~- 0 I0:: 0-I I".l0-1 ! i D .••J i, I ! 0. 0'+ Appendix Figure A-23.Proportion of cells with juvenile chum salmon present by percent cover category (bars)in the lower Susitna River. 1984 and fitted suitability index (line)calculated for the middle Susitna River.1983. - "'......~l -7.Q I--'! I.:!:.Standard Error N =Number of C 11 S 1 d 0> C C C 0 ~0...,~O>c ...,~s.. ::s ---'-U+>c'"C...,QJ '"U QJQJ QJ ~'"Qla;"'~w'">....s..",....-0 :;:;...,...,.1::'-O>l-'0 '-W.><.0...."'w en>s..",'-QJ u .0 "'C C .o::s .0 ::SO>s..",QJCL QJen <II c'"00 ::s CT<II ",s..>~E<II 0 C ~co uco ox <>-At!)oox .....>z COVER TYPE I e 5 amp e ~T I T 1I~II i I I "rt-"<,.--f-.---I .,.. ! .L T - c-'-- ,.---I -1 +--! I ,t I ..'I'"~M :<rJ ~--JI ~I r<rl II)~'*, ~ I'~II ,I T .1 tl,I ~<-.....<<"<.-"-<I .... r-!I!--I.--:-1 •I l-I I !I c-, !I , I .....!; ..L 1 i j 0.1 -- .... I-z..... V).....<:r: Q.. ::E: ~::cu ::c I-.... 3 V) ...J ...J..... U LL. 0 Z 0.... "..,. I- OX 0 Q.. 0 OXa.. '"'"' - Appendix Figure A-24.Proportion of cells with juvenile chum salmon present by cover type (bars)in the lower Susitna River,1984~ - A-Lf DRAFT/PAGE 21,4/29/85 4/1/85,4/15/85 NUM2B/Appendix A The distribution of chum proportional presence by depth intervals in 1984 was similar to that found in the 1983 studies (Appendix Figure .~ A-25).Since the distributions were similar,the criteria fit in 1983 was used to test for the value of depth in increasing the associations with chum catch.Therefore velocity was first used alone and then with depth to form categories which were associated with chum proportional presence. Although composite weighting factors calculated by velocity alone and velocity and depth together were both significantly associated with chum proportional presence,the composite weighting factor calculated by depth and velocity together seemed to fit better (Appendix Table A-9). Therefore both velocity and depth suitabilities will be used to model chum salmon habitat. Summary A summary table of revisions of the middle river suitability criteria for use in the lower river reveals that abqut half the criteria were not changed or changed only slightly (Appendix Table A-I0).The velocity and percent cover relationships were often not changed while the depth and cover.type criteria have often been modified greatly.Point specific values for all the suitability criteria developed for use in the lower river are presented in AppendiX Table A-II. - - ><.... Cl Z I._ ][~Standard Error N =Number of Cell sSampled 1983 >-I---J--0::0 c>: I--:::l --r-I-1:....,1--,.'----,-1.l...1_,r---...:!I~~O.110 <n ..of rio I.S ~z.z.'1>31 /.t. Ir-o - DEPTH (ft) Appendix Figure A-25.Proportion of cells with juvenile chum salmon present by depth intervals (bars)tn the lower Susitna River,1984 and fitted suitabil ity index (line)deve loped for the middle Susitna River,1983. Appendix Table A-9.Proportional presence of chum salmon fry associated with several composite weighting factors. Composite Composite Proportion Weighting Weighting Total With Factor Factor Number Fish Calculation Interval of Cells Present Chi-Square Velocity o -0.55 49 0.20 x.~=34.3 0.60 -0.81 51 0.49 p<O.OOl 0.86 82 0.24 0.93 -1.00 69 0.64 Velocity*Depth 0 -0.32 71 0.10 0.34 -0.49 54 0.43 '-2 =46.8 0.50 -0.73 60 0.42 P <0.001 0.76 -1.00 66 0.67 -- A u.'-',-IJ--_._-,------ DRAFT/PAGE 1 4/2/85,4/29/85 NUM2/Table A-8 Appendix Table A-I0.Summary of reV1Slons of 1983 middle river juvenile salmon criteria for use in the lower Susitna River, 1984. ~ Spe<;ies Velocity Percent Cover Cov~r'Type Depth '"'" Chinook Turbid chinook Same as 1983 Modified Modified (clear)criteria developed in 1983 used Chi noo,k Same as 1983 Same ~s 1983 Modified Modified (turbi d) Coho Same as 1983 Same as 1983 Modified Same as 1983 Sockeye Modified Same as 1983 Modified Modified Slightly Slightly Slightly Chum Same as 1983 Modified Modified Modified (Set to 1.0)(Set to 1.0)- - 1 )})1 'J -l J -1 )J 1 1 .'.....I l }j 1DkAr-}/PAbt • 4/2/85,4/15/85 NUM2/Table A-9 Appendix Table A-II.Suitability indices for juvenile s~lmon for velocity,depth,and cover in the lower Susitna River,1984 .. VELOCITY . Chinook Coho Sockeye Chum Velocity.'Suita-Velocity Suita-Velocity Suita-Velocity Suita- (ft/sec)bility (ft/sec)bi 1ity (ft/sec)bil ity (ft/sec)bi 1ity 0.00 0.42 0.00 0.29 0.00 1.00 0.00 0.86 0.05 1.00 0.05 1.00 0.05 1.00 0.05 1.00 0.35 1.00 0.35 1.00 0.20 0.71 0.35 1.00 0.50 0.80 0.50 0.88 0.50 0.48 0.50 0.87 p 0.80 0.38 0.80 0.55 0.80 0.35 0.80 0.70 I 1.10 0.25 1.10 0.32 1.10 0.14 1.10 0.56 -,\:1.40 0.15 1.40 0.12 .1.30 0.00 1.40 0.37 0,1.70 0.07 1.70 0.04 1.70 0.15 2.00 0.02 2.00 0.01 2.00 0.03 2.30 0.01 2.10 0.00 2.10 0.00 2.60 0.00 DEPTH Chinook (turbid)Chinook (clear)Coho Sockeye Chum Depth Suita-Depth Suita-Depth Suita-.Depth Suita-Depth Suita- (ft)bil ity (ft)bi 1ity (ft)bil ity (ft)bil ity (ft)bi 1ity 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.10 0.29 0.15 0.00 0.14 0.00 0.10 1.00 0.10 1.00 . 0.30 1.00 0.20 0.25 0.15 1.00 10.00 1.00 0.50 1.00 1.50 1.00 1.50 0.25 10.00 1.00 0.80 0.68 1.80 0.33 1.80 0.80 1.30 0.50 10.00 0.33 2.10 1.00 1.80 0.38 10.00 1.00 10.00 0.38 Appendix Table A-II (Continued) DRAFT/PAGE 2 4/2/85,4/15/85 NUM2/Table A-9 Percent Chinook Chinook Cover Type Cover (turbid)(clear)Coho Sockeye Chum No cover 0-5%0.15 0.01 0.00 0.18 1.00 Emergent Vegetation 0-5%0.23 0.11 0.05 0.39 1.00 6-25%0.30 0.33 0.14 0.54 1.00 26-50%0.33 '0.55 0.24 0.70 1.00 51-75%0.39 0.78 0.33 0.85 1.00 76-100%0.40 1.00 0.42 1.00 1.00 Aquatic Vegetation 0-5%0.23 0.10 0.04 0.23 1.00 _'"f:'::,.6-25%0.30 0.32 0.13 0.32 1.00 I 26-50%0.33 0.53 0.21 0.41 1.00-c 51-75%0.39 0.76 0.30 0.50 1.00d-76-100%0.40 0.97 0.38 0.59 1.00 Debris or Deadfall 0-5%0.15 0.05 0.08 0.21 1.00 6-25%0.20 0.17 0.24 0.29 1.00 26-50%0.20 0.28 0.39 0.37 1.00 51-75%0.20 0.39 0.55 0.45 1.00 76-100%0.20 0.50 0.70 0.53 1.00 Overhanging Riparian 0-5%0.15 0.04 0.07 0.25 '1.00 Vegetation 6-25%0.20 0.13 0.20 0.34 1.00 26-50%0.20 0.21 0.33 0.44 1.00 51-75%0.20 0.30 0.46 0.54 1.00 76-100%0.20 0.38 0.59 0.63 1.00 Undercut Banks 0-5%0.23 0.11 0.12 0.25 1.00 6-25%0.30 0.33 0.34 0.34 1.00 26-50%0.33 0.55 0.56 0.44 1.00 51-75%0.39 0.78 0.78 0.54 1.00 76-100%0.40 1.00 1.00 0.63 1.00 """)..,......)J ....]•••,..-.,...)"......1 J •J J I I }»-1 j 1 1 ).)1 1 J 1 ORAl"1'/PAbt .]}1 } 4/2/85,4/15/85 NUM2/Table A-9 Appendix Table A-II (Continued) Percent Chinook Chinook Cover Type Cover (turbid)(clear)Coho Sockeye Chum Large Gravel (1-3 11 )0-5%0.15 0.02 0.02 0.18 1.00 6-25%0.20 0.08 0.06 0.24 1.00 26-50%0.20 0.13 0.10 0.32 1.00 51-75%0.20 0.18 0.14 0.38 1.00 76-100%0.20 0.23 0.18 0.45 1.00 Rubble (3_5 11 )0-5%0.15 0.03 0.02 0.18 1.00 6-25%0.20 0.10 0.06 0.24 1.00 26-50%0.20 0.17 0.10 0.32 1.00 :t:'"51-75%0.20 0.23 0.14 0.38 1.00 76-100%0.20 0.30 0.18 0.45 1.00 I ·"r~Cobble or Boulder 0-5%0.15 0.03 0.02 0.18 1.00--.J (>5 11 )6-25%0.20 0.11 0.06 0.24 1.00 26-50%0.20 0.18 0.10 0.32 1.00 51-75%0.20 0.25 0.14 0.38 1.00 76-100%0.20 0.32 .0.18 0.45 1.00 - DISCUSSION Chinook Salmon """1 DRAFT/PAGE 24,4/29/85 4/1/85,4/15/85 NUM2B/Appendix A - The turbid water velocity criteria developed in 1983 were used for both clear and turbid chinook distributions in the lower river in 1984.The reason that there was no shift in velocity optimum from clear to turbid water may be due to several factors.In the middle river,substrate is much 1a rger and therefore,on the average,chi nooks may fi nd hi gher velocities as suitable as there is always some substrate cover to hide - under or behind.In the lower river,however,very little substrate cover is 'present and therefore chinook use lower velocity water much ..... more. In the lower river,cover suitabilities were also found to be somewhat different.Part of this difference may be due to the actual cover in cover type categories being of a different type.For instance,the aquatic vegetation in Caswell Creek which harbored large numbers of chinook fry was not present in any of the sampled streams in the middle river.Also the debris cover type in the lower river was often much more silted in than in the middle river and therefore less suitable. The primary cover type is associated with a variety of secondary cover types and it is likely that,on the average,secondary cover types associ ated with a prima ry cover type in the lower ri ver are di fferent than the secondary cover types most common in the mi ddl e ri ver.If these secondary cover types are more suitable for fi sh then they mi ght raise the suitability of the primary cover type.- DRAFT/PAGE 25,4/29/85 4/1/85,4/15/85 NUM2B/Appendix A - - -- - Most notable in the analysis of chinook suitability criteria was the effect of depth upon the distribution of chinook salmon.In clear water in the lowet river,chinook salmon found deep water much more suitable (Appendix Figure A-7).This is probably due to the tributaries in the lower river having a turbidity of approximately 10 to 20 NTU and there- fore depth might have a cover value in deeper waters.In the upper river much of the data was collected in Portage Creek and Indian River and other areas where the turbidity was usually less than 5 NTU and depth would not provide cover at depths which can be sampled.Sometimes/-----~------------------~. juvenile salmon thought to be chi.nook fry could be seen feeding on 'the surface in tributary mouths such as Rolly Creek where depths were greater than 5.0 ft. Inturbi dwater,on the other hand,depths greater than 1.5 ft were less suitable than shallower cells (Appendix Figure A-ll).This trend was also found in 1983 although discounted at the time.This difference may be due to fish reacting to high suspended solid concentrations by staying near the surface (Wallen 1951 as cited in Beauchamp et al. 1983).It also could be due to fish not being able to feed at depths where there is very little light,whereas in shallower water a small amount of light may enable fish to feed. Coho Salmon The suitability criteria developed for coho salmon juveniles in the middle river were modified only slightly in cover suitability for use in the lower reach.The fit of the data to the composite weighting factor A I",1_-·...,;1..,-1 }J L ---~---.........._--------------------------------- DRAFT/PAGE 26,4/29/85 4/1/85,4/15/85 NUM2B/Appendix A was not very high (r=O.33)however,which suggests that coho respond to other factors than those studied.These factors might include food supply or seasonal movements. Sockeye Salmon Since sockeye normally rear in lakes,it is not surprising that velocity is one of the most important variables affecting their distribution.In both the lower and upper river,no sockeye have been captured in cells with velocities greater than 1.2 ft/sec.The highest catches of sockeye were made at Beaver Dam Slough,which is always a backwater with minimal velocity. Instream cover does have some importance in juvenile sockeye salmon distribution and it appears they also use turbidity as cover (Section 3.2.4).Cover type suitabil ities were somewhat different in the lower reach than in the mi ddl e reach,perhaps due to differences in the primary or secondary cover type within the categories between the two reaches. Chum Salmon Chum salmon,in contrast to the other species,did not show any positive response to the presence of cover.The response shown,which is a negative one,is probably partly a function of gear efficiency.They did respond to velocity and depth,however.The lack of relationship A',-A .\""'''-.......v' - - - JP!'Il, - -- -- - DRAFT/PAGE 26.1~4/29/85 4/1/85 t 4/15/85 NUM2B/Appendix A with cover may partly be a function of schooling behavior which reduces the need for cover.It is also possible that since chum fry rear in fresh water for only a short period they usually are searching for food instead of hiding in cover. The heavier use of shallower depths by chum juveniles found in both years is due to unknown factors.This could be due to a use of shallow depths and low velocities in side channels where some of the suspended solids may settle out.Perhaps these areas also are somewhat warmer than adjacent areas as the sunl i ght strikes the substrate and ;s absorbed heating the water above. DRAFT/PAGE 28,4/29/85 4/1/85,4/15/85 NUM2B/Appendix A LITERATURE CITED Beauchamp,D.A.,M.F.Sheperd,and G.B.Pauley.1983.Species pro- files:Life histories and environmental requirements (Pacific Northwest)chinook salmon.U.S.Department of the Interior,Fish and Wildlife Service.FWS/OBS-83/1. Bovee,K.D.1982.A guide to stream habitat analysis using the instream flow incremental methodology.Instream Flow'Informati on Paper.No.12.U.S.Fish and Wildlife Service.FWS/035-82/26. E.Woody Tri hey and Associ ates and Woodward-Clyde Consul tants.1985. Instream flow relationships report.Vol.1.Working Draft. Alaska Power Authority Susitna Hydroelectric Project.Report for Harza-Ebasco Susitna Joint Venture,Anchorage,AK.1 vol. Hale,S.S.,P.M.Suchanek,and D.C.Schmidt.1984.~1odelling of juvenile salmon and resident fish habitat.Part 7 in Schmidt, D.C.,S.S.Hale,D.L.Crawford,and P.M.Suchanek (ed5:").1984. Resident and juvenile anadromous fish investigations (May -October 1983).Susitna Hydro Aquatic Studies.Report No.2.Alaska Department of Fish and Game.Anchorage,Alaska. Marshall,R.P.,P.M.Suchanek,and D.C.Schmidt.1984.Juvenile salmon rearing habitat models.Part 4 in Schmidt,D.C.,S.S.Hale,D.L. Crawford,and P.M.Suchanek,(ed5:'l.1984.Resident and juvenile anadromous fish investigations (May -October 1983).Susitna Hydro Aquatic Studies.Report No.2.Alaska Department of Fish and Game.Anchorage,Alaska. Nie,N.H.,C.H.Hull,J.G.Jenkins,K.Steinbrenner,and D.H.Bent. 1975.Statistical package for the social sciences.2nd ed. ~McGraw-Hill Book Co.,New York,USA. Suchanek,P.M.,R.P.Marshall,S.S.Hale,and D.C.Schmidt.1984. Juvenile salmon rearing SUitability criteria.Part 3 in Schmidt,. D.C.,S.S.Hale,D.L.Crawford,and P.M.Suchanek (ed5:").1984. Resident and juvenile anadromous fish investigations (May -October 1983).Susitna Hydro Aquatic Studies.Report No.2.Alaska Department of Fish and Game.Anchorage,Alaska. Wallen,I.E.1951.The direct effect of turbidity on fishes.Doctoral dissertation.University of Michigan,Ann Arbor,Michigan,USA. A-57---... - - - DRAFT/PAGE 1 4/29/85 NUM2B/Appendix B - APPENDIX B ,~ MODELLED SITE TURBIDITIES,JUVENILE SALMON CATCHES,AREAS,SIDE CHANNEL FLOWS, WEIGHTED USABLE AREAS,AND HABITAT INDICES- -- ........_--,-,----------------------_........._--------- ·""'" - ,~ ""'" • -. DRAFT/PAGE 2 4/29/85 NUM2B/Appendix B ,,-, ..- This appendix is a compilation of data arranged into a number of graphs and tables.The first three tables (Appendix Tables B-1,B-2,and B-3) present:modelled side channel turbidities;modelled site catches and CPUE's of juvenile salmon;and lengths of RJHAB model sites;respec- tively.Appendix Table B-4 presents modelled side channel flows as a function of mainstemdischarge at 3,000 cfs increments. Next weighted usable areas and habitat indices are presented by species in the following order: Chinook Salmon Tabulation of weighted usable areas and habitat indices for 18 sites (Appendix Table B-5). Graphs of weighted usable area versus rna instem discharge for si tes not presented in Section 3.3: -- - - Caswell Creek Mouth Beaver Dam Slough Hooligan Side Channel Bearbait Side Channel Last Chance Side Channel Rustic Wilderness Side Channel Island Side Channel Mainstem West Bank Goose 2 Side Channel 13 -I (Appendix Figure B-1) (Appendix Figure B-1) (Appendix Figure B-2) (Appendix figure B-2) (Appendix Figure B-3) (Appendix Figure B-3) (Appendix Figure B-4) (Appendix Figure B-4) (Appendix Figure B-5) Circular Side Channel Sauna Side Channel Bearbait Side Channel Sunset Side Channel Sunrise Side Channel Trapper Creek Side Channel Coho Salmon DRAFT /PAGE 3 4/29/85 NUM28/Appendix 8 (Appendix Figure 8-5) (Appendix Figure 8-6) (Appendix Figure B-6) (Appendix Figure 8-7) (Appendix Figure 8-7) (Appendix Figure 8-8) - Tabulation of weighted usable areas and habitat indices for three sites (Appendix Table 8-6). Chum Salmon Tabulation of weighted usable areas and habitat indices for 15 sites (Appendix Table 8-7). Graphs of weighted usable area versus mainstem discharge for sites not presented in Section 3.3: Hooligan Side Channel (Appendix Figure 8-9) Kroto Slough Head (Appendix Figure 8-9)""" Bearbait Side Channel (Appendix Figure 8-10) """'1 Island Side Channel (Appendix Figure 8-10) Mainstem West Bank (Appendix Fi gu re B-11 ) Goose 2 Side Channel (Appendix Figure 8-11) Circular Side Channel (Appendix Fi gure 8-12)- .,-. 6-i.. - Sauna Side Channel Sucker Side Channel BeaverDam Side Channel Sunrise Side Channel Sockeye.Salmon DRAFT /PAGE 4 4/29/85 NUM2B/Appendix B (Appendix Figure 8~12) (Appendix Figure 8-13) (Appendix Figure B-13) (Appendix Figure B-14) ,....Tabulation of weighted usable areas and habitat indices for seven sites (Appendix Table 8-8). Graphs of weighted usable area versus mainstem discharge for sites not presented in Section 3.3: ,~ Caswell Creek Mouth Beaver Dam Slough (Appendix Figure B-15) (Appendix Figure B-15) '\(Nil)) DRAFT/PAGE 1 4/11/85,4/29/85 NUM2/Tab 1e B-1 Appendix Table B-1.Turbiditiesi\within modelled side channels of the lower Susitna River,June through August,1984.Values within parentheses~ere calculated by inputting the overall mean for all the side channels during a given two week period. Site June 1-15 June 16-30 July 1-15 July 16-30 Aug 1-15 Aug 16-30 Mean West Bank Side Channels /-~-._-'--.......<':~> Kroto Side Channel (64)394 (369)Q72,70':!)784 126 388 Bear Bait Side Channel (64)392 284 ---~f2 328 142 ,-,254 Mainstem West Bank (64)(227)(369)368 324 ,r •.32.!t..279 Sauna Side Channel 120 (227)496 364 244 (J~~,.?56:266 Trapper Side Channel 96 576 940 470 306 608 499 Middle Side Channels Hooligan Side Channel (64)365 288 296 704 544 377 Last Chance Side Channel (64)(227)296 672 352 576 365 Island Side Channel 55 126 334 336 228 r~(209)215 Circular Side Channel 89 122 592 288 216 08,304/241 Sucker Side Channel 26 64 276 118 292 ~44,163)140 Sunrise Side Channel 18 112 180 88 280 4'4,J2~'121 OJ I East Bank Side Channel ,,[Rusti c Wi 1derness Side Channel (64)120 130 160 196·38 118 Coose Side Channel 41 140 384 300 188 i 64,244')194 Sunset Si de Channel (64)(227)(369)114 100 \.,,4J.!lJ6J 152 Beaver Dam Side Channel (64)90 224 134 170 /150 139 OVERALL MEAN 64 227 369 312 314 /209 ! _,_i r ) i ! 0../\,( 0--(t C;:,\ __,J ,I ,I J ,)1 • )J -B )I )))»J )J )1 1 1 1 Appendix Table 8-2.Catch and catch per cell (CPUE)of juvenile salmon within lower Susitna River sampling sites,1984.Cells have been standardized to an area of 300 ft 2 . I~D.~.l·t c"d Is Chinook Coh,:!Chum Lockeyo Chinook Coho Chum So(:keye 31 t.~sampled Col tel.C.at.:tl catch catch CPUI.'::CPUE CPlJE Cf'lJL ----_._----_._--------.--------------,-0_--"._-_...~.........._......----....,_....._......----.,-.--..-._..-_._._.__._-- Hooligan Side Channel "77 21 0 70 3 0.27 O.(1I)1.01 0.04 Eagles Nest Side Channel 30 5 0 0 0 0.17 0.00 0.00 0.00 (),}Kroto Slaugh Head 56.5 4 0 1 -,0.07 0.00 0.02 0.(14... \Rolly Creek Mouth 91 53 39 ,-,87 0.58 0.43 0.02 0.96.. 0\Bearbait Sida Channel 49.4 4 0 3 0 1).08 0.00 0.06 0.00 Last Chance Side Channel 50 (I 0 1 (I 0.00 0.00 O~O2 0.00 Rustic Wilderness Side Channel 65 55 1 11 0 0.85 0.02 O.17 0.00 Caswell Creek "Iouth '74 419 24.5 0 21 5.6¢>:3.:::.~1 0.00 0.28 Island Side Channel 82 :59 1 74 2 0.4B 0.1)1 0.90 0.02 f1ainstem West Sank 45 7 0 (I 1 0.16 0.00 0.00 0.02 Goose 2 Side Channel 82 7il-1 30 -,0.90 0.01 0.37 0.02""- Circular Side Channel 88 28 0 114 6 0.32 0.00 1.30 0.07 Sauna Side Channel 44 :5 0 41 5 0.07 0.00 0.93 0.11 Sucker Side Channel 77.1 e>'"0 112 15 0.30 0.00 1.45 0.19...~ Beaver Dam Slough 83 14 67 (I 101 0.17 0.81 0.00 1.22 Bea.ver Dam Side Channel 102 153 9 23 71 1.50 0.09 0.23 0.70 Sunset Side Channel 73.5 121 0 I)1 -:I 1.65 0.00 0.00 O.16.... Sunrise Side Cha.nnel 73 120 1 43 8 1.64 0.01 0.59 0.11 Birch Creek Slaugh 96 2:;:;''7t 45 29 0.24 0.74 0.47 0.30 Trapper Creek Side Cha.nnel 96 4:~\2 20 4 0.45 0,.02 0.21 0.04 SUEnOTHL 1434.5 1209 437 598 369 0.84 0.30 0.42 0".26 Opportunistic sites 16:5.7 249 5 10 43 1 !=.I'''J O~O3 0.06 0 4 26.;:,..::. TOTAL 1598.2 1458 442 608 412 0.91 0.28 ().38 0.26 DRAFT/PAGE 5 4/29/85 NUM2B/Appendix B Appendix Table B-3.Lengths of RJHAB model sites in the lower Susitna River,1984.- Site Hooligan Side Channel Eagle's Nest Side Channel Kroto Slough Head Rolly Creek Mouth Bearbait Side Channel Last Chance Side Channel Rustic Wilderness Side Channel Caswell Creek Mouth Island Side Channel Goose 2 Side Channel Sucker Side Channel Beaver Dam Slough Beaver Dam Side Channel Sunrise Side Channel Birch Creek Slough Trapper Creek Side Channel Length (feet) 1377 490 748 1437 496 961 1169 712 769 1030 658 436 608 1003 841 968 - - - ""'" - ]1 )1 j J )j ]J '~J 'J )1 1 Appendix Table 8-4.Side channel flows at the 15 modelled side channels in the lower Susitna River aSI function of mainstem discharge.1984.Flows calculated from rating curves presented in Quane et al.(1985).Flows marked with an asterisk were not reliably modelled according to hydraulic criteria (Appendix D).Discharges for which NA is presented for flow are "not available"because rating curves were not developed at these discharges. HoaLl GMJ S.Co KROTO SLDUGH HEAD BEARBAIT SIDE CHANNEL LAST CHANCE S.C.RUSTIC WILDERNESS S.C. ------------------~-------------------~------------------------------------------ MINSWI SITE SITE SITE SITE SITE l)!SCH{\RSE flREH FLOW AAEli FLOW AREA FLOW AREA FLOfJ AREi!FLOW 1200(1 63400 0 48200 0 3100 0 17500 (l 4800 0 15(1)(1 6340(J lj 48200 0 3100 0 17500 0 4!WO I) 1800')63400 0 48200 (I 3100 (I 17500 (I 4800 0 txt 21000 63400 0 48200 (I 3100 (I 17500 (I 31900 54 I 24000 71800 50 48200 0 3100 (I 17500 1 49500 76 -:\27('00 8690(i n 48200 (I 3100 0 31700 3 60700 103. 30000 90800 100 4B20(l 0 3100 0 50600 :5 69700 134 3300(i 96500 135 48500 0 3100 \)63900 a nBOO 171 360no 104800 178 57900 0 5700 33 73200 13 83300 213 31j'(ItlO 113700 229 67900 74 10800 48 80000 21 89900 261 42000 122900 2aB 7750(1 98 ,14600 67 85900 31 97000 315 45000 131300 358 86800 128 17900 93 9(1600 46 104000 375 48000 141200 •439'95100 163 21100 125 94000 67 109000 442 51000 J52oo0 531 102200 206 23800 Ibb 96300 95 114000 516 54(01)163000 636 10b700 255 26400 217 98500 131 117400 596 57000 174100 753 110200 314 29000 279 100200 178 119200 bB4 60000 186800 885 113500 381 31500 354 101800 238 120700 779 bJi)OO 200800 10J2 116600 459 33900 445 103200 314 121700 NA MOOO 213300 1194 119000 547 36300 552 104400 408 122200 NP: 69(it{1 226000 1373 120100 648 38300 NA 105500 526 122700 riA 72000 239(tOO 1570 121000 761 40000 Nfl 106300 669 123000 riA '15000 250~O(l 1785 121400 88~41500 NPI 107000 B44 123500 NA Appendix Table 8-4.Continued. ISLAND SIDE CHANNEL HAINSTEti WEST FANr~600SE 2 SIDE CHANNEL CIRCULAR SIDE CHANNEL SAUNA SIDE CHANNEL ---~-~~-----------------------------~-------------------------*---~-------------- M/mSTE~l SITE SITe SITE SITE SITE DISCHARGE AREA FLOW AREA FLOW AREA FLOW AREA FLOW AREA FLOW 0 12000 31500 6 61603 6 (I 0 59~64 6 42093 5 15000 31500 6 61603 b (I (l 59464 6 42093 5 CAt 18000 31500 6 61603 6 0 I)59464 6 42093 5 I 210(1(1 315Ci(i b 73426 19 (I (I 59464 b 420'13 5 0<)24000 31500 II B0904 53 I)0 59464 6 42093 5 27(l()0 31500 6 93353 13~0 (I 51464 6 42093 5 30000 31500 II 106613 3(1]9600 0 594b4 6 42093 5 33000 31500 6 114738 470 21500 24 59464 b 41611 14 3600(1 44600 6~'117696 559 34300 32 71590 27 48790 19 39000 48100 94 120505 6<;7 47800 41 76534 38 49127 21u, 42000 53200 126 123397 762 01400 52 80557 54 49758 25 45000 58900 166 1292'11 674 72(f(lO 65 85140 73 50289 ?"'..7 48000 65500 215 133649 995 81400 81 92944 98 50889 34 51000 72000 273 136885 1123 *"87800 98 102530 129 51451 39 541)00 79400 342 140761 InCI *'93200 118 113323 167 52011 44 57000 86700 424 144269 1404 *'971(1)141 125753 213 52678 50 60(10(1 93100 520 147899 155~;..99900 166 134218 268 53294 56 63(JOO 99800 631 151842 1715 *'102000 195 143575 334 54275 63 66000 106200 7SB 154205 1882 *1(13200 226 150869 412 *55184 70 69(101)111900 904 156425 20St'it HI4200 261 154657 503 *56053 77 i20(1(i I1B200 1070 *158522 2241 f.104800 300 157074 610 *57142 135 7500(1 123300 1258 1.160818 2431 '*105100 342 1592H 733 it 61018 93 J I J ~l ,j ~.J J ••I »)] 1 J ]1 ))1 1 )1 -1 .J 1 1 Appendix Table B~4.Continued. BUCKER51DE CHANNEL BEAVER DAn SIDE CHANNEL SUNSET SIDE CHANNEL SUNRiSE SIDE CHANNEL TRAPPER CREEK S.C. ---------------------------------------------------------~-----------~----------- MINSm1 SITE SITE SITE SITE SITE DISCHARGE AREA FLOW AREA FLOW AREA FLOW AREA FLOW AREA FLOW 120(1)i)(I IB900 <1 49562 5 0 0 73300 14 15000 (I 0 1B900 (1 49562 5 I)0 73300 14 b:j 18000 0 \)1B900 {1 49562 \')(I (j 73300 14 I 21000 (I 0 IB900 {1 49562 5 0 Ii 73300 14 :...0 24000 l)0 1&900 {1 49562 5 0 0 7330(1 14 2700v 1600 0 18900 (1 64118 15 l'i (I 0 73300 14 3000(1 8500 13 18900 <1 69129 25 *(I 0 73300 14 33000 H900 18 18100 {1 784aB 47 0 ()73300 14 36000 16900 24 18900 <1 89472 bS 19000 19 75600 26 39(100 19400 31 16900 {l 91943 96 53900 29 85100 28 42000 23600 39 18900 {1 106320 132 7B500 41 97100 30 450(10 29600 48 19900 <1 122338 178 97100 58 10B700 39 48000 37100 59 22~OO 7 135476 235 115400 79 ImOO 72 51(100 4/:.600 71 28000 11 14924B 305 131100 10&128900 129 54000 57900 86 32600 IB 165990 390 146900 139 137400 221 57000 6&900 101 3570(1 29 173483 492 160600 181 143300 370 bOOM 71300 119 38000 45 18S419 614 175600 233 148600 564 63000 H900 139 39600 68 194419 757 192000 295 154fWO 683 b6(IO(1 75900 161 408(10 101 2030iJO 925 207300 370 1b0700 819 69000 77300 185 4150(~148 20im2 1119 t 22140(1 .459 166\00 975 t nooo 78100 211 41900 213 21072B 1345 *229000 564 169800 1151 i 75000 78300 240 42100 ;)02 215661 1603 *23330f)688 172600 1351 i Appendix Table 8-5.Weighted usable areas and habitat indices for juvenile chinook salmon in lower Susitna River model sites,1984. lIIIO!l JIOlII..16All SIDE CllAlllIEl ~P.D10 SlOU6H HEAD BEAReAIT sm CHAhHEl -------_..--------...------------------------_...._--------------------------_.....--------...--....---------- Ml~STE"SITE CHINOOK CltlliOOK ItAIHSTU SITE CHlffOOK CHIHoa.:""'"1Iil1llSTEIl SITE CHINOOK CHIIiOllK DISCHARSE AREA WIlA H•.1.CISCHARSE AREA i1UA H.I.DISCHARGE AHEA IiUil H.1. 12000 &J400 500 O.IH 12000 48200 100 .00 1211W 3!O'J 20 0.')1 1500(,\moo 500 0.01 15000 48200 100 .00 15000 HOii Ll)V.v1 18000 &3400 500 0.01 18000 4a100 Il'O •',Iv 18\100 310,'~':''J.'J1 # 21000 63400 500 0.01 210(1)4820(1 WO .00 2H'OfJ 3i(1l)Z{)O.VI 24000 19Bi.10 7il60 0.10 24000 48200 100 .01)2400Q 3100 20 0.01 21000 810900 7200 o.oa 21000 49200 100 .')'1 27000 310(}it'Q.OI 30000 90800 10100 O.Oi 3(001)45200 100 .00 30000 3100 20 v.Ol 33000 910500 6100 0.010 33000 48500 100 .00 33000 3100 20 0.01 36000 104800 5500 0.05 310000 57900 2700 U5 31>000 5100 200 0.04 39000 Hmo 4900 0.04 39000 moo 4800 fj.f)7 moo 10800 350 0.03 .2000 122900 4200 0.03 42000 77500 10200 0.08 42000 141>00 530 0.04 4SOOO \:tUOO 3600 0.0:1 45000 861100 7300 0.08 45000 11900 1050 0.04 48000 141200 2'100 0.02 ,fSOOO 95100 8100 0.09 48000 21100 720 0.03 51000 mooo 2200 0.01 51000 102200 7901l 0.08 51000 23800 790 0.03 54000 11>3000 2000 0.01 54000 1010700 6'100 0.06 54000 26400 800 0.03 51000 114100 2000 0.01 57000 110200 &000 0.05 57000 29000 750 0.03 60000 186800 1900 0.01 100000 113500 5100 0.04 &0000 31500 700 0.02 63000 200800 1800 0.01 63000 1110600 4300 0."4 63000 33'100 650 0.02 6&000 213300 1800 O.O!&&000 119000 3400 0.03 66000 3&300 1010 0.02 69000 226000 1800 0.01-69000 121HOO 2900 0.02 109000 3130(1 590 0.02 72000 mODO 1800 0.01 72000 121VOO 2500 0.02 72000 40000 5iO 0.01 15000 250'100 1800 O.ul ~ooo 121400 2200 0.02 15000 41500 560 0.01 B-Io ~ ....T£1l IIEST ..IOlISl 1 51.CIIMIlft tlllCllLM SlDE ~ _'0".______________......._.._...-------...-------_._._----_._-_..._---_....__...------.....--------_.._-_._------ itA lttS f Ell !tTE tlllNlJllt:I:IUIIlIIlII ItltlHSTEIl SHE CIUNOOK CIUHOllK IIAINSTEIt sm CHINOOK CltINOllt: nSCHARGE AREA llllA H.I.DISCHAIlG£AREA WIlA H.I.DISCHARSE AREA Willi H.I. 12000 61/103 1082 0.02 (2001)0 0 0.00 12000 :i9~o4 747 0.01 I~OOO 61603 1082 0.02 15000 0 a 0.00 15000 5'1U4 Hi 0.01 18000 01603 1082 0.02 laooo a 0 0.00 18000 S94H 747 0.(11 21000 73426 10041 0.14 21000 0 0 0.00 210UO 594114 747 ("'01 240(1)80904 8m 0.10 24000 I)0 0.00 24000 59404 747 0.01 21000 93353 5224 0.06 27000 0 0 0.00 27000 594H 741 0.01 3ססoo HlSIII3 4045 0.04 ~000 '1600 Isao 0.10 30000 S'I4b4 H7 v.Ol 33000 114738 3959 0.03 33000 21500 2'100 0.13 33aoo SH04 747 o.al 3.'l000 U11090 3S01 0.03 36000 34300 4000 0.12 361100 m90 8717 0.12 moo 120SO~ms 0.03 39000 41800 :il00 0.11 ·3%00 76534 8404 0.11 42000 123m 385~0.03 42000 61400 blOO 0.10 42000 80557 8013 0.10 45000 129211 4UJ 0.03 .45000 12000 6900 0.10 45000 85140 1472 0.0'1 48000 ·13364'1 %30 0.03 48000 81400 1000 0.09 48000 92944 7077 0.08 51000 13b88S 50S0 0.04 :11000 B7BOO 6700 0.08 51000 102530 0998 0.07 54000 H01lt!5554 0.04 54000 93200 6000 0.06 54000 113m 10m O.UIt 57000 1«2109 (,211 0.04 57000 97100 4600 0.05 51000 125753 0634 0.05 60000 1478'/9 ons 0.05 60000 9'f900 3100 0.03 60000 134218 1>516 0.05 63000 l~l842 7092 0.05 maG 102000 2700 0.03 UOOO 143575 6'106 0.05 6/1000 154205 7598 G.05 60000 103200 2400 0.02 610000 ISOB09 7926 0.05 ~6'1000 \5&425 7913 0.G5 69000 104200 2100 0.02 109000 154651 SS6l 0.0& 72000 15S522 S078 0.05 72000 104800 lSOO 0.02 72000 157074 88~O 0.06 75000 lD081S 9438 0.05 moo 105100 1(,00 0.02 75000 \59211 8154 0.06 ..... B-1! Appendix Table 8-5.Continued. - SAUIIA SIDE CHANNEL sucm SIDE ClWlHEL SlAVER 01\11 SI DE CHANNEL .._..-..----------_.._....._-------------------...-----------..----------_..----..-------------...-_.._--------------- MIIlSTEIl SITE CHINOIJl(CHlNlllll ftAINSTEIl SHE t1I111011l CHINOllX llAilISTEIl SHE CHIIlOOK CHINOO~ DISCHARGE AREA WUA H.I.DISCHARGE ARE':'IIUA H.L DISCHARGE AREA 11M H.J. 12000 42093 165 .00 12000 0 0 0.00 I~OOO 18900 50 ,.00 150%42093 1115 .00 15000 0 0 0.00 15000 18900 50 .00 18000 42093 165 .00 18000 0 0 0.00 10000 t8900 50 .00 21000 42093 165 .00 21000 0 0 0.00 21000 t8900 50 .00 24000 42093 1~5 .00 24000 0 0 0.00 24000 18900 50 .00 27MO 420'i3 1115 .00 21000 bOO 0 0.•00 27000 18900 SO .00 lOGOO 42093 165 .00 30000 8500 1060 0.12 lOOOO 18900 ~.00 33000 47611 5410 0.11 33000 14900 1000 O.ll 3lO00 18900 50 .00 30000 48190 5113 0.12 36000 16'100 1570 0.09 36000 t8iOO 50 .0(1 39000 49121 5159 0.12 39000 19400 1510 0.08 39000 18900 SO .00 4Z000 49758 5140 0.12 42000 73600 1450 0.06 42000 18900 50 .00 45000 50289 5503 0.11 45000 29600 1550 0.05 45000 moo 50 .00 48000 50889 4980 0.10 48000 37100 2010 0.06 48000 22400 B20 0.04 51000 5145t 4470 0,09 51000 41MlOO 2940 0.06 51000 28000 2310 0.0fl ~54000 52011 4&46 0.08 54000 51'l00 4210 0.07 54000 moo 3560 0.11 57000 5267S 3645 0.07 57000 &b900 4680 0.07 57000 35700 3S40 O.H 60000 5329~3365 0.06 60001>11300 44'10 0.06 110000 38000 3570 0.09 63000 54275 3116 0.06 63000 73900 4230 0.06 63000 moo 3060 0.09 66000 55184 2947 O.O~66000 moo 3940 0.05 MlOOO 4&800 2510 0.011 ~ 69000 Sb053 2757 0.05 69000 7i300 3610 0.05 69000 41500 22[,0 0.05 72000 57142 2678 0.05 120000 moo 3271>0.04 72000 41'100 2100 0.05 75000 61019 2714 0.04 15000 78300 ·3010 0.04 75000 42100 2000 0.05 ~~ SIIIsn SIll(CHillII£l SUIlR ls[SlDE C!WIIEL TRAPPER CREUS.C.-...----------.----...----------------..._-----------------------_.-...--------------...--------_...-.....--...-------..--IIAIHSJEIl SITE CHINOOK CHINOOX "~!NSTEII SITE CHIMOQ;:CHINOOK IIAINSTEII SITE CHINOOK CHINOO~.DISCHARGE AREA IIUA H.I.DISCHARGE AREll WUA K.I.DISCHARGE AREA WUA H.J.12000 m62 568 fi.OI 12000 U 0 0.00 12000 73300 1100 0.02150004'1562 568 (l.01 moo 0 0 &.00 t500(1 moo !l(\(l 1}.O2l80004956251180.01 18000 0 0 0.01>lBOOO moo !lOt)0..°2 """'1 21000 49562 568 O.Ot 21~OO I)0 O.ilO 2100iJ 7:5300 1100 0.02240004956256S0.01 24000 0 I)O.vO 24000 iS3QO 1100 0.02270006mamB0.06 2700&0 v.oo 270~(l moo 1100 0.02300006'1129 4091 0.06 31>000 Q 0 O.fJl>31>000 73300 HOO 0.02330007849843790.06 33000 0 0 0.00 33000 moo 1100 0.023bOOO89m44200.05 36000 1'1000 610 0.03 361)00 751100 2000 O.O~39000 97943 4630 1).05 39000 53900 3250 0.06 moo B5100 91110 0.114200010632049B4&.OS 421>00 78500 5660 0.07 42000 moo 8300 O.Oi45OGO122m54360.04 .45000 'HOG 6090 0.06 45000 10B700 7loo 0.0741000135m58460.04 48000·115400 4270 0.04 48000 119100 5700 0.055100014924B58680.04 51000 131100 3820 0.03 51000 128900 4000 0.0354GOO16599057680.03 54000 146900 3540 0.02 54000 137400 2700 0.02510G0173m54870.03 57000 160600 3250 0.02 57000 143300 1800 0.01 ., 60000 188419 sm 0.03 60000 175600 3180 0.02 60000 148000 1300 0.01 I6300019441'1 r.ooo 0.03 63000 192000 3460 0.02 63000 154800 1300 0.01 "&bOOO 203000 6231 0.03 66000 207300 3160 0.02 66000 160100 1300 0.0110'1000 206972 6263 0.03 69000 221400 4080 0.02 6.'1000 166100 1300 0.01 ~~72000 210728 6157 0.03 72000 229000 U'10 O.\l2 72000 I69t100 1300 0.01·15000 215861 5848 (1 •.03 15000 233300 mo 0.02 75000 1721100 1300 0.01 ~.., '8-/2. CH-I ~-~()()t<\NUA. CASWELL CREEK MOUTH 7050 (T!:lousands) BEAVER DAM SLOUGH -------------------- .30 ---r---'---r-----r-----,--'-----r------~-----. .0 ----I '9 -8 - v 7 -,~ -~ «u.t.....-...6u::'ll«"V C UJ 0 5-oJ '!) :Il:l <I ~)- V).c -,I-4--...._. Q wr 3 -T (:1' w ~- -9 0 1 0 6 -- ---5 i ---t - as l- I ''=> ~Z - --- --G-------i:r- -,r-----------,--------- 10 30 50 (Thousands) MAINSTEM DISCHARGE AT SUNSHINE (cf:;) 70 Appendix Figure B-1.Weighted usa:ble area fol'"juvenile chinook salmon at the caswell Creek and Beaver Dam tributary study sites as a function of mainstem discharge. B-13 CHINOO~<\/I/U/.\ HOOLIGAN SIDE CHANNEl. 905070 {Thousgo~~l BEAR BAIT SIDE CHAN NEL 30 -.-Bre"ched -,-------------------------,,-------------<8 7 .-, .;: 6 - ;,- .~ '-' .q:5w'--ik r" 4"0 C l.J 0 4-..Jfh ". Q)j«C' tit f:. :J.........3 -a l.J t- T 0 2 - Ul ~ 1 - 0 10 .... -"-- 30 50 70 (Thousands) MAINSTEM DISCHARGE.AT.SUNSH fNE (cfs) Appendi.x Figure 8..2..Weighted usable area for Juvenile chinook salmon at the Hool1gan "nd Bearbait Side Channel study sites as a function of mainstem discharge. ;B-1Y CHII'-lOC)r<\NU,.6, LAST CHANCE SIDE CHANNEL 28 ---..--.---------------- ·~r__---r--··--·-··· 907050 (Thousands) 30 Breached t ----8 o -t----e.,----,----r-----,----,..-----,- 10 2.fa - -~ RUSTlC WILDERNESS SIDE CHANNEL6-.--........-------'----------'---------'--1 o --=='-"-=-""---r- 10 ___8reached ---J -,-----,-----,------r----...,,~---- 30 50 70 (Tho u sands) MAINSTEM OI.SCHARGEALSUNSHINE (cfs) 4.- 5........-- - - ,Appendix Figure 8-3.Weighted usable area for juvenile chinook salmon at the Last Chance and Rustic Wilderness Side Channel study sites as a function of malnstem discharge. 'B-I.6 CHI ~,~()()K \,N l.J lA ISLAND SIDE CHANNEL' S,5 --------------------------------,-.---.--.-,.. .-,.~,::5 . 5 - --- 1 - 3 ..j.- 1,5 - .3.S-- wv ~{'Jo W~ 40 in-J:.--,t- -.( ~.IJi ...,-. 0:.:N) .....f~ '- 0,5 --BI-------""~Breached 0-+--·---,-----.-------'-1--------,--.--,---..--....".._..,-------'j 10 .30 50 70 (ThQusands) MA(NSTEM WEST BANK - - -[] 70.30 50 (ThQusands) MAINSTEM DISCHARGE AT SUNSHINE (efa) "t.Breached ...---~---------------------_._-------_.,1 1 10 -""9-- lY 8fII '-' « ILI-70:::"'.:("0 C ILl 0 6-1r4 lD:J«0V)-C ......1-S-''''- a ILl,...41. C> W .3~ 2 10 Appendix Figure B-4.Weighted usable area for juvenile chinook salmon at the Island Side Channel and Ma1nstem West Bank study sites as a function of mafnstem discharge. 8-/b CHI N ()O}<\N U l\ GOOSE Z SiDE CHANNEL.. 8 ,, --- ',-----1------ 70 ,, I, I, I,,, I, I I 30 50 (Thousands) Breached t I,, I I I I I o -t-~-~--.,......--..-..'.....,..------_,_---___,r_I'J 10 4 5 .3 - 7 - ·2 - ~- - CIRCULAR SIDE:CHANNEL..'-,---------------------------------------- -- 10 9 .-,8-- 0-7In---<l iIoV""-60::i/) <l"c:wO 5..3[1) IU:J«0 1f)J::. ::>..':,..4- 0 W t-.3I <:> w 2~t Brea<:hed o +-----,..-------r------_,_------,,-------,------,---- 10 ·30 50'70 (Thousands) todAlNSTEM DISCHARGE AT SUNSHINE:,(cfs) --,Appendix Figure 8-5.Weighted usable area for juvenil e chinook salmon at the Goose Z and Circular Side Channel study sites ilS a function of main stem discharge. "... '8-11 CHINOOK vVUA SAU NA SIDE CHANNEL. 2 - - .,-._------ 70SO (Thousanu\I) t Breached 30 --- --o ... 10 4- .3 - 5 6 ...,------------------------------.----_·-----1 I-- BEAVER DAM SIDE CHANNEL. 4 ----------. -=-=-·-=-::..;-=-;i~-=-=-:...:-:....::-:...:-::...;:i::..;-=-=-:..§!::::::=:::::;:==:!...--_,----·---.-----,.----....--------1 30 50 (Thousands-) MAINSTEM DISCHARGE AT SUNSHINE (ch) <( lIJ,....,.· 'Ii.,/) «~ wC; -J I;' ,~::; «Q ifj .t: =-t", o wr Lo l.LI ~ .3 2 ..5 - 1.5 1 - D.5 - o 10 Breached \ \ \ "b, 70 - Appendix Figure B-6.Weighted usable area for juvenile chinook salmon at the Sauna and Beaver Dam Side Channel study sites as a function of mainstem discharge.- CHINOOK WUA SU.NSET SIDE CHANNEL. 7 - ..-...-- 6 - 703050 (Thousands) SUNRiSe:SIDE CHANNEL. ._-----------'---------_._---~---~ Breached t 4- O+--'-'-~--r---,----'---r----':"---,r----r-":-_~--~ 10 2 5 - .3 - 7.-. - I-----...... I ! 70 Breached t :-I O 'l!....__._'.__.._I...r---....-.-~...,....-._f-_._rl----_r_----r-----~-----I 10 30 50 (Thousands) MAINSTEM DISCHARGE AT $UN5HINE (<:fs) 4 5 .- 6- "... Appendix Figure B-7.Weighted usable area for juvenil e chinook salmon at the Sunset and Sunrise Side Channel study sites as a function of mainstem discharge. - 10 - 9 ,-.. ;::8 0- (/) -'7 -«w,-... Q:<II -e("'O 6c: l.IJO -J Lit iD j«0 5v,.J: --',1- ~-.....- 0 4l.IJ - l- I 0 !oJ 3 ~ 2 10 CHI f'J (J C)f<\l/IJ/\ TRAPPER CREEK SIDE CHAN NE.L Projected WUA • (Head barely overtopped) Breached t 30 50 (Thou~ands) MAINSTEtot DISCHARG~A"L_:?_UNSHINE:(<;fs) 70 - ~ .1 Appendix Figure B.a.Weighted usable area for juvenile chinook salmon at the Trapper Creek Side Channel study site as a function of mainstem discharge. B-2.0 - - - Appendix Table 8-6.Weighted usable areas and,habitat ind!ces for juvenile coho salmon in lower Susitna Rlver model sltes,1984. ~. Appendix Table·B-7.Weighted usable areas and habitat indices for juvenile chum ~salmon in lower Susitna River model sites,1984. ~____'___"__'.'_C__"'__~_'0',.,-'-'.-'- t«Illt.l SAIl 51 ~C!IAIllI£t.KIlOTG S1.11116H HEI\O B£ARilU SUE c-.a.--_....__._---.-----_.....-------------..-----------------------IIIHHSTflf SItE CHUIt tHJllt MIMSlEll SITE CHUM ClIIJII IfAIIfSTElI SITE CHUIl CHUII DlSCHAR6E AREA ~iJA It.1.DISCHARGE AREA ~UA H.I.DISCHARGE AREA MUA K.I. 12000 63400 28500 0.45 12000 48200 moo 0.82 12000 3100 1300 0.42 15000 1>3400 28500 0.45 1SOO0 48200 :19600 0.82 15000 3100 1300 0.42-18000 63400 28500 0.4S 18000 48200 moo 0.82 111000 3100 1300 0.42 21000 1>3400 28500 1),45 21000 48200 39600 0.82 21000 3100 1300 0.42 24000 79800 moo 0.60 24000 48200 39600 0.92 24000 3100 1300 0.42 27000 811900 46700 0.54 27000 49200 ;moo 0.82 27000 3100 1300 0.42 30i100 90800 44000 0.49 30000 48200 39600 o.B2 30000 3100 1300 0.42 33000 96500 41700 0.43 33000 41lSoo 39600 0.82 33000 3100 1300 0.41 3&000 104800 3&400 0.37 36000 57900 41000 0.71 36000 5700 1400 0.25 39000 113700 34700 0.31 39000 moo 42900 0.03 3'1000 10800 1900 0.18 42000 122900 30300 0.25 42000 71500 44500 0.51 42000 14600 2600 0.18 ,-.45000 131300 26100 0.20 4SOOO 116800 46100 0.53 45000 17900 3300 U8 48000 141200 21900 0.16 48000 95100 47600 0.50 49000 21100 4100 0.1' 51000 1S2000 18900 0.12 51000 102200 46S00 0.45 51000 23900 5300 0.22 54000 163000 18100 O.t!54000 106700 42300 0.40 54000 26400 5700 0.22 $7000 174100 17600 0.10 57000 110200 38300 0.35 57000 29000 5500 0.19 60000 1811900 17200 0.0'60000 113500 344i10 0.30 60000 31500 5100 ~.16 63000 200800 h900 0.08 63000 116600 moo·0.2S 63000 33'100 4700 0.14 &6000 213300 16700 0.08 66000 11'1000 24100 0.20 66000 36300 4400 0.12 6'1000 22l>OOQ 16400 0.01 69000 128100 I'1BVO Q.1i 69000 38300 4200 0.11 ~71000 239000 1.100 0.01 72000 121000 .17900 O.IS nooo 40000 4100 0.10 75000 250900 15800 1>.06 75000 121400 moo 0.13 75000 41500 4000 0.10 B-21________________'"--.e--'.~_'__..;.........,---------------'--~~------------------_--'- Appendix Table B-7.Continued.- ." LAST OtAlICE S.C.RUSnc NllOERNESS S.C,ISlMlD SIDE CMANNEl---------------_....-_..__............---...._--...-----_.._..---_........_..-..~---_...-...--------------------------_......_-..---------~ IIA IHSTEK SITE CHUM CHUM MIHSTEK SITE .CHUIl CMUIl IlAIHSTEIl SITE CHUM eHUIl 01 SClti<llG£AREA IIUA H.I.DISCMARSE AllEA NUA H.I.DISCHARGE AREA NUA H.l. 12000 17500 moo 0.06 12000 4800 3bQO 0.75 12000 moo 19300 0.61 -tSOOO 17500 mO{l {l.O&15000 480')3&00 0.15 15000 moo 1930(,0.01 ISOw 17500 11500 {l.oe.18000 4800 3&00 1).75 18000 31500 moo 0.01 21000 11500 11500 0.60 21000 moo 3080il 0.97 21000 moo lnoO 0.&1 24000 17500 moo 0.60 24000 49500 32500 0.66 24000 315/)0 moo 0••1 21000 31700 11500 0.36 27000 00100 27600 0.45 27000 31500 19300 0.61 30000 50600 11500 0.23 30000 moo 22700 0.33 3000il 31500 19300 0.&1 33000 63900 11500 0.18 3301)0 76800 ISIOO 0.24 33000 31500 19800 O.U 36000 73200 11500 0.16 36000 moo moo 0.16 36000 44600 28100 0.63 -,39000 '80000 11500 0.14 39000 89900 Wooo 0.12 39000 48100 28800 D.60 42000 85900 USOO 0.13 42000 97000 BBOO 0.09 42000 53200 25800 0.48 45000 90600 1150G G.13 45000 104000 7400 0.07 45000 58900 22700 0.3'1 48000 .94000 WOO 0.12 48000 109000 SBOO 0.05 48000 05500 19100 0.30 51000 '1l300 moo 0.16 51000 114000 4200 0.04 51000 72000 17400 0.24 ~ 54000 98500 20200 0.21 54000 111400 3300 0.03 54000 moo 15100 0.19 57000 100200 19500 0.19 57000 119200 3000 0.03 51000 86700 13200 0.15 60000 101800 18000 0.18 60000 120701)3000 0.02 60000 93100 12400 0.13 63000 103200 16200 0.16 moo 121701)3000 0.02 63000 9'JBOO 12700 0.13 06000 104400 13600 0.13 6&01)0 122200 3000 0.92 66000 106200 13000 0.12 69000 105SOO IOSOO 0.10 69000 122700 3000 0.02 '69000 111900 133C")0.12 720~106300 B800 0.08 72000 123000 3000 0.02 72000 118200 13000 0.12 7S00a 107000 7000 t007 75000 123500 3000 0.02 7~00 123300 13~OO v.lI IhUIlSTtII lIEST BAlIK 6OOS£2 SUlE CIIAIlllEl.CIRCIUI SIH C1lIltIlEl--_....--.._---....-..._-_..-----------_..------------------------_.-.._-------_._------_._-_..._----------..----- IlAI/lSTEIl SITE CHIJ!f CHUM "AlNSTE~SITE CHU~CHUIl IlAINSTEII SITE CHUIl CHUIl DISCHARGE AREA lIIlA H.J.DISCHAR6E AREA WIIA H.I.liiSCHMIl6E AREA NllA H'I. 121lOO 61603 41090 0.16 12000 0 0 0.00 12000 59464 46109 0.78 15000 6101l3 m90 0.7i>15000 0 0 0.00 15000 59464 40109 0.78 lBOOO i>I&03 47090 0.76 18000 0 0 0.00 18000 59464 46109 0.18 21000 73426 5m5 0.73 21000 0 0 0.00 21000 59464 4&109 0.18 24000 80904 43ZS9 0.54 24000 0 0 0.00 24000 59464 46109 0.78 27000 93353 31bil6 0.34 21000 {I 0 0.00 27000 59464 46109 0.78 30001>10B613 21151 0.25 30000 9600 4900 0.51 30000 59464 46109 0.18 ~ 33000 114738 23420 0.20 33000 21500 11000 0.51 moo 59464 46109 0.78 , 36000 117696 21182 0.19 36000 moo 17400 0.51 36000 71590 44495 0.62 3'lO00 120505 21096 0.18 39000 41800 25500 0.53 39000 7.534 4460b 0.58 42000 123391 21218 0.11 42000 01400 31BOO 0.52 42000 811551 42269 0.52 45000 129211 22389 0.17 45000 72000 moo .0.53 4SOOO 85140 42176 0.50 I'IiI1! 48GOO 13364'1 26770 0.20 48000 81400 41600 O.SI 48000 92944 43074 0.46 SIOOO 13688S 276~1 0.20 51cno 91900 moo 0.4'1 51000 102530 4S026 0.4. 54000 140761 303B2 0.22 54000 moo 40700 0.44 54000 113323 50073 0.44 '7000 144269 31815 0.22 57000 97100 33400 0.34 57000 125153 50248 11.40 'I!llI'I &0000 147899 13950 0.23 60000 99900 24000 0.24 60000 134m mas 0.34 &3000 151S42 35953 0.24 i>3000 102000 18600 0.18 03000 143m 49339 0.34 6ilOOO 154205 3648'1 0.24 66000 103200 13B06 0.13 66000 15086'1 49565 0.33 69000 156425 36211 0.23 69000 104200 10400 0.10 09000 154657 5034b 0.33 12000 158522 31029 0.23 72000 104800 11300 0.08 72000 157074 4"8491 0.31 75000 160818 36B09 0.23 15000 105100 7400 0.07 15000 1S92H 4&1'i7 0.29 - - B-22 .-Appendix Table B-7.Conti nued. SAUNA SIDE CHAN:'EL SUC~ER SIDE CHilHNfL B£AVER OAK SlllE CHANlIEL --_..----_.__._------------_._..----------..,.----------------------------------_..-----------------------_...------_.- "msm SITE CHUd CHU~"AlIlSTE"SITE CHU"CHU""AIMS1E~SUE CHun ~HUH mCHARSE AREA WUA H.L DiSCHARGE liREll Willi H.r.OISCHliii:6E AREA WUA H.L ~20DO 42M3 31754 0.75 1201)0 0 0 0.00 12000 18900 moo v.63 150"0 mn 31754 0.75 15000 0 0 0.00 15000 18900 moo 0.63 1BOOO 420n 31754 0,75 IBOOO Q 0 0.00 18000 IB900 11900 0.63 2[000 42093 31754 0.75 21000 0 0 0.00 21000 18900 11900 0.63 .f1"I'i"lI>,24000 42093 31754 0.75 24000 0 0 0.00 24000 IB901}11900 0·.63 moo 42093 31154 0.75 27()1j1)1600 0 ()~OO 27000 1890;0 1190£1 0.63 30000 42093 31754 0.75·30000 BSOO 1300 D.B6 30000 18'100 '11900 C/.63 33000 .47/011 28574 0./00 3~OOO 14900 moo ();19 33000 IB900 11900 n.63 !..~36000 48790 21855 0.57 36000 1(,900 12100 0.75 3/,000 IB900 11900 0.03 39000 49121 21307 0.56 39000 l'1400 13200 0.68 39000 18900 11900 0.63 42000 49158 26m 0.53 42000 23600 13400 0.51 42000 18900 moo n.63 45000 50289 25204 MO 45000 29600 moo 0.48 45000 19900 11900 0./'0 48000 5OBB9 236]0 0.47 4S000 31100 19900 0.54 4BOOO 22400 moo 0.59.....M4 51000 4b600 27700 0.59 51000 28000 15700 0.56510005145122565 54000 52011 21836 0.42 54000 57900 moo o.se 54000 32600 17~0 0.54 57000 5mB 2lIB 1 0.41 57000 11/,900 34400 0.51 57000 35700 18800 0.53 110000 53294 20990 0.39 60000 71300 32900 0.46 60000 38000 IB200 0.4B 113000 54275 206&9 0.3B 63000 73901i 30800 0.42 63000 39MlO 16400 0.41 G6oo0 55184 20938 0.38 6/,000 75900 28200 0.31 IIMlOO 40800 14000 0.34 69000 56053 21017 0.37 6'1000 77300 25000 0.32 119000 4l500 12100 0.29 72000 57142 Zll53 0;37 72000 78100 21600 0.28 12000 moo 11300 0.27 75000 6lOl8 23015 0.38 75000 moo 19200 1J.25 75000 42100 10700 0.25 - _T Sill tIlAIlIlEl SUIlIUSE SIDE OtANIlEL mfPER CRED:S.C.---------------""._'--'!"-_._----------...------_.-...._-.._---.__...._-_.....------..._-----_...-_._-.-----_..._--...-._- "AIIlSTE"SITE CHU"CIllJIl IlAIMSTEIl S!Tf CHUIl CHU"IlAIHSTEK sm CHUIl CHUIl n1SCHARGE AREA IIUA H.I.OIStlfARSE AREII IltlA II;I.DISCHARGE AREA IIUll H.I. 12000 fflb2 27135 0.55 12000 0 0 0.00 12000 73300 45400 0.62 ~15000 49562 27135 0.55 15000 0 0 0.00 15000 moo 45400 0.62 18.000 49562 27135 0.55 lBOc/O 1)0 0.01i laooo mfjO 4540il Ct.0: 21001}4m2 27135 0.55 21000 0 0 0.00 210vO 73300 45400 0.62 24000 495112 27135 0.55 24000 0 0 0.00 24000 moo 454tii)0.02 27000 64118 30457 0.48 27000 0 0 0.00 27000 milO 45400 \I.~2.-30000 69129 325an 0.47 30000 0 0 0.00 30000 73300 45400 0.62 l3000 ]8488 34059 0.43 33000 0 0 0.00 33000 moo 45400 0.&2 36000 89412 34808 0.39 36000 190011 6200 0.33 31100!)75000 44700 0.59 39000 97943 31649 0.38 390\)0 53900 32400 0.&0 39000 85100 moo 0.51 ',".2000 106320 39BBB 0.38 42000 78500 46400 M'1 42000 97100 40500 C/.42 4SOOO 122338 46370 0.38 45000 97100 49700 0.51 45000 108700 36900 0.34 .8000 135476 ~iJtB5 0.38 48000 115400 44500 0.39 48000 119100 32100 0.2] 51000 149248 52671 0.35 51000 131100 31500 0.29 51000 128900 25700 0.20 54000 165990 5378/0 0.32 54000 146900 31100 0.21 54000 137400 l'I400 0.14 51000 113483 48410 11.28 :11000 160600 26000 0.17 57000 143300 13800 0.10 /,0000 188419 50093 0.27 60000 175600 25200 0.14 60000 148800 10600 11.117 63000 194419 43299 0.22 63000 192000 25300 0.13 63000 154800 10100 0.07 &6000 203000 41715 0.21 66000 207300 26200 0.13 &6000 160100 9700 0.06 69000 20&972 37100 0.18 69000 221400 moo 0.13 69000 16G100 9500 0.06 72000 210ns 33481 G.t/'12000 229000 28500 0.12 72000 169800 9400 0.06 75000 215BGI 32949 0.15 75000 233300 29000 0.12 75000 172600 9400 0.05 C__::r--!I.J [\/1 \/\/I J p" ____HOOLIGAN SIDE CHANNEL ~:.()---f------------------------------------------~----------------------------------.----------.--~ i n - - - .,A ',," ,I ~.J90 -..------r----r-.-------,------,----------; [] +Breached 1D .----~-------r-------,-·-----:T------,-- 10 30 50 70 (ThQuscnds) ,t:"t ~.-", ;:-I.40 a- U> '-' «35 --w'-, Q::'".~~ W Ci l!>30 --w -' «c.' Vi ~- ."'_.~2-5 -- Q ~ ~ I 1...:._.20 - '....5- i 5 - 50 ..- 45 40 35 30 25 -- Breached + KROTO SLOUGH HEAD l ::;::20 t- Io l.LI 15 ~ 10 -- --.--------,·-------,1,------.---,...----- 70 903050 (Thousands) WAINSTEM DISCHARGE AT SUNSHINE (cfs) 5 -- 10 Appendix Figure 8-9.Weighted usable area for juvenile chum salmon at -the Hooligan Side Channe'l and Kroto Slough Head study sites as a function of mainstem discharge. - B-2Y CH I__J tvl \'\/tJ l\ BEARDAIT SIDE CHANNEL "j - I - - - I-- I I 70 I .-----..--.---.-------J Breached t --------8----------' o -------r-.---:----,------, 10 .30 50 (ThQusands) 2 4-- 5- 6 - .:: <..( ·1..&..1',-"'........ iJ:~')<tV C 'loLl 0 ~-S 4.(0 V)..!: -~.t-- ISLAND SiOt CHANNEL 30-r-- 34 - 1 t"j G ------o~----------~wr-i 8 -x "...,,' w n ....J :"A ill:J <-{C, tj~:,~ 30 - 28 - 26 - Z~- Breached 20 - 14- 12 -- Appendix Figure 8-10.Weighted usable area fOT Juvenile chum salmon at the Bearbait and Island Side Channel study sites as a function of mainstem discharg.e. 705050 (ThQ u san ds) MAINSTEM DISCHARGE AT SUNSHINE kfs\.....~ 10 ------r-----,-- 10 -- B-2.o C 1'1 U r\,~\N UA, MAINSTEM WEST BANK 50 (Tho u sands) ..... ..... --r-----r----~---___r_-.........-.. 7030 20 .. 10 "....L.L - 24 .. 2403 - 4·8 40 - 46 . za _. .3-0 . r. 'n--, ...--...- l.LIO ....J fl ) IDJ 4:C'.",.e::."\-,,,/ GOOSE 2 SIPE CHANNEL 4.5 ..._--.------~---'--------------:----------- 25 J --- f-:-----,---...,..-.,L..l--..--·---:---l------~--·----,-------------,---... 30 50 70 (Thousands), MAINSTEM DISC.HARGE AT :::>UNSHINE (cfs) 10 Breached t o 1 D 20 - 4.0 - ,"-;'l .~.- --us 'J->I~~ m;j "'l'!:- 11').1= -~~--...._- '-'30 Appendix l='igure B·11.Weighted usable area for juvenile chum salmon at the Mainstem West Bank and Goose 2 Side Channel study sites as a function of mainstem discharge. B-26 ::;:::i ..r'....__._. ::'~i" CHU!\,'1 ViUA CIRCULAR SIDE:CHANNEL-------'-----------_._--_.... .r.:':;.'1'-'-' 5::- « 1.1.1"....• 0::;:'? «~ wO -.....J !,')m:;; 400.c-I-_.J ...._ o W l- T {'j \101 5 51 -I 50 I I 49 -I 48 i I 4.7 .J I +ti ·1 [j....+)--0 0 0 ·~·5 . BreaChedt E3 El 7030 +(1 ...-------, 10 4-1 50 (ThQusands) SAUNA SIDE CHANNEL ,3,5-·-~r··~~·--.·---.~-~._.--------. - -33 ~"Breached t ..-----r------ 703050 (Thousands) MAINSTEM DISCHARGE:AT SUNSHINE (cfs) ~-,".~-",.-1 ~~.i1 '.f "1----.Ir.' 11'> '~?,Q « LNF""'·29 ..(>:.2!«'-',-:z.cJLLICi .J 1:- III :l ~. l -I«0 "- </}.c :)1-2:6 .. '../ Q ...../:':. \.oJ "-.;;l - I- J 2:~(:> w 23>---22 - ~1- Appendix Figure 8-12.Weighted usable area for juvenile chum salmon at the Circular and Sauna Side Channel study sites as afunc:Uon of main stem discharge.' .- 8-2'7 -l-O 3:1 ,.-... :.::: 30 C!:.'}-._' «75 - lLI"'~u:.'1)I.r.{~ w 0 20 -I-I '1)<p ::; <.(0 t.'-~..c -.~t-.'.....15 -c.Breached tw t- I \')10 - w S 1 a 50 C:HU tv1 \NU~·, SUCKER SIDE CHANNEl. 50 (Thousands) BEAVER DAM SIDE CHANNEL 70 - - - «:!",Lj,--, ,,-~4~ wi3 -1 Ii') Ul '", <I''J '!'I"C-.~..-_...._~ :-0 'T--- 1 ~J ··f 1 ~:--I "1 f - ~~~--~. 1:;- ,.-:' ;~--------- '0 -I Breached t _____-fCt------.J .......... - - - 30 50 (Thousands) MAINSTEM DISCHARGE AT SUNSHINE (cfs) Je I .:1 ···1 o -'1:;-..-...---...----,...--------,- 10 ,~-_"'-J.-----.,----',---_.,--.-- 70 - i:::g Append1x F1gure 8-13.We1ghted usable area for juvenl1e chum salmon at the Sucker and Beaver Dam Side Channel study sites as a funct10n of mainstem discharge. 8-28 - - - CHU~!1 \JVUA SUNRISE SIDE CHANNEL.cO - ---- 70 Breached I o .l------~_r_~---·--r~-f---...---....;...---r---'---..-------...-----J . 10 30 50 (Thousands) WAINSTEM DISCHARGE AT SUNSHINE (cfs) 10 30 40 50 Appendix Figure 8-14.Weighted usable area for juvenile chum salmon at the Sunrise Side Channel study site as a function of rnainstem discharge. - 13-29 - Appendix Table B-8.Weighted usable areas and habitat indices for juvenile sockeye salmon in lower Susitna River model sites,1984. ""'"ROlLY CREEK IUlUTH CASIiHl mn llOUTH BUYER 1lH"SLOUlill----_._-------_....._-----_.._---...._--------------------------------------------------.._----------..._-------_........- IIAINSTE~sm .SOCKElE SOCKEYE ~AINSTE~SITE SOCKEYE SOCKEYE "AIHSTE"SHE SOCKHE seC~,E,E J)ISCHARGE AREA MUA H.I.DISCHARGE AREA WUA H.I.DISCHARGE AREA ~UA H.I. 12000 8490.0 10600 0.12 12000 16200 1350 0.08 12DOO 111.00 1>200 D.53 15000 84900 1060D 0.12 1500D lh200 1350 0.08 15000 11600 6200 0.5.3 1800;)8mo 10600 0.12 18000 16200 1350 0.08 18000 WOO 6200 0.~3 21000 84'/00 W600 0.12 21000 16200 1350 Me 2lQOO 11700 blvO 0.53 24000 85300 10600 0 0.12 24000 16200 1600 0.10 24000 11900 62CWJ (1.5: 1.7000 88100 11000 0.12 27000 163M 1100 1).10 27QOQ 12200 64C1i o.~: 30000 moo 13400 0.14 31)01)0 16700 1900 0.11 301i00 moo 6600 ~~5J 33000 99800 moo 0.18 33000 moo 2300 0.11 moo 13000 6700 0.5,;'. 36000 108900 22800 0.21 36000 18000 2;'00 0.14 36000 13400 7000 0.51 39000 121000 28900 0.24 39000 18900 3100 0.16 3900(1 moo 1100 0.5: 42000 115~.35500 0.26 42000 19800 3100 0.19 42000 14400 1300 0.51 45000 152600 43400 0.28 45000 21000 4300 0.20 45000 15000 750Q 0.5& 48000 178500 52100 0.29 48000 moo 5000 0.23 48000 15700 1100 o.n 51000 198800 64400 0.32 51000 22700 5100 0.25 51000 16300 8000 0.49 54000 213000 15300 0.35 54000 moo 6400 0.21 54000 J6800 8200 0.49 51000 223200 82800 0.31 57000 24600 1200 0.2'i 51000 moo 8600 0.4~ 60000 '229B00 8821)0 0.38 60000 25S00 1900 0.31 60000 18500 8900 0.45 63000 235000 Q3000 0.40 61000 26300 8600 0.33 63000 moo 9400 v.4S b600Q 218100 moo 0.41 1>600D 21200 '1200 0.34 6bOOO 20800 10200 C.4~ 69000 241600 99900 0.41 69000 21'100 10000 0.36 69000 21600 10800 0.5~ 72000 243200 100700 0.41 72000 28'100 10~0.31 72000 22100 llOOO 0.5C 75000 243600 101500 0.42 15000 2'1100 11400 0.38 15000 moo I WOO O.4~ SutlER me CllAlllIEL SEAVER Oll"SIDE C!fANltEL---------------_._---_...-----------------------_.._..._----_.. ~lHSTEII SHE socmE SOCKEYE ~AIIiSTE~SiTE SDCKHE SOCKEYE OISCHARGE "REA IIUA H.I.DISCHilR6E AREA MUll II.I. 12000 0 0 0.00 120M 19900 3000 0.16 15000 1)0 0.00 150;)0 111900 3000 0.11>~ 18000 1)1)0.00 18000 IS900 3000 0.16 211>00 0 0 O.OC-21000 IBM 3000 0.10 24000 0 0 0.00 moo 181011 3000 0.16 21000 1600 0 0.00 21000 1891i0 3000 0.16 ~oo 85-)0 1200 0.14 30000 moo 3000 O.lb 33000 a900 1800 a.12 33000 18900 3000 0.16 1'000 16'100 1100 0.10 36000 18'100 3001}0.16 39000 19400 1500 0.08 39000 :9900 3000 0.16 42000 23600 1200 0.05 42000 19900 3000 0.16 45000 29600 1200 0.04 45000 19900 3000 0.15 48000 31100 2600 0.01 49000 22400 3200 0.14 51.000 46bQO 4000 Q.O'I 5tOOO 29000 3100 0.13 54000 51900 5000 0.09 54000 moo 4100 0.13 51000 116900 5300 0.08 57000 35700 4300 0.12 60000 11300 54(10 O.OB 60000 38001;4200 0.11 b1400 moo 5500 0.01 61001;moo 3900 0.10 66000 15900 5600 0.01 66000 40900 3600 0.09 69(100 11300 S600 0.07 69000 41500 3200 0.08 ~ 12000 19100 5600 0.01 moo 41900 3000 0.01 75000 18300 5600 0.07 75000 42100 2800 0.01 B-30 - - Appendix Table B-8.Continued. B-31 C::'j ()(',<E"y'E~-\AI'I J!:l~"-_I r "~.('l ',_l \ CASWELL CREEK MOUTH 14 - 8 - - '..J.~~ '/' ru -' <.l '.'u>...c--,I-".--..--"' 7 4 -. _3 ,-, ~ o --------------r------,------r--·---~T------····----r-·-·-·-- 10 ~30 50 70 (Thousand!3) I I i <t '-'J.-. ~_':2«'~ w::O ...lCb Ul~«0 ~')~-·.r-~'--a 13 12 -- Ii - i 0 - 9 - BEAVER DAM SLOUGH---------------_.-----------.--- ,-- 10 30 50 (Thousands) MAINSTEM DISCHARGE AT SUNSHINE (cfs) 70 Appendtx Figure 8-15.Weighted usable area for juvenile sock.eye salmon at the Caswell Creek.and Beaver Dam tributary study sites as a function of mainstem discharge. 13-32. - ..... APPENDIX C COMPARISON OF THE IFIM AND RJHAB MODELLING TECHNIQUES AT TWO SELECTED SITES DRAFT/PAGE 1 4/11/85,4/29/85 NUM2/APPENDIX C - - INTRODUCTION MEtHODS CRAFT DRAFT/PAGE 2 4/11/85,4/29/85 NUM2/APPENDIX C Trapper Creek Side Channel (RM 91.6)and Island Side Channel (RM 63.2) were selected as sampling sites for this comparative study because they represent two different channel types of the lower Susitna River. Trapper Creek Side Channel is a simple straight channel.Island Side Channel is a more complex,winding channel.Further descriptions and photos of these two sites are contained in Quane et al.(1985). Descriptions of the two modelling techniques will not be presented here. Detailed descriptions of the IFIM are presented in Appendix D of this report and Bovee (1982),and summarized in Section 2.0 of this report. The original RJHAB model was first developed and described in Marshall .C-I DRAFT/PAGE 3 4/11/85,4/29/85 NUM2/APPENDIX C et al.(1984)and modifications were described in Section 2.0 of this report. Both techniques entail taki ng depth,velocity and cover or substrate measurements spaced at intervals acrOS$transects running at right angles to the channel.Far fewer measurements are taken for the RJHAB model than for the IFIM models.A hydraulic model is developed with the IFIM and the model is run on a main frame computer.No hydraul ic model is developed by the RJHAB and the model runs on a spreadsheet with a microcomputer.The IFIM model can generate estimates of equivalent optimum habitat called weighted usable areas (WUA's)with any flow .,., within its calibration range,while the RJHAB model only calculates WUA's at discharges for which measurements are taken.Therefore,it is necessary to interpolate between point measurements generated by the RJHAB model.The RJHAB model does have the advantage of be"jng able to run in areas heavily influenced bymainstem backwater or sloughs with flows less than 5 cfs.The measurements and data analysis for the RJHAB model were taken by different investigators than those who took the IFIM measurements and analyzed them. The RJHAB model uses measurements at an additional upper transect within each of the sites.This upper area was very similar to lower sections of the site,and therefore waul d not change comparabi 1ity of the two methods.The IFIM presents results of the analysis on the basis of a 1000 foot reach,whi 1e the RJHAB model presents WUA's for the si teo Therefore,the length of each site as used in the RJHAB model was calculated and WUA's were adjusted to the basis of a 1000 foot reach. C-2. - - - - - - .- - - - DRAFT/PAGE 4 4/11/85,4/29/85 NUM2/APPENDIX C At Island Side Channel,two additional partial transects were put in for IFIM analysis of the site,and no RJHAB measurements were taken at these transects.A trial run which minimized the effect of these two addi- tional transects showed only very minor changes in WUA. RESULTS An IFG-2 IFIM model was run at Island Side Channel and hydraul ic data werecolletted at a side channel flow of 338 cfs (Appendix D).At Trapper Creek Si de Channel,hydraul ic data for an IFG-4 IFIM model were collected at flows of 16,32,and 389 cfs.Habitat data for the RJHAB model were collected four times at Trapper Creek Side Channel and five times at Island Side Channel and the RJHAB models at both sites were evaluated as It goo d"(Table 6J. The modelled response of area at the Trapper Creek and Island side channel sites to changes in discharge was almost identical for both the IFIM and RJHAB modelling techniques (Appendix Figure C-l).Differences in areas below the overtopping flow at Island Side Channel are probably due to the IFIM not being able to model flows below 5 cfs while the RJHAB WUA was measured at a flow of less than one cfs.Other C-3 TRAPPER CREEK SIDE CHAf'-J~jEL (11"1104 vs.RJHAS COMPARISON) - -50 (Thousands) 30 OIFlM •RJHAB 13 8e 0 BOB B' -t----.------.-----.----.....----r--,--- 70 T80 170 T60 T50 .-..140.-=-8. "I;:1300"0a...'-"~ t 20«0w.c:~t-«'-'110 100 90 80 70 10 ISLAND SIDE CHANNEL :: no t 60 150 140 l30 ,-..120.,-.. -",-".,;::1100"0a",'-",100«0 "'..::~90« 80 70 60 - 50 40 10 o IFlt! •RJIlAB 30 50·70 (Thousands) WAINSTEM DISCHARGE AT SUNSHINE (cfs) l - Appendix Figure C-1.Comparison of site areas calculated with the RJHAB and IFIM rnodel1 ingtechniques for the Trapper Creek and Island Side ChanAel study sites. - - - DRAFT/PAGE 5 4/11/85,4/29/85 NUM2/APPENDIX C At Trapper CreekSide Channel,the shape of the WUA curves for both species were basically the same for both modelling methods (Appendix Figure C-2).The RJHAB model ap~ears to consistently underestimate the amount of WUAin comparison to the IFIM model.The underestimation of WUA by the RJHAB model leads to smaller habitat indices although the shapes of the habitat index curves are similar for both techniques {Append;x Fi gu reC-3)• At Island Side Channel,on the other hand,WUAs from the two modelling methods do not compare closely (Appendix Figure C-4).The chinook and chum WUA response curves look more s'imi 1ar to each other than do the modelling techniques.Peaks in WUA for the RJHAB model occur at approx- imately 40,000 cfs while the IFIM model predicts a peak WUA at approxi- mately 60,000 cfs.The IFIM model does predict a chinook salmon WUA of 6,230 ft 2 to 6,600 ft 2 at side channel fl owsof 6to 11 cfs which corresponds to the peak in the RJHAB model where a measurement was taken at a side channel flow of approximately 10 cfs. When habitat indices are calculated for both methods at Island Side Channel,differences between the two techniques appear smaller (Appendix Figure C-5).The RJHAB model shows a peak habitat index for chinook salmon at approximately 39,000 cfs which the IFIM model would also show at si de channel flows of 6 to 11 cfs.Chum habi tat indi ces for both techniques decrease after overtopping although the RJHAB habitat indices drop off more steeply. (-5 TRAPPER CREEK SIDE CHANNEL CHINOOK WUA COMPARISONS 12 11 ~10 OIFlM--9 +RJHAB tT Itt........8 ...:w........ It:"7...:"c wO 6....."ID:J...:0 In.t:5:JI-......, 0 4:w l- X C).5W 3; %:~..~~.~~~y 0 10 .50 50 (Thousands) 70 - Appendix Figure C-2.Comparison of weighted usable areas calculated with the RJHAB and IFIM modelling techniques for juvenile chinook and chum salmon at Trapper Creek Side Channel.1984. 60- .-..50--tT fit-40 -c:a IFlM·W0" It:"+RJIlAB"':C wD 30....... ID:J ...:0It).t::Jt 0 20w'!i C'w J::10 0 10 CHUN WUA COMPARISONS 30 50 70 (Thous':lnds) ......INSTEW DISCHARGE AT !»UNSHINE (ct.) c.-~ - - ,.., - - - -TRAPPER CREEK SIDE CHANNEL CHINOOK HABITAT INDICES ..- ...... I 0.15 0.12 0.1 t 0.1 0.09 w 0.08 Qz 0.07 ~c.0.06I-mc.0.05I. 0.04 0.03 0.02 0.01 0 10 01FIM +RJHAB 30 70 CH UN HABIT AT INDICES 30 50 70 (Thousands) MAtNSTEW DISCHARGE AT ~UNSHINE (efs) OIFl" +RJHAB 04-------,..---.......-----,..----r-.-..,;--,---'----,;---- 10 0.1 0.3 0.%- 0.5· 0 ..0& 0.7 -r-a~'-.-8--8-8--:..-8-8-----------------:..---·--'-"1' 8 's., ·_-11"--_+--+--+--__.....~_.~ 0.6 - ! I I i ~.... ~ iiicx \of C. Z - ...... - Appendix Figure C-3.Comparison of habitat indices calculated with the RJHAB and IFIM model 1 ing techniques for juvenile chinook salmon at Trapper Creek Side Channel,1984. C-'7 7 6 --:-o IFlH--+RJIlAB a-Slit....... <4: W";'~"4"cwC) ~1It m::J CO 3ll')J:JC 0 W t-2:I: U'I3-GW ~ ISL.AND SIDE CHANNEL CHINOOK WUA COMPARISONS \ - o 10 30 50 (ThQu.ands) CHUW WUA COtoCPARISONS 70 42 40 38 o lFIJI--36 +R...AIl-a-34lit....... <4:32IoV"'~1It C"C 30lola ~'"m::J 28<4:0ll')J: :::It-26....... 0w 24to- I U'22W ~ 20 18 t. 10 30 50 70 (Thousands) toCAlNSTEtoC DISCHARGE AT SUNSHINE (c:fs) \ '"'"" Appendix Figure C-4.Comparison of weighted usable areas calculated with the RJHAB and IFIH modelling techniques for juvenile chinook and chum salmon at Island Side Channel,1984. c.-'6 - - ISLAND SIDE CHAf',lNEL CHINOOK HABITAT INDICES 0,1 -'------------'-7---------------------, 50 (Thousctnds) 30 o IF I" +RJIIAB 10 0.01 -t-----,-----.-----r-----r----..----.,-----..- 70 0.02 0.03 O.O...~...J l3-<J.0E36 60 .~ 0.0.7 .. 0.09 0.08 . w Q Z 0.06- .':; ~0.05 «r - - CHUW HAmTATINDtCES - l I '19, '~\~~~ ~.b ',~-t_J 30 50 70 (Tho u scmds) toUUNSTEW DtSCHARGE AT SUNSHINE (.;ts) AppendfxFiguN!C-5.Comparison'of habitat indices calculated with the RJHABand IFIK model1 ing techniques for juvenile chinook and chum salmon at Island Side Channel,1984. C--9 DRAFT/PAGE 6 4/11/85,4/29/85 NUM2/APPENDIX C DISCUSSION The two modelling methods compared very favorably at calculating areas within the two sites.The shape of the chum and chinook WUA and habitat index responses at Trapper Creek Side Channel were very similar.The RJHAB model consistently underestimated WUA in comparison to the IFIM model.This is probably due to the R~IHAB model not taking into account - the area between the shoreline cell and the cell located one-third of the way across the channel.This area is often marginal habitat with barely suitable velocities. At Island Side Channel,large differences in WUA can also be attributed,~ in part,to the RJHAB model not taking into account peripheral marginal habitat more than six feet from shore.This difference is also reflect- ed in the habitat indices where the proportion of usable area drops off more quickly for the RJHAB model.The differences in,WUA below the overtopping flow can be attributed to the fact that the IFIM model does not run at flows less than five cfs while actual flows at discharges below the overtopping one are less than one cfs (Quane et al.1985). Other sources of differences between the two methods may be attributed to errors in the rating curves between side channel flow and mainstem - discharge.For example,the rating curves calculated a side channel flow of 49 cfs at a mainstem discharge of 33,000 cfs for Island Side Channel.Small changes in mainstem discharge near the overtopping flow lead to big changes in WUA which chopped off the top of the chinook salmon WUA peak for the IFIM method. ~-tO - .-, - DRAFT/PAGE 7 4/11/85 t 4/29/85 NUM2/APPENDIX C The effects of sampling errors in data collection on WUA estimates from both the RJHAB and IFIM techniques are unknown.Since many more measure- ments are taken for the IFIM t it should be less susceptible to sampling errors.Because only one IFIM measurement was taken at Island Side Channel at a flow of 338 cfs t however t the reltability of .modelling flows as small as 5 cfsis unknown.It seems reasonable to assume that an IFG-4model at Island Side Channel would have given somewhat differ.. ent results than did the IFG-2 model.The RJHAB model works well in situations where the primary effect of discharge is due to backwater and the IFIM model cannot be used or works poorly. In summary,the RJHABmodelgenerally gives lower WUA estimates than does the IFIM methodology.Also peaks in WUA are often narrower for the RJHAB model.Both models show the same general trends in the habitat indices for chum and chinook salmon although.the RJHAB model is more sensitive to increases in velocity and depth which decrease the habitat indices more quickly.Since the habitat indices for both sites cal- culated using both techniques are not appreciably different t analysis of trends and optimal flows by use of habitat indices woul d 1ead to simi 1ar conclusions using both methods.Comparisons of the IFIM with other instream flow methodologies have also shown differences in output t and no one method has yet been proven best (Annear and Conder 1984). C-n -DRAFT/PAGE 1 4/11/85,4/29/85 NUM2/Literature Cited ~ LITERATURE CITED Annear,T.C.,and A.L.Conder.1984.Relative bias of several fisheries instream flow methods.North American Journal of Fisheries Management 4:531-539. Bovee,K.D.1982..A guide to stream habitat analysis using the instream flow incremental methodology.Instream Flow Information Paper.No.12.U.S.Fish and Wildlife Service.FWS/035-82/26. .Hale,•S.S.,P.M.Suchanek,and D.C.Schmidt.1984.Modelling of juvenile salmon and resident fish habitat.Part 7 in Schmidt. D.C.,S.S.Hale,D.L.Crawford,and P.M.Suchanek (eds.).1984. Resident and juvenile anadromousfish investigations (May -October 1983).Susitna Hydro Aquatic Studies.Report No.2.Alaska Department of Fish and Game.Anchorage,Alaska. Marshall,R.P.,P.M.Suchanek,and D.C.Schmidt.1984.Juvenile salmon rearing habitat models.Part 4 in Schmidt,D.C.,S.S.Hale,D.L. Crawford,and P.M.Suchanek (eds:).1984.Resi dent and juveni le anadromous fish investigations (May -October 1983).Susitna Hydro Aquatic Studies.Report No.2.Alaska Department of Fish and Game.Anchorage,Alaska. Quane,T.,P.Morrow,1.Queral,T.Keklak,and T.Withrow.1985. Technical memorandum in support of Task 14 (Lower River Resident and Juvenile Anadromous Fish Studies)Alaska Department of Fish and Game,SusitnaAquatic Studies.Anchorage,Alaska. .C-I2.. - - - - DRAFT APPENDIX 0 DRAFT/PAGE 1 5/1/85,5/2/85 ANDY/Abstract HYDRAULIC MODELS FOR USE IN ASSESSING THE REARING HABITAT OF JUVENILE SALMON IN SIX SIDE CHANNELS OF THE LOWER SUSITNA RIVER James Anderson, Jeffrey Bigler,and Andrew Hoffmann of Alaska Department of Fish and Game Susitna Hydro Aquatic Studies Third Floor,Michael Building 620 East Tenth Avenue Anchorage,Alaska 99501 ABSTRACT Six side channels (IsJand,Mainstem West Bank,Circular,Sauna,Sunset, and Trapper Creek)in the lower reach of the Susitna River were evalu- ated using an Instream Flow Incremental Methodology (IFIM)physical habitat simulation (PHABSIM)model 1 ing approach to evaluate the effects that site flow and mainstem discharge have on rearing juvenile salmon habitat.These sites were thought to contain potential habitat con- ditions for rearing juvenile salmon and were chosen to range greatly in size,shape,and overtopping discharge. Six hydraul icsimulation models (either IFG-2 or IFG-4)were calibrated to simulate depths and velocities associated with a range of site- specific flows at these six modell ing study sites.Comparisons between 'I)-I D;.:tAFT DRAFT/PAGE 2 5/1/85,5/2/85 ANDY/Abstract - corresponding sites of simulated and measured depths and velocities indicate that the calibrated models provide reliable estimates of depths and velocities within their recommended calibration ranges.- The recommended cal ibration ranges over which these model scan hydrau- ,. lically simulate the habitat of rearing juvenile salmon is:Island Side Channel from 35,000 to 70,000 cfs mainstem discharge;Mainstem West Bank ~ Side Channel frOlll 18,000 to 48,000 cfs;Circular Side Channel from 36,000 to 63,000 cfs;Sauna Side Channel from 44,000 to 63,000 cfs; Sunset Si de Channel from 32,000 to 67,000 cfs;and Trapper Creek Side Channel from 20,000 to 66,000 cfs. ... ..... - - 'J)-/I """ .... - VtiAFT TABLE OF CONTENTS ABSTRACT TABLE OF CONTENTS LIST OF APPENDIX FIGURES LIST OF APPENDIX TABLES INTRODUCTION METHODS Analyt~ca 1 Approach Study Site Selection General Techniques for Data Collection General Techniques for Calibration General Techniques for Verification RESULTS Island Side Channel Site Description Da ta Co 11 ected Cal ibrati on Verification Application Mainstem West Bank Side Channel Sit~Description Data Collected Calibration Verification Appli cation Circular Side Channel Site Description Data Collected Calibration Verification Application ... b-III DRAFT/PAGE 1 5/1/85 ANDY/Table of Contents TABLE OF CONTENTS (Continued) Sauna Side Channel Site Description Data Collected Cal ibration Verification Application Sunset Side Channel Site Description Data Collected Calibration Verification Application Trapper Creek Side Channel Site Description Data Collected Cal ibratfon Verification Application SUMMARY ACKNOWLEDGEMENTS LITERATURE CITED APPENDIX ATTACHMENT 1 -1)-tV C;:\AFT DRAFT/PAGE 2 5/1/85 ANDY/Table of Contents """I i I -i -. j DRAFT/PAGE 1 . 4/19/85,5/1/85,5/2/85 ANDY lAppendi x Fi gures LIST OF APPENDIX FIGURES Appendix Figure 0-1 Location of the six IFG hydraulic model 1 ing sites in the lower Susitna River. 0-2 Overview of Island Side Channel (RM63.2). 0-3 Location of Island Side Channel study site (RM 63.2). 0-4 .Compari son of rating curves.for Island Si de Channel transect 6 (Q site)(from Quane et al.1985). 0-5 0-6 0-7· 0-8 0-9 Cross section of transects 1,lA,2 and 3 at Island Side Channel (adapted from Quane et al.1985). Cross section of transects 4,4A,5 and 6 at Is1 and Side Channel (adapted from Quane et al.1985). Comparison of observed and predicted water surface profiles from calibrated model and surveyed·thalweg profile at Island Side Channel (adapted from Quane et al.1985). Appl icationrange of the calibrated hydraulic model at Island Side Channel. Overview of Ma i nstem West Bank Si de Channel (RM 74.4). 0-10 .- Comparison of rating curves for Mainstem West Bank Side Channel transect 1 (Q site)(from Quane et al.1985)• 0-11 Location of Mainstem West Bank Side Channel study sHe (RM 74.4). 0-12 Cross section of transects 1,2 and 3 at Mainstem West Bank Side Channel (adapted from Quane et al.1985). - 0-13 Cross section of transects 3A and 4 at Mainstem West Bank Side Channel (adapted from Quane et al.(1985). DRAFT/PAGE 2 4/19/85,5/1/85,5/2/85 ANDY/Appendix Figures -LIST OF APPENDIX fIGURES (Continued) 0-15 0-14 Compari son of observed and predicted water surface profiles from calibrated model and surveyed thalweg at Mainstem West Bank Si de Channel (adapted from Quane et ale 1985). Application range of the calibrated hydraulic .model at Mainstem West Bank Side Channel. """ Overview of Circular Side Channel (RM 75.3). Comparison of ratin9 curves .for Circular Si.de Channel transect 4 lfrom Quane et ale 1985). 0-18 Location of Circular Side Channel study site (RM 75.3). D-16 D-17 D-19 Cross section of transects 1,2 and 2A at Circular Side Channel (adapted from Quane et al.1985). 0-20 Cross section of transects 3,4 and 5 at Circular Side Channel. 0-21 0-22 Compari son of observed and predicted water surface profi 1esfromca1i brated model and surveyed thalweg profile at Circular Side Channel (adapted from Quane et al.1985). Application range of the calibrated hydraulic model at Circular Side Channel. 0-23 Overview of Sauna Side Channel (RM 79.8). 0-26 0-24 Compari son of rati ng curves from Sauna Side Channel transect 2 (from Quane et al.1985). 0-25 Locations of Sauna Side Channel study site (RM 79.8). Cross section of transects 1,2,3 and 4 at Sauna Side Channel (adapted from Quane et ale 1985). D-27 Comparison of observed and predicted water surface profiles from calibrated model and surveyed thalweg profile at Sauna Side Channel (adapted fromQuaneet ale 1985). .... , D-'1/1 DRAFT/PAGE 3 4/19/85,5/1/85,5/2/85 ANDY/Appendix Figures LIST OF APPENDIX FIGURES (Continued) D~28 Application range of the calibrated hydraulic model at Sauna Side Channel. 0-29 Overview of Sunset Side Channel (RM 86.4). -0-30 0-31 Compari son of rating curves from Sunset Si de Channel at transect 1 (from Quane et ale 1985)• Location of Sunset Side Channel study site (RM 86.9). 0-32 Cross section of transects 0,1,.2 and 3 at Sunset Side Channel (adapted from Quane et al.1985). 0-33 Cross section of transects 4,5,and 6 at Sunset Side Channel (adapted from Quane et a 1.1985). 0-34 Comparison of observed and predi cted water surface profi 1es from calibrated model and surveyed thalweg profile at Sunset Side Channe 1 (adapted from Quane et a 1.1985). Application range of calibrated hydraulic mode.lat Sunset Side Channel. Overview of Trapper Creek Si de Channel (RM 91.6).'. Comparison of rating curves from Trapper Creek Side Channel transect 4 (from Q,uane et ale 1985). Location of Trapper Creek Side Channel study site (RM 91.6)• Cross section of transects 1,2,3 and 4 at Trapper Creek Side Channel (adapted from Quane et ale 1985). r 0-40 Comparison of observed and predicted water su rface profi 1es from ca 1i brated model and surveyed thalweg prof;1e for Trapper Creek Side Channel (adapted from Quane et ale 1985)• 1)-Vri DRAFT/PAGE 4 4/19/85,5/1/85,5/2/85 ANDY/Appendix Figures LIST OF APPENDIX FIGURES (Continued) D-41 Application range of the calibrated hydraulic model at Trapper Creek Side Channel. D -VHf ... - 0-1 DRAFT/PAGE 5 4/19/85,5/1/85,5/2/85 ANDY/Appendix Figures LIST OF APPENDIX TABLES Table The six lower river IFG modelling sites with corresponding river mile location. 0-2 Percent cover and cover type categories. 0-3 Substrate classifications. 0-4 The six lower river side channel IFG model- ling sites with type of hydraulic model used, dates calibrations flows measured,and corresponding site specific flows and mafnstem discharges for the open water period in 1984. Comparison of field measured and model predicted water surface elevations at the calibration flow of 338 cfs for Island Side Channel. 0-7 - - Comparison between observed and predicted water surface elevations,discharges and velocities for 1984 Mainstem West Bank Side Channel hydraulic model. Comparison between observed and predicted water surface elevations,discharges,and velocities for 1984 Circular Side Channel hydraulic model. 0-8 Compari son of fi el d measured and model predi cted water surface elevations at the calibration flow of 52 cfs for Sauna Side Channel. 0-9 The effects of the backwater at Sauna Si de Channel ~information obtained from transect 2. ~, - .-- 0-10 0-1l Comparison between observed and predicted water surface elevations,discharges and velocities for 1984 Sunset Side Channel hydraul ic model. Differences between stages of zero flow input into the model and Quane et al.(1985) thalweg survey at Sunset Side Channel . DRAFT/PAGE 6 4/19/85,5/1/85,5/2/85 ANDY/Appendix Figures LIST OF APPENDIX TABLES (Continued) 0-12 Comparison between observed and predicted water.surface elevations,di scharges,and velocities for 1984 Trapper Creek Side Channel hydraulic model. - - DRAFT/PAGE 1,4/30/85, 4/18/85,4/9/85 ANDY /Do'c 1 INTRODUCTION -- - - - - About 40%of the annual discharge of thelowerSusitna River at Park's Highway bridge originates from the mainstem Susitna RiVer above the confluence of the Talkeetna and Chulitna Rivers.Thus,operation of the proposed hydroelectric project wi 11 alter the naturalfl ow regime of this lower river reach beyond the normal weekly variations in flow which occur naturally during the open water season. One of the predominate aquatic habitat types in this lower river reach which maybe affected by such flow alternations are side channels.Side channel areas in this river reach currently provide habitat for rearing juvenile salmon.The quantity and quality of juvenilesalmonid rearing habitat in side channels in this river reach is dependent on a multitude of interrelated habitat variables,including water depth and velocity, which are i ntimatelyrel ated to mainstem discharge. This appendix presents resul ts of the physical habitat modell ing simu- 1ation efforts that Al aska Department of Fi sh and Game (ADFG)Su Hydro personnel conducted in the open water season of 1984.The objective of the study was to provide calibrated hydraulic simulation models for selected Task 14 lower river juvenile salmon habitat modelling study sites.The approach of the study was to apply a methodology which utilizes water depth and velocity as the dominant hydraulic variables to quantify the responses of rearing habitat to changes in site flow and mainstem discharge.The methodology used was the system developed by the U.S.Fish and Wildlife Service (USF&WS)Instream Flow Group (IFG) using the Instream Flow Incremental Methodology (IFIM)Physical Habitat DrtAFT DRAFT/PAGE 2,4/30/85, 4/18/85,4/9/85 ANDY /Do'c 1 Simulation (PHABSIM)modelling system (IFG 1980,Bovee 1982).The calibrated hydraulic simulation models will be utilized to assess how site flows and mainstem discharge affect juvenile salmon rearing habitat in side channel habitats of the lower Susitna River reach. METHODS Analytical Approach The current most accepted methodology used for assessing habitat re- sponses to flow variations is the USFWS,IFIM,PHABSIM modelling system. The IFIM,PHABSIM modelling system is a collection of computer programs used to simulate both the available hydraulic conditions and usable habitat at a study site for a particul ar speci es/l ife phase as a func- tion of flow.It is based on the theory that changes in riverine habitat conditions can be estimated from a sufficient hydraul ic and bi.ologic field data base.It is intended for use in those situations where flow regime and channel structure are the major factors influenc- fngriver habitat conditions. The modelling system is based on a three step approach.The first step uses field data to cal ibrate hydraul ic simulation model s to forecast anticipated changes in physical habitat variables important for the species/life phase under study as a function of flow.The second step involves the collection and analysis of biological data to determine the behavioral responses of a particular species/l ife phase to selected physical habitat variables important for the species/l ife phase under - - ..... - - - """ ..- - ,.... I I - ..... DRAFT/PAGE 3,4/30/85, 4/18/85,4/9/85 ANDY/Doc 1 study.This information is used to develop weighted behavioral response criteria curves (e.g.,utilization curves,preference curves,or suit- ability curves).The third step combines information gained in the first two steps to calculate weighted usable area (WUA)indices of habitat usability as a function of flow for'the species/life phase under study. Hydraulic modelling is of central importance to the PHABSIM system.The primary purpose of incorporating hydraulic modelling into the analytical approach is to make the most efficient use of limited field observations to forecast hydraul ic attr'lbutes of riveri ne habitat (depths and veloc- ities)under a broad range of unobserved streamflow conditions. The IFG speci'fically developed twohydrauTic models .(IFG-2and IFG-4) during the late 1970's to assist fisheries biologists in making quanti- tative evaluations of effects of streamflow alterations on fish habitat. TheIFG-2 hydraulic model is a water surface profile program that is based on open channel flow theory and formulae.The IFG-2 model can be used to predict the horizontal distribution of depths and mean column velocities at 100 points along a cross section for a range of stream- flows with only one set of field data.The IFG-4 model provides the same type of hydraul ic predictions as the IFG-2 model,but it is more strongly based on field observations and empiricism than hydraulic theory and formulae.Although a minimum of two data sets are required for calibrating the IFG-4 model,three are recommended.Either model can be used to forecast depths and·velocities occurring in a stream channel over a broad range of streamflow conditions. b--3 DRAFT/PAGE 4,4/30/85, 4/18/85,4/9/85 ANDY/Do'c 1 The IFG-4 model,which is·based upon a greater number of observed sets of field data (i.e.flow levels),generally can be used to model a greater range of flow conditions than the IFG-2 model.Additionally, since the IFG-4 model is more dependent upon observed depths and veloc- ities than the IFG-2 model,,predicted depths and velocities can be directly compared with the observed values. tool for verifying the models. This comparison isa useful Both models are most applicable to streams of moderate size and are based on the assumption that steady flow conditions exist within a rigid stream channel.A stream channel is rigid if it meets the following two criteria:(l)it must not change shape d~ring the period of time over which the calibration data are collected,and (2)it must not change shape while conveying streamflows within the range of those that are to be simulated.Thus a channel may be II r igid ll by the above definition, even though it periodically (perhaps seasonally)changes course. Streamflow is defined as "s teady U if the depth of flow at a given location in the channel remains constant during the time interval under consideration (Trihey 1980). In this analysis,all streamflow rates were referenced to the average daily discharge of the Susitna River at the U.S.Geological Survey (USGS)stream gage at Sunshine,Alaska (station number 15292780).This location was selected as the index station primarily because it is the gage located near the center of the river segment that is of greatest interest in this particular analysis.The target mainstem discharge range for data collection was from 12,000 to 75,000 cfs. b-t.J - ....... , - DRAFT/PAGE 6,4/30/85, 4/18/85,4/9/85 . ANDY /Do'c 1 species-specific life history requirements.Criteria for application of the representative concept are less restrictive,enabl ing this concept to be used when only limited biological information is available or when critical habitat conditions cannot be identified with any degree of certainty. In the critical concept,a study area is selected because one or more of the physical or chemical attributes of the habitat are known to be of critical importanc~to the fish resource.That is,recognizable phys- i calor chemical characteristics of the watershed hydrology,i nstream hydraulics,or water quality are known to control species distribution or relative abundance within the study area.Because of this,an eval- lJation of critical areas will provide a meaningful index of species response in the overall critical study area. The representative concept acknowl edges the importance of phys ica 1 habitat variables throughout the entire study stream for sustaining fish populations.Thus,under the representative concept approach,study areas are selected for the purpose of quantifying relationships between streamflow and physical habitat conditions important for species/life phase under study at selected key locations (representative reaches) that collectively exempl ify the general habitat characteristics of the entire river segment inhabited by the species/life phase under study. For this study,an adaptation of the representative concept was the approach used to assess how mainstem discharges affect the rearing habitat of juvenile salmon in the side channel habitat of the Kashwitna b-.S - DRAFT/PAGE 7,4/30/85, 4/18/85,4/9/85 ANDY/Doc 1 to Talkeetna reach of the Susitna River.The six specific sites modelled in this study were chosen by ADF&G Su Hydro Resident and - Juvenile Anadromous (RJ)project personnel in conjunction with ADF&G Su Hydro Aquatic Habitat and Instream Flow Study (AH)project and E.Woody Trihey and Associates (EWTA)personnel from lower river side channels which met the following basic criteria: 1.The sites were chosen to range greatly in size,shape,and overtopping discharge; 2.The sites were thought to contain potential habitat conditions for rearing juvenile salmon; 3.The sites were judged by AH project and EWTA personnel to be readily modelled using IFG methods; 4.The sites were accessible by boat at normal mainstem dis-~ charges during the open-water season;and, 5.The sites were above Kashwitna landing and therefore much easier to sample for logistical purposes. The six sites chosen for modelling complemented other sites modelled using the RJHAB method.All were side channels as the majority of potential habitat in the lower river is composed of side channel habi- tat,and much of the other habitat is affected primarily by mainstem backwater which is difficult to model with the IFG model. - DRAFT/PAGE 8,4/30/85, 4/18/85,4/9/85 ANDY/Doc 1 Appendix Figure D-1 shows the location of each of these six study sites selected for study based on the above criteria.The river mile location of each of the six sites is presented in Table 0-1. General Techniques for Data Collection A study reach was selected for detailed evaluation in each of the six side channel sites.The leDgth of the reach was determined by placing enough transects within the area to adequately represent the major macrohabitat types of the particular side channel area. Cross sections were located within each study reach following field methods described in Bovee and Mil hous (1978)and Trihey and Wegner (l98t).Cross sections were located to facilitate collection of hydrau- lic .and channel geometry measurements of importance in evaluating flow effects on salmon rearing habitat.Field data were obtained to describe a representative spectrum of water depth and veloeity patterns ,cover, and Substrate composition at each side channel reach. The number of cross sections established at the study reaches varied from four to eight.The end points of each cross sections were marked with 30-inch steel rods (headpins)driven approximately 28 inches into the ground.The elevation of each headpin was determined by differen- tial level ing using temporary benchmarks set at assumed elevations of 100.00 feet. COOK INLET PARKS HIGHWAY BRIDGE~oc:::::::::ZUSGS STATION (15292780) .' 10 I - - - - Appendix Figure D-l.Location of the six IFG hydraul ic modell ing sites in the lower Susitna River. - ~-8 .- -- - DRAFT/PAGE 1 4/19/85 t 5/2/85 ANDY/Tables Appendix Table D-1.The six lower river IFG modelling sites with corresponding river mile location • 0-q DRAFT/PAGE 9,4/30/85, 4/18/85,4/9/85 ANDY /Do'c 1 Cross section profiles were measured with a level,survey rod,and fiberglass tape.Horizontal distances were recorded to the nearest 1.0 foot and streambed elevations to the nearest 0.1 foot.Water surface elevations at each cross section in the study site were determined to the nearest 0.01 feet by differenti al level ing or reading staff gages located on the cross section. Streambed elevations used in the hydraulic models were determined by maki ng a compari son between the surveyed cross section profi 1e and the cross section profil es derived by subtracting the flow depth measure- ments at each cross section from the surveyed water surface elevation at each calibration flow (Trihey 1980). A longitudinal streambed profile (thalweg profile)was surveyed and plotted to scale for each modeling site (Quane et al.1985). The water surface elevation at which no flow occurs (stage of zero flow) at each cross section in the study site was determined from the stream- bed profile.If the cross section was not located on a hydraulic control,then the stage of zero flow was assumed equal to that of the control immediately downstream of the cross section. Discharge measurements were made using a Marsh-McBirney or Price AA velocity meter,topsetting wading rod,and fiberglass tape.Discharge measurements were made using standard field techniques (Buchanan and Somers 1969;Bovee and Milhous 1978;Trihey and Wegner 1981).Depth and velocity measurements at each calibration flow were recorded for the 'I)...../O - - ~, .- ..... - DRAFT/PAGE 10,4/30/85, 4/18/85,4/9/85 ANDY/Doc 1 same respective points along the cross sections by referencing all horizontal measurements to the left bank headp"in. Cover and substrate values were also determined for each cell along modelling transects.Methods described in Schmidt et ale (1984)were used to code cover (Appendix Table 2).Substrate categories were clas- sified by visual observation employing the substrate classifications presented in Appendix Table 3.The distribution of various substrate types was indicated on field maps~Substrates were classified using a single or dual code.In those instances that a dual code was used,the first code references the most predominant (i.e.,70%rubble/30%cobble =RU/CD). General Techniques for Calibration The calibration procedu refor each.of the hydraul ic model s was preceded by field data collection,data reduction,and refining the input data. The field data collection entailed establishing cross sections along which hydraulic data (water surface elevations,depths,and velocities) were obtained at each of the different calibration flows.The data reduction entail s determining the streambed and water surface el e- vations,velocity distribution and stage of zero flow for each cross section;and,determining a mean discharge for all the cross sections "in the study site.Refining the input data entailed adjusting the water surface elevations and velocities so that the forecasted data agreed more closely to the observed.A model was considered calibrated when: 1)the majority of predicted water surface profiles were within ±0.05 ft 1)-1/ DRAFT/PAGE 1 1/31/85 1985A/Table 2 Table 0-2.Percent cover and cover type categories. Substrate Code %Cover Code silt,sand (no cover)1 0-5 .1 emergent vegetation 2 6-25 .2 -aquatic vegetation 3 26-50 .3 1-3 11 gravel 4 51-75 .4 - 3-5"rubble 5 76-100 .5 -5"cobble,boul der 6 debris 7 -overhanging riparian vegetation 8 undercut bank 9 - - D-/J.. DRAFT/PAGE 1 1/31/85 1985A/Table 1 Appendix Table 0-3.Substrate classifications. - - Substrate b-/3 Particle Size Classification -.;,.",..-,..............,....;&-------..;.;.-----'---'--'--..."1"'1""-----.........----------_ DRAFT/PAGE 11,4/30/85, 4/18/85,4/9/85 ANDY/Dnc 1 of the observed elevations and 2)the majority of predicted velocities were within ±0.10 ft/sec of the measured velocities.A calibrated IFG-4 model gives velocity adjustment factors in the range of 0.9 to 1.1,and relatively few velocity prediction errors.The velocity adjustment factor is the ratio of the computed (observed)discharge to the predict- ed discharge. An IFG-2 model ~oes not have velocity adjustment factors and must be reviewed with the observed data before its considered calibrated. General Techniques for Verification The verification of how well each of these six hydraulic models simulat- ed their respective site flows was performed by hydraul ic engineers of EWT&A.The approach they used to assess the quality of each model was based on two levels of criteria.The first was qualitative evaluation of four separate sub-criteria.These sub-criteria were: 1.How well does the model conform to the established IFG and EWT&A guidelines? 2.How well does the extrapolation range of the model conform to the desired range? 3.Are the models appropriate for the species and life stage being considered? - ,.... - -! - - I~ ..- ~- - DRAFT/PAGE 12,4/30/85, 4/18/85,4/9/85 ANDY/Doc 1 4.How well do the ranges of depth and velocities of the fore- casted data conform to the ranges of depth and velocity of the suitability criteria curves being considered based on a "visual"evaluation? After the first level of qualitative evaluation was performed,an overall rating was given to the various segments of each model.The ratings given were excellent,good,acceptable,and unacceptable. Figures depicting these rating are presented for each site in the results section.The second level in the verification process required a statistical analytical evaluation of the models calibration.It was only performed when the forecast capabi1 ities of either the IFG-2 and IFG-4 model were not given an excellent rating in the level one eval- uation.For a detailed explanation of the verification ana 1ysi s see Appendix Attachment 1. RESULTS The results of the physical habitat simu1 ation modelling studies are presented below by study site.The six lower river side channel IFG modelling sites with type of hydraulic model used,dates calibration flows were measured,and corresponding site specific flows and mainstem discharges for the open water period in 1984 are presented·in Appendix Table D-4.For each study site,a general site description,a summary of data collected at the study sites,a description of the model calibration procedures used to calibrate the model for the study site, the verification of the model at the study site,and the recommended application of the model for the study site are presented. Appendix Table D-4. DRAFT IPAGE 1 5/2/85 AH~R!Table 4 The six lower river side channel IFG modelling sites with type of hydraulic model used,dates calibration flows measured,and corre- sponding site specific flows and mainstem discharges for the open water period in 1984. - Mainstem Date .Site Discharge Type of Calibration Specific at Side Channel Hydraulic Flow Flow Sunshine Site (RM)Model Measured (cfs)(cfs) Island Side Channel (63.2)IfG-2 July 25 338 56,100 Mainstem West Bank (74.4)IFG-4 September 2 450 32,000 September 20 310 30,500 September 25 6 Circular Side Channel (75.3)IFC-4 July 24 204 55,200 August 17 50 42,500 Sauna Side Channel (79.8)IFG-2 July 23 52 52,000 Sunset Side Channel (86.9)I FC-4 July 22 496 57,800 August 17 127 42,500 Trapper Creek Side Channel (91.6)IFG-4 September 18 16 20,900 -August 16 32 44,000 July 21 389 57,700 - ..... .0--/4, DRAFT/PAGE 1,4/30/85 4/18/85,4/10/85 ANDY/Doc 3,5/2/85 Island Side Channel (RM 63.2) Site Description Is1 and 5i de Channel is located on the east bank of the rna in channel of the 5usitna River at river mile (RM)63.2 (Appendix Figure D-2).This side channel is located downstream of a braided,vegetated floodplain and is not directly connected to the main channel Susitna River.It is approximately 0.7 miles in length with both the mouth and head portions adjoining side channel networks •Breachi ng f1 owsin thissi de channel result from overtopping of the head by an adjoining larger side channel. Prior to breaching,flow in the side channel is greatly reduced with a series of pools remaining (Quane etal.1985). The IFG modell ing site selected for Island Side Channel during the 1984 open waterfield season,was 735 feet in length and was located in the lower portion of the side channel (Appendix Figure D-3).The site generally consists of a pool-riffle-poo1 sequence.Based on assessments by Quane et al.{l985},an area of backwater extends through the study site to a point at least 1,100 feet upstream from the mouth of the side channel at a nan-breaching rnainstem discharge of 35,000 cfs.During mainstern discharges of 38,000 to 66,700 cfs,the area of backwater extends throughout the study site. The right bank of the study site is steep,being approximately five feet high,and results from erosional effects.The primary riparian vege- tation along this bank is alder.There are also two side pocket areas, 0-11 ~.. E9 River Mile o 2000~~ f I ~~r j FEET /J (Approximate Scale)r.......,\.". .....=.:al'.·~r",t~.'·.·).:.'1;j,~,t.·r<·~.'.'-,?~1~);.p-<.. '. ".~:~., 1°"··'.':.. ,..'.f:..., Appendix Figure 0-2.Overview of Island Side Channel (RM63.2). D--f8 - ~. - - - J 1 1 j 1 1 .)~)J )J ])J CJ I--0 ~ppendix Figure 0-3.Location of Island Side Channel studysfte(RM 63.2). .'..;.\"';'~,.':::j'~,~4'." .;.../~·ir.-Jt~.;...r ""';~"~'l7 ., ......-. ~. •'.I. ··,·U:.'·.'····',:..~_-1 -• • ' J t '.!<~~(.f",j~,;;...I ~!.:J DRAFT/PAGE 2,4/30/85 4/18/85,4/10/85 ANDY/Doc 3,5/2/85 along this bank,which during higher site flows (about 400 cfs), become slow velocity slack water areas.In contrast,the left bank of the study site consists largely of a gently sloping depositional bank. The riparian vegetation on this bank is sparse and consists primarily of shrub willow. Substrate at the study site consists primarily of gravels,cobbles,and rubbles,with substrate changing to sand and silt in slackwaterareas. The thalweg gradient of the side channel is 15.6 ft/l11i1e (Quane et a1. 1985).Breaching of Island Side Channel is the result of overtopping of the head by an adjoining side channel.From an evaluation of field observations,aerial photography and the stage/discharge relationship developed for this side channel,an initial breaching discharge has been estimated to occur at 34,000 cfs (Quane et a1.1985). Based on a review of available rating curves (Appendix Figure 0-4)it has been determined that at mainstem discharges exceeding 35,000 cfs, the hydraulics within this side channel become directly controlled by mainstem discharge (Quane et a1.1985).A side channel streamflow estimate of 43.5 cfs has been estimated to occur at a mainstem discharge of 35,000 cfs (Quane et a1.1985). Eight cross sections were surveyed within this site during 1984 to define channel geometry (Appendix Figures D-5 &6).The upper two transects (5 and 6)were located in primarily pool habitat.Transects 4A and 4 represent primarily riffle habitat in the main portion of the channel.Transect 4Awas placed as a partial transect originating from the ri ght hand bank.It represents the larger of the two sl ack water ! I t I 10',... I~ I.:, I~ I~ 0=43.5 cis '"•looO·ron (liSE\.•"''.rut ,2._0..1t _.,~......__...L-....-..............,.--....-....-_- a.. .,.,. I t•~Ig, q,1 It) '::1 °1 r ~WSEL:91.79 I CO«fltOl.lEtl J5.ooo ~q ~U.1ClO cf, W5U •11I'o.0,~0 1.1»1)•It r 1 .'Oa~ ISLAND SIDE CHANNEL TR6 GAGE 63.256 ".800 ~.0 ~J5._<f' wsn •111,0.01"aO.llo1 •It 111'"•0 .." -,"... "':loOt(lO IIIXNl'lCt lIT :IUOh1J<IlllClOCnl - .....Appendix Figure 0-4.Comparison of rating curves for Island Side Channel transect 6(Q site)(from Quane et . a1.1985). O-dl CROSS SECTION I STAnON 0 +00 280240200lISa120ISO40o lOll j104 CROSS SECTION IA I 10;5 STATION 0+44 102 101 100 9 iii iii IS 117 91S iii 15 iii 4 93 liI2 iii 1 90 ISIiI aa 1S7 ae8~1 iii I ~iii ii,iii i I zo ~> W ...J W ..., > i= <t ...J..., a: '; .:! 21S02402001110120ao40 10~ 104-10.3 ~102 ~101 100 Z n 0 III ~1i17 ~liII W 1iI~ oJ iii + W 11;5 W 1i12> ~III cl:1iI0 U ,.J 8 iii...,aaa: I a7 SL' 86 ~ 8~ 0 DISTANCE FROM LEFT BANK HEADPIN Ifutl DISTANCE FROM LEFT BANK HEADPIN I 'eet I J 10~ 104-10;5-102~.:101...100 Z u Q IiIII ~1i17 ~iii I...,1iI~ .J iii 4..., 1iI;5 w liI2 >iii 1 ~1iI0 <t IlliI..J...,1111 a:117 811 1115 0 CROSS SECTION 2 STATION ttl2 40 110 120 180 200 240 2110 .. .:!... z 2 ~ ~ W ...J W ..., > ~ ...J..., II: lOll .,.---,-.,.- 104 CROSS SECTION 3 103 STATION 2+55 102 101 100 99 98 97 98gt ~j_338.h iii 1 90 89 a8 a7::t-ii,Iii ,i 5 •i B iii o 40 80 120 1 ISO 200 240 2110 DISTANCE FROM LEFT BANK HEADPIN Ileat)DISTANCE FROM LEFT BANK HEADPIN (feet) Appendix Figure 0-5.Cross section of transects l,lA,2,and 3 at Island Side Channel (adapted from Quane et. al.1985). J )I -)~J I J 1 I 1 I }l -}-}] ,OS,os CROSS SECTION 4 '04 j CROSS SECTION 44'04 '03 STATION 4+31 . 'o~STATION !Hi2 ... t02.........,oa ~'01~.!•101 -too...-'0(1 Z 1111z:Ill/ 0 980118~97i=II"-r 116-r 1111 >95:> II,lII,l liS -'114-'1/4 ~u ~..""SIJW SIJ ""92""112 >lit>III j::j::liDliD-r 811-r -'-'811 W 88""88 0::870::87 86C)lIll 8588 0 40 80 t20 180 200 240 280I04080t211leo20024028(1 9-J OISTANCE FROM ,"EFT BANK HEAOPIN (feel)DiSTANCE FROM '"EFT BANK HEADPfN UIIIIeU '05 105\04 CROSS SECTION 5 '04 j CROSS SECTION 6 '03 STATION 5+65 103 STATION 7.35-102-'02 ••'0''0'•~...100 -100--gil Z 99Zgil98e0....g7 I-117c(81 -r gil>115 >98l&I UJ..,J 114 nile"..J 94 1 ~-I ))8 ~"UJ U UJ 93 l&I U UJ 92>III >91....110 ~90c(1111 -r 89..,J ..IUJ118 UJ 880::67 0::87lIll 86118 8804080120160200240280 0 40 80 120 1 liD 200 240 280 DISTANCE FROM '"EFT BANK HEADPIN (fee I)DISTANCE FROM LEFT BANK HEADPIN (ful) Appendix Figure 0-6.Cross section of transects 4,4A.~·~and 6 at Island Side Channel (adapted from,Quane et. a1.1985). $ DRAFT/PAGE 3,4/30/85 4/18/85,4/10/85 ANDY/Do~3,5/2/85 areas in this reach.The four downstream most transects"are primarily in pool type habitat.Transect lA was also a partial transect,repre- ,senting the smaller slack water area along the right bank. Data Collected Hydraulic data were collected at a site flow of 338 cfs (Appendix Table 0-4).The .mean daily discharge for the Susitna River on the date the cal ibration data were collected at the study site was 56,100 cfs as determined from provisional USGS streamflow data. Calibration Calibration data available at the close of 1984 field season was limited to that obtained for a side channel flow of 338 cfs (56,100 cfs mainstem discharge).As a result,an IFG-2 model was ~sed to forecast instream hydraul ics based on this single cal ibration flow.The streambed pro- file,.stages of zero flow,and observed and predicted water surface elevations for this study reach are plotted to scale in Appendix Figure 0-7". The original fiel d water surface elevations (WSEL I s)were compared to the model predicted WSEL's for the calibration flow of 338 cfs (Appendix Table 0-5).At transect lA,the original field WSEL was surveyed at 93.46 feet.In examining the WSEL I s of transects 1 and 2 (93.33 and 93.41 feet in el evati on respectively ),it was fel t that an error in 0-d~ - ..... 1 1 j J .~I 1 )t j 1 I J ~) I ~ '0+00 ~~ STlltAIolBED STATION 'h ... ~I~ t.va ISLANOSIDE CHANNEL Thalweg Profile with Observed'ond Predicted Woler Surfoce ProWn Thol.'V Grlldi,nt.i6,6fulllftlle Obu,ved Wot,'S"foc.Elevotion Simulolld Wol,f S~rfou Elevotion E;xlropoloted Wot.,Surflic,EI.votion Elevotion 01 Ztrll Flow .......d Thol.'v Prof ii' W////P_._.-.-.-._._.-._._._.-.-._.-._._.-.-.-._._._.-.~._._._._._.-._._.-.-._.-._._._,,} •••."••III ell E.'t"14!eli ••tlilt_,."'lI'.....tNIII.' ---.-----._.---._._._._.----_._.--_._--.-.--_._._._._._.-...----_.---._._."-_._..",. .._-------- ~",,,..«._-~ n""un.TII.a".,c-f..fItA"IUI,"'...ner"'It"llle'.'.'"'In 4A fltAlIIIlel.n ....tc...I.'4e ,...I:I ,'SLANg SIDE CHANNEL \~)--:2 1 \\I/"'~"":~~'("' j A ~r .? I ~~::::::s:::;...-?\".-~~'\A'•~oo;'£4 ,,:::::::::/'*:' h...;~. IAwoI·III:0411 .. .. ~ .! z 0;::..>..III ..IIII...>.. S ..I Ii!.. .. .. 0.00 tJ I ~.J <Jl 1 Appendix Figure 0-7.Comparison of observed and pred1cted water surface profil es from calibrated model and surveyed thalweg profile at Island Side Channel (adapted from Quane et~al.1985). DRAFT /PAGE 6 4/19/85,5/2/85 ANDY/Tables Appendix Table D-5.Comparison of field measured and model predicted water surface elevations at the calibration flow of 338 cfs for Island Side Channel .. Water Surface Elevation (ft) Transect Fiel d Model Predicted Di fference 1 1A 2 3 4 4A 5 6 93.33 A93.46 93.41 93.44 93.48 93.52 93.56 93.55 93.33 93.36 93.36 93.40 93.46 93.50 93.53 93.56 0.00 0.05 0.04 0.02 ,0.02 0.03 0.01 - A Water surface elevation reduced by 0.1 feet to 93.36 feet. O-'d-(P - DRAFT/PAGE 4,4/30/85 4/18/85,4/10/85 ANDY/Doc 3,5/2/85 surveyi ng occurred at transect lAo As a resul t,the WSEL for thi s transect was lowered by 0.1 feet to 93.36 feet.For all other transects,the difference between the fiel d WSEL I S and the model pre- dicted WSEL's for the calibration flow were 0.05 ft.or less. The two partial transects (lA and 4A)which represent slackwater habitat were extended out to the principal velocity~filament.In order to complete the data sets for these two partial transects for use in the model,the associated data from transects 1 and 4 were used.At partial transect lA,the vel ociti es were all negati vee In order to use thi s information in the model,these velocities were treated as positive,as it was felt that the direction of the current would not influence the utilization of this area by juvenile salmon.With respect to the amount of water flowing through this section,it amounted to only 6.5 cfs or about 2%of the flow. Verification Based on the first level of verification conducted by EWT&A,the model does an excellent job of simulating hydraulics between 35,000 and 56,000 cfs mainstem discharge (69 and 416 cfs site flow).Above 56,000 cfs, however,the simulated depth and velocity distributions begin to deteri- orate in quality.As a result,the model simulations were rated good between 56,000 and 64,000 cfs(416 and 692 cfs site flow),acceptable between 64,000 and 70,000 cfs (692 and 984 cfs site flow),and unaccept- able above 70,000 cfs mainstem.Below35,000cfs mainstem"insufficient Application Range of the at Island Cali bra ted Hydraulic Model Side Channel RM (63~2) Site Specific Flow.cfs o B 38 115 271 545 984 1283 . D \ P CP I I ,I ~ II -'"'"'"'"" I I I I o 10 20 30 4'0 50 60 70 75 Mainstem Discharge at Sunshine Station.cfs x 1000 1m Excellent II Good •Acceptable 0 Unacceptable Appendix Figure 0-8.Application range of the calibrated hydraulic model at Island Side Channel. I I J.~J c ..1 1 1 J 1 - ""," - DRAFT/PAGE 5,4/30/85 .4/18/85,4/10/85 ANDY/Doc 3,5/2/85 data was available to evaluate the performance of the model.These ratings are depicted graphically in Appendix Figure 0-8. The second level of the verification has not been performed as of this time. Ap P1teat ion For habitat simulation modelling.purposes,the hydraul ic simulation model developed for Island Side Channel·can simulate channel flows in the mainstem discharge range of J5,OOO to 70,OOOcfs. DRAFT/PAGE 1,4/30/85 4/18/85,4/10/85 ANDY/Doc 2,5/2/85 Mainstem West Bank Side Channel (RM 74.4) Site Description Mainstem West Bank Side Channel is located on the west bank of the main channel Susitna River at river mile 74.4 (Appendix Figure D-9).It is approx~mately 2.2 miles in length.Both the mouth and head of the side channel directly connect to the Susitna River.Two heads,both located approximately 1.5 miles upstream of the study site,connect this side channel to the mainstem (Quane et al.1985). The IFG modelling site within this side channel during the 1984 open water field season was 930 feet in length and was located in the lower portion of the side channel (Appendix Figure 0-9).The side channel within the study site is confined on the west by a steep bank and on the east by a well vegetated island which separates it from the mainstem. The upper portion of the side channel upstream of the study site is separated from the mainstem by a network of side channels and well vegetated islands.A minor channel is located within the study site on the east bank of the side channel.During nonbreached conditions, the side channel primarily consists of a series of pools and small riffles.Groundwater provides the major contribution of flow prior to breaching of the head (Quane et al 1985). Breaching of Mainstem West Bank Side Channel occurs as the result of overtopping by the mainstem of at least one of the two side channel heads located approximately 1.5 miles upstream of the study site.The D-30 - I 1 1 1 )I }})l 1 l ~1 ~:.....~ .:). .'•.~~...~.,.'!',;"', o, w--.. Appendix Figure 0-9.Overview of Mainstem West BankSfdeChannel (RM 74.4). DRAFT/PAGE 2,4/30/85 4/18/85,4/10/85 ANDY/Doc 2,5/2/85 side channel has been estimated to be -initially breached at a mainstem discharge of 19,000 cfs (Quane et al.1985). Based on a review by Quane et al.(1985)of the stage versus mainstem discharge rating curve (Appendix Figure D-10),it has been determined that at mainstem discharges greater than 19,600 cfs,the hydraulics within this side channel are directly controlled by.mainstem discharge. The site flow that.occurs at 19,600 cfs was measured to be 5.7 cfs. Located within this study site were five transects (I,2,3,3A,4)in the main channel and three transects (2A,3 in part,38)in a minor side channel from which hydraul ic information was gathered (Appendix Figure 0-11).The corresponding cross sections are presented in Appendix Figure 0-12 &13. The lower two transects (1,&2)bisect pri.marily pool-run type habitat where the banks are gently sloping on both sides.On the upper three transects {3,3A,&4)the left bank consisted of an erosional bank and was primarily bordered by alder.For modelling purposes,transects 3 and 3A were ended on a finger-liKe gravel bar on the right bank which longitudinally bisected the site with the main channel on the left and a minor channel on the right which was free flowing at high flows, backwater at median flows,and dry at low flows.Th-is bar began downstream from transect 4 and ended between transects 2 and 3. Transect 3A was placed in order to obtain a better representation of the slow water debris-strewn habitat along the left hand bank.The main D-3~ - - - - .. ! Q=5.7 cis -,--,~ MAINSTEM WEST BANK SIC TR I GAGE 74.451 -i ,---o,.c .,0-'·JU'ClJ.s Z•aazl r Z .0." ,'·····.t ..I':ir gr U) ~I ;,1 I l:l....t - - - - ••IX "".>eIt.'\o:~"""""'lie f10ll0D":1 \ - - Appendix Figure 0-10.Comparison of rating curves for Mainstem West Bank Side Channel transect l(Q site)(from Quane et.al.1985). o I ·t~J.z. ,.,..""'Io_~,",:",'Jt 6 Stall Gage Traneect o 260 I I FEET (Approximate Sca'e) '....\~~ .- ...~~. ~ _J Appendix Figure 0-11. t Location of Mainstem West Bank Side Channel ~.t~cJjY S••i~jJ (R~.Zj·41..cl .)J I J J I l 1 }~l 1 ~)J l J 1 1 400300 10.' 104 1CROSS·SECTION 2 103 STATION 1+66-102OJ OJ 101-100 Z llll 0 1111 j::117 ~e8 I<J II'....114I<J 113 I<J 112 ~111 I-110octe9.... I<J 1111 II:e7 1111 811 0 100 200 300 400 DISTANCE FROM LEFT BANK HE.ADPIN (fe.11 200 '-.*'i lih...........:> 100 200 300 400 DISTANCE FROM LEFT BANK HEADPIN (f.ell lOll 104 103 4i 102..101..100 911 Z 9110 j::97 oct 911 >11&W....94 I&J 93 I<J 92 >III I-110 oct 1111.... I<J e8 II:87 811 e, 0 100 CROSS SECTION I STATION 0+00 10~ 104 103 '0 10~ 0)101...100 :i!: 1111 0 e8 ~117 oct 8lI :>II~ I<J -J II' I<J e~ I<J 112 :>111 f-eo oct 811..J I<J 88 0 II:87 116 IIll W 0 Ul DISTANCE FROM LEFT BANK HEADPIN (fut) Appendix Figure 0-12.Cross section of transects 1,2,and 3 at Mainstem West Bank Side Channel (adapted from Quane et.a1.1985). D'ISTANCE FROM LEFT BANK HEADPIN (feet) 400300200100 DISTANCE FROM LEFT BANK HEADPIN (feet) a 105 j104 CROSS SECTION 4 I 103 STATION 9+32 102 101 100 99 118 117 \lIS P5 04 OJ 02 Sll 80 88 88 117 88 8!i -I iii i i 8 I I ......-U 011... zo I- ~ W ...J LLJ LaJ>" I- <t ...J Wc:: 300200100o 105 j . 104.CROSS SECTION.3A I 10J.STATION 5+62 102 101 100 ::~.,.97 98"..."....94 310 cit 93 6 ~. 92 91 90 89 88 87 86 85 I I Ii,iii' 400 zo i=<t>LaJ ..J LaJ W> I-<t ..J LLJc:: ......., III....... o ., Wo Appendix Figure 0-13.Cross section of transects 3A and 4 at Mainstem West Bank Side Channel (adapted from Quane et. al.1985). )j )J .... - :~ ..... DRAFT/PAGE J,4/30/85 4/18/85,4/10/85 ANDY/Doc 2,5/2/85 channel habitat of these three transects (3,3A,&4)consisted of run-riffle type habitat. Substrate at this site primarily consisted of rubble and cobble.The thalweg gradient of the side channel is approximately 12.3 ft/Tnile (Quane et aT.1985). Data Collected Hydraulic data were collected for model calibration at three discharges: 6,310,and 450 cfs (Appendix Table 0-4).Mean daily discharges for the Susitna River on the dates that calibration data were collected of this study site were 19,600;30,500,and 32,000 cfs,representively as determi ned from provi si onal USGS streamflow data. Calibration Calibration data available at the close of the 1984 field season includ- 0-31 EUT "N'"'I/OR eNANNEL ~ ..400 n ...uc,...,...""ICT • '........et a. I"IN'."~ nllE/HOEO STATION lIull If !..•5!..:tJ J...........t.;: c... ~ It ••00 Thol ..,S.,v.,D.I.·141010 .n.I••,G,041 ••,.la.1 ,..11..11. O~..,..4 W.'"SU,'.0.,".v.ll0. Si...,01.4 WOI ..$0"000 11••011 .... 1lllOpolOI04 Wol ..S.rfo.,1E1••01l00 1'''01100 0':"0 Flo.'w#Tllal""P,olllo MAINSTEM WEST BANK SIDE CHANNEL TlIalw'lI Ptolil.with Ob ••rved and Pr.dlc'ed WOler Surfoc,ProW ......,,~."Of ,..."pi CM""'./ J r ••,-_.._._._.._.-._._.-._._._._..-._._..-~lif!:::r ef ;--_._---;~-------~~_._._._.-._.-._.-._._.~~.....,~._._.-._.-._._.-.-._._._.-.-.-._.-.-._._._._._._._._.-.--:_____Ml...~;:~~ ...,.,....,./~~T7......" 1 UANlle,I / AO,••tdt-,..... 'U ..,t:e,• lAO'.'•..:...,..tt·tt'''.US ,....,. II t_.ltICC,.,.....Un ....I '••IIUel'• I',....I ••• ..00 10-+00 noUIISEO STATION 11..11 Appendix Figure 0-14.Comparison of observed and predicted water surface profiles from calibrated model and surveyed thalweg at Mainstem West Bank Side Channel (adapted from Quane et.all 1985). !J I J J J J J I •1 J ) - ".... - ~. - DRAFT/PAGE 4,4/30/85 4/18/85,4/10/85 ANDY/Doc 2,5/2/85 To evaluate the performance of the hydraul ic model,observed and pre- dicted water surface elevations,discharges,and velocity adjustment factors were compared (Appendix Table D-6).The 15 sets of observed and predicted WSEL's for the five transects of the 3 calibration flows were all within ±0.02 ft.of each other except for 2 sets which were within ±0.10 feet of each other.All the observed and predicted discharges were within 10%of each other and an velocity adjustment factors were within.the good range of 0.9 to 1.1.Additionally,the stage information of the model was compared to available rating curves (Appendix Figure D-I0). To represent the slackwater debris area along the left bank of the upper portion of this stUdy site,a partial transect (3A)was placed about 60 feet upstream from transect 3.In order to complete this data set for transect 3A for use in the model,.the velocity information from transect 3 for the two site flows of 310 and,450 cfs were incorporated into transect 3Across sectional area and water surface elevations.After incorporating this information into transect 3A,the discharge for the 310 cfs site flow,however,did not fall within 10%of the respective discharge that was calculated at the discharge transect.As a result, velocities for the 310 cfs site flow were adjusted upward by 17%. At the low flow measurement of 6 cfs,the velocity measurements were made completely across transect 3A.The discharge calcu·lated at this site was 18%higher than calculated at the discharge transect.The velocities at this transect were therefore reduced by 15%. ------_......._-------'--'-----...,.-------....:....,-~--~---,------_......._-- - DRAFT/PAGE 2 4/23/85 AHMR/Appendix Tapl~3 Appendix Table 0-6.Comparison between observed and predicted water surface elevations,discharges,and velocities for 1984 Mainstem West Bank side channel hydraulic model.- Streambed Water Surface Station Elevation Discharge Velocity -Observed Predicted Observed Predicted Adjustment ! (ft)(ft) (ft)(cfs)(cfs)Factor 0+00 92.85 92.86 6.0 6.3 1.005 1+66 92.86 92.87 6.9 7.2 .991 5+08 93.25 93.26 6.9 7.2 1.004 -5+62 93.51 93.52 5.8 6.1 .996 9+32 95.06 95.06 5.1 5.4 1.013 Qo =""6:0 Qp =6.0 - 0+00 94.62 94.61 312.8 315.7 1.030 1+66 94.64 94.64 301.3 307.5 1.024 5+08 94.85 94.86 306.4 318.2 1.007 5+62 94.93 94.99 292.8 288.6 .993 Qo =301.0 Qp=308.0 0+00 94.97 94.98 460.4 457.0 .974 1+66 95.00 95.00 446.1 438~2 .975 5+08 95.19 95.18 470.6 455.2 .994 -5+62 95.29 95.23 409.6 415.3 1.001 9+32 96.54 96.45 473.9 451.9 .969 Qo =452.0 Qp =444.0 00 is the mean observed calibration discharge. Qp is the mean predicted calibration discharge. - - - D-4 0 .- - ..... ..... - DRAFT/PAGE 5,4/30/85 4/18/85,4/10/85 ANDY/Doc 2,5/2/85 At transect 4 the water surface elevations were not similar across the transect at the 6 cfs flow measurement.Therefore,a weighted average water surface elevation was calculated for this transect. At higher site flows several small low velocity side channel/backwater areas existed.It was felt that this habitat,which was not represented in the IFG-4 analysis,would be an important area to assess.Because of this,three transects were placed across one of these minor side chan- nels.These transects were to be used to hand calculate the habitat in this area.However,because this side channel area is so small compared to the total area being modelled using the IFG-4,it was felt that including this area in the total weighted usable area calculations would not truly reflect the value of this habitat.For this reason,hand calculations of these areas were notdbne. Verification Based on the first level of verification by EWT&A,the model does an excellent job of simulating channel hydraul i cs between 18,000 and 21,000 cfs mainstem discharge (6 and 20 cfs site flow)(Appendix Figure 0-15). Above 21,000 cfs,simulated water surface profiles deviate somewhat from field observations.As a result,the model was rated good between 21,000 and 28,000 cfs mainstem discharge (20 and 200 cfs site flow),and between 28,000 and 34,000 cfs mainstem discharge (200 and 500 cfs site flow)the model again was rated excellent.Two calibration data sets were collected within this range.Above 34,000 cfs,the quality of the D-J...}I Application Range of fhe Calibrated Hydraulic Model at Mainste-m .West Bank RM (74.4) Si te Specific Flow.cfs 0 13 307.690 1080 .1555 .2118 2431 I I .I. 0, ...t 't' I I 111111'0 111 1111111111111 .ir. 0 10 2()30 ,...~()50 60 70 75 Mainstem .Discharge at,Sunshine Station,cfs x 1000 1m Excellent -Good -Acceptable 0 Unacceptable Appendix Figure 0-15.Application range of the calibrated hydraulic model at Mainstem West Bank Side Channel. J J I J .",.~J J J J •.1 (I .,J J I I ]..) .~ .- - - .- DRAFT/PAGE 6,4/30/85 4/18/85~4/10/85 ANDY/Doc 2,5/2/85 hydraulic simulations begins to deteriorate as the slope of the site flow versus WSEL relationship flattens as a result of channel'geometry. The deviation between the regression line developed within the model and that of the rating curve developed independently for the site increases with dischar;-ge until t~e model simulations are no longer acceptable. The model simulations were rated good between 34,000 and 41;000 cfs (500 and 727 cfssiteflow),acceptable between 41,000 and 48,000 cfs (727 and 1000 cfssite flow),and unacceptable above 48,000 cfs mainstem dis- charge • Overall,the model simulations were rated excellent between 18,000 and 21,000 cfs (6 and 20 cfs)and 28,000 and 34,000 cfs (200 and 500 cfs), .good between 21,000 and 28,000 cfs (20 and 200 cfs)and 34,000 and 41,000 (500 and 727 cfs).They were acceptable between 41,000 and 48,000 cfs (727 and 1,000 cfs)and unacceptable over 48,000 cfs. As of this time,the second level of the verification has not been performed. Application For habitat simulation modell ing purposes,the hydraul ic simul ation model developed for Mainstem West Bank Side Channel can simulate channel flows in the mainstem discharge range of 18,000 to 48,000 cfs. 0-~3 DRAFT/PAGE 1,4/30/85 4/18/85,4/11/85 ANDY/Doc 6,5/2/85 - Circular Side Channel (RM 75.3)- Site Description - Ci rcular Side Channel is located on "the west bank of the Sus itna Ri ver at river mile 75.3 (Appendix Figure 0-16).It is approximately 0.9 miles long and is separated from the mainstem by a large well vegetated island.Both the mouth and head of this side channel are connected to - - the mainstem Susitna River.An extensive backwater area has been observed to occur in the lower portion of the study site.A network of small channels at the head provide mainstem flow into the site after breaching.Prior to breaching,flow is greatly reduced and the channel "'IIIl is composed of large pools connected by small riffles (Quane et al. 1985). Breaching of Circular Side Channel is the result of direct overtopping of the head by themainstem Susitna River,and has been estimated to be initially breached at a mainstem discharge of 36,000 cfs (Quane et al. 1985).It has been determined that the hydraulics within this side channel become governed by mainstem discharge at mean daily mainstem discharge exceeding 36,000 cfs.The site flow that occurs at this mainstem discharge has been estimated to be 26.8 cfs (Appendix Figure 0-17)(Quane et al.1985). - - Based on assessments by Quane et al.(1985),backwater has not been observed to occur during non-breaching mainstem discharges.At breach- ;ng rna;nstem di scha rges of 55,200 to 66,700 cfs,however,an area of 1 1 J J ,-))1 ])1 J ]l -)]J o \ L (}) .'- ,,/~.>t..-'").,-,,:,-.;.t ..' " '1'1 ..,~ "..f'c ~), ..,Ir. Appendix Figure 0-16.Overview of Circular Side Channel (RM75.3). !,.....-.'~-~..-..-oIL..."\•'"~t".1 ....~3'.....CC1,.:r.~ ! '"', •• .0=2.6.8 cfs .-------.---t I I~ I~Ig I~ ! 0=26.8 cis . ! CIRCULAR SIDE CHANNEL TR4 GAGE 75.354 - Appendix Figure 0-17.Comparison of rating curves for Circular Side Channel transect 4 (from Quane et.al.1985). - -- DRAFT/PAGE 2,4/30/85 4/18/85,4/11/85 ANDY/Doc 6,5/2/85 backwater was found to occur upstream toa point approximately 90 feet above transect 2A.At a mainstem discharge of 42,500 cfs,backwater has been determined to extend slightly past transect 2. The IFG modelling study site within Circular Side Channel was 820 feet in length and was located in the upper half of the side channel (Appendix Figure D-18).The thalweg gradient of this study site is 14.3 ft/mil e (Quaneet al.1985).Riparian vegetation along both banks of this study si te consists mostly of al de.r and cottonwood.Substrate within the lower reaches of the study site consisted predominately of silts,sand~,and gravels changing to rubbles at the upper reaches.Six transects from which hydraul ic information was gathered for the model were located within this study site (Appendix D..,18).The channel is relatively straight and the cross~ections are generally box shaped in configuration {Appendix Figures 19 &20}.Transects 1 and 2 were located in~hallow pool habitat,created by the backwater.·Transect 2A was located in transitional habitat which became run-like habitat at higher flows.Transect 3 was located in riffle habitat.Transect 4 was located in a run area at the end of a pool area which transect 5 also bisects. Data Collected Hydraulic data were collected at two cal ibration discharges:50 and 204 cfs (Appendix Table D-4).Mean daily disc;harges for the Susitna .Rive.r on the dates that calibration data were collected at the Circular d •• .J:.. (P ...~...·..,'.i,,'j.~,. 91"'",·····;'. ,I )' .:.~" ·1'\.:,'"..../"''''..... .'... ·r .. .w.-.,::~ :~r;~,'(.;~~.'........,.'~.,;:~.,,~•~./"'.''-.''..,..,-,:..1-I{.~y:~~ _.,,:.~A~'.',...p,",'15.3~1~"'::'"'l":',.~":"i ,........_',.~...,',..--.... '.-.'":-';''':- .~."\ ... "'t.......'0'.;·.~.:~.,.".~."'L"·~~·,..... •,.Y ~,.,:". ,. ",~'t-~:::,~"'1>,>'1>'/ Appendix Figure DdS.Location of Circular Side Channel study site (RM 75.3)• .1 J )"I )J j J }}1 -B )]1 )1 ))J ")J :zeo240;lQQ1110120 2110 110 ;Z40 40 DISTANCE FROM LEFT BANK H~ADPINH ••IJ ;zoo \.I=.<:c:ii3--....I.,Q4.I'....--~~o.,. CROSS SECTION 2 STATION 1+98 180120 :laQ DO40 240 CROSS SECTIOH2A STATION 2+65 ;ZClO100 '05 'CIt-'0)..'02.:!lOt-11l1l· ~u· 911 •j:",.:~a . IloI '5..I 'tILl 8'ILl U>8,j: 80« -'08 ILl lIDII:117 so as 0 no11040 DISTANCE fROM LEfT BANK HEADPIN n.oll CROSS SECTION I STATION 0+00 'CIS tat 10)..t02 ~lOt-100 :l:u g ,.....:87 ~o ILl 85 ,.I ILl 8t ILl 83 >82 ;:81 «80 ..J 08 (') III lID Q:.7 • U u ..L a -0 DlliTANCE FROid LEFT BANK HEADPIN tint) Appendix Figure 0-19.Cross section of transects 1,2,and 2A at Circular Side Channel.(adapted from Quane at.al.1985). 2110240200111012011040 ~~~~ "£~~ CROSS SECTION 4 STATION 6t63 lOS CROSS SECTION 3104 las-10J STATION 4+33 104 -102 10J .. ~101 ..102 100 •101...Z 88 -100 2 811 Z lit ....87 0 88 ~ j:: >118 117 ....liS ~118 ..J....114 IlJ liS 83 ..J 114 \.tJ ....>112 Il3i=111 204 .,.....112> ~110 ~O .,. t=111 ..J 811 110 ..../18 ~ 0:: ..J 811 0 87 IlJ /18It: «/Ill 87<I,1I11 1111 0 40 110 120 1/10 200 240 2110 liS0 0DISTANCEFROMLEFTBANKHEADPIN('ee') DISTANCE FROM LEFT BANK HEADPIN (teel) CROSS SECTION IS STATION 8 t 20 lOS 104 103-102-..101.....100 lit Z 8110117i=lI/I~liS....114..J....Il3 ....112 >111 5 110 /Ill..J 1111.... It:117 1111 liS 0 40 110 120 1110 200 240 2110 DISTANCE FROM LEFT BANK HEADPIN Oeel) Appendix Figure 0-20.Cross section of transects 3,4,and 5 at Circular Side Channel. t j J •))-~J t )I - - DRAFT/PAGE 3,4/30/85 4/18/85,4/11/85 ANDY /Do.c 6,5/2/85 Side Channel study site were 42,500 and 55,200 cfs as determined from provisional USGS streamflow data. Calibration Calibration data were available at the close of the 1984 field season for side channel flows of 50 and 204 cfs.An I FG-4 mode 1 was u sed to forecast instr~am hydraulics based on these two calibration flows.The streambed profile,stages of zero flow,and observed and predicted water surface elevations for the study reach are plotted to scale in Appendix Figure D-21.The two data sets were used to predict hydraulic informa- tion from side channel flows of 6 to 733 cfs (mainstem discharges of 25,500 to 75,000 cfs). To evaluate the performance of the hydraul ic model,observed and pre- dicted water surface elevations,disch~rges,and velocity adjustment factors were compared (Appendix Table D-7).Because of the 2 cali- bration flows only a 2 point rating curve was formulated.In evaluating the performance of the model,observed and predicted WSEL I sand di s- charges were the same because of this rating curve.Velocity adjustment factors were all within the good range of 0.9 to 1.1.Additionally,the stage information of the model was compared to the rating curves estab- lished by Quane et al.1985 (Appendix Figure 0-17). At the high flow measurement of 204 cfs,the original field measured discharge at transect 2 was 34%lower than that calculated at the discharge transect.In order to use this information in the model,the D-sl -Ji1lMI_..__.--~~---""""""'------'-'-------------___ '0'•••at«:u.ln-.--- Thol.'1I GradiuI'"14.3 ,..t/mil' ObJ.,"'d Wolo,Surface EI",olion Simulol.d WOI.,$ur'Ol;'Elnotlon E"rapolol.d WoI.r $ur'oc;.Elnollon ElnQlion a'Z~ra Flo. Thal.'11 Pralll.•._._._."',,,} •....__............-..204 1;'.E••'8'I f ,:_._._._._.;._~_.;;..._._._;~:::~r····II .::.-.-.-.-."..-.-."".-------------~-------:------..--.._.~._..,....-_.--~ CIRCULAR SIDE CHANNEL Thalwell Profil.wilh Obs.rv,d and Predieled Wahlr Surfae.Profil n ~ ;! o lOll" '---',..,..""".....~.. " ,. to. 01 " ...>.. ~......a:,0 z ~:...... w 0;:;VI ~ CJ, 10 'U"S~CJ •'''''Nsce ..I ,.ANueT J ,,,,,..steT 4 ,,,,,,,sur,.....ue'•.& JEll ••,F...-I,.I . ,i •iii , i ; •i.i'I iii ,.1 'i"iii ' i ' , ,i ,i ; , •6 a", 0 ...00 I "'tOO to+OO .....00 10-t00 .,+00 STREAMBED STATIO"I h.,I Appendix Figure 0-21.Comparison of observed and predicted water surface profiles from calibrated model and surveyed thalweg profile at Circular Side Channel (adapted from Quane et.al.1985). I J J _c .J '...~.J J J J )J Appendix Table 0-7. DRAFT/PAGE 4 4/23/85 AHMR/Appendix Table 3 Comparison between observed and predicted water surface elevations t discharges t and velocities for 1984 Circular Side Channel hydraulic model. Streambed Water Surface Station Elevation Discharge .Vel od ty Observed Predicted Observed Predicted Adjustment (ft)(ft)(ft)(cfs)(cfs)Factor jllll:llllllr, 0+00 89:28 89.28 44.4 44.4 1.000 1+98 89.30 89.30 47.9 47.9 .998 2+65 89.41 89.41 56.0 56.0 1.000 ~4+33 90.20 90.20 43.7 43.7 1.000 6+63 90.60 90.60 50.9 50.9 .997 8+20 90.62 90.63 53.6 53.6 1.000-QO =49:1i Qp ='1f9':O 0+00 90.29 90.29 202.8 202.8 .998 ,.-1+98 90.27 90.27 203.1 203.1 .987 2+65 90.31 90.31 198.4 198.4 .999 4+33 90.66 90.66 176.9 176.9 .998 6+63 91.29 91.29 199.9'199.9 1.000 8+20 91.32 91.32 194.2 194.2 1.000 Qo =196.0 Qp =196.0 Qo is the mean observed ca librationdischarge. -Qp is the mean predicted calibration discharage. -- --_._-------........,-.........--..,"--'-'--'--.......,.""'----_.........------'-----------....-;"'--- DRAFT/PAGE 4 t 4/30/85 4/18/85 t 4/11/85 ANDY/Doc 6 t 5/2/85 - individual velocity measurements were all adjusted upwards by 52%.Why there was such a large discrepancy between flows at this particular ~ transect when the four other transect flow measurements were within 9% of the discharge transect measurement is unknown. At transect 5 there was a change in the channel cross section from when the actual cross section survey was done and when the two calibration f1 ows were made.Between the cross sect i on su rvey of September 5,1985 t and the two calibration flow measurements July 24 and August 17,1984,a flood event occurred on August 26,1984.After this flood t the right side of the channel at transect 5 was scoured out.In order to avoid violating one of the underlying assumptions of the model,(i.e.,that a rigid stream channel exists)the cross section determined from the two calibration flows was used in the model. During the 50 cfs ca,libration flow measurement a water surface elevation was not surveyed for transect 5.In order to obtain a water surface """'!i Uh elevation for the model,a value was calculated from the average of the" depth measurements added to the corresponding cross section elevations of the 50 cfs flow measurement. Verification Based on the first level of verification by EWT&A t the model does an excellent job of simulating channel hydraulics between 39,000 and 57,000 cfs,mainstem discharge (38 and 213 cfs site flow)(Appendix Figure 0-5~ - - DRAFT/PAGE 5,4/30/85 4/18/85,4/11/85 ANDY/Do~6,5/2/85 D-22).Above 57 ,000 cfs,the simul ated depth and vel ocity distributions begin to deteriorate in qual ity.The model simulations were therefore rated good between 57 ,000 and 60,000 cfs (213 and 268 cfs site flow), .\WlIii1Il acceptable between 60,000 and 63,000 cfs (268 and 334 cfs site flow), - ,- and unacceptable above 63,000 cfs mainstem discharge (Appendix Figure ". 0-22).Below 39,000 cfs,the model simulations were also .rated less than excellent as forecasted velocity and depth distributions deteri- orated in quality.The model simulations were rated good between 36,000 and 39,000 cfs mainstem discharge (27 and 38 cfs site flow).(Appendix Figure 0-22).Below 36,000cfs mainstem (controlling discharge), insufficient information is available to evaluate the model. The second level of the verification has not been performed as of this time .. Application For habitat simulation modell ing purposes,the hydraul ic simul ation model developed for Circular Side Channel can simulate channel flows in the mainstem discharge range of 36,000 to 63,000 cfs. 0-55 ___~_","""'~Iil """__,,;,-,~,,,,,:--,-,--_ ~~..~.,~Pt-.._A..-...,.....~,..~-,.~,.~._,--~"";"'-"""",.~_._,..•...~'-.,-.-.........-.~.."'..,-,..~, Application Range of the Calibrated Hydraulic Model at Circu Ia r-Sid~Channa I RM (15.3) Si te Specific Flow,cfs 0 2 12 43 ..U8 268 .537 733 I I ,,..,EEi3lln UI 1111111 1111 III III 1l'IIH f111111111111 g d, VI h"- I i I I 0 10 20 30 40 50 60 70 75 Mainstem Discharge at Sunshine·Station,cfs x 1000 I ~Excellent II Good -Acceptable 0 Unacceptable, Appendix Figure 0-22.Application range of the calibrated hydraulic model at Circular Side Channel. .J )J "I I I I I )I J J ~J .1 J J ),J DRAFT/PAGE 1,4/30/85 4/14/85,4/19/85 ANDY/Doc 9,5/2/85 Sauna Side Channel (RM 79.8) Site Description Sauna Side Channel is located on the west bank of the Susitna River at river mile 79.8 (Appendix Figure 0-23).It is approximately 0.2 miles ,long.Both the mouth and head of the side channel·are connected to a larger side channel of the mainstem Susitna River.For the most part, the si de channel is confinedonthe<west side by ahi gh bank and on the east by a large sparsely vegetated gravel bar.A smaller side channel enters just below the head of Sauna Side Channel on its west bank.This side channel conducts flow to the study site during high mainstem discharges,but dewaters before the head of Sauna Side Channel becomes unbreached.Breadling flows result from overtopping of the side.channel that adjoins the head on the east bank of Sauna Side Channel.Prior to breaching,the channel is composed of two large interconnected pool s whose water level s are maintained from ground water seepage originating from the vicinity of the head.An extensive log jam exists at the head of Sauna Side Channel that likely influences the flow into this side 'channel. Based on assessments by Quane et al.1985 breaching of Sauna Side Channel is the result of overtopping of the head of the side channel by the adjoining side channel.Based on fiel d observations and stage/discharge relationships,the mainstem discharge estimated to initially breach Sauna S1 de Channel was 37,000 cfs.A controll 1ng discharge of 38,000 cfs was determined for this side channel also based - - - " RM79E;9 ...,.~1 /'~.~ ./i".....,....'./. -., '.~ \ .~ " •I'. ." .., .)/.'Y .-~..\..y''':'.' "';' ,j.~ E9 River Mile o 500 ! ,I FEET (Approx.lmate Scale' Appendix Figure 0-23.Overview of Sauna Side Channel (RM 79.8). n~>·~...~· -. - .- ,.,., I - DRAFT/PAGE 2,4/30/85 4/14/85,4/19/85 ANDY /Do·c 9,5/2/85 on this stage/discharge relationship.A side channel flow of 22.5 cfs has been estimated to occur at the 38,000 cfs mainstem discharge as derived from the stage versus streamflow rating curve (Appendix Figure" D-24).Based on a review of the 1984 stage data and thalweg elevations by Quane et al (1985),it has been determined that backwater does not occur in Sauna Side Channel during non-breaching mainstem discharges. Duri ng breaching di scha rges of 54,600 to 56,700 cfs,however,the a rea of backwater was observed to occur throughout the Sauna Side Channel study site.The IFG modelling site within this s.ide channel during the 1984 open water fiel d season was 480 feet in 1ength and located approxi- mately 2,000 feet from the mouth of the side channel (Appendix Figures D-23 &25).The thalweg gradient at this site is 10.4 ft/mile (Quane et a 1.1985)wi th'substrates throughout this site cons;sting primari ly of sands and silts.The water is slow moving with"velocities usually less than l.0 ft/sec.The left bank at this site is a erosional bank with a height exceeding five feet.Riparian vegetation along this bank consists of alder and birch,in contrast,the left bank is a depositional bank with no riparian vegetation. Four transects were located within this study site (Appendix Figure D"26).Transects 1 and 2 were located in shallow pool habitat whereas transects 3 and 4 were located in deeper pools. Data Collected Hydraulic data were collected at a calibration di.scharge of 52 cfs (Appendix Table D-4).The mean daily discharge for the Susitna River on !! On •IClO.lZ~(1/SEl,-151 Z.1IS9Z ,Z 0 a.9Z Q:22.5 cfs 0u 0 ur"IZ14 0..Z.ll" ,z •0.10 .. I~ 18. I~ 1° l! ~r~~ .~'% C C ~~ ....i3~ ~c~. ..,t-...L..~..~---------_...,J-.--ll.-_ MI'lSlC11 atStI\.~OlIG::ftf ~11C n'l.':lQCn'~\0 =rtty,.. COIITROI.lO _.CIIIO ~CI ~11,1011 ef~ vsn •10·J·1I06«r'554 •IS ,z CI.t) lICIT COIITlQlLlCl 15.DOO~0 ~Ja.roo eft 1IO£~1l1ll OE'tlOP£O SAUNA SIDE CHANNEL TR2 GAGE 79.8S2 -.~.--------~~-~--------- 1!'I-(t'4)J[""O[X'1l'MtX itT su--"....I"'C t It:J00Cf"51 Appendix Figure.0-24.Comparison of rating curves from Sauna Side Channel transect 2 (from Quane et.al.1985). 0-tt,D FEET (Approximate Scale) - o I Staff Gage Transect. 200 I Appendix Figure 0..25.location of Sauna Side Channel study site (RM79.8). 10:1 10:1._ .10.CROSS SECTION I 10.jCROSSSECTION 2 10J STATION 0.00 ...10J '.STATIONItBI---102 III \02..101 ~10\..-100 ....100 gg Z g8 Z gil 0 0 i=911 ~g7 4;i? gil >811 >8.:1 ILl UILl.J .J II.ILl 84 ILl iJ 83 \Ill ILl 92ILl>>III j:;81 l'\,I :12 ihi=110 4;80 J oct IlSI .J.J LII 119 \LI lUI II:1111 0 Q:i 117 117 IlII IlIIt,:I 11:1 ()'a 20 .0 110 110 100 120 \<40 1110 1110 2QD a 20 40 110 110 100 \20 140 leo 1110 200 P DISTANCE FROM LEFT BANK HEADPIN (lull OISTANCEFROM LEFT BANK HEADPIN (feet) 10:1 101l~104 CROSS SECTION ;J \04 CROSS SECTION 4 103 STATION :h77 103 STATION 4+81-..IQ2 ;102 "-to'..101....tOO -....100 Z III Z 119!2 .1111 0 llll I-'117 ~97~1I11 116 ILl \1:1 >9:1oJILl W 9...J 94 llJ \LI IIJ ILl 112 W 92> j:;lit 112 oro >91 oct 110 ;:90 ..J 1I1i 4;1I9 W ..J II:n \LI IIlI 1I7 a:67 1111 IIIl 1I:1 11:1 a 20 <40 110 110 100 120 \<40 ISO Ilia 200 a 20 <4D 110 1I0 100 120 140 1110 Ilia 200 DISTANCE FROM LEFT BANK HEADPIN (,..,)DIST ANtE FROM LEFT BANK HEADPIN Out I Appendix Figure D-26.Cross section of transects 1,2,3,and 4 at Sauna Side Channel (adapted from Quane et.al.1985). I -I J },J - ,- oRAFT/PAGE 3,4/30/85 4/14/85,4/19/85 ANOY/Oo~9,5/2/85 the date that the calibration data were collected at the Sauna Side Channel study site was determined to be 52,000 cfs,based on provisional USGS streamflow data. Calibration Calibration data available at the close of the 1984 field season consisted of that for a side channel flow of 52 cfs.Based on thi s cal ibration flow,an IFG-2 model was used to forecast instream hydrau- lies of this study site.The streambed profile,stage of zero flow,and observed and predicted water surface elevations for the study reach are plotted to scale in Appendix Figure 0-27.This data set was used to predict hydraulic information from side channel flows of 5 to 93 cfs (mainstemdischarges of 21,000 to 75,000 cfs).To evaluate the perfor- mance of the hydraul ic model fiel d observed and model predicted water surface elevations were compared (Appendi~Table 0-8).Additionally, the stage information of the model was compared to the rating curves establ ished by Quane et al.(1985)(Appendix Figure 0-24). It was difficult to hydraulically calibrate this site as only very limited field data were available.A site flow WSEL rating curve could only be developed for transect 2 (Appendix Figure 0-24).The IFG-2 model is essentially a water surface profile model and a critical variable for calibrating it,is the water surface elevations of simulat- ed flows.Data,however,is only available for transect 2 and not for any of the other three transects.The actual velocity measurements from other measured field flows at the discharge transect,however,can be 0-(03 ---'-........,--;--_.~---_......_-~-...:...-_--------'--~~--=------------ i2 .1I,lIIllIe,.. '0"'00 ,.....uc,J'ItANlle,I '+00 "."tlCl I ,-._._.-,,_._.~~~~_•••_-._---_._.a_a _.._.....-._.-._.._._••II :J:(••,.,.,.U_'Ill•• •....-._.~._._.._.-.-._._.-.-.-._.-.-.-._._._._.-.-._.-.-.-I.efl .,lI,th ••II."'...., ,"'0 'nl .., Thol.t,Su,.t1 Dolt.841009 Thol.tQ G,oditnl.·IO.4 ft'"milt Ob ..,••d Walt.'Slirfact Elnotian SllIIuloltd Wollr Surfelct EI."olion Eltt,opololtd Wo'"$uFfoet Elt.olliln "Elnollon II'Zt'll FI~. Thlll.tO Profilt SAUNA SIDE CHANNEL Thalweg Profile with Ob"rved and Predicted Water Surface ProW ... • ~ ~~------~l~f~ .. .. -;... :!... ;,;OQ Cl i= ~II... ..J '"'">.. ~ ..J '""'0: or 10 0;00 CJ, b ;...}:: STREAMBED STATION 1100" Appendix Figure 0-27.Comparison of observed and predicted water surface profiles from calibrated model and surveyed thalweg profile at Sauna Side Channel (adapted from Quane et.al.1985)• .t 1 J j } DRAFT/PAGE 10 4/19/85,5/2/85 ,~ANDY/Tables Appendix Table D-8 •Comparison off;e1d measured and model predicted water surface elevations at the calibration flow of 52 cfs for Sauna Side Channel. - Water Surface Elevation (ft) Ori gina 1 Model Transect Field Modified Field*Predicted 1 90.70 90.60 90.61 2 90.71 90.61 90.62 3 90.72 90.62 90.63 4 90.69 90.59 .90.63 *Field water surface elevations were reduced by 0.1 feet. - - 0-i.D'5 DRAFT/PAGE 4,4/30/85 4/14/85,4/19/85 ANDY/Doc 9,5/2/85 used to compare to the model predicted velocities for those same flows. At the discharge measurement for transect 2,however,there were only two flows that were far enough away from the 52 cfs measurement to be able to do this (38 and 68 cfs).Thus,the information available to hydraulically calibrate the IFG-2 model for this site consists of the water surface elevations and velocity measurements for all four transects at the calibrating flow of 52 cfs and water surface elevations ~ and velocities for the two other site flows of 38 and 68 cfs at transect 2. Overall,the site is hydraulically quite homogenous being influenced to a great deal by backwater (i.e.,all predicted velocities were less than 1.0 ft/sec).The effects of the backwater seem more pronounced at the 68 cfs flow.From the field data,the observed top width is greater by 20 feet,the water surface elevation is 0.93 feet higher and the average velocity is 0.20 ft/s~c slower than predicted by the model (Appendix Table 0-9).At the 38 cfs flow the effect seems to have reversed,with ~ the observed widths being similar,the WSEL 0.08 feet lower,and the average velocity 0.09 fUsec faster than predicted by the model (Appen- dix Table 0-9). In the calibration process,the original field WSEL was reduced by 0.1 feet.This adjustment was made in order to obtain water surface ele- > vations that agreed more closely at the lower site flows.It was felt that this adjustment would make the model,in terms of predictability, more sensitive at the lower site flows.By reducing the WSEL of transect 1 by 0.1 feet,the difference between the WSEL of the field - }1 l -}~1 )'})j DRAFT/PAGE 11 4/19/85 ANDYiTables J']J i ,j J A Calibration flow B Original field WSEL input into model The effects of the backwater at Sauna 51 de Channel ,information obtained from transect 2. Top Width eft) Field Model 0.52 0.49 0.42 0.32 0.53 0.51 Average Velocity (ft/sec) Field Model 55.0 53.0 52.0 77 .0 53.5 50.5 Original Modified Site WSEL eft)WSEL{ft) Flow (cfs)Field Model Field Model 68 91.85 91.06 91.85 90.92 52 A 90.71 B 90.74 90.61 C 90.62 38 90.24 90.42 90.24 90.32 Appendix Table 0-9. C Field WSEL reduced by 0.1 ft DRAFT/PAGE 5,4/30/85 4/14/85,4/19/85 ANDY/Doc 9,5/2/85 and the model at the 38 cfs site flow was reduced from 0.18 feet,when the calibration discharge WSEL was 90.71,to 0.08 feet,when the cali- bration discharge WSEL was 90.61 feet (Appendix Table D-9). As a result of a flood on August 26,sediments were deposited in the study site resulting in changes in all the cross sections derived from the calibration flow on July 23.Asa result,the cross sections obtained during the September 15 survey were used in the model until the water's edge of the calibration flow was reached when then the cross section from the calibration flow waS used. When measuring the velocities and depths at each of the transects,the discharge calculated at transect 4 was 16%lower than the 52 cfs site flow calculated at the discharge transect.In order to utilize this information in the model,the velocities were adjusted upwards by 16%. There was not a stage-site flow rating curve developed for transect l. When inputting other flows into the model,the IFG-2 requires either the associated WSEL for this flow or the slope.Because the WSEL could not be obtained for these other flows at this transect,a slope value of 0.00005 was input instead.This value was generated by the model from transectl at the calibration flow of 52 cfs. Verification The dominant influence of backwater on channel hydraulics makes the site a poor candidate for application of IFG-2 modeling techniques.However, '0-~8 - - - - ..... - DRAFT/PAGE 6,4/30/85 4/14/85,4/19/85 ANDY/Doc 9,5/2/85 because only one data set was collected,appl ication of the IFG-4 hydraulic model was not an option. Based on the first level of verification by EWT&A,the IFG-2 model for this site does an excellent job of simulating channel hydraulics between 48,000 cfs and 58,000 cfs mainstem discharge (34 to 52 cfs site flow) (Appendix Figure 0-28).Within this range,predicted WSEL IS,depths, a.nd velocities are in close agreement with field information (evaluated at 38 cfs by discharge measurement made by Quane et al (1985).The predictive capabil i.ty of the model within this range provi des evidence that the backwater influence within the study site is lessening with decreasing discharge. Below48,000cfsmainstem"there is increasing disagreement between the WSEL IS predicted by the model and those extrapolated from the rating curve.At 23 cfs site flow,the difference in predicted WSEL between model and rati ng curve equation has increased to approximately one foot at transects 1 and 2.Although there is evidence that suggests that the model maybe a more accurate predictor of WSEL I s than the rating curve equations below 48,000 cfs mainstem,insufficient information exists to resolve the difference with confidence.Since depths become shallow within this range,predictive errors in WSEL can result in significant errors in predicted depths and velocities.For this reason,the recom- mended extrapolation range is limited below 48,000 cfs. Above 48,000 cfs mainstem,there is increasing,disagreement between the WSEL's predicted by the model and those observed in the field.One of " Application Range of the Calibrated Hydraulic Model at Sauna -Side ChQnne I RM(79.8) Si t e S pecif,ic Flow t c fs 5 12 22 37 ~6 80 93 I I I I "~llllllllllllltllllllHl~I I d, .,J 0 I I I I I 0 10 20 30 40 50 60 70 75 Mainstem Disch or ge at ,Sunshine station t cfs x 1000 mIl Excellent _Acceptable ,II GoodoUnacceptable - Appendix Figure 0-28.Application range of the calibrated hydraulic model at Sauna Side Channel. ...,,],J J J'I I -- - - - -.. DRAFT/PAGE 7,4/30/85 4/14/85,4/19/85 ANDY/Doc 9,5/2/85 the premi ses of the hydraul ic theory that is the bas is of the I FG-2 model is that the water surface profile of the study reach is controlled by its slope.This premise is violated when the water surface profile is influenced by mainstem backwater.From examination of discharge measurements made at 48 and 68 cfs it is apparent that the influence of backwater is increasing with stage above 58,000 cfs mainstem. Overall,the recommended extrapolation range is 1 imited above 58,000 cfs.The model simulations were rated excellent between 48,000 and 58,000 mainstem discharge (34 to 52 cfs site flow).Good between 46,000 and 48,000 (31 to 34 cfs)and from 58,000 to 60,000 cfs (52 to 58 cfs). Acceptable between 44,000 and 46,000 cfs (28 to 31 cfs)and 60,000 to 63,000 cfs (58 to 62 cfs).The model was rated unacceptable below 44,000 cfs and above 63,000 cfs mainstem discharge (Appendix Figure 0-28). The second level of the verification procedure has not been performed as of this time. Application For habitat simul ati on modell i ng purposes the hydraul i c s imul ation model developed for Sauna Side Channel can simulate channel flows in the mainstem discharge range of 44,000 to 63,000 cfs. [)-i-J DRAFT/PAGE 1,4/30/85 4/11/85,4/21/85 ANDY/Doc 5,5/2/85 Sunset Side Channel (RM 86.9) Site Description Sunset Side Channel is located on the east bank of the Susitna River at river mile 86.9 (Appendix Figure 0-29).It is approximately 1.1 miles in length and is separated from the main channel Susitna River on the west by a network of vegetated islands and side channels.The channel is confi ned on the east by a hi gh cut bank.Pri or to breachi ng,the side channel is composed of a sequence of pools and riffles.During this period,flow is maintained in the main channel by groundwater seepage and upwelling.Subsequent to breaching,flows up to 3,900 cfs have been measured (Quane et al 1985). Breaching of Sunset Side Channel results from the direct overtopping of the head of the side channel by the mainstem SlJsitna River.Based on assessments by Quane et ale 1985 the side channel has been estimated to be initially breached at 31,000 cfs and controlled at a mainstem discharge of 32,000 cfs.The associated site flow has been estimated to be 45.8 cfs (Appendix Figure 0-30).This compares to an estimated flow of 41.1 cfs derived from the flow versus mainstem discharge rating curve presented in Appendix Figure 0-30 (Quane et ale 1985). Based on assessments by Quane et ale (l985)a backwater area does not occur in this side channel during unbreached conditions.But at breach- ing mainstem discharges ranging from 56,000-66,700 cfs,an area of })r I }l 1 1 J 1 } D.Staff Gage Tranaect ffi River Mile o 1000 I I FEET (Approximate Scale) .,<t .~.. $- .'':'),;/:..",~"SIlS\'f~P.1\\'1EI\.. I • EBRM 87 ·f '''5~,~'f$·i '1~'~.f..~J,.~;'1 l).,.J,~"'Jk,;.,:,I,.,'"I"\t:j······'.;,:1 '.,;,"I'/',.;r:.i.~··/;.J.;;ril.:.'.[:~I ...:t!~);.V.t ...:\!•.'i .\",'.~\."'/:"';'*'Y~I$!""."4",~".,.".1",,"8(1"'"•~,%I.·t·'"'..,.•-".'.~.:~;:::~1t.W;)."f'~~;:gf::~;;;.[.~'I;W:>l!I'~~·:"H(/I.!~.i'l;'I'(.·.'t.,...~l'~!;·.• ,.t'II"'.['(f,"I:~;,ti,;_~.~('J.•••(y.::,::·,·(..",,if.;'",:!.''''t.:~·~:...t ,,IJ ,:r!.',. j.jtf ...,."Jr \-"f~"1,.~.\.,•.,I,,"...".,··..F<\·.,;,'f·.':'1·"·...••,.•,,'f.'Off:""~(':..,,:-...\1 1 ;1'1-;;i'r·,·'···Io'tr~::,),··IJJ.*:'I'r:·...·;l,;i 1 '1:.,I'-,',........,.l.'.("......,"...•••...;'~.'.t'.,,:;p.,.r':;:.}'I'i:.!.j"'I:;j,/.IIl'"i I~:..~.~~>;~~~/~I;/I tt.!~;:::.~::~:'N;~;:1 :~J...".,...·f ,tf.,cti'/~.•.',".:.I';'f"'J :';;.,r:.'r·t . •',y,,,l,,\./'{···,I'•.••'f:'~;:"""..t,.p,.':'i~i~t:i~)..l ~j./'f"• tJ, IE Appendix Figure D-29.Overview of Sunset Side Channel (RM 86.4). - - - / /;: .j' t I 10 l::ien I.:. p:~- I~ IlXI.r·9C rU;11 o •10·0.5JJ5 IIlSEl •,o1 '.JO" Ie .l •0." 0:45.8 cfs SUNSET SIDE CHANNEL TRI GAGE 86.9Sf ! _-, lOOT coonOUEO 17.'00 S 0 S.lZ.OOC c'. lIS[l •100 •11'1 QO.OH5 •'l(' G oc •10.17 •75,......]1)0.1 ,l.G." 1'"+-__CflllnOl.tEO J2.00(I Se So Illt.OOO <'I WIEl •10.1•7766 ~o.951J •9C .Z •0.99 Q:41.1 cfs -.. :. - Appendix Figure 0-30.Comparison of rating curves from Sunset Side Channel at transect 1 (from Quane et.al.1985).- - -- - - -. DRAFT/PAGE 2,4/30/85 4/11/85,4/21/85 ANDY/Doc 5,5/2/85 backwater was observed to extend upstream approximately 1,100 feet to a point between transects 1 and 2. The IFG modelling site within Sunset Side Channel during the 1984 open water field season was located in the lower portion of the side channel and was 1410 feet in length (Appendix Figures 0-29 &31).Seven tran- sects from which hydraulic information was collected were located within this study site (Appendix Figures 0-32 &33).The channel within the study site has a gradual bend.The right bank from transects 2 to 6 is erosional in nature becoming less steep and depositional in nature at transects 0 and 1.On the left bank from transects 2 to 6 is primarily depositional in nature becoming steep and erosional in the areas of transects 0 and 1.At the transect 2 on the left bank a small side channel area enters through which water was never observed running (Appendix Figure 0-31).The thalweg gradient within the study site is 9.5 ft/mile (Quane et al.1985).Riparian vegetation along the right bank is primary birch and spruce whereas on the left bank it is alder. Transect 0 is located in shallow pool type habitat and has substrates of sand and small gravels.At transects 1 (the discharge site)and 2,the primary habitat type is run,and the substrate is small gravel.At transect 3,the habitat changes to run-shallow pool habitat,with the predominate substrates being small and large gravels.The hydraulic control for transects 5 and 6 is transect 4.This transect represents riffle habitat,with substrates composed mostly of small and large gravels.Transects 5 and 6 are located in deep pool habitat,with substrates being composed of mostly small and large gravels. D-t5 -I Wz.z«z o wo-en wen-a:z ~en .. ~Staff Gage Transect oI , FEET (Approximate Scale) _.-:il,~ D-tC, - -O"l \0 00 :E:a::- (1J ~-VI ~ ::::l ~ VI """lIl-(1Jc:c:co..c:u (1J "0..... V') ~-(1J VIc: ::::l V')-.... 0 c: 0.... ~cou 0 -J -..... (Y) I 0 (1Js... ::::l O'l..... LL. X..... "0 C (1J 0. 0. e:t: l 1 -)-I }.~-)J )}))J }}) 110110CROSSSECTION01091 CROSS SECTION I'09 108 STATION 2.23STATION0'00 ~108 -107-;107 ..106:!106 ~-10:'105 104Z104Z~103 2 103....10:l ....102<I: <I:101>101 >IU 100 W 100.J 99 .J IIl1w W98 98IU97IU97>>118;:96 9!!....9!!<t 127 ell «94.J 94 .JW93IU 9JCJIt:92 a:1129'111,90 110 0 '00 200 300 400 0 100 200 .lOa 400--tJrr DISTANCE FROM LEFT BANK HEADPIN (feell DISTANCE FROM LEFT BANK HEADPIN (feef) 110 CROSS SECTION 2\09 STATION 4 +75 110'08 lOll j CROSS SECTION 3;107 108 STATION 1+58..108 --111'1-105 .. '04 ::106Z 0 103 -10:! i=102 Z 104 -j t<I:101 0 '03>....'02lIJ'00 <I:.J 99 >'0'W 98 IU ".~IU .J IIl197 IU>96 496 .ro 98i=IU 11795l2.t eft,><I: ~::~~I 498 ell.J 94 """":::::W 93 <t V 127 .faIt:92 .J 94 W 93II'II:90 92 a 100 200 JOO 400 91 lIO DISTANCE FROM LEFT BANK HEADPIN (Ieet)a 100 200 JOO 400 DISTANCE FROM LEFT BANK HEADPIN (fnt) Appendix Figure 0-32.Cross section of transects 0,1,2,and 3 at Sunset Side Channel (adapted from Quane et.al.1985). 110 ]. 1011 CROSS SE~TION fl IDe STATION II:·53..107•108.....lOll.... Z 104 2 103 I-102 ~101 ""100 .J 1111 UJ lie ""117 >118 i=1111«114.J UJ 113 0::112 111 110 400 0 100 aDo JOO 400 DISTANCE FROM LEFT BANK HEADPIN ('ull :SOD200100 DISTANCE FROM LEFT BANK HEADPIN (,.., I o 110 j109 CROSS SECTION 4 I 108 STATION 9 ..10 107 106 10:1 104 103 102 101 100 99 98 j97 \ 96 ~-~--4911 olt 95 ..........---.....1---121 .It 114 93 92 91 90 Iii iii i I I ••...- %o i=«> UJ .J UJ UJ> I-« ...I..... It:CJ, --J-l Cb 400300aDo100 '")----496 olt,In olt" CROSS SECTION 6 STATION 14 ..10 110 1011-IDe...107•..lOll.......10:1 Z 104 0 103 i=102 ~101 ""100 .J 1111 ""lie ""117>lie i=1I11«114.J ""113 It:112 111 110 0 DISTANCE FROM LEFT BANK HEADPIN (,..t) Cross section of transects 4,5,and 6 at Sunset Side Channel (adapted from Quane et.al.1985). J Appendix Figure 0-33. ,)J J J J - - - - - - - DRAFT/PAGE 3~4/30/85 ,4/11/85,4/21/85 ANDY/Doc 5,5/2/85 Data Collected Hydraulic data were collected at two calibration discharges:127 and 496 cfs (Appendix Table 0-4).Mean daily discharges for the Susitna River on the dates that cal ibration data were collected at the Sunset Site Channel study site were 42~500 and 57,800 cfs,respectively as determined from provisional USGS streamflow data. Calibration Calibration data were available at the close of the 1984 field season for side channel flows of 127 and 496 cfs.Based on these two cali- bration flows,an IFG-4 model was used to forecast instream hydraulics at this study site.The streambed profile,stage of zero flow,and observed and predicted water surface elevations for the study reach are plotted to scale in Appendix Figure 0-34.Both calibration data sets were used to predict hydraulic information from side channel flows of 7 to 1,603 cfs (mainstem discharges of 21,000 to 75,000 cfs). To evaluate the performance of the hydraulic model,observed and predicted water surface elevations,discharges,and velocity adjustment factors were compared (Appendix Table 0-10).The hydraul ic model at Sunset Side Channel is similar to Circular Side Channel.Because of the 2 calibration flows,only a 2 point rating curve was formulated.In evaluating the performance of the model,observed and predicted WSEL's and discharges were the same because of this rating curve.Velocity D-t1 ~ ~ U~U.II;Cl •",..Nucr ,..."uc,..,a""lIte,I,.ANJ[et I1'•.aNSEt;,I Thah"Q $urny Doh'840929 Tholw'Q Grodi.n"9.5 fllllmil. Obll,..d Wah,Surfac.Elovollol'l $imulo,.d Wo,er $u''OCt 1::1 ..01101'1 E.'ropolahd Wo,.,Surfac.EIlYolion EI.vo'ion of Zero Flow Tho'W'Q Profil. TIIl""sECv 0 :-._._._._._.:._._._._._._._._._._._._._.:.:_._._._.-.-.--:-;-------_._-_.__.., ___---------. : ::'",OJ___.~--.------.--.-.-._.•K".,,0_'~~l~---·--'::=-".-"'~:."u_~ SUNSET SIDE CHANNEL Thalweg Profile with Observed and Predic'ed Waler Surface Profiles • ~ to '" II ee .: zo i=...>... -'...... >;:... -'... If. o I COo "'t0010.00 I_''''0 ,....I I ••Iii iii iii iii iii ,Ii.iii i j i ,iii Iii iii iii ii'iii iii iii i • • • , ," , 0""00 '.00 1()t00 'Si'OQ STREAMBED STATlOII It..tl Appendix Figure 0-34.Comparison of observed and predicted water surface profiles from calibrated model and surveyed thalweg profile at Sunset Side Channel (adapted from Quane et.al.1985). 0]J ..J I .1 ]'~.'".J J ..J J ..~.J oJ -DRAFT/PAGE 3 4/23/85 AHMR/Appendix Table 3 Appendix Table 0-10.Comparison between observed and predicted water surface elevations,discharges,and velocities for 1984 Sunset Side Channel hydraulic model. ,~-, Streambed Water Surface Station Elevation Discharge Velocity Observed Predicted Observed Predicted Adjustment-(ft)(ft) (ft)(cfs)(cfs)Factor (:IIIIt'I\0+00 94.27 94.27 132.7 132.4 1.000 2+23 94.34 94.34 131.7 131.3 .999 4+75 94.69 94.69 133.6 133.3 1.000 7+58 94.97 94.97 127.2 126.9 .998 9+10 95.54 95.54 136.4 136.3 1.000 11+53 95.98 95.98 125.5 125.2 .999 14+10 95.97 95.97 129.9 129.6 Qo =131.0 Qp =131.0 0+00 95.62 95.62 462.3 462.3 1.000 2+23 95.67 95.67 500.0 500.0 .999 4+75 95.75 95.75 504.6 504.6 1.000 7+58 95.87 95.87 438.1 438.1 1.000 I~'9+10 96.18 96.18 507.2 507.2 .993 11+53 96.64 96.64 469.9 469.9 .999 14+10 96.63 96.63 492.0 492.0 1.000 Qo =482.0 Qp =482.0 Qo is the mean observed calibration discharge. ~ Qp is the mean predicted calibration discharge. ;P'- D-8J ------,......-,---------------,..".""-- - DRAFT/PAGE 4,4/30/85 4/11/85,4/21/85 ANDY/Doc 5,5/2/85 -adjustment factors were all within the good range of 0.9 to 1.1. Additionally,the stage information of the model was compared to the rating curves established by Quane et al.(1985)(Appendix Figure 0-30). In the model,the stages of zero flow are not the same as those deter- mined from the thalweg survey by Quane et al.1985 (Appendix Table 0-11).The stage of zero flow values,input into the model,were derived from the thalweg points of the model input cross sections of transects 0,1,2,and -4.The reason for this change in thalweg elevations is likely the result of the flood event.All the points used in the model were from measurements made before the flood,whereas the Quane et ale (1985)thalweg survey was done after the flood event. At transect 6,the velocities at the high calibration flow measurement (496 cfs)were adjusted upwards by 15%and at the low calibration flow measurement (127 cfs)adjusted downwards by 21%.Because this transect bisects a deep pool with eddies,it is difficult to obtain an accurate discharge measurement.The eddy effect was much more pronounced at the high cal ibration flow measurement,as there was a section of about 40 feet in which the velocities were negative.Because of its depth and slow velocities this area was considered as valuable habitat for rearing juvenile salmon.In order to facilitate us-ing these negative velocity .... values in the model these measurements were treated as positive.... At transect 3 there was a di fference in WSEL r s at the 127 cfs cali- bration flow.WSEL at the left bank was 95.03 feet whereas at the right bank it was 94.90 feet.As the staff gage WSEL was 94.93 feet and the -. 0-8d-. ,~ Appendix Table D-11. DRAFT/PAGE 14 4/19/85 ANDY/Tables Differences between stages of zero flow input into the model and Quane et al.(l985)thalweg survey at Sunset Side Channel. ,~ Stage of Zero Flow (ft) Transect Model Input Thalweg Survey- 0 92.30 92.50 -. 1 92.60 93.00 2 93.40 93.60 ~ 3 93.40 93.60 4 94.20 94.40 5 94.20 94.40 6 94.20 94.40 J~ ~- D-83 - DRAFT/PAGE 5,4/30/85 4/11/85,4/21/85 ANDY/Doc 5,5/2/85 majority of flow occurred along this right side a WSEL of 94.93 feet was used in the model. At transect 4 there was a large discrepancy (0.54 ft)in WSEL's across the transect at the cal 'i brat;on flow of 127 cfs.Thi s was because the section of the channel where a majority of the flow occurred was higher in elevation and separated by a gravel berm from a lower elevation minor channel where the staff gage was located.In order to utilize this cross section in the model,the channel cross section of the minor channel was elevated upwards by 0.6 feet. At a section of transect 3 the individual velocity measurements for the ~ 127 cfs site flow were greater than the corresponding velocity measure- ments at the higher 496 cfs site flow.If these original values were to be used in the model,the simulated velocities would decrease with increasing site flows.This realistically does not occur.In order to amend this situation,the velocities were adjusted such that the rela- tionship would simulate a positive increase in velocities with corre- sponding increases in site flow. Verification Based on the first level of verification by EWT&A,the model does an excellent job of simulating channel hydraulics between 50,000 and 61,000 cfs,mainstem discharge(275 and 649 cfs site flow)(Appendix Figure 35). Above 61,000 cfs,the simulated depth and velocity distributions begin to deteriorate in qual ity.The model simulations were rated good _ o-8J../ - - ..... - )~ DRAFT/PAGE 6,4/30/85 4/11/85,4/21/85 ANDY/Doc 5,5/2/85 between 61,000 and 64,500 cfs (649 and 850 cfs site flow),acceptable between 64,500 and 67,000 cfs (850 and 1,000 cfs site flow),and unac- ceptable above 67,000 cfs mainstem discharge (Appendix Figure 0-35). Below 50,000 cfs,the model simulations were also rated less than excell ent,primari ly because of reduced effectiveness in predi cti ng water surface profiles as compared to field observations.The model simul ations were rated good between 38,000 and 50,000 cfs (89 and 275 cfs site flow),acceptable between 32,000 and 38,000 cfs (41 and 89 cfs site flow),and unacceptable below 32,000 cfs mafnstem discharge (Appendix Figure 0-35). Overall,the model simulations were rated excellent between 50,000 and 61,000 cfs (275 and 649 cfs)and good from 38,000 to 50,000 cfs (89 to 275 cfs)and from 61,000 to 64,500 cfs (649 to 850 cfs).They were acceptab 1e between 32,000 and 38,000 cfs (41 and 89 cfs)and between 64,500 and 67,000 cfs (850 and 1,000 cfs),and became unacceptable at mainstem discharges below 32,000 cfs and above 67,000 cfs • The second level of verification has not been performed as of this time. Application For habitat simul ation Illodell"i ng purposes the hydraul ic simul ation model developed for Sunset Side Channel can simulate channel flows in the mainstem discharge range of 32,000 to 67,000 cfs. D-85 ---,-----~.--g-------------------------------- Application Range of the Calibrated Hydraulic Model at Sunset-Side Channel RM (86.9) Site Specific Flow,cfs o 7 31 101 260 614 1191 1603 CJ I co b I I I r I I ~~ ~,,,~I ~ I I I I I I o 10 20 30 40 50 60 70 75 Mainstem Discharge at Sunshine Station,cfs x 1000 '.Excellent •Acceptable II Good D Unacceptable Appendix Figure 0-35.Application range of calibrated hydraulic model at Sunset Side Channel. .,)-']J -J ~~.J J }J - - DRAFT/PAGE 1,5/2/85 4/10/85,4/21/85 ANDY/Doc 4,5/1/85 Trapper Creek Side Channel (RM 91.6) Site Description Trapper Creek Side Channel is located on the west bank of the Susitna River and is approximately 5.0 miles in length (Appendix Figure 0-36). It has a relatively uniform,broad,and flat bottomed alluvial channel which is fed by multiple heads.It is separated from the mainstem Susitna River by a complex of sand bars,small channels,and vegetated islands.The head portion of this side channel is located in a complex of small channels and vegetated islands making it difficult to identify the origin of breaching flows (Quane et a1.1985). During unbreached conditions flows in Trapper Creek Side Channel are principally due to Cache Creek and groundwater occurring in the upper reaches of the side channel.Breaching of Trapper Creek Side Channel is the result of the direct overtopping of the multiple heads of the side channel by the mainstem Susitna River.Based on assessments by Quane et a1.(1985),the channel is estimated to be initially breached at a mainstem discharge of 43,000 cfs.Based on the comparison of the stage versus mainstem discharge r~ting curve for transect 4 (Appendix Figure 0-37)by Quane et a1.1985,a discharge of 44,000 cfs was selected as the controlling breaching discharge.This mainstem discharge - corresponds to a streamflow measurement of 31.4 cfs. Based on assessments of backwater by Quane et a1.(1985),an area of backwater has not been observed during other breaching and nonbreaching D-81 Appendix Figure 0-36.Overview of Trapper Creek Side Channel (RM 91.6). D-88 - - - ! Q:31.4 cfs ~Ole.10·z•GSZl ellS[\,.tol I.ZlU ~.z .O.tt ~~ e 3 ~ ! 0"•10·'"t5U 0..J.USI ..z •0." ,----0"•10.41 •SIOI 0..t.l'n r Z •G.tt •! ~ ~ ~ ~I e r~'"i ~ - ~ ,,~ ..'~t-----_..L._-.-o:....:o:-----"""-""""""""""::!_.,-J------.....--J-------.--------:'.,. /'IfIt>Ole<:lX1"'1l1l;t ~:-.tCllI>C f1aoe::;r-,l "30.I-'lO rCfl ...llOO S 0 S 93.Joo efo I/SEl •10···,ess gl.1S6S •90 r2'•0.95 .... - - i:i... .t 'UI g, ~I.,. ~I °1 I w~~"Or to'fTllOI.l(~ ts.ooa S.4!S.",000 e', WUl •10°·01101 0°·0156 •90 r l •0.8l TRAPPER CREEK SIC TR4 GAGE 91.651 .,- Appendix Figure 0-37.Comparison of rating curves from Trapper Creek Side Channel transect 4 (from Quane ~t.al.1985). D-8~ DRAFT/PAGE 2,5/2/85 4/10/85,4/21/85 ANDY/Ooc 4,5/1/85 mainstem discharges.But at mainstem discharges ranging from 15,700 to 22,700 cfs,pooling was observed at transects 1,2,and 3 which resulted from the control located about 370 feet downstream from transect 1. The IFG modelling site selected for Trapper Creek Side Channel during the 1984 open water field season was 790 feet in length and was located in the lower portion of the side channel in a broad open channel area (Appendix Figures D-36 and 0-38).Four cross sections were surveyed within this area to define channel geometry (Appendix Figure 0-39).The upper two transects were si tuated ina run,whereas the lower two transects were in a backwater pool influenced by a downstream control. Substrate within the study consisted primarily of cobbles and gravel s with some sand at the first transect.The thalweg gradient of the side channel is 12.1 ft/mile (Quane et al.1985). Data Coll ected Hydraulic data were collected at three calibration discharges:16,32, and 389 cfs (Appendix Table 0-4).Mean daily discharges for the Susitna River on the dates that calibration data were collected at the Trapper Creek study site were 20,900;44,000;and 57,700 cfs respectively as determined from provisional USGS streamflow data. Calibration Calibration data were available at the close of the 1984 field season for side channel flows of 16,32,and 389 cfs.Based on these - - - I .]1 )'J ))j I I, i I )1 J ))}., d, -D ~, Appendix Figure 0-38.Location of Trapper Creek Side Channel (I~M 91.6). ~ I study sHe ,__;;tW"'<""'~ 105\05 CROSS SECTION I 10...j CROSS SECTION 210~STATION 0+00 103 STATION 2 +89-103 --102 Qi 102GI..101 II 101---100 ....100 Z 99 Z 911 0 98 0 98 I-97 ~97<l 116>96 >W 95 W 115..J ..J "'"W 94 LLI 113 ~....."J ~B9 ch937W LLI 112 32 c"16 eta>92 .....::::>91 .>111I-90 I-900<1:oct..J 89 ..J 811 W Be w 8S0:e7 a:8708S8S I 85 85 a 100 200 300 400 0 100 200 300 ...00-0 'P DISTANCE FROM LEFT BANK HEADPIN (feet I DISTANCE FROM LEFT BANK HEADPIN (feet) 105 ~.-.__._----,- 105104CROSSSECTION3 10"']CROSS SECTION 4103STATION5+76 103 STATION 1+90-....102 -102-IIII101III 101u...100 -........10099 Z 911Z 0 1111 0 liS ~117 i=117116~118>115LLI LLI 115 ..J 9 .....J 9 ...W 93 LLI 113 92 J "Ut Ji"16 ciaW112 W '"C::7'>91 >91~110 ~110811811..J ..JW88 I LLI S8a:87 I 0:87116868S850100200300...00 0 100 200 300 ...00 DISTANCE FROM LEFT BANK HEADPIN (feet)DISTANCE FROM LEFT BANK HEADPIN (feet) Appendix Figure 0-39.Cross section of transects 1,2,3 1 and 4 at Trapper ,I J "J J Cr,e,W k S'~.~3 Chan"j 1 (~~:ftedJ::im o·LA,~e et ,,'1 '":'~,~f).J .J J -J J J - .-. DRAFT/PAGE 3,5/2/85 4/10/85,4/21/85 ANDY/Doc 4,5/1/85 calibration flows an IFG-4 model was used to forecast instream hydraulics for this study site.The streambed profile,stages of zero flow,and observed and predicted water surface elevations for the study reach are plotted to scale in Appendix Figure 0-40.All three data sets were used to predict hydraulic information for side channel.flows from 9 to 1,351 cfs (mainstem discharges of 12,000 to 75,000 cfs). To evaluate the performance of the hydraulic model,observed and predicted water surface el evati ons,di scharges,and velocity adjustment factors were compared (Appendix Table D-12).Of the 12 sets of observed and predicted WSEL's,six sets were within ±O.02 feet of each other and the other six sets were within ±0.05 feet of each other.All the observed and predi cted di scharges were wi thin 10%of each other except for one set in which there was an 11%difference.All velocity adjustment factors were within the good range of 0.9 to 1.1. Additionally,the stage information of the model was compared to the rating curves established by Quane et al.(1985)(Appendix Figure D-37)• Between the time period when the first two calibration flows (389 and 32 cfs)were made and the last calibration flow of 16 cfs was made the channel cross section at transect 1 was scoured by a flood event.In order to utilize this information in the model the cross section deter- mi ned from the survey and the 16 cfs flow measurement were IJsed,the WSELls of the two calibration flows (389 and 32 cfs)were then reduced by 0.37 'feet. TRAPPER CREEK SIDE CHANNEL ThalweQ Profile with Observed and Predicted Water Surface ProW .. CJ I -0 -l: / ..----; i ~"...CANt . ~ Tllial ••g Surnr Daft.840"3 Thal••o 'readiIR"12..1 "."",it, Otl ..,nd WI'"$U,'OC.I £If,allofll Si"'tllot.d Wat.,S"r'O't fltwaUoll [Itropolalld Wa'"Surfaci f'notloft EIIYoUo,.0'ZltO 'low nol...,P,ow. ·"'1 Uti 9Uk AO'.,Ol It . ,......UC,I ,.....uet,.......c.J '......C'IIII·,..,...---I ·~~oo iii 'i .':"00 'ii''0':'00 iii ,.t.!.DQ • _._._._._._._._._._._._._._._.".._.no."] _.._._._._.-._._._._._.._.-._._..-._.•l"lIh 1."'H"u",....... :IE:..._.-.-..-._.-.-.•"'I'"_,....1..~.I~...1:.::::.:::=.:=.::.::.:::.-:.::-.:=..::.:.:,:,:::·-·=,:==-===:-·-·-·-·-·-·m",-·_·_·;-";;;;:;;:·:::::~-~':I'~ ~.~w ~ ~~.... ..N.... STREAMBEO STATION '''.U Appendix Figure 0-40.Comparison of observed and predicted water surface profiles from calibrated model and surveyed thalweg profile for Trapper Creek Side Channel (adapted from Quane et.al.1985). .J .J J ;.J ]i J I it j ].J J I """ - - Qo is the mean observed calibration discharge. Qp is the mean predicted calibration discharge. DRAFT/PAGE 4,5/2/85 4/10/85,4/21/85 ANDY/Doc 4,5/1/85 Transect 1 was determined to be a poor site for measuring discharge as it was a pool area affected by a downstream control.The velocities for ~ the 32 cfs calibration flow were therefore adjusted upwards by 27%and for the 16 cfs calibration flow by 20%. Verification Based on the first level of verification by EWT&A the model does a good job of simulating channel hydraulics between 20,000 cfs and 54,000 cfs mainstem discharge (15 and 220 cfs site flow)(Appendix Figure 0-41). There are sufficient deviations in water surface elevation and discharge between predicted and observed values within this range to preclude attainment of the excellent rating.This is because the model is approximating a portion of the rating curve described by two adjoining linear relationships with a single line. Between 54,000 cfs and 58,000 cfs mainstem (220 and 460 cfs site flow) the model does an excellent job of simulating channel hydraulics. Beyond 58,000 cfs mainstem,the quality of the simulations begins to deteriorate as the slope of the stage/discharge relationship for the site flattens with a change in channel geometry.The deviation between the regression line developed within the model and that of the rating curve increases with discharge until the model simulations are no longer acceptable.The model simulations were rated good between 58,000 cfs and 61,000 cfs (460 and 600 cfs site flow),acceptable between 61,000 cfs and 66,000 cfs (600 and 820 cfs site flow),and unacceptable above 66,000 cfs mainstem (Appendix Figure 0-41). - - - 1 1 )))J J !j )»!1 ,1 1 1 ) A pp Ii cat ion Ran 9 e 0 f t·h e Ca lib rat e d H y d r a u lie Mod eI . at Trapper Creek Side Channel .- RM (91.6) Site Specific Flow,cfs d, --t) -f-(o 10 15 20 22 30 29 40 107 50 564 60 1030 1351 70 75 Mainstem Discharge at Sunshine Station.cfs x 1000 •Excellent _Acceptable II Good D Unacceplabie AppendiX Figure 0·41.'Application range of the calibrated hydraulic model at Trapper Creek Side Channel. DRAFT/PAGE 5,5/2/85 4/10/85,4/21/85 ANDY/Doc 4,5/1/85 The second level of the verification has not been performed as of this time. Overall,the model simulations were rated excellent from 54,000 to 58,000 cfs (220 to 460 cfs)and good from 20,000 to 54,000 (15 to 220 cfs)and from 58,000 to 61,000 cfs (460 to 600 cfs).They were acceptable from 61,000 to 66,000 cfs (600 to 820 cfs),the simulations became unacceptable below 20,000 cfs and above 66,000 cfs. Application For habitat simulation modelling purposes the hydraulic simulation model ~ developed for Trapper Creek Side Channel can simulate channel flows in the mainstem discharge range of 20,000 to 66,000 cfs. SUMMARY Island Side Channel (RM 63.2) An IFG-2 hydraulic model was used to hydraulically simulate site flows of this study site based on one field measured flow of 338 cfs was.The calibrated IFG-2 model simulated site flows excellently in the mainstem ~ discharge range of 35,000 to 56,000 cfs and good in the range of 56,000 to 64,000 cfs.The acceptable range was from 64,000 to 70,000 cfs.For habitat simulation modelling purposes the Island Side Channel hydraulic model can simulate channel flows in the mainstem discharge range of 35,000 to 70,000 cfs. D-er8 ..... - - ..... """ DRAFT/PAGE 6~5/2/85 4/10/85~4/21/85 ANDY/Doc 4~5/1/85 Mainstem West Bank Side Channel (RM 74.4) An IFG-4 hydraulic model was used to hydraulically simulate site flows at this study site based on field measured flows of 6~310,and 450 cfs from which simulated flows were based.The IFG-4 model developed for this site simulated site flows excellently in the mainstem discharge range of 18~000 to 21,000 cfs and from 28~000 to 34~000 cfs.It predicted good in the range of 21~000 to 28,000 cfs and from 34~000 to 41~000 cfs.The acceptable range was from 41~000 to 48~000 cfs.For habitat simulation modelling purposes the Mainstem West Bank Side Channel hydraulic model can simulate channel flows in the mainstem discharge range of 18~000 to 48~000 cfs. Circular Side Channel (RM 75.3) An IFG-4 hydraulic model was used to hydraulically simulate site flows at this study site based on field measured flows of 50 and 204 cfs from which simulated flows were based.The IFG-4 model simulated site flows excellently in the mainstem discharge range of 39,000 to 57,000 cfs.It predicted good in the range of 36~000 to 39,000 cfs and from 57~000 to 60,000 cfs.The acceptable range was from 60,000 to 63,000 cfs.For habitat simulation modelling purposes the Circular Side Channel hydraulic model can simulate channel flows in the mainstem discharge range of 36,000 to 63,000 cfs . DRAFT/PAGE 7,5/2/85 4/10/85,4/21/85 ANDY/Doc 4,5/1/85 Sauna Side Channel (RM 79.8) An IFG-2 hydraulic model was used to hydraulically simulate site flows at this study site based on one field measured flow of 52 cfs from which simulated flows were based.The IFG-2 model simulated site flows excellently in the mainstem discharge range of 48,000 to 58,000 cfs and good in the range of 46,000 to 48,000 cfs and from 58,000 to 61,000 cfs. The acceptable range was from 44,000 to 46,000 cfs and from 61,000 to 63,000 cfs.For habitat simulation modelling purposes the Sauna Side Channel hydraul ic model can simulate channel flows in the mainstem discharge range of 44,000 to 63,000 cfs. Sunset Side Channel (RM 87.0) An IFG-4 hydraulic model was used to hydraulically simulate channel flows at this study site based on field measured flows of 127 and 496 cfs from which simulated site flows were based on.The IFG-4 model simulated site flows excellently in the mainstem discharge range of 50,000 to 61,000 cfs.It predicted good in the range of 38,000 to 50,000 cfs and from 61,000 to 64,500 cfs.The acceptable range was from 32,000 to 38,000 cfs and from 64,500 to 67,000 cfs.For habitat simul ati on modell i ng purposes the Sunset Si de Channel hydraul i c model can simulate channel flows in the mainstem discharge range of 32,000 to 67,000 cfs. D-\00 - - - - - DRAFT/PAGE 8,5/2/85 4/10/85,4/21/85 ANDY/Doc 4,5/1/85 Trapper Creek Side Channel (RM 91.6) An IFG-4 hydraulic model was used to hydraulically simulate channel flows at this study site based on field measured flows of 16,32,and 389 cfs from which simulated flows were based.The IFG-4 model simulated site flows excellently in the mainstem discharge range of 54,000 to 58,000 cfs.It predicted good in the range of 20,000 to 54,000 cfs and from 58,000 to 61,000 cfs.The acceptable range was from 61,000 to 66,000 cfs.For habitat simulation modelling purposes the Trapper Creek Side Channel hydraulic model can simulate channel flows in the mainstem discharge range of 20,000 to 66,000 cfs. 0-/0 \ DRAFT/PAGE 1 5/1/85 ANDY/Acknowledgements --ACKNOWLEDGEMENTS The authors express their appreciation to the following for their assistance in preparing this report: The other ADF&G Su Hydro Aquatic Studies Program staff who provided their support to this study.For collection of field data:Fred Metzler,Pat Morrow,Isaac Queral,Glenn Freeman, and John McConnaughey.To Paul Suchanek for collection of the cover information used in assessing the weighted usable areas of the models.In reduction of the data,making the many computer runs,and helping prepare this appendix:Fred Metzler,Mary Shiffer,Dan Kingsley,and Kathy Dugan.To Tim Quane,Pat Morrow,and Isaac Queral for use of much of their findings and figures from Task 36 support technical report - Hydrological Investigations at Selected Lower Susitna River Study Si tes.To the edi tors:Doug Vi ncent-Lang,Ti m Quane and Paul Suchanek.For the cartography on the final figures: Carol Hepler and Rox Ann Peterson. To E.W.Trihey and Associates particularly Bob Aaserude and Diane Hilliard for the valuable expertise in the collection of - data,the calibration,and the verification of the hydraulic models.It is felt that Bob Aaserude's involvement in the preparing of this appendix deserves an authorship. 1)-IOd- (~DRAFT/PAGE 1 5/1/85 ANDY/Literature Cited LITERATURE CITED Bovee,K.D.,and R.Milhous.1978.Hydraulic simulation in instream flow studies:theory and techni ques.Instream Flow Information Paper No.5.Instream Flow Service Group.USFWS.Ft.Collins, Colorado. 1982.A guide to stream habitat and analysis using instream flow incremental methodology.Instream Flow Information paper No. 12.Coop.Instream Flow Service Group.USFWS.Colorado. Buchanan,T.J.,and W.P.Somers.1969.Techniques of water resources investigations of the United States Geological Survey.Chapter A8. Discharge measurements at gaging stations.USGS.Washington DC. Estes,C.C.,D.S.Vincent-Lang,(eds)1984.Aquatic Habitat and Instream Flow Investigations.Alaska Department of Fish and Game. Su Hydro Aquatic Studies Report Series No.3.Alaska Department of Fish and Game.Anchorage,Alaska. Quane,T.,P.Morrow,and I.Queral.1985.Hydrological Investigations at Selected Lower Susitna River Study Sites.Alaska Department of Fish and Game.Su Hydro Aquatic Studies Task 36 Support Technical Report.Alaska Department of Fish and Game.Anchorage,Alaska. D -103 "--------"---, DRAFT/PAGE 2 5/1/85 ANDY/Literature Cited Schmidt,D.C.,S.S.Hale,and D.L.Crawford (eds.).1984.Resident and Juvenile Anadromous Fish Investigations (May -October 1983). Alaska Department of Fish and Game.Su Hydro Aquatic Studies Report Series No.2.Alaska Department of Fish and Game. Anchorage,Alaska. Trihey,E.W.1979.The IFG incremental methodology.In G.L.Smith, ed.Proceedings of the Instream Flow Criteria and Modeling Work- shop.Colorado Water Resources Research Institute,Colorado State University.Pages 24-44.Information Series No.40.Fort Collins,Colorado. 1980.Field data reduction and coding procedures for use with the IFG-2 and IFG-4 hydrau1 ic simulation models.'Instream Flow Service Group,USFWS.Fort Collins,Colorado. and D.L.Wegner.1981.Field data collection procedures for use with the physical habitat simulation system of the Instream Flow Group.Instream Flow Service Group.USFWS.Fort collins, Colorado. D-ID~ - - Appendix Attachment 1 Technical Memorandum Extrapolation Limits of the 1984 Middle River IFG Models ,...0-IO~ E..\VOODY TRII-lEY &ASSOCIATES .NSTREAM PLOW AND RIVERINEUARITAT ASSESSMENTS P.o.BOX 11177.. ANCHORAGE.ALASKA 09511 (007)562·7707 Technical Memorandum Extrapolation limits of the 1984 Middle River IFG Models by - --,~..N.Diane Hilliard E.Woody Trihey and Associates Apd 1 8.1985 0-IOlo - '"'" - - The 1984 middle river IFG hydraulic models have been calibrated -1nd their extrapolation ranges evaluated.The IFG-4 models were calibrated using both the IFG and E'tlT&A guidelines.The IFG-2 models were c-1librated using a variable Manning's n approach.Hith an increase in the depth of flow, there is a corresponding decrease in Manning's n values.The depth and velocity information collected at each site was classified as either ca 11 bra ti on or shore 1i ne da ta.The ca 1i bra tion da ta was collected across the entire cross section.Shore1ine data were collected from each bank .- out into the channel until either the depth or velocity was limiting to field personnel.Site-specific flow values.as determined by either the water surface elevation versus site flow or site flow versus mainstem discharge relationships are presented for mainstem discharges from 5,000 to. 35,000 cfs.Wi thin thi srange of ma i ns tem di scharges,several study si tes transform from clear water side sloughs to turbid water side channel s to mainstem channels.Baseline flows have been estimated for the sttes when , theY"are not controlled by the ma instem. The quality of each model was based on two levels of criteria.The level one criteria is a qualitative evaluation of four separate criteria.The models were given a numeric rating of compliance for each criteria whenever possible.When it was not possible to routinely assign a numeric rating through a comparison of model performance with criteria.a numeric rating was assigned based on professional judgment.Application of professional judgment requires:an understanding of open channel hydraulics. familiar'ity with the study site.experience with the models,and O(nowledge of how the model will De used in the habitat analysis . 0-/01- r~umeric ratings for each of the four cri teria are 2,I,or O.The models received cl rating depending on how well they :net the criteria.By summing the individual ratings.an overall rating was calculated for each model. Using the overall rating,models were evaluated according to the following sca le: - Excellent Good Acceptable Unacceptable 8 7 5-6 <5;or zero for any evaluation category - The level t",o criteria are based on analytical approach and ",ill only be made when a model is not considered excellent in the level one evaluation. 1. LEVEL ONE EVALUATION FOR IFG MODELS How well does the model conform to the IFG and EWT&A calibration guidelines? -- -- Compare predicted depths and velocities for calibration flows with observed field da ta. , ,Are the vel oci ty profi les rea li s ti c? Are there more than a few outl iers for the extrapolated flows? Do the predicted di scharges agree wi th the di scharges measured in the field (IFG-4 model only)for each transect? Are the predicted water surface elevations for a broad 'range of discharges coincident with the ra ting curves for each si te? Plot the water surface profiles.stage of zero flow,and thalweg. Are they reasonable?To be reasonable,the water must flow downhill; an increase in discharge should cause the pool riffle sequence to drown out and the wa ter surface profi le to become more un i form in gradient;a decrease in discharge should cause the water surface D-108 '7 -- - .~ - -\ - - profile to more acutely reflect changes in stream bed gradient and riffle pool profiles. 2 =A mode 1 tha t can foreca s t bo th wa ter sur face e~eva ti ons and ve 1oei ti es accura te 1y. 1 =A model that can define water surface elevations and velocities accurately at the calibration flows but may not be able to reliably define both WSEL and velocities for the extrapolated flows. 1)=A model that can not reproduce depths or velocities accurately at the ca 1ibra tion flow or throughout the extrapola tion range. 2.How well does the extra pol a ti on range of the mode 1 confqrm to the desired range? The first assumption rna'de in this evaluation is that the rating curves (site flow versus mainstem discharge and water -surface elevations'versus \ mainstem discharge for the site are accurate.The ability to evaluate the forecasting capabil i ties improve wi th an increase in number of transects which have well-defined rating curves.By reviewing aerial photography and incorporating field experience,determine if there are dramatic changes in the channel geometry or local flow patterns (such as ·other channels be- coming overtopped at higher mainstem discharges)that may cause a signifi- cant change in the site flow versus mainstem discharge relationship above the range of available data.The number of hydraul ic models required to describe the full spectrum of hydraulics in the site can be determined from this analysis (one for each straight-line portion of tile site now versus mainstem discharge plot).Low flow models should be able to describe the baseline flow conditions.High flow models describe t~E tlreached condi- tions and can be checked by comparing the water surface profiles and velocities with observed data. 2 =A model that can accurately define both water surface elevations and velocities accurately. 1 =A model tha t can describe ei ther veloci ties or wa ter surface profiles accurately. o =The model cantt describe depth and velocity for the defined range. 3.Are the mode 1 s appropri a te for the speci es and life stage be i ng considered? Cross sections should be ~ocated to accurately define cover.substrate.or other habi tat parameters whi ch are of importance to the species and/or 1 He stage of interest.Study sites set up for a particular species or life stage may not accurately represent the habitat conditions for a second - - - species or life stage. ~.,., -'.. •-.-~-Jo-~.,.".." Hydraulic models for juveniles should accurately define low velocity areas «0.8 ft!sec).but need not be as accurate when velocities exceed 2 fps. Depth needs onl y to be approxi rna te above 0.15 feet.and is of 1i ttle consequence in steep-sided channels where an error will not cause a notable change in top width. Hydraulic models for adults should accurately define velocities up to 2 ft/sec.and depths up to 1.0 feet. 2 =A model that provides sufficient precision in hydraulic forecasts to be applied to evaluation of adult and juvenile life phases wi til all equal level of confidence. 0-110 ,( - 1 =The model provides:a higher level of precision for evaluation of ei tiler adul t or juveni le 1i fe phase.The grea test accuracy of the model is for the life phase for which it was originally established but resulting hydraulic forecasts are sufficiently accura te to be acceptable for other 1 i fe phases.H'ld the study site been laid out differently,additional data collected or a separa te hydraul i c mode 1 ca 1i bra ted,an excellent ra ting would have been possible. o =Insufficient data were collected to calibrate the model in the flow range of interest for the species/life stages to be eva 1ua ted. 4.How well do the ranges of depth and velocities of the forecasted data conform to the ra.nges of depth and veloci ty of the sui tabil i ty criteria curves being considered based on a "visual"evaluation? Do ~he predicted hydraulic variables associated with a 'high percent error fall within the a.b.or c limits of the suitabili~curves? 1.0 \.0 ~at\b ~b c.-.-Z d ...C ~ ?:1\"t ..l J ~~r l- ':I I \ ;:: -''0(\-4\ 'L~ 0 ...........o j c.rl ~02L I.0 F'I C~l.JU t::2. 0-I/{ ~ Even though the model is not accurately reproducing depths or velocities from d hydraulic viewpoint.the predicted suitability indices could fall within.a range that is nat sensitive to errors in one of these indices. The calibrated model is linked with the habitat model and weighted usable area versus site flow plots are developed.Are the \-lUA projections continuing on the same trend beyond the extrapola tion range or is there a change in the trend? When there is a change in the WUA versus site flow relationship.similar to Figure 4.an upper limitstiould be established at the lowpofnt in the curve. - - c+--------------o-t------------- "'"" o o 2 =An accura te description of all ranges of depths and velaci ties present in the study site. 1 =Forecasting capabilities of the model are adequate when it accura tely describes two of the three ranges of the sui tabil i ty curve. a =When one or no ranges of the suitability curve are describ~d accura te 1 y. D-II'C1- .ft - LEVEL TWO EVALUATION FOR IFG MODELS Use of the level two criteria requires an analytical approach and should be applied when the forecast capabilities of either the IFG-2 and IFG-4 model are not given an excellent rating in the level one evaluation.These techniques can be incorporated as an additional step in the calbration procedure for future studies.The best method of evaluating the predictive capabilities of the hydraulic models is to collect an additional data set at each cross section that is not used in the calibration procedure and compare it to the model predictions.The test could not be applied, however,because·of the limited field data that were available.All data sets that were collected were used to calibrate the models. The analytical procedure presented has been suggested for use in geographic models which face simila.r pr.oblems in evaluating the differences between. ob~~r.ved and predicted data.To date,this is the most appropriate method to use in place of collecting an additional data set. A visual comparison is made between scatter plots of the observed and predicted depths and velocities at all cross sections for each calibration flow.The standard USGS discharge measurement procedure requires at least 20 -25 verticals where depth and velocity data are collected.For a particular channel the verticals at higher flows are spaced further apart than at low flows.Because a cell-by-cell comparison is made for the IFG-4 model,velocities must be assigned to the same cells at the same flows. The velocities are interpolated between -=.r.j-3cent cells for the high flows and used as i npu t for the mode 1.The !;~-4 ':'lode 1 with two or i!lOre flow s 0-113 7' generally has a larger number of verticals than the IFG-2 model suggesting this method of evaluation is more appropriate for the IFG-4 model. Scatterplot evaluations provide i)Qualitative assessment of the forecast capabilities of the model.A QU-intitative assessment can be made by computing several statistics which describe the differences between observed and predicted values {Willmott 1981}.Pearsonts Product-Moment - Correlation Coefficient (r).Coefficient of Determination (r2).the slope (b)and intercept (a)of a least squares regression between observed and predicted values are reported as the reliable measures of a model's predictive capabilities.Willmott has suggested computing additional stafistics to better evaluate the predictive capability of the model. These variables incl ude the systematic and unsystema tic components of the root mean square error RMSE S and -'as well as the where: JJ RH.SE U =[N-1 ~(Pi -(a +bO.»2 ]0 ~5 lei .1 total root mean square error tJ RMSE =[N-I ,?:(Pi -0i)2 ]0.5 I.e I i =1.2 •.••••••.n (sample size of the number of predicted cells) o =Observed or field measured data P =Model predicted data. An index of agreement (d)may also be calculated to determine the degree to which a model's predictions are error free.The index of agreement is computed by ~ d =1 -?-(?.-0.)2lei1] "J ]2)'[p.- 0 +O·- 011 ,...1 D-IIJ..{ ~ - - - - .- - ...The value of d varies between 0.0 and 1.0 where a computed value of 1.0 indica tes perfect agreement between the observed and predicted observa- tions,and 0.0 denotes complete disagreement. A visual comparison can-be ~ade-of the observed and predicted velocity distribution plots for the IFG-2 models,where much of the data is along the shorelines only.In general,the cells in the IfG-2 model do not coincide with verticals where field measurements were made,but rather with distinct changes in channel geometry,roughness,or habitat suitability.A representative velocity distribution "shape"was developed for each cross section.using the calibration flow data.which typically extended the full width of the cha-nnel.Where only shorel ine data was ava ilable.the-shape of the veloci ty profile was modeled after ei ther a similar cross section at the site where a complete data set was available.or by simply developing a shape based on the channe_t~eometry (i.e.,the highest velocities shoul~ correspond to the dee"pest portion of the channel).This isa reli_able.-,- fa··_~~I -._. ~thod.since cross-sectional area and discharge are fixed and therefore the average channel velocity is defined. Operating the IFG-2 model at discharges other than th.e calibration flow produces velocity profiles similar in-shape to that of the calibration flow.When inconsistencies between field data and predicted velocities occurred at high flows.a second model was developed.Generally,the high flow model predicts velocity profiles that are steeper near the water's edge than the corresponding low flow models. The level two analyses are nearly complete anCl ",ill be included in the draft report.Each of the models were evaluateu JOC rated using Chinook 0-1\5 )Y -----,-,-----------------..,....--------- juvenile rearing cr-i teria.A separa te evalua tion using the chum spawning criteT"ia ....ill be discussed in a later memor-andum after up ....elling informa- tion is col1ected.A summary of the "application ranges of the calibrated model.s wit~their associated ratings is presented in F'gurO!l.The hydraulic relationships used in the calibration effort are listed in the Appendix tables and should be used in the habitat modeling and flow dura- ti on ana 1y sis. SITE-SPECIFIC EVALUATIOUS site'S The specific evaluations 6~the middle riverAare not given because they -. are not applicable to the lower river study. - - ~. D-\\1.D L------------------ -- - DRAFT/PAGE 1 4/20/85 3/7/85,3/15/85,4/30/85,5/1/85 NUM3/Title Page Ci1AFT PART 3 Resident Fish Distribution and Population Dynamics in the Susitna River below Devil Canyon """". - - - - CrtAFT DRAFT/PAGE 2 4/20/85 3/7/85,3/15/85,4/30/85,5/1/85 NUM3/Title Page ", RESIDENT FISH DISTRIBUTION AND POPULATION DYNAMICS IN THE SUSITNA RIVER BELOW DEVIL CANYON Report No.7,Part 3 by Richard L.Sundet and Stuart D.Pechek Susitna Aquatic Studies Program Alaska Department of Fish and Game 620 E.10th Avenue,Suite 302 Anchorage,Alaska 99501 ABSTRACT Studi es of res i dent fi sh were conducted in both the lower and l11i ddl e, Susitna River in 1984.Primary emphasis in the middle river was to determine the seasonal distribution,t"iming of spawning,and spawning areas of ra"inbow trout,and to monitor 13 index sites as part of the long term monitoring effort.Most of the rainbow trout data was col- lected by use of radio telemetry.Results showed that rainbow trout are relatively few in numbers and that spawning occurs at selected areas which are influenced by lakes.Much of the rainbow trout population in the middle river probably originates in lakes which outlet to middle river tributaries.Lakes where rainbow trout are abundant and which probably contribute heavily to middle river populations were at the headwaters of Fourth of July Creek and in the upper reaches of Portage Creek.Rainbow trout were also found to use Portage Cre€k more exten- sively than previously thought.Spawning occurred during the first week of June.All rainbow trout move out of tributaries by early October CflAFT CrtAFT DRAFT/PAGE 3 4/20/85 3/7/85,3/15/85,4/30/85,5/1/85 NUM3/Title Page (probably triggered by low fall discharges),and most overwinter in the mainstem Susitna River slightly downstream (0.1-4.0 miles)of the tributary where they were captured.Other middle river studies suggest Arctic grayling overwinter in the mainstem Susitna then ascend and spawn in tributaries in late May.Arctic grayling also outmigrate from tributaries at the same time as rainbow trout.Catch data at middle river index sites in 1984 was similar to 1982 and 1983 findings. Studies in the lower river reinforces the belief that some humpback whitefish are anadromous,and rainbow trout and Arctic grayling outmigrate from most east side tributaries in September.Lower river studies also found that burbot move into the Deshka River in mi d-September. - r r r r r DRAFT/PAGE 1,4/20/85,4/30/85,5/1/85 3/7/85,3/15/85, 3/20/85,4/8/85 NUM3/Table of Contents, TABLE OF CONTENTS ABSTRACT LIST OF FIGURES LIST OF TABLES LIST OF PLATES LIST OF APPENDIX FIGURES LIST OF APPENDIX TABLES 1.0 INTRODUCTION 2.0 METHODS 2.1 Study Locations 2.1.1 Relative abundance measurements 2.1.2 Population estimates 2.1.3 Radio telemetry 2.2 Data Collection 2.2.1 Relative abundance 2.2.2 Population estimates 2.2.3 Radio telemetry Equipment Transmitter implantation Tracking 2.3 Data Recording and Analysis 3.0 RESULTS 3.1 Distribution and Relative Abundance of Resident Fish on the Lower Susitna River 3.1.1 Rainbow trout 3.1.2 Burbot 3.1.3 Arctic grayling 3.1.4 Round whitefish 3.1.5 Humpback whitefish 3.1.6 Longnose suckers ----,~,--------·---_".i-..------~--------------- DRAFT/PAGE 2,4/20/85,4/30/85,5/1/85 3/7/85,3/15/85,3/20/85,4/8/85 NUM3/Table of Contents, TABLE OF CONTENTS (Continued) 3.1.7 Other species Dolly Varden Northern pike Threespine stickleback Ninespine stickleback Arctic lamprey 3.2 Resident Fish Index Site Monitoring on the Middle Susitna River 3.2.1 Tributary mouth sites 3.2.2 Slough sites 3.2.3 Mainstem sites 3.3 Radio Telemetry Studies of Selected Resident Fish on the Middle Susitna River 3.3.1 Rainbow trout 3.3.2 Burbot 3.3.3 Arctic grayling 3.4 Other Resident Fish Studies on the Middle River 3.4.1 Lake surveys 3.4.2 Tag-and-recapture studies Rainbow trout Burbot Arctic grayling Round whitefish Humpback whitefish Longnose suckers Dolly Varden 4.0 DISCUSSION 4.1 Lower Susitna River 4.1.1 Rainbow trout 4.1.2 Burbot 4.1.3 Arctic grayling 4.1.4 Round whitefish 4.1.5 Humpback whitefish 4.1.6 longnose suckers 4.1.7 Other species Dolly Varden Northern pike Threespine stickleback Ninespine stickleback Arcti c 1amprey - - .... DRAFT/PAGE 3,4/20/85,4/30/85,5/1/85 3/7/85,3/15/85, 3/20/85,4/8/85 NUM3/Table of Contents,. TABLE OF CONTENTS (Continued) 4.2 Middle Susitna River 4.2.1 Rainbow trout 4.2.2 Burbot 4.2.3 Arctic grayling 4.2.4 Round whitefish 4.2.5 Humpback whitefish 4.2.6 Longnose suckers 4.2.7 Other species Dolly Varden Lake trout Threespine stickleback 5.0 .CONTRIBUTORS 6.0 ACKNOWLEDGEMENTS 7.0 LITERATURE CITED 8.0 APPENDICES Appendix A-Floy anchor tag retention rates Appendix B -Radio tagged fish tagging and habitat data Appendix C -Population and biological characteristics Appendix D-Population estimates Appendix E -Middle river index site catch data and descriptions,and spawning rainbow trout habitat data - - DRAFT/PAGE 1,4/30/85,5/1/85 3/15/85,3/20/85,4/8/85,4/20/85 NUM3/List of Figures LIST OF FIGURES ..... Figure 1 2 3 4 5 6 7 8 9 10 11 12 13 Title Resident fish study sites on the Susitna River between the Chulitna River confluence and Devil Canyon,1984 Fourth of July Creek drainage Lower Portage Creek drainage Movement of 12 radio tagged rainbow trout in the Susitna River below Devil Canyon,June 1983 to May 1984. Movement of five radio tagged rainbow trout in the Susitna River below Devil Canyon,Septem- ber 1983 to October 1984 Movement of four radio tagged rainbow trout in Fourth of July Creek,May to October 1984 Movement of six radio tagged rainbow trout in Indian River,May to October 1984 Movement of five radio tagged rainbow trout in the Susitna River,May to December 1984 Movement of five spawning rainbow trout, tagged at TRM 2.3 Portage Creek,in Portage Creek and then in the mainstem Susitna River, May to December 1984.At capture,all five fish were pre-spawners. Movement of five rainbow trout and one Arctic grayling in Portage Creek and then in the mainstem Susitna River,May to December 1984. Movement of four radio tagged rainbow trout in Fourth of July Creek and then into the rna i n- stem Susitna River,May to December 1984 Frequency distribution of radio tagged rainbow trout locations in tributaries,at tributary. mouths,and in the mainstem Susitna River during 1984 Mean surface water temperatures in Fourth of July Creek correl ated to four spawni ng radi 0 tagged rainbow trout's movement from the Susitna River into Fourth of July Creek,1984 1 1 3\ 4-0 LIST OF FIGURES (Continued) Figure 14 DRAFT/PAGE 2~4/30/85~5/1/85 3/15/85~3/20/85~4/8/85~4/20/85 NUM3/List of Figures Title Mean surface water temperatures and relative depths in Portage Creek correlated to seven radio tagged rainbow trout1s maximum movement and their outmigration into the ma'instem Susitna River~1984 - - ~, - ..., DRAFT/PAGE 3,4/30/85,5/1/85 3/15/85,3/20/85,4/8/85,4/20/85 NUM3/Li st of Fi gures LIST OF TABLES - - Table 1 2 3 4 Title Resident fish sampling schedule at tributaries in the middle reach of the Susitna River, 1984. Habitat data collected at RM 57.2 where 50 ninespine stickleback were captured,August 5, 1984. Boat electrofishing catch and catch per unit effort (CPUE)of four resident fish species in the middle Susitna River by three macrohabitat types. Catch data of resident fish species in the middle river at opportunistic and lake sites, May to October 1984 ---------_._---------------~--------------- ""'" - - - DRAFT/PAGE 4,4/30/85,5/1/85 3/15/85,3/20/85,4/8/85,4/20/85 NUM3/List of Figures LIST OF PLATES - Plate 1 2 3 Title Boat electrofishing in the mainstem Susitna River at RM 147.8 and angl ing at TRM 2.3 of Portage Creek,June 1984 Locating a radio tagged rainbow trout in the mainstem at RM 111.4 and measuring water velocities at this location,February 1984 Implanting a radio tag into the abdomen of a rainbow trout (on left)and externally radio tagging a rainbow trout (on right) ---I-,---------------r-I-----.........----------------- - - - - - DRAFT/PAGE 5,4/30/85,5/1/85 3/15/85, 3/20/85,4/8/85,4/20/85 NUM3/List of Figures LIST OF APPENDIX FIGURES Appendix Figure Title C-1 Length frequency composition of rainbow trout captured in the Susitna River between Cook In 1et and Devi 1 Canyon by all gear type,May to October 1984 ",.. C-2 C-3 C-4 C-5 C-6 C-7 C-8 C-9 Length frequency composition of burbot cap- tured in the Susitna River between Cook Inlet and Devil Canyon by all gear types,May to October 1984 Length frequency compos i ti on of Arct ic gray- ling captured in the SusitnaRiver between Cook Inlet and Devil Canyon by all gear types, May to October 1984 Length frequency composition of round white- fish captured in the Susitna River between Cook Inlet and Devil Canyon by all gear types, May to October 1984 Length frequency compos iti on of humpback whitefi sh captured in the Susitna River between Cook In 1et and Dev i 1 Ca nyon by all gear types,May to October 1984 Length frequency compos i ti on of longnose suckers captured in the Susitna River between Cook Inlet and Devil Canyon by all gear types, May to October 1984 Tim"ing of 1984 rainbow trout and Arctic grayling spawning in the middle Susitna River determi ned by the i nci dence of pre-to post-spawners Age and length relationships for spawning rainbow trout captured "in the Susitna River between the Chulitna River confluence and Devil Canyon,May 17 through June 27,1984 Age and length relationships for spawning Arctic grayling captured in the Susitna River between the Chulitna River confluence and Devil Canyon,May 17 to June 5,1984 (.,-3 (.;-5 DRAFT/PAGE 6,4/30/85,5/1/85 3/15/85,3/20/85,4/8/85,4/20/85 NUM3/List of Figures LIST OF APPENDIX FIGURES (Continued)- Appendix Figure C-I0 C-ll C-12 C-13 Title Age and length relationships for spawning round whitefish in the Susitna River between Cook Inlet and Devil Canyon,October 9 to October 15,1984 Age and length relationships for pre-spawning rainbow trout captured in lakes C and D at the headwaters of Fourth of July Creek,September 14,1984 Age and length relationships for rainbow trout captured in the Susitna River between the Chul itna Ri ver confl uence and Devil Canyon, May to October 1984 Survival rate curves for rainbow trout cap- tured in the Susitna River between the Chul itna River confl uence and Devi 1 Canyon, 1983 and 1984 - - - DRAFT/PAGE 7,4/30/85,5/1/85 3/15/85, 3/20/85,4/8/85,4/20/85 NUM3/List of Figures LIST OF APPENDIX TABLES ,.... Appendix Table Title B-2 r- ~"'~-C-1 C-2 B-1 Summary of tagging data for radio tagged rainbow trout captured on the Susitna River between the Chulitna River confluence and Devil Canyon,May to July 1984. Radi a tagged rainbow trout habitat measure- ments taken at their relocations in January and February 1984. Length data for resident fish captured on the Susitna River,1984. Fork lengths (mm)of sexually mature and immature resident fish captured on the Susitna River,1984. C-(" - C-3 0-1 0-2 E-1 E-2 Rainbow trout age-length relationships on the Susitna River between the Chulitna River confl uence and Devi 1 Canyon,May to October 1984. Seber-Jolly method-population,survival and recruitment from a four-catch or longer experiment. Bailey's deterministic method-population, survival and recruitment from a three-catch experiment. Boat electrofishing catch and catch per unit effort (CPUE)of four resident fish species at 13 index sites in the middle reach of the Susitna River in 1984. Habitat characteristics and measurements taken at spawning rainbow trout sites in 1984. (,-)1..-- 0-] £-\ ---,-,--------------------------------------- - CRAFT 1.0 INTRODUCTION DRAFT/PAGE 1,4/20/85, 4/11/85, 4/8/85,3/7/85,3/15/85,3/20/85 NUM3/Introduction,4/30/85,5/1/85 .... - - Since November 1980,Resident Fish Studies have been conducted in the Susitna River drainage to more accurately determine the distribution and relative abundance.of resident fish.In 1982,resident fish abundance were compared at rna i nstem,slough,and tri buta ry mouth macrohabi tats (ADF&G 1983d).In 1983,studi es focused more on the mi ddl e reach of river because construction of the proposed hydroel ectric dams woul d affect this reach of river most.Microhabitat suitability criterias and population estimates were developed for rainbow trout (Salmo gairdneri Richardson)and burbot (Lota lota Linnaeus)in selected areas during 1983 (Suchanek et al.1984;Sundet and Wenger 1984).Rainbow trout and burbot movements were more clearly defined using radio telemetry in 1982 and 1983 (ADF&G 1983c;Sundet and Wenger 1984).Data from these studies and·catch data has shown burbot most often reside in the mainstem in summer and in both mainstem and some tributaries in winter.These data This report primarily addresses resident fish studies conducted during the 1984 open-water season.However,radio telemetry results also present fish movements from December 1,1983 to the open-water season (for winter monitoring of 1983 radio tagged fish)and after the open- water season to December 1,1984 (to show movement patterns of summer CRAFT DRAFT/PAGE 2,4/20/85, 4/11/85, 4/8/85,3/7/85,3/15/85,3/20/85 NUM3/Introduction,4/30/85,5/1/85 1984 tagged fish during the transition period from open-water to winter conditions).Although sampling primarily occurred in the mainstem Susitna River between Cook Inlet and Devil Canyon,sampling also occurred at several tributaries in this reach of river. The primary emphasis of Resident Fish Studies in 1984 was to further determine the seasonal distribution,timing of spawning,and locations of spawning for rainbow trout above Talkeetna using radio telemetry. Studies in 1984 also furthered our knowledge of resident fish distribution in the Susitna River below Talkeetna. The radio telemetry program in the Susitna River below Talkeetna in 1984 monitored rainbow trout and burbot movements.Fish below Talkeetna as well as several fish in the middle river were radio tagged after August 1984.This report will include radio tagged fish monitoring data from December 1983 to December 1984.Since data from fish radio tagged after August is 1 imited,these data wi 11 be presented ina report to be published in November 1985 covering winter 1984-85 studies. In addition to radio telemetry studies,13 resident fish index areas between Talkeetna and Devil Canyon were sampled regularly during 1984 to evaluate trends in abundance of resident fish.These 13 index areas, which have been sampled each year since 1982,encompass three macro- - .- - - - habitats (tributary mouths,sloughs,and mainstem).By annual k a monitoring these sites where larger concentrations of resident fish have traditionally been found in comparison to other sites,the effects of the proposed hydroelectric dams to resident fish populations can be - \vN-.,a """'1.. .-/- assessed by habitat type. DRAFT/PAGE 3,4/20/85, 4/11/85, 4/8/85,3/7/85,3/15/85,3/20/85 NUM3/Introduction,4/30/85,5/1/85 These post-project .~lfg£l§resul t from vJkj·J ~~__,,,,o,_,,-,,,,,=_,,,,,,,,,,,,,,,,,,-,",,,,";,,",,,,,,,",......"_,.~.~.,•....-.,.,.",,'c""',CJ~"'" changes in water temperature,flow,turbi dity,and other water qual ity - -- ""'" parameters. 3 DRAFT/PAGE 1 4/11/85,4/23/85 3/7/85,3/15/85,3/20/85,4/8/85 NUM3/Methods,4/30/85,5/1/85 2.0 METHODS 2.1 Study Locations 2.1.1 Relative abundance measurements Thirteen index sites between Talkeetna and Devil Canyon were sampled ~ twice per month by boat electrofishing to monitor seasonal trends in relative abundance of resident fish (Figure 1).Site descriptions of Skull Creek (RM 124.7),Susitna Mainstem -West Bank (RM's 137.3-138.3), Susitna Mainstem (RM's 147.0-148.0),and Susitna Mainstem -Eddy (RM 150.1)are provided in Appendix E,while the remaining nine site ~ descriptions are presented in Appendix F of Aquatic Instream Flow Studies,1982 (ADF&G 1983f).In addition,other mainstem,side channel, slough,and tributary sites on the Susitna River between Cook Inlet and Devil Canyon were sampled intermittently. Several tributaries in the middle reach of the Susitna River were sampled during 1984 to determine the extent of rainbow trout and Arctic grayling (Thymallus arcticus Pallus)spawning and rearing in these tributaries (Table 1).These tributaries were selected because of their size,their proximity to Devil Canyon,or due to their relatively high ~ abundance of these fish species. SA Slough 20 Susitno Moins1em-Wesl Bank 113.6 124.7 131.1 138.6 144.5 148.8 101.2 112.3 125.3 140.1 137.3 -138.3 147.0 -148.0 150.1 Slough Sftes Whiskers Creek Slough -ROUth Slough 6A Slough SA Slough 20 -Moutb MlIfnstem Sftes Susftna Mafnstem -West Ban£ Susftna Mainstem Susitna Mainstem -Eddy River Mile Tributary Mouth Sites Lane Creek Skull Creek Fourth of July Creek Indian River Jack Long Creek Portage Creek R. t ~. ,~ Figure 1.Resident fish study sites on the Susitna River between the Chulitna River confluence and Devil Canyon,1984. 5 DRAFT/PAGE 2 4/11/85,4/23/85 3/7/85,3/15/85,3/20/85,4/8/85 NUM3/Methods,4/30/85,5/1/85 Table 1.Resident fish sampling schedule at tributaries in the middle reach of the Susitna River,1984. RM TRM May Jun Aug Oct Whi skers Creek 101.4 7.0 X Fourth of July Creek 131.1 0.1 -3.5 X X X Indian River 138.6 0.1 -9.0 X X Portage Creek 148.8 0.1 -11.6 X X In the lower Susitna river,the upper reaches of Kashwitna River (RM 61.0),Sheep Creek (RM 66.1),Goose Creek (RM 72.0),and Montana Creek (RM 77.0)were sampled during early September to determine the extent of rainbow trout and Arctic grayling rearing in these tributaries. Attempts were made to radio tag fish in these tributaries to determine the timing of their fall outmigration.Radio telemetry results from fish tagged in these areas will be presented in a later report. Six lakes with outlets to the middle Susitna River were surveyed in 1984.These surveys were done to determine if rainbow trout populations existed in these lakes and if these fish migrate to or from the mainstem ~ Susitna River.Surveys were conducted at four lakes at the headwaters of Fourth of July Creek [Lakes A,B, C,and D (Figure 2)J,Miami Lake which outlets into Indian River at TRM 4.5,and an unnamed lake which outlets into Portage Creek at TRM 2.3 (Figure 3). Resident fish catches recorded at five fishwheel sites,two outmigrant trap sites,a fyke net weir site,and 20 juvenile salmon rearing study sites (JAHS)were also examined to evaluate trends in relative abundance - -, - - Figure 2. Figure 3. Fourth of July Creek drainage. PORrAS€CR€EK (RM 148.81 Lower Portage Creek drainage. 1 •CABIN •RIV£R "I~E o I I •, "'1.£5 DRAFT/PAGE 3 4/11/85,4/23/85 3/7/85,3/15/85,3/20/85,4/8/85 NUM3/Methods,4/30/85,5/1/85· and seasonal movements.In addition,several east side tributaries such as Kashwitna River were monitored in April and May to determine timing of immigration of resident fish from the mainstem Susitna River. 2.1.2 Population estimates Population esimates were made using multi-year data (1981-84)for four resident fish species tagged in the middle river (see Appendix D). 2.1.3 Radio telemetry Selection of radio tagging sites in the mainstem Susitna between the Chulitna River confluence and Devil Canyon were based on resident fish distribution data collected during the 1981,1982,and 1983 open water field seasons (ADF&G 1981b,1983b;Sundet and Wenger 1984).Primary efforts to capture and radio tag rainbow trout and Arctic grayling in the mainstem were focused at the mouths of Whiskers Creek,Lane Creek (RM 113.6),Fourth of July Creek,Indian River,and Portage Creek.Fish were also caught and radio tagged in the upper reaches of Fourth of July Creek,Indian River and Portage Creek at the same time the spawning studies were done. 2.2 Data Collection 2.2.1 Relative abundance Resident fish were collected at mainstem and tributary sites primarily with a boat mounted electrofishing unit (Plate 1).A Coffelt Model - - - -Plate 1.Boat electrofishing in the mainstem Susitna River at RM 147.8 and angling at TRM 2.3 of Portage Creek,June 1984. ______._......._,~_~l,,-------"""""--- DRAFT/PAGE 4 4/11/85,4/23/85 ., 3/7/85,3/15/85,3/20/85,4/8/85 NUM3/Methods,4/30/85,5/1/85 VVP-3E boat electrofishing unit powered by a 2,500 watt Onan generator was used for boat electrofishing and using techniques described in ADF&G (1983a).Secondary gear types used included outmigrant traps at RM 22.4 and at RM 103.0,a fyke net weir at TRM 2.5 of the Deshka River, backpack electrofishing units,gill nets,hook and line,hoop nets,and trotlines. Biological data (age,length,sex,and sexual maturity)were collected as outlined in ADF&G (1983a).Scales for age determination were taken Scales were also taken from spawning Arctic grayl i ng and round whitefi sh to determi ne ages of spawners for these from a representative sample of rainbow trout captured above the 1.., S Chulitna River confluence. :f~"ti~.species.L]~e:u-~\-A.r~<t/ti)L~~. ArtL ~ Survival rates were calculated for rainbow trout above the Chul i tna River confluence in 1984 using catch and age data and methods presented in Everhart et ale (1975). Habitat parameters were measured at locations where spawning rainbow trout were found.These parameters included water velocity,water depths,substrate type,and general water quality.Specific habitat data collection methodology are summarized in ADF&G (1983a). Habitat parameters were also measured at radio tagged fish relocation sites during winter 1983-84.During ground surveys in January and February,some radio tagged fish were located to within a four-foot- radi us and habitat measurements were made as close to the si gnal as - -- DRAFT/PAGE 5 4/11/85,4/23/85 3/7/85,3/15/85,3/20/85,4/8/85 NUM3/Methods,4/30/85,5/1/85 possible (Plate 2).Habitat parameters measured included those taken at fish spawning sites as well as ice thickness and the presence or absence of slush ice.In January and February the fate of each radio tagged fish surveyed was also determined. A tag-and-recapture program was continued in 1984 to monitor the season- al movements of adult resident fish.Floy anchor tags were used to tag seven species of adult resident fish:rainbow trout,Arctic grayling, humpback whi tefi sh (Coregonus pi dschi an Gmel in),round whi tefish, burbot,longnose suckers (Catostomus catostomus Forster),and Dolly Varden (Salvelil1us malma Walbaum).All resident fish that appeared hea lthy after capture and were 1arge enough to accommodate a tag were tagged.Burbot with a total length (TL)greater than 225 millimeters (mm)were tagged and other resident species with fork lengths (FL) greater than 225 mm were tagged. Tag recoveries were made by the resident fish study group,the adult salmon fishwheel crews,and the angling public. 2.2.2 Population estimates Population estimates were made for adult (~200 mm)rainbow trout, Arctic grayling,round whitefish,and longnose suckers in the middle river using the multiple year (1981-84)tagging and recapture data (see Appendix D). 11 .~ ""..'·1··..~.t.~,. I¥ i" "\~ ,I :"'''''1-i ~J,Ji ~ Plate 2.Locating a radio tagged rainbow trout in the mainstem at RM 111.4 and n~asuring water velocities at this location,February 1984.. '~ I _J ....... - - DRAFT/PAGE 6 4/11/85,4/23/85 3/7/85,3/15/85,3/20/85,4/8/85 NUM3/Methods,4/30/85,5/1/85 2.2.3 Radio telemetry Most fish which were radio tagged were captured by boat electrofishing or by hook and line (Plate 1). Equipment Radio telemetry receiving equipment used in this study was developed by Smi th-Root Incorporated in Vancouver,Washi ngton.Receivi I1g equi pment consisted of a low frequency (40 MHz)radio tracking receiver (Model RF-40)and scanner (Model SR-40),and a loop antenna (Model LA-40). Radio transmi tters used in 1984 were manufactured by Advanced Telemetry Systems (ATS)of Bethel,Minnesota.Two types of radio tags were used: internal and external.Internal radio tags with 6-11 month life expec- tancies were implanted in rainbow trout.External r?dio tags were attached to several pre-spawning rainbow trout and one Arctic grayling. External tags were used on several pre-spawning rainbow trout since it was believed the condition of some of these fish would be unacceptable for internal implants.An external radio tag was used on the Arctic grayling since past efforts to internally radio tag this species have failed (ADF&G 1983c). Radio tags used in rainbow trout and burbot tagged in 1983 and monitored during 1984 were somewhat different.Refer to Suchanek et al (1984)and ----_._---------=------_._._----------------_...... DRAFT/PAGE 7 4/11/85,4/23/85 3/7/85,3/15/85,3/20/85,4/8/85 NUM3/Methods,4/30/85,5/1/85 Sundet and Wenger (1984)for methods of tagging,transmitter types,bio- logical characteristics,and summer movement of fish tagged in 1983. Advanced Telemetry Systems'internal transmitters (Model 10-35)used in 1984 were identical to those used in 1983 studies except pu1 se rates were slightly higher [1.0 and 2.5 pulses per second (pps)]. Advanced Telemetry Systems'external radio tags (Model RM 625)are rectangularly shaped,encapsulated in epoxy,and have slightly flexible 24.0 centimeter (cm)external antennas.The external transmitters are 0.9 cm high,1.5 cm wide,3.0 cm long,and have a dry weight of approxi- mately 9.5 grams.The power source for the transmitters is a 1.4 volt mercury battery providing life expectancies of 90 days. for these tags are 2.4 pps. The pulse rates -The same transmitter frequencies (40.600-40.770 MHz)were used in 1984 as in 1983 studies.All radio tags were immersed in cold water (105°C)- for 48 hours to ensure they were transmitting properly before they were -. implanted in fish. Transmitter implantation Based on personal communications with Carl Burger (USFWS)and experience gathered from the previous three years of radio telemetry studies,the minimum fork lengths of rainbow trout and Arctic grayling radio tagged was set at 380 mm (ADF&G 1983b,1983c;Sundet and Wenger 1984). It- DRAFT/PAGE 8 4/11/85,4/23/85 3/7/85,3/15/85, 3/20/85,4/8/85 NUM3/Methods,4/30/85,5/1/85 Internal radio tags were implanted using the same procedures described in Ziebell (1973).Before surgery,fi sh were anesthetized wi th MS-222 (tricaine methane-sulfonate). ,..... Prior to attaching external tags to fish,two Peterson disc needles were epoxied to one side of each radio tag so the needles were perpendicular to the length of the tag.Fish to be externally radio tagged were anesthetized and then the exte.rnal tag was attached just below the dorsal fin (Plate 3).This method was similar as attaching Peterson disc needles described in the adult anadromous section of the 1983 .- - ..... ,'".., procedures manual (ADF&G 1983a).This was accomplished by using pliers to push the two Peterson needles through the dorsal portion of the fish . After the needles were through the fish,one Peterson disc was attached to each needl e.The radio tag and Peterson di scs were then pushed tightly next to the fish and excess needle (over 1.0 cm past the discs) were cut off with pliers.The remaining needles were then rolled with the tip of the pliers to tighten and secure the tag in place. After radio.tagging,the fish were placed into a live box and held upright until they regained their equilibrium.The fish were then held overnight whenever possible for observation.The following day the sutures were checked and the transmitter's signal was tested before releasing each radio tagged fish near their point of capture. it) ----_._-----------,-------_._--------~------- (S" Plate 3.Implanting a radio tag into the abdomen of a rainbow trout (on left)and externally radio tagging a rainbow trout (on right). ...J ..J J .J J J -. j~ - DRAFT/PAGE 9 4/11/85,4/23/85 3/7/85,3/15/85, 3/20/85,4/8/85 NUM3/Methods,4/30/85,5/1/85 Tracking Biologists radio tracked fish primarily by fixed wing aircraft and boat. Aerial radio tracking was done using methods described in Adult Anadromous Investigations 1981 (ADF&G 1981c).Radio tracking was conducted over the mainstem Susitna by fixed wing from December 1983 to December 1984.Between December 1,1983 and May 1,1984,radio tracking was done between RM 40.6 and RM 152.0 approximately every 20 days.From May 1 to August 30,1984 radio tracking flights were made every 10-14 days between RM 97.0 (Talkeetna)and RM 152.0.From September 1 to December 1,1984 radio tracking was conducted between RM 0.0 and R~1 152.0 every 14 days.Fixed wing tracking was also done over various tributaries such as Fourth of July Creek and Portage Creek from May 1 to December 1,1984. Aerial tracking by helicopters was also done occasionally during winter 1983-84 and in May and June 1984.During the summer,radio tracking was also done by boat during each field trip to pinpoint radio tagged fish. "------_._----------------------------- DRAFT/PAGE 1,4/11/85,4/23/85 3/7/85,3/15/85,3/20/85,4/8/85 NUM3B/Results,4/23/85,5/1/85 3.0 RESULTS 3.1 Distribution and Relative Abundance of Resident Fish on the Lower Susitna River 3.1.1 Rainbow trout Rainbow trout were first recorded captured on May 6 at the Deshka River. Other early immigration of rainbow trout were reported by sport fishermen at Kashwitna River between May 16-19 (David James,pers. comm.).·During this time,water temperatures were 6.0°C (on May 6)at the Deshka River and 8.2°C (on May 10)at Kashwitna River (TRM 1.5). A total of 155 rainbow trout were captured by all methods in 1984.The highest rainbow trout catches (62 fish)were at the Deshka River. Relatively high catches were made during boat electrofishing sampling in the mainstem Susitna River between RM 30.0 and RM 98.5 in early September (31 fish)and at the mouth of Little Willow Creek (RM 50.3,14 fish)during late September.Only nine rainbow trout were captured in upper reaches of east side tributaries during early September. In 1984,73 rainbow trout were Floy anchor tagged in the lower river and four fish were recaptured.All four fish were tagged and recovered in the Deshka River in 1984 and moved a maximum of 3.5 miles. )~ - - -I - - DRAFT/PAGE 2,4/11/85,4/23/85 3/7/85,3/15/85, 3/20/85,4/8/85 NUM3B/Results,4/23/85,5/1/85 3.1.2 Burbot A total of 334 burbot were captured in 1984.Of these,the 282 adult burbot (~200 mm)were caught at the Deshka River between TRM's 0.0 and 6.0.Adult catches at the Deshka River were high in May,September and October,however,little sampling was done between these time periods. Few juvenile burbot (4 200 mm)were captured.Twenty-one juveniles were captured at JAHS and the Deshka River weir sites,while none were captured by outmigrant traps at RM 22.4. In 1984,197 burbot were Floy anchor tagged in the lower river.Most of these burbot (178)were captured at the Deshka River between TRM 0.0 and TRM 6.0.Twenty-five recoveries were made in 1984 from 23 different fish.All 25 recoveries were made in the Deshka River between TRM 0.0 and TRM 6.0.Nearly all (23)of these recoveries were from burbot tagged in this reach of the Deshka River in September or October 1984. Another recovered burbot was tagged in the Deshka River in May.The remaining recovery was from a burbot tagged at Anderson Creek (RM 23.8) on June 22,1981 and recaptured on October 15,1984 at the mouth of the Deshka River (RM 40.6).During the interim,it grew from 488 mm to 581 mm (TL). 3.1.3 Arctic grayling In 1984,Arctic grayling immigration from the mainstem Susitna River were fi rst reported by 1oca 1 fi shermen at Grey I s Creek (RM 59.5)on DRAFT/PAGE 3,4/11/85,4/23/85 3/7/85,3/15/85,3/20/85,4/8/85 NUM3B/Results,4/23/85,5/1/85 April 28,and Rabi deux Creek (RM 83.1)on May 1.Peak catches for Arctic grayling were reported at Grey's Creek on May 7 - 8 and at Kashwitna River (RM 61.0)from May 16-19 (David James pers.comm.). A total of 271 Arctic grayling were captured in 1984.Most (60%)of the fish were captured during September and early October boat electro- fishing efforts.The maximum catch during these efforts was 69 fish at mainstem sites between RM 60.1 and RM 98.5.Only two Arctic grayl"ing were captured in upper reaches of east side tributaries during early September. Ninety-nine Arctic grayling were Floy anchor tagged in the lower river and no recaptures were made in 1984. 3.1.4 Round whitefish -; A total of 1,195 round whitefish were captured in 1984.Round whitefish were most abundant at JAHS sites.Eight hundred three juvenile (<:200 mm)and one adul t round whi tefi sh (~200 mm)were captured at 20 regul arly sampl ed sites and several opportuni sti c sites ~Catches over 100 round whitefish were recorded at four side channel sites:Sucker'" (RM 84.5),Beaver Dam (RM 86.3),Sunset (RM 86.9),and Sunrise (87.0). Adult catches were made mostly by boat electrofishing in the mainstem between river miles 60.1 and 98.5 where 72 fish were captured. DRAFT/PAGE 4,4/11/85,4/23/85 3/7/85,3/15/85,3/20/85,4/8/85 NUM3B/Results,4/23/85,5/1/85 In October,19 sexually ripe round whitefish were captured at six sites. Most spawning fish (8)were found in the mainstem at RM 70.0.Several individual or pairs of spawning fish were also caught in the mainstem between RM 50.5 and RM 84.0. In 1984,113 round whitefish were tagged in the lower river and four round whitefish were recovered.Three of the fish (one tagged each year in 1981,1982,and 1984)were recaptured less than 5.0 miles from their tagging site.The remaining fish moved from Montana Creek (RM 77.0)to TRM 1.5 of the Kashwitna River (RM 61.0)in two years. 3.1.5 Humpback whitefish Six hundred eighty-seven.humpback whitefish were captured in 1984. Most (94.2%)of the fish were captured by fishwheels or by outmigrant traps.Outmigrant traps located at Flathorn Station (RM 22.4)captured 71 juvenile humpback whitefish with 26.8%of the catch occurring in early September.No adult humpback whitefish (2200 mm)were captured at the outmigrant trap site. Fi shwheel catches of humpback whitefi sh were the greatest at Fl athorn Station (RM 22.4)where 369 adults were captured.Fishwheels at Sunshine (RM 79.0)and Yentna (RM 28.5,TRM 4.0)stations also captured over 70 humpback whitefish.The maximum seasonal catch at all fishwheel sites was during late August with relatively high catches also in early July to early August. ~"."....---,--------------_..---------~---_......._------ DRAFT/PAGE 5,4/11/85,4/23/85 3/7/85,3/15/85,3/20/85,4/8/85 NUM3B/Results,4/23/85,5/1/85 Boat electrofishing humpback whitefish catches were the highest (16 of 27 fish)at mainstem sites between RM 30.0 and RM 98.5.No catches were made at JAHS sites,but some round whitefish catches may have been misidentified and were actually humpback whitefish (Paul Suchanek,pers. comm.). In 1984,261 humpback whitefish were tagged in the lower river and three were subsequently recaptured by fi shwheel s.All three recaptures were tagged at Flathorn Station (RM 22.4).One fish was recovered at Flathorn Station eight days later.The other two were recovered at Yentna Station (RM 28.5,TRM 4.0).The time between tagging and recovery of these two fish was 2 and 30 days. 3.1.6 Longnose suckers Eight hundred sixty-two longnose suckers were captured in 1984.Most longnose sucker catches (326 fish)were made at JAHS sites and all but a few were juveniles «200 mm).Catches at JAHS sites were the highest at Sunrise Side Channel (RM 87.0,53 fish).Catches over 40 fish were also made at Hooligan (RM 35.2)and Sucker (84.5)side channels. Higher boat electrofishing catches (145 ·of 191 fish)were recorded at mainstem sites between RM's 30.0 and 60.0 compared to other boat electrofishing sites.Relatively high catches were also made at the Deshka River and fishwheel sites (226 and 76 fish,respectively).Most (50%)fishwheel catches occurred during early July or late August and' - - - r·~ DRAFT/PAGE 6,4/11/85,4/23/85 3/7/85,3/15/85,3/20/85,4/8/85 NUM3B/Results,4/23/85,5/1/85 most (87%)Deshka River catches were in Mayor September.Thirty-five juvenile longnose suckers were also captured by outmigrant traps at RM 22.4. During 1984,283 longnose suckers were tagged in the lower river and~ y ~,.,/''ItA -f.e-..t ro ne-~ -1-efl~'9Ol4'n"""o..es'D'e-~-emuIT'C1r.1o:r"wa s reca ptu reo/-:..-J'R-"fi ve month s ~~i-t ~Ja:8 ~ ~~.7 miles from where it was tagged in the Deshka River:U: ~n~.:rc~..e (RM 40.6,TRM 1.0).t 3.1.7 Other Species Dolly Varden Thlrty-one Dolly Varden were captured in the Susitna River during 1984 with the highest catch at fishwheel sites (15 fish). Concentrations of Dolly Varden were found at the mouth of the Talkeetna River and some at Kashwitna River (TRM1.2)between April 30 and May 6. After this time,sport fishermen's catches sharply declined indicating that the fish had moved elsewhere.Sport fishermen reported Dolly Varden catches at Clear Creek (TRM 6.0 of Talkeetna River)-increased near May 6.Dolly Varden were the first observed resident species to immigrate the Kashwitna River and Talkeetna River in 1984.At this time,both of the summer glacial rivers were still clear and much shelf ice was present.Mid-day water temperatures at Talkeetna River on May 2 was 3.5°C and on May 8 was 4.5°C,while at Kashwitna River on May 6 was 6.5°C. - DRAFT/PAGE 7,4/11/85,4/23/85 3/7/85,3/15/85,3/20/85,4/8/85 NUM3B/Results,4/23/85,5/1/85 During 1984,five Dolly Varden were tagged in the lower river and none were recovered. Northern pi ke In 1984,three northern pike (Esox lucius Linnaeus)were captured with all three being caught at Flathorn Station (RM 22.4).Two adult fish (2:200 mm)were captured in late August,one by a fishwheel and one by an putmigrant trap.The other fish was a juvenile «200 mm)captured by an outmigrant trap in mid-August. Threespine stickleback . A total of 8,775 threespine stickleback (Gasterosterus aculeatus Linnaeus)were captured during 1984.Outmigrant traps at Flathorn Station captured 7,765 threespine stickleback.The highest seasonal outmigrant trap catches (37.1%)occurred during early August.Juvenile salmon crews (JAHS)captured the remaining fish.At JAHS sites,the maximum catch (915 of 1,010)of threespine stickleback was recorded at -, Beaver Dam Slough (RM 86.3).Over 95 percent of the catch at all sites were young of the year stickleback (20-40rnm). Ninespine stickleback In 1984,50 ninespine sticklebacks (Pungitius pungitius Linnaeus)were captured by a Juvenile Anadromous Habitat Study crew sampl ing at an ..,., DRAFT/PAGE 8,4/11/85,4/23/85 3/7/85,3/15/85,3/20/85,4/8/85 NUM3B/Results,4/23/85,5/1/85 opportunistic site and another ten by outmigrant traps at RM 22.4.Fish caught by the JAHS crew were captured by beach seining at an unnamed upland slough on the west side of the Susitna River at RM 57.2.No lengths were taken but two age classes were observed.One age class was approximately 25 mm,while the other was approximately 50 mm in total length.Table 2 lists the habitat parameters and corresponding measure- ments taken where the sticklebacks were captured. Table 2.Habitat data collected at RM 57.2 where 50 ninespine stickleback were captured,August 5,1984. Water Measurements: Velocity Depth pH Temperature DO Conductivity Substrate Composition Cover Characteristics: Type %cover Arctic lamprey 0.1 fps 3.5 ft 6.6 7.5°C 6.5 mg/l 160 uhos/cm Mud emergent vegetation 96-100% A total of 425 Arctic lamprey (Lampetra japonica Martens)were captured in 1984.A fyke net weir on the Deshka River (RM 40.6,TRM 2.5) captured most Arctic lamprey (336 fish).Five of these fish were adults (310-600 mm)and the remainder juveniles (<.200 mm).Arctic lamprey were caught at the Deshka River from late May to mid-August with most being captured during mid-May (32.7%)or late July (66.9%). DRAFT/PAGE 9,4/11/85,4/23/85 3/7/85,3/15/85,3/20/85,4/8/85 NUM3B/Results,4/23/85,5/1/85 Outmigrant traps at RM 22.4 captured 22 Arctic lamprey.The remaining Arctic lamprey were captured at JAHS sites,with nearly all (55 of 57) the catch being at Birch Creek Slough (RM 88.4),or in the Deshka River by hoop nets. 3.2 Resident Fish Index Site Monitoring on Middle Susitna River A total of 1,409 resident fish were captured by boat electrofishing at 13 index sites in 1984.Seven species of fish were captured:rainbow trout,burbot,Arctic grayling,round whitefish,humpback whitefish, longnose suckers,and Dolly Varden.Most of these fish (56.6%)were captured at tributary sites.Mainstem sites and slough sites accounted for 35.3%and 8.2%of the catch,respectively.Relatively high catches at the combined sites were recorded for rainbow·trout,Arctic grayling, round whitefish,and longnose suckers.Catch data for these four species are presented by site in Appendix Table E-1 and by macrohabitat type in Table 3.Less than 20 fish of each of the other resident fish species were captured at index sites.Because catches of these species were so small,no further catch data of these species will be presented. 3.2.1 Tributary mouth sites Round whitefi sh and Arcti c grayl in9 were captured more frequently at tributary mouth sites (83.5%)in 1984 than rainbow trout or longnose suckers (Table 3).Round whitefish were captured in greatest numbers at Indian River (136 fish)(Appendix Table E-1).Arctic grayling were - )i 1 I J i ]]1 1 J i j Table 3.Boat electroflshlng catch and catch per unit effort (CPUE)of four resident fish species In the middle Susltna River by three macrohabltat types. CPUE 15 In parenthesis,and the units are catch per minute. --------------------------_._-----------~---------------------------------------------~------------------------------------------ >4AV 16-]1 JUN 1-15 JUN' 16-.10 JUL 1-15 JUL 16-.1\ "UG 1-15 "UG 16-]1 SEP 1-15 SEP 16-]0 OCf 1-15 fOfAL SPECIES MACROHABITAT TYPE Tributary Mouths 111\IN80W TROUT 151 .011 31 .081 31 .061 ---1----1 _1.121 ---1----1 21 .061 171 .251 81 .161 11 .0_1 531 .111 ROUND Wlil TEF I S'"631.311 211.701 6911.911 ---1----1 251.741 ---1----1 HII.5_1 _61.611 8311.641 201.7413711.741 ARcnc GRAVLING 6\1 .301 141.361 211 .581 ---1----1 301 .BBI ---1----1 311.931 7511.091 361.711 91 .331 2B]1 .561 LONGNOSE SUCKER 010.001 21 .051 31 .061 ---1----1 311 .911 ---1----1 211 .631 161 .231 21 .04 I II .0_1 161 .151 1391 .661 _611.191 12212.621 ---1----1 9012.651 ---1----1 7212.151 1".>412.231 12912.551 illl.151 18311.551 ~ ~ TOTAL RAIN80W fROUf ROUNO ....1 fEF I SH ARC n C GRAVL I HG LONG"lOSE SUCKER 101 .20 I 61 .121 421 •tiS I 21 .0-1 010.001 411.00 I 21 .50 I II .251 010.001 ---1----1 51 .201 ---1----1 010.001 ---1----1 51 .201 ---1----1 Sloughs II .081 ---1----1 010.001 ---1----1 II •DB I ---I ----I II .061 ---1----1 010.001 31 .21 I 21 .14 I 61 .411 21 .121 51 0301 11 .421 51 .30 I II .141 ---1----1 I-I .111 010.001 ---1----1 231 .181 010.001 ---1----1 541 .421 010.001 ---1----1 201 .161 TOTAL !>011.211 711.751 101 .-11 ---1----1 31 .251 ---1----1 III .761 1911.151 Malnstem II .141 ---1----11111 .871 RAIN80"fROUf 31 .04 I 010.001 010.001 ---1----1 010.001 ---1----1 010.001 51 .091 III .IBI 51 .101 241 .071 ROUND ....1 fEFI SH J61 .531 6511.491 161 .621 ---1----1 31 .1'11 ---1----1 tll.501 6011.091 6411.0BI 251 .4912711 .e41 \ ARC fl C GRAVLI HG 211 .401 201 .461 2411.231 ---1----1 2111.?41 ---1----1 61 .381 401 .731 141 .241 41 .0BI 1561 .471 LONGNOSE SUCKER 010.001 Jl .011 51 .261 ---1----1 61 .351 ---1----1 010.001 91 0161 51 .081 21 .041 301 .091 fOfAL No effort. 661.961 8B12.021 4512.311 ---1----1 3011.151 ---1----1 141 .861 1\412.071 'HII.561 361 .70146711.461 DRAFT/PAGE 10,4/11/85,4/23/85 3/7/85,3/15/85, 3/20/85,4/8/85 NUM3B/Results,4/23/85,5/1/85 captured mostly at Indian River (l04 fish)or Portage Creek (l04 fish). Most rai nbow trout and longnose suckers were captured at Indi an Hi ver - (26 fish)and Jack Long Creek (21 fish),respectively. 3.2.2 Slough Sites grayling (48.7%)were found more often at slough other species:rainbow trout (12.6%),round ('T4 b\c:~) whitefish (20.7%),and longnose suckers (l8.0~).Thirty-eight of 41 "- Arctic grayling captured were caught at Whiskers Creek Slough in late .In I'iS'f) +aele 3 3he.,sA Arctic sites lin 19~than May (Appendix Table E-l). 3.2.3 Mainstem sites - Round whitefish (56.9%)and Arctic grayling (32.0%)were captured more often at mainstem sites than rainbow trout or longnose suckers (Table 3).Most round whitefish (178 fish)were captured between RM 147.3 and RM 147.4,while most (72 fish)Arctic grayling were captured between RM 137.3 and 138.3.Rainbow trout and longnose suckers were both captured in the highest numbers at RM 150.1 where 19 and 34 fish were captured, respectively. - ,~ - -- DRAFT/PAGE 11,4/11/85,4/23/85 3/7/85,3/15/85,3/20/85,4/8/85 NUM3B/Results,4/23/85,5/1/85 3.3 Radio Telemetry Studies of Selected Resident Fish of the Middle Susitna River 3.3.1 Rainbow trout Eighteen of 26 radio tagged rainbow trout tagged in 1983 were monitored during the winter of 1983-84 until their batteries expired.The remain- ing eight fish died or their batteries expired between JulY.and early December 1983.Biological data,tagging data,and monitoring data collected between May and early December 1983 is presented in Suchanek et al.(1984)and Sundet and Wenger (1984). Movements of the 1983 tagged fish from mid-December through late April were minimal with most (13 of 18 fish)moving less than 2.0 miles from where they were found in early December (Figures 4 and 5).The maximum winter movement was shown by rainbow trout 729-1.0 which moved 46.5 mi 1es in 23 days.Because it moved downstream much more rapi dly than the others,it was believed to have died between early January and early February.The remaining four fish moved from 2.8-8.1 miles between mid-December and late April.The maximum upstream movement of radio tagged rainbow trout during this time was 2.8 miles. On January 11 and 12,fourteen of 17 fish whose transmitters were still functioning were located by helicopter in open-water areas of the main- stem Susitna.The open water prevented sampling and biologists were unable to determine the fate of these fish. 600-1.0 606-1.2 607-1.5 x =TAG a RELEASE SITE ••MAJNSTEM SUSJTNA RIVER LOCATION I, \ \. \ \ \ \ \ \ \ \ \ \.----...._------ ---.----...--..------..--....-------626-1.2 140 136 134 132 4th of JULY CR 130 +--........,.-...-.....---r-........-...-.,.-...-.....--,---,--r----r-.,....-,...--,--...,....-r- SEPr5 OCT IS NOV 15 DEC 15 JAN 15 FEB 15 MARI5 APR IS MAY IS 150 PORTAGE C •__0 718-1.5 RECAPTURED AT TRM 3,6 of INDIAN R on 7/6/84 x .TAG a RELEASE SITE Q :MAINSTEM SUSITNA RIVER LOCATION ooTRIBUTARY LOCATION*.NO LOCATIOOS BETWEEN THESE POINTS -----597-1.3:\ 6700 1·'-..,.-.....--,~__,_~-,--.......,--.-_,_~-.--.......,-1---r-'_.,----,---,-_-.--...--,-.•....•.-~729-10 i I JUNI5 JULf5 AUGIS SEPI5 OeTl5 NO\ll5 DEC 15 JAN IS FEBI5 MARI5 APR IS MAY'S JUNIS JUL 15 140 INDIAN R Figure 4.Movement of 12 radio tagged rainbow trout in the Susitna River below Devil Canyon,June 1983 to May 1984. Q :J i >}}I·•]t »~l I>;i 150 767-1.0 x =TAG 8 RELEASE SITE •=MAINSTEM SUSITNA RIVER LOCATION 0=TRIBUTARY LOCATION *=NO LOCATIONS BETWEEN THESE POINTS ,-749-1.0, I I, ------..----.--------..---.------.-------~639-1.0 \ \ \ \. '"....., ~FISH FOUND DEAD ,Q-O···••··O---..ON BANK AT TRM a 1 i 670-1.4 RECAPTURED J OF LITTLE PORTAGE'CR (608-1.7) l._TAG REPLACED TO 6ei8-1.7~j ·......0.......e : '-e-*~..-.-.--.-........-.-.-...-.-.-.-._._.,'.:•It ..729-1.3 SEP 15 OCT 15 NOV 15 DEC 15 JAN 15 FEB 15 MAR 15 APR 15 MAY 15 JUN 15 JUL 15 AUG 15 SEP 15 OCT 15 670-1.4 TAGGED 6/7/83 /AT TRM 0.2 WHISKERS CR /'..*o~··· ··".....•. :-lv...-"'---.- JACK LONG CR W -.J ::e ~f JULY .CR a::130 W> a:: ~I «-z.... U> ~ (/) LANE CR- 110 Figure 5.Movement of five radio tagged rainbow trout in the Susitna River below Devil Canyon, September 1983 to October 1984. DRAFT/PAGE 12,4/11/85,4/23/85 3/7/85,3/15/85, 3/20/85,4/8/85 NUM3B/Results,4/23/85,5/1/85 The other three rainbow trout were initially located by helicopter then pinpointed during ground surveys in areas of the mainstem covered by ice.Rainbow trout 670-1.4 was pinpointed at RM 101.1,this fish had stayed within a 0.5 mile radius during 1983-84 monitoring.It was in 2.5 feet of slush free water with an estimated velocity of 1 to 2 feet per second (Appendix Table B-1).Rainbow trout 767-1.5 was pinpointed at RM 114.8,it had ranged 1.2 miles from its tagging location.Rainbow trout 767-1.5 was located in one foot of water with six inches of slush ice over the water.The fate of both of these fish (670-1.4 and 767-1.5)was not determined since no movement was detected after ice augering in the areas where their transmitted signals were the strong- est.The fate of the remaining rainbow trout (597-1.3)located during the ground survey was not determined.This was due to a peculiar null from its signal which prevented pinpointing the fish. Surveys on February 15-17 found 15 fi sh whose transmitters were sti 11 functioni ng.A11 but five fi sh were found in areas of open water. Three (670-1.4,709-1.5,and 729-1.3)of the five fish found under ice cover moved from 10-200 feet after drilling with an ice auger over them. These three fish were found in water depths averaging 6.2 feet and flows averaging 2.2 fps.Appendix Table B-1 lists habitat measurements taken at the three locations. The other two fish found during ground surveys in mid-February were believed to be dead.No movement was detected for either rainbow trout 767-1.5 or 729-1.0.Rainbow trout 767-1.5 was found in only one foot of - .- DRAFT/PAGE 13,4/11/85,4/23/85 3/7/85,3/15/85,3/20/85,4/8/85 NUM38/Results,4/23/85,5/1/85 water and 729-1.0 in 1.5 feet of water.Water velocities near these two fish were less than 0.5 fps. All ten fish found in open water during mid-February were in areas with no anchor ice.Three of these fish were in pools and the others were in riffles. Six 1983 radio tagged rainbow trout were found after April but batteries from transmitters of three of these fish expired by mid-May.The remaining three fish (639-1.0,749-1.0 and 767-1.0)were monitored until August or September 1984 (Figure 5).One of these fish (767-1.0)moved into Fourth of July Creek on May 24,then moved up and into Indian River in early July (Figure 6 and 7).The other two fish al so moved into Indian River during the summer (Figure 7). Two rainbow trout tagged in 1983 were recovered in 1984.Transmitters in both fish had been dead for over one month.Rainbow trout 718-1.5 was recaptured at TRM 3.6 of Indian River by a sport fisherman on July 6 and rainbow trout 670-1.4 was recaptured at Lane Creek on May 18 by a boat electrofishing crew.At this time,the transmitter was replaced with tag 608-1.7.On May 31,the fish was located at the mouth of Little Portage Creek (RM 119.0).On June 27 the tag was found along the bank at TRM 0.1 of the creek and the fi sh was suspected to have been eaten by a predator. In 1984,23 rainbow trout were radio tagged in the middle river between May 17 and July 23 with 82.6%being tagged by June 10.Fourteen of the 0.0 I 1'\1 ~,,'0 i 0 t I'Of' ~ fOi·Z.4.PR£.SPlI,WN£R fll-1-6 ,,[J.U&~a CONOITION fn.l.~.UNKNOWN HO·I ".NON·""&WN[III Uf.l.l,....[."'&WNl" ••TAG.R£LEAS£S,tt ••"'''INSltM IUS""'1111\'[111 LOCATION 0·TIII18O'&'"LOC&'ION "..-.--.....-."'·1,1 .'I I I I I ~,..~.~,................._UO-I.4 .,\'oJ :1 ~?Ot ..2.• ..\to ..,:;?llt.'.t I''/::I:' I : I \I:.,...•I:; !I ,'''! IIQ 1...,\j '\•i \,:\-.4' l '~ 'lA...o,S II(W __("."'.'::1'21.1.0 10 ~i ••••e ".:;.,JJ~ot~~.,.0..0····.0·..0,~~ CHUllTN&\••••u; 1II111[ft 0"0'''':0'o Movements of five radio tagged rainbow trout in the Susitna River,May to December 1984. UQ~~~~~~I~St'~~le~~D£CIS WHISI([fIS CR_ 100 CHUlITfr4&" '.~ '"0 .I.t,(;,K LONG Cit,.. ::11. I- iii::> 'II 1$1: 4'''""Ul'l'CJlI Figure 8. IfrO&"ft ....I"o! :II '"wilt >.. -------------------- ' e.t.lA!NSTEM SUSITNA IlIVER LOCATION o 'INOIAN RIVER LOCATiON ......639·1.0,TAGGED 9/2163 -,-,709-2.4 ---140·IA ••-••749-1.0,TAGGED 9/1/8) --757-1.1 ...........761-1.0,TAGGEO 10/4/83 O''0·'00 ,····,·§······i·········, 00 •..,,.....~ :b' Movement of six radio tagged rainbow trout in Indian River, May to October 1984. 1.0 5.0 3.0 4.0 6.0 2.0 9·f"i~"~ :/"I;.",i if \i:/0':.:\~\\1 :~\i:j;\j -/.:"lJ ~ 0.0 , I ,.I i "'j e'1~i I I i MAY I~JUN 15 JUL I~AUG I~SEP 15 OCT 15 NOV 15 0: W ;> 0: Figure 7. z« Ciz .... .J ::i SEP 15 ••TAG a RELEASE SITE O'SPAv/NING FISH LOCATION O.MAINSTEM SUSITNA RIVEIl OIl SID(TRIBUTARY FISH LOCATION AT CAPTURE .-.-.-709-i.•,PIlE-SPAWNEIl --nO-i.•,PRE-SPAWNEIl ........nO-i,4,PilE-SPAWNER -----767·1.0,SEXUAL CONDITION UNKNOWN,TAGGED 10/./8) AUG 15JUL15JUN10MAY10 Figure 6.Movement of four radio tagged rainbow trout in Fourth of July Creek,May to October 1984. UPSYR[AM FISH 8ARRIEflc!!!'....TEflFALLl 1.8 1.6 ~I.. i 0::I,iw> ii 1<:I.W W 0:: 0 ).0.8oJ ~UNNAMED..."TRI8UTAflY ~I IL 0 0.6 :r l-e: ~ 0 0 .• IL O.i l-I i ,"J l ~,~;,.'" DRAFT/PAGE 14,4/11/85,4/23/85 3/7/85,3/15/85,3/20/85,4/8/85 NUM3B/Results,4/23/85,5/1/85 fish radio tagged were pre-spawners.All but two of the fish were ) captured in /tributaries or at tributary mouths with most being in Fourth of July Creek (9),Portage Creek (8),and Indian River (4).No tracking data is provided for one fish tagged at Fourth of July Creek because it apparently died after tagging.Appendix Table B-2 presents a summary of capture and biological data for the other 22 radio tagged fish.Indi- vidual fish movements of the 22 radio tagged rainbow trout from time of tagging to November 30,1984 are presented in Figures 6 to 11.Most (21 of 25)radio tagged fish monitored during the summer appear in two or more figures since they moved in the mainstem as well as in tributaries such as Portage Creek or between tributaries.Most (I6)of the 1984 tagged fish showed large summer movements between 5.0-15.0 miles from where they were tagged.All but one fi sh showed an upstream movement over 1.0 mile.The maximum movement was shown by rainbow trout 598-1.6 which moved 101.2 miles (Figure 8).Because it moved so much and so rapidly after August 13,it was believed to have died close to this time.Another fish (670-1.2)was believed to have died because of spawning (Figure 9). Only one of the 25 monitored radio tagged rainbow trout was not found in a tributary during the summer of 1984.Ten fish ascended Portage Creek, six ascended Fourth of July Creek,four ascended Indian River,two ascended both Fourth of July Creek and Indian River,one fish ascended Little Portage Creek,and one ascended Whiskers Creek.The fish which moved into Whiskers Creek,later moved into the Chulitna River (Figure 8).In 1984,434 locations of 39 different radio tagged rainbow .....613·10 '.",,'.,"\I',\ ,I I \,..,. I \1 1 I \1 , :~"T+,T7"10-2.4 .., : \ i .. I :"...i:'I!.749-10 . •'f ':I.~.!o~.-,·I-l-"-599·11.-.T'--................•..•.•. I I I I I I I I I I I I I I I I I .j? I I I ,.670.1.2:('..-..-~.-,/ \/ \/ LI ••TAG 8 RELEASE SITE ••MAINSTEM SUSITNA RIVER LOCATION O.TRIBUTARY LOCATION MAY 15 JUN 15 JUL 15 AUG 15 SEP 15 OCT 15 NOV 15 DEC 15 - 136 147 137 • 148- 149·· 150-' W ..J ~ a:w > a: «z.... lI) ::> lI) POHTAGE CR-l OCT 15SEP15 r, --599-1.1 ",•.•613-1.3 670-1.2 749-1.8 710·24 AUG 15 ••TAG 8 RELEASE SITE ••MAINSTEM SUSITNA RIVER LOCATION o.PORTAGE CREEK LOCATION JUL 15 9,,'.,",'.,T:? ,I I • ,I,.,.,\ I • ::"P... P':'C('-,..(j''''.....,,'U 0':"',b.\:,t ~....:,I','.\ !f \'~ I ":'~",:\ ,,''\I:'·•,,',.,:":\" 0 ........(\a."(."~./"'\Cl 'I\:i",:' JUN 15 "~!Ill) "", ~--~ 613-1.3 LOCATED \ 0.1 MILE UP OUTLET OF LAKE \ ~ MAY 15 o I I Ii'.iii I.I *1 •i 8 4 14 AT FORKS 12 10 6 UNNAMED OUTLET FROM A LAKE UNNAMED OUTLET FROM A LAKE 2 MOUTH OF THOROUGHFARE CR W ..J- :::E a: w I ~~0:: ~::.:: w wa:u w 19«.... 0:: 0a.. Figure 9.Movement of five spawning rainbow trout,tagged at TRM 2.3 Portage Creek,in Portage Creek and then in the mainstem Susitna River,May to December 1984.At capture,all five fish were pre-spawners. I -J }j l )1 1 )1 -~l 1 J f J J -) ••TAG 8 RELEASE SITE ••MAINSTEM SUSITNA RIVER LOCATION O'TRIBUTARY LOCATION .·NO LOCATIONS BETWEEN THESE POINTSI, ~,...-.......... ,J ..",~l "•.620-1.2 I I I, I, I,, I, lr·_·_·.."...709-1.2 728-1.0 (RECAPTURED Ii ACCIDENTLY___.~9-1.2 RELEASED AT RM 138.6 ~(PROBABLY DEAD THE SAME DAY)"'~AFTER JULY) ~'~~--O-L-Q--".I 'r ."I,'~630-1.7 . I \'I .t. I \ !t ~ !I \ I I ~ j I ! ·\I I . · I \!I \ I I ~·\ !I "I I \·; *f1 i II i : i \.'·726-1.0 \ \ \ 140 142 144 1:36 146 134 INDIAN R 138 150 W ..J ::::li PORTAGE CR 148 0:: W > a:: «z l- I/) ::> I/) --D 659-1.2 (PROBABLY DEAD AFTER JULY) ••TAG 6 RELEASE SITE O'SPAWNING FISH LOCATION ••NON-SPAWNING OR UNKNOWN If SPAWNING FISH LOCATION O'MAINSTEM SUSITNA RIVER LOCATION AT CAPTURE -----620-1.2.NON-SPAWNER --630-1.7.PRE-SPAWNER - -659-1.2,POST-SPAWNER -.-.709-1.2,PRE-SPAWNER .........728-1.0,NON-SPAWNER ••••••740-2.3.PRE-SPAWNER ~. j \ i \ . \.q!'cr ~ I \ i \ i " i "i \I '.\ 9.6• :9_,D.-a.._ \ 740-2.3 Q'.!o.o.;-:""•. ARCTIC ~GRAYLINGi!f ~~ • I .~.',I •.,:p:i 1 :1*, t4 ....~i~.ri:".~.: 01 ' •..I •.::-0:.~:1 ~\.:~~1~:/.:..\.:,. . d \:0 . :i ~ :j \ :;p-oo --0-i ~ ~~'Q--0-1-:0- I ~4 J _--e--___~'~o 'i I 0 --I i 1 .....-0'IOi I 6 4 B 10 12 14 AT FORKS UNNAMED OUTLET FROM A LAKE 2 UNNAMED OUTLET FROM A LAKE MOUTH Of THOROUGHFARE CR W ..J :I 0:: ~I w>-..J 0:: ~ W W 0:: U W C) ~ I-a:: 0 Q. MAY 15 JUNI5 JUL 15 AUG 15 SEP 15 OCT 15 MAY 15 JUN 15 JUL 15 AUG 15 SEP 15 OCT 15 NOV 15 DEC 15 Figure 10.Movement of five rainbow trout and one Arctic grayling in Portage Creek and then in the mainstem Susitna River,May to December 1984. ,......-..-~ \ \ \ \ \ \ \ \ 10630·1.0 x ·TAG 8 RELEASE SITE •'MAINSTEM SUSITNA RIVER LOCATION o •TRIBUTARY LOCATION 598-1.6 !'..l a •• I :1':·...·...608-1.9 \: I \ I 'I \ I \,, \ ~ 32 30 134 132 4th 01 JULY CR 130 128w .J :E a::126 w > a: «124 Z ~ III :::l 122III 34 AI CAPTURE --598-1.6,PRE·SPAWNER ........608-1.9,POST -SPAWNER --.--630-1.0,NON-SPAWNER -.-.760·2.4.POST-SPAWNER.,,'\' /1"\/~ I \" I 1\ I',\ I.:~ :~l, I I 1, '1 I I x •TAG 8 RELEASE SITE o •SPAWNING FISH LOCATION ••NON-SPAWNING FISH LOCATION o •MAlN5TEM SUSITNA RIVER LOCATION \ \ \. \I ;Qx....l.'J·.......~ i I ". i i " i i i i i i 598-1.6 RADlO-TAGGED MMILE ""\""UTA:' 1.4 1.6 1.2 I. 0.6 0.2 0.4 UPSTREAM FISH BARRIER Ql5'WATERFALL) ~1.8 W .J i a:: w ~a:: ~ ~I w w ~ a::u >-.J ::> ""'3 IJ,. 0 J: ~a:: ::> 0 IJ,. 0.0 Iii I 0;'0...jt O i I b i -·b.i MAY 15 JUN 15 JUL 15 AUG 15 SEP 15 MAY 15 JUN 15 JUL 15 AUG 15 SEP 15 OCT 15 NOV 15 DEC 15 Figure 11.Movement of four radio tagged rainbow trout in Fourth of July Creek and then into the mainstem Sus1tna River,May to December 1984. J ~._,..."J J 1 )'c .•1 I I t •I - - - ..- - DRAFT/PAGE 15,4/11/85,4/23/85 3/7/85,3/15/85, 3/20/85,4/8/85 NUM3B/Results,4/23/85,5/1/85 trout were made.Movements of the 1984 monitored radio tagged rainbow trout can be pl aced into four major categori es based on thei r annual life history:(1)those associated with overwintering (December - April),(2)those associated with spawning (May and June),(3)those associated with summer rearing (July -Septemb~r),and (4)those associ- ated with the transition period between fall and w"inter (October and November).The distribution of 1984 monitored fish changed by macro- habitat as the season changed.Between December and April,nearly all (90.7%)rainbow trout were found in the mainstem (Figure 12).During May and June,62.2 percent of the radio tagged rainbow trout locations were in tributaries.The majority (58.9%)of fish locations from July to September were also in tributaries. By October 6,all but one radio tagged rainbow trout had outmigrated tributaries to their mouths or the mainstem.The one fish (659-1.2) that remained in a tributary after September was believed dead since it had not moved recently (Fi gure 9).In October and November,72.1%of the fish locations were associated with the mainstem with the remaining locations being at tributary mouths. 3.3.2 Burbot Two sampling trips made in mid-January and mid-February 1984 to locate and determine the fate of burbot radio tagged in summer 1983 yielded limited data.In January,only one burbot was pinpointed during ground surveys.Two more radio tagged burbot were located in open-water areas - - OVERWINTERING,FOR 1983 TAGGED FISH (DECEMBER -APRIL)n.75 IN SLOUG) 100 >~80 ILl=>~60 a:: lL ...4 Z ILl ~ZO ILl ll. >80 SPAWNING ~(MAY -JUNE) ~60owff:4... ~20 (,) a:: ~0 n -135 >8 SUMMER REARING ~(JULY -SEPTEMBER) ~o ILla:: IL... ~2 (,) II:: ILl 0.. n =163 LOCATIONS IN MAINSTEM AT TRIBUTARY MOUTHS TRANSITION PERIOD BETWEEN FALL AND WINTER (OCTOBER -NOVEMBER)n.~I LOCATIONS IN TRIBUTARIES OR SLOUGHS >80 ~ ILl :Jo LaJg:4 I- ~2 Ua:: ~ Figure 12.Frequency distribution of radio tagged rainbow trout locations in tributaries, at tributary mouths,and in the mainstem Susitna River during 1984. of the mainstem during January. still functioning. DRAFT/PAGE 16,4/11/85,4/23/85 3/7/85,3/15/85, 3/20/85,4/8/85 NUM3B/Results,4/23/85,5/1/85 In February,only one radio tag was - Burbot 670-3 was located in the mainstem at RM 87.0 in January.Biolo- gists measured six inches of water with zero velocity and two feet of slush ice between the water and the ice at this location.This burbot did not move when ice augering was done in the vicinity of its strongest radio tag signal and therefore its fate was not determined.Two burbot sets made in the area overnight failed to catch any fish.Both this fish and the other two fish (639-3.0 and 720-3.0)located in mid-January were found to have moved downriver less than 1.0 mile from where they were found on December 1,1983.Burbot 639-3.0 was located at RM 131.1 in waters approximately 4.0 feet deep and velocities of 1-2 fps. Habitat measurements taken near burbot 639-3.0 showed a water tempera- ture of 0.2°C,conductivity of 247 umhos,pH of 7.9,and DO of 13.6 mg/1.In February,burbot 720-3.0 was found 0.2 miles downriver where it was pinpointed in January in an open-water channel off Slough 10 (RM 133.8) in water approximately 4.0 feet deep.Burbot 720-3.0 was last found on March 13,at RM 133.7. 3.3.3 Arctic grayling One radio tagged Arctic grayling was monitored between May and September 1984.This fish was captured by boat electrofishing,tagged and re- 1eased at RM 150.1 on May 22.At capture,it was found to be a pre- spawning male,410 mm (FL)long and 10 years old.Two days after DRAFT/PAGE 17,4/11/85,4/23/85 3/7/85,3/15/85,3/20/85,4/8/85 NUM3B/Results,4/23/85,5/1/85 tagging,it moved into Portage Creek (RM 148.8)for the summer (Figure 10).The maximum recorded upstream location of this fish was TRM 8.7 on August 6.Shortly thereafter,the fish began to move out of Portage Creek.This fish was last found on September 6 at TRM 3.8/of Portage - Creek. 3.4 Other Resident Fish Studies on the Middle River Tab 1e 4 provi des catch data of res i dent fi sh captured in the mi ddl e river in 1984 at sites other than the 13 boat electrofishing index sites.Population estimates were made using provisional data;numbers used and population estimates made are provided in Appendix D. Four sites were located in 1984 where rainbow trout probably spawned. These sites were at Portage Creek TRM 2.3 and TRM 5.1,at Fourth of July Creek TRM 0.7 and at TRM 0.5 of an unnamed sid~tributary outletting at TRM 0.7 of Fourth of July Creek.Appendix Table E-2 provides habitat measurements taken at these sites as well as general comments about each site. 3.4.1 Lake surveys A total of 390 resident fish were captured in six lakes surveyed in 1984 (Table 4).Most (86.1%)fish captured were rainbow trout.Lake B was ~ the only lake surveyed where no rainbow trout were captured.Effort was similar in all lakes except at the lake outletting at TRM 2.3 of Portage J )J J 1 J J 1 1 I 1 J DRAFT/PAGE 18,4/11/85,4/23/85 3/7/85,3/15/85,3/20/85,4/8/85 NUM3B/Results,4/23/85,5/1/85 Creek.Much less effort was expended in this lake compared to the other five lakes. 3.4.2 Tag-and-recapture studies Rainbow trout In 1984,153 ra i nbow trout were Floy anchor tagged and 30 recoveri es were made from rainbow trout tagged in the middle river.Seven recoveri es were made at the tagging sites and 15 rainbow trout were recovered within 5.0 miles of their tagging sites.The remaining eight fish moved an average of 27.4 miles from where they were tagged.Most (63%)recoveries were from fish tagged in 1983.Eighty-seven percent of the recoveries were made in or at the mouths of tributaries such as Fourth of July Creek (14,RM 131.1),Indian River (6,RM 138.6),and Portage Creek (3,RM 148.8).·The longest move recorded for a Floy anchor tagged rainbow trout in the Susitna River is 55.7 miles.This fish was tagged in Fourth of July Creek in 1983 and recovered in an unnamed tributary of the Chulitna River (TRM 23.1)during 1984. Burbot In 1984,six burbot were Floy anchor tagged in the middle Susitna River and there were no recaptures. -! - -, - - - DRAFT/PAGE 19,4/11/85,4/23/85 3/7/85,3/15/85,3/20/85,4/8/85 NUM3B/Results,4/23/85,5/1/85 Arctic grayling During 1984,425 Arctic grayl ing were tagged and 44 recoveries of 43 different fish were made from fish tagged in the middle river.This included one fish tagged at Cheechako Creek (RM 152.4)in August 1982 and recovered in May 1984 at Portage Creek (RM 148.8).The 44 recaptured fish ranged 0.0 to 95.8 miles from the time they were tagged. Eight fish were recaptured at their tagging sites.Twenty-one Arctic grayling recaptured were caught within 5.0 miles of their tagging site. The remaining 15 fish moved an average of 20.0 miles.The maximum movement for an Arctic grayling to date is 95.8 miles.This fish was tagged in Portage Creek (RM 148.8,TRM 6.0)in 1983 and recovered in Kashwitna River (RM 61.0,TRM 2.0)in 1984.Most (61.4%)recoveries were made from fish tagged in 1983.Twenty-one of the recoveries were made in tributaries or at tributary mouths with another 17 recaptured in the mainstem. In addition to these recoveries,one Arctic grayling was recaptured at Portage Creek which had been tagged at Tsusena Creek (RM 181.3)in 1982. Round whitefish In 1984,481 round whitefish were tagged and 76 recoveries of 72 differ- ent fish were made from fish tagged in the middle river.Most of the recoveries were made near the tagging sites with 25 at the tagging sites and 35 recoveries made within 5.0 miles of the tagging sites.The remaining 16 fish moved an average of 16.0 miles with a maximum movement -4-~ DRAFT/PAGE 20,4/11/85,4/23/85 3/7/85,3/15/85,3/20/85,4/8/85 NUM3B/Results,4/23/85,5/1/85 of 55.7 miles.Sixty-one percent of the recoveries were made from fish tagged in 1983.Forty-two of the recoveries were made at tributary _ mouths and 32 recoveries were made in the mainstem. Humpback whitefish In 1984,a total of 25 humpback whitefish were tagged in the middle Susitna River.No humpback whitefish were recaptured in 1984. Longnose suckers - A total of 158 longnose suckers were tagged in 1984.Thirteen longnose suckers were recaptured in 1984 of which all were tagged in the middle river.Four fish were recaptured at their tagging sites,another four were recaptured within five miles of their tagging sites,and the remaining five fish moved an average of 16.5 miles.Most of the fish recaptured were tagged in 1982 (five fish)or in 1983 (six fish). Dolly Varden -, During 1984,eight Dolly Varden were tagged in the middle river.Three recoveries of two different fish were made in 1984 with both fish being tagged in 1984.One fish was tagged at RM 139.4 and recaptured one month later at RM 136.7 and again three months later at Indian River (RM 138.6).The other fish was tagged and recaptured,12 days later,at Indian River. - ......, - DRAFT/PAGE 1,4/8/854/11/85, 3/10/85,3/17/85, 3/20/85,4/23/85 NUM3/Discussion,4/30/85,5/1/85 4.0 DISCUSSION 4.1 Lower Susitna River 4.1.1 Rainbow trout .e.-j./~ Studies done in 1984 show rainbow trout immigrate lower river tributaries up to 10 days earlier than middle river tributaries.Water temperatures taken at Deshka River and Kashwitna River in 1984 shows the immigration occurs at similar temperatures (6.0°C and 8.2°C, respectively)as Fourth of July Creek (7.7°C)in the middle river. Catch data from 1981 also shows rainbow trout immigrate lower river tributaries early in May (ADF&G 1981b). After rearing in tributaries during summer,rainbow trout are believed to move out of most east side tributaries in September or October and overwinter in the mainstem Susitna River.The fall outmigration is probably triggered by discharge and/or temperature.In 1984,the primary cause of the outmigration was probably a flood in late August (refer to pa rt 4.2.1 and Fi gu re 14).Su rveys conducted in the upper reaches of several of these tributaries in .early September found very few rai nbow trout.Sport fi shermen have reported summer rainbow trout populations in these tribu~aries such as the North Fork of the Kashwitna River and Montana Creek to be high (Dave Watsjold,pers.comm.). Radio tagging data from winter 1981-82 support the hypothesis that some rainbow trout overwinter in the lower mainstem Susitna River (ADF&G DRAFT/PAGE 2,4/8/854/11/85, 3/10/85,3/17/85,3/20/85,4/23/8£ NUM3/Discussion,4/30/85,5/1/85 -1983b).Recoveries of Floy anchor tagged fish show some Talkeetna River rainbow trout also overwinter in the mainstem Susitna River (Sundet and Wenger 1984).Because few adult rainbow trout were captured in early spring or late fall in the Deshka River in 1981 or 1984,adult rainbow trout may overwinter in the upper reaches of the Deshka River. Likewise,because catch rates were relatively high for juvenile rainbow trout during these times on the Deshka River,it appears some upper Deshka River juveniles outmigrate to the mainstem or the lower Deshka River for overwintering. 4.1.2 Burbot Susitna River burbot reside mostly in mainstem influenced areas (ADF&G 1981b, 1983b,1983c,1983d).This relationship is probably due to high turbidities or low light penetration in the mainstem which the light sensitive burbot prefers (ADF&G 1983e;Suchanek et ale 1984).Studies (1981-84)have shown that burbot are found in much higher concentrations below Talkeetna than above it (ADF&G 1981b,1983b).This is probably due to a greater frequency of preferred habitat.Captured burbot and radio tagged burbot have been found most often in backwater areas of varying depths (ADF&G 1983c, 1983e,Suchanek et ale 1984). Our data indicates that burbot readily use mainstem influenced areas in the lower river for spawning.Capture rates have been reported by sport fishermen to be high at the mouths of the Deshka River (RM 40.6)and Alexander River (RM 10.1)during mid-December to early February (ADF&G 1983b,1983c).Radio tagged burbot data has also shown that all - - - the mainstem Susitna. - - ..... - DRAFT/PAGE 3,4/8/854/11/85, 3/10/85,3/17/85,3/20/85,4/23/85 NUM3/Discussion,4/30/85,5/1/85 11 burbot monitored duri ng January and February 1982 and 1983 have remained in the mainstem between RM's 26.0 to 89.6.Since approximately 85%of burbot over 400 rrm total 1ength a re spawners for a given yea r (ADF&G 1983c),it is likely several of the radio tagged fish spawned in , r~~~ Burbot spawning in mainstem influenced areas is of particular impor~;rpJ tance °O!'e_£,,:,j;=proJ_ect effect of h~'!roe]_"Et:_i~_d"I~L~LJ1Q.W_~peak i ng ~: wh_!:~,.,._~~R~.~~t !.~._~~.~.L!~.!~S wa!er levels.Because burbot are demersal ~ spawners,this effect could substantially decrease burbot populations in ~..'•..~A ..•._ mainstem influenced areas of the Susitna River by desiccating eggs.~~ Eggs of other species such as salmon could possible survive this effect ~ if redds a.re .....in...a..r.eas of grOUn,dwater o ..'.....l / ......,'---__,~....[k~'~Jk.~~*~l Jv-~ Burboit·---'most 1i kel,Y}use the Deshka River and Al exander River for \"",_"",..,,_,.__~_,.~,_~""-"-~~C" spawn i ng.The pre-spawni n9 movement into these tri butari es has been reported to begin in November,but from studies done in 1984 this movement appears to begin slightly earlier (ADF&G 1981b;1983b).During 1984,burbot were captured by hoop nets between TRM's 0.0-6.0 on the Deshka River with catch rates increasing substantially from mi d-September to mi d-October.The hi ghest burbot catch rates duri ng May,September and October 1984 sampl i ng were between October 11-15 which were the last days of the open-water field season.Local resi- dents on the Deshka River reported that slush ice flowed from October 16 to November 11 and on November 11 the river froze over (Leon Dick,pers. comm.). DRAFT/PAGE 4,4/8/854/11/85, 3/10/85, 3/17/85, 3/20/85,4/23/85 NUM3/Discussion,4/30/85,5/1/85 Whil e few burbot were captured at the farthest upriver Deshka River sites (between TRWs 5.0 to 6.0)in early September,by mid-September the catch and catch rates were much higher.This also suggests that Susitna burbot may spawn in the Deshka River further upriver than thought.Data from one burbot tagged and recovered in 1984 and catch data from 1981 also supports this this hypothesis.The recaptured fish was tagged at TRM 1.5 on September 13 and recovered at TRM 6.0 on October 13.Catch data in 1981 at TRM 4.5 (site C)showed no adult burbot until .late August (AOF&G 1981b).Between then and mid-October catch rates increased. Young-of-the-year burbot have been seldom captured;however,·during mid-June 1984 several thousand approximately 15 mm (TL)burbot were observed-along the shoals of the Deshka River at TRM 1.9.A similar timing of hatching was reported in 1982 at Slough 9 (RM 129.2)where several dozen of the same size fish were captured (ADF&G 1983b). 4.1.3 Arctic grayling Arctic grayling spring movements into lower river tributaries usually begins in early Mayor up to 20 days earlier than in most middle river tributaries such as Portage Creek.The earlier immigration in tributaries below RM 98.5 than above is probably due to warmer water temperatures.Daytime surface water temperatures in 1984 was 8.2°C at Kashwitna River on May 10 and 0.8°C at Portage Creek on May 9. - - - DRAFT/PAGE.5,4/8/854/11/85, 3/10/85,3/17/85,3/20/85,4/23/85 NUM3/Discussion,4/30/85,5/1/85 After rearing in tributaries during the summer,Arctic grayling apparently move out of most east side tributaries into the mainstem to overwinter.In larger west side tributaries such as the Deshka River, most adult Arctic grayling are suspected to overwinter in the upper reaches of those tributaries.Catch rates were only slightly higher in early spring and late fall compared to other times at the Deshka River (TRM's 1.5-6.0)in both 1981 and 1984.Two recaptures in 1981 indicate that some Arctic grayling in the lower river may overwinter far downriver in the mainstem from their summer rearing tributaries.These fish were tagged in May 1981 and were later recovered 9.9 and 32.5 miles upriver (ADF&G 1981b). 4.1.4 Round whitefish Boat electrofishing in the lower river was resumed in 1984.Data collected in this area showed round whitefish distribution in 1984 was similar to 1982 findings (ADF&G 1983b).Because a number of sexually ripe round whitefish have been found in the mainstem in October,round whitefish are thought to use this habitat for spawning at this time. 4.1.5 Humpback whitefish Two stocks of humpback whitefish appear to be in the Susitna River below Devil Canyon.One stock is anadromous while the other remains in the river all year.Scale analysis of fish captured in 1983 at Yentna River fishwheels showed that many (l9.2%)exhibited periods of high growth compared to most (4.9%)fish captured above the Chulitna River DRAFT/PAGE 6,4/8/854/11/85, 3/10/85,3/17/85,3/20/85,4/23/8£ NUM3/Discussion,4/30/85,5/1/85 confluence.This suggests that a high percentage of humpback whitefish in the lower river overwinter in the estuary.Relatively high juvenile humpback whitefish catches at outmigrant traps at RM 22.4 in 1984 also lends support to this hypothesis. Fishwheel catch data shows adult humpback whitefish begin their spawning run in June and it continues through September.With the addition of fishwheels at Flathorn Station {RM 22.4}in 1984 more knowledge was gathered on the timing and behavior of humpback whitefish during the spawning migration.As the season progressed,humpback whitefish catches similiarly progressed up the river.Catches were high at Flathorn station from early July to late August,at Yentna Station (RM 28.5,TRM 4.0)from early July to early September,and at Sunshine Station (R~l 79.0)in late August and early September.During this time,fishwheel effort was approximately the same at all stations except during spring. Compari son of catches between the three si tes suggests many humpback whitefish migrate up the Yentna River and into areas between these stations to spawn.One of the suspected spawning areas between stations is Anderson Creek (RM 23.8).Large numbers of humpback whitefish were gill netted in this tributary in 1981 (ADF&G 1981b). Tag-and-recapture data also support evidence of the spawning run orig- inating from or near the estuary.In 1984,two fish were recovered at Yentna Station in late August and early September which had been tagged two and 30 days earlier at Flathorn Station. - - """" - ,~ ..- - ..... DRAFT/PAGE 7,4/8/854/11/85, 3/10/85,3/17/85, 3/20/85,4/23/85 NUM3/Discussion,4/30/85,5/1/85 Exact spawning locations for humpback whitefish have not been found,but it is highly suspected that they neaxly--_excJusjY~lysp~wrLjJ:LtLtb.~,-----_..,..-.----..-."_.,"- /tarle._~.Support for th is hypothes i sis provi ded by the few numbers of humpback whitefish captured in mainstem influenced areas in either 1982 or 1984.Large numbers of pre-spawners have been captured,however,at Anderson Creek.Spawning is presumed to occur from mid to late October. 4.1.6 Longnose Suckers Longnose suckers.are distributed widely throughout the Susitna River. Boat electrofishing catch data from 1982 and 1984 show they are the most abundant resident fish species (except sculpins and sticklebacks)in the lower river (ADF&G 1983b). Recapture data from 1981 to 1984 suggests most adult longnose suckers move little in the summer.Only one of 12 suckers recaptured from 1981 to 1984 moved over 5.0 miles from their tagging sites (ADF&G 1981b, 1983b). Late May spawning occurs for longnose suckers in tributaries as well as in the mainstem Susitna River (ADF&G 1983b).Although Morrow (l980) reports longnose suckers are only spring spawners,high numbers of male and several female pre-spawners have been captured in the mainstem Susitn~River during September and October.These fish have been captured throughout the lower river above RM 35.4 but are most numerous between RM 35.4 and RM 60.0.These data suggest that there may be two DRAFT/PAGE 8,4/8/854/11/85, 3/10/85,3/17/85,3/20/85,4/23/8~ NUM3/Discussion,4/30/85,5/1/85 spawning periods of longnose suckers in the Susitna River with one in late May and the other in October or November.Because no practical ~ sampling occurs during freeze-up (between mid-October to December)this hypothesis will be difficult to prove. 4.1.7 Other Species - Dolly Varden Dolly Varden are widely distributed in the Susitna River but few have been captured.They have been captured most frequently near the mouths of the Kashwitna River and Talkeetna River.- In tributaries which have sizable populations of Dolly Varden,studies done in 1984 show that this species is the first resident species to immigrate from the mainstem during spring.Limited data from 1984 suggests that this migration occurs at colder temperatures than for - other speci es • Northern pi ke Northern pike are scarce in the Susitna River with only five fish being captured since 1981.1 Since all of the fish have been captured between RM's 22.4 and 36.3,including Yentna fishwheels (RM 28.5,TRM 4.0),it is most probable that these fish have emigrated from the Yentna River 1 The reference to the Susitna River includes only where fish have been collected for the past four years.Studies have been con- ducted at TRM 4.0 Yentna River but no further upriver than that. )t - DRAFT/PAGE 9,4/8/854/11/85, 3/10/85,3/17/85, 3/20/85,4/23/85 NUM3/Discussion,4/30/85,5/1/85 drainage.Northern pike populations in the Yentna River system,partic- ularly around Skwentna,has increased dramatically in the last decade (Kelly Hepler,pers.comm.).These fish are believed to have been originally illegally stocked (Stan Kubik,pers.comm.). Threespine stickleback Populations of threespine stickleback appear to be very variable in the Susitna River.In 1981,large numbers of Age 3 stickleback were cap- tured throughout the lower reach above RM 10.1.Although 1 imi ted sampling was done in 1982,relatively few threespine sticklebacks (mostly Age 1)were captured.Prior to the 1984 season,we theorized that catches would be high for threespine stickleback if the offspring of the 1981 spawners returned in 1984 as Age 3 fish.However,there was no large increase in Age 3 stickleback catches in 1984.Over 95%of the 1984 catch was Age I fish.It is not known what has caused the decrease in threespine stickleback numbers. Capture data in both 1981 and 1982 suggests that threespine stickleback are anadromous and an upriver spring migration begins from the estuary in 1ate May (ADF&G 1981b,1983b). Ninespine stickleback Catch data from 1981-84 reflects the scarcity of this species in the lower river.In 1984,a small concentration of ninespine sticklebacks were captured in a low oxygen and heavily vegetated slough in the lower DRAFT/PAGE 10,4/8/854/11/85,~ 3/10/85,3/17/85,3/20/85,4/23/85 NUM3/Discussion,4/30/85,5/1/85 Susitna at RM 57.2.In 1981,only two fish were captured with both captures occurring in the Deshka River (unpublished data). Arctic lamprey Arctic lamprey have been found to be much more abundant in the lower river than the middle river.Most Arctic lampreys have been captured in muskeg or shallow lake draining tributaries such as Birch Creek or Deshka River. Arctic lamprey populations can be anadromous or resident (Morrow 1980). Both forms have been reported to be parasi ti c,but the more bl unt teeth found in the freshwater form suggests that it is nonparasitic.McPhail and Lindsey (1970)report that while the largest Alaskan Arctic lamprey found is 411 mm,most adults of the nonparasitic form rarely exceeds 180 mm.In·four years,only 16 adul ts over 180 mm have been captured. One fish captured in 1984 at the Oeshka River may have been a Pacific lamprey.It measured approximately 60.0 centimeters (cm)in length and 9.0 cm in diameter.Other adults captured have been smaller in length and much less in diameter. 4.2 Middle Susitna River 4.2.1 Rainbow trout In 1984,our knowledge of the seasonal distribution of rainbow trout in ~I - the middle river was greatly increased.Primary sampling emphasis was -- - -. DRAFT/PAGE 11,4/8/854/11/85, 3/10/85,3/17/85, 3/20/85,4/23/85 NUM3/Discussion,4/30/85,5/1/85 placed on radio tagging fish earlier in the year,May-June,to learn more about the timi ng and 1ocati ons of spawni ng.Emphasi s was also placed on tagging fish from the upper section of the middle river: Portage Creek,Indian River and Fourth of July Creek.Previously few rainbow trout have been captured early in the season,however,18 fish were successfully tagged in 1984 before June 16 of which 14 were pre- spawners.Several pre-spawners from a school of 30 probable spawners were captured in Portage Creek a river which was previously believed to harbor few rainbow trout (Sundet and Wenger 1984).Previous success at capturing rainbow trout in Portage Creek may have failed because most of the adults move up into the creek before sampling normally occurs (after ice-out in the mainstem).Since sampling was done during this time from TRM 0.0 to TRM 11.6 and only these fish were observed or captured,it is highly possible that these fish represented the majority of spawning rainbow trout in Portage Creek.Four tagged fish apparently spawned at TRM 2.3 and another 0.5 miles up the small side tributary which enters at TRM 2.3 during early June 1984 (Figures 9).Movements of other pre- spawning radio tagged fish shows spawning also occurs at TRM 5.1 and at the same time (Figure 10). Other pre-spawning radio tagged rainbow trout show that a similar timing of spawning occurs in Fourth of July Creek at TRM 0.7 and TRM 0.8 in a side tributary which enters at TRM 0.7 of Fourth of July Creek,and probab ly at Litt 1e Portage Creek duri ng the same peri od as those in Portage Creek (Figure 6;Sundet and Wenger 1984;Figures 11 and 5). DRAFT/PAGE 12,4/8/854/11/85, 3/10/85, 3/17/85, 3/20/85,4/23/8S NUM3/Discussion,4/30/85,5/1/85 Water surface temperatures correlated to movement of four radio tagged rainbow trout into Fourth of July Creek in 1984 shows spawning rainbow trout move into that tributary in late May when temperatures are 6.7°C to 8.5°C (Figure 13).Slightly cooler temperatures (2.8°C)"in this tributary was observed in 1983 during the immigration (Mark Wenger, pers.comm.). Both TRM 4.3 and TRM 5.1 of Portage Creek,and TRM 0.7 of Fourth of July Creek are similar in all three have an outlet from a lake flowing into it.It is suspected that rainbows in both Portage Creek and Fourth of July Creek use the confluences of the lake outlets to spawn at because of the outlets warmer water temperatures.During the first week of June <If!lo\ 1984,the mainstem of both Fourth of July Creek and Portage Creek were approximately 8°C,while the outlets at TRM 0.7 of Fourth of July and TRM 2.3 of Portage Creek were 12°C.Little Portage Creek is also influenced by lakes,however,no temperatures have been taken at this creek in June. Follow-up studies in 1984 on the lakes which flow into Fourth of July Creek (Figure 2)and Portage Creek at TRM 2.3 (Figure 3)showed that they were abundant with rainbow trout.Since few juvenile (~200mm) rainbow trout have been captured in Portage Creek over the past four years,it is suspected that the lake which flows into Portage Creek at TRM 2.3 acts as a nursery.No sampling was done in the lakes which outlet at TRM 5.1,but a similar phenomenon probably occurs there. ")J ~l J )J ))1 J ))1 J TEMPERATURE OF TRIBUTARY x FLOWING INTO FOURTH OF--'" JULY CREEK AT TRM 0.7 147 13.1 •I AT THIS TIME,THE fiSH ,ENTERED fOURTH OF 12.1 -l JULY CREEK IN 1984 11.1 10.1 9.1 ~B.l I.lJ ~1.1 I- ~ 0:6.1 I.lJ ~I 0..:e ~I.lJ 5.1 I- 47 .I ., 27 I ., 0., I ~"""I , JAN I fEB I MAR I APR I MAY I JUN I JUL I Figure 13.Mean surface water temperatures in Fourth of July Creek correlated to four spawning radio tagged rainbow trout's movement from the Susitna River into Fourth of July Creek~1984.. DRAFT/PAGE 13,4/8/854/11/85, 3/10/85,3/17/85, 3/20/85,4/23/85 NUM3/Discussion,4/30/85,5/1/85 -Although many juveniles have been captured in Fourth of July Creek,it is suspected that the lakes which outlet to this tributary contribute heavily to the ra"inbow trout population in Fourth of July Creek below TRM 1.8.Support for this hypothesis is provided by:(l)little spawning gravel has been found in Fourth of July Creek between TRM 0.0 and TRM 1.8,(2)several small adult rainbow trout were gill netted above the barrier at TRM 3.5 in 1984 and (3)better scale analysis of rainbow trout in 1984 showed that many fish captured below TRM 1.8 of Fourth of July Creek had the same stunti ng scale patterns as those captured in lakes at its headwaters.Scale analysis of mainstem Fourth of July Creek fish also showed they were usually the smaller sized fish ,in each age class compared to other middle river stocks.For example, ..fl1--~~~f~-Age._5_fish,five fiSh from_Eo~rth of July Creek were the smallest. ~~ While spawning has been found to occur in both Fourth of July Creek and Portage Creek,little spawning evidence has been documented in Indian River.Only one pre-spawning radio tagged rainbow is believed to have spawned in Indian River (Fish 757-1.1 in Figure 8).Other pre-spawners captured at Indian River have moved elsewhere to spawn.A juvenile salmon crew captured over 30,000 juvenile salmon in Indian River during the open-water season of 1984 and only one juvenile rainbow trout.The ~.(apparent low success of rainbow trout spawning in Indian River may be D .due to no acceptable passage for rainbow trout from Miami Lake or othertiE..;,.elakes draining into Indian River.,~~'L ~~Db ./~a';-'f'! ~,~.J-The overall low number of juveni 1es found in the mi ddl e ri ver 'j nfers{~~~. C;;f~that either egg or juvenile survival is extremely low.One reason for ~~+'4Il;-J-~~'+~IS~,rk+-P1/~'~~..~U~I ~V J - - - -- ,~ - -- DRAFT/PAGE 14,4/8/854/11/85, 3/10/85,3/17/85,3/20/85,4/23/85 NUM3/Discussion,4/30/85,5/1/85 this low survival may be the cold temperatures of these tributaries. After spawning in early to mid-June,fish in different systems have been found to act differently for the rest of the summer.In Fourth of July Creek most fish stay between TRM 0.4 and TRM 1.8.However,a number of fi sh have been found to move out 'and into other areas as the mouth, nearby sloughs or into Indian River (Figure 7,Sundet and Wenger 1984).~o In Portage Creek,most radio tagged rainbow trout move further upriver in that tributary after spawning (Figures 9 and 10). Fish tagged in 1984 showed a simi'lar summer association with spawning salmon as did fish tagged in 1983,many of which were found close to spawning chum and pink salmon (Sundet and Wenger 1984).More fish were tagged earlier in 1984 than in 1983 and their movement coincides with the earl i er adult chi nook salmon movement.Thi s movement was most evident at Indian River and Portage Creek which have the greatest number of spawning chinook salmon above Ta"lkeetna;escapement in 1984 was over 1,500 chinook salmon in each river (Barrett et al.1985).As in 1983, this close association with spawning salmon is probably due to rainbow trout utilizing salmon eggs as a primary summer food source.This was substantiated during a late June 1984 helicopter survey.At this time, fish were pinpointed to an exact location in a pool or riffle and in all cases,radio tagged fish were within 100 feet of adult chinook salmon. Most rainbow trout fourid in Portage Creek and Indian River,remain in the tri buta ri es through peak sa 1mon spawning peri ods.Peak ch i nook spawning is in late July and peak pink and chum spawning is in late August.Apparently,however,rainbow trout do not rely on coho salmon hI DRAFT/PAGE 15 t 4/8/854/11/85 t ~ 3/10/85 t 3/1l/85 t 3/20/85 t 4/23/8~ NUM3/Discussion t 4/30/85 t 5/1/85 -eggs as a food source because most fish have moved out before coho peak spawning.Peak coho spawning is in late September. By October 6 in both 1983 and 1984 t all radio tagged fish outmigrated tributaries.Ground surveys conducted in early October 1984 between TRM's 0.1 and 1.0 of Fourth of July Creek also support radio tagged fish outmigration findings.At this timet no rainbow trout were captured and only one was observed. Water surface temperatures and relative depths were correlated to the outmigration of seven radio tagged rainbow trout from Portage Creek to the mainstem in 1984.Water depths appeared to influence the outmi- gration more than temperature.A late season flood between August 20 and August 30 apparently triggered the final outmigration of six fish (Figure 14).Downstream movements had begun much earlier for most of the fish with the maximum upriver locations for four fish being on July 6.The slower downriver movement between then and late August is probab ly due to other reasons such as food supply more than temperature or water depths. Winter monitoring movement of radio tagged fish over three years show nearly all rainbow trout move slightly downstream (0.1-4.0 miles)after October before hol ding (ADF&G 1983b t 1983c;Sundet and Wenger 1984; Figures 4 and 5).This downstream movement is believed to occur because they are searching for acceptable overwintering areas.During all three years t however t several rainbow trout have remained all winter near the tributary at which they were tagged.From 1984 taggings t it appears - 4.0 z 3.0 q I If) \D ,..: I I QlIX) ~'",...,... ~ '"Ia,...'",... aia-- o ~2.~ Q:: I- ~2.0 l-x: C) UJ 1.5 x: UJ ~1.0 C) \L \L ;:O.~ (/) - 0.0 ..I---,..-----,r-----r--...,......--~--,..._-__r--___..--"""'T"-- JUN I JUL I AUG I SEP r OCT I AT THIS TIME,THE FISHES MAXIMUM =UPSTREAM LOCATION IN 1984 I AT THIS T1MEl.THEIFISHWEREFIHST =FOUND AT PORTAGEICREEKMOUTH ..: a , Qla,... X",-TEMPERATURE OF TRIBUTARY FLOWING INTO PORTAGE CREEK AT TRM 2.3 4 8 6 2 14 12 10 , If) \D O,.....L..--.....---.,........----....---,-----.---........--.,..------.----.-- JUN I JUL r AUG I SEP I OCT I Figure 14.Mean surface water temperatures and relative depths in Portage Creek correlated to seven radio tagged rainbow trout's maximum movement and their outmigration into the mainstem Susitna River,1984. -,=_,-----""""".......__........,-~----------------1 DRAFT/PAGE 16,4/8/854/11/85, 3/10/85, 3/17/85, 3/20/85,4/23/8~ NUM3/Discussion,4/30/85,5/1/85 that fish from Portage Creek may overwinter near that tributary mouth. Dur"ing winter sampling in the lower river in 1981-82,although none of the radio tagged fish were captured,a high catch per unit effort was recorded for other rainbow trout in the vicinities of the tagged fish (ADF&G 1983b).Due to this reason,it is suspected that rainbow trout concentrate in small numbers and use specific areas of the lower mainstem Susitna for overwintering.A simi1iar phenomenon probably occurs in the middle river. Winter sampling at radio tagged fish relocations have provided little information to characterize rainbow trout overwintering areas.This is mostly due to the.difficulty in winter sampling.In winter 1983-84 most of the river was open where radio tagged fish were found.Habitat data collected at these areas has varied except for conductivity.In all cases where radio tagged fish have been found in winter,the con- ductivity was relatively high (above 125.0 uhos/cm)indicating that the - areas they seek out are influenced by groundwater.Radio tagged rainbow trout also appear to prefer areas of moderate water velocities.Limited data from winter 1982-83 shows similar conclusions (ADF&G 1983c). ~\, ~",~\"btseveral radio tagged rainbow trout known to be alive in September and ~t~-~October,have been found dead when samp1 ing in January or February. !\..'t1...Thi s supports the theory that there is a heavy overwi nteri ng mortality. ------",...---_•.._.•_._"~.>._,,_.•~.••.-~----""'~~..,.~............ Needham and Jones (l959)and Needham and 51 ater (l94sj report""that overwintering mortalities are often high due to physical catastrophes such as dewatering,collapsed snow banks,and anchor ice formation. ,1;t:~~~h '1,'fr«Jr.~~_.~ Radio tagged fish data reflects the high DRAFT/PAGE 17,4/8/854/11/85, 3/10/85, 3/17/85,3/20/85,4/23/85 NUM3/Discussion,4/30/85,5/1/85 mortal ity rate for rainbow ""." - trout.In 1983,middle river rainbow trout mortality rate was only 33.3%and in 1984 was 42.7%.1 In three years,sport fishermen have reported catching two radio tagged fish,three have been found eaten by predators,and one died after being trapped in a side channel which dewatered during fall.Other mortalities include one attributable to post-spawning die-off and three to overwintering. 4.2.2 Burbot Catch data from 1981-83 shows burbot are much 1ess abundant in the middle river than the lower river (AD¢&G 1981b,1983b,1983c).Although food and rearing habitat could possibly be l"imiting factors in the numbers of burbot found between the Chulitna River confluence and Devil Canyon,biologists observed in winter 1983-84 another reason why there· may be is less burbot in thi s reach.Duri ng aeri a.l tracki ng surveys for radio tagged fish in January,a large section of the river from RM 123.0 to RM 150.0 was noti ced to have rema i ned open.It was observed that approximately half of the area that was open had anchor ice on the substrate and occasionally the anchor ice would free itself and float to the top.5i nce burbot spawn from mi d-January to early February in the 5usitna River,the formation and movement of anchor ice could disrupt the success of spawning.With several poor years of spawning and given the fact most radio tagged burbot have not migrated far or frequently, no new individuals would be recruited to the existing population in 1 The higher survival rate is somewhat biased because in 1984 we sought out more large adults to radio tag.Therefore the survival rate is probably closer to 1983 findings.--G/) DRAFT/PAGE 18,4/8/854/11/85, 3/10/85,3/17/85,3/20/85,4/23/8~ NUM3/Discussion,4/30/85,5/1/85 this area. Although spawning is probably limited in the middle river,one radio tagged fish monitored overwinter in 1983-84 stayed near Slough 10 indicating that some spawning may occur there.The capture of juveniles near Slough 9 in 1982,suggests spawning may also occur there (ADF&G 1983b)• 4.2.3 Arctic grayling - -The general distribution and abundance of Arctic grayling in the middle Susitna River in 1984 was found to be similar to 1981-83 findings.~ PopJJ1ation estimates ~multiple-year data show Arctic grayl ing are------------ the third most abundant resident fish species in the middle'river (other ~---------'-_..__._-~-._--___.__._-'___.~_._-_.. than sculpins and sticklebacks).During all four years,Arctic grayling -'-~--_.~""'--....-_~~--------------~- have been found most numerous at Indian River (RM 138.6)and Portage Creek (RM 148.8).High catches at Whiskers Creek Slough (RM 101.2), Lane Creek (RM 113.6),Fourth of July Creek (RM 131.1),and a mainstem site at RM 150.1 are recorded only in May,June or September.Catches at all sites over all four years have generally been the highest in the spring and fall. Until 1984,our knowledge of Arctic grayling immigration and outmigra- tion from tributaries to the mainstem was based on tag-and-recapture and catch per unit effort data.These data show Arctic grayling move into tributaries in May and early June to spawn,then begin to outmigrate in mid-September (ADF&G 1981b,1983b;Sundet and Wenger 1984).In 1984, -I -I - DRAFT/PAGE 19,4/8/854/11/85, 3/10/85,3/17/85,3/20/85,4/23/85 NUM3/Discussion,4/30/85,5/1/85 the ~i~~~pre-spawning radio tagged fish a..d~d 2.. ~.~Sid::~2!-~o our knowledge of Arctic grayling movement and spawning.~~ ThiSfuh ascended Portage Creek two days after bei ng tagged in the ~~ ",,~J.ect'J~. mainstem on May 22.Other studies done in 1984 also shows the Spring~cf>f\t-immigration to Portage Creek occurs at this time.No fish were observed~(J;"t.,,'" or captured between TRM 1.5 and TRM 3.2 from May 9 to 25,but surveys '"( ,~ conducted later in the upper reaches of Portage Creek showed Arctic grayling were found upriver to TRM 11.6.During early to mid-June, catch rates became higher in the upper reaches"indicating Arctic grayling were ascending Portage Creek. The radio tagged fish apparently spawned at TRM 0.4 between May 22 and 30.A similar timing of spawning was shown by 82 other Arctic grayling in 1984 which were captured and examined for sexual conditions above RM 125.0.Other fish examined in 1984 below RM 125.0 shows timing of spawning there is about seven to ten days earlier than above RM 125.0 (Appendix Figure C-7).The difference in \~iming".ef-{~'p~~~·i~g)between ......,.-.._"~,.~"".._''',...."'..-.,.....",~.,...-'~ these sections is probably due to tributaries warmer water temperatures below RM 125.0 than above.~.similar~i;i-rig"\&f-'spawnin.~J occurred in "\.",,,,«,.,"""'''."".",.".«.••-'' 1983 (Sundet and Wenger 1984). Mean surface water temperatures "j n Portage Creek duri ng 1ate May 1984 when Arctic grayling began to immigrate and spawn were slightly higher than those reported elsewhere in Alaska.Portage Creek water temperatures were 5.1°C on May 26 while Tack (1980,1973)reports the spawning run begins when water temperatures are 1°C and spawning occurs at 4°C in Interior Alaska. DRAFT/PAGE 20,4/8/854/11/85, 3/10/85,3/17/85,3/20/85,4/23/8~ NUM3/Discussion,4/30/85,5/1/85 Catch and recapture data from 1981-84 as well as the radio tagged fish movement data,shows most large adult Arctic grayling remain in tribu- taries through early September.Smaller adults and juvenile Arctic grayling have been captured at tributary mouths during the summer, probably because they were displaced in the~pper re~.~~s a resul~of territorialism (ADF&G 1983b).~~~res-:;~agged fish 2- ShOW~dUlts move little during th~~~mmer.(A?0(~~r~'1 ~(~,~-1,ji~v', It was speculated in 1983 that some adult Arctic grayling may overwinter in Portage Creek,however,evidence from the radio tagged fish suggests that hypothesis is not true.After early August,the fish began to move rapidly downstream along with radio tagged rainbow trout.Unfortu- nately,the tag's battery expired before the Arctic grayling reached the mainstem.Radio tagged rainbow trout which were outmigrating at the same time were all found to be out of Portage Creek and in the mainstem by October 6.The outmigration of the radio tagged Arctic grayling,as well as radio tagged rainbow trout,was apparently triggered by a late season flood between August 20 to 30 (Figure 14).- Little information except what can be speculated from tag recapture data"----.""" has been obtained on the winter distribution of Arctic grayling in the mainstem.Several tag recoveries in 1983 and 1984 from fish tagged during those years suggest some Arctic grayling may overwinter far downriver in the mainstem from their summer rearing tributaries. Several fish tagged at Portage Creek in late May to July have been recovered 10 to 20 miles downriver.Three fish were recovered in late May 1984 25 to 37 miles downriver from where they were tagged in 1982 or 'L DRAFT/PAGE 21,4/8/854/11/85, 3/10/85,3/17/85,3/20/85,4/23/85 NUM3/Discussion,4/30/85,5/1/85 September 1983 and 1984. _. 1983.It is possible these three fish were recovered before they returned upriver or that they had simply relocated during the interim. Most {fsh tagged at Portage Creek or nearby,however,indi cate they may"'1--_.,,_.._-_,.,....,..··,..·..'.M.',•••,"_,__",,_••,.,_",.,••,.,"",..".,, , ~0~~:,~,~.~ter between RM 146.0 toRM14.R.O.or-a·t,·RM,15,Q,•.L ."...,.~Q~t /-hAAJ - electrofishing,gill net,and hoop net catch rates were very high at v)l..·Z these areas during mid-late May and at RM 146.0 to RM 148.0 in late C0I.Jl.~ Irec-lIW/J.'itI L,",JL. In 1984,two Arctic grayl ing were recovered which were tagged in or above Devil Canyon.While no recoveries have been made above the canyon from those tagged below,this is the 'first evidence that fish ~- populations can successfully migrate downriver through Devil Canyon. 4.2.4 Round whitefish Population estimates made in 1984 using multiple-year data show that nd whitefish are the most abundant resident fish species in the middle river.Catch data from 1982-84 show the highest concentrations .of round whitefish occurs between RM 132.6 and RM 150.1,and round whitefish are much more abundant in the middle river than in the lower river. I\,'POOled CPUE rates based on boat e1ectrofi shi n9 data from 1982-83 ShO~r that round whitefish CPUE's at tributary or slough sites ~e €SY higher)~{:l?} -------l!-han_~itL"~~.:!,"nstem sites.Relatively high-CPUE's at mainstem sites,f~''"\'r~.)% however,are in June and September of those years and al so ~~,:!~~.!:1.'~~'.(~·1tt7-( (1984.Juvenile round whitefish captured at JAHS sites have been found ! """"'......-'''''''',..._._''",... ~ff<.PU!&q /U9 7 v.-..-l(Q '( DRAFT/PAGE 22,4/8/854/11/85, 3/10/85,3/17/85,3/20/85,4/23/8E NUM3/Discussion,4/30/85,5/1/85 more often at turbid mainstem and slough sites than at tributary sites (Suchanek et ale 1984).Juvenile round whitefish prefer these areas because of lower water velocities and higher turbidities which they use as cover. While a definite fall downriver movement was shown by recaptured round whitefish in 1981 and 1982,only a slight downstream movement throughout the summer was shown by recovered fish in 1983 and 1984. Extremely sexually ripe round whitefish have been captured in the mainstem Susitna or at the mouths of several middle river tributaries each year from 1981-84 during early October.This suggests that round whitefish extensively use mainstem influenced areas for spawning.Peak spawning probably occurs from mid to late October.Several areas have been found since 1981 where round whitefish specifically spawn at such as the mouths of Lane Creek,Indian River,and Portage Creek and in the mainstem at RM 147.0.Other round whitefish close to spawning have also been found scattered throughout the middle river in pairs or small groups. Whil e it is unknown where adul t round whi tefi sh overwi nter,early spri ng catch and recapture data suggests they may overwinter near their summer rearing areas which ;s primarily in the mainstem above RM 132.0 (ADF&G 1983b;Sundet and Wenger 1984). - ,.... -.. - DRAFT/PAGE 23,4/8/854/11/85, 3/10/85,3/17/85, 3/20/85,4/23/85 NUM3/Discussion,4/30/85,5/1/85 4.2.5 Humpbackwhitefish Catch'data from 1981-84 shows humpback whitefish are relatively scarce in the middle river.Most humpback whitefish in the middle river probab ly overwi nter in that reach.However,increased outmi grant trap catches of humpback whitefish at RM 103.0 in the fall and high growth rates shown in scales of several adults suggests that some humpback whitefish may outmigrate from the middle river to overwinter in the estuary (Sundet and Wenger 1984). 4.2.6 Longnose suckers Catch data shows little differences in distribution or abundance of longnose suckers between 1982,1983 and 1984.Recapture data from these years show longnose suckers generally move little in summer,but a spring upriver and a fall downriver migration may occur.Several 1984 recaptured fish also shows some fish move upriver during June through August. 4.2.7 Other species Dolly Varden Dolly Varden have been found mostly at,Lane Creek,Indian River,and Portage Creek.Catch data suggests Dolly Varden move into tributaries before late June (ADF&G 1983b,1984b).It is believed that they stay in DRAFT/PAGE 24,4/8/854/11/85, 3/10/85,3/17/85,3/20/85,4/23/8! NUM3/Discussion,4/30/85,5/1/85 the tributaries through October at whi ch time they spawn and then outmigrate to overwinter in the mainstem. Dwarf populations of Dolly Varden are found in the upper reaches of several tributaries (ADF&G 1983b).These populations are believed to remain in these tributaries year-round. Lake trout Two populations of lake trout were found in the middle river drainage during lake surveys for rainbow trout in 1984.While Miami Lake had been known to have lake trout in it,no information was known on a lake feeding Fourth of July Creek (Lake B in Figure 2).Although only juveniles were cap,tured,Lake B may harbor a sizeable population of adul t 1ake trout.A steep shorel i ne and no boat prevented proper setting of gillnets in this lake.No depths were measured,however,the lake was extremely clear and appeared to be over 50 feet deep. Threespine stickleback Threespine stickleback are less numerous in the middle reach of the Susitna River than the lower reach.However,they have been caught in relatively large numbers between the Chulitna River confluence and RM 120.0 (ADF&G 1981b). -, i - - .- .... DRAFT/PAGE 1,4/12/85 3/18/85,3/20/85 NUM3/Contributors 5.0 CONTRIBUTORS Data was primarily collected by Richard Sundet and Stuart Pechek.Data were also collected by fishwheel,outmigrant,and JAHS crews. Dana Schmidt provided the study design and assisted with making population estimates.Drew Crawford edited the draft. Data processing was done by Allen Bingham,Alice Freeman,Kathrin Zosel, Donna Buchholz,and Dan Sharp. Drafting was done by Carol Hepler and Roxanne Petersen. Typing was done by Skeers Word Processing • DRAFT /PAGE 1 3/18/85 t 4/12/85 NUM3/Acknowledgements 6.0 ACKNOWLEDGEMENTS Funding for this study was provided by the State of Alaska t Alaska Power Authority. We would like to express our gratitude to all the people and organiza- tions that provided information or assistance to the Resident Fish Study during the past years.We would also like to thank Leon Dick and David James for describing resident fish catch rates on the Deshka River and east side tributaries t respectively. 'We are especially grateful to Carl Burger (USFWS)for his technical expertise and advice on radio telemetry investigations.We would also like to thank Carl for providing photographs showing the implantation of radio tags into rainbow trout. - - ..... - - DRAFT/PAGE 1,4/8/85,5/1/85 3/18/85, 3/20/85,4/12/85 NUM3/Literature Cited,4/30/85, 7.0 LITERATURE CITED Alaska Department of Fish and Game (ADF&G).1981a.Aquatic studies procedures manual.Phase 1 (1980-81).Subtask 7.10.Alaska Department of Fish and Game Susitna Hydro Aquatic Studies. Anchorage,Alaska. 1981b.Phase 1 final draft report.Subtask 7.10.Resident fish investigation on the Lower Susitna River (November 1980-0cto- ber 1981).Alaska Department of Fish and Game Susitna Hydro Aquatic Studies.Anchorage,Alaska. 1981c.Phase 1 final draft report.Subtask 7.10.Adult anadromous fisheries project.Alaska Department of Fish and Game Susitna Hydro Aquatic Studies.Anchorage,Alaska. 1983a.Aquatic studies procedures manual.Phase II (1982-83). Subtask 7.10.Alaska Department of Fish and Game Susitna Hydro Aquatic Studies.Anchorage,Alaska. 1983b.Susitna Hydro aquatic studies phase II basic data report.Volume 3 (1 of 2).Resident and juvenile anadromous fish studies on the Susitna River below Devil Canyon,1982.Alaska Department of Fish and Game Susitna Hydro Aquatic Studies. Anchorage,Alaska. DRAFT/PAGE 2,4/8/85,5/1/85 3/18/85,3/20/85,4/12/85 NUM3/Literature Cited,4/30/85 -1983c.Susitna Hydro aquatic studies phase II data report. Winter aquatic studies (October 1982 -May 1983).Alaska Department of Fish and Game Susitna Hydro Aquatic Studies. Anchorage,Alaska. "'"" 1983d.Use of major habitat types by juvenile salmon and resident species.Appendix G in Susitna Hydro aquatic studies Phase II report.Synopsis of the 1982 aquatic studies and analysis of fish and habitat relationships (2 of 2:Appendix A-K).Alaska Department of Fish and Game Susitna Hydro Aquatic Studies. Anchorage,Alaska. 1983e.Influence of habitat parameters on distribution and relative abundance of juvenile salmon and resident species. Appendix F in Susitna Hydro aquatic studies Phase II report. Synopsis of the 1982 aquatic studies and analysis of fish and habitat relationships (2 parts).Alaska Department of Fish and Game Susitna Hydro Aquatic Studies.Anchorage,Alaska.- 1983f.Habitat location description and photos.Appendix F in Susitna Hydro aquatic studies phase II basic data report.Volume 4 (3 of 3:Appendices D-J).Aquatic habitat and instream flow ., studies,1982.Alaska Department of Fish and Game Susitna Hydro Aquatic Studies.Anchorage,Alaska. 1984. DRAFT/PAGE 3,4/8/85,5/1/85 3/18/85,3/20/85,4/12/85 NUM3/Literature Cited,4/30/85 Susitna Hydro aquatic studies (May 1983 -June 1984) ,- procedures manual (l of 2).Alaska Department of Fish and Game Susitna Hydro Aquatic Studies.Anchorage,Alaska. Barrett,8.M.,F.M.Thompson,and S.N.Wick.1985.Adult anadromous fish investigations:May -October 1984.Alaska Department of Fish and Game Susitna Hydro Aquatic studies Report No.1.Prepared for Alaska Power Authority.Anchorage,Alaska. Everhart,W.H.,A.W.Eupper,and W.O.'Youngs.1975.Principles of fishery science. Kingdom. Cornell University Press.London,Uni ted McPhail,J.D.,and C.C.Lindsey.1970.Freshwater fishes of north- western Canada and Alaska.Bulletin of the Fisheries Research Board of Canada 173:1-381. Morrow,J.E.1980.The freshwater fishes of Alaska.Alaska Northwest Publishing Company,Anchorage,Alaska. Needham,P.R.,and D.W.Slater.1945.Seasonal changes in growth, mortality and condition of rainbow trout following planting. Transactions of the American Fisheries Society.73:117-124. Needham,P.R.,and A.C.Jones.1959.Flow,temperature,solar radiation,and ice in relation to activities of fishes in Sagehen Creek,California.Ecology 40(3):465-474. -':< DRAFT/PAGE 4,4/8/85,5/1/85 3/18/85,3/20/85,4/12/85 NUM3/Literature Cited,4/30/85 R&M Consultants.1982.Tributary stability analysis report.Susitna Hydroelectric Project.Prepared for the Alaska Power Authority. R&M Consultants.Anchorage,Alaska. Ricker,W.E.1975.Computation and interpretation of biological statistics of fish populations.Bulletin of the Fisheries Research Board of Canada.Bulletin 191.Ottawa,Canada. Suchanek,P.M.,R.L..Sundet,and M.N.Wenger.1984.Resident fish habitat studies.Part 5 in Schmidt,D.C.,S.S.Hale,,D.L. Crawford,and P.M.Suchanek (eds.).1984.Resident and juvenile anadromous fish investigations (May -October 1983).Susitna Hydro Aquatic Studies.Report No.2.Alaska Department of Fish and Game.Anchorage,Alaska. Sundet,R.L.,and M.N.Wenger.1984.Resident fish distribution and population dynamics in the Susitna River below Devil Canyon.Part 5 in Schmidt,D.C.,S.S.Hale,D.L.Crawford,and P.M.Suchanek (eds.).1984.Resident and juvenile anadromous fish investi- gations (May -October 1983).Susitna Hydro Aquatic Studies. Report No.2.Alaska Department of Fish and Game.Anchorage, Alaska. Scott,W.B.,and E.J.Crossman.1973.Freshwater fishes of Canada. Bulletin of the Fisheries Research Board of Canada 184.Ottawa, Canada. - - - - - DRAFT/PAGE 5,4/8/85,5/1/85 3/18/85,3/20/85,4/12/85 NUM3/Literature Cited,4/30/85 Tack,S.L.1973.Distribution,abundance,and natural history of the Arctic grayling in the Tanana River drainage.Alaska Dept.of Fish and Game,Federal Aid in Fish Restoration,Annual Report of progress,1972-1973,Project F-9-5,14(R-I):34 p. 1980.Migrations and distribution of Arctic grayling in interior and Arctic Alaska.Alaska Dept.of Fish and Game,Federal Aid in Fish Restoration,Annual Report of Progress,1979-1980, Project F-9-12,21(R-I):32 p. Ziebell,C.D.1973.Ultrasonic transmitters for tracking channel catfish.The Progressive Fish Culturist.35(1):28-31. ,~ - - - - DRAFT/PAGE 1,4/8/85,4/23/85 3/18/85,3/20/85,4/30/85 NUM3/Appendix A,5/1/85 APPENDIX A Floy Anchor Tag Retention Rates - ..... DRAFT/PAGE 2,4/8/85,4/23/85 3/18/85,3/20/85,4/30/85 NUM3/Appendix A,5/1/85 METHODS The external Floy anchor tag (model FD-67)has been used to tag resident fish since January 1981 to determine seasonal and yearly movements.The dimensions of the tag and tagging procedure are explained in the 1981 procedures manual (ADF&G 1981a).Di sc dangl er tags were used to tag burbot for several months during 1981 and spring 1982. Floy anchor tag retention rates were evaluated for Arctic grayling, round whi tefi sh,and longnose suckers by compa ri ng the number of fi sh with tag scars to the total number.of fish with tag scars and Floy anchor tags of that species recaptured in 1983.1 By subtracting this ratio from 1.00,Floy anchor tag retention rates were determined.Tag retenti on rates fol"rai nbow trout were not determi ned because the smaller scales of this species regenerate rapidly and make it difficult to detect tag scars.In 1983,no captured longnose suckers showed a tag scar. RESULTS The Floy anchor tag retention rate for Arctic grayling during 1984 was 75.9%with 15 of 58 recaptures showing a tag scar.The 1984 tag retention rates for round whitefish and longnose suckers were 83.7%and 92.9%respectively,with 15 of 92 round whitefish and one of 14 longnose suckers showing a tag scar. 1 Only those fish recaptured by a resident study crew were used. Other groups such as sport fi shermen or fi shwheel crews di d not look for scarred fish. fJ -I DRAFT/PAGE 3,4/8/85,4/23/85 3/18/85,3/20/85,4/30/85 NUM3/Appendix A,5/1/85 DISCUSSION We believe that improper tag placement has been the primary cause of tag loss in our studies.In 1982,tags were injected into the dorsal musculature.After 1982,tags were anchored at the base of the dorsal fin through the interneural rays.We have noticed increased tag retention rates each year thereafter.Tag retention rates for Arctic grayling increased from 69.4%in 1983 to 75.9%in 1984.Tag retention rates for round whitefish showed similar improvement,increasing from 77.5%in 1983 to 83.7%in 1984. - - - - - ".... - DRAFT/PAGE 4,3/20/85,4/23/85 3/10/85, 3/18/85,4/30/85 NUM3/Appendix C,5/1/85 APPENDIX B Radio Tagged Fish Capture,Biological, and Winter Habitat Data - - )J ))J )i i ]1 M 1 1 )-~J ]]j DRAFTIPAGE 1 4/23185,4/30/85 NUM3/Tab1e B-1 Appendix Table B-1.Summary of tagging data for radio tagged rainbow trout captured on the Susitna River between the Chulitna River confluence and Devil Canyon,May to July 1984. Radio Frequencyl Type of Fork Implant:F10y Location Lengths Internal Tag Agel Spawning Method Captured River Date Date (mm)or External Number Sex Condition Captured and Re 1eased Mile Captured Released 599-1.1/485 Internal 17457 -1M pre-spawning HL Portage Creek TRM 2.3 6/1 6/1 598-1.6/410 Internal ---6/M pre-spawning HL Tributary Mouth at TRM 0.5 6/5 6/5 TRM 0.7 of Fourth of July Creek 613-1.0/475 Internal 17453 81M pre-spawning HL Portage Creek TRM 2.3 6/1 6/1 608-1 .7/391 Internal 11948 8/F pre-spawning EF Lane Creek 113.6 5/17 5/18 ~(recap was 670-1.4) I 608-1 .9/400 Internal 17757 6/F post-spawned HL Fourth of July Cr.TRM 0.4 6/27 6/27- 620-1~2/390 Internal 17492 6/M non-spawner EF Portage Creek 148.8 6/4 6/4 630-1.0/450 Internal 17761 -1-non-spawner HL Fourth of July Cr.TRM 0.8 6/27 6/27 6~0-1 .7/495 Internal 17753 -1M pre-spawning HL Portage Creek TRM 5.1 6/20 6/20 659-1.2/435 Internal 12615 7/-post:"spawned EF Indian River 138.6 5/26 5/27 670-1.2/450 Internal 17456 -IF pre-spawning HL Portage Creek TRM 2.3 6/1 6/1 709-1.2/433 Internal 5825 6/F pre-spawning EF Indian River 138.6 5/22 5/24 709-2.4/450 External 17598 -IF pre-spawning EF Fourth of July Cr.131.1 5/28 5/31 719-1.6/405 Internal 1243 6/---EF Indian River 138.6 7/23 7/23 720-2.4/400 External 5823 7/F pre-spawning HL Fourth of July Cr.131.1 5/23 5/24 728-1.01392 Internal 5091 -1M non-spawner HL Portage Creek TRM 5.1 6/5 6/5 not sexed or not aged EF =Electrofishing HL =Hook and U ne HN =Hoop net TRM =Tributary River Mile Appendix Table B-1 (Continued). Radio Frequency Type of Fork Implant:Floy Lengths Internal Tag Age/Spawning (mm)or External Number Sex Condition 729-1.5/420 Internal 5107 7/- 730-2.4/485 External 15464 -/M pre~spawning 740-1.4/425 Internal 5867 7/-non-spawner 749-1.8/418 Internal 5092 11M pre~spawning 757-1.1/424 Internal 5817 7/M pre~spawning 760-2.4/400 External 16066 7/M post-spawned ~770-2.4/462 External 17454 8/F pre-spawning I ~Total =22 Fish =not sexed or not aged EF =Electrofishing HL =Hook and Une HN =Hoop net TRM =Tributary River Mile DRAFT/PAGE 2 4/23/85,4/30/85 NUM3!Table B-1 Location Method Captured River Date Date Captured and Released Mile Captured Released EF Whiskers Creek Slough 101.2 5/21 5/21 HL Fourth of July Cr.131.1 5/18 5/18 HN Mainstem 136.7 6/5 6/5 HL Portage Creek 2.3 6/5 6/5 EF Indian River 138.6 5/22 5/24 HL Fourth of July Cr.0.7 6/3 6/3 HL Portage Creek 2.3 6/1 6/1 I " ,.)J ~]J J -=...J j n Radio tagged rainbow trout habitat measurements taken at their relocations were tagged in 1983. }) Appendix Table B-2. 1 J ..-J ]1 j 1 1 ]-»}) DRAFT /PAGE 1 4/23/85 1 4/30/85 NUM3ITable B-2 in January and February 1984.Fish Ice OpenTO)Fhh Radio Covered Movement Depths (ft)VelocHy Wa ter Qua 1i t, Frequency Date RM (c)(;n ft)Water Ice Slush (fps)Substrate Temp be DO mg/l umhos cm pH 670-1.4 1/11 101.1 c 0.0 2.5 2.5 0.0 *1~2 rubble/cobble 2/14 100.7 c -50.0 5.7 2.5 0.0 2.5 rubble/cobble +0.1 14.0 236.0 7.6 709-1.5 2/15 116.5 c -10.0 3.0 2.0 0.0 1.5 rubble/cobble -0.1 13.1 256.0 7.5 718-1.5 1/11 131.1 0 --*4.0 0 0 --rubb 1e/cobb 1e +0.2 13.6 247.0 7.9 729-1.01 2/16 64.8 c 0 1.5 2.5 3.0 0.0 cobble 729-1.3 2/15 111.4 c -200.0 10.0 2.0 0.0 2.5 cobble -0.1 13.0 212.0 7.5 767-1.51 1/11 114.8 c 0 1.0 2.0 0.5 0 2/15 114.5 c 0 2.0 2.0 0.5 0.5 ~ 1 Fish believed dead~ \.J *Estimated measurement because meter did not work or too deep --No movement or no measurements taken - ""'" ,.... DRAFT/PAGE 1,3/20/85,4/23/85 3/10/85, 3/18/85,4/30/85 NUM3/Appendix C,5/1/85 APPENDIX C Population and Biological Characteristics -. DRAFT/PAGE 2,3/20/85,4/23/85 3/10/85,3/18/85,4/30/85 NUM3/Appendix C,5/1/85 A total of 6,941 resident fish of eleven species were measured for length on the Susitna River from May to October 1984.Appendix Table C-1 presents the range and means of fish measured and Appendix Figures C-l to C-6 present length frequency compositions of six of these species. Sexual maturities of four resident fish species were determined in 1984. Appendix Table C-2 presents lengths of fish examined for sexual condi- tion and Appendix Figure C-7 illustrate the timing of spawning for two of these species:rainbow trout and Arctic grayling. Ages were determined for spawners of three species "in 1984 (Appendix Figures C-8 to C-10).Ages were also determined for spawning and non-spawning rainbow trout captured in lakes C and D at the headwaters of Fourth of July Creek (Figure C-11). The ages of 147 rainbow trout captured on the Susitna River between the Chulitna River confluence and Devil Canyon in 1984 were determined by scale analysis.These rainbow trout ranged in age from Age 1 to Age 9. Ages 4 (l7.0%),5 (26.5%),6 (23.1%),and 7 (20.4%)were sampled most often (Appendix Table C-3).The ages of thirty-five rainbow trout captured in lakes A,C,and 0 ranged from Age 2 to Age 7.In these lakes,ages 3 and 4 were sampled more frequently than others.A graph- ical representation of age-length data shows rainbow trout captured in the mainstem Susitna and its tributaries are approximately 60 mm larger at a given age class than fish captured in Lakes A,C,and 0 (Appendix Figure C-12). C--} Appendix Table C-1. DRAFT/PACE 1 4/30/85 NUM3B/Table C-1 length data for resident fish captured on the Susitna River,1984.- length (mm) Speci es Sampling Period n Range Mean -Rainbow Trout lower Susitna 105 27 -530 281.9 (Fl)Middle Susitna 227 27 -545 282.7 lakes in Middle Susitna 200 70 -360 180.7 Combined Total 532 27 -545 244.2 - Burbot lower Susitna 217 209 -701 406.9 (Tl)Middle Susitna 15 42 -475 241.2 ~ Combined Total 232 42 -701 396.2 Arctic Grayling lower Susitna 197 89 -392 251.2 -(Fl)Middle Susitna 641 40 -427 255.4 Combined Total 838 40 -427 254.5 iO""l Round Wh;tefi sh lower Susitna 301 40 -469 222.8 (Fl)Middle Susitna 1729 20 -410 140.8 Combi ned Total 2030 20 -469 153.0 Humpback Whitefish lower Susitna 348 30 -510 286.2 (Fl)Middle Susitna 298 25 -410 87.6 Combined Total 646 25 -510 194.2 Longnose Suckers lower Susitna 377 30 -447 297.3 (Fl)Middle Susitna 490 24 -392 148.8 Combi ned Total 867 24 -447 213.4 Dolly Varden lower Susitna 6 163 -366 247.3 (Fl)Middle SusHna 16 119 -457 249.5 lakes in Middle Susitna 20 92 -190 125.9 Combined Total 42 92 -457 190.3 lake Trout lakes in Middle Susitna 3S 60 -468 109.6 -(Fl) Threespine Stickleback lower Susitna 1271 20 -90 28.8 (Tl)Middle Susitna 337 15 -74 32.0 Combined Total 1608 15 -90 29.5 Northern Pike lower Susitna 3 83 -713 433.7 (Fl) Arctic lamprey lower Susitna 21 113 -162 134.7 (Tl)Middle Susitna 87 74 -290 126.1 Combined Total 108 74 -290 127.8 Fl =Fork length Tl =Total length - e..--1" RAINBOW TROUT IS ~te ev >uc ~ ) (T ~ L to. X S Fork length n •532x'"244.2 range"'27-545 8 S8 188 lS8 288 258 388 358 ....S8 S"SS8 11811 Lenglh (",,,,) 8'..L---r-..L-,.--,..--,....--..,..-----,----r-....,.---r--.---;::=--\--..,.....- Appendix Figure C-l.Length frequency composition of rainbow trout captured in the Susitna River between Cook Inlet and Devil Canyon by all gear types.May to October 1984 • ..... BURBOT 15 E 18 e v Total length n =232x=396.2 range =42-701 > ()c ~ J (] 11/ L u. X 5 8-'----......J--.--+-'--.----.--,.--r-..--.....,..-..---r-,..--r-.,---r...l-J,--+-.,--- A SA lIlA 15A 2AIt 25A lltA 3SA <AA _Sit SAIl sse 6"8 6SA 71llt 75"RAA RSA lenglh <",,,,) Appendix Figure C-Z.Length frequency composition of burbot captured in the Susitna River between Cook Inlet and Devil Canyon by all gear types.May to October 1984. em M 'U ~10 ev )- Uc: iJ J r:J'•.. IA. X 5 ARCTIC GRAYLING Fork 1ength n ..838 i ..254.5 range ..40-427 0 .........-"""T""".........--.,.--.----...---"""T"""---.--.----.----r--'=::::;.--.--- 50 100 150 280 250 308 358 480 458 508 lenglh <.....> Appendix Figure C-3.length frequency composition of Arctic grayling captured in the Susitna River between Cook Inlet and Devil Canyon by all gear types,May to October 1984. ROUND WHITEFISH 35 3e 25 10 5 Fork length n '"2030x'"153.0 range =20-469 -iii 50 llile 150 280 250 3ee 350 400 451i1 500 550 Lenglh <.....) Appendix Figure C-4.length frequency composition of round whitefish captured in the Susitna River between Cook Inlet and Devil Canyon by all gear types,May to October 1984. HUMPBACK WHITEFISH- 25 20 Fork length ,..,.!!...646 E x ..194.2 E range ..25-510 .v 15 >- U C '11 J (J QI '0I. lL. ::<: 5 I0.L.--r-.L--r---r---r--=~=---r---r--,.--,.---\==::;="'--.,..-- o 50 199 150 260 Z50 300 350 499 450 S00 SS0 Lenglh (mm) Appendix Figure C-5.Length frequency composition of humpback whitefish captured in the Susitna River between Cook Inlet and Devil Canyon by all gear types,May to October 1984. LONGNOSE SUCKER 2S ZIl - ,.., E E v '5 >-o C QI J (J ~10 lL. Fork length n ..867x..213.4 range =24-447 ~.....---,-_......;:=::::;:=~--,.---,-----.---.--!:::==r---,- o 50 tile ,50 200 2S0 3ell 3S11 40e 4S0 500 Lenglh (mm) Appendix Figure C-6.Length frequency composition of longnose suckers captured in the Susitna River between Cook Inlet and Devil Canyon by all gear types,May to October 1984. ------------_....._------------------------------------------ Appendix Table C-2. DRAFT /PAGE 1 4/23/85,5/2/85 NUM3/Table C-2 Fork lengths (mm)of sexually mature and immature resident fish captured on the Susitna River,1984. Spawning Males Females Combined Sexes Species Sampling Area Dates Condition n Range x n Range x n Range x Rainbow trout Middle Susitna reach 5/17-6/27 1 7 320-495 428 11 345-462 406 18 320-495 415 5/26-6/27 2 5 306-370 332 9 310-450 404 14 306-450 378 5/17-6/27 3 12 306-495 388 20 310-462 405 32 306-495 399 6/5-6/28 4 ---- -- ------9 255-392 332 Fourth of July 9/14 1 29 160-310 217 20 173-270 222 49 160-310 219 Lakes C and D 9/14 4 ------------15 95-240 155 -- Arctic grayling Middle Susitna reach 5/17-6/2 1 19 322-417 370 26 265-395 321 45 265-417 341 5/19-6/28 2 27 302-427 355 29 305-405 352 56 302-427 353 5/17-6/28 3 46 302-427 361 55 265-405 337 101 265-427 348 5/27-5/28 4 ------------4 276-290 283 - r Round whitefish Combined lower and 9/26-10/15 1 13 244-390 324 20 292-425 334 33 244-425 330 middle Susitna 10/13-10/15 2 ------4 305-394 349 4 305-394 349,reaches 9/26-10/15 3 13 244-390 324 24 292-425 336 37 244-425 332 "Longnose suckers Combined lower and 5/19-10/15 1 57 273-390 333 1 --345 58 273-390 333 middle Susitna 6/28-6/29 2 2 322-330 326 ------2 322-330 326 reaches 5/17-10/15 3 59 273-390 333 1 --345 60 273-390 333 ~Pre-spawners 3 Post-spawners 4 Combined pre-and post-spawners Non-spawners --No sample or sex undetermined )I I •I J .1 I )I.. _. ...., ..... - ~u Z 1LI :::> 0 1LIa: LL I- Z 1LI Ua: 1LI :so a. 20 10 0 RAINBOW TROUT Ed sPRE-SPAWNERS ~aPOST-SPAWNERS ~o Z ILl ;:)a ILla:: L ~ Z ILloa:: ILl C1. 100 90 AHCrn-GRAYUNG Appendix Figure C-7.Timing of 1984 rainbow trout and Arctic grayling spawning in the middle Susitna River determined by the incidence of pre-to post-spawners. (.,-7 Appendix Figure C-8.Age and length relationship for spawning rainbow trout captured in the Susitna River between the Chulitna River confluence and Devil Canyon, May 17 through June 27,1984. """ "... - n-35 i·358mm N K C n-35 "i.332 mm ••mean I·ranQ8 8911• •w-, til. c c 561 ,A •II) 109 'A E' 1 8 =til. c 6 c 410 380 390 370 360 400 3~0 330 340 300 310 270 290 260 250 280 .320~a:o lL. -E E- - Spawnl~Males Spawning Females Appendix Figure C-9.Age and length relationships for spawning Arctic grayling captured in the Susitna River between the Chulitna River confluence and Devil Canyon, May 17 through June 5,1984. 410 390 370 -(I) -(I) 350 ..:1 -(I) E c E ..-330 c :r ~I....... C). '"c Z 310 . c l&J ..J -(II ~cr 290 0 li.. 270 n=10 n=22 x=311 mm x·332mm 250 -.mean -(I)I·range 230 ""'"I 5 6 7 8 9 5 6 7 8 9 10 12,'AGE'I I i ,'AGE',iW" 210 Spawning Moles Spawnin(J Females Appendix Figure C-I0o Age and length relationships for spawning round whitefish in the Susitna River between Cook Inlet and Devil Canyon,October 9 to to October 15,1984. G--(0 7654 -260ee- :x::240... (!) Z 220 l&J ...J ~200 a:: 0 LL.180 160 ~=NON -SPAWNERS III =SPAWNING FEMALES 140 ~=SPAWNING MALES I =RANGE 120 •.MEAH 100 80- - .~ 360 I .-340 320 ~ 300 - AGE Aopendix Figure C-11.Age and length relationships for pre-spawning rainbow trout captured in lakes C and D at the headwaters of Fourth of July Creek, September 14,1984. Appendix Table C-3. DRAFT/PAGE 1 4/23/85,4/30/85 NUM3/Table C-3 Rai nbow trout age-l ength re1ati onshi p5 on the Sus i tna Ri ver betwee~the Chuli tna Ri ver confl uence and Devil Canyon,May to October 1984.Fish aged were captured by all methods. Numbers Len§th (mm) Age of Fish Standard 596 Conti dence (Years)Sampled Mean Deviation Intervals Range Fish captured in the mainstem Susitna and its tributaries ·1 1 155 2 1 216 3 9 232 28.36 210 -254 186 -282 4 25 293 33.40 279 -307 245 -395 5 39 328 31.80 317 -338 255 -375 6 34 380 34.72 368 -392 291 -436 7 30 414 31.83 402 -425 360 -468 8 6 475 22.26 452 -498 452 -517 9 2 538 9.99 450 -626 531 -545 TOTAL 147 353 186 -545 0 Fish captured in three lakes at the headwaters of Fourth of July Creek (RM 131.1) \117 2.88 114 -120 115 -12125 ~3 8 185 22.26 166 -204 160 -215 4 8 223 8.61 215 -231 208 -235 5 6 246 18.54 226 -266 218 -265 6 6 294 20.43 276 -316 263 -310 7 2 355 7.07 325 -385 350 -360 TOTAL 35 223 115 -360 1 Methods of capture were by:boat electrofishing,gill net,hoop net,and hook and line. J I ..._J .1 !&;t •.~I ,.... 'j'i'll'lllll- 600 ~~o -:1 ~oo !l 4~0 0 ~-1fte••e .- 400 I"l-. 0: I 9 j 8 I 7 -.•..... T t.I • .l.e I_FISH .N THE IIAINSTEM SUSITNA RIVER AND ITS TRIBUTARIES,n -147 FISH IN LAKES AT HEADWATERS OF FOURTH OF JULY CREEK (RM 131.1),ft-35 RANGE OF FISH MEAN _0 I·...... 100 I· •= ~O~ 7 I ,I I 2 3 4 ~6 AGE :J: i-3~0 •'"C)• Z • 0 LaJ • .,.~..J 300,. C ~:1 T a::2~0 I 0 f • ~T.I •..••.....e e ..L C 200 ':'..-e I :. I~O -'- -Appendix Figure C-12.Age and length relationships fer rainbow trout captured in the Susitna River between the Chulitna River confluence and Devil Canyon, May to October 1984. "....c...-1J DRAFT/PAGE 3,3/20/85,4/23/85 3/10/85,3/18/85,4/30/85 NUM3/Appendix C,5/1/85 One hundred forty-three of the 147 rainbow trout aged in the mi ddl e river were captured by hook and line or by boat e1ectrofishing.1 The instantaneous survival rate for rainbow trout in this reach of river captured by these two methods was calculated at 42.7%(Appendix Figure C-13). 1 To be consistent with 1983 instantaneous survival rate calcu- lations,one Age 6 and three Age 7 fish were not used in 1984 calculations because they were captured by gear other than boat electrofishing and hook and line.Fish used in 1983 calculations were only used if they were captured by boat e1ectrofishing or by hook and line. ~ 1984 DATA POINTS ,1 •0.'4 41 •6&22.53 II •:-0.85 SURVIVAL •42.7 ~ II -143.-'984 DATA POINTS,.-c-19M POINTS USED I"OR REGRESSiON AHALYSIS 100 --ItE6R£SSlOM UN£ 1983 DATA POINTS ~,1 _0 ••7•II -!e8H.&9 50 .11--1.10 ••SURVIVAL •'3.'~•II -244 _.I 983 DATA POINTS 0 o -I 983 POlNTS USED 1"011 W REGRESSION AHALYSIS ~•- •REGRESSION LJH£ :r: 5Qu..•.-u..100 •(f)a::w CD 5~ :::J •Z ~ • ,.... 2 3 4 5 6 7 8 9_.AGE Appendix Figure C-13.Survival rate curves for rainbow trout captured in the Susitna River between the Chulitna River confluence and Devil Canyon,1983 and 1984. - - - - - APPENDIX 0 Population Estimates DRAFT/PAGE 1 4/30/85,5/1/85 NUM3B/Appendix 0 - DRAFT/PAGE 2 4/30/85,5/1/85 NUM3B/Appendix D Methods Population estimates were made of adult (e200 mm)rainbow trout,Arctic grayling,round whitefish,and longnose sucker populations in the middle river using the mul tipl e year (1981-1984)tagg"j ng and recapture data. Since only adult fish have been tagged,population estimates generated are applicable only for fish above 199 mm fork length.Seber-Jolly and Bailey's methods (Ricker 1975)were set up on a commercial microcomputer spreadsheet program.The number of recaptures of each species was adjusted by the tag retention rate for that species (refer to Appendix A and Recapture Data,below,for tag retention rates). Results and Discussion Population estimates from the Seber-Jolly method (Appendix Table D-1) and Bailey's method (Appendix Table D-2)must be considered tentative at this time.The number of recaptures are too low,leading to large confidence intervals on the population estimates using Bailey's method. We are presently working out the confidence intervals for the Seber- Jolly method.For the long term monitoring program,efforts will be redirected so that we can get a higher recapture rate. Recapture Data Floy anchor tag retention rates were calculated in ·1983 and 1984 for several resident fish species.This was done by comparing the number of AppW/ldlo hbl.0-1.Siber-Jolly IIethod-popul.tlon.lurvlv.1 .nd rlcrultllent (I'0Il •four-cltch or 10n,Ir loplrlMnt.fhh tl"ed were captured In tho "Iddl,rtYlr f .....1981-84. Fhh N."ly Fi.h check",._capt.ur••gf Fi ah .uk.d .tDATEPQII...t.Uon R.~,.u't..KArk-.d IMerk.19811982 1983 Tot.)KI RAINBOW TROUT ltal NA 'i'2 NIl NA NA NIl NA NA 1982 1~6S Sll9 144 i7.7 HA NIl 7 2 1983 10:16 0 274 ;'12 2 4 HA 6 I lu,'toto 2 ~b 1984 "NIl 1.7 0 I I.17 NA It.toil.:J n Tut..l NIl NA 9 :I 14 NA NA SurY~v.12 0.12 AIlClTC GRAYLlNil UU ...49 NA NA NA NA NA NA t?1m 2765 :121.400 410 6 NA NA b •1983 6797 <I 70:1 91:J """NO 33 10 ~.t.2 47 I IJ84 0)NA :lS3 j 9 34 44 NA Bet.,1 L5,~. rot.a ~'.A h':1":.I"NA OJ,.~~ll""......1 Z-0.57~ RIIlJNII WHITEfISH 1981 '''''.e NA NA NA 'In NA 'In 1M2 _3 327.7~0 7!l1 0 HA 'Ill 0 I 1983 7~64 0 1079 1112 I s,)NA ~I 14 "'t_2 .2 1!l84 "NA 64u "14 :1:1 69 NA !:I.til ~;'~:2 lat..!....NA I 6'5:1 'In NA :.:il..~·vl '0'...12 ,-,.44 lO/lGNOSE SUCKERS 1911 NA 80 NA NA NA NO NA NO 19112 6:S:l:l 3:!U,41B 437 ~NA lUI ~2 1983 7.13 0 434 447 ~14 NA I":s n.t.:2 .,. 1984 "NA ~I:S 0 '5 7 12 NO Bet.:I 2S9 Tot.l NA NIl 4 19 7 'In 'In 5urviv~12 0.63 NIl •Iiclt IPpllcable. JlCg.[;;;;;;.'\'l.~~ I '--~'I " 'I I I -1 1 1 ))1 1 -<II j i 1 ..)1 1 j J j ]j Appendix Table 0-2.Bailey's deterministic method-population,survival and recruitment from a three-catch.. experiment.Fish tagged were captured in the middle river from 1981-84. Date Population Recruits Fish Newly Fish c:hec:ked Rec:apt.ur"es Marked IMOlrks marked at: 1981 1982 92 NA 144 NA NA 179 7 NA .312 ;2 42.43 4.081296 1027 RAINBOW TROUT,1981-83 1981 1982 1983 S.D. Surviva12 0.94 5.0 0.85 RAINBOW TROUT,1982-84 1982 1983b1982· 144 NA NA Wi198310091.2'7 274 312 4 NA \1984 197 1 16 \.;oJ S.D.1009 0.90 Surviva12 0.22 S.D 0.22 ARCTIC GRAYLING,1981-83 1981 19821981 49 NI-l NA NA198222"73 3.34 400 410 6 NF,1983 913 :;~30S.D~1496 1.93 Surviva12 1 ..05 S.D 0.84 Appendix Table 0-2 continued. O.:ate Popul ilti on Rec:rui ts Fish Newly Marked F.ish checked IMarks ReCilpture. marked'at; ARCTIC GRAYLING.1982-84 1982 1983 5800 1984 1.92 1982 1983 400 NA Nt\11lf.\ 765 913 :':;0 N'~ 583 9 34 S.D.2:~::54 0.67 Burvi \.tl;:\l.'~::0 ..55 S.D 0.30 ROUND WHITEFISH,1981-83 1981 1982 a 1981 4EJ NA NA Ni\1982'1(1616 O.I:H:)'720 751 (I Nf.\\1983 11 "12 1 :::)0-t S.D.10616 0.00 Survi val :;)(J II ~59 S.D'U.59 ROUND WHITEFISH.1982-84 1982 1983 1982 720 NA NA Nf.\ 1983 6:,~04 1.82 1079 1172 50 Nf.\ 1984 640 14 =.r;::,:j S.D.2(i06 0,,52 Burv 1 V,'\12 0.40 S.D O.1f.~ J J •J J )I 1 1 f -J )))J 1 ]j 1 )}1 } Appendix Table 0-2 continued. lONGNOSE SUCKERS,1981-83 1981 1982 8137 1983 s.o.6548 0.68 0.54 eo 418 NA 437 447 1981 NA '.,'"2 1982 NA NA 14 Surviva12 1.05 8.0 0.93 lONGNOSE SUCKERS,1982-84 1982 1983 1982 418 NA NA NA 1983 8101 1.13 434 44'7 14 NA 1984 215 :5 '7 S.D.466'7 0.51 ~SLirvivca12 0.78,S.D 0.55V', S.D.=Standard deviation. NA =Not applicable. DRAFT/PAGE 3 4/30/85,5/1/85 NUM3B/Appendix D fish recaptured with tags to the total number of fish recaptured (recaptured tagged fish plus those showing a tag scar).Refer to Appendi x A for further methods.Tag sca rs were not recorded for fi sh captured in 1982 so actual tag retention rates are unavailable for that year.However,since retention rates are known for 1983 and assuming there was little change in retention rates between 1982 and 1983,1983 retention rates were applied to recaptures made in 1982. - - DRAFT/PAGE 4 4/30/85,5/1/85 NUM3B/Appendix D Rainbow trout:100%tag retention rates for years 1981-1984. Arctic grayling:69.4%tag retention rate for year 1982 Year Tagged Actual No.Recaps Adjusted No.Recaps 1981 4 6 1982 31 44 Total 35 50 Arctic grayling:69.4%tag retention rate for year 1983 Year Tagged Actual No.Recaps Adjusted No.Recaps 1981 2 3 1982 21 30 1983 14 20 Total 37 53 Arctic grayling:75.9%tag retention rate for year 1984. Year Tagged Actual No.Recaps Adjusted No.Recaps 1981 1 1 1982 7 9 1983 26 34 1984 8 11 Total 42 55 Round whitefish:77.5%tag retention rate for year 1982. Year Tagged Actual No.Recaps Adjusted No.Recaps 1981 0 0 1982 32 41 Total 32 41 Round whitefish:77.5%tag retention rate for year 1983. Year Tagged Actual No.Recaps Adjusted No.Recaps 1981 1 1 1982 39 50 1983 32 42 Total 72 93 0-7 Round whitefish:83%tag retention rate for year 1984 DRAFT/PAGE 5 4/30/85,5/1/85 NUM3B/Appendix 0 - Year Tagged Actual No.Recaps Adjusted No.Recaps 1981 0 0 1982 12 14 1983 46 55 1984 18 22 Total 76 91 Longnose suckers:100.0%tag retention rate for years 1982 and 1983. Year.Tagged Actual No.Recaps Adjusted No.Recaps 1981 0 0 1982 5 5 1983 6 7 1984 2 2 Total 13 14 - - APPENDIX E Middle River Index Site Catch Data and DRAFT/PAGE 1,5/2/85 4/12/85,4/30/85 NUM3/Appendix E ~, - Descriptions,and Spawning Rainbow Trout Habitat Data - - - - 1 J »1 J 1 )1 i 1 ~J 1 )1 1 DRAFT /PAGE 1 4/24/85,4/30/85 NUM3IT.ble E-l Appendix Table E-l Boat electrofishing catch and catch per unit effort (CPUE)of four resident fish species .t 13 index sites in the midddle reach of the Susitna River in 1984. CPUE is in p.rentheses,and the units are catch per minute. River MAY JUN JUN JUL JUL AUG AUG SEP SEP OCT TOTAL Locati on Mile Sped es 16-31 1-15 16-30 1-15 16-31 1-15 16-31 1-15 16-30 I-IS CATCH Tri butary Mouth'Sites Lane Creek 113.6 R.inbow trout 5 (0.2)0 0 0 -0 1 (0.3)--6 Arct Ie gr.yl i ng 16 (0.5)4 (0.7)3 (0.4)-3 (0.8)-0 2 (0.6)~-2& Round whi tefi sh 5 (0.2)5 (0.8)2 (0.3)-2 (0.5)-1 (0.3)2 (0.6)--17 long nose sucker 0 2 (0.3)0 8 (2.0)-9 (3.0)2 (0.6)- -21 Skull Creek 124.7 Rainbow trout 1 (0.3)-0 -0 -0 --.1 Arct i c grayli ng 5 (1,4)-3 (0.6)-2 (0.7)·6 (3.0)---16 Round ..hi tef ish 1 (0.3)-2 (0.4)-0 -1 (0.5)---4 Longnose sucker 0 -0 -3 (1.0)1 (0.5)-- - 4 Fourth of Ju 1y Creek 131.1 Ra i nbow trout 2 <0.1)1 (0.1)2 (0.3)-1 (0.2)·1 (0.2)1 (0.1)1 (0.11 -9 Arct i c gr.y 1;ng 0 0 1 (0.1)-2 (0.4)-1 (0.2)7 (0.5)1 (O.ll -12 Round whitefish 4 (0.1)4 (0.5)6 (0.8)-7 (1,3)5 (0.9)10 (0.8)1 (0.1)37 Longnose sucker 0 0 0 -3 (0.5)-2 (0.4)11 (0.6)0 -16 Indian River 136.6 Rai nbow trout 6 (0.11 1 (0.1)0 -3 (0.4)-1 (0.1)9 (0.3)6 (0.3)0 26 Arctic grayli ng 10 (0.2)1 (0.1)7 (0.6)-6 (0.8)·5 (0.6)47 (1,4)26 (1,6)0 104 0\Round whitefish 21 (0.3)2 (0.3)33 (3.0)-6 (0.6)2 (0.3)16 (0.5)45 (2.5)11 (0.9)136 Longnose sucker 0 0 1 (0.1)-6 (0.8)-4 (0.4)0 1 (O.ll 1 (0.1)13 I J.ck Long Creek 144.5 R.i nbow trout 0 0 0 -0 -0 0 0 0 0.-Arct ie grayl'ng 2 (0.1)1 (0.2)5 (1.0)10 (2.9)-0 1 (0.1)0 0 19 Round whi tefi sh 7 (0.3)7 (1,3)25 (5.0)5 (1.4)2 (1.3)12 (1.6)18 (3.3)0 76 Longnose sucker 0 0 2 (0.4)-0 0 1 (0.11 0 0 3 Port.ge Creek 148.6 Ra i nbow trout 1 (0.0)1 (0.1)1 (0.1)-0 -0 6 (0.5)1 (0.11 1 (0.11 11 Arct i c gray1 i ng 28 (0.6)8 (0.7)6 (0.8)-7 (0.7)·19 (1.0)18 (1.6)7 (0.4)9 (0.9)104 Round whi tefi sh 25 (0.5)9 (0.8)21 (2.11 5 (0.5)-7 (0.7)6 (0.5)19 (1.0)9 (0.9)101 Longnose sucker 0 0 0 -11 (1.1)-5 (0.5)2 (0.2)1 (0.1)0 19 Sl0!!ilh Sites Whiskers Creek Slough 101.2 Ra i nbow trout 3 (0.11 0 .1 (0.3)0 1 (0.2).-5 f10uth Arctic gr.yl;ng 38 (1,2)0 1 (0.3)-I (0.3)1 (0.21 41 Round whitefish 3 (0.1)-0 0 ·1 (0.3)2 (0.3)--6 Longnose sucker 1 (0.0)1 (0.1)1 (0.3)5 (1.4)0 .8 Slough 6A 112.3 Ra i nbow trout 0 -0 0 0 -0 Aret;c grayl i ng 4 (0.3)-0 -0 0 --4 Round whi tet;sh 3 (0.2)-3 (0.3)0 1 (0.2)--7 Longnose sucker 1 (0.1)-3 (O.~)-0 -0 -4 DRAFT /PAGE 2 4/24/85,4/30/85 NUM3/Table E-l Appendix Table E-l (Continued). River MAY JUN JUN JUL JUL AUG AUG SEP SEP OCT TOTAL Location Mile Specie~16-31 1-15 16-30 1-15 16-31 1-15 16-31 1-15 16-30 1-15 CATCH Slough 8A 125.3 Rainbow trout 7 (2.0)-0 -0 -0 1 (0.2)--8 Arctic grayling 0 -0 0 -I (0.3)0 --1 Round whi tef Ish 0 .0 ·0 1 (0.3)3 (0.5)-4 Longnose sucker 0 -1 (0.2)0 -0 0 --1 Slough 20 Mouth 140.1 Rainbow trout -0 0 -0 -0 0 1 (0.1)-1 Arctic grayling 2 (0.5)0 0 -0 6 (1.2)0 -8 Round whi tefi ~h -4 (1.0)2 (0.6)·0 .0 0 0 -6 Longnose sucker -1 (0.3)0 0 -1 (0.5)5 (1.0)0 -7 Ma i n~tem 5i tes ~Susitna Mainstern 137.3-Rainbow trout 2 (0.2)0 0 0 0 0 0 -2 West Bank 138,3 Arctic grayling 2 (0.2)1 (0.1)22 (1.8)-21 (2.0)-6 (0.6)13 (1.2)7 (0.6)-72 I Round wh;tef;~h 2 (0.2)8 (0.71 7 (0.6)-2 (0.2)-4 (0.4)7 (0.6)17 (1.4).47 Longnose sucker 0 0 4 (0.3)-4 (0.4)0 2 (0.2)1 (0.11 -11 ~Sus;tna Mainstem 147.0-Rainbow trout 0 0 0 0 -0 2 (0.1)0 1 (0.0)3 148.0 Arctic grayling 9 (0.2)17 (0.6)2 (0.3)-'0 -0 23 (0.9)5 (0.2)4 (0.1)60 Round whitefi~h 34 (0.9)56 (2.0)9 (1.2)1 (0.2)4 (0.8)33 (1.3)25 (1.11 16 (0.5)178 Longnose sucker 0 3 (0.11 1 (0.1)·2 (0.3)0 5 (0.2)4 (0.2)2 (0.1)17 Mainstern 150.1 Rainbow trout 1 (0.1)0 ---3 (0.2)11 (0.5)4 (0.2)19 Arctic grayling 16 (1.0)2 (0.41 -4 (0.2)2 (0.11 0 24 Round whi tefi sh 0 1 (0.2)-·.--20 (1.0)12 (0.9)9 (0.5)52 Longnose sucker 0 0 ---2 (0.11 0 0 2 0.0 =Trace. J dJ J I ))1 1 )])1 j »)1 )}] DRAFT /PAC£1 4/24/85 NUM3/Table £-2 l Appendix Table £-2.Habitat characteristics and measurements taken at spawning rainbow trout sites in 1984. ( Locati on RM TRM Date Depth (ft) Water Mean Velocity (ft per sec) Temp·c Water Quality 00 Conducti vity pH mg/l umhos/cm Substrate,Comments (rJ r vJ Portage Creek Portage Creek (along the east shore) 148.8 2.3 148.8 5.1 6/2 6120 4.5 0.2 " 6.4 7.6 7.1 12.4 11.8 96.0 66.0 large gravel 70t small cobble 25t bedrock 5\ large gravel 90t boul ders 10\ Seven spawning fish were captured on June 1 and"5 from a school of approximately 30 adult rainbow trout.Five of these fish were radio tagged. This site is characterized by a pool with a tribu· tary outlettlng from the west.The pool's maxI- mum depth was estimated at 20 ft.Most spawning probably occurred in the pool near the side tributary's outlet where estimated water veloci- ties were 0.5-1.5 fps and depths 1.5-2.5 ft. One spawning fish was captured and radio tagged on June 20.One other spawning radio tagged fish was located at this site during this time.This site is characterized by low velocities due to large boulders and by a small tributary flowing Into It.Most spawning probably occurred along the east side versus the west side.Water veloci- ties were much greater along the west side. Fourth of July Creek Unnamed side tri butary outletting at TRM 0.7 of Fourth of Jul y Creek 131.1 0.7 0.5 6/3 6/5 2.8 1.4 2.5 0.5 9.3 12.2 6.8 11.4 21.0 large gravel 10\ sma 11 cobb 1e 60\ large cobble 20\ small gravel 60\ large gravel 40\ One post-spawned fish was captured and radio tagged on June 3.Two spawning radio tagged fish were also located at this site during this time. This site is characterized by a side tributary out- letting into'it on the east side. One pair of spawning fish were observed.The male was captured and radio tagged.Other fish were observed but it was unknown if they were spawners.This side tributary was 10-15 feet wide.Limited undercut banks provided primary cover . •-=No measurements taken. DRAFT/PAGE 2,5/2/85 4/12/85,4/30/85 NUM3/Appendix E INDEX SITE DESCRIPTIONS (the other nine site descriptions are provided in ADF&G 1983f) Skull Creek -Mouth Skull Creek is a small clearwater tributary with a summer discharge of approximately 50 cfs which empties into the east side of the Susitna River at RM 124.7 (R&M 1982).The mouth of the creek is characterized by shallow water depths,low water velocities and small cobble sub- strate.No object cover other than the substrate usually occurs. Susitna Mainstem -West Bank The index site Susitna Mainstem -West Bank is along the west bank of the mainstem Susitna River between RM 137.3 and RM 138.3.This site is characterized by low to moderate water depths and velocities,and has large cobble substrate.No object cover other than the substrate usually occurs. Susitna Mainstem This mainstem index site is between RM 147.0 and RM 148.0 and includes both sides of a large island.The area along the east and west banks of the mainstem river is characterized by steep banks and moderate water depths and velocities.Along the island,the shorelines are gently f.,-r - - - sloping and water velocities are'low to moderate. DRAFT/PAGE 3,5/2/85 4/12/85,4/30/85 NUM3/Appendix E The substrate at this - - - - site is predominately large cobble which acts as the only object cover. Susitna Mainstem -Eddy This site is a back eddy along the east bank of the mainstem Susitna River at RM 150.1.The area is characterized by steep banks,moderate water depths and low velocity.The substrate is sand and rock. .... I " GRAFT PART 4 DRAFT/PAGE 1,5/1/85 4/29/85,4/25/85 NUM4/Abstract !""'" i - -, Time Series Analysis of Juvenile Salmon Outmigration,Discharge,and Turbidity in the Susitna River,Alaska CrtAFT - CrtAFT DRAFT/PAGE 2,5/1/85 4/29/85,4/25/85 NUM4/Abstract TIME SERIES ANALYSIS OF JUVENILE SALMON OUTMIGRATION, DISCHARGE,AND TURBIDITY IN THE SUSITNA RIVER,ALASKA Report No.7,Part 4 by:Stephen S.Hale Alaska Department of Fish and Game Susitna Aquatic Studies Program 620 East 10th Avenue,Suite 302 Anchorage,Alaska 99501 ABSTRACT ,,"-Ouring ""i \middle I. \ the three years of study of juvenile salmon outmigration from the [=r or e !a..-+l ~ reach of the Susitna River,a correspondence has been noted between the peaks of river discharge and the peaks of outmigration. Further investigation of the relationship of outmigration to discharge was required because two large hydroelectric dams have been proposed for an area above the rearing area of salmon.These dams will markedly change the downstream discharge and turbidity regimes,factors which~ ! \influence not only salmon outmigration,but almost all fish species and ,1 "'·Jjl~stages.j Box-Jenkins model s were developed for the 1983 and 1984 '''-''''.".-~.".'.-.-.-../' time series of river discharge,turbidity,and chinook and sockeye salmon fry outmigration rates in order to statistically describe the natural conditions.Bivariate transfer function models were constructed_?,l ~~tV....ai\... for turbidity and outmigration rate wA+efl explain present values of these variables in terms of their own past values as well as past values of discharge.The time series examined were described by relatively ;;'::'--'''-"" si e model s usi ng mostly fi rst-order autoregressi ve terms./Al though \, _.,','-~ the time series plots of discharge and outmigration appeared to be ) J;' different between the two years,the underlyi ng stochasti c processes / which generated these series'were the same. f;RAFT - CRAFT TABLE DF CONTENTS ABSTRACT LIST OF FIGURES 1.0 INTRODUCTION 1.1 Time Series Analysis 1.2 Applications of Time Series Analysis 1.3 Obj ecti ves 2.0 METHODS DRAFT/PAGE 1 4/29/85,4/25/85 NlIM4/Table of Contents 2.1 Time Series Models 2.2 The Data 2.3 Identification and Estimation of Time Series Models 2.4 Transfer Function Models 3.0 RESULTS 3.1 Univariate Model for Mean Daily Discharge 3.2 Univariate Model for Turbidity 3.3 Univariate Model for Age 0+Chinook Salmon Outmigration 3.4 Univariate Model for Age 0+Sockeye Salmon Outmigration-3.5 Discharge -Turbidity Transfer Function Model 3.6 Discharge -Chinook Transfer Function Model 3.7 Discharge -Sockeye Transfer Function Model ,~ 4.0 DISCUSSION 5.0 ACKNOWLEDGEMENTS 6.0 LITERATURE CITED ,~ _____w ----_ - DRAFT/PAGE 1,5/1/85 4/29/85,4/25/85 NUM4/List of Figures LIST OF FIGURES ~r~Figure 1 !""'"2 o!'l!'~3 4 5 6..... 7 §,i;a, 8- 9 10 ,~ 11 ,.-. 12 13 14 Map of the Susitna basin study region. Discharge,turbidity,and chinook and sockeye salmon outmigration rate,1983. Discharge,turbidity,and chinook and sockeye salmon outmigration rate,1984. Susitna River discharge time series at the Gold Creek gaging station,1983 and 1984. Plots of autocorrelations and partial auto- correlations for 1983 discharge time series. Log-transformed discharge time series,1983 and 1984 • Plots of autocorrelations and partial auto- correlations for 1983 log-transformed discharge time series. Spectrum of 1983 discharge time series. Plots of autocorrelations and partial auto- correlations for 1984 discharge time series. Plots of autocorrelations and partial auto- correlations for 1984 log-transformed discharge time series. Turbidity time series at Talkeetna Station, 1983 and 1984. Plots of autocorrelations and partial auto- correlations for 1983 turbidity time series. Differenced turbidity time series,1983. Plots of autocorrelations and partial auto- correlations for differenced 1983 turbidity time series. 15 Age 0+chinook salmon outmigration rate time series,1983 and 1984. 16 Plots of autocorrelations and partial auto- correlations for 1983 chinook salmon outmi- gration time series. 19 DRAFT/PAGE 2,5/1/85 4/29/85,4/25/85 NUM4/List of Figures 17 Log-transformed age 0+chi nook salmon outmi- gration rate,1983 and 1984. 18 Plots of autocorrelations and partial auto- carrel ati ons for log-transformed 1983 chi nook salmon outmigration time series. Plots of autocorrelations and partial auto- correlations for log-transformed 1984 chinook salmon outmigration time series. 20 Age 0+sockeye salmon outmigration rate time series,1983 and 1984. - 21 Plots of autocorrelations and partial auto- correlations for 1984 sockeye salmon outmi- gration time series. 22 Plot of cross correlations between the resi- duals of the ARMA (1,1)discharge model and the prewhitened turbidity time series,1983 data. 23 Plot of cross correlations between the residu- als of the ARMA (1,1)discharge model and the prewhitened chinook salmon outmigration time series,1983 data. 24 Plot of cross correlations between the residu- als of the ARMA (1,1)discharge model and the prewhitened sockeye salmon outmigration time series,1984 data. - On-AFT DRAFT/PAGE 1,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Hale 1.0 INTRODUCTION During the course of examining the plots of daily catch rate of out- migrating juvenile salmon at the Talkeetna Station outmigrant traps,it was noted that there was an apparent correspondence between the peaks of the time series of mean daily discharge and the time series of salmon outmigration (ADF&G 1983;Roth et al.1984;Roth and Stratton 1985). Correlation analysis showed that there was a relatively strong relation- ship between discharge and the outmigration rates of certain species/age classes of salmon.The term outmigration rate is used here to mean the -I number of outmigrating fry captured at the traps per hour,not the distance travelled per hour.This relationship is not simply a matter of a greater volume of water being fishecL at higher discharges.The correlations of catch rate of age 0+salmon with water velocity at the mouths of the traps were not significantly different from zero (Roth et al.1984,Appendix A). 2 D Skeena River in British Columbia (with regard t~:"lateral dis-\"",~,,.,' '""/£;t\"'\...~_/)('J".>'-' tribution),the greater the density of sbcj<eye and pink fry per unit '"""--"",,,- volume of water.The correspondence betwe~!1-lfi·;~hiir~fera:t-eand salmonid ---;:,""-"'--"'-",,,\.,-,-,'.....__."..__.~---~,-~-~.~..--~--''''-'''','--"..-_..,./'; outmigration has ~sobeen'~;ported by other investigators (Cederholm'"/,--"'. and Scarlett 1982 -coho salmon;Congleton et al.1982 -chum and Similarly,McDonald (1960)found in the ~-,.." that the greater the water veT~Ci~ ',,-._-,.// chinook salmon;Godin 1982;Grau 1982;Solomon 1982b).The selective advantages of this behavior,according to Solomon (1982b),include easier passage over long distances or shallow areas and protection from predators provided by increased turbidity and by the large numbers DRAFT/PAGE 2,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Hale resulting from a coordinated mass migration in response to an environ- mental cue. There are probably two mechanisms which account for this relationship in the Susitna River.One is that the fish,which have gradually become physiologically ready for outmigration by growth and in response to photoperiod and temperature,are stimulated by a rise in mainstem discharge to begin that outmigration (Grau 1982).The second mechanism is that high flows physically displace the fish downstream.This latter mechanism may frequently occur for fry rearing in side sloughs,particu- larly for chum salmon (Oncorhynchus ketal and sockeye salmon (Q.nerka). The natal sloughs for many chum and sockeye salmon have berms at the heads which prevent water from the mainstem from entering the site at low levels of discharge.When high flows occur,the slough heads are overtopped and the fry which had been rearing in low velocity water are subjected to a strong current. Because two large hydroelectric dams have been proposed for the Susitna River in an area upstream of the rearing areas of the juvenile salmon (Fi g.1)and because these dams woul d rna rkedly alter the natural di s- charge and turbidity regimes,it is necessary to quantify the relation- shi p between the di scharge and turbidity regimes and the outmi gration patterns of the juvenile salmon.After the dams begin operation,the annual patterns of river discharge and turbidity level would be smoothed -both would be lower than normal in the summer and higher than normal in the winter.Also,the high frequency (daily)oscillations of these two time series would be dampened;there would be less day to day w l ~-}1-})j!i J i --1 }]#~ 1 j/j •10 Rive'mil'Incnmanll Figure 1.Map of the Susitna basin study region. Data Center). (Source:Arctic En~ironmental Information DRAFT/PAGE 3,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Hale A post-project d~~~h~~9~regime which will have no delete- rious effects on salmon outmigration needs to be determined.If such a ...."".........................~--_..-._,-~,.._-~"._~~"-,_~.~,---,.•..-__.-----_.'.• discharge regime is not feasible,therJth~~ffg_cts of a given discharge ._••..c,J --_.",.,-~.._"""'""'''.-....."··~-._~_J_.~."O'"._"."_.~~._,,_.~.~£'"-.,,,.~.._ ....-_..-.."-'""....... There are many factors other than discharge and turbidity wnte~affect the outmigration timing of juvenile salmon including time of year,size of fish,photoperiod,light intensity,and temperature (Brannon and Salo 1982);however,discharge and turbidity bear further investigation because of the changes in these two variables which would be caused by the proposed dams.Potential negative effects of an altered flow regime include accelerated or delayed timing of outmigrations.Changes in outmigration timing may place the fish in their rearing areas at an unfavorable time from the standpoint of food supply,which could cause reduced survival (Hartman et al.1967).Lower discharge levels can result in a shorter distance covered per day (Raymond 1968).Decreasing mainstem flows can lead to stranding of fish in pools which have been isolated from the mainstem (Solomon 1982a).Lower flows and clearer water than normal may also result in increased predation (Stevens and Miller 1983). Turbidity level in the Susitna River probably does not have much effect on the daily number of fry which outmigrate or on the initiation of outmigration.In clear water streams,however,an increase in turbidity level can directly increase the number of outmigrating salmon by providing cover from predators (Solomon 1982b).Turbidity level in the Susitna River does change outmigration timing because fry in turbid - - - d (~'"/J n::-.~.-~~~~~W 1Jvt-/~~~..- Lf DRAFT/PAGE 4,5/1/85 4/19/85,4/25/85,4/29/85 \d ,>~~ NUM4/Hale '1L~.~~"Jr '" \2 water outmigrate during the day as well as during the night (Godin 1982;~ Roth et al.1984).Clearing of the water could force the fry to shift to a nocturnal outmigration to avoid predators. T'rl---t-· However,this would be - - - of marginal benefit for fry during the~co~tinuous daylight in June and July at 63°N latitude.f')~"'\ To avoid or alleviate the above problems,it is necessary to understand the mechanisms producing the present discharge,turbidity,and outmigration regimes.Knowledge of the discharge outmigrationl relationships will be useful in trying to establish a post-project flOW!I )0\ regime which will not interfere with the natural outmigration timing.1 ~. In addition to the effects of'discharge and turbidity level on juvenile salmon outmigration,the effects of these two variables on juvenile salmon rearing and overwintering must be considered because changes in river flow can affect the survival rate of young salmon (Stevens and Miller 1983).The effect of variations in discharge on juvenile salmon habitat of the Susitna River has been modelled (Hale et ale 1984)as has the relationship with turbidity level (Suchanek et ale 1984).The current discharge and turbidity regimes that are driving these models must be statistically described. 1.1 Time Series Analysis The statistical methods collectively known as time series analysis are a logical choice for analyzing the present discharge,turbidity,and outmigration regimes.A time series is a collection of observations " DRAFT/PAGE 5,5/1/85 ~ 4/19/85, 4/25/85,4/29/85 NUM4/Hale - ordered in time such as daily water temperature measurements.Time series are shaped by both deterministic and stochastic (random)events. Future values have a probability distribution which is conditioned by past values.Random events (or "s hocks")operating on the time series have a "memory",that is,the effect of these di sturbances may be apparent for several time units after the event occurred. Time series analysis consists of removing deterministic trends from a time series so that the values fluctuate around a mean level.A trans- formation may be necessary to ensure a constant variance.The random .- processes that generated the observed series can then be mathematically defined.The residuals left over after this model is fitted should be "white noise"(completely random)if the model is adequate.With a white noise time series,past values provide no linear information about future values. Time seri es can be passed through a mathemati ca 1 fi 1ter whi ch changes the form of the input series.A "l ow pass filter"dampens high frequency perturbations and allows low frequency.perturbations to pass unchanged. This is useful in smoothing noisy time series so that the basic pattern ~ may be more readily observed.The low pass von Hann filter was applied to the salmon outmigration time series in Roth and Stratton (1985). High pass filters are used when it is desirable to remove obvious (low frequency)trends in order to focus on the high frequency events. Time series analysis allows the construction of mathematical models using only the information contained in the t'ime series itself.For - ..... -. - DRAFT/PAGE 6!5/1/85 4/19/85!4/25/85!4/29/85 NUM4/Hale example!although the discharge time series results from several in- dependent variables including rainfall!air temperature!and solar inso- lation on the glaciers!it is not necessary to quantify these inputs in order to model the output (di scha rge).Informati on on the effects of all the inputs is already contained in the past history of the discharge record. Time series analysis includes frequency domain (spectral analysis)and time domain (parametric)problems.Spectral analysis is concerned with transforming a time series with a Fourier transform to a sum of sines and cosines (see Priestley 1981)and is appropriate with periodic series such as the classical example of the Canada lynx/snowshoe hare ten year cycle (Bulmer 1978).Methods for time domain problems (or Box-Jenkins models)are referred to as ARIMA (autoregressive!integrated!moving average)models (Box and Jenkins 1976).ARIMA models have been used extensively in economic forecasting (Nel son 1973;Granger and Newbold 1977)• These models can be extended to what Box and Jenkins (1976)have called transfer function models.This is a model where an output series is a function of one or more independent input series as well as its own past history.Time series models do not require information on input series but if such information is available!then it may be possible to obtain a model with more predictive power. DRAFT/PAGE 7,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Hale There are no replicates in time series analysis.An observed series is one realization of all possible time series which could have been generated from a random process.Time series analysis examines the nature of the probabl isti c process that generated the observed seri es. The mode]should have similar'properties to the generating mechanisms of the stochastic process (Granger and Newbol d 1977).Then,one can fonn summary statistics about the series and make inferences about the nature of the stochastic process.After a model has been developed,it can be used to test some hypothesis about the generating mechanism of the time series,to forecast future values of the series,or to make decisions on how to control future values of the series (Granger and Newbold 1977). 1;2 Applications of Time Series Analysis Time series analysis has been extensively used in examining physical data,particularly in oceanography.Salas and Smith (1981)demonstrated that ARIMA model s can be used to model the time series of annual flows in streams.Srikanthan et ale (1983)analyzed the time series of annual flows in 156 streams in Australia and found that most of the ones which were non-random could be modeled by an autoregressive model.Time series models have also been used to examine the effect of the Aswan dam on the discharge of the Nile River and the effect of a hydroelectric dam on the discharge regime of the Saskatchewan River (Hipel et al.1978). - - Time series analysis methods have been also been used in examining time ~ series of abundance and catch in marine fisheries (Van Winkle et ale 1979;Botsford et ale 1982;Peterman and Wong 1984;and Taylor and r DRAFT/PAGE 8,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Hale Prochaska 1984).These ~thods have been used by Saila et ale 1980; Mendelssohn 1981;Stocker and Hilborn (1981),Kirkley et al.(1982),and Jensen (1985)for forecasting future abundance or catch of marine fish stocks.Mendelssohn (1981)used transfer function models in addition to univariate Box-Jenkins model s to forecast fi sh catch.Botsford et al. (1982)focused on searching for causal mechanisms of observed cycles in salmon fisheries in California rather than on defining models for the fisheries. Applications to freshwater fish ecology problems are much more limited. Saila et ale (1972)used time series methods to cross correlate upstream migration activity of the alewife to solar radiation and water tempera- ture.O'Heeron and Ellis (1975)considered a time series model for "judging the effects of reservoir management on fish.Applications of spectral analysis to ecological problems have been reviewed by Platt and Denman (1975)and time series analysis in ecology was the subject of a symposium proceedings edited by Shugart (1978). 1.3 Objectives The objective of this paper is to develop mathematical models for the times series of mean daily Susitna River discharge at the Gold Creek gaging station (river mile 136.7),daily turbidity level,and daily outmigration rates of chinook salmon (Oncorhynchus tshawytscha)and sockeye salmon (Q.nerka)at the Talkeetna Station outmigrant traps (river mile 103.0)during the open water seasons of 1983 and 1984. Because time series analysis can provide an efficient summarization of a q ----------------~------.;.,...,-------------------- DRAFT/PAGE 9,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Hale data set by a few parameters (Hipel et al.1978),these models will be used to statistically describe the present conditions.The discharge and turbidity information will be useful for examining their relation- ship with salmon fry outmigration as well as with other species and life history events.In addition,transfer function models for discharge- turbidity,discharge-chinook outmigration and discharge-sockeye outmi- gration are developed to describe the relationship between these vari- abl~and to be used as a possible technique to forecast future values or 1\ to examine the probable effects of the proposed dams. Turbidity was chosen as a variable of interest because of its strong relationship with discharge,because of its importance in determining the distribution of rearing juvenile salmon (Suchanek et al.1984),and because trapping of suspended sediment from the glaciers in the dam impoundment would create a turbidity regime SUbstantially different from the present regime.Chinook salmon were chosen because this species rears in sloughs and side channels affected by mainstem discharge and b~i.rw.ok_~_~~_~ee!:_se 1ec.~~~~_~v~~~~~.:t:.iQIL'§.Q~£i.~.Q1..... the impact assessment study (EWT&A 1985).The sockeye salmon time series '<~_,~,_""..,..,...",~..~......-"~.,~._..~""._...c-."'."",,",,"",,,,,-,,,,,",,,,,-,>;,,,,~""~""'h'"''''''~'~''''''_'''''''''''''''?JD'-'-.l'_''''~'~_''''''''''''''''''-''-''··'·"·>'''''''''''''''-<"."""""~"~" was chosen because mainstem discharge affects sloughs which are both natal and rearing areas for this species.While chinook salmon spawn mainly in tributaries in this system,sockeye salmon spawn mostly in mainstem sloughs. to - DRAFT/PAGE 10,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Hale 2.0 METHODS All of the time series work was done using the BMDP statistical package (Dixon et al.1981).Univariate models were developed for the four time seri es:di scharge,turbidi ty,and chi nook and sockeye outmi gratian. Then,transfer function models were constructed for turbidity and the two salmon time series as output with the discharge series as input. 2.1.Time Series Models Box-Jenkins models can be summarized as follows (Box and Jenkins 1976; McCleary and Hay 1980;Chatfield 1984).Suppose there is a time series Yot ' t =1..N.Then Yt is a moving average process of order q (or an MA(q)process)if .'.-+e'b'II t-~ - where 6 4 are constants and 6 0 =1.The term at is a ~~~~ cess.White noise consists of a series of random shocks,each dis- ~ tributed normally and independently about a zero mean with a constant variance.Yt is an autoregressive process of order p (or an AR{p) process)if where <Pi.are constants.This is similar to a multiple regression model except that y t is regressed not on independent variables but on past DRAFT/PAGE 11,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Hale values of itself.A first order autoregressive process,AR(l),has the form: Box and Jenkins (1976)define a backward shift operator Bas: For m =1, - lIIOI\ I or,the previous value. Using B,the AR(l)equation can be written: - l -C!>.B Time series resulting from a mixture of AR and MA processes are called ARMA(p,q)models and have the form: 1't -cp,1't-1 T '..+ep.,1t-p +Zlt + e l llt_1 +~..+e t a....C-t Ii. I~ DRAFT/PAGE 12,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Hale Using the backward shift operator B,an ARMA (1,1)may be written as: ARMA (p,q)models are appropriate only when the time series is station- ary.Stationary means that there is no systematic change in the mean or the variance over time and that there are no strictly periodic var- iations (Chatfield 1984);in other words,the mean,variance,and autocovariance are not dependent on time.Time series which are not stationary can sometimes be handled by "differencing"the series.Taking the difference of adjacent values gives a differencing order,d,of one: -- J - Such models are said to be "integrated"and are denoted by ARIMA(p,d,q) where p is the order of the autoregressive component,d is the order of differencing,and q is the order of the moving average component. Time series with seasonal variations,such as would occur in a multiple year series of daily water temperature measurements,can be made sta- tionary by seasonal differencing.For example,the value for April 15 of one year is subtracted from the value for April 15 of the following year,and so on for all days of the year. ________--~_l3 DRAFT/PAGE 13,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Hale It has been assumed above that the time series had a mean value of zero. With stationary time series which have a non-zero mean (which is the case for all time series discussed in this paper)the mean has to be subtracted from every Yi term.For example,the form of an AR(l)model would be: The autocorrelation function plays a major role in identifying and building time series models.A regular correlation coefficient measures the correlation between N pairs of observations on two variables.The autocorrelation coefficient is somewhat similar except that it measures the correlation between all observations of the same variable at a given distance apart in time (that is,between Yt and Yt~i\for all values of t).Also,the covariance is estimated only over N-k pairs of obser- vations (McCl eary and Hay 1980).Autocorrel ation coeffi ci ents at dif- ferent lags indicate the extent to which one value of the series is related to previous values and can be used to evaluate the duration and the degree of the "memory"of the process.The autocorrelation function (ACF)is the set of autocorrelation (AC)coefficients at different lags associated with a time series;a plot of the ACF is called a correlogram (Chatfield 1984). The ACF is defined as: - --covat-iall\ce (Yt )Yt:tA') Va 1--1 ~",ct:l r t ) DRAFT/PAGE 14,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Hale - and is estimated by: ~1\CYt -Y)(~t~-n t:.I .. • N N-A - A partial autocorrelation (PAC)coefficient measures the excess corre- lation at lag k which is not accounted for by an autoregressive model of order k-l.The set of PAC's at different 1ags associ ated wi th a time series is called.the partial autocorrelation function (PACF). There are three steps in developing an ARIMA model:model identifica- tion,parameter est"imation,and diagnostic checking (Box and Jenkins 1976).ARIMA model building is an iterative process.The first thing to do is to look at a plot of the time series.Time series that are not stationary must be made so by trend removal which can be accomplished by ei ther differencing the seri es or by polynomial regressi on.Exami nation of the autocorrelation function (ACF)and the partial autocorrelation function (PACF)of a stationary series helps to identify a possible ARIMA model.The next step is to estimate the parameters of the model and again examine the ACF and PACF plots,this time on the residuals from the model.This.process is repeated until the residuals show no significant AC's or PAC·s at any lag,which indicates that the residuals are generated by a white noise process. 15 DRAFT/PAGE 14.1,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Hale -2.2.The Data Mean daily 'discharge values for 1983 and 1984 (Fig.2,Fig.3)were obtained from the U.S.Geological Survey gaging station on the Susitna River at Gold Creek,river mile 136.7 (Still et ale 1984;U.S.Geolog- ical Survey provisional data,1984).The time series frame examined was May 18 to August 30 (l05 observations).Discharge levels begin to decline in September when glacier melting decreases;hence,a longer series would not be stationary.Throughout this paper,the unit for discharge is one thousand cubic feet per second,except when logarithmic -transformation is used -then,the unit is cubic feet per second. Outmigration rate (Fig.2,Fig.3)was measured by two outmigrant traps, one on each bank,located at river mile 103.0 (Roth et ale 1984;Roth - and Stratton 1985).The rate is reported as number of fish per trap hour with catch from the two traps combined.Only age 0+fry were used in the analysis because the traps were not efficient at capturing age 1+ fry and,consequently,the numbers were low.Further,age 1+chinook and sockeye salmon have essentially completed their outmigration from this reach of river by the end of July so the time series are shorter. The chinook salmon time series for 1983 runs from May 18 (shortly after ice-out)to August 30 (when outmigration ;s winding down),a total of 105 observations.The 1983 sockeye salmon data was not examined.There were six days during the 105 day series when the outmigrant traps were J" ao,-----:::::-::=::-:-:--=-=-=-=-::------, SUSITNA RIVER DISCHARGE 10 O-lrm.......mrm~~~mrm~nnnnnm~"""'~nnmmnn:""'mn1 ,)un 1 Ju"15 Jut 1 Ju.1~Auq'Au9 15 1983 500,---------... SUSITNA RIVER TURBIDITY 100 O-!mr.......rmmmnn111TlTmm"mnnrmm......mTlTrmm"""'nnnmTITrmm........."".,j Jun 1 .hIn 1~.Jut 1 .Jut,~Af.I9 1 AU9 1~ 1983 '983 lIS .,------:-=c:=-::::--="..,.,,:-:-::-=-__--r-----, 17 AGE 0+CHINOOK 'I! IS 14 13 '2 "'0 "IS 7 I! "~ 3 2 1 O~1Tmtffm....;Tt;nr;:;;;;".mnnrmmnnnnnii:nmn"""';;;;;;;;."","'"""""'~ lIS ,-----.-;;;;;;;-:=c-~':':===-----.... 17 AGE 0+SOCKEYE'.,,, 14 13 12 11 10 "IS 7 a "4 3 2 1 o ~\Tnn~?""'~~mm:~nnm~~",m.nnm"'"""""'...,.;j;,llfI ..lui 1~AU9 1 Au.;1~ 1983 - Figure 2.Discharge,turbidity,and chinook and sockeye salmon outmigration rate,1983. 60...-------~----------_SUSITNA RIVER DISCHARGE '0 o -bmonmmm"""'onm"""'................lTn"m"""'nrmmrm"""'''''''''''''''.......rmnl ,JU"1 .Ju"1S Jut 1 .Jul tS ....U!J,Aut;15 . 1ge. ~-,-------------~=-------,SUSITNA RIVER TURBIDITY 400 '00 o-lnn.........""';'"","""",rmm""",,"",,,,,,,,,,,,,,,,,,,,,,mnrmrmrmm,,,,,,,mrm......ri Jun l'Jun 15 Jut I Jut 1:5 Au9 1 A4.l0 15 '91>4 "',..-----=-=:---=-=--.,,---------.'7 AGE 0+.CHINOOK,. HI '4 ,,J '2 "'0 g •7 S 5 4 ,J 2 I O-bm......."""''''''''m6nrmmrmmrmm............'''''''......nmn''''''''''''''''''''''''''''rmi Jun 1 Jun 15 .Jul 1 Jut 15 oI4.t9 1 AU9 15 '9104,.,-----------------------..., '7 AGE 0+SOCKEYE,. '5 14 1,J '2 "'0 g "7 "5 4 .J 2 • o -Mlimm,Tiftlm;",rmm"""'mnr........."""''''''''nrm....................;m;I"'''''mIll1 ",un 1 Jwn lS Jul 1 .lUI 15 AuQ 1 Ave;lS "84 Figure 3.Discharge,turbidity,and chinook and sockeye salmon outmigration rate,1984. l8 91 - - - - DRAFT/PAGE 15,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Hale not fished - a one day,a two day,and a three day period.Although values for gaps in time series can be estimated by a spline method,the gaps in the outmigration series are short enough so that a simple interpolation of values is sufficient (Sturges 1983). In 1984,the traps were continuously operated from May 14 to October 6. However,the seri es were cut off at the end of August in order to be comparable to 1983 and to achieve a stationary series.About 98%of the cumulative outmigration of age 0+chinook and sockeye fry in 1984 had occurred by the end of August (Roth and Stratton 1985). Daily water samples for turbidity (Fig.2,Fig.3)were taken at the outmigrant trap station and measured with an HF Instruments Model No. ORT-15S field turbidometer (Roth et al.1984;Roth and Stratton 1985). Uni ts are in nephel ometri c turbi dity units (NTU).Only the 1984 tur- bidity series was examined. 2.3.Identification and Estimation of Time Series Models The methods described above were applied to the four time series.The AC and PAC plots were examined to help identify possible AR and MA com- ponents.A tentative model was developed and the parameters estimated. The BMDP Box-Jenkins program estimates parameters by both the condition- all east squares method and the backcasti ng method.The estimates ~(' chosen for this paper were from whichever method gave the lowest residu-~ al mean square.Insignificant components were removed from the model.~ - DRAFT/PAGE 16,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Hale The residuals were checked to see if there was significant departure from the assumption that they were white noise.If not,the model was adequate. The time series of mean daily discharge from May 18 to August 30ap- pea red to be stationary so no differencing was done.A plot of the range of sub-groups of the series against the mean of the sub-groups (as suggested by Hoff (1983))indicated that a log transformation of the data would be helpful in stabilizing the magnitude of the fluctuations throughout the series;therefore,a model was also developed for the log ~ transformed data.As the turbidity time series was questionably station- ary,models were developed for both the original series and for a differenced series. Models were developed for the chinook and sockeye salmon outmigration ~ rate time seri es on both the raw data and on data transformed by 1n (x +1).The transformation In(x +1)was used to avoid taking 10ga- rithms of zero;there was zero catch on some days. 2.4.Transfer Function Models When there is an independent variable·which is also a time series,a transfer function model can be developed.This model consists of the transfer function component from the independent variable as well as the - DRAFT/PAGE 17,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Hale ARIMA component (or noise component)from the dependent variable (McCleary and Hay 1980)and can be represented as: --f (X ~-b )+ where:Yt is the output time series Xt is the input time series f(Xt.~)is the transfer function component Nt is the noi se or ARIMA component - Transfer function models can be bivariate (when there is one independent variable)or multivariate (more than one independent variable). The steps to take in developing a transfer function model (Box and Jenkins 1976;McCleary and Hay 1980;Dixon et al.1981)are:(1)develop an ARIMA model for the input series,obtaining the pre-whitened input (residuals),(2)filter the output series by the model for the input series,(3)cross-correlate the residuals from the first two steps,(4) identify the form of the transfer function component from the cross correlation function,(5)assuming the errors are white noise,estimate the values for the parameters,(6)identify an ARIMA model for the residuals,(7)if the ARIMA component is not white noise,combine the ARIMA component with the transfer function component to form a new model,(8)estimate the parameter values,and (9)examine the ACF and .21 DRAFT/PAGE 18,5/1/85 4/19/85, 4/25/85,4/29/85 NUM4/Hale PACF plots on the residuals from the new model to see if the model is adequate. Transfer function model s were developed in thi s fashion for di scharge/ turbidity,discharge/chinook outmigration,and discharge/sockeye out- migration.Only one input (discharge)was used.Multiple input transfer function models (L iu and Hanssens 1980)or multivariate time series models (Mendelssohn 1982)can be developed;but are substantially more complex. - - - - - - - DRAFT/PAGE 18.1,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Hale 3.0 RESULTS 3.1.Univariate Model for Mean Daily Discharge The time series of mean daily discharge during the summer of 1983 is shown in Fig.4;the log-transformed data are in Fig.6.ACF and PACF plots for the raw data are given in Fig.5 and for the 10g-transformed data in Fig.7.In all the ACF and PACF plots,the "+11 symbol on either side of the vertical axis indicates the 95%confidence interval.The first order autoregressive component is strong in both the raw and the transformed series.The ACF and PACF plots for the raw data indicate that a moving average component is required.Models containing various combinations of first and second order AR and MA terms were examined.Of the acceptable models identified,the model with the lowest standard errors on the parameter estimates and the least significant residuals was an ARMA(2,2).However,the ARMA(l,l)was nearly as good as the ARMA (2,2)so,in keep"ing with Box and Jenkins'(1976)advice that a par- simonious model (i .e.,the one with the fewest possible parameters)is desirable,the ARMA(l,l)is considered the "best ll model for the non- transformed data.Parameter estimates are: ~A ¢.=.992 with std.error of .0135 A 6 1 =-.580 with std.error of .0807 The model is: ..-, - - Susitna River Discharge.1984 60 IIlIII'l 50 -cc 400u Jl-iii' 8.1 30_:I 00~i£......-u :ii 20:I 0 10 8!t 0 Jun 1 Jun 15 Jul 1 Jul 15 AU9 1 AU9 15 -1984 Figure 4.Susitna River discharge time series at the Gold Creek gaging station,1983 and 1984. [;tiAFT Pl..OT OF AUTOC~O~R~R~E:.::L",A::.T~I"O",N:.::5,-_ -&.0 -.8 -.6 -..-.2 .0 .2 ••.6 .8 1.0 LAG CORR..----.----+----+----.----+----+----.----.----+----+ 1 C .•859 IXXXX,;XXl(IlXXXXXXXXIlIlXX ' 2 .627 •txxXXXXX.XXXXIlXXX 3 .444·•IXXXXXXXX.XX 4 .266 •IXXXXXXX • 5 .057 •IX • 6 -.122 •XXXI • 7 -.230 •XXXIlXXt • 8 -.288 •XIlXIlXXXI • 9 -.317 •.IlXXXXIlXXl 10 -.324 •XXXXXXXIlI II -.281 •XXXXXXXI ....... • • • • • •• • • ..- -_._-----------~.__._'.. .. 12 -.202 ..XIlIll(XI 13 -.134 ..XXXI 14 -.077 ..IlXl 15 .002 ..I ....,,16:..,,~.D?5~~"'~.-"';';"';·c.:·,,·-'...;.--....'rxx"--"----..,.,...,..-,-.._.--" '17 •094:.;;.~_:c_..'-_-I;-X:;:X:'-----_..'_:_-_--_--- 18 .055 IX 19 .014 •t 20 -.021 XI 21 -.061 •XXI 22 -.097 ..XXI 23 -.132 _.""'X"X::;X~I-_-__,_'C..-~'--~---_---- 24 -.149 ..XIlIlXI • 25 -.138 •XIlIlI 26 -.141 ..XXIlXI -27'-'-.160 ..XIlXXI 28 -.135 ..IlXXI 29 -.082 ..IlXI '30 --;:;030-'•X I JI .037 •IX 32 .11_8_____.__.__"__.!.)(~,,~_.. PLOT OF PARTIAL AUTOCORRELATION5 - -1.0 ~.8·~.6 -.4 ~.2 .0 ~2 .4 .6 .8 1.0 LAG;-.CORA.,....i---~:..+~:-..-.-.-:--~.~.--_.-.-:~~+---. I I .859 ..IXXXX.XXXXXXXXXIlXXXXXX 2 -.425 XXXXXX+-XXXXI +- .- - 3 .196 ..IXXXXX ..-.32:5 XXX.XXXXI • 5'"-.141 .XltXltl·.. 6 -.022 •XI + 7 -.013 ..I .. 8 -.023 •XI .. 9 -.009 • I .. 10 -.092 ..'JIlt I .... II .128 ~IXXl(.. 12 -.050 •XI • 13 -.031 ..XI • 14 .026 ..IX • ''--15'"".07''''--~,~=-•.--I'XX'.-_.,'.'..---.-...-". 16 -•.064 ..XXI • 17 -·.0f'>6 •XXI .. 18 -.158 "XXXXI • 19 .059 ..IX .. 20 -.093 •.XXI .. - 21 .067 IXX .. 22 -.015 • I • 23 -.107 •XXXI ., 24 .029 •IX .. 25 .021 IX • _2'""6.-_-_•..,2""0;:3...."'X.=.X::.X::.X.=.X~I..---:.-.-----~.-. 27 .036 •IX • 28 .050 •[X • 29 -.048 •Xl • -"jo';'1'29-•I xiol • 31 .040 • I X • 32 -.0_1_9 •.=.[•_ Figure 5.Plots of autocorrelations and partial autocorrelations for 1983 discharge time series. Log-transformed discharge.1983 12 11.5 11 10.5...... 0\..c ~~<>II 1015 '3 9.5 9 6.5 8 Jun 1 Jun 15 Juf 1 Jut H5 1983 Aug 1 Aug 15 LOG-TRANSFORMED DISCHARGE,1 984 12,.....-----------------------, 11.5 '1 'GJ 10.5 C>a:: ~100 !l1 0....., 3 9.5 9 8.5 8 - - - Jun 1 Jun 15 Jul 1 Jul 15 1984 Aug 1 Aug 15 Figure 6.Log-transformed discharge time series,1983 and 1984. C;1AFT PLOT OF AUTOCQRR€LATIOHS ---_._._---~_._-----~---_._---- -1.0 -.8 -.6 -._-.2 .0 ~2 "._.6 .8 1.0 LAG CORR.+----+----+----+----+----+----+----+----.----.----. + :t._ + • + _.+._---- e + ;tl'>-.I~O 27..-•.1....12... 2B -.1611 29 -.112 _..JQ._""..oSO_.__. 31 .023 32 .100 I '.849'+IXXXX+xxxxxXXXXXXXX1CXXX' '2 .•644'+U:XX'xxxx+XXXXXXXX 3 AS&__.________•.~.__ 4 .2aO +IXXXXXXX + 5 .081 IXX + _......b __,~.D...SL.."--_.__._. .~_,_..x.x..1..~~._+_ 7 -.208 +XXXXXI 8 -.276 +XXXXXXXI ----9_-=-.ll1l..~XllXXXll.L_ 10 -.315 +XXXXXXXXI II -.268 XXXXXXXI -.l.2---.J AA __,__.~~.,--.XJl.x..x.x..1..-...__~~__•_._ 13 -.113 XXXI + 14 -.059 .)(1" IS .0'A .--*-------.L ..f:._...._......._ 16 .079 1 xx 17 .0qO IXX '6 .049 ....l...X-,~._.. 19 .004 I 20 -.034 +Xl ._21 ",-..0.7...3--..__..._~.._...__..)uu_. 22 -.112 XXXI 23 -.ISI XXXXI_2.."'.._.c."•.LZ-L _+...XXXX!. 25 -.172 •XX~~t XXXXI .)OIXXlt! XXXXI x~'tl ..Xl IX IXXX.- PLOT OF PARTIAL AUTOCQRRELATlONS -1.0 -.8 -.6 -._-.2:.0 .2 .4 ..6 .8 1 ..0 LAG COR~..----.~---+----.----.----+-++----+.~--+ • e + ----ll[...------..- +'lltXXX+X~XXXXX'()O(XXXXXXX XXKXXX+XXXX I _____..•IllX)C)CX XXX.xXXXI + +XXXI -......-----1---_1 __.. •XI e Xl .....~..._.-'t__.._L-.._...____._.~..._~._..._~__., e XX I 1><><XXe __.J_.----X...1-....,,: +J(Jl:l +IX -....__¥_.__l.XX--_.•- •KX I +Xl .•.xX.xx.l.e.._ +XX e XII:( .•._-,-~X--..-.._+. XI )0<\ .1_ I .~XXXY,1 I.X I ~ 1 lXX.x ... 1 + XI ".1-.86<;1 2 .-.452 3.192 29 -e017 30-...1.24- 31 •017 32 -.024 _-.301 5 -.137 __...b-..--..-lll5 -._ .7 -.O~Jl 8 -.029 .__.!l.~O'LL_.. 10 -.072 11 .159 ---l2-..-_-,,__0.4.5__ 13'-.06. I_.037 -_1S---..ll.7.J3.- 16 -.097 17 -.0.9 IS ._._-·.1".1._ 19 .o_a 20 -.066 ._-2-1..~.o.46-_.. 22 -.02''11 23 -.JoSi··~ 2".012 2'5 .OOH .26--..t~J 2.7._.0,2.& 28 .0:<4 - Figure 7.Plots of autocorrelations and partial autocorrelations for 1983 log-transformed discharge time series. DRAFT/PAGE 19,5/1/85 4/19/85, 4/25/85,4/29/85 NUM4/Hale Neither the mean nor any of the autocorrelations or partial auto- correlations of the residuals was significant;therefore,the model is adequate. The plots of both the ACF and PACF on the res i dua 1s from th is model showed a slightly significant spike at a lag of 15 or 16 days.This could indicate that the discharge time series shows a periodicity of ""'" about 15 days,or slightly more than two weeks.This possibility was further examined by spectral analysis.The spectrum of discharge (Fig. 8)does in fact indicate a peak at a frequency of .065 (a period of 15 days).It is not known at this time if this periodicity is "rea l".It may be related to weather patterns in the basin which control solar - insolation (cloud cover)and rainfall.A much longer time series of discharge would have to be examined to answer this question.A periodic term could be added to the ARMA(l,I)model (Box and Jenkins 1976)but, given the low significance level of the periodicity,it does not seem appropriate at this stage of model development. Carrying the idea of parsimony a step further,it can be seen that an ARMA(I,O)model using the log-transformed data is adequate and has the lowest number of parameters.The parameter estimates for this model are: A ~\=.994 with std.error of <.00005 giving --~7 '1 l ~~)1 1 J 1 .+••••+••••+••••+••••+••••+•••••••••+•••a+••••+••••+••••+••••+••••+00 ••+••••+••••+••••+••••+••••+••••+. Z.It +••• • M'J»"..., ••• .....•••• •-.~~.•••••• ..• ..•••• _.,-~----_..._--_..---.--~--_._.--.._-_....-~--. --..-60---..----------------..--~--....-.------.~..---.---- ...-...~.~~-•..!·~._--.~----_.-------~~-~----...-:-~~-_._~:---~~......,..--.-~-_..• ·-h---------:---.---G 1.2 +·S •p._..-...•. E • C .60 + T •• 1-o_~~-I__.-------:--~--=:v~---:--._.._--.-,--_-_-,-••,.~-'__.'_.~'__:~_~:~:-:_ IU .0 I •-.·+···:62;··+···:~j;··+···:i2;······:i75···+···:22;······:2j5······:;2;······:;7;······::2;·····~::1;····· 0.00 .050 .100 .150 .200 .250 .300 .350 .400 .450 .500___.."__".•~..'.___0_'•••••_....••_•.,._.•.___~J _••••.'.L.,.__....._~__•.•__••_...__,_.__.'._• FlEa Figure B.Spectrum of 1983 discharge time series. DRAFT/PAGE 20,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Hale A ~ The parameter as.is very close to unity.If </J,were equal to 1.000,the model would be reduced to a random walk model (Chatfield 1984).That is, the log of the discharge for today is the same as the log of the dis- A charge for yesterday plus a random error term.When¢,approaches 1.000 in a model with only one AR term,the series could be non-stationary (Hoff 1983).To test this,the series was differenced.The residuals from an ARIMA{I,I,O)model showed significant spikes,so the differenc- ing did not help;the ARIMA(l,O,O)model is better. The AC's on the residuals of the ARMA(l,O)model were a little better r than those of the ARMA(l,1)on the non-transformed data.However,the mean of the residuals was slightly significant,so the ARMA{l,l)model on the raw data is probably superior to this one. The 1984 discharge time series is shown in Fig.4 and Fig.6.The ACF and PACF plots (Fig.9)are similar to those of 1983.An ARMA(I,l)model on the 1984 raw data was adequate,as it was in 1983.Parameter esti- _~A mates are:y =23.2;(/;,=.808 (std.error =.0638);and ~=-.692 (std.error =.0750).An AR{l)model on the log-transformed data was also adequate but,again,had a slightly significant mean residual.The ACF and PACF plots,using log-transformed data (Fig.10),are similar to those of 1983,but perhaps show less indication of a moving average process.The estimate for ¢l is .994 (exactly the same as the 1983 data),with a standard error of 0.0001 and for y,is 10.0. 30 _f· ri J- CnAFT +. + ._-._-----_._---~"---"_..-~._---- + + • + + -1.0 -.8 -.6 -.4 -.2 .0 .2 .4 .6 .8 1.0 "~~~.__~Q~R~_.----+----+-----+----+----+----+----+----+----.----+--'-'''-----_.',--.----_.----------..--"--'-'.-----i---".----.~-,...-.-.-_..._..---.----...~\--. I .885 +I KXXX.XXXXXXXXXXXXXXXXX 2 .700 +r XXXlOIXX+XXXXXXXXXX 3 .556 +IXXXJ;XXXX+XJ;XXX 4 .430 +IXXXXXXXXX+. __S ~_3~£...__J~!P~~••XX _~._ 6 .268 +IXXXXXXX + 7'.214 +IXXXXX + __8 .15L +{XXXX 9 •.097 +IXX + 10 .062 •IXX + __~J__._.__.933 ...,__~~_x_..,_~._ 12 -.015 •1 + 13 -.073 +XXI + 14 -.137 •XXXI +--xs--:-_20..------•XXXXXI •'~--'-"------ 16 -.273 XXXXXXXI 1·7 -.333 •XXXXXXXXI-1e-=-:363----------.--X"xxxxxxix( 19 -.366 +KXXXXXXXXI ~_~E>~.•XXXXXXX~JU +_ 21 -.343 +XXXXXXXXXI 22 -.297 •XXXXXXXI _:l_~_._=-~:l_4.,_+icx~lllllC_L 2.-.199 IlIlXXXl 25 -.165 XXXX I 26 - •I 3.II XX I-2'7----;:'.09-7 -..XX I 26 -.059 +XI 29 -.025 +Xl 30 -.001 +1 31 -.013 I .?.2 __..:::•020_+X 1. ~1.0 -.8 -.6 -.4 -.2 .•0 .2 .4 .6 .8 1.0 _LA~Q.~~.~-'!.~~~":"-.-:~!..::~-=::.!.~..:-'~-~~~.:_.:=:~~~~_ 1 1 .885 +IXXXX+XXXXXXXXXXXXXXXXX 2 -.379 XXXX+XXXXI • 3 .195 ..IXXXXX. 4 -.163 .XXXXI • 5 .115 +IXXX +----6'---~;o-li'--+··-·-i---'+----··------·---··-·-~--- 7 -.033 •XI • 8 -.061 +XXI • + ••• 9 -.002 +I • 10 _052 +IX • -_...~~---=.!~!~.,.~_....!-_._.~JICJ._.~_._ 12 -.103 •XXXI + 13 -.048 •XI + 14 -.105 +XXXI IS,-;049 •XI 16 -.121 +XXXI 17 -.063 ...!_.__~X_I --I8""-~041 •I X 19 -.034 +KI 20 -.043 +XI •--za-----:-oiii-----·----------------.---l"i("-~------..-------.-.------.-- 22 .0.2 IX 23 .032 •IX •--24--:'-;029---'----.-----------.-"--..·xl--..---.-·---..·------------- 25 -.008 I 26 .015 +I -27 .osi'1-0..• 28 -.OI.1 29 -_015 1 30 -.077 xxi 31 -.065 XXI 32 .OOI I-- .-. - Figure 9.Plots of autocorrelations and partial autocorrelations for 1984 discharge time series. 31._--------------.......;;........--- --Pt.O.:f--Of'---JUU:OC'ORRELAT.I'O'...S---_.._ CrtAFT 1.0 + .6 * + • + +-.. .-----~ -_.---._!,-_.- • + + ~------_._-_. • .2 .4 +==+- .. ....~---~---------------- .0 * -.....-.2 +=----+ -1.0 -.8 -.6CORR.......-+__.LAG 1 I .906 +IXXXX+XXXXXXXXXXXXXXXXXX __2....-=a.42.O"'1!Jl.~I + 3 .14()*IXXX + ...-.071 •XXI-• 5 .016 1'__.J._j,_.__ 6 -.008 +I + 7 -.Q99 +XXI • __~_t..O18 +.-J 9 .060 •IX + 10 -.033 ..XI + __1_1__~~~~___.•__~~__!:._ 12 -.077 +XXI • 13 -.073 ..XXI • ~_-=_lll.L.._____________"_X2l.t 15 -.155 1'XXX~1 16 -.069 •XXI + __.U :-_~~.23__.+_JH-.__.._~__ 113 .090 +I XX 19 -.062 •XXI 2.0 .~~3:~_!'_.~~~__._.. 21 .039 +IX 22 .090 +IXX + 23 __:-_,_O~\)__X.I 24 .002 • I 25 -.019 1 26 -.023 •~I 27 .014 Y:XX 2B .Old I 29 -.058 ..XI_ JO -.OJO +Xl 31 -.023 XI __3_~__~O_~J'_..._!.l'!J_ _~J_':;.--o _-"';';";';-~~:.6 .-.4--.2 .0'."'."2 .4 .6 .8 1.0 ~.GGi--'-,jC~OP;RUD....._'l*"'_,,·=".".,...."""'-"-"'-'"'"'••=.-=.."'-,,-,....,1'=-=-"'-"'-"'.~==.,"',;.,,.==t+=,,-'"'-=..++=-=.,,~o=++=',=,"-,-",-='++=-='~'=+~__ I I .~06 +IXXXX.XXXXXXXXXXXXXXXXXX ~~.zA.J'.____•---Ul.XXlUUC.lt..,XJlXXXXXXXXX _ 3 .601+lXXXXXXXX.XXXXXX 4 .478 IXXXXXXXXX+XX ---5..··375 ~------.-J.X..XXXll.....f:__...... 6 .292 ';-+IXXXXXXX + 7 .ZII •IXXXXX ----8__---LJJ:__.•I lOUL....L._ 9 .OB8 +IXX + 10 .057 +IX • ___ll-__.J~__________----XlL....______1'_ 12 -.019 +I + 13 -.073 +~XI • __1_~=-._1.3_5____ .:t..10l;JU._....!:_..._ 15 -.209 •XXXXXI * 16 -.Z86 +XXXXXXXI _L7_.,-.349..•__.ll:ll.!tXXXXXX_I_ 18 -.376 *XXXXXXXXXI 19 -.378 XXXXXXXXXI _ZO..~:.u_IL ..!:..._JIJ!ll:l!_XXX.XX_L * 21 -.350 •_XXXXXXXXXl .. 22 -.306 +XXlClCXXXX I 23 ~.257 ..XXXXXXI 24 -.211 XXXXXI 25 -.171 +XXXXI 26 -.14-6 XXl\XI 27 -.123 XXxI 28 -.OdS •XXI 29 .,..1155_._.,1.1 30 -.036 XI 31 -.026 •XI __3.2 -.0.18 ....._~1__._._ --I!Ll1t_e»=--PARllAL_AlJIOCORRELAI LONS..._,.__ Figure 10+Plots of autocorrelations and partial autocorrelations for 1984 log-transformed discharge time series. .- .- DRAFT/PAGE 21,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Hale 3.2.Univariate Model for Turbidity The time series for turbidity in 1983 (Fig.11)was more complex than that of discharge.The ACF and PACF plots (Fig.12)indicate a strong AR(1)component.However,AR(l),AR(2),and ARMA(l,l)models were not adequate to explain the series. The series appears to border on being non-stationary because it in- creases in the spring as glacier melt increases and then declines in the fall.(This series would certainly be non-stationary over a longer time frame because the turbidity level is very low in the winter).The slow decay of the autocorrelations in the ACF (Fig.12)also indicates non-stationarity. Further investigation using the raw data showed that the series has a significant second orderMA term,while the first order MA term ;s not significant.Both first and second order AR terms are significant.This gives the model: "1t ~11f,.l +.t:t't ('i't-I +.2?>11'-\..t-~ with std.errors:on 1.=.0122 J\ on ¢~=.0234 1\ on e =.0988 ~ 33 11,./)+.0"(?,t.-~-n-,.,) t-a....t Susitna River Turbidity.1 983 400 ~300 '-' ~ .'2 of 200~ 100 ., 0 ~Jun 1 Jun lS Jul 1 Jul 15 Aug 1 Aug 15 1983 ~ Susitna River Turbidity.1984 ~, soo ~ 400 -,..., ~300 '-' ~ .'2 ..Q '-200.= ~100 ~ 0 Jun 1 Jun 15 Jul 1 Jul 15 Aug 1 Aug 15 1984 Figure 11.Turbidity time series at Talkeetna Station,1983 and 1984. 3'1 caAFT ,-'PLOT OF AUTQCO~R~L4fIO~5 ._-_..-_._---_..•---_.._--------- LAG -1.0 -.~'-.6 -.4 -.2 .0 .2 .4 .6 .8 1.0 CORR.+----+----.----+----+----+----+----+----+----+--~-+ I 1 .~24 IXXXX-XXXXXXXXXXXXXXXXXX 2 .850 IXXXXXXX+xxxxxxxxxxxxx ;].8l4 -IXXXXXXXXX_XXXXXXXXXX 4 .781 lXXXXXXXXXX-XXXXXXXXX 5 .731 -IXXXXXXXXXXX+XXXXXX 6 .666 rxxxxxxXXXxxx_XXXX 7 .588 lXXXXXXXXXXXXX+X 8 .524 lx.xxxxxXX.xxxxx . ".470 IXXXXXXXXXXXX 10 .417 IXXXXXXXXXX 11 .353 txxxxxxxxx 12 .302 IXXXXXXXX 13 .232 lXXXXXX 14 .164 -[XXXX 15 .138 ----------1'1<.XX 16 .133 IXXX 17 .106 I.xXX I~.06'1'--'--'---·'1 xx·----- 19 .O~'r ~ 20 .041 IX .21~-~-(f5-6-·_··_. Ix .... ..-22 .053 IX 23 .036 IX '24'---'~ffJ I """ 25 -.()Ol 2~.,..OO? .l'"7 -..,)U 1 .2.,3 -.Ul I ""-'On'.;:;I 3"~.-.6 fi:~'·xx I 31 -.Od2 ;0:'( 32 -.0<17 X.q - PLOT OF :>ARTIAL l\ulOCCRRELATfC'l5 -1.0 -.6 -.6 -.4 -.2 .0 .2 .q .~.a 1.0 CORR ••----+----+----+----.----+----.----+----+----.----+II 1 ."24 IXXXX_XXXXXXXXXXXXXXXXXX 2 -.025 XI LAG --- ~, - ::3 .2190 IXXXX-)( 4 .007 I 5 -.010 +XXI 6 -.135 -------------::.-oXoX::=:X::"":'I----------------- 7 -.17,;).XXX)t I 8 -.017 I g -.."11 1 10 .014 I II -.018 I 12 .005 1----:-- 1.3 -Q 136 ...xxx,,"( 14 -.075 XXI 15 .204 _rxxxxx 16 0155 rxxxx+ 17 -.013 I--1-0--.o:n .---~~---------'-1--------..----.--------~- 19 -.·117 J 20 .012 I 21 .07'J------·~---------..---·Jx·x--·-·---.-------~---- 22 -.Q65 Xl<I 23 -.Jl~f ~--·-.=:_l~--~----~-------...--x~Df-"·----;--~~--.-------.----.---" 2'5 -...}()!;:.X)I".I 26 .at..1 I"x - 2"7"-.:"~-6f ",I "tJ ..:)d f •I 2.CJ -.t·:;,.."»+.~_'(o{;o:I 30 .J c...I . 31 .04:'·I x 32 -.I I I -X X'I and partial 1983 turbidity Plots of autocorrelations autocorrelations for time series. Figure 12. 35 DRAFT/PAGE 22,5/1/85 4/19/85, 4/25/85,4/29/85 NUM4/Hale While this ARMA model is adequate for the time frame examined,in general,an integrated model (i.e.,one with a differencing operation) is probably more appropriate.The differenced series (Fig.13)is clearly stationary with a mean close to zero.The ACF and PACF plots for the differenced series (Fig.14)verify that the differencing was successful.The differenced series could be adequately modeled with just the second order MA term;the first order autoregression term was not significant in the differenced series.The equation is: =.0972 and the mean of the residuals insignifi-"with std.error on 6l:a, cant. +2 t-,t-.23 a t-~ W"u··(.: Z t :: The second order moving average term is likely related to the random "s hock"caused by a rising discharge (which is in turn caused by rain- fall)which resuspends sediment.It takes a few days after the rainfall is over for this perturbation in turbidity level to drop to the pre- rainfall level. 3.3.Univariate Model for Age 0+Chinook Salmon Outmigration The time frame chosen for Age 0+chinook salmon was the same as that of discharge (Fig.15).The plots of the ACF and the PACF for 1983 (Fig. 16)show a strong first order autoregresssive component.In fact,an ARMA(1 ,0)model ,abbreviated as AR(1),adequately represents the data. - 3~ 1 )),1 -1 )I 1 )J 1 )1 ))iii j 1 I:..• •• • (0 r I gi na I 0 alt a I •, I • .c• •.. • ,. • 0'1 ;I\.f oil • I•II• I••~I I..•• • I- II .:l 'I ~L. S I I···1 I • 1 eccce o(c I •----.---1-·-1---1+--,--+1_--I I ..1 CI "0II- -.. >-I--C-EOa: ::,) t- -:::) t- Z- t'J :2J ~ ~ c. ! •I I ••I ,, CC~CCIlll .r c r'i ~ I Ie • I I • I., i i • ;~i II, :-t--·_:r-:~--~-~--g----~ 1 I I ! - - CJ::»c ..J::»., II . (Firat Dlffereoc~sl I j I.f II.'I'.1 I .iICC',c...• , .1 reli ;c c c:<c'"rei I c I,.;,r ,••I ., :·f i:'•I,•'"i :i,.,!I ·C..eel;!'I .•'I iI!•I I 'i:I I I .I .I '---;[--1;-_0,_--;'--f i ----;---;I--i;-I--F--~-;IO- DAYS 1 I .~.. • I I I.-10-- =II !I I I • z::»., '"I c r- -;f-- )0 <:=: )0 Z ..J CJ<::»::»::» 2".,< ·I, !011 /\1~~l} a 11o{,'"ICI I •i .~.•. I 'C C ca•• 'I 1 ' :I I I 'Ill .':I .!'I'.,."".1•••1.._11I.__ )..I .~ I ! >t--C-EOa: :::) t- -:::) t- Z- W 'fJ Figure 13.Turbidity time series,1983.(Day l''"~~a'y 18,Day 105 =Auqust 30). CriAFT ?LOr OF AUTOCONRELATIO~S IX}l.X + lX)(X + IlO< xx ~~~--=-~~__~~_ ..XXXI I -~~_!_-------~----------- l(I IX)(X + IX e -.110 9 -.010 10 .053 11 -.029 12 .121 13 .059 -1.0 -.5 -.6 -.4 -.2 .~.2 .4 .6 .8 ~•• LAG caRR.+----+----.----+----+----+----+----+----+----.----+ I I XX"XXXXI I l -.015 2 -.271 3 -.005_ ___~~_._IU 5 .101 6 .083 7 -.076 -------------~---- 14 -.297 15 -.162 16 .167 ~-1-7----.143 18 -.06>1 19 -.117 O!O -0126 21 -.009 22 0153 :2.3 0045 24 -.00] 2S -..l~"t 3D .3\ --~i e f}4 7 ...167 -~')1'': -...l.J4 XX+XXXX I +XX-XXI txxxx + ._---~-----"-----_.._._.._,---~-------- IXXXX + XX I XX'{1 ------~-xx-xi-----~-----~------~-- I +IXXXX + --_._~._-_._-------_._----_...-.~-,---'"--..- [< 1 +XX;.()I't 1'<. 1'( J x.:<}()(-'Co 1 .1(:(X J IX PLOT OF PARTIAL AUT.:lCOQREL.TIO>OS -1 ~o -.8 -.6 -."-.2 .0 .-2 ~4·-.6 .8 1.0 LAG CORR.+----.----+----.----+----.----+----+----+----+--__+ IX :x XXI +XXXI I I r ~.1(x '.It. ...'J(1(~.I L''C('j ~ ..)C A,I(( xxt [X;(X,l(~ ;~4 ·l~)1 ,,-OJ -.I U "'t ·I ,JC ~r -..J 1 1 .!~·1'..1 I 29 -.I I 7 30 -.!JoO .31 ·146 I I -.0 15 :>~~~I_ 2 -.Z71 Xx"xxx)tJ' 3 -.016 1 4 .o~o lX' 5 .107 IXxX + 5 .140 IXXX + 7 -.015 1B -.071 -~--'---:X:::X:::I:-·--~~---~---~--- 9 -.06'1 Xx \ __lg~~._O_3_1_______________.x_~.__ J.l -.06~)'l(~I 12 .163 [~XXX+ __-l_3__.-!.1 04 ..,__._~__~[x xx • 14 -.22~X.xxxxl 15 -.\8R X XXX,..I _____.j._~~2.?~___+.__._._~J_ 17 .06~lXX 16 ..1)25 19 .027 20 -.018 21 -.104 22 -.'109 ?3 ..'1~~) Figure 14.Plots of autocorrelations and partial autocorrelations for differenced 1983 turbidity time series. 38 Age 0+Chinook Salmon,1983 18 17 16 1S 14 13 12 F"L. ::l 110 J:10 U 9CL .t::8u '0 7u 6 5-4 .3 2 1 0 ..Iun 1 ..Iun 1S ..lui 1 ..lui 15 Aug 1 Aug 15 1983.- Age 0+Chinook Salmon,1984 18 17 16 15 14 13-12 L. :J 110 J:10U9CL-.t::8u...70 U 6 5 4 .3 2 0 Jun 1 Jun 15 ..luI 1 ..luI 1S Aug 1 Aug 15 1984- ..- Figure 15.Age 0+chinook salmon outmigration rate time series,1983 and 1984. 3q CriMT PLOT OF AUTOCOR,~R",E"L,-,A~T~IO=N':;!S,--.._ -1.0 -.8 -.6 -.4 -.2 .0 .2 .4 .6 .8 1.0 LAG CORR:....----+--+--+----+---+----+---+---+---+__+ I I .657 +tXXXX+XXXXXXXXXXX 2 .398 +IXXXXXK+XXX 3 .284 + I XXXXXXX 4 .182 +IXXXXX + 5 .090 +IXX + ~~~29 •IX + 7 .005 I + 8 -.001 +I + ~ I ! 9 -.041 +XI + 10 -.047 +XI ~ II -.075 •XXI • 12 -.087 +XXI • 13 -.103 •XKXI + 14 -.119 XXXI 15 -.075 -+~--:~X~Il:';I~--~-=+------------- 16 ..-.052 XI. 17 -.099 +XXI +--.8----.160 ------------.--xJCxx.-----.-----.--------. 19 -.179 +XXXXI + 20 -.156 +XXXXI •-'21.----.145 •-XXX:J(I---~~'-~-----"------,-.--". 22 -.141 +XXXXI • 23 -.127 •XXXI +----i4-----::-;t2-3----·-------------.--·xx·xi---------".--".-.--.----.-- 2S -.124 •XXXI • 26 - • I 22 +XXX I + 27 ..:..;I i'i •XXX I 28 -.117 XXXI 29 -.124 •XXXI 30 -.137 xxi<I 31 -.124 +XXXI 32 -.101 XXXI - PLOT pF PARTIAL AUTOCORRELATIONS LAG -1.0 -.8 -.6 -.4 -.2 .0 .Z .4 CoAA.+---+--+---+---...----+--+--+ .6 • I 1 .657 +lXXXX+XXXXXXXXXXX 2 -.060 +XI + + 3 .083 •IXX + 4 -.044 +Xl 5 -.034 +XI + + + •• + + • + - Plots of autocorrelations and partial autocorrelations for 1983 chinook salmon outmfgration time series. 6 -.025 +XI 7 .009 •I 8 .007 +I 9 -.062 •xXI 10 .016 +t -II -.071 •XXI 12 -.001 •I 13 -.049 +Xl __14 -.031 +Xl + 15 .059 •--IX'+ 16 -.020 •I • 17 -.099 +XXI + -t8-----~IOO---------------..--.-XiX~---- 19 -.036 •Xl • 20 .006 • ( • --ii--~";'-:-0-22 --.-------xr---+---------- 22 -.031 •XI + 23 -.034 XI +-;2-..--.04"6"'----•i--'--. 2S -.042 •Xl • 26 -.O~I •XI • -27 --:..-~034------------.----XI 28 -.045 XI 29 -.050 Xl30--.oil -------~-------+XlCl-----.- 31 -.034 Xl 32 -.02Q •XI Figure 16. ,- DRAFT/PAGE 23,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Ha1e Although the plot of the range of sub-groups against the mean ;ndic~ted the need for a logarithmic transformation,the residual ACts of an AR(l) model on the 10g-transformed data (Fig.17)were slightly larger (but still insignificant)than those of the AR(l)model on the raw data.The A standard error on ~,however,was lower with the log-transformed data. ACF and PACF plots for the log-transformed data are shown in Fig.18. The AR(l)model for the raw data was: 1\ with standard error on <P.=.0743. The AR(l)model for the log-transformed data was: '"with standard error on 90.=.0363. The mean of the residuals was not significant. The time series plot for age 0+chinook salmon outmigration in 1984 (Fig.15)shows a different pattern from that of 1983.The fry did not begin to migrate in 1984 until about June 12.The low level of out- migration early in the season causes a time series which ;s non- stationary.To avoid this problem,the dates selected to be included in 4-1 Log-transformed chinook,1983 4 .3.5 3 .-'2.5 ""': + :I:2~ (J 3 1.5 0.5 - Jun 1 Jun 15 .luI 1 .lui 15 1983 Aug 1 Aug 15 LOG-TRAt'--ISFORMED CHINOOK,1984 4.-----------, - Aug 1 Aug 15Jul1.luI 15 1984- 1 °-tm"",mrnmrrrnrrrrfnrrmrrmTTTTT1TlT1nTmrmnmrrmmmrrrnrrl1TFTrrmnmmnlTTTlm Jun 1 Jun 15 .3.5 0.5 3 "2.5.- + :I:2~ (J Z ...J 1.5 Figure 17.Log-transformed age 0+chinook salmon outmigration rate,1983 and 1984. [;iiAFT ....PLOT OF AUTOCORR€LATION5 r I~ ---------~l-"O--.....5-.--.6--~.-4-·'-_2 ..-.-0--.__-2----4--_I!>---.a---1-.O· LAG CORR.+----+----+----+----+----+----+----+----+----+----+ I 1 .84L-_________-.-·_"----IJl.XJtJt..xJtlUtxJtJtXXXJl.Jl.JUtXX.."__. 2'.695 +I XXXXXX+XXXXXXXXXX 3 .559 +IXXXXXXXX"XXXXX .4 ....5 ..._,,--_...-UtXXXXXXXX ....._ 5 .343 +IXXXXXXXXX+ 6 .238 +IXXXXXX 7 165 i .__.._...__-4XX>'xxX -"'*c...._ 8 .115 +IXXX .. 9 .028 +IX + 10 -.OQ~_.__.._.,---------*._l.__.~_.~. 11 -.053 +X I .. 12 -.084 +XXI 13 -.J 1~._.__.__~~___Xx..x L ..._."'----__.."__ I 4 .- • [48 +XX XX 1 + 15 -.156 +XXXXI . 16 -.146-.~.__-:...L ____X:x.XX.l_--..:..__~__.:..:~.:..:=___~::::._:.::.::..·---~ 17 -.196 +XXXXXI + 18 -.249 +XXXXXX[+ __-L9 ~28lL_.__.__L_X.lOOUOUt.1-__._.__.__.,__..__~..__ 20 -.298 +XXxxxxxI + 21 -.292 XXXXXXXI.· ~_-.....2aa-_______....XXXXXXXI._.+_.__.. <!3'-.26<!XXXXXXXI 24 -.<!51 +XXXXXXI + .._25--__=.24>l.-___.xXXXXXI_to . 26 -.<!,"S XXXXXX I + ;£7 -.241 +XXXXXX[+ .2.8 -.24.5 XUX.XX 1 29 -.2bO X)(J(KX;tt I + 30 -.273 XlCXlCXXXI ~!-.2lo"XXXXlCX I 32 -.205 lClClClOl I ---------------------_._-_._---- PLOT OF PARTIAL AUTOCORRELATIONS 1.9 .8 .6 .4 .2 .9 .~.,.6 .~ LAG CORR..----+----.-~--.~--+__--+----++.~~_+----. I I 847 •IXXx.x.->xxxxxxxxx~~ 2 -.076 ..XXI + 3 -.035 +XI 4 010 ..1 .. + + + + + + _t 5 -.033 +XI +, 6 -.086 +XXI + 7 037 ~--'lI~X'--"._ 8 .016 ...J + 9 -'.I 83 XXXXX I .. 10 .122--_...._---I-XJt.J;_"-•._ 11 - •093 +XX I .. 12 -.010 +I 13 -.o~_..XI _~__~_ 14 -.022 +.lCl + 15 .003 ..I --U>------.D4-2---.--A I-lt.---"---- 17 -.2<!6 X+XXXlCI 18 -.099 +XXI ----1-.9-.-.024 .---"--.._'.-I-lt...__._.+_ 20 -.021 +XI 21 .015 +1 _-22 ....0~"'-.____.._.+_---Xl.-.-.__+.__. 23 .015 +I 24--.118 •XXXI _25......•01.6 +I 026 -.089 XlC( 27 -.0<!4 X I 26 -.094 lCXI 29 -.073 XXI 30 -.049 xl _J.I .~0.41 IX 32 .029 IX - """' -Figure 18.Plots of autocorrelations and partial autocorrelations for log-transformed 1983 chinook salmon outmigration time series. DRAFT/PAGE 24,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Hale 1984 ran from June 12 to August 31 (79 cases).Analysis of this shorter series is not as strong as that of the longer series in 1983 but the series is long enough from a statistical point of view;Hoff (l983) suggests that about 40 or 50 observations is the minimum necessary for attempting an ARIMA model.Although logarthmic transformation did not appear to be strictly necessary for the 1983 data,it was required (to produce an AR(l)model)with the 1984 data,perhaps because of the shorter time series in 1984. The ACF plot for 1984 on the log-transformed data (Fig.19)is similar to that of 1983,although it does decay a little more quickly.The 1984 PACF plot (Fig.19)is very similar to that of 1983 in indicating a strong AR(l)"component.The estimated value of~in 1984 is 0.973 (very 1\ close to that of 1983),with a standard error of 0.0265.The 1984 model is: The mean of the residuals is insignificant.This model does not differ from that of 1983. 3.4.Univariate Model for Age 0+Sockeye Salmon Outmigration Age 0+sockeye salmon outmigration was examined from May 23 through August 31,1984 (Fig.20).This time series shows a strong AR(l}compo- nent (Fig.21),similar to that of the chinook salmon time series. 4't- - -i - -1.0 -.8 -.6 -.4 -.2 .0 .2 .4 .6 .8 1.0 ~~~-*~-_~_~.=.~~~.J~~--+~~-=-""'=-~~----"----.~~i..- I .812 ->IXXXXX->XXXXXXXXXXXXXX 2 63'..+_------1.XJOl.XXXJt+..xx..xxx.x.xx.-..._ 3 .442 ->IXXXXXXXXX->X 4 .285 ->I XXXXXXX -> ___5~_--..L.Z9____. .->._.•J.XXXX..._.~_._ 6 .132 ->IXXX -> 7 .072 ..I XX -> __.B-.....-_aAZ J .._._.I.lL-~._ 9 .080 +(XX + 10 .14.->IXXXX-> ...-ll__._~20.0 .....J.I UXX1L....~._.__.. ..__.... 12 .226 ->IXXXXXX-> 13 .204 ->IXXXXX-> I 6:~.J51~n_':"':":..:....::...:.:~::.~=-:_:.:....-.:.:..~.:_:.:.._-_.__.:..::~~x.xx ...:.!'....:........:.._ 15 .110 ->IXXX -> 16 .096 I XX .17 .._.055 ......__1lC._->_ IB -.023 ->XI 19 -.026 XI .• .__2.0 -~0..87__.._.___XX I .+. 21 -.147 ->XlC.XlC.1 22 -.139 XXXI .23 .-.•.125._........->.lC.XXI -> 24 -.112 XXXI-> 25 -.0~7 XXI 26 -.067 XXI-> 27 -.007 XXI 2B -.07\XXI 29 -.034 X'(I 30 -.107 XXX! 31 -.16'1 XXXXI _.32._.._~•.20~_'.XXXXX.I __•. _.PLOI ~2ARI~ALAutOCDRRELAT~ON5 -1.0 -.6 -.6 -.4 ~"t..~=_-~~,_~~~.:__+- -.2-+--.0 .2 +----+ .6 • .8..... -> .~. 1 I .B12 ->t XXXXX->XXXXXXXXXXXXXX __._..2-_-=..o..OIU__._.0 .J_.)l.X 1 !.-._.~._. 3 -.13.XXXI .. "-.034 ..XI .. __......5_._---"-0_31__._..__"t..__._._U.~_·....... _ 6 .079 ..lXX. 7 -.104 ..XXXI -> __.!l __~_Q.3~._.. .!..I_l<•...._ 9..1"7 •IXXXX .. 10 .116 •IXXX .. ---.l.1 •.9_U_..__._.....!_..L __~..!....__. .._ 12 -.045 ->XI 13 -.0.7·->XI ..•._...._.....~ --y4":,;...;",...:,;;,.~-:;;~·o·O(:f .__,_~__.__I~~~.2 ._ IS -.003 ..I .. 16 .075 •IXX ....1.1'_.~_.083 _.->.XXI 18 -.156 •XXXX I 19 .197 IXXXXX" .._,20 -:..•.19~__...XXXX~r • 21 -.156 •XXXXI .. 22 ~126 IXXX -> 23 .018 _._J • 24 -.001 I • 25 -.Ob6 XXI • 26 .010 I .- 27 .012 I 28 -.t}d6 XXI 29 -.033 XI 30 .00b I 31 -.153 •XXXKI _.,_~,__'"_....LO§_I....~_._._•_.__1_~__ Figure 19.Plots of autocorre1ations and partial autocorre1ations for log-transformed 1984 chinook salmon oUbnigration time ser'ies. Figure 20.Age 0+sockeye salmon outmigration rate time series,1983 and 1984. a - ..... CrtAFT Pl:;OT OF AUTOCORRELATIONS -~.o ,·-.8 -.6 -._-.2 .0 .2 .4 .6 .8 1.0 ~t.!::AG~~-~::c..,:,_C~lJRA!!!!!.~~'~=':·~''::::::::=.':::=::::-'':::=='':::=~'~=::''':::=::''':::=::'':::=~'• I I .799 I IXXXX'XXXXXXXXXXXXXXX 2 .543 •I XXXXXX.XXXIUlXX 3'.4_• .I XXXXXXXIXXX· •..15 •IXXXXXXXX.X 5 .361 •IXXXIUOCXXX 6 .275 ----.-------.-------iXxxxxxX .----------.- 7 .229 •IXXXXXX. 8 .190 •IXXXXX. 9'.126 •IXXX. 10,:'.093 •lXX',<. 11 .101 •IXXX • 12 .127---·--------.-IXXX •~------- 13 .136 •IXXX. 14 .126 •IXXX. IS...12.•.IXXlIt·.,.._c__·_ 16 .139 •IXXX • __~~__~~~_____;-}~:X-----;_---- 19 .021 •IX • 20 .003 I I •-----_._-------_._~--._-._------------ 21 .0.9 •IX • 22 .019 •I • 23 -.053 ._____...X_'_• -24C:--~-;;063 XX I • 25..-.052 •XI • 26'-.0.3 •XI • ---27---:;'~iJ59-'•XI • 28 -~06~•XXI • 29 -.107 •XXXI 3-0'"':'.166 •XXXX,• 31 -.180 •XXXXI • 32 -.177 •XXXXI • PLO'f OF PARTIAL AUTOCORRELIlT''''ON=S:::::.-_ .~&-' • .,-1:-.0 ,-.8··-.6;-"..~.2>:.O:'~--:.'::.2'..••.. L'AG,"CORR._~.'.1"• I t..799 •'XXXX.XXXXXXXXXXXXXXX 2 -.267 XX.XXXXI • 1.0 j-t' 3 .329 ..I.XXXX.XXX- •-._....X.·",,-.·::~"·- 5 .028 - •:Ix"•. ---6--~7'l'--.---------------.-XX-,--..------------·------- 7 .099 •IXX·. 8 -.116 •XXXI • 9 -.016 ...,-..•• Ut..059 ...IX •.- 1'1 .027.•I'X".'" -U---.067 •IXX" 13 .01S • 1 •1"-;'015 ..--~----·-·--·-·-.--·-·-·--I..·-·~--..·• IS .012 •I·• 16 .OS5 •I'll;• 17 .-.071 '.XXI •--'8--:;'-;'0'-2----------.--.---..---~·--.--Xl----.-.-.----""--'-.--------.- 19 -.13.•XXXI • 20 .129 •IXXX •~-I----;040---------------'--.--.---=I""'X=:..,--.---- 22 -,.1 a.lUtXXX I'•. 23 :.048 •IX • 24 -.04:3 •'1 X'-• 25 -.083 •XXI • 26 .082 •I XX • 27'-.064 •-iXl • 28 -.025 •Xl • 29 -.189 XXXXXI • '30'--';-047'•IX • 31 -.030 •XI • _32__:,~_~65 •XX_I ..!..._.__ Figure 21.Plots of autocorrelations and partial autocorrelations for 1984 sockeye salmon outmigration time series. '-11 DRAFT/PAGE 25,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Hale However,neither an AR(I)model on the raw data or on the log- transformed data was adequate.A MA(I)component was also significant in the raw data,leading to the model: A A The standard error on ~(.775)was .0681 and on ~(-.567)was .0883. Although the mean of the residuals was slightly significant,none of the autocorrelations or partial autocorrelations were,so the model is reasonable. Examination of the autocorrelation coefficients of the four time series - presented above at lag =1 day (adjacent values)gives an idea of the smoothness of the time series.Typically,the coefficient for physical/~. chemical variables is higher than that of the biological variables and the time series for discharge (Fig.4)and turbidity (Fig.11)are less jagged than those for chinook salmon outmigration rate (Fig.15)and sockeye salmon outmigration rate (Fig.20).Saila et al.(1972)reported similar results for the autocorrelations of alewife upstream migration activity in relation to incident solar radiation and water temperature. The square of the autocorrelation coefficient at lag =1 gives a measure of the percentage of the variance of the value for today which is expl ained by what was measured yesterday (Murray and Farber 1982).In 1983,{.86)2 =74%of the variability of discharge for one day was - 48 -,,;;,,'---------------------- - - DRAFT/PAGE 26,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Hale explained by the value for discharge on the previous day.The percentage for turbidity was (.92)2 =85%while,for chinook salmon outmigration rate,it was only (.66)2 =44%,and,for sockeye salmon,(.65)2 =42%. 3.5.Discharge-Turbidity Transfer Function Model The cross correlations for the residuals from the discharge series and the turbidity series,both filtered by the ARMA(l,l)model for discharge,had a significant spike at lag =1 day (Fig.22).This suggests a candidate model (Box and Jenkins 1976;McCleary and Hay 1980): where:Ytis the output series (turbidity) W o and ~are transfer function parameters xtis the input series (discharge) Neis the noise component,an ARIMA model - The assumption that the ARIMA component of the model was white noise leads to significant AC·s in the residuals series and must be rejected. The ACF and PACF plots on the residuals from this model suggest an AR(l) model for the N component,leading to the full model: - 1-cr,8 - l -¢.B Ltq ------------------- l;i1AFT - -1.0 -.~-.4 -.?e0 .2 ~~.6 .~1.0 LAG CORH ••----.----~----+----.----.----.----.----.----~----+ + • ... I -2U -.Q2~xl - I q -.tIS ..)l,~)(I -Id -.022 ~l -17 -.076 Xxi ·--l·6--·-.1-77 (){}(X;(+ -15 -.102 •XXKI -14 -.010 I ---1--~---...-l-64 _·....x-XX~I---.-----'-'.---------------"',- -12 .062 IXX -11 -.023 •xl --·_·-l.Q,---..~-·--_.-.-.-----_..----·-----+--X·x.-1-------....----'"0 •.•--_._••••-.-----.,.-,--- -9 -.0!>1 XXI • -8 -.009 I ~~~--~7~~--.-l24--~----~-_.---~~...-XXXI -6 -.009 I -5 -.050 +XI 4 •Q.2-l--.-----------.--~.-__~.-1~-··.~.-~------------.-----.-- -3 .037 IX -2 -.050 +XI 1 .06;!-·---..----------__~-XJH---~.....-.----------------- o .077 +lXX 1 .349 IXXXX+XXXX 2 .1-Q9----.--------K~X-l • 3 -.006 I ~ •.016 •I --'-----S-•138 ~KXX--------.--_. 6 -.132 •XXXI 7 -.002 •I • 8 -.02L .~_"~-------_....--'--__.._ 9 -.0!>9 •XXI + 10 -.016 I , I •005 •1 _--J _ 12 -.072 •XXI 13 -.029 XI ____..l.-4-."!!!..••026-..__+-.x-r 15 -_146 +XKXXI 16 -.052 +XI _...lZ.-..__1..1_4____~.• ..1 XXx.._+ 18 .019 •I • 19 -.235 X+XXXX)+ --ZQ ---t-ZS-.-__..._~.+._tJUUX•.._ Figure 22.Plot of cross correlations between the residuals of the ARMA (1,1)discharge model and the prewhitened turbidity time series,1983 data. - 50 DRAFT/PAGE 27,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Hale Parameter estimates were: 1\ W o =8.349 with std.error of 1.7044 1\ ~,=-0.559 with std.error of 0.1718 ¢.=0.993 with std.error of 0.0009 The t statistic for each of these estimates was significant,leading to the conclusion that discharge and turbidity are related by the equation: "j't - 8.3S 8 +-~5f.B ;- - The ACF and PACF plots on the residuals from this model showed no significant spikes;therefore,the model is adequate. This model does not necessarily imply that discharge level is a causal factor for turbidity.These two variables are correlated largely because when glacial melting is high,both discharge and turbidity are high. Thi s phenomenon provi des the seasonal trend of the two seri es;the discharge of clear water tributaries such as Portage Creek and Indian River (w~ich increase discharge but not turbidity)is a noise component. Discharge does have some causal effect of turbidity by resuspending sediments and other particles during a rapid rise in discharge level. 51 DRAFT/PAGE 28,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Hale The value of the model is that it allows levels of turbidity for a few days ahead to be predicted from past values of both turbidity and discharge. 3.6.Discharge-Chinook Transfer Function Model After both the input series (discharge)and the output series (chinook salmon outmigration rate)were filtered by the ARMA(l,l)model for the discharge series and the residuals from both series were cross correlat- ed,there was a significant correlation at lag =1 day (Fig.23).This suggests the transfer function model,as given by McCleary and Hay (1980): - t or,using the backward shift notation of Box and Jenkins (1976): This model implies that the current day's discharge rate has an effect on the next day's outmigration rate.The estimate ofwowas 0.02.The residual ACF using this model suggested that the assumption of white noise for the N component was not valid;it appeared that an ARMA(l,O) model would be appropriate.The full model is: 'rt -w.B ~t - -¢,8 - CriAFT PLOT OF C~055 CO~RELATIONS .0 .2 ..6 •'I I .? ".... l~;'\(;":(;,1.......----.----+----..----..----..----.----II-----+----.----+ -<'0 .(}!<'1 -19 -.34U -. -iJ}-.032 Xl -17 -.04~xl -16 .096 ~1:(:(~ ____-_I2.~~~_'_>lX_ -14 .-.0""'6 Xl -1.J .007 1 -12 -.022 __~__~.__;(__t_.__+__._.__,__~__~_'. -11 .025 IX -10 -.04+~I + -9 ~.077 ~_..l!X.L ___'"_ -8 .011 ..1 -7 -.020 I -6 -.003 _,..._.._"_+1._ -5 -.009 1 -4 .033 +I X -3 .085 .._....Jxx __+_ -2 -.062 XXI -I -.010 1 .. I>.1>17 + , +_ 1 .220 +IXXXX+X 2 -.032-XI + __3__--"007 -..L.__~_ 4 -.004 +I + 5 .078 +IXX + 6 -.087 +Xk I • 7 .016 +1 .. 8 -.0-47 +XI + 9 .062 +1 !QL_+-_ 10 -.063 +XXI + 11 • 1 34 +1 X XX + 12 -.107 +XX~J.+_ 13 .025 IX + 14 -.065 XXI 15 -.024 ..+__!1...~_ 16 -.045 Xl 17 .019 I + __l~_-..!.LtL..__.__...!.-1_~X~__~_ 19 -.166 +>lXXXI 20 .066 +IXX ---------- - Figure 23.Plot of cross correlations between the residuals of the ARMA (1,1)discharge model and the prewhitened chinook salmon outmigration time series,1983 data. 53 DRAFT/PAGE 29,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Hale The parameters for this model were estimated as: 1\ GUo =.025 with std.error of .0249 ~t =.667 with std.error of .0751 The t stati sti c on the estimate for W o was not si gnifi cant.However, because the practice of prewhitening the output series with the model.. for the input series tends to underestimate the significance of the results (Botsford et al.1982)and because there was a significant cross .correlation between discharge and outmigration rate at a lag of one day, it seemed best to leave this term in the model.This would have to be verified with more years of data.The model is: - +.. The ACF and PACF for the residuals from this model show no significant spikes so we may conclude that the model is adequate. This model does not imply that the discharge series is a strong predic- tor for the outmigration series.But adding discharge does result in an expression which has more predictive value than would be obtained by looking at the outmigration series by itself. 3.7.Discharge-Sockeye Transfer Function Model As with the discharge-chinook relationship,the cross-correlations of the discharge and sockeye series,filtered by an ARMA(l,l)model for :,! DRAFT/PAGE 30~5/1/85 4/19/85~4/25/85~4/29/85 NUM4/Hale ..... I - discharge~showed a significant spike when the sockeye series was lagged one day behind the discharge series (Fig.24).This spike was stronger for sockeye than it was for chinook.A candidate model (Box and Jenkins 1976;McCleary and Hay 1980)is: The ACF and PACF plots on the residuals from this model suggest an ARMA(1~1)model for the Nt component~leading to the full model: Parameter estimates were: - ,- (1-e,13) (,-<P,B) ,.... I\.Wo =.08 with std.error <.00005 J\ d.=-.73 with std.error .2205 (P,=.69 with std.error .0878e=-.57 with std.error .0957I ,.- -, The t statistic for each of these estimates was significant~giving: (I- ( I +--t-.51 B)--.,q 8) ~;1'I.t;,=~cA~X 10-3 .08 8 +-.13 B ,..." I:~AFT PLOT OF CROSS CORRELATIONS -1.0 -.8 -.6 -.4 -.2 .0 .2 .4 .6 .8 1.0 LAG CORA..----+----+----.----+----+----+----+----.----.----+ I -20 -.027 +XI + -19 -.099 +XXI + -18 .059 +IX + -17 -.007 •I • -16 -.059 +XI • -I~.021 •J_It__--:_~_____._ ~.-~~047------·---------.---XI -13 .020 •IX • -12 -.066 ._XXL__+_ -11 .040 •IX + -10 -.101 •IXXX • -9 -.003 • I •.--8 -.102------------.XlCXl---.------------- -7 .034 •--IX • -6 .020 • I -5 .012 •I + -4 .009 +I • ":3 .035 •IX-. ~-~i06------------·-----.----ixXX ;.------------.-----.- -1 -.279 XX.xilXXI • o .096 +IXX + I .467 + -IXXXX+XXXXXXX 2 -.002-• I + 3 -.007 . .__._•.L ~_ --~4--;'O~-• 1 • 5 .126 •IXXX + I>-.168 .XXXXI • 7 .063 •IXX.- 8 .039 •IX • 9 .054 •IX •-10 ---"~~'052·-.-------•"ii -.~----. 11 .029 •IX. 12 ":.092 +XXI •--i3------;i36---'·----·-------------.----Ixxi-.- 14 -.000 • I • 15 -.041 •XI • -iiS----·'.-05Cf ..'--.-..-'-+Ix ---'. 17 .025 •IX • 18 -.IIT +XXXI •-i-9---.-Z36--------------.---lixxx+i ------.----------.--_.--- 20 -.248 X+XXXXI +- Figure 24.Plot of cross correlations between the residuals of the ARMA (1,1)discharge model and the prewhitened sockeye salmon Dutmigration time series,1984 data. - - - - .- ,~ - DRAFT/PAGE 31,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Hale The ACF and PACF plots on the residual series from this model showed no significant spikes and the mean of the residuals was barely significant; therefore,the model is adequate. DRAFT/PAGE 32,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Hale 4.0 DISCUSSION Time series analysis is a useful method for dealing with time ordered data sets,including ones that do not appear to make much sense at first glance because they are too noisy or because they drift as a result of random events.The modeling effort helps us to understand why the plots look as they do and what factors shape them.It also is useful in trying to understand what effect a change in the contro~1 i ng factors mi ght produce. The segments of the time series examined (discharge,turbidity,chinook and sockeye salmon outmigration)were described by relatively simple Box-Jenkins model s,using mostly fi rst-order terms.None of the series appeared to require d~enCing (although turbidity was on the border- ~.--// line)to achieve stationarity nor did they appear to have a periodic component (discharge being a possible exception)which would require seasona 1 differencing.All of the series showed a very strong fi rst order autoregressive term,indicating that the value for anyone day is strongly influenced by the value for the previous day.Similar results for the flow regimes of several streams in Austral ia was reported by Sri kanthan et a1.(1983),who found that most of those streams with a non-random process had a first order autoregressive term. By building Box-Jenk"ins models for these time series,a better under- standing of the processes which control these variables was developed in .,/"\r' that the structure of the random processes which generate an observed -':'~''''''-~~-""-'-------'~--.·....-._"'~-'''--.."'....._..M_~~'.'._.·__~_~_·_''''"__,_·_''"~.~"'..~--..,~.,-~""",-..._.','-.'..'~,~-"'-,.~,..",-,.-;",,,,---"-0__• 58 - -1 DRAFT/PAGE 33,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Hale :.:_~.~:.~.,(t~:'."Xt£~/k'l911!L~It is import,~t/to explore the effect on salmon ./ outmigration of a construction'(~roject which will change the basi~/ rul es,that is,change the underlyi ng physi ca 1 processes.Whereas the present discharge regime can be described as a mixed first order auto- regressive and moving average process,the discharge regime under a .between the two years.This was true for both discharge and for chinook outmigration was examined.Even though the discharge peaks do not match salmon outmigration;only a single year of turbidity and sockeye salmon post-project scenario could include entirely different terms. 1 /;1 An important point is that ~~;!.~2~E_P!~;~~~!~,(theautoregressive (,.and movi ng average components)were the same in 1983 and in 1984 even /1 /\'0 though the actual time series,or "realizations,"looked very different I / / I at all we 11 between the two years,the process whi ch generated these pass model. peaks in both years was the same and can be described by an ARMA(l,l) OtnJ. In a sense,the proposed dams - fil ter on the di scharge regime,dampeni ng out the hi gh frequency pertur- bations and letting the low frequency (annual cycle)events pass. However,this is an oversimplification because a new element would be present if the dams are built -namely,power demand.Power demand is, not t~phase with the natural discharge fluctuations0and dam operation .\to accommodate power demand wi 11 change the mecham sms whi ch generate ,the current discharge regime. - DRAFT/PAGE 34,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Hale The important question is,how would the outmigration rate be affected if these discharge spikes were not present,as with a dam-regulated discharge regime?Further,what effects would these changes have on the population survival rate?Relatively high levels of discharge,and possibly four or five day peaks,in the late spring and early summer may ~ be necessary to facilitate normal outmigration timing of juvenile salmon.On the other hand,very high discharge levels at this time of year,as do occur naturally,may be harmful to juvenile chinook salmon if these floods displace the fry downstream from what would otherwise be their rearing areas. in future reports. ,.,."-,.,-,~,~".~~,,~--~..~"'-~_.'_'_~'-_•.~_..~ rTime series analysis is a powerful statistical tool which has many f \applications to the Susitna Aquatic Studies Program.The present paper ''''"--, only begins to use the full potential of the method.It could be useful to build Box-Jenkins models for the 36 year record of discharge at Gold Creek gaging station.Because this information is continuous,it can be digitized as monthly,weekly,daily,or even hourly means.Turbidity, temperature,and dissolved gas time series could also be modeled in this manner.Develop"ing time series models for the proposed post-project di scharge regime woul d be interesting to see whether the post-project discharge regime is also an ARMA(l,1)process.Intervention analysis, which is an extension of Box-Jenkins models concerned with a natural or human caused change to a system,would be an appropriate method to use (Box and Tiao 1975;Hipel et ale 1978;Thompson et ale 1982).One could determine if the intervention (construction of the dams)would have a DRAFT/PAGE 35,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Hale - significant effect on the time series processes.This method has been used to model the effect of the Aswan dam on the Ni 1e Ri ver and the Gardiner dam on the South Saskatchewan River in Canada (Hi pel et ale 1978).Before and after mean 1eve 1s can not be compared us i ng normal analysis of variance because the observations are serially correlated. Some preliminary work was done with the 36 year record of mean weekly discharge.This time series definitely requires differencing to achieve -stationarity because a period of discharge "activity"occurs from only about mi d-May to mi d-September (four months);duri ng the rest of the year,discharge level is low and has relatively little variance.The 1,768 records of mean weekly discharge from 1950 to 1983 are not plotted here,but the reader can picture what the series looks like by imagining v~? along snake whose practi ce it is to eat a ni ce fresh r~~bit_.every thi rd ~f\_.t.-.r'',(,..'' day.The rabbits travel down the length of the snake,undigested in this ~ ---0----/\----/\-- ~11 bbi t case.These rabbits come in all shapes -short and skinny,short and fat,tall and skinny,and tall and fat -but they were all from the same litter so the underlying processes which created them is the same.If the dams are built,the parents or even the species will change;the time series would perhaps look as if the snake had swallowed a weasel every third day:- Post-project DRAFT/PAGE 36,5/1/85 4/19/85,4/25/85,4/29/85 NUM4/Hale -- - With regard to the Susitna fish data,the work described in this paper on outmigration rates could be expanded to other species,years,and input variables.An excellent use of time series analysis would be to develop forecast models for the annual return of adult salmon or the annual total number of outmigrants.The adult salmon return of a par-- ticular year is strongly related to the return of the previous year (that is,when catch is high one year,it tends to be high for several years)and there is probably a periodicity component based on strong age classes.With such a model,one ,could predict the size of next year1s adult salmon return,a piece of information which would be very useful to both fishery and hydroelectric dam managers.However,the time series of adult salmon return to the Susitna River is not long enough (only seven or eight years of data).A minimum of about 40 or 50 observations is necessary (McCleary and Hay 1980;Huff 1983),although the method has been applied by Jensen (1985)to fish catch data with as few as 32 observations.The ann.ual abundance of adult chinook and coho salmon in the California fishery has been examined with time series analysis by Botsford et a1.(l982)and Peterman and Wong (1984)have looked at sockeye salmon cycles in British Columbia and Bristol Bay.For the present,analysis of salmon time series in the Susitna River will have to be restricted to daily r~tes of a single year. l'~"rFJ~~~~~~~- #~h~~,:{(tL~;f-R -h)tfLr - ,..... - DRAFT/PAGE 1,5/1/85 4/25/85,4/29/85 NUM4/Acknowledgements 5.0 ACKNOWLEDGEMENTS I would like to thank Kent Roth,who has run the outmigrant operation since its beginning in 1982,and Dr.Dana Schmidt,former Project Leader of the Resident and Juvenile Anadromous Fish project,for their valuable discussions on some of the ideas in this report.Allen Singham and Paul Suchanek also made helpful comments on a draft copy of the report,and Drew Crawford helped with the figures. Much of thi s work was done as a project for a course on time seri es analysis taught by Dr.J.Horowitz of the Department of Mathematics and Stati sties,University of Massachusetts.Hi s stimul ati ng course and assistance with the time series analysis are appreciated. I am grateful to Mary Ferber of the Alaska Resources Library for conducting a computerized literature search on ecological and fisheries applications of time series analysis,and to Skeers Word Processing Services for the typing of this report. DRAFT/PAGE 1,5/1/85, 4/19/85,4/25/85,4/29/85 NUM4/Literature Cited 6.0 LITERATURE CITED Alaska Department of Fish and Game.1983.Habitat relationships of juvenile salmon outmigration.Appendix H in Synopsis of the 1982 aquatic studies and analysis of fish and habitat relationships. 1983.Susitna Hydro Aquatic Studies,Alaska Department of Fish and Game,Anchorage,Alaska. Botsford,L.W.,R.D.Methot,Jr.,and J.E ..Wilen.1982.Cyclic co- variation in the California king salmon,Oncorhynchus tshawytscha, silver salmon,Q.kisutch,and dungeness crab,Cancer magister, fisheries.Fishery Bulletin 80:791-801. Box,G.E.P.,and M.Jenkins.1976.Time series analysis.Forecasting and control.Holden-Day,San Francisco. Box,G.E.P.,and G.C.Tiao.1975.Intervention analysis with applica-II!II'l tions to economic and environmental problems. American Statistical Association 70:70-79. Journal of the $4$ Brannon,E.l.,and E.O.Salo.(eds.).1982.Proceedings of the salmon and trout migratory behavior symosium.June 3-5,1981.University of Washington,Seattle,Washington. Bulmer,M.G.1978.The statistical analysis of the ten year cycle. Pages 141-153 .i!!.H.H.Shugart,Jr.(ed.).Time Series and Eco- logical Processes.SIAM-SIMS Conf.Sere 5.Society for Industrial and Applied Mathematics,Philadelphia.- ~, DRAFT/PAGE 2,5/1/85, 4/19/85,4/25/85,4/29/85 NUM4/literature Cited Cederholm,C.J.,and W.J.Scarlett.1982.Seasonal immigrations of juvenile salmonids into four small tributaries of the Clearwater River,Washington,1977 -1981.Pages 98 -110 ~E.l.Brannon and LO.Salo (eds.'.Proceedings of the salmon and trout migratory behavi or symposium.June 3-5,1981.University of Washington, Seattle,Washington. Chatfield,C.1984.The analysis of time series:an introduction. Chapman and Hall.london. Congleton,J.L.,S.K.Davis,and S.R.Foley.1982.Distribution, abundance and outmigration timing of chum and chinook salmon fry in the Skagit salt marsh.Pages 153 -163 ..i.!!.LL.Brannon and LO. Salo (eds.).Proceedings of the salmon and trout migratory behav- ior symposium.June 3-5,1981.Universit~of Washington,Seattle, Washington. Dixon,.W.J.,M.B.Brown,L.Engelman,J.W.Frane,M.A.Hill,R.I. Jennrich,and J.D.Toporek.(eds.).1981.BMDP statistical software. Cal ifornia. 1981.University of California Press,Berkely, EWT&A.1985.Instream Flow Relationships Report.Volume No.1. Prepared for Harza-Ebasco Susitna Joint Venture by E.Woody Trihey and Associates and Woodward-Clyde Consultants,Anchorage,Alaska. DRAFT/PAGE 3,5/1/85. 4/19/85,4/25/85,4/29/85 NUM4/literature Cited Godin,J-G.Y.1982.Migrations of salmonid fishes during early life history phases:daily and annual timing.Pages 22-50 in LL. Brannon and E.O.Salo (eds.).Proceedings of the salmon and trout migratory behavior symposium.June 3-5,1981.University of Washington,Seattle,Washington. Granger,C.W.J.,and P.Newbold.1977.Forecasting economic time series.Academic Press,New York.' Grau,E.G.1982.Is the lunar cycle a factor timing the onset of salmon migration?Pages 184-189 in E.L.Brannon and LO.Salo (eds.). Proceedings of the salmon and trout migratory behavior symposium. June 3-5,1981.University of Washington.Seattle,Washington. Hale,S.S.,P.M.Suchanek.and D.C.Schmidt.1984.Modelling of juvenile salmon and resident fish habitat.Part 7 in D.C.Schmidt, S.S.Hale,D.L.Crawford,and P.M.Suchanek.(eds.).1984. Resident and juvenile anadromous fish investigations (May -October 1983).Susitna Hydro Aquatic Studies.Report No.2.Alaska Department of Fish and Game,Anchorage,Alaska. Hartman,W.l.,W.R.Heard,and B.Drucker.1967.Migratory behavior of sockeye salmon fry and smolts.Journal of the Fisheries Research Board of Canada 24:2069-2099. -1 - ~, - ..- I DRAFT/PAGE 4,5/1/85, 4/19/85,4/25/85,4/29/85 NUM4/Literature Cited Hipel,K.W.,D.P.Lettenmaier,and A.I.McLeod.1978.Assessment of environmental impacts.Part one:intervention analysis.Environ- mental Management 2:529-535. Hoff,C.1983.A practical guide to Box-Jenkins forecasting. Wadsworth,London. Jensen,A.L.1985.Time series analysis and the forecasting of menhaden catch and CPlIE.North Ameri can Journal of Fisheri es Management 5:78-85. Kirkley,J.L,M.Pennington,and B.L Brown.1982.A short-term forecasting approach for analyzing the effects of harvesting quotas:application to the Georges Bank yellowtail flounder (Limanda ferruginea)fishery. 40:173-175. J.Cons.into Explor.Mer. -Liu,L-M.,and D.M.Hanssens.1980.Identification of multiple-input transfer function model s.BMDP stati sti cal software.Techni ca 1 Report No.68.Los Angeles. McCleary,R.,and R.A.Hay,Jr.1980.Applied time series analysis for the social sciences.Sage Publications,Beverly Hills,California. DRAFT/PAGE 5,5/1/85, 4/19/85, 4/25/85,4/29/85 NUM4/Literature Cited McDonald,J.1960.The behaviour of Pacific salmon fry during their downstream migration to fresh water and salt water nursery areas. Journal of the Fishery Research Board of Canada 17:655-676. Mendelssohn,R.1981.Using Box-Jenkins models to forecast fishery dynamics:identification,estimation,and checking.Fishery Bulletin 78:887-896. Mendelssohn,R.1982.Environmental influences on fish population dynamics:a multivariate time series approach.Paper presented at a meeting of the American Statistical Association.August,1982. Cincinnati,Ohio. Murray,L.C.,and R.J.Farber.1982.Time series analysis of an historical visibility data base. 16:2299-2308. Atmospheric Environment Nelson,R.1973.Applied time series analysis for managerial forecast- ing.Holden-Day,San Francisco,California. O'Heeron,M.K.,Jr.,and D.B.Ellis.1975.A comprehensive time series model for studyi ng the effects of reservo;r management on fi sh populations.Transactions of the American Fisheries Society 104:591-595. ('8 - DRAFT/PAGE 5.1,5/1/85, 4/19/85, 4/25/85,4/29/85 NUM4/Literature Cited Peterman,R.M.,and F.Y.C.Wong.1984.Cross correlations between reconstructed ocean abundances of Bristol Bay and British Columbia sockeye salmon (Oncorhynchus nerka).Canadian Journal of Fisheries and Aquatic Sciences 41:1814-1824. Platt,T.,and K.L.Denman.1975.Spectral analysis in ecology. Annual Review of Ecology and Systematics 6:189-210. Priestley,M.B.1981.Spectral analysis and time series.Vol 1: univariate series,Vol 2:multivariate series,prediction and control.Academic Press,London. Raymond,H.L.1968.Migration rates of yearling chinook salmon in relation to flows and impoundments in the Columbia and Snake Rivers.Transactions of the American Fisheries Society 97:356-359. Roth,K.J.,D.C.Gray,and D.C.Schmidt.1984.The outmigration of juvenile salmon from the Susitna River above the Chulitna River confluence.Part 1 in D.C.Schmidt,S.S.Hale,D.l.Crawford and P.M.Suchanek (eds.).1984.Resident and juvenile anadromous fish investigations (May -October 1983).Susitna Hydro Aquatic Studies.Report No.2.Alaska Department of Fish and Game. Anchorage,Alaska. Roth,K.J.,and M.E.Stratton.1985.The migration and growth of juvenile salmon in the"Susitna River.Part 1 in Resident and juvenile salmon fish investigations (May -October 1984).Susitna Aquatic Studies.Report No.7.Alaska Department of Fish and Game,Anchorage,Alaska. DRAFT/PAGE 7,5/1/85, 4/19/85,4/25/85,4/29/85 NUM4/Literature Cited Saila,S.B.,T.T.Polgar,D.J.Sheehy,and J.M.Flowers.1972.Corre- lations between alewife activity and environmental variables at a fishway.Transactions of the American Fisheries Society " 101:583-594. Saila,S.B.,M.Wigbout,and R.J.Lermit.1980.Comparison of some time series models for the analysis of fisheries data.J.Cons. into Explor.Mer.39:44-52. Salas,J.D.,and R.A.Smith.1981.Physical basis of stochastic models of annual flows.Water Resources Research.17:428-430. Shugart,H.H.,Jr.(ed.).1978.Time series and ecological processes. Proceedings of SIAM-SIMS Conference.Society for Industrial and Applied Mathematics,Philadelphia. Solomon,D.J.1982a.Migration and dispersion of juvenile brown and sea trout.Pages 136-145 in LL.Brannon and E.O.Salo (eds.). Proceedings of the salmon and trout migratory behavior symposium. June 3-5,1981.University of Washington,Seattle,Washington. Solomon,D.J.1982b.Smolt migration in Atlantic salmon (Salmo salar .h.)and sea trout (Salmo trutta .h.).Pages 196-203 in LL. Brannon and E.O.Salo (eds.).Proceedings of the salmon and trout migratory behavior symposium.June 3-5,1981.University of Washington,Seattle,Washington.- - - - - - - .-. DRAFT/PAGE 8,5/1/85, 4/19/85, 4/25/85,4/29/85 NUM4/Literature Cited Srikanthan,R.,T.A.McMahon,and J.L.Irish.1983.Time series analyses of annual flows of Australian streams.Journal of Hydrol- ogy.66:213-226. Stevens,D.E.,and L.W.Miller.1983.Effects of river flow on abun- dance of young chinook salmon,American shad,longfin smelt,and delta smelt in the Sacramento-San Joaquin River system.North American Journal of Fisheries Management 3:425-437. Still,P.J.,R.D.Lamke,J.E.Vaill,B.B.Bigelow,andJ.L.VanMaanen. 1984.Water resources data.Alaska.Water year 1983.U.S.G.S. Water-Data Report AK-83-1.U.S.Geological Survey,Anchorage, Alaska. Stocker,M.,and R.Hilborn.1981.Short-term forecasting in marine fish stocks.Canadian Journal of Fisheries and Aquatic Science. 38:1247-1254. Sturges,W.1983.On interpolating gappy records for time series analysis.Journal of Geophysical Research.88:9736-9740. Suchanek,P.M.,R.P.Marshall,S.S.Hale,and D.C.Schmidt.1984. Juvenile salmon rearing suitability criteria.Part 3 in D.C. Schmidt,S.S.Hale,D.L.Crawford,and P.M.Suchanek (eds.).1984. Resident and juvenile anadromous fish investigations (May -October 1983).Susitna Hydro Aquatic Studies.Report No.2.Alaska Department of Fish and Game.Anchorage,Alaska • 7-1 DRAFT/PAGE 9,5/1/8f, 4/19/85,4/25/85,4/29/85 NUM4/Literature Cited Taylor,T.G.,and F.J.Prochaska.1984.Incorporating unobserved cyclical stock movements in fishery catch equations:an applica-.... tion to the Florida blue crab fishery. Fisheries Management 4:67-74. North American Journal of Thompson,K.W.,M.L.Deaton,R.V.Foutz,J.Cairins,Jr.,and A.C. Hendricks.1982.Application of time series intervention analysis _ to fish ventilatory response data. and Aquatic Sciences 39:518-521. Canadian Journal of Fisheries - Van Winkle,W.,B.L.Kirk,and B.W.Rust.1979.Periodicities in Atlantic Coast striped bass (Morone saxatilis)commercial fisheries '~ data.Journal of the Fisheries Research Board of Canada 36:54-62.