HomeMy WebLinkAboutSUS378..
Fish Habitat and Instream Floi Relationships
in the Middle Reach of the Susitna River :
An Extrapolation r1ethodo lo gy
Clevel and R. Stewar d
and
Woody T ihey
Abstract
3
This document outlines a methodology for extrapolating habita-data
obtained at intensively studied areas to the r emainde r of the middle
Susitna River in order to describe the integrated response of fish habitat
\-li thin the river segment to streamflow variations under ice-free
conditions. It is assumed that the habitat availability and responses
determined at intensively studied sites are representative of habitat
conditions in all nonstudied sites within the same category. The
extrapolation is based on the supposition that the presence of upwelling is
essential for the successfu l spawning of chum and sockeye salmon , and that
rearing fish respond directly to instream hyaraulic and water quality
conditions.
The extrapolation method is applicable to evaluating existing and with-
project habitat.potential for a road range of habitat catcgor'es, species,
and life stages. At present we feel that only slough and side channel
habitats, chum and sockeye spawning, and chum and chinook rearing may be
profitably addressed on a quantitative basis.
. . .
Introduction
This document outlines a methodology for eva luating the av ailabi lity of
rearing and spawning habitat for sa lm on within the Talk eetna-to-Devi l
Canyon segme nt, also known as the midd le ·reach, of the Susitna River. Ou r
intent i s to provide a mea~s of ext~apolating habitat data obtaine~ at
intens ively studied areas to the re mainder of the midd l e river in order to
describe system r esponse to streamflow variations under ice-free condi -
tions. The method ranks study sites at which salmon utilization and
ha bitat data have been collected into dis c rete categories based upon
'several r e lated physica l and biological c rite ria •. Areas in the middle
river for .which little or -no fisheries data exists have been grouped with ·
intensively stu.died sites having simiiar phys ica l characteristics based on
field observati ons and a n examination of aerial photographs . It is impor-
tant that a positive relationship be demonstrated between salmon utiliza-
tion or habitat avai l ab ility and the hydraulic, geomo rphic, and hydrologic
characteristics used to rank studied and non-studied areas ·into distinct
categories such that s pawning and r ea ring habitat avail.ability indices
developed for the intensively studied sites may be considered representa-
tive of associated non-studi e d sites. Since an estimate of the surface
a r ea of all sites is availabl e for a wide range of ma instem discharge s , a
habitat avail ability index . determined separately for spawn ing and rearing
sa lm on, may be expressed fo r each category as a function of st r eamflow.
When habitat ava il abi li ty indices fo r all habitat categories are combined,
a composite picture eme r ges of the existing relationsh ip between habitat
avai l abi l ity and discharge fo r the entire middl~ reach of the Susitna
River .
This approach has the additiona l merit of being applicable to with-
'project impact analys es since the abiotic envi r onment r esu l t i ng from the
with -project flow regimen may be forecast with a comparativel y high degree
of confidence. The a ssumpt i on requi red is that expected changes i n habitat
qu a li ti; a nd quantity will be attended by adjustments in the dist ri bution
a nd r e l at i ve abundance of fis h popu l ations. Based on ou r cur r ent know l edge
'of annua l variations in habitat uti l i zat i on wi t h i n specifi c areas as a
fun ction of year-to-year va r iations i n discharge, thi s assumption appea r s
j us':ifieel .
Mention shoul d be made of the te r mi no l ogy use~ i n this paper . We are
concern ed with f i sh hat>itat, that is, the milieu of environmental
conditions to which a typica l i ndividua l of the species in question
responds both be havioral l y and physiologically. t1ore specifically, we are
interested in the environmenta l variables which i nfl uence the growth,
reproduction, and surviva l of the fish. Important biologi c al factors
i nc l ude food availabi l ity, parasitism o r -disease, and p r edation . It is
general ly recognized that temperature, wa t e r depth and velocity, cover or
.shelter, and strea mbed material are the most impo r tant physical variables
affecting the amount and quality of instream fis h habi tat (Hynes 1972).
Although it may be assumed that varying these physica l va r iables in time
a n d space has direct consequences in terms of fish distribution and
abundance , it shou l d be emphasized that habitat variabl es are usually not
independent of one another aod must be considered in combination. Under
some circumstances, however , the utility of specific areas as fish habitat
2
may be determined by one or two dominant environmental factors whose
importance overshadows the combined effects of a ll other biologic factors
and physical variables . An ~xample is the overriding importance of
adequate passage depths for adult salmon downstream of spawning areas. In
many cases , the facto r s which contro l or limit the fi s h population may not
be kno~n , primarily because their effects are exerted at locations outside
the w ~t~rshed or at times when no data are co ll ected. Flooding, str eambed
i nstability, anchor ice bu ildup, and ice floe scouring are transient yet
recurrent phenomena within the S usitna River which affect the l ong-term
quality and persistence of fish habitat.
Care must be taken to di st i ngui sh bet\'leen fish habit at and habitat ~
The latter term designates major categories of aquat ic habitat having
visually recognizable hydraulic and mo r phologic characteristics that are
apparent in aerial photography (F igure 1). Six habitat ty pes have bee n
identified within the middle reach of the Susitna Rive r: mainstem, side
channe l , side s l ough , upland slough , tributary, and tributary mo uth (ADF&G
1983). The geographica l location and persistence of ce rtain habitat types,
such as tributa r ies and their mouths, are genera lly fixed. In other
in stances , a given sect i on of the river may exist as one habitat type at
high discharges and as another at lower f1 ows . An examp l e is the trans-
form ation of some side channe l s into s i de sloughs as mainstem stage r ecedes
below the thalweg elevation at their heads . An important characteristic of
these sites, in regard to their value as fish habitat, appears to be the
frequency and duration of time they exist as side channe l s or side sloughs.
3
c .... ,... ••• ,
lla•Mt
'-lldJ &.t , ..
W.ntot tl'lu• b:r
fT'IIoi'Uitllft w.a ttr
lr.t rf.JU t lt "''UI.on
Figu re 1.
Cl u,...alttp,h.l1n41
or no•llb t.,
'"~ti'IUI Df
rNln,IUBWit U
t wttttl cln.s llo,.
Classification of aqua t ic habitat types
reach of the Susitna River . Source : E.
for
Sl,lflfC.:t ltU
ugtt•lta
hl.af'd • tnll
,~.,.., uo~rldot
t he Talkeetna -to -Devil
T r i hey & Associ ate s ~~ oody
S\lrflc:t ltU
upoucr
l ll t tm.~.t "-""
I I'IIIS,.;uwt:l~lfl
Canyon
( 19 84 ) •
Utilization of Habitat Types
Uti li zation data availab l e from 1981-B3 spawne rs surveys by the Alaska
Department of F.i sh and Game (ADF&G 1984a) suggest that t ri butaries , side
s lou ghs a nd , to a lesser extent, side chann els are the primary spawning
areas of the five spec ies of sa lmon which occur in the Susitna Rive r
(Figur e 2). A comparatively sma ll number of fis h spawn in mainstem , upl and
slough. and tributary mouth habitats. Since the ext ent and quality of
tr ibut a ry habitat is basi cally unaffec t ed by mainstem discharge and
temperature, we have chosen to omit evaluation of tributary habitat f rom
the extrapol ation analysis.
Chum a nd sockeye sa l mon a r e . the most abundant of the three species which
spa\'ln in habitat types other than tribu tari es· in the T a lk eetna-to -Devil
Canyon reach of the Susitna River . Small numbers of pink salmon utilize
side channels and . side sloughs for spa\-tning du ring even numbered years and
are thought to outmigrate within 3 to 5 days after eme r gence from spawning
gravels . Therefore, pink salmon are not considered significant in an
analysis of existing habit at cond i tions .
Of the chum salmon spawni n g observed withi n mainstem , side channel, and
side s l ough areas, the latter habitat type appea r s to be the most pre-
ferred. Approximately 80~ of all chum sa l mon spawning outside of
tributaries has been documented in side sloughs (ADF&G 1981, 1982, 1984a).
Side channel and mainstem areas, however , are often char acteri z ed by highly
turbid water in which spawning fish or their r edds are difficult to detect,
possibl y causi ng an underestimate of their va lu e as spawning habitat .
Tables 1 and 2 summa r ize spawne r survey information obtained for side
4
Figure 2 .
MS-MAINST E M
SC-SIDE CHANNEL
SL-UPLAND and SIDE SLOUGHS
T -TRIBUTARIES
i!f'-PRIMARY SPAWNING HABITAT
t -SECONDARY SPAWNING HABITAT
4
'-I INCIDENTAL SPAWNING HA-BITAT
MS SC S L T
~
\
'
COHO
MS SC SL T
CHINOOK
MS SC SL T
\ \
\
\
SOCKEYE
MS S C SL T
PINK
MS SC S L T
CHU 1
Relative distribution of salmon spawning within different
habi at types of the middl Susitna River (ADF&G 1984c).
TaDle 1. Percent distribution of chum and sockeye (second run)
salmon reported for side sloughs in the middle Susitna
River based on data averaged for a lhrec-year period
(1981-83). Oata obtained from AOF&G (1984a).
Percent Distribution
Slough River Mile Chuca Salmon Sockeye Sa 1 con
1 99 .6 0.1 0
2 100 .2 1.3 0
38 101.4 * 0.3
7 113 .2 0 0
8 1:.3 .7 4 .9 0
8C 121.9 0.9 0 .1
m 122 .2 3.0 0.3
f1oose 123 .5 4.2 1.3
8A 125 .4 16 .1 13 .7
B 126 .3 1.6 0.6
9 128 .3 ll .8 0.7
9!\ 133 .8 6.6 0 .1
11 135 .3 lJ.O 69 .7
13 135 .9 0 .1 0
14 135 .9 0 0
16 137 .3 * 0
17 138 .9 2.5 0.5
2U 140 .0 1.8 0 .1
21 14 1.1 21.5 12 .6
22 144.5 5.5 0
21A 145 .3 0 .1 0
-----
* Tr ace
Tab l e 2 . Chum sa l mon spawning reported for mainstem and side channel
areas in the middle Susitna River, 1981 -83 . Data obtained
from ADF&G (1981, 1982 , 1984a).
----
Approximate Habitat Seawner Utilizatio~
Category!! 19iff 1~8~ 1~83 River 1·1il e
1v0 .5 II + 0 0
114 .9 II 0 ++ ++
115 .1 II 0 0 ++
119.0 0 0 ++
128 .6 II 0 ++ 0
129 .2 Vll + 0 0
129 .8 II + + 0
130 .5 II + 0 0
131.1 VII + + +
131.3 IV 0 ++· +
136 .0 II + ++ +++
136 .8 0 0 ++
137 .4 11 0 ++ 0
138 .2 0 + 0
138 .9 u 0 ++ ++ ~/
148 .2 0 +++ 0
l/ See Table 5 for habitat category descriptions . Si tes which are not
assigned a category number are found in areas \·lhich are classified
as mainstem habit~t at both 23 ,000 and 9 ,000 cfs .
?:./ Utilization
u
+
++
+++
Codes:
No spawners or redds reported
Less than 10 spawners or redds
10 to 100 spawners repor ted
Over 100 spawners reported
reported
!I Eleven spawning sockeye salmon observed 9/15/83
slough, side channel , and mainstem areas vlithi n the middle r each during
1981 -83 . Th e number of chum salmon r eported from these three habitat types
averaged 2,300 fish/year ove r this time period.
In 1983 , 11 sockeye and 56 chum salmon adults were obser ved spawning in the
mainstem Susitna Ri vr:i' i mmediately upstream of the mout h of the Ind ian
River (AOF&G 19 84 a}. This is the only recorded occurrence of sockeye
spawning in areas other than side s lou gh habitats. In r ega rd to side
s lou gh spaw ni ng, a n average of 760 sockey e spawned a nnu ally in t he
Ta l keetna-to-Oevil Canyon reach. These fish were distributed amo ng 12 of
the 2 1 s ide s l oughs found in the 50-mile l o ng r each of the midd l e river
(Table 1). It sho ul d be noted that chum and sockeye ~almon spa wn ing a r eas
overlap ped wi t hin all of the s ide sloughs. in which sockeye redds were fou nd
(ADF &G 19t34a).
Juvenil e ch u m and chinook sa l mon are the most abundant salmonid spec i es
which rear in the s i de slough and side channel habi tats of the midd l e
Susitna Rive r (Figure 3). Th ey a r e therefore most suscepti bl e in ter ms of
overa ll numbers affected to rearin g habitat perturbations. For this
redson, these two species have been selected for eva luating r earing habitat
with 1n the entire middle reach of the Sus itna River. Habitat for juveni l e
sa lm on i s gene r a ll y provided for by a ll habitat types ; however, fi s h
densities a r e usua lly highest in side slough ~nd side channel areas. The
sole exception is coho salmon, which r ea r predominatly in upland s l oughs.
Extensive sampling fo r j~veniles has not been conducted in mainstem habi-
tats, large ly due to sampl i ng gear ineff i ciency in the typical l y deep, fast
and turbid waters of the mai nstem river. Therefore, uti 1 i zat ion of the
5
Trlbutarloa
59.6~
Sl do Chonnols ---~
4.0~
Chinook
Sldo Channols
21 .7 ..
Upland Sloughs
6.5 ..
Sldo Sloughs
54 .4 ..
Trlbulorlos
37.0 ..
RELATIVE ABU DANCE
OF JUVE ILE SALMON
n• aeso
Trlbulorloa __ ,._
0.5~
Sockoyo
Coho
Sldo Chonnols
Upland Sloughs
49.7"'
Figure 3 . Relative abundance and distributio~ of juvenile salmon within different
habitat types of the midd l e Susitna River (AOF&G 1984b).
lateral margins of these habitats by juvenile salmon may be gre ater than
indicated by the avai l ab le data.
Surface Area Response of Habitat Types
The total su rf ace area or each habitat type in the Talkeenta-to-Devil
Canyon r each has been estimated for mainstem discharges ranging from 9,000
to 23,000 cfs (USGS gage 15292000) using digital measurements on
1 inch= 1,000 feet aerial photographs (Figure 4). The s urface areas
assoc i ated with upland sloughs, tributaries and tributary mouths
co llectively represent l es s than 1.3% of the tota l surface a r ea of the
middle reach, and habitat types exhibit little change in response to
mainstem discharge. At times surface areas of these habitat types may
respond mo r e to seasonal patterns of local precipitation and runoff than to
var i ations in mainstem discharge .
Compa r atively large differences in surface areas of mainstem, side channel ,
and side s lough habitat is apparent between mainstem discharges of 9,000
and 23 ,000 cfs . From an inspect i on of Figure 4 it may be seen that side
channe l and side slough surface areas are inversely related. Fish distri -
buti on data a l so indicate side sloughs and side channels are the most
extensively utilized portions of the river corr i dor. He nce, it is these
habitat types which are of principal interest in terms of assessing
existing and potential fisheries values.
6
Iii"
C!l ...
u
(II -Q
~
<{
C!l u
0 .._
:J
CJ)
ttl
0
1-
Figure 4.
2500
MAINSTEM
100
500
400 -I
0 300 ~
10 0
(C
"' Q c.
5 CJ)
c ...
c;'
0
Q
> ...
Q
so tJ -40 :J
...!!
30
UPLAND SLOUGH
20
0.5
10
L-~---L--~--~_J __ _L __ _L __ ~--L-~---L--~ __ L__j0.1
9 10 11 12 1 J 14 15 16 17 18 19 20 21 22 23
Mainstem Discharge at Gold Creek (x10l, cfs)
Surface area responses of habitat types to mainstem discharges
in the middle Susitna River . Source: E . Woody Trihey &
Associates (1984}.
Habitat Attribute Preferences
Considerable information has been gained by AOF&G studies of the habitat
preferences exh i bited by spawning chum and sockeye salmon (ADF &G 1984c}.
Preference for a given habitat va ri ab l e is exp r essed in the form of a
suitability function which stochastical l y describes the relat i onship
between the va r iable and fish behavio r (Baldrige and Amos 1981). Species-
specific suitability functions, or criteria, developed for spawning chum
and sockeye salmon are based on a large number of measurements obtained at
redd sites in side slough and side channel a r eas of the middle Susitna
River. These data are modified slightly to account for the proportional
d i stribution of acceptable habitat within the immediate a r eas in which
redds were located. Suitab~ lity criteria have been defined for spawning
chum and sockeye for several habitat attributes, including depth, velocity,
substrate and upw e lling (Figures 5 and 6). For both species, depths
exceeding 0.8 feet were found to have a negligible effect on redd site
selection in side sloughs and side channels. Velocities selected most
frequently by chum and sockeye salmon fall within the range of 0.0 to
1.0 feet/second. Accordingly, maximal suitability values are assigned to
these velocities. Utilization declines gradually at higher velocities but
rapidly at lower velocities, resulting in slightly skewed, bell-shaped
suitability curves . Substrate sizes preferred by the two species are
similar, although chum salmon are capable of excavating larger bed
materials than sockeye due to their larger body size. The presence of
groundwater upwelling has been directly linked with redd site selection by
both chum and sockeye salmon spawning within the middle reach of the
Susitna River. Since measurements of upwelling rates are diff1cult to
7
1 .0 1.0-
I
oc 0 .8 0.8-
0
'Q
~ ,.. -0..6 0.6-
.0 • -:;
(I) 0 .4 0.4-
0.2 0.2-v 0 A 0
0 2 3 • 5 0 \ 2 y • 8 9
Velocity (ftlsec) Depth (ft)
1.0
oc 0 .8 • 'Q
~ ,..
0.6
.0
~ :;
(I) 0.4
0.2
0
10 12
Substrate
Figure 5 . Velocity. depth. and substrate suitability criteria for
spawning chum salmon i n side sloughs and side channels of
the middle Susitna Rive r. Modified from ADF&G (l984c).
1.0 1.0-
>( 0.8 0.8-
0
'0 .s
>-O.G 0.6-
-D
!! :; 0.4 0.4-en
0.2 Q.2 -
0 0 ~
0 0 I l T
2 3 4 5 1 2 6 9
Ve l ocity {ft/sec) Dopth {ft)
1 .0
>( 0.6
0
'0 .s
?: 0.6
-D
"'
:;) 0.4 en
0.2
0
2 6 10 1~
Substrate
Figure 6 . Velocity, depth, and substrate suitabi lity criteria for
spawning sockeye salmon in side s loughs and s id e channels
of the middle Susitna River . Modified from ADF&G (1984c}.
obtain in the field, a simple binary criteria indicating preference or no
preference for areas in which upwelling is present or absent has been
assigned to both species.
Suitabi Hty functions similar to those described above for spawning have
been developed to assess rearing habitat availability in side sloughs and
side channels for juvenile chinook and c~um salmon (Figure~ 7 and 8). The
physical variables generally considered important to rearing salmon include
water depth, velocity, and the type and amount of cover present. Cover is
used by salmonid juveniles as a means of avoiding predation and unfavorable
water velocities. Instream objects, s~ch as submerged macrophytes, large
substrates and organic debris, and overhanging vegetation in near shore
zones provide shelter for juvenile salmonids. A positive correlation
between chinook juvenile densities and turbidity levels has a lso been
reported, suggesting that highly turbid water may be preferred by this
species for its cover value {AOF &G 1984b).
Habitat Availability (Spawning and Rearing WUA)
Su fficient data has been obtained to effectively model the availability of
spawni ng and rearing habitat at several side slough and side channel study
sites. The Weighted Usable Area {WUA)--an index of habitat availability--
was calculated for each.species/life stage and discharge of interest at
each study site. The calculation of WUA roughly equates the area of sub-
optimal fish habitat within the study site to an equivalent area of optimal
habitat. A sample total surface are~ and WUA resp.onse curve (i.e., WUA
expressed as a function of mainstem discharge} is presented in figu re 9
for chinook salmon rearing at Slough 21. Also shown in Figure 9 is the
8
1.0
J( 0.8
G
~ = !: 0.6
.Q
Cl
:;
0.4 (I)
0.2
1 .0
J( 0.8
G
~ E.
!: 0.6
I:J
Cl -~
(I) 0.4
0.2
0
1 2 3
Velocity (lileec)
0-5 6-25 26-50 51-75 76-100
Percent Cover
1.0-
0 .8-
0.6-
0.4-
0.2-
0
0
1.0
o.e
0.2
A
v
A
I l -, I
1 2 8
Depth (ft)
/
12 3 4 56 78 9
Cover Type
Figure 7. Velocity. depth and cover suitability criteria for juvenile
chinook salmon under clear water (solid line) and turbid
water (dashed line) conditions i n side sloughs and side
channels of the middle Susitna River. Cover type codes may
be found in ADF&G (1984b).
1.0
>< 0 .8
CD
"0
.5
>-
0.6
.D
Cll -::s
(I) 0.4
0.2
1.0
>< 0.8
CD
"0
.5
~ 0.6
:0
Cll
:;
(I) 0.4
0.2
0
1.0-
0.8-
0.6-
0.4-
0.2•
0
1 2 3
Velocity (ft/sec)
Percent Cover
0
A • ~ y A 1
Depth (ft)
12 3 45 6 7 8 9
Cover Type
9
Figure 8. Velocity, depth and cover suitability criteria for juvenile
chum salmon in side sloughs and side channels of the middle
Susitna River. Cover type codes may be found in AOF&G (1984b).
30~------------------------------------------------------~
25 Controlling Olschorgo
....
0
0
0 ...
I( 20 .. -....
Ill
41 .. <
Q 15
:0
Ill ,,
::> ,
41 --a. 10
'Qj
~
Initial Broaching Olacharga
~ '' '4 ' ', .. ---·
----::---____ ... -'
L--.:.:-.::-:::.:.:::.::::: ----. . .
5 Ran;a In WUA when alough !Iowa aro
not controllod by malnetem dlacharge
0 5 10 15
,'
,
I
I ,
20 25
Molnstcm Discharge (cfa x 1000)
30 35 40
Figure 9. Habitat response curve for juvenile chinook salmon at Slough 21 (RM 141.8)
within the extrapolation limits of the habitat model. Clear and turbid
water suitability criteria were used to simulate habitat availability for
discharges less than 18,000 cfs and greater than 22,700 cfs, respectively.
The range in WUA indicated for the unbreached condition is based on slough
flows of 5 to 10 cfs. Modified from ADF&G (1984b).
mainstem discharge at ~othich the head of Slough 21 is overtopped. It can be
seen that the WUA for ch inook juveniles i s maximal wh en the study site
possesses side channel characteristics.
A total of thre e side s l o u gh and fo ur s ide c hannel study s ites have been
evaluated to dat e for chum a nd chinook r e?.ri ng and chum and sockeye
spawning habitat ava i 1 abi 1 ity . The rearing and spawning WUA present at
each of these sites is li sted in Tables 3 and 4 for ma instem discharges of
9,000 , 12,500, 16,000 , and 23,000 c.fs. Habit a t could not be modeled for
several flow-site co mbinations due to hydraulic data limitations . The
general impression impa rted by the tabled va lues is that both rearing and
s pawning WUA tend to peak in the 16,000 to 23,000 cfs r ange for mos t study
sites .
There a r e two dist inct advantages associated with the use of WUA as an
index of available fish habitat . The first is th a t a \olide r ange of fl ow
conditions may be simulated and compa r ed , including flows typical of wet ,
normal, and dry water years. I t is therefore possible to evaluate habitat
availability under project ed post-project flow conditions. A second
advantage to mode li ng WUA is the modest expenditure of time and money it
r equi r es relative to an extens ive fish samp ling program, often spread out
ove r sever al yea r s, which attempts to define habitat quality on the bas ~s
of utilization data. For a river as l a r ge and complex as the Susitna, an
exhaustive su rvP.y of fish populations i s cost p rohibitive . Suf f i ~i ent
fi she1 i es data ha s been co 11 ected , however, to cone 1 ude that fish
distribution and abundance varies cons ide r ably between sites within each
habitat type . Supe ri mposed on this spatial variabi l ity are short-and
9
,
Table 3. Chinook and chum salmon rearing habitat WUA determined for selected modeling sites
in the middle Susfnta River at mainstem discharges of 9,000, 12,500, 16,000 and 23,000 cfs .
The maximum WUA dOd the associated mafnstem discharge (Qmax) for each study site is indicated.
Base Over-WEIGHTED USABLE AREA (x 11 000}
Mode 11 ncJ-1
Slough topping
·flow Discharge Qmax Maximum
Site (cfs) (cfs) Species (cfs) WUA 9,000 12,500 16,000 23 ,000
----··-
Slough 8A 10 33,000 chinook 2.1 2.1 2.1 2.1
chum 22.2 22.2 22.2 22.2
Slough 9 10 16,000 chinook 21,700 33.4 1.8 1.8 30.2 30.4
chum 22,900 25.8 22.3 22.3 22.3 25.8
Side Channel 10 5 19,000 chinook 21,100 16.6 3.7 3.7 3.7 15.2
-chum 21,600 17.1 10.8 10.8 10.8 16.5
Lower Side 5,000 chinook 5,900 27.0 25.4 15.0 11.8
Channel 11 chum 5,900 37.2 35.9 21.0 14.0
Upper Side 6 13,000 chinook 16,000 32.5 10.1 10.1 32.5 25.6
Channe l 11 chum 18,'000 31.7 22.9 22.9 27.3 26.4
Side Channel 21 20 9,000 chinook 14,900 33.5 31.9 31.9 30.6 25.1
chum 14,900 42.3 40.6 40.6 39.9 32.2
Slough 21 5 18,000 chinook 25,000 25.2 1.2 1.2 1.2 20.8
chum 25,700 20.7 ~6.4 16.4 16.4 17.2
l/ Or.iy those sites for which hydraulic simulation data were obtained are presented.
Table 4 . Chum and sockeye salmon spawning habitat WUA determined for selected modeling sites in the middle
Susitna River at mainstem discharges of 9 ,000 , 12 ,500 , 16,000 and 23 ,000 cfs . The maximum HUA
and the assoc iated mainstem discharge (Qmax) for each study site is indicated .
_ ...... _ ··----
Base Over-WEIGHTED USABLE AREA {x 1 2000)
1·1ode 1 i ngll
Slough topping
Flow Discharge Qmax 14aximum
Site (cfs} (cfs) Species (cfs) WUA 9,000 12 ,500 16,000 23 ,000
Slough 8A 10 33,000 chum 5.1 5.1 5.1 5.1
sockeye 6 .0 6 .0 6 .0 6 .U
Slough 9 10 16,000 chum 26,700 9.1 3.4 3.4 3.4 8.5
sockeye 24 ,800 7.0 5 .6 5.6 5.6 6.8
Side Channel 10 5 19,000 chum 24,900 6.1 0.4 0 .4 0.4 3 .5
sockeye 22 ,900 7,3 1.0 1.0 l.U 7.3
Lower Side 5,000 chum 5,900 32 .8 27 .2 24 .4 19 .3
Channel 11 sockeye 5,900 28 ,2 20 .8 16 .6 12 .8
Upper Side 5 13,000 chum 22 ,800 14.4 5.7 5.7 6.1 14 .3
Channel 11 sockeye 20,600 14 .4 8 .2 8.2 9 .4 11 .8
Side Channel 21 20 9 ,000 chum 12,700 3.8 3 .0 3.5 3.2 1.3
sockeye 12 ,000 4.8 4 .4 3.8 2 .6 0 .8
Slaugh 21 5 18 ,000 chum 28,700 16.4 6 .9 6.9 6.9 5.9
sockeye 27,300 13 ,7 8 .0 8.0 8 .0 7.5
--·-.. -
.!. Only those sites for which hydraulic simulat i on were obtained are presen ted .
long-term temporal fluctuations in population sizes as well as sampling
biases associated with deep,· fast. and turbi d water.
The apparent heterogeneity among study sites within each habitat type is
corroborated by the differences observed in WUA estimates. Side channels,
for example, do not pro vi de spawni n.g or rearing habitat which is
proportional to their wetted surface area or the volu~~~e of water which they
convey. Simnarly. habitat availability varies considerably among the
different side sloughs studied. To illustrate this point. chum salmon
spawning WUA is plotted in Figure 10 as a function of surface area for six
mo~eled sloughs at typical clear water base flows for each site. The
WUA :surface area ratio may be viewed as an efficiency index since it
imp11e$ that the availability of habitat may be m<?re economical or
productive with regard to stream surface area at certain streamflows.
Figure 10 indicates that Slough 10 contains far less WUA per unit surface
area than do Sloughs 21, SA, and Upper Side Cha nnel 11 {the latter site is
a slough at mainstem discharges of less than 16.000 c~s). Note that the
general ranking of sloughs based on their efficiency inde x values is
similar to their percentile ranking based on utilization data {c.f.,
Table 1).
Extrapolation Method
Due to the natural variability within habitat types, we have divided non-
tributary areas of the middle rfver into discrete categories, each con-
sisting of a population of sites having similar large-scale physical
characteristics. A necessary assumption is that t·he biological potential
of all sites within a category may be accurately aescribed by habitat
10
6.0~-----------------------------------------------,
"" 0
0
0
)(
5 .0 -
~ 4.0
:::>
~
0 :z
z 3: 3.0-
< 0.
U)
:z
0
~ 2.0 -J <
U)
~
2 ......
0 1.0.
0 S L OUCH 21
0 SLOUGH 0
UPPER SIDE CHANNEL 1 1 0 0 SLOUGH 8A
SIDE CHANNEL 2 1 0
,... SIDE CHANNEL 10 0~------~.------~.~~~~.~~~,r-------,r------i
0 20.0 40.0 GO 0 80.0 100.0 120.0
SIDE SLOUGH SURFACE AREA (x 1000)
F;gure 10. Relationship between WUA and surface area at typical base
slough flows for six mode ling sites within the middle
Susitna River.
indices determined for one or more representative study sites. This .
assumption 1s valid ,f (1) the phy sica 1 vari ab 1 es incorporated into the
habitat model are the dominant environmental factors affecting fish di stri-
bution. and (2) the suftability functions relating fish behavior to the
physical variables are accurate. In cases where the availability of habi-
tat is determined by a single controlling factor, this factor will be used
to initially screen sites prior to the application of modeling results.
For example. chum and sockeye . sockeye spawning haibtat wi 11 be evaluated
for each site only if it is determined that passage depths are suitable and
upwelling is present.
The physical data necessary to stratify side channel and side slough
habitat types falls into two general categories. The first category
includes existing data which may be compiled from published and unpublished
sources. These data and the pre 1 i mi 1nary ana lyses conducted with them are
discussed below in the context of _study site selection. A second category
consists of physical and biological data which may be collected during the
1984 field season. These .include variables identified as important irr the
preliminary analy ~is. and additional information to be gathered at .both
modeled and unmodeled sites. As discussed below. the second category of
data will also be used to assess the representativeness of the selected
modeling sites.
Analytical C~nstraints
The .h"abitat types which are to be initially evaluated for spawning and
rearing habitat availability include side sloug'ls ~~t:l side channels. At
11
present. these are the only habitat types meeting the following criteria:
(1) they represent a significant proportion of total spawning and rearing
habitat within the middle reach of the Susitna River; (2) their
distribution and cumulative surface areas may be expected to change
significantly under post-project flow conditions; and (3) the existing data
base is sufficient to support a quantitative analysis. It is anticipated
that selected mainstem and upland slough areas will be included as the
ana lysis progresses.
Chum and sockeye salmon are the primary species of interest in regard to
spawning habitat availability within side sloughs and side channels.
Rearing habitat availability will be evaluated for chum and · chinook salmon
within these habitat types. These species have been initially selected due
to their relative abundance w1thin side sloughs and side channels. and
because habitat suitability criteria are available for use in estimating
WUA. Spawning and rearing life stages are to be evaluated for similar
reasons. On a population level. the perpetuation of these life history
phases at levels .supported by existing side sloughs and side channels is of
critical importance to the maintenance of salmon stocks within the middle
Susitna River.
At present. we feel that only side slough and side channel habitats. chum
and sockeye spawning. and chum and chinook rearing may be profitably
addressed on a quantitative basis. It should be stressed that the
extrapolation method is theoretically applicable ·to a much wider range of
hal>i tat types. species and 1 i fe stages. Given reasonab 1 e cause · and
s ufficient data. additional habitat types and species/life stages can be
added to the analysis at a later date.
12
Analysis of ADF&G data and ae ri a l reconnaissance photography has revealed
consistent patterns in the morphological and hydrological f eatures crf side
slou gh and side ch ann e l sites located in the middle river. The ob s erved
patterns form the basis for a preli minary stratification of these habitat
types into severa l categories , and will be discussed separately be l ow for
spaw ning and r earing habitat evaluations. They should not be construed as
the fi nal array of catego~ies to be used in the extrapolation ana lysis.
Th e classification represents an initial attempt at s tratifi cation and its
principal value at this time i s to facilitdte study site selecti on for FY85
fi e ld studies in the middle river. The study sites a r e cu rren tly being
investigated for rearing and spawni n g habitat utili zation and avai l ab ility
following procedures which are consistent with the extrapola tion
methodology .
Rearing Habitat
Si te-specific i nvestigations of r earing habitat have indicated that rearing
fish are directly influenced by cover and velocity. These habitat
att ribut es are functions of streamflow, cha nn e l st ructure and , in the
Sus itna River, turbidity. Hence a f u ndamenta l assumption for extrapo lating
site-specific habitat responses to nonstudied a r eas i s that portions of th~
river with channel structu r e , hydraulic cha r acteristics and turbidity
levels similar to the studied areas will possess similar habitat potential
and responses .
Based on this a s sumption, slough , side channel, and ma i nstem areas
pertinent to the evaluation of existing and potential r eari ng habitat were
13
categorized us ing various mo rphol ogic and hydraulic features discernib l e in
aeria l photogr aphy obtained at mainstem discharges of 23,000, 16,000,
12,500, and 9,000 cfs. Prima ry emphasis was placed on the transformation
occu rrin g to mainstem and side channe l areas in tile 23 ,000 and 9,000 cfs
pho tog ra phy . Th ese f l ows fall within the range of moderate to loH d i s -
charges conveyed by the midd l e Susitna River during the ice-free months of
the year. Aerial photogr aphs obtai ned !1arch 2, 1983 when the river was
covered with ice were also inspected and ope n leads which appea r ed to be
caused by upwelling were ide ntified. A vi sual compariso~ of the three sets
of photographs provided the basi s for a preliminary categorization of more
than 100 sites. A description of the categori es and numbe r of sites wi thin
each of the categories is prese nted in Tab l e 5. The cat ego ri es are a rrange d
i n descending order of impo rt ance based on the following criteria :
(1) relevance to analyses of existlng and potential (i.e., post-project)
rearing habitat; (2) total number of sites and surface areas affected; and ,
(3) ease and r eliabi li ty of mode l app li cation to rep r esentative study
sites. Also indicated is the number of sites fo r which chum and chinook
salmon rearing models have been developed and habitat availability indices
have bee n calculated. Given s ufficient t i me and money, we would r ecommend
that a mini mum of t hree habitat modeling sites be established for each
category. Resource constraints, however, dictate that a smaller number of
categories and study sites be samp l ed.
Habitat mode lin g results for intensively s tudied sites can be us ed to
estimate the total amount of r earing habitat presently available for
juvenile chum and chinook salmon at simila r locations within the middle
river. For this analysis the ratio between WUA and total su rface area of
the site will be dete r mined at four mainstem discharges (9,000, 12 ,500,
14
Table 5. Rearing habitat ca .egories , the approxi ma te numbe r of middle ri ver sites within each cate gory ,
and the nu mbe r of habitat modeling sites completed and reco mmended for future study for each
category.
-------------------~------------------·--------·----------A~p--prox-i~m-a_t_e----~N~u-m -be_r __ o~f~M~o~d~e~li~n-g-S~~it~e-s
Category
II
Ill
IV
v
VI
VII
VIII
I X
Description
Numoer
of Sites
-----------#----·--------------------------·-
Distinct hanne l s with clear water visible in 23 ,000 and
9,000 cf photography and apparent th erma l leads in
March, ~83 photography .
35
Dist ac t s i de channel ar eas at 23,000 cfs wh ich contain 21
cl ear \<tater at 9,000 cfs and have apparent therma l leads
in t1 arch photography .
Distinct side channel areas at 23,000 cfs which contain 14
clear water at 9,000 cfs without apparent thermal leads
in March photography .
Distinct mafnstem or s iQe channel areas at 23 ,000 cfs 18
which become or r ema in side channe l s at 9,000 cfs .
Inaist1nct ma instem or side channel areas (sh oals) at 14
23 ,000 cfs which become distinct side channels at 9,000 cfs.
Indistinct ma instem or side c11annel areas (shoa ls) at 11
23,000 cfs which remain indistin ct at 9,000 cfs .
Indistinct mainstem or side channel areas (shoals) at 5
23,000 cfs which contain clear water at 9,000 cfs and
have appa ren t leads in Narch photography.
Indistinct ma fnstem or side channel areas (shoa ls) at 3
23,000 cfs which contain clear water at 9,000 cfs without
apparent therma 1 1 eads in t1arch photography.
Distinct and indi s tinct side chann e l areas at 23 ,000 cfs 9
whicn become dew atered at 9,000 cfs.
Comp letea Reco rmlended !I
6 a Io
4 0 0 1
7 0 1 1
1 1 2 2
1 2 2 2
0 1 1 1
0 1 1 1
0 1 1 1
0 0 0 1
0 0 0 0
l/ Recommended habi tat mode l i ng sites are based on possible t otals of 6, 8, or 10 mode li ng sites .
16,000, and 23 ,000 cfs). Total WUA for each category will be estimated by
multiplyin g the mean WUA:surface area ratio determined at intensively
studied sites by the cumulative surface area of all si tes within the same
catego ry. Category HUAs will be summed t o esti mate the total amount of
rearing hab itat ava il ab l e i n the middle rive r for juvenile chum and chinook
salmon at each disch a rge.
The information us e d to stratify the midd l e rive r and evaluate the habitat
potentia l of variou s categories will be considerabl y r efined on the ba s is
o f data obtained in FY SS. It will be necessary to verify the preli minary
c l assi ficati on scheme , de termine the representativeness of modeling s ites,
and defi ne existing r elationships with i n nonstudied categories.
Spawni ng Habitat
A suff ici ent number of s i de slough study sites have been evaluated in
previous ADF&G investigations to suppo rt an ext r apolation of ch u m and
sockeye spawning WUA determined fo r these sites to the r ema inde r of the
s i de s l oughs in the middle river havi ng s imilar morpholo gica l an d hydro-
logi cal characte ri stics. Th ese studi es conc lude that upw e llin g is a pre-
r eq ui s ite for successful chum and sockeye spaw ning, with substrate , depth
and ve l oci ty bei ng i mpo r tant secondary co nsiderations.
The ext r apo l ation methodology for chum and sockeye salmon spawning is based
on the premise that successful spawning under exis ti ng str eamflow , therma l,
and sediment r egimes is dependent upon the presence of upwelling and
conditioned by subst rate, aepth, and ve 1 oci ty attributes. However,
15
spawning can only occur in those portions of side sloughs or side channels
possessing adequate passage dep ths.
High resolution aerial photographs of the middle Susitna River were
obtained on March 2, 1983 \'lhen the river was covered with ice. All side
slough areas in which open leads are visible have been categorized as
possessing an active groundwater source. These sites will be re-examined
in aeria l photography obtained when the ma in stem d i scharge was 23 ,000,
16,000, 12,500 and 9,000 cfs in order to i dent i fy their overtopping
discharge and flow characteristics such that they can be stratified using
the same methods and classification scheme used to stratify rearing sites
(see Table 5).
The categorization and stratification of both modeled and nonmodeled side
slough sites will be further refined on the basis of site-specific
hydraulic, morphologic and hydrologic data available in project r eports
issued by AOF&G, EWT&A , and "R&M Consultants, Inc. In addition to access
and upwelling, site-specific attributes of particu lar interest include the
frequency o f overtopping, hydraulic slope, top width o r surface area,
subs trate composition, and the velocity and depth distribution at
r ep r esentative transects under various flow conditions. The analys is of
data pertaining to these attributes will be used to interpret and qualify
WUA forecasts available for the mode led side sloughs within similar
categories.
The evaluation of chum and sockeye spawning habitat availability in side
channel and peripheral mainstem areas will also be founded on the
assumption that only those locations where upwelling exists are capable of
16
suppo rting spawning activiti es , and then only if access , substrate composi-
tion, and velocity and depth conditions are suitable. A visua l analysis of
the March 2, 1983 aerial photography r evealed 45 mainst e m or side chann e l
sites with open leads that are l ikely to result from upwelling.
Comparisons between these sites and chum and sockeye spawning locations
reported by the AOF&G (AOF&G 1981, 19 82, 1984a) indicate that open lead
areas exist at 10 of 13 r epo rted mainstem spawning sites.
The 1984 Task ~2 field studies relating to middle river chum and sockeye
spawning habitat will focus on known spawning sites and suspected upwelling
areas whe r e spawning has not been reported. A total of 48 candidate sites
exist; 13 known spawni ng sites, including three l ocations for which open
leads are not apparent in the March photography, and 35 potential spawning
sites where no spawni ng has been reported but upwelling is suspected. The
known spawning sites have been tentatively stratified using the same
c la ss·ification scheme described above for re a rin g sites (Tab l e 2). At
present, the 48 candidate sites are believed representati ve of known or
potential chum and sockeye spawning sites within mainstem and side channel
areas that might be directly affected by st r eamflow alterations.
All 48 locati ons will be visited at least once quring FY85 to collect
spawn e r utilization and channel structure data an d to confirm the presence
of upwelling. Sixteen habitat sites have been selected for detailed study;
eight of these will be locations where chum o r sockeye spawning has
occurred at least once during the 1981-83 period. Habitat modeling data
will also be collected at eight locations where upwelling is present but
spawning has not been reported. A quantitative comparison will be made of
17
the hydraulic and mo rphologic attr ibutes of both modeled and nonmodeled
sites in an attempt to furt~er refine the stratification of known or
potential spawning sites, and to identify fac tors which may be res ponsible
for the long-term absence or year-to-year variation of spawning at certain
sites under existing conditions. This information will be used in combina-
tion with W~ and s urface area estimates from modeled sites to assess chum
a nd sockeye salmon spawn ing habitat availability in mainstem and side
channel a reas of midd l e Susitna River at discharges of 9,000, 12,500,
16,000 , and 23,000 cfs .
Sununary
In order to validate the classification and stratification of study sites
within the midd le Susitna River, r econnaissance grade field surveys wil l be
conducted during 1984 at a large number of s ites within each category,
including all candidate spawning and rear-ing study sites. Habitat inven-
tory procedures have been developed as a systematic, cost-effective means
of obtaining a semi-quantitative description of the physica l attributes
present at each site. Figure 11 indicates the principa l habitat inventcry
form to be completed at each surveyed site. Supp l emental fo~ms allow for
detailed remark s , photographs, and sketches of site-specific obse rv ations.
Our intent is to use this information to desc ri be habitat attributes which
appear to be important to the distribution and abundance of salmonid
populations, such that nonmodeled $ites can be linked to modeled sites.
Whereas the primary focus of the extrapolation methodology is its utility
in describing existing habitat conditions within the middle river, the
method appears to be well-suited to forecasting with-project effects. This
18
Figure 11.
Habitat Inventory
Crew: ______________________________ __ Date:
Time:
A.M.:
Location: ----------------Category~------
Maln:uom Discharge:
Me n Reach Velocity:
Site Specific Discharge:
Does Upwelling Occur?
Broached? Yeo/No
Estim ted/Measured
Estlm ted/Meas ured
Yes /No/Cannot Be Detected VIsually
Do Tributaries Enter tho Slough or Side Channel? Yes/No
If Ye~. De.scTiptlon of Tributary(Gize,loe tlon,habltat): -------
Head Gogo; ______ -WSEL: ----
Mid-Roach Gage; WSEL:
Mouth Gage: WSEL: ----
Substrate: 1 2 3 4 56 7 8 0 10 11 12 13
Substrate Embeddodneu:
Dontlnont Cover Codll:
Pereant Ccvor:
1 2 3
12345678 9
Remarks:
Stroa.mbank Slope:
123456
1 2 3 Stable/Unstable
Stroambank Vegetation: 1 2 3 4
Representative Top Widttc -----
Representative Depth:
Bankfull Top Width: ------
Bankfull Depth;
Socchl Disk Measurement: 1st: ----2nd: ___ Average: ___ _
Length of Ba.ckw ter(non-breachod): Estimated/Measured
Were Fish Observed or S lned? Y s/No
Adult: Chinook __ Coho --Sockeye __ Chum ___ Pink----
Juvonl:o: Chinook __ Coho __ Sockeye __ Chum __ Pinlt --
Remartcs:
EWT&A
Primary data recording form to be used in 198 field surveys of
mainstem and side ch~nnel sites in the middle Susitna River .
is particularly true if the present status of fish habitat within the river
has been adequately documented, and the relationship between discharge and
habitat availability is known . Because the stratification and extrapola-
tion concepts outlined in this paper represent a logical and effective
means of assessing existing and potential habitat availability, we
recommend their adoption as a framework for future studies within the
midd le Susitna River.
19
0 -• • •
Literature Cited .
Alaska Department of Fish and Game (AD F&G). 1981. Adult anadromous
f isheries project, phase I final draft report. Alaska Department of
Fish and Game. Su Hydro Aquatic Studies Program. Anchorage, Alaska.
• 1982. Adult anadromous fish studies, volume 2, phase II final data
-report. Alaska Department of Fish and Game, Su Hydro Aquatic Studies
Program. Anchorage, Alaska .
• 1983. Susitna hydro aquatic studies, phase II basic data report.
--""v·ol. 4: Aquatic habitat and instream flow stu1ies, 1982. Anchorage,
Alaska.
• 1984a. Susitna hydro aquatic studies. Report No. 1. Adult
---'anadromus fish investigations .(Hay-october 1983). Anchorage, Alaska.
---=· 1984c~ (In press). Sus"itna hydro aquatic studies. Report No. 3.
Aquatic haMtat and instream flow investigations (Hay-October 1983).
Anchorage, Alaska.
Baldrige, J .E., and D. Amos. 1981 •. A technique for determining fish
habitat suitabi 11 ty criteria: a comparison between habitat
utilization and availability. Pp. 251-258 in N.B. Armantrout (ed.),
Aquisition and Utilization of Aquatic Habitat Inventory Information.
American Fisheries Society. 376 pp.
E. Woody Trihey and Asso.ciates. 1984. Response of aquatic habitat surface
areas to mainstem discharges in the Talkeetna to Devil Canyon reach of
the Susitna River, Alaska. Susitna Hydroelectric Proj. Doc. No. 1693.
15 pp.
Hynes, H.B.N. 1972. The Ecology of Running Waters. Uni v. of Toronto
Press. 555 pp.