HomeMy WebLinkAboutSUS701PRELIMINARY DRAFT
IMPACT ASSESSMENT TECHNICAL MEMORANDUM
SUSPENDED SEDIMENT AND TURBIDITY
Prepared By:
ERWIN E. VAN NIEUWENHUYSE
ARCTIC ENVIRONMENTAL INFORMATION AND DATA CENTER
UNIVERSITY OF ALASKA -FAIRBANKS
707 "A" STREET
ANCHORAGE, ALASKA 99501
Submitted To:
HARZA-EBASCO SUSITNA JOINT VENTURE
711 "H" STREET
ANCHORAGE, ALASKA 99501
1985
TABLE OF CONTENTS
INTRODUCTION .............................................................
BASELINE CONDITIONS
METHODOLOGIES ............................................................
Predicting Impacts •••••••••••
Assessing Impacts •••••••••••
REFERENCES CITED .........................................................
33RA/012a i
1
1
6
6
10
12
METHODS FOR PREDICTING AND ASSESSING DOWNSTREAM IMPACTS
'RESULTING FROM IMPOUNDMENT OF THE UPPER SUSITNA RIVER
SUSPENDED SEDIMENT AND TURBIDITY
INTRODUCTION
This technical memorandum presents a discussion of the ways in which
environmental impacts associated with an altered suspended sediment regime in
the middle Susitna River can be predicted and assessed. It begins with a
description of baseline suspended sediment transport and our present
understanding of how it influences habitat quality and biological productivity
in the middle Susitna River. This is followed by a discussion of how this and
other information presently being gathered can best be used to make rational
predictions regarding post-impoundment conditions and some methods available
to test these predictions should the proposed project be licensed and built.
BASELINE CONDITIONS
l.ft 'Q
Glaciers cover approximately ~% of the Susitna River drainage area
above the USGS gauging station at Gold Creek (RM 137). About 13% of the
annual runoff measured at this station (over 150 miles downstream from the
nearest glacier) is contributed by glacial wasting
1/et..-.,:s~"'
(Willi&mSQR and R&M
Consultants, Inc. 1982). The estimated mean annual suspended sediment
discharge at Gold Creek was calculated by Harza-Ebasco as 7. 26 x 10 6 t/yr
(Harza-Ebasco 1984). Using data collected during 1952-1957, Borland (1961)
~to"
estimated the mean annual sediment load at Gold Creek as 3. 29,.._ t/yr with an
6 average annual runoff of 7.00 x 10 ac-ft. Knott and Lipscomb (1983), on the
basis of weekly sampling from May-September 1982, estimated the total
't f(} e:.
suspended sediment load for that period in 1982 as 2.81 tons with a water
A
33RA/012 - 1 -
6dischargeof6.2 x 10 ac-ft.The total suspended load during the 1983 water
'X IOh
6year(October 1982 -September 1983)was 3.46~t~at 6.45 x 10 ac-ft total
discharge (Knott and Lipscomb 1985).Knott's figures compare well with
Borland's estimates even though Knott's sampling transect was located 35 river
miles downstream from the Gold Creek station.
Almost all (99.6%)of this suspended sediment transport takes place
between May and September.During spring breakup and flood events throughout
the open water season,the relative amount of sand transported exceeds the
silt-clay particle size fraction transported.Typically,by late June the
onset of glacial wasting in the headwaters,combined with higher summer flows,
increases the proportion of silt-clay in the suspended sediment load.From
late June to mid-September 1982,the proportion of silt-clay fluctuated around
78 ±8.5%of the total suspended load.The TSS concentration during this
.-
1
period averaged 372 ±160 mg/l (Knott and Lipscomb 1983),while turbidities
ranged from 74 to 730 NTU (R&M Consultants,Inc.1982).By late September or
early October,TSS concentrations and turbidities typically drop below 20 and
during the winter months (December-March),TSS concentrations are usually
<3 mg/l and turbidities <lNTU (Knott and Lipscomb 1985,R&M Consultants,Inc.
1981).
The physical processes of flow and TSS transport during the open water
season and of temperature and ice during the winter exert powerful limitations
on the ability of the middle river to support fish production.Temperature
and ice processes probably limit production in most streams throughout Alaska
(clearwater as well as glacially influenced)and are not exceptional or
particularly destructive in the middle river.From the standpoint of
density-independent regulation,the two most distinctive limiting factors in
-33RA/012 - 2 -
--
the Susitna River are:1)the highly erratic flow regime;and 2)the
relatively high suspended sediment load during the open water season.
In our opinion of the two,sediment load (or more accurately,turbidity)
is the more important in terms of limiting fish production principally because
it limits fish food production in all areas affected by mainstem flow.In
other words,if the flow regime were regulated to optimize fish habitat
conditions,but turbidity levels remained unchanged,fish production would be
enhanced far less (if at all)than if turbidity were substantially reduced or
eliminated under the existing flow regime.
Most of the biological studies conducted to date have been directed
toward the goal of describing the relationship between fish habitat and
mainstem flow in the middle river.This information can be used to define the
with-proj ect flow regime best suited to maintain or enhance existing fish
habitat conditions in the middle river.This emphasis on flow and fish
habitat stems not only from the important role flow plays in fish production,
but also from the fact that the amount of flow released by the proposed
project is negotiable.The quantity and particle size distribution of
suspended sediment transported by the regulated flow,as well as the turbidity
it will exert,are not negotiable;they are simply the inevitable consequences
of damming a glacial river.Thus,from the standpoint of predicting the
potential harmful or beneficial effects associated with reducing and
redistributing the annual transport of suspended sediment throughout the
middle river,the question of how turbidity exerts its influence on the basic
biological productivity of a lotic system requires a quantitative answer.
This quantitative response must be analyzed in the context of both prevailing
natural conditions and the anticipated with-project TSS regime.
33RA/012 - 3 -
.-.
.~!
Based on 1984 field observations and preliminary data collected before
breakup (1985),it is clear that the natural TSS regime just described exerts
a strong influence on both the seasonal timing and the extent of primary and
aquatic insect (especially clinonomid and plecopteran)adults observed from-
secondary production in the middle river.The large numbers of emerging
mid March to mid-May 1985 throughout the middle river,indicate that
substantial secondary production takes place during the winter months
(Van Nieuwenhuyse 1985).This insect growth appears to have been based
-,,"
largely on the algae production which took place during the previous fall
transition period (i.e.,when turbidities fell to below 20 NTU).March-May is
also the period when recently emerged chum fry feed in preparation for
outmigration and that chinook juveniles display a significant growth spurt
(ADF&G/Su-Hydro 1982,1985a).
Most of the chum and 1+chinook have migrated out of the middle river by
mid summer.At this time,substantial numbers of 0+chinook juveniles begin
to migrate out of their natal tributary streams (most notably Indian River and
Portage Creek)and redistribute themselves principally in turbid side
channels.As autumn approaches,juvenile chinook densities in the tributaries
decline and densities in sloughs increase (ADF&G/Su-Hydro 1982).During the
summer,aquatic insect densities in turbid Susitna River habitats are two
orders of magnitude lower than in clearwater habitats during spring
-(ADF&G/Su-Hydro 1985b;Van Nieuwenhuyse 1985).During the fall,when 0+
!~
(
chinook are preparing to overwinter in slough and tributary mouth habitats,
the mainstem and peripheral channels along the entire length of the middle
river display a large pulse of primary production which is estimated to exceed
the spring pulse by at least two orders of magnitude on a daily systemwide
basis (Van Nieuwenhuyse 1985).The task of processing this large quantity of
33RA/012 - 4 -
--
-
organic carbon (an educated guess would conservatively place the figure
somewhere between 100,000-500,000 tons)falls primarily on the cold-adapted
chironomid larvae.Thus,the fall appears to be the major period of insect
food production,while much of the winter is characterized by food processing.
This basic pattern is not significantly different form the patterns
displayed by many clearwater streams elsewhere in Alaska or in temperate
latitude streams in general (Anderson 1984,Ball and Barn 1975,Lowe and Gale
1980,and Tett et ale 1978).What is distinctive about the middle river is
that the bulk of the primary production it supports is restricted to a very
brief period in autumn when turbidity levels fall below 20 NTU.The 20 NTU
figure characterized the fall transition period of 1984 and does not
necessarily represent the maximum turbidity level at which substantial primary
production may take place in the middle river.If the period between glacial
melt "shutdown"and river freezeup had been characterized instead by a higher
turbidity level (e.g.,50 or even 100 NTU),it is conceivable that a
substantial increase in primary production over summer levels would still have
..-
""
occurred.This is because the rate of primary production is largely a
function of the amount of useful light energy available to the benthic algae
community.
With reasonably accurate forecasts of with-project turbidity levels and a
quantitative understanding of the relationship between algae production and
light energy at depth,'it will be possible to objectively predict the impacts
associated with impoundment.With-project turbidity levels will be deduced on
the basis of output from the DYRESM modeling effort currently under way.This
task will require a model which will predict turbidity on the basis of
forecasted suspended sediment concentrations generated by the DYRESM model.
33RA/012 - 5 -
----~---------------------_._-~-------
METHODOLOGIES
Predicting Impacts.The previous discussion of baseline conditions
provides a reasonably accurate description of natural patterns of suspended
sediment transport and juvenile fish distribution in the middle river.Less
accurate is our quantification of the relationships between TSS and turbidity
and between turbidity and light extinction.We have essentially no hard data
on primary and secondary production during spring and fall.
While we know that chironomid dipterans constitute the single most
numerous food item in the diet of juvenile chum,chinook,and coho salmon
(ADF&G/Su-Hydro 1982),we are only now collecting data on the density,
distribution,and species composition of these and other aquatic insect
communities during the spring when juvenile salmonid growth rates appear to be
at their highest.Previous efforts have been limited to only part of the open
water season (ADF&G/Su-Hydro 1982,1985b).Likewise,baseline data collection
on benthic algae biomass,species composition,and productivity began for the
first time in March 1985 (Van Nieuwenhuyse 1985).This monitoring effort
(Task 31)will continue until freezeup and should yield the information
required to fill the most critical gaps in our knowledge of the basic
biological processes currently supporting fish production in the middle river.
Considerable effort has been devoted to modelling the "weighted usable
area (WUA)responses"of selected channels throughout the length of the middle
river to mainstem discharge according to the Instream Flow Incremental
Methodology.Rather than address fish production per se,this methodology
provides a relative index of how much preferred habitat a given flow creates
in a given channel for a given fish species and lifestage.The WUA response
for a given channel is derived from empirical relationships between hydraulic
and structural features of the channel and catch statistics.The means by
which the information gleaned from these site-specific models will be applied
:~33RA/012 - 6 -
to the entire middle river will be described in forthcoming reports by
E.Woody Triheiand Associates.
The resulting integrated model will not,however,take into account the
effects of mainstem flow and TSS concentration on lower trophic levels (i.e.,
the food base for fish production).The positive effect of increased algae
productivity on fish production is well documented (LeCren 1972).To make
......
,~
-
objective predictions about how impoundment will effect middle river salmonid
production will thus require the following additional information:
1.DYRESM model forecasts of weekly (or monthly)reservoir outflow TSS
concentrations and particle size distributions;
2.a model relating suspended sediment concentration to turbidity;
3.baseline patterns of algae biomass,macroinvertebrate densities,and
community composition during at least one annual cycle;
4.quantification of the relationship between light at depth and gross
primary productivity in glacial systems;and
5.refined quantification of the relationship between turbidity and light
extinction.
A field monitoring program (Task 31)designed by AEIDC to meet two of
these information needs (3 and 4 above)has been underway since March 1985.
Some preliminary data collected to date are reported in Van Nieuwenhuyse
33RA/012 - 7 -
,....
(1985).Until now,the search for a reliable method to predict TSS concentra-
tion on the basis of turbidity (or vice versa)(2 above)has been hampered by
a failure to consider the particle size distribution of the suspended sediment
load.Since clay and silt-sized particles «6211),impart a proportionally
greater cloudiness to water than sand-sized or larger particles (>6211),a
simple linear regression model relating turbidity and TSS concentration will
not explain an acceptably high percentage of the variance between these two
parameters (especially in a glacial system).By incorporating a second inde-
pendent variable (e.g.,the percentage of TSS <6211 in diameter)and by using a
multiple regression model,a much greater degree of reliability is attained.
This approach was used to relate the TSS and particle size distribution
data presented in Knott and Lipscomb (1983)with simultaneous turbidity
measurements made by R&M Consultants (R&M Consultants 1982)for the Susitna
River at Talkeetna,the Talkeetna River,the Chulitna River,and the Susitna
River at Sunshine.The simple linear regression model yielded the following
equation:
where y
2logy =-1.2147 +1.351 log xl (n=58;r =0.65)
turbidity (NTD)and xl =TSS (mg/l).
Equation 1.
....ti
Using the same data,the multiple regression method yielded the equation:
,-
2logY =-1.226 +0.9884 log xl +0.0161 x 2 (n=58;r =0.93)
where X z =proportion (%)of TSS concentration <6211 in diameter.
Equation 2.
33RA/0l2 - 8 -
-
..-
..-
i~
-
-
Incorporatihg-Ene----second--parameter results in a 28%increase in the amount of
variation explained.Perhaps by adding other parameters to represent smaller
particle sizes,the amount of variation explained could be increased further,
but the fit for Equation 2 is adequate to demonstrate the utility of the
multiple regression approach.
A recent technical memorandum (Reub and Wilkinson 1985)provides
preliminary site-specific models relating mainstem discharge and turbidity to
euphotic area models in six middle river channels representing a broad
spectrum of morphological and hydraulic characteristics.These models operate
similarly to the IFG models and are also capable of describing the photic
energy available for photosynthesis at any given depth in a channel given the
mainstem discharge,turbidity,and incident light intensity.As more data on
productivity at different underwater light levels become available from the
Task 31 monitoring effort,these models will be refined to yield estimates of
expected primary production rates under a given flow and TSS/turbidity regime •
Thus,the regression equation derived for the relationship between the
quantity of useful light energy received by the benthic algae community of a
stream and the community's metabolism will be analogous to the "suitability
criteria curves"used in IFG analyses.The principal difference will be that
light at depth alone will very likely explain 60-90%of the variation in
productivity,whereas suitability criteria curves developed for fish spawning
or rearing are often fit by eye and display considerable variation about the
expected mean value.
Until reasonably well circumscribed forecasts of with-project seasonal
TSS outflow concentrations are available,current educated guesses regarding
with-project downstream trophic status must be couched in terms of
33RA/012 - 9 -
.-
hypothetical ranges and must rely on the very limited amount of information
presently available (e.g.,Harza-Ebasco 1985,EWT&A and Woodward Clyde 1985)•
Predictions regarding the trophic status of Watana and Devil Canyon
Reservoirs were made by Peterson (1982)based on the Vollenweider phosphorus
loading model.For a discussion of the methods used and their limitations,
the reader is referred to this report.
Assessing Impacts.If the proposed project is licensed and built,
impacts on downstream trophic status can be assessed by one or more of the
following methods:
1.Monitoring community metabolism of the middle river through continuous
measurements of dissolved oxygen (DO)concentration,and water
-"'"
2.
temperature at a mainstem site in the lower portion of the middle river
(i.e,just upstream of the confluence area at Talkeetna).
Monitoring community metabolism by tracking pH,alkalinity (or perhaps
conductivity),and water temperature at the Gold Creek bridge.
3.Periodic sampling of the benthos in selected habitats to estimate algae
and macroinvertebrate biomass and species composition.
Monitoring community metabolism rates could be accomplished with a
minimum of labor using equipment already owned by the Alaska Power Authority.
Given the lower amplitude of discharge fluctuations expected after
impoundment,a suitable site for placing two dissolved oxygen/temperature
probes could be found well below the reach of the middle river which will
33RA/012 -10 -
--_._------------------------------------
still bear oxygen-supersaturated water under with-project conditions.These
data,combined with data from nearby weather stations and a program to monitor
incident light,water quality (e.g.,T85,particle size distribution,
turbidity,conductivity,pH,alkalinity,macronutrient concentrations,trace
,...
-';lNl
metal concentrations,etc.),salmonid smolt outmigration,and adult
escapements,would provide a basis for comparing with-project conditions to
baseline conditions.This program would be required until at least three
.....~}
.....
(
brood years of chinook outmigrants had returned to spawn.
Theoretically,community metabolism rates could be monitored even in
oxygen-supersaturated water using simultaneous measurements of pH,alkalinity,
and water temperature,but this method has yet to be applied in Alaska and has
been used far less frequently than the dissolved oxygen method elsewhere.
Periodic benthic sampling in habitats selected on the basis of use by
rearing fish or some other set of criteria,would be far more labor-intensive
and less reliable than the community metabolism approach if alone it had to
serve as the basis for impact assessment.It would,however,provide a way to
detect any substantial shifts in community structure or in the seasonal
pattern of algae and macroinvertebrate production caused by impoundment and
would complement the community metabolism approach very well.
The application of anyone of these techniques,combined with long-term
monitoring of smolt outmigration or some other index of fish production,would
do much to further our understanding of lotic ecosystems •
33RA/012 -11 -
REFERENCES CITED
Alaska Dept.of Fish and Game/Su-Hydro.1982.Susitna Hydro Aquatic Studies,
Phase II Basic Data Report.Vol.3.Resident and juvenile anadromous
fish studies on the Susitna River below Devil Canyon,1982.Preliminary
draft report.Anchorage,AK.Alaska Power Authority.Report for Acres
American,Inc.2 vol.
Alaska Dept.of Fish and Game/Su-Hydro.1985a.
anadromous fish investigations (May-October
Anchorage,AK.Alaska Power Authority.1 vol.
Resident and
1984).Draft
juvenile
report.
Alaska Dept.of Fish and Game/Su-Hydro.1985b.Availability of invertebrate
food sources for rearing juvenile chinook salmon in turbid Susitna River
habitats.Report for Alaska Power Authority.Anchorage,AK.1 vol.
Anderson,P.R.1984.
subarctic streams.
AK.113 pp.
Seasonal changes of attached algae in two Alaskan
Master's thesis.University of Alaska,Fairbanks,
Ball,R.C.and T.G.
Michigan.In:
California Press.
Bahr.1975.Intensive survey:Red Cedar River,
River Ecology,B.A.Whilton (ed.).University of
Berkeley.711 pp.
-,"
.-.
Borland,W.M.1961.Sediment transport of glacier-fed streams in Alaska.
Journal of Geophysical Research 66(10):3347-3350.
E.Woody Trihey and Assoc.and Woodward-Clyde Consultants.1985.Instream
Flow Relationships Report.Vol I.Report for Harza-Ebasco Susitna Joint
Venture,Anchorage.1 vol.
Harza-Ebasco Susitna Joint Venture.1984.Reservoir and river sedimentation.
Prepared for Alaska Power Authority,Anchorage,AK.1 vol.
Harza Ebasco Susitna Joint Venture.1985.Technical Report Series.Water
Quality.Report for Alaska Power Authority.Anchorage,AK.1 vol.
Knott,J.H.and S.W.Lipscomb.1983.Sediment discharge data for selected
sites in the Susitna River basin,Alaska,1981-82.U.S.Geological
Survey Open-File Report 83-870.45 pp.
Knott,J.M.and S.W.Lipscomb.1985.Sediment discharge data for selected
sites in the Susitna River basin,Alaska,October 1982 to February 1984.
U.S.Geological Survey Open-File Report 85-157.68 pp •
LeCren,E.D.1972.
Productivity of
Academic Press.
Fish production in freshwaters.In:Conservation and
Natural Waters,R.W.Edwards an~D.J.Garrod (eds).
London.p.115-131.
Lowe,R.L.and W.F.Gale.1980.Monitoring periphyton with artificial
benthic substrates.Hydrobiologia 69:235-244.
33RA/012 -12 -
....
Peterson,L.A.and R&M Consultants,Inc.1982.Water quality effects
resulting from impoundment of the Susitna River.Alaska Power Authority •
Susitna Hydroelectric Project.Report for Acres American,Inc.1 vol.
R&M Consultants,Inc.1982.Water quality annual report 1982.
AK.Alaska Power Authority.Susitna Hydroelectric Project.
Acres American,Inc.1 vol.
Anchorage,
Report for
R&M Consultants,Inc.1981.Review of existing Susitna River basin water
quality data.Anchorage,AK.Alaska Power Authority.Susitna
Hydroelectric Project.Report for Acres American,Inc.1 vol.
Reub,G.and C.Wilkinson,1985.Preliminary Analysis of the influence of
altered middle river discharge and turbidity regimes on the surface area
of the emplistic zone.Technical memorandum to Harza-Ebasco Susitna
Joint Venture,Anchorage.13 pp.
Tett,P.,C.Gallegos,M.G.Kelly,G.M.Hornberger,and B.J.Cosby.1978.
Relationships among substrate,flow,and benthic microalgal pigment
density in the Mechums River,Virginia.Limnol.Oceanogr.24(4):785-797.
of Task 31 primary
Arctic Environmental
Technical memorandum
Susitna basin glacier
Alaska Power Authority,
Van Nieuwenhuyse,E.E.1985.Preliminary results
production monitoring effort,March-June 1985.
Information and Data Center.University of Alaska.
to Harza-Ebasco Susitna Joint Venture.
f/((r"'I'50 1"'\
\Jil litl:ms Qu.I,W.D.and R&M Consultants,Inc.1982.
studies.Prepared for Acres American,Inc.
Susitna Hydroelectric Project.22 pp •
.-
".,
.-
~-,-:
....,.
33RA/012 -13 -