HomeMy WebLinkAboutSUS257" 11
I ·'
I
II
•
; . ..
SUSITNA HYDROELECTRIC PROJECT
SLOUGH GEOHYDROLOGY STUDIES
Prepared by
Harza-Ebasco Susitna Jo i nt Venture
for the
Alaska Power Auth ority
February 1984
$'U 5
3 '{-s-
]
TABLE 0~ COli TENTS
SECTION/TITLE
LIST OF TABLES
LIST OF FIGURES
1.0 INTRODOCTION
2.0 METHODOLOGY
2.1 Data Compilation and Review
2.2 Site Visits
2.3 Agency and Subcontractor Contacts
2.4 Data Analyses
2.4.1 Aquifer Properties
2.4.2 Aerial Photograph Interpretation
2.4.3 Field Data Reduction
~.5 Mathematical Modeling
·2 .5 .1 Datw Correlations
2.5.2 Two-Dimensional Cross-Sections and Profiles
3.0 RESULTS
3.1 Hydrogeologic Setting
3.1.1 Regional Geology
3.1.2 Interpretation of Aerial Photographs
3.1.3 Slough Runoff Estimates
3.1.4 Groundwater Underflow Estimates
3.2 Aquifer Properties
3.2.1 Talkeetna Pumping Test
3.2.2 Talkeetna Specific Capacity Data
3.2.3 Slough 9 Surface Water -Groundwater Correlation
3.-J·-uaca Correlations
3.3.1 Slough Discharge Data
3.~.2 · ·Seepage Meter Data
3.3.3 Temperature Data
3~ Analytical Modela
3.4.1 Groundwater Level Variation•
3.4.2 Temperature Variations
ii
iii
1
1-6
6-23
4.0 ~NCEPTUAL SYSTEM MODEL 23-24
5.0
6.0
EIFECTS OF PROJECT OPERATION
SUM MAllY
REFERENCES
TABLES
FIGURES
24-26
26
27
i
NO.
1
) . ..
L IST OF TABLES
TITLE
Transmissivity Est i mates Based on Specific Capacity Data
for Talkeetna We lls
ii
·"
!!2.:.
1
2
3
4
5
6
7
8
9
10
11
12
n
14
15 I .., ..
).
I
LIST OF FIGURES
TITL!
Groundwater Contours and Flow Lines. Susitna River at
Slough SA
Groundwater Discharge vs. Hainstem Discharge, Slough SA in
1982
Groundwater Discharge and H~inste• Discharge vs. Time,
Slough 8A in 1982
Aquifer Test Data, Talkeetna Fire Hall Well, March 1981
Pumping Test
Groundwater Level Variations in Response to River Stage
Fluctuations
Hainstem and Slough Discharge vs. T1 .~
Slough 11 Discharge vs. Hainstem Discharge at Gold Creek
Slough 8A Discharge vs. Hainstem Discharge at Cold Creek
Seepage Rate vs. Hainstem and Slough Discharge at Sloughs
8A • 9 • 11, and 21
Slough 8A Water Temperatures
Slough 9 Surface Water and Groundwater Temperatures
Slough ll Water Temperatures
Slough 21 Water Temperatures
Simulated Groundwater Level Variations in Response to River
Stage Variations
Simu lated Groundwater Levels vs. Distance from River
iii
1.0 INTRODUCTION
This r~ort providea results of a study begun in September 198J .into • hydrolq&ic conditions affectin& side sloughs of the Susitna River between
Devil Canyon and Talkeetna. downstreaa of the propoaed Susitna Hydroelectric .
Project. Because of the importance of these sloughs as salmon spawnin& and
rearing areas 1 and the possibility that groundwater discharge to the sloughs
is derived from the aainstea 1 the current study involves investigations into
hydraulic and thermal relationships between mainstem flows and slough flows.
The basic objective of this study ia to predict possible variations in the
amount and temperature of groundwater discharge to the sloughs as a result of
variations in mainstea flows and temperatures induced by project operations.
The current study is based on existing data collected during 1982 and 1983 by
R&M Consultants and the ADF&C SuHy dro Aquatic Studies Group. Those data have
been used in a variety of statistical and other mathematical analyse• in an
attempt to identify signfficant interrelationships between mainstem andeugh
hydrologic conditions. No new data have been generated during this study.
other than observations made during field reconnaissance trips and information
gleaned from published reports.
2.0 METHODOLOGY
2.1 Data Compilation and Review
A variety of ~urface water. groundwater. and water quality data have been
compile~~~~-sources such as R&H Consultants. ADF&G, U.S . Geological Survey.
and published and unpublist ed
inoludJ"the follovina:
reports. The types of data which are available
..,.,
d
~·
Aquifer teat data. specific capacity data. and well logs from shallow
wells in the Ta l keetna area.
DRAFT 2/21/84
-1-
o Cfoundwater level data -occasional water
l982 fro• sixteen wells near ~iou~ aAand
~ 9; continuous dat•pod water level r~a
wells near alou&h 9 ••
o Aerial photocrapha.
level measurements durina
sixteen wells near slouah
durin& 1983: fro• three
o Mainstea discharge data -da_ily records from the USGS gaging station
at Cold Creek for 1982 and 1983.
o Mainstem water surface elevation data -occasional 19S2 and 19S3
recorda from 33 stations within and in the vici~ity of Sloughs SA, 9,
11, an~ 21; water surface profiles predicted by hydraulic modeling.
o Slough discharge data -daily records during the summer of 1982 froa
gaging stations in sloughs 9 and 11, and daily records during the
summer of 1983 from gaging stations in sloughs 8A, 9, and 11.
o Seepage meter data -occasional summer 1983 readings from nine
seepage meters in slcughs SA, 9, 11, and 21.
o Summer 1982 and 1983 weather data from th~ Sherman weather station.
o Groundwater temperature data -occasional temperature measurements
4uring 1982 from fifteen wells near slough 8A and from fourteen wells
near-slough 9; continuous datapod records during late 19S2 through
1~83 fro~-three wells near slough 9.
ol Occasional
locationa,
1982 temperature measurements at various mainstem (two
near each of sloughs SA and 9) and slough (sloughs 6A, SA,
~ • 9, 9A, 91, 10, 11, 20, 21, and 22) locations •
P:
DRAFT 2/21/84
-2-
;/
o Interaittent maiastea temperature data for the summer of 1982 throuah
the su..er of 1983 (seventeen locaL~ans between Talkeetna and Devil
!V Canyon); intermittent slouah temperature data for the :vint er and
autumn of 1982 throuah the summer of 1983 (sloughs 8A, 9, 11, 16, 19,
and 21).
o Miscellaneous water quality data from several mainstea and slough
locations.
2.2 !!!! Visits
A site reconnaissance trip was conducted on September 21 and 22, 1983. The
visits were made durin& a period of relatively low mainstem discharae
(approximately 10,000 cfs), so the influence of groundwater discharge on
s1ouah conditions was more apparent.
Durin& the afternoon of September 21, helicopter flyovers of several sloughs
between Talkeetna and slough 11 were made, with stops at slouahs 8A, 9, and 11
for more direct observations. In these sloughs, several observations were
made of seepa&e and upwelling. In addition, instrumentation including staff
gases, staae recorders, and seepage meters was observed on the ground, and
monitoring wells at slough 9 were observed from the air. Lover reaches of
alouJh 11 were toured on foot, and the servicing of instrumentation at well
9-lA was observed. Several alouahs upstream of slough 11, and Devil Canyon,
~ere oh~~~~ from the air in flying to Watana Camp at the end of the day.
On Sept~~ 22~ -servicina of the staae recorder at Deadman Cre~k was
observed. The lower reaches of slouJb 9 were later toured on foot. Seepage
meter 4(asurements were observed at slough 11, and side slough 10 was visited
brief~durin& the return to Talkeetna by boat •
•
DRAFT 2/21/84
-3-
Agency ~ Subcontractor Contacts ·'
Followina the site visit des.~ribed above, a number of knowledaable individuals
and orasnizations were contacted in order to obtain published and unpublished
info~atioo which •ilht be available, and to elicit any co..ents or
sugaestions which miaht affect future studie1. Oraanizations contacted
include the Harza-Ebalco Joint Venture, R&M Consultant•, the Ala1ka Power
Authority, Trihey & Associates, AEIDC, u.s. Geoloaical Survey, Alaska
Geological an~ Geophysical Survey1,-and the U.S. Fish and Wildlife Service.
2.4 B!!! Analyses
2.4.1 Aguifer Propert i es
Results of aquifer tests and specific capacity data in the Talkeetna area have
been obtained from uses files. These data have been subjected to standard
hydroloaic analyses for estimation of aquifer properties for the alluvial
mat~rials at that site. The resulting properties should be similar to those
of the valley-fill materials further upstream, in the vicinity of the side
sloughs.
Datapod hydrographs have been provided for mainstem stage and groundwater
levels in wells at slough 9. Attem9ts have been made to interpret these data
by applyin& published (8)!/ techniques for estimating aquifer properties
based on arouadwater variations in response to stream stage variations •
. -,. -
2.4.2 Aerial Photograph Interpretation
Availa~e aerial pbotoarapbo have been interpret•d to identify probable . ...,
contacts be~en bedrock, &lacial detritus, and alluvial materials. Locations
of rep~ted seeps and upwellings have been compare~ with the inferred
surficiai geoloay to seek any obvious relationships between aeoloaic contacts
and location• of aroundwater discharae to sloughs.
!/ Refers to the numbers in "References" at the end of the text.2.4.3
DRAFT 2/21/14
-4-
2.4.3 Field Data Re'duetion ,, -:----
' Tbe reduction of available ~~eld data has involved the tabulation, plotting,
aod coaputer storaae of selected data. Data collected durin& 1983 has been
emphasized because of the variety of data available and the existence of
relatively larae amounts of continuous or partially-continuous data. Where
possible, mean daily values of parameters such as water level, discharge,
temperature, and precipitation have been plotted versus time, and the
resulting gra~s compared to ascertain possible correlations. Parameters
suspected. of being stronaly correlated have been pl,cted aaainst each other on
linear and logarithmic paper to determine the probable functional form of any
relationships between the variables. During the course of the statistical
analyses discussed below, much of the 1983 data has also been input to
computer files, basically in the forw of time series, in order to facilitate
the statistical analyses and other mathematical analyses. It must be
recoanized that much of the 1983 data is provisional and subject to chanae as
the data are reviewed and further reduced. However, these data should still
be adequate to illustrate major trends and interrelationships.
2.5 Mathematical Modeling
2.5.1 Data Correlations
A variety of statistical correlations of existing time-series data (water
levels, discharge rates, temperatures, other water quality parameters) have
been performed: These activities were conducted to attempt to ascertain
~igni~~nt correlations amona the various parameters for which data are
availa,e.
~ • In general, these activities have included autoregression of time series data
to asclrtain preexisting trends; transformation of .data so that nonlinear
regression analyses can be performed, including lagging the data with respect
to time; and multiple linear regression of transformed and nontransformed
data. Transforma tions of the data were based in part on knowledge of the
aeneral hydroloaical setting of each slouah. The objective of these analyses
DRAFT 2/21/84
-s-
.1
.·
was to aacertain sianifi~ant relationship• a~on& variable• such as slouah
~"' . dischatae and teaperatu~e. mainstea discharge and staae, air tempe~ature,
mainstea water temperature, .precipitation, etc.
2 .5.2 Two-Dimensional Cross-Sections and Profiles
Simplified analytical models of flow ~nd thermal transport in vertical
sections normal to t tae rive~ have been used in analyzin& e~istin& data for the
slou&h hydrolagic r~6ime. Computer-programs were p~epared based on published
analytical solut i ons to relevant flow problema (1, 6).
Simulations of the groundwater surface between the mainstea and the slougha,
and variation of that surface with variations in mainstem water levels, within
a two-dimensional vertical section extendin& from the river to the slouah,
were conducted by applyin& the convolution integral app~oach outlined by Hall
and Moench (6). Althou&h this approach presumes symmetry with respect to the
dimension normal to the vertical section, and is thus only an approximation,
it is believed to provide a reasonable estimate of the relationship between
variations in mainstem stage and groundwater levels. Similar analyses were
carried out for groundwater temperature variations, by applying the
convolution integral approach of Hall and Moench (6) to the coupled thermal
and groundwater flow solution developed by Acres American (1).
;.1 Hydrogeologic Setting
J. 1.1 I Res ional Ceo 1osz
""";!'r
The r~ional geologic setting
Talkeelna has previously been
3.0 RESULTS
of the Susitna River between Devil Canyon and
described in several .works (5, 7, 9), and those
descriptions will not be repeated in detail here. However, basic
characteristics of regional geology relevant to the present study are briefly
disc u ssed below for the sake of completeness.
DRAFT 2/ll/84
-6-
As desCfibed by R&H Consultant• (9), ,.
" all aloupu along the rivet-are part of the modern floodplain of the
S~aitna River [which] conaista predominately of cobbly sandy gravela with
ailty mantles in areas between and adjacent to the main channels. Above
and immediately adjacent to the modern floodplain lie a aerie• of fluvial
and glaciofluvial terraces deposited ••• following the later Wisconsin
glaciations of Southcentral Alaska. The terrace deposits generally
consi ~t of coarse sandy sravels overlain by a few feet of sandy silt and
silt overbank depoaits ••• The valley floors and side walls above the
terraces are thought to consist of glacial tills composed of gravel, sand
and silt~ •• Older ••• glacial and glaciofluvial drift may underlie t~e
terraces and modern floodplains. Redrock underlies the unconsolidated
materials at an undeter:n ined depth."
Available geologic mapping (10, 13) suggests that the unconsolidated fluvial
and glaciofluvial deposits are confined to a very narrow interval along the
river valley, with consolidated bedrock located on both sides of the river
between Devil Canyon and Talkeetna. Interpretation of aerial photographs
suggests that the width of the valley-fill sediments in the reach between
sloughs 11 (near Gold Creek) and 8A i$ relatively consistent, averaging
approximately 3,000 feet.
3 .1.2 Interpretation of Aerial Photographs
The following discussion of the slough environment has been inferred from
aerial photographs of the Susitna River and sloughs, at a scale of
approximately 1 inch • 1000 feet, and various project reports.
Ccdi ... .::n:s in th·e -River and alough regions consist of materials deposited
within the active channel of the Susitna river (c~annel sediments) and
mater~~~ forming the valley walls (valley wall deposits ). Valley wall
deposJPs may include bedrock, terrace deposits formed during past higher river
·~
level~,· and till deposits , which reportedly cap the entire region.
};·
Sloughs are generally found on the left descending bank, with mainstem flow
generally, but not consistently, a l ong the right descending bank. Slough
areas are generally well vegetated, except within the channel of the slough
itself. Slough areas are generally contiguous with the valley wall area,
occasionally separated by a tributary stream. The photographs were inspected
DIU.FT 2/21/84
-7-
..
for evi~nee of uniformity in paleo-channel ~1dth, as might be inferred froa
terrae~ or valley vall position. There ~a s z~me consistency noted in channel
width io the segment examined between Gold Creek and slough sf At Cold Creek,
the apparent paleo-channel widens substan t ially, perhaps as a result of Gold
Creek flow and sediment contributions. The river appears to have adjusted to
a pa~te rn lying between that of a braided stream and that of a meandering
stream. Rela~ively steep terrace (?} valley walls are observed on the south
and east s t.ores (left descending bank) while the north and west shores (right
descending b ~rrk) appear from the photographs to exhibit generally undulating
topography, gently rising with distance from the river. However, field
ob ~e rvations suggest that the right descending valley wa~l has about the same
s t eepness as the left descending wall, particularly in the vicinity of slough
9. Many abandoned channel scars are evident in the channel fill materials
forming the small islands and lowermost floodplains adjacent to the river.
Vegetation is generally absent within these scars.
Upwellings (groundwater discharge withing the sloughs) are occasionally, but
not consistently, visible on the photographs. There is no discernible
relationship among the locations of the areas of upwellings, and the river
morphology, distribution of river sedi ments, or the floodplain configuration.
At several sloughs there is a distinct boundary at the mouth of the slough ,
separating dark (probably clear, silt free) water discharging from the slough,
from the gray (probably turbid) water of the mainstem. In some cases, a zone
of mixing of these waters can be observed extending downriver within the
mainstem. There may be some suggestion of upwelling within the mainstem, as
evid.:!oc~u;,., spot~-of dark water apparent within t he turbid mainstem flow.
3 .1. 3~ Slough Runoff Estimates
~
One p~tentiel source of at least part of the discharge from individual sloughs
is dirlct precipitation on the drainage area of the slough. While no attempt
has been made to generate synthetic storm hydrographs for each slough, total
precipitation on the drainage area of a particular slough over relatively long
periods of time (several months) has been compared with slough discharge over
the same time periods. This approach was based on the rather simplistic
DRAFT 2/21/84
-8-
assumption that cumulative precipitation ove~ Telativelr lone periods will
.; approxi~te the sua of surface runoff and aroundwater infiltration within a
basin. In this manner an es~iaate can be made of the proportion of slouch
discharse derived froa localized sources, such as direct precipitation on the
slouch drainaae area p us intearated groundwater recharae within the drainaae
area, relative to the amount of slough discharge derived from external sources
such as localized groundwater transport from the mainstem, or more regional
groundwater underflow within the river basin.
The results of the s e analyses suggested that only very small proportions (of
the order of a few per cent) of slough discharge could be attributed to
precipitation, either directly as runoff or indirectly as infiltration and
subsequent groundwater discharge to the sloughs. It is recognized, however,
that these calculations are no substitute for the more detailed generation of
synthetic storm hydrographs which are being developed by others. Nonetheless,
based on these preliminary estimates, subsequent analyses were based on the
working hypothesis that most of the discharge from sloughs 8A, 9, and 11 was
derived from sources such as direct discharge from the mainstem as a result of
overtopping of berms, regional groundwater underflow within the Susitna River
a.lluvium, or more l ocalized (and probably relatively shallow) lateral flow
from the river toward t he sloughs.
3.1.4 Groundwater Underflow Estimates
Based on estimates of aquifer properties (as discussed in more detail bel~w)
cud the average· d~stream groundwater level gradient within the Susitna River
Valley , an estimate has been made of the volumetric rate of groundwater
transpt in the downstream direction within the Susitna River alluvium. For
an ass d hydraulic conductivity of 500 gallons per day (gpd) per square ·-,.r
foo•, l saturated thickness of 100 feet, an aquifer width of 3000 feet
(incluJing the active channel and the alluvial floodplain), and an average
downstream groundwater level gradient of 0.003, the average rate of downstream
transport of groundwater would be about 0.7 cubic feet per second (cfs). Even
if this estimate is low b y an order of magnitude, it would appear that
regional groundwater transport within the Susitna River alluvium would not be
sufficient to provide all of the groundwater discharge apparently observed in
J
DRAFT 2/21/84
-9-I . ', I .J
the var~oua aloucha. Thra tends to support. ~ bypothesi• that a larce
proportion cf tbe tlouch dischacce may be derived from shallow· lateral flow ·
••1 from the river, rather than ~ecional groundwater
~
/ \;liver valley-fill aaterial•··~
<: 1 ~ ,, • ·-~ T~ \ j lc-~.-r ':.....-'-l .r
' L• "-"' . •
It
., ..
underflow within the Susitna
I v I ' ;_I . --{ ..
/t<-. / r
Another aspect of groundwater underflow was considered by referring to the
aap1 of croundwater contours at slough• 8A and 9 for variout date• in 1982
presented by R&H Consultant• (9, Figure• 3.4 through 3.21). Assumin&
I _' I
I ;··'
homogeneout add isotropic aquifer mat~rialt, groundwater flow linet were drawn
normal to the water level contour lines shown on those mapa. The flow linet
suggested flow from a side channel of the river toward a portion of the right
descendin& bank in the upper reachet of slouch 8A (see, e.,., Fig. 1), and
toward slough 9B and a portion of the left descending bank in the upper
reaches of slough 9. Assuming the same saturated thickness and hydraulic
conductivity as noted above, the groundwater discharce through each inferred
flow tube (see Fig. 1) was calculated. By summing the discharges within the
several flow tubes, an estimate was obtained of the total groundwater
discharge to that reach of the slough fed by the several flow tubes. This was
converted to a unit flow by dividing by the total length of slough bank at the
terminus of all of the flow tubes.
Since no 1982 discharge measurement s were available for slough SA, the
calculated unit flow• (i.e., discharge per length of slough bank) were
compared with mainstem discharge at the Gold Creek gage for selected dates
{Figs. 2, 3).-As can be seen from Fig. 2, there is no obvious correlation
betwccu ·Lt.c: ""'\l-;.:.\:harge per unit bank length and the mainstem discharge.
However, from Fig. 3 it appears that ther~ might be ~ time-series correlation
wi th •l:ag of several days between the two discharges (i.e., in early
Septem er, the unit slough discharge increases as the mainstem d i scharge ,..
increa\ea, while in early October a decrease in mainstem discharge is fol l o wed
severa#-~days later by a decrease in unit slough discharge). However, no
definite·: conclusions can be drawn from this very limited set of data.
DRAFT 2/21/84
-10-
uain& a ~•i•ilar approach, eatimatea of .the total aroundvater diacharae to
alough• 9 and 9A ~re co•pared with measured discharae fro. alo~&h 9. For
June 23, 1982, when the .. i n~te• dischar&e at Cold Creek vaa 25,000 cfs and
the s l ou&h 9 ber8 vaa probably overtopped, the estimated slough discharae vas , J , -
1.44 cfs and the .easured diat.harae vaa 180 ch. For October 7, 1982, when t :·( ... ,-
the mainste• discharae at Gold Creek vaa 8,480 cfs, the estimated slouah •J
discharae vaa 1.43 cfa and the measured discharae vas 1.0 cfa. Ho definite
conclusions can be dra wn from these observationa, except t~at the approximate
aroundvater d~charae toward slouah-9 appears to be of the same order of
magnitude as the observed discharae from the slough durin& conditions of
low-flow on the mainstem.
3.2 Aquifer Propertiea
3 .2.1 Talkeetna Pumping Teat
I . ,\
In Karch of 1981, a 100-foot deep well was constructed at the Talkeetna Fire
Hall. A constant-rate pumping test of the well was performed on March 10-11,
1981. The well vas pumped at a constant rate of 310 gallons per minute (gpm)
for a period of twenty-nine houra, and water levels were periodically measured
in the well. Water levels in the pumping well stabilized within about an
hour, and remained essentially constant for the duration of the test.
The pumpina test data were obtained during a search of u.s.G.S. files in
Anchorage. !fie data were plotted on semi-logarithimic and full-logarithmic
~t-e-r, ..... tJ standard analyse• were conducted (11, 12). The Jacob straight-line
analysis of the semi-logarithmic data plot (Fig. 4) yielded a transmissivity
of ap~ximately 13,900 gpd/ft during the early period of the test, before
stabi~ation of water levels in the well. The full-logarithmic data plot
~
could~ be matched by either the Theis or Hantush type curves, so no aquifer
properlies could be inferred in this manner.
Assuming a saturated thickness of approximately 21 feet based on we ll loga,
the calculated transmissivity for this teat would give a hydraulic
conductivity of approximately 630 gpd/ft2 •
DRAFT 2/21/84
-11-
_ .....
; .
. .
The atabilizatio of wateT level• in the pumP.e~ well indicate• ao.e kind of
.;
recharie to the teated aquifer, aa a reault of delayed yield froa storaae,
leakaae froa adjacent water-bearin& units, or induced infiltration fro• the
river. Well loa• indicate that the unit tested ia probably confined
(arteaian), so delayed yield froa atoraae by aravity drainaae is unlikely.
The inability to match the field data with the Rantush leaky-artesian type
curvet suggests that leaka&e ia alao relatively unlikely. Thus, the most
probable cause of the water-level stabilization ia indu c ed infiltration froa
the river, suggestio& hydraulic connection between the aquifer and the river.
However, the actual cause of this phenomenon can be neither confirmed nor
quanti f ied because of the lack of ob ervation well data during the teat.
3.2.2 Talkeetna Specific Capacity D•ta
Aquifer transmissivity can also be estimated from specific capacity data (the
ratio of total water level drawdown to .pumping rat e) collected during well
drilling and testing. Such data are available for six wells in the Talkeetna
area, and have been obtained fro~ U.S.G .S. files. Utilizing graphs presented
by Walton (11, 12), the estimated transmissivity determined from these data
ranges fro• 2,400 to 14,000 gpd/ft assuming water table conditions, and from
4,400 to 27,000 gpd/f assuming artesian conditions. The results are
summarized on Table 1.
Of the six wells for which specific capacity data are available, well depths
were reported-for only thLee. All three wells were only 17 feet deep, and
tl.ua 'ir"Vwl-J -toe ex.pected to exhibit water-table conditions in this environment.
By dividing the estimated transmissivity by the original saturated thickness
in ea~of these three wells, hydraulic conductivity values rangin& from 240
to 1~~ gpd/ft2 are obtained, with a mean of 710 gpd/ft2 • This compares
quite'fav~ably with the value of 630 gpd/ft2 inferred from the pumping test
data a\· the Talkeetna Fire Hall.
DIW'T 2/21/84
-12-
:.
3.2.3 • Sloush ! Surface ~ -Groundwater ·eorrelat'ion
Atte111pts have been aade to e.~ti~~~ate aquifer properties froa correlationa of
river s t aae and aroundwater level variations at slouah 9. The data were
analyzed accordina to •ethoda deacribed by Pinder et al. (8). However, the
field data could not be matched to the theoretical type curves generated by
the methods of Pinder et al. (8), regardleas of the values assumed for aquifer
~ properties. In general, the field data curves had substantially d i fferent
~~r -slopes than tne theoretical curves for all values of aquifer diffusivity (Fig.
0.~ 5). In particular, data from borehole 9-5 showed a 111ore rapid rise earl)' in
time, b u t a substant i ally lower peak value, than predict~d by the theory (Fig.
5).
I ~ appears that the hydrol ogic conditions affectins the wells near slough 9
are considerably diffe rent than those assumed in the theory. For example, the
theory is based on the assumpt ion that all recharge to the aquifer durin&
passage of a flood peak o n the river is derived fro111 l a tera l inflow from the
river to the aquifer. At slough 9, it is possible that groundwater levels are
also affected by regional water level variations and possibly by groundwater
underflow originating far upriver from the slough or from the bedrock areas
southeast of the slough. I t is also possible that the groundwater level data
we r e affected by recharge both from the ma i ns t em and from the slough, since
the slough 9 berm was overtopped d u ring much of the summer of 1983. The
beaver dam located near the mouth of slough 9B could also affect local
groundwa ter c~nditiona, particularly nea r borehole 9-5, by raisin& local
groundwater leveii and perhaps moderating the influence of v ariations in r iver
stage.
3.3 nlr. C~rrel a t i ono
~ •
A varil·ty of correlation• between slough and mainstem data have been
attemptea. These have included merely comparing graphs of time-s eri es data,
plott i ng var i ables versus each other on linear, semi-logarithmic and full
logarithmic paper, and utiliz i na a standard statistical analysis compute r
program to perform multiple linear regre s sion and cross-correlation analyses
of transformed and raw data. I l l gene ral, the analyses conducted to date have
employed •e an daily v a lues of relevan t parameter••
DRAFT 2/21/84
-13-
The .ore fo~l linear· rearession ~nd ~rosa~orrelation analyses which have
i
been conducted have used the KINITAI computer proara• developed .at
pennsylvania State Universit7. HINITAI is a aeneral purpose' statistical
coaputin& syste•, includin& recently-imple .. nted routines for ti.e series
analysis based on techniques described by Box and Jenkins (4). The fairly
vide usaae of KlNITAI, and its beses in standard statistical techniques,
confer a considerable dearee of reliability on results of its application.
3.3.1 Slough Discharge B!!!
A variety of correlations have been drawn between slough discharge data for
sloughs SA, 9, and 11 and several other parameters such as mainstea discharge,
mainstem stage, water temperature, and precipitation. No general
relationships have been observed. In many important respects, the three
sloughs for which most data are available behav~ differently.
The general relationship between slough and mainstem discharge is illustrated
by Figure 6, which shows discharge versus time for the mainstem at Gold Creek
(provisional 1983 USGS data) and for sloughs SA, 9, and 11 (provisional 1983
R&H Consultants data). There generally appears to be a correspondence at
least between major peaks in the slough and mainstea discharge measurements.
For example, the higher mainstem flows observed in early June, early August,
and late August are fairly well reflected in the data from slough• 8A and 9.
The slough 9 discharge appears to correlate very well with even less
significant variations in aainatem discharge. This would be expected,
ltu~~v~, b~cau~;the slough 9 berm was overtopped approximately half the time
peri:t.eflected in Figure 6, so slouah 9 actually acts as a side channel to
the stem durin& much of this period. Slough 11 exhibit• very little
varia on in discharge at the scale plotted on Figure 6. Nonetheless, the
~
1lou~ 9 discharge also appears to reflect the relatively high mainstem flows
observ~ in early June, and the steadily declining .mainatea flow observed in
mid-September.
DIAFT 2/21/84
-14-
...
In gene~al , utilizing MINITAI routine•~ ~the . ~ischa~ge at alough 11 correlate•
fairly well with aainatea discharge or stage, with correlation :~oefficientl in
exce11 of 90% for linear re-~ess i on• with alough 11 diacharge as the dependent
variable. Multiple linear rearession involving parameters such aa temperature
or precipitation had only alightly hiaher correlation coefficient• than when
aainatea discharge or staae vas the only independent variable. Furthermore, a
plot of alou&h 11 discharae veraus mainstea diacharge exhibits a linear fora
with a poaitive slope (Fi&• 7). In contrast, linear regr e ~sions involvina
alou&h 8l discharge as the dependent variable exhibited correlation
coefficients of the order of 25 -55%. Addition of other parameter• increased
the values of these correlation coefficients, but that m~y repreaent only the
effect of correlating two time series which exhibit similar seasonality in
their variat i ona. Linear regress i ons involv i ng slough 9 discharge as the
dependent variable exhibited corre lation c oefficient• in the range of 65 to
90%. However, these regressions generally i ncluded mainstem discharae as an
independent variable, and thus are probably biase~ since alough 9 waa
reportedly overtopped during much of the s ummer of 1983.
It i s perhaps noteworthy that slough 11 , whose discharge is moat readily
correlated to tha t of the mainstea, is perhaps the simpleat of the three
alougha studied in detai l . The surface drainage area of this slough ia
extremely Dmall, so that slough discharge is less likely to include surface
runoff aa a complicating factor. Furthermore, the ae.rial photograph
interpretation discussed above noted that the :iver valley seems to widen
considerably lt Gold Creek, juat above slough 11, and to maintain a fairly
con•is tent~i~ib .in the vicinity of slough• 8A through 11. Thua, it may be
that groundwater recharge from the mainstem becomes substantially more
signi;ifant below Gold Creek than above Gold Creek because of thia change in
morph~IY•
'1ft
' It shoJtd alao be noted that wh~reaa a plot of slough 8A discharae versu1
mainste~=diacharge shows conaiderable acatter and can not be readily
~epresented by a single fun c tional form, some of the data can be seamented
into different time period• durin& each of which a fairly strong linear
r elationahip between slough discharae a nd mainstem discharge roan be observed
(Fig. 8). The time perioda illuatrated in Figure 8 are distinguiahable by the
DRAFT 2/21/84
-lS-
fact th~ eacb of thea·i• either a period of .aenerally· risin& river staae. or
aenerally fallin& river staae. Furthe~ora . linear fits to the data durin&
different perioda of fallina.river staae (Auaust 14-20 and September 2-17)
aenerally have about the saae slope. while fits to the data durin& different
periods of fallina staae have substantially different slopes (durin& the
period Auaust 20-25, while the river stage was riaina. the sl~uah discharae
was actually decreasina). This information sugaests that, at least at slouah
SA, phenomena such as bank storage may be significant in con trollin& slouah
discharae. Si~ce s imilar relationships have not been o b served in the data
froa sloughs 9 or 11, this phenomenon may be localized to the vicini ty of
slouah 8A.
3.3.2 Seepage ~ ~
The seepage meter data are aenerally consistent with ~he slough discharae
correlations discussed above. Figure 9 shows plots of seepage meter data
versus both mainstem and slough discharge data. The seepage rates at meter'
8-1, 8-2, 9-1, and 9-3 are generally positively correlated with either
mainstem or slough discharge, although the data are rather widely scattered
about the line~r regression fit to the data (Figs. 9a -9Jf. However, seepaae
rates at meter 9-2 seem to be uncorrelated with either mainstem or slough
discharge (Fi&• 9b). At slough 11, t he seepage rates at both meters 11-1 and
11-2 are very we l l correlated with bot h mainstem and slough d i scharge. This
tends to confira the previous observations that discharge at slough 11 is
str ongly correlated with m3 instem dis charge, and there is a good likelihood
that up~elliial at -slouab 11 is derived rather directly from mainste m recharge
to the~cal groundwater aquifer.
Seepa~meter data at slough 21 suggest that this slough i s substantially
diff~r•nt frea those belo w Gold Creek. Seepage rates appear to be negatively
correllted to aainstea discharge at meter 21-1, with seepage rates decreasina
as mainstea diacharae increases. At seepage meter 21-2, there appears to be
no correlation between seepage rates and mainstem discharge. At slough 21,
the river valley is narrower and the valley walls somewhat steeper than
further downstream. Thus, a relatively high proportion of the aroundwater
discharae at this slou&h may originate from infiltration of precipitation on
the surroundin& uplands, rather than aroundwater underflow from the river.
DRAFT 2/21/14
-16-
0 0 .
).J .J Te•pe~ature ]!!!
Analrs es of teaperature dat~ have been l i •ited to considerin& plots of daily
mean teaperatures at various points, primarily usin& 1983 data. Li•ited plots
of slou&h temperature versus mainstea teaperature have also been ••de. These
analyses have used provisional 1983 temperature data provided by the Alaska
Department of Fish and Game. In some cases, ADF&G was gracious enough to
provide data which had not even been full y reduced, in order to expedite the
present study' Thus these data are-subject to revision, and some error may
even have been introduced durin& our reduction of the data. Nonetheless, it
ia believed that t he present data are sufficient to illu~trate general trends
in the water temperature data, and thus support t he following discussion.
At slough SA, data are pri~rily available from intragravel and surface water
measuring points at the mi ddle a nd in the upper reaches of the slough (Fig.
10). The intragravel datoa show essentially the same behavior, with
temperatures gradually r i sing from a bo ut 3°C in early Hay to about 5° C in
late July , and then fairly rapidly fa l ling to about 4° in late Aug u st (Fig.
lOb). Temperatures i n the middl e of the slough are generally higher than
those at the upper end of the slough , except in the latter half of July. The
intragravel temp erat u res gene r all; appear to be subdued reflections of the
surface water tempe r atu res at corresponding points. However, surface water
temperatures for the middle of the slough exhibit greater variations, rising
a s high as 14° C i n late July (Fig. lOb). Surf ace water temperatures at the
upper end of the slough only rise to about 7.5 °C, but show the same gener al
'\ t:rC&r<!S -~ -~ the . middle of the slough. Since this slough W&l reportedly not
t-·-.~ overtoied in t he 1983 record, the high temperatures observed in the surface ~.-:1:
\' water the middle of the slough can probably be attributed to solar heating, J) :" 0
.-
1. rather han aroundwater inflow or surf a ce water discharge as a result of ~~
' ~ -l..Av-~-overto)pi... It should also be noted that the maximum surface water "0 1 , • \ .l
'temper.ture at river cross-s ection LRX 29 during the summer of 1983 was also
' 0 about 14°: "'c in late July, comparable to the maximum slough surface water
I' I .
I I temperature.
DRAFT 2/21/84
-17-
I
l
At sloufb 9, data are av•ilable f o r surface ~ater and intraaravel measurin1
. .
points within the slouCh. surface water and intraaravel measurioa points on
the .. instea, and fro. three_aroundvater vella (Pi&• 11). Both mainstea
probes, as vell as the surface water probe within the slou&h, shov essentially
the sa .. behavior: winter teaperatures are near zero, with the intraaravel
temperature about a degree highe~ than the surface water temperature at the
mainstem durin& late September and October of 1913; teaperatures at all three
points begin to increase in mid-Kay and reach maximums of about 13° in late
June. and perristinG through .July; temperatures then fall to near zero by late
September. In contrast, the intragravel measurements at slough 9 remain
essentially constant at about 3.5°C from mid-March through late August, vith
temperatures exceeding 4°c on only two occasions, and falling to 3° only
once (Fig. 11). The groundwater data show considerably more variation than
the slough intragravel data. At borehole 9-lA, which is nearest to the river,
temperatures reached a low of about 2.5 ° in late February, and then rose to
over 5° in early September. At borehole 9-5, near slough 98, temperatures
fell from 4° in early January to 2.5° during April, and then rose to about
5.5° in early October before again falling. At boreho l e 9-3, temperatures
were relatively stable, varying between 3.5° and 4.5°. However, in
general, during the winter period January to Hay, temperature variations in
9-3 were opposite those in the other two wells, rising when they vere falling,
and vice versa. During the summer, temperatures in all three vella generally
rose (Fig. 11).
In very generl l terms, the groundwater temperatures at slough 9 appear to be
very ~u~d~ ~efi;ctions of surface water temperatures in the vicinity of
slouJhJ9, with peak aroundwater temperatures lagging peak s urface ~ater
temper urea by two to four months. However, it has not been determined
wheth the groundwater temperatures actually reflect changes due to the
~
infil~atio.of river water into aquifer materials, or whether the groundwater
merely~~eflects seasonal variations in parameters such as air temperature or
solar r~aiation.
DRAFT 2/21/84
-11-
• \' <
I""
\ ''. L-.~
\_" 0 •
At slou&b 11, data are·av.ilable fot surface water and intragravel measurina
points within the sloulh. and surface water measurin~ points on .the mainstea
(Fi&• 12). The intraaravel ~emperature within the slough is rather unifora,
increasina slightly froa about 3°C in January to 3.5°C in early Hay, and
then re.ainina essentially constant through late August. The surface water
temperature within the slou&h is approximately the same as the intragravel
temperature through late April, but then increases and varies between 5 and
7°c from Hay through August. There is no apparent relationship between
aainstea and rlou&h water temperatures, in striking contrast to the fairly
strona cc•rrelation between mainstea and slough discharge at slough 11.
At slough 21, data are available for surface water and intragrvvel measurina
points on the mainstem and at the mouth and in the upper reaches of the slough
(Fig. 13). In t ragravel temperatures at the mouth of the s l ough were
approximately constant at 3.5°C from January through April, then gradually
increased to almost 4°C by late August. Intragravel temperatures in upper
reaches of the slough varied around 3°C from January through April, but then
increased to about 6.5°C from early June through mid-August, with
I ;
considerable temperature variation. Except at the mouth of slouah 21,
intragravel temperatures were essentially t h e same as surface water
temperatures at comparable points, suggest ing that the intragravel water may
result from downwelling of surface water rather than upwellina of cooler
t-JAL.-I .• ~ ......
!)'-~ t ( .
groundwater.
3.4 Ar.alyticil Models . -~
Limited mathematical modelin& of groundwater levels and temperatures baa been
perfo~d during this study. The basic objective of this modeling was to
invesJPgate the rate at which changes in mainstem stage or temperature might ,.
be pr~ ... ~ed toward the sloughs through the groundwater regime. No attempt
was .. -. to actually simulate groundwater dischara~ to the alouaha, or the
temperature of such discharge. To this end, some simple one -dimensional
analytical models were applied.
DRAFT 2/21/84
-19-
I
3.4.1 Groundwater Level· V~riationa
..
As descti bed by Hall and Moench (6), flow and head variations in station~ry
linear strea.-aql>ifer syste111a can be simulated by application .o(the
convolution integral. Head ~luctuationa in a semi-infinite aquifer due to an
arbitrarily varying flood pulse oo the streaa can be expressed aa ao integral
involving the stream stage and various aquifer properties. Th e integral
so l ution can then be expressed in approximate fora by a finite series which ~&
convenient f or computer evaluation.
In its simplest form, the solution presented by Hall and Moench (6) can be
expressed ~• follows :
t
h(x,t) • fF(~)U(x, t -"t)d~, (1)
I)
where h(x,t) is the groundwater e l evat i on at distance x f rom the stream and at
timet since the simulat i on began; F(t)•H(t), the river stage at timet; and
U(x,t), the instantaneous u n it impulse response function, is given by (6)
(2)
and Ol is the aquifer diffusivity, given by the ratio of transmissivity to
storage coefficient. Equation (1) can be approximated by the finite seriea
,
I
h(x,t)~ L F(k)U[x, (i-k+l)~t] ~t
t:l
(3)
A computer pro&ra~ has been written to evaluate equation (3) for a variety of
values of the input parameters. Io general, it has been assumed that the
aquife~ydraul ic c onductivity is 500 gpd/ft2 , aquifer thickness is 100
feet , ~d the storage coefficient va r ies between 0.0002 for a rt esian
condi~ns and 0.2 for water table conditions.
'
DRAFT 2/21/84
-20-
-.
Ficure 14 sh~• the •i•'~ted croundwater le~~l as a function of ti.e at
vaf ious ~ia t ancea fro• the river. The surface water hydrocraph ,utilized was.
the water level at the Suaitna River sidechannel above alou&h 9 for the ti•e
period Kay 25 throu&b June 10, 1983 (R&M Consultant• ?roviaional data). Five
da ~a points per day were interpolated fro• arapbs of the side channel ata&e
durin& that period, The observed water level variation• at boreholes 9-lA and
9-5 have also been plotted on Ficure 14. It ia interestin& to note that the
~bserved &roundwater l~vels are most closely matched by simulated curves for
a r tesi ~n con i~iona, rathe~than water table conditions (i.e., for a stora&e ~
coeffic 'ent o1 0.0002 rather than 0.2). However, the data for borehole 9-lA,
located about 700 feet from the river, are most closely ~tched by the
si~lated water level at a distance of about 2000 feet from the river, while
the data for borehole 9-5, located about 1500 feet from the river, are moat
c losely m ~tched by the simulated water level at a distance of about 1000 feet
from the river. As noted previoualy, water levels at borehole 9-5 are
probably affected by slou'gh 9B and the beaver dam at the mouth of 9B, and thua
would not be expected to readily fit the present theory. These results
suggest tha t the groundwater aquifer in the vicinity of borehole 9-lA may
behave somewhat as an arte s ian aquifer rather thau a water table aquifer.
However, well logs in the vicinity of slough 9 (9) would suggest water table
conditions. It is possible that local overbank silt deposits or relatively
thin layers of fine-grained materi als may act to partially confine coarser
water-bearing layer s in the area, thus resulting in localized or short-ter.
hydraulic behavior as an artesian aquifer.
I . I : .. ; ..
Fiaures 15a 'throuah 15d show the simulated aroundwater level as a function of
distalc away from the river for variou ~ times and various values of aquifer
diffus ity. Theae figures aenerally illustrate that as diffusivity geta
larae (i.e., the storage coefficient gets smaller), the effects of variations
~
in riv\r ltaae are more rapidly propagated into the aquifer toward adjacent
slou ah~~ For example, Figure 15d shows that for fully artesian conditions,
small variation• in river stage could be very quickly transmitted, as a
pressure wave, a distance of over 4000 feet into the aquifer within one day.
Thua, for fully artesian conditions, changes in river stage could influence
groundwater upwellin& to the sloughs almost instan t aneously. On the other
DRAFT 2/ 'l.l/84
-21-
I
hand, Figure 15a suggest' that for water tab~e conditions, variations in river
stage ai&ht not have an appreciable effect on groundwater cond i tions except ·
very near the river. Consequently, under water table conditio ns." variations
in •iver stage aight not be expected to significantly affect averaae
groundwater upwelling ·to the sloughs unless the areas of upwelling were
rela tively near the river.
3.4.2 Temperature Variations
Groundwater temperature variations have been considered by a process similar
ro that used to analyze water level variat.ions. Acres American (1) presented
an analysis of coupled thermal and groundwater f l ow for a single squ are-wave
temperature pulse representing the average river water temperature. By
applying the convolution integral approach of Hall and Moench (6), the
analysis of Acres American (1) can be extended to consider shorter time frame
variations in river tempera ture.
Equation (1) can again be applied, with F(c) nov being given by the river
water temperature. The instantan~ous unit impulse response function U(x,t)
can be derived from the unit step response function P(x,t) by differentiation
with respect to time (6). P(x,t) is essentially the solution given by Acres
American (1),
T(x,t} • 0.5 erfc [(x-v t)/2(Dt)l/2] r (4)
wilt:lt: 't(x, t) is the groundwater temperature at time t and distance x away f roa
the ri.·v r due to a un i t step increase in river water temperature (1); vr is
the a age retarded velocity of the mean temperature, which accounts for heat
excha. e between the groundwater and the soil skeleton of the aquifer (1); and
D is rbe coefficient of hydrodynamic dispersion, which accounts for the
)
temperature dissipation as a result of mechanical dispersion durin& transport
through ·the porous medium ( 1).
DRAFT 2/21/84
-22-
Results of thi1 analysis -generally confir.ed · the result ~ of the aiailar s t udy
perf or8ed by Acres American (1): as a result of heat tr a nafer .a~d aechanical.
dispersion durina flow throuah the aroundwater reai ... aho rt-tera variations
in river teaperature are rapidly damped. Consequently, by the ti ..
aroundwater has traveled froa the river to a nearby alouah, ita teaperature
coul d easily be approximately equal to the aean annual river temperature.
This conclusion is consistent with t.he bs ~r vations noted previoui!Y that
slou~h intragravel temperatures. vhicb probably represent the temperature of
upwelling groundwater, are relatively constant throughout the year, and are
approximately equal to mean annual river water temperature.
4.0 CONCEPTUAL SYSTEM~
The results of the present study do not permit a single model to be formulated
which can describe the discharge and temperature variations wh i ch are observed
at the various sloughs studied. The hydraulic and t hermal behavi r of each
slough is substantially different from that of the other sloughs studied. The
discharge at slough 11 seems to correlate very well with mainstem discharge,
while the discharge at slough 9 is largely con trolled by mainstem overtoppina
of the berm and t he discharge at slough 8A may be complicated by factors such
as surface runoff and groundwater underflow from sources other than the
ma i n s teDl of the Susitna River.
Reg3rdless of the complicatina factors affecting discharge from each slough,
the available-data suggest that the temperature of upwelling groundwater
remaina -f~lyconstant throughout the year, at a temperature approximately
equal;{ the mean annual mainstem temperature. Th\s study has tended to
confi previous conclusions that heat exchange between groundwater and soil
mater~ s, and mechanical dispersion during groundwater transport through the
aquif~, •~e reasonable mechanisma to account for the observed groundwater
temperttures.
DRAFT 2/21/84
-23-
1 d ~ .J1
It is doubtful that adaitional studies vithi~ ·project constraints can improve
aianificantly oa the current 1tatua of knovledae reaardina the · alouaha. ·
However. one additional field study which miJht provide aianificant add i tional
info~ation with a relatively aaall investment of project resources would be
additional atteapta at aquifer teatina. utilizina exiatina vella. Available
data indicatea that no successful aquifer testina has been conducted at any of
the project well location• on the Susitna River below Devil Canyon. Fallina
head pe~ability testa were reportedly attempted at the deeper vella at
slouab 9. but~be testa were not successful because of the high permeability
of the material tested. Successful testing of these wells might require
su1tained puapina at a relatively high rate for a period of several hours or
days. This would require the use of pumping equipment. el£ctrical generatina
equipment to operate the pump, and probably fuel for a aenerator. Such
aquifer tests, or additional attempts at fallirlJ head testa ot" siailar in-situ
permeability testing, could help confirm the nature of local aquifer material•
\e.g., water table or partially confined) and quantify the degree of hydraulic
connection between the river and t h e groundwatet" aquifer. Such kn~wledae
could help refine present estimates of the rates at which changes in mainstea
hydraulic or thermal river conditions are propagated tt.rough the groundwater
regime toward the sloughs.
5.0 EFFECTS ~ PROJECT OPERATION
Th~ results of the present study do not permit any detailed projections to be
made of the slough discharge or temperature variations which might result from
clr.;n,ac'lt in main.at;m condition• aa a reault of project ope rat ion. Because of
the substantial differences among the sloughs in their hydraulic and thermal
behav~, it ai&ht be necessary to construct a model of each individual slough
in or~r to make detailed prediction s of the effects on the sloughs of changes
in ma~atea conditions. However, some general conclusion• can be drawn based
on the~reaulta of this study.
DRAFT 2/21/84
-24-
So.e alou&ha, such as •louah 11, will probab~y respond fairly directly to
chana•• in aainatea diach•ra•· Slou&h 11 discharae is correlat•d fa i rly well
with .. iastea dit~;a., so ~ny lona-te~ increase or decrease in aainstea
~~~~ ~rae could result in a siailar increase or decrease in averaae slouah
dischar&•• However, any sue~ relationship can not be quantified based on
available data.
Soae slouJhs, such as slough 9 during the s ummer of 1983, will be ove rtopped
durin& auch o~ the t~ as a result-of hi&h river staae or ice staging. Such
sloughs aigbt be effectively considered as side channels of the river, rather
than sloughs, during such periods. To the extent that the mainstea flow which
will result in overtoppina of the berms of a particular slough is known,
projections of project flows can be used to estimate ~hat proportion of the
tiae such sloughs will carry predominantly mainstem flow (at mainstea
temperatures), rather than groundwater discharge.
Howe ver, most sloughs will probably be similar to slough 8A in that it will
not be possible to sepa r ately determine each factor contributing to the
discharge of the slough without conducting very extensive additional field
investigations at each such slough. It is proba ble, however, that for sloughs
which are as complicated as slough SA, the contribution to slough discharge as
a resu t of groundwater underflow originating at the river will be small
enough that project variations in mainste~ discharge will not significantly
affect the slough discharge under moat conditions. However, it is not
possible with~present i nformation to either confirm or quantify any such
~elations.
I Temper~urea of groundwater discharge to the sloughs appears to be reasonably
approxl!ated by the mean annual river temperature. It is likely that any
~
variat\ons in mean annual river temperature as a result of project operation
will aiao result in a similar change in the temperature of groundwater
upwellini to the sloughs, to the extent that such upwelling is derived froa
the mainstem (e.g., as is probably the case at slough 11). Similarly, for
sloughs such as slough 9, which are frequently overtopped, any changes in
mainstem temperature will al.to result in similar changes in the mainstem flow
DRAFT 2/21/84
-25-
vhich is diverted dova· the alouah durin& ove~t~ppina. Thi• could induce
dovnwellina of river water durin& overtopped perioda, which would have so ..
influence on the averaae te.perature of aroundwater which ia'discharaed to the
slouah. Aaaia. it it not possibl~ with present infor.ation to quantify such
effecta.
6.0 SUMHAIY
This study provides a review of much available hydraulic and thermal data
reaardina the~ischarae and temperature of side slouahs tributary to the
Susitna River between Devil Canyon and Talkeetna. This revie~ of the data has
served to illustrate the complexity of hydraulic conditions at the sloughs.
It has not been possible to formulate a sinale conceptual model which can
serve to describe each individual slough. On the contra~, each of the
sloughs studied in detail differs significantly f rom the other sloughs in one
or more important respect. Because of these complexities, it is not possible
to quantitatively predici the changes in slough discharge or temperatures
which might result from changes in mainstem conditions as a result of project
operation.
f The discharge from some individual sloughs (such as slough 11) can probably be
_sorrelated fairly well with mainstem di i ~t·.q., so that projections could be
made of the changes in slough discharge which would result from changes in
main stem discharae. However, the discharge from most sloughs will probably be
influ~nced by diversions froa the mainstem as a result of overtopping,
)fOVerland runoff and tributary discharge, and other factors which will precluae
.V 9etail o~:<l -·..,C"ttjec .tions of discharae for each slough in the study reach.
. .. I -+-~·-·' /T.:: \ I'.). rf:Y· .. ./ . -
~j The tett ~~ature of aroundwater discharge to the sloughs does appear to remain
~· ,' relat{ely c..>nstant at a temperature approximately equal to the mean annual
, .. : river l:emper•ture. However, without knowing the proportion of discharge froa
~:i . an individual slough which can be attributed to such gr~undwat~r discharge, it
r:11,• it not possible to project the time-variation of heat which is .nailable for
;.' salmon incubation at a particular slough.
DIW7. 2/21/84
-26-
1.
z.
3.
4.
s.
6.
7.
8.
9.
10.
11.
12.
13.
REFERENCES ·
TITLES ·· · · ·
Acres AMric:aa Incorporated, "Suaitna Hydroelectric: Project, Slouah
Hydroceoloay lleport," prepared for Alaska Power Authority, March 1983.
Alaska Department of Fish and Ca.e, "Suaitna Hydro Aquatic: Studies, Phase
II Basic: Data Report, Volu.e 4: Aquatic: Habitat and lnstreaa Flow
Studies, 1982. Appendix C -·Temperature Data ." 1983.
Alaska De"partment of Fish and Came, "Susitna Hydro Aquat i c: Studies, Phase
I I Data Report, Winter Aquatic: Studies (October, 1982 -May, 1983).
Appendix A-Continuous Surface and lntraaravel Temperatures," 1983.
Box, C.E.P., and C.M. Jenkins, Time Series Analysis, Revisec Edition,
Holden Day Publishing Company:-1976.
Freethey, c.w., and D.R. Scully, ·~ater Resources of the Cook Inlet
Basin, Alaska," U.S. Geologica l Survey ilydrologic Investigations
Atlas HA-620, 1980.
Hall, F.R., and A.F. Moe nch, "Application of the Convolution Equation
to Stream-Aquifer Relationships," Water Resources Research, Vol. 8,
No. 2, April 1972. -
Pewe, T.L •• "Quaternary Geology of Alaska," u.s. Geological Survey
Profess ional Paper 835, 1975.
Pinder, C.F., J.D. Bredehoeft. and H.H. Cooper,Jr., "Deterlll ination of
Aquifer Diffusivity from Aquifer Respo nse to Fluctuations in River
Stage,"~ Resources Research, Vol. 5, No. 4, August 1969.
R&M Consultants Incorpor ated, "Susitna Hydroelectric: Project, Slough
·Hy.!rology Interim Report," prepared for Acres American Incorporated,
Dece~ber 1982.
Tuck, R., "The Curry District, Alaska," U.S. Geological Survey Bu l letin
857-C, 1934.
--'ton, W.C., "Selected Analytical Methods for Well anci Aquifer w,. ~valuation," Illinois State Water Survey Bulletin 49, 1962. --l --W4 ~ton, W.C., Groundwa t er Resource Evaluation, McGraw-Hill Book
Y:. Company, New York , 1970. .
Weber, F.R., "lleconnaissance Engineering C'.eology for Selection of
Hi ghway Route from Talkeetna to McGrath , Alaskll," u.s. Geological
Survey Open-File Report, 1961.
DRAFT 2/21/84
-27-
, . . ; .. f j I . I . ·, I ' I. : -J -...l..-,-)-.-1 • I --I j I I I J--+ --. -~--"-,~ -~ I I I __ L ,--,--a · ,-,-1 __ I , ~-... --' -,--~--, : ~ y --~-·-·--. -., I I i _[_......J __ :-.-o--z;)·'l .I. ; -·-L---~-1 . I . ~ I f}::;. --•·-• 0 I ' ,._,--.·-1-a ·.., 7 Y _t_ I. I -O~D-. ...!._j _Q ·-r ·-~-I I _ ~ -.:; r-'o : ~,.. -.:.._ ..f ..l~;.s ~ II!· +--o, -~--. .,.. :• .,_ ·"-·' ..a -. ('\ .; ' ..: : ~ -. ~ ·-! ¥ c -; --; ~. if.! ~_, ~---"c(')-C _.. ; 1 _1 Jf( __;_-J; '-' --I . -'-I ----I ' -.-~ -~-I . I '·----.-~ .... ~ : _1 __ .-:___l. ~· -~ I ·-~ o ' ' ' . 0 I . '
I < _j_ • ---·-. I I , ...... :....--~---~-· . I ' I j ----. ....... I
. I ' I . ---~ I I I . --• : ___ ._ :-;-I I I I • I I
: ...... ~ ' • I ;__ ~-i I I ___;_ __
; __ ; ·-I 1 I • -;-i : _
• W. ' ' -·-: 1 0 I 0 '--1--;---j-.1 --·:~ -;a..!o _•_o :-. ~r -o-·o ·: o ~_! __ -!-_l...._~--
-.1 0 8 8 + -0 ··; 0 ·• ...(-! ! I I
0 ' · . -~~ A' I ·" I --·--·oo-,.-. . r · •'' .11 '---I : - - -• 0 • :~-~ -···~ ·;-! j I : -
--·--: I I ~--·-·~ --. -' ' .,.,. /"'' --·· ----~-....
(.. _:_ -'-·· --~ ;::' ~ ~ -/"\-• __ ---1 'oJ
.. ,-· I I ; . I .. • ~ ------'C.' ' ! ..... ~ ·--.... -' . .J ' .• ---~-~~C)~:-~--5 -1 '1 -~r ·-~' -~-~J·_ ~ ""~ --~ --.:::1 . -,
4 -~.d\ ' . 0 -:'I 1 -. ! .... .;-'-.) _....:.. ~ ~---. -,-,• : f ' ":'-----'---: ill: ;>I . ;., . ; ~ __ -__ · -'· --> i 'i ; _.&;: ..9 . ..: -
. _..li# --: '" • ..<· .( J# -u : I ' .... }..__J. _ _!_ -J ---~-~'U Cf "1
1
_ ·-' J -: c -~--I . • -• • I -~~~~-6: ~L. .v~ ... 1..$ :} i _a)_ I!-....:,-! ~· :»1_
. I J ~--:r; . . ~ ~--.) l"-~., . or :_,. ! . __ . : , , . ; , ' ...... f;l
. .,J ..... ~ -, --. ' : . . ~-'-~-.... -~-
._7 .-. ; :_.: ~-.---~·-:~--~-;-:-; .L ~.~ .... .n * --· . r-. ,-, . ~-.J .... -;:::; -,.-: , J.-· ---· -..1 ---~ -I I i
' \1 ~-·--: --:--~ : . I ---!---, 1 ~ •• --, • ..... 0 I
1
° . -·--·--· I 'I' ----~~--~! --•
-··---·---
11. --~-. I .-, .i I . I ' I I ' · --f I I ~ . .........J-,-, . -:--~ . ! I -~ : : ({l:-"""j I I ; : . 0 -·~ _,.. • (\ I 1\__l_. -. r: ! ~· · ~-.:..6 ::-1 -:--«':; I ~ ; .
... ·--~-. -~ • I -: I ./\• I _j_ -· ...... , --·'loJ/ '~' ' .. a ------'-"-=---. .
• I .~.-__:..... • ; ~ ~ ~ ! j . --
--------:-I ! _l _.....__! .... ;
' ! ; j I .... : . ; ..... -
·-~ _......._. -· --~-----·~ ~--;.. ---7 -. . -=-·--.-.. -·--0 .... :---"1 -
-·: --; -~~ •. --cD
-·--4lfll" • .,-•
. . ~
. J -•. -:
.(» . ..-.. --: --··-0 . -~ .~~ . ., --
~
"' ),.._ ~ " "" 1.::' -"'
_ ..
E " 0 -"" '--\) c ~ ~ ~ ~ (.) \. -3 J 1-
• -~ "' 1:: \J ~ -1 ~
"' ~ ~ c
~
E) ~
c 0 .:; • = ·.~ ... -: ~ . c L. 0 ! -. ~ ~ > '0 L. c . ~ ~ e o a j . • 0
0
U) =
~ I 0 N I ., • ~ ..
0
0
0
0
tl
...
0 0 0 -• "' .. u
·C)
0 "'' " ~ -.....,J
l
-..! l( ~
1 ....I
l'
"" ~
! ~
~ ~ ~
\.J
~ l
" ~ ~ ,_
~ ~
1 ~ r
:S +-
~
,_
"' ~ ::s
"' • -
" ~
.;
"' ·-~
0 .., "'""'\ J ., , t-.l 't ~a. ... ~
~ ·-... c-'-' ~ ~ ~ -
•
v ~ " ~ ~ . ~ v ,.. " ~ ~..s .. u }' ~ -S ~ ~ .....
~ ~"" Q
""' u
} 'I t'
" ll' -' " ~~ ... -:s ~ 0 """ .. -~~
~ \.:)
~ •
~ ~ ·-• ~ ~ ~
j
G> . --
-,.--..
"' ·-~
Q
0
..: -
1.
1: ..
·' ...
-.. -.. ...:. T /
.......
' '
I
~ . --~ • /
' ' ' ..........
' '
l -
,J -I
/
/.
(',
~-'·
''-" ' '
' ' '
I
I
I
~ /
'""' ..... I
.-
........_
I
I
. . . . . ,,• ... · : .. o · ... -I
'4l
I
I
I
"" s
•
M
0
I t. ~; -· -•---7 __ ~ , __ _ 5 __ _ 4 __ _ , __ _
2 __ _
'-•--7 __
6 __
5 __
4 __
I I
I
I
I
I I
I I I
I
I
-~--=
I '
,,_.
I I I I
I
I I
I I I J'l
_l Il l _l_l
l J u _!__!__!
_111 I I I I
I ·~ 3__ ... ,;
-2 - 2 __ ' ' I . ' ' I I I
' ' '
~
~
t=
a_
7_
6 _
5_
I
~
4_ §
3_
2-
il I I
i I
I I
' UJ I
I~ COl
I I
t I
I I I
I I I
I I I
I
I I
I
I ' I I
I
y
I I _!I
I I I I I
f I t t
I I
I I I I _I
L ~~~~~t+~l-t~~t+~~~~ ·~t+~l-r+1~rf~1rr1+1 ~1-t~Hr~1-t+~~-r~,'r+'~·~ ' II I I __!_
I I I! I _1_
I 1 _1 I _!_lt I I
--c'
~
I -.I !)._,~~
I ~ I I ~~ ·
I· f I I I
I IJ I I I 1 I
I I I
I
I·~ I I I I
i.J
" " y... Ill IY
~ ..... J •. o
~
cu -
9 IH I ~~ VI I
"" ~ O.l
,'( I /1 V I I ~ I
(~ II 1/i I I I Y I
I I l I I I /I I l ,_... I
l•) I II I I ... /' I I /
j II A I J .J~~A 7
.l
t ·-o.,
\1
l'
G
~ u -
o.f
7 r I VJ.C I I A
I 11
J
J7 I II I I ~ I /1
l I I 1 I Y IY..
/I // ~ 1/ I I V A I
1/ IN I I I J 1/ 1/
I
·~ o.2. l l / I I
I J I ). J1 I I V Y.
II 1/,. 1*1 V I ! ~ tr I
l .
! I fl I /• ~~ / 17 I j I
I I I
I Vi I I J'l 1/1/1 $ /: I Vt I I I
I I i I
0 0 /0 2(} 30
E { D..f.J ~ t,.""' ~ "oK-r.S
X : J.;rtQ"<.c frc,..,. r:v~ ba.t\k
v s Afl4•"14r-J.;.f!Fv..slvrty) r~,. of -tr~~f1141Sfl'Yrly (T) ~
~+•r.:/ rvdy ( r)
Fl,urc ). 6-rrJv.kJ.~v.t~ Lt.vd VA.r;-..1~1 ;.., Kt-Ij>f~MJt!!
-to R. j v ..,_ )j~ ~ . FICA c.tiA aT f'nt\J
... ,
R fk Cuv.s<A{f.,. ... t.s
(,(SC-S
/'ff.3
AU OCTO WI
~
:5I
:s
::t ...
1
'l\..
~
" ;.-..
.:0
" ,.
~
"' ~
Si:.
~ ...
'\
" ~"<-
·' ~ :; .;
:-··~· .::
J;\:',
I
;
... ·-;~
" ~
\J
~' ~
~
~ u
"" ,_
\S
-r
~
:S
0 -""
I I ...
·.
•
, . .. . ........ . ... . .
• . . . . . . ·-• •
f f-. .. . . ' ... • •
••••• • • • • • .... ~ . . . .. .. .
• •• • •• • ••
• • • •
3
•
1-
••
• • • • •
• • • • •
••
~ L---L-·--------~·----------~·----------~·----------~·----------~·L---~
/0 IJ .l,..O LJ )~ 1r
M a.. ; ""rl ~""' J i f c.."' ar-.7 ~ 4 (;., IJ. C r ~ e..k/ to 1 c f 5
7. 5 / o "'"J L.. II
D; ..s c."'ta ? ~
0 ) .s c.k a.r.1 ~ Vf ,
A-7-G-u l,l Cr e e-1(
...
•
• 0 ••• ,. • •... : ··: •
·-:=
)0
~ Jl
't •
t!") ,, z.f" -~J -" c
~ ""'~ \'t ~" ~~ ..
-S '1c1 If'" • • • -~· \l
"' • • • •
~ • 0
-t t(
"" •
~ • f
\) • • • e • • -•
~ '/0
• • •
• •
• • \ ... ••
• • • • • • • • • ... ~·· •
0 '" ,; ~ 'Lt' )c> Jr
JA ~l~sf'~-J. I.S"c:..h-.rJ ~ o--f 6ulJ Cr~e-k1 /0 3 cfs
F~7urc.. fl s I~ "'"1 "' r-~ /) Jscha ,.., ~ vr . Mo.. i"'st~""" .. ..
O;s-c~a. rJ~ aT 6-· (il Cre.•k.
• • •
II I
I
Ill
• ~.
~ II
••
'1.
0
;· ..
. " : -~ + . :.
+rt: . :" :: ....... ··-;
........ . . .
. . . ,, ..
s 10 n 20 a
JULY
11VL
................... . ... ... .. . .
..... -.
. . : 1 !:=-:
• • • • • • • .. • • ::: • • •• ~ : ;J -. 'I
.. ··..;: .. :.·· ;· ..
!'-~:i:~ .
-~~'........... " . -~:,·
Tn :· • .·f
. . .... . . . '·
. ;...; . . . . . :-
.:.·~:·· ..... . . .. . . ..
0 • : ••. •• • + .. . . . . . . . .
. . ...........
" .
"
s 10 u 20 n
SlPUMBEJI
. . . ............
..... . ........ .
... . . . .... ... ... . •• . . • •• • • ••• ...u
-: :::: :jB : . > ::.: .: .. : ::: :.
• ••••••••••• ,. 0 ••• ::. ~ . 0.. :: : ~. :.. . ..
.. ;:;1
..· ~1 1 .. . . ..
:·· .... ,
. . . . . . . : . . .. . . . . .. : : ~ ! .
. : . : . :.. . . . " " :: : ;::~l ... -l' .. I . . ;l . =~ . . :-ij
. . ::t
;:;! " " . .. ,....
. ij
!I 10 u 20 a
OCTOBEJI
!I 10 u 20 a
NO'o'lloiBIJI
S 10 IS 20 B
O(C(MBEJI
5a«.rt~: ,f[)F~fi-(3)
!rPF ~c.-Provin DkJ !'H'3 [)a:f,._
I 'P'3
f;Ju.r~ /0 b. 5"/~"'f L.. ?A
~""'f!o:/~r
Sou....-us: lt-DF ~c-(3)
/r/)fi-(; foviJ~ l<tKJ P.JA
fZ +k G-.--.r "-ft._vfr ProviJI o1-1.J /Cff'") [).q.,..
1/. ){o "Jf.. 'l JCA.rfHt:.
tva/"~...-OA-J C r-Du..Jw4~...-
/e """f e r..:f~M-,. r, I 'fir J
5 Dl.Vf"c.~r: A-t> F.J-~ ( 3)
A-D F If 6-f'n vii ;,-.,.d /'ji'J /)o:/A..
I i ~ 3
FiJL<.rt! 12.. 5/o"-/ ~ If lv...t-e.--
r~r...pe,.,._l-u..-~~., l'/?3
JANU...,.
(1...
JNIUAiff f[SRUAII'I'
A-D F i-6-c:J )
A-DF + {r Prwr.riovo..J. 110 P4tA..
/'Pi3
•ucUST
F ;;r.r~ /], 5fCJu)l, 2/ !v-,.fey-
Te"'l'r41M'~5./ /<t.f.::J
.. '
~ " t .........
'to> .,
\1
~
..,.
~
" \.
" ~ .,
• ~
" -
\. .. ....
' ).
610
J01
l<¥1
j,D]
606
6or
(
6o{
El t:t.(I ~J.
F;;v..-t.. 1-f. S;,..,..._f..r~ c-.-
Va..-;,.;f~U>V i"'
River Sf~L v
...
~,
~
! ,..,
'j.)
" " c~
....
~
~
' -,)
~ ::..
"
~
i--:;
,.,
,.,
,.,
I flo
I•(
I of
to)
1
I>A-. Siw-c.t(4~ C-rou.J..~v-4~
Lt.v dr YS', lhsrtl~Lc f,..,.
~iv~ S~ ~ Co efiV~J..¢
.o1
I
6oJ 0 I
I
I
I
2 !1' ((WI (1.1' z.l.IO 21!» l?{fJ
riv ~ fe ~ ~
Fi;u,-e. Js-b. S iMu[af~ ··cn ·~-~.Jwer-
Lcvch VI, P1JTtJ.....ce. frT>-
R_;ve,..J S-for-,c. Gcf/rcJc><J-:: tJ.oL-
--o\
~~
I '3 I
I
I
I
tJ zSO ]UtJ . )>I" -_)]tv --. -
JSC. Sin.. fA. far '-l Cnu.11./t,;Aj-er-Lev d.r
·vr. DTJra.ttcc. -Fnw. ~iv~.,./
S for~ l' Co e-ff ro 8,¢ -= 0. on_
f;J~ /S"J., s;.,_J..¢4( · (;yr;V.n}r..Ad~r lev dJ
VJ. DTJfa,a. .frn-. R. i V(/.1
rn~-,.: ~erf1.-c..:~"'t =-o .ooo2_