REGION X
UIRVE1LLA
I
EVALUATION OF
MILNER WATER
MODEL
ENVIRONMENTAL
PROTECTION
AGENCY
SEATTLE, WASHINGTON
WORKING PAPER NO
PA°910
o
8°
7
o
01
-------
EPA 910/8-75-092
July, 1975
EVALUATION OF LAKE MILNER
WATER QUALITY REPORT
PREPARED BY:
JOHN YEARSLEY
E.P.A. REGION X
1200 6th AVENUE
SEATTLE, WASHINGTON 98101
I
-------
THIS DOCUMENT IS AVAILABLE IN LIMITED QUANTITIES
THROUGH THE U.S. ENVIRONMENTAL PROTECTION AGENCY,
SURVEILLANCE AND ANALYSIS DIVISION, 1200 SIXTH
AVENUE, SEATTLE, WASHINGTON 98101
8 1J1.
-------
A WORKING PAPER PRESENTS RESULTS OF
INVESTIGATIONS WHICH ARE, TO SOME EXTENT,
LIMITED OR INCOMPLETE. THEREFORE,
CONCLUSIONS OR RECOMMENDATIONS
EXPRESSED OR IMPLIED — MAY BE TENTATIVE.
3
-------
4
-------
TABLE OF CONTENTS
CHAPTER PAGE
INTRODUCTION 13
RESULTS OF FIELD STUDIES 17
Hydrologic Data 17
Geometric Characteristics 24
Municipal and Industrial Waste Discharge 24
Tributary Waste Loadings 26
In-Stream Water Quality 28
THEORETICAL ANALYSIS ^5
COMPARISON OF THEORETICAL ANALYSIS & FIELD STUDIES ^
PREDICTION OF DISSOLVED OXYGEN AT LOW FLOW 67
FINDINGS 77
BIBLIOGRAPHY 8'
5
-------
6
-------
LIST OF FIGURES
FIGURE PAGE
1 Map of Upper Snake Subbasin 15
2 Profile of vertically averaged velosity at 19
Snake River Mile 852,3
3 Map showing locations of irrigation return 21
flows in the Milner Reach of the Snake River
4 Location of major industrial and municipal 27
discharges in the Lake Milner reach of the
Snake River
5 Location of receiving water quality sampling 29
stations in the Lake Milner reach
6 Observed temperatures in the Milner reach 31
7 Observed dissolved oxygen in the Milner reach 32
8 Observed D.O. saturation in the Milner reach 33
9 Profile of D.O. Milner Dam 10/22/74 31*
10 Profile of temperature Milner Dam on 10/22/74 35
11 Profile of temperature at Miler Dam on 10/23/74 36
12 Profile of D.O. at Milner Dam on 10/23/74 37
13 Profile of temperature at Milner Dam 38
on 10/24/74
14 Profile of D.O. at Milner Dam on 10/24/74 39
15 Profile of D.O. at R.M. 652.3 on 10/22/74 *°
16 Profile of temperature at R.M. 652.3 on 10/22/74
17 Profile of D.O. at R.M. 649.5 on 10/22/74 *1
18 Profile of temperature at R.M. 649.5 on 10/22/74 ^
7
-------
FIGURE PAGE
19 Profile of D.O. at R.M. 654.0 on 10/22/74 42
20 Profile of Temperature at R.M. 654.0 on 42
10/22/74
21 Observed BOD at Snake R.M. 640.0 50
22 Observed BOD at Snake R.M. 647.2 51
23 Observed BOD at Snake R.M. 654.0 52
24 Predicted and observed diurnal D.O. 57
variations at R.M. 640
25 Predicted and observed diurnal D.O. 58
variations at R.M. 653
26 Predicted D.O. in the Lake Milner reach of 60
the Snake River showing the effect of
photosynthesis
27 Predicted and observed D.O. in the Lake 63
Milner reach
28 Predicted and observed D.O. in the Lake Milner 64
reach (includes estimated photosynthesis)
29 Dissolved oxygen as a function of flow for 70
various loading levels
8
-------
LIST OF TABLES
TABLE PAGE
1 Frequency with which minimum daily
dissolved oxygen was In a given range
during water year 1970
2 Observed flow in the Snake River at
R.M. 674.7, 652.3 and 640.0
3 Average velocity of the Snake River at 20
Highway 27 Bridge (R.M. 652.3), October
2224, 1974
4 River velocities in the Milner reach of the 20
the Snake River estimated from the travel
time of oranges, October 23, 1974
5 Gaged and estimated irrigation return flows 23
between Minidoka and Milner Dams, October 1974
6 Geometric characteristics of the Milner reach 2k
of the Snake River
7 Average observed flow, temperature, BOD^, 25
NH3-N for point source discharges in the Milner
reach of the Snake River and tributary streams
8 Carbonaceous BOD (5-day) and anmonia nitrogen 26
for surface and groundwater return flows to
the Lake Milner reach
9 Observed minima, mean and maxima for receiving kk
water quality sampling stations in the Milner
reach
10 Boundary conditions for the dissolved oxygen 5k
model of the Milner reach of the Snake River
at R.M. 654.0
11 Sediment oxygen demand values used in the 55
simulation of dissolved oxygen in the Milner
reach
9
-------
TABLE
PAGE
12 Production and respiration rates in the 61
Lake Milner reach
13 Proportionality constants used to estimate 62
standard deviation of important parameters
14 Contributions of major dissolved oxygen 66
sources and sinks to the total oxygen
budget of the portion of the Milner reach
of the Snake River from R.M. 654.0 to R.M.
640
15 Variation of mean minimum simulated 68
dissolved oxygen and maximum standard
deviation with flow in the Milner reach
16 Organic waste loading levels for point
sources in the Milner reach assuming
various waste treatment strategies
68
17 Improvement in dissolved oxygen at various 71
flows resulting from various treatment
strategies
18 Reaeration rates in Lake Milner as a function 72
of river flow
10
-------
11
-------
12
-------
LAKE MILNER REPORT
I. INTRODUCTION
The Mllner reach of the Snake River, between Minidoka Dam and
Milner Dam (see Figure 1), is classified as being water quality limited.
One of the important limiting water quality parameters is dissolved
oxygen, as monitored by the Federal Water Quality Administration (FWQA)
and the Environmental Protection Agency (EPA) at Milner Dam, show extended
periods of low dissolved oxygen. Conditions have been particularly
critical during periods of low flow when the discharges from municipal
and industrial waste sources were at their peak. For example, Table 1
gives the frequency with which the minimum daily dissolved oxygen was
below 6.0 mg/1* during the 1970 water year. Only those months when the
average monthly flows at Minidoka Dam were below 3000 c.f.s. are included
in this table. The effects of low dissolved oxygen upon aquatic life have
reached serious proportions. Major fish kills occurred in the Milner
in 1960, 1961 and 1966 during the food processing season. In addition to
the discharge of organic wastes from industrial and municipal sources,
the oxygen demand associated with return flow from irrigation wasteways,
*The State of Idaho (1973) water quality criterion for dissolved
oxygen states: "No wastewaters shall be discharged'—which—will cause
-—the DO concentration to be less than 6 mg/1 or 90 per cent of saturation,
whichever is greater.
13
-------
flow from irrigation wasteways, decay of algae in impoundments and
bottom sediments all contribute to the observed dissolved oxygen problems.
Reductions in waste discharge since 1971, coupled with above-average
flows in the Snake River, have resulted in substantial improvement in the
dissolved oxygen of the Milner reach. No dissolved levels below 6.0 mg/1
have been observed since 1971. However, dissolved oxygen levels below
90% saturation were measured in October and March of 1973 and January and
March of 1974.
Table 1
Month
October
November
December
January
February
March
April
Frequency with which minimum daily dissolved oxygen was
in a given range during water year 1970. Only those
months for which the minimum flow at Minidoka Dam was
less than 3000 c.f.s. are shown.
Snake River Flow
@Minidoka Dam c.f.s.
Avg.
2480
1394
1366
3118
2614
1182
7061
Min.
324
296
1290
1420
1340
135
1500
No. of Days in Month Minimum DO
(mg/1) was in Indicated Range
0.0-2.0 2.0-4.0 4.0-6.0 6.0
(mg/1) (mg/1) (mg/1) (mg/1)
1
13
2
10
3
1
7
1
22
6
26
31
27
22
29
-------
#r«*\ I
• -j .
1 v, • t/>^L
f ) w>r.
— ,-t -B c h ijT V > l l i\,
¦H ^ ' S ^v.
FIGUP.F 1
ENVIRONMENTAL PPOrgCf^/i A i'.'.f
WATER OuAU** f*' i
A«?EA 10
UPPER SNAKE SU3BAS
-------
In the fall of 1974, the Idaho Operations Office of the
EPA Region X drafted National Pollutant Discharge Elimination
System (NPDES) permits for the industrial waste sources J.R.
Simplot and Ore-Ida in the Burley-Heyburn area. Mathematical
modelling methods (Yearsley 1974) were used to support the
permit writing process. The result of the modelling study
showed that the point source discharges would contribute mea-
surably to the violation of dissolved oxygen standards at low
river flows, even if the discharge satisfied the appropriate
guidelines for best practicable treatment (BPT). The Idaho
Operations Office requested a water quality survey in the
affected portion of the Snake River. The purpose of the survey
was to provide data for validating the mathematical model,
as well as to assess ambient water quality during low flow.
A compliance monitoring program for industrial and municipal
waste sources in the area was designed to be conducted concur-
rently with the in-stream survey.
The Surveillance and Analysis Division of EPA Region X,
in a cooperative program with the National Field Investigation
Center (Denver) of EPA and the State of Idaho Department of
Health and Welfare, planned and carried out this survey during
the period October 21-28, 1974.
This report compares the results from the fieid observa-
tions with those predicted by modelling techniques, as used
in the permit writing process.
16
-------
II. RESULTS OF FIELD STUDIES
The data used to evaluate the mathematical model Includes
biologic data, geometric data, point and non-point source data,
and receiving water quality data. Data collected during other
surveys, including those of other agencies, have been used
where applicable.
Hydrologic Data
The U.S. Geological Survey records the flow of the Snake
River just downstream from the Minidoka Dam. They also monitor
the diversion from Lake Hilner, as well as the flow just down-
stream from Milner Dam. During the field study, EPA and State
of Idaho personnel gaged the flow in the Snake River at River
Mile 652.3 (Highway 27 Bridge). The flow at Minidoka and
Milner Dam, according to preliminary estimates by the U.S.
Geological Survey, and the results of the measurements made
by the survey team, are shown in Table 2. The U.S. Bureau
of Reclamation maintained constant flows from Minidoka Dam
during the period October 18 - 25, 1974.
17
-------
Table 2 Observed flow in the Snake River at River Miles 674.7,
652.3 and 640.0*
Observed Stream Flow (c.f.s.)
Minidoka Dam Highway 27 Bridge Milner Dam
Date River Mile 674.7 River Mile 652.3 River Mile 640.0
10/21/74 2730 3168
10/22/74 2680 3221 3102
10/23/74 2700 3127
10/24/74 2730 3204
Cross-stream variation of the vertically averaged stream
velocity at the Highway 27 Bridge is shown in Figure 2. This
data is from the measurement made on October 23, 1974. Average
velocities, U, as estimated from:
o_
tK
CD
where,
Q " the river flow, cubic feet/second
A - the cross-sectional area, Square feet
are shown in Table 3 for each, day that a gaging measurement was made.
18
-------
ej2
V
E
L
o
c
z
T
V
F
T
~
9
E
G
0.70
0.Q0
10
jit
. » . .
. i
* » *
I . . .
DISTANCE FROM NORTH BANK -
FIOURE £ PROFILE OF VERTICALLY AVERAGED VELOCITY
AT SNAKE RIVER MILE 9S2.3. I0/23/7 4
-------
Table 3 Average velocity of the Snake River at Highway
27 Bridge (R.M. 652.3), October 22 - 24, 1974.
Date
Average Velocity (feet/second)
10/22/74
10/23/74
10/24/74
0.52
0.53
0.54
Additional measurements of river velocities were made
at other locations in the river. These measurements were made
by following a tracer and observing the travel time. The
tracers used in this experiment were oranges. Velocity data
collected in this manner is shown in Table 4.
Table 4 River velocities in the Milner Reach of the Snake
River estimated from the travel time of oranges,
October 23, 1974.
River Mile Date Velocity (feet/second)
Irrigation return flows from ditches and wasteways were also
measured during the survey. Standard stream gaging tachniques
were used for some of the return flows, while others were simply
estimated. The results of these measurements for all the known
surface return flows between Minidoka and Milner Dams are shown
in Table 5. Locations of the surface drains are given in Figure 3.
659.6-659.2
656 -653.8
642.5-642.3
10/23/74
10/23/74
10/23/74
0.91
0.92
0.37
20
-------
3 Ma 9 ^VU>ouiO«r~ U*. AT VOO-% o t X f2(2-ifc A.n ioO OFTUi?i^ FL
-------
The average Snake River flow at Milner Dam In Table 2 was 3125
c.f.s. and 3184 c.f.s. at Highway 27 Bridge. Assuming that the hydro-
logic regimen of the river had reached a steady-state condition, the
data implies that the groundwater return to the Snake River is negligible
in this segment. The major contribution or return flow, both from surface
and groundwater sources, appears to be in the segment between Minidoka
Dam (R.M. 675.0) and the Highway 27 Bridge (R.M. 652.3). The average
flow at Minidoka Dam, from Table 2 was 434 c.f.s. less than the flow at
the Highway 27 Bridge. Measured sources in this segment include the City
of Rupert STP and surface returns as given in Table 5. These sources
account for 133 c.f.s. Assuming that the ground water return accounts
for the deficit, 321 c.f.s. can be ascribed to groundwater return in
the segment between Minidoka Dam and the Highway 27 Bridge.
22
-------
Table 5 Gaged and estimated (est) irrigation return flows
between Minidoka and Milner Dams, October, 1974.
Approx. Flow
Mile
#
Name
Date
(cfs)
644.9
1
Unknown
10/24
1415
dry
646.7
2
G-20 Lateral
10/24
1405
.5
est
648.0
3
Main Drain
10/22
1415
32.6
10/24
1200
31.7
649.0
4
D-17 Drain
10/22
1430
1
est
650.3
5
Unknown
10/24
1430
.2
est
653.1
6
Unknown
10/24
1200
5
est
654.2
7
Unknown
10/23
1400
4
est
655.0
9
Goose Creek
10/22
1440
12.3
10/24
1300
12.3
655.2
8
D-16 Drain
10/22
1630
8.9
657.3
10
Unknown
10/23
1500
.5
est
659.3
11
Duck Creek
10/22
1520
7.1
659.6
12
Spring Creek
10/22
1540
13.5
660.6
13
Marsh Creek
10/22
1600
22.2
10/24
1340
22.2
662.6
14
Unknown
10/24
1250
1
est
663.1
16
D-5 Drain
10/24
1300
7.2
663.6
15
F Waste Canal
10/24
1330
11.7
663.6
17
Unknown
10/24
1310
.5
est
664.3
18
D-4 Drain
10/23
1520
10
est
664.7
19
Unknown
10/23
1540
dry
666.0
20
Unknown
10/23
1550
.5
est
667.8
21
D-3 Drain
10/23
1620
8.7
671.6
22
Unknown
10/23
1650
.1
est
23
-------
Geometric Characteristics
Important geometric characteristics of the river, which are
needed in the modelling process include the river width, depth
and cross-sectional area. Preliminary results of work done by
the U.S. Geological Survey, as well as soundings made during the
October, 1974 survey were used to estimate these parameters. Table
6 gives the estimates for river width, depth and cross-sectional
area for segments within the Milner reach.
Table 6 Geometric characteristics of the Milner reach of the
Snake River (1Preliminary measurements by the U.S.
Geological Survey.
Survey teams).
"Soundings by EPA water quality
River Mile
674.71
671.21
668.61
665.21
659.81
654.61
649.51
647.32
645.0*
641.52
River Width
Cross-sectional Area
Hydraulic Radius
(feet)
(square feet)
(feet)
460
2,200
4.8
930
2,200
2.9
805
3,030
3.8
820
2,420
3.0
900
3,460
3.8
760
4,940
6.5
790
8,090
10.2
900
8,010
8.9
800
8,640
10.8
1,000
15,000
15.0
Municipal and Industrial Waste Discharge
Discharges of carbonaceous and nitrogenous BOD to Lake Milner
from municipal and industrial sources were measured by EPA-NFIC
(Denver). These data are summarized in Table 7. These summaries
include the average loadings of BOD3 and NH3 from each of the
-------
Table 7 Average observed flow, temperature, BOD5, NH3-N for
point source discharges in the Milner reach of the
Snake River and tributary streams 10/22/74—10/28/74.
Discharger Flow Temp. B0D5 NH3-N
(mgd) ( C) tng/1 lbs/day mg/1 lbs/day
1. Amalgamated Sugar Co.
0.86
36.1
4,600
35,000
20.0
140
2. Rupert STP
2.64
20.1
590
11,000
15.7
300
3. J.R. Simplot
#001
4.67
14.9
100
4,400
6.0
250
#002
0.94
9.1
23
180
0.2
1
4. Hey bum STP
0.20
17.6
230
410
22.3
40
5. Burley STP
1.73
9.8
17
240
7.3
110
6. Bryant's Meats
0.036
14.5
560
200
22.6
6
7. Ore-Ida
#001
4.42
15.6
38
1,550
3*
100
#002-cooling water only 0.78
18.2
4
30
0.4*
2
#002-cooling water and
1 est
16.6
51
430
3.6
30
silt water
#003
0.2*
19.3
5
0.02
21
#004
1.49
13.5
40
510
0.1
1
^Variation betweem samples was more
than two orders of magnituc
25
-------
municipal and industrial sources, as measured by EPA-NFIC (Denver).
Locations of discharges are shown in Figure 4.
Tributary Waste Loadings
Loadings to Lake Milner from surface and ground water return
flow were estimated from the water quality of what were felt to
be representative sources. A summary of the carbonaceous BOD (5-day)
and ammonia nitrogen for those drain and wells which were sampled
is given in Table 8. The Main Drain, which is not a representative
surface return is also given in Table 8, since it does have a
substantial impact on the Milner reach of the Snake River.
Table 8 Carbonaceous BOD (5-day) and ammonia nitrogen for sur-
face and groundwater return flows to the Lake Milner
reach of the Snake River 10/22/74—10/27/74
Surface Returns Number Flow BODc NHo~N
(see table) (cfs) mg/I mg/1
Goose Creek 9 12 0.6 0.01
Flume @ R.M. 657.3 10 0.5 est 1.4 0.59
Marsh Creek 13 22 0.6 0.02
Drain @ R.M. 663.6 17 0.5 est 1.6 0.10
D-4 Drain 18 10 2^5 0.05
Averages 1.3 0.15
Main Drain 3 32 125 0.7
Groundwater Return
Spring below Minidoka Dam 0.4 0.01
Shallow well @ Rupert STP 0.6 0.26
Rupert well K & 3rd 0.6 0.23
Averages 0.5 0.17
26
-------
Amalgamated Sugar
Rupert STP
J. R. Simplot
Heyburn STP
Burley STP
Bryant s Meats
Ore-Ida
Figure ^ Location of major industrial and municipal discharges in the
Lake Milner reach of the Snake River, Idaho (October 1974).
-------
In-stream Water Quality
Water quality parameters measured during the October, 1974 survey
include:
1)
Temperature
2)
Dissolved Oxygen
3)
Conductivity
4)
PH
5)
Biological oxygen demand (2,5,10 and 20-day)
6)
Suspended Solids
7)
Total Kjeldahl nitrogen
8)
Ammonia nitrogen
9)
Organic nitrogen
10)
Nitrite and Nitrate nitrogen
11)
Total phosphorus
12)
Chlorophyll a
The first four parameters in the list above were measured in-situ
with a Hydrolab Surveyor II. The remaining parameters were measured in
the laboratory. The laboratory methods are reported by EPA-NFIC (Denver)
(1974).
The location of receiving water quality stations are shown in Figure
5. These stations were chosen to define the water quality at important :
oundary points. The station at Minidoka Dam (R.M. 675.0) provided a
measure of the water quality of Lake Walcott discharge. The station at
the U.S. 30 Bridge (R.M. 654.0) provided a reference point for conditions
just upstream from those discharges for which NPDES permits were drafted.
This station also provided a means for estimating non-point source contri-
butions between Minidoka Dam and the Burley-Heyburn area. The stations at
R.M. 652.3 (Highway 27 Bridge) and R.M. 647.2 provided a check for water
quality just downstream from major waste discharge sources. The station at
Milner Dam (R.M. 640.0) defined the downstream boundary of the reach and was
also at the point where we expected the most impact upon water quality waste
discharges. The Main Drain constitutes a major waste source, and a station
there was necessary to determine the BOD and ammonia loadings.
28
-------
Figure 5" Location of receiving water quality sampling stations in the Lake
Milner reach of the Snake River, Idaho. EPA - State of Idaho survey
October 1974.
-------
Temperature, dissolved oxygen concentration, pH, conductivity and
nutrient levels were measured twice a day at each station. Samples were
also taken at 1,2,5 and 10 meters, river depth permitting. Chlorophyll
a concentrations were measured at the surface at each of the Snake River
stations. Cross-sections were made at each station, except for the fA
inidoka Dam and Main Drain stations. The purpose of the cross-sections
was to determine the extent of lateral variations in water quality.
Diurnal measurements of temperature, dissolved oxygen concentration, pH
and conductivity were made at two locations. One at Milner Dam from 1810
on 10/22/74 to 910 on 10/23/74, and one at Tom's Marina from 1745 on
10/24/75.
Temperature, dissolved oxygen concentration and dissolved oxygen
saturation levels are plotted as a function of river mile in Figures 6,
7 and 8. Minima, means and maxima for temperature, dissolved oxygen
concentration, BOD (2, 5, 10 and 20-day), ammonia nitrogen and chlorophyll
a, as measured during the survey, are given in Table 9.
Contours showing the lateral variations of temperature and dissolved
oxygen at those stations where measurements of this type were made are
shown in Figures 9 through 20.
In general, the dissolved oxygen levels were at or above the satu-
ration value during the period 10/22/74—10/24/74. Only at Milner Dam
were dissolved oxygen levels observed below 100% saturation for this
period.
30
-------
IS
14.
13
12
11
10
8
e
7
e
9
4-
3
2
1
f
!
Q
I B 6
OO QJ
o o
o o
snake river mile
[CURE & OBSERVED TEMPERATURES IN THE MlLNEi
REACH OF THE SNAKE RIVER
10/22/74- 10-2 4-/74. .
-------
13
12
LI
10
9
8
7
e
s
4-
3
2
1
I
O
O
8
o
» i
• i » » « « i
i i i i i i i «
¦ I
i » i i i
i i i i ¦ i » » i
070 S3
SNAKE RIVER MILE
IQURE T
OBSERVED DISSOLVED OXYGEN IN THE
MILNER REACN OF THE SNAKE RIVER
-------
140
Co
CO
X>
O
3
A
T
U
R
A
T
I
O
N
110
70
E0
B4S 090 OSS 060 BBS 070 873
SNAKE RIVER MILE
FIGURE 6 OBSERVED D.O. SATURATION IN THE
MILNER REACH OF THE SNAKE RIVER
10/22/7 10/24/7 4..
-------
FIGURE 9 Profile of D.O (mg/1) Milner Pam (Snake River Mile MO.O) on 10/22/7^
-------
Figure 10 Profile of Temperature (C ) at Milner Dam (Snake River Mile f^O.O) on 10/22/74
-------
0
10
Depth, feet
20
CO
Cr.
30
Figure
11
Profile of Temperature (C ) at Mfiner Ham (Snake P.iver Mile £**0.0) on 10/23/7^
-------
Of
v
*t Mf
e/-
(s,
t »?re ,
'e <%.
o)
on
to/-
-------
Figure 13 Profile of Temperature (C ) at Milner Dam (Snake River Mile 6*0.0) on 10/2V7^
-------
Figure 14 Profile of Temperature (C°) at Milner Dam (Snake River Mile 6k0.0) on 10/2^/7^
-------
Figure 15 Profile of D.O. (mg/1) at Snake River Mile 652.3 on 10/22/7^
Figure 16 Profile of Temperature (C ) at Snake River Mile 652.3 on 10/22/74
40
-------
Depth, feet 10
20fa
Figure 17 Profile of D.O. (mg/l) at Snake River Mile 6^9.5 on 10/22/74
-------
Figure 19
Profile of D,0. (mg/1) at Snake River Mile 65^.0 on 10/22/7**
42
-------
In none of the observations, were dissolved oxygen levels below
the 90% level. The Snake River In the Milner reach was generally
a greenish brown as a result of the algal growth. Secchi disk readings
ranged from 2.7 to 3.3 feet.
Cross-sectional and vertical variations of temperature were
less than 0.5 C at all stations, including the one at Milner Dam.
The dissolved oxygen varied vertically and laterally as much as 0.8
mg/1 at the shallower stations, and as much as 1.4 mg/1 at Milner
Dam.
Solar heating and photosynthetic oxygen production are both
surface-related phenomena. For this particular time, however, the
data suggests that photosynthesis had a more important role in the
dissolved oxygen budget than solar heating had in the heat budget.
-------
Table 9
Observed minima, mean, and maxima for receiving water sampling stations in
the Milner reach of the Snake River 10/22/74 — 10/24/74.
River Mile
674.9
654.0
652.3
647.2
640.0
Min
Mean
Max
Min
Mean
Max
Min
Mean
Max
Min
Mean
Max
Min
Mean
Max
Temperature (C)
9.0
9.3
10.0
8.5
9.0
9.6
8.0
8.8
9.4
8.0
8.8
9.8
7.5
9.1
10.6
Dissolved Oxygen (mg/1)
9.3
10.3
11.0
10.8
11.7
12.0
11.0
11.5
12.0
10.4
11.1
12.1
9.5
10.4
11.4
60D (2-day) (mg/1)
0.5
0.7
0.9
0.7
1.0
1.2
0.7
0.8
1.1
0.8
1.0
1.4
1.0
1.2
1.3
BOD (5-day) (mg/1)
1.3
1.5
1.7
2.5
2.8
3.5
2.4
2.7
3.1
2.4
3.0
3.6
2.7
3.2
3.9
BOD (10-day) (mg/1)
2.5
2.8
3.5
2.6
3.1
3.6
2.7
3.2
3.9
BOD (15-day) (mg/1)
2.7
5.0
6.4
9.8
9.9
10.0
4.4
6.1
7.0
NH3-N (mg/1)
0.01
0.01
0.03
20.01
20.01
0.02
20.01
0.02
0.04
0.01
0.03
0.06
0.01
0.02
0.0*
Chlorophyll a (mg/1)
30.4
32.2
33.5
32.3
34.8
37.1
28.7
32.8
38.0
30.9
35.9
41.4
46.5
55.4
61.4
-------
(2)
III. THEORETICAL ANALYSIS
The steady-state dissolved oxygen budget for a vertically and
laterally well-mixed stream, in which diffusion and dispersion processes
are neglected can be written:
U d c _ _ Xz ( C-Cs*^ _ ^ v I- ^ c _ ^ C
(j x " " S6HOO
where,
u - the stream velocity, feet/second
C ¦ the dissolved oxygen concentration mg/1
x ¦ the distance along the axis of the stream, positive downstream,
feet
K2 = the reaeration rate constant, days'" *
CSat » the saturation concentration of dissolved oxygen, mg/1
K ¦ the deoxygenation rate, days"^-
L « the carbonaceous biological oxygen demand (BOD), mg/1
- the dissolved oxygen sources, mg/l/second
¦ the dissolved oxygen sinks, mg/l/second
Similarly, the carbonaceous biological oxygen demand (C-BOD), L, is
udL _ + T n (3)
dl* S
-------
The nitrogenous biological oxygen demand (N-BOD), N, is
u
d x
where,
K3 - the deoxygenation rate for nitrogenous BOD, days"*
& - the N-BOD sources, mg/1/second
r\j 0 the N-BOD sinks, mg/l/second
The factor, 86400, in equations (2), (3) and (4) is for the purpose
of making the dimensions homogeneous. This is a result of expressing the
rate constants, Kj and in units of days"* and all other parameters in
units of seconds.
Sources,for dissolved oxygen included in this model are:
1) Surface and groundwater return flow
2) Photosynthetic, P, oxygen production by algae
Sinks, rc , for dissolved oxygen included in this model are:
1) Oxygen demand, S, associated with bottom sediments
2.) Respiration, R, of algae
The sources for BOD are accounted for as boundary conditions, rather
than as internal sources. For example, if a BOD source is discharging at
the rate, Qw, with a concentration, Lw, into a river with a flow, Qr, and
a concentration, Lr, then the effect of the source upon the river, at the
point of discharge, is given by computing a new in-stream concentration, Lr
according to:
-------
L._ _ Q^gLiO + £2* L
+¦ <$?~
y = _ (4)
Equation (3) implies that the waste mixes instantaneously and uniformly
across the river at the time o£ discharge.
No sinks for BOD are included in the analysis.
The solutions to equations (2), (3) and (4) are, respectively:
C = C*., - CC^-CVe^
- T e ^ - e^j - »- e -*%r ) <5>
-i
-------
where
Po
the maximum rate of dissolved oxygen production by algse,
mg/l/day
P
the fraction of the day during which photosynthesis occurs
t
the fraction of the day elapsed from midnight
tO
the time that photosynthesis begins, as a fraction of
the day from midnight
IV. COMPARISON OF THEORETICAL ANALYSIS AND FIELD STUDIES
The steady-state and time models described in Chapter III were
used to simulate the depth and average dissolved oxygen, C-BOD, and
N-BOD in the Lake Milner reach of the Snake River. Mathematically modelling
was done only for that portion of the Milner reach between R.M. 654.0 and
R.M. 640.0. Waste loadings to this reach of the Snake River were well-
defined and the number of receiving water quality stations was adequate.
In addition, this reach contained those industrial and municipal discharges
examined in the initial study (Yearsley 1974).
In order to estimate concentrations with Equations (5), (6) and (7),
the values of certain parameters must be specified, as well as the bound-
ary conditions for river flow and water quality. Those condt&nts required
for the steady-state simulation Include: the deoxygenatlon rate, Ki, the
reaeration rate, K2, the nitrification rate, K3, the oxygen production, »
the oxygen sinks, , the river flow, Q, river width, W, and average
depth, H.
48
-------
The deoxygenation rate, K^, was assumed to be 0.15 days"*, base
This value is the same as that used in permit analysis (Yearsley 1974).
Long term measurement of BOD were made at R.M, 654.0, R.M. 647.2 and
R.M. 640.0 The theoretical BOD, using a rate constant of 0.15 days"*
(base e), is compared to these field observations in Figures 21, 22, and
The reaeration rate, K^, was computed from the equation by
O'Connor and Dobbins (1958):
y U.q u'/l
^ = H9/l (9)
where,
K2 » the reaeration rate, days"** (base e)
U » the average stream velocity, feet/second
H m the stream depth, feet
The nitrification rate, K3, was assumed to be 1.5 days"1 (base e).
The deoxygenation rate, Klt and the intrification rate, K3> were
adjusted for temperature using the following, Fj:
(T-20)
7l - 1.047
(10)
where,
Fj - the temperature factor
T - the stream temperature,°C
49
-------
~
o
n
Q
S
L
CJ1
—>
*
FIGURE. 21 OBSERVED BOS AT SNAKE R . M . 84-0.0.
6FA-NFZC
10/82/7 4-10/23/7
-------
10
0
o
•D
M
Q
/
L.
en
0= Observed
7 -
L=Lo(l-e"nj5t)
¦8"
-
FIGURE ££ OBSERVED BOB AT SNAKE R • M • 04-7.2
EPA-NFIC(DENVER) DATA
12/82/74.-10/25/74.
-------
D
'D
n
o
~
u
CJ1
to
* O = Observed
X -
L=L0(l-e-°-15V
IS3
IS
TIME - DAYS
FIGURE 23 OBSERVED BOD AT SNAKE R. fi. BS4..0.
EPA-NFIC(DENVER> DATA
10/22/74- 10/25/74*
-------
The reaeration rate, K2, was adjusted for temperature using the
factor, F2:
F2 = 1.024
(T-20)
(11)
Initial water quality conditions at the upstream boundary of
the area modeled (R.M. 654.0) were taken as the average of receiving water
quality measurements at the U.S. 30 Bridge. The average carbonaceous BOD
loading, Wp, at this location was estimated from:
_ 5.4 * G? * V-
"D
- e )
(12)
where,
Q « the river flow, cubic feet/second
L5 ¦ the average BOD at R.M. 654.0 (Table 10), mg/1
Kj - the deoxygenation rate, days"* (base e)
Furthermore, the carbonaceous BOD at this point was assumed to consist
of 20,000 lbs/day from the Rupert STP and 72,000 ll?s/day from other sources,
including discharge from Minidoka Dam and surface and groundwater return
flow. The Rupert STP portion of the carbonaceous BOD was treated in the
model as a controllable point source, while the remainder was considered to
be the background BOD in the river. It is the latter value which is given
in Table 11 with the other boundary conditions for flow and water quality.
-------
Table 10 Boundary conditions £or the dissolved oxygen model
of the Milner reach of the Snake River at R.M. 654.0,
October 22 - 24, 1974.
Parameter
Units
Value
Flow
c.f.s.
3.84
Temperature
C
9.0
Dissolved oxygen
mg/1
11.7
Carbonaceous BOD
mg/1
4.2
Nitrogenous BOD
mg/1
0.0
Bulk river velocities, u, were estimated from Equation (1).
The river flow, Q, in any segment below the Highway 27 Bridge (R.M.
654.0) was assumed to be equal to the average measured river flow at the
Highway 27 Bridge (R.M. 652.3) plus the discharge from any surface returns
or point sources discharges downstream from R.M. 652.3 (Tables 5 and 7).
The cross-sectional area for each river segment is given in Table 6.
The only dissolved oxygen source identified in the Milner reach of
the Snake River was the photosynthesis production of dissolved oxygen
by algae. Dissolved oxygen sinks included the respiration of algae and
the benthic oxygen demand associated with organic sediments.
The demand associated with bottom sediments was determined from measure-
ments made by Kreizenbeck (1974). The values used for each reach are shown
in Table 11. These values are similar to those used in the permit analysis.
54
-------
A number of different methods were used to estimate the net oxygen
produced by algae. First of all, previous research (Bain 1967) has indi-
cated that photosynthesis and respiration by algae can be related directly
to cholorphyll a concentrations, in the following manner:
Pmax ¦ 0.24 (13)
R - 0.024 (14)
Table 11 Sediment oxygen demand values used in the simulation
of dissolved oxygen in the Milner reach of the Snake
River. Values are based upon measurements made by
Kreizenbeck (1974).
River Mile Sediment Oxygen Demand
(grams/meter2/day)
654-652
652-650
650-648
648-646
646-644
644-641
641-640
0.89
1.04
1.04
1.85
1.85
1.85
5.32
These rates were considered by the investigator to be applicable
when algae are actively growing, and at water temperatures of approximately
20°C. There was no way of assessing the first criterion with available
data. The water temperature during the October 1974 survey was generally
about 9°C (see Figure 6). Photosynthetic and respiration rates for the
October 1974 survey, as computed from Equations (13) and (14) should,
therefore, be considered as estimates only.
-------
Time series observations of dissolved oxygen at R.M. 640.0 and
R.M. 654.0 were also used to estimate photosynthetic rate. Assuming
that the diurnal variation at the uppermost river boundary (Minidoka
Dam at R.M. 675.0) is small, and that reaeration rates between Minidoka
Dam are high, the diurnal variation in dissolved oxygen is given by
Equation 8.
The photoperiod, p, for the Milner reach during October 22-24,
1974 was assumed to be 0.4 (9.6 hours). The time at which photosynthesis
began, t, was assumed to be 0.3 (0700) Under these conditions, the
values of maximum photosynthesis, Pmax, which appear to best fit the
data (Figures 24 and 25) were found to be 15/mg/l/day at Milner Daur
(R.M. 640) and 10 mg/l/day at Tom's Marina (R.M. 654).
56
-------
T"
o=
Pred!cted
"
x=
Observed
a
>
o
O
O
O
0
0
0
0
0
0
o
o
X
X
o
o
o
o -
X c
<
X
O
O
X
O
X
X
X
X
XXX
X
O
# g O
;
I
G00 1200 1800 2400
TIME - HOURS
FIGURE^ PREDICTED AND OBSERVED DIURNAL
D.o. VARIATIONS AT R.M. S4.0.
EPA SNAKE RIVER SURVEY 10/23/74.
-------
13
C-n
D
I
9
9
O
L
V
&
D
O
X
Y
Q
E
N
M
Q
/
L
12
11
10 -
O = Predicted
^ = Observed
O o
o Ox X
o
O XX qX
o
\
o ox
X O X
O x
° o °
x X X °XoX x O X X
X o x X
o
J I.
o
x°x,
TIME - HOURS
IQURE £5 PREDICTED AND OBSERVED DIURNAL
D.O. VARIATIONS AT R . M . 853.
£PA SNAKE. RIVER SURVEY 10/24-/7-
-------
The average photosynthesis, PAV, can be estimated from:
?ao - * •*?
(15)
where,
P
photoperiod, fraction of a day
Pav
the daily average photosynthetic production of
oxygen, mg/l/day
Pmax «
the maximum photosynthetic production of oxygen,
mg/l/day
The permit analysis (Yearsley 1974) was made with the assumption
that there was no production or respiration by algae. For the waste
loads as given in Table 7, the velocities as estimated by Equations,
initial water quality from Table 10 and sediment oxygen demand from
Table 11, the simulated dissolved oxygen (Equation 5) without photosyn-
thesis or respitation is given by the dashed line in Figure 26. Proper
choice of photosynthesis and respiration rates, keeping all other inputs
constant improved the simulation as shown by the solid line in Figure 26.
The rates of photosynthesis and respiration, as predicted by these
various methods are shown in Table 12. A great deal of significance should
not be given to the fact that the numbers computed in these different ways
agree so well. It does indicate, however, that major sources and sinks for
dissolved oxygen, as well as the value of important, rate constants, have
been identified.
-------
FIOURC 26 PREDICTED D.O. IN THE LAKE MILNER
REACH OF THE SNAKE RIVER 9HOWINO
THE EFFECT OF FHOTOSVNTME9X9.
-------
Table 12 Production and respiration rates in the Lake Milner
reach of the Snake River from: (1) Comparison of
observed and predicted dissolved oxygen in Lake Milner,
(2) Chlorophyll a measured by EPA-NFIC (Denver) during
the October 1974 survey. (3) Diurnal observations of
dissolved oxygen during the October 1974 survey.
Production(P) and Respiration(R)
(mg/l/day)
River Mile (1) (2) (3)
P
R
P
R
P
R
654-652
2.1
0.84
2.1
0.84
2.0
652-650
2.1
0.84
2.0
0.79
-
-
650-648
2.1
0.84
2.2
0.86
-
-
648-646
2.1
0.84
-
-
—
646-644
2.1
0.84
-
-
-
-
644-641
2.1
0.84
-
-
—
-
641-640
3.3
0.84
3.4
1.33
2.6
-
Having established that the model included the major contributions
to the dissolved oxygen—BOD budget, we then estimated the sensitivity
of the model to random error in important parameters. These parameters
included the sediment oxygen demand, S; the net oxygen production by
algae, P-R; the deoxygenation rate, Kj; the reaeration rate, K£; and the
longitudinal river velocity, u. It was assumed that each of these para-
meters consisted of a mean plus a random component. The means were com-
puted as described in Tables 11, 12 and Equations (10), (9) and (1),
respectively. The random component was assumed to be normally distributed
with a standard deviation proportional to the mean value. The proportio-
nality constants were subjectively chosen. The proportionality constants
are shown in Table 13. Twenty-five simulations were performed, with all
inputs the same as previously described, except that the random component
was added to each of the indicated parameters.
61
-------
Table 13 Proportionality constants used to estimate standard
deviation of important parameters. Standard deviation »
x mean value of the parameter.
Mean values of the dissolved oxygen, with a band of one standard
deviation, as determined from these twenty-five simulations are shown
in "Figure 27. The minimum, mean and maximum observed dissolved oxygen is
also shown.
The random error of these parameters does not explain the variability
in the observed data, even though it does estimate the mean of the observed
values to within 0.2 mg/1. However, the variability can be accounted for
by including the diurnal fluctuations in dissolved oxygen due to photo-
synthesis. This is shown in Figure 28 where the diurnal variations
estimated with Equation 8 have been added to the random variations in the
steady-state simulations, as predicted above.
The worBt error in predicting the mean of the steady-state dissolved
oxygen concentration occurs at the downstream boundary, R.M. 640.0. The
difference between the observed mean and one standard deviation below the
predicted mean is 0.4 mg/1.
Parameter
Proportionality constant
Sediment demand, S
Net algal oxygen production, P-R
Deoxygenation rate, Kj
Reaeration rate, K2
River velocity, U
0.2
0.2
0.5
0.5
0.1
62
-------
05
Co
D
£
S
S
o
L
v
£
D
O
X
V
G
E
N
M
Q
/
L
13
12
11
:tC
10
i r
Mean - Predicted
T ¦"
+
-Mean - Observed
6
B4S
f>d j|
na n
SNAKE RIVER MILE
FIGURE 27 PREDICTED AND OBSERVED D.O. IN THE
LAKE MILNER REACH OF THE SNAKE RIVER.
SURVEY ON 10/22/V10/24>/74-.
-------
13
C75
r»
i
s
3
O
L
V
E
D
O
X
V
8
E
N
M
G
/
L
+ 6
IZ -
\
11
O
10 -
..4>~
O
?an Observed
Mean Predicted
B42
r>rt ii
CrHr
rtii ry
646
SNAKE. RIVER MILE
FIGURE E8 PREDICTED AND OBSERVED D.O. IN THE
LAKE MILNER REACH OF THE SNAKE RIVER
-------
The mean contribution of each dissolved oxygen sink and percent of
total contribution is given in Table 14. Also shown in Table 14 is the
net oxygen produced by the difference between photosynthesis and respira-
tion of algae. Discharge from point sources account for 38% of the total
demand on the oxygen resource. In-stream or non-point source BOD accounts
for 28% of the demand, deaeration 23% and sediment demand 11%. Net oxygen
produced by algal activity during this period was 43% of the total oxygen
demand. While the contribution by algae at this time appears to be a
benefit, the deposition of these same algae may be responsible for the
sediment oxygen demand in Lake Milner and downstream reservoirs.
65
-------
Table 14 Contributions of major dissolved oxygen sources and
sinks to the total oxygen budget of the portion of
the Milner reach of the Snake River from R.M. 654.0 to
R.M. 640. Based upon October, 1974 water quality
and hydrologic conditions.
Dissolved oxygen sink Total oxygen demand
(mg/1) (% of total demand)
Surface transfer
0.59
23
In-stream (background) BOD*
0.72
28
Sediment demand
0.28
11
Point source BOD*
2.57
38
Dissolved oxygen source Total oxygen demand
(mg/1) (% of total demand)
Net algal oxygen production 1.13 44
66
-------
V. PREDICTION OF DISSOLVED OXYGEN AT LOW FLOW
The results of comparing the data observed during October 1974
with mean values predicted by the steady-state dissolved oxygen model,
indicate that the major oxygen sources and sinks have been identified
in the Milner reach of the Snake River. For the river flow waste dis-
charge rates of October 1974, State of Idaho water quality standards for
dissolved oxygen were not violated. However, as the river flow decreases,
the likelihood of standards violations will increase, as long as the
waste discharge rate remains constant. Once the validity of the model-
ling process has been established, the model should be used to predict
water quality which occurs at lower flow regimes.
Initially, the effect of lower flows upon the Snake River was
examined for those waste discharges and boundary conditions as observed
in October 1974. Random variations, associated with error in the same
parameters described previously, were also determined as a function of
flow. The mean and standard deviation of the simulated dissolved oxygen
are given in Table 15 for various flows between 3184 c.f.s. and 750 c.f.s.
These results indicate that the error associated with the simula-
tion Increases as the flow decreases.
67
-------
Table 15 Variation of mean minimum simulated dissolved oxygen
and maximum standard deviation with flow in the
Mllner reach of the Snake River. Base upon October
1974 water quality conditions.
River Flow Mean Minimum Maximum Standard
(c.f.s.) dissolved oxygen deviation
(mg/1)
(mg/1)
3184
10.24
0.23
2000
9.04
0.36
1000
4.55
0.90
750
1.11
1.09
Various waste treatment strategies were also examined. The
strategies included:
1) Eliminating the discharge from Simplot and Ore-Ida only
2) Reducing the discharge from all controllable point sources
by 50%
3) Eliminating the discharge from all controllable sources
The loading levels from the various dischargers under these condi-
tions are:
Table 16
Organic waste loading levels for point sources in
Mllner reach of the Snake River, assuming various
waste treatment strategies: 1) No discharge from
Simplot and Ore-Ida 2) Reduce all point sources by
50% 3) Eliminate all point source discharges
Level 1
bod5
(lbs/day)
Level 2
B0D5
(lbs/day)
Level 3
B0D5
(lbs/day)
Rupert STP
11,000
5,500
0
J.R. Simplot
0
2,200
0
Heyburn STP
0
205
0
Burley STP
Bryant's Meats
0
120
0
200
100
0
Ore-Ida
0
1,260
0
Main Drain
41,000
20,500
0
68
-------
The effect of these strategies upon minimum dissolved oxygen in
the Milner reach of the Snake River is shown in Figure 29, as a function
of flow. Table 17 gives an estimate of the dissolved oxygen increase,
D.O., resulting from each of the strategies. The base waste discharge
conditions is taken as October 1974. The dashed line in Figure 29
corresponds to the no discharge condition analyzed in the permit analysis
(see Table 6 in Yearsley 1974). Comparison of the no discharge conditions
using October 1974 results with that of the permit analysis indicate
that water quality is better at higher flows, with the October 1974
results, but poorer at the lower flows. The major differences between
the two conditions are associated with in-stream, or background, BOD
and the assumptions regarding groundwater return flow. The in-stream
ultimate BOD was found to be 4.2 mg/1 from the October 1974, but was
assumed to be 1.5 mg/1 in the permit analysis. The groundwater return
flow in the segment of the Snake River from R.M. 654—640 was found to
be negligible from the results of the hydrologic studies. The permit
analysis assumed that the groundwater return flow was 15 c.f.s. per mile
or 210 c.f.s. for the 14 mile segment. The dilution effect of groundwater
return flow, in addition to the low background BOD, would bias the permit
analysis in favor of higher water quality at the low flows. At the high
flows, the results of the October 1974 study would be biased in the same
direction due to higher initial dissolved oxygen and. net production of
oxygen by algae. The latter was not considered in the permit analysis.
69
-------
le
Ll
10
a
s
7
9
S
4k
3
2
1
QURC as DISSOLVED OXYOCN AS A FUNCTION OF
FLOW FOR VARIOUS LOADINQ LCVCLS.
MtLNCR RS. AOH OF TM6 SNAKE RIVER.
-------
Table 17 Improvement in dissolved oxygen, D.O., at various flows,
resulting from various treatment strategies.
River Flow
(c.f.s.)
Improvement in dissolved oxygen, D
0., (mg/1)
Level 1
Level 2
Level 3
3,184
2,000
1,000
750
0.15
0.27
0.94
1.28
0.28
0.66
2.16
3.23
0.56
1.30
4.25
6.35
In order to use the model as a predictive tool, it is necessary to
examine the range of validity of the assumptions. The concepts upon
which the model described in Chapter IV are based have been verified at
low flows in other river systems, e.g., O'Connor and Di Toro (1970). It
has not been done in the Milner reach of the Snake River. Important
assumptions or principles developed in Chapter IV, which may be affected
by reducing the flow include:
1) Reaeration rate, K2, is a function of river temperature,
velocity and depth Equations (9) and (11) only.
2) Deoxygenation rate, Kj, is a function of temperature only.
3) Dispersion effects are negligible.
4) Thermal stratification is unimportant.
5) Sediment demand is a function of temperature only.
The reaeration rate, K2, is an impoundment river, decreases rapidly
as the flow decreases. In the last segment of Lake Milner (R.M. 641.0—
640.0), for the October temperature of 9.0°C and for a constant average
depth of 15.0 feet, the reaeration rate, K2, at various river flows, as
71
-------
computed from Equation (9) is given in Table 18.
Table 18 Reaeration rates in Lake Milner (R.M. 641.0—640.0)
as a function of river flow. Average depth - 15.0
feet and river temperature 9.0°C.
River Flow
(c.f.s.)
(feet/second)
(days"*)
5,000
0.33
0.10
3,184
0.21
0.08
2,000
0.13
0.06
1,000
0.07
0.04
750
0.03
0.03
According to Bennett and Rathburn (1972), the range of depth for
which Equation (9) has been verified varies from 4.0 to 24.2 feet. The
range of velocities varies from 0.19 feet/second to 4.20 feet/second.
The minimum reaeration rate was 0.32 days"* (base e). Fortescue and
Pearson (1967) reported successful use of a similar formulation for
predicting reaeration rates as lows as 0.04 days-*. Hydroscience (1971)
suggests that for water depths greater than approximately 10 feet the
reaeration rate is a function of depth, H, only and suggests that the
relationship is:
s.
L
K
(16)
Busch (1972) suggests that the relationship should be:
-------
^ --
cO.*>
(17)
The lower value is conservative in favor of improved water quality.
Values for reaeration given in Table 17 are within the range of values
predicted from Equation (17). A possible source for reaeration in the
Milner reach, other than turbulent transfer due to river velocity, is
that of wind-mixing. The research available is not conclusive regarding
the quantitative nature of this source. Furthermore, while wind speeds
observed during the October 1974 survey averaged between 10 and 15 miles
per hour for extended periods during the day on Lake Milner, there were
equally long periods at night when calm or semi-calm conditions prevailed.
The deoxygenation rate, Kj_, is generally assumed to be a function
of temperature only. Research indicates, however, that the turbulence
associated with high stream velocities may increase deoxygenation rates.
Hydroscience (1971) reports experimental results showing that the deoxyge-
nation rate, Kj, varies from 0.1 to -.5 days—1 (base e) for rivers with
depths of 10 feet or greater. The values obtained from the laboratory
should provide a lower bound on the rate since turbulent mixing is at a
minimum.
Deposition rates of suspended organic material may increase as the
river flow decreases. This results in an apparent increase in the con-
sumption of BOD, but is not reflected in the dissolved oxygen budget as
in-stream biological oxygen demand. This increased rate does contribute
73
-------
to the sediment oxygen demand so that while the reaction time may be
longer, there will still be the same total oxygen demand.
The effects of longitudinal dispersion were assumed to be negligible
in the development of Equations (2), (3) and (4). As shown by Thomann
(1973), the maximum dissolved oxygen deficit is similar in dispersive
and non-dispersive systems, only when the frequency associated with the
discharge is small. That is, when there is no time variation of the
discharge rate. The difference between non-dispersive systems increases
as the discharge frequency increases. The dispersive nature of the
system is measured by a dimensionless number, P:
P - 0.1. Figure 8 in Thomann (1973) shows that the maximum dissolved
oxygen deficit associated with a period of about six (6) days would be
attenuated 60% if dispersion were included. The amount of attenuation
increases as the flow decreases. There are not adequate time series of
all the discharge data during the October, 1974 survey. However, if the
mean values for waste discharge, given in Table 7 are accurate, then the
simulation associated with those values will be essentially the same
whether dispersion is included or not. This is true because there is
(18)
Assuming that the dispersion coefficient E ¦ 300 feet^/day,
U - 0.2 feet/second and Kj - 0.1 days"1 (1,16 x 10~5 seconds"1), then
74
-------
very little attenuation of the steady-state or long term (i.e., periods
greater than approximately 30 days) discharge, when dispersion is included.
Thermal stratification also reduces the vertical mixing and prevents
atmospheric oxygen from diffusing to the bottom of the river. In many
reservoirs, this results in anoxic conditions in the bottom. WRE (1969)
has provided a crude criterion for estimating when a river may be thermally
stratified. This criterion is based on a Rayleigh-like number,
= t#I
where,
L
-
reservoir length
D
-
reservoir depth
Q
m
Volume discharge rate
V
-
reservoir volume
t>
m
water density
p
-
Average density gradient
g
m
gravitational constant
For certain simplifying assumptions regarding the density profile,
Equation (19) can be reduced to:
75
-------
W " 3ZO ^
^ (20)
Assuming, L ¦ 73920 feet (14 miles),
D * 10 feet
V ¦» 1 x 10 cubic feet (24,000 acre feet), then
»"*" » 1.2 at 500 c. f. s. and
» 7.5 at 3184 c.f.s.
As a rule of thumb, WRE (1969) suggests that for values of PF much
less than I W- (0.318), the water body will be thermally stratified; for
values approximately equal to , weakly stratified; and for values
much greater than t, the water body will be well-mixed. The results
for Lake Milner indicate that thermal stratification will not be signifi-
cant for the flow range 500—3184 c.f.s.
Accumulation of organic sludge on the river bottom increases as the
river velocity decreases. Krenkel et al (1969) describe the work of Velz
(1958) in which he showed that acute BOD effects from bottom deposits may
be expected when the river or reservoir velocity is reduced below 0.6 feet/
second. Krenkel at al (1969) also indicate that the oxygen demand associated
with these deposits is significantly affected by ion change through the
soil, microbiological action and leaching of organic and mineral substances.
In summary, there are certain dangers in extending the mathematical
model to lower flow conditions. In the case of deoxygenation rate, Kp
7K
-------
and the reaeratlon rate, K2, the justification for doing so is based
upon the success of research in other river systems, or maintaining a
conservative waste loading. Analysis of important dimensionless para-
meters indicates that dispersion effects and thermal stratification will
not seriously affect the results for flows ranging between 500 and 3184
c.f.s. The percent contribution from sediment demand decreases as the
flow decreases, so that the order of the error would be 10% or less,
assuming that the sediment demand was in error by 100%.
VI. FINDINGS
The results of the October 1974 field study and mathematical model-
ling showed:
1. Dissolved oxygen standards were not violated in the Snake
River during the period October 22—24, 1974.
2. Groundwater return flow was negligible between R.M. 654.0 and
640.0 of the Snake River during the October, 1974 survey.
3. Oxygen production by algae contributed significantly to the
dissolved oxygen budget of Lake Milner. The total contribu-
tion amounted to an estimated 1.1 mg/1.
4. Major sources of oxygen demand during the October 22-24, 1974
survey included organic wastes from municipal and industrial
wastes sources, surface transfer of oxygen, in-stream (back-
ground) biological oxygen demand, sediment oxygen demand and
respiration of algae.
5. The steady-state mathematical model for dissolved oxygen simu-
lated the mean observed dissolved oxygen with a maximum dif-
ference of 0.2 mg/1. The estimated error associated with
7
-------
the simulation had a maximum standard deviation of 0.24 mg/1.
Discharges from controllable point sources contributed 38%
of the total oxygen demand In the Mllner reach of the Snake
River. Maximum Impact occurred in the last six (6) miles
(R.M. 646 to 640) of the river.
Extrapolation of the mathematical model to low flows indicates
that violations of the State of Idaho water quality criterion
for dissolved oxygen would occur at river flows of 2,000 c.f.s.
at waste loadings equal to those observe during October, 1974.
Estimated standard deviation of the simulations increases as
the river flow decreases. At 3184 c.f.s. the estimated stan-
dard deviation was 0.24 mg/1 and 0.90 mg/1 at 750 c.f.s. for
October, 1974 discharge rates and water quality boundary
conditions.
The permit analysis (Yearsley 1974) predicted better water
quality at low flows than the extrapolation of results from
the October, 1974 survey. The discrepancy is due to the
lower in-stream BOD and higher groundwater return flow assumed
in the permit analysis.
-------
CONCLUSIONS
Dissolved oxygen levels In the Milner reach are influenced
strongly by river hydrology and point source discharge of
organic wastes. Greatest impact from the point source dis-
charges is felt in the last six miles (R.M. 646—640) of
the segment. Non-controllable factors, including sediment
demand, surface transfer and non-point source BOD, are also
important.
Steady-state mathematical modelling methods provide reliable
means for estimating the impact of waste discharges, sediment
oxygen demand, photosynthesis and respiration and surface
transfer, when supported by an adequate field study program.
The extrapolation of model results to low flow conditions
should be done with care. Particularly close attention should
be given to the determination of reaeration rates and sedimenta-
tion rates at low flows.
-------
80
-------
BIBLIOGRAPHY
Bain, R.C. 1967. Predicting Diurnal Variations in Dissolved Oxygen
Caused by Algae in Estuarine Waters. National Symposium on
Estuarine Pollution. ASCE and Stanford University. August 1967;
250-279.
Bennett, J.P., and Rathbun, R.E. 1972. Reaeration in Open-Channel Flow.
U.S. Geological Survey Professional Paper 737. U.S. Government
Printing Office. Washington, D.C. 1972; 75 p.
Busch, A.W. 1972. A Five-Minute Solution for Stream Assimilative
Capacity. Journal of Water Pollution Contract Federation, 44(7):
1453-1456.
Fortescue, G.E. and Pearson, J.R.A. 1967. On Gas Absorption Into a
Turbulent Liquid. Chem. Eng. Sci., 22: 1163-1176.
Krenkel, P.A., Thackston, E.L. and Parker, F.L. 1969. Impoundment and
Temperature Effect on Waste Assimilation. Journal of Sanitary
Engineering Div. ASCE. SAI: 37-64.
Kreizenbeck, R.A. 1974. Milner Reservoir Benthic Oxygen Demand Study.
Region X, EPA. August 1974.
O'Connor, D.J., and Di Toro, D.M. 1970. Photosynthesis and Oxygen Balance
in Streams. Journal of Sanitary Engineering Div. ASCE. SA2: 547-571.
O'Connor, D.J., and Dobbin, W.E. 1958. The Mechanism of Reaeration in
Natural Streams. ASCE Trans. 123: 641-666.
Thomann, R.V. 1973. Effect of Longitudinal Dispersion on Dynamic Water
Quality Response of Streams and Rivers. Water Resources Research,
9(2): 355-366.
Velz, C.J. 1958. Significance of Organic Sludge .Deposits in Oxygen Rela-
tions in Streams. Technical Report W58-2. Taft Engineering Center.
Yearsley, J.R. 1974 Dissolved Oxygen Analysis for Lake Milner Permit.
Report to Idaho Operations Office, Region X, EPA.
1973. Rules and Regulations for the Establishment of Standards of Water
Quality and for Wastewater Treatment Requirements for Waters of the
State of Idaho. Idaho Board of Environmental and Community Services,
State of Idaho, Boise.
1971. Simplified Mathematical Modeling of Water Quality. Prepared by
Hydrosclence, Inc. for the Environmental Protection Agency, Washington,
D.C. 127 p.
1969. Mathematical Models for the Prediction of Thermal Energy Changes
in Impoundments. Prepared by Water Resources Engineers, Inc., Walnut
Creek, California.
81
-------
82
------- |