WATER POLLUTION CONTROL RESEARCH SERIES
16010 DMG 12/71
 FLUSHING OF SMALL SHALLOW LAKES
U.S. ENVIRONMENTAL PROTECTION AGENCY

-------
          WATER POLLUTION CONTROL RESEARCH SERIES
The Water Pollution Control Research Series describes the
results and progress in the control and abatement-of pollution
in our Nation's waters.  They provide a central source of
information on the research, development and demonstration
activities in the Environmental Protection Agency,  through
inhouse research and grants and contracts with Federal, State,
and local agencies, research institutions, and industrial
organizations.

Inquiries pertaining to Water Pollution Control Research
Reports should be directed to the Chief, Publications Branch
(Water), Research Information Division, R&M,  Environmental
Protection Agency, Washington, D,C. 20^60.

-------
                FLUSHING  OF  SMALL SHALLOW LAKES
                               by
                        Claud C. Lomax
                        John F. Orsborn
           The R. L. Albrook Hydraulic Laboratory
          College of Engineering Research Division
                 Washington State  University
                  Pullman,  Washington 99163
                            for the
              Office  of Research and Monitoring

               ENVIRONMENTAL PROTECTION AGENCY
                      Grant #16010  DMG
                         December  1971
For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, B.C., 20402 - Price 80 cents

-------
          EPA Review Notice

This report has been reviewed by the
Environmental Protection Agency and
approved for publication.  Approval
does  not  signify that the contents
necessarily reflect the view and pol-
icies of the Environmental Protection
Agency, nor does mention of trade
names or commercial products consti-
tute endorsement or recommendation
for use.
                  ii

-------
                            ABSTRACT
Restoration of quality to polluted lakes by inflows of clean water with
simultaneous outflow of polluted water was investigated.  Elliptical
basins were used in the laboratory to simulate shallow lakes.  The in-
vestigation determined the influence of selected geometric and flow
parameters on the flushing efficiency.

The theoretical analyses are combined with the experimental results to
obtain equations for predicting the flushing curves.  Application of
these equations will give the potential flushing efficiency of a pro-
posed flushing scheme.

The width of the basin perpendicular to the axis of flow is an impor-
tant parameter.  In general, narrower basins have better flushing
action and less erratic flow patterns than wider basins.  The pollution
remaining in a basin after any interval of flushing is primarily depend-
ent on the time of the flushing interval divided by the detention time
of the basin.

This report was submitted in fulfillment of Grant Number 16010DMG, under
the partial sponsorship of the Water Quality Office, Environmental Pro-
tection Agency.
                                  iii

-------
                          CONTENTS
Conclusions                                            1




Recommendations                                        3




Introduction                                           5




Experimental Apparatus                                13




Test Procedure                                        15




Data Analysis                                         17




Test Results                                          21




Discussion of Test Results                            23




Acknowledgments                                       29




References                                            31




Appendices                                            33

-------
                        FIGURES
                                                       Page
 1  NOMENCLATURE AND FLOW ZONE SKETCHES                  6

 2  RESIDUAL CONCENTRATION AND AVERAGE RATE  OF
    CONCENTRATION REMOVAL                                7

 3  GENERALIZED BOUNDARY CONDITIONS FOR  FLUSHING
    EFFICIENCY                                           7

 4  SMOOTHED OUTFLOW CONCENTGRAPH                        8

 5  PREDICTED VALUES, CR/CQ                             10

 6  FLUSHING PARAMETER, td/t   , AS A FUNCTION OF
    ASPECT RATIO, Y/X, AND INLET WIDTH,  Wj_              11

 7  SCHEMATIC OF TEST APPARATUS                         13

 8  TYPICAL OUTFLOW HYDROGRAPH, C/CQ VERSUS  t/td        17

 9  CONCENTRATION VERSUS TIME                           18

10  FLUSHING CURVES FOR 10-FOOT BY 4-FOOT BASIN        21

11  FLUSHING CURVES FOR 10-FOOT BY 6-FOOT BASIN        21

12  FLUSHING CURVES FOR 10-FOOT BY 10-FOOT BASIN        22

13  FLUSHING CURVES FOR 6-FOOT BY 10-FOOT BASIN        22

14  IDENTICAL RUN COMPARISON. RUNS 60 AND 61           24

15  EXAMPLE OF WELL-DIFFUSED JET                        26

16  COMPARISON OF AVERAGE FLUSHING CURVES               27
                          VI

-------
                        TABLES








NO.                                                 Pa?e



 1  Predicted Values, CR/CQ , from Eqs. 9 and 14     10



 2  Summary of Test Conditions                       15



 3  Data and Computed Values                         19



 4  Data Summary for 10-Foot by 4-Foot Basin         36



 5  Data Summary for 10-Foot by 6-Foot Basin         37



 6  Data Summary for 10-Foot by 10-Foot Basin        38



 7  Data Summary for 6-Foot by 10-Foot Basin         39
                           Vll

-------
                           CONCLUSIONS


1.  The most important parameter in determining the shape of the flush-
ing curve, CR/C0 versus t/td,  is the variable behavior of the inflow
jet.

2.  Inlet velocity and depth of flow have no separable effects on the
flushing curves.

3.  For a predictable jet behavior, the residual concentrations are
primarily a function of the time ratio, t/td, and the basin shape (as-
pect ratio, Y/X).

4.  The theoretical flushing efficiency as measured by ts/td is influ-
enced directly by the inlet width, V^, inversely by the aspect ratio,
Y/X (Eq. 2),

5.  Although the results presented have been derived to apply to flush-
ing pollutants from a lake, the same information can be used to evalu-
ate the degree of increased pollution versus time where a polluted
stream (inflow) is permitted to enter a lake.

6.  Perhaps the most significant conclusion reached in this study is
that a jet entering a shallow basin with a solid boundary (the bottom)
and an air-water interface (the free surface), does not follow predicted
patterns  for a jet discharging  into an infinite medium (1). Considerable
work remains to be done to describe the behavior of such a jet includ-
ing the influences of density differentials.

-------
                           RECOMMENDATIONS
Although  this program was conducted under controlled  laboratory condi-
tions, valuable  information has been developed for planning the enhance-
ment of water quality in polluted lakes situated within reasonable dis-
tances of flushing water sources.  Factors such as wind and tidal action
were not considered in this program, but their relative effects can be
relatively estimated.  Wind action would tend to promote more complete
mixing in shallow lakes, and tidal action would tend  to oppose the
flushing action.  Temperature differentials between the inflow and the
lake were not investigated.

Therefore, field pilot studies under natural conditions are an obvious
recommendation.  Many natural situations ,of various sizes already exist
which could be monitored.  Care should be taken to avoid systems which
are so large that automatic, continuous monitoring becomes too expensive.

A system analysis study should be conducted,  using the results of this
study combined with natural physical and climatological factors, to de-
velop an optimum pilot investigation plan.  This analysis should be con-
ducted prior to the initiation of any field experiments.  Based on the
hydraulic and quality parameters of the flow system,  the system analysis
would be used to determine required sampling location and density, data
types and probable scaling factors to be evaluated as a result of dif-
ferent system sizes.

The results of these pilot tests would provide a significant contribu-
tion to water resources system design information for the enhancement of
quality in polluted lakes as guided by results of this laboratory study.

-------
                            INTRODUCTION
 Flushing of a lake means reducing the pollution by clean inflow with an
 equivalent outflow of polluted water.  In the process the clean water
 both displaces and mixes with the polluted water.

 Many factors influence the effectiveness of the cleansing stream.  This
 study was conducted to investigate those parameters believed to be most
 important and manageable under laboratory conditions.  The parameters
 selected for study were:  1) the inlet velocity, 2) the inlet width,
 3) the depth, and 4)  the basin shape.  Testing was conducted on two
 depths, two inlet widths, three inlet velocities, and four elliptical
 basins.  The primary purpose of the project was to evaluate the various
 parameters to determine their influences on flushing efficiency, and
 to develop prediction equations based on the geometric and flow charac-
 teristics of the systems tested.

 Analyses  were completed to:   1)  develop a test program,  2)  analyze the
 system for comparison with experimental results,  and 3)  develop pre-
 diction equations which incorporate  the analytical and experimental re-
 sults  of  the study.

 A  dimensional analysis of the various geometric,  flow,  and  fluid parame-
 ters was  completed to isolate the more important variables  as  a guide
 for establishing the  test program.   The average rate of  removal of pol-
 lutant, Qr,  was  of primary interest  and is  a varying percent of the in-
 flow rate, Q.|_, as a function of time.   The  following set of dimension-
 less ratios  was  developed:
                        V ' Y ' X ' d

Terms are defined in Figure 1 and in  the Nomenclature.  Fluid properties
such as density and viscosity were considered constant  throughout  the
system.

Grouping the first four dimensionless terms  in Eq. 1 yields an expression
for the average rate of soluble pollutant removal during passage of some
time, t, such that

                                      W?X
                                                                    (2)

K' is a dimensionless coefficient containing  4/TT  for elliptical basins
and is a function of not only the geometric and flow parameters,  but
also time.  K1 was found to be independent of the jet momentum force,  T?±,
and the Reynolds number.

For the general case of flushing lakes for various geometric conditions,
Qr should approach Qt as Wt approaches Y.  Also,  as X becomes very

-------
 large with respect to W^
 and Y, Qr should approach
 Qi-

 The coefficient K' is used
 to relate Qr to the actual
 flushing curve at any rela-
 tive flushing time, t/td,
 in Figure 2 for a particular
 lake geometry.  An important
 time ratio for evaluating
 flushing efficiency is ts/td,
 or the relative time at
 which the real flushing
 curve departs from the ideal
 flushing curve.

 Noting that the volume of the
 lake is V = 0.25irXYd and hold-
 ing d  constant in Eq. 2,  it
 would be expected that as W^
 approaches Y,  the average
 rate of pollutant removal Qj.
 approaches Q^.   At the other
 extreme,  W^ approaches zero,
 Qi>  Qr and the flushing effi-
 ciency should all tend towards
 zero.   These  general  boundary
 conditions are shown  graphi-
 cally  in Figure 3.

 It was  found  in these tests
 that for small  values of
 and  low inflow discharges,
 the  flushing  action can move
 through the basin as  a wave
 front  and be  quite  efficient.
As  long as Q^  is  finite,  the
 actual  flushing  curve  will
 follow  the  ideal  flushing
 curve  for a time,  ts/td,  if
external  mixing  forces  such
as wind  are neglected.
             10 feet maximum
(a) Plan View of a Longitudinal Basin-
        Hydraulically Long Lake
(b) Elevation, Common to (a) and (c)
   Developing  Developed
   Jet Zone-,  £L2°™  ,-SinkZone
(c) Plan View of a Transverse  Basin-
       Hydraulically Short Lake

Fig.  1.   Nomenclature and Flow
          Zone  Sketches
A theoretical analysis of  the  flushing was  developed to relate concen-
tration to time.  The displacement  phase  of pollutant removal continues
for a time interval, tg/td, with  the  outflow concentration equal to
C/C0 = 1.0.  The remaining outflow  concentration  is  assumed to follow
some form of an exponential expression.   Figure 4 shows the assumed
curve.  The area under the curves to  any  time, t/td,  is proportional to
the pollutant removed from the basin.

-------
                      Qr = QO between
                   1.0              2.0             3.0
                t/td,  Relative  Length of Flushing Time

      Fig. 2.   Residual Concentration  and Average Rate
                 of Concentration Removal
                                                  4.0
     1.0
    0.8
« ^ 0.6
» 5 O-4
o ^
    0.2
       •Wj/B
          /
          /
        _/

        /
Y/X-
                                   General parameter being a
                                   function of lake shape and
                                   equal to K1 at t/tj = 1.0.
                 ASSUMED:
                 Volume of  lake constant.
                 Depth of water constant
                 in lake, inlet and outlet.
                                            l
       0                             1.0
          •Relative Inlet Width, Wj/Y	^
             -Aspect Ratio, Y/X, for Small, Constant Inlet Width, Wj-

      Fig. 3.   Generalized Boundary Conditions
                 for Flushing Efficiency
                                                  2.0

-------
                             Ideal Flushing Curve
                            Smoothed  Outflow Concentgraph
The initial quantity of dye  in  the  basin  equals  1.0 KV,  where K is a dye
removal coefficient.  The area  under  the  curve from t = 0 to t = °°  must
equal KV.  Integrating the equations  to find  the area and determine "a",
                      f Is- +  f
                      L d   »*-
                             •-<
;a(t-ts)/td d(_t_;
                                                           (3)
On the right side of Eq. 3,  K  is multiplied by td to obtain the correct
units for new time base.  Within the  brackets,  the first term,  tg/tj*
accounts for area under  the horizontal  part  of  the curve;  that is,  where
C/CO = 1.0.  The^area under the exponential  decay part of  the curve is
determined by Jt /t,.  Eliminating K  and  dividing by Q.^ to get
V.
Qi
                                  J_Id   a(t-ts)/td
                                   a  t
                                (4)
Divide by td.  The lower limit gives one  and  the  upper  limit gives zero
for "e" raised to its power.  Therefore,
                        1  =  ~d+~
Solve for
                        a  =
                                   -1
                              1 -
                                                           (5)


                                                           (6)

-------
   Residual in Basin  =  Total at the start minus that removed.
                                                   'ts/td

 Substitute limits and cancel K.  Divide by Q^ to get td = V/Q.,  and
 divide by tj.                                                1
                                                                     (8)


 Now substitute  a = -l/[l-(tg/td)]  and simplify.
                co
This equation applies for  t/td > tg/td.  To obtain  CR/CQ  when  t/td <
ts/t, use
  sd

                            CR
 The most significant  time  ratio  in  the  right  side of Eq. 9  is  t  /tH.
 Expressing  this  ratio in  terms of the geometric  and flow parameters, and
 inverting to  give whole numbers  leads to
The  experimental value of  tg  is equal to the X distance divided by the
average velocity of the fastest flow-through element.  If

                            v  =  bVi                              (12)

then for elliptical basins, Eq. 11 becomes
Eq. 13, when inverted and simplified, gives
                           tg     4bWt
                           ~d  =  ~W~       '

In this form, b can be selected to fit any condition.
It is feasible to assume a value of b for Eq.  14 and compute .t /tH and
the values of CR/CO can then be computed for values of t/t, using Eq.  9
First the limits will be examined.  If ts/td = 1.0, then the polluted
water is all displaced by the clean inflow and the ideal flushing curve

-------
is followed.   Obviously,  tg/t^ cannot exceed 1.0.  Assuming  instantane-
ous mixing of  the  inflow with the basin water,  ts/tj =0, and the expo-
nential curve  starts at t/t^ = 0.   The assumption of an exponential  decay
curve implies  perfect  mixing.  This curve and the ideal flushing curve
are plotted for Figure 5.   It is  apparent that any value of  tg/t^  be-
tween zero and one will give a curve lying between these two curves.
    100
 _   80
 2 S
     60
 4- C
 a o

 & 1 40
 O m
O  O
     20
      a
J Ideal Mixing
  t,/td = 0
         Ideal Flushing
        - t,/td - 1.0
                       1.0              2.0              3.0
                        t/tj,  Relative Length of Flushing Time
                       Fig.  5.   Predicted Values, CR/C0
                                     4.0
Table 1 gives the values  calculated for predicting CR/CO for various
t/tj and ts/td values.  Although  Table 1 gives valid limits for the
assumed conditions, there  are  other possibilities which are amenable  to
theoretical analysis.  Conditions may change with time so flushing may

                               Table  1.
              Predicted Values, CR/CO, from Eqs.  9 and  14
                                      t/td
£d
(1)
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0
(2)
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1
(3)
0.368
0.331
0.294
0.258
0.221
0.184
0.147
0.110
0.074
0.037
0.000
2
W
0.135
0.109
0.084
0.062
0.042
0.025
0.012
0.004
0.001
0.000
0.000
3
(5)
0.050
0.036
0.024
0.015
0.008
0.003
0.001
0.000
0.000
0.000
0.000
4
(6)
0.018
0.012
0.007
0.004
0.002
0.001
0.000
0.000
0.000
0.000
0.000
                                    10

-------
be either:  1) displacement, 2) near perfect mixing, or  3) very  poor
mixing to the extent of clean water short  circuiting through  the basin
to the exit.  The experimental values cannot be  below  the  ideal  flush-
ing curve in the lower limit curve as given by values  in Table  1.   Each
curve consists of a short section of the ideal flushing  curve plus  an
ideal mixing curve which starts at tg/td.  It is recognized that  neither
ideal mixing nor a perfect short circuit (upper  limit  of no flushing)
are possible.  Variability in the experimental values  of t /td  can  be
attributed to:  1) the jet expansion and 2) the  type of  start,  "static"
or "running."

Examination of the limits of Eq. 14 reveals some of the  complexity  of
establishing b values.  td/tg cannot be less than unity. When Y/W£  =  1,
td/ts - 1 and b must equal 4/ir.  The minimum value of  Y/X  also  occurs
when Y = W^.  If Y = °°, then Y/W^ = °°, td  = °°, and td/ts = <». However,
ts is finite.  If V^ =0, ts = 0, td = <*>,  and td/ts =  °o.   In  this case,
Eq. 14 is not valid.
The experimental data have been plotted  in Figure  6  to  show the  depend-
ence of td/tg on aspect ratio Y/X  and  inlet width  Wj_.   These data  are
averages for the combined data   (4
from two depths and  three ve-
locities.  Within each group
there is random scatter of ±10
percent except for two readings
which were ±25 percent from the
average.  It should  be noted
that the running start data
gave t,j/ts values which were
from two to five times those
for the static starts.

The curves shown in  Figure 6
can be combined with Eq. 14.  ~
                               TJ
        10Y
                  = 8 in.) (16)
with Wi, X, and Y  in feet.  In
Eqs. 15 and 16, b  = 0.29X and
0.24X, respectively, for a
typical jet expansion rate.

Considerably more  data, pref-
erably under natural condi-
tions, are required to extend
with confidence Eq. 15 beyond
the limits of  these tests.
                                   td
                 Limit Ideal Flushing — =1.0
Fig. 6.  Flushing Parameter, td/ts, as a
         Function of Aspect Ratio, Y/X,
         and Inlet Width, WA
                                11

-------
                       EXPERIMENTAL APPARATUS
 Figure  7  is  a  schematic  of the test apparatus.   The test basins were
 constructed  on a  12-foot by 12-foot wood platform.   The entrance still-
 ing box,  gate,  and depth measuring piezometer tap were common for all
 tests.  At the exit,  the mixer,  Fluorometer tap,  control gate,  and
 measuring weir were common.   The sides  of each  basin were fabricated
 with sheet metal  fastened to a plywood  template of  the desired ellipti-
 cal shape.   The entrance and exit sections were faired to the elliptical
 basins  by a  2-inch radius.   Sections were provided  for an 8-inch exit
 width and for  both an 8-inch and a 4-inch entrance  width for each basin.

 A constant head tank supplied the clean water.   After the desired flow
 rate was  established with regulating valves,  a  shut-off valve could be
 closed  and reopened without  affecting the flow  rate.   Flow rates were
 measured  with  the  90° V-notch weir.   Repeated readings on the point gage
 established  the constancy of flow rate  during a test  run.   A tail-water
 gate at the  exit of the  discharge channel controlled  the depth.   Time-
 lapse,  motion  pictures in color  were taken at intervals.   Rates  of one
 picture per  second,  per  two  seconds,  or per four  seconds were used.

 Samples of the  outlet channel  flow were pumped  at a  constant rate through
 a heat  exchanger for  temperature stabilization,  then  through the Fluorom-
 eter.   The Fluorometer readings  were recorded with a strip  chart  recorder.
 Temperatures were  sensed  at  the  outlet  of the flow-through door  with a
 thermistor and  recorded whenever changes  occurred.  The  inlet and outlet
 temperatures of the basin were monitored  for  a  few  of  the  final  tests.
 The water in the basin was usually warmer  than  the  inlet  supply  by 2  to
 5 degrees Celsius.  The flow was warmed about 1.5 degrees  Celsius by
heat  from the air  and  lighting during passage through  the  basin.
                          TEMPERATURE
                           INDICATOR
                          PUMP

                          CHILLER
                  12" GRID
                RECORDER
                                                             FLUOROMETER
/  CONSTANT
 TEMPERATURE
    BATH
         ,  'i i — HEAT
         P  J   EXCHANGER
                                                           TAIL-WATER!
                                                           CONTROL—'
       "BAFFLES
  1—ENTRANCE BOX
                              PLAN  VIEW
                Fig. 7.  Schematic of Test Apparatus
                                  13

-------
                          TEST PROCEDURE


1.  A flow rate was established with the upstream regulating valves.
2.  The tailgate was adjusted to establish the desired depth.

3.  Flow was shut off without disturbing the regulating valves.
4.  The basin was filled to the test depth with water polluted with
fluorescent (Rhodamine WT) dye.  This dyed water extended to the control
gates in the inlet and outlet channels.

5.  The entrance box was filled to the necessary depth with clean water.
6.  After the basin water became quiescent the upstream control gate was
opened.

7.  The shutoff valve and the downstream control gate were opened simul-
taneously.

8.  The Fluororaeter, timing, and camera systems were started with the
start of flow into the entrance channel.

At the beginning of each test the basin concentration was adjusted to be
within the range  85 to 100 percent on the Fluorometer scale being used.
The test was terminated when the outflow percentage reached about five
percent.  Table 2 summarizes the parameters studied in this investigation.
               Table 2.  Summary of Test Conditions
Variable
(1)
Inlet width, /
in inches X
Inlet velocity, f
in feet per <
second I
Depth, in f
inches L
Running Start
also tested
Basin Size, in feet, Y by X
4 by 10
(2)
4.0
8.0
0.1
0.2
0.4
2.0
4.0
No
6 by 10
(3)
4.0
8.0
0.1
0.2
0.4
2,0
4.0
Yes
10 by 10
(4)
8.0
0.1
0.2
0.4
2.0
4.0
No
10 by 6
(5)
4.0
8.0
0.1
0.2
0.4
2.0
4.0
No
                                 15

-------
                          DATA ANALYSIS
The raw data from the strip charts were corrected  for  temperature  vari-
ations in the sample temperature by  the technique  given  by Cobb  and
Bailey (2).  Time adjustments were made to  correct the raw data  for
transit time in the inlet, and  the delay due  to  passage  through  the  out-
let channel and the tube  leading to  the Fluorometer.   Although a small
error exists due to instrument  response lag,  no  correction was made  for
it.

In Figure 8 the data were reduced to a common comparison base by using
the relative concentration; that is, measured concentration  divided  by
initial concentration and multiplied by 100 to return  to percentages.
A dimensionless time equal to the time from the  start  of flushing  (when
clean water enters the basin) divided by the  basin detention time  has
also been used for all comparisons and evaluations.  Figure  8 may  be
thought of as a dimensionless hydrograph of the  outflow  concentration.
               i   i   i   i   1   i   i   ill  ii
            0    0.4   0.8    1.2    1.6   2.0   2.4   2.8   3.2
        Fig.  8.  Typical Outflow  Hydrograph,  C/CQ  Versus  t/td
The area under the curve when multiplied by the proper factors gives the
quantity of dye removed.  This subtracted from the initial quantity gives
the residual concentration.  Considerable local variation from this aver-
age value will occur throughout the basin.
                                17

-------
 Referring  to  the  simplified concentration versus  time  curve in Figure  9:
                 Initial quantity of dye  =  KG V
                                            dt
               Dye removed  =  K/  Q.C
                                 t              «•
Residual quantity of dye  =  KCRV  =  K(CDV - f  Q.C
                                         R    /    i n
                                                          dt)
                 Basin volume
                             V  =  0.257rXYd
            Basin detention time  =  t.  =  —-  =
         (17)

         (18)

         (19)

         (20)

         (21)
               Fig. 9.   Concentration Versus Time
Eliminate K, substitute Q^d for V, divide by Co and introduce  td into
the third term of  Eq. 19 to obtain:
                           fC0
                        =  k
                                        tn/td
                                        t0/td
(f-)      (22)
Simplify to:
                                                                   (23)
When multiplying by 100 to get the relative concentration on a percentage
basis, Eq. 23 is applicable to the curve shown in Figure 8.

Because the experimental Cn versus t curve is not mathematically defin-
able, discrete steps as shown in Figure 9 must be summed to obtain the
area under the curve.  Equation 24 shows this summation as it was used
in the computer program to obtain the relative residual concentration,
CR/C0.
                                18

-------
 CR
(7-) 100
 uo
={l-
                       o,
Wl
 i
100    (24)
The required time and temperature corrections were applied to the data
read from the strip chart before they were entered into a program on the
GE 235.  Table 3 is a printout from this program.  The first two columns
are the inputs of time and fluorescence concentration in the outflow.
The third column shows the relative fluorescence of the outflow, (100C/CO).
The fourth column gives the ratio of time to detention time, and the
last column shows the mean residual relative concentration in the basin,
(100CR/C0).

The residual concentration calculated for the terminal time can be com-
pared with the residual measured with the Fluorometer after thoroughly
mixing the basin contents at the end of the test.  The measured value
was four percent and the calculated value was one percent for this example.

                  Table 3.  Data and Computed Values
t, In
seconds
(1)
•
68.
77.
79.
82.
85.
88.
90.
95.
98.
101.
103.
105.
106.
108.
110.
112.
113.
115.
118.
120.
122.
126.
131.
166.
182.
200.
240.
348.
368.
434.
469.
478.
C, tn
percent
(2)
94.
94.
92.
90.
88.
86.
84.
82.
80.
78.
76.
74.
72.
70.
68.
66.
64.
62.
60.
58.
56.
54.
52.
50.
50.
48.
46.
44.
44.
42.
42.
40.
38.
c/c0

(3)
100.
100.
98.
96.
94.
91.
89.
87.
85.
83.
81.
79.
77.
74.
72.
70.
68.
66.
64.
62.
60.
57.
55.
53.
53.
51.
49.
47.
47.
45.
45.
43.
40.
t/*d

(4)
.00
.09
.10
.11
.11
.12
.12
.12
.13
.13
.14
.14
.14
.14
.15
.15
.15
.15
.16
.16
.16
.17
.17
.18
.23
.25
.27
.33
.47
.50
.59
.64
.65
CR/CO t, in
seconds
(5) (1)
100. 648.
91. 660.
90. 691.
89. 725.
89. 743.
89. 767.
88. 814.
88. 855.
87. 910.
87. 960.
87. 980.
86. 1008.
86. 1071.
86. 1160.
86. 1182.
86. 1304.
86. 1312.
85. 1427.
85. 1481.
85. 1555.
85. 1602.
85. 1638.
84. 1713.
84. 1780.
82. 1955.
80. 2004.
79. 2266.
77. 2340.
70. 2677.
68. 2837.
64. 2965.
62. 3000.
62. 3305.
C, in
percent
(2)
38.
36.
34.
32.
30.
30.
28.
28.
26.
26.
24.
24.
22.
22.
20.
20.
18.
16.
16.
14.
14.
12.
12.
10.
10.
8.
8.
6.
6.
4.
4.
3.
3.
C/C0

(3)
40.
38.
36.
34.
32.
32.
30.
30.
28.
28.
26.
26.
23.
23.
21.
21.
19.
17.
17.
15.
15.
13.
13.
11.
11.
9.
9.
6.
6.
4.
4.
3.
3.
t/'d

(4)
. .88
.90
.94
.98
1.01
1.04
1.10
1.16
1.23
1.30
1.33
1.37
1.45
1.57
1.60
1.77
1.78
1.94
2.01
2.11
2.17
2.22
2.32
2.42
2.65
2.72
3.07
3.18
3.63
3.85
4.02
4.07
4.48
CR/CO

(5)
53.
52.
50.
49.
48.
47.
45.
43.
41.
39.
39.
38.
35.
33.
32.
28.
28."
25.
24.
23.
22.
21.
20.
19.
16.
15.
12.
12.
9.
8.
7.
7.
5.
                                  19

-------
                             TEST RESULTS
The computed values  for each  run were plotted and the  smooth curves
shown  in Figures  10  through 13  were drawn.   The solid  line was  drawn
through the visual average of the data.   The dashed  lines show  the
range  for two identical runs  having the greatest spread,  and the  dotted
lines  border the  inclusive field for all  data taken  on the basin.
      100
                                       DESCRIPTION
                                 Visual average  of all data
                                  Inclusive field  for all data
                        	Identical runs with, widest spread
            Idea!  "|
            Flushing^
            Curve  J
o
                         1.0              2.0              3.0          3.8
                           Relative Length of Flushing Time
                 Flushing Curves  for 10-Foot by 4-Foot Basin (X-Y)
      100
                                       DESCRIPTION
                                 Visual average of all  data
                                 Inclusive field for all data
                                 Identical runs with widest spread
           Ideal  1
           Flushing}-
           Curve  J
                         1.0              2.0              3.0          3.8
                         ,  Relative Length of Flushing Time
            11.   Flushing  Curves for  10-Foot by 6-Foot Basin (X-Y)
                                   21

-------
      100
       80
      60
  i  §
      40
 O 
  >
  "5 o
  5
    40
     c
     a>
     8
  a
      20
  0

Fig.
                               CURVE
                                           DESCRIPTION
                                     Visual average of all data
                                      Inclusive field for all data
                                      Identical runs with widest spread
         Ideal  1
         Flushing^
         Curve  J
               i   i
                                                     3.0         3.8

         13.  Flushing Curves for 6-Foot by  10-Foot Basin (X-Y)
                        1.0             2.0
                        t  Relative Length of Flushing Time
Most of  the  testing was done by introducing  a clean stream into a pol-
luted static basin and removing an equal  quantity of polluted flow  at
the opposite end of the basin.  However,  several running starts were
made.  Flow  through the basin was established and dye fed into the  in-
flow.  When  the concentrations of dye  in  the outflow, basin, and inflow
were the same,  the dye feed was stopped to start the run.  During the
running  starts  the basin circulation pattern was established by the  time
the clean water was introduced.  Therefore,  the transit time across the
basin was much shorter and, consequently, the initial flushing was  less
efficient for the running starts  than  for the static starts.  The  time
required to  reach a Cjj/Co value of five percent was essentially  the
same for both types of starting.
                                  22

-------
With  a  data  spread  for  identical  conditions nearly  as  great  as  the  in-
clusive field  for all tests  of  a  basin,  the effect  of  the  test  parame-
ters  is either completely masked  or barely discernible.

Motion  pictures of  identical conditions  show  completely  different jet
behavior.  Selected frames from the motion pictures do not show the com-
plete jet  action and the dispersion with time but they do indicate stages
of  the  dispersion and jet movement and permit comparisons  to be made.
Figure  14  shows the persistence of "hot" islands of pollutant for Run I
that  do not  occur in identical  Run II.

In  the  first run of another  test, the inflow  jet spread  laterally almost
across  the basin, mixed with the  dye, and swept the pollution toward the
outlets in a "front."   In an identical run the inflow  jet  remained  iso-
lated and  mixed only at the  edges so that a clean or slightly mixed
stream  penetrated to the outlet.  Whether such jet  behavior  is  a random
phenomena  or is triggered by subtle differences in  the test  conditions
remains to be  determined.
                    DISCUSSION OF TEST RESULTS
Two distinct yet related flushing actions take place.  Initially, the
clean water, displaces polluted water and this is  the most efficient
flushing possible.  In fact, if a barrier could isolate the clean water
from the polluted water and prevent mixing as the jet spread laterally,
the basin would be cleansed in a time, t/tj = 1.0.  Figures 10 through
13 show this ideal flushing curve as a straight line between Cn/Co = 100,
t/td = 0.0; and CR/CO - 0.0, t/td = 1.0.  The efficiency of the flushing
can be judged by how close the experimental curve comes to this ideal
curve.

However, mixing does occur and this modifies the flushing curve. Ideally,
the incoming water should spread across the basin and mix uniformly at
the front.  This would accomplish two desirable objectives:  1) the time
of the displacement phase would be long, and 2) there would be no iso-
lated islands of highly polluted water left in the basin.

Figure 15 shows the strong demarcation that occurs between the original
basin water and the inflow when the jet diffuses across the basin later-
ally.  This run had a long initial displacement time, t
Figure 16 shows a reproduction of the average curves for the basins
studied.  The wider the basin, the greater the relative time, t/t^, re-
quired to reach a given relative residual concentration.  The time dur-
ing which the flushing curve coincides with the ideal flushing curve is
inversely proportional to the width of the basin.
                                23

-------
   Run I
Run II

       CS
                            ^ ..__
                                     360  SECS.
Fig. 14.   Identical  Run Comparison, Runs 60 and 61

          X by Y = 6 ft by  10 ft,
          Wi =8 in.,  d ••-- 2 in., v - 0.1 fps.
          Note dye islands  in Run I.
          Arrows indicate circulation
                      24

-------
         Run I
                                   Run II
                                      . 6
      1300 SECS.

 DYE ISLANDS  FORMED


           i  H

       •

 L  DYE i:




                 i
                • *
                Hm
      2420 SECS.
DYE  ISLAND PERSISTS
 1670 SECS.

NO ISLAN.
              Fig. 14.  (continued)
                     25

-------
 50  SECS.
 3  SECS.

193 SECS.
260 SECS.
     Fig. 15.  Example of Well-Diffused Jet
                     26

-------
    100
_    80
o
ZJ
•o
£
   240
   c
   
-------
                         ACKNOWLEDGMENTS


The co-investigators on this project were Claud C. Lomax and John F.
Orsborn.  A significant contribution to the analysis and data reduction
was made by S. T. Chen and C. Y. Eric Shih, Research Assistants in the
Albrook Hydraulic Laboratory.  This study was performed under Grant
16010 DMG by the Federal Water Quality Administration with grantee con-
tributions by the College of Engineering Research Division of Washington
State University, Pullman, Washington.  The Project Officer was Charles
F. Powers, Pacific Northwest Water Laboratory,  Water Quality Office,
Environmental Protection Agency, Corvallis, Oregon.
                              29

-------
                            REFERENCES
 1.  Abramovich, G.  N.,  The  Theory of  Turbulent  Jets.  M.I.T.  Press
    Cambridge, Mass.,  1963.

 2.  Cobb, E. D.,  and Bailey, J. F.,  "Measurement  of  Discharge  by
    Dye-Dilution  Methods, "  Hydraulic  Measurement  and Computation.
    Book  1, Chapter 14  (Supplement No.  1), U.S. Department of  the
    Interior, Geological  Survey Publication,  1965.

 3.  Albertson, M. D., et  al..  "Diffusion of Submerged Jets," Trans ->-"
    actions of the  American Society of  Civil  Engineers. Vol. 115
    1950, pp. 639-664.                                           '

4.  Abraham, G.,  "Jet Diffusion in Stagnant Ambient Fluid,"  Delft
    Hydraulic Laboratory  Publication  29. Delft, The Netherlands,
    July, 1963.

5.  Abraham, G.,  "Entrainment Principle and Its Restrictions to   ,/-
    Solve Problems  of Jets, " Journal  of Hydraulic Research.  IAHR
    Vol. 3,  No. 2,  1965, pp. I^23~.

6.  Abraham, G.,  "Horizontal Jets  in  Stagnant Fluid of Other Density,"
    Journal of the Hydraulics Division. ASCE,  Vol. 94, No. HY4, July
    1965, pp.  139-154.                                               '

7.  Daily, J.  W.,  and Harleman, D.R.F., "Turbulent Jets and Dif-
    fusion Processes," Fluid Dynamics, Addison -Wesley Pub. Co.,
    Reading, Mass.,  1966.
                               31

-------
                    APPENDICES
A.  Notation . .

B.  Data Summary
    Table 4:  Data Summary for 10-Foot by
              4-Foot Basin 	
    Table 5:  Data Summary for 10-Foot by
              6-Foot Basin 	
    Table 6:  Data Summary for 10-Foot by
              10-Foot Basin  	
    Table 7:  Data Summary for 6-Foot by
              10-Foot Basin  	
Page No,

   34

   36


   36

   37

   38

   39
                      33

-------
                       APPENDIX A.  NOTATION



The following symbols are used in this paper:



     a   coefficient in exponential equation



     b   coefficient for ratio of transit velocity to inlet velocity



     C   concentration, in percent



    C    outflow concentration at time, t , in percent



    C    initial concentration in the basin,  in percent



    CL   residual concentration in the basin, in percent
     K


     d   depth of flow or depth of lake,  in feet



     e   2.718



     f   function of



    F.   momentum force in inflow, in pounds



     K   coefficient used to obtain the absolute amount of dye

         removed from the concentration measured,  in percent



    K1   general pollutant removal coefficient used in Eq. 4



    Q    flow entrained by jet



    Q.   inflow rate,  in cubic feet per second (Qo)



    Q    outflow rate,  cubic feet per second



    Q    removal rate,  in cubic feet per second



    Q    average removal rate,  cubic feet per second



    Q    Q.  + Q ,  in cubic feet per second



     R   aspect ratio



     t   time,  in seconds



    t^   basin detention time = V/Q.,  in seconds



    tQ   time measured  from start of clean flow into basin,

         in  seconds



    tQ   time zero
                                 34

-------
ts   basin transmit time (time from the start of clean flow  into
     basin until the outflow concentration shows its effect  in
     that Cn is less than C0), in seconds

 v   mean basin flow through velocity = v^/2, in feet per second

ve   entrainment velocity, in feet per second

Vf   velocity in the inflow channel, in feet per second

 V   volume of basin, in cubic feet

W.j_   width in inlet, in feet

W0   width in outlet, in feet

 X   length of ellipse parallel to the direction of flow, in feet

 Y   width of ellipse perpendicular to the direction of flow,
     in feet

AC   change in concentration in time integral At

At   time integral

  y  absolute viscosity of liquid, assumed constant, in
     pound-second per square foot

  p  mass density of liquid, assumed constant for inflow and
     lake, in slugs per cubic foot
                            35

-------
U)
                                                    Table 4.
                                    Data  Summary  for  10-Foot by  4-Foot  Basin



In order of
Hour-Day-
Honth-Year
(1)
1420-16-5-69
1555-16-5-69
0850-20-5-69
0950-20-5-69
0855-27-5-69
1000-27-5-69
0910-4-6-69
1545-4-6-69
1315-6-6-69
1530-6-6-69
1555-9-6-69
1135-10-6-69
1400-10-6-69
1145-13-6-69
1350-13-6-69
1445-17-6-69
1613-17-6-69



Run
Num-
ber
(2)
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38



x,
in
feet
(3)
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10



Y,
in
feet
(4)
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4



"i,
in
inches
(5)
8
8
8
8
8
8
8
8
8
8
8
8
8
4
4
4
4



d,
in
inches
(6)
2
2
2
2
2
2
4
4
4
4
4
2
2
2
2
2
2


v§ In
feet
per
second
(7)
0.1
0.1
0.2
0.2
0.4
0.4
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.4
0.4
0.1
0.1

Ql,in
cubic
feet
per
second
(8)
0.011
0.011
0.022
0.022
0.040
0.040
0.044
0.044
0.044
0.044
0.044
0.022
0.022
0.022
0.022
0.0056
0.0056



ts>
in
seconds
(9)
202
242
93
100
60
61
110
120
124
103
77
109
105
65
74
345
310



*
in
seconds
.(10)
472
458**
224**
222**
130
134**
250**
250**
236**
236**
221**
225**
221**
262**
247
828
828
Relative Residual Concen-
tration, CR/CO, in percent

Terminal
Meas-
ured
(ID
2
4
3
4
2
3
2
i
1
2
2
2
3
4
3
3
4
Calcu-
lated
(12)
2
4
3
4
1
3
2
1
1
2
2
3
3
4
4
2
3
* Maximum
Extrapolated Minimum
**
Adjusted Average
Spread

At t/td -

2.5
(13)
3
3*
10
11
9
10
5
5
5
9
7
6
5
14
13
6
7
14
3
8
11

3.0
(14)
2*
2*
6
8
6
7
3
3
3
6
4
4
4
10
9
3
4
10
2
5
8

3.5
(15)
1*
1*
4
5
3
6
2*
2
1
4
3
2*
2*
6
6
2*
3*
6
1
3
5

4.0
(16)
1*
1*
3*
3*
2
5
1*
1
1*
3
1*
1*
1*
5
• 4
1*
2*
5
1
2
4

-------
                Table 5.
Data Summary for 10-Foot by 6-Foot Basin



In order of
Hour -Day -
Month-Year
(1)
1030-5-8-69
1500-5-8-69
1445-6-8-69
0845-12-8-69
1115-12-8-69
1345-12-8-69
1600-14-8-69
0900-15-8-69
1030-15-8-69
1340-15-8-69
1445-15-8-69
1000-2-9-69
1130-2-9-69
1430-2-9-69
1600-3-9-69
1045-4-9-69
1420-4-9-69
0900-5-9-69
1125-5-9-69
1020-8-9-69
1355-8-9-69



Run
Sum-
[>er
(2)
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59



x,
in
feet
(3)
10
10
10
10
10
10
10
. 10
10
10
10
10
10
10
10
10
10
10
10
10
10



Y/
in
feet
(4)
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6



vit
in
inches
(5)
8
8
8
8
8
8
8
8
8
8
8
4
4
4
4
4
4
4
4
4
4



d
in
inches
(6)
2
2
2
2
2
2
2
2
2
4
4
2
2
2
2
4
4
2
2
2
2


v. in
feet
per
second
(7)
0.2
0.2
0.2
0.1
0.1
0.1
0.4
0.4
0.4
0.2
0.2
0.4
0.4
0.4
0.4
0.2
0.2
0.2
0.2
0.1
0.1

Qi> in
cubic
feet
per
second
(8)
0.0222
0.0222
0.0222
0.0111
0.0111
0.0111
0.0444
0.0444
0.0444
0.0444
0.0444
0.0222
0.0222
0.0222
0.0222
0.0222
0.0222
0.0111
0.0111
0.0055
0.0055



ts,
in
seconds
(9)
107
99
39**
220
200
58**
55
55
27**
111
66**
77
77
20**
72
217
159
145
150
405
335



td,
in
seconds
(10)
342
-342
342
722
722
722
176
176
176
352
352
381
381
402
402
804
804
701
701
1512
1676
Relative Residual Concen-
tration, CR/CO, in percent

Terminal
Meas-
ured
(11)
7
6
5
5
6
4
4
2
5
3
4
5
7
4
4
8
5
5
6
5
5
Calcu-
lated
(L2)
2
2
6
7
5
4
2
1
4
9
7
3
5
1
5
5
0
5
6
6
6
^ Maximum
Extrapolated Minimum
** Average
Running Start gprcad

At t/td =

2.5
(13)
16
18
18
10
8
10
11
13
20
13
20
12
13
13
13
11
13
15
13
7
9
20
7
13
13

3.0
(14)
12
14
14
7
5
7
7
9
15
11
14
9
9
8
8
7
8
10
8
4*
6*
15
4
9
11

3.5
(15)
9
11
11
4*
3*
4
4
6
11
9
10
5
5
5
6
4*
5
6
6*
3*
4*
11
3
6
8

4.0
(16)
7
8
8
3*
2*
3*
3
3
8
6*
8
3
2*
2
2*
3*
3
3*
2*
2*
3*
8
2
4
6

-------
00
                                                     Table 6.
                                    Data  Summary for 10-Foot by 10-Foot Basin



In order of
Hour-Day-
Month-Year
(1)
1100-26-3-69
1000-1-4-69
0915-2-4-69
1415-3-4-69
1045-4-4-69
1040-11-4-69
1435-14-4-69
1030-16-4-69
1040-18-4-69
1450-18-4-69
1030-25-4-69



Run
Num-
ber
(2)
11
12
13
14
15
16
17
18
19
20
21



x,
in
feet
(3)
10
10
10
10
10
10
10
10
10
10
10



Y,
in
feet

10
10
10
10
10
10
10
10
10
10
10



Wi,
in
inches
(5)
8
8
8
8
8
8
8
8
8
8
8



d,
in
inches
(6)
2
2
4
2
2
2
2
2
4
4
4


v, in
feet
per
second
(7)
0.4
0.4
0.2
0.2
0.2
0.1
0.1
0.1
0.1
0.1
0.2

Q. in
cubic
feet
per
second
(8)
0.0388
0.0444
0.0444
0.0222
0.0222
0.0111
0.0111
0.0111
0.0222
0.0222
0.0444



tst
in
seconds
(9)
58
50
103
102
106
310
217
218
213
244
105



cd>
in
seconds
(10)
330**
278**
484**
572**
590
1298**
1263**
1274**
1133**
1180
590
Relative Residual Concen-
tration, CR/CQ, in percent
Terminal


Meas-
ured
(11)
7
8
3
6
6
7
8
6
11
9
12

Calcu-
lated
(12)
8
9
3
6
6
6
7
5
11
8
11
# Maximum
Extrapolated Minimum
A"A A \7tf*T" SJCTO
Adjusted Average
J Spread
At t/td =



2.5
<13)
17
17
13
10
10
5
10
7
13
14
25
25
5
1 1
LJ
20


3.0
(14)
13
12
9
7
7
4
7*
5*
10*
10
21
21
4
i n
1U
17


3.5
as)
9
9
6
5*
5*
3*
6*
4*
8*
8*
16
16
3

13


4.0
(16)
8*
8*
5
4*
4*
2*
5*
3*
7*
7*
14
14
2

12

-------
GJ
                                                     Table  7.
                                     Data Summary for 6-Foot by  10-Foot Basin



T r\t*Aevv f>£
Hour -Day
Month -Year
(1)
1335-15-9-69
1555-15-9-69
1600-16-9-69
0945-17-9-69
1320-17-9-69
0930-18-9-69
1135-18-9-69
0846-19-9-69
1129-19-9-69
0930-23-9-69
1321-23-9-69
1522-23-9-69



~*
Num-
ber
(2)
60
61
62
63
64
65
66
67
68
69
70
71



X,
"7
in
feet
(3)
6
6
6
6
6
6
6
6
6
6
6
6



y,
* 7
in
feet
(4)
10
10
10
10
10
10
10
10
10
10
10
10



Wtt
" JL9
in
inches
(5)
8
8
8
8
8
8
8
8
8
4
4
4



d.
^* t
in
inches
(6)
2
2
2
2
2
2
2
4
4
2
2
2


v, in
f f-
per
second
(7)
0.1
0.1
0.2
0.2
0.2
0.4
0.4
0.2
0.2
0.2
0.2
0.2

Q
cubic
f
per
second
(8)
0.0111
0.0111
0.0222
0.0222
0.0222
0.0444
0.0444
0.0222
0.0222
0.0111
0.0111
0.0111



t.,
S"
in
seconds
(9)
82
90
47
48
47
24
25
43
54
47
55
68



t A
in
seconds
(10)
687
687
342
342
342
178
178
356
356
737
737
737
Relative Residual Concen-
tration, CR/CQ, in percent
Terminal

Meas-
ured
(11)
7
6
4
2
6
4
9
6
7
6
6
6
Calcu-
lated
(12)
8
1
2
1
1
1
3
8
7
4
6
5
Maximum
Minimum
Average
Spread
At t/t. =
Q

2.5
(13)
34
21
19
27
23
27
28
20
20
18
22
20
34
18
23
16

3.0
(14)
29
14
13
20
17
22
22
16
17
12
16
13
29
12
18
17

3.5
(15)
27
8
9
15
11
17
17
13
12
8
11
10
27
8
13
19

4.0
(16)
26
5
7
11
8
13
13
11
9
6
9
7
26
5
10
21

-------
1
Accession Number
w
5
2

Subject Field &. Group
SELECTED WATER RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
Organization
WaaTni not-nn Sf-nt-n TTn-f ^roY•a^ f-v Pullman UaaVil not-nn QQIfi^
                 College of Engineering Research Division
                 Albrook Hydraulic Laboratory	
     Title
                 Flushing of Small Shallow Lakes
| Q Authors)
Lomax, Claud C.
Orsborn. John F.
16

21
Project Designation
EPA, WQO No.
16010 DMG
Note
 22
     Citation
 23
     Descriptors (Starred First)
               *Lakes,  *Hydraulics, Water Quality, Hydraulic Models,
 25
     Identifiers (Starred First)
              *Inflow,  *Flow Characteristics, Enhancement, Tests
 27
         Restoration of quality to polluted lakes by inflows of  clean water with
    simultaneous  outflow of polluted water was investigated.  Elliptical  basins were
    used in  the  laboratory to simulate shallow lakes.  The  investigation  determined
    the influence of selected geometric and flow parameters on the  flushing efficiency,
         The theoretical analyses are combined with the experimental results to obtain
    equations  for predicting the flushing curves.  Application of these equations will
    give the potential flushing efficiency of a proposed flushing scheme.
         The width of the basin perpendicular to the axis of flow is an important
    parameter.   In general,  narrower basins have better flushing action and less
    erratic  flow patterns than wider basins.  The pollution remaining in  a  basin
    after any  interval of flushing is primarily dependent on the time of  the flushing
    interval divided by the detention time of the basin.
         The report was submitted in fulfillment of Grant Number 16010 IMS,  under
    the partial  sponsorship of the Water Quality Office, Environmental Protection
    Agency.
Abstractor
    Ifolin F. Orsborn
                               Institution
           Washington State University
 WR:t02 (REV. JULY 1969)
 WRSIC
SEND WITH COPY OF DOCUMENT. TOl WATER RESOURCES SCIENTIFIC INFORMATION CENTER
                          U.S. DEPARTMENT OF THE INTERIOR
                          WASHINGTON, D. C. 20240
                                                                               * SPO: 1970-389-930

-------