Ecological Research Series
PHOSPHATE REDUCTION AND RESPONSE OF
PLANKTON POPULATIONS IN KOOTENAY LAKE
Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Corvallis, Oregon 97330
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into five series. These five broad
categories were established to facilitate further development and application of
environmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ECOLOGICAL RESEARCH series. This series
describes research on the effects of pollution on humans, plant and animal
species, and materials. Problems are assessed for their long- and short-term
influences. Investigations include formation, transport, and pathway studies to
determine the fate of pollutants and their effects. This work provides the technical
basis for setting standards to minimize undesirable changes in living organisms
in the aquatic, terrestrial, and atmospheric environments.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/3-76-063
May 1976
PHOSPHATE REDUCTION AND RESPONSE OF PLANKTON
POPULATIONS IN KOOTENAY LAKE
by
Richard A. Parker
Washington State University
Pullman, Washington 99163
Grant Number R-800430
Project Officer
D. Phillips Larsen
Corvallis Environmental Research Laboratory
Corvallis, Oregon 97330
U. S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND DEVELOPMENT
CORVALLIS ENVIRONMENTAL RESEARCH LABORATORY
CORVALLIS, OREGON 97330
-------
DISCLAIMER
This report has been reviewed by the Corvallis Environmental Research
Laboratory, U.S. Environmental Protection Agency, and approved for
publication. Approval does not signify that the contents necessarily
reflect the views and policies of the U.S. Environmental Protection
Agency, nor does mention of trade names or commercial products
constitute endorsement or recommendation for use.
ii
-------
ABSTRACT
The purpose of this research project was to determine the effects
of reducing by 90 percent the soluble inorganic phosphate input to
Kootenay Lake, British Columbia, Canada, a significant link in the
Columbia River system governed by United States-Canadian treaty.
Measurements on temperature, transparency, inorganic and organic
phosphate, nitrate, ammonium, chlorophyll j., copepods, and cladocerans
were made during 1971-75, and compared with observations made for three
years prior to phosphate reduction in 1969. Only a slight decrease in
chlorophyll occurred, although there were important changes in species
composition. Total zooplankton levels were not-affected, however one
genus (Daphnia) has virtually disappeared. Data indicate that primary
production in the lake was nitrogen limited prior to 1969, phosphate
limited two years later. These changes may also be related to the
completion in 1972 of Libby Dam in Montana, upstream from the lake on
the Kootenay River. An overview of the results is reported, with
detailed accomplishments reported in Parker 1»2>3»4»5>6 incorporating
the unpublished work of J. E. Cloern, J. R. Davis, J. R, Hargis, and
K. G. Taylor.
This report was submitted in fulfillment of grant number R800430
by Washington State University under the partial sponsorship of the
Environmental Protection Agency. Work was completed as of July 1975.
iti
-------
CONTENTS
Page
Abstract ill
List of Figures vi
List of Tables vii
Sections
I Introduction 1
II Conclusions 3
III Recommendations 4
IV Field Observations and Laboratory Analyses 5
V Simulation Model 32
VI References 49
VII Appendix 53
-------
FIGURES
No.
1 Sampling stations on Kootenay Lake 6
2 Flow of the Kootenay River at Station 1 8
3 Mean temperatures at Stations 1-5 9
4 Extinction coefficients 10
5 Soluble inorganic phosphate concentrations 12
6 Soluble organic phosphate concentrations 13
7 Particulate phosphate concentrations 14
8 Nitrate concentrations 16
9 Ammonium concentrations 17
10 Phytoplankton levels 18
11 Copepod levels 21
12 Cladoceran levels 22
13 Total zooplankton densities at Station 2 24
14 Gravid zooplankton densities at Station 2 25
15 Total zooplankton densities at Station 3 26
16 Gravid zooplankton densities at Station 3 27
17 Total zooplankton densities at Station 4 28
18 Gravid zooplankton densities at Station 4 29
19 Total zooplankton densities at Station 5 30
20 Gravid zooplankton densities at Station 5 31
21 Simulated soluble inorganic phosphate concentrations 37
22 Simulated nitrate concentrations 39
23 Simulated ammonium concentrations 41
24 Simulated phytoplankton levels 42
25 Simulated copepod levels 43
26 Simulated cladoceran levels 44
27 Velocity and volume flow at Station 1 45
28 Velocity at four depths at Station 2 47
29 Mean northerly velocity at Station 2 along 48
north-south axis of lake, and flow at Station 1
vi
-------
TABLES
No.
1 Drainage characteristics and nutrient loadings 2
for Kootenay Lake
2 Algal genera identified from Kootenay Lake for 20
1965-66 by Pillion e_t al. (unpublished) and for
1973-75 by Cloern and Prescott (unpublished)
3 Variables used in the nutrient-plankton model of 34
the mixed layer in Kootenay Lake
4 The model system 36
5 Constants used in the simulation model 38
vii
-------
SECTION I
INTRODUCTION
The effects of an Increase in the nutrient budget on lakes
have been veil documented in several cases (see for example Hasler7,
Edmondson et al. , Edmondson , Beeton * , Verduin ). There
have been fewer reports of the response by an enriched lake after
reduction in the nutrient income (Edmondson13> 14> Ahlgren15, Malueg
at _al.16). Kootenay Lake, a large, generally oligotrophic lake
in southeastern British Columbia, Canada, received significant
quantities of inorganic phosphate via the Kootenay River from 1953
until 1969. This material originated primarily from the activities
of a fertilizer plant located on a tributary of the Kootenay River
near Kimberley, British Columbia. Early in 1969 there was a large
reduction (80-90%) in the amount of phosphate being put into the
river and thus into the lake.
Prior to phosphate reduction, Taylor17 and Parker1 conducted a
limnological study from 1966 until 1969, emphasizing nutrient levels
and plankton densities. About the same time, Fillion (unpublished)
focused attention on the phytoplankton composition. His samples
have subsequently been examined by Northcote et al. (unpublished)
at the University of British Columbia. ZyblutT8~eviewed long-term
changes in the zooplankton population of Kootenay Lake, and North-
cote1^ provided a historical account of the salmonid fishery as it
related to human activities and nutrient loading. Further study
was begun in 1971 and continued through May 1975. Results from the
second period will be compared with those given by Taylor17, focusing
largely on changes in nutrient concentrations and plankton populations,
It also should be pointed out that changes caused by reduced phosphate
income have been compounded by completion of Libby Dam, located
several hundred kilometers upstream on the Kootenay River in Montana.
Shortly after leaving the Lake, the Kootenay River joins the Columbia
River, providing about 30 percent of the total low flow at that point.
The general drainage characteristics of Kootenay Lake, as well as
nutrient loading before and after 1969, are sunmarized in Table 1.
-------
Table 1. DRAINAGE CHARACTERISTICS AND NUTRIENT
LOADINGS FOR KOOTENAY LAKE
Drainage Characteristics
2
Drainage Basin (km ) 2 45,669
Lake Surface Area (km ) 417
Annual Inflow £n ) 26 x 10
Lake Volume (m ) 40 x 10
Retention Time (days) 566
Lake Mean Depth (m) 96
Nutrient Loadings
2
Nitrate - Nitrogen (g/m /yr)
1966-69 0.45
1971-74 2 0.48
Ammonium - Nitrogen (g/m /yr)
1967-69 0.25
1971-74 2 0.18
Phophate - Phosphorus (g/m /yr)
1966-69 0.68
1971-74 0.051
-------
SECTION II
CONCLUSIONS
Seasonal flow patterns of the Kootenay River have been altered con-
siderably by the completion of Libby Dam in 1972. The peak values
during the spring run-off period in late May and early June have
been reduced approximately 50%, with the balance of the flow being
distributed over the summer period. Temperatures in the lake are
somewhat lower than before the construction of the dam, and the
transparency of the lake has been increased by a factor of two
during late spring and early summer, coincident with lower volume
flows.
Soluble inorganic phosphate entering the lake has decreased approx-
imately 90% since 1969. There has also been a concurrent decrease
in the soluble organic phosphate fraction as well as in the parti-
culate phosphate. The soluble organic fraction has, however, not
decreased as much as the inorganic fraction. Furthermore, the
decrease in the particulate fraction can be attributed largely to
a decrease in silt load brought about by reduced flows of the
Kootenay River. Nitrate and ammonium concentrations have changed
little since 1969, although reductions during the summer period due
to algal growth no longer appear to be limiting to the phytoplankton
populations. Rather phosphate has become the nutrient of prime
consideration.
Chlorophyll a. levels in Kootenay Lake have been reduced 10-15% as
a result of reduced phosphate input and somewhat lower temperatures.
There has been a change in the species composition of the phyto-
plankton, with diatoms playing an increaalngly important role relative
to blue green algae. Daphnia has been virtually eliminated from
Kootenay Lake for reasons that are unclear. Preferential feeding
by landlocked sockeye salmon in a more transparent lake could be
responsible. This would of course require that Diaphanosoma not
be preyed upon heavily since this genus has become the dominant
cladoceran in the lake. Larger copepod populations could be brought
about by reduced grazing pressure by Daphnia on important food items
like Cyclotella and Cryptomonas.
-------
SECTION III
RECOMMENDATIONS
Although phosphate input into Kootenay Lake has been decreased
from 0.68 to 0.05 g/m /year, attempts to predict the consequences
on plankton populations have only been partially successful. The
differential equations used to simulate the system suffer from lack
of an adequate hydrodynamic component, as well as reliable estimates
of essential parameters. If the model developed for this lake is to
be applied further, or to other lakes, inclusion of a well-defined
hydrodynamic sub-model is recommended. In addition, rate constants
for phytoplankton populations should be measured in the field, in-
cluding dependence on light intensity, temperature, and nutrient
concentrations. Other types of mathematical models (e.g. multi-
variate regression) should also be investigated to improve prediction
accuracy needed for sound management.
-------
SECTION IV
FIELD OBSERVATIONS AND LABORATORY ANALYSES
METHODS
Sampling stations on Kootenay Lake were the same for 1971-74 as those
used by Taylor17, with the exception that Station 3 was discontinued in
1973 (Figure 1). Velocity of river input was measured during 1974-75
at Station 1 with an Ott current meter at 1 m, and at Station 2 by
means of drogues. Temperature and light penetration were measured by
Whitney underwater instruments. Readings were taken at 1-m intervals
from the surface to 10 m at Stations 2-5, and to 5 m at Station 1.
Zooplankton samples were collected at Stations 2-5 using a Clarke-
Bumpus sampler with a number 10 nylon net towed at approximately 3
km.p.h. Two samples were collected at each station and each of three
depths (1, 5, 10 m). The samples were preserved in formalin and
counted subsequently in the laboratory. Five 1-ml subsamples of a
known dilution of the original sample were removed by piston pipette
and counted on Sedgewick-Rafter cell. The density of a given zooplankton
species was then calculated as follows:
Number/liter = total no. in subsamples x vol. of dilution
5 4 x no. of revolutions ^ '
Water samples were collected at each station by Kemmerer bottle. At
Station 1 two samples were taken at 1 m, and at Stations 2-5 two samples
were collected at each of the three depths. Two subsamples of known
volume (300-500 ml) from each original sample were filtered using a
0.45n membrane filter (Gelman, 47 mm dia.). One filter was analyzed
for particulate phosphate, and one analyzed for total chlorophyll a..
The chlorophyll a. concentrations were measured by soaking the filter in
10 ml of 100% methanol for 3-4 hours, then reading the optical density
at 660 nm with a 1 cm path length. To convert the optical density to
chlorophyll a. concentration in yg /liter, the optical density was
multiplied by 13.9 x vol. of water filtered x vol. of solvent used
(10), as suggested by Tailing and Driver20.
The phosphate concentration was measured using the method of Strickland
and Parsons21. Nitrate was analyzed using the phenoldisulfonic acid
method of APHA, AWWA, and WPCF22; ammonia determinations were made
using the method of Solorzano23.
Samples ordinarily were collected weekly during the summer, biweekly
in the spring and fall, and monthly in the winter. Mean values over 10
meters will be used as the basis for subsequent discussions of all
variables considered.
-------
N
0 10 20 30 km
Figure 1. Sampling locations on Kootenay Lake
6
-------
PHYSICAL CHARACTEBISriCS
River Flow
The seasonal pattern of flow of the Kootenay River was altered con-
siderably following the completion of Libby Dam in 1972 (Figure 2).
During 1966-1969, peak flow occurred early in June and was approxi-
mately two million liters per second. In 1972 the peak was approximately
1.6 million liters per second, less than 0.5 million liters per second
in 1973 and about 1.2 million liters per second in 1974. Attempts to
disburse accumulated water from the reservoir behind.Libby Dam over a
longer period (and more gradually) were moderately successful.
This new flow pattern has had a considerable effect on the temperature
regime of the lake, as well as on the transparency of the water.
Nevertheless the water flows northward along the eastern shore of the
lake from the point of entrance of the river at the south end. In fact,
this water may travel all of the way to the north end of the lake, where
it mingles with water entering in the Duncan and Lardeau Rivers, before
moving southward and out through the west arm.
Tempejrature
Water is released from behind Libby Dam at various depths below the
surface of the impoundment in order to maintain acceptable temperatures
in the Kootenay River. The fundamental seasonal pattern has not been
altered in Kootenay Lake, although there has been a small decrease in
mean values (Figure 3). Recall that Station 1 is located at the point
where the Kootenay River enters the lake, and that Station 2 through 5
are distributed over the south half of the lake.
Transparency
Figure 4 illustrates well the effect of changing the seasonal flow
pattern on the transparency of water in and entering Kootenay Lake.
The suspended silt load in the river (Station 1) has been reduced
markedly by lowering peak flows. This has had a pronounced effect on
the extinction coefficients for the upper 10 meters as is clearly
shown in Figure 4. This reduction in turn has altered the seasonal
growth pattern of the phytoplankton. Prior to 1972, Kootenay Lake
typically had a spring bloom followed by substantial reduction during
June and another bloom in August and September. Since 1972, phyto-
plankton levels in the lake have been more evenly distributed through-
out the growing period. Extinction coefficients in the lake (on a per
-------
2.H
S 1.8
v,
C30
0.6
1966 1967 1968 1969
1970
YERR
1971 1972 1973 1974
Figure 2. Flow of the Kootenay River at Station 1.
-------
;o o
i*
3; -i
-o
ro
O
.
3
a
oo
u
CO
UJ
I
8
15
15
15
15
5
D
STfi 2J
STfl 3-
SIR
STfl 5
1966
1967
_^
1968
1969
1970
YEflR
1971
1973
1973
1974
Figure 3. Mean temperatures at Stations 1-5.
-------
SIR 1
3.0
1.0
3.0
1.0
3.0
1.0
3.0
1.0
3.0
1.0
0
SIR 2
STfl 3
STfl
STfl 5
1966
1967
1968
1969
1970
YEflfi
1971
1972
1973
19714
Figure 4. Extinction coefficients.
-------
meter basis) frequently reached 2.0 in June; since 1972 values rarely,
if ever, exceed 1.0. This means, of course, that the depth to which
one percent of the surface light is transmitted during the most
turbid periods has been increased from about two meters to over four
meters.
NUTRIENTS
Phosphorus
Prior to 1969 the amount of inorganic phosphorus entering Kootenay
Lake at Station 1 ordinarily exceeded a concentration of 12 micromoles
per liter (yM/1) during the winter months. Since 1969, concentrations
in entering river water have not exceeded 2 yM/1 (Figure 5). Although
concentrations were obviously low during the winter months, maximal
daily inputs occurred during periods of peak flow. For example, about
15 metric tons entered the lake on 30 May 1966. Following input
reduction, a maximum value of 0.75 metric tons entered on 20 May 1972.
Average entrance over the year has been reduced approximately 10 fold,
and mean values in the lake have been reduced from approximately 3
yM/1 to 0.3 yM/1. The amount of soluble organic phosphate is shown
in Figure 6. This quantity does not appear to change significantly
during the year, although there has surely been a reduction from
about 1 yM/1 prior to 1969 to about 0.2 yM/1 since. Note that the
inorganic phosphate pool in the lake is reduced during the summer
period of peak algal growth, particularly during the period of 1971-
1974. Prior to 1969, the only significant summer depressions occurred
in 1966 and 1967. Particulate phosphate (Figure 7) is composed of
phosphate adsorbed on silt particles as well as phosphate incorporated
in living organisms. A review of the seasonal distribution for this
variable indicates that the highest levels are coincident with peak
river flow, thereby suggesting that much of the particulate phosphate
is in fact adsorbed material. Unfortunately no attempt was made to
distinguish between particulate phosphate in living organisms from
that adsorbed to silt particles and, consequently, one cannot use
the seasonal distributions as indicative of plankton population
fluctuations.
Nitrogen
The highest nitrate concentrations in the Kootenay River occur
during winter and spring. These high levels are reflected in
values measured at Station 2-5. As with phosphate, maximum total
11
-------
STfl 1
10
5
10
STR 2
STR 3
10
ro
STfl 4
10
STfl 5
10
O1—
1966 1967 1968 1969
1970
YEflR
1971 1972 1973 1974
Figure 5. Soluble inorganic phosphate concentrations.
-------
10
5
10
5
y 10
o
10
5
10
5
0
STP 1
STfl 2
STfl 3
STfl 14
STfl 5
1966
1967
-t-M-fc.
1968
1969
1970
YEflR
1971
1972
1973
1974
Figure 6. Soluble organic phosphate concentrations,
-------
HO
20
40
20
40
20
40
20
SIR 1
20
0
SIR 2
SIR 3
SIR
SIR 5
_L
1966
1967
1968
1969
1970
YEPR
1971
1972
1973
1974
Figure 7. Participate phosphate concentrations.
-------
daily Input coincides with periods of peak river flow. During 1966
and 1967 when inorganic phosphate levels were somewhat depressed,
nitrate concentrations were reduced to near zero, indicating that
nitrate was in fact limiting the growth of phytoplankton prior to
1969 (Figure 8). On the other hand, during the period 1971-74,
nitrate was never reduced to zero during the year, whereas inorganic
phosphate frequently could not be detected during the same period,
Kootenay Lake probably became phosphate limited in terms of primary
production (Chu2\ Goldberg et_ al.25, Fogg26, Lackey27, Thomas28).
One should also look to ammonium concentrations (Figure 9) since algal
cells may take up ammonium preferentially over nitrate for at least
some species (Eppley29). An examination of the data shows that in
1967 ammonium levels reached zero at times when nitrate was still
present. During 1968, nitrate never reached low levels whereas
ammonium frequently was depressed. The same can be said for 1969.
During 1971-74 ammonium was reduced during the summer months but
was always measurable.
BIOLOGICAL POPULATIONS
Phvtoplankton
Chlorophyll a. readings have been converted to milligrams per liter
dry weight of algae assuming a one-half % concentration of chlorophyll
based on series of measurements made during the month of May. This
proportion undoubtedly does not remain constant throughout the year,
however conversion to dry weight facilitates comparison with similar
quantities for zooplankton. There has been a reduction in the number
of major phytoplankton peaks since 1969 (Figure 10), but the annual
means have decreased only about W%. Figures for 1971 and 1972 sug-
gest that the phytoplankton levels remain reasonably high through-
out the summer, although in 1973 there were distinct peaks in the
spring and late summer at all stations. Since there was no distinct
period of peak flow or abnormal temperature patterns in 1973, one must
look to nutrients and zooplankton for the cause. Here phosphate was
essentially zero in midsummer (between peaks), and ammonium levels
remained low. Nitrate, on the other hand, seemed to mirror the
phytoplankton peaks. This observation suggests that the phytoplankton
species present had taken up most of the ammonium and were required
to rely on nitrate as a prime source of nitrogen for growth. Since
the nitrate levels did not in fact go to zero, one is forced to
conclude that phosphate ultimately limited the growth and that perhaps
temperature changes the uptake kinetics in such a way that two peaks
were produced. One cannot avoid considering a further possibility,
that is, grazing pressure brought about by a massive Increase in the
cladoceran population.
15
-------
ID
LU
d
C
oc
1966
1974
Figure 8. Nitrate concentrations.
-------
STfl 1
i
g
10
5
10
5
10
5
10
STfl 2
STfl 3
STfl
STfl 5
10
1966 1967 1968 1969
1970
YEflR
1971 1972
1973 1974
Figure 9. Ammonium concentrations.
-------
<
1966
1974
Figure 10. Phytoplankton levels.
-------
Little detail is available on algal composition in Kootenay Lake.
During 1965-66, D. Pillion made extensive collections at the lake;
however, he did not identify many of the species present in his
samples. Subsequent work by G. Ennis, also from the University of
British Columbia (unpublished), yielded a list of species present
in the lake at that time. This effort was followed in 1973-75 by
the unpublished results of Cloern and Prescott. Since the material
collected in 1965-66 was not examined by Cloern and Prescott, nor
was the material collected by Cloern examined by those reviewing
the 1965-66 samples, only generic comparisons seem warranted
(Table 2). Unfortunately, specific population densities are not
available for further comparison. Lately, among the larger forms,
Aaterionella is dominant during May and June, followed by Anabaena.
Phormidium. and Fragilaria during July, August, and September. Of
those genera important in the diet of zooplankton, Cyclotella and
Cryptomonas are prevalent from May through August. Of significance
here is the fact the Anabaena did not reach bloom proportions in
1973-75, as it did traditionally each August during 66-69. It should
also be pointed out that Oscillatoria was not identified for 1973-
75, as it had been for 1965-66. Furthermore Stephanodiscus did
not appear in 1973-75 as it had earlier. Considerable care must
be taken in evaluating the generic lists for the two periods,
since the outflow of Duck Lake at the south end of Kootenay Lake
empties into the Kootenay River and contributes algal species that
may not be typical open water forms. For example, Chiamydomonas»
Eudorina, Pandorina, and Volvox were identified in Kootenay Lake,
but should not be considered typical of the Kootenay Lake plankton.
Zooplankton
The observed copepod and cladoceran quantities in milligrams
per liter were calculated by multiplying numerical densities
by 2.5 and 3*5 micrograms per individual, respectively. As
with chlorophyll, these values do not remain constant, part-
icularly during periods of peak reproductive activity. The
total copepod population tended to peak during late summer
and early fall, although they were represented in the lake
throughout the year (Figure 11). The cladocerans, too, peaked
in late summer; however, they were virtually absent from the
lake during the balance of the year (Figure 12).
19
-------
Table 2. Algal genera identified from Kootenay Lake for 1965-66 by Fillion et^ al (unpublished) and
for 1973-75 by Cloern and Prescott (unpublished)
1965-66
1973-75
[O
o
Cyanophyta
Anabaena
Microcystis
Merismopedia
Oscillatoria
Phormidium
Cryptophyta
Chroomonas
Cryptomonas
Pyrrophyta
Ceratium
Gymnodinium
Peridinium
Peridinlopsis
Chrysophyta
Dinobryon
Mallomonas
Chlorophyta
Ankistrodesmus
Botryococcus
Chlamydomonas
Cladophora
Cosmarium
Crucigenia
Dictyosphaerium
Dispora
ElakatothrJbc
Eudorina
Gemellicystis
Kirchneriella
1965-66
1973-75
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Lagerlieimia
Oocystis
Pandorina
Scenedesmus
Staurastrum
Tetraedron
Trepomonas
Ulothrix
Vqlvox
Bacillariophyta
Achnanthes
Asterionella
Caloneis
Cocconeis
Cyclotella
Cymbella
Diatoma
Epithemia
Fragillaria
Gomphonema
Hannaea
Melosira
Meridion
Navicula
Nitzschia
Opephora
Rhizosolenia
Stephanodiscus
Synedra
Tabellaria
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
-------
IN3
1966
1974
Figure 11. Copepod levels.
-------
ro
0.2
0.1
0.2
0.1
£ 0-2
| 0.1
t—•
x.
0.2
0.1
0.2
0.1
Q
STfl 1
STfl 2
-STfl 3
.STfl
.STfl 5
1966
1967
A.
999
V
1968
1969
1970
TERR
1971
1972
1973
1974
Figure 12. Cladoceran levels
-------
Two copepods are common in Kootenay Lake, Cyclops bicuspidatus. and
Diaptomus ashlandi, and three cladocerans, Daphnia galeata, Diaphanosoma
leuchtenbergianum, and Bosmina coregoni. A detailed evaluation of
the seasonal reproductive activity of each of these five species
was made. Figure 13 presents the total densities of each species
at Station 2. This is followed in Figure 14 by a summary of those
cladocerans with eggs and copepods with egg sacks. Note that during
1966-69, Diaptomus reached essentially the same level each summer.
Cyclops on the other hand appeared to be increasing during the three-
summer study. The years 1971-73 saw increasing summer peaks of
Diaptomus as well as Cyclops. Daphnia reached substantial densities
during 1967, but was relatively low during 1966 and 1968. This genus
has continued to decline and during 1971-74 was essentially absent
from the lake. Diaphano somat in contrast, has increased markedly
during 1971-73, reaching levels second only to Diaptomus. Bosmina
never has contributed substantially to the total zooplankton pop-
ulation, but recently it has been as abundant as Daphnia. These
changes could be induced oy selective predation by landlocked sockeye
salmon (Brooks and Dodson3", Brooks31, Galbraith3^, Warshaw33, Wells31f)
or by competition among zooplankters (Hall jet al..3% Hazelwood and
Parker36).
Although the total copepod density did not vary considerably from
summer to summer during 1966-68, a larger number of individuals
were gravid during 1967 than during the other two years. Further-
more, even though 1972 was a year of high Diaptomus populations,
only a relatively small proportion of the population was carrying
egg sacks. Cladocerans also, as exemplified by D iaphano s oma,
exhibited a rather erratic relationship between the size of the
population and the number of gravid members in it. This may be
due to differences in number of young per brood. Other possibil-
ities include lower mid-summer natural mortality rates as well as
lower grazing pressure. Information comparable to that given for
Station 2 in Figures 13 and 14 is given in Figures 15 and 16 for
Station 3, Figures 17 and 18 for Station 4, and in Figures 19 and
20 for Station 5. Recall that Station 5 is located slightly north
of the point where water leaves the lake via the west arm.
Cladoceran populations are remarkably similar at all four Stations,
although the copepod populations at Station 5 do not appear to
follow the same pattern as those at Stations 2-4 during 1971-74.
Noteworthy is the fact that the population of Diaptomus was
approximately twice as high in 1973 at Stations 2-4 than at
Station 5.
23
-------
PG
50 -
1966
1967
1968
1969
1970
TEflR
1971
1972
1973
1974
Figure 13. Total zooplankton densities at Station 2.
-------
5
1966
1967
1968
1969
1970
YEflR
1971
1972
1973
1974
Figure 14. Gravid zooplankton densities at Station 2.
-------
ro
en
1966
1967
1968
1969
1970
YEflR
1971
1972
1973
Figure 15. Total zooplankton densities at Station 3.
-------
ro
cc
UJ
CO
0
3
0
0.3
0
1966
1967
1968
1969
1970
YEflR
1971
1972
1973
1974
Figure 16. Gravid zooplankton densities at Station 3.
-------
ro
oo
50
E
fc
30
20
10
0
30
20
10
1966
1967
1968
1969
1970
YEflR
1971
1972
1973
1974
Figure 17. Total zooplankton densities at Station 4.
-------
ro
10
1966
1967
1968
1969
1970
YEPR
1971
1972
1973
1974
Figure 18. Gravid zooplankton densities at Station 4.
-------
U)
3
50
40
30
20
10
0
30
20
10
0
3
1966
1967
1968
1969
1970
YEflR
1971
1972
1973
1971*
Figure 19. Total zooplankton densities at Station 5.
-------
0.3
1966
1967
1968
1969
1970
TEflR
1971
1972
1973
1974
Figure 20. Gravid zooplankton densities at Station 5.
-------
SECTION V
SIMULATION MODEL
PARAMETER ESTIMATION
All models of ecological systems contain various rate and proportionality
constants, as well as other coefficients, that must be given specific
values before simulated results can be generated. Many of these
parameters may be estimated from field or laboratory observations, and
others taken from similar situations reported in the literature.
Unfortunately some quantities cannot be measured, although they exist
and are conceptually important in model development. In these cases,
it becomes desirable to obtain estimates by fitting part or all of the
model to data gathered from the system under study. The general
procedure follows .
Suppose the model system is described by the set of differential
equations
7i - fi , I - l,...,n (2)
where y' is the time derivative of the i state variable y, an element
of the vector y_, and £ is a vector of parameters. For fixed £, the
system may be viewed as consisting of ordinary differential equations.
If, on the other hand, £ is allowed to vary, y! is more appropriately
viewed as the partial differential equation
The effect of changing p^ (j = l,...,m) on the solution y of this
equation is found by first writing
8y m
3(-g~-)/3pj = |(9fi/3yk)(3yk/3pj) + 3fi/9pj-
But the order of differentiation can be reversed yielding
. (5)
V9P/' k=l
Now let 9y,/9p., the sensitivity of yfc to p , be designated as
and rewrite as
m
= ^(8f./8yt) u, + 3f./3p.. (6)
k=l
32
-------
Taken over all i and j, this produces a system of linear differential
equations in u that can be solved in concert with the original system
describing state variable behavior through time.
With this background the approach used to estimate £ proceeds along
the following lines. Suppose £ are the observed values of _y_ at
sampling time T. Suppose further, that an objective function of the
form
II
Ti
- jt>
(7)
or
Ti
Ti
(8)
(9)
is selected for minimization (others could be justified including the
use of y . in the denominator to reduce computation) . Now differentiate
D with respect to p , D say, and expand in a Taylor Series around
initial approximations
^(2+ ApJ - J, ApJ(3Dk/3p.J)
solution) to give
= 0.
(10)
The resulting system is linear in, and can be solved repeatedly for
A£ to update £. This approach should be recognized as Newton's method.
Still other techniques can be applied including conjugate gradients
and steepest descent (see, e.g., Powell37).
Now observe, for example, that
D = 3D/9p (11)
Ti
and D (£ + A£>
1C J- 1
y1)aukj/3p.J -
0.
(12)
Here one is required to compute values of the second derivatives of y.
with respect to all p . > that is , a massive system of n state variable
equations (differential) plus nm sensitivity equations plus nm2 second
derivative equations must be solved at every T, usually numerically.
The last ^
in D and
, .
set of nm^ equations can be avoided by expanding y. about p_
finding the derivative of D with respect to Ap (D' say) :
33
-------
Ti k j
and Dl(p + Ap) * IU&. - $Ap a )» - 0, L - 1,...,m. (14)
J LT1 i k kncj ^
To avoid inverting ill-conditioned matrices, the p should be scaled so
that all parameters are of the same order of magnitude. Furthermore,
if the magnitudes of the state variables are not of the same order, care
must be taken to scale the n components of the objective function D.
VARIABLES INCLUDED
Parker38*39'1'2 has attempted to create a reliable series of predictive
models that relate the state and driving variables in Kootenay Lake.
The variables used in the most recent version of the model (Parker1)
are given in Table 3.
Table 3. VARIABLES USED IN THE NUTRIENT PLANKTON
MODEL OF THE MIXED LAYER IN KOOTENAY LAKE
Driving Variables State Variables
Inputs Summer algal assemblage (group 1)
Volume flow Spring algal assemblage (group 2)
Inorganic phosphate Inorganic phosphate
Nitrate Nitrate
Ammonium Ammonium
Copepods
Temperature Cladocerans
Solar Radiation
Transparency
GROWTH AND DEATH RATES
Dynamic models based on differential equations require mathematical
descriptions of growth and death processes in terms related to
the driving variables. The growth rate of the two algal groups
34
-------
considered (spring and summer assemblages) were made functions of
solar radiation, temperature, and the three nutrients: phosphate,
nitrate, and ammonium. In addition, mortality rates due to natural
causes and predation were also functions of temperature. Cladoceran
and copepod growth rate was largely a function of grazing activity
and temperature, mortality rates largely attributed to temperature.
Specific functions selected to relate algal growth to light intensity
and temperature are discussed in Parker1'^. Monod relationships
were used to describe dependence on the nutrients, with pooled
nitrogen used as a single variable. All of the underlying relation-
ships used to describe growth and mortality rates are given in Table
4. In addition, the differential equations used to describe the
seven state variables in the system are tabulated. Note that hori-
zontal transport due to incoming Kootenay River water, as well as
vertical transfer, have been incorporated in these equations.
Essentially, the differential equations for the three nutrients are
mass balance forms which assume that all of the nutrients Ingested
but not incorporated into new zooplankton are immediately recycled.
Also any nitrate that has been reduced and used by phytoplankton
for growth will be returned to the system as ammonium. Specific
values used for all of the constants shown in Table 4 are summarized
in Table 5.
COMPARISON WITH OBSERVED VALUES
The model outlined in Table 4 was developed primarily from informa-
tion gathered during the period 1966-69 while phosphate input was
high. The primary objective of the modeling effort was to predict
the consequences of reduced phosphate input. Once the model was
completed and "tuned" to 1966-69 data (Parker 1>39), it was applied
using input data from 1971-74. Only one constant was changed in
Table 5 for this application. For use over 1966-69, the half-
saturation constant for phosphate was 1.0 for both algal groups.
This is, of course, a very high figure, its use being consistent
only with very high ambient phosphate levels prior to 1969. It
was assumed that phytoplankton in the system would slowly adjust
to the new phosphate levels by gradually reducing the half-satur-
ation constant to a more typical value. A review of model behavior
under different assumptions of reduced half-saturation constant
suggested as appropriate a linear decrease from 1.0 in 1971 to
0.2 in 1974. Failure to reduce the half-saturation constant resulted
in the virtual loss of the phytoplankton populations in 1973 and 1974,
Simulated values for inorganic phosphate, nitrate, and ammonium are
presented in Figures 21, 22, 23, respectively. These figures should
be compared to Figures 5, 8, and 9, respectively.
35
-------
Table 4. THE MODEL SYSTEM
Growth rate (G)
Mortality rates
Natural Predation
Algal group 1
Algal group 2
Cladocera (C.)
Copepoda (C
(I)f(T)g
TB(C11C1+C21C2)
C13T
C23T
C14TC1
C24TC2
I = light intensity
T - temperature
S. = nutrients
f±(T) = [(T/T8l)exp(l-T/Tal)
,°1
B = expt-k(A1+A2)]
' A1(GA1 '
Differential equations
- T3Ai/SY
Y
"l"
A2(GA2 - ^2 - M2A2> ' VSV9Y
ci(Gci - Mici - 'W - vaci/3Y
C//n \r \f \ W^f* J *^V
2( Cl ~ H.C2 ~ 2C2' ~ V3C2/3Y
q(A1M2Al + A2M2A2 - C^ - C2GC2) - qC^G^ + ^^
m31ll/9Z ' V3N1/3Y
A2M2A2) - n2(ClGCl +.C2GC2> " nl(A!GAl * A2GA2)N3/(N2 + N3)
-------
sin i
10
5
10
5
10
SIR 2
SIR 3
CO
STfl
10
STP 5
10
1966 1967 1968 1969
1970
TEflR
1971 1972 1973 197U
Figure 21. Simulated soluble inorganic phosphate concentrations.
-------
Table 5. CONSTANTS USED IN THE SIMULATION MODEL
an „ 35. CI;L = 0.70 Kn = K12 = 1*
a12 = 0.02 = a21 c12 =0.10 K21 = K22 = 6
&21 = 8- C13 =0.03 q = 0.3
IS1 = IS2 = 1<6 C14 = !-5 nl = 2'°
TS1 = 20. cn = 0>3 n2 „ 5.0
TS2 = 11. c22 = 0.15 m = 0.15 x 103
a1 = 4.80 c23 = 0.015
a2 = 1.45 C24 = °*5
bl = b2 = 0.47
k =0.35
K12 decreased linearly from 1 to 0.2 during 1972-74,
The agreement between predicted and observed values for phosphate
is reasonably good except during the summers of 1966 and 1967.
In 1966 there was an influx of phosphate in the Kootenay River
which was not accounted for by the model. Still, there was more
spring phytoplankton (group 2) model growth than in the lake.
Nitrogen metabolism in the lake is assessed largely in terms of
nitrate and ammonium concentrations. Examination of Figure 22
and the corresponding Figure 8 for observed values indicates that
the largest deficiency in model output for the period 1966-69
is that the simulated values are too low at times when the
observed nitrate concentrations are high in the lake, resulting
in model values being drawn down too low during mid-summer (see
for example 1968 and the spring of 1969). During 1971-74 the
model is not as responsive as appears necessary to mimic the actual
situation in Kootenay Lake. For example predicted winter and
spring values for 1972 are low, and the summer values in 1972 and
1973 appear too high. Examination of Figures 23 and 9 shows model
38
-------
GO
ID
1966
1967
1968
1969
1970
TERR
1971
1973
1973
1974
Figure 22. Simulated nitrate concentrations.
-------
values being consistent with observed values during periods of low
concentration but substantially too low during periods of high
ammonium concentration. This may be due to inadequate contributions
by zooplankton during peak periods of growth and contributions from
the sediments during the continuous period of winter circulation.
Figure 24 presents the simulated values for the two algal assem-
blages in Kootenay Lake. Comparison with Figure 10 shows that the
largest problem area lies in simulating the late summer algal
population. Assuming that the two assemblages provide a reasonable
breakdown for phytoplankton in Kootenay Lake, the model algal group
1 rarely ever reaches the levels attained in Kootenay Lake. Appar-
ently nutrient levels in the model are not adequate to produce the
required magnitude in late summer.
Simulated copepod densities are given in Figure 25, for comparison
with observed values in Figure 11. Although the model values are
reasonably good for 1966-1969, they are substantially low for 1971-
1974. This is particularly evident in 1972 and 1973. Again the
cause could lie with the model's failure to produce higher algal
peaks during August, thus producing an Inadequate food supply for
the copepods. The same problem appears to exist for the model
cladoceran population (Figures 26 and 12).
Observed quarterly means for variables used in the simulation
model are presented in Tables A1-A9 of the Appendix.
HYDRODYNAMIC CONSIDERATIONS
It is clear from the water retention time (566 days) that advection
plays a dominant role in the dynamic aspects of Kootenay Lake.
Figure 27 is indicative of the flow of the Kootenay River at its
point of entrance into the lake. During periods of peak volume
flow, the velocity of the river reaches approximately one meter
per second. If the river continued up the lake as a slug of water,
it would move over 600 kilometers in one week. Conversely, if
mixing were complete, a corresponding input of 1,500 cubic meters
per second would displace one-fortieth of the lake's volume in
one week. Since the lake is on the order of one hundred kilometers
long, this would suggest a weekly movement of about two and a half
kilometers. Obviously, the true value lies somewhere between
these two extremes. Because velocity measurements were not made
during much of the study (then only at Station 2), the model used
the lower bound to calculate expected velocities at each point.
This approach required use of a vertical transfer coefficient
40
-------
STfl 1
10
5
10
5
10
5
10
5
10
STfi 2
STfl 3
STfl
"N.
STfl 5
1966
1967
1968
1969
1970
TERR
1971
1972
1973
19714
Figure 23. Simulated ammonium concentrations.
-------
STfl 1
- STfl 2
ro
, - STfl 3
-STfl 4
-STfl 5
I
0
1966 1967 1968 1969
1970
YEflR
1971 1972 1973 1974
Figure 24. Simulated phytoplankton levels.
-------
0.2
0.1
0.2
0.1
0.2
SIR 1
-STR 2
-SIR 3
~ 0.1
>^J^J^^
0.2
-STfl 4
0.1
fs
o.;
-STR 5
0.1
L
'
1966
1967
1968
1969
1970
TERR
1971
1972
1973
1974
Figure 25. Simulated copepod levels.
-------
0.2 -SIR 1
0.1
0.2
0.1
0.2
0.1
0.2
0.1
0.2
0.1
0
-STfl 2
-SIR 3
STfl
-STfl 5
1966
1967
1968
1969
1970
YEflR
1971
1972
1973
19714
Figure 26. Simulated cladoceran levels.
-------
1.2
0.9
0.6
en
X X
0.3
XX
X
» X
X
300
600 900
CUBIC METERS / SECOND
1200
1500
Figure 27. Velocity and volume flow at Station 1.
-------
—1 3
of approximately 0.15 cm wk x 10 , even though studies of some lakes
(Jassby and Powell^0} suggest that this value may be low by an order
of magnitude; molecular transfer is approximately 0.07 cm wk""l x
103.
In order to establish the actual velocities existing in the lake
under different river flows, two additional sampling stations were
established near Station 2. The first was located midway between
Station 2 and the west bank, and the second was located midway
between Station 2 and the east bank. Drogue measurements were made
at one, five, and ten meters (occasionally at twenty meters) at
Station 2 as well as at the two new stations. Figure 28 presents
the results of these measurements for nine days during 1974-75.
These have been ordered according to river flow in an attempt to
focus on patterns of horizontal flow in the vicinity of Station
2. Recalling that the main axis of the lake lies in a north-
northwest south-southeast direction, it is apparent at once that
the turbulent motion in the lake is extreme. Only occasionally
are the directions of flow at the three or four sampling depths
coincident, although the magnitudes are similar. A further con-
solidation of the results is presented in Figure 29. Clearly the
mean northward flow is not coupled strongly with input volume.
There is, however, evidence to support a counterclockwise motion
in the lake, that is, the river water flows more strongly along
the east shore and returns to some extent along the west shore.
This provides for little motion in the center where Station 2 is
located. Examination of these results point to a rather haphazard
relationship between velocity and volume flow at Station 2, and at
other times southward flow occurs. In the model, of course, south-
ward flow was never permitted since the river waters were always
presumed to contribute to a northward displacement. Actual measure-
ments during periods of high flow (greater than 1,500 cubic meters
per second) suggest a northward flow at Station 2 of about 0.2
meters per second, approximately ten times more than that calculated
by volume displacement. More realism can be expected in the model
if the velocity component is modified accordingly, and the vertical
transfer coefficient changed as well.
46
-------
(M
|
RIVER FION
VOLUie 1262 ttVSEC
VELOCITT 0.74 H/SEC
omt 6 JUH TV
V
d
-P» >-
RIVER FLM
VOLUME «S5 M'/3EC
VELOCJTT O.K7 N/SEC
DB1E ITn JU. 74
RIVER FLOH
vaue siv
VELOCITT 0.24 M/SEC
WTE 29 WC 74
RIVER FLOW
VOLlfC 1503 HV3EC
VELOCITT 0.87 K/SEC
OflTE 15 JUN 711
VOLUHC
vaacin
DRTE 20
286 «'/SEC
o.w H/3EC
L TV
/ A
RIVER FLOH
VOLIflC
VELOCITT
ORTC 12
199 HV9EC
0.18
74
RIVER FLOH
VOLUME 1186 HV3EC
VELOCm 0.60 M/SEC
ORTE 22 JUH 7U
RIVER FLOH
VOLUME 489 HV9CC
VELOCITY 0.91
ORTE 30 JUL
RIVER FLOH
VOLUC IU09 tt'/3EC
VELOCITT 1.10 H/SEC
ORTC IS NflT 75
Figure 28. Velocity at four depths at Station 2.
-------
1,0
UJ
to
tn
-------
SECTION VI
REFERENCES
1. Parker, R. A. Some problems associated with computer simulation
of an ecological system. In M. S. Bartlett and R. W. Hiorns
(eds.) . The Mathematical Theory £f_ the Dynamics of_ Biological
Populations. Academic Press, London, pp. 269-288 (1973).
2. Parker, R. A. Capabilities and limitations of a nutrient-plankton
model. In E. J. Middlebrooks, D. H. Falkenborg, and T. E.
Maloney (eds.). Modeling the Eutrophication Process. Utah
Water Research Laboratory, Logan, pp. 121-130 (1973).
3. Parker, R. A. Some consequences of stochasticizing an ecological
system model. In P. van den Driessche (ed.). Mathematical
Problems in Biology» Victoria Conference. Lecture Notes in
Biomathematics. Springer-Verlag, Berlin. 2:174-183 (19747.
4. Parker, R. A. Empirical functions relating metabolic activity
in aquatic systems to environmental variables, ^J. Fish.
Res. Bd. Caii. 31(9): 1550-1552 (1974).
5. Parker, R. A. Stability of a nonautonomous ecosystem model.
Intern. J_. Systems Scl. 6:197-200 (1975).
6. Parker, R. A. The influence of environmental driving variables on
the dynamics of an aquatic ecosystem model. Verh. Intern. Verein
Limnol. 19:47-55 (1975). '
7. Haeler, A. D. Eutrophication of lakes by domestic drainage.
Ecology 28:383-395 (1947).
8. Edmondson, W. T., C. C. Anderson, and D. R. Peterson. Artificial
eutrophication of Lake Washington. Limnol. Oceanogr.
1:47-53 (1956).
9. Edmondson, W. T. Changes in Lake Washington following an
increase in the nutrient income. Verh. Int. Ver. Limnol.
14:167-175 (1961).
10. Beeton, A. M. Environmental changes in Lake Erie. Trans Amer.
Fish. Soc. 90:153-159 (1961).
11. Beeton, A. M. Eutrophication of St. Lawrence Great Lakes.
Limnol. Oceanogr. 10:240-254 (1965).
12. Verduin, J. Phytoplankton communities .of western Lake Erie
and the C02 and 0_ changes associated with them. Limnol.
Oceanogr. 5:372-380 (1960).
49
-------
13. Edmondson, W. T. Water-quality management and lake eutrophication:
The Lake Washington case. In T. H. Campbell and R. 0.
Sylvester (eds.). Watery Resources Management and Public
Policy. U. of Washington Press, Seattle (1968).
14. Edmondson, W. T. The present condition of Lake Washington.
Verh. Intern. Verein. Limnol. 18:284-291 (1972).
15. Ahlgren, I. Changes in Lake Norrviken after sewage diversion.
Verh. Intern. Verein. Limnol. 18:335-361 (1972).
16. Malueg, K. W., D. P. Larsen, D. W. Schults, and H. T. Mercier.
A six-year water, phosphorus, and nitrogen budget for
Shagawa Lake, Minnesota. J_. Environ. Quality 4:236-242 (1975),
17. Taylor, K. G. Limnological Studies on Kootenay Lake, British
Columbia, Canada. Ph.D. Thesis, Washington State
University, Pullman (1972).
18. Zyblut, E. R. Long-term changes in the limnology and
macrozooplankton of a large British Columbia lake.
£• Fish. Res. Bd. Can. 27:1239-1250 (1970).
19. Northcote, T. G. Some effects of mysid introduction and nutrient
enrichment on a large oligotrophic lake and its salmonids.
Verh. Intern. Verein. Limnol. 18:1096-1106 (1972).
20. Tailing, J. R., and D. Driver. Some problems in the estimation
of chlorophyll-^ in phytoplankton. In R. A. Vollenweider
(ed.), 1969, A Manual on Methods for Measuring Primary
Production in Aquatic Environments. F. A. Davis, Co.,
Philadelphia. 213 pp. (1969).
21. Strickland, J. D. H., and T. R. Parsons. A Manual of Sea-
water Analysis. Fish. Res. Bd. Can. Bull. No. 125.
Queens's Printer, Ottawa. 203 pp.(1965).
22. Standard Methods for the Examination of Water and Wastewater.
American Public Health Association, Washington, B.C.
874 pp. (1971).
23. Solorzano, L. Determination of ammonia in natural waters by
the phenolhypochlorite method. Limnol. Qceanogr.
14:799-801 (1969).
24. Chu, S. P. The influence of the mineral composition of the
medium on the growth of planktonic algae. Part II. The
influence of the concentration of inorganic N and
phosphate P. 'U. Ecol. 31:109-148 (1943).
50
-------
25. Goldberg, E. D., T. J. Walker, and A. Whisenand. Phosphate
utilization by diatoms. Biol. Bull. 101:274-284 (1951).
26. Fogg, G. E. Algal Cultures and Phytoplankton Ecology. U. of
Wisconsin Press, Madison. 126 pp. (1965).
27. Lackey, J. B. Plankton as Related to Nuisance Conditions in
Surface Waters. In F. R. Moulton and F. Hitzel (eds.)
Limnological Aspects of Water Supply and Waste Disposal.
American Association for the Advancement of Science,
Washington, D. C. 87 pp. (1949).
28. Thomas, W. H. Phytoplankton nutrient enrichment experiments off
Baja California and in the eastern equatorial Pacific Ocean.
£. Fish. Res. Bd. Can. 26:1133-1145 (1969).
29. Eppley, R. W., J. N. Rogers, and J. J. McCarthy. Half-saturation
constants for uptake of nitrate and ammonium by marine
phytoplankton. Limnol. Oceanogr. 14(6):912-920 (1969).
30. Brooks, J. L., and S. I. Dodson. Predation, body size, and
composition of plankton. Science 150:28-35 (1965).
31. Brooks, J. L. Eutrophication and changes in the composition
of the zooplankton. In Proceedings of a Symposium,
Eutrophication; Causes, Consequences, Correctives.
National Academy of Sciences, Washington, D. C.
pp. 236-255 (1969).
32. Galbraith, M. G., Jr. Size-selective predation on Daphnia by
Rainbow Trout and Yellow Perch. Trans. Amer. Fish. Soc.
96:1-10 (1967).
33. Warshaw, S. J. Effects of Alewives (Alosa pseudoharengus) on the
zooplankton of Lake Wononskopomuc, Connecticut. Ljjmol.
Oceanogr. 17:816-825 (1972).
34. Wells, L. Effects of Alewife predation on zooplankton populations
in Lake Michigan. Limnol. Oceanogr. 15:556-565 (1970).
35. Hall, D. J., W. E. Cooper, and E. E. Werner. An experimental
approach to the production dynamics and structure of
fresh-water animal communities. Limnol. Oceanogr.
15:839-928 (1970).
36. Hazlewood, D. H., and R. A. Parker. Population dynamics of
some freshwater zooplankton. Ecology 42:266-274 (1961).
51
-------
37. Powell, M. J. D. A FORTRAN subroutine for solving systems of
non-linear algebraic equations. UK Atomic Energy Research
Establishment Report R 5947, Harwell (1968).
38. Parker, R. A. Simulation of an aquatic ecosystem. Biometrics
24(4):803-821 (1968).
39. Parker, R. A. Estimation of aquatic ecosystem parameters.
Verh. Intern. Verein. Limnol. 18:257-263 (1972).
40. Jassby, A. and T. Powell. Vertical patterns of eddy diffusion
during stratification in Castle Lake, California. Limnol.
Oceanogr. 20(4):530-543 (1975).
52
-------
SECTION VII
APPENDIX
OBSERVED QUARTERLY MEANS FOR KOOTENAY LAKE
VARIABLES USED IN SIMULATION MODEL
Table Page
A-l Quarterly means (number of observations) for Kootenay 54
River flow (m3/sec) at station 1
A-2 Quarterly means (number of observations) of 55
temperature at 1 M (station 1) and mean temperature
over 10 M (stations 2-5)
A-3 Quarterly means (number of observations) of the 56
extinction coefficient (corrected for algal content)
over 5 M at station 1 and over 10 M at stations 2-5
A-4 Quarterly means (number of observations) of inorganic 57
phosphate as uM/L at 1 M (station 1) and mean
inorganic phosphate over 10 M (stations 2-5)
A-5 Quarterly means (number of observations of nitrate 58
concentration as \iVLfL (station 1) and mean nitrate
concentration over 10 M (stations 2-5)
A-6 Quarterly means (number of observations) of ammonium 59
concentration as yM/L at 1 M (station 1) and mean
ammonium concentration over 10 M (stations 2-5)
A-7 Quarterly means (number of observations) of mean 60
phytoplankton density over 10 M as mg/L (stations
2-5)
A-8 Quarterly means (number of observations) of mean 61
copepod density over 10 M as mg/L (stations 2-5)
A-9 Quarterly means (number of observations) of mean 62
cladoceran density over 10 M as mg/L (stations 2-5)
53
-------
Ln
*>
Table A-l. QUARTERLY MEANS (NUMBER QF OBSERVATIONS) FOR KOQTENAY RIVER FLOW On3/sec) AT STATION 1
Qtr.
1
2
3
4
1966
1300(4)
366(7)
139(3)
1967
207 (1)
1659 (4)
579(8)
163(4)
Year
1968 1969 1971
165(3) 143 C2)
1245(4) 1668(3)
507(8) 484(12)
249(4) 158(2)
1972
899(5)
657(13)
729(4)
1973
140 C3)
247(6)
326(10)
376(4)
1974
842(2)
1344(4)
-------
Table A-2. QUARTERLY MEANS (NUMBER OF OBSERVATIONS) OF TEMPERATURE AT 1 M (STATION 1)
AND MEAN TEMPERATURE OVER 10 M (STATIONS 2-5)
Ol
Ui
Sta.
Qtr.
1
2
3
4
5
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1966
8.6(4)
17.7(9)
9.9(5)
7.2(4)
16.9(10)
11.6(5)
7.4(4)
17.4(8)
12.7(4)
7.6(4)
17.3(9)
13.0(4)
1967
1.5(1)
10.2(4)
17.5(4)
11.6(4)
5.3(2)
8.7(5)
16.4(4)
13.9(4)
5.1(2)
9.0(5)
17.2(4)
11.3(4)
5.1(3)
9.1(5)
17.2(5)
14.9(3)
31.1(2)
17.3(5)
14.6(4)
1968
3.4(3)
8.0(4)
16.8(8)
5.5(4)
7.9(3)
16.4(8)
8.3(4)
3.4(2)
8.0(4)
16.3(7)
7.3(3)
4.3(1)
7.8(4)
16.4(7)
7.9(3)
3.2(3)
8.8(4)
16.6(8)
11.2(4)
Year
1969
1.5(1)
6.5(3)
2.8(1)
6.9(3)
2.8(2)
7.0(3)
2.3(1)
7.2(3)
2.6(2)
8.4(3)
1971
17.2(11)
5.2(2)
15.7(12)
9.7(2)
16.5(10)
10.5(2)
16.7(10)
10.5(2)
17.0(10)
13.1(1)
1972
8.7(5)
15.2(12)
11.4(4)
7.2(5)
15.5(12)
9.9(4)
7.3(5)
15.5(13)
10.6(4)
7.7(5)
15.6(13)
11.3(4)
7.6(3)
15.7(11)
10.6(3)
1973
0.8(2)
11.2(5)
17.2(8)
7.4(2)
3.2(2)
9.1(5)
17.6(8)
8.2(2)
3.9(3)
5.8(3)
3.6(2)
9.5(5)
17.3(8)
10.0(3)
3.4(1)
11.8(3)
17.1(8)
1974
1.6(1)
8.2(3)
2.2(1)
6.6(3)
1.8(1)
7.4(3)
9.9(2)
-------
Table A-3. QUARTERLY MEANS (NUMBERS OF OBSERVATIONS) OF THE EXTINCTION COEFFICIENT (CORRECTED
FOR ALGAL CONTENT) OVER 5 M AT STATION 1 AND OVER 10 M AT STATIONS 2-5
ui
•en
Sta.
Qtr.
1
2
3
4
5
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1966
1.92(2)
0.84(9)
0.85(9)
1.16(4)
0.25(10)
0.17(5)
0.74(4)
0.28(8)
0.23(4)
0.56(3)
0.34(9)
0.24(4)
1967
1.10(1)
3.13(4)
0.98(8)
0.52(4)
0.30(2)
1.15(6)
0.70(8)
0.29(4)
0.23(2)
0.73(6)
0.58(8)
0.42(4)
0.16(3)
0.60(6)
0.57(8)
0.26(3)
0.46(3)
0.56(8)
0.25(4)
1968
1.75(3)
2.05(2)
0.87(8)
0.67(3)
0.92(3)
0.36(8)
0.18(4)
0.17(2)
0.63(4)
0.33(7)
0.18(3)
0.10(1)
0.47(4)
0.32(7)
0.18(3)
0.25(3)
0.34(4)
0.27(8)
0.15(4)
Year
1969
1.54(2)
1.93(3)
0.23(1)
1.01(3)
1.10(2)
0.15(1)
0.91(2)
0.09(2)
0.60(2)
1971
0.70(11)
0.79(2)
0.34(12)
0.20(2)
0.31(10)
0.21(2)
0.32(10)
0.18(2)
0.27(10)
0.16(1)
1972
1.90(5)
1.20(12)
0.69(4)
0.37(5)
0.42(12)
0.29(4)
0.36(5)
0.41(13)
0.39(3)
0.34(5)
0.42(13)
0.28(3)
0.26(3)
0.37(11)
0.24(3)
1973
1.32(2)
1.32(6)
1.21(8)
1.75(2)
0.23(2)
0.32(5)
0.40(8)
0.31(2)
0.24(3)
0.24(3)
0.24(2)
0.40(5)
0.34(8)
0.22(2)
0.22(1)
0.33(3)
0.34(8)
0.22(2)
1974
2.58(2)
2.56(4)
0.36(2)
0.33(4)
0.27(2)
0.35(4)
-------
Table A-4. QUARTERLY MEANS (NUMBER OF OBSERVATIONS) OF INORGANIC PHOSPHATE AS yM/L
AT 1 M (STATION 1) AND MEAN INORGANIC PHOSPHATE OVER 10 M (STATIONS 2-5)
Ul
Sta.
Qtr.
1
2
3
4
5
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1966
4.01(4)
7.70(9)
11.54(5)
1.54(4)
3.38(10)
3.00(5)
1.62(4)
3.19(8)
2.73(4)
1.80(4)
2.54(9)
2.14(4)
1967
12.28(1)
2.58(5)
4,28(8)
11.01(4)
2.85(2)
2.40(6)
1.62(8)
2.17(4)
2.78(2)
2.29(6)
1.16(8)
1.99(4)
2.41(3)
2.16(6)
0.94(8)
1.27(3)
1.21(3)
0.51(8)
0.66(4)
1968
9.02(3)
3.65(4)
5.64(8)
3.38(4)
2.66(3)
3.02(8)
2.77(4)
2.91(2)
2.65(4)
2.63(7)
2.57(3)
2.56(1)
2.55(4)
2.37(7)
2.40(3)
2.18(3)
1.55(4)
1.67(8)
1.90(4)
Year
1969
4.73(2)
0.86(3)
3.04(1)
1.93(3)
2.75(2)
1.79(3)
2.93(1)
1.92(3)
2.45(2)
1.60(3)
1971
0.66(11)
1.23(2)
0.40(12)
0.32(2)
0.39(10)
0.38(2)
0.38(10)
0.47(2)
0.27(10)
0.00(1)
1972
0.13(5)
0.07(12)
0.25(4)
0.50(5)
0.06(12)
0.32(4)
0.48(5)
0.03(13)
0.16(4)
0.57(5)
0.03(13)
0.12(4)
0.50(3)
0.01(11)
0.10(3)
1973
1.27(2)
0.48(5)
0.26(8)
0.47(2)
0.60(2)
0.12(5)
0.05(8)
0.36(2)
0.55(3)
0.28(3)
0.54(2)
0.13(5)
0.03(5)
0.18(3)
0.30(1)
0.03(3)
0.02(8)
0.04(2)
1974
0.36(2)
0.24(4)
0.48(2)
0.32(4)
0.42(2)
0.29(4)
-------
Table A-5. QUARTERLY MEANS (NUMBER OF OBSERVATIONS) OF NITRATE CONCENTRATION
AS viM/L AT 1 M (STATION 1) AND MEAN NITRATE CONCENTRATION OVER 10 M (STATIONS 2-5)
Ln
00
Sta.
Qtr.
1
2
3
4
5
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1966
1.10(6)
1.20(5)
1.27(8)
1.67(5)
1.56(7)
1.36(4)
1.38(8)
1.43(4)
1967
31.36(1)
8.36(5)
1.88(8)
2.14(4)
3.57(2)
4.48(6)
2.07(8)
3.36(4)
3.71(2)
3.69(6)
2.07(8)
3.79(4)
3.38(3)
3.01(6)
1.99(8)
2.62(3)
1.79(3)
2.02(8)
3.05(4)
1968
7.93(3)
5.45(4)
3.70(8)
4.75(4)
5.31(3)
3.74(8)
4.67(4)
7.50(2)
4.45(4)
3.98(7)
4.60(3)
5.00(1)
3.98(4)
3.62(7)
4.02(3)
6.05(3)
2.53(4)
3.63(8)
4.09(4)
Year
1969
8.71(2)
8.36(3)
7.71(1)
7.52(3)
6.85(2)
7.83(3)
7.57(1)
7.24(3)
7.07(2)
6.26(3)
1971
5.44(11)
3.03(2)
6.24(12)
5.82(2)
5,71(10)
4.72(2)
5.70(10)
3.79(2)
4.46(10)
0.99(1)
1972
6.09(5)
3.39(12)
3.67(4)
8.06(5)
3.17(12)
4.95(4)
7.31(5)
2.91(13)
3.75(4)
7.33(5)
2.86(13)
4.02(4)
5.48(3)
2.36(11)
4.40(3)
1973
8.38(2)
2.90(6)
4.16(8)
5.99(2)
7.66(2)
2.70(5)
3.73(8)
5.92(2)
7.95(3)
4.13(3)
7.59(2)
3.42(5)
3.28(7)
4.10(3)
6.28(1)
1.79(3)
2.96(8)
3.94(2)
1974
17.14(2)
6.64(4)
8.62(2)
6.09(4)
8.71(2)
5.44(4)
-------
Table A-6. QUARTERLY MEANS (NUMBER OF OBSERVATIONS) OF AMMONIUM CONCENTRATION AS yM/L AT 1 M
(STATION 1) AND MEAN AMMONIUM CONCENTRATION OVER 10 M (STATIONS 2-5)
sta.
i
1
2
3
4
5
Qtr. 1966
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1967
6.56(1)
4.83(2)
0.82(2)
2.34(5)
2.99(4)
0.42(2)
1.82(6)
1.82(2)
2.71(3)
1.47(7)
4.20(3)
3.03(3)
3.35(5)
1.39(4)
1968
8.16(3)
2.61(2)
2.99(3)
12.35(4)
1.69(3)
1.03(7)
7.51(4)
1.06(1)
2.93(4)
1.52(4)
8.30(3)
0.14(1)
1.66(4)
1.48(5)
7.81(3)
0.61(3)
1.94(4)
0.76(5)
7.81(4)
Year
1969
6.03(2)
1.43(3)
4.02(1)
1.41(3)
7.40(2)
1.03(3)
2.19(1)
1.37(3)
8.64(2)
1.41(3)
1971
5.55(11)
0.90(2)
3.59(12)
0.91(2)
4.50(10)
0.72(2)
4.90(10)
0.77(2)
3.16(10)
1.02(1)
1972
1.91(5)
1.47(12)
2.19(4)
1.61(5)
1.14(12)
2.27(4)
1.85(5)
1.34(13)
2.18(4)
1.54(5)
1.79(13)
2.05(4)
1.25(3)
1.55(11)
2.41(3)
1973
6.00(2)
1.97(6)
1.65(8)
3.25(2)
4.53(2)
2.14(5)
1.34(8)
1.18(2)
2.62(3)
2.24(3)
6.57(2)
2.21(5)
1.32(7)
1.11(3)
5.10(1)
2.08(3)
1.24(8)
1.33(2)
1974
1.95(2)
3.55(4)
1.40(2)
1.65(4)
1.59(2)
1.78(4)
-------
Table A-7. QUARTERLY MEANS (NUMBER OF OBSERVATIONS) OF MEAN PHYTOPLANKTON
DENSITY OVER 10 M AS mg/L {STATIONS 2-5)
Sta.
Qtr.
2
3
4
5
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1966
0.64(4)
1.30(10)
0.42(5)
0.26(4)
1.14(8)
0.36
0.22(4)
0.87(9)
1.00(4)
1967
0.20(1)
0.90(6)
0.61(8)
0.60(4)
0.10(1)
0.54(5)
0.41(8)
0.61(4)
0.70(1)
0.60(4)
0.46(8)
0.45(3)
0.35(3)
0.49(8)
0.54(4)
1968
0.90(3)
0.77(8)
0.79(4)
0.42(2)
0.96(4)
0.59(6)
0.83(3)
0.55(1)
0.81(4)
0.58(6)
0.98(3)
0.47(3)
0.88(4)
0.57(7)
0.71(4)
Year
1969
0.05(1)
1.42(3)
0.75(2)
1.52(3)
0.50(1)
1.32(3)
0.90(2)
1.68(3)
1971
0.77(12)
0.67(2)
1.31(10)
0.55(2)
0.86(10)
0.73(2)
1.08(10)
1.48(1)
1972
0.58(5)
0.76(12)
0.37(4)
0.53(5)
0.79(13)
0.47(4)
0.37(5)
0.87(13)
0.43(4)
0.68(3)
0.93(11)
0.54(3)
1973
0.17(2)
1.00(5)
0.73(8)
0.32(2)
0.30(3)
0.85(3)
0.32(2)
0.75(5)
0.53(7)
0.49(3)
0.35(1)
0.73(3)
0.38(8)
0.58(2)
1974
0.20(2)
0.63(4)
0.30(2)
0.87(4)
-------
Table A-8. QUARTERLY MEANS (NUMBER OF OBSERVATIONS) OF MEAN COPEPOD DENSITY
OVER 10 M AS mg/L (STATIONS 2-5)
Sta.
2
3
4
5
Qtr.
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1966
0.017(3)
0.033(9)
0.029(4)
0.007(3)
0.041(7)
0.636(3)
0.013(3)
0.063(9)
0.037(3)
1967
0.008(2)
0.039(6)
0.075(8)
0.046(4)
0.013(1)
0.030(6)
0.076(8)
0.035(4)
0.010(3)
0.041(6)
0.057(8)
0.044(3)
0.034(3)
0.099C8)
0.45C4)
1968
0.014(3)
0.068(7)
0.050(4)
0.007(2)
0.018(4)
0.067(6)
0.038(3)
0.006(1)
0.054(4)
0.111(6)
0.045(3)
0.062(3)
0.109(8)
0.044(4)
Year
1969
0.002(1)
0.002(1)
0.009(2)
0.005(3)
0.026(1)
0.004(3)
0.021(3)
1971
0.066(11)
0.054(2)
0.068(10)
0.073(2)
0.083(9)
0.042(2)
0.154(10)
0.130(1)
1972
0.018(4)
0.076(12)
0.025(4)
0.008(5)
0.077(13)
0.066(4)
0.018(4)
0.084(13)
0.050(4)
0.010(3)
0.076(10)
0.050(3)
1973
0.014(2)
0.107(5)
0.155(7)
0.013(2)
0.018(3)
0.018(2)
0.018(2)
0.049(5)
0.119(6)
0.030(3)
0.007(1)
0.049(2)
0.096(6)
0.098(2)
1974
0.014(2)
0.005(4)
0.031(2)
0.014(4)
-------
Table A-9. QUARTERLY MEANS (NUMBER OF OBSERVATIONS) OF MEAN CLADOCERAN
DENSITY OVER 10 M AS mg/L (STATIONS 2-5) .
Ol
Sta.
Qtr.
2
3
4
5
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1966
0.000(3)
0.026(9)
0.002(4)
0.000(3)
0.035(7)
0.004(3)
0.000(3)
0.024(9)
0.003(3)
1967
0.000(2)
0.000(6)
0.060(8)
0.048(4)
0.000(1)
0.000(6)
0.036(8)
0.029(4)
0.000(3)
0.000(6)
0.050(8)
0.044(3)
0.000(3)
0.043(8)
0.011(4)
1968
0.000(3)
0.038(7)
0.010(4)
0.000(2)
0.000(4)
0.020(6)
0.001(3)
0.000(1)
0.000(4)
0.015(6)
0.001(3)
0.000(3)
0.010(8)
0.001(4)
Year
1969
0.000(1)
0.000(3)
0.000(2)
0.000(3)
0.000(1)
0.000(3)
0.000(3)
1971
0.015(11)
0.004(2)
0.012(10)
0.005(2)
0.014(9)
0.003(2)
0.020(10)
0.007(1)
1972
0.000(4)
0.016(12)
0.001(4)
0.000(5)
0.010(13)
0.007(4)
0.000(4)
0.009(13)
0.007(4)
0.000(3)
0.009(10)
0.001(3)
1973
0.000(2)
0.000(5)
0.070(7)
0.004(2)
0.000(3)
0.000(2)
0.000(2)
0.000(5)
0.063(6)
0.008(3)
0.000(1)
0.000(2)
0.026(6)
0.036(2)
1974
0.000(2)
0.000(4)
0.000(2)
0.000(4)
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/3-76-063
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Phosphate Reduction and Response of Plankton
Populations in Kootenay Lake
5. REPORT DATE
1976
May
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Richard A. Parker
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
I
Washington State University
Pullman, Washington 99163
10. PROGRAM ELEMENT NO.
1BA608
11. CONTRACT/GRANT NO.
R 800430
12. SPONSORING AGENCY NAME AND ADDRESS
Corvallis Environmental Research Laboratory
Environmental Protection Agency
200 S.W. 35th St. Corvallis, Oregon 97330
13. TYPE OF REPORT AND PERIOD COVERED
Final Report
14. SPONSORING AGENCY CODE
EPA-ORD
15. SUPPLEMENTARY NOTES
16. ABSTRACT
The purpose of this research project was to determine the effects of
reducing by 90 percent the soluble inorganic phosphate input to Kootenay
Lake, British Columbia, Canada, a significant link in the Columbia River
system governed by United States-Canadian treaty. Measurements on temperature,
transparency, inorganic and organic phosphate, nitrate, ammonium, chlorophyll a_,
copepods, and cladocerans were made during 1971-75, and compared with
observations made for three years prior to phosphate reduction in 1969. Only
a slight decrease in chlorophyll occurred, although there were important
changes in species composition. Total zooplankton levels were not affected,
however one genus (Daphnia) has virtually disappeared. Data indicate that
primary production in the lake was nitrogen limited prior to 1969, phosphate
limited two years later. These changes may also be related to the completion
^n 1972 of Libby Dam in Montana, upstream from the lake on the Kootenay River.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFJERS/OPEN ENDED TERMS
c. COS AT I Field/Group
Plankton response
phosphate reduction
Kootenay Lake
Q8H
8. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
72
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
63
U.S. GOVERNMENT PRINTING OFFICE: I976-S97.3I6/95 REGION 10
------- |