EPA-600/3-82-096
December 1982
EXPERIMENTS AND EXPERIENCES IN B1 OMAN IPULATI ON
STUDIES OF BIOLOGICAL WAYS TO REDUCE ALGAL
ABUNDANCE AND ELIMINATE BLUE-GREENS.
by
1
Joseph Shapiro
and
Bruce Forsberg , Vincent Lamarra , Gunilla Lindmark ,
5	6	7
Michael Lynch , Eric Smelt2er , George Zoto .
R 803870
Project Officer
Charles Powers
Corvallis Environmental Research Laboratory
Corvallis, Oregon 97330
CORVALLIS ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CORVALLIS OREGON 9733O
1.	Limnological Research Center, University of Minnesota, Minneapolis
2.	Instituto Nacional de Amazonia, Manaus, Brazil
3.	Department of Wildlife Science, Utah State University, Logan
k.	Institute of Limnology, University of Lund, Lund, Sweden
5.	Department of Ecology, University of Illinois, Urbana
6.	Department of Water Resources and Environmental Engineering, Montpelier,
Vermont
7.	Research Department, New England Aquarium, Boston

-------
t
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing!
1. REPORT NO. 2.
EPA-600/3-82-096
3. RECIPIENT'S ACCESSION NO.
PB83 148 09 8
4. TITLE AND SUBTITLE
Experiments and Experiences in Biomanipulation--
Studies of Biological Ways to Reduce Algal Abundance and
Eliminate Blue-Greens
5. REPORT DATE
December 1982
5. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Joseph Shapiro- and Bruce Forsberg, Vincent Lamarra,
Gunilla Lindmark, Michael Lynch, Eric Smeltzer,George Zot
8. PERFORMING ORGANIZATION REPORT NO.
0
9. PERFORMING ORGANIZATION NAME AND ADDRESS
\imnological Research Center
University of Minnesota
Minneanolis, Minnesota
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
R 803870
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Corvallis, Oregon 97333
13. TYPE OF REPORT AND PERIOD COVERED
Report 107^—1 Q78
14. sponsoring"Agency code
EPA-600-02
15. SUPPLEMENTARY NOTES
Project Officer: Charles Powers, Environmental Research Laboratory, Corvallis, OR 97333
16. ABSTRACT
Studies have been done to find alternatives to restoring or managing lakes by control!inr
external sources of nutrients. The guiding principle has been to understand and use bio-
logical interactions within lakes. This process is called biomanipulation and it is
clear from the results that algal abundance and type can be varied substantially by one
or r.ore of the following procedures:
Elimination of benthivorous fish which recycle phosphorus from sediments.
\ Manipulations of algal populations by lowering pll, causing artificial
circulation.
Increasing abundance of larger herbivorous zooplankters by reducing predation
on them, by eliminating planktivores entirely or, by providing refuges from
planktivores.
17.	KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
b. identifiers/open enoed terms
c. COSAT! I'ield/Group



18. D'.STRIBUT ON STATEMENT
Release to public.
19. SECURITY CLASS (This Report,
Unclassified
21. NO. of PAGES
25?

20. SECURITY CLASS tThispagel
Unclassified
22. PR'CE
EPA Form 2220-1 (Rev. 4-77) poevioys e3«t:on is obsolete ^

-------
.DISCLAIMER
Although the research described in this report has been funded by the
U.S. Environmental Protection Agency through grant R-803870 to the University
of Minnesota, it has not been subjected to the Agency's oeer and policy review
and therefore does not necessarily reflect the views of the Anency and no
official endorsement should be inferred.

-------
THE AQUATIC FOOD CHAIN
(Not to scale)

PISCIVOROUS
FISH
eat
PLANKTIVO ROUS
FISH
eat
• ' .<
V-fcA
'fv ,H(
.'4"
NUTRIENT S
I
HERBIVORES
eat
I
ALGAE
use
~
nutrients
recycle
I
BENTHIVOROUS
FISH
iii

-------
ABSTRACT
Studies have been done to find alternatives to restoring or managing
lakes by controlling external sources of nutrients. The guiding principle
has been to understand and use biological interactions within lakes. This
process is called bi oman i pu1 ati on and it is clear from the results that algal
abundance and type can be varied substantially by one or more of the
following procedures.
1)	Elimination of benthivorous fish which recycle phosphorus from
sed iments.
2)	Manipulations of algal populations by lowering pH, causing
artificial circulation, or stimulating algal viruses.
3)	Increasing abundance of larger herbivorous zooplankters by reducing
predation on them, by eliminating planktivores entirely or, by providing
refuges from planktivores.
This report was submitted in fulfillment of grant R 803870 by Joseph
Shapiro, Ph.D., under the sponsorship of the U.S. Environmental Protection
Agency. This report covers a period from 9/15/75 to 9/15/78 and work was
completed as of 9/80.
iv

-------
CONTENTS
Fronti spiece	i i i
Abstract	i v
Contents	v
Ac know! edgements	vi
I ntroduct i on	1
Project Summary	k
I.	Biological Effects on the Si2e of the Nutrient Pool	12
II.	Algal Manipulations
A.	The Effects of pH and Carbon Dioxide on Algal Communities	30
B.	The Effects of Artificial circulation on Algal Populations	^6
C.	Effects of Environmental Stresses on the Relationship between
P lectonema boryanum and Cyanophage LPP-1	129
III.	Direct Manipulation of Zooplankton Populations	155
IV.	Manipulations of P1anktivorous Fish - Effects on Zooplankton and
Phytoplankton.
A.	Experimental Studies	158
B.	The Effects of Fish Toxicants on Phytoplankton
File Data	190
Effects of Rotenone in Wirth Lake	195
The Influence of Fish on the Abundance of Algae in Clear Lake..208
C.	Effects of Winter-kill on Minnesota Lakes in 1978-1979	210
D.	Effects of Winter-kill in Lake of the Isles	219
V.	The Role of Physica1-Chemical Conditions in Affecting
Algal Abundance	235
VI.	Conclusions and Recommendations	251
v

-------
ACKNOWLEDGEMENTS
Many people and organizations have contributed directly or indirectly to
these studies, and many are acknowledged in individual sections. Among the
others who deserve thanks are: the residents surrounding Lake Emily, Little
Lake Johanna and Twin Lake; the North Oaks Company for permitting us to use
Pleasant Pond; North Hennepin Community College for allowing us to use Loch
Loso; the Minneapolis Park and Recreation Board which cooperated in use of
lakes under its jurisdiction, as did the Minnesota Department of Natural
Resources; the Minnesota DNR which allowed use of its files. Preparation of
the report was helped greatly by Cora Barr, Edward Swain, Richard Darling and
Valerie Laser-Cartier. The several reviewers and Charles Powers are commended
for their fortitude and patience.
vi

-------
INTRODUCTION
For the last two decades or so limnologists appear to have been divided
into two groups seemingly working in different directions — the first to
understand the basic physical, chemical, and biological processes operating
in lakes, and the second to find ways of preventing, curing, or alleviating
eutrophication of lakes. Those working in the first direction generally seem
to look upon themselves as theoretical limnologists and by-and-large have
left the "problems" of lakes to the second group. Those interested in
eutrophication have for the most part taken a narrower view of lakes and have
concentrated their efforts on identifying limiting or potentially limiting
nutrients and their sources and on determining ways to reduce nutrient inputs
or concentrations. This distinction has continued despite such early
demonstrations as that by Hrbacek et a 1. (1961) that alterations in the fish
populations could result in fewer algae; and that by Symons et a 1. (1969)
that artificial circulation of stratified lakes could result in reduced
abundance of blue-green algae. In other words it has long been known that
key elements of eutrophic systems, indeed the very elements that are most
undesirable -- abundant algae with their resulting decreased transparency,
blue-green algae with their peculiarly undesirable characteristics — are not
strictly functions of nutrient inputs or concentrations. The possibility
exists to alleviate the symptoms of eutrophication through avenues other than
control of nutrient input from outside sources. The notable lack of
exploration of such alternative courses probably resulted from at least three
causes: the fact that Hrbacek et a 1. did not state their case forcefully!
the fact that Symons et al., and indeed many who used artificial circulation
subsequently, were unable to explain their results; and the fact of the
remarkable recovery of Lake Washington so energetically chronicled by
Edmondson (1977)- In effect nutrient manipulation seemed to work and the
need for alternative measures to restore lakes may not have been recognized.
It was in this spirit that we approached our study of the Minneapolis
Chain of Lakes (Shapiro and Pfannkuch 1973) > a group of five lakes fed by
urban storm drainage. To no one's surprise the eutrophic condition of these
lakes was to a great extent a function of the large nutrient input.
However, elimination of these inputs was and still is not possible for many
reasons — economic, energy requirements, etc. Also, various in-lake
treatments such as nutrient precipitation were not likely to be of use
because of the continuous nature of the nutrient inputs. Fortunately, at the
same time that we were studying the lakes and beginning to appreciate these
problems, we were doing other things. First, we were artificially
circulating lakes, e.g. Lake Calhoun. From this we learned that circulation,
although in this case it increased the abundance of blue-greens, can have
biological effects of potential significance to lake restoration. Notably
the circulation caused an increase in the number of large herbivorous
zooplankters and this, as shown by Hrbacek et a 1 . (19&1) , could result in
fewer algae. Second, we were investigating the effects of increased carbon
dioxide concentrations and of reduced pH on algal populations. From this we
learned that either treatment leads to an increase in green algae at the
expense of blue-greens, and we tentatively tied this to the sometimes success
of artificial circulation. Third, we were studying the likelihood that fish
removal by winter-kill or by fish toxicants could be beneficial because of
removal of benthivorous fish, which we speculated might act as nutrient pumps
(Lamarra 1975)- Soon afterward we looked at the other possibility — that
fish removal might be beneficial also by allowing herbivorous zooplankters to
increase in size and abundance. More recently in considering the mechanisms
1

-------
of circulation and the reasons for blue-green reduction we have studied the
possible role of cyanophage.
Thus there developed the possibility that a complex of methods other
than nutrient (N and/or P) manipulation per se could be used to help restore
lakes. As the methods all involved manipulating the biota in various ways,
the term "biomanipulation" was coined to describe the approach. Thus,
biomanipulat ion is not a method — it is an approach -- an attitude as it
were. The main precept is that control of outside sources of nutrients is
not the only way to restore lakes which are in a eutrophic condition, but
that there exist within lakes biological phenomena that act to increase
nutrient levels and to emphasize the effects of high nutrient levels and that
if we understand the phenomena we can use them to our advantage.
This report is a compilation of studies we have made on these biological
phenomena. They are, as with all ongoing research, in various stages of
resolution. Some have been wholly funded by the U.S. Environmental
Protection Agency; and some only partially, with the remainder coming from
the National Science Foundation and the Department of the Interior. Our
approach has been to use any and all methods to obtain information that could
bear on the role of the biota in establishing conditions in lakes that would
lead to methods for manipulating the biota in an appropriate fashion.
Consequently, included in this report are:
results of laboratory experiments,
results of small-scale field experiments,
results of large-scale field experiments,
detailed observations on lakes unintentionally manipulated
by wi nter-k ill,
detailed observations on lakes intentionally manipulated
by fish toxicants,
detailed observations on lakes not manipulated in any way,
selected observations on lakes studied as part of a mass survey,
selected observations on lakes from the files of the Minnesota
Department of Natural Resources, the Minnesota Pollution
Control Agency, and by others.
The manipulations to be described may be categorized as follows:
I.	Those involving biological effects on the size of the nutrient
pool .
II.	Those involving algal abundance and type without regard to the
nutrient pool. These include:
1.	manipulations of the algae themselves
2.	manipulations of zooplankters
3- manipulations of planktivorous fish
k. manipulations of piscivorous fish
5. manipulations of the physical environment.
REFERENCES
Edmondson, W.T.	1977* Trophic equilibrium of Lake Washington. U.S.
Environmental	Protection Agency. EPA-600/3-77*087.
Hrbacek, J., M.	Dvorakova, V. Korinek, and L. Prochaskova. 1961.
2

-------
on the species composition and
plankton association. Verb.
carp as a major contributor to
Internat. Verein. Limnol.
Demonstration of the effect of fish stock
the intensity of metabolism of the whole
Internat. Verein. Limnol. 1^:192-195.
Lamarra, V.A. 1975- Digestive activities of
the nutrient loading of lakes. Verh.
19:2^61-2468.
Shapiro, J. and H.O. Pfannkuch. 1973- The Minneapolis Chain of Lakes; a
study of urban drainage and its effects. Interim Rept. no. 9.
Limnological Research Center, University of Minnesota. 250 pp.
1969. Water quality behavior in reservoirs
research papers. U.S. Dept
Symons, J. M. and others,
compilation of published
and Welfare., Public Health Service. Cincinnati, Ohio.
Hea 1
616
th
pp.
a
Educat i on
3

-------
PROJECT SUMMARY
Introduction
Lake restoration ventures designed to reduce the abundance of undesirable
algae almost always are based on the premise that reduction of nutrient input
from external sources or from anoxic sediments is the key to success. Rarely
have biological interactions within lakes been exploited deliberately to
reduce or help in the reduction of such algal populations. However, consider-
ation of such an approach, termed biomanipulation, as opposed to nutrient
manipulation, indicates that it has great potential alone or in combination
with nutrient manipulation.
Figure 1 shows some of the possibilities. Note that although the end
goal is reduction of algal biomass, none of the possible manipulations involve
nutrients directly. Most deal with changing the quantitative and qualitative
relationships among the biota so that the desired end is achieved. It should
be evident that while some of the possible manipulations are more likely to
succeed than are others, most are likely to be more feasible and less
expensive than direct reduction of the nutrients. What we do not know is the
extent to which the manipulations can be successful, their duration of
effectivenessj or their freedom from unexpected consequences.
This report is a summary of work done on biomanipulations at the
Limnological Research Center up until 1980.
Among the possibilities for such manipulation that we have studied are:
I.	Elimination of bottom-feeding fish which, through their feeding activi-
ties, increase the nutrient concentrations and thereby the abundance of
algae in lakes in which they abound;
II.	Manipulations of algal populations to change species composition and/or
reduce abundance by (A) lowering pH, (B) causing artificial circulation,
(C) stimulating activity of viruses that attack blue-green algae;
III.	Direct manipulations of zooplankton populations to increase abundance of
herbivorous species and therefore grazing on the algae;
IV.	Indirect manipulations of zooplankton herbivores by manipulating their
predators — planktivorous fish — by (A) experimental additions, (B)
elimination of planktivores by rotenone treatment, and (C) elimination of
planktivores by winter kill;
V.	Modifications in oxygen concentrations that may lead to large changes in
algal populations via their effects on refuges for zooplankters.
A

-------
BIOMANIPUL ATION
formation of
refuges
predation
blue-green to
green shift
ncrease
reduction of
CIRCULATION
AND/OR AERATION
ADDITION OF
PISCIVOROUS FISH
of large herbivores
ELIMINATION OF
BENTHIVOROUS FISH
FISH TOXINS
FISH DISEASES
WINTER-KILL
CAPTURE
Reduction of Algal Biomass
Increase in Transparency
Fig.1. Some aspects of biomanipulation.
The central goals of reduction of algal biomass,and
increased transparency, are achieved through a variety
of manipulations such as those shown in darker type.
Mechanisms are indicated in lighter type.
5

-------
' E1imi nat ion of bottom feedinq f i sh
The work of Lamarra showed that bottom-feeding fish excrete phosphorus
and nitrogen compounds and that the rate of excretion depends on the fish
size, the temperature and the type of lake sediment. Lamarra hypothesized
that such input could be a significant contribution to the total nutrient
loading of lakes. An opportunity to test this idea arose when the Minnesota
Department of Natural Resources decided to restructure the fish population
in Lake Marion, a shallow (mean depth = I.98 meters) large (172 ha) lake in
south central Minnesota. Estimates of the fish population and its
characteristics were made before the rotenone treatment using mark-recapture
methods, and after the treatment using shore census of the dead fish. The
latter method gave much higher values. Using these values, annual inputs for
fish excretion were calculated at 88 mg/m2 per year for phosphorus and 270
mg/m2 per year for nitrogen. The Minnesota Pollution Control Agency had
previously calculated the total phosphorus loading rate to the lake from its
primarily agricultural watershed as 84 mg P/m2 per year. Therefore, the fish
provide about half the phosphorus input to the lake. Based on two years'
study of the lake before treatment, predictions have been made regarding
reductions in algal biomass and productivity and increases in transparency
expected from reduction in total phosphorus. Data to test these predictions
have been collected and are in process of analysis.
'' A Man i pu1 at i on of a 1 gal popu1 at i ons by pH 1ower i nq
Preliminary studies had confirmed the hypothesis that green algae are
favored over blue-greens at lower pH values. In these preliminary
experiments, nutrients plus C02, or nutrients plus acid, added to the waters
of Lake Emily caused green algae to become dominant, while adding nutrients
alone caused the blue-greens to increase. A total of "JO experiments have now
been done in the field and in the laboratory and the conclusions are as
follows:
a)	The phenomenon is reproducible. In every case (19) in Lake Emily where
C02 was added with nitrogen and phosphorus, the blue-green to green shift
occurred.
b)	The phenomenon is not limited to Lake Emily. Ten other lake waters have
been tested, and the shift took place in all of them.
c)	Additions of HC1 generally had the same effect as additions of C02, but
exceptions occurred.
d)	The shift occurred whether field experiments were begun in June or late
September.
e)	Field experiments with pH controlled enclosures showed that the shift
from blue-greens to greens occurred at pH values of 5*5 ~ 8.5. when C02 was
used, and at pH values of 5-5 - 7-5 when HC1 was used.
f)	In most of the experiments the green algae resulting from the shift were
Scenedesmus and Ch1orella, and in one experiment there were 22 species and
subspecies of Scenedesmus.
g)	The shift from blue-greens to greens seemed to be more rapid in spring
and fall than in summer. This may be related to the size of the inoculum of
greens as experiments with different sizes of inoculum showed the rate
increased with a higher initial proportion of greens.
h)	The shift often seemed to occur precipitously, and it involved almost all
species of blue-greens in the lakes tested. However, some blue-greens
remained, for if the pH was raised, they again began to regain dominance.
i)	The reason for the shift is obscure. It may involve competition between
6

-------
the two types of algae, but the increase of the greens occurs after the
decrease of the blue-greens. Thus the two phenomena are distinct. As the
blue-greens disappear, phosphate and ammonia are found in the water, but
disappear as the greens grow.
j) One possibility is that the blue-greens are affected by algal viruses at
lowered pH. This is suggested by the manner in which the filaments break up.
The role of nutrients appears to be important to the greens. If arsenate,
which reduces phosphate uptake by the greens, is added the shift is delayed
or prevented. Chlorine additions at high pH also cause the shift, presumably
by a different mechanism.
I I B Hani pulat i on of algal populations by artificial ci rculation of
1 akes
Artificial circulation, frequently termed aeration, has been a lake
restoration technique of limited value probably because of the lack of a
proper theoretical framework for its use. To remedy this situation we
constructed two such frameworks, one to explain the shift from blue-green
algae to greens that is often observed, and one to explain the diminution in
algal biomass that sometimes occurs. The hypotheses were tested in two lakes
in a series of eight experiments utilizing a total of J(> enclosures that were
1 meter in diameter and extended from the surface through the thermocline to
a depth of 8 meters. Some of the "bags" were open at the bottom while others
were sealed and filled with surface water only, so that following temperature
stratification in them by conduction the chemistry of their bottom waters
could be adjusted. The enclosures were circulated by air to different depth
and with different intensities and the results determined.
1. Species composition
Response of the phytoplankton at the species level appeared to depend
primarily on changes in water chemistry in the euphotic zone during mixing.
In the eutrophic lake with lower alkalinity, deep rapid mixing which
increased nitrogen, phosphorus, and C02 levels in the euphotic 2one led to a
shift from blue-green algae to greens and diatoms. Deep slow mixing, which
also increased nitrogen, phosphorus, and C02 levels in the euphotic zone
resulted in increases in blue-greens. However, in the case of rapid mixing,
carbon dioxide was introduced into the euphotic 2one rapidly enough to lower
pH values, while in the slow-mix enclosures pH remained high. This result is
consistent with those of the previous sections.
Not only did the green algae benefit from rapid circulation, but diatoms
also showed increases. As this occurred during shallow mixing as well,
without the increase in nutrients and C02, the mechanism must be different -
possibly related to turbulence preventing the diatoms from sinking out of the
euphotic zone.
Circulation in the higher alkalinity eutrophic lake was not as
successful in shifting algal species composition. Not only was the water more
buffered against pH change, but the concentration of C02 in the hypolimnion
was lower. Consequently, pH values did not decrease significantly during
circulation. Furthermore the lake contained a metalimnetic population of
Osc i11ator i a rubescens which generally increased in abundance in proportion
to the total phosphorus increases in the euphotic zone resulting from mixing,
and probably also as a result of increased light and temperature.
7

-------
2. Community response
Data from the circulation experiments were also used to construct and
test a mathematical model describing response of the total algal community.
The most important variables in the model were found to be Zm, the mixed
depth, and TP, total phosphorus, which have opposite effects on the maximum
concentration of chlorophyll in the mixed layer during circulation. An
increase in Zm causes a decrease in chlorophyll and an increase in TP causes
an increase in chlorophyll. The relative magnitude of these changes therefore
determines whether the chlorophyll concentration will increase or decrease.
Furthermore the size of the chlorophyll change wi11 be a function of such
other factors as the N/P ratio, as at higher ratios the yield of
chlorophy11/P is greater; the loss rate as a result of death, sinking, and
grazing - factors also affected by circulation; and the extinction
coefficient, Ew, of the water as it is determined by non-algal substances
dissolved or suspended in it.
These results demonstrate why without an adequate theoretical framework
it has been difficult to predict and/or understand the qualitative and
quantitative changes that have occurred in lakes during circulation. The
results obtained here will be useful in designing future attempts.
II c. Han i pu1 at i on of a 1qa1 popu1 at i ons through the use of spec i f i c
vi ruses
Attempts have been made without success to control blue-green algae in
lakes by utilizing the known capacity of several viruses to lyse them. As
part of an investigation into the mechanism of the shift from blue-greens to
greens at lowered pH, we studied, in the laboratory, the relationship between
algal viruses and their hosts. We used the blue-green P1ectonema boryanum
and the Cyanophage LPP-1. Among the factors studied were: 1) the effect of
pH alterations, 2) the effect of algal host age and density, 3) the effect
of nutrient concentration, and k) the effect of other algal species. The
most relevant observation was that the alga thrives at both high (>10.5) and
low (<7.5) pH values in the absence of the virus but it is lysed at the lower
values in the presence of the virus. The implication is that lowering pH by
artificial circulation of a lake may result in lysis of the blue-greens by
viruses present in the system. However, more work needs to be done to
determine whether this is what actually happens
I I I Di rect mani pulat ion of zooplankton populations
Decreases of algal abundance could result from increases in herbivore
abundance. Therefore, experiments were conducted on the feasibility of using
pantothenic acid, previously reported to be effective, to increase Daphnia
abundance. Results were negative and it is concluded that such
manipulations, including attempts to add herbivores directly to lakes, would
be ineffectual. It is pointed out that certain pesticides may be exceedingly
effective in eliminating Daphnia, however.
IV Ind i rect man i pulat i on of zooplankton populat i ons via piankt i vorous
f i sh
A. Experimental
Experiments in which different densities of planktivorous fish were
8

-------
studied for their effects on zooplankton and algal populations were carried
out in enclosures, divided ponds and whole ponds.
1. Enclosure experiments
In the enclosure experiments (1 meter diameter, 1 meter deep)
additions of bluegill sunfish eliminated such herbivores as Daphnia pulex and
Daphni a ga1eata while allowing the smaller species Daphn i a amb i qua and
Daphnia parvula, to develop. Effects on the algae were dramatic with algal
biomass in the enclosures with fish averaging, in one series, 16-fold that in
the enclosures without fish. In some of the experiments, as fish predation
intensity increased, filamentous blue-greens became relatively more abundant.
In these experiments in the absence of fish predation the predominant algae
were greens. The effects on algal biomass were the result of fish predation
on the zooplankton, rather than fertilization of the enclosures by fish
excreta. This was shown by experiments in which nutrients were added
intentionally to the enclosures.
2.	Divided pond experiments
These were done by dividing a small pond (0-5 ha) with polyethylene
sheeting. One half contained numerous fathead minnows and the other half
contained a few larger fish. As in the enclosures, Daphnia pulex was
eliminated in the half containing the minnows and Daphnia ambiqua and
Daphnia parvula appeared. Consequently the algal biomass in this half of the
pond averaged five times as high as that in the other half during July. This
was not a result of greater phosphorus availability as the phytopiankton/P
ratio was an average of 3-^ times as high in the "minnow" half.
3.	Whole pond experiments
Two ponds side by side, which normally winter-killed, were used. One
was stocked with mature perch and bluegill sunfish. One year later the ponds
differed greatly in their zooplankton communities. In the stocked one,
Daphnia pulex was absent, chlorophyll concentrations were high and
transparency was low. The pond not stocked had large populations of Daphnia
pulex, generally low chlorophyll concentrations and high levels of
transparency.
During these investigations it was discovered that under certain
circumstances the presence of Daphnia pulex appears to result in an abundance
of Aphanizomenon flos-aquae in the form of large flakes not grazeable by the
Daphnia. This has been noted in other studies as described later.
IV B. Man i pulat i on of planktivore populat ions wi th f i sh tox i ns.
1. Previous experiences in Minnesota
Examination of the files of the Minnesota Department of Natural
Resources revealed 13 lakes which had been treated in previous years with
fish toxicants, and for which pre- and post-treatment transparency data were
available. Of these, 7 had higher transparencies after treatment, 2 probably
increased in transparency and b showed no change. In some of the lakes, the
effect appeared to last for several years.
9

-------
2. Effect of rotenone in Wirth Lake
This (16 ha, *»<3 m) lake is eutrophic from storm drainage input. Over a
period of several years the lake received a variety of ameliorative
techniques such as nutrient export, artificial circulation and piscivore
stocking. However, beneficial effects were minimal (circulation actually
increased nutrients and algal concentrations) until the lake was treated with
rotenone in fall, 1977- In 1978» Daphnia pulex became abundant and, despite
the circulation-caused high nutrient levels, it kept algal concentrations
very low and transparency high until August. In August, Aphanizomenon
flos-aquae became abundant in the flake form, disappearing only in September
when Daphni a pu1 ex were also absent. Evidence suggests that had the lake not
been treated with rotenone, the piscivores would have controlled the
planktivore populations and Daphni a pulex might have become abundant for this
reason.
3. Effect of rotenone in Clear Lake
This small lake divided by a roadway was treated with rotenone	after one
year of study. The half which had not previously winter-killed was	affected
most by the rotenone treatment. Daphnia pulex became abundant	and algal
biomass declined sharply.
IV C. Han i pu1 at i on of p1ankt i vores by wi nter-k i11
1.	Lakes affected in 1978-79-
Many Minnesota Lakes winter-killed in 1978-79- Nineteen lakes including
non winter-kill controls were sampled four times during spring and summer of
1979 to determine the effects. Of eight lakes suspected of hard winter-kill,
four had Daphnia pulex in them, and in three it was the dominant crustacean,
averaging 19~33/1 - Daphnia pulex also appeared in two lakes suspected of
partial winter-kill and in one lake known to be low in panfish.
Chlorophy11/TP ratios in the four lakes with abundant |K pu1 ex averaged
less than .132 + .046. Among the remaining fifteen lakes, ch1orophy11/TP
averaged .362 + .136.
Transparencies of the four D^_ pu1 ex lakes averaged greater than 2.07 +
.57 m and that of the remaining fifteen lakes 1.63 + >68 m. For three
control lakes, for which pre- and post-winter-kill transparency data were
available, no transparency changes were noted following winter-kill, but for
three partial winter-kill lakes, transparency doubled after the winter-kill.
With regard to the algal population, three of the four lakes in which D.
pulex were abundant were characterized by an abundance of Aphani zomenon
flos-aquae in its flake form.
2.	Effect of winter-kill in Lake of the Isles
In 197&~77. Lake of the Isles in Minneapolis suffered a severe
winter-kill. This storm drainage-fed eutrophic lake had perennially
developed large crops of blue-green algae and low transparency during summer.
In 1977, transparencies were so high that macrophyte problems prevailed
requiring mechanical harvesting. The high transparencies were probably
caused by grazing by Daphn i a pulex which became abundant in the lake
following the demise of the pianktivores. At the same time as [L pulex
10

-------
appeared, IL magna was found in the lake, and D_^ galeata increased in size
over previous years. Although some of the increase in transparency resulted
from the decrease in chlorophyll, part of the increase rests on the fact that
much of the remaining chlorophyll was present in Ceratium. Not only do these
organisms not attenuate light effectively but they were most abundant at some
distance below the lake surface.
V. Effects of phys i ca1-chemi ca1 cond i t ions on algal populations
Lake Harriet (H3 ha; 8.8 m mean depth) in Minneapolis perennially
produces lower algal concentrations than expected from its nutrient
concentrations. This discrepancy has been attributed to grazing by the
abundant Daphnia ga1eata, and indeed low chlorophyll concentrations have been
correlated with a high proportion of phaeophytin -- evidence of such grazing.
In 1974, summer chlorophyll concentrations in the lake suddenly increased
from their usual 5 ug/1 to as high as ^7 ug/1 . Algal volumes increased in
proportion and transparencies decreased. The situation ameliorated in 1975.
and by 1976 was "normal". In recent years, the same phenomenon appears to be
recurr i ng.
The explanation for the high chlorophyll in 197^ appears to lie in the
reduced numbers of Daphnia present that year. The decreased numbers of
Daphn i a may have resulted indirectly from the somewhat higher concentration
of phosphorus in the lake in 197^- That is, we hypothesize that the
increased phosphorus levels, too low to raise algal abundance by more than 20
or 30 percent, nonetheless allowed primary production in the euphotic zone
(not measured) to increase to the extent that dissolved oxygen concentrations
in the upper part of the hypolimnion (measured) became too low to allow the
Daphn i a to retain the zone as a refuge from fish predation. Consequently the
Daphn i a were forced to inhabit the waters above, where predation depleted
their numbers and released the algal population from their herbivory. Hence
the algal increase. If this hypothesis is correct it will represent the
first true threshold effect of nutrients in stimulating algal biomass in a
lake. It also opens the possibility that, if the upper portion of the
hypolimnion of such a lake were to be oxygenated artificially, Daphnia could
find a refuge from the fish, and remain abundant enough to limit the size of
the algal population.
11

-------
I. BIOLOGICAL EFFECTS ON THE SIZE OF THE NUTRIENT POOL.
THE ROLE OF BENTHIVORES IN LAKE MARION, MINNESOTA.*
It has been observed that the removal of fish from a lake, either
artificially with fish toxicants or naturally by winter-kill, often results
in reduced phytoplankton abundance and increased transparency. This
observation has been documented by Hrbacek et ah 096l), Schindler and
Comita (1972), Stenson et a_K (1976). and others. One possible explanation
for this phenomenon follows from the work of Lamarra (1975a)- Lamarra
demonstrated that benthivorous fish regenerate considerable quantities of
nutrients from the bottom sediments by their feeding and digestive
activities. Removal of such fish could eliminate a source of internal
nutrient loading to the lake and lead to lowered alga! standing crops.
Management of the fish community has been proposed as a technique to control
algal abundance (Shapiro et al. 19755- Successful development of this
technique will require detailed study of the mechanisms involved.
Andersson et a 1. (1978), like Lamarra, found higher levels of phosphorus
and nitrogen in experimental enclosures containing benthivorous fish than
were found in fish-free controls. However, the significance of benthivorous
fish excretion in the nutrient budget of an entire lake has never been
directly evaluated. In order to do this, an experiment was designed
involving Lake Marion, in McLeod County, Minnesota. Lake Marion, in 1978.
had a fish community dominated by benthivores such as black bullheads
(tctalurus melas) and carp (Cypr i nus carpio). An opportunity to study the
effects of removing these fish existed when the Minnesota Department of
Natural Resources planned to treat the lake with the piscicide, rotenone, for
recreational fisheries management purposes, in Fall 1978. However,
unforeseen increases in the price of the rotenone caused a postponement of
the treatment until Fall 1979- Thus, two seasons of field work (during 1978
and 1979) have been carried out on Lake Marion in order to establish baseline
data and predictions against which the response of the lake, in 1980, to fish
removal will be compared. Although completion of the experiment awaits the
I98O season, the results to date are presented below.
METHODS
Samplinq Procedure. All water samples were taken during the mid-morning
hours at a single station in Lake Marion located over the deepest area
of the basin (Fig. 1) .
Temperature and Pi ssolved Oxygen. Depth profiles of temperature and
dissolved oxygen were obtained with a Yellow Springs Instrument Co.
Model 5^ ABP dissolved oxygen meter.
Transparency. Secchi disc measurements (SD) were taken using a 20 cm white
d i sc_j_
Phosphorus. Total phosphorus (TP) in depth-integrated whole water samples
was digested with potassium persulfate, and analyzed col orimetrica11y by
the ascorbic acid method (APHA 1971)- Dissolved phosphorus (OP) was
measured on filtrates by the same method. Particulate phosphorus (PP)
is obtained by difference. Orthophosphorus (P04-P) was measured on
undigested filtrates using the ascorbic acid method. All values
reported represent the mean of at least three replicate analyses.
N i trogen. Total nitrogen (TN) in depth-integrated whole water samples was
analyzed by alkali-persul fate digestion, followed by nitrate reduction
rtby Eric Smeltzer and Joseph Shapiro
12

-------
and colorimetric analysis as described by D'Elia et al. (1977) • In
order to avoid clogging of the cadmium columns, an ammonium chloride
buffer solution was used during nitrate reduction in place of the boric
acid buffer employed by D'Elia £t aj_. This substitution was shown not
to alter the results. Nitrate in filtered samples was analyzed
colorimetrically by the cadmium reduction method (APHA, 1970• The
values reported represent the sum of nitrate plus nitrite (N03"N and
N02-N). Ammonia (NH3~N) in filtered samples was analyzed
colorimetrically by the method of Chaney and Marbach (1962). All values
reported represent the mean of at least three replicate analyses.
Photosynthes i s and Respi ration. Rates of planktonic photosynthesis and
respiration were measured J_n s i tu by the light and dark bottle oxygen
method. Glass 300 ml B.O.D. bottles were filled with water from depths
of 0,.5, 1.0,1.5(2.0, and 3>0 m and incubated at depth for the 6-hour
period preceding solar noon. Oxygen changes in the light and dark
bottles were analyzed by the azide modification of the Winkler method
(APHA, 1970. Daily rates of photosynthesis and respiration were
obtained by multiplying the 6-hour rates of gross photosynthetic oxygen
production and respiratory consumption by factors of 2 and 4,
respectively. Conversion of oxygen changes into units of carbon
assimilation and loss was done by assuming a photosynthetic quotient of
1.2 moles of oxygen evolved per mole of carbon assimilated, and a
respiratory quotient of 1.0 moles of carbon respired per mole of oxygen
consumed (Strickland, i960). Integral rates of photosynthesis (Aint)
and respiration (Rint) (mgC m-2 day-1) were obtained by numerical
integration of the photosynthesis-depth profiles. "Light-saturated"
volumetric rates of photosynthesis (Aopt, mgC m~3 day-1) were obtained
by using the maximum volumetric rate observed in the
photosynthesis-depth profile.
P1ank ton i c B i omass. Chlorophyll a concentrations (CHL) were determined with
904 acetone extraction of samples ground on glass fiber filters,
followed by trichromatic spectrophotmetric analysis (Strickland and
Parsons, 1968). No correction was made for the presence of
phaeopigments. Particulate organic carbon (P0C) was measured by the
dichromate oxidation method (Strickland and Parsons, 1968). Because
naturally-occurring particulate organic carbon is in a chemically more
reduced state than the glucose used to calibrate the POC analysis, a
conversion factor of .83 mg natural POC per mg POC as glucose was
applied. This factor was obtained by assuming that the reduction state
of naturally occurring POC reflected the photosynthetic quotient of 1.2.
All CHL and POC values reported represent the mean of at least three
replicate analyses.
Algae. Depth-integrated water samples were preserved in Lugol's solution for
qualitative analysis.
Zooplankton. Depth-integrated zooplankton samples were obtained from single
vertical tows with a Wisconsin plankton net (17 cm diameter opening, 64
mesh size). The animals were preserved in a sucrose-formalin mixture
for microscopic identification and enumeration.
F i sh Populat i on Est imates. The abundance of bullheads and carp in Lake
Marion was estimated by two methods.
1. A multiple-census mark-recapture estimate was made, using the
results of nettings carried out on four dates in August and
September, 1979- The fish were captured in trap nets set for 48
hours with their leads extending into shore. Captured fish were
13

-------
marked by removal of the right pectoral fin, and were returned to
the water at a location near the center of the lake. The population
numbers were estimated for each species and for each size class
within a species according to the method of Schumacher and
Eschmeyer (19^3)•
2
Em c
N = Im r
(1)
N = the estimated population number
m = the cumulative number of previously marked fish
r = the number of marked fish recaptured on a given sample date
c = the total number of marked and unmarked fish captured
on a given sample date
T. indicates a summation over all sample dates
Separate estimates were made for two si2e classes of bullheads. It
proved impossible to obtain a mark-recapture estimate of carp
numbers because these fish rarely entered the trap nets.
2. The second method of estimation involved a count of dead fish washed
ashore five days after the lake was treated with rotenone in
October, 1979- Length intervals along the shore were marked off
systematically around the entire lake. Complete counts were made
of the fish within each interval, with the assistance of about 50
seventh-grade students and their teachers from the Hutchinson,
Minnesota, Junior High School. By knowing the length of shoreline
contained within these intervals (20% of the total shore length),
the total number of fish on the shore could be estimated. This
number is an underestimate of the true number of fish in the lake
because an unknown fraction of the fish never came ashore.
Separate counts were made for each species. Length-frequency
distributions were also obtained on samples of dead fish in order
to develop estimates of numbers in each si2e class. Using this
information, along with length-weight relationships for the two
species, it was possible to make separate biomass estimates for 17
weight classes of bullheads and 7 weight classes of carp.
RESULTS
Lake Marion is a large, shallow, eutrophic lake located within the
watershed of the South Fork of the Crow River in McLeod County in
south-centra! Minnesota (T. 115,116N,R. 29.30W). A map of the lake is given
in Figure 1. Agriculture is the major land use in the Lake Marion watershed.
The main (south) basin of the lake is 172 ha in surface area with a mean
depth of 1.98 m, and is drained by a single outlet at the south end. Depth
profiles of temperature and dissolved oxygen indicate that the lake does not
stratify during summer. The small north basin (3^ ha) is dominated by
14

-------
South Bosm
4.0 m.
max.
Sample Station
Outlet
Fig. 1 Map of Lake Marion, Mcleod County, Minnesota.
15

-------
emergent vegetation and drains into the main basin through a narrow culvert.
The north basin was not sampled, and the results presented below apply to the
mai n basin only.
Secchi disc transparency in Lake Marion was low, averaging .89 m in 1978
and .64 m in 1979 (June-September means). Algal biomass was moderately high
during this period, with CHL averaging 36 mg m~3 in 1978 and 43 mg m~3 in
1979- Blue-green algae such as Hicrocyst is sp. and Osci1lator ia sp. were the
dominant phytoplankton. Theoretical considerations (Brezonik, 1978; Megard
et a 1., I98O) indicate that the reciprocal of the Secchi disc depth (SD-1)
should be a linear function of the concentration of light-attenuating
substances such as chlorophyll. This relationship holds true for Lake Marion
in 1978 and 1979 as can be seen in Figure 2. The intercept of the regression
of SD-1 on CHL can be interpreted as the average amount of "background"
attenuation by materials not associated with the algae. When this intercept,
with a value of .41 m-1 is compared with the average value for SD-1 of
1.40m-l, it can be seen that most (71%) of the light attenuation was caused
by materials directly associated with the algae.
The suggestion that increased transparency and reduced algal abundance
following fish kill are a result of reduced nutrient loading by benthivorous
fish excretion carries with it the assumption that algal biomass and
productivity are regulated primarily by nutrient concentrations. Plots of
CHL and Aopt vs. TP and TN for Lake Marion (Figs. 3 and 4) give no apparent
relationships, within seasons, between the nutrients and algal biomass or
productivity. However, factors such as sampling error, temperature
variations, and time lags in the response of the algal community to changes
in nutrient levels could obscure such relationships. Better results have been
obtained when season means of CHL and Aopt, for a large number of lakes, are
plotted against TN and TP (Sakamoto, 1966; Dillon and Rigler, 1974; Jones and
Bachman, 197&; Smith, 1979)- The implication of these relationships is that
nutrients such as nitrogen and phosphorus exert primary controlling influence
on algal biomass and productivity. Figures 5 and 6 show that season mean
values from Lake Marion fit well when compared with these established
relationships. It is therefore reasonable to predict that a reduction in
nutrient loading to Lake Marion, and a corresponding drop in nitrogen and
phosphorus concentrations, should lead to reduced average summer algal
biomass and productivity.
One way of determining the effect of benthivorous fish on levels of
nutrients is to correlate past changes in fish abundance with changes in
nutrient concentrations. Table 1 summarizes all the data that is available
for Lake Marion. The period of 1975~1979 was a time of low winter dissolved
oxygen concentration, chronic winter-kill, and dominance of the fish
community by black bullheads. The sharp decline in bullhead abundance
between 1977 and 1979 was probably a consequence of winter-kill conditions.
The bullhead decline was accompanied by a pronounced drop in TP and a small
drop in TN. This trend, though somewhat circumstantial, provides evidence
that nutrient levels in Lake Marion may depend on benthivorous fish
abundance.
A second way of determining the significance of benthivorous fish in the
nutrient budget of Lake Marion is to compute the quantities of phosphorus and
nitrogen excreted by these fish. To do this, the following information was
assemb1ed.
1. The specific excretion rates of nitrogen and phosphorus (ug/g wet wt/hr)
by carp and black bullheads were obtained from the data of Lamarra
(1975b). By measuring specific excretion rates of fish taken from Union
16

-------
2.0
• •
SD = .0249 CHL+ .413
0.5
0.0
30
CHL (mg m"3)
40
50
60
20
0
Fig. 2 The relationship between SD~^ and CHL in
Lake Marion, 1978 and 1979.
17

-------
. lOOr
50f-
C7»
E
5 10-
ic
lOOr
197B

50 IOC
TP {mg m"3)
too
D>
E
50-
X
O 10
1000
lOOr
1978
1500 2000
_-3\
T N (mg rrr
cn
E
!P70 «•
o>
E
50-
1979
_J
X
o
50 100
TP (mg m"3)
I
o
1000
1500 2000
TN (mg rrf3}
Fig. 3 The relationships between CHL, TP and
TN in Lake Marion within the 1978 and
1979 seasons.
_3000r
'&• I
"O
E looo- ^ 978
50	!00
,-3\
TP (mg m" )
1979
50	100
_-3\
TP {mg rrr
3000
a
T>
2000
o
w
£
n
o
<
ooc
3000.-
fO
'e
o
E
a
o
<
2000-
•000-
1000
1976
! 000
15C0 2000 2500
TN (mg m'3)
'979
1500 2000
TN (mg m"3)
2500
Fig. 4 The relationships between Aopt, TP and
TN in Lake Marion within the 1978 and
1979 seasons.
18

-------
100
1979-
1978
O
100
10
1000
TP (mg m"3)
ic
100
TN (mg rrf3)
Fig. 5 July-August means of CHL, TP arid TN in
Lake Marion, 1978 and 1979, compared with the
empirically established relationships of A. Jones
and Bachman (197$), and B. Sakamoto (1966J. The
regression equation for "CHL vs. TN was obtained
from Sakamoto's Fig. 4.
1000-
Aopt ^ IS8 TF-77
0	SO	100
TP (mg m"2)
' 3003^- : ^ TN ~ 49
>%
O
ID
2000-
0
cn
£
1
<
100:
2000
TN (mg rrf3)
Fig. 6 June-Sept, means of Aopt, TP and TN in
Lake Marion, 1978 and 1979, compared with the
empirically established relationships of Smith
(1979).
19

-------
Table 1. Correlation of past changes in relative bullhead abundance in Lake
Marion with changes in nutrient concentrations (mg m"3). Bullhead
values represent the mean number of bullheads captured per trap
net per day of set.
Relative Bullhead
Year	Abundance	TP TN	Source
1975	—	93 1920 Minnesota Pollution Control
Agency. Single sample date
Sept. 1975.
1977	152	-- -- Minnesota Dept. of Natural
Resources. Mean of 5 trap
nets on a single August
sample date.
1978	—	107 1960 Present study. July - Sept.
means.
1979	60	63 1760 Present study. Nutrient
values are July - Sept.
means. Bullhead number is
the mean for 22 nets distri-
buted over 4 Aug. - Sept.
sample dates.
20

-------
Lake, Minnesota, and placed in small plastic enclosures for one-hour
time intervals, Lamarra developed the following regression equations for
carp.
log 10 E(DP) = -.49 loglO W + .027 T + .77	(2)
loglO E(NH3) = -.38 loglO W + .04 T + 1.03	(3)
E(DP) = specific excretion rate of dissolved phosphorus
(tig g-1 hr-1)
E(NH3) = excretion rate of ammonia
W = wet weight of fish (g)
T = temperature (°C)
Lamarra's specific excretion rate experiments on black bullheads
were not extensive enough to permit the inclusion of temperature as a
second variable in the regressions. However, by assuming the same
temperature coefficient for black bullheads as for carp, the following
equations for black bullheads can be obtained from Lamarra's data.
loglO E(DP) = -.347 loglO W + .027 T + .344	(4)
loglo E(NH3) = -.701 log!0 W + .04 T + 1.08	(5)
It will be assumed that other variables affecting the specific
excretion rates, such as sediment phosphorus content, were similar in
Lake Marion to levels found in Lamarra's Union Lake.
2.	In order to simplify calculations of loading rates, the year was divided
into two 6-month seasons, summer and winter, during which it was assumed
that representative water temperatures were 20 and k degrees C,
respect i vely.
3.	The biomass of benthivorous fish in Lake Marion in 1979 was estimated by
the mark-recapture and the shore census methods. The results are shown
in Table 2. The sharp discrepancy between the bullhead estimates by the
two methods is disturbing. There are some possible reasons for the
systematic error in the mark-recapture process which may have led to the
underestimate, but they will not be discussed here. The results of the
shore census will be used for the purpose of calculating nutrient
loading to Lake Marion by benthivorous fish, although as noted before,
this is an underestimate of the true biomass of fish in the lake.
Annual rates of nutrient loading by benthivorous fish excretion were
then estimated as follows. Specific excretion rates for each weight class of
the two species were calculated using Lamarra's equations, the mean weight of
individuals in each weight class, and the temperature during each season.
Loading rates for each weight class at each temperature were calculated as
the product of the specific excretion rate and the total biomass of the
weight class in the lake. These loading rates were summed over all weight
classes and temperature periods to obtain the total annual nutrient loading
21

-------
rates by benthivorous fish of each species. The results are given in Table
3-
DISCUSSION
Bottom-feeding fish such as carp and bullheads are commonly found in
lakes with obvious manifestations of eutrophication. Moyle (1968) noted a
correlation between the presence of "rough fish", algal blooms, and low
transparency in Minnesota lakes. The cause and effect relationships are not
obvious however. One interpretation is that bottom-feeders promote water
fertility by "stirring up" nutrients from the sediments. Lamarra (1975a)
disagreed with this interpretation and demonstrated that stirring of the
sediments did not produce increases in total phosphorus and chlorophyll the
way fish activity did. A second mechanism linking benthivorous fish to
turbidity might be the resuspension of light-scattering particles from the
bottom by fish movements. In Lake Marion, the strong dependence of light
attenuation in the water column on chlorophyll concentration (Fig. 2) is
evidence that factors other than algae are of minor importance in causing low
transparency. If benthivorous fish are, in fact, partly responsible for the
eutrophic condition of Lake Marion, then the causal mechanism is probably
nutrient regeneration by excretion, as suggested by Lamarra, rather than the
stirring up of sediments.
In order to judge the significance of benthivorous fish in the overall
nutrient budget of Lake Marion, the calculated nutrient loading rates from
fish excretion must be compared with other sources of loading including
external loading from the watershed, precipitation, and direct release from
the sediments. Benthivorous fish excretion supplied phosphorus to Lake
Marion, in 1979. at a rate of approximately 88 mg m-2 yr-1 (from Table 3)•
An estimate of phosphorus loading from external sources can be obtained from
the Minnesota Pollution Control Agency (MPCA, 1975» unpublished report). The
MPCA estimated phosphorus loading rates to Lake Marion from external sources,
including drainage, direct precipitation, and septic tank seepage. The
necessary values for nutrient inputs from the various sources and for the
hydrologic factors were obtained indirectly from regional averages. The MPCA
estimated that the external phosphorus loading rate to Lake Marion was about
8i» mg m-2 yr-1. Apparently, fish excretion was as important as external
sources in supplying phosphorus to the lake.
To determine whether fish excretion represents a significant fraction of
the total loading, it is necessary to evaluate the rate of direct sediment
phosphorus release. Values reported in the literature for phosphorus release
rates from aerobic lake muds under laboratory conditions vary over a range of
-2.0 - 9-^ mg m-2 day-1 (Kamp-Nielsen, 197^; Banoub, 1975: Ryding and
Forsberg, 1977)- This range is too broad to develop a reliable approximation
for phosphorus release rates from Lake Marion sediments. For example, use of
these literature values suggests that direct sediment release could be a
considerable proportion of the total loading, even under aerobic conditions.
Thus, Ryding and Forsberg (1977) quote a range of 0-3 ~ 2.0 mg m-2 day-1 for
sediments derived from three shallow, highly eutrophic Swedish lakes. When
expressed on an annual basis, this range of 110 -730 mg m-2 yr-1 can be
compared with the loading rates to Lake Marion from benthivorous fish (88 mg
m-2 yr-1) and from external sources (8*) mg m-2 yr-1). In this manner one can
speculate that direct sediment release of phosphorus was of such overwhelming
importance that removal of benthivorous fish might have no significant impact
on phosphorus levels in the lake.
On the other hand, Carlson (1975) measured phosphorus release rates from

-------
Table 2. Estimates of Benthivorous Fish Biomass in Lake Marion, 1979.
Method
Species
Mark - Recapture Bullheads
Total Number
25,050
(23,090 - 27,400)*
Total Wet
Weight (kg)
1,860
Area! Dens
(kg ha'1
11
Mark - Recapture Carp
Shore Census
Bui 1 heads
145,800
10,100
59
Shore Census
Carp
18,700
7,300
43
* Precision limits given for the mark - recapture bullhead estimate are the
95% confidence intervals, calculated according to Ricker (1975).
Table 3. Estimates of nutrient loading rates to Lake Marion during
summer 1979 by benthivorous fish excretion.
LOADING RATES
Species
Bullheads
Carp
TOTAL:
Phosphorus
kg yr
103
46
149
-1
-2 -1
mg m yr
61
27
88
Nitrogen
kg yr
202
263
465
-1
-2 -1
mg m yr
117
153
270
23

-------
aerobic sediment obtained from eutrophic Cottonwood Lake, Minnesota, a lake
similar in morphometric and biological characteristics to Lake Marion.
Carlson found no net phosphorus release from aerated muds. It is
conceivable, therefore, that benthivorous fish excretion may indeed have
supplied as much as half the total loading of phosphorus to Lake Marion. It
may be of significance that the laboratory phosphorus release rates reported
by Ryding and Forsberg (1977) greatly underestimated the net flux of
phosphorus from their sediments J_n s i tu, as calculated from the nutrient
budgets of their lakes. This discrepancy suggests that biological fprocesses
such as benthivorous fish excretion, that are not duplicated under laboratory
conditions, may be of considerable importance in the internal phosphorus
loading of their lakes as well.
The effect of removing the portion of the phosphorus loading derived
from fish excretion can be analyzed by modifying Vollenweider1s (19&9)
phosphorus balance model to account explicitly for internal loading.
L(E) ~ L(F) + L(S) Tp Tp	(6)
dTP/dt =	-	alP plP
where dTP/dt = the change in total phosphorus concentration
per unit of time (mg m-3 yr-1).
L(E) = the areal rate of phosphorus loading from
external sources (mg m-2 yr-1).
L(F),L(S) = areal rates of internal phosphorus loading from
fish excretion and from direct sediment re-
lease, respectively (mg m-2 yr-1).
TP = concentration of total phosphorus (mg m-3).
2 = mean depth of the lake (m).
c = specific phosphorus sedimentation rate (yr-1).
p = hydraulic flushing rate (yr-1).
Under steady-state conditions (dTP/dt * 0), equation 6 can be solved for the
equilibrium phosphorus concentration (TP*) as follows.
1(E) + L(F) + L(5)	(7)
TP*= Z* (a + p)
Of the terms in equations 6 and , L (S) and have yet to be measured in
Lake Marion. Field studies of sediment release and phosphorus sedimentation
rates are planned for Lake Marion in 1980 in order to evaluate one or both of
24

-------
these unknowns. When this is accomplished, it will be possible to use an
approach similar to that of Dillon and Rigler (1975)» with equation 7» to
predict the decline in TP levels to be expected after the elimination of
benthivorous fish excretion. Such a prediction can be verified by observing
the lake's response in 1980 to fish removal.
Reductions in algal biomass and productivity can be expected if declines
in TP levels do, in fact, occur. This prediction depends on the assumption
that phosphorus, rather than nitrogen, is the variable controlling Chi and
Aopt. This assumption is justified, given the observed mean TN/TP ratios of
l8.lt and 28.k in 1978 and 1979- These ratios indicate phosphorus limitation,
according to Sakamoto (1966) and Smith 0979) • The input ratio of nitrogen
(NH3)/phosphorus (DP) by benthivorous fish excretion is 3-1 (see Table 3)•
Therefore, elimination of the fish should produce an even greater degree of
phosphorus limitation in Lake Marion.
There is one factor that could cause a delay in the expected nutrient
decline in the lake following benthivorous fish removal. This is the
potential for fish carcasses to release nutrients upon decomposition. Bull
and Mackay (1976) indicated that carp contain about 0.5% phosphorus and 2.6%
nitrogen, on a wet weight basis. By assuming that these values are
approximately true for both carp and bullheads, and by using data given in
Table 2, it can be calculated that decomposition of the carp and bullheads
would release 52 kg of phosphorus and 435 kg of nitrogen. These quantities
are significant when compared with the amounts of the nutrients excreted by
the fish when they were alive (see Table 3)- As a consequence, nutrient
levels may not decline as rapidly as expected. However, inputs of nutrients
from fish decomposition constitute a single pulse of loading, in contrast to
excretion, which was a chronic source. Rapid sedimentation in the shallow
Lake Marion basin should insure that the effects of this pulse are
short-1i ved.
In addition to the elimination of benthivorous fish excretion, there is
a second mechanism that could explain reduced algal abundance following fish
kill. P1anktivorous fish exert size-selective predation pressure on the
zooplankton community (Brooks and Dodson, 1965)» and may prevent the
dominance of large-bodied species. Many investigators, including Hrbacek and
Novatna-Dvorakova (19&5). Galbraith (19^7)» and Stenson (1972, 1976) have
documented the occurrence of numerous, large, herbivorous zooplankton such as
Daphnia sp. following fish kill, and some have linked reduced algal densities
in these situations to increased grazing losses. Addition of pianktivorous
fish such as the fathead minnow (P imepha1es promelas) can reverse the effect,
causing increased algal abundance, as shown by McNabb (1976), Helfrich
(1976), and Lynch and Shapiro (1980). Fathead minnows were the dominant
planktivore in Lake Marion in 1978 and 1979. and their removal by the
rotenone treatment should result in greater zooplankton abundance and
increased average body si2e of the zooplankters. Once the response of Lake
Marion in 1980 to fish kill is known, it will be possible to study the
significance of the grazing mechanism by observing changes in the zooplankton
community and by evaluating the role of grazing losses in regulating algal
biomass and growth rates. It will also, of course, be possible to verify the
expectations about the significance of benthivorous fish excretion in the
nutrient budget of Lake Marion.
All data collected on Lake Marion during the course of this project in
1978 and 1979 are tabulated in the Appendix (Table 3a).
25

-------
Table 3a (Appendix) Measurements on Lake Marion in 1978 and 1979.
Temp* SD Chi	POC	Aopt	Aint
Date (°C) (nn) (mg m-3) (mg m-3) (mgC m-3 day-1) (mgC m-2 day
6/10/78
20.8
.9
23
5900
—
—
6/17/78
20.8
1.5
16
3400
719
1360
6/27/78
22.6
1.2
32
3140
1440
2930
7/8/78
22.4
.7
54
--
2980
2550
7/14/78
23.2
. 6
49
5023
2544
2600
7/23/78
22.4
.8
45
--
2220
2410
8/4/78
21.6
.8
31
4840
1480
2190
9/7/78
24.1
.6
41
7860
1640
1810
5/1/79
9.9
.8
57


_
6/11/79
19.2
.65
43
—
--
--
6/29/79
20.9
.84
24
378
1560
2220
7/16/79
24.6
.64
43
4690
2590
2960
8/1/79
23.7
.53
52
4880
2290
2130
8/15/79
19.9
.70
45
—
1220
1480
9/3/79
22.4
.62
44
4370
2080
2020
9/14/79
16.8
.53
50
4620
1560
1640
* Average temperature in water column.
26

-------
Table 3a (Appendix) continued
Rint	TP	DP	P04-P	TN (N03+N02)-N NH3-N
(mgC m-2 day-1) (mg m-3) (mg m-3) (mg m-3) (mg m-3) (mg m-3) (mg m-3)
1560
96	40
89	30	7
2490 85	28	9	—	125
1920 101	25	5	2080	13	19
1245 103	34	-	1870	8	0
1350 107	78	3	1790	2	0
1230 116	32	1	2100	2	5
83	--	-	2580
78	—	-	2310
1740	82	19	-	2460
2820	64	22	-	1650
2490	55	12	-	1690
1860	68	19	-	1990
2240	64	16	-	1740
1910	66	16	-	1740
27

-------
ACKNOWLEDGEMENTS
Financial support was contributed by the U.S. Environmental Protection Agency
through a grant to Dr.
Joseph Shapiro. The assistance of the Minnesota Department of Natural
Resources, Section of Fisheries is gratefully acknowledged. Many individuals
participated in the field work, including Peter Berglund, Bruce Forsberg,
Elizabeth Kellogg, Dale Smeltzer, Edward Swain, David Wright, and about fifty
students and their teachers from the Hutchinson, Minnesota Junior High
School. Data analysis was aided by a grant from the University of Minnesota
Computer Center.
REFERENCES
Andersson, G., H. Berggren, G. Cronberg, and C. Gel in. 1978. Effects of
planktivorous and benthivorous fish on organisms and water chemistry in
eutrophic lakes. Hydrobio1ogia 59:9-15-
American Public Health Association. 1971. Standard Methods for the
examination of water and wastewater. 13th ed. Washington, D.C. American
Public Health Association. 874 pp.
Banoub, M. 1975- Experimental studies on material transactions between mud
and water of Gnadensee (Bodensee) . Verh. Int. Ver. Limnol.
19:1263-1271•
Brezonik, P.L. 1978. Effect of organic color and turbidity on secchi disc
transparency. J. Fish. Res. Bd. Can. 35:1^10-1416.
Brooks, J.L. and S.I. Dodson. 1965* Predation, body size, and composition
of plankton. Science 150:28-35-
Bull, D.J. and W.C. Mackay. 1976. Nitrogen and phosphorus removal from
lakes by fish harvest. J. Fish.. Res. Bd. Can. 33:1374-1376.
Chaney, A.L. and E.P. Marbach. 1962. Modified reagents for determination of
urea and ammonia. Clinical Chemistry 8:130-132.
Carlson, R.E. 1975- Phosphorus cycling in a shallow eutrophic lake in
southwestern Minnesota. Ph.D. Thesis. Univ. Minnesota, Minneapolis,
Mi nn.
D'Elia, C.F. 1977- Determination of total nitrogen in aqueous samples using
persulfate digestion. Limnol. Oceanogr. 22:760-763.
Dillon, P.J. and F.H. Rigler. 1974. The phosphorus-chlorophyll relationship
in lakes. Limnol. Oceanogr. 19:767_773-
Dillon, P.J. and F.H. Rigler. 1975- A simple method for predicting the
capacity of a lake for development based on lake trophic status. J.
Fish. Res. Bd. Can. 32:1519"1531•
Galbraith, M.L. 1967- Size-selective predation on Daphn i a by rainbow trout
and yellow perch. Trans. Am. Fish. Soc. 96:1-10.
Helfrich, L.A. 1976. Effects of predation by fathead minnows, Pimephales
promelas, on planktonic communities in small, eutrophic ponds. Ph.D.
Thesis. Michigan State University. E. Lansing, Michigan.
Hrbacek, J., M. Dvorakova, V. Korinek, and L. Prochaskova. 1961.
Demonstration of the effect of fish stock on the species composition and
the intensity of metabolism of the whole plankton association. Verh.
Internat. Verein. Limnol. 14:192—195•
Hrbacek, J. and M. Novotna-Dvorakova. 1965- Plankton of four backwaters
related to their size and fish stock. Rozpravy Cesk. Akad. Ved., Matem.
pr i r. ved. 75:1-65.
Jones, J.R. and R.W. Bachman. 1976. Prediction of phosphorus and
chlorophyll levels in lakes. Jour. Water. Poll. Control Fed.
28

-------
148:2176-2182.
Kamp-Nielsen, L. 1974. Mud-water exchange of phosphate and other ions in
undisturbed sediment cores and factors affecting the exchange rates.
Arch. Hydriobiol. 73^218-237•
Lamarra, V.A. 1975a* Digestive activities of carp as a major contributor to
the nutrient loading of lakes. Verh. Internat. Verein. Limnol.
19;2461-2468.
Lamarra, V.A. 1975b- Experimental studies of the effect of carp (Cypr i nus
carpio) on the chemistry and biology of lakes. Ph.D. Thesis. University
of Minnesota. Minneapolis, Minn.
Lynch, M., and J. Shapiro. 1980. Predation, enrichment, and phytoplankton
community structure. Limnol. Oceanogr. In press.
McNabb, C.D. 1976. The potential of submersed vascular plants for
reclamation of wastewater in temperate 2one ponds. J_n Tourbier, J. and
R.W. Pierson (eds.) . Biological Control of Water Pollution. Univ.
Penn. Press, p. 123~132-
Megard, R.O., H.A. Boyer, and J.C. Settles. 198O. Light, secchi disks, and
trophic states. Limnol. Oceanogr. In Press.
Moyle, J.B. 1968. Notes on some characteristics of Minnesota lakes having
blue-green algal blooms. Minn. Dept. Conserv. Spec. Publ. 52.
Ricker, W.E. 1975- Computation and interpretation of biological statistics
of fish populations. Bull. Fish. Res. Bd. Can. 191:1-382.
Ryding, S.O. and C. Forsberg. 1977• Sediments as a nutrient source in
shallow polluted lakes. pp. 227-234. J_n Golterman, H.L. (ed.)
Interactions Between Sediments and Fresh Water. Proc. Int. Symp.
Amsterdam, Netherlands, Sept. 6-10, 1976.
Sakamoto, M. 1966. Primary production by the phytoplankton community in
some Japanese lakes and its dependence on lake depth. Arch. Hydrobiol.
62:1-28.
Schindler, D.W. and G.W. Comita. 1972. The dependence of primary production
upon physical and chemical factors in a small, senescing lake, including
the effects of complete winter oxygen depletion. Arch. Hydrobiol.
69:413-451•
Schumacher, F.X. and R.W. Eschmeyer. 19^3- The estimation of fish
population in lakes or ponds. Jour. Tenn. Acad. Sci. 18:228-249.
Shapiro, J., V. Lamarra and M. Lynch. 1975- Biomanipulation: An ecosystem
approach to lake restoration, pp. 85-96. J_n Brezonik, P.L. and Fox,
J.L. (eds.). Proc. symp. on water quality management through biological
control. U.S. EPA Rept. No. ENV-07"75-l.
Smith, V.H. 1979- Nutrient dependence of primary productivity in lakes.
Limnol. Oceanogr. 24:1051-1064.
Stenson, J.A.E. 1972. Fish predation effects on the species composition of
the zooplankton community in eight small forest lakes. Rept. Inst.
Freshwater Res. Drottningholm. 52:132-148.
Stenson, J.A.E., T. Bohlin, L. Henrickson, B.I. Nilsson, H.G. Nyman, H.G.
Oscarson and P. Larsson. 1976. Effects of fish removal from a small
lake. Verh. Internat. Verein. Limnol. 20:794-801.
Strickland, J.D.H. i960. Measuring the production of marine phytoplankton.
Bull. Fish. Res. Bd. Can. 122:1-172.
Strickland, J.D.H. and T.R. Parsons. 1968. A practical handbook of seawater
analysis. Bull. Fish. Res. Bd. Can. 167:1_311«
Vollenweider, R.A. 1969- Moglichkeit en und Grenzen elementarer Modelle der
Stoffbilanz von Seen. Arch. Hydrobiol. 66:1-36.
29

-------
II. ALGAL MANIPULATIONS
IIA. THE EFFECTS OF PH AND CARBON DIOXIDE ON ALGAL COMMUNITIES*
in 1973 Shapiro described experiments, done in 1971 and 1972, in which
he had caused shifts in algal populations from predominantly blue-greens to
greens simply by adding nutrients and carbon dioxide to the systems. Because
of the potential to use such shifts as a lake restoration technique, as well
as because of the theoretical implications, it was desirable to learn more
about the phenomenon. For example, does it occur only in Lake Emily, the
original test site? How reproducible is it? Is it reversible? At what pH
values does it occur? Are there resistant species? Above all it was
desirable to learn the mechanism of the shift — what causes it? It was to
answer these questions that further experiments were done between 1972 and
1976.
METHODS
The early experiments had been done in the field in plastic "bag"
enclosures. Because of their success, and our desire to maintain nearly
natural conditions, most of the work to be described was done similarly. The
physical setup was that shown in Figure 7- The "bags", of six mil
polyethylene, were made by cutting appropriate lengths of extruded tubing,
approximately 1 meter in diameter, and sealing the bottom ends. Heat sealing
was unsatisfactory so the bottoms were sealed by folding over five
centimeters, taping with waterproof "Griffolyn" tape, and folding and taping
twice again. The tops of the bags were held open by folding the plastic
outward over a 1-meter diameter hoop of 1-inch polyethylene tubing.
Generally, to fill the bags they were taken about 20-30 feet from the raft
and towed, by means of a three-point harness on the top hoop, toward the
raft. In this way they filled with water from the first meter. They were
then pushed under the outer rim of the outrigger, raised up, and the hoops
fixed by nails to wooden blocks about inches above the water surface. No
metal was allowed to remain in contact with the water in the bags. As the
rafts were east-west oriented and the bags were on the south side, no shading
occurred. Generally the bags were about 1.5 meter deep and thus contained
about 1 cubic meter of water. On occasion certain bags were filled by pumping
water through a 2-inch diameter gasoline-operated pump that was first
flushed. Usually the bags were topped off using the pump or a plastic
bucket, and on occasion bricks wrapped in polyethylene were put in the bags
to keep the bottoms down. Sampling was done by dipping from the surface
after the first meter or so was well mixed with a wooden canoe paddle. Care
was taken to not disturb the sedimented material on the bottoms of the bags.
Nutrients were added in solution. Phosphorus was added as KH2P04, generally
over the first five days to a final concentration of 100 micrograms P/liter,
and nitrogen was added similarly as NH4N03 to a final concentration of either
700 or 1000 micrograms N/1 iter. pH was adjusted either by adding dilute HC1,
dilute NaOH, or by bubbling in 100% carbon dioxide gas. Other additions will
be described later. pH measurements were made with a variety of field
i nstruments.
Most of the work was done on Lake Emily, which is, as noted in the
earlier paper (Shapiro, 1973). a small lake just north of St. Paul,
Minnesota. The lake is shallow for most of its expanse and many of the
experiments were performed near the shallow southern end. However, the
experiments at different pH values were done at the northern end in about h~S
*by Joseph Shapiro, Vincent Lamarra and George Zoto
30

-------
Fig. 7 Physical arrangement used in the enclosure studies.
31

-------
meters of water. The lake is exceedingly eutrophic, being fed by storm
drainage from the surrounding suburban areas.
RESULTS
Reproduc i b i1i ty
Although the carbon dioxide-induced shift had been demonstrated first in
1971 and confirmed in 1972, every time a series of experiments was done at
least one pH 5-5 bag, continuously bubbled with C02 and with nutrients added,
was used along with an appropriate control with nutrients only to confirm the
continued capacity of the population to shift. Such experiments are noted by
aster i sks in Table k.
Of these 20 comparable lake experiments using additions of C02 to pH
5.5, and additions of N and P, 19 showed the shift from blue-greens to
greens. The sole exception was experiment F19 in which N and P were added
six days after the C02 treatment was begun. In some of the experiments
stirring with air was continuous while in others stirring occurred only at
sampling times. No difference in result was noted. It appears therefore
that, as the controls never went over to greens and the C02/nutrient
treatments virtually always did, the phenomenon is highly reproducible.
On one occasion (Table h, J (1)6) C02 treatment was continued only long
enough (about 30 minutes) to lower the pH to 5-5, whereupon the C02 was
turned off. In this case only a partial shift occurred.
Tests in Other Lakes
The phenomenon appears not to be limited to Lake Emily. On several
occasions (H, I, J2) various other lake waters were tested by incubating them
in the laboratory (300 foot candles continuous light, 22 degrees C) with N
and P and C02. Lake Emily water was treated similarly for comparison. Not
only did the shift occur in Lake Emily under the laboratory conditions, but
it took place in all 10 other lake waters tested (Table k) . In 3 of these
tests the controls, containing N and P only, also changed. This may have
been caused by C02 leaking into the incubator from those systems being
bubbled with it. They represent the only controls that shifted during all of
the studies.
Time of Year
In addition to the phenomenon being reproducible, and applying to other
lake waters and their algae, it appears to extend over most of the growing
season for the algae. Experiments in Lake Emily have been begun as early as
June 2k and as late as September 2k, and have gone on as late as October 25,
a 1 I successfu11y.
Effect of HC1 versus C02
Because one effect of the C02 addition is to lower the pH to about 5*5,
and because an understanding of the mechanism of the shift necessitates
knowing whether lowering the pH artificially without C02 addition is
effective, three field experiments (Table 4, C(l) 5, 6; and Dll) were done in
which N and P were added and the pH was lowered to 5*5 with HC1. The shift
took place in all three. However, in two laboratory tests with water from
Quarry Lake the shift did not occur (Table k, I).
Experiments at Different pH Values
The original intent of the work had been to test a hypothesis based on a
publication of King (1970) on the relative ability of green and blue-green
32

-------
TABLE 4
Experiments in which either CO2 or HCL was used to lower the pH to
5.5 - 6.0. Experiments marked * are those where CO2 and N & P were
added to enclosures in the field.
SERIES &
EXPERIMENT
DATE
LAKE
AGENT
NUTRIENTS
SHIFT
* B 2
7/29/71
Emily
co2
+
Yes
B 3
7/29/71
Emily
co2
-
Partial
ciD2
7/23/72
Emily
HC1
-
No
cd)5
7/23/72
Emily
HC1
+
Yes
C(l)6
7/23/72
Emily
HCl
+
Yes
* c(l)9
7/23/72
Emily
C°2
+
Yes
* C(l)l°
7/23/72
Emily
co2
+
Yes
C(l)11
7/23/72
Emily
co2
-
Yes
C(l)12
7/23/72
Emily
co2
-
Yes
* C(2)8
8/9/72
Emily
co2
+
Yes
* D 2
6/22/74
Emily
co2
+
Yes
D 11
6/22/74
Emily
HCl
+
Yes
* E 2
7/16/74
Emily
co2
+
Yes
* F 2
8/21/74
Emily
co2
+
Yes
* F 3
8/21/74
Emily
co2
+
Yes
* F 15
8/21/74
Emily
co2
+
Yes
* F 16
8/21/74
Emily
co2
+
Yes
* F 17
8/21/74
Emily
co2
+
Yes
* F 18
8/21/74
Emily
co2
+ (a)
Yes
* F 19
8/21/74
Emily
CO 2
+ (a)
No
* F 22
8/21/74
Emily
C02
+
Yes
* F 13
8/21/74
Emily
C02
+
Yes
* F 14
8/21/74
Emily
C02
+
Yes
(a) = CO2 begun day 1;
nutrients
added days 6
- 10.

COMMENTS
1 X	inoculum
1 X	inoculum
1 X	inoculum
1 X	inoculum
1 X	inoculum
1 X	inoculum
1 X	inoculum
1 X	inoculum
5 X	inoculum
20 X	inoculum
33

-------
Table 4 - continued
SERIES &
EXPERIMENT
DATE
LAKE
AGENT
NUTRIENTS
SHIFT
COMMENTS
* G 2
9/24/74
Emily
C02
+
Yes

H
10/74
Emily
C02
+ (b)
Yes
lab
H
10/74
Pleasant
C02
+ (b)
Yes
lab
H
10/74
Powderhorn
C02
+
Yes
lab
H
10/74
Como
C02
+ (b)
Yes
lab
H
10/74
Emily
C02
+
Yes
lab 1 X
H
10/74
Emily
C02
+
Yes
lab 2 X
H
10/74
Emily
C02
+
Yes
lab 4 X
(b) = control also
shifted.




I
10/74
Quarry
C02
+
Yes
lab
I
10/74
Quarry
C02
+
Yes
lab
I
10/74
Quarry
HC1
+
No
lab
I
10/74
Quarry
HCl
+
No
lab
j(d'
j(d2
*	Jd )4
*	J(D5
J(1)6
*	J(D7
7/28/75	Emily
7/28/75	Emily
7/28/75	Emily
7/28/75	Emily
7/28/75	Emily
7/28/75	Emily
C02
-
Partial
co2
-(c)
No
C02
+
Yes
co2
+
Yes
co2
+ (c)
Partial
C02
+
Partial
(c) = CO2 added only briefly at the beginning.
(2)
'(2)
'(2)
'(2)
'(2)
'(2)
7/76
Ardmore
C02
+
Yes
lab
7/76
Lobe
C02
+
Yes
lab
7/76
Ryan
C02
+
Yes
lab
7/76
Half Moon
C02
+
Yes
lab
7/76
Dickey
C02
+
Yes
lab
7/76
Shady Oak N.
C02
+
Yes
lab
34

-------
algae to compete under various conditions. However, it soon became obvious
that the algal shift might have a practical use in lake restoration. It was
also obvious immediately that one cannot lower the pH of a whole lake to 5»5
as a practical procedure, so the extent to which the shift would occur at
higher pH values was of interest. Accordingly, several series of field
experiments were done, all at Lake Emily.
The experimental setup was similar to that used in the pH 5-5
experiments. Control of pH was accomplished through addition of C02 to one
set of bags and HC1 to a second set. The pH adjustments were accomplished as
follows: pH electrodes, emerging through rubber stoppers in the ends of one
foot long sections of one-inch diameter plastic tubing, were inserted several
inches below the surface of each bag to be regulated. Each electrode was
connected to a pH stat housed in a waterproof shelter on the central raft.
Each pH stat, which was wired into a central multiple cam timer, controlled a
stainless-steel solenoid valve. The valves were used to meter either
compressed C02, or HCI (10-30% depending on circumstances), into the bags.
Running to the raft from a small shed on shore were a 110-volt A.C. power
line (connected to a ground fault interrupter), a 1/4-inch polyethylene tube
carrying compressed C02 at 20 PSIG from cylinders in the shed, and a 1/2-inch
compressed air line from a 1/2 HP electric compressor.
Every day at 10 a.m. and 2 p.m. a timer in the shed sent power to the
timer on the raft. Two pH stats were thus activated to deliver either C02 or
HCI (from a pressurized glass bottle on the raft) to the 2 bags to which
their solenoids and pH electrodes were attached, to adjust the pH levels down
to the appropriate pre-set values. At the same time compressed air was
released from tubes on the bottoms of the bags in sufficient volume to mix
the water thoroughly so that the pH adjustment would be accurate. After four
minutes all units shut off and after two more minutes the next pair of bags
was circulated and adjusted. This cycle went on until all the bags were
done, whereupon the whole system shut down until the next cycle. No pH
adjustments were made at night.
Three such graded pH experiments were done in 197^: numbers D, E, and
F, beginning on June 22, July 16, and August 21 respectively. As in the pH
5-5 experiments, N and P were added to all of the bags. The experimental
conditions and results are shown in Table 5-
The results were best in experiment D, where the bags adjusted with C02
to pH 5.5, 6.5. 7-5» and 8.5 all showed the shift, although it was incomplete
at pH 7.5. The bags adjusted with HCI showed the shift at pH values of 5-5i
6.5. and 7-5- The bag adjusted to pH 8.5 with HCI retained its blue-green
population, as did the control bag, #12, where the pH exceeded 9.5 as a
result of photosynthesis.
In experiment E the bags adjusted to pH 6.5 and 7-5 with C02 failed to
shift and the pH 8.5 C02 bag and pH 7-5 HCI bag showed partial shifts only.
The control remained blue-green.
In experiment F both pH 5-5 bags shifted. Neither bag at pH 6-5 did so.
At pH 7.5 one bag remained predominantly blue-green and the other showed a
partial shift. Of the bags at pH 8.5 one did not shift and the other did.
However, the former actually was at pH 7-5 and the latter at 6.5 because of
poor control by the pH stats. The control remained blue-green.
Thus the results were not clear-cut and seem to show an element of
chance. However, they do show that in nine out of sixteen cases the change
occurred, completely or partially, at pH values above 5-5 and at values as
high as 8.5. Thus the possibility of shifting algal populations from
blue-greens to greens as a practical measure is reinforced.
35

-------
TABLE 5
Experiments in which pH was maintained at predetermined values. N and P
added in all cases.
SERIES &
EXPERIMENT
DATE
pH
AGENT
SHIFT
D 2
6/22/74
5.5^
C02
Yes
D 4
6/22/74
6.5
CO2
Yes
D 5
6/22/74
7.5
CO2
Partial
D 6
6/22/74
8.5
C02
Yes
D 11
6/22/74
5.5^b^
HC1
Yes
D 7
6/22/74
6.5
HC1
Yes
D 8
6/22/74
7.5
HC1
Yes
D 9
6/22/74
8.5
HC1
No
D 12
6/22/74
>9.5^
-
No
E 2
7/16/74
5.5^
C02
Yes
E 3
7/16/74
6.5
C02
No
E 5
7/16/74
7.5
co2
No
E 7
7/16/74
8.5
co2
Partial
E 8
7/16/74
7.5
HC1
Partial
E 1
7/16/74
>9.5^
-
No
F 2
8/21/74
5.5(a)
co2
Yes (d)
F 3
8/21/74
5.5^
C02
Yes (d)
F 5
8/21/74
6.5 (7.0)
C02
No (d)
F 6
8/21/74
6.5 (7.0)
C02
No (d)
F 7
8/21/74
.7.5
C02
No (d)
F 8
8/21/74
7.5
co2
Partly(d)
F 9
8/21/74
8.5 (7.5)
co2
No (d)
F 10
8/21/74
8.5 (6.5)
C02
Yes (d)


>9.5^
-
No (d)
( )	=	actual pH
(a)	=	excess CO2, no stat
(b)	=	HC1, no stat
(c)	=	no stat, high pH from photosynthesis
(d)	=	1 x inoculum
36

-------
Biological Details of the Shift
At this point it is worth considering the nature of the shift in some
detail. Table 6 is a listing of the algae predominating at the beginning of
each series of experiments, and of the algae making up the bulk of the
population after the shift had occurred. The point of greatest interest here
is that filamentous and gelatinous blue-greens give way to mostly greens of a
few restricted genera. The general effect is best portrayed by a photograph
as in Figure 8. In the case shown (Expt. D2 Table the shift occurred at a
pH of 5.5 and resulted in mostly Chlorella and Scenedesmus. The same initial
algal population was subjected to pH 8.5 (Expt. D6 Table 5) and again the
resulting algae were mostly Chlorella and Scenedesmus. These changes at pH
8-5 are summari2ed in Figure 9.
Frequently there is more than one species of the dominant green genus.
For example, in the pH 8.5 experiment described above, following the shift
there were no fewer than 22 species and subspecies of Scenedesmus alone.
Rate of the Shift
The rate at which the shift occurs is variable. We believe that, in
general, experiments done in the late spring and perhaps in the fall result
in quicker shifts than do those done in mid-summer. This may be related to
temperature, but it may also be because a larger inoculum of green algae is
present earlier and later. In one case (Series F), suspecting too few green
algae to begin with, we inoculated all of the bags with green algae (mostly
Scenedesmus and Ch1 ore!1 a) that had resulted from a shift in the previous
experiment. The normal inoculum used was a 2-gallon pail of the "greens" to
one bag, or about 1 volume/125 volumes. To determine the effect of inoculum
size we inoculated two other bags (F13 and Fl^) with 5x and 20x this quantity
respectively. The bags with 1x and 5x inoculum appear to have undergone the
shift after approximately 18 days treatment with C02 and nutrients at pH 5-5.
but the bag with 20x inoculum had shown a clear shift by day 13.
Obviously then, it is necessary to have greens present to grow. But not
all of the greens do respond. On occasion some have not changed in abundance
while others have even decreased in number.
In looking at the effects of pH on the rate of the shift the data are
equivocal. In Experiment D (Table 5) where pH values of 5>5» 6-5> anc' ®*5
were achieved with C02, the shift occurred in 11, 10, and 10 days
respectively. When the pH was maintained at 5-5. 6-5> and 7-5. with HC1, the
shift required 9. 10, and 13 days respectively — suggesting, under these
conditions, a slightly faster response at the low pH.
One fact about the shift is that it often appears to occur somewhat
precipitously after a variable period with no apparent change. For example,
we have on several occasions examined the bags on say, Friday evening, and
decided that little change had occurred, while on the following Monday the
algae seemed to be almost all Chlorella and Scenedesmus.
Resistant Algae
Of perhaps as much interest as what species of algae become abundant is
the question, what species of blue-greens if any are resistant to the shift?
As noted earlier, in 19 of 20 cases where nutrients were added and the pH was
lowered to 5*5 with C02, the shift occurred. However, in several instances
where other pH levels were used, the shift did not occur. Does this mean
that certain species of algae are resistant? Probably not. For example, in
experiments D2 and D6 carried out with C02 at pH 5-5 and 8.5 respectively,
all of the following disappeared to the point where they were below
enumeration levels: Osci1lator ia, Anabaena, Chroococcus, Gomphosphaer iurn.
37

-------
TABLE 6
Algae predominating before and after the shift.
(* = shift done with chlorine instead of pH change)
Experiment Series
B
E
F
Emily
Emily
Pleasant
H*
Pleasant
Powderhorn
Como
Como
(1)
Algae before shift
Oscillatoria
Anabaena
Microcystis
Oscillatoria
Anabaena
Aphanizomenon
Oscillatoria
Anabaena
Gomphospheria
Oscillatoria
Aphanizomenon
Oscillatoria
Oscillatoria
Scenedesmus
Oscillatoria
Scenedesmus
Oscillatoria
Ankistrodesmus
Oscillatoria
Ankistrodemus
Oscillatoria
Scenedesmus
Oscillatoria
Scenedesmus
Oscil1atoria
Scenedesmus
Chroococcus
Microcystis
Anabaena
r(D
Chroococcus
Microcystis
Anabaena
Algae after shift
Scenedesmus
Chlorella
Chlorella
Dictyosphaerium
Scenedesmus
Chlamydomonas
Chlorella
Scenedesmus
Coelastrum
Scenedesmus
Chlorella
Scenedesmus
Stichococcus
Chlorella
Scenedesmus
Chlorella
Chlorococcus
Scenedesmus
Chlorel la
Oscillatoria
Mougeotia
Scenedesmus
Scenedesmus
Chlorella
Scenedesmus
Chlorella
Nitzschia
Chlorococcus
Cosmarium
Selenastrum
Dictyospherium
G1eocysti s
Cryptomonas
Ankistrodesmus
Selenastrum
Scenedesmus
38

-------
Table 6 - continued
Experiment Series
^(2) Ardmore
J(2) Lobe
J(2) Ryan
°(2) Half Moon
J(2) Dickey
^(2) Shady Oak N.
Algae before shift
Aphanizomenon
Oscillatoria
Merismopedia
Anabaena
Gleocystis
Anabaena
Tribonema
Microcystis
Oscillatoria
Chroococcus
Oscillatoria
Chroococcus
Algae after shift
Closteriopsis
Chlorella
Chlorella
Ankistrodesmus
Chlorella
Ankistrodesmus
Ankistrodesmus
Nitzschia
Closteriopsis
Selenastrum
Scenedesmus
Sphaerocystis
In certain of the experiments Phormidium may have been present with
Oscillatoria. We have used the name Oscillatoria alone.
39

-------
F1g. 8. Photographs showing the shift
induced by addition of N & P and CCL to a
pH of 5.5 (Expt. D2, Table 4). L
40

-------
and Aphanizomenon. Thus, these species were vulnerable at both pH levels.
However, examination revealed that Osci1latoria was beginning to reappear
after a week at pH 5-5 but remained absent at pH 8.5. Apparently, as with
most organisms, some individuals are more resistant than their fellows and,
if in a particular experiment the shift does not occur, it probably is for
some reason other than that the algae are resistant species.
Reversibility of the Shift
Several attempts were made to determine if the shift was reversible,
i.e. if once again raising the pH or allowing it to increase by itself
through photosynthesis would bring back the blue-green dominance. One such
attempt was experiment CI (6) This experiment was begun on J/27/72 with the
algal population consisting of Microcystis. Osci11atoria, Anabaena, and
Aphan izomenon. The enclosure Was treated with N and P and HC1 at pH 5-5 and
by 8/3 Scenedesmus and D i ctyosphaer i um were abundant. By 8/8 Scenedesmus was
predominant^ On 8/8 the pH was raised to 3.2 with KOH and more N and P were
added. On 8/16 Scenedesmus was still the dominant alga but by S/2k Anabaena,
Osc i11ator i a, and Aphan i zomenon had regained dominance.
Perhaps the best such experiment from the technical point of view was
that begun on August 21, 197^- The contents of bag E2, which had been
treated with C02 and N and P at a pH of 5-5. had become predominantly green
(Scenedesmus). between 7/16 and 8/8. From 8/8 to 8/21 the bag was allowed to
sit with no further additions of C02. It remained green. On 8/21 it was
divided into two bags. One had C02 and N and P added as originally. The
other had its pH raised to 9>5 with KOH. The first retained its dominance of
Scenedesmus until the experiment was ended on 9/18, although a few filaments
of Osc i11ator i a did remain. The second bag, with the high pH, behaved
similarly at first but began to develop more and more filamentous
blue-greens. By 9/19 the population, although it still had many Scenedesmus,
was once again rich in filamentous blue-greens.
In another example, bag CIO, which had been shifted to greens at pH 5-5
with C02 and N and P, had its pH raised with KOH but, by chance, only to 6.7.
Some reversion did occur but the population remained about half greens and
half filamentous blue-greens.
Finally, one case exists in which reversion occurred without addition of
KOH. This was experiment C9- Treatment begun on 7/23/72 caused the
blue-greens to shift over to ChI ore 1 I a and Scenedesmus and by 8/16 a notation
was made that the bag contained "all greens, no blue-greens". On 8/25 the
C02 was shut off. By 9/2b Scenedesmus was still abundant but Chroococcus and
many filamentous blue-greens were present. The pH had not been monitored but
undoubtedly it must have risen as a result of photosynthesis.
Mechanism of the Shift
The reason for the shift is still not known. One might presume that
lowering the pH to 5-5 with C02 or HC1 would be, as suggested by Brock
(1973). detrimental to the blue-green algae and in fact might kill them.
However, notwithstanding this possibility, we have been able to culture
blue-greens, including Osc i11ator i a, successfully at pH 5-5- Furthermore,
the shift has been made to occur at pH values as high as 8.5. and surely this
cannot be claimed to be a physiologically damaging pH for blue-greens. The
fact that the shift occurs only sometimes at a pH value above 5*5 also
suggests the view that it is not the pH per se that is important.
Originally we believed that the shift was due to the superior C02 uptake
kinetics of the blue-green algae, and that reversal of competition to favor
the greens was brought about by supplying C02 either directly by bubbling it
41

-------
in, or by lowering the pH so as to decompose the native alkalinity of the
lake waters. Although Long (1976) recently has confirmed this difference in
C02 uptake ability, we do not believe that this is the sole explanation. That
is, the change does not appear to be one in which green algae slowly replace
blue-greens. In most cases where the shift has occurred there has been a
rapid decrease of the blue-greens — as indicated by changes in numbers and
decreases in chlorophyll -- before the greens have begun growing rapidly.
The blue-green decrease and the green increase thus appear to be separate
phenomena. Furthermore, phosphorus and perhaps nitrogen appear to be
involved. When the blue-greens decrease high concentrations of P04-P and
NH3-N are found in the water. These substances are not left over from the
initial addition that is usually made, but apparently have been liberated
from the blue-green cells. That is, in control systems where the pH remains
high and no shift occurs, phosphate does not appear in solution. However,
where the shift occurs, phosphate and ammonia do show up (Fig. 10). Thus, of
42 examples where the data are available, in 17 there was a positive
correspondence between the shift and the release of phosphate. In 18 there
was a positive correspondence between no shift and no release of phosphate;
and in the remaining 7 cases there was neither correspondence. However, in k
of the last 7 the release began but the experiment was terminated, and in
these the shift might have taken place if given time.
Thus the release of nutrients from the blue-greens apparently must occur
before the greens grow. The nutrients then are, of course, used by the
greens. Curiously enough, the addition of nutrients at the beginning of the
experiment, even when greens are rare, facilitates the shift. If N and P are
not added, the shift, even at pH 5-5. is slow and partial. But if N and P
are added, even though they are absorbed rapidly by the blue-greens, the
eventual shift is more dramatic.
In one series of experiments the C02 and nutrients were added beginning
at different times, e.g. C02 from the beginning, nutrients at 6-10 days, or
nutrients at the beginning and C02 from the sixth day, etc. In most of these
cases the shift occurred. However, in one case (F19) where the N and P were
added over the period 6-10 days while the C02 was begun on day 1, the shift
did not occur, although in the duplicate (Fl8) it did.
Experiment J(1) included two bags in which C02 treatment was done for
only about one half hour at the beginning of the experiment. Bag J{1) 2 was
treated in this way but with no N and P and failed to shift. Bag J(1) 6,
which had N and P added showed a partial shift.
The C02 also acted as a nutrient in the sense that generally more algae
developed when C02 was used to lower the pH than when HC1 was used. But this
is understandable as at the high concentrations of phosphorus (about 200
micrograms/1) and nitrogen (1000 micrograms/1), C02 could be limiting,
particularly in a system isolated from the rest of the lake.
One hypothesis about what might be happening to cause the shift is that
the blue-greens are being destroyed by cyanophage or bacteria as a result of
the pH manipulations (Lindmark, 1979. Lindmark and Shapiro, in prep.), and
that once the blue-greens are gone th the few greens initially present no
longer have competition for the nutrients. Furthermore, green algae
generally require a higher concentration of phosphate for rapid uptake
(Shapiro, unpubl.) and now have it available to them. The reason that
nutrients are required may be to stimulate the blue-greens to active growth
in which condition they may be more susceptible to the destructive agent.
This hypothesis is supported by certain observations. For example, when
Osc i11ator i a disappears, the filaments do break up first as though certain
cells were being lysed. Also not all blue-greens are affected. Thus,
42

-------
A
A /I
/>'
/f/jk,
/; /:/
//I / /
A#/
a/Mi)
y'in
ti'j
5 ' /'
Q

J "OT.il
y
-------
Raph i d i ops is, for example, may continue to grow and increase in numbers even
though Osc i11ator i a is decreasing in numbers. Also, even when Osci1latoria
has decreased it sometimes regrows in the same system, e.g. D2, indicating
that it is not the pH, which remains low, that is the important feature.
An observation that bears on the mechanism of the shift concerns the
effect of arsenic on the phenomenon. In earlier work (Shapiro, unpubl.) it
has been shown that whereas blue-green algae distinguish between phosphate
and arsenate, taking up the former and leaving the latter, green algae do not
make this distinction and arsenate slows their uptake of phosphate.
Accordingly, we reasoned, if phosphate transfer from lysed blue-greens to
greens is important, would the addition of arsenate slow or stop the shift?
Two experiments were done to test this. The first was begun on	Two
bags were set up at Lake Emily, both treated with C02 to pH 5>5 and with N
and P added as usual. However, at the beginning of the experiment 100
micrograms/1iter of arsenic as sodium arsenate was added to one of the bags,
with the other serving as a control. Both bags behaved similarly until about
the fourteenth day, i.e. chlorophyll fell to a low level by day 10 and PO^t-P
was found in solution as expected. However, the control bag shifted to
greens with a maximum of chlorophyll by day 17. while the "arsenate" bag did
not reach its peak of green algae until day 2^. Thus the presence of the
arsenate caused a one-week delay although the shift did occur and the final
chlorophyll levels reached were similar.
The second experiment, begun on 8/21/7^, was conducted in the same way
as the first. Again, the control and experimental bags behaved similarly for
the first two weeks. However, the control bag shifted shortly thereafter
with a significant increase in chlorophyll (to about 90 micrograms/liter)
while no shift took place in the presence of the arsenate, and the
chlorophyll reached only 32 micrograms/liter. Thus, in this case the
arsenate completely prevented the shift from occurring.
Finally, in considering the mechanism of the shift, the effects of
chlorine must be taken into account. It had previously been shown that
chlorine limited the rate of phosphate uptake by blue-green algae more than
by green algae (Shapiro, unpubl.). Therefore it was reasoned, if phosphate
is added to a mixture of algae in the presence of chlorine, the green algae
should have an advantage and the shift from blue-greens to greens might occur
even at a high pH.
In the first such experiment, begun	750 micrograms/1 of
chlorine were added to the experimental bag along with N and P while the
control bag had only N and P added. The two bags behaved similarly in that
both developed crops of algae containing approximately 300 micrograms/1 of
chlorophyll. No shift occurred in either one although phosphate was released
and later reabsorbed in the chlorine bag. Both had pH values of 10.
The second chlorine experiment was begun on 8/21/7^ and was done in
duplicate. In both cases the blue-greens were greatly diminished in the
presence of chlorine and there was a shift to greens even though the pH was
between 10 and 11. As in the previous experiment P04 was released in both
experimental bags. Another field experiment in Lake Emily in July 1975 and k
of 5 laboratory experiments on other lakes were all successful in that
chlorine caused the shift at high pH.
Thus, these experiments along with the arsenate work support the idea
that the phosphate uptake by greens and the subsequent growth of the greens
is a phenomenon separate from the demise of the blue-greens. However, the
reason for the blue-green decline still remains unknown. Recently we have
begun using autoradiography with C14 and P32to help elucidate the mechanism.
44
-------
APPLICATIONS
As noted earlier the work described has both theoretical and practical
significance. From the theoretical standpoint the work may support the idea
of better C02 uptake and use kinetics in blue-green algae. It certainly
appears to support the notion that if phosphate becomes available those algae
that are able to out-compete others for it will become predominant. From the
practical standpoint, the work may have great value in lake restoration. As
described in a previous paper (Shapiro et aj_., 1975) we foresee the day when
it will be possible to manipulate small or moderate-sized lakes to improve
them. Part of the manipulation may involve converting the population of
algae from forms inedible by zooplankton to forms that can be eaten by them.
In other words, we consider it possible to bring about a blue-green to green
shift in whole lakes. The relationship of our studies to such shifts brought
about by artificial circulation has already been described (Shapiro et a 1.,
1975) and our continuing studies indicate that the same factors operating to
cause the shifts in the work reported herein are responsible for shifts in
such circulation ventures.
REFERENCES
Brock, T.D. 1973• Lower pH limit for the existence of blue-green algae:
evolutionary and ecological implications. Science 197JU80-ii83•
King, D.L. 1970. The role of carbon in eutrophication. J. Water Poll.
Control Fed. 42:2035-2051.
Long, E.B. 1976- The interaction of phytop 1ankton and the bicarbonate
system. Ph.D. Thesis, Kent State University.
Lindmark, G. 1979* Effects of environmental stresses on the relationship
between P1ectonema boryanum and cyanophage LPP-1. Ph.D. Thesis,
Institute of Limnology, University of Lund, Lund, Sweden.
Lindmark, G. and J. Shapiro. 1980. The effect of pH on the relationship
between a blue-green alga and its virus. In preparation.
Shapiro, J. 1973- Blue-green algae: why they become dominant. Science
179:382-384.
Shapiro, J., V. Lamarra, and M. Lynch. 1975- Biomanipu1 ation: an ecosystem
approach to lake restoration, pp. 85-96. J_n Brezonik, P.L., and J.L.
Fox (eds.), Proceedings of a Symposium, "Water Quality Management
through Biological Control". Dept. Environmental Eng. Sci., Univ.
Flor ida.
45
-------
II.B. THE EFFECTS Of ARTIFICIAL CIRCULATION ON ALGAL POPULATIONS*
INTRODUCTION
Artificial circulation has often been proposed as a method for
controlling alga! blooms in lakes and reservoirs. However, the response of
the phytoplankton to destratification has been quite variable. A number of
investigators have noted changes in species composition, and a shift from
predominantly blue-green algae to green species is often reported (Symons et
aK 1969; Malueg et aK 1971; Haynes 1971, 1973, 1975; SirenkoetaK 1972;
Weiss and Breedlove 1973)- However, in other cases, blue-green species have
increased following circulation attempts (Ridley et aj_. 1966; Lackey 1973;
Knoppert £t aj_. 1970)• Certain species such as Osc i11ator i a rubescens
(Bernhardt 1967, Weiss and Breedlove 1973) and Ha 11omonas (Lackey 1973)
appear to be particularly sensitive to destratificat ion and are quickly
eliminated. Changes in total algal biomass have also been variable, with some
authors reporting an increase (Hooper et aj_* 1952; Ridley et a_U 1966;
Knoppert et aj_. 1970; Drury et aj_. 1975) . and others a decrease (Bernhardt
1967; Robinson et £l_. 1969; Malueg et aj_. 1971) in algal biomass following
destratificat ion. It is clear, then, that whole lake circulation can produce
a variety of responses in the phytoplankton, not all of which are beneficial,
and if circulation techniques are to be used effectively we must first
understand the mechanisms underlying these responses. The primary objective
of our research has been to investigate these response mechanisms. We
recognized that these mechanisms would probably involve complex interactions
among many physical, chemical and biological factors, and that the response
of the algal community could, therefore, only be understood within the
context of a larger theoretical framework consistent with both the
destratification experience of others and our knowledge of the structure and
function of aquatic ecosystems. Accordingly, we began our investigations by
constructing such a framework. Potential response mechanisms were then
identified and evaluated in controlled field experiments.
THEORETICAL FRAMEWORK
Effects of Destratification on Species Composition
Changes in algal species composition usually reflect the differential
effects of various growth factors. A number of these factors can change as a
direct consequence of artificial circulation.
During circulation hypolimnetic water, rich in C02 and nutrients, is
introduced into the euphotic 2one. If C02 is introduced faster than it is
consumed during primary production or vented to the atmosphere it will
accumulate, causing a shift in the carbonate equilibria and a corresponding
drop in pH. The magnitude of this pH drop will depend primarily on the
amount of C02 present in the hypolimnion of a lake and on the buffering
capacity of the water. Several investigators have reported a drop in pH
during artificial circulation and in many cases this change was accompanied
by a shift in the algae from blue-green to green species (Symons et a]_. 1969;
Haynes 1971, 1973, 1975; Malueg et aj_. 1971; Weiss and Breedlove 1973). It
is also possible for the pH to increase or remain high following
destratification if autotrophic consumption of C02 exceeds the rate of
introduction via mixing. Knoppert and associates (1970) reported very high
pH levels and an increase in the abundance of blue-green algae following
destratification. There is considerable support for the idea that
*by Bruce Forsberg and Joseph Shapiro
46
-------
blue-greens have a competitive advantages over green algae when the pH is
high and that the advantage shifts in favor of the green algae in a low pH
environment. King (1970) found that algal communities in sewage lagoons were
dominated by blue-greens when the pH was high and by green algae when the pH
was low. Shapiro (1973) demonstrated that the composition of natural
phytoplankton communities could be shifted from predominantly blue-green
algae to green algae by the direct addition of C02. However, it is not clear
whether these shifts are a result of competition for C02 as suggested by King
(1970) or the result of some other phenomenon. For instance, Lindmark (1979)
has shown that low pH can stimulate cyanophage activity, giving non-bluegreen
species a selective advantage. Hypolimnetic water introduced into the
euphotic 2one during circulation can also contain high concentrations of Si,
N, P, and other growth factors. Several investigators have noted an increase
in the surface concentration of Si after the onset of mixing (Ridley et a 1.
1966; Hedman and Tyley 19&7).
This increased availability of Si may offer a selective advantage to
diatoms which have a unique (Ridley et aj_. 19&6; Hedman and Tyley 19&7)
requirement for this mineral. An increase in the silica concentration during
the mixing of one British reservoir was followed by a bloom of the diatom
Aster ione 11 a formosa (Anon. 1965~66) .
Artificial destratification is often accompanied by an increase in
euphotic 2one concentrations of nitrogen (Brezonik et aj_., 1969» Haynes
1970; arid phosphorus (Hooper et ak 1952, Wirth and Dunst 19^7» Slack and
Ehrlich 1967. Leach et a]_. 1970, Weiss and Breedlove 1973)* However, some
investigators have reported a decrease in TP during mixing (Haynes 1975,
Drury et aj_. 1975. Weiss and Breedlove 1973) presumably due to chemical
co-precipitation with iron (Lee 1970). The response of each algal species to
changes in the availability of phosphorus and nitrogen will depend on its own
physiological requirements as well as competitive interactions with other
species.
Concentrations of Fe, Mn, B, Al, Cu, and H2S may also change following
des trat i f i cat i on (Toetz et £_[. 1972) but the effect of these substances on
algal growth is not well understood.
Changes in water temperature during destratification can also influence
species composition. Surface temperatures generally decrease after the onset
of mixing (Toetz £t aj_. 1972). The rates of many physiological processes
such as photosynthesis should decline with temperature and any differences
between species in these temperature effects will tend to produce changes in
species composition.
Artificial circulation can also influence the light climate of the algae
by decreasing the proportion of the population which resides in the euphotic
zone thereby reducing the relative availability of light. Assuming that the
algae are uniformly distributed in the mixed layer, the proportion of the
population residing in the euphotic 2one is equivalent to the ratio of the
depth of the euphotic zone to the depth of the mixed layer, Ze/Zm. This
ratio can be further described, according to the Lambert Bouger Law, by the
equat ion:
Ze _ 1n(lo/Ize) . 1
Zm ~ (aEc)/Zm + Ew Im
4.6
aEc + Ew Zm
0)
47
-------
where,	¦=	surface irradiarice
Ize	=	irradiance at the depth of the euphotic zone = 1%
Ec	= partial extinction coefficient for chlorophyll a
(m2 mg Chi a-1)
Ew	= residual extinction coefficient of the water (m-2)
a	= area! standing crop of chlorophyll a (mg Chi a m-2)
Ze	-	depth of the euphotic zone in meters
2m	=	depth of the mixed layer in meters.
Equation 1 indicates that as the mixed depth is increased both the depth of
the euphotic zone and 2m will increase. However, the net result will always
be a reduction in the ratio Ze/Zm. Figure 11 indicates the change in Ze/Zm
expected as the mixed depth is increased, assuming constant values for a, Ew
and Ec. It is clear the Ze/Zm will decrease as a continuous function of Zm
and that the most significant decrease in light availability will occur at
shallow mixed depths. This reduction in relative light availability
following destratificat ion should result in lower integral rates of
photosynthesis, TT (mg C m-2 d-1). Although this is not always observed,
Haynes (1973) reported lower integral rates of photosynthesis following
destratificat ion in Kezar Lake, New Hampshire. The importance of this change
to the phytoplankton depends on the relative magnitude of algal loss factors
(e.g. respiration, excretion, sinking and grazing). If Zm is increased to
the point where TT equals the total integral loss rate, L, then growth will
cease. This mixed depth was defined by Sverdrup (1953) as the critical
depth. When the mixed depth exceeds the critical depth the population will
decline. The response of each population to changes in light availability,
then, will be expressed as a change in the ratio, IT /L. If light is the
primary factor limiting photosynthesis and L remains constant, TT /L should
decrease following destratification and the response should be different for
each species. This differential effect of light limitation should influence
species composition during mixing.
Two additional factors which can change as a consequence of artificial
circulation and influence species composition are the loss rates due to
sinking and grazing. In the absence of turbulence, the sinking loss rate for
a population is equal to the terminal sinking velocity divided by the mixed
depth (Bella 1970, Bannister 1974b). As the mixed depth is increased then
the sinking loss rate should decline. In addition, the increased turbulence
resulting from artificial destratification should also reduce the sinking
velocity and the net effect of decreasing the sinking velocity and increasing
the mixed depth should be to greatly reduce the sinking loss rate. This
effect should benefit heavier species, such as diatoms, the most and may
result in an increase in their relative abundance.
Environmental variables affecting herbivorous zooplankton can also
change dramatically after destratificat ion. In lakes with anoxic hypolimnia
zooplankton are normally restricted to the strata above the thermocline.
After destratification, when bottom waters are no longer anoxic, zooplankton
have been observed to increase their depth distribution (Brown et aj_. 1971»
Fast 1971. Brynildson and Serns 1977)- In addition to increasing the volume
of zooplankton habitat, circulation may also reduce the adverse effects of
predation by creating a dark refuge from visual predators and also by
increasing the distance between predator and prey. These changes should lead
to an increase in zooplankton abundance with larger organisms such as
Daphnia, showing the greatest response. Several investigators have reported
an increase in zooplankton abundance following destratification (Riddick
1957; McNall 1971; Brown et al. 197W Brynildson and Serns 1977» Shapiro and
48
-------
4-


•=100 mg Chl-m"2
3-
Ec = 0.02 m2- mgChl-1

Ew= 0.7
2-

1-

0-

0	10	20	30
Z m meters
Fig. II. Theoretical relationship between the
ratio euphotic depth/mixed depth, and the mixed
depth, assuming constant values for Ec, Ew and
aerial standing crop. The figure indicates an
increase in light limitation with increasing
mixed depth.
49
-------
Pfannkuch 1973) < The increase in grazing pressure accompanying this change
can have a great effect on the algal community.
Each of the factors described above can influence species composition
and each can change as a direct consequence of destratificat ion. Many of
these factors are interrelated and all can act at the same time. We have
tried to construct a theoretical framework which reflects the complexity of
these interactions. Figures 12a and 12b summarize what we feel are the most
important mechanisms through which whole lake circulation can influence
species composition. We have tried to emphasize in these figures how the
final result can depend on the mixing rate.
Effects of Destratification on Algal Community Standing Crop
The response of the entire algal community to artificial
destratification is defined here as the change in algal standing crop summed
for all species. Algal standing crop is equivalent to algal biomass and can
be expressed on either a volumetric (e.g. mg Chi a m-3) or an areal
(e.g. mg Chi a m-2) basis.
Bannister (197^b) has shown that the growth of a phytopiankton
population will always be a complex function of many factors, operating
simultaneously, and that the effect of a change in any one factor can only be
understood within the context of a larger theoretical framework which
considers all of them. A number of authors have developed mathematical
models of algal growth in order to provide such a framework (Bella 1970,
Lorenzen and Mitchell 1973* Megard and Smith 197^t Bannister 197^a» Lehman
et a 1 ¦ 1975* Oskam 1978, and see Patten 1968 for a review of pre-1965 models)
and, of these a few (Bella 1970, Lorenzen and Mitchell 1973> Oskam 1978) were
developed specifically to describe the response of algal communities during
artificial circulation. While we also recognized the need for a broad
theoretical framework, we found the latter models lacking in several
important respects. First, these models are all difficult to evaluate in the
field and often require separate laboratory experiments to determine some
model parameters. Second, they usually ignore algal losses due to sinking
and grazing (except Bella 1970) which can be quite important in the community
response. Finally, they either totally ignore (Bella, Oskam) or give only a
crude treatment (Lorenzen and Mitchell) of the effects of nutrient
limitation. In particular, nutrient limitation was described by a separate
empirical yield term in the Lorenzen and Mitchell model which made it
impossible to consider the effects of nutrient and light limitation
simultaneously or to determine the effects of changes in nutrient
concentration which might occur during mixing. In an attempt to remedy these
deficiencies we have developed a new model of algal growth which has the
following features:
1.	All model parameters can be easily evaluated in the field.
2.	All loss factors are considered including respiration, excretion,
sinking, and grazing.
3.	The effects of nutrient and light limitation are considered
simultaneously.
k. The effect of changes in total nutrient concentrations is considered
d i rectly.
In essence, the model treats the algal community as a carbon pool, with
carbon entering the pool through phytosynthesis and leaving through
respiration, excretion, sinking and grazing. Algal growth then is described
as the net result of both production and loss processes.
50
-------
Aerobic Resp j :
of Ovaries ir j •'
Increased;
'SHOCK
pH decrease
CIRCULATION
Species
Shift
Fig, 12 a. A diagrairmatic representation of some
of the results of whole-lake circulation and
their role in causing a shift from blue-green
algae to greens and diatoms.

-------
C02 competition
pH increase
C02 decrease
Nutrient increase
in euphotic zone
increase m
primary productivity
increase in proportion
of blue-green algoe
Inadequate
CIRCULATION
Fig. 12 b. The probable manner by
which inadequate circulation changes
the proportion of algal types.
52
-------
R - B - D	(2)
where, R «= carbon specific growth rate (d-1)
B = carbon specific production rate (day-1)
D * specific loss rate (day-1)
The value of B depends on several additional factors including the
availability of light and nutrients. The effects of these factors will be
considered below. The daily integral rate of photosynthesis If(mg Cm-2 d-1)
can be set equal to the area of a rectangle (Fig. 13) with one side equal to
the value of the maximum daily volumetric rate of photosynthesis (p-max;
mg C m~3 d-1) and the other side equal to the depth 2' (Tailing 1957a.b)•
If * p-max z'	(h)
The parameter p-max can be redefined as the product of the maximum daily
specific rate of photosynthesis (P-max; mg C mg Chl-1 d-1) and the
concentration of chlorophyll a in the layer (c; mg Chi m~3).
T = P-max c 2'	(5)
The parameter z1 is derived empirically and has been found to depend (Tailing
1957a«b) °n the intensity of photosynthetically active irradiance at the
surface (lo), the intensity of active irradiance at the depth 21 (12') » the
extinction coefficient of chlorophyll a (Ec; m2 mg Chl-1), the concentration
of chlorophyll a, c, and the residual extinction coefficient of the water
(Ew; m-1) such that:
In (Io/Iz1)	(6)
Z' = cEc + Ew
The nutrient dependence of light saturated rates of photosynthesis was
recently recognized by Senft (1978). He proposed a hyperbolic function
similar to that of Droop (197^) to describe the relationship between light
saturated rates and cellular nutrient quotas in laboratory cultures.
Similarly, Forsberg (1980) has found that the value of P-max obtained from i n
s i tu incubations with natural populations depends on the ratio of the
concentration of total phosphorus to chlorophyll (TP/c) according to the
equat ion:
P-"1" = CaJ <1 - T& >	<7>
SM*
where, P-max = the maximum daily specific rate of photosynthesis under
conditions of saturating nutrients (mg C mg Chl-1 d-1)
kq = the subsistence quota of TP or the ratio, TP/c, at which
the value of P-max is zero (mg TP mg Chl-1) .
After substitution equation 5 yields:
TT
ln(Io/Iz 1 )c P-
sat
max
cEc + Ew
i_ J
-------
p mflC m 3 d 1
p- max
0
0
Z' p- max
Z'
Fig. 13. A typical photosynthesis-depth pro-
file showing the equivalence between the depth
integrated rate it (mgC m~2 d~^) and the product
p-max V .
54
-------
B is then determined by dividing 7T by the area! standing crop of carbon
©Zm C; mg C m-2).
ln(Io/Iz' )c P-
sat
max
cEc + Ew
1-
Kq
TP/C
Zm9c
(9)
where,0= the carbon:ch 1 orophy 1 1 ratio in the phytopiankton (mg C mg Chl-1)
Zm = depth of the mixed layer (m).
The complete growth model is given by:
1n(Io/Iz')
cEc + Ew"
i _ M
1 TP
(10)
This model is similar to that presented by Megard and Smith (197^) • The only
significant difference is our inclusion of the nutrient dependence terms
which modifies the value of P-W$x. As pointed out by Megard and associates
(1979) this model requires that the algae are uniformly distributed in the
mixed layer and that saturating light intensities are achieved below the
surface of the water column. The upper limit to algal standing crop can be
determined by setting equation 10 equal to 2ero and solving for the
equilibrium concentration of chlorophyll, c*.
ln(lo/Iz') P Hi - DsZmEw	(11)
___	iTsaX
c a +
EcDoZm + (1 n(10/121) P"^ Kq)/TP
This equation shows explicitly how the peak concentration of chlorophyll a
depends on TP, Zm, algal loss factors, several light limitation parameters
and ©.
The equation can be used to predict how a change in any of these factors
following destratificat ion will influence algal standing crop. An increase
in Zm, after destratification, will act to decrease standing crop while an
increase in TP, which often accompanies destratification, will act to
increase standing crop. However, the actual change in standing crop will
depend on the combined action of all of the factors in equation 9. A number
of these factors are sensitive to changes in algal species composition which
might occur following destratification. For example, circulation may result
in an increase in the abundance of blue-green algae (Fig. 12a). Since
blue-green algae generally have lower sinking rates and are considered less
edible than other groups, the loss rate, D, should decrease. This in turn
would increase the yield of chlorophyll by increasing C*. Alternatively,
mixing could result in a shift to green algae and diatoms. Since these
species tend to be more easily grazed (Porter 1973) and have the highest
55
-------
sinking rates (Smayda 1970) this shift would lead to an increase in 0 and a
decrease in C*. Changes in species composition may also influence the values
of kc, Pmax, kq, and even In (I0/I2') since these characteristics can vary
between species. Unfortunately we do not know enough about the variability
in these latter parameters between species to predict how changes in species
composition might affect their values at the community level. Several other
factors can indirectly influence the levels of parameters in equation 10
during mixing. The survival and feeding rates of zooplankton may increase if
the pH is lowered during mixing (Ivanova and Klekowski 1972; Kring and
O'Brien 1976) and this might result in a higher grazing rate and larger value
of D. In addition, the TP concentration may decline if the mixing rate is
sufficient to produce oxidizing conditions in the water column, since these
conditions might cause phosphorus to co-precipitate with iron (Wet2el 1975)
and result in an oxidized microzone at the sediment surface which would
prevent the release of phosphorus from the sediments (Lee 1970). A change in
surface temperature during mixing may also have significant effects on both
P-max (Tailing 1957a,b) and zooplankton survival (D) . In summary, Figures Uta
and 14b provide a diagrammatic representation of the mechanisms through which
whole-lake circulation might influence the level of algal community standing
crop. Again in these figures we have emphasized the potential importance of
the mixing rate to the community response.
METHODS
General Design
The mechanisms proposed above were investigated in controlled field
experiments. Eight experiments were carried out over a period of three years
in two Minnesota lakes. Little Lake Johanna, a small (7-3 ha.)
hyper-eutrophic lake with a maximum depth of 13 meters, was the site of
experiments 1, 2, h, 5, and 7. The remaining experiments (3, 6, and 8) were
carried out on Twin Lake, a slightly larger (15 ha.) eutrophic lake with a
maximum depth of 12 meters. Additional details on the limnology of Twin Lake
can be found in A1 lott (1978). In each experiment "]-)(> polyethylene cylinders
were suspended from outriggers attached to rafts. A diagram of the general
apparatus is shown in Figure 15- The rafts were always anchored in at least
9 m of water. All enclosures were made of 6 mi I extruded clear polyethylene
cylinders with a diameter of 1 meter and a depth of 8 meters. Each was open
at the top, reinforced with PVC tubing on the sidewalls and either open or
closed at the bottom. Open bottom enclosures were used to simulate natural
lake conditions during artificial circulation. These enclosures were held
open at the bottom by weighted (sand-filled) PVC hoops and lowered slowly
from the surface to entrain an undisturbed water column. Closed bottom
enclosures were used to study the effects of selected hypolimnetic
constituents which are brought to the surface during artificial circulation.
These enclosures were first filled with surface water, sunk to position, and
then allowed to stratify thermally by conduction with their surroundings.
(The thermocline was generally at a depth of 3 meters in Little Lake Johanna
and 5 meters in Twin Lake during the experiments.) Once stratified, water
was withdrawn from the hypolimnetic portion of an enclosure and after a
variety of additions the water was then returned to the artificial
hypolimnion. All additions were completed before mixing began and included
various combinations and concentrations of the following substances.
56
-------
Decreased
Ircreosed
Wixsc Deoln
shifi
Decreased
P/R
Increased
Zcoplcik'cn
si;e - rjmbe
Zoopiarkton
Habi'ct
Inc-eosed
CIRCULATION
Decreased
Bicmass
Fig. 14 a. A diagrammatic representation of the
manner in which whole-lake circulation may act
to reduce algal biomass.
57
-------
increased sbundonee

of olgae

mostly

bluegreens

Inadequate
CIRCULATION
fculnef mcrees*
ifi ftJpnoiiC
dec?ecw iri
dissolved o*yg»
in hypolimmo.-
IFKf8GS€ in
prinory proflutfivty
CO? aecreose
decrease ir *ish and
KKsptarklo* hct 'C
pH incense j............
aceose * 57?
select ^
CO j ccjrpelf'icn '
aecreose -n
locplonk'r luruivfl
tncrec&e m
of b-ue-qreen Glgoe
dec-ease n
grajifva -ates I
decrees? foopknMo
$i??s one absence
decease
rornq irtgrn^
Fig. 14 b. Possible ways by which inadequate cir-
culation may result in increased abundance of
algae.
58
-------
P-K2HP04
N-NH^+C 1
A1kali n i ty-NaHC03
C02-C02 gas
The final concentrations of these substances in the artificial hypolimnion
could be controlled to simulate the natural hypolimnetic levels. Enclosures
were mixed with diffused air released from a centrally located polyethylene
air line. The rate of mixing and the depth of the mixed layer in each
enclosure could be varied by adjusting the rate of air flow and the depth of
air release, respectively (Fig. 15). Air flow was regulated at the surface
with needle valves (Hoke, Inc.). The needle valves were connected to a
manifold which received air from either an on-shore compressor (1/2 hp) or a
compressed air tank on the raft. In a few enclosures carbon dioxide gas was
also released at the diffuser head. After the first two experiments an
attempt was made to remove planktivorous fish which were often trapped
accidentally in the enclosures. Before the air was turned on each enclosure
was slowly seined with a 1-meter diameter screen to remove any fish which
might have been entrained. In one experiment planktivorous fish were also
added to several enclosures to reduce the effects of herbivorous zooplankton.
By varying the mixing rate, mixed depth, and additions it was possible to
simulate a wide range of mixing conditions which could then be used in
controlled experimental designs aimed at investigating specific response
mechanisms. A more detailed description of each experimental design is
presented in the text.
Sampling and Measurement
The enclosures were monitored at regular intervals (1-10 days). Special
sampling techniques were employed to avoid disturbing the water column in
stratified control enclosures. Temperature and oxygen profiles were obtained
with a YSI probe. Transparency was determined with a Secchi disc which could
be lowered side-ways and then flipped into a horizontal position to take the
reading. Air flow was measured with a Gilson air flow meter either at the
surface after disconnecting the air line from the needle valve (expts. 1, 2,
3, *0 or at 7 meters with an "in line" meter (expts. 5.	7» 8). Water
samples were taken at the surface in experiments 1 and 2 and with an
integrated tube sampler (Lund 19^9) in experiments 3~8. The depth of the
water column sampled with the tube sampler was constant for each experiment
but was varied between experiments so that the entire euphotic 2one was
usually sampled. Water from integrated samples was routinely analyzed for
TP, alkalinity, pH, acid corrected chlorophyll a, and occasionally for
chlorophyll b, chlorophyll c, N02, N03> NH3, C02, and Particulate Organic
Carbon. Particulate Organic Carbon, Chi a, Chi b, Chi c, were determined
according to Strickland and Parsons (1968). TP was measured by ascorbic acid
methods following persulfate oxidation (Menzel and Corwin 1965) • The cadmium
reduction technique (APHA 1971) was used to determine N02 and N03 and the
Chaney and Marbach (1963) method was used to measure NH3. Alkalinity was
determined according to Standard Methods (APHA 1971)- pH was measured with a
Beckman pH meter immediately after the samples were taken. When C02 was
added to artificial hypolimnia in closed enclosures the C02 concentration was
determined by field titration (APHA 1971)- C02 concentrations were also
calculated from measurements of alkalinity, pH and temperature made on
routine samples (Stumm and Morgan 1970)- A subsample from each surface or
integrated sample was preserved with Lugol's solution for algal counts; 10
mis of each subsample was concentrated by centrifugation and photographed at
59
-------
closed
air line
diffuser
Fig. 15. Experimental apparatus for suspend-
ing polyethylene enclosures.
60
-------
lOOx and 400x with a Leitz Ortholux photomicroscopic system. These
photomicrographs were used to identify sampling dates where major changes in
species composition or abundance occurred. The phytoplankton on these dates
were then counted using an inverted microscope (Utermohl 1958). This
procedure greatly reduced the number of phytoplankton counts necessary to
describe changes in the phytoplankton. Phytoplankton numbers were converted
to algal volume using average cell volume values calculated for each species.
The average cell volume was calculated by determining the dimensions of
10-100 cells, calculating the volume for each cell using a simple geometrical
formula and then averaging the volumes. Cellular carbon for the
autoradiographic analyses was calculated from cell volume using the
regressions of Strathman (19&7) •
Rates of primary production were determined at both the community and
species level in experiments 3 and 6. In experiment 3 samples were collected
from several depths, spiked with C1 ^-bicarbonate and incubated in duplicate
300 ml glass bottles, J_n s i tu. An initial subsample from each depth was also
taken to determine concentrations of total inorganic carbon. This procedure
was found to be too time consuming when incubations were carried out
simultaneously in several enclosures. Consequently a different method was
developed and applied in experiment 6. In this method an integrated water
sample from the mixed layer was collected, spiked with Cli»-bicarbonate and
incubated in duplicate 7 meter clear Tygon tubes (12 mm ID). (The depth of
the euphotic zone was always less than 7 meters.) A subsample of the
original integrated sample was also retained for determination of TP, Chi a,
P0C and total inorganic carbon. Aside from these differences, the procedures
followed in both experiments were essentially the same. The incubations
lasted from l-i< hours at which point Lugol's solution was added to the
incubation mixture to kill the phytoplankton. Community C14 activity was
determined with the acid bubbling technique (Schindler et a_l_. 1972) using an
Amersham Searle scintillation counter. Total inorganic carbon and rates of
carbon fixation were calculated by methods described in Bueltman et a 1¦
(1969)- In experiment 3 the integral rate of photosynthesis (mg C m-2 d-1)
for the community was determined by integrating the volumetric rates of
photosynthesis (mg C m-3) found in the bottles with depth. In experiment 6
the integral rate of photosynthesis for the community was obtained by
multiplying the volumetric rate of photosynthesis found in the Tygon tube
(mg C m-3) by the length of the Tygon tube (meters). In a parallel
incubation using both tubes and bottles the tube method gave an integral
approximately higher than the bottle method. This difference was most
likely due to the inaccuracies inherent in the bottle technique. Since
bottles were placed at discrete depths there was a high probability that the
depth corresponding to the maximum photosynthetic rate was missed resulting
in an underestimate ofJT . This was not a problem with the tube method since
all depths are equally represented. Continuous light readings from a station
5 miles from the incubation site were used to calculate incident irradiance
associated with an incubation period. Daily integral rates of photosynthesis
were then determined from the product.
61
-------
Daily integral photosynthesis
integral photosynthesis
during incubation
daily irradiance
incubation irradiance
Rates of photosynthesis were also determined at the species level using
the technique of track autoradiography. Samples for autoradiographic analysis
were taken from the Lugol's preserved incubation samples. In experiment 3.
subsamples were taken from each depth and combined to give a single
integrated sample. Incubation samples from experiment 6 were already
integrated. Species specific rates of photosynthesis were then determined
according to the method described by Knoechel and Kalff (1975)- The
photosynthetic activity determined from these integrated samples represents
the average activity of the mixed layer population.
When possible, zooplankton tows were taken in the experimental
enclosures at the end of each experiment. Zooplankton tows were also taken
initially in the closed bottom bags before any additions were made. However,
initial tows were not taken in the open bottom bags to avoid disturbing
chemical stratification. Instead, a tow was taken in the lake shortly after
the open bottom bags were lowered and this sample was assumed to be
representative of the populations in the enclosures. Zooplankton samples
were preserved with a sucrose-formalin solution (Haney and Hall 1973)- The
abundance of herbivorous species was determined by direct counts using a
Sedgewick-Rafter cell at lOOx.
Special Analytical Methods
Statistical procedures used throughout are taken from Snedecor and
Cochran (1973)•
The grazing potential associated with each zooplankton community was
calculated using species filtering rates obtained from the literature. A
regression of filtering rate on organism length was determined for
cladocerans using the literature data (r2 «= .60). This regression was used
to calculate filtering rates for cladoceran species when literature values
were not available. A list of species-specific filtering rates and the
regression equation used in the calculation of community filtering rates is
presented in Table 16.
Instantaneous algal growth rates were determined at both the species and
divisions level in experiments 1 and 6 using the following equation:
r = In (A(t2)/A(tl))/(t2-tl)
where r ¦ instantaneous specific growth rate (day-1)
ti « sampling time; (days)
Ati = algal biomass at time i (biomass units m-2)
(12)
These instantaneous specific growth rates, r, should not be confused with
the finite specific rates, R, B, and D, which were presented in the
theoretical framework. Although both types of rates can be used to describe
the growth of an algal population they are not interchangeable. The basic
62
-------
difference is that the finite rates are compounded at relatively large time
intervals while the instantaneous rates are compounded at infinitesima1ly
small time intervals. Algae grow more or less continually and algal growth is
therefore best described by instantaneous rates. We have used instantaneous
rates whenever possible in our analysis. It was necessary, however, to use
finite rates at the community level in experiment 6 in order to evaluate the
model presented in the theoretical framework. The finite specific growth
rate, R, was calculated as:
= A(t2) - A(tl)
(t2-t1) A
(13)
where A ¦ the average level of algal biomass	between sampling times 1 and 2
(biomass units m-2). The finite specific	rate of production, B, was
calculated as: —
B = —	(H)
A
where TT ¦ the average daily integral
dates (mg C m-2 d-l) and the finite
d i fference:
D
rate of photosynthesis between sampling
specific loss rate was obtained by
B - R
(15)
Algal carbon was used as the biomass unit in these
Algal carbon was determined as the product of the chlorophyll
and the carbon to chlorophyll ratio in the phytop Iankton,©.
calculations,
concentrat ion
EXPERIMENTS AND RESULTS
limnology of Enclosures
Before the results of our experiments in polyethylene enclosures could
be compared to those obtained in whole-lake circulation experiments, it was
first necessary to demonstrate that the limnological environments inside and
outside of the enclosures were sufficiently similar. The similarity of these
environments was determined by comparing limnological conditions in the open
bottom control enclosures (not mixed) with those found in the surrounding
lake. Data from experiment 3 were used for this comparison. Figures 16a and
16b show the depth-time distributions of isotherms and isopleths of oxygen
for Twin Lake during this experiment. The thermocline was fairly constant at
approximately 5 meters. A metalimnetic peak in oxygen concentration is
evident during the month of August. This maximum coincided with the
development of a metalimnetic population of Osc iI 1ator i a rubescens . The
disappearance of this oxygen maximum in late August coincided with a slight
increase in the depth of the mixed layer. The portion of the metalimnion
which was eroded at this time apparently contained the photosynthetically
active segment of the Osc i11ator i a population. The depth-time distributions
of isotherms and isopleths of oxygen for one of the open bottom control
enclosures (bag 1) in experiment 3 are shown in Figures l?a and 17b,
respectively. The distribution of isotherms in the control enclosure is
virtually identical to that of the lake. The distribution of oxygen isopleths
in the control enclosure is also quite similar to that of the lake indicating
that the processes of atmospheric exchange and photosynthesis responsible for
63
-------
27
26
24
22
E
20
20
a
¦o
August
	9
E
<.1
August
Fig. 16. Top,(a) isotherms (°C) and bottom,(b) isopleths of oxygen in
Twin Lake during experiment 3.
64
-------
27
E
£
+*
a
4>
-o
August
September
		10
E
^4
Q.
«
¦D
20 O
•<.2'
September
August
Fig. 17. Top, (a) isotherms (°C) arid bottom, (b) isopleths of oxygen
in unmixed control bag 1 during experiment 3.
65
-------
the observed oxygen distributions were similar in both environments.
Significantly, the metalimnetic population of Osci11atoria and associated
oxygen maximum which were found in the lake also developed in the control
enclosure.
The disappearance of the oxygen maximum occurred at approximately the
same time in both systems. A number of other important variables including
Chi a, TP, Secchi depth, pH and C02 also followed similar trends in both the
control bag and the lake (Fig. 25). The correspondence between conditions in
the lake and control enclosures was not as good in Little Lake Johanna. This
is primarily a result of the high variability in limnological conditions
characteristic of this lake. Little Lake Johanna receives a great deal of
storm water runoff from an artificially enlarged drainage basin which during
storms results in large fluctuations in lake chemistry and biology. Once the
experimental enclosures were set in place the enclosed water columns were
essentially isolated from these chemical and biological changes and
limnological conditions within the enclosures tended to diverge from those in
the lake. This was not a problem in Twin Lake because the drainage basin was
small and the lake was relatively undisturbed.
It was also important to determine whether the changes in enclosure
limnology resulting from artificial circulation were similar to those
reported in destratified lakes and reservoirs. Data from experiment 3 were
again used for this purpose. Figures l8a and 18b show the depth-time
distribution of isotherms and oxygen isopleths in an open bottom enclosure
circulated with compressed air which was delivered at a rate of 100 mis
air/minute (measured at the surface). It is evident from the isotherm
distribution that complete thermal destratification was not achieved at this
mixing rate. This was surprising since this rate of air release per unit of
lake surface area (.127 liters per minute/m-2) was almost 13 times the level
recommended by Lorenzen and Fast 097&) (-01 liter per minute/m-2) to achieve
destratification in lakes and reservoirs. Thermal stratification is
apparently maintained in the mixed enclosure by efficient thermal conduction
through the side walls. Flow rates in excess of 1000 mis air/minute were
required to completely destratify the enclosures. However these higher
mixing rates resulted in the resuspension of bottom sediments, which
prohibited their use. Although thermal destratificat ion was not achieved at a
flow rate of 100 ml /min, Figure 18b indicates that sufficient mixing
occurred to produce a uniform depth distribution of oxygen. This mixing rate
was also found to produce a uniform distribution of algal cells, total
phosphorus and pH in the mixed layer. Isometric depth distributions of
chemical and biological variables are also characteristically found in
destratified lakes and reservoirs (Toetz et a_K 1972). We generally used the
lower mixing rates (10-100 ml/minute) in our experiments in order to avoid
resuspending sediments and also because the destratification of chemical and
biological variables which was achieved at these rates was more important
than thermal destratificat ion in our theoretical framework.
During the experiments, there were certain effects on the
phytoplankton which could be traced to the enclosures themselves. It will be
important to distinguish these "enclosure effects" from the effects of the
various treatments in each experiment. One direct effect of enclosure was to
reduce turbulence in the water column. This tended to increase the effective
sinking rates of the phytoplankton in the control enclosures which often had
a lower algal standing crop at the end of an experiment than the lake. In
addition to reducing turbulence, the walls of the enclosures also provided a
substrate for the growth of periphyton. These periphyton competed with the
phytoplankton for light and nutrients introducing some ambiguity into the
66
-------
22
27,
26
3-
22
E
	18
a
TJ
— 6
September
August
air
on
101214 16^16 14
12
1291
<¦2
<.2
August
September
on
Fig. 18. Top, (a) isotherms (°C) and bottom, (b) isopleths of oxygen
in bag 1 (100 ml min~') during experiment 3.
67
-------
interpretation of experimental results. The periphyton growth was most severe
in the deep mixed enclosures because of the high nutrient levels produced in
the euphotic zone. Periphyton growth was noticeable after the first two
weeks in most experiments and changes in the phytoplankton which occurred
after this period were therefore considered less significant. These and
other "enclosure effects" will be addressed in the text when they influence
the interpretation of experimental results.
Experiment 1
Our purpose in this experiment was to simulate the range of mixing
conditions commonly found in artificially circulated lakes and to determine
the response of the phytoplankton to these conditions. Open bottom enclosures
were used to entrain "natural" water columns. Different mixing conditions
were simulated by varying both the rate and the depth of mixing. The mixing
rate was controlled by regulating the rate of air flow through the diffuser
and the mixed depth was determined by the depth of air release. The shallow
mixed depth was set at 3 meters to coincide with the depth of the mixed layer
before circulation. This was done to prevent the introduction of
hypolimnetic water into the mixed layer so that the direct effect of
turbulence could be examined separately. The experiment was conducted on
Little Lake Johanna during the period June 19 - June 28, 1976. Seven open
bottom enclosures were suspended from two connected outriggers. The
treatment associated with each enclosure is given in Table 7- Samples were
taken at the surface and analyzed for TP, pH, alkalinity, C02, and algal
biomass, both as Chi a and algal volume. Temperature profiles, D.O. profiles
and Secchi disc readings were also taken at regular intervals. Zooplankton
tows were taken initially in the lake and in all of the enclosures at the end
of the experiment. The depth of air release was raised to 7 » on June 22 in
bags 2, k, 5. and 7 when it was evident that sediments were being stirred up
by the diffusers. The experimental results indicate that we were able to
simulate a wide range of mixing conditions.
Complete thermal destratification was achieved in bag J at a mixing rate
of 1000 ml/min. Unfortunately this rate also resulted in the resuspension of
bottom sediments which continued even after the diffuser head was raised to 7
meters. The air release rate of 100 mls/min. in bags k and 5 was not
sufficient to produce thermal destratificat ion. However, this rate of mixing
was rapid enough to produce a uniform depth-distribution of all chemical and
biological parameters. The air release rate of 20 mls/min. maintained in
enclosure 2 was too low to eliminate even chemical + biological
stratification although considerable mixing did occur.
Changes in surface values of pH, C02, TP, SD and algal biomass
associated with each treatment are shown in Figure 19. There was an inverse
relationship between C02 and pH in this and all subsequent experiments which
was due to interactions within the carbonate buffering system. There is also
a consistent inverse relationship between S.D. transparency and chlorophyll
concentrations in each experiment as predicted by the Lambert-Bouguer Law
(Megard et a]_. 1979)' C02 increased and pH decreased as the mixing rate
increased in deep mixed bags (2, k, 5. 7). TP also increased in these
enclosures as the mixing rate increased resulting in higher chlorophyll
concentrations and lower transparency. The levels of C02, pH, TP, and SO
achieved at 1000 mg/min. (bag 7) were very similar to those attained at 100
mls/min. (bags 4 and 5) indicating that complete mixing occurred at 100
mls/min. Chlorophyll concentrations were lower in bag 7 than in bags k and 5
even though the TP levels were similar. This may indicate some inhibition of
68
-------
Table 7: Treatments for enclosures in Experiment 1, Little Lake Johanna
Air Flow
(ml nn'n"^ Air release
Ba^
Air on
Air off
@ surface)
depth (m)
Text designation
1
--
--
0
__
Control
2
6/19/76
6/28/76
20
8
Deep slow mixed
3
"
u
20
3
Shallow 20
4
n
M
100
8
Deep 100
5
M
11
100
8
Deep 100
6
H
tl
1000
3
Shallow 1000
7
II
11
1000
8
Deep 100
69
-------
algal growth at the highest mixing rate. The levels of most parameters in
the shallow mixed enclosures (bags 3 S 6) did not differ greatly from the
control and lake values, although chlorophyll a concentrations were somewhat
lower in the control bag (bag 1). The average chlorophyll a concentration
determined for each enclosure was an increasing linear function of the
average TP concentration as indicated in Figure 20. (The numbers on this
figure refer to the different enclosures.) The regression was significant at
the .05 level. The fast-deep mixed enclosures all had relatively high
average TP and chlorophyll values. The average Chi a value in the fastest
mixed enclosure (7) fell below the regression line indicating possible
inhibition. The control bag, shallow mixed bags, slow mixed bags and the
lake all had relatively low average TP and Chi a values. The control bag (1)
also fell below the line. Reduced turbulence in this enclosure may have acted
to increase the effective sinking losses of the algae and lower the yield of
chlorophyll. An increase in the mixed depth of the fast-deep mixed bags
should have resulted in greater light limitation. The negative effect of
light limitation on algal growth was apparently offset by the positive effect
of increased nutrient availability in these bags.
Since the experiment was relatively short, the final species composition
in each enclosure was generally similar to the initial composition (cf. Fig.
19) which made it difficult to determine the differential effects of
treatment conditions at the division level. This problem was overcome by
using the instantaneous specific growth rates for each division instead of
the final species composition as an indicator of the algal response.
Instantaneous specific growth rates were calculated for the diatoms, green
and blue-green algae in each enclosure for the experimental time period of 10
days. The calculations were based on the change in algal volume summed for
all species within a division. These growth rates are shown in Figure 21
along with average values of alkalinity, pH, C02, temperature, TP measured
and phosphorus availability measured at the surface during the same period.
The ratio TP/Chl a was chosen as a measure of phosphorus availability since
it was shown (Forsberg 1980) to be analogous to the cellular nutrient quota
term, particulate phosphorus/chlorophyll a, used by Senft (1978).
There was a general increase in the growth rates for all divisions in
the shallow mixed bags (3 and 6) as the mixing rate was increased. Although
there were small increases in TP and C02 in these bags, the average levels of
phosphorus availability did not increase significantly over the control bag
and this growth response apparently reflects the direct effect of increased
turbulent mixing. While the mechanism involved is not clear, the effect of
increased turbulence may have been to reduce the sinking rates of the algae.
That the greatest stimulation in growth was found for the heavier diatoms
seems to support this hypothesis. The additional effects of deep mixing can
be seen in the results for bags 2, k, 5> and ].
Green algae and diatoms showed a similar response to deep mixing. Their
growth rates both increased with the mixing rate reaching a peak at 100
ml/min and then declined again at the fastest mixing rate (bag 7). However,
the diatoms showed a greater increase than the green algae. The blue-greens
showed a different response pattern with their growth rates dropping steadily
as the mixing rate was increased. The decline in growth rates for all
divisions in bag 7 may have been related to the low surface temperatures
produced in this enclosure by thermal destratification. (This may also
explain the lower yield of Chi a found in bag 7 cf. Fig. 20.) With the
exception of the results from bag 7> the growth rates of greens and diatoms
generally exceeded those found at similar flow rates in the shallow mixed
enclosures and this increase was probably due to the higher levels of
70
-------
Iftka
1
fit
«•?*
T*
bag 1 control
4*1 y 4* j is
l*L. Mi
bag 2 20 datp
•*ri *»y»
l^n A.
bag 3 20 ahaitow
•ayi OyM





	

V












		 "



—	-

,




dftyi
•tf on ^
10 1
#
10 1
e
'e *'
I 2
0'
n
V50
o 25
•
S
0
V
s 2
fi 3.
Y 10000'
E
a iooo
• IOC'
E 1.
n
'e 100-
cf ">¦
% .•
E 0.1.
»-
bag 4 100 daap
1
lUn

baas 100 d««p
"r' i"±
	1L
day 1
iir on W
10 1
bag6 1000 ihatlow
<«ri
ritA

bag y 1000 daap
<¦* i	«r io

Jk.
r
10 1
#
'V
Fig. 19. Changes in pH, C02> TP, Secchi disc transparency
chlorophyll a_ and algal volume in the experimental
enclosures and in the lake during experiment 1,
(g = greens, b = blue-greens, o = other).
71
-------
nutrient availability observed in the deep mixed bags. A similar increase
was found for blue-greens in the slow mixed enclosure (bag 2). However, the
growth rate for blue-greens decreased at 100 mls/min (bags k and 5) in spite
of high levels of nutrient availability. This inhibition of growth was
apparently related to the low pH levels found at this mixing rate.
These results indicate that each division responds in a unique way to
artificial circulation. This differential growth response can have an
important effect on competitive relationships. Kalff and Knoechel (1978)
have suggested that the competitive advantage of one group over another can
be viewed as the difference between their observed growth rates. When the
difference is positive the first group has the advantage because it grows at
a faster rate but when the difference is negative the advantage shifts to the
other taxonomic group. Applying this criterion to Figure 21 it is clear that
diatoms maintained a competitive advantage over both greens and blue-greens
in all of the enclosures. Similarly, blue-greens outcompeted greens under
all treatment conditions. However, it should be noted that the competitive
advantage of blue-greens over greens was dramatically reduced at the faster
deep mixing rates. This latter shift in competitive advantage was the
combined result of the stimulation of greens and inhibition of blue-greens
found under these conditions.
The initial and final quantities of herbivorous zooplankton for
experiment 1 are shown in Figure 22. There was a dramatic reduction in
herbivore abundance in all enclosures by the end of the experiment. However,
this response cannot be attributed to any treatment effect since the
zooplankton also crashed in the lake.
In summary, the results from this experiment demonstrated the advantages
of using a controlled experimental approach. We were able to create a wide
range of mixing conditions resembling those commonly found in destratified
lakes. The range of algal response associated with these conditions was also
similar to that described in the destratification literature and the results
were generally consistent with our theoretical framework. Slow deep mixing
resulted in increased TP and pH levels at the surface. These conditions
increased the competitive advantage of blue-green over green algae.
Rapid-deep mixing resulted in low pH and high TP levels at the surface. These
conditions increased the competitive advantage of diatoms over blue-greens
and reduced the advantage of blue-greens over green algae. However, the
blue-greens maintained a slight advantage over the greens so a quantitative
shift
toward greens was not observed. A general increase in total algal standing
crop was correlated with an increase in TP in the rapidly mixed enclosures.
This increase in biomass occurred despite an increase in the mixed depth
indicating that the depressive effect of light limitation on algal growth was
less important than the positive influence of nutrient enrichment. Although
these results were encouraging several important limitations of the
experimental design were encountered. Our inability to attain thermal
destratification at relatively high mixing rates presented a serious problem.
Because of the limitation we could not investigate the direct effect of
thermaI destratification on the algae. However it was possible to achieve
isometric depth distributions of most chemical and biological variables and
since these were the changes most likely to influence the phytoplankton we
decided to continue using these flow rates in subsequent experiments. Another
limitation of the experimental design was that it was difficult to separate
the effects of different hypolimnetic factors such as C02 and TP. This
problem was primarily a result of using open bottom bags. The hypolimnion
naturally contained a large number of chemical factors which could affect the
72
-------
20-
Chl
mg m
10

5
Experiment 1


4

7
i
r2 = 0.87

Chi = 0.013 TP + 4.46
500
TP mg m
1000
Fig. 20. Relationship between average concentra-
tions of TP and chlorophyll a^ in the experimen-
tal enclosures and in the lake during experiment
1 (numbers refer to different enclosures, L =
lake).
0.6
0.5
0.4
0.3-
b»9
flow rate ml min
20
1000
20
100
1000
100
meters
8.9
8.6
8.9
7.2
7.2
7.1
8.7
TP ug l
49
55
74
807
1096
946
66
69
69
107
71
115
CP2 uM 1-1
227
294
370
Temp C
23.0
23.0
12.4
24.5
23.2
24.2
22.7
9.0
9.3
8.8
43.5
31.3
66.7
Fig. 21. Composite instantaneous growth rates
calculated in each enclosure for the first 10
days of experiment 1. Physical and chemical para-
meters shown represent the average values in
each enclosure for the 10-day period.
-------
E « perl merit \
Experiment 4
tfay 1 fifty 30
beg
Experiment 6
20«
in
o
n
I
E
«
CB
m
20

Experiment 7
o
>
2
z
beg
beg
Fig. 22. Changes in the abundance of herbivo-
rous zooplankton observed in the enclosures and
lake during experiments 1, 2, 4, 5, 6, 7, and 8.
74
-------
algae and these factors all tended to co-vary when they were brought to the
surface during mixing. An attempt was made in later experiments to separate
the effects of several of these factors using closed bottom enclosures and
controlled hypolimnetic additions.
Experiment 2
The second experiment immediately followed the first (July 6 - Aug. 28,
197^) and was also carried out on Little Lake Johanna. The results of
experiment 1 suggested that changes in pH, C02 and TP during mixing were
important factors in the algal response. Consequently, experiment 2 was
specifically designed to investigate the response of the phytoplankton to
changes in pH, C02, and TP levels. A range of C02, pH, and TP levels was
created primarily by varying the air flow rate which determined the rate at
which hypolimnetic TP and C02 were introduced into the euphotic zone. C02
was also added directly to some enclosures to obtain a very low pH. The
influence of the rate of change in pH on the phytoplankton was also
investigated in this experiment to test the hypothesis that a very rapid drop
in pH might shock the blue-green algae resulting in some physiological damage
perhaps making them more susceptible to cyanophage infection. To simulate
this effect two of the enclosures (b and 10) were circulated very rapidly for
a brief period of time at the beginning of the experiment and were then
allowed to restratify. All treatments are summarized in Table 8. Treatments
were randomly assigned to the enclosures. Routine sampling procedures and
analyses were identical to those used in experiment 1. Several treatment
changes were made during the course of the experiment. Airflow was increased
to 20 mis per minute in enclosures 2 and 7 on day 7 in an attempt to
increase the mixing rate slightly. C02 flow was increased to 150 mis. per
minute in bag 6 on day 3 in order to produce a lower pH.
Trends in the surface values of chemical and biological variables in
each enclosure are shown in Figure 23- As in the first experiment, pH
decreased and C02 increased at the faster deep mixing rates. The greatest
drop in pH was observed in enclosure 16 which was supplied with additional
C02 gas. The pH in this bag dropped from S.h to 5-8 in the first seven days
and remained low throughout the experiment. Once the pH dropped in the
continuously mixed bags it remained low throughout the experiment. As
expected, the most rapid pH drop was observed in bags k and 10. Thermal
destratification was achieved after only 22 minutes in these enclosures. The
observed drop in pH during this period was from 9-^ to 6.9 in bag I4 and from
S.k to 7.0 in bag 10. C02 and pH returned to control levels in these bags
after only 5 days and remained there for the rest of the experiment. pH
levels remained high in the slow mixed and control enclosures.
Surface values of TP also increased, as in the first experiment, with
increases in the mixing rate. TP increased quickly in the rapidly mixed
enclosures (1», 5» 6, 8, 9 2nd 10) and gradually in the slow mixed enclosures
(2 and 7)* Once the TP reached its peak in the continuously mixed enclosures
it tended to decline very slowly. The decline was much faster in enclosures
and 10. This latter effect was most likely due to the quiescent conditions
and presumably greater sinking rates which occurred in these bags after they
restratified. As in the previous experiment, concentrations of chlorophyll a
were correlated with TP. Figure 2k shows the linear relationship between
average TP and Chlorophyll concentrations determined during the experimental
period. The regression is just barely significant at the 5% level. Most of
the variance in this relationship is due to the slow mixed bags (2 and 7) •
These enclosures were both dominated by the blue-green alga Hicrocyst i s
75
-------
Table 8: Treatments for enclosures in Experiment 2, Little Lake Johanna
Bag Air on
1
Air flow C02 flow
(ml min-"! (ml min"'
Air release
Air off @ surface) § surface) depth (m)
0
Text designation
Control
2
7/10/76
8/28/76
10(20)
—
7
20 deep
3
7/10/76
8/28/76
100
--
3
100 shallow
4
7/10/76
(22 min)
7/10/76
4000
1000
7
4000 deep
instantaneous
5
7/6/76
8/28/76
100
--
7
100 deep
6
7/10/76
8/28/76
50
50(150)
7
50 deep + CO2
7
7/10/76
8/28/76
10(20)
__
7
20 deep
8
7/10/76
8/28/76
100
—
7
100 deep
9
7/10/76
8/28/76
50
—
7
50 deep
10
7/10/76
(22 min)
7/10/76
4000
--
7
4000 deep
instantaneous
Table 9: Treatments for enclosures in Experiment 3, Twin Lake
Bag
1
2
3
4
5
6
7
8
Air on Air off
7/30/76 9/20/76
Air flow
(ml min~1
@ surface)
0
100
20
100
0
50
20
50
Air release
depth (m)
7
7
7
7
7
7
Text designation
Control
100
20
100
Control
50
20
50
76
-------
1
JZ
V
10
'e 1000
ts 100
S '0
E 1
0.1
7
5.5
. bag 1
big 3
fV
bag 5
.

9

;-wvA






r~~

^—

Vv^
control
100 thatiow
pHahock +C02
100 deep
0, 25 50 I
15 "J
^ 25 SO l
25 50
j.. b»g6
M
-fc
k bag 8
IV
bag 9
i '
bag 10
¦'IL,


9



	
.!
r-



r*v_


50 + CO2 deep
:L~/
u/^
100 deep
SO  t
o
o>
E
£
D
O
O
Fig. 23. Changes in pH, CO2, TP, Secchi disc
transparency, chlorophyll a_ and algal volume
in the experimental enclosures and in the lake
during experiment 2 (g = greens, b = blue-greens,
d = diatoms, 0 = other).
77
-------
20-
15
10-
S-
0
300'
200'
o
V)
a.
t-
P0
I
E
(M
O
u
»
E
10
10'
0.1
X
a
5.5
days
take
¦e
bag 2
IN.
bag 7

















—\
Ilk*
20 daap
20 daap
¦ir on
*
25
50
^ 25 50 ^ 25 50
Fig. 23. Cont'd. Changes In pH, C02» TP, Secchi
disc transparency, chlorophyll a and algal volume
in the experimental enclosures and in the lake
during experiment 2 (g = greens, b = blue-greens,
d = diatoms, o = other).
78
-------
150-
Exp 2

7
100-

Chl

mg m 3
2
50-
jr 5

/ 4 „

9 _ t - 0.36

3 10 8 Chi = 0.128 TP + 10.27
n-
1
0	500	1000
TP mg m~3
Fig. 24. Relationship between average concentrations of
TP and chlorophyll a in the enclosures during
experiment 2 (numbers refer to different
enclosures).
79
-------
aeruginosa which, due to its buoyancy, was able to accumulate near the
surface. This resulted in a shallow effective mixed depth for this
population which, according to equation 11 should have resulted in a higher
yield of chlorophyll a per unit of total phosphorus. It is significant to
note that bag 7« which deviated the most from the regression line also had
the largest bloom of M i crocyst i s aeruq inosa . This experiment was long enough
in duration so that changes in species composition associated with the
different treatments could be easily discerned. However, caution had to be
exercised in interpreting phytoplankton changes which occurred late in the
experiment because bag effects such as periphyton growth (predominantly
diatoms) became increasingly more important at this time. For this reason
the analysis presented here will be restricted primarily to the first half of
the experimental period. All enclosures were dominated initially by green
algae (refer to Figure 23, predominantly by species ofSphaerocyst i s,
Schroeder i a. and Coelastrum spp.) . There was a shift to blue-green algae in
the slow mixed bags W and 7) » which had both high pH and TP levels. As
indicated above, Mi crocyst i s aerug i nosa became the dominant species in these
enclosures. The response of blue-greens to low pH was inconsistent in this
experiment. Blue-greens declined in bag 5 but only after an initial peak.
The biomass of blue-greens remained fairly constant in bags 6 and 9 while
their relative abundance increased dramatically in bag 8. There was some
evidence that the rate of drop in pH may be important in determining the
blue-green's response. Blue-greens were almost completely eliminated in bags
k and 10 (determined from photomicrographs) shortly after the initial rapid
drop in pH. However, they appeared to recover once the pH and C02 levels had
returned to the control levels. The response of the green algae to low pH
was also less consistent than in the first experiment. There was a small
peak of chlorophyll dominated by green algae shortly after the pH drop in
enclosures h and 10 (determined from photomicrographs). These peaks
disappeared, however, after the pH and C02 returned to control levels. The
relative abundance of green algae also increased in bag 5 but decreased in
bags 6, 8, and 9 in spite of the low pH levels observed. Diatoms did not
appear to out-compete the other divisions as well as they did in the first
experiment. There were, however, two peaks of diatoms (identified from
photomicrographs) in enclosure 16 which had the highest C02 and TP levels.
These peaks correspond to the two chlorophyll maxima shown in Figure 2}. In
addition, there was an increase in the relative abundance of diatoms in bag 5
which also had high C02 and TP levels.
Changes in the abundance of herbivorous 2ooplankton in the experimental
enclosures are shown in Figure 22. The zooplankton standing crop in the
lake, the control (bag 1) and the slow mixed enclosures (2 and 7) all
increased to approximately double the initial lake value. There was a
decrease in herbivore standing crop in the shallow mixed bag (bag 3). the
rapid mixed bag with the lowest pH (bag 6) and both bags which received the
rapid pH drop treatment (bags k and 10). The greatest increase in herbivore
standing crop occurred in the three rapid mixed bags which experienced only a
moderate drop in pH (5, 6, and 9) •
It is significant to note that the bags with the highest herbivore
standing crop (5, 8, and 9) also fell considerably below the TP vs. Chi
regression line in Figure 2k indicating that the lower yield (Chl/TP) in
these bags may have been due to increased grazing pressure. Selective grazing
on the more edible green algae may also have prevented the blue-green to
green shift from occurring in bags 8 and 9-
Several of the results in this experiment were consistent with the
theoretical framework. Slow deep mixing resulted in high surface values of TP
80
-------
arid pH favoring the development of blue-green algae. Blue-green algae
decreased and green algae increased shortly after a rapid drop in pH and
increase in TP in bags k and 10. There was also a slow shift in relative
abundance from blue-greens to greens associated with high TP and low pH in
bag 5- At the community level, algal standing crop increased with TP in the
mixed layer as predicted by equation 11. Low pH levels generally led to
increases in zooplankton abundance. There were a number of results however,
which did not appear consistent with our theoretical model. A shift in
relative abundance from blue-green to green algae did not always occur in the
bags with low pH and high TP levels. This may have been the result of the
increased activity of herbivorous zooplankton in these bags. The zooplankton
may have been selectively grazing the green algae which would have prevented
the shift. This hypothesis was investigated in subsequent experiments. It
was clear from experiment 1 that blue-greens can lose some of their
competitive advantage to green algae at low pH and high nutrient availability
and yet still remain competitively superior. This may explain the absence of
a blue-green to green shift in some of the rapidly mixed enclosures in
experiment 2. The response of zooplankton to low pH was also inconsistent.
Extremely low pH apparently reduced zooplankton survival in bag 6. According
to our framework (Fig. lita) a reduction in pH should increase zooplankton
survival. It is more likely, though, that there is probably an optimum pH
for feeding and survival, below which survival declines. Evidence from the
literature (Ivanova 1969; Kring and O'Brien 1976) suggests that this optimum
is usually quite low. For example Ivanova found that the optimum pH for
filtering in Cer i odaphn i a was 7 and the rate drops steeply both above and
below this point. It is quite likely then that the unnaturally low pH
attained in bag 6 (low of 5-8) was below the optimum for most herbivores.
Experiment 3
Our purpose in experiment 3 was to determine whether the general
response of the phytopiankton to artificial circulation differs significantly
between lakes. Twin Lake, another small eutrophic lake was chosen for this
experiment. The experimental design was similar to those in the first two
experiments. Eight open bottom enclosures were suspended from two connected
outriggers. The design was set up for four different treatments; unmixed
(control), slow mix (20 mls/min), medium mix (50 mls/min) and rapid mix
(100 ml/min). Each treatment was duplicated and all treatments were randomly
assigned to enclosures. Treatment assignments are indicated in Table 9.
Air was released from a depth of 7 meters in all enclosures. The experiment
began July 30> 1976 and was terminated on September 20, 1976 after vandals
destroyed the bags. Small pianktivorous fish were occasionally trapped in
enclosures as they were lowered during the first two experiments. These fish
may have contributed to the variance in zooplankton and phytoplankton between
bags. This problem was avoided in experiment 3 and all subsequent
experiments by passing a screen through the enclosures several times after
they were in place. This was done slowly to avoid disrupting stratification.
Integrated water samples were taken with a tube sampler to a depth of 5
meters. This was the average thermocline depth in unmixed enclosures.
Routine measurements and analyses were identical to those in the first
experiment. In addition, species specific rates of photosynthesis were
determined in several of the enclosures using the technique of
autoradiography. (Refer to the Methods section for a discussion of this
technique.). No zooplankton data were collected since it was not possible to
obtain samples in the enclosures after they were vandalized.
8]
-------
Changes in pH, C02, TP, SD, Chi a, and algal volume for each enclosure
are shown in Figure 25- The pH dropped and C02 increased in all of the mixed
bags. The pH drop was approximately the same at all mixing rates falling
from an initial value of 8.9 to approximately 7-9« The pH did not remain low
in the mixed enclosures but returned to the high control levels shortly after
the initial drop. The pH dropped again in all of the mixed enclosures
towards the end of the experiment. TP and chlorophyll concentrations
increased in all of the circulated enclosures. Again the average mixed layer
concentrations of TP and chlorophyll were positively correlated. This
relationship is shown in Figure 26. The regression is significant at the .05
level. The points corresponding to the fast mixed bags (2 and 4) seem to fall
slightly below the regression line indicating a relatively low yield of
chlorophyll per unit of TP and, perhaps, an inhibitory effect at these high
mixing rates. There were very few changes in species composition associated
with the different treatments. The phytoplankton community was dominated by
Osci1latoria rubescens which usually accounted for more than 95% of the total
algal biomass in all of the enclosures. In the stratified lake and control
bags it occurred as a dense metalimnetic population while in the mixed bags
it had a uniform depth distribution. The metalimnetic population in the lake
and control bags were slowly eroded as the mixed layer increased naturally in
the fall resulting in higher densities near the surface. The only
significant change in species composition was found in bag 2 where a dramatic
decline in Osc i1 Iator i a rubescens and a shift to green algae and diatoms
occurred toward the end of the experiment. It should be noted, however, that
periphyton growth on the side walls of this and other mixed enclosures became
very dense toward the end of the experiment and most of the "planktonic"
green algae and diatoms were periphytic forms. A decline in Osc i11ator i a was
also observed in bags k, 6, and 7 toward the end of the experiment but this
species still remained dominant in these enclosures.
I n s i tu C-14 incubations for the autoradiographic analysis were carried
out in the lake and bags I, 2, k, 5» and 7 on August lBth. A series of 300-ml
bottles with standard C-14 activity were incubated at 0, 2, U, 5. and 6
meters for 2 hours. A single depth integrated sample was obtained from each
set of bottles after the incubation and preserved for the laboratory
analysis. Osc i11ator i a rubescens was the only species for which satisfactory
track counts could be obtained. The other species were so rare that 95^
confidence intervals were always greater than the mean track counts. Relative
photosynthetic activity was expressed as DPM per 100 um of filament length.
This should be interpreted as the average photosynthetic activity of
Osc i11ator i a filaments in the water column. The results from this analysis
are shown in Figure 27* The average activity of Osc i11atori a rubescens in
the lake and control enclosures was not significantly different. This result
provides additional support for our assumption that the limnological
environment within the enclosures is representative of natural lake
conditions. There was also no significant difference in activity between the
two rapidly mixed bags (2 and 4). The highest activity was found in the
rapidly mixed enclosures while the lowest occurred in the control enclosures.
The slow mixed enclosures (bag 7) had an intermediate level of activity. The
higher activity in the mixed enclosures is difficult to explain. There was
very little difference in pH between these enclosures at the time of the
incubation. Greater nutrient availability might have led to higher specific
photosynthetic activity but the ratio of TP/chlorophyll was actually lowest
in the mixed enclosures. The most likely explanation for this difference in
photosynthetic activity involves the stratification of Osc i11ator i a
82
-------
SO
M-
JC-
€ «¦
i
10-
0-
7 iso
€ ioo
5 w
p .
* !¦
T Si'
£
p. 1SC-
I«
Iftfc*
»
bag i
tenttoi
bag 2
YOO
A
bag
20
3
/
k

lig
100
4
•




A—



r
r -
¦	
,	



6 106'
o «
"i i
^ -
^

r^
/V


l,l—"" —



d»y» o 2$ 5a
• tr on ^ 4
25 SO fl
i
1 25 50 C
> 4
25 SO
y i
25 SO
i
. 6»fl S
control
/ »

big e
so
-•
bag 7
20
h

bag 1 /
50
*^jy
f »




/V*

r


^	-


r^



A	^
-w



0 25 50 C
t
L 25 50 I
y 4
I 25 50 a
i
L 25 SO
k
Fig. 25. Changes in pH, CO2, TP, Secchi disc
transparency, chlorophyll a^ and algal volume in the
enclosures and the lake during experiment 3 (g =
greens, b = blue-greens, d = diatoms, o = other).
83
-------
Exp 3
Chi 50
mg m
Chl = 0.784 TP - 8.65
100
200
TP mg m~3
Fig. 26. Relationship between average concentrations of
TP and chlorophyll £ in the enclosures and the
lake during experiment 3 (numbers refer to
different enclosures, L = lake).
84
-------
o
*»
K
c 8^
«
E
•
= 6h
o
V"
3
e
S 2-
2
Cl
° 0.

lak(
rh
s
t
tag
Hh
1
1
tag
rh
2
t
tag
r+-
4

lag
rh
5
I
tag
rf
7
treat-
ment
lake
cont rol

lOOmla-mln"1
control
20mli-min"1
pH
8.63
6,92
8 72
8.08
8.93
8.87
TP/chl
2.20
2.16
1.51
1.51
1.82
1. 40
Fig. 27. Average photosynthetic activity of Oscillatoria
rubescens determined autoradiographically in several
enclosures and in the lake on August 18, 1976,
experiment 3. Associated values of pH and phosphorus
availability, TP/chl, are also indicated. (Vertical
lines on bars represent 95% confidence intervals.)
85
-------
populations in the control enclosures. Since the populations in the lake and
control enclosures occurred at a depth near the bottom of the euphotic zone
their rate of photosynthesis was probably light limited. The principal
effect of mixing, then, was to bring these populations in contact with the
higher light intensities near the surface. Therefore, the increased activity
in the mixed enclosures was probably a result of greater light availability.
This may also explain the lower nutrient availability TP/Chl in the mixed
enclosures. Without the constraints of light limitation the algae were free
to grow until nutrients became limiting. This may, in part, explain why
higher algal standing crops were observed in the mixed enclosures.
Several of the results of this experiment were inconsistent with both
our theoretical framework and, also, with the results of the previous two
experiments on Little Lake Johanna. Mixing did result in a decrease in pH
and an increase in TP levels as expected but a clear shift from blue-green to
green species only occurred in one enclosure and that result is suspect
because of the potential influence of periphyton. This shift did occur in
the first two experiments and the important difference may be that the actual
drop in pH associated with mixing was much greater in Little Lake Johanna
than it was in Twin Lake. This difference was partly a result of the higher
alkalinity and, hence, greater buffering capacity of Twin Lake water. The
alkalinity in mixed Little Lake Johanna enclosures was in the range of 1.5 ~
2.5 meq/1, which compares with 2.1 - 3-3 meq/1 for Twin Lake. A second
reason for the larger pH drop in Little Lake Johanna enclosures was that the
level of C02 was much higher in these bags. The C02 concentrations in Twin
Lake mixed bags seldom reached 100 ug/1 while levels in mixed Little Lake
Johanna bags often exceeded 500 ug/1. Both the initial alkalinity and C02
levels in the hypolimnion, then, determine the pH drop following
destratification. The high alkalinity and low C02 levels in Twin Lake may
also explain the rapid return to high pH levels after the initial drop in the
mixed enclosures. According to our theoretical framework, light limitation
should increase as the mixed depth increases. The exact opposite effect was
observed for Osc i11ator i a in the mixed enclosures in experiment 3- This
result occurred because a fundamental assumption of our theoretical growth
model (equation 10) was violated. The model assumes a uniform distribution
of algae in the mixed layer but the Osc i11ator i a population in Twin Lake was
concentrated in the metalimnion. The effect of mixing then was to reduce
light limitation instead of to increase it.
Some results at the community level in experiment 3 agree with those
found in the previous two experiments. An increase in TP in mixed enclosures
and a positive correlation between TP and Chi a were observed in all three
experiments. However, the slope of the TP vs Chi a regression for
experiment 3 was much steeper than those observed for the first two
experiments. This may have resulted from the combined effect of nutrient
enrichment and a release from light limitation on algal growth in the mixed
bags in experiment 3-
Experiment k
The results of the first three experiments suggested that changes in
C02, TP, and other nutrients during mixing might be influencing the
phytoplankton response. However these results were somewhat ambiguous since
it was not possible to separate the effects of these factors from other
hypolimnetic factors which covaried in the open bottom enclosures.
Experiment k was the first of three experiments which utilized closed
bottom enclosures and controlled chemical additions in an attempt to separate
86
-------
the effects of phosphorus, nitrogen and C02 from other hypolimnetic factors
during circulation. The experiment was carried out from September 1 to
September }0, 1976 on Little Lake Johanna. Eight closed bottom enclosures
were used. Various combinations of phosphorus (as K2HP01*) , nitrogen NH^Cl)
and carbon dioxide (gas) were added to the artificial hypolimnia of these
bags in amounts designed to simulate natural hypolimnetic concentrations. C02
was added by bubbling samples of hypolimnetic water until they were presumed
saturated and then returning them to the hypolimnion. The additions were
varied between enclosures to produce a range of pH and nutrient levels so
that the effects of these two factors could be separated. NaHC03 was also
added to most of the enclosures to simulate the expected alkalinity in the
mixed lake. All additions were made before the air was turned on. Open
bottom bags were also included in the experimental design so the closed
bottom bags could be compared to a less controlled natural system. One of
these bags was not mixed and served as the control for both open and closed
bottom bags. Air was released at 7 meters at a rate of 100 mls/min (measured
at surface) in all mixed bags. A summary of treatments is given in Table
10. Integrated water samples were taken to a depth of 3 meters which was the
approximate depth of the thermocline in the lake. Routine measurements and
analyses were the same as those described for experiment 1 except for the
determination of algal biomass. Algae were not counted and, therefore, algal
volume was not determined in this experiment. Chlorophyll a and chlorophyll
b provided the only quantitative measures of algal biomass. Chlorophyll a
was assumed to represent total algal biomass. Chlorophyll b only occurs in
the Chlorophyta and euglenoids. Since euglenoids were never observed in any
of the enclosures the concentration of chlorophyll b was assumed to be a
measure of the biomass of green algae. As in previous experiments, changes in
species composition were also determined from photomicrographs of the algal
samples. Zooplankton tows were taken initially in the lake and in all of the
closed bottom bags. Final tows were taken in all bags and in the lake. All
bags were also screened initially to remove fish. Changes in pH, C02, TP, SD
and algal biomass in the enclosures are indicated in Figure 28. A rapid drop
in pH was observed in all of the closed bags which had C02 additions (bags 3.
6, 8, 9) with the greatest drop occurring in the bag with the largest C02
addition (bag 9)• The pH in bag 9 fell from an initial value of 8.6 to J.0
one day after the air was turned on. The pH in the mixed open bottom bag
(bag 10) also fell rapidly reaching a low of 7-2. Unfortunately, the pH was
also low in the other bags which did not receive C02 as well as in the lake
for most of the experiment. TP increased in all of the closed bottom bags
which received nutrient additions (3. 7» 8, 9) but the maximum level
reached (approximately 380 ug/1) was considerably less than the peak observed
in the mixed open bottom bag (bag 10, Gk~! ug/1). The level of TP remained
high in the closed bottom-nutrient addition bags but dropped rapidly in the
open bottom-mixed bag. TP levels in the bags which did not receive nutrient
additions were similar to the control and lake levels. Chlorophyll a peaked
in all enclosures and in the lake on day 6. This peak was followed by a
rapid decline which was also independent of treatment conditions. The
synchrony of the peak and crash in all of the bags and in the lake indicates
that some common factor was responsible for this change. Since the
experiment was begun late in the growing season the decline in chlorophyll a
may have been the result of cooler temperatures or shorter day length. This
may also explain why TP and Chi a concentrations were not as well correlated
in this experiment as they were in the previous three. The relationship
between average chlorophyll a and TP concentrations in all enclosures is
shown in Figure 29- The regression is not significant at the .05 level.
87
-------
Table 10: Treatments for enclosures in Experiment 4, Little Lake Johanna
Air flow Air
(ml min~l release Bag
Bag Air on Air off 0 surface) depth (m) Type Additions* Text designations
1	0 -- open	--	open control
2	9/4/76 9/30/76	100 7 closed A	closed high pH
3	"	"	100 7 closed A+N+P+lowC closed nutrients
low pH
4	"	"	100 7 closed A+N+P closed nutrients
high pH
5	"	"	100 7 closed none closed mixed
control
6	"	100 7 closed A+N+P+lowC closed nutrients
low pH
7	"	"	100 7 closed A+N+P closed nutrients
high pH
8	"	"	100 7 closed A+N+P+lowC closed nutrients
low pH
9	"	"	100 7 closed A+N+P+highC closed nutrients
very low pH
10 "	"	100 7 closed -- open mixed
*A(A1 kalinity,NaHC03), N(Nitrogen, NH^Cl), P(Phosphorus, K^HPO^), CfC^)
Table 11: Treatments for enclosures in Experiment 5, Little Lake Johanna
Bag
Air on
Air off
Air flow
(ml min"'
@ 7m)
Air
release
depth (m)
Bag
Type
Additions*
Text designation
1
7/8/77
8/15/77
50
7
open
—
open mixed
2
__
--
0
--
open
--
open control
3
II
II
50
7
closed
A+C
closed low pH
4
II
<1
50
7
closed
A+N+P
closed nutrients
high pH
5
--

0
--
closed
none
closed control
6
It
II
50
7
closed
A+N+P+C
closed nutrients
low pH
7
11
H
50
7
closed
A
closed high pH
^(Alkalinity, NaHC03), N(Nitrogen, NH4C1), P(Phosphorus, K2HP04), C(C02)
»»
-------

150

bag 1
o, control
bag 2
C, A
bag 3
C.N+PfA+C
bag 4
c.N+P+A
k
bag 5
c, no add
bag 6
c.A+C
f>
1
E
100



l\
1
k
K

£
u
9
E
50-




k
k
k
/V











1-
E
ci2'
« 3






\
\
A
400








E
a.
K
Dl
E
300
200-
100-
\	.



p
r

	
t
E
O
o
*
100-
10'



r-


r~
E
X »¦
s
7-




u



day* 0 15 30 0 15 30 0 15 30 0 15 30 0 15 30 0. 15 30 0 15 30
• if on f	f	#	#	#	f	t
7
days 0
air on
bag 7
C, N + P + A
bag 1
c.N+P+A + C
bag 9
e.N+P+A+fi
K
bag 10
o m i*td

x--




r^~
* \
6
-------
There was only a slight increase in the average chlorophyll concentrations in
the closed bottom bags which received nutrient additions. The yield in the
mixed open bottom enclosure (10) was particularly low. This suggests that
some factor other than mixing, TP, pH, alkalinity or C02 may have inhibited
algal growth in this bag. There were no major changes in species composition
during this experiment. The flora in all enclosures and in the lake was
dominated by green algae throughout the experiment and the green algae were
not replaced by other species after their populations crashed. There was
also no consistent effect of circulation on the abundance of herbivorous
zooplankton. The changes which occurred in each enclosure are shown in
Figure 22. When these zooplankton numbers were converted to community grazing
rates (see Methods section), however, it was evident that the herbivores were
influencing the final yield (Chi a) in the different enclosures. Figure 30
shows the relationship between the ratio TP/Chl a (inverse of yield) and the
community grazing rate calculated on the day of the final zooplankton tow for
all of the enclosures. The numbers on the graph again represent the
different enclosures. The regression is significant at the .05 level. It is
apparent from this Figures that those enclosures with the highest grazing
rates also had the lowest yield.
This experiment primarily served as a test of the new controlled
experimental design. We were able to separate the effects of pH, C02, TP,
nitrogen, alkalinity, physical mixing, and light limitation from the effects
of a large number of unknown factors associated with the hypolimnetic milieu.
However, we were not entirely successful in simulating natural levels of
controlled substances. We were able to reproduce the drop in pH and increase
in C02 observed in the open bottom mixed bag but our nutrient additions were
apparently insufficient to simulate the natural increase in TP observed
during mixing. The response of the biota in this experiment was somewhat
anomalous. However, we did not expect the results to be very meaningful
since the experiment was begun so late in the growing season. Perhaps, the
most significant result was the extremely low yield observed in the open
bottom mixed enclosure. This result provides evidence for the existence of
an unknown hypolimnetic factor, not adequately addressed in our theoretical
framework, which has an inhibitory effect on algal growth.
Experiment 5
Experiment 5 was carried out on Little Lake Johanna between July 8 and
August 15. 1977. As in the previous experiment closed bottom enclosures and
controlled chemical additions were used to separate the effects of N, P and
C02 from other hypolimnetic factors during mixing. Open bottom enclosures
were also used for comparison. Hypolimnetic additions were then varied
between these enclosures in a factorial design to separate the effects of
nutrients (N and P) and pH (C02) during mixing. Two levels of nutrients
(high N+P, low N+P) and two levels of pH (high pH, low pH) were used as
indicated below.

low pH
hiqh pH
low N + P
Bag 3
Bag 7
high N+P
Bag 6
Bag 4
The calculations and procedures involved in the addition of hypolimnetic
90
-------
40-
Chl 20-
mg m 3

4
Exp 4

2
9 3 _

				 u
" ¦""" L

6


10

r2 - 0.10

Chi = 0.018 TP + 24.96
TP mg m
-3
200
400
Fig. 29. Relationship between average concen-
trations of TP and chlorophyll a_ in the enclo-
sures and lake during experiment 4 (numbers
refer to different enclosures, L = lake).
o>20 -
F ( m3*m 2
Fig. 30. Relationship between the total
herbivore filtration rate (F) and the
ratio TP/chlorophyll a in the enclosures
and lake on the final day of experiment 4.
91
-------
nutrients to the closed bottom bags were modified slightly in experiment 5-
Integrated samples were taken to a depth of seven meters (the final mixed
depth) in both open and closed bottom enclosures and analyzed to determine
what the concentrations of N, P, C02 and alkalinity would be after mixing.
Additions of N, P, C02 and alkalinity to the closed bottom bags were then
calculated so that the final mixed-concentration would be the same as the
average mixed-concentration found in the open bottom enclosures. Nutrients
were added to the hypolimnia in the same manner as in experiment k except for
C02 which was added in the following way. Hypolimnetic water was first
collected from a closed bottom bag and placed in a container of known volume.
The concentration of C02 was determined initially and then again after the
water was bubbled with C02 gas to determine the exact quantity of C02 added
before the water was returned to the hypolimnion. In this way we could
accurately control the amount of C02 added to each hypolimnion.
Air was released at a depth of seven meters in all of the mixed
enclosures. Variability in air flow caused some problems in previous
experiments. This variability was primarily due to the poor quality of the
needle valves used. These old valves were replaced in this experiment with
more accurate microadjusting valves. The new valves were also used in all
subsequent experiments. Part of the variability in air flow rates was also a
result of measuring air flow at the surface. It was found that air flow
measured at the surface did not accurately represent the flow at the depth of
air release which was much lower due to hydrostatic pressure. The air flow
also varied due to kinks in the air lines and small differences in the
release depth. To resolve these problems an in-line flow meter was
constructed. The meter, which was used in this and all subsequent
experiments, was designed to measure the air flow at the point of release.
The air flow rate of 100 ml/min at the surface, used in previous experiments,
was found to be equivalent to a flow rate of 50 ml/min at seven meters. This
air flow rate was used in all mixed enclosures. All treatments were randomly
assigned to enclosures. A summary of these treatments is given in Table 11.
Routine measurements and analyses were the same as those described in
Experiment 1. Integrated water samples were taken to a depth of 5*5 meters.
Zooplankton tows were taken initially in the lake and in all of the closed
bottom bags. Final tows were taken in all bags and in the lake. All bags
were also screened initially to remove fish.
Changes in pH, C02, TP, SD, Chla, and algal volume in each enclosure are
shown in Figure 31* The pH dropped to a low of 7-1 within six days in both
closed bottom bags which received C02 additions {bags 3 and 6). This was
slightly less than the low of 7*3 observed in the open bottom mixed enclosure
(bag 1). The pH returned gradually in bag 6 and quickly in bags 1 and 3 to a
level near that of the other bags. A similar drop in pH was not observed in
the lake, the open control bag (bag 2) or in the other closed bottom bags
However, the pH levels in all bags except the closed control (bag 5) were
initially quite low and the "high pH" treatment was only slightly different
from the "low pH treatment". The initial pH in the closed control (bag 5)>
was quite high (9). TP increased to 1 it 15 ug/1 and 1205 ug/1 in the closed
bottom enclosures which received nutrient additions {k and 6 respectively)
exceeding the maximum value of 1060 ug/1 observed in the open mixed bag (1).
The TP level remained high in bags k and 6 but declined rapidly in bag 1.
This rapid decline of TP in the open bottom bag may represent a loss of
particulate phosphorus by sedimentation. Particulate phosphorus could not
drop out of the closed bottom bags and was probably resuspended in the water
column. The levels of TP in the closed bottom enclosures which did not
receive nutrients (bags 3> 5. 7) were similar to those in the lake and open
92
-------
bag 7
bag 3
c, A + C
bag 5
c, control
bag 2
o, control
bag 1
200-
100
50'
a. 1000-
9-
20
20
20
20
20
20
20
0
20
da ys 0
air on ^
0.
0
0
Fig. 31. Changes in pH, CO2, TP, Secchi disc transparency, chlorophyll
and algal volume in the enclosures and lake during experi-
ment 5. (algal volume: g = greens, b = blue-greens, d =
diatoms, 0 = other; treatment designations: 0 = open bottom,
c = closed bottom, N = nitrogen, P = phosphorus, A = alkalin-
ity, C = CO2-)
93
-------
control (bag 2).
The response of the phytoplankton in this experiment was not entirely
consistent with our theoretical framework. Osc i11ator i a spp. dominated in
all of the enclosures initially. A definite shift to green algae occurred in
four out of the five closed bottom enclosures (i», 5. 7)- The pH was not
consistently low in these enclosures but it did drop below pH 8 in all of
them at least once during the experiment. Since it is not clear whether
continued low pH or just a sudden drop is required to produce the blue-green
to green shift these results are somewhat ambiguous. High levels of
available nutrients, TP/Chl, which are also apparently required for the
blue-green to green shift (Shapiro 1973) » were observed in all of these
enclosures just before the greens bloomed. It is important to note that the
blue-green to green shift did not occur in the open bottom enclosures (1 and
2) even though the levels of pH and TP were similar to those in the closed
bottom bags. This result seems to indicate that some unknown factor
necessary to maintain blue-green dominance was excluded from the closed
bottom bags. This factor might have a stimulatory effect on blue-greens, an
inhibitory effect on greens or both. The community level response was
consistent with previous results. Chla levels were again correlated with TP
concentrations. The relationship between average TP and Chla concentrations
is shown in Figure 32. This regression is significant at the .05 level. The
highest chlorophyll levels were found in the open mixed bag (1) and the two
closed bottom enclosures which received nutrient additions (k and 6). As in
previous experiments the stimulatory effect of nutrient enrichment was
apparently more important than the negative influence of light limitation
when the mixed depth was increased.
The results of the factorial experiment were inconclusive. An analysis
of variance was performed to determine the effects of pH and TP on three
different dependent variables: The average Chla concentration, the volume of
blue-green algae on day 19 and the volume of green algae on day 19- There was
no significant effect of pH on any of these variables (p = .05)• This is not
surprising considering the small difference between high and low pH
treatments. TP also had no significant effect on these variables. An effect
of TP on Chla was expected since a significant linear relationship was
already demonstrated between average TP and Chla (Fig. 32). The failure to
find a significant effect was primarily due to the small number of degrees of
freedom available for the analysis of variance.
Changes in the abundance of herbivorous 2ooplankton observed in the
enclosures are shown in Figure 22. The sample from the initial lake tow was
lost. Therefore it is not possible to determine whether a change in
herbivore numbers occurred in the lake or open bottom enclosures. However,
there was a consistent increase in herbivore numbers in all of the closed
bottom bags. The smallest increase occurred in the control bag which had the
highest pH levels, as predicted in the theoretical framework.
Experiment 6
Experiment 6 was set up on Twin Lake and lasted from August 5 to
September 9> 1977• This was the largest and most ambitious experiment
undertaken using 16 enclosures suspended from four interconnected outriggers.
The previous experiments indicated that changes in pH and nutrient levels
during artificial circulation might be important factors in the phytoplankton
response. It was apparent after the first three experiments that the effects
of these factors could not be entirely separated in a natural system where
many other variables also influence the phytoplankton. As a result, the
94
-------
25-
r* = 0.55
Chi = 0.028 TP + 7.79
500
TP mg m ®
1000
Fig. 32. Relationship between average concentra-
tions of TP and chlorophyll a in the enclosures and
lake during experiment 5 (numbers refer to different
enclosures, L = lake).
95
-------
relationships observed in these experiments were generally ambiguous.
Subsequent experiments were designed to separate the different factors as
much as possible using closed bottom enclosures and controlled chemical
additions. However, the closed bottom bags were only successful in
separating different hypolimnetic factors and there were still some
uncontrolled factors which made the results difficult to interpret (e.g.
zooplankton, light limitation), in experiment six we decided to take a
somewhat different approach. Instead of trying to study the effect of each
factor separately an attempt was made to investigate the interaction between
factors and their combined effect on the phytoplankton. In particular, the
experiment was designed to investigate two different interaction mechanisms
which might be important during artificial circulation: 1) the interaction
between grazing and pH in the blue-green to green shift mechanisms, and 2}
the interaction between all of the factors in Equation 11 in the community
response mechanism.
The interaction between grazing and pH in the blue-green to green shift
mechan i sm
The results of previous experiments suggested that even when conditions
favor a blue-green-green shift (i.e. low pH, high nutrient availability)
herbivores may prevent it from occurring by grazing selectively on the green
algae (cf. expt. 2, bags 8 and 9)- A factorial experiment was designed to
test this hypothesis using eight closed bottom enclosures. Two levels of pH
and two levels of grazing were used. The pH was varied by hypolimnetic C02
additions. The abundance of herbivores was reduced in the "low grazing"
treatments by adding four pianktivorous fish (Lepomis qibbosus) to each of
these enclosures. Nitrogen, phosphorus and alkalinity were added to all of
these to simulate natural hypolimnetic levels and all of these were mixed
to seven meters with an air flow rate of 50 ml/minute. Each treatment was
duplicated so that interaction effects could be determined and to raise the
number of degrees of freedom for the analysis of variance. Mixed, open
bottom bags with and without fish were also included for comparison. The
general experimental design and enclosure numbers associated with each
treatment are indicated below.
Closed Bottom Enclosures
H i qh pH	Low pH
High grazing	9. 2	5«
Low gra2 i ng	3.13	11,8
Open Bottom Enclosures
H i gh graz i ng	1,7
Low Grazi ng	6, 16
The effect of these treatments on the blue-green-green shift was measured in
several ways. The growth rate was determined for individual green and
blue-green species. Composite growth rates were also calculated at the
division level and the difference between the growth rates of blue-greens and
greens was used to indicate competitive advantage. In addition, rates of
photosynthesis were determined for a number of blue-green and green species
autoradiographical1y in bags 3. 5» 9 and 11.
96
-------
The community response mechanism
The second part of the experiment was an investigation of the
interaction among all factors in equation H during artificial circulation
and their combined effect on algal community standing crop. This general
mechanism was investigated by I) evaluating all of the parameters of equation
11 with data from several different enclosures representing a range of mixing
conditions, 2) testing the reliability of the model on other enclosures and
3) using the model to simulate the effects of a wide range of mixing
conditions on algal standing crop. All of the closed bottom bags in the
factorial design (described above) were monitored along with two unmixed
control bags (10,15) in order to evaluate the model parameters (Ec and Ew).
The evaluation of the remaining parameters was based on measurements
restricted to bags 3. 5. 8, 9. 10, and 11. Bag 8 was included after bag 11
developed a leak. All treatments were randomly assigned to enclosures. A
complete summary of treatments is given in Table 12. Most of the sampling
procedures and analyses used in experiment 6 were the same as those described
in experiment 5* However, several modifications and additional procedures
were required. Routine integrated samples were taken to a depth of 5*5
meters. Initial and final zooplankton tows were taken in a 11 of the
enclosures and in the lake. Photosynthetic rates were measured at both the
community and species levels approximately every four days in bags 3» 5» 8,
9, 10 and 11. Seven-meter integrated samples taken for the productivity
measurements were also analyzed for pH, Chi a, POC, TP, and alkalinity.
Light measurements were recorded continuously at a site 5 miles southwest of
the lake.
Changes in pH, C02, TP, SD, Chi a and algal volume observed in the top 5
m of each enclosure are indicated in Figure 33-
The results of the factorial analysis of the effects of grazing and pH
were inconclusive. The only effects considered in the analysis were those
observed during the first 6 days since this was the only period during the
experiment where the pH in the "high" and "low" pH treatments differed
significantly (cf. Fig. 33)- Six-day growth rates were calculated for
blue-greens and greens at both the division and species level while species
specific rates of photosynthesis were determined on day 6. These values
represent the average growth and photosynthetic rates of populations in the
mixed layer. The competitive advantage of greens over blue-greens was also
calculated from these six-day growth rates. A separate analysis of variance
was done for each of these dependent variables. The significance of each main
(pH or grazing) and interaction effect was determined with an F-test at the
.05 level. A summary of the results of these analyses is presented in Table
13- None of the F-tests were significant at the .05 level. There are several
reasons why no significant effects of either grazing or pH were found. The
difference between "high" and "low" pH treatments was generally less than 1
pH unit. (cf. Fig. 33). This change was apparently too small to affect the
competitive advantage of green over the blue-green algae. As in the previous
Twin Lake experiment, this small drop in the pH was primarily due to the high
alkalinity and relatively low hypolimnetic C02 concentrations characteristic
of the lake. A similar drop in pH was observed in the open bottom mixed bags.
The lack of a significant grazing effect in the analysis was apparently a
result of the failure of pianktivorous fish in the "low grazing" treatments
to remove herbivorous zooplankton. Their general ineffectiveness was
indicated by the lack of any significant difference between the final
zooplankton counts in enclosures with and without fish (Fig. 22). In
addition, the grazing and pH effects may have been confounded by the high
97
-------
Table 12: Treatments for enclosures in Experiment 6, Twin Lake
Air flow Air
ml min-1 Release Bag
Bag Air on Air off § 7 m depth (m) Type Additions'
1 8/5/77 9/9/77
4
5
6
7
8
10
11
12
13
14
15
16
50
50
50
50
50
50
50
50
50
50
50
50
50
7 open	none
7 closed	N+P+A
7 closed	N+P+A+F
open	none
7 closed	N+P+A+C
7 open F
7 open	none
7 closed	N+P+A+C+F
7 closed	N+P+A
closed	none
7 closed	N+P+A+C+F
7 open	none
7 closed	N+P+A+F
7 closed	N+P+A+C
closed	none
7 open F
Text designation
open low pH
high grazing
closed high pH
high grazing
closed high pH
low grazing
open control
closed low pH
high grazing
open low pH
low grazing
open low pH
high grazing
closed low pH
low grazing
closed high pH
high grazing
closed control
closed low pH
low grazing
open control
closed high pH
low grazing
closed low pH
high grazing
closed control
open low pH
low grazing
*A(A1 kalinity, NaHCOo), N(Nitrogen, NH^Cl), P(Phosphorus, K2HPO4), C(C02),
F(Fi sh).
98
-------
Table 13: Factorial analysis for the effects ofpH
Experiment 6.
high pH
Treatment high qraz.
Dependent var.	Bag	9	2
r
chl a

.2870
.2780
r
phytoplankton

.0569
.1696
r
greens

.0306
.0970
r
blue greens

.0087
.1710
r
diatoms

.2587
.2750
r
d-b

.2500
.1040
r
b-g

.0393
.0740
r
d-g

.2893
.1780
r
Oocysti s

.0400
-.0330
r
Sphaerocysti s

.0511
-.0350
r
bg coccoids

.1474
.0931
r
monad

-.2897
-.0980
r
Cryptomonas

.0028
.2030
r
Mel osi ra

.3563
.3690
r
Fragil aria

.1428
.1810
r
Oscillatoria

.0439
.2380
r
Anabaena

.2130
.1950
r
Synedra

.2035
.2730
r
Aphanizomenon
Anab. (gC cell"'
f1} 1
T f1)
.0286
.1446
P
182.7

P
Aphan. (gC lOOum"
Frag. (gC cell"]
Crypt (gC cell"]
Mel. (gC 1OOum"'
Sphaer (gC cell"'
1404.0

P
d
d" )
d"1}
68.3

P
256.0

P
247.1

P
d )
74.1

* + = yes, - = no
and grazing on algal growth and photosynthesis in
Test of Significance*
high pH

low pH

low pH
at
.05 level

low graz.
hign qraz.
low graz.

Effects

3
13
5
14
11 8
pH
qraz.
inter
.2423
.2450
.1989
. 2384
m .2615

-

.2560
.1117
.1800
.0984
m .2062
-
-
-
.1197
.1715
.0376
-.0396
m .0936
-
-
-
.2084
.1487
. 181 3
.0999
m .1579
-
-
-
.3989
.4421
.5101
.4464
m .2805
-
-
+
.1905
.2934
.3288
.3465
m .1226
-
-
-
.0887 -
.0228
.1437
.1395
m .0643
-
-
-
.2792
.2706
.4725
.4860
m .1869
-
+
+
.3119
.0900
.1846
.1500
m .2871
-
-
-
.0823
.1775
.1563
-.1340
m -.0280
-
-
-
.2593
.1305
.3240
-.1819
m .0736
-
-
-
.2735 -
.2094
-.1017
m
m .1881
-
-
-
.2640
.2506
.1987
.0004
m .1795
-
-
-
.5530
.3090
.4934
.1969
m .2912
-
-
-
.2200
.3925
.5534
.3041
m .2764
-
-
-
.0692
.1404
.2330
.0625
m .2366
-
-
-
.2630
.0852
.1761
.1102
m .0340
-
-
-
.0578 -
.0492
.2936
-.3900
m .0578
-
-
-
.1923
.1607
.1315
.0639
m .2202
-
-
-
30.8

41 .8

36.0
-
-

213.2

245.3
391 .8
-
-

29.5

47.8

85.2
-
-

267.7

202.6
246.6
-
-

112.9

130.1
294.0
-
-

28.0

23.3

34.3
-
-

-------
30
1
E
i 10
0
« 200-
E
0	100-
01
E
0
*'¦
«j
'E 300-
ft. 200*
£ ico
? 0-
lake
««y 1 «•»»
b b
SJK
bag 1
iar 1
b
Lsw

bag 2
1 1
itlV, Ffere.
bag 3
«•) 1 4»y 6
fail ftfB
bi
ig 4
i
l£JL
«»»«
¦ do



XV .





———
-ws__


rv_

E
# 100
O to-
0*
e i-
	
A	



x »•
° 7-

0
C, N+P + A
c.N+p+A+F
'o^onUo^*^"*
daysO 20 40 0 20 40
air on C C \
k ,D 40
^ 20 40 1
^ 20 40
c. N+P+A+C
Fig. 33. Changes in pH, CO2, TP, Secchi disc trans-
parency, chlorophyll a^ and algal volume in the enclo-
sures and lake during experiment 6. (algal volume: g =
greens, b = blue-greens, d = diatoms, 0 = other; treat-
ment designations: 0 = open bottom, c = closed bottom,
N = nitrogen, P = phosphorus, A = alkalinity, C = CO?,
F = fish)'
100
-------
0
200
._E3U
bag 10
dat ^ 4*y t
_OS4
bag 11
¦•y 1	day •
b
jgfe
bag 12
«*i 1
bJL
bag 13
«•* 1 «»T •
fan
Jkn
bag 14
HtrW
C 1-
o

a 00
E
« 0
E 100
o 1°-
u
a 1-
&
x ¦'
*7-
6

C, control
C.N + P+A + C4F
c. N+P+A+F
C.N4P4A+C
days 6 20 40
air on ^
t
20 40 ft 20 40 0 23 40 0
$ % ~

20-
n

E

_
10-
E


0-
n
200-
E

2:
100 -
O

a

E


0-

i 1.

0

« i-
r>
1
f
300'
fib
200


01
100
F

r>
0
E
100
n
O
10
O
1
a
bag 15
d» yl	B»y6
IK&.	fhin
c, control
d.y. i
uron C
bag 16
*
Fig. 33 Cont'd. Changes in pH, CO2, TP, Secchi disc
transparency, chlorophyll a and algal volume in the enclo-
sures and lake during experiment 6. (algal volume: g =
greens, b = blue-greens, d = diatoms, 0 = other; treat-
ment designations: 0 = open bottom, c = closed bottom,
N = nitrogen, P = phosphorus, A = alkalinity, C = CO2,
F = fish)
101
-------
variability in nutrient levels observed in the closed bottom bags (Fig. 33)-
The analysis of community response mechanisms began with an evaluation
of several factors in equation 11 which were relatively constant. The
extinction coefficient of chlorophyll, Ec, and the residual extinction
coefficient of the water, Ew, were determined from a regression of the total
extinction coefficient, E , on the concentration of chlorophyll a. The total
extinction coefficient was determined from Secchi disc readings using the
Lambert-Bouguer Law and assuming that the light intensity at the Secchi
depth, Isd, was 15% of the surface intensity, lo (Megard, personal
commun i cat i on) .
tr+ _ 1 n( Io/Isd) _ 1.897	(16)
SD	" ~~W
The regression is shown in Figure 3^- The slope of the regression line,
0.023 ">~2 mg Chl-1, is an estimate of Ec. The y-intercept, O.78 m-1, is an
estimate of Ew. The light limitation factor, In {f0/121), is normally
calculated from the maximum daily volumetric rate of photosynthesis, p-max,
the integral rate of photosynthesis,TT , and the total extinction coefficient,
E , as
1n(Io/Iz ) = p_max	(17)
Unfortunately, In (lo/lz1) could not be calculated from our data because
p-opt was not measured directly. Megard et aj_. (1979) reported an average In
(lo/lz1) value of I.9+O.I8 for populations in Halstead's Bay, Minnesota.
Megard et a_[. (1978) also showed that the value of In	(lo/lz1)
constant (1.9-2.0) when daily incident irradiance	exceeded
day — 1) . Since daily incident irradiance was greater	than 380
for all incubations in experiment 6 we have assumed a	constant
for In (lo/lz1) in the present analysis. The nutrient
specific rate of photosynthesis, P-max, and the
phosphorus, Kq, were determined from a regression of
specific rate of photosynthesis, P-max, on the ratio	c/TP.
measured directly but was calculated as:
was relatively
320 Langleys
Langleys day-1
va1ue of 1.9
saturated maximum daily
subsistence quota of
the maximum daily
P-max was not
P-max =
u Et
1.9c
(18)
Only data from the first 17 days were included in this regression.
Periphyton growth on the sidewalls of the enclosures became a significant
problem past this point and the photosynthesis results were inconsistent. The
regression (shown in Fig. 35) was significant at the .05 level and represents
the linear transform of equation 7'
n sat c , D sat	(la's
P-max = P-max Kq jp + P"max
The y-intercept, 140 mg Cmg Chl-1 d-1, is an estimate of P-max, and the
102
-------
3
E 2
••
•>
iu
,\
• •
1
0
80
Fig. 34. Relationship between the
total extinction coefficient and
chlorophyll a in the closed bottom
enclosures during experiment 6.
0>1OO-
50
max -140
0.50
Fig. 35. Relationship between
the maximum daily specific rate
of photosynthesis and the ratio
chlorophyll a/TP in several of
the closed bottom enclosures
during the first 20 days of
experiment 6.
103
-------
inverse of the x-intercept, l.Bl mg TP mg Chl-1, is an estimate of Kq. The
carbon to chlorophyll ratio in the algae,©, was determined from a regression
of the concentration of particulate organic carbon on the concentration of
chlorophyll measured in the enclosures (Fig. 36). The slope of this
regression line, Al mg C mg Chl-1, is an estimate of&. The finite specific
loss rate, D, was determined at frequent intervals in the enclosures where
photosynthesis was measured. It was calculated as the difference between the
finite specific growth rate and the finite specific rate of carbon fixation
(see the Methods section for a more detailed discussion of this calculation).
The average value of D determined during the experiment was 0.110 + 0.032
day-1. The two remaining parameters in equation 11, TP and Zm, were
considered variables in the analysis. The equi1ibrium model (equation 11)
and a summary of parameter values are given below:
c* =	ln(Io/Iz')	- DeZmEw	(20)
EcDeZm + (ln(Io/Iz1) P-"* Kq)/TP
fflaX
P-max ¦ UO mg C mg Chl-1 d-1
Kq = 1.81 mg TP mg Chl-1
Ec = 0.023 ni-2 mg Chl-1
Ew = O.78 m- I
In (10/121) « 1.9
D - 0.110
The reliability of the model was tested by comparing the equilibrium
concentration of chlorophyll predicted by the model to the actual chlorophyll
concentration observed at the point of maximum yield (C/TP) in each closed
bottom enclosure. The TP concentration at the point of maximum yield and
mixed depth in each enclosure were used in the model prediction. All closed
bottom enclosures except bag 11 (which leaked) were considered in the
analysis, including those which were not involved in the evaluation of P-max,
Kq, and$.
The relationship between predicted and observed equilibrium chlorophyll
concentrations is shown in Figure 37- If the agreement between observed and
predicted results was perfect all points would fall on a line with slope
equal to 1 and which intersects the origin (shown by the dashed line in Fig.
37)- The slope of the actual regression line (not shown) was not
significantly different from 1 and the y-intercept was not significantly
different from 0. There was also a linear relationship between equilibrium
chlorophyll and TP concentrations (Fig. 38). Both predicted and observed
equilibrium chlorophyll concentrations showed linear relationships to TP and
the slopes and intercepts were not significantly different at the .05 level.
As in previous experiments, a similar relationship was also found between
average concentrations of TP and Chi a in all of the enclosures in experiment
5 (Fig. 39)« These results indicate that the mathematical model presented
here provides a consistent theoretical framework for addressing community
response patterns.
After these preliminary tests the model was used to simulate the
relationship between maximum chlorophyll concentrations, c* and the mixed
depth at several levels of TP. The results are shown in Figure ^0a. The TP
values chosen were those observed in several of the enclosures and the black
circles represent the observed results. The model predicts that, at a single
level of TP, the maximum chlorophyll concentration will decrease as the mixed
104
-------
6
4
2
0
Fig. 36. Relationship between the con-
centrations of chlorophyll a^ and particu-
late organic carbon in several closed
bottom enclosures during experiment 6.
Cobs ugChi 1-1
Fig. 37. The agreement between the max-
imum observed concentration of chlorophyll
a_ and that predicted by the model in the
closed bottom enclosures and in the lake
during experiment 6 (the dotted line
represents perfect agreement).
105
-------
o predicted
_ 80- «observed
40-
200
100
ugTP-1"1
Fig. 38. The relationship between TP and the
maximum concentration of chlorophyll a_ in the
closed bottom bags and in the lake during
experiment 6. Both predicted (open circles)
and observed (closed circles) results are
shown along with their associated regression
Iines.
EXP 6
100"
Chl 50-
mg m
r* = 0.7S
Chi - 0.39> TP - 6.39
TP mg m~3
200
100
Fig. 39. The relationship between average concen-
trations of TP and chlorophyll a^ in the enclosures
and lake during experiment 6 (numbers refer to
different enclosures, L = lake).
106
-------
TP
211
100-
145
101
50-
600-
TP
211
-------
depth is increased due primarily to increased light limitation. However, if
the concentration of TP rises during mixing the maximum chlorophyll
concentration may increase, as it did in the experimental enclosures in spite
of the light limitation effect. If the TP increases from 18 to 100 ug PI — 1
during mixing, the mixed depth would have to become greater than 30 meters
for a reduction in c* to occur.
The effect of changes in Zm and TP on maximum total algal standing crop
(C*Zm) was also simulated and the results are shown in Figure 40b. Again the
black circles represent values observed in the experimental enclosures. The
model predicts that at a given TP concentration, the maximum total standing
crop will increase with the mixed depth to a depth of about 15 meters and
then decline steadily to a depth of 43 meters below which the community is
eliminated. As with c*, an increase in TP will always result in a higher
value of c*Zm than would otherwise be expected. This is especially true at
relatively shallow mixed depths such as those used in experiment 11. C*Zm
increased dramatically after the mixed depth was increased from 5 to 7 meters
in the experimental enclosures due primarily to an increase in TP.
Although D was assumed to be constant in this analysis there was some
indication that differences in herbivorous grazing between bags had a
significant effect on the yield of chlorophyll. As in experiment k, a linear
relationship was found between TP/Chl a (inverse of yield) and community
grazing rates determined at the end of the experiment (Fig. 41). The lower
yield associated with the highest grazing rates suggests that D was
significantly higher in these bags.
Experiment 7
There was evidence in several of the previous experiments that rapid
mixing can occasionally inhibit algal growth or reduce the yield of
chlorophyll. This effect was most obvious in the open bottom enclosures (for
example bag 7 in experiment 1 and bag 10 in experiment 4). These results led
us to conclude that we may have overlooked some important effects of
artificial circulation by using closed bottom enclosures and restricting our
air flow rates to relatively low levels. The last two experiments were
designed to take another look at the effect of mixing rate on algae. Only
open bottom bags were used and the fastest mixing rates were 5 to 6 times
greater than those commonly used in the previous experiments.
Experiment 7 was set up on Little Lake Johanna and lasted from July 6 to
July 26, 1978. The experiment was primarily designed to determine the mixing
rate required to produce the "inhibition effect" which was often observed in
rapidly mixed open bottom bags. The design therefore consisted of a broad
range of mixing rates (0-300 ml/min. § 7m in open bottom bags). The effect
of "light limitation" was also investigated. An attempt was made to simulate
the degree of light limitation expected in an exceptionally deep lake
following destratification. To accomplish this two of the enclosures (13 and
14) were constructed of a combination of black and clear polyethyelene. The
top meter of both enclosures was clear and the lower 7 meters was black.
This had the effect of reducing the proportion of the population within the
euphotic zone. A complete list of treatments is given in Table 14.
Integrated samples were taken to a depth of 4 meters in this experiment.
All other sampling and analytic procedures were the same as in experiment 1.
Changes in pH, TP, SD, and Chi a in each enclosure are shown in Figure 42.
The pH was relatively low in all enclosures during the experiment. TP was
generally higher in the mixed bags but was not correlated to Chi a as it was
in previous experiments. Figure 43 shows the relationship between average TP
108
-------
2	4
F ( m^*m—^ )
Fig. 41. Relationship between the total herbivore
filtration rate (F) and the ratio of TP/chlorophyll
a in the enclosures and lake on the final day of
experiment 6.
109
-------
Table 14: Treatments for
enclosures
in Experiment
7, Little Lake Johanna
Bag Air on Air off
Air flow
(ml min
0 7m)
Air release
depth (m)
Bag
type
Text designation
1 7/6/78 7/26/78
10
7
open

10
2 " "
20
7
open

20
3
50
7
open

50
4 " "
75
7
open

75
5
100
7
open

100
6
125
7
open

125
y ii it
150
7
open

150
8
175
7
open

175
g H I'
200
7
open

200
10
0
--
open

Control
n
300
7
open

300
12
0
--
open

Control
13
50
7
open
w/black
(7m
50 black
black/lm
14
75
7
open
w/black
(7m
75 black
black/lm
Table 15: Treatments for enclosures	in Experiment 8, Twin Lake
Air flow
(ml min	Air release Bag
Bag Air on Air off 0 7 m)	depth (m)	type Text designation
1	8/11/78 9/13/78 250	7	0	250
2	0	--	0	Control
3	8/11/78 9/13/78 250	7	0	250
4	0	--	0	Control
110
-------
Table 16: Herbivore grazing rates obtained from the literature or calculated
from a regression of literature values.*

mm
ml/animal/dy

Species
L
F
Source
Daphnia rosea
1.45
4.6
Haney 1973, Burns & Rigler 1967
D. galeata
1.60
5.1
Haney 1973, Burns & Rigler 1967
D. parvula
0.95
3.8
Haney 1973
D. longispina

2.3
Nauwerck 1963
D. rerrocurva 0.90
(carapace)
2.8
Regression*
D. pulex
1.75
5.4
Regression*
D. ambigua
0.88
2.7
Regression*
Ceriodaphnia quadrangula
0.80
4.6
Haney 1973
Diaphanosoma brachyurum
1.15
1.6
Haney 1973
Bosmina longirostris
0.50
0.4
Haney 1973
Chydorus sphaericus
0.15
0.2
Haney 1973
Camptocerus rectirostris
1.00
3.1
Regression*
Pleuroxus procurvus
0.50
1.5
Regression*
Scapholebris Kingii
0.90
2.8
Regression*
Simocephalis vetulus
3.00(max)
9.4
Regression*
S. serrulatus
2.90
9.1
Regression*
Diaptomus siciloides

2.0
Comita 1964
Regression for Cladocerans
F = 3.16 L - .09.	r^ = .60	n = 8
111
-------
Fig. 42. Changes in pH, TP, Secchi <
in the enclosures and lake
enclosures utilizing black
isc transparency and chlorophyll a
during experiment 7 (w/blk indicates
plastic).
112
-------
Exp 7
13
3
5

6
2
L .
14

1

11
8
9

Chi = 0.046 TP ^ 7.22
r2 = 0.095

0	100	200	300
TP mg m 3
43. Relationship between average concentrations of
TP and chlorophyll a^ in the enclosures and lake
during experiment 7 (numbers refer to different
enclosures, L = lake).
113
-------
and chlorophyll concentrations in the enlcosures. The regression is not
significant. It is important to note that the lowest yield (Chl/TP) occurred
in bags 8, 9. and 11, which had the fastest mixing rates. The correlation
becomes significant if these points are omitted. This provides additional
support for the hypothesis that some unknown factor associated with rapid
mixing can inhibit the algae. The factors most likely to be involved in this
inhibition effect will be considered in the discussion section. The species
composition changed dramatically in several of the enclosures. All
enclosures were initially dominated by the blue-green alga, Anabaena
ci rcinalis. The composition in most of the bags switched to predominantly
diatom and green species by the end of the experiment. The switch occurred
first in the rapidly mixed bags and then eventually occurred in 11 out of the
14 bags (4, 5. 6. 7. 8, 9. 10, 11, 12, 13. 1*0 including both controls. A
similar shift did not occur in the lake. The abundance of herbivorous
zooplankton increased in all enclosures except bag 12 (control) during the
experiment (Fig. 22). The fact that this increase was observed under all
treatment conditions as well as in the lake indicates that some uncontrolled
factor common to all of the bags was responsible. This factor may have been
pH which was low in all of the enclosures during the experiment. Our
theoretical framework suggests that lower pH should be favorable to the
development of herbivorous zooplankton.
Experiment 8
Experiment 8 was set up on Twin Lake and lasted from August 11 to
September 13, 1978. Four open bottom enclosures were suspended from a single
raft. The experiment was designed to determine whether the mixing rates
which inhibited algae in experiment 7 would have a similar effect on Twin
Lake algae. Only one mixing rate (250 ml/min. @ 7m) was tested. Treatment
assignments are listed in Table 15*
Integrated samples were taken to a depth of 5«5 meters. Zooplankton
tows were taken three times during the experiment. All other procedures were
the same as in experiment 1. Changes in pH, TP, SD, and Chi a are indicated
in Figure 44. The pH dropped initially about 1 unit in both rapid mix
enclosures (1 and 3) while the pH increased in the control bags (2 and 4).
As in previous experiments TP increased in the mixed bags and TP was highly
correlated with Chi a. Figure 45 shows the relationship between average TP
and Chi concentrations in the enclosures. The regression is highly
significant and similar to those determined in previous Twin Lake experiments
(experiments 3 and 6). No inhibition was evident as was found in experiment
7. All enclosures were dominated by Osci1latoria rubescens throughout the
experiment. Initial zooplankton samples were lost. However, the abundance
of herbivorous zooplankton was higher in the mixed enclosures than in the
controls and the lake on the other two sampling dates (see Fig. 22).
Circulation apparently had a stimulatory effect on the zooplankton as
predicted in the theoretical framework.
DISCUSSION
The experimental results presented in this section provide evidence for
the action of a number of specific response mechanisms affecting algal
species composition and abundance during artificial circulation.
1X4
-------
bag 1
lake
250
bag 2
control
CO
60-
40-
20-
CO
100-
50-
bag 4
20
bag 3
250
control
30
60-
40-
20-
100-
50-
days 0
20
30
20
30
a
Fig. 44. Changes in pH, TP, Secchi disc transparency and chlorophyll in
the enclosures and lake during experiment 8.
115
-------
40
Exp 8
Chi 20
mg m
Chi = 0.56 TP - 4.33
50
0
100
TP mg m
Fig. 45. Relationship between average concentrations of
TP and chlorophyll a in the enclosures and lake
during experiment 8 (numbers refer to different
enclosures, L = lake).
116
-------
Response Mechanisms Affecting Species Composition
The response of the phytoplankton at the species level seemed to depend
primarily on changes in water chemistry which occurred in the euphotic zone
during mixing. These changes in water chemistry depended, in turn, on the
mixing rate used and on individual lake chemistry. This was most clearly
demonstrated in the open bottom bags of experiments 1, 2 and 3- Rapid mixing
generally resulted in high concentrations of C02 and nutrients at the surface
accompanied by reduced pH levels. These conditions seemed to favor the
development of green algae and diatoms while suppressing the growth of
blue-green algae (cf. experiments 1 and 2). This shift to greens and diatoms
occurred most frequently in rapidly mixed Little Lake Johanna enclosures.
This was apparently a result of the lower pH and higher nutrient levels
achieved during mixing due to lower alkalinity and higher hypolimnetic
levels of C02 and nutrients. Nutrient concentrations also increased at the
slower mixing rates. However, these mixing rates were apparently insufficient
to produce a significant increase in C02 at the surface and therefore, the pH
usually remained high. These conditions of both high pH and high nutrient
levels seemed to favor the development of blue-green algae. The blue-green,
Hicrocystis aeruginosa grew particularly well under these conditions reaching
high densities in both of the slow mixed enclosures in experiment 2. The
shift in competitive advantage from blue-greens to greens which was observed
as the mixing rate was increased was apparently related to changes in pH (see
Fig. 21). While we were able to isolate the effects of pH and nutrients from
other hypolimnetic factors by using closed bottom bags in experiments *~, 5
and 6, the details of this pH-shift mechanism are still unclear. There was
some indication that the shift to greens at low pH may have been due to the
selective inhibition of blue-green algae. This can be seen in the results for
experiment 1 (Fig. 21) where the growth rates for blue-greens decreased
steadily as the pH dropped in the deep mixed enclosures. Shapiro (1973) also
found that blue-greens were suppressed at low pH in his field experiments.
The mechanism for this suppression may involve the activity of cyanophage.
Lindmark (1979) demonstrated a dramatic increase in the incidence of viral
infection as the pH was lowered in laboratory cultures of blue-green algae.
Alternatively, blue-green to green shifts observed may involve the
differential effects of carbon limitation for these two divisions. King
(1970) suggested that, at a given level of alkalinity, carbon limitation
should increase with pH and that blue-greens were better adapted to low
alkalinity and high pH while greens were better suited to conditions of low
pH and high alkalinity. Long (1979) confirmed King's hypothesis in a series
of laboratory growth experiments. He found that the half saturation constant
for carbon uptake, Ks, increased with pH for a large number of species, and
that, in general, blue-greens had lower Ks values than greens at a given pH.
He also demonstrated the superior uptake characteristics of green algae at
low pH and high alkalinity levels and predicted a shift in dominance from
blue-green to green algae as the alkalinity increases and pH decreases.
Figure ^6 presents a summary of the conclusions of Long and King as they
relate to the results obtained in experiments 1, 2, and 3- The dotted line
was calculated by Long from his laboratory results. Blue-greens should
outcompete greens above the line and greens should dominate below it. The
analogous "transition zone" suggested by King (1970) from his observations on
sewage lagoons is also shown. Long felt that the difference between his
dividing line and King's was due to the higher levels of mixing which he used
in his laboratory experiments compared to the stagnant conditions in the
lagoons. King (1970) showed that the effects of carbon limitation could be
greatly reduced by increasing the mixing rate in laboratory cultures. The
117
-------
V 2-L
2-7
2-2
2-10
2-4
2-1
«s£
01-3
3-5
-L
1-6
B.5-
3 7 3-8
3-6 3-2
3-3
pH
3-4
2-8
2-9
7.S"
2-5
1-4
1-5
1-7
7.0-
6.5-
2-6
90 100 110 120 130 140
50
80
60
70
40
Alkalinity mg-I-1 as CaC03
Fig. 46. Expected effect of pH and alkalinity on the competi-
tive relationship between greens and blue-greens.
(Shaded region = transition zone of King (1970),
dotted line = transition boundary of Long 1978).
Values of average pH and alkalinity for experiments
1, 2 and 3 are indicated by 1-J, where !•= experi-
ment number and J = enclosure number (L = lake).
118
-------
average levels of alkalinity and pH observed in enclosures in experiments 1,
2 and 3 are also indicated on the Figures. In general, deep mixing resulted
in an increase in alkalinity and decrease in pH. A wide range of alkalinity
and pH levels was obtained in Little Lake Johanna (expt. 1 and 2) while a
smaller range was observed in Twin Lake (expt. 3)- Changes in species
composition observed in these experiments were not entirely consistent with
the predictions of Long and King. The values of pH and alkalinity in all of
the enclosures for all three experiments were within the range which Long
predicted should lead to dominance by green algae. However, while shifts in
either the relative abundance or competitive advantage of greens were
observed in several of these enclosures there were also many instances, most
notably in bags 2 and 7 in experiment 2 and all of the bags in experiment 3.
where blue-green algae dominated. The results are somewhat more consistent
with the predictions of King in that the bags in experiment 2 which were
dominated by blue-greens occurred above his transition zone. However, the
alkalinity and pH levels for Little Lake Johanna in experiments 1 and 2 were
also in this region (1-L, 2-L) and these communities were dominated by green
algae. In addition, most of the enclosures in Twin Lake, which were
dominated by blue-greens still fell within the "green" region. While the
lack of agreement between experimental results and the predictions of Long
and King is difficult to explain, there was one trend in the experimental
data which suggests that carbon limitation may still be involved in the pH
shift mechanism during artificial circulation. This was the shift in
competitive advantage from blue-greens to greens which occurred in the deep
mixed enclosures in experiment 1 (Fig. 21). This shift was accompanied by a
large drop in pH and increase in alkalinity. While green algae never fully
gained the advantage over the blue-greens at the faster mixing rate, this
result suggests that a shift in species composition might occur if the pH
fel1 much further .
An increase in diatoms was also observed in several of the rapidly mixed
enclosures. Similar increases in diatom species have also been reported in
whole lake destratificat ion experiments (Bernhardt 1967 > Haynes 1975.
Knoppert e_t aj_. 1970, Lackey 1973. Weiss and Breedlove 1973) • As with the
green algae, the advantage of diatoms at the faster mixing rates may have
been related in part to a reduction in carbon limitation resulting from lower
pH and higher C02 concentrations. However, the results obtained in
experiment 1 indicate that another mechanism may have also been involved.
The growth rates of all species increased dramatically in the "shallow mixed"
enclosures without a major change in the levels of pH, C02, alkalinity or
TP/Chl a (Fig. 21) suggesting that neither carbon nor phosphorus limitation
were involved. These increased growth rates were apparently related to the
direct effects of turbulent mixing. However, the mechanism involved is not
clear. Increased turbulence may have reduced nutrient limitation indirectly
by breaking down the zones of nutrient depletion which can surround algal
cells under quiescent conditions (King 1970. Munk and Riley 1952).
Alternatively, the sinking velocities of algal cells may have been reduced as
the mixing rate increased resulting in lower sinking losses. This latter
mechanism would have had the most significant effect for the heavier diatom
species. The fact that diatoms showed the greatest increase in growth rate
in the shallow mixed bags suggest that this might be the principal mechanism
involved. Knoechel and Kalff (1975) found that seasonal changes in turbulence
had a significant differential effect on the sinking losses of a blue-green
and a diatom species in Lac Hertel, Quebec, resulting in a dramatic shift in
their relative dominance. It is also possible that, in our system, the
increase in diatoms at the faster mixing rates was an "enclosure effect"
caused by the dislocation and suspension of periphytic species growing on the
119
-------
side walls. The enclosure walls often became covered by a dense layer of
periphyton, predominantly diatoms, by the end of an experiment. This
increase in periphyton often coincided with the shifts to diatom species
observed in the plankton. The fact that many of the diatom species which
occurred in the plankton were also periphytic forms (e.g. Synedra spp.
N i tszch i a spp.) suggests that these shifts may have been artificial.
Osc i11ator i a rubescens was the dominant alga in Twin Lake during all of
the Twin Lake experiments. This alga commonly resided in the metalimnion and
was therefore only observed in the open bottom enclosures, and in the lake,
it generally responded favorably to mixing and its population usually
increased in proportion to the total phosphorus concentration (Figs. 26, 39i
kS) • Smith et a_[. (1975) also found that metal imnetic populations of
Oscillatoria rubescens increased following destratification in eutrophic
Mirror Lake, Wisconsin. However, Bernhardt 0967) observed a dramatic
decline in this species following destratificat ion in Wahnbach Reservoir and
a similar reduction was observed in a metalimnetic population of Osc i11ator i a
tenu i s during artificial circulation (Weiss and Breedlove 1973)* The
autoradiographic results obtained in experiment 3 indicate that the average
photosynthetic rate of the Osc i11ator i a population increased dramatically
after destratification (Fig. 27). Baker et a 1. (1969) also found an increase
in photosynthesis when they took samples from a metalimnetic population of
Osc i11ator i a aqardh i i and incubated them near the surface. In a similar
experiment Watanabe (1979) observed higher growth rates for Osc i11ator i a
mougeot i a when samples taken from the metalimnion were incubated near the
surface. He felt that these higher growth rates were due to either the
higher light levels or higher temperatures found near the surface, or both.
The increase in photosynthesis observed for Osc i11ator i a rubescens in the
mixed bags, then, was apparently the result of increases in both the
temperature and available light which occurred as the metalimnetic population
was brought to the surface. Once it was "released" from light limitation, the
population grew to a level determined primarily by the concentration of total
phosphorus in the mixed layer (Fig. 26). The response of Osc i11ator i a
rubescens was not consistent with the prediction of equation 11 that the
population should decrease as the mixed depth is increased due to increased
light limitation. This contradiction occurred because one of the assumptions
of the model was violated. The model assumes that the algae are uniformly
distributed through the mixed layer while, in this case, they were
concentrated in the metalimnion.
Mechanisms Involved in the Community Response During Artificial Circulation
The evaluation of the mathematical model presented above with data from
experiment 6 indicates that this model provides the basic theoretical
framework necessary to understand the community response during artificial
circulation. This framework is summarized by equation 11. The parameters in
this equation, most likely to change during destratification are the mixed
depth, Zm, and the concentration of total phosphorus, TP, in the mixed layer.
The mixed depth always increases during artificial circulation while the
results of our experiments suggest that TP also usually increases. Equation
11 predicts that these two changes will have opposite effects on the maximum
concentration of chlorophyll attained in the mixed layer, c*. An increase in
TP will cause c* to increase while an increase in Zm causes c* to decrease.
These contradictory effects of Zm and TP provide a mechanistic explanation
for the ambiguous results obtained in previous whole lake destratification
experiments. It is clear that c* can either increase, decrease or remain the
120
-------
same following destratification depending on which factor, TP or Zm, has the
greater effect. This was clearly demonstrated in the simulation for Twin
Lake (Fig. ItOb) . Since destrat i f i cat i on of the experimental enclosures
resulted in a large increase in TP and only a small increase in Zm, C*
increased dramatically. As with the population density, c*, the total algal
population in the mixed layer, c*Zm, is also expected to increase if TP
increases following destratification. However, c*Zm, may either increase,
decrease or remain the same, as Zm changes depending on both the initial and
final value of Zm. This general response pattern was shown in the simulation
for Twin Lake (Fig. 40b) . The small increase in Zm and the large increase in
TP observed in the mixed experimental enclosures both acted to increase the
value of c* Zm and it is expected that a similar result would be obtained if
the whole lake was destratified. An analysis similar to the one given for
Twin Lake was presented by Lorenzen and Mitchell (1975) for Kezar Lake, New
Hampshire. They considered the effects which nutrient and light limitation
would have on algal growth during artificial circulation and although they
used a different model their results were qualitatively the same. They also
showed that peak biomass, c^Zm, would either increase, decrease or remain the
same depending on the initial and final value of Zm. As in the analysis for
Twin Lake, they predicted that peak biomass, c*Zm, would reach a maximum at a
specific mixed depth which they had defined in a previous paper (Lorenzen and
Mitchell 1973) as the optimal mixed depth, Z-opt. However, the value of
Z-opt determined for Kezar Lake (approx. 1-5 m) was much less than the value
determined for Twin Lake (approx. 15 m). This lower value of Z-opt in Kezar
Lake means that a much smaller increase in Zm would be required to cause a
reduction in c'
-------
3 [46]
6 [313
75
CO
O)
^ 50"
o
2 [7]
25-
100
TP mg-rn"^
150
50
Fig. 47. Regression lines from relationships between average
TP and chlorophyll s concentrations determined for
experiments 1, 2, 3, 5, 6 and 8 (numbers in paren-
thesis indicate the initial ratio of inorganic
nitrogen/inorganic phosphorus in each experiment).
122
-------
than 5, and could be limited by either if IN/IP was in the range 5~12. It is
clear, then, that the phytoplankton in Twin Lake were limited by phosphorus
while the low yield and low IN/IP observed in Little Lake Johanna suggests
that these phytoplankton were limited by nitrogen. These results indicate
the importance of knowing which nutrient limits algal growth since this will
be the nutrient which mainly influences the community response during mixing.
The mathematical model presented above will have to be modified to describe
the community change in N-limited lakes during artificial destratification.
This may be done by re-defining the model parameters, TP, Kq and (*mcfx in
terms of total nitrogen. However a test of this "total nitrogen" model will
have to be made before it can be recommended for routine use.
In addition to Zm and the concentration of limiting nutrient, the loss
rate D may also change during artificial circulation. In the theoretical
framework an increase in D would result in a decrease in c* (equation 11).
Although D was fairly constant in experiment 6, small differences in the
grazing rate between enclosures did affect the final yield of chlorophyll
(Fig. 41) with the lowest yield observed in the enclosure with the highest
grazing rate. Oskam (1978) also found an inverse correlation between the
abundance of zooplankton and the yield of chlorophyll a in several
destratified eutrophic reservoirs. Several investigators have reported
increases in the abundance of herbivorous zooplankton during whole lake
circulation (Riddick 1957. Brown et a_K 1971. McNall 1971. Shapiro and
Pfannkuch 1973)- These changes might also be expected to result in lower
yields of phytoplankton. However, the final concentration of chlorophyll
achieved will always depend on the combined effect of changes in D, Zm and
the concentration of the limiting nutrient as indicated in equation 11.
The constants: Pmax, Kq,©, In (lo/lz1) and Ec in equation 11 should
change with community composition while the constant Ew will depend on the
amount of non-algal material dissolved or suspended in the water. These
"constants" determine how changes in the concentration of limiting nutrient,
Zm and D which occur during mixing will influence the community response.
Since community composition and Ew can vary significantly both between lakes
and within a given lake it will be necessary to perform a separate analysis
for each lake and community considered. It will also be necessary to
consider the effects which changes in species composition during artificial
circulation will have on the model constants.
The low yield of algal biomass per unit of TP which was often observed
at the fastest mixing rates (experiment 1, bag 7; experiment 4, bag 10;
experiment 7. bags 8, 9. and 11) is difficult to explain. This apparent
inhibition of growth may have been related to the low surface temperatures
achieved in these bags. A surface temperature of 12.4 degrees C was produced
in bag 7 of experiment 1 at a mixing rate of 1000 ml min-1. This was much
lower than the value of 23 degrees C obtained at a mixing rate of 100 ml
min-1 (bag 5). although the surface levels of pH and nutrients achieved in
the two bags were quite similar (c.f. Figure 21). If temperature were the
only factor responsible for the difference in composite growth rates in
Experiment 1 then a simple temperature correction made by assuming a Q10
value of 2.18 (Goldman and Carpenter 1974) and normalizing all growth rates
to a common temperature should give similar results in both enclosures. When
the composite growth rates for diatoms in bags 5 and 7 are both normalized to
23 degrees C, they are, in fact, quite similar (0.54 and 0.57 day-1,
respectively). However, when the same correction is applied to the composite
growth rates of blue-greens and greens a similar correspondence is not
obtained (for bags 5 and 7. respectively: blue-greens: 0.13 and 0.022 day-1;
greens: 0.06 and 0.018 day-1). There is apparently some other factor in
123
-------
addition to temperature which is contributing to the lower growth rates for
these latter divisions at the faster mixing rate. The fact that all of the
rapidly-mixed low-yield enclosures were open at the bottom suggests that some
hypolimnetic factor may be involved. Symons et al. (1969) also observed some
inhibition of algal growth during whole lake circulation. They suggested
that increased concentrations of hydrogen sulfide which they noted at the
surface during mixing may have been responsible for the effect. The fact that
hydrogen sulfide was also found at the surface of bag 7 during experiment 1
suggests that a similar mechanism may have been involved in the inhibition of
blue-greens and greens in this enclosure. Clearly, further research into the
effects of temperature and hypolimnetic constituents on algal growth is
needed before this problem can be fully resolved.
ACKNOWLEDGEMENTS
We gratefully acknowledge E. Smeltzer, M. Lynch, V. Smith, N. Allott,
E. Swain, and B. Monson for their assistance during the course of these
experiments. D. Baker provided the solar irradiance data used in the
productivity analysis. The patience and assistance of the residents
surrounding Twin Lake and Little Lake Johanna is also greatly appreciated.
REFERENCES
Allott, N. 1978. Recent paleolimnology of Twin Lake near St. Paul,
Minnesota, based on a transect of cores. M.S. Thesis, University of
Minnesota. 102 pp.
American Public Health Association, 1971- Standard Methods for the
examination of water and waste water. 13th ed. Washington, D.C. American
Public Health Association, 8?4 pp.
Baker, A.L., A.J. Brook and A.R. Klemer 19&9- Some characteristics of a
naturally occurring population of Osc i11ator i a aqardh i i Gamont: Limnol.
Oceanogr., 14:327*333 -
Bannister, T.T. 1974a. Production equations in terms of chlorophyll
concentration, quantum yield, and upper limit to production. Limnol.
Oceanogr. 19:1-12.
Bannister, T.T. 1974b. A general theory of steady state phytoplankton growth
in a nutrient saturated mixed layer. Limnol. Oceanogr. 19:13—31 -
Bella, D.A. 1970. Simulating the effect of sinking and vertical mixing on
algal dynamics. J. Water Poll. Contr. Fed. 42:140—152.
Bernhardt, H. 1967- Mog1ichkeiten der verbinderung anaerober verhaltnisse in
einer trinkwasser talsperre warend der sommerstagnation. Arch.
Hydrobiol. 63(3):404-428.
Brezonik, P.L., J. Delfino and G.F. Lee 19&9- Chemistry of N and Mn in Cox
Hollow Lake, Wisconsin, following destratification. J. San. Eng. Div.
ASCE 95 (SA5):929-940.
Brown, D.J., T.G. Brydges, W. Ellerington, J.J. Evans, M.F.P. Michalaski,
G.G. Hitchin, M.D. Palmer and D.M. Veal 1971- Progress report on the
destratificat ion of Buchanan Lake. Ont. Water Res. Comm., AID for Lakes
Program.
Burns, C.W. and F.H. Rigler. 1S6J. Comparisons of filtering rates of Daphn i a
rosea in lake water and in suspensions of yeast. Limnol. Oceanogr. 12:
492-502.
124
-------
Brynildson, O.M. arid S.L. Serns 1977- Effects of destrat i f icat i on and
aeration of a lake on the distribution of planktonic Crustacea, yellow
perch and trout. Wise. Dept. of Nat. Res. Tech. Bull. No. 99. 22p.
Bueltman, C.G. (chairman) 1969. Provisional Algal Bioassay Procedure. Joint
Industry/Government Task Force on Eutrophication, New York, NY.
Chaney, A.L. and E.P. Marbach 1962. Modified reageants for determination of
urea and ammonia. Clinical Chemistry 8:130-132.
Comita, G.W. 1964. The energy budget of piaptomus siciloides, Lilljeborg.
Verh. Internat. Verein. Limnol. 15*-653•
Droop, M.R. 1974. The nutrient status of algal cells in continuous culture.
J. Mar. Biol. Ass. U.K. 54:825-855-
Drury, D.D., D.B. Porcella and R.A. Gearheart 1975- The effects of
artificial destratification on the water quality and microbial
populations in Hyrum Reservoir, Utah Water Res. Lab., Utah State
University, Logan, Utah. PR EJEWO 11-11.
Fast, A.W. 1971. The effects of artificial aeration on lake ecology. U.S.
Environmental Protection Agency. Water Pollution Control Series 16010
EXE 12/71. 470 pp.
Forsberg, B.R. I98I. A general theory of phytoplankton growth for field
applications. Ph.D. Thesis, University of Minnesota.
Forsberg, C., S. Ryding, A. Claesson and A. Forsberg 1978. Water chemical
analyses and/or algal assay-sewage effluent and polluted lake water
studies. Mitt. Internat. Verein. Limnol. 21:352-362.
Goldman, J.C. and E.J. Carpenter 197^+- A kinetic approach to the effect of
temperature on algal growth. Limnol. Oceanogr. 19:756-766.
Haney, J.F. and D.J. Hall. 1973' Sugar-coated Daphnia: A preservation
technique for Cladocera. Limnol. Oceanogr. 18:331~333-
Haynes, R.C. 1971. Artificial circulation of a small eutrophic New Hampshire
lake, Ph.O. Thesis, University of New Hampshire
Haynes, R.C. 1973- Some ecological effects of artificial circulation on a
small eutrophic lake with particular emphasis on phytoplankton, I. Kezar
Lake Experiment, 1968. Hydrobiologia 43:463-504.
Haynes, R.C. 1975* Some ecological effects of artificial circulation on a
small eutrophic lake with particular emphasis on phytoplankton. II.
Kezar Lake Experiment, 1969* Hydrobiologia 46:141-170.
Hedman, E. and S. Tyley 196 7 - Elimination of stratification of Lake Cachuma.
California. USGS Water Resources Div. Menlo Park, Calif. 40 pp.
Hooper, F.F., R.C. Ball, and H.A. Tanner 1952. An experiment in the
artificial circulation of a small Michigan Lake. Trans. Amer. Fish. Soc.
82:222-241 .
Ivanova, M.B. and R.Z. Klekowski 1972. Respiration and filtration rates in
S imocepha1 us vetu1 us (O.F. Muller) (Cladocera) at different pH. Pol.
Arch. Hydrobiol. 19:303~3^8.
King, D.L. 1970. The role of carbon in eutrophication. J. Water Poll.
Contr. Fed. 42:2035-2051.
King, D.L. 1979- Lake Measurements, In: Lake Restoration. Proceedings of a
National Conference. Minneapolis, Minn., 1978.
Kalff, J. and R. Knoechel. 1978. Phytoplankton and their dynamics in
oligotrophic and eutrophic lakes. Ann. Rev. Ecol. Syst. 9:475"495-
Knoechel, R. and J. Kalff 1975- Algal sedimentation: the cause of a
diatom-bluegreen succession. Verh. Internat. Verein. Limnol.
19:7^5-754.
.125
-------
Knoppert, P.L., J.J. Rook, T. Hotker and G. Oskam 1970. Destratificat ion
experiments in Rotterdam. J. Amer . Water Works. 62:•
Kring, R.L. and W.J. O'Brien 1976. Accommodation of Daphnia pulex to altered
pH conditions as measured by feeding rate. Limnol. Oceanogr.
21:313-315.
Lackey, R.T. 197 3• Artificial reservoir destratification, effect on
phytoplankton. J. Water Poll. Contr. Fed. :668-673-
Leach, L.E., W.R. Duffer, and C.C. Harlin Jr. 1970. Induced hypolimnetic
aeration for water quality improvement of power releases. U.S.
Environmental Protection Agency, Water Pollution Control Series 16080.
Ada, Oklahoma. 32 pp.
Lee, G.F. 1970. Factors affecting the transfer of materials between water
and sediments. Literature Review No. 1 Eutrophication Information
program, University of Wisconsin, Madison. 35 PP-
Lehman, J.T., D.B. Botkin and G.E. Likens 1975- The assumptions and
rationales of a computer model of phytop 1ankton population dynamics.
Limnol. and Oceanogr. 20:3^3"36^.
Lindmark, G. 1979* Interaction between LPP-1 virus and PIectonema boryanum.
Ph.D. Thesis, University of Lund., Sweden.
Long, E.B. 1979- The interaction of phytopiankton and the bicarbonate
system. Ph.D. Thesis, Kent State University, 112 pp.
Lorenzen, M. and R. Mitchell 1973- Theoretical effects of artificial
destratification on algal impoundments. Envir. Science and Tech.
7:939-9^.
Lorenzen, M. and R. Mitchell 1975- An evaluation of artificial
destratification for control of algal blooms. J. Amer. Water Works
Assoc. 67:372-376.
Lorenzen, M. and A.W. Fast 1976. A guide to aeration/circulation techniques
for lake management. U.S. Environmental Protection Agency.
EPA-600/3-77-004.
Lund, J.W.G. 19^9' Studies on Asterionella. I. The origin and nature of the
cells producing seasonal maxima. J. Ecol . 37: 389_1+19 *
Malueg, K., J. Tilstro, D. Schultz and C.F. Powers 1971• The effect of
induced aeration upon stratification and eutrophications processes in an
Oregon farm pond. Internat. Symp. on Man made Lakes. KnoxviUe,
Tennessee.
McNall, W.J. 1971' Destratificat ion of lakes. Job Progress Report C-8,
Statewide Fisheries Investigations. New Mexico Federal Aid Proj. No.
F-22-R- 11 . 31 PP.
Megard, R.O. and P.D. Smith 197^- Mechanisms that regulate rates of growth
of phytopiankton in Shagawa Lake, Minnesota. Limnol. and Ocean.
19:279-296.
Megard, R.O., W.S. Combs Jr., and J.C. Settles 1978. Planktonic algae,
photosynthesis, and oxygen in the Mississippi River at Minneapolis and
St. Paul, Minnesota, Final Report, Metropolitan Waste Commission
MWCCi*69.
Megard, R.O., W.S. Combs Jr., P.D. Smith and H.S. Knoll 1979- Attenuating
light and daily integral rates of photosynthesis attained by planktonic
algae. Limnol. and Oceanogr. 2^:1038-1050.
Menzel, D.W. and N. Corwin 1965- The measurement of total phosphorus in
seawater based on the liberation of organically bound fraction by
persulfate oxidation. Limnol. Oceanogr. 10:280—283•
126
-------
Munk, W.H., and G.A. Riley 1952. Absorption of nutrients by aquatic plants.
J. Mar. Res. 11:215-240.
Nauwerck, A. 19^>3• Die Beziehungen zwischen Zooplankton und Phytoplankton im
See Erken. Symb. Bot. Upsal. 17:163 pp.
Oskam, G. 1978. Light and 2ooplankton as algae regulating factors in
eutrophic Brisbosch Reservoirs. Verh. Internat. Verein. Limnol.
20:1612-1618.
Patten, B.C. 1968. Mathematical models of plankton production. Int. Rev.
Gesamten Hydrobiol. 53:357-408.
Porter, K.G. 1973* Selective grazing and differential digestion of algae by
zooplankton. Nature, 244:179~l80.
Riddick, T.M. 1957- Forced circulation of reservoir waters yields multiple
benefits at Ossining. N.Y. Water and Sew. Works 104:231-237-
Ridley, J.E., P. Cooley and J. Steel 1966. Control of thermal stratification
in Thames Valley reservoirs. Proc. Soc. Water Treat. Exam. 15:225.
Robinson, E.L., W.H. Irwin and J.M. Symons 1969- Influence of artificial
destratification on planktonic populations in impoundments. Trans.
Kent. Acad, of Sci. 30:1 -18.
Schindler, D.W., R.V. Schmidt, and R. Reid 1972. Acidification and bubbling
as an alternative to filtration in determining phytoplankton production
by the 14C method. J. Fish. Res. Bd. Canada 29:1627-1631.
Senft, W.H. 1978. Dependence of light-saturated rates of algal
photosynthesis on intra-ceI 1u1ar concentrations of phosphorus. Limnol.
Oceanogr. 23:585~840.
Shapiro, J. 1973- Blue-green algae: why they become dominant. Science
179:382-384.
Shapiro, J. and H.O. Pfannkuch 1973* The Minneapolis chain of lakes. A
study of urban drainage and its effects 1971-1973- Interim Rep. 9»
Limnol. Res. Center, Univ. Minnesota, 250 pp.
Sirenko, L.A. and others 1972. Effect of artificial water aeration on basin
algal flora. Gidrobiol. Zh. 8:68-72.
Slack, K.V. and G.G. Ehrlich 1967- Water quality changes in a destratified
water column enclosed by polyethylene sheet. U.S.G.S. Prof. Paper
575-6:235-239-
Smayda, T.J. 1970. The suspension and sinking of phytoplankton in the sea.
Oceanogr. Mar. Biol. Ann. Rev. 8:353~4l4.
Smith, S.A., O.R. Knauer and T.L. Wirth 1975- Aeration as a lake management
tool. Technical Bull. No. 87. Wisconsin Dept. Nat. Resour., Madison, 40
PP.
Snedecor, G.W. and W.G. Cochran 1973- Statistical Methods. Iowa State
University Press. Ames, Iowa. 593 PP-
Strathmann, R.R. 1967- Estimating the organic carbon content of
phytoplankton from cell volume or plasma volume. Limnol. and Oceanogr.
12:411-418.
Strickland, J.D. and T.R. Parsons 1968. A practical handbook of seawater
analysis. Bull. Fish. Res. Bd. Can. 167-
Stumm, W., and J.J. Morgan 1970. Aquatic Chemistry. John Wiley and Sons,
Inc., N.Y. 583 pp.
Symons, J.M., W.H. Irwin, E.L. Robinson and G.G. Robeck 1969- Impoundment
destratificat ion for raw water quality control using either mechanical
1	or diffused-air pumping. J_n, Water Quality Behavior in Reservoirs,
James M. Symons (ed., U.S. Dept. of Health, Education and Welfare, Paper
No. 12, pp. 389-427.
127
-------
Tailing, J.F. 1957a* Photosynthetic characteristics of some freshwater
plankton diatoms in relation to underwater radiation. New Phytol.
56:29-50.
Tailing, J.F. 1957b. The phytopiankton population as a compound
photosynthetic system. New Phytol 56:133"1^9•
Toetz, D., J. Wilhm and R. Summerfelt 1972. Biological effects of artificial
destratificat ion and aeration in lakes and reservoirs - analysis and
bibliography. U.S. Dept. of Interior REC-ERC-72-33.
Utermohl, H. 1958. Zur Vervol1kommnung der quantitativen
PhytopIankton-Methodik. Mitt. Internat. Verein. Limnol. 9:1 -38•
Watanabe, M.F. 1979* Studies on the metalimnetic bluegreen alga Osc i11ator i a
mouqeot i a in a eutrophic lake with special reference to its population
growth. Arch. Hydrobiol. 86:66-86.
Weiss, C.M. and B.W. Breedlove 1973. Water quality changes in an impoundment
as a consequence of artificial destratification. Water Resources Res.
Inst., University of N. Carolina, Report No. 80.
Wetzel, R.G. 1975- Limnology. W.B. Saunders Co., N.Y., 7^3 PP*
Wirth, T.L. and R.C. Dunst 1967- Limnological changes resulting from
artificial destratification and aeration of an impoundment. Wisconsin
Conserv. Dep., Fish. Res. Rep. No. 22.
128
-------
II. C. EFFECTS OF ENVIRONMENTAL STRESSES ON THE RELATIONSHIP BETWEEN
PLECTONEMA BORYANUM AND CYANOPHAGE LPP-1*
INTRODUCTION
For many years a reduction of nutrient input to lakes was considered the
universal remedy to combat the cultural eutrophication created during the
last decades. Although meaningful, the measures were not sufficient in
several cases (Bjork et aj_. 1972, Ryding and Forsberg 197&) » and additional
alternative approaches are in demand (Shapiro 1977. 1978).
Changing phytopiankton communities from blue-green algal dominance to
green algae with the intention of improving the algal food resource for
higher organisms in the ecosystem is an attractive approach which was
successfully accomplished by Shapiro (1973) and Shapiro et aK (19755 * In
their experiments, performed on a small scale in the field (plastic
enclosures), a rapid shift from blue-green algae to green algae followed an
injection of C02 and the accompanying drop in pH. Whether the supply of C02
favored the green algae in competition with blue-green algae for carbon, or
whether other organisms were affected and involved in the mechanism of the
shift is still unclear. Artificial circulation of whole lakes has sometimes
resulted in a shift from blue-green to green algae or diatoms. The
successful ventures generally were accompanied by a lowering of pH (Shapiro
1977) .
The recent attention focused on the microbial pathogens of blue-green
algae and their interrelations has exposed an attractive field where many
questions remain despite intense research efforts (Safferman and Morris T9^3»
Stewart and Brown 19&9. Safferman 1973b, Padan and Shilo 1973« Daft et a 1.
1975, Stewart and Daft 1977)- Several investigators (Safferman and Morris
196^, Jackson and Sladecek 1970, Cannon et aj_. 197^» Cannon 1975) have
commented on the use of viruses as a means of controlling algal blooms.
However, Goryushin (1976) warns of the ability of cyanophages to induce
mutation in higher organisms. Nevertheless, the presence of cyanophages in
environments common to many blue-green algae, e.g. at high pH, must be
significant as a selective factor.
The present study is an attempt to clarify whether relations between
viruses (cyanophages) and their algal hosts can be affected by manipulations
in the environment. Is it possible to activate cyanophages and accelerate
lysis of blue-green algal populations or to enhance the resistance of
blue-green algae to attack from cyanophages?
The experiments presented here were performed under laboratory
conditions with a well-known algal-cyanophage system, PIectonema boryanum and
cyanophage LPP-1 (attacking strains of Lynqbya. Phormidiurn and P1ectonema).
The work was done in close connection with field experiments on natural
blue-green algae communities, however, because the nature of the induced
blue-green algal collapse in plastic enclosures suggested lysis of the algal
cells (Shapiro et aj_. 1975. and pers. commun.) .
The rate of LPP-1 cyanophage replication and lysis of PIectonema was
studied in relation to:
a)	pH alterations by C02/air additions
b)	algal host culture age and density
c)	nutrient concentrations
d)	presence of additional algal species
The results will be discussed in relation to observations from field
experiments (Shapiro 1973. Shapiro 1977)-
*by Gunilla Lindmark and Joseph Shapiro
129
-------
MATERIALS AND METHODS
Algal and Cyanophaqe Strains
The filamentous blue-green alga, Plectonema boryanum IU 581 (Indiana
Univ. Culture Collection) and its cyanophage LPP-1 (kindly supplied by Dr.
R.E. Cannon at the University of North Carolina, Greensboro) were the main
organisms selected for the experiments. Morphological and physicochemica1
characteristics of the LPP cyanophage and its algal hosts and the process of
viral infection are reviewed by Padan and Shilo (1973)- Additional algal
strains used were cultures of the blue-green alga Anabaena c i rc i na1i s (Wis
1038) and the green alga Scenedesmus sp. (probably S_;_ obi iquus, Rhee 1972).
Culture Methods
Algal stock cultures were grown in a synthetic nutrient solution
(modified Gorham's medium, Table 17). and algal cells were routinely
transferred into fresh nutrient solution at intervals of 10 days. Cells from
12-day old stock cultures (density ca. 20 million cells/ml) were used as
inoculum in the experiments. Both stock and experimental algal cultures were
incubated at 2b± 2 degrees C and were constantly illuminated by cool-white
fluorescent lamps with an intensity of 1500 lux. The stock cultures were
hand shaken once daily while cultures in the experimental flasks were mixed
for 7 minutes once an hour by teflon-coated magnetic stirring bars.
Cyanophage suspensions were obtained by lysis of a host algal culture
and filtered (Millipore 0.22 um). The suspensions were stored refrigerated
and in the dark to prevent losses in viability (Padan and Shilo 1973)- The
titer of the cyanophage suspension was assayed prior to every experimental
series. The requirement of the LPP-1 cyanophage for 10 mM MgC12 in the
medium was fulfilled in all experimental cultures.
The experiments were run as batch algal cultures with S00 ml of nutrient
solution in 1-liter Erlenmeyer flasks. The flasks with nutrient solution
were autoclaved with two Pasteur pipets inserted through the cotton plug and
with tygon tubing attached to one of the pipets for sampling (cf. Fig. 1*8).
Sampling was performed by inserting sterilized Pasteur pipets into the tubing
and removing culture suspension by suction immediately after mixing. No
attempts were made to keep the cultures completely free from bacteria.
Depending on the design of the experiment, the cultures were incubated for
8-25 days. Due to limited experimental culture volumes and the time factor
in viral replication, it was necessary to set priorities in the analytical
program to give maximum information in a limited number of experiments.
Description of a Culture Unit
A schematic diagram of one culture unit is shown in Figure 48. In each
experimental series six or seven culture flasks with separate pH meters and
valves for C02/air (5:95. v/v) were connected in parallel to a timer and the
C02 air tank. The pH meters were operating constantly during the experiment
and functioned to hold pH levels below preset values. The timer activated
stirring of the culture solutions and opened the valves for C02/air bubbling
seven minutes per hour. The valves were secondarily controlled by pH in the
cultures, and gas was provided if pH exceeded the value set on the pH meter
for a particular culture. As soon as pH levels were corrected the valves
were closed. The tubing for gas supply was transparent tygon and glass when
possible. Due to a pressure decrease from loss of gas by diffusion through
the walls of the tubing when no gas was required for a longer time (ca. 5~6
hrs), culture solution was sucked up into the inlet pipet and occasionally
130
-------
Table 17: Composition and concentration of nutrient solution*
Compound	mg/1	mM
KC1
34
0.46
Na^iO^- 9^0
58
0.20
Na2C03-10H20
54
0.19
MgS04
75
0.62
CaCl 2
36
0.48
NaN03
496
5.80
kh2po41^
4.7
0.032
FeCl3-6H20
2.8
0.01
EDTA
3.7

Trace solution (Hoagland's A-Z)
* modified from Hughes, Gorham and Zehnder (1958)
^concentrations modified in some experiments
131
-------
valve	inlet CO2 + air
sampling
cotton
filter
cotton plug
PH
controller
tank
electrode
timer
magnetic
stirrer
Fig. 48. Schematic diagram of one experimental unit.
132
-------
into the tubing. Various kinds of tubing were tested but none worked
satisfactorily. Therefore, either the gas supply mechanism was interrupted
during the night or the outlets (pipet tips) were raised above the surface of
the culture suspension. In the latter case the gas supplied was not always
enough for complete correction of pH increases in the cultures.
Analytical Methods
Physical and chemical analyses
pH was recorded every second to third hour during daytime. P04-P was
analyzed as soluble reactive phosphorus according to Murphy and Riley 09&2)
and by the stannous chloride method. NH^-N was determined according to
Chaney and Marbach (1962), and N03~N was analyzed as N02-N (Bendschneider and
Robinson 1952) after reduction by a copperi2ed Cd column. For chlorophyll a,
5-25 ml of algal suspension was filtered under vacuum on a 2.5 cm glass fiber
filter (Whatman GF/C) and washed with a small volume of distilled water. The
filter was ground in 90% acetone, and chlorophyll was extracted for 12 hrs in
the dark at U-6 C. Estimation of chlorophyll a followed the method of
Lorenzen (19&7)•
Cyanophage assays
For plaque assays, 0.5 ml portions of diluted viral suspension were
mixed with 2.0 ml of a dense algal host culture and 2.5 ml of melted nutrient
agar (nutrient solution in 1% agar) and plated on Petri dishes over a base of
15 ml solidified nutrient agar (nutrient solution in 1.5% agar) according to
the double-agar layer technique (Safferman 1973a)- The plates were inverted
and incubated under continuous illumination (1500 lux) at 22+ 1 degree C, and
the number of cyanophage particles was estimated by counting clear plaques on
the algal "lawns" after k-$ days incubation. Appropriate cyanophage titer
for assay was obtained through tenfold dilutions in a saline-magnesium
solution (0.2 g MgC12. 6H20 and 5-85 g NaCl per liter distilled water). Each
plaque developed on the plate was considered to originate from one
cyanophage, and the titer of the assayed suspension (plaque-forming units
(PFU)/ml) was evaluated using the proper dilution factor.
EXPERIMENTAL DESIGN AND RESULTS
The experiments were performed in four successive series with six or
seven culture flasks in each series. Results from one series aided in
determining the next. Since the addition of C02/air to reduce pH did not
show any differences in the nature or rate of lysis of P1ectonema cultures
when compared to the use of HC1 in a preliminary study, addition of C02/air
was selected for all experiments to avoid changes in alkalinity (Stumm and
Morgan 1970, Brewer and Goldman 1976) .
Series 1
This experiment was done to determine whether pH has an effect on the
a 1ga-cyanophage relationship. P1ectonema was grown in the nutrient solution
(at 1 mg P04-P/1) for 11 days with no pH adjustment. The cyanophage was then
inoculated, and pH adjustment with C02/air was begun (Fig. ^9)- Controls
received no C02/air.
133
-------
12 n
i- -
11 -
10 -
8 -
7
i
-i
I
W
5-;

0	1	2	3	4	5	6
Days
Fig. 49. pH in cultures of Plectonema infected with cyanophage LPP-1
(	, 1-4) and cyanophage-free (—, 5 and 6). Arrow
(day 0) indicates inoculation with cyanophage and start of
C02/air addition (1-3 and 5) to 11-day old cultures.
134
-------
Culture
no.
pH adjusted with
CQ2/air to:
LPP-1 added
(PFU/ml)
3
i.
2
5-5
6.5
7-5
100,000
100,000
100,000
100,000
5
6
control
5-5
control
0
0
When pH was lowered from 11.5 to about 7*5 (Fig. ^9) in cultures of
P1ectonema simultaneously inoculated with cyanophage, replication of the
cyanophage was strongly enhanced (Fig. 50). In the culture where pH was
lowered to 5-5. replication was delayed. After three days of incubation,
however, the cyanophage concentration was similar to that in the other
cultures with pH adjustment. In the culture where pH was not adjusted (Fig.
49, culture k) on the other hand, replication of the cyanophage declined
early and ceased at a comparably lower density (Fig. 50).
The response of PIectonema to the presence of the cyanophage was obvious
after only 2k hours (Fig. 5D in cultures where C02/air was supplied and the
pH lowered, while algae in the control culture (without C02/air addition)
maintained a healthy population despite an increase in cyanophage particles
of several orders of magnitude (Fig. 51> cf. Fig. 50).
PIectonema cultures free from cyanophage showed the same healthy growth
irrespective of the change in pH (Fig. 5D> The only difference noted
between the two cyanophage-free cultures was that the algal filaments became
short (5-30 cells) and straight in the culture with adjusted pH, while
filaments in the control culture were long (more than 30 cells) and twisted
into bundles. According to Padan and Shilo (1973)» however, the length of
the filaments has no effect on the adsorption rate of cyanophage to
P1ectonema or the reproductive cycle of the cyanophage.
Two days after the viral infection of PIectonema, only one- and two-cell
trichomes were left in the cultures with adjusted pH, but long (more than 20
cells) and bright green trichomes were present in the control cultures. The
only symptom of viral infection in the control was that the cells swelled
somewhat and were thus abnormally distinct in the filaments. After four days
there were no detectable PIectonema cells in cultures with adjusted pH and
cyanophage. However, 17 days after the infection (adjustment of pH stopped
on day 6) some long, fine filaments were observed in the lysed cultures
indicating regrowth of a "new" PIectonema population. pH had by then
increased to about 9. This behavior can be compared with the repeated
oscillations in P1ectonema density and cyanophage LPP-1 concentration
observed in continuous cultures (Cowlishaw and Mrsa 1975, Cannon et. al.
When the addition of C02/air was interrupted during the night, pH
increased to the same levels in cultures with cyanophage as in
cyanophage-free cultures (Fig. 49). In control cultures (no C02/air
addition) the presence of the cyanophage did not affect pH conditions during
the experiment (Fig.	cultures k and 6). According to Padan et al¦
(1970), cyanophage infection begins to inhibit photoassimilation in
PIectonema cultures early in the infective process. Wu and Shugarman (1967).
however, stated that the rate of photosynthetic oxygen evolution does not
change until the infected cells rupture.
Thus, the present results show clearly that although the alga can live
at low pH, and the alga and cyanophage can coexist at high pH, lysis occurs
when the pH is lowered in an infected culture.
1976).
135
-------
Cyanophage LPP-1
series 1
RFU/ml
1010-
inoculum, 105 PFU/ml
5.5
-•*-
-a- as
-O- 7.5
control
0
3
4
5
1
2
Days
Fig. 50. Replication of cyanophage LPP-1 after
inoculation on day 0 in 11-day old cultures of
Plectonema. Control ( • ) pH adjusted a, o ) .
Culture numbers (1-4) indicated.
136
-------
Plectonema boryanum
hours
Reproduced from
best available copy.
0 6 9
• • •
• • •
pH	5.5 6.5 7.5 control 5.5 control
cyanophage LPP-1 LPP-1 LPP-1 LPP-1 — —
Fig. 51. Appearance of Plectonema cultures (on 0.22 ym
Millipore membrane filters) growing under controlled and
drifting pH conditions with and without inoculated cyano-
phages (LPP). Hour 0 indicates cyanophage inoculation and
addition of C0?/air to varying pH in the 11 days old
cultures (Series 1).
72
100
137
-------
Phosphate concentrations were below 10 ug P0U-P/1 in all systems
throughout the experiment (i.e. for 6 days after cyanophage inoculation).
Series 2
The purpose of this experiment was to repeat parts of series 1 using a
more dilute algal culture and to determine whether continuous pH adjustment
is necessary to cause lysis of the host alga. In addition, a more dilute
medium was used. Fewer cyanophage particles were added to give a number of
phage particles per algal cell similar to that in series 1.
P1ectonema cells (35 ml of a 12-day old stock culture) were inoculated
into a half-strength (0.5 mg P04-P/1) nutrient solution (Table 17)- This was
stirred 3 h f°r pH stabilization before cyanophage LPP-1 was inoculated and
C02/air addition begun (Fig. 52). In two cultures C02/air injections were
performed only one and three times, respectively.
Culture	pH adjusted with	LPP-1 added
no.	C02/air to:	(PFU/ml)
1	6	2000
2	7	2000
3	6 (3 times)	2000
k	6 (once)	2000
5	control	2000
6	6	0
7	control	0
In this series the fastest replication of cyanophage occurred in the
culture where no C02/air was added (Fig. 53. control culture). Because of
the dilution of the alga? inoculum, however, pH values in the cultures were
more than 2 pH units lower than in the first series at the time the
cyanophage was inoculated (Fig. 52, cf. Fig. i»9) • Furthermore, pH was
controlled by cyanophage activity after one day of incubation, since
photosynthetic activity was affected (Fig. 52). In the cyanophage-free
control culture pH increased as a function of algal growth (Fig. 52, cf. Fig.
5W •
In the cultures with pH adjustment cyanophage replication was delayed,
but it later progressed at a similar rate in all cultures (Fig. 53)- As in
the earlier experiment (series 1), the "most acid" culture (pH ca 6)
contained a lower concentration of cyanophage during the first two days of
viral infection than did the cultures where pH was kept around 7 (Fig- 53)-
Chlorophyll a development in the cultures was inversely related to
cyanophage replication (Figs. 53 and 5M • The rapid lysis of P1ectonema cells
in the control culture (no C02/air addition) directly reduced the uptake rate
of P04-P (Fig. 55, culture 5)• Cultures with one or three injections of
C02/air were able to absorb more P0^-P before being affected by cyanophage
activity (Fig. 55> cultures 3 artd k) . In the cultures with pH adjustment and
cyanophage, only small amounts of P04-P were left in solution, and only a
slight increase in POh could be seen on about the seventh day of the
infection period (Fig. 55)• The concentration of phosphate was initially 500
ug P/l; however, three hours after addition of algal cells, i.e. at the time
for cyanophage inoculation, only 200 ug P0^-P/1 remained in solution. The
cyanophage-free cultures depleted PO^-P from the nutrient solution within
three days (Fig. 55)•
Two days following cyanophage inoculation NH4-N appeared in solution in
some cultures (Fig. 55)* Since the culture which lysed first showed the
138
-------
12-|
series 2
cyanophage-free
0	1	2	3	4	5	6
Days
12-
I
series 2
11 -
cyanophage -free
1
U
1
T
5
Days
Fig. 52. pH in cultures of Plectonema infected with
cyanophage LPP-1 (1-3 and 5) and cyanophage-free
(6 and 7). Arrow (day 0) indicates inoculation with
cyanophage and start of CC^/air addition (1-3 and 6).
Culture numbers are indicated.
139
-------
Cyanophage LPP-1
series 2
PFU/ml
>10_
P 2
106-
-f— 6
-o- 7
-a- 6 (3)
6 (D
control
3
0
2
1
Chlorophyll a
.7
'n6
' PH
control
-n- 6
100-
50 -
0
2
4
6
Days	Days
Fig. 53. (left). Replication of cyanophage LPP-1 in Plectonema cultures
after inoculation on day 0. Control (•) , pH adjusted (to value shown)
(, O / A ,~ ) . Culture numbers (1-5) shown. (1) and (3) indicate 1 and
3 injections of CC^/air.
Fig. 54. (right). Chlorophyll iL in cultures of Plectonema infected with
cyanophage LPP-1 { q.' O A / * r ®)and cyanophage-free (~,¦') pH condi-
tions shown here and in Fig. 53. Culture numbers (1-7) indicated.
-------
lig/i
-
-------
earliest indication of NH4-N, this was probably connected
activity. No NH4-N was found in uninfected cultures (Fig.
with frequent input of C02/air showed a rapid increase to high
of NH4-N, most probably due to algal lysis. The nitrogen
nutrient solution was N03. and the N:P ratio (by weight) was 80
The replication rate of cyanophage in the cultures was not
making the C02/air injections once, three times or frequently,
to P04-P and NH4-N release following lysis, on the other hand,
of C02/air addition was critical (cf. Fig. 55)-
to cyanophage
55). Cultures
concentrations
source in the
: 1.
affected by
With respect
the frequency
Series 3
Because of the differences in results between series 1 and 2, i.e. the
virus replicated more at ambient pH in series 2, the algal cultures used in
this series were again older and more concentrated than in series 2.
P1ectonema (10 ml inoculum) was allowed to grow in a half-strength
nutrient solution modified to 5 mg N03-N/1 and 0.35 fg P04-P/1 for k.S days
before cyanophage was inoculated and C02/air addition was started (Fig. 56).
In this series 1 ml of 12-day old stock cultures of Scenedesmus or Anabaena
was added simultaneously with the cyanophage
LPP-1 added
(PFU/ml)
inoculation.
cu1ture
no.
pH adjusted wi th
C02/air to:
1
2
3
4
5
6
7
6
6
6
6
control
control
control
10,000
10,000
10,000
0
10,000
10,000
10,000
Addi tional alga
Scenedesmus
Anabaena
Scenedesmus
Anabaena
The replication of cyanophage was much slower in these cultures than
when the host algae had approached a stationary phase of growth (Fig. 57» cf.
Fig. 50) or when a surplus of inorganic nutrients was present in the culture
(cf. Fig. 53). As in series 1 where no external nutrients were present, a
faster replication of cyanophage LPP-1 occurred in the cultures with pH
adjustment. In addition, pH levels in the cultures were similar to those in
series 1 at the time of cyanophage inoculation (Fig. 56, cf. Fig. k$),
Anabaena or Scenedesmus added to the P1ectonema culture simultaneously
with cyanophage in this series did not significantly affect the cyanophage
replication pattern (Fig. 57)- Nor did chlorophyll a concentrations in the
cultures show any differences during the first three days that could be
related to algal species composition (Fig. 58). The main differences were
due to the C02/air addition and the related variation in cyanophage activity.
Neither P04-P nor NH4-N could be detected during the first three days of
infection. On the eighth day, however, 210 ug/1 and 190 ug/1 of NH^-N were
found in the cultures with pH adjustment containing PIectonema and P1ectonema
+ Anabaena, respectively. On the other hand, no NH^-N was found in the
culture with pH adjustment containing P1ectonema and Scenedesmus or in any of
the cultures where pH was high. N03"N and N02-N were below 25 ug/1 and 10
ugN/1, respectively, in all cultures. Initial input of N03~N and P04-P was
much lower in this series, i.e. 5 mg N/1 and 0.35 mg P/l.
After two days of incubation with cyanophage, Plectonema consisted of
short (3~10 cells) filaments in the cultures with pH adjustment; Scenedesmus
looked healthy, while only a small number of Anabaena cells remained viable.
In the control cultures (no C02/air addition) the Plectonema filaments after
142
-------
12 n
PH
series 3
11 -
10 h
9-
max
mean
min
8
7-
6-
-//-
10
12
Days
Fig. 56. pH in cultures of Plectonema before and after inoculation
with cyanophage LPP-1 and Scenedesmus or Anabaena. Day 0 (upper scale)
indicates inoculation of fresh nutrient solution with PIectonema cells
and day 0 (lower scale) indicates start of CC^/air addition (4 cultures)
and inoculation with cyanophage (6 cultures). Dots represent mean
values and vertical bars the range.
143
-------
PFU/ml
Cyanophage LPP-1
series 3
10'
— O —
P+S
P+A
A —
10 -
2
3
0
1
4
5
Days
Fig. 57. Replication of cyanophage LPP-1 in
algal cultures with pH adjustment (open symbols,
pH ca. 6) and in controls (closed symbols).
Plectonema + Scenedesmus (P+S) and Plectonema
+ Anabaena (P + A). LPP-1 inoculumn was l(r
PFU/ml. Culture numbers (1-7) indicated.
144
-------
150- Chlorophyll a
100-
P+S
P+A
2
4
5
0
3
1
Days
Fig. 58. Chlorophyll in cultures of PIecto-
nema (P), Plectonema + Scenedesmus (P+S) and
PIectonema + Anabaena (P+A) infected with cyano-
phage LPP-1 and grown with pH adjustment (open
symbols) or as controls (closed symbols). Scene-
desmus and Anabaena were inoculated simultaneously
with the cyanophage (day 0) in 4.5-day old
Plectonema cultures (series 3).
145
-------
two days were long (more than 30 cells) and twisted into bundles when no
additional algal species were present, and long, but not twisted, when
Scenedesmus or Anabaena was present. After four days of incubation, when
P1ectonema in the control cultures consisted of 1~3 cell filaments and lysis
was in progress, Anabaena in particular looked very healthy. Scenedesmus was
more abundant in the culture with pH adjustment than in the control culture.
pH in the control cultures decreased after 3 days of incubation (Fig. 56).
Small amounts of bacteria were found in the lysed cultures after eight days
of incubation.
Series k
Since the age of the algal host culture seemed to affect the rate of
cyanophage replication, identical nutrient solutions inoculated with
PIectonema cells were allowed to reach different growth phases (considered as
host age), and different densities and pH (Fig. 59) before cyanophage LPP-1
was inoculated.
P1ectonema (10 ml inoculum) was allowed to grow in a ha 1f-strength
nutrient solution modified to 3 mg N03~N/1 and 0.3 mg PO^-P/l for 1.5. 6 and
11 days before cyanophage inoculation and the start of C02/air addition (Fig.
59) .
Culture	pH adjusted with	LPP-1 added Algal host culture
no.	C02/air to:	(PFU/ml)	age (days)
1	7	1000	1.5
2	control	1000	1 .5
3	7	1000	6
k	7 (once)	1000	6
5	7	iooo	n
6	8 (once)	1000	11
As found in earlier experiments, the cyanophage attacking algae in the
exponential phase of growth showed the slowest rate of replication (Fig. 60).
Again, algae in the lag phase of growth induced a rapid increase in
cyanophage, indicating that variation in algal density was less important for
viral replication than the physiological state of the host. The chlorophyll
a concentration was 2.6 and J,.2 times higher after 6 and 11 days of initial
incubation when compared with the value after 1.5 days.
The difference in cyanophage replication between one or frequent
additions of C02/air was much smaller than that caused by algal host age
(Fig. 61). Again, the infection of the "youngest" PIectonema culture
resulted in a release of NH^-N (250 ug/1 on day 10 and 510 ug/1 on day 15) in
the culture with pH adjustment. P04-P was below 5 ug/1 up to day 10 in all
cultures, but on day 15 it ranged between 10-21 ug PO^-P/I with no relation
to host culture age or CQ2/air input.
DISCUSSION
Since Safferman and Morris (1963) isolated the first cyanophage,
designated LPP since it lyses species of the blue-green algal genera Lynqbya,
Phormidium and PIectonema, additional cyanophages have been discovered and
investigated (cf. reviews by Padan and Shilo 1973. Stewart and Daft 1977)-
The main emphasis has been on cyanophage characteristics, growth cycles and
interactions with host algal metabolism, mostly under standard photosynthetic
conditions (Wu and Shugarman 1967. Padan, £t aj_. 1970. Padan and Shilo 1973.
146
-------
+ LPP-1, 4C02/aJr
1 series 4
10-
24
20

t
8
12
14
!
T
n
4
0
2
6
10
Days of incubation
Fig. 59. pH in cultures of Plectonema before and after inoculation
with cyanophage LPP-1 and start of CC^/air addition (day 0, lower scale)
Algae were grown for 1.5, 6 and 11 days before the cyanophage was added.
Cultures with pH adjusted to ca. 7 are omitted in the diagram. Culture
numbers (1-6) are indicated.
147
-------
10® 1 PFU/ml
—O"
10®-
Cyanophage LPP-1
series 4
10®-
-a- 1.5 days
1.5 »
-fi- 6	M
—o—
10 -I
6
8
14
10
2
4
0
Days
Fig. 60. Replication of cyanophage LPP-1 in
cultures of Plectonema grown for 1.5, 6 and 11
days before inoculation with the cyanophage.
Open symbols indicate cultures with frequent
addition of C02/air (pH ca. 7), and closed sym-
bols indicate no (¦)' and one ( • , ~) injection
of C^/air.
148
-------
PFU/ml	Series 4
11 da:
7 days
4 days i
0W ; CO2 /air
W 3 days | addition:
LPP
inoc.
•one injection
o frequent
onone
21 hrs
10 -
12
14
10
6
8
0
2
4
Culture age. days
Fig. 61. Concentration of cyanophage LPP-1 in
relation to Plectonema culture host age and time
following cyanophage inoculation. Shaded areas
indicate difference between cultures receiving
no, one and frequent injections of CC^/air.
149
-------
Stewart and Daft 1977). It is clear that viral replication within the algal
host is regulated largely by host metabolism, and that viral infection in
turn affects the physiological state of the host alga.
Interpretation of the results from this study will be centered on how a)
input of C02/air and accompanying changes in pH, b) inorganic nutrients, and
c) physiological state of the host alga (host culture age) affected the
replication of the cyanophage LPP-1 in P1ectonema boryanum cultures. As the
conditions simulated in the laboratory were to a large extent derived from
treatments used in plastic bag experiments (Shapiro 1973. 1975). the
laboratory results are compared with the outcome of field experiments. Only
those cyanophage - alga interactions that are more or less directly related
to the areas defined above will be considered.
Dependence on pH and C02 Input
Since growth of blue-green algae frequently occurs under high pH
conditions (King 1970). and collapse can be induced by a lowering of pH
(Shapiro 1973). it was interesting to find that P1ectonema continued to grow
at high pH levels where lysis was avoided, even though the cyanophage
concentration initially increased. On the other hand, the addition of
C02/air to a culture containing the cyanophage induced a rapid lysis of the
PIectonema population despite the fact that the alga could grow well at low
pH in the absence of the cyanophage.
LPP cyanophage activity is reported to be stable from pH 5 to 11 (pH
adjusted with HC1 or NaOH in filtered viral suspensions) by Safferman and
Morris (1964). The same authors reported that PIectonema develops readily at
pH 7-11, whereas no, or poor growth occurs below pH 7 (cf. Padan and Shilo
1973)• In my experiments the growth of Plectonema was not particularly
affected when pH was lowered by addition of C02/air to as low as 5-5 or
allowed to rise to nearly 12. Instead, high (greater than 11) and very low
(5.5-6) pH values seriously affected cyanophage replication, and the reduced
or delayed infection of the algal cells could favour algal growth and
development of chlorophyll a (cf. Fig. 51 and 5M • Between pH 6 and 11 there
were no significant variations in cyanophage replication patterns that could
be related to pH conditions solely. According to Padan et aj_. (1970).
replication of cyanophage LPP-1G was identical in the presence or absence of
C02.
In lysed cultures with no pH adjustment, pH stabilized between 8 and 9-
Addition of Na2C03 to lysed cultures to raise the pH accelerated a regrowth
of PIectonema, while addition of nutrient solution had no effect on regrowth.
Attempts to isolate P1ectonema from sites where LPP cyanophages were found
have proved fruitless (Cannon et, aj_. 197^)-
In field experiments (plastic bags) the induced collapse of the
blue-green algal populations and the shift in species dominance occurred at
pH values as high as 8.5 (Shapiro 1975)- Furthermore, a reversal of the shift
(to blue-greens) required re-inocu1 at ion of blue-green algae as well as
increase in pH (Shapiro op. cit.) .
LPP cyanophages naturally occur mostly in polluted environments, e.g.
waste stabilization ponds, fish ponds during periods of fish kill, polluted
rivers and rice fields (cf. Safferman and Morris 19&7. Shane et aj_. 1972,
Singh 1973, Cannon et aj_. 197^)- Host algae have generally not been observed
in samples that showed a presence of LPP virus (Safferman and Morris op.
cit.). For example, along the polluted Christina River (in Delaware,
U.S.A.), host algae were found only occasionally in samples where LPP
cyanophages were present. A virus-resistant Plectonema laboratory culture
150
-------
grew well in the river water. However, the river pH was reported to be
relatively stable, varying between 6.6 and 7.8, and concentrations of
inorganic N and P0*t-P were high (Shane et aj_. 1972). Similar observations
were made in a pond close to Minneapolis (Minnesota, U.S.A.) (Lindmark
unpubl.). Survival and tolerance of the LPP cyanophages apparently enable
them to control their algal hosts in nature and thus eliminate the
development of blooms.
Other cyanophages discovered later have been found to possess somewhat
different requirements, e.g. the SM-1 cyanophage (attacking strains of
Synechococcus) requires C02 for replication (cf. Stewart and Daft 1977) -
Nevertheless, additional cyanophages need to be isolated and characterized.
Just as there are specific bacteriophages for a large number of bacteria, it
seems possible that there is a cyanophage for each blue-green alga.
Dependence on Inorganic Nutrients
Whether the addition of nutrients or the associated decline in pH (from
11.5 to ca. 9) favored cyanophage replication in the control culture (no
addition of C02/air) in series 2 as shown in Figure 53, is unclear. The
inoculation of cyanophage in relation to algal density was about one third of
that used in experiments where no nutrients were added before cyanophage
inoculation. In spite of this, the amount of cyanophage (PFU) released after
2k hours was considerably higher than in series 1 (cf. Fig. 50 and 53)* The
addition of C02/air in the presence of nutrients, however, resulted in a
short delay in viral replication that for a short time favored the algae (cf.
Figs. 53 and 5^0 .
These results partly agree with observations from field experiments
where a combined reduction in pH and addition of nutrients (N and P)
accelerated the blue-green algal collapse (Shapiro 1973)- In the laboratory
systems the pH was lowered by the addition of nutrient solution.
From the analysis of NH4-N and P04-P concentrations following lysis of
Plectonema, it seems that the mode and amount of C02/air input are important
(Fig. 55) • 'r> the field the decline of blue-green algae is often accompanied
by release of PO^-P (Shapiro et aj_- '975) •
When cells of Scenedesmus were present during lysis of Plectonema. no
NH4-N was detected at low pH; it was most likely utilized by Scenedesmus¦ In
cultures with Anabaena and without additional algal species, the released
NH4-N remained in solution. It could not be determined whether low pH
solely, or in combination with other factors e.g. bacteria or lysing agents
(Gunnison and Alexander 1975) following lysis of PIectonema, was the reason
that Anabaena failed to survive. However, at high pH Anabaena remained
v i ab1e.
These findings may be useful in experiments in natural ecosystems, e.g.
artificial circulation of whole lakes, with respect to nutrient requirements
and competition between blue-green algae (N2 fixers in particular) and green
algae.
Dependence on Age of Host Algal Culture
It is evident that P1ectonema in the exponential growth phase showed the
strongest "physiological resistance" to lysis (cf. Fig. 61). In the k.S and
6-day old culture of Plectonema in series 3 and k, respectively, pH was
around 11 at the time for cyanophage inoculation. The addition of C02/air to
the cultures stimulated cyanophage replication in both series. However, the
resistance of the algae to lysis due to culture age was more pronounced than
151
-------
that due to the varying pH adjustments in series k (cf. Fig. 57 and 61). In
quasi-continuous cultures Cowlishaw and Mrsa (1975) found that up to 0.3$ of
the algal cells survived the first round of lysis and suggested that this
could not be due solely to mutation. Safferman and Morris (19&^) reported
that seven days were necessary to achieve lysis in a four-day culture of
PIectonema.
Higher resistance to viral attack during the exponential phase of growth
can be compared to the usual pattern of development of blue-green algae. In
Lake Erken (Sweden) Aphanizomenon fIos-aquae steadily increased during summer
before a viral infection terminated the bloom in September (Granhall 1972).
Physical and chemical environmental factors (no pH values reported) varied
little during the critical period; however, the abundance of Aphanizomenon
was extremely low the following summer.
CONCLUSIONS
Although several interactions in the cyanophage-alga system studied
paralleled observed events in field experiments with similar external
stresses and variations, the data are far from sufficient to allow definite
conclusions. On the other hand, the results are encouraging, and more
experiments along similar lines should be conducted. One of the difficulties
with the work reported here is that, because of time restrictions, it was not
possible to repeat each experiment enough to establish strict relationships.
For example, if the results of series 1 could be validated by sufficient
repetition, then the effects of small environmental changes could be more
readily evaluated. In this manner it might be possible to determine, for
example, whether artificial circulation of a lake, with subsequent lowering
of pH and elevation of nutrient concentrations, would be more likely to cause
a shift from blue-green algae to green algae at one time in the history of an
algal bloom than at an earlier or later point in that history. Such
information would be of immense practical value and would be a step toward an
ecological approach to lake restoration (Shapiro 1978).
The use of cyanophages is an attractive method for biological control,
in that it is catalytic and avoids poisoning the general environment. The
strong specificity of the cyanophages could provide selective control of
blue-green algae, although development of resistant algal clones might be
expected after large viral infections.
The need to understand the cause of algal blooms -- both biologically
and chemically — remains (Walsby 1970, Reynolds and Walsby 1975) -
ACKNOWLEDGEMENTS
The work reported above was carried out in the laboratories of the
Limnological Research Center, University of Minnesota, and was supported by
research grants to Dr. J. Shapiro from the National Science Foundation, the
Office of Water Resources Research and the Environmental Protection Agency.
Additional grants to G. Lindmark were made by the American-Scandinavian
Foundation and the Fulbright Commission.
152
-------
REFERENCES
Bendschneider, R. and R.J. Robinson. 1952. A new spectrophotometric
determination of nitrite in sea water. J. Mar. Res. 11:87-96.
Bjork, S., L. Bengtsson, H. Berggren, G. Cronberg, G. Digerfeldt, S.
Fleischer, C. Gelin, G. Lindmark, N. Maimer, F. Plejmark, W. Ripl, and
P.O. Swanberg. 1972. Ecosystem studies in connection with the
restoration of lakes. Verh. Int. Verein. Limnol. 18:379~387•
Brewer, P.G. and J.C. Goldman. 1976. Alkalinity changes generated by
phytoplankton growth. Limnol. Oceanogr. 21:108—117•
Cannon, R. 1975- Field and ecological studies on blue-green algal viruses.
Proc. Water Quality Management through Biological Control, Jan. 23-30,
1975. Univ. of Florida, Gainesville.
Cannon, R.E., M.S. Shane, and E. DeMichele. 197^• Ecology of blue-green
algal viruses. J. Env. Eng. Div. EE6 pp. 1205-1211.
Cannon, R., M.S. Shane, and J.M. Whitaker. 1976. Interaction of PIectonema
boryanum (Cyanophyceae) and the LPP-cyanophages in continuous cultures.
J. Phycol. 12:418-421.
Chaney, A.L. and E.P. Marbach. 1962. Modified reagents for determination of
urea and ammonia. Clin. Chem 8(2):130-132.
Cowlishaw, J. and M. Mrsa. 1975- Co-evolution of a virus-alga system.
Appl. Microbiol. 29:234-239-
Daft, M.J., S.B. McCord, and W.D.P. Stewart. 1975• Ecological studies on
algal-lysing bacteria in fresh waters. Freshwat. Biol. 5'577-596.
Goryushin, V.A. 1976. Interaction between viruses and b!ue-green algae.
Abstract from SIL Symp. Experimental Use of Algal Cultures in Limnology.
Oct. 26-28, 1976, Norway.
Granhall, U. 1972. Aphan i zomenon flos-aquae: infection by cyanophages.
Physiol. Plant. 26:332-337-
Gunnison, D. and M. Alexander. 1975- Resistance and susceptibility of algae
to decomposition by natural microbial communities. Limnol. Oceanogr.
20:64-70.
Hughes, E.O., P.R. Gorham. and A. Zehnder. 1958. Toxicity of a unialgal
culture of Mi crocyst is aeruq i nosa. Can. J. Microbiol. 4:225~236.
Jackson, D.F. and V. Sladecek. 1970. Algal viruses — eutrophication
control potential. Yale Sci. Mag. 44:16-22.
King, D.L. 1970. The role of carbon in eutrophication. J. Water Poll.
Control Fed. 42:2035-2051.
Loren2en, C.J. 1967- Determination of chlorophyll and pheo-pigments:
spectrophotometric equations. Limnol. Oceanogr. 12:343-346.
Murphy, J. and J.P. Riley. 1962. A modified single solution method for
determination of phosphate in natural waters. Anal. Chim. Acta
27:31-36.
Padan, E., D. Ginzburg, and M. Shilo. 1970. The reproductive cycle of
cyanophage LPP1-G in PIectonema boryanum and Its dependance on
Photosynthetic and respiratory systems. Virology 40:514-521.
Padan, E. and M. Shilo. 1973- Cyanophages -- viruses attacking blue-green
algae. Bacteriol. Rev. 37:343"370.
Reynolds, C.S. and A.E. Walsby. 1975- Water-blooms. Biol. Rev. 50:437~48l.
Rhee, G.-Yull. 1972. Competition between an alga and an aquatic bacterium
for phosphate. Limnol. Oceanogr. 17:505"514.
153
-------
Ryding, S.-O. and C. Forsberg. 1976. Six polluted lakes: a preliminary
evaluation of the treatment and recovery processes. Ambio 5 W '• 151 ~ 156.
Safferman, R.S. 1973a- Special methods -- virus detection in Cyanophyceae.
i n Stein, J., (ed), Handbook of Phycological Methods: Culture Methods
and Growth Measurements. Cambridge Univ. Press.
Safferman, R.S. 1973b. Phycoviruses. In Carr, N.G. and B.A. Whitton
(eds.): The Biology of Blue-Green Algae. Univ. Calif. Press, Berkeley
and Los Angeles, Calif.
Safferman, R.S. and M.E. Morris. 1963* Algal virus: isolation. Science
140:679-680.
Safferman, R.S. and M.E. Morris. 1964. Control of algae with viruses. J.
Am. Water Works Assoc. 56:1217-1224.
Safferman, R.S. and M.E. Morris. 1967- Observations on the occurrence,
distribution and seasonal incidence of blue-green algal viruses. Appl.
Microbiol. 15:1219-1222.
Shane, M.S., R.E. Cannon, and E. DeMichele. 1972. Pollution effects on
phycovirus and host alga ecology. J. Water Poll. Control Fed.
44:2294-2302.
Shapiro, J. 1973- blue-green algae: why they become dominant. Science
179:382-384.
Shapiro, J. 1975* Biomanipulation of lakes. Biennial Report from
Limnological Research Center, Univ. of Minnesota, Minneapolis.
Shapiro, J. 1977* Biomanipulation — A neglected approach? Plenary address
delivered at the 40th meeting of the American Society of Limnology and
Oceanography, Lansing, Michigan, June 1977-
Shapiro, J. 1978. The need for more biology in lake restoration. Contr.
No. I83 from the Limnological Research Center, Univ. of Minnesota,
Minneapolis. EPA 440/5~79~001.
Shapiro, J., V. Lamarra, and M. Lynch. 1975- Biomanipulation — an
ecological approach to lake restoration. Proc. Water Quality Management
through Biological Control, Jan. 23_30, 1975. Univ. of Florida,
Ga i nesv i11e.
Singh, P.K. 1973- Occurrence and distribution of cyanophages in ponds,
sewage and rice fields. Arch. Microbiol. 89:l69~172.
Stewart, J.R. and R.M. Brown. 1969- Cytophaga that kills or lyses algae.
Science 164:1523-1524.
Stewart, W.D.P. and M.J. Daft. 1977* Microbial pathogens of cyanophycean
blooms. In Droop, M.R. and H.W. Jannasch (Eds.): Advances in Aquatic
Microbiology. Academic Press, London.
Stumm, W. and J.J. Morgan. 1970. Aquatic Chemistry. An introduction
emphasizing chemical equilibria in natural waters. Wiley Interscience,
New York.
Walsby, a.E. 1970* The nuisance algae: curiosities in the biology of
planktonic blue-green algae. Water Treatm. and Exam. 19 (4) : 359** 37 3 -
Wu, J.H. and P.M. Shugarman. 1967- Effects of virus infection on the rate
of photosynthesis and respiration of a blue-green alga, P1ectonema
boryanum. Virology 32:166-167-
154
-------
III. DIRECT MANIPULATION OF ZOOPLANKTON POPULATIONS.*
The manipulation of herbivore abundance, as will be described later, can
be done in a variety of indirect ways. However, work by Fritsch (1953) and
Murphy (1970) indicated that a direct approach might be possible. Fritsch
had found that additions of yeast to the Ch1amydomonas he was feeding to
cultures of Daphnia pulex had the effect of increasing the lifespan of the
cladoceran. Investigating further, he was able to show that pantothenic
acid, a constituent of yeast, had the same effect, and could, at a
concentration of 200 ug/1, increase the life span of the Daphnia by
threefold. He also claimed that pantothenic acid increased egg production.
Murphy's results with D^ pulex were somewhat different. He found that
although the Daphnia would survive on a diet of Ch1amydomonas and
Scenedesmus. along with calcium acetate, bovine albumin and several
antibiotics, they produced infertile eggs. However, addition of calcium
pantothenate caused a very striking increase in the number of viable young,
with the optimum concentration ranging from 0.2 to 2.0 mg/1. He did not show
any effect on the life span, however.
Although both of these experimenters worked with 1ess-than-natura1
conditions, Fritsch with Chiamydomonas as the sole food, and Murphy with a
mixture of algae and certain supplements, it seemed possible that
panthothenic acid additions to a natural system might be one way to increase
the Daphnia population substantially. Accordingly, in July of 1976 we set our
four enclosures in Little Lake Johanna a small lake just north of St. Paul,
Minnesota. The enclosures were 1-meter diameter polyethylene bags sealed at
the bottom and open at the top and holding approximately 1350 liters of lake
water. The experiment ran from July 16 to September 9- Pantothenic acid was
added to the bags once on July 16, as a solution of calcium pantothenate, to
give final concentrations of 0.02, 2.0, and 20 mg/1 respectively. The
cladocerans were sampled periodically by vertical net tows with a plankton
net. Results are shown in Tables 18 and 19.
It is clear that neither the Daphnia nor the total cladocerans showed
increases as a result of the pantothenic acid additions. In fact, it almost
appears as though the additions reduced abundance of the animals. Our
results may indicate that pantothenic acid is not required by Daphnia under
natural conditions. Certainly its effects, if any, were insufficient to
warrant further study.
It should be mentioned that there are other chemical treatments that
directly reduce cladoceran abundance. For example, rotenone is known to kill
many of the microcrustaceans when applied in concentrations used to eliminate
fish (Hoffman and Olive, 1961; Kiser	aj_. 1963) (see section IV B for
effects in Wirth Lake). However, when used, as in Minnesota, in the fall,
its effects are probably gone by spring.
Pesticides also are very effective at eliminating, especially,
cladocerans. This is particularly true for the organophosphorus pesticides,
many of which have EC-50 values for Daphnia pulex of less than 1 ug/1
(Shapiro 1980). Pesticides may in fact be even more detrimental than this
would indicate. Gliwicz (personal communication) is finding that
concentrations of Lindane only about 1% as high as the EC-50 significantly
reduce fecundity of Daphnia¦
The possibility of adding zooplanktonic herbivores to a lake to increase
their abundance directly does not appear to be really worthwhile, except
perhaps under circumstances where all previously existing individuals have
been killed by some substance such as rotenone. This has in fact been done.
In 1950 the Canadian Wildlife Service treated a small lake in the Canadian
*by Joseph Shapiro
155
-------
Rockies with rotenone. One week later zooplankton were netted in nearby
Moraine Lake and transported in milk cans to the treated lake and dumped in.
Whether they survived is unknown. Under such circumstances the introduced
individuals could act as an innoculum, but under "normal" circumstances it is
doubtful whether one could add enough herbivores to affect the abundance
meani ngful1y.
Furthermore, there is good evidence that the presence of a given
zooplankter in a given natural system does not depend on whether or not the
organism has been put into the system, but depends on whether or not the
organism can survive in the system. An example is the chain of closed lakes
in the Grand Coulee of western Washington State (Shapiro, unpub.) . In this
system, Lake Lenore had long contained two species of D i aptomus but, probably
because of its greater salinity, nearby Soap Lake did not support members of
this genus. However in the 1950s Soap Lake gradually became less saline
because of leakage of gresh water into it from a large irrigation project.
Accordingly, in the late '50s large populations of Piaptomus appeared in Soap
Lake. Experiments in which Soap Lake water was concentrated an additional
10% showed that the organisms would not then live in it, substantiating the
fact that it was the dilution that allowed the crustaceans to survive.
Another example in the same area is Lake Washington, which for reasons
unknown has lately become inhabited by a sequence of species of Daphnia. All
of them had been present in surrounding waters and therefore had the
opportunity to become abundant in the lake, but hitherto had not done so
(Edmondson, 1979)•
in our own studies, Lake of the Isles (Section IV D) presents a good
example. Daphni a magna is not normally looked upon as a lake species, and
O'Brien (pers. comm.] feels that it cannot coexist with planktivorous fish.
In winter 197&-77 Lake of the Isles suffered a winter-kill, and in 1977
Daphnia magna was found on several occasions.
Thus it appears as though efforts to manipulate herbivore populations
directly are probably not likely to succeed. However, as will be shown
below, in the following sections, great success can be expected from
manipulations of the predators of herbivores.
REFERENCES
Edmondson, W.T. 1979- More on Daphn i a and D i aptomus in Lake Washington.
Paper presented at 42nd annual meeting of ASLO. Stony Brook. N.Y.
Fritsch, R.H. 1953* Die lebensdauer von Daphnia spec, bei verschiedener
Ernahrung, besonders bei Zugabe von Pantothensaure. Z. Wiss. Zool.
157:35-56.
Hoffman, D.A. and J.R. Olive. 19&1. The effects of rotenone and toxaphene
upon plankton of two Colorado reservoirs. Limnol. Oceanogr. 6:219-222.
Kiser, R.W., J.R. Donaldson, and P.R. Olson. 19&3- The effect of rotenone
on zooplankton populations in freshwater lakes. Trans. Amer. Fish. Soc.
92:17-24.
Murphy, J.S. 1970* A general method for the monoxenic cultivation of
Daphnidae. Biol. Bull. 139:321-332.
Shapiro, J. 1980. The importance of trophic-1evel interactions to the
abundance and species composition of algae in lakes. j_n Hypertrophic
Ecosystems. Mur, L. and J. Barica, (eds.), W. Junk Publishers. The
Hague. 330 PP-
156
-------
Table 18: Numbers of cladocerans* per net haul in enclosures treated with
calcium pantothenate at the concentrations indicated.
Sample dates
0, Control
0.02 mq/1
2.0 mg/1
20 mg/1
July 16
1755
1528
4028
1200
July 19
2236
3094
3692
3450
July 30
2822
2154
3128
1443
August 2
5334
2926
3959
1346
August 6
5576
2393
3249
2172
August 29
3275
4879
3198
3795
September 9
2641
3264
4687
7332
X 7/30 - 9/9
3929
3123
3646
3217
*C1 adocerans include Daphnia, Bosmina, Ceriodaphnia, Diaphanosoma,
Chydorus, and Scapholebris
Table 19: Numbers of Daphnia per net haul in enclosures treated with
calcium pantothenate at the concentrations indicated.
Sample dates
0, Control
ojD2 .ma/1
2.0 mg/1
20 mg/1
July 16
390
24
315
270
July 19
52
208
552
253
July 30
1054
329
1088
333
August 2
1155
152
294
137
August 6
1579
66
361
194
August 29
2095
1230
1784
1139
September 9
931
648
3397
2867
x 7/30-9/9
1362
485
1384
934
157
-------
IV. MANIPULATIONS OF PLANKTIVOROUS FISH -
EFFECTS ON ZOOPLANKTON AND PHYTOPLANKTON*
IV A. EXPERIMENTAL STUDIES
Our first studies of the effects on algae of manipulating populations of
planktivorous fish were done in enclosures in two small bodies of water
Lake Emily and Williams Pond -- both in St. Paul, Minnesota. These
experiments showed that the effects of the fish were dramatic (Shapiro £t al¦
1975)- We decided to extend the studies and examine the phenomenon in more
detail in Pleasant Pond. We also attempted an experiment in Loch Loso,
(Minneapolis) in which, after adding planktivores for a period, piscivores
were added to reverse their effects. The results of these studies are
presented below.
PLEASANT POND
Studies in this pond were done in 1975 and 197&. As two detailed papers
describing the results are published (Lynch 1979^-ynch and Shapiro 1981) what
follows is a summary of the material relevant to this report.
Pleasant Pond (about 0.25 hectares, mean depth 2.5 meters) is about 10
km north of St. Paul, Minnesota. The basin, excavated about 1965» has no
surface inlet or outlet and lies in a sandy outwash plain of glacial origin
surrounded by oak forest and secondary growth. Because of its shallowness
the pond normally winter-kills and was devoid of fish when our work began.
Three studies are described below — one in which fish were added to
enclosures, one in which fish and/or nutrients were added to enclosures, and
one in which the whole pond was divided, with fish added to one half.
Enclosure Studies with Planktivorous Fish
On June 15. 1975. 12 1-m diameter 0.8 m deep polyethelene enclosures,
closed at the bottom, were suspended in the pond and filled with surface
water. Ten were stocked with five different densities of bluegill sunfish
(Lepom i s macroch i rus) in duplicate, the remaining two serving as controls.
Two of the fish enclosures failed to survive the experiment and fish escaped
from two others. Thus results are available from only eight enclosures
(Table 20). Furthermore, as the single fish in one of the enclosures died
after five days, this enclosure was treated as another control. During the
experiment which lasted until July 25. zooplankters were sampled weekly by
vertical tows in the enclosures using a Wisconsin net with a 15 cm diameter
opening and 64 um netting. Average coefficients of variation for four
samplings on three occasions were, O.36 for cladocerans, 0.45 for nauplii,
0.59 f°r copepods, and 0.59 for rotifers. All samples were fixed immediately
with formalin/sucrose (Haney and Hall 1973)- Phytoplankton samples taken
from near the surface of the enclosures were fixed with Lugol's solution.
Counts were made with a Wild inverted microscope and were converted to cell
volumes using means of measurement of 10-50 individuals. Primary production
was measured by the C-14 method with incubation at a depth of 0.5 meters for
2k hours. All incubations were on cloudless days. All oxygen and
temperature measurements were made with a model 5*» -YSI meter. pH was
determined in the laboratory with a Beckman meter. Total phosphorus, soluble
reactive phosphorus, nitrate, nitrite, and alkalinity were determined
according to Standard Methods (APHA 1971)- Ammonia was determined as in
Chaney and Marbach (1962) .
>>by Michael Lynch and Joseph Shapiro
158
-------
Table 20: Bluegill sunfish added and recovered from enclosures. Enclosures
1 and 7 were controls. Since its one fish only lived for 5 days,
enclosure 2 is treated as a control.
Enclosure
Added - 18 Jun
Length
(mm)
Wt
isi
Recovered - 2 Aug
Length
(mm)
Wt
JrL
2
8
9
3
11
76
76
83
70
108
64
101
87
76
75
114
108
102
89
79
7.8
11.0
4.8
23.2
1.8
20.4
14.0
7.8
7.8
26.2
23.2
20.4
14.0
9.2
101
102
80
108
96
101
87
80
75
114
103
102
88
84
21.7
22.5
9.8
24.3
16.8
19.1
10.2
10.0
6.4
24.9
20.0
19.0
13.2
9.9
159
-------
Adding fish to the enclosures had a dramatic effect on the zooplankton
community. As shown in Figure 62, three species completely disappeared from
all fish enclosures — Daphnia pulex, Daphnia ga1eata mendotae and Chaoborus.
Three others appeared only in fish enclosures at the end of the experiment --
Daphn i a amb i qua, Daphni a parvula, and Asplanchna pr i odonta. While
Cer i odaphn i a increased at the lowest fish density, it remained in low
abundance at higher fish densities. Cyc1 ops verna1is appeared to be immune
to fish predation, as was Bosm i na long i rostr i s which increased with intensity
of fish predation. Several rotifers also increased with increasing predation
(Fig. 63) .
The effects of the fish treatments on the phytop 1ankton community were
just as dramatic. Not only were the algal volumes in the bags with the fish
about 16 times as great, on the average, as in the bags without fish, but
there were great changes in species composition as well (Table 21). In the
enclosures without fish the predominant algae were Cryptomonas, CeratIum,
C1osteri um, and Pen i um. At the lowest level of fish predation, the
gelatinous form Oocystis became dominant along with Scenedesmus, and, as fish
predation increased, filamentous blue-greens became more predominant in the
community with Aphan i zomenon f 1 os-aquae making up ~}2% of the volume in the
enclosure with 5 fish. Indeed the density of filamentous blue-greens
(Anabaena and Aphanizomenon) shows a good inverse relationship to the
abundance of Daphn i a plus Cer i odaphn i a (Fig. 64). The implication is that
there is a threshold in density of these cladocerans below which such algae
can escape grazing. The abundance of flagellates shows a similar
relationship to the density of Daphn i a and Cer i odaphn i a, (Fig. 65), but the
flagellates continue to be an important part of the algal community even at
the highest concentrations of the cladocerans. Presumably this group of
algae is able to maintain itself under intense grazing conditions (Fott, in
press) .
There is little doubt that the changes in zooplankton abundance resulted
from predation by the fish, and the relationships in Figures 64 and 65
suggest strongly that algal changes were a result of the zooplankton changes
i.e. as the major herbivores were eliminated the algal biomass became
greater.
Certainly there is no evidence that phytopiankton alterations resulted
from changes in nutrients because of the treatments. Table 22 shows that
total phosphorus was at much the same concentration in all enclosures and the
pond, and in fact averaged about 10% greater in the fishless enclosures than
in those with fish. Neither ammonia nor nitrate became undetectable and the
differences did not seem systematic. The high concentrations of P04-P in the
pond and fishless enclosures presumably result from the rapid turnover of the
algal population by the abundant grazers and the inability of the few
remaining algae to remove all the available phosphate from solution.
Enclosure Studies with Planktivores and/or Nutrients
Despite the fact that there was no evidence that the effects of the fish
treatments were related to nutrients, a second series of experiments was done
in 1976 to test the idea directly. This experiment, from May 29 to July 15»
was begun with twelve 1.8 m deep enclosures, six containing two small (60-70
mm) Lepom i s macroch i rus each, and six without fish. Three levels of
enrichment were used so that for enclosures with and without fish two were
left as controls (LN), two received 10 ug/1 phosphorus plus 110 ug/1 nitrogen
per week (MN), and two received 20 ug/1 phosphorus and 330 ug/1 nitrogen per
week (HN). Nutrients were added as solutions of KH2P04 and KH4N03 and
160
-------

s
\
.<§
c:

100
0
10
0
I Dapht
Mn
Daphnia pu/ex
I
i±M	
LJi
Daphnia gateata mendotae
2_ Daphnia
I
f parwla \ rh
0
1,000
0
3,000r
0
2or Daphnia I
| ambigua |
A
hr+in|m
Ceriodaphnia
reticulata
Bosmina ]
longirostris
iih
T>
C
o
CL
7 2 i 8 9 3 II 6
Increasing
No FishiFish
jPredation
,nnr_ Diaptomus^
n sicihides,
rh	, I
0
2,000
0
1,000
0
10


Nauplii j
ft] rh rh ^ rh ^ rh ^ rfi
Cyclops
vernalis
kti

h.
LLk
j Chaoborus
I americanus
'
2Qr Asptanchna
J Priodonta ^
ElL
"2 1 7 2 8 9 3 116
8.
| Increasing
No Fish|Fish —
! Predation
Fig. 62. Mean and range for abundance of crustaceans, Chaoborus
and Asplanchna for last three sampling dates of enclosure
experiment.
161
-------
<\J
5
\
:§


2,000
0
600
0
2,000
0
200
0
4,000
0
Brachionus
angufaris
Brachionus
quadridentato
Kerote/la
cochfearfs
Lecane /una
Monosfy/a Sp.
"2172
o
Q-
No Fish
[Q	dLi
m.

m _ rH
Jltl.
SR
8 9 3 II
Increasing
Fish
Predation
Fig. 63. Mean and range for abundance of predomi-
nant rotifers for last three sampling dates
of enclosure experiment.
162
-------
Table 21: Effects of increasing levels of fish predation on mean total biomass
(lCP/unP/ml), and proportional abundance of five dominant species.
(Data are means for the last 3 sampling dates.)

POND	15.91 + 3.59
Cryptomonas tetrapyrenoidosa	.50
Ceratium hirundinella	.18
Heterochromonas globosa	.14
Cryptomonas marsonii	.07
Botrydiopsis arhiza	.05
ENCLOSURE 1	6.01	+ 3.57
Ceratium hirundinella	.46
Cryptomonas tetrapyrenoidosa	.21
Heterochromonas globosa	.10
Closterium moniliferum	.06
Cosmarium sexnotatum	.04
ENCLOSURE 7	4.97	+ 2.00
Cryptomonas tetrapyrenoidosa	.43
Closterium moniliferum	.31
Ceratium hirundinella	.11
Cosmarium sexnotatum	.05
Aphanizomenon flos-aquae	.02
ENCLOSURE 2	8.77+1.23
Penium sp.	.27
Cryptomonas tetrapyrenoidosa	.26
Closterium moniliferum	.22
Ceratium hirundinella	.16
Cosmarium reniforme	.02
FISH ENCLOSURES
ENCLOSURE 8 (1 fish) 114.73	+ 4.21
Oocystis lacustris	.52
Scenedesmus armatus	.20
Sphaerocystis schroeteri	.08
Ankistrodesmus falcatus
var. mirabilis	.03
Schizochlamys planctonica	.03
ENCLOSURE 9 (2 fish) 106.60 + 37.85
Anabaena circinalis	.47
Cryptomonas tetrapyrenoidosa	.37
Cosmarium sexnotatum	.04
Pediastrum boryanum	.03
Schizochlamys planctonica	.02
ENCLOSURE 3 (2 fish) 83.41	+ 36.18
Oocystis lacustris	.46
Anabaena circinalis	.09
Cosmarium sexnotatum	.08
Heterochromonas globosa	.07
Sphaerocystis schroeteri	.07
ENCL0SUREJ1 (4 fish) 88.78	+ 26.37
Anabaena circinalis	.64
Cryptomonas tetrapyrenoidosa	.30
Aphanizomenon flos-aquae	.02
Treubaria crassispina	.01
Sphaerocystis schroeteri	.01
ENCLOSURE 6 (5 fish) 308.05	j_ 248.80
Aphanizomenon flos-aquae	.72
Heterochromonas globosa	.08
Anabaena circinalis	.07
Sphaerocystis schroeteri	.04
Cryptomonas tetrapyrenoidosa	.02
163
-------
30
£

\


\




a
20
to

c:



V

1

-------
Daphnia + Ceriodophnia /m (x 10 )
Fig. 65. Mean density (last 3 sampling dates) of
flagellates vs. mean density of Daphnia
and Ceriodaphnia in enclosures.
165
-------
Table 22: Chemical conditions in enclosures with and without fish
(means for last 3 sampling dates).
CONTROLS	FISH ENCLOSURES
Enclosure i
POND
1
7
2
8
9
3
11
6
Number of fish
--
0
0
0
1
2
2
4
5
Total Phosphorus
(ug/l)
160
100
119
134
100
121
106
92
128
P04 - P (ug/l)
84
23
38
43
4
3
4
2
9
NH3 - N (ug/l)
26
26
31
38
2
7
20
48
0
N02 - N + N03 - N
(ug/l)
14
13
13
19
30
37
12
26
30
PH
8.02
8.00
8.10
8.00
8.18
9.00
8.26
8.41
8.28
166
-------
stirred in after weekly samples had been taken. The N/P ratio was chosen to
equal that of the pond at the onset of the experiment. Two enclosures failed
to survive and three had only one fish left alive at the end of the
experiment. Nutrient concentrations in the enclosures are shown in Table 23.
It is clear from Figure 66 that nutrient enrichment was not the major
determinant of zooplankton community composition. In all enclosures without
fish Daphni a pulex was the dominant herbivore with Cer iodaphnia, Cyclops,
nauplii and rotifers remaining rare. In the enclosures with fish, Cyc1 ops
and nauplii appear to have increased at the medium levels of enrichment, and
rotifers increased at medium and high levels. However the presence of the
fish resulted in the removal of Daphn i a as well as reduction in the abundance
in D i aptomus at all levels of enrichment.
As in the first experiment there were accompanying changes in the
abundance and community structure of the algae (Table 24). On the average,
algal volumes in the fish enclosures were 2.5 times as high as in those with
no fish but in neither case did the algal abundance bear a relationship to
the nutrient status of the enclosures.
In the enclosures without fish, enrichment caused increases in Cosmar i um
and in Chroococcus, two algae of relatively poor nutritional quality for
herbivores. In the presence of the fish, desmids increased at the higher
nutrient levels. Qualitative differences between the enclosures with fish
and those without were not as striking as in the first enclosure experiment.
Indeed there were more blue-greens in the enclosures without fish than in
those with fish, but as noted above these were mainly Chroococcus, a
gelatinous blue-green. Perhaps the most striking effect was the appearance
of Cerat i um in the enclosures with fish.
As in the first experiment, then, the fish removed the larger herbivores
and the increase in algal biomass in the presence of fish was apparently not
nutrient-related. In the second experiment, the increase in algal volume
resulting from fish addition was smaller than in the first experiment. This
was in agreement with the fact that in the second case enclosures without
fish contained only about one-fourth as many large herbivores as in the first
exper iment.
Pond Separation Experiment
To test the effects of adding fish on a larger scale than hitherto, on
April 27. 1976, Pleasant Pond was divided in two. We used a double curtain
of nylon-reinforced polyethylene sheeting weighted firmly into the sediments
with steel chain and held about 25 cm above the water surface by a cable
stretched above the pond, and by styrofoam floats attached to the upper edge
of the curtain. The north half (PPN) was stocked by the Minnesota Department
of Natural Resources on Hay 5 with 10,000 walleye fry (Stizostedion vitreum).
Despite our seining efforts throughout the summer we neither saw nor captured
any of these fish. However, in early October the DNR was able to capture 306
fish (total weight 7-7 kg) after two days of intensive trap netting. Thus it
appears that most of the fish had not survived. Fortunately on June 5
several mature fathead minnows (Pimephales promelas) were sighted in the
southern half of the pond (PPS) and by June 13 schools of newly-hatched
minnows were visible from shore. On August 3 sampling of a 19 m2 area with a
3 mm mesh seine yielded 500 minnows of average length four cm. The origin of
these fish is unknown but since they were present in PPS only, and since PPN
had very few walleye, we merely redesignated the ends of the ponds so that
PPN was the fishless control and PPS was the end wi th fish.
As we had expected the zooplankton population responded essentially as
167
-------
Table 23: Chemical conditions, algal biomass, and photosynthetic rates in
control and enriched enclosures with and without fish.
Enclosure #
1
7
3
9
5
2
8
4
6
12
Number of fish
0
0
0
0
0
2
2
2
2
2
Nutrient regime
LN
LN
MN
MN
HN
LN
LN
MN
HN
HN
Total Phosphorus
(ug/1)
34
24
60
44
58
29
44
61
108
108
PO4 - P (ug/1)
2
4
5
3
5
2
3
18
14
10
NH3 - N (ug/1)
40
23
56
58
76
16
59
96
111
70
NO? - N + NOo - N
(ug/1)
6
22
76
34
196
6
8
36
186
134
PH
9.5
9.0
9.4
9.7
9.9
9.2
9.1
9.7
9.8
10.0
mg C fixed/liter/
day
.07
.09
1.11
.07
.06
.36
.15
2.02
6.56
2.35
10"^ pg C fixed/u"^
phytoplankton/day
0.8
0.8
1.6
0.3
0.4
0.7
0.3
4.40
13.48
1.95
lO^u^ phytoplankton/
ug total phosphorus
2.5
4.9
11.5
5.1
2.4
17.6
11.7
7.5
4.5
11.2
168
-------
x6
42
S3
£
200
0
80-
0-
20-
0-
400-
0
1000-
0-
40,000
0
j r
1 Daphnia put ex
1 I 1
rn r+.
rh rh i
I Ceriodophnio
I reticulata
PI PI fh rh rh
Diaptomus
davipes
	rh i4-t	
0
Cyclops
vernalis
rh H-I	I—rh rh
I Nouplii
J	rh rb
Rotifers
I
n rh
400-1 Asptanchna
priodonta
rtiftim
Chaoborus
americanus
I 7 3 9 5 12
L L M M H I L
No Fish !
8 4 6 12
L M H H
Fish
Fig. 66. Mean and range (last 3 sampling dates)
for abundance of zooplankton species in enclosures
at different nutrient concentrations. Nutrient
levels are L - low (controls); M - medium; H - high.
169
-------
Table 24: Effects of fish predation on mean total phytoplankton biomass
(10^ um^/ml) and proportional abundance of five dominant species
at different levels of enrichment. (Data are means for last 2
sampling dates.)
NO FISH

FISH ENCLOSURES (2 fish eac
h)
ENCLOSURE 1 - LN
8.55
ENCLOSURE 2 - LN
51.05
Closterium moniliferum
.62
Staurastrum paradoxum
.57
Staurastrum paradoxum
.13
Ceratium hirundinella
.29
Ulothrix oscillarina
.10
Cosmarium sexnotatum
.05
Chroococcus dispersus
.10
Cyclotella sp.
.03
Heterochromonas globosa
.01
Gloecystis vesiculosa
.01
ENCLOSURE 7 - LN
11.71
ENCLOSURE 8 - LN
51 .55
Rhodomonas minuta
.35
Ceratium hirundinella
.43
Staurastrum paradoxum
.31
Closterium moniliferum
.20
Chroococcus dispersus
.12
Cyclotella sp.
.10
Pediastrum boryanum
.04
Cosmarium sexnotatum
.08
Gloeocystis vesiculosa
.04
Staurastrum paradoxum
.06
ENCLOSURE 3 - MN
69.06
ENCLOSURE 4 - MN
45.88
Chroococcus dispersus
.60
Staurastrum paradoxum
.77
Cosmarium reniforme
.17
Cosmarium sexnotatum
.10
Cosmarium sexnotatum
.10
Ceratium hirundinella
.08
Aphanizomenon flos-aquae
.05
Chroococcus dispersus
.02
Sphaerocystis schroeteri
.02
Chlamydomonas sp. 1
.01
ENCLOSURE 9 - MN
22.38
ENCLOSURE 6 - HN
48.67
Chroococcus dispersus
.92
Staurastrum paradoxum
.61
Straurastrum paradoxum
.05
Cosmarium sexnotatum
.33
Ulothrix oscillarina
.01
Chroococcus dispersus
.03
Heterochromonas globosa
.01
Pediastrum boryanum
.01
Oscillatoria tenuis
<.01
Trachelomonas sp. 1
.01
ENCLOSURE 5 - HN
13.88
ENCLOSURE 12 - HN
120.38
Chroococcus dispersus
.49
Scenedesmus denticulatus
.76
Cosmarium sexnotatum
.38
Staurastrum paradoxum
.17
Staurastrum paradoxum
.05
Scenedesmus quadricauda
.02
Oscillatoria tenuis
.03
Cosmarium sexnotatum
.02
Ulothrix oscillarina
.01
Oscillatoria tenuis
.01
170
-------
it had in the enclosures (fig. 67)- By the end of June the fathead minnow
population had become well established and the zooplankton populations in the
two halves of the pond began to diverge. In Pleasant Pond South, Daphnia
pulex was eliminated by the end of July, and Ceriodaphnia and Diaptomus
clavipes remained very rare. At the end of summer the two small daphnids, D.
ambiqua and D^ parvu 1 a appeared as they had in the "fish" enclosures in
experiment 1. Nauplii were unaffected by the predation but rotifers became
very abundant. In PPN, Daphnia pu1 ex became very abundant through July then
gave way to Cer i odaphn i a. Rotifer numbers remained low.
Coincident with these differences in zooplankton there were large
differences in the algal biomass and species composition. Until mid-June
(Table 25) the two halves of the pond maintained essentially equal
concentrations of algae. However, from June 25 on, algal volumes in the
south end were greater, and for the period June 25 to July 23 averaged 5
times as great. The picture is complicated by the fact that total phosphorus
in both halves of the pond increased until mid-June, remained high, and fell
again in August (Table 25)* The reasons for these increases are unknown.
1976 was a particularly dry year and part of the increase may be a
consequence of evaporation. Certainly there was no runoff from June through
August. Also as the ponds stratified in 1976 and the bottom waters became
anoxic, sediment P release may have contributed. Furthermore the minnows in
the south half may also have increased phosphorus concentrations there
through excretion, as when the zooplankters became very scarce the minnows
began to feed on the sediments. In the northern half this explanation was
not tenable but dense stands of the macrophyte Ca11i tr iche developed, and
they may have transferred nutrients from the sediments to the water.
In any event it is clear from Table 25 that increases in total
phosphorus notwithstanding, from June 25 to early August phytop 1 ankton/tota 1
P ratios were higher in PPS, averaging $.k times as great as those in PPN.
The low ratios and low algal abundances in the north end were not a result of
nutrient limitation. As shown in Table 25 from June 12 to July 23 specific
rates of carbon fixation in the north end were equal to or even greater than
those in the south end.
Composition of the algal community in both ponds was similar through May
consisting mainly of dense populations of flagellates (Chlamydomonas spp.,
Cryptomonas obovata, and Phacus spp.). During June both pond halves
developed blooms of Aphani zomenon flos-aquae. In the south half the bloom
crashed and was replaced in July by a dense bloom of the green alga
Ank i strodesmus, following which Osc i11ator i a tenu i s and Anabaenops i s became
abundant through August. In the north half the Aphan i 2omenon bloom
collapsed, recovered, and finally disappeared by late-July. During August a
variety of forms were present including flagellates, diatoms, Sphaerocyst i s
and the filamentous Osc i11ator i a and Ulothr i x, but total biomass remained
1 ow.
Thus as in the enclosure experiments, addition of fish had the effect of
eliminating the larger herbivores and increasing the algal biomass many-fold.
Nutrients did not seem to be involved as rates of primary production were
equally high in the pond halves with and without fish. Furthermore in all
cases increased fish predation led to increased algal biomass despite greater
abundance of small herbivores.
LOCH LOSO
Loch Loso is a small body of water consisting of two interconnected
ponds having a total area of about 1/2 hectare. The ponds are on the grounds
171
-------
APRIL , MAY , JUNE , JULY , AUG.
200

$
\


Dophma pulex
Ceriodaphnia reticulata
0
4p Daphnia parvuta
0I	L		L
4r Daphnia ambigua
0I	1 ' 1
00
oL-o-
jy.-°
-o-a
rr'a«
& ro
o
100
o
200
0
2000
0
40
0
Dtaptomus clavipes
—•—<1 < » ***&££
Cyclops verna/is
	1 » 0 *
6173
Nauplii
Rotifers
A
—•'•QeWC > t	#
Chaoborus americanus
Fig. 67.
APRIL MAY JUNE JULY AUG.
1976
The 1976 abundances of main zooplankton species
in Pleasant Pond North having no fish (solid lines)
and Pleasant Pond South in which fathead minnows
began to appear in June (dashed lines).
172
-------
Table 25: Total P, phytoplankton, community arid specific photosynthetic rates and biomass/TP ratios in
Pleasant Pond, 1976.
Date
May 14
May 27
June 12
June 25
July 9
July 23
August 20
TP mq/m3
PP N	PP S
33	26
43	47
120	71
119	193
109	119
1 02	1 57
12	39
phytoplankton
107 um^/l
PP N PP S
83	55
52	47
288	220
762	4130
207	690
61	345
254	332
mq C fixed/1/dy
PP N	PP S
0.05	0.06
0.06	0.10
1.40	0.86
1.15	0.87
0.74	0.93
0.23	0.92
0.11	1.66
10~4 pg	C fixed/um^
phytoplankton/dy
PP N	PP S
0.7	1.0
1.2	2.2
4.8	3.9
1.5	0.2
3.5	1.4
3.7	2.7
0.4	5.0
107 um^
ph.ytoplankton/ug TP
PP N PP S
2.5	2.1
1.2	1.0
2.4	3.1
6.4	21.4
1.9	5.8
0.6	2.2
21.2	8.5
-------
of Hennepin Community College in Minneapolis and are used to collect storm
runoff. As a result they are productive. The larger half (HCC1) encompasses
about 2/3 of the total area, and has a maximum depth of 3 meters, The smaller
half (HCC2) has a mean depth of about 2 meters. Being shallow, the ponds
normally winter-kill. As a result when we first inspected them late in 197^
the ponds had substantial populations of Daphnia pulex. They also had a
so-called "flake" or "grass blade" bloom of Aphani2omenon.
Our intention was to divide the ponds early in 1975. stock the larger
half with planktivorous fish (using the other as a control), and determine
the effects of the fish on the zooplankton and algae. Then we proposed to
add piscivorous fish to the larger half early in 1976 and follow the results
for a year more. This plan was followed but not exactly as we had hoped.
1975 was a wet year, and although we did separate the ponds and add
planktivores to HCC1, (350 mature perch and 100 bluegill sunfish) the ponds
soon became joined as a result of greatly increased water levels and remained
so until July until after the fish reproduced. Therefore in late 1975 we
removed the barrier and allowed the fish to equilibrate to both sides. On
April 9, 1976, the ponds were divided again and 3^ northern pike (Esox
1ucius) about 13 inches long were added to HCC1. However the planktivorous
fish had become so abundant that the predators had little effect and even
trapping fish from HCC1 and placing them in HCC2 did not lower the
planktivore abundance in HCC1 low enough to allow Daphni a to reoccur.
The results of periodic vertical hauls for zooplankton are shown in
Tables 26 and 27. Daphnia pulex was very abundant in both sides of the pond,
in 1975 from March to early August in HCC1, and in 1975 from March to late
August in HCC2 (Figs. 68 and 69). Smaller cladocerans such as Bosmi na and
Chydorus were relatively rare. In both halves D i aptomus c1av i pes were
present early but gave way in June to large populations of [k s i c i1oi des.
Rotifers were rare throughout the whole period. Clearly the planktivorous
fish had little effect in 1975- However by 1976 the picture was different
(Tables 26 and 27. Figs. 68 and 69). pulex was completely absent from
HCC2 and was found at low concentrations in HCC1 on only two dates. Bosmi na
which had been rare in 1975. became abundant in July and August, and Chydorus
sphaer icus became the dominant cladoceran. Diaptomus c1avipes was absent and
D. s i c i1oi des became abundant in both parts of the pond. In contrast to 1975.
rotifers became extremely abundant. The shift from Jk c1av i pes to D.
s i c i1oi des in 1976 may have resulted from predation by the planktivores, and
the brief appearance of D_j_ pul ex in HCC1 in 1976 may have indicated the
effect of the northern pike. However, it is clear that in this system, as
opposed to Pleasant Pond and the enclosure systems, there was a lag of about
a year in the effect of the planktivore introduction. The effect of the
piscivore introduction could not be studied further in 1977 as a nearly
complete fish-kill occurred in 1976-77•
Phytop 1ankton response to the zooplankton was as expected. in 1975.
except for periods when Aphan i zomenon flos-aquae formed "flake" blooms,
chlorophyll concentrations were very low (Fig. ~J0 and Table 28). They were
not low for lack of nutrients (Table 28) and in fact only on June 26 when
chlorophyll concentrations were relatively high were there any indications
that available nitrogen might be low enough to be limiting. Rather, the high
concentrations of free phosphate, as in the enclosures in Pleasant Pond
(above) implicate the high zooplankton grazing rate as the reason for the
sparse algae. As a result of the low algal populations Secchi disc
transparency (Figures 71 and 72 and Table 28) was high and remained so even
on June 26 1975 when Aphanizomenon was present in flake form. Chlorophyll
samples were collected in 1976 but lost before analysis. However,
174
-------
Table 26: Zooplankton (thousands of individuals/m^) in vertical net hauls in HCC1, 1975 and 1976.
Bosmina longirostris
Ceriodaphnia quadrangula
Chydorus sphaericus
Daphnia pulex
Cyclops vernal is
Mesocyclops edax
Diaptomus clavipes
Diaptomus siciloides
Naupli i
Asplanchna priodonta
Brachionus angularis
Brachionus calycifloris
Brachionus piaticatilis
Brachionus quadridentata
Colurella sp.
Epiphanes brachionus
Filinia sp.
Kellicottia longispina
Keratella cochlearis
Keratella quadrata
Keratella vulga
Lecane luna
Monostyla sp.
Mytlina sp.
Platyias sp.
Polyarthra sp.
Synchaeta sp.
1975
March	April	May	June
4 11 26 2 9 25 2 7 14 23 28 6 15 20 26
.994	3.82
2.39	.866
.303	11.4	1.28	5.20
1.00 2.86 5.23 2.00 .697 2.73 12.1 86.2 154 18.7 35.8 81.3 21.9	19.1	31.2
.061 .636 .224 .333 .697 .606 .187	8.11 6.97 23.5 19.9 28.0 43.3
3.33 7.32 .299 .333 .460 .187
.348 .697 1.21 .374
1.84 7.43 .713 1.99 .866
8.56 33.8 36.9 58.9
4.14 52.3 27.4 20.7 70.5 121 104
.667
.232
,333 1.27
.369 .929
1.38 3.64 3.25
.713
,309
1.99
Chaoborus
.398
-------
Table 26 Continued
1975	1976
July	Aug.	Sept. Oct. Dec. Jan.	Feb. April
3 11 18 25 1 12 1 9 27 10 17 1 12 12 23 6 9 17 28
B.	longirostris	1.49	3.39
C.	quadrangula	.978	.994
C.	sphaericus	3.91 2.98	37.7 283 2134 2151 711 8.80 4.28 2.51 6.21 17.3
D.	pulex	6.85 53.7	53.2 12.9 15.8	5.34
C.	vernalis	39.1 41.8	136 1.99 2.26 25.4 25.9 48.7 63.3 48.1 22.4 2.93	5.77
M.	edax 8.27 5.77
D.	clavipes	1.96 16.4	5.21 3.98 3.39
D.	siciloides	18.6 28.3	8.34 15.9 20.3 13.9 40.1 81.9 240 336 183 20.5 4.62 188 40.7 12.5 12.4 326
Nauplii	47.9 79.0	45.9 4.97 23.7 41.6 58.9 345 538 395 58.2 4.62 8.56 7.53 352 335
S:aPn~s	8.56 7.41 5.02 ,68 3,7
B. calycifloris	42.0 583 10.7
B. platicatilis
B. quadridentata	11.3
Colurella sp.
Fi1iniahspnUS	370 1055 711 411 547 3756 20"2
K. longispina
K. cochlearis	2.36	202 85.4 8.56 11.1	5238
K. quadrata	16 2 171 744	2Q7
K. vulga
L. luna	.978 1.04
Monostyla sp.	.994
Mytilina sp.
Platyias sp.
Poiyarthra sp.	522 3606 2243 1019 1556 3822 20-2
Synchaeta sp.
Chaoborus
-------
Table 26 Continued
B.	longirostris
C.	quadrangula
C.	sphaericus
D.	pulex
C.	vernal is
M. edax
D.	clavipes
D.	siciloides
Naupl i i
A.	priodonta
B.	angularis
B. cal.ycifloris
B. piaticatilis
B. quadridentata
Colurella sp.
E.	brachionus
Filinia sp.
K. longispina
K. cochlearis
K. quadrata
K. vulga
L. luna
Monostyla sp.
Mytilina sp.
Platyias sp.
Polyarthra sp.
Synchaeta sp.
Chaoborus
1976
May	June	July	Aug.
7 14 20 27 4 11 17 25 3 3 16 22 30 5 13 20 27
3.18 6.93 10.2 207 283 272 743 3141 1371 4525
4.62 6.93 12.3 2.09
2.45 2.24 8.29 7.50 11.8 20.6 126 194 67.5 29.3 23.8 15.2 7.94






1 .54

2.04







10.7
2.45
2.24












2.36


2.24

18.5
23.8
26.2
194
161
205
188
111
36.5
126
91 .0
84.8
243
258
204
82.9 5.63


16.9
8.66
14.3
10.5
9.14
3.37
1.59


2.36
681
363
505
451 91.9
205
197
180
355
472
310
283
104
105
81 .8
203
292
331
88.3


15.2
194
127
120
877
15.6
528
511
425
3.66
314
10.1
7.94
3.99
6.21
2.36
80.8
64.1 29.5	41.3 6.73
182
34.7
17.2
4.47






2.09

9.98


2.45
11.2
852 6640
1542
1585
114
50.2
77.7
71.1
71 .4
53.9
8.27
5.34
7.36
6.71
19.4 33.8
3.37
20.6
26.2
8.66
10.2



2.36



5.63
50.5
1.59
95.4
575
613
3925 241
226 133
216
314 238
4.76	52.3	6.93	6.28 1.68	2.36
39.7	782	1510 16.4 2.09	1.59
p649	5.53	316	459	53.7 4.09 1.83 1.59
71 .4
-------
Table 27: Zooplankton (thousands of individuals/m2) in vertical net hauls in HCC2, 1975 and 1976.
Alonella sp.
Bosmina longirostris
Chydorus sphaericus
Daphnia pulex
Cyclops bicuspidatus
Cyclops vernal is
Mesocyclops edax
Diaptomus clavipes
Diaptomus siciloides
Naupli i
Asplanchna priodonta
Brachionus angularis
Brachionus calycifloris
Brachionus quadridentata
Filinia sp.
Keratella cochlearis
Keratella quadrata
Keratella vulga
Lecane luna
Mytilina sp.
Platyias sp.
Polyarthra sp.
Synchaeta sp.
1975
March	April	May	June
4 11 26 2 9 25 2 7 14 23 28 6 15 20 26
.030
.020 .247	1.92 4.30 6.83 20.2
7.15 5.09 .636 .783 1.98 1.01 2.43 61.5 98.2 19.1 101 20.2 34.4 44.4 21.4
2.15
.314 .030 .040 .247 .040 1.09 2.67 3.32
.962 1.07 8.54 10.7
2.23 1.82 .091 .040	.040	.222 1.42
1.07 3.42 1.19
.030	.673 .747 .586 2.55 22.2 12.8 53.9 85.9 49.5 34.4
.447
2.23 .364 .061
,020
.187	.475
,187 2.18 2.89 1.42 1.92
2.15
1 .07
-------
Table 27 Continued
Alonella sp.
B.	longirostris
C.	sphaericus
D.	pulex
C. bicuspidatus
C.	vernal is
M. edax
D.	clavipes
D. siciloides
Naupli i
A.	priodonta
B.	angularis
B. calycifloris
B. quadridentata
Filinia sp.
K. cochlearis
K. quadrata
K. vulga
L. luna
Mytilina sp.
Platyias sp.
Polyarthra sp.
Synchaeta sp.
1975
July
11
Aug.
18 25 1 12
Sept.
19 27 10 17
1976
Oct. Dec. Jan.
1
12 12
Feb. April
23 6 9 17
18
46.5 2.15	6.93 66.7 20.6 280
5.05 116 46.2 57.3 44.4 100 44.4
751 381 37.5 26.6
2.20
2.70
8.08 46.3 4.29 3.58 5.39 5.20
1 .47
5.05 16.0 4.29 7.17 13.5 34.6 22.2 3.44 14.7 112
135 242 44.0 74.1 6.73 31.2 62.9 13.8 88.3 138
243 32.9
102
125
6.59
9.86 1659 7.65 12.7 3.27
9.86
5.92
834 2260 1170 864
70.9 26.6
5.92 16.6 3.82
3.32	22.9 2.55
191
8.81
29.2 17.6
31 .5 103
2.94
2492 13.2
32.3
103 3253 1294 1048 177 7017 5672
15.9	23.5
-------
Table 27 Continued

1976
May
7
14
20
27
June
4
i
11
17
25
July
3 8
16
22
30
Aug.
5
13
20
27
Alonella sp.










10.8
1 .65
29.6
7.86

1.83
3.08
B. longirostris











6.59
6.73
2.62
16.7
1 .83
23.1
C. sphaericus
1.66
4.40
1.78
2.14

18.2
293
127
164
42.7
195
36.2
1 .35
5.24

1 .83
4.62
D. pulex

















C. bicuspidatus

















C. vernal is


12.5
8.56

4.98
3.01
3.90

3.56







M. edax






6.01
33.1
77.3
90.7
152
219
64.6
128
80.2
85.9
159
D. clavipes

















D. siciloides
49.8
7.34
1.78
4.28
5.10
11.6
7.52
5.84
1.21
3.56
14.4
1.65
5.39
7.86

1.83

Naupl i i
49.8
116
51.6
64.2
125
87.9
176
171
485
493
1053
196
246
315
410
583
363
A. priodonta
9.95

16.0
139
1277
662
21.0
19.5

1.78







B. angularis



92.0







11.5
8.08
115
8.36


B. calycifloris
9.95
33.7
8.90
51.4
828
13.3











B. quadridentata






263
5.84
3.03



21.5
21.0

11.0
27.7
Filinia sp.
8.30
98.3
171
21.4
12.7


1.95






1.67


K. cochlearis

51 .3
1335
1600
449
28.2
76.7
27.3
1.52
1.78





1.83

K. quadrata
254
3192
1808
2525

3.32
1.50

21 .2
10.7


1 .35
13.1



K. vulga




5.10


3.90
4.55
5.34
54.1
85.7
7143
3980
76.9
40.2
41.6
L. luna

















Mytilina sp.




2.55



13.6
12.5
18.0
3.30
6.73


5.48
3.08
Platyias sp.




5.10
31 .5
1840
3238
2627
626
7.22






Polyarthra sp.
2276
478
92.5
44.9
2.55
360
1544
195
1.52








Synchaeta sp.
56.4
95.4
299
19.3













-------
Fig. 68. Daphnia pulex densities in HCC1
1975 & 1976.
? 60
M
Fig. 69. Daphnia pulex densities in HCC2,
1975 & 1976.
-------
.55 2|7
HCCI
40
30
'112
HCC2
A
M
J
J
A
Fig. 70. Chlorophyll concentrations in HCCI
& HCC2, 1975.
182
-------
Table 28: Chemical conditions and chlorophyll in
44/um mesh. 1 = HCC1 2 = HCC2
1975	March	April


4
11
26
2
9
25
P04-P
1
20
20
268
272
218
104
(ug/i)
1-bot


144
292
192
98

2
29
20
246
216
196
80

2-bot


298
214
264
65
Total Dis.-P
1
51
56
74
92
43
102
(ug/1)
1-bot


32
102
33
107

2
82
59
54
48
30
86

2-bot


56
55
53
66
Dis. Org. P
1
31
36



0
(ug/1)
1-bot





9

2
53
39



6

2-bot





1
Part. P
1
53
44
28
2
0
8
(ug/1)
1-bot


23
2
0
23

2
99
279
27
29
0
26

2-bot


20
6
0
36
Total P
1
104
100
102
94
38
110
(ug/1)
1-bot


55
104
33
130

2
181
238
81
77
29
112

2-bot


76
61
49
102
Chi a<44um
1
0.5
0
0.4
0.3
1.0
5.0
(ug/1)
2
0.2
0
0.8
0.2
5.0
4.0
Chi a">44um
1
0
0.1
0
0
0
0.1
(ug/1)
2
0.4
0.1
0.1
0
0.7
0.2
Tot. Chi a
1
0.5
0.1
0.4
0.3
1.0
5.1
(ug/1)
2
0.6
0.1
0.9
0.2
5.7
4.2
and HCC2, 1975. Chlorophyll was partitioned with a
May
2
7
14
23
28
June
6
15
20
26
78
79
13
6
18
24
46
31
3
80
61
14
18
21
22
45
39
33
75
61
10
20
50
32
43
4
9
77
62
10
27
58
32
47
45
44
64
97
28
23
36
52
77
56
30
64
92
37
34
41
41
68
61
50
63
97
30
38
65
55
64
30
40
69
91
30
42
74
59
109
46
65

18
15
17
18
28
31
25
27

31
23
16
20
19
23
22
17

36
20
18
15
23
21
26
31

29
20
15
16
27
62
1
21
5
7
34
3
5
0
2
6
59
10
11
29
5
1
18
11
8

12
4
31
2
9
12
9
67
73
5
37
26
8
4
0
0
26

69
104
62
26
41
52
79
62
89
74
103
66
39
42
59
79
69

75
101
61
40
74
67
73
97
113
74
128
56
50
78
59
91
72

1.5
2.1
11
1.2
2.8
1.7
1.3
1.0
1.4
3.0
2.4
24
1.8
0.9
4.4
0.9
0.8
1.3
0.1
0.3
1.0
0.5
0.2
0.3
1.0
0.4
23
1.0
0.8
1.9
1.1
0
0.2
1 .9
39
32
1.6
2.4
12
1.7
3.0
2.0
2.3
1.4
24.4
4.0
3.2
25.9
2.9
0.9
4.6
2.8
39.8
33.3
-------
Table 28 Continued
1975
July
Aug.
11 18 25 1 12
P04-P
1

5
18
30
50
32
5
(ug/1)
1-
•bot
35
37
62
47
43
10

2

2
30
21
54
42
26

2-
¦bot
52
37
27
77
54
72
Total Dis.-P
1

51
57
73
100
47
41
(ug/1)
1-
•bot
72
78
107
84
69
37

2

43
120
67
102
56
69

2-
•bot
84
80
85
114
65
112
Dis. Org. P
1

46
39
43
50
15
36
(ug/1)
1-
•bot
37
41
45
37
26
27

2

41
90
46
48
14
43

2-
¦bot
32
43
58
37
11
40
Part. P
1

10
36
13
14
9
61
(ug/1)
1-
•bot
15
20
15
21
23
59

2

11
0
18
22
12
41

2-
¦bot
7
41
10
13
13
30
Total P
1

61
93
86
114
56
102
(ug/1)
1-
-bot
87
98
122
105
92
96
2

54
104
85
124
68
110

2-
-bot
91
121
95
127
78
142
Chi a <44um
1

2.6
1 .7
1.1
1.7
5.4
14.2
(ug/1)
2

1.9
1.1
0.8
0.5
1.9
2.6
Chi a>44um
1

1.1
8.0
0.4
0.1
4.6
141
(ug/1)
2

0.6
10
4.1
0.7
4.5
40
Tot. Chi ci
1

3.7
9.7
1.5
1 .8
10.0
155
(ug/1)
2

2.5
11.1
4.9
1.2
6.4
42.6
IP 27
10	12
9	12
117	64
120	68
43	60
41	60
151	98
158	103
33	48
32	48
34	34
38	35
177	72
115	79
116	122
124	109
220	132
156	139
267	220
282	212
21.7
3.7
195
108
217
112
-------
Table 28 Continued
ICE 	
1975

March

April

May



June



4
11
26
2
9
25
2
7
14
23
28
6
15
20
NO3 - N
1
592
816
390
342
395
360
335
385
315
240
220
188
155
138
+ N02 - N
1 -bot


720
278
545
435
440
415
235
240
220
208
150
155
(ug/1)
2
448
550
330
288
390
425
380
425
300
280
230
196
175
120

2-bot


270
308
390
315
410
400
245
280
220
196
155
165
NH3 - N
1
210
84
895
550
680
420
300
293
16
86
84
190
173
156
(ug/1)
1 -bot


1060
550
570
366
260
335
54
96
68
180
165
156

2
206
132
840
540
590
366
340
312
30
146
120
84
118
17

2-bot


765
520
480
423
330
335
30
160
216
90
130
156
PH
1
7.21

6.91
7.30
7.32
7.90
7.32
7.96
9.10
8.51
7.81
7.93
7.75

1 -bot


6.91
7.10
7.25
7.43
7.23
7.68
9.03
9.01
7.72
7.83
7.77


2
7.03
7.43
6.86
7.20
7.20
7.40
7.30
7.69
9.20
7.89
7.61
8.05
7.61


2-bot


6.74
7.20
7.20
7.29
7.30
7.72
9.22
7.98
7.60
8.06
7.56

A1kalinity
1


1.02
0.86
0.90
0.90
0.75
0.93
0.90
0.85
0.82
1.06
0.91

(meq/1)
1 -bot


1.64
0.63
1.14
0.86
0.69
0.78
0.82
0.86
0.78
1.00
0.90


2

1.49
0.59
0.69
0.79
0.67
0.70
0.82
0.82
0.83
0.67
0.90
0.76


2-bot


0.59
0.67
0.78
0.79
0.63
0.82
0.80
0.85
0.67
0.87
0.75

Dis. Oxygen
1
2.85
1.86
9.89
9.01
8.00
7.90
9.50
9.40
12.35
7.38
7.10
6.61


(mg/1)
1 -bot
6.09
1.15
9.40

2.99
7.65
9.55
9.26
12.34
9.00
6.81
7.03



2
3.27
1.75
10.45
7.97
8.75
8.11
9.66
9.23
12.20
6.35
6.07
7.82



2-bot
4.40
1 .82
1.35
7.06
6.56
5.17
9.60
9.10
12.25
5.70
5.65
7.82


Secchi Disc,
1





.46
.91
1.55
1.52
2.74
2.90
2.54


(m)
2





.33
.86
1 .52
1.24
1.52
2.44
1.96


( = bottom)
-------
Table 28 Continued
1975	June July
26 3 11 18 25
N03 - N
+ N02 -
(ug/1)
NHo - N
(ug/1)
PH
A1 kalinity
(meq/1)
Dis. Oxygen
(mg/1)
Secchi Disc,
(m)
( = bottom)
1
16
60
188
19
24
1-bot
6
49
45
13
18
2
2
53
720
124
150
2-bot
2
112
384
140
151
1
0
29
6
22
58
1-bot
105
50
27
75
46
2
1
12
28
31
65
2-bot
1
70
41
38
82
1
9.18
7.73
7.46
7.35
7.76
1
cr
o
7.20
7.32
7.31
6.85
7.70
2
9.80
7.72
7.50
8.15
8.03
2-bot
7.50
7.15
7.40
7.90
7.97
1

0.78
0.86
0.90

1-bot

0.91
0.81
0.86

2

0.71
0.75
0.96

2-bot

0.45
0.78
0.90

1

6.45
7.00
6.49
5.80
1-bot





2

6.76
7.52
7.26
5.07
2-bot





1
1.98
2.57
1.98
2.74
2.54
2
1.68
2.59
1 .80
2.26
2.29
Aug.
1 12 19 27
19
18
75
92
95
255
163
173
8.08
7.67
8.20
7.97
1.04
1.02
1.31
1.35
8
2
20
72
10
19
35
216
9.82
8.74
8.62
7.87
0.94
0.83
1.14
1 .20
13
5
21
14
131
118
187
42
9.33
9.38
7.95
8.07
1.06
1.02
1.39
1.37
9.27
9.23
5.72
6.11
24
21
25
20
28
28
33
45
9.30
9.35
8.82
8.50
1.04
1.04
0.81
0.80
7.06	18.45
5.25	8.44
5.70 10.05
5.30 5.55
2.13 0.94
1.98 1.85
0.71 0.79
0.89 0.63
-------
1976
HCCI
1975
M
Fig. 71. Secchi disc transparency in HCC1,
1975 & 1976. (— = bottom.)
1976
HCC2
1975
Q
o 2 0
a>
cn
2 2
24
2 6
2 8
A	M	J	J	A
Fig.
72.
Secchi disc transparency in HCC2,
1975 & 1976. (— = bottom.)
-------
transparency measurements survive and show (Table 29) that transparencies
were very much lower in 1976 than in 1975- Furthermore, although algae were
abundant, Aphani zomenon f1os-aquae did not form grass blades in 197& in the
absence of Daphn i a pu1 ex.
CONCLUSION
These experiments demonstrate the dramatic effectiveness of
p 1 anktivorous fish in removal of large herbivores and the consequent dramatic
effects on the phytopiankton. They also show that the system is complex, as
witnessed by the fluctuations of the Aphani2omenon. The experiment on
control of the pianktivorous fish was terminated by the winter-kill, and so
no conclusions can be drawn as to its ultimate success or failure.
ACKNOWLEDGEMENTS
We gratefully acknowledge the management of North Oaks Development for
allowing us to use Pleasant Pond, and Hennepin Community College for use of
Loch Loso.
REFERENCES
American Public Health Association. 1971* Standard methods for the
examination of water and wastewater. 13th ed. Washington, D.C. American
Public Health Association. pp.
Chaney, A.L. and E.P. Marbach. 1962. Modified reagents for determination of
urea and ammonia. Clinical Chemistry 8:130-132.
Fott, J. In press. A comparison of the growth of flagellates under heavy
grazing stress with a continuous culture.
Haney, J.F. and D.J. Hall. 1973- Sugar-coated Daphnia: A preservation
technique for Cladocera. Limnol. Oceanogr. l8:331~333-
Lynch, M. 1979- Predation, competition, and zooplankton community structure
-- an experimental study. Limnol. Oceanogr. 24:253"262.
Lynch, M. and J. Shapiro. 1981. Predation, enrichment, and phytoplankton
community structure. Limnol. Oceanogr. 26:86-102.
Shapiro, J., V. Lamarra and M. Lynch. 1975- Biomanipu1 ation: an ecosystem
approach to lake restoration, pp. 85-96 J_n Brezonik, P.L. and J.L. Fox
(eds.). Proc. symp. on water quality management through biological
control. USEPA Rept. No. ENV-07-75'1.
188
-------
Table 29: Secchi disc transparency (m) HCC1 + HCC2, 1976.
Date, 1976	HCC1	HCC2
April 17
1.22
.76
28
1.09
.76
Hay 7
1.17
.63
14
.69
.69
20
.76
.74
27
1.22
.63
June 4
1.22
.91
12
.59
.46
17
.46
.28
26
.33
.30
July 3
.23
.30
9
.25
.28
16
.28
.23
23
.20
.25
30
.18
.23
Aug. 5
.18
.15
13
.18
.18
27
.20
.15
189
-------
IV. B. THE EFFECTS OF FISH TOXICANTS ON PHYTOPLANKTON*
FILE DATA
One of the best ways to look at the roles of fish in lakes is to remove
the fish and see what happens. As described in Section I, if benthivores
were abundant, one of the effects might be to lower nutrient concentrations
and, therefore, reduce the algal biomass. Another effect, if many
planktivores were present initially, might be an increase in the numbers of
larger herbivores which would lower the algal biomass. As many lakes in
Minnesota have been "reclaimed" by the Department of Natural Resources, i.e.
treated with fish toxicants, in the last 30 years, we examined existing
records to see if any pattern would emerge. In all we found that 229 lakes
had been treated between 19^5 and 1976. Of these 138 were "cold water" lakes
managed for trout, and 91 were "warm water" lakes managed for species such as
bass, northern pike, and walleye. Because the effects we were looking for
are unlikely in cold water lakes, we limited our search to the 90 warm water
lakes. Furthermore, as little data other than fish information generally was
collected we attempted to use Secchi disc transparency measurements where
possible to determine whether the treatment had resulted in greater clarity
of the water, presumably as a result of lessened fertility or increased
herbivore abundance.
Sources for our data were as follows:
1.	Minnesota Department of Natural Resources
-- files of the Section of Fisheries
-- files of the Section of Ecological Services
--Carl's Lake progress reports.
2.	Fremling, C., Lake Winona report, Winona State University 1973
3- Limnologica! Research Center, University of Minnesota,
-- Transparency in Minnesota Lakes -- reports from the Secchi disc
program.
Of the 91 lakes for which data was sought, 78 had to be eliminated
because either pre or post-transparencies were unavailable or the lake was
shallow enough so it winter-killed annually and was managed for water fowl
rather than for fish. This left 13 lakes, for which the data are shown in
Table 30. Using the data with care, i.e. giving little weight to Secchi disc
values other than those for July through September, it appears as though
seven of the lakes showed increases in transparency following treatment, two
probably increased in transparency, and four probably showed no change. In
some of the lakes the results were dramatic and long lasting. For example,
transparency in Scandinavian Lake increased from about 1 m in August 19&5»
before treatment, to an average of 2.3 meters in August 1973 and 1975. and
still averaged 1.4 meters in 1977- In Graham Lake transparency in early
August before treatment was less than 0.3 meters. Three years after
treatment transparency in early August was listed as greater than 1.5 meters.
In other lakes there were large increases that probably lasted for
several years although the data do not exist to verify this. For example, in
Fish Lake the average transparency for July and August before treatment was
about 1.0 meter. In 1961, in the second post-treatment year, the July 6
value was h.O meters. Little Mud Lake on July 31» 1970 was listed as
"turbid". On August 12, 1971, the year following treatment, the transparency
was A.k meters.
Two of the lakes, Ham and Christina, showed no increases in the first
years after treatment but did eventually become much more transparent. We
"by Joseph Shapiro and Eric Smeltzer
190
-------
know of no reasons for the increase other than the treatment. It may be that
those lakes in which planktivorous fish were abundant responded more quickly,
because of the short generation time of the herbivorous zooplankters, than
those lakes in which the effect was caused by the elimination of
bottom-feeding fish. In the latter elimination of the nutrients by the fish
would not be felt immediately, but the lake response would be a gradual
lessening of fertility.
Clear Lake is an interesting example where the lake did respond to
toxaphene, but unfortunately the year after treatment the lake was restocked
with bottom feeders and planktivores and so it did not remain transparent.
in addition to the 12 lakes listed, Minnesota DNR personnel can list a
number of lakes that they recall as having responded to treatment with
increased transparency, as well as some lakes that did not respond in this
way. However, data are not available to document these observations.
In balance, then, it would appear that use of fish toxicants does reduce
algal abundance and that in some cases the effect has been longlasting.
191
-------
Table 30: The effect of fish toxicants on lake transparency: data from the
files of the Minnesota Department of Natural Resources.
Lake,
County,
Date of Treatment,
Toxicant
Date
Secchi
Di sc
Transparency
(ft)
Comments
Scandinavian,
Pope
11-6-67
Toxaphene
6-23-55
8-65		
7-69
8-28-73
5-28-74
8-28-75
1977 mean
4.0
	323_
15.5
7.5
13.0
7.5
4.5
large increase in
transparency lasting at
least 8 years
Fish,
Jackson
11-4-59
7-16-47
8-15-50
8:20:56	
2.9
4.1
	2.7_
large increase in
transparency lasting at
least 2 years
Toxaphene
7-6-61
13.0

Carls,
Scott
9-22-75
6-23-58
1974	mean
1975	mean
2.0
1.9
4.1
transparency increase
after winterkill 1974-5
and after rotenone
Rotenone
1976 mean
6.3
treatment 1975
Little Mud,
7:31:70	
	"turbid"_
transparency increased
Meeker
9-15-70
Rotenone
5-29-71
6-19-71
8-12-71
11.5
13.8
14.5

Graham,
8-2-54
<1.0
transparency increased
Carl ton
9-23-55
Toxaphene
8-7-58
>5.0
greatly, possibly
lasting for at least
3 years
CI ear,
Waseca
10-25-63
Toxaphene
7-22-51
6-24-57
9-19-60
6-11-63
8-12-63
1963 mean
6-16-64
8-4-64
2.5
2.5
1.5
1.3
1.3
	118_
5.0
3.1
transparency increased
siightly
Bear,
Freeborn
11-13-58
Toxaphene
7:23:57	
8-59
6-24-61
	210_
>3.5
1.2
transparency increased
siightly
192
-------
Table 30 Continued
Lake,

Secchi
County,

Disc
Date of Treatment,

Transpa
Toxicant
Date
(ft)
Winona, Lower Lake
7-21-68
2.5
Winona
10-10-72
2.0
9-17-73
9:17:73	
	K7
Rotenone
10-15-73
3.5

5-8-74
14.8

5-15-74
4.8

6-17-74
8.5

6-20-74
9.1

6-26-74
6.0

8-8-74
3.0

9-17-74
1.8
Bud,
7-31-47
2.7
Martin
7-31-54
2.0
9-10-67
4-19-67
1.5
Rotenone
5-22-67
2.8

6-20-67
2.8

7-18-67
10.5

8:23:67	
	5;0

5-15-68
10.0

6-12-68
>13.0

7-16-68
3.4

6-5-70
11.5

1975 mean
3.0
Christina,
?-47
5.0
Douglas
?-50
3.0
11-4-65
?-52
6.0
Toxaphene
?-56
4.0

?-59
0.7

?-61
0.5

?-63
0.8

8-22-67
1.0

?-68
1.6

8-12-72
6.5

9-8-76
3.0
Ham,
8-24-26
2.0
Anoka
9;21-48	
	1:0
10-18-55
8-22-56
1.9
Rotenone
8-1-74
3.7

1975 mean
7.0

1976 mean
6.0
Comments
high transparency in
early summer 1974, but
transparency declined to
pre-treatment levels
in the summer
high transparency in
early summer 1968, but
declined to pre-treatment
levels late in summer
possible delayed
transparency increase
possible delayed
transparency increase
193
-------
Table 30 Continued
Lake,
County,
Date of Treatment
Toxicant	
Anka,
Douglas
11-4-65
Toxaphene
Date
1947 range
§:18:50	
8-22-67
?-68
?-72
Secchi
Disc
Transparency
(ft)	
8
- 10.5
.liL_
1.7
1.7
4.5
Comments
no transparency increase
Winona, Upper Lake
Winona
9-17-73
Rotenone
7-21-68
10-10-72
?:1Z:ZL.
5-15-74
6-6-74
6-13-74
6-20-74
6-26-74
7-8-74
10-7-75
2.3
1.8
_1.3_
4.0
1.9
2.8
2.5
2.1
1.5
12.7
no transparency increase
194
-------
EFFECTS OF ROTENONE IN WIRTH LAKE*
One lake in which we have been able to study the effects of rotenone
treatment in detail is Wirth lake in Minneapolis. This was not our original
intent, which was to study the effects of added piscivores. The situation
became complicated, but the results are of interest.
Wirth Lake (area 16 ha, mean depth 4.3 m) is managed by the Minneapolis
Park and Recreation Board as a swimming and fishing facility. Because of its
location it has long received considerable amounts of urban runoff as well as
periodic inputs of nutrient-1aden Bassett Creek water during floods. In
consequence it is eutrophic with the characteristic features of low
transparency, large populations of algae dominated by blue-greens, anoxic
hypolimnion, etc. Our first study of the lake, where these factors were
described, was done in 197^ as part of a survey of lakes controlled by the
Minneapolis Park Board (Shapiro 197^+) - In 1974, upon learning that the
Minnesota Department of Natural Resources planned to remove all adult
northern pike (Esox 1uc i us) from Lake Harriet, another Park Board lake, in
Spring 1975* we asked them to stock a portion of the pike into Wirth Lake.
This was to determine whether these predators could lessen the abundance of
planktivorous fish in the lake, thereby relieving the larger herbivorous
zooplankters in turn from predation. If this were to occur, the resulting
increase in herbivore activity could materially lower algal abundance,
thereby improving the lake for the recreational activities. Accordingly,
plans were made, and in April, 1975. 212 adult northern pike were stocked
into Wirth lake. The intent was to continue to monitor the appropriate
aspects of the limnology of the lake to evaluate the efficacy of the
procedure. This plan was carried out through 1976, but in 1977 several
things were done to Wirth Lake that made it difficult to follow the original
plan: In August 1977 the Park Board, wishing to alleviate the eutrophic
state of the lake, hired an engineering consultant who built a peat bed
beside the lake and began to pump hypolimnetic lake water through it in an
attempt to remove phosphorus from the water. Unfortunately, at the same
time, the consultant installed eight aerators on the lake bottom and put them
into operation. Because of this, hypolimnetic phosphorus, which normally
would not enter the euphotic zone, became diluted throughout the lake.
Furthermore in September, 1977. the DNR after removing some of the fish from
the lake treated the lake with rotenone. Restocking began in October 1977
and continued into 1978. Examination of stocking records also reveals that
stocking had begun in 197^ and continued after 1975- Thus a great deal has
happened to the lake. However, we have continued to follow phenomena in the
lake in an effort to evaluate the effects of the manipulations. A summary of
the events is presented below:
A.	Studies by Limnological Research Center
197^ — June-September, temp, 02,	transparency, Chi, N, P, algae,
zooplankton, echosounding
1975	~~ June-September,	same	as	above
1976	" May-August,	same	as	above
1977	~~ June-November,	same	as	above
1978	— May-September,	same	as	above
B.	Fish stocked by Minnesota DNR
1974	— 150 northern pike -- yearlings
1975	Spring — 212 northern pike adults
1976	January -- 115 northern pike — yearlings
1976 March — 52 northern pike -- adults
*by Joseph Shapiro
195
-------
C.	Fish removed by net by Minnesota DNR
1977 August-September, 5500 crappies, 4200 sunfish, northern pike
D.	Rotenone treatment by Minnesota DNR
1977 — September 13
E.	Fish restocked by Minnesota DNR
1977	— October, 9^7 walleye fingerlings, 1050 large mouth bass,
fingerlings, 2390 Channel catfish
1978	— Spring, carp, adult (by fishermen)
1978 -- May, 11,200 b1uegill, adults
1978 June, 9600 largemouth bass, fry, 15.000 walleye, fry
F.	Lake level change
In 1977 a barricade was built to prevent Bassett Creek from overflowing
into the lake and to isolate the fish population. The lake level was thus
raised 1.2 m and its area increased.
G.	Artificial circulation and aeration
1977 — August, circulators first turned on.
1977~7B -- Winter, circulation done to prevent winter-kill by low oxygen
under ice resulting from flooding of land area by level increase.
H.	Phosphorus removal
1977 ~~ August, water was pumped through the peat bed but measurements by
the Minnesota Pollution Control Agency indicated little, if any, removal of
phosphorus.
Phytoplankton
Because the purpose of the original stocking with northern pike in 1975
had been to reduce the algal abundance, it is appropriate to look first at
what did happen to the algae.
ChIorophy11
Surface chlorophyll concentrations for 1971978 are shown in Table 31
and plotted in Figure 73» Chlorophyll generally was at moderate levels of
10-25 mg m-3 in Spring and rose during July, August and September to as high
as 69 mg/m-3 in 1974 and to 85 mg m-3 in 1978. The increase in 1978 may have
been occasioned to a degree by phosphorus entering the epilimnion from the
hypolimnion. However, generally, total P in the epilimnion remained constant
through mid-August (Table 31, Fig. 7k) and stratification was stable until
about the same time. Thus there was a progressive increase in the
chlorophyl1/P ratio (Table 30 • As the ratio increased from about 0.2-0.5
towards 1.0, either some other substance, such as nitrogen, was becoming
available by runoff or fixation, and making the use of P more efficient, or
some agent resulting in reduction of chlorophyll was becoming less effective.
Fixation is possible, as species capable of it were present, and in 197^_1975
when nitrate levels were measured they were very low, although during the
same years ammonia was present in the surface waters in significant amounts.
A ch1orophy11-reducing agent may be found in the population of herbivorous
zooplankters and, as seen in Table 32 there is, in recent years, including
1978, a tendency for cladocerans to decline from Spring to mid-August. In
any event, in 1975 and 1978 the pattern was different. In Spring 1978 the
chlorophyll levels were very low, and in both years chlorophyll
concentrations fell precipitously in late summer.
Algal Species
No determinations were made for 1975. but in 197^. 197&. and 1977.
blue-greens were predominant. In 197^. Aphanizomenon flos-aquae and various
species of Anabaena were dominant until September, when Osci1lator ia became
196
-------
abundant. In 1976, Aphanizomenon gave way to a mixture, Aphanizomenon and
Anabaenops i s. then of Aphan i zomenon, Anabaenops i s and Osc i 1 1ator i a. In
June-August, 1977, Osci1lator'a was abundant, followed by Aphanizomenon and
Anabaena. However, in 1978 the picture was different. Through early July
blue-greens were very scarce, with most of the few algae present being greens
— cryptomonads or Schroeder i a. By July 20 however, flakes of Aphan i zomenon
f1os-aquae dominated the now-abundant algae, and continued to do so along
with some Cerat i urn and Mi crocyst i s until August 20 when the flakes
disappeared, reappearing again as smaller flakes on August 30.
Transparency
The changes in algal abundance, and to an extent in species composition,
manifested themselves in changes in Secchi disc transparency, as shown in
Table 31 arid Figure 75- In 1974, 1975 and 1977, summer transparencies were
less than a meter and as low as 0.6 m. However 1976, and especially 1978,
transparencies were very much greater. In fact, in 1978, Secchi disc
transparency exceeded 2 m into August and was as high as 4.6 m on July 4.
One apparent result was an increase in macrophyte abundance with reports of
floating masses of weeds.
Total phosphorus
Surface total P concentrations are shown in Table 31 and Figure 7^*
Until August 1977* concentrations seemed relatively stable at 40-60 mg m~3.
As 1975 had more rainfall than either 197^ or 1976, its higher values
probably reflect the greater amount of urban runoff. However there is no
doubt that the very high values in August-September 1977. and from Hay onward
in 1978, were brought about by the air-induced circulation. Not only did
they occur after circulation began, but Wirth Lake has exceptionally high
concentrations of P in its hypolimnion. Figure 76 shows a plot of total P on
September 15, 1975- In fact at one time because of the abundance of P in the
hypolimnion, we considered using an Olzewski tube (Olzewski 1959) to drain P
from the hypolimnion and export it to Bassett Creek.
Dissolved oxygen
Dissolved oxygen data are shown in Figure 77» where vertical profiles
for mid-August are drawn for the three years when the data were taken.
Clearly, operation of the aeration devices resulted in the deeper waters
becoming aerobic, and in a considerably greater fraction of the lake
sediments being exposed to oxygenated water in 1977 and 1978 than in
1974-1976.
Zooplank ton
Examination of Table 32 shows that Wirth Lake has a varied zooplankton
population. A total of 11 cladocerans have been found, of which 6 are species
of Daphnia. In addition, 4 copepods and at least 11 rotifers make up the
community. Because of their role in controlling the algae, the larger
Daphnia are of particular interest. Figure 78 shows the abundance of Daphnia
qa1eata and Daphnia pulex from 1974-1978. Daphnia qa1eata was present in all
five years, generally most abundant in Spring, reached its highest population
of 486,000 m-2 in May 1976 and its lowest -- undetectable — in August, 1978.
Mean body length in this species has been as follows:
197
-------
Table 31: Transparency, phosphorus and chlorophyll in Wirth Lake, 1974-1978.
Secchi Disc,
m
1974
June 26	1.13
July 16	.64
August 14	.61
September	10 .67
1975
June 9	1.22
July 1	.70
July 23	.79
August 15	.76
September	9 1.07
September	15
Chi
Total P a^
mg m-3 mg m-3 Chi/P
43	12.3	.29
42	31.5	.75
42	49.4	1.18
80	68.7	.86
60	14.0	.23
58	33.0	.57
38	20.0	.53
59
52	10.0	.19
68
1976
May 18	3.14
June 18	1.52
July 15	1.52
August 17	.61
1977
June 14	.76
July 7	.70
August 1	1.01
August 24	.85
September 23	.88
November 8	.91
46	22.4	.49
44	19.9	.45
36	16.7	.46
67	59.1	.88
58	22.0	.38
55	27.0	.49
47	38.0	.81
160	43.0	.27
176	65.0	.37
182	80.0	.96
1978
May 29	3.66
June 14	3.51
July 4	4.57
July 20	2.13
August 2	2.29
August 18	1.37
September 15	1.46
148	4.1	.03
220	8.3	.04
224	3.6	.02
204	33.8	.17
227	43.6	.19
215	85.0	.40
317	26.0	.08
198
-------
T	1
.	i
cr
E
U ^ r. 6 "7 6 ? ' b ' k ' 6~ 9 ? " 6 7 6 9* 6 7 8" 9~~0 " 6 ~ 7 8 ~9
c/>	1974	1975	1976	IS77	1978
Fiq. 73. (Top) Chlorophyll a concentrations
in Wirth L., 1974 - 1978.
Fig. 74. (Center) Total phosphorus concen-
trations in Wirth L., 1974 - 1978.
Fig. 75. (Bottom) Secchi disc transparencies
in Wirth L., 1974 - 1978.
199
-------
to
O
O
0
- 3
-C 4
"q.
CD
O 5
Total P mg I
.5	1.0
Fig. 76. Total phosphorus profil
L., Sept. 15, 1975.
Dissolved 02 mg I 1
E
JC.
Q.
0)
Q
Fig. 77. Dissolved oxygen profiles in Wirth
L. before (1974 & 1975) and after
(1978) aeration.
-------
Table 32: Zooplankton (thousands of individuals m-2) in Wirth Lake, 1974-1978.


1974



1975



1976


June
July Aug Sept
June July

Aug
Sept May
June July Aug
CLADOCERA:
26
16
14
10
9
1
23
15
9
18
18
15
17













Daphnia pulex




13
9
2






Daphnia galeata
27
7
24
152
125
169
93
26
46
486
108
88
58
Daphnia retrocurva


4

4




7
1
8
8
Daphnia longiremis




15




183

3

Daphnia parvula




4






3

Daphnia ambigua









65



*Bosmina longirostris
151
3
110
359
275
594
5
362
307
1131
4

175
Ceriodaphnia sp.
25
209
420
182
2
2
4
2
32
46


60
*Chydorus sphaericus
47
91
20
43

6
40
7
52

6
23
142
Diaphanosoma 1euctenbergianum
3
10
157
113

4
44
155
115

6
248
258
*Leptodora kindtii













Total Cladocera less *
55
226
605
447
163
184
143
183
193
787
115
350
384
COPEPODA:













Cyclops bicuspidatus
30


17
218
4


9
135
6
2
22
Cyclops vernal is

7


5

4
4
14



29
Mesocyclops edax
130
33
43
107
32
165
84
57
83
32
117
110
315
Diaptomus siciloides
66

55
36
42
86
45
81
129
23
24
84
39
Nauplii (copepodids)
3
333
177
197
51
117
58
44
195
303
223
488
416
ROTIFERA:













Asplanchna

23


51
15



3



Conochilus

688

48

1506







Filinia

20





2



1
143
Kellicottia longispina
2
3

54





166
42
3
2
Keratella cochlearis

20
4
3
35
6

2
3
8015
45
191
601
Keratella quadrata
2



9




42
16


Polyarthra vulgaris

3
24
5





3
35
85
195
Trichocerca multicrinis

7



2







Trichocera similis












12
Platyias patulus













Brachionus sp.













-------
Table 32 Continued
CLADOCERA:
Daphnia pulex
Daphnia galeata
Daphnia retrocurva
Daphnia longiremis
Daphnia parvula
Daphnia ambigua
*Bosmina longirostris
Ceriodaphnia sp.
*Chydorus spaericus
Diaphanosoma 1euctenbergianum
*Leptodora kindtii
N3
m Total Cladocera less *
COPEPODA:
Cyclops bicuspidatus
Cyclops vernal is
Mesocyclops edax
Diaptomus siciloides
Nauplii (copepodids)
ROTIFERA:
Asplanchna
Conochi1 us
F i 1 i n i a
Kellicottia longispina
Keratella cochlearis
Keratella quadrata
Polyarthra vulgaris
Trichocerca multicrinis
Trichocera similis
Platyias patulus
Brachionus sp.
June July Aug
14 7 1
51
33
35
24
50
142


3
9

3
21

32
539
97
61


19
42
50
142


3
665
230
386


3
30
52
176
88
78
43
312
240
473


5
185
50
16

105
104
1132
190
524
15
10

9
10
3


37

14

977
1978
24
Sept
23
Nov May
8 29
June
14
July
4 20
Aug
2
18
Sept
15



145
441
182
471
233
209

79

56
143
12
9
17
6


130





3



66

6
2
2
2
20
9
5
103
9









158



2

12
9
32
53
149





3
1
11
142









5
367
0
56
288
453
191
494
240
220
142
28
12
1575
83


3



222
2
3

45
52
64
10
5
11
98



17
56
84
53
27
319
436

97
143
143
871
1294
329
268
500
6








5237








11
5
3








66
949









468
2

5
2



11




88
7




566
247





3
6
510
36
28





O


58
145
-------
-i	t	1	r
400-

A
1
1 1 I 1
1
i
. 1
i
i
i i i
i i i i
|
t |l D pui$*
!1 .1
U i i
i i ; 1
1 1 1 ;

- i
i
i
¦ i
-
! \ « 1
! ! 1
J \
i 1 ' "
\
\
\
\
V
V
			
- k -
¦
V
/
. t
\
\ i -
\ D. go feat a s
\ \
\
6 : e
1974
6 ! e 9 5 6 .'8
1975	1976
6 7 S 9
1977
6 7 8 9
1978
Fig. 78. Abundance of Daphnia pulex and Daphnia galeata in Wirth L.,
1974 - 1978. (Rotenone treatment was done on Sept. 13, 1977.)
203
-------
Year	ttean Body Length, mm
19714	.883 (July and August only)
1975	-992 (June, July and August)
1976	.878 (June, July and August)
1977	*791 (June, July and August)
1978	1.00 (May only)
Too few specimens were present in 1978 to determine body lengths for other
than the May sample. Daphnia pu1 ex was found in only two of the five years
1975 when it occurred at a maximum of 13,000 m-2 and was gone by
mid-August, and 1978, when it occurred at up to 1*71.000 m-2 disappearing
probably in late August. Body lengths were considerably greater than those
for Daphnia qaleata, as follows:
Date	1978 Mean Body length, mm
5/39	1.39
6/14	l.kk
7A	1.70
7/10	1.38
8/7	1.46
8/18	1.32
Ten days following rotenone treatment on September 13, 1977* no Daphni a were
found -- indeed the only zooplankters recorded were a few Cyclops,
Mesocyclops, and Keratella. However, by November 8, 1977. Daphnia qaleata
was found in respectable numbers and Bosmi na and D i aptomus had recovered as
well. Cyc1 ops b i cuspi datus was present in very great abundance, possibly
because of the abundance of food and absence of competitors and predators.
Several species seem to show a distribution inverse to that of [h pulex.
For example, in 1978, Bosm i na and D i aphanosoma were rare until D^ pu 1 ex
disappeared. Similarly, Cer i odaphn i a was abundant except for 1975 when it
was sparse, and 1978 when it was absent. Kel1icotia was also present except
for the two years when D_^ pu1 ex appeared, and Chydorus sphaer icus was rare in
the early part of 1978. Even. JK qaleata was in least abundance in 1978 when
D. pulex was abundant.
DISCUSSION
Wirth Lake has undergone a variety of manipulations and has exhibited a
number of changes. The questions are, what are the causal relationships, and
what can be learned? The most obvious change in the lake was the exceedingly
small algal population during early 1978, which resulted in the very great
transparencies during the same period. Two possible explanations present
themselves -- either the growth rate of the algae was low, or the loss rate
was high. It is unlikely that the growth rate was low. Rainfall and
therefore runoff to the lake during early 1978 were above average and, in
addition, operation of the circulators increased the nutrient concentrations.
Furthermore, it is not likely that circulation caused light limitation as
transparency was very high, and the algae increased greatly during July and
August even though circulation continued and transparency declined. The most
likely explanation lies in the grazing activities of the very large
population of [h pulex, which averaged 256,000 m-2 or, considering the depth
of the 20oplankton tow (8 m) and the completely mixed water column, 32
individuals per liter. This concentration is within the range of that capable
of removing from the water essentially all filamentous blue-greens (Lynch and
204
-------
Shapiro 1980, see Section IV.A.). In fact, very few blue-greens were present
during the early part of 1978, being replaced by green algae that must have
been growing very rapidly to be able to maintain themselves in the face of
such intense grazing. This is similar to what happened in Severson Lake,
Minnesota, as a result of a large population of	pulex (Schindler and
Comita 1972). Low phytoplankton standing crops with no blue-greens but with
diatoms and green flagellates were reported.
Two of the manipulations may have contributed to the large population of
D. pulex in 1978 -- the circulation and the rotenone treatment. The
circulation may have helped because, as shown by echosoundings taken in Wirth
Lake in June 1977 and as expected from the dissolved oxygen distribution, the
fish population had been confined to the epilimnion. Also, because of the
anaerobic conditions in the hypolimnion, the Daphnia, being without a refuge
would also be in the epilimnion and subject to predation. Circulation,
because of dilution of both populations, would be expected to reduce this
predation and result in an increase in the Daphnia population as it did in
Lake Calhoun during circulation there (Shapiro and Pfannkuch 1973)- Had
sampling been done in September 1977, after circulation began, but before the
rotenone treatment, any such increase in Daphnia could have been detected.
Unfortunately, this was not done and so the question is moot. On the other
hand, there is little doubt that the elimination of the fish (of which 50%
were crappies, 25% bluegill sunfish, 15% carp, and 10% perch, bullhead,
sucker, northern pike, and largemouth bass) by the rotenone treatment was
largely responsible for allowing JK pulex to exist and become abundant in the
lake in 1978. The appearance of this species is common following reduction
of predation on it because of either winter-kill (Schindler and Comita 1972;
Shapiro, Section IVC) or circulation (Shapiro, and Pfannkuch 1973 Lake
Calhoun); as is its rapid disappearance following the addition of predators
(Lynch 1979)• What makes pulex so sensitive is its size and therefore its
visibility to the fish. For example the IK pulex in Wirth Lake in 1975
averaged 1.^5 mm body length -- compared with the average of O.89 mm for the
0. qaleata in the lake from 197^~1977- On Hay 29, 1978 IK qa1eata were
larger than usual, averaging 1.00 mm long -- higher than on May 18, 1976
(0.86 mm), or on June 14, 1977 (0.82 mm). One interesting feature is that 0.
qa1eata did not reach very great abundance in November 1977, following its
recovery from rotenone treatment, even though it seemingly had few
competitors. Furthermore, JK qa1eata has existed in the lake continuously in
the face of fish predation. It thus seems as though this species, although
it may respond somewhat by increasing its size in the absence of predation
is, by contrast with IK pu1 ex, controlled by factors other than predation, at
least under normal predator levels.
The original intent in studying Wirth Lake was, as noted, to evaluate
the effect of adding carnivorous fish. Although the later manipulations did
result in more dramatic changes, there is evidence that the stocking with
northern pike may have had some effect. Northern pike had been added in
1974.	1975 and 1976. These stockings, particularly of adults, may have been
efficacious in reducing the abundance of one or more planktivores. In July
1975,	northern pike in Wirth Lake were rated by the DNR as being few in
number, by September 1975 they were near the state average and by August 1977
they were well above the average. At the same time, perch which were
abundant in September 1975 were, by August 1977. reduced in numbers. One
might have expected such a progression to lead eventually to the appearance
of D_;_ pulex in the lake, assuming the perch were their chief predator. The
rotenone treatment made it impossible to tell. Strangely enough, D. pulex
did appear — but in 1975 — suggesting a great decline in planktivore
205
-------
abundance or activity in that year. Further evidence of such a decline lies
in the fact that JK qa1eata in 1975 averaged .99 mm in length, compared with
.85 in 197^. 1976, and 1977- Chlorophyll a also was different in 1975.
showing an uncharacteristic decline (Fig. 73). To what extent these
phenomena are related to the stocking of northern pike is unclear. It is
possible that the pike stocked in 197^ and 1975 were the cause for the D.
pulex increase in 1975. but the uncharacteristic decline in chlorophyll came
about during the period when th pulex was rare or absent.
The other anomalous relationship occurred in summer 1978. Until July k,
and probably for some time afterwards, blue-green, and in fact all algae,
were scarce, probably because of the grazing pressure of the large population
of |L pu1 ex. However, on July 20 chlorophyll levels were high and remained
so until the end of sampling on September 15- The explanation lies in the
ability of Aphanizomenon to maintain so-called "grass-blade" blooms in the
presence of large Daphn i a. such as [h pul ex. Lynch (1980) believes that once
Aphanizomenon colonies reach a size large enough to escape predation by
Daphnia, they prosper as the Paphnia remove algal competitors and release
nutrients from them. In order for the colonies to become large enough to
escape predation in the first place the algae must be able to grow where they
will not be grazed. Lynch believes that this may occur at the sediment
surface where new colonies may form from akinetes. This idea is strengthened
by the failure of Aphani2omenon to form such large colonies in enclosures,
where the water is isolated from the sediments, or in places where the
hypolimnion is anoxic. In other words, two circumstances favor the
production of large ungrazable Aphan i zomenon colonies -- many large Daphn i a
such as IK pu1 ex, and an oxic hypolimnion. Both were present in Wirth
Lake in 1978, the first as a result of the rotenone, the second because of
artificial circulation. It is possible that had artificial circulation not
been done, the Daphnia would have been able to control the algae beyond
mid-July. Interestingly, on September 15. 1978, when no Paphni a were present,
the grass-blade colonies had disappeared, and the transparency which had been
high despite the abundant chlorophyll present in the large algal colonies,
increased only slightly even though the chlorophyll declined to less than a
third of that on August 15*
CONCLUSIONS
The conclusion from this study is that, although Wirth Lake was
beginning to respond to the stocking of northern pike, in that the number of
perch declined, the greatest influence on the lake resulted from the rotenone
treatment. By eliminating planktivorous fish [h pulex was allowed to
increase and began to control the algal population to the point where
transparency increased greatly despite the large increase in phosphorus
concentrations resulting from circulation. The circulation had two
detrimental effects: it made it more difficult for the peat bed to remove
any significant fraction of the phosphorus in the lake and it allowed
Aphan i zomenon to form grass blade blooms.
REFERENCES
Lynch, M. 1979- Predation, competition, and zooplankton community structure
-- an experimental study. Limnol. Oceanogr. 2k:25}~2(>2.
206
-------
	. 1980. Aphanizomenon blooms: Alternate control and cultivation by
Daphnia pulex. p. 299-304.j_n ASLO Special Symp. No. 3- W.C. Kerfoot,
(ed.). p. 299-304. University Press of New England. Hanover, New
Hampshire. 793 PP>
Lynch, M. and J. Shapiro. I98O. Predation, zooplankton, grazing, and
phytoplankton community composition. Limnol. Oceanogr. In Press.
Olzewski, P. 1959- Attempt to siphon hypolimnetic water out of a lake. Verh.
Internat. Verein. Limnol. 14:855-
Schindler, D. W. and G. W. Comita. 1972. The dependence of primary
production upon physical and chemical factors in a small, senescing
lake, including the effects of complete winter oxygen depletion. Arch.
Hydrobiol. &9:413~451-
Shapiro, J. and H.O. Pfannkuch. 1973* The Minneapolis Chain of Lakes. A
study of urban drainage and its effects. Interim Report No. 9«
Limnological Research Center. University of Minnesota.
Shapiro, J. 1974. Report to the City of Minneapolis Park and Recreation
Board on the condition of lakes in Minneapolis, 1974. Interim Report
No. 12, Limnological Research Center. University of Minnesota.
207
-------
THE INFLUENCE OF FISH ON THE ABUNDANCE OF ALGAE IN CLEAR LAKE, MINNESOTA*
It has been observed that lakes often respond to fish-kills with reduced
phytoplankton abundance the following year. To study the ecological
mechanisms underlying this phenomenon, a whole-lake experiment was designed
involving Clear Lake, Minnesota. The purpose of the study was to assess the
relative importance of two possible mechanisms linking fish-kills to algal
dec 1i nes:
1.	Removal of planktivorous fish results in increased zooplankton
abundance and/or increased dominance by large-bodied 2ooplankton species.
Algal abundance declines in response to increased grazing pressure.
2.	Removal of benthivorous fish eliminates a source of nutrient
regeneration from the sediments to the water. Algal abundance declines in
response to a reduced supply of nutrients.
Clear Lake (21 ha, 2 m max. depth) is divided into two basins, North (N)
and South (S), by a road and causeway. Clear S, in 1978, contained
populations of benthivorous black bullheads (I etalurus melas) and
planktivorous fathead minnows (Pimephales promelas)¦ Clear N, in 1978, was
void of fish as a result of a previous winter-kill. Both basins were treated
with the fish toxicant rotenone in Fall, 1978, so that fish were virtually
absent in 1979* Thus, the natural fish-kill during winter in Clear N and the
artificial fish-kill by rotenone created a controlled experimental situation
in which the response of the two basins to fish removal could be studied.
The removal of fish from Clear Lake both by winter-kill and by rotenone
treatment resulted in sharp declines in algal abundance. Increased grazing
of algae by zooplankton in the absence of fish was an important mechanism
leading to the alga! decline. This conclusion was supported by the following
observat i ons.
1.	The zooplankton community was dominated by large herbivores in the
f i sh-k ill si tuat i ons.
2.	Algal losses to non-respiratory causes increased in importance in
the production and loss dynamics of the algal community after fish-kill.
3.	Non-respiratory loss rates were correlated with herbivorous
zooplankton biomass.
h. The population of Daphn i a pu1 ex present in Clear S after fish-kill
was physically capable of filtering food particles at a rate sufficient to
account for the observed non-respiratory losses suffered by the algae.
5- The specific rate of photosynthesis increased after fish-kill.
6. There was a shift in the nutrient pools in favor of the dissolved
fractions after fish-kill.
There was, however, some evidence against the importance of grazing in
the fish-kill situations.
1.	No inverse correlations were observed (within a season) between
algal and zooplankton abundance.
2.	Bag experiments failed to demonstrate a positive growth response by
algae to release from grazing.
Reduced levels of TP and TN were observed in the absence of fish. The
decline in nutrient levels may have accounted for about half of the extent of
the algal decline in Clear S following fish-kill. However, the elimination
of nutrient regeneration from benthivorous fish excretion could not have
caused the observed nutrient decline. This conclusion is supported by the
following evidence.
1. Changes in TP levels during 1978 were synchronous between North and
South basins, and were correlated with rainfall.
>'
-------
2. The nutrient decline following rotenone treatment occurred in Clear
N, the control basin, as well as in Clear S.
3- The quantity of phosphorus excreted by bullheads to Clear S during
1978 was a very small fraction of the total annual load of phosphorus to that
bas in.
h. The decline in the quantity of phosphorus present in Clear S between
1978 and 1979 could not be accounted for by the loss of fish excretion as a
phosphorus source.
The possibility remains that fish excretion could be an important
mechanism linking fish kills to reduced algal abundance in lakes with more
substantial benthivorous fish populations.
209
-------
IV.C. EFFECTS OF WINTER-KILL ON MINNESOTA LAKES IN 1978-1979*
Because the winter of 1978-1979 was particularly long, with snow
covering the ice for most of the period, a considerable number of Minnesota
Lakes suffered winter-kill. Therefore we took the opportunity to study a
number of these lakes along with what appeared to be control non winter-kill
lakes to determine the effects of the fish removals from the ecosystem. The
lakes are listed in Table 33 along with notes on their presumed condition as
provided by Duane Shodeen, Metro Area Fisheries Manager. Each lake was
sampled four times between June 12 and August 30> 1979• Measurements made
included Secchi disc transparency; temperature and dissolved oxygen (by
probe); chlorophyll, algae, and total phosphorus (upper 3 meters by tube
integrator); and zooplankton (vertical haul 17 cm Wisconsin net). Recording
echo sounder scans were done to evaluate the size of the fish population.
CHARACTERISTICS OF THE LAKES
Because very little previous data was available for most of these lakes,
a direct comparison with earlier years was not possible for most. Rather
than before and after comparisons therefore, we looked for characteristics
common to winter-kill lakes. Specifically, we were interested in:
1.	Zooplankton populations -- do winter-kill lakes have different
and/or larger herbivores?
2.	Chlorophyll/TP ratios — are they lower in lakes with larger
herbivores?
3.	Transparency -- is it generally greater in winter-kill lakes?
k. Algal populations -- are they different in winter-kill lakes?
Because test nettings were not done in all of the lakes, and because the echo
sounder malfunctioned, we do not actually know which lakes had complete kills
or which had partial or no kills. Therefore there is a danger of circularity
in the interpretation. Nevertheless, it is instructive to analyze the
results, which are summarized in Table 3^»
Zooplankton
Of the eight lakes suspected of hard winter-kill, (Table 33) three of
them, Rebecca, Hydes, and Powderhorn were dominated by Daphnia pulex to the
virtual exclusion of all other cladocerans. Although Rebecca and Hydes did
have diaptomids present, they were absent from Powderhorn Lake. O'Dowd Lake
had Daphn i a pulex on one date when it shared dominance with Daphn i a gaIeata
mendotae. The lake had, in addition to D i aptomus, Cer iodaphnia and
D i aphanosoma¦ Thus four of the eight suspected hard-kill lakes had D. pu1 ex
in them, and in three of the lakes these were the dominant form. Furthermore,
in these three the Daphnia pulex were abundant, averaging 19~33 1-1
throughout the water column. In the four remaining suspected hard-kill lakes
the main species of Daphnia were D. retrocurva, D. parvula or D. galeata.
In the lakes suspected of only partial or no winter-kill the main
species of Daphnia were 0. qa1eata and D. retrocurva. However in four of the
lakes, D. pu1 ex made its appearance -- as a co-dominant in Sarah Lake and on
one date in Bass Lake; and as the sole dominant in Ryan Lake, at a
concentration of 2 1-1 and in Cedar Lake at a concentration of 20 1-1. Cedar
Lake was listed as a control but Duane Shodeen had noted that it was low in
panfish. Thus in four lakes suspected of hard winter-kill, in two lakes
suspected of partial winter-kill, and in one lake known to be low in panfish
D. pulex was the dominant or abundant. This situation presumably resulted
*by Joseph Shapiro
210
-------
Table 33: Lakes examined in winterkill study.
Lake
County
Comments
Presumed "Hard-Kill" Lakes
Rebecca
Snail
01Dowd
Thole
Eagl e
Hydes
Sunset
Powderhorn
Hennepin
Ramsey
Scott
Scott
Carver
Carver
Washington
Hennepin
infrequent kills
infrequent kills
killed in '75 back to
killed in '75 back to
frequent recent kills
infrequent kills
killed in '75
capacity
capacity
Presumed "Partial-Kill" Lakes
Owasso	Ramsey
Josephine	Ramsey
Bass	Hennepin
Ryan	Hennepin
¦ined No Kill Control Lakes
Sarah
Turtle
Cedar	low in panfish, aerated in '79
Fish	high in panfish
Bavaria
Pierson
CI ear
211
-------
from the reduction of fish predation. Only in Sarah Lake was the presence of
D. pulex unexpected.
The case of Eagle Lake is unique. Although it had a consistently large
population of Daphni a, averaging 5^ 1 -1• the only species present were D.
qaleata and 0. parvula¦ Thus the size of the population suggests that the
lake did winter-kill but the species suggest otherwise. The average number
of Daphnia among the other lakes, omitting those seven in which D. pulex
dominated, was 9 1~1-
Chlorophyll/phosphorus Ratios
Many of the data in this report suggest that Daphnia, especially D.
pu1 ex, are effective in reducing algal populations. Accordingly, it is of
interest to compare chlorophyll/phosphorus ratios in the lakes containing
many Daphnia with the ratios in the other lakes. Among the four lakes having
D. pul ex as dominant and in abundance, Rebecca, Hydes, Powderhorn, and Cedar,
the mean chlorophyll/TP was .132 + .046. If the single exceptionally high
values (Table 3^) for Rebecca, Hydes, and Powderhorn are omitted, the ratio
is even lower, .064 + .030. Among the remaining fifteen lakes chl/TP was .362
+ .136. Interestingly, Eagle Lake, despite its great abundance of D. qaleata
and D. parvula. had a high ch1orophy11/TP ratio (.313)- The data suggest,
therefore, that the lakes with predominantly D. pulex have significantly
lower algal crops than the other lakes. This shows up well when the data are
plotted as chlorophyll vs. phosphorus in Figure 79* The D. pu1 ex lakes fall
on a line very different in slope from those lakes having other species of
Daphn i a.
Transparency
The transparency of a lake is determined only in part by the chlorophyll
content of the water. Nevertheless, comparison of the four D. pulex lakes
with the other fifteen lakes shows the former to have a mean Secchi disc
transparency of 2.07 + -57 m (2.30 + .82 m if the single exceptionally low
values for Rebecca and Powderhorn lakes are omitted) as compared with I.63 +
.68 m for the non-D. pulex lakes. Again Eagle Lake, despite its large
population of Daphnia has a -ow transparency (0.57 m).
A limited amount of earlier transparency data was available for direct
comparison as six of the lakes had been involved in a Secchi disc survey
within the period 1973-1978. These lakes are compared for the period
July-August in Table 35- The three "control" lakes, Sarah, Turtle, and Cedar
had very similar transparencies in both periods suggesting that their
herbivore populations were stable. The three "partial winter-kill lakes",
Owasso, Ryan, and Bass, doubled in transparency. As noted earlier, Ryan and
Bass did have low concentrations of D. pulex, but Owasso had none. Data for
Bass and Owasso are lacking, but observations on Ryan Lake in 1977 and 1978
showed Daphn i a virtually absent from the lake.
A1 qae
The algal samples have not been investigated in detail, but certain
observations have been made. In particular we have been interested in the
occurrence of so-called "flake" blooms of Aphanizomenon and, as such flakes
tend to break apart during preservation, samples of water were also filtered
through glass fiber filters to be used for assessment of the abundance of the
flakes. The results of these filtrations are shown in Table 36. Although
212
-------
Table 34: Chemical and biological data for lakes examined in the winterkill
study. Daphnia species are: P = pulex; R = retrocurva;
G = galeata; V = parvula; u = small unknown, not pulex.


Rebecca


Snail


6-_27
7-20
8-_8
8-30
6-13
7-4 7-27
8-13
T.P.
106
37
63
205
33
35 31
37
Chi .a
33.5
3.4
5.1
25,0
7.5
9.9 9.5
13.1
S.D.
.94
4.60
4.11
2.29
4.05
2.29 2.16
2.29
D.O.
.10
.05
.05

.10
.30 .10

Daph. sp.
RP
PG
P
P
RG
RG u
u
Daph. #
38
13
n
14
26
5 3
.3
Chi/TP
.32
.09
.08
.12
.23
.28 .31
.35


01Dowd


Thole


6-25
7-18
8-6
8-29
6-25
7-18 8-6
8-29
T.P.
60
62
no
147
114
175 172
162
Chi .a
29.9
31.1
89.0
118
58.3
50.5 91.1
96.9
S.D.""
.98
1.01
.61
.61
.88
.64 .55
.43
D.O.
.50
.05
.05

3.0
.10 .20

Daph. sp.
G
GP
G
u
u


Daph. #
28
8
6
5
1
0 0
0
Chi/TP
.49
.50
.81
.80
.51
.29 .53
.60


Eagl e


Hydes


6-18
7-13
8^3
8-22

7-13 8^3
8-29
T.P.
315
301
427
447
450
486 522
582
Chi .a
69
71
193
151
1.6
104 51
22
S.D."
1 .07
.52
.27
.43
2.67
2.29 1.22
2.32
D.O.
.10
.10
.10

5.6
.10 .05

Daph. sp.
u
VG
6
GV
P
P P
P
Daoh. #
128
39
18
29
49
17 46
21
Chi/TP
.22
.24
.45
.34
.004
.21 .10
.04
T.P. = total phosphorus mg m~3; Chi a = chlorophyll mg m"3
S.D. = Secchi disc transparency, m; D.O. = dissolved oxygen mg 1
Daph. # = # of Daj)hnia per liter of the whole water column
213
-------
Table 34: Continued



Sunset


Josephine


6-21
7^8
7-30
8-21
6-13
hi
7-27
8-13
T. P.
161
78
104
82
37
59
55
65
Chi.a
23.3
20.2
61.7
38.4
6.7
11.1
14.8
3.6
S.D.
2.35
1.55
.79
1.22
2.29
1.34
.94
1.34
D.O.
7.0
5.1
8.2

.01
.10
.05

Daph. sp.
G
G

u
GR
GR
RG
RG
Daph. 4
114
6
0
.5
14
2
2
3
Chl/TP
.14
.26
.59
.47
.18
.19
.27
.06


Powderhorri


Bass


6-21
7-8
7-30
8_-21_
6-27
hi0
8-8
8-30
T. P.
283
310
376
133
40
54
73
109
Chi .a
2.9
225
8.0
4.7
4.2
19.0
29.4
58.5

2.56
.91
1.49
1.71
4.51
1.83
1.28
.85

1.1
0.4
.05

.10
.05
.05

sp-
P
P
u
P
G
GP
u
u
§
24
32
.03
32
30
7
4
5
Chl/TP
.01
.73
.02
.04
.11
.35
.40
.54


Owasso


Ryan


6£1_3
7-3
7-27
Ml
til
ZilP_
8-8
8=30
T.P.
60
62
79
91
139
104
108
122
Chi .a
13.4
8.0
9.5
14.7
13.4
19.8
38.9
39.9
S.D.
2.90
3.72
2.47
1.86
2.56
1.52
1.10
1.10
D.O.
.10
.20
.05

.05
.05
.10

Daph. sp.
GR
G
GR
u
P
P
u
u
Daph. #
26
13
.6
1
7
1
.5
.5
Chl/TP
.22
.13
.12
.16
.10
.19
.36
.33
214
-------
Table 34: Continued
Sarah

6-27
7-20
8-8
T.P.
123
89
75
Chi. a
12.5
47.3
53.2
S.D.
3.32
.79
.82
D.O.
.10
.05
.01
Daph. sp.
PG
PG
G
Daph. #
8
2
2
Chl/TP
.10
.53
.71



Turtle

6-13
7-4
7-27
T.P.

24
20
Chi .a

5.6
4.7
S.D.

2.01
1.98
D.O.

3.9
.10
Daph. sp.

R
R
Daph. #

5
4
Chl/TP

.23
.24


Cedar

6-25
7-18
8-6
T.r
443
562

Ch; .
59.2
44.1

S.D. ~
1.52
1.58

D.O.
7.4
2.1

Daph. sp.
P
P

Daph. #
10
46

Chl/TP
".13
.08



Clear

6-21
7-8
7-30
T.P.
89
62
45
Chi.a
28.7
12.6
10.8
S.D.
.98
1.16
1.52
D.O.
5.0
.05
.05
Daph. sp.
G
GR
GR
Daph. tr
22
3
7
Chl/TP
.32
.20
.24


Fish

8-30
6-25
7-18
8-6
8-29
125
44
35
46
47
120
22.6
16.2
27.6
27.2
.79
1.49
1.19
.88
1.25

.05
.05
.05

PG
G
RG
R
R
2
7
1
3
17
.96
.51
.46
.60
.58


Bavaria

8-21
6-18
7-13
8-3
8-22
23

48
33

9.4

27.5
3.4

1.68

1.07
3.47



.20
.05

RG

GR
RG

8

2
5

.41

.57
.10



Pierson

8-29
6-18
7-13
8-3
8-22
654
32
28
27
25
37.9
13.4
6.2

7.8
1.43
2.04
2.32
2.71
2.53

6.7
.25

.05
P
G
G
RG
RG
4
5
3
6
2
.06
.42
.22

.31
8-21
60
30.1
1.25
RG
6
.50
215
-------
* PowJenorn
/	— ~"t * Cedor
- r- "" "	Hydes
r -.30?
Rebecca
	i. .
200
Total P mg m~
Chlorophyll/TP relations in lakes with (X) and without (•)
abundant Daphnia pulex.
216
-------
Table 35: Secchi disc transparencies of
for the period 1973-1978, and
Values are July-August means.
partial winter-kill and control lakes
for 1979, following winter-kill:
Transparency (in)
1973-78	1979
Control Lakes
Sarah	0.99	0.80
Cedar	1.22	1.51
Turtle	1.83	1.90
Partial-Kill Lakes
Ryan	0.61	1.24
Bass	0.59	1.32
Owasso	1.33	2.68
Table 36: Abundance of Aphanizomenon flakes in the lakes containing them
1979. 0 = absent, 1 = very sparse, 2 = moderate, 3 = very
abundant.
in
Lake
Pier son
Sarah
Thole
Hydes
Powderhorn
Cedar
Sampling period (see Table 2 for date)
A	B	C	D
0
1
1
0
1
3
0
0
0
2
3
2
0
0
0
3
2
1
0
0
2
2
1
217
-------
six of the nineteen lakes had some flakes during the sampling period, only
three of them had flakes in abundance. Interestingly, these three lakes,
Hydes, Powderhorn, and Cedar were three of the four lakes that were dominated
by abundant Daphnia pulex (means = 33. 22, and 20 1-1 respectively). The
fourth such lake, Rebecca, in which D. pulex averaged 19 1 — 1» had no flakes.
However, Sarah Lake which had a few flakes early in the season also had D.
pulex.
The lakes characterized by abundant flakes were also those with (except
for Eagle Lake) the highest total phosphorus - 5^0, 278, and 553 ug 1-1
respectively. Again Rebecca Lake had a low value -- 103 ug !"!• Eagle Lake,
despite its high total phosphorus and its abundance of Daphni a (5^ 1-1) had
no flakes. However, the Daphni a were, as noted, D. qaleata and D. parvula,
not D. pulex. Thus lakes with abundant flakes appear to be characterized by
abundant D. pu1 ex and high total phosphorus.
Finally, as noted by Lynch (1980), dissolved oxygen may play a role in
formation of flakes by Aphani2omenon. Of the four lakes with many D. pulex,
all but Rebecca had significant amounts of dissolved oxygen near the bottom
during the first three quarters of the sampling period -- specifically,
Hydes, 1.9 mg 1-1; Powderhorn, 0.52 mg 1-1; Cedar, 4.8 mg 1-1. Rebecca, on
the other hand, had only 0.07 mg 1-1.
SUMMARY
In summary, lakes which winter-kill or are otherwise low in panfish
develop large populations of D. pu!ex leading to low Chl/TP ratios and high
transparencies. These lakes may also have Aphanizomenon in the form of large
flakes. The presence of the flakes may be attributed to high concentrations
of TP in the water and to adequate dissolved oxygen near the sediments.
REFERENCES
Lynch, M. I98O. Aphanizomenon blooms: alternate control and cultivation by
Daphnia pulex. In Evolution and Ecology of Zooplankton Communities.
Special Symp. Vol. 3- ASLO. W.C. Kerfoot, (ed) . Univ. Press of New
England, Hanover, N.H.
218
-------
iV.D. EFFECTS OF WINTERKILL IN LAKE OF THE ISLES*
This lake lies within the city limits of Minneapolis and is administered
by the Minneapolis Park and Recreation Board. It is M.3 ha in area, has a
mean depth of 2.7 m, and is connected on the west to Cedar Lake and on the
south to Lake Calhoun. Lake of the Isles has been described in detail by
Shapiro and Pfannkuch (1973) and Shapiro (197^)• In essence it is highly
eutrophic, receiving a phosphorus loading of about 2 g m-2 yr-1, virtually
all from urban runoff. In 1930-3 ^» although the lake had blue-green algae in
it, they were not as abundant as at present and Ceratium formed a large
proportion of the biomass. The Minnesota Department of Natural Resources
records show that in July 1927, the Secchi disc transparency was about 2.5 ra-
in the past few years, it has been as low as 0.4 m. Daphn i a pulex and
D i aphanosoma were the only two cladocerans reported in 1930-31 (Klak 1933)-
in recent years Daphnia pu1 ex has been scarce, and there are now 9~11 species
of cladocerans present -- all, except for the few Leptodora, smaller in si2e
than D_. pu lex.
The Park Board hopes to restore the lake by diversion of nutrients from
it through use of first flush storm water diverters. An EPA lake restoration
grant testing the idea was implemented beginning in 1979-
This section describes the effects of a winter-kill that occurred during
the winter of 1976-1977-
Ch1orophy11
Surface chlorophyll concentrations from 1971-1978 are shown in Figure
80. (Where they have been taken, depth integrated samples compared very
closely.) In general chlorophyll is low in May and especially in June,
increases throughout July to a peak in August, and begins to decline before
the end of August to a lower value during September. Concentrations in
1977-78 were lower than during the preceding six years.
Algal Species
Normally blue-green algae are relatively scarce during May and June but
become abundant in July and August. In fact blue-greens may become so scarce
in spring and early summer as to be undetectable. In place of the
blue-greens are usually found a variety of apparently ungrazable greens. In
1975 Sphaerocystis dominated beyond July 1, and in 197& Oocystis and
Sphaerocystis co-dominated with the few blue-greens. These were accompanied
by numerous other greens such as Scenedesmus. Eudor i na. Tetraedron.
Cosmar i um. Pandor i na, and Chlorel la. The situation was somewhat different in
1977- On May 20, although there were no blue-greens there were also no large
ungrazable greens, the algal population consisting exclusively of
Ch1amydomonas, Schroeder i a, green microf1agel1ates, cryptomonads, and
Euq1ena. By June 15 there were a very few filaments of Anabaena flos-aquae,
and Osc i11ator i a Iimnetica. but the same green algae were present. A few
diatoms had appeared and Cerat i um h i rund i ne11 a, which had been found in
moderate numbers in earlier years was present in low numbers.
Ceratium continued to increase in abundance reaching a high of 1800 m1 -1
and making up by far the bulk of the algal biomass on August 1, 1977.
coincident with the maximum abundance of blue-greens which had been
increasing. Cryptomonads were more abundant in 1977» being in excess of 1500
m1 -1 until early July. Although Aphanizomenon flos-aquae became abundant in
the lake during August, it did not form so-called grass blade blooms nor had
it in previous years.
On July 19. 1977. when transparency was 2.4 m, a study was made of the
vertical distribution of the algae. The results for the most important
*by Joseph Shapiro
219
-------
»0t>
75-
197;
A
r
/ \
/ t
t !
1
\ A !
¥ \ ;
V
/
C*
E
-i	1	i	1	r-
1972
A S
/\n
/ i \
/
t
\
t\
v\i
1973
1974
A
/ \
¦250
-<500
-1750
-11000
-1250 E
r->-
zt>i
1975
T'
A3
I \<
1
/ V
/ ^
/ *
¥ \
1976
/ \
I \
a
JU
1977
4?v

X
.ft1
\\
1978
T
' A /
; / V
' i
\ '
i	/
. /
ii
y**- +
.1
*
_J	L.
C
o
w
O
u
o
T3
- 250 o
o
«rt
567B9I0 66 78 9 10 56 769 10 567891
- '000
il25C
— 1500
Fig. 80. Concentrations of chlorophyll (•) in surface waters and
cladocerans (-) in the water column, L. of the Isles,
1971 - 1978.
220
-------
species are shown in Table 37* The surface sample was lost. Note the
different distributions. Both species of Osci11atoria were more abundant at
intermediate depth, as was Cerat i urn. Mer i smoped i a was virtually absent in the
upper b meters of the water column but was very abundant at 5 meters. The
green microflagellates on the other hand were found throughout the 5 m depth
but were especially abundant in the upper 2 m.
Transparency
Transparencies in Lake of the Isles have ranged from as low as .40 m to
as high as U.2 m (Fig. 81). In general summer, i.e. July, August, values, are
less than 1 m but seem to have increased gradually from 1971 and 1972 to 1975
when they again declined. The 1977 season is clearly different with
transparencies always exceeding 1 m and in fact exceeding 2 m except for a
1-month period from late July to late August. It is also clear from a
comparison of the shapes of the Figuress (80 and 81) that there is a close
relationship between transparency and chlorophyll in Lake of the Isles.
Regression of 1/Secchi disc against chlorophyll for all data of 1972-1978 (n
= 67) yields an r2 value of O.656. Regression of only the 1977 data yields
an r2 value of O.858. Thus the high transparencies in 1977 did result from
low algal populations.
Total Phosphorus
Total phosphorus has been measured in the lake since 1971* Surface
concentrations shown in Figure 82 follow a curious pattern. In 197W2 they
averaged about 9° ug 1-1 but from 1973"1975 the concentrations were lower at
about 6O-7O ug 1-1. In 1976 concentrations were a bit higher — 70-80 ug
1-1. They were lower again in 1977. about 50-60 ug 1-1 and rose in 1978 to
~ry high concentrations — up to 210 ug 1-1. There have been no events
rig place in the watershed which could explain these changes, but
.jmination of rainfall records may shed some light. From 1970 to 197^
precipitation declined continuously from 30.5 in yr-1 to 19-1 in yr-1. As
the lake depends on runoff for its phosphorus load and the average
concentration of runoff to Lake of the Isles is in excess of 600 mg m~3
(Shapiro and Pfannkuch 1973) a shortage of this runoff could explain the
decline in concentration. Unfortunately this relationship breaks down, as, in
1975 when rainfall reached the high level of 35-15 in yr-1, total phosphorus
concentration in the lake did not change appreciably. Furthermore 1976, when
concentrations rose somewhat above those in 1975. was an especially dry year,
with only 16.5 inches of precipitation. Finally 1977 (3^*9) and 1978 (27-5+)
do not fit the relationship either. In particular the 1978 (27-5+) values
lack an explanation. A complicating factor in this lake is the large carp
population. These fish, which are capable of pumping phosphorus from the
sediments (Lamarra 1975). were abundant in the lake in 1975 but many died off
in winter 1976-77- At this time therefore the phosphorus concentrations
cannot be explained satisfactorily.
Dissolved Oxygen
Anoxia in the bottom waters of Lake of the Isles has been detected as
early as May 23. 1972 and generally by mid- to late-July dissolved oxygen has
disappeared entirely from the hypolimnion. Some differences from year to
year do occur, however, as shown in Figure 83.
Zooplankton
The zooplankton of Lake of the Isles is diverse with at least 11
cladocerans, k copepods and about a dozen rotifers (Table 38). In 1933 there
221
-------
Table
37: Vertical distribution of the most abundant
July 19, 1977. (numbers of individuals or
algae in Lake of the Isles
filaments)
Depth Oscillatoria Oscillatoria green	Merismopedia Ceratium
(m) Agardhii tenuis	 microflagellates tenuissima hirundinella
0

„
--
	
	
1
36
72
324
0
135
2
54
171
225
0
315
3
54
414
117
9
2000
4
324
324
126
0
729
5
99
9
90
4500
315
222
-------
971
1972
1973
1978
1977
1975
1976
Fig. 81. Secchi disc transparencies in L. of the Isles, 1971 - 1978.
223
-------
-I	1	r
120
IIC
100
90
80
70
60
50
fO
£
f UO-
Q- I
,20-
c
3 or
100-
9C-
ac-
70-
6Cr
50
40
\ A
\
1971
x ,1\
1972
1975
/\
/' >
/ \
30
Vi
1976

1973
\.
1977
i\j
•V '
56769 10 b 6789 IC 56789 10 56769 iO
Fig. 82. Concentrations of total phosphorus in surface waters of L.
of the Isles, 1971 - 1978.
224
-------
Dissolved oxygen mg I
Fig. 83. Profiles of dissolved oxygen in L. of the Isles on
approximately the same date in June for five years.
225
-------
Table 38: Lake of the Isles Zooplankton ^thousands •
.
June	July Aug Sept Oci
8 23 14 21 31 21 28
CLADOCERA:
Daphnia pulex
Daphnia galeata
120
83
11
15
9
7
176
Daphnia retrocurva
147
92
23
117
42
87
27
Daphnia parvula







Daphnia ambigua
4
5





Diaphanosoma 1euctenbergianum
31
28
170
68
79
58
5
Ceriodaphnia lacustris
4
9
370
57
61
40
14
Total "arazers"
306
217
574
257
191
192
222
Bosmina longirostris
1905
18
306
887
108
918
1817
Chydorus sphaericus
89
28
272
279
84
69
578
Leptodora kindtii







Pleuroxus sp.







C0PEP0DA:







Cyclops bicuspidatus
333
9
5
169
7
174
438
Cyclops vernal is
9
18

15
22
58
44
Diaptomus siciloides
9
5
11
26

7
18
Naupli i
400
111
687
49
500
687
511
Mesocyclops edax
44
262
158
77
262
163
44
R0TIFERA:







Asplancha priodonta
9


19
19
11
5
Conochiloides sp.
4


15
3541
36
23
Keratella cochlearis
382
64
87
8
206
195
1288
Keratella quadrata
626
32




50
Kellicottia longispina
4


8
14
29
285
Polyarthra vulgaris
297
184
283

136
69
330
Trichocerca multicrinis




93


T. similis
Lepadella sp.
Myti1inia sp.
Filinia sp.	196 4
Monostyla sp.	14
Platyias patulus
Testudinella patina
Euchlanis dilata
1972
May June July Aug	Sept
18 13 13 18 29 13 20

148
5
22


62

148
1
3
101
7


198

114
23





38
32
86
150
260
110

120
240
460
71
320
260
0
568
304
676
221
580
630
42
1900
140
250
610
2400
270
21
13
59
220
120
510
16
730
5
1500 800 930 1100 980 1700 820
-------
Table 38 Continued
n AnnrFRA-
July
22
1974
Aug Sept
20 16
1975
June July
9 1 23
Aua
15
Sep1
9
May
18
1976
June July Aug
18 15 17
May
20
1977
June
15
23
28
30
v L. r* \J U O L r\ f \ •

















D. pulex













1
31
19
39
D. galeata
82
28
156
331
68
22
25
65
68
155
28
14
120
27
107
45
128
D. retrocurva
17
52
112
214
25
19
50
827
272
54
19
77
2


4
4
D. parvula
8

18
34
3
6

3
8



13




D. ambigua








94








D. leuctenbergianum 297
156
4
5
167
174
233
124

44
107
395




5
C. lacustris
42

8
37
18
6
25
41



7





Total "arazers"
446
236
298
621
281
227
333
1060
442
253
154
493
135
28
138
68
176
B. longirostris

48
189
380


6
140
700
74

19





C. sphaericus .
328
256
55

3
407
308
385

208
512
72
2

6
2
1
L. kindtii
3












1

1

Pleuroxus sp.

















COPEPODA:

















C. bicuspidatus
8
14
33
825

11
9
29
1046


51
420


2

C. vernal is
149
369
166
69
51
41
68
113


25
47


11
2

D. siciloides
173
73
17
190
170
112
133
51
205
94
38
68
89
98
275
135
201
Nauplii
133
86
181
5
60
62
83
255
1233
407
667
1363
810
128
554
568
779
M. edax
66
88
20
25

104
25
49
20
27
66
166
79
70
108
65
165
ROTIFERA

















A. priodonta
5

17
9

3
6
3









Conochiloides sp.
8




3
0
u


3







K. cochlearis
3
14
25
14
3
9

3
2337
91
244
414
58
58
332
526
1192
K. quadrata



51




38
13



2



K. longispina
8
10



6

6
38

231
98
2




P. vulgaris
8




3



10
47
21





T. multicrinis






2


13
47
12
2




T. similis

















Lepadella sp.








8








Mytilinia sp.










3






Filinia sp.










117
5





Monostyla sp.

















P. patulus

















T. patina

















E. dilata
-------
Table 38 Continued
/> i a r-. a a n n R
July
6 13
15
19
21
Aug
1
1977
5 24
Sept
7 13
21
Nov
8
1978
May June July
29 14 4
20
CLADOCERA:
















D. pulex
15
3
3

6



25







D. galeata
31
6
3


6
6
52
33
86
195
6

3

D. retrocurva





6
3
42
12
19
56

73
323
105
28
D. parvula








3





9

D. ambigua










3

106
17
6

D. 1 euctenbergianum
31
257
269
207
230
255
863
98
43
81
25


3
17
48
C. lacustris


3


3
3
199
333
1112
589
3
3
252
256
122
Total "grazers"
77
266
278
207
236
270
875
391
416
1245
759
198
188
595
396
198
B. longirostris





3
16
3777
133
780
336
184
54
95
26
17
C. sphaericus



3
3
10
29
52
37
24
34
2
3
24
40
187
L. kindtii
















Pleuroxus sp.















3
COPEPODA:
















C. bicuspidatus
12

3
14

3

21
65
30
15
132
221
34


C. vernalis




3

19

31
42
22



6

D. siciloides
225
215
314
238
130
158
159
70
97
143
142
99
209
235
145
57
Naupli i
994
788
744
521
796
661
863
899
700
651
285
5
926
704
1040
215
M. edax
108
152
191
110
97
87
115
7
43
570
268
2
36
71
71
6
ROTIFERA:
















A. priodonta



9
3
100
10
59
56
276
46
11


54
6
Conochiloides sp.














66
351
K. cochlearis
3014
1580
1219
445
899
606
745
269
136
304
290

2905
1115
165
710
K. quadrata










3

70
14
9

K. longispina

15
24
9
109
364
500
38
6

3

85
17
17
45
P. vulgaris


3

57
77
131
1241
1292
1079
241

24
126
111
226
T. multicrinis




12
16
38
7
3
14


3



T. similis





3










Lepadella sp.





3






6
7
3
3
Mytilinia sp.




15
3










Filinia sp.







17
3







Monostyla sp.








6







P. patulus






3








8
T. patina















31
E. dilata
















-------
Table 38 Continued
1978
Aug Aug Sept
CLADOCERA:
2
18
15
D. pulex



D. galeata
59
137
111
D. retrocurva
23

45
D. parvula
5

3
D. atnbigua



D. leuctenbergianum
147
185
50
C. lacustris
98
107
47
Total "arazers"
332
429
256
B. longirostris
33
81
26
C. sphaericus
347
465
92
L. kindtii



Pleuroxus sp.
3


COPEPODA:



C. bicuspidatus


10
C. vernalis
21
55

D. siciloides
72
83
67
Naupli i
412
713
612
M. edax
67
44
29
ROTIFERA:



A. priodonta


2
Conochiloides sp.
824
26
36
K. cochlearis
363
486
50
K. quadrata
3

2
K. longispina
3

2
P. vulgaris
185
63
14
T. multicrinis
5
9

T. similis



Lepadella sp.

2

Mytilinia sp.



Filinia sp.
8

5
Monostyla sp.



P. patulus

2

T. patina

2

E. dilata

2

-------
were fewer cladocerans, with only two species, Daphnia pulex and Diaphanosoma
brachyuium. reported by Klak (1933)* In 1977 Daphnia magna was found on
several occasions although it had not been found in other years. Daphnia
pulex, although being reported in 1933, was absent in 1971~1976, but appeared
in moderate concentrations (up to 39,000 m-2) in June 15~July 21 of 1977 and
reappeared on two occasions in 1978. Daphnia retrocurva showed an inverse
relationship to [K pul ex, but [L galeata, generally the most abundant
Daphnia, co-existed successfully with Daphnia pulex in contrast to Wirth Lake
where pu 1 ex was much more abundant (IV B 2). The abundance of |h qa1eata
in 1977 was no different from previous years. Certainly, however, qa1eata
is least abundant in July and August and most abundant in June.
In addition to D_^ retrocurva being least abundant when IK pulex was
present, D i aphanosoma. Bosm ina. and Cer i odaphn i a also showed this
relationship. However, Cer iodaphni a was also in low abundance in 1976 so the
relationship is not perfect. Bosmi na can co-exist with th galeata.
The distribution of Chydorus sphaer i cus is interesting in that it is
abundant only when filamentous blue-greens abound. This is in agreement with
many similar distributions in the literature and probably occurs because C.
sphaer i cus moves into open water by using filaments of blue-greens as rafts.
A correlation between the numbers of filaments of Aphanizomenon f1os-aquae
and the numbers of Chydorus yields an r value of O.38. However it may be
more significant that out of 19 pairs, in 7 where Aphani2omenon was rarer
than 1300 filaments per milliliter, Chydorus was rarer than 3 m-2. Thus
there may be a threshold concentration of filamentous blue-greens above which
the abundance of Chydorus in the plankton is a function of its abundance in
the littoral areas but below which it either does not become transported to
the limnetic regions or only does so in small numbers.
The body sizes of the daphnids are of interest. Figure 84 shows that in
May and June 1977 IL. ga1eata tended to be larger than in 197&.
In addition to changes in cladocerans, D i aptomus s i c i1o ides was
unusually abundant in 1977*
F i sh
A study by the Minnesota Department of Natural Resources in 1975 showed
that, whereas the bullhead, northern pike, and yellow perch populations were
near the state means, the carp population was high -- above the state-wide
median -- and the crappie and bluegill populations were high. During the
winter of 1976~77» dissolved oxygen under the ice declined to very low levels
and an estimated 95% of the fish died, including many of the carp (MDNR
records). In April 1977 although there was no flow into the lake from Cedar
Lake because of its low level and because the channel from Lake Calhoun was
dammed off, another survey showed adult suckers, bullheads, northern pike,
crappie and golden shiner but no carp. Apparently some fish had survived the
winter kill, although they were few. In August 2, 1977* adult and young of
the year crappie were detected along with bluegills and adult carp. The
latter may have entered from Lake Calhoun when the barricade was removed, but
the crappies probably resulted from spawning in the lake. There also was
evidence of spawning by northern pike. Restocking of the lake took place
during May 1977 when 20,000 largemouth bass fry and 1*5.000 walleye fry were
placed in the lake; and in August 1977 when 5500 crappies , 4200 sunfish, and
14 adult northern pike removed from Wirth Lake were put into Lake of the
Isles. Finally, 4300 northern pike fingerlings were added in May 1978.
There was no winter-kill in the winter of 1977~78.
230
-------
2.00r
160
E i.20
E
CP
c
a;
.80
.40
5/18
1976

i >
\

1 \
1 \
i 1976 ,
\
n	
6/15
1
1
1977
-l 5/20

197'
6/18
1976
8/24
J/17 1977
1976
May
June
July
August
Fig. 84. Comparison of mean body lengths of Daphnia galeata in
1976 and 1977. Vertical bars are t 1 standard deviation.
231
-------
DISCUSSION
The most obvious change in Lake of the Isles during the 8-year period
has been the high transparencies of 1977* So high were they that macrophyte
problems -- partially caused from various Potamoqetons -- were experienced
and mechanical harvesting was done. The significant relationship in this
lake between transparency and chlorophyll leaves no doubt that the cause of
the high transparency was the low chlorophyll concentrations. The question
is, why were the chlorophyll concentrations so low in 1977? To some extent
the low algal populations probably relate to the low phosphorus
concentrations i.e. while it is difficult to compare the data from
year-to-year because of its nature, if the highest summer chlorophyll
concentrations are compared with a visual average for total P, the
correlation is good, except for 1978 when exceptionally high total phosphorus
prevailed (Table 39).
That is, in 1977. total phosphorus was lowest and overall algal
abundance was lowest. However it is likely that this is only part of the
reason and that grazing by herbivores was also important in controlling the
algal population. Thus if one compares, for each year from 1971~1976, the
shape of the chlorophyll curve with the shape of the curve representing
abundance of Daphn i a plus D i aphanosoma plus Cer i odaphn i a. as in Figure 80,
there is a close resemblance, particularly for 197^~^975t and 1976. At the
same time the 1977 data show a rather good pos i tive relationship between
chlorophyll and zooplankton abundance. That is, on the one hand there is
evidence that low chlorophyll is caused by grazing while on the other hand
the high chlorophylls of 1977 occurred with high zooplankton. This paradox
may result from the fact that Cera t i um made up 72-30% of the algal biomass in
Lake of the Isles between June 15 and August 2^, 1977 whereas in earlier
years its contribution was much less — averaging 27% in 1976, 8% in 1975.
and 2% in 197^* Because Ceratium is too large to be eaten by most
herbivores, its abundance probably increases when nutrients are released as
i ~er algae are consumed. Thus the positive 2ooplankton/chlorophyl1
at i onsh i p.
This situation is analogous to that in Wirth Lake where algal abundance
was low in the presence of the large population of Daphnia pulex until the
grass blade blooms of Aphan i zomenon f1os-aquae developed. It is analogous
too in that the "facilitated" algae — in Lake of the Isles, Cerat i um, and in
Wirth Lake, Aphanizomenon flos-aquae -- both permit greater transparency than
would be expected from consideration of chlorophyll alone. This is
especially true for Cerat i um which reached its maximum abundance not at the
surface but at some distance beneath it. Thus the zooplankton, both by
reducing algal abundance and by fostering "large" algae, increase
transparency.
Of the reason for the presence of the large grazers there can be little
doubt. The fish-kill in 1976-77 was severe enough to eliminate even carp.
This may also explain why total phosphorus was so low during 1977* More
importantly the elimination of the planktivores removed the predation
pressure from the zooplankton allowing EL pulex, which had not been recorded
from 1971 onwards, to appear, and even allowing Eh magna, never recorded in
the lake before, to be present. Even the [h qa1eata were larger in early
1977 than at any time in 1976. Interestingly, here and in Wirth Lake, D.
galeata was present through all of the years studied even though predation
pressure was high enough to eliminate pulex
One additional feature that may have helped Daphnia pulex to be present
in 1977. and to the extent it did in 1978, is the distribution of dissolved
232
-------
Table 39: "Visual averages" of total phosphorus concentrations, and maximum
chlorophyll a concentrations in L. of the Isles. 1971-1978
(mg m - 3).
"Average"	Maximum
Year	Total P	Chlorophyll a
1971	90	72
1972	90	111
1973	60-70	58
1974	60-70	58
1975	60-70	70
1976	70-80	72
1977	50-60	48
1978	>90	41
233
-------
oxygen. As shown in Figure 83 during both of these years dissolved oxygen
existed well below the thermocline of 3~5 m early in the year providing a
refuge for the Daphnia from the few remaining warmwater fish. This situation
existed in 1972 also but presumably [h pulex did not appear because predator
pressure was still too high.
REFERENCES
Klak, G.E. 1933- A comparative study of summer plankton from 22 bodies of
water in the vicinity of St. Paul, Minnesota. Master's Thesis.
University of Minnesota 1933-
Lamarra, V.A. 1975- Digestive activities of carp as a major contribution to
the nutrient loading of lakes. Verh. Int. Verein. Limnol. 19:2461 -
Shapiro, J. and H.O. Pfannkuch. 1973* The Minneapolis Chain of Lakes. A
study of urban drainage and its effects. Interim Report No. 9«
Limnological Research Center. University of Minnesota.
Shapiro, J.	Report to the City of Minneapolis Park and Recreation
Board on the condition of lakes in Minneapolis, 197^» Interim Report
no. 12. Limnological Research Center. University of Minnesota.
234
-------
V. THE ROLE OF PHYSICAL-CHEMICAL CONDITIONS IN AFFECTING ALGAL ABUNDANCE
—LAKE HARRIET*
Lake Harriet (area 1^3 ha, mean depth 8.8 m) is administered by the
Minneapolis Park and Recreation Board. The sole sources of water to the lake
are precipitation, surface runoff, and groundwater which is known from
previous studies to influence the lake (Shapiro and Pfannkuch 1973)• The
lake has no functional outlet. Interest attaches to this lake because of all
the Minneapolis lakes it is normally the clearest and apparently the least
eutrophic. Vet the phosphorus concentrations in the lake are high enough to
support much larger algal populations than normally exist. The following
discussion describes the probable reasons for this condition of the lake and
the event that occurred in 197*t» when the algal population rose to very high
levels. Lake Harriet is the subject of an experiment beginning Spring 1979>
in which, under an EPA lake restoration grant, the drainage area around the
lake is being swept periodically to prevent nutrients from reaching the lake.
ChlorophyI 1
Surface chlorophyll concentrations from 1971 to 1978 are shown in Figure
85- Generally, although concentrations may be much higher in spring, they
have always been below 10 mg m~3 and usually below 5 mg m~3 >n summer. This
pattern changed in 197^ when concentrations were always above 10 and as high
as kj mg m~3 during the summer. Concentrations in 1975 were much lower than
in 197^ but still higher than usual. The data on algal biomass in Table 40
support the chlorophyll results in that 197^ had a very high algal crop and
that in 1975 was still unusually high.
Phaeophyt in
In 1977 phaeophytin measurements were made. The relationship between
chlorophyll and phaeophytin is shown in Figure 86. High chlorophyll values
are associated with high chlorophyl1/phaeophytin ratios and low chlorophyll
is associated with a higher proportion of phaeophytin. The r2 value is O.63.
Algal Species
In 1971 green algae predominated through July, but from mid July the
domi nants were the filamentous blue greens — Aphan i zomenon and Anabaena. In
1972 the same situation occurred, but the greens persisted a few weeks
1onger.
In 197^ the most abundant algae were the blue greens Anabaena,
Aphanizomenon, and Gomphosphaeria. In 1975 Anabaena, Aphanizomenon, and
Osc i11ator i a abounded, in 1976 Aphan i zomenon and Anabaena and in 1977
Aphan i zomenon. Although Aphan i zomenon f1os-aquae was identified, it was
never found to form "grass blade" or "flake" blooms.
Transparency
Secchi disc transparencies in Lake Harriet (Fig. 87) have ranged from a
high of S.U m in 1972 to a low of 1.06 m in 197^* In general summer values
are least with the very high transparencies occurring only in spring. The
lowest July-August transparencies coincided with the high algal populations
of 197U and 1975.
Total Phosphorus
Surface concentrations of total phosphorus are shown	in Figure 88.
July-August values appear to have dropped from about h2 mg	m~3 in 1971 to a
stable level of 3^~37 mg m~3 in subsequent years, except for	197^ and 1976-
*by Joseph Shapiro
235
-------
Table 40- Algal biomass in surface waters of L. Harriet, July-September,
1971-1977.
Year
1971
Date
7-19
8-	2
8-24
9-13
mq-T1
4.40
1.10
0.61
mean
2.04
1972
7- 6
7-24
8-	9
8-22
9-13
0.61
0.33
1.10
1.50
0.70
0.85
1974	7-22
8-20
9-16
10.16
11.96
10.23	10.78
1975
7- 1
7-23
8-15
9-	9
1.80
1.51
8.17
2.97
3.61
1976
1977
7-14
8-17
8-26
7- 7
7-13
8-	1
8-24
0.17
0.83
1.30
0.50
0.65
2.18
1.75
0.77
1.27
236
-------
/ \ 1974
1977
5 6 7 6 9 IC
Fig. 85. Concentrations of chlorophyll in surface waters of L. Harriet,
1971 - 1978.
>> r
si 5
Q.
O
o>
2
Cl
o
10
12
Chlorophyll, mg m~3
Fig. 86. Chlorophyl1/phaeophytin relations in L. Harriet, 1977,
237
-------
-i	1	1	1	r
1971
"I	1	1	1	r-
1973
1972
1 •'••• U
1 '14
I /1 / ^
\!
*
t
A ¦
E
^ t
o 5
Qt	1—	f	

t
n

V
V
1974
-4	'	!	h
-i	1	1	(-
1977
2-
1975
\
1976
i \
1978
* ' \
V ^
r" v.
. *
	1	L	l	
56789 10 56789 lC 56 789 (0 5678
Fig. 87. Secchi disc transparencies in L. Harriet, 1971 - 1978.
238
-------
r —r- * -I- 1 ?
1971
* \
v—
	:	1		«	r ¦
1972
*
h
tit i i
1973
¦ '
\
	1	r	1	1	
1974
1975
¦ v\ ¦
1976
i i i i i
1977
r ^ A
W I
i > i i i
1978
	<
\ / "
V
I'll
56 7 B9 10 56789 10 56789 10 56789 iC
Fig. 88. Concentrations of total phosphorus in surface waters of L.
Harriet, 1971 - 1978.
239
-------
In 1976 values were very low probably because of the extremely low rainfall.
However, in 197^ the highest July-August values occurred in spite of a very
low rainfall. Thus it is unlikely that rainfall is the only determinant of
total phosphorus from year to year. No explanation can be found for the high
1974 concentration.
D i ssolved Oxygen
The hypolimnion of Lake Harriet becomes anoxic with the production of
H2S, and the rate of depletion of oxygen does parallel the algal abundance.
For example, in the period July 19-24 (Fig. 89) dissolved oxygen
concentrations at 6 m were highest in 1972, intermediate in 1971 arid lowest
in 1974, 1975. and 1978. These correspond to 1974~75~78 having the highest
chlorophyll concentrations, 197' being next and 1972 being least. For the
period August 15-26 (Fig. 89), the correlation is even better. The rank
orders are:
Estimated dissolved oxygen
	at 6 m		Summer chlorophyll concentrations
1972
lowest 1972
1971
1971
1978
1978
1975
1975
1974
highest 1974
In any event, 1974 appears to have had the least concentration of oxygen
at 6 m, i.e. the rate of depletion during 1974 seems to have been the
greatest, corresponding with the greatest algal abundance and the highest
concentration of total phosphorus.
Zooplankton
Lake Harriet has a varied zooplankton assemblage of about 9 cladocerans,
6 copepods and 9 rotifers, as listed in Table 41. Paphnia pulex. which is
of such interest in Wirth Lake and Lake of the Isles was not present in Lake
Harriet in 1930~1931 (Klak 1933) and played little if any role from
1971-1978. The ones found in 1977 were immature and may not have been D.
pu1 ex. D parvu1 a was also not abundant, but D. retrocurva was present and
abundant most of the time. D_;_ qaleata, largest except for CL pul ex was
present and abundant, except that in 1974 it was sparse and in part of 1978
it was absent (Table 41). [K retrocurva was also sparse in 1974 but in 1978
it was perhaps more abundant than usual in the absence of EK ga1eata. Except
for a brief appearance in 1974 and 1975. Cer i odaphni a was absent in contrast
to its abundance in Wirth Lake and Lake of the Isles. Also, Bosmi na,
abundant in 1971 and 1972, almost disappeared thereafter. Several rotifers
normally abundant were sparse during 1974 and 1975* Thus 1974 and 1975# and
particularly the former, seem to have been an unusual year for the
zooplankton community in Lake Harriet.
Determinations of body length of D_^ ga1eata show no significant
differences in si2e from year to year.
F i sh
A Department of Natural Resources survey on May 21-23, 1958, showed that
the northern pike population was low (0.8 per net lift) and the perch
-------
Dissolved oxygen mg l~'
1-26
8-20
8-24
8-22
1975 1978
1971
1972
1974
7-19
7-23
7-22
7-20
7-24
1971
1974
1975 1978
1972
Fig. 89. Profiles of dissolved oxygen in the upper portion of L.
Harriet during the month of July, 1971 - 1978, and during the
month of August, 1971 - 1978.
241
-------
I?!?!? Zooplankton (thousands m~2) in L. Harriet,
CLADOCERA:
Daphnia galeata mendotae
Daphnia retrocurva
Daphnia parvula
Daphnia pulex
Total Daphnia
Ceriodaphnia lacustris
Diaphanosoma
1euctenbergianum
Bosmina longirostris
Chydorus sphaericus
Leptodora kindtii
COPEPODA:
Cyclops bicuspidatus
thomasi
Cyclops vernal is
Mesocyclops edax
Diaptomus siciloides
Diaptomus oregonensis
Diaptomus clavipes
Nauplii
ROTIFERA:
Asplanchna priodonta
Conochilus sp.
Kellicottia longispina
Keratella cochlearis
Keratella quadrata
Filinia longiseta
Polyarthra vulgaris
Trichocera multicrinis
T. simi 1 is
1971
June
4
23
July
14
30
84
173
66
90
9
87
4
56
13

0
3
106
260
170
149
4
12
366
137
2972
215



3
4
9

3


3551
1016
364
209
71
18
6
14
63
77
153
189
254
51
51

702
206
15
6


913
1317
58
143
4
56
409
842

6
960
367
12

382
69



3


1971-1978.
Aug.
29
31
Sep1
21
: May
19
June
13
1972
July
13
Aug
9
18
45
36
0
36
40
0
35
31
8

124
23
3
267
108
14
240
94
26
114
52
14
81
76
74
0
150
379
360
180
81
59


25
230
77
170
116
163
20
367
35
31
P
33
65
3900
P
250
28
190
38
P
140
170
27
78
212
1300
2800
170
350
102
163
17
43
142
16
65
670
390
375
125
19
12
290
18
350
381
2300
2300
420
920
860
380
155
83
211
7
7
277
46
30
8
48
90
24
149
8
12
7
88
835
180
3399
248
2017
552
4011
9
1099
114
884
54
68
1192
58
3978
54
333
547
30
4
935
170
-------
Table 41 Continued


1972


1974


Aug
Sept

July
Aug
Sept

29
13
20
22
20
16
CLADOCERA:






D. galeata

30

32

51
D. retrocurva
?
0
?
5

6
D. parvula





2
D. pulex






Total Daphnia
110
30
70
37
0
59
C. lacustris





2
D. 1 euctenbergianum
58
85
62
340
173
259
B. longirostris
180





C. sphaericus
280
92
110
13
37
253
L. kindtii






C0PEP0DA:






C. bicuspidatus
34
—
131


55
C. vernal is
30

49
33
78
366
M. edax

--

6
86
183
D. siciloides
92
--
39
189
33
254
0. oregonensis
138

84



D. clavipes

--
13



Naupli i
490
--
390
18
139
369
R0TIFERA:






A. priodonta

--




Conochilus sp.

--

315
34
220
K. longispina
48
--
18


7
K. cochlearis
404
—
333


4
K. quadrata
17
—
3


F. longiseta

—




P. vulgaris
547
—
348

24
4
T. multicrinis






T. similis






1975
June July
9 1 23
Aug Sept
15 9
31
66
4
1006
12
104
528
4
46
39
193
238
101 431
13
17
35
4 26
201
81
488
73 144
13
171
17
4
63
67
63
192
21
21
154
17
50
3
82
85
487
55
4
4
15
115
19
27
1
1
0
16
16
5
17
21
61
21
79
57
80
1
1976
May June
18 22
699
7
3
3053
10
155
111
410
269
119
709 119
2
2
75
5
79
978 246
38
88
157
3
8
3
-------
Table 41 Continued

1976


1977




1978




July
Aug
June
July
Aug

May
June
July
Aug

Sept

15
17
15
7
13
1
24
29
14
20
2
26
15
CLADOCERA:













D. galeata
0
43
356
75
52
26

1014
491




D. retrocurva
112
153
9
57
26
90
38
17
52
57
162
109
40
D. parvula













D. pulex


11?




3





Total Daphnia
112
196
376
132
78
116
38
1034
543
57
162
109
40
C. lacustris













D. 1euctenbergianum
92
129
6
93
131
107
42

a
24
74
104
11
B. longirostris







3





C. sphaericus
18
134
11
10

86
85
9
4
35
105
70
77
L. kindtii













COPEPODA:













C. bicuspidatus
9
70
413
26

13
34
416
550
3


14
C. vernalis
9
12

10


13


3
9
3
14
M. edax
53
34
40
60
22
9
4

11
84
54
18
11
D. siciloides
282
161
198
122
179
43
106
359
188
127
103
75
69
D. oregonensis













D. clavipes













Naupli i
561
292
531
353
467
261
344
932
639
808
692
226
204
ROTIFERA:













A. priodonta







23
4




Conochilus sp.
172
1177
152
6274
1509
612
102


2129
587
44
66
K. longispina
24
460
23
18
22
47
641
88
63
5
9
29
106
K. cochlearis
123
65
1211
244
205
206
518
2213
2637
480
450
174
419
K. quadrata
2

129
3



926
41
8
3


F. longiseta
15

132





4
5
6
21
49
P. vulgaris
31
137
353
41
22
154
310
877
920
198
111
75
63
T. multicrinis

62
55
8
22
43
13



3
41
57
T. similis
2

6


4


4


10
17
-------
population was high (168 per net lift. By 196^. after stocking with pike,
pike were more abundant, nearly 3*3 per lift, while perch had declined to 28
per lift. On July 1, 197^ northern pike were at 9*3 per lift and perch had
declined further to 7-3 per lift. From 1975 to 1978 a total of 13&3 adult
northern pike were removed as part of a program to replace them with
muskellunge. Stocking of the muskellunge began in September 1974 when 295
were put into the lake and continued along with walleye stocking as follows:
Year	fluskel1unqe	WalI eye
1974	295 adult	1300 fingerlings
1975	200 fingerlings	1000 fingerlings
1976	100 fry	1100 fingerlings
1977	102 fingerlings	1500 fingerlings
1978	100 yearlings	1100 fingerlings
The 197^ stocking with muskellunge was not successful and in 1978
observations by the DNR indicated a large perch population in the lake.
DISCUSSION
The main question to be answered is what happened in 197^? Why did the
algal population increase so dramatically, as evidenced by measurements of
chlorophyll, biomass and transparency. Some of the increase may have come
about because, as seen in Figure 88, phosphorus was higher during 197^ than
any other year from 1971 to 1978. The correlation coefficient between
average surface total phosphorus values during July through September and the
corresponding chlorophyll values is O.69 (Table 42). However, if 1974 and
1975 are omitted, the r value drops to -.06. That is, the chlorophyll does
not appear to be controlled ordinarily by concentrations of phosphorus.
Indeed, if chlorophyll were controlled directly by phosphorus i.e. if P were
the limiting factor for the algae, then the chlorophyll increase during 197^
should have been much lower. The real cause for the increase in chlorophyll
seems to be that the ch1orophy11/P ratio increased. This can be seen in
Table 43 and Figure 90 where the chlorophyl1/P ratios are shown for each
year from 1971 through 1978. Only two things can account for such an
increase in the ratio — an increased supply of limiting nutrients other than
P, or a decrease in the rate of grazing by the zooplankton.
The first is unlikely as no known changes occurred in the drainage area
and it is difficult to understand how something such as nitrogen could
in--~ose 5 to 10-fold without an obvious change. The second explanation is
v	iikely as the number of Daphn i a did decrease dramatically in 1974, as
set;,! in Table 41. Table 1*3 also compares the chlorophyll/P ratios for
1971-1978 with various concentrations of Daphnia and other herbivorous
zoopIankters. Comparison of the chlorophyll/P ratio with all herbivores
gives an r value of + .34, clearly inconsistent with the phenomenon of
grazing. However, a comparison of the ratio with Daphni a qaleata
concentrations gives an r value of -.3&» and comparison with all Daphnia
together gives an r of -.58. Interestingly Daphni a retrocurva must be an
important grazer along with Daphnia galeata as the addition of Daphnia
retrocurva, the chief "other" Daphni a. raises the r value considerably.
One problem with this analysis is the high chlorophyll/P ratio in 1975
when, as the data in Table 41 show, the mean Daphn i a population was large
(150,000 m-2) and presumably, therefore, grazing rates were high. However,
as the data in Table 41 show, the concentrations of Daphni a on July 1, 1975.
were exceptionally high and if these are omitted the mean becomes much less
245
-------
e 42: Relationship between surface concentrations of chlorophyll and
total phosphorus in L. Harriet, 1971-1978.
Puly-Sept. Mean (mg . m"^)
Year
ChK
TP
1971
3.1
42
1972
3.5
38
1973
3.9
38
1974
29.0
50
1975
14.0
36
1976
3.6
27
1977
4.8
32
1978
7.2
37
All data, r-= 0.685
Minus 1974 and 1975, r = -0.57
246
-------
Table 43: Chlorophyll/phosphorus ratios in L. Harriet 1971-1978 compared
with those in Wirth L. 1974-1978; relationship of chl/p ratios
in L. Harriet to various groupings of grazers.
mean concentrations of various grazing
cladocerans, thousands m~ , July-Sept.
mean chl/TP
July-Sept.
Daphnia
Ceriodaphnia
All
Year
Harriet
Wirth
& Diaphanosoma
Daphnia
1971
.073

239
110
1972
.092

302
188
1973
.105

--
--
1974
.592
.930
291
32
1975
.413
.430
296
150(56)
1976
.135
.670
265
154
1977
.156
.490
184
91
1978
.210
. 170*
145
92
D. galeata
74
104
28
50(2)
22
38
0
* = year following rotenone treatment of Wirth L.
( ) = value without data from July 1, 1975
r
r
r
chl/TP vs. Daphnia + Ceriodaphnia + Diaphanosoma = +.34
chl/TP vs. all Daphnia = -.58 (-.82)
chl/TP vs. D. galeata = -.36 (-.51)
247
-------
.3-
CL
1973
1975
1977
1978
1971
1972
1974
1976
Years
Fig. 90. Mean summer values for the ratio chlorophyll/TP in L. Harriet
surface waters, 1971 - 1978.
248
-------
(56,000 m-2) which is in much more agreement with the other results. Then,
the r values for correlation of chlorophyll/P ratios with Daphnia qa1eata and
with all Daphnia are -.51 and -.82 respectively. Again addition of D.
retrocurva betters the correlation significantly.
It is interesting to note (Table 3) that the chlorophyll/P ratios in
Lake Harriet for 1971-1973 were similar to those in Wirth Lake in 1978 i.e.
when herbivores became abundant in Wirth Lake as a result of rotenone
treatment the year before. Furthermore the 1973 and 1975 ratios in Lake
Harriet were similar to those in Wirth Lake before its rotenone treatment.
Further evidence of the "grazing" hypothesis comes from the
chlorophy11/phaeophytin ratios found in Lake Harriet in 1977 which suggests
fairly convincingly (r 2 ¦ 0.6>3t r ¦ 0.79) that the normally low
concentrations of chlorophyll in the surface waters of Lake Harriet are
accompanied by high percentages of phaeophytin — a phenomenon consistent
with the high grazing rate.
Thus it appears that the high concentrations of chlorophyll in Lake
Harriet in 197^ and 1975 probably resulted from the unusually low
concentrations of herbivorous zooplankters in those years.
The question then arises, why were the herbivores less abundant during
those two summers? According to the Department of Natural Resources fish
survey data, the perch population had been declining from 1964 to 1974 and
probably, therefore, selective predation on the larger herbivores had been
declining as well. Thus one would have expected that in 1974 herbivores
would have been at a peak, or at least more abundant than in 1971 and 1972.
However, this assumes that abundance of the fish is the sole indicator of
their importance in predation and clearly this is not so. The degree of
predation must depend on the spatial relationship between the fish and their
prey — the Daphnia. If the Daphnia had a refuge from the fish they would
face less predation, but if the refuge were to vanish, their numbers could
decline. This may be what happened in Lake Harriet. The dissolved oxygen
data show, for the years when it is available, that in 197^ and 1972 and to
some extent in 1978, when herbivores were relatively abundant, the upper part
of the hypolimnion maintained higher dissolved oxygen concentrations than
during 1974 and 1975 when the herbivores were least abundant (Fig. 89). That
is, in 197^ and 1975 the refuge vanished when the Daphni a, unable to survive
in the lower colder waters because of the low oxygen content, were forced to
move upwards into the warmer epilimnion where predation by warm water reduced
thei r numbers.
Once again, therefore, the question is, why did the oxygenated refuge
vanish? Increased concentrations of total phosphorus in 1974 may provide the
explanation. Primary production is directly related to total phosphorus
(Smith 1979) and hypolimnetic oxygen depletion is directly related to primary
production. Thus the decrease in dissolved oxygen very likely resulted from
the increase in total phosphorus in 1974. Although the total phosphorus in
1975 was not high the oxygen deficit was still such that the refuge for the
Daphnia did not exist.
The explanation detailed above, if proven true, would probably be the
first demonstration of a threshold effect in eutrophication, or at least in
the response of a lake to a nutrient. It would not have been expected from
the empirical relationships available to us from the work of Dillon and
Rigler (197^) or Jones and Bachman (1976) that such a small increase in
phosphorus could cause such a large increase in algal biomass. However,
considering the biological interaction makes it not only possible but likely
that this did happen. It would be worth testing this idea in other lakes,
even in Lake Harriet itself, by adding sufficient nutrients to increase
249
-------
depletion of dissolved oxygen in the upper part of the hypolimnion. As the
amount of phosphorus reaching Lake Harriet from "normal" sources is in the
range of 800 kg yr-1, the amount that would have to be added is within
practical 1imi ts.
The reason for the high chlorophy11/P ratio in 1975 may have been the
carryover of the 197^+ oxygen deficit. However, another reason may lie in an
increased number of perch in the lake, following the removal of their
predators, and the inability of the introduced muskellunge to control the
perch. These changes are reflected in the higher chlorophyl1/P ratios from
1976-1978 than from 1971-1972. Because of the vulnerability of Daphnia to
predation by perch, disappearance of their refuge under these conditions of
high perch populations could result in a dramatic increase of chlorophyll in
Lake Harriet. Already mean chlorophyll values in 1977 and 1978 are higher
than for any years except for 1974 and 1975-
One difficulty with the scheme as outlined is that one would have
expected the Daphni a to be smaller as well as rarer in 1974 and 1975 *
However, this does not seem to have happened. Perhaps under certain
circumstances predation may be non size-selective.
REFERENCES
Dillon, P.J. and F.H. Rigler. 1974. The chlorophyll-phosphorus relationship
in lakes. Limnol. Oceanogr. 19'7k7"773*
Jones, J.R. and R.W. Bachmann. 1976- Prediction of phosphorus and chlorophyll
levels in lakes. Wat. Poll. Contr. Fed. 48:2176-2182.
Klak, G.E. 1933- A comparative study of summer plankton from 22 bodies of
water in the vicinity of Minneapolis and St. Paul, Minnesota. Master's
thesis, University of Minnesota 1933-
Shapiro, J. and H.O. Pfannkuch. 1973- The Minneapolis Chain of Lakes. A
study of urban drainage and its effects. Interim Report No. 9»
Limnological Research Center, University of Minnesota.
Smith, V. H. 1979* Nutrient dependence of primary productivity in lakes.
Limnol. Oceanogr. 24:1051-1064.
250
-------
Vi. CONCLUSIONS AND RECOMMENDATIONS
In previous publications (Shapiro, et al. 1975» Shapiro, 1980a, 1980b)
we have attempted to convince those involved in lake restoration to think in
ecological terms. That is, we have urged utilization of trophic
relationships existing within lakes as adjuncts or alternatives to nutrient
control. The work presented in this report reinforces our belief that such
an approach — biomanipulation — could be feasible and effective. We
believe that by limiting lake restoration to nutrient control, we are
abrogating our role as ecologists and making it impossible to restore many
lakes which might otherwise be benefited. Clearly, we do not have all the
answers. We do not know whether such unexpected consequences as the
appearance of Aphanizomenon following planktivore reduction will be common,
and we do not know the duration of effectiveness of any of the measures
proposed. However, such measures as rotenone treatment to eliminate
planktivores and/or benthivores, stocking of piscivores, and formation of
zooplankton refuges are generally much less expensive than the measures for
nutrient reduction, and should be investigated before such expensive ventures
are entered into. Similarly, changes that might dramatically reduce algal
abundance, such as increases in nutrient loading sufficient to eliminate
zooplankton refuges, and stocking with planktivores, should be avoided.
Recent deliberate whole-lake manipulations such as those of Benndorf et al .
(198l) and Shapiro and Wright (1982) are encouraging. We hope this report
will stimulate others to attempt lake restoration by biomanipulation.
REFERENCES
Benndorf, J., D. Uhlmann, and K. Putz, 1981. Strategies for water quality
management in reservoirs in the German Democratic Republic. J_n
Eutrophication: a global problem -- part 1. World Health Organization
Water Quality Bulletin Volume VI. Environment, Canada. ISSN 0706-8158.
Shapiro, J., V. Lammara and M. Lynch, 1975- Biomanipulation: an ecosystem
approach to lake restoration, pp 85-96. J_n Brezonik, P.L. and Fox, J.L.
(eds.) . Proc. symp. on water quality management through biological
control. U.S. EPA Rept. No. ENV-07"75"1.
Shapiro, J., 1980a. The need for more biology in lake restoration. J_n Lake
Restoration: proceedings of a national conference. Minneapolis. EPA
U/5-79-OOI.
Shapiro, J., 1980b. The importance of trophic-1 eve 1 interactions to the
abundance and species composition of algae in lakes. J_n Hypertrophic
Ecosystems. Junk. The Hague.
Shapiro, J. and D.I. Wright, 1982. Lake restoration by biomanipulation -
Round Lake, Minnesota. Accepted by Freshwater Biology.
251
-------