United States
Environmental Protection
Agency
Environmental Research
Laboratory
Corvallis OR 97333
Research and Development
EPA-600/S3-82-096 Apr. 1983
Project Summary
Experiments and Experiences in
Biomanipulation: Studies of
Biological Ways to Reduce
Algal Abundance and
Eliminate Blue-Greens
Joseph Shapiro, Bruce Forsberg, Vincent Lamarra, Gunilla Lindmark, Michael
Lynch, Eric Smeltzer, and George Zoto
Studies were made to find alterna-
tives to restoring or managing lakes by
controlling external sources of nutri-
ents. The guiding principle was to under-
stand and use biological 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
more of the following procedures: elim-
ination of benthivorous fish which re-
cycle phosphorus from sediments; ma-
nipulations of algal populations by low-
ering pH, causing artificial circulation;
increasing abundance of larger herbiv-
orous zooplankters by reducing preda-
tion on them, by eliminating plankti-
vores entirely or, by providing refuges
from planktivores.
This Project Summary was developed
by EPA's Environmental Research Lab-
oratory, Corvallis, OR, to announce key
findings of the research project that is
fully documented in a separate report of
the same title (see Project Report order-
ing information at back).
Introduction
Lake restoration projects designed to
reduce the abundance of undesirable
algae usually are based on the premise
that reduction of nutrient input from
external sources or from anoxic sedi-
ments 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, consideration of such an ap-
proach, termed biomanipulation. as op-
posed to nutrient manipulation, indicates
that it has great potential alone or in
combination with nutrient manipulation.
Figure 1 shows some of the possibil-
ities. Note that although the end goal is
reduction of algal biomass, none of the
possible manipulations involve nutrients
directly. Most manipulations listed deal
with changing the quantitative and quali-
tative relationships among the biota so
that the desired reduction 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 is not known is the extent to which
the manipulations might be successful,
the duration of their effectiveness, or the
unexpected consequences from their
use. Figure 2 illustrates the aquatic food
chain.
This report is a summary of work done
on biomanipulations at the University of
Minnesota Limnological Research Cen-
ter, ending in 1980.
Among the possibilities for such manip-
ulation are:
• Elimination of bottom-feeding fish
which, by their feeding activities,
increase the nutrient concentrations
and thereby the abundance of algae
in lakes.
-------
BIOMANIPULATION
formation of
refuges
ADDITION OF
PISCIVOROUS FISh
t
reduction of
planktivorous fish
1
decrease in
*~ predation
1
FISH TOXINS
FISH DISEASES
WINTER-KILL
CAPTURE
CIRCULATION
AND/OR AERATION
increase in proportion
of large herbivores
blue-green to
green shift
increase in
grazing intensity
Reduction of Algal Biomass
Increase in Transparency
decrease in nutrient
recycling
ELIMINATION OF
BENTHIVOROUS FISH
Figure 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 capital letters; mechanisms are indicated in lower case type.
• Manipulation of algal populations to
change algae species' composition
and/or reduce algae abundance by
lowering pH, causing artificial circula-
tion, stimulating activity of viruses
that attack blue-green algae.
• Direct manipulations of zooplankton
populations to increase abundance
of herbivorous species and therefore
grazing on the algae.
• Indirect manipulation of zooplankton
herbivores by manipulating their
predators—planktivorous fish—by ex-
perimental additions, elimination of
planktivores by rotenone treatment,
or elimination of planktivores by
winter kill.
• Modifications in oxygen concentra-
tions, possibly leading to large chang-
es in algal populations via their
effects on refuges for zooplankters.
Elimination of Bottom-Feeding
Fish
The work of Lamarra (1975) showed
that bottom-feeding fish excrete phos-
phorus and nitrogen compounds and that
the rate of excretion depends on the fish
size, the temperature, and the type of lake
sediment present. Lamarra hypothesized
that such input could make 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 restruc-
ture the fish population in Lake Marion, a
shallow, large lake (mean depth 1.98 m,
172 ha area) in south central Minnesota.
Estimates of the fish population and its
characteristics were made before and
after the rotenone treatment using mark-
recapture methods and shore census of
the dead fish, respectively. 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 agricul-
tural watershed as 84 mg P/m2 per year.
Therefore, fish excretion provides 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 bio-
mass and productivity and increases in
transparency expected from reduction in
total phosphorus. Data to test these
predictions have been collected and are
being analyzed.
Manipulation of Algal
Populations by pH Lowering
Preliminary studies had confirmed the
hypothesis that green algae are favored
over blue-greens at lower pH values. In
these preliminary experiments, nutrients
plus COz 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 70 experiments
have now been done in the field and in the
laboratory and the conclusions are as
follows:
1. The phenomenon is reproducible. In
every case (19) in Lake Emily where
COz was added with nitrogen and
phosphorus, the blue-green to green
shift occurred.
2. The phenomenon is not limited to
Lake Emily. Ten other lake waters
have been tested, and the shift took
place in all of them.
3. Additions of HCI generally had the
same effect as additions of COz, but
some exceptions did occur.
4. The shift occurred whether field
experiments were begun in June or
in late September.
5. Field experiments with pH-controlled
enclosures showed that the shift
from blue-greens to greens occurred
at pH values of 5.5 to 8.5, when COz
was used, and at pH values of 5.5 to
7.5 when HCI was used.
6. In mostof the experiments, the green
algae resulting from the shift were
Scenedesmus and Chlorella; in one
experiment, there were 22 species
and subspecies of Scenedesmus.
7. 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: experiments with different
sizes of inoculum showed the rate
increased with a higher initial pro-
portion of greens.
8. The shift often seemed to occur
precipitously, and it involved almost (
all species of blue-greens in the lakes
-------
The Aquatic Food Chain
(Not to Scale)
Piscivorous
Fish
eat
\
Planktivorous
Fish
eat
\
Herbivores
eat
\
.
**
Nutrients
Algae
I
use
t
Nutrients
*
recycle
I
Benthivorous
Fish
Figure 2. 7/?e aquatic food chain (not to scale).
tested. However, some blue-greens
remained, for if the pH was raised,
they again began to regain domi-
nance.
9. Thereasonfortheshiftisobscure.lt 10.
may involve competition between
the two types of algae, but the
increase of the greens occurs after
the decrease of the blue-greens.
Thus, the two phenomena are dis-
tinct. As the blue-greens disappear,
phosphate and ammonia are found in
the water but disappear as the
greens grow.
One possibility is that the blue-greens
are affected by algal viruses at low-
ered 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 arsen-
ate, which reduces phosphate up-
take by the greens, is added the shift
is delayed or prevented. Chlorine
additions at high pH also cause the
shift, presu mably by a different mech-
anism.
Manipulation of Algal
Populations by Artificial
Circulation of Lakes
Artificial circulation, frequently termed
aeration, has been a lake restoration
technique of limited value probably be-
cause of the lack of a proper theoretical
framework for its use. To remedy this
situation, two such frameworks were
constructed: one to explain the shift from
blue-green algae to greens that is often
observed, and the other to explain the
diminution in algal biomass that some-
times occurs. The hypotheses were tested
in two lakes in a series of eight experi-
ments utilizing a total of 76 enclosures
one meter in diameter, which extended
from the surface through the thermocline
to a depth of eight 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.
Species Composition
Response of the phytoplankton at the
species level appeared to depend pri-
marily 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 eupho-
tic zone led to a shift from blue-green
algae to greens and diatoms. Deep slow
mixing, which also increased nitrogen,
phosphorus, and CO2 levels in the eupho-
tic zone resulted in increases in blue-
greens. However, in the case of rapid
mixing, carbon dioxide was introduced
into the euphotic zone rapidly enough to
lower pH values, while in the slow-mix
enclosures pH remained high. This result
agrees with results of the previous sec-
tions.
Not only did the green algae benefit
from rapid circulation, but diatoms also
increased. As this occurred during shal-
low 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 COz
in the hypolimnion was lower. Conse-
quently, pH values did not decrease
significantly during circulation. Further-
more, the lake contained a metalimnetic
population of Oscillatoria 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.
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 chloro-
phyll concentration will increase or de-
crease. Furthermore, the size of the
chlorophyll change will be a function of
such other factors as the N/P ratio, as at
higher ratios the yield of chlorophyll/P is
greater; the loss rate as a result of depth,
sinking, and grazing—factors also affec-
ted by circulation; and the extinction
coefficient, Ew, of the water as it is
determined by non-algal substances dis-
solved or suspended in the water.
These results demonstrate why, with-
out an adequate theoretical framework, it
has been difficult to predict and/or under-
stand the qualitative and quantitative
changes that have occurred in lakes
during circulation. The results obtained
here will be useful in designing future
attempts.
Manipulation of Algal
Populations Through the
Use of Specific Viruses
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 Tower
pH, laboratory studies were used to
explore the relationship between algal
viruses and their hosts. The blue-green
Plectonema boryanum and the Cyano-
phage LPP-1 were used. 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 4) the effect of other
algal species. The most relevant obser-
vation 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 determi ne
whether this is what actually occurs.
Direct Manipulation of
Zooplankton Populations
Decreases of algal abundance could
result from increases in herbivore abun-
dance. Therefore, experiments were con-
ducted on the feasibility of using panto-
thenic acid, previously reported to be
effective, to increase Daphnia abundance.
Results were negative, and it is con-
cluded that such manipulations, includi ng
attempts to add herbivores directly to
lakes, would be ineffectual. Certain pesti-
cides may be exceedingly effective in
eliminating Daphnia, however.
Indirect Manipulation of
Zooplankton Populations via
Planktivorous Fish
Experiments in which different densi-
ties of planktivorous fish were studied for
their effects on zooplankton and algal
populations were carried out in enclo-
sures, divided ponds, and whole ponds.
Enclosure Experiments
In an enclosure measuring one meter
in diameter and one meter deep addi-
tions of bluegill sunfish eliminated such
herbivores as Daphniapu/ex and Daphnia
galeata while allowing the smaller spe-
cies, Daphnia ambigua 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 than in the enclosures
without fish. In some of the experiments,
as fish predation intensity increased,
filamentous blue-greens became rela-
tively more abundant. In the absence of
fish predation the predominant algae
were greens. The effects on algal bio-
mass were the result of fish predation on
the zooplankton, rather than fertilization
of the enclosures by fish excreta. This
result was achieved by experiments in
which nutrients were added intentionally
to the enclosures.
Divided-Pond Experiments
These experiments were done by divid-
ing a small pond (0.5 ha) with poly-
ethylene 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 ambigua and
Daphnia parvula appeared. Consequent-
ly, 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
since the phytoplankton/P ratio was an
average of 3.4 times as high in the
"minnow" half.
Whole-Pond Experiments
In these 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 zoo-
plankton communities. In the stocked
lake, Daphnia pulex was absent, chloro-
phyll concentrations were high and trans-
parency 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 circum-
stances 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.
Manipulation of Planktivore
Populations with Fish Toxins
Previous Experiences in
Minnesota
Examination of the files of the Minne-
sota Department of Natural Resources
revealed 13 lakes which had been
treated in previous years with fish
toxicants; for all 13, pre- and post-
treatment transparency data were avail-
able. Seven had higher transparencies
after treatment, two probably increased
in transparency and four showed no
change.
Effect ofRotenone in W/rth Lake
This (16 ha, 4.3 m mean depth) lake is *
eutrophic from storm drainage input. I
Over a period of several years, the lake
-------
was treated by a variety of ameliorative
techniques: nutrient export, artificial cir-
culation, and piscivore stocking. However,
beneficial effects were minimal (circula-
tion 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 nutri-
ent 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 Daphnia pulex were also absent.
Evidence suggests that had the lake not
been treated with rotenone, the pisci-
vores would have controlled the plank-
tivore populations and Daphnia pulex
might have become abundant for this
reason.
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 previous-
ly winter-killed was affected most by the
rotenone treatment. Daphnia pulex be-
came abundant and algal biomass de-
clined sharply.
Manipulation of
Planktivores by Winter-Kill
Lakes Affected in 1978-79
Many Minnesota lakes winter-killed in
1978-79. Nineteen lakes including those
with no 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/I. Daphnia pulex also appeared in
two lakes suspected of partial winter-kill
and in one lake known to be low in
panfish.
Chlorophyll/TP ratios in the four lakes
with abundant D. pulex averaged less
than .132 ± .046. Among the remaining
fifteen lakes, chlorophyll/TP averaged
.362 ±.136.
Transparencies of the four D. pulex
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 trans-
parency data were available, no trans-
parency 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 Aphanizomenon flos-aquae
in its flake form.
Effect of Winter-Kill in
Lake of the Isles
In 1976-77, 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 trans-
parency during summer. In 1977, trans-
parencies were so high that macrophyte
problems prevailed, requiring mechanical
harvesting. The high transparencies were
probably caused by grazing by Daphnia
pulex which became abundant in the lake
following the demise of the planktivores.
At the same time as D. pulex appeared, D.
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.
Effects of Physical-Chemical
Conditions on Algal Populations
Lake Harriet (143 ha; 8.8 m mean
depth) in Minneapolis perennially pro-
duces lower algal concentrations than
expected from its nutrient concentrations.
This discrepancy has been attributed to
grazing by the abundant Daphnia galeata,
and indeed low chlorophyll concentra-
tions have been correlated with a high
proportion of phaeophytin—evidence of
such grazing. In 1974, summer chloro-
phyll concentrations in the lake suddenly
increased from the usual 5 fjg/\ to as high
as 47 UQ/\. 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 re-
curring.
The explanation for the high chloro-
phyll in 1974 appears to lie in the reduced
numbers of Daphnia present that year.
The decreased numbers of Daphnia may
have resulted indirectly from the some-
what higher concentration of phosphorus
in the lake in 1974. That is, the hypothesis
was made that the increased phosphorus
levels, too low to raise algal abundance by
more than 20 or 30 percent, nonetheless
allowed primary production in the eu-
photic zone (not measured) to increase to
the extent that dissolved oxygen concen-
trations in the upper part of the hypo-
limnion (measured) became too low to
allow the Daphnia to retain the zone as a
refuge from fish predation. Consequently,
the Daphnia 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 stimulat-
ing 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 artif ically, Daphnia could find
a refuge from the fish and remain
abundant enough to limit the size of the
algal population.
Reference
Lamarra, V. A. Experimental studies of
the effect of carp (Cyprinus carpio) on
the chemistry and biology of lakes.
Ph.D. Thesis. University of Minnesota,
Minneapolis. 1975.
. S. GOVERNMENT PRINTING OFFICE: 1983/659-095/1922
-------
Joseph Shapiro is with the University of Minnesota, Minneapolis, MN; Bruce
Forsberg is with the Institute Nacional de Amazonia, Manaus, Brazil; Vincent
Lamarra is with Utah State University, Logan, UT; Gun/Ha Lindmark is with the
University of Lund, Lund, Sweden; Michael Lynch is with the University of
Illinois, Urbana, IL; Eric Smeltzer is with the Department of Water Resources
and Environmental Engineering, Montpelier, VT; and George Zoto is with the
New England Aquarium, Boston, MA.
Charles Powers is the EPA Project Officer (see below).
The complete report, entitled "Experiments and Experiences in Biomanipulation:
Studies of Biological Ways to Reduce Algal Abundance and Eliminate Blue-
Greens," (Order No. PB 83-148 098; Cost: $22.00, subject to change} will be
available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Environmental Research Laboratory
U.S. Environmental Protection Agency
200 SW 35th Street
Corvallis. OR 97333
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $30O
PS 0000329
AG£NCY
------- |