JournalofPlanktonReseardiVoI.19no.il pp.1687-1711, 1997
Carbon dynamics in the 'grazing food chain' of a subtropical lake
Karl E.Havens and Therese L.East
Ecosystem Restoration Department, South Florida Water Management District,
West Palm Beach, FL 33416-4680, USA
Abstract. Studies were conducted over a 13 month period at four pelagic sites in eutrophic Lake Okeechobee, Florida (USA), in order to quantify carbon (C) uptake rates by size-fractionated phytoplankton, and subsequent transfers of C to zooplankton. This was accomplished using laboratory I4C
tracer methods and natural plankton assemblages. The annual biomass of picoplankton (<2 p.m),
nanoplankton (2-20 u.m) and microplankton (>20 u.m) averaged 60, 389 and 100 jig C I"1, respectively, while corresponding rates of C uptake averaged 7,51 and 13 u.g C I"1 rr1. The biomass of microzooplankton (40-200 u.m) and macrozooplankton (>200 jj.m) averaged 18 and 60 u,g C t~',
respectively, while C uptake rates by these herbivore groups averaged 2 and 3 ng C I"1 rr1. There were
no strong seasonal patterns in any of the plankton metrics. The ratio of zooplankton to phytoplankton
C uptake averaged 7% over the course of the study. This low value is typical of that observed in
eutrophic temperate lakes with small zooplankton and large inedible phytoplankton, and indicates
ineffective C transfer in the grazing food chain. On a single occasion, there was a high density (>40
H) of Daphnia lumholtzii, a large-bodied exotic cladoceran. At that time, zooplankton community C
uptake was >20 u.g C 1"' rr1, and the ratio of zooplankton to phytoplankton C uptake was near 30%.
If D.lumholtzii proliferates in Lake Okeechobee and the other Florida lakes where it has recently
been observed, it may substantially alter planktonic C dynamics.
Introduction
Pelagic carbon (C) dynamics have been a major focus of limnological research in
recent decades (e.g. Hillbricht-Ilkowska, 1977; Sommer, 1989; Weisse and Stockner, 1992). The underlying goal has been to understand better the food web that
ultimately supports fish and other higher trophic levels. Much of the research has
focused on complex interactions that occur near the base of the web, among
phytoplankton, bacteria, protozoa and zooplankton. Several major paradigms
have emerged from that work, including: (i) that picoplankton (planktonic
autotrophs <2.0 u,m in diameter) can account for substantial fractions of total
phytoplankton biomass and productivity (Stockner and Shortreed, 1989); (ii) that
microzooplankton (rotifers, copepod nauplii and ciliate protozoa) can account for
a considerable fraction of total zooplankton community grazing (Mazumder et al,
1990; Lair, 1991); and (iii) that substantial amounts of C and energy may flow
through microbial-based pathways (Sanders et al, 1988). These features may vary
with trophic state. For example, the relative productivity of picoplankton appears
to be higher in oligotrophic (Stockner and Shortreed, 1989) than in mesotrophic
and eutrophic lakes (Sondergaard et al., 1988; Weisse, 1988; Szelag-Wasielewska,
1997), and the relative importance of microzooplankton grazing is highest in
eutrophic, cyanobacteria-dominated lakes, especially when the plankton do not
include large daphnids (Gliwicz, 1990; Havens, 1991). Microbial pathways may
also become more important, relative to direct phytoplankton-zooplankton links,
in hypereutrophic lakes (Gliwicz, 1990; Weisse and Stockner, 1992).
These advances in understanding of the planktonic community have come primarily from research on temperate ecosystems, including the Laurentian Great
© Oxford University Press
1687
K.E.Havens and TL-East
Lakes, various smaller lakes in North America, Europe and Asia, and the marine
environment. Few comparable studies have been conducted in tropical and subtropical regions. Crisman (1990) reviewed some of the features of lowland subtropical lakes, which generally are shallow, eutrophic, and have plankton
dominated by large cyanobacteria and small-bodied zooplankton. Large daphnids
are conspicuously absent. On the basis of conceptual models developed for European lakes nearly 30 years ago (Gliwicz, 1969), and recently validated for temperate lakes in North America (Porter, 1997), we predict that the conditions in
subtropical lakes will result in a low efficiency of Cflowin the grazing food chain.
In the present study, we tested this prediction by conducting simultaneous
measurements of photosynthetic C uptake by phytoplankton and C transfers to
grazing zooplankton. Ratios of zooplankton to phytoplankton C uptake were
considered in the context of plankton biomass and community structure.
Method
Study site
Lake Okeechobee (Figure 1) is a large (A<, ~ 1800 km2), shallow (zmean = 2.7 m)
lake located at 26°58'N, SO^CW, at an elevation of 0.0 m above sea level. The
natural lake was formed several thousand years ago during oceanic recession. It
is polymictic, with temperature variation between 18°C in winter and 32°C in
summer, and eutrophic, with mean total phosphorus (TP), chlorophyll a and
Secchi disk values of 100 u.g I"1, 30 u.g I"1 and 50 cm, respectively (Havens et al,
1994). The phytoplankton are dominated by low-light-adapted cyanobacteria, but
blooms of high-light-adapted diazotrophic taxa occur at times and locations of
increased light availability (Havens et al, 1997). The zooplankton are dominated
by rotifers and calanoid copepods (Beaver and Havens, 1996), as well as a recent
immigrant — the exotic cladoceran Daphnia lumholtzii. This species reaches a
length of >1.5 mm (not including spines) and, as such, is much larger than any
native cladoceran found in Florida lakes (Crisman and Beaver, 1990).
Sampling
We sampled four sites (Figure 1) that represented the documented range of
pelagic limnological conditions in Lake Okeechobee. The North site is nutrient
rich and displays high levels of algal biomass. The Central site is also nutrient rich,
but displays low algal biomass due to frequent light-limiting conditions caused by
wind resuspension of mud bottom sediments (Phlips et al., 1997). The South site,
located near the pelagic-littoral interface, has lower nutrient concentrations,
greater light availability, and highly variable levels of algal biomass that may be
controlled in part by pulse inputs of nutrients from rainfall (Phlips et al., 1994)
and/or interactions between phytoplankton and benthic algae (Havens et al,
1996a). The West site occurs in a transitional region with characteristics intermediate to those of the central lake and littoral margin zones.
Sampling was performed in August and November 1995, and January, March,
May and August 1996, in order to account for seasonal variation in the measured
1688
Carbon dynamics in grazing food chain
N
A
0
5
10
15 Kilometers
Fig. 1. Map of Lake Okeechobee, Florida, USA, showing the locations of the four sampling sites. The
shaded region is a littoral zone with emergent macTophytes.
states and rates. At each sampling site, Secchi depth was measured with a 20 cm
black and white disk, and water temperature at mid-depth using a Hydrolab Surveyor water analyzer. Integrated water samples were collected using a 3-cmdiameter PVC tube that was extended from the surface to -0.5 m above the
sediments. A sufficient number of tubes was collected to fill two 20 1 Nalgene
carboys. From each carboy, a 100 ml aliquot was removed and fixed with acid
Lugol's solution for phytoplankton enumeration. The carboys were then placed
under a tarp to prevent exposure to direct sunlight until return to the laboratory.
At each site, the contents of two additional carboys were passed through a 35 u.m
plankton net to collect zooplankton. The collected animals were preserved with
cold formalin-sucrose for enumeration and body size measurements.
1689
K.E.Havens and T.L.East
Overview of tracer studies
Microscopic analyses and 14C tracer studies were used to quantify five state variables and six rates, including: C standing stocks and uptake rates for picophytoplankton (PICO), nano-phytoplankton (NANO) and micro-phytoplankton
(MICRO), C standing stocks for micro-zooplankton (MICZ) and macrozooplankton (MACZ), and C uptakes by MICZ and MACZ during grazing on
phytoplankton. It was not possible to fractionate PICO from NANO for grazing
studies without damaging the nanoplankton cells. Therefore, grazing rates were
determined on the combined 0.2-20 jim size class, and the >20 u.m size class. The
fractionation for phytoplankton followed that suggested by Sieburth et al. (1978).
Zooplankton fractionation was consistent with that used by Havens (1993) for
studies of C transfer in temperate lakes. In Lake Okeechobee, preliminary studies
using different mesh sizes (300, 200,100 and 40 \i.m) indicated that the >200 u.m
size class (MACZ) contained >95% (by dry weight) copepod adults, copepodids
and cladocerans, while the 40-200 u.m size class (MICZ) contained rotifers and
nauplii.
Laboratory methods
The biomass and C content of each phytoplankton size class were determined
from chlorophyll a concentrations measured on fractionated water samples. We
used this approach, rather than estimating biomass directly from microscopic
counts, because preliminary evaluation of phytoplankton counts gave a much
higher degree of variability (among replicates of a given sample) than did chlorophyll a measurements. More importantly, the magnitude of that variability was
influenced by the composition of the phytoplankton assemblages (i.e. it was lower
in samples dominated by unicells, and higher in samples dominated by filaments
or colonies). Nevertheless, the taxonomic composition of phytoplankton can significantly affect grazing interactions with zooplankton. Thus, we performed
microscopic counts using the procedure of Lund et al. (1958), in order to identify
the numerically dominant taxa at each site.
For chlorophyll a analysis, a measured volume (generally 100 ml) of whole-lake
water from each carboy was passed through a 20 jxm Nitex® screen, with the filtrate passing into a filter funnel with a Whatman GFF filter, which has an effective porosity of 0.2 u,m (Chavez et al., 1995). The cells retained on the 20 u.m
screen were back-flushed onto a GFC filter, and both filters were evacuated.
These two fractions represented PICO+NANO and MICRO, respectively. A
third aliquot of water was passed through a 2.0 u.m Poretics polycarbonate filter,
and the filtrate collected onto a GFF filter for the determination of PICO chlorophyll. NANO chlorophyll was then calculated by difference. All chlorophyll concentrations were determined according to standard spectrophotometric methods
(American Public Health Association, 1993), after grinding with a tissue grinder
and overnight acetone extraction. Phytoplankton C standing stocks were estimated using a C to chlorophyll a conversion factor of 25:1, which falls near the
middle of the range reported in the literature (Banse, 1977; Bowie et al, 1985;
Cloern etal., 1995).
1690
Cartron dynamics in grazing food chain
Phytoplankton C uptake rates were determined using the standard 14C light
and dark bottle method (Vollenweider, 1974). Production bottles (two light and
one dark 150 ml borosilicate glass bottles) were spiked with 0.5 ml of
[14C]bicarbonate having a specific activity of 1 jxCi ml"1, and incubated for 2 h at
ambient lake temperature and a constant irradiance of 150 \LE m~2 s"1. This
approximates irradiance in the upper euphotic zone of the lake, and recent photosynthesis-irradiance curves (Gu et ai, 1997) indicate that rates measured herein
approximate optima, rather than water column means for phytoplankton photosynthesis.
Following the incubation period, production bottles were held in a dark box
and covered with ice. The water in each bottle was size fractionated, using successive 20,2.0 and 0.2 |xm porosity filters. For the two smaller size fractions, small
volumes (<10 ml) were filtered, and in all cases a low vacuum (<0.1 atm) was
applied, in order to minimize damage to cells. All filtration procedures were performed in near total darkness (<10 (xE m~2 s"1). Filters containing phytoplankton
fractions were fumed over concentrated HC1 to remove unassimilated 14C and
placed into vials with Ecolume® liquid scintillation cocktail. The samples were
counted on a Rack Beta scintillation counter, with correction for background and
color quenching. Carbon uptake rates (^g C Hh" 1 ) for each phytoplankton size
class were calculated according to Vollenweider (1974), with the amounts of available 12C determined from alkalinity, pH and temperature measurements of incubated samples. Net C uptake rates were calculated as the difference between light
and dark bottles.
Recently, there has been a controversy regarding whether one should size fractionate phytoplankton before or after productivity incubations. Proponents of the
pre-incubation fractionation approach (Fahnenstiel et al., 1994) argue that commonly used post-incubation membrane filtration procedures damage cells and
cause loss of label into smaller size fractions. Proponents of post-incubation fractionation (Fumas, 1987) argue that their approach minimizes stress to cells before
incubation, which presumably could cause unnatural C uptake rates. Given the
importance of the fractionation procedure to the present study, we conducted
preliminary comparisons of C uptake rates using both approaches. These studies
also permitted an examination of time-specific uptake rates, to confirm that a 2 h
incubation was appropriate for our full-scale studies. Incubations were performed
in May 1995 using water collected at the South site in Lake Okeechobee, from
which split samples were fractionated either before or after incubations lasting
60, 160 and 240 min. Time-specific C uptake rates were calculated as described
above.
In contrast to the results of previous studies, no significant differences were
found for any size fraction between the pre- and post-incubation procedures.
Furthermore, time-specific uptake rates did not vary among incubation periods.
Therefore, we selected post-incubation fractionation because it is a simpler
approach.
The C contents of MACZ and MICZ were estimated from microscopic counts
and measurements, after fractionation with a 200 u.m Nitex® screen. For the
MACZ fraction, aliquots were counted at 50x magnification until at least 400
1691
K.E.Havens and T.UEasi
individuals had been enumerated, and taxon densities (numbers I"1) were calculated based on sample volumes (20 1) and the percentages of the volumes actually counted. If samples contained <400 individuals, entire contents were
enumerated. At least 10 individuals of each taxon (copepod adults and copepodids treated as separate taxa) were also measured, and their dry weights approximated from lengths using the equations of Culver et al. (1985). Where necessary,
equations for taxa of similar body morphology were used, since that study did not
include some of the species encountered in Lake Okeechobee. Taxon dry weights
were converted to C content using a multiplication factor of 0.48 (Andersen and
Hessen, 1991), and population C standing stocks were determined as density
times mean individual C content. The same methods were used to determine the
C content of MICZ, except that counts and measurements were carried out at
100X magnification. Dry weights were determined from MICZ biovolumes,
assuming that the animals had unit density and that dry weight = 0.1 wet weight
(Pace and Orcutt, 1981). Biovolumes were estimated by approximating body
shapes to regular geometric solids.
Zooplankton C uptake rates were measured using the procedure of Bogdan
and McNaught (1975), as modified by Havens (1993). The natural phytoplankton
were labeled with 14C and used as the food source for measuring community
grazing rates by MICZ and MACZ fractions. On each sampling date, whole-lake
water was poured through a 40 n,m Nitex® screen to remove zooplankton. One
liter aliquots of the filtrate from each site were dispensed into two glass jars, one
of which was spiked with 10 n-Ci of [14C]bicarbonate solution. The jars were
capped and incubated for -20 h at ambient lake temperature and 150 ^E m~2 s"1
irradiance.
Prior to conducting grazing studies, zooplankton from the carboys (stored
overnight in a dark room) were collected into 'grazing chambers', opaque plastic
cylinders (30 cm long X 5 cm diameter) with bottoms covered with 40 or 200 n,m
Nitex® screen. Transfer of the water and animals was accomplished by gentle
siphoning, while the chamber was immersed in a 4 1 plastic bottle. Siphoning into
the chamber continued until the bottle was full (at which time, 4 I of zooplankton had been captured inside each chamber, with phytoplankton and water
passing through the mesh base). The chambers were placed in a bucket partially
filled with lake water and allowed to equilibrate for -20 min.
Once the zooplankton chambers were established, phytoplankton from the
same site were size fractionated. The contents of each jar of filtered lake water
(one spiked with 14C) were passed through a 20 p,m Nitex® screen, with the filtrate (PICO+NANO) exiting into a plastic beaker. The retained MICRO cells
from the 14C-labeled water were back-flushed into the unlabeled PICO+NANO
fraction, while the MICRO cells from the unlabeled water were back-flushed into
the 14C-labeled PICO+NANO fraction. This procedure resulted in two complete
algal assemblages that differed only in terms of the fraction carrying the 14C label.
In a preliminary control study, we conducted an isotope balance and found that
the combined activity of the two size fractions accounted for at least 92% of the
total activity of unfiltered water from each lake site. This provides evidence that
algal cells were not ruptured during the 20 jjun size-fractionation process.
1692
Carbon dynamics in grazing food chain
Grazing experiments commenced immediately after the algal assemblages
were reconstituted and, for each site, were completed within 30 min. Measurements of size-fractionated samples collected from the algal assemblages indicated
that only a small percentage of the particulate label was transferred between the
two phytoplankton size classes during this time period.
To determine MACZ grazing rates, triplicate grazing chambers (200 u.m) were
gently placed in each algal assemblage and grazing allowed to occur for 5 min.
This time period is less than the reported gut-passage times of most crustacean
and rotiferan zooplankton (Downing and Rigler, 1984). One chamber from each
triplicate set was immersed in boiling water to kill the zooplankton before the 5
min incubation. This chamber served as a control to correct for non-grazing
uptake of 14C by animals' carapaces, which can account for a considerable fraction of the total uptake (Havens, 1991). During the incubation period, 10 ml
samples of the algal assemblages were collected, filtered onto Poretics 0.2 (xm
polycarbonate filters, fumed over HC1, and placed in vials with scintillation cocktail for the determination of feeding suspension activities. At the end of the
grazing period, the chambers were raised from the water and the animals rinsed
onto 20 jj,m nylon niters, which were immediately evacuated and placed into vials
with scintillation cocktail. Activities were determined by scintillation counting
with correction for background and color quenching. Carbon uptake rates were
determined as:
ug C I"1 h"1 = [(d.p.m. I"1 live) - (d.p.m. H killed control)] X T X C
(d.p.m. ml"1 feeding suspension)
where T is a temporal correction factor (per hour) of 12 and C is the calculated
C content of the PICO+NANO or MICRO phytoplankton fraction.
The same procedure was used to measure grazing by MICZ, except that 40 u.m
grazing chambers were used, and grazing was measured only on the
PICO+NANO algal fraction. Previous studies have shown that the microzooplankton taxa occurring in Lake Okeechobee (Keratella, Polyarthra, Brachionus,
Filinia and copepod nauplii) generally do not consume filamentous and colonial
MACRO algae.
In our opinion, the laboratory method used here was superior to the more commonly used in situ Haney chamber method (Haney and Hall, 1975) because our
approach allowed for use of natural phytoplankton assemblages as the food
source, whereas the Haney chamber approach estimates the community filtering
rate (f, ml"1 I"1 time"1) by allowing zooplankton to graze on radiolabeled 'model'
algal cells. In most published studies, the model has been a small flagellate such
as Chlamydomonas or Cryptomonas, which may not resemble the naturally
occurring algae. To estimate the C uptake rate, one multiplies/by the measured
C standing stock of natural phytoplankton (u.g C I"1). The underlying assumption,
that the model cells are grazed at the same rate as natural algae, may not be valid,
especially if large, inedible taxa predominate. By using radiolabeled natural
plankton to estimate /, we overcome this problem. Nevertheless, there are some
1693
K.E.Havens and T.L.East
concerns regarding this approach. First, it is possible that during the time required
to radiolabel natural phytoplankton (20 h in this case), some of the label may have
been excreted as dissolved organic I4C, and subsequently taken up by bacteria
and protozoa in the microbial loop. If this occurred, the calculated filtering rates
may have reflected grazing on a broader size range of natural particles than
desired. Nevertheless, we still view this as superior to the results based on some
laboratory-cultured model. Furthermore, size fractionation of the water prior to
grazing studies, as done herein, can ameliorate problems associated with label
transfer from large to small particles.
Our grazing studies may also be criticized because they subjected zooplankton
to unnatural conditions, which may have affected their filtering rates (Downing
and Rigler, 1984). We took care to minimize disturbance of the zooplankton, conducting all transfers of animals using gentle siphoning of water between containers, and allowing time for acclimation. To address these possible concerns, and to
determine whether our grazing experiments gave rates that were comparable to
those derived from a Haney chamber approach, we conducted paired measurements at all four sites during March 1996. Haney chambers were loaded with
either 14C-labeled Chlamydomonas sp. (5 p-m), to estimate filtering rates on
PICO+NANO phytoplankton, or 14C-labeled Anabaena limnetica, to estimate filtering rates on MICRO phytoplankton. Duplicate in situ grazing experiments
were run for 5 min, after which we size-fractionated the MACZ and MICZ.
Laboratory studies, using the natural phytoplankton as the feeding suspension,
were performed as described above. When the laboratory community filtration
rates (ml I"1 h"1) for animals grazing on natural PICO+NANO were compared to
Haney chamber rates for animals grazing on Chlamydomonas sp., there was good
agreement of results: MACZ (lab) = 0.9 ± 0.7 (SD) versus MACZ (Haney) = 0.8
± 0.5 ml I-1 h-1 and MICZ (lab) = 0.9 ± 0.2 versus MICZ (Haney) = 1.1 ± 0.4 ml
I"1 h->. This reflects the fact that the natural NANO+PICO was strongly dominated by nanoflagellates and other similar sized nanoplankton algae, i.e. the algal
model was a realistic one in this case. The similarity between laboratory and field
measurements of grazing on MICRO was not as striking, although there was not
a statistically significant difference. MACZ grazing on 14C-labeled A.limnetica
averaged 2.7 ± 1.5 ml H h"1, while in the laboratory the rate was 5.6 ± 6.0 ml H
h"1. Because the MICRO phytoplankton fraction was dominated by relatively
small colonies of Merismopedia and Aphanocapsa, high filtration rates were
measured in the laboratory, while lower rates were measured using the large, and
almost certainly less edible, A.limnetica. These results underscore the advantage
of our approach.
Methodological limitations
Before considering the results of this investigation, certain limitations of the
methods must be considered. First, it must be recognized that the C flow models
derived herein are periodic 'snapshots' of the community that do not take into
account possible changes of the system in response to short-term variations of
forcing functions (pulse nutrient inputs, high inflow events, wind storms, etc.), or
1694
Carbon dynamics in grazing food chain
diurnal variation in phytoplankton C uptake or zooplankton grazing. Furthermore, estimates of C uptake are based on daytime measurements. Because phytoplankton take up C only during the 12-14 h of daylight, while herbivores may graze
for 24 h per day, ratios of zooplankton to phytoplankton C uptake could be underestimated. However, this may be offset by the use of optimal light levels in the
primary productivity incubations. The use of a fixed C to chlorophyll ratio (25:1)
introduces further uncertainty, because a wide range of values (from 5 to 200:1)
have been reported in the literature. In summary, the C flux models presented
herein provide relatively coarse-scale information for comparing major patterns
across time and space within the lake, and with other published results. They probably do not contain a sufficient level of detail (or accuracy) for constructing mechanistic models of lake C dynamics, unless additional measurements are performed.
Results
Physical conditions at the study sites
Water temperatures ranged from >30°C in August to <20°C in January (Figure
2A). On any given date, temperatures measured (near mid-day) at the four sites
varied by <2°C. Secchi transparencies varied between 18 and 80 cm (Figure 2B),
with the lowest values recorded in winter (November-February), the windy
season in subtropical Florida, and the highest values in summer (May-September). On most occasions, transparencies were lowest at the Central site and
highest at the South or North sites.
Plankton taxonomic composition
Filamentous and colonial non-heterocystous cyanobacteria (Lyngbya, Oscillatoria, Merismopedia and Microcystis) dominated the phytoplankton on most
sampling dates (Table I). However, in November and January, small nanoflagellates (3-5 ^m) were common at the North and Central sites, and in August, the
diazotrophic cyanobacteria Anabaena circinalis and A.limnetica were among the
dominants at the North, West and South sites.
Microzooplankton biomass was dominated by cyclopoid and calanoid nauplii,
and six rotifer taxa, whose relative biomass varied with season and location
(Table II). The non-loricate Gastropus spp. were dominant only from November
to March, Keratella cochlearis was a dominant from November to August, and
Polyarthra vulgaris was a dominant on all sampling dates. Brachionus havanaensis and Filinia longiseta were never found at the Central pelagic site, a pattern
that was identified in a more comprehensive study of zooplankton community
structure between 1987 and 1992 (Beaver and Havens, 1996).
Macrozooplankton biomass was dominated by Diaptomus dorsalis in 22 out of
24 samples examined (Table III). Cyclopoid copepodids (Acanthocyclops vernalis, Mesocyclops edax and Tropocyclops prasinus) were often co-dominant, as
was the medium-sized cladoceran (0.5—1.0 mm) Diaphanosoma brachyurum. In
August 1996, high densities of D.lumholtzii were recorded at the Central (12
animals I"1), West (41 animals I"1) and South (7 animals I"1) sites.
1695
ICE.Havens and T.LEast
30
EESC
t
f
n
CENTRAL
•
f
g
DC * "
CD
LU
Q
•
NORTH
ff
WEST
10
SOUTH
n
A S O N D J F M A M J J A
MONTH
B100
80
O
•
CENTRAL
60
•
LU
Q
O
O
LU
CO
T
•
:
T
•
n
20
•
•
B
T
T
•
NORTH
•
WEST
•
•
SOUTH
•
A S O N D J F M A M J J A
MONTH
Fig. 2. Lake water temperatures (A) and Secchi depths (B) at the four pelagic sites on the six sampling dates in 1995 and 1996.
Carbon flux summary
Lake-wide C fluxes (Figure 3), averaged over all sites and sampling dates, indicate three major features of the grazing food web: (i) MACZ dominate zooplankton biomass, but account for a similar portion of total C uptake from
phytoplankton relative to MICZ; (ii) NANO dominate both the biomass and C
uptake of phytoplankton; (iii) a relatively small fraction of the C taken up by
phytoplankton in photosynthesis is transferred directly to zooplankton. These
1696
Carbon dynamics in grazing food chain
Table I. Dominant (>10% total numerical abundance) phytoplankton taxa at the four pelagic
sampling sites in Lake Okeechobee on six sampling dates from August 1995 to 1996. Taxa are listed in
order of decreasing density, and do not include picoplankton
Month
Central
North
West
South
August 95
LLIM
MINC
MTEN
MINC
LLIM
OLIM
MTEN
OLIM
LLIM
MTEN
MINC
OLIM
LLIM
MINC
November 95
NFLA
MTEN
NFLA
MTEN
MTEN
OLIM
OLIM
MTEN
January 96
MTEN
NFLA
ADEL
NFLA
ADEL
MTEN
LLIM
OLIM
NFLA
MTEN
NFLA
March 96
SSCH
NFLA
LLIM
OLIM
NFLA
MTEN
ADEL
ADEL
MTEN
LLIM
NFLA
ADEL
ADEL
LLIM
MTEN
NFLA
OLIM
MTEN
NFLA
OLIM
LCON
ALIM
OLIM
ACIR
LLIM
ALIM
OLIM
ALIM
MTEN
May 96
August 96
LLIM, Lyngbya limnetica; MINC, Microcystis incerta; MTEN; Merismopedia tenuissima; OLIM,
Oscillatoria limnetica; NFLA, nanoflagellates; ADEL, Aphanocapsa delicatissima; LCON, Lyngbya
contorta; SSCH, Sphaerocystis schroeteri; ACIR, Anabaena circinalis; ALIM, A limnetica.
features were generally seen across all sites and dates, although there were some
interesting exceptions (see below).
Phytoplankton biomass and carbon uptake
Total phytoplankton biomass varied from below 200 to >1800 fxg C h1 (Figure 4A).
There was no lake-wide seasonal pattern. On any given sampling date, maximal
values were most often found at the South or West sites. In August 1996, a bloom
(that included O.limnetica, A.circinalis, L.limnetica and A.limnetica) at the West
site had a total biomass that was 2-fold higher than any other observation. The
chlorophyll a concentration associated with this bloom was in excess of 70 p-g I"1.
The relative contributions of three algal size classes to total phytoplankton biomass
(Figure 4B) ranged from 0 to 55% for MICRO, from 25 to 100% for NANO and
from 0 to 22% for PICO. There was not a strong seasonal pattern, although the
relative contribution of NANO tended to be highest during January and March.
Phytoplankton C uptake rates displayed a weak seasonal pattern, with evidence of lower lake-wide rates in January and March, and varied from 15 to 160
jig C I"1 h"1 (Figure 5A). However, conclusions about seasonality should be considered with caution, given that only a single year of data were collected. Rates
exceeding 100 u.g C I"1 h~' occurred in every month except January and March.
The high rates in August 1995, when biomass was relatively low, reflect an active
1697
K.E.Havens and T.L.East
Table II. Dominant (>10% total biomass) microzooplankton taxa at the four pelagic sampling sites
in Lake Okeechobee on six sampling dates from August 1995 to 19%. Taxa are listed in order of
decreasing biomass, and copepod adults are summed with their copepodids
Month
Central
North
West
South
August 95
CYCN
CYCN
PVUL
CYCN
BHAV
CYCN
PVUL
November 95
CYCN
CYCN
CYCN
PVUL
GAST
CYCN
PVUL
GAST
January %
GAST
PVUL
GAST
CYCN
GAST
GAST
KCOC
March 96
CYCN
CALN
CALN
CYCN
PVUL
GAST
CALN
PVUL
CYCN
CALN
CYCN
KCOC
GAST
May%
CALN
CYCN
KCOC
CALN
CYCN
KCOC
PVUL
KCOC
CYCN
CALN
BANG
KCOC
CYCN
KCOC
PVUL
August %
PVUL
KCOC
CALN
CYCN
CALN
KCOC
BHAV
PVUL
BHAV
FLON
CYCN
KCOC
BHAV
CYCN
KCOC
FLON
CYCN, cyclopoid nauplii; PVUL, Polyarthra vulgaris; BHAV, Brachionus havanaensis; GAST,
Gastropus spp. KCOC, Keratella cochlearis; CALN, calanoid nauplii; BANG, Brachionus angularis;
FLON, Filinia longiseta.
Table HI. Dominant (>10% total biomass) macrozooplankton taxa at the four pelagic sampling sites
in Lake Okeechobee on six sampling dates from August 1995 to 19%. Taxa are listed in order of
decreasing biomass, and copepod adults are summed with their copepodids
Month
Central
North
West
South
August 95
DDOR
DBRA
DDOR
AVER
DDOR
AVER
DDOR
AVER
November 95
DDOR
DDOR
DDOR
TPRA
DDOR
TPRA
January %
DDOR
MEDA
DDOR
DDOR
MEDA
MEDA
DDOR
March %
DDOR
MEDA
DDOR
ETUB
DDOR
DDOR
MEDA
May%
DDOR
DBRA
DDOR
DBRA
DDOR
DBRA
DDOR
August %
DLUM
DDOR
DDOR
DBRA
MEDA
DLUM
DDOR
DDOR
DLUM
DDOR, Diaptomus dorsalis; DBRA, Diaphanosoma brachyurum; AVER, Acanthocyclops vemalis;
TPRA, Tropocydops prasinus; MEDA, Mesocyclops edax; ETUB, Eubosmma tubicen; DLUM,
Daphnia lumholtzil
1698
Carbon dynamics in grazing food chain
(
ZOOPLANKTON:
MICZ
1
(18±17ugC/L) I
(
MACZ
1
(60±22MgCVL) I
f
PHYTOPLANKTON:
PICO
(6O ± 91
]
'
NANO
^ I
(
MICRO
M
+ 78jjgC/L)
(389 + 204 pgC/L) I
7 + 6ugC/L/hr
51+25ugC/L/hr
[ (100 ±
/
13+15fjgC/L/hr
I DIC (24 + 7 mgCA.) 1
Fig. 3. Summary of carbon flux rates and pools, expressed as means (± 1 SD) across all sites and sampling dates during the study period. Acronyms are defined in the text.
phytoplankton assemblage. The relative contributions of three algal size classes
to total C uptake (Figure 5B) ranged from 0 to 45% for MICRO, from 50 to 100%
for NANO and from 0 to 45% for PICO.
Zooplankton biomass and carbon uptake
Total zooplankton biomass varied from 10 to 500 (xg C I"1 and did not display a
pronounced lake-wide seasonal pattern (Figure 6A). High values occurred in at
least one sample in August, January, May and August, and, overall, the West site
had the highest zooplankton biomass. The high values in August 1996 corresponded with the unusually high densities of D.lumholtzii at the West, Central
and South sites. Densities at the West site were especially high, at >40 animals \~x.
The body length of these large grazers averaged between 1.5 and 2.0 mm. The
relative biomass of MICZ ranged from 10 to 95% (Figure 6B). The strongest
dominance by MICZ was observed in March and May, while strongest MACZ
dominance occurred in August 1996, coincident with the D.lumholtzii peak.
There was a weak seasonal pattern in zooplankton C uptake rates, with
maximal values in August (1995 and 1996) and minimal values in January and
March (Figure 7A). The unusually high grazing rate in August 1996 at the West
station corresponded to the D.lumholtzii peak. We estimate that these animals
had filtering rates in excess of 1 ml animal"1 h"1. In August 1995, grazing was
dominated by MICZ at three of the four stations, while during other months,
MACZ were the dominant grazers, accounting for on average 60% of the C
uptake (Figure 7B). The greatest relative contribution of MACZ to total community grazing occurred in August 1996.
1699
K.E.Hivens and T.L.East
TOTAL PHYTOPLANKTON BIOMASS
2000
1500
1
1000
500
B
C N W S
C N W S
C N W S
AUG
NOV
JAN
C N W S
C N W S
MAR
MAY
i
C N W S
AUG
RELATIVE PHYTOPLANKTON BIOMASS
100
MICRO
NANO
PICO
Fig. 4 Total phytoplankton biomass, calculated from chlorophyll a concentrations (A) and relative
contributions of MICRO, NANO and PICO to total biomass (B) at the four sites on six sampling
dates in 1995 and 1996. In (A), the vertical bars are ± 1 SD of the mean.
1700
Carbon dynamics in grazing food chain
A
TOTAL PRIMARY PRODUCTIVITY
200
160
t 120
80
40
B
ll
C N W S
C N W S
C N W S
C N W S
C N W S
AUG
NOV
JAN
MAR
MAY
C N W S
AUG
RELATIVE PRIMARY PRODUCTIVITY
100
80
•
MICRO
60
HI
occ
w
Q.
40
NANO
20
C N W S
C N W S
C N W S
C N W S
AUQ
NOV
JAN
MAR
I
C N W S
MAY
I
PICO
C N W S
AUG
Fig. 5. Total phytoplankton productivity (A) and relative contributions of M ICRO, NANO and PICO
to total productivity (B) at the four sites on six sampling dates in 1995 and 1996. In (A), the vertical
bars are ± 1 SD of the mean.
1701
K.E.Havcns and T.UEast
TOTAL ZOOPLANKTON BIOMASS
300
200
O)
100
C N W S
MAY
C N W S
AUG
RELATIVE ZOOPLANKTON BIOMASS
MICZ
MACZ
C N W S
AUG
C N W S
C N W S
C N W S
C N W S
NOV
JAN
MAR
MAY
C N W S
AUG
Fig. 6. Total zooplanlcton biomass, calculated from dry weights (A) and relative contributions of
MICZ and MACZ to total biomass (B) at the four sites on six sampling dates in 1995 and 1996. In
(A), the vertical bars are ± 1 SD of the mean.
1702
Carbon dynamics in grazing food chain
COMMUNITY GRAZING RATES
C N W S
C N W S
C N W S
C N W S
C N W S
AUG
NOV
JAN
MAR
MAY
C N W S
AUG
RELATIVE GRAZING RATES
MICZ
MACZ
C N W S
C N W S
C N W S
C N W S
C N W S
AUG
NOV
JAN
MAR
MAY
C N W S
AUG
Fig. 7. Total zooplankton community grazing (A) and relative contributions of MICZ and MACZ to
total community grazing (B) at the four sites on six sampling dates in 1995 and 1996. In (A), the vertical bars are ± 1 SD of the mean.
1703
K.E.Havens and T.LEasi
Zooplankton.-phytoplankton carbon uptake
The percentage of phytoplankton C uptake that is passed on to zooplankton has
previously been used as a rough estimate of ecological transfer efficiency
(Havens, 1992), recognizing that it does not account for differential losses of C
via excretion or respiration. In the present study, this transfer efficiency was 7%,
when averaged over all sites and dates (Figure 8). There was evidence of a weak
seasonal pattern, with maximal values in August and minimal values in March.
Highest efficiencies corresponded with large D.lumholtz'u populations at the
Central and West sites in August 1996. Ratios of zooplankton:phytoplankton
biomass (data not shown here) were similarly low, averaging 12%. Biomass
ratios, however, are more difficult to interpret, given the substantial differences
in growth rates and generation times between phytoplankton and their grazers.
Discussion
Three general features of the Lake Okeechobee plankton web are indicated by
the results of this investigation, (i) Most of the primary productivity is associated
with phytoplankton >2 u>m in size, i.e. picoplankton productivity is relatively low
in this eutrophic system, (ii) Although macrozooplankton are the dominant
grazers, microzooplankton account for a large fraction of the total C flux from
phytoplankton to zooplankton. (iii) Only a small fraction of the C taken up by
the phytoplankton in photosynthesis is subsequently transferred to zooplankton
in grazing.
RELATIVE CARBON TRANSFER
C N W S
AUG
C N W S
NOV
C N W S
JAN
C N W S
MAR
C N W S
MAY
C N W S
AUG
Fig. 8. Ratio of total zooplankton community grazing to total phytoplankton productivity, at the four
sites on six sampling dates in 1995 and 1996. The vertical bars are ± 1 SD of the mean.
1704
Carbon dynamics in grazing food chain
Picoplankton relative productivity
The relatively small contributions of picoplankton to total phytoplankton
biomass and productivity in eutrophic Lake Okeechobee are consistent with the
notion that picoplankton importance declines along the trophic gradient. The
11% value for relative picoplankton productivity obtained here is similar to that
reported by Weisse (1988) for meso-eutrophic Lake Constance, but it is considerably lower than the 40-60% values reported by Stockner and Shortreed
(1989) for oligotrophic lakes in British Columbia. Likewise, Sondergaard et al.
(1988) documented that in P- and N-enriched enclosures in Lake Almind,
Denmark, picoplankton accounted for <10% of total phytoplankton C uptake,
but in the oligotrophic waters of the open lake, they accounted for >30%. Burns
and Stockner (1991) reported a significant inverse correlation between picoplankton relative biomass and trophic state (measured as chlorophyll a) in six
New Zealand lakes. Recently, Szeleg-Wasielewska (1997) found a 'significant
negative correlation between picoplankton contribution to total biomass and
total biomass' in Polish lakes. In highly eutrophic Lake Mikolajskie, Poland,
picoplankton have been observed to account for only 1% of the total phytoplankton biomass (Jasser, 1997).
One explanation for this relationship is that larger phytoplankton are superior
competitors for nutrients (Agusti et al., 1990) or less susceptible to grazing by zooplankton (Porter, 1973) in nutrient-rich lakes, while smaller phytoplankton
(picoplankton-size cells in particular) are superior competitors where nutrients
are scarce (Stockner and Shortreed, 1989).
Microzooplankton relative grazing
Microzooplankton grazing in Lake Okeechobee, as a percentage of total zooplankton community grazing, is typical of what has been reported for temperate
lakes with similar limnological features. Comparable values include: 55% in
shallow hypereutrophic Old Woman Creek Estuary, Ohio (Havens, 1991); up to
100% in eutrophic Lake Aydat, France (Lair, 1991); and 80% in eutrophic Star
Lake, Vermont (Bogdan and Gilbert, 1982).
Shallow, turbid eutrophic lakes with high densities of cyanobacteria are an
unfavorable habitat for large Daphnia, but may support high densities of rotifers
and other small herbivores (Gliwicz, 1990; Havens, 1991; Noges, 1997). This
reflects the fact that filamentous cyanobacteria can interfere with the grazing
activities of large cladocerans (Gliwicz and Lampert, 1990). Kirk and Gilbert
(1990) also showed that large cladocerans are negatively affected by high concentrations of fine suspended inorganic matter (because they consume the particles,
expend energy, but obtain little nutritional value), while small rotifers such as Keratella are not. Gliwicz (1986) observed that cladocerans with large amounts of clay
in their guts sink more rapidly, and must expend a greater amount of energy in
order to maintain their position in the water column. In Lake Okeechobee, high
abiotic turbidity occurs frequently, due to wind resuspension of bottom sediments.
This is a common feature of very shallow lakes (Nixdorf and Deneke, 1997).
1705
K.E.Havens and T.LEast
Two other variables, fish predation and water temperature, make eutrophic
subtropical and tropical lakes unfavorable for large daphnids, but suitable for
smaller zooplankton (Crisman 1990). Conditions that are deleterious to large
cladocerans may actually benefit rotifers, whose populations can be dramatically
impacted by predation and mechanical damage caused by large Daphnia species
(Gilbert, 1985).
In the present study, the lowest contribution of microzooplankton to total zooplankton grazing was observed in August 1996, when the large exotic D.lumholtzii
was abundant. In contrast to the native D.ambigua, D.lumholtzii can tolerate high
water temperatures (Venkataraman, 1981), and its long terminal spines may offer
some protection from fish predation. The biomass of rotifers was very low on this
date, especially at sites where the Daphnia occurred.
The high relative grazing rates of MICZ in August 1995 require some
additional discussion, because they coincided with relatively high total zooplankton community grazing rates, low total biomasses and low relative MICZ
biomasses. One possible explanation for these contrasting results is that there
were additional micrograzers present in the plankton that were not quantified as
part of the MICZ enumerations and biomass calculations. In particular, large ciliates were not counted here, yet they have been shown to be abundant at certain
times of the year in this shallow lake (Havens and Beaver, 1997), and the taxa
that occur include large oligotrichs, which consume small phytoplankton (Hecky
et ai, 1978).
Zooplankton:phytoplankton carbon uptake
The low ratios of zooplankton to phytoplankton C uptake are consistent with the
conceptual models developed for European lakes (Gliwicz, 1969, 1990). In the
absence of large effective cladoceran grazers, there is little direct transfer of C
from phytoplankton to macrozooplankton, especially when the phytoplankton
are dominated by filamentous and/or colonial cyanobacteria. Microzooplankton,
which accounted for a large portion of total zooplankton biomass in Lake Okeechobee, generally graze on small algal cells and bacteria (Pourriot, 1977; Gilbert
and Bogdan, 1981).
In a comparable study conducted at mesotrophic East Twin Lake, Ohio,
Havens (1992) measured a transfer efficiency of near 1%. The plankton community was dominated by large cyanobacteria (Anabaenaflos-aquaeand Dictyosphaerum ehrenbergianum), cyclopoid and calanoid copepods, and relatively
small cladocerans {Eubosmina coregoni and D.retrocurva). Lyche et al. (1996)
recently constructed C budgets for the plankton of mesotrophic Svinsjoen,
Norway, where the phytoplankton was dominated by diatoms and the zooplankton was dominated by cyclopoid copepods. Their model showed 58 u,g C I"1 day 1
entering phytoplankton via photosynthesis, and 2.6 and 2.5 u,g C H day 1 passing
directly to micro- and macrozooplankton, respectively. The ecological transfer
efficiency calculated from these values was 11%. The authors concluded that in
lakes lacking populations of large Daphnia, much of the C and energy that enter
the system in phytoplankton photosynthesis are lost as algal particulate material
1706
Carbon dynamics in grazing food chain
and dissolved organic carbon, which enters the microbial loop and subsequently
exits as bacterial and protozoan respiration. Recently, Noges (1997) observed
that in eutrophic Lake Vortsjarv, Estonia, with cyanobacteria-dominated phytoplankton and rotifer-dominated zooplankton, the ratio of zooplankton to phytoplankton C uptake was only 2%. The author concluded that, in that lake, the
'detrital food chain is prevalent'.
In contrast, high ecological transfer efficiencies have been reported where
small phytoplankton and Daphnia (even moderate-sized species) co-occurred.
Gamier and Mourelatos (1991) reported a value of 75% for Certil Lake, France,
where the dominant phytoplankton taxa were 'small edible' Chlamydomonas,
Cryptomonas, Cyclotella and Oocystis, and the dominant zooplankton were
D.cucullata and D.hyalina. Daphnia can also form an effective linkage between
the microbial loop and higher trophic levels in the grazing food web (Riemann,
1985).
Pathways for carbonflowin subtropical plankton webs
When the results of this study are placed in the context of other research conducted on subtropical Florida lakes and temperate lakes having similar plankton
structures, a hypothetical model of Cflowemerges (Figure 9). The present study
demonstrated that little C flowed directly from phytoplankton to zooplankton.
However, high densities of bacterioplankton and ciliate protozoa in Lake
Okeechobee and other subtropical lakes (Beaver et al, 1988) suggest that the
microbial loop may be a relatively more important pathway. Given the high
Fig. 9. A simple model of carbon flaxes in the pelagic food web of shallow subtropical lakes. The
arrows indicate pathways that are presumed to be major (thick arrows) and minor (thin arrows). Pathways (1) and (4) are based on the results of the present study. Pathways (3), (6) and (10) are based
on previous studies conducted in Lake Okeechobee (Crisman and Beaver, 1990; Havens et al, 1996b),
and all other pathways are based on published results from other eutrophic lakes, as cited in the text.
Not all pathways of carbon flux are shown.
1707
K-E-Havens and T.LEasi
abundances of large-bodied calanoid and cyclopoid copepods encountered in the
lake, important pathways may also include predation by these animals on small
rotifers and nauplii, which in turn consume nanoplankton, picoplankton and bacteria (Gilbert and Bogdan, 1981). Large calanoid copepods, in the size range
encountered in Lake Okeechobee, may also graze directly on ciliate protozoa
(Burns and Schallenberg, 1996), which in turn consume pico- and nano-sized
plankton.
It has been suggested (Crisman and Beaver, 1990) that, in subtropical lakes,
fish consumption of phytoplankton is another important pathway for C flow. At
present, however, this hypothesis is based solely on the results of uncontrolled
experiments. When the plankton of Lake Okeechobee were excluded (inside
mesocosms) from fish predation, both phytoplankton and zooplankton increased
in density. This response is consistent with the fact that the fish assemblage contains a high density of gizzard shad (Dorosoma cepedianum). These fish have the
ability to filter plankton over a wide size range, from large macrozooplankton to
microphytoplankton, and as Crisman and Beaver (1990) noted, they 'prey upon
the small-bodied macrozooplankton and feed on the algal size spectrum that
would otherwise not be grazed due to the absence of large-bodied macrozooplankton'.
Recent studies of connectance webs in Lake Okeechobee (Havens etai, 1996b)
indicate that detrital resources may also represent a major source of C to higher
trophic levels. Consensus is building that this is a general feature of aquatic and
terrestrial food webs (Polis and Strong, 1996).
No doubt there is a considerable amount of C reaching the higher trophic levels
in Lake Okeechobee and in other subtropical lakes, because they often display
high densities of fish (Crisman, 1990), including large top predators. From a fisheries management standpoint, understanding the structure and function of the
plankton food web is critical. Fisheries models based on the traditional view that
zooplankton eat algae, and subsequently are eaten by fish, simply may not be
effective in subtropical lakes. Furthermore, as Crisman and Beaver (1990) concluded, the recently popularized notion that phytoplankton blooms may be controlled by 'biomanipulation' offish stocks is also probably not reasonable in these
lakes, where there is little capacity for the development of zooplankton large
enough to graze the large cyanobacteria. One might conclude that the invasion
of Florida lakes by D.lumholtzii could alter this situation, because the animals fall
into the size range of daphnids shown previously to have dramatic grazing
impacts on phytoplankton (Lampert et ai, 1986). In a year-long survey of
D.lumholtzii, at over 40 pelagic sites in Lake Okeechobee, we have not encountered high densities. The August 1996 event in the present study was a striking
contrast, and at that time there was indeed a dramatic increase in the efficiency
of transfer of phytoplankton C to zooplankton. However, at the West site, where
D.lumholtzii achieved a density of >40 animals I"1, we also recorded >70 p,g I"1 of
chlorophyll a, due to a bloom of cyanobacteria. This finding appears to contradict the prediction that this animal can control the lake's phytoplankton;
however, it reflects just one observation. More research will be needed to reach
a definitive conclusion.
1708
Carbon dynamics in grazing food chain
Acknowledgements
The authors are grateful to Andrew Rodusky and Bruce Sharfstein for assistance
in the field and laboratory, and to Peter Doering, Soon-Jin Hwang, R.Thomas
James, Paul McCormick, Alan Steinman, Barry Rosen and two anonymous referees for constructive comments on a draft manuscript.
References
Agusti.S., Duarte.C.M. and Canfield,D.E.,Jr (1990) Biomass partitioning in Florida phytoplankton
communities. / Plankton Res., 13, 239-245.
American Public Health Association (1993) Standard Methods for the Examination of Water and
Wastewater, 17th edn. APHA, Washington, DC.
Andersen,T. and HessenJD.O. (1991) Carbon, nitrogen and phosphorus content of freshwater zooplankton. Limnol. Oceanogr., 36, 807-813.
Banse,K. (1977) Determining the carbon-to-chlorophyll ratio of natural phytoplankton. Mar. Biol,
41, 199-212.
BeaverJ.R. and Havens.K.E. (1996) Seasonal and spatial variation in zooplankton community
structure and their relation to possible controlling variables in Lake Okeechobee. Freshwater Biol.,
36, 45-56.
BeaverJ.R., Crisman,T.L. and Bienert.R.W.Jr (1988) Distribution of planktonic ciliates in highly
coloured subtropical lakes: comparison with clearwater ciliate communities and the contribution of
mixotrophic taxa to total autotrophic biomass. Freshwater Biol., 20, 51-60.
Bogdan.K.G. and GilbertJJ. (1982) Seasonal patterns of feeding by natural populations of Keratella,
Polyarthra, and Bosmina: clearance rates, selectivities, and contributions to community grazing.
Limnol Oceanogr., 27, 918-934.
Bogdan,K.G. and McNaught.D.C. (1975) Selective feeding by Diaptomus and Daphnia. Verh. Int. Ver.
Limnol, 19, 2935-2942.
Bowie.G.L. et al. (1985) Rates, Constants, and Kinetics Formulations in Surface Water Quality Modeling, 2nd edn. United States Environmental Protection Agency, Athens, GA.
Burns.C.W. and StocknerJ.G. (1991) Picoplankton in six New Zealand lakes: abundance in relation
to season and tropic state. Int. Rev. Ces. HydrobioL, 76, 523-536.
Burns.C.W. and Schallenberg,M. (1996) Relative impacts of copepods, cladocerans and nutrients on
the microbial food web of a mesotrophic lake. /. Plankton Res., 18, 683-714.
Chavez,F.P., Buck.K.R., Bidigare.R.R., Karl,D.M., Hebel,D., Latasa,M., Campbell,L. and NewtonJ.
(1995) On the chlorophyll a retention properties of glass fiber GFF filters. Limnol. Oceanogr., 40,
428-^33.
CloernJ.E., Grenz,C. and Vidergar-Lucas.L. (1995) An empirical model of the phytoplankton
chlorophyll:carbon ratio—the conversion factor between productivity and growth rate. Limnol.
Oceanogr., 40,1310-1321.
Crisman,T.L. (1990) Natural lakes of the southeastern United States: origin, structure, and function.
In Hackney,C.T, Adams.S.M. and Martin,W.H. (eds), Biodiversity of the Southeastern United States
— Animal Communities. John Wiley and Sons, New York.
Crisman,T.L. and BeaverJ.R. (1990) Applicability of biomanipulation for managing eutrophication
in the subtropics. Hydrobiologia, 200,177-185.
Culver.D.A., Boucherle,M.M., Bean,DJ. and Fletcher J.W. (1985) Biomass of freshwater crustacean
zooplankton from length-weight regressions. Can. J. Fish. Aquat. ScL, 42, 1380-1390.
DowningJ.A. and Rigler ,F.H. (1984) A Manual on Methods for the Assessment of Secondary Productivity in Fresh Waters. IBP Handbook No. 17. Blackwell Scientific, Oxford.
Fahnenstiel.G.L., Redalje,D.G. and Lohrenz,S.E. (1994) Has the importance of photoautotrophic
picoplankton been overestimated? Limnol. Oceanogr., 39,432—438.
Fumas,MJ. (1987) Effects of prescreening on productivity of size-fractionated phytoplankton.
Limnol. Oceanogr., 32, 483-^491.
GarnierJ. and Mourelatos.S. (1991) Contribution of grazing in phytoplankton overall losses in a
shallow French lake. Freshwater Biol, 25,515-523.
GilbertJJ. (1985) Competition between rotifers and Daphnia. Ecology, 66,1943-1950.
GilbertJJ. and Bogdan,K.G. (1981) Selectivity of Polyarthra and Keratella for flagellate and aflagellate cells. Verh. Int. Ver. Limnol, 21,1515-1521.
1709
K.E.Havens and T.L.East
Gliwicz,Z.M. (1%9) Studies on the feeding of pelagic zooplankton in lakes with varying trophy. Ekol.
Pol., 17,663-708.
Gliwicz,Z.M. (1986) Suspended clay concentration controlled by filter-feeding zooplankton in a
tropical reservoir. Nature, 323, 330-332.
Gliwicz,Z.M. (1990) Why do cladocerans fail to control algal blooms? Hydrobiologia, 200, 83-97.
Gliwicz,Z.M. and Lampert,W. (1990) Food thresholds in Daphnia species in the presence and absence
of blue-green filaments. Ecology, 71,691-702.
Gu,B., Havens.K.E., Schelske.C.L. and Rosen.B.H. (1997) Uptake of dissolved nitrogen by phytoplankton in a eutrophic subtropical lake. / Plankton Res., 19, 759-770.
HaneyJ.F. and Hall.DJ. (1975) Die! vertical migration andfilter-feedingactivities of Daphnia. Arch.
Hydrobiol, 75, 413-441.
Havens.K.E. (1991) The importance of rotiferan and crustacean zooplankton as grazers of algal
productivity in a freshwater estuary. Arch. Hydrobiol, 122, 1-22.
Havens,K.E. (1992) Acidification effects on the algal-zooplankton interface. Can. J. Fish. Aquat. ScL,
49, 2507-2514.
Havens.K.E. (1993) An experimental analysis of macrozooplankton, microzooplankton and phytoplankton interactions in a temperate eutrophic lake. Arch. Hydrobiol., 127, 9-20.
Havens.K.E. and BeaverJ.R. (1997) Consumer and resource control of ciliate protozoa: experimental
studies in a shallow, copepod-dominated subtropical lake. Arch. Hydrobiol., in press.
Havens,K.E., Hanlon,C. and James,R.T. (1994) Seasonal and spatial variation in algal bloom
frequencies and intensities in Lake Okeechobee, Florida, USA. Lake Reserv. Manage., 10,
133-138.
Havens.K.E., East.T.L., Meeker.R.H., Davis,W.P. and SteinmanAD. (1996a) Phytoplankton and
periphyton responses to in situ experimental nutrient enrichment in a shallow subtropical lake. /
Plankton Res., 18, 551-566.
Havens.K.E., Bull,L.A., WarTen.G.L., Crisman.T.L., Phlips.EJ. and SmithJ.P. (1996b) Food web
structure in a subtropical lake ecosystem. Oikos, 75, 20-32.
Havens,K.E., Phlips,E J. and Cichra,M.F. (1997) Light availability as a possible regulator of cyanobactena species composition in a shallow subtropical lake. Freshwater Bioi, in press.
Hecky,R.E., FeeJEJ., Kling.H. and RuddJ.W.M. (1978) Studies on the planktonic ecology of Lake
Tanganyika. Fisheries and Marine Service Technical Report No. 816.
Hillbricht-Ilkowska,A. (1977) Trophic relations and energyflowin pelagic plankton. Pol Ecol. Stud.,
3,3-98.
Jasser.I. (1997) The dynamics and importance of picoplankton in shallow, dystrophic lake in comparison with surface waters of two deep lakes with contrasting trophic state. Hydrobiologia, 342,
87-93.
Kirk.K.L. and Gilbert J J . (1990) Suspended clay and the population dynamics of planktonic rotifers
and cladocerans. Ecology, 71, 1741-1755.
Lair,N. (1991) Comparative grazing activities of the rotifer and crustacean communities in a eutrophic
lake. Verh. Int. Ver. Limnol, 24, 924-927.
Lampert,W., Fleckner.W., Rai,H. and Taylor.B.E. (1986) Phytoplankton control by grazing zooplankton: a study of the spring clear-water phase. Limnol. Oceanogr, 31,478-490.
LundJ.W.G., Kipling.C. and LeCren.E.D. (1958) The inverted microscope method of estimating algal
numbers and the statistical basis of estimations by counting. Hydrobiologia, 11, 142-170.
LycheA-, Andersen,T., Christoffersen.K., HessenjS.O., Berger Hansen,P.H. and Klysner,A. (1996)
Mesocosm tracer studies 2. The fate of primary production and the role of consumers in the pelagic
carbon cycle of a mesotrophic lake. Limnol Oceanogr., 41, 475-487.
Mazumder,A., McQueen.DJ., Taylor,W.D., Lean.D.R.S. and Dickman.M.D. (1990) Micro- and
mesozooplankton grazing on natural pico- and nanoplankton in contrasting plankton communities
produced by planktivore manipulation and fertilization. Arch. Hydrobiol, 118, 257-282.
Nixdorf,B. and Deneke.R. (1997) Why 'very shallow' lakes are more successful opposing reduced
nutrient loads. Hydrobiologia, 342, 269-284.
Noges.T. (1997) Zooplankton-phytoplankton interactions in lakes Vortsjarv, Peipsi (Estonia) and
Yaskhan (Turkmenia). Hydrobiologia, 342,175-183.
Pace,M.L. and OrcuttJ.D. (1981) The relative importance of protozoans, rotifers and crustaceans in
a freshwater zooplankton community. Limnol Oceanogr., 26, 822-830.
PhlipsJEJ., Aldridge^FJ. and Hanlon.C. (1994) Review of a seventeen-year chlorophyll record as it
pertains to the trophic status of Lake Okeechobee, Florida, USA. Water Res. Bull, 30, 229-238.
Phlips.EJ., Cichra,M.F., Havens.K.E., Hanlon,C, Badylak.S., Reuter,B., Randall,M. and Hansen,P.
(1997) Relationships between phytoplankton dynamics and the availability of light and nutrients
in a shallow subtropical lake. /. Plankton Res., 19, 319-342.
1710
Carbon dynamics in grazing food chain
Polis,G.A. and StrongJJ.R. (1996) Food web complexity and community dynamics. Am. Nat., 147,
813-846.
Porter,K.G. (1973) Selective grazing and differential digestion of algae by zooplankton. Nature, 244,
179-180.
Porter,K.G. (1997) Integrating the microbial look and the classic food chain into a realistic planktonic
food web. In Polis.G.A. and Winemiller.K.O. (eds), Food Webs — Integration of Patterns and
Dynamics. Chapman and Hall, New York.
Pourriot,R. (1977) Food and feeding habits of Rotifera. Arch. HydrobioL, 8,243-260.
Riemann3- (1985) Potential importance of fish predation and zooplankton grazing on natural populations of freshwater bacteria. Appl. Environ. Microbiol., 50,187-193.
Sanders,R. W., Porter.K.G., Bennett.S J. and DeBiase A E . (1989) Seasonal patterns of bacterivory by
flagellates, ciliates, rotifers, and cladocerans in a freshwater planktonic community. Limnol.
Oceanogr., 34, 673-687.
Sieburth^.M., Smetacek,V. and LenzJ. (1978) Pelagic ecosystem structure: heterotrophic compartments of the plankton and their relationship to plankton size fractions. Limnol. Oceanogr., 23,
1256-1263.
Sommer.U. (1989) Plankton Ecology — Succession in Plankton Communities. Springer-Verlag,
Berlin.
Sondergaardjvl. et aL (1988) Pelagic food web processes in an oligotrophic lake. Hydrobiologia, 164,
271-286.
StocknerJ.G. and Shortreed.K.S. (1989) Algal picoplankton production and contribution to food
webs in oligotrophic British Columbia lakes. Hydrobiologia, 173, 151-166.
Szelag-Wasielewska,E. (1997) Picoplankton and other size groups of phytoplankton in various
shallow lakes. Hydrobiologia, 342, 79-85.
Venkataraman,K. (1981) Field and laboratory studies on Daphnia carinata King (Cladocera:
Daphnidae) from a seasonal tropical pond. Hydrobiologia, 78, 221-225.
Vollenweider.R.A. (1974) A Manual on Methods for Measuring Primary Production in Aquatic
Environments. IBP Handbook No. 12. Blackwell Scientific, Oxford.
Weisse,I. (1988) Dynamics of autotrophic picoplankton in Lake Constance. / Plankton Res., 10,
1179-1188.
Weisse.I. and Stockner J.G. (1992) Eutrophication: the role of microbial food webs. Mem. 1st. ltal.
idrobioL, 52, 133-150.
Received on March 19, 1997; accepted on July 14, 1997
1711