Impact of fish predation on cladoceran body weight distribution and zooplankton grazing in lakes during winter

Impact of fish predation on cladoceran body weightdistribution and zooplankton grazing in lakes duringwinter

ERIK JEPPESEN,* , † JENS PEDER JENSEN,* MARTIN SØNDERGAARD,* MORTEN

FENGER-GRØN,* METTE E. BRAMM,* KJELD SANDBY,‡ POUL HALD MØLLER §

AND HELLE UTOFT RASMUSSEN–

*Department of Freshwater Ecology, National Environmental Research Institute, Silkeborg, Denmark†Department of Plant Ecology, University of Aarhus, Nordlandsvej, Risskov, Denmark‡County of Funen, Damhaven, Vejle, Denmark§County of Vejle, Ørbækvej, Odense SØ, Denmark–County of Frederiksborg, Kongens Vænge, Hillerød, Denmark

SUMMARY

1. It is well accepted that fish, if abundant, can have a major impact on the zooplankton

community structure during summer, which, particularly in eutrophic lakes, may cascade

to phytoplankton and ultimately influence water clarity. Fish predation affects mean size

of cladocerans and the zooplankton grazing pressure on phytoplankton. Little is, however,

known about the role of fish during winter.

2. We analysed data from 34 lakes studied for 8–9 years divided into three seasons:

summer, autumn/spring and winter, and four lake classes: all lakes, shallow lakes without

submerged plants, shallow lakes with submerged plants and deep lakes. We recorded how

body weight of Daphnia and then cladocerans varied among the three seasons. For all lake

types there was a significant positive correlation in the mean body weight of Daphnia and

all cladocerans between the different seasons, and only in lakes with macrophytes did the

slope differ significantly from one (winter versus summer for Daphnia).

3. These results suggest that the fish predation pressure during autumn/spring and winter

is as high as during summer, and maybe even higher during winter in macrophyte-rich

lakes. It could be argued that the winter zooplankton community structure resembles that

of the summer community because of low specimen turnover during winter mediated by

low fecundity, which, in turn, reflects food shortage, low temperatures and low winter

hatching from resting eggs. However, we found frequent major changes in mean body

weight of Daphnia and cladocerans in three fish-biomanipulated lakes during the winter

season.

4. The seasonal pattern of zooplankton : phytoplankton biomass ratio showed no

correlation between summer and winter for shallow lakes with abundant vegetation or for

deep lakes. For the shallow lakes, the ratio was substantially higher during summer than in

winter and autumn/spring, suggesting a higher zooplankton grazing potential during

summer, while the ratio was often higher in winter in deep lakes. Direct and indirect

effects of macrophytes, and internal P loading and mixing, all varying over the season,

might weaken the fish signal on this ratio.

Correspondence: Erik Jeppesen, Department of Freshwater Ecology, National Environmental Research Institute, PO Box 314,

DK-8600 Silkeborg, Denmark. E-mail: [email protected]

Freshwater Biology (2004) 49, 432–447

432 � 2004 Blackwell Publishing Ltd

5. Overall, our data indicate that release of fish predation may have strong cascading

effects on zooplankton grazing on phytoplankton and water clarity in temperate, coastal

situated eutrophic lakes, not only during summer but also during winter.

Keywords: community structure, fish, grazing, PEG, winter, zooplankton

Introduction

Zooplankton community structure and biomass are

determined by food availability, temperature, preda-

tion and interactions between different zooplankton,

including inter-specific and interference competition

(Gliwicz, 1985; Sterner, 1989; Gliwicz, 2003). Accord-

ingly, zooplankton community structure varies sea-

sonally. The plankton ecology group (PEG) model

(Sommer et al., 1986) describes a major increase in

cladoceran abundance in late spring, which in some

lakes often leads to a high grazing pressure and a

clear water phase. During summer, the grazing

pressure may decline as a result of predation or

propagation of inedible algae and high abundance of

cyanobacteria. In autumn, the grazing pressure on

phytoplankton may rise again, but without reaching

the level of the peak observed in late spring (Sommer

et al., 1986). This pattern does not, however, apply to

all lake types. Jeppesen et al. (1997) have shown that

concurrently with increasing nutrient levels, or

increased predation pressure from fish on zooplank-

ton, the spring and autumn peaks decline and

ultimately disappear at the highest nutrient and fish

densities. Other investigations (Deneke & Nixdorf,

1999) and minimal models (Scheffer et al., 1997) also

lend support to the hypothesis that decreasing

zooplankton grazing occurs at high fish densities.

Conversely, fish kills or low fish recruitment may lead

to high grazing pressure on phytoplankton and clear

water conditions throughout the summer (Kubecka &

Duncan, 1994; Gliwicz, 2003).

Little is known of the relative importance of

resource and predator control of zooplankton during

winter. Usually resource control is considered to be

the most important factor. The PEG model states that

(i) ‘herbivore biomass decreases as a result of reduced

fecundity because of lower food concentration as well

as decreasing temperature’, and (ii) ‘towards the end

of winter, nutrient availability and stronger light

permit unlimited growth of phytoplankton’. The

suggested lack of grazer control in spring is often

attributed to a resource and temperature mediated

delay in zooplankton growth (Sommer et al., 1986),

but it could also be a result of high predation on

zooplankton by fish. In fact, some experimental

studies (Tempte et al., 1988; Vanni et al., 1990; Rud-

stam, Lathrop & Carpenter, 1993) and minimal model

exercises (Scheffer et al., 1997) suggest that the spring

clear water phase might occur earlier in the season if

planktivorous fish density is low. Similarly, Jeppesen

et al. (1997) showed that potential zooplankton gra-

zing pressure on phytoplankton in eutrophic lakes

with high densities of planktivorous fish was low, not

only during summer but throughout the year. This

also indicates fish control of zooplankton during

winter.

Whether the regulating influence of fish on trophic

structure is more important in summer than winter is

unknown. Attention may be drawn both to factors

reducing and increasing the predation on zooplank-

ton during winter. Among the factors reducing the

predation risk is aggregation of planktivorous fish,

such as bream (Abramis brama L.), near the bottom, in

the littoral zone or in adjacent streams (Jepsen & Berg,

2002). Secondly, their activity on a diel basis may be

reduced (Jacobsen et al., 2002). Thirdly, a gradual

decline in abundance of underyearling fish usually

occurs during autumn because of fish predation and,

accordingly, the predation pressure on zooplankton

most likely decreases. Finally, the food intake by fish

declines during winter (Kitchell et al., 1977), although

the growth of prey zooplankton is also reduced

(Bottrell et al., 1976). By contrast, several factors may

enhance the predation pressure on zooplankton dur-

ing winter. First, transparency is higher during winter

when phytoplankton biomass is low, and this may

increase the predation risk from visually hunting fish.

Secondly, as a result of an autumn overturn in

summer-stratified lakes, zooplankton cannot use the

hypolimnion as a refuge against potential predators.

Thirdly, a major loss of refuges for zooplankton may

also occur in lakes with high summer coverage of

submerged macrophytes (clear shallow lakes and

Fish predation in lakes in winter 433

� 2004 Blackwell Publishing Ltd, Freshwater Biology, 49, 432–447

some small-sized deep lakes) when plants senesce in

autumn.

Unfortunately, most studies of fish-zooplankton

interactions and the cascading effects on phytoplank-

ton are restricted to the summer season. Outdoor field

experiments with zooplanktivorous fish Menidia ber-

yllina (Cope) conducted in Texas, U.S.A., showed

depression of large-bodied zooplankton during all

seasons, including winter when the temperature

occasionally dropped to 3–4 �C (Drenner, Threlkeld

& McCracken, 1986). These results indicate that fish

potentially may depress zooplankton grazing during

winter. Several field studies have revealed low

zooplankton grazing in winter. Haney (1973) recorded

grazing rates of zooplankton by fish in Heart Lake,

U.S.A., of <10% per day in winter and early spring,

while the rates were comparatively high in autumn

(20–50% day)1). Gulati (1978) found grazing rates of

4.5% day)1 in autumn and winter and 22% in early

spring (March to April). Yet, it is noteworthy that

zooplankton grazing, expressed as the percentage of

phytoplankton biomass, was reduced from summer to

winter. This does not, however, necessarily imply

lower grazer control of phytoplankton, as the growth

rate of phytoplankton also decreases in winter. Thus,

Garnier & Mourelatos (1991) showed that zooplank-

ton consumption only declined from 85 to 43% of the

potential phytoplankton production, suggesting a

high grazing pressure also during winter, although

the filtering rate of zooplankton underwent a fivefold

reduction from summer to winter.

To further elucidate the role of fish in determining

cladoceran specimen body weight and grazing capa-

city of zooplankton, we analysed data from 34 Danish

lakes with contrasting nutrient levels and densities of

planktivorous fish. We also analysed data from three

lakes with artificially low fish densities following

biomanipulation or fish kill.

Methods

We sampled 34 lakes for 8–9 years (1989–97/98)

(Table 1) fortnightly in summer (1 May to 1 October)

and nine times in winter when not ice covered. Depth-

integrated zooplankton samples were taken with a

Patalas sampler at three stations placed randomly in

areas representing 80% of maximum depth, then

pooled. Depending on the total phosphorus (TP) level,

between 4.5 and 9 L of the pooled sample were

filtered through an 80-lm net and fixed in Lugol’s

iodine (1 mL, 100 mL tap water) and 0.5–1 L was

settled overnight in Lugol’s solution. Rotifers and

small nauplii were counted from the settled samples,

while all other zooplankton were counted from net

samples. At least 100 individuals of the dominant

zooplankton species were counted. Length–weight

relationships, according to Dumont, Van De Velde &

Dumont (1975) and Bottrell et al. (1976), were used to

estimate biomass. If possible, up to 50 individuals

were measured. A pooled water sample from the

photic zone was analysed for TP (Søndergaard,

Kristensen & Jeppesen, 1992) and chlorophyll a

(Jespersen & Christoffersen, 1987).

The composition and relative abundance of pelagic

fish in the lakes was determined by standardised

fishing (Mortensen et al., 1990) with multiple mesh-

sized gill nets (6.25, 8, 9, 12.5, 16.5, 22, 25, 30, 33, 38,

43, 50, 60 and 75 mm). The length and depth of each

section of mesh were 3 and 1.5 m, respectively.

Fishing was conducted in each lake between 15

August and 15 September, as previous trial fishing

indicated that the distribution of the fish populations

was most even then (Mortensen et al., 1990). More-

over, young-of-the-year fish were also large enough

to be included in the catch by this time. The nets

were set in the late afternoon and retrieved the

following morning. Catch per unit effort (CPUE) of

Table 1 Physical and chemical characteristics (mean ± SD) of the lakes included in the analysis. Total phosphorus (TP),

chlorophyll a (Chl a) summer means (1 May to 1 October)

Lake type n

Surface area

(km2)

Mean depth

(m)

Maximum depth

(m)

TP

(mg L)l)

Chl a

(lg L)l))

Shallow with submerged macrophytes 10 0.9 ± 1.0 1.8 ± 1.0 4.1 ± 2.1 0.06 ± 0.04 18 ± 21

Shallow without submerged macrophytes 15 3.1 ± 10.2 1.8 ± 0.9 3.4 ± 2.0 0.35 ± 0.22 145 ± 84

Deep lakes 9 3.7 ± 4.3 8.2 ± 4.7 19 ± 10 0.09 ± 0.06 36 ± 21

All lakes 34 2.6 ± 7.0 3.5 ± 3.8 7.7 ± 8.7 0.20 ± 0.20 79 ± 83

n ¼ number of lakes.

434 E. Jeppesen et al.

� 2004 Blackwell Publishing Ltd, Freshwater Biology, 49, 432–447

planktivorous fish was calculated as mean catch per

net.

We divided the data into three periods: summer (1

May to 1 October), winter (1 November to 1 January),

and spring/autumn (March to April and October). A

pre-analysis revealed no differences between spring

and autumn data and the data were therefore pooled

to obtain a larger data set for analyses. Data from

January and February were omitted as sampling was

limited because of frequent ice cover and in recent

years because of omission of sampling. We shifted the

spring months a year back in time as we assumed that

fry recruitment in summer has an important influence

on the predation pressure the following spring. For

each lake and year, we then calculated a time-

integrated average for the abundance and total

biomass of zooplankton, cladocerans, Daphnia and

phytoplankton for the three time periods studied.

Then, we calculated the zooplankton : phytoplankton

ratio and the mean weight of cladocerans and Daph-

nia. We have chosen this method instead of average

ratios for the different sampling dates to avoid the

dominance of a few extremely high averages during

the clear water period. We analysed the total data set

and then divided it into three categories: (i) lakes with

mean depths <4 m with submerged macrophytes,

(ii) lakes with a mean depth <4 m, but with no

submerged macrophytes, and (iii) lakes with a mean

depth >4 m with or without submerged macrophytes.

Furthermore, we analysed the seasonal dynamics

of three lakes in which significant changes in the

zooplankton community occurred during the 9-year

study period as a consequence of fish stock mani-

pulation or fish kill. The three lakes are all shallow

and eutrophic (Table 2). In Lake Engelsholm, 19 t

cyprinids were removed by netting during April to

September 1992. Thereby the calculated biomass of

cyprinids was reduced from 675 to 150–300 kg ha)1

(Møller, 1998). In Lake Arreskov, a major fish kill

occurred during autumn and winter (1991). An

additional 4 t cyprinids were removed in 1995, and

during 1993 and 1995 the lake was stocked

with underyearling pike amounting to totally

141 individuals (ind) ha)1 . Cyprinid biomass was

calculated as 172 kg ha)1 in 1987 and 71 kg ha)1 in

1995 (Sandby, 1998). In Lake Bastrup, 7 t cyprinids

were removed during 1995–97. The cyprinid biomass

was calculated to 300–400 kg ha)1 before biomanip-

ulation and to 150 kg ha)1 afterwards (Frederiksborg

County, 2000).

Statistical analyses

The variables compared are assumed to have approxi-

mately similar errors so that a classic linear regression

cannot be used, as it requires negligible variation of

the independent variable. Instead, we used the

method described by Fuller (1987, p. 30).

Results

Multi-lake analysis

Boxplot of the surface water temperatures for the

three selected periods and the three lake types is

shown in Fig. 1. Mean temperature was 17 ± 3.1 (SD),

17.2 ± 3.1 and 16.5 ± 3.2 �C during summer in the

surface waters of shallow lakes with, without macro-

phytes and in deep lakes, respectively. In spring/

autumn the same figures are reduced to 7.5 ± 2.9,

7.8 ± 2.9 and 7.4 ± 3.3 �C, and during winter to

4.3 ± 1.7, 4.4 ± 2.0, and 5.4 ± 2.0 �C. For comparison,

the limited data from January to February showed an

average of 2.9 ± 1.5 �C. Minimum values recorded

during the selected winter period (November to

December) were 0.5–1.0 �C. During the study period

1989–98 we did not record any significant changes in

temperature with time for any of the periods (P > 0.4).

So, we can exclude confounding effects of climate

change on the results presented.

Daphnia and cladoceran (other than Daphnia) mean

size decreases with increasing fish predation because

Table 2 Morphometric data and annual

mean hydraulic retention time and total

phosphorus concentrations in the epilim-

nion of the three lakes in which the fish

stock was reduced

Surface

area (km2)

Mean

depth (m)

Maximum

depth (m)

Retention

time (days)

Total phosphorus

(mg P L)1)

Lake Engelsholm 0.44 2.6 6.1 65–88 0.07–0.12

Lake Arreskov 317 1.9 3.7 510–770 0.10–0.23

Lake Bastrup 0.32 3.5 7.0 600–2200 0.03–0.10

Fish predation in lakes in winter 435

� 2004 Blackwell Publishing Ltd, Freshwater Biology, 49, 432–447

of size-selective predation (Brooks & Dodson, 1965).

We therefore elucidated how body weight of Daphnia

and cladocerans varied among the three selected

seasons. For all lake types there was a significant

positive correlation in the mean body weight of

Daphnia (Fig. 2) and all cladocerans (Fig. 3) between

the different seasons, and only in shallow lakes

without macrophytes did the slope differ significantly

from one in one case (winter versus summer for

Daphnia) (Table 3).

The biomass ratio of zooplankton to phytoplankton

has earlier proven to be closely related with the

number of planktivorous fish caught in multi-mesh

sized gill-nets in Danish and New Zealand lakes

(Jeppesen et al., 2000a,b). We therefore elucidated how

the ratio in Danish lakes varied between the three

selected seasons (Fig. 4). The full data set had a

significant positive relationship between winter and

summer mean zooplankton : phytoplankton ratio,

with the slope being significantly higher than one

(2.75) (Table 3). This relationship was largely deter-

mined by shallow lakes without submerged macro-

phytes (slope 2.79), as no significant relationship

between summer and winter was found for shallow

lakes with macrophytes or deep lakes (Table 3). The

relationship between spring/autumn and summer

zooplankton : phytoplankton ratio for the different

lake types largely followed the pattern for summer

versus winter. By contrast, the relationship between

winter and spring/autumn was significant for all lake

types, the slope being significantly higher than one in

all cases (1.56–1.66) (Fig. 4; Table 3).

Mean body weight of cladocerans in summer and

winter was generally positively related to the

zooplankton : phytoplankton ratio for the same sea-

sons (Fig. 5; Table 4). For winter, the slope was signi-

ficantly lower than one (0.22–0.40) for all lake types. By

contrast, the slope for the summer was not significantly

different from zero for shallow lakes with submerged

macrophytes and for deep lakes, whereas the slope for

shallow lakes without submerged macrophytes devi-

ated significantly from zero, but not from one.

For summer, the mean body weight of cladocerans

and the zooplankton : phytoplankton ratio was signi-

ficantly (P < 0.05–0.0001) negatively related to CPUE

of planktivores caught in multiple mesh-sized gill nets

during test-fishing conducted between 15 August and

15 September both on the full data set and when

divided into seasons (except for the zooplankton :

phytoplankton ratio during winter) (Fig. 6).

Examples of effects of significant fish stock changes

In three eutrophic lakes with major changes in the fish

stock, the reduction in the planktivorous fish biomass

resulted in an overall major increase in mean body

weight of Daphnia and other cladocerans and the

zooplankton : phytoplankton biomass ratio, followed

by large intra-annual variations (Fig. 7). In Lake

Arreskov, a major shift in dominance occurred in

1991 from small-sized cladocerans to Daphnia, which

resulted in a pronounced increase in mean body

weight of cladocerans from 9.7 lg ind)1 in late

November 1991 to 19 lg ind)1 in mid-January 1992

(Fig. 6). The increase occurred primarily during win-

ter 1991/1992 after fish kills. A second increase took

place in winter 1994–95 after a significant decrease in

summer and autumn 1994. Median body weight

increased from 4.0 lg ind)1 in mid-September to 13

and 36 lg ind)1 in mid-December and mid-March,

respectively. In Lake Engelsholm, the mean body

weight of cladocerans rose sharply during autumn

1992 and again in autumn 1993 after a summer

decrease. In Lake Bastrup, a major increase in cladoc-

eran mean body weight occurred from 2.2 lg ind)1 in

November 1997 to 10 lg ind)1 in March 1998 and

from 2.9–6.8 lg ind)1 during November to March

1998/1999 although the temperature did not exceed

3.6 �C. In all three lakes, the level of chlorophyll a and

the chlorophyll a to TP ratio were generally lower (also

during winter) and zooplankton : phytoplankton was

Tem

pera

ture

(°C

)

Shallow + macrophytesShallow – macrophytes

Deep

Summer Spring/autumn Winter

0

5

10

15

20

25

Fig. 1 Boxplot (5, 25, 75 and 95% percentiles) of water tem-

perature measured during the samplings in summer (1 May to 1

October), spring and autumn (March, April, October) and winter

(November to December) in the three selected lake categories.

Line goes through the medians.

436 E. Jeppesen et al.

� 2004 Blackwell Publishing Ltd, Freshwater Biology, 49, 432–447

Spring/autumnWinter Winter

Spring/autumnWinter Winter

Spring/autumnWinter Winter

Spring/autumnWinter Winter

–1

0

1

2

3

4

5

SummerSpring/autumnSummer–1 0 1 2 3 4 5–1 0 1 2 3 4 5–1 0 1 2 3 4 5

–1

0

1

2

3

4

5

–1

0

1

2

3

4

5

–1

0

1

2

3

4

5All lakesDaphnia body weight (µg ind–1)

Shallow lakes with macrophytes

Shallow lakes without macrophytes

Deep lakes

Fig. 2 Plots of mean body weight of Daphnia (lg DW ind)1) between seasons (log base e) for various lake types in 34 Danish lakes,

covering 8–10 years. Also shown are a regression line, if significant (Table 3) (thick line), and the 1 : 1 line (thin line). Each dot

represents one lake year.

Fish predation in lakes in winter 437

� 2004 Blackwell Publishing Ltd, Freshwater Biology, 49, 432–447

Spring/autumnWinter Winter

Spring/autumnWinter Winter

Spring/autumnWinter Winter

Spring/autumnWinter Winter

SummerSpring/autumnSummer–1 0 1 2 3 4–1 0 1 2 3 4–1 0 1 2 3 4

–1

0

1

2

3

4

–1

0

1

2

3

4

–1

0

1

2

3

4

–1

0

1

2

3

4

Shallow lakes with macrophytes

Shallow lakes without macrophytes

All lakesCladoceran body weight (µg ind–1)

Deep lakes

Fig. 3 Mean body weight of cladocerans (lg DW ind)1) during winter and spring/autumn versus summer and spring/autumn (all

loge-transformed) for various lake types in 34 Danish lakes, covering 8–10 years. Also shown are a regression line, if significant

(Table 3) (thick line), and the 1 : 1 line (thin line). Each dot represents one lake year.

438 E. Jeppesen et al.

� 2004 Blackwell Publishing Ltd, Freshwater Biology, 49, 432–447

Table 3 Results of regression analyses (assuming similar variation on the dependent and independent variables) when testing the

differences among seasons of cladoceran and Daphnia body weight and the zooplankton : phytoplankton ratio

Intercept T0

Test for intercept

equal to 0 Slope T1

Test for slope

equal to 1

Daphnia (mean weight)

Winter versus summer

All )0.13 ()0.4–0.2) )0.8 ns 1.09 (0.9–1.3) 1.0 ns

Shallow with macrophytes 0.28 ()0.1–0.7) 1.4 <0.05 0.95 (0.8–1.1) )0.5 ns

Shallow without macrophytes )0.84 ()1.7 to )0.1) )2.0 ns 1.56 (1.1–2.1) 2.2 <0.03

Deep 0.17 ()0.3–0.7) 0.7 ns 0.88 (0.6–1.1) )1.0 ns

Winter versus spring/autumn

All 0.28 (0.1–0.5) 2.8 <0.007 0.93 (0.8–1.1) )1.3 ns

Shallow with macrophytes 0.61 (0.3–1.0) 3.5 <0.002 0.85 (0.7–1.1) )1.7 ns

Shallow without macrophytes 0.006 ()0.4–0.4) 0.02 ns 1.07 (0.8–1.3) 0.6 ns

Deep 0.36 (0.1–0.7) 2.5 <0.02 0.88 (0.7–1.0) )1.7 ns

Spring/autumn versus summer

All )0.19 ()0.4–0.1) )1.5 ns 1.05 (0.9–1.2) 0.8 ns

Shallow with macrophytes )0.14 ()0.7–0.4) )0.5 ns 1.08 (0.8–1.4) 0.5 ns

Shallow without macrophytes )0.21 ()0.6–0.2) )1.1 ns 1.11 (0.9–1.3) 0.9 ns

Deep )0.36 ()0.8–0.1) )1.4 ns 1.06 (0.8–1.3) 0.6 ns

All cladocerans (mean weight)

Winter versus summer

All 0.15 (0.01–0.3) 2.1 <0.04 0.97 (0.8–1.1) )0.4 ns

Shallow with macrophytes )0.19 ()0.8–0.4) )0.7 ns 1.21 (0.6–1.8) 0.7 ns

Shallow without macrophytes 0.27 (0.1–0.4) 3.8 <0.0002 1.05 (0.8–1.4) 0.4 ns

Deep )0.18 ()0.8–0.5) )0.5 ns 1.13 (0.7–1.5) 0.6 ns

Winter versus spring/autumn

All 0.13 (0.03–0.2) 2.7 <0.08 0.96 (0.9–1.0) )1.0 ns

Shallow with macrophytes 0.04 ()0.2–0.3) 0.4 ns 0.86 (0.7–1.0) )1.5 ns

Shallow without macrophytes 0.14 (0.05–0.23) 3.1 <0.03 0.95 (0.8–1.1) )0.9 ns

Deep 0.28 (0.03–0.5) 2.2 <0.03 0.97 (0.8–1.1) )0.3 ns

Spring/autumn versus summer

All )0.02 ()0.2–0.1) )0.2 ns 1.06 (0.9–1.2) 0.8 ns

Shallow with macrophytes )0.22 ()0.7–0.2) )1.0 ns 1.39 (0.9–1.8) 1.7 ns

Shallow without macrophytes 0.11 ()0.1–0.3) 1.6 ns 1.13 (0.9–1.4) 0.9 ns

Deep )0.87 ()1.7 to 0.1) )2.2 <0.03 1.43 (0.9–1.9) 1.7 ns

Zooplankton : phytoplankton

Winter versus summer

All )5.57 ()3.1 to )8.0) )4.5 <0.0001 2.75 (2.0–3.5) 4.4 <0.0001

Shallow with macrophytes )70 ()577–436) )0.3 ns 19.4 ()114–152)ns 0.3

Shallow without macrophytes )4.8 ()2.9 to )6.7) )5.0 <0.0001 2.79 (2.1–3.5) 5.0 <0.0001

Deep )6.2 ()17–4) )1.1 ns 3.06 ()0.3–3.1)ns 1.2

Winter versus spring/autumn

All )1.50 ()0.8 to )2.2) )4.0 <0.001 1.66 (1.4–1.9) 4.9 <0.0001

Shallow with macrophytes 1.93 ()0.3 to )3.6) )2.3 <0.04 1.56 (1.1–2.1) 2.2 <0.04

Shallow without macrophytes )1.07 ()0.2 to )1.9) )2.5 <0.02 1.56 (1.2–1.9) 3.1 <0.003

Deep )4.78 ()10.2–0.7) )1.7 ns 2.89 (1.1–4.7) 2.1 <0.05

Spring/autumn versus summer

All )1.16 ()0.5 to )1.8) )3.7 <0.0003 1.25 (1.1–1.4) 2.5 <0.02

Shallow with macrophytes )3.89 ()8.2–0.4) )1.8 ns 1.87 (0.7–3.0) 1.5 ns

Shallow without macrophytes )2.46 (1.3 to )3.6) )4.3 <0.0001 1.81 (1.4–2.2) 3.8 <0.0003

Deep 0.51 ()0.4–1.5) 1.1 ns 0.76 (0.5–1.0) )1.7 ns

Intercept and slope are given with 95% CL in parentheses. T0: test value for intercept equal to 0. T1: Test value for slope equal to 1

(T-distribution, n ) 2 d.f.). As the data set consists of eight to 10 samples per lake, significance levels >0.005 must be interpreted with

care. ns: not significant.

Fish predation in lakes in winter 439

� 2004 Blackwell Publishing Ltd, Freshwater Biology, 49, 432–447

Spring/autumnWinter Winter

Spring/autumnWinter Winter

Spring/autumnWinter Winter

Spring/autumnWinter Winter

SummerSpring/autumnSummer0 1 2 3 4 5 6 70 1 2 3 4 5 6 70 1 2 3 4 5 6 7

0

1

2

3

4

5

6

7

0

1

2

3

4

5

6

7

0

1

2

3

4

5

6

7

0

1

2

3

4

5

6

7

Shallow lakes with macrophytes

Shallow lakes without macrophytes

All lakesZooplankton: phytoplankton biomass ratio

Deep lakes

Fig. 4 The zooplankton : phytoplankton biomass ratio (%) during winter and spring/autumn versus summer and spring/autumn

ratio (all loge-transformed) for various lake types in 34 Danish lakes, covering 8–10 years. Also shown are a regression line, if

significant (Table 3) (thick line), and the 1 : 1 line (thin line). Each dot represents one lake year.

440 E. Jeppesen et al.

� 2004 Blackwell Publishing Ltd, Freshwater Biology, 49, 432–447

higher after the decline in fish abundance than when

small-sized cladoceran species dominated at high fish

predation (Figs 7 & 8).

Table 4 Results of regression analyses of mean cladoceran body weight during summer and winter versus the zooplankton : phy-

toplankton ratio during the same seasons (see also legend of Table 3)

Intercept T0

Test for intercept

equal to 0 Slope T1, T0

Test for slope

equal to 1

Summer cladoceran mean weight versus summer zooplankton : phytoplankton ratio

All )1.9 ()2.7 to )1.1) )4.9 <0.0001 0.87 (0.6–1.1) )1.1 ns

Shallow with macrophytes )3.7 ()9.0–1.8) )1.3 ns 1.2 ()0.3–2.6)ns 0.2

Shallow without macrophytes )1.9 ()2.5 to )1.2) )5.3 <0.0001 0.84 (0.6–1.1) )1.2 ns

Deep 0.3 ()1.8–1.0) 0.3 ns 0.35 ()0.3–1.0)ns )2.0

Winter cladoceran mean weight versus winter zooplankton : phytoplankton ratio

All )0.20 ()0.5–0.5) )1.3 ns 0.36 (0.27–0.45) )13.8 <0.0001

Shallow with macrophytes )0.44 ()1.5–0.7) )0.8 ns 0.40 (0.1–0.7) )3.6 <0.0005

Shallow without macrophytes 0.06 ()0.3–0.3) 0.4 ns 0.22 (0.1–0.3) )15.8 <0.0001

Deep 0.20 ()1.1–0.6) 0.3 ns 0.33 (0–0.0.7) )4.0 <0.0001

As the data set consists of eight to 10 samples per lake, P levels >0.005 must be interpreted with care. ns: not significant.

Log e

cla

doce

rans

(µg

ind

–1)

–0.5

0

0.5

1.0

1.5

2.0

2.5

Fish CPUE (no. net –1 night

–1)

WinterSpring/autumn

Summer

0–100 100–200 >200

Zoo

plan

kton

: phy

topl

ankt

on0

0.5

1.0

1.5

Fig. 6 Boxplot (median, 25, 75, 10 and 90% quartiles) of mean

body weight of cladocerans (loge-transformed) and the biomass

ratio of zooplankton : phytoplankton for samples collected

during summer and winter versus the catch per net at night in

multiple mesh-sized gill nets in 34 lakes monitored one to

three times during the study period (n ¼ 56–74). Fish sampling

occurred in all years and lakes between 15 August and 15

September.

0 1 2 3 4 5 6 70 1 2 3 4 5 6 7

–2

0

2

4

0

2

4–2

–2

0

2

4–2

0

2

4S

umm

er m

ean

body

wei

ght o

f cla

doce

rans

(µg

ind–

1 )

Win

ter

mea

n bo

dy w

eigh

t of c

lado

cera

ns (

µg in

d–1 )

Sha

llow

lake

s w

ith m

acro

phyt

esS

hallo

w la

kes

with

out m

acro

phyt

esA

ll la

kes

Dee

p la

kes

Zooplankton: phytoplankton biomass ratio

Summer Winter

Fig. 5 Mean body weight of cladocerans (lg DW ind)1) versus

zooplankton : phytoplankton biomass ratio (all loge-trans-

formed) for the various lake types during summer (left) and

winter in 34 Danish lakes, covering 8–10 years. Regression line

is shown for significant relationships (see Table 4).

Fish predation in lakes in winter 441

� 2004 Blackwell Publishing Ltd, Freshwater Biology, 49, 432–447

Discussion

Our results suggest an overall positive relationship

between summer and winter levels of the mean body

weight of Daphnia and other cladocerans as well as the

zooplankton : phytoplankton biomass ratio, though

lake-type specific differences occurred. A decline in

size of cladocerans and Daphnia during summer is

often ascribed to increased predation by fish (Hrbacek

et al., 1961; Brooks & Dodson, 1965). Accordingly, we

observed an inverse relationship between summer

and winter mean body weight of pelagic cladocerans

versus CPUE of planktivorous fish caught during

surveys in August to September. Also, however, we

found a significant relationship between summer,

spring/autumn and winter levels of body weight of

both Daphnia and other cladocerans, and generally

the slope was not significantly different from one

(Table 3). Thus, apparently, winter predation pressure

on zooplankton overall mirrored summer predation,

though high residuals suggest large lake or year-to-

year variations. It could be argued that the overall

small changes in cladoceran and Daphnia mean body

weight from summer to winter reflect low renewal of

the cladoceran community from summer to winter,

determined by low winter fecundity mediated by food

shortage and low temperatures, or by low winter

hatching of resting eggs. However, the data from

lakes in which planktivorous fish abundance declined

because of fish kill or cyprinid removal showed that

dominance by small-bodied specimens in summer

does not automatically lead to dominance of small-

bodied specimens during winter. In these lakes,

significant shifts from small to large specimens were

occasionally seen during autumn and winter (Fig. 7)

following changes in fish abundance. For the 34 lakes,

the intercepts were in a few cases slightly, albeit

significantly, greater than zero, particularly for winter

89 90 91 92 93 94 95 96 97 98 99 89 90 91 92 93 94 95 96 97 98 99

Dap

hnia

(µg

ind–

1 )

Wat

er te

mpe

ratu

re (

°C)

Cla

doce

rans

(µg

ind–

1 )

Chl

orop

hyll

a (µ

g l–

1 )

0

10

20

30

40

50

0

10

20

30

0

10

20

30

0

10

20

30

40

0

10

20

30

40

0

10

20

30

40

0

10

20

30

40

50

0

10

20

30

0

5

10

15

20

25

0

50

100

150

200

0

50

100

150

200

0

25

50

75

100

Lake Arreskov

Lake Engelsholm

Lake Bastrup

Lake Arreskov

Lake Engelsholm

Lake Bastrup

Fig. 7 Seasonal dynamics in mean body weight of Daphnia, herbivorous cladocerans, water temperature and chlorophyll a in three

lakes in which major changes occurred in the cyprinid fish biomass because of fish kill and ⁄or biomanipulation of fish (indicated by

arrows).

442 E. Jeppesen et al.

� 2004 Blackwell Publishing Ltd, Freshwater Biology, 49, 432–447

versus spring/autumn, which may indicate a slightly

lower fish predation during winter. The zooplank-

ton : phytoplankton ratio in the different seasons

adds further support to the hypothesis of a relation-

ship between the summer and winter predation

pressure by fish, as the ratio during both summer

and winter was correlated with CPUE of planktivor-

ous fish caught in gill nets in late summer (Fig. 5).

The mean body weight of cladocerans and the

zooplankton : phytoplankton ratios were generally

positively correlated both for the summer and winter

seasons; yet there were differences between lake

categories. For shallow lakes with macrophytes we

found no relationship between winter and summer

ratios or between mean cladoceran body weight and

the zooplankton : phytoplankton ratio during sum-

mer. In these lakes, the zooplankton : phytoplankton

ratio tended to be higher during summer than in

autumn/spring or winter in most lakes, possibly

indicating a potentially higher zooplankton grazing

pressure on phytoplankton in summer. The higher

ratio in summer may be attributed to various factors.

First, a higher plant biomass in summer may yield

greater protection against fish predation than later in

the season and therefore higher survival and higher

grazing pressure on phytoplankton (Timms & Moss,

1984; Lauridsen et al., 1996; Jeppesen et al., 2002) as in

the plant-free seasons. Yet, if protection were of

general importance, we would also have expected

summer mean cladoceran and Daphnia body weights

to be higher than during winter, which was not the

case (Figs 2 & 3). Secondly, in plant-rich lakes other

filter feeders occur, of which some are associated with

plants (e.g. Sida) and others are benthic (e.g. mussels).

Such filter feeders may exert a high grazing pressure

on phytoplankton (Ogilvie & Mitchell, 1995; Stansfield

et al., 1997) and thereby indirectly enhance the

zooplankton : phytoplankton ratio. Thirdly, increased

nutrient limitation of phytoplankton in these clear-

water lakes, because of nutrient uptake by submerged

macrophytes (Wetzel, 1983) and benthic algae

(Jansson, 1989; Hansson, 1992), may potentially

affect the phytoplankton and thus indirectly the

ratio during summer. Finally, by feeding on plant

surfaces or benthic-derived organic matter (Jones &

Waldron, 2003), zooplankton may maintain a high

density and accordingly a high ratio (Jeppesen et al.,

2002). That submerged plants play a role, whether

direct or indirect, is supported by the fact that the

zooplankton : phytoplankton ratio during summer

and winter was significantly correlated for shallow

lakes without submerged macrophytes, and that good

correspondence in the ratio was found outside the

macrophyte season (spring/autumn versus winter)

also in lakes with submerged macrophytes.

In the shallow lakes without macrophytes, the

zooplankton : phytoplankton ratio during winter

and spring/autumn was generally lower than during

summer at low summer ratios, and higher at high

ratios (intercept negative, slope >1). In Danish lakes

and abroad, low summer ratios are usually associated

with strong fish predation on zooplankton (Jeppesen

et al., 2000a, 2003; Fig. 5). It is therefore likely that the

proportionally lower winter and spring/autumn

ratios in lakes with low ratios during summer occur

because the more transparent water prevailing in

winter and autumn/spring increases the predation

risk, as compared with the more turbid summer

season. By contrast, in lakes with high summer

zooplankton : phytoplankton ratios (and thus relat-

ively few fish), zooplankton are less affected by fish,

and zooplankton density remains high during winter

and spring/autumn despite the reduced phytoplank-

ton biomass (lower nutrient level, less light). Thus, in

these lakes, the ratio is higher during winter than

during summer, which is supported by the results

obtained from the three lakes in which the fish stock

declined (Fig. 8). Apart from the spring clear water

phase, the ratio was particularly high in winter after

fish manipulation. Accordingly, the Chl a : TP ratio

tended to be lower than before manipulation, which

may indicate higher grazer control of phytoplankton.

However, contradicting this view, the body weight of

cladocerans and Daphnia did not differ among sum-

mer and winter in the biomanipulated lakes except

for March in Lake Arreskov (Fig. 8), and data on the

34 lakes showed no differences either. Therefore, the

slope and the regression line between cladoceran

mean weight and zooplankton : phytoplankton ratio

were substantially lower during winter than summer

(Figs 2 & 3). The high slope of the ratio (winter versus

summer) may alternatively be explained by differ-

ences in internal loading, with loading being higher in

lakes with high grazer control of phytoplankton

in winter than in lakes with low grazer control

(Søndergaard, Jensen & Jeppesen, 1999). Accordingly,

nutrient constraint on phytoplankton during winter is

reduced in lakes with low, rather than high,

Fish predation in lakes in winter 443

� 2004 Blackwell Publishing Ltd, Freshwater Biology, 49, 432–447

grazer control of phytoplankton, which leads to a

lower ratio.

For deep lakes no relationship was found for

zooplankton : phytoplankton ratio between winter

and summer and the ratio was generally higher in

winter than during summer. A possible explanation is

that zooplankton are more exposed to fish predation

during summer as both fish and zooplankton are

restricted to the epilimnion and metalimnion because

of low oxygen concentrations in the hypolimnion.

Nevertheless, we did not find lower mean body

weight of Daphnia and other cladocerans in summer.

Moreover, winter zooplankton : phytoplankton also

tended to be higher than in spring/autumn where the

water masses are fully mixed. An alternative explan-

ation is that light limitation of the phytoplankton

during the winter months with short days character-

ised by low irradiance in Denmark, and the effect of

light limitation would, of course, be much stronger in

deep lakes than in comparable shallow lakes because

of deep mixing.

The large scatter in the relationships observed in

Figs 2–5 may be attributed to various factors. First,

planktivorous fish abundance varies over the season

Dap

hnia

(µg

ind–

1 )C

laco

cera

ns (

µg in

d–1 )

Zoo

plan

kton

: phy

topl

ankt

onC

hla:

TP

0

10

20

30

40

50

0

10

20

30

40

50

0

1

2

3

4

5

0.0

0.5

1.0

1.5

AfterBefore

M A M J J A S O N D M A M J J A S O N D M A M J J A S O N D

Lake Arreskov Lake Engelsholm Lake Bastrup

Fig. 8 Seasonal dynamics in Daphnia and cladoceran mean weight, zooplankton : phytoplankton ratio and the chlorophyll a : total

phosphorus ratio (Chl a : TP) (monthly mean ± SD) in three Danish lakes before and after reduction in the abundance of planktivorous

fish because of biomanipulation and ⁄or fish kills.

444 E. Jeppesen et al.

� 2004 Blackwell Publishing Ltd, Freshwater Biology, 49, 432–447

and from year to year depending on a number of

factors such as recruitment success of planktivores and

piscivorous fish (Mills & Forney, 1983; Cryer, Peirson

& Townsend, 1986; Persson & Crowder, 1997), food

availability (Persson & Greenberg, 1990), macrophyte

coverage (Jeppesen et al., 2002) winter starvation and

fish kills under ice (Jackson, 2003; Meding & Jackson,

2003) and summer temperature (Mehner, 2000), which

may induce seasonal and inter-annual variations in

predation threat to zooplankton. Secondly, survival of

cladocerans during winter depends on food conditions

and temperature. In a long-term study in Esthwaite

Water, U.K., George & Hewitt (1999) found higher

abundances of Daphnia in cold winters, while Eudiap-

tomus dominated during warm winters, which they

attributed to differences in phytoplankton composi-

tion and abundance and to temperature-dependent

differences in maintenance costs for the two crusta-

ceans. George & Hewitt (1999) further showed that

high winter survival had a large influence on the

spring cohort of Daphnia. Thirdly, the bottom-up effect

on phytoplankton may vary seasonally and be

mediated by variation in, for instance, the extent of

external and internal loading of nutrients. Fourthly, in

some lakes the zooplankton : phytoplankton ratio lies

systematically above or below the regression lines, but

otherwise follows the general pattern (differ in inter-

cept but not in slope, data now shown). It is, however,

outside the scope of this paper to deal with this scatter

in more detail.

Our results that indicate potential strong predator

control during winter when planktivorous fish abun-

dance is high are supported by an enclosure experi-

ment in the highly eutrophic Lake Søbygaard (T.

Sørensen, G. Muldrij, E. Jeppesen and M. Sønder-

gaard, unpublished data). Presence of fish (roach,

Rutilus rutilus, and three-spined sticklebacks, Gaster-

osteus aculeatus) in high densities (totally 8 m)2) resul-

ted in low densities of large-bodied zooplankton and

high chlorophyll a (50–120 lg L)l) and low zooplank-

ton : phytoplankton ratio throughout the winter,

while in enclosures without fish different daphnids

including Daphnia magna were abundant, chlorophyll a

low (typically <5–10 lg L)l) and the zooplankton :

phytoplankton ratio was high. Identical experiments

in a less eutrophic lake revealed no significant

differences in chlorophyll a among treatments despite

major differences in zooplankton community struc-

ture. Thus, the zooplankton response to fish and the

cascading effects on phytoplankton during winter

apparently decline with decreasing TP, as earlier

demonstrated in several studies conducted during

summer (reviewed by Pace et al., 1999) and empirical

data from numerous lakes (Jeppesen et al., 2003). The

strong effects in eutrophic lakes depend, however, on

the presence of large-bodied cladocerans. Experiments

in Lake Søbygaard the following year, when cyclopoid

copepods dominated the zooplankton community,

showed comparatively minor, though still significant,

cascading effects of fish on phytoplankton during

winter (M. Bramm, unpublished information).

In summary, we have shown an overall close

relationship between mean specimen body weight of

all cladocerans and Daphnia during summer and

winter, which we attribute to a comparable fish

predation pressure during the two seasons. The

response of the zooplankton : phytoplankton ratio

was less clear, and most likely influenced by direct

and indirect effects of seasonal changes in submerged

macrophyte abundance and resource control of phy-

toplankton. Our conclusion is supported by data from

lakes with drastic changes in fish abundance during

the study period and from ongoing enclosure experi-

ments. We emphasise, however, that our coastal-

situated lakes are often only temporarily ice-covered

during winter or in cold years only for 1–3 months.

Moreover, because of an insufficient amount of data

we excluded the coldest months, January to February,

which had on average 1.5 �C lower temperatures than

the ‘winter’ data presented here. Different results may

be obtained in a more continental climate with longer

term ice cover.

Acknowledgments

We are grateful to the Danish counties for access to

some of the data used in the analysis and thank Kathe

Møgelvang and Anne Mette Poulsen for skilful

technical assistance. The study was supported by the

research programme ‘The role of fish in ecosystems,

1999–2001,’ funded by the Danish Ministry of Agri-

culture, Fisheries and Food and by the project ‘Con-

sequences of weather and climate changes for marine

and freshwater ecosystems’ (CONWOY, http://

www.conwoy.ku.dk) funded by the Danish Scientific

Research Council and the EU Eurolimpacs project. We

thank Thomas Davidson and Leland J. Jackson for

valuable comments.

Fish predation in lakes in winter 445

� 2004 Blackwell Publishing Ltd, Freshwater Biology, 49, 432–447

References

Bottrell H.H., Duncan A., Gliwicz Z.M., Grygierek E.,

Herzig A., Hillbrich-Ilkowska A., Kurasawa H., Lars-

son P. & Weglenska T. (1976) A review of some

problems in zooplankton production studies. Norwe-

gian Journal of Zoology, 24, 419–456.

Brooks J.L. & Dodson S.I. (1965) Predation, body size and

composition of plankton. Science, 150, 28–35.

Cryer M., Peirson G. & Townsend C.R. (1986) Reciprocal

interactions between roach Rutilus rutilus, and zoo-

plankton in a small lake: prey dynamics and fish

growth and recruitment. Limnology and Oceanography,

31, 1022–1038.

Deneke R. & Nixdorf B. (1999) On the occurrence of

clearwater phases in relation to shallowness and

trophic state: a comparative study. Hydrobiologia, 408/

409, 251–262.

Drenner R.W., Threlkeld S.T. & McCracken M.D. (1986)

Experimental analysis of the direct and indirect effects

of an omnivorous filter-feeding clupeid on plankton

community structure. Canadian Journal of Fisheries and

Aquatic Sciences, 43, 1935–1945.

Dumont H.J., Van De Velde I. & Dumont S. (1975) The

dry weight estimate of biomass in a selection of

Cladocera, Copepoda and Rotifera from plankton,

periphyton and benthos of continental waters.

Oecologia, 19, 75–97.

Frederiksborg County (2000) Bastrup Sø, tilstand og

udvikling 1999. pp. 83. Frederiksborg County, Denmark

(In Danish).

Fuller W.A. (1987) Measurement Error Models. John Wiley

& Sons, New York.

Garnier J. & Mourelatos S. (1991) Contribution of grazing

in phytoplankton overall losses in a shallow French

lake. Freshwater Biology, 28, 515–523.

George D.G. & Hewitt P. (1999) The influence of year-

to-year variations in winter weather on the dynamics

of Daphnia and Eudiaptomus in Esthwaite Water,

Cumbria. Functional Ecology, 13, 45–54.

Gliwicz Z.M. (1985) Predation or food limitation: an

ultimate reason for extinction of planktonic cladoceran

species. Archiv fur Hydrobiologie. Beiheft/Ergebnisse der

Limnologi., 21, 419–430.

Gliwicz Z.M. (2003) Between hazards of starvation and

risk of predation: the ecology of offshore animals. In:

Excellence of Ecology, Book 12 (Ed. O. Kinne). Interna-

tional Ecology Institute, Oldendorf/Luhe.

Gulati R. (1978) Vertical changes in filtering, feeding and

assimilation rates of dominant zooplankton in a

stratified lake. Verhandlungen Internationale Vereinigung

der Limnologie, 20, 950–956.

Haney J.F. (1973) An in situ examination of the grazing

activities of natural zooplankton communities. Archiv

fur Hydrobiologie, 72, 87–132.

Hansson L.-A. (1992) Effects of competitive interactions

on the biomass development of planktonic and

periphytic algae in lakes. Limnology and Oceanography,

73, 241–247.

Hrbacek J., Dvorakova V., Korinek V. & Prochazkova L.

(1961) Demonstration of the effect of the fish stock on

the species composition of zooplankton and the

intensity of metabolism of the whole plankton associa-

tion. Verhandlungen Internationale Vereinigung der Lim-

nologie, 14, 192–195.

Jackson L.J. (2003) Macrophyte-dominated and turbid

states of shallow lakes: evidence from Alberta lakes.

Ecosystems, 6, 213–223.

Jacobsen L., Berg S., Broberg M., Jepsen N. & Skov C.

(2002) Activity and food choice of piscivorous perch

(Perca fluviatilis) in a eutrophic shallow lake: a radio-

telemetry study. Freshwater Biology, 47, 2370–2379.

Jansson M. (1989) Role of benthic algae in transport of

nitrogen from sediment to lake water in a shallow

clearwater lake. Archiv fur Hydrobiologie, 89, 101–109.

Jeppesen E., Jensen J.P., Søndergaard M., Lauridsen T.,

Pedersen L.J. & Jensen L. (1997) Top-down control in

freshwater lakes: the role of nutrient state, submerged

macrophytes and water depth. Hydrobiologia, 342/343,

151–164.

Jeppesen E., Jensen J.P., Søndergaard M., Lauridsen T. &

Landkildehus F. (2000a) Trophic structure, species

richness and biodiversity in Danish lakes: changes

along a phosphorus gradient. Freshwater Biology, 45,

201–213.

Jeppesen E., Lauridsen T., Mitchell S.F., Christoffersen K.

& Burns C.W. (2000b) Trophic structure in the pelagial

of 25 shallow New Zealand lakes: changes along a

nutrient and a fish gradient. Journal of Plankton

Research, 22, 951–968.

Jeppesen E., Søndergaard Ma., Søndergaard Mo.,

Christoffersen K., Jurgens K., Theil-Nielsen J. &

Schluter L. (2002) Cascading trophic interactions in

the littoral zone of a shallow lake. Archiv fur Hydro-

biologie, 153, 533–555.

Jeppesen E., Jensen J.P., Jensen C., Faafeng B., Hessen

D.O., Søndergaard M., Lauridsen T.L., Brettum P. &

Christoffersen K. (2003) The impact of nutrient state

and lake depth on top-down control in lakes: study of

466 lakes from the temperate zone to the Arctic.

Ecosystems, 6, 315–325.

Jepsen N. & Berg S. (2002) The use of winter refuges by

roach tagged with miniature radio transmitters. Hydro-

biologia, 483, 167–173.

446 E. Jeppesen et al.

� 2004 Blackwell Publishing Ltd, Freshwater Biology, 49, 432–447

Jespersen A.-M. & Christoffersen K. (1987) Measure-

ments of chlorophyll a from phytoplankton using

ethanol as extraction solvent. Archiv fur Hydrobiologie,

109, 445–454.

Jones I.J. & Waldron S. (2003) Combined stable isotope

and gut content analysis of food webs in plant-

dominated, shallow lakes. Freshwater Biology, 48,

1396–1407.

Kitchell J.F., Johnson M.G., Minss C.K., Loftus K.H.,

Greig L. & Olver C.H. (1977) Percid habitat; the river

anology. Journal of the Fisheries Research Board of Canada,

34, 1959–1963.

Kubecka J. & Duncan A. (1994). Low fish predation

pressure in the London reservoirs: I. Species composi-

tion, density and biomass. Internationale Revue der

gesamten Hydrobiologie, 79, 143–155.

Lauridsen T.L., Pedersen L.J., Jeppesen E. & Søndergaard

M. (1996) The importance of macrophyte bed size

for cladoceran composition and horizontal migration

in a shallow lake. Journal of Plankton Research, 18,

2283–2294.

Meding M.E. & Jackson L.J. (2003) Biotic, chemical, and

morphometric factors contributing to witner anoxia in

prairie lakes. Limnology and Oceanography, 48, 1633–1642.

Mehner T. (2000) Influence of spring warming on the

predation rate of underyearling fish on Daphnia – a

deterministic simulation approach. Freshwater Biology,

45, 253–263.

Mills E.L. & Forney J.L. (1983) Impact on Daphnia pulex of

predation by young yellow perch in Oneida Lake, New

York. Transactions of the American Fisheries Society, 112,

154–161.

Møller P.H. (1998) Engelsholm sø. In: Sørestaurering i

Danmark: metoder, erfaringer og anbefalinger (Eds M.

Søndergaard, E. Jeppesen & J.P. Jensen ), pp. 145–153.

Miljøstyrelsen, Miljønyt. 28 (In Danish).

Mortensen E., Jensen H.J., Muller J.P. & Timmermann M.

(1990) Fiskeundersøgelser i søer: Undersøgelsesprogram,

fiskeredskaber og metoder. Danmarks Miljøundersøgelser,

teknisk anvisning nr. 3. pp. 60. (In Danish).

Ogilvie S.H. & Mitchell S.F. (1995) A model of mussel

filtration in a shallow New Zealand lake, with

reference to eutrophication control. Archiv fur Hydro-

biologie, 133, 471–482.

Pace M.L., Cole J.J., Carpenter S.R. & Kitchell J.F. (1999)

Trophic cascades revealed in diverse ecosystems.

Trends in Ecology and Evolution, 14, 483–488.

Persson L. & Crowder L.B. (1997) Fish-habitat interac-

tions mediated via ontogenetic niche shifts. In: The

Structuring Role of Submerged Macrophytes in Lakes (Eds

E. Jeppesen, Ma. Søndergaard, Mo. Søndergaard & K.

Christoffersen). Springer-Verlag, New York.

Persson L. & Greenberg L. A. (1990) Juvenile competitive

bottlenecks: the perch (Perca fluviatilis) – roach (Rutilus

rutilus) interaction. Ecology, 71, 44–56.

Rudstam L.G., Lathrop R.C. & Carpenter S.R. (1993) The

rise and fall of a dominant planktivore: direct and

indirect effects on zooplankton. Ecology, 74, 303–319.

Sandby K. (1998) Arreskov Sø. In: Sørestaurering i Dan-

mark: metoder, erfaringer og anbefalinger (Eds M.

Søndergaard, E. Jeppesen & J.P. Jensen), pp. 105–115.

Miljøstyrelsen, Miljønyt. 28. (In Danish).

Scheffer M., Rinaldi S., Kuznetson Y.A. & Van Nes E.H.

(1997) Seasonal dynamics of Daphnia and algae

explained as a periodically forced predator-prey

system. Oikos, 80, 519–532.

Søndergaard M., Kristensen P. & Jeppesen E. (1992)

Phosphorus release from resuspended sediment in the

shallow and wind exposed Lake Arresø, Denmark.

Hydrobiologia, 228, 91–99.

Søndergaard M., Jensen J.P. & Jeppesen E. (1999) Internal

phosphorus loading in shallow Danish lakes. Hydro-

biologia, 408/409, 145–152.

Sommer U., Gliwicz Z.M., Lampert W. & Duncan A.

(1986) The plankton ecology group model of seasonal

succession of planktonic events in fresh waters. Archiv

fur Hydrobiologie, 106, 433–472.

Stansfield J.H., Perrow M.R., Tench L.D., Jowitt A.J.D. &

Taylor A.A.L. (1997) Submerged macrophytes as

refuges for grazing Cladocera against fish predation:

observations on seasonal changes in relation to

macrophyte cover and predation pressure. Hydrobiolo-

gia, 342/343, 229–240.

Sterner R.W. (1989) The role of grazers in phytoplankton

succession. In: Plankton Ecology. Succession in Plankton

Communities (Ed. U. Sommer). Springer-Verlag, New

York.

Tempte J., Allen-Rentmeester Y., Luecke C. & Vanni M.

(1988) Effects of fish summer kill on zooplankton and

phytoplankton populations in Lake Mendota. Annual

International Symposium on Lake and Watershed Manage-

ment 8.

Timms R.M. & Moss B. (1984) Prevention of growth of

potentially dense phytoplankton populations by zoo-

plankton grazing in the presence of zooplanktivorous

fish, in a shallow wetland ecosystem. Limnology and

Oceanography, 29, 472–486.

Vanni M.J., Luecke C., Kitchell J.F., Allen Y. & Temte J.

(1990) Effects on lower trophic levels of massive fish

mortality. Nature, 344, 333–335.

Wetzel R.G. (1983) Limnology, 2nd edn. pp. 860. Saunders

College Publishing, Philadelphia.

(Manuscript accepted 21 January 2004)

Fish predation in lakes in winter 447

� 2004 Blackwell Publishing Ltd, Freshwater Biology, 49, 432–447