Effects of habitat complexity on community ... - Wiley Online Library

Freshwater Biology (2007) 52, 1009–1021

doi:10.1111/j.1365-2427.2007.01748.x

Effects of habitat complexity on community structure and predator avoidance behaviour of littoral zooplankton in temperate versus subtropical shallow lakes M A R I A N A M E E R H O F F , * , †, ‡, § C A R L O S I G L E S I A S , § F R A N C O T E I X E I R A D E M E L L O , § , – J U A N M . C L E M E N T E , ‡, § E L I S A B E T H J E N S E N , * T O R B E N L . L A U R I D S E N * A N D E R I K J E P P E S E N * , † *Department of Freshwater Ecology, National Environmental Research Institute, University of Aarhus, Silkeborg, Denmark † Department of Plant Biology, Institute of Biological Sciences, University of Aarhus, Aarhus, Denmark ‡ Seccioń Limnologıá, Departamento de Ecologıá, Facultad de Ciencias, Universidad de la Repu´blica, Montevideo, Uruguay § Grupo de Investigacioń en Ecologıá Ba´sica y Aplicada, Asociacioń Civil Investigacioń y Desarrollo (I + D), Montevideo, Uruguay – Grupo de Investigacioń en Ecotoxicologıá y Quı´mica Ambiental, Facultad de Ciencias, Universidad de la Repu´blica, Montevideo, Uruguay

SUMMARY 1. Structural complexity may stabilise predator–prey interactions and affect the outcome of trophic cascades by providing prey refuges. In deep lakes, vulnerable zooplankton move vertically to avoid fish predation. In contrast, submerged plants often provide a diel refuge against fish predation for large-bodied zooplankton in shallow temperate lakes, with consequences for the whole ecosystem. 2. To test the extent to which macrophytes serve as refuges for zooplankton in temperate and subtropical lakes, we introduced artificial plant beds into the littoral area of five pairs of shallow lakes in Uruguay (30–35S) and Denmark (55–57N). We used plants of different architecture (submerged and free-floating) along a gradient of turbidity over which the lakes were paired. 3. We found remarkable differences in the structure (taxon-richness at the genus level, composition and density) of the zooplankton communities in the littoral area between climate zones. Richer communities of larger-bodied taxa (frequently including Daphnia spp.) occurred in the temperate lakes, whereas small-bodied taxa characterised the subtropical lakes. More genera and a higher density of benthic/plant-associated cladocerans also occurred in the temperate lakes. The density of all crustaceans, except calanoid copepods, was significantly higher in the temperate lakes (c. 5.5-fold higher). 4. Fish and shrimps (genus Palaemonetes) seemed to exert a stronger predation pressure on zooplankton in the plant beds in the subtropical lakes, while the pelagic invertebrate Chaoborus sp. was slightly more abundant than in the temperate lakes. In contrast, plantassociated predatory macroinvertebrates were eight times more abundant in the temperate than in the subtropical lakes. 5. The artificial submerged plants hosted significantly more cladocerans than the freefloating plants, which were particularly avoided in the subtropical lakes. Patterns indicating diel horizontal migration were frequently observed for both overall zooplankton density and individual taxa in the temperate, but not the subtropical, lakes. In contrast, patterns of diel vertical migration prevailed for both the overall zooplankton and for most individual taxa in the subtropics, irrespective of water turbidity.

Correspondence: Mariana Meerhoff, National Environmental Research Institute, University of Aarhus, Vejlsøvej 25, 8600 Silkeborg, Denmark. E-mail: [email protected]; [email protected] 2007 The Authors, Journal compilation 2007 Blackwell Publishing Ltd

1009

1010

M. Meerhoff et al. 6. Higher fish predation probably shapes the general structure and dynamics of cladoceran communities in the subtropical lakes. Our results support the hypothesis that horizontal migration is less prevalent in the subtropics than in temperate lakes, and that no predatoravoidance behaviour effectively counteracts predation pressure in the subtropics. Positive effects of aquatic plants on water transparency, via their acting as a refuge for zooplankton, may be generally weak or rare in warm lakes. Keywords: diel horizontal migration, diel vertical migration, free-floating plants, refuge effect, subtropical lake

Introduction Trophic cascades (indirect effects of carnivores on plants or algae mediated by herbivores) may have profound impacts on the structure and functioning of several ecosystems, particularly so in aquatic environments (Shurin et al., 2002). Considerable debate still exists about the ultimate mechanisms causing variation in the trophic cascade, whether predators indirectly affect the plants by reducing the numbers of their herbivorous prey or by inducing antipredator behaviour in the prey (Lima & Dill, 1990). In several types of ecosystems, the last mechanism seems to be more important than the classical numerical effects (Schmitz, Krivan & Ovadia, 2004). In lakes, the spatial distribution of fish can have important consequences for the spatial distribution of their prey and may modify the expected outcome of direct interactions (Romare & Hansson, 2003). However, structural complexity can strongly mediate both competitive and predatory interactions by providing prey refuges, thus stabilising predator–prey interactions and, potentially, sustaining more diverse communities. Timms & Moss (1984) first suggested that plants, though not essential for cladoceran growth, could be necessary for their survival in the presence of fish predators. Since then, increasing research has focused on testing the hypothesis that aquatic plants provide refuge for large-bodied grazers against fish predation in shallow lakes (Lauridsen & Lodge, 1996; Burks et al., 2002). In these systems, the pelagic zooplankton often moves horizontally into the littoral area [diel horizontal migration (DHM)], seeking daytime refuge from predators, primarily fish (Lauridsen & Buenk, 1996; Burks, Jeppesen & Lodge, 2001a). This refuge effect of submerged macrophytes, enhancing the survival of pelagic cladocerans with consequent stronger grazing pressure on phytoplankton, helps maintain water transparency (Timms & Moss, 1984;

Scheffer et al., 1993), and indirectly diversity (Declerck et al., 2005), in temperate shallow lakes. Furthermore, the mobility of the pelagic zooplankton may enhance the coupling between pelagic and littoral habitats (Van de Meutter, Stoks & De Meester, 2004), as has been indicated for fish (Schindler & Scheuerell, 2002). By contrast, in deep lakes, vulnerable zooplankton typically move to the hypolimnion during the day, where the colder, darker and less oxygenated water offers refuge from visual predators, and move upwards at night [diel vertical migration (DVM)] (Lampert, 1993; Ringelberg & Van Gool, 2003). Whereas DVM has been documented in many studies, DHM seems far less predictable, however (Burks et al., 2002). The refuge effect of plants for cladocerans, particularly for the important grazer Daphnia, varies with the composition of the potential predators (Jeppesen et al., 1997a), as the plants can also provide a refuge for zooplanktivorous juvenile fish against piscivores (Persson & Eklo¨v, 1995) and shelter many predatory invertebrates (Burks, Jeppesen & Lodge, 2001b). The refuge effect also seems to depend on plant architecture (Nurminen & Horppila, 2002), bed size (Lauridsen et al., 1996), density (Burks et al., 2001a) and volume (Schriver et al., 1995), and on the trophic state of the lake (Lauridsen et al., 1999). Under low-nutrient conditions, high transparency and low density of plants enhance fish predation pressure, while, the refuge effect is also weak under hypertrophic conditions because of the often high density of planktivorous fish and scarcity of submerged plants (Jeppesen et al., 1997b). Several key questions remain about the role of macrophytes in regulating these trophic interactions. The effect of macrophytes under different climates is an issue of increasing importance, particularly when considering the potential impact of global warming on lakes (Burks et al., 2006). Most studies on DHM have focused on Northern temperate lakes (Burks et al.,

2007 The Authors, Journal compilation 2007 Blackwell Publishing Ltd, Freshwater Biology, 52, 1009–1021

Community structure and predator avoidance of zooplankton in warm lakes 2002). Freshwater lakes located in warmer areas typically differ from temperate lakes in several key aspects of their fish communities, which probably exert stronger predation pressure on the zooplankton than in equivalent temperate lakes (Jeppesen et al., 2005; Meerhoff et al., unpubl. data). Presumably as a result, large-bodied pelagic cladocerans are generally rare in (sub)tropical systems (Fernando, 2002), and Daphnia spp. in particular are absent or scarce (Mazumder & Havens, 1998; Pinto-Coelho et al., 2005). Some research suggest a weaker effect of submerged plants as a zooplankton refuge in the subtropics, although only few studies, in eutrophic lakes (Meerhoff et al., 2003; Iglesias et al., 2007) and in the laboratory (Meerhoff et al., 2006), have been conducted hitherto. Although far less studied than the submerged plants, plants of a different architecture (e.g. large free-floating plants) are particularly important in warm climates and may become more important with increasing winter minimum air temperatures. In laboratory behavioural experiments, Daphnia avoided free-floating plants more than submerged plants, although all plant types were avoided even in the presence of alarm signals from crushed conspecifics (Meerhoff et al., 2006). However, it is still unclear whether these results can be generalised to a wider variety of zooplankton species and environmental conditions. Here, we analysed the refuge hypothesis of plants for zooplankton in relation to two main factors: climate (temperate versus subtropical) and plant architecture (submerged versus free-floating) across a gradient of turbidity. We hypothesised that the refuge effect for zooplankton would be substantially lower in a warmer climate, as a result of the higher densities of zooplanktivorous fish, and would decline with decreasing plant structural complexity and with increasing turbidity.

1011

Methods Design and sampling methodology We selected a set of five pairs of shallow lakes along a gradient of water transparency in Uruguay (30–35S) and Denmark (55–57N). We paired the lakes in both countries in terms of Secchi depth (SD): 1.6, 1.0, 0.7, 0.4, and 0.20 m SD (summer means). Within the pairs, the lakes had comparable size, nutrient concentrations and macrophyte cover (range: 0–70% PVI, percent volume inhabited sensu, Canfield et al., 1984) (Table 1). In each lake, we introduced artificial plant beds mimicking submerged and large free-floating plants (four replicates of each) (Fig. 1). The use of artificial substrata inevitably fails to replicate any chemical interactions between plants and animals; however, we ensured that the initial amount and quality of substratum was identical. The modules consisted of 1-m diameter PVC plastic rings with an attached net from which hung the artificial plants. The same plastic material (originally green Christmas tree decorations) was used for both ‘submerged’ and ‘free-floating’ plants. In each submerged plant module, we used 100 0.8–1.0-m long plants, with an architecture resembling Cabomba or Elodea spp. (3.5-cm long ‘leaves’). The freefloating plants (40 per module) consisted of a 15-cm diameter plastic disc to limit light transmission in a patchy manner, with a total length of 2 m of plastic material (with 1.5-cm long ‘root hairs’) arranged in shorter pieces to mimic the root network of Eichhornia crassipes (Mart.) Solms. Both types of habitat modules therefore had a similar structure (between 80 and 100 m of plastic material in total), and 49% and 30% PVI for submerged and free-floating plants, respectively. We calculated the % PVI of the free-floating plants by multiplying the area covered by the discs by the average length of the root network (0.33 m). We placed the habitat modules at 1-m depth in sheltered

Table 1 Main characteristics of the lakes included in the study in the temperate (Temp.) and subtropical (Sub.) regions, ordered by increasing water turbidity (from 1 to 5) 1

2

3

4

5

Turbidity

Temp.

Sub.

Temp.

Sub.

Temp.

Sub.

Temp.

Sub.

Temp.

Sub.

Secchi depth (m) Area (ha) TN (lg L)1) TP (lg L)1) % PVI

1.6 70 340 14 7

1.6 30 350 21 15

1.0 11 420 46 5

1.0 11 410 42 10

0.7 3 1000 76 70

0.7 2.5 1107 159 70

0.4 22 2090 54 5

0.4 28 926 68 10

0.2 2 1050 97 0

0.2 2 839 47 0


1012

M. Meerhoff et al.

Fig. 1 Photograph of the artificial submerged (left) and free-floating (right) plant beds used in the study.

and plant-free areas. The sampling campaigns took place in summer (January 2005 in Uruguay and July 2005 in Denmark, average water temperature 27 C and 18 C, respectively), about 4 weeks after the introduction of the habitat modules in the lakes, thus allowing periphyton and invertebrates to colonise the plastic structures. We sampled the lakes consecutively. In open water, we took water samples for the analysis of total phosphorus, total nitrogen (Valderrama, 1981; Søndergaard, Kristensen & Jeppesen, 1992), chlorophyll-a (Jespersen & Christoffersen, 1987) and alkalinity, and measured in situ parameters (transparency as Secchi depth, PAR (Photosynthetically active radiation) light attenuation with Licor Li-250 radiometer, and pH, temperature and conductivity with Horiba field sensors). We took day and night whole column (from surface to a few centimetre above the sediment) samples for zooplankton with a 1.0-m long tube (6-cm diameter) from each module and from four open water sites nearby (8 L each, filtered through a 50-lm mesh size and preserved with Lugol 4%). We removed one ‘plant’ per module to collect potentially predatory macroinvertebrates (sieved through 500-lm mesh size, and preserved in 70% alcohol). By night, we sampled the fish (and other nekton groups if present, such as shrimps) strictly associated with each plant module using a cylindrical net (1.20-diameter, mesh size 0.3 cm), which was previously placed over the sediment and below each module and subsequently lifted quickly from the boat with a 1.5-m long hook. The same

modules were used in both countries. Before shipping them, they were washed by using high water pressure, disinfected with concentrated chlorine solution, rinsed thoroughly and sun-dried. We identified all cladoceran zooplankton to genus or species and counted at least 50–100 individuals of the most abundant taxa. Copepods were counted as cyclopoids or calanoids (including only adults and copepodites). We classified the cladoceran genera Bosmina, Ceriodaphnia, Daphnia, Diaphanosoma, Leptodora, Moina, Scapholeberis, Simocephalus and Polyphemus as free-swimming/pelagic, and the others as benthic/ plant-associated.

Statistical analyses We analysed the effects of climate and habitat complexity on each target variable, after matching the lakes by their turbidity level, by applying three-way factorial A N O V A (factors: ‘climate’, two levels; ‘habitat’, three levels in the case of zooplankton, i.e. both plant architectures plus open water; ‘turbidity’, five levels). In the case of significant interactions between climate and other factors, we estimated the overall levels of each target variable within climate zones by LSMEANS (PROC MIXED, SAS Institute Inc., 2004), averaging over significant main and interactive effects of turbidity, habitat and time if appropriate. To analyse broad diel spatial patterns of cladocerans, we included ‘time’ as a fourth factor in the A N O V A tests, whereas we studied diel changes in the density of individual taxa at each turbidity level with an exten-


Community structure and predator avoidance of zooplankton in warm lakes Results

ded two-way A N O V A (factors: ‘habitats’ and ‘time’), correcting for potential ‘set-up’ random effects (i.e. correlation between day and night measurements inside the same module). We classified observed patterns as ‘classic’ DHM when the densities in the (submerged) plants decreased in the night with a concomitant, even slight, increase in the open water (significant interaction ‘habitat’ · ’time’ in the A N O V A tests). We classified patterns as ‘classic’ DVM when we found a nocturnal density increase in all habitats (significant effect of ‘time’). This implies that the animals were not caught by the sampling device during the day, suggesting that they were in or just above the sediments (DeStasio, 1993). Diel patterns contrasting those described above were classified as ‘reverse’ (RHM, RVM). However, a nocturnal decrease in all habitats can be interpreted also as extreme DHM, i.e. if the animals were dispersed beyond the sampled littoral area, such as into the true pelagic zone. Data were prior log10 (x + 1)-transformed to fulfil the requirements of homoscedasticity (Cochran’s test) and normal distribution of residuals. Post hoc analyses were made by using the Tukey HSD tests.

Community structure and abundance: the role of climate We found remarkable differences in the structure of the zooplankton communities between climate zones, in terms of taxon-richness, composition and density (Fig. 2). Taxon richness (genera) of Cladocera was significantly higher in temperate lakes (LSMEANS, P < 0.0001). Daphnia spp., while frequent and abundant in the temperate lakes, was found in only one lake in the subtropical set and even there was rare. Small-bodied Diaphanosoma, Bosmina, Moina and Ceriodaphnia (ordered by their frequency of occurrence), and in lower densities the large-bodied Simocephalus, typically comprised the free-swimming Cladocera in the subtropical lakes. Furthermore, the benthic/plantassociated cladocerans comprised a more diverse assemblage (at the genus level) in the temperate lakes, including the large-bodied, plant-attached Sida crystallina Mu¨ller and Eurycercus lamellatus Mu¨ller. The chydorids Chydorus, Pleuroxus, Acroperus, Alona and Alonella were also frequent (Fig. 2). By contrast, in the subtropics, we found fewer taxa, Chydorus, Alona and

Free-floating plants

Submerged plants

1013

Open water Subtropical Temperate

Ceriodaphnia Diaphanosoma Bosmina Daphnia Simocephalus Leptodora Polyphemus Scapholeberis Moina

Pelagic free-swimming

Alona Pleuroxus Acroperus Disparalona Chydorus Sida Eurycercus Alonella Pseudochydorus Iliocryptus Camptocercus Leydigia Macrothrix Graptoleberis

Benthic/ plant-associated

Cyclopoid copep. Calanoid copep. 0

20

40

220 0

5

10

15

100 0

5

10

15

20

25

Density (nos. L–1)

Fig. 2 Mean density of free-swimming/pelagic and benthic/plant-associated cladoceran genera and adult copepods (calanoid and cyclopoid) in temperate and subtropical lakes in each of the three sampled habitats. The means represent the overall average of day– night mean densities in each lake (±1 SE). 2007 The Authors, Journal compilation 2007 Blackwell Publishing Ltd, Freshwater Biology, 52, 1009–1021

M. Meerhoff et al.

2.4

1.2

2.4

2.0

1.0

2.0

0.8

1.6

0.8

1.6

0.6

1.2

0.6

1.2

0.4

0.8

0.4

0.8

0.2

0.4

0.2

0.4

40

40

30

30

20

20

10

10

1.0

(b)

0

0 3500 3000 2500 2000 1500 1000 500 0 60 50

500

(c)

400 300 200 100 0

(d)

40 30

40 30

20

20

10

10 0 350 300 250 200 150 100 50 0

0

0

0

(e)

160 120 80 40 0

60

40

50

30

40 30

20

20

10

10

Calanoids (nos. L–1)

Chaoborus (nos. L–1)

1.2

0

0 60

16

50

12

40

8

30 20

4

10

Benthic clad. (nos. L–1)

Fi s h (nos. m–2)

50

Shrimps (nos. m–2)

0

700 600 500 400 300 200 100 0

60

50

0

Pred. macroinvert. (nos. m–2)

Subtropical 700 600 500 400 300 200 100 0

(a)

Polyph. + Lept. (nos. L–1)

Temperate 60

Cyclopoids (nos. L–1)

Pleuroxus being the most representative benthic/ plant-associated genera. The density of cladocerans was also significantly higher in the temperate lakes, when considering both the total and the free-swimming genera (LSMEANS, P < 0.0001, on average 5.5-fold higher) (Fig. 2). The density of free-swimming genera varied from 4.4 (±0.6 SE) to 46.5 (±9.1 SE) L)1 in the subtropical lakes, and from 24.4 (±3.9 SE) to 284.1 (±65.1 SE) L)1 in the temperate lakes (whole lake sample means, i.e. average of the day and night densities in the different habitats). The density of the benthic/plant-associated genera was also higher in the temperate lakes, the magnitude of the differences being affected by ‘turbidity’ and ‘habitat’ (significant interaction terms in A N O V A test). We found more cyclopoid than calanoid copepods in the temperate lakes (23.1 ± 1.9 SE and 3.2 ± 1.0 SE L)1, respectively, average of whole lake sample means across turbidity levels). However, similar average densities of cyclopoid and calanoid copepods occurred in the subtropics (10.1 ± 1.1 SE and 9.6 ± 1.2 SE L)1, respectively, whole lake sample means across turbidity levels). These patterns were maintained for copepodites as well. The number of genera and the density of free-swimming and benthic/plant-associated cladocerans varied along the turbidity gradient (Fig. 3e). However, because of the significant interaction between ‘turbidity’ and ‘climate’ or ‘habitats’ in the A N O V A tests, no direct effect of turbidity on cladoceran density could be identified. The assemblage structure and density of zooplankton predators also differed substantially between climate zones (Fig. 3a–d). The subtropical fish communities associated with the plant beds were dominated by small-bodied individuals and were much more numerous than in the equivalent temperate lakes (Meerhoff et al., unpubl. data). The omnivorous shrimp Palaemonetes argentinus Nobilli (Collins, 1999) was often present and numerous in the subtropics. In almost all lakes in both climate zones, we found some invertebrate predators in the pelagic, adding to the predation pressure exerted by fish. The predatory midge Chaoborus was more frequent and slightly more abundant in the subtropical lakes (0.31 ± 0.05 SE versus 0.19 ± 0.04 SE L)1, ‘climate’ effect, F(1,206) ¼ 3.74, P ¼ 0.054). However, the pelagic cladoceran predators Leptodora kindtii Focke (giant water flea) and Polyphemus pediculus L. occurred only in the temperate lakes and showed a pattern quite opposite

Pelagic clad. (nos. L–1)

1014

0

0

1 2 3 4 5

1 2 3 4 5

Turbidity level

Turbidity level

Fig. 3 Changes in mean density (±1 SE) along a categorical turbidity gradient in the artificial plant beds (average of all habitats and times for zooplankton) in temperate (left) and subtropical (right) lakes of: (a) fish and shrimps (nos. per m2), (b) pelagic invertebrate predators (i.e. Chaoborus; Leptodora + Polyphemus, nos. per litre), (c) littoral predatory macroinvertebrates (nos. per m2), (d) adult calanoid and cyclopoid copepods (nos. per litre), and (e) free-swimming/pelagic and benthic/plantassociated cladocerans (nos. per litre). The lakes are ordered by increasing turbidity level (actual Secchi depths in Table 1). Notice the different scales for the two climate zones.

to that of Chaoborus (Fig. 3b). The plant-associated predatory macroinvertebrates were more diverse and numerous (c. eight-fold more dense) in the temperate lakes than in the equivalent subtropical lakes. The abundance of none of the potential predators could be directly related to turbidity, and neither could cladoceran or copepod density (Fig. 3).


Community structure and predator avoidance of zooplankton in warm lakes

1015

The role of plant architecture In both climate zones, we found the highest proportion of cladocerans (total and free-swimming) among the submerged plants, although the magnitude of the ‘habitat’ effect was affected by ‘climate’ and ‘turbidity’ (i.e. significant interaction terms). In the temperate lakes, 55.0% (±10.5 SE) of the free-swimming cladocerans occurred in the submerged plants (Tukey’s test, post hoc comparison between plant types, P ¼ 0.029), while 46.8% (±7.7 SE) did so in the subtropical lakes (Tukey’s test, P ¼ 0.008). In the temperate lakes, the density of cladocerans among the free-floating plants was intermediate between submerged plants and open water whereas, in the subtropical lakes, density was always lowest in the free-floating plants (Fig. 4). The benthic/plant-associated cladocerans were also more numerous among the submerged plants in both climate zones (three-way A N O V A , ‘habitat’ effect, F(2,225) ¼ 82.4, P < 0.001, Fig. 4). In general, the cyclopoid copepods were more abundant with increasing structure. That is, the highest densities occurred among the submerged plants followed by free-floating plants and, lastly, open water. By contrast, the calanoid copepods presented the opposite pattern in both climate zones. Most of the potential predators also used the two plant types differently. Despite often being more numerous in open water, the densities of Chaoborus did not differ significantly among habitats in any of the climate zones. L. kindtii had higher densities in open water, while P. pediculus appeared only in the plant beds and during the day. In the temperate zone, fish were more pelagic, as indicated by their stronger association with the free-floating plants, whereas fish were significantly associated with the submerged plant beds in the subtropics (three-way A N O V A , interaction ‘plants’ · ‘climate’, F(1,63) ¼ 5.85, P ¼ 0.018, Fig. 4). The plant-associated predatory macroinvertebrates had higher densities in the free-floating plants (per unit of plant weight) in both climate zones.

Diel movement patterns The diel changes in total density of the free-swimming cladocerans differed remarkably between climate zones, although the magnitude was affected by turbidity (four-way A N O V A , significant third order interaction, F(8,184) ¼ 2.19, P < 0.0293). We found higher

Fig. 4 Spatial distribution of fish, and diel spatial distribution of free-swimming/pelagic and benthic/plant-associated cladocerans in temperate and subtropical lakes. The data represent the sample means (±1 SE) of five lakes (average of lake averages). FF, free-floating plants; S, submerged plants; OW, open water. Notice the different scales for both climate zones.

relative density of cladocerans during the day in the temperate lakes (60.7% ± 7.4 SE), but during the night-time in the subtropics (61.0% ± 9.1 SE). In the temperate zone, the most frequently observed patterns were consistent with the predictions of the classic DHM: the density of free-swimming cladocerans in the submerged plants decreased during the night and


M. Meerhoff et al.

1016

Predicted migration patterns DHM

DVM

Temperate 1

2

10

1000

40

800

30

8 6 4 2 0

Turbidity level

4

Free-swimmning cladocerans (nos. L–1)

600

3

20

400

10

200 0

0

500

0

140 120 100 80 60 40 20 0

1000

20

800

15

400 300 200 100

600

10

400

5

200

5

Subtropical

30 25 20 15 10 5 0

0

0

700 600 500 400 300 200 100 0

50 40 30 20 10

Day

Open water

Night Free-floating

0

Day

Night

Submerged plants

Fig. 5 Predicted patterns of ‘classic’ diel horizontal migration (DHM) and diel vertical migration (DVM) (above), and the observed diel changes in density in the open water, free-floating plants and submerged plants of the population of free-swimming/pelagic cladocerans, in temperate (left) and subtropical (right) lakes along a categorical turbidity gradient (increasing from top to bottom) (below). We did not have a priori expectations about the behaviour of cladocerans regarding the freefloating plants in a DHM context. Notice the different scales for both climate zones.

increased slightly in the open water (Fig. 5, Table 2). In the subtropical lakes, however, we found clear signs of the classic DHM in only one lake, whereas in three lakes we found significant evidence of DVM. These

DVM patterns occurred at both extremes of the turbidity gradient in the subtropics, while in the temperate zone DHM seemed to occur at all turbidity levels except in the clearest lake (Table 2). In this lake, the total density of free-swimming cladocerans varied according to a reverse horizontal migration (RHM) with higher density by night in the free-floating and submerged plants and lower density in open water. Cyclopoid copepods were often more numerous at night in all habitats in both climate zones, which fit with the DVM pattern. The calanoids displayed DVM in most temperate lakes, whereas patterns were more diverse in the subtropical lakes (Table 2). In some cases, the behaviour of the individual taxa most susceptible to predation differed from the overall population pattern (Table 2). We observed a broad set of responses by the same taxa, both across and within each climate zone (Table 2). However, when considering the behaviour of individual herbivorous cladoceran taxa, some general patterns emerged. Firstly, the most frequent response of the largest taxa in each lake was DVM (in seven out of 10 lakes). In the temperate set, DVM was the pattern displayed by the largest taxa in four lakes covering the whole turbidity gradient. In the subtropical set, DVM by the largest taxa present occurred under medium-to-high turbidity conditions. In the clearest lake, the most abundant Bosmina showed significant DVM, whereas the slightly larger but less abundant Ceriodaphnia displayed DHM. Secondly, DHM represented the commonest behaviour performed by individual herbivorous taxa in the temperate lakes (71% of the statistically significant patterns in the A N O V A tests), whereas DVM was most common in the subtropics (67% of the significant responses) (Table 2). Other spatial patterns, such as those fitting with RHM and reverse vertical migration, were seldom found (Table 2). In a few cases, we interpreted the latter pattern as a nocturnal dispersal of the animals beyond the sampled area (i.e. into the true pelagic zone). This was the case for Ceriodaphnia in Lake Stigsholm (Denmark, turbidity level 4), where densities in the three sampled habitats decreased by 50% at night-time. Chaoborus performed DVM under all environmental conditions (climate and turbidity level), with significantly higher densities at night. The other pelagic predators, L. kindtii and P. pediculus, appeared mostly during the day.


Community structure and predator avoidance of zooplankton in warm lakes

1017

Table 2 Migration patterns of the total assemblage of free-swimming cladocerans and of the most abundant zooplankters in each pair of lakes (temperate and subtropical) along the turbidity gradient (increasing from 1 to 5). The herbivorous cladoceran taxa are ordered by decreasing body size. Bold letters indicate significant results in the two-way A N O V A tests. Symbols: DVM, diel vertical migration; DHM, diel horizontal migration; RHM, diel reverse horizontal migration; RVM, diel reverse vertical migration; No M, no signs of migration; –, genus absent; d, genus present in just one habitat or too scarce numbers for statistical tests; ?, interpreted pattern despite differing from predictions (e.g. a nocturnal decrease in density in submerged plants and open water was interpreted as DHM). The last four rows indicate the total number of each significant pattern by the herbivorous cladoceran taxa that occurred in each lake (nonsignificant patterns in parentheses) 1

2

3

4

5

Turbidity

Temp.

Sub.

Temp.

Sub.

Temp.

Sub.

Temp.

Sub.

Temp.

Sub.

Clad. population Sida Eurycercus Daphnia Simocephalus Diaphanosoma Ceriodaphnia Moina Bosmina Calanoid cop. Cyclopoid cop. Predators Leptodora Polyphemus Chaoborus Total herb. DHM Total herb. DVM Total herb. RHM Total herb. RVM

RHM D D – D D DVM – RHM – DVM

DVM – – d d d DHM – DVM DVM DVM

DHM? DVM d DHM No M DHM No M – d DVM DVM

DHM – – – RVM DHM – – – DHM DHM

DHM DVM DHM DVM DHM DHM DHM – DVM RVM DHM

No M – – – DVM DHM DHM RHM d DHM DVM

DHM? DHM DHM DHM? – DHM DHM? – DHM DVM DHM

DVM – – – – DVM – DVM DVM RVM DHM

DHM DVM d – d DHM DHM – DHM DVM DVM

DVM – – – – DVM – DVM – RVM RVM

DVM – – 0 1 (1) 0

– – DVM (1) 1 0 0

RHM RVM DVM 1 (1) 1 0 0

– – DVM 1 0 0 (1)

– – DVM 2 (2) 1 (2) 0 0

– – DVM 1 (1) (1) (1) 0

RHM RVM – 1 (5) 0 0 0

– – DVM 0 2 (1) 0 0

– – DVM 1 (2) (1) 0 0

– – – 0 1 (1) 0 0

Discussion We found notable differences in the zooplankton (taxon richness, density and size) and its dynamics between the two climate zones, regardless of the gradient in water transparency and other environmental variables (such as nutrient status and lake area). The subtropical zooplankton was much less dense and less diverse, and the cladocerans were smaller than in similar temperate lakes. Although features other than climate that may affect zooplankton could differ between our two sampled zones (e.g. biogeographical processes), our results are in line with various pieces of evidence that suggest strong climate-related differences in the structure of the zooplankton communities. The cladoceran community structure found in the littoral area of the subtropical lakes concurs with other studies suggesting that large-bodied pelagic cladocerans, and particularly Daphnia spp., are infrequent in the (sub)tropics (Mazumder & Havens, 1998; Fernando, 2002; Pinto-Coelho et al., 2005). However, there are

exceptions to this rule of thumb. Thus, remarkably rich Daphnia communities have appeared in the sediment records of Lake Naivasha, Kenya (Mergeay et al., 2004). In addition, large-bodied Daphnia, even larger than predicted by latitudinal models (Gillooly & Dodson, 2000), are abundant as long as fish are absent in Lake Rivera, Uruguay (N. Mazzeo & G. Lacerot, Universidad de la Republica, Montevideo unpubl. data), while ephippia and remains of Daphnia and other large cladocerans have been found in the sediments of several of the shallow lakes studied in Uruguay (K. Jensen, NERI, Silkeborg unpubl. data), even though these taxa were absent in the contemporary samples. This indicates that (sub)tropical lakes can have large cladocerans, but are apparently unable to sustain them for long periods or in high numbers. Interestingly, we found that the relative share of total copepod density attributable to calanoids was higher in the subtropics. Calanoid copepods are often relatively abundant in warm lakes (G. Lacerot, unpubl. data), although this pattern is not universal (Pinto-Coelho et al., 2005).


1018

M. Meerhoff et al.

The same important structural differences between both climate regions occurred in the benthic/plantassociated cladoceran communities. Because of the sampling device used, we may have underestimated the density of the strictly benthic or plant-associated animals. However, the comparison between both climate regions is valid as we used the same methods in all lakes. Our findings differ from those of a study conducted along a climate gradient in Europe, where no climate effect was found for the plant-associated cladocerans (Gyllstro¨m et al., 2005). These authors argue, however, that the higher plant %PVI found in warmer lakes could have masked potential climate effects. Although very important for cladoceran physiology and competition (Moore, Folt & Stemberger, 1996), higher temperature per se does not seem to explain directly the low abundance and size of cladocerans in the (sub)tropics. In warm lakes in southern Europe (Gyllstro¨m et al., 2005), Daphnia spp. appeared scarcely, while fish biomass and the fish:zooplankton biomass ratio were higher than in the cold and the temperate lakes. Taking into account the high density of fish and shrimps in the littoral of our subtropical lakes, it is reasonable to assume that the lower number of genera and lower density of both freeswimming/pelagic and benthic/plant-associated cladocerans observed here are the result of a very high predation pressure. The lower periphyton biomass in the subtropical lakes (Meerhoff et al., unpubl. data) also probably influenced the assemblage of plant associated cladocerans. In agreement with previous suggestions (Dumont, 1994; Fernando, 2002), our results therefore support the hypothesis that higher predation pressure in warmer lakes is the key factor for the observed patterns in the assemblage composition and taxon richness of cladoceran communities. According to our results, plant architecture and climate affect the use of macrophytes as a refuge by the zooplankton. In support of our hypothesis, we found a higher density of cladocerans among submerged than free-floating plants in both climate zones. In the temperate lakes, the use of free-floating plants by cladocerans followed intermediate diel trends and densities were intermediate between those in submerged plants and open water. The more complex structure of the submerged plants seemed to offer better conditions for the free-swimming cladocerans, even in the subtropics, despite the high

density of fish among the plants (Meerhoff et al., unpubl. data). Despite supporting lower densities of fish than the submerged plants, free-floating plants were strongly avoided by subtropical cladocerans. This agrees with previous studies in single lakes with real plants (water hyacinth E. crassipes), in which crustaceans were less numerous under free-floating than among submerged plants (Meerhoff et al., 2003) or in open water (Brendonck et al., 2003). Our results also give field support to laboratory behavioural experiments (Meerhoff et al., 2006), in which Daphnia obtusa Kurz avoided the free-floating plants to a greater extent than the submerged plants. Factors other than direct macrophyte cues (as the ‘plants’ were plastic) must account for the avoidance by cladocerans in our study. The free-floating plants may produce differences in the light environment, which could be used by cladocerans as a proximate cause of migration, as occurs in DVM (Siebeck, 1980; Ringelberg & Van Gool, 2003). A lower food quantity because of shading of phytoplankton and persistently low oxygen concentrations under dense free-floating mats, in addition to the high predation risk, may be the ultimate causes of the strong aversion of subtropical cladocerans towards the free-floating plants. Patterns of DHM and DVM occurred simultaneously in both sets of the studied shallow lakes. However, our study support the hypothesis that DHM is less prevalent in the subtropics than in temperate lakes (Burks et al., 2002). Diel variation in the risk of fish predation is implicated as the main factor explaining horizontal density gradients in the zooplankton (Gliwicz & Rykowska, 1992) or their horizontal migration (Lauridsen & Buenk, 1996). We lack data on fish densities in the open water during these field studies; however, fish densities are usually very high in all habitats in the Uruguayan lakes, particularly in summer (Iglesias et al., 2007), compared with temperate Danish lakes (Jeppesen et al., 2003). Besides the risk from fish in the pelagic, swimming in open water also exposes cladocerans to encounters with Chaoborus in the subtropical lakes (Iglesias et al., 2007). We found no thresholds in the densities of fish or other potential littoral predators leading to particular predator-avoidance behaviours of cladocerans. However, two lines of evidence (i.e. the overall patterns of the free-swimming Cladocera assemblages under


Community structure and predator avoidance of zooplankton in warm lakes contrasting climates, and the behaviour of the most vulnerable individual taxa in specific lakes) support the idea that DVM represents the commonest response of both individual and overall cladocerans when the predation risk, or the perception of predation, is very high. In most lakes in both climate regions, the largest cladocerans appeared to be close to the sediment surface during the day, thus avoiding visual predators (DeStasio, 1993). In experimental studies with the tropical Daphnia lumholtzi Sars, pronounced DVM occurred in the absence of predator cues (Havel & Lampert, 2006). The authors suggest that an innate response may be more profitable than an inducible response under the continuously strong predation pressure in the tropics. Our results suggest that the DVM response might be the default under high predation risk conditions, although it may not necessarily be innate for all cladocerans. From a broad perspective, our study indicates that DHM occurs under conditions of reduced fish predation (or perception) risk, even if high densities of predacious macroinvertebrates occur in the littoral (which theoretically decreases the likelihood of DHM, Burks et al., 2001b), as we observed in the temperate lakes in contrast to the subtropics. In partial agreement with a proposed model for temperate shallow lakes (Jeppesen et al., 1997b), temperate cladocerans overall and most individual cladoceran taxa used submerged plants as a daytime refuge under mesotrophic and eutrophic conditions, but not under oligotrophic conditions where fish predation risk is expectedly higher (Jeppesen et al., 2003). However, we found evidence of DHM even in very turbid lakes in the temperate region. In contrast to the above-mentioned model, and our own hypothesis, we observed no relationship between the antipredator behaviour and turbidity in the subtropical lakes, as patterns of DVM occurred under both oligotrophic and eutrophic-hypertrophic conditions. DHM, the predator-avoidance behaviour most frequently described for temperate shallow lakes, clearly seems insufficient in the subtropics. Despite the occurrence of DVM, the density and structure of the cladoceran community indicate that no predator-avoidance behaviour can effectively counteract the effects of the high predation pressure in subtropical lakes. To conclude, our results indicate that any positive effect of aquatic plants on water transparency, mediated through their use as a refuge by the zooplankton,

1019

may be weak or rare in warm lakes. This supports previous investigations in single subtropical (Meerhoff et al., 2003; Iglesias et al., 2007) and Mediterranean lakes (Castro, Marques & Gonçalves, 2007), reviews (Jeppesen et al., 2005) and laboratory studies (Meerhoff et al., 2006). The lack of an efficient refuge among the plants may also be a key factor explaining the minor effect of submerged plants on water clarity observed in many subtropical lakes in Florida (USA) (Bachmann et al., 2002) compared with temperate lakes with the same nutrient concentrations (Jeppesen et al., 2007).

Acknowledgments We warmly thank R. Ballabio, L. Boccardi, A. Borthagaray, C. Bruzzone, H. Caymaris, L. Cavada, C. Fosalba, G. Goyenola, N. Mazzeo, W. Noordoven, T. Olivera and the personnel of Aguas de la Costa S.A. (in Uruguay), and to M. Bramm, T. Buchaca, L.S. Hansen, K. Jensen, T.B. Kristensen, F. Landkildehus, T. Larsen, Z. Pekcan-Hekim, K. Stefanidis, J. Stougaard Pedersen, K. Thomsen and Y.S. Tunali (in Denmark) for their help in the field or in the laboratory during the sampling campaigns. Thanks also to A.R. Pedersen for statistical advice, and to A.M. Poulsen and J. Jacobsen for editorial assistance. We acknowledge the constructive comments by reviewers and editor A. Hildrew. The landowners and managers are acknowledged for their permission to enter the lakes (IMM, MGAP, OSE, Aguas de la Costa S.A., Cabanã Tropicalia S.A. & Rossi family). This work was funded by the Danish Research Agency. EJ and TLL were supported by the projects ‘Conwoy’ (Danish Natural Science Research Council), ‘Clear’ (a Villum Kann Rasmussen Centre of Excellence Project) and ‘Eurolimpacs’ (EU).

References Bachmann R.W., Horsburgh C.A., Hoyer M.V., Mataraza L.K. & Canfield D.E. Jr. (2002) Relations between trophic state indicators and plant biomass in Florida lakes. Hydrobiologia, 470, 219–234. Brendonck L., Maes J., Rommens W. et al. (2003) The impact of water hyacinth (Eichhornia crassipes) in a eutrophic subtropical impoundment (Lake Chivero, Zimbabwe). II. Species diversity. Archiv fu¨r Hydrobiologie, 158, 389–405.


1020

M. Meerhoff et al.

Burks R.L., Jeppesen E. & Lodge D.M. (2001a) Littoral zone structures as refugia for Daphnia against fish predation. Limnology and Oceanography, 46, 230–237. Burks R.L., Jeppesen E. & Lodge D.M. (2001b) Pelagic prey and benthic predators: impact of odonate predation on Daphnia. Journal of the Northern American Benthological Society, 20, 615–628. Burks R.L., Lodge D.M., Jeppesen E. & Lauridsen T.L. (2002) Diel horizontal migration of zooplankton: costs and benefits of inhabiting littoral zones. Freshwater Biology, 47, 343–365. Burks R.L., Mulderij G., Gross E., Jones J.I., Jacobsen L., Jeppesen E. & Van Donk E. (2006) Center stage: the crucial role of macrophytes in regulating trophic interactions in shallow lake wetlands. In: Wetlands: Functioning, Biodiversity, Conservation, and Restoration (Eds R. Bobbink, B. Beltman, J.T.A. Verhoeven & D.F. Whigham), pp. 37–59. Springer-Verlag, Berlin Heidelberg. Canfield D.E. Jr., Shireman J.V., Colle D.E. & Haller W.T. (1984) Prediction of chlorophyll-a concentrations in Florida lakes: importance of aquatic macrophytes. Canadian Journal of Fisheries and Aquatic Sciences, 41, 497–501. Castro B.B., Marques S.M. & Gonçalves F. (2007) Habitat selection and diel distribution of the crustacean zooplankton from a shallow Mediterranean lake during the turbid and clear water phases. Freshwater Biology, 52, 421–433. Collins P.A. (1999) Feeding of Palaemonetes argentinus (Decapoda: Palaemonidae) from an oxbow lake of the Parana´ River, Argentina. Journal of Crustacean Biology, 19, 485–492. De Stasio B.T. (1993) Diel vertical and horizontal migration by zooplankton - population budgets and the diurnal deficit. Bulletin of Marine Science, 53, 44–64. Declerck S., Vandekerkhove J., Johansson L. et al. (2005) Multi-group biodiversity in shallow lakes along gradients of phosphorus and water plant cover. Ecology, 86, 1905–1915. Dumont H.J. (1994) On the diversity of the Cladocera in the tropics. Hydrobiologia, 272, 27–38. Fernando C.H. (2002) Zooplankton and tropical freshwater fisheries. In: A Guide to Tropical Freshwater Zooplankton. Identification, Ecology and Impact on Fisheries (Ed. C.H. Fernando), pp. 255–280. Backhuys Publishers, Leiden. Gillooly J.F. & Dodson S.I. (2000) Latitudinal patterns in the size distribution and seasonal dynamics of new world, freshwater cladocerans. Limnology and Oceanography, 45, 22–30. Gliwicz Z.M. & Rykowska A. (1992) Shore avoidance in zooplankton: a predator-induced behavior or predator-

induced mortality. Journal of Plankton Research, 14, 1331–1342. Gyllstro¨m M., Hansson L.-A., Jeppesen E. et al. (2005) The role of climate in shaping zooplankton communities of shallow lakes. Limnology and Oceanography, 50, 2008–2021. Havel J.E. & Lampert W. (2006) Habitat partitioning of native and exotic Daphnia in gradients of temperature and food: mesocosm experiments. Freshwater Biology, 51, 487–498. Iglesias C., Goyenola G., Mazzeo N., Meerhoff M., Rodo´ E. & Jeppesen E. (2007) Horizontal dynamics of zooplankton in subtropical Lake Blanca (Uruguay) hosting multiple zooplankton predators and aquatic plant refuges. Hydrobiologia, in press. Jeppesen E., Jensen J.P., Søndergaard M., Lauridsen T., Pedersen L.J. & Jensen L. (1997a) Top-down control in freshwater lakes: the role of nutrient state, submerged macrophytes and water depth. Hydrobiologia, 342/343, 151–164. Jeppesen E., Lauridsen T.L., Kairesalo T. & Perrow M. (1997b) Impact of submerged macrophytes on fishzooplankton interactions in lakes. In: The Structuring Role of Submerged Macrophytes in Lakes (Eds E. Jeppesen, Ma. Søndergaard, Mo. Søndergaard & K. Cristoffersen), pp. 91–114. Ecological Studies. Springer Verlag, New York. Jeppesen E., Jensen J.P., Jensen C., Faafeng B., Hessen D.O., Søndergaard Ma, Lauridsen T.L., Brettum P. & Christoffersen K. (2003) The impact of nutrient state and lake depth on top-down control in the pelagic zone of lakes: a study of 466 lakes from the temperate zone to the Artic. Ecosystems, 6, 313–325. Jeppesen E., Søndergaard M., Mazzeo N., Meerhoff M., Branco C., Huszar V. & Scasso F. (2005) Lake restoration and biomanipulation in temperate lakes: relevance for subtropical and tropical lakes. In: Restoration and Management of Tropical Eutrophic Lakes (Ed. V. Reddy), pp. 341–359. Science Publishers, Inc., Enfield. Jeppesen E., Søndergaard M., Meerhoff M., Lauridsen T.L. & Jensen J.P. (2007) Shallow lake restoration by nutrient loading reduction – some recent findings and challenges ahead. Hydrobiologia, in press. Jespersen A.M. & Christoffersen K. (1987) Measurements of chlorophyll-a from phytoplankton using ethanol as extraction solvent. Archiv fu¨r Hydrobiologie, 109, 445–454. Lampert W. (1993) Ultimate causes of diel vertical migration of zooplankton: new evidence for the predator-avoidance hypothesis. Archiv fu¨r Hydrobiologie Ergebnisse der Limnologie, 39, 79–88. Lauridsen T.L. & Buenk I. (1996) Diel changes in the horizontal distribution of zooplankton in the littoral


Community structure and predator avoidance of zooplankton in warm lakes zone of two shallow eutrophic lakes. Archiv fu¨r Hydrobiologie, 137, 161–176. Lauridsen T.L. & Lodge D.M. (1996) Avoidance by Daphnia magna of fish and macrophytes: chemical cues and predator-mediated use of macrophyte habitat. Limnology and Oceanography, 41, 794–798. 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. Lauridsen T.L., Jeppesen E., Mitchell S.F., Lodge D.M. & Burks R.L. (1999) Diel variation in horizontal distribution of Daphnia and Ceriodaphnia in oligotrophic and mesotrophic lakes with contrasting fish densities. Hydrobiologia, 409, 241–250. Lima S.L. & Dill L.M. (1990) Behavioural decisions made under the risk of predation: a review and prospectus. Canadian Journal of Zoology, 68, 619–640. Mazumder A. & Havens K.H. (1998) Nutrient-chlorophyll-Secchi relationships under contrasting grazers communities in temperate versus subtropical lakes. Canadian Journal of Fisheries and Aquatic Sciences, 55, 1652–1662. Meerhoff M., Mazzeo N., Moss B. & Rodrı´guez-Gallego L. (2003) The structuring role of free-floating versus submerged plants in a shallow subtropical lake. Aquatic Ecology, 37, 377–391. Meerhoff M., Fosalba C., Bruzzone C., Mazzeo N., Noordoven W. & Jeppesen E. (2006) An experimental study of habitat choice by Daphnia: plants signal danger more than refuge in subtropical lakes. Freshwater Biology, 51, 1320–1330. Mergeay J., Verschuren D., Kerckhoven L.V. & De Meester L. (2004) Two hundred years of a diverse Daphnia community in Lake Naivasha (Kenya): effects of natural and human-induced environmental changes. Freshwater Biology, 49, 998–1013. Moore M.V., Folt C.L. & Stemberger R.S. (1996) Consequences of elevated temperatures for zooplankton assemblages in temperate lakes. Archiv fu¨r Hydrobiologie, 135, 289–319. Nurminen L. & Horppila J.A. (2002) A diurnal study on the distribution of filter feeding zooplankton: effect of emergent macrophytes, pH and lake trophy. Aquatic Science, 64, 198–206. Persson L. & Eklo¨v P. (1995) Prey refuges affecting interactions between piscivorous perch and juvenile perch and roach. Ecology, 76, 70–81. Pinto-Coelho R., Pinel-Alloul B., Methot G. & Havens K.E. (2005) Crustacean zooplankton in lakes and reservoirs of temperate and tropical regions: variation

1021

with trophic status. Canadian Journal of Fisheries and Aquatic Sciences, 62, 348–361. Ringelberg J. & Van Gool E. (2003) On the combined analysis of proximate and ultimate aspects in diel vertical migration (DVM) research. Hydrobiologia, 491, 85–90. Romare P. & Hansson L.-A. (2003) A behavioural cascade: top-predator induced behavioural shifts in planktivorous fish and zooplankton. Limnology and Oceanography, 48, 1956–1964. 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. SAS Institute Inc. (2004) SAS OnlineDoc 9.1.3. SAS Institute Inc., Cary, NC. Scheffer M., Hosper S.H., Meijer M.-L., Moss B. & Jeppesen E. (1993) Alternative equilibria in shallow lakes. Trends in Ecology and Evolution, 8, 275–279. Schindler D.E. & Scheuerell M.D. (2002) Habitat coupling in lake ecosystems. Oikos, 98, 177–189. Schmitz O.J., Krivan V. & Ovadia O. (2004) Trophic cascades: the primacy of trait-mediated indirect interactions. Ecology Letters, 7, 153–163. Schriver P., Bøgestrand J., Jeppesen E. & Søndergaard M. (1995) Impact of submerged macrophytes on fishzooplankton-phytoplankton interactions: large-scale enclosure experiments in a shallow eutrophic lake. Freshwater Biology, 33, 255–270. Shurin J.B., Borer E.T., Seabloom E.W., Anderson K., Blanchette C.A., Broitman B., Cooper S.D. & Halpern B.S. (2002) A cross-ecosystem comparison of the strength of trophic cascades. Ecology Letters, 5, 785–791. Siebeck O. (1980) Optical orientation of pelagic crustaceans and its consequences in the pelagic and littoral zones. In: Evolution and Ecology of Zooplankton Communities (Ed. W.C. Kerfoot), pp. 28–38. The University Press of New England, Hannover. Timms R.M. & Moss B. (1984) Prevention of growth of potentially dense phytoplankton populations by zooplankton grazing, in the presence of zooplanktivorous fish, in a shallow wetland ecosystem. Limnology and Oceanography, 29, 472–486. Valderrama J.C. (1981) The simultaneous analysis of total N and total P in natural waters. Marine Chemistry, 10, 109–122. Van de Meutter F., Stoks R. & De Meester L. (2004) Behavioral linkage of pelagic prey and littoral predators: microhabitat selection by Daphnia induced by damselfly larvae. Oikos, 107, 265–272. (Manuscript accepted 29 January 2007)
