Habitat selection and diel distribution of the crustacean
zooplankton from a shallow Mediterranean lake during
the turbid and clear water phases
BRUNO B. CASTRO, SE´RGIO M. MARQUES AND F. GONC¸ALVES
Department of Biology & Centre for Environmental and Marine Studies (CESAM), University of Aveiro, Aveiro, Portugal
SUMMARY
1. The fish fauna of many shallow Mediterranean Lakes is dominated by small-bodied
exotic omnivores, with potential implications for fish–zooplankton interactions still largely
unknown. Here we studied diel variation in the vertical and horizontal distribution of the
crustacean plankton in Lake Vela, a shallow polymictic and eutrophic lake. Diel sampling
was carried out on three consecutive days along a horizontal transect, including an open-
water station and a macrophyte (Nymphaea alba) bed. Since transparency is a key
determinant of the predation risk posed by fish, the zooplankton sampling campaigns
were conducted in both the turbid (autumn) and clear water (spring) phases.
2. In the turbid phase, most taxa were homogeneously distributed along the vertical and
horizontal axes in the three consecutive days. The only exception was for copepod nauplii,
which showed vertical heterogeneity, possibly as a response to invertebrate predators.
3. In the clear water phase, most zooplankton taxa displayed habitat selection. Vertically,
the general response consisted of a daily vertical migration (DVM), despite the limited
depth (1.6 m). Horizontally, zooplankters showed an overall preference for the pelagic
zone, independent of the time of the day. Such evidence is contrary to the postulated role
of macrophytes as an anti-predator refuge for the zooplankton.
4. These vertical (DVM) and horizontal (macrophyte-avoidance) patterns were particularly
conspicuous for large Daphnia, suggesting that predation risk from size-selective predators
(fish) was the main factor behind the spatial heterogeneity of zooplankton in the spring.
Thus, the difference in the zooplankton spatial distribution pattern and habitat selection
among seasons (turbid and clear water phases) seems to be mediated the predation risk
from fish, which is directly related to water transparency.
5. The zooplankton in Lake Vela have anti-predator behaviour that minimises predation
from fish. We hypothesise that, due to the distinct fish community of shallow
Mediterranean lakes, aquatic macrophytes may not provide adequate refuge to zoo-
plankters, as seen in northern temperate lakes.
Keywords: alternative states, freshwater zooplankton, migration, patchiness, spatial distribution
Introduction
The distribution of aquatic organisms, and partic-
ularly plankton, has long been known to be hetero-
geneous. Spatial heterogeneity is a common feature in
all ecosystems and is the result of many interacting
physical and biological processes (Pinel-Alloul, 1995;
Folt & Burns, 1999; Hu¨lsmann et al., 1999; Thackeray
et al., 2004). Physical processes, such as wind-induced
turbulence or currents (Thackeray et al., 2004), seem to
be the main drivers of large-scale heterogeneity. As
the spatial scale decreases, the importance of biologi-
cal processes increases and may overlap that of the
physical processes (Folt & Burns, 1999). The formation
Correspondence: Bruno B. Castro, Departamento de Biologia da
Universidade de Aveiro, Campus Universita´rio de Santiago,
3810-193 Aveiro, Portugal. E-mail: brunocastro@bio.ua.pt
Freshwater Biology (2007) 52, 421–433 doi:10.1111/j.1365-2427.2006.01717.x
 2007 The Authors, Journal compilation  2007 Blackwell Publishing Ltd 421

of swarms or aggregations of plankton is considered
to be a consequence of the search for food (Jensen,
Larsson & Ho¨gstedt, 2001; Roozen & Lu¨rling, 2001)
and mates (Folt & Burns, 1999), and to predator-
avoidance (Kvam & Kleiven, 1995; Roozen & Lu¨rling,
2001). Predatory pressure, in particular, has received
considerable attention as a major driving force of
zooplankton patchiness. Examples of predator-in-
duced aggregations, resulting in spatial patterns of
freshwater zooplankton, include avoidance of the
shore (Romare & Hansson, 2003; Pinel-Alloul, Me´thot
& Malinsky-Rushansky, 2004), diel vertical migration
(DVM) (Gilbert & Hampton, 2001; Han & Strasˇkraba,
2001; Hembre & Megard, 2003; Ringelberg & Van
Gool, 2003) and diel horizontal migration (DHM)
(Kvam & Kleiven, 1995; Lauridsen & Buenk, 1996;
Burks et al., 2002; Wojtal et al., 2003).
Diel vertical migration is one of the best-studied
antipredator defences of zooplankton. The most usual
pattern of DVM is for a population of zooplankton to
dwell in deeper waters during the day and then to rise
towards the warm surface waters for a few hours at
night, thus avoiding visually-oriented predators dur-
ing the day (Dodson, 1990; Han & Strasˇkraba, 2001).
This behavioural pattern is triggered by light changes at
dusk and dawn (Ringelberg, 1999; Ringelberg & Van
Gool, 2003), and is related to light penetration of the
water column (Dodson, 1990; Alonso et al., 2004). The
phototactic reactions leading to DVM are strongly
enhanced by predator signals, particularly fish kairo-
mones (Loose & Dawidowicz, 1994; Ringelberg, 1999;
Ringelberg & Van Gool, 2003). In shallow lakes,
however, there may be no hypolimnetic refuge avail-
able (Burks et al., 2002). Han & Strasˇkraba (2001)
suggested that aggregation near the sediment during
the day may be used as a defence strategy in shallow
lakes, as confirmed in the field by Jeppesen et al. (2002).
Numerous studies have also shown that large zoo-
plankters often aggregate in vegetation stands, swim-
ming out only at night to the adjacent open water
(Kvam & Kleiven, 1995; Lauridsen & Buenk, 1996;
Masson et al., 2001; Wojtal et al., 2003). This phenom-
enon – DHM (Burks et al., 2002) – depends on the
effectiveness of macrophytes as a refuge, which seems
to result from the balance between predator pressure
and macrophyte density (Schriver et al., 1995; Jeppesen
et al., 1998; Perrow et al., 1999; Burks et al., 2002). In
addition, macrophyte stands also represent a potential
hazard for the zooplankton, which can suffer severe
predation losses to macrophyte-associated biota (Hu¨ls-
mann et al., 1999; Perrow et al., 1999; Masson et al.,
2001). This is particularly important in the case where
planktivorous fish (especially juveniles) also use macr-
ophytes as a refuge from piscivores (Burks et al., 2002;
Romare et al., 2003; Romare & Hansson, 2003).
Predation gradients are usually coupled with
marked physical gradients, both vertically (epilim-
nion/hypolimnion) and horizontally (near-shore/off-
shore). In shallow polymictic lakes, however, such
gradients may be less well defined, due to the poorly
marked separation between littoral and pelagic zones
and to the constant mixing of the water column.
Furthermore, in shallow Mediterranean lakes, the fish
community is dominated by omnivorous fish (Blanco
et al., 2003) that prey heavily on zooplankton, inclu-
ding in littoral areas (Garcı´a-Berthou, 1999; Garcı´a-
Berthou & Moreno-Amich, 2000), thus precluding the
littoral refuge effect of aquatic vegetation. Therefore,
the spatial heterogeneity of zooplankton in such lakes
may differ from what has been observed in northern
temperate lakes, where most of the studies on DVM
and DHM have been conducted.
Given the distinctive characteristics of shallow
Mediterranean lakes, we conducted a study on the
spatial distribution patterns of the zooplankton from
Lake Vela (Portugal) under contrasting conditions
(turbid vs. clear water phase). The main aim was to
assess if zooplankters exhibited antipredator beha-
viour, such as habitat selection and/or diel migration
pattern, as a response to predation pressure from fish.
More specifically, we intended: (i) to determine
whether DVM is constrained in this shallow and
polymictic lake; (ii) to assess if macrophytes provide
refuge for zooplankters against littoral-foraging pred-
ators. To do so, we performed 3-day diel sampling
cycles at littoral (macrophyte bed) and pelagic samp-
ling sites at two depths (near the surface and near the
bottom). The study was conducted during the turbid
and the clear water phases, since these two alternative
states represent contrasting conditions in terms of
water transparency and fish predation pressure.
Methods
Site description
Lake Vela is a small (maximum floodable area 70 ha),
shallow (0.9 m average depth; 2.4 m maximum depth)
422 B.B. Castro et al.
 2007 The Authors, Journal compilation  2007 Blackwell Publishing Ltd, Freshwater Biology, 52, 421–433