Indirect enhancement of large zooplankton by …investigadores.uncoma.edu.ar/Lab_Limnologia/publicaciones/Gonzalez... · brates as predators of zooplankton (ARN~R et al. 1998, HIROVEN

Arch. Hvdrobiol. 158 4 551-574 Stuttgart, December 2003

Indirect enhancement of large zooplankton by consumption of predacious macro i nverte b rat es by littoral fish

Maria de 10s Angeles Gonzalez Sagrario’ * and Esteban Balseiro2 * Universidad Nacional de Mar del Plata, Argentina

With 6 figures and 4 tables

Abstract: Food webs in the littoral zones of shallow lakes are inherently complex. Submerged macrophytes are considered refuge areas although they host potential predators for zooplankton. During spring : summer of 2000/2001, we carried out two mesocosm experiments in Los Padres Lake, a shallow macrophyte dominated lake of the Argentine Pampa plain. We investigated the effect of littoral fish predation on zooplankton and the role of submerged macrophytes and benthic macroinvertebrates in the zooplankton-fish relationship. Treatments differed in macrophyte cover (0-65 % PVI) and fish presence. In both experiments, we determined zooplankton abundance and body size distribution, as well as macroinvertebrate abundance. The addition of the small littoral fish Astyanax eigenmanniorum (Characidae, Tetragonopterinae) to the enclosures had either a positive or a negative effect on large zooplankton, especially on calanoid copepods, depending on macroinvertebrate densities and availability. Fish did not impact on small size Cladocera. Instead, littoral fish preferred macroinvertebrates to zooplankton, and thus, the impact of fish predation on plankton community depended largely on the macroinvertebrate abundance. Therefore, our results suggest that macrophyte-fish complex could enhance or depress zooplankton abundance and that the effect is related with predatory macroinvertebrates associated with vegetation.

Key words: enclosure experiment, benthic macroinvertebrates, macrophyte refuge, invertebrate predation.

Authors’ address: Universidad Nacional de Mar del Plata, Dpto. Biologia, CC

Universidad Nacional del Comahue, Centro Regional Universitario Bariloche, 1245, (7600) Mar del Plata, Argentina; E-mail: [email protected].

(8400) San Carlos de Bariloche, Argentina. * CONICET (Consejo Nacional de Investigaciones Cientificas y Técnicas).

DOI: 10.1 127/0003-9136/2003/0158-0551 0003-9136/03/0158-0551 $6.00 0 2003 E. Schweizerbart’sche Verlagsbuchhandlung, D-70176 Stuttgart

552 Maria de 10s Angeles Gonzalez Sagrario and Esteban Balseiro

Introduction

Predation is one of the key factors governing patterns in natural ecosystems (SIH et al. 1998). Fish predation is a size-selective process, and zooplankton accordingly display alternative strategies and behaviours to avoid it, including reduction of body size (HAMBRIGHT 1994, BERTOLO et al. 1999) and pigmenta- tion (ZARET & KERFOOT 1975). In shallow lakes, die1 horizontal migration (DHM) from pelagic to littoral zones may serve as an effective mechanism against fish predation (LAURIDSEN & BUENK 1996, LAURIDSEN et al. 1996, LAURIDSEN et al. 1998, STANSFIELD et al. 1997). For example in eutrophic Danish lakes, both large and small cladocerans (like Duphnia, Cerioduphniu and Bosmina) migrate during day-time to submerged macrophyte stands (LAURIDSEN & BUENK 1996, LAURIDSEN et al. 1996, LAURIDSEN et al. 1998).

Submerged macrophytes are considered refuge areas against fish predation due to their structural complexity (CARPENTER & LODGE 1986, BURKS et al. 2001). In addition, they host a rich epiphytic fauna (KORNIJ~W & KAIRESALO 1994), including macroinvertebrate predators (SIH 1987, JOHNSON 1991, PA- TERSON 1994), young of the year (YOY) fish and littoral fish (WHITESIDE

1988, DIEHL & EKLOV 1995, HALL & RUDSTAM 1999). Besides searching for their own refuge in littoral areas, YOY and littoral fish also forage among macrophytes and therefore constitute a risk for zooplankton, particularly for large individuals (BURKS et al. 2002). Macrophytes thus mediate multiple processes like predation, cannibalism and competition. Whether macrophytes constitute an anti-predator refuge or potential risk areas likely depends on the local composition of the littoral food web.

Macroinvertebrate predators have been overlooked as key organisms in littoral zones. Macroinvertebrate predators have a central role in the littoral food web because they prey on smaller organisms, like cladocerans, but also serve as food for fish and in this way mediate the cascading effect to lower levels. Several laboratory or field studies evidenced the role of benthic macroinvertebrates as predators of zooplankton ( A R N ~ R et al. 1998, HIROVEN 1999, ZIM- MER et al. 2001), but most of the field experiments concerning predation in the littoral zone were focused on zooplankton-fish or macroinvertebrate-fish interactions (MITTELBACH 1988, Moss et al. 1998, DIEHL & KORNIJ~W 1998). Un- der this scenario, the study of JOHNSON et al. (1996) constituted a rare example, as they focus on the strength of the effect of fish and dragonfly on benthic macroinvertebrates in the littoral food web. However, studies like this, considering the zooplankton-macroinvertebrate-fish interaction strength, still need to be considered in detail.

The Pampa plain of Argentina contains a series of poorly studied shallow lakes heavily vegetated with different types of macrophytes. Fish communities are extremely rich (RINGUELET 1975), among this assemblage Astyunux eigen-

Indirect enhancement of large zooplankton 553

manniorum (COPE) (Characidae, Tetragonopterinae) is considered an abundant small littoral fish that could be included into the zoobenthivorous niche as macroinvertebrates are, mainly, part of its diet (GROSMAN et al. 1996, VILELLA

et al. 2002). In this study, we investigate the trophic interactions in the littoral zone of a

typical shaIlow, macrophyte-dominated lake of the Pampa plain, Los Padres Lake. Our objectives included: 1) to test whether predation by the fish A. eigenmanniorum impacted zooplankton abundance and size structure, 2) to examine the effect of submerged macrophytes on A. eigenmanniorum predation on zooplankton, and 3) to examine the role of macroinvertebrates in relation to zooplankton and fish predation.

Study area

Los Padres Lake is located in the Pampa plain of the Buenos Aires Province (Argen- tina) at 37" 56's and 57" 44'W. This shallow lake (Area = 2 km2; Mean Depth = 1.8 m) has a polymictic thermal regime and alkaline waters (pH = 8.6). The transparency is variable with turbid and clear periods alternating from year to year (Secchi disk depth ranged from 40 to 90cm during 1998/2002). Although no summer mean Chlorophyll-a and TP concentrations are available, reported data showed that both parameters varied inter-annually. Values ranged from 20 up to 9OpgL for Chl-a and from 100 up to 400vg/L for TP during 1999-2001 (M. A. G O N Z ~ L E Z SAGRARIO, unpubl. data). These values allow us to consider this lake as eutrophic.

In the littoral zone of Los Padres Lake, Schoenoplectus californicus (MEYER) STEUD. constitute an outer ring around the whole lake while in the inner part, different species of submerged macrophytes dominate, being Potarnogeton pectinatus RUIZ and P A V ~ N and Ceratophyllurn dernersurn L. the dominant species. Vegetated areas host small fish species, birds and a rich macroinvertebrate assemblage. Jenynsia lineata (JENYNS), Cnesterodon decenrnaculatus (JENYNS), A. eigenmanniorum and juvenile Cichlasorna faceturn (JENYNS) constitute the littoral fish assemblage that are forming schools and swimming among the edge and inner part of the macrophyte stands or like C. fasceturn which is nesting there. Fish are generally omnivorous preying on zooplankton and macroinvertebrates. Coots (Fulica leucopteru VIEILLOT) and black- necked swans (Cygnus rnelancoryphus MOLINA) are the mainly vegetarian and more abundant birds in the lake. Ducks (Anas versicolor VIEILLOT and A. cyanoptera VIEIL- LOT) are also present and could impact on aquatic vegetation as well as on macroinvertebrates, while Podiceps rolland QUOY and GAIMARD (white-tufted grebe) feed mostly on fish and macroinvertebrates and is also abundant in the littoral zone (DEL HOYO 1987, CANEVARI et al. 1991). Macroinvertebrate assemblage is very rich and comprise water mites (mainly Pionidae), Palaemonetes argentinus (NOBILI) (grass shrimp) and Zygoptera larvae (Coenagrionidae) as the most abundant predators (GONZ~LEZ SAG- RARIO, unpubl. data). Littoral rotifers (mainly Euchlanis), cladoceran (like Alona and Chydorus), and ciclopoid copepods dominate the microinvertebrate assemblage of the


littoral zone, but pelagic cladoceran species like Ceriodaphnia, Moina or Bosmina are also present in C. demersum stands (GONZALEZ SAGRARIO, unpubl. data).

In open waters, the zooplankton community is mainly composed by medium to small sized cladocerans like Bosmina (Neobosmina) huaronensis DELACHAUX, Cerio- daphnia dubia RICHARD, Moina micrura KURZ, and Diaphanosoma brachyurum (LIE- VIN). During short periods the large cladoceran species Daphnia (Ctenodaphnia) spinulata B I R A B ~ N also occurs. Two species of calanoid copepods are present: Notodiap- tomus incompositus (BRIAN) and Boeckella bergi RICHARD, with the former being dominant. Acanthocyclops robustus (SARS) dominates the cyclopoid copepod assemblage.

Materials and methods

Experimental setup and sampling

During spring 2000 and summer 2001, we performed two mesocosm experiments in an area of 1.30m in depth covered by C. demersum. Each treatment occurred in closed enclosures of plexiglass (100 pm in thickness; 1 m in diameter and 2 m in height) (N = 3). Enclosures had hoops at each end, so they were open both to the atmosphere and the bottom sediment and were suspended 40cm above the water surface from a steel frame weighed with bricks.

We conducted the first experiment between 28 November and 6 December 2000 with two treatments: without fish (treatment 1) and with 5 fish per enclosure (treatment 2), reaching a density of 6.35 fish/m2. All the enclosures contained macrophytes (C. demersum) with a percent of volume infested (PVI) = 65 and natural lake zooplankton.

We then conducted a second experiment in the end of summer, between 12 and 26 March 2001, with a higher fish density and three treatments. In the fishless treatment (TI), we filled enclosures with lake water and natural lake zooplankton. In the fish treatment (T2), 10 fish per enclosure (equivalent to 12.74 fish/m2) occurred with lake water and natural lake zooplankton. In the third treatment (macrophyte+fish) (T3), we placed 10 fish per enclosure (equivalent to 12.74 fish/m2) with a submerged macrophyte (C. demersum) and natural lake zooplankton. We placed the plexiglass enclosures in an area free of macrophytes.

In both experiments, we used A. eigenmanniorum, a littoral small fish of 4 to 5 cm in length. Enough C. demersum was added to the enclosures to achieve 65 PVI, corre- sponding to 558.79g of dry biomass/m2. The PVI performed in the enclosures ex- ceeded the threshold of 15-20% reported by SCHRIVER et al. (1995) for macrophytes to bring a refuge effect to zooplankton, and resembles natural lake conditions. We washed the plants and rinsed off macroinvertebrates in experiment 2 to avoid a difference among enclosures withlwithout (bare sediment) vegetation. At the beginning of both experiments, we enhanced zooplankton concentration in the enclosures by adding zooplankton collected from the lake with a plankton net of 65 pm mesh size in a relation of 2 x in respect to the background conditions. Unfortunately, these methods failed because initial zooplankton concentrations resemble to the one in the lake.


We sampled each enclosure and the lake every two or three days for the complete duration of the experiments, except for the second experiment where the last two measurements occurred after 4 and 6 days. We assessed initial conditions through sampling in day 0 and testing statistical differences. We collected composite samples of 12 L with a 6 L Schindler Patalas trap at 0.5 m and 1 m depth, and collected 1 L for chemical analysis and filtered the remaining volume through a 65 pm mesh net for zooplankton quantification. We sampled macroinvertebrates qualitatively, dragging a net of 500pm mesh size from the bottom to the water surface (and through macrophytes when present, knocking the net against them) along each enclosure covering a total volume of 20 L. We invested the same amount of time in each sampling. We pre- served all samples in 4 % formaldehyde solution.

Laboratory methods

We quantified zooplankton in a Bogorov chamber of 5 ml, and macroinvertebrates under a stereomicroscope. We subsampled zooplankton if necessary and counted all macroinvertebrates present in the sample. Also, we measured crustacean body length on 20 randomly-selected individuals per species and per sample. We categorized sepa- rately calanoid copepods in copepodites (I, 11, 111, IV+V) and adults (male and female).

We determined total phosphorus (TP) and total dissolved phosphorus (TDP) respectively on 100ml of unfiltered and GF/C filtered lake water. We digested TP and TDP samples with potassium persulphate at 125 "C and 1.5 atm for 1 h, and determined phosphorus concentration by the ascorbate-reduced molybdenum blue method (APHA 1989). We measured phytoplankton biomass as Chlorophyll-a concentration (Chl-a), filtering a volume of 150-400 ml through a Whatman GF/C filter and extracting the pigment in 90 % acetone. We calculated Chl-a concentration according to the mono- chromatic method of LORENZEN (1967) following absorbance readings at 665 and 750nm. For Chl-a and TDP determination we used Whatman GF/C filters with 1.2pm a pore size of. We also measured Chl-a, TP and TDP in the lake as a control of our experiment.

Statistical methods

We tested differences on initial conditions on data for zooplankton and macroinvertebrate abundances and Chl-a, TP and TDP concentrations using t-test (experiment 1) or One-way ANOVA (experiment 2). Thereafter, we analysed treatment effect using a two-way RM-ANOVA (Reapeted Measures ANOVA) considering treatments as the factor and time as the repeated measures. For post hoc comparisons we performed the Newman-Keuls test (UNDERWOOD 1997). To compare correlation among prey abundance and macroinvertebrate predators we performed Pearson correlations. For all tests, we fixed the level of significance at 0.05. We previously tested normality and ho- mogeneity of variance and when data did not meet these assumptions we performed log 10 or square root transformations depending the on case.

We also estimated the net effect of predation (NE) by determining the net effect of each predator treatment on each prey taxa following JOHNSON et al. (1996). Net effects


are the percentage by which each prey densities in treatments without predator change in predator treatments, NE = 100 * (Y-y)/y, where Y is the prey abundance in the predator treatment and y the prey abundance in the predator free treatment. We charac- terized net effects as weak (INEI <25 %), moderate (25 % I lNEl150 %) and strong (lNEl> 50 %). We calculated NE using untransformed data from all enclosures in each treatment and for all sampling dates. According to JOHNSON et al. (1996), we associated significance of NE with statistically significant treatment effect in RM-ANOVA, maintaining an experimentwise error rate of a=0.05.

Results

Experiment 1

Initial zooplankton abundances on day zero were not significantly different between treatments (t-test: N. incompositus: females: t = 0.08, V+IV stages t = 1.17, and I1 copepodite stage: t = 0.86; B. huuronensis: t = 1.8, as some examples, df = 4 and P >0.05 for all cases). Neither were Chl-a concentration (t- test: t = 1.88, df = 4, P >0.05) nor Secchi disc depth (t-test: t = 7, df = 4, P > 0.05).

At the beginning of the experiment, C. dubiu and B. huuronensis dominated the zooplankton community, reaching abundances higher than 300 and 100 ind./L, respectively. Their populations decreased significantly towards the end of the experiment in both treatments (Fig. 1, Tables 1 and 3), independent of fish presence. C. dubiu declined in the enclosures correlated with the decrease in C. dubiu lake population (Pearson’s correlation: 1-2 = 0.85, F3.3 = 24, Pc0.07). In addition, no clear fish predation effect occurred for cladoceran populations, not even for Duphniu spinuluta (Fig. 1, Tables 1 and 3). More- over, no significant correlation occurred for mean D. spinulutu abundance and mean abundance of other predators found in the enclosures like water mites (Pearson’s correlation: 1-2 = 0.17, F8.8 = 2.4, P >0.05) or damselfly (Pearson’s correlation: 1-2 = 0.02, F8.8 = 1.4, P >0.05).

We detected a significant fish effect on N. incompositus population (Table 3), finding higher abundances of some stages of this calanoid copepod in the fish treatment than in fishless enclosures (Fig. 2). In particular, a significant difference between treatments occurred for females and the copepodites stages IV+V and I1 (Fig. 2, Tables 1 and 3). We found adults and cyclopoid copepodites evenly distributed between treatments (Fig. 2, Tables 1 and 3). We recorded no significant differences in cladoceran body size or in copepod age distribution between treatments for any of the sampling dates (some examples for the last measurement: B. huuronensis: t = 0.34, df = 19; C. dubiu: t = 0.69, df = 16; D. spinuluta: t = 1.45, df = 21; N. incompositus: female: t = 0.45, df = 29; male: t = 1.26, df = 28; V+IV copepodite stage: t = 0.16, df = 18; P>0.05 in all the cases).


40

30

20

10

0

ci > g 450 G

P) o 300 G cb -0 c 3 150

._ W

2 0

150

100

50

0

Daphnia spinulata T

Ceriodaphnia dubia T

Bosmina huaronensis T '

Fishless Treatment (Tl) - Fish Treatment (T2)

Fig. 1. Cladoceran species mean abundances in Experiment 1 in the fish and fishless treatments. Error bars represent standard error.

Net fish effect (NE) on zooplankton was positive in the majority of the cases. N. incompositus females, copepodite stages V+IV and I1 experienced a strong to moderate positive significant net fish effect (NE: 57.3, 46.3 and 38.04 %, respectively). For the rest of the calanoid sized fractions we found moderate non significant net fish effect and in the particular case of D. spinu- Zutu, NE accounted 60 %, but based on RM-ANOVA this effect was not significant.


Table 1. Mean values and standard error of zooplankton and macroinvertebrate abundances recorded in experiment 1 and 2. Experiment 1: T1: fishless treatment, and T2: fish treatment. Experiment 2: T1: fishless treatment, T2: fish treatment and T3: macrophyte + fish treatment. N = 3 in both experiments. Abundances are expressed in indi- vidualk. nd means not determined and -, not recorded in that treatment.

Experiment 1 Experiment 2

TI T2 TI T2 T3 Mean f St. ermr Mean f St. error Mean f St. error Mean f St. error Mean f St. error

Zooplankton D. spinulata C. dubia D. brachyurum M. micrura B. huaronensis N. incampositus female N. incompositus male Calanoid Copepcdites V + IV Calanoid Copepodites 111 Calanoid Copepcdites I1 Calanoid Copepcdites I Adult Cyclopoid Copepods Cyclopoid Copepodites

Macroinvertebrates Zygoptera larvae

P. argentinus Acarina (Pionidae) Total Macroinvertebrate

Predators Chironomidae Oligwhaeta (Naididae) Ephemeroptera

(Oniscigastridae) Ostracoda Amphipoda Total Non Predatory

(Coenagrionidae)

Macroinvertebrates

6.0f2.0 90.4f3 1.0 - -

69.2f24.7 4.5f1.0 4.08f 1 .OO 11.7f2.1 6.7f1.0 3.1 f0.5 1.7f0.5 3.0f1.7 16.1f4.1

0.006f0.01

-

0.044fO.02 0.05f0.03

0.52f0.15 0.056f0.03 -

0.031 f0.02 0.156f0.05 0.76f0.25

9.0f3.0 107.6f48.2 -

- 50.0f 12.0 7.0f1.3 6.2f1.7 17.0f3.0 9.3f1.8 5.2f0.8 3.7f0.7 1.0f0.3 16.7f6.3

0.035 fO.0 I

-

0.035f0.01 0.07f0.02

0.37f0.13 0.023f0.01 -

0.017f0.01 0.151 f0.06 0.56f0.21

-

5.08f0.73 37.3f11.4 12.6f3.3 - 4.8fl.l 2.7f1.4 12.5 f 1.9 9.2f1.2 7.7f0.9 7.0f1.3 6.2f0.8 45.1f6.5

-

-

-

0.OOf0.00

0.012f0.01 -

-

O.O1fO.O1 0.01f0.01 0.03f0.03

-

4.8f1.0 42.18f 16.0 26.0f6.7 -

2.8f0.8 2.3f0.5 15.7f2.8 9.7f1.8 7.7f1.0 6.0f0.9 6.4f1.1 48.9f6.3

0.01 fO.01

0.02f0.01 -

0.04+0.02

0.023f0.01 0.01 fO.O1 0.01 fO.O1

-

0.01 fO.01 0.08f0.04

-

4.0fl.l 54.3f32.1 16.6f4.3 - 3.8f1.6 2.9f0.7 8.2f1.6 7.7f2.3 6.0f1.4 5.1 f 1.2 6.9f1.4 71.1f18.2

0.04f0.02

0.03f0.02 - 0.07f0.04

0.75f0.03 0.02f0.01 0.02f0.02

0.01 fO.01 -

0.2f0.07

We found macroinvertebrates dominated by chironomid larvae in both treatments. However, several predators such as Zygoptera larvae and acari (Pionidae) also occurred in the assemblage. Oligochaete abundance (repre- sented by Pristina and Chaetogater) decreased significantly in presence of A. eigenmanniorum. Other macroinvertebrate taxa like Zygoptera and chironomid larvae showed a significant interaction between fish presence and time, meaning that fish effect was depending on time and furthermore, that the increase in time exposure increases prey mortality. For Zygoptera and chironomid larvae, abundances in the fish treatment decayed after the first day and thereafter remained low (Fig. 3, Tables 1 and 3). We did not find a significant

Indirect enhancement of larae zoodankton 559

Table 2. Mean values and standard error of physical-chemical parameters recorded in experiment 1 and 2. Experiment 1: TI: fishless treatment, and T2: fish treatment. Ex- periment 2: T1: fishless treatment, T2: fish treatment and T3: macrophyte + fish treatment. N = 3 in both experiments. nd means not determined.

Experiment 1 Experiment 2

T1 T2 T1 T2 T3 Mean f St. error Mean f St. error Mean f St. error Mean f St. error Mean f St. error

Chlorophyll-a (F~/L) 19.5k3.3 19.6k3.0 35.0k2.0 34.0f2.0 45.0f6.0 Secchi Depth (cm) 44.0f5.0 44.0f5.0 56.0*5.0 63.3 f4.4 74.3k7.1 Total Phosphorus (Fg/L) nd nd 117.Of6.1 99.0f7.7 123.1 f 13.2 Total Dissolved nd nd 53.6k 10.0 49.0f6.3 50.6f7.6

Phosphorus (pg/L)

Table 3. Results of the two-way repeated measures analyses of variance from the mesocosm experiment. * P-value <0.05; ** P-value<0.01; *** P-value <0.001; ns = not significant.

Effects off

Treatment Time Interaction

F1.4 P-value F4.16

Zooplankton D. spinulata 0.8 ns 3.7 C. dubia 0.2 ns 24.2 B. huaronensis 1.6 ns 17.6 N. incompositus female 14.7 ** 3.8 N. incompositus male 3.0 ns 14.9 Calanoid Copepodites V + IV 8.6 1.4

Calanoid Copepodites I1 14.5 1.8 Calanoid Copepodites I 1.8 ns 0.6 Adult Cyclopoid Copepods 1.9 ns 3.7 Cyclopoid Copepndi tes 0.0 ns 2.6

* Calanoid Copepodites I11 4.1 ns 1.1 *

P-value F4.16 P-value

ns 1 .o ns 0.8 ns 2.88 ns 0.5 ns 2.4 ns

ns 2.2 ns ns 0.7 ns ns 0.3 ns ns 1.6 ns

1.8 ns ns 0.5 ns

*** *** * ***

*

Macroinvertebrates Zygoptera larvae 6.3 ns 2.9 ns 4.4 Acari 3.3 ns 6.3 ** 2.0 ns Chironomidae 0.7 ns 8.2 *** . 3.2 Oligochaeta 100 ** 23.5 *** 10.8 **

*

*

Amphipoda 0.0 ns 3.2 ns 0.6 ns Ostracoda 0 ns 1.1 ns 2.2 ns Physical-chemical Parameters Chlorophyll-a (pg/L) 0.0 ns 9.3 ns 1 .o ns Secchi Depth (cm) 0.0 ns 68.4 ** 4.6 *


Notodiaptomus incompositus 12

9

6

3

0

2 30 2

g 20

.4 W

cd -0 E:

P 1 10 6

0

15

10

5

0

Copepodites I I

Copepodites 111 I

I I Adult Males

Cyclopoid copepodites

._ " 4 5 4 I W V

5 30

15

0 6

Copepodites I1 'IP

TCopepodites IV + V

Adult Females

T T 10

5

0

- Fishless Treatment (Tl) - Fish Treatment (T2)

Fig. 2. Calanoid (N. incompositus) and cyclopoid copepod mean abundances in Exper- iment 1 in the fish and fishless treatments. Error bars represent standard error.


2.5

2.0

1.5

1 .o

.g 0.5 2

: 0.0 : 3 2.0

W

0

4 1.5

1 .o

0.5

0.0

Fishless Treatment (T 1)

Fish Treatment (T2)

0 Zygoptera rzzzzza Hydracarina

Oligochaeta - Ostracoda mza~ Amphipoda E Chironomidae

Fig. 3. Macroinvertebrate mean abundances in Experiment 1 in the fish and fishless treatments.

fish effect on acari abundance, nevertheless mean overall abundance was lower under fish treatment (Table 1) in which N. incompositus female and V +IV copepodite abundances enhanced significantly (Table 3, Fig. 2). More- over, we found a significant negative correlation between the abundance of female and copepodite stages V+IV of N. incompositus and mean acari abundance (Pearson’s correlation for: female: i‘- = 0.4, F ~ . ~ = 4.26, P = 0.05; for V+IV copepodite stage: ? = 0.4, F8.8 = 4.41, P<0.05).

Net fish effect was negative for most macroinvertebrates. Fish produced a strong negative significant NE in oligochaetes (NE = -61.5 %), a moderate non-significant effect for chironomids (NE = -30.05 %) and a weak effect for acari (NE = -21.3 %).

We recorded an increase of Microcystis aeruginosa KUTZING and Botryo- coccus braunii KUTZING during the experiment and these species accounted for the major fraction of the phytoplankton biomass and assemblage. No difference in chlorophyll-a or Secchi depth occurred between treatments (Tables 2 and 3).

562 Maria de 10s hgeles Gonzalez Sagrario and Esteban Balseiro

Experiment 2

Initial experimental cladoceran and calanoid concentrations, tested by One- way ANOVA, revealed no significant difference between treatments on day 0 (e.g.: D. brachyurum: F2.6 = 0.6, M. micrura: F2.6 = 0.6, N. incompositus: F2.6

= 0.2, in all cases P >0.05). The cladoceran species D. brachyurum and M. micrura dominated the zoo-

plankton community, reaching 100 and 60 ind./L, respectively at the beginning

250

200

150

100

50

0

rl \ 120

90 Q) 0

5 60 a G 3 30 6

0

h

a

15

10

5

0

Diaphanosoma brachyurum

T

Moina micrura T

Ceriodaphnia dubia

- Fishless Treatment (Tl) - Fish Treatment (T2) - Macrophytes + Fish Treatment (T3)

Fig.4. Cladoceran species mean abundances in Experiment 2 in the fishless, fish and macrophytes + fish treatment. Error bars represent standard error.


Table 4. Results of the two way repeated measures analyses of variance from the mesocosm experiment 2. Treatment comparisons assessed by Newman-Keuls test, only significantly different treatments are displayed. * P-value < 0.05; ** P-value < 0.01; *** P-value <0.001; ns = not significant. T1: fishless treatment, T2: fish treatment, T3: macrophyte + fish treatment. ': only case where comparison was made between T2 and T3, df of treatment: 1,4; time: 4, 16; interaction: 4, 16.

_.Y

Zooplankton D. brachyurum 2.2 ns C. dubia 0.6 ns M. micrura 1.2 ns N. incompositus female 2.1 ns N. incompositus male 0.8 ns Calanoid Copepodites V + IV 5.2 * Calanoid Copepodites 111 3.4 ns Calanoid Copepodites I1 0.8 ns Calanoid Copepodites I 0.5 ns Adult Cyclopoid Copepods 0.9 ns Cyclopoid Copepodites 2.1 ns

Macroinvertebrates Non Predatory 6.4 *

Zygoptera larvae' 12.5 * Physical-chemical Parameters

Secchi Depth (cm) 9.1 * Total Phosphorus (pg/L) 4.6 ns Total Dissolved 9.6 *

Macroinvertebrates

Chlorophyll-a (pg/L) 53.5 ***

Phosphorus (pg/L)

Effects off Treatment

Treatment Time Interaction comparisons

F?A P-value F4.24 P-value Fs,24 P-value

17.2 *** 2.5 ns 8.2 *** 1.8 ns 1.4 ns

12.7 *** 8.7 *** 1.6 Ns 2.2 Ns 8.5 ***

20.2 ***

2.1 Ns

0.9 Ns

92.5 *** 27.0 *** 18.8 *** 20.8 ***

1.0 Ns 2.3 Ns 0.8 Ns 0.6 Ns 1.1 Ns

0.9 ns 1.0 ns 0.5 ns 3.4 ** 1.2 ns

3.2 * T2 - T3

0.6 ns Tl-T3

0.8 ns T2-T3

27.9 *** Tl-T3/T2-T3 11.4 *** Tl-T3/T2-T3 12.1 *** 8.6 *** Tl-T2/Tl-T3

of the experiment (Fig. 4). We also found C, dubia, but in a lower abundance in respect to the other cladoceran species present in this experiment (Fig. 4).

We did not detect a clear fish predation effect for cladoceran populations (Tables 1 and 4). D. brachyurum exhibited significantly lower densities under fish presence compared to fishless treatment only on 14 March (i.e. after fish introduction in the enclosures) (One-way ANOVA: F2.6 = 13.0, P<O.Ol), but not afterwards (Fig. 4, Table 4). D. brackyurum and M. micrura populations decreased significantly over time in the experiment (Table 4, Fig. 4). The temporal dramatic decrease in D. brackyurum and M. micrura densities occurred after the first 2 days of the experiment, as Newman-Keuls test reported no significant differences after these sampling dates (P > 0.05). Moreover, the decline of both cladocerans correlated significantly with lake population decline

564 Maria de 10s hgeles Gonzalez Sagrario and Esteban Balseiro

Notodiaptomus incompositus

3-

20

15

10

5

0 ' a 40 C

30 a, u

"a

.-

5 20

P 5 10 d

0

12

9

6

3

0

Cyclopoid Copepodites t&

Copepodites I I T -

Copepodites 111

T

I Adult Males

1 I

Copepodites I1 7 t: 10

5

0 Copepodites IV + V

Er. 30

0 0

J 2o G

10 p- i? v n

I " Adult Females

T T t l 2

2 90

u 60

"a 5 30 P d

0

.d W

a,

5

- Fishless Treatment (Tl) I-Lzza Fish Treatment (T2) fisi!~~ Macrophytes + Fish Treatment (T3)

Fig. 5. Calanoid (N. incompositus) and cyclopoid copepod mean abundances in Exper- iment 2 in the fishless, fish and macrophytes+fish treatment. Error bars represent standard error.


0.4

0.3

0.2

0.1

n cl -.. 2 0.0 ._ W

P) 0

a c f P

P)

0.3

0.2 + e 2 0.1 > c .- 2 u 0.0

5 0.3

0.2

0.1

0.0

(Tl)

0 Zygoptera Ostracoda

EC~%%%! Amphipoda Palaemonetes

E Chironomidae D Ephemeroptera

Oligochaeta

Fish Treatment (T2)

L

MacroDhvtes + Fish Treatment

Fig. 6. Macroinvertebrate mean abundances in Experiment 2 in the fishless, fish and macrophytes + fish treatment.

(Pearson's correlation for: D. bruchyurum: 8 = 0.96, F3.3 = 79, P <0.01; M. micruru: 2 = 0.98, F3.3 = 399, P<O.OOl).

In general, fish predation did not impact significantly on the copepod assemblage (Tables 1 and 4). However, the calanoid copepod N. incompositus


showed differences among treatments as the V+IV copepodite stage decreased significantly under macrophytes +fish treatment (T3) (Fig. 5, Table 4). Moreover, a net predator positive effect occurred when comparing the copepodite V+IV stage between fishless treatment (Tl) and fish treatment (T2) (NE = 25.5 %), while a significant negative net effect accounted in the macrophyte+fish treatment (T3) (NE = -34.4 %) (P <0.05 in all cases). On the con- trary, A. eigenrnanniorum produced no clear effects on the other fractions of the calanoid population (Fig. 5, Tables 1 and 4). Cyclopoid copepodite abundance changed significantly during the experiment when considering the temporal scale (Fig. 5, Tables 1 and 4).

We recorded significantly more macroinvertebrates in the macrophytes + fish treatment (Tables 1 and 4). Zygoptera, oligochaetes and Chironomidae larvae dominated the assemblage at the beginning of the experiment (Fig. 6). Qualitative trends showed that macroinvertebrate abundance declined in the fish treatment after addition of A. eigenmanniorum (Fig. 6) . Alternatively, in the macrophytes +fish treatment both groups of macroinvertebrates, predators and non-predators increased their abundances significantly (Tables 1 and 4). In the case of predators like Zygoptera larvae, the presence of macrophytes re- sulted in a positive effect, preventing fish predation (Fig. 6, Table 4). Also in the macrophyte+ fish treatment, the abundance of calanoid copepodite stage V + IV, as described above, decreased significantly compared to fishless and fish treatments, and a negative significant correlation occurred between the mean abundance of copepodite stage V+IV and mean Zygoptera abundance (Pearson’s correlation: 3 = 0.32, F13.p, = 3.65, P<0.05).

During the experiment, the lake experienced an increase of the cyanobacteria Anabaena spp. The presence of C. demersum in the third treatment created a positive effect in water transparency (Tables 2 and 4). We found higher chlorophyll-a, TP and TDP concentration in this treatment (Table 4), especially after the set up of the experiment (One way ANOVA: Chl-a: F2.6 = 72.08, TP: F2.6 = 31.23, TDP: F2.6 = 9.18; in all cases P<O.OOl). The significant higher values in the macrophytes+fish treatment are the result of the periphy- tic chlorophyll associated to C. demersum that gave off when we introduced the plants into the enclosures, wind-driven resuspension or to a sampling pro- cedure. We did not record a difference in TP concentration between treatments, but we observed higher levels of TDP in T 1 (Table 2 and 4).

Discussion

Considerations about the experimental design

The duration of our first and second experiment was 9 and 14days, respectively. Unfortunately, strong wind storms precluded the continuation of both


experiments. This time duration can be considered too short to detect significant indirect results. However, we believe that the two experiments showed the direct and combining effect of a littoral fish, macrophytes and macroinvertebrates on zooplankton community and size structure. Therefore, our results indicated that the complexity of the littoral food web and the general state- ments as “macrophyte as refuge areas” should be considered in Pampa plain lakes, or other lakes, with a rich macrophyte-macroinvertebrate assemblage.

The scale of our closed-enclosure system proved sufficiently large as no biases toward increasing significant effects were found. Caged-effects were expected including periphyton development on the walls of the enclosures, but not a negative effect on the zooplankton or fish due to increase of temperature. Moreover, the decline on Cladocera populations in the experiments may be associated with an intrinsic/extrinsic (like cyanotoxins) factor(s), as the same populations fallen also occurred in the lake. In addition, any removal sampling effect is diminished in this case. It is possible that Cladocera decline could be associated to the increase of cyanobacteria reported in the experimentshke, as the noxious effect of cyanotoxins on these microcrustaceans is well docu- mented (FERRAO-FILHO et al. 2000).

Even when enclosures in the second experiment were set-up in a littoral area free of macrophytes, what intended to homogenize macroinvertebrate distribution among treatments, macroinvertebrate distribution was biased. The fishless treatment (T 1) sustained lower abundances than the rest, and the macrophyte+fish treatment (T3) higher ones. This differential macroinvertebrate distribution did not allow the detection of a fish predation effect on macroinvertebrates or an indirect effect on zooplankton. Nevertheless, the ex- istence of a macroinvertebrate abundance gradient permitted to establish a correlation between macroinvertebrate and zooplankton abundance, determining a negative correlation of macroinvertebrate predators on each zooplankton Prey t Y Pea

Direct and indirect fish predation effects

Indirect effects occur when the impact of one species on another requires the presence of a third one. They can arise, for example, through linked chains of direct interaction. Trophic cascades and exploitative competition are examples of indirect interaction chain effects (WOOTON 1994). In complex structured areas as littoral zones, alternative prey may mediate a refuge effect for large zooplankton against fish predation or furthermore, fish could release zooplankton from intermediate predators. Likely, in our first experiment we found that larger size-classes like N. incompositus females (1.1 mm) and different copepodite stages increased in fish treatment and also that fish had a moderate to strong positive net effect on this prey. Moreover,. qualitative and quantita-


tive trends support the idea of a predation fish effect on macroinvertebrates. For example, oligochaetes declined significantly under fish treatment, a negative moderate to strong fish effect occurred on different macroinvertebrate taxa, and the interaction of fish presence and time led to significantly lower chironomid and Zygoptera larvae abundances. These findings evidence that size-selective predation is at work across taxonomic group and this trend sug- gests that A. eigenmanniorum preferred or foraged more efficiently on larger prey such as macroinvertebrates instead on larger zooplankton. Cladoceran assemblage in Los Padres Lake consisted of small and medium-sized species. B. huaronensis, C. dubia and M. micrura are below 0.6mm in body length, and this size is likely to be effective in escaping from predation of visually oriented planktivores (BROOKS & DODSON 1965). HAMBRIGHT & HALL (1992), BEKLIOGLU & Moss (1996) and Moss et al. (1998) also document the lack of fish effect on small cladoceran, but SCHRIVER et al. (1995) and LIESCHKE & CLOSS (1999) found a high fish predation pressure on small zooplankton like Ceriodaphnia or Bosmina. In this sense, in the first experiment performed in Los Padres Lake, the lack of an effect of A. eigenmanniorum on the small and large size fraction of the zooplankton populations must be related to the shift to larger prey like macroinvertebrates and not to a visual detection issue.

In our second experiment, we did not observe a fish predation effect on the different zooplankton size fractions. Under the fish treatment, macroinvertebrates declined and we found no clear effect of fish on cladocerans or copepods. Instead, under the macrophyte+fish treatment, macroinvertebrates were more abundant, most likely because C. demersum acted as a macroinvertebrate refuge against fish predation and in this case, a negative impact on copepodites stage V+IV occurred. Fish diets are very broad and dynamic as the result of the variability and availability of larger prey, even in the case of fish with clear preferences (EGGERS 1982). In addition, the ontogenetic size-dependent shifts from microcrustaceans to macroinvertebrates may be highly variable and this shift in juvenile fish is constrained by competition with other planktivorous and benthivorous fish (GARC~A-BERTHOU 1999, GARC~A-BERTHOU & MORENO-AMICH 2000, HJELM et al. 2000, PERSSON & BRONMARK 2002). In particular, Astyanax species are described as omnivorous, highly opportunistic and with a broad dietary spectra highly dependent on resource availability (Es-

VILELLA et al. 2002). In our experiments, we used fish ranging on 4.5-5cm in length. Based on the facts that A. eigenmanniorum is a more insectivorous species (VILELLA et al. 2002) and that ontogenetic diet shift occurs in some Astyanax species when a size class higher than 5-6cm is reached (ESTEVES

1996), we expected a fish predation impact either on zooplankton or macroinvertebrates depending, mostly, on the availability of these food items. In both experiments, fish did not impact on the different zooplankton fractions, more-

TEVES 1996, DA LUZ & OKADA 1999, LOBON-CERVIA & BENNEMANN 2000,


over qualitative and quantitative trends in experiment 1 suggest that A. eigenmanniorum depressed macroinvertebrates. Furthermore, net fish predation effects were significantly strong to moderate positive on copepods in experiment 1, while in experiment 2 we found a positive net effect in fish treatment (T2) for the V+IV copepodite stage, but a negative in macrophyte+fish treatment (T3) that sustained significantly more Zygoptera larvae in respect to T2. In addition, significant effects on larger size classes of the calanoid populations, like N. incompositus females, copepodite stages V+IV and 111, were negative and strongly correlated with predacious macroinvertebrates abundance (acari in experiment 1 or Zygoptera larvae in experiment 2). Accor- dingly, A. eigenmanniorum produced an indirect effect on zooplankton through the regulation of intermediate predators, like macroinvertebrates, and cascading down the direct fish effect on macroinvertebrate prey to the zooplankton trophic level.

The structural complexity of littoral zones may favour daphnids and other zooplankton to seek refuge from fish predation. But macrophytes are indeed the habitat of epiphytic and benthic invertebrates (KORNIJ~W & KAIRESALO 1994, ZIMMER et al. 2001) and many of them are considered potential zooplanktivorous predators, such as odonate larvae (LOMBARDO 1997, HIRVONEN

1999, BURKS et al. 2001), notonectids (ARNBR et al. 1998, HAMPTON et al. 2000), water mites (BUTLER & BURNS 1991, BALSEIRO 1992), flatworms (BLAUSTEIN & DUMONT 1990, BEISNER et al. 1996) and also, in our lakes, grass shrimps (COLLINS 1999). Under this scenario, zooplankton must face multiple predators and the benefit of escaping from pelagial predators must be balanced with the cost of confronting predators in the littoral zone (Moss et al. 1998, SCHEFFER 1998, BURKS et al. 2001, BURKS et al. 2002). Particularly, the shallow lakes of the Pampa plain, like Los Padres Lake, contain an extended littoral zone rich in both small littoral omnivorous fish, like A. eigenmanniorum, and macroinvertebrate predators. As littoral fish are forced to the littoral areas by piscivorous fish, macroinvertebrates have to coexist with their predators and must develop associated adjustments to avoid fish predation as increasing hiding rates (DIONNE et al. 1990) or reducing movements and respon- siveness toward prey (MCPEEK 1990). In the case of zooplankton, similar associated trade off must exit to avoid macroinvertebrate predators associated with littoral zones in Los Padres Lake. Small cladocerans are evenly distributed between vegetated areas and open water, while the adults and copepodites of N. incompositus avoid the inner part of the macrophyte beds as well as the edge (GONZALEZ SAGRARIO, pers. obser.). In our experiments, small cladoceran did not experience a direct or indirect fish predation effect, instead the different stages of the N. incompositus population were sensible to indirect predation effects and this might explain the avoidance of the littoral area by the different calanoid size fraction observed in the field.

570 Maria de 10s Angetes Gonzalez Sagrario and Esteban Balseiro

An unexpected result from experiment 2 was the significant enhance in chlorophyll-a concentration in the macrophyte + fish treatment compared with treatments without vegetation. This increase might be associated with the de- tached epiphytic algae, probably by wind-driven resuspension, that con- tributed to the water column phytoplankton biomass in the enclosures. AHN et al. (1997) and SCHWEIZER (1997) reported that benthic diatoms accounted for more than the 50 % to the phytoplankton biomass in littoralhearshore waters. Our experimental results support the importance of the epiphytic origin of phytoplankton in littoral areas.

Conclusions

According to our study, the potential role of littoral zones as refuge areas for zooplankton may be highly dependent on the specific trophic composition of the littoral food web, especially in the richness of multiple intermediate predators that could mediate fish impact on zooplankton. Macroinvertebrate predators, such as Zygoptera larvae, play a central role because they act as prey for fish and also might exert a high predation pressure on zooplankton. If fish impact is weak on invertebrate predators, macrophytes could constitute risky areas for large zooplankton like calanoid copepods. Littoral fish such as A. eigenmanniorum play an indirect positive effect on zooplankton releasing macroinvertebrate predation on large individuals and promoting macrophyte refuge effect. Therefore, we can conclude that trophic interactions in the littoral zone are complex, and whether or not macrophytes could be used as refuge areas by large zooplankton is likely to depend on direct fish effect on the density of multiple macroinvertebrate predators that inhabit the littoral and could impact on zooplankton populations.

Acknowledgements

We would like to thank N. SAGRARIO, N.,GONZALEZ, L. GONZALEZ SAGRARIO and L. D’ALESSIO for field assistance during the set up of the experiments. J. CARRETO, C. QUEIMALIROS and M. REISSIG provided their facilities and helped during chlorophyll and nutrient analysis. We also would like to thank H. BENAVIDES for his unconditional support and the Club de Pesca Atlantic0 for the use and free access to boats in Los Padres Lake. Finally, thanks to R. BURKS, E. JEPPESEN, B. MODENUTTI, C. DE FRAN- CESCO, E. SPIVAK, L. SANTAMAR~A, and the two anonymous reviewers for their critical comments that greatly improved this manuscript. M. A. GONZ~LEZ SAGRARIO is sup- ported by a fellowship of CONICET.

References AHN, 1.-Y., CHUNG, H., KANG, J.-S. & KANG, S.-H. (1997): Diatom composition and

biomass variability in nearshore waters of Maxwell Bay, Antartica, during the 1992/93 austral summer. - Polar Biology 17: 123-130.

Indirect enhancement of larae zoodankton 571

APHA (1989): Standard methods for the examination of water, sewage, and waste- water. - lTh Ed. American Public Health Association, Washington, D.C., pp. 1550.

A R N ~ R , M., KOIVOSTO, S., NORBERG, J. & KAUTSKY, N. (1998): Trophic interactions in rockpool food webs: regulation of zooplankton and phytoplakton by Notonecta and Daphnia. - Freshwat. Biol. 39: 79-90.

BALSEIRO, E. G. (1992): The role of pelagic water mites in the control of cladoceran population in a temperate lake of the southern Andes. - J. Plankton Res. 14: 1267- 1277.

BEISNER, B. E., MCCAULEY, E. & WRONA, F. J. (1996): Temperature-mediated dynamics of planktonic food chains: the effect of an invertebrate carnivore. - Freshwat. Biol. 35: 219-232.

BEKLIOGLU, M. & Moss, B. (1996): Mesoscosm experiments on the interaction of sediment influence, fish predation and aquatic plants with the structure of phytoplankton and zooplankton communities. - Freshwat. Biol. 36: 315-325.

BERTOLO, A., LACROIX, G., LESCHER-MOUTOU~, F. & SALA, S. (1999): Effects of physical refuges on fish-plankton interactions. - Freshwat. Biol. 41: 795 -808.

BLAUSTEIN, L. & DUMONT, H. J. (1990): Typhloplanid flatworms (Mesostoma and related genera): mechanisms of predation and evidence that they structure aquatic invertebrate communities. - Hydrobiologia 198: 61-77.

BROOKS, J. L. & DODSON, S. I. (1965): Predation, body size, and composition of plankton. - Science 150: 28-35.

BURKS, R. L., JEPPESEN, E. & LODGE, D. M. (2001): Pelagic prey and benthic predators: impacts of odonate predation on Daphnia. - J. N. Amer. Benthol. SOC. 20: 615-628.

BURKS, R. L., LODGE, D. L., JEPPESEN, E. & LAURIDSEN, T. L. (2002): Die1 horizontal migration of zooplankton: cost and benefits of inhabiting the littoral. - Freshwat. Biol. 47: 343-365.

BUTLER, M. I. & BURNS, C. W. (1991): Prey selectivity of Piona exigua, a planktonic water mite. - Oecologia 86: 210-222.

CANEVARI, M., CANEVARI, P., CARRIZO, G. R., HARRIS, G., RODR~GUEZ MATA, J. & STRANECK, R. J. (1991): Aves Argentinas. Tom0 11. - Fundacibn Acindar, pp. 497.

CARPENTER, S. R. & LODGE, D. M. (1986): Effects of submerged macrophytes on eco- system processes. - Aquat. Bot. 26: 341-370.

COLLINS, P. A. (1999): Feeding of Palaemonetes argentinus (Decapoda: Palaemonidae) from a oxbow lake of the Paranfi River, Argentina. - J. Crustacean Biol. 19: 485- 492.

DA Luz, B. D. G. & OKADA, E. K. (1999): Diet and dietary overlap of three sympatric fish species in lakes of the Upper Paranfi River floodplain. - Braz. Arch. Biol. Technol. 42: 441-447.

DEL HOYO, M. L. (1987): Caracterizacidn ecol6gica de las aves acufiticas de la Laguna de Los Padres, Pdo. Gral. Pueyrredbn, Prov. Buenos Aires. - Tesis de Licencia- tura, Universidad Nacional de Mar del Plata.

DIEHL, S. & EKLOV, P. (1995): Effects of piscivore-mediated habitat use on resources, diets, and growth of perch. -Ecology 76: 1712-1726.

DIEHL, S. & KORNIJ~W, R. (1998): Influence of submerged macrophytes on trophic interactions among fish and macroinvertebrates. - In: JEPPESEN, E., S~~NDERGAARD,

Submerged Macrophytes in Lakes. - Ecological Studies 131, Springer, New York, M., SONDERGAARD, M. & CHRISTOFFERSEN, K. (eds.): The Structuring Role of

pp. 24-46.


DIONNE, M., BUTLER, M. & FOLT, C. (1990): Plant-specific expression of antipredator

EGGERS, D. M. (1982): Planktivore preference by prey size. - Ecology 63: 381-390. ESTEVES, K. E. (1996): Feeding ecology of three Astyanax species (Characidae, Tetra-

gonopterinae) from a floodplain lake of Mogi-GuaGd River, Parani River Basin, Brazil. - Envir. Biol. Fish 46: 83-101.

FERR~O-FILHO, A., AZEVEDO, S. M. F. 0. & DEMOTT, W. R. (2000): Effects of toxic and non-toxic cyanobacteria on the life history of tropical and temperate cladoce- ram. - Freshwat. Biol. 45: 1-19.

GARC~A-BERTHOU, E. (1999): Spatial heterogeneity in roach (Rutilus rutilus) diet among contrasting basins within a lake. -Arch. Hydrobiol. 146: 239-256.

GARC~A-BERTHOU, E. & MORENO-AMICH, R. (2000): Food of introduced pumpkinseed sunfish: ontogenetic diet shift and seasonal variation. - J. Fish Biol. 57: 29-40.

GROSMAN, M. F., CASTELAIN, J. R. G. & USUNOFF, E. J. (1996): Trophic niches in an Argentine pond as a way to assess functional relationships between fishes and other communities. -Water SA 22: 345-350.

HALL, S. R. & RUDSTAM, L. G. (1999): Habitat use and recruitment: a comparison of long-term recruitment patterns among fish species in a shallow eutrophic lake, Oneida Lake, N.Y. USA. - Hydrobiologia 408/409: 101-1 13.

HAMBRIGHT, K. D. (1994): Morphological constraints in the piscivore-planktivore interaction: implications for the trophic cascade hypothesis. - Limnol. Oceanogr.

HAMBRIGHT, K. D. & HALL, R. 0. (1992): Differential zooplankton feeding behaviour, selectivities, and community impacts of two planktivorous fish. - Environ. Biol. Fish. 35: 401-41 1.

HAMFTON, S. E., GILBERT, J. J. & BURNS, C. W. (2000): Direct and indirect effects of juvenile Buenoa macrotibialis (Hemiptera: Notonectidae) on the zooplankton of a shallow pond. - Limnol. Oceanogr. 45: 1006-1012.

HIROVEN, H. (1999): Shifts in foraging tactics of larval damselflies: effects of prey density. - Oikos 86: 443-452.

HJELM, J., PERSSON, L. & CHRISTENSEN, B. (2000): Growth, morphological variation and ontogenetic niche shifts in perch (Perca puviutilis) in relation to resource availability. - Oecologia 122: 190-199.

JOHNSON, D. M. (1991): Behavioural ecology of larval dragonflies and damselflies. - Trends Ecol. Evol. 6: 8-13.

JOHNSON, D. M., MARTIN, T. H., CROWLEY, P. H. & CROWDER, L. B. (1996): Link strength in lake littoral food webs: net effects of small sunfish and larval dragonflies. - J. N. Amer. Benthol. SOC. 15: 271-288.

KORNIJ~W, R. & KAIRESALO, T. (1994): Elodea canadensis sustains rich environment for macroinvertebrates. - Verh. Internat. Verein. Limnol. 25: 2270-2275.

LAURIDSEN, T. L. & BUENK, I. (1996): Die1 changes in the horizontal distribution of zooplankton in the littoral zones of two shallow eutrophic lakes. - Arch. Hydro- biol. 137: 161-176.

tal migration of zooplankton: predator-mediated use of macrophyte habitat. - In:

(eds.): The Structuring Role of Submerged Macrophytes in Lakes. - Ecological Studies 131, Springer, New York, pp. 233 -239.

behaviour by larval damselflies. - Oecologia 83: 371-377.

39: 897-912.

LAURIDSEN, T. L., JEPPESEN, E., SONDERGAARD, M. & LODGE, D. M. (1998): Horizon-

JEPPESEN, E., SONDERGAARD, M., SONDERGAARD, M. & CHRISTOFFERSEN, K.


LAURIDSEN, T. L., PEDERSEN, L. J., JEPPESEN, E. & S~NDERGAARD, M. (1996): The importance of macrophyte bed size for cladoceran composition and horizontal migration in a shallow lake. - J. Plankton Res. 18: 2283-2294.

LIESCHKE, J. A. & CLOSS, G. P. (1999): Regulation of zooplankton composition and distribution by a zooplanktivorous fish in a shallow, eutrophic floodplain lake in south east Australia. -Arch. Hydrobiol. 146: 397-412.

LOBON-CERVIA, J. & BENNEMANN, S. (2000): Temporal trophic shifts and feeding di- versity in two sympatric, neotropical, omnivorous fishes: Astyanax bimaculatus and Pimelodus maculatus in Rio Tibagi (Paranh, Southern Brazil). - Arch. Hydro- biol. 149: 285-306.

LOMBARDO, P. (1997): Predation by Enallagma nymphs (Odonata, Zygoptera) under different conditions of spatial heterogeneity. - Hydrobiologia 356: 1- 9.

LORENZEN, C. J. (1967): Determination of chlorophyll and pheo-pigments: Spectro- photometric equations. - Limnol. Oceanogr. 12: 343 -346.

MCPEEK, M. A. (1990): Behavioral differences between Enallagma species (Odonata) influencing differential vulnerability to predators. - Ecology 71: 1714-1726.

MITTELBACH, G. G . (1988): Competition among refuging sunfishes and effects of fish density on littoral zone invertebrates. - Ecology 69: 614-623.

MOSS, B., KORNIJ~W, R. & MEASY, G. (1998): The effects of nymphaeid (Nuphar lu- tea) density and predation by perch (Perca Juviatilis) on the zooplankton communities in a shallow lake. - Freshwat. Bibl. 39: 689-697.

PATERSON, M. J. (1994): Invertebrate predation and the seasonal dynamics of micro- crustacea in the littoral zone of a fishless lake. - Arch. Hydrobiol. 99: 1-36.

PERSSON, L. & BRONMARK, C. (2002): Foraging capacities and effects of competitive release on ontogenetic diet shift in bream, Abramis brama. - Oikos 97: 271-281.

RINGUELET, R. A. (1975): Zoogeografia y ecologia de 10s peces de aguas continentales de Argentina y consideraciones sobre las Areas ictioldgicas de AmCrica del Sur. - Ecosur 2: 1-122.

SCHEFFER, M. (1998): Ecology of Shallow Lakes. - lSf ed. Chapman and Hall, Suffolk,

SCHRIVER, P., B~GESTRAND, J., JEPPESEN, E. & S~NDERGAARD, M. (1995): Impact of submerged macrophytes on fish-zooplankton-phytoplankton interactions: large- scale experiments in a shallow eutrophic lake. - Freshwat. Biol. 33: 255-270.

SCHWEIZER, A. (1997): From littoral to pelagial: comparing the distribution of phytoplankton and ciliated protozoa along a transect. - J. Plankton Res. 19: 829-848.

SIH, A. (1987): Nutritional Ecology of aquatic insect predators. - In: SLANSKY, F. Jr. & RODRIGREZ, J. G. (eds.): Nutritional Ecology of Insects, Mites, Spiders and related invertebrates. -John Wiley & Sons, Inc., New York, pp. 579-607.

SIH, A., ENGLUND, G. & WOOSTER, D. (1998): Emergent impacts of multiple predators on prey. -Trends Ecol. Evol. 13: 350-355.

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. - Hydrobiologia 3421343: 229-240.

UNDERWOOD, A. J. (1997): Experiments in ecology: their logical design and interpreta- tion using analysis of variance. - Cambridge University Press, pp 504.

VILELLA, F. A., BECKER, F. G. & HARTZ, S. M. (2002): Diet of Astyanax species (Teleostei, Characidae) in an atlantic forest river in southern Brazil. - Braz. Arch. Biol. Techn. 45: 223-232.

pp. 357.

574 Maria de 10s Anaeles Gonzafez Saarario and Esteban Balseiro

WHITESIDE, M. C. (1988): O+ fish as major factors affecting abundance and patterns of littoral zooplankton. - Verh. Internat. Verein. Limnol. 23: 1710-1714.

WOTTON, J. T. (1994): The nature and consequences of indirect effects on ecological communities. - Annu. Rev. Ecol. Syst. 25: 443-466.

ZARET, T. M. & KERFOOT, W. C. (1975): Fish predation on Bosmina longirostris: body-size selection versus visibility selection. - Ecology 56: 232-237.

ZIMMER, K. D., HANSON, M. A., BUTLER, M. G. & DUFFY, W. G. (2001): Size distribution of aquatic invertebrates in two prairie wetlands, with and without fish, with implications for community production. - Freshwat. Biol. 46: 1373-1386.

Submitted: 30 January 2003; accepted: 7 September 2003.

Documents

Indirect enhancement of large zooplankton by …investigadores.uncoma.edu.ar/Lab_Limnologia/publicaciones/Gonzalez... · brates as predators of zooplankton (ARN~R et al. 1998, HIROVEN