Alternative Stable States in Size-Structured Communities: Patterns, Processes, and Mechanisms

Home , Alternative stable state, Journal of Ecology, Priority effect

Arne Schröder

2008

Akademiska Avhandling som med vederbörligt tillstånd av rektorsämbetet vid Umeå Universitet för erhållande av filosofie doktorsexamen i ekologi kommer att offentligen försvaras på engelska fredagen den 23 maj 2008, kl. 13:00 i Lilla hörsalen, KBC-huset.

Examinator: Prof. Dr. Lennart Persson, Umeå Universitet

Opponent: Prof. Dr. Tim Benton, Faculty of Biological Sciences, University of Leeds, Leeds, United Kingdom

Department of Ecology and Environmental Science Umeå University SE-901 87 Umeå Sweden

Organisation: Document name: Dept. of Ecology and Environmental Science Doctoral Dissertation Umeå University Date of issue: SE-901 87 Umeå, Sweden 30th of April 2008

Author: Arne Schröder

Title: Alternative Stable States in Size-Structured Communities: Patterns, Processes, and Mechanisms.

Abstract: Alternative stable states have been, based on theoretical findings, predicted to be common in ecological systems. Empirical data from a number of laboratory and natural studies strongly suggest that alternative stable states also occur in real populations, communities and ecosystems. Potential mechanisms involve population size- structure and food-dependent individual development. These features can lead to ontogenetic niche shifts, juvenile recruitment bottlenecks and emergent Allee effects; phenomena that establish destabilising positive feedbacks in a system and hence create alternative stable states. I studied the consequences of population size-structure for community dynamics at different scales of system complexity. I performed laboratory and ecosystem experiments. Small poecilliid fishes and planktonic invertebrates with short generation times and life spans were used as model organism. This allowed me to assess the long-term dynamics of the populations and communities investigated. The main experimental results are: (a) An ontogenetic niche shift in individuals of the phantom midge Chaoborus made the population vulnerable to an indirect competitive recruitment bottleneck imposed by cladoceran mesozooplankton via rotifers. Consequentially the natural zooplankton food web exhibited two alternative attractors. (b) Body size determined the success of Poecilia reticulata invading resident population of Heterandria formosa and thus the type of alternative stable state that established. Small invaders were outcompeted by the residents, whereas large invaders excluded their competitor by predating on its recruits. (c) External juvenile and adult mortality altered the internal feedback structure that regulates a laboratory population of H. formosa in such a way that juvenile biomass increased with mortality. This biomass overcompensation in a prey population can establish alternative stable states with top- predators being either absent or present. The major conclusion is that size-structure and individual growth can indeed lead to alternative stable states. The considerations of these ubiquitous features of populations offer hence new insights and deeper understanding of community dynamics. Alternative stable states can have tremendous consequences for human societies that utilise the ecological services provided by an ecological system. Understanding the effects of size- structure on alternative stability is thus crucial for sustainable exploitation or production of food resources.

Key words: biomass overcompensation, Chaoborus, emergent Allee effect, Heterandria formosa, juvenile recruitment bottlenecks, ontogenetic niche shifts, Poecilia reticulata

Language: English ISBN: 978-91-7264-499-1 Number of pages: 11 + 4 papers Signature: Date: 15th of April 2008

To my father

Food is the first thing…

Bertholt Brecht (The Three-Penny-Opera)

This thesis is based on the following papers, which will be referred to in the text by their Roman numerals:

I Schröder, A., Persson, L. & de Roos, A.M. 2005. Direct experimental evidence for alternative stable states: a review. Oikos 110, 3 – 19.

II Schröder, A., Westman, E., Persson, L. & de Roos, A.M. Alternative zooplankton community attractors revealed in a whole-lake perturbation experiment. (unpublished manuscript)

III Schröder, A., Nilsson, K.A., Persson, L., & van Kooten, T. Invasion success depends on invader body size in a size-structured mixed predation – competition system. (unpublished manuscript)

IV Schröder, A., Persson, L., & de Roos, A.M. Positive effects of mortality: Biomass overcompensation in response to experimental culling. (unpublished manuscript)

Paper I is reproduced with the kind permission from the publisher.

Contents

Introduction 1

Objectives 3

Material and Methods 4

Major Results and Discussion 5

Summary, Conclusions and Perspectives 6

Acknowledgements 6

References 7

Thanks 10

Appendices: Paper I – IV

Introduction When the Canadian government closed the cod fisheries in the North-West Atlantic in 1992, in the hope to open it up again in a few years when the overexploited stocks would have had recovered, they were in for an unwelcome surprise: nowadays, 16 years later, stock density is still as low as at the start of the moratorium. Why? When in the seventies, throughout the Western Hemisphere, blue, sparkling clear lakes turned into green, repellent looking (and smelling) basins, authorities attempted to tackle the problem with the construction of sewage plants and by constraining the use of agricultural fertiliser. They did so often enough in vain, as many lakes remained eutrophied. Why? When a modest increase in their wealth led pastoralists in the Sahel zone to invest into larger herds, desertification rates multiplied and utilisable grazing areas turned into barren ground. Today, many societies in this region are, despite several ecological restoration and developmental aid programs, in more precarious situations than a few decades ago. Why? This thesis is not about answering these questions as they have been asked and answered before. No, it is about something, the answers to these seemingly unrelated problems all have in common, namely the concept of alternative stable states. As explained in more detail in the introduction to paper I (text and Fig. 1 therein), alternative stable states mean that an ecological system can be in several contrasting states (e.g. high versus low fish stock density, transparent versus turbid water, vegetated versus barren ground etc.) despite that the external environmental conditions are identical. Furthermore, these alternative states are stable in the sense that they (a) exhibit the long- term dynamics of each state as determined by its underlying attractor rather than transient dynamics and (b) that the system is irreversibly locked in each contrasting state and can not shift, without external disturbances strong enough to initiate such a state transition, to the alternative state. Such state irreversibility can have, as the above examples demonstrate, wide-ranging implications, not only ecologically but also socially and politically. Often, different ecosystem states are differently beneficial for the human societies that utilise and rely on the ecosystem services, such as food production or recreation. For example, when overgrazing by too large cattle herds removes the vegetation cover, the initiated soil erosion and reduction in water retention potential shifts the whole ecosystem irreversibly to a desert, where plants (and cattle) can not live anymore. The disrupting effects on society can be tremendous. Understanding the causes and predicting the effects of alternative stability in ecological systems is thus crucial for a sustainable utilisation and management of ecological resources, which are despite all technological progress still the foundation of all our lives. The general mechanisms behind alternative stables states involve positive feedbacks where the system component of interest (as described by the state variable, for example population density, water clarity or vegetation cover) has a positive effect on itself mediated by other components of the system (DeAngelis, Post & Travis 1986, Wilson & Agnew 1992). In ecological systems where dynamics ultimately are driven by organisms and the populations they constitute, positive feedbacks originates from positive density dependence in population regulation, i.e. per-capita population growth rate increases with population size (before negative density dependence cuts in and it finally decreases again at even higher density) (Berryman, Lima & Hawkins 2002). Positive density dependence

can have many different reasons (see Courchamp, Berec & Gasgoine 2008 for a comprehensive overview). It can arise from beneficial modifications individuals impose on their physical environment, for example when denser vegetation cover leads to higher soil water retention allowing more plants to establish (Rietkert et al. 2004) or when algal/bacterial bio films stabilise sediments and promote further encrustation (van de Koppel, Rietkert & Weissing 1997, van de Koppel et al. 2001). Another possibility for positive density dependence arises when two competing species each have a stronger negative influence on the other species than on conspecifics. Such an interaction may stem from interference between species (Begon, Harper & Townsend 1996). The initially more abundant species, independent of its identity, will than benefit from a priority effect and increase further in density and hamper the establishment of the other species. Social mechanisms which increase survival can also lead to positive density dependence. This may happen more passively, for example when prey aggregation dilutes predator effects, or actively, like when increased vigilance in larger groups reduces predator attack success (Stephens & Sutherland 1999, Courchamp et al. 2008). When positive density- dependence is strong enough so that population growth is even negative at low densities, a certain minimum number of individuals is required for establishment and persistence of the population at a site. Once the population size drops below this threshold, the population is not viable anymore and the community the population is a part of exhibits two alternative stable states: a state with the population present, and one without it. But why community size-structure? Body size is one of the most influential characteristics of an individual as many physiological and ecological traits scale with body size. The metabolic costs for homeostasis and maintenance, the attack rate and handling time of resource items, the vulnerability to predation or abiotic mortality factors are all size-dependent. Therefore, body size determines to a very large extent the ecological role and performance (as defined by the set and strength of trophic interactions they are involved in) of an individual. Size-dependent processes can have profound consequences for the interaction of small and large hetero- or conspecific individuals and the dynamics of the communities they are living in. Size-dependent interactions can lead to positive feedbacks and hence to alternative stable states. For example, Chase (2003a) could in an experiment with two differently sized freshwater snail show that depending on the initial prey : predator ratio two contrasting communities developed. The species which traded-off growth to a size refuge from predation for early reproduction and thus spend longer time vulnerable to predation before maturation became the dominant species when the initial ratio was high and per-capita predator effects thus low. In contrast when the prey : predator ratio was low, the earlier maturing but smaller species became dominant (overall snail biomass, remained, however, low and predator control high). However, differences in body size between species are only one aspect of size- structure. The individuals of the majority of animal taxa grow substantially between birth or hatching and their maturation or death (Werner 1988). Accordingly, for the same reasons as discussed above for between-species differences in body size, an individual’s ecological role and performance change often profoundly through its ontogeny. It is not uncommon for individuals of many species to change their resource use and feeding strategies with increase in body size over their ontogeny (Werner & Gilliam 1984, Neill 1988, Wilbur 1988). Also vulnerability to predation changes over ontogeny as predation is usually (negatively) size-selective. These ontogenetic niches shifts can lead to juvenile recruitment bottlenecks where a species may be limited in its population growth by low

maturation or survival to the adult stage (Neill 1975, Walters & Kitchell 2001). Maturation rate is low when juveniles experience high intra- or interspecific competition. Strong competition might also initiate low starvation-induced survival as is possible by predation pressure on juveniles. These recruitment bottlenecks can lead to a number of interesting phenomena. When heterospecific individuals at a certain life history stage resemble each other more in their resource use and niche occupation than their conspecifics at another size the sign of their trophic interaction changes through ontogeny. Many predator species grow through size ranges where they have to compete with their later prey for food, but are due to morphological pre-adaptations to their later life style less efficient foragers on the shared resource (Wilbur 1988). Thus the prey can impose a competitive recruitment bottleneck on its predator and may even exclude it from the community. However, once predator individuals make it through this bottleneck, they reduce the density of the prey and even drive it to extinction, thereby enhancing the recruitment of their own offspring. This leads to another stable state in such a mixed predation – competition system, where only the predator persists (van de Wolfshaar, de Roos & Persson 2006. Ontogenetic shifts in predation vulnerability can also have implications for a top predator size-selectively feeding on a prey. By relaxing competition (thus opening the bottleneck) among the surviving prey individuals, a predator can change the feedback structure regulating its prey population density and size-structure (de Roos et al. 2007). As different size-classes of the prey will respond differently to the relaxed competition, a predator can, depending on its size-selectivity and the exact prey bottleneck, actually increase the density of the size-class it is feeding on, thereby stimulating its own population growth. This positive density dependence can again lead to alternative stable community structures with and without the predator (de Roos & Persson 2002). Predation-induced shifts in prey population size-structure may also facilitate other predators that feed on size-classes at the opposite end of the size spectrum than the predator that is triggering the shift (de Roos et al. unpublished manuscript). All in all, differences in body size between hetero- or conspecific individuals may have wide-ranging effects on communities and food webs. Further insights how size- structure can determine dynamics and patterns in ecological systems are required.

Objectives The objective of this thesis was to experimentally gain insights into the effects of population size-structure on community dynamics with special attention to the occurrence of and mechanisms behind alternative stable states. The notion of alternative stable states goes back at least to the late sixties (Lewontin 1969, Holling 1973, May 1977) and its relevance for natural systems has been strongly debated (Connell & Souza 1983, Peterson 1984, Souza & Connell 1985, Sutherland 1990), so I started out with the questions: How to actually look for alternative stability? What possible manipulations can be applied to test for alternative stable states? And more, what is the actual experimental evidence for them in the literature? The answers to these questions led to paper I. It turned out, that positive evidence was restricted to laboratory microcosms and field enclosures, so I next assessed whether alternative stable states occur on an appropriate scale for size-dependent ecological interactions. The whole-lake experiment conducted

led to paper II. Whole-ecosystem studies are the acid test for any ecological theory (Carpenter 1996, 1998), however, logistic difficulties and the inherent uncontrollable stochasticity sometimes make it hard to evaluate a certain mechanism. I therefore used a laboratory model system to study size-dependent interactions and their dynamical consequences in more detail. The results of this experiment are reported and discussed in paper III. Predation and harvesting can alter the internal feedback structure that regulates a size- structured population and can thereby shift the whole community the population is embedded in to an alternative stable state (de Roos & Persson 2002, Persson et al. 2007, de Roos et al. unpublished manuscript). I investigated the hypothesised mechanism behind such potential state transitions again in a laboratory aquarium set-up. This study produced paper IV.

Material and Methods Ecological studies can be performed at many spatial and temporal scales. These scales range from the measurement of individual traits such as attack rates in tiny arenas during a few hours, over microcosms under highly controlled laboratory conditions or within-generation experiments in field enclosures subjected to the natural background noise in environmental stochasticity to whole ecosystem studies involving the complete set of environmental drivers, system components, and internal feedbacks. Each approach has its unique set of heuristic advantages and disadvantages in terms of repeatability, mechanistic scrutiny and ecological relevance. I believe that scientific progress is best made by combining these approaches and I accordingly conducted experiments at different scales. In particular, I perturbed two whole lakes to assess whether the internal feedback structure (driven by the body sizes of the dominant species, the phantom midge Chaoborus) has changed and whether this affected the recovery potential of the pre- perturbation state on the scale of a whole ecosystem (paper II). I further used laboratory aquaria to investigate the implications of size-structure and food dependent development for community dynamics and the potential mechanisms under more controllable conditions (paper III and IV). Experiments testing for alternative stable states require by definition studies that are capable of informing about the long-term dynamics and the stability of the system under investigation. Practically, that means following a population or community for at least one whole life-span of the involved organisms. I therefore worked with short-living species with fast generation times. In particular, I chose dipteran insects, crustacean zooplankton and rotifers (for the whole-lake study of paper II) or two viviparous poecilliid fish species, namely Heterandria formosa (or Least Killifish) and Poecilia reticulata (or Common Guppy) (for the laboratory experiments of paper III and IV). Especially the latter two species are optimal for laboratory studies as they are not only small and fast but in addition allow the measurement of important individual traits such as growth rate, attack rate and fecundity, something that is not feasible with many other ecological model organisms like zooplankters or ciliates. Such quantification offers the opportunity to argue more mechanistically and base conclusions not solely on patterns obtained.

Major Results and Discussion After hot debates on what constitutes an alternative stable system (Connell & Souza 1983, Peterson 1984, Souza & Connell 1985, Sutherland 1990) and how to test for them (Connell & Souza 1983, Grover & Lawton 1994, Petraitis & Dudgeon 2004, Scheffer & Carpenter 2003, paper I, Schröder 2008), over the last few years several reviews have compiled substantial empirical evidence for their existence in ecological systems (Scheffer et al. 2001, Didham, Watts & Norton 2005, Chase 2003b). However, most of the cited studies used non-manipulative approaches or short-term studies that lack inferential rigor compared to experimental long-term studies. In paper I, we compiled studies that employed direct manipulations and long-term experiments to assess alternative stability. The results show that alternative stable states clearly exist in a wide variety of ecological systems, including terrestrial or freshwater habitats, communities of plants, unicellular organisms, insects or sessile invertebrates and also different trophic configurations. Size-structure and its contribution to the bistability was (although presumably important), apart from a very few studies, not explicitly considered, making it difficult to draw any general conclusions. The results from the literature review also shed new light on findings from non-manipulative approaches and short-term experiments, making their conclusions on the occurrence of alternative stable states more convincing (at least for the general question on the principle possibility). However, at the time of this study, no positive experimental evidence for alternative stable states on a natural scale (whole ecosystem) had been published, although many field enclosure studies strongly implied that they occur in the systems investigated (e.g. Handa, Harmsen & Jefferies 2002, Schmitz 2004, Chase 2003a). Nowadays, 3 years later, this picture has changed: Persson et al. (2007) demonstrated two contrasting attractors in the fish community of an arctic lake and in paper II, we report contrasting attractors in a natural zooplankton community of two small bog lakes. Both cases are strongly coupled to size-dependent interactions and food-dependent development in at least the dominant species. In the second case, it is an ontogenetic niche shift of the top-predator that makes it vulnerable to an indirect juvenile recruitment bottleneck imposed by its later prey (paper II). As the precise outcome of these experiments would not be explainable without major dynamical consequences of population size-structure and food-dependent development, these results underscore the importance of these features for the dynamics of whole food webs and ecosystems. An ontogenetic niche shift in combination with a classical juvenile recruitment bottleneck lies also behind the reported alternative stable states in the mixed predation – competition system investigated in paper III. This study shows how size- structure and size-dependent scaling of competitive strength and predation vulnerability can alter and even reverse the dynamical outcome of invasions. Depending on their body size, invaders either fail (when small) or succeed (when large). This makes the establishment of a local community not only dependent on the numbers of invaders but also on the body size of the colonising individuals. So far, size differences of colonisers have been largely neglected in community assembly, where priority effects are usually considered as being density-mediated and not size-mediated. In paper IV we could demonstrate how external size-selective mortality, as it may result for example from harvesting or predation, alters the internal feedback structure that regulates a size- structured prey population. Relaxed competition with increased external mortality led to increased adult fecundity. The higher reproductive output consequentially translated into a higher juvenile biomass at intermediate mortality pressure not only when adults where targeted but also when juveniles themselves were subjected to mortality. Such an

overcompensation in prey size class can, as has been demonstrated in a whole-lake experiment (Persson et al. 2007), lead to increased population growth with increased population size in a size-selectively feeding predator. This results from that a certain total consumption rate is required to initiate the beneficial shift in prey biomass distribution over size classes. This lead to that predators once excluded from system, can not re- invade in small numbers and that the system exhibits two alternative stable states: one with prey and predator and one without the predator (de Roos & Persson 2002).

Summary, Conclusions and Perspective My results clearly show that alternative stable states exist in ecological systems, and that they also occur in nature. These findings are rather unsurprising. Far more importantly, it is evident that size-structure and food-dependent development, two ubiquitous features of natural populations and individuals, which however are often overlooked in theoretical ecology, can lead to these patterns by creating the potential for ontogenetic niche shifts and juvenile recruitment bottlenecks. Moreover, these individual- and population-level processes propagate through more complex systems as communities, food webs and whole ecosystems. As a main conclusion, I would like therefore to emphasise the relevance of population size-structure and food-dependent individual performance for a more complete understanding of ecological systems at any level of organisation and scale. We only have started to look more detailed into this and many more questions remain un-answered, probably even un-asked, concerning especially the relevance and implications of these features for more complex systems. How do food webs behave when everything is size- structured and develops food-dependently? What are the effects of density/food- dependence in both juvenile and adult stages for population dynamics? How far does the presence of alternative stable states in a top-predator influence the rest of food web? When, under what circumstances and in what kind of systems are alternative stable states more likely to occur? Alternative stable states can have tremendous consequences for human societies that utilise the ecological services provided by an ecological system. And as size-structure increases the likelihood of the occurrence of alternative stable states, a deeper understanding the effects of size-structure on alternative stability is thus crucial for sustainable exploitation or production of food resources.

Acknowledgements I am indebted to many people for help in several ways. Joseph Travis and David Reznick were so kind to send us the fish. Lars-Ola Westlund, William Larsson, Lars Lundmark and Kenneth Österlund offered much needed technical support. Krister Mattsson, Mårten Söderquist, Emma Göthe, Christian Tiedemann and Torunn Skau helped with sampling, measuring and maintenance of the experiments. Pär Byström and Tobias van Kooten gave valuable comments on earlier drafts of this thesis. The Memorial of J.C. Kempe Foundation supported several parts of the work reported in this thesis financially.

References

Begon, M., Harper, J.L. & Townsend, C.R. (1996) Ecology. Individuals, Populations and Communities. Blackwell Science, Oxford. Berryman, A.A., Lima Arce, M. & Hawkins, B.A. (2002) Population regulation, emergent properties, and a requiem for density dependence. Oikos, 99, 600- 606. Carpenter, S.R. (1996) Microcosm experiments have limited relevance for community and ecosystem ecology. Ecology, 77, 677-680 Carpenter, S.R. (1998) The need for large-scale experiments to assess and predict the response of ecosystems to perturbation. Successes, Limitations, and Frontiers in Ecosystem Science (eds M.L. Pace & P.M. Groffman) pp. 287-312, Springer-Verlag Berlin. Chase, J. M. (2003a) Experimental evidence for alternative stable equilibria in a benthic pond food web. Ecology Letters, 6, 733-741. Chase, J.M. (2003b) Community assembly: when should history matter? Oecologia, 136, 489-498. Connell, J.H. & Souza, W.P. (1983) On the evidence needed to judge ecological stability or persistence. The American Naturalist, 121, 789-824. Couchamp, F., Berec, L. & Gascoigne (2008) Allee effects in ecology and conservation. Oxford University Press, Oxford. DeAngelis, D. L., Post, W. M. & Travis, C. C. (1986) Positive Feedbacks in Natural Systems. Springer-Verlag, Berlin Heidelberg. De Roos, A. M. and Persson, L. 2002. Size-dependent life-history traits promote catastrophic collapses of top predators. Proceedings of the National Academy of Science, 99, 12907-12912. De Roos, A.M., Schellekens, T., van Kooten, T., van de Wolfshaar, K., Claessen, D. & Persson, L. (2007) Food-dependent growth leads to overcompensation in stage- specific biomass when mortality increases: The influence of maturation versus reproduction regulation. The American Naturalist, 170, E59-E76. De Roos, A.M., Schellekens, T. van Kooten, T., & Persson, L. Stage-specific predator species help each other to persist while competing for a single prey. (unpublished manuscript). Didham, R.K., Watts, C.H. & Norton D.A. (2005) Are systems with strong underlying abiotic regimes more likely to exhibit alternative stable states? Oikos, 110, 409-416. Grover, J. P. & Lawton, J. H. (1994) Experimental studies on community convergence and alternative stable states: comments on a paper by Drake et al. Journal of Animal Ecology, 63, 484-487. Handa, I. T., Harmsen, R. & Jefferies, R. L. (2002) Patterns of vegetation change and the recovery potential of degraded areas in a coastal marsh system of the Hudson Bay lowlands. Journal of Ecology, 90, 86-99. Holling, C.S. (1973) Resilience and stability of ecological systems. Annual Reviews of Ecology and Systematics, 4, 1-23.

Lewontin, R.C. (1969) The meaning of stability: Diversity and Stability in Ecological Systems. Brookhaven Symposium in Biology 22, 13-24. May, M.R. (1977) Thresholds and breakpoints in ecosystems with a multiplicity of stable states. Nature, 269, 471-477. Neill, W.E. (1975) Experimental Studies of Microcrustacean Competition, Community Composition and Efficiency of Resource Utilization. Ecology, 56, 809-826. Neill, W.E. (1988) Community response to experimental nutrient perturbations in oligotrophic lakes: the importance of bottlenecks in size-structured populations. Size-structured Populations:Ecology and Evolution (eds B. Ebenman & L. Persson), pp. 236-255. Springer-Verlag, Berlin. Persson, L., Amundsen, P.-A., de Roos, A.M., Klemetsen, A., Knudsen, R. & Primicerio, R. (2007) Culling prey promotes predator recovery – alternative stable states in a whole-lake experiment. Science, 316, 1743-1746. Peterson, C. H. (1984) Does a rigorous criterion for environmental identity preclude the existence of multiple stable points? The American Naturalist, 124, 127-133. Petraitis, P. S., & Dudgeon, S. R. (2004) Detection of alternative stable states in marine communities. Journal of Experimental Marine Biology and Ecology, 300, 343 – 371. Rietkert, M., Dekker, S.C., de Ruiter, P.C. & van de Koppel, J. (2004) Self- organized patchiness and catastrophic shifts in ecosystems. Science, 305, 1926-1929. Scheffer, M., Carpenter, S., Foley, J.A., Folke, C. & Walker, B. (2001) Catastrophic shifts in ecosystems. Nature, 413, 591-596. Scheffer, M. & Carpenter, S. R. (2003) Catastrophic regime shifts in ecosystems: linking theory to observation. Trends in Ecology and Evolution, 8, 275 – 279. Schmitz, O. J. (2004) Perturbation and abrupt shift in trophic control of biodiversity and productivity. Ecology Letters, 7, 403-409. Schröder, A. (2008) Inference about complex ecosystem dynamics. New Models for Ecosystem Dynamics and Restoration. (eds R. Hobbs & K. Suding), in press. Island Press, Washington. Sousa, W. P. & Connell, J. H. (1985) Further comments on the evidence for multiple stable points in natural communities. The American Naturalist, 125, 612-615. Stephens, P.A. & Sutherland, W.J. (1999) Consequences of the Allee effect for behaviour, ecology and conservation. Trends in Ecology and Evolution, 14, 401-405. Sutherland, J. P. (1990) Perturbations, resistance, and alternative views of the existence of multiple stable points in nature. The American Naturalist, 136, 270-275. Van de Koppel, J., Herman, P.M.J., Thoolen, P. & Heip, C.H.R. (2001) Do alternate stable states occur in natural ecosystems? Evidence from a tidal flat. Ecology, 82, 3449-3461.

Van de Wolfshaar, K.E., de Roos, A.M., & Persson, L. (2006) Size-dependent interactions inhibit coexistence in intraguild predation systems with life- history omnivory. The American Naturalist, 168, 62-75. Walters, C. & Kitchell, J.F. (2001) Cultivation/depensation effects on juvenile survival and recruitment: implications for the theory of fishing. Canadian Journal of Fisheries and Aquatic Sciences, 58, 39-50 Werner, E.E. & Gilliam, J.F. (1984) The ontogenetic niche and species interactions in size-structured populations. Annual Reviews of Ecology and Systematics, 15, 393- 425. Werner, E.E. (1988) Size, scaling, and the evolution of complex life cycles. Size- structured Populations: Ecology and Evolution (eds B. Ebenman & L. Persson), pp. 60-81. Springer-Verlag, Berlin. Wilbur, H.M. (1988) Interactions between growing predators and growing prey. Size-structured Populations: Ecology and Evolution (eds B. Ebenman & L. Persson), pp. 157-172. Springer-Verlag, Berlin. Wilson, J. B. & Agnew, A. D. Q. (1992) Positive-feedback Switches in Plant Communities. Advances in Ecological Research, 23, 263-336.

Thanks

First of all, I want to thank my supervisors, Lennart and André. Working with you has been, to say the least, inspiring. I learned an amazing lot, not just technical details, but also, what really matters, how to do ecological research in a “no old science”-way. Your enthusiasm, your creativity was so contagious. I will always remember how we seriously discussed filtering a whole lake! However, supervisors provide only some part of a PhD students’ education. A lot comes from other students and Post-Docs, so I would like to thank especially Jens, Tobias, Karin and Zlatko for all the fun discussions we had and the help you gave me for understanding things, from experimental set-ups over fish biology and population modelling to statistics. The R-study group (Ullis, Lena, Johan, Zlatko) has to be mentioned, too, of course. Jens further helped me so much in getting started here, that I will be eternally grateful. But what is more, you guys have not only been great colleagues but also dear friends. Thanks for being yourself, Tobias (can one be more willing to have a beer?), Karin (can one be more eager for discussions?), Zlatko (can one be more focused and straight?), Irina, Ines, Martin (can one be more politically incorrect?), Carolyn (can one be more cheerful?), Magnus, Jenny, Per, Darius, and many more…! You made the sunny side of the street so much sunnier. And when things appeared to go down the drain, you helped me to pull through. Further thank goes to everyone, who ever helped me out with tips, material and experiments, especially to David for his hospitality and his immeasurably valuable advises on experimentation with poecilliids. I would like to say again thank you to Emma, Torunn, Krister, Christian and Mårten for help in the field and in the lab. You guys really made things happen! Sometimes you need a change of scenery to give your creativity a kick. I found this at two places here: Thanks, Kafé Schmäck and Brobergs Café for a relaxing and inspiring atmosphere (and good coffee); many of lines in this thesis have been written there. Finally, thank you to my four families for all the love and support: Karin, Lasse and Klaus in Cuxhaven, Birte and Jason in Oakland, Detlev, Roswitha and Lars in Erkrath and Volker and Inge in Berlin! And last but not least, Esther, thank you for being there!