MODELKEY Workpackage Effect3: Canonical community and food chain effect modelling Required amount of research work: 4 year PhD plus 2 year Postdoc Motivation: The translation of effects of chemical compounds from observations in the laboratory in a set of single-species toxicity experiments to expected effects in the field, given a specified emission, is far from straight foreword. The reason is that the biological, chemical and physical situation is immensely more complex in the field, while the scales in space and time are also widely different. The environment is variable in time and spatially inhomogeneous, the compound has a complex dispersal and transformation behavior, biological species are present whose sensitivities are unknown, and processes such as population dynamics, species interactions, adaptation and selection are involved that have not been studied in the laboratory under controlled conditions. This list is just a sample of the problems that complicate the translation of observed effects in the laboratory to expected effects in the field. Yet, this translation is vital to proper environmental management, while the theoretical/conceptual tools we have available are very limited indeed. This workpackage aims at contributing to the translation toolbox in two respects: - the development of appropriate concepts to quantify effects in very simple theoretical communities; this knowledge will be helpful to recognize effects in the field, where feedbacks and species interactions complicate the recognition of small effects on integrated systems. - the modelling of the flux of compounds through a realistic food chain, and the effects of these compounds; this knowledge will be helpful to interpret the field data that are colleted in the SITE WP's. Specification: Within WP Effect3 we have two related studies: the theoretical community and the more practical food chain study. Community study: The purpose of this modelling study is to evaluate how effects in well-integrated ecosystems would show up. The study will deal with the problem how to translate toxic effects on individuals to that on ecosystems, including the compensating adaptations that will occur if one or more of the functional aspects of the ecosystem will be affected. To study these mechanisms in a systematic way, it is essential to simplify the ecosystem to an extend that is much simpler than the field situation. We will use a theoretical metaphor for this purpose: the canonical community; this is a theoretical community that consists of three groups of organisms (producers, consumers and decomposers) in a homogenous environment that is closed for mass, and open for energy (light comes in, and temperature is constant). The word canonical means "the simplest representative that still has all essential properties"; it is widely used in mathematics. Dynamic Energy Budget (DEB) theory will be used to structure the model and to quantify all rates (Kooijman, 2000, 2001; Kooijman et al., 2003). Much research is already invested in the theory of canonical communities (Kooijman & Nisbet, 2000; Kooijman et al., 2002), but the application to the evaluation of toxic effects will be new. The DEB theory relates effects of toxicants to internal concentrations via changes of appropriate parameter values; the time it takes for an organism to equilibrate its internal concentration of some compound with the environment increases with its body size. Although the community study does not aim to mimic field situations, there is still a need for biological realism of the simplified system. Therefore we specify the species (although the DEB theory applies to all species), and select appropriate parameter values. Moreover we will try to understand the effects in the biofilms that are studied in SITE 2. To this end we will organize a close collaboration with ECT (Dr. Knacker, SME) and USZ (Ms Dr. Schmidt, SITE 2) to link experimental and theoretical work, starting with a kick-off crash course in DEB theory in Amsterdam to ensure a sound theoretical foundation for experimental work, followed by working visits during the research. The first choice of canonical communities is: nitrogen (as DIN, a frequently limiting nutrient), bacteria (as decomposers), alga (as producers), paramecium (a ciliate that consumes algae). After a analysis of the behavior of this system in the blanc, an in response to toxic stress, we extend the system in two steps by introducing didinium (a ciliate that preys on paramecium), and/or a rotifer (Brachiones rubens, which competes with paramecium). These organisms are all of small body size, which allow to test model predictions at the ecosystem level in dynamic detail. It also means that the process of toxicokinetics is less important, because we can assume that the internal concentration is fast in equilibrium with the external one. This combination of species is well-studied in the literature (e.g. Holyoak, 2000; Doucet & Maly, 1990). We will extend the choice of species with dominant ones in biofilms and estimate the parameters on the basis of a series single-species toxicity tests that are generated in USZ and ECT, judge the goodness of fit, and adjust modelling details where necessary, select the appropriate parameter values and compare computer simulations of the canonical communities with data from experimental communities. We aim at closed systems (with full nutrient recycling), but if these turn out to be experimentally too unstable, we will use chemostat setups. We hope to find out in this study how integrated ecosystems will respond to toxic stress imposed by chemical compounds with a variety of modes of actions (such as an increase of maintenance requirements, or a reduction of growth rate, or of the nutrient/food uptake rate). In other words: we want to develop a search image for mild forms of toxic stress in ecosystems, that allows us to recognize such stress in real-world ecosystems. Functional aspects (i.e. nutrient recycling) depend on structural aspects (biomass amounts in the various trophic levels), but how mild toxic effects will show up in both aspects is presently not known. Complex feedback mechanisms in these integrated systems make application of a modelling framework essential. Food chain study: The purpose of this study is to assist the interpretation of the food chain data generated in SITE 1, 3, 4 and 5, in the light of the qualitative results that come from the community study. We focus on a benthos-feeding fish as top predator, benthic macro-invertebrates, algae and bacteria. Since we here deal with organisms of larger body size, so toxicodynamics is more important than in the canonical community study. The choice of the specific species and chemical compounds will be made consistent with choices in the SITE WP's. We will model the flux of compounds through the food chain in the framework of the DEB theory (Kooijman, 2000). This includes the incorporation of the various uptake and elimination routes for the compound, and how this is modified by the nutritional state of the organisms (lipid content; body size and lipid content tend to increase with food chain node); this effort will be linked to WP EXPO 3. We will intensively collaborate with setting up a shared data base for data from literature, field measurements and experimental data. Our modelling effort has elements in common with EXPO 3, but here we will study dynamics aspects (changing population sizes, and species interactions). Moreover, toxic effects will be taken into account, that are based on internal concentrations. To this end, we will analyse a set of bioassay from ECT on toxicokinetics and effects, using the Software package DEBtox (Kooijman & Bedaux, 1996) to extract the required parameter values. We still evaluate the goodness of fit, and adjust the model details if required. The will then use these parameters to interpret the data from the field, in the light of the results of the community study. In a later stage of the project we will try to link the biological effect module to the exposure module to be developed in EXPO 4. One the basis of this study, we hope to be able to interpret the field data, and evaluate the strength of the toxic stress. The strength of this approach is - effects of toxicants on sublethal endpoints are weighed with respect to the overall effect on a (simple) community - effect studied are fully integrated with biodegradation studies. - effects are followed during exposure; this allows the evaluation of peak emissions versus dispersed continuous emissions Deliverables: Month 1 - 18: 1) Literature study on theory for and experiments with simple closed communities. 2) Collection of data for eco-physiology of relevant species and toxico-kinetics and effects of target toxicants; both from literature and from experiments at ECT and USZ. 3) Computer simulation studies of effects of toxicants with various modes of action in canonical communities. 4) Computer simulation studies of toxico-kinetics and effects in food chains, using parameter values of selected target organisms and toxicants in the SITE studies. Month 19 - 36: 5) Parameter estimation via model application on experimental results of ECT and USZ. 6) Model studies of extensions of canonical communities with (simple) food webs (predators), and competing producers. 7) Simulation of effects of target compounds in biofilm dynamics. 8) Creating theory that quantifies the effects of toxicants on the link between ecosystem structure (biomass distribution) and function (nutrient recycling). 9) Creating theory for effects of toxicants in food chain that is based on internal concentrations that can vary in time. Month 37 - 60: 10) Linking a trimmed effect module to the exposure module developed in EXPO 4 11) Linking between theory and experimental and in situ data of SITE WP's. 12) Continuation of the systematic analysis of possible effects of toxicants in extensions of canonical communities. Comparison of results with data and concepts in the literature. 13) Evaluation of our fundamental understanding of observed effects of toxicants on biologically integrated systems. References: Kooijman, S. A. L. M., Auger, P., Poggiale, J. C. and Kooi, B. W. (2003) Quantitative steps in symbiogenesis and the evolution of homeostasis. Biol. Reviews, 78: 435--463 Kooijman, S. A. L. M., Dijkstra, H. A. and Kooi, B. W. (2002) Light-induced mass turnover in a mono-species community of mixotrophs. J. Theor. Biol. 214: 233--254 Kooijman, S. A. L. M. (2001) Quantitative aspects of metabolic organization; a discussion of concepts. Phil. Trans. R. Soc. B, 356: 331--349 Kooijman, S. A. L. M. and Nisbet, R. M. (2000) How light and nutrients affect life in a closed bottle. In: J{\o}rgensen, S. E. (ed) Thermodynamics and ecological modelling. CRC Publ., Boca Raton, FL, USA, pp: 19--60 Kooijman, S. A. L. M. (2000) Dynamic Energy and Mass Budgets in Biological Systems. Cambridge University Press, pp 424 Kooijman, S. A. L. M. and Bedaux, J. J. M. (1996) The analysis of aquatic toxicity data. VU University Press, Amsterdam, pp 149 Doucet, C. M and Maly, E. J. (1990) Effect of copper on the interaction between the predator Didinium-nasutum and its prey Paramecium-caudatum. Can J Fish Aquat Sci 47: 1122 - 1127 Holyoak M., Lawler S.P. and Crowley, P.H. (2000) Predicting extinction: Progress with an individual-based model of protozoan predators and prey Ecology 81: 3312 - 3329 ------------------------ Experimental efforts that will help the modelling in WP EFFECT3 Species and compound selection: consistent with SITE WP's For food chain modelling we need bioassays on toxic effect on - biodegradation + effects on heterotrophic bacteria (co-metabolism of degradation of toxic compounds and other organic matter) - algal growth - benthic invertebrate and fish: survival, feeding, growth and reproduction, in relation to internal concentrations. So we need to study toxico-kinetic here, and quantify the different uptake and elimination routes For canonical community modelling: - effects of the compound on survival, growth and reproduction of paramecium, didinium and brachiones (assuming that the effects on the alga is included in the food-chain package), and a choice of dominant species in the biofilms studies in SITE 2. Effects of food levels on energetics and toxic effects will be integrated in the study (since they are basic to the DEB theory) What we need here is the combination between toxico-kinetics and effects through exposure time, especially for the large specimens (benthic invertebrate, fish). The required length of the experiments depend on the choice of the compound, species and sizes of the individual. We need to follow the process till the internal concentrations are in equilibrium. The elimination rate is very important to know. -------------- SITE1: Establishment of river basin specific data base (Eric de Deckere, Burkhard Stachel, Sergi Sabater) SITE2: Pollution induced tolerance of microbenthic communities (Mechthild Schmitt) SITE3: Macroinvertebrate communities (Eric de Deckere) SITE4: Fish community (Helmut Segner) SITE5: Chemical assessment of biomagnification and bioaccumulation (Anton Kocan, Jan Petrik) KeyTox1: Sampling and extraction (Kevin Thomas) KeyTox2: Toxicant isolation (Werner Brack) KeyTox3: Toxicant identification(Pim Leonards) KeyTox4: Bioassays (Marja Lamoree) KeyTox5: Database (Joop Bakker) KeyTox6: Identification of river-basin specific key toxicants (? maybe part of KeyTox5?) EXPO1: modelling sedimentation rates and sediment stability (Bernhard Westrich) EXPO2: Bioavailability assessment (Jussi Kokkonen) EXPO3: Bioaccumulation and food chain assessment (Bert van Hattum) EXPO4: Integration and application of exposure formulations and transport modelling (Arthur Baart) EFFECT1: Models for sensitivity analysis of toxic substances (Sovan Lek) EFFECT2: Integrative modelling of the effect of toxic substances on community patterns (Dick de Zwart) EFFECT3: Canonical communities (Bas Kooijman) DECIS1-5: Decision making (Claudio Carlon). WP leaders unclear. -------------