From adaptive behavior to communities and ecosystems in a changing world: Challenges for individual-based modeling in the next years

Background/Question/Methods A main task of current ecology is to understand and predict the effects of environmental change on biodiversity and its role for ecosystem services. Existing descriptive, empirical, and highly aggregated models are of limited use for predicting the effects of unprecedented environmental conditions. Thus, mechanistic models are needed where low-level processes are represented and tested for a wide range of environmental conditions. This includes individual-based models (IBMs), which represent local interaction, adaptive behavior and decision making. IBMs are now an established tool in population ecology. However, IBMs are still, with the exception of forest ecology, not used very often to address questions regarding communities and ecosystems. The main reason for this might be the notion that the complexity of IBMs, which already is a challenge for population IBMs, has to be multiplied for communities and ecosystems so that the resulting models cannot be handled any more.

Results/Conclusions I will present examples of successful IBMs of communities and ecosystems. The most important modeling strategies employed so far are: modeling functional groups instead of species and focusing on key species. I will list the challenges for putting these and further new approaches into practice. One of the most important challenges is to develop models that allow including both plants and animals in a generic way. Approaches like the field-of-neighborhood (FON) approach have this potential. In conclusion, I will argue that we badly need fully-fledged IBMs of communities and ecosystems, but that we also need simpler models in order to explore the significance of individual-based aspects.