Scientific Opinion on good modelling practice in the context of mechanistic effect models for risk assessment of plant protection products

The tissue traits and architectures of plant species are important for land-plant ecology in two ways. First, they control ecosystem processes and define habitat and resources for other taxa; thus, they are a high priority for understanding the ecosystem at a site. Second, knowledge of trait costs and benefits offers the most promising path to understanding how vegetation properties change along physical geography gradients. There exists an informal shortlist of plant traits that are thought to be most informative. Here, we summarize recent research on correlations and tradeoffs surrounding some traits that are prospects for the shortlist. By extending the list and by developing better models for how traits influence species distributions and interactions, a strong foundation of basic ecology can be established, with many practical applications.

Statistical models are the traditional choice to test scientific theories when observations, processes or boundary conditions are subject to stochasticity. Many important systems in ecology and biology, however, are difficult to capture with statistical models. Stochastic simulation models offer an alternative, but they were hitherto associated with a major disadvantage: their likelihood functions can usually not be calculated explicitly, and thus it is difficult to couple them to well-established statistical theory such as maximum likelihood and Bayesian statistics. A number of new methods, among them Approximate Bayesian Computing and Pattern-Oriented Modelling, bypass this limitation. These methods share three main principles: aggregation of simulated and observed data via summary statistics, likelihood approximation based on the summary statistics, and efficient sampling. We discuss principles as well as advantages and caveats of these methods, and demonstrate their potential for integrating stochastic simulation models into a unified framework for statistical modelling. © 2011 Blackwell Publishing Ltd/CNRS.