| Context: In environmental systems it is often very difficult to design predictive models by using physical concepts alone. This is especially the case in systems that are highly variable in both space and time and where many (potential) explanatory factors are involved. Examples of such systems are spreading and germination of seed, erosion and sedimentation, local movements of birds or insects. In our research we use some large datasets of the aforementioned processes, where many abiotic factors are available to explain few biologic processes. This research focuses on developing rigorous statistical tools to identify the key relationships in these systems and then develop stochastic dynamic models. Aim: Developing statistical methods to identify stochastic models for highly variable dynamic environmental systems. Methods: data analysis and statistical modelling. |
