Prediction of the biogeochemical cycles of even-aged forest ecosystems by a model coupling growth and yield, process-based and biogeochemical approaches
Résumé
Essentially, forest ecosystems are complex systems. Their study holds several distinct scienti c elds leading to varied conceptual approaches. However, environmental issues and sustainable development policies are pushing managers to understand the functioning of these ecosystems in all their components and to overcome any conceptual boundary. Thus, it is necessary to identify and quantify the interactions between many processes by combining modeling and statistical analysis. The objective of our work is to propose a dynamic model of forest growth and biomass production suitable to varied ecosystems (di erent species and soil and climate growth conditions) and forest managements. This model combines di erent approaches (growth and yield, process-based and biogeochemical cycles) to take into account carbon, water and nutrient cycles and to include several processes such as wood production, transpiration, litterfall, litter decomposition or losses of nutrients by drainage. Such a model integrates di erent concepts. The most important motivation is to propose a mutual enrichment instead of a simple juxtaposition of di erent equations. Such a model is necessary to anticipate and adapt forest management under di erent environmental scenarios (global changes). The study of such complex soil/plant interactions pushes the modeler to integrate, and even construct, varied mathematical analysis tools such as: design and analyse dynamic systems, model calibration, analyse and minimize the propagation of the associated error, data reconstruction and implementation of numerical methods. Then, the combination with other scienti c elds becomes possible in this mathematical framework, because even complex, the conception of such model rely on rigorous and synthetic methods of analysis and investigation of the forest ecosystem functioning. This approach can also be used to highlight unexpected behavior between di erent scales of processes and can provide a hierarchical analyses of the main processes at a given scale.