DFG Research Training Group 1829 "Integrated Hydrosystem Modelling”: Bayesian multi-purpose modeling of processes in the soil-crop-atmosphere nexus on the landscape scale

The knowledge about the fate and behaviour of agrochemicals (nitrogen, pesticides) at the landscape scale is still limited. Observations of transformation rates in laboratory experiments are often in stark contrast to rates documented in natural systems, the heterogeneity of which is typically only vaguely known. Soil-plant-atmosphere processes are a key control for water and element cycles in the landscape and feedbacks are known to be strong. To describe these coupled processes, a spatially and temporally explicit, yet one-dimensional process model representation is adopted. For an encompassing representation of the associated spatio-temporal uncertainties, the connected collaborative research centre (CRC 1253 CAMPOS) will set up a stochastic modelling framework on the catchment scale. The framework will be complemented by Bayesian modelling to better describe the uncertainties of near surface processes, the system’s state space, soil and land use variability, the adequacy of modeling approaches, and to assimilate remote sensing data.