Bayesian calibration of a grassland model under water stress conditions

Bayesian calibration of model parameters holds potential to reduce and understand the uncertainty surrounding model outputs. With focus on a complex model of grassland systems (PaSim – Pasture Simulation model), and under conditions of altered climate (i.e. precipitation reduction), Ben Touhami and Bellocchi showed that Bayesian calibration can help in reducing uncertainties associated with model parameters by updating a distribution function from a prior knowledge. The analysis performed offered insights on potential causes of the model deviation from the measured data (likely due to an incomplete knowledge of the species composition of grassland vegetation).

Posted in Uncategorized.