Bayesian integration of flux tower data into process-based simulator for quantifying uncertainty in simulated output

This research implemented a Bayesian statistical method to calibrate a widely used process-based simulator BIOME-BGC against estimates of gross primary production (GPP) data. Six parameters of BIOME-BGC were calibrated, which were also allowed to vary month-by-month to investigate the hypothesis that the phenology exhibited a seasonal cycle that was not accurately reproduced by the simulator. Time varying parameters substantially improved the simulated GPP as compared to GPP obtained with constant parameters.

Process-based simulator, BIOME-BGC, Gross primary production, Bayesian calibration, uncertainty estimation