Abstract
Accurate crop status forecasting benefits from assimilating remote sensing observations and crop model simulations. When conducting data assimilation (DA) using an Ensemble Kalman filter (EnKF), arbitrary inflation factors are normally adopted to account for unspecified uncertainties, thus avoiding filter divergence. Here, we developed a Bayesian methodology in which the uncertainties were systematically quantified by combining disparate methods in one framework. Its applicability and performance with crop model GECROS using the EnKF framework were tested against data collected from two years of field experiments, in which aboveground biomass (W above), grain weight (W grain), aboveground nitrogen (N) content (N above), grain N content (N grain) and leaf traits like leaf dry weight, leaf N content and leaf area index were measured for rice. Using only the observations from the first year, the uncertain parameters in GECROS were calibrated by a Markov Chain Monte Carlo approach, while the parameters in the assumed error model that describes the uncertainties of crop model simulations were estimated simultaneously. The calibrated model parameters performed well in the validation year, except for the simulated leaf traits (Normalized Root Mean Squared Error (NRMSE ) > 0.38). Remotely sensed leaf traits predicted by a Gaussian Process Regression (GPR) model were more accurate (NRMSE< 0.34), with uncertainties of the remote sensing observations estimated from the GPR model itself. Assimilating simulated and predicted leaf traits with their estimated uncertainties into EnKF prevented filter divergence, and the forecast accuracy of crop model improved in the validation year. Compared with simulation without assimilating in-season remote sensing observations, the NRMSE of updated whole-season W above andN above both decreased from 0.37 to 0.20; and those of updated W grain andN grain at harvest decreased from 0.40 and 0.28 to 0.22 and 0.19, respectively. The developed method contributes to systematic uncertainty analysis in DA and accurate forecasting of in-season and end-of-season crop carbon and N status for smart farming.