Technical Note: Calibration and validation of geophysical observation models

[摘要] We present a method to calibrate and validate observationalmodels that interrelate remotely sensed energy fluxes togeophysical variables of land and water surfaces. Coincidentsets of remote sensing observation of visible and microwaveradiations and geophysical data are assembled and subdividedinto calibration (Cal) and validation (Val) data sets. EachCal/Val pair is used to derive the coefficients (from the Calset) and the accuracy (from the Val set) of the observationmodel. Combining the results from all Cal/Val pairs providesprobability distributions of the model coefficients and modelerrors. The method is generic and demonstrated usingcomprehensive matchup sets from two very differentdisciplines: soil moisture and water quality. The resultsdemonstrate that the method provides robust model coefficientsand quantitative measure of the model uncertainty. Thisapproach can be adopted for the calibration/validation ofsatellite products of land and water surfaces, and theresulting uncertainty can be used as input to dataassimilation schemes.

[学科分类] 地球化学与岩石