Incorporating Surface Soil Moisture Information in Error Modeling of TRMM Passive Microwave Rainfall

This study assesses the significance of conditioning the error modeling of The National Aeronautics and Space Administration (NASA)´s Tropical Rainfall Measurement Mission Microwave Imager rainfall algorithm (2A12) to near-surface soil moisture data derived from a land surface model. The term “conditioning” means the model parameters´ dependence on soil wetness. The Oklahoma (OK) region is used as the study area due to its relatively low vegetation and smooth terrain and the availability of high-quality in situ hydrometeorological data from the Mesonet network. The study period includes two warm seasons (March to October) from 2009 and 2010. The National Oceanic and Atmospheric Administration/National Severe Storms Laboratory ground radar-based National Mosaic and Quantitative Precipitation Estimation system (NMQ/Q2) is used as high-resolution (5-min/1-km) reference rainfall. The surface wetness conditions (wet, dry, and normal) were determined from surface soil moisture fields simulated by the NASA Catchment Land Surface Model forced with Q2 rainfall fields. A 2-D satellite rainfall error model, SREM2D, is used to provide the ensemble error representation of 2A12 rainfall using two different parameter calibration approaches: conditioning the SREM2D parameters to the surface soil wetness categories versus not. The statistical analysis of model-generated ensembles and associated error metrics show better performance when surface wetness information is used in SREM2D. In terms of quantification, the ensemble rainfall from the conditional SREM2D parameter calibration shows better reference rainfall encapsulation. The conditioning of SREM2D to soil wetness can apply to rainfall rate estimates from other microwave sensors on board low Earth orbiting satellites and is valuable for the forthcoming missions on precipitation (Global Precipitation Measurement) and soil moisture (Soil Moisture Active Passive).