Calibration and Verification of Land Surface Temperature Anomalies Derived from the SSM/I

Abstract

The current network of internationally exchanged in situ station data is not distributed evenly nor densely around the globe. Consequently, the in situ data contain insufficient information to identify fine spatial structure and variations over many areas of the world. Therefore, satellite observations need to be blended with in situ data to obtain higher resolution over the global land surface. Toward this end, the authors calibrated and independently verified an algorithm that derives land surface temperatures from the Special Sensor Microwave/Imager (SSM/I). This study explains the technique used to refine a set of equations that identify various surface types and to make corresponding dynamic emissivity adjustments. This allowed estimation of the shelter height temperatures from the seven channel measurements flown on the SSM/I instrument. Data from first?order in situ stations over the eastern half of the United States were used for calibration and intersatellite adjustment. The results show that the observational difference between the in situ point measurements and the SSM/I?derived areal values is about 2°C with statistical characteristics largely independent of surface type. High?resolution monthly mean anomalies generated from the U.S. cooperative network served as independent verification over the same study area. This verification work determined that the standard deviation of the monthly mean anomalies is 0.76°C at each 1° ? 1° grid box. This level of accuracy is adequate to blend the SSM/I?derived temperature anomaly data with in situ data for monitoring global temperature anomalies in finer detail.