Improvement of the Rain/No-Rain Classification Method for Microwave Radiometers Over the Tibetan Plateau

摘要: This letter identifies causes of the deterioration Global Satellite Mapping Precipitation rain detection over Tibetan Plateau during summer monsoon season. Using rain/no-rain classification (RNC) method Plateau, observed brightness temperature (Tb) at 21 (23) GHz from Tropical Rainfall Measuring Mission Microwave Imager (the Defense Meteorological Program Special Sensor Imager) [Tb(21 V) and Tb(23 V)], surface emissivity ( $\varepsilon )$ was substituted for temperatures (Ts) to exclude areas low Ts as snow cover because it is difficult distinguish between scattering signals precipitation cover. A case study demonstrates that systems are excluded often below threshold due use an inadequate value $ (constant throughout year), even though a signal high-frequency channels under no evident. After replaced with values monthly mean satellite observation-based land database, rainfall improved. In addition, suggested database RNCs should also consider diurnal variations in Tb(21 large differences Plateau.