Performances of six reanalysis profile products in the atmospheric correction of passive microwave data for estimating land surface temperature under cloudy-sky conditions

Abstract

Passive microwave (PMW) observations from the Advanced Microwave Scanning Radiometer 2 provide a way to obtain cloudy land surface temperatures (LSTs). However, atmospheric corrections must be performed on cloudy LSTs due to the cloud effect at higher frequencies. In this paper, six reanalyzed profiles, including the fifth-generation European Centre for Medium-range Weather Forecasts Reanalysis (ERA5), Interim Reanalysis (ERA-Interim), Japanese 55-year Reanalysis Data (JRA-55), Modern-Era Retrospective analysis for Research and Application V2 (MERRA2), National Centers for Environmental Prediction (NCEP) /Final Operational Global Analysis (FNL), and NCEP/Global Forecasting System (GFS), were compared with 2829 radiosonde profiles derived from the University of Wyoming. Then, their performances in correcting the atmospheric effects of LSTs at cloudy skies were investigated. Results showed that the ERA5 had the best accuracy in revealing the actual atmospheric conditions, and the RMSEs of transmittance, downward radiance, and upward radiance were about 0.007, 2.01, and 1.89 K, respectively. The RMSEs between the estimated LSTs and referenced LSTs varied from 3.15 K of the ERA5 to 6.12 K of the NCEP/FNL, indicating the ERA5 can be recommended for the atmospheric correction of PMW-based LST retrievals. Additionally, transmittance accuracy plays an essential role in impacting the LST retrievals in any weather.

Keywords

• reanalysis product
• land surface temperature
• atmospheric correction
• amsr2
• evaluation