Estimating All-Weather Surface Longwave Radiation from Satellite Passive Microwave Data

Surface longwave radiation (SLR) is an essential geophysical parameter of Earth’s energy balance, and its estimation based on thermal infrared (TIR) remote sensing data has been extensively studied. However, it difficult to estimate cloudy SLR from TIR measurements. Satellite passive microwave (PMW) radiometers measure under the clouds therefore can in all weather conditions. We constructed retrieval models using brightness temperature (BT) Advanced Microwave Scanning Radiometer 2 (AMSR2) a neural network (NN) algorithm. European Centre for Medium-Range Weather Forecasts Reanalysis v5 (ERA5) product was used as reference. NN-based were able reproduce well spatial variability ERA5 at global scale. Validations indicate reasonably good performance found land sites, with bias 1.32 W/m2, root mean squared error (RMSE) 35.37 coefficient determination (R2) 0.89 AMSR2 surface upward (SULR) data, −2.26 RMSE 32.94 R2 0.82 downward (SDLR) data. SULR SDLR accuracies higher oceanic biases −2.98 −4.04 RMSEs 6.50 13.42 values 0.83 0.66, respectively. This study provides solid foundation development PMW model applicable scale generate long-term continuous products multi-year satellite future research spatiotemporal resolution.