Does ERA5 outperform satellite products in estimating atmospheric downward longwave radiation at the surface?

Abstract

Atmospheric downward longwave radiation (DLR) is a key component of the surface energy budget in the Earth system. Satellite retrievals and atmospheric reanalysis estimates are the two typical approaches to obtaining a spatio-temporally continuous DLR product. In this study, we evaluated the DLR product from the latest ERA5 atmospheric reanalysis and the well-known Clouds and Earth's Radiant Energy System (CERES) satellite retrievals, against high-quality observations collected at 46 Baseline Surface Radiation Network (BSRN) stations over land surfaces and at 9 Global Tropical Moored Buoy Array (GTMBA) buoy stations. The accuracy of the ERA5 DLR product over land was found to be higher on average than that of CERES at hourly to monthly time scales. Conversely, ERA5 performed slightly worse than CERES-SYN when estimating DLR over the ocean surface. This is the first time that atmospheric reanalysis has performed better than satellite retrievals in estimating DLR over the land surface, demonstrating the potentially extensive application prospects for ERA5 as well as setting new challenges for quantitative remote sensing research.