Comment on acp-2022-794

Abstract

<strong class="journal-contentHeaderColor">Abstract.</strong> Downward longwave radiation (DLR) affects energy exchange between the land surface and the atmosphere, and plays an important role in weather forecasting, agricultural activities, and the development of climate models. Because DLR is seldom observed at conventional radiation stations, numerous empirical parameterizations have been presented to estimate DLR from screen-level meteorological variables. The reliability and representativeness of parameterization depend on the coefficients regressed from the simultaneous observations of DLR and meteorological variables. Only a few previous studies have attempted to build parameterizations over regions in China such as the Tibetan Plateau and East China. In this study, a long-term (2011&ndash;2022) hourly dataset of DLR and meteorological elements, obtained from seven stations of the China Baseline Surface Radiation Network, was used to recalculate the coefficients of the Brunt and Weng models, and to develop a new model. Results showed that the mean bias error (MBE) and relative MBE (rMBE) between the measured clear-sky DLR and that estimated using the Brunt, Weng, and new models were &minus;4.3, &minus;5.1, and 3.7 W m^&minus;2 and &minus;1.5 %, &minus;1.8 %, and 1.3 %, respectively. The root mean squared errors (RMSEs) where in the range of 13.8&ndash;14.3 W m^&minus;2 and the relative RMSEs (rRMSEs) were approximately 5.0 %. The MBEs (rMBEs) of the Brunt, Weng, and new models under all-sky conditions were &minus;2.8 W m^&minus;2 (&minus;1.0 %), &minus;6.1 W m^&minus;2 (&minus;2.1 %), and &minus;1.5 W m^&minus;2 (&minus;0.5 %), respectively. The RMSE (rRMSE) of the parameterization models in retrieving all-sky DLR was ~17.5 W m^&minus;2 (~6.1 %). Therefore, the models are considered suitable for retrieval of DLR over China.