This study is a global assessment of the performance of two empirical models for estimating daily global solar radiation (SR) in the present-day climate, as well as the applicability of these models for near-future climates. One empirical model is used to estimate SR from a diurnal temperature range (DTR) and is called the DTR-SR model, while the other is used to estimate SR from relative humidity (RH) and is called the RH-SR model. Using data from the reanalysis and atmosphere-ocean coupled global climate model (GCM) known as MIROC4, we found that the DTR-SR model operates more accurately over land than the RH-SR model, though the RH-SR model outperformed the DTRSR model over the ocean. Based on an assessment using data from the decadal prediction performed by the GCM, the DTR-SR model could be applied to estimate GCM-simulated SR in altered climates until the year 2035 considered in this study, given the comparatively limited performance of the model during the winter and in arid and semi-arid regions. This suggests that the DTR-SR model is likely useful for data imputation if GCM-simulated SR data are unavailable.
ASJC Scopus subject areas