Understanding the surface temperature cold bias in CMIP5 AGCMs over the Tibetan Plateau

Recent studies have demonstrated that the majority of the Phase-5 Coupled Model Inter-comparison Project (CMIP5) models underestimate annual and seasonal mean surface air temperatures (Ta) over the Tibetan Plateau (TP). In addition, half of the models underestimate annual and seasonal mean surface temperatures (Ts) over the TP. These cold biases are larger over the western TP. By decomposing the Ts bias using the surface energy budget equation, this study investigates the contributions to the cold surface temperature bias on the Tibetan Plateau from various factors, including the surface albedo-induced bias (SAF), surface cloud radiative forcing (CRF), clear-sky shortwave (SW) radiation, downward clear-sky longwave radiation (DLR), surface sensible heat flux and latent heat flux, and heat storage. The results suggest that SAF and DLR are the main factors causing the cold surface temperature bias. Because SAF and DLR are respectively affected by the snow coverage fraction and water vapor distribution produced by the models, these results then imply that the snow coverage fraction parameterization and water vapor distribution over the TP require further improvements.