Under the Coordinated Regional Climate Downscaling Experiments‐East Asia (CORDEX‐EA‐II), the outputs from two regional climate models (RCMs) driven by four global climate models (GCMs) are used to investigate the characteristics and possible mechanisms of the projected elevation‐dependent warming (EDW) over the Tibetan Plateau (TP) under the Representative Concentration Pathway emission scenario 8.5 (RCP8.5). Results show that widespread warming over the TP is projected with considerable disagreements in warming intensity and the maximum warming center among RCMs. The largest spread in the surface air temperature (T_as) projections is found above 5,000 m, indicating that a large uncertainty exists over the higher elevations. A marked EDW signal over the TP is simulated under the RCP 8.5 by the multi‐RCM ensemble mean for all seasons, particularly in autumn. Based on the analysis of the surface energy budget, it is found that the surface albedo feedback (SAF) is the primary contributor to EDW and acts as the main source of uncertainty in EDW projections among RCMs. The downward longwave radiation (DLW) is found to be the dominant factor in regulating T_as change over the TP, and its contribution to EDW is model‐dependent. Furthermore, the structure and magnitude of projected EDW are sensitive to the RCM physics and driving GCM, as they can alter the projections of snow cover and albedo, which modulate the simulated SAF and its effect on EDW. Additionally, RegCM4 shows a higher sensitivity to the anthropogenic greenhouse forcing than WRF, evidenced by the larger temperature projections and stronger EDW signal in RegCM4.

Using the results from three global climate models (GCMs) and seven regional climate models (RCMs), summer monsoon climate changes during 2041–2060 over Indian Peninsula are projected based on the Intergovernmental Panel on Climate Change A1B emission scenario. For the control climate of 1981–2000, most nested RCMs can improve the temporal‐spatial distributions of temperature and precipitation over Indian Peninsula compared to the driving GCM of European Centre/Hamburg Fifth Generation (ECHAM5). Most nested RCMs produce advanced monsoon onset for control climate, which is similar to the result of driving GCM of ECHAM5. For future climate widespread summer warming is projected over Indian Peninsula by all climate models, with the Multi‐RCMs ensemble mean (MME) temperature increasing of 1°C to 2.5°C and the maximum warming center located in northern Indian Peninsula. The disagreement in precipitation changes projected by RCMs indicates that the surface climate change on regional scale is not only dominated by the large‐scale forcing which is provided by driving GCM but also sensitive to RCM' internal physics. Overall, wetter condition is shown in MME with significant increase of monsoon rainfall over southern India, with intermodel spread ranging from −8.9% to 14.8%. Driven by same GCM, most RCMs project advanced monsoon onset while delayed onset is found in two Regional Climate Model (RegCM3) projections, indicating uncertainty can be expected in the Indian Summer Monsoon onset. All climate models except Conformal‐Cubic Atmospheric Model with equal resolution (referred as CCAMP) and two RegCM3 models project stronger summer monsoon during 2041–2060.

To assess the influence of future land use and land cover change (LULCC) on temperature projections over East Asia, and quantify its contribution relative to the increased greenhouse gas (GHG), two simulations, with and without LULCC, are performed by 4th regional climate model (RegCM4) driven by two global climate models for the year 2041–2060 under the Representative Concentration Pathway4.5 (RCP4.5) scenario. Results show that LULCC induces about the changes of 0.1–0.3 °C in surface air temperature over areas undergoing the conversion between forest and crop with clear regionality and seasonality. The responses of extreme temperature are more sensitive to LULCC than those of mean temperature, especially for the daily minimum temperature (Tmin)-related indices. Disagreements in the temperature responses between the two models are simulated during most of the years, particularly over east China during summer. Such uncertainties can be explained by the diversity responses of available energy, which is related to large scale atmospheric circulations produced by the RCM. The extent of LULCC can determine both the magnitude and sign of changes in temperature. LULCC-related changes of Tmax and Tmin can be as large as 25% of those by GHG. LULCC contributes less than 10% to the total projected warming under RCP4.5 when averaged over sub-regions but can reaches up to about 30% on local scales. Those results highlight the important role of LULCC in projecting the regional to local scale temperature changes, and emphasize the necessity of including LULCC forcing in future regional modeling studies.

To explore the driving mechanisms of elevation-dependent warming (EDW) over the Tibetan Plateau (TP), the output from a suite of numerical experiments with different cumulus parameterization schemes (CPs) under the Coordinated Regional Climate Downscaling Experiments-East Asia (CORDEX-EA-II) project is examined. Results show that all experiments can broadly capture the observed temperature distributions over the TP with consistent cold biases, and the spread in temperature simulations commonly increases with elevation with the maximum located around 4000–5000 m. Such disagreements among the temperature simulations could to a large extent be explained by their spreads in the surface albedo feedback (SAF). All the experiments reproduce the observed EDW below 5000 m in winter but fail to capture the observed EDW above 4500 m in spring. Further analysis suggests that the simulated EDW during winter is mainly caused by the SAF, and the clear-sky downward longwave radiation (LW_clr) plays a secondary role in shaping EDW. The models’ inability in simulating EDW during spring is closely related to the SAF and the surface cloud radiative forcing (CRFs). Furthermore, the magnitude and structure of the simulated EDW are sensitive to the choice of CPs. Different CPs generate diverse snow cover fractions, which can modulate the simulated SAF and its effect on EDW. Also, the CPs show great influence on the LW_clr via altering the low-level air temperature. Additionally, the mechanism for different temperature changes among the experiments varies with altitudes during summer and autumn, as the diverse temperature changes appear to be caused by the LW_clr for the low altitudes while by the SAF for the middle-high altitudes.