| Global warming is one of the most significant changes in the Earth’s system,while the Earth’s surface warming rate by external forcing shows a geographical-dependent feature.The Arctic is the area with the fastest warming worldwide.Meanwhile,the warming amplification,or greater surface warming than the global average,occurs in the Tibetan Plateau(TP)and has resulted in melting permafrost,retreating glaciers,and decreasing biodiversity,which has a significant impact on the ecological environment.Many studies focused on the Arctic warming mechanisms;however,there is a lack of systematic research on what controls TP warming amplification.This study examines the major drivers of TP warming amplification in recent decades(1979–2020)and under future climate scenarios(2061–2100)by applying multiple radiative kernels and local energy budget diagnosis based on state-of-the-art reanalysis datasets and the Coupled Model Intercomparison Project Phase 6(CMIP6)data.The main conclusions are as follows:Based on observational analyses(1979–2020),the results show that TP warming amplification exists in all seasons with seasonal differences in amplitude.The TP warming rate in winter is larger than that in summer.Both the snow-albedo feedback and cloud-radiation feedback are the major contributors to the seasonality of TP warming.The snow cover and albedo decreased significantly in winter but showed only small changes in summer.Meanwhile,the total and low-level cloud cover increased(decreased)in summer(winter),leading to cooling(warming)of the TP in summer(winter).The attribution analyses based on the pi Control and Historical simulations of CMIP6 models confirmed that TP warming is caused by external forcing.Because the external forcing associated with anthropogenic activity is increasing,TP warming will continue to the end of the 21 st century.Under future(2061–2100)warming scenarios,the TP shows a higher warming rate than 88% regions worldwide.The results of radiative kernel diagnosis suggest that surface albedo feedback is the leading contributor to future TP warming during winter due to the reduction in snow cover and albedo over the TP.Under SSP2-4.5(SSP5-8.5)scenario,the TP warming induced by surface albedo feedback is about 38.9%(37.5%)of TP total warming.Although CMIP6 models projected a consistent signal of future TP warming,the amplitudes of warming magnitude are diverse.The uncertainty of TP warming during winter is about 0.92(0.74)times the TP warming magnitude projected by the CMIP6 ensemble mean under the SSP2-4.5(SSP5-8.5)scenario.Based on the emergent constraint method and the key physical process(snow-albedo feedback)of TP warming,this study further reduces the uncertainty of projected TP warming.Under the SSP2-4.5(SSP5-8.5)scenario,the uncertainty of future projected TP warming reduces by 39.8%(30.3%).This is mainly due to the systematic cold bias of TP surface air temperature projected by CMIP6 models,which leads to the underestimation of the local melting rate of snow over TP and thus the TP warming rate. |