Evaluation of CMIP5 Earth System Models in Reproducing Leaf Area Index and Vegetation Cover over the Tibetan Plateau

• Published in: Acta meteorologica Sinica Vol. 28; no. 6; pp. 1041 - 1060
• Main Author: 鲍艳 高艳红 吕世华 王青霞 张少波 许建伟 李瑞青 李锁锁 马迪 孟宪红 陈昊 常燕
• Format: Journal Article
• Language: English
• Published: Heidelberg The Chinese Meteorological Society 01.12.2014; Key Laboratory of Land Surface Process and Climate Change in Cold and Arid Regions, Chinese Academy of Sciences, Lanzhou 730000%Key Laboratory of Land Surface Process and Climate Change in Cold and Arid Regions, Chinese Academy of Sciences, Lanzhou 730000; Earth System Modeling Center, Nanjing University of Information Science & Technology, Nanjing 210044
• Subjects: Atmospheric Protection/Air Quality Control/Air Pollution; Atmospheric Sciences; Earth and Environmental Science; Earth Sciences; Freshwater; Geophysics and Environmental Physics; Meteorology; 叶面积指数; 地球系统; 指数和; 植被模型; 植被覆盖; 模拟显示; 评估; 青藏高原; China, People's Rep., Xizang, Tibetan Plateau; China, People's Rep., Huang He R; Tibetan Plateau; vegetation cover; earth system model (ESM); dynamic global vegetation model (DGVM); Coupled Model Intercomparison Project Phase 5 (CMIP5)
• ISSN: 0894-0525; 2095-6037; 2198-0934; 2191-4788
• DOI: 10.1007/s13351-014-4023-5

The abilities of 12 earth system models（ESMs） from the Coupled Model Intercomparison Project Phase5（CMIP5） to reproduce satellite-derived vegetation biological variables over the Tibetan Plateau（TP） were examined.The results show that most of the models tend to overestimate the observed leaf area index（LAI）and vegetation carbon above the ground,with the possible reasons being overestimation of photosynthesis and precipitation.The model simulations show a consistent increasing trend with observed LAI over most of the TP during the reference period of 1986-2005,while they fail to reproduce the downward trend around the headstream of the Yellow River shown in the observation due to their coarse resolutions.Three of the models：CCSM4,CESM1-BGC,and NorESM1-ME,which share the same vegetation model,show some common strengths and weaknesses in their simulations according to our analysis.The model ensemble indicates a reasonable spatial distribution but overestimated land coverage,with a significant decreasing trend（-1.48%per decade） for tree coverage and a slight increasing trend（0.58%per decade） for bare ground during the period 1950-2005.No significant sign of variation is found for grass.To quantify the relative performance of the models in representing the observed mean state,seasonal cycle,and interannual variability,a model ranking method was performed with respect to simulated LAI.INMCM4,bcc-csm-1.1m,MPI-ESM-LR,IPSL CM5A-LR,HadGEM2-ES,and CCSM4 were ranked as the best six models in reproducing vegetation dynamics among the 12 models.