Anatomy and assessment of surface water and energy balance components simulated by CMIP6 models in Pan Third Pole

• Published in: Journal of hydrology (Amsterdam) 2025-05, Vol.652, p.132656, Article 132656
• Main Authors: Liu, Zhu, Huang, Bohan, Liu, Su, Fan, Xuewei, Yin, Jina, Duan, Qingyun
• Format: Article
• Language: English
• Subjects: CMIP6; Cryosphere; Energy balance; Model bias; Pan Third Pole; Water balance
• ISSN: 0022-1694
• DOI: 10.1016/j.jhydrol.2024.132656

•Overall, CMIP6 models simulate energy balance components better than water balance components.•Runoff is simulated the worst and the relative bias is around −40 %∼60 % in PTP.•Long term mean seasonal water storage change and ground heat flux change are not negligible.•Snow and frozen soil play significant roles in the seasonal imbalance of water and energy in PTP. The Pan Third Pole (PTP) contains Tibet Plateau and its surrounding regions such as Hindu Kush, Karakoram and Pamirs and it is one of the highest and most fragile as well as rapidly developing regions of the Earth. Besides, PTP origins Yangtze River, Yellow River and Lancang-Mekong River, which feed millions of people downstream. Therefore, climate change in PTP has significant impact on livings and water supply of local residents. The surface thermal condition constrains the evolution of climate and provides the basic driving force for the land surface processes such as evapotranspiration as well as snowmelt and sublimation, which significantly affect the water balance and its partition. In this study, simulations of water balance components including precipitation, evapotranspiration, runoff and surface energy balance components including net radiation, latent heat and sensible heat from 20 CMIP6 models are evaluated with ERA5-Land reanalysis data during 1981–2010 over PTP region and its three subregions. Specifically, the annual and seasonal relative bias of different components as well as water and energy balance are evaluated. The novelty of this study is that we conduct a comprehensive evaluation of both annual and seasonal characteristics of all major water and energy balance components simulated by CMIP6 models rather than any individual component as well as investigate whether the water and energy cycle is still closed by the raw biased key variable simulations from each model. In addition, the potential influence of melting snow and frozen soil to water and energy balance is assessed. Results show that the spatial patterns of relative bias of water and energy components are quite different among models and significant bias exist for seasonal model simulations. In general, CMIP6 models simulate water balance worse than energy balance especially for winter and runoff has larger relative bias (relative bias ranges from −40 %∼60 %) compared with precipitation and evapotranspiration. Moreover, water and energy imbalance reflected by long term mean water storage change and ground heat flux c