Evaluation of global terrestrial evapotranspiration in CMIP6 models

Evapotranspiration (ET), which relates to hydrology and the energy cycle process of land surface, is a key link of the water cycle and an important item of expenditure in energy balance. There have been multiple state-of-the-art climate models that were included in Phase 6 of the Coupled Model Intercomparison Project (CMIP6), but the performance in estimating the ET of the models is still unclear. Thus, this study evaluated the global terrestrial ET of CMIP6 models. The performance of the models and CMIP6 ensemble mean was compared with the GLEAM v3.3a dataset, and the uncertainties of the ensemble were evaluated using the signal-to-noise ratio (SNR). The results show that there was no perfect model that could perform optimally in all aspects of the comparison, and the performance of the CMIP6 ensemble was better than that of a single model. MIROC6, CESM2, and EC-Earth3 performed satisfactorily in some aspects, whereas the performance was poor in other aspects. GFDL-ESM4 exhibited a relatively poor performance. Most models and the CMIP6 ensemble overestimated ET, and the estimations of different models varied greatly, but the results of most models showed an increasing trend. The CMIP6 ensemble overestimated ET in most regions of the world and may have smoothened the variation in the model estimations. In high latitudes such as northern parts of North America and Eurasia, results of the CMIP6 ensemble and GLEAM were approximately the same. The uncertainty of the CMIP6 ensemble was generally low and the estimation reliability varied according to the geographical region.

Automated summary

This study evaluated the global terrestrial ET of CMIP6 models and found that no single model could perform optimally in all aspects of the comparison. The performance of the CMIP6 ensemble was better than that of individual models, but most models and the ensemble overestimated ET. The uncertainty of the CMIP6 ensemble was generally low, and the estimation reliability varied by geographical region.