Assessment of actual evapotranspiration variability over global land derived from seven reanalysis datasets

• Published in: International Journal of Climatology
• Main Authors: Feng, Taichen, Su, Tao, Zhi, Rong, Tu, Gang, Ji, Fei
• Other Authors: National Natural Science Foundation of China, Postdoctoral Science Foundation of Jiangsu Province
• Format: Article in Journal/Newspaper
• Language: English
• Published: Wiley 2019
• Subjects: Atmospheric Science; Arctic; Global warming
• Online Access: http://dx.doi.org/10.1002/joc.5992; https://api.wiley.com/onlinelibrary/tdm/v1/articles/10.1002%2Fjoc.5992; https://onlinelibrary.wiley.com/doi/pdf/10.1002/joc.5992; https://onlinelibrary.wiley.com/doi/full-xml/10.1002/joc.5992; https://rmets.onlinelibrary.wiley.com/doi/pdf/10.1002/joc.5992

With the motivation to identify whether global warming would lead to an increase in the rate of actual evapotranspiration (AE) over land, an assessment is made of AE variability from an ensemble of seven reanalyses (NCEP‐NCAR, NCEP‐DOE, MERRA, MERRA V2, ERA‐Interim, JRA‐55, and CFSR). By subdividing global land into nine climatic regions using the aridity index (AI), we examine the variability and long‐term trend of regional AE and various related factors. Results indicate that AE shows a significant increasing trend during 1979–2015 in the humid, wet, and arctic regions. Overall, three main parameters, namely, precipitation, surface net solar radiation, and wet‐day frequency, strongly influence the temporal variations of AE over most regions. The possible causes of trends in AE are examined in view of the linear trends of the related meteorological variables. Accordingly, increasing potential evapotranspiration and increasing vapour pressure are the main contributors to increasing AE in humid regions. Increasing temperature is the main contributor to increasing AE in wet regions. Both increasing energy supply and water supply on AE largely explain the significant increasing trend of AE in arctic regions. In view of data availability, a correlation analysis was conducted between AE in individual reanalyses and the results from the Mezentsev–Choudhury–Yang equation and the ensemble of the seven reanalyses. On average, NCEP‐NCAR is the worst among the seven reanalysis datasets in revealing temporal characteristics of AE.