Using Geo‐detector to attribute spatio‐temporal variation of pan evaporation across China in 1961–2001

Declining pan evaporation (Epan) across many continents in the past decades despite global warming is gaining attention of hydrology and climate science communities. While Geo‐detector is considered useful for attributing the spatio ‐ temporal variation of a certain geography‐related element, it has not been applied for Epan attribution. Based on the Geo‐detector, we first attribute the spatio ‐ temporal variation of Epan across China in 1961–2001. The results show that relative humidity is the primary factor influencing the temporal dynamics of Epan, followed by sunshine duration, maximum air temperature and wind speed. In the northwest China (south China and east China), wind speed, air temperature and relative humidity (relative humidity, sunshine duration, wind speed and air temperature) are the dominant factors of Epan variations. The results obtained are comparable to the previous studies using other statistical methods (i.e., the partial differential method, Pearson correlation coefficient and multiple regression), which reveal the usefulness of Geo‐detector for attributing and diagnosing the spatio ‐ temporal variations of Epan. Moreover, more detailed spatial information and interactions among meteorological elements can be obtained using Geo‐detector. These findings should provide helpful references for understanding the mass and energy budgets at different spatial–temporal scales under global warming.