Spatial pattern of reference evapotranspiration change and its temporal evolution over Southwest China

Due to the close relationship of climate change with reference evapotranspiration (ET o ), detecting changes in ET o spatial distribution and its temporal evolution at local and regional levels is favorable to comprehensively understand climate change-induced impacts on hydrology and agriculture. In this study, the objective is to identify whether climate change has caused variation of ET o spatial distribution in different analysis periods [i.e., long- (20-year), medium- (10-year), and short-term (5-year)] and to investigate its temporal evolution (namely, when these changes happened) at annual and monthly scales in Southwest China (SWC). First, we estimated ET o values using the United Nations Food and Agriculture Organization (FAO) Penman-Monteith equation, based on historical climate data measured at 269 weather sites during 1973–2012. The analysis of variance (ANOVA) results indicated that the spatial pattern of annual ET o had significantly changed during the past 40 years, particularly in west SWC for the long-term analysis period, and west and southeast SWC in both medium- and short-term periods, which corresponded to the percent area of significant differences which were 21.9, 58.0, and 48.2 %, respectively. For investigating temporal evolution of spatial patterns of annual ET o , Duncan’s multiple range test was used, and we found that the most significant changes appeared during 1988–2002 with the significant area of higher than 25.0 %. In addition, for long-, medium-, and short-term analysis periods, the spatial distribution has significantly changed during March, September, November, and December, especially in the corresponding periods of 1988–1997, 1983–1992, 1973–1977, and 1988–2002. All in all, climate change has resulted in significant ET o changes in SWC since the 1970s. Knowledge of climate change-induced spatial distribution of ET o and its temporal evolution would aid in formulating strategies for water resources and agricultural managements.