Changes in reference evapotranspiration over the non‐monsoon region of China during 1961–2017: Relationships with atmospheric circulation and attributions

Abstract

AbstractReference evapotranspiration (ET0) is an essential component of the hydrological cycle and is crucial to water resources management and assessment. Spatiotemporal variations in ET0 and their attribution to five climatic variables (maximum and minimum air temperatures, solar radiation, vapour pressure, and wind speed) were estimated for the non‐monsoon region of China from 1961–2017 meteorological data by using the Penman–Monteith equation and the partial‐differential method. The association between ET0 and atmospheric circulation patterns was explored by applying Pearson's correlation analysis. The annual ET0 series had two earlier shift points, the first showing a significant increase from 1961 to 1973 (period I) and the second showing a significant decrease from 1974 to 1994 (period II). ET0 has significantly increased again since 1994 (period III). Wind speed has been the dominant contributing factor to ET0 followed by maximum air temperature during all three periods, but the contribution gap between minimum and maximum air temperatures is narrowing. ET0 has been strongly correlated with eight selected teleconnection indices in some regions, but the locations and spatial extents have varied considerably for different indices. Eight relationships with ET0 were divided into four groups: North Atlantic Oscillation and Arctic Oscillation (mainly the northwest non‐monsoon region), Pacific North American Index and Pacific Decadal Oscillation (mainly the northern non‐monsoon region), Antarctic Oscillation and Atlantic Multidecadal Oscillation (mainly the Hexi Corridor and the southern Xinjiang), and East Asian Summer Monsoon Index and South China Sea Summer Monsoon Index (mainly the southeast non‐monsoon region). The Antarctic Oscillation, Atlantic Multidecadal Oscillation, and South China Sea Summer Monsoon Index indicated more potential for predicting ET0. When we used Pearson's correlation analysis to analyse the relationship between atmospheric circulation and the dominant factors affecting ET0, we found that the circulation patterns are likely to affect the variation in ET0 by influencing the dominant climatic variables.