Transition features of surface evapotranspiration responding to climate warming with spatial precipitation-based climate type in northern China

Abstract

Global surface evapotranspiration is one of the most significant components of the response of the water cycle to a warming climate. However, trends in surface evapotranspiration have varied widely from the trend in climate warming according to recent studies, and some studies have even shown an opposing trend. The reason for this difference in the response of surface evapotranspiration to climate warming is still not completely understood. We validated the gridded FLUXNET evapotranspiration dataset and the Global Land Data Assimilation System (GLDAS) against evapotranspiration data observed in Northern China by eddy covariance systems. The response of evapotranspiration to increases in temperature varies with climate type in Northern China and there is a correlation with the amount of precipitation. Climate types with a mean annual precipitation of 200–400 mm (P250 and P350) are a sensitive interval in which the climatological trend of evapotranspiration changes from negative to positive and the response of evapotranspiration to an increase in temperature is less obvious. In more humid climates, evapotranspiration increases with increasing temperature, whereas in drier climates evapotranspiration decreases with increasing temperature. Similar transition zones for surface evapotranspiration are also seen in other regions. This is a new understanding of global changes in surface evapotranspiration and can explain the different trends in the response of surface evapotranspiration to temperature reported previously. Although the methods and data used are different from these earlier studies, the differences between precipitation-based climate types are clearer. These transition zones can be explained by the different mechanisms for the effect of temperature on evapotranspiration. Temperature directly affects the potential evapotranspiration and an increase in temperature will increase the potential evapotranspiration. However, this increase in evapotranspiration can result in a decrease in the soil moisture content, which will restrict the growth of vegetation, which in turn reduces surface evapotranspiration. In humid climates, where the soil moisture is high, the direct temperature effect is dominant and an increase in temperature will enhance surface evapotranspiration. In drier climates where the soils are deficient in water, reduced vegetation growth means that evapotranspiration is limited and an increase in temperature weakens surface evapotranspiration due to soil moisture stress. The prominent effect of the summer monsoon in Northern China results in a wide range of precipitation over a large spatial area and the climate transition zone is clearly defined. Thus, the effect on evapotranspiration of different climate types classified according to the amount of precipitation is significant in this area. This has deepened our understanding of the mechanisms influencing evapotranspiration in similar areas. In the areas where transition between climate types is obvious, the response of surface evapotranspiration to climate warming is complex and there is large spatial variation in the warming trend for evapotranspiration. Therefore, the variation in regional evapotranspiration cannot be determined by the variation in the point or local evapotranspiration. This will significantly affect the assessment of regional water resources and drought monitoring in these transitional areas. For such areas, the analysis of trends in evapotranspiration should be at a fine resolution to meet the technical needs for the assessment of water resources and drought monitoring. The factors influencing surface evapotranspiration are complex. In addition to temperature and precipitation, other factors such as wind speed, solar radiation and atmospheric humidity also affect surface evapotranspiration to varying degrees. The results of this study were affected by these factors, resulting in a certain degree of dispersion and bias in the results, which will in turn affect the objective quantitation analysis.