Seasonal divergence in the sensitivity of evapotranspiration to climate and vegetation growth in the Yellow River Basin, China

Abstract

Seasonal variations in terrestrial evapotranspiration (ET) in the Yellow River Basin (YRB) have crucial impacts on the seasonal trajectories of the regional water cycle, vegetation growth, and local climate feedback. However, the possibly divergent roles of climate and vegetation growth variations in controlling seasonal ET patterns remain poorly quantified. This study therefore quantifies the interannual sensitivity and attribution of ET to climate and vegetation growth variations in different seasons and different biomes in the YRB in China between 1982 and 2011, using the satellite‐derived normalized difference vegetation index (NDVI), FLUXNET‐based upscaled ET, and concurrent climate data. The results reveal a clear seasonal divergence in the interannual sensitivity of ET to climate and vegetation growth variations in the YRB. Interannual precipitation and NDVI variations play a dominant role in controlling seasonal ET variations in the YRB, with temperature having a marginal effect. Interannual ET sensitivity to precipitation weakens with an increasing mean annual precipitation gradient in almost all seasons, especially in summer and autumn. More importantly, a seasonally varying role of vegetation growth in mediating seasonal ET was discovered, and a crucial role of late‐growing‐season vegetation growth in controlling the seasonal trajectory of regional ET was explicitly identified. These results suggest that ongoing intensive vegetation restoration has crucial impacts on seasonal water‐cycle patterns and consequent terrestrial‐atmospheric biogeochemical feedback in the YRB.

Land–Atmosphere Exchange of Water and Heat in the Arid Mountainous Grasslands of Central Asia during the Growing Season

“…The Le was higher than the He at the end of May and lower than the He with an overall decreasing trend after August. Overall, the Le was higher than the He from June to July, suggesting high ET in this period, and this result is consistent with the vigorous vegetative growth during this period, which was associated with relatively high transpiration [40,41]. Different results were obtained for the Le and He series from April to October in 2014 compared with the values from 2013 and 2015; the Le was lower than the He in June and July 2014, indicating relatively low ET in this period.…”

“…The Le was higher than the He at the end of May and lower than the He with an overall decreasing trend after August. Overall, the Le was higher than the He from June to July, suggesting high ET in this period, and this result is consistent with the vigorous vegetative growth during this period, which was associated with relatively high transpiration [40,41] (Table 1), during which the ET in the normally vigorous growing season was relatively low [40,41].…”

“…Different results were obtained for the Le and He series from April to October in 2014 compared with the values from 2013 and 2015; the Le was lower than the He in June and July 2014, indicating relatively low ET in this period. This is consistent with the low precipitation, severe drought and poor vegetation growth in 2014 (Table 1), during which the ET in the normally vigorous growing season was relatively low [40,41]. …”

“…Monthly variations in the ET were closely related to vegetation growth during the growing period in this region; in 2013 and 2015, the ET rapidly increased with vegetation growth from April to July and reached its maximum value in July. Most of the vegetation began to wilt or enter dormancy in August [40,41]; therefore, transpiration was reduced, rapidly decreasing ET. In 2014, significantly low P, severe drought and poor vegetation growth resulted in relatively low grassland ET during the normally vigorous growing season (Figure 4).…”

Seasonal Divergent Tree Growth Trends and Growth Variability along Drought Gradient over Northeastern China

“…where, n is the number of years from 1982 to 2015, and AI i is the aridity index in ith year. The PET was calculated with the adjusted Thornthwaite equation [36,37], which has been widely used in large-scale research, where detailed parameters for the Penman-Monteith equation were sometimes hard to obtain [37,38]. The annual total PET was calculated by summing monthly PET (PETm), which was calculated using Equation 3:…”