蒸汽压差
大气科学
蒸发
化学
环境科学
潜热
潜在蒸发
含水量
干旱
水文学(农业)
蒸腾作用
光合作用
气象学
物理
生态学
岩土工程
工程类
生物
生物化学
作者
Bing Tong,Jianping Guo,Hui Xu,Yinjun Wang,Huirong Li,Lingen Bian,Jian Zhang,Shenghui Zhou
标识
DOI:10.1016/j.scitotenv.2022.157890
摘要
Surface energy partitioning is one of the most important aspects of the land-atmosphere coupling. The objective of this study is to examine how soil moisture (SM) and atmospheric conditions (net radiation, Rn and vapor pressure deficit, VPD) affect surface evaporation fraction (EF, determined by LE/(LE + H), where LE and H are latent and sensible heat flux, respectively) with measurements at a semi-arid grass site in China during the mid-growing season, 2020. The three factors (SM, Rn, and VPD) were divided into different levels, and then their effects on EF were investigated qualitatively using a combinatorial stratification method and quantificationally using a path analysis. Generally, the results indicated that the effect of one factor of SM, Rn and VPD on EF was influenced by the other two factors. EF tended to increase with increasing SM. Increased VPD (Rn) enhanced (weakened) the SM-EF relationship. When soil was dry, EF tended to decrease with increasing VPD; when soil was wet, EF initially levelled off and then decreased with increasing VPD. Increased Rn enhanced (weakened) the positive (negative) effect of VPD on EF when soil was wet (dry). In terms of Rn effect, EF tended to decrease as Rn increases. Further, path analysis suggested that SM, Rn, and VPD not only directly affected EF, but also indirectly affected EF, mainly through canopy conductance (Gs) and temperature difference between land surface and air (∆T). The direct effect of SM accounted for >50 % of its total effect on EF, while the total effects of Rn and VPD on EF were dominated by their indirect effects. These observational evidences may have implications for improving representation of land-atmosphere coupling in atmospheric general circulation models over the semi-arid regions covered by grass.
科研通智能强力驱动
Strongly Powered by AbleSci AI