Ten-year variability and environmental controls of ecosystem water use efficiency in a rainfed maize cropland in Northeast China

• Published in: Field crops research Vol. 226; pp. 48 - 55
• Main Authors: Wang, Yu, Zhou, Li, Ping, Xiaoyan, Jia, Qingyu, Li, Rongping
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.09.2018
• Subjects: Drought effects; Eddy covariance; Indirect effect; Structural equation modeling; Eddy covariance; Structural equation modeling; Indirect effect; Drought effects
• ISSN: 0378-4290; 1872-6852
• DOI: 10.1016/j.fcr.2018.07.006

•Structural equation model was used for analyzing the environmental control of WUE.•Daily WUE was regulated by LAI, followed with Ta and VPD in maize cropland.•Distinct controlling mechanisms of daily WUE were detected for dry and wet years.•Precipitation frequency is a good predictor of annual WUE in maize cropland. As a critical link between carbon and water cycles, water use efficiency (WUE) is an important metric for assessing ecosystem response to climate change. However, the controlling mechanism of WUE is still unclear because a number of environmental factors are usually cross-correlated in natural systems. Using the structural equation modeling (SEM) method, this study investigated the seasonal and inter-annual variations of WUE and their controlling mechanism based on 10 years of eddy covariance data in a rainfed maize cropland. The results showed that annual WUE varied between 2.1 g C kg−1H2O and 3.6 g C kg−1H2O, with a multi-year mean of 2.8 g C kg−1H2O. At a daily timescale, leaf area index (LAI) was the primary controlling factor of WUE, while air temperature (Ta) and vapor pressure deficit (VPD) were both shown to have significant indirect effects on WUE through regulating LAI. Distinct controlling mechanisms of daily WUE were detected for years with different hydro-climatic conditions. The primary controlling factors were generally consistent in the 3 dry years (LAI, Ta and VPD). Moreover, WUE consistently showed negative response to soil water content in dry years, while an opposite response was found in wet years. Mean annual Ta was shown to explain 54% of the inter-annual variation of WUE. Despite the insignificant relationship between WUE and precipitation amount at an annual scale, the precipitation frequency was found to be a good predictor of the annual WUE. These findings provide important implications for the accurate simulation of ecosystem WUE, especially under drought conditions in semi-arid regions.