Search

• Ecosystem WUE remained stable in 1985–1999 but increased in 2000–2015 on CLP. • Climate control the temporal dynamics in WUE via effects on GPP/ T before 2000. • Revegetation-induced increase in LAI caused increase in WUE via T /ET after 2000. Ecosystem water use efficiency (WUE), defined as the ratio of gross primary productivity (GPP) to total ecosystem evapotranspiration (ET), is an important indicator of the coupling between the terrestrial carbon and water cycles. China's Loess Plateau (CLP) has been experiencing climatic warming in recent decades and large-scale revegetation since the early 2000s. Understanding the combined effects of revegetation and climate change on ecosystem WUE patterns is important for a better prediction of future changes of the water and carbon cycles in the region. In this study, we decompose ecosystem WUE into a two-component process, i.e., the ratio of carbon uptake to plant transpiration, GPP/ T , and the ratio of plant transpiration to total evapotranspiration, T /ET. Based on this, we investigate the temporal variations of ecosystem WUE and examine the underlying controls of WUE dynamics on CLP during 1985–2015. We find that ecosystem WUE on CLP remains more or less unchanged during 1985–1999 as a result of an increased T /ET that is largely offset by a decreased GPP/ T induced by increases in atmospheric vapor pressure deficit. In comparison, ecosystem WUE significantly increases during 2000–2015, owing to the increased leaf area index, which promotes the fraction of plant transpiration over total ET (i.e., T /ET). In facing the upcoming peak of greenness and continuing climate warming, our results suggest that CLP's ecosystem WUE may experience a downward trend in the future due to diminishing positive effects of leaf area index but increasing negative effects of atmospheric vapor pressure deficit. Our study highlights the importance of decomposing ecosystem WUE into component processes to better understand the mechanisms that underlying the changes of ecosystem WUE and to predict future ecosystem WUE dynamics. [ABSTRACT FROM AUTHOR]