| Abstract： |
Changes in ecosystem water use efficiency (WUE) and evapotranspiration (ET) that are driven by vegetation change can be used as metrics to quantify ecosystem carbon and water exchange. The effect of vegetation on WUE has been explored in the past through consideration of total ET. However, interception (Ei) doesn't directly participate in ecosystem carbon and water exchange, and uncertainty of modeled Ei, resulting in great uncertainty of ET partitioning and ecosystem WUE. Here, we estimated Ei in China using Gerrits' model, replacing the algorithm for Ei in the Priestley-Taylor Jet Propulsion Laboratory (PT-JPL) model. The modified PT-JPL model was then incorporated into the analytical WUE model to estimate WUE over China for 2003-2011, and the results were validated against literature data (WUELiterature). To evaluate the net effect of vegetation restoration on WUE, we simulated WUE in three ways: with vegetation change; without vegetation change; and with vegetation change, but without any Ei changes caused by vegetation changes. In the study presented here, we demonstrate the WUE simulated by the analytical WUE model is significantly correlated with the WUELiterature (R-2 = 0.76, p-value < 0.001), and leaf area index contributes the most to the spatial variability of Ei/ET, ranging from 10% to 55%. Increased Ei weakened the effect of vegetation change on WUE by similar to 51.3% during vegetation restoration in China; a reduction that was particularly marked in southern China. These findings highlight the importance of considering Ei when assessing the effects of vegetation change on ecosystem carbon and water exchange, and provide insights that are important for the sustainable management of vegetation change in China. |
