Unraveling the relative impacts of climate change and human activities on grassland productivity in Central Asia over last three decades.

Climate change (CC) and human activities (HA) have severely influenced grassland productivity in Central Asia since the 1980s. However, the relative impacts of CC and HA on grassland productivity are not adequately documented, especially over the past three decades. In this study, we adapted the Ensemble Empirical Mode Decomposition (EEMD) to reveal potential timescales at which grassland productivity varied in Central Asia and to investigate the spatiotemporal variations of grassland productivity during 1982-2015. We developed a quantitative method that incorporated the EEMD, along with six scenarios, to disentangle the effects of CC and HA on grassland productivity in Central Asia. Results showed that grassland productivity in Central Asia trended upward significantly at a rate of 0.66 gC m yr and was dominated by a 3-year time scale oscillation. The impacts of CC and HA on grassland productivity varied significantly over space and time. CC mainly facilitated grassland productivity restoration, whereas HA decreased grassland productivity in Central Asia. Besides, varied HA in six regions of Central Asia were due to different policy implementations across these regions. In particular, HA in Xinjiang significantly promoted grassland restoration, accounting for 22.5% of the total human-affected area, mostly because of the implementation of the Grazing Withdrawal Program (GWP), while HA significantly accelerated grassland productivity degradation in Uzbekistan and Turkmenistan over last three decades. Additionally, HA promoted the restoration of grassland productivity in Kazakhstan in a short period due to the disintegration of the Soviet Union, but degraded it at long-term scale. Further, precipitation was found to be the main climatic factor while grazing be the main human factor for controlling grassland productivity variations in Central Asia, respectively. Overall, our study provides not only a novel way of quantifying the impacts of CC and HA on vegetation variations but also new insights into mechanisms mediating grassland productivity in Central Asia.