Ecological Flow Management Identified as Leading Driver of Grassland Greening in the Gobi Desert Using Deep Learning.

This study develops a convolutional recurrent deep learning model to accurately predict fine‐resolution spatiotemporal changes in grass coverage in arid regions. Applying the model to the Gobi Desert reveals that ecological flow regulation contributes to 61.8% of the total increase in grass cover (130.6 km2) in the study area (40,423 km2) over 2005–2015, nearly triple the contribution of local climate change (+23.0%). The transboundary hydrological impact (+32.4%) and interactions between drivers (−17.2%) are also significant. In an intermediate future climate change scenario, we found no statistically significant trend for the total grass‐covered area due to the counteracting effects among different drivers. The study findings suggest that timely, adaptive and spatially heterogeneous ecological flow management is crucial for addressing grassland degradation in arid regions. This study provides a promising approach to land surface modeling under climate change and human disturbance and expands the existing understanding of the global greening process. Plain Language Summary: The Earth's recent greening (i.e., vegetation increase) has boosted the terrestrial carbon sink, intensified the hydrologic cycle, and contributed to mitigating global warming. CO2 fertilization has been identified as the primary cause at the global scale, while climate change, land use changes, and nitrogen deposition also contribute. However, there are still many gaps in our understanding of how grasslands in arid areas are being affected. Studies in Central Asia identified precipitation as the dominant factor affecting grassland change and grazing as the main influence from humans. Based on satellite data, we employ a novel deep learning (DL) approach to emulate spatiotemporal changes in grass coverage at a 1‐km resolution over an area of 40,423 km2 in the Gobi Desert in northwestern China. Our study identified previously overlooked mechanisms contributing to the increase in grass coverage, including ecological flow management and transboundary hydrological impact. Ecological flow management was found to be responsible for almost three times the increase in grass coverage over 2005–2015 than that of local climate changes. Our findings suggest that timely, adaptive, and spatially targeted ecological flow management is urgently needed to prevent further grassland degradation in arid regions, and DL‐based artificial intelligence can offer valuable management suggestions. Key Points: Deep learning (DL) precisely emulates spatiotemporal changes in grass coverage at a 1‐km resolution over an area of 40,423 km2DL interprets ecological flow regulation (EFR) as the leading cause of grassland greening in the Gobi DesertTimely, adaptive and spatially heterogeneous ecological flow management is desired to prevent grassland degradation in arid regions [ABSTRACT FROM AUTHOR]