Sustainability Assessment and Risk Zone Identification of Irrigation-Driven Regional Greening in Northwest China

Irrigation-driven greening is essential for northwest China’s dryland ecosystems, where vegetation growth depends on key hydrological factors, including precipitation (PRE), evapotranspiration (ET), soil moisture (SM), and irrigation water use (IWU), which affect water availability to a certain extent. To assess greening sustainability, a 1 km IWU dataset was created for 2001–2022 by combining remote sensing and ancillary data using machine learning, overcoming limited irrigation records. By linking IWU with the normalized difference vegetation index (NDVI) and analyzing trends in irrigated areas, we implemented a regional zonation approach to identify specific risk areas and evaluated both greening sustainability and vegetation responses using water balance (WB) and various hydrological variables. The results show that NDVI has increased widely over the past two decades, with sustained positive WB and stable irrigation, indicating improved water availability. However, spatial differences exist: 35.98% of irrigated areas have rising NDVI but falling IWU, especially in the east, where higher NDVI, IWU, WB, PRE, and ΔSM (soil moisture difference between growing season end and start) reflect favorable climate and hydrology; attention should also be directed toward potential deep percolation and saline sinks. In contrast, areas with high IWU often displayed elevated NDVI but declining water availability, suggesting unsustainable greening due to excessive irrigation. In addition, the SCDIWU-SCDNDVI class dominates among significant NDVI-IWU trends, indicating potential for sustainable irrigation under certain drought and climate conditions. Overall, the northwestern portion of the study area exhibits the lowest water availability; cities such as Urumqi warrant particular attention. These findings identify at-risk areas and those with better water resilience, supporting targeted water–vegetation management.