Meteorological to agricultural drought propagation and vegetation response across China

Study region This study covers mainland China, spanning the Eastern Monsoon Region, the Northwest Arid and Semi-Arid Region, and the Qinghai-Tibet Alpine Region. Study focus Meteorological drought does not translate linearly into agricultural drought, and lagged vegetation responses remain poorly constrained. We combined Moderate Resolution Imaging Spectroradiometer (MODIS) normalized difference vegetation index (NDVI), satellite solar-induced chlorophyll fluorescence (SIF), and drought diagnostics based on the Standardized Precipitation Evapotranspiration Index (SPEI) and Standardized Soil Moisture Index (SSMI) to quantify meteorological-to-agricultural drought propagation, drought duration and severity, and vegetation response memory across China. We examined long-term trends and propagation lags, evaluated cumulative and delayed vegetation responses using correlation and wavelet coherence analyses, and used machine-learning attribution to identify the dominant controls on vegetation variability. New hydrological insights for the region Agricultural drought lasts longer and is more severe than the meteorological drought that triggers it, with a duration approximately one month longer, reflecting the drought amplification effect within the soil-vegetation system. Drought propagation time varies significantly across China, with a median lag of 3–5 months, and the longest lags occur in inland semi-arid regions. Vegetation exhibits a pronounced drought memory effect: NDVI and SIF responses lag approximately 5 months behind drought onset. Compared with NDVI, SIF is more sensitive to short-term water stress. In most parts of China, temperature and evapotranspiration dominate vegetation dynamics, while hydrological factors exert a stronger influence in the humid southeastern regions.