Vegetation drought condition index for probabilistic monitoring and forecasting of vegetation drought

As the impacts of meteorological drought on vegetation have intensified, there is a growing need for a system that can quantitatively assess and forecast vegetation drought. This study proposes a vegetation drought monitoring and forecasting framework utilizing a copula-based probabilistic approach to address this need. By constructing a joint probability distribution between a meteorological drought index and a vegetation index, we developed the Vegetation Drought Condition Index (VDCI), which was then integrated with numerical weather prediction data to establish a probabilistic vegetation drought forecasting framework. The VDCI is capable of selectively detecting vegetation stress caused by meteorological conditions and enables the quantitative assessment of drought severity through a four-level vegetation drought classification criteria. Spatial and temporal analyses confirmed that the VDCI can identify vegetation drought more clearly than individual indices. Moreover, the probabilistic forecasting framework demonstrated excellent forecasting performance with an average F1-score of approximately 0.9 across all pixels. This study proposes a framework enabling quantitative monitoring and forecasting of vegetation drought based on the probabilistic relationship between meteorological drought and vegetation response, and is expected to contribute to the development of ecosystem-based drought early warning and response strategies in the future. • Developed the Vegetation Drought Condition Index using a copula-based probabilistic approach. • Proposed a new framework for spatiotemporal forecasting of vegetation drought. • Overcame uncertainty in weather predictions and the lack of vegetation forecasts. • Quantified the propagation time and lag effects of meteorological drought on vegetation. • Identified climate and land cover type influences on vegetation drought vulnerability.