Development of an indicator system for solar-induced chlorophyll fluorescence monitoring to enhance early warning of flash drought

In recent years, flash droughts (FD) and slow droughts (SD) have increasingly occurred interconnectedly, leading to significant losses in the water-energy-food systems of the Middle-lower Yangtze Plain. This is primarily due to the complexity of drought events with varying onset rates, which present severe challenges to traditional drought early warning systems. Therefore, developing a comprehensive early warning system capable of effectively forecasting multi-temporal scale drought events is urgently needed. In this study, we analyzed the frequency, duration, and spatial extent of drought events with different onset rates (SD, SFD, and FD) in the Middle-lower Yangtze Plain from 2000 to 2019, based on the GLDAS-Noah soil moisture dataset. Using these identified drought types as case studies, we proposed a novel multi-temporal scale drought early warning system (MSDEWS), which relies on the multi-metric (time, threshold, and variability) response of the solar-induced chlorophyll fluorescence rapid change index (SIF RCI). Furthermore, we compared the performance of SIF RCI with traditional meteorological drought indices in FD early warning, highlighting the impact of multi-source SIF datasets (CSIF, GOSIF, and RTSIF) and vegetation type response sensitivity on the MSDEWS's forecasting ability. The results demonstrate that monitoring the response dynamics of SIF RCI in shrublands and broadleaf forests significantly improves the accuracy of FD forecasts. SIF RCI can detect FD onset signals 4–5 pentads in advance. We observed significant differences in the minimum values, variability, and response times of SIF RCI across drought events with varying onset rates. The early warning effectiveness of the SIF RCI for FD is attributed to its capacity to respond to meteorological drought stress before significant soil moisture loss, offering higher sensitivity compared to changes in vegetation structure. The proposed MSDEWS shows considerable potential for application in other global FD hotspots, although its forecasting capabilities and quantification thresholds require further investigation and localization. Monitoring agricultural droughts using the vegetation response of SIF RCI provides new perspectives and solutions for addressing the increasingly complex multi-temporal-scale drought risks.