Towards near real-time drought stress assessment in Europe’s temperate forests – comparing remote sensing time series with continuous in-situ tree-level measurements

• Sentinel-2 enables robust seasonal comparisons with in-situ monitored tree water deficit. • NDWI captures tree water stress in European Beech but not in Norway Spruce. • Vegetation indices do not linearly capture tree water stress throughout the year. • Temporal aggregation improves tree water stress monitoring accuracy. Increasingly frequent droughts in Central Europe underscore the need for regular forest health monitoring. For decades, vegetation indices like Normalized Difference Vegetation Index (NDVI) and Normalized Difference Water Index (NDWI) have been used as an indicator of forest condition and these indices are the focus of next-generation near real-time forest water stress monitoring which aims to update stress information coincident with satellite revisit times. However, few studies have robustly correlated vegetation indices with continuous ground-based monitoring tracking the water stress of individual trees. This study examines the relationship between 7 years of Sentinel-2 derived vegetation index time series (NDWI, NDVI, Chlorophyll Red Edge: CIRE, Chlorophyll Carotenoid Index: CCI, and Enhanced Vegetation Index: EVI) and continuous stem-level tree water deficit measurements from 41 European Beech ( Fagus sylvatica L.) and 61 Norway Spruce ( Picea abies (L.) H. Karst.) trees in Switzerland’s temperate forests. Results show both an inter- and intra-seasonal correlation between tree water deficit and vegetation indices at a monthly time step (linear fit R 2 values reaching 0.62 for Norway Spruce and 0.58 for European Beech), but it is more challenging to resolve recoverable tree water deficit at a finer temporal resolution. Common satellite-derived vegetation indices are unlikely to fully replace the sensitivity of in-situ tree water stress measurements but some indicators like EVI for Norway Spruce or NDWI for European Beech show potential for estimating broad tree water stress trends during the vegetated period. Monthly, or ongoing rolling average estimates could offer a scalable approach for regular drought stress assessment.