Detecting forest disturbances in mid-latitude ecosystems using Sentinel-2 time series of functional traits anomalies

Forest are facing increasing stress from climateinduced droughs and extreme weather events, threatening ecosystem function and resilience. Timely detection of these disturbances is crucial for effective forest management. Remote sensing offers scalable monitoring of forest condition and dynamics across large areas with consistent spatial and temporal resolution. While most forest disturbance mapping approaches rely on indirect proxies like vegetation indices, this study explores the potential of Sentinel-2 derived plant traits, Canopy Chlorophyll Content (CCC) and Leaf Area Index (LAI), which offer more direct links to vegetation function and structure. Two documented disturbance types were analyzed: the 2022 summer drought and 2023 windthrow. A rolling regression slope analysis was applied to cumulative anomalies of functional traits to assess the timing, intensity, and persistence of the disturbances across ground-validated plots. Results show droughts generate isolated anomaly peaks during the event, with a stronger impact on leaf pigmentation (average reduction in CCC of -41.53% vs. -31.93% in LAI). In contrast, treefall produces a comparable reductions in both traits (exceeding 30.00%), indicating both functional and structural canopy loss. In the year following the event, a prolonged plateau of anomalies is observed, with partial midseason attenuation linked to understory dynamics. These findings highlight the potential of the proposed approach to effectively support long-term forest monitoring and conservation strategies.