Spatio-temporal forest structure dynamics in Germany: A synthesis of remote sensing products

Abstract

Germany’s forests have undergone unprecedented structural transformations in recent years, driven by the 2018-2020 drought and the following increased frequency of disturbances such as windstorms, bark beetle outbreaks and forest fires. This thesis analyses these dynamics, delivering a comprehensive, high-resolution, spatio-temporal assessment of forest structure changes between 2017 and 2023 by synthesising three remote sensing products, provided by the German Aerospace Center. These products have a 10 m spatial and a monthly to yearly temporal resolution, covering the entire study area of Germany. Weak to moderate correlations between forest structure and forest condition suggest that disturbances affect forest structure in diverse ways. This thesis distinguishes distinct disturbance patterns and integrates them with forest structural data, providing insights into key aspects of forest structure dynamics — from pre-disturbance vulnerabilities to direct disturbance impacts and post-disturbance recovery. The national analysis, based solely on the remote sensing products, is complemented by two regional analyses that incorporate ground-truth reference data to define disturbance patterns and confirm the characteristics observed at the national scale. Main findings reveal considerable effects on the forest structure, which vary depending on the prevailing disturbance patterns. Both national and regional analyses show that standreplacing disturbances lead to significantly greater reductions in canopy cover, canopy height, aboveground biomass and foliage height diversity than non-stand-replacing disturbances, with wind, bark beetle and large-scale disturbances causing the most severe impacts. Among all disturbance patterns, pre-disturbance structural decline, particularly in aboveground biomass, is observed compared to undisturbed reference areas. Post-disturbance recovery varies notably, with canopy cover showing the fastest rebound. This thesis demonstrates the potential of remote sensing to support adaptive forest management, enhance forest inventories and enable early-warning systems by detecting early signs of stress before visible disturbances occur. The findings also contribute to a deeper understanding of forest regeneration patterns and could guide targeted reforestation efforts, promoting ecosystem resilience in the face of ongoing environmental change.