Beyond Keywords: LLM-Based Content Analysis of Climate Adaptation Strategies in German Regional Planning Documents
This thesis evaluates the integration of climate mitigation and adaptation strategies within German regional planning documents using Large Language Model (LLM)-based content analysis. It investigates the standalone spatial distribution and temporal evolution of climate topics, as well as their correlation with socio-economic and environmental indicators. Findings reveal uneven emphasis: energy supply and flood protection show universal engagement driven by federal mandates, while more recent topics like carbon sinks and heat mitigation remain limited or geographically concentrated. Temporal analysis shows limited impact from legislative changes, given the long feedback and amendment cycle in Germany’s top-down planning scheme. Spatially, urbanization correlates with heightened climate challenges but also increased capacity and awareness. The study recommends enhancing data infrastructure, mandating measurable local targets, and improving multi-level governance coordination to strengthen the effectiveness of spatial planning for climate resilience.