From temperature observations to mitigation measures: Data-driven approaches to multi-scale climate adaptation in Hesse, Germany

As urban temperatures rise and the demand for urban densification increases, climate adaptation has become a significant concern for policymakers. However, we face a challenge in today's data-rich environment: How can we effectively manage the vast amounts of information available to make informed decisions for our cities and communities? Accordingly, urban agglomerations in Germany are struggling in allocating measures.

In collaboration with the federal state of Hesse, we have developed a data-driven decision-making method to address two critical questions: Which locations should we focus on for heat mitigation and the protection of cool oases? And which climate adaptation measures are most suitable for each location?

Our method consists of four steps:

• Identification of Hot Spots and Cold Spots: We identify universal hot and cold spots that consistently experience high or low temperatures during both day and night. This assessment is based on various temperature metrics that capture heat stress and imbalances in surface heat fluxes, which we harmonize using a 100 m grid. We validate our findings using local climate zones and CORINE land cover data.
• Prioritization of Locations: We prioritize hot and cold spots using a merit- and penalty-based system. Rather than focusing solely on the hottest areas, we emphasize locations where heat impacts vulnerable populations or disrupts cooling patterns over broader regions.
• Assessment of Mitigation Potentials: Each grid cell is evaluated for its deficits and potentials for climate adaptation using a multitude of contextual data. This includes detailed land cover information obtained from high-resolution aerial imagery segmentation, along with green and blue indicators (such as NDVI, green volume, and water availability), and settlement structure (e.g., types  of historic urban centers or single family housing).
• Suitability of Adaptation Measures: Based on the results from steps 2 and 3, we rank the suitability of various adaptation measures (e.g., green facades, increased tree coverage) for each grid cell using a second merit- and penalty-based system.

In a close loop with practitioners from local to regional authorities, this integrated approach enables us not only to identify areas in need of climate action, but also to recommend specific, actionable measures. By leveraging data at this level of detail and scale, we facilitate informed, targeted climate adaptation strategies.