Variability of cooling distances and intensities of urban green spaces over six hot summer days affects their relevance for UHI mitigation in Berlin, Germany
Abstract
In the context of global urbanization and increasingly frequent heatwaves, urban green spaces (UGS) play a crucial role in mitigating urban heat islands (UHI). However, the variability and reliability of these cooling services is often overlooked. This study addresses the temporal and inter-site variability of cooling effectiveness across different UGS types in Berlin, Germany. We analyzed inner (land surface temperature, LST) and outer (park cooling distance, PCD; park cooling intensity, PCI) cooling parameters for 228 parks, 512 allotments, and 143 cemeteries using six cloud-free Landsat 8 OLI scenes from hot summer days (Tmax > 25°C). While LST showed low variabilities between sites, PCD and PCI fluctuated more, especially in allotments. Only 9–16% of the UGSs exhibited stable cooling distances over time, with parks achieving the largest PCD and cemeteries providing the most intense and reliable cooling. Allotments were the least effective and most variable. To investigate drivers of temporal PCD variability, we conducted regression analyses using vegetation indices, spatial metrics, and urban structure. However, the models explained little of the variance (R² up to 0.3), indicating complex external influences not fully captured by the available data. To analyze the multifaceted role of UGS in UHI mitigation, we propose the novel concept of “Cooling Relevance,” which integrates intensity, spatial extent, reliability, and the number of potential beneficiaries. Parks emerged as the most relevant UGS type under this framework. Our findings highlight the need to incorporate reliability into urban greening strategies to support equitable and climate-resilient urban planning.
Fields of Study & Subjects
Urban Heat Island Mitigation
Smart Materials for Construction
Urban Green Space and Health
Publication details
- Journal / source
- Sustainable Cities and Society
- Publisher
- Elsevier BV
- Volume
- 137
- Pages
- 107139–107139
- Publication date
- 9 January 2026
- Language
- EN