Warm streets, cool leaves? Exploring vegetation phenology with daily heating patterns: Brussels 2022

Sammanfattning

Urban vegetation is a key strategy for climate adaptation in increasingly dense and warming cities. While some studies look at urban seasonal trends in plant phenology, fewer look at continuous dynamic phenology measures. This study investigates the relationship between the Plant Phenology Index (PPI) and Land Surface Temperature (LST) across Brussels in 2022, using satellites Sentinel-2-derived PPI and Landsat 8 and 9 LST. The land surface temperatures of green spaces were consistently lower than those of urban areas, exhibiting differences of up to 8.5°C. While forests were particularly effective in cooling, showing temperatures up to 6.1°C lower than those in parks. Despite these patterns, overall results revealed weak spatial correlations between PPI and LST, with no significant association (ρ = -0,033, p > 0,05). More pronounced negative correlations, though still moderate, were found in urban (ρ = -0.55, p < 0.001***) and vegetative (ρ = -0.41, p < 0.001***) areas between the two variables. When examining LST and PPI dynamics within green areas and in a 100 m buffer outside them, the expected trends were observed: lower LST and higher PPI within vegetated environments and vice-versa outside. However, a two-way ANOVA revealed this difference was not statistically significant (p > 0.05). Although PPI decreases more outside parks and LST increases, the temperature rise is not simply due to reduced phenology index. Instead, vegetation type and spatial context play a critical role in mitigating urban heat. This research offers insights into the dynamics of PPI as a remote sensing metric in urban environments but also highlights the need to incorporate vegetation type and climate parameters in future studies to better understand urban heat island vegetation mitigation strategies.