Global enhanced vegetation photosynthesis in urban environment and its drivers revealed by satellite solar-induced chlorophyll fluorescence data

Investigation on the future impacts of climatic and environmental change on vegetation photosynthesis has been largely restricted to controlled field experiments, which can hardly be extended to global scale due to limited spatial, species and ecosystem coverages. However, in urban areas plants experience altered environments that mimic potential future conditions, with higher air temperature, atmospheric carbon dioxide (CO2) concentration and pollution levels. Cities can therefore be used as global, unplanned experiments for assessing the photosynthetic response to multiple climatic and environmental drivers. Following this logic, here we investigate the urbanization impact on vegetation primary productivity and its drivers at global 160 mega-cities, using highspatial resolution satellite solar-induced chlorophyll fluorescence (SIF) data as the proxy of photosynthesis. SIF enhancements were observed across most of the urban-rural gradients, accounting for more than 85% of the investigated land pixels. More importantly, SIF enhancements due to indirect urbanization impact (i.e., the impacts of climatic and environmental factors on vegetation growth) offset approximately 47% of the SIF reductions due to land cover change, a value significantly higher than that observed for a greenness spectral index (Enhanced Vegetation Index, EVI) (30%). Atmospheric CO2, air temperature, radiation and atmospheric nitrogen dioxide (NO2) were found to be the main drivers accounting for the enhanced SIF in urban areas. These results prove a dominant and global enhancement of vegetation photosynthesis in urban conditions, and reveal the specific contribution of climatic and environmental factors. Our findings can help to forecast the impacts of future environmental conditions on vegetation photosynthesis, and enhance our knowledge on the overall response of terrestrial biomes to climate change.