UrbAlytics aims to bridge Artificial Intelligence with Earth Observations, producing information layers that can support city planners and decision-makers in the context of climate resilience and related challenges in urban areas. This research investigates, thanks to the joint expertise of the partners Latitudo 40 and LAND Research Lab®, the Urban Heat Island (UHI) effect evaluating its impacts on cities, assessing Ecosystem Services provided by Blue and Green Infrastructures and proposing a set of Nature-Based Solutions (NBS) for climate adaptation and extreme heat mitigation.
What is the challenge that is being addressed?
The main challenges addressed by the project is evaluating Urban Heat Island effect, its impact and the impact of several Climate adaptation solutions, at urban scale combining data from various sources, including Earth observations, open data databases, and AI-generated information in order to realize a fully automated tool potentially replicable worldwide.
What is the AI solution the project has implemented?
The developed service includes several AI algorithms that are applied to produce the final analysis. First of all, a land-cover classification model based on multitemporal Sentinel-2 images has been developed. In addition, a tree cover density regression model has been developed using Sentinel-2 satellite imagery. Finally, an algorithm to downscale Land Surface Temperature at 10 m spatial resolution combining Landsat 8/9 and Sentinel-2 images has been created in order to better highlight heat phenomenon at urban scale.
Who helped implement the AI solution?
This solution is implemented in the context of Urbalytics, a winning project from the AI4Copernicus 4th Open Call, by the Latitudo 40 and LAND Italia Srl companies. The results of the project are available here.