AI Support for Transforming Surfaces into Multifunctional Green Spaces

Urbanization has led to a rapid increase in hard surfaces and a reduction in green areas, which impacts the hydrological cycle and ecosystem functions. This leads to decreased ecosystem services and more frequent floods in cities, especially with increased rainfall and insufficient stormwater systems. To counteract this, multifunctional green and blue structures are needed. Nature-based solutions (NBS) can regulate water flows, lower urban temperatures, enhance biodiversity, and provide recreational opportunities. Green areas affect both citizens' well-being and the city's environmental performance.

The project aims to develop an AI- and GIS-based decision support tool to optimize the transformation of impervious surfaces into green spaces. The tool will integrate goals for climate adaptation, biodiversity, and urban planning into a single platform, using AI and GIS to analyze data and suggest effective locations for green transformation. By utilizing machine learning trained on remote sensing, geographical data, and other spatial sources, the model can identify areas with high potential for green space transformation and consider factors such as ecosystem services, flow regulation, and recreation.

The tool will be tested in Malmö and Uppsala, and the experiences from these cities will improve the model to create a scalable solution for other cities. By combining AI and urban planning, cities can better adapt to climate change, improve quality of life, and strengthen biodiversity. The tool can also help real estate companies comply with EU directives such as the EU taxonomy and CSRD, and contribute to changing land use policies, for example, by transforming parking lots into green spaces. The goal is to create more resilient cities that are cost-effective and better for people, animals, and the climate.