As climate change exacerbates environmental challenges associated with urban growth, green stormwater infrastructure (GSI) is a prevalent stormwater mitigation strategy to provide resilience and mitigate the impacts of development on flooding. In parallel, fully sustainable GSI systems must confront the challenges of historically unequitable distribution of infrastructure. The current data revolution has reached municipal stormwater programs; however, these programs are limited by a lack of knowledge of GSI life-cycle dynamics, high performance and emerging computational tools, and how to integrate new science into design and planning decisions. There is a scientific gap in the space formed among GSI design, performance function, and planning decisions that requires bridging hydrologic science, urban planning, and data analytics. This project leverages innovations in artificial intelligence (AI), advancements in the empirical and theoretical understanding of urban hydrologic science, and social data to produce a new model of GSI dynamics that considers social and environmental equity issues. This model will flip the paradigm of infrastructure planning and put the impact on society and the environment on par with engineering solutions to flooding. The model will be made available for use by public and private practitioners to plan, develop, and manage more sustainable and equitable GSI, and by researchers to deepen convergent knowledge of the complex social issues associated with urban flooding. 

The current state of GSI research is ripe for the application of AI techniques to advance GSI knowledge to discern key parameters, optimize GSI design and development, and enable future performance forecasts in a changing environment. For this project, civil engineers, computer scientists, and geographers are joining together to produce a new platform that uses AI in a dynamic environment with multiple data modalities, ranging from their spatial and temporal characteristics to data types. The research framework acknowledges the wider implications of GSI and its high interdependency and connection to the surrounding community and aims to improve social justice of GSI design through an equity-aware AI model. This project will use a large GSI monitoring relational database (housed at Villanova University) by combining GSI performance data and city-wide open data and applying machine learning methods to develop predictive models applicable across the US. This work targets advancing understanding of GSI dynamics by forecasting the performance of GSIs for a given array of conditions and constraints in urban settings to equitably maximize GSI community benefits. The project will support a diverse faculty team and engage students, urban communities, and industry and academic colleagues by: (1) creating state-of- the-art research and mentoring opportunities for graduate and undergraduate students from underrepresented backgrounds, (2) developing and delivering GSI learning modules for practitioners, and (3) integrating and promoting issues of equity and sustainability within urban stormwater management.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.