CAREER: Bridging Data, AI, Theory, and Education to Advance Global Urban Biodiversity Research

More than half of the world's people now live in cities, a share projected to reach roughly 70% by 2050. As cities expand, they profoundly reshape the biodiversity that underpins ecosystem services essential to human health and well-being, such as clean air and water, pollination, and climate resilience. Yet scientists still lack a clear understanding of what drives biodiversity patterns in cities worldwide, which species persist in urban environments, and how urban biodiversity will change under future environmental conditions. Progress has been limited because the necessary data are fragmented, inconsistent, and difficult to analyze at a global scale. This CAREER project addresses these gaps by integrating ecological theory, large-scale data, and artificial intelligence (AI) to transform the study of urban biodiversity and to make the resulting data products and tools openly available to the broader research community. The project advances the national interest by strengthening the scientific foundation for sustaining biodiversity in the cities where most Americans live, and by training a diverse, data-capable scientific workforce. Its integrated education program partners with K–16 schools in Madison, Wisconsin, training teachers to use motion-activated trail cameras and Raspberry Pi-based kits for biodiversity monitoring, engaging students in collecting and analyzing real data, and incorporating biodiversity data science into undergraduate and graduate curricula. These activities broaden participation in science, particularly in under-resourced classrooms, and cultivate the next generation of environmental data scientists. The project pursues three research objectives. First, it will build the first centralized, multi-taxon, open-access database of urban biodiversity spanning more than 1,400 cities in over 150 countries, integrating species occurrence records for plants, birds, mammals, and amphibians with functional traits, phylogenies, and environmental and socioeconomic variables. The database harmonizes data from global biodiversity repositories, community-science platforms, and structured surveys, following FAIR (Findable, Accessible, Interoperable, Reusable) principles. Second, the project will develop and openly share a generative AI forecasting framework that produces probabilistic forecasts of urban species assemblages and biodiversity metrics under multiple future climate scenarios through 2050, using high-throughput computing and validation against historical data. Third, leveraging this infrastructure, the project will enable community-driven research into the ecological, evolutionary, and socioeconomic processes that shape urban biodiversity, including how cities filter species and how broad-scale patterns generalize across regions and taxa. Together, these objectives advance ecological theory of community assembly in human-dominated landscapes, expand biodiversity forecasting capacity, and provide lasting open data and computational tools for macroecology and conservation planning. The use of generative AI to forecast biodiversity reflects current NSF priorities in artificial intelligence. 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.