Uneven biodiversity sampling across redlined urban areas in the United States

Citizen science data has rapidly gained influence in urban ecology and conservation planning, but with limited understanding of how such data reflects social, economic, and political conditions and legacies. Understanding patterns of sampling bias across socioeconomic gradients is critical to accurately map and understand biodiversity patterns, and to generating representative and just environmental knowledge. In this study we explore how historic racially-explicit zoning policies (redlining) relate to biodiversity data collection across and within 195 metropolitan areas in the United States covering >30 million people across 38 states. We specifically look at birds, as they are the most widely studied group of animals, and hundreds of thousands of citizen scientists collect biodiversity data each year. We consistently find uneven bird observation sampling density across redlined areas, with so-called ‘desirable’ areas (i.e. historically white areas) having more than twice the density than areas redlined as ‘hazardous’. We further estimate the degree to which historically redlined areas are surveyed sufficiently and identify regions across all metro areas in need of enhanced bird surveying. After accounting for differences in vegetation, open space, and climate, we find significantly lower sampling density and sampling completeness in these redlined neighborhoods. Our results shed light on the importance of considering socio-political conditions in the interpretation of urban biodiversity estimates. We conclude by discussing specific policy implementations– such as the Justice 40 Initiative and propose a new EPA indicator – and opportunities in collaborating with bottom-up community and social justice organizations for a more representative understanding of biodiversity in urban areas.