Bird biodiversity sampling shows increasing geographical disparities associated with historical redlining in the United States

Historic segregation and inequality are critical to understanding modern environmental conditions. Race-based zoning policies, such as redlining in the United States during the 1930s, are associated with racial inequity and adverse multigenerational socioeconomic levels in income and education, and disparate environmental characteristics including tree canopy cover across urban neighborhoods. We quantify the association between redlining and bird biodiversity sampling density and completeness - two critical metrics of biodiversity knowledge - across 195 cities in the United States. We show that historically redlined neighborhoods remain the most under-sampled urban areas for bird biodiversity today, potentially impacting conservation priorities and propagating urban environmental inequities. The disparity in sampling across redlined neighborhoods grades increased by approximately 40% over the past 20 years. We identify specific areas critical for reducing uncertainty and increasing equity of sampling of biodiversity in urban areas. Our findings highlight how human behavior and past social, economic and political conditions segregate not just our built environment, but may also leave a lasting mark also the on the digital information we have about urban biodiversity. Methodology: The Global Biodiversity Information Facility (https://gbif.org) provided us with georeferenced biodiversity records for bird species found inside HOLC-defined neighborhoods in 195 cities. We searched for and downloaded bird recordings with geographic coordinates that were gathered after 1932 up to October 2022 using the rgbif package. The terms "observation," "live specimen," "human observation," "preserved specimen," and "machine observation" were all used to describe the recordings we compiled. Please see the README_data.md for description on the biodiversity data and the README_code.md for a description on the code used in this manuscript. For additional questions feel free to reach out to diego.ellissoto@yale.ed