Measuring Access to Amenities and Exposure to Risks among Manufactured Housing Units

In this study, we conduct the first national analysis of environmental amenities and risks among manufactured housing units in the United States. We use building footprints to develop a machine learning model that predicts the location of single-wide manufactured housing units nationwide. We use publicly available datasets to measure exposure to risks, including air pollution, facilities/sites regulated by the EPA, floodplains, and industrial jobs. We also measure access to amenities, including different types of land cover (e.g., green space, trees, and water), schools, SNAP retailers, transit stops, and walkability. Finally, we examine differences in access/exposure by census region, city governance and population, and neighborhood income. Our results highlight substantial differences in access and exposure to amenities and risks between manufactured housing units and other buildings. These risks vary by neighborhood income and city size, as well as across incorporated cities and unincorporated areas.