Improving the Understanding of the U.S. Urban Racial Geography and Its Temporal Change Using the Racial Landscape Method

Abstract

Racial landscape (RL) is an innovative methodology for studying racial geography that offers several advantages over current approaches. This article aims to highlight two key features of RL that have the potential to significantly impact the discipline. First, the RL approach introduces a fundamentally different method for assessing segregation compared to existing methods. The RL method allows us to calculate segregation for any arbitrary area without the need for subdivisions, diversity measures, reference regions, or reliance on census geography. Importantly, by using data from fifty-one Metropolitan Statistical Areas (MSAs) across the United States, we demonstrate that the RL’s segregation metric produces comparable rankings of segregation among MSAs when compared to existing segregation indexes. Thus, although the RL expands the scope of problems where segregation can be quantified, it remains compatible with current segregation assessment practices. Second, we use data from the core parts of four selected MSAs in 1990 and 2020 to showcase how high-resolution RL-based racial maps can be employed for spatially explicit visual analyses of racial change. We discuss the potential impact of RL on the field, particularly in relation to segregation assessment and the evaluation of spatially explicit models of racial dynamics.