Empirically-derived, locally responsive travel time thresholds for optimal geographic supermarket access using national commuting data

Abstract

This study introduces a novel method for assessing spatial access to supermarkets in the United States, addressing limitations of the US Department of Agriculture Food Access Research Atlas definition. Our method eliminates the food desert poverty requirement and derives empirical time-based thresholds that reflect a nuanced urban/rural spectrum and regional differences in normative drive times. Using these thresholds, we identify more than 66 million individuals experiencing suboptimal food access, distributed across 30,011 census tracts. Travel time thresholds ranged from 2.93 to 7.17 min (sd = 1.09) in urban blocks, 5.86 to 19.21 min (sd = 1.93) in suburban blocks, 2.68 to 12.98 min (sd = 1.81) in large rural blocks, and 9.74 to 28.13 min (sd = 3.39) in small rural blocks. The shortest thresholds were found in urban centers around New York City, Chicago, and Los Angeles, while the longest were in sparsely populated regions of Northern California, the Southwest, and the Great Plains. Our approach represents a significant methodological advancement in food access research and can be applied to measure variation in geographical access to other community resources and services.