Abstract
"Food deserts" are usually defined as geographic areas without local access to fresh, healthy food. We used community ecology statistics in supermarkets to quantify the availability of healthy food and to potentially identify food deserts as areas without a diverse selection of food, rather than a binary as to whether fresh food is present or not. We test whether produce diversity is correlated with neighborhood income or demographics. Abundance and diversity of fresh produce was quantified in supermarkets in Broward County, Florida, USA. Neighborhood income level and racial/ethnic makeup were retrieved from the U.S. Census and American Community Survey. Although diversity varied, there were no communities that had consistently less available fresh food, although the percent of a neighborhood identifying as "white" was positively correlated with produce diversity. There may be fewer choices in neighborhoods with a higher proportion of minorities, but there were no consistent patterns of produce diversity in Broward County. This method demonstrates an easy, inexpensive way to characterize food deserts beyond simple distance, and results in precise enough information to identify gaps in the availability of healthy foods.