Environment/Climate Change
Spatial Patterns and Predictors of Illegal Dumping in Mississippi Nina Franzen Lee* Nina Lee Marcos Luna Sarah Kountouris Cristina Nica Erica Walker
Background Illegal dumping of waste is an increasingly difficult problem to solve and persists despite strategies to curb the prevalence. Dumping sites can contaminate soil and water, increase health risks for neighboring communities, and dampen economic investment in already vulnerable areas through visual pollution. Mississippi is an understudied area with a long history of environmental health challenges and injustices that routinely appears at the bottom of state health and economic rankings. Aim To identify patterns and drivers of uncontrolled illegal dumping sites across MS at the neighborhood level to better understand this exposure and provide foundational data for future epidemiologic analyses and interventions. Methods Publicly available data on illegal dumping sites were combined with sociodemographic and spatial predictors at the site and census block group level. Sites were mapped, and a clustering analysis was performed. Descriptive statistics explore key predictors and outcomes. A negative binomial model was employed to account for the excess variance in the count data. Model selection was guided by AIC, Moran’s I, and LRT (p-value: 0.952) and a standard model was used with no spatial lags. Results The model has a pseudo R2 of 0.64. Significant predictors include proximity to hazardous waste facilities (positive), number of wood and paper manufacturing facilities (positive), presence of Environmental Justice categories exceeded (positive), higher percentage of group quarters populations (positive), and support from waste tire assistance grants (negative). Conclusion There are spatial patterns of illegal dumping in MS potentially aligning with industry, existing policy interventions, and environmental vulnerability. These results suggest that with improvements in illegal dumping data, higher resolution vulnerability predictions can be made, with potential for making progress on much-needed epidemiological studies and assessing policy interventions.