Examining the relationship between tree canopy coverage and crime: An examination of within-neighborhood change

Abstract

A bourgeoning area of research suggests that tree canopy coverage can reduce crime in neighborhoods. However, nearly all prior studies have analyzed the effects of tree canopy coverage on crime across neighborhoods at a single timepoint. Thus, it is necessary to examine change within neighborhoods of the relationship between tree canopy coverage and aggregate crime outcomes. For the present study, we conduct a longitudinal block group analysis of tree canopy coverage, sociodemographic characteristics, and crime in Washington, DC. To model within-neighborhood change, we employ an unconditional Poisson-based regression estimator with dummy variables to represent fixed-effects. The findings show that tree canopy coverage can indeed reduce crime within neighborhoods over time, and with substantive magnitude. Yet, these salutary effects pertain to violent crimes only. In fact, we determine that tree canopy coverage leads to more motor vehicle thefts. The implications of these findings for criminology, urban studies, and public policy are discussed.