Geospatial Associations between Property Crimes and Land-use Features: A Case of Nashik, India

Understanding how configurations of urban land-use features influence the spatial concentration of property crimes is vital for developing evidence-based safety and planning strategies. This study evaluates the geospatial associations between eighteen distinct land-use features and three types of property crimes: robbery, burglary, and theft in Nashik, India. Using property crime records from 2022 and locations of land-use features extracted through the Google Maps API, the research integrates Kernel Density Estimation (KDE), Location Quotient (LQ), and Multiple Linear Regression (MLR) modelling to uncover spatial patterns and statistically significant relationships.

Using KDE, property crime subzones were identified to capture local-scale variations of property crimes and land-use features. LQ analyses revealed uneven geographic concentrations of property crimes and land-use features across subzones. MLR models revealed that several land-use features, including ATMs, banks, hospitals, police stations, recreational places, shops, and transit places, significantly influence property crime occurrences. However, the magnitude of influence varies across different types of property crimes. One-way ANOVA test results confirmed that the MLR models were statistically significant, validating the geospatial associations between property crimes and land-use features.

The findings underscore the importance of integrating spatial analytics with urban planning to enhance safety. By demonstrating how geospatial patterns of land-use features influence property crimes, this study contributes to urban geoscience research and provides actionable insights for urban planners, urban designers, and policymakers. Future research could extend the analysis to spatiotemporal patterns or apply the methodology to cities with different built environment morphologies.