Unveiling the green shield: How green spaces thwart urban crime

Discover how parks and greenery combat crime, creating safer urban environments through social cohesion and ecological balance.

Environmental criminology aims to understand crime by examining time and space. This helps develop crime prevention measures for communities, and advances in mapping technology make features like green spaces more tangible. These include areas with trees, parks, and gardens. In cities like Chicago, where crime is a major concern, exploring how greenery affects crime is crucial for public safety.

As more nuanced research in this area has expanded over the last twenty years, there has been strong evidence to suggest that unmaintained and blighted vegetation (e.g., vegetation found within vacant lots) is associated with criminogenic activity, and maintained and healthy vegetation (e.g., vegetation found within park areas) is regarded as being protective against crime. These findings have been observed in urban settings across the United States, including in Chicago. Research has indicated that greening is associated with reductions in assaults and other violent crimes, as well as reductions in burglaries and thefts.

How does greenness deter criminal behaviour in neighbourhoods?

Busy Streets Theory suggests that social interaction fosters a positive atmosphere, building social cohesion, trust, social capital, and collective efficacy in communities. Busy streets are represented by actively maintained, organised spaces and visible social interaction, and can be manifested through reduction of blight and vacant areas through greening interventions.

Busy streets are described  by three components which provide the context for busy streets and community empowerment: i) the intra community component, which represents the social relationships among neighbourhood residents; ii) the interactional component, which represents individual and organisational social interactions that promote trust and social capital; and iii) the behavioural component, which represents organisational collaborations between residents and organisations within communities and collective actions taken to improve a neighbourhood.

Normalised difference vegetation index – A novel approach for measuring greenness

Captured by remote sensing satellite imagery from the United States Geological Survey, normalised difference vegetation index (NDVI) is a quantitative metric used to determine the density of greenness on a patch of land by observing the distinct colours (wavelengths) of visible and near-infrared sunlight reflected by vegetation. NDVI offers a level of granularity beyond low-level aerial photography estimation, capturing greenness in a precise manner (compared to solely examining park acreage or tree canopy percentage from low-altitude aerial photography, for example) because every pixel in a high-resolution satellite image contains a unique vegetation index quantity. NDVI values range from -1 to +1, with values closer to +1 representing denser, healthier vegetation. Negative values or values near zero represent built-up areas or water.

Leveraging and integrating unique and novel data sources for quantifying features of the physical space will aid ongoing efforts to prevent crime in neighbourhoods.

Shaun Bhatia

The effects of greenness on crime in Chicago

In our study, we examined the effects of greenness, as measured using NDVI, on crime in Chicago census block groups. Both greenness and crime were measured by meteorological season between 2014 and 2018, and a longitudinal statistical strategy was utilised. NDVI by block group was calculated by averaging the NDVI values for individual pixels encompassed in a block group. Geocoded crime incident data from the Chicago Police Department was similarly joined to each individual block group, and subsequently converted to rates (crime incidents per population size).

Our analysis factored in various neighbourhood traits, like disadvantage and population density. We also considered “spatial dependency,” where nearby areas influence each other’s crime rates. This gave us some assurance that our findings were accurate and unbiased. Additionally, we used a “two-way fixed effects” method to address space-time changes in crime rates that were not linked to greenness or other factors we studied.

Figure 1. Average total crime rate (left) and average normalised difference vegetation index (right), Chicago, 2014-2018
Credit. Author

In Figure 1 (left side), we observe the average crime rate per 1,000 to be largely concentrated in the south and west of the city, regions also known to have high poverty levels, unemployment, and housing disinvestment. NDVI (Figure 1, right side) tended to be more evenly dispersed than crime (Figure 1, left side), with pockets of high NDVI throughout the city. When we controlled for neighbourhood characteristics and spatial dependency, we observed an inverse association between NDVI and crime rate, suggesting that as greenness in Chicago increases, the crime rate decreases.


The goal of this study was to explore the relationship between greenness and crime from a city-wide perspective, and expand upon the wealth of cross-sectional research that has been conducted in this area by using a longitudinal study design. Priority was placed on the precision of unit of place using block groups as our spatial unit, while utilising a gradient approach for measuring greenness using NDVI, and a spatial analytic strategy.

Using a longitudinal approach in our study had two particular benefits. Firstly, from a within-year perspective, we were able to account for seasonality, which is important in cities such as Chicago, where seasonal traits may be instrumental in predicting crime rates. Secondly, from a between-year perspective, the use of a longer study period and two-way fixed effects specification allowed us to estimate long-term population-level processes that were potentially omitted in our analyses.Β 

Greening interventions have shown great promise in reducing crime in cities across the United States. They represent potentially cost-effective, and long-lasting strategies for enhancing social cohesion and reducing crime. We hope this novel work will build upon existing research, continue to promote future inquiry in this space of environmental criminology research, and help inform future evaluation of greening intervention in urban settings.


  • Local governments should continue to promote green infrastructure, particularly in neighbourhoods with historical and systemic disinvestment.
  • Programs may consider engaging community members in greening activities to improve neighbourhood trust and empowerment.
  • Evaluation of the public safety implications of greening programs should be considered vital in the implementation and maintenance of interventions.

πŸ”¬πŸ§«πŸ§ͺπŸ”πŸ€“πŸ‘©β€πŸ”¬πŸ¦ πŸ”­πŸ“š

Journal reference

Bhatia, S., & Jason, L. (2023). Greenness and crime in the city: an investigation using remote sensing and spatial panel models. Applied spatial analysis and policy16(1), 229-257. https://doi.org/10.1007/s12061-022-09477-9

Shaun Bhatia was born and raised in Chicago, Illinois. He attended the University of Illinois at Urbana-Champaign, where he earned his Bachelor of Science in Statistics and Psychology, and his Master of Science in Epidemiology. He obtained his Doctor of Philosophy in Community Psychology from DePaul University in Chicago. During his professional career, he has worked primarily as a quantitative researcher in industry, non-profit, government, and academic settings. His primary research interests are in the domains of community violence prevention evaluation, spatial econometrics, and environmental criminology. He is currently a postdoctoral research fellow at the University of Michigan, researching firearm injury prevention in community settings.