Risk Terrain Modeling
Risk Terrain Modeling for Crime Prediction and Prevention · Also known as: environmental criminology, RTM analysis, crime risk mapping
Risk Terrain Modeling (RTM) is a geospatial crime prediction method that identifies high-risk locations by analyzing environmental and geographic features that attract or facilitate crime. Developed by Joel Caplan, Lichen Kennedy, and James Miller in 2011, RTM bridges environmental criminology theory with geographic information systems (GIS) to create predictive risk maps. Unlike methods that predict offender location (e.g., geographic profiling), RTM predicts where crimes are likely to occur based on terrain characteristics, infrastructure, and social environmental factors.
Read the full method
Sign in with a free account to read this section.
Method map
The neighbourhood of related methods — select a node to explore.
When to use it
Risk Terrain Modeling is most valuable for place-based crime prevention in urban and suburban areas with available geospatial infrastructure data and sufficient historical crime records (minimum 3–5 years). It works best for predicting acquisitive crimes (robbery, burglary, auto theft) where environmental factors play a strong causal role. RTM is less effective for predicting purely interpersonal crimes (domestic violence, sexual assault by acquaintances) where victim-offender relationship dominates location factors. Requires reliable GIS data layers and that environmental factors remain relatively stable during the prediction period.
Strengths & limitations
- Provides actionable, place-based insights for focused crime prevention and resource allocation
- Incorporates multiple environmental and contextual factors into a single predictive model
- Statistically grounded: uses hypothesis testing to identify significant risk factors, avoiding arbitrary factor inclusion
- Separates place-based risk (terrain characteristics) from person-based risk (offender location), enabling distinct interventions
- Applicable to varied crime types and geographic scales (neighborhoods, city districts, regional analysis)
- Requires comprehensive, accurate geospatial data; missing or outdated data layers degrade model quality
- Cannot account for temporal dynamics such as seasonal crime variation or rapid environmental change
- Ecological fallacy: high-risk locations identified by RTM may not reveal the actual mechanisms driving individual offender decisions
- Model performance depends on historical crime data quality; biased data collection or under-reporting produces skewed risk surfaces
Frequently asked
How is Risk Terrain Modeling different from crime hotspot identification?
Hotspot analysis identifies where crimes have clustered in the past (descriptive). RTM explains why crimes concentrate in certain areas by analyzing the environmental characteristics that make those places crime-prone (predictive and causal). A hotspot and a high-risk terrain may overlap, but RTM also identifies emerging high-risk locations with few past crimes but conditions favorable for future crime.
Can RTM predict individual offender locations?
No. RTM predicts where crimes are likely to occur based on place characteristics. Geographic profiling predicts where an offender likely resides. The two methods answer different questions and can be used complementarily in an investigation: geographic profiling narrows offender location; RTM narrows crime occurrence location.
What environmental factors are most commonly significant in RTM?
For acquisitive crimes, factors often include proximity to major roads and highways, density of bars and nightlife, presence of vacant properties, population density, and proximity to transit stations. For violent crime, factors include social disorder indicators (visible signs of crime), retail gun dealers, and areas with high population turnover. Factors vary by crime type and local context.
How is the composite risk score calculated in RTM?
After identifying significant risk factors, RTM typically uses weighted regression coefficients to combine factors. Each factor is assigned a weight based on its statistical relationship to crime density. The composite score is the sum of weighted factor values. Alternatively, kernel density estimation weights all factors equally, smoothing crime concentration across space to create a continuous risk surface.
How often should an RTM model be updated?
RTM models should be refreshed annually or biannually to account for environmental changes and crime pattern shifts. If a major environmental change occurs (opening of a new transit station, demolition of a significant structure, zoning change), the model should be re-estimated sooner. Risk predictions degrade rapidly if underlying conditions change without model updating.
Sources
- Caplan, J. M., Kennedy, L. W., & Miller, J. (2011). Risk terrain modeling: Brokering criminological theory and GIS methods for crime forecasting. Journal of Research and Practice in Criminal Justice, 17(1), 56-69. link ↗
- Kennedy, L. W. (2008). Crime and Environment. Routledge. link ↗
- Brantingham, P. J., & Brantingham, P. L. (1991). Environmental criminology. Sage Publications. link ↗
How to cite this page
ScholarGate. (2026, June 3). Risk Terrain Modeling for Crime Prediction and Prevention. ScholarGate. https://scholargate.app/en/forensics/risk-terrain-modeling
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
- Crime Linkage AnalysisForensics↔ compare
- Geographic ProfilingForensics↔ compare
- Network Analysis of Case LawForensics↔ compare