Table of Contents
- Exploring Patterns of Crime In Los Angeles County Using A Density Surface Model
- Setting up Your Workspace
- Preparing the Data
- Skill Drill: Clipping the LA City Boundaries to the Census LA County Boundary
- Skill Drill: Adding XY Data
- Skill Drill: Clipping the Crime Data to the Census LA County Boundary
- Skill Drill: Creating a Subset of the Data Based on Crime Category
- Changing Global Environment Settings for Raster Processing
- Creating a Density Surface Model Using the Simple Method
- Creating a Density Surface Model Using the Kernel Method
- Skill Drill: Creating A Density Surface Model Based on Your Criteria
Skill Drill: Creating a Map of the Results
A surface model is an approximation of a surface based on a set of sample points. It represents a phenomenon that can be measured continuously across a surface. Elevation, slope, temperature, precipitation, population, crime, and disease are all examples of surface models that you may be familiar with or have seen before. There are two primary surface modeling techniques used to create surfaces, density, and interpolation.
Interpolation allows you to create a surface from point locations. It attempts to predict values at specific locations, in between known values. For example, a temperature surface may be interpolated from a series of weather stations across the state of California. A surface model using interpolation attempts to predict the specific temperature values at all locations across the entire surface.
Like interpolation, density also allows you to create a surface from point locations. However, density does not try to predict specific values at a location but instead communicates geographic trends. For example, a map of cancer density in a region does not predict that someone developed cancer at a specific location. Instead, it communicates the patterns and perhaps the likelihood of cancer over a region. Density is a function that distributes the quantity or magnitude of a phenomenon over a unit of area, such as population per square kilometer. It is derived from the quantity measured at each location and the spatial relationship of the locations of the measured quantities.
Estimated time to complete this tutorial: 4 hours
- Review how to acquire data from a public source
- Discover clustering patterns in point data
- Create a density surface using the simple method
- Create a density surface using the kernel method
- Create a density surface based on feature attributes
In this scenario, the Los Angeles County Sheriff’s Department is interested in analyzing patterns of crime to make administrative decisions on how to allocate resources. You will attempt to assist the Sherriff’s department by making a series of density surface models.
You will use the following criteria in your analysis:
- A history of incident responses
- A search neighborhood based on a manual inspection of the data
- A search neighborhood based on spatial autocorrelation
- A comparison between simple density and kernel density
Conduct this analysis using the Universal Transverse Mercator (UTM) system along with the North American Datum of 1983 (NAD83). Los Angeles County lies in Zone 11 of the UTM system. All of your data must be in this spatial reference system at the start of your analysis. Create working copies of your data in this spatial reference system using the Project tool in ArcMap as needed.