Creating a Density Surface Model Using the Simple Method
A density surface model calculates the density of a phenomenon by using known values within a user-defined search neighborhood. The search neighborhood you define will have the most significant impact on the resulting surface. It is also the most arbitrary factor when creating a density surface model. Determining the most appropriate search neighborhood for your dataset can take time, research, and a good understanding of the data. There are also some tools you can use to determine an optimal size. In this example, your data is non-weighted. Non-weighted means that each data point represents only one incident. One point is no more important or has no more weight than any other point. For situations like this, using the average distance between data points to determine the size of the search neighborhood is an excellent place to start.
From the Analysing Patterns toolbox under Spatial Statistics tools, launch the Average Nearest Neighbor tool. The Average Nearest Neighbor tool determines the average distance from each feature to its nearest neighboring feature. In this instance, it will calculate the average distance between data points.
For the Input Feature Class, choose your drunk driving data. Leave all other settings as default and click OK.
If you remembered to disable background geoprocessing, the results appear in the geoprocessing window. If not, you may have to search for your geoprocessing results under the Geoprocessing menu from the main menu in ArcMap. Make a note of the observed mean distance. This distance is a good starting point when determining the size of the search neighborhood for non-weighted data.
When using the Simple Method, the cell values in a density surface model are a function of the number of data points in the search neighborhood divided by the area of the neighborhood. The value of each cell is calculated individually based on the number of data points in the search neighborhood. From the Density toolbox under Spatial Analyst Tools, open the Point Density tool.
For the Input Point Features, choose your drunk driving incidents data. Save the Output raster to your working folder. For the Output cell size, enter 30. The smaller the cell size, the smoother the surface will look. However, the file size will also be more substantial. In this example, 30 meters is a good balance given the size of Los Angeles County. Under Neighborhood, choose Circle as the shape. Next, to the Radius, enter the Observed Mean Distance value from the results of your Average Nearest Neighbor tool. Next to Area units, change the unit of measurement to SQUARE METERS. This setting will ensure that the units match the results of the Average Nearest Neighbor tool. Leave all other settings as default and click OK.
A new density surface model is added to the Table of Contents. In this example, ArcMap used a default blue color scheme. The layer is symbolized using the Equal Interval classification method and nine classes. To observe patterns using a more familiar color scheme for hot-spot analysis, you can change the color ramp.
Open the layer properties for the simple density surface model layer and navigate to the Symbology tab. For the Color Ramp choose the Cold to Hot Diverging color scheme. You can view the color ramp labels by right-clicking on the color ramp and un-checking Graphic View. When you are ready, click OK.
The results will be a hot-spot map of incidents related to drinking and driving over the past 30 days based on the Simple Density method.