What does the cell value mean on a point/line density output raster?
Last Published: April 25, 2020
The cell values of a point density output raster represent the number of points that fall within the neighborhood per area unit (without specifying a population field). For line density, they represent the length of the portion of each line segment that falls within the neighborhood per area unit (if not specifying a population field).
The density area unit is selected by default upon the linear unit of the projection from the input feature dataset (the feature dataset must be in a projected coordinate system in order to choose a different area unit from the tool drop-down list box):
• SQUARE_MAP_UNITS — If the units are unknown, points, or decimal degrees • SQUARE_MILES — For feet, yards, miles, or nautical miles • SQUARE_KILOMETERS — For meters or kilometers • SQUARE_INCHES — For inches • SQUARE_CENTIMETERS — For centimeters • SQUARE_MILLIMETERS — For millimeters
For example, if inputting a point dataset that is in feet, the area unit will be square miles by default. So a pixel value of 10 on the output raster means there are about 10 points per square mile within the searched neighborhood. Whereas for line input features, a pixel value of 10 means the average line length is 10 feet per square mile within the search neighborhood.
The pixel value does not have much to do with the output cell size of the raster or the projection of it. Output cell size is the resolution of the dataset, and its linear unit depends on the output raster projection. For example, a point density raster can contain density values measured per square mile yet the dataset is in feet (i.e., each pixel covers an area of 100 ft. by 100 ft. on the ground). It makes sense to modify it if planning to overlay the data with others for further analysis and would like to keep them consistent, or the raster is to be converted to vector data and it is necessary to control the smoothness of the vector geometry. The raster resolution has an affect on the output data file size (i.e., the finer the resolution, the larger the file size, in general).
The population field parameter can be used to weight some features more heavily than others, depending on their meaning, or to allow one feature to represent several observations. For example, one address might represent a condominium with six units, or some crimes might be weighted more heavily than others in determining overall crime levels.
For line features, a divided highway probably has more impact than a narrow dirt road, and a high-tension line has more impact than a standard electric distribution line. If a population field other than 'None' is specified, the 'length of the line' is considered to be its actual length times the value from the population field for that line record, and so for the Point Density tool.