Understanding analysis in the ArcGIS Spatial Analyst extension

The easiest way to understand cell-based modeling is from the perspective of an individual cell (the worm's-eye approach) as opposed to the entire raster (the bird's-eye approach). To do so, think of yourself as a cell in a raster dataset. You represent a location, and you have a value. All ArcGIS Spatial Analyst extension tools will ask you to manipulate or retain your value based on a defined series of rules.

Calculating an output value for each cell

To calculate an output value for your specified location (cell) using any Spatial Analyst tool, there are three things you need to know:

• The value of your specified location (cell)
• The manipulation that will be performed on the value
• Which other cell locations and their values to include in your calculations

How do you determine these three things?

• You automatically know what the value is for your location (your input cell value).
• Each tool in Spatial Analyst manipulates the value at your location in different ways. The documentation for the tool indicates the logic behind how the value is manipulated.
• With some Spatial Analyst tools, you can calculate an output value by knowing only the value of your location, such as raising your value by a specified power. With certain other tools, you can determine the values of other locations within the raster dataset to which your specified location belongs, such as looking in a neighborhood around you (a focal tool) or including cell locations and their values defined by other raster datasets (zonal tools).
• This three-step process occurs for each location (cell) in the raster dataset within any Spatial Analyst tool. All tools work on a cell-by-cell basis, and each calculation for each cell requires the value of the cell, the manipulation that is being applied, and other cell locations to include in the calculations. The Spatial Analyst tools are grouped into categories based on how they manipulate values—you only need to understand how the cell values are manipulated in the different categories.
• For many tools, you can refine how the manipulation (the calculations) will be performed through user-defined parameters. For example, the cells to include in each calculation for a focal tool may vary based on the neighborhood that is defined.

The three-step process considering several tools

• Power tool: When applying this tool to your raster dataset, you need to know how to take the value of your location's value and raise it to the power defined by the value of the same location your cell represents in the other raster dataset to return an output value for your location.
• Cos tool: When applying this tool to your raster dataset, you must know the value of your specified location and how to take the cosine of that value to return an output value.
• Euclidean Distance tool: When applying this tool, you must know your location and determine how far your location is from the closest source (which is defined by a source dataset) to return an output value for your location.
• Focal Statistics tool: When applying this tool to determine the maximum value within a 3-by-3 neighborhood, you must know your location's value and the values of the eight immediate neighbors around your location. You will calculate a specified statistic for the nine values and apply the resulting value to the output at your location.
• Zonal Statistics tool: When applying this tool with the Mean option, you must know your location's value and take the mean of all the values of the cells that belong to the same zone as your cell, which is defined by a zone raster dataset.

The above logic occurs for each cell in the input raster.