essedesignguide.org

There are several methods popularly used for spatial data analyses

Spatial data analyses go a step ahead of spatial data mapping. Spatial data analyses include all methods applied to spatial data that entail adding value to spatial data; converting data to information; revealing and explaining patterns and anomalies in the data set.

Spatial data analyses ultimately support decision making. Spatial data analyses mimic the way the human mind thinks. The human mind perceives space and instantly looks for patterns and anomalies, and tries to provide an explanation for the observations. There are several techniques for spatial data analyses. Here only the popular spatial analyses techniques used in a GIS environment, such as plotting distribution, querying, buffering, overlaying, and other complex analyses, are discussed.

Plotting distribution is similar to putting color pins on a map and finding patterns.  
 

Querying can be location based or attribute based. Location based or spatial query is the process of selecting features based on location or spatial relationship. For example you can run a spatial query to find out how many hospitals are located in a selected area. In a GIS environment you can frame a wide variety of queries to select features in one layer based on their location relative to features in another layer. Attribute based query is sometimes also referred to as aspatial query. It is being discussed here for the sake of completeness of data analysis techniques. With attribute based query one can retrieve and display all the geographic features that meet some criteria based on information stored in a layer’s attribute table.    

Example of attribute query: Plot epicenters of all earthquakes occurring from 1963 through 1998 AND having magnitude greater than 3.5

Buffering is a method that uses buffers or zones at specified distances around features or attributes. Buffers can be both constant and variable-width and a user can generate a single buffer zone or multiple buffer zones. A user’s interest may be inside the buffer zone or outside the buffer zone. For example, for the purpose of disaster preparedness, a user may be interested in finding out the sphere of influence of a volcano and may also want to know (spatial query) if there are any towns located within the sphere of the volcanoes influence. 

A buffer zone around known volcanoes in the Alaska – Kamchatka region created to delineate regions of potentially greatest impact during a volcanic eruption

Overlaying offers a powerful means to retrieve desired information from multiple sources of geospatially referenced data sets in a GIS environment. A user can choose to extract features that are common or unique in different information layers. For example a city GIS may contain information layers on property prices, transportation networks, natural waterbodies, etc. A user may want to buy property in an area that is close to the bus route, but far away from a potential flood zone, and yet is in a price range that he/she can afford. In this case, using available data layers and carrying out overlaying operations can serve as a powerful decision support mechanism.

GIS tools offer several options for information extraction from georeferenced data layers. More complex selection criteria can be defined depending on user needs. For example, a user may want to select all features that:

- intersect a named feature
- are within a specified distance of an identified feature
- completely contain certain features
- have their center in a particular region
- share a line segment with a particular polygon
- touch the boundary of a particular polygon
- are identical to a named feature
- are crossed by the outline of an identified region
- contain, or are contained by other features or regions
- etc.

 Reference: Center for Spatially Integrated Social Science
Bailey and Gattrel, 1995, Spatial Data Analysis by Example