Introduction to Data Analysis Using Geographic Information Systems

Capabilities of a GIS

Spatial Questions That Drive an Analysis

GIS applications are the result of spatial questions. The following questions are commonly asked:

1. What exists at a particular location on the face of the earth (locational analysis)?
2. When is a specific spatial condition satisfied?
3. What has spatially changed over time?
4. What kind of pattern will emerge from geographical data?
5. What will happen if certain phenomena are entered into predetermined scenarios?

Locational Analysis

Mapped data primarily indicates where objects are located, but cannot explain why. For example, an aerial photo may show that corn is growing vigorously in certain sections of a field, but cannot explain why it does not grow well in other areas. GIS analysis, on the other hand, may show a connection between corn growth, soil type, and available water by simultaneously examining computerized crop, soil, and soil moisture maps.

Spatial analysts continually seek patterns in mapped data. Defining the distribution of a crop creates a geographical pattern. Investigating the commercial feasibility of a potential crop may require an evaluation of how production, transportation, and markets are related to each other. The GIS is able to store the necessary data and to study these complex relationships in a fast, flexible manner.

Satisfying a Spatial Condition

Frequently a GIS user wants to discover whether the mapped data will meet certain conditions. Suppose someone wants to know where to situate new groundwater wells. The wells need to be located within 10 miles of a particular subdivision and need to occur along or adjacent to pipelines already serving that location. The general spatial conditions would be used to define a broad area in which the wells could be situated. Then, the specific location of each new well would be based, in part, on locational analyses performed using well-siting criteria.

Temporal Analysis

Both our perception of the world and the world itself are constantly changing. The growth of suburban cities and the effects on metropolitan land-use have changed dramatically in the past decade. Knowledge that is restricted to a single point in time may not be meaningful for answering some questions. For example, an analyst might look at the relationship between changing land-use practices and zoning law changes over many years. By storing and comparing maps of various dates, GIS can perform temporal analyses.

Emerging Patterns

GIS users seek spatial patterns. In other words, they want to know if two or more things vary similarly in space. They may want to know, for example, if there are a proportionately larger number of traffic fatalities along roads with posted speeds of 65 m.p.h. compared to those posted at 55 m.p.h. If there are, other questions such as, "Which stretches of roads are the most hazardous?" may follow.

Evaluating Different Scenarios

"Scenario building" is the result of "What would happen if . . . ." type questions. For example, what would happen to coastal areas if global temperatures increased, the icecaps partially melted, and sea levels were elevated? Here, the user employs a model designed to forecast and map the potential impact of climate changes on sea level in coastal areas. Application of such a model allows the user to construct a hypothetical situation and forecast the outcome.

Sometimes a change in assumptions results in a forecast that is substantially different than expected. For this reason, it is important to test a model's assumptions through a process known as "sensitivity analysis." Sensitivity analysis can be used to determine how much the outcome depends on (or is sensitive to) assumptions used by the model.

In the above example, the outcome (what would happen to coastal areas) partially depends on a change in sea elevation. Change in sea elevation depends on the degree to which icecaps are melted and changes in rainfall and evaporation which may result from a change in global temperature. If subsequent analyses are performed with the model assuming different feasible changes in melting, rainfall, and evaporation for a range of global temperature changes, how different are the projected effects on coastal areas? By testing different scenarios that vary your assumptions and evaluating the response, it is possible to determine how sensitive the model is to changes in assumptions.

Basic Database Requirements

A GIS must allow the operator to: (1) incorporate (import) data from outside sources, (2) easily update and alter data, and (3) ask data-related questions of (or query) the database. The database management system (DBMS) software that is a part of a typical GIS provides these capabilities. Also, commercially available database management software programs can be programmed to perform these same tasks outside of the GIS.

Importing and Expanding the Database

Data imported into a GIS often comes in the form of standard ASCII (American Standard Code for Information Interchange) files. ASCII is a standardized code that can be read by nearly all computer systems. A GIS may also be capable of importing data files that are in other formats.

Another method of expanding the database is to manually enter data. Every DBMS has the capacity to create new geographical objects of interest or records. This technique is commonly used to add a relatively small amount of new objects (a few printed pages) to the database.

The DBMS also allows for error checking as new records are created or existing ones are updated. Not all errors can be eliminated in this way, however, so care must be taken when collecting, automating, and changing the database.

A GIS must also provide the ability to create data files that can be exported to other systems. During the exporting process, data files are written in a common format (e.g., ASCII) to a file that can then be imported by other systems.

Updating Attributes

Another common task is updating or editing the database. Since no user can foresee all future data needs and applications, a GIS must provide ways to easily modify, refine, or correct the database. Attribute data are seldom static. Therefore, maintaining the currency of the data depends on updating capability.

Query the Database

Manipulating the database to answer specific data-related questions is accomplished through a process known as database analysis. Tabular output is the result of a database analysis query. Those output products might become part of a summary report or they might be imported into spreadsheet software where further analyses might be performed.

To query the database, logical expressions that impose limits or conditions on the database search are defined. These logical expressions specify which geographical objects are to be included in the analysis and/or how that data is to be analyzed. A subset of the database is produced. Some logical expressions are simple and require only one condition while others are very complex and contain multiple conditions.

For example, using a database that contains data about the location (township, range, and section), ownership, and size of lakes, the user could ask the simple question: "Where are all the lakes in my county?" In searching for the answer, the DBMS creates a data subset that is presented in tabular format (rows and columns of information). The subset created to answer this question meets a single condition: the condition that the data apply to lakes in that particular county. In contrast, the question "Where are the private lakes in section 34 of township 61 range 19 that are larger than 15 acres?" contains multiple conditions. As a result, the database creates a subset of the data which meets all of these conditions.

Another type of question that can be asked using logical expressions requires that mathematical analysis be performed. These questions prompt the DBMS to perform such functions as calculating population density, generating descriptive statistical summaries, and translating between measurement systems (e.g., converting area defined in square meters into acres or square miles). These types of operations are common when the database contains numeric data and a statistical analysis of the data is required.

Produced by Communication and Educational Technology Services, University of Minnesota Extension.

In accordance with the Americans with Disabilities Act, this material is available in alternative formats upon request. Please contact your University of Minnesota Extension office or the Extension Store at (800) 876-8636.

University of Minnesota Extension is committed to the policy that all persons shall have equal access to its programs, facilities, and employment without regard to race, color, creed, religion, national origin, sex, age, marital status, disability, public assistance status, veteran status, or sexual orientation.