Data analysis with GIS

When trying to relate wetland maps to rainfall amounts recorded at different points, it can be difficult to do so, especially in places such as airports, schools, and television stations. A GIS can be used to visualize two-dimensional and three-dimensional characteristics of the Earth’s surface. This also includes the atmosphere and subsurface from informational points as well. GIS can quickly generate map data from contour lines that have the ability to give the indications of differing amounts of precipitation. This type of map is called a rainfall contour map.
Many methods are able to estimate the characteristics of surfaces from limited point measurements and require a high level of sophistication to do so accurately. Two-dimensional contour maps that are created from the surface modeling precipitation points can be overlaid with any other map in GIS covering the same areas. This derived map is able to provide additional information; in this case, this would be the potential viability of water power as a renewable energy source. GIS can be used to compare many other renewable energy resource viability options for any geographic region.
Additionally, from a series of three-dimensional points, elevation contours can be generated from slope analysis, which would make it to easily define watershed locations by computing all the areas uphill from any point of interest. An expected line connecting the lowest points of successive cross sections along the course of a valley or river can be computed from elevation data as well.

From all of this, we are able to ascertain that there are five main steps in the data analysis process:

1. Framing the question
2. Exploring and preparing the data
3. Choosing the methods of analysis and the tools
4. Performing the analysis
5. Examining and refining the results

Framing the question: It is a good idea to frame the question to make the subsequent steps easier to go through, for example, frame the question in a manner that helps determine which GIS tools and methods will be used for analysis.

Exploring the data: This step is known to be the most time-consuming, and you aren’t guaranteed to have all the data needed for the analysis. It is a good idea to know the data format that will be used, how current that data is, the scale and detail of the data, the coordinate system used, whether the data uses any geometry work with the analysis, whether the data has the attributes needed, and whether the data has any access or usage constraints.

Preparing the data: During this step, the data format to use will be extremely important to know as it will determine which set of tools will need to be used. Make sure the data is organized, the data is readily extractable, and there are no errors that occur when using the data in the tools that will be used.

Choosing the methods of analysis and the tools to be used: The methods and tools should be readily and easily defined by the question framed. Generally, the question should have a direct one-to-one for the methods and tools, and having a simple diagram for the analysis is considered good practice. A simple example is provided in the following image:

Performing the analysis: Since diagramming is considered good practice, all that needs to be done here is to follow the tasks in sequence. The more complex the analysis, the more it may be necessary to create a model with ModelBuilder to automate the process, which will make it easier to change a parameter and run the model again for different scenarios.

Examining and refining the results: This step is just to look at the results and see if there are additional parameters missed in the original question and add some tweaks to better fit the original vision of the question.

For a more detailed look, along with some tutorials on the steps in the data analysis process, visit http://www.bcps.org/offices/lis/researchcourse/data_process.html.