The first tome of this series (Developing a new GIS, August 2007) stressed the importance of the data model. Remember the data model?

Figure 1. Example of a land base data model.

The logical data model describes a complex version of the real world in a database. Physically, a data model appears as a list of fields and descriptors in layout form. But the data model ultimately drives the bus.

Once we establish our data model, what do we do with it? It’s about the data, right? So how do we work with data in our new GIS? Let’s look at how the data model works.

Figure 2. Overview of an urban data model.

Figures 2 and 3 illustrate the need to approach GIS with three-dimensional thinking. A large part of the negative criticism about GIS data is rooted in this lack of understanding. Figure 3 describes how the different elements of a base map relate to each other. Visualize the layered structure. Then look again at how the tables in the land base data model relate to each other.

In this process, we first need to develop our data model. Next, we need to acquire our data. Then we need to put that data someplace where we can manage it and use it effectively. We require a container for the data in our model before it can become useful as GIS data. In this example, we will create and load a standard ESRI geodatabase (.gdb) file.

Figure 3. View of a data model showing relationships.

The ESRI manual1 gives 11 steps in geodatabase design:

1. Identify the information products that you will create and manage with your GIS.
2. Identify the key data themes based on your information requirements.
3. Specify the scale ranges and spatial representations of each data theme at each scale.
4. Decompose each representation into one or more geographic data sets.
5. Define the tabular database structure and behavior for descriptive attributes.
6. Define the spatial behavior and integrity rules for your datasets.
7. Propose a geodatabase design.
8. Design editing workflows and map display properties.
9. Assign responsibilities for building and maintaining each data layer.
10. Build a working prototype and review and refine your design.
11. Document your geodatabase design.

Figure 4. ArcCatalog displaying new geodatabase menu.

Many of these steps were performed or discussed as part of our plan development in the second part of this series (October 2007). We are now ready to test our plan with a practical exercise. So let’s create a geodatabase. To do that, we need to launch ArcCatalog2.

In ArcCatalog, we navigate to the folder in the catalog tree where we want to locate our new geodatabase and right-click on it. We then click on “New.” In ArcGIS 9.2and newer versions, there are two geodatabase choices: Personal or File Geodatabase. Let’s choose File Geodatabase. A new folder appears in the tree. We now name our new geodatabase. (See Figure 4.)

Figure 5. “Import Feature Class” dialog box displaying “Query Builder.”

OK, we have now created our new geodatabase (GDB) container. When you view the folder where we created the GDB in ArcMap, it appears as a standard “container” icon. Currently, our container is empty. We need to add the data layers to the geodatabase to make it functional. To do that, we execute the “Import” command in ArcCatalog by right-clicking on the container icon and bringing up the menu. We then select “Import.” That action activates the submenu, which lists the types of data available for import.

We will select “Feature Class (single).”3This action brings up the “Import” dialog box. In the top box, “Import Features,” click on the file icon and navigate to the data source you wish to import. The data source can be Shape files, coverages or CAD files. In the “Output Feature Class,” give it a unique name. You can select only the portions of the data set you wish to import by using the “Expression” button, which activates the “Query Builder.” You can also add or delete fields from your “Output Feature Class” using the “+” or “X” buttons in the dialog. Click “OK” to import your new feature class. (See Figure 5.)

Figure 6. View of complete Parcel Feature Class.

We need to repeat these steps for each of our data feature classes. The result is that we have loaded a Feature Data Set4in our geodatabase. So, we have constructed and populated our geodatabase. Now, what can we do with it? Before we explore that, let’s take a brief look at the “Query Builder” function.

ArcGIS uses a standard computer SQL (Structured Query Language) for accessing and managing its databases. SQL is not used to add or delete any data in ArcGIS. But there are a variety of ways SQL Expressions are used in ArcMap and ArcCatalog to manipulate data. In ArcGIS, the dialog for creating an SQL Expression is the Query Builder.

Figure 7. Layer Properties and Query Builder dialogs.

In ArcCatalog, the user might build an “Expression” to identify items to exclude from a feature class. The result of that action tends to be invisible. To better demonstrate how the Query Builder works, we will use an example where the result is displayed. Open ArcMap, and in the Map View, you’ll see the Parcel Layer displayed. (See Figure 6.)

Then in the data frame, right-click on the layer we want to query. From the menu, choose “Properties” then “Definition Query.” Then click the “Query Builder” button. (See Figure 7.)

Figure 8. ArcMap view displaying queried parcel data set.

We are going to use the Query Builder to change the Map View to display only the parcels with areas greater than two acres. We simply double-click on the [ACREAGE] field and click on the “>“ operator and enter the number 2. Click “OK,” “Apply” and “OK.” That changes the display. (See Figure 8.) To return to the original data view, simply clear the Expression.

Using GIS tools to view our data in this dynamic fashion is called geovisualization.5This process demonstrates the power of working with geographic data in a system as opposed to viewing it on a two-dimensional analog map product.

Now that we have our GIS data all neatly organized in a container, what’s next in our plan? Our data may be ready for our users, but are our users ready for our data? That will be determined in the training plan--the next installment of this series to be explored in August.


1. This online file can be accessed from the Help menu in ArcGIS 9.2.

2. If you need to refresh ArcCatalog basics, review the article “Data management with ArcCatalog,” POB, April 2006.

3. A feature class is a collection of geographic features with the same geometry type (such as point, line or polygon), the same attributes and the same spatial reference. Feature classes can be stored in geodatabases, Shape files, coverages or other data formats. Feature classes allow homogeneous features to be grouped into a single unit for data storage purposes. For example, highways, primary roads and secondary roads can be grouped into a line feature class named “roads.” In a geodatabase, feature classes can also store annotation and dimensions.

4. A feature data set is a collection of feature classes stored together that share the same spatial reference.

5. Geovisualization (short for geographic visualization) refers to techniques and tools designed to interactively “visualize” spatial phenomena.

To view the first three installments of this series, visit and search on “Developing a New GIS.”