Creating a feature dataset
Setting up a feature dataset involves three primary tasks:
- First, create a new feature dataset.
- Add the set of feature classes that will be held in your feature dataset and populate their contents (that is, add features). See Creating feature classes. Note:
Only the person who created the feature dataset can add feature classes to it.
- Add optional geodatabase objects to your feature dataset to model relationships and add rich behavior. Options include adding a relationship class, a topology, a network dataset, a terrain, or a geometric network.
- In the Catalog tree, right-click the geodatabase in which you want to create a new feature dataset.
- Click New > Feature Dataset.
- Type a name for the feature dataset.
- Navigate to the spatial reference you want to use. Alternatively, click Import and navigate to the feature class or feature dataset that uses the spatial reference you want to use as a template.
- Use Modify if you want to change any parameters in the coordinate system you've chosen. Edit the coordinate system's parameters and click OK.
- If your data requires a vertical coordinate system for z-units, you may import one from another feature class or feature dataset; otherwise, select None. Click Modify if you want to change any parameters in the coordinate system you've chosen. Edit the coordinate system's parameters and click OK.
- Enter values for the xy-tolerance, z-tolerance, and m-tolerance, or accept the default value, which is the equivalent of 1 mm in real-world units.
- Choosing the z-tolerance and m-tolerance should follow a strategy similar to that used for setting the xy-tolerance (that is, choosing the default in most situations works well).
- By default, the Accept default resolution and domain extent check box is checked. If you want to manually adjust the domain and extent values of your new dataset, uncheck this box. If you are creating data in a pre-9.2 geodatabase or chose Unknown for the horizontal coordinate system, you should uncheck this box and confirm that the default values are appropriate.
- If you have unchecked the Accept default resolution and domain extent check box, there is an additional dialog box in the wizard that allows you to change the xy-, z- and m-resolution as well as the minimum and maximum z- and m-values.
- Click Finish.
- You may want your most frequently used coordinate systems to appear at the top of the list in the wizard so you don't have to navigate down through the tree. First, quit out of the wizard. Now, in ArcCatalog, turn on the Coordinate Systems folder (Customize > ArcCatalog Options under the General tab) if it is not already turned on. Inside the Coordinate Systems folder, drag and drop, or copy and paste, the coordinate systems you use most frequently into the top level of the folder. For example, copy a coordinate system from inside one of the subfolders,then right-click the Coordinate System folder and choose paste. (If you drag and drop coordinate systems to the top level of the Coordinate System folder, they are automatically copied, not moved, by ArcCatalog, so they will still be accessible via their original subfolder too). You can also create new subfolders or reorganize the existing subfolders. Now, when you go into the New Feature Class wizard again, you'll see those coordinate systems that you moved to the top of the Coordinate Systems folder at the top of the list.
- Click Import to populate the Spatial Reference Properties dialog box with information from another feature class. You can then customize the template's spatial reference.
- To modify a predefined (or a template's) coordinate system or to define a custom coordinate system from scratch, click Custom on the Coordinate System dialog box.
- Since the size of the spatial domain is dependent on the value of precision, when the precision is changed, the maximum z-value will change to fit within the size of the spatial extent. Similarly, when the maximum z-value is changed, the precision will change to fit the domain extent.
Choosing the default value for each tolerance is a very good choice in all but extreme cases.
The xy-, z-, and m-tolerances are used to set the distance that coordinates are allowed to be adjusted by the software during various processing operations.
The default x,y-tolerance is set to 0.001 meters or its equivalent in the units of the dataset's coordinate system. For example, if your coordinate system is recorded in feet, the default value is 0.003281 feet (0.03937 inches). If your x,y coordinates are stored in decimal degrees, the default value is 0.0000000556.
Using the default xy-tolerance works extremely well in almost all situations unless you have an extreme case in which you have incredibly precise data (i.e., coordinate accuracy better than survey-grade coordinates). If you need to consider an alternative xy-tolerance value, see Coordinate Tolerances for more information.