Metadata

Metadata is commonly called "data about data." More precisely, metadata is information that describes, or documents, a geographic dataset.

In the real world, examples of unofficial metadata can be found almost everywhere (the handwriting on the back of a photograph is one example). The type of metadata used to describe data used in a GIS, however, is official, or "standardized metadata." (Around the world, there are government organizations—like the Federal Geographic Data Committee in the U.S.—whose mission is to create rules for standardizing metadata.)

The writing on the back of a photograph is an unofficial metadata record that describes the contents and context of the photo.

A metadata record for a GIS dataset can be very detailed, but it typically includes information about why the data was collected (its purpose), what geographic area the data covers (its geographic extent), who collected the data, when the data was collected, what processes were performed on the data, and who should be contacted for more details about the data.

You create, edit, and view metadata in ArcCatalog. ArcCatalog automatically derives and documents some data properties, such as the geographic extent. Other properties, such as when and how the data was collected, must be documented by the data creator.

ArcCatalog provides different stylesheets for viewing metadata. The FGDC ESRI stylesheet consists of three tabs: Description, Spatial, and Attributes. Mouse over the bullets below to see an example of each tab.

The Description tab displays a thumbnail graphic of the data as well as keywords, an abstract, and a purpose statement (not shown here). Publication information is also included in the Description tab.

Members of the GIS community like to share data and methods. Metadata is critical for sharing data—people who are thinking about using a dataset for a project first view its metadata in order to determine whether the dataset is appropriate for the project. If a dataset lacks metadata, it may be used inappropriately—and any analysis results or measurements made with the data may be inaccurate.