How a GIS represents and models geographic information
How a GIS represents and models geographic information

All geographic information is represented and managed using three primary GIS data structures:

  • Feature classes
  • Attribute tables
  • Raster datasets

These three fundamental data types can be extended with additional capabilities to manage data integrity, model geographic relationships (such as network connectivity and flow), and add important geographic behavior.

Each GIS has a collection of datasets

Typically, a GIS is used for handling several different datasets where each holds data about a particular feature collection (for example, roads) that is geographically referenced to the earth's surface.

A GIS database design is based on a series of data themes, each having a specified geographic representation. For example, individual geographic entities can be represented as features (such as points, lines, and polygons); as imagery using rasters; as surfaces using features, rasters, or TINs; and as descriptive attributes held in tables.

In a GIS, homogeneous collections of geographic objects are organized into data themes such as parcels, wells, buildings, orthoimagery, and raster-based digital elevation models (DEMs). Precisely and simply defined geographic datasets are critical for useful geographic information systems, and the design of layer-based data themes is a key GIS concept.

GIS datasets are logical collections of geographic features

A dataset is a collection of homogeneous features for each theme. Geographic representations are organized in a series of datasets or layers. Most datasets are collections of simple geographic elements such as a road network, a collection of parcel boundaries, soil types, an elevation surface, satellite imagery for a certain date, well locations, or surface water.

In a GIS, spatial data collections are typically organized as feature class datasets or raster-based datasets.

Many data themes are best represented by a single dataset such as for soil types or well locations. Other themes, such as a transportation framework or surface elevation, are often represented by multiple datasets. For example, transportation might be represented as multiple feature classes for streets, intersections, bridges, highway ramps, railroads, and so on. The table below illustrates how surface elevation might be represented using multiple datasets.

Raster datasets are used to represent georeferenced imagery as well as continuous surfaces such as elevation, slope, and aspect.

Common GIS representations


Geographic representation



Large water bodies




Urban areas


Road centerlines


Administrative boundaries


Well locations




Satellite imagery


Surface elevation

DEM rasters

Contour lines

Elevation points

Shaded relief rasters

Land parcels


Parcel tax records


Thematic layers become datasets. This is the key organizing principle in a GIS database.

Each GIS will contain multiple themes for a common geographic area. The collection of themes acts as a stack of layers. Each theme can be managed as an information set independent of other themes. Each has its own representation (as a collection of points, lines, polygons, surfaces, rasters, and so on).

Because layers are spatially referenced, they overlay one another and can be combined in a common map display. In addition, GIS analysis tools, such as polygon overlay, can fuse information between data layers to discover and work with the derived spatial relationships.

Thematic layer organization of GIS data

Any effective GIS database will adhere to these common principles and concepts. Each GIS requires a mechanism for describing geographic data in these terms along with a comprehensive set of tools to use, manage, and share this information.

How GIS users work with geographic information

Users work with geographic data in two fundamental ways:

  • As datasets, which are homogeneous collections of features, rasters, or attributes, such as parcels, wells, buildings, orthophoto imagery, and raster-based digital elevation models
  • As individual elements or subsets, such as the individual features, rasters, and attribute values contained within each dataset

Working with GIS datasets

In ArcGIS, homogeneous collections of geographic objects are organized into datasets about common subjects, such as parcels, wells, roads, buildings, orthophoto imagery, and raster-based DEMs.

Many of the operations that users perform in ArcGIS work on datasets as inputs or create new datasets as results. Datasets also represent the most common method for data sharing among GIS users.

Datasets provide the primary data sources for each of the following:

Maps, globes, and 3D scenes: These views provide the principal display of geographic information as a series of map layers. Each map layer references a specific GIS dataset and is used to symbolize and label the dataset. In this way, map layers help bring your GIS datasets to life in your GIS.

Map layers in a map of Oregon.Map layers—such as 3D buildings, imagery, and surface elevation

Map layers in 2D maps and 3D scenes are used to symbolize and label GIS datasets. This map has layers for cities, highways, state and county boundaries, water bodies, and streams. Each of these layers is used to portray a GIS dataset.

Geoprocessing inputs and derived datasets: GIS datasets are common data sources used for geoprocessing and are useful for automated data processing and GIS analysis. Datasets are used as inputs and new datasets are derived as results for various geoprocessing tools.

Geoprocessing helps you to automate many tasks as a series of operations so that they can be run as a single step. This helps to create a repeatable, well-documented data processing workflow.

Users also work with ArcGIS datasets to perform spatial analysis.

A geoprocessing model used to identify and rank potential sites for new parks

This model illustrates how to identify and rank potential sites for new parks. Good candidate locations must have high population counts and not be too close to existing parks.

Working with individual features and elements in datasets

In addition to working with datasets, users also work with the individual elements contained in datasets. These elements include individual features, rows and columns in attribute tables, and individual cells in raster datasets. For example, when you identify a parcel by pointing at it, you're working with the individual data elements in a dataset:

Clicking on an individual feature accesses its parcel information.

You work with individual data elements when you edit features—as in this example for editing road centerlines:

Individual vertices being added to represent the shape of a road centerline

When working with tables, users work with descriptive information contained in rows and columns, as illustrated here:

Selected rows are highlighted in blue.