1.5: Concepts

Maps as a Model of Reality

The real world is too complex and unmanageable for direct analysis and understanding because of its countless variability and diversity. It would be an impossible task to describe and locate each city, building, tree, blade of grass, and grain of sand. How do we reduce the complexity of the Earth and its inhabitants, so we can portray them in a GIS database and on a map? We do it by selecting the most relevant features (ignoring those we do not think are necessary for our specific research or project) and then generalizing the features we have selected. Chapter 6, as well as later portions of this chapter, covers the selection and generalization process in more detail. For now, let’s focus on features.

Features

As described in Definition #2 (and Figure 1.2), conceptually, there are two parts of a GIS: a spatial or map component and an attribute or database component. Features have these two components as well. They are represented spatially on the map and their attributes, describing the features, are found in a data file. These two parts are linked. In other words, each map feature is linked to a record in a data file that describes the feature. If you delete the feature’s attributes in the data file, the feature disappears on the map. Conversely, if you delete the feature from the map, its attributes will disappear too.

Features are individual objects and events that are located (present, past or future) in space. In Figure 1.2, a single parcel is an example of a feature. Within the GIS industry, features have many synonyms including objects, events, activities, forms, observations, entities, and facilities. Combined with other features of the same type (like all of the parcels in Figure 1.2), they are arranged in data files often called layers, coverages, or themes. In this text, we use the terms feature and layer.

In Figure 1.4 below, three features—parcels, buildings, and street centerlines—of a typical city block are visible. Every feature has a spatial location and a set of attributes. Its spatial location describes not only its location but its extent. While “location” may be simple to grasp, it is difficult to locate features accurately and precisely. Accuracy and precision are examined in Chapter 2, but, in brief, precision deals with the exactness of the measurement. For example, some input devices, like GPS, have a certain error. They may be precise within a certain accuracy range if used correctly. Accuracy is the degree of correspondence between the data and the real world.

Figure 1.4: Each feature in the layers above has a spatial location and attribute data, which describes the individual feature.

Besides location, each feature usually has a set of descriptive attributes, which characterize the individual feature. Each attribute takes the form of numbers or text (characters), and these values can be qualitative (i.e. low, medium, or high income) or quantitative (actual measurements). Sometimes, features may also have a temporal dimension; a period in which the feature’s spatial or attribute data may change.

As an example of a feature, think of a streetlight. Now imagine a map with the locations of all the streetlights in your neighborhood. In Figure 1.5, streetlights most are depicted as small circles. Now think of all of the different characteristics that you could collect relating to each streetlight. It could be a long list. Streetlight attributes could include height, material, basement material, presence of a light globe, globe material, color of pole, style, wattage and lumens of bulb, bulb type, bulb color, date of installation, maintenance report, and many others.

Figure 1.5: Location of street lights, represented with a red circle, and their attributes.

The necessary streetlight attributes depends on how you intend to use them. For example, if you are solely interested in knowing the location of streetlights for personal safety reasons, you need to know location, pole heights, and bulb strength. On the other hand, if you are interested in historic preservation, you are concerned with the streetlight’s location, style, and color.

Now continue thinking about feature attributes, by imagining the trees planted around your campus or office. What attributes would a gardener want versus a botanist? There would be differences because they have different needs. You determine your study’s features and the attributes that define the features.

Points, Lines and Polygons

Now think of the feature’s shape on a map. Single or multiple paired coordinates (x, y) locate individual features in space and define their unique shape. The x and y values of each coordinate pair are associated with real world coordinate systems, which are discussed in Chapter 3. For now, let’s focus on the shape of features, which take the generalized form of points, lines, and polygons (see Figure 1.6).

Figure 1.6: Each feature has a spatial position.

Points

Points are zero dimensional features (meaning that they possess only one x, y coordinate set) whose location is depicted by a small symbol. What you represent as a point depends on your study. Examples include streetlights, individual trees, wells, car accidents, crimes, telephone polls, earthquake epicenters, and even, depending on scale, buildings and cities.

Lines

Lines are formed from a sequence of at least two paired coordinates. The first pair starts the line and the last ends it. Two coordinate pairs form a straight line. Additional paired coordinates can form vertices between the starting and ending points that allow the line to bend and curve. Having length (which can be measured) but no width, a line feature is one-dimensional. Again, what is represented as a line depends on your study, but street centerlines, utility lines, canals, railroad tracks, rivers, flight paths, and elevation contour lines usually form lines.

Polygons

Polygons are features that have boundaries. Formed by a sequence of paired coordinates, polygons differ from lines in that the starting point is also its ending point. This provides polygons with both length and width, so these two-dimensional features can calculate the area contained within the feature. What is represented as a polygon differs from study to study, but examples include lakes, forest stands, buildings, counties, countries, states, and census districts.

Topology

One of the most important concepts associated with GIS and other geotechnologies is topology. As features are added to a GIS, they form spatial relationships—called topology—with each other (both with features within the same layer and with features in different layers). You might find topology a confusing term partly because it has both spatial and mathematical properties. For our purposes, you can define it as the spatial relationships among features. It deals with where features are in relation to one another and how they are related to one another. These relationships take the form of simple distance calculations from one feature to another, but also include the more complicated issues of adjacency and connectivity.

Figure 1.7: Fire hydrants are located along streets (so fire trucks can connect) and adjacent to structures that can burn.

1. Distances between features. The geographer Waldo Tobler created what some call the “first law of geography”, which states, “Everything is related to everything else, but near things are more related than distant things.” (1970, 236). This type of topology looks at the spatial relationships of where features are located. Consider the spatial locations of streets, bike lanes, sidewalks, and streetlights. They are positioned to work together. This is a type of topology; a relationship exists. Notice the relationship between the fire hydrant, building, and street in Figure 1.7.
2. Adjacency. Adjacency focuses on a single type of feature (like streets or buildings) and whether parts of two or more individual features are shared (or contained). Think of an individual street segment, and how it is most likely physically connected to at least one additional street segment at one or both of its ends. These adjacent street segments are in turn connected to additional segments, which in turn are connected to streets, forming a network. When a single point or line (like a boundary between two parcels) is shared by at least two features, the spatial data file stores only a single point or a single line to prevent duplication that could lead to errors. This topological relationship describes how features are related.
3. Connectivity. Also focusing on how features are related, connectivity specifies the way features are linked in a network. Even though a couple street segments may be physically connected in space, that does not mean that traffic can go in both directions. These are topological relationships that you can specify. Differing from adjacency, connectivity can include multiple feature types. For instance, you can determine the flow of water through connected pipe and valve features.