Scale in GIS

Map Scale

In map design and production with GIS, it is very useful to think of scale as the ratio of distance on the map to distance on the ground. This is the representative fraction, and with a map of 1:2,500, 1 inch on the map equals 2,500 feet on the ground (metric units may be used as well). A map at 1:2,500 is known as a large-scale map because the fraction (½, 500) is large compared with other possible map scales. Large-scale maps are particularly good at showing local-level details over a relatively small area.

As seen in Figure 1, land parcel boundaries for a neighborhood are mapped at a scale of 1:2,500. It is easy to pick out the detail and characteristics of the parcel objects at this scale. If we change the scale to 1:35,000, which is generally considered to be a medium scale, many more of the parcel boundaries come into view for this particular area. But it becomes very difficult to visualize subtle differences in the parcel objects’ shapes. Some parcel boundaries are visible at this scale, although there is a great deal of obfuscation in denser areas. Changing the scale to 1:1,000,000, we can no longer see any detail within the study area, and the general shape of its exterior boundary is no longer crisp. As such, maps with scales in the neighborhood of 1:1,000,000 are known as small-scale maps and are good at showing less detailed features over large areas. To put this scale into better context, the U.S. counties surrounding this neighborhood are also shown in the figure. The many large counties are well represented at this scale, but the neighborhood-level detail is lost. Therefore, a key issue in dealing with scales is to select one that is appropriate, given the mapping task at hand.

Spatial Scale

The spatial scale of a given data set is the smallest discernable unit of geography contained therein.

It is also known as spatial resolution. Imagery data, for example, are widely evaluated based on how small a unit area one of its individual pixels or cells represents. Imagery captured from aerial photography routinely has a high resolution such that pixels represent areas of less than 1 m2 (square meter). Imagery captured from a satellite station may also approach the resolution of imagery taken via aerial photography, but many satellite platforms offer much coarser resolutions, some greater than 1,000 m. The temporal scale of a data set, or the frequency of data collection for a given area, is also of relevance. Depending on the area, aerial photography, for example, can be quite variable in terms of how often new data are collected. On the other hand, many satellite-based forms of spatial data collection follow a regular temporal cycle. Spatial data collected at such regular cycles are useful in many applications, particularly those where careful monitoring of the changes in the landscape is required. In this way, users must not only consider the level of detail associated with a data set, they must also consider its frequency of collection if there is a temporal dimension to their study.

...