Real World GIS: Points, Lines, Polygons

If you’re a reader of maps, then you’re already familiar with the common graphical devices map makers (cartographers) use to represent real world features on a two-dimensional surface. If you look at a common reference map, like a road map, you may see rest-stops displayed as a point of some kind; highways and roads displayed as lines colored and sized by their type; and perhaps a city metropolitan area displayed as an enclosed polygon delineating the extent of the urban area.

So how does this relate to GIS? A Geographical Information System stores and represents real world features using the same graphical methodology. Every discrete feature on the Earth’s surface can be represented as a point, line, or polygon. And by discrete, I refer to features that have a distinct boundary, like power-plants, substations, transmission lines and counties. Examples of continuous features - features that exist “everywhere”- include wind speed, elevation, and temperature. For now we’ll keep our focus on discrete data that can be represented with geometric shapes (vectors) and save our “continuous data” discussion for another day.

So how do we decide how to represent a feature within a GIS? The short answer is… it depends upon how close or how far away we are from the object, aka “map scale”. I’ll discuss map scale in more detail in my next post. In the meantime, here’s an example. If you’re looking at a map showing the entirety of North America, New York City would be best shown as a single point. After all, from that far away, even a polygon or line would look like a point, so it makes sense to use a point. On the flip side, if you were looking at a road- map of the state of New York, New York City would be better represented as polygon showing the boundary of the urban area. See the relationship? The important concept to remember here is that all features of the discrete type are displayed using points, lines, or polygons (nothing else), and that the symbol type used is scale dependant.

In my next post I’ll discuss “map scale” in more detail and how it relates to the GIS layers developed and offered by Platts Geospatial.