What is a GIS Database?
Think about a map for a second, a state road map for example. If you take away all the borders, legends, north arrows, scales, and titles, what you’re left with is just the actual map which represents features on or near the earth’s surface (in space) and probably some labels that depict what’s what. Now take away all the labels. The features on the map are most likely represented with a variety of colors and symbols that can still give you a lot of information. You would know that a blue blob was probably a lake, a black line probably a road, and a large dot a city or town. Creating the map with symbols that give the reader some more information about a particular feature was the job of the cartographer. We’ll discuss map symbols and cartography later in the semester.
Now, take away all the symbology in the map. Make everything the same color. Make all the lines the same thickness. Make all the dots the same size. What you are left with is the fundamental building blocks of data for GIS -- points, lines, and polygons. Any feature or phenomenon on the earth’s surface can be reduced to one of these three shapes, or feature types.
Suppose we took our map, covered it with a piece of clear plastic, and traced all the streams onto the plastic. Suppose we did the same for the roads, the towns, forested areas, and wetlands, placing each type of information on a separate piece of plastic. Each plastic sheet would now contain only one shape, or feature type (all points, all lines, or all polygons), which is essentially a layer of the original map. This is the basis for GIS datasets.
GIS datasets go by many names. Spatial data, spatial database, geospatial data, geodataset, map layer, coverage, and theme are all common terms used to refer to data for GIS. The software we will be using in this class, ArcView, uses the term theme, so we will use that term throughout this manual.
Let’s go back to our example map. We’ve got all our themes on separate pieces of plastic, big deal. Well, suppose you were interested in determining the effect of roadways on the fragmentation of wetland areas. No problem. Pull out the sheet with wetlands and the one with roads and lay them on top of each other. Now you’ll be able to look at roads and wetlands at the same time without any other features cluttering things up. This process of overlaying two or more themes to examine relationships between them is fundamental to working with GIS.
Our example, however, is still missing something. We have each theme separated out and we can overlay them in any combination or order that we need. But, we can’t tell the difference between Pomona Road and the Garden State Parkway or between Nacote Creek and the Mullica River. We’ve lost all the information that was present in the symbols and labels of the original map. This is where the database part of GIS comes in to play; this is also where our example falls apart and we need to discuss strictly computer based GIS.
Every theme has a database attached to it that contains information about the features in that theme. This database, called the attributes of the theme, is organized in a way so that each individual feature in the theme is linked to one specific line, or record, in the database. For example, a theme representing the counties of New Jersey would have 21 polygons, one for each county. The attributes of this theme would have 21 records; one record for each polygon (county). Each type of information in the attribute table gets its own column in the database, or field. In our county example, our database would contain the fields: county name, population, area, perimeter, and any other information included in this dataset. The database can have as many fields as are needed, and we’ll learn how to create new fields and add data later in the semester. This is where the real power of GIS comes from.
Using GIS, it is conceivable to work with 10 themes, each with 50 or so fields in their attributes and hundreds or thousands of records. Think about how cluttered our paper map would be. We can ask questions of the database to pull out, or reselect, only those features that we’re interested in, based on some criteria. For example, we can show all the counties in NJ with a 1990 population of less than 150,000. With the idea of overlaying themes, we can ask questions based on the spatial distribution of data, such as, "Show me all the residential areas that fall partially or wholly within the flood prone zone". Or, "Show me all the sand & gravel mines that are greater than 2 acres, that are not within the Pinelands National Reserve, that are within 300 feet of freshwater wetlands, that are in municipalities with population densities greater than 200 people/sq. mi., and tell me the names and addresses of those mines". The complexity of the questions you can ask is a function of the size of the database (and the processing ability of the computer). Of course this makes sense; the more information that is available, the more questions you can answer. Keep in mind that in this class we will learn how to add information to an existing database. You can always add information as long as the spatial data -- the map part -- already exists. If you’re interested in learning how to create spatial datasets from scratch, enroll in Introduction to Geographic Information Systems.