GLG 490-- Using a Geographic Information System
written by George Hilley in 1997.



Table of Contents:

Introduction
The Basics of a Geogrpahic Information System
Some Common GIS Software Packages
Basic Elements of ArcView That You Should Know


Introduction
    We learned how to use databases such as Filemaker Pro in this class.  These databases are relational databases and can be used to link records and abstract information in various ways.  However, these databases do not know where the data is located on the earth.  While we can include information such as the latitude and longitude of a measurement in our relational database, the database cannot calculate the length between observations, the area around observations, or perform any type of analysis where the location of the measurement is a concern.  As geologists, a key piece of information about data, the location of the data, is lost in a relational database such as Filemaker Pro.
    The Environmental Systems Research Incorporation (ESRI) circumvented this problem in the late 1960s when they invented the concept of a Geographic Information System (GIS).  A GIS records and understands the significance of the spatial location of data when performing analyses.  In this way, GIS is a hybrid between CAD and Filemaker Pro- in CAD, little if any information can be associated with a spatial position recorded by a point, line, or area; however, Filemaker Pro can record a large amount of information but has no information about its spatial position.  A GIS contains both pieces of information and so we can create points, lines, and areas and associate those spatial features with data.  This is a huge conceptual leap in database design that has many applications in geology.
    A good example of a set of data in which spatial information is important is a map.  Features on a map represent different pieces of data that may include things like faults, fold axes, contacts, assay measurements, and remotely sensed imagery.  The spatial location of each of these pieces of data is essential in understanding how these data relate to each other.  Using a CAD program to store the map data loses information about the data, furthermore using a relational database program to store the map data loses essential information about the data as well.  Therefore, the data only make sense when both the spatial position of the data points and what those points are are considered.  This is a perfect application of a Geographic Information System.

The Basics of a Geographic Information System
     A GIS is "An organized collection of computer hardware, software, geographic data, and personnel designed to efficiently capture, store, update, manipulate, analyze, and display all forms of geographically referenced information" (ESRI, 1993).  There are two different types of GIS that are currently available- a vector-based GIS and a raster-based GIS.  The vector-based GIS stores information as a series of nodes and vertices.  From these basic two units, vectors are defined by creating a line between two nodes.  The vertices act to better detail the shape of the line, whereas the nodes define both ends of the line and serve as a container for attributes related to the line.  From the vector units, areas may be defined by labeling closed areas created by the intersections of many vectors (ESRI, 1993)  In this way, all three types of spatial data (point, line, polygon) may be represented in the vector GIS (Figure 1.1).  The ARC module of ARC/INFO is the most popular and widely used vector GIS system.  The second variety of GIS are the raster GIS.  In this type of GIS, map data are rasterized and converted into a series of discrete square elements, called pixels.  These pixels are in turn organized as a grid of these squares, with a square’s location being represented by a row number and a column number.  Each of these pixels holds a value representing some sort of data at that point on the grid (Figure 1.2).  These data values contained within the pixels are often called feature types and may represent a point, an element of a line, or part of an area.  For example, if a road exists at a particular location, the grid elements through which the road passes will be assigned a numerical value corresponding to a "road" designation.  In this way, spatial information is represented in much the same way as graphical image data in computers are represented (Star et al., 1990).  This allows for the easy conversion of processed image data into a raster GIS format.  IDRISI and the GRID module of ARC/INFO are two popular raster GIS systems.  Due to a recent strive to incorporate both raster and vector systems into a single GIS, IDRISI has added fairly limited vector functionality to their software package.  Each of these types of GIS has its particular strengths and weaknesses.  The vector GIS is good at handling discontinuous data (i.e.. property lines, geologic units, etc.) while the raster GIS is good at handling continuous surfaces of data (i.e.. elevation data, etc.).




Some Common GIS Software Packages
    Several Geographic Information Systems are available.  ARC/INFO, ArcView, MapInfo, ATLAS*GIS (recently acquired by ESRI), and Intergraph are some common GIS software packages.  In addition, some remote sensing packages such as ERMapper and ERDAS incorporate a limited number of features into their analysis tools.  The software package that we will be using in this class is ArcView 3.0b.  This is a nice, user-friendly GIS that avoids many of the troubles and pains of past GIS software which are primarily command-line programs.

Basic Elements of ArcView That You Should Know
    There are several elements of ArcView with which you should be familiar.  We will elaborate more on these elements and on more sophisticated elements in class; however, these are the following are the four elements that you must understand in order to get started with ArcView.  The elements are a view, a theme, a table, and a layout.
    A view is a place where you can organize and display all of your geographic data.  Many different data sets can be part of a view and various combinations of these data sets can be displayed within the view.  A view is created by either 1) double clicking on the "View" button in the project window, or 2) selecting the "View" button in the project window and pressing the "New" button on the project window toolbar.
    A theme is simply a geographic data set.  Examples of themes would be geologic maps, annotation layers, hydrologic maps, contour maps, digital elevation models, and remotely sensed imagery.  To add a theme to your project, either 1) click the add theme button on the toolbar, or 2) select the "Add Theme" option from the "Theme" menu.  Both of these operations must be carried out in the "View" window.  To turn the theme on, simply click on the check box next to the theme's name in the "View" window and the theme will appear.
    A table is a tabular representation of your geographic data.  Each theme has an associated table which contains the attributes of each geographic element in your data set.  Although a theme must be associated with a table, a table doesn't necessarily have to be associated with a theme.  Therefore, both data associated with spatial locations and data unassociated with spatial locations can be stored in a GIS.  A table can be created by either 1) double clicking on the "Table" button in the project window, or 2) clicking on the "Table" button in the project window and then clicking the "New" button on the project toolbar.
    A layout is a map that you can create out of your geographic data.  Layouts can contain scale bars, keys, north arrows, views, and other objects you may wish to include on a map.  Therefore, the Layout area is where you can design the final mode of presentation of your geographic analysis.  A layout can be created by either 1) double clicking on the "Layout" button in the project window, or 2) clicking on the "Layout" button in the project window and then clicking the "New" button on the project toolbar.
 
 


Pages maintained by
Prof. Ramón Arrowsmith

Pages last modified on Tues Nov 11 1997.