VISUALIZING GEOGRAPHY: USING A GIS TO TURN SPATIAL CONCEPTS INTO REALITY 
Lawrie E. Jordan, III, and Stephen L. Sperry 
ERDAS, Inc. 
2801 Buford Highway NE, Suite 300 
Atlanta, Georgia 30329-2137 USA 
404/248-9000 
Fax: 404/248-9400 
COMMISSION IV 
ABSTRACT 
In the past, the development of GIS software functions has concentrated on replicating manual mapping and analysis 
procedures. More complex analysis functions have been limited largely by the incompatibility of data formats. Recent GIS 
software developments have allowed users to access both raster and vector data at the same time. This began a trend 
toward true data merging. Now users can visualize geography rather than computer files, and much more easily realize 
solutions to complex spatial problems. This paper will discuss recent trends in geoprocessing and describe software 
functions, user interfaces and applications which point GIS in the direction of a true geography visualization and analysis 
tool. 
INTRODUCTION 
The concept of visualization has its roots in the GIS of the 
1970's when manual techniques were used to ingest various 
data sources into a digital environment. Analysts could use 
mylar drawn grids to interpret soils information or to capture 
data from a USGS quadrangle map, aerial photograph, or 
other source and produce a black and white, or gray-scale, 
computer generated map. Rather sophisticated analytical 
techniques were accomplished in the absence of the 
technology to augment those analyses. 
In the late 1970's and early 1980's advances in hardware and 
software eliminated the need for hand drawn overlays. Digital 
image processing systems could display and manipulate data. 
These systems allowed users to classify digital satellite data 
and aerial photographs into distinct vegetation and landcover 
classes. These new data layers could be digitally overlaid to 
show new areas that met several criteria. All visualization was 
still in two dimensions. 
It has been only recently that true three-dimensional analyses 
have been possible. Through improved hardware and 
sophisticated software, users can take a drive through the 
landscape without ever leaving the office. Digital data can be 
modeled to answer any number of "what if..." questions. 
Modeling was done in the early days, but steps were recorded 
in a notebook. Now steps are selected through user-friendly 
graphical user interfaces and the solutions displayed almost 
instantaneously on the computer screen. As parameters 
change, the model can be updated and re-run immediately. An 
analysis that took weeks to perform in the 1970's now takes 
just a few minutes; and a three-dimensional view of that 
analysis can be generated quickly and easily. 
The concept of motion and animation in the GIS environment 
is also relatively new. However, the applications for them are 
expanding. As GIS is applied toward global research, models 
are created for weather patterns, carbon dioxide emissions 
and other global trends. The ability to see the changes in the 
earth in an animated sequence of images provides scientists 
and decision makers with undisputed evidence that the future 
of the planet is in danger. The real power of GIS is in 
prediction models. Based on past and current data, models 
can be developed to show how the environment will change if 
current conditions persist, or if these conditions worsen, and 
how this future might be altered if preventative measures are 
899 
enacted. It is one thing to speculate on the effects of global 
warming, population increases, deforestation and other global 
problems, it is quite another to see the effects of these 
situations graphically on an interactive display. A dramatic 
image can speak much louder than columns of statistics. 
CHANGING THE DEFINITION OF GIS 
Before GIS can be used for true geographic visualization, the 
concept of the traditional GIS must be broadened to include 
all data types-raster, vector, attribute, spreadsheet, ancillary, 
etc. To derive solutions to the complex problems users bring 
to GIS requires all available information. Most users now 
recognize the value of integrating raster and vector data into a 
GIS. The timeliness and accuracy of satellite imagery and 
aerial photographs makes data base updates easier and more 
cost-effective than creating new vector data from ground truth 
information. The image processing functions available with 
raster data provide a new level of analysis that adds depth to a 
GIS. Images can be classified to show wildlife habitats, soils 
information, vegetative health and numerous other categories 
that add to the completeness of a GIS. 
AVHRR Data 
The increased use of Advanced High Resolution Radiometer 
(AVHRR) data from the National Oceanic and Atmospheric 
Administration (NOAA) satellites is making global studies 
easier. The extensive coverage of these data give a picture of 
an entire continent at one time. Data sets from several years 
can be used to study change in overall vegetation and land use 
over time. These multitemporal techniques have been used 
extensively in South America to track deforestation. 
AVHRR data are also being widely used by the United 
Nations Environment Programme's (UNEP) Global 
Resource Information Database (GRID). GRID was 
developed to bridge the gap between the scientific 
understanding of earth processes and sound management of 
the environment at national, regional and global levels. GRID 
provides GIS data layers to users all over the world. These 
layers cover both regional and global information on political 
boundaries, elevation, soils, vegetation, watersheds, rainfall, 
surface temperature, ozone distribution, population density 
and much more. These data sets are compiled by inputting 
AVHRR and other satellite data into a GIS. Although GRID