Figure 2: A data flow-based visualization system (IRIS Explorer). 
3.0 GIS AND SCIENTIFIC VISUALIZATION 
Geographic information systems provide processing and analysis functions for spatially distrib 
uted data. The data (from maps, aerial photos, satellites, relational databases) is introduced and 
stored in a special-purpose database in which a common spatial coordinate system is the primary 
means of reference. GIS can output the results of the spatial analysis in the form of maps, plans, 
and reports. 
Scientific visualization is a method of comprehending large, often complex and multi-dimensional 
data sets generated by simulation or physical measurement by converting them into color im 
ages [DeFanti et al. 1990]. Some visualization systems offer the user the capability to explore and 
create new kinds of visualization. These systems are called application builders or data flow systems. 
In data flow systems, visualization is broken down into discrete modules, each of which performs 
a specialized task. The modules are assembled together into a data flow network, the output of 
which is an image plus widgets to control the processing parameters of the modules (Figure 2). 
Some of the weaknesses that are inherent in traditional GIS are the very strengths of scientific 
visualization systems [Kleinfeldt et al. 1996b]. Foremost among these is the ability of visualization 
systems to support n-dimensional data representation. Scientific visualization systems are extremely 
interactive, with control widgets that allow real-time adjustments to color, scaling, position, and 
other nuances that greatly affect a researcher's ability to perceive important data patterns. In