WÊÊÊÊÊÊÈËttÈÊÊÊÊÊÊtÊtÊ& 
conceptual models involving geographic relationships 
can be performed (i.e. land suitability/capability). 
This facilitates both scientific investigation and 
policy analysis criteria over large areas in short 
periods of time. 
7) Change analysis can be efficiently performed 
for two or more different time periods. 
8) Interactive graphic design and automated 
drafting tools can be applied to cartographic design 
and production. 
9) Certain forms of analysis can be performed cost 
effectively that simply could not be done 
efficiently if performed manually. 
10) There is a resultant tendency to integrate data 
collection, spatial analysis, and decision-making 
processes into a common information flow context. 
This has great advantage in terms of efficiency and 
accountability. 
Evaluating the capability of the techniques, as 
well as understanding the limitations of the 
systems, will permit the spatial scientist 
(assessor) to determine if a GIS can provide answers 
to spatial assessment questions. Knowledge of the 
basic data elements will help in the design of 
future GIS and avoid duplication or "reinvention of 
the wheel." The capabilities of a GIS for 
geographic problem solving are tremendous if systems 
are designed properly and output is optimally 
utilized. Uses of Geographical Information Systems 
are then only limited to the articulation of the 
problem to be studied and the availability of 
reliable data. 
Many GIS have six primary functions in common: 1) 
data entry, 2) encoding, 3) preprocessing, 4) data 
base management, 5) spatial/statistical analysis and 
modeling and 6) statistical/graphic output (Myers, 
1985, Brumfield, 1983, 1985) (fig. 2; see next 
page). 
5 DATA GATHERING 
Data gathering (i.e., selection) procedures of GIS 
development are the most important methodological 
consideration for GIS design outside of establishing 
the overall information needs of the user. This 
step is vital, because the type of data included and 
the scale of data resolution will determine the 
quality and usefulness of the information generated. 
Data accuracy must also be considered in terms of 
data gathering procedures, for accuracy is relative 
to the scale of data compilation (fig. 3). 
6 DATA ENTRY/ENCODING/PREPROCESSING 
The largest portion of time and energy invested in 
developing a GIS system is in obtaining, entering, 
converting, and storing data. Useful data exist in 
various forms such as tabular, graphical, digital 
and remotely sensed video/digital. A primary 
problem of a GIS system, therefore, becomes the 
integration of these various forms into a single 
computer compatible format (fig. 4). 
Spatial data are referred to as layers, fields or 
variables. Representation of these spatial entities 
must retain two basic characteristics of the 
original when the information is to be utilized in 
automated GIS processing: 1) the actual variable or 
characteristic, such as its name or value, and 2) 
its spatial location, or where it resides in 
geographical space (Dangermond, 1984). 
Spatial data layers exist as information in one of 
four possible states: points, lines, surfaces, and 
polygons. These layers have their geographic 
components encoded by one of two basic techniques: 
vector/polygon format or raster/grid cell format. 
Vector form is created by assigning x,y coordinates 
to various points along a line or polygon. 
Cartographic entries into digital format are 
translated point for point and line for line (Marble 
and Peuquet, 1983). Raster form is located by its 
position on a predetermined spatial lattice or grid 
cell system. 
These two forms of data transformation or geo 
coding each have their advantages and limitations. 
Vector format depicts spatial information more 
accurately since it is point referenced data in N 
dimensional math coordinate space (e.g. x,y) and is 
more compatible with statistical analysis (Marble 
and Peuquet, 1983). Vector format results in less 
data volume, however, it requires more computer 
processing time for analysis. Raster format results 
in a loss of geographic specificity, however, 
manipulation and processing efficiency is higher. 
Vector representation refers to points whose 
positions are defined by the number of axis in 
coordinate space. Raster format represents a 
defined area, including the point position in 
coordinate space (e.g., AVHRR pixels). Spatial data 
modeling procedures are less complicated with raster 
because it provides more accurate registration, yet, 
Figure 4 
564 
is posit 
both fon 
utilized 
types of 
from one 
Tomlins oi 
AISC, ve 
and somi 
multivar: 
soil moi 
raster f 
and dis] 
formats 
modeling 
7 DATA 
A GIS 
numerous 
multiple 
system 
update 
DATA 
MANAGI 
SUBSY 
7.1 Data 
Pertinent 
capabilit 
particula 
specific 
processit 
national 
not ali 
in terms 
or comput 
tion lit 
enables 
processit 
capabilit 
telecommt 
informati 
in comple 
7.2 Data 
Data anal 
statistic 
informati 
interpret 
overlayin 
functions 
models. 
applied 
grouping 
feature 
understan 
particula 
phenomena 
Analysi 
requires 
procedure