3 Fundamentals of GIS
A GIS is best described as a system of hardware, software, data, people, organizations, and institutional arrangements for collecting, storing, analyzing, and disseminating information about areas of the earth [6]. According to Congalton and Green
[7], GIS analysis can be divided into four main categories: overlaying, buffering,
modeling, and network analysis. These four components of analysis represent the
basic tools of GIS. Each one of these tools is simple, yet they can be combined to
produce complex, spatial analysis.
The process of overlaying usually involves combining two or more different
layers of a particular area to create a new layer that combines the attributes of
individual layers or compares the layers to calculate summary statistics. Various
types of overlay operations exist. Layers can be added, subtracted, divided, or
multiplied according to the values of certain attributes. The two datasets shown in
Fig. 5.2 depict the conditions of a floodplain in Kentucky during flood and post-flood
periods. The (D:\BOOK\7append\glossary.html) - polygon polygon datasets shown
in Fig. 5.2c and d were derived from Landsat imageries and indicate the wet and land
areas in the floodplain at the different discharge conditions. If we overlay the two
layers and compare the floodplain attributes (by subtracting the values of Fig. 5.2c
from Fig. 5.2d), we can identify areas where the floodplain has changed in the
period. After the subtraction operation, the new data layer (Fig. 5.2e) which was
created shows change as a positive or negative and no change as zero value. As
demonstrated in the figure, the overlay results quickly show where in the floodplain
changes have occurred due to flow changes. Overlaying is frequently combined with
other analysis methods to produce even more valuable results. For instance, one
could combine this operation with a model that automatically calculates the water
surface elevation from existing data.
Buffering is an important pre-analysis technique of identifying objects within a
specified distance of a reference object. It combines spatial data query techniques
and cartographic modeling. The reference object may be a point location, a line, or a
polygon. A simple environmental example would be to create a buffer of groundwater pollution zone around a waste site. This buffer could be used to assess health
risks to the affected population. The buffering operation can also be used to predict
coastline erosion by estimating the distance of the coastline at any time from the
buffer.
One of the most useful transformations in GIS has been the incorporation of
modeling. This involves linking the GIS database to a computer model of some
process. The GIS prepares input data for the model by combining all the relevant
data for every object. This allows spatial data to be processed in large quantities
using powerful, complex algorithms. There are two types of modeling: simulation
and predictive. Generally simulation modeling requires a high degree of technical
expertise and involves using GIS to simulate a complex phenomenon in nature.
Predictive modeling is a more powerful modeling tool than the simulation modeling.
202
S. O. Darkwah et al.
A GIS is best described as a system of hardware, software, data, people, organizations, and institutional arrangements for collecting, storing, analyzing, and disseminating information about areas of the earth [6]. According to Congalton and Green
[7], GIS analysis can be divided into four main categories: overlaying, buffering,
modeling, and network analysis. These four components of analysis represent the
basic tools of GIS. Each one of these tools is simple, yet they can be combined to
produce complex, spatial analysis.
The process of overlaying usually involves combining two or more different
layers of a particular area to create a new layer that combines the attributes of
individual layers or compares the layers to calculate summary statistics. Various
types of overlay operations exist. Layers can be added, subtracted, divided, or
multiplied according to the values of certain attributes. The two datasets shown in
Fig. 5.2 depict the conditions of a floodplain in Kentucky during flood and post-flood
periods. The (D:\BOOK\7append\glossary.html) - polygon polygon datasets shown
in Fig. 5.2c and d were derived from Landsat imageries and indicate the wet and land
areas in the floodplain at the different discharge conditions. If we overlay the two
layers and compare the floodplain attributes (by subtracting the values of Fig. 5.2c
from Fig. 5.2d), we can identify areas where the floodplain has changed in the
period. After the subtraction operation, the new data layer (Fig. 5.2e) which was
created shows change as a positive or negative and no change as zero value. As
demonstrated in the figure, the overlay results quickly show where in the floodplain
changes have occurred due to flow changes. Overlaying is frequently combined with
other analysis methods to produce even more valuable results. For instance, one
could combine this operation with a model that automatically calculates the water
surface elevation from existing data.
Buffering is an important pre-analysis technique of identifying objects within a
specified distance of a reference object. It combines spatial data query techniques
and cartographic modeling. The reference object may be a point location, a line, or a
polygon. A simple environmental example would be to create a buffer of groundwater pollution zone around a waste site. This buffer could be used to assess health
risks to the affected population. The buffering operation can also be used to predict
coastline erosion by estimating the distance of the coastline at any time from the
buffer.
One of the most useful transformations in GIS has been the incorporation of
modeling. This involves linking the GIS database to a computer model of some
process. The GIS prepares input data for the model by combining all the relevant
data for every object. This allows spatial data to be processed in large quantities
using powerful, complex algorithms. There are two types of modeling: simulation
and predictive. Generally simulation modeling requires a high degree of technical
expertise and involves using GIS to simulate a complex phenomenon in nature.
Predictive modeling is a more powerful modeling tool than the simulation modeling.
202
S. O. Darkwah et al.