ArcGIS for Architects GE2304

CHAPTER TWO
GEOGRAPHIC DATA MODEL
2.1. Data Model
Data in a GIS represent a simplified view of the real world. Physical entities or
phenomena are approximated by data in a GIS. These data include information on the
spatial location and extent of the physical entities, and information on their non-
spatial properties. GIS does not store a map in any conventional sense. Instead GIS
stores the data from which we can draw a desired view to suit a particular purpose.
Spatial Models are important in that way in which information is represented, affects
the type of analysis performed and types of graphic display that can be obtained.
In order to represent the spatial information and their attributes, the Data Model
represents the linkages between the real world domain of geographical data and the
computer and GIS representation of these features. As a result, Data Model, not only
helps in organizing the real world geographical features into a systematic
storage/retrieval mechanism, but also helps in capturing user’s perception of these
features. The two data models common in GIS are Vector data model and Raster data
Model.
2.1.1 Vector data model
A vector data model uses sets of coordinates and associated attribute data to define
discrete objects. Groups of coordinates define the location and boundaries of discrete
objects, and these coordinate data plus associated attributes are used to create vector
objects representing the real-world entities.
There are three basic types of vector objects (entities) along with their attributes:
points, lines, and polygons. A point uses a single coordinate pair to represent the
location of an entity that is considered to have no dimension. Gas wells, light poles,
accident location, and survey points are examples of entities often represented as
point objects in a spatial database. Some of these have real physical dimension, but
for the purposes of the GIS users they may be represented as points. In effect, this
means the size or dimension of the entity is not important spatial information, only
the central location. Attribute data are attached to each point and these attribute data

ASTU Architecture Dept. 1

ArcGIS for Architects GE2304
record the important non-spatial characteristics of the point entities. When using a
point to represent a light pole, important attribute information might be the height of
the pole, the type of light and power source, and the last date the pole was serviced.
Linear features, often referred to as arcs, are represented as lines when using vector
data models. Lines are most often represented as an ordered set of coordinate pairs.
Each line is made up of line segments that run between adjacent coordinates in the
ordered set. A long, straight line may be represented by two coordinate pairs, one at
the start and one at the end of the line. Curved linear entities are most often
represented as a collection of short, straight, line segments, although curved lines are
at times represented by a mathematical equation describing a geometric shape. Lines
typically have a starting point, an ending point, and intermediate points to represent
the shape of the linear entity. Starting points and ending points for a line are
sometimes referred to as nodes, while intermediate points in a line are referred to as
vertices. Attributes may be attached to the whole line, line segments, or to nodes and
vertices along the lines.
Area entities are most often represented by closed polygons. These polygons are
formed by a set of connected lines, either one line with an ending point that connects
back to the starting point, or as a set of lines connected start-to-end. Polygons have an
interior region and may entirely enclose other polygons in this region. Polygons may
be adjacent to other polygons and thus share “bordering” or “edge” lines with other
polygons. Attribute data may be attached to the polygons, e.g., area, perimeter, land
cover type, or county name may be linked to each polygon.

ASTU Architecture Dept. 2

ArcGIS for Architects GE2304

Figure 5.1 Points, nodes and vertices define points, line, and polygon features
in a vector data model.

Figure 5.2 Vector representation

2.1.2 Raster data model
Raster data models define the world as a regular set of cells in a grid pattern.
Typically these cells are square and evenly spaced in the x and y directions. The
phenomena or entities of interest are represented by attribute values associated with
each cell location. The information content in one cell depends upon the size of the

ASTU Architecture Dept. 3

ArcGIS for Architects GE2304
cell. If the cells are sufficiently small, the information present in one cell will be
more. This is called resolution of the image.
Raster data models are the natural means to represent “continuous” spatial features or
phenomena. Elevation, precipitation, slope, and pollutant concentration are examples
of continuous spatial variables. These variables characteristically show significant
changes in value over broad areas. The gradients can be quite steep (e.g., at cliffs),
gentle (long, sloping ridges), or quite variable (rolling hills). Because raster data may
be a dense sampling of points in two dimensions, they easily represent all variations
in the changing surface. Raster data models depict these gradients by changes in the
values associated with each cell.
In raster structure a single cell represents a point, lines by sequences of neighboring
cells and area by collections of contiguous cells.
In Raster data structure represent points by single cells, lines by sequences of
neighboring cells, and areas by collections of contiguous cells.

Figure 5.3 Raster representation

ASTU Architecture Dept. 4

ArcGIS for Architects GE2304
2.1.3 A Comparison of Raster and Vector Data Models
The question often arises, “which are better, raster or vector data models?” The
answer is neither and both. Neither of the two classes of data models is better in all
conditions or for all data. Both have advantages and disadvantages relative to each
other and to additional, more complex data models. In some instances it is preferable
to maintain data in a raster model and in others in a vector model. Most data may be
represented in both, and may be converted among data models. As an example,
elevation may be represented as a set of contour lines in a vector data model or as a
set of elevations in a raster grid. The choice often depends on a number of factors,
including the predominant type of data (discrete or continuous), the expected types of
analyses, available storage, the main sources of input data, and the expertise of the
human operators.
2.1.4 Advantage/disadvantages of Raster Data Model over Vector Data Model

to vector data.

merging etc. are complex in raster data model as compared to vector.

wever, multi-theme overlay operations are more easy in raster data model.
-based model.
Vector Data Model
Advantages:
Good and real representation of geographic data
Compact data structure
Topology can be completely described
Accurate for graphic output
has less storage space
Disadvantages:
Data structure is complex
Combination of several vector polygons create difficulties in handling
Simulation is difficult because each unit has a different topological form

ASTU Architecture Dept. 5

ArcGIS for Architects GE2304
Display and plotting are expensive
Raster Data Model
Advantages:
Simple data structure
Overlay of mapped data with remote sensing data is easy
Simulation is easy because each spatial unit has same size and shape
Good for multiplayer overlay
Disadvantages:
Data is voluminous and require large storage space
Use of large cell to reduce data volume loses significant information
Crude raster maps have ugly look
Network linkages are difficult to establish.
2.2. GIS Spatial Analysis
Spatial analysis is a set of techniques for deriving new information and knowledge
from spatial data. It is distinguished from other types of analysis because the results
of spatial analysis are dependent on the locations of the objects being analyzed.
Spatial analysis is useful for gaining an understanding of where geographic features
and phenomena are located and the spatial relationship or spatial interactions between
the features and phenomena. The recent development of geographic information
technologies has facilitated the widespread application of spatial analysis, making it
an increasingly important analytical tool for scientific research and decision‐making.
The Common Spatial Data Analysis:
1. Query
There are two types of query for GIS
a) Attribute ( Aspatial ) query
 It retrieves a data subset from a map by working with its attribute
data by asking logical questions
E.g. Select the Towns with populations over 500,000 in the Ethiopia.
b) Spatial Query
 Involves selecting features based on location.

ASTU Architecture Dept. 6

ArcGIS for Architects GE2304
E.g. select high schools within 2 kms of downtown Adama is an
example of a spatial query because it uses the spatial concept of
distance.
2. Measurements (lengths, perimeters, areas)
 Only an approximation of the true distance on the ground
Measurements of Vector Data
 Vector data are stored as line segments even curved lines are stored as short
line segments
 Measurements can be stored as attributes in vector GIS
 Distance measurement (Euclidean distance) – obtained by using Pythagorean
Theorem
 Perimeter - sum of the straight line lengths of the boundary
 Area - sum of areas of simple geometric shapes formed by subdividing the
feature of interest
Measurements of Raster Data
 Raster data are approximated using a grid cell representation
Distance Measurement of Raster Data
 Euclidean distance – shortest distance/path
B

 Manhattan distance – distance along raster cell sides from one point to
another

ASTU Architecture Dept. 7

ArcGIS for Architects GE2304
 Proximity – concentric equidistant zones are established around the start
point A, and the distance to B is taken to be the corresponding radius of the
circle that intersects it

Perimeter Measurement of Raster Data

 Number of cell sides that make up the boundary of the feature is multiplied
by the grid resolution
Perimeter = ((number of cell sides) x (resolution)
Area Measurement of Raster Data
 Number of cells of a feature is multiplied by the known area of an
individual grid cell
Area = (number of cells) x (area of individual cell)
3. Reclassification
 An important variation of querying in GIS
 Can be used in place of querying for raster GIS
 Uses reclassification rules to reclassify or assign new values to cells
 Results to a new image
 May be Boolean or Weighted
Boolean Reclassification
 Produces a two-coded image from a complex image
 Original image is reclassified to an image with only 0 or 1 as cell values
Example, if the production forests are of interest, all cells representing forestry can
be assigned a value of 1, and all other cells a value of 0.

ASTU Architecture Dept. 8

ArcGIS for Architects GE2304
Reclassification by Weighting
 A different weight is assigned to different feature types or classes, based on the
purpose of the reclassification
 Higher weights can be assigned to priority classes while lower weights to those
of lower priority
Example, if the purpose is for forest conservation, a weight of 5 can be assigned to
forestry and lower weights to other features
Reclassification Example
Land use Original value New value New value
Boolean Weighted
Forestry 10 1 4
Water 11 0 2
Settlement 12 0 1
Agricultural 13 0 3
Land

4. Proximity Analysis (Buffering)

 A buffer zone is often treated as a protection zone and is used for planning and
regulation.
 Creation of buffer zone around selected features within specified distance
 Selected map features may be points, lines or polygons.
Point entity: circular buffer zone

ASTU Architecture Dept. 9

ArcGIS for Architects GE2304

Line entity: elongated buffer zone

Polygon entity: buffer zone has the same shape as original polygon, but larger

Example:
 How many houses lie within 100 m of this water main?
 What is the total number of patients within 10 km of the healthcare
facility?
5. Spatial overlay
This technique involves combining multiple spatial datasets to identify spatial
relationships, intersections, and overlaps. It can be used to analyze the interaction

ASTU Architecture Dept. 10

ArcGIS for Architects GE2304
between different geographic features, such as land use, infrastructure, and
environmental factors.
 Integrating data from two or more sources to produce a new map layer

6. Spatial interpolation
Spatial interpolation is the process of estimating values at unsampled locations
based on known values at sampled locations. Interpolated data is only an
approximation of the true value. This can be used to:
 Create continuous surfaces from discrete point data, such as weather
observations or soil characteristics.
 fill the gaps between observed data points
 construct height contours
7. Surface analysis
 Slope and aspect calculation
 Visibility analysis: Identification of areas of terrain that can be seen from a
particular point on the terrain surface
ASTU Architecture Dept. 11
ArcGIS for Architects GE2304

8. Network analysis
 Involves linear features or network of linear features that are topologically
structured (roads, rivers, pipelines, cables, etc.)
Examples:
 shortest path
 time of travel
 route tracing
9. Suitability analysis
suitability analysis is used to identify areas within planning area that are best
suited to particular land-use such as settlement, agriculture ,national park, and
other uses by considering different factors.
Example: Architects can identify the most suitable location for a new building
or development by considering different factors such as topography, slope, soil
conditions, environmental constraints, accessibility, views, and other site-
specific characteristics.

ASTU Architecture Dept. 12

ArcGIS for Architects GE2304
Spatial analysis can be done in two ways:
1. Vector Based Spatial Data Analysis
Buffering
 A buffer zone is often treated as a protection zone and is used for planning
or regulation.
 Buffering uses distance measurements from map features to create the
buffer zones.
 The GIS user must therefore know the measurement unit of map features,
for example, meters or feet prior to buffering.
Selected map features for buffering may be:
 points,
 lines, or
 Areas
Buffering around points creates circular buffer zones.

Buffering around lines creates a series of elongated buffer zones.

Buffering around polygons creates buffer zones extending outward or from the
polygon boundaries.

ASTU Architecture Dept. 13

ArcGIS for Architects GE2304

The buffer distance

 does not have to be constant it can vary according to the values of a given field
Example1:
Streams may be buffered with 200 meters along major streams and 100 meters
along tributaries

ASTU Architecture Dept. 14

ArcGIS for Architects GE2304
Example2: A factory may be buffered with distances of 5, 10, 15, and 20 meter, thus
forming multiple rings around the factory

Map Overlay
Map overlay combines the geometry and attributes of two feature maps to create the
output.

Feature Type and Map Overlay

Map overlay can be grouped by feature type into:
 In a point-in-polygon operation
The same point features in the input are included in the output but each point is
assigned with attributes of the polygon within which it falls.

ASTU Architecture Dept. 15

ArcGIS for Architects GE2304

 In a line-in-polygon operation
The output contains the same line features as in the input but each of them is
separated by the polygon boundaries on the overlay map.

 In polygon on-polygon overlay operation.

 Involve two polygon maps
 the output combines the polygon boundaries from the input and overlay
maps to create a new set of polygons

ASTU Architecture Dept. 16

ArcGIS for Architects GE2304

Map Overlay Methods

There are three common map overlay methods
1. UNION
Preserves all map features from the input and overlay maps by combining the area
extents from both maps

2. INTERSECT
Preserves only those features that fall within the area extent common to both the
input and overlay maps

ASTU Architecture Dept. 17

ArcGIS for Architects GE2304

3. IDENTITY
Preserves only map features that fall within the area extent defined by the input
map. Features of the overlay map that fall outside the area extent of the input map
are left out of the output .The input map for IDENTITY may contain points,
lines, or polygons.

Map Manipulation
The Geo Processing extension tools
Dissolve
 removes boundaries between polygons that have the same value of a selected
attribute

ASTU Architecture Dept. 18

ArcGIS for Architects GE2304

 Dissolve removes these unnecessary boundaries and creates a new, simpler map
with the classification results as its attribute values.
Example: Addis Ababa wereda to Zone

Clip
 Creates a new map that includes only those features of the input map that falls
within the area extent of the clip map.

ASTU Architecture Dept. 19

ArcGIS for Architects GE2304

Merge
 Creates a new map by piecing together two or more maps.

2. Raster Based Spatial Data Analysis

In raster analysis, geographic units are regularly spaced, and the location of each unit
is referenced by row and column positions. Grids are the common format for
numerous sources of spatial information including satellite imagery, scanned aerial
photos, and digital elevation models. These data sources have been adopted in many
GIS projects and have become the most common sources of major geographic
databases.
Measurement - Distance, Area Size
Cost Surface (e.g. travel costs)
 Distance measure is combined with a fixed cost per unit distance to calculate
travel cost
Re Classification
 assign new values to cells

ASTU Architecture Dept. 20

ArcGIS for Architects GE2304

Overly
 Mathematical operators are performed on corresponding cells from two or
more layers to produce an output value
 Input layers may be added, subtracted, multiplied, or divided to produce output
data
Mask
 Creates a new map that includes only those features of the input map that falls
within the area extent of the mask/clip map.
Surface Analysis
 Slope and Aspect: Calculated from a grid of elevations (a digital elevation model-
DEM)
2.3 Concept of DTM (Digital Terrain Model)
DTM represents the elevation of the Earth's surface using a grid of cells or pixels.
Each cell in the grid contains a numerical value representing the elevation at that
location. The size of the cells determines the resolution of the dataset, with smaller
cells providing higher resolution and more detailed information. While they are
related, they have distinct differences in terms of data content and applications.
2.3.1. Digital Terrain Model (DTM)
A DTM is a more refined version of a DEM (Digital Elevation model) that represents
the bare ground surface by removing or minimizing the presence of vegetation,
buildings, and other above-ground features. A DTM aims to provide a more accurate
representation of the bare terrain by filtering out non-ground objects.
ASTU Architecture Dept. 21
ArcGIS for Architects GE2304
2.3.2 DTM Data Sources
DTM can be derived from various data sources, including:
1. LiDAR (Light Detection and Ranging): LiDAR is a remote sensing technology
that uses laser pulses to measure the distance between the sensor and the
Earth's surface. LiDAR data can provide highly accurate and detailed elevation
information, making it a preferred source for generating DTM.
2. Photogrammetry: Photogrammetry involves capturing overlapping aerial or
satellite images and using computational techniques to extract elevation
information. Photogrammetric techniques can be used to generate DTM,
although the level of accuracy and resolution can vary depending on the quality
and coverage of the imagery.
3. Ground Surveys: Ground-based surveys, such as GPS (Global Positioning
System) or total station measurements can also be used to collect elevation data
at specific locations. These measurements can be combined to create DTM for
limited areas or for ground truth and validation purposes.
2.3.3 Applications of DTM data
DTM data is often used in various applications, such as:
1. Terrain Analysis
For experts to understand the complicated topography of landscapes, DTM and are
vital tool. These models make it easier to identify slopes, aspects and curves by
depicting the Earth’s surface as a grid of elevation points. This information is vital
and useful for land-use planning, agriculture and environmental management.
2. Modeling hydrological systems
Watershed analysis, flood risk assessment and water resource management all depend
on an understanding of water flow. By defining drainage networks, locating
watershed boundaries and forecasting the movement of water over various terrains,
DEM provides substantial contributions. This information is essential to reduce the
danger of flooding and develop efficient storm water management systems.

ASTU Architecture Dept. 22

ArcGIS for Architects GE2304
3. Infrastructure Planning
Infrastructure development and urban planning need precise elevation data. The
utilization of DTM facilitates the identification of appropriate construction sites,
evaluation of the effects of elevation on transportation networks and enhancement of
infrastructure project architecture which guarantees that developments are both
environmentally friendly and functional.
4. Agricultural Precision
In today’s era, technology plays a bigger role in agriculture. DTM supporst precision
agriculture by giving farmers precise elevation data, enabling them to plan effective
use of land, optimize irrigation and evaluate the risk of soil erosion. This maximizes
agricultural productivity while promoting sustainable farming methods.
5. Management of Natural Resources
DTM provide insights into the three-dimensional structure of the terrain, which helps
with natural resource management, like forestry, mining or wildlife conservation.
Such insights provided by DTM aids in planning extraction operations, keeping an
eye on deforestation and determining how terrain affects biodiversity.

ASTU Architecture Dept. 23