Fundamentals of Geographic

Information System
Session 3

Spatial Data Models

Learning Objectives
Nidhi Misra • Concepts of spatial data
Senior Geologist models
Centre for Geoinformation Management Training (CGMT) • Concepts of Tessellation
Geological Survey of India Training Institute Hyderabad • Representation of geographic
fields and objects, Raster,
Vector, TIN, etc.
GSITI
 What is a Data Model?

Model:
• Miniature representation of a system whose
properties are to be studied

Data Model:
• Specifies structure or schema of a data set
• Documents data

Ref: ESRI help files for Arc GIS

GSITI
 Raster and Vector

• Geographic objects are discrete entity with a defined

boundary and are generally represented by vector
• Geographic fields are continuous entity with values
everywhere and are generally represented by raster
• Thus, Raster & Vector represent different ways to
encode and generalize geographic phenomena
• However based on requirement both raster and vector
may be used to code both fields and discrete objects
Ref: ESRI help files for Arc GIS

GSITI
 Raster and Vector based representation of Geographic
objects

Real World

0 1 2 3 4 5 6 7 8 9
0 R T
1 R T
2 H R
point
3 R line
4 R R
5 R
6 R T T H
7 R T T polygon
8 R
Ref: ESRI help files for Arc GIS
9 R
Raster Representation Vector Representation
GSITI
 Data Sources for Raster and Vector GIS

Raster Vector

Remote Sensing Digitizing

Digital Elevation Models Coordinate
(DEM)
Raster Graphics Geometry
Rasterized Vector Surveying
Features
Images GPS
Tables Vectorization of Raster
Datasets
Ref: ESRI help files for Arc GIS

GSITI
 Tessellation Approach

Definition

• Tessellation (or tiling) is the partition of space into disjoint

cells

• Each cell contains information that characterize that part of

space

• Regular tessellation (all cells same)

• Irregular tessellation (cells may be different)

Ref: ESRI help files for Arc GIS

GSITI
 Regular Tessellation

• Cells are the same shape and size

• Field value assigned to a cell is associated with
entire area occupied by that cell
Types:

Any geometrical
shape
which completely
occupies the entire
space.

Square cells most used: simple georeferencing, also called

Ref: ESRI help files for Arc GIS
raster (or grid)

GSITI
 Regular Tessellation

Advantage:

• Prior knowledge of space partition and therefore

fast computations

Disadvantage:

• Not adaptive to geographic phenomenon we want

to represent (smaller raster size)
e.g.: representation of elevation in flat terrain v.s.
steep terrain
Ref: ESRI help files for Arc GIS

GSITI
 Irregular Tessellation

• Partition into disjoint cells with varying shape and size

• Field value assigned to a cell is associated with entire

area occupied by that cell

• Many types. eg. region quadtree

Based on regular tessellation of square cells, but
neighbouring cells with same value will be represented
as one bigger cell

Ref: ESRI help files for Arc GIS

GSITI
 Irregular Tessellation

Advantages:

• Reduction in amount of memory used to store the

data
• Adaptive: more variation in values, more cells
• Quick computation of areas with same field value

Disadvantages:

• More complex computations, algorithms

Ref: ESRI help files for Arc GIS

GSITI
 Raster Generalization
The two rules for raster generalization

Largest share rule Central point rule

Ref: ESRI help files for Arc GIS

GSITI
 Quiz
Which of the following rule is used for Raster Generalization ?

Options
1. Central Point Rule 2. Corner Point Rule

Correct Answer: Option 1. Central Point Rule

Remarks: The two rules used for Raster Generalization are

1. Central Point Rule and 2. Largest Share Rule

GSITI
 Raster Data Model
• Area is covered by grid with equal-sized,
square cells
• Attributes recorded by assigning each cell
a single value based on majority feature
• Image data is a special case of raster data
in which the “attribute” is a reflectance
value from the electromagnetic spectrum

Ref: ESRI help files for Arc GI

GSITI
 Grid Data Structure

Grid size is defined by extent, spacing and no data value

information
• Number of rows, number of column
• Cell sizes (X and Y)
• Top, left , bottom and right coordinates

Grid values
• Real (floating decimal point)
• Integer (may have associated attribute table)

Ref: ESRI help files for Arc GIS

GSITI
 Rows and Columns in Raster data
Raster Data Model
0 1 2 3 4

0
1
3,2

2
Row (y)
(x,y)

3
4

Column (x) Ref: ESRI help files for Arc GIS

GSITI
 Slope Float Raster sans Table

Ref: ESRI help files for Arc GIS

GSITI
 Slope Integer Raster with Table

Ref: ESRI help files for Arc GIS

GSITI
 Representation of Geographic Fields
Digital Elevation Models

Vector: Triangulated Irregular Network (TIN)

Raster: Grid

Ref: ESRI help files for Arc GIS

GSITI
 Field represented by Vectors

Ref: ESRI help files for Arc GIS

GSITI
 Quiz
Which of the following is true about attribute table in a Float
Raster ?
Options
1. Attribute Table has 2 Prefilled columns 2. Attribute table does not exist

Correct Answer: Option 2. Attribute table does not exist

Remarks: A Float Raster does not have an Attribute Table.

An Integer / Discrete Raster has an Attribute Table.

GSITI
 धन्यवाद / Thank You
GSITI