GIS DATA MODELS

Emmanuel A. Mensah
Osman Adams

1
Learning Outcomes
At the end of the module participants would be
able to:
 Explain the importance of data to Geographic information
Systems
 Identify the characteristics and sources of geographic data
 Explain the concept and purpose of data models
 Distinguish between spatial and attribute data
 Explain differences between vector, raster and TIN data
models and explain how they model real-world features in
a GIS.
 Understand the criteria for choosing a data model
 Explain how relational Data models store attributes
2
Content
 Information Systems and Data
 Characteristics and sources of Geographic data
 Spatial and Attribute Data
 Spatial Data Models
 Vector Data Model
 Raster Data model
 TIN Data model
 Criteria for choosing a spatial data model
 Attribute models
 Relational and object relational data model

3
Digital Information System

 An organized combination of computer hardware,

software, DATA, trained personnel and
communication networks solely for the purpose of
1. collecting,
2. managing,
3. transforming and
4. disseminating information.

O’Brien 2003
4
 Information System Components and Activities

Pr So
og
rc es ialist ram ftwar
u
so spe
c me e R
R e s a es o
ple nd IS nd urc
o a pro es
Pe ers ce
s du
ndU r es
E

Control of system performance

Dat

me es
Input of

dia
Processing Output of
aa

rc
Dat Know

data

es esou
of data into information
nd
aR

resources information products

and
R
es o dge b

Ma ware
u rc
le

chin
d
es s es

Ha r
Storage of data resources
a

Network Resources
Communication media and support
O’Brien 2003 5
Geographic Information Systems

 A computer based system to aid in the collection of,

maintenance, storage, analysis, output and
distribution of spatial data and information (Bolstad,
2002).

 A GIS is an information system that handles

GEOGRAPHIC DATA.

6
Nature of Geographic Data
 Spatial component
 Geometry (Dimensions)
 coordinate system (Location)
 Spatial relationship/Topology
 Descriptive component (Attributes)
 Describes spatial objects
 Time component
 temporal element
Types of Data in a GIS

 Spatial data ~ which describes location (where)

 Attribute data ~ which specifies characteristics at that
location (what, how much, and when)

Description (Attribute /Aspatial) Data

Name of Owner
Year built
Property Value
Name of Architect
GIS Data
Location (Geographical/Spatial/Geo)
Data
Accra
6o25’23.65’’ N 1o30’28.05’’ W
8
Geographic Data Layers
Sources of GIS Data
 Existing data
 Digital
 Map and
plan
 Paper files

 Acquisition
 Remote
sensing
 Photogramm
etry
 Field survey
10
Data Models

A data model is a set of constructs/structures for describing and representing

selected aspects of the real-world in a computer (Longley et al, 2005)

11
Spatial Data Models/structures
Spatial Data model is the means to represent
geographic aspects of the real world and defines the
type of operations that can be performed.

1. Vector data model

2. Raster data model

3. TIN data model 12

Vector Data Model
Dimensions of features (Objects) and
Corresponding symbols

13
Vector Data Model
Point - geometric description only refers to the
position of the object by a pair of coordinates (x,
y); used for trees, posts etc. Zero dimensional Point

Line - geometrically described by recording the

coordinates of a series of points and the
sequence in which these points should be linked
Line
(1D; eg., roads, canals, rivers)

Polygon - a line with the same start and end,

usually refers to an enclosed region (2D; eg.,
lakes, ponds, farm plots) Polygon
 Vector Data Model

NODE VERTEX

VERTEX

NODE
NO
DE

ARC
Vector model
7,10
5,9
10
9,8
A
4,7
polygon
1,6 8,6
6,6
5 2,5

line 5,4

point
2,2 4,1

5 10
16
VECTOR DATA – LAND USE

17
VECTOR DATA – POINTS,LINES,POLYGONS

18
Types of Vector Models

Non Topological (Spaghetti model)

 Geometry
 Annotation

Topological vector data

 Geometry
 Attributes
 Spatial Relationships
19
Topology and Spatial Relationships

Topological invariants

Interior Boundary
Exterior

Source: Wolfgang Kainz

Topology and Spatial Relationships

disjoint covered by

meet contains

equal covers

inside overlap

Source: Wolfgang Kainz

File Formats for Spatial Vector Data
KML/KMZ: Keyhole Markup Language file from Google Earth
DxF: Data interchange file
GPX:GPS Exchange format
GeoJason:
MDB: ESRI Personal Geodatabase format

Shape ‘file’: vector data format introduced with ArcView in 1993

 comprises several (at least 3) physical disk files (with
extension of .shp, .shx, .dbf), all of which must be present
 openly published specs so other vendors can create shape
files
CSV: Comma Separated Value File
TXT: Text File (Tab delimited)
XLS: Excel files
WFS: Web feature Service
Vector Data Model

Advantages Disadvantages
 It provides a more compact  It is a more complex data
data structure than the raster structure than a simple
model raster

 It provides efficient encoding  Overlay operations are more

of topology and as a result difficult to implement
more efficient implementation
of operations that require  The representation of high
topological information, such spatial variability is
as network analysis inefficient
 The vector model is better  Manipulation and
suited to supporting graphics enhancement of digital
that closely approximate hand- images cannot be effectively
drawn maps done in the vector domain
Raster Data Model
Raster datasets represent geographic features by dividing the
world into discrete square or rectangular cells laid out in a grid.
Each cell/PIXEL has a value that is used to represent some
characteristic of that location.
Raster Data Model
The thematic characteristics are what we
refer to as attributes. For example,
Attributes can be stored for features
represented in raster formats.
RASTER DATA – ORTHO PHOTO
RASTER DATA - DEM

27
RASTER – DEM IN 3D

28
RASTER DATA – Satellite Image

29
File Formats for Raster Data
1. TiFF: Tagged Image File Format (Geo TiFF’s)
2. IMG: ERDAS IMAGINE Data format
3. DAT: ENVI Data format
4. MPR: ILWIS Data format
5. HDF: Hierarchical Data Format
1. JPG2000: JPEG 2000 Raster Data Format
6.
2. ASCIIPortable
PNG: Grid Network Graphics
7.
3. Multi-Resolution
BIL: SeamlessBy
ESRI Band Interleaved Image Database (MrSID)
Line Format
8.
4. National
BMP: Imagery
Bitmap ImageTransmission
format Format (NITF)
9. National Land Archive Production System (NLAPS)
 GIF —Graphic Interchange Format for raster datasets

1. DEM: USGS Digital Elevation Model

2. XYZ: Gridded ASCII
3. SID: Multi Resolution Seamless Raster
4. GRID —Esri Grid raster dataset format
Advantages/Disadvantages Raster Models

 Simple data structure  Inefficient use of

 Easy and efficient computer storage
overlaying  Errors in perimeter and
 Compatible with Remote shape
Sensing imagery  Difficult to perform
 Simple for programming network analysis
by user  Inefficient projection
 Same grid cell definition transformations
for various attributes  Loss of information when
using large pixel sizes
 Less accurate and less
appealing map output
Hybrid Models
 Triangulated Irregular Network (TIN)
 Built from set of x,y locations with measured value
 Construct irregular tessellation of triangles

Delaunay triangulation:
- as much as possible equal-sided triangles
- circumcircles around triangles do not contain other anchor points
TIN DATA

33
When to use Raster or Vector

 Geographic Fields – Use Raster or TIN

 Have values everywhere
 May be discrete or continuous fields
 Example Land use, air temperature, atmospheric
pressure

 Geographic Objects – Use Vector

 Sparse in the study area (Empty spaces between)
 Have sharp boundaries.
 Example buildings,roads,street lamps, car parks, sports
field
34
 Attribute Data Types
DATA TYPE MAGNITUD Equal Absolute
E Interval Zero
Nominal

Ordinal
√
Interval
√ √
Ratio
√ √ √
Attribute Data Types
Categorical (Name): Numerical (Known difference
between values)
1. Nominal – String Values
• land use types, District 3. Interval
names  Magnitude
2. Ordinal  Equal interval
• Date (from 9999 BC to 2013
 Magnitude AD)
• road class; stream class • Latitude (from +90° to −90°)
• Health (healthy vs. sick)
• Opinion ('completely
agree'/ 'mostly 3. Ratio – Integer / Floating
agree'/'mostly point
disagree'/ 'completely  Absolute zero
disagree') •  income, age(from 0 to 99
years)   may be expressed
as integer [whole number] or
floating point [decimal
fraction]
Attribute Data Storage
Flat files (Example Spreadsheets)

Database – An integrated collection of logically

related records or objects shareable between one or
more applications.

Database Management System (DBMS) –

software to create, maintain and access
databases
GIS and Database Management Systems

System Task
• Data loading
GIS Software • Editing
• Visualization
• Mapping
• Analysis

Database Management • Storage

System • Indexing
• Security
• Query
Attribute Spatial
Data Data
Types of DBMS Models

Relational – RDBMS
Stores data in many sets of relations
Relies heavily on keys (unique identity fields) to dynamically link
records.
Current and most used

Object-relational – ORDBMS
Models entities as objects thus data for objects are stored in
tables within the relational database and behaviour of objects are
stored within the application as application logic.
Object – relational databases add rich behavior to spatial and
attribute data
Relational Data Model
 Relation (table)
Attribute (column)  domain is ?
Table name

Book BookId Title Author Year

1 Principles of GIS Burrough 1998
2 Introduction to DBMS Date 1990
3 Basics of remote sensing Joyce 1993
4 Principles of GIS Green 1990

Tuple (record, row) Attribute value

Borrow BookId StudentId DateBorrowed
1 11 21-09-1999
3 34 20-08-1999
4 11 17-08-1999
Relational Data Model
 Links between relations (related tables)

Book BookId Title Author Year

1 Principles of GIS Burrough 1998
2 Introduction to DBMS Date 1990
3 Basics of remote sensing Joyce 1993
4 Principles of GIS Green 1990

Primary key
Foreign key
Borrow BookId StudentId DateBorrowed
Author 1
Author_Name 11 Telephone City21-09-1999
3 34 20-08-1999
Burrough
4 11 030-123456 Utrecht
17-08-1999
Date 0430-4445556 New York
By 053-4874444 Enschede

Student StudentId Lastname Initials Course

11 Dawson N UPLA
Advantages of a Database
1. Provides concurrent Access – Simultaneous
Access for multiple users

2. Facilitates storage of large datasets – Can

store and manage Gigabytes, terabytes and
petabytes of data

3. Provides data Integrity – enforced through

key uniqueness, referential integrity,
Attribute domains and data types
Advantages of DBs (continued)
4. Provides redundancy control – established
through normalization of relations
5. Provides backup and recovery

6. Data model for definition of database schema

7. High level declarative language for database
queries.

8. Improved Access control

 Questions?
Email: [email protected]
Mobile:0208017140

44