Data Models

Chapter 2
Pp 29-53
Lecture 2 1
 R eal W orld
D ata M odel D ata
S tructure
1.2 , 4.7
5.8 , 3.6
8.9 , 7.2
.
.
Phenomena/entities
that exist in the real Computer
world Representation
Machine Code
10011101

An abstraction, relevant phenomena 00110110

and properties Lecture 2
10110100 2
 Data Model
• Objects in a spatial
database.
• The spatial data
model provides a
formal means of
representing and
manipulating
spatially-referenced
information.
Lecture 2 3
 Coordinates
• Coordinates are used to define the
location and extent of spatial objects.

https://www.codeproject.com/Articles/44389/Build-a-
Desktop-GIS-Application-Using-MapWinGIS

Lecture 2 4
 Attributes
• Attributes describe the spatial object.
• Attribute data complements the coordinate data to define the spatial object.

5
Lecture 2
 Thematic Layers
• A logical
separation of data
according to
theme.
• Each layer reflects
a particular use or
characteristic.
• Overlays.
https://jessyfaas.wordpress.com/2016/07/01/gis-in-het-voortgezet-onderwijs-wat-is-gis/comment-page-1/

Lecture 2 6
 Coordinate Data
• GIS often uses a
Cartesian coordinate
system.
• Cartesian Coordinate
System (x & y or x, y & z)
• Usually specified as
decimal numbers that
increase from bottom to
top.

Lecture 2 7
 Coordinate Data
• Latitude & Longitude
– Origin (intersection of
the Equator and
Greenwich meridian)
• Spherical Coordinates
– Deg., min., sec. (DMS)
– Decimal degrees (DD)

Lecture 2 8
Nomenclature of Geographic
Coordinate System

http://www.justtrails.com/nav-skills/navigation-skills-latitude-and-longitude/

Lecture 2 9
 Conversion from Spherical to
Cartesian

Tissot’s Indicatrix

http://www.perrygeo.com/tissot-indicatrix-examining-the-
distortion-of-map-projections.html

Lecture 2 10
 Conversion
from DMS to DD

Convert: 68o 40’ 19.2” W to decimal degrees:

MIN SEC
DD  DEG  
60 3600
40 19.2
68  
60 3600
- 68.67200

Lecture 2 11
 Conversion
from DD to DMS 68o 48’ 57”

Convert: -68.67200 to degrees, minutes, seconds:

D  68
M  INT (0.67200 * 60)  INT (40.3200)  40
S  (0.3200 * 60)  19.2
DMS  68o 40'19.2"W

Lecture 2 12
 Converting Arc to Surface
Distance

d  r 
Where d is the ground distance, r is the
radius of the sphere and the angle is
specified in radians.

360  2 radians
o

Lecture 2 13
Converting Arc to Surface
Distance

At the equator, one degree

of longitude is about 111km

Lecture 2 14
An Ellipsoidal Earth

http://qcoherent.com/newsletter/December2014/December2014_2.html

Lecture 2 15
 Great Circle
A great circle is defined as any line resulting from a
plane passing through the center of the globe.

http://www.shsu.edu/~dl_www/bkonline/GenRef111/Reading/RefL01Maps/
Lab01index.htm

Lecture 2 16
 c=a-b

c a=90o-a
b=90o-b a
b

Latitude ϕ, Longitude λ

d  r cos [sin a sin b cos a cos b cos(a  b )]

1

Lecture 2
17
 Conversion from Geographic to
Cartesian Coordinates
x  R cos  cos 
y  R cos  sin 
z  R sin 
R  6371 km
This formula requires latitude and longitude to be in radians.

Lecture 2 18
Geographic & Magnetic North

http://www.ingv.it/ufficio-stampa/research-areas/the-earth/characteristics-of-the-earth-s-magnetic-field/view?
set_language=en
19
Lecture 2
 Types of Attribute Data
• Attribute data record the non-spatial characteristics of an
entity.
• Attributes have values
– Observed
– Measured

Llbean.com
Lecture 2 20
 Measurement of Attributes
• Physical scientists define measurement as
the comparison of an object to a standard
object.
• They define two types of measurements
– Extensive/Fundamental Properties (feet)
– Derived – by combining extensive properties
(feet/second)

Lecture 2 21
 Stevens’ Levels of
Measurement
• Social scientists weren’t satisfied with this
classification
• Stanley Stevens (1946) proposed a
framework for measurement types based
upon “levels of measurement”.
• He defined measurement as being the
assignment of classes or scores to
phenomena according to a set of rules.

Lecture 2 22
 Stevens’ Levels of
Measurement
• There are four basic levels according to
Stevens:
– Nominal – provides descriptive information.
– Ordinal – implies a rank order.
– Interval – implies order and difference in
magnitude.
– Ratio - implies order and difference in
magnitude and has an absolute 0.

Lecture 2 23
 Ordinal
• You can do comparisons:
– If A>B and B>C then the correct increasing
order is C, B, A; i.e., establish order
• You can establish equality between two
orders.

Lecture 2 24
 Interval
• Operations include:
– Count
– Equality
– Order
– Addition and Subtraction

Lecture 2 25
 Ratio
• Operations:
– Count
– Equality
– Order
– Addition and Subtraction
– Multiplication and Division
– Higher order operations

Lecture 2 26
 Additional Levels of
Measurement
• Nicholas Chrisman includes several more in his
textbook "Exploring Geographic Information
Systems."
• I will add those here:
– Absolute scales – scales bounded on both ends like
probability
– Cyclical measures – angular measure
– Counts are misfits. They are not continuous, but
otherwise behave as a ratio scale
– Graded membership in categories – Fuzzy set theory;
i.e., not all membership within a class must be equal.

Lecture 2 27
Common GIS Data Models
Spatial Data

Vector Data Raster Data

Non-topological Topological

Higher-level Data
Simple Data

Regions Dynamic
TIN
Segmentation

Lecture 2 28
Two Most Common
Spatial Data Models

Lecture 2 29
Triangulated Irregular Network
(TIN)

http://www.geosolutions.com/3d/analyse/images/interpolate_tin.gif

Lecture 2 30
 Regions

https://gis1.usgs.gov/csas/gap/viewer/land_cover/Images/Regions.png

Lecture 2 31
 Dynamic Segmentation
Dynamic segmentation is
the process of computing
the map location (shape)
of events stored in an
event table. Dynamic
segmentation is what
allows multiple sets of
attributes to be
associated with any
portion of a linear http://resources.arcgis.com/en/help/main/10.1/
index.html#//003900000026000000
feature.
Lecture 2 32
Contour Lines over a Raster

http://www4.ncsu.edu/~hmitaso/gmslab/hohen2/w2elev10.gif

Lecture 2 33
 Vector & Raster
• Vector is better at representing discrete
features.
• Raster is better at representing continuous
features
• A project may contain both vector and
raster layers.
• Spatial operations can only be performed
on one type of layer.
Lecture 2 34
 Vector & Raster (cont’d)
• The best data model for a given layer
depends upon the operations, the
experience and the views of the user.
• No decision is final, as one can be
converted to the other.

Lecture 2 35
 Vector Terminology
The terms
polygon and
area will be used
interchangeably.

Lecture 2 36
Multiple Representations

Lecture 2 37
Vector Model

Lecture 2 38
Single Part Features

1 to 1 Relationship

Lecture 2 39
Multipart Features

Many to 1 Relationship

Lecture 2 40
Multipart to Single-Part Problem

Lecture 2 41
Polygon Inclusion

Lecture 2 42
 Polygon Inclusions
• Areas in polygons that are part of the
polygon, but different from the rest of the
polygon: e.g. Islands in a lake.
• Solutions:
– Create separate polygons for each inclusion.
– Create an attribute column for coding
inclusions.

Lecture 2 43
 Boundary Generalization
• The incomplete representation of
boundary locations.

http://www.i95exitguide.com/side-trips/the-maine-coastline/

Lecture 2 44
 Vector Topology
Topology – geometric properties that to not change
with shape: Adjacency, Connectivity, Containment

Lecture 2 45
 Topology
Topology in the object data model is a set
of rules and software tools to define spatial
relationships an behaviors, such as:
– Polygons must not overlap within a dataset.
– Lines must not overlap themselves within a
data set.

Lecture 2 46
Three Types of Vector Features
Organization
1 Point ID X Y
2
Points 1 32.7 45.6
4
2 76.3 19.5
3 3 22.7 15.8
etc…..

1
C Line Begin End
6 A
Lines B
ID Point Point
A 6 9
9 239
B 9 1
C 239 1
etc…..
12 13
Polygon
11 22 ID Lines
52
Polygons A 11, 12, 52, 53,
53 54
41 B 52, 53, 9, 41,
54 9 22, 13

47
Lecture 2
 Advantages of Topology
• Maintain correct data spatial relationship (Find
errors)
• Efficient data storage (quickly process large
data sets)
• Facilitate spatial analysis (Network analysis,
Adjacent area analysis, overlay analysis

Lecture 2 48
Encoding Topological Primitives

Polygon Bounding Arcs

A (e,f,g,i,j) (h) (k)
B (a,b,c,-i)
C (-c,d,-j)
D (-k)

Lecture 2 49
Arc Bounding Nodes Left Poly Right Poly
a 1,2 - B
b 2,3 - B
c 3,5 C B
d 3,4 - C
…
Lecture 2 50
k 9,9 D A
Nodes Co-bounding Arcs
1 a,i,-g
2 -a,b
3 -b,d,c
4 -j,d,e
5 -c,j,-i
6 -e,f
Lecture 2 51
Vector Features, Tables &
Structures

Lecture 2 52
 Assignment

• Read Chapter 2 up to page 54

• End-of-chapter problems:
– 4, 6, 8, 9, & 12
– And the following problems:

Lecture 2 53
Calculate the distance between Bangor and
Glasgow using the Great Circle Formula and
compare your answer to the estimate below.

Bangor, Maine - 44.8012° N, 68.7778° W

Glasgow, Scotland - 5.85781° N, -4.24253° E

Estimated distance: 2,837 miles

Lecture 2 54
Encode the Topology for the
Diagram below:

Lecture 3 55