8/2/2019 Spatial Analysis 6th semester
1/51
Foundations for Applied GIS
Spatial Analysis
Geog-3205
Khurram Chohan
8/2/2019 Spatial Analysis 6th semester
2/51
Spatial Analysis
Geographic information analysis is concerned withinvestigating the patterns that arise as a result ofprocesses that may be operating in space.
Representation, Description , Measurement,Comparison , and generation of spatial pattern arethe main techniques / methods to GeographicInformation Analysis.
2
8/2/2019 Spatial Analysis 6th semester
3/51
Literature Review
Spatial data manipulation, usually in a geographicinformation system (GIS), is often referred to asspatial analysis, particularly in GIS companies'promotional material. Your GIS manuals will give
you a good sense of the scope of these techniques,as will texts by Tomlin (1990) and, more recently,Mitchell (1999).
Buffering
Point in Polygon..Queries
3
8/2/2019 Spatial Analysis 6th semester
4/51
Literature Review
Spatial data analysis isdescriptiveandexploratory. These are important first steps in allspatial analysis, and often all that can be done withvery large and complex data sets. Books by
geographers such asUnwin (1981), Bailey andGatrell (1995), and Fotheringham et al. (2000)
4
8/2/2019 Spatial Analysis 6th semester
5/51
Literature Review
Spatial statistical analysis employsstatisticalmethodsto interrogate spatial data to determinewhether or not the data are "typical" or"unexpected" relative to a statistical model.
By Ripley (1981, 1988), niggle (1983), and Cressie(1991).
5
8/2/2019 Spatial Analysis 6th semester
6/51
Literature Review
Spatial mode involves constructing models topredict spatial outcomes.
In human geography, models are used to predict
flows of people and goods between places or tooptimize the location of facilities (Wilson, 1974,2000)
In environmental science, models may attempt tosimulate the dynamics of natural processes (Ford,1999).
6
8/2/2019 Spatial Analysis 6th semester
7/51
Spatial Data Types
Raster / Image Data
Vector Data
7
8/2/2019 Spatial Analysis 6th semester
8/51
Spatial Data Types
Vector View It records locational coordinates of points , Lines, and
areas.
The vector model conforms to an object view of the world,where space is thought of as an empty container occupiedby different sorts of objects.
8
8/2/2019 Spatial Analysis 6th semester
9/51
Spatial Data Types
Raster / Image View Instead of starting with objects on the ground, a grid of
small units of Earth's surface (called pixels) is defined.
For each pixel, the value, or presence or absenceof something of interest, is then recorded.
9
8/2/2019 Spatial Analysis 6th semester
10/51
Higher Level Abstraction of Spatial Data Types
An Object
The Object View
The Field View
10
8/2/2019 Spatial Analysis 6th semester
11/51
Higher Level Abstraction of Spatial Data Types
The Object View
world is considered as a series of entities located inspace.
Entities are (usually) real
You can touch them, stand in them, perhaps evenmove them around.
Example..Places can be occupied by any number ofobjects. A house can exist in a census tract, which mayalso contain lampposts, bus stops, road segments, parks,and so on.
11
8/2/2019 Spatial Analysis 6th semester
12/51
The Object View
The object view has advantages when well-definedobjects change in time: for example, the changing
data for a census area object over a series ofpopulation censuses.
12
Higher Level Abstraction of Spatial Data Types
8/2/2019 Spatial Analysis 6th semester
13/51
Higher Level Abstraction of Spatial Data Types
An Object
An object is a digital representation of all or part ofan entity. Objects may be classified into different
object types: for 'example, intopoint objects, lineobjects, and area objects.
Example------woods and fields
13
8/2/2019 Spatial Analysis 6th semester
14/51
The Field View
In the field view, the world is made up of propertiesvarying continuously across space.
Example-----Earth SurfaceWhere ElevationVaries
Similarly, we can code the ground in a grid cell as
either having a house on it or not. The result is alsoa field, in this case of binary numbers where1=house and 0= no house.
14
Higher Level Abstraction of Spatial Data Types
8/2/2019 Spatial Analysis 6th semester
15/51
Raster Data Model is not only the one way torepresent the geographic variations where data arerepresented in regularly shaped Pixels.
An alternative way to represent surface in a meshof non-overlapping triangles called TriangularIrregular Network (TIN)
A good example is a map of soil type. Everywherehas a soil, so we have spatial continuity, and wealso have self-definition by the soil type involved, sothis is a field view.
15
Higher Level Abstraction of Spatial Data Types
8/2/2019 Spatial Analysis 6th semester
16/51
16TIN
RASTERSOIL MAP
Higher Level Abstraction of Spatial Data Types
8/2/2019 Spatial Analysis 6th semester
17/51
Land Use Maps
These types have been given different names: k-color maps , Black and White
K-Color technique, each type is assigned a specificcolor required to show the variation.
17
Higher Level Abstraction of Spatial Data Types
8/2/2019 Spatial Analysis 6th semester
18/51
Right Choice to representation of spatial world
Real world is represented in the form of elementsand these elements are stored in database.
Entity Vs. Object
Entity must be identifiable; if you can not see it,you can not record it.
Entity must be relevant, and of interest.
Entity must be describable. Therefore, it must haveattribute / characteristics so that we can record it.
18
8/2/2019 Spatial Analysis 6th semester
19/51
Right Choice to representation of spatial world
Definition of an Entity: Entity is defined as a phenomenon of interest in reality
that is not further subdivided into phenomena of the samekind.
Examples:
Road Network Vs. Roads
Forest Vs. stands.
19
Ri h Ch i f b i f
8/2/2019 Spatial Analysis 6th semester
20/51
What you want to do with the data entered into thecomputer system?
What should be the map design and objectives?
What should be the model to represent real world?
20
Right Choice for better representation ofspatial world
8/2/2019 Spatial Analysis 6th semester
21/51
Types of Spatial Objects
Point: an object with no length,L0
Line : an object having the same spatial dimensionas any simple length, that is, L 1.
Area : an object with spatial dimension lengthsquared, or L2
Cartographic conventions.
21
8/2/2019 Spatial Analysis 6th semester
22/51
Types of Spatial Objects
Theoretical and Practical Issues
There is a need to separate the representation ofan object from its fundamental spatial
characteristics. Forexample, a line object may beused to mark the edge of an area.
Geographic scale is important.
A railway station may be represented as a point, a set oflines, or an area.
The objects discussed are often two-dimensional
Need to represent Elevation / Depth22
8/2/2019 Spatial Analysis 6th semester
23/51
Types of Spatial Objects
This view of the world is a static one. This is fine for some problems, but in many applications
our main interest is in how things evolve and change overtime.
23
8/2/2019 Spatial Analysis 6th semester
24/51
Scales for attribute description
24
8/2/2019 Spatial Analysis 6th semester
25/51
It is understood that reality represented in point,line, polygon.
Geometric entities are embedded with aspatial
data. The range of possible attribute is huge: Color, age
,use, ownership, and so on.
Attributes can be classified into types based ontheir level of measurements.
25
Scales for attribute description
8/2/2019 Spatial Analysis 6th semester
26/51
Nominal Scale Each value is a distinct category, serving only to
label or name the phenomenon.
We call certain buildings "shops" and there is noloss of information if these are called "category 2"instead.
Categories must be inclusive / Mutually exclusive
26
Scales for attribute description
8/2/2019 Spatial Analysis 6th semester
27/51
Scales for attribute description
Ordinal Scale Nominal scale does not imply relationships between
classes.
Ordinal scale rank classes according to somecriterion.
Classification of land according to its agriculture potential
Note: Attributes measured on the nominal and ordinal scales
are often referred to collectively as categorical data. 27
Ordinal (Ranks)
Good Better
Best
8/2/2019 Spatial Analysis 6th semester
28/51
Scales for attribute description
Interval Scale: The interval level of measurement has the property
that differences or distances between categories
are defined using fixed equal units.
interval scales lack an inherent zero
Thermometers typically measure on an interval scale,ensuring that the difference between, say, 25 and 35"C is
the same as that between 75.5F and 85.5F. 28
Interval (numeric)
Age
Income
8/2/2019 Spatial Analysis 6th semester
29/51
Scales for attribute description
Ratio Scale
Example
If place A is 10 km (6.2137 miles) from B and 20 km(12.4274 miles) from C, the ratio of the distances is
distance AB /distance AC = 10 / 20
= 1/2
29
Ratio (scale)
Length
Area
8/2/2019 Spatial Analysis 6th semester
30/51
Spatial Data Types In Everyday Life
30
8/2/2019 Spatial Analysis 6th semester
31/51
Spatial Data Manipulation and Analysis
31
8/2/2019 Spatial Analysis 6th semester
32/51
Mapping and Visualization Geometric Intersections, Buffering, and
Point-in-Polygon Tests
Map Overlay
Linking GIS and Spatial Anaysis
32
Spatial Data Manipulation and Analysis
8/2/2019 Spatial Analysis 6th semester
33/51
Mapping and Visualization
maps have been used for centuries as a datastorage and access mechanism for topographic andcadastral (land ownership) information.
Thematic Maps are being used to display statisticdata and result of other systematic surveys.
Maps created specifically to highlight the distribution of aparticular phenomenon or theme are called Thematic
Maps
a
33
8/2/2019 Spatial Analysis 6th semester
34/51
Mapping and Visualization
Population change in the United States, by county,from 1990 to 2000.(Data from 1990 & 2000 decennial censuses).
34
8/2/2019 Spatial Analysis 6th semester
35/51
Mapping and Visualization
35
8/2/2019 Spatial Analysis 6th semester
36/51
Mapping and Visualization
A "dot density" map that depicts count data.Cartography by Geoff Hatchard.
36
8/2/2019 Spatial Analysis 6th semester
37/51
Mapping and Visualization
A "proportional circle" map that depicts count data.Cartography by Geoff Hatchard.
37
8/2/2019 Spatial Analysis 6th semester
38/51
Mapping and Visualization
A "pie chart " map that depicts rate data.Cartography by Geoff Hatchard.
A pie chart is a circular chart
which is divided into sectors,
illustrating proportion
In a pie chart, the arc length
of each sector is proportional
to the quantity it represents
38
8/2/2019 Spatial Analysis 6th semester
39/51
Mapping and Visualization
A "bar/column chart" map that depicts rate data.Cartography by Geoff Hatchard.
A bar chart or bar graph is a
chart with rectangular bars
with lengths proportional to the
values that they represent.
The bars can be plotted
vertically or horizontally.
39
8/2/2019 Spatial Analysis 6th semester
40/51
Mapping and Visualization
A "graduated color" (choropleth) map that depictsdensity data. Cartography by Geoff Hatchard.
A choropleth map is a thematic map
in which areas are shaded
or patterned in proportion
to the measurement of
the statistical variable
being displayed on the
map, such as population
density or per-capita income.
40
8/2/2019 Spatial Analysis 6th semester
41/51
Mapping and Visualization
A "unique values" map that depicts density data. Notethat the legend, which in the original shows onecategory for each state, is trimmed off. Cartography byGeoff Hatchard.
Logically or not, peopleprefer colorful maps.
For this reason some
might be tempted to
choose Arc Map's Unique Values
option to map rates, densities,
or even counts.
This option assigns a unique
color to each data value 41
8/2/2019 Spatial Analysis 6th semester
42/51
Broad Street cholera outbreak in Soho, London 1854.
42
Geometric Intersections, Buffering, and point-
8/2/2019 Spatial Analysis 6th semester
43/51
Geometric Intersections, Buffering, and pointin-polygon Tests
Is a measure of the distance between features. It is most commonly measured in units of length but
can be measured in other units.
Such as travel time or noise level
Four parameters must be specified to measure
43
Parameters Parameters (Examples)
Target Location A Road, A hospital, or A park
A unit of measure Distance in meters
Function to calculate proximity. Straight line distance, travel time.
The area to be analyzed The area to be analyzed
Geometric Intersections, Buffering, and point-
8/2/2019 Spatial Analysis 6th semester
44/51
we can easily determine how many cases of adisease occur within various distances of certainkinds of factory or other facility.
We need geocoded data for cases of the diseaseand also for the facilities.
Point-in-polygon operations allow us to determinehow many cases of the disease occur in the
relevant buffer areas
44
Geometric Intersections, Buffering, and pointin-polygon Tests
8/2/2019 Spatial Analysis 6th semester
45/51
45
8/2/2019 Spatial Analysis 6th semester
46/51
4/6/201
2
46
8/2/2019 Spatial Analysis 6th semester
47/51
Map Overlay
In a map overlay two or more layers are overlaid inorder to produce a new layer.
47
8/2/2019 Spatial Analysis 6th semester
48/51
48
Map Overlay
8/2/2019 Spatial Analysis 6th semester
49/51
49
Map Overlay
8/2/2019 Spatial Analysis 6th semester
50/51
50
8/2/2019 Spatial Analysis 6th semester
51/51
Linking GIS and spatial analysis
The GIB view of spatial data and that of spatialanalysis are different.
Spatial analysis is not widely understood.
The spatial analysis perspective can sometimesobscure the advantages of GIS.
Top Related