SPATIAL DATA ANALYSIS ........ Modified2

V.
SPATIAL DATA ANALYSIS
Contents
 Classification of analytical GIS capabilities
 Retrieval, classification & measurement
 Overlay functions
 Neighbourhood functions
 Network analysis
Compiled by Nigatu W. HU-IoT-BEE-July, 2016 1

 So far, we have discussed the nature of spatial data, notions of data acquisition and
preparation.
 Here, it is important to make a distiction b/n analytical models (often referred to as
models) and analysis (or analytical operations).
 By analysis we mean only a subset of what is usually implied by the term, but not all
types of analysis (ex. Statitical analysis will not be dealt with).
 All knowledge of the world is based on models of some kind, whether they are simple
abstractions or complex equations that describe a physical phenomena.
 Spatial data/map itself is a kind of model of some part of the real world.
 In this topic, we will focus on analytical functions that can form the building blocks for
application models. Source: Principles of GIS
2
Compiled by Nigatu W. HU-IoT-BEE-July, 2016
 Therefore, here Analysis = analytical operations

 Definition: Analysis of spatial data can be defined as computing new information that
provides new insight from the existing, stored data.
Source: Ellen-Wien Augustijn & Principles of GIS

 Classification of analytical GIS capabilities
 There are many ways to classify the analytical functions of a GIS.

 One can classify the analytical GIS capabilities into the following groups:
1. Classification, retrieval, and measurements
 All functions in this category are performed on a single (vector or raster) data, often using the associated
attribute data.
2. Overlay functions
 These allow the combination of two or more spatial data layers treating features at the same location.
3. Neighbourhood functions, and

 Neighbourhood functions evaluate the characteristics of an area surrounding a feature’s location.
4. Connectivity functions
 These functions work based on networks/connectivity between features, such as road network.
Source: Principles of GIS
4
1. Classification, retrieva, and measurement functions
 Classification: allows the assignment of features to a class on the basis of attribute values or
attribute ranges (definition of data patterns). On the basis of reflectance characteristics found in
a raster, pixels may be classified as representing different lannd cover types.
 Retrieval functions: allow the selective search of data. We might thus retrieve all agricultural
fields in a set of land cover types.
 Measurement function: allow the calculation of distances, lengths, or areas.
 Generalization: is a function that joins different classes of objects with common characteristics
to a higher lelel (generalized) class. For instance, we might generalize fields where open and
dense forests are found as simply ‘FOREST’.
 In geography, the term ‘aggregation’ is often used to indicate the process of generalization ISPIT
 Measurement operations
 Geometric measurement on spatial features include (excluding attribute data measurement

which will be done in query language):
 Distance and area size computations in practical #2, length, area, perimeter
 For simplicity, such measurements will be seen in a planar spatial reference system.
 Measurements can be done on both vector and raster data.
Ex. Finding sum is attribute data measurement, not

seen in geometric measurement
Source: Ellen-Wien Augustijn & Principles of GIS Compiled by Nigatu W. HU-IoT-BEE-July, 2016 6
 Measurements-Vector
 The primitives of vector data sets are point, (poly)line and

polygon.
 Related geometric measurements in vector data are location,
length, distance b/n two features and area size. In some cases,
minimal bounding box is used.
 Location:
 Stored in a GIS by a single co-ordinate pair (X,Y) for a point
 A list of pairs for a line (segment) or polygon boundary

 Measurements-Vector Pythagorean distance
function:
 Length:
 Stored as a single line
 As a group of lines
 Polygon boundary (perimeter)
 Area calculation:
 Area size is associated with polygon
features. Again, it can be stored with
the polygon as an extra attribute
value.
 Measurements-Vector
 Another geometric measurement used by the GIS is the minimal bounding box (MBB) computation.
 It applies to polygons, and determines the minimal rectangle-with sides parallel to the axes of the
spatial reference systems-that covers the feature.
 It is an important support function for the GIS: For instance, if the bounding boxes of two polygons do
not overlap, we know the polygons cannot possibly intersect eachother.
 Since polygon intersection is a complicated function, it will help to check the presence of intersection
of polygons. Polygons intersect,
MBB overlap
Polygons do not
intersect, MBB
overlap
 Measurements on Raster data
 Measurements on raster data layers are simpler because of the regularity of the cells
 The area size of a cell is constant and is determined by the cell resolution.
 The so-called anchor point and resolution are the geometric information stored with the
raster data
 The anchor point is fixed by convention to be the lower left (sometimes upper left) location
of the raster.

 Raster measurements include: location, distance and area size.

 Location of an individual cell derived from raster’s anchor point and the resolution
 The cell’s location can be its lower left corner (X: 25546, Y:468678) or midpoint
Resolution: 20 x 20 meters

 The distance b/n two raster cells is the standard distance function applied to the
locations of their respective mid-points, taking into account the cell resolution.
Cell Resolution=30m
ISPIT
𝑧= 302 + 302 = 2(302 )
67.08m
Source: Ellen-Wien Augustijn & Principles of GIS Is this right?
 The area size of a selected part of the raster (a group of cells) is calculated as the
number of cells multiplied by the cell area size.
 The number of cells is also called the frequency count.
Area=
ISPIT

 Spatial selection query
 When exploring a spatial data set, the 1st thing one usually wants is to select certain
features interactively, to restrict the exploration.
Selection
 In interactive spatial selection, one defines the selection condition by pointing at or object is
polygon
drawing spatial objects on the screen display, after having indicated the spatial data
layer(s) from which to select features.
 The interactively defined objects are called selection objects.
 The GIS then selects the features in the indicated data layer(s) that overlap (i. e. intersect,
meet, contain, or are contained in)
 Features within the selection objects are selected and highlighted (also in attribute table).
 Interactive spatial selection answers questions like ‘’What is at ….?’’

 Example of interactive spatial selection
 What is at …..? What is at a specific location?
 selection in which you select features by

clicking on the screen (on the feature to
select) or
 drawing a graphic, to select all objects

within this graphic

 Spatial selection by attribute conditions
 Define a selection condition on the features attributes in a query language, such as SQL.
 Display the result both on the map and in the attribute table.
 Answer questions : Where are the features with …… ?
 Atomic conditions use predicate symbols (Area < 4000)
 A condition that tests a single criterion is called an atomic or single condition.
 Single conditions make use of operators (arithmetic & relational):
Relational
Arithmetic = equal to
+ addition < less than
- subtraction <= less than or equal to
* multiplication > greater than
/ division >= greater than or equal to
Source: Ellen-Wien Augustijn & Principles of GIS <> not equal to
 Example of attribute condition
Select all the land use areas with a size > 4,000,000 m²
 Combining attribute conditions
 When multiple criteria have to be used for selection, they can be expressed in a series of
single conditions.
 Single conditions can be combined into composite conditions using logical connectives
(operators):
BOOLEAN OPERATORS
 AND
 OR

Known as
XOR
 NOT
ISPIT
 Combining attribute conditions
Syntax – logical connectives
Composite conditions use logical connectives:

 AND (returns true if both expressions a and b are true)
 OR (returns true if one or both of the expressions a and b is true)
 NOT (returns true if expression is false)
 Area < 400,000 (atomic condition)

ISPIT
 (Area < 400,000) AND (landuse =80) (composite cond.)
 (Area < 400,000) OR (landuse = 80) (composite cond.)
 NOT (landuse = 80) (negate condition) Source: Ellen-Wien Augustijn & Principles of GIS

 Example of combining attribute conditions
Select all the land use areas with a size > 4,000,000 m² AND land use = 70
 Spatial selection using topological relations
 Various forms of topological relationships between spatial objects were discussed in earlier
topics.
 These relationships can also be useful to select features as well.
 The spatial operations are:
 Containment relationship (inside): i. e. selecting features that are inside selection
objects. Ex. Polygon can contain-polygon, lines and points.
 Intersect relationship (overlap): i. e. selecting features that intersect. ISPIT
 Meet relationship (adjacent): i. e. selecting features adjacent to selection objects.

 Distance: i. e. selecting features based on their distance. Ex. Searching within a given
distance from the selection objects or at a given distance.
 Spatial selection using topological relationships
 Spatial selection using topological relationships-
 Data retrieval involves the selective search of data.
 Example: Data retrieval with

topological CONTAINMENT
RELATIOHSHIP
Select all clinics

in Ward “A”.
Selection 1: Ward A
Selection 2: Clinics
 Classification
 Classification is a technique of purposefully removing detail from an input data set, in the
hope of revealing important patterns (of spatial distribution).
 In the process, we produce an output data set, so that the input data set can be left intact.
 This is done by assigining a characteristic value to each element in the input set, which is
usually a collection of spatial features that can be raster cells or points, lines, or polygons.
 If the number of characteristic values is small in comparison to the size of the input set, we
have classified the input set.
 Ex1. Household income of a city can be classified into few categories: say, low. Below
average, above average, and high.
 Ex2. Classify an elevation map into classes with intervals of 50 m.
 Classification
 The input data set may have itself been the result of a classification, and in such a case we
call it a reclassification.
 (Re) classification will produce a new output data set.
 For instance, we may have a soil map that shows different soil type units and we would
like to show the suitability of units for a specific crop.
 In this case, it is better to assign to the soil units an attribute of suitability for the crop.
Generally: ISPIT
 Classification  Reclassification
 The number of classes is reduced or increased  The original number of classes does
 May have geometric editing after the classification not change (ex. Soil type & suitability)
 Also called generalization, aggregation, merge  No geometric change on vector data
and dissolve operations  Different classification systems
 Reclassification
Soil unit map Reclassify the soil unit map into a PH map
4 5
2
3 6 Note:
1 No change in the
# of classes
Soil PH map
6.2 7.4 8.0
5.8 6.3 7.2
SPATIAL DATA ANALYSIS User-controlled classification:
 User defined (controlled) classification from 8 land use classes into 4 more
general classes (spatial merging)
 In user-controlled classification we indicate the
classification attribute and the classification
method
 This is normally done via a classification table.
CODE OLD TYPE NEW TYPE

10 Planned Residential Residential
Merged 20 Industry Commercial
30 Commercial Commercial
4 classes 40 Institutional Public
50 Transport Public
60 Recreational Public
70 Non Built-up Non Built-up
80 Unplanned Residential Residential
 In this section, we will look at techniiques of combining two spatial data layers and
producing a third from them.
 The binary operators that we discus are known as spatial overlay operators.
 Standard overlay operators take two input data layers, and assume they are
georeferenced in the same system, and overlap in study area. ISPIT
 The principle of spatial overlay is to compare the characteristics of the same location in
both data layers, and to produce a new output value for each location.

 Vector overlay functions ISPIT
 Intersection: The potato fields on clay soils.
 Union: The fields where potato or maize is the crop.
 Difference: The potato fields not on clay soils.
 Complement: The fields that do not have potato as crop.

 Overlay operations - Intersect

 Only the features inside the common area of the two input data are kept in the output.
 One input data can be point, line or polygon feature type, the other must be a polygon data set.

 Overlay operations - Clip
• Cookie cutter
• Restricts the spatial extent to the outer boundary of
the second layer
• No other polygons from the second layer play a role.

 Raster Overlay operations
 Vector overlays are geometrically complicated and may result in poor performance.
 Raster operators do not suffer from such disadvantage, as the computations are
performed cell-by-cell, thus they are fsater than vector. ISPIT
 GIS raster processing have a language to express operations on rasters.

 These languages are referred to as map algebra or sometimes raster calculus.
 They allow a GIS to compute new rasters from existing ones, using a range of functions
and operators.

 Arithmetic operators
 Various arithmetic operators are supported. The standard ones include:

 addition (+),
 Subtraction (-),
 Multiplication (*), and
 Division (/).
 Other arithmetic operators may include:
 modulo division (MOD): Ex. Modulo division returns the remainder of division: 10 MOD 3
return 1 as 10-3x3=1
 Integer division (DIV)
 goniometric operators: sin, cos, tan, asin, acos, atan. Source: Principles of GIS

 The following are some simple map algebra assignments.

 The assignment C1:=A+10, this will add a constant factor of 10 to all cell values of raster

There are also other raster overlay operators, such as:

 Comparison and logical operators
 Conditional expressions, and
 Decision table
Please refer to these operators in the reference and text books
Assignment 1

 Neighbourhood functions
 The principle of neighbourhood in spatial analysis is to find out the characteristics of the vicinity of
a location.
 Here the question is not only ‘what is at the location?’, but also ‘what is near the location?’.
 Thus, neighbourhood analysis deals with questions on what is near a specific location?
 To perform neighbourhood analysis, we must:
 State which target locations are of interest to us, and define their spatial extent
 Define how to determine the neighbourhood for each target, and
 Define which characteristic(s) must be computed for each neighbourhood.
 For instance, our target might be a medical clinic. Its neighbourhood could be defined as:
 An area at 2km distance as the crow flies, or (air)
 All roads within 500m travel distance, etc.

 As a characteristics, total population of the area can be analyzed. Source: Principles of GIS
 Neighbourhood operations
 Neighbourhood operations include:
 Proximity functions
. makes use of the geometric distance function
 Buffer (vector and raster)
 Thiessen polygon generation (vector)
 Spread functions (raster) PROJECT
. assumes that the phenomenon spreads in all directions, but not necessarily equally easily in all directions
 Seek functions (raster)
. the phenomenon will choose a least-resistance path

 Proximity functions
 Definition of neighbourhoods of features based on

distance
(distance in units of length, travel time, etc.)
 Target locations can be point, lines or polygons
 Typical examples:
Determination of buffer zones along groundwater
exploration wells
Construction of Thiessen polygons
Source: ITC & Principles of GIS
 Thiessen polygons
Example of Thiessen rainfall map
 Set of points as target locations
 Question to be answered:
Given any location in a study area, to which
target location is it closest ?
 Polygons are created indicating which area

“belongs” to which target point.
Source:ITC & Principles of GIS

 Neighbourhood operations Spread of pollution
 Spread (diffusion) computation
 In spread computation the neighbourhood of a target

location not only depends on distance but also on direction
and differences in the terrain
 This is especially the case when a target location contains a
“source material” that spreads over time.
 Examples of source materials:
 Air (temperature, humidity, pollution)
 Good example is spreading of pollution
 Water flow or soil contaminants

 Seek (flow) computation
 Spread computation determines how a phenomenon

spreads over an area in all directions but with different
difficulty or resistance
 In some cases the phenomenon does not spread in all

directions, but only along a chosen ”least-cost path”
 This is when we use SEEK computation
 Typical example is determination of the path of water

flow in a catchment area
Source:ITC & Principles of GIS Compiled by Nigatu W. HU-IoT-BEE-July, 2016 41
Elevation raster
 Seek (flow) computation cont’d
 Input for a seek computation is an elevation raster
Accumulated flow count

 For each cell the steepest downward slope to a
Flow direction raster
neighbouring cell is determined
 The direction of this downward slope is stored in the
flow direction raster
 The value of the accumulated flow for each cell is the
number of cells that flow into this particular cell. ISPIT
Accumulated flow count raster
 Cells with a value 0 have no other cells flowing into
them, and represent higher areas. ISPIT

Raster based surface analysis –different types and include:

 Connectivity/network functions
 A netwok is a connected set of lines, representing some geographic phenomena, typically of

the transportation type.
 The ‘goods’ transported can be almost anything:
 People, cars and other vehicles along a road network
 Phone calls along a telephone network,
 River network, etc.
 Network analysis can be performed on either raster or vector data layers, but more commonly done on
vector data layers, as line features can be associated with a network.
 A fundamental characteristic of any network is whether the network lines are considered directed or not.
 Directed networks associate with each line a direction of transportation (one directional)
 Undirected networks do not, that means the goods can be transported along a line in both directions.

 Connectivity functions
 Network analysis analyzes the way ‘goods’ can be transported along these lines.
 Types of networks (network can be)
 Directed network (transportation in one line) and undirected network
 Planar network and non-planar Network
 Two types of operations:
 Optimal path finding
 Ordered
 Unordered
 Network partitioning
 Network allocation
 Trace analysis Compiled by Nigatu W. HU-IoT-BEE-July, 2016 45

 Planar and non-planar networks  NON-PLANAR NETWORKS
 PLANAR NETWORKS  These are multi-level crossings
 Planar is that can be embedded in a two- with underpasses and overpasses
dimensional plane.  Examples are a large-scale traffic
 Examples are stream/river networks network
www.brombeer.net700 × 525
Source:ITC & Principles of GIS www.skyscrapercity.com1107 × 642


SPATIAL DATA ANALYSIS ........ Modified2

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

SPATIAL DATA ANALYSIS ........ Modified2

Uploaded by

Copyright:

Available Formats

V.

SPATIAL DATA ANALYSIS

Compiled by Nigatu W. HU-IoT-BEE-July, 2016 1

 Therefore, here Analysis = analytical operations

Source: Ellen-Wien Augustijn & Principles of GIS

Compiled by Nigatu W. HU-IoT-BEE-July, 2016 3

 There are many ways to classify the analytical functions of a GIS.

3. Neighbourhood functions, and

 Geometric measurement on spatial features include (excluding attribute data measurement

Ex. Finding sum is attribute data measurement, not

 The primitives of vector data sets are point, (poly)line and

Source: Ellen-Wien Augustijn & Principles of GIS

Compiled by Nigatu W. HU-IoT-BEE-July, 2016 7

Source: Principles of GIS

 Raster measurements include: location, distance and area size.

Source: Ellen-Wien Augustijn & Principles of GIS

Compiled by Nigatu W. HU-IoT-BEE-July, 2016 11

𝑧= 302 + 302 = 2(302 )

Source: Ellen-Wien Augustijn & Principles of GIS

Compiled by Nigatu W. HU-IoT-BEE-July, 2016 13

Compiled by Nigatu W. HU-IoT-BEE-July, 2016 14

 What is at …..? What is at a specific location?

 selection in which you select features by

 drawing a graphic, to select all objects

Source: Principles of GIS

Compiled by Nigatu W. HU-IoT-BEE-July, 2016 15

Syntax – logical connectives

Composite conditions use logical connectives:

 Area < 400,000 (atomic condition)

Compiled by Nigatu W. HU-IoT-BEE-July, 2016 19

 Meet relationship (adjacent): i. e. selecting features adjacent to selection objects.

 Example: Data retrieval with

Select all clinics

6.2 7.4 8.0

5.8 6.3 7.2

 This is normally done via a classification table.

CODE OLD TYPE NEW TYPE

Source: Principles of GIS

Compiled by Nigatu W. HU-IoT-BEE-July, 2016 28

 Vector overlay functions ISPIT

 Intersection: The potato fields on clay soils.

 Union: The fields where potato or maize is the crop.

 Difference: The potato fields not on clay soils.

 Complement: The fields that do not have potato as crop.

Source: Ellen-Wien Augustijn & Principles of GIS

Compiled by Nigatu W. HU-IoT-BEE-July, 2016 29

 Overlay operations - Intersect

Source: Ellen-Wien Augustijn & Principles of GIS

Compiled by Nigatu W. HU-IoT-BEE-July, 2016 30

 Overlay operations - Clip

Compiled by Nigatu W. HU-IoT-BEE-July, 2016 31

 Raster Overlay operations

 GIS raster processing have a language to express operations on rasters.

Source: Principles of GIS

Compiled by Nigatu W. HU-IoT-BEE-July, 2016 32

 Various arithmetic operators are supported. The standard ones include:

Compiled by Nigatu W. HU-IoT-BEE-July, 2016 33

 The following are some simple map algebra assignments.

Source: Ellen-Wien Augustijn & Principles of GIS

Compiled by Nigatu W. HU-IoT-BEE-July, 2016 34

There are also other raster overlay operators, such as:

Please refer to these operators in the reference and text books