Spatial Manipulation and Analysis

1
Lecture 5

• Spatial Manipulation and Analysis

2
Last week
Working with tables (attribute data)

• Viewing, editing, summarising attribute data.
• Joins and relates
• Geocoding

How can I enhance the appearance of my map?
Tables, querying, displaying, presenting
Adding map elements, labels, graphs etc
How do I select data to match certain criteria?
What makes a good map ?
Select Features, Select by Attribute, Select by
Location
Thematic mapping, projections, Layouts
3
Overview of Weeks 2-6 of GEOG5040
Week 2 A first look at ArcGIS and
ArcInfo Workstation
Week 6 Rasters Surfaces
Weeks 7-11 Adapting ArcGIS (with Dr Andy Evans)
USING ARCGIS
Week 3 Data in, out and shake it all about!
Week 5 Spatial Manipulation Analysis
Week 4 Tables, Querying, Displaying and
Presentation
4
This Lecture

• Spatial manipulation
• Combining data.
• Append
• Dissolve and Eliminate
• Spatial Analysis
• Overlay functions
• Clip, Erase, Identity, Intersect and Union
• Proximity functions
• Near, Pointdistance and Buffer
• Network Analysis.

5
Week 5 Spatial Manipulation and Analysis
Spatial manipulation
Spatial manipulation
Network Analysis
Spatial Manipulation Analysis
Spatial Manipulation Analysis
Spatial statistics
Overlay Functions
Overlay Functions
Proximity Functions
6
Aims and Objectives of Week 5

• The aims of this lecture/practical are to
  introduce you to
• The spatial analysis functions of ArcToolbox
• Network Analysis
• At the end of this lecture/practical you should
  be able
• Operate the suite of ArcToolbox tools for spatial
  analysis
• Have tested your geodatabase network (built in
  practical 3) to run a flow and trace task.

7
ArcToolbox and Spatial Analysis/Data Management

• Most data manipulation and analysis functions we
  have looked at so far have been accessed through
  the toolbars in ArcCatalog and ArcMap
• Now looking at the operations that are available
  through ArcToolbox
• Require relativley few and straightforward inputs
• Wizards make them easy to use.

8
Selecting the right toolCoverages vs Feature
Classes

• ArcToolbox has Tools for Coverages and Tools for
  Geodatabases/Feature Classes
• Need to choose correct Tool
• Get to know your way around ArcToolbox
• Analysis Tools will work for both models in
  general
• However, a Tool is not always available in one of
  the data models

9
Spatial Manipulation - Combining Data

• It is likely that before analysis, youll have to
  combine several datasets.
• Weve looked at combining data that is about the
  same place. You also may need to combine
  geographical data that are adjacent into a single
  dataset.
• Having it in one dataset will allow us to analyse
  it.

10
Append - Coverages

• Makes a new Coverage by joining two or more
  Coverages usually next to each other.
• If the Feature types and Attribute Tables are the
  same, these joined too.
• Linked Polygon edges need same codes.
• First pick Coverages.

11
Append Coverage Wizard
Choose append Method
Decide how tics and coverage features will be
numbered
12
Append Geodatabase Wizard
13
Dissolve

• Merges features that share a value for an
  Attribute so there are no internal boundaries.
• E.g. dissolve on shared class in Attributes.

A
A
C
B
C
A
B
B
C
14
Dissolve in ArcToolbox

• Data Management gt Generalization gt Dissolve.

15
Eliminate

• ArcToolbox tools (under Data Management gt
  Generalization) that make new Feature Classes
  from old by combining and deleting.
• Merges Polygons based on either the longest
  shared Arc or largest Area.
• Used to eliminate slivers e.g. merge all small
  Polygons with their neighbours with a larger
  area.

16
Eliminate

• Options to
• Ensure outer borders of Coverages/Feature classes
  kept.

17
Spatial Analysis

• Answer such questions as
• If smallpox has a spread rate of 10 km per day
  from a postal sorting office into the surrounding
  population, how many people will be infected in 3
  weeks?
• If we close all farms within 2km of a
  foot-and-mouth infected farm, which farms do we
  close?
• If we need to remove a 100m wide strip of trees
  to make a motorway, how much profit can we make
  by selling the wood associated with 10km of new
  road?
• All these questions can be answered by overlaying
  and cutting out bits of datasets.

18
Analysis Tools in ArcToolbox
19
Extract Tools

• Analysis Tools gt Extract
• Clip keeps Features falling within clipping
  Coverage Polygons opposite of Erase (an Overlay
  tool).
• Select
• Split used for splitting Features into their own
  Coverages/Feature classes

20
Extract Tools Clip

• Only keeps the features inside the clip feature.

21
Extract Tools Split

• Clips the input Features/coverages and stores
  them in multiple output feature classes/output
  coverages

22
Overlay

• Combination of attributes
• e.g. a property developer may be interested in
  land NOT in flood zone but for sale
• Overlay Take two layers and create a third
  which meet certain criteria
• One layer must be a polygon feature
  class/coverage
• Other layer can be a polygon, line or point
  feature class
• Pre-GIS it would have been done using transparent
  maps and a light table

23
Overlay tools - Erase
Original Coverage Erase Coverage Final
Coverage

• Cuts out and removes areas of Features that fall
  inside Polygons in a erase Coverage.

24
ArcToolbox Erase
25
Other Overlay Tools

• Combine two Coverages/feature classes, deleting
  different sections.
• All overlay tools have similar dialogs to Erase.
• Attributes usually combined and repeated.

A
A
A
1
2
A1
A2
B2
B1
B
B4
B3
3
4
A3
A4
26
Overlays

• Types

Output
Input
Original
Identity (Input determines output area)
27
Proximity

• Nearness in terms of physical distance
• E.g. House buying close to nice things such as
  friends, shops, countryside but long way from
  unpleasant things such as a smelly factory
• E.g. Power Station may need to be close to
  infrastructure (roads etc) and water but far away
  from protected natural beauty spots

28
Proximity in GIS

• GIS implements proximity using buffers
• A set distance around a point (circle)
• A set distance on one side or both sides of a
  line (polygon)
• ArcGIS creates a buffer and then allocates a
  field within the attribute which defines its
  status (inside or outside the buffer)

29
ArcToolbox Proximity Tools

• Find or delineate Features based on distance from
  others.
• Analysis Tools gt Proximity

30
Buffer

• Puts an area around a given Feature, some
  distance out from it.
• Multiple ring buffer does several bands

31
Overlaid and Buffered Attributes

• All the Overlay and Buffer Tools have different
  ways of dealing with attributes.
• For example, if we split a Polygons Area, do we
  rescale the Attribute data in each or not? Do we
  even keep the data?
• Need to think carefully about whether attributes
  have been altered.
• Do you need to import the old area for a polygon
  and rescale the data associated with new Polygons
  by, e.g. assuming its evenly spread
  geographically.

32
Create Thiessen Polygons

• Makes Thiessen Polygons, ignoring/ditching Points
  Proximal Tolerance or less apart.

33
Other Proximity Tools

• Near finds the distance between a Point in one
  Feature Class and the nearest Point in another,
  as long as they are below Search Radius apart.
• PointDistance finds the distance between all
  Points in one Feature Class and all Points in
  another can produce very large Tables.

34
Statistics

• In ArcToolbox Analysis gt Statistics
• Analyses Attribute Tables and puts results in
  another.
• Frequency,
• Summary Statistics Means, Standard Deviation etc.

35
Spatial Statistics

• Spatial Statistics tools analyse the geographic
  distribution of data features across a map
  surface and in relation to one another
• Intended for application to point data sources

36
Analysing patterns submenu

• Average Nearest Neighbour
• Manhattan distance
• Euclidean distance

Analysing patterns gt Average Nearest Neighbour.
Source ESRI Help
37
Other Spatial Statistics tools

• Mapping Clusters
• E.g.Hotspot Analysis
• Measuring Geographic Distributions
• E.g. Central Feature
• Directional Distribution
• Linear Directional Mean
• Mean Centre
• Standard Distance
• Utilities
• E.g. Calculate Areas
• Collect Events

38
Combining Tools Spatial Analysis

• Using the tools in combination allows us to
  select and amalgamate data.
• This is the basis of GIS Spatial Analysis.

39
Forest loss

• If we fell a 100m width of forest to build a new
  motorway, how much wood do we lose?

• We have data on the motorway route and forests.
• Buffer 50m around motorway.
• Convert buffer to Polygon Feature Class.
• Clip forests from the forest Feature Class using
  the new buffer Polygons.
• Aggregate the new clipped Polygons AREAs.

40
Hospital markets

• Were evil NHS destroyers. We have layers of
  different affluence areas and a Point layer of
  hospitals. We want to find out the levels of
  affluence and poverty near our hospitals so we
  can convert some to private health care.

• Make Thiessen Polygons around hospitals to see
  which one people in an area go to.
• Use Intersect tool to combine the new Polygons
  with the affluence Polygons.
• This makes a new dataset each Polygon having its
  affluence and a hospital reference.
• Use Statistics tools to summarise affluence for
  each hospital.

41
Activity Spatial Analysis using ArcToolbox

• Follow the instructions on the activity sheet to
  experiment with the Dissolve tool, Proximity tool
  and Clip tool

42
Network Analysis

• Whats the fastest way to
• A point on an electronic circuit.
• How much will flow down
• A road or sewage pipe.
• Where is there a blockage
• in a motorway system.
• Whats downstream from
• a pollution point in a braided river system.

43
Things We Can Model As Networks

• Transport Systems
• Roads, rivers, hill slopes.
• Utility Networks
• Sewage, water, electricity, gas.
• Knowledge Networks
• The spread of information, prices, policy effects.

44
Networks in ArcGIS

• Stored as Geodatabases, therefore Topology built
  during editing, and constraints added.
• We can guarantee a Network Object will only
  respond in a certain way (e.g. water pressure in
  a pipe will be under some limit).
• Stored in a special, constricted Feature Dataset
  element the Geometric Network.
• Consists of edge and junction elements which must
  be used to link each other up i.e. a edge cant
  connect directly to another edge, it must go
  through a junction.

Geometric Network
Geodatabase
Feature Class
Features
Feature Dataset
45
Classes

• Junction Feature Classes inherit from
• ESRI Simple Junction Feature.
• Edge Feature Classes inherit from
• ESRI Simple Edge Feature or
• ESRI Complex Edge Feature.
• Complex Edge Features are groups of Features that
  act together.
• Network Features are like other Features, only
  they are assigned to a Geometric Network Object.

46
Making Geometric Network Objects

• Right-click the Feature Dataset gt New gt Geometric
  Network wizard.

47
Weights

• The Wizard allows you to say that some elements
  will have Weights attached.
• Weights control the flow rate through an edge or
  junction, e.g. stream width, or maximum traffic
  speed.
• You dont need to tell it the Feature class and
  attribute to use, just a name.
• You use the name to set an Attribute to a weight
  when you make the Feature Class.
• This way you can have the same weight name for
  different Attributes depending on the Feature,
  e.g. Maximum Flow can be used for Stream
  discharge and Traffic movement.

48
Setting up Weights

• Integer is Long Integer elsewhere.

49
Making Network Feature Classes

• Slight differences from other Feature Classes.
• Pick a ESRI Network Class to inherit (usually a
  Simple one)
• Pick the Network to add to (needs to be planned
  and made first).

50
Sources and Sinks

• You can set a Junction to be a net Source or
  Sink, i.e. where materials enter or leave the
  system.
• This generates a flow direction for each Feature.
• This is used in combination with weights to
  determine flows.

51
Setting Weights

• You get the option to set the Attributes to use
  as particular Weights.

52
Table of Contents

• You should end up with something like the
  following.

Network
ArcDesktop will make default Junction Classes for
you.
But you can make your own.
53
Setting Connectivity Constraints

• You can constrain which lines and junctions can
  connect in the Geometric Networks Properties.

54
Adding Network Features

• You can add elements in ArcMap as usual, but you
  need to remember to add junction elements between
  edges.

• Source/Sink elements can have their AncillaryRole
  Attribute set to source/sink/neither. All
  Features also get an ENABLED Attribute which
  determines if they take part in the Network.

55
Network Analysis

• Add the Utility Network Analysis toolbar. (Viewgt
  Toolbars)
• This gives you the ability to put in start and
  end flags for flow analysis.
• Also lets you block flows with flags.
• In the Analysis gt Options you can set the
  analyses to use Selected or unselected Features.
• You can also control flows by enabling and
  disabling Features in their Attributes.

56
Flow and Tracing

• You can see the net result of sinks, sources and
  weights by putting on flow arrows. Flows can be
  indeterminate and uninitialized as well as
  determinate.
• However, the Trace Task list allows you to do
  more complex analysis between start and end
  flags.

57
Trace Tasks

• Find Common Ancestors
• Common upstream points of flags.
• Find Connected
• Maps edges/junctions that can be reached from
  flags.
• Find Loops
• Finds Network loops.
• Find Path
• Finds the shortest path by length or the weights
  set in Options.
• Trace Downstream / Trace Upstream
• Maps areas Down/Upstream of flags.
• Can use blocking flags etc. to prevent some areas
  being accessed.

58
Solve and Clear Flags

• To set an analysis running, press the Solve icon.
• To clear flags, select the menu item under
  Analysis.

Solve
59
Example Find Path

• Find the quickest route from a water entering a
  glacier to an exit point on the forefield, given
  certain stream diameters etc.

60
Example Trace Upstream

• Find out where pollutants could be entering a
  stream system.

61
Other Examples

• Find out the maximum flow points on a Network.
• Work out where a Network is damaged.
• Calculate the effect of a Network blockage.
• Calculate the potential spread of pollutants.
• Calculate the shortest distance between two
  cities on a road network.

62
Summary

• To use Features in a Network, you have to build
  the Geometric Network.
• You can set names for the weights to be used with
  the Network.
• You can then add Feature Classes to it and assign
  the weights to Attributes.
• Features should inherit from ESRIs Simple or
  Complex Edge or Junction Classes.
• Junctions can be set as Source / Sink Objects,
  and Edges and Junctions can be given weights.

63
Summary

• You add Network Features in ArcMap as usual, but
  edges should be linked by junctions.
• You can enable and disable Features and set
  Junctions as Sources / Sinks to determine the
  flows around the Network.
• You can use the flows and links to do analysis.

64
Activity 2

• Use the sewer network you built in Practical 3 to
  test out the flows and trace tasks.
• http//www.geog.leeds.ac.uk/courses/postgrad/geog5
  040/lecture5/Act5_08/

65
Week 5 Summary

• Combining data
• Append
• Dissolve and eliminate

Spatial manipulation
Network Analysis

• Building a network
• Flow and trace analysis

Spatial Manipulation Analysis
Spatial statistics
Overlay Functions

• Analysing patterns
• Mapping clusters
• Measuring geographic distributions
• Utilities

Proximity Functions

• Clip
• Erase
• Identity, Intersect, Union

• Buffer
• Near, Pointdistance

66
Next Lecture

• Raster and 3D Data
• Practical - Return to Soapland! Use Spatial
  Analysis tools to quantify the terrible effects
  of an explosion in Albert Square