1. Introduction to Spatial Statistical Analysis

2. Spatial Statistics
The GIS dictionary (Wade and Sommer, 2006) define spatial statistics as “the field of study concerning statistical methods that use space and spatial relationships (such as distance, area, volume, length, height, orientation, centrality and/or other spatial characteristics of data) directly in their mathematical computations.

3. Types of Spatial Data
Event
Fixed Location
Exhaustive

4. Event Data Something has occurred in an area
Unique attribute is its spatial location
Examples:
Crimes in a city; Accidents on the roads
Tree growing in an area
An animal stopping to graze (or move through)
Volcano (eruptions or locations)
May have associated attribute data (continuous or discrete)
Analysed through spatial point patterns

5. Event Data figure

6. Fixed Location Result from a designed sample survey
One or more variables sampled at a number of fixed locations (defined by their spatial co-ordinates)
Sampling may be systematic or random in space
The measured variable may be continuous or discrete
Examples include:
Weather stations
Measuring of/for pollutants or their concentrations
Ore grade or soil samples

7. Fixed location example

8. Exhaustive
Contains value for the variable of interest for the entire study area
Can be continuous or discrete, and represented by polygonal thematic map, or raster
Examples include:
DEM/DTM
Aerial photos/Satellite images
Geological Map, Census data
Area data are considered exhaustive. Spatial data that is aggregated to area units (as in census data)

9. Exhaustive example

10. Spatial Analysis Types
Point Pattern and Kernel density
Geostatistics and Interpolation methods

11. Focus on Event Data
Point Pattern Analysis

12. Point Pattern Analysis
Objectives of point pattern analysis are:
Estimate intensity of point pattern
Expected number of events per unit area
Enables visualisation of the events
Study structural characteristics of the spatial distribution
Determine if the events are random, clustered or regular
Indicative of a process that is underlying the occurrence of events

13. Point Pattern Analysis
Is the study of the spatial arrangements of points in space.
Is the evaluation of the pattern, or distribution, of a set of points on a surface.
It can refer to the actual spatial or temporal location of these points or also include data from point sources.

14. Analysing Point Patterns
Global and local study
Looking for outliers & clusters
Global – study properties for the entire dataset
Spatial statistic -> Avg nearest neighbour, Getis Ord,
Local – looking at behaviour related to neighbours
Spatial Visual -> Voroni maps (Outliers, Smoothing, Variation)
Spatial statistic -> Local Moran’s I (Anselin)

15. Global

16. 1. Average nearest Neighbour
Calculates a nearest neighbour index based on the average distance from each feature to its nearest neighbouring feature.

17. 1. Average nearest Neighbour

18. 1. Average nearest Neighbour

19. 1. Average nearest Neighbour

20. 1. Average Nearest Neighbour
Example 1. 5 10
+ 3
+ 3
+ 4
+ 8
+ 5
+ 15
Do = 7 8m 6
20 m 3 5m
De = ∗30 3m 4 4m
15m 2
10m 7 1
30 m

21. 2. Getis Ord- Hot Spot Analysis
This tool identifies statistically significant spatial clusters of high values (hot spots) and low values (cold spots).

22. 2. Getis Ord- Hot Spot Analysis
This tool works by looking at each feature within the context of neighboring features.
A feature with a high value is interesting, but may not be a statistically significant hot spot.
To be a statistically significant hot spot, a feature will have a high value and be surrounded by other features with high values as well.
The local sum for a feature and its neighbors is compared proportionally to the sum of all features;
When the local sum is much different than the expected local sum, and that difference is too large to be the result of random chance, a statistically significant Z score results.

23. 2. Getis Ord- Hot Spot Analysis

24. 2. Getis Ord- Hot Spot Analysis
Example 2. Manual 6 3
i=3
i=2
1 m
2 m
1.5 m 4
i=4 1
G1 = ∗ ∗ ∗3 −4 ( ) 𝑆 − ( ) 2 3
i=1

25. 2. Getis Ord- Hot Spot Analysis
Potential Applications Applications can be found in crime analysis, epidemiology, voting pattern analysis, economic geography, retail analysis, traffic incident analysis, and demographics

26. Getis Ord

27. Local

28. Voroni maps Thiessen (Voronoi) polygons
Polygons are created by joining together the sample points and drawing the perpendicular bisectors of the joining segments. The intersections of these bisectors determine the sides and the number of polygons, so that the perimeters of the generated polygons are equidistant from designated points and the area of influence.
Initially Thiessen polygons were used for meteorological data analysis (rainfall stations), but nowadays they are also applied as GIS analytical tool in studies to determine areas of influence (hospitals, fire stations, subway stations, shopping malls, air traffic control, mobile, analysis of populations of plant species, etc.).

29. Voronoi Maps
1. Link a point to its neighbouring points by straight lines.
2. Construct the perpendicular in the middle of each line.
3. The polygons defined by the intersection of this perpendicular are the Thiessen polygons.

30. Voronoi Maps

31. Local Morans Anselin
Given a set of weighted features, the Cluster and Outlier Analysis tool identifies clusters of features with values similar in magnitude. The tool also identifies spatial outliers. To do this, the tool calculates a Local Moran's I value, a Z score, a p-value, and a code representing the cluster type for each feature. The Z score and p-value represent the statistical significance of the computed index value.

32. Local Morans Anselin

33. Local Morans Anselin Example 2. Manual6 3
i=3
i=2
1 m
2 m
1.5 m 2
i=4 1
i=1
G1 = 1−3 𝑆 2 ( − − ( 2−3) )

34. Local Morans Anselin