Lesson 12 Flashcards
- The checking if the location of the points tends to cluster together, whether they’re more uniformly distributed or arranged randomly
- It provides fundamental clues about the underlying spatial processes and relationships
Point Pattern Analysis
If the point pattern is identified as nonrandom, it can be further classified as more clustered than random or more
dispersed than random. (T or F)
True real omsim
Results from a contagion process where one location attracts a number of points. Points are close to one another and there are large patches that does not contain any points.
Clustered Point Pattern
Commonly results from some form of competition in space, where points repel one another. The dispersed patterns are seemingly regular over space; hence, it is called uniform.
Uniform Point Pattern / Dispersed Point Pattern
Happens when points repel one another
Negative Autocorrelation
Generated by an independent random process, wherein every location (or small area) of a study area has an equal probability of receiving an event or point, and for which the location of an event is independent of the location of all other events. In other words, the underlying generating process has no spatial logic.
Random Point Pattern
Often engaged in examining departures from complete spatial randomness.
Point Pattern Analysis
The distribution of points throughout a given study region follows a homogenous Poisson process
Complete Spatial Randomness (CSR)
Properties of CSR
- Each event has an equal probability of occurring at any position in the study region
- The position of any event is independent of the position of any other.
CSR Challenge: When examining absolute location, this is the variation of the observations’ density across a study area (no equal probability).
First-order property of pattern
CSR Challenge: When exploring interactions between locations, this is the observations’ influence on one another (no-independence).
Second-order property of pattern
Point Pattern Technique: Characterize the pattern in terms of its distribution vis-a-vis the study area (takes the first-order property).
Density Methods
Point Pattern Technique: The interest lies in how the points are distributed relative to one another (takes the second-order property)
Distance-based Methods
Density Method that measures a pattern’s overall density. Ratio of observed number of points, 𝑛 to the study region’s surface area, 𝑎
Global density
Density Method: A point pattern’s density measured at different locations within the study area
Local density
This is an example of a local density method that focuses on changes in the density of points across a study region i.e. the frequency of points occurring in various parts of the area. It is performed by overlaying a regular grid on the region of interest and then counting the number of points found in each quadrat (cell) of the grid.
Quadrat Analysis
Limitations of the Quadrat Analysis
- It is a measure of dispersion rather than pattern (based primarily on the density of points)
- Results in a single measure for the entire distribution, variations within the region are not recognized. Thus, markedly different point patterns can give rise to identical frequency distributions of the points per quadrat
- Sensitive to grid size—changing the size of the quadrats may also affect the significance
This method provides an alternative to quadrat analysis. It focuses on the distances between points rather than the density of points in an area. However, this statistic cannot be used to compare point pattern maps.
Nearest Neighbor Analysis
A way to understand the geographical pattern or spatial order of phenomena across space.
Point Pattern Analysis
Three Types of Point Patterns
Clustered, Dispersed, and Random.
In theory, the distribution of geographical points follows the core principles of ___?
Complete Spatial Randomness (CSR).
An example of a local density-based method
Quadrat Analysis
An example of a distance-based method
Nearest Neighbor Analysis
Quadrat Method Results Interpretation:
VMR > 1 to Infinity indicates ?
Indicates a clustered point pattern (a good
deal of variation in the number of points per cell)
Quadrat Method Results Interpretation:
VMR = 1 indicates ?
Indicates a random point pattern
Quadrat Method Results Interpretation:
VMR < 1 to 0 indicates ?
Indicates a uniform or dispersed point pattern (little variation of points from cell to cell)
Nearest Neighbor Analysis Results Interpretation:
R Values close to 0 indicates a ?
Clustered point pattern
Nearest Neighbor Analysis Results Interpretation:
R Values close to 2.15 indicates a ?
Dispersed point pattern
Nearest Neighbor Analysis Results Interpretation:
R = 1 indicates ?
Indicates consistency with a random
pattern
