Week 9: Point Pattern Analysis Flashcards
How do we work out if there is a higher probability of points occurring in one area compared to another?
Carrying out quadrat analysis
How do we determine a random distribution using quadrat analysis?
If the average number of points per cell is roughly equal to the variance of the number of points per cell.
How do we determine a clustered distribution using quadrat analysis?
If there is a large amount of variability.
List some limitations of Ripley’s K(d).
- Dependent on study boundary area
- Circles go outside study area
- Each point has a unit value, there’s no magnitude or quantity
What is nearest neighbour analysis?
Measures the distance of each points to its nearest neighbour, determines the mean distances between neighbours and compares the mean distance to what would have been expected in a random distribution.
What are 3 ways of defining distance?
- Direct (standard)
- Indirect (linear)
- Based on a network
Why is Kernel Density Estimation (KDE) useful?
- Generalises data over larger regions
- Good method to show estimation
- Provides a risk surface more than an actual picture of what is occurring.
How does KDE work?
- Every point has a density estimate, and when you overlay a grid on top of a map, you transfer those density estimates on top of the grid.
- Measures the distance between each grid cell centerpoint and each incident data point and determines the cell weight for that point
List 3 factors that the weight of each cell depends on.
- Distance fro the grid cell centerpoint to the incident data point
- Size of the radius around each incident data point
- Method of interpolation
What is bandwidth?
Refers to the size of the cone of the kernel and is specified by the user.
What does a bell curve shaped bandwidth represent?
- peaks and declines rapidly with no defined radius
- Continues across entire grid
What does a uniform (flat) distribution bandwidth represent?
- All points in radius equal
- Represented by cylinder
