- Information
- AI Chat
Spatial Analysis - Lecture notes 8
Geographical Perspectives (ENV-4010Y)
University of East Anglia
Recommended for you
Preview text
Geographical Perspectives – ENV-4010Y
Spatial Analysis
Spatial analysis is concerned with “... investigating the patterns that arise as a result of processes that may be operating in space.” (O’Sullivan and Unwin, 2010, p. 4).
Elements of Spatial Patterns
Spatial analysis involves:
Representing patterns (e. through visualisation)
Describing patterns (e. by identifying or summarising trends)
Predicting patterns (e. from known criteria or understanding embodied in a model)
Explaining patterns (e. through comparison of observations with hypothesised outcomes from a model)
Maps Represent Patterns
Spatial Data are Special
Two
important geographical considerations are spatial heterogeneity and spatial dependence.
Challenges that arise when using spatial data include:
- Spatial autocorrelation
- Modifiable Areal Unit Problem (MAUP)
- Ecological fallacy
IDW interpolated surface example:
Pollution across Paris
Kernel Density Estimation (discrete data)
“Many of the standard techniques and methods documented in statistics textbooks are found to have significant problems when we try to apply them to the analysis of spatial distributions” (O’Sullivan and Unwin, 2010, p. 34).
For what reasons are spatial data special?
Conventional statistical analysis imposes conditions or assumptions on the data it uses and spatial data almost always violate the requirement of a random sample.
Modifiable Areal Unit Problem
If the spatial units in a study were specified differently, we might see different patterns.
Patterns can change markedly according to the degree of aggregation or zoning used.
Spatial Analysis - Lecture notes 8
Module: Geographical Perspectives (ENV-4010Y)
University: University of East Anglia
