Spatial Data Analysis in Ecology and Agriculture Using R

Working with Spatial Data
Analysis of Spatial Data
Data Sets Analyzed in This Book

R Programming Environment
R Basics
Programming Concepts
Handling Data in R
Writing Functions in R
Graphics in R
Other Software Packages

Statistical Properties of Spatially Autocorrelated Data
Components of a Spatial Random Process
Monte Carlo Simulation
Review of Hypothesis and Significance Testing
Modeling Spatial Autocorrelation
Application to Field Data

Measures of Spatial Autocorrelation
Preliminary Considerations
Join-Count Statistics
Moran’s I and Geary’s c
Measures of Autocorrelation Structure
Measuring Autocorrelation of Spatially Continuous Data

Sampling and Data Collection
Preliminary Considerations
Developing the Sampling Patterns
Methods for Variogram Estimation
Estimating the Sample Size
Sampling for Thematic Mapping
Design-Based and Model-Based Sampling

Preparing Spatial Data for Analysis
Quality of Attribute Data
Spatial Interpolation Procedures
Spatial Rectification and Alignment of Data

Preliminary Exploration of Spatial Data
Data Set 1
Data Set 2
Data Set 3
Data Set 4

Multivariate Methods for Spatial Data Exploration
Principal Components Analysis
Classification and Regression Trees (aka Recursive Partitioning)
Random Forest

Spatial Data Exploration via Multiple Regression
Multiple Linear Regression
Building a Multiple Regression Model for Field 4.1
Generalized Linear Models

Variance Estimation, the Effective Sample Size, and the Bootstrap
Bootstrap Estimation of the Standard Error
Bootstrapping Time Series Data
Bootstrapping Spatial Data
Application to the EM38 Data

Measures of Bivariate Association between Two Spatial Variables
Estimating and Testing the Correlation Coefficient
Contingency Tables
Mantel and Partial Mantel Statistics
Modifiable Areal Unit Problem and Ecological Fallacy

Mixed Model
Basic Properties of the Mixed Model
Application to Data Set 3
Incorporating Spatial Autocorrelation
Generalized Least Squares
Spatial Logistic Regression

Regression Models for Spatially Autocorrelated Data
Detecting Spatial Autocorrelation in a Regression Model
Models for Spatial Processes
Determining the Appropriate Regression Model
Fitting the Spatial Lag and Spatial Error Models
Conditional Autoregressive Model
Application of SAR and CAR Models to Field Data
Autologistic Model for Binary Data

Bayesian Analysis of Spatially Autocorrelated Data
Markov Chain Monte Carlo Methods
Introduction to WinBUGS
Hierarchical Models
Incorporation of Spatial Effects

Analysis of Spatiotemporal Data
Spatiotemporal Cluster Analysis
Factors Underlying Spatiotemporal Yield Clusters
Bayesian Spatiotemporal Analysis
Other Approaches to Spatiotemporal Modeling

Analysis of Data from Controlled Experiments
Classical Analysis of Variance
Comparison of Methods
Pseudoreplicated Data and the Effective Sample Size

Assembling Conclusions
Data Set 1
Data Set 2
Data Set 3
Data Set 4
Conclusions

Appendices
Review of Mathematical Concepts
The Data Sets
An R Thesaurus
References
Index