In this chapter, we learned how to test for autocorrelation in spatial data using Moran's I index. We also learned how to model autocorrelation using a SAR model. This was followed by modeling count data using a Poisson GLM. After that, we learned the basics of geostatistics, with variograms and kriging, in particular. Then, we learned how to understand the degree of spatial dependence in a random field using variograms. We learned about the different ways of plotting and modeling spatial dependence using the geoR and gstat packages. We then learned how to predict or interpolate data from covariograms using a method called kriging, and we learned how to compute exceedance probability and looked at checking the residuals of a prediction.
In the upcoming chapters, we'll learn how to automate different spatial processing and analysis tasks. We'll then learn...
