Date of Award
Master of Science
Faculty of Science, Technology and Engineering
Dr Lyn Bloom
This research is in the area of geostatistics and consists essentially of two parts. The first is an investigation of the variogram and cross variogram and the associated kriging and cokriging methods of spatial prediction and the second is an application of these in the analysis of two (original) data sets. In the first part (chapter 1 to chapter 5), the focus is on summarising and illustrating the various techniques of Exploratory Data Analysis (EDA) and some methods used to estimate and model the experimental variograms and cross variograms for a given data set, together with some of the geostatistical methods of kriging and cokriging used for prediction purposes. The research also illustrates some of the many applications of this theory in the earth and environmental sciences. The second part of the thesis (chapter 6) is an application of these geostatistical techniques to the analysis of two (new) data sets. The first data set consists of the Available Phosphate (in ppm) and Potassium (in ppm) from two fields (one cropped and one uncropped) in the Jimperding Brook area of Western Australia. The second data set consists of the number of species of Banksia at various locations within a region of Southwestern Australia.
Tripathi, S. (1996). Variograms and kriging in the analysis of spatial data. Retrieved from http://ro.ecu.edu.au/theses/970