Morgan, Craig John2006-11-162006-11-162006-11-16http://hdl.handle.net/10539/1768Faculty of Science School of Statistics snd Acturial Science 9907894x craig.morgan@goldfields.co.zaThis dissertation presents a detailed study of geostatistics. Included in this work are details of the development of geostatistics and its usefulness both in and outside of the mining industry, a comprehensive presentation of the theory of geostatistics, and a discussion of the application of this theory to practical situations. A published debate over the validity of geostatistics is also examined. The ultimate goal of this dissertation is to provide a thorough investigation of geostatistics from both a theoretical and a practical perspective. The theory presented in this dissertation is thus tested on various spatial data sets, and from these tests it is concluded that geostatistics can be effectively used in practice provided that the practitioner fully understands the theory of geostatistics and the spatial data being analyzed. A particularly interesting conclusion to come out of this dissertation is the importance of using additive regionalized variables in all geostatistical analyses.35328 bytes2285585 bytes91614 bytes497321 bytes3033248 bytes1390221 bytes721981 bytes1059171 bytes626543 bytes685731 bytes406319 bytes440340 bytes1230601 bytesapplication/vnd.ms-excelapplication/pdfapplication/pdfapplication/pdfapplication/pdfapplication/pdfapplication/pdfapplication/pdfapplication/pdfapplication/pdfapplication/pdfapplication/pdfapplication/pdfenGeostatisticsKrigingConcept of supportAdditivityStationarityThesis