Geostatistics for Engineers and Earth Scientists

Rating:
Summary: A fatally flawed variant of classical statistics
Review: Geostatistics is a fatally flawed variant of classical statistics because it violates the requirement of functional independence and ignores the concept of degrees of freedom. Without degrees of freedom unbiased statistical inferences are impossible to obtain.

Rating:
Summary: Good reference for Geostat
Review: I strongly recommend this book for reference and self study. It contains tons of useful definitions, theorems,lemmas, etc.., all with proofs. The algorithms are presented in a very clear and detailed way. The assumptions are stated clearly. It helps a lot with the math and derivations.

Rating:
Summary: From "Stochastic Environmental Research and Risk Assessment," 14(3)207-209
Review: One could wonder how a new textbook in geostatistics would have a chance to offer something new and fresh to the reader, when there are so many of them already available on the market. The answer is contained in a sentence that could not be better and more clearly summarized than by the author himself, in the introductory chapter: "Derivations in mathematical books are unnecessarily general, resulting in a complexity that tends to intimidate the uninitiated, while explanations in geostatistical books are cryptically short--to the point that most readers end up taking the derivations for granted."

This single sentence is also the philosophy of this book. By keeping a constant link and perfect balance between theoretical and methodological concerns, the book allows the novice reader to discover step by step the exciting realm of geostatistics and its applications. As explained by the author, this textbook is the result of a long teaching experience to engineering students at the graduate level. After reading this book, I have to say that I wish I could have had the chance to be one of his students when I was struggling myself with theory as a Ph.D. student a few years ago.

A requirement for reading the book is a college-level knowledge in matrix algebra, probability, and statistics. I have found that in spite of its obviousness, this clarification is sometimes lacking in other introductory books, which tend to present the material in such a way the reader could wrongly believe that he or she will be able to grasp the basics and even the more advanced results without the necessity of having a basic command of the disciplines (probability and statistics) that underlie geostatistics. The author also emphasizes that "many of the suggestions of my colleagues were incorporated into the final text, but I reserve the right, and assume the blame, for disagreeing in a few instances." This sentence shows us that beyond the various streams that exist in the field, the author had in mind the concern of presenting a view of the field as objective as possible to the newcomer, which is a very responsible attitude.

The book includes 14 chapters. Each chapter is organized in a very clear fashion, by making distinctions between what is a definition, lemma, theorem, corollary, assumption, or algorithm. This helps the reader avoid confusion about a pure theoretical property and widely accepted methodological ways of having the job done. A proof for each lemma, theorem, and corollary, as direct as possible, is provided each time and allows the curious reader to discover why things are the way they are. This theoretical presentation is accompanied in many places by practical guidelines and discussions that help the reader to make the step from theory to practice. Numerous, very practical, exercises are proposed, as well as commented and illustrated solutions. Only the most classical techniques are presented, allowing the reader to focus on the essentials, rather than being overwhelmed by numerous variations on the same topic and the unavoidable and often confusing discussion about advantages and disadvantages of each of them. Adequate additional references are provided for the reader who wants to go further herself or himself.

I would conclude by saying that this book, in my opinion, is one of the best introductions to geostatistics that I have ever read. It is a perfect balance between theory and methodological considerations. The presentation of the various techniques is made with a permanent concern for the comprehension of the underlying theory, without neglecting the illustrative part that helps the reader to fully understand the implications of the developments. It is concise and yet complete enough for helping the newcomer to find his or her way in the meanders of this rapidly evolving discipline. I am sure that even the more experienced geostatisticians will read it with great pleasure and will find for it a very special place on their bookshelf. It is the kind of book that you enjoy recommending to anyone who wants a good starting point in a field. I will personally use it as reference book for my students. In summary, I very strongly recommend this book to anyone who wants to have an introductory, consistent, and unbiased view of geostatistics.