Stochastic Modeling and Geostatistics: Principles, Methods, and Case Studies, Volume II
Edited by T. C. Coburn, J. M. Yarus, and R. L. Chambers


PREFACE

The basic ideas of geostatistics have been known since the earliest writings of Krige, Matheron, and their contemporaries; but it was not until the mid- to late-1980s that geostatistical methodology would undergo a major expansion that would ultimately change the face of data analysis in all disciplines that use spatial data. While many of the new developments initially took form in academic centers like Stanford University, the University of Texas, and Ecole des Mines in Fontainbleau, they were given life in the research centers of companies like BP/Amoco, Marathon Oil, and ExxonMobil through practical application to applied geoscience and petroleum-related problems.

By the early 1990s, geostatistics had reached a certain level of maturity and it was beginning to be integrated into a number of reservoir studies. The idea for a book on geostatistics began to germinate at Amoco and Marathon Oil; and in 1994, the American Association of Petroleum Geologists (AAPG) published Stochastic Modeling and Geostatistics: Principles, Methods, and Case Studies (SMG) as the third contribution to its series on computer applications in geology. The book presented both a reflective review of developments in geostatistics up to that time and a forward look at the application of a number of new developments which were just beginning to be come to the forefront.

Unbeknownst to the editors and to AAPG, SMG appeared at a time that coincided with the chronological cusp of an even greater explosion in spatial thinking, stochastic modeling, and geostatistical data analysis. Whether intended or not, from that point forward the practice of geoscience experienced a rather remarkable evolution, as did many other scientific disciplines. Since its publication, SMG has become one of the most highly regarded, oft-quoted, and visible texts on geostatistical methodology, and it has served to stimulate countless new ideas, applications, and techniques.

It has now been more than ten years since SMG was published, and the discipline of geostatistics has undergone tremendous expansion and development since that time. In particular, the fundamental ideas of geostatistics have become even more intertwined with the broader notions of stochastic modeling and spatial statistics that, in their own right, have undergone unparalleled expansion and growth. In this same timeframe, computational power has mushroomed and new information technology (e.g., geographic information systems) has literally transformed the practice of geoscience.

Because much has happened in the realm of geostatistics since 1994, the task of summarizing and characterizing the signal accomplishments would be an almost insurmountable task. Yet, we believe it is paramount that the geostatistics community—in particular, the earth science part of that community—once again step back, reflect on its past achievements, and look forward to further developments and enhancements. Like sequels to good movies, sequels to good books are difficult to create; and so it is with considerable trepidation that we have once again undertaken the assignment of compiling a body of information about geostatistics and stochastic modeling that is both representative of achievements throughout the past decade and forward-looking enough to catapult the discipline to a new level. The challenge of preparing yet another high-profile book on geostatistics is quite daunting when so many good ones have been published in recent years.

The idea for this second volume of SMG took root late in 2000 at an AAPG-sponsored Hedberg conference held at The Woodlands Conference Center outside Houston, Texas. Many of the presentations from that conference were subsequently revised and formalized, and are included as part of this volume. The Hedberg conference papers have been supplemented by other selected reports and essays that span the continuum of methodology and applications from recent years, as well as some relatively new ideas and technologies that will likely experience further development and implementation in the near future.

The volume was initially compiled in the summers of 2001–2003 while the senior editor served as visiting scientist at the U.S. Geological Survey in Reston, Virginia, and at Sandia National Laboratories in Albuquerque, New Mexico. This was a challenging, transitional timeframe marked by changes in thought, practice, and profession from which we, like many others, were not exempt, and so this effort has required much more time to complete than originally envisioned.

The principal title of the book is once again Stochastic Modeling and Geostatistics. Whereas “geostatistics” is a fairly well defined notion, “stochastic modeling” has become a somewhat more nebulous term due to the corresponding development of probabilistic approaches that are not specifically spatial in nature. Consequently,in this second volume of SGM, as in the first, stochastic modeling is assumed to involve geostatistical or related spatial techniques. We acknowledge both the interrelatedness of the two disciplines and the diverging lines of thought and practice.

In addition to the application of some relatively new methodologies, readers of SMG II will no doubt recognize the use of several familiar ideas—possibly even some early vintages of such ideas—in both well-known and novel situations. One recurring theme is the integration of seismic information, reflecting the significant growth that has occurred in this area in recent years. In addition, the book includes material that is not necessarily petroleum-related and information that is not solely intended for use in the earth sciences. We have purposely broadened the scope of the text to underscore the expanding influence of geostatistics in disciplines other than the geosciences.

In recent years the field of geostatistics has been blessed with the addition of numerous new researchers and practitioners—both young and old alike—and readers will note the expanded collection of authors represented here. These men and women constitute a remarkable cross-section of the human capital devoted to geostatistical endeavors in academia, industry, and government, and our hats go off to them for their continued commitment to excellence and creativity, and to their patience with us as editors.

The book is divided into five sections. Each section is accompanied by its own short introduction. As a general introduction, Section I contains reflective essays pertaining to the state of geostatistics and stochastic modeling, and addressing their interrelatedness and their connection to one another. Section II is devoted to theoretical developments and the advancement of fundamental principles. Section III focuses on novel applications of methodology in a variety of settings and situations. Fully developed case studies are grouped together in Section IV. While there is a fine line between methodological applications and case studies, the chapters in this section represent more comprehensive investigations encompassing the technical, practical, and operational aspects of conducting a geostatistical investigation. Section V contains papers on systems and resources that are pertinent to the continuing evolution of geostatistics and stochastic modeling. The resulting collection of chapters represents what we believe to be a good mix of theory, methods, applications, and philosophy.

While embracing diversity of ideas, applications, methodologies, and situations, we felt it necessary to impose some editorial consistency on the final version. Good communication is essential to the successful completion of any geostatistical project, and the use of common terminology promotes better understanding among all parties involved. Hence, we have made an extensive effort to standardize language and word usage throughout the book, and to impose a consistent presentation format. For example, we have chosen to use the word “semivariogram” as opposed to “variogram,” we have chosen “data” to be plural rather than singular, and we have required some rigor in the use of the terms “model” and “modeling.” In the editing process we have made a conscious effort to enforce these rules, but in an effort of this size, complete adherence is essentially impossible. No doubt disparities remain. Finally, we have made every effort to eliminate spelling and grammatical mistakes, but some will have gone undetected.

Throughout the book there are references to vendors and products that are of a proprietary nature. We have made every attempt to give credit where credit is due, and we apologize in advance to those who may not have received proper acknowledgement. Reference to specific commercial or governmental entities by various authors, or mention of products or services available from such organizations, does not necessarily represent our views or opinions, or those of the publisher, nor does it in any way constitute an endorsement.

From the beginning of this project we have had three main objectives: (1) to advance the general knowledge of geostatistics; (2) to benchmark the state-of-the-art in geostatistical thought and practice, principally within the geoscience community at the beginning of the 21st century; and (3) to foster an understanding in the commercial world that employing geostatistical technologies can positively impact the bottom line. As the old adage goes, “nothing succeeds like success;” and so we hope the readers see the successful implementation of geostatistical tools in every chapter of this book. Whether or not we have achieved our objectives remains an open question— one that only our readers can address. We also hope SMG II will be as informative and enduring as the first volume has been, and that it will be a springboard for greater expansion of the geostatistics discipline.

Timothy C. Coburn, Abilene, Texas
Jeffrey M. Yarus, Houston, Texas
Richard L. Chambers, Tulsa, Oklahoma