Automated Stratigraphic CorrelationBy
- F.P. Agterberg, Mathematical Applications in Geology Section, Geological Survey of Canada, Ottawa, Canada,xxxx
This book provides an introduction to recent developments in automated stratigraphic correlation of fossil data, using computer programs for ranking and scaling of stratigraphic events. Mainframes or microcomputers can be used to aid the stratigrapher during data inventory for a region or time period, for construction of a biozonation based on stratigraphic events, (such as the latest appearance datum of a fossil species), and for automated correlation. The book is intended for advanced geology students, research workers and teachers with a background in stratigraphy and an interest in using computer-based techniques for problem-solving.
Developments in Palaeontology and Stratigraphy
Published: April 1990
The concept of probabilistic stratigraphy, and thus of RASC, is both intuitively simple and intellectually appealing, and, although the emphasis in Agterberg's book is on mathematics, the latter is not particularly fearsome. The evolving state of both RASC and CASC has been extensively described in a series of earlier publications and readers who are familiar with them will find a corresponding amount of familiar material here. Nevertheless, this book is obviously the presently definitive version and should be the point of departure for those seriously interested in experimenting with this approach to biozonation and correlation. It is right up to date, with a number of 1990 items in the reference list. Although ASC is mainly about RASC, there is a useful review of some other methods early in the book, and elements of some of these are indeed incorporated into the RASC/CASC package. Agterberg (himself a mathematician, but working within a Mathematical Applications section in the Geological Survey of Canada, and in close collaboration with Dr. F.M. Gradstein, a biostratigrapher) has performed a considerable service to the science of stratigraphy by developing Hay's initial idea to its fullest extent; few stratigraphers are sufficiently numerate to have attempted such a task.
Automated Stratigraphic Correlationby Fritz Agterberg not only contains a lot of music, but qualifies as a full-blown symphony both in the complexity and the orchestration of his presentation. This book is not an introductory overture or "how-to" manual for the uninitiated, but a high-level exposition and review of the state-of-the-art of quantitative biostratigraphic correlation.... the book may be viewed as an expanded and detailed program document of the highest order. The programs are available on request from the Bedford Institute of Oceanography in Dartmouth, Nova Scotia.This book presents detailed scientific analysis, not watered down introductory computer applications for non-computer people. This text is a refreshing look into the rigorous development of geoscience as well as a major summary of biostratigraphic advances in the last 25 years through the UNESCO-sponsored International Geological Correlation Program 148. The impact of quantitative methodology on stratigraphic thought is only beginning, and I believe this effort will be instrumental in the future discussions of the Code of Stratigaphic Nomenclature as well as how we go about our science on a daily basis.
Brian Shaw, BHP Americas, Inc., Houston, TX, USA, The American Association of Petroleum Geologists Bulletin
The book is an impressive presentation of computer techniques applied to stratigraphical correlation and the reader is left in no doubt that the people mainly concerned with the research involved have been very active indeed.
R. Reyment, Earth-Science Reviews
...the book provides a thorough rationale, discussion, and exploration of a quantitative stratigraphic correlation methodology that is achieving ever wider recognition and use. The problems that were inherent in quantitative stratigraphic correlation early in its development have been solved, and the book provides an excellent review of the pitfalls and how they were overcome. It belongs on the shelf of everyone involved in stratigraphic correlation.
- 1. PROBABILISTIC METHOD FOR AUTOMATED STRATIGRAPHIC CORRELATION. Introduction. IGCP Project 148. Quantitative biostratigraphy. Quantitative chronostratigraphy. Quantitative lithostratigraphy. Recent developments in stratigraphy. 2. PRINCIPLES OF QUANTITATIVE STRATIGRAPHY. Introduction. Zones in biostratigraphy. Quantitative versus qualitative stratigraphy. Local versus regional ranges of taxa. Estimation of the highest and lowest occurrences of taxa. The frequency distributions of highest and lowest occurrences of taxa. 3. APPLICATIONS OF MATHEMATICAL STATISTICS AND COMPUTER SCIENCE TO ZONATION, CORRELATION AND AGE INTERPOLATION. Introduction. Binomial test for randomness. Binomial distribution model for microfossil abundance data. Multiple pairwise comparison. Applications of graph theory. Use of cubic smoothing splines for removing ``noise'' from microfossil abundance data. Biostratigraphic correlation between Tojeira 1 and 2 sections in central Portugal using E. mosquensis abundance data. Multivariate methods. Research on time-scales. Computer simulation experiments on estimation of the age of chronostratigraphic boundaries. Smoothing of time-scales with the aid of cubic spline functions. Statitistical significance of ages. 4. CODING AND FILE MANAGEMENT OF STRATIGRAPHIC INFORMATION. Introduction. Five basic types of files. Hay example as derived from the Sullivan database: Lower Tertiary nannoplankton in California. Partial DAT file for the Hay example. DAT files constructed by Guex and Davaud. Gradstein-Thomas database: Cenozoic foraminifera in Canadian atlantic margin wells. Characteristic features of Gradstein-Thomas database. Frequency of occurrence of taxa of cenozoic foraminifera along the northwestern Atlantic margin. Artificial datasets based on random numbers. 5. RANKING OF BIOSTRATIGRAPHIC EVENTS. Introduction. Hay's original method. Algorithmic version of Hay's original method. Uncertainty ranges for events in the optimum sequence. Other ranking algorithms. Conservative ranking methods. Three-event cycles. Higher-order cycles and pseudo-cycles. The influence of coeval events. 6. SCALING OF BIOSTRATIGRAPHIC EVENTS. Introduction. Scaling versus ranking. Statistical model for scaling of stratigraphic events. Artificial example. Computer simulation experiments. Normality test. Marker horizon option of the RASC method. Unique event option of RASC program. Binomial and trinomial models for scaling. Application of Glenn and David's trinomial model. Comparison of observed and estimated probabilities. 7. RANK CORRELATION AND PRECISION OF SCALED OPTIMUM SEQUENCE. Introduction. Rank correlation coefficients. RASC step model. Presorting and ranking by Harper. Precision of the scaled optimum sequence. 8. NORMALITY TESTING AND THE MODIFIED RASC METHOD. Introduction. Autocorrelation of the second-order differences. Unitary Associations and RASC methods applied to Drobne's alveolinids. Application of RASC and normality test to Palmer's database for the Riley Formation in central Texas. Modified RASC method. Application of modified RASC to the Gradstein-Thomas database. Frequency distributions of stratigraphic events. Application of modified RASC to Drobne's alveolinids. Comparison of range charts for Palmer's database. 9. EVENT-DEPTH CURVES AND MULTI-WELL COMPARISON. Introduction. Principles of correlation and scaling in time and comparison to composite standard method. Generalized description of the CASC method. Statistical selection of optimum spline-curves. Cross-validation method. Jackknife method. Computer simulation experiment for event-depth spline fitting with error analysis. Regional application of RASC and CASC. Application of RASC and CASC to Hibernia oilfield. Application of CASC to Palmer's database. Benthic foraminiferal zonation, central North sea. Integration of forminiferal and dinoflagellate datasets, Labrador shelf-grand banks. 10. COMPUTER PROGRAMS FOR RANKING, SCALING AND REGIONAL CORRELATION OF STRATIGRAPHIC EVENTS. Introduction. Summary of contents of the 12 modules of micro-RASC. List of decisions to be made by user of the RASC computer programs. Brief history of the development of RASC and CASC. References. Index.