Geostatistics pp 143-161 | Cite as

Geological Models in Transition

  • W. C. Krumbein
Part of the Computer Applications in the Earth Sciences book series


The past decade was a time of transition from dominantly univariate statistical procedures in geology to increasingly multivariate techniques as the influence of the computer spread through the science. The accompanying rapid rise of the model concept in geology brought with it a diversity in methods of structuring data for search procedures and statistical predictor models. The general linear model, with its great flexibility, easily rose to dominance in this domain, both for research and practical application.

But other transitions, mostly unanticipated a decade ago, were also on their way. Simulation models, introduced for computer experimentation in several fields, called attention to the need for more critical and analytical structuring of geological models. The newer models in large part specify (or at least suggest) the types of observational data needed to test their relevance to real-world phenomena. This is particularly true for probabilistic models, in which conventional time-honored observations do not always provide critical data for discrimination among competing models. The spatial implications of the models require critical examination supported by new types of field observations not normally made in a search approach. Population densities of simulation input variables and operational definitions for their measurement also need reevaluation. This paper discusses the problems in the context of stratigraphic analysis and of stream networks in geomorphology.


Markov Chain Stream Network Topological Class Topological Form Distinct Network 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. James, W. R., 1967, Nonlinear models for trend analysis in geology: Kansas Geol. Survey Computer Contr. 12, p. 26–30.Google Scholar
  2. James, W. R., 1968, Development and application of nonlinear regression models in geology: unpubl. doctoral dissertation, Northwestern Univ.Google Scholar
  3. James, W. R., 1970, Regression models for faulted structural surfaces: Amer. Assoc. Petroleum Geologists Bull., v. 54, no. 4, p. 638–646.Google Scholar
  4. James, W. R., 1970a, The geological utility of random process models, in Geostatistics, Plenum Publ. Co., N.Y., p. 89–90.Google Scholar
  5. Krumbein, W. C., and Dacey, M. F., 1969, Markov chains and embedded Markov chains in geology: Jour. Intern. Assoc. Math. Geology, v. 1, no. 1, p. 79–96.CrossRefGoogle Scholar
  6. Krumbein, W. C., and Shreve, R. L., 1970, Some statistical properties of dendritic channel networks: U. S. Office of Naval Research, Tech. Rep. 13, ONR Task No. 389–150, Northwestern Univ., 117 p.Google Scholar
  7. Shreve, R. L., 1966, Statistical law of stream numbers: Jour. Geology, v. 74, no. 1, p. 17–37.CrossRefGoogle Scholar
  8. Shreve, R. L., 1967, Infinite topologically random channel networks: Jour. Geology, v. 75, no. 2, p. 178–186.CrossRefGoogle Scholar
  9. Smart, J. S., 1969, Topological properties of channel networks: Geol. Soc. America Bull., v. 80, no. 9, p. 1757–1774.CrossRefGoogle Scholar
  10. Vistelius, A. B., 1949, On the question of the mechanisms of the formation of strata: Doklady Acad. Nauk. SSSR, v. 65, p. 191–194.Google Scholar

Copyright information

© Plenum Press, New York 1970

Authors and Affiliations

  • W. C. Krumbein
    • 1
  1. 1.Northwestern UniversityUSA

Personalised recommendations