Abstract
An interpolation method based on a multilayer neural network (MNN), has been examined and tested for the data of irregular sample locations. The main advantage of MNN is in that it can deal with geoscience data with nonlinear behavior and extract characteristics from complex and noisy images. The training of MNN is used to modify connection weights between nodes located in different layers by a simulated annealing algorithm (one of the optimization algorithms of the network). In this process, three types of errors are considered: differences in values, semivariograms, and gradients between sample data and outputs from the trained network. The training is continued until the summation of these errors converges to an acceptably small value. Because the MNN trained by this learning criterion can estimate a value at an arbitrary location, this method is a form of kriging and termed Neural Kriging (NK). In order to evaluate the effectiveness of NK, a problem on restoration ability of a defined reference surface from randomly chosen discrete data was prepared. Two types of surfaces, whose semivariograms are expressed by isotropic spherical and geometric anisotropic gaussian models, were examined in this problem. Though the interpolation accuracy depended on the arrangement pattern of the sample locations for the same number of data, the interpolation errors of NK were shown to be smaller than both those of ordinary MNN and ordinal kriging. NK can also produce a contour map in consideration of gradient constraints. Furthermore, NK was applied to distribution analysis of subsurface temperatures using geothermal investigation loggings of the Hohi area in southwest Japan. In spite of the restricted quantity of sample data, the interpolation results revealed high temperature zones and convection patterns of hydrothermal fluids. NK is regarded as an interpolation method with high accuracy that can be used for regionalized variables with any structure of spatial correlation.
Similar content being viewed by others
Explore related subjects
Find the latest articles, discoveries, and news in related topics.REFERENCES
Aarts, E. H. L., and Korst, J. H. M., 1990, Simulated annealing and Boltzmann machines: John Wiley & Sons, New York, 267 p.
Ahmed, S., and Marsily, G. D., 1987, Comparison of geostatistical methods for estimating transmissivity using data on transmissivity and specific capacity: Water Resources Res., v. 23, no. 9, p. 1717–1737.
Allard, D., 1994, Simulating a geological lithofacies with respect to connectivity information using the truncated gaussian model, in M. Armstrong and P. A. Dowd, eds.: Geostatistical simulations: Kluwer, Dordrecht, p. 197–211.
Briggs, I. C., 1974, Machine contouring using minimum curvature: Geophysics, v. 39, no. 1, p. 39–48.
Casanova, P. G., and Alvarez, R., 1985, Splines in geophysics: Geophysics, v. 50, no. 12, p. 2831–2848.
Cristakos, G., 1990, A Bayesian/maximum-entropy view to the spatial estimation problem: Math. Geology, v. 22, no. 7, p. 763–777.
Dowd, P. A., and Sarac, C., 1994, A neural network approach to geostatistical simulation: Math. Geology, v. 26, no. 4, p. 491–503.
Deutsch, C. V., and Journel, A. G., 1998, GSLIB: Geostatistical software library and user's guide: Oxford University Press, New York, 369 p.
Gambolati, G., and Volpi, G., 1979, Groundwater contour mapping inVenice by stochastic interpolators: 1. Theory: Water Resources Res., v. 15, no. 2, p. 281–290.
Gómez-Hernández, J. J., and Srivastava, R. M., 1990, ISIM3D: An ANSI-C three-dimensional multiple indicator conditional simulation program: Computers & Geosciences, v. 16, no. 4, p. 395–440.
Goovaerts, P., 1997, Kriging vs stochastic simulation for risk analysis in soil contamination, in A. Soares, J. J. Gómez-Hernández, and R. Froidevaux, eds., geoENVI—Geostatistics for Environmental Applications: Kluwer, Dordrecht, p. 247–258.
Gu, B., Koike, K., and Ohmi, M., 1997, Interpolation of randomly distributed data by artificial neural network incorporating characteristics of spatial correlation: Geoinformatics, v. 8, no. 4, p. 269–279 (in Japanese with English abstract).
Hecht-Nielsen, R., 1992, Neurocomputing: Addison-Wesley Longman, New York, 536 p.
Hohn, M. E., 1999, Geostatistics and petroleum geology: Kluwer, Dordrecht, 235 p.
Huang, Z., Shimeld, J., Willianson, M., and Katsube, J., 1996, Permeability prediction with artificial neural network modeling in the Venture gas field, offshore eastern Canada: Geophysics, v. 61, no. 2, p. 422–436.
Kanevsky, M., Arutyunyan, R., Bolshov, L., Demyanov, V., and Maignan, M., 1995, Artificial neural networks and spatial estimations of Chernobyl fallouts: Proc. IAMG 1995 Ann. Conf., Osaka, Japan, 29 Oct.-2 Nov. 1995, p. 27–29.
Kapageridis, I. K., and Denby, B., 1998, Ore grade estimation with modular neural network systems. A case study, in G. N. Panagiotou and T. N. Michalakopoulos, eds., Information technologies in the minerals industry: A. A. Balkema Publishers, Rotterdam, p. 52.
Koike, K., and Ohmi, M., 1995, Consideration on interpolation methods to characterize a subsurface structure from irregularly spaced data: Geoinformatics, v. 6, no. 3, p. 133–146 (in Japanese with English abstract).
Koike, K., Gu, B., and Ohmi, M., 1998, Three-dimensional distribution analysis of phosphorus content of limestone through a combination of geostatistics and artificial neural network: Nonrenewable Resources, v. 7, no. 3, p. 197–210.
Koike, K., Shiraishi, Y., Verdeja, E., and Fujimura, K., 1998, Three-dimensional interpolation and lithofacies analysis of granular composition data for earthquake-engineering characterization of shallow soil: Math. Geology, v. 30, no. 6, p. 733–759.
Laine, E., 1996, Quality mapping of the Ryytimaa dolomite in western Finland: Math. Geology, v. 28, no. 4, p. 477–499.
Lajaunie, C., 1990, Comparing some approximate methods for building local confidence intervals for predicting regionalized variables: Math. Geology, v. 22, no. 1, 123–144.
Lajaunie, C., Courrioux, G., and Mauel, L., 1997, Foliation fields and 3D cartography in geology: principles of a method based on potential interpolation: Math. Geology, v. 29, no. 4, p. 571–584.
Lee, Y.-M., and Ellis, J. H., 1997, On the equivalence of kriging and maximum entropy estimators: Math. Geology, v. 29, no. 1, p. 131–152.
Matheron, G., 1981, Splines and kriging: Their formal equivalence, in D. F. Merriam, ed., Downto-earth-statistics: Solutions looking for geological problems, Syracuse University Geological Contributions, Syracuse, NY, p. 77–95.
Miller, D. M., Kaminsky, E. J., and Rana, S., 1995, Neural network classification of remote-sensing data: Computers & Geosciences, v. 21, no. 3, p. 337–386.
Mitasova, H., and Mitas, L., 1993, Interpolation by regularized spline with tension: I. Theory and implementation: Math. Geology, v. 25, no. 6, p. 641–655.
Raghavan, V., Wadatsumi, K., and Masumoto, S., 1993, Automatic extraction of lineament information of satellite images using digital elevation data: Nonrenewable Resources, v. 2, no. 2, p. 148–155.
Ripley, B. D., 1981, Spatial statistics: John Wiley & Sons, New York, 252 p.
Rizzo, D. M., and Dougherty, D. E., 1994, Characterization of aquifer properties using artificial neural networks: Neural Kriging: Water Resources Res., v. 30, no. 2, p. 483–497.
Rumelhart, D. E., Hinton, G. E., and Wiliams, R. H., 1986, Learning representations by backpropagating errors: Nature, v. 323, no. 9, p. 533–536.
Sakamoto, M., Shiono, K., Masumoto, S., and Wadatsumi, K., 1993, Acomputerized geologic mapping system based on logical models of geologic structures: Nonrenewable Resources, v. 2, no. 2, p. 140–147.
Salu, Y., and Tilton, J., 1993, Classification of miltispectal image data by the binary diamond neural network and by nonparametric, pixel-by-pixel-methods: IEEE Trans. Geosci. Remote Sens., v. 31, no. 3, p. 606–617.
Sigeno, H., Abe, K., and Noda, T., 1985, Conceptual hydrothermal system model for the Hohi area based on geochemistry of geothermal fluids: Report of the Geological Survey of Japan, no. 264, p. 285–302 (in Japanese with English abstract).
Singer, D. A., and Kouda, R., 1996, Application of a feedforward neural network in the search for Kuroko deposits in the Hokuroku district, Japan: Math. Geology, v. 28, no. 8, p. 1017–1023.
Tamanyu, S., Yoshizawa, M., and Nomura, K., 1995, Deep subsurface temperature distribution pattern estimated from many temperature logging data: Examples of Hohi geothermal area, Kyushu, Japan: Bull. Geol. Surv. Japan, v. 46, no. 7, p. 313–331 (in Japanese with English abstract).
Wang, L., 1996, Modeling complex reservoir geometries with multiple-point statistics: Math. Geology, v. 28, no.7, p. 895–907.
Zhu, H., and Journel, A. G., 1993, Formatting and integrating soft data: Stochastic imaging via the Markov-Bayes algorithm, in A. Soares, ed., Geostatistics Tróia '92: Kluwer, Dordrecht, p. 1–12.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Koike, K., Matsuda, S. & Gu, B. Evaluation of Interpolation Accuracy of Neural Kriging with Application to Temperature-Distribution Analysis. Mathematical Geology 33, 421–448 (2001). https://doi.org/10.1023/A:1011084812324
Issue Date:
DOI: https://doi.org/10.1023/A:1011084812324