Elsevier

Computers & Geosciences

Volume 31, Issue 1, February 2005, Pages 35-43
Computers & Geosciences

An effective method for 3D geological modeling with multi-source data integration

https://doi.org/10.1016/j.cageo.2004.09.005Get rights and content

Abstract

The existing 3D geological modeling systems rely heavily on large numbers of borehole and cross-section data. However, it is well known that the available geological data are generally sparse and undersampled. In this paper, we propose a stepwise refinement method for 3D modeling with multi-source data integration. The method can naturally simulate geological structures no matter whether the available geological data are sufficient or not. By stepwise refinement on multiple data, the method increases the accuracy of 3D models gradually and effectively. In addition, the mechanisms used in the method for organizing and manipulating information can have an equally important impact upon geologists’ thought, the interpretation of geological data, and 3D modeling methodology. A concrete example of using the method to Huai Bei fault and fold belt shows that the method can be applied to broad and complex geological areas.

Introduction

The technology of 3D geological modeling will bring about great changes in the methods of acquiring, storing, processing and displaying geological data. The 3D modeling methods frequently used are mainly based on plenty of 2D/3D seismic data (Zehnder and Allmendinger, 2000; Duvinage and Mallet, 2000; Sirakov and Muge, 2001) or borehole data with cross-sections between boreholes (Lemon and Jones, 2003). However, in many situations seismic data is insufficient and borehole data is very sparse, thus it is very difficult to effectively establish geological models with these methods. The data and system integration in the geosciences have been addressed extensively. Some examples are as follows. The authors in paper (Breunig, 1999) presented a sampling complex approach that facilitates the unified representation of irregular geo-objects in 3D and implements the integration of spatial data and systems for a 3D geo-information system. Based on vertical and oblique photogrammetry, Lebel et al. outlined a prospective 3D model construction method that depends on a 2.5D geological mapping and data extraction technique (Lebel et al., 2001). An open CORBA-based system architecture was presented that connects two existing geoscientific software tools—the geological 3D modeling and visualization tool (GOCAD) and a geophysical 3D modeling tool (IGMAS)––via a 3D Geo-database kernel (Shumilov and Breunig, 2000). However, no comprehensive yet convenient software system has been developed since geological data reflects geological entities and bears the features of diversity, uncertainty, and complexity.

In this paper, we propose a stepwise refinement method with multi-source data integration and present a comprehensive yet convenient 3D modeling system. The method can naturally simulate geological structures no matter whether the available geological data is sufficient or not. By stepwise refinement on multiple data, the method increases the accuracy of 3D modeling gradually and effectively. In addition, the mechanisms used in the method for organizing and manipulating information can have an equally important impact upon geologists’ thought, the interpretation of geological data, and 3D modeling methodology since the method initiates the idea of utilizing various data and other mathematic means to gradually refine 3D models. Based on the proposed method, a 3D subsurface visualization system (SVS) has been developed. The SVS implemented in Visual C++ 6.0 and OpenGL graphics library and being able to run on the PC platform, provides a comprehensive yet convenient environment, which makes it possible for the user to build 3D models in oil exploration where there is enough available data, as well as in geological survey and mineral extraction where the geological data is insufficient. A concrete example of using the method and SVS to Huai Bei fault and fold belt shows that the method/SVS can be applied to broad and complex geological areas.

The rest of the paper is organized as follows. Section 2 introduces the major steps of the proposed method and Section 3 discusses different types of data. We discuss in detail the proposed method in two sections: Section 4 for integration architecture and Section 5 for 3D modeling methods. We demonstrate the application of the proposed method/SVS to Huai Bei fault and fold belt, China in Section 6. Finally, Section 7 concludes the paper.

Section snippets

Steps involved in the method

The proposed method builds 3D models from multi-source data in a stepwise manner and involves six steps: (1) integration of 2D/2.5D data, (2) supplement of cross-sections, (3) simulation of faults, (4) definition of a template, (5) construction of horizons, and (6) representation of solids.

Data of different types and qualities are first merged into a 3D geologic model by data conversion interfaces. In the areas where borehole data and cross-sections are not sufficient, supplementary

Multi-source data

The data for representing geology are diverse and include boreholes, cross-sections, geological maps, structural geology maps, DEM, etc. Since some geological phenomena bear the characteristics of complexity and uncertainty, the geological data are incomplete and heterogeneous. In the paper, the diverse data are classified into three types: direct data, indirect data, or assistant data, depending on the type and the purpose of the information and application.

  • Direct data, such as borehole data

Integration architecture

The main difficulty raised by geological modeling is the diversity of the initially available data, both in terms of nature and reliability (Winkler et al., 1999). To utilize the geological data of different types and qualities for 3D modeling, and to keep these data consistent in models, we design an architecture of data integration and discuss it in this section. The integration mainly involves two aspects: source-oriented integration and object-oriented integration (Fig. 1).

3D modeling methods

The research in the past has focused on generating 3D models from borehole and cross-section data. This has limitations, however, since these data are lacking in many areas. In this section, we emphasize on how to use data of different types and qualities, as well as various mathematical methods to gradually refine 3D models to a desired accuracy.

Modeling in Huai Bei, China

In this section, we present a concrete example to demonstrate the application of the proposed method to 3D modeling.

The geological structure in Huai Bei, China is very complicated. The rocks are folded several times and the number of faults is 136 in the study area. (The basic properties of 79 out of the 136 faults have been obtained.) On the other hand, there are numerous valuable deposits. The most important deposits are the anthracite coal fields in the middle of the area. There are numerous

Conclusions

The stepwise refinement method with multi-source data integration is a significant improvement for 3D geological modeling. Uncertainty and inaccuracy caused by sparse original data cannot be improved by advanced interpolation methodology only. Supplementing cross-sections by a CSCAD tool is a simple and powerful way of filling the sparse data gap. However, drawing numerous cross-sections in a complicated area is tedious and unrealistic. Therefore, an efficient and robust algorithm for fault

Acknowledgements

The research presented is funded by the Research Project of National Resource Board (Grant No. 200010301-06). The authors would like to thank John H Mammoser, the editor and anonymous reviewers for their constructive suggestions.

References (14)

There are more references available in the full text version of this article.

Cited by (0)

View full text