As presented here, when fused with 2.5-D high-resolution terrain renderings, photogrammetrically-derived, and/or simple 2-D geology maps become more precise and telling representations of geology than ordinary 2-D maps. These representations are highly technological and it is easy to loose track of the science has to underpin each view angle. The extrapolation of these 2.5-D elements to a 3-D model needs to be guided by a geologist, using available and well established 2-D cross-section construction rules and an empirical knowledge of geological element behaviour (e.g. regional stratigraphic thickness variations, fault intersection behaviour) to detect possible anomalies. Balanced cross-sections are tedious to construct and several workers have developed cross-section balancing software, to alleviate the complexity of the task of managing a stratigraphic framework, following the section balancing rules and using seismic, borehole and simple geology data. Our study demonstrates that much detail could be derived from the careful study of aerial photographs of rugged areas, to create a three-dimensional geology framework. Visualizations of complex geological objects are tentative representations of reality. The «truth value» of these visualizations is dependent on the amount of control data that is available. Given that the mapping of the surface of the earth is much more economical than subsurface surveys, much remains to be done to integrate surface data into 3-D geological models derived from geophysical and borehole data. Hence, future work needs to focus on the geometrical integration of subsurface and surface data in thrust-fold belts into comprehensive 3-D geological volumes, that could be used for example to enhance seismic processing.

Many technological hurdles also need to be lifted. Data exchange between photogrammetric, cross-section construction, geological modelling, and reservoir flow modelling software are still tedious to operate and time-consuming due to a lack of a uniform data model for geological features. These software are also generally costly and beyond the reach of many academic researchers. In addition, these programs are fairly difficult to master and often lack some basic data input/output tools (e.g. we had to write a DXF translator to import-export data to gOcad). However, the biggest hurdle to 3-D geology data integration is the lack of an intuitive workflow, and of clearly stated scientific rules guiding geology modelling. If a better understanding of 3-D geological structures is to be developed, more case studies and step-by-step instructions are needed to lay out consistent methodological guides that go beyond the level of software instruction manuals. A methodological guide does not replace a good teacher however, and as demonstrated here, a computer animation can be an excellent teacher prop to emphasize the importance of sound geoscientific concepts. We can only hope that these computer animations will become more available and will accompany both software manuals and classroom textbooks. To a fair extent, these notes and animations will determine the success of universities and industry in gaining the combined technological and scientific skill base necessary to integrate leading-edge 3-D geological modelling concepts and software into the common geoscientific culture.