Ten years ago, at the Exploration ’77 meeting in Ottawa, Allan Spector and Wilfred Parker presented a paper entitled Computer Compilation and Interpretation of Geophysical Data. At that time, the geophysical community was on the threshold of developing new “tools” for use in both airborne and ground magnetics. Higher- resolution instrumentation for airborne surveys; the development of microprocessor-controlled field magnetometers and gradiometers of much greater accuracy, reliability and data storage capability; and the development of reliable, cost-effective techniques for data enhancement and modelling: These were representative of the emerging “tools.” Since then, many of these developments have been realized — in particular, the improvements in over-all survey quality, improvement, in magnetometer accuracy and navigational positioning, and the complete conversion to digital acquisition by all the survey companies. Microprocessor-based ground magnetometers are now being produced by all major geophysical instrument manufacturers. The production of data diskettes and the storing of surveys in bubble memories is as routine today as were electronic notebooks and cartridge tapes in 1977.
Data-enhancement techniques are of particular interest to an industry bent on finding mineable mineral deposits. Data-enhancement means the process of highlighting or subduing a particular desired pattern or signature within the over-all magnetic field that has been sampled. Since the early 1970s, examples of these data-enhancement images have been relatively common. Total field contour maps have been used for more than 30 years. Today’s technology enables geophysicists to enhance shallow features at the expense of deeper source signatures. An artificial sun illumination can also be used to enhance features using an Applicon ink-jet color plot, for example. After 1977, these processed images became available over selected areas using what digital data sets were available. They were produced by relatively few companies.
The production and availability of these images (or data sets) in the past 10 years and the ease and reliability of data-enhancement has contributed greatly to our progress in aeromagnetic interpretation.
On a super-regional scale, extremely large data sets have been completed or are in the process of being compiled. The recently-completed anomaly map of the entire North American continent is a prime example of this type of compilation, allowing processing and interpretation at a scale only dreamed of in the 1970s.
Projects for similar compilations over the African and South American continents and over continental margins are at the initial stages. The completion of these projects will facilitate large-scale basin analyses and the interpretation of complete geological provinces that, until now, have not been contemplated. With the advent of these large databases, the time period and cost for acquiring the basic geological framework for a prospective area has been greatly reduced.
The display of magnetic data using image analysis and data-enhancement techniques has advanced significantly with the advent of readily accessible computer hardware and work stations. An Intergraph work station tied to a vax 780 computer has been installed at Leeds University in the U.K. It is being used to compile the huge continental, geophysical databases mentioned earlier.
Image analysis techniques can also be applied to aeromagnetic data in a personal computer (pc). Advances in image-processing systems and the accessibility of personal computers as work stations and terminals have brought about significant improvements in the quality of aeromagnetic interpretation.
Within the past 10 years, computer hardware and software technology has moved away from the “white coats in the air-conditioned back room,” so that geophysicists and geologists may now generate processed or enhanced products at the routine initial stage of an aeromagnetic interpretation.
The available algorithms that may be used for aeromagnetic interpretation have also improved with the advent of new techniques for variable depth susceptibility mapping, methods for continuing fields on to uneven surfaces, and methods for integration or “layering” of digital magnetic images with complementary data sets such as topography, gravity, geochemistry and, most recently, with satellite and airborne radar images. These processes are but the first step in the important task of geological interpretation, by which we mean:
* the determination of the distribution and geometry of magnetic sources; and
* the recognition of structural features and lithologic boundaries through changes in magnetic trends and textures.
The determination of the distribution of magnetic sources and their pertinent parameters has been revolutionized on two fronts:
* the availability of fully interactive modelling programs on the pc; and
* the capability for the systematic modelling of a large aeromagnetic survey. In this case, thousands of line kilometers of airborne data may be processed automatically on an anomaly-by-anomaly basis, leading to a magnetic body plot that facilitates the structural interpretation of the area.
In a more detailed sense, modelling may be combined with other data- enhancement techniques to lead to an enhanced lithologic interpretation. When the improved magnetic image (such as a susceptibility map) is combined with the hard data from automatic modelling results, a more reliable lithologic interpretation results.
The recognition of structural features and lithologic boundaries now results in the routine creation of pseudo-geology maps. These are based on the classification and identification of magnetic signature parameters such as lineation, texture, amplitude and shape. The major contribution of the improved image analysis and data-enhancement techniques in the past decade has been to supply the interpreter with a much clearer picture of these parameters and hence to allow a more definitive interpretation.
Even when the regional geological framework of an area is known, aeromagnetics can help find a particular lithological/structural target. Over a deep carbonate basin, for example, target signatures on a high-resolution, total magnetic field map (shown on page 33) suggest zones associated with basement faults and late-stage volcanics that may act as controls for Sedex-type mineralization. The first step in such a case would be to create a regional-residual separation in order to remove the large regional component caused by extrusive rocks near the basement surface. To obtain a better resolution of the residual features, this image can then be poll-reduced and continued downward, close to the basement surface. When this procedure was followed in an actual case, east-west structures were successfully identified along with clear indications of faulting sub-parallel to stratigraphy and numerous cross-trending features. Features that had been previously interpreted as basement faults were in fact due to minor errors in the original line-to-line data levelling that was not evident on the total field map.
To eliminate these spurious artifacts and to provide still better resolution, a directional filter was applied and the first vertical derivative calculated. This resulted in the final image, shown on page 28. Faults, lithological boundaries and some important intrusive events were identified in the interpretation phase. When compared with the original image, it is clear that the final product has greatly facilitated the interpretation phase.
The evolution towards the final image in this example involved enhancing the geophysical signatures of known or suspected geological patterns. Herein lies the major benefit of the recent advances in aeromagnetic interpretation: the ability to generate images that can be used to interpret lineation, textures and shapes that are much closer in appearance to their true geological sources.
The above interpretation and processing example embodies a further important progression tha
t has occurred in the past decade. The geophysicist must now be concerned not only with interpreting or proposing a distribution of magnetite that best fits the observed data or enhanced image, but with a distribution that is consistent with the mapped geology and with reasonable geological models. Mathematically, this means that geophysicists have been able to reduce the non-uniqueness problem by greatly reducing the number of possible answers. Realistically, what is happening is that a good interpretation geophysicist today must become more and more part-economic and part-structural geologist. In fact, many of today’s geologists are carrying out the geophysical interpretation of aeromagnetic images.
The closer integration that has been achieved between the enhanced images of the magnetic field and the underlying geological sources has initiated developments in the examination of the relationships between magnetite distribution and regional/ detailed and economic geology. These somewhat less spectacular developments have occurred on two fronts. In the fields of theoretical and experimental petrochemistry and mineralogy, progress has been made in understanding the stability and distribution of the magnetite family of minerals in various geochemical and metamorphic environments. This has helped us to relate magnetic textures to geological processes, including both genesis and metamorphic over-printing. On another front, the sampling and analysis of rock magnetic susceptibilites and remanence characteristics have changed from the traditional study of iron ore to a more statistically significant sampling of rocks for geological interpretation. Recently (in Scandinavia, for example) more regionally based data sets of susceptibility have become available. Within the next decade, as we obtain access to larger and more diverse data sets on the distribution of magnetite within various geological settings, we will be able to further reduce the ambiguity of our interpretations.
As mentioned previously, the introduction of microprocessor-based magnetometers has advanced the art of ground follow-up or aeromagnetic targets. In the majority of today’s cases, these targets are no longer simply anomalous “bull’s-eyes” or linear iron-formations. We have progressed to a level of sophistication where we are able to design follow-up programs to delineate mafic-felsic contacts within a volcanic sequence, as well as derived sediments; we can recognize and delineate alteration zones and regions of major faulting and minor crosscutting shear zones; and, in many cases, we can outline intrusive features and indicate their geometric relationships, age and zonation. In medium-to-high-grade metamorphic environments, we are able, with the use of sensitive gradiometer measurements, to separate the various schists and gneisses and indicate their probable ancestry back through the metamorphic event. Instruments in use today have the capability of measuring the vertical gradient and storing and automatically correcting at least 10-20 km of readings.
Within the past two years, the advances in image analysis have joined this new generation of ground magnetometers to produce final enhanced products in the field. Both total field and enhanced maps that were generated using mainframe computers and a large color plotter in 1977 are being produced by field-portable personal computers and off-the-shelf printer plotters today.
At the conclusion of the paper on aeromagnetics at Exploration ’77, the authors forecasted that “computer languages (would) continue to become more powerful and easier to use and (that) computer operating systems and user programs (would) allow greater usage of interactive terminals by geophysicists.” The industry has exceeded this forecast.
Our forecast for advances in the next decade are focused on five areas of research:
* We will have completed possibly three continent-wide compilations of aeromagnetic data.
* A more standardized and representative physical database will become available.
* What we will call universal geophysical databases will be more readily available and will incorporate topography, gravity and radar and other satellite images with the aeromagnetic data.
* By the end of the next decade, we will be generating complex structural and possibly lithological maps based on interpretations, with the aid of pattern recognition algorithms and artificial intelligence. James Misener is vice-president of Paterson, Grant & Watson Consulting Geophysicists in Toronto. This is an abbreviated version of a paper presented at Exploration ’87. The complete paper will appear in the proceedings volume, to become available from the Ontario Geological Survey in 1988.
Be the first to comment on "DECADES OF CHANGE"