ArticlePDF Available

Abstract and Figures

Traditional methods of geological mapping were developed within the inherent constraints imposed by paper-based publishing. These methods are still dominant in the Earth Sciences, despite recent advances in digital technology in a range of fields that include Geographical Positioning Systems, Geographical Information Systems, 3-D computer visualisation, portable computer devices, knowledge engineering and Artificial Intelligence. Digital geological mapping has the potential to overcome some serious limitations of paper-based maps. Geological maps are usually highly interpretive, but traditional maps show little of the raw field data collected or the reasoning used during interpretation. In geological mapping, interpretation typically relies on the prior experience and prior knowledge of the mapper, but this input is rarely shown explicitly with the final map. Digital mapping techniques open up new possibilities for quantifying many types of uncertainty associated with the mapping process, and using this uncertainty to evaluate the validity of competing interpretations.
Content may be subject to copyright.
Digital mapping: quantifying uncertainty 1 Jones et al 25.06.03 14:51
Digital field data acquisition: towards increased quantification
of uncertainty during geological mapping
CognIT a.s, Postboks 610, 1754 Halden, Norway
Department of Geological Sciences, University of Durham, Durham, DH1 3LE, U.K.
Corresponding author:
Number of words: 5977 (including all text, references, captions, tables).
Abbreviated title: "Digital mapping: quantifying uncertainty" (max. 40 chars)
Abstract __ Traditional methods of geological mapping were developed within the inherent
constraints imposed by paper-based publishing. These methods are still dominant in the Earth
Sciences, despite recent advances in digital technology in a range of fields that include
Geographical Positioning Systems, Geographical Information Systems, 3-D computer
visualisation, portable computer devices, knowledge engineering and Artificial Intelligence.
Digital geological mapping has the potential to overcome some serious limitations of paper-based
maps. Geological maps are usually highly interpretive, but traditional maps show little of the raw
field data collected or the reasoning used during interpretation. In geological mapping,
interpretation typically relies on the prior experience and prior knowledge of the mapper, but this
input is rarely shown explicitly with the final map. Digital mapping techniques open up new
possibilities for quantifying many types of uncertainty associated with the mapping process, and
using this uncertainty to evaluate the validity of competing interpretations.
Although the methodology of geological mapping has remained largely unchanged since William
Smith’s pioneering work two hundred years ago (Smith 1801, 1815), recent developments in
digital technology have the potential to revolutionise the way in which geological field data is
gathered, stored, processed, displayed and distributed. The digitalisation of field mapping is set to
occur through advances in GPS (“Geographical Positioning Systems”), GIS (“Geographical
Information Systems”), highly portable hand-held PDA devices (“Personal Digital Assistants”),
high powered 3-D computer graphics, and satellite communication equipment. In this paper we
present a justification of the fundamental importance of how this new technology can help to
improve some of the inherent limitations of traditional methods of geological mapping.
Digital mapping: quantifying uncertainty 2 Jones et al 25.06.03 14:51
Prior Information
In common usage the concept of "Prior Information" conveys a variety of meanings, largely
dependent upon context, ranging from philosophical to mathematical. This paper is not intended
as an exhaustive discussion on the semantics of prior information, but rather to focus
pragmatically on specific issues that have direct relevance to the process of geological mapping.
Relationship between information and knowledge
In relation to geosciences it is convenient for us to consider “prior information” in a broad way
that also encompasses related concepts such as “data” and “knowledge”. During the last decade
many business sectors, including the hydrocarbon and mining industries have focused heavily on
maximising re-use of corporate knowledge and expertise. This highly profit-driven process has
given rise to very pragmatic research in areas of applied knowledge engineering. Workers in these
fields (e.g. van der Spek & Spijkervet 1997; Liebowitz & Beckman 1998) generally view
“information” as just one part of a knowledge hierarchy that ranges from basic data input to
sophisticated interpretation by an expert (Table 1).
Interdependence of data, information and knowledge
In traditional views concerning the way in which scientific progress occurs, heavy emphasis is
placed on the role of induction to derive scientific theories based on observed phenomena (Fig. 1).
Because there was generally little consideration that sensory inputs might be misleading, objective
“scientific” observations could be treated as facts. This view of scientific methodology was
largely dominant until early in the twentieth Century, and still remains influential in much of
Earth Science education today.
The inductivist view of scientific methodology has been refuted in several ways, and individual
scientific theories are now more generally viewed as belonging to larger knowledge structures
(the “paradigm” of Kuhn 1970, or the “research programme” of Lakatos 1974). Of central
importance in the rejection of induction is that observations cannot be made independently of
prevalent theory, and that the formation of individual theories builds upon prior scientific
knowledge. The significance of this is that there is a solid philosophical foundation for stating that
prior information (sensu lato) is always an influential factor during scientific research, including
geological mapping and other disciplines within the Earth Sciences. Put more simply, this
discussion merely emphasises something that most geologists will take for granted: the end
product of a geological mapping process is a non-unique, subjective interpretation that is
markedly influenced by the previous background, prior experience, and expertise of the geologist,
and is an interpretation made within the context of current geological understanding. Although in
Digital mapping: quantifying uncertainty 3 Jones et al 25.06.03 14:51
general this is implicitly understood by field geologists, there is rarely any attempt to state
explicitly the assumptions or inferences made, or to quantify any uncertainty associated with
Accessibility of knowledge
Researchers who have studied the transfer of knowledge between workers in modern industry
have recognised that there is a great variety in the accessibility of an individual’s knowledge and
expertise (Nonaka 1994; Nonaka & Takeuchi 1995). Whilst some knowledge often exists in an
explicit format that is readily available and that can be communicated and understood by other
colleagues, there is usually also a large amount of implicit knowledge that is not (yet) in a format
that is easily accessible to others (Table 2). Implicit knowledge exists within the head of a person,
but can readily be made available to others (i.e. made explicit) through query and discussion, or
through a conscious decision to document what one knows about a subject. In contrast, a large
amount knowledge possessed by humans is usually tacit (Polanyi 1958, 1966). Tacit knowledge
exists within the head of a person, but is generally in a form that is not easily accessible to other
people, either because its owner is not aware that they possess the knowledge, or will not be able
to express it in a useful, understandable format. An example is the knowledge of how to ride a
bicycle: although you may know how to cycle without having to think, it is very difficult to give
an explicit description of how to perform the task to someone who has never tried themselves.
Traditionally the transfer of tacit knowledge of this type is achieved by repeated learning through
“trial and error”, often in a teacher/pupil or master/apprentice situation. More recently, specialised
knowledge elicitation techniques have been devised that extract tacit knowledge and represent it
in more explicit forms (e.g. Boose 1986; Kidd 1987; Ford & Sterman 1998; White &
Workers who are considered to be “experts” within a particular domain typically possess high
levels of tacit knowledge. This is often particularly acute within geological mapping, where the
quality of output will be greatly dependent upon the skill and expertise of the mapper, although
their proficiency is rarely immediately obvious simply by looking at the final result.
Conditional probability
Ideally, when faced with uncertainty associated with interpretation of individual field
observations, the geologist should attempt to make further field observations in order to resolve
any outstanding issues. In practice, this process will almost always be restricted by limitations
imposed through lack of exposure and limited resources (manpower, time, money), and the
geologist is almost always obliged to supplement direct observation with a mixture of insight,
guile, guesswork, and "gut-feeling". These aspects reflect the geologist’s personal bias, and
Digital mapping: quantifying uncertainty 4 Jones et al 25.06.03 14:51
represent a prior belief which if it can be expressed explicitly, can be used to predict the
likelihood that a particular interpretation is true. This forms the basis of Bayesian statistics, in
which the probability that hypothesis H is true given the occurrence of event E (in context I) is
given by:
p(H | EI) = ( p(H | I) * p(E | HI) ) / p(E | I).
Thus Bayesian probability can provide a mathematical framework for expressing the uncertainty
associated with geological interpretation, and offers a possible way to extend and improve
traditional methods of geological mapping.
Traditional Geological Mapping
The ability to produce accurate field maps and to record associated observational data in a
notebook lies at the core of all Earth Science activities (e.g. Barnes 1981; McClay 1987), and
forms the basis by which geological maps are constructed. Whilst a geological map is a two-
dimensional (2-D) representation of the distribution of rock formations in an area, it also conveys
through symbols and graphics the three-dimensional (3-D) geometry and form of the rocks and
structures in the area. The gathering of field data occurs in a broad range of natural environments
and is typically carried out by individuals or small teams of geoscientists working on foot or using
various forms of transport: this requires that any equipment or techniques used to gather
information are highly mobile and easy to maintain. For these reasons, it perhaps not surprising
that field mapping has since its inception remained as a paper-based activity using maps, field
notebooks and compass-clinometers.
The scientific aims of a study and the time available for fieldwork – which is often dictated by
funding – will determine the type of geological mapping to be carried out. Barnes (1981)
identifies four main types of geological mapping activities:
a) Reconnaissance mapping typically covers a large area and is carried out to find out as much as
possible about a poorly-known region over a short period of field study. Significant amounts of
work may be done using remote sensing techniques or interpretation of aerial photographs.
b) Regional mapping typically results in geological maps at 1:50,000 scale recorded on an
accurate topographic base map. Such mapping is generally the result of systematic programmes
of field-based data gathering, fully supported by photogeological interpretation and integration
of other sub-surface geological or geophysical datasets.
c) Detailed mapping generally refers to maps made at 1:10,000 or larger scales and in many cases
are produced to document key geological relationships in detail. Many require the field
Digital mapping: quantifying uncertainty 5 Jones et al 25.06.03 14:51
geologist to produce their own base map simultaneously using planetable-, chain- or cairn-
mapping techniques (see Barnes 1981 and references therein).
d) Specialized mapping where maps are constructed for specific purposes and do not necessarily
include all aspects of the observed geology. These include mine plans and maps showing
geotechnical, geophysical or geochemical data.
The following information is particularly critical to all field-based data-gathering and
observational activities: accurate location, geological context, and the spatial/temporal
relationship to other data or observations gathered at that or other location(s). In addition, all field
maps and notebooks must be legible, be readable by another geologist and must clearly
distinguish between observed facts and inferences drawn from those facts (Ramsay & Huber
1987). Generally speaking, observations and data are gathered at a series of localities, the
locations of which are marked by hand onto a topographic or aerial photographic base map, with
all data measurements and observations being recorded simultaneously in a field notebook.
Ideally, the extent of visible outcrop (“green-line mapping”) and location of mappable geological
boundaries – including some indication of how well constrained these boundaries are in terms of
the available exposure (using solid, dashed or pecked lines) - will be added to the base map by the
field geologist as they move between localities. Most geologists are encouraged to interpret their
observations and measurements as they map and to modify these interpretation as more
information is acquired. Thus field mapping is a uniquely iterative exercise in which both data-
gathering and interpretation occur simultaneously. This also means, however, that field-based data
gathering presents a number of very significant challenges viewed from a prior information
perspective. In particular, we would highlight four main problems:
Field mapping involves extensive use of tacit knowledge, in which the a priori assumptions
made both when making interpretations or even when gathering data are either not stated or
may not even be considered – thus ‘facts’ and ‘inferences’ are often not clearly separated in
many, if not all studies.
The workflow from field-mapping to published map is generally a complex process involving
data collection, interpretation, data reduction and final map drafting. The map is an abstraction
at one specific scale of a large amount of data collected at the outcrop scale. Therefore the vast
majority of ‘inputs’, ‘data’ and ‘information’ are typically excluded from the final result as
they generally cannot all be incorporated into the final paper map. Most maps are therefore
dominated by interpretations (‘knowledge’ + ‘expertise’). In many cases, some of the
interpretation is made in locations far removed from the field either before or after the actual
data-gathering was carried out.
Digital mapping: quantifying uncertainty 6 Jones et al 25.06.03 14:51
All paper maps are inherently limited in terms of what they show by their scale. In many cases
this means that they lack precise spatial accuracy, meaning that reproducible observations or
measurements are often difficult or impossible.
The final map generally shows little expression of the uncertainties involved in its production,
and where uncertainty is depicted it is primitive, ad-hoc, and qualitative. A basic representation
of uncertainty is often indicated through the use of different line symbols (e.g. solid lines for
exposed boundaries vs. dashed lines for inferred or conjectural). Thus traditional geological
mapping remains a highly interpretative, subjective art form in which uncertainty is difficult to
quantify in any statistically meaningful way.
Digital Geological Mapping
GIS has evolved from its early use as a computer mapping system and is now defined as ‘an
information management system for organizing, visualizing and analysing spatially-orientated
data’ (Coburn & Yarus 2000 p.1). Since GIS became commercially available in the 1980s, GIS
products are now used in a large number of applications that deal with spatial data, including
social and economic planning, marketing, facilities management, environmental and resource
assessment (Rhind 1992; Longley et al. 2001). In its original guise, GIS largely dealt with 2-D
data that was mapped onto the Earth’s surface (Rhind 1992). It was recognised that in order to
deal with volumetric spatial information or 3-D geometries from sub-surface data, a 3-D GIS or a
GSIS (GeoScientific Information System) was required, and such systems (e.g. GOCAD) have
now been developed for commercial purposes (Turner 1992; Mallet 1992; Turner 2000).
Bonham-Carter (2000) described the core GIS activities in a geoscience project as being:- 1) data
organisation, 2) data visualization, 3) data search, 4) data combination, 5) data analysis and 6)
data prediction and decision support. The combination of these capabilities and the ability to
handle large databases (up to a terabyte) indicate the power of the GIS approach for handling
spatial data and its attraction for geoscience users such as the petroleum and mining industries.
Digital Geological Mapping (DGM) is a methodology by which a geologist collects GPS-located
field data in a digital format. The method has been adapted from digital mapping and surveying
techniques and is becoming widely used in construction, engineering and environmental
industries. In Earth Science fieldwork, apart from a few specific studies, such as determining the
3-D architecture of fault zones and sandstone bodies (Maertens et. al. 2001; Xu et al. 2001), the
methods have yet to be widely adopted. The DGM we describe here is a mapping system that
would be suitable for most geological purposes. The system involves the integration of three key
technological components; 1) a GPS receiver usually capable of obtaining differential correction
data that enable sub-metre positional accuracy; 2) a PDA or other digital data-logger and 3)
mobile GIS software. Mobile GIS is a specialised version of PDA software that can exchange
Digital mapping: quantifying uncertainty 7 Jones et al 25.06.03 14:51
information with more general purpose desktop GIS. When used in 3D mode we suggest that the
DGM provides an onshore equivalent to a wider definition of digital mapping as used by
petroleum industry to make 3D structural interpretations in sub-surface data.
In a GIS, information is usually displayed as a series of layers that can be superimposed with each
comprising a single type of data. Typically this may comprise features or objects that have distinct
shape and size or field data that vary continuously over a surface (Longley et al. 2001), as
summarised in Table 3.
The advantage of a GIS-based mapping system is that any number of different types of data may
be georeferenced and included as a separate layer in the database. These can then be displayed and
analysed in conjunction with newly acquired field data. A generalised work flow for digital
geological mapping, some typical data inputs and an example GIS database are shown in Fig. 2.
Examples of data that may be included are; regional geophysical maps, aerial photography,
satellite imagery, topographic data, previously digitised geological information, sample
catalogues, geochronological data, geochemical data.
2D, 3-D and 2.5D data
In DGM the data are collected in either 2-D or 3-D modes. If the objective is purely to produce a
map of the region then the point data may be stored using an x and y coordinate and lines and
polygons stored as lists of x and y coordinates. Real-time Differential GPS systems regularly give
precision to approximately 1m in the horizontal plane and survey systems that post-process
positional data can attain cm-scale accuracy. The positional precision and accuracy that may be
achieved using GPS is dependant on variations in the input satellite configuration (an error
summarised by the Dilution of Precision statistic calculated continuously by GPS receivers).
Topography or buildings can limit the number of input satellites available to a GPS receiver and
thus accurate positioning near a cliff or a in a deep valley may be difficult to achieve.
Geological information gained from traditional geological mapping has been displayed on 2-D
representations such as geological maps, however this format has disadvantages as described
above. Whilst GIS software products are often used to produce traditional geological maps the
systems also allow more flexible methods of visualisation that can be easily tailored to individual
requirements. For example, on screen data can be viewed at different scales using the zoom and
pan functions with different combinations of data layers visible as required.
GIS data may be overlain or ‘draped’ onto a digital elevation model, in the form of a surface fitted
to a raster map of elevation values, to produce a display that has been referred to as a 2.5D
representation (Longley et al. 2001). These data may then be displayed using a 3-D viewer that
allows rotation to different vantage points as well as zoom and pan. One particularly useful
Digital mapping: quantifying uncertainty 8 Jones et al 25.06.03 14:51
geological application of 2.5D displays is to study how geological formations and structure are
related to topography (Fig. 3).
For accurate 3-D reconstructions of geological architectures, the z-coordinate or elevation for all
positions is essential. Most mobile-GIS applications allow this to be incorporated in the data
table. Despite GPS having poorer resolution in the z direction, in good conditions, a differential
GPS can give a vertical precision of approximately 1m . Alternatively, 2-D data may be converted
to 3-D by locating the positions onto a digital elevation model, however the resolution is limited
by that of the elevation data (typically 50 m).
The advantages of DGM over traditional mapping include improved time efficiency (especially on
data management, analysis and output); with elevation data DGM is inherently 2.5D or more and
the data have high spatial precision and thus a significant reduction in location error and
uncertainty can be achieved.
DGM is capable of incorporating prior information for the following reasons :-
Scalability in GIS means that an a priori framework, newly collected data, and
interpretation can all be maintained in a single digital model;
Several types of data can be stored together, all tied to their geospatial position within the
GIS model: attribute data, metadata, photos, sketches, ideas, notes, video, speech etc;
Having data in digital format is starting point for quantification of uncertainty (see below).
In order to improve the inclusion of prior information in DGM workflows, more work needs to be
done to integrate the various steps involved in the process of data acquisition, interpretation, and
final model. . User-friendly data gathering methods need to be developed to make it possible for
geologists to capture information in ways that are more intuitive, rather than ways dictated by the
non-flexibility of existing hardware and software.
Quantification of Uncertainty in Geological Mapping
Irrespective of whether a mapper uses traditional or digital methods, the acquisition and
interpretation of field data inevitably involves a wide range of different types of uncertainty
(Table 4). Uncertainties accrue from the onset of data acquisition, and accumulate and propagate
throughout the overall interpretation process. Some sources of uncertainty can readily be
expressed in terms of a quantitative assessment of the precision of measurement for a piece of
equipment (e.g. error tolerance of GPS or clinometer measurements). Other uncertainties can be
reduced and quantified by repeating observations so that a measure of variance can be ascertained
(e.g. variation of dip and strike at a single exposure).
Digital mapping: quantifying uncertainty 9 Jones et al 25.06.03 14:51
Geological mapping tends to produce sparse datasets. This is usually because the amount of
exposure is limited, but even when there are high levels of exposure it is generally impractical to
study all exposed rock in detail. Therefore one of the most important aspects of creating a
geological map involves the interpolation of data to fill the areas between intermittent data points
actual measured. Interpolation below (and above) the surface of the earth is of course also central
in producing 2-D cross-sections and 3-D models. Most GIS systems have in-built geostatistical
analysis tools for the interpolation of geospatial point data across a topographical surface (e.g.
“kriging”), and which also provide a statistical measure of uncertainty across the whole surface
(e.g. “variograms”). The most basic approaches to kriging incorporate rather simplistic probability
distributions, and the values of uncertainty derived are simply based on the sparseness of data.
With simple kriging methods there is generally no opportunity for the involvement of prior
information to prime the interpolation, and the kriged surface that is derived might not always be
geologically realistic. In particular, simple kriging will generally attempt to smooth out sharp
variations between adjacent data points, so is not well suited for areas containing discontinuities
such as faults. More sophisticated approaches such as disjunctive kriging are generally more
suitable for most geological interpolations. Future work should concentrate on developing
workflows that makes it easier for the geologist to influence the interpolation process by
specifying prior inputs that will pre-condition the kriging.
Qualitative uncertainties
Many types of uncertainty are difficult to express quantitatively, and are more suited to a
qualitative evaluation by the geologist. Although this may be abhorrent to inductivists that believe
that science consists only of quantitative, objective measurement, a subjective statement such as
“this rock looks sheared and I am reasonably confident that it is a mylonite” is a more useful and
representative observation statement than having to make a binary choice between “this is a
mylonite” and “this is not a mylonite”. An obvious strategy to tackle this situation would be for
field geologists to specifically record an estimate of confidence with every observation, as a
matter of routine. However, most geologists will perceive this data as superfluous, and gathering
it as an additional, unnecessary burden, because there is a general lack of methodology developed
within geological mapping which allows such information to be used in a systematic way.
The potential now exists for this situation to change, following recent advances in mathematics
and computer technology, especially within several branches of artificial intelligence (AI). In
particular, developments in Fuzzy Logic provide a formal framework in which relative terms
(such as “quite sheared”, “very fractured”) can be transformed to discrete numerical values and
represented within binary computer code. Other branches of AI (e.g. Bayesian networks, neural
Digital mapping: quantifying uncertainty 10 Jones et al 25.06.03 14:51
networks, genetic algorithms, constraint satisfaction techniques) have potential for finding
solutions for complicated non-linear models involving very many variables. All these
methodologies have a proven track record of finding good solutions to real problems with a much
shorter amount of computer processing than traditional approaches.
Future Trends in Geological Mapping
For the last two decades the growth in Information Technology (IT) has generally been so great
that geologists have struggled to keep abreast of technological advances. Within geological
mapping earth scientists have been slow to improve workflows and methods of interpretation that
exploit the newly developing technologies. There is no indication that the current rate of growth
within IT is set to diminish, and the following trends are likely to provide improved opportunities
for geoscientists to improve the process of geological mapping:
portable equipment will continue to become lighter, cheaper, more robust, more powerful,
more intuitive and user-friendly, and more integrated with the user
an increased number of 2-D and 3-D analytical tools will be incorporated into existing GIS
software to provide an single integrated tool for geological mapping and interpretation.
interpretation tools will propagate information about uncertainties through the modelling
process so that various interpretations can be tested in parallel and an indication of overall
uncertainty can be given for each interpretation.
satellite communications technology combined with GRID facilities will bring super-
computing power to field geologists (a “PersonalGRID”). This will increase the possibilities
for ongoing iterative interpretation of field data whilst still in the field.
speech recognition software combined with semantic based search technology can help to
encourage the geologist to verbalise (= make explicit) more of the decision-making processes
involved in mapping.
Although traditional processes of geological mapping have a proven track record established over
two hundred years, there are nevertheless important methodological shortcomings seen from a
scientific perspective:
paper-based published maps generally show only a fraction of the field data that have been
collected and used as the basis for the map’s creation. Other data remain hidden in the field
notebook and in the head of the mapper.
Digital mapping: quantifying uncertainty 11 Jones et al 25.06.03 14:51
published maps rarely make reference to the reasoning used during interpretation of the basic
field data. Reasoning typically relies heavily on prior information and knowledge that
represents the experience of the geologist before the onset of the mapping project.
paper-based maps are by necessity published at a fixed scale. The skill of the cartographer is to
present as much relevant information as possible whilst maintaining legibility at the chosen
scale, but inevitably there is a loss of precision especially with respect to geospatial
with traditional maps there are generally only very limited possibilities for expressing any
uncertainty concerning the given interpretation, and these are not quantitative.
Whilst the above limitations have always been acceptable as long as there were no viable
alternatives to paper-based publishing, today’s information technology makes it possible to store
the entire life’s work of a geologist on a compact disc that costs less than a pair of boot laces!
Digital mapping has the potential to improve the scientific validity of the mapping process in the
following ways:
collected field data can be stored together with the interpreted map in a single digital model
within a GIS. In the future it should be possible to capture and store a wider range of
multimedia datatypes (including metadata, photos, sketches, ideas and notes, video, speech
etc.) all tied to the appropriate geospatial position within the GIS model
prior inputs used as the basis for interpretations can be stated explicitly within the same GIS
the inherent scalability of a GIS model makes it possible to store unlimited data for any
geospatial locality, so that the amount of data is no longer restricted by the scale of the final
published map
as progressively more analytical tools are incorporated into GIS software, geospatially
referenced data can be analysed and interpreted within a single software environment
many types of uncertainty that arise during the mapping process can either be quantified or can
be estimated qualitatively in a way that can be represented digitally (using AI techniques).
Digital geological mapping is still in its infancy. Future work should concentrate on the following
the digital workflow should be continuously improved to make it more intuitive and quicker to
capture field data in digital format
more integration of analysis tools and GIS is needed
Digital mapping: quantifying uncertainty 12 Jones et al 25.06.03 14:51
methodologies should be further developed that use the uncertainties associated with
individual data or interpretations as the input to produce an overall estimate of uncertainty
associated with a given model. Alternative interpretations can be modelled simultaneously,
with uncertainty calculated for each
efforts should be made to combine portable GIS with GRID technology.
Acknowledgements __ Thanks to Jonny Imber, Nicola De Paola, Phil Clegg and other members of
RRG and Bernt Bremdal at CognIT for fruitful debate. Steve Freeman and others at RDR made
early progress with DGM. Much of this work was part of the Ocean Margins Link project
(NER/T/S/2000/01018), funded by Statoil UK Ltd, BP and NERC.
Barnes, J.W. 1981. Basic Geological Mapping. Geological Society of London Handbook Series,
Open University Press, 112pp.
Boose, J H. 1986. Expertise Transfer for Expert System Design. Elsevier, New York.
Bonham-Carter, G. 2000. An overview of GIS in the geosciences. In T.C. Coburn and J.M. Yarus,
(eds) Geographic information systems in petroleum exploration and development. AAPG
Computer Applications in Geology, 4, 17-26.
Coburn, T.C. & Yarus, J.M. 2000. Geographic information systems in petroleum exploration and
development. AAPG Computer Applications in Geology No. 4, 315pp.
Ford, D. & Sterman, J. 1998. Expert Knowledge Elicitation for Improving Mental and Formal
Models. System Dynamics Review. 14, 309-340.
Kidd, A.L (ed.) 1987. Knowledge Elicitation for Expert Systems: A Practical Handbook. Plenum
Press, New York.
Kuhn, T. S. 1962. The Structure of Scientific Revolutions. University of Chicago Press, Chicago.
Lakatos, I. 1974. Falsification and the Methodology of Scientific Research Programmes. In:
Lakatos, I & Musgrave, A. (eds) Criticism and the Growth of Knowledge. Cambridge
University Press, Cambridge, 91-196.
Liebowitz, J. & Beckman, T. 1998. Knowledge Organizations. CRC Press, Florida.
Longley, P.A., Goodchild, M.F., Maguire, D.J. & Rhind, D.W. 2001. Geographic Information
Systems and Science. Wiley & Sons Ltd, Chichester.
Maerten, L., Pollard, D.D. & Maerten, F. 2001. Digital Mapping of three dimensional structures
of the Chimney Rock fault system, central Utah. Journal of Structural Geology, 23, 585-
Mallet, J.L. 1992. GOCAD: A computer aided design program for geological applications. In
Turner, A.K. (ed) Three-Dimensional Modeling with Geoscientific Information Systems.
NATO ASI Series C: Mathematical and Physical Sciences, vol. 354. Kluwer Academic
Publishers, Dordrecht, The Netherlands, 123-141.
Digital mapping: quantifying uncertainty 13 Jones et al 25.06.03 14:51
McClay, K.R. 1987. The Mapping of Geological Structures. Geological Society of London
Handbook Series, Open University Press, 161pp.
Nonaka, I. 1994. A Dynamic Theory of Organizational Knowledge Creation. Organization
Science, 5, 14-37
Nonaka, I. & Takeuchi, H. 1995. The knowledge-creating company: how Japanese companies
create the dynamics of innovation. Oxford University Press.
Polanyi, M. 1958. Personal Knowledge: Towards a Post-Critical Philosophy. Routledge &
Kegan Paul, London.
Polanyi, M. 1966. The Tacit Dimension. Routledge & Kegan Paul, London
Ramsay, J.G. & Huber, M.I. 1987. The Techniques of Modern Structural Geology. Volume 2:
Folds and Fractures. Academic Press, 700pp.
Rhind, D.W. 1992. Spatial data handling in the Geosciences. In Turner, A.K. (ed) Three-
Dimensional Modeling with Geoscientific Information Systems. NATO ASI Series C:
Mathematical and Physical Sciences, vol. 354. Kluwer Academic Publishers, Dordrecht,
The Netherlands, 13-27.
Smith, W. 1801. General Map of Strata in England and Wales.
Smith, W. 1815. A Delineation of the Strata of England and Wales, with part of Scotland.
Turner, A.K. 1992. Three-Dimensional Modeling with Geoscientific Information Systems. NATO
ASI Series C: Mathematical and Physical Sciences, vol. 354. Kluwer Academic Publishers,
Dordrecht, The Netherlands, 443pp.
Turner, A.K. 2000. Geoscientific modeling: past, present and future. In T.C. Coburn and J.M.
Yarus, (eds) Geographic information systems in petroleum exploration and development.
AAPG Computer Applications in Geology, 4, 27-36.
White, S. & Sleeman, D. 1999. A Constraint-Based Approach to the Description of Competence.
In: Proceedings of EKAW-99, Lecture Notes in AI. Springer Verlag.
van der Spek, R. & Spijkervet, A. 1997. Knowledge Management: Dealing Intelligently with
Knowledge. CIBIT, Utrecht.
Xu, X, Battacharya, J.A., Davies, R.K. & Aiken, C.L.V. Digital Geologic Mapping of the Ferron
sandstone, Muddy Creek, Utah, with GPS and reflectorless Laser Rangefinders. GPS
Solutions, 5, 15-23.
Figure Captions
Fig. 1. Traditional inductivist view of scientific methodology.
Fig. 2. a) Digital Geological Mapping work flow indicating the relationship between different
activities, b) some input data types, c) a display from a GIS database for Achmelvich Bay, NW
Fig. 3. Digital geological mapping of part of the foreland to the Moine Thrust at Loch Assynt, N.
Scotland. a) Map with boundaries displayed on aerial photograph, b) oblique 2.5D view from
south of same data showing that boundary A is sub-horizontal and boundary B dips to the east.
Digital mapping: quantifying uncertainty 14 Jones et al 25.06.03 14:51
Fig. 1.
Fig. 2.
observation = fact scientific theory prediction
induction deduction
Digital mapping: quantifying uncertainty 15 Jones et al 25.06.03 14:51
Fig. 3.
Level Example
5 Expertise (= ability to apply knowledge
effectively to perform tasks accurately and
efficiently within existing constraints)
the finished geological map!
a geological summary of the mapped area
scientific paper describing new fossil found in area
4 Knowledge (= application of information
to be able to make decisions, solve
problems, or perform tasks)
boundary between rock types A and B is in valley
rock A is granite
mineral P is biotite
3 Information (= data in a specified context,
data with meaning) rock A is observed at location GR 234 789
rock A contains minerals P, Q, R
mineral P has colour that is black
2 Data (= inputs that can be represented
explicitly in symbolic form) colour is black
lustre is shiny
GPS location at time T is 123456 from satellite 1
GPS location at time T is 123678 from satellite 2
1 Inputs (= sensory signals, machine
measurements) geologist receives visual sensory signals
GPS unit receives satellite signals
Table 1. The knowledge hierarchy with examples from geological mapping
Knowledge accessibility General examples Examples from mapping
Explicit scientific papers published in journals.
instruction manuals.
detailed outcrop map (“green-line”
Implicit unpublished observations.
working hypotheses.
undocumented troubleshooting fixes.
observations not recorded explicitly
on final map.
general geological theory that has
influenced specific interpretation.
Tacit instant recognition of minerals based on “look
and feel” rather than explicit physical tests.
interpretation of seismic sections.
expertise of the mapper.
the mapper’s preconceived bias,
insight, gut-feeling and intuition.
Table 2. Different levels of accessibility of knowledge
Digital mapping: quantifying uncertainty 16 Jones et al 25.06.03 14:51
GIS data Type Geological data Specific example
Point Object Locations of structural stations,
anywhere a measurement is made
or a sample is taken
Bedding strike & dip
Gravity measurement
Geochemical sample
Line Object Boundaries between areas or
linear objects Contact between rock units
Fault trace
Polygon Object Areas of rock units Formation extent
Area of igneous intrusion
Raster Field data sampled on a matrix of
equally sized squares Elevation (Digital Elevation
LandSat™ image
Table 3. Data types in a typical GIS system
Level Type of uncertainty Examples
Positional “how sure am I of my current location?”
“how reliable is my base map?”
“what is the precision of my GPS measurements?”
“is the borehole straight, or has it deviated without me knowing?”
Measurement “what is the precision of my clinometer?”
“what is the accuracy of my dip/strike readings?”
Scale-dependant variability “how much does the dip and strike vary over the scale of the outcrop?”
“is my reading representative of the surrounding area?”
Observational “is this rock best described as a granite?”
“is this fossil the brachiopod pentamerus?”
“is that a stretching lineation or an intersection lineation?”
Temporal “how reliable is this way-up criteria?”
“is the relative age of these structures identified correctly?”
Data acquisition
Sampling bias “is my data biased by the natural predominance of sub-horizontal exposures?”
“has my sampling been skewed by me focusing only on the zones of high strain?
Correlation “is this limestone the same unit as the limestone at the last outcrop?
“is it valid to correlate the S2 fabric here with the S2 fabric observed on the other side
of the area?”
Interpolation “how likely is it that all the ground between these two outcrops of slate also consists
of slate?”
“how much control do I have over the geometry of this fold trace?”
Primary interpretation
Inference from topography “is there really a fault running along the unexposed valley floor?”
“does this sharp change in slope correspond to a lithological boundary?”
Finished 2D map
Geological cross-section
3D structural model
“how can I quantify the uncertainty associated with this sophisticated interpretive
model that I have slowly built up through a long iterative process of data collection
and individual primary interpretations?!”
Table 4. Examples of different types of uncertainty arising during geological mapping.
... The Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER), Landsat series of satellites, Hyperion, and Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) are extensively utilized in lithological mapping (Gad and Kusky 2007;Pour and Hashim 2014a;Kumar et al. 2015;Guha et al. 2021). However, lithological or rock type classification becomes more challenging where the geological site consists of several rock types of similar mineral composition and weathered surfaces, which diminish the intra-rock variability (Jones et al. 2004). A few recent studies demonstrated the advantages of integrating physical characteristics such as morphological and textural information with spectral properties of surface for improving the intra-rock variability and classification accuracy (Othman and Gloaguen 2014;Wei et al. 2016;Masoumi et al. 2017;Othman and Gloaguen 2017). ...
... The lithological classification becomes a more challenging task in a complex geological setting and inadequate rock exposure (Jones et al. 2004). Furthermore, homogeneity in major mineral composition of different rock types and weathered surface cover makes rocks' spectral characteristics less distinctive and illustrative to be used as primary input features in the lithological classification. ...
We present an optimal integration of multi-sensor datasets, including Advanced Spaceborne Thermal and Reflection Radiometer (ASTER), Phased Array type L-band Synthetic Aperture Radar (PALSAR), Sentinel-1, and digital elevation model for lithological classification using Machine Learning Models (MLMs). Different input features such as spectral, spectral and transformed spectral, spectral and morphological, spectral and textural, and optimum hybrid features were derived and evaluated to accurately classify different rock types found in the Chhatarpur district (Madhya Pradesh), India using the Support Vector Machine (SVM) and Random Forest (RF). The SVM achieves better classification accuracy and shows less sensitivity to the number of samples used in model development. The optimum hybrid features outperform other input features with an overall accuracy and κ coefficient of 77.78% and 0.74, which is around 15% higher as obtained using ASTER spectral data alone. Thus, the proposed multi-sensor optimal integration approach is recommended for successful lithological classification using MLMs.
... The central role of geographic information systems (GIS) in the creation of geospatial databases is an undeniable fact [3][4][5]. GIS applications flourished in the 1980s when the first commercial GIS packages were released, such as ARC/INFO developed by the Environmental System Research Institute (ESRI), and then GRASS (geographic resources analysis support system), the first free and open-source software (FOSS) for GIS, was made available [6]. The emergence of web and mobile technologies inherently fueled the development of collaborative web mapping and volunteered geographic information (VGI). ...
Full-text available
A collaborative open-source IT infrastructure is designed and implemented to optimize the process of geological field data collection, integration, validation, and sharing. Firstly, field data collection is carried out by multiple users using free and open-source GIS-based tools for mobile devices according to a predefined database structure; then, data integration is automatically performed in a central server, where the collected geological information is stored and validated; finally, data are shared over the Internet, providing users with up-to-date information. The IT infrastructure is currently being employed to accomplish surveys for the realization of the “Brescia” geological map within the New Geological Map of Italy, scale 1:50.000 (CARG Project). Users are only required to run the field data collection application on their mobile devices, add different geometric features to predefined thematic layers and fill in the dialogue forms with the required information to store the new structured and georeferenced data in the central database. The major advantage of the proposed IT infrastructure consists of guaranteeing the operational continuity between field surveys and the finalization of geological or geothematic maps leveraging field data collection tools that are operational both online and offline to ensure the overall system resilience.
... Data acquisition methods, however, may affect DOM resolution, that in turn affect the accuracy and precision of derived measurements (Casini et al., 2016;Cawood et al., 2017;James et al., 2017). Thus, their efficacy, like traditional field mapping (Jones et al., 2004), may be limited by several types of error and uncertainties thereof (Buckley et al., 2008;Rarity et al., 2014). ...
In this study, we examine the errors and uncertainties associated with orientation measurements collected from digital outcrop models using the geometrical property, collinearity. Collinearity is expressed as the characteristics of a set of points lying on a single straight line, and is beneficial because a trace far from collinear is required for obtaining accurate orientation measurements of planar geological bodies from digital outcrop models. We, thus, demonstrate this relationship, as well as assess the impact any associated errors have on orientation distribution forms using orientation measurement traces collected from sandstone intrusions with a digital outcrop model of the Panoche Giant Injection Complex acquired through light detection and ranging and photogrammetry techniques. Our experiments highlight how in addition to sampling bias and sample size, unreliable orientation estimates can negatively affect interpretation. We show that the distribution of orientation for a network of geological structures (e.g., fractures, sandstone intrusions) in a particular region can be altered with the inclusion of erroneous measurements. From our case study, it was noted that the high proportion of unreliable orientation measurements obtained when using a digital outcrop model of the study area resulted in inconsistencies in the structural analysis that could have been attributed to other factors. Thus, without putting into consideration the reliability of the samples, a sampling issue will still be faced, regardless of using the appropriate sample size and technique.
... Various attempts have been made to utilize these early digital data capture systems to assist geologists in their routine fieldwork (e.g. Brimhall, 1998;Brimhall and Vanegas, 2001;Jones et al., 2004;McCaffrey et al., 2005;De Donatis and Bruciatelli, 2006;Clegg et al., 2006;Pavlis et al., 2010). The introduction of the Apple iPad in 2010 and the subsequent proliferation of similar tablet devices utilizing the Android operating system represented another major step forward in the digitization of fieldwork (e.g. ...
Major advances in smartphones and tablets in terms of their built-in sensors (esp. cameras), available computational power and on-board memory are transforming the role of such devices into the key digital platform around which geological fieldwork is redesigning itself. This digital transition is changing how geoscientists collect and share multimodal-multidimensional field datasets, which can now be readily distributed via standardized exchange formats and online data repositories. The increased accessibility of digital field datasets means that such data products are no longer the sole preserve of geospatial/geoscience specialists, but also students, stakeholders and the general public alike, providing great opportunities for knowledge transfer over the entirety of the research value chain. In the wake of this digital transition, the geological community has welcomed with enthusiasm and curiosity the introduction during 2020 of a native LiDAR scanner equipped on both the iPad Pro and the iPhone 12 Pro. This scanner offers a potential paradigm shift in digital geological fieldwork and puts these devices at the forefront of smartphone assisted fieldwork. In this work, we review progress in smartphone/tablet assisted geological fieldwork and test the iPhone 12 Pro's effectiveness as a replacement for conventional geological field tools. Specifically, we evaluate the geo-location accuracy of the iPhone's Global Navigation Satellite System (GNSS) receiver, the effectiveness of its inertial measurement unit (IMU) and magnetometer for orientation data collection, its photo-video imaging capabilities, and the performance of the device's newly equipped LiDAR in the field. We demonstrate that the performance of the iPhone for orientation and raster image data capture is high, being comparable to analog compass-clinometers and reflex/mirrorless cameras. Whilst location error is within the order of a few meters, the level of accuracy and the fast stabilization of the signal means that, beyond survey grade applications, the iPhone's geo-location capabilities are acceptable for most field cases. With regards to the iPhone's built-in LiDAR scanner, it is an excellent tool for depth assisted camera focusing and for casual 3D outcrop sharing, especially for ‘soft’ applications such as geo-heritage documentation and the production of teaching materials (here we also propose a simple mode of uploading outcrops models in Google Maps). However, the generated 3D models in some cases may be considered overly crude for detailed interrogation, particularly where the fidelity of the surface reconstruction is critical to the analysis (e.g. mesh facet orientation estimation). Based on our review of the evolution of digital field acquisition technologies and on our extensive field testing of the sensor suite integrated within the iPhone 12 Pro, it is clear that the digital transition of geological fieldwork is already mature, whereby smartphone devices have become as indispensable in the field as the geologists' traditional hammer and hand lens.
... Celles-ci peuvent être de différents types, à différentes échelles (Figure 1.3), d'une valeur plus ou moins quantitative [Koch and Link, 2002]. Leur information peut être bruitée, leur qualité plus ou moins grande, leurs incertitudes respectives peuvent avoir différents ordres de grandeur [Jones et al., 2004]. Leur distribution spatiale est également parfois très inégale. ...
Full-text available
Les méthodes de modélisation géologique 3D ont pour but de construire des modèles numériques cohérents du sous-sol à partir de données ponctuelles échantillonnées sur le terrain ou en profondeur. Très populaires de nos jours, les méthodes dites implicites permettent de construire plusieurs champs scalaires agencés les uns par rapport aux autres et construits à partir de données de contacts d’unités géologiques et de leurs orientations respectives. Les surfaces géologiques sont ensuite extraites comme iso-potentielles de ces champs. Dans ce cadre, la Méthode du Potentiel, proposée il y a plus de 20 ans par l’École des Mines et le BRGM, utilise les outils géostatistiques, comme l’interpolation par co-krigeage, afin de reconstruire ces champs scalaires et ces surfaces. Bien qu’éprouvées, ces méthodes de modélisation montrent encore leurs limites face à quelques modèles complexes. Certaines structures géologiques, telles que les plis non cylindriques, les filons minéralisés ou les réseaux fluviatiles par exemple, présentent une structuration suivant une direction préférentielle (anisotropie) clairement identifiable localement mais variant spatialement. Très souvent, le nombre ou la répartition des données disponibles initialement ne permet pas de caractériser cette anisotropie variable correctement. Ces travaux de thèse visent à pallier ce manque en intégrant cette anisotropie comme donnée d’entrée au sein de la modélisation. Pour ce faire, deux approches ont été développées :(1) Une première approche exploite les données de dérivées premières (tangentes) ou dérivées secondes du champ de potentiel, permettant de contraindre localement l’anisotropie du champ scalaire. Cette approche est développée dans le cadre de la modélisation de plis poly-phasés, exemple emblématique de la problématique de l’anisotropie variable et de la nécessité d’action sur la courbure de surfaces. L’apport et l’usage de chacun des types de données sont comparés et discutés dans ce cadre.(2) Une seconde approche plus globale interprète le potentiel comme la convolution d’un bruit blanc par un noyau gaussien. Cette méthode permet d’introduire une expression de l’anisotropie sous forme explicite, pouvant être interpolée depuis des données d’anisotropie échantillonnées ou construite comme un a priori géologique. Enfin, le contexte de déploiement respectif de l’une ou l’autre approche développée est discuté en regard du cas d’application considéré.
... In order to address these challenges, the use of massive borehole data and the digitalization of geological mapping in a more general sense (Jones et al. 2004) and, more concretely, geological 3-D models, in their different forms and approaches, are gaining importance over the last decadesnot only for visualization purposes but also as a planning and forecasting tool. This is evidenced by the fact that geological mapping is currently undergoing a transformation from traditional 2-D to 3-D, with geological 3-D modeling becoming a key priority for national geological surveys (NGS) (Berg et al. 2011). ...
Full-text available
The coexistence of a wide variety of subsurface uses in urban areas requires increasingly demanding geological prediction capacities for characterizing the geological heterogeneities at a small-scale. In particular, detrital systems are characterized by the presence of highly varying sediment mixtures which control the non-constant spatial distribution of properties, therefore presenting a crucial aspect for understanding the small-scale spatial variability of physical properties. The proposed methodology uses the lithological descriptions from drilled boreholes and implements sequential indicator simulation to simulate the cumulative frequencies of each lithological class in the whole sediment mixture. The resulting distributions are expressed by a set of voxel models, referred to as Di models. This solution is able to predict the relative amounts of each grain fraction on a cell-by-cell basis and therefore also derive a virtual grain size distribution. Its implementation allows the modeler to flexibly choose both the grain fractions to be modeled and the precision in the relative quantification. The concept of information entropy is adapted as a measure of the disorder state of the clasts mixture, resulting in the concept of “Model Lithological Uniformity,” proposed as a measure of the degree of detrital homogeneity. Moreover, the “Most Uniform Lithological Model” is presented as a distribution of the most prevailing lithologies. This method was tested in the city of Munich (Germany) using a dataset of over 20,000 boreholes, providing a significant step forward in capturing the spatial heterogeneity of detrital systems and addressing model scenarios for applications requiring variable relative amounts of grain fractions.
... Attempting to reduce impacts to the final model, some methods focus on the variability of uncertainty introduced by humans or computation during data collection, processing, and representation (Thore et al., 2002;Jones et al., 2004;Tacher et al., 2006;Lindsay et al., 2012). Some other methods attempt to quantify impacts of errors and uncertainties in final 3D models (Caumon et al., 2009;Jessell et al., 2010;Wellmann et al., 2010;Wellmann and Regenauer-Lieb, 2012). ...
Full-text available
To visualize and analyze the impact of uncertainty on the geological subsurface, on the term of the geological attribute probabilities (GAP), a vector parameters-based method is presented. Perturbing local data with error distribution, a GAP isosurface suite is first obtained by the Monte Carlo simulation. Several vector parameters including normal vector, curvatures and their entropy are used to measure uncertainties of the isosurface suite. The vector parameters except curvature and curvature entropy are visualized as line features by distributing them over their respective equivalent structure surfaces or concentrating on the initial surface. The curvature and curvature entropy presented with color map to reveal the geometrical variation on the perturbed zone. The multiple-dimensional scaling (MDS) method is used to map GAP isosurfaces to a set of points in low-dimensional space to obtain the total diversity among these equivalent probability surfaces. An example of a bedrock surface structure in a metro station shows that the presented method is applicable to quantitative description and visualization of uncertainties in geological subsurface. MDS plots shows differences of total diversity caused by different error distribution parameters or different distribution types.
... These developments are also closely tied to major methodological improvements for virtual outcrop model (VOM) interpretation. All these advancements have accelerated the use of digital outcrop data capture and analysis in field geology, transforming what was principally a visualization medium into fully interrogatable quantitative geo-data objects (Jones et al., 2004;Bemis et al., 2014;Howell et al., 2014;Hodgetts et al., 2015;Biber et al., 2018;Bruna et al., 2019;Caravaca et al., 2019;Thiele et al., 2019;Triantafyllou et al., 2019). Initially, close-range remote-sensing studies seeking to reconstruct and analyze rock outcrops were dominantly built around terrestrial laser scanning systems (terrestrial lidar), which became commercially available around two decades ago (e.g., Bellian et al., 2002). ...
Full-text available
Since the advent of affordable consumer-grade cameras over a century ago, photographic images have been the standard medium for capturing and visualizing outcrop-scale geological features. Despite the ubiquity of raster image data capture in routine fieldwork, the development of close-range 3D remote-sensing techniques has led to a paradigm shift in the representation and analysis of rock exposures from two- to three-dimensional forms. The use of geological 3D surface reconstructions in routine fieldwork has, however, been limited by the portability, associated learning curve, and/or expense of tools required for data capture, visualization, and analysis. Smartphones are rapidly becoming a viable alternative to conventional 3D close-range remote-sensing data capture and visualization platforms, providing a catalyst for the general uptake of 3D outcrop technologies by the geological community, which were up until relatively recently the purview of a relatively small number of geospatial specialists. Indeed, the continuous improvement of smartphone cameras, coupled with their integration with global navigation satellite system (GNSS) and inertial sensors provides 3D reconstructions with comparable accuracy to survey-grade systems. These developments have already led many field geologists to replace reflex cameras, as well as dedicated handheld GNSS receivers and compass clinometers, with smartphones, which offer the equivalent functionality within a single compact platform. Here we demonstrate that through the use of a smartphone and a portable gimbal stabilizer, we can readily generate and register high-quality 3D scans of outcropping geological structures, with the workflow exemplified using a mirror of a seismically active fault. The scan is conducted with minimal effort over the course of a few minutes with limited equipment, thus being representative of a routine situation for a field geologist.
... In recent decades, the approaches of digital geological mapping (DGM) have attracted increasing global attention. Many computer scientists and geologists have worked together to build a data framework of DGM and develop hard-and software systems to facilitate field geological survey and education [2][3][4][5][6][7][8][9][10][11][12][13][14][15]. ...
Full-text available
The development of innovative information technologies has improved the geological mapping process through the use of smart and portable equipment to collect field data, build a geological database and produce geological maps. This revolution has also brought great influence and challenges to practical training in field geology. In this paper, we present our workflow and application of the Digital Geological Survey System (DGSS) during field geology training for undergraduates in Zhoukoudian. The DGSS employs a Point-Routing-Boundary (PRB) model to reform the methods of geological section survey and geological mapping in terms of data collection and map-making and provides a pipelined solution from field data collection to map-making. The experiences of data collection, geological mapping, cross-section survey, and production of stratigraphic histograms and cross-section maps prove that DGSS can save time and reduce labor intensity for undergraduates during learning field geology. Based on the field practice of undergraduates in Zhoukoudian, the influence of the DGSS in promoting field geological teaching and the students’ feedbacks to DGSS are discussed. Overall, the DGM system is more popular than the conventional notebook and toolbox. The experience in Zhoukoudian proves that digital devices are efficient and useful for geological practical training of field geology for undergraduates.
Purpose A principle prerequisite for designing and constructing an underground structure is to estimate the subsurface's properties and obtain a realistic picture of stratigraphy. Obtaining direct measure of these values in any location of the built environment is not affordable. Therefore, any evaluation is afflicted with uncertainty, and we need to combine all available measurements, observations and previous knowledge to achieve an informed estimate and quantify the involved uncertainties. This study aims to enhance the geotechnical surveys based on a spatial estimation of subsoil to customised data structures and integrating the ground models into digital design environments. Design/methodology/approach The present study's objective is to enhance the geotechnical surveys based on a spatial estimation of subsoil to customised data structures and integrating the ground models into digital design environments. A ground model consisting of voxels is developed via Revit-Dynamo to represent spatial uncertainties employing the kriging interpolation method. The local arrangement of new surveys are evaluated to be optimised. Findings The visualisation model's computational performance is modified by using an octree structure. The results show that it adapts the structure to be modelled more efficiently. The proposed concept can identify the geological models' risky locations for further geological investigations and reveal an optimised experimental design. The modifications criteria are defined in global and local considerations. Originality/value It provides a transparent and repeatable approach to construct a spatial ground model for subsequent experimental or numerical analysis. In the first attempt, the ground model was discretised by a grid of voxels. In general, the required computing time primarily depends on the size of the voxels. This issue is addressed by implementing octree voxels to reduce the computational efforts. This applies especially to the cases that a higher resolution is required. The investigations using a synthetic soil model showed that the developed methodology fulfilled the kriging method's requirements. The effects of variogram parameters, such as the range and the covariance function, were investigated based on some parameter studies. Moreover, a synthetic model is used to demonstrate the optimal experimental design concept. Through the implementation, alternative locations for new boreholes are generated, and their uncertainties are quantified. The impact of the new borehole on the uncertainty measures are quantified based on local and global approaches. For further research to identify the geological models' risky spots, the development of this approach with additional criteria regarding the search neighbourhood and consideration of barriers and trends in real cases (by employing different interpolation methodologies) should be considered.
Building an expert system involves eliciting, analyzing, and interpreting the knowledge that a human expert uses when solving problems. Expe­ rience has shown that this process of "knowledge acquisition" is both difficult and time consuming and is often a major bottleneck in the production of expert systems. Unfortunately, an adequate theoretical basis for knowledge acquisition has not yet been established. This re­ quires a classification of knowledge domains and problem-solving tasks and an improved understanding of the relationship between knowledge structures in human and machine. In the meantime, expert system builders need access to information about the techniques currently being employed and their effectiveness in different applications. The aim of this book, therefore, is to draw on the experience of AI scientists, cognitive psychologists, and knowledge engineers in discussing particular acquisition techniques and providing practical advice on their application. Each chapter provides a detailed description of a particular technique or methodology applied within a selected task domain. The relative strengths and weaknesses of the tech­ nique are summarized at the end of each chapter with some suggested guidelines for its use. We hope that this book will not only serve as a practical handbook for expert system builders, but also be of interest to AI and cognitive scientists who are seeking to develop a theory of knowledge acquisition for expert systems.
Knowledge acquisition (KA) is a crucial stage in the development of an expert system. As a process, it involves eliciting, analyzing, and interpreting the knowledge that a human expert uses when solving a particular problem and then transforming this knowledge into a suitable machine representation. KA is critical since the power and utility of the resulting expert system depend on the quality of the underlying representation of expert knowledge. The aim of this book is to provide the builders of expert systems with some practical advice on what KA involves and some of the methodologies and techniques that have been developed to aid its effectiveness.
Achievements of the Workshop definition of the problem existing three-dimensional geoscientific information systems three-dimensional data structures and display methods applications of three-dimensional geoscientific modelling transcriptions of conference committee discussions.
The GOCAD research program is specially devoted to the Geometric Modelling of complex geological objects in connection with Geophysical, Geological and Reservoir Engineering applications. It is based on a new interpolation technique called "Discrete Smooth Interpolation' (DSI) designed to account for the heterogeneous and imprecise data encountered in geology. The basic principle of the DSI method and the main specifications of the associated GOCAD software are presented. -Author
This chapter emphasizes application of expert systems for reference and locational maps. The broader question of expert systems for the design of more abstract maps, such as statistical maps, has seldom been addressed in the literature. It seems clear that an expert system for design of reference maps, locational maps, and some navigation aids may be more feasible in the short term than a system to generate locational maps as well as purely thematic and analytical graphics. In part this is because the symbology of locational maps is more standardized and thus design constraints may be more readily identified. The specificity of map purpose for locational maps also places finite bounds on the range of appropriate symbology, without limiting map projection, scale, use of color, and level of generalization. The purpose of this chapter is to present design criteria for a cartographic expert system that focuses on the problems of map execution, including compilation and production of base features and thematic overlays of locational information. Expert systems and artifical intelligence for other cartographic processes, such as digital scanning, or knowledge-based geographic information systems will not be considered. The discussion remains on a conceptual level. The components of a system and the required functionalities are identified, and the relationships between the two discussed. Progress to date on each phase of the cartographic process will be assessed through a review of the literature. First an overview of expert systems design is given. -from Authors