Trinks 2005 Virtual Outcrops
Trinks 2005 Virtual Outcrops
Trinks 2005 Virtual Outcrops
DOI 10.1007/s10069-005-0026-9
O R I GI N A L A R T IC L E
Abstract Laser scanning is a very ecient way to generate realistic, high-resolution digital models of 3-D
geological outcrops. This paper discusses the methodologies involved in the creation and analysis of virtual
outcrops, based on laser scanner data. The visualisation
of the laser scanner data as a photorealistic 3-D object is
described. Geological features picked out on the virtual
outcrop (e.g. fractures, faults or bedding planes) can be
extrapolated outward, into space, and inward, into the
subsurface, using tension surfaces.
Keywords Visualisation Outcrop Laser
scanning Virtual reality
Introduction
Traditional methods of mapping and analysing geological outcrop data are based on a combination of geological surveying, photo-mosaics, cairn mapping,
structural eld data, stereo plots, rose diagrams and 3-D
Electronic Supplementary Material Supplementary material is
available in the online version of this article at http://dx.doi.org/
10.1007/s10069-005-0026-9
Reviewed by: J.D. Clemens, D. Yuen
I. Trinks (&) R. Jones R. Hobbs N. Holliman
e-Science Research Institute, University of Durham,
DH1 3LE, Durham, UK
E-mail: [email protected]
P. Clegg B. Holdsworth J. Imber R. Wilson
Reactivation Research Group, Dept. of Earth Sciences,
University of Durham, DH1 LE, Durham, UK
E-mail: [email protected]
K. McCarey R. Jones
Geospatial Research Ltd., Dept. of Earth Sciences,
University of Durham, DH1 3LE, Durham, UK
S. Waggott
Halcrow Group Ltd, Arndale House, Otley Road,
LS6 2UL Leeds, UK
block diagrams. Not only are these methods time consuming, but, in addition, the nal output is generally in
the form of a series of two-dimensional, largely paperbased images that attempt to depict a three-dimensional
world. This approach can make it dicult to adequately
visualise and analyse the spatial and temporal relationships between dierent geological structures, requiring
the geologists to develop a 3-D and often 4-D (temporal)
picture within their imagination. This must then be
transcribed into some form of representative model,
such as a block diagram or cartoon. This process is
highly subjective and relies on the skill and ability of the
geoscientist to imagine the observed data and to reproduce it in a form presentable to others (e.g. see Jones
et al. 2004).
Increasingly, geologists are utilising the benets of
digital mapping technology (e.g. Xu et al. 2000, Edmondo2002, McCarey et al. 2003, Jones et al. 2004,
Clegg et al. 2005, McCarey et al. 2005, Wilson et al.
2005) whereby data is captured using Global Positioning
Systems (GPS) and laser scanning equipment, and
viewed and analysed in Geographical Information Systems (GIS). These new technologies permit fast and
accurate data acquisition, and provide additional
important geospatial information by allowing a variety
of data types to be viewed simultaneously. However, the
data are still generally analysed and presented in a twodimensional framework. Ideally, the geologist needs a
means of capturing the 3-D topography and spatial
geometry of an outcrop in a digital format that can be
viewed and analysed, like a real outcrop, in three
dimensions on a computer workstation or laptop, with
the additional benet of sophisticated data processing
and visualisation algorithms.
The Reactivation Research Group (RRG) in the
Earth Sciences Department, in collaboration with Geospatial Research Ltd (GRL) and the e-Science Research
Institute at the University of Durham are incorporating
photo-realistic virtual outcrops into their well established digital mapping methodology, GAVA (Geospatial
Acquisition, Visualisation & Analysis). Photo-realistic
14
15
Fig. 1 3-D laser scanner
(RieglLMS Z360i) with digital
camera mounted on top, in
front of the Cullercoats
outcrop, located on the coast
northeast of Newcastle, UK
RRG/GAVA:
Geospatial Research:
Halcrow:
Riegl 3D Laser Scanner:
Visualisation Toolkit (VTK):
VTK 3D data le formats:
Paraview:
Generic Mapping Tools (GMT):
Cocone/SuperCocone/TightCocone:
Polyworks:
Gawk:
Head mounted displays:
http://www.dur.ac.uk/react.res/RRG_web
http://www.dur.ac.uk/grl
http://www.halcrow.com
http://www.riegl.co.at
http://public.kitware.com/VTK
http://www.vtk.org/pdf/le-formats.pdf
http://www.paraview.org
http://gmt.soest.hawaii.edu
http://www.cse.ohio-state.edu/tamaldey/cocone.html
http://www.innovmetric.com
http://www.gnu.org/software/gawk/gawk.html
http://www.inition.co.uk/inition/products_hmds.htm
16
17
Fig. 3 Photograph of outcrop
at Cullercoats (NE England).
The Permian sandstone is
bedded and densely fractured.
This outcrop was digitised using
a 3-D laser scanner. Colour
values were recorded in
addition to the spatial position
for each scan point with a
digital camera. This
photograph was taken in the
summer, while the scanning
took place in spring, which
explains the dierences in
colour when compared with
Fig. 4
18
Fig. 6 Example of automated
tting of a tension surface
through picked points. The
spatial information contained
in the 3-D pick point locations
is used to extrapolate the
surface into space and the subsurface. A comparison of the
tension surface with a best-t
plane may provide uncertainty
estimates for the surface tting.
Note the individual scan points
close to the observer, which
merge into a continuous surface
at greater distance
or the surface expression of faults and fractures. Standard geological mapping techniques include the measurement of strike and dip of geological interfaces in the
eld. Similar measurements can be made for the virtual
outcrop model either directly or by adding in eld
measurements located in their correct geospatial position using high-precision GPS equipment. In addition,
we can use surface tting algorithms to extrapolate the
surface expression of geological interfaces, dened by a
set of picked points, into the subsurface. We make use of
the Generic Mapping Tools(GMT) (Wessel and Smith
1991) surface command, a gridding algorithm for the
generation of an adjustable tension surface of continuous curvature through a set of 3-D points (Fig. 6). The
correct extrapolation of tension surfaces that t a chosen
set of pick points is dependent on the picking error, the
distance from the outcrop surface, the continuity of the
geological surface and the mathematical function used.
A comparison with a best-t plane through the pick
points may provide a quantitative measure for the
uncertainty associated with the surface approximation.
Picking of geological surfaces such as bedding, faults
and fractures in virtual reality permits the generation of
Fig. 7 The picking and surface tting process may be repeated for
several structures to generate a complex 3-D fracture model. This
model can be used as basis for qualitative and quantitative
geological analysis of deformation processes in 3-D and 4-D, and
Outlook
Digital mapping allows the capture of a much larger
amount of information than traditional methods, and
provides geoscientists with new tools for data processing
and analysis. The creation of realistic 3-D models of
geological structures, using laser scanning, and their
spatial interpretation in computer-generated virtual
scenes, is still an emerging technology. Future develop-
19
References