ArticlePDF Available

Abstract and Figures

This paper presents a comparative study about of 3D reconstruction based on active and passive sensors, mainly LiDAR – Terrestrial Laser Scanner (TLS) and raster images (photography), respectively. An accuracy analysis was performed in regard to the positioning of outcrop point clouds obtained by both techniques. To make the comparison feasible, datasets were composed of point clouds generated from multiple images in diff erent poses using a consumer digital camera and directly by a terrestrial laser scanner. After preprocessing stages to obtain these point clouds, both were compared using positional discrepancies and standard deviation. A preliminary analysis showed that the use of digital images for 3D reconstructions is a feasible method for digital outcrop modeling, with a low cost of data acquisition and without a signifi cant loss of accuracy compared to LiDAR.
Content may be subject to copyright.
Reconstrução 3D de Modelo Digital de A oramento Baseado em Múltiplas
Imagens e Laser Scanner Terrestre
Reginaldo Macedônio da Silva1,2, Maurício Roberto Veronez1,2,
Luiz Gonzaga Júnior1,3, Francisco Manoel Wohnrath Tognoli1,2,
Marcelo Kehl de Souza1,2 & Leonardo Campos Inocencio1,2
1Universidade do Vale do Rio dos Sinos – UNISINOS
2Advanced Visualization Laboratory - VIZLab
Campus São Leopoldo: Av. Unisinos 950, Cristo Rei, São Leopoldo/RS, Brasil
{macedonios; veronez; ftognoli; lcinocencio}, {lgonzagajr; marcelo.k.souza}
2Universidade do Vale do Rio dos Sinos – UNISINOS
Graduate Program on Geology – PPGEO
Campus São Leopoldo: Av. Unisinos 950, Cristo Rei, São Leopoldo/RS, Brasil
3Universidade do Vale do Rio dos Sinos – UNISINOS
Applied Computer Science Graduate Program – PIPCA
Campus São Leopoldo: Av. Unisinos 950, Cristo Rei, São Leopoldo/RS, Brasil
Received on March 16, 2016/ Accepted on April 14, 2016
Recebido em 16 de Março, 2016/ Aceito em 14 de Abril, 2016
This paper presents a comparative study about of 3D reconstruction based on active and passive sensors, mainly LiDAR
– Terrestrial Laser Scanner (TLS) and raster images (photography), respectively. An accuracy analysis was performed
in regard to the positioning of outcrop point clouds obtained by both techniques. To make the comparison feasible,
datasets were composed of point clouds generated from multiple images in diff erent poses using a consumer digital
camera and directly by a terrestrial laser scanner. After preprocessing stages to obtain these point clouds, both were
compared using positional discrepancies and standard deviation. A preliminary analysis showed that the use of digital
images for 3D reconstructions is a feasible method for digital outcrop modeling, with a low cost of data acquisition
and without a signifi cant loss of accuracy compared to LiDAR.
Keywords: LiDAR, 3D Reconstruction, Digital Outcrop Model, Terrestrial Laser Scanner, Digital Image.
Esse artigo apresenta um estudo comparativo sobre reconstrução 3D baseado em sensores ativos e passivos, principal-
mente LiDAR (Terrestrial Laser Scanner) e imagens raster, respectivamente. Uma análise de exatidão foi realizada para
o posicionamento das nuvens de pontos para ambas as técnicas. Para tornar a comparação possível, nuvens de pontos
foram geradas a partir de várias imagens tomadas de diferentes locais utilizando câmeras digitais de alta resolução. Após
o pré-processamento para obter as nuvens de pontos, estas foram comparadas com as nuvens de pontos obtidas com o
Brazilian Journal of Cartography (2016), Nº 68/6, Special Issue GEOINFO 2015: 1203-1210
Brazilian Society of Cartography, Geodesy, Photgrammetry and Remote Sense
ISSN: 1808-0936
Silva. R. M. et al.
1204 Brazilian Journal of Cartography, Rio de Janeiro, Nº 68/6 p. 1203-1210, Jun/2016
Laser Scanner Terrestre por meio da análise de discrepâncias posicionais e desvio-padrão. Os resultados mostram que
a utilização de imagens digitais para reconstruções 3D é adequada para a modelagem digital de afl oramentos e tem
como vantagens a rapidez e o baixo custo de aquisição dos dados sem perda de exatidão signifi cante quando comparada
com os modelos digitais resultantes da técnica LiDAR.
Palavras chaves: Reconstrução 3D, Modelo Digital de Afl oramento, Laser Scanner Terrrestre, Imagem Digital.
Increasing advances in new technologies
have produced a couple of unexplored new
opportunities in the field of technologies
applied to geosciences. Thus it is important
to test and evaluate the best way to use these
technologies. Currently, in geology, we have
effi cient tools to obtain three-dimensional (3D)
data that include color and intensity, allowing
accurate measurements of layers thicknesses
for inaccessible places, for example, outcrops.
Three-dimensional digital models, especially
those obtained from a terrestrial laser scanner,
and more recently from multiples digital images
have been intensively employed.
One technique that has quickly evolved
is georeferenced geological information by the
GNSS (Global Navigation Satellite System). This
system has allowed more effi cient integration,
both in accuracy and in time gain, of the diff erent
products in a single geological reference system,
ensuring greater reliability in the processes of
generation three-dimensional geological models
(PRINGLE et al., 2004; THURMOND et al.,
2005; WHITE & JONES, 2008).
The use of digital mapping technologies
has grown in the last ten years, in particular the
use of terrestrial laser scanners and topography
equipment, integrated systems with satellite
navigation and geographic information (XU
et al., 2001; ALFARHAN et al., 2008), thus
replacing numerous photographic mosaics that
are routinely used in the interpretation of large
Terrestrial laser scanners are able
to capture a few hundreds of millions of
georeferenced points. This device, to defi ne
the three-dimensional coordinates of points on
a surface, emits laser pulses with the aid of a
scanning mirror. When a pulse hits an object,
a portion of the energy returns back to the
equipment. The distance between the sensor and
object is measured based on the time lag between
the emission and return of the pulse. Calculation
of the coordinates of each point, obtained by
the laser scanner is possible from a point with
known coordinates in the source pulse. Thus, the
study of outcrops is stimulated by the ability to
quantify the data estimated or ignored due to the
lack of access.
The use of LiDAR technology, especially
terrestrial laser scanner, in studies of outcrops
is expanding due to the ease of acquisition of
precise, fast and automated georeferenced data.
This technology has been used for this purpose
for a decade (BELLIAN et al., 2002), but only
in recent years has the number of scientific
articles increased signifi cantly. However, the
topics of interest are quite diverse, and include:
methodological approaches (BELLIAN et al.,
2005; ABELLAN et al. 2006; ENGE et al.,
2007; BUCKLEY et al., 2008; FERRARI et
al., 2012), reservoirs (PRINGLE et al., 2004;
et al., 2009; ROTEVATN et al., 2009; ENGE
al, 2009, 2010), fractured rocks (BELLIAN
et al., 2007; OLARIU et al., 2008; JONES
et al., 2009; ZAHM & HENNINGS, 2009),
erosion rates (WAWRZYNIEC et al., 2007), a
synthetic seismic model (JANSON et al., 2007),
orientation of basaltic lava fl ows (NELSON et
al., 2011) and classifi cation of spectral patterns
(INOCENCIO et al., 2014).
Photo-realistic 3D modeling is a research
topic that addresses the quick generation of three-
dimensional calibrated models using a hand-
held device (SE & JASIOBEDZKI, 2006). This
technique allows for the creation of 3D models,
both for visualization and measurements, based
on multiple images. Several studies (LEUNG,
2006; ALIAGA et al., 2006) have used this
photogrammetry technique for the reconstruction
of 3D models, and have analyzed the eff ects and
Brazilian Journal of Cartography, Rio de Janeiro, Nº 68/6 p. 1203-1210, Jun/2016
3-D Reconstruction of a Digital Outcrop Model
methods for image-based modeling from multiple
images (SZELISKI, 2010). In geology, our goal
is that this technique will be able to applied to the
analysis of outcrops in three dimensions in the
laboratory at low cost compared with LiDAR.
In addition to, it should be able to be used to
improve and facilitate virtual interpretations
(BALTSAVIAS et al, 2001; ENGE et al, 2007).
Thus the aim of this study is to quantify,
through control points, the positional error of
outcrops mapped by an image-based modeling
technique and by LiDAR, as well as to perform
a comparison of the positional errors.
Using multiples images (photographs), we
can (re)construct three-dimensional models. This
is the reverse process of obtaining photographs
from 3D scenes. When a 3D scene is projected a
2D plane depth is lost. A 3D point corresponding
to a specifi c image point is constrained to be
in the line of sight. From a single image, it is
impossible to determine a point in the line of sight
that corresponds to the image point. However,
if two images are available, then the position of
a 3D point can be found at the intersection of
the two projection rays. This process is called
triangulation. Therefore, this process requires
the multiple pass approach that begins with
the camera calibration process to relate the
measuring range of the sensor to the real world
quantity that it measures. It is necessary to fi rst
understand the mathematical model of a camera
to calibrate it. For this purpose, we have adopted
a projective camera model (pinhole camera),
which has been widely adopted as the camera
model in computer vision, since it is simple and
accurate enough for most applications.
The pinhole camera is illustrated in Figure
1 (a), while a slightly diff erent model, in which
the image plane is in front of the center of the
projection, is expressed in Fig. 1 (b).
To understand multi-view geometry, we
must first consider the relationship between
two cameras (or sequentially moving one
camera), which is actually called epipolar
geometry. Epipolar geometry is the geometry of
intersecting planes of images. Using the common
points between the images, along with the
intersection of planes, it is possible to calculate
the 3D position of objects in the scene.
Image Plane
Real Target
Pinhol e
X = ( )X, Y , Z
u = (x, y)
Image Coordinates
Camera coordinates
Fig. 1 – Pinhole camera (A), Model (B).
We have shown that there is a geometric
relationship between corresponding points in
two images of the same scene. This relationship
depends only on the intrinsic parameters of the
two cameras and their relative translation and
rotation (Figure 2).
epipo lar
plane 2
plane 1
center 2
center 1
Fig. 2 – Two cameras with epipolar constraints.
Consider a single camera viewing a
3D point w in the world passing through x1
and optical center c1. From one camera, it is
impossible to identify the point in the ray. The
projection of the ray in image plane 2 defi nes
an epipolar line. Therefore, the point in the
rst image plane (Camera 1) corresponds to
a constrained line in the second image plane
Silva. R. M. et al.
1206 Brazilian Journal of Cartography, Rio de Janeiro, Nº 68/6 p. 1203-1210, Jun/2016
(Camera 2). This relationship is called an epipolar
constraint. The constraint on corresponding
points is a function of the intrinsic and extrinsic
parameters. If intrinsic parameters are given,
then the extrinsic ones can be determined as
well as the geometric relationship between the
cameras. Another advantage is that, given the
intrinsic and extrinsic parameters of the cameras,
the corresponding point of one image can be
found easily through a 1D search along the
epipolar line in the other image.
A mathematical model can capture the
relationship between two cameras (two images)
and can provide 3D point determination. In a
general context, the mathematical constraint
between the positions of the corresponding
points x1 and x2 in two normalized cameras
can be obtained by an essential matrix (note that
either camera calibration or a diff erent matrix –
fundamental matrix is required). Details about
the essential matrix can be obtained from any
good computer vision literature (HARTTLEY
R. & ZISSERMAN, 2004). This matrix can
provide the above described parameters, mainly
the camera matrices (resectioning process) and
their parameters. Using a series of 3D-2D image
plane correspondences, it is possible to compute
the camera pose estimation, which utilizes the
camera parameters of the right camera that
minimize the residual error of the 3D-point
In another approach, three or more
cameras, instead of two can be considered.
In three views, there are six measurements,
therefore three degrees of freedom. However,
it is for lines that there is the more signifi cant
gain. In two-views, the number of measurements
equals the number of degrees of freedom of
the line in 3D-space, i.e., four. Consequently,
there is no possibility of removing the eff ects of
measurement errors. However, in three views
there are six measurements on four degrees
of freedom therefore, a scene line is over-
determined and can be estimated by a suitable
minimization over measurement errors.
For the purpose of computation, we
implemented this sequence of concepts in an
in-house computer vision library using OpenCV
(BRAHMBHATT, 2013) – for computer vision
and image processing support, Google Ceres-
Solver library1 – for modeling and solving large
complicated nonlinear least squares problems
and Eigen library2a high-level C++ library of
template headers for linear algebra, matrix and
vector operations, and numerical solvers.
The following subitems describe
the materials and and methods used in the
development of this work.
3.1 Materials
The study area is an outcrop of the Rio
Bonito Formation, Lower Permian of the Paraná
Basin, called Morro Papaléo and located at
Mariana Pimentel, Rio Grande do Sul state,
southern Brazil (Figure 3), between the geodetic
coordinate, latitudes 30°18’10”S and 30°18’40”S
and between longitudes 51°38’40”W and
51°38’30”W in the datum SIRGAS2000. The
area is an abandoned quarry that was originally
exploited for kaolin. It is a three-dimensional
outcrop with a good exposure of rocks such as
fossiliferous siltstone, carbonaceous siltstone,
pebbly mudstone and sandstone.
We implemented points that server as
a support for the georeferencing of the point
clouds obtained by LiDAR and the image-based
modeling technique. The georeferenced points
were tracked with Hyper-RTK GNSS equipment
and were supported by geodetic points layered on
top of the outcrop. These georeferenced points
(P1 and P2, Table 1) were used as a support for
measuring coordinates of points on the surface
Fig. 3 - Location map of the study area.
Brazilian Journal of Cartography, Rio de Janeiro, Nº 68/6 p. 1203-1210, Jun/2016
3-D Reconstruction of a Digital Outcrop Model
outcrop, which were subsequently used to
analyze the positional error. As a result tracking
points (P1 and P2) were obtained in the system
coordinates (Figure 3, & Table 1) UTM:
Table 1: Plane coordinates in the utm projection
of the points of support for surveying via the
central meridian at 51° sirgas2000 reference
system (geocentric reference system for the
POINTS E (m) N (m)
height – h
P1 438125,808 6646812,115 136,775
P2 438135,602 6646873,338 137,468
To obtain the coordinates on the surface
of the outcrop we used a total station (Leica
Viva TS15, Fig. 4 - Tracking of points P1 and
P2 with the use of GNSS-RTK (A). The points
of the surface outcrop were measured with Total
Station (B).
Fig. 4 - Tracking of points P1 and P2 with the
use of GNSS-RTK (A). Points on the surface
outcrop measured with Total Station (B).
This was adopted as a criterion for the
selection of local points emphasizing on the
contrast of colors and other well-defined
characteristics. This facilitated the identifi cation
of the point cloud, both in a terrestrial laser
scanner and image-based modeling. With the
total station, 21 points on the surface of the
outcrop were measured, as illustrated in Figure
4B. These coordinates were used as parameters
to determine the positional quality of the outcrop
For imaging the outcrop, we used a Leica
Scanner Station C10, with a resolution point
cloud ranging between 2mm and 4cm.
The point cloud was processed to eliminate
unnecessary information such as vegetation
and fallen rocks in front of the outcrop. In the
outcrop, sandstone predominates in Morro
Papaléo and these rocks are in the point cloud
shown in Figure 5.
F ig. 5 - Point cloud obtained with the terrestrial
laser scanner.
The same outcrop was photographed
with a Nikon D3000 digital camera at a
resolution of 7 Megapixels. The procedure
for the collection of photos in the eld was
adopted to maintain approximately the same
distance between the camera and outcrop
(Figure 6). Another procedure was adopted to
consider the top and bottom of the outcrop in
the same photo. The photos were taken from
diff erent positions to obtain approximately
60% overlap between the images.
The processing of digital photos and
reconstruction of the outcrop were an image-
based modeling technique. We reconstructed
the 3D outcrop and generated a cloud of the
points and georreference following the same
procedures used in the generation of Digital
Outcrop Model (DOM) obtained with the TLS.
Silva. R. M. et al.
1208 Brazilian Journal of Cartography, Rio de Janeiro, Nº 68/6 p. 1203-1210, Jun/2016
By comparing the results for the generation
of the DOMs based on the LiDAR technique,
and reconstruction of 3D objects from photos,
we determined that the image-based modeling
(Figure 7A) for photos allowed a visual resolution
of better quality. However, the model generated by
the terrestrial laser scanner (Figure 7B) allowed
the spacing (resolution) of the points in the point
cloud to be controlled, whereas, there is no such
control in image-based modeling from photos.
Fig. 6 – Pictures obtained with the camera (A).
Positions of the camera (B).
Fig. 7 3D Reconstruction from photos (A) and
the terrestrial laser scanner (B).
Assembling the image-based modeling
and terrestrial laser scanner point cloud, the Chi-
Square test indicated a 95% confi dence level for
georeferencing the diff erences, indicating that
there was no signifi cant diff erence between the
control data and techniques evaluated. In the
comparison of the relative error models of the
techniques used, we observed that the diff erence
became smaller than 5 cm, as shown in Table 2.
T able 2: Diff erence between linear measurements
obtained from the models generated.
Laser Scanner
Diff erence
11.6134 1.6375 -0.0241
22.3313 2.3451 -0.0138
31.8380 1.7960 0.0420
41.7010 1.6892 0.0118
52.8580 2.8669 -0.0089
The digital outcrop modeling technique
can assist in outcrop interpretation, mainly for
places that are hard to reach due to the large size
and height of an outcropping, or for security
reasons. This papers results have shown that
the image-based modeling techniques can be
feasible in this application instead of LiDAR
because the average linear error is under 40 cm.
The cost of LiDAR equipment is much higher
than that of a digital camera; hence, image-based
modeling can provide good quality results at a
lower cost as well.
The relative precision measurements
performed from the point cloud obtained from
the image-based modeling had an error below
5 cm (Table 2) than that for the point cloud
obtained from a terrestrial laser scanner, which
allows the geological features to be analyzed for
data modeling.
This study argue that image-based
modeling techniques can assist in obtaining
a point cloud in places with occlusions from
shading or obstructions around the object of the
study, which is not possible to obtain using the
LiDAR technique.
The georeferencing of the point clouds
from the image-based modeling technique
allowed overlapping of the point cloud from
Brazilian Journal of Cartography, Rio de Janeiro, Nº 68/6 p. 1203-1210, Jun/2016
3-D Reconstruction of a Digital Outcrop Model
the LiDAR technique, proving that the model
generated from photos can be associated with a
reference system. This, in turn, allows integration
of other information obtained from other data
AIKEN, C. Laser rangefinders and ArcGIS
combined with three-dimensional photorealistic
modeling for mapping outcrops in the Slick
Hills. Oklahoma. Geosphere, June 1, 2008;
4(3): 576 - 587.
Simplifying the Reconstruction of 3D Models
using Parameter Elimination. Computer Vision,
2007. ICCV 2007. IEEE 11th International
Conference on, 2007. 14-21 Oct. 2007. p.1-8.
A.; BOSH, H.; PATERAKI, M. Digital surface
modelling by airborne laser scanning and
digital photogrammetry for glacial monitoring.
Photogrammetric Record, n. 17, p. 243-273,
C. Analysis of hyperspectral and LiDAR
data: Remote optical mineralogy and fracture
identifi cation. Geosphere, December 1, 2007;
3(6): 491 - 500.
B.; LARUE, D., 2002, 3-Dimensional digital
outcrop data collection and analysus using
eye-safe laser (LiDAR) technology: American
Association of Petroleum Geologists (AAPG).
Search and Discovery Article 40056, (http://
D. C. Digital outcrop models: applications
of terrestrial scanning LiDAR technology in
stratigraphic modeling. Journal of Sedimentary
Research, n. 75, p.166–176. 2005.
BRAHMBHATT S. Practical OpenCV. Apress.
November 13, 2013
H.D; KURZ, T. H. Terrestrial Laser Scanning
in Geology: Data Acquisition Processing and
Accuracy Considerations. Journal of the
Geological Society, London; 2008, v. 165;
ISSUE: 3, p. 625-638. DOI: 10.1144/0016-
A.; HOWELL, J. A. From outcrop to reservoir
simulation model: Workfl ow and procedures.
Geosphere, December 1, 2007; 3(6): 469 - 490.
ENGE, H. D. & HOWELL, J. A. Impact of
deltaic clinothems on reservoir performance:
Dynamic studies of reservoir analogs from the
Ferron Sandstone Member and Panther Tongue,
Utah. AAPG Bulletin, February 1, 2010; 94(2):
139 - 161.
REDFERN, J. A new approach for outcrop
characterization and geostatistical analysis of
a low-sinuosity uvial-dominated succession
using digital outcrop models: Upper Triassic
Oukaimeden Sandstone Formation, central High
Atlas. Morocco. AAPG Bulletin, June 1, 2009;
93(6): 795 - 827.
REDFERN, J. Integration of digital outcrop
models (DOMs) and high resolution
sedimentology - workfl ow and implications for
geological modelling: Oukaimeden Sandstone
Formation, High Atlas (Morocco). Petroleum
Geoscience, May 1, 2010; 16(2): 133 - 154.
View Geometry in Computer Vision. Cambridge,
Cambridge University Press, 2 edition, April
2004. 670p.
FITCHEN, W. Three-dimensional geological
and synthetic seismic model of Early Permian
redeposited basinal carbonate deposits, Victorio
Canyon, west Texas. AAPG Bulletin, October
1, 2007; 91(10): 1405 - 1436.
K. J. W. Quantitative analysis and visualization
of nonplanar fault surfaces using terrestrial laser
scanning (LiDAR)--The Arkitsa fault, central
Greece, as a case study. Geosphere, December
1, 2009; 5(6): 465 - 482.
Silva. R. M. et al.
1210 Brazilian Journal of Cartography, Rio de Janeiro, Nº 68/6 p. 1203-1210, Jun/2016
Calibration and validation of reservoir models:
the importance of high resolution, quantitative
outcrop analogues. Geological Society, London,
Special Publications, January 1, 2008; 309(1):
87 - 98.
X. Improving fractured carbonate-reservoir
characterization with remote sensing of beds,
fractures, and vugs. Geosphere, April 1, 2009;
5(2): 126 - 139.
LEUNG, C. W. Y., Effi cient methods for 3d
reconstruction from multiple images 3D,
2006, 263p., Ph.D, Thesis, Scholl of Information
Technology and Electrical Engineering, Depto
of Engineering, University of Queensland,
February 2006.
Reconstructing fl ood basalt lava fl ows in three
dimensions using terrestrial laser scanning.
Geosphere, February 1, 2011; 7(1): 87 - 96.
L. V.; XU, X. Outcrop fracture characterization
using terrestrial laser scanners: Deep-water
Jackfork sandstone at Big Rock Quarry. Arkansas.
Geosphere, February 1, 2008; 4(1): 247 - 259.
PHELPS, R. M. & KERANS, C. Architectural
Characterization and Three-Dimensional
Modeling of a Carbonate Channel Levee
Complex: Permian San Andres Formation, Last
Chance Canyon, New Mexico, U.S.A. Journal
of Sedimentary Research, November 1, 2007;
77(11): 939 - 964.
3D high-resolution digital models of outcrop
analogue study sites to constrain reservoir model
uncertainty: an example from Alport Castles,
Derbyshire, UK. Petroleum Geoscience, 10,
343–352. 2004.
J. A.; FOSSEN, H. Overlapping faults and their
eff ect on fl uid fl ow in diff erent reservoir types:
A LiDAR-based outcrop modeling and ow
simulation study. AAPG Bulletin, March 1,
2009; 93(3): 407 - 427.
SE, S. & JASIOBEDZKI P. Photo-realistic 3D
Model Reconstruction. IEEE International
Conference on Robotics and Automation,
Orlando, Florida, USA., May 1, 2006; 3076 -
SZELISKI R. Computer Vision: Algorithms
and Applications (Texts in Computer Science).
London, Springer, 2011 edition, October 2010.
X. Building simple multiscale visualizations
of outcrop geology using virtual reality
modeling language (VRML). Computers and
Geosciences, 31, 913–919. 2005.
Chronotopographic analysis directly from
point-cloud data: A method for detecting
small, seasonal hillslope change, Black Mesa
Escarpment. NE Arizona. Geosphere, December
1, 2007; 3(6): 550 - 567.
WHITE, P. D. & JONES, R. R. A cost-effi cient
solution to true color terrestrial laser scanning.
Geosphere, June 1, 2008; 4(3): 564 - 575.
ZAHM, C. K. & HENNINGS, P. H. Complex
fracture development related to stratigraphic
architecture: Challenges for structural
deformation prediction, Tensleep Sandstone at
the Alcova anticline, Wyoming. AAPG Bulletin,
November 1, 2009; 93(11): 142
ResearchGate has not been able to resolve any citations for this publication.
Full-text available
Fracture prediction in subsurface reservoirs is critical for exploration through exploitation of hydrocarbons. Methods of predicting fractures commonly neglect to include the stratigraphic architecture as part of the prediction or characterization process. This omission is a critical mistake. We have documented a complex heterogeneous fracture development within the eolian Tensleep Sandstone in Wyoming, which arguably is one of the least complex reservoir facies. Fractures develop at four scales of observation: lamina-bound, facies-bound sequence-bound, and throughgoing fractures that span the formation. We documented a detailed facies and fracture-intensity model using LIDAR-scanned outcrops located at the Alcova anticline in central Wyoming. Through this characterization, we reveal the existence of a striking variability in fracture intensity caused by original depositional architecture, overall structural deformation, and diagenetic alteration of the host rock.
Full-text available
Many fault surfaces are noticeably nonplanar, often containing irregular asperities and more regular corrugations and open warping. Terrestrial laser scanning (light detection and ranging, LIDAR) is a powerful and versatile tool that is highly suitable for acquisition of very detailed, precise measurements of slip-surface geometry from well-exposed faults. Quantitative analysis of the LIDAR data, combined with three-dimensional visualization software, allows the spatial variation in various geometrical properties across the fault surface to be clearly shown. Plotting the variation in distance of points from the mean fault plane is an effective way to identify culminations and depressions on the fault surface. Plots showing the spatial variation of surface orientation are useful in highlighting corrugations and warps of different wavelengths, as well as cross faults and fault bifurcations. Analysis of different curvature properties, including normal and Gaussian curvature, provides the best plots for quantitative measurements of the geometry of corrugations and folds. However, curvature analysis is highly scale dependent, so requires careful filtering and smoothing of the data to be able to analyze structures at a given wavelength. Three well-exposed fault panels from the Arkitsa fault zone in central Greece were scanned and analyzed in detail. Each panel is markedly nonplanar, and shows significant variation in surface orientation, with spreads of ~±20°-25° in strike and ±10° in dip. Most of the variation in orientation reflects decimeter-and meter-scale corrugations and longer wavelength warps of the fault panels. Average wavelengths of corrugations measured on two of the panels are 4.04 m and 4.43 m. Whereas the orientations of the three fault surfaces show significant variations, the orientations of fault striae are very similar between the panels, and are tightly clustered within each panel. Fault-slip analysis from each panel shows that the local stress field is consistent with the regional velocity field derived from global positioning system data. The oblique slip lineations observed on the fault panels represent a combination of two contemporaneous strain components: north-northeast-south-southwest extension across the Gulf of Evia and sinistral strike slip along the west-northwest-east-southeast Arkitsa fault zone.
Full-text available
The advent of high-resolution, precise, back-pack portable terrestrial lidar scanners (TLS) provides a revolutionary new tool for obtaining quantitative, high-resolution (2-mm to 30-mm point spacing) measurements of landscape surface features. Moreover, data collected using these instruments allow observation of geomorphic processes in systems that can experience change on a daily basis. We have introduced TLS techniques in ongoing investigations of semiarid landscapes associated with weakly cemented sandstones along part of the Black Mesa escarpment of NE Arizona. Clay-cemented, Jurassic sandstones exposed along this escarpment are sensitive to moisture, and thus climate, via hydration-expansion weathering of interstitial clay. Sediment shed from weathered slopes has caused locally rapid valley fl oor aggradation and upper basin slope vertical denudation rates of 2-3 mm/yr over 10- to 100-yr timescales, as indicated by dendrochronology coupled with soil geomorphic analysis. These rates suggest rapid hillslope denudation rates. Employing the University of New Mexico Lidar Laboratory Optech Ilris 3D TLS, we are constructing a high-resolution model of two major basins along the escarpment. Focusing on a single, small (30 × 60 m) area of a mostly non-vegetated, steep slope (>35°), we demonstrate in this paper a method of comparative analysis of pointcloud data sets that can detect subcentimeter change resulting from a single season of monsoon precipitation along the escarpment. Using repeat scans can provide an empirical evaluation of single season erosion rates in the study site, and because our observations are geospatial in nature, we can also document the parts of the slopes that make the greatest contribution to local valley fl oor aggradation. In demonstrating the utility of this method, we expect that continued investigation of this site will provide insight to the key processes associated with soil-mantled versus bedrockdominated slopes during modern escarpment retreat and hillslope modifi cation, which, in turn, may further elucidate the impacts of Holocene climate change on this rapidly evolving landscape.
Full-text available
Complete understanding of outcrop-scale stratal architectures requires extrapolation of two-dimensional depositional cross sections into three dimensions. This study integrates a high-resolution digital outcrop model with outcrop observations to create a three-dimensional geologic model of distal outer-ramp carbonate stratigraphy. Data were taken from sinuous canyon-wall exposures of the Permian San Andres Formation of Last Chance Canyon, New Mexico. A series of laterally extensive carbonate benches previously interpreted as constructional sponge mud mounds are here modeled as a channel-levee complex that is characterized by an alternating history of aggradation punctuated by erosional sediment bypass. Levee mudstones have an erosional base, contain dolomitized peloidal thin beds, and have limited faunal constituents. Primary levees were initiated as a carbonate apron and transformed into a low-relief channel-levee system as a result of deposition by dilute turbidity currents. Subsequent sediment bypass incised through the pre-established channels and into the underlying facies. The fill consists of skeletal packstone lags and fine peloidal packstones, which are correlated to coral- and sponge-bearing secondary levee deposits that drape the primary levees. Debrites, also found within some channelized areas, are thought to represent channel-margin oversteepening and slumping. Channel widths range from approximately 300 to 800 in, and channel-to-levee relief reaches up to 40 in. Three-dimensional surface models representing architectural elements of the channel-levee complex confirm outcrop observations. Surfaces that were modeled as a best fit to the three-dimensional outcrop tracings reveal development of low-sinuosity channels in previously established bathymetric lows created by earlier phases of sediment bypass and erosion. Sequence stratigraphic interpretations suggest that the channel-levee complex developed during the highstand part of a longterm transgression. The overall transgressive setting led to generation of substantial quantities of carbonate mud in more proximal areas of the ramp. Accumulated mud in the source region was transported down depositional dip during highstand parts of high-frequency cycles and resulted in growth of the levees. Carbonate channel-levee complexes are not likely to form during highstand sequences in which the source sediment is grain-dominated. Leveed channels are a viable component of deepwater carbonate settings and may be more common than has been previously recognized. Complex stratal architectures found in this channel-levee complex could be analogous to apparently gullied slopes of muddy carbonate ramps, especially those developed in transgressive sequences.
Full-text available
The Lower Permian outcrops of Victorio Canyon, in the Sierra Diablo Mountains in west Texas, show undisturbed stratigraphy of carbonate toe-of-slope and basinal deposits. These rocks consist of a vertical stack of carbonate debris-flow deposits and hyperconcentrated density-flow deposits thick skeletal-ooid grainstone de, posited as basin-floor fans, various reworked muddy carbonate deposits, and fine siliciclastic siltstones. This succession was deposited within five sequences that include the upper Hueco Formation (late Wolfcampian) through the Bone Spring Formation (middle Leonardian). Using a ground-based light detecting and ranging-generated high-resolution digital outcrop model (DOM) aged as a template, we mapped and digitized the stratigraphy of the toe-of-slope and basinal deposits on the 5-cm (2-in.) precision DOM. On the basis of the digitized stratigraphic contacts, several regional surfaces were constructed, and a 3-D geocellular model was built. Facies information within this model is extrapolated from measured section data using both a kriging algorithm and stochastic simulations. Using impedance values extracted from a subsurface analog, a 3-D impedance model was created for both the kriged and the stochastic models. Both models incorporate fine-scale stratigraphic architecture. In addition, the stochastic impedance model incorporates spatially correlated noise, resulting in more realistic synthetic seismograms. Three-dimensional synthetic seismograms were calculated at 20, 40, and 80 Hz. The reservoir-prone facies is skeletal-ooid grainstone deposited as a 1.5-km x 750-m (0.93-mi x 2460-ft) basin-floor fan up to 15 m (49 ft) thick. This basin-floor fan is subtly imaged in vertical seismic section at a frequency below 80 Hz. It is, however, better recognized on time slices with peak frequency as low as 20 Hz and even better delineated on horizon slices that parallel the stratigraphy.
We present a new method for reconstructing flood basalt lava flows from outcrop data, using terrestrial laser scanning (TLS) to generate three-dimensional (3D) models. Case studies are presented from the Faroe Islands and the Isle of Skye (UK), both part of the North Atlantic Igneous Province (NAIP). These were analyzed to pick out lava flow tops and bases, as well as dykes, lava tubes, and sedimentary layers. Three-dimensional surfaces were then generated using modeling software, and 3D geological models constructed. Finally, the models were interrogated to give data on flow thickness and crust-to-core ratio. The aim of this research is to obtain quantitative data on the internal heterogeneity of a sequence of flood basalt lava flows, and to provide high-resolution information about flow geometries and volcanic facies variations in 3D. Lava flow sequences display complex stacking patterns, and these are difficult to understand from photos or outcrop observations. Laser scanning allows us to study inaccessible outcrops, while avoiding the perspective distortion in conventional photography. The data from this study will form parts of larger models of flood basalt provinces, which will be used to improve seismic imaging in areas of basalt cover, and aid our understanding of facies architecture in flood basalts.
Subsurface reservoir models are typically limited by a lack of spatially accurate geometric data on bedform architecture and geometry. These factors are key controls on fluid flow. Outcrop analogs have long been used as a source of such data, but the capture of sufficiently precise outcrop data is a challenge. The study presented in this article used highly accurate geometrical digital geological outcrop data collected using ground-based laser scanning (light detection and ranging [LIDAR]) to build and test three-dimensional geocellular models of deltaic reservoir analogs. Two well-exposed ancient river-dominated delta systems, the Panther Tongue and the Ferron Sandstone Member, which both crop out in central Utah, were digitally mapped to precisely recreate their clinothem and clinoform geometries in geocellular reservoir modeling software. Such clinoforms are commonly draped with low-permeability mudstones that produce reservoir heterogeneity by subdividing the deltaic sand body into a series of dipping sandstone beds (clinothems). A key aspect of the modeling was to accurately capture these geometries and their effect on simulated fluid flow. Portions of the two deltaic systems were dynamically analyzed in a reservoir modeling software by simulating production in 41 models. These models tested a range of mudstone barrier continuities and permeabilities. Results quantify how the continuation of the heterogeneities governed the production rate and recovery factor in the Panther Tongue models. Mudstone permeability values were more important in the Ferron Sandstone models with steeper dipping and closer spaced clinothems, although production was still influenced by the continuation of the heterogeneities.
The use of digital outcrop models in combination with traditional sedimentological field data improves the accuracy and efficiency of qualitative and quantitative characterization of outcrop analogs for reservoir modeling purposes. In this article, we apply an innovative methodology of outcrop characterization to an Upper Triassic fluvial-dominated system, exposed in extensive outcrops with limited three-dimensional (3-D) exposure. Qualitative analysis of the study outcrop allows the subdivision of the formation into three architectural intervals. Each interval can be further subdivided into subintervals on the basis of architectural style. This subdivision provides information on reservoir compartmentalization, which is used for zonation of the geocellular model. Qualitative analysis also provides valuable information on reservoir facies distribution. A new technique termed "perpendicular projection plane" is presented as a tool for quantitative analysis of outcrops with reduced 3-D exposure. This technique improves the accuracy of apparent width measurements of geobodies exposed in outcrops, which are subparallel to paleoflow. The quantitative analysis provides a detailed data set of geobody dimensions to use as conditioning data for analog reservoir models. Statistical analysis of the dimensions provides empirical relationships to apply in subsurface analog systems to reduce uncertainty related to stochastic modeling approaches.
Advances in data capture and computer technology have made possible the collection of 3D high-resolution surface and subsurface digital geological data from outcrop analogues. This paper describes research to obtain the 3D distribution and internal sedimentary architecture of turbidite channels and associated sediments at a study site in the Peak District National Park, Derbyshire, UK. The 1D, 2D and 3D digital datasets included Total Station survey, terrestrial photogrammetry and remote sensing, sedimentary logs and a Ground Penetrating Radar (GPR) dataset. A grid of 2D GPR profiles was acquired behind a cliff outcrop; electromagnetic reflection events correlated with cliff face sedimentary horizons logged by Vertical Radar Profiling. All data were combined into a Digital Solid Model (DSM) dataset of the site within reservoir modelling software. The DSM was analysed to extract 3D architectural geometries for petroleum reservoir models. A deterministic base model was created using all information, along with a suite of heterogeneous turbidite reservoir models, using 1D, 2D or 3D information. The model suite shows significant variation from the deterministic model. Models built from 2D information underestimated connectivity and the continuity of geobodies, but overestimated channel sinuosity. Advantages of using 3D digital outcrop analogue data for 3D reservoir models are detailed.