Remote sensing techniques for onshore oil and gas exploration
Ben Lord1
Abstract
Remote-sensing data are highly useful for the oil and gas
industry. Remote sensing has proven to be an integral tool for
Downloaded 12/27/16 to 132.239.1.231. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
downstream and upstream oil and gas operations through evaluation
of infrastructure for well-site planning and for exploration through
large-scale regional reconnaissance. Spectral analysis is a key form
of processing to evaluate for surface outcrops and surface hydro-
carbon seepage. We will show methodologies using remote-sensing
data and how they assist in the overall understanding of the hy-
drocarbon system. A number of studies are shown in which hy-
perspectral and multispectral data sets have proven useful in un-
derstanding the geology and inferring information about the
subsurface. Case studies are presented from Pennsylvania, USA;
Neuquén province, Argentina; and Lebanon. Advantages and
disadvantages of the use of these methodologies and the future of
remote sensing in the oil and gas industry are discussed.
Introduction and methods
Remote sensing has proven to be an effective tool for monitor-
ing a plethora of different earth phenomena. Whether it be
monitoring of environmental qualities in various ecosystems or
urban planning, remote sensing has found its place as a key map-
ping tool in a variety of disciplines. One of the main applications
of remote sensing is that of oil and gas resource mapping. Remote
sensing has been used for both downstream and upstream applica-
tions, such as pipeline mapping, wellsite planning, and environ-
mental change detection. Remote sensing also has been used
extensively in oil and gas exploration, utilizing everything from
high-resolution visual data sets such as WorldView or RapidEye
to radar data and multispectral/hyperspectral data sets. Through
spectral analysis, it is possible to see and map potential geologic
structures and gain a better understanding of the hydrocarbon
system beneath the earth’s surface. Several studies have indicated
that hyperspectral images can be used to identify onshore and
offshore hydrocarbon seepage (Ellis et al., 2001; van der Meer et
al., 2002; Prelat et. al, 2013). This has proven useful for rapid
mapping in remote areas (Ducart et al., 2016) and to be an ex-
tremely cost-effective approach (Lisica, 2015) for high-grading
study areas. With seismic data and field crews yielding financial
burdens and health and safety risks, remote sensing is becoming
a go-to tool for exploration. Some remote sensing practices will
be highlighted here, together with case studies of previous work
done by NEOS GeoSolutions to demonstrate how remote sensing
data assisted the overall understanding of the study areas.
Hyperspectral wavelengths
Remote sensing encompasses and utilizes information derived
from surface reflectance in the electromagnetic spectrum. Spectral
analysis involves the evaluation of energy absorbed and reflected in
various wavelengths. In effect, everything on the earth’s surface
reflects and absorbs various wavelengths differently. The ability to
1
NEOS GeoSolutions.
24 THE LEADING EDGE January 2017
Figure 1. Spectral profiles of Canadian bitumen sand samples of different
percentages in the visible near infrared (VNIR) and shortwave infrared (SWIR)
bands. Notice key absorption at 2310 nm (figure taken from Curtis, 2013).
analyze and characterize these reflectance spectra provides informa-
tion on land-cover types and, in this case, clues about the overall
hydrocarbon system at the surface. Spectral classifications applied
here were trained using reference endmembers generated from field
operations collected by various ground crews operating field spec-
troscopy devices, as well as some key published reference spectra.
Figure 1 shows one such published spectra example, that of bitumen
samples (Curtis, 2013) taken from Canadian oil sands. The bitumen
samples have spectral absorption located at 2310 nm. Analysis of
these specific wavelengths helps to generate final classifications.
Hyperspectral sensors
Sensors used for the following case studies ranged from mul-
tispectral (1–10 bands) to hyperspectral (hundreds of bands), with
most of the sensors focused predominantly on the visible near in-
frared (VNIR) and shortwave infrared (SWIR) parts of the elec-
tromagnetic spectrum. Data were either attained from free resources
such as the United States Geological Survey (USGS) (Landsat8
and ASTER) or collected from airborne platforms by third-party
vendors (SPECTIR and ITRES). The airborne platforms were
flown at optimum collection times during the year and during
optimum times of day, taking into account sun inclination and sun
angle durations. The systems ranged in spatial resolutions from 3
to 30 m depending on sensor type in use and the study’s objectives,
with the airborne hyperspectral sensors having 3 m resolution and
the majority of satellite sensors in the range of 30 m resolution.
Methodologies used in case studies
After radiometric and geometric correction, all data sets were
processed through Harris Geospatial’s ENVI program where first
a number of atmospheric correction algorithms were applied to
the data. These corrections ranged from QUAC (quick atmospheric
correction), used primarily for VNIR and SWIR data sets based
http://dx.doi.org/10.1190/tle36010024.1.
Special Section: Remote sensing
 on the proven basis that the average reflectance of various materials
within a scene will be scene independent (Agrawal and Sarup,
2011), to ENVI’s In-Scene Atmospheric Correction (ISAC) for
thermal data sets to calculate emissivity based off of work by
Young et al. (2002). ENVI-based classifications were also used.
These algorithms were used to smooth and extrapolate data before
comparing them to known spectral samples and a priori knowledge
for the final classifications. The classifications used incorporated
Downloaded 12/27/16 to 132.239.1.231. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
whole-pixel and subpixel analysis, analyzing angular differences
in spectra and overall wavelength land-cover contributions to
in-scene pixels. These classification algorithms included spectral
angle mapper (SAM), linear spectral unmixing (LSU), continuum
removal, and band ratios. These were all combined into a decision
tree approach to create final classifications.
Spectral angle mapper (SAM). Spectral signature matching
is very important in the identification of earth surface objects.
Endmember spectra collected in the field (NEOS spectral library)
as well as spectra from the USGS spectral library were used to
identify oil seep and other target pixels such as rock types and
mineral alterations in the images. A SAM algorithm, which is a
spectral-based classifier, was used to compare pixel values to
user-derived endmembers or signatures using the various spectral
libraries, both USGS and proprietary (Kruse et al., 1993). SAM
compares the angle between reference spectra vectors and spectra
from each pixel in an image. It compares each pixel in the image
with every endmember for each class in a spectral library and
assigns a weighting value between 0 and 1. Endmembers can be
taken directly from the image or from signatures measured directly
in the field or laboratory. The smaller the angle, the closer the
relationship between a given reference spectrum and a pixel. The
main disadvantage of the SAM algorithm is that it does not
consider the subpixel attributes. The two main outputs of SAM
analysis are a SAM classification (which is generated based on
one particular threshold value of angle calculation) and a gray-scale
rule image where the lower values signify angles closest to the
input spectra.
Linear spectral unmixing (LSU). LSU outputs the same prod-
ucts as the SAM classification (a classified image and a gray-scale
rule image) but specifically takes into account the spectral mixing
that can take place within one pixel since inevitably, despite the
highest resolutions capable by sensors, there will be spectral vari-
ability within the pixel itself (Plaza and Plaza, 2011). The method
assumes that within one scene pixel there are many different factors
that can affect the average endmember spectra that come out of
that particular pixel. That is to say that in one pixel there will be
a series of different landforms that can affect the overall spectra
of that location, including vegetation, different mineral alterations,
and different soil and ground properties. LSU investigates these
aspects and analyzes the images incorporating these factors based
on the endmember spectral library used in the analysis.
Continuum removal. Throughout the images, background
spectra from vegetation and general quaternary cover can mix
with some of the more pure spectra expected to be classified. One
approach to minimize this effect is via linear spectral unmixing
suggested previously; another method to highlight some of the
more subtle variability in in-scene spectra is via continuum removal
Special Section: Remote sensing
(Clark and Roush, 1984). The continuum-removal process helps
to differentiate target spectra from background spectra. The
continuum consists of background absorption, which is necessary
to extrapolate the baseline of the target’s general spectra. These
continuum-removed images were later used for highlighting
absorption features through band ratios and indices.
Band ratios. Minerals and land-cover types of interest have
specific absorptions that need to be highlighted. Using classifiers
such as SAM and LSU, the entire spectrum is analyzed and may
not capture some of the key wavelengths that need to be considered.
Band ratios allow for this specific wavelength analysis. For this
approach, the band-math preset function in ENVI was used, and
mathematical indices were created and applied in order to best
highlight these specific wavelengths in the spectrum. The math-
ematical equations created for each target endmember calculates
the depth of the specific absorption as well as gradient changes
of the spectra. The output of these ratios are single-band images
in gray scale (values between 0 and 1), also known as a rule image,
in which the brightest sections (values closer to 1) signify the
deepest section of endmember absorption dip.
Spectral endmembers used as a basis for creating these indices
and ratios were taken from known spectra from field operations
or spectral libraries, similar to the reference endmembers in the
previous classification algorithms. These endmembers were used
as references and are compared to spectral profiles from the study
area for generating the surface hydrocarbon classifications.
Combined decision tree approach. All gray-scale rule images
produced by the SAM, LSU, and band ratio images put into a
decision tree matrix where thresholds were applied. These thresh-
olds are derived from and compared with SAM, LSU, and ratio
values from known targets from a priori knowledge (ground truth
operations, expert geologic knowledge, known geologic map, etc.)
and applied in the decision tree. This decision tree process produces
final hydrocarbon seepage classifications, mineral alteration clas-
sifications, and lithology classifications. Figure 2 shows an example
decision tree path classifying for alteration minerals.
Indirect mineralogy (satellite/airborne)
In addition to the direct detection of hydrocarbon spectra, both
mineral analysis and surface geochemistry play a large role in the
overall understanding of the hydrocarbon system. With respect to
indirect mineralogy, a number of relationships can be established
between surface hydrocarbon seepage and mineral occurrence. The
classification of minerals in the various study areas becomes part
of the indirect hydrocarbon detection methodology. The indirect
detection method for remote sensing data relies on the detection
of sediments and rocks altered at the surface by the microseepage
of hydrocarbons. The hydrocarbons that escape from the under-
ground reservoir cause an oxidation-reduction reaction either in
situ or along vertical migration pathways (Saunders et al., 1999).
The resulting mineral alterations can be detected at the surface in
sediments and soils. The mineralogical alteration induced by hy-
drocarbon seepages include bleaching of red beds, enrichment of
ferrous iron, and the alteration of clay minerals and carbonates
leading to mappable alteration zones above and near subsurface
accumulations. Figure 3 shows an example diagram of key surface
observations for this microseepage model. It is through these mineral
January 2017 THE LEADING EDGE 25
 classifications that an indication of potential surface hydrocarbon
seepage points can be observed. In addition, tonal anomalies and
subtle changes in the spectral content are used as important indica-
tors of seepage (Almeida-Filho et al., 1999).
Case study: Pennsylvania, USA
Downloaded 12/27/16 to 132.239.1.231. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
transform infrared radiation (FTIR) system fitted onto a helicopter
platform. The FTIR sensor collected data to test for potential
methane plumes and was flown at lower elevations to create the
smallest and most precise pixels for direct detection.
A calibration and validation operation was performed in which
two separate field visits were conducted during acquisition and

Although many areas already have been well researched and postprocessing/classification to validate the results. Classifications
highly explored for oil and gas, remote sensing can still play a were done for hydrocarbon seepage as well as methane gas seepage
helpful role in these areas. The study area of Tioga County in from the combined SWIR spectra from the airborne platform
Pennsylvania has produced hydrocarbons for more than 100 years, and the thermal data collected from the helicopter-based FTIR
and many of the well sites are in full operation. However, a number platform. Field crews were also deployed to take analytical spectral
of wells and pipelines have been decommissioned, plugged, or device (ASD) readings of oil seepage from documented abandoned
abandoned. Some of these sites have since started leaking, adding wells. Figure 4 shows ASD readings taken from land-cover types
to the natural seepage already occurring in the area. Companies surrounding seepage areas. These readings were later used as
wishing to sustain ongoing production require the ability to drill reference spectra input into ENVI classification algorithms such
wells while taking into account the environment and overall safety as SAM and LSU.
of the local population. In Pennsylvania,
this, in effect, means avoiding pipelines
through all potential pathways of meth-
ane, whether they be man-made or
natural (McKee and Beasley, 2013), in
order to protect fresh ground-water
aquifers. The ability to map these hy-
drocarbon occurrences using remote
sensing technology has proven to be
highly useful for environmental analysis
and for future infrastructure planning
as described by Smejkalova and Bujok
(2012) and Schneising et al. (2014).
For this study, a SPECTIR airborne
hyperspectral survey was conducted in
Figure 2. Example decision tree constructed for alteration mineral classifications associated with hydrocarbon
late summer 2011 over the entire county. microseepage. As annotated above, each decision node has thresholds within it to characterize the pixel. The output
During the same period, a smaller test of this decision tree node can lead into another node (leaf node) or an edge, which can be a classified pixel or,
area was covered using a Fourier eventually as you go down the nodes, an unclassified pixel.

Figure 4. ASD sample readings taken from an abandoned oil well location
Figure 3. Diagram illustrating microseepage in oxidizing zones and accompanied with surrounding land covers, including soil and vegetation altered by oil
minerology alterations (modified from Schumacher, 1996). contamination.

26 THE LEADING EDGE January 2017 Special Section: Remote sensing

 Thermal remote sensing sensors like the FTIR measure the
energy emitted from the earth’s surface in the thermal infrared
region of the electromagnetic spectrum (3000–15000 nm). The
radiant temperature of a given object depends on many factors,
such as emissivity, conductivity, capacity, diffusivity, and thermal
inertia. Representative emittance and emissivity spectrums were
Downloaded 12/27/16 to 132.239.1.231. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
generated at each X and Y location of the FTIR data for further
analysis and mapping purposes. The thermal emissivity data were
gridded and stacked using proprietary NEOS algorithms, creating
thermal image files for each flight.
Methane gas shows characteristic absorption at the 7800 nm
region of the emissivity spectrum (Figure 5). Thermal data wave-
length ranging from 7500 to 12,000 nm were used to detect
methane plumes. A band ratio was calculated at 7800 nm from
the thermal image files. Band threshold parameters as well as
absorption characteristics of CH4 spectra and surrounding areas
(vegetation, soil, etc.) were used in the binary decision tree algo-
rithm described in Figure 2 to yield possible CH4 plume polygons
for the study area.
Ground-truthing verification work was performed to validate
oil classification points. Field validation sites were selected based
on a subset of classified locations with regard to accessibility. Eleven
classified locations were targeted to validate the hyperspectral
classification. Field visits were conducted with a ground crew using
Figure 5. Example methane spectra with key target absorption at 7800 nm.
Figure 6. Seep location field photo taken over brown algal soil surrounding an
abandoned well.
Special Section: Remote sensing
a methane gas detector (model Sensit HXG-3) and spectral field
device to compare classified spectra with known spectra.
Results from the field validation campaign showed numerous
classified areas to have visible oil sheen on standing water, a notice-
able brown algae mixture within the standing water (Figure 6),
stained hydrocarbon-bearing soil, or extremely high methane
readings (up to 19,660 ppm). Other areas where origins were
determined to be natural had oil sheens in stagnant water sur-
rounded by vegetation and usually were located on the banks of
wetlands or springs.
Results from spectral mapping coupled with field observations
discovered seepage occurring from broken and abandoned well
sites as well as natural seepage indicators in numerous positions
that at the time of the survey had not been documented. Figure 7
shows some of the key classified discoveries in the survey area
differentiated between liquid hydrocarbon and gas hydrocarbon
seep endmembers.
In summary, the project allowed for better understanding of
where surface seepage was occurring. It also provided real-time
analysis of potential leakage in already developed areas, making
it an ideal tool for future pipeline monitoring because of its rapid
and noninvasive nature (Schneising et al., 2014). The area was
also an ideal location for sampling surface leakage and measuring
spectral characteristics of the leakage.
Case study: Argentina
Remote sensing greatly assists in the overall understanding
of frontier areas from a regional setting (Prost, 2002) and can be
used prior to leasing or staking claims to establish which areas
contain favorable qualities worth pursuing. A study located in the
northern regions of Neuquén province in Argentina acquired
multiple airborne hyperspectral data sets incorporating the VNIR-
SWIR portions of the electromagnetic spectrum as well as thermal
data using ITRES’ CASI, SASI, and TASI systems. These data
were collected in order to better understand and characterize the
Figure 7. Tioga County study area showing classified liquid hydrocarbon seepage
points in green and gas hydrocarbon seepage points in purple. Ground survey
crews were deployed in a number of these positions to verify classification targets.
Surveys showed an 89% success rate with the classification between airborne and
ground-based methods.
January 2017 THE LEADING EDGE 27
 geology, petroleum systems, and hydrocarbon prospectivity of The spectral information divergence classification method
the fold and thrust belt located in the basin. The study’s overall (Du et al., 2004) was used in which divergence of in situ spectra
aim was to better characterize the petroleum potential of the area is measured in comparison to an input reference spectrum. Pixels
of interest. The high spatial and spectral resolutions of the data
sets allowed for very good characterization of varying lithology
as well as surface hydrocarbon seepage.
Downloaded 12/27/16 to 132.239.1.231. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/

The Neuquén Basin is known for its unconventional reservoirs,

which are dominated by the Vaca Muerta Formation. Understand-
ing the area’s geology and rock properties is vital to successful
exploration. The study area is located in an arid environment fairly
devoid of vegetation with a highly complex and mappable geology
outcropping at the surface. Geology in the region consists of a
series of anticlinal and synclinal structures, which allowed for
variation in lithology along vertical dipping beds to be captured
at outcrop locations via remote sensing imagery. This allowed for
detailed mapping of lithologic variation, specifically the identifica-
tion of sandstone, clays, and carbonates. Figure 8 shows an example
of some of the key clay and carbonate alterations occurring on
the edges of the main anticlinal formation in the block identified
through decision tree classification as described above.
A series of field studies was conducted in the area taking Figure 8. Neuquén Basin study area with hyperspectral as base gray-scale
samples of spectra using an ASD Fieldspec Pro system. Samples imagery. Classifications of minerology are overlaid. Notice how mineral
were taken of different formations such as the Vaca Muerta and alterations occur on the flanks of the anticline structure. This was confirmed with
the surrounding Tordillo as well as at documented historical hydrocarbon seepage locations in the field.
surface hydrocarbon seepage locations. (The locations of the latter
were obtained from published articles as well as colloquial knowl-
edge from local sources.) Areas of known seepage were found to
have visible iron staining associated with the seepage, confirming
the correlation in the field with the known microseepage mineral
alteration model. The existence in the classification of illite, ka-
olinite, and irons identified areas of potential microseepage. The
USGS mineral spectra for each of the target minerals are displayed
in Figure 9.
The interpreted result was that seepage is occurring along
geologic contact boundaries and fault zones (associated with the
known migration pathways) and causing multiple mineral altera-
tions identifiable by the hyperspectral analysis. This is illustrated
in the schematic cross section in Figure 10. The study gave insight
into the variability within lithologic sections and their contribu-
tions to source rocks and seepage associated with the source rocks. Figure 9. USGS spectra taken for key minerals associated with hydrocarbon
Classification of shales from the Vaca Muerta was also per- microseepage in Neuquén Basin.
formed using total organic carbon
(TOC) measurements. TOC in a major-
ity of cases can be linked to source rocks
(Naeth et al., 2005), is useful in assess-
ing overall quality of these rocks, and
has been used in the evaluation of shales
in some unconventional reservoirs mak-
ing it extremely useful for the Neuquén
Basin study area and Vaca Muerta
analysis. Here, field samples were col-
lected from various rock formations and
taken for lab geochemical work to
measure TOC. These samples also were
Figure 10. Diagram showing a schematic cross section illustrating the relationship between hydrocarbon migration
taken for spectral lab work where spec- and hydrocarbon indicators. Hydrocarbon indicators (H) occurring along contact boundaries (especially the Vaca
tral measurements were taken from Muerta) and along faults with accompanied mineral alterations of three or more minerals (3) occurring in similar
samples with varying degrees of TOC. areas along contacts and seeping directly through the sedimentary fill.

28 THE LEADING EDGE January 2017 Special Section: Remote sensing

 and spectra with smaller divergence are said to be most similar
to the reference spectra. In this case, a threshold is applied, and
if divergence is higher than this specified threshold, the pixel is
not classified. In this instance, the reference spectra were based
on in-scene regions of interest (ROI) derived from field sample
locations. The reference spectra are the average spectra generated
Downloaded 12/27/16 to 132.239.1.231. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
was used, consisting of two sample spectra from locations in North
America as well as known bitumen absorptions from Curtis (2013)
(Figure 1) (to analyze surface hydrocarbon seepage). These spectra
were considered representative of the target seepage to be classified
and were used in the absence of any sample spectra available from
Middle Eastern regions. The hyperspectral data were run through

from all pixels in the ROI. Sampling locations were divided into
various land-cover types based on known geology and TOC
sample values. TOC samples were grouped into values of high,
medium, and low TOC based on TOC lab readings. These were
grouped as follows:
a decision tree analysis to produce highest-scoring rule images
and generate a final classification of only pixels closely matching
the endmember input spectra.

• high: TOC values higher than 0.8 wt. (%);

• medium: TOC values between 0.4 and 0.8 wt. (%); and
• low: TOC values less than 0.4 wt. (%).

Figure 11 shows example endmembers obtained in surface

TOC measurements from various formations, the key one being
the Vaca Muerta.
Once classification outputs were generated for these designated
target zones, remaining TOC samples not used in the training
were incorporated to validate results, making sure TOC values
fit correctly with their appropriate category (high, medium, low).
Figure 12 shows an example output from the classification. Figure 11. ASD spectral samples of shale formations with low and high TOC.
Some pitfalls do exist with regard to this TOC analysis, Notice the difference between the high TOC Vaca Muerta and Agrio samples (red
given the overall complexity of the geology in the area. It is and yellow) versus the low TOC samples of the same formations (blue and teal).
first assumed that shales exist throughout the study area, and
therefore only certain subsets of the study area were picked for
this analysis. These selected areas consist of known high TOC
shale outcrops and areas where known ground samples were
collected that had high TOC values (higher than 0.8 wt [%]).
There is also the complication of inferring subsurface TOC
from these surface TOC readings due to the location’s geologic
complexity. Making subsurface assumptions from this approach
could be a risky and complicated process. However, based on
work from Veiga et al. (2005), there is a very high correlation
between the hydrocarbon index and TOC levels, making the
areas of high TOC classified to be high-priority areas for
further study.

Case study: Lebanon

Remote sensing is especially applicable in areas where little
ground data are available (Taranik and Crosta, 1996). The low
accessibility of these frontier areas can be due to lack of road access
or geopolitical reasons. This is especially true for the following
case study in Lebanon — an area of little infrastructure and, as
of recent times, some geopolitical issues. SPECTIR hyperspectral
VNIR-SWIR imagery was collected for the northern onshore
parts of Lebanon for hydrocarbon evaluation and classification of
known seepage. In the study area, no field data were collected due
to geopolitical health and safety issues, with the only safe zone
area being declared as the city of Beirut (out of which the majority
of operations were based for this project). This meant classifications
were done without a priori knowledge in the form of field data.
Figure 12. Example surface TOC map with high TOC values classified in red,
Fortunately, two locations in the area were documented historical medium TOC values classified in yellow, and low TOC values classified in blue.
hydrocarbon show points. These were used to compare with The green dot symbolizes the ground survey area where numerous geochemical
processing and classification results. A spectral library database samples were taken and used as training for the hyperspectral data.

Special Section: Remote sensing January 2017 THE LEADING EDGE 29

 The two sample spectra that were used generated varying
results in classification. The geospatial intersect function in ArcGIS
was used to find areas of overlap between the two classifications.
The highest densities of this final classification coincided with
known mapped faults and contact boundaries of older exposed
strata, potentially indicating the existence of buried mature source
Downloaded 12/27/16 to 132.239.1.231. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
frontier areas where little is known about the area. This has been
shown in case studies such as that of Lebanon where known field
endmembers are not already properly established.
Detection of surface seepage and surface lithology can have
its disadvantages and limitations. What must be taken into account
during the evaluation of surface lithology and surface seepage is

rock. The majority of seepage points occurred on Jurassic and that these are measurements of the surface that do not in any way
Late Cretaceous rock. The highest densities of this classification
coincided well with the historical seepage points (Figure 13).
For this project, hyperspectral data were combined with 200 m
resolution radiometric data to obtain enhanced surface lithologic
mapping. An averaged 22-band mosaic was stacked with a 3-band
radiometric data set. An unsupervised isodata classification was
applied and compared with known surface geology maps for in-
terpretation and to attempt further enhancement of detail to the
surface geology already mapped. For this process, each class output
from the unsupervised classification was compared with the known
geology map, and links were made as to the overall surface variation
and the area’s geologic setting (Figure 14).
The geologic setting will impact the overall spectral signature
collected from the sensor, whether it is from variations in vegeta-
tion, variations in soil composition, or variations in surface lithol-
ogy characteristics. From these correlations, a surface variation
map was made, leading to an enhanced surface lithology map
(Figure 15). This product yielded improved insights on where the
mature source rock might be and higher precision on the exact
contact boundaries between geologic units.

Conclusions and discussion

Remote sensing has proven to be a highly effective tool for Figure 14. Method for overlaying various land cover classifications in comparison
oil and gas exploration by increasing the understanding of the with known geology map. Here the classified area in green is most associated with
hydrocarbon system from the surface. It is a highly useful tool for Lower Cretaceous.

Figure 13. Bitumen classification map using SPECTIR hyperspectral data. Two Figure 15. Final surface variation map where variations in color signify
areas pointed out are known documented hydrocarbon seepage locations, which differences in surface lithology obtained by combining radiometric and
coincide well with the classification results. hyperspectral data sets.

30 THE LEADING EDGE January 2017 Special Section: Remote sensing

 penetrate below the surface (Agar and Coulter, 2007). The use of
hyperspectral and multispectral technology cannot directly infer
information from the subsurface. This means that, in complex
geologic systems, what can be seen on the surface does not neces-
sarily reflect what might be happening directly below; this is
usually true for scenarios in which the formations at the surface
Downloaded 12/27/16 to 132.239.1.231. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
are younger than that of the latest tectonic event. For example,
this holds true for surface seepage, as varying migration pathways
can mean that reservoirs or producing source rocks are not always
directly below the seepage point. Similarly with outcropping rocks,
due to high complexity of variance in directional stresses, the
outcropping rock may be dipping away (Abrams, 2005) in another
direction, or may be completely folded over as we have seen in
Argentina. In the analysis of the hydrocarbon system in these
situations, additional remote sensing data types such as magnetic
or gravity data assist in creating a better understanding of sub-
surface anomalies that would cause some of these migration
pathways to form. Adding seismic data to the integration with
remote sensing data can enhance areas of potential faulting and
identify where the structures extend beyond the surface. An
understanding and integration of both subsurface and surface data
is of great assistance to any analysis (Schumacher, 2016). As previ-
ously mentioned, however, this is expensive; what remote sensing
can do is establish high-priority areas for further study.
This being said, remote sensing is a tremendously useful form
of data analysis for the earth’s surface in the oil and gas realm. In
Pennsylvania, USA, it was shown that seepage classifications
provided insights into leaking abandoned wells in addition to
naturally occurring seepage. Individual surveying of these numer-
ous wells by foot can be challenging. Therefore, being able to map
and locate seepage that could prove harmful to the populous and
the environment (regardless of the source being natural or man-
made) in a quick and cost-effective way can be extremely useful
for oil and gas planning and services.
In Argentina, hyperspectral airborne data was used to map
hydrocarbon seepage and mineral alteration associated with
microseepage. The lithologic variance allowed for key formations
such as the Vaca Muerta to be mapped and classified. The com-
bination of field data and expert geologic knowledge resulted in
the advancement in detail of lithologic mapping in the area. Also,
with the combined assistance of TOC lab data (analyzed from
field samples) surface TOC was mapped, establishing a better
assessment in the quality of potential source rocks and assisting
in the evaluation of shales and their locations.
Finally, in the Lebanon study, it was shown that a great deal
can be gained from remote sensing — even with a lack of field
data. This was shown through classification of hydrocarbon
seepage using bitumen samples and previously attained hydro-
carbon spectra. This was validated with the correlation of the
classification with known historical hydrocarbon show points.
However, issues do exist when picking reference spectra. In the
most ideal cases, reference endmembers should be collected from
the scene itself for direct correlation between field and airborne
spectra. For the Lebanon study, this was not available and hence
reference spectra were taken from regions outside of the study
area; specifically, two spectra from North America were used.
This can lead to additional uncertainty in the classification as
Special Section: Remote sensing
spectra from outside the study will not directly match with
spectra from the study area representing the “true” landcover
types (Lillesand et al., 2009). The analysis also included bitumen
samples taken from Curtis (2013) in which key absorptions were
established prior to classification. This was followed by intersect
functions in ArcGIS to determine areas where multiple seepage
classifications existed, thereby increasing the likelihood of its
existence. Then, similarly to the Argentina case study, compari-
sons were made with known geologic maps of the region, and,
using unsupervised classifications of hyperspectral data, greater
detail was given to the surface lithology, which in turn resulted
in improvements in the geologic maps. The Lebanon project is
an ideal study showing the applicability of remote sensing in
frontier areas where little ground data exists.
Advancing technology with newer sensors and newer satellite
payloads, with higher spatial and spectral resolutions, will only
increase the overall applicability of remote sensing (Lisica, 2015).
Lisica also makes reference to new superspectral sensors such as
that of WorldView3 (launched in August 2014) that combine
higher spatial/spectral resolutions in larger footprints, making
this new technology ideal for basin-wide and regional analyses.
Remote sensing, as shown here and in countless other studies,
has proven to be a unique, cost-effective way of understanding
broad frontier areas and high-grading areas for further study.
Corresponding author: blord@neosgeo.com
References
Abrams, M. A., 2005, Significance of hydrocarbon seepage relative
to petroleum generation and entrapment: Marine and Petroleum
Geology, 22, no. 4, 457–477, http://dx.doi.org/10.1016/j.marpet-
geo.2004.08.003.
Agar, B., and D. Coulter, 2007, Remote sensing for mineral explo-
ration – A decade perspective 1997-2007, in B. Milkereit , ed.,
Proceedings of Exploration 07: Fifth Decennial International
Conference on Mineral Exploration, 109–136.
Agrawal, G., and J. Sarup, 2011, Comparison of QUAC and
FLAASH atmospheric correction modules on EO-1 Hyperion
data of Sanchi: International Journal of Advanced Engineering
Sciences and Technologies, 4, no. 1, 178–186.
Almeida-Filho, R., F. P. Miranda, and T. Yamakawa, 1999, Remote
detection of a tonal anomaly in an area of hydrocarbon micro-
seepage, Tuscano basin, north-eastern Brazil: International Jour-
nal of Remote Sensing, 20, no. 13, 2683–2688, http://dx.doi.
org/10.1080/014311699212029.
Clark, R. N., and T. L. Roush, 1984, Reflectance spectroscopy:
Quantitative analysis techniques for remote sensing applications:
Journal of Geophysical Research, 89, no. B7, 6329–6340, http://
dx.doi.org/10.1029/JB089iB07p06329.
Curtis, B., 2013, Quantifying bitumen in natural oil sands using
NIR technolog y: ASD Inc., http://w w w.asdi.com/
getmedia/63538819-9aca-40fc-aa57-82a2e58c5a1c/Quantify-
ing_bitumen_in_natural_oil_sands_using_NIR_technology.pdf,
accessed 2 December 2016.
Du, Y., C. Chang, H. Ren, C. Chang, J. O. Jensen, F. M. D’Amico,
2004, New hyperspectral discrimination measure for spectral
characterization: Optical Engineering, 43, no. 8, 1777–1786,
http://dx.doi.org/10.1117/1.1766301.
January 2017 THE LEADING EDGE 31
 Ducart, D. F., A. M. Silva, C. L. B. Toledo, and L. M. de Assis, 2016, Mapping iron ox-
ides with Landsat-8/OLI and EO-1/Hyperion imagery from the Serra Norte iron de-
posits in the Carajás Mineral Province, Brazil: Brazilian Journal of Geology, 46, no.
3, 331–349, http://dx.doi.org/10.1590/2317-4889201620160023.
Ellis, J. M., H. H. Davis, and J. A. Zamudio, 2001, Exploring for onshore oil seeps with
hyperspectral imaging: Oil and Gas Journal, http://www.ogj.com/articles/print/vol-
ume-99/issue-37/special-report/exploring-for-onshore-oil-seeps-with-hyperspectral-
imaging.html, accessed 12 November 2016.
Downloaded 12/27/16 to 132.239.1.231. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/

Lillesand, T. M., R. W. Keifer, and J. W. Chipman, 2009, Remote sensing and image in-
terpretation: John Wiley & Sons, Inc.
Lisica, R., 2015, A new age for oil and gas exploration remote sensing data and analytics
are changing the industry: Earth Imaging Journal, http://eijournal.com/print/articles/a-
new-age-for-oil-and-gas-exploration-remote-sensing-data-and-analytics-are-chang-
ing-the-industry, accessed 12 November 2016.
Kruse, F. A., A. B. Lefkoff, and J. B. Dietz, 1993, Expert system-based mineral mapping
in northern Death Valley, California/Nevada using the airborne visible/infrared imag-
ing spectrometer (AVIRIS): Remote Sensing of Environment, 44, nos. 2–3, 309–336,
http://dx.doi.org/10.1016/0034-4257(93)90024-R.
McKee, B. and C. Beasley, Use of remote sensing technologies to detect surface and near-
surface stray gas occurrence and potential migration pathways in Tioga County, Penn-
sylvania, 2013, Search and Discovery Article #80317.
Naeth, J., R. di Primio, B. Horsfield, R. G. Schaefer, P. M. Shannon, W. R. Bailey, and
J. P. Henriet, 2005, Hydrocarbon seepage and carbonate mound formation: a basin
modelling study from the Porcupine Basin (offshore Ireland): Journal of Petroleum
Geology, 28, no. 2, 147–166, http://dx.doi.org/10.1111/j.1747-5457.2005.tb00077.x.
Plaza, A., and J. Plaza, 2011, Parallel implementation of linear and nonlinear spectral un-
mixing of remotely sensed hyperspectral images: proceedings of High-Performance
Computing in Remote Sensing, http://dx.doi.org/10.1117/12.897326.
Prelat, A., Gunaratne, S., Huebner, L., Freeman, C., Cook, A., and C. Soriano. 2013.
Airborne hyperspectral detection of natural offshore and onshore hydrocarbon seeps,
in F. Aminzadeh, T. B. Berge, and D. L. Connolly, eds., Hydrocarbon Seepage:
From Source to Surface: SEG/AAPG Geophysical Developments Series 16, 171–
182, http://dx.doi.org/10.1190/1.9781560803119.ch11.
Prost, G. L., 2002, Remote sensing for geologists: A guide to image interpretation: Tay-
lor & Francis.
Saunders, D. F., K. R. Burson, and C. K. Thompson, 1999, Model for hydrocarbon mi-
croseepage and related near surface alterations: AAPG Bulletin, 83, no. 1, 170–185.
Schneising, O., J. P. Burrows, R. R. Dickerson, M. Buchwitz, M. Reuter, and H. Bovens-
mann, 2014, Remote sensing of fugitive methane emissions from oil and gas produc-
tion in North American tight geologic formations: Earth’s Future, 2, no. 10, 548–558,
http://dx.doi.org/10.1002/2014EF000265.
Schumacher, D., 1996, Hydrocarbon-induced alteration of soils and sediments, in D.
Schumacher and M. A. Abrams, eds., Hydrocarbon Migration and Its Near-Surface
Expression: AAPG Memoir 66, 71–89.
Schumacher, D., 2016, Improving success of surface geochemical surveys: 7 pitfalls to
avoid: Search and Discovery Article #80504.
Smejkalova, E., and P. Bujok, 2012, Remote sensing methods in the identification of oil con-
taminations: GeoScience Engineering, 58, no. 1, 24–33, http://dx.doi.org/10.2478/v10205-
011-0010-6.
Taranik, J. V., and A. P. Crosta, 1996, Remote sensing for geology and mineral resources,
an assessment of tools for geoscientists in the future: International Archives of Pho-
togrammetry and Remote Sensing, 31, B7, 689–699.
van der Meer, F., P. Van Dijk, H. Van Der Werff, and H. Yang, 2002, Remote sensing
and petroleum seepage: a review and case study: Terra Nova, 14, no. 1, 1–17, http://
dx.doi.org/10.1046/j.1365-3121.2002.00390.x.
Veiga, G. D., L. A. Spalletti, J. A. Howell, and E. Schwarz, 2005, The Neuquen Basin,
Argentina: A case study in sequence stratigraphy and basin dynamics: Geological So-
ciety of London, Special Publications, 252, 1–14.
Young, S. J., R. B. Johnson, and J. A. Hackwell, 2002, An in-scene method for atmo-
spheric compensation of thermal hyperspectral data: Journal of Geophysical Research,
107, D24, ACH 14-1–ACH 14-20, http://dx.doi.org/10.1029/2001JD001266.

32 THE LEADING EDGE January 2017 Special Section: Remote sensing