Available via license: CC BY 4.0
Content may be subject to copyright.
Citation: Carrillo Ramirez, A.;
Gonzalez Penagos, F.; Rodriguez, G.;
Moretti, I. Natural H2Emissions in
Colombian Ophiolites: First Findings.
Geosciences 2023,13, 358. https://
doi.org/10.3390/geosciences13120358
Academic Editors: Jesus
Martinez-Frias and Fedor Lisetskii
Received: 26 September 2023
Revised: 16 November 2023
Accepted: 17 November 2023
Published: 22 November 2023
Copyright: © 2023 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
geosciences
Article
Natural H2Emissions in Colombian Ophiolites: First Findings
Alejandra Carrillo Ramirez 1,2, Felipe Gonzalez Penagos 3, German Rodriguez 3and Isabelle Moretti 2, *
1Facultad de Minas, Universidad Nacional de Colombia, Medellín 050034, Colombia; ncramirez@univ-pau.fr
2UPPA-LFCR, Rue de l’Université, 64013 Pau, France
3Atlas Research Group, Bogota 77152, Colombia
*Correspondence: isabelle.moretti@univ-pau.fr
Abstract:
The exploration of natural H
2
or white hydrogen has started in various geological settings.
Ophiolitic nappes are already recognized as one of the promising contexts. In South America, the
only data available so far concerns the Archean iron-rich rocks of the Mina Gerais in Brazil or the
subduction context of Bolivia. In Colombia, despite government efforts to promote white hydrogen,
data remain limited. This article introduces the initial dataset obtained through soil gas sampling
within the Cauca-Patia Valley and Western Cordillera, where the underlying geology comprises
accreted oceanic lithosphere. In this valley, promising areas with H2potential were identified using
remote sensing tools, in particular vegetation anomalies. The Atmospherically Resistant Vegetation
Index (ARVI) appears to be well adapted for this context and the field data collection confirmed the
presence of H
2
in the soil in all pre-selected structures. The valley undergoes extensive cultivation,
mainly for sugar cane production. While H
2
emissions lead to alterations in vegetation, unlike reports
from other countries, they do not result in its complete disappearance. Soil gas measurements along
the thrusts bordering the Cauca Valley also show high H
2
content in the fault zones. In the valley, the
presence of sedimentary cover above the ophiolites which are presumably the H
2
generating rocks,
which addresses the possible presence of reservoirs and seals to define potential plays. Drawing
parallels with the Malian case, it could be that the intrusive element could serve as seals.
Keywords: Colombia; natural H2; Cauca-Patia Valley; ophiolites; vegetation index
1. Introduction
1.1. Natural H2within the New Energy Mix
Within the context of the race to decarbonize the world energy mix, the dihydrogen,
which we will call hydrogen or H
2
, has a growing role both as an energy vector but also as
energy source. Natural H
2
is present in the subsurface from where it could be produced
and not manufactured. Exploration for natural hydrogen underway in several countries
including Australia, USA, Spain, and France [
1
]. Production has been a reality in Mali,
West Africa, for a decade but remains unfortunately limited due to security and political
issues [
2
]. The expectations regarding natural H
2
are high since the expected price for this
carbon-free H
2
is lower than the cost for green hydrogen which is manufactured through
water electrolysis. However, given that natural H
2
is a natural resource, its production
relies on appropriate regulations and various issues need to be addressed regarding its
implementation. Knowledge on H
2
generation/migration/accumulation processes is
progressing but acquisition of significant new data needs to continue. The advantages
and disadvantages of natural H
2
, its environmental impact, and its renewability must
be evaluated globally on a case-by-case basis. Life Cycle Analyses significantly hinge on
specific actions taken, but overall, it is expected to be more favorable than the production of
green hydrogen through electrolysis [
3
]. The goal of this paper is to contribute to the global
knowledge on natural hydrogen by sharing new information and data from Colombia.
In this country, which is rich in oil, coal, and other natural resources, natural H
2
is an
emerging topic of interest for geologists as well as regulators alike, especially in areas off
Geosciences 2023,13, 358. https://doi.org/10.3390/geosciences13120358 https://www.mdpi.com/journal/geosciences
Geosciences 2023,13, 358 2 of 17
the grid. After briefly reviewing of the ongoing changes in the Colombian sub-surface
legislation, we will discuss the H
2
natural systems known in other countries and present
the first finding on a similar system in the Cauca-Patia Valley of Colombia.
1.2. Natural H2a New Resource Recognized by the Law in Colombia
To promote a just energy transition and diversify the country energy matrix, Colom-
bia’s national development plan, approved for 2022–2026, includes tax benefits of natural
hydrogen as a non-conventional renewable energy source (FNCER in Spanish). This initia-
tive aims to encourage the exploration and production of natural resources for production
and development of previously unproductive land. Areas without in situ resources or
that are non-connected to the grid are particularly targeted for the exploration of these
low-carbon resources. This regulatory framework has given the Colombian Ministry of
Mines and Energy the power to authorize interested parties to develop natural hydrogen
and related natural gas projects. The Ministry is also responsible for determining the guide-
lines, requirements, technical, economic, financial, and legal conditions for the granting
and executing such authorization.
2. Natural H2Systems within the Colombian Geological Context
2.1. H2Prospectivity in Colombia
The rocks that generate H
2
in subsurface started to be rather well known and could
be classified following [
4
] in 4 categories: oceanic and mantellic rocks (H
2
-GR1), iron-rich
sediments and intrusive rocks (H
2
-GR2), radioactive rocks (H
2
-GR3), and organic-rich
source rocks, especially coal (H
2
-GR4). In the first three cases, the H
2
comes from water
(redox reactions or radiolysis, respectively), and in the last category, from organic matter. In
ophiolitic contexts such as Oman or New Caledonia, the main H
2
generating rock category
is H
2
-GR1 [
5
]. In southern and western Australia, and within the Mina Gerais in Brazil, the
oxidation of Banded Iron Formation (BIF) is likely the major source of H
2
[
6
,
7
]. In South
Africa and in Canada, H
2
-GR3 [
8
,
9
]) has been proposed. In the Songliao Basin in China, or
in the Cooper Basin in Australia, the authors favor H
2
-GR4 [
10
,
11
]. Some other reactions
may generate H
2
but do not look currently the most promising in terms of volume in the
zones of active H2exploration and production.
In Colombia, due to the accretion of oceanic terranes during the subduction along the
Pacific coast (Figure 1), at least three of the H
2
Generating Rocks are present: ophiolites,
coal, and the iron-rich facies. Figure 2shows a schematic cross-section from the Pacific to
the Amazonian Craton through the various geological provinces of Colombia where the
search for H2could be targeted.
Geosciences 2023,13, 358 3 of 17
Geosciences 2023, 13, x FOR PEER REVIEW 3 of 19
Figure 1. Geological terranes of Colombia and main domains, modified from [12].
Columbian geology is complex and characterized by active tectonics with two sub-
duction zones. To the west along the Pacific coast, the Nasca plate is subducting, and to
the north, the oblique subduction of the Caribbean plate is taking place toward the south-
southeast (Figure 1). The roughly N–S oriented main reliefs consist of the Eastern Cordil-
lera, an inversed Mesozoic back arc basin, and the Central and Western Cordillera which
are accreted terranes and contain ophiolitic series (Figure 2). To the east, the Llanos Basin
is the foreland of the Eastern Cordillera thrust belt. It contains a large part of the Colom-
bian oil resources [13] and references therein. The Magdalena Valley between the Eastern
and Central Cordillera is also an oil and gas (O&G) province [14]. Westward, the Cauca-
Patia Valley is in another geological context where accretional material and ophiolites are
dominant. Despite some exploration efforts and acquisition of subsurface data, the Cauca-
Patia Valley has never been successful for O&G; however, the valley could be promising
regarding natural H
2
.
Figure 1. Geological terranes of Colombia and main domains, modified from [12].
Columbian geology is complex and characterized by active tectonics with two sub-
duction zones. To the west along the Pacific coast, the Nasca plate is subducting, and
to the north, the oblique subduction of the Caribbean plate is taking place toward the
south-southeast (Figure 1). The roughly N–S oriented main reliefs consist of the Eastern
Cordillera, an inversed Mesozoic back arc basin, and the Central and Western Cordillera
which are accreted terranes and contain ophiolitic series (Figure 2). To the east, the Llanos
Basin is the foreland of the Eastern Cordillera thrust belt. It contains a large part of the
Colombian oil resources [
13
] and references therein. The Magdalena Valley between the
Eastern and Central Cordillera is also an oil and gas (O&G) province [
14
]. Westward,
the Cauca-Patia Valley is in another geological context where accretional material and
ophiolites are dominant. Despite some exploration efforts and acquisition of subsurface
data, the Cauca-Patia Valley has never been successful for O&G; however, the valley could
be promising regarding natural H2.
Geosciences 2023,13, 358 4 of 17
Geosciences 2023, 13, x FOR PEER REVIEW 3 of 19
Figure 1. Geological terranes of Colombia and main domains, modified from [12].
Columbian geology is complex and characterized by active tectonics with two sub-
duction zones. To the west along the Pacific coast, the Nasca plate is subducting, and to
the north, the oblique subduction of the Caribbean plate is taking place toward the south-
southeast (Figure 1). The roughly N–S oriented main reliefs consist of the Eastern Cordil-
lera, an inversed Mesozoic back arc basin, and the Central and Western Cordillera which
are accreted terranes and contain ophiolitic series (Figure 2). To the east, the Llanos Basin
is the foreland of the Eastern Cordillera thrust belt. It contains a large part of the Colom-
bian oil resources [13] and references therein. The Magdalena Valley between the Eastern
and Central Cordillera is also an oil and gas (O&G) province [14]. Westward, the Cauca-
Patia Valley is in another geological context where accretional material and ophiolites are
dominant. Despite some exploration efforts and acquisition of subsurface data, the Cauca-
Patia Valley has never been successful for O&G; however, the valley could be promising
regarding natural H
2
.
Figure 2.
Schematic E–W cross-section showing structural features of central part of Colombia at
the level of Cali. The vertical exaggeration is about 1.5. Westward, oceanic terranes related to the
various accretions could be prospective H
2
provinces due to the presence of ophiolitic nappes. The
mantellic wedge between the Nazca plate and South America may also play a role. The Cauca-Patia
Basin is the area studied in this paper. This conceptual cross-section is based on the work of [15] for
the western part and for the eastern part from [16].
2.2. Serpentinization and H2Generation
The oceanic lithosphere generates H
2
during its alteration by hot water. The main
reactions could be summarized as follow:
Olivine + water => serpentinite + magnetite + brucite + H2
Orthopyroxene + water => serpentine + magnetite + H2 + silicon dioxide
These reactions are widely described in detail in the literature [
4
,
17
,
18
] and references
therein. To sum up, olivine oxidizes to serpentinite, water is reduced, and H
2
is released.
The global reaction takes place in several stages: oxidation of olivine generates magnetite,
itself rich in Fe
2+
, which is then oxidized. This reaction has been first described in the Mid
Oceanic Ridge (MOR) [
19
], the kinetic of this reaction has been studied and happen to be
optimal around 300–340
◦
C but started before [
20
]. The olivine content is usually high in
the mantle part of the oceanic lithospheres, low in the basalts of the upper oceanic crust and
intermediary in the gabbros of the lower oceanic crust. As a result, the H
2
content in the
hydrothermal vents is rather high in the slow spreading center where the mantle reaches
the sea bottom [
21
]. When the oceanic lithosphere is getting colder and moves away from
the MOR, the serpentinization process is usually not complete and additional H
2
could be
generated when the Pression and Temperature (PT) conditions are becoming optimal again.
These various steps of serpentinization have been described and even dated in Turkey [
22
].
In Oman, as in New Caledonia, a blocked subduction induces the thrusting of an
oceanic lithosphere above sedimentary rocks. High H
2
content has been measured in
various bubbling sources, over 70% in Oman [
23
]. The gas however is often a blend
between H
2
, N
2
, and CH
4
[
5
,
24
]. Onshore, the reactant is supposed to be rain water,
but when in the MOR, it is sea water [
25
,
26
]. The temperature and depth of the onshore
serpentinization could be questioned. If the 300
◦
C suggested for the MOR corresponds to
a depth of about 10 km in a regular continental geothermal gradient, alternative reactions
such as the alteration of magnetite may be efficient to generate H
2
at lower temperature [
27
].
In this case, hydrogen could be generated in the first kilometers of the curst. Since ophiolites,
and more generally, pieces of accreted oceanic lithosphere are present in Colombia, we first
focused our study on determining if H
2
could be generated there by any of the processes
listed above.
2.3. H2Exploration Workflow
The H
2
systems, encompassing generation, migration, and accumulation, are not yet
fully understood, but it has been noted by various authors that H
2
gas seeps are rather
numerous and can be used to start the natural H
2
exploration [
28
–
30
]. The emissions of H2
have an impact on vegetation, and this effect can be discerned through satellite imagery
Geosciences 2023,13, 358 5 of 17
or by employing the vegetation index derived from Landsat images [
31
]. Mapping these
emissions facilitates the precise targeting of field acquisitions. A methodology, outlined by
the mapping of H2-generating rocks, examination of vegetation anomalies, and a review of
existing datasets, has been detailed [4] and implemented in the Cauca-Patia Valley.
In this article, we first present the geological setting of the studied area with existing
data on the oceanic lithosphere involved in the terranes of west Colombia. The pre-
fieldwork study conducted on the identification of vegetation anomalies will then be
described. Next, we present the soil gas survey results and assess the potential for natural
H2resources of the Cauca-Paita Valley.
3. Pre-Field Trip: Colombian Ophiolites of the Cauca Valley
3.1. The Accreted Ophiolites of the Western Part of Colombia
Western Colombia is composed of terranes accreted to the Amazonian craton during
several geological periods from the Neoproterozoic to the Miocene. These terranes were
named after pre-Columbian ethnic groups to avoid confusion with names of Colombian
formations, groups, or geological provinces [
12
]. The main oceanic terranes, Mesozoic
in age, are the Calima and Cuna terranes which have been accreted to South America
during the Paleocene and Oligocene [
32
]. In addition, northward, in the Caribbean Basin,
rocks of oceanic origin are also present [
33
], which are considered to be part of the Tairona
terrane [32].
Western Colombia can be divided into several structural complexes arranged as
parallel strips bounded by fault zones. From east to west, we find the Quebradagrande
Complex, the Arquía Complex, both of which are part of the Eastern Cordillera, the
Amaime Complex, which is the basement of the Cauca Paitia Valley, the Western Cordillera
Complex, and the ChocóComplex, which is part of the Cuna Terrane (Figure 1). Each
of these complexes has a different geological origin and evolution. The Quebradagrande
Complex represents remnants of deposits that originated between the northwestern edge
of South America and the Amaime–Chaucha volcanic arc, as well as fragments of oceanic
crust from the Proto-Caribbean. The Arquía Complex is a tectonic mixture of blocks from
different origins, produced by shearing between the Cretaceous Caribbean-Colombian
Igneous Province and the western margin of Colombia during the Cretaceous. The Amaime
Complex is composed of fragments from the Amaime–Chaucha volcanic arc, oceanic
plateau basalts, and volcanic sedimentary rocks that were accreted diagonally to the
western edge of Colombia during the Late Cretaceous. The Amaine Fm is dated from
90.6 Myr
±
2.5 Myr [
34
]. The Western Cordillera Complex corresponds to a portion of
the Caribbean plate that thickened (oceanic plateau) and began to accrete to the western
margin of Colombia starting from the Paleocene [35].
The Colombian geodynamic framework has allowed the accretion of ophiolitic rocks
over time. The focal point of interest for this study is the Ginebra Ophiolitic Massif, as
defined by [
36
]. This constitutes a block of ultramafic and mafic rocks of the Jurassic to the
Lower Cretaceous age. Located on the western flank of the Central Cordillera, it is bounded
to the east by the Guabas–Pradera faults and on the west by the Palmira–Buga faults [
37
].
These ophiolites are composed of three primary rock groups: amphibolites, gabroic rocks,
and ultramafic rocks. Amphibolites are found at the southern and southwestern limits of
the Ginebra Ophiolitic Massif and are primarily composed of hornblende and plagioclase.
Gabbroic rocks encompass troctolites, gabbronorites, gabbros, and hornblende gabbros.
The ultramafic peridotitic bodies, including harzburgites, lherzolites, and wehrlites, along
with pyroxenites, are situated in the central and southern portions of the Ginebra Ophiolitic
Massif [38].
The Buga Batholith is an intrusive igneous body located in Western Colombia. It
covers an area of 200 km
2
and is considered one of the oldest in the region with an age
of
90.6 ±1.3 Ma [39]
, which is contemporaneous of the Amaime Fm [
34
]. It outcrops in
the municipalities of Buga, San Pedro, and Tuluáin the Valle del Cauca department and
intrudes two main units: the Ginebra Ophiolitic Massif and the basalts of the Amaime
Geosciences 2023,13, 358 6 of 17
Formation [
40
]. According to [
34
], three distinct facies are defined within this unit of
holocrystalline phaneritic rocks. The first facies is composed of leucocratic tonalites, which
contain medium-to-coarse crystals of quartz and plagioclase, with mafic minerals, pre-
dominantly biotite. The second facies consists of mesocratic tonalites, which contain
medium-sized crystals with quartz, plagioclase, hornblende, and biotite occurring in simi-
lar proportions. The third facies corresponds to melanocratic diorites, quartz-diorites, and
tonalites, which are composed of medium to fine-sized crystals of amphiboles and biotite.
Another unit corresponding to the study area is the Amaime Formation, named by [
41
].
This Amaime Formation is a set of basic volcanic rocks found on the western flank of the
Central Cordillera in the Valle del Cauca department [
37
]. These rocks consist of a series of
massive toleitic basalts with abundant pillow lava horizons located to the west of the Cauca–
Almaguer fault [
42
,
43
]. These basalts have a porphyritic texture and a more predominant
aphanitic texture. Additionally, they exhibit a degree of crystallinity ranging from glassy to
holocrystalline [42].
Finally, there are the alluvial deposits of the Quaternary system, which contain sands,
silts and clays deposited in river channels, terraces, and flood plains. These deposits
are unconsolidated, heterometric, and heterogeneous with variable thicknesses and are
generally unstratified [43].
3.2. Remote Sensing Approach of the Cauca-Paita Valley
Anomalies of vegetation that could be related to gas escape were identified in Cauca-
Patia Basin through satellite imagery from various providers, such as the U.S. Geological
Survey (USGS), the National Aeronautics and Space Administration (NASA), and the
European Space Agency (ESA), which offer high spatial resolution. These anomalies
correspond to small depressions in the soil, also known as fairy circle or Sub-Circular
Depressions (SCD) in the literature [
31
]. They have been previously studied in Russia [
28
],
the United States [
29
], and Brazil [
44
]. For this article, the term “fairy circles” has been
used for the studied Colombian structures, as it is more illustrative and because the areas
of vegetation anomalies do not exhibit depressed areas or if so, only very subtle ones.
In the Cauca-Patía Valley, 23 fairy circles have been identified (location on Figure 3
and satellite image Figure 4), ranging in size from 10 to 60 m
2
. Some of these circles are
located in sugarcane fields. The pattern is different from that of other fairy circles around
the world: vegetation grows within these fairy circles. We note different configurations:
(1) in some sugarcane fields, the height of the canes is slightly less than that of the canes
outside the circles (Figure 4a–c), and (2) there is something else, planted or naturally grown
within the circle (Figure 4d). The images show a combination of reduced vegetation in a
relatively circular area and a rectangular or square area framing a depression. The fairy
circles looked relatively circular, but the farmer clearly does not try to cultivate the area
around. They may have known from the previous years that it was not useful.
These fairy circles are rather small, but they are also clearly visible on Google Earth
satellite images. They are less visible on the Landsat image because the resolution of our
dataset is of 30 m. We use of the multispectral data attempting to get an automatic detection
of the gas micro-seepage areas (Figure 5). The method which has been proposed to date
was based on the vegetation indexes, such as the soil adjusted vegetation index (SAVI)
computed from Landsat infra-red data [
4
,
31
]. We have used Sentinel-2 to get a better
resolution (10–13 m instead of 30). We also extended the computed parameters as shown
below and synthetized Table 1.
Geosciences 2023,13, 358 7 of 17
Geosciences 2023, 13, x FOR PEER REVIEW 7 of 19
Figure 3. Map of the Cauca-Patia Valley studied area with the locations of the cross-sections. The
red dots are the locations of soil gas measurements, the orange points are the location of the vege-
tation anomalies where soil gas measurement have not been performed. Initial data from map 300
from the Geological Service of Colombia. Projection WGS84.
Figure 4. Four of the vegetation anomalies in the Cauca-Patia Basin (a–c) are located in sugar cane
field; (d) is the ST3, the circle is full of bamboo.
Figure 3.
Map of the Cauca-Patia Valley studied area with the locations of the cross-sections. The red
dots are the locations of soil gas measurements, the orange points are the location of the vegetation
anomalies where soil gas measurement have not been performed. Initial data from map 300 from the
Geological Service of Colombia. Projection WGS84.
Geosciences 2023, 13, x FOR PEER REVIEW 7 of 19
Figure 3. Map of the Cauca-Patia Valley studied area with the locations of the cross-sections. The
red dots are the locations of soil gas measurements, the orange points are the location of the vege-
tation anomalies where soil gas measurement have not been performed. Initial data from map 300
from the Geological Service of Colombia. Projection WGS84.
Figure 4. Four of the vegetation anomalies in the Cauca-Patia Basin (a–c) are located in sugar cane
field; (d) is the ST3, the circle is full of bamboo.
Figure 4.
Four of the vegetation anomalies in the Cauca-Patia Basin (
a
–
c
) are located in sugar cane
field; (d) is the ST3, the circle is full of bamboo.
Geosciences 2023,13, 358 8 of 17
Geosciences 2023, 13, x FOR PEER REVIEW 10 of 19
Figure 5. Eight Spectral Filters in the Cauca-Paita Valley highlighting the vegetation anomalies. See
text and Table 1 for acronym definitions. Yellow arrows indicate the North.
4. Soil Gas Measurements
The data of the soil gas measurement are presented Table 2. Due to the hardness of
the soils, or rocks near the faults, it was not possible to drill down to the standard 80/100
cm. The “depth” column indicates the depth at which the gas was pumped. The hours
have been also noted for future work. In this study, sufficient time in the field was not
allocated to collect time series to evaluate the gas-content variations around a full day
period. Consequently, the soil gas data at each sampling location represent only a one-
time sampling event. At first glance, one may note that H
2
is present in all the selected
zones. The highest concentrations of hydrogen were along the fault zone and elevated
concentrations of H
2
content were also detected in the fairy circles.
Table 2. Soil gas measurement, BIOGAS 5000 (GA).
Station Sample Date 2022 Hou
r
Depth (cm) O
2
(%) CH
4
(%) CO
2
(%) H
2
ppm CO ppm H
2
S ppm BAL (%)
H
2
_ST1 scd
1 12/10 10:10 40 20.2 0 0.1 1 1 1 79.7
2 12/10 10:22 40 19.9 0 0.3 22 1 1 79.9
3 12/10 10:25 80 19.7 0 0.4 14 1 2 80
4 12/10 10:35 80 8.5 0 15.4 18 3 2 76
5 12/10 10:36 80 0.2 0 25.6 33 1 2 74.2
6 12/10 10:47 80 18.6 0 3.4 137 17 2 79.8
7 12/10 10:51 80 19.4 0 1.3 320 8 7 79.6
8 12/10 10:56 80 17 0 12.8 289 7 2 79.3
9 12/10 11:00 60 19.9 0 0.7 54 2 1 79.7
10 12/10 11:07 80 19 0 8.3 255 5 2 79.3
11 12/10 11:14 80 19.9 0 1.8 40 3 1 79.4
12 12/10 11:18 75 7 0 15.7 140 8 2 78
Figure 5. Eight Spectral Filters in the Cauca-Paita Valley highlighting the vegetation anomalies. See
text and Table 1for acronym definitions. Yellow arrows indicate the North.
Table 1. Vegetation indexes: spectral band used and calculation.
Vegetation Index Spectral Bands Calculation Range of Values Interpretation
NDVI (Normalized Difference
Vegetation Index) NIR and R (NIR −R)/(NIR + R) −1 to 1 High values indicate the
presence of vegetation
SAVI (Soil-Adjusted
Vegetation Index) NIR and R
(NIR
−
R)/(NIR + R + L)
×
(1 + L)
−1 to 1
High values indicate the
presence of vegetation,
even on soils with high
reflectance
NDWI (Normalized Difference
Water Index) NIR and SWIR (NIR −SWIR)/(NIR + SWIR) 0 to 1 High values indicate the
presence of water
GLI (Green Leaf Index) B, G, and R (NIR + R + B + G)/4 0 to 1 High values indicate the
presence of chlorophyll
EVI (Enhanced
Vegetation Index) NIR, R, B, and G 2.5 ×(NIR −R)/(NIR + 6 ×R−
7.5 ×B + 1) −1 to 1 High values indicate the
presence of leaf biomass
ARVI (Atmospherically
Resistant Vegetation Index) NIR and R (NIR −R)/(NIR + R) −(NIR −
SWIR)/(NIR + SWIR) −1 to 1
High values indicate the
presence of vegetation,
even under adverse
atmospheric conditions
GNDVI (Greenness
Normalized Difference
Vegetation Index)
G and R (G −R)/(G + R) −1 to 1
High values indicate the
presence of vegetation,
even under adverse
atmospheric conditions
NBRI (Near-Infrared and
Shortwave Infrared Burned
Area Index)
NIR and SWIR (NIR −SWIR)/(NIR + SWIR) 0 to 1 High values indicate the
presence of burned areas
Geosciences 2023,13, 358 9 of 17
The Normalized Difference Vegetation Index (NDVI) is an important multispectral
index for tracking the physiological dynamics of key plant features such as biomass,
nitrogen levels, and leaf area. It is the most widely used remote sensing vegetation index
and is based on different ratios of reflected energy in the near-infrared reflectance and red
proportion of the light spectrum. It has been used to predict crop yield in the field and
its relationship with yield under conditions of drought, heat, and biotic stress has been
demonstrated [45].
The Atmospherically Resistant Vegetation Index (ARVI) is a multispectral vegetation
index used to monitor and measure biomass, vegetation density, and crop health. The
main advantage of ARVI lies in its ability to better correct for atmospheric effects compared
to NDVI. ARVI’s atmospheric resistance is achieved through a self-correction process for
the atmospheric effect in the red channel. It uses the difference in radiance between the
blue and red channels to correct the red channel’s radiance. This allows for more accurate
vegetation measurements in areas with varying atmospheric conditions. Simulations
utilizing radiative transfer calculations in arithmetic and natural surface spectra, under
diverse atmospheric conditions, show that ARVI has a similar dynamic range to NDVI but
is, on average, four times less sensitive to atmospheric effects than NDVI. This makes ARVI
particularly useful in regions with high atmospheric variability and in studies that require
precise measurements [46].
The Greenness Normalized Difference Vegetation Index (GNDVI) is a multispectral
vegetation index used to measure the amount of vegetation in a given area. Unlike NDVI,
GNDVI is based on reflectance in the green band (G) rather than the red band (R). The
idea behind this modification is that the green band is less sensitive to environmental
and atmospheric effects, allowing for more precise vegetation measurements. GNDVI has
been successfully used to monitor and predict crop yield under various environmental
conditions, such as drought, heat, and biotic stress. Additionally, GNDVI is less sensitive
to the effects of direct sunlight and humidity compared to other indices [47].
The Short-Wave Infrared and Near-Infrared Burned Area Index (NBRI) is a multispec-
tral vegetation index used for the detection and monitoring of burned areas. Its calculation
is based on the B8a near-infrared and B11 short-wave infrared bands, normalized using the
same spectral bands. NBRI relies on the premise that burned areas exhibit distinct spectral
characteristics compared to green areas, making it a valuable tool for highlighting these
regions in satellite images. It has been successfully employed in detecting burned areas in
forests and natural environments and is considered a valuable tool for wildfire monitoring
and management. Furthermore, NBRI is less sensitive to environmental and atmospheric
effects compared to other vegetation indices, enabling more accurate measurements of
burned areas [48].
The Green Leaf Index (GLI) is a measure of reflectance in the blue, green, and red
bands within the visible spectrum, and is used to assess the total chlorophyll content of
leaves. It is highly correlated with maize nitrogen content and has long been used to
monitor whether crops have sufficient nitrogen. By quantifying visible color differences
related to nitrogen content, the GLI could serve as an additional factor to amplify nitrogen-
based differences between crops. Blue reflectance has also been used to detect differences
in chlorophyll concentrations or other carotenoids related to nitrogen and water stress. It
is included in the calculation of the Enhanced Vegetation Index (EVI), which has proven
effective in distinguishing between soil and vegetation, suggesting that blue reflectance
might be more sensitive to stress affecting vegetation cover rather than color [49].
The Enhanced Vegetation Index (EVI) is a multispectral vegetation index used to
monitor and measure biomass, vegetation density, and crop health. It was developed
to extend sensitivity to high leaf biomass while minimizing the impact of aerosols and
background uncertainty sources. EVI has been successfully used to monitor and predict
crop yields under different environmental conditions, such as drought, heat, and biotic
stress. Additionally, EVI has been employed to differentiate between soil and vegetation,
making it particularly useful in remote sensing studies and vegetation cover monitoring.
Geosciences 2023,13, 358 10 of 17
Overall, EVI is considered one of the most accurate and reliable vegetation indices available
for crop improvement and vegetation monitoring studies [50].
Based on this analysis, soil gas measurements were planned for the accessible vegeta-
tion anomalies in the valley and the faults near Ginebra (see Figure 3).
4. Soil Gas Measurements
The data of the soil gas measurement are presented Table 2. Due to the hardness
of the soils, or rocks near the faults, it was not possible to drill down to the standard
80/100 cm. The “depth” column indicates the depth at which the gas was pumped. The
hours have been also noted for future work. In this study, sufficient time in the field was
not allocated to collect time series to evaluate the gas-content variations around a full
day period. Consequently, the soil gas data at each sampling location represent only a
one-time sampling event. At first glance, one may note that H
2
is present in all the selected
zones. The highest concentrations of hydrogen were along the fault zone and elevated
concentrations of H2content were also detected in the fairy circles.
Table 2. Soil gas measurement, BIOGAS 5000 (GA).
Station Sample Date
2022 Hour Depth
(cm) O2(%) CH4(%) CO2(%) H2ppm CO ppm H2S
ppm BAL (%)
1 12/10 10:10 40 20.2 0 0.1 11 1 79.7
2 12/10 10:22 40 19.9 0 0.3 22 1 1 79.9
3 12/10 10:25 80 19.7 0 0.4 14 1 2 80
4 12/10 10:35 80 8.5 0 15.4 18 3 2 76
5 12/10 10:36 80 0.2 0 25.6 33 1 2 74.2
6 12/10 10:47 80 18.6 0 3.4 137 17 2 79.8
7 12/10 10:51 80 19.4 0 1.3 320 8 7 79.6
8 12/10 10:56 80 17 0 12.8 289 7 2 79.3
9 12/10 11:00 60 19.9 0 0.7 54 2 1 79.7
10 12/10 11:07 80 19 0 8.3 255 5 2 79.3
11 12/10 11:14 80 19.9 0 1.8 40 3 1 79.4
12 12/10 11:18 75 7 0 15.7 140 8 2 78
13 12/10 11:23 80 19 0 3.8 300 11 2 79.2
14 12/10 11:27 80 19 0 2.5 112 4 1 78.9
H2_ST1
scd
15 12/10 11:31 30 20.5 0 0.4 40 11 1 79.2
H2_ST2
scd
1 12/10 16:11 40 21.1 0 0.4 29 8 0 78.5
2 12/10 16:17 50 20.6 0 6 250 4 1 78.4
3 12/10 16:19 40 21.2 0 0.2 30 1 0 78.6
4 12/10 16:27 30 21.2 0 1.1 14 3 1 78.6
5 12/10 16:30 30 21.1 0 0.1 82 1 78.7
6 12/10 16:34 30 21.1 0 0,2 71 1 78.8
7 12/10 16:38 25 20.6 0 1.1 21 1 78.4
1 12/10 13:52 80 20.7 0 0.9 19 5 0 78.4
2 12/10 13:56 80 20.2 0 0.9 24 15 1 78.6
3 12/10 14:00 50 20.5 0 1 13 1 0 78.5
4 12/10 14:03 80 20.5 0 1 10 1 0 78.5
5 12/10 14:06 40 20.7 0 0.7 11 1 0 78.6
6 12/10 14:09 50 20.1 0 1.2 60 17 1 78.7
7 12/10 14:13 70 20 0 1.4 73 1 78.6
8 12/10 14:16 80 19.6 0 2 94 1 78.4
9 12/10 14:19 80 20.5 0 1.7 64 7 1 78.6
H2_ST3
scd
10 12/10 14:23 80 20.3 0 1 10 1 1 78.6
Geosciences 2023,13, 358 11 of 17
Table 2. Cont.
Station Sample Date
2022 Hour Depth
(cm) O2(%) CH4(%) CO2(%) H2ppm CO ppm H2S
ppm BAL (%)
H2_ST4
1 13/10 9:30 40 16.2 0.1 3.2 220 10 1 80.5
2 13/10 9:40 40 19.4 1 1.3 239 4 1 79.3
3 13/10 9:41 40 19 0 1.2 22 2 1 79.8
4 13/10 9:48 20 18.5 0 1.8 42 6 1 79.6
5 13/10 9:55 10 20.6 0 0.2 72 1 79.3
6 13/10 10:00 15 20.3 0 0.2 22 8 1 79.4
7 13/10 10:04 15 20 0 0.8 29 16 1 79.3
8 13/10 10:09 80 18.6 0 2.2 18 7 1 79.3
9 13/10 10:16 25 20 0 0.7 22 5 1 79.3
1 13/10 10:54 40 20.3 0 0.6 15 3 1 79.1
H2_ST5 2 13/10 10:59 40 18.7 0 2.6 10 3 1 78.8
H2_ST6
1 13/10 11:12 40 20.6 0 0.5 29 5 1 79
2 13/10 11:16 25 20.7 0 0.2 12 1 1 79.1
3 13/10 11:19 30 20.5 0 0.8 35 13 1 79
1 13/10 11:27 70 19.3 0 2.2 88 22 1 78.9
H2_ST7 2 13/10 11:30 50 20.6 0 0.7 142 10 1 79
H2_ST8
1 13/10 11:42 25 17.6 0 3.6 72 4 1 78.8
2 13/10 11:44 25 17.5 0 4 87 16 1 78.8
3 13/10 11:47 40 19.8 0 1.7 90 7 1 78.9
4 13/10 11:51 45 17.8 0 3.1 52 4 1 78.8
1 13/10 13:21 20 21.1 0 0.1 20 2 0 78.9
2 13/10 13:24 20 20.9 0 0.5 330 9 1 78.8
3 13/10 13:28 30 20.8 0 0.5 >1000 11 2 78.9
H2_ST9
4 13/10 13:36 40 20.6 0 0.7 22 3 1 78.8
H2_ST10 1 13/10 13:59 25 21.1 0 0.1 50 15 1 78.8
2 13/10 14:05 30 19.2 0 1.6 17 4 1 79.2
1 13/10 14:26 30 20 0 1.4 72 8 1 78.8
2 13/10 14:49 20 21.1 0 0.3 86 2 1 78
H2_ST11
3 13/10 14:59 35 21.5 0 0 110 2 0 78.5
H2_ST12
1 14/10 9:08 40 20.8 0.1 0.3 47 13 1 79
2 14/10 9:12 30 20.2 0.1 0.9 38 13 1 79
3 14/10 9:18 40 20.7 0 0.3 33 2 1 79.1
4 14/10 9:20 35 20.3 0 0.5 13 9 1 79.2
1 14/10 9:36 30 18.7 0.1 2.6 41 1 78.8
H2_ST13 2 14/10 9:39 35 20.1 0 0.8 17 7 1 79
H2_ST14
1 14/10 9:47 15 20.5 0 0.3 10 1 1 79.2
2 14/10 9:52 20 20.6 0 0.4 42 1 79.1
3 14/10 9:54 30 19.3 0 1.1 41 1 79.6
1 14/10 10:30 35 16 0.1 3.2 77 10 1 81.9
2 14/10 10:35 30 16.6 0 2.8 58 4 1 80.4
3 14/10 10:39 30 18.6 0 1.8 38 5 1 79.5
4 14/10 10:39 30 12.6 0 6.9 74 5 1 80.4
5 14/10 10:41 45 8 0 13.3 17 3 1 79.4
6 14/10 10:49 35 20.1 0 1 60 7 1 78.9
H2_ST15
7 14/10 10:52 40 19.7 0 1.2 82 1 79.1
H2_ST16
1 14/10 12:19 40 20.3 0.1 0.7 65 7 1 79
2 14/10 12:22 80 11.2 0 9 60 1 79.8
3 14/10 12:25 70 20 0 0.7 33 3 1 79.1
4 14/10 12:28 50 20.4 0 0.3 31 2 1 79.2
4.1. Zones of Vegetation Anomalies
Due to access difficulties and time constraints, gas was only measured in 3 of the
13 structures identified by remote sensing. As shown in Table 2, values exceed 200 ppm ST1
and ST2. Overall, the values are more homogeneous that the ones published in other SCD’s
Geosciences 2023,13, 358 12 of 17
(Namibia or Brazil) and much higher than those for Australia [
51
]. The methane content is
neglectable, lower than 1%, and the CO
2
content is rather variable with a maximum value
of 25%. This fairy circle (SCD-1) is located on a farm where various animals including
horses and cows are present.
4.2. Soil Gas Measurement on the Fault Zones
Two cross-sections have been realized to measure the gas near the fault system that
border the Mesozoic ultramafic body of Ginebra. The stations H2_ST4 to 11 are in the
southern B-B’ section and the ST12 to ST16 are in session C-C’. We tested the presence
of H
2
within the fault zone, considering that it could be a migration pathway for gases
generated at greater depths. We also aimed to measure the disparity between the hanging
wall and the footwall, specifically between the valley where a sedimentary layer overlays
the ophiolites and the ultramafic body itself. Table 2provides the presented values, and the
H2content along the section is illustrated in Figure 6.
Geosciences 2023, 13, x FOR PEER REVIEW 13 of 19
4.2. Soil Gas Measurement on the Fault Zones
Two cross-sections have been realized to measure the gas near the fault system that
border the Mesozoic ultramafic body of Ginebra. The stations H2_ST4 to 11 are in the
southern B-B’ section and the ST12 to ST16 are in session C-C’. We tested the presence of
H2 within the fault zone, considering that it could be a migration pathway for gases gen-
erated at greater depths. We also aimed to measure the disparity between the hanging
wall and the footwall, specifically between the valley where a sedimentary layer overlays
the ophiolites and the ultramafic body itself. Table 2 provides the presented values, and
the H2 content along the section is illustrated in Figure 6.
As shown in Table 2, it was sometimes very difficult to drill within the rock and some
of the data have been collected very near the surface. In that case, from our experience,
the H2 content is often rather low due to the quantity of air in the shallow soil. However,
as for the zones of vegetation anomalies, H2 presence was detected in all soil gas samples
with concentration varying from 6 to 330 ppm. The methane content remains null, and the
CO2 was also variable reaching a maximum of 13%. H2S levels were very low, always be-
tween 1 and 2 ppm) and the CO content was also low but varied from 0 to 22 ppm.
Figure 6. Schematic cross-sections B-B’ and C-C’ with the measurement points. The H2-ST4 to 6 are
located south of section B-B’ near point 8. The hydrogen (GA) concentration surpasses 1000 ppm at
the H2_ST9 on the eastern border of a laterite outcrop. Due to the lack of subsurface data, the fault
dip evolution versus depth is unknown.
There is no correlation between the CO2 and the H2 concentrations (Figure 7) and in
the absence of methane, one may note a direct correlation between the CO2 and the O2
contents. In only two cases, both in the SCD-1, the sum O2 + CO2 exceed 24% resulting in
a deficit in nitrogen. However, the Bio GA does not analyze all gases in the field, in par-
ticular nitrogen; so, the value “balance” in Table 2, often interpreted as nitrogen, is just
inferred from the sum of all other gas concentrations.
Figure 6.
Schematic cross-sections A-A’ and B-B’ with the measurement points. The H2-ST4 to 6 are
located south of section B-B’ near point 8. The hydrogen (GA) concentration surpasses 1000 ppm at
the H2_ST9 on the eastern border of a laterite outcrop. Due to the lack of subsurface data, the fault
dip evolution versus depth is unknown.
As shown in Table 2, it was sometimes very difficult to drill within the rock and some
of the data have been collected very near the surface. In that case, from our experience, the
H
2
content is often rather low due to the quantity of air in the shallow soil. However, as for
the zones of vegetation anomalies, H
2
presence was detected in all soil gas samples with
concentration varying from 6 to 330 ppm. The methane content remains null, and the CO
2
was also variable reaching a maximum of 13%. H
2
S levels were very low, always between 1
and 2 ppm) and the CO content was also low but varied from 0 to 22 ppm.
There is no correlation between the CO
2
and the H
2
concentrations (Figure 7) and in
the absence of methane, one may note a direct correlation between the CO
2
and the O
2
contents. In only two cases, both in the SCD-1, the sum O
2
+ CO
2
exceed 24% resulting in a
deficit in nitrogen. However, the Bio GA does not analyze all gases in the field, in particular
Geosciences 2023,13, 358 13 of 17
nitrogen; so, the value “balance” in Table 2, often interpreted as nitrogen, is just inferred
from the sum of all other gas concentrations.
Geosciences 2023, 13, x FOR PEER REVIEW 14 of 19
Figure 7. Correlation between various gases based on Table 2 data. The group “Valley” corresponds to
the data within the vegetation anomalies zones and the group ’’Relief’’ represents the data on the two
cross-sections.
5. Discussions and Conclusions
5.1. Sub Circular Depressions and Gas Escape
The relationship between the vegetation anomalies and the H
2
escape is a research
topic still subject to debate. Some researchers consider that the H
2
explorationists give too
much importance to this subject, others believe it forms the foundational basis for initiat-
ing the definition of a play [4]. There is truth in both points of view and the quantitative
values of H
2
in the soil may be affected by many processes [52]. Once again, in Colombia,
by selecting the area for soil gas analysis based on the presence of H
2
potential generating
rocks and vegetation anomalies, we found relatively high H
2
content in the shallow soil.
The second preselected zone, near Ginevra, was the ophiolites thrust sheets; a geological
context where H
2
is expected to be present. We can therefore consider that our under-
standing of hydrogen systems is now reasonably established for the initial basin selection
stage in order to be quick and efficient [4]. Just as oil seeps do not indicate potential drill-
ing locations, H
2
gas surface emanations do not indicate where and if there could be un-
derground commercial accumulation of natural hydrogen. Mapping of the SCD and the
H
2
-GRs form a starting basis for further exploration.
However, the vegetation anomalies in this humid tropical zone of Colombia are
slightly different from previous observations in Brazil [6,44], in Namibia [31] and in Aus-
tralia [51,53]. These last three zones are rather dry with a tropical to Mediterranean cli-
mate, and they are located in the southern hemisphere at 16°, 20°, and 31° South, respec-
tively. During the rainy season, the SCD could be full of water, but the area is very dry for
most of the year, making agriculture almost impossible. Trees do not grow in the SCD,
and perhaps they are dead like in Russian or North American SCD’s [28,29]. The Cauca-
Patia Valley is at the opposite in the wet equatorial zone, at 5°N of latitude, about 1000m
above sea level. The valley is highly cultivated, with sugarcane plantations covering the
entire valley. In this context, gas emanations affect the plants, especially the sugarcane
crops, but without killing them. Additionally, as illustrated in Figure 4, there are instances
where the observed anomaly is simply another plants, particularly bamboo in our case.
We have not carried out a soil survey and this is not our area of expertise, but these obser-
vations deserve an explanation. What makes soil containing a high H
2
content or H
2
-con-
suming microorganisms less suitable for sugarcane but compatible with the growth of
bamboo?
These observations also weaken the “absence of vegetation H
2
related SCD” link
observed elsewhere. It would seem that we should be systematically looking for abnormal
vegetation, and so indexes linked to infra-red datasets (Landsat, Sentinel, and other pro-
grams) rather than only the absence of such vegetation which could be easily mapped
using Google Earth satellite images.
Figure 7.
Correlation between various gases based on Table 2data. The group “Valley” corresponds
to the data within the vegetation anomalies zones and the group “Relief” represents the data on the
two cross-sections.
5. Discussions and Conclusions
5.1. Sub Circular Depressions and Gas Escape
The relationship between the vegetation anomalies and the H
2
escape is a research
topic still subject to debate. Some researchers consider that the H
2
explorationists give too
much importance to this subject, others believe it forms the foundational basis for initiating
the definition of a play [
4
]. There is truth in both points of view and the quantitative values
of H
2
in the soil may be affected by many processes [
52
]. Once again, in Colombia, by
selecting the area for soil gas analysis based on the presence of H
2
potential generating rocks
and vegetation anomalies, we found relatively high H
2
content in the shallow soil. The
second preselected zone, near Ginevra, was the ophiolites thrust sheets; a geological context
where H
2
is expected to be present. We can therefore consider that our understanding of
hydrogen systems is now reasonably established for the initial basin selection stage in order
to be quick and efficient [
4
]. Just as oil seeps do not indicate potential drilling locations,
H
2
gas surface emanations do not indicate where and if there could be underground
commercial accumulation of natural hydrogen. Mapping of the SCD and the H
2
-GRs form
a starting basis for further exploration.
However, the vegetation anomalies in this humid tropical zone of Colombia are
slightly different from previous observations in Brazil [
6
,
44
], in Namibia [
31
] and in
Australia [51,53]
. These last three zones are rather dry with a tropical to Mediterranean
climate, and they are located in the southern hemisphere at 16
◦
, 20
◦
, and 31
◦
South, respec-
tively. During the rainy season, the SCD could be full of water, but the area is very dry for
most of the year, making agriculture almost impossible. Trees do not grow in the SCD, and
perhaps they are dead like in Russian or North American SCD’s [
28
,
29
]. The Cauca-Patia
Valley is at the opposite in the wet equatorial zone, at 5
◦
N of latitude, about 1000 m above
sea level. The valley is highly cultivated, with sugarcane plantations covering the entire
valley. In this context, gas emanations affect the plants, especially the sugarcane crops, but
without killing them. Additionally, as illustrated in Figure 4, there are instances where
the observed anomaly is simply another plants, particularly bamboo in our case. We have
not carried out a soil survey and this is not our area of expertise, but these observations
deserve an explanation. What makes soil containing a high H
2
content or H
2
-consuming
microorganisms less suitable for sugarcane but compatible with the growth of bamboo?
These observations also weaken the “absence of vegetation
⇔
H
2
related SCD” link
observed elsewhere. It would seem that we should be systematically looking for abnor-
mal vegetation, and so indexes linked to infra-red datasets (Landsat, Sentinel, and other
programs) rather than only the absence of such vegetation which could be easily mapped
using Google Earth satellite images.
Geosciences 2023,13, 358 14 of 17
5.2. Potential of the Cauca-Paita Valley
This work demonstrates that H
2
emanations exist in the Cauca-Paita Valley and has
provided solid evidence of the existence of a hydrogen system in the region and more
generally in Colombia.
Through analogy with other countries, like Oman, New Caledonia, and the Philip-
pines [
5
], we consider that ophiolites, geological formations present in and around the basin,
are likely the primary source of hydrogen in this region. An alternative will be the Buga
Batholite alteration, the facies of this batholith is biotite-rich and its alteration may generate
H
2
as suggested in the case of the Rhine Graben [
54
]. However, we cannot disregard the
influence of the mantle wedge just below the oceanic lithosphere involved in the accretion
(see Figure 2). In the Pyrenees, the H
2
emanations are considered to be sourced by a mantle
wedge located only a few kilometers below the surface and which is currently being altered
by serpentinization [
55
]. In that Pyrenean case, the migration pathway is supposed to be
the deeply rooting faults and the targeted reservoirs are compressive structures below a salt
layer which could be the seal, which is a classical pattern for a foreland. However, often the
known H
2
generating rock is not so deep; in Oman and New Caledonia, the ophiolites are
outcropping, and it is unclear if reservoirs may exist within the thrust sheets. The setting of
the Cauca-Paita Valley seems more favorable for H
2
accumulations since in the valley the
sedimentary formations covering the ophiolites may contain reservoir rocks. Structural
traps may exist due to compression, and alongside these, stratigraphic traps can also be
encountered. Through analogy in the Magdalena Valley or in the Llanos Basin, the facies
variation within the tertiary deposits results in stratigraphic traps [56].
Figure 8synthetizes our current view of the H
2
system within the Valley. Meteoric water
infiltration in the relief results in a downward circulation of the water that reacts with the
ophiolites or other iron-rich rocks such as the intrusion at a depth large enough to allow fast
reactions. It could be classical serpentinization or magnetic oxidation at low temperature
as proposed by [
27
]. The generated H
2
migrates through the faults or in a dissolved phase
within the aquifers. It could be trapped in the valley in case of presence of reservoir and
seal. H
2
which is not trapped continues to migrate to the surface and results in the measured
emanations within the fairy circles. The presence of seals is not sufficiently constrained within
the Valley system, but in Mali, intrusions of dolerite sills provide an excellent seal of H
2
reservoirs [57]. Thus, a similar case could be proposed as sketched on Figure 8.
Geosciences 2023, 13, x FOR PEER REVIEW 15 of 19
5.2. Potential of the Cauca-Paita Valley
This work demonstrates that H2 emanations exist in the Cauca-Paita Valley and has
provided solid evidence of the existence of a hydrogen system in the region and more gen-
erally in Colombia.
Through analogy with other countries, like Oman, New Caledonia, and the Philip-
pines [5], we consider that ophiolites, geological formations present in and around the
basin, are likely the primary source of hydrogen in this region. An alternative will be the
Buga Batholite alteration, the facies of this batholith is biotite-rich and its alteration may
generate H2 as suggested in the case of the Rhine Graben [54]. However, we cannot disre-
gard the influence of the mantle wedge just below the oceanic lithosphere involved in the
accretion (see Figure 2). In the Pyrenees, the H2 emanations are considered to be sourced
by a mantle wedge located only a few kilometers below the surface and which is currently
being altered by serpentinization [55]. In that Pyrenean case, the migration pathway is
supposed to be the deeply rooting faults and the targeted reservoirs are compressive
structures below a salt layer which could be the seal, which is a classical paern for a
foreland. However, often the known H2 generating rock is not so deep; in Oman and New
Caledonia, the ophiolites are outcropping, and it is unclear if reservoirs may exist within
the thrust sheets. The setting of the Cauca-Paita Valley seems more favorable for H2 accu-
mulations since in the valley the sedimentary formations covering the ophiolites may con-
tain reservoir rocks. Structural traps may exist due to compression, and alongside these,
stratigraphic traps can also be encountered. Through analogy in the Magdalena Valley or in
the Llanos Basin, the facies variation within the tertiary deposits results in stratigraphic traps
[56].
Figure 8 synthetizes our current view of the H2 system within the Valley. Meteoric
water infiltration in the relief results in a downward circulation of the water that reacts
with the ophiolites or other iron-rich rocks such as the intrusion at a depth large enough
to allow fast reactions. It could be classical serpentinization or magnetic oxidation at low
temperature as proposed by [27]. The generated H2 migrates through the faults or in a
dissolved phase within the aquifers. It could be trapped in the valley in case of presence
of reservoir and seal. H2 which is not trapped continues to migrate to the surface and
results in the measured emanations within the fairy circles. The presence of seals is not
sufficiently constrained within the Valley system, but in Mali, intrusions of dolerite sills
provide an excellent seal of H2 reservoirs [57]. Thus, a similar case could be proposed as
sketched on Figure 8.
Figure 8.
Schematic model of the Cauca-Patia Valley H
2
system. Below the dashed line, everything
is speculative. The generation zones are likely below the two cordilleras and lateral migration,
especially through fault zones, results in H2presence in the valley.
Geosciences 2023,13, 358 15 of 17
5.3. Conclusions
Whilst this research is still preliminary, we provide the first evidence of natural
hydrogen emissions in the Colombian accreted terranes. There are hydrogen emissions on
the faults bordering the Cauca-Paita Valley and microseepages within the valley itself.
This confirmation marks a significant milestone in understanding the geological and
geochemical processes in this region and its potential in holding H
2
resources as a new
primary carbon-free energy source.
Future work should focus on data acquisition to image the subsurface in the valley in
an attempt to identify possible reservoirs, seals, and traps. A better knowledge of the iron
content of the various ultramafic rocks will also provide more quantitative data and if the
results were inconclusive, to our understanding, would not represent a negative factor for
continued exploration. A study of the ground water and its gas dissolved phase should
also provide useful and relevant data as it has been carried out in Kosovo [22]
Globally for Colombia, the presence of natural hydrogen emissions in the Cauca-Patia
basin suggests that there could be new carbon-free energy resources in this region, which
currently lacks conventional fossil fuel reserves.
Author Contributions:
The intuition that the Cauca-valley could have H
2
resource emerged from
discussion between all authors. This paper synthetizes the A.C.R. master work. G.R. provided some
of the data. The remote sensing analysis has been done by A.C.R. The soil gas measurements by
A.C.R., F.G.P. and I.M. I.M. and A.C.R. prepared the first version of the paper. All authors have read
and agreed to the published version of the manuscript.
Funding:
The field acquisition has been funded by Atlas Research. The work done in France by UPPA.
Data Availability Statement: All gas data collected are in the Table 2.
Acknowledgments:
Alejandra’s Master thesis has been conducted at the National University under
the supervision of Marion Weber. The field trips were carried out with the Juan Carlos Molano and
Ariel Cadena. We would like to thank Philippe Dubreuilh, Gabriel Pasquet, and Vincent Roche, as well
as the two anonymous reviewers for a careful review of the preliminary version of this manuscript.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
Rigollet, C.; Prinzhofer, A. Natural Hydrogen: A New Source of Carbon-Free and Renewable Energy That Can Compete With
Hydrocarbons. First Break 2022,40, 78–84. [CrossRef]
2.
Diallo, A.; Cissé, C.S.T.; Lemay, J.; Brière, D.J. La découverte de l’hydrogène naturel par Hydroma, un «Game Changer» pour la
transition énergétique. Ann. Mines-Réalités Ind. 2022,2022, 154–160. [CrossRef]
3.
Brandt, A. Greenhouse gas intensity of natural hydrogen produced from subsurface geologic accumulations. Joule
2023
,7,
1818–1831. [CrossRef]
4.
Lévy, D.; Roche, V.; Pasquet, G.; Combaudon, V.; Geymond, U.; Loiseau, K.; Moretti, I. Natural H
2
exploration: Tools and
workflows to characterize a play. Sci. Technol. Energy Transit. 2023,78, 27. [CrossRef]
5.
Vacquand, C.; Deville, E.; Beaumont, V.; Guyot, F.; Sissmann, O.; Pillot, D.; Arcilla, C.; Prinzhofer, A. Reduced gas seepages
in ophiolitic complexes: Evidences for multiple origins of the H
2
-CH
4
-N
2
gas mixtures. Geochim. Cosmochim. Acta
2018
,223,
437–461. [CrossRef]
6.
Moretti, I.; Prinzhofer, A.; Françolin, J.; Pacheco, C.; Rosanne, M.; Rupin, F.; Mertens, J. Long term monitoring of natural hydrogen
superficial emissions in a Brazilian cratonic environment. Sporadic large pulses versus daily periodic emissions. Int. J. Hydrogen
Energy 2021,46, 3615–3628. [CrossRef]
7.
Geymond, U.; Ramanaidou, E.; Lévy, D.; Ouaya, A.; Moretti, I. Can Weathering of Banded Iron Formations Generate Natural
Hydrogen? Evidence from Australia, Brazil and South Africa. Minerals 2022,12, 163. [CrossRef]
8.
Sherwood Lollar, B.; Onstott, T.C.; Lacrampe-Couloume, G.; Ballentine, C.J. The contribution of the Precambrian continental
lithosphere to global H2 production. Nature 2014,516, 379–382. [CrossRef]
9.
Truche, L.; Joubert, G.; Dargent, M.; Martz, P.; Cathelineau, M.; Rigaudier, T.; Quirt, D. Clay minerals trap hydrogen in the Earth’s
crust: Evidence from the Cigar Lake uranium deposit, Athabasca. Earth Planet. Sci. Lett. 2018,493, 186–197. [CrossRef]
10.
Horsfield, B.; Mahlstedt, N.; Weniger, P.; Misch, D.; Vranjes-Wessely, S.; Han, S.; Wang, C. Molecular hydrogen from organic
sources in the deep Songliao Basin, P.R. China. Int. J. Hydrogen Energy 2022,47, 16750–16774. [CrossRef]
Geosciences 2023,13, 358 16 of 17
11.
Boreham, C.J.; Edwards, D.S.; Feitz, A.J.; Murray, A.P.; Mahlstedt, N.; Horsfield, B.; Boreham, C.J.; Edwards, D.S.; Feitz, A.J.;
Murray, A.P.; et al. Modelling of hydrogen gas generation from overmature organic matter in the Cooper Basin, Australia. APPEA
J. 2023,63, S351–S356. [CrossRef]
12.
Restrepo, J.J.; Toussaint, J.F. The geology of Colombia. Chapter 3: Tectonostratigraphic terranes in colombia: An update first part:
Continental terranes. In The Geology of Colombia, Volume 1 Proterozoic—Paleozoic; Gómez, J., Mateus–Zabala, D., Eds.; Servicio
Geológico Colombiano: Bogotá, Colombia, 2020. [CrossRef]
13.
Gonzalez-penagos, F.; Moretti, I.; Guichet, X. Fluid Flow Modeling in the Llanos Basin, Colombia. In Petroleum Systems Analysis—
Case Studies: AAPG Memoir; AbuAli, M.A., Moretti, I., Bolås, H.M.N., Eds.; The American Association of Petroleum Geologists:
Washington, DC, USA, 2017; Volume 114, pp. 191–217.
14.
Moretti, I.; Charry, G.R.; Morales, M.M.; Mondragon, J.C. Integrated exploration workflow in the South Middle Magdalena Valley
(Colombia). J. S. Am. Earth Sci. 2010,29, 187–197. [CrossRef]
15.
Lopez-Ramos, E. Origin of a Double Forearc Basin: The Example of the Tumaco—Manglares Basin, Northwestern Southamerica.
CT&F—Cienc. Tecnol. Futuro 2020,10, 67–92. [CrossRef]
16.
Bayona, G.; Cortés, M.; Jaramillo, C.; Ojeda, G.; Aristizabal, J.J.; Reyes-Harker, A. An integrated analysis of an orogen– sedimentary
basin pair: Latest Cretaceous—Cenozoic evolution of the linked Eastern Cordillera orogen and the Llanos Foreland Basin of
Colombia. Geol. Soc. Am. Bull. 2008,120, 1171–1197. [CrossRef]
17.
Marcaillou, C.; Muñoz, M.; Vidal, O.; Parra, T.; Harfouche, M. Mineralogical evidence for H
2
degassing during serpentinization at
300 ◦C/300 bar. Earth Planet. Sci. Lett. 2011,303, 281–290. [CrossRef]
18.
McCollom, T.M.; Klein, F.; Moskowitz, B.; Berquó, T.S.; Bach, W.; Templeton, A.S. Hydrogen generation and iron partitioning
during experimental serpentinization of an olivinepyroxene mixture. Geochim. Cosmochim. Acta 2020,282, 55–75. [CrossRef]
19.
Charlou, J.L.; Donval, J.P.; Fouquet, Y.; Jean-baptiste, P.; Holm, N. Geochemistry of high H
2
and CH
4
vent fluids issuing from
ultramafic rocks at the Rainbow hydrothermal field (36◦140N, MAR). Chem. Geol. 2002,191, 345–359. [CrossRef]
20.
Klein, F.; Bach, W.; McCollom, T.M. Compositional controls on hydrogen generation during serpentinization of ultramafic rocks.
Lithos 2013,178, 55–69. [CrossRef]
21.
Cannat, M.; Rommevaux-Jestin, C.; Fujimoto, H. Melt supply variations to a magma-poor ultra-slow spreading ridge (Southwest
Indian Ridge 61◦to 69◦E): Melt supply variations. Geochem. Geophys. Geosyst. 2003,4. [CrossRef]
22.
Lévy, D.; Callot, J.P.; Moretti, I.; Duttine, M.; Dubreuil, B.; de Parseval, P.; Boudouma, O. Successive phases of serpentinization
recorded in the ophiolite of Sivas (Turkey), from oceanic crust accretion to post-obduction alteration. BSGF
2022
,193, 12.
[CrossRef]
23.
Zgonnik, V.; Beaumont, V.; Larin, N.; Pillot, D.; Deville, E. Diffused flow of molecular hydrogen through the Western Hajar
mountains, Northern Oman. Arab. J. Geosci. 2019,12, 71. [CrossRef]
24.
Deville, E.; Prinzhofer, A. The origin of N
2
-H
2
-CH
4
-rich natural gas seepages in ophiolitic context: A major and noble gases study
of fluid seepages in New Caledonia. Chem. Geol. 2016,440, 139–147. [CrossRef]
25.
Combaudon, V.; Moretti, I.; Kleine, B.I.; Stefánsson, A. Natural hydrogen emissions in Iceland and comparison with the
Mid-Atlantic Ridge. Int. J. Hydrogen Energy 2022,47, 10217–10227. [CrossRef]
26.
Pasquet, G.; Hassan, R.H.; Sissmann, O.; Varet, J.; Moretti, I. An Attempt to Study Natural H
2
Resources across an Oceanic Ridge
Penetrating a Continent: The Asal–Ghoubbet Rift (Republic of Djibouti). Geosciences 2021,12, 16. [CrossRef]
27.
Geymond, U.; Briolet, T.; Combaudon, V.; Sissmann, O.; Martinez, I.; Duttine, M.; Moretti, I. Reassessing the role of magnetite
during natural hydrogen generation. Front. Earth Sci. 2023,11, 1169356. [CrossRef]
28. Larin, N.; Zgonnik, V.; Rodina, S.; Deville, E.; Prinzhofer, A.; Larin, V.N. Natural Molecular Hydrogen Seepage Associated with
Surficial, Rounded Depressions on the European Craton in Russia. Nat. Resour. Res. 2015,24, 369–383. [CrossRef]
29.
Zgonnik, V.; Beaumont, V.; Deville, E.; Larin, N.; Pillot, D.; Farrell, K.M. Evidence for natural molecular hydrogen seepage
associated with Carolina bays (surficial, ovoid depressions on the Atlantic Coastal Plain, Province of the USA). Prog. Earth Planet.
Sci. 2015,2, 31. [CrossRef]
30.
Moretti, I.; Brouilly, E.; Loiseau, K.; Prinzhofer, A.; Deville, E. Hydrogen Emanations in Intracratonic Areas: New Guide Lines for
Early Exploration Basin Screening. Geosciences 2021,11, 145. [CrossRef]
31.
Moretti, I.; Geymond, U.; Pasquet, G.; Aimar, L.; Rabaute, A. Natural hydrogen emanations in Namibia: Field acquisition and
vegetation indexes from multispectral satellite image analysis. Int. J. Hydrogen Energy 2022,47, 35588–35607. [CrossRef]
32.
Toussaint, J.F.; Restrepo, J.J. The geology of Colombia. Chapter 7: Tectonostratigraphic terranes in colombia: An update second
part: Oceanic terranes. In The Geology of Colombia, Volume 2 Mesozoic; Gómez, J., Pinilla–Pachon, A.O., Eds.; Servicio Geológico
Colombiano: Bogotá, Colombia, 2020. [CrossRef]
33.
Weber, M.; Cardona, A.; Paniagua, F.; Cordani, U.; Sepúlveda, L.; Wilson, R. The Cabo de la Vela Mafic-Ultramafic Complex,
Northeastern Colombian Caribbean region: A record of multistage evolution of a Late Cretaceous intra-oceanic arc. Geol. Soc.
London, Spéc. Publ. 2009,328, 549–568. [CrossRef]
34.
Nivia, A. Geología y Geocronología del Batolito de buga y el Macizo Ofiolítico de Ginebra, Colombia. 2017. Conference: XVI
Congreso Colombiano de Geología at: Santa Marta—Magdalena (Colombia). Available online: https://www.researchgate.net/
publication/319631780 (accessed on 20 September 2023).
35.
Moreno, M.; Pardo, A. Stratigraphical and sedimentological constraints on western colombia: Implications on the evolution of
the Caribbean plate. Am. Assoc. Pet. Geol. Mem. 2003,79, 891–924.
Geosciences 2023,13, 358 17 of 17
36.
McCourt, W.J.; Aspden, J.A.; Brook, M. New geological and geochronological data from the colombian andes: Continental growth
by multiple accretion. J. Geol. Soc. Lond. 1984,141, 831–845. [CrossRef]
37.
Nivia, A. Mapa geológico departamento del Valle del Cauca escala 1:250000, memoria explicativa 2001. 2001. Available
online: https://www.researchgate.net/publication/275971711_Mapa_Geologico_Departamento_del_Valle_del_Cauca_Escala_
1250000_Memoria_Explicativa_2001 (accessed on 20 September 2023).
38.
Ossa, C.; Concha, A.H. Petrogénesis de las rocas del macizo ofiolítico de Ginebra entre las veredas la honda (Ginebra) y el diamante
(Buga) en el departamento del Valle del Cauca. 2007. Available online: https://dialnet.unirioja.es/servlet/revista?codigo=19929
(accessed on 20 September 2023).
39.
Tapias, J.G.; Ramírez, N.E.M.; Gutiérrez, F.A.A.; Hernández, J.A.C. Catálogo de dataciones radiométricas de colombia en arcgis
y google earth. 2015. Servicio Geológico Colombiano, Publicaciones Geológicas Especiales. Available online: https://www2
.sgc.gov.co/MGC/Documents/Compilando/04_Gomez-et-al-2015_Catalogo-de-dataciones-radiometricas.pdf (accessed on 20
September 2023).
40.
Palomino, M. Caracterización petrográfica del Batolito de Buga en la Cordillera Central, Valle del Cauca. 2017. Available online:
https://repositorio.uniandes.edu.co/server/api/core/bitstreams/3d70fc54-a922-4b8a-baa4-14d11eae79bf/content (accessed on
20 September 2023).
41. McCourt, W.J.; Verdugo, G. Mapa Geológico de Colombia, Plancha 300, Cali; BogotáIngeominas: Ingeominas, Bogotá, 1985.
42.
Barbosa, G. República de Colombia, Ministerio de Minas y Energía. Instituto de investigación e información geocientífica,
minero-ambiental y nuclear ingeominas. Memorias mapa geológico del departamento del Cauca. 2003. Available online:
https://recordcenter.sgc.gov.co/B4/13010040019624/documento/pdf/0101196241101000.pdf (accessed on 20 September 2023).
43.
ANH. Available online: https://www.anh.gov.co/en/hidrocarburos/informacion-geologica-y-geofisica/estudios-integrados-
y-modelamientos/ (accessed on 20 September 2023).
44.
Prinzhofer, A.; Moretti, I.; Françolin, J.; Pacheco, C.; D’Agostino, A.; Werly, J.; Rupin, F. Natural hydrogen continuous emission
from sedimentary basins: The example of a Brazilian H
2
-emitting structure. Int. J. Hydrogen Energy
2019
,44, 5676–5685. [CrossRef]
45.
Hassan, M.A.; Yang, M.; Rasheed, A.; Yang, G.; Reynolds, M.; Xia, X.; Xiao, Y.; He, Z. A rapid monitoring of NDVI across the
wheat growth cycle for grain yield prediction using a multi-spectral UAV platform. Plant Sci. 2019,282, 95–103. [CrossRef]
46.
Kaufman, Y.J.; Tanre, D. Atmospherically resistant vegetation index (ARVI) for EOS-MODIS. IEEE Trans. Geosci. Remote Sens.
1992,30, 261–270. [CrossRef]
47.
Posada-Asprilla, W.; Medina-Sierra, M.; Cerón-Muñoz, M. Estimación de la calidad y cantidad de pasto kikuyo (Cenchrus
clandestinum (Hochst. ex Chiov.) Morrone) usando imágenes multiespectrales. Rev. UDCA Actual. Divulg. Científica
2019
,22.
[CrossRef]
48.
Ávila Vélez, E.F. Propuesta metodológica para cuantificar áreas afectadas por incendios forestales utilizando imágenes satelitales
sentinel-2. caso de studio páramo del almorzadero, Colombia. 2019. Available online: https://www.academia.edu/44443374
/Propuesta_metodol%C3%B3gica_para_cuantificar_%C3%A1reas_afectadas_por_incendios_forestales_utilizando_im%C3%A1
genes_satelitales_Sentinel_2_Caso_de_estudio_p%C3%A1ramo_del_Almorzadero_Colombia (accessed on 20 September 2023).
49.
Becker, T.; Nelsen, T.S.; Leinfelder-Miles, M.; Lundy, M.E. Differentiating between nitrogen and water deficiency in irrigated
maize using a uav-based multi-spectral camera. Agronomy 2020,10, 1671. [CrossRef]
50.
Boegh, E.; Soegaard, H.; Broge, N.; Hasager, C.; Jensen, N.; Schelde, K.; Thomsen, A. Airborne multispectral data for quantifying leaf
area index, nitrogen concentration, and photosynthetic efficiency in agriculture. Remote Sens. Environ. 2002,81, 179–193. [CrossRef]
51.
Frery, E.; Langhi, L.; Maison, M.; Moretti, I. Natural hydrogen seeps identified in the North Perth Basin, Western Australia. Int. J.
Hydrogen Energy 2021,46, 31158–31173. [CrossRef]
52.
McMahon, C.J.; Roberts, J.J.; Johnson, G.; Edlmann, K.; Flude, S.; Shipton, Z.K. Natural hydrogen seeps as analogues to inform
monitoring of engineered geological hydrogen storage. In Enabling Secure Subsurface Storage in Future Energy Systems; Miocic, J.M.,
Heinemann, N., Alcalde, J., Edlmann, K., Schultz, R.A., Eds.; Special Publications, 528; Geological Society: London, UK, 2022. [CrossRef]
53.
Frery, E.; Langhi, L.; Markov, J. Natural hydrogen exploration in Australia—State of knowledge and presentation of a case study.
APPEA J. 2022,62, 223–234. [CrossRef]
54.
Murray, J.; Clément, A.; Fritz, B.; Schmittbuhl, J.; Bordmann, V.; Fleury, J.M. Abiotic hydrogen generation from biotite-rich granite:
A case study of the Soultz-sous-Forêts geothermal site. France. Appl. Geochem. 2020,119, 104631. [CrossRef]
55.
Lefeuvre, N.; Truche, L.; Donzé, F.V.; Ducoux, M.; Barré, G.; Fakoury, R.A.; Calassou, S.; Gaucher, E.C. Native H
2
exploration in
the western Pyrenean foothills. G-Cubed 2021,22, e2021GC009917. [CrossRef]
56.
Caballero, V.M.; Rodríguez, G.; Naranjo, J.F.; Mora, A.; De La Parra, F. From facies analysis, stratigraphic surfaces, and depositional
sequences to stratigraphic traps in the Eocene—Oligocene record of the southern Llanos and northern Magdalena Basin. In The
Geology of Colombia, Volume 3 Paleogene—Neogene; Gómez, J., Mateus–Zabala, D., Eds.; Publicaciones Geológicas Especiales 37;
Servicio Geológico Colombiano: Bogotá, Colombia, 2020; 48p. [CrossRef]
57.
Maiga, O.; Deville, E.; Laval, J.; Prinzhofer, A.; Diallo, A.B. Trapping processes of large volumes of natural hydrogen in the
subsurface: The emblematic case of the Bourakebougou H2 field in Mali. Int. J. Hydrogen Energy 2023. [CrossRef]
Disclaimer/Publisher’s Note:
The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.
