Journal of Petroleum Geology, Vol.

27 (4), October 2004, pp 321- 333

321

THE GIBRALTAR DISCOVERY, NORTHERN LLANOS

FOOTHILLS, COLOMBIA:
CASE HISTORY OF AN EXPLORATION SUCCESS
IN A FRONTIER AREA
T.Villamil 1*, J. Muoz, J. Snchez 2, J. J. Aristizabal 2, J.Velasco 2,
P. E. Luna 2, A. Mantilla 2, A. Fajardo 3, L. E. Pea 4, M. G. Paz 2,
O. Silva 2, E. Snchez 2 and N. Meza 5
Ecopetrols recent Gibraltar discovery was made in a previously little known area of the northern
Llanos foothills of the Eastern Cordillera of Colombia. The structure at Gibraltar is complex and
comprises a thrust stack with which are associated hangingwall anticlines providing four-way
closures (or three-way against a sealing fault).The Gibraltar-1 well was originally drilled by Occidental
in 2002. According to Ecopetrols subsequent structural reinterpretation, this well penetrated the
back limb of a fault propagation fold. This structural model was confirmed with a sidetrack well
which penetrated the Palaeocene Barco Formation and demonstrated the presence of light
hydrocarbons.
A second objective of well re-entry at Gibraltar was to investigate the Eocene Mirador Formation,
a major reservoir unit elsewhere in the Llanos foothills, in case of missed or bypassed pay.Tests of
the Mirador Formation verified the presence of high quality 57o API degree condensate that
flowed at a rate of 690 barrels per day together with 44 million cubic feet of rich gas per day.
The success at Gibraltar will hopefully encourage further exploration developments in this
remote and structurally-complex region.
INTRODUCTION
Few significant structural targets remain to be drilled
in the Worlds major oil-producing provinces.
Remaining targets tend to be located in regions of
extreme extension or compression, or in regions in
which exploration activity is difficult due to economic
Ecopetrol. Calle 37 No. 8-43, Bogot, Colombia.
Current address: Lukoil-Overseas, Carrera 4 No. 72-35,
suite 600, Bogot, Colombia.
1

* Author for correspondence: [email protected]

Ecopetrol. Calle 37 No. 8-43, Piso 8, Edificio Colgs,
Bogot, Colombia.
3
Ecopetrol-ICP, Piedecuesta, Colombia. Current address:
Hocol S.A. Colombia.
4
Ecopetrol. Calle 37 No. 8-43, Bogot, Colombia, currently
independent.
5
Ecopetrol. Calle 37 No. 8-43, Bogot. Current address:
ChevronTexaco Colombia.
2

or other factors. Extensional areas such as the West

African - South American margin and the Gulf of
Mexico have attracted significant attention in recent
years. Compressional regions in onshore foldbelts are
generally considered to be less appealing for a variety
of reasons including their structural complexity, the
absence of high-quality seismic data, and possible
difficulties associated with drilling and logistical
operations. However, these areas may have significant
exploration potential. An example is the northern
Llanos foothills of the Eastern Cordillera of Colombia
(Fig. 1), an area which is currently undergoing
something of a revival in exploration (Villamil, 2004).
The emergence of this area was confirmed with
the discovery of the giant Cusiana field (1.6 B brl
o.e.) in 1991, followed by Cupiagua (900 MM brl
o.e.) in 1993 and Florea, Pauto, Volcanera and Dele
between 1994 and 1997 (Cooper et al., 1995a, b;
Martinez, 2003 and references therein). Exploration
then slowed due to economic and contractual factors
but was revived from 1999/2000 onwards with the

322

The Gibraltar discovery, Llanos foothills, Colombia

reactivation of various major projects, some of which

had been dormant for almost 10 years. Among these
was the exploration of the Samor (or Sirir) area in
which the Gibraltar field, the subject of this paper, is
located.
The purpose of this paper is to review the discovery
of the Gibraltar field as a case history of an
exploration success in a challenging and structurally
complex onshore foldbelt. More details including
structural and geochemical data will be presented in
a follow-up paper (Aristizabal et al., in prep.) which
is being completed.
Structurally, the Llanos foothills area can be
divided into three regions:
(i) A relatively simple frontal zone with adequate
to good seismic coverage and good seismic images
(see examples in Martinez, 2003; Estrada and
Jaramillo, 2003). This structural setting occurs to the
east of a long, narrow syncline known as the El
Descanso syncline in the Gibraltar area and by other
names (e.g. the Nunchia syncline) elsewhere (Fig. 2).
(ii) Zones of intermediate structural complexity are
located towards the hinterland (Fig. 2), for example
beneath the axis of the Nunchia syncline. An example
of a field in this setting is Cupiagua. Seismic imaging
is generally poor in these areas, and reliable maps can
only be constructed with 3D acquisition and refined
processing techniques (Estrada and Jaramillo, 2003).
(iii) Areas of high structural complexity include
hinterland triangle zones characterized by the
occurrence of antiformal thrust stacks (Fig. 2). This
is the structure at the Florea and Pauto fields as well
as at Gibraltar.
In spite of these variations in structural style, fields
in the Llanos foothills show a number of features in
common. Reservoir rocks are of low porosity, low to
high permeability and high deliverability (Cazier et
al., 1995, 1997; Pulham, 1994; Pulham et al., 1997 a,
b; Rathke and Coral, 1997; OLeary et al, 1997). A
range of hydrocarbon fluid types are present, from
poor liquid yields of gas condensate (e.g. Pauto,
Volcanera fields) to very rich yields of gas condensate
in a single unstratified phase (Cupiagua), to volatile
oil with a gas cap (Cusiana) (Rathke and Coral, 1997;
Martinez, 2003).
Background to the Gibraltar project
The Gibraltar-1 well was originally drilled by
Occidental in 2002 and confirmed the reservoir
potential of the Eocene Mirador Formation, providing
strong indications of the presence of light
hydrocarbons (volatile oil or gas). Occidental then
relinquished the well due to commercial
considerations, but Ecopetrol decided to re-enter it
and to drill sidetracks based on a scenario of untested

or bypassed pay. The presence of light hydrocarbons

was confirmed in the Barco Formation in sidetrack
(ST) 3 and the Mirador Formation produced 690
barrels per day and 44 million cubic feet of gas per
day.
Several examples of bypassed pay have been
documented in the Llanos foothills. Perhaps the best
known are the Cusiana and Cupiagua fields, both
drilled to reservoir level in the 1970s by Ecopetrol
but not discovered until the 1990s by BP. Similarly,
the Tame structure was drilled by Shell in 1965, then
by Exxon in 1987, and finally discovered by Repsol
and Total in 2002. The Gibraltar discovery can be
added to this list.
The decision to re-enter the well was based on a
number of criteria including:
(i) Electric logs with deep resistivity readings of
more than 1,000 ohm-m, suggesting the presence of
light hydrocarbons in the Mirador Formation (Figs. 3
and 4). In addition, oil and gas shows were reported
while drilling, and also during later analyses of
sidewall cores and cuttings from both the Mirador
Formation and the Paleocene-Eocene Cuervos
Formation. Detailed analyses indicated compositional
variations with depth, and also GOR variations similar
to those recorded in the Cusiana and Cupiagua fields.
(ii) Modular Dynamic Formation Tester (MDT)
data which showed a gas gradient in the upper portion
of the Mirador Formation.
(iii) A gas leak out of casing which reached the
surface and had to be flared during the entire drilling
operation.
(iv) A re-analysis of the local structure (Figs. 5
and 6) and of the fields geological and stratigraphic
characteristics, and also of the lithology of the cuttings
recovered.
GEOLOGICAL BACKGROUND
Stratigraphy and regional geology
Based on regional considerations, the principal source
rock for the oil encountered in Gibraltar-1 ST-5 is
inferred to be the early Late Cretaceous La Luna
Formation, although this was not penetrated by the
drill there. The La Luna Formation and equivalent
units were deposited during Albian to Santonian times
in a shallow epicontinental sea which covered most
of Ecuador, Colombia, Venezuela, Trinidad, parts of
Peru, Guyana and Suriname (Macellari and DeVries
1987; Macellari, 1988; Villamil and Arango, 1998;
Villamil et al., 1999). The La Luna is generally
considered a high-grade source unit (e.g. Arthur et
al., 1987) with TOC values of 1-6% and hydrogen
index values in excess of 500 mg/g (Villamil, 2003).
The formation is up to 600m thick in the Lake
Maracaibo and Middle Magdalena Valley Basins. Two

T. Villamil et al.

323

Fig. 1. Topographic map of Colombia and western Venezuela showing the location of the Llanos foothills
structural trend and the Gibraltar structure. The Sirir region, where the Gibraltar well is located, is in an area
where the Eastern Cordillera turns sharply to the NW, probably reflecting the dog leg configuration of the
Jurassic rift system which was inverted during Neogene uplift of the Eastern Cordillera.
CF: Central Foothills, NF: Northern Foothills.

324

The Gibraltar discovery, Llanos foothills, Colombia

Fig. 2a. Structural setting of the Llanos foothills exploration trend showing the three structural zones of
differing complexity (modified from Martinez, 2003; Estrada and Jaramillo, 2003). The eastern zone is
composed of frontal structures which are relatively simple and easy to image with conventional seismic.
The central zone is more complex with highly folded and faulted structures; conventional seismic imaging is
poor and imaging with 3D seismic and refined processing techniques is necessary to confidently map the
traps. The western zone is composed of stacked sheets or horses in an internal triangle zone. Seismic imaging
is poor and wells target structural models rather than traps mapped with seismic; dips are steep and drilling
is expensive.
Fig. 2b. Complex foothills structures (e.g. Florea, Gibraltar) to the west of the axis of the Nunchia El
Descanso syncline.
Fig. 2c. Structure at the Cupiagua field beneath the axis of the Nunchia syncline.
Fig. 2d. Structure at the Cusiana field which is to the east of the Nunchia syncline but still in a sub-thrust
setting.
Figs 2b-d are from Martinez (2003).

intervals are particularly rich in organic matter and

correspond to oceanic anoxic events in the late Albian
and at the Cenomanian-Turonian boundary,
respectively. The latter interval is of more regional
extent (Villamil and Arango, 1998), and probably
sources most of the hydrocarbons at Gibraltar and
associated structures, although this requires verification
in the field.
Regionally overlying the La Luna Formation is the
Campanian-Maastrichtian Guadalupe Formation (Fig.
3) which was not penetrated by Gibraltar-1 but which
is inferred to be present there. The formation is
composed of prograding shallow-marine deposits
(Cooper et al., 1995 a,b; Pulham et al., 1997 a,b) and
may have some reservoir potential. It represents the
final stages of infilling of the Cretaceous epicontinental
basin, and the sediments of which is it composed were
derived from both east and west. Sediments derived
from the east are quartz-rich but include phosphatic
intervals which correspond to episodes of oceanic
upwelling. Sediments derived from the west have a
greater content of lithic fragments reflecting their

derivation from uplands composed of metamorphic,

igneous and sedimentary protoliths. The presence of
the Guadalupe Formation in the northern foothills is
also predicted from regional palaeogeographic maps
of Colombia (Villamil, 1999) and Venezuela (Pindell
et al., 1998). These suggest that the Guadalupe
Formation in the Gibraltar area is thicker and finergrained than in the central Llanos foothills.
The overlying Palaeocene Barco Formation (Fig.
3) was penetrated and tested by Ecopetrol with
Gibraltar-1 ST-5. This was the first time that the
formation had been tested in the northern Llanos
foothills. The Barco Formation is composed of
sediments derived from the Guyana Shield located
to the east of the Llanos foothills. It is an important
reservoir unit in the central and southern foothills,
and is one of the main producers at the Cusiana and
Cupiagua fields. In the northern foothills, the
presence, distribution and quality of the Barco
Formation were not known prior to the drilling of
Gibraltar-1. The formation is somewhat thicker here
than in the southern and central Llanos foothills. Its

T. Villamil et al.

325

Fig. 3. Generalised stratigraphic section and petroleum systems chart of the central and northern Llanos
foothills. The Gachet Formation, coeval with the La Luna Formation but of different facies, is the main
regional petroleum source rock. The principal reservoirs in the Gibraltar area are the Mirador and Barco
Formations; the Guadalupe Formation may be the target of future wells.

reservoir quality was found to be excellent. A drill

stem test (DST-1) was conducted in open-hole
conditions and the Barco produced more than 5,000
b/d of low salinity water and light hydrocarbons.
The Late Palaeocene-earliest Eocene Cuervos
Formation (Fig. 3) is widely distributed in the Llanos
foothills, Llanos Basin and Eastern Cordillera, but had
not been penetrated in the northern foothills before
the drilling of Gibraltar ST2. The formation provides
a local seal to the underlying Barco Formation and
forms an intra-reservoir barrier to lateral and vertical
migration. The Cuervos Formation posed significant
problems during drilling of the Gibraltar well. The
lower Cuervos is composed of thin coal seams
interbedded with fine-grained siliciclastics which at
Gibraltar dip approximately 65o to the west. This
lithology and dip gave rise to unstable drilling
conditions, which ultimately caused the first two
sidetracks to fail.

Overlying the Cuervos Formation is the Eocene

Mirador Formation (Fig. 3) which is the principal
reservoir and exploration target in the Llanos foothills
and Llanos Basin (Cazier et al., 1995, 1997). It is
coeval with the Misoa and Mirador Formations of the
prolific Maracaibo Basin in Venezuela and NE
Colombia (Villamil, 1999; Pindell et al., 1998 and
references therein). In the Llanos foothills, the Mirador
Formation is mainly composed of well-sorted, finegrained fluvial quartzarenites derived from the Guyana
Shield. The formation is intensely micro-fractured and
this contributes to its high permeability. Net pay in
Mirador sandstones can reach several hundreds of feet.
Overlying the Mirador Formation is the Oligocene
Carbonera Formation (Fig. 3) which is dominated by
shales and represents a marked increase in
accommodation space and associated regional
flooding. The formation forms an excellent seal
capable of retaining significant columns of

The Gibraltar discovery, Llanos foothills, Colombia

Fig. 4. Example of Mirador Formation well logs from Gibraltar-1. The reservoir is a naturally-fractured finegrained and very clean sandstone derived after several sedimentary cycles from the Guyana Shield to the east
of the foldbelt. Note saturations in deep resistivity, neutron-density crossover showing hydrocarbons and the
PEF (Photoelectric Factor) log with the FMI showing the fractured nature of the reservoir.

326

hydrocarbons; at the Cupiagua field, for example, the

Carbonera holds more that 6,000 ft of gas condensate
(Martinez, 2003).
Overlying the Carbonera Formation are molasse
deposits derived from the west as a consequence of
the latest stages of uplift of the Eastern Cordillera.
Structural geology
The highly-deformed Sirir area is located close to
the easternmost margin of the Eastern Cordillera, an
inverted Cretaceous graben system. Cross-sections

through this fold-and-thrust belt have provided a range

of shortening values (e.g. Colletta et al., 1990; Dengo
and Covey, 1993; Cooper et al., 1995 a, b; Roeder
and Chamberlain, 1995; Corredor, 2003) indicating
the complexity of the regional structure (Fig. 5). Fig.
6 shows the presence at Gibraltar of a stack of three
or more horses interpreted from seismic information
and structural modelling. The thrust stack involves
the Mirador, Barco and possibly the Guadalupe
Formations. Beneath the stack is a possibly larger and
younger thrust sheet which was the original target of

T. Villamil et al.

327

Fig. 5. Simplified geological map of the Sirir block and Gibraltar structure. Note opposing vergence of faults
defining a tectonic window into the uppermost portion of the stacked horses within the triangle zone to the
west of the El Descanso syncline. Fig. 5a is a photograph of the Gibraltar area, looking north. Mountains to the
right are composed of Neogene rocks dipping to the east, and represent the stratigraphy above the passive
roof fault that bounds the triangle zone. Strata to the left are Paleogene rocks generally dipping to the west;
these strata lie above the uppermost thrust, verging to the east, of the triangle zone. The valley in the centre
of the photo represents a tectonic window into the upper triangle zone. The photo corresponds
approximately to the surface features of the section presented in Fig. 6 whose profile is marked on the map.

328

The Gibraltar discovery, Llanos foothills, Colombia

Fig. 6. Structural cross-section with uninterpreted and interpreted seismic for the Gibraltar structure. Note
the poor quality of the seismic data; drilling is therefore performed using detailed structural models rather
than on well-imaged seismically-derived traps. The Gibraltar-1 well penetrated the uppermost position of the
crest of the structure (see Fig. 7 for details). Vertical scale is approximate and in feet; the crest of the Mirador
Formation in the tested Gibraltar structure is constrained by two wells; the crests of Gibraltar deep 1 and 2
are based on prognoses and are approximate.

T. Villamil et al.

329

Fig. 7. Dipmeter analysis of the Gibraltar-1 well. The well penetrated the top of the Mirador Formation in the
eastern or steep limb of the anticline. A few feet into the Mirador, dips change to horizontal and define the
axis of the fold. The lower portion of the Mirador and the entire Cuervos Formation show dips towards the
west, defining the western or gentle dip of the fold.
Some of the main uncertainties in reserves estimation and phase of hydrocarbons in the Gibraltar structure
are the consequence of the fact that the well drilled the crest rather than a flank. The oil-down-to is
therefore relatively limited and the proven phase is the lighter of the structure.

the Gibraltar well; this structure, now termed

Gibraltar deep, is at a depth of about 15,000 ft.
Above the stack is thrust a shallower nappe (4,500ft
deep). The fold associated with this overlying fault
can be observed in the surface geology. Closures are
either four-way or three-way against a fault (Fig. 6).
The thrust sheets were transported to the east
before being stacked. The thrust stack is bounded to
the east by a passive roof fault which separates it from
the El Descanso Syncline (Figs. 5 and 6). This passive
fault is an intra-Carbonera unconformity which serves
as a dcollement surface (Corredor, 2003).
Before Ecopetrol initiated operations in the
Gibraltar area, it was not known whether the well
drilled by Occidental had penetrated an abnormally
thick section of the Cuervos Formation, a steep

forelimb of a fault-propagation fold, or the back limb

of a fault-propagation fold; i.e. it was not clear whether
the eastern or the western flank of an anticline had
been drilled. Several companies evaluated the block
and concluded that Gibraltar-1 had penetrated the
eastern, more steeply-dipping limb. Based on an
alternative interpretation of dipmeter data and a
different structural model, Ecopetrol concluded that
the well had penetrated the western, more gentle limb
(Fig. 6). A sidetrack was drilled towards the east and
reached the target Barco Formation at a depth of
11,500 ft after two failed attempts, verifying the
structural model.
Multiple thrust faults and associated anticlines host
commercial accumulations to the south of Gibraltar
(e.g. at Florea and Pauto fields), and also of course

330

The Gibraltar discovery, Llanos foothills, Colombia

Fig. 8. The figure presents the results of Gibraltar initial tests. Test results were constrained by gas separation
and storage capacity; numbers presented in the figure are minimum cases rather than expected flow rates.
The maximum condensate flow was measured at 690 barrels per day and the average gas flow was of 44
million cubic feet of gas per day. Reservoir pressure was maintained and recovered immediately after periods
of closure and build-up demonstrating that the well perceives no reservoir boundaries.

in analogous fold-and-thrust belts such as the

Canadian and northern US Rockies. Interpretation of
the Gibraltar structure was part of a regional study of
an area covering about 500km which used both the
available, relatively poor-quality seismic data (230km)
together with field geology and Landsat-Radarsat
images. The structural interpretation was aided by the
Fullbore Formation Micro Imager tool (FMI) and an
analysis of Resistivity At Bit (RAB), which provided
a high resolution image of the borehole. These logs
together with the available data allowed major
lithologic discontinuities such as formation boundaries
and faults to be identified, and a relatively consistent
interpretation of dip domains to be made (Fig. 7). Dips
were found to steepen from the top of the Mirador
Formation to the TD of the well, and it was concluded
that the well navigated sub-parallel to stratigraphy
through the Cuervos Formation (Fig. 7).
The thrust system in the Gibraltar area is expressed
at the surface as two different structural trends. The
first of these is characterized by west-dipping thrust
faults with which are associated asymmetric hangingwall anticlines of relatively low amplitude (about
2km); the flanks of these anticlines show higher dips
and generally plunge to the south. The second trend,
above the passive roof fault, is characterized by dips
to the east of up to 70o in a high-amplitude monocline

or syncline, and involves much younger strata.

Deformation and shortening are not constant along
strike and tear faults compensate for differences in
displacement (Fig. 5). At Gibraltar, multiple thrust
sheets cause the Mirador Formation to be repeated at
depth; the top of the formation at the crest of the
structure is at 6,500ft measured depth (Fig. 6).
OPERATIONAL PROCEDURE
AND RESULTS
Based on a re-interpretation of the structural
configuration at Gibraltar (Fig. 6) and the prediction
of a new reservoir interval in the Barco Formation,
together with the hydrocarbon shows in wells
Gibraltar-1, Gibraltar-1ST1 and Gibraltar-1ST2, a
new sidetrack was drilled in 2002-2003. The plan was
to re-enter the Gibraltar well and to locate a whipstock
tool at 10,462ft, drill though a 9 7/8 casing and
sidetrack towards the centre of the anticline in order
to reach older rocks.
The Barco Formation was penetrated at the third
attempt at 12,050 ft and showed the same reservoir
quality as it does at Florea field, approximately
170km to the SW of Gibraltar. The well found
commercial volumes of hydrocarbons in the Mirador
Formation which was tested and produced 690 b/d of

T. Villamil et al.

57o API oil and 44 million cubic ft/d of rich gas (Fig.
8). Its flow was restricted due to separation capacity
and oil storage limitations. The tests did not show the
boundaries of the reservoir.
The sidetrack was drilled in a direction opposite
to the natural tendency of the Gibraltar-1 well
(azimuth: 70o; dip 25o); the whipstock exit trajectory
was to 231o with a dip of 70o. Drilling with oil-based
mud was necessary to prevent high torque and
diminish possible sticking problems. Major problems
were caused while drilling due to the unexpected
presence of over 20 thin coal seams separated by
mudstones and coaly shales, dipping 65o, which
caused the first two sidetracks to fail. For the third
attempt, the mud weight was then increased to the
limit allowed by the fracture gradient, and the drilling
velocity was reduced to 2ft per hour to allow the
release of gas. Tar was also used during drilling to
avoid wall collapse and to seal possible opened
fractures. Reaming and back reaming was routinely
applied as the coals were being drilled. The Barco
Formation was then drilled with triconic drill bits and
a mud weight of 8.8 ppg. It was found to have excellent
reservoir quality and to contain hydrocarbon shows.
P50 reserves estimates at Gibraltar are considered to
be above 200 MM brls o.e.
Although Gibraltar-1 and the sidetracks confirmed
the presence of oil in the structure, other questions
remained unanswered and a new well (Gibraltar-2)
was therefore planned. This well will attempt to
identify the depths of the gas-oil and oil-water
contacts, and to investigate whether and in what way
the fluids change phase with depth. The well will also
assess whether productivity varies in the flank of the
structure relative to the crestal locations. If Gibraltar2 produces significant volumes of gas, this could be
reinjected through Gibraltar-1.

331

to other fields in the foothills. Because of this, wells

will drain the structure rapidly and efficiently.
(iv) Gibraltar is located around 2km from the Cao
Limn -Coveas pipeline which has available capacity,
permitting rapid and economic transportation of
produced oil.
(v) Given the fact that Occidental had invested
significant funds in well location and construction,
finding costs for Ecopetrol were low (approximately
7-10 cents per brl o.e.).
(vi) Perhaps most important, the Gibraltar
discovery has opened an exploration frontier in the
Llanos foothills.
CONCLUSIONS

DISCUSSION

1.The Gibraltar gas condensate discovery has opened

a new exploration frontier in the northern Llanos
Foothills of Colombia. The main characteristics of
Gibraltar are shared by other fields along similar
trends in the area and include: complex structure,
complex reservoirs, and complex compositions of
hydrocarbon fluids.
2.The Gibraltar discovery well produced
approximately 690 barrels of 57 API oil and 44 million
cubic feet of gas per day; the trap is close to available
pipeline infrastructure and production can be delivered
easily to the export terminal.
3. Currently the well is producing approximately the
same volumes during an extended test. Pressure has
not decreased, confirming originally estimated
volumes.
4. Because of the geological complexity of the
Foothills, missed or bypassed pay opportunities are
the norm rather than the exception here. Cusiana,
Cupiagua, Capachos and Gibraltar are examples of
this type of opportunity. However, drilling and
operational best practices are required for these
developments.

Reservoir rocks in the Llanos foothills area are nonconventional in that they have low porosity although
permeabilities can be high; tests indicated that
permeability at Gibraltar-1 averaged >2D.
Exploration and field development can have uncertain
outcomes, but at Gibraltar, however, a number of
factors combined to enhance the discoverys economic
viability:
(i) The condensate is of high quality, with an API
of 57o.
(ii) The crest of the Gibraltar structure is at a depth
of 6,500ft, making it the shallowest in the Llanos
foothills trend; drilling costs and duration of drilling
and completion are therefore relatively low.
(iii) DST-2 showed that the reservoir has an
average permeability of 2D, which is high compared

Note added in proof: well Gibraltar-2

At the time of submission of this paper, Ecopetrol was
drilling well Gibraltar-2 A-1 from the same location
as Gibraltar-1 to target the western limb of the fold,
downdip from the position where the original well
penetrated the top of the reservoir. The well aimed to
reach the top of the Mirador Formation approximately
500ft below the top of the Mirador drilled by
Gibraltar-1 and several hundred metres to the NW.
The well was to navigate sub-parallel to stratigraphy
within the Mirador and exit the Mirador approximately
1,000 ft below the lowest-known hydrocarbon level
determined by Gibraltar-1.
Operationally, Gibraltar-2 was successful and the
well is currently in extended tests. Gibraltar 2,
however, did not reach the original target box but

332

The Gibraltar discovery, Llanos foothills, Colombia

ended up testing the steep eastern limb of the fold.

Because of this, the well could not navigate within
the Mirador Formation in order to extend the oil
column downwards in a significant manner and could
not find variations in fluid composition and fluid
contacts. In brief, the technical results of Gibraltar-2
are as follows:
(i) The entire reservoir section was hydrocarbon
saturated and the LKH (lowest known hydrocarbon
level) was lowered by an additional 1,000 ft. The total
height of the hydrocarbon column is 2,140 ft. No oilwater contact was encountered and reservoir
boundaries were not found at the lowest tested interval.
(ii) It was demonstrated that the richness of the
condensate increased slightly with depth, suggesting
however that the original reserves estimates will
probably increase.
(iii) Dips encountered by the well resulted in the
slight modification of the structural model, also
impacting positively on reserves estimates.
(iv) The well has been in extended test for the past
few months, production of about 580 barrels and about
48 million cubic feet has been stable and little drop in
pressure has been measured.
(v) The current proven reserves estimation, based
on six months of extended tests and on the minor
changes in reservoir pressure associated with draining
of the reservoir, is that the reservoir has at least 120
million barrels of oil equivalent (15 MM brls oil and
630 B cu ft gas) in just the northern part of the
structure. These numbers have a certainty of 90%. The
well has extracted 5 B cu ft and 67,000 barrels of 53
API oil which have been exported. The P50 estimated
geological recoverable reserves remain above 200
million barrels of oil equivalent, the hydrocarbonwater contact has not been encountered, and the
southern part of the structure has not been considered
in volumetric calculations.
ACKNOWLEDGEMENTS
Alberto Caldern, former President of Ecopetrol,
Vctor Eduardo Prez, former senior E&P Vice
President, Luisa Fernanda Lafourie, former Minister
of Energy and Mines, Luis Ernesto Meja, current
Minister of Energy and Mines, and Isaac Yanovich,
current President of Ecopetrol, provided support and
encouragement for the Gibraltar project for which they
are thanked. Jaime Martnez (Ecopetrol security),
Luci Alcira Pachn (Ecopetrol Community Relations)
and Enrique Gallardo from the division of contracts
managed above-ground issues. Hector Alfonso, Alba
Mesa and Carlos Jaramillo assisted with geophysical
processing, petrology and biostratigraphy,
respectively. Isaac Yanovich and Mauricio Salgar
(Chief Operating Officer of Ecopetrol) kindly granted

permission for publication of the paper. The senior

author also thanks Claudia Arango and Camila
Villamil for their support. Miguel Ramirez, Martin
Keeley and Andy Pulham provided valuable
comments during journal review which are
acknowledged with thanks.
REFERENCES
ARISTIZABAL, J. J., SANCHEZ, J., PAZ, M. G., MUOZ, J.,
VILLAMIL, T., JARAMILLO, C. and RUEDA, M., in
preparation. GOIS, a method for the geological control of
complex frontier areas: details from the Gibraltar discovery.
Journal of Petroleum Geology.
ARTHUR, M. A., SCHLANGER, S. O., and JENKYNS, H. C.,
1987.The Cenomanian Turonian oceanic anoxic event I, II.
In: Brooks, J. and Fleet, A. (Eds), Marine petroleum source
rocks. Spec. Publ. Geol. Soc. London 26, 216-218.
CAZIER, E. C., COOPER, M. A., EATON, S. G., and PULHAM,
A. J., 1997. Basin development and tectonic history of the
Llanos Basin, Eastern Cordillera, and Middle Magdalena
Valley, Colombia. AAPG Bull., 81, 1332-1335.
CAZIER, E. C., HAYWARD, A. B., ESPINOSA, G., VELANDIA,
J., MUGNIOT, J-F. and LEEL,W. G., 1995. Petroleum Geology
of the Cusiana field, Llanos Basin foothills, Colombia. AAPG
Bull., 79, 1444-1463.
COLLETTA, B., HEBRARD, F., LETOUZEY, J., WERNER, P. and
RUDKIEWICKZ, J., 1990. Tectonic style and crustal
structure of the Eastern Cordillera (Colombia) from a
balanced cross section. In: Letouzey, J. (Ed.), Petroleum and
Tectonics in Mobile Belts Paris, Editions Technip, 81-100.
COOPER, M. A., ADDISON, F. T., ALVAREZ, R., HAYWARD, A.
B., HOWE, S., PULHAM, A. J. and TABORDA, A., 1995 a.
Basin development and tectonic history of the Llanos Basin.
In:Tankard, R., Suarez, S. and Welsink, H. J., (Eds), Petroleum
Basins of South America. AAPG Memoir 62, 659-665.
COOPER, M. A., ADDISON, F. T., ALVAREZ, R., HAYWARD, A.
B., HOWE, S., PULHAM, A. J. and TABORDA, A., 1995 b.
Basin development and tectonic history of the Llanos Basin,
Eastern Cordillera, and Middle Magdalena Valley, Colombia.
AAPG Bull. 79, 1421-1443.
CORREDOR, F., 2003. Eastward extent of the late Eoceneearly Oligocene onset of deformation across the northern
Andes: constraints from the northern portion of the
Eastern Cordillera fold belt, Colombia, and implications
for regional oil exploration. VIII Simposio Bolivariano,
Exploracin Petrolera en las Cuencas Subandinas, 34-45.
DENGO, C. A. and COVEY, M. C., 1993. Structure of the
Eastern Cordillera of Colombia: implications for trap styles
and regional tectonics. AAPG Bull. 77, 1315-1337.
ESTRADA, C. A. and JARAMILLO, J. A., 2003. Making the
difference: a story of pitfall and successes in seismic imaging
in a thrust belt environment. VIII Simposio Bolivariano,
Exploracin Petrolera en las Cuencas Subandinas, 371-379.
MACELLARI, C. E., 1988. Cretaceous Paleogeography and
Depositional cycles of Western South America. Journal of
South American Earth Sciences, 1, 373-418.
MACELLARI, C. E., and DE VRIES, T. J., 1987. Late Cretaceous
upwelling and anoxic sedimentation in northwestern South
America. Palaeogeography, Palaeoclimatology, Palaeoecology,
59, 279-292.
MARTNEZ, J. A., 2003. Modelamiento structural 3D y
aplicaciones en la exploracin y explotacin de
hidrocarburos en el cinturn de cabalgamiento del
Piedemonte Llanero, Cordillera Oriental, Colombia. VIII
Simposio Bolivariano, Exploracin Petrolera en las Cuencas
Subandinas, 106-117.
OLEARY, J.,WARREN, E., GEEHAN, G., and HERBERT, R., 1997.

T. Villamil et al.

Evaluation of reservoir quality in the Llanos Foothills,

Colombia. VI Simposio Bolivariano Exploracion en las Cuencas
Subanadinas, Tomo I, 163-166.
PINDELL, J. L., HIGGS, R., and DEWEY, J. F., 1998. Cenozoic
palinspastic reconstruction, paleogeographic evolution and
hydrocarbon setting of the northern margin of South
America. In: Pindell, J. L., and Drake , C., (Eds),
Paleogeographic Evolution and non-Glacial Eustasy,
northern South America. SEPM Special Publication 58, 4585.
PULHAM, A. J., 1994. The Cusiana field, Llanos Basin, eastern
Colombia: high-resolution sequence stratigraphy applied
to late Paleocene-Early Oligocene estuarine, coastal plain
and alluvial clastic reservoirs. In: S. D., Johnson, (Ed.), High
Resolution Sequence Stratigraphy: innovations and
applications. University of Liverpool, Abstract Volume, 63-68.
PULHAM, A., MICHELL, D., MACDONALD, D., and DALY, C.,
1997 a. Reservoir modelling in the Cusiana Field, Eastern
Colombia; Characterization of a deeply-buried, low
porosity reservoir. VI Simposio Bolivariano Exploracion en
las Cuencas Subanadinas, Tomo I, DK:198-216.
PULHAM,A.J., MITCHELL,A., MACDONALD, D. and DALY, C.
1997 b. Sequence stratigraphic evaluation and production
characteristics of the Eocene Mirador Formation, Cusiana
Field, Eastern Colombia. In: Shallow Marine and Nonmarine
Reservoirs: Sequence Stratigraphy, Reservoir Architecture
and Production Characteristics, K. W. Shanley and B. F.
Perkins, (Eds.), GCSSEPM Foundation 18th Annual Research
Conference, Shallow Marine and Nonmmarine Reservoirs, 225233.

333

RATHKE, W. W. and CORAL, M. G., 1997. Cupiagua Field,

Colombia: Interpretation case of a large, complex thrust
belt gas condensate field. VI Simposio Bolivariano Exploracion
en las Cuencas Subanadinas, Tomo I,119-128.
ROEDER, D. and CHAMBERLAIN, R. L., 1995. Eastern
Cordillera of Colombia: Jurassic-Neogene crustal evolution.
In: Tankard, A. J., Suarez, S., Welsink, H. J., (Eds) Petroleum
Basins of South America, AAPG Memoir 62, 633-645.
VILLAMIL,T., 2004. Regional hydrocarbon systems of Colombia
and western Venezuela: their origin, potential, and
exploration. Chapter 32 in: Bartolini, C., Buffler, R. and
Blickwede, J., Eds: The Circum-Gulf of Mexico and the
Caribbean: Hydrocarbon habitats, basin formation and
petroleum. AAPG Memoir 79, 697-734.
VILLAMIL, T., 1999. Campanian-Miocene tectonostratigraphy,
depocenter evolution and basin development of Colombia
and Western Venezuela. Palaeogeography Palaeoclimatology
Palaeoecology, 153, 239-275.
VILLAMIL, T. and ARANGO, C., 1998. Integrated stratigraphy
of latest Cenomanian and early Turonian facies of Colombia.
In: Pindell, J. L., and Drake, C., (Eds.), Tectonostratigraphic
studies in northern South America: basin histories and
implications for eustasy during non-glacial times. SEPM
Special Publication 58, 129-159.
VILLAMIL,T.,ARANGO, C. and HAY,W.W. , 1999. Plate tectonic
paleoceanographic hypothesis for Cretaceous source
rocks and cherts of northern South America. In: Barrera
E. and Johnson, C. C., Evolution of the Cretaceous OceanClimate System, GSA Special Paper 332, 110-121.