DOVE ENERGY

Structural Analysis of Highly Fractured, Heterogeneous Basement,

Sayun-Masila Basin, Yemen
ANN MURRAY, Task Geoscience, Aberdeen, United Kingdom, DAVE MONTGOMERY and DONALD
MILNE, Dove Energy, Aberdeen, United Kingdom
Introduction
Fractured basement rocks have become an increasingly common target for hydrocarbon production in
the Republic of Yemen. The aim of this study is to determine basement characteristics with regard to
predicting reservoir potential within the Bayoot Field of Block 53, Yemen (Figure 1).
Geological context
The Bayoot Field is located on the northern margin of the Sayun Masila Basin. The basin is a NW-SE
trending graben formed during Late Jurassic to Early Cretaceous rifting related to Gondwana breakup
(Menzies et al., 1992, Harris et al., 2003). A further phase of rifting with associated transtensional and
transpressional structures occurred during the Oligocene and Miocene as a result of opening of the Red
Sea and Gulf of Aden (Ahlbrandt, 2002, Sakr et al., 2005). Structures formed during these phases were
principally controlled by rejuvenation of the Late Proterozoic Najd shear fault system (Beydoun et al.,
1996). The resulting structural framework comprises arrays of parallel and kinked en-echelon faults that
further compartmentalise the graben into sub-basins. Complex secondary structures including strikeslip, oblique-slip, thrust and normal faults, folds and dyke swarms are associated with some of the major
faults, particularly near their terminations (McMahon Moore, 1979 and Bott et al., 1992).
Five wells were drilled into a basement high termed the Rudood Ridge, which is located in the footwall
of a locally significant SW-dipping transtensional fault (Figure 1). In the study area crystalline basement
is thought to be of Archaean age, but its composition is poorly constrained. It has probably endured
substantial periods of sub-aerial exposure, which has caused weathering of its upper reaches. In
addition, the various phases of deformation have led to pervasive fracturing together with rift-related
intrusions. Hydrocarbon emplacement is through fault juxtaposition of the fractured basement against
Late Jurassic organic rich-shale source rocks of the Madbi formation.
Database
Comprehensive petrophysical log suites, production log data, mud logs, cuttings, junk basket and
sidewall core basement samples, 3-D seismic and 3430 m of borehole image were made available for
interpretation (Table 1). Interpretation of some datasets is on-going.
Table 1.

Well data summary. Petrophysical logs and 3-D seismic data were available for all wells. Note
that the study wells are variably deviated. + Mark of Baker Atlas, *mark of Schlumberger.

Well

Wellbore deviation (from

vertical)

Borehole image log type

Production log data

Petrography

Bayoot SW-1

Sub-vertical

STAR+

Not acquired

Whole rock junk basket and

cuttings

Bayoot SW-2

20/021

FMI/UBI*

Bayoot SW-3
Bayoot SW-4
Bayoot SW-5

52/349
51/020
51/030

Interpreted

Cuttings

STAR/CBIL

Interpreted

Not acquired

STAR/CBIL

Acquired (work in progress)

Not acquired

STAR/CBIL

Not yet acquired

Side wall cores (work in

progress)

Figure 1.

Location and top basement depth structure maps showing the study wells.

Methodology
Borehole image logs permitted seven different fracture types to be identified and oriented. Where
acoustic transit time images were available, possible open fractures were classified and their apparent
along-hole widths were visually ranked into fine, medium and wide. Fractures were displayed and
analysed stereographically and corrections for borehole bias were applied (after Terzaghi, 1965).
Fracture intensity statistics, per metre, were calculated.
A structural image facies scheme was developed to aid in constraining fracture style and intensity,
potentially vuggy and brecciated intervals, and zones of degraded borehole image quality. Its purpose is
to highlight sections where fractures are undersampled (i.e. where borehole image quality is poor or
where fracturing is too intense to accurately trace and orient features with confidence).
Present day in-situ stress features oriented from the images comprise drilling induced tension fractures
and borehole breakout. With the exception of the sub-vertical Bayoot SW-1 well, the study wells are
strongly deviated. This has implications for direct inference of principal horizontal stress axes from the
borehole indicators (c.f. Mastin 1988). Further geomechanical modelling, which is outside of the current
study scope, would be required to interpret these features with confidence. Nonetheless, the most
confident stress indicators display reasonably consistent axes. Sub-sets of part open fractures dipping in
excess of 50 and striking 030 to the derived principal horizontal axis were filtered from the fracture
databases. These features could be critically stressed and thus more likely to contribute to fluid flow if
plumbed into the flowing fracture network (c.f. Rogers, 2003). Critically stressed fracture occurrence
was compared against production log data to establish whether any correlations existed.
Petrographical description, point counting and X-ray diffraction (XRD) of cuttings, junk basket rock
samples and sidewall cores were undertaken and rock types were determined. A 14-fold rock typing
scheme was developed using the petrophysical logs and petrography results and was applied across the
five wells. Relationships between rock type, fracturing style and structural image facies classification
were explored.
Interpretation summary
Overall, fracture intensity was found to be very high (in excess of 20 fractures per metre) although rare
zones of low fracture intensity occur locally. Fractures were found to be extremely scattered in terms of
inclination, azimuth and strike; no clear fracture sets were evident and any apparent preferred
orientations can be explained by borehole bias effects. Whilst seven different fracture types were
identified from the images, it was not possible to further categorise the features as primary phase
cooling fractures, secondary tectonic fractures or exhumation (unloading) fractures. A similar
conclusion was reached by Genter et al. (2000) in their study of Rhine Graben granites; core was
required for full fracture typology.
Present day in-situ stress indicators were found to support a NE-SW principal horizontal stress axis;
orthogonal to the local fault strike and Rudood Ridge axis (Figure 1). Whilst direct inference of
horizontal stresses from the deviated wellbores is deemed low confidence, the results are in agreement
with published studies (c.f. Al-Kotbah and Al-Ubaidi, 2001 and Sakr et al. 2005). An exception is well
Bayoot SW-1 where the inferred principal axis is oriented NW-SE, parallel to the local bounding fault.
Due to the complex structural framework, rotations in local stress fields are likely.
Part open fractures that are inferred to be critically stressed together with sections of vuggy image
facies were found to correlate with zones of formation fluid inflow as interpreted from the production

log data. Fluid inflow is greatest within the upper reaches of the basement. Seismic data show that
splays from the locally significant fault (Figure 1) intersect the wells just above the top basement pick
where mud log reports show evidence of notable weathering. It is possible that both the vicinity to the
fault together with enhancement and improvement of fracture interconnectivity through weathering
promotes flow through this section. Indeed a notable seismic anomaly is also observed here, but it is
unclear whether this is due to a distinct change in rock properties or the presence of hydrocarbons.
Mud losses recorded at greater depths also correlate with part open fracture occurrence. Furthermore,
they were identified where there were no recorded fluid inflows or losses. This may suggest that these
particular fractures are short-lived and discontinuous and/or poorly interconnected.
Petrophysical log responses show the basement to be highly heterogeneous; a total of fourteen
different rock types were constrained from the logs acquired from the five study wells. There are
similarities in the rocks encountered between the closely spaced wells, but rock unit continuity is
complex. The more distal well, Bayoot SW-1, penetrates rocks with distinctly different log responses.
Petrographical and XRD analysis of cuttings, junk basket and sidewall core rock samples show them to
be plagioclase-, quartz-, hornblende- and biotite-rich. Alteration is common and probably indicative of
significant hydrothermal activity. A variety of rock types were suggested based on point count and XRD
results, but it was not possible to categorically classify cuttings and junk basket rock samples as being of
igneous or metamorphic origin; however, the sidewall cores are considered to be igneous. The
determined rock types include; quartz micro-monzodiorites, micro-diorites, quartz micro-diorites, microgranodiorites, micro-tonalites and quartz-rich microgranities if igneous in origin; and amphibolites,
gneisses or quartzites if metamorphic in character. Taking into account both log and petrography
evidence it is likely that well Bayoot SW-1 penetrates metamorphic rocks.
Alkali feldspars are very rare in the Bayoot SW-1 and -2 samples, however provisional observations of
sidewall cores from Bayoot SW-5 suggest an abundance of alkali granite. Rock types determined from
log responses and sidewall core compositions correlate well.
Both filled and open fractures are clearly visible from the whole rock specimens and the rare crystal
aggregates present in cuttings. Fracture fill material includes pyrite, calcite and secondary, highly
sericitised plagioclase. Some thin sections reveal part open fractures up to 20 m wide and very rare
secondary porosity following grain dissolution. No samples were acquired from the upper, possibly
weathered basement section. Relationships between imaged fracturing style and occurrence with
respect to the derived rock types are unclear.
Summary
Basement in the Bayoot Field is intensely fractured and heterogeneous. No clear relationships between
the imaged style of fracturing and occurrence with respect to rock type were found. A dominant NE-SW
principal horizontal stress axis inferred from imaged indicators agrees with published results.
Hydrocarbons appear to be confined to the upper reaches of the basement. It is inferred that prolonged
sub-aerial exposure has enhanced the fracture network through weathering in this zone. The locally
dominant fault that intersects the wells directly above the top basement has probably acted as a direct
conduit between the juxtaposed Late Jurassic source rocks and the fractured basement.

References
AHLBRANDT, T.S., 2002. Madbi Amran / Qishn total petroleum system of the MaRib-Al Jawf / Shabwah, and Masila-Jeza Basins,
Yemen. U.S. Geological Survey Bulletin 2202-G, Version 1, 28pp. Available on-line at: http://pubs.usgs.gov/bul/b2202-g/b2202g.pdf
AL-KOTBAH, A.M. and AL-UBAIDI, M.R. 2001. Principal stress orientation of Yemen faults in the Mesozoic age.
University, Faculty of Science Bulletin, 14, 85-103.

Sanaa

BEYDOUN, Z. R., AS-SARURI, M. A. L., EL-NAKHAL, H., AL-GANAD, I. N., BARABA, R. S., NANI, A. S. O. and AL-AAWAH, M.H. 1996.
International lexicon of Stratigraphy, v. III, Asia, Republic of Yemen, International Union of Geological Sciences Publication, 34,
Fascicule 10b2, 245p.
BOTT, W.F., SMITH, B.A., OAKES, G., SIKANDER, A.H. and IBRAHAM, A.I. 1992. The tectonic framework and regional
hydrocarbon prospectivity of the Gulf of Aden. Journal of Petroleum Geology, 15, 211-243.
GENTER, A., TRAINEAU, H., LEDESERT, B., BOURGINE, B. and GENTIER, S. 2000. Over 10 years of geological investigations within
the HDR Soultz project, France.
Proceedings of the World Geothermal Congress, 2000.
Available on-line:
www.geothermie.de\egec-geothernet\prof\hdr\0710.pdf.
HARRIS, R., COOPER, M., SHOOK, I. 2003. Focusing oil and gas exploration in eastern Yemen by using satellite images and
elevation data alongside conventional 2D seismic. Focus Article (modified version of an article that appeared in World Oil,
November 2002). Available online at:
http://www.cseg.ca/recorder/pdf/2003/02feb/feb03_03.pdf
McMAHON MOORE, J., 1979. Tectonics of the Najd transcurrent fault system, Saudi Arabia. Journal of the Geological Society,
136, 441-452.
MASTIN. L. 1988. Effect of borehole deviation on breakout orientations. Journal of Geophysical Research, 93 (B8), 9187-9195.
MENZIES, M. A., BAKER, J., BOSENCE, D., DART, C., DAVISON, I., HURFORD, A., ALKADASI, M., MCCLAY, K., NICHOLS, G.,
ALSUBBARY, A. and YELLAND, A.. 1992. The timing of magmatism, uplift and crustal extension: preliminary observations from
Yemen. Geological Society, London, Special Publications 1992, 68, 293-304.
ROGERS, S.F. 2003. Critical stress-related permeability in fractured rocks. In AMEEN, M (ed.) Fracture and in-situ stress
characterisation in hydrocarbon reservoirs. Geological Society, London, Special Publications, 209, 7-16.
SAKR, K., ABDEL-MONEM, S.M., KHALIL, H., MAHMOUD, S., HAMIMI, Z., AL-AYDRUS, A., AL SUBBARY, A., AL-KOTTBAH, A., EL
GANAD, A., AL-GABARY, A. and AL-WOSABI, M.A. 2005. A study of crustal deformation along the Red Sea region using geodetic
and seismic data from Egypt and Yemen. Acta Geophysica Polonica, 53, 75-85.
TERZAGHI, R.D. 1965. Sources of error in joint surveys. Geotechnique, 15, 287-304.

Acknowledgements
Many thanks to Philip Whiteley and Clive Simpson of Goodland Ventures and Heike Delius and Brian
Callaghan of Task Geoscience for all of their help and support.