Jan.
2001 Vol. 222 No. 1 Feature Article
|
NATURAL GAS: Igniting New Markets Part 1:
Exploration Methods
Large gas-prospective areas indicated by bright spots
Results from combined vertical-incidence and
wide-angle seismic, as well as ODP data, argue in favor of gas-charged reservoirs in the deepwater Southern
Canary basin
Christian Müller, Friedrich Theilen and
Bernd Milkereit, Department of Geosciences Geophysics, University of Kiel, Germany
eismic
evidence acquired on RV Poseidon in 1997 and 1999 indicates the presence of large, deepwater gas
reservoirs in Middle Miocene-age sediments in the Southern Canary basin. Pronounced, seismic bright-spot
reflections in the sedimentary column with areal extents of more than 50 km2 are
well imaged on multichannel reflection seismic sections (MCS) and Ocean Bottom Hydrophone (OBH) records. This
article examines that evidence that these bright-spot areas are indeed gas prospective.
Normal-incidence reflection coefficients of R =
0.4, calculated from MCS shot records, indicate strong, negative acoustic-impedance contrasts. Further, mud
diapirs near bright spots outline vertical gas-migration paths. However, the nature and origin of these
prominent bright spots remain unresolved. Circumstantial evidence, such as occurrence of bright spots within
the hydrate-stability-zone and existence of large magma volumes in the lower crust and upper mantle, points
toward methane or CO2-rich volatiles trapped in the sedimentary sequence.
Introduction Reconnaissance
seismic surveys were conducted to image deepwater basins at the transition between continental margins off
West Africa and oceanic crust. The Canary Islands add complexity to the study area. Ocean island systems such
as Hawaii and the Canaries have been built by large volumes of magma intrusions and lava extrusions based on
intraplate volcanism. The Canary Islands, off the northwest African continental margin (Fig. 1), have been
formed by hot-spot volcanism, where subaerial volcanic activity started on Fuerteventura around 20 million
years ago (Ma).1
|
Fig. 1. Location map of survey area
showing reflection seismic lines and OBH stations acquired by RV Poseidon in 1997 and 1999. |
|
Today, most investigations focus on the subaerial part
of the islands and their submarine flanks.2 Large sediment flows have been found on the flank of
Tenerife by the Teide Group, indicating past catastrophic sediment failures.3 Recent megaslides
have also been studied west of El Hierro Island.4 Seismic surveys in deepwater areas focus on
landslide distribution and structure in the Northern Canary basin.5 A series of well-stratified
layers down to basement levels can be interpreted from MCS results and four ODP sites in the Northern and
Southern Canary basins.6
The main unconformities correlate with major phases of
volcanic activities in the area, e.g., Roque Nublo phase (4.3 3.4 Ma) and Mogan Group (14 to 13.3 Ma)
on Gran Canaria.7 Stratigraphic sequences are mainly formed by debris avalanches in the vicinity
of the islands. However, no indications for bright-spot reflections in this area have been reported in
literature.
During two cruises of
RV Poseidon in 1997 and 1999, a number of pronounced seismic bright spots were imaged on a dense grid
of 2-D reflection seismic lines, 1,000 to 2,600 ft (300 to 800 m) below seafloor at water depths of more than
12,500 ft (3,800 m) in the Southern Canary basin. The deepwater basin is located at the transition between
oceanic crust and the continental margin, indicated by a 175-Ma magnetic anomaly (S1) east of Fuerteventura.8
Bright-spot evaluation using forward modeling,
Amplitude Versus Offset (AVO) analysis and true-amplitude data-processing techniques require detailed
background P-wave velocity information. Most seismic surveys in deep water lack wide-angle reflections for
velocity analysis. The MCS maximum shot-receiver distances of 2,600 ft (800 m) at water depths of more than
12,500 ft (3,800 m) do not provide reliable subsurface velocites.
ODP Leg 157, located on the submarine flank of Gran
Canaria, provided limited stratigraphic and well control of the main bright-spot areas south and southwest of
El Hierro. This was helpful in interpreting major stratigraphic changes in the Southern Canary basin indicated
on seismic line 28/97, Fig. 2.
|
|
Fig. 2. (Top) Reconnaissance deepwater
seismic line 28/97 showing six, prominent bright-spot reflections in the Southern Canary basin. (Bottom)
ODP Site 956, southwest of Gran Canaria, provide limited stratigraphic and well control (V.E.=33). |
|
Velocity, density and porosity logs from these sites
are available for the upper 2,300 ft (700 m) of sediments, whereas the lowermost 1,000 ft (300 m) mainly
comprises volcaniclastic deposits originating from Gran Canaria. Pressure and temperature data provide
additional constraints for calculation of the stability field of volatiles.6 For acquiring local,
subsurface velocity structures in bright-spot areas, 15 OBH stations were deployed on the seafloor.
Vertical-Incidence Reflection Data
The multichannel reflection seismic data has been
processed (12 fold) to preserve relative-amplitude ratios. Seismic sections exhibit a number of pronounced
bright-spot reflections with high acoustic-impedance contrasts and clear lateral terminations embedded in
well-stratified sediments of low reflectivity, Fig. 3. The bright-spot reflection shows a clear phase reversal
compared to the seafloor reflection.
|
Fig. 3. Seismic section from line 17/99
showing bright-spot V embedded in well-stratified sediments of low reflectivity. |
|
Based on the dense grid of 2-D seismic profiles, the
bright-spot structure (line 17/99) south of El Hierro reveals an areal extent of about 25 km2.
Bright-spot reflections generally tend to follow sedimentary bedding, thus distinguishing them from bottom
simulating reflectors (BSRs). A second bright-spot structure southwest of El Hierro shows an areal extent of
more than 50 km2.
A prestack MCS shot record, where bubble pulses have
been suppressed by applying predictive deconvolution, is presented in Fig. 4. Phase reversal and bright-spot
reflection amplitude of about twice the seafloor-reflection amplitude are clearly visible. A seafloor
reflection coefficient of about 0.2 has been calculated from primary and multiple reflections in OBH records,
as well as velocity and density logs from ODP Site 956. From this, bright-spot reflection coefficients of
about 0.4 were calculated, indicating a strong decrease in acoustic impedance and P-wave velocity.
|
Fig. 4. Shot record clearly showing
phase reversal of bright-spot reflection at 5.35 s TWT compared to the seafloor reflection at 4.75 s
TWT. |
|
Seismic attribute mapping, particularly amplitude
response, has been applied on stacked MCS sections. For example, the average ratio between bright-spot
amplitude and seafloor reflection amplitude for bright-spot structure V, south of Tenerife, clearly exceeds
2:1. Amplitude ratios displayed in Fig. 5 show little variation throughout the entire bright-spot area,
whereas a slight amplitude decrease observed between traces 200 and 250 is caused by another high-reflective
bright spot located directly above this structure. Hereby, the local character of these amplitude anomalies
against sedimentary reflections is clearly seen.
|
Fig. 5. Amplitude ratio between bright
spot and seafloor reflection calculated from true-amplitude processed and stacked MCS data. Average
amplitude ratio of 2.23 is indicated by horizontal line. |
|
One-dimensional forward modeling of the near-vertical
bright-spot reflection response, based on Zoeppritz-equations, indicate that preferably negative impedance
contrasts (trapped volatiles) produce the high-reflection amplitudes observed in the data. Large, positive
impedance contrasts (e.g., sills, lava flows) including velocity and density constraints calculated
from ODP Site 956 do not produce sufficient reflection amplitudes.
Mud diapirs (seismic chimneys) give further evidence of
gas-saturated sediments.9 Three mud diapirs, interpreted as representing different developmental
stages, indicate vertical gas / fluid-charged sediment mobility from bright-spot depths into shallow sediments
and to the seafloor, where mud volcanoes are observed, Fig. 6.
|
Fig. 6. Seismic line 19/99 showing
bright-spot V near mud diapirs, indicating vertical gas / fluid-charged sediment mobility. |
|
Wide-Angle Seismic Data
Velocity analysis of the deepwater MCS reflections
provided poor resolution. Therefore, an AVO study using OBH stations at the seafloor was designed. In 1999, a
bright-spot location southwest of El Hierro Island was covered with 10 OBH stations on two lines at distances
of 2 to 4 mi. OBH records with high S/N ratio allowed access to P-wave structure in the sedimentary column and
upper oceanic crust, Fig 7. Strong, refracted energy indicates P-wave velocities of about 6 km/s for the
acoustic basement at 1-s TWT below seafloor. Here, the phase-reversed bright-spot reflection can be traced to
incidence angles of more than 40°.
|
Fig. 7. Wide-angle OBH record displayed
in reduced time with a reduction velocity of 6.0 km/s after suppression of the direct water wave.
Shallow primary reflections can be identified. |
|
Shallow sedimentary reflections in the OBH records are
superimposed with high-amplitude direct arrivals comprising the primary pulse and a sequence of bubble pulses
dominated by low frequencies. Application of predictive deconvolution was not successful, which was obviously
due to trace-to-trace variation of the source signature and the dominating direct-arrival amplitudes.
|
Fig. 8. Preferred bright-spot
velocity-depth model derived from semblance analysis and modeling of vertical-incidence reflection
coefficients. |
|
Application of a band-pass filter of 25 to 100 Hz,
followed by subtraction of an average estimate of the direct arrival, led to a significant improvement of S/N
ratio for primary wide-angle reflections. Subsequently, semblance-based velocity analysis was performed.
Considering P-wave velocity constraints from ODP Site 956 and Hamilton,10 a subsurface velocity
model for the target zone was established, Fig. 8.
Since OBH records are common receiver gathers,
shot-to-shot variation in source strength has to be corrected before any further AVO is performed. These
corrections have been applied after spherical-divergence correction based on amplitude variation of the direct
arrival. OBH records from shallow bright spots south of El Hierro provide the full range of incidence angles
for AVO analysis. Airgun-OBH geometry implies that reflection points are spread out over about 1 km, but no
significant changes in reflection characteristics at the bright-spot interface are observed from amplitude
analysis.
Fig. 9 shows the AVO response corrected for spherical
divergence and non-elastic attenuation at OBH 04, line 14/99. AVO modeling reveals the best fit for a constant
Poissons ratio of 0.35 across the interface, indicating presence of liquid volatiles rather than free
gas. Local pressure and temperature conditions, calculated from ODP Site 956 parameters, require CO2
and CH4 to be in liquid phase. Other investigations confirm that CH4 and CO2
in liquid phase can cause this pronounced decrease in bulk P-wave velocity.11
|
Fig. 9. AVO data calculated from OBH
record at bright-spot location V, plotted with synthetic AVO responses. Best fit is achieved for
constant Poissons ratio of 0.35, indicating dissolved volatiles rather than free gas. |
|
Large, negative normal-incidence reflection
coefficients that increase slightly with offset lead to definition as a Class III reservoir.12
Conclusions
The observed phase reversal, high amplitude ratios and
modeling point toward pronounced low-acoustic impedance layers caused by local fluid / gas accumulations in
the deepwater sediments. Lack of a lower boundary reflection is interpreted as a P-wave gradient zone caused
by a decreasing amount of dissolved gas below the bright spots, comparable to BSRs and gas hydrates at shallow
and intermediate water depths. At present, origins of these bright-spot reflections in the deepwater
environment between the Canary Islands and West Africa remain unresolved.
In this environment, CO2 and CH4
as main gas components are considered. During island formation, large amounts of CO2-rich magma
has been transported from deep crustal and mantle reservoirs to shallow magma chambers, while CO2
escapes due to decompression.13, 14 Large amounts of methane-rich gas have been drilled at ODP
Site 955, southeast of Gran Canaria, originating from organic-rich sediments transported from the African
continental margin, and bright spots following the sedimentary bedding point to low-permeability sequences as
reservoir seal.
Acknowledgment
This work was financially supported by the
Bundesministerium für Bildung, Wissenschaft, Forschung und Technologie BMBF (contract number 03G0530B)
and the Deutsche Forschungsgemeinschaft DFG (MI 558/4-1 and MI 558/4-2).
Literature Cited
- Coello, J., et al., "Evolution of the eastern
volcanic ridge of the Canary Islands based on new K-Ar data," J. Volcanol. Geotherm. Res., Vol.
53, pp. 251 274, 1992.
- Funck, et al., "Reflection seismic
investigations in the volcaniclastic apron of Gran Canaria and implications for its volcanic evolution,"
Geophysical Journal International, Vol. 125, pp. 519 536, 1996.
- Teide Group, "Morphometric interpretation of
the northwest and southeast slopes of Tenerife, Canary Islands," Journal of Geophysical Research,
Vol. 102, pp. 20,325 20,342, 1997.
- Urgeles, R., et al., "The most recent
megalandslides of the Canary Islands: El Golfo debris avalanche and Canary debris flow, west El Hierro
Island," Journal of Geophysical Research, Vol. 102, pp. 20,305 20,323, 1997.
- Watts, A. B. and D. Masson, "A giant landslide
on the north flank of Tenerife, Canary Islands," Journal of Geophysical Research, Vol. 100, pp.
24,487 24,498, 1995.
- Schmincke, H.-U., Proc. ODP, Initial Reports, Ocean
Drilling Program Vol. 157, College Station, Texas, 1995.
- Schmincke, H.-U. and M. Sumita, "Volcanic
evolution of Gran Canaria reconstructed from apron sediments: Synthesis of VICAP Project drilling,"
Proceedings of the Ocean Drilling Program, scientific results, Vol. 157, pp. 443 469, 1998.
- Verhoef, J., et al., "Magnetic anomalies off
West-Africa (20° 38°N)," Marine Geophysical Research, Vol. 13, pp. 81
103, 1991.
- Heggland, R., "Detection of gas migration from
a deep source by the use of exploration 3-D seismic data," Marine Geology, Vol. 137, pp. 41
47, 1997.
- Hamilton, E. L., "Sound velocity gradients in
marine sediments," J. Acoust. Soc. Am., Vol. 65(4), pp. 909 922, 1979.
- Wang, Z., and A. Nur, "Aspects of rockphysics
in seismic reservoir surveillance," In Sheriff, R. E.: Reservoir Geophysics, SEG, Tulsa, Oklahoma, 1992
- Rutherford, S. R. and R. H. Williams, "Amplitude-versus-offset
variations in gas sands," Geophysics, Vol. 54, pp. 680 688, 1989.
- Schmincke, H.-U., Vulkanismus,
Wissenschaftliche Buchgesellschaft Darmstadt, 1986.
- Gerlach, T. M. and E. J. Graeber, "Volatile
budget of Kilauea volcano," Nature, Vol. 313, pp. 213 277, 1985.
The authors |
Christian Müller
received his undergraduate degree from the University of Kiel in 1997 where he worked on
application of ground-probing radar techniques. Since 1998, he has been working toward his PhD at the
University of Kiel interpreting bright-spot reflections in the Southern Canary basin and applying various
seismic techniques such as AVO, inversion and modeling. He is student member of SEG, AGU, EAGE, EEGS, EGS
and DGG. |
Dr. Friedrich Theilen
is a marine geophysicist at the Department of Geosciences of the Christian-Albrechts-University in Kiel.
His main research focuses on marine seismology and development of methods for acquiring physical
parameters of marine sediments. Major ongoing projects are related to investigation of the
seismostratigraphic situation in the Canary basin and the Tjoernes Fracture Zone north of Iceland. He is
member of SEG, AGU, EEGS and DGG. |
Bernd Milkereit
is currently director of the geophysical section of the Department of Geosciences at the University of
Kiel. He graduated in 1981 from Kiel, where he worked until 1985. He earned his PhD in 1984, joined the
Geological Survey of Canada from 1985-96, and returned to the University of Kiel in 1996. His main
interests are 3-D seismic data acquisition, processing and interpretation. He is member of CSEG, EAGE, AGU
and SEG. |
|
|
|
Part 2: North
America Outlook |
|
Part 3: HP/HT
Drilling and Completions |
|
Part 4: Tight
Formation Stimulation |
|
Part 5: Sour
Gas Handling |
|
Part 6:
Anomalously pressured zones |
|
Part 7:
Gathering and Compression |
|
Part 8:
Monetizing Stranded Gas |
|