Click to view page images in PDF format
INITIAL GEOPHYSICAL AND GEOLOGICAL ASSESSMENT OF AN INDUSTRY 3D SEISMIC
SURVEY COVERING THE JAPEX/JNOC/GSC MALLIK 5L-38 GAS HYDRATE RESEARCH WELL
T. A. Brent1, M. Riedel2, M. Caddel3, M.
Clement4, T. S. Collett5, and S. R. Dallimore2
1 Geological Survey of Canada, 3303 - 33rd Street NW, Calgary, AB T2L 2A7
2 Geological Survey of Canada, 9860 W. Saanich Rd, Sidney, BC, V8L 4B2
3 BP Canada Energy Company, Calgary 240- 4th Avenue SW, Calgary, AB, T2P 2H8
4 Veritas GeoServices, 715 Fifth Avenue SW, Suite 2200, Calgary, AB, T2P 5A2
5 United States Geological Survey, Denver Federal Center, Box 25046, MS-939, Denver, Colorado 80225
An industry 3D seismic
survey (Mallik 3D), covers 126 km including Mallik 5L-38 and three
industry exploration wells (Figure 1).
The data was recorded in 2002 on Richards Island over a lake-covered,
marine-inundated, permafrost terrain.
The acquisition was optimized for deeper
conventional hydrocarbons and thus
limited shallow spatial resolution, fold,
and frequency (60 Hz maximum).
Shallow data gaps reach 700 ms while
irregular permafrost ice creates complex
static solutions, degraded velocity
control, and energy transmission losses.
Data volumes illustrated come from two
processing flows including one intended
to produce a near-true-amplitude output.
The imaging quality of the Mallik 3D improves below about 700 metres, and
the wide aerial coverage reveals the
broader observations of geology and gas
hydrate
seismic
expression.
Three downhole velocity surveys tie to the Mallik 3D at Imperial Mallik J- 37, P-59 and A-06. Anomalously elevated velocities at these locations infer gas hydrate thickness of 225, 105 and 135 meters respectively, in addition to the 116 m meters known at Mallik 5L-38.
The Mallik 3D lies within a
regional fault zone (Taglu Fault Zone)
which is at least 150 km long and on
strike with a major hinge line of syntectonic
Tertiary deposition into the
Beaufort Sea Basin. The Mallik 3D
structural interpretation indicates both
compressional and extensional features.
A major down-to-north normal fault
(named F1, see Figure 1) strikes EW and
divides the Mallik 3D into a north block containing Mallik wells 5L-38 and P-59, and a south block containing Mallik wells J-37 and A-06. The
two southern block wells are drilled near the crest of a NW-striking thrust anticline. The major normal
fault has apparent throw, near 5L-38, of 800 meters, interpreted on seismic
at a level (marker “G”, see
Figure 2) deeper than the gas hydrate zones by about 700 m. Down-to-north offset of a biostratigraphic
datum indicative of Middle-to-Late Eocene age strata is 705 m, between 5L-38 and J-37, and occurs
about 600 m below the
seismic
correlation.
Seismic
character correlation across the fault is uncertain
and is explained by changing stratal thickness, due to fault growth possibly through Late Eocene or
Oligocene time, which would reduce fault throw upsection at the level of hydrate deposits of 5L-38, by
several hundred meters. This implies the hydrate-hosting sediments in Mallik wells at 5L-38 and P-59,
are younger and have a different geologic history than the sediments with inferred gas hydrate at Mallik
J-37 and A-06.
Stacked data of the Mallik 3D indicate data contamination which might potentially confuse gas hydrate interpretation, these include: 1.) energy reverberations generated from impedance at air/ground, top/base ice-bonded permafrost, intra-permafrost taliks, and at the base-Iperk sequence, and; 2.) Amplitude and frequency degraded zones (wash-outs) caused by lake-controlled near-surface conditions creating signal attenuation. Wash-outs are observed as deep as one second two-way-time, and are directly beneath and spatially coincident with the outline of specific lakes.
Anomalous seismic
energy from suspected gas hydrate is evident in the Mallik 3D. Observed
are numerous, mostly weak, discontinuous, near-horizontal
seismic
energy or amplitude changes which
appear to cut across or be superimposed on dipping primary geologic reflectors. The two-way-time
range to expect
seismic
evidence of the theoretical base of methane hydrate stability is estimated
between 825 and 1035 milliseconds. Pronounced amplitude “blanking” is apparent on
near-true-amplitude
processed data above levels within this time range. Throughout much of the Mallik 3D,
where primary reflectors are dipping, there are commonly several near-parallel energy levels
interpretable as being near the base of gas hydrate stability, unlike the classic
seismic
expression on
marine data which exhibit a single bottom-simulating reflector.
Other observations of gas hydrate interest include: 1.) apparent east dip of suspected
base-hydrate
energy into the anticlinal fold in the southern part of the Mallik 3D; 2.) subsurface positions of
certain major faults appearing to underlie the general trend of lakes at surface and 3.) Major faults
exhibiting a generally vertical
zone of amplitude/frequency degradation and/or chaotic
seismic
nature
around them in the subsurface.
Figure 1. Map showing trace of the Mallik anticline and faults (F1 to F6) interpreted at
seismic
marker “G”. Also shown are the location of Mallik area
wells and
seismic
lines A-B and C-D seen in Figure
2. Overlain is culture data including marine shoreline, lakes, rivers and 100 foot elevation
contours (thin dark lines). Note the general coincidence of lakes over fault F5 and southeast part of fault F4.
Figure 2. Interpreted Mallik 3D
seismic
lines A-B and C-D (location in
Figure 1). Overlain at Mallik area well locations are
checkshots or VSP-corrected velocity curves, depth scales (m KB), and biostratigraphic datum. Blue arrows indicate
base of ice-bonded permafrost and red arrows indicate weak
seismic
expression closest to the estimated base of methane
hydrate stability.
Vertical
zones of amplitude /frequency /coherency “wash-out” are outlined with red-dashed lines. Yellow
arrows indicate speculated
seismic
multiples, top GH energy, paleo-GH levels or diagenetic acoustic overprints possibly
the legacy of deeper paleo-hydrate levels.
Seismic
data are from pre-stack time migrated processing.