Geophys.J.Int.(2006)166,155–169doi:10.1111/j.1365-246X.2006.02805.x G J I S e i s m o l o g y
Three-dimensional ismic reflection investigation of the upper
crustal Winagami sill complex of northwestern Alberta,Canada J.Kim Welford 1∗and Ron M.Clowes 2
1Department of Earth and Ocean Sciences,University of British Columbia,V ancouver,B.C.,Canada 2Lithoprobe Secretariat,University of British Columbia,V ancouver,B.C.,Canada
Accepted 2005September 20.Received 2005August 29;in original form 2005February 23
S U M M A R Y
Using a 3-D ismic reflection data t recorded by the Canadian petroleum exploration in-
dustry to 5.1s two-way traveltime (∼15km depth)in northwestern Alberta,the Winagami
reflection quence (WRS),an interpreted sill complex previously identified on Lithoprobe
2-D multichannel ismic reflection lines,is investigated to determine its 3-D geometry and
reflective characteristics.Clear evidence of the top of the WRS emerges from the data t.Data
ctions outline a 3-D reflective sheet dipping to the southeast.From polarity comparisons with
reflections from the dimentary quence,the reflective signature of the deep body is inferred
to result from higher impedance material than the surrounding host rocks,thus reinforcing
earlier interpretations that the deep reflections result from doleritic sills intruded into gneissic
crystalline bament.Simple 1-D forward modelling reveals that the thickness of the sheet is
on the order of 100m throughout the survey region.From 3-D Kirchhoff forward modelling,a
model of the reflective sheet is constrained between 11and 16km depth along the northwestern
edge of the survey area.The abnce of the Winagami reflections,and bament reflectivity in
general,in the southeast of the survey region coincides with a positive aeromagnetic anomaly
inferred to be caud by magmatic rocks.The prence of the magmatic rocks may have either
influenced the geometry of the intrusion of the sills,overprinted their reflective signature or
been co-magmatic,dependent upon the relative timing of emplacement of the two features.
The Winagami sill complex appears to have been intruded horizontally into the crust during a
period of tectonic compression.The emplacement of such deep Precambrian sills reprents
an under-appreciated mode of magmatic addition into the crystalline crust.
Key words:Alberta,continental evolution,deep ismic reflection,mafic sills,Precambrian,
ismic modelling.
1I N T R O D U C T I O N From 2-D multichannel ismic reflection lines acquired in west-ern Canada by Lithoprobe,quences of coherent upper crustal reflectivity have been obrved within the crystalline crust of north-ern Saskatchewan (Mandler &Clowes 1997),southern Alberta (Mandler &Clowes 1998)and northwestern Alberta (Ross &Eaton 1997).Due to similarities in the reflective geometries and char-acteristics of the quences,imaged cross-cutting relationships in northwestern Alberta and nearby surface outcrops in northern Saskatchewan,the reflector quences have been interpreted as mafic sill complexes.One of the quences,the Winagami reflec-tion quence (WRS)of northwestern Alberta,equivalently referred to herein as the Winagami
sill complex,is characterized by one to ∗Now at:Department of Earth Sciences,Memorial University of Newfoundland St.John’s,NL,Canada.E-mail:kwelford@esd.mun.ca five bands of bright coherent reflections confined within a depth
range of 9–24km (3–8s two-way traveltime (TWTT))and appears to曲谱怎么学
extend over an area of more than 120000km 2(Ross &Eaton 1997).If the sill interpretation for the WRS is correct,this deep igneous feature is comparable in size to many of the world’s large Phanero-zoic igneous provinces (Coffin &Eldholm 1994).The considerable lateral extent of the inferred sill complex and its location within the upper crystalline crust underlying the Western Canada Sedimentary Basin,a locus of extensive oil and gas exploration,provided the im-petus for eking an industry-acquired 3-D ismic reflection data t to investigate the quence.While studies of the crystalline crust have generally been restricted to 2-D surveys,which produce cross-ctional images of the subsurface,the u of industry 3-D ismic reflection data permits the investigation of the deep sill bodies as
3-D features.Following negotiations with BP Canada in 2002,a
3-D data t that overlapped the Winagami quence was located in the Pine Creek region of north
western Alberta (Fig.1).Although the data t was acquired strictly for exploration purpos and so C 2006The Authors 155 by guest on May 10, fordjournals/Downloaded from
156J.K.W elford and R.M.
Clowes
Figure 1.Detailed bament domain map of the region surrounding the Pine Creek 3-D survey area.The locations of the CAT’92and PRAISE’942-D ismic reflection lines collected by Lithoprobe (with line numbers shown in white circles)are also shown.The inferred bifurcation of the Snowbird Tectonic Zone is shown in the eastern part of the map.Bament domains are bad on our interpretation of the aeromagnetic data.
was limited to recording times of 5.1s TWTT (∼15km depth),the nearby Lithoprobe evidence for the WRS indicated that at least the top of the Winagami complex could be imaged and mapped in 3-D using the Pine Creek data t.This study prents the results of our analys of this data t.Along with 3-D ismic investigations around the German KTB borehole (Stiller 1991)and studies of the Head-Smashed-In reflector of southern Alberta,Canada (Welford &Clowes 2004),this work reprents one of the pioneering efforts to examine upper–middle crustal structure to depths on the order of 15km using industry-style 3-D ismic reflection data.2B A C KG R O U N D I N F O R M AT I O N The WRS is obrved on four of the ten multichannel Lithoprobe lines from the 1992Central Alberta Tranct (CAT,lines 1–10)ex-periment and on eight of the ten lines from the 1994Peace River Arch Industry Seismic Experiment (PRAISE,lines 11–20)(Fig.1).Confined to the crystalline bament beneath the Western Canada Sedimentary Basin (WCSB),the sill complex extends throug
h v-eral Proterozoic bament domains.Most relevant to the Pine Creek survey are the Wabamun and the Chinchaga domains,ranging in age from 2.0to 2.4Ga (Ross 2002).Their boundaries,along with tho of most bament domains below the WCSB,have been determined on the basis of aeromagnetic data and K–Ar and U–Pb dating of drill cores that interct bament (Burwash et al.1962;Ross et al.1991;Villeneuve et al.1993;Pilkington et al.2000).From cross-cutting relationships with dipping crustal fabric and reflection offts from
dated brittle events,the emplacement of the Winagami sill complex
is thought to have occurred between 1890and 1760Ma (Ross &
Eaton 1997).As such,it postdates the asmbly of most of the do-
部分提前还款计算器mains in the study region and was perhaps coeval with tectonism as-sociated with the opening and closing of the Thorsby basin/domain to the southeast (Fig.1).Since the Winagami sills were intruded prior to the deposition of the Western Canada Sedimentary Basin,
C 2006The Authors,GJI ,166,155–169
by guest on May 10, 2014
3-D ismic investigation of the Winagami sill complex 157
满保their prent-day depth distribution between 10and 20km does not reprent the depth at which they were intruded.Compensating for the thickness of the dimentary cover and ignoring erosion of the crystalline bament prior to diment deposition,Ross &Eaton (2002)suggest that sill emplacement occurred between 6.5and 16.5km depth,at and above the brittle–ductile transition depending on the geotherm during emplacement.The emplacement of basaltic rock into the continental upper crust results from the upward transport of magma from the ba of the crust or the upper mantle along vertical conduits or dykes.Magma is intruded along a plane perpendicular to the least principal compres-sive stress (Gretener 1969;Parsons et al.1992)such that extensional environments will favour vertical intrusions while compressional zones will favour horizontal sheet intrusions.According to Zoback et al.(1989),most intraplate regions of the world are prently char-acterized by compressional stress regimes with extension only oc-curring in thermally uplifted regions.Assuming that such conditions also prevailed during the Proterozoic,occurrences of horizontal sills
should be widespread in intraplate regions of Precambrian crust that
除湿器原理
have experienced magmatism.The bands of bright reflections making up the WRS are locally horizontal to subhorizontal but together form a large-scale anasto-mosing fan that converges to the southeast (Ross &Eaton 1997)(Fig.2).Since tectonic stress typically are uniform over distances many times the thickness of the elastic part of the lithosphere (down to 10–35km)(Zoback et al.1989),large-scale intrusional patterns will be dictated by the stress distribution within plate interiors im-pod by the regional tectonics.Meanwhile,local stress variations will control the small-scale variations in the intrusional
pattern.Figure 2.Unmigrated stack of the Winagami reflection quence on the CAT and PRAISE ismic lines.On all plots,approximate depths are bad on an average velocity of 6km s −1.Since velocities in the diments are slower,the ba of the diments (which occurs between 1.8and 2.2s TWTT on the plots)is not as deep as shown.The ismic lines on which the Winagami is en are indicated on the location plot.Tho displayed are highlighted in black.Table 1.
Acquisition parameters for the Pine Creek 3-D data t in northwestern Alberta.
Total number of shots 3724Shot size/source 4kg dynamite Shot depth 18m Shot interval 140m Source line spacing 630m Receiver group interval 70m Receiver line spacing 560m Maximum sourc
e–receiver offt 16190m Average source–receiver offt 2904m Maximum number of live receiver lines 11王乐义案
Maximum number of live channels 2410
Geophones per receiver station 6Geophone spacing 4m Sample rate 2ms Record length 5.108s 3T H E P I N E C R E E K 3-D D ATA S E T The Pine Creek 3-D data t was collected by Veritas Energy Services Inc.in 2001January for BP Canada.The acquisition pa-rameters for this survey are prented in Table 1.The acquisition geometry of the Pine Creek data t is shown in Fig.3.For this 3-D survey,3724shots were recorded by patches of 5–11receiver lines each consisting of 12–254receiver groups.Receiver lines were spaced 560m apart and the receiver groups were connected at a spac-ing of 70m.Due to this variable acquisition,each shot was recorded by between 240and 2410receiver groups.
C 2006The Authors,GJI ,166,155–169
by guest on May 10, 2014
158J.K.W elford and R.M.
Clowes
Figure 3.Survey geometry for the Pine Creek 3-D data t.Data from the Pine Creek survey are of good quality.The processing flow applied to the data t consisted of trace editing,refraction stat-ics,application of a t 1.5geometrical spreading correction,spiking deconvolution,NMO correction,residual statics,trace
balancing,Figure 4.Velocity analysis example for the Pine Creek data t showing (a)a CMP gather,and (b)its mblance velocity spectrum.The picked velocity profile is shown by the dashed light grey line in (b).A 1000-ms AGC gain has been applied to the ismic data plot.
stack and post-stack 3-D f–k migration (using migration velocity
values corresponding to 90per cent of the average interval veloci-
ties computed for the entire data t).To highlight deep reflections
in the displayed stacked volume,f–x deconvolution was applied in
both the inline and crossline directions.
4E V I D E N C E F O R D E E P R E F L E C T O R S
I N T H E P I N E C R E E K R E G I O N
Bright Winagami-like reflections occurring between 4and 5.1s
闭上你的嘴TWTT are evident throughout most of the Pine Creek pre-stack
data t on both shot and CMP gathers (Fig.4a).Velocity m-
blance spectra plots point to high stacking velocities for the deep
reflections,which suggest that they originate from within the crys-
talline bament (Fig.4b).The deep reflections are unlikely to be
originating from scatterers outside the survey area due to their con-
sistent arrival times across the survey and the fact that the apexes
of the reflection hyperbolas on both the shot and CMP gathers are
腊肉炒蒜苗
generally at or near zero offt.
Throughout the stacked data volume,deep reflections are easily
obrved between 4and 5.1s TWTT on inline (Fig.5)and crossline
(Fig.6)ctions.While the consistency in the arrival times and
the relatively high amplitudes of the deep reflections obrved on
most of the CDP gathers result in brute stacks,which provide an
interpretable 3-D image of the deep reflectors,the application of
both refraction and residual statics helped improve the reflector’s
coherency.
Fig.5shows six inline ctions through the migrated Pine Creek
data volume.Progressing from the southwest to the northeast of the
data cube (ctions a–f),the ctions consistently show Winagami-
like horizontal to subhorizontal reflections between 4.5and 5.1s TWTT.For Sections (a)–(c),the horizontal to subhorizontal re-flections are truncated to the southeast.Moving northeast through the volume,Sections (d)–(f)provide a clearer image of a band of C 2006The Authors,GJI ,166,155–169
by guest on May 10, 2014
3-D ismic investigation of the Winagami sill complex
159
Figure 5.Inline ctions through the migrated Pine Creek data volume plotted between 2and 5.108s TWTT (approximate depth of 6–15km).The locations of the ctions with respect to the data cube are shown at the top using the lettered labels.The ctions have not been horizontally or vertically exaggerated.The locations of the reflections attributed to the Winagami quence are shown with the black arrows.The smaller white arrows point to shallower instances of bament reflectivity not interpreted to be part of the Winagami quence.Apart from the geometrical spreading correction,no gain has been applied to the data ctions.On all plots,approximate depths are bad on an average velocity of 6km s −1.Since velocities in the diments are slower,the ba of the diments (which occurs between 2and 2.5s TWTT on the plots)is not as deep as shown.
reflectivity dipping to the southeast.On Sections (d)and (e),the re-flector appears to have undulating topography.Evidence of a cond reflective band is obrved at the limits of recorded time on inline Section (a).The crossline ctions shown in Fig.6show a coherent reflection at roughly 4.5s TWTT in the northwesternmost ctions (a–c).Moving southeastward (ctions d–f),this reflection begins to dip southwestward,eventually disappearing beyond the temporal extent of the recorded traces.Given the southeastward dip of the reflections along the inline ctions (Fig.5),crossline ctions further southeast than Section (f)show no Winagami-like reflections.Of note,there is evidence for a cond band of reflections at the limits of recorded C 2006The Authors,GJI ,166,155–169
by guest on May 10, 2014
160J.K.W elford and R.M.
Clowes
Figure 6.Crossline ctions through the migrated Pine Creek data volume plotted between 2and 5.108s TWTT (approximate depth of 6–15km).The locations of the ctions with respect to the data cube are shown at the top using the lettered labels.The ctions have not been horizontally or vertically exaggerated.The locations of the reflections attributed to the Winagami quence are shown with the black arrows.The smaller white arrows point to shallower instances of bament reflectivity not interpreted to be part of the Winagami quence.Apart from the geometrical spreading correction,no gain has been applied to the data ctions.On all plots,approximate depths are bad on an average velocity of 6km s −1.Since velocities in the diments are slower,the ba of the diments (which occurs between 2and 2.5s TWTT on the plots)is not as deep as shown.
time on both crossline Sections (a)and (b).If this cond band of reflections corresponds to a cond sill,the temporal proximity of the reflected arrivals to the overlying ones suggests that the two sill sheets are clor spaced than tho imaged to the north by Lithoprobe and that the anastomosing convergence of the sills extends to the ba of the 3-D volume.In Fig.7,three different chair-cut views through the ba of the data volume provide a link between instances of deep reflectivity on the inline,crossline and time ctions.In plot a,the crossline ction (WSW to ENE)is dominated by a horizontal reflection.The interction of that reflection with the inline ction (WNW to ESE)is sugges
tive of a sheet dipping to the southeast.Similar patterns are evidenced in plots b and c.In plots b and c,the time slice shows where the sheet cuts through the bottom of the data volume.The Winagami-like reflections obrved throughout the data vol-ume have geometries that greatly differ from the horizontal reflec-tions from the overlying diments (arriving before 2.5s TWTT)on all of the vertical data plots.Conquently,the later arrivals are likely not multiples but are reflections that originate from within the crys-talline bament.This interpretation is supported by the velocityqq解冻官网
C 2006The Authors,GJI ,166,155–169
by guest on May 10, fordjournals/Downloaded from
