Back to Exploration – 2008 CSPG CSE

Back to Exploration – 2008 CSPG CSEG CWLS Convention 771
Centrifuge Simulations of the Interaction Between Folding,
Faulting and Diapirism During Regional Extension
Elena Konstantinovskaya*
INRS-ETE, Québec, QC
Lyal Harris, Benjamin Carlier, Audrey Lessard-Fontaine, Jimmy Poulin and Adrien Handschuh
INRS-ETE, Québec, QC, Canada
lyal [e-mail]
Eric L. Johnson and Nichola Thomas
Hartwick College, Oneonta, NY, United States
and
Sylvie Daniel
Université Laval, Québec, QC, Canada
Summary
Centrifuge modelling is shown to be a powerful technique for simulating (i) the interaction between
faults and diapirs during rifting, (ii) the geometry of diapir-related folds, including sub-diapir
structures, (iii) active folding associated with displacement on normal faults, and (iv) folds and faults
related to gravitational collapse between lateral ramps in deltas. X-Ray computed tomography (CT
scanning) and gOcad visualization permit the progressive 3D development of structures in
centrifuge models to be studied. In aiding the interpretation of seismic and field data and providing
a means of developing and testing structural hypotheses, centrifuge modelling can provide a
valuable addition to a petroleum exploration program.
Introduction
Through investigating the behaviour of rheologically and dynamically similar systems, physical
(analogue) models provide a better understanding of the progressive development and 3D
geometry of structures in sedimentary basins and fold-thrust belts. Whilst sandbox models replicate
brittle, upper crustal deformation, simulations using a high-acceleration centrifuge are often better
suited to study dynamic systems where body forces (e.g. due to density differences) are important,
or where active folding due to rheological contrasts between layers is significant. A pilot study was
undertaken to develop new modelling materials and CT scanning techniques to simulate the
interaction between diapirs and faults and the 3D geometry of structures in rift basins, deltas and
passive margins through centrifuge analogue modelling.
Back to Exploration – 2008 CSPG CSEG CWLS Convention 772
Method
In diapir models, thin layers and “micro-laminates” of modelling clays and modelling clay-silicone
putty mixes were used to simulate a sequence of inter-bedded sedimentary strata. Less dense/more
ductile silicones simulating mobile salt horizons were placed at different levels in some models. The
rheology and X-Ray properties of these modeling materials were quantified, and the ability to
distinguish individual layers in multilayer packages through X-Ray computed tomography (“CT
scanning”) established. Prescribed cuts in some models represented fault planes of pre-determined
initial dip and /or orientation (e.g. Fig. 1a). Other models comprising layer packages of different
thickness and composition without initial structures were used to test the effects of lithology on
deformation and diapir formation. In models of collapse structures between lateral ramps in deltas,
layers were constructed between fixed wooded walls representing basement-controlled lateral
ramps of differing geometry (Fig. 1b). Upper layers overlaid the entire wooden base. All models
were placed in a rotor with layers vertical (Fig. 1c) and allowed to extend along one axis whilst
undergoing an acceleration of ca. 950 G normal to the initial layering. CT scanning of models was
performed at successive stages of their deformation and final models were serially sliced.
Reconstructions of the 3D geometry of diapir margins, marker horizons, and faults in gOcad from
digitised serial CT scans (e.g. Fig. 2) enabled the progressive development of models to be studied.
Optimum CT scanner configurations and methodologies for scanning silicone-modelling clay layers
were developed.
(a) (b) (c)
Figure 1: (a) Example of multilayered model prior to deformation where prescribed cuts represent faults. (b) Schematic diagram of
lateral ramp models. Basal multilayers of modelling clay, silicone putty and modelling clay-silicone mixes were positioned between
fixed wooden walls representing lateral ramps. Upper layers covered the entire base. (c) Models are positioned in the centrifuge
rotor with layers vertical so that G forces act normal to initial layering.
Examples
(i) Diapir emplacement during extensional faulting. Displacement initially occurred on prescribed
faults. Synthetic and antithetic listric normal faults that soled out on ductile horizons subsequently
developed. Diapirs preferentially intruded normal faults, deforming fault planes and locally reversing
fault dips leading to a termination of displacement on initial faults (Fig. 2). Model layers were folded
producing domal culminations above diapirs and/or in the footwalls of normal faults. Broad synforms
developed between normal faults of opposing dip and between diapirs. In models with a transfer
fault deformation was compartmentalized: diapirs were better developed in the block that underwent
less layer-parallel extension, and diapirs diverged away from the transfer fault and “walls” parallel
but away from transfer faults (Fig. 2). Syn-extensional diapir models display many similarities with
salt-related structures (e.g. Niger delta).
Back to Exploration – 2008 CSPG CSEG CWLS Convention 773
Back to Exploration – 2008 CSPG CSEG CWLS Convention 774

Figure 2: Top: CT scans parallel (left) and perpendicular (right) to the extension direction illustrate the geometry
of silicone diapirs (tops in green and purple, right scan) and folds in adjacent intelayered modelling
clay+silicone/silicone layers, and the deformation of early faults (red and blue). Bottom: gOcad reconstructions
based on serial CT images of diapirs (gold) emplaced during dispacement on normal faults and a transfer fault.
(ii) The geometry of diapir-related folds, including sub-diapir structures. A variety of fold styles
develop associated with diapir emplacement (Fig. 3). Recumbent isoclinal folds develop ajacent to
the diapir trunk beneath the overhanging “tongue” of mushroom-shaped diapirs and tight to isoclinal
upright to inclined synforms form between diapirs. Open domes form above diapirs. Models
showed that layers beneath diapirs were also folded into broad antiforms, with associated parasitic
folds in thinner layers. Viscous layers beneath synforms may thin/neck.
Figure 3: Fold styles in modelling clay+silicone and silicone (pink) layers associated with two
stages in the development diapirs (less dense and more ductile, yellow silicone) during
layer-parallel extension. Broad folds with minor parasitic folds develop beneath diapirs.
(iii) Folds associated with displacement on normal faults. Simple open folds develop in the footwall
to extensional faults (Fig. 4 left). Tighter, more complex folds (Fig. 4 right) develop in footwall layers
with strong mechanical anisotropy when competent dense horizons are displaced. Reverse drag
Back to Exploration – 2008 CSPG CSEG CWLS Convention 775
(upper layers, left), zero drag (lower layers, left) and normal drag (right) may occur in hangingwall
layers depending on layer rheology and density.
Figure 4: Folding of footwall and hanging wall layers during displacement on normal faults. The
model at right comprises thick red and green competent and dense modelling clay layers, and lower
modelling clay+silicone layers with a greater viscosity contrast in comparison to those in the left
model. Ductile, less dense silicone (silver left, dark green right) separates basal and overlying layers.
(iv) Collapse between lateral ramps. A model of structures formed during collapse between
convergent lateral ramps (Fig. 5) simulates structures that may form in deltas where basement
structures localise transfer faults. Faults with normal and strike-slip components formed oblique to
the extension direction in the areas adjacent to the fixed end wall and in the central part of the
model. Folds with axes both perpendicular and parallel to the transport direction developed in the
region closest to the moving end wall (equivalent to the delta toe area in nature) forming elongate
domes and basins that fold isoclinal folds. Figure 5 illustrates the fine details of structures formed in
“microlaminate” layers simulating interlayered sandstone and shales.
Figure 5: Oblique view of upper surface and vertical cut through a model simulating collapse between convergent
lateral ramps showing changes from extensional to contractional deformation similar to that found in some deltas.
Conclusions
Centrifuge modelling is shown to be a powerful technique to simulate a variety of structures in rift
basins, passive margins and deltas that canot be easily simulated in sandbox models, such as
where diapirism and active folding of rheologically stratified sequences are important structural
elements. Whilst results presented only illustrate structures formed in extensional settings,
centrifuge modelling can equally well be applied to structures in contractional settings such as fold-
Back to Exploration – 2008 CSPG CSEG CWLS Convention 776
thrust belts. Model materials, CT scanning techniques, and image analysis software developed in
this study allow fine details of structures to be developed and imaged and for 3D visualizations to be
constructed in gOcad. Non-destructive visualization of models through CT scanning permits the
progressive development of structures and superposed events to be studied. Centrifuge models aid
interpretation of reflection seismic data and the determination of structural histories and deformation
mechanisms in sedimentary basins.
Normal faults and/or fractures in competent horizons are shown to localise diapir intrusion then be
subsequently deformed during diapir ascent. A variety of fold styles is associated with diapirs.
Modelling results have implications for the study of sub-salt structures that represent important
exploration targets in some basins (e.g. Gulf of Mexico) in being ab

ผลลัพธ์ (ไทย) 1:

กลับไปสำรวจ – 2008 CSPG CSEG CWLS ประชุม 771เครื่องหมุนเหวี่ยงจำลองการโต้ตอบระหว่างพับFaulting และ Diapirism ระหว่างภูมิภาคต่อKonstantinovskaya เอเลนา *INRS-ETE, Québec, QCแฮ lyal เบนจามิน Carlier, Audrey Lessard ฟงแตง จิมมี่ Poulin และเอเดรียน HandschuhINRS-ETE, Québec, QC แคนาดาlyal [e-mail] L. Johnson และ Thomas Nichola Hartwick วิทยาลัย นีออน NY สหรัฐอเมริกาและDaniel SylvieUniversité Laval, Québec, QC แคนาดาสรุปเครื่องหมุนเหวี่ยงแบบจำลองแสดงเป็น เทคนิคที่มีประสิทธิภาพสำหรับการจำลอง (i) การโต้ตอบระหว่างdiapirs ระหว่าง rifting, (ii) เรขาคณิตของ diapir เกี่ยวข้องพับ รวม diapir ย่อยและข้อบกพร่องโครงสร้าง พับใช้งานอยู่ (iii) เกี่ยวข้องกับปริมาณกระบอกสูบปกติ บกพร่อง และพับ (iv) และข้อบกพร่องที่เกี่ยวข้องกับความโน้มถ่วงยุบระหว่างทางลาดด้านข้างใน deltas ภาพรังสี (CTการสแกน) และแสดงภาพประกอบเพลง gOcad 3D พัฒนาก้าวหน้าของโครงสร้างในเครื่องหมุนเหวี่ยงรุ่นได้ศึกษา ในการตีความข้อมูลธรณีวิทยา และฟิลด์ที่ช่วยงาน และให้หมายถึงการพัฒนา และทดสอบสมมุติฐานโครงสร้าง การสร้างแบบจำลองเครื่องหมุนเหวี่ยงสามารถให้การนอกจากนี้มีโปรแกรมสำรวจปิโตรเลียมแนะนำผ่านการตรวจสอบพฤติกรรมของระบบแบบไดนามิก และ rheologically คล้าย จริงแบบจำลอง (อนาล็อก) ให้มีความเข้าใจอันดีของการพัฒนาที่ก้าวหน้าและ 3Dgeometry of structures in sedimentary basins and fold-thrust belts. Whilst sandbox models replicatebrittle, upper crustal deformation, simulations using a high-acceleration centrifuge are often bettersuited to study dynamic systems where body forces (e.g. due to density differences) are important,or where active folding due to rheological contrasts between layers is significant. A pilot study wasundertaken to develop new modelling materials and CT scanning techniques to simulate theinteraction between diapirs and faults and the 3D geometry of structures in rift basins, deltas andpassive margins through centrifuge analogue modelling. Back to Exploration – 2008 CSPG CSEG CWLS Convention 772MethodIn diapir models, thin layers and “micro-laminates” of modelling clays and modelling clay-siliconeputty mixes were used to simulate a sequence of inter-bedded sedimentary strata. Less dense/moreductile silicones simulating mobile salt horizons were placed at different levels in some models. Therheology and X-Ray properties of these modeling materials were quantified, and the ability todistinguish individual layers in multilayer packages through X-Ray computed tomography (“CTscanning”) established. Prescribed cuts in some models represented fault planes of pre-determinedinitial dip and /or orientation (e.g. Fig. 1a). Other models comprising layer packages of differentthickness and composition without initial structures were used to test the effects of lithology ondeformation and diapir formation. In models of collapse structures between lateral ramps in deltas,layers were constructed between fixed wooded walls representing basement-controlled lateralramps of differing geometry (Fig. 1b). Upper layers overlaid the entire wooden base. All modelswere placed in a rotor with layers vertical (Fig. 1c) and allowed to extend along one axis whilstundergoing an acceleration of ca. 950 G normal to the initial layering. CT scanning of models wasperformed at successive stages of their deformation and final models were serially sliced.Reconstructions of the 3D geometry of diapir margins, marker horizons, and faults in gOcad fromdigitised serial CT scans (e.g. Fig. 2) enabled the progressive development of models to be studied.Optimum CT scanner configurations and methodologies for scanning silicone-modelling clay layerswere developed. (a) (b) (c)Figure 1: (a) Example of multilayered model prior to deformation where prescribed cuts represent faults. (b) Schematic diagram oflateral ramp models. Basal multilayers of modelling clay, silicone putty and modelling clay-silicone mixes were positioned betweenfixed wooden walls representing lateral ramps. Upper layers covered the entire base. (c) Models are positioned in the centrifugerotor with layers vertical so that G forces act normal to initial layering.Examples(i) Diapir emplacement during extensional faulting. Displacement initially occurred on prescribedfaults. Synthetic and antithetic listric normal faults that soled out on ductile horizons subsequentlydeveloped. Diapirs preferentially intruded normal faults, deforming fault planes and locally reversingfault dips leading to a termination of displacement on initial faults (Fig. 2). Model layers were foldedproducing domal culminations above diapirs and/or in the footwalls of normal faults. Broad synformsdeveloped between normal faults of opposing dip and between diapirs. In models with a transferfault deformation was compartmentalized: diapirs were better developed in the block that underwentless layer-parallel extension, and diapirs diverged away from the transfer fault and “walls” parallelbut away from transfer faults (Fig. 2). Syn-extensional diapir models display many similarities withsalt-related structures (e.g. Niger delta). Back to Exploration – 2008 CSPG CSEG CWLS Convention 773Back to Exploration – 2008 CSPG CSEG CWLS Convention 774Figure 2: Top: CT scans parallel (left) and perpendicular (right) to the extension direction illustrate the geometryof silicone diapirs (tops in green and purple, right scan) and folds in adjacent intelayered modellingclay+silicone/silicone layers, and the deformation of early faults (red and blue). Bottom: gOcad reconstructionsbased on serial CT images of diapirs (gold) emplaced during dispacement on normal faults and a transfer fault.(ii) The geometry of diapir-related folds, including sub-diapir structures. A variety of fold stylesdevelop associated with diapir emplacement (Fig. 3). Recumbent isoclinal folds develop ajacent tothe diapir trunk beneath the overhanging “tongue” of mushroom-shaped diapirs and tight to isoclinalupright to inclined synforms form between diapirs. Open domes form above diapirs. Modelsshowed that layers beneath diapirs were also folded into broad antiforms, with associated parasiticfolds in thinner layers. Viscous layers beneath synforms may thin/neck.Figure 3: Fold styles in modelling clay+silicone and silicone (pink) layers associated with twostages in the development diapirs (less dense and more ductile, yellow silicone) duringlayer-parallel extension. Broad folds with minor parasitic folds develop beneath diapirs.(iii) Folds associated with displacement on normal faults. Simple open folds develop in the footwallto extensional faults (Fig. 4 left). Tighter, more complex folds (Fig. 4 right) develop in footwall layerswith strong mechanical anisotropy when competent dense horizons are displaced. Reverse drag Back to Exploration – 2008 CSPG CSEG CWLS Convention 775(upper layers, left), zero drag (lower layers, left) and normal drag (right) may occur in hangingwalllayers depending on layer rheology and density.Figure 4: Folding of footwall and hanging wall layers during displacement on normal faults. Themodel at right comprises thick red and green competent and dense modelling clay layers, and lower
modelling clay+silicone layers with a greater viscosity contrast in comparison to those in the left
model. Ductile, less dense silicone (silver left, dark green right) separates basal and overlying layers.
(iv) Collapse between lateral ramps. A model of structures formed during collapse between
convergent lateral ramps (Fig. 5) simulates structures that may form in deltas where basement
structures localise transfer faults. Faults with normal and strike-slip components formed oblique to
the extension direction in the areas adjacent to the fixed end wall and in the central part of the
model. Folds with axes both perpendicular and parallel to the transport direction developed in the
region closest to the moving end wall (equivalent to the delta toe area in nature) forming elongate
domes and basins that fold isoclinal folds. Figure 5 illustrates the fine details of structures formed in
“microlaminate” layers simulating interlayered sandstone and shales.
Figure 5: Oblique view of upper surface and vertical cut through a model simulating collapse between convergent
lateral ramps showing changes from extensional to contractional deformation similar to that found in some deltas.
Conclusions
Centrifuge modelling is shown to be a powerful technique to simulate a variety of structures in rift
basins, passive margins and deltas that canot be easily simulated in sandbox models, such as
where diapirism and active folding of rheologically stratified sequences are important structural
elements. Whilst results presented only illustrate structures formed in extensional settings,
centrifuge modelling can equally well be applied to structures in contractional settings such as fold-
Back to Exploration – 2008 CSPG CSEG CWLS Convention 776
thrust belts. Model materials, CT scanning techniques, and image analysis software developed in
this study allow fine details of structures to be developed and imaged and for 3D visualizations to be
constructed in gOcad. Non-destructive visualization of models through CT scanning permits the
progressive development of structures and superposed events to be studied. Centrifuge models aid
interpretation of reflection seismic data and the determination of structural histories and deformation
mechanisms in sedimentary basins.
Normal faults and/or fractures in competent horizons are shown to localise diapir intrusion then be
subsequently deformed during diapir ascent. A variety of fold styles is associated with diapirs.
Modelling results have implications for the study of sub-salt structures that represent important
exploration targets in some basins (e.g. Gulf of Mexico) in being ab

ผลลัพธ์ (ไทย) 3:

กลับไปสำรวจ– 2551 cspg cseg cwls ประชุม 771
centrifuge การจำลองปฏิสัมพันธ์ระหว่างพับและขยายในภูมิภาค diapirism ตลอด

เอเลน่า konstantinovskaya *
inrs-ete คูจาก BEC , QC
ไลเอิล แฮร์ริส , เบนจามิน carlier ออเดรย์ เลสซาร์ด ฟอนเทน จิมมี่ และเอเดรียน พูแลน handschuh
inrs-ete คูจาก BEC , QC , แคนาดา
ไลเอิล _harris @ จ้า . inrs . CA
อีริคจอห์นสันและโทมัส
ว่า Lฮาร์ตวิกวิทยาลัย , นีออน , นิวยอร์ก , สหรัฐอเมริกา


และ Sylvie แดเนียลมหาวิทยาลัยลาวาลคูจาก BEC , QC , แคนาดา

เพื่อสรุปแบบแสดงเป็นเทคนิคที่มีประสิทธิภาพเพื่อจำลอง ( 1 ) ปฏิสัมพันธ์ระหว่าง
ความผิดพลาดและ diapirs ในระหว่าง rifting ( 2 ) รูปทรงของ diapir ที่เกี่ยวข้องเท่า รวมถึง โครงสร้าง diapir
ย่อย ( 3 ) พับใช้งานที่เกี่ยวข้องกับการเกี่ยวกับความผิดปกติและ ( 4 ) พับและข้อบกพร่อง
เกี่ยวข้องกับแรงโน้มถ่วงระหว่างด้านข้างทางลาดในสันดอน . การถ่ายภาพรังสีส่วนตัดอาศัยคอมพิวเตอร์ ( CT สแกนและการ gocad
) อนุญาตให้ก้าวหน้าการพัฒนาโครงสร้างในรูปแบบ 3D
ซึ่งต้องศึกษา ในการตีความของแผ่นดินไหวและเขตข้อมูลและให้
ความหมายของการพัฒนาและการทดสอบสมมติฐานโครงสร้างซึ่งแบบจำลองสามารถให้
เพิ่มคุณค่าโปรแกรมการสำรวจปิโตรเลียม .

ผ่านการตรวจสอบพฤติกรรมของระบบ และแบบไดนามิก rheologically คล้ายคลึงทางกายภาพ ,
( อนาล็อก ) แบบให้ความรู้ความเข้าใจในการพัฒนาความก้าวหน้าและเรขาคณิต 3 มิติของโครงสร้างในแอ่งตะกอน
และพับสายกระตุก รุ่น sandbox ขณะที่เลียนแบบ
เปราะด้านบนของเปลือกโลกการใช้เครื่องจำลองการเร่งสูงมักจะดีกว่า
เหมาะในการศึกษาระบบแบบไดนามิกที่บังคับร่างกาย ( เช่น เนื่องจากความแตกต่างของความหนาแน่น ) มีความสําคัญ ,
หรือที่ปราดเปรียวพับเนื่องจากการความแตกต่างระหว่างชั้นสำคัญ การศึกษานำร่องคือ
) การพัฒนาวัสดุแบบใหม่ และ CT สแกนเทคนิคเพื่อจำลอง
ปฏิสัมพันธ์ระหว่าง diapirs และข้อบกพร่องและเรขาคณิต 3 มิติของโครงสร้างในระแหงแอ่ง deltas และ
ขอบเรื่อยๆผ่านอนาล็อกเพื่อการสร้างแบบจำลอง
กลับไปสำรวจ– 2551 cspg cseg cwls ว่าด้วยวิธีการในรูปแบบ diapir 772

ชั้นบางและ " ไมโครลามิเนต " ของการสร้างแบบจำลองดินเหนียวและดิน
แบบซิลิโคนผงผสมถูกใช้เพื่อจำลองลำดับระหว่างเตียงเป็นชั้น ความหนาแน่นน้อยกว่า / มากกว่า
อ่อนซิลิโคนมือถือจำลองขอบเขตเกลือไว้ในระดับที่แตกต่างกันในบางรุ่น
สมบัติและคุณสมบัติของวัสดุเหล่านี้รังสีเอกซ์แบบปริมาณและความสามารถในการแยกชั้น ในแต่ละชั้น
แพคเกจผ่านการถ่ายภาพสามมิติ (
" กะรัต