Tectonophysics 502 (2011) 336–350

Contents lists available at ScienceDirect

Tectonophysics

journal homepage: www.elsevier.com/locate/tecto

Thrust wedges with décollement levels and syntectonic erosion: A view from analog models

E. Konstantinovskaya a,⁎, J. Malavieille b a Institut national de la recherche scientifique, Centre Eau, Terre et Environnement (INRS-ETE), 490, de la Couronne, Quebec City (QC), Canada G1K 9A9 b Géosciences Montpellier, CNRS UMR 5243, Université Montpellier 2, 34095 Montpellier, Cedex 5, France article info abstract

Article history: Analog sandbox models have been set up to study the impact of syntectonic erosion on thrust wedges with Received 19 August 2010 one and two décollement levels. Different accretion mechanisms are activated depending on interactions Received in revised form 7 January 2011 between surface processes and wedge mechanics: frontal accretion, backthrusting, underthrusting and Accepted 14 January 2011 underplating due to décollement induced duplex formation at depth. These mechanisms may function Available online 15 February 2011 simultaneously, being located at different parts across the wedge. For all the experiments, a high friction is Keywords: imposed at the base of models and the volume of eroded material remains equal to the volume of newly Thrust wedge accreted material, maintaining a constant surface slope during the shortening. Erosion limits the forward Antiformal stack propagation of thrust wedges and favors the underthrusting of basal layers allowing duplex formation. Analog modeling Erosion promotes development of major backthrusts in the thrust wedges without or with one décollement, Erosion but no backthrusts was formed in the wedges with two décollements. Slow erosion allows lower extent of Exhumation basal underthrusting in comparison with regular-rate erosion. Variations in the erosion taper lead to changes Canadian Rockies in duplex geometry and exhumation rate in thrust wedges with one or two décollements. The 6° erosion taper promotes formation of antiformal stack at the rear part of thrust wedge, high rate of basal underthrusting and high extent of erosional removal. The cover layers are nearly completely eroded above the antiformal stack and form the synformal klippe in frontal part of thrust wedges. The 8° erosion taper allows development of individual ramp-anticlines and active forward thrusting of cover layers above the décollement and low rate of basal underplating below it, with consequent low extent of erosional removal. The results of our experiments support the observations on structural evolution and erosion in the Alberta Foothills of the Canadian Rockies. Crown Copyright © 2011 Published by Elsevier B.V. All rights reserved.

1. Introduction Glodny et al., 2005; Gutscher et al., 1996, 1998; Kukowski et al., 2002; Platt, 1988, 1990). Combined with erosion, it allows exhumation of deep Décollements are very common in foreland thrust belts and their rocks in compressional orogens (Avouac, 2003; Bonnet et al., 2007, role has long been studied. They are responsible for duplex formation, 2008; Konstantinovskaia and Malavieille, 2005; Malavieille, 2010; which evolution and geometry vary in styles and greatly influence the Osborn et al., 2006; Simoès et al., 2007). Thus, surface processes play a dynamics of thrust wedges. For examples, antiformal stacks with major role in the growth of aerial thrust wedges influencing their forward thrusting in the cover and significant underthrusting of basal dynamics and long term evolution. Numerous analog modeling studies units are characteristics of the southern Pyrenees in Spain (Vergés and have been devoted to the understanding of these complex interactions Martinez, 1988), spaced ramp anticlines with folding in the cover and at different scales. Some consider thrust development at regional scale small extent of basal underthrusting are typical for the San Andean (Barrier et al., 2002; Casas et al., 2001; Cobbold et al., 1993; Del Castello thrust belt in northern Argentina (Belotti et al., 1995), and duplex styles et al., 2004; Larroque et al., 1995; Leturmy et al., 2000; Malavieille et al., may vary laterally along the same orogenic front as shown in the 1993; Marques and Cobbold, 2002; Merle and Abidi, 1995; Mugnier Canadian Rocky Mountains (Fermor and Price, 1987; Lebel et al., 1996; et al., 1997; Persson and Sokoutis, 2002; Persson et al., 2004; Storti and McMechan, 2001; Price, 1986, 2001; Price and Fermor, 1985; Soule and McClay, 1995; Storti and Poblet, 1997) or at the scale of the orogen Spratt, 1996; Stockmal, 2001). The synchronous play of frontal accretion (Bonnet et al., 2007, 2008; Davy and Cobbold, 1991; Hoth et al., 2006; and underthrusting of duplex units at depth has been shown to be an Konstantinovskaia and Malavieille, 2005; Koons, 1989; Malavieille, important mechanism for the material transfer in thrust wedges (e.g. 2010; Malavieille and Konstantinovskaya, 2010; Persson and Sokoutis, 2002), suggesting that synkinematic erosion and sedimentation influence fault propagation and geometry of thrust wedges favoring ⁎ Corresponding author. Tel.: +1 418 654 2559; fax: +1 418 654 2600. a punctuated thrust activity, alternating frontal thrusting, out-of- E-mail address: [email protected] (E. Konstantinovskaya). sequence thrusting and backthrusting.

0040-1951/$ – see front matter. Crown Copyright © 2011 Published by Elsevier B.V. All rights reserved. doi:10.1016/j.tecto.2011.01.020 E. Konstantinovskaya, J. Malavieille / Tectonophysics 502 (2011) 336–350 337

Analog modeling approaches have been used for a long time in 2. Experimental method complement with geological studies to analyze the effects of frictional and ductile décollements on duplex styles. Presence of frictional (glass We analyze the effects of syntectonic erosion on fault propagation, microbeads) décollements in the accreted series of purely brittle duplex geometry, material transfer and exhumation in thrust wedges thrust wedge models allows underplating of thrust units developing using 2D sandbox experiments. This study focusing on the role of an antiformal stack, whose growth and location is favored by erosion décollements is complementary of an earlier work of Konstantinovskaia (Bonnet et al., 2007, 2008; Konstantinovskaia and Malavieille, 2005; and Malavieille (2005), in which the style of fault propagation and the Malavieille, 2010). Presence of ductile décollements affects the diversity of exhumation patterns were studied in eroded thrust wedges growth and evolution of sand–silicone experimental wedges. It was as a function of type of basal friction (high and low), thickness of shown that internal deformation style and forward propagation of accreted series and the presence of subduction window. In our previous structures in brittle–ductile wedge models are strongly dependent work it was shown that the uplift of material occurs along a cluster of upon the brittle–ductile coupling (Bonini, 2003, 2007; Costa and subvertical thrusts in the middle part of the eroded thrust wedge with Vendeville, 2002; Mugnier et al., 1997; Smit et al., 2003). Strong low basal friction. The material is exhumed along a series of inclined décollements in sand–silicone wedges with two décollements (20°–50°) thrusts in the rear of the high-friction wedge. The vertical (Couzens-Schultz et al., 2003) favor local underthrusting of the component of exhumation is generally higher for the wedges with high cover, development of individual ramp-anticlines, internal deforma- basal friction than for low-friction wedges, and it is amplified by the tion of thrust sheets and low layer parallel shortening, whereas weak presence of décollements. The addition of décollement layer in a sand décollements enable forward thrusting of the cover, antiformal stacks, pack requires high basal friction in order to create a difference in rate of coeval growth of structures, low internal strain and lower layer lateral material transfer across the growing thrust wedge. Thus, only parallel shortening that occur later. Otherwise, very weak silicone high basal friction wedges are discussed here. The effect of slower rate of décollements may produce a localization of deformation at long lived erosion on fault kinematics and exhumation in a thrust wedge is newly detachment folds above a floor thrust tip (Bonini, 2003). It was tested in the present study (MW3) in order to compare it to the noticed that backthrusts (hinterland verging) and forethrusts (fore- previously obtained model wedge (MW2) with regular erosion land verging) may develop in model thrust wedges depending on the (Table 1). The former experiment with a single décollement in model relative strengths of the basal décollement and overlying cover and on wedge is reproduced in the present study (MW5) to be compared to basal friction (Bonini, 2007; Chapple, 1978; Davis and Engelder, 1985; new experiments of model wedges with two décollements (MW6–7). Mandl and Shippam, 1981). The basic device (Fig. 1) is made by a flat basal plate bound by two The impact of sedimentation on thrust wedge geometry and fault lateral glass walls (see detail in Bonnet et al., 2007, 2008; kinematics was studied by Bonnet et al. (2007), Konstantinovskaya et al. Konstantinovskaia and Malavieille, 2005; Malavieille, 1984). To (2009), Mugnier et al. (1997), Smit et al. (2010), Storti and McClay reduce the amount of sidewall friction, a lubrication of glass walls was (1995), and Storti et al. (2000). Diffuse sedimentation results in large done before sand deposition. The deformation box is 200 cm forethrust spacing and a more diffuse deformation style (Selzer et al., (length)×10 cm (width) providing 150 cm of total shortening. A 2007; Simpson, 2006; Storti and McClay, 1995). High-rate syntectonic stepper motor moves a basal plastic sheet that is pulled beneath a sedimentation towards the front of the wedge also lowers the taper vertical rigid backstop at one side of the box. The rough surface of the angle and focuses deformation towards the rear of the wedge (Simpson, plastic sheet allows simulation of a high basal friction at the base of the 2006; Storti and McClay, 1995). The addition of syntectonic sediments layered incoming sand. The stable backstop represents the upper plate during the evolution of sand–mica thrust wedges result in increase of against which the thrust wedge develops. The analog granular materials the retrovergent thrusting stage that occurred along a long-lived ramp have frictional properties satisfying the Coulomb theory and they whose lower tip was located at the subduction slot (Storti et al., 2000). In correctly mimic non-linear deformation behavior of crustal rocks in the contrast, syntectonic sedimentation in foreland areas of thrust wedges brittle field (Dahlen, 1984; Dahlen et al., 1984; Lohrmann et al., 2003). with décollements favors the activation of a weak décollement layer at The Aeolian sand used in the experiments is rounded with a grain size of the base of a cover sequence, leading to the development of a piggyback less than 300 mm and a density of 1690 kg/m3. The internal coefficient basin (Konstantinovskaya et al., 2009; Mugnier et al., 1997). of friction is 0.57 and the cohesion Co=20 Pa. A basal friction Erosion restrains the forward propagation of thrust wedge, induces corresponding to these parameters is around 24° for the high basal development of major back thrusts, enhances material transport across friction models. Successive colored sand layers are put on the plastic the wedge, and focuses the internal deformation and exhumation close sheet simulating sedimentary sequences. The initial thickness of sand to the backstop (Bonnet et al., 2007; Hoth et al., 2006; Konstantinovskaia layers is 3.6 cm. A proto-wedge is built before shortening to rapidly and Malavieille, 2005; Selzer et al., 2007; Simpson, 2006; Willett et al., obtain a thrust wedge at critical taper. It is 10.4 cm high and its slope is 1993). The application of both syntectonic erosion and sedimentation 15°. The passive marker particles composed of colored sand were (Konstantinovskaya et al., 2009; Mugnier et al., 1997) results in distributed at each 5 cm along the basal layer of model wedges. Tracing development of “passive-roof duplex” (or triangle zones) in thrust of displacement paths of the marker particles from serial experimental wedges with décollements. In thrust belts formed by a succession of photos promotes reconstruction of material transfer through the thrust ramp anticlines, exhumation is induced by erosion that mainly wedge during continuous shortening. postdates the tectonics. In a passive-roof duplex, the exhumation is The weak décollement levels are created by introducing thin mainly controlled by the erosion that continuously balances thrust (1–2mm)layersofglassmicrobeadsatdifferentlevelsofthesandcake. tectonics. A series of factors (relative strength of a décollement horizon, They are a Coulomb material and their density and size are almost the strain rate and basal friction) leading to the development of backthrusts, same as those of dry sand, however due to their close to perfect passive roof duplexes and triangle zones has been widely investigated roundness their coefficient of internal friction is about 23% smaller (0.44), by physical modeling approaches (Bonini, 2007; Couzens-Schultz et al., with cohesion almost negligible. The successive colored sand layers are 2003; Gutscher et al., 1998; Koyi et al., 2000; Mulugeta and Koyi, 1987). accreted in front of the proto-wedge developing a Coulomb thrust wedge Our modeling study focuses on the impact of frictional décollements during convergence. Scaling factor is 105:1cminexperimentisroughly on duplex formation, subsequent underthrusting (high basal coupling) equivalent to 1 km in nature. Scaling, and characterization of models and and underplating (different decoupling levels acting at different depths analog materials used in sand-box modeling are discussed in Dahlen within the wedge) and material transfer in eroded sand thrust wedges. (1984), Dahlen et al. (1984), Davy and Cobbold (1991), Gutscher et al. Results are then applied to better understand complex structures (1996, 1998), Kukowski et al. (2002), Lallemand et al. (1994),and observed in natural example of the Canadian Rocky Mountains. Lohrmann et al. (2003) with a synthesis given in Graveleau (2008). 338 E. Konstantinovskaya, J. Malavieille / Tectonophysics 502 (2011) 336–350

Table 1 Initial setup characteristics, rates of erosion, duplex types, extent of basal underthrusting and erosional removal in the experimental thrust wedges.

Thrust Erosion taper Erosion rate, cm Number of Presence Presence of duplex Exhumation of Extent of erosional Ratio of basal Total a a wedge angle, ° of shortening décollements of major and its type basal material removal ,Eeros,% underthrusting , shortening,

backthrust Rund,% cm MW1 No No No Yes No No 0 6 105 MW2 8 2 No Yes No Yes 35 38 125 MW3 8 4 No Yes No Yes 28 33 119 MW4 6 2 1 Yes Antiformal stack Yes 26 25 115 MW5 6 2 2 No Antiformal stack No 23 30 86 MW6 8 2 3 No Upper ramp-anticlines No 18 16 86 and lower normal duplex

Eeros =Serod /(Sinit +Sinp)×100%; Serod =Sinit +Sinp −Sfinal;Rund =Sund/Sfinal ×100%, where Sinit, initial wedge area; Sinp, total area added as “input” at the front (or accreted at the base) of the wedge; Sfinal, thrust wedge area at the end of shortening; Serod, area of eroded material, Rund, ratio of basal underthrusting, Sund, area of basal underthrusting; Sinit,Sfinal and Sund are measured from digitized photos of model wedges, Sinp is calculated from experiment parameters. a Values are taken at the end of shortening.

Erosion is performed by hand with a thin metal plate (the sand experiment with slow erosion (MW3) could be compared with natural being removed using a vacuum cleaner) to maintain the slope of the cases in which a higher resistance of composing rocks and/or wedge at a constant angle reflecting the mean taper angle imposed by unfavorable climatic conditions occurred. Three last experiments were wedge mechanics (Bonnet et al., 2007, 2008; Konstantinovskaia and run to study effects of one (MW4) or two (MW5–6) décollements in a Malavieille, 2005). The erosion surface was projected on the glass wall thrust wedge under erosion. The erosion taper angle was 6° for the after 20 cm of shortening in the way that the profile runs along the thrust wedges with one (MW4) and two décollements (MW5) to reflect thrust wedge and cuts X axis at its toe. Erosion of the units was applied low basal friction at the base of the cover layers above the décollements in a constant manner, independently of their compositional nature, as (Konstantinovskaia and Malavieille, 2005). The imposed slope of the a function only of topography. Erosion is made gradually each 2 cm of erosion surface was 8° in the experiment MW6 with two décollements. shortening. Between erosion steps, the wedge is allowed to try to The variation of boarding conditions in the six experiments (Table 1) obtain its own critical taper. Thus higher topographies and topo- was set to study their effect on thrust wedge kinematics, fault graphic anomalies were eroded leading to erosion that is distributed propagation, underthrusting and exhumation rates. The experiments and linearly dependant on elevation. Generally this means that were repeatedly run to ensure that similar deformation is reproduced erosion is increasing towards higher topographies. This is supported under the same setup and that it was the changes in the initial setup that by other analog models (e.g., Cruz et al., 2008; Hoth et al., 2006), and affected variations in model wedge evolution. The total amount of also by observations from natural situations where erosion can be shortening in the presented experimental runs varies about 105±20 cm positively correlated with elevation (e.g., Summerfield and Hulton, (Table 1). The moment to stop the experiments was chosen once internal 1994). The results of models are useful to discuss the effects of several wedge kinematics reached equilibrium and total length of sand layers first order mechanical parameters on the deformation and structural was accreted. The average value of total shortening 105 cm characterizes evolution of orogenic wedges submitted to erosion. the thrust wedge without erosion (MW0). The most important total There is no sedimentation applied in the experiments of the shortening (125 cm) is observed in eroded model wedge without present study. This setting likely corresponds to the orogen wedges décollements (MW2) that is slightly higher than the one (120 cm) in the where eroded material is transported further away from the frontal wedge with slower erosion (MW3). The eroded wedges with one and thrusts toward the foreland basin. In natural example of Canadian two décollements (MW4–6) demonstrate the lowest values of total Rockies compared to our experiments, the material eroded from the shortening that is 115 and 86 cm, respectively (Table 1). Mesozoic thrust wedge of the Alberta Foothills is transported to the east and deposited in the Alberta Planes. 3. Results The first model (MW1) was set with no décollements, and no erosion was applied through the experiment to establish a critical accreting 3.1. Experiments without and with erosion taper angle (8°) in the models with high basal friction (Table 1). Two subsequent experiments were run to test the effects of erosion on the 3.1.1. No erosion thrust wedges without décollement (MW2–3, Table 1). The imposed The model wedge MW1 (Fig. 2) is growing by continuous frontal slope of the erosion surface was 8° determined from the first accretion and by slight basal underthrusting under the rear part of the experiment. The erosion rate was two times slower in the experiment wedge. Forward propagation of thrust faults leads to constant lateral MW3 in comparison to MW2 that was simulated by erosion and sand growth of the model wedge. Frontal thrusts are relatively flat (20–25°) removal each 4 cm of shortening (MW3) instead of 2 cm (MW2). The and become stepper (35–40°) with time being transferred and

Fig. 1. Scheme of the experimental device. The initial thickness of sand layers is 3.6 cm. The height of the proto-wedge is 10.4 cm and its slope is 15°. The plastic sheet of rough surface simulates high basal friction in the models. The angle of erosion slope α constitutes 8° (critical taper) and 6° in different experiments. The cross sections presented in this study (Figs. 2–7) are lateral photographs taken through the sidewall glasses. E. Konstantinovskaya, J. Malavieille / Tectonophysics 502 (2011) 336–350 339

Fig. 2. Stages of deformation of the thrust wedge MW1, high basal friction, no erosion, no décollements: photos of the model (a–d) and interpretation of the wedge structure at the end of shortening (e). The thrust wedge grows laterally and thickens through the shortening (b–d). No exhumation is observed. The wedge slope 8° (e) represents the critical accreting taper angle for the thrust wedges with high basal friction. deformed toward the rear part of the wedge. A major backthrust (31– 1 and 2, Fig. 3), while the later accreted basal material is exhumed 38°) displaces the units of thrust wedge over the toe of the proto-wedge. (particle 3, Fig. 3). Exhumation occurs along a series of steeply (50– The accreted basal material is never exhumed remaining at the low 64°) inclined forward thrust faults at the rear of the wedge (Fig. 3e). structural levels of the wedge (particles 1 and 2, Fig. 2). The shallow seated backthrusts (44–72°) cut the frontal part of the The measured surface slope of the non-eroded thrust wedge MW1 proto-wedge at the back side of the exhumation area to accommodate is 8° that is used as a critical taper angle of erosion for other model the backward displacement of the growing thrust wedge (Fig. 3b–e). wedges with high basal friction. 3.1.3. Slow erosion (material removal each 4 cm of shortening) 3.1.2. Regular erosion (material removal each 2 cm of shortening) The thrust wedge MW3 (Fig. 4) is restrained to the constant The geometry of model wedge MW2 (length, thickness and slope geometry with the beginning of erosion (8°), similar to the model profile) remains constant (Fig. 3) once a critical erosion taper (8°) is wedge MW2. Frontal thrusts (9–28°) become steeper (55–62°) in the imposed. New material is accreted along frontal thrusts (13–27°) and central part of the wedge being inclined by rotation or by a series of transferred to the inner basal part of the wedge by underthrusting of paired backthrusts (25–34°). The backthrusts that develop at the rear accreted units. Forward thrusts steepen (30–38°) with depth and part of the wedge have a lower angle dip (40–44°) in comparison to toward the rear part of the wedge. Smaller backthrusts (30–38°) are the similar backthrusts of the model wedge MW2. The accreted associated with frontal forward thrusts. Through shortening, upper material is exhumed mostly in the central part of the wedge (particle sand layers are eroded and basal layers are piled up at the rear part of 3, Fig. 4) along steep (55–62°) forward thrusts. the wedge where the domain of maximum exhumation is located The exhumation area has essentially the same width at the surface (Fig. 3). The early accreted basal material is uplifted to the high in the experiments MW2 and MW3 but its location is different: it is structural levels of the wedge but does not reach the surface (particles restricted to the rear of the wedge MW2 (Fig. 3e) and found in the 340 E. Konstantinovskaya, J. Malavieille / Tectonophysics 502 (2011) 336–350

Fig. 3. Stages of deformation of the thrust wedge MW2, the imposed erosion slope 8°, no décollements: photos (a–d) and interpretation of the wedge structure at the end of shortening (e). The thrust wedge retains the same geometry through shortening (b–d). Exhumation of basal layers is observed at the rear part of the wedge at the end of shortening. middle of the wedge MW3 (Fig. 4d). The different location of below the décollement and by frontal accretion of the cover layers exhumation is likely related to development of larger area backthrusts above it. At the beginning of shortening, the lower sand layers of the in MW3 that demonstrate shallower dip (40–44°) than in the thrust wedge accrete to form a duplex structure under the décolle- backthrusts in MW2 (44–72°). The more extensive backthrust ment in the rear part of the wedge (Fig. 5b). The cover sand layers development at the rear of the wedge MW3 is likely favored by above the décollement are accreted along the forward propagating lesser rate of erosion that provided more time to the wedge to reach thrusts (FT1–FTn) to form an upper thrust wedge. The forward thrusts critical taper between each step of erosion. (16–20°) are associated with paired backthrusts in the upper wedge, typical of thrust wedges with low basal friction (Konstantinovskaia 3.2. Experiments with erosion and presence of one and two décollements and Malavieille, 2005). Continuous erosion (6°) leads to progressive growing of duplex The thrust wedges MW4 with one décollement (Fig. 5) and MW5 under the décollement to form an antiformal stack in the thrust with two décollements (Fig. 6) were submitted to erosion along wedge MW4 (Fig. 5c–d). The inclined forward thrusts (20–50°) in the profile of 6° to reflect low friction along décollements at the base of duplex (Fig. 5c) with progressive shortening (Fig. 5d) become steeper cover layers. The erosion in MW4–6 is applied each 2 cm of shortening (60–90°) and even get a negative plunge (−15°) being arched as in MW2. upward above the piled up basal tectonic slices. Both the early and the later accreted basal material were brought to the surface in the 3.2.1. Erosion 6° (one décollement) domain of major exhumation at the rear part of the thrust wedge The thrust wedge MW4 (Fig. 5) with one low friction décollement (particles 1–3, Fig. 5d). The upper thrust wedge of cover layers creeps (thin layer of glass microbeads) grows both by basal underthrusting along the décollement (Fig. 5c) being finally compressed in a E. Konstantinovskaya, J. Malavieille / Tectonophysics 502 (2011) 336–350 341

Fig. 4. Stages of deformation of the thrust wedge MW3, the imposed erosion slope 8°, no décollements: photos (left) and their interpretation (right). The erosion is two times slower than in the experiment MW2 (Fig. 3). Exhumation of basal layers is observed at the middle part of the wedge at the end of shortening.

synformal klippe in front of the growing antiformal stack (Fig. 5d). MW5 (Fig. 6c–d). The cover layers of the upper duplex (between two New frontal thrust Fn′ that develops in front of the synformal klippe is décollements) are nearly completely eroded above the growing lower merged to the décollement at the base of the cover sand layers duplex at the end of experiment (Fig. 6d). The upper thrust wedge (Fig. 5d). The major backthrust (40°) grows at the rear of the wedge at composed of the cover sand layers above lower décollement is the final stages of shortening, contributing to further basal material characterized by a series of paired forward and back thrusts (32°). The exhumation (Fig. 5e). upper wedge is compressed in the synformal klippe (Fig. 6d) at the final stages of shortening. No major backthrust occurs during the thrust wedge formation. 3.2.2. Erosion 6° (two décollements) The thrust wedge MW5 with two décollements (Fig. 6) is run with the imposed erosion taper (6°), similar to the model wedge MW4. The 3.2.3. Erosion 8° (two décollements) presence of two décollements in the wedge MW5 changes the fault The thrust wedge MW6 with 2 décollements (Fig. 7)was propagation inducing the development of independent system of submitted to erosion along the 8° slope taper that is a critical taper thrusts and pop-up structures (PU1 and PU2) above each décollement angle for high basal friction wedges. The higher-angle erosion taper (Fig. 6b, red square) with underplating of material at different applied to this thrust wedge leads to changes in fault propagation, structural levels. The antiformal stack (Fig. 6c, d) is formed under material transfer and exhumation if compared to other models with the lower décollement at the rear part of the thrust wedge, but it décollements (MW4 and MW5). remains wider and lower than the domal structure in the model MW4 The small-scale duplex is formed by underthrusting of basal (Fig. 5c–d; Video 1). The upward transfer of basal material within the material under the lower décollement at the rear of the wedge, in the lower duplex occurs along more flat (35–40°) thrusts comparing to domain of maximum exhumation (Fig. 7c–d, Video 2). It composes the one-décollement wedge MW4. Neither early, nor later accreted only about 1/3 of the wedge thickness at the end of experiment basal material reaches the surface being uplifted to the medium depth (Fig. 7d). Paired forward and back thrusts develop in the cover layers level of the wedge (particles 1 and 2, Fig. 6d). The area of maximum being frequently independent below and above the upper décolle- exhumation is located at the rear of the thrust wedge (Fig. 6d). The ment (Fig. 7b, red square). With continued shortening, the cover upper forward thrusts are very shallow dipping (3°) or even obtain a layers above the lower décollement are accreted along a series of negative plunge (−12°) at the upper structural level of the wedge low-angle forward thrusts, which form individual ramp-anticlines 342 E. Konstantinovskaya, J. Malavieille / Tectonophysics 502 (2011) 336–350

Fig. 5. Stages of deformation of the thrust wedge MW4, the imposed erosion slope 6°, one décollement (white glass microbeads) is introduced at the base of the model: photos (left) and their interpretation (right). Exhumation of basal layers is observed in the dome-like antiformal stack at the rear of the wedge at the end of shortening (d). Thrust faults steeply plunge down the section at the frontal part of the growing antiformal stack (c–d). The cover layers above décollement are completely detached from the basal layers and compressed in synformal klippe (c–e).

(Fig. 7c–d). The frontal thrusts (32°) become steeper (43°) when the upper décollement being eroded (Fig. 7d). No major backthrust migrating toward the center part of the wedge with continuous occurs during the thrust wedge formation. shortening. At the rear part of the model, the thrust faults in the cover layers become flat or get negative plunge (−5°) being curved around 3.3. Marker particle displacement paths the growing lower duplex (Fig. 7d). The basal material is transferred through the lower duplex but it is locked under the décollement and The marker particles at the base of accreted sand layers were dis- never exhumed (particle 1, Fig. 7d). The higher duplex between two placed through thrust wedges with continuous shortening. The displace- décollements is not eroded in this model, only the cover layers above ment paths of marker particles were traced for each steps of shortening. E. Konstantinovskaya, J. Malavieille / Tectonophysics 502 (2011) 336–350 343

Fig. 6. Stages of deformation of the thrust wedge MW5, the imposed erosion slope 6°, two décollements (white glass microbeads) are introduced in the model: photos (left) and their interpretation (right). The independent system of thrusts develops above each décollement (red square). The basal layers below the lower décollement form antiformal stack growing through shortening (b–d) but they are not exhumed (d). The cover layers above the lower décollement are nearly completely eroded above the stack (d). Thrust faults steeply plunge down the section at the frontal part of the growing antiformal stack (d). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

The thrust wedge MW1 without erosion is characterized by very along 6° profile. All marker particles accreted at the base of the thrust slight basal underthrusting. The marker particles (Fig. 2e) are wedge are exhumed at the rear part of the wedge, in the area of active transferred along the base of the wedge to its rear part and then they growing of the antiformal stack (Fig. 5c–d). are slightly uplifted mostly by movement along the major backstop. The thrust wedges MW5 and MW6 with two décollements are The thrust wedge MW2 with regular erosion at critical taper (8°) characterized by transfer of marker particles throughout the basal demonstrates active material uplift and exhumation with shortening. duplexes but they stay locked under the lower décollements and The basal material accreted during the first stages of shortening is never reach the surface (Figs. 6d and 7d). The higher vertical uplift of uplifted along the steep forward faults in front of the proto-wedge, marker particles is observed for the model wedge MW5 (erosion 6°), but it does not reach the surface (particles 1 and 2, Fig. 3e). The basal within which the antiformal stack was formed at the rear of the wedge material accreted after the onset of erosion is exhumed along a series (Fig. 6d). The basal material is uplifted very close to the surface in this of inclined forward thrusts at the rear of the wedge, in the domain of model with upper duplex being nearly completely eroded. The model maximum exhumation (particle 3, Fig. 3e). wedge MW6 (erosion 8°) is characterized by formation of small-scale The thrust wedge MW3 with slow erosion along critical taper profile basal duplex (Fig. 7d) and vertical transfer of basal material is limited (8°) has the similar exhumation paths of the marker particles (Fig. 4d) if by about 1/3 of the wedge thickness. compared to the wedge MW2 (Fig. 3e). But the domain of maximum exhumation is stabilized in the central part of the thrust wedge. 3.4. Extent of erosional removal and ratio of basal underthrusting The thrust wedge MW4 with one décollement demonstrates the most pronounced exhumation of basal material in comparison to all As follows from the previous sections, erosion and variations of other experiments. Exhumation is started with the onset of erosion critical taper angle affect the fault propagation and duplex geometry 344 E. Konstantinovskaya, J. Malavieille / Tectonophysics 502 (2011) 336–350

Fig. 7. Stages of deformation of the thrust wedge MW6, the imposed erosion slope 8°, two décollements (white glass microbeads) are introduced in the model: photos (left) and their interpretation (right). Thrusts are frequently independent below and above the upper décollement (red square). The basal layers below the lower décollement form small-scale normal duplex slightly growing through shortening (b–d). Basal material is never exhumed (d). The cover layers form individual ramp-anticlines (c–d). Thrust faults are slightly inclined down the section at the frontal part of the growing basal duplex (d). Only cover layers above the upper décollement are eroded above the duplex (d). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

in the model thrust wedges. To quantify the effect of these parameters into account the displacement of passive marker particles of colored on basal underplating and exhumation in the thrust wedges, we sand distributed at each 5 cm along the basal layer. calculated extent of erosional removal (Eeros) and ratio of basal The extent of erosional removal (Eeros) is estimated as follows: underthrusting (Rund)(Table 1). To estimate the extent of erosional removal (Eeros), we identified Eeros = Serod Sinit + Sinp ×100%; the following areas (Fig. 8a): (1) S , initial wedge area; (2) S , total init inp . area added as “input” at the front (or accreted at the base) of the where the area of eroded material (Serod) is calculated as follows: wedge; (3) Sfinal, thrust wedge area at the end of shortening; and (4) Serod, area of eroded material. Sinit and Sfinal are measured from Serod = Sinit + Sinp–Sfinal: digitized photos of the model wedges. Sinit is equal the sum of the area of proto-wedge and of the undeformed sand layers. The length of fl undeformed sand layers corresponding to Sinit is taken at the tip of the The ratio of basal underthrusting (Rund) re ects the part of fi final thrust wedge. Sinp is calculated from experimental parameters accreted basal material with respect to the nal area of eroded wedge being equal the product of thickness (3.6 cm) multiplied by length of (Fig. 8b). The ratio is calculated as follows: the accreted sand layers. The length of accreted layers is calculated by analyzing the serial experimental photos of model evolution taking Rund = Sund = Sfinal ×100%; E. Konstantinovskaya, J. Malavieille / Tectonophysics 502 (2011) 336–350 345

fi Fig. 8. (a) Proportions of initial (Sinit), eroded (Serod), and nal (Sfinal) areas in model wedges used to calculate extent of erosional removal (Eeros). Serod =Sinit +Sinp −Sfinal. (b) Ratio of basal underplating (Rund) calculated as a ratio of the area of basal layers (Sund) with respect to final area (Sfinal) in model wedges. Sinit,Sfinal and Sund are measured from digitized photos of the model wedges. Sinp is calculated from experimental parameters. Critical taper angle of our model wedges is equal to the imposed erosional slope (α′) because basal detachment is horizontal. (c) Critical taper angle (α+β) composed of the dip angle of basal detachment β and the taper angle α determines the geometry of a thrust wedge, modified after Davis et al. (1983).

where the area of basal underthrusting (Sund) is measured from 4. Discussion digitized photos of model wedges. It is defined as the area composed of basal layer material within eroded wedge. The basal layer is detected 4.1. Structural evolution of model thrust wedges due to its specific color or it is situated below the lower décollement (Fig. 8b). Two main tectonic processes account for the growth of thrust The extent of erosional removal (Fig. 9a, Table 1) increases with wedges without erosion: frontal accretion, which is more pronounced shortening for all eroded thrust wedges. This parameter is higher in for the low basal friction wedges, and underthrusting that char- the thrust wedges without décollements (MW2–3) than in the thrust acterizes the high basal friction wedges (e.g. Lallemand et al., 1994; wedges (MW4–6) with décollements. The highest quantity (35%) of Malavieille, 2010). The last case is well illustrated by the experiment removed material occurred in the thrust wedge MW2 that was eroded without erosion, where the thrust wedge MW1 grows constantly by regularly (each 2 cm of shortening), higher than Eeros (28%) in the underthrusting of tectonic slices (Fig. 2). Repeatedly during the wedge with slow erosion (each 4 cm of shortening). Between the shortening, structural thickening of the wedge leads to increase of thrust wedges with décollements, the higher erosional removal was shear stress at its base, and deformation propagates forward with occurred in the wedges MW4–5 (with one and two décollements) initiation of a new frontal thrust. The major backthrust at the rear of eroded at the 6° erosion taper (Fig. 9a). The thrust wedge MW6 (with the wedge and internal diffuse deformation helps to reach the critical 2 décollements) eroded along the 8° slope has the smallest extent of taper. erosional removal between all eroded thrust wedges. Thrust wedges submitted to constant erosion never reach the The ratio of basal underthrusting (Fig. 9b) is the highest in the critical equilibrium in their internal part because wedge thickness eroded thrust wedges without décollements (MW2–3) with lower remains constant and the deformation cannot propagate toward the value of Rund in the wedge submitted to slow erosion (MW3). The foreland (Mugnier et al., 1997). It is confirmed by our experiments, ratio of basal underthrusting is lower in the thrust wedges with one or where the eroded thrust wedges MW2–6 do not grow laterally two décollements (MW4–5) than in the wedges without décolle- preserving their constant geometry (slope, thickness and length) ments (MW2–3). The thrust wedge with one décollement (MW4) is through progressive shortening (Figs. 3–7). characterized by stabilization of Rund at the end of shortening (Fig. 9b) It was observed that slower erosion (MW3) leads to lesser extent because the basal material reached the surface and began to be eroded of erosional removal and lesser extent of basal underthrusting (Fig. 9) at these stages (Fig. 5d). Between all eroded model wedges, the in comparison with the experiment under regular erosion (MW2). smallest basal underthrusting occurred in the thrust wedge with two Correspondingly, the increase of erosional removal promotes further décollements (MW6) and eroded along the 8° slope (Fig. 9b). increase of basal underthrusting. This conclusion supports the The thrust wedge without erosion (MW1) has no erosional removal idea of a strong feedback connection between surface processes and of material and it is characterized by the lowest vertical transfer through internal wedge dynamics (Avouac, 2003; Bonnet et al., 2007, 2008; the model wedge (Fig. 2e) and the smallest ratio of basal material Konstantinovskaia and Malavieille, 2005; Malavieille, 2010; Osborn accretion (Fig. 9) in comparison with all other experiments. et al., 2006; Simoès et al., 2007). Variations in ratio of basal underthrusting are in positive linear A combination of low angle (6°) erosion and high basal friction in correlation with the extent of erosion removal in the examined thrust the thrust wedges with décollements (MW4–5) favors the develop- wedges (Fig. 9c). When more material is removed by erosion from the ment of highly exhumed antiformal stacks (Figs. 5 and 6). The wedges surface, then more material is accreted at the base of model wedges. are in constant increase of structural thickening by basal underplating 346 E. Konstantinovskaya, J. Malavieille / Tectonophysics 502 (2011) 336–350

Fig. 9. Diagrams showing the variations of the extent of erosional removal Eeros (a) and ratio of basal underthrusting Rund (b) in thrust wedges with continuous shortening given for each step of deformation as a percentage of total shortening reached in each experiment (Figs. 2–7, Table 1). Correlation between parameters Eeros and Rund (c). Eeros and Rund are higher in the eroded model wedges MW2–3 without décollements if compared to the eroded wedges MW4–6 with décollements (a–b). Only very slight basal underthrusting occurs in the model wedge MW1 without erosion (b). The model wedges MW4–5 eroded along 6° profile (antiformal stacks) are characterized by higher basal underplating than the model wedge MW6 eroded along 8° profile (small scale normal basal duplex). Higher erosion removal promotes higher basal underplating (c). in order to reach the critical taper (8°) required by high basal friction. high basal friction in thrust wedges eroded along the equilibrium The tectonic slices within the stacks are piled up and arched above critical slope (8°) favors small thrust slip and development of normal each other (Figs. 5e and 6d). Antiformal duplexes are interpreted to be duplexes while low basal friction under the same circumstances could formed when the amount of displacement along a thrust fault is produce forward-dipping duplexes. essentially equal to the length of the next lower thrust slice (McClay, Thus, the experiments with model wedges constructed with 1992). Such a relationship seems to characterize the development of décollements (MW4–6) provide the evidence that the erosion slope antiformal stacks in the model wedges MW4–5. may influence kinematics of fault propagation and duplex geometry. Erosion along the critical slope (8°) applied to the thrust wedge Different accretion mechanisms (Fig. 5) are then activated depending with décollements (MW6) promotes forward propagation of thrusts on interactions between surface processes and wedge mechanics: in the cover layers located between and above two décollements and frontal accretion, duplexing and underplating, and backward thrust- favors the formation of individual ramp-anticlines (Fig. 7). Being in ing. These mechanisms may function simultaneously, being located at equilibrium between the basal friction and the erosion slope, the different parts across the wedge (Fig. 5e). wedge does not need to grow vertically. Thus, less structural thickening occurs under the lower décollement, where the normal 4.2. Natural example: Southern Foothills of the Canadian Rocky Mountains duplex is growing mostly laterally (Fig. 7c–d). Tectonic slices in the duplex dip gently backward, in the direction opposite to the overall The multiple detachments and duplexes of different structural thrust transport. Length of tectonic slices in normal duplexes is known styles are well known in the southern Foothills of the Canadian to be greater than the slip along thrust faults in contrast to forward- Rockies (Langenberg et al., 2006; Lebel et al., 1996; McMechan, 2001, dipping duplexes where the slice length is lesser than the displace- 2002; Price, 1986, 2001; Price and Fermor, 1985; Price and Monger, ment along thrust faults (McClay, 1992). It may be inferred that the 2000; Soule and Spratt, 1996; Stockmal, 2001). E. Konstantinovskaya, J. Malavieille / Tectonophysics 502 (2011) 336–350 347

In the southern Alberta Foothills, the supracrustal wedge consists The critical taper angle of this foothills thrust wedge (α+β=5°) is of a very thin Lower Paleozoic to Middle Jurassic platform succession close to that one of the eroded model wedges MW 4–5 with that is overlain by a thicker late Jurassic to Paleocene foreland basin décollements (Figs. 5 and 6), in which imposed erosional taper succession (Price and Monger, 2000). The structure of the Foothills angle is (α′=6°) and basal detachment is horizontal (Fig. 8c). The (Fig. 10a) is dominated by closely spaced, steeply dipping, imbricate, critical taper of the whole Canadian Rockies thrust belt is estimated to listric thrust slices, which give rise to the topography dominated by be 5.2° in case of syntectonic and 8.9° in case of post-tectonic erosion ridges that are held up by sandstones and valleys that are eroded in (Osborn et al., 2006). shales (Price and Monger, 2000). At the surface, the thrust faults are The Moose Mountain and Limestone Mountain Culminations closely spaced and generally dip steeply (N45°) to the southwest (Fig. 10b–d) in the western Foothills belt represent the examples of (Fig. 10a–b); but at depth they flatten and merge with each other, the structures with antiformal stacks (Begin and Spratt, 2002; with a regional detachment near the top of the Paleozoic rocks, and Newson and Sanderson, 1996; Price and Monger, 2000; Repol et al., eventually, with the basal detachment of the foreland thrust and fold 2010; Soule and Newson, 2000). The stacks are formed by Paleozoic belt, below which the Paleoproterozoic crystalline basement and (Cambrian to Carboniferous) limestone thrust sheets that are flanked overlying supracrustal rocks are undeformed. The basal detachment by Jurassic and Cretaceous clastic sequences. Hydrocarbon reservoirs in the Alberta Foothills dips gently (β=3°) to the southwest that (Fig. 10b–c) are associated with Paleozoic carbonates in the together with the taper angle (α=2°) determines the geometry of antiformal stack of the Moose Mountain culmination (Price and the thrust wedge in the frontal part of the Canadian Rockies (Fig. 10a). Monger, 2000; Soule and Newson, 2000). The oil and gas pools are

Fig. 10. (a) Structural cross-section across the southern Alberta Foothills, south of the Bow Valley, after Ollerenshaw (1978); Price and Monger (2000); 1, Tertiary; 2, Upper Cretaceous; 3, Upper–Lower Cretaceous and Jurassic; 4, Mississippian; 5, Upper Devonian; and 6, Cambrian strata. (b) Simplified structural sections: Jumping Pound, and Jumping Pound West gas fields, after Bruce, et al. (1995); Price and Monger (2000); 1, Tertiary; 2, Upper Cretaceous; 3, Lower Cretaceous and Jurassic; 4, Mississippian and Upper Devonian; and 5, Cambrian strata. Wells are shown. (c) Simplified structural section through the Moose Mountain Culmination, after Newson and Sanderson (1996), Soule and Newson (2000). The Moose Mountain thrust cuts down section over the highest point and the leading edge of the basal duplex; 1, Jurassic to Cretaceous; 2, Mississippian; 3, Devonian; and 4, Cambrian strata. Wells are shown. (d) Structural cross-section across Limestone Mountain Culmination, after Begin and Spratt (2002). The antiformal stack is a foreland-dipping duplex. 1–2, Cretaceous Edmonton (1) and Belly River (2); 3, Cretaceous Blairmore and Blackstone and Jurassic Fernie and Kootenay; 4, Mississippian; 5, Devonian; and 6, Cambrian strata. LD — Low Detachment. 348 E. Konstantinovskaya, J. Malavieille / Tectonophysics 502 (2011) 336–350 sealed by the evaporitic Mississippian Formation emplaced by the et al., 2006). The antiformal stack composed of Paleozoic arched thrust Moose Mountain thrust (Soule and Newson, 2000). slices is localized in front of the McConnel Thrust, similar to the model The antiformal stack of the Moose Mountain culmination is limited wedges MW4–5, in which antiformal stacks of basal layers are at the base by a regional main detachment level (Fig. 10c), lodged at exhumed in front of the proto-wedge toe (Figs. 5c and 6d). It seems the base of the Cambrian (Newson and Sanderson, 1996). The Moose that proto-wedge localizes the subsequent deformation ahead of its Mountain Thrust Fault separates the antiformal stack from the upper toe, without being affected significantly by thrust faults during model imbricate system (Newson and Sanderson, 1996; Repol et al., 2010; deformation. Such a relationship is likely observed in the frontal part Soule and Newson, 2000). The Moose Mountain thrust surface (Soule of the Canadian Rockies, where the older thrust wedge of Paleozoic and Newson, 2000) ramped up over the western flank and top of the strata above the McConnel Thrust is not affected by the deformation Moose Mountain culmination and cuts down into the footwall duplex propagating more externally into the Mesozoic strata of the Alberta on its eastern flank (Fig. 10c). Foothills as the accretionary wedge grows. The exhumation of It was suggested that the Moose Mountain thrust developed Paleozoic antiformal stack occurs at the rear part of the newly formed simultaneously (alternatively) with the underlying duplex, with the thrust wedge, in front of the older wedge toe. majority of displacement on the thrust occurring before the footwall The Canadian Rockies are characterized by the typical presence of a duplex emplacement (Soule and Newson, 2000). Price (2001) also frontal passive-roof duplex (or triangle zone) structures (Fig. 10a–b) noted that displacement in the Front Ranges north of Banff, Alberta (Gordy et al., 1977; Hiebert and Spratt, 1996; Lebel et al., 1996; Price, occurred simultaneously on several major faults that are distributed 1981; Stockmal et al., 2001). The formation of triangle zones in sand across the thrust wedge. These observations are in good correspon- thrust wedges was previously obtained mostly in presence of ductile dence with our experimental models, in which displacement occurs décollements, being sensitive to basal shear stress, variation in strain simultaneously along two major décollements and in the underlying rate, strength and dip of the décollement, strength of the wedge, and duplexes (Figs. 5–7). the presence of syntectonic erosion (Bonini, 2007; Cotton and Koyi, The structure of the Limestone Mountain Culmination (Fig. 10d) is 2000; Couzens-Schultz et al., 2003; Gutscher et al., 2001; MacKay, dominated by two lithotectonic packages separated by a major 1995; Mugnier et al., 1997). In the present study, no triangle zone was detachment (Begin and Spratt, 2002). The lower structural package is obtained in eroded model wedges contained or not frictional glass represented by a SW-plunging antiformal stack of four thrust sheets of microbeads décollements. However, triangle zones were observed in the Paleozoic carbonate platform rocks. The upper structural package eroded sand thrust wedges with frictional glass microbeads décolle- comprises Mesozoic siliciclastic foreland basin rocks, deformed into ments (Konstantinovskaya et al., 2009). That experiment set differs an NE-verging thrust-and-fold belt (Begin and Spratt, 2002). The from the present study by the simultaneous application of syntectonic antiformal stack is detached from the upper package by a decoupling erosion and sedimentation and by dip of basal detachment (β=2°) surface of the Roof Thrust. The imbricate fan of the upper package is and décollements, although the critical taper 6–7° (basal detachment passively folded and tilted toward the foreland above the antiformal dip β=2°+erosional slope α=3–4°) was the same as in the model stack. Our model wedges with one and two décollements also wedges MW4–5(α′=6°). Syntectonic sedimentation in foreland demonstrate the passive folding and tilting of the cover layers areas of thrust wedges favors the activation of a weak décollement above the roof décollements toward the frontal (foreland) side of the layer at the base of a cover sequence (Konstantinovskaya et al., 2009; antiformal stacks (Figs. 5–7), similar to natural structures of Moose Mugnier et al., 1997), while syntectonic erosion delays forward Mountain and Limestone Mountain Culminations (Figs. 10c–d). propagation of the deformation front and results in the development The effects of syn- and post-tectonic erosion on exhumation in the of out-of-sequence thrusting, duplexing and exhumation of the rear Canadian Rocky Mountains were studied by Osborn et al. (2006) by part of wedges, (Bonnet et al., 2007; Davis et al., 1983; Horton, 1999; comparison of physiography between end-of-Laramide time and the Hoth et al., 2006; Konstantinovskaia and Malavieille, 2005; Willett present day. It was concluded that erosion triggered exhumation in the et al., 1993). The combination of the both processes in thrust wedges rear of the thrust wedge with a subsequent forward shift of active thrust with slightly inclined décollements is favorable for the triangle zone faults. For example, once Paleozoic rocks had been exposed in the development. western front ranges (Bourgeau thrust sheet), a new, more easterly Based on the example of Canadian Rockies, Osborn et al. (2006) interface between mountains and foothills was created. This interface calculated that in case of the post-tectonic erosion, the critical taper of likely jumped still farther eastward to the McConnell thrust sheet as the the thrust belt might have been 8.9°, and 6–8 km of material could Mesozoic rocks were progressively eroded (Osborn et al., 2006). have been eroded with the erosion rate 0.13 mm/yr. In case of In our experiments, the domains of maximum exhumation are syntectonic (syn-orogenic) erosion, the critical taper of the thrust belt located at the rear part of the accretionary wedges while the frontal might have been 5.2°, and 2–4 km of material could have been eroded thrusts continue to propagate toward the foreland (Figs. 5–7), similar with the erosion rate 0.065 mm/yr. Thus, the erosion slope likely to the Canadian Rockies (Osborn et al., 2006)(Fig. 10a–b). The upper affects the extent of material removal in the orogenic belt. imbricate wedge located above the roof décollement is folded into the Our experiments support the conclusions of Osborn et al. (2006). synformal klippe between the exhumed basal layers and the newly Firstly, variation in erosion slopes applied to identical models with two accreted frontal tectonic slices (Fig. 5). The thrust slices in the upper décollements (MW5–6) resulted in changes of fault kinematics (Figs. 6 thrust wedge above the décollement propagate much farther toward and 7, see text above) and consequently — in changes of the extent of the foreland if compared to the thrust slices of the basal antiformal basal material underplating and erosional removal (Fig. 9). The 6° angle duplex (Fig. 5c–d). The similar suggestion was made by Repol et al. of erosion profile promoted formation of antiformal stack (Fig. 6d) with (2010) for the Moose Mountain culmination: the thrust system high extent of basal underplating (MW5), and the 8° erosion slope underlying the Moose Mountain Thrust (Fig. 10c) is characterized by provided formation of individual ramp-anticlines and small scale basal thrusts of smaller size and it likely suffered the significantly less duplex (Fig. 7c–d) with low extent of the basal underplating (MW6). tectonic transport when compared to the thrust system above the Secondly, higher rate of erosion (Fig. 9) leads in the models wedges Moose Mountain Thrust, for which persistent structural emplacement (MW2–3) to higher extent of basal material exhumation. along strike is observed. The model proto-wedge in presented experiments (Fig. 6d) may 5. Conclusions represent in nature the Paleozoic strata above the McConnel Thrust (Fig. 10a) that consist of the older thrust wedge formed previously to Interactions between surface processes and wedge mechanics the Mesozoic imbricate thrust sheets of the Alberta Foothills (Osborn have important consequences on the structure and evolution of E. Konstantinovskaya, J. Malavieille / Tectonophysics 502 (2011) 336–350 349 foreland thrust belts. Different accretion mechanisms are thus com- Cobbold, P.R., Davy, P., Gapais, D., Rossello, E.A., Sadybakasov, E., Thomas, J.C., Tondji Bijo, J.J., de Urreiztieta, M., 1993. Sedimentary basins and crustal thickening. bined to account for wedge growth: frontal accretion, backthrusting, Sediment. Geol. 86, 77–89. underthrusting and underplating due to décollement induced duplex Costa, E., Vendeville, B.C., 2002. Experimental insights on the geometry and kinematics formation at depth. These mechanisms may function simultaneously, of fold-and-thrust belts above a weak, viscous evaporite décollement. J. Struct. Geol. 24, 1729–1739. being located at different parts across the wedge. Cotton, J.T., Koyi, H.A., 2000. Modelling of thrust fronts above ductile and frictional Our experiments clearly reflect these complex feedback mecha- detachments: application to structures in the Salt Range and Potwar Plateau. nisms. A thrust wedge without erosion constantly grows by basal Pakistan. Geol. Soc. Am. Bull 112 (3), 351–363. underthrusting and forward propagation of frontal thrusts. The major Couzens-Schultz, B.A., Vendeville, B.C., Wiltschko, D.V., 2003. Duplex style and triangle zone formation: insights from physical modeling. J. Struct. Geol. 25, 1623–1644. backthrust at the rear of the wedge helps to rapidly reach the critical Cruz, L., Teyssier, C., Perg, L., Take, A., Fayon, A., 2008. Deformation, exhumation, and taper. Addition of erosion limits thrust wedge forward propagation topography of experimental doubly-vergent orogenic wedges subjected to and thickening. The eroded thrust wedges in our experiments do not asymmetric erosion. J. Struct. Geol. 30, 98–115. Dahlen, F.A., 1984. Non-cohesive critical Coulomb wedges: an exact solution. J. Geophys. grow laterally preserving the constant geometry (slope, thickness and Res. 89, 10,125–10,133. length) through shortening. The higher rate of erosion leads to higher Dahlen, F.A., Suppe, J., Davis, D., 1984. Mechanics of fold-and-thrust belts and extent of exhumation of basal material. Slow erosion leads to lesser accretionary wedges: cohesive Coulomb theory. J. Geophys. Res. 89, 10,087–10,101. Davis, D.M., Engelder, T., 1985. Thin-skinned deformation over salt. In: Lerche, I., extent of basal underthrusting. Variations in erosion slopes led in O'Brien, J.J. (Eds.), Dynamical Geology of Salt and Related Structures. Academic changes of fault kinematics and of the extent of basal material Press, San Diego, California, pp. 301–337. exhumation. The imposition of erosion taper with lower (6°) angle Davis, D., Suppe, J., Dahlen, F.A., 1983. Mechanics of fold and thrust belts and accretionary wedges. J. Geophys. Res. 88, 1153–1172. than the critical value (8°) in the models with décollements promoted Davy, P., Cobbold, P.R., 1991. Experiments on shortening of a 4-layer model of the formation of antiformal stack with high extent of exhumation. The 8° continental lithosphere. Tectonophysics 188, 1–25. erosion profile (equal to critical slope) provided formation of Del Castello, M., Pini, G.A., McClay, K.R., 2004. Effect of unbalanced topography and overloading on Coulomb wedge kinematics: insights from sandbox modelling. individual ramp-anticlines in the upper wedge above the décollement J. Geophys. Res. 109, B05405. doi:10.1029/2003JB002709. and small scale normal duplex below it with low amount of basal Fermor, P.R., Price, R.A., 1987. Multiduplex structure along the base of the Lewis thrust underplating. The results of our experiments conform well to a natural sheet in the southern Canadian Rockies. Bull. Can. Petrol. Geol. 35 (2), 159–185. example of Canadian Rockies. Glodny, J., Lohrmann, J., Echtler, H., Gräfe, K., Seifert, W., Collao, S., Figueroa, O., 2005. Internal dynamics of a paleoaccretionary wedge: insights from combined isotope tectonochronology and sandbox modelling of the South-Central Chilean forearc. Earth Planet. Sci. Lett. 231, 23–39. Acknowledgments Gordy, P.L., Frey, F.R., Norris, D.K., 1977. Geological guide for the C.S.P.G. and 1977 Waterton-Glacier Park Field Conference. Can. Soc. Petrol. Geol. Calgary Ab. 1–93. We particularly acknowledge Stephane Dominguez for help and Graveleau, F., 2008. Interactions Tectonique, Erosion, Sédimentation dans les avant- pays de chaînes: Modélisation analogique et étude des piémonts de l'est du Tian constructive comments during experimental work and Christian Shan (Asie centrale). Université Montpellier II — Sciences et Techniques du Romano for technical support. Elena Konstantinovskaia benefited of Languedoc, Thesis. 487 p. grants from the French MENRT to fund the Associate Professor Gutscher, M.-A., Kukowski, N., Malavieille, J., Lallemand, S., 1996. Cyclical behavior of thrust wedges: insights from high basal friction sandbox experiments. Geology 24 position in Montpellier (2000–2003). Our sincere thanks to F. Storti, (2), 135–138. M. Bonini and an anonymous reviewer for useful and constructive Gutscher, M.-A., Kukowski, N., Malavieille, J., Lallemand, S., 1998. Episodic imbricate comments that aid us to improve the manuscript. thrusting and underthrusting: analog experiments and mechanical analysis applied to the Alaskan Accretionary Wedge. J. Geophys. Res. 103 (B5), 10,161–10,176. Gutscher, M.-A., Klaeschen, D., Flueh, E., Malavieille, J., 2001. Non-Coulomb wedges, wrong-way thrusting, and natural hazards in Cascadia. Geology 29 (5), 379–382. Appendix A. Supplementary data Hiebert, S.N., Spratt, D.A., 1996. Geometry of the thrust front near Pincher Creek. Ab. Bull. Can. Petrol. Geol. 44 (2), 195–201. Supplementary data to this article can be found online at Horton, B.K., 1999. Erosional control on the geometry and kinematics of thrust belt development in the central Andes. Tectonics 18 (6), 1292–1304. doi:10.1016/j.tecto.2011.01.020. Hoth, S., Adam, J., Kukowski, N., Oncken, O., 2006. Influence of erosion on the kinematics of bivergent orogens: results from scaled sandbox simulations. In: Willett, S.D., Hovius, N., Brandon, M.T., Fisher, D.M. (Eds.), Tectonics, Climate, and Landscape References Evolution. Geol. Soc. Am. Spec. Paper, 398. Penrose Conference Series, pp. 201–225. Konstantinovskaya, E.A., Rodriguez, D., Kirkwood, D., Harris, L.B., Thériault, R., 2009. Avouac, J.P., 2003. Mountain building, erosion, and the seismic cycle in the Nepal Effects of basement structure, sedimentation and erosion on thrust wedge Himalaya. Adv. Geophys. 46, 1–80. geometry: an example from the Quebec Appalachians and analogue models. Bull. Barrier, L., Nalpas, T., Gapais, D., Proust, J.N., Casas, A., Bourquin, S., 2002. Influence of Can. Petrol. Geol. 57 (1), 34–62. syntectonic sedimentation on thrust geometry. Field examples from the Iberian Konstantinovskaia, E.A., Malavieille, J., 2005. Erosion and exhumation in accretionary Chain (Spain) and analogue modelling. Sediment. Geol. 146, 91–104. orogens: experimental and geological approaches. Geochem. Geophys. Geosyst. 6 Begin, N.J., Spratt, D.A., 2002. Role of transverse faulting in along-strike termination of (2), Q02006. doi:10.1029/2004GC000794. Limestone Mountain Culmination, Rocky Mountain thrust-and-fold belt, Alberta. Koons, P.O., 1989. The topographic evolution of collisional mountain belts: a numerical Canada. J. Struct. Geol. 24, 689–707. look at the Southern Alps, New Zealand. Am. J. Sci. 289, 1041–1069. Belotti, H.J., Saccavino, L.L., Schachner, G.A., 1995. Structural styles and petroleum Koyi, H.A., Hessami, K., Teixell, A., 2000. Epicenter distribution and magnitude of occurence in the Sub-Andean thrust belt of northern Argentina. In: Tankard, A.J., earthquakes in fold-thrust belts: insights from sandbox models. Geophys. Res. Lett. Suarez, S., Welsink, H.J. (Eds.), Petroleum Basins of South America: Am. Assoc. 27, 273–276. Petrol. Geol. Mem., 62, pp. 545–555. Kukowski, N., Lallemand, S.E., Malavieille, J., Gutscher, M.A., Reston, T.J., 2002. Bonini, M., 2003. Detachment folding, fold amplification, and diapirism in thrust wedge Mechanical decoupling and basal duplex formation observed in sandbox experi- experiments. Tectonics 22 (6), 1065. doi:10.1029/2002TC001458. ments with application to the Mediterranean Ridge accretionary complex. Mar. Bonini, M., 2007. Deformation patterns and structural vergence in brittle–ductile thrust Geol. 186, 29–42. wedges: an additional analogue modelling perspective. J. Struct. Geol. 29, 141–158. Lallemand, S.E., Schnurle, P., Malavieille, J., 1994. Coulomb theory applied to doi:10.1016/j.jsg.2006.06.012. accretionary and non-accretionary wedges: possible causes for tectonic erosion Bonnet, C., Malavieille, J., Mosar, J., 2007. Interactions between tectonics, erosion, and and/or frontal accretion. J. Geophys. Res. 99, 12,033–12,055. sedimentation during the recent evolution of the Alpine orogen: analogue Langenberg, W., Pana, D., Stockmal, G., Price, R., Spratt, D., 2006. The structure of the modeling insights. Tectonics 26, TC6016. doi:10.1029/2006TC002048. Crowsnest Pass Transect. Field trip guide. The 26th Canadian Tectonic group Bonnet, C., Malavieille, J., Mosar, J., 2008. Surface processes versus kinematics of thrust Workshop. 14 – 15 October 2006, Crowsnest Pass, Alberta, Canada. 37 p. belts: impact on rates of erosion, sedimentation, and exhumation — insights from Larroque, C., Calassou, S., Malavieille, J., Chanier, F., 1995. Experimental modeling of analogue models. Bull. Soc. Géol. France. 179 (3), 179–192. forearc basin development during accretionary wedge growth. Basin Res. 7, Bruce, C.J., Frey, F.R., Graff, G.W., Tippet, C.R., 1995. Geological guidebook, 1995 C.S.P.G. 255–268. Student-industry field trip, in Calgary, Alberta. Can. Soc. Petrol. Geol. 261 p. Lebel, D., Langenberg, W., Mountjoy, E.W., 1996. Structure of the central Canadian Casas, A.M., Gapais, D., Nalpas, T., Besnard, K., Roman-Berdiel, T., 2001. Analogue models Cordilleran thrust-and-fold belt, Athabasca – Brazeau area. Ab. Bull. Can. Petrol. of transpressive systems. J. Struct. Geol. 23, 733–743. Geol. 44, 258–268. Chapple, W.M., 1978. Mechanics of thin-skinned fold-and-thrust belts. Geol. Soc. Am. Leturmy, P., Mugnier, J.L., Vinour, P., Baby, P., Colletta, B., Chabron, E., 2000. Piggyback Bull. 89, 1189–1198. basin development above a thin-skinned thrust belt with two detachment levels as 350 E. Konstantinovskaya, J. Malavieille / Tectonophysics 502 (2011) 336–350

a function of interactions between tectonic and superficial mass transfer; the case Price, R.A., 1981. The Cordilleran foreland thrust and fold belt in the southern Canadian of the Subandean Zone (Bolivia). Tectonophysics 320, 45–67. Rocky Mountains. In: McClay, K.R., Price, N.J. (Eds.), Thrust and Nappe Tectonics: Lohrmann, J., Kukowski, N., Adam, J., Oncken, O., 2003. The impact of analogue material Geol. Soc. London Spec. Publ., 9, pp. 427–448. properties on the geometry, kinematics and dynamics of convergent sand wedges. Price, R.A., 1986. The southern Canadian Cordillera: thrust faulting, tectonic wedging, J. Struct. Geol. 25, 1691–1711. and delamination of the lithosphere. J. Struct. Geol. 8, 239–254. MacKay, M.E., 1995. Structural variation and landward vergence at the toe of the Price, R.A., 2001. An evaluation of models for the kinematic evolution of thrust and fold Oregon accretionary prism. Tectonics 14 (5), 1309–1320. belts: structural analysis of a traverse fault zone in the Front ranges of the Canadian Malavieille, J., 1984. Modélisation expérimentale des chevauchements imbriqués: Rockies north of Banff. Ab. J. Struct. Geol. 23, 1079–1088. application aux chaînes de montagnes. Bull. Soc. Geol. Fr. 26, 129–138. Price, R.A., Fermor, P.R., 1985. Structure section of the Cordilleran foreland thrust and Malavieille, J., 2010. Impact of erosion, sedimentation and structural heritage on the fold belt west of Calgary, Alberta. Geol. Surv. Can. Pap. 14–84 1 sheet.. structure and kinematics of orogenic wedges: analog models and case studies. Geol. Price, R.A., Monger, J.W.H., 2000. A Transect of the Southern Canadian Cordillera from Soc. Am., Account GSA Today 20 (1), 4–10. doi:10.1130/GSATG48A.1. Calgary to Vancouver: Vancouver, British Columbia. Geol. Ass. Can, Cordilleran Malavieille, J., Konstantinovskaya, E., 2010. Impact of surface processes on the growth of Section. 164 p. orogenic wedges: insights from analog models and case studies. Geotectonics 44 (6), Repol, D., Kraemer, M., Stephenson, B., Eggenkamp, I., 2010. Moose Mountain: new 541–558. insight into its internal structure and relative timing of deformation. GeoCanada Malavieille, J., Calassou, S., Larroque, C., 1993. Modélisation expérimentale des relations 2010 Conference. Calgary, Alberta, Abstract, p. 524. tectonique sédimentation entre bassin avant-arc et prisme d'accrétion. C. R. Acad. Selzer, C., Buiter, S.J.H., Pfiffner, O.A., 2007. Sensitivity of shear zones in orogenic wedges Sci. Paris 316, 1131–1137. to surface processes and strain softening. Tectonophysics 437, 51–70. Mandl, G., Shippam, G.K., 1981. Mechanical model of thrust sheet gliding and Simoès, M., Avouac, J.P., Beyssac, O., Goffe, B., Farley, K.A., Chen, Y.-G., 2007. Mountain- imbrication. In: McClay, K.R., Price, N.J. (Eds.), Thrust and Nappe Tectonics: Geol. building in Taiwan: a thermo-kinematic model. J. Geophys. Res. 112, B11405. Soc. London Spec. Publ., 9, pp. 79–97. Simpson, G.D.H., 2006. Modelling interactions between fold-thrust belt deformation, Marques, F.O., Cobbold, P.R., 2002. Topography as a major factor in the development of foreland flexure and surface mass transport. Basin Res. 18, 125–143. arcuate thrust belts; insights from sandbox experiments. Tectonophysics 348, Smit, J.H.W., Brun, J.P., Sokoutis, D., 2003. Deformation of brittle–ductile thrust wedges 247–268. in experiments and nature. J. Geophys. Res. 108 (B10), 2480. doi:10.1029/ McClay, K.R., 1992. Glossary of thrust tectonic terms. In: McClay, K.R. (Ed.), Thrust 2002JB002190. Tectonics. Chapman and Hall, London, pp. 419–433. Smit, J., Burg, J.-P., Dolati, A., Sokoutis, D., 2010. Effects of mass waste events on thrust McMechan, M.E., 2001. Large-scale duplex structures in the McConnel thrust sheet, wedges: analogue experiments and application to the Makran accretionary wedge. Rocky Mountains. Southwest Ab. Bull. Can. Petr. Geol. 46 (3), 408–425. Tectonics 29, TC3003 10.1029/2009TC002526. McMechan, M.E., 2002. Structural geometry in the Carbon Creek area of the Rocky Soule, G.S., Newson, A.C., 2000. Footwall decapitation and contemporaneous thrust Mountain Fold and Thrust Belt, northeastern British Columbia. Bull. Can. Petr. Geol. fault motion in the Moose Mountain duplex, Rocky Mountain Foothills of Alberta. 50 (3), 407–418. Can. Soc. Explor. Geophys., Archives 2000. 2000. Merle, O., Abidi, N., 1995. Approche expérimentale du fonctionnement des rampes Soule, G.S., Spratt, D.A., 1996. En echelon geometry and two-dimensional model of the émergentes. Bull. Soc. Géol. Fr. 166 (5), 439–450. triangle zone, Grease Creek, Syncline area. Ab. Bull. Can. Petrol. Geol. 44, 313–323. Mugnier, J.L., Baby, P., Colletta, B., Vinour, P., Bale, P., Leturmy, P., 1997. Thrust geometry Stockmal, G.S., 2001. Abrupt changes in structural style and associated stratigraphic controlled by erosion and sedimentation: a view from analogue models. Geology thickness across cross-strike lineaments, Rocky Mountain Foothills, northeastern 25 (5), 427–430. British Columbia. Bull. Can. Petrol. Geol. 49 (4), 497–512. Mulugeta, G., Koyi, H., 1987. Three-dimensional geometry and kinematics of Stockmal, G.S., McMechan, M.E., Lebel, D., Mackay, P.A., 2001. Structural style and experimental piggyback thrusting. Geology 15, 1052–1056. evolution of the triangle zone and external Foothills, southwestern Alberta: Newson, A., Sanderson, D., 1996. Moose Mountain: an example of an oil and gas pool in implications for thin-skinned thrust-and-fold belt mechanics. Bull. Can. Petrol. the overthrust belt of the Canadian Rocky Mountains. Canadian Society of Geol. 49 (4), 472–496. Petroleum Geologists Field Trip. Institute of Sedimentary and Petroleum Geology, Storti, F., McClay, K., 1995. Influence of syntectonic sedimentation on thrust wedges in Calgary. analogue models. Geology 23, 999–1002. Ollerenshaw, N.C., 1978. Calgary, Alberta-British Columbia: Geological Survey of Storti, F., Poblet, J., 1997. Growth stratal architectures associated to décollement folds Canada, Map 1457A, 1:250 000 scale map and structure sections. and fault-propagation folds. Inferences on fold kinematics. Tectonophysics 282, Osborn, G., Stockmal, G., Haspel, R., 2006. Emergence of the Canadian Rockies and 353–373. adjacent plains: a comparison of physiography between end-of-Laramide time and Storti, F., Salvini, F., McClay, K., 2000. Synchronous and velocity-partitioned thrusting the present day. Geomorphology 75, 450–477. and thrust polarity reversal in experimentally produced, doubly-vergent thrust Persson, K.S., Sokoutis, D., 2002. Analogue models of orogenic wedges controlled by wedges: implications for natural orogens. Tectonics 19 (2), 378–396. doi:10.1029/ erosion. Tectonophysics 356, 323–336. 1998TC001079. Persson, K.S., Garcia-Castellanos, D., Sokoutis, D., 2004. River transport effects on Summerfield, M.A., Hulton, N.J., 1994. Natural controls of fluvial denudation rates in compressional belts: first results from an integrated analogue-numerical model. major world drainage basins. J. Geophys. Res. 99 (B7), 13 871–13 884. J. Geophys. Res. 109, B01409. doi:10.1029/2002JB002274. Vergés, J., Martinez, A., 1988. Corte Compensado del Pirineo oriental: geometria de las Platt, J.P., 1988. The mechanics of frontal imbrication: a first-order analysis. Geol. cuencas de antepais y edades de emplazamiento de los mantos de corrimiento. Acta Rundsch. 77 (2), 577–589. doi:10.1007/BF01832399. Geol. Hisp. 23, 95–106. Platt, J.P., 1990. Thrust mechanics in highly overpressured accretionary wedges. Willett, S.D., Beaumont, C., Fullsack, P., 1993. Mechanical model for the tectonics of J. Geophys. Res. 95 (B6), 9025–9034. doi:10.1029/JB095iB06p09025. doubly vergent compressional orogens. Geology 21, 371–374.