© A. Farkhutdinov et al., Published by EDP Sciences 2025

1 Introduction

The Urals is an example of a Paleozoic orogen which extends for nearly 2,500 km and represents the geographic border between Europe and Asia. Geographically (from north to south) it is divided into Polar, Cis-Polar, Northern, Middle and Southern Urals. One of the most geologically complicated structures of the Southern Urals is the Karatau structural complex (Fig. 1). The history of its study is associated with the discovery of the thrust-and-fold structure of the Ural Mountains, which refuted the concept of their vertical-block structure (‘geosynclinal theory’). The first thrust in the Urals was identified back in 1927 by the Soviet geologist and paleontologist Georgiy Nikolaevich Frederiks (1889–1938) (Frederiks, 1927), after him allochthonous structures were described by a number of other researchers. Around these discoveries, fierce disputes flared up, which at first were scientific in nature, but later (1930s) the mobilists became victims of political repression as supporters of a hostile bourgeois doctrine (Farkhutdinov et al., 2019). As a result, the geosynclinal theory of the formation of the Urals remained dominant until the second half of the 20^th century (Farkhutdinov et al., 2017).

There are three points of view on the formation of the Karatau structural complex: (1) it arose due to vertical movements exclusively (Shatskiy, 1945; Tetyaev, 1938); (2) it is thrusted as a result of compression (Kamaletdinov, 1974); (3) it arose due to a flexural bulge as a result of a load on the lithosphere (Kissin, 2008). The diversity of views on the formation of the territory confirms the relevance of these studies.

The aim is to propose a possible kinematic and mechanical evolution of the Karatau structural complex using analogue sandbox modelling based on a more precise geological interpretation.

Fig. 1 Map showing the different zones of the Urals and its geographic divisions (a). Geological map of the Southern and part of the Middle Urals showing the location of the URSEIS seismic profile (b). Main faults are: Tt − Tashly thrust, At − Alatau thrust, Zt − Zilmerdak thrust, Zf − Zuratkul fault, MUf − Main Uralian fault, Emf − East Magnitogorsk fault, Tf − Troisk fault. The box (c) indicates the area discussed in this paper (Adapted from Brown et al., 2011). Carte montrant les différentes zones de l’Oural et ses divisions géographiques (a). Carte géologique de l’Oural du Sud et d’une partie de l’Oural Moyen montrant l’emplacement du profil sismique URSEIS (b). L’encadré (c) indique la zone discutée dans cet article.

2 Geological settings

2.1 Regional geological settings

The Urals have formed as a consequence of the Uralian ocean closure and involves geodynamic episodes of island arcs accretion (namely: Tagil and Magnitogorsk arcs), obduction of oceanic crust and, at the end, continent-continent collision between Baltica, Kazakstania and Siberia continental blocks. From west to east Uralides are divided into a number of tectonic zones: the undeformed foreland basin, the foreland fold and thrust belt, the Magnitogorsk-Tagil Zone, the East Uralian Zone and the Trans-Uralian Zone (Fig. 1) (Brown et al., 2011).

The Southern Urals is one of the most studied part of the Ural mountains and was investigated during the multicomponent Urals Seismic Experiment and Integrated Studies (URSEIS) survey in 1995. URSEIS displayed the first complete seismic image of the crustal and upper mantle architecture of this intact Paleozoic orogen (Brown et al., 2006) (Fig. 2).

The Foreland fold and thrust belt is composed of shortened proximal continental margin with an Archean basement to the west and, to the east, of an accretionary complex including ophiolites as remnants of an oceanic domain. The Magnitogorsk-Tagil Zone corresponds to accreted island arcs between two continents and is bounded by two prominent crustal faults: to the west, the Main Uralian fault (MUf) limits the arcs and the Foreland fold and thrust belt and to the east, the East Uralian fault (EUf) marks the boundary of the eastern continental crustal blocks.

The Urals have a long-lasting geodynamic history with five main structural and depositional stages (Puchkov, 2013), observed in the whole territory and summarized in the figure 3. Archean-Paleoproterozoic is a time of formation of the Volgo-Uralia continent and its amalgamation with other blocks into Baltica continental block. During Meso- and Neoproterozoic eras (known locally as Riphean (R) and Vendian (V)), different extensional stages took place and ended locally with the closure of the pre-Uralian ocean during a convergent event (Cadomian orogeny) between Baltica and Arctida continental blocks (Kuznetsov et al., 2010), building the Timanides. The Paleozoic-Early Mesozoic stage, corresponds to the development of the Uralides (during a period equivalent to the west-european Variscan orogeny). The Mid-Jurassic-Miocene times were those of a quiet platform stage. Finally, the Pliocene-Quaternary period is the neo-orogenic stage with some uplift of the structures resulting in some denudation of the western fold and thrust belt and exhumation to the present (Glasmacher et al., 2002). Present day seismicity accompanies this evolution − the most recent earthquake, with a magnitude of 5.4, occurred in 2018 in the center of the Bashkirian Megazone (Fig. 1b). The focal mechanism was mainly compressional with a strike-slip component (Tevelev et al., 2019).

To summarize, Ural Mountains encompass a long and polyphase tectonic evolution including two main convergent events: the Cadomian (600–550 Ma) and the Variscan (405–280 Ma) orogeneses (Peive et al., 1977), creating Timanian and Uralian orogens respectively (Fig. 4).

In the Karatau domain of the Southern Urals, a wide graben-like structure known locally as the Kaltasin Aulacogen, was formed during the Early Riphean extension episode (1720–1700 Ma) at high angle to the Urals boundary. It is traced in a southeastern direction (in modern coordinates here and hereinafter in the text) from the platform to the territory of the Urals (Figs. 5 and 6). Another episode of extension (same direction) occurred in the Middle Riphean (1380 ± 20–1366 ± 12 Ma) (Puchkov et al., 2007).

During Late Riphean times (720 ± 7–709.9 ± 7.3 Ma) (Puchkov et al., 2007), a new stage of extension and rifting took place in the NW-SE trending Kaltasin Aulacogen and potentially simultaneously along the nearly N-S trending pre-Uralian ocean margin (Fig. 6), resulting in a rift-rift-rift triple junction similar to the Labrador Atlantic triple junction during Early Cenozoic. The rift in the Karatau structural complex domain differs from the Labrador example that the former has not evolved into the ocean. On the contrary eastern part of the triple junction arrived to the Pre-Uralian ocean formation with subsequent closure and formation of the Timanian orogen. The Zilmerdak thrust (Figs. 5 and 6) in the Southern Urals marks Timanian orogen current western boundary (Brown et al., 1999). Later, erosion and denudation took place, and in the Late Cambrian a new stage of extension led to the Uralian ocean formation.

The closure of this Paleozoic ocean happened during the Variscan orogeny with the Main Uralian fault as a suture zone. It was a WSW-ENE compression, almost parallel to the Kaltasin Aulacogen (Fig. 6). During the neo-orogenic stage (Pliocene-Quaternary) Ural Mountains got their present topography, when Variscan structures were uplifted. A possible cause of this uplift is the remote influence of stresses from the Alpine-Himalaya collision belt (Petrov et al., 2021).

Such a geological history has given a number of features to the Urals, one of which is high amount of basement involvement with localization of thrusting by reactivation of two sets of pre-existing structures in the basement (Brown et al., 1999; Perez-Estaun et al., 1997; Giese et al., 1999).

Tectonic inversion of extensional structures at a high angle to the Urals resulted in the development of lateral structures and along-strike changes (Perez-Estaun et al., 1997; Brown et al., 1999). One of excellent examples of a lateral structure across which displacement is transferred toward the foreland is the Karatau structural complex (Brown et al., 2006).

Fig. 2 Interpreted line drawing of the depth migrated URSEIS reflection seismic profiles (Brown et al., 2008). Complete names of the faults are given in Figure 1. Tracé interprété des profils de sismique réflexion URSEIS migrés en profondeur.

Fig. 3 Geodynamic and structural stages of the Urals (according to Puchkov, 2013, Maslov et al., 2010, Petrov et al., 2021 with modifications). Étapes géodynamiques et tectoniques de l’Oural.

Fig. 4 Orogens of the East European Craton (according to Pease et al., 2021 with slight modifications according to Baluev, 2006 and Kuznetsov and Romanyuk, 2021). Orogènes du craton Est-Européen.

Fig. 5 Superimposed Riphean-Vendian sediments thickness (Masagutov, 2002) and major tectonic faults on aeromagnetic map of the Southern Urals (extracted from EMAG2v3 global magnetic grid (Meyer et al., 2017)). Complete names of the faults are given in Figure 1 description. Superposition de l’épaisseur des sédiments du Riphéen-Vendien et des principales failles tectoniques sur la carte aéromagnétique de l’Oural du Sud. Les noms des failles sont détaillés dans la légende de la Figure 1.

Fig. 6 Approximate directions and ages of orogeny and rifting periods in the Karatau structural complex area (age is given according to Puchkov et al., 2007). Directions et âges approximatifs des périodes d’orogenèse et de rifting dans la région du complexe structural de Karatau.

2.2 The Karatau structural complex: definition and description

The Karatau structural complex (KSC) which stands out by its topography (Fig. 7) is part of the Bashkirian Megazone (whose outcrops are mostly composed of the Riphean-Vendian deposits with age of 1650–540 Ma, Fig. 1b).

At first glance, the KSC is outlined on geological maps by the apparent offset of the Riphean-Permian main boundary fault (marked as Kf in the Fig. 8). The Karatau structural complex is bounded to the southwest by a prominent left-lateral strike-slip fault, the Asha fault and to the northeast by a high-angle right lateral Yuruzan strike-slip fault,both acting as a transfer faults. The Asha fault damage zone ranges from 20 to 200 m whereas for the Yuryuzan fault it is more than 500 m (Kamaletdinov, 1974). To the northwest, the complex ends in a major thrust of the Upper Riphean over the Permian strata known as the Karatau fault. The southeast boundary of the KSC seems to be a southwest to northeast trending right-lateral strike-slip fault becoming more complex to the northeast. This complex zone is known as the Pervomay transpression zone, which consists of a bundle of right-lateral strike-slip faults (Tevelev et al., 2018). The KSC itself can be divided in two parts, the southwestern half showing a series of four folds (including the following anticlines from north to south: Karatau, Vorobyinogorsk, Azhigardak, and Berezovogorsk) associated to northwestern verging thrusts. In the northeastern part of the KSC, between the frontal thrust and the large Suleimanovo anticline, there is a large basin filled with Permian deposits and known as the Sim trough (the possible piggy-back basin) (Kamaletdinov, 1974).

The Karatau structural complex includes Lower Riphean to Late Paleozoic rocks (Fig. 9). Outcropping Upper Riphean-Vendian formations are represented mainly by terrigenous rocks with fragments of quartz and feldspar. Paleozoic deposits begin with the Devonian terrigenous-carbonate, terrigenous-clayey and carbonate layers with a thickness of 630 m. Thickness of the Carboniferous, which is predominantly carbonate, is 830 m. Above, Permian sandstones, siltstones and mudstones are rhythmically interbedded up to 1900 m thick. Permian deposits exist only in the south-eastern part of the Karatau complex (Sim trough) and contain horizons of large-block formations − olistostrome. The total thickness of Paleozoic sediments ranges from 1460 to 3360 m (Kamaletdinov, 1974; Kazancev et al., 1999; Puzhakov et al., 2019).

Gravity surveys conducted in the 1970s estimated the depth to the basement to be 10-12 km within the Karatau complex (Shakurov et al., 1988). More recent seismic studies conducted near Karatau showed an even greater depth, which can be estimated about 20 km within the complex itself.

Fig. 7 Topographic map and digital elevation model of the Karatau structural complex territory (according to Shuttle Radar Topography Mission (30 m resolution)). Carte topographique et modèle numérique de terrain du territoire du complexe structural de Karatau.

Fig. 8 Geological map of the Karatau structural complex territory (Knyazev et al., 2013 with modifications). Carte géologique du territoire du complexe structural de Karatau.

Fig. 9 Representative stratigraphy of the Karatau structural complex (south-eastern part) (according to Puzhakov et al., 2019 with modifications). Stratigraphie représentative du complexe structural de Karatau (partie sud-est).

2.3 Previous interpretations: Karatau and the mobilist paradigm

One of the fragments of the KSC, the Karatau Ridge (Fig. 7), has long been considered as a vertically elevated block. In 1930, under the leadership of its director, Academician Dmitry Vasilyevich Nalivkin (1889–1982), a geological expedition from the Leningrad Institute of Geological Map of the Academy of Sciences of the USSR worked on the Karatau Ridge. The conclusion of the expedition members was a pure vertical (‘no thrust’) origin of the Karatau Ridge structure. Academician Nikolay Sergeevich Shatskiy (1895–1960) later also considered the Karatau as an uplifted block of the East European platform basement and denied thrusts. The first to declare the allochthonous nature, despite the generally accepted fixist paradigm, was Murat Abdulhakovich Kamaletdinov (1928–2013), who mapped the Karatau thrust in 1954. It is shown on the State Geological Map N-40-X (1:200 000 scale) (Sinitsyn and Sinitsyna, 1956), with reference to Kamaletdinov’s 1954 and 1956 reports (Kamaletdinov, 2011). The interpretation of the Karatau ridge as a thrust sheet was made a year after Stalin’s death, when there was no longer death penalty for mobilist views in the Soviet Union, but their supporters still aroused suspicions as enemy elements, and as for thrust structures, they were denied (Kamaletdinov, 2007; Farkhutdinov et al., 2017). This did not stop Kamaletdinov, who continued to study the Urals from a mobilist point of view. As a result of detailed geological survey work, drilling of structural exploration and parametric wells in the 1960s, the fold and thrust structure of the Urals was established and the thrust-nappe theory was endorsed (Kamaletdinov, 1965a, 1965b; Farkhutdinov et al., 2019). The Karatau thrust, as a part of the Karatau structural complex, has a historical importance in this major discovery.

Besides the “fixist" and “mobilist" interpretations, a recent interpretation by Kissin (2008) considers the KSC as a vertically uplifted block, due to a flexural bulge resulting from the Uralian load on the lithosphere. As it is based solely on map analysis this hypothesis is difficult to evaluate because of the lack of cross-section and stratigraphic logs.

3 Structural interpretation and working hypothesis

3.1 Revised tectonic interpretation of the Karatau structural complex

As previously stated, the tectonic complex is limited by the Karatau thrust, left-lateral Asha and right-lateral Yuryuzan strike-slip faults (Fig. 8).

In order to build geological cross-sections in the KSC, we had access to seismic data in the foredeep basin southwest of the Asha fault and northeast of the Yuruzan fault, including a few deep regional lines corresponding to the sinuous path of section IJ (line published in Lozin, 2015) (Fig. 10) and to the north-west part of section GH (unpublished). In these domains, the décollement can be inferred from the seismic data as well as the actual deformation front even when buried beneath Paleozoic sequences. Inside the KSC on the other hand, we had to rely on the published geologic maps (Sinitsyn and Sinitsyna, 1956; Knyazev et al., 2013; Moseychuk et al., 2017; Puzhakov et al., 2019) to deduce the sediment thicknesses down to the Upper Riphean and to infer the décollements from the construction of balanced cross-sections.

The Karatau thrust has a shape of an irregular arc curved to the north on a map view. In a vertical section, it is slightly flattening out and merging with a décollement at depth. By consistency with a ramp and flat model, we end up with a décollement situated at approximately 6–7 km depth within the bottom of the Upper Riphean (red arrow at 6820 m depth, Fig. 9), so that thickness of the Karatau allochton is increasing from north-west to south-east (Fig. 11, cross-sections AB and CD).

Our interpretation is presented in a series of cross-sections in Figure 11 (legend and location are in Fig. 8) as well as in a 3D schematic view (Fig. 12), given the complex geological features of the KSC. The most striking feature is the rapid change in Upper Riphean thickness from less than 1000 m outside the KSC up to 4000 m inside. This is in good agreement with a narrowing of the Kaltasin Aulacogen rift between the Asha and Yuruzan faults during the Late Riphean extensional episode as can be seen in Figure 13.

Another important feature is the change in the foredeep structure across the Yuruzan fault. Northeast of this fault, a very thick (up to 4000 m) flexural basin with a typical wedge section developed during the late Carboniferous, whereas thicknesses of the Carboniferous remain smaller than 1000 m in sections AB and CD within the KSC (Fig. 11). We have chosen however to ignore the differences and to group the Paleozoic sequences in the transverse section evolution of Figure 13.

We performed a restoration of the slips on the faults and of the folds in these cross-sections. The amount of shortening varies greatly on these cross-sections (from north to south): IJ − 4.3 km, CD (the KSC territory, bounded by strike-slip reverse fault at south-east) − 30.4 km, AB (the KSC territory, bounded by strike-slip reverse fault at south-east) − 43.1 km, GH − 11.5 km. The reason for such differences is that cross-sections AB and CD are covering mostly the fold and thrust belt and IJ, GH completely and mostly foredeep basin respectively.

From a structural perspective, we observe the changes in the depth of décollement and the location of the deformation front both within and outside the Karatau. The deformation front is not outcropping but covered by Paleozoic sediments on cross-section GH (Fig. 11). Outside of the KSC to the south-west the main décollement level (at the base of the Upper Riphean) passes from 6 (Karatau) to 3.5 km and to the north-east from 6.5 (Karatau) to 6 km. Reconstruction at the end of the Late Riphean shows the vertical offset of the Upper Riphean base from within (3–4 km thicknesses) to outside (0.7–1 km thicknesses) of the KSC (Fig. 13). It can be explained by the fact that Asha fault during Late Riphean was a major normal fault and one of the borders (along with the Yuruzan fault) of the Upper Riphean rift basin at the territory of the Karatau structural complex (Fig. 13).The fact that there is no transverse structure crossing the Zilmerdak fault (Zf in the Fig. 11) to the S-E from the KSC suggest strongly that it marks the limit of the Karatau Upper Riphean rift as expected in rift-rift-rift triple junction configuration.

Several factors contribute to the high interest and complexity of the Karatau structural complex territory: NE-SW strike (unlike other major structures of the Southern Urals), greater advancement of deformation front, division of the Pre-Uralian foredeep into two parts, tectonic faults limiting structural complex on all sides, presence of Proterozoic rocks at the surface.

Fig. 10 Part of the regional seismic line (Lozin, 2015) corresponding to the section IJ (Fig. 8 and 11) with interpretations. Partie de la ligne sismique régionale (Lozin, 2015) correspondant à la section IJ (Fig. 8 et 11) avec interprétations.

Fig. 11 Cross-sections for Karatau structural complex geological map (location of the sections and legend in Fig. 8). Coupes transversales de la carte géologique du complexe structural de Karatau (localisation des coupes et légende dans la Fig. 8).

Fig. 12 3D schematic representation of the Karatau structural complex geological cross-sections (location of the sections and legend in Fig. 8). Représentation schématique en 3D des coupes géologiques du complexe structural de Karatau (localisation des coupes et légende dans la Fig. 8).

Fig. 13 Schematic block-diagram along EF section crossing the strike of the Karatau structural complex (Permian and Riphean time). The late Variscan décollements are shown as horizontal thick red dashed lines. Due to the vertical exaggeration, the faults bounding the basin appear vertical but we have no constrains concerning their precise high angle. Bloc-diagramme schématique le long de la coupe EF coupant transversalement le complexe structural de Karatau (respectivement aux périodes du Riphéen en bas et du Permien au-dessus). Les failles bordières du bassin Riphéen sont dessinées verticales en raison de l’exagération verticale bien que nous ne connaissions pas précisément la valeur de l’angle.

3.2 Working hypothesis

The greater advancement of fold and thrust belt’s deformation front during compression (orogeny) in general can be caused by various reasons:

• Local presence of low friction basal décollement (represented by salts or clay), along which the masses of rocks ‘slip’ and advance more than others (for example, the territory of the Jura mountains in Switzerland and France (Davis and Engelder, 1985)).

• Existence of basement highs which can act as a barrier for moving orogen masses (for example, Peikang and Kuanyin basement highs in Taiwan (Hung et al., 1999)).

• Pre-existing structures at a high angle to compression which are reactivated with a strike-slip component (for example, the territory of the Sulaiman-Kirthar arcuate fold and thrust belt in Pakistan (Macedo and Marshak, 1999)).

• Change in décollement depth due to strong lithological heterogeneities or sedimentary cover thickness variation (for example the Northern Apennines in Italy (Pieri and Groppi, 1981)).

Based on Kamaletdinov’s (1974) and Brown’s (1999, 2006) ideas, it is assumed that formation of the Karatau structural complex was influenced by at least the factors 3 and 4 above. The Asha and Yuruzan faults where originally normal faults, reactivated as strike-slip faults during the Variscan compression. The décollement’s larger depth within the Karatau structural complex territory (≈6.5 km within the Karatau, while 3.5 km to the south-west and 6 km to the north-east) was acquired during the Upper-Proterozoic NE-SW extension (720–709 Ma, Fig. 6). In other words, the formation of the Upper Riphean rift basin within the KSC territory, oriented at a high angle to the subsequent compression, predetermined the advancement of this territory. To test this hypothesis, analogue sandbox modelling was carried out. The goal is to reproduce the peculiar geometry of the Karatau complex standing ahead of the regional deformation front of the Urals and described by our cross-sections of Figures 11 and 12. The models, in their initial state, correspond to the inherited structures at the beginning of the Variscan orogeny. The applied shortening corresponds to the Variscan compression.

4 Analogue sandbox modelling

4.1 Description of experiments

Nowadays, analogue modelling is widely used to study the dynamics of the lithosphere deformation at various spatial and temporal scales. The present experiments are part of the wide theme of the analogue modelling of orogens (Graveleau et al., 2012). We follow a simple, rather classic approach, using a rigid box with fixed side walls except the so called moving back wall that is translated towards the center of the box to produce a horizontal shortening of the material inside. We present here three experiments using three slightly different models. However, these three models are chosen among twenty five experiments exploring essentially the effects of the geometry of pre-existing normal faults. Top view videos of the remaining twenty two models are not shown here and are provided in the supplementary materials available on Zenodo (DOI: 10.5281/zenodo.15013162).

The chosen length scale is l* = 2.5 × 10⁻⁶, or 1:400,000 (10 mm in the model represent 4 km in nature). At this scale, rock cohesion may be neglected and dry sand, or other cohesionless granular materials, constitute adequate analogue materials of the brittle upper crust (Hubbert, 1951). There is no other scaling relationship to use. In particular, the time scale is not constrained because the materials used are not sensitive to the strain rate, and we do not add rate sensitive processes (like sedimentation or erosion for example).

We used a CV32 aeolian quartz sand (98% of quartz) with grain sizes between 150 μm and 425 μm (median 250 μm). This sand has a homogeneous density of 1697 ± 7 kg/m³ when deposited by sieving and its angle of internal friction drops from 44° (static friction) to 33° (dynamic friction) (Maillot et al., 2007). The same sand was dyed with oxides to serve as passive marker of deformation. To model weak décollement layers we used glass beads (median grain size of 100 μm and a friction angle of 12° to 20°). Sand and glass beads were sieved manually using a 600 μm sieve.

The inside of the box is 530 mm long and 437 mm wide (Fig. 14) representing a 212 × 175 km area centered on the Karatau complex (Fig. 8). The basal plate, moving back wall and front wall are made of PVC and the side walls of glass. The side walls were treated with a carbon based water repellent (“Rain-X”) before each experiment to reduce friction with the experimental material. Two rigid plates made of foam board are put on the base of the sandbox (light grey regions labelled "Left" and "Right" in Fig. 14c), leaving a deeper central part corresponding to the territory of the KSC. Based on the schematic cross-section EF at Permian time (Fig. 13), thickness of these blocks is 8.5 mm (3.4 km in nature), except in model 3, where the "Right" part is 3.5 mm (1.4 km in nature) thus creating an asymmetry due to a deeper décollement to the NE of the Yuruzan fault (about 6 km depth) than in the SW of the Asha fault where it is estimated at a depth between 3 and 4 km. The central trough represents a pre-existing ‘rift basin’ bordered by the Asha and Yuruzan normal faults that pre-date Paleozoic tectonics.

Figure 15 shows the initial inner structure of the three models. In model 1, the central and back regions are covered with a thin layer of glass beads that acts as a weak décollement. In model 2, the décollement is even weaker in the central region (glass beads on a vinyl film), but stronger in the back region (no glass beads: sand rests directly on the PVC). As there is no data on the nature of the mechanical resistance of the décollement level (we have proposed shale interlayers, Fig. 9), we tried various basal friction values through the use of various materials. Finally, model 3 varies from model 2 by a deeper décollement in the Right part (3.5 mm instead of 8.5 mm).

The back region represents mechanically the compressed and thrusted sedimentary materials present to the SE of the rift basin (recall the triple junction configuration prior to Variscan orogeny) when the deformation front of compression was about to reach the basin. For this reason, the region had a pre-existing relief in the form of a wedge, 210 mm in length (scaling to 84 km) and 70 mm (28 km) in sand thickness (resulting in a slope angle of α = 18.4°, Fig. 14b). That thick wedge is not intended to be geometrically realistic, but only mechanically realistic in that it slides on its base concentrating deformation at its frontal tip where the useful model lies. It may deform internally so that it transfers compressive stress in a more uniform manner than a rigid vertical back stop would.

The box is horizontal. The maximum displacement applied on the moving back wall is 123 mm, corresponding to a shortening of 49.2 km in nature (Model 3). Model 2 was stopped at 57 mm of shortening (22.8 km in nature), and Model 1 at 100 mm (40 km). The motor moves the back wall at the velocity of 0.5 mm/sec, with top view photographs taken every two seconds, at 1 mm displacement intervals (0.4 km in nature). From these top photographs we were able to calculate the displacement field of the surface of the model throughout the experiments. Following the completion of models 2 and 3, the sand was humidified through capillarity, and then cut parallel to the shortening direction to reveal cross-sections that were photographed.

Fig. 14 Experimental set-up: (a) Perspective view of the initial state of the experimental box. (b) Side view from the left side common to all experiments (c) Top view of the empty box showing the four regions (Left, Central, Right, Back) of the experimental box. Dispositif expérimental : (a) Vue en perspective de l’état initial de la boîte expérimentale. (b) Vue latérale du côté gauche commune à toutes les expériences. (c) Vue de dessus de la boîte vide montrant les quatre régions de la boîte expérimentale.

Fig. 15 Vertical sections in each of the four regions of the box showing the materials and markers used for a) model 1 (experiment E561 of the data base of Cergy laboratory), b) model 2 (experiment E564) and c) model 3 (experiment E573). The grey color indicates rigid foam board. Without vinyl film or foam board, the material is in contact with the box made of PVC. Coupes verticales dans chacune des quatre régions de la boîte montrant les matériaux et les marqueurs utilisés pour a) le modèle 1 (expérience numéro E561 dans la base du laboratoire de Cergy), b) le modèle 2 (expérience E564), c) le modèle 3 (expérience E573). La couleur grise indique un carton bulle rigide. En l’absence d’un film vinyle ou d’un carton, le matériau est en contact avec la base de la boite en PVC.

4.2 Experimental results

Model 1 is only illustrated with a top view of the final state (Fig. 16a). The surface trace of forethrusts (black lines) are roughly symmetrical along strike with respect to a line passing through the center of the box as expected from the symmetrical initial structure of the model. In the two side regions, there are six to eight thrust traces while there are only four in the central region (labelled 1 to 4). The thrusts formed in a normal, forward sequence, starting near the end of the initial wedge at the surface and the basement elevations (ramps) on the side regions. In the central region, thrusts numbered 1 to 4 propagate much more rapidly to the foreland and associated backthrusts are visible (probably associated to thrusts 1 to 3). The thrusts are more spread apart, forming longer thrust sheets. This is due to the deeper décollement level (and the resulting thicker sand layer) in the central part as is well known (e.g., Philippe (1995) or Macedo and Marshak (1999)).

On the side regions, the ramps imposed by the rigid basal plates have focused the thrusts near them, generating backthrusts that outcrop in the back region near the pushing wall. Transition between the side, central and back regions are marked by curved backthrusts, probably with a strike-slip component. Our major interest is however on the forethrusts geometry that exhibits a frontal advancement of the deformation front in the central region of 51 mm (20 km) compared to the left and right regions, and thicker thrust sheets marked by more distant outcrops. This compares well with the Karatau complex that advances the deformation front by approximately 20 to 25 km across the Asha and Yuruzan faults.

We continue with model 2 which, we recall, differs from the previous one by a stronger basal friction in the back region, a lower basal friction in the central region, and a total shortening of 57 mm (22.8 km). Again, surface trace of thrusts are symmetrical, with three thrusts on the side regions and only two, more spaced, in the center, that propagate further forward, showing a relative advancement of 32 mm (13 km) of the deformation front. Photographs of cross-sections in each region and their interpretation (Fig. 17a–c) show steeper thrusts and narrower thrust sheets in the side regions than in the central one, as expected from the difference of basal friction. They also shed more light on the structure of the backthrusts.

We conclude from these experiments that the existence of a lower basement décollement in the central part, with a lower friction, produces a deformation front that has some of the features of the Karatau region. In an attempt at capturing in addition the asymmetry between the south and north flanks of the Karatau region, we performed model 3^*, identical to model 2, but with a lower imposed ramp on the right (north) region as suspected from the SW-NE cross section reconstructed at Permian time (Fig. 13), and with a shortening comparable to that of model 1.

Top, final view of model 3 shows an advancement of the deformation front of 72 mm (29 km) of the central region with respect to the left (south) region, and 39.5 mm (16 km) with respect to the right (north) region (Fig. 16c). Cross-sections (Fig. 17d–f) also confirm that there are more numerous and more closely spaced thrusts in the left region (six thrusts), followed by the right region (five thrusts) and finally the central one (three thrusts): the deeper the décollement, the fewer and longer thrusts sheets form.

The distribution of shortening was estimated on photos of the cross-sections by restoring the black marker to horizontal straight lines (Fig. 17). Such restoration showed only 36–75% of total shortening which is equal to the moving back wall displacement (57 and 123 mm for models 2 and 3 respectively). The rest − 25–64%, comes from horizontal sand compaction and vertical thickening. In central regions, of lower basal décollement friction (Fig. 17b, e) (glass beads on a vinyl film in the center) most part of the shortening goes to folding and thrusting and less to sand compaction. Another important parameter was the ramp thickness: the lower it is (from 8.5 mm in Fig. 17f to 3.5 mm in 17d), the less shortening goes to sand compaction, as it plays a role of a rigid block in the direction of shortening.

The surface distribution of horizontal cumulative displacement in the direction of the moving back wall in models 2 and 3 (Fig. 18) was calculated from the correlation of successive top view photographs taken throughout the experiment, using PIVlab (Thielicke and Sonntag, 2021). As expected, displacement is null in the foremost region of the box that was left undeformed, and reaches the imposed amount of 57 mm and 123 mm near the back wall for the experiments 2 and 3 respectively. Between these regions, displacement progresses by jumps corresponding to the previously identified thrusts (Fig. 18a,c). The three profiles taken in the left, central and right regions (Fig. 18b,d) help identify the thrusts.

In the model 2, profile AB (Fig. 18b) shows two major accidents, i.e. a roughly flat zone of the curve, at 6 mm of cumulative displacement on the vertical axis, and a peak at 44 mm followed by a trough that corresponds to a backthrust (also visible on cross-section, Fig. 19b, yellow line). Profile IJ (right region) and GH (left region), are almost identical as they pass ramps of the same thickness (8.5 mm). Three steps at 4, 18, 34 mm of displacement corresponding to three thrusts are observed. Small disturbances on GH curve from 44 to 50 mm is due to the sand landslide (Fig. 16b).

In the model 3, profile AB (Fig. 18d) shows three major steps respectively at 10, 59 and 106 mm of displacement. Profile IJ (right region) shows four steps at 9, 30, 60 and 99 mm and those of profile GH (left region) are located at 15, 32, 50, 67 mm (the others not showing clear gradients of the profile). Looking at the abscissae of the profiles (Fig. 18d), we see that the steps of profile AB are located further ahead from the back wall than those of profile IJ, themselves slightly ahead of the profile GH, thus quantifying the observations made on Figures 16 and 17d, e, f.

Since we are interested in the regions ahead of the pre-existing topographic wedge, we compare the displacement at x equal to 227 mm and 220 mm (vertical dashed lines, Figs. 18b and 18d) from the front wall for models 2 and 3 respectively, corresponding roughly to the end of the slope of the pre-existing wedge. At that location, for model 2 displacement in the central region is 44 mm (17.6 km) while the side regions have absorbed only 20 to 22 mm (8 to 8.8 km). As for model 3 displacement in the central region is 106 mm (42.4 km) while at the side regions − 68 mm (27.2 km). This constitutes another observable to be compared to the structures of the Karatau region.

Fig. 16 Interpreted top view photographs of the final states of all three experiments: in (a) model 1, in (b) model 2 and in (c) model 3. The white dashed lines indicate the four regions of the model (Fig. 12c). The black lines with chevrons follow the outcrops of forethrusts while the yellow lines indicate the outcropping backthrusts. Total shortenings are given at the bottom right and shown with large grey arrows. The checkerboards have 10 × 10 mm squares. Photographies interprétées des états finaux des trois expériences : dans (a) modèle 1, dans (b) modèle 2 et dans (c) modèle 3. Les lignes tiretées blanches délimitent les quatre régions du modèle (Fig. 12c). Les lignes noires avec des chevrons indiquent les chevauchements et les lignes jaunes, les rétro-chevauchements. Les raccourcissements totaux sont indiqués en bas à droite et matérialisés par une grosse flèche grise. Les damiers ont des carrés de 10 × 10 mm.

Fig. 17 Photographs of cross-sections with interpretations of final states of Models 2 and 3. (a), (b), (c) cross-sections with interpretation of Model 2 in the Right part (corresponding to section IJ in Fig. 8), Central part (section AB in Fig. 8) and Left part (section GH in Fig. 8), respectively. (d), (e), (f), same legend, for Model 3. In each image, the dashed grey transparent line following the lowermost black marker is restored into a straight line in order to estimate the shortening accommodated by the various thrusts. This does not add up to the total imposed shortening. The missing shortening was accommodated as diffuse deformation in the sand. Estimated proportions are indicated in each case. Photographies de coupes transversales avec interprétations des états finaux des modèles 2 et 3. (a), (b), (c) coupes transversales avec interprétation de modèle 2 dans la partie droite, la partie centrale et la partie gauche, respectivement. (d), (e), (f), même légende, pour le modèle 3. Dans chaque image, la ligne pointillé transparente grise qui suit le marqueur noir le plus bas est restaurée en une ligne droite pour estimer le raccourcissement accomodé par les chevauchements. Ce raccourcissement n’est pas égal au raccourcissement total imposé. Le raccourcissement manquant est accomodé en déformation diffuse dans le sable. Les proportions estimées sont indiquées dans chaque cas.

Fig. 18 Final surface distribution of total horizontal displacement in the direction of shortening at the end of experiment 2 (a) and 3 (c). The total imposed displacement at the back wall is 57 and 123 mm, in experiments 2 and 3 respectively. Three mean displacement profiles for experiment 2 (b) and 3 (d) are averaged from the shaded regions labelled IJ, AB and GH in (a) and (c). Abscissae are counted in mm from the fixed front end of the box. Distribution de la surface finale du déplacement horizontal total dans la direction du raccourcissement à la fin de l’expérience 2 (a) et 3 (c). La moyenne de trois profils de déplacement moyen pour les expériences 2 (b) et 3 (d) est calculée à partir des régions ombrées étiquetées IJ, AB et GH dans (a) et (c).

Fig. 19 Comparison between geological (IJ, AB, GH, Fig. 11) and Model 3 cross-sections (Fig. 17d, e, f). They are superimposed by matching the deformation fronts of model and natural sections AB at the point indicated by a nail. Comparaison entre les coupes géologiques (IJ, AB, GH, Fig. 11) et les coupes modèle 3 (Fig. 17d, e, f). Elles sont superposées en ajustant les fronts de déformation du moèle et naturel de la coupe AB au point indiqué par un clou.

4.3 Discussion on analogue models and the Karatau region

The Karatau region constitutes a salient in the western front of the Urals fold-and-thrust belt bounded to the North and South by transfer zones. Transfer zones were extensively studied by Philippe (1995) with analogue models and examples from the Jura and Vercors Alpine massifs. Of interest to us are Philippe’s tests on lateral changes in basal friction only (Philippe, 1995, Figs. III-97 and III-126), lateral changes in the thickness of the sedimentary cover associated with changes in basal friction (Fig. III-36), lateral changes in amount of shortening or in shape of back stop (Figs. III-87 and III-89). The general conclusions are that the thickness of the cover and the mechanical weakness to shear of the basal surface are the most important parameters and promote thicker, longer thrust sheets and rapid advancement of the deformation front. Macedo and Marshak (1999) propose a systematic analysis of fold-and-thrust belt salients and classify them in terms of structures pre-existing the fold-and-thrust belt (oblique basins, structural highs (see also Farzipour-Saein et al., 2013), continental margin, strike-slip faults, variations of décollement material, indenter shape) that precise geological conditions in which cover thickness and décollement strengths may vary. In a series of analogue models they produce salients due to pre-existing basins with various shapes. Some of these models are very close to ours but remain qualitative regarding the specific setup of the Karatau region.

An efficient way to produce a salient consists in using a viscous material below the sedimentary cover of the salient. This can produce a salient bounded by strike-slip faults or at least narrow shear zones (Cotton and Koyi, 2000) as observed in the Karatau region. However, there is no evidence of any substantial layer of viscous rocks like salts or thick overpressured shales in the stratigraphy of the region revealed by the existing wells (although overpressures could well have disappeared since the Variscan compression).

These experimental studies and the available geological information on the Karatau region suggest that it should be possible to reproduce quite precisely our geological interpretations with models involving pre-existing normal faults perpendicular to the compression and the associated changes of depth of the décollement and of the sedimentary cover thickness. However, because the geological information is quite sparse (recall we only have a seismic line (IJ section), and one part of GH section (not published), plus the wells No 47, 1 LEU and 7 MSG on line IJ, and another one further south to help complete the log, and the surface geological maps.), we explored this question with twenty five models. These models allowed us to discard geometrical features and helped constructing our final cross-sections. We tried various geometries of the steps defining the central and right regions (width, heights, lengths). then we varied thickness of the sand, the initial wedge slope, the slope of the basal décollement (4.5°, by lifting the frontal end of the box), and the material in contact with the basal décollement to vary its strength. We set a décollement in the right region deeper than the central one, or a short central décollement that stops half way towards the front end wall. Videos of top views of these models are available in the electronic supplement.

The conclusions from these essays is that the KSC needs to be a graben defined by the Asha and Yuruzan faults, before compression. The other important ingredient is the relative weakness of the décollement within the KSC. The décollement dip does not affect substantially the results.

Among all twenty five models we tested, and the three models we present here, model 3 is not only the most relevant, but it can even be compared quantitatively to our interpretation of the Karatau region. In order to help comparison, we superimpose the cross-sections IJ, AB and GH (Fig. 11) to the equivalent cross-sections of model 3 (Fig. 17d–f) brought to the natural scale and using the most frontal thrust of the section AB as the reference point. All cross-sections are thus aligned without other adjustments (Fig. 19).

Using the same reference, we superimposed the geological map (Fig. 9) with the surface thrust traces of model 3 (Fig. 16c) to produce Figure 20. Using Figures 19 and 20, the comparison between the experimental outcome of model 3 and the geological structures of the Karatau region will focus on the following criteria: the position of the deformation front, the number of thrusts formed, their map and cross-section traces, and the horizontal shortening accommodated.

We first see (Fig. 19) that the fronts of deformation have the same tendencies: they match on sections IJ between model and nature, while section GH shows a natural front slightly ahead of the model front (recall that Carboniferous and Permian sediments buried thrusts to the south-west of the KSC, in particular the deformation front is actually situated near well No 47 which was added on Fig. 20).

The IJ cross-section has four to five thrusts (not all outcropping), matching the five thrusts of the experimental IJ section. The AB section has four thrusts within the KSC, and the experimental one, three. The GH section exhibits seven thrusts (most of them are buried), for six thrusts in the experimental equivalent. The number of thrusts are thus reasonably matching.

Regarding shortenings, we followed the procedure of Koyi et al. (2003) to distinguish between layer parallel compaction (that can be high in the sand models) and shortening associated to folding and faulting. We have restored the cross-sections of Figure 11 to obtain figures in the KSC and the neighboring regions (Tab. 1, column 2). The same restoration performed on the models cross-sections (Figs. 17d–f), yielded the shortenings displayed on Table 1, column 3. The total cumulative displacement obtained from PIV analysis of the surface displacement of model 3 is given in Table 1, column 4.

First, the IJ geological cross-section has a windy course following a seismic line. It is entirely in the foredeep territory where there is very little deformation. Two thrusts are located just to the south of the section (Fig. 20) so their shortenings do not appear on the section. This is why section IJ displays very little shortening (4.3 km). Section IJ of the models has a different course. It is straight and perpendicular to the shortening. Therefore the much greater shortenings in the model 3 section IJ (25.2–27.3 km), are probably more representative of the actual shortenings North-East of the KSC. The PIV surface cumulative displacements are always greater than those deduced from cross-section restoration. This is because they take into account all types of shortening, compaction included, while the other two columns of Table 1 account for fault slip and folding only. Upon comparison, we conclude that the model 3 does reproduce a greater displacement in the central region compared to the side regions, although displacements in the side regions are too high.

Finally, a few discrepancies should be outlined. First, since our models do not include redistribution of granular material by erosion and sedimentation, we did not include section CD in the comparison. The piggy-back basin (Sim trough labelled 6 on Fig. 8), the Suleymanovo anticline (labelled 5) and the Pervomay transpression zone (labelled 7) are not reproduced in our models. Syntectonic sedimentation would have promoted the formation of a long thrust sheet in front of the Suleymanovo anticline that our models cannot reproduce. More complex boundary conditions are also probably needed to account for the transpression zone. Second, the change in the basal step from 8.5 mm (models 1 and 2) to 3.5 mm (model 3) is not quite at scale with cross-section IJ: the step should be 1.2 mm (∼500 m) instead of 3.5 mm (∼1.4 km). Such a small step (1.2 mm) does not produce any differences in the thrusting sequence with respect to the central region without a step. Thirdly, we never obtained clear strike-slip faulting bounding the Karatau region as observed on the geological map. The ingredients to produce strike-slip faults in the models are basement highs (Macedo and Marshak, 1999) or very strong mechanical contrast of the décollement (Cotton and Koyi, 2000). Pre-cuts using dental floss in the sand above the lateral ramps delimiting the central regions were not sufficient to produce strike-slip faults.

Fig. 20 Superposition of the Karatau structural complex geological map (Fig. 8) and the interpretation of the final state of the experimental model 3 (Fig. 16c). Same legend as figures 8 and 16. Superposition de la carte géologique du complexe structural de Karatau (Fig. 8) et de l’interprétation de l’état final du modèle expérimental 3 (Fig. 16c).

5 Conclusion

The Proterozoic polyphase rifting plays a major role in the formation of the KSC. It has created the Upper Proterozoic sediment depocenter limited by normal faults on its future territory. Our cross-sections, constructed based on seismic profiles (in the vicinity of the KSC) and on surface geology (within the KSC) showed thickening of these sediments. We propose the Upper Riphean basal horizon as a local décollement level, which to south-east changes to the Lower Riphean in a classical flat-ramp geometry. This interpretation differs from a previous cross-section of the Southern Urals crossing the KSC made by Brown (1999, 2006) with a gradually deepening décollement from Middle Riphean to Lower Riphean.

The experimental outcomes of three analogue models (selected from twenty five models using different set-ups) support the hypothesis of the formation of the Karatau structural complex as a part of the Bashkir Megazone which produced a salient deformation front during the horizontal compression of the Variscan orogeny. The major features that caused the relatively greater advancement of this territory are the greater depth of décollement within the complex and pre-existing normal faults. The relative advancement of the deformation front, the number of thrusts and the shortenings accommodated in the Karatau complex relative to the NE and SW regions were reasonably reproduced by the modelling.

The absolute amounts of shortening were however less comparable and various geometric features are missing. This is because of the boundary conditions which precluded strike-slip oblique components or block rotations, and because of the absence of surface processes. All these ingredients would probably be necessary to reproduce the superimposed Sim trough (labelled 6 on Fig. 9), the Suleymanovo anticline (labelled 5) and the Pervomay transpression zone (labelled 7).

Acknowledgments

Anvar Farkhutdinov benefited from the financial support of CY Advanced Studies (Fellows-in-Residence program), and from the combined support of the PAUSE program of the Collège de France and the SFRI program of CY Cergy Paris Université. We thank Dominique Frizon de Lamotte and Geoffroy Mohn for inspiring discussions. We acknowledge the useful review by Hemin Koyi and the support of the editors of the BSGF.

References

All Tables

All Figures

Fig. 1 Map showing the different zones of the Urals and its geographic divisions (a). Geological map of the Southern and part of the Middle Urals showing the location of the URSEIS seismic profile (b). Main faults are: Tt − Tashly thrust, At − Alatau thrust, Zt − Zilmerdak thrust, Zf − Zuratkul fault, MUf − Main Uralian fault, Emf − East Magnitogorsk fault, Tf − Troisk fault. The box (c) indicates the area discussed in this paper (Adapted from Brown et al., 2011). Carte montrant les différentes zones de l’Oural et ses divisions géographiques (a). Carte géologique de l’Oural du Sud et d’une partie de l’Oural Moyen montrant l’emplacement du profil sismique URSEIS (b). L’encadré (c) indique la zone discutée dans cet article.

In the text

	Fig. 2 Interpreted line drawing of the depth migrated URSEIS reflection seismic profiles (Brown et al., 2008). Complete names of the faults are given in Figure 1. Tracé interprété des profils de sismique réflexion URSEIS migrés en profondeur.
In the text

	Fig. 3 Geodynamic and structural stages of the Urals (according to Puchkov, 2013, Maslov et al., 2010, Petrov et al., 2021 with modifications). Étapes géodynamiques et tectoniques de l’Oural.
In the text

	Fig. 4 Orogens of the East European Craton (according to Pease et al., 2021 with slight modifications according to Baluev, 2006 and Kuznetsov and Romanyuk, 2021). Orogènes du craton Est-Européen.
In the text

Fig. 5 Superimposed Riphean-Vendian sediments thickness (Masagutov, 2002) and major tectonic faults on aeromagnetic map of the Southern Urals (extracted from EMAG2v3 global magnetic grid (Meyer et al., 2017)). Complete names of the faults are given in Figure 1 description. Superposition de l’épaisseur des sédiments du Riphéen-Vendien et des principales failles tectoniques sur la carte aéromagnétique de l’Oural du Sud. Les noms des failles sont détaillés dans la légende de la Figure 1.

In the text

	Fig. 6 Approximate directions and ages of orogeny and rifting periods in the Karatau structural complex area (age is given according to Puchkov et al., 2007). Directions et âges approximatifs des périodes d’orogenèse et de rifting dans la région du complexe structural de Karatau.
In the text

	Fig. 7 Topographic map and digital elevation model of the Karatau structural complex territory (according to Shuttle Radar Topography Mission (30 m resolution)). Carte topographique et modèle numérique de terrain du territoire du complexe structural de Karatau.
In the text

	Fig. 8 Geological map of the Karatau structural complex territory (Knyazev et al., 2013 with modifications). Carte géologique du territoire du complexe structural de Karatau.
In the text

	Fig. 9 Representative stratigraphy of the Karatau structural complex (south-eastern part) (according to Puzhakov et al., 2019 with modifications). Stratigraphie représentative du complexe structural de Karatau (partie sud-est).
In the text

	Fig. 10 Part of the regional seismic line (Lozin, 2015) corresponding to the section IJ (Fig. 8 and 11) with interpretations. Partie de la ligne sismique régionale (Lozin, 2015) correspondant à la section IJ (Fig. 8 et 11) avec interprétations.
In the text

	Fig. 11 Cross-sections for Karatau structural complex geological map (location of the sections and legend in Fig. 8). Coupes transversales de la carte géologique du complexe structural de Karatau (localisation des coupes et légende dans la Fig. 8).
In the text

	Fig. 12 3D schematic representation of the Karatau structural complex geological cross-sections (location of the sections and legend in Fig. 8). Représentation schématique en 3D des coupes géologiques du complexe structural de Karatau (localisation des coupes et légende dans la Fig. 8).
In the text

Fig. 13 Schematic block-diagram along EF section crossing the strike of the Karatau structural complex (Permian and Riphean time). The late Variscan décollements are shown as horizontal thick red dashed lines. Due to the vertical exaggeration, the faults bounding the basin appear vertical but we have no constrains concerning their precise high angle. Bloc-diagramme schématique le long de la coupe EF coupant transversalement le complexe structural de Karatau (respectivement aux périodes du Riphéen en bas et du Permien au-dessus). Les failles bordières du bassin Riphéen sont dessinées verticales en raison de l’exagération verticale bien que nous ne connaissions pas précisément la valeur de l’angle.

In the text

Fig. 14 Experimental set-up: (a) Perspective view of the initial state of the experimental box. (b) Side view from the left side common to all experiments (c) Top view of the empty box showing the four regions (Left, Central, Right, Back) of the experimental box. Dispositif expérimental : (a) Vue en perspective de l’état initial de la boîte expérimentale. (b) Vue latérale du côté gauche commune à toutes les expériences. (c) Vue de dessus de la boîte vide montrant les quatre régions de la boîte expérimentale.

In the text

Fig. 15 Vertical sections in each of the four regions of the box showing the materials and markers used for a) model 1 (experiment E561 of the data base of Cergy laboratory), b) model 2 (experiment E564) and c) model 3 (experiment E573). The grey color indicates rigid foam board. Without vinyl film or foam board, the material is in contact with the box made of PVC. Coupes verticales dans chacune des quatre régions de la boîte montrant les matériaux et les marqueurs utilisés pour a) le modèle 1 (expérience numéro E561 dans la base du laboratoire de Cergy), b) le modèle 2 (expérience E564), c) le modèle 3 (expérience E573). La couleur grise indique un carton bulle rigide. En l’absence d’un film vinyle ou d’un carton, le matériau est en contact avec la base de la boite en PVC.

In the text

Fig. 16 Interpreted top view photographs of the final states of all three experiments: in (a) model 1, in (b) model 2 and in (c) model 3. The white dashed lines indicate the four regions of the model (Fig. 12c). The black lines with chevrons follow the outcrops of forethrusts while the yellow lines indicate the outcropping backthrusts. Total shortenings are given at the bottom right and shown with large grey arrows. The checkerboards have 10 × 10 mm squares. Photographies interprétées des états finaux des trois expériences : dans (a) modèle 1, dans (b) modèle 2 et dans (c) modèle 3. Les lignes tiretées blanches délimitent les quatre régions du modèle (Fig. 12c). Les lignes noires avec des chevrons indiquent les chevauchements et les lignes jaunes, les rétro-chevauchements. Les raccourcissements totaux sont indiqués en bas à droite et matérialisés par une grosse flèche grise. Les damiers ont des carrés de 10 × 10 mm.

In the text

Fig. 17 Photographs of cross-sections with interpretations of final states of Models 2 and 3. (a), (b), (c) cross-sections with interpretation of Model 2 in the Right part (corresponding to section IJ in Fig. 8), Central part (section AB in Fig. 8) and Left part (section GH in Fig. 8), respectively. (d), (e), (f), same legend, for Model 3. In each image, the dashed grey transparent line following the lowermost black marker is restored into a straight line in order to estimate the shortening accommodated by the various thrusts. This does not add up to the total imposed shortening. The missing shortening was accommodated as diffuse deformation in the sand. Estimated proportions are indicated in each case. Photographies de coupes transversales avec interprétations des états finaux des modèles 2 et 3. (a), (b), (c) coupes transversales avec interprétation de modèle 2 dans la partie droite, la partie centrale et la partie gauche, respectivement. (d), (e), (f), même légende, pour le modèle 3. Dans chaque image, la ligne pointillé transparente grise qui suit le marqueur noir le plus bas est restaurée en une ligne droite pour estimer le raccourcissement accomodé par les chevauchements. Ce raccourcissement n’est pas égal au raccourcissement total imposé. Le raccourcissement manquant est accomodé en déformation diffuse dans le sable. Les proportions estimées sont indiquées dans chaque cas.

In the text

Fig. 18 Final surface distribution of total horizontal displacement in the direction of shortening at the end of experiment 2 (a) and 3 (c). The total imposed displacement at the back wall is 57 and 123 mm, in experiments 2 and 3 respectively. Three mean displacement profiles for experiment 2 (b) and 3 (d) are averaged from the shaded regions labelled IJ, AB and GH in (a) and (c). Abscissae are counted in mm from the fixed front end of the box. Distribution de la surface finale du déplacement horizontal total dans la direction du raccourcissement à la fin de l’expérience 2 (a) et 3 (c). La moyenne de trois profils de déplacement moyen pour les expériences 2 (b) et 3 (d) est calculée à partir des régions ombrées étiquetées IJ, AB et GH dans (a) et (c).

In the text

Fig. 19 Comparison between geological (IJ, AB, GH, Fig. 11) and Model 3 cross-sections (Fig. 17d, e, f). They are superimposed by matching the deformation fronts of model and natural sections AB at the point indicated by a nail. Comparaison entre les coupes géologiques (IJ, AB, GH, Fig. 11) et les coupes modèle 3 (Fig. 17d, e, f). Elles sont superposées en ajustant les fronts de déformation du moèle et naturel de la coupe AB au point indiqué par un clou.

In the text

	Fig. 20 Superposition of the Karatau structural complex geological map (Fig. 8) and the interpretation of the final state of the experimental model 3 (Fig. 16c). Same legend as figures 8 and 16. Superposition de la carte géologique du complexe structural de Karatau (Fig. 8) et de l’interprétation de l’état final du modèle expérimental 3 (Fig. 16c).
In the text