Understanding the Deformation Structures and Tectonics of the Active Orogenic Fold-Thrust Belt: Insights from the Outer Indo-Burman Ranges

Understanding the structural evolution of the outer Indo-Burman Rages (IBR) is crucial to reconstructing the Cenozoic geodynamics of the Bengal Basin and the complex development of collisional events between the Indian and Burmese plates [1]. The plate boundary region between the India and Burma plates in the outer IBR region consists of fold belts, and regional scale thrust (Figure 1) is a prominent earthquake-prone area, where deformation and kinematics are governed by oblique subduction of the Indian Plate beneath the Burma plate [1–6]. This oblique convergence-related transpressional tectonics in orogenic belts is a common phenomenon giving rise to complex deformation patterns (e.g., [7–11]). Transpressional deformation involves both strike-slip and dip-slip components. However, the strike-slip deformations can deviate from simple shear when shortening oblique to the deformation zone [10]. At the obliquely convergent plate boundaries, deformation occurs both locally and regionally and produces a range of deformation structures [9, 12]: folds, thrusts, and shear zones, which are essential for oil and gas exploration [13, 14] and active tectonic studies [15, 16].

Tectonic deformation of the western outer edge of the young and active IBR, known as the Chittagong Tripura Fold Belt (CTFB), is mainly attributed to the oblique India-Burmese convergence. This oblique convergence-related transpressional tectonics splits along different tectono-geomorphic units separated by large-scale thrust and dextral strike-slip faults and produces a fold-thrust belt, in which the western limit is marked by the “Deformation Front” (Figure 1) [2, 17]. The convergence is accommodated by different tectonic structures where the deformation intensity and age gradually decrease from IBR in the east to the CTFB in the west. According to Maurin and Rangin [18], a total of about 11 km east-west shortening occurs in the last 2 Ma, indicating ~5 mm/year long-term shortening rate. Two regional scale active faults and thick-skinned tectonics of the CTFB constantly accommodate the stress associated with the Indo-Burma oblique subduction and, therefore, are responsible for the kinematic evolution of the CTFB [18, 19]. The two regional faults are the Chittagong Coastal Fault (CCF) to the west and Kaladan Fault to the east, and these faults are running approximately N-S (Figure 1). The CCF is approximately a coast parallel structure where an overthrusting hanging wall concealed major parts of the western flank of the Sitakund, Inani, and Dakhin Nhila anticlines (Figure 2). Recent geodetic measurements suggest the tectonic convergence with overall E-W shortening in the CTFB area occurred at the rate of ~5 mm/year [3, 5, 19]. This east-west shortening produces approximately a north-south complex fold and overlapping thrust systems in the CTFB area. The plunging fold system consists of a series of NNW-SSE trending anticlines and synclines, where the geometry is predominantly expressed by NNW-SSE striking reverse faulting [20]. Moreover, the presence of active macro-gas seepages in the CTFB area is reported and encountered during the fieldwork. The presence of such seepages suggests hydrocarbon accumulation in the subsurface [21]. Therefore, the proper understanding of the deformation structures (e.g., fold and fault) of the study area will facilitate having a clear insight regarding the hydrocarbon system of the area.

The folds, faults, joints, fractures, and their geometry can be used to interpret the regional deformation and kinematic history. Fractal dimensional analyses of these structures over 2D space can reveal their formation processes and kinematic evolution [22–25]. For example, the lineament/fracture network revealed from the Advanced Land Observing Satellite (ALOS) images is the imprint of two-dimensional features in the ground surface [26, 27]. Rocks exhibit anisotropy when subjected to stress, and the resultant cracks/fractures are assumed to have preferential orientation [28–31]. The study of such fracture networks, their size, and their spatial relationships are important to establish the regional stress field, rock strength and its deformability [32], and kinematic model [33, 34]. In this study, the fractal dimension anisotropy of the fracture system (i.e., fracture spacing and orientation in a given direction) has been analyzed to understand the spatial distribution of the fractures and associated stress field.

Understanding the evolution of geological structures and their dissemination with progressive deformation is the fundamental aspect in inaugurating the tectonics of any region [35, 36]. This leads to a better appraisal of the state of paleostress conditions that led to the deformation on account of regional tectonics. Therefore, it becomes essential to study the paleostress condition in accordance with the kinematics of the structures. The previous studies on CTFB have only been restricted to geochemistry and provenance studies [37–41]; seismic hazards [2, 4–6, 19, 42–44]; and geodynamics, tectonics, and geophysical investigations [1–3, 18–20, 42, 45, 46]. Except Betka et al.’s [17] slip-partitioning study that carried out locally in Tripura, no attempt has been made to comprehensively decipher the deformation structures, related kinematics and stress regime, and balanced geological cross-sections of the CTFB’s Tertiary sediments and amount of strain in terms of field investigation. The present paper aims to fill this lacuna by presenting a complete structural map of the CTFB area of the outer IBR (Figure 2) for the first time based on several comprehensive geological fieldwork carried out since 2008 and other published maps and data sets ([2, 20, 47] and references there in). It also provides evidence for structural deformations and interprets their kinematics in the exposed Tertiary sedimentary rocks in the area. Structural interpretations are then constrained by in-depth analysis of the folds, faults, and fractures/lineaments geometry to understand the prevailing stress regime in the area.

Three balanced geological cross-sections from the northern, central, and southern parts of the study area illustrate the geometry and kinematics of the structures developed in the exposed Tertiary sediments and the total amount of strain. The fault-slip analysis reveals the paleostress direction that led to the formation of deformation structures in the CTFB. The tectonic evolution of the CTFB in response to the oblique convergence along the Kaladan Fault and CCF is also discussed. This study helps characterize the fault, lineament/fracture disposition patterns in CTFB and provides a clearer understanding into the deformation kinematics and tectonics of a young and active orogenic belt.

Extending from the northern end of the Arakan Trench in the south to the juncture of the Dauki, Halflong, Naga, and Disag thrusts to the north, the CTFB has been recognized as the outer edge of the western Neogene belt of the Indo-Burman Ranges (IBR), which separated from the eastern Palaeogene belt of the IBR along the Kaladan Fault [17–19]. Geologically, the CTFB is bounded by the Foredeep basin part of the Bengal Basin to the west and the IBR to the east and considered to be an active orogenic belt in the world (Figure 1). To the east, it is structurally separated from the IBR by the approximately N-S running Kaladan thrust (also known as Tut Fault). To the west, the boundary of the CTFB with the Foredeep basin is assumed to be along the edge of the convex shape “Thrust Front” (Figure 1) [17, 20], represented by the CCF. The CCF is a regional thrust fault, running NNW-SSE along the northeastern coastline of the Bay of Bengal (Figures 1 and 2). According to Maurin and Rangin [18] both the Kaladan and CCF have a dextral-slip component. In between the CCF and the Kaladan Fault, this ~100 km wide deformation/transpressional zone (i.e., the CTFB) is framed by en echelon plunging folds and faults [20].

Before the convergence of the India-Burma plates, the CTFB area was an integral part of the Foredeep basin of the Bengal Basin [48]. The commencement of the CTFB development in the eastern part of the Foredeep basin is mainly presumed to be started from the Pliocene as an accretionary wedge due to the incessant subduction of the Indian Plate’s oceanic crust beneath the Burmese Plate [4, 18, 39, 41]. With continued collision, the eastern part of the Foredeep basin evolved as a fold belt and form folded flank of the Bengal Basin, currently known as the CTFB (Table 1). Since last 2 Ma, the rapid development of the CTFB and its westward propagation is facilitated from upliftment of the Shillong Plateau and consequent deepening and sedimentary filling of the Sylhet Trough (Figure 1) [18]. Due to this rapid westward propagation since Pliocene, thin-skinned tectonics of the CTFB is gradually engraved westward by thick-skinned tectonics of the IBR [18, 49]. Therefore, the deformation age of the structures gradually increases eastward from the CTFB to the IBR.

Based on the geometry and deformation intensity of the folded structures, field observation, Shuttle Radar Topography Mission (SRTM) image analysis, and published data set, the CTFB area has been divided into two tectonic zones (Figure 2): (a) western fold-thrust zone and (b) eastern highly compressed fold-thrust zone [20]. To the west, subduction-related deformation continues from the CTFB towards the current Foredeep basin, but the intensity of the deformation gradually decreases from the east to the west and ceases along the Deformation Front. The transitional boundary of these two tectonic units is marked by the “Thrust Front” [17]. The continued deformation from the Deformation Front to the Thrust Front produces an approximately 80 km wide young subsurface fold belt, which is known as the Emerging Fold Belt [20].

The CTFB area consists of late Miocene–Quaternary accreted sediments of ~5 km thickness (Figure 2), where the sedimentary facies ranges from shallow marine through deltaic to fluvial environments [40, 48]. Within the CTFB, the Miocene Surma Group consists of shallow marine shelfal and intertidal deposits and is divided into two formations: Bhuban at the bottom and Boka Bil at the top (Table 2). Overlying Boka Bil Formation, the Tipam Group comprised of Pliocene fluvial deposits and is divided into two formations: Tipam Sandstone at the bottom and Girujan Clay at the top. Above the Tipam Group, the Pliocene–Pleistocene Dupi Tila Group consists of meandering channels, floodplain, and alluvial deposits [2, 40, 48].

In the CTFB (i.e., the western prolongation of the Indo-Burman accretionary prism), the upper deltaic sediments of the Surma, Tipam, and Dupi Tila groups occur (~up to 5 km) above a major basal detachment fault and deformed [42]. In the last 2 Ma, these deformations rise to a series of elongate, ~NNW–SSE trending curvilinear plunging folds, indicating the combination of thick-skinned and thin-skinned tectonic processes [2, 17, 18]. These folds are arranged in a set of alternating low relief ridge-forming fault-cored anticlines and valley-forming synclines. The anticlines are continuous along-strike for tens to more than 100 km. Moreover, the geometry of the CTFB fold ridges is attributed to the free westward migration of fold, longitudinal thrust, and multi-phase deformation of the Tertiary sediments [2, 46].

The study connects the outcrops-scale surface geology and fault kinematics data with large-scale kinematics and tectonics to determine the overall geometry and kinematics of the CTFB. The bedding attitude data (from fifteen anticlinal structures), lineament/fracture data (from two structures), fault-slip data (from four structures), exposed deformation structures, and stratigraphic and lithologic data from the CTFB were measured and recorded. A geologic map of the CTFB area that defines the outer belt was constructed (Figure 2). Outcrops are sparsely exposed due to ever-increasing settlement, heavy vegetation cover, and rapid erosion. Field data were collected only from the road-cut, river, and lake outcrops along transects that cross each anticline during several fieldwork conducted in the last decade. Lithologic and fault contacts detected along each transect were extrapolated manually on both sides using a hillshade digital elevation model based on the SRTM images (ver. 3.0 with 1° × 1° tiles at 1 arc second, i.e., 30 m resolution). Secondary tectonic and structural data sets from the published maps are digitized, scaled, and integrated with the primary field data to get the total coverage of the study area.

To quantify the fold belt geometry, we have analyzed the bedding attitude data (total $n=1356$⁠) individually for each anticlinal structure and all of them together (Figures 3 and 4). The orientations of the studied structure were determined using a cylindrical best fit to poles to bedding using the software Stereonet 11 (Figure 3) [50]. Determination of the bedding to mean pole follows Right Hand Rule (RHR), and pi axis orientation (trend and plunge) of each segment was determined using the same software.

The primary data set used to detect fold-related fracture (i.e., lineaments) are ALOS, SRTM-based digital elevation models (DEMs), and Google Earth. After initial processing of the ALOS images (e.g., spatial enhancement and hillshade) in ArcGIS 10.3 environments (Figures 5(a) and 5(d)), lineament/fracture demarcated, and digitized manually through visual interpretation to avoid the manmade features (e.g., road and foot track). During manual extraction, the utmost care has been given to maintain the same scale/zooming level. Features more than 500 m in length and having structural significance such as fracture, faults, abrupt changes of river courses, and linear scarp faces were digitized as lineament/fracture and stored in a geographic information system (GIS) database. For the comparison and necessary corrections of lineaments obtained by remote sensing, the resulting lineaments maps (Figures 5(b) and 5(e)) were cross-checked with the geological map, topographic map, SRTM-based DEM, Google Earth, and field observations to remove roads, lithological boundaries, and any other linear artifacts. After all these processing and ground checks, a total of 483 and 345 lineaments/fractures were mapped for the Sitakund and Sitapahar structures, respectively (Figures 5(b) and 5(e)).

We used the AMOCADO software [51] to analyze the possible relationship between the extracted fracture/lineament (Figures 5(b) and 5(e)) anisotropy with the stress filed of the area. AMOCADO is based on the modified Cantor dust method [52], where a set of scanlines (parallel) are superimposed on the lineament/fracture pattern. The number of segments Ns that intercept the lineament/fracture and segment-length s are automatically measured and plotted as a log cumulative Ns vs. s in the graph to determine the slope m (i.e., fractal dimension). Consequently, the scanline set is rotated in 1° incremental steps up to 180° to produce the complete result. Here, the AMOCADO method [51] is used to quantify the fractal dimension anisotropy (insets in Figures 5(b) and 5(e)) of the extracted fracture/lineament system of the Sitakund and Sitapahar anticlines. The FracPaQ, a MATLAB™ open-source toolbox [53], was used to quantify fracture/lineament patterns (Figures 5(b) and 5(e)), including their distributions and spatial variation (Figures 5(c) and 5(f)).

The rose diagrams (Figure 6) of the extracted lineaments were plotted and compared with the general trend of the CTFB. The total fractures obtained from the remote sensing technique were divided into different sets on the rose diagrams according to their angle and direction relative to the fold axis [54–56]. If a fracture set has a trend parallel with the fold axial trend (i.e., within 0-15° of the fold axial trace), it is defined as the “axial fracture set.” On the other hand, if a fracture set is perpendicular to the fold axial trend (i.e., 80-90° from the trend of the related fold axis), it is defined as a “cross-axial fracture set.” In this study, the axial fracture set is detonated by S1 (Figure 6(e)). Fractures having greater angles (i.e., >15° and <80°) with the trend of the fold axis were called an “oblique fracture set.” Two oblique fracture sets defined with S2 and S3 (Figures 6(d) and 6(f)) were identified in this study and are presented in Figures 5 and 6, respectively. The angle of each set (i.e., axial, cross-axial, and oblique fracture sets) was carefully compared with the fold axial trace. Finally, the bisector of the two oblique fracture sets was regarded as parallel to subparallel of the maximum shortening direction (i.e., σ₁) [57, 58].

Kinematic analysis of the mesoscale deformation structures (Figure 7) was performed based on field geological investigations, oriented field photographs, geometric measurements of the bedding and fault-slip data (e.g., rake, displacement), and their mutual relationship with respective anticlinal structures of the CTFB.

To evaluate the structural pattern and shortening of the Tertiary sediments in the outer belt of the Indo-Burma Ranges, we constructed a balanced cross-section based on our field data and seismic profiles of the study area from the previous literature [18, 39, 42, 59]. Three probable balanced sections (Figure 8) are drawn in the northern, middle, and southern part of the study area along AA$′$⁠, BB$′$⁠, and CC$′$ traverse, respectively. In these profile sections, the lines are drawn perpendicular to the general trend of the fault planes along the ENE-WSW direction. Here, the cross-section lines of the formation layers and fault planes are primarily oriented perpendicular to the principal direction of shortening. Apparent dips of each layer and fault planes were calculated from the field data and projected along with strikes on the balanced cross-section. To understand the subsurface structure, available seismic sections are used, and the surface data is projected downward using the kink methods [60, 61]. Basal detachment fault is identified from the different seismic sections [42]. To construct the cross-section, it is assumed that there was no transport of material; hence, the volume was maintained during the deformation as the domain suffered major tectonic deformation after the deposition of Tipam Formation.

All fault-slip data (orientations of fault planes and slip vectors) recorded during the field works in the four different structures (Bandarban, Sitapahar, Gobamura, and Dakhin Nhila) were used for paleostress determination (Figure 9). Common kinematic indicators such as Riedel shears, shear bands, and bedding offset (generally well preserved ∼0.1–1 m of displacement between fault blocks) were used to decipher the sense of slip.

Following Angelier [62], a total of 46 fault data are collected from the study area and classified as an oblique-slip thrust fault. There are several methods suggested for paleostress investigation using fault-slip data (e.g., [62–79]). The paleostress analysis using fault-slip data can be done using kinematic or dynamic methods based on the following assumptions [67, 77, 80]: (1) The bulk state of stress is assumed to be constant, and any displacement on the fault planes are independent; (2) the slip on the fault plane ensues the direction of resolved shear stress maximum under a certain state of stress condition (Wallace–Bott Hypothesis); and (3) the faults are believed to be homogeneous and formed by the same tectonic event [62, 67, 77, 81]. We resolved the paleostress condition in this study, using fault-slip data collected from the studied region by Right Dihedron methods. Although some of the fault planes show a wide range of strike orientations, most of them are oriented along the NNW-SSE (maxima) or NNE-SSW (submaxima). The collected fault data sets are arranged into homogeneous data subsets using the software program “Win_Tensor” (version 5.8.6; [82, 83]). This process eliminates all probable heterogeneities from the initial data. All the data are presented in a single data set, and the initial separation has been completed using the Right Dihedron method. Further, the data is filtered based on stress ratio (R), stress axes orientation, and symmetry of the measured sets. Following Delvaux and Sperner [83], the best possible data are recognized with a low value of “counting deviation” and “nominal counting” values of 0 and 100 for σ₁ and σ_3, respectively. The detailed result of paleostress analysis is presented in Section 4.5.

Bedding attitude measurements (⁠$n=1356$ data points) of 15 anticlines show almost consistent NNW strike with variable dip amounts (Figure 3). Only a few of these measurements are shown in Figure 2 due to limited space. Table 3 summarizes the main geometrical features of the fifteen anticlines from the two zones of the CTFB area, and Figure 3 represents the density contour of the poles to bedding with a cylindrical best fit line axial plane and fold axis for each studied anticline of the CTFB area. Figure 4(d) presents plots of fold interlimb angle vs. axial plane dip. However, for the eastern highly compressed zone, plunges vary from 0.9° to 5.3°, and for the western fold-thrust zone, plunges vary from 0.3° to 8.5°. The standard deviation of the trend and plunge of the fold axis is 7.93° and 2.21°, respectively, (Figures 4(e) and 4(f)). Out of fifteen anticlines (Figure 3), about a half (seven) plunges to SSE, while another half (eight) plunges to NNW.

The dip of the axial surface of the studied anticlines varies from 68° to 89°. For the eastern highly compressed fold-thrust zone, the dip of the axial plane varies from 76° to 88°, and for the western fold-thrust zone, the axial plane dip varies from 68° to 89°. The standard deviation of the axial plane dip is 6.00. Out of fifteen studied anticlines, six of their axial planes dip to the east, whereas the remaining axial planes dip to the west (Figure 3, Table 3). Interestingly, the axial plane of most of the anticlines of the eastern highly compressed zone is dipping to the west except one, whereas in the western fold-thrust zone, both west and east-dipping axial planes have been observed.

The interlimb angles of the analyzed folds show a distinct difference between the eastern highly compressed fold-thrust zone and the western fold-thrust zone (Figures 4(e) and 4(f)). Generally, the folds in the western zone show a larger interlimb angle (average is 125°), whereas folds in the eastern compressed zone have a smaller interlimb angle (average is 97°). However, one fold from each zone (Utan Chatra, eastern zone; Sitakund, western zone) shows different results (Table 3).

A combined analysis of the entire bedding attitude measurements (⁠$n=1356$⁠) of the CTFB area shows an axial plane orientation (Figures 4(a)–4(b)) and strike trend (Figure 4(c)). A weak linear relationship has been revealed based on graphical analysis of the fold interlimb angle vs. an axial surface dip of the same folds (Figure 4(d)), except for Barkal, Kasalong, and Gobamura. Shear strain, γ, has been calculated for the fifteen studied folds of the CTFB area based on the following equation: $γ=tan90°−α$⁠. Here, $α$ is the dip of the axial plane [84]; the result is reported in Table 3 and shows a considerable variation of the shear strain from 0.009 to 0.41 (Supplement 1, Figure S1). The highest value (0.41) has been calculated for Sitakund Anticline, whereas the lowest value is calculated for Semutang Anticline, and interestingly, both are situated at the western fold-thrust zone.

Elongated folds dominate the structural geology of the CTFB with minimum aspect ratio (i.e., length of the hinge line divided by the width of the fold) of more than 3 (Figure 2). Fold-related fractures/lineaments have been analyzed from the two anticlines within the western fold-thrust zone and had a geometrical relation to the fold elements. For both anticlines, three distinct fracture/lineament sets (S1-S3) were identified. These fracture/lineament sets can be attributed to syn-folding and/or post-folding regimes. The chronological and geometrical studies can reveal the relation between fractures and the fold [85, 86]. The fold-related fractures are visualized using the rose diagram (Figure 6).

Based on these diagrams and following the RHR, for Sitakund anticline, S1 fractures include the NNW-SSE fracture set with the mean direction of N155°±5, the S2 fractures include NE-SW fracture set with the mean direction of N45°±5 and S3 fractures with almost E-W orientation with the mean direction of N95°±5 (Figures 5 and 6). For Sitapahar anticline, S1 fractures include the NNW-SSE fracture set with the mean direction of N155°±5. The S2 fractures include NE-SW fracture set with the mean direction of N37°±5 and S3 fractures with almost E-W orientation with the mean direction of N90°±5 (Figures 5 and 6). With respect to the Sitakund and Sitapahar fold axial trace, these fractures are divided into two major groups: (a) axial fractures S1, which are parallel to the fold axial trace, and (b) oblique fractures S2 and S3 with an angle of 120° and 65°, respectively, relative to the fold axial trace (Figure 6). The acute angle (α) of the two intersecting S2 and S3 fracture sets is 50°±5.

The results of the AMOCADO analysis of the fracture system are presented in Figure 5 and Table 4. The axial ratios (i.e., a/b) of the fractal dimension anisotropy ellipse, which represents anisotropy intensity for the fracture system of the Sitakund and Sitapahar anticlines are found to be 1.16 and 1.27, respectively. The azimuthal trends of the major axis (a) for Sitakund and Sitapahar anticline fracture systems are found to be 82° and 79°, respectively, which are approximately orthogonal to their respective anticline trend (Table 4).

Among the two broad zones of the CTFB, brittle (fault, joint, and cleavage) and brittle-ductile (boudins and shear bands) deformations are dominant in the eastern highly compressed fold-thrust zone compared to the western fold-thrust zone. Only compressional/transpressional deformation has been observed in the two deformation zones of the CTFB. The observations from field and satellite images on different scales suggest a fold parallel to subparallel thrust with dominant NNW-SSE to NNE-SSW trend (Figure 7).

Both the east- and west-dipping thrusts at different scales with highly variable dipping angles are omnipresent in the study area (Figure 7). Observed displacement along the fault plane ranges from a few cm to tens of m. Evidence of synthetic (Figure 7(a)) antithetic (Figure 7(b)) thrusts, thrust-shear zone (Figure 7(c)), fault propagating kink fold (Figure 7(e)), and thrust fault with dextral shear (Figure 7(f)) are present in both deformed zone of the CTFB, but dominantly to the eastern zone. The age and relief of the exposed Tertiary rocks increase towards the east, and the oldest exposed formation in the study area is the Bhuban Formation, mainly exposed at the eastern anticlines of the CTFB. However, the exceptions are the Sitakund, Inani, and Dakhin Nhila anticlines, whose western flanks are underthrusted by the east-dipping and west verging CCF (Figure 2).

The field data sets that were used to generate the balanced cross sections are presented on the map (Figure 2) and stereographic projections (Figure 3), accompanied by a description of field observations in Table 3 and a field photo showing the structural relationship in Figure 7. Sections are drawn parallel to the principal direction of shortening to estimate the shortening of the Tertiary sediments. Three ENE-WSW oriented balanced cross-sections from the northern (AA$′$⁠), middle (BB$′$⁠), and southern part (CC$′$⁠) of the study area (Figure 2) were constructed to examine the tectonic features on the surface and the subsurface across the CTFB (Figure 8 and Figure S2 of Supplement 2). Sections are drawn parallel to the principal direction of shortening to estimate the shortening of the Tertiary sediments (Figure 8(d)). Here, most of the anticlines are bounded by single thrust fault on both sides except the Sitakund anticline (Figure 8(a)). The western flank of the Sitakund anticline is affected by imbricate thrust faults. Therefore, the Sitakund anticline is more tightly folded than the other folds in the western fold-thrust zone. All the anticlines show gentle to open fold in the surface, and the interlimb angle decreases with depth as well as towards the east. The shortening estimates in the northern part of the study area along AA$′$ are ~16 km for the 94 km section. In the central part along BB$′$ section, the shortening is ~4.2 km for the 68 km section. In the southern part along CC$′$ section, the estimated shortening is ~1.5 km for the 7.5 km section. The percentages of shortening of the Tertiary sediments in AA$′$⁠, BB$′$⁠, and CC$′$ sections are ~15%, ~6%, and ~16.7%, respectively (Supplement 3, Figures S3 and S4).

The stress tensor obtained by processing all fault data from CTFB gives a NE-SW to ENE-WSW directed compression. In Figure 9, a relative position of the principal stress axes and stress ratio (R) is provided (also see Supplement 4, Tables S1 and S2). It may be noted that the maximum principal stress axes in all locations are found to be horizontal, while the minimum principal stress axes are vertical, consistent with a compressive environment. This suggests that all the faults recorded in the study area result from thrust due to NE-SW to ENE-WSW directed compression. The fault-slip analysis also helps to analyze the tectonic stress regime, including any variation in the stress regime. The stress regime is determined with the help of the stress regime index (R$′$⁠) invoked in the program “Win_Tensor.” The R$′$ is related to $R$ values depending on the relative orientations of the principal stress axes and includes a range of values. Delvaux et al. [87] have quantified the R$′$ values in detail for various stress regimes. These stress regimes are generally recognized as radial extension, pure extension, transtension, pure strike-slip, transpression, pure compression, and radial compression [72, 83]. The stress regimes for all faults have been evaluated based on the R$′$ value and the principal stress orientations (see Supplement 4, Table S1). The stress regime indicates that in all the locations, the faults lie within the pure compressional domain.

The western Indo-Burman Ranges (i.e., CTFB) has developed an exemplary linear NNW-SSE oriented doubly plunging fold belt. These folds are arranged in a set of alternating valley-forming synclines and ridge-forming anticlines [2, 88]. To the west, the surficial expressions of the CTFB folds are limited chiefly along with the Chittagong Coastal Fault (CCF).

The bedding attitude analysis (Figure 3) reveals that the axial planes of the eastern highly compressed zone anticlines are dipping primarily to the west whereas the axial planes in the western fold-thrust zone dipping both east and west. The larger interlimb angles were observed in the western zone (average 125°) compared to that of the eastern compressed zone (average 97°) except for Utan Chatra and Sitakund anticlines (Table 3; Figure 3). A weak linear relationship exists between fold interlimb angle and axial surface dip of the same folds (Figure 4(d)), except for the Barkal, Kasalong, and Gobamura, which are bent down below, possibly due to proximity to the Kaladan Thrust. Although the calculated shear strain (⁠$γ$⁠) of the CTFB area shows a small variation (Table 3), anticlines in the eastern zone have a relatively higher shear strain than the western zone. Based on the interlimb angle and shear strain of the studied anticlines, it is evident that the intensity of the folding in general decreases towards the west: the direction of the fold belt propagation (Figures 4(e) and 4(f)). The exceptional values for the few anticlines in the western zone are possibly related to the deformation caused by CCF. The CCF is a major boundary thrust known as “Thrust Front” that separates the “Emerging Fold Belt” to the west and CTFB to the east [17, 20]. The western flank and part of the axial zone of a few of the anticlines (e.g., Sitakund, Inani, and Dakhin Nhila) are highly deformed and under thrusted by the CCF and, therefore, show deviation from the general fold parameter results (Figure 8(a)). Overall, Hossain et al.’s [44] classification of CTFB into two broad zones seems to be supported by the current study with few exceptions. In the balance geological cross-section, it is also evident that there is a sharp variation in the spacing of anticlines. At the western fold-thrust zone, spacing between anticlines is higher than the eastern highly compressed zone, whereas the curvature of the synclines is higher in the eastern highly compressed zone.

Multicurved geometry of some of the CTFB fold ridges (e.g., Sitapahar, Changotaung; Figure 2) are attributed to the westward migration of fold, multiphase deformation of soft sedimentary rocks, and longitudinal thrust as well as transverse faulting [2, 88]. In the western fold-thrust zone, the curvature of the anticlines is higher than the synclines, whereas in the eastern highly compressed zone, the curvature of the synclines and the anticlines are more or less similar (Figure 8). It indicates that the development of anticlines in the western fold-thrust zone is more fault-controlled, while the anticlines in the eastern highly compressed zone are superposed in nature and result from buckling as well as fault propagation. This is a typical structural style in fold-and-thrust belts [89–92].

At low to very low metamorphic levels, Ramsay [93] classifies the common rock units from the highest (with value 1) to the lowest competence (with value 10). As the exposed Tertiary rock units in the core of the CTFB anticlines are at a very low metamorphic level, therefore, Ramsay’s [93] classification can be applied. The folded rock units in the CTFB area are mainly quartz sandstone (competence value 3), siltstone (competence value 7), and shale (competence value 9). In the CTFB area, the presence of less competent rock units (e.g., shale, mud, and silty shale) in the form of separation surfaces and lateral changes in sedimentary facies can cause the overall changes in fold wavelength, amplitude, and style (Figure 3).

The mechanical properties of rock units involved in folding have profound effect on the fold geometry. If similar sized-layers of incompetent and competent rock units are forced to deform at a given rate, the differential stress will less in the incompetent unit compared to the competent unit. As the rock units in the study area are subjected to layer parallel compressive stress, buckling should be dominant mechanism. In fold-thrust belts, the compressive stress that drives thrusting causes the rock units to buckle and thereby shortening and thickening of the rock units. The buckling can occur either through flexural shear/flow and/or flexural slip depending on the competency of the rock units. According to Donath and Parker [94], the type of mechanism of folding is related to the competence contrast and mean competence of the rock units involved in folding. In the CTFB area, the average competency of the rock units (Table 2) is medium to high, and the competence contrast is also medium to high as compared to Ramsay’s [93] classification. Hence, by considering the Donath and Parker [94] kinematic model and Ramsay’s [93] competence classification, the dominant mechanism of folding in the CTFB area is possibly a flexural shear with some influence of a flexural slip.

In a fold-thrust belt, two major fracture systems are generally developed, which are fold-related and fault-related fractures [55, 56, 95]. Fractures with different orientations can be attributed to changes in the stress field during folding [95]. Fold-related fractures may be axial, cross-axial, and oblique fractures [55, 56]. In the CTFB area, a good relation prevails between fold-related fractures and the fold axes (Figure 6). The S1 fracture set (i.e., axial extensional fracture) has the mean direction of N120°. Fold-related conjugate oblique fractures set S2 and S3 have the trend of ~NNE and ~E, respectively. The bisector azimuth of these two conjugate fracture sets (i.e., S2 and S3) is almost perpendicular to the mean trend of the axial fracture set [95, 96].

The geometric relation between these fracture sets and axial plane suggests that the maximum shortening stress (σ₁) that formed the CTFB anticlines also formed the fractures/lineaments simultaneously. Therefore, it can be argued that the fracture sets S2 and S3 are the members of a conjugate shear system, where the horizontal shortening is perpendicular to the S1 fracture set. If so, then the σ₁ (i.e., maximum principal stress) bisects the acute angle between S2 and S3, while the σ₃ (i.e., minimum principal stress) bisects the obtuse angle between the fracture planes. However, the underlying assumption is that the fractures which formed in homogenous materials under homogenous stress field and in a single deformation event do not significantly perturb the stress field in their neighborhood and their orientation has not rotated significantly since their commencement [55, 57, 58, 74, 97]. In this study, the S2 and S3 conjugate shear fractures in the CTFB area are probably formed simultaneously under the same stress conditions. As these fracture sets were formed as a conjugate pair, the orientation of the bisector angle of the two sets (S2 and S3) of the Sitakund (70°) and Sitapahar (63°) would show the maximum principal compressive stress (σ₁). The mean shortening (σ₁) direction measured from the fracture/lineaments of the two anticlines is 67°, which is in good agreement with the shortening direction (65°) determined from the axial plane orientation.

The directions of maximum anisotropy intensity of the Sitakund and Sitapahar anticlines fracture systems are 82° and 79°, respectively, which are approximately parallel to the mean shortening direction maximum compressive stress (σ₁). This AMOCADO-based results of maximum fractal anisotropy intensity direction (insets of Figures 5(b) and 5(e)) reasonably fit with the regional compressional direction of the CTFB (Figure 10).

The fault-slip data collected from the four different anticlines (i.e., Gobamura, Sitapahar, Bandarban, and Dakhin Nhila) of the fold belt (Figure 2) are used to reveal the paleostress direction which prevailed during deformation. It may be noted that Gobamura is situated in the eastern compressed zone, while Sitapahar, Bandarban, and Dakhin Nhila are from the western fold-thrust zone of the CTFB. The results reported in Section 4.5 reveal an NW-SE to ENE-WSW directed maximum compression, which remains consistent in all the four anticlines mentioned above. It is envisaged that all the thrust faults are related to CTFB deformation on account of NW-SE to ENE-WSW directed maximum compression, which is in good agreement with the paleostress direction obtained from conjugate fracture set as well as fractal anisotropy intensity direction.

Moreover, it is possible to determine the paleostress of a folded area based on the axial plane orientation of the fold [98]. The axial plane orientation of the mapped (Figure 2) and analyzed anticlines (Figure 3) has been used to determine the paleostress of the CTFB. In general, the compressional axis that creates the fold is equal to the maximum principal stress axis (σ₁). This compressive axis is aligned perpendicular to the axial plane of the folds. The average trend of the fold axial plane is ~155° (see Table S2; Figure 10(d)), and therefore, the trend of the maximum principal stress (σ₁) is ~65°, which supports the result obtained through other methods.

The mesoscale analysis of deformation structures performed in this study is mainly based on the field kinematics of the observed thrust faults (Figure 7 and Figure S5 of Supplement 5) and their relation with the map patterns of the anticlines (Figure 2). The presence of approximately east and west verging thrust parallel to subparallel of the NNW-SSE trending fold indicates the west-verging fault-cored anticlines with ENE a subhorizontal shortening. In general, bedding dip distributions, fault kinematics data, and overall map patterns indicate that the subhorizontal shortening axes are approximately perpendicular to the axial trace of the CTFB (Figures 2 and 10). The faults observed locally in both western and eastern zones of the study area (Figure 2) are most likely related to ENE subhorizontal shortening due to westward propagation as the accretionary prism of the Indo-Burmese oblique subduction zone [2, 49]. In general, the deformation pattern shows more complexity in the eastern zone and gradually becomes less intense towards the propagation front of the western IBR, i.e., CTFB, and the observation is consistent with previous studies [17, 20]. The fault-slip analysis also helps analyze the tectonic stress regime and the detailed account of the stress tensor orientation. The tectonic stress regime of the four anticlines has been determined (see Section 4.5), revealing the variation in stress regime in the CTFB from the northern to the southern part. The evaluation of the stress regime (Supplement 4, Table S1) for the respective faults indicates that most faults lie within the pure compressional domain. The obtained stress regime also fits well with the regional tectonics of the CTFB [1, 3, 99].

Three balanced geological cross-sections are drawn along the mean compressive stress direction in the northern, middle, and southern part of the study area (Figure 2), and the estimated shortening percentage of the Tertiary sediments are ~16 km (15%), ~4.2 km (6%) and ~1.5 km (16.7%), respectively. This variation may be due to the fold morphology of the different sedimentary layers. The northern AA$′$ section covers part of the eastern highly compressed zone where the folds are open to gentle and the overall compression rate is high. The middle and the southern part of the study area represent the western fold-thrust zone where the folds are gentle and the percentage of shortening decreases. In the present study, the rate of shortening is not measured because data on thermochronology and geochronology is not available. Maurin and Rangin [18] estimated a crustal shortening of 11 km in the IBR with a 5.5 mm/yr rate along a composite section. Betka et al. [17] reported a westward propagation rate of ≥15 km/Myr. The estimated shortening of the Tertiary sediments in the present study is quite similar to the previous works. The balanced cross-sections of the present study suggest that the CTFB is a thin-skinned fold-thrust belt that reflects an active accretion of deeper Foredeep basin sediments of the Bengal Basin above a ~5–5.5 km deep décollement. We suppose that the décollement, which is at ~7–7.5 km depth in the eastern part of CTFB [42], is laterally continuous and rises to ∼2 km shallower in the western margin of the CTFB. Betka et al. [17] also suggest that the décollement laterally continuous to the west and rises to a shallower level.

Structural data and kinematic analyses of this study suggest that the CTFB was developed due to an approximately east-trending subhorizontal shortening that is normal to the fold axial plane (Figures 2, 3, 9, and 10). Although Maurin and Rangin [18] reported a dextral shear in the western IBR, we found no such field evidence in the IBR outer belt (i.e., CTFB). Our geologic mapping and paleostress analysis (see Supplement 4, Table S2) in the CTFB area revealed no evidence of significant transpression or strike-slip faulting in the study area. The measured ENE shortening orientations based on the fold axial plane, fractures/lineaments (Figure 10), fracture anisotropy (Figure 5), and paleostress analysis (Figure 9) is consistent with the maximum horizontal stress orientations derived from earthquake focal mechanism solutions and with the absolute plate motion direction of the Indian Plate to Eurasian Plate (Figure 10) in the CTFB area. Though only one earthquake focal mechanism solution in the middle of the area shows evidence of dextral strike-slip (see in Figure 10), all other evidences suggest compression rather than transpression is dominant in the study area. The study conducted by Betka et al. [17] in the Tripura Fold belt area and Wang et al. [19] in the Myanmar region also found no evidence of dextral shear in the IBR outer belt. Moreover, the characteristics of synthetic Fault Plane Solution (FPS) from a recent dynamic model involving India-Burma oblique motion are consistent with those derived from our observations [6]. The synthetic source mechanisms from the dynamic model reveal reverse faulting in the CCF and Kaladan faults.

Our observations are also consistent with the shallow seismic events, maximum principal stress orientations (i.e., σ₁) extracted from the World Stress Map database [99], and GPS-derived present-day stress field (Figure 10). The apparent inconsistency between the oblique GPS vector velocity motion of the Indian Plate in the Bengal Basin (Figure 1) and the geologic shortening axes normal to the structural trend of the CTFB (Figure 10) might result from strain partitioning of dominantly pure shear regimes perpendicular to the frontal/outer IBR margin structures. The structural analyses suggest a nearly complete strain partitioning in the frontal part of the IBR (i.e., the CTFB). The component of the plate motion which is normal to the fold belt is absorbed by the CTFB, whereas the component parallel to the fold belt is taken up along ~NS-striking major dextral faults (i.e., Sagaing Fault, Churachandpur-Mao Fault, and Kabaw) in the inner part of the IBR [5, 17, 100]. A similar strain partitioning has also been reported at other oblique subduction zones where the axial traces of accretionary wedge fold belt are usually trench parallel [101, 102]. Therefore, we can rationally conclude that the CTFB absorbs normal component (with respect to the fold belt orientation) of India-Burma oblique vector velocity motion and developed as a nearly fully partitioned accretionary wedge. We assumed that the partitioning between the outer belt of the IBR (i.e., CTFB) and the inner belt occurred along the Kaladan Fault (Figure 1).

Folds, fault-bounded anticlines, thrusts, and factures/lineaments are the major deformation structures of the outer belt (i.e., CTFB) of the IBR. The development of the anticlines in the western fold-thrust zone of the CTFB is more controlled by buckling than the anticlines in the eastern highly compressed fold-thrust zone, which are superposed in nature and developed firstly due to buckling and later through fault propagation mechanism.

The geometric relation between these fracture sets, axial plane, and axial trace suggests that the maximum shortening stress (σ₁) direction measured from the fracture/lineament of the two anticlines is 67°, which is in good agreement with the shortening direction (65°) determined from the orientation of the fold axial plane. The stress tensors deduced from fault-slip analysis of four anticlines show an NW-SE to ENE-WSW directed maximum compression and remain consistent for the analyzed anticlines of the area. It is envisaged that all the thrusts of the CTFB are related to the NW-SE to ENE-WSW directed maximum compression, which is in good agreement with the other analyzed results in this study.

The geologic shortening percent of the Tertiary sediments are ~13.8%, ~8.4%, and ~10.2% for the northern, middle, and southern parts, respectively, of the outer fold-belt component. Kinematics analysis also suggests that the CTFB is a thin-skinned fold-thrust belt that reflects active accretion of the Foredeep basin sediments of the Bengal Basin above a deep regional décollement. Geologic mapping and kinematic analysis of the present study revealed no evidence of significant transpression, which is consistent with previously published studies and, therefore, suggest complete strain partitioning.

The interpreted tectonic stress regime and resulted fold-thrust system of the region will significantly improve the understanding of deformation of any young and active orogenic fold-thrust belt. Finally, it is the first attempt to provide a detailed geological mapping of the CTFB along with a comprehensive analysis of the exposed deformation structures and their stress regime. This study would help to further develop the kinematic and dynamic model of the outer IBR in terms of present-day collision of the eastern and northeastern parts of the Indian Plate.

All analyzed data are accessible at https://osf.io/x6ye9/?view_only=84091640f750442781e3f2ad8716727f.

Key Points. (i) We present a complete structural map of the Chittagong Tripura Fold Belt (CTFB) of the IBR outer belt for the first time. (ii) The maximum compressive stress (σ₁) shows eastward shortening normal to the fold axial plane, and the estimated shortening is ~10.8%. (iii) Due to strain partitioning, the CTFB absorbs the component of plate velocity motion that is normal to mean anticlinal trend.

The authors declare that they have no conflicts of interest.

The study is financially supported by the National Natural Science Foundation of China (41888101 and 41822204), the Science and Technology Major Project of Xinjiang Uygur Autonomous Region, China (2021A03001) and the Jahangirnagar University Research Grants through the Faculty of Mathematical and Physical Sciences. Md. Sakawat Hossain was supported by the CAS PIFI Visiting Scientist Fellowship (2019VCB0011) at the Institute of Geology and Geophysics, Chinese Academy of Sciences. A special thanks to the China-Pakistan Joint Research Center on Earth Sciences for supporting the implementation of this study (131551KYSB20200021).