Modelling of subsurface geological structure related to anomalous magnetotelluric phase data exceeding 90° in Central India

SUMMARY

We note anomalous phase (>90°) in magnetotelluric (MT) measurements at several stations along the Narmada river course normal to the Tan Shear Zone (TSZ) in central India. We made efforts to derive a possible cause of the anomalous phase by comparing the results of numerical modelling based on the tectonics of the TSZ with the measured impedance tensor data. A zone of multiple reactivations of the TSZ leads to the formation of a damage zone and conjugate Riedel shear faults parallel and normal to the TSZ, respectively. The conjugate Riedel shear faults act as a pathway for the Narmada River course. The multiple tectonic activities associated with the TSZ induce surficial heterogeneity that results in distortion in the MT data. The Mohr circle and phase tensor analysis establish the effect of distortion in the data. To understand the cause of distortion as well as anomalous phase in the present data, we carried out synthetic 3-D modelling. We modelled the damage zone and conjugate Riedel shear faults by a cross shape. From the results, we noted that the near-surface 3-D heterogeneity across the TSZ critically distorted the YX component of the MT impedance tensor due to the strong current channelling. When the current flows across the conjugate Riedel shear zone (Narmada river course) charges are accumulated along the boundaries of near-surface heterogeneity. The charges associated with the current channelling brought out the reversion of the electric field direction that is reflected in the form of an anomalous phase in the YX component of the impedance tensor in measured data.

1 INTRODUCTION

An anomalous phase in the magnetotelluric (MT) measurements is generally described as a phase exceeding 90° in the off-diagonal elements of the impedance tensor (Heise & Pous 2003). According to the principle of causality in electromagnetic induction, the MT phase does not exceed 90° in a 1-D or 2-D model. However, the near-surface inhomogeneous conditions induce current channelling that critically distorts the electrical fields, resulting in anomalous phase values, that is >90° (Egbert 1990). Several synthetic and field data studies across the globe reported an anomalous MT phase due to the current channelling, anisotropy or strong resistivity contrast (Pous et al. 2002; Weckmann et al. 2003; Heise & Pous 2003; Pous et al. 2004; Ichihara & Mogi 2009; Pavan Kumar & Manglik 2012; Pina Varas & Dentith 2018). Ichihara & Mogi (2009) proposed an L-shape 3-D model that consists of a regional conductor and attached a local conductor. The local conductor induces a reverse current direction generating an anomalous phase near the local conductor.

MT study (Raju et al. 2022) was conducted across the eastern segment of the central India tectonic zone (CITZ) to understand the tectonics associated with the collision process. Raju et al. (2022) reported anomalous MT phases at several stations, which are located across the major tectonic boundaries such as the Narmada-Son lineament (NSL) central India shear zone (CIS) and Tan shear zone (TSZ) in the CITZ. The anomalous phase data set was not considered for the 2-D modelling, while in the 3-D inversion modelling, the anomalous phase data could not be fitted well with inversion results (Raju et al. 2022). Therefore, in this study, we made an effort to derive the possible cause of the anomalous MT phases across the CIS and TSZ. Following Ichihara & Mogi (2009), and considering tectonic and geological constraints we present a cross-shape 3-D model, and discussed the possible cause of the anomalous phase in central India.

2 GEOLOGY AND TECTONICS OF STUDY AREA

The Grenvillian and Rodian global tectonic events during the Archean and Proterozoic times results in suturing of different protocontinents (such as Bundelkhand Craton, Bastar Craton, Dharwar craton, and Singhbum craton) of India that brought out the formation of the ENE–WSW trending CITZ (Naqvi et al. 1974; Acharyya & Roy 2000; Roy & Prasad 2003 and references therein). The CITZ has differential crustal movement and weakness, which are manifested in the form of several tectonic boundaries such as the NSL, CIS and TSZ, and the eruption of Deccan volcanism (Bhattacharji et al. 1996). The NSL marks the northern boundary of the CITZ; whereas the CIS is considered the southern limit. The TSZ runs parallel to the CIS.

The CIS is a continental collision-suture zone developed under a compressional regime, and the entire crustal column is disturbed due to the collision process (Mall et al. 2008). The occurrence of a wide range of mylonitic structures along the CIS indicates ductile deformation, and its initial brittle deformation is wiped out by plastic deformation (Jain et al. 1991). Another major tectonic boundary, the TSZ is developed under the bulk compressional regime resulting in strong deformation along the TSZ trend (Roy & Prasad 2001). The structural features of mylonitized granites and gneisses indicate the strike-slip nature of shearing emplaced along the TSZ. The shear zone deformation is mainly defined as ductile; however, the brittle nature of reactivation is also noted at some places along the boundary. Pt-M layers with mylonite and ultra mylonite indicate strong internal plastic deformation. The presence of the Pt-C matrix indicates the generation of melt within the TSZ and that describes the multiple reactivations of the TSZ (Chattopadhyaya & Bhattacharjee 2019).

3 ANOMALOUS MT PHASE ACROSS THE CENTRAL INDIA SHEAR ZONES

In the study area, we note an anomalous phase in the YX component of the impedance tensor at 13 MT stations (Fig. 1) along the Narmada river course in the region of CIS and TSZ. The measured axes of MT impedance tensor XY and YX components are situated in N–S and E–W directions, respectively. In the measured data, the anomalous phase in the YX component starts at the period range of 10–100 s on either side of the TSZ as shown in Figs 1 and 2; however, two stations (R18 and R19) near/on the TSZ do not show the phase above 90°.

4 DISTORTION ANALYSIS USING MOHR CIRCLE AND PHASE TENSOR ELLIPSE

Electrical anisotropy analysis (Raju et al. 2022) of present MT data suggests that electrical anisotropy is not the cause of the anomalous phase in the study area. Furthermore, the study indicates that near-surface heterogeneity across the area critically distorts the data that reflect in the form of an anomalous phase. Therefore, we analyse the anomalous phase data set using the Mohr circle and phase tensor to understand the distortion in the data. Mohr circle is used to depict the dimensionality of the subsurface structure, and distortion in the MT data (Lilley 1993a). In this study, we consider the real part of the impedance tensor to plot the Mohr circle and note a consistent pattern with the increasing period for all the stations as shown in Figs 3 and S1. The consistent variation of the Mohr circle with period suggests that the MT data are less contaminated with noise (Raju & Patro 2020). The Mohr circles cut the vertical axis in the longer period range of the most of stations (Figs 3 and S1) which shows the distortion and/or phase 90° in the MT data (Lilley 1993a; Lilley & Weaver 2009). This kind of behaviour is reflected in all the MT stations including R18 and R19 (for example, see R18 station in Fig. 3) which have phases <90°. Therefore, the behaviour of the Mohr circle indicates distortion rather than the signature of phase >90° in the MT data. Furthermore, the regional geological trend is indicated through the direction of the radial arm in the Mohr circle (Lilley 1993b). In this study, the radial arm changes with period indicating a strong influence from a 3-D structure. From the results of the Mohr circle, we suggest that all the MT stations data including those whose phases within the 90° are critically distorted by near-surface heterogeneity. The phase tensor (PT) plot also gives insight into understanding the distortion in the MT data (Caldwell et al. 2004). Phase tensor ellipses with filled values of the maximum principal axis are shown in Fig. 4. In the present PT plot, we observe high principal values in the short period as well as the longer period range at all MT stations. In the short period range (0.001–0.01 s), high values could be caused by near-surface 3-D structure, and in the longer period range (1–1000 s), a regional 3-D structure may cause high values of the maximum principal axis in the PT. We conclude from the Mohr circle and PT analysis that near-surface 3-D conductive structures critically distort the present data set.

5 NEAR SURFACE HETEROGENEITY IN CENTRAL INDIA

Usually, inland areas, the distortion is low in the sedimentary environment whereas, in the crystalline rock areas such as Precambrian regions, the distortion could be extreme in the MT data (Jones 2011). In this study area, the large resistivity contrast between the Precambrian age rocks such as Gneiss overlies by the Deccan volcanic rocks and near-surface unconsolidated sediments along the Narmada river course may distort the regional response of the YX component of the impedance tensor. The distortion appears in the form of an anomalous phase as shown in Fig. 2. Development and multiple reactivations of the strike-slip faults along the TSZ contribute to the formation of a fault damage zone. Furthermore, the strike-slip movement of the TSZ develops Riedel shear and conjugate Riedel shears or antithetic faults (Fig. 5). The conjugate Riedel shear developed simultaneously with Riedel shear on the right angle to the normal movement of fault (Tchalenko 1970).

The volume of a fault zone, minor faults, fractures, cleavage and folds across the fault core are collectively called fault damage zone. The open and filled fractures in the damage zone yield a heterogeneous and anisotropic permeability structure association of fluid flow (Caine & Forster 1999). Fluid distribution within the damage zone generally explains fault zone conductivity as those, which are observed in California (Unsworth et al. 2000; Bedrosian et al. 2002) and Chile (Hoffmann-Rothe et al. 2004), and the source of fluid in these zones are mostly meteoric origin. In this study area, multiple reactivations of the strike-slip nature of the TSZ generate lineaments or fractures (damage zone) across it. The presence of sediments associated with the meteoric origin of fluid in the damage zone cause high conductivity along the TSZ (E–W direction) at a shallow level. Multiple reactivations (pre and post-Deccan trap period) accompanied by crushing of the rocks along the TSZ lead to a decrease in the grain size of the rocks. Therefore, more fluids infiltrate through pore space in the damage zone than adjacent areas, and that gives rise to a high conductivity in shallow levels along the TSZ.

Antithetic faults in the N–S direction are developed during the strike-slip movement of the TSZ (Fig. 5) bringing out a pathway for the Narmada river course. Weathered rocks and sediments associated with fluids (river origin) could be formed as near-surface high conductive structures (heterogeneity) along the river course. To understand the possible cause of the distortion association of the anomalous phase in the present data set, we model the damage zone and antithetic faults of the TSZ.

6 3-D FORWARD MODELLING

3-D forward modelling was performed using the MT3FWD algorithm implemented in the WinGlink software package (Mackie et al. 1994), and modular electromagnetic 3-D inversion code (ModEM; Egbert & Kelbert 2012; Kelbert et al. 2014) based on the non-linear-conjugate-gradient method. The MT3FWD model mesh is extended to 76.6 km (x-axis) × 72.6 km (y-axis) × 101 km (z-axis), and it contains 38 (x-axis) × 54 (y-axis) × 65 (z-axis) layers including 10 air layers in the z-axis. We have considered an initial model with the 1000 Ωm background resistivity and added local topography into the model. The ModEM model mesh comprises 77 km (x-axis) × 73 km (y-axis) × 100 km (z-axis) length in three different axes and the model consists of 122 (x-axis), 62 (y-axis) and 53 (z-axis) layers. The features in 3-D forward models are taken from 3-D inversion results (Raju & Patro 2020) and local geology as well as measured data. From −0.9 to −0.3 km (above mean sea level) depth, the resistivity (4.6–342 Ω.m) values are assigned to both models owing to the measured data (Fig. 6). From shallow level to 100 km (see Fig. 7), the models consist of three different resistivities 700–1000, 100 and 10 000 to 10 292 Ωm from 3-D inversion results (Raju et al. 2022). Based on the characteristics of the TSZ damage zone and its antithetic faults, we assume that the damage zone forms as a regional, and the antithetic fault acts as a local conductor normal to the regional one. In the models, the regional conductor (RC in Fig. 8a) is extended 63 km y-direction with 8 km width (in the x-direction) below the R18 and R19 stations (along the surface expression of the TSZ). We assume that the regional conductor is extended infinity distance on the right-hand side of the y-axis (eastern extension of TSZ); hence, the dimension of the conductor would be quasi-2-D (see Fig. 8a). Two local 3-D conductors (LC1 and LC2 in Fig. 8a) are attached in the middle of the regional conductors in the x-direction parallel to the antithetic faults (along the Narmada river course) of the TSZ. The conductor ‘LC1’ covers 3 km on the y-axis and 9 km on the x-axis and the ‘LC2’ is extended from 3 to 5 km on the y-axis and 15 km on the x-axis. The conductors RC, LC1 and LC2 all together formed a cross shape 3-D model as shown in Fig. 8(a). The response of the cross shape 3-D model is computed with a wide range of resistivity and thickness of the regional and local conductive structure. We note a phase above 90° on the right-hand side of the two local conductors (Fig. 8b) for the resistivity value of 0.2 and 2 Ωm for regional and local conductors, respectively, at the depth range of −0.3 to −0.1 km (above mean sea level).

Near-surface inhomogeneity often comprises significant charges along the boundaries when current flows across. The charges could be a cause of current channelling, deflection of telluric current, and provide the secondary field that is added to the primary field vectorially (Jones 1983). In such conditions, the transverse component (measured across the boundaries) of impedance tensor data, is critically distorted by the galvanic charges that are reflected sometimes in the form of an anomalous phase. In the present MT data, the YX component of the impedance tensor is polarized normally to the Narmada River course. According to distortion theory, surface charges are induced across the near-surface heterogeneity structures along the Narmada River course. The presence of charges across the boundaries tends to current channelling in the Y-component of the electric field that is evident in the form of anomalous phase in the YX component of measured and synthetic MT data set as shown in Figs 2 and 8(b). Since the XY component is parallel to the river direction, there are no charges expected along the boundaries thus no distortion occurred in the XY component (see Fig. 2). Finally, the locations of synthetic phase data responses have well coincided with the measured phase data response as shown in Figs 8(b) and 9.

7 DISTRIBUTION OF ELECTRIC (E) FIELD VECTORS

We demonstrate electric (E) field vector distribution that can explain the cause of the anomalous phase along the Narmada River. The MT3FWD forward modelling returns (E and H) field values corresponding to the period (T). The E-field values are used to compute the distribution of E-field vectors that are discussed as follows. For example, a schematic view of a sinusoidal variation of the Ex and Ey fields at period (T) 1 s is propagating along the x-axis with time (t) as shown in Fig. 10. The time lapse between the Ex or Ey real and Ex or Ey imaginary components is used to derive the phase lag (⁠${\phi }_{\mathrm{x}}or{\phi }_{\mathrm{y}}$⁠) as presented in Fig. 10. In the study, the real part of E-field (Ex' and Ey') values are used to compute the E-field vectors for the period 1000 s at two time (t) steps, that is 0 and 500 s with the help of the following equations:

where K-wave number; T is period of the wave; t is propagation time and φ-phase (radians)

In E-field vector distribution (see Fig. 11), the Y-polarized E-field vectors (align in E–W geographical direction or y-axis in the Cartesian coordinate system) dominate in the regional conductor, and the X- polarized E-field vectors (align in N–S geographical direction or x-axis in the Cartesian coordinate system) prevail in the local conductor. On the left-hand side of the local conductors, the Y-component of the E-field dominates and is parallel to the field, which is in the middle of the regional conductor. Whereas, on the outside of the right edge of the local conductors LC1 and LC2, the Y-polarized E-field is reversed corresponding to the regional conductor (RC). The change in direction of the E-field vectors leads to a change of polarity in impedance tensor that is reflected in the form of anomalous phase in the Y-polarized E-field, that is YX component in real data as well as synthetic studies. There is no telluric current deviation in the regional conductor; hence, we do not observe the anomalous phase over the regional conductor in the measured data as well as synthetic data. However, the presence of near-surface heterogeneity also distorted the R18 and R19 stations' MT data as discussed in Section 4. From the above description, we conclude that when the current flows across the Narmada River course in central India, charges are built up along the boundaries. The current direction is reversed due to the charges that manifested in the form of an anomalous phase in the YX component of the measured data.

8 DISCUSSIONS AND CONCLUSIONS

For the first time, MT studies (Raju et al. 2022) are conducted in the eastern segment of CITZ to understand the tectonics associated with the collision process. The MT profile, which is in the N-S direction, is covered all major tectonic boundaries such as the NSL, TSZ and CIS in the CITZ. The anomalous MT phase is observed at several stations that spatially coincide with the NSL, TSZ and CIS. In this study, numerical modelling is carried out to understand the possible cause of anomalous phase across the CIS and the TSZ with constraints from measured MT data, geological conditions, and 3-D inversion results from Raju et al. (2022). Reactivation of the TSZ at multiple episodes results in a heterogeneous near-surface structure and 3-D nature of the lower crustal level (Raju et al. 2022). The multiple reactivations of the TSZ are evidenced by the damage zone as well as antithetic faults across it. The Narmada River course is completely controlled by the antithetic faults of the TSZ in the study area. Deposition of sediments as well as weathered rocks in the damage zone and antithetic faults added near-surface heterogeneity across the shear zone. Thus, the MT data across the TSZ are critically distorted by the surficial 3-D conductive structures. We demonstrated distortion in the measured data set using the Mohr circle and PT approaches. Further, using the position of the damage zone and antithetic faults we constructed a cross-shape 3-D model. In the cross-shape model, the damage zone is considered a regional conductor, and antithetic faults are attributed to local conductors. The orientation of the observed XY and YX components are parallel and perpendicular to the antithetic fault (local conductor), respectively. The 3-D response is computed at different depth levels and the anomalous phase is noted on the right-hand side of local conductors at the depth level of −0.3 to −0.1 km (above mean sea level), and the response is correlated well with measured data. From the response of the 3-D model, we suggest that the YX component of the impedance tensor in the measured data is critically distorted due to the accumulation of charges across the Narmada River course when the current is flowing across it. The charges associated with current channelling cause changes in the direction of the E-field manifesting as the anomalous phase in the data measured along the river course.

SUPPORTING INFORMATION

Figure S1: Mohr circle for MT stations phase above 90° and phase below 90° (R19). The Mohr circle crosses the vertical axis which may indicate the distortion in the data.

Please note: Oxford University Press is not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the paper.

ACKNOWLEDGEMENTS

We thank Director CSIR-NGRI for the constant encouragement and permission to publish this work. We are also thankful to Chinna Reddy, K. and Ujjal Kumar Borah for participating in MT data acquisition. PKP thanks Gary Egbert for providing the ModEM code and Naser Meqbel for sharing 3-D Grid software. The study was supported by a CSIR 12th 5-yr plan project SHORE (PSC-0205) and MLP-6404–28 (BPK). Khasi Raju further acknowledges the CSIR for its Senior Research Fellowship and the Next Generation EU-sponsored RETURN PNRR project for their support of his research. We thank S. Bulent Tank and the anonymous reviewer and the editor Ute Weckmann for their constructive suggestions, which have improved the manuscript.

DATA AVAILABILITY

Data used for the study are available with CSIR-National Geophysical Research Institute and can be made available under the institutional data policy.

COMPETING INTEREST

The authors declare that they have no competing interests.

References