1 Introduction

Mapping the galleries and checking their ground conditions near the railway tracks are always challenging. It is highly prioritized in the coal mining sector since it is related to the safety and security of the environment and lives. It adds an immense sense of responsibility when these galleries are in the vicinity of the railway track. Railway tracks can be considered as the veins of any state/country for different livelihood transportation. Railway contributes a lot to the development of any country for lives and economy. The present study area is located near railway tracks in Tundu of Jogidih Colliery, Jharia Coalfield, Dhanbad (figure 1). According to an old surface mine plan, some mine galleries are expected at a certain distance away from the railway tracks. These galleries are expected to be ~30 years old and are associated with coal seam-X and seam-XA. These coal seams are expected at a depth of approximately 10–15 m, as indicated in the drilled hole data of the nearby area (figure 2). Subsequently, galleries may have been developed between 10 and 20 m depth through the bord and pillar method. In the bord and pillar method, it is assumed that galleries were excavated for coal production and coal pillars were left for the roof support (Coal Atlas of India 1993). So, we are motivated to perform this study for the safety and security of the railway tracks and nearby areas.

Figure 1
figure 1

(a) Generalized geological map of the Jharia Coalfield (modified after Chandra 1992) and (b) schematic layout plan showing tentative locations of ERT profiles.

Full size image
Figure 2
figure 2

Represents the nearby area bore hole logging data of the Tundu (modified after Das et al. 2017).

Full size image

Therefore, parallel to sub-parallel profile lines of ERT were selected for mapping the possible extension of the cavity/gallery/goaf using Flash RES Universal 61 channel instrument, and data were processed using a 2.5D resistivity inversion program. A magnetic survey was performed using a GSM 19T standard portable proton magnetometer with GPS, and data were analyzed using Oasis Montaj software. Both techniques are unified to understand better the complex nature of cavities/galleries with unknown depth, dimension, extension, and surrounding area conditions.

Magnetic study with other geophysical methods proved very efficient for cavity/gallery/void detection. Mochales et al. (2008) performed a combined study using magnetic, gravity, and ground penetrating radar (GPR) techniques to detect underground cavities in the Zaragoza area of Spain. Near-surface investigation of karstic terrains in Ireland has been achieved successfully using magnetic and resistivity methods by Gibson et al. (2004). An integrated geophysical study comprising magnetic, gravity, electromagnetic and resistivity methods was used for the detection of caverns in sandstone by Chamon and Dobereiner (1988). Geophysics offers non-invasive indirect characterization of complex subsurface formations, urging integrated geophysical investigation for higher accuracy (Horo et al. 2020, 2021, 2023; Kannaujiya et al. 2021).

Different researchers have executed successful studies for detecting cavities/voids using ground-penetrating radar and microgravity (Beres et al. 2001; Leucci and De Giorgi 2010). Scholars also accomplished the detection of cavities in the subsurface using microgravity and gravity techniques (Butler 1984; Bishop et al. 1997). Self-potential has been used well for cavern studies by Lange (1999). Miller and Steeples (1991) used seismic reflection to detect voids in the coal seams. Miller and Steeples (1994) applied seismic surveys for environmental issues. Scholars like Grandjean and Leparoux (2004) proved the efficacy of the seismic method for the detection of cavities and buried materials. Ge et al. (2008) utilized a seam-seismic technique to locate voids in an anthracite mine. Debeglia et al. (2006) applied microgravity and MASW for karst investigation in Orléans, France. The MASW method has been evaluated for the assessment of steeply dipping cavities by Xu and Butt (2006). SRT has been used for karst cavities, as conducted by Sheehan et al. (2005). Integrated geophysical investigation is quite popular for coal mining studies (Wu et al. 2016; Luo et al. 2019). In Raniganj and Jharia Coalfields, different geophysical studies have been carried out for sustainable coal mining and mine hazard analysis, mostly using resistivity (Verma and Bhuin 1979; Verma et al. 1982; Singh et al. 2004; Bharati et al. 2015, 2016b, 2019, 2021; Das et al. 2017; Srivastava et al. 2020; Kumar et al. 2021) and magnetic (Vaish and Pal 2015a, b; Pal et al. 2016) methods, separately. The present study mainly focuses on mapping the old mine galleries and checking the stable ground formation considering their extension near railway tracks using combined resistivity and magnetic study.

2 Geological setup of the study area

Our study area is a part of Tundu, Jogidih Colliery, Jharia Coalfield, which is west of Dhanbad (figure 1). This study is related to the gallery associated with coal seam-X and seam-XA, which belong to the Barakar Formation. The most prominent coal seams are from the Barakar Formation, which consists of nearly 18 major coal seams. Barakar Formation mainly comprises sandstone of variable grain size, argillaceous sandstone, intercalation of sandstone and shale, carbonaceous shale, jhama, mica-peridotite, and coal seam (Chandra and Chakraborty 1989; Vaish and Pal 2015a, b). Lower Gondwana series includes Talchir, followed by Barakar (lower coal series), then barren measures and uppermost is the Raniganj series (upper coal series), all formations lying above the Archean basement (Fox 1930; Chandra 1992). Jharia Coalfield covers an area of around 450 km2, a sickle-shaped area with an extension of approximately 19 km from north to south and 38 km from east to west.

3 Methodology

A resistivity survey was performed in the vicinity of the railway tracks to check the stable homogeneous formation and extension of the gallery zone near the railway track. Resistivity data was acquired through 61-channel (64 electrodes) Universal Flash RES instrument with Wenner, Schlumberger, Gradient and Dipole–Dipole arrays. A total of seven profiles were selected for ERT data acquisition (figure 1). The electrode spacing for selected profiles AA′ and FF′ was 2.5 m, covering a length of 158 m. ERT data were collected with 2 m electrode spacing in profiles BB′, CC′, DD′, EE′, and GG′, covering a total distance of 126 m (figure 1b). All the profiles were initiated from the culvert side and terminated at the Chandrapura station side, as indicated in figure 1(b). Wenner array is sensitive to vertical variations and is preferred for finding the horizontal structure (Loke 1999). Schlumberger array is discreetly sensitive to both horizontal and vertical structures (Loke 1999). Dipole–Dipole array is sensitive to horizontal variations, so it is preferred to find a cavity or dyke like structure (Loke 1999). Gradient array with multiple electrode combinations best suits the subsurface structure resolution (Dahlin and Zhou 2004, 2006; Bharti et al. 2016b). Resistivity data were filtered with the Flash RES Universal data checking program and processed through a 2.5D resistivity inversion program. After processing all the arrays, the data were combined for a joint inversion to achieve more effective results. It has been established that the inversion of collective datasets acquired by different arrays in the same profile delivers relative benefits for all arrays and yields better results than the individual (De la Vega et al. 2003; Stummer et al. 2004; Athanasiou et al. 2007; Bharti et al. 2016a; Das et al. 2017). Signal-to-noise ratio and resolution significantly enhanced in the inversion of combined data (Zhou and Greenhalgh 2000; Dahlin and Zhou 2006).

Magnetic data was acquired using a GSM 19T standard portable proton magnetometer with GPS. The sensitivity of the instrument is 0.15 nT/Hz with a resolution of 0.01 and 0.2 nT absolute accuracy. Dense magnetic measurements were performed on and around the expected area for mapping the underground mine gallery. Magnetic data were obtained using a proton precession magnetometer at 2 m data spacing for five profiles spaced 10 m apart, generating high-resolution data. Repeated base readings were taken every two hours, and the magnetic data were corrected for the diurnal variation. Magnetic data utilized in this study were of 99 signal strength.

Further, magnetic data were gridded and processed in the Oasis Montaj software. A total magnetic anomaly (TMA) map was developed by subtracting the IGRF value from the total magnetic intensity (TMI) map. The first-order vertical derivative of TMA data was computed to highlight shallow structures, reducing anomaly complexity and allowing precise imaging of the causative structures (Rao et al. 1981; Pal and Majumdar 2015; Vaish and Pal 2015b; Pal et al. 2016). The TMA was corrected using the reduced-to-pole (RTP) technique, ensuring the anomaly center on top of its causative sources (Roy and Aina 1986; Ganguli et al. 2021a, b, 2022).

4 Results

4.1 Resistivity study

In figure 3, 2D ERT cross sections of (a) Wenner, (b) Schlumberger, (c) gradient, (d) dipole–dipole, and (e) joint inversion of all arrays for Profile AA′ are plotted. However, for interpretation purposes, the joint inversion section has been selected, as indicated in figure 10(a). The 2D ERT section shows subsurface resistivity anomaly distributions in Profile AA′ (figure 10a). AH1 indicates a relatively high resistivity (~300–800 Ωm) feature with a horizontal extension of ~0–60 m and depth of ~13–32 m, which may be inferred as high resistive compact ground. AL1 indicates a relatively low resistivity (~150–200 Ωm) feature with a horizontal extension of ~60–76 m and depth of ~15–27 m, which may be inferred as goaf/voids/cavities filled with loose material of relatively higher moisture content. AH2 represents a relatively high resistivity (~250–350 Ωm) feature with a horizontal extension of ~70–100 m and depth of ~8–22 m, which may be inferred as relatively high resistive compact ground. AL2 represents a relatively low resistivity (~100–150 Ωm) feature with a horizontal extension of ~128–160 m and depth of ~10–25 m, which may be inferred as goaf/voids/cavities filled with loose material of relatively higher moisture content. Details of resistivity features are provided in table 1.

Figure 3
figure 3

Represents the 2D ERT cross sections of (a) Wenner, (b) Schlumberger, (c) gradient, (d) dipole−dipole, and (e) joint inversion of all arrays for Profile AA′.

Full size image
Table 1 Describes the approximated resistivity value, extension and depth of the anomalous features of the joint inversion cross-section for different profiles in Part A.
Full size table

In figure 4, 2D ERT cross sections of (a) Wenner, (b) Schlumberger, (c) gradient, and (d) joint inversion of all arrays for Profile BB′ are plotted. However, the joint inversion section has been selected for interpretation purposes, as indicated in figure 10(b). 2D ERT section shows subsurface resistivity anomaly distributions in Profile BB′ (figure 10b). BH1, a relatively high resistivity feature of ~280–460 Ωm, has been identified with a horizontal extension of ~0–33 m and a depth of ~6–27 m, which may be inferred as high resistive compact ground. BL1, a relatively low resistivity feature of ~120–160 Ωm, has been observed with a horizontal extension of ~34–42 m and a depth of ~8–14 m, which may be inferred as goaf/voids/cavities filled with loose material of relatively higher moisture content. BH2, a relatively high resistivity feature of ~200–260 Ωm, has been delineated with a horizontal extension of ~44–108 m and a depth of ~6–27 m, which may be inferred as relatively high resistive compact ground. BL2, a relatively low resistivity feature of ~140–160 Ωm, has been delineated with a horizontal extension of ~80–84 m and a depth of ~10–12 m, which may be inferred as goaf/voids/cavities filled with loose material of relatively higher moisture content. Details of resistivity features are provided in table 1.

Figure 4
figure 4

Represents the 2D ERT cross sections of (a) Wenner, (b) Schlumberger, (c) gradient, and (d) joint inversion of all arrays for Profile BB′.

Full size image

In figure 5, 2D ERT cross sections of (a) Wenner, (b) Schlumberger, (c) gradient, (d) dipole–dipole, and (e) joint inversion of all arrays for Profile CC′ are plotted. However, for interpretation purposes, the joint inversion section has been selected, as indicated in figure 10(c). 2D ERT section indicates subsurface resistivity anomaly distributions in Profile CC′ (figure 10c, table 1). CH1 represents a relatively high resistivity (~200–460 Ωm) feature with a horizontal extension of ~0–46 m and a depth of ~7–27 m, which may be inferred as high resistive compact ground. CL1 represents a relatively low resistivity feature of ~160–180 Ωm with a horizontal extension of ~47–58 m and a depth of ~16–25 m, which may be inferred as goaf/voids/cavities filled with loose material of relatively higher moisture content. CH2 represents a relatively high resistivity feature of ~180–320 Ωm with a horizontal extension of ~59–130 m and a depth of ~5–27 m, which may be inferred as relatively high resistive compact ground. CL2 indicates a relatively low resistivity feature of ~40–80 Ωm with a horizontal extension of ~82–96 m and a depth of ~9–14 m. CL3 shows a relatively low resistivity feature of ~40–80 Ωm with a horizontal extension of ~105–117 m, and a depth of ~9–16 m is observed; both features may be inferred as possible topsoil covered with loose material of relatively higher moisture content.

Figure 5
figure 5

Represents the 2D ERT cross sections of (a) Wenner, (b) Schlumberger, (c) gradient, (d) dipole−dipole, and (e) joint inversion of all arrays for Profile CC′.

Full size image

In figure 6, 2D ERT cross sections of (a) Wenner, (b) Schlumberger, (c) gradient, (d) dipole–dipole, and (e) joint inversion of all arrays for Profile DD′ are plotted. However, for interpretation purposes, the joint inversion section has been selected, as indicated in figure 10(d). The 2D ERT section represents subsurface resistivity anomaly distributions in Profile DD′ (figure 10d, table 1). DH1, a relatively high resistivity (~200–520 Ωm) feature, has been delineated with a horizontal extension of ~0–56 m and a depth of ~8–27 m, which may be inferred as high resistive compact ground. DL1, a relatively low resistivity (~100–180 Ωm) feature, has been identified with a horizontal extension of ~61–70 m and a depth of ~7–20 m, which may be inferred as goaf/voids/cavities filled with loose material of relatively higher moisture content. DH2, a relatively high resistivity (~220–300 Ωm) feature, has been delineated with a horizontal extension of ~73–130 m and a depth of ~16–27 m, which may be inferred as relatively high resistive compact ground. DL2, a relatively low resistivity (~100–140 Ωm) feature, has been delineated with a horizontal extension of ~0–130 m and a depth of ~0–10 m, which may be inferred as possible loose material of relatively higher moisture content.

Figure 6
figure 6

Represents the 2D ERT cross sections of (a) Wenner, (b) Schlumberger, (c) gradient, (d) dipole−dipole, and (e) joint inversion of all arrays for Profile DD′.

Full size image

In figure 7, 2D ERT cross sections of (a) Wenner, (b) Schlumberger, (c) gradient, (d) dipole–dipole, and (e) joint inversion of all arrays for Profile EE′ are plotted. However, for interpretation purposes, the joint inversion section has been selected, as indicated in figure 10(e). 2D ERT section shows subsurface resistivity anomaly distributions in Profile EE′ (figure 10e, table 1). EH1, a relatively high resistivity feature of ~150–180 Ωm, has been observed with a horizontal extension of ~25–130 m and a depth of ~4–10 m, which may be inferred as a possible homogeneous compact formation (first layer). EH2 represents a relatively higher resistivity feature of ~190 Ωm with a horizontal extension of ~0–130 m and a depth of ~10–27 m, which may be inferred as a possible homogeneous compact formation (second layer).

Figure 7
figure 7

Represents the 2D ERT cross sections of (a) Wenner, (b) Schlumberger, (c) gradient, (d) dipole−dipole, and (e) joint inversion of all arrays for Profile EE′.

Full size image

In figure 8, 2D ERT cross sections of (a) gradient, (b) dipole–dipole, and (c) joint inversion of both arrays for Profile FF′ are plotted. However, for interpretation purposes, the joint inversion section has been selected, as indicated in figure 10(f). 2D ERT section indicates subsurface resistivity anomaly distributions in Profile FF′ (figure 10f, table 2). FH1, a relatively high resistivity (~100–120 Ωm) feature, has been delineated with a horizontal extension of ~0–160 m and a depth of ~0–13 m, which may be inferred as a possible homogeneous compact formation (first layer). FL1, a relatively low resistivity (~70–85 Ωm) feature, has been identified with a horizontal extension of ~0–160 m and a depth of ~15–32 m, which may be inferred as possible goaf/voids/cavities filled with loose material of relatively higher moisture content (second layer).

Figure 8
figure 8

Represents the 2D ERT cross sections of (a) gradient, (b) dipole−dipole, and (c) joint inversion of all arrays for Profile FF′.

Full size image
Table 2 Describes the approximated resistivity value, extension, and depth of the anomalous features of the joint inversion cross-section for different profiles in Part B.
Full size table

In figure 9, 2D ERT cross sections of (a) Wenner, (b) Schlumberger, (c) gradient, (d) dipole–dipole, and (e) joint inversion of all arrays for Profile GG′ are plotted. However, for interpretation purposes, the joint inversion section has been selected, as indicated in figure 10(g). The 2D ERT section represents subsurface resistivity anomaly distributions in Profile GG′ (figure 10g, table 2). GL1 indicates a relatively low resistivity feature of ~40–120 Ωm with a horizontal extension of ~0–160 m and a depth of ~0–6 m, which may be inferred as topsoil covered with loose material of relatively higher moisture content (first layer). GH1 indicates a relatively high resistivity feature of ~160–320 Ωm with a horizontal extension of ~0–160 m and a depth of ~6–27 m, which may be inferred as a possible homogeneous compact formation (second layer).

Figure 9
figure 9

Represents the 2D ERT cross sections of (a) Wenner, (b) Schlumberger, (c) gradient, (d) dipole−dipole, and (e) joint inversion of all arrays for Profile GG′.

Full size image
Figure 10
figure 10

Represents joint inversion of profiles AA′, BB′, CC′, DD′, EE′, FF′, and GG′ for figures (a), (b), (c), (d), (e), (f), and (g), respectively.

Full size image

4.2 Magnetic study

The TMA map of the present study area is shown in figure 11. A low magnetic anomaly over the expected area has been observed, possibly indicating gallery locations marked by the black dashed line. Magnetic anomaly varies from –350.8 to –176.1 nΤ near the possible gallery/void.

Figure 11
figure 11

Total magnetic anomaly map of the study area. A black dashed polygon shows a possible broad zone of underground galleries.

Full size image

The RTP of the TMA map is shown in figure 12. A broad low magnetic anomaly zone (–517.6 to –200 nΤ) has been identified over the expected area of the possible old gallery, marked by a black dashed polygon.

Figure 12
figure 12

Reduced to pole (RTP) of total magnetic anomaly (TMA) map. A black dashed polygon shows a possible broad zone of underground galleries.

Full size image

The first vertical derivative (FVD) of the TMA map is shown in figure 13, which clearly delineates 11 low magnetic anomalies (–24.5 to –10.9 nΤ/m) over the expected area of possible gallery/void shown by the dashed circle with 1–11 numbering for each possible gallery. The low anomaly may be due to natural wear and tear in the galleried zone over time. This phenomenon of low anomaly may be due to the loose material or absence of material.

Figure 13
figure 13

First vertical derivative of total magnetic anomaly map.

Full size image

5 Discussions

A comprehensive analysis has been carried out using ERT and magnetic methods. For simplification, the entire area has been studied using ERT considering two parts, Part A and Part B. 2D ERT sections were generated along parallel and sub-parallel profile lines on Part A and Part B to detect cavity/gallery/goaf with their extension. Our concern for the study is also to check the stable subsurface formation of the area within ~25 m from the railway tracks, considering homogeneous underneath formation. The details of distinct resistive anomalous features delineated in Part A and Part B have been summarized in tables 1 and 2, respectively. Based on the variation of relative resistivity anomaly, different zones have been identified, including (i) a possible solid high-resistive compact ground, (ii) a possible cavity/gallery/goaf filled with water or loose material with relatively higher moisture content, and (iii) a possible topsoil covered with higher moisture content or saturated by water.

A schematic geoelectrical model of Part A generated using 2D ERT sections of subsurface resistivity (Ωm) variation for profiles AA′, BB′, CC′, DD′, and EE′ indicating the tentative extension of subsurface cavity/gallery/goaf filled with water or loose material of relatively higher moisture content, is presented in figure 14. A schematic geoelectrical model covering parts of Part A and Part B is shown in figure 15. 2D ERT sections of subsurface resistivity (Ωm) variation delineates the tentative location of solid homogeneous formation. Imprints of the possible galleries have been identified at a certain distance away from the railway tracks using magnetic, which corresponds well with the schematic old mine plan of BCCL, as shown in figure 16. A relatively high magnetic anomaly zone was observed beside railway tracks, which may indicate the absence of galleries/voids/goaves, and the presence of homogeneous compact ground. No recent drilling was performed for the report verifications; however, the approximate depth range identified in the present study corroborates well with the old drilled hole data available around the study area, indicating coal seam-X and seam-XA with depth ranging from about 10 to 15 m.

Figure 14
figure 14

An integrated model of 2D ERT sections of subsurface resistivity (Ωm) variation for joint inversion profiles AA′, BB′, CC′, DD′, and EE′ (Part A) showing the tentative extension of subsurface cavity/gallery/goaf filled with loose material of relatively higher moisture content.

Full size image
Figure 15
figure 15

An inferred model of 2D ERT sections of subsurface resistivity (Ωm) variation showing the tentative location of solid homogeneous materials.

Full size image
Figure 16
figure 16

Correlation of the galleries from (a) schematic tentative galleries surfaces mine plan of the site in Tundu Khaas, Jogidih Colliery, source (BCCL) to (b) first vertical derivative map.

Full size image

Using the ERT study, it has been inferred that ground within 20–25 m from the railway line in Part A is a possible homogeneous compact formation without any cavity/gallery/goaf. Meanwhile, the ground within 25–30 m from the railway line in Part B has also been inferred to be a possible homogeneous compact formation without any cavity/gallery/goaf. So, unifying ERT and magnetic studies, it is inferred that ground within the area of ~20 m from the railway tracks is stable due to homogeneous compact ground as there is no gallery.

6 Conclusions

  • The comprehensive ERT study suggests different subsurface zones, such as (i) relatively high resistive feature possibly indicates homogeneous/solid compact ground, (ii) relatively low resistive feature possibly indicates cavity/gallery/goaf filled with water-saturated formation/loose formation of relatively higher moisture content, and (iii) moderately low resistive feature indicates possible topsoil covered with higher moisture content/saturated by water; considering relative resistivity distribution.

  • A broad low magnetic anomaly zone has been identified that corresponds well with the broad low resistivity anomaly pattern associated with possible galleries of the study area. The first vertical derivative of the TMA map delineates 11 low magnetic anomalies, indicating possible location and extension of galleries/voids, which corresponds well with the schematic old mine plan of BCCL.

  • ERT sections failed to delineate prominent imprints of the 11 galleries separately as identified in the magnetic field, as the ERT lines may not align along the suspected area observed in the first vertical derivative of the TMA. Moreover, the imprints of the galleries may be partially averaged out/smoothened, leading to an extended zone of low resistivity in the ERT section AA′ and FF′.

  • The ground within 20–25 m from the railway line in Part A and the ground within 25–30 m from the railway line in Part B are characterized by relatively higher and homogeneous resistivity with higher magnetic anomaly distribution, possibly indicating homogeneous compact formation without any cavity/gallery/goaf.

  • So, unifying ERT and magnetic studies, it is inferred that ground within the area of ~20 m from the railway tracks is stable due to homogeneous compact ground as there is no gallery.