0016-7622/2017-90-2-000/$ 1.00 © GEOL. SOC. INDIA JOURNAL GEOLOGICAL SOCIETY OF INDIA Vol.90, August 2017, pp. Abandoned mine galleries detection using Electrical resistivity tomography method over Jharia coal field, India Prasenjit Das, S. K. Pal * , P.R. Mohanty, Piyush Priyam, Abhay Kumar Bharti and Rajwardhan Kumar Department of Applied Geophysics, Indian Institute of Technology (ISM), Dhanbad– 826004, India *E-mail: sanjitism@gmail.com ABSTRACT Land subsidence is a serious problem in Indian coalfields due to old underground mine workings. Unfortunately, most of these are uncharted as no mine plans are available. The hidden galleries, goafs, shafts etc. may pose great threat for future mine development as well as to the local environment. The mine workings should be charted to undertake an effective preventive action. In the present study, 2D electrical resistivity tomography (ERT) technique has been used to detect underground mine workings, mainly air or water filled galleries. Initially, the whole exercise has been executed through a synthetic model study. Gaussian random noise of 5mV/A has been added with synthetic data to demonstrate field condition which provides realistic results. ERT survey was conduc- ted over a part of Jogidih coal mine of Jharia coal field in India for a first time. Four electrode configurations, Wenner, Schlumberger, dipole-dipole and gradient were considered for this study. The results indicate the presence of sub-surface water and air filled cavity due to high resistivity contrast with surroundings. INTRODUCTION Electrical resistivity tomography (ERT)/electrical resistivity imaging (ERI) method has wide application in the field of near surface geophysics. This imaging method is capable of detecting sub-surface cavity and geological features (such as faults, formation boundaries and depth of bed rock) based on the sub-surface resistivity distribution. Detection of sub-surface cavities of old coal workings is very important for coal mining from deeper seams as well as for mitigating probable regional environmental hazards. Sub-surface cavities may be air filled or water filled that could be easily detected due to high resistivity contrast with surroundings (Zhou et al., 2002; van Schoor, 2002). This technique is most popular due to its quick cost effective result (Griffiths and Barker, 1993; Zhou et al., 2000). ERT has already been successfully used by various researchers to detect sub-surface sinkhole and cavities (Kruse et al., 2006; Leucci et al., 2004; Dobecki and Upchurch, 2006; Martinez et al., 2009; Bharti et al., 2015, 2016). A few studies on identification and delineation of mine workings in Raniganj and Jharia coalfield were carried out in recent past by different geophysical methods (Krishnamurthy et al., 2009; Singh, 2003, 2013a; Bharti et al., 2015, 2016). Maillol et al. (1999) used ERT for detection of uncharted mine galleries in West Bengal, India. Singh (2013b) used 2D resistivity imaging for delineation of water logged area in inaccessible underground workings at Hingir Rampur colliery. Most of these studies were conducted with a single array for detection of either one or two mine workings. The present study deals with two fold objective, viz., (i) synthetic model study to characterize a numbers of water filed cavity (WFC) and air filled cavity (AFC) mine workings using different arrays and (ii) mapping of the exact ground locations of old mine galleries/goafs and coal pillars over a part of Jogidih coal mine for ground stability treatment using sand stowing. In general, the Wenner array is good in resolving vertical changes i.e., horizontal structures, but relatively poor in detecting horizontal changes i.e. narrow vertical structures (Loke, 1999). The Schlumberger array is moderately sensitive to both horizontal and vertical structures. The dipole-dipole array is very sensitive to horizontal changes in resistivity, but relatively insensitive to vertical changes in the resistivity. Thus, it is good in mapping vertical structures, such as dykes and cavities. This array has better horizontal data coverage than the Wenner (Loke, 1999). Dahlin and Zhou (2004) have shown that the gradient array with multiple current-electrode combinations is best among the electrode arrays in terms of resolution of sub-surface structures and it is clearly superior to the commonly used Wenner, Schlumberger, dipole-dipole, and pole-dipole and pole-pole arrays in most of the modeled cases. So we have chosen four electrode arrays as gradient, Wenner, Schlumberger and dipole-dipole for acquisition of field data. STUDY AREA The present study has been carried out over a part of Jogidih coal mine, Jharia coalfield, India. This coal mine was closed 25-30 years ago. This area is located at about 24km west of Dhanbad town. The rock formations of Jharia coalfield unconformably overlying the Archean basement, mainly belong to the lower Gondwana group rocks of Permian age comprising Talchir, Barakar, Barren measures and Raniganj formations, from bottom to top (Chandra, 1992; Pal et al., 2016; Srivardhan et al., 2016). The Barakar Formation is main coal bearing formation. The rocks of Barakar Formation are mainly sandstone of variable grain size, argillaceous sandstone, intercalation of sandstone and shale, carbonaceous shales, jhama, mica-peridotite and coal seams (Chandra and Chakraborty, 1989; Vaish and Pal, 2015, 2016). The location map of study area with geological map of Jharia coal field (after GSI, 1964) and surface mine plan of a part of Jogidhi coal mine is shown in Fig.1. Verma and Bhuin (1979) observed in their electrical resistivity study over parts of Jharia coalfields, India that the coal bearing strata show relatively high resistivities in the range of about 700 Ωm to 1140 Ωm whereas, Barakar Formation comprising sandstone, sandy shale and shale show relatively low resistivities in the range of about 300 Ωm to 400 Ωm. SYNTHETIC MODEL Based on the possible geological condition of study area (Fig.2) a numerical model has been designed for better interpretation of field data. Initially, forward model has been computed using RES2DMOD software package (Loke and Barker, 1996). Finite difference algorithm has been used for forward modelling (Loke et al., 2003).This algorithm divides the model sub-surface into a number of rectangular blocks. The model (Fig. 3a) consists of 50 equally spaced electrodes with 2.5m separation. It contains four sub-surface layers (Table 1) viz., (i) alluvium/weathered layer, (ii) sandstone formation, (iii) coal layer with alternative seven coal pillar and seven gallery/goaf and (iv) shale.The