Identification of shale gas existence based on 2-dimensional magnetotelluric data in the Kutai basin, East Kalimantan, Indonesia

Research on unconventional methods in the shale gas exploration process has begun to be developed to meet the demand for oil and natural gas sustainability. The research area is located in the Kutai Basin, East Kalimantan, Indonesia. The current study from the Geological Survey Center shows that the Kutai Basin has 46.79 TCF of shale gas potential. The magnetotelluric (MT) method was applied in this study to describe the distribution of subsurface resistivity values. The inversion schemes performed in the MT method are 1-dimensional and 2-dimensional inversions. In this study, there are 8 measurement stations in one line. The presence of low resistivity values or conductive zones identifies the potential for black shale layers. The potential for shale gas is suspected to be detected at station KT29, which is included in the Pamaluan Formation, and at stations KT12, KT31, KT13, and KT15, which are included in the Pulaubalang Formation, because it is associated with source rock from petroleum in the Kutai Basin system, as supported by previous research data.


Introduction
To supply Indonesia's growing demand for oil and natural gas, production of both has been increasing [1], [2].The development of oil and gas exploration must continue to be carried out to process new energy sources that can be developed from now to the future in order to create sustainable national development.Shale gas is unconventional method that is beginning to be developed in the exploration process that can be used to increase oil and gas production [3].Shale gas is a gaseous hydrocarbon that forms and becomes trapped in layered rock formations that are highly impermeable and non-porous [4].
Exploration and production in several countries in Southeast Asia, such as Indonesia, Malaysia, and Thailand, has substantial quantities of shale gas and strong possibilities for it [5].One of the prospective hydrocarbon producers in Indonesia is located in the Kutai Basin, East Kalimantan [6], [7].The stratigraphy of the Kutai Basin consists of four main formations, namely the Pamaluan, Pulaubalang, Balikpapan, and Kampung Baru formations, with the Pamaluan Formation showing maturity achievements with mature to over-mature categories based on maturity analysis through rock-eval and pyrolysis [8], [9].
In oil and gas exploration, many kinds of methods are applied, one of which is the magnetotelluric (MT) approach.MT is a method that detects subsurface conditions from a depth of a few meters to tens of kilometers using natural electromagnetic (EM) waves [10].Because of its very deep subsurface penetration and adequate vertical resolution for describing the distribution of subsurface  [11].This paper aims to gather subsurface information on the shale gas potential of the surrounding area by conducting a 2-dimensional inversion of MT data.

Methods
2.1.Fundamental theory 2.1.1.Kutai basin, East Kalimantan, Indonesia The Kutai Basin is located in East Kalimantan, which is thought to have been formed by the rifting process in the Middle Eocene due to the expansion of the Sulawesi Sea and the North Makasar Strait [12].Sedimentation in the Kutai Basin is divided into two phases: the Paleogene sedimentation phase, which is transgressive, and the Neogene sedimentation phase, which is regressive [13].The Kutai Basin has a dominant geological structure of folds and faults that mostly runs northwest (NE) to southwest (SW) [14].The geologic map of the SN Traverse of Kutai Basin, East Kalimantan, is shown in Figure 1.
The stratigraphy of the Kutai Basin consists of the Pamaluan Formation with quartz sandstone interbedded with limestone, claystone, siltstone, and black shale.In addition, there is the Bebuluh Formation, with reef limestone interbedded with passive limestone and black shale.There is also the Pulaubalang Formation with greywacke sandstone and quartz sandstone interbedded with limestone, claystone, coal, and dacite tuff.Next, there is the Balikpapan Formation, with sandstone and claystone interbedded with siltstone, shale, limestone, and coal.Then there is the Kampungbaru Formation, with quartz sandstone interbedded with claystone, shale, siltstone, and lignite.And there are alluvium deposits in the form of gravel, sand, silt, sandstone, and mud [15].From the stratigraphy of the Kutai Basin, the Pamaluan Formation as the source rock and shale gas carrier is in the mature to over-mature category based on maturity analysis through rock-eval and pyrolysis [8], [9].Apart from that, almost all routes in the Kutai Basin have oil and gas fields because of the anticlines in the Samarinda anticlinorium route, whether faulted or not, starting from land to offshore [14].Based on the results of research conducted in 2017 by the Central of Geological Agency, 46.79 TCF of shale gas potential is in the Kutai Basin.
2.1.2.Unconventional oil and gas Oil and gas are extracted directly from source rock formations, which are fine-grained rock layers with very low hydraulic conductivity, in unconventional oil and gas exploitation [16].Unconventional oil and gas reservoirs are formed in low-energy, oxygen-anaerobic environments rich in organic matter and clay minerals [17].Shale oil and gas, tight oil and gas, coal bed methane, and gas hydrates are included in unconventional hydrocarbon resources [18].
2.1.3.Shale gas Shale gas is a type of unconventional oil and gas obtained from source rock or source rock in the form of shale trapped in the source rock [3], [2].Therefore, in order to produce shale gas that has been trapped in the source rock, artificial fracturing is required [8], [19].Shale is a type of rock in the form of clay particles compacted in mudstone that has the ability to store gas in the pore volume of the matrix and on the surface area of the pores [9], [20], [21].In order to evaluate the potential of shale gas, there are contributing aspects, namely the depositional environment, depth, total organic carbon (TOC), maturity, and geographic location [2].

Magnetotelluric (MT) method
The MT method estimates the resistivity distribution of rocks in the subsurface using natural EM fields [10], [22].The MT method is widely applied in various geophysical explorations, such as geothermal exploration studies [23], volcanic systems [24], and fault structure investigations [25].The various resistivity values in depth are produced through MT measurements.In EM phenomena, the equations of the MT method are arranged in Maxwell's equations.In this study, the Maxwell equations used are Faraday's law and Ampere-Maxwell's law. ∇ where B = magnetic induction, J = current density, and D = electric displacement.The bold notation represents vectors in three-dimensional space E = ̅ = (Ex, Ey, Ez), which means that the components of E, H, B, J, and D have components , , dan .Equations ( 1) and ( 2) describe that changes in the magnetic field cause an electric field and vice versa.
The relationship between the electric and magnetic fields and the material's response to the EM field is described by the constitutive equation.: where  = conductivity,  = 1/ = resistivity,  = dielectric permittivity, dan 0 = magnetic permeability.
Because the MT method is in the time domain, it needs to be converted into the frequency domain.The H and E field equations then have the same form.: 2) In the MT method, there is a range of rock conductivities that use a quasi-static approach.In this approach, the conductivity term () has a greater effect than the displacement current term () which is used to inject the current and can be ignored [26].Then Maxwell's equations ( 1) and ( 2) change to: (5.2) From the equations (5.1) and (5.2), the curl operation is carried out.The electric field E changes to: For the magnetic field H it changes to: From equations (6.1) and (6.2), wave equation is obtained: where  2 = iωµ 0 σ + ω 2 µ 0 ε 0 (complex number).In the MT method, the frequency used is ~10 4 Hz, then σ ≫ ε = iωµ 0 σ ≫ + ω  µ 0 ε 0 and then used in determining skin depth.The skin depth phenomenon, which is the depth in a homogeneous medium where the EM wave amplitude attenuates, will occur for electromagnetic waves that have been reduced to 1/e of their amplitude at the earth's surface (ln e = 1 or e = 2,718 …).To determine the depth of penetration or the depth of an EM wave inquiry, the quantity of skin depth is used, which is stated as with skin depth in meters.Mostly, the estimated depth of investigation from skin depth is too large, so a more accurate and effective depth is used: (10) The comparison of the electric field E with the magnetic field H in the MT method is called impedance Z.In a homogeneous medium, the amplitude of the EM field at the earth's surface is still present in both fields, so it must be removed to become with k = ωµ 0  .This equation has a relationship between impedance and resistivity: where Z is a complex number.From equation (13), we also get the phase quantity, which is derived from the comparison of the imaginary component Z with the real component Z is: () ]= 45°, (for a homogeneous medium) (14a) () ] ≠ 45°, (for a non-homogeneous medium) (14b) because resistivity is not constant and varies in depth, the apparent resistivity symbol ρa is represented as a function of depth ().

One-dimensional and two-dimensional inversion
One-dimensional inversion is conducted using the Bostick and Occam algorithms in the WinGLink application.The Bostick curve is in the form of apparent resistivity, while the Occam curve is in the form of real resistivity [27].Occam's algorithm is the main emphasis of this model since it is more accurate and has a smaller error value than Bostick's algorithm [28].
Two-dimensional inversion was carried out using the WinGLink application.The inversion carried out is invariant mode, or a combination of TE (transverse electric) and TM (transverse magnetic) modes.One-dimensional inversion in the form of resistivity variations in the form of depth sounding and the direction of the electric field and magnetic field are of the same magnitude (  =  [28].The resistivity and thickness of each layer are parameters in 1-dimensional modeling [37].The results of 1-dimensional inversion modeling are in the form of a resistivity profile for each measurement point, while for 2-dimensional modeling, they are in the form of modeling the distribution of subsurface resistivity [38], [39].

Research methodology 2.3.1. Magnetotelluric (MT) data acquisition
The Kutai Basin MT data used in this research comes from country data acquired by the MT survey team of the Center of Geological Survey, Ministry of Energy and Mineral Resources.MT data includes a total of 40 measurement points.In this study, MT data was analyzed at 8 points out of 40 measurement points.The MT data covers an area of around 40×15 km 2 , with a distance between measurement points of 2-5 km 2 .

Magnetotelluric (MT) data processing
MT data processing is carried out using several software packages, including SSMT2000 software v. 0.6.0.65 (Phoenix Geophysics), MT-Editor v. 0.99.2.90 (Pheonix Geophysics), and WinGLink software v. 2.18, (Center for Geological Survey).When processing MT data, the SSMT2000 program is applied.The primary raw data (.TS3), (.TS4), (.TS5), (.TBL), (.CLC), and (.CLB) are processed in the first phase to transform the time domain to the frequency domain.The data (.TS) is the frequency range value data.In (.TS3) data has a frequency range of 40-320 Hz.In (.TS4) data has a frequency range of 5.6-33 Hz.In (.TS5) data has a frequency range 0.0034-4.7 Hz.The data of (.TBL) is the coordinate data of the measurement location.Next, there is calibration data contained in (.CLC) data as sensor calibration and (.CLB) data as tool calibration.The output of this process is sampling rate frequency data in the form of frequencies in the format fc3, fc4, fc5, fc6, and fc7.After that, a robust process was carried out to determine the best coherence of No Wight (NW), Rho Variance (RV), and Ordinar Coherence (OC) data using the MTeditor applicationm.The robust process is a statistical processing method using iterative weighting of residuals to identify and remove deviant data, or what is usually called noise [40].Coherence is the ratio of signal to noise that interferes with the recording.The output of this process is high-frequency data in the format (.MTH) and low-frequency data (.MTL), which contains information about resistivity and apparent phase.Then an XPR (crosspower selection) process is conducted on the data to remove frequencies that have low weight and produce the best coherence.After that, the file is converted into (.edi) format.The next step is to enter the file (.edi) into the WinGLink application for inversion.Before inversion, the trajectory is checked with the data using the Maps menu to find out the appropriate trajectory.
Inversion of 1-dimensional into 2-dimensional through different processes.For 1-dimensional data, the data is masked and smoothed to produce very smooth data.After that, a 1-dimensional inversion is carried out.For 2-dimensional inversion, this is done by entering the inversion parameters, namely the TAU value, and iterating to produce the smallest possible RMS.After that, 2-dimensional inversion is carried out.The output from 1-dimensional and 2-dimensional inversion is in the form of 1-dimensional and 2-dimensional cross sections, which are then interpreted by considering regional geological data.The diagram of MT data processing is displayed in Figure 2:

Results and discussion
There are eight measuring points for the MT data measurement stations, which form 1 measuring track in a northwest-southeast direction with a track length of 42×15 km 2 .The eight measuring points are at stations KT22, KT31, KT29, KT12, KT14, KT13, KT15, and KT40.From the station data, data processing is carried out to produce 1-dimensional and 2-dimensional inversion contour models.

One-dimensi inversion
One dimensional inversion has the following 2 processes 1. Sounding.
In this process, iteration is used at each station with 30 iterations, and the inversion used is Occam's inversion.

X-Section
At this stage, it functions to display the 1-dimensional results shown in Figure 3.The resistivity value in 1-dimensional inversion only varies vertically with depth.Figure 3 shows the different depths for each measurement station.Low resistivity values, or so-called conductive zones, were detected at depths of 1000-5000 m at stations KT22, KT31, KT29, KT12, KT14, KT15, and KT40.Apart from that, low resistivity is also found at depths of 8000-12000 m at KT13 station.There are also moderate resistivity values detected at almost every station.In addition, high resistivity values were detected at depths of 3000-5000 m at station KT13 and at depths of 1000-5000 m at stations KT22, KT29, and KT40.

Two-dimensional inversion
In order to get the best inversion results, the optimization process is conducted using TAU values and iterations.The TAU value used is 3, and the iteration is 30 iterations, which are expected to obtain the smallest RMS value.The data processing results obtained the smallest RMS of 1.4%.The results of the 2-dimensional inversion are shown in Figure 4.
The results of the 2-dimensional inversion show that the upper part of the layer has a low resistivity of 1 Ωm to 10 Ωm at a depth of about 1000-5000 m.The depth of the varying resistivity values is influenced by the Samarinda anticlinorium pattern.The pattern of the trajectory is interpreted to have a folding and incision structure.The low resistivity/conductive layer forms a structural pattern in the form of a faulted anticlinorium.It is this structure that identifies the reason new conductive anomalies are detected when the depth is more than 5000-10000 m for KT31, KT29, and KT13.For KT22, the conductive anomaly can already be detected when the depth is 2000-4000 m.For KT12, KT14, and KT15, the conductive anomaly can already be detected when the depth is 1000-4000 m.And, for KT40, the conductive anomaly can only be detected at a depth of 8000-11000 m.
The existence of low resistivity values or conductive zones can identify the potential for black shale layers.Station KT29 is included in the Pamaluan Formation, and stations KT31, KT12, KT13, and KT15 are included in the Pulaubalang Formation, which has the potential to store shale gas based on the inversion results because it has a low resistivity value and also because it is associated with the source rock of the Kutai Basin petrolium system (source rock), supported by previous research data Zajuli & Wahyudiono's research, 2018 [8] and research by Weckmann et al.2007 [41].

Conclusion
This research concludes that 1-dimensional inversion seems to be fairly well correlated with the 2dimensional model depicted in Figures 3 and 4. All stations in this study are located in an area suspected of being an unconventional reservoir by considering the results of low resistivity values of 2 ohm.m to 4 ohm.m at each measurement station.Shale gas may have been detected at station KT29, according to information, which belongs to the Pamaluan Formation, and at stations KT31, KT13, and KT15, which belong to the Pulaubalang Formation, because it is associated with the host rock of the petroleum system of the Kutai Basin, as supported by previous research data.Supporting data needed to increase the confidence value of the results of geological information on the surface, such as drill data and surveys using other methods.

Figure 1 .
Figure 1.Geological map of research area, Kutai Basin, East Kalimantan