Abstract
Two hundred km of 2D seismic survey was carried out at the Lesser Himalayan Thrust Belts in Dailekh district, western Nepal. The main motivation is to elucidate the geologic relationship between the known oil and gas seeps, subsurface structure, and stratigraphy in the area. This is a challenging task which is from its extreme structural and geological complexity such as thrust faulting, tight folding, steep dip layers, and strong lateral variations in seismic velocity. Seismic data were acquired with SERCEL 428XL system and processed by GEOEAST computer software. In order to increase signal-to-noise ratio (SNR), suppress interference, and search for optimum acquisition parameters, a series of comparative tests on the different charge depth and size, group interval, CDP fold, geophone array, and single high-sensitivity geophone were conducted. We also tested 2S3L (two lines shooting and three lines receiving) wide line profiling. The results indicate that single hole with charge depth of 12 m, 4-16 kg charge size (less charge size for the densely populated areas), single high-sensitivity geophone, and 1S2L wide line profiling with 132 folds are the optimum acquisition parameters. On the basis of comparative process experiment, data processing workflow consisting of data preparation, prestack denoising, amplitude compensation, deconvolution, tomography static correction, velocity analysis, residual static correction, CRS stack, poststack migration, prestack time migration (PSTM), and prestack depth migration (PSDM) was selected. Maybe affected by problem of conflicting dip in complex media, CRS stack section does not show satisfactory geological characteristics. PSTM profile has moderate signal-to-noise (S/N) ratio; the shallow, medium, and deep continuous reflections can be observed in section. More details of the geological structures can be observed in PSDM section, especially in medium and shallow layers (less than 3000 ms or 4000 m), but PSDM method is more expensive and highly time consuming than that of CRS stack and PSTM. So, the PSTM section can be reasonably used for geological interpretation. By reference to field mapping, thrust characteristics, and MT data, the final interpretation to the PSTM section identified the interfaces of 6 geological units (Paleoproterozoic Nabhisthan Fm., Paleoproterozoic Dubidanda Fm., Neogene to Late Cretaceous Surkhet group, Late Carboneferous to Early Cretaeous Gondwana group, Mesoproterozoic Upper Lakharpata group, and Lower Lakharpata group) and delineated Main Boundary Thrust (MBT), Ramgarh Thrust (RMT), Padukasthan Thrust (PT), and Dailekh Thrust (DT). The bottom of Surkhet group which is our top target zone is about 4250 meters deep.
1. Introduction
This work is part of the ongoing joint project “Oil and Gas Resources Survey in Nepal” conducted by the Chengdu Center Of China Geological Survey (CGS) and the Petroleum Exploration Promotion Project/Department of Mines and Geology (PEPP/DMG) of Nepal. The main goals of the project are to better understand the petroleum geology and the subsurface structure of western Nepali Lesser Himalayan Zone and to define its hydrocarbon potential by combining existing data with new information obtained from this study and then to select one or two subsurface structure target for drilling.
The project was a team effort consisting of CGS (Chengdu Center) professionals in the fields of geology, geophysics (seismic, magnetotelluric method (MT)), sedimentology, petroleum geochemistry, petrography, and photogeology along with geophysicist from Qinghai Geophysical Prospecting Division, BGP Inc., CNPC, and geologists from the Petroleum Exploration Promotion Project of Nepal. Petrographic and geochemical sample collecting, laboratory analysis, interpretation, and modeling were conducted by the Chengdu Center Of China Geological Survey (CGS). Seismic data acquisition, process and interpretation were conducted by Qinghai Geophysical Prospecting Division, BGP, CNPC. The work done by the team consisted of (a) reviewing all petroleum-related data on Nepal; (b) conducting field trips throughout western Nepal to collect samples for geochemical and petroleum geological analyses; (c) collecting and analyzing oil and gas seep samples; (d) mapping field structures for selected critical areas (Dailekh and Tansen) in western Nepal; (e) implementing 200 km 2D seismic survey in Dailekh area; (f) completing 800 km MT survey in whole Nepal.
Finally, the regional thrust tectonic system including MBT, RMT, MT, DT, and PT was established in Dailekh. The sequence and age of the main strata Lakharpata group, Gondwana group, and Surkhet group have been redefined by new data. And it was reasonably considered that Lakharpata group, Gondwana group, and Surkhet group may exist under the RMT fault.
The Swat Formation marine shale of Surkhet group has a thickness of 50 m is the main source rock, its organic matter type was classified as I-II type, TOC content of it ranging from 0.7% to 2.24% and its value ranging from 1.83% to 2.07%, which indicate that organic matter was in the evolution stage of condensate to wet gas.
The Suntar Formation fluvial sandstone and Meipani Formation shore sandstone of Surkhet group are the main reservoir rocks. The sandstone of Suntar Fm. has a thickness of 233 m is medium grained feldspathic quartz sandstone. The pore type of it is intergranular and intragranular dissolved pores, with an average porosity of 9.2%. The thickness of Melpani Fm. shore quartz sandstone was measured to be about 50 m. The porosity of quartz sandstone ranging from 7% to 10% and asphalt fillings can be observed in the sandstone.
Seismic data were collected using dynamite (at depth of 12 m high-speed layer excitation, single hole, 4 kg-16 kg charge size) as a source of seismic energy. Acquisition geometry adopts 1S2L wide line profiling with 132 folds, single high-sensitivity geophone reception, 30 m channel spacing, and maximum offset is 5925 m.
Seismic data were processed by a processing workflows of tomographic statics, energy compensation, surface wave attenuation, abnormal amplitude attenuation, abnormal amplitude compensation, linear noise attenuation, surface consistent deconvolution, stacking velocity analysis, residual statics, predictive deconvolution, comprehensive global optimization residual statics, CRS stacking, CMP stacking, PSTM velocity field, and PSTM migration.
Finally, high-quality seismic sections that can be used for geological interpretation were obtained; six sets of geological units (Paleoproterozoic Nabhistan Fm., Paleoproterozoic Dubidanda Fm., Surkhet group, Gondwana group, Upper Lakharpata group, and Lower Lakharpata group) and thrust faults such as MBT, RMT, MT, DT, and PT were identified. The trap structures that can be drilled were selected. The Nabhishtan Fm. (phyllite), Dubidanda Fm. (quartzite), RMT fault, Suntar Fm. (feldspathic quartz sandstone), Swat Fm.(shale), and Melpani Fm.(quartz sandstone) were encountered from top to bottom. The bottom of the main target layer Melpani Fm. of Surkhet group was predicted to be 4250 m deep.
Unfortunately, due to the pandemic, the drilling work originally scheduled to be completed in 2021 has not been completed yet.
2. Geographical/Geological Background
The Federal Democratic Republic of Nepal is between China to the north and India to the south, as shown in Figure 1. From east to west, it is about 800 km with width from 130 km to 230 km. The main thrusts in the Himalayan fold thrust belt include the Main Frontal Thrust (MFT), Main Boundary Thrust (MBT), Main Central Thrust (MCT), and South Tibetan Detachment (STD) (not shown on Figure 1). These faults divide the region into five geographic/tectono-stratigraphic zones which include Terai plain, Subhimalayan (or Siwalik Fold Belt), Lesser Himalayan, Greater Himalayan, and Tethyan (or Tibetan) Himalayan zone (not shown on Figure 1) [1, 2].
The Terai plain is a part of the Indo-Gangetic Plain that extends from the southern Indian Shield to the northern Siwalik Fold Belt (Subhimalayan). The plain is underlain by a thick sequence of Mid-Late Tertiary molasse (Siwalik group), which unconformably overlies sediments of early Tertiary to Proterozoic Surkhet group, Gondwana group, etc., and igneous and metamorphic rocks of the Indian Shield [3].
The Subhimalayan Zone (or Siwalik Fold Belt) ranges in width from 5 km to 45 km is bounded by Main Frontal Thrusts (MFT) to the south and Main Boundary Thrusts (MBT) to the north. It consists of thick beds of folded and fault-repeated Tertiary molasse (Siwalik group) [4]. The Siwalik group has been informally subdivided into three units, and the entire package of sediments in the Siwalik group is unmetamorphosed and coarsens upward [5].
The Lesser Himalayan is a wide, complicated structural zone. It is located between the Main Boundary Thrust (MBT) in the south and the Main Central Thrust (MCT) in the north. Separated by regional unconformities, this zone is divided into three stratigraphic successions, which are the Lesser Himalayan sequence, the Gondwana sequence, and the Eocene-Lower Miocene foreland basin sequence. Detrital and igneous zircon U-Pb ages indicate that the Lesser Himalaya in Nepal is Proterozoic in age [6]. Numerous other thrusts and normal faults cut Lesser Himalayan rocks, most importantly the Ramgarh thrust [7, 8].
The Greater Himalayan Zone (Higher Himalayan Zone) lies between the Main Central Thrust (MCT) to the south and South Tibetan Detachment System (STDS), which is not shown in Figure 1) to the north. The zone is composed of metamorphic Proterozoic rocks overlain by a conformable sequence of Cambrian to Eocene Tethyan sediments.
Within different Himalayan tectono-stratigraphic zones, the stratigraphy has been informally subdivided into several units. Researchers use different nomenclature to describe those units. In this paper (see Table 1), we employ the description of the stratigraphic units described mainly by GEOLOGICAL MAP of Nepal (scale 1 : 1000000), unpublished Geological Map of Parts of DAILEKH and SURKHET Districts (compiled by PEPP/DMG, scale 1 : 50000), and other authors [1, 7, 9, 10].
3. Seismic Data Acquisition
Before 1994, a total of 5263 km length of 2D seismic profiles were obtained from four geophysical surveys in Terai plain [11, 12]. In eastern Terai plain, the seismic control is quite densely (approximately ), and the other part of the Terai plain is covered by the grid () of seismic lines [9]. There is no reflection seismic profile in Himalayan Thrust Belts.
Nepal is essentially unexplored. The only well was drilled in the far eastern Terai plain of the country near Biratnagar. The well was abandoned at 3530 m; only small amount of background (mud) gas were encountered [9].
The Dailekh region, which belongs to the Lesser Himalayan Thrust Zone, is a seismic exploration blank area in western Nepal. The Precambrian-Lower Paleozoic strata outcropped to the surface. The topography is characterized by high mountains with a drop of 400-1500 m, steep slopes, narrow valleys, deep ditches, and dense forests. Therefore, the seismic operation is very challenging due to strong topography.
The geological/geophysical characteristics of Dailekh Lesser Himalayan Thrust Zone is similar to that of the thrust belt on the northern margin of Qilian Mountains (the cradle of Chinese petroleum industry) in northwestern China: (a) the landform features are basically the same; (b) both of them belong to thrust nappe structure with high and steep fracture; (c) Precambrian-Mesozoic old strata or metamorphic rocks are mainly exposed in the region; (d) the anisotropy of velocity field is strong [13, 14].
Thousands of kilometers of 2D seismic profile and hundreds of square meters of 3D seismic survey with high SNR and high fold were conducted by oil company in Qilian Mountains thrust belt. So, the major acquisition parameters of Dailekh seismic profiling were developed based on the results of previous experience and field tests. Improving the reflected energy of the footwall of overthrust nappe and the imaging of folded deformation strata are the key problems in this area. The conventional wide line and large combination seismic acquisition technology has shortcomings in improving the seismic wave energy of the overburden zone, the adequacy of wave field sampling, the accuracy of seismic data, and the in-phase superposition of high and steep structures. So, low-frequency acquisition, symmetrical sampling, and wide line observation techniques are proposed. Low frequency acquisition is used to improve the reflected energy of the footwall of the nappe. Symmetrical sampling is used to realize full sampling of shot gather and trace gather [15]. Single-point excitation and reception provide high fidelity first break and reflection data for improving the accuracy of tomographic static correction and the imaging effect of folded strata. The combination and superposition of small and wide line bin can improve the signal-to-noise ratio.
The layout of seismic lines is shown in Figure 2. Seven of ten seismic profiles acquired in this region are oriented in north-northeast to south-southwest, which is roughly orthogonal to the strike of main structure. The other three seismic profiles are orthogonal to these seven lines. Hard conditions and high mountains prevent us to extend the survey line farther to the northeast or southwest. The acquisition parameters were optimized by a series of tests with line DK2019-5SN, which is more than 20 km long and traverses across main kinds of protoliths and gas seeps.
3.1. Source Test
The site for source test is located at a place where Nabhisthan (Nb) Fm. protoliths (phyllites and quartzites) are exposed. The dynamite source was used due to its lower cost and uniform source signature. The civil buildings in the working area are simple, and antivibration is not strong, so a test with small charge size in shallow well was also carried out to provide information for the safety in the populated areas. Data were recorded with SERCEL 428XL and analyzed by KL Seis software. More details are listed on Table 2.
With 16 kg (excited in the high-speed old formation) charge size, we tested different well depths of 10 m, 12 m, 15 m, and 18 m. A single shot record and quantitative analysis results are shown in Figure 3. The results indicate that effective reflection can be seen in raw seismic shot gather. There is no significant difference of seismic data with different source depths at 10 m, 12 m, 15 m, and 18 m. Quantitative analysis shows that energy and SNR with 10 m source depth are slightly poor, and frequency band width is relatively narrow, while energy and SNR with 12 m, 15 m, and 18 m are very similar and better than those with 10 m source depth.
We tested different charge sizes of 10 kg, 12 kg, 14 kg, 16 kg, and 18 kg by fixing the source depth of 12 m. An example shot record and quantitative analysis results are shown in Figure 4. The results show that the seismic data quality with charge sizes of 10 kg, 12 kg, 14 kg, 16 kg, and 18 kg is equivalent, and seismic energy increases with larger charge size, while the SNRs are very similar.
Four smaller charge sizes of 1 kg, 2 kg, 3 kg, and 4 kg with shallow source depth were also tested. The segmental frequency scanning results of single shot and quantitative analysis are shown in Figure 5. The first break of raw data is clear at near offsets and weak at farther offsets due to energy decay. The main frequency is below 15 Hz, and the effective frequency band of raw data is 0-23 Hz. The bigger the charge size, the stronger the seismic energy. Interestingly, the SNRs are very similar.
Two types of geophone arrays were tested. In both cases, conventional vertical geophones with 30DX-10 Hz and individual sensitivity at 70% of damping of 28 v/m/s were used. One array used rectangle pattern while the other is rectangle pattern with 5 m geophone spacing. For comparison, a single vertical geophone with SG5 Hz and individual sensitivity at 70% of damping of 80 v/m/s is also tested (Figures 6 and 7).
Overall three shot records look very similar, although they have subtle difference. The dominant frequency of geophone array can suppress the ambient noise better, but there is a small time shift and frequency content difference with geophone arrays compared to single geophone receiver. The spectrums of three shot records seems have same dominant frequency, while the spectrum of the single high-sensitivity sensor shows more subtle features and changes. The 20-geophones array has the highest signal-to-noise ratio (SNR), and the SNR of the 10-geophones array is a little lower than that of single high-sensitivity sensor. Both types of geophone arrays have the equal RMT amplitude levels but the single high-sensitivity sensor has the highest RMT amplitude value.
3.2. Wide Line Profiling Test
The line DK2019-5SN of 20.46 km was chosen as wide line profiling for comprehensive evaluation of acquisition parameters based on technological and economical consideration. The survey specifications for the wide-line profiling test are summarized in Table 3.
3.2.1. Analysis of Different Fold and Different Geometry
The stack sections with different folds are presented in Figure 8. The six sections look very similar, although fine distinctions could be existed among them. From top to bottom, the reflections could be observed in all sections, but detail of structures could be revealed only in these sections with higher fold such as 132 folds. With the number of fold coverage increased to more than 132 times, there is no obvious discrepancy existed among these sections. The quantitative analysis of the sections with different folds are shown in Figure 9. The results indicated that there is a significant positive correlation between the number of fold coverage and signal-to-noise ratio, the higher the number of fold coverage, and the higher the signal-to-noise ratio. And compared with single line acquisition scheme, wide line measurement shows great advantages in improving signal-to-noise ratio. Interestingly not only the six spectrum curves of sections with different folds look very similar but also the six RMS amplitude of sections with different folds almost have the similar values.
The stack sections acquired by different layout chart were presented in Figure 10. Several reasonable reflections could be revealed in stack sections by different geometry of 1S1L, 1S2L, 1S3L, 2S2L, and 2S3L. It seems that with the increase of shot line or receiving line, the reflection characteristics of middle and deep layers become much better. So, 1S2L geometry was chosen for final acquisition.
3.2.2. Analysis of Maximum Offset
The sections with different maximum offsets were presented in Figure 11. The comparison of profiles with different maximum offsets indicates that with the increasing of the maximum offset, the reflection characteristics of the middle and deep layers are slightly improved, but there is no significant difference once the maximum offset reaches 5925 m or longer. The quantitative analysis of the seismic sections with different maximum offsets displayed that with the increasing of the maximum offset, the value of signal-to-noise ratio increases a little (Figure 12), and the three spectrum curves and three RMS amplitude values of sections with different maximum offsets look very similar, although they have subtle difference.
Based on the above analysis, the final survey specifications for the 2D seismic lines at Dailekh area are listed in Table 4. Finally, high-quality raw single-shot records are obtained in Less Himalayan Thrust Belt, which lays a good foundation for further data processing and interpretation.
4. Seismic Data Process
Seismic data processing workflows is presented in Figure 13. The seismic data processing software package GEOEAST (developed by BGP, CNPC) was used for all of data quality control and processing.
The emphasis of data processing focus on static correction, prestack fidelity denoising and high-precision imaging.
Strong topography, noise, complicated lithology, and structure at near surface deteriorate seismic data quality. The near-surface tomography velocity model inverted from the first arrive travel time picks can depict the near surface structures more accurate and in more details [16–18]. Elevation statics correction was performed to correct the mistakes associated with the variable source and receiver elevations by setting the data onto a common datum. Refraction statics corrections were used to solve the problem associated with the variable thickness and velocity of surface weathered strata [19]. Tomography static correction was used to solve the medium and long wavelength static problem. The stack sections by applying elevation statics, refraction statics, and tomography statics corrections are shown in Figure 14. Detailed comparison of the three sections shows that tomography static correction improves the S/N ratio and continuity of reflections.
To noise which amplitude is several times larger than the average energy threshold of the signal, we take small threshold value to suppress it firstly, the smaller the threshold value is, the stronger the suppression is. Then, the surface consistency compensation is carried out, and the amplitude difference will be smaller, and the threshold strain will be larger. Then, the surface consistent deconvolution is performed to improve the influence of wavelet difference caused by the surface.
Why do we use two kinds of deconvolution methods?
The surface excitation lithology in this area is mainly metamorphic rock, and there are some differences in the transverse direction of wavelet. Therefore, the surface consistent deconvolution (24 ms predictive distance, 256 ms operator length) was used to solve the wavelet difference caused by the surface. The predictive deconvolution (12 ms predictive distance, 256 ms operator length) was carried out to compress the wavelet to improve the resolution of low frequency data. Integrative global optimal residual static correction was performed to further improving profile quality.
To improve the imaging quality, we applied the CRS algorithm to the conventionally processed data immediately prior to stacking [20–22]. As an extension of the travel time equation to fully exploit the information contained in the prestack data, the CRS stack strategy was developed by Jäger et al. [23] and was originally introduced as a data-driven method to simulate zero-offset sections. In the CRS stack method, the multiparameter moveout equation describes a surface in the time-midpoint-half offset domain (in 2D), rather than a trajectory in the conventional common midpoint (CMP) stack [24]. The CRS hyperbolic equation with its three attributes in 2D reads [25]:
Here, is the half-offset, is the midpoint displacement with respect to the considered CMP position, corresponds to the ZO two-way travel time, and is the near-surface velocity. The emergence angle of the central ray is given by , and the radius of curvature of the normal (N) wave and the normal incidence point (NIP) wave are indicated by and , respectively. In application of the CRS stack, optimum values for these three parameters () are automatically determined independently for each sample (). This is realized by varying the three parameters () and thus the operator shape, and performing a coherence analysis along the stacking operator in the multicoverage data. The three parameters yielding the highest coherence are the desired parameters [26].
The CMP stack and CRS stack results were presented in Figure 15. From top to bottom and from right to left edge, the reflections could be clearly observed in CRS stack section, but same appearance could not be observed in CMP stack section. CRS section obviously has a better signal-to-noise ratio than that of CMP section. Although CRS section could reveal more detail of the cross reflection, little faults, and mixing reflection, the diffraction wave cannot be well positioned, and imaging precision was not significantly improved. It seems reasonably to suppose the CRS stack method face with some problem in imaging of complex structures, especially in the existence of conflicting dips.
For the needs of high-precision imaging of complex overthrust nappe structures in the study area, we used the Kirchhoff integral prestack time migration, which has simplicity, efficiency, feasibility, and target-orientated property, low sensitivity to velocity model, to improve imaging accuracy [27]. Meanwhile, the experimental application of Kirchhoff integral prestack depth migration was carried out in this area.
To perform PSTM, a sufficiently accurate RMS velocity model was needed. The velocity model was obtained by iterative velocity analysis method [28]. First, use the stack velocity model to perform the prestack time migration and check whether the effective reflection events in the output CRP gathers are flattened. If the CRP gathers are uneven, Inverse NMO will be done with migration velocity. Repeat the above work until the effective reflection events of CRP gathers are flattened to obtain the final migration velocity. In consideration of the strong regional heterogeneity and large velocity variation of the studied area, the final migration velocity is obtained by two estimation stages: through the iterative velocity analysis method, the RMS velocity that can flatten the CRP gathers within 30 degrees is obtained. Then, the isotropic prestack time migration is performed using the RMS velocity field to obtain the migrated gathers. On the gathers, ETA spectrum analysis is performed to flatten all the CRP gathers. To the prestack depth migration velocity analysis, the velocity model was obtained by iterative layer velocity of constrained velocity inversion (CVI) technique [29, 30]. This prior information had been used as constraints like as RMS velocity, knowledge of the overthrust nappe structures, and layer interpretation. The procedure had three main stages: layer velocity conversion by CVI, iterative velocity model along layer, and final model generation. During iteration, repeat picks of CIP residual velocity and prestack depth migration until all the CIP gathers flattened. The final velocity model is shown in Figure 16.
(a)
(b)
The bend-ray traveling time computation that is based on hypothesis of horizontal layer, is more close to the real situation of earth structure, and can improve the precision of traveling time [31]. Kirchhoff bending rays prestack time migration (PSTM, velocity analysis density is 20 CMP; migration aperture is 12000 m; antialiasing distance is 15 m) and Kirchhoff integral wave equation prestack depth migration (PSDM; grouping offset is 30 m; migration aperture is 7000 m; interval velocity is by stack velocity spectrum) were applied to seismic data to overcome many typical drawbacks of CMP processing such as positioning errors and collapse diffractions and to improve lateral and vertical resolution of images [32–34].
The results of prestack time migration (PSTM) and prestack depth migration (PSDM) are presented in Figure 17. The reflection characteristics of depth migration (PSDM) are better than that of time migration (PSTM), especially in medium and shallow layers (less than 3000 ms or 4000 m). Because there is no well log data to constrain velocity and horizon updating in depth migration, geological structures may still not be correctly located.
The finally obtained prestack time migration profile has moderate S/N ratio; shallow, medium, and deep reflection (the deepest reflectivity down to 8 s) is complete. The effective wave band ranges from 3 to 40 Hz.
5. Seismic data Interpretation
Final prestack time migration (PSTM) sections were used for seismic interpretations. Because there is no well log in this area, and the surface is outcropped basically by Paleoproterozoic Nabhistan Formation, time migrated imagines cannot be corrected by well data information. Tracing seismic horizons was difficult and less accurate in some areas due to locally poor quality of seismic data. In some seismic sections, the seismic solution is very low. In some profiles, quality was highly influenced by faults which cause some ambiguities in geological interpretation [35]. Based on the theory of thrust, geological outcrop, stratigraphic thickness, interval velocity analysis, and seismic reflection characteristic (blank reflection or weak reflection is the seismic characteristics of mudstone or siltstone; low frequency and strong reflection is generally of large gravel layer; random reflection is the seismic characteristics of metamorphic rock), the interfaces of 6 geological units are inferred, and several kinds of thrust faults are recognized in sections. The key point of the interpretation scheme is that the Paleoproterozoic metamorphic strata (Nabhisthan Formation and Dubidanda Formation) are superimposed on the Neogene to Late Cretaceous sedimentary strata (Surkhet group) along the RMT fault. But this interpretation result is needed to be verified by drilling.
There are two seismic lines (DK2019-5SN and DK2019-2EW) passing through the core part of the Dailekh anticline structure, and the intersection point is the selected drilling well location. The NNE-SSW trending section DK2019-5SN has 20.46 km long (Figures 2 and 18) in Dailekh district. The seismic section DK2019-2EW is a 25.32 km-long seismic line trending in near east-west direction perpendicular to the DK2019-5SN. Six different geological horizons with different seismic properties have been identified in both sections, ranging in age from Neogene to Paleoproterozoic. These horizons, from top to bottom, consist of Nabhisthan Fm., Dubidanda Fm., Surkhet group, Gondwana group, Upper Lakharpat group, and Lower Lakharpat group. The stratum of Surkhet group is slightly uplifted and has the thickness of about 2100 m, and a gentle anticline was observed in Surkhet group sequences, while the Lower Lakharpata group stratum is characterized by the depression. It is speculated that the stratum of Gondwana group here has reached the maximum thickness. Two large thrust faults (MBT and RMT) and several small thrust faults with little offset are observed in seismic sections. Both MBT and RMT have typical fault plane reflection characteristics. MBT is obviously inclined to the north in section DK2019-5SN.
MBT (Main Boundary Thrust) is a relatively high and steep thrust nappe fault near the surface. It is a large thrust fault with NE dip and NW-SE strike. On the profile, the fault extends into the basement, and the buried depth is more than 15000 m (6000 ms). Due to the limitation of topography, the seismic line cannot pass through the exposed area of MBT, the direct connection between the reflection and MBT outcrops is not seen, so the MBT can only be interpreted speculatively according to the burial depth and the reflection characteristics.
The RMT (Ramgarh Thrust) is a thrust fault located at the southwest part of the working area. The faulting surface is dipping in NE direction and striking in NW-SE direction. The Oligocene-Early Miocene Suntar Formation (footwall) exposed in the southwest part of the fault, and the Paleoproterozoic Dubidanda Formation (hanging wall) exposed in the northeast part of the fault. The fracturing distance of the fault is quite large, and the fracturing surface is the bottom of Dubidanda Formation.
PT (Padukasthan Thrust) fault is located in northern part of the working area and exposed partly to the surface. DT (Dailehk Thrust) fault is a large reverse fault extending to the surface, which is distributed around the working area and is generally dipping in northward direction. It is speculated that PT fault and DT fault may be the secondary faults of RMT fault. However, it can be seen from several locations that PT fault cuts off RMT fault. Therefore, PT fault should be formed very lately.
6. Discussion
In the first stage of petroleum field exploration activities in complex structure belt such as Dailehk district, the detail of subsurface structure, geometry, steep dip faults, and the distribution of target reservoir formation could be revealed by geophysical modeling. Conversional integration and inversion methods of magnetic and gravity data would result in uncertainty and inconsistency in complex geological media, because magnetic model could reveal near-surface structure, and deep anomalies could be modeled by gravity data. A novel strategy by simultaneous gravity and magnetic inversion followed by fusion procedure was developed to construct a geological model for the Makran subduction zone in southeast of Iran [36]. While preserving information both from near surface and subsurface structures, the fused model could be able to provide a better view of Makran subduction for geological interpretation.
Generally speaking, in initial stage of petroleum exploration, precise reservoir characterization is impractical due to presence of insufficient data. But an artificial intelligent (AI) method called knowledge-based seismic inversion could be used in reservoir geological characterization modeling [37]. In the proposed approach, they use seismic data to construct the porosity distribution model for the selected petroleum reservoir using the AI method. The final porosity model obtained by knowledge-based seismic inversion AI method is practically applicable for fractured reservoir characterization and petroleum fields with spatially sparse wells.
For geologic and stratigraphic interpretation of thin resource or reservoir layers and fine-scale geologic feature such as shale or sandstone intervals, fractures, domes, and wedge shapes, the seismic vertical resolution enhancement (VRE) technique could be applied. A variety of VRE methods and strategies are available to solve the problem of thin-layer interpretation in seismic data [38]. Alaei et al. [39] proposed a novel strategy of nonstationary scale transformation filter for VRE. The application of this method to three seismic data sets showed that it not only increases the bandwidth frequency of the data but it also increases the dominant frequency of bandwidth compared with the Gabor deconvolution method. The result showed an enhancement in the resolution adequate for thin-layer seismic interpretation.
In addition to the methods and technologies mentioned above, the problems existing in seismic exploration should be carefully analyzed and solved in Dailekh district. Here, the surface is outcropped basically by High-Speed Paleoproterozoic Nabhistan Formation metamorphic rocks. The stack velocity spectrum clearly shows the characteristics of transition from high-velocity layer of metamorphic rock to low velocity layer of sedimentary rock. The magnetotelluric (MT) profile along the line DK2019-5SN also shows that the near-surface high-resistivity strata overlays upper the underground low-resistivity strata. It seems reasonable to suppose that the low-speed layers of the Neogene to Late Cretaceous sedimentary rocks may exist deep underground.
The excitation lithology is basically hard metamorphic rock. The strong impedance between the near-surface high-speed metamorphic rocks and the underlying low-speed sedimentary rocks prevents the penetration of seismic energy into target zones [40, 41]. The shielding effect of nappe and the rapid attenuation of high-frequency vibration make the high-frequency energy very difficult to be transmitted into the underlying strata (Meihou [42]). As a result, the seismic reflection frequency band is narrow, and frequency is low. Therefore the S/N ratio of seismic data is not very high.
How to improve the effective frequency of seismic wave in nappe structural belt?
At first, small charge size and multiple excitation [43–45] can be used to reduce the energy wasted due to the damage of the borehole wall rock and the source interference as well. But the cost is very high; secondly, to use the high-sensitivity geophone with lower frequency to expand the effective frequency band of seismic wave and improve the signal-to-noise ratio of data.
The original seismic data in this area have moderate signal-to-noise ratio. The imaging accuracy of PSTM section is generally better than that of poststack migration section, but the improvement is not significant. On PSTM section, the reflections are often interfered by many cross branches, distortions, and discontinuities. The PSTM section shows that the structure of the area is relatively complex, the stratigraphic reformation is very strong; folds and small faulting blocks develop. The prestack depth migration (PSDM) can improve the imaging of steep dip structures, but at the same time, migration smiles are more obvious [46]. The accuracy of seismic imaging needs to be improved.
To improve the imaging accuracy of seismic data in thrust belt, not only common diffraction surface (CDS) stack method [47], common-offset common diffraction surface (CO CDS) stack method [48], partial diffraction surface stack method, and 3D partial diffraction surface stack method [26] but also the 3D seismic survey [49, 50] and 3D prestack depth migration (PSDM) processing could be applied.
7. Conclusions
Seismic exploration in Himalayan thrust zone faces many unprecedented difficulties, including superimposed multistage thrusting structures, strong near-surface heterogeneity, complicated seismic reflection wave-field, and a densely inhabited urban environment. Based on the experience of seismic exploration in the Qilian Mountain Thrust Belt in Western China, we have conducted a series of field tests on the charge size, charge depth, the pattern of geophone array, and the wide line profiling to obtain the optimum acquisition parameters and technical schemes, which are technically, economically, and safely feasible. The prestack denoising, tomography statics, CRS stack, velocity analysis, prestack time migration (PSTM), and prestack depth migration (PSDM) are several major steps in data processing. Among these stack and migration sections, the imaging of CRS stack section is not very good, the PSDM section is good, but this method is too expensive to be applied, the prestack time migration (PSTM) section is more desirable for geological interpretation.
A reasonable geological interpretation scheme was suggested on the basis of field map, thrust characteristics, and MT materials. Several kinds of thrust faults such as MBT (Main Boundary Thrust), RMT (Ramgarh Thrust), and DT (Dailekh Thrust) were recognized, and six major reflections (T-Nabhisthan Fm., T-Dubidanda Fm., T-Surkhet Group, T-Gondwana Group, T-Upper Lakharpata Group, and T-Lower Lakharpata Group) were identified. The bottom of the T-Surkhet Group, our top target zone, is about 4250 m in depth.
Data Availability
The raw seismic data used to support the findings of this study have not been made available because the agreement signed between the Chinese government and the Nepalese government prohibits the disclosure of research data.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
Acknowledgments
The project was funded by the Chinese Government and supervised by the Development Research Center of China Geological Survey (DRC/CGS) on behalf of China International Development Agency. This work was operated by the Chengdu Center of China Geological Survey (CC/CGS) and Petroleum Exploration Promotion Project/Department of Mines and Geology (PEPP/DMG) of Nepal. We thank all project partners for their collaboration and support. Dr. Shaowu Wang and Dr. Paul Li helped to greatly improve the manuscript.