1 INTRODUCTION

Underground hot water, as a kind of multifunctional mineral resource, has been widely used in industry, agriculture, and daily life. It is a multipurpose natural resource that integrates water, minerals, and heat or geothermal energy, which is a renewable and clean energy. This geothermal energy has a wide spatial distribution, good stability, and high efficiency. The utilization of geothermal energy has a long history. Geothermal energy can potentially contribute to the national strategy of “carbon neutrality” and “carbon peak” (Wang et al., 2023; Xie et al., 2022). Thus, exploitation of geothermal resources can play an important role in the development of a low-carbon economy (Li et al., 2020).

Geothermal resources are distributed in convergent plate margins, rifts, oceanic island hotspots, and large-scale extensional tectonic activity areas (Simmons, 2020) and various nondestructive geophysical exploration methods have been used to identify geothermal resources (Hudson et al., 2022; Susilawati et al., 2023; Uchôa et al., 2023; Yasin et al., 2023). With obvious advantages in deep exploration, geophysical technology can effectively detect deep hidden fault structures (Di et al., 2020), obtain important information such as strata and burial depth, and provide important information for the location selection of geothermal wells. Therefore, exploration of geothermal resources is the first step for development of geothermal resources in the future (Hu et al., 2011; Li et al., 2021; Zhang et al., 2021).

Geothermal resources have been explored by geophysical technologies such as gravity, magnetic, electric, seismic, radiological, and other methods and technologies. Pamukcu et al. (2007) used the significant negative correlation between free air gravity anomalies and aeromagnetic anomalies as an important indicator for regional underground geothermal prospects. The controlled source audio magnetotelluric technology (CSAMT) has been widely used to identify favorable structures for deep geothermal energy, thus indirectly finding geothermal resources (Arafa-Hamed et al., 2023; Di et al., 2020; Kumar et al., 2017; Wang et al., 2017). Erkan et al. (2008) used the temperature and pressure data of exploration wells to study the Chena Hot Spring in Alaska. They established a geothermal reservoir model and examined the fluid flow characteristics. Abiye and Haile (2008) investigated fluid circulation in the Boku geothermal system by using an integrated geophysical method with heat, gravity, magnetism, and electricity, and thus reconstructed the geometric structure of underground aquifer. Different geophysical methods have their respective application scope, which means that for any geophysical problem, there may be multiple solutions. Accordingly, use of a single exploration method has its limitations and uncertainties. The use of an integrated geophysical method is particularly important to overcome these limitations and uncertainties (Diao & Du, 2019; Han et al., 2018; Kouadio et al., 2020; Wan & Wang, 2023).

This paper describes successful application of an integrated geophysical technology to explore the geothermal resources in the Yueliang Bay of the Binhai County in Jiangsu Province. Both technology improvement and application procedure of this integrated geophysical technology can serve as good references for future exploration of geothermal resources.

2 GEOLOGICAL OVERVIEW OF THE EXPLORATION AREA

The exploration area of this project is located in the east of Binhai County in Jiangsu Province. Its strata belong to the Yangtze stratum area. The surface is covered by Cenozoic strata, but no bedrock outcrop is observed in this area. The loose strata (Quaternary Q + Neogene N) are about 300–500 m in thickness. As observed through geological drilling, the bedrock strata in the exploration area from old to new are Ordovician (O) shallow Marine carbonate rocks, Silurian-Devonian (S-D) Marine clastic rocks, Carboniferous-Permian (C-P) shallow Marine carbonate rocks, clastic rocks with coal-bearing clastic rocks, Cretaceous (K), and Paleogene (E) continental clastic rocks. Tectonically, the exploration area is located in the coastal uplift of the Subei Basin of the Lower Yangtze landmass and roughly bounded by F5 (Batan-Xiaojie fault) and the A3 fold (South Batan-Dayujian anticline). A2 (Binhai-Huaihe inverted steering cline) is located on the north and south sides of F5 (Figure 1). The F5 fault is the focus of this survey because it damages the integrity of A2 and A3 of the fold. Its strike is about 60° dip to the southeast, with the dip angle being about 50°–60°, and its length is about 34 km. It controls the boundary between the Ordovician system (O) and the Silurian system (S), and the southern side is mostly covered by the Paleogene system (E) (Liu et al., 2021).

The lithosphere thickness of the exploration area is about 80–100 km; the depth of the interior (560°C) is about 25–30 km; the earth heat flow value is about 70 mW/m2; and the geothermal gradient is lower than 27°C/km. According to the local geothermal gradient of 27°C/km, excluding the influence of other activities, the geothermal water temperature can theoretically reach about 42°C when the aquifer is at the burial depth of 1000 m. When the burial depth is 2000 m, the geothermal water temperature can reach up to 69°C approximately. The fault structures in the area are well developed, among which the NE trending fault F5 is a heat- and water-control structure in the exploration area. The deep part of the exploration area is the Ordovician limestone stratum, and the fissure formed under the action of a fault structure is a good water-bearing hot reservoir. The fissure layer with a fracture structure is the best target horizon for geothermal water resource exploration in this exploration area. The geothermal cover mainly includes the upper member of Yancheng Formation of Quaternary and Neogene, with a thickness of 240–380 m, and is interdeposited with clay and sand layers. Among them, the cohesive soil shows strong water insulation, relatively low thermal conductivity, and good thermal insulation. The above analysis indicates that the exploration area has favorable geothermal geological conditions for the formation of underground hot water in four aspects, namely, “source, ventilation, storage, and cover.”

3 PHYSICAL CHARACTERISTICS OF STRATA

The integrated statistics of regional resistivity and wave velocity are listed in Table 1. As is shown, the resistivity of strata rocks shows the following characteristics: the clastic rocks of Cenozoic and Upper Paleozoic have low resistivity, while the carbonate rocks have high resistivity. The resistivity of strata changes from low to medium, from low to high to medium, and from low to high from top to bottom; it decreases obviously when the strata are broken. The wave velocity shows the following characteristics: the difference between bedrock and overlying strata is obvious, and the wave velocity gradually increases from shallow to deep stratum. When the fracture causes the bedrock to move or break obviously, the stratum appears to move or miss, with the wave velocity declining noticeably. The above physical characteristics provide a good electrical prerequisite for the use of the electromagnetic method and micromotion to explore the formation structure and fault structure, and then infer the geothermal resources.

4 CSAMT AND TECHNIQUE

4.1 Implementation procedure for geophysical exploration

CSAMT has high construction efficiency and low cost, thus being suitable for large-area construction. Because micromotion construction is characteristic of low efficiency and high cost, this study used the CSAMT exploration as the primary construction for a rough survey and micromotion detection as a secondary review and evaluation for fine location identification. The exploration process was as follows: First, the abnormal location was preliminarily determined based on the results of the CSAMT method. Then, micromotion detection was carried out near the well location to further determine the geothermal well location.

Three parallel survey lines were arranged, all perpendicular to the fault structure F5 (Figure 2). According to the results of the CSAMT method, the location of geothermal wells was preliminarily determined, and micromotion detection was carried out near the well location (line L2) to further determine the well location of the geothermal well.

4.2 CSAMT method

The CSMAT method has two key processes: technical parameter determination and data processing.

4.2.1 Technical parameters of CSAMT

The CSAMT method uses an equatorial dipole device to observe the H y horizontal component of the magnetic field orthogonal to the field source and the E x horizontal component of the electric field parallel to the field source. The apparent resistivity of Carneia ${\rho }_{\mathrm{s}}$ is calculated as (Tang & He, 2005)

$${\rho }_{\mathrm{s}}=\frac{1}{5f}\frac{{\left|E_x\right|}^2}{{\left|H_y\right|}^2},$$

where f represents the frequency.

In this study, the CSAMT method adopts the V8 electric method system produced in Canada. Its transmitting dipole AB is 1.75 km long and it is arranged parallel to the measuring line. The distance between AB and the measuring line is within 8–10 km. The acquisition frequency used in measurement this time is 0.125–9600 Hz. The maximum transmitting current reaches 18A and the receiving point is 50 m away. Its field working layout is shown in Figure 3.

4.2.2 Data processing in CSAMT

In the data processing of CSAMT, AMT was introduced to carry out auxiliary correction and inversion constraint on near- and transition-field curves. This can restore their authenticity as far as possible (Wang et al., 2015).

Figure 4 shows a comparison of the curve at a point on the L2 survey line before and after correction with the AMT method. The original apparent resistivity curve at this point has severe distortion below 20 Hz, and the CSAMT curve after the full frequency domain apparent resistivity correction still has the problem of insufficient correction, although the data in the transition field and near field are corrected to some extent (Zhao et al., 2008). Combining the AMT apparent resistivity curve at this point, after correcting the corrected apparent resistivity curve again, the curve in the transition zone is smoother and the curve shape is closer to the AMT apparent resistivity curve. The apparent resistivity curves of CSAMT data after the correction of the transition field and the near field are basically equivalent to the AMT measurement results (Luan et al., 2018). A relatively perfect AMT processing method can be used to invert the corrected CSAMT data.

The effect of this correction on the identification of field structures is presented here. Figure 5a shows the CSAMT inversion profile without near-field correction. A high-resistivity basement is formed at an exploration depth of about 900 m. Caused by near-field data distortion, this obvious data distortion results in inconspicuous characteristics of a low-resistivity fault structure at 1.5 km below the profile. The shallow part is the data from the remote area, and the response to the formation and structure is still credible. Figure 5b shows a cross-section map of data inversion corrected by the apparent resistivity method of the full frequency domain that is directly calculated with the measured Carneia apparent resistivity. Compared with Figure 5a, the reflected high-resistivity basement depth is relatively increased, and the low-resistivity structure covered by the high-resistivity basement also appears initially. The stratigraphic boundary is more noticeable, but the depth of stratum is still somewhat distorted. Figure 5c shows the data inversion cross-section of the corrected apparent resistivity curve again after correction in combination with the apparent resistivity curve of AMT. After correction by the above method, the profile as a whole is well improved under the influence of the near-field effect. The characteristics of the low-resistivity fault structure below 1.5 km of the profile are clearer, while the characteristics of the pseudo-structure below 2.5 km of the profile are effectively suppressed, which improves the display of deep effective geological information and increases the effective detection depth. The electrical layer reflected by its resistivity is highly consistent with the stratigraphic division through later drilling.

4.3 Micromotion detection

The micromotion detection obtains the dispersion curve of the surface wave (Rayleigh wave) in the weak vibration signal through the circular array. It captures the velocity characteristics of the S-wave below the array and thus speculates stratigraphic and structural characteristics (Liu et al., 2019). An American A-tom single-station seismometer was used for this micromotion detection. During the observation, a 0.1 Hz geophone was used for signal collection. Thirteen collectors were set up around each measurement point to collect data by a quadruple circular array observation system, with a radius of 375, 750, 1125, and 1500 m. As shown in Figure 6, a total of six measurement points were arranged.

At the same time, the regular quadruple circular array was used to detect the Rayleigh wave dispersion curve in the micromotion signal by spatial autocorrelation (Xu et al., 2013). Figure 7 shows the dot dispersion curve diagram at point DK01. Its low frequency is around 0.3–0.5 Hz, and the dispersion curve is smooth overall, with good quality. On the basis of the dispersion curve calculation, the velocity structure of the apparent S-wave below the center point of the array was obtained by this micromotion detection. The velocity profile of the apparent S-wave was obtained by interpolating the velocity of multiple array measurement points (Xu et al., 2012).

5 INTEGRATED INTERPRETATION OF THESE DATA

5.1 Interpretation of CSAMT data

The three CSAMT results after auxiliary correction and inversion constraint of near- and transition-field curves are shown in Figure 8. The stratigraphic and structural characteristics within 3000 m depth are clearly reflected. Combined with the geological and physical property data in the exploration area, the strata within 3000 m reflected in the inversion resistivity cross-section can be roughly divided into three large electrical layers. From top to bottom, the first layer is about 300 m thick and has low-resistivity characteristics, which is presumed to be the reflection of the Quaternary and Neogene (Q+N) loose layer, and the thickness gradually decreases from the small end of the section to the large end, that is, from northwest to southeast. The thickness of the second layer is about 900 m, showing low-resistivity characteristics, which is speculated to be a reflection of Silurian (S) Marine clastic rock strata, from the profile of the small end to the large end. In other words, the thickness from the northwest to southeast is gradually becoming thicker. The third layer is thicker and has high-resistivity characteristics, which is speculated to be the reflection of Ordovician shallow Marine carbonate rocks with intercontinental clastic rock strata. The above results clearly indicate that the third layer is a water-bearing thermal reservoir and is the target of this geothermal exploration.

On the inference of fault structure, the sudden change of resistivity in the transverse or the steep low-resistivity anomaly in the longitudinal section of the resistivity profile is generally the main indication of the existence of a fault structure. There are transverse abrupt changes in resistivity on the inversion resistivity profile. These changes are caused by faults (Zhang et al., 2014). According to the three CSAMT inversion resistivity profiles, there are obvious signs of fault structures, and the fault structures on the three profiles correspond well in a spatial position, which can be inferred to be the same fault, namely, DF1 fault. The fault traverses 1.2 km of line L1, 1.25 km of line L2, and 1.2 km of line L3 and shows obvious low-resistivity anomalies. Among them, line L3 passes through wide water surface and construction site, and the interference is severe and the measurement line cannot be fully distributed, so there is still a lack of reflection of the fault structure. There are NW dip faults at the large end of line L1 and line L3, which can be caused by interference according to the site conditions.

5.2 Interpretation of micromotion detection data

Figure 9 shows the interpretation section diagram of the micromotion S-wave velocity to only reflect the formation relative velocity. Combined with the integrated analysis of geological data in the exploration area, the velocity within the micromotion profile presents an obvious stratified reflection. The velocity in the first layer is less than 470 m/s and the thickness of the layer is about 300 m, which is presumed to be the reflection of the Quaternary and Neogene (Q+N) loose layer. The second layer has a velocity of 470–1050 m/s and a thickness of about 800 m, which is presumed to be the reflection of Silurian (S) Marine clastic rock strata. The velocity of the third layer is higher than 1050 m/s, which is presumed to reflect the Ordovician shallow Marine carbonate strata. Near the depth of hole DR01, the velocity shows an obvious low speed, which is speculated to be caused by the strong water-rich formation caused by the fractures of rock layer caused by DF1 fault. These results, mutually confirmed with the results of CSAMT, are in good agreement.

5.4 Analysis of model features

The geothermal water in the exploration area is mainly replenished by atmospheric precipitation, followed by overflow replenishment of weak aquifers and vertical replenishment within the tectonic fracture zone. When the local water is stored in the tectonic fissure and karstic fissure, it can receive the heat from deep earth statically in situ (formed by natural warming, and the temperature and depth are roughly linear) and the trace elements in the surrounding rock are continuously absorbed. The controlling fault F5 (Batan-Xiaojie fault) is a northeast-trending tensile fault and in the open state of the fault, which is highly favorable to the migration and storage of underground hot water. This geothermal reservoir model in the exploration area is shown in Figure 10.

Geothermal sources generally belong to the normal geothermal field area. The geothermal source is mainly formed through natural warming and manifested in the form of water fluid. The deeper the water reservoir is buried, the higher the temperature. According to the local geothermal gradient of 27°C/km, excluding the influence of other activities, the geothermal water temperature can theoretically reach about 42°C when the aquifer is at 1000 m depth in this area. When the depth is 2000 m, the geothermal water temperature can reach about 69°C. The deep heat source forms the main geothermal channel along the fault or from the magmatic zone to the shallow part, and the water source migrates through the fault or fissure. The fault F5 in the northeastward direction is the heat- and water-control structure in the exploration area. Accompanied by long-term tectonic activities, it not only strengthens the communication with deep heat sources but also enhances the water richness of deep aquifers (Wang et al., 2020). At the same time, the deep Ordovician limestone strata form geothermal reservoirs. The abundant fissure is a good foundation for water-bearing thermal reservoirs. The Quaternary and Neogene deposits with the interlayer of clay and sand and the thickness of 240–380 m are good thermal insulation because of their strong water insulation and relatively low thermal conductivity (Tao, 2000), thus forming the geothermal cover in this area.

6 BOREHOLE VERIFICATION

Based on the results of the CSAMT method and micromotion detection, it is revealed that the stratigraphy and the inferred fault structure are basically the same. The results are in agreement with each other, achieving the effect of fretting detection on further evaluation and verification of the CSAMT method and making the preset borehole more reliable. DR01 drilling verifies the development of fractures and fissures in the 1217–2919 m well section. There are 16 layers of fissures with a cumulative thickness of 274.30 m, which can be used as the main water-bearing section of the geothermal well. The water temperature at the wellhead of this well is 51°C; the daily water volume reaches more than 2000 m3; and the depth of the final hole is 2919 m. The drilling situation is shown in Table 2. Figure 11 shows the comparison between the integrated interpretation profile of line L2 of the CSAMT method and the column chart of the well formation. As can be seen, the geophysical results are in good agreement with the data of the geothermal well. The inferred formation thickness and location of the water-bearing fracture zone are highly consistent with the borehole data.

The verification process also shows that the cations in geothermal fluid are Na+ and the anions are Cl−; thus, the geothermal fluid is Cl–Na-type water in water chemistry. The hot water in this area has a salinity of 3388 mg/L, which is classified as salt water according to salinity. The pH value is 7.5, indicating weak alkaline water according to the pH classification. The millimole equivalence of calcium and magnesium ions in the hot water is 11.41 mmol/L, which is classified as very hard water according to groundwater hardness. The water temperature of the well reaches 51°C. According to the temperature of geothermal resources, it is classified as warm water in low-temperature geothermal resources, which can be used for physiotherapy, bathing, heating, greenhouse, breeding, and so on.

7 CONCLUSIONS

This study applied an integrated method combining CSAMT for large-scale surveys with micromotion detection technology for local geothermal resource identification of geothermal resources in Binhai County, Jiangsu Province, China. A geothermal resource model was established and verified by the geological investigation of a drilling well. The findings of this study reveal that the introduction of Audio Magnetotelluric sounding Technology into the CSAMT method can largely improve the auxiliary correction to the curve of the near field and the transition field and the processing means of inversion constraint. It can improve the reflection ability of the results to the deep structure and strata, provide more accurate inversion inference results, and increase the effective detection depth. With CSAMT as the main method for large-scale exploration and micromotion detection as an auxiliary method for local recognition, the multiscale methods can verify each other and obtain deep electrical velocity and model features accurately, thus aiding in exploration of geothermal resources in the exploration area. As is revealed from the geothermal wells extracted based on the exploration results, the location and stratification of the water-bearing fault zone are highly consistent with the integrated geophysical exploration results. This confirms the effectiveness of this method and technology. The combination of CSAMT and micromotion detection shows obvious advantages in multi-scale geophysical exploration, and can be used as a preferred approach for further geothermal resource exploration.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflict of interest.