Introduction

Time domain induced polarization (TDIP) is a geophysical exploration method widely used in various fields such as metallic and non-metallic mineral exploration1, groundwater search2,3, oil and gas development, geothermal and urban geological survey, etc., and has been the subject of many technical studies, innovations, and improvements over the years4,5,6,7,8,9,10,11,12,13. The depth and accuracy of exploration targets have been placed under higher requirements with the current new round of prospecting work14,15,16,17,18. The conventional IP gradient array survey is characterised by high efficiency. However, it is limited by the use of a single fixed transmitter electrode spacing, which is used mainly for reconnaissance as it takes less time than most other resistivity/IP arrays. It only reflects the characteristics of electrical distribution at a certain depth of the subsurface. It is unable to depict the changes in the electrical structure at different depths and is not comprehensive and detailed enough to show the distribution of the electrical resistivity and polarisation of the subsurface medium19,20,21. Other arrays using multiple transmitters along line that provide information about source depth (like conventional dipole–dipole surveys) need to prepare many transmitter locations at fixed measuring points, which has a low efficiency. In recent years, we have worked on the development of multi-transmitter gradient surveys utilising a number of groups of transmitter electrodes with different bipole spacing. This type of array takes advantage of the characteristics of the IP profile survey and IP sounding. This is especially useful when combined with high powered current-supply. This method efficiently collects information at different depths in the study area with high spatial resolution and large detection depths22,23.

The Huaniushan mining area is a major polymetallic (silver-lead–zinc-barite) metallogenic belt in the Beishan area of Gansu Province. The mineral district has experienced much attention and exploration activity in recent years. It has been the subject of many recent studies using a variety of methods24,25,26. Most of them are based on geological and geochemical exploration methods, with fewer studies that involve gravity and magnetic geophysical surveys. Huaniushan mining is located in the Gobi Desert area at the junction of Gansu Province and Xinjiang Province. It has a continental semi-desert climate, perennial aridity, almost no surface water, and a high near-surface resistivity. These factors make it difficult to carry out electrical measurements in the area. Power output from many common, conventional IP systems is limited. Transmitter currents are often very low where transmitter injection sites are in highly resistive, dry surface layers, making it challenging to obtain good quality resistivity and polarisation data. In order to address the challenges of low power supply, poor penetration depth and sparse spatial resolution of the conventional IP method in the Gobi Desert, we utilize high-power survey equipment combined with multi-spacing measurement electrodes, arranged in a 'multi-transmitter gradient surveys array’ configuration to carry out 3D IP measurements in the Huaniushan mining area. We obtain geoelectrical property information at a range of depths by arranging power supply electrodes with different pole distances in the observation area. Subsequently, 3D inversion modelling was performed using the acquired TDIP measurements to obtain a high-resolution 3D electrical structure of the subsurface in the study area. Additionally, we combined geological and drilling information from the mining area to construct a 3D geological-geophysical model, interpreting the favorable layers and spatial distribution of the polymetallic deposits.

Methodology

The development of multi-transmitter gradient surveys array is based on the conventional gradient array method. Depending on the target body direction and the working conditions in the field area, the transmitter electrodes may be laid perpendicularly to the direction of the tectonic structure (transverse), or parallel to the direction of the tectonic structure (longitudinal) or both, i.e. parallel and perpendicular (transverse, longitudinal). By changing the distance between the power supply electrodes, it is possible to obtain electrical information at different depths underground. This methodology is particularly suitable for high-power transmitters.

When collecting data using the multi-transmitter gradient array the two main parameters acquired by 3D excitation measurements are apparent resistivity (ρs) and apparent polarisation (ηs). The apparent resistivity is calculated as:

$$\rho_{s} = K(A,B,M,N)\frac{{\Delta U_{t} }}{I}$$
(1)

Equation (1) ρs represents the apparent resistivity in ohm.m, A and B are the power supply electrodes, M and N represent the measurement electrodes. K(A,B,M,N) represents the configuration coefficient, which is dimensionless and is related to the distance between the power supply electrodes and the position of the measured electrode distance. ∆Ut represents the potential difference between electrodes M and N, measured in V, and I is the transmitte current, measured in A27.

The raw data obtained by TDIP system used in this study collects full waveform data, which has two complete charging and discharging processes in one cycle (Fig. 1). The adopted frequency band of power is 16 s and the power outage delay time is 140 ms.

Fig. 1
figure 1

TDIP complete transmitter waveform and received signal in a transmitter cycle. (a) is the full waveform of one transmitter cycle, (b) is two complete charging and discharging processes, I is the transmitter current, ∆Ut is the total potential, Up is the primary potential, Us is the secondary potential and Usp is the self-potential.

Full size image

The formula for calculating the apparent polarisation rate ηs is as follows:

$${\eta }_{s}=\frac{{\Delta U}_{s}}{{\Delta U}_{t}}\times 100\text{\%}$$
(2)

Equation (2) where ηs is a dimensionless value, ∆Ut is the total field potential in mV, ∆Us is the secondary field potential in mV27.

The instruments and equipment used in this data acquisition work are the large-scale 3D electromagnetic detection system, developed by the Institute of Geophysical and Geochemical Exploration of the Chinese Academy of Geological Sciences and produced by the Chongqing Geological Instrument Factory in China.. It is equipped with high power supply, notch filter and multi-parameter extraction. The high-power multifunctional electromagnetic transmission system consists of the EM-T160 transmitter, rectified power supply and 200 kW diesel generator set. The parameters of the high-power launch system are shown in Table 1. The system has a maximum transmitting voltage of up to 1600 V and a maximum power of 160 kW. This enables ultra-high-power power supply, ensuring strong signal strength and improved signal-to-noise ratio for better signal detection at depth. The receiving system utilizes multiple sets of receiving equipment to complete the electromagnetic measurement. Each multi-function receiver is three electrical channels, allowing for the measurement of electrical signals in multiple directions and at multiple measurement points. The parameters of the multifunctional receiving system are shown in Table 2.

Table 1 High-power multi-function transmitter technical parameters.
Full size table
Table 2 Multi-function receiver technical parameters.
Full size table

We carried out multi-spacing gradient TDIP data collection in the Huaniushan polymetallic mining area; field photos are shown in Fig. 2. The 3D TDIP data collection program consists of 20 lines. The distance between them is 200 m and covering an area of 10.64 km2. Among these, the 10 westernmost lines, ie. line numbers 220 to 400 station spacings were 50 m, while the remaining 10 lines from line number 420 to 600 station spacings were 100 m. In total, there are 440 measurement positions. To accommodate the large span of the survey area, a set of transmitter electrodes, labelled ① in Fig. 3, is used for lines 220 to 400, while another set of transmitter electrodes, labelled ② in Fig. 3, is used for lines 420 to 600. We completed a total of 624 TDIP measurement points, with 30 quality checks performed, resulting in a quality check rate of 4.8%. The mean square relative error of apparent resistivity was 3.14%, and that of apparent polarizability was 2.09%.

Fig. 2
figure 2

Photographs of field TDIP data collection at the Huaniushan polymetallic mining area.

Full size image
Fig. 3
figure 3

Arrangement of TDIP observation in Huaniushan polymetallic mining area.

Full size image

For the gradient IP measurements collected during this project, the power is supplied in the X-direction. For both electrode setups, transmitter electrode spacings are 2000 m, 2500 m, 3000 m, 3500 m and 4000 m. The receiving electrodes are M1N1 = M2N2 = 50 m. The transmitter period is 0.0625 Hz, 50% duty cycle. We chose the 0.0625Hz transmitter period to optimise the excitation effect and improve the efficiency of field measurements, while considering the size of the ore body in the study area and the receiving distance of the electric field, the MN spacing of 50 m is the most suitable. The power supply currents corresponding to each transmitter electrode spacing are as follows: Transmitter electrode ①: 12 A, 17 A, 14 A, 24 A and 10 A in the X direction. For transmitter electrode ②: 18 A, 23 A, 15 A, 16 A and 11 A in the X direction.

Regional geological overview and geophysical characteristics

Geological features

As shown on the geological map in Fig. 4, the upper rock group of the Jixianian Pingtoushan Formation in the test area is flanked on the north by the second lithological section of phyllite and carbonaceous slate. The part of the contact zone between the two is the location of the No. 1 ore belt. The No. 2 ore belt is located in the south-eastern corner of the test area, at the transition between the second lithological section of phyllite and the third lithological section of phyllite with marble. There is a nearly north–south trending fault developed in the area (F3). There are different degrees of fault breccias and mylonite in the faults, and the sections are generally straight, with sections being variable. There are a number of veins that occur along the fault, which control the spatial distribution of the ore body. Magmatic activities in the ore bodies appear to have been strong and frequent, mainly manifested in the large-scale magmatic intrusion in the early and middle Variscan and Indochina period. The intrusive event most closely related to mineralization is mainly the Indosinian granite. There are various types of vein rocks developed in the area, mainly acidic vein rocks, followed by medium basic vein rocks. They are often distributed in groups spatially, and are mainly controlled by the near east–west and north-west ruptures and fissures. This system appears to have been active at various times in geological history, again, with much occurring in the Variscan and Indosinian periods.

Fig. 4
figure 4

Location and Simplified Geological Map of the Huaniushan Mining Area. (a) Map of the People’s Republic of China. This map is based on the standard map (Approval No. GS(2016)1569) downloaded from the Standard Map Service website of the National Administration of Surveying(http://bzdt.ch.mnr.gov.cn/), Mapping and Geographic Information, with no modifications made to the base map. (b) Geological sketch map of Huaniushan mining area.

Full size image

Geophysical characteristics

The rock (ore) resistivity and polarizability data of Huaniushan mining area, presented in Tables 3 and 4, were obtained from the project of "Integration and Application of New Technology of Physical and Chemical Prospecting for Typical Ore Deposits in Liuyuan Area, Gansu, China (Project number: 12120113100500)" report. Physical data statistics for the test area were obtained from physical measurements based on drill core and mine specimens, supplemented by surface specimens.

Table 3 Statistical table of resistivity of major rocks and ores of the geological target bodies in the test area28.
Full size table
Table 4 Statistics on the polarizability rates of the main rock and ore in the geological target body of the test area28.
Full size table

In summary, the electrical properties of rocks in this experimental area have the following characteristics. Compared to strata and rock masses, pyrrhotite, pyrite, and lead–zinc ore all have the characteristics of high polarizability and low resistivity.These characteristics are useful when identifying potential ore zones in the resistivity/IP data sets. The electrical resistivity of phyllite itself is not significantly different from that of marble, but both have high polarizability and low resistivity electrical characteristics. Unfortunately, the phyllite (carbonate) unit has similar characteristics to the ore body in this area, i.e. the characteristics of low resistivity and high polarizability, which can easily become an interference factor in mineral exploration.

Data collection and analysis of results

Apparent resistivity and IP results from this study are shown as plan maps for each.

transmitter spacing in Figs. 5 (IP) and 6 (resistivity). Overall, the regional patterns observed in these figures, both in the apparent resistivity and the IP contour plots, are similar. The distribution of apparent polarizabilities at different transmitter electrodes (Fig. 5) shows that the background field of apparent polarizabilities in the survey area is generally low (< 6%) and smooth. Additionally, the values of apparent polarizabilities in the western part of the survey area are generally higher than those in the eastern part. However, with increased depth (i.e. increased transmitter bipole spacing), highly-polarised zones become more obvious. Examining the distribution characteristics of apparent resistivity at different transmitter electrode spacings (Fig. 6), the apparent resistivity in the western part of the survey area is lower than that in the eastern part, and the pattern of apparent resistivity at different transmitter spacing in the survey area varies significantly. The contact between the northwest phyllite (carbonate) and marble in the study area is readily apparent in both data sets.

Fig. 5
figure 5

Contours showing anomalies of apparent polarizability at different transmitter electrode distances of TDIP in the Huaniushan mining area. (a) AB = 2000 m; (b) AB = 2500 m; (c) AB = 3000 m; (d) AB = 3500 m; (e) AB = 4000 m.

Full size image
Fig. 6
figure 6

Contours of apparent resistivity anomalies at different transmitter electrode distances of TDIP in the Huaniushan mining area. (a) AB = 2000 m; (b) AB = 2500 m; (c) AB = 3000 m; (d) AB = 3500 m; (e) AB = 4000 m.

Full size image

The widely distributed phyllite (carbonate) unit in the northwest corner of the survey area (NO.1 ore belt) coincides with the location of the ferromanganese cap and lead–zinc mineralisation in the central part of the survey area (NO.2 ore belt). According to the metallogenic law of the area, it can be observed that the ore bodies in this area are mainly found in the contact zone of the marble and the phyllite. This observation is also consistent with the junction of the low and medium resistivity embodied in the central part of the survey area. A comprehensive analysis of the electrical anomalies and geological data is shown in Table 5.

Table 5 Analysis and interpretation of IP anomalies in the Huaniushan mining area.
Full size table

3D inversion interpretation of IP data

The incomplete Gauss–Newton method22 was used to perform 3D inversion of the obtained IP data in the mining area. According to the regularization theory of Tikhonov et al.29, the objective function of the 3D inversion problem was defined30 as follows:

$$f = \frac{1}{2}\phi_{d} + \frac{\lambda }{2}\phi_{m}$$
(3)

where ϕd and ϕm in Eq. (3) are the data fitting term and model constraint term respectively, with λ is the regularisation factor. We can adjust λ to change the weight of the data fitting term and model constraint term in the objective function. The value of λ can be selected in the range of 1 ~ 10, and the larger the value of λ, the greater the model constraint term is in the objective function. Using smaller values for λ makes the weight of the data fitting term significant. The optimal λ is generally determined by experimenting with a number of values31. However, when selecting the λ value, we relied on our self-developed inversion software (EM3DSPI, proprietary in-house developed), which was tailored based on geological information and other data from the study area. Through repeated testing and validation, we specifically determined the λ value for this research, ultimately adopting the default value of 5 in the software.

Figure 7 shows the fitted curves of apparent resistivity and forward apparent resistivity for different supply pole distances, and it can be seen that the fitting error for different supply pole distances is between 5 and 9%, which is a good fit and indicates that the 3D inversion results are reliable.

Fig. 7
figure 7

Fitting curves of measured and forward simulated apparent resistivity for different transmitter electrode distances. (a) AB = 2000 m; (b) AB = 2500 m; (c) AB = 3000 m; (d) AB = 3500 m; (e) AB = 4000 m.

Full size image

The inversion depth is 570 m and the inversion ran for five iterations. The polarisability model is shown in Fig. 8 and the resistivity model is shown in Fig. 9. The polarizability isosurface is set at 25.5% , and the resistivity isosurface is selected as 100 Ω.m.

Fig. 8
figure 8

3D geological-geophysical model constructed based on the polarizability rate information of the Huaniushan mining area.

Full size image
Fig. 9
figure 9

3D geological-geophysical model constructed based on resistivity information of Huaniushan mining area.

Full size image

Based on the distribution of the survey line layout of the mining area combined with the geological map (Fig. 4), We use geological data to construct a 3D geological-geophysical model based on the polarizability and resistivity of the mining area.

The 3D geological-geophysical models constructed based on the polarizability and resistivity of the mining area, combined with relevant geological and drilling data (Fig. 8 and Fig. 9). From the two figures, it can be seen that there is a pattern similarities between the resistivity parameter model and the polarizability parameter model, and a number of low resistivity zones also exhibit high polarizabilities. During the geological interpretation process, the model can be divided into eight low resistivity and high polarizability anomalous bodies numbered 1–8 in Figs. 8 and 9. We observe that the highly polarised anomalous area in Fig. 4 is almost at the contact between the two lithologies. So, we speculate that at least some of these 8 highly polarizability anomalous zones are potential ore zones.

In general, anomalies 1 to 8 are low resistivity and high polarizability anomaly areas. Among them, the anomalous zone 1, 5 and 6 are closed, while the anomalous zones 2, 3,4 7 and 8 are not closed. Area 7 is located in the within a zone rich in marble and hornfels. Anomaly 6 (which is relatively large and characterised as low resistivity) is located in a zone rich in granite, hornblende, and marble, the response is assumed presumed to be due to the presence of sulfides. The phyllite formation (i.e. anomaly 1) is divided into two relatively independent anomalies separated by the F3 structure of the nearly north–south trending fault. However, this faulting is not obvious in the resistivity data in this area. The strike position of the F3 fault from near north–south direction is visible in the TDIP 3D detection results (Figs. 8 and 9).

Integrated interpretation of resistivity/IP data with drillhole data

Figure 10 shows the geological results and ore body location along IP line 400 (see.

Fig. 10
figure 10

Comprehensive geological and borehole interpretation profile of the L400 line.

Full size image

Figure 4 for line and borehole locations. Figure 11 superimposes the borehole locations over the inverted resistivity and polarisation results for Line 400. This line runs directly over anomaly Zone 4. As can be seen from the geological and drilling composite profile of the study area (Fig. 10), the depths of the drill holes range from 198 to 600 m, and the samples collected from the five drill holes are mainly marble and phyllite, with most of the lead–zinc mineralisation observed at depths between 100 and 150 m.

Fig. 11
figure 11

Composite profile of TDIP exploration on the L400 line in Huaniushan polymetallic mining area.. (a) L400 line inversion resistivity section. (b) L400 line inversion polarizability section. Borehole locations (black lines) and interpreted ore body locations (red lines) described in Fig. 10 are highlighted.

Full size image

While it is obvious from a comparison of the ore zones outlined by the bore results shown in Fig. 10 and the resistivity and polarisation results shown in Fig. 11 is that the multi-spacing gradient results shown here do not do a perfect job of defining the shape of the ore zones in this area. This is not unexpected as the gradient survey configuration used here is unlikely to ever produce that high a level of resolution. Nevertheless, the practical question that most explorationists need to answer is: would I have drilled in the right location, based on the available results? Or at least: were the results interesting enough to do further work in the area? In both cases, it is likely that the answer would be “yes” based on the results shown here. Furthermore, selecting lambda during inversion is both difficult and critical. We anticipate that more scholars will investigate this mathematical constraint in subsequent research.

Conclusions

In this paper, we present the results of experimental work on a high-power Induced Polarization (IP) method using a multi-transmitter gradient surveys array in the Huaniushan mining area. The results are consistent with drilling data in the area and reveal the underground electrical structure in three dimensions. The multi-spacing gradient method used in this experiment has been shown to be effective. The conclusions are as follows:

  1. (1)

    The 3D geological-geophysical model constructed on the basis of the 3D electrical structure of the Huaniushan mining combined with the relevant geological and drilling data clearly shows the locations of a number of anomalous responses. One of the local anomalies coincides with the location of the known mining sites in the mining area. In addition, eight potential mineralisation sites have been discovered, which provides a basis for the evaluation of the potential mineral resources in the mining area.

  2. (2)

    The improved gradient IP methodology shown here,based on multi-transmitter gradient transmitter spacings, can quickly and efficiently obtain 3D resistivity and polarizability information of the survey area in the high contact resistivity environment typical of the, Gobi desert and other areas with high contact resistances. This method has the advantages of providing information at a number of depths as well as a large detection depth.

  3. (3)

    Based on the deep geological structure revealed by 3D exploration, this methodology may be useful in locating mineral-rich areas and provide a basis for the deployment of mineral resources exploration.