The project site under consideration is located in the Al-Qassim region of the Kingdom of Saudi Arabia. It covers an area of 120×80 m on which a large underground water tank is constructed, with the 120 m sides aligned in the east-west directions. The overall height of the tank is 6 m of which 4 m are below ground level. The tank consists of two main chambers, one on the north side and the other on the south side. Structural distress was noticed in the tank, which called for investigating the soil underneath.

A total of eight boreholes (BH1-BH8) were drilled on the north and south sides of the tank, as shown in Fig. (1), to determine the natural and engineering properties of the soil profile at the water tank site. Four of these boreholes are 10 m deep and the other four are 15 m deep. Boreholes BH2, BH4, BH5, and BH6 were drilled in February, 2016, while boreholes BH3 and BH7 were drilled in March, 2021, and boreholes BH1 and BH8 were drilled in April, 2021.

The necessary geotechnical field and lab tests of soil samples collected from these boreholes were conducted. SPT was carried out in accordance with BS 1377-9:1990 [23] and representative disturbed and undisturbed samples were collected at various depths. These samples were used to complete the geotechnical investigation of the soil profile at the site.

Electrical resistivity surveys were used to investigate the ground subsurface of the project site. The locations of survey lines were chosen at the north and south faces of the tank so that each line would coincide with the existing line of boreholes conducted at the project site, as shown in Fig. (1). The Dipole-Dipole array was used for the electric resistivity survey since it has relatively reliable horizontal coverage, considerably higher signal strength, and it is not very sensitive to electrical noise. The electrical imaging system was performed with a multi-electrode resistivity meter system. Such survey used a number of 48 electrodes placed in a straight line of 141 m in length at a constant spacing of 3 m. The 2-D electrical resistivity tomography (ERT) was performed with a multi-electrodes resistivity using Molisana Apparecchiature Elettroniche (MAE), advanced geophysics instruments, and MAE-X612 meter system. ZondRes2d (2021), a geophysical software, was utilized to analyze the collected data and interpret its results [24].

Fig. (1). The investigated project site with locations of the boreholes and resistivity lines.

Fig. (2). Dipole-Dipole configuration implemented in soil resistivity survey.

In the Dipole-Dipole resistivity survey, the basic idea was to use four electrodes arranged, as shown in Fig. (2). Using the value of the voltage and the current measured between electrodes M, N, A, and B respectively, the resistivity at point A () is calculated according to Equation 1. This configuration was repeated throughout the 48 electrodes and observed apparent resistivity data were recorded.

(1)

where V is voltage, I is current, n is an integer multiplier, and a is the distance between the dipole electrodes.

In this study, the MAE electrometer instrument was used to conduct the electrical resistivity survey, as shown in Fig. (3). MAE electrometer instrument was located at the midpoint between electrodes number 14 and 25. The MAE built-in software was run to start recording the resistivity data. Fig. (4) shows the north line of the electric resistivity survey at the site adjacent to the water tank. The recorded data were then inverted to resistivity images of the soil profile by ZondRes2d (2021) [24]. ZondRes2d provides an inverse model that estimates the actual subsurface structure based on the measured electrical resistivity values at the site.

The field values of the standard penetration test (SPT) for soil samples collected from the boreholes at the site are given in Table 1 with N referring to the number of blows. The results from soil boreholes are as follows:

Fig. (3). The MAE electrometer instrument used in the study.

Fig. (4). North side of the tank with the resistivity cable extending for 141 m.

• At a depth of 1.5 m, the soil is medium in consistency and relative density at the location of boreholes No. 1 to No. 6.
• At a depth of 3.0 m, the soil is soft in consistency and loose in relative density at the location of boreholes No. 2, No. 4 to No. 6, while the soil is very stiff and dense at the location of the rest boreholes (i.e. boreholes numbers 1, 3, 7, and 8).
• At a depth of 4.5 m, the soil is hard and very dense in all boreholes, except for boreholes No. 2 and No. 6, the soil is very stiff and with medium density.
• From a depth of more than 4.5 m to the end of the depth of the boreholes, the soil is hard and very dense.

The laboratory test results of samples extracted from the site are given in Table 2, while the type of soil layer and their depths for the eight boreholes are listed in Table 3. The results in these two tables show that there is homogeneity and continuity in the sub-surface soil at the site, and in general, the soil at the studied tank site consists of three main layers:

• Backfill material: A layer of backfill consisting of poorly graded sand and silty clay mixed with gravel and pebbles, brown in color, with low to medium density. This layer appeared in almost all boreholes and extended up to a depth of 4.0 to 4.5 m from ground level.
• Gray stiff to hard clay with expansive nature: It appeared in 5 boreholes with a thickness of 2.0 m to 4.50 m. In this layer, clay percentage is 60% and plasticity index is 37% with a high potential for swelling.
• Light color interaction between highly weathered limestone and marl: It appeared in all boreholes, extending to the end of the 15 m depth boreholes and existing at the end of 10 m depth boreholes as well. The marl deposits are clay deposits that have a high percentage of salt and have the potential for high swelling too [25].

Within these three main layers comprising the soil profile at the site, a couple of pockets of lean clay mixed with sand appeared in BH3 and BH7, which did not appear throughout the soil profile.

The electrical resistivity surveys were conducted for the north side of the tank in two stages almost two years apart, February, 2018, and October, 2020. For the south side of the tank, the electrical resistivity survey was conducted in February, 2021. The measurements of apparent resistivity were collected using MAE and advanced geophysics instruments, and the measured electric resistivity data were recorded by the MAE built-in software. These recorded data were then inverted to resistivity images through the ZondRes2d (2021) software, providing an inverse model that approximates the soil subsurface profile. The non-uniqueness and smoothing condition of the inverted results may impose some limitations. The inversion results, however, can give an excellent resistivity image of the soil. All three electrical resistivity surveys revealed that there are three distinct soil layers, as shown in Figs. (5-7). The limits of the underground water tank location are also highlighted in red dashed lines in these figures. For interpretation of the presented electrical resistivity images, Table 4 shows the ranges of electrical resistivity values for some soil types, according to Loke (2000) [26]. The interpreted electrical resistivity values were compared with the extracted soil samples from boreholes to corroborate and validate the results.

Fig. (5). Electrical resistivity values at the north side of the tank in 2018.

Fig. (6). Electrical resistivity values at the north side of the tank in 2020.

Fig. (7). Electrical resistivity values at the south side of the tank in 2021.

The electrical resistivity values for the cross-section of the soil profile of the north side performed in 2018 are shown in Fig. (5). In the middle layer, the measured resistivity values are the lowest compared to the top and bottom layers. The center portion of the soil profile of the middle layer showed resistivity values of less than 3 Ωm for a distance of around 50 m, which falls into the range of clay soil type. These measurements of electrical resistivity of low value indicate clay soil type, which matches well with the extracted samples from soil boreholes. The bottom layer shows resistivity values in the range between 1.76 Ωm to 59.05 Ωm, which falls within the range of the clay, marls, and limestone soil types. Limestone and marl soils were also characterized by boreholes tests at these layer depths, as shown in Table 3.

Electrical resistivity results that were performed in 2020 for the same cross-section at the north face of the tank are shown in Fig. (6). For the 2020 survey of the north side, the resistivity values in the top layer are close to those measured in 2018 with no major changes. For the middle layer, the low resistivity values in the middle portion of the layer extended toward the east side of the tank (i.e. at a horizontal distance from 0 to around 85m). The resistivity values in this region, which are less than 3 Ωm, fall in the range of the clay soil resistivity values. The reason for this change in soil resistivity values from the prior scan in 2018 might be attributed to a water leakage that occurred in this area of the tank. For the bottom layer, the resistivity values showed a considerable change with time in terms of an increase in resistivity values except for the range between 100 to 120 m. Nevertheless, these values fall within the range of the resistivity values of clay, marl, and limestone soils, and these types of soil were also confirmed by the soil samples from boreholes tests.

For the 2021 survey of the south side of the tank, electrical resistivity values for the cross-section of the soil profile are shown in Fig. (7). The resistivity values in the top layer are non-consistent to a certain extent throughout the overall width of 141 m as compared to the north face of the tank. The middle layer showed similar resistivity values to the surveys of the north side, ranging from 1.53 Ωm to 3.82 Ωm, with the lowest values at a distance between 50 to 75 m. For the bottom layer, resistivity values range from 5.17 Ωm to 72.43 Ωm, which fall within the range of the electrical resistivity values of clay, marl, and limestone soils.

A summary of the ranges of all measured resistivity values beneath the water tank for all three soil surveys is given in Table 5, and average resistivity values for the three soil layers are compared in Fig. (8). The resistivity values summarized in Table 5 show reasonable agreement with those available in the literature [26], as shown in Table 4. Clay and marl deposits typically have resistivity values of 0.1 Ωm – 50 Ωm, and 2 Ωm – 50 Ωm, respectively. This is apparent from soil resistivity Figs. (5-7), showing the middle layer to yield resistivity values in this range. It is also affirmative of the results of the eight drilled boreholes at the site. Moreover, the middle layer shows resistivity values starting from 1 Ωm to 20.55 Ωm, which fall within the range of the electrical resistivity of the clay soil as characterized by the boreholes investigations. Limestone resistivity values range from 20 Ωm up to almost 400 Ωm, as shown in Table 4. The geotechnical investigation through the boreholes of the current location shows that the limestone is highly weathered, thus having low resistivity values (4.23 to 250.52 Ωm). Similarly, marl is expected to give resistivity values in the range between 2 to 50 Ωm, which compares well with the extracted samples from soil boreholes. Overall, the ranges of soil resistivity values, given in Table 4, are in acceptable agreement with the results of the eight soil boreholes shown in Table 3. Hence, these findings reflect the reliability of using electrical resistivity tests in conducting a preliminary investigation of the soil profile at a project site.

Fig. (8). Average resistivity values for all three soil layers based on the conducted three surveys.