Audio‐Magnetotelluric Survey for Groundwater Investigation in the Al‐Jaww Plain in Eastern Abu Dhabi, Al‐Ain, United Arab Emirates

The United Arab Emirates (UAE), located in an arid climate zone with low rainfall, relies on shallow aquifers for freshwater. Understanding the depth and extent of such aquifers is crucial for meeting water supply needs. The UAE's hydrogeology is influenced by neighboring mountains in Oman. The Al‐Jaww Plain in southeast of Al‐Ain city is an essential groundwater source, characterized by a large, flat area of gravel and sand deposits from the Oman Mountains. This study aims to map groundwater aquifers in the Al‐Jaww Plain by integrating the audio‐magnetotelluric (AMT) method, seismic reflection profiling, and borehole data. AMT data were collected along an 11‐km ENE–WSW profile and a 2D resistivity model was generated. The resulting model delineates three distinct geo‐electrical zones from the surface to a depth of 5 km. First, a shallow layer with low resistivity (0–15 Ωm) represents the Quaternary and Pliocene aquifers, in addition to the Upper Cretaceous Simsima and Tertiary groundwater aquifer zone, extending to a depth of 1.5 km. Second, a moderately resistive layer (15–250 Ωm) is recorded beneath the first layer, corresponding to the Upper Cretaceous Aruma foreland basin sequence. Finally, a high‐resistivity region (>250 Ωm) at depths exceeding 3 km is attributed to the allochthonous Hawasina thrust sheet, which is associated with Late Cretaceous obduction of the Semail ophiolite. These findings have practical implications for managing groundwater resources in Al‐Ain.


10.1029/2023EA003181
2 of 15 Geophysical techniques play a major role in locating and assessing potential of groundwater aquifers.However, relatively few geophysical studies have been previously conducted on the Al-Jaww plain subsurface.Woodward (1994), Ali et al. (2008), Ali and Watts (2009), Cooper et al. (2014), and Bruno and Vesnaver (2021) used seismic geophysical techniques to investigate the geological structures and stratigraphy beneath the Al-Jaww Plain.Ali et al. (2008) interpreted seismic horizons in this area and classified them into four different bedrock units: (deepest) Mesozoic shelf carbonates and allochthonous marine sequences, unconformably overlain by a clastic and carbonate foreland basin sequence, followed upwards by a Tertiary dominantly carbonate sequence.Gravity and magnetic surveys were also conducted over the Al-Jaww Plain by Ali et al. (2008) to constrain the subsurface geological structure of the study area.The Bouguer anomalies of the Al-Jaww Plain are represented by lower values than those of the surrounding regions, with values ranging from −76 to −70 mGal explained by thickening of the crust in this area (Ali et al., 2008).
El Mahmoudi and Gabr (2006) studied the hydrogeology and subsurface structure of wadis on the plain to better understand the Quaternary alluvial aquifer configuration and groundwater flow patterns.They applied the Electrical Resistivity Tomography (ERT) method to obtain a two-dimensional resistivity profile (600 m long in N-S direction) across Wadi Muraykhat and Wadi Sa'a (refer ERT profile in Figure 1b), which was supported by available borehole data, to characterize the Quaternary aquifer investigated in their study.An unconformity at the base of the Quaternary alluvium was identified in their final resistivity profile.This unconformity was interpreted to outline paleochannels in the bedrock formed by Neogene erosional processes.The resistivity profile of El Mahmoudi and Gabr (2006) showed a resistive layer (>100 Ωm) extending from the surface to a depth of 40 m, which they interpreted as dry alluvium.This was underlain by a groundwater-bearing conductive layer of clay and mudstone located between 40 and 120 m depths (El Mahmoudi & Gabr, 2006).(Ali et al., 2008) showing the locations of audio-magnetotelluric stations, seismic, and borehole data.The black dashed line is the Electrical Resistivity Tomography profile (El Mahmoudi & Gabr, 2006).The insert shows a regional map of the United Arab Emirates highlighting the study area (red star).The color bar shows elevation.Panel (a) is generated by GMT (Wessel et al., 2009).Note that MD-1 is the deep hydrocarbon exploration well used to depth convert the seismic profile (Figure 7).

10.1029/2023EA003181
3 of 15 In this study, we use a passive geophysical technique, the audio-magnetotelluric (AMT) method, to map the subsurface electrical structure of the Al-Jaww Plain.In addition, in order to extend the results of El Mahmoudi and Gabr (2006), we have decided to have a longer geophysical profile (11 km) and ENE-WSW oriented traverse to make sure that the totality of Al Jaww plain is investigated.Magnetotelluric technique was chosen because it gives both shallow and deep electrical resistivity information, as opposed to ERT, which is restricted to shallow investigations.
AMT measurements are highly sensitive to fluid distribution and, thus, are a powerful tool for imaging groundwater-bearing strata in the subsurface (Erdoğan & Candansayar, 2017).There have been numerous studies from various countries regarding the application of the AMT method for groundwater investigations (Bernard et al., 1990;Falgàs et al., 2009Falgàs et al., , 2011;;Meju et al., 1999;Nenna et al., 2013).Low resistivity areas may indicate the presence of saturated formations, suggesting potential aquifer zones.Additionally, a seismic reflection profile across the Al-Jaww Plain and borehole data from a groundwater program are used to gain insights into the area's shallow and deep subsurface structures, stratigraphy, and groundwater conditions.
This study aims to investigate the subsurface electrical structure of the Al-Jaww Plain, southeast of Al-Ain, Abu Dhabi, UAE, using the AMT method to delineate potential shallow and deep saturated zones.Electric studies will be supplemented by a seismic reflection profile across the Al-Jaww Plain, along with borehole data, in order to gain insights into the region's shallow and deep subsurface structures, stratigraphy, and groundwater conditions.Understanding the depth and extent of these aquifers is crucial for addressing urban water supply needs in this arid environment.

Geological and Hydrogeological Setting
The northern Oman Mountains comprise a series of thrust sheet complexes composed of Late Cretaceous Tethyan oceanic crust and mantle that were thrusted against the Arabian carbonate platform.These allochthons include the Semail ophiolite, a composite body consisting of Cenomanian upper mantle and oceanic crust with a thickness ranging from 8 to 15 km.Other allochthons include the Haybi nappe, composed of Upper Permian to Cenomanian oceanic seamounts and alkaline volcanics, the Hawasina allochthon, consisting of distal to proximal oceanic sedimentary rocks, and the Sumeini complex, which comprises proximal carbonate slope sediments (Glennie et al., 1973;Searle & Ali, 2009).
During the Cenomanian-Turonian period, loading of the Semail ophiolite onto the Arabian passive margin resulted in down-flexing of the margin and formation of the Aruma foreland basin, while a concurrent flexural forebulge formed to the west (Ali & Watts, 2009;Boote et al., 1990;Warburton et al., 1990).The formation of this forebulge induced regional uplift, leading to erosion and removal of a portion of the shelf carbonate deposits and the subsequent formation of the Wasia-Aruma break, which demarcates the boundary between the foreland sequences and underlying shelf margin deposits (Glennie et al., 1973).Subsequently, during the late Coniacian-Campanian, the foreland basin was filled with deep marine mudstones of the Juwaiza and Fiqa formations (Boote et al., 1990;Patton & O'Connor, 1988;Robertson, 1987;Warburton et al., 1990).These formations were later overlain by conglomerates and shallow marine limestone deposits of the upper Maastrichtian Simsima and Qahlah formations (Nolan et al., 1990).The northeastern Arabian continental margin then remained stable during the Paleocene-Oligocene, resulting in the deposition of a transgressive sequence comprising the Umm Er Radhuma, Rus, Dammam, and Asmari formations (Glennie et al., 1973;Searle, 2019;Searle & Ali, 2009).
The Central Iran-Arabia collision that occurred in the late Oligocene-early Miocene ultimately resulted in the closure of the Neo-Tethys Ocean.This collision triggered extensive thrusting in the study area, leading to the emplacement of allochthonous Permian-Cenomanian carbonates over the autochthonous foreland sequences (Ali et al., 2018;Searle, 1988Searle, , 2007)).Additionally, this tectonic event caused prominent west-verging fold formation along the western flank of the northern Oman Mountains, reactivating Mesozoic faults within the fold-and-thrust belt and culminating in the deposition of the Pabdeh foreland basin sequences (Abdelmaksoud et al., 2022(Abdelmaksoud et al., , 2023;;Boote et al., 1990;Dunne et al., 1990;Searle, 1988).
The major fold structures formed in the study area during the late Oligocene-Miocene period include Jabal Hafit, Jabal Malaqat, and Jabal Mundassa (Figure 1).Jabal Hafit, situated southeast of Al-Ain city, is an elongated anticline whose surface expose measures 29 km in length and 5 km in width.This structure exhibits a surface elevation expression of 1,160 m, plunging southeast into the Omani borders and northwest into the UAE (Abou El-Enin, 1993;Ali et al., 2009).Adjacent to Jabal Hafit, the Al-Jaww Plain is located in the western foothills of the Oman Mountains (Figure 1).The exposed strata in the west of the plain are Eocene (Damman and Rus formations), Oligocene (Asmari Formation), and Miocene (Fars Formation) in age (Figures 1 and 2).On the eastern side of Al-Jaww Plain, Upper Cretaceous to Lower Eocene rocks are exposed in the outcrops of Jabal Malaqat and Jabal Mundassa (Figure 1).These units comprise the Qahlah, Simsima, Umm Er Radhuma, Rus, and Dammam formations (Abd-Allah et al., 2013).The Qahlah and Simsima Formations exhibit an unconformable contact with the underlying Semail ophiolite.The Semail ophiolite, which consists of Late Cretaceous serpentinite mantle rocks, is prominently exposed on the western side of Jabal Malaqat and Jabal Mundassa outcrops.The Maastrichtian Simsima Formation overlies the Semail ophiolite and underlies the Umm Er Radhuma (Muthaymimah) Formation in these outcrops.
Despite being characterized by an arid climate, Al-Jaww Plain experiences flash floods due to its proximity to the Oman Mountains, which supply groundwater to the aquifers of Al-Ain.Rainfall during the Quaternary period transformed the area into a major oasis in the Arabian Peninsula, supporting life and agriculture through the availability of shallow fresh groundwater.
A key groundwater source in the Al-Ain region is the Quaternary aquifer, part of the Western Gravel Aquifer, with thicknesses ranging from 60 to 400 m (Elmahdy & Mohamed, 2015).The Quaternary alluvium, comprising sand and gravel layers with interbedded mud, serves as the water-bearing lithostratigraphic unit for this aquifer.These clastic units are predominantly clay-rich and calcareous, and the alluvium constituting these deposits was deposited in flood plains draining from the western Oman Mountains (El Mahmoudi & Gabr, 2006).In terms of other key aquifer sequences, Rizk et al. (1997) identified the primary productive aquifers in the UAE as fractured peridotite in the Oman Mountains, gravel aquifers in the Oman Mountains, and sand dunes in the western part of the UAE.
The Al-Jaww Plain aquifer is replenished by precipitation from the Oman Mountains, primarily through the flow of water in wadis such as Wadi Muraykhat and Wadi Sa'a (Gabr et al., 2013).Recharge of the Quaternary aquifer underlying Al-Jaww Plain also occurs due to infiltration of rainfall and movement of water from the northwestern part of the Oman Mountains through subsurface pores and fractures (Gabr et al., 2013).Additionally, other potential sources of aquifer recharge include irrigation, the water transfer from deeper saturated layers through fault zones, and infiltration of water leaks from major water transmission lines and urban areas.Groundwater measurements from Al-Jaww Plain exhibit Total Dissolved Solids (TDS) content values ranging from 500 to 7,000 mg/L (Environmental Agency Abu Dhabi (EAAD), 2023; UN-ESCWA & BGR, 2013), with permeability values ranging from 2.9 × 10 −4 to 4.6 × 10 −3 m 2 /s (Patton & O'Connor, 1988).The elevation of the water table in Al-Jaww Plain ranges from 250 to 300 m above mean sea level (EAAD, 2023).

The AMT Method
The AMT method is an electromagnetic passive-source inductive method used to infer the Earth's subsurface electrical conductivity (Simpson & Bahr, 2009).The subsurface electrical conductivity is calculated from temporary surface measurements of the magnetic (H) and electric (E) fields.The electromagnetic signals measured in the AMT and magnetotelluric (MT) are naturally occurring signals generated by worldwide lighting activities and by the interaction of solar activities with the magnetosphere.Low frequencies signals (<1 Hz) are generated by the interactions between the Earth's magnetosphere and solar wind, sunspot activity, and auroras.The high-frequency sources (>1 Hz) in the audio range used in AMT surveys are generated by lightning activities worldwide (Simpson & Bahr, 2009).The interactions of these electromagnetic signals create telluric currents that flow horizontally in the Earth's crust (Simpson & Bahr, 2009); these currents then produce the magnetic signals measured in MT and AMT surveys.
The electric and magnetic variations are measured in the field using magnetic and electric sensors in different orientations.The data are recorded as a time series, later converted to the frequency domain during the data processing stage.The transfer function (i.e., the complex impedance matrix) is then estimated from the measured field in the frequency domain, as shown in Equation 2. The apparent resistivity and phase can then be calculated from the impedance components for the selected frequencies.Once the MT data are processed and analyzed, forward modeling and inversion algorithms are used to produce resistivity distribution models (1D, 2D, and 3D).Equation 1shows the relationship between the electric (E) and magnetic (H) field vectors, where the ratio of the H and E fields represents the impedance tensor (Z).Equation 2shows the relationship between the components of the impedance tensor (Zxx, Zxy, Zyx, and Zyy) and the electric and magnetic fields measured in the x (north-south) and y (east-west) directions.

𝐄𝐄 = Z𝐇𝐇
(1) The apparent resistivity (ρ a ) and phase (φ) are then calculated from the impedance components as follows: where ω is the angular frequency (ω = 2πf, where f is the frequency in Hz), μ is the magnetic permeability of free space (1.2566 × 10 −6 H/m), Z ij represents one of the components of Z, and ImZ ij and ReZ ij represent the imaginary and real parts of the impedance components, respectively.

AMT Data Analysis
Analysis of MT data is a common procedure used to infer the main properties of subsurface geoelectric structures, such as the strike direction or the presence of superficial distorting bodies and enables the most appropriate modeling approach to be determined (Groom & Bailey, 1989).Prior to generating a subsurface model from MT data during the inversion stage, it is important to understand the degree of distortion and dimensionality of the data sets (Groom & Bailey, 1989;Marti, 2006;Smith, 1995).For the AMT data measured in this study, we used two-dimensionality tools to check the dimensionality of the surface structures, the phase-sensitive skew (Bahr, 1988(Bahr, , 1991) ) parameter and phase tensor analysis (Caldwell et al., 2004).

Phase Sensitivity Skew
Phase Sensitivity Skew (η) is a dimensionality tool that represents a measure of the skew of the phases of the impedance tensor (Bahr, 1988(Bahr, , 1991)).This parameter is thus unaffected by the distortion effect, unlike amplitude based dimensionality tools, and is given by: where: where, Im denoted Imaginary part and   * and   * are the complex conjugates of Z yx and Z xy , respectively.
Phase-sensitive skews (η) less than 0.1 indicate 1D, 2D, or distorted 2D (3D/2D) cases, whereas values of η > 0.3 indicate 3D structures.Values of η between 0.1 and 0.3 indicate regional 2D and 3D local distortion.The AMT data from the Al-Jaww Plain acquired in this study show η > 0.3 in the short periods around the AMT dead-bands and at around 1 s.However, most of the measured phase-sensitive skew data generally indicate a regional 2D structure (Figure 3).The red colored line in Figure 3 indicates the 0.3 dimensionality threshold value.Most of the η value estimates calculated at different periods from all the stations fall below the threshold, indicating a 1D or 2D character of the subsurface structure below our study area.However, the few higher frequencies indicate a 3D effect, especially within the AMT deadband (Figure 3).These elevated values within the deadband may be attributed to noise present in the AMT deadband.Robust processing of the time series helps reduce this noise.Additionally, before inversion, a consistency assessment was conducted on the MT data using the D+ function (Parker, 1980) to discard inconsistent data points.

Phase Tensor Maps
Phase tensor analysis is a commonly used dimensionality tool in MT data analysis (Booker, 2014;Caldwell et al., 2004).Phase tensor analysis makes no prior assumptions regarding the regional structure and is unaffected by galvanic distortions as it is derived from the phase component (Caldwell et al., 2004).The phase tensor Φ is defined as the ratio of the real (X) and imaginary parts (Y) of the complex impedance tensor, Z: The phase tensor can be described graphically using ellipses, expressed in terms of the minimum (Φ min ) and maximum (Φ max ) principal axes and the skew angle (β) and it is expressed by the following formula (Booker, 2014;Caldwell et al., 2004): where R(α + β) is the rotation matrix and R T is the transposed or inverse rotation matrix.The strike of the ellipse's major axis is given by α − β.Booker (Booker, 2014) suggested that the phase tensor skew angle (β) is important in deciding whether 2D or 3D interpretation is required.Large skew angles (more than 5° considering the data error in the measurement) are considered a robust test for the presence of 3D regional structure, with skew angles less than 5° used to justify a 2D interpretation.In addition, in the presence of a 3D structure, the phase tensor ellipses are asymmetric and show large skew angle values.
The phase tensor maps produced for some selected periods for our AMT data reveal the dimensionality of the subsurface (Figure 4).In Figure 4, the phase tensor is represented by ellipses, where the color filling the ellipses represents the skew angle.The phase tensor shape and the skew angle value reveal the dimensionality of the subsurface, where small skew angles and circular phase tensors indicate 1D and 2D structures, while large skew angles and elliptical phase tensors indicate 3D structures.In general, the phase tensor analysis from our study area indicates regional 1D and 2D structures characterized by small skew angles between −5° and 5° (Figure 4).

2D MT Inversion
The main goal of any geophysical inversion is to produce a final subsurface model that represents the ground truth.Geophysical inverse problems generally involve determining subsurface structures from observations obtained on the Earth's surface (Pedersen & Hermance, 1986).The general geophysical inversion procedure involves a simple initial model from which the theoretical geophysical response is computed and compared with the observed data.This procedure is repeated for various models through an iterative process until the minimum difference between the computed and observed responses is achieved.Depending on the algorithm used for the inversion, the error vector will be minimized with some type of damping factor or regularization parameter that stabilizes the inversion process.The data misfit is characterized by an objective function, which is minimized using a suitable optimization algorithm.The problem of data prediction is called forward modeling, in which data are calculated based on known values of model parameters.The estimated model in the final iteration is accepted as the solution to the inversion and represents the geological structure.
We perform a 2D inversion of the AMT data to resolve subsurface resistivity structures in the study area.Before generating a 2D model, the data were rotated to the estimated electrical strike direction to identify the regional 2D MT response.The strike angle (N5°E) was calculated with the MTpy code (Krieger & Peacock, 2014) from the phase tensor.In the 2D case, the resistivity varies with depth and along the profile, whereas the resistivity is assumed to be constant perpendicular to the profile.To generate the 2D AMT model from our measured data, we performed a joint inversion of the transverse electric (TE) and transverse magnetic (TM) modes using WinG-Link processing software developed by Geosystem Inc. (WinGLink, 2018).This program uses the Rodi and Mackie (2001) smooth model inversion routine, which finds regularized solutions to the 2D inversion problem using the non-linear conjugate gradient method.A finite-difference formulation with the option to incorporate topography into the model mesh is used for forward model calculation (WinGLink, 2018).The parameters used in the inversion process were as follows: (a) the initial model was set at 20 Ω.m (a homogenous half space), (b) both TE and TM modes were inverted, (c) the frequency range was set to 10,400-0.35Hz, (d) the error floor was set at 10% Rho and 5% Phase.To deal with any static shift effect in the data, the 2D Inversion routine of the WinG-Link program (WinGLink, 2018) incorporates a static shift (Bahr, 1988(Bahr, , 1991;;Caldwell et al., 2004;Goldstein & Strangway, 1975;Groom & Bailey, 1989;Jiracek, 1990;Li et al., 2011;Marti, 2006;Simpson & Bahr, 2009;Smith, 1995) correction in the AMT data inversion process.At the end of the inversion process, the RMS error reached a value of 3.358 at iteration number 30.In general, a good fit between the observed and modeled data was achieved by the final iteration at all AMT stations.

Groundwater Borehole Data
The UAE conducted a comprehensive Ground Water Program (GWP) from 1988 to 2013, in collaboration with the National Drilling Company, Abu Dhabi, and the US Geological Survey (USGS).The program's initial primary objective was to assess the UAE's groundwater resources in the UAE; however, over time, the program expanded to groundwater monitoring and research.As part of this program, a total of 1,240 boreholes were drilled across the UAE to investigate groundwater conditions (Kress, 2017).The drilling of these boreholes was guided by the interpretation of subsurface geophysical data obtained using the electrical resistivity method.In the context of Al-Jaww Plain, a geological cross-section was constructed using data from 11 shallow groundwater boreholes acquired as part of the GWP.This cross-section extends from the southwestern region near Jabal Hafit to the northeastern area near Jabal Malaqat (Figure 1).

Seismic Reflection Profile
The seismic reflection profile IQS1 was acquired in 1982 as part of a regional seismic survey for hydrocarbon exploration in the Al-Ain area.This profile is approximately 21 km long and crosses the Al-Jaww Plain in a WSW-ENE orientation (see Figure 1).The seismic data were recorded using a vibroseis seismic source, a maximum two-way-travel time record length of 5 s, and a sample interval of 4 ms.
In 2007, the IQS1 seismic profile was reprocessed by WesternGeco.Various processing steps were applied to the raw SEG-Y seismic data to enhance their quality and interpretability.These steps included geometry assignment to ensure accurate positioning of the seismic traces, field static corrections to correct for variations in arrival times caused by near-surface conditions, trace editing to remove any data artifacts or inconsistencies, and FK filtering to attenuate coherent noise.Further processing steps included amplitude balancing to normalize the seismic amplitudes, deconvolution filtering with an operator length of 160 ms to enhance the seismic resolution and initial velocity analysis to estimate the subsurface velocity field.Pre-stack time migration was then used to accurately position the seismic reflectors vertically, post-stack migration velocity analysis was applied to refine the velocity model, and common mid-point stacking was used to enhance the signal-to-noise ratio and improve seismic imaging.Overall, these processing steps were used to improve the overall quality and interpretability of the seismic data, thus allowing for better subsurface imaging and geological analysis of the Al-Jaww Plain.The seismic profile was depth converted based on a velocity profile obtained from check shot data from a nearby deep exploration well (MD-1, refer to Figure 1).Moreover, we used the lithology and properties of the layers penetrated by the groundwater borehole data to interpret the seismic profile.However, these wells did not penetrate the Hawsina thrust.Therefore, there is uncertainty on the thickness of this unit.

AMT Data Acquisition and Processing
In this research, data were acquired along an 11-km-long ENE-WSW profile in Al-Jaww Plain (east of Al-Ain city) at 10 AMT stations (Figure 1).We used an MTU5A receiver to measure electric and magnetic AMT signals.This instrument is a V5 System 2000 for MT measurements developed by Phoenix Geophysics Ltd.In the field, the measurement station locations were selected carefully before data acquisition to avoid the presence of high-tension power cables, fences, underground pipelines, and any noise that can affect the AMT signals.We measured the x (N-S) and y (E-W) horizontal components of the magnetic (H x and H y ) and electric (E x and E y ) fields at each recording station to obtain high-frequency AMT signals.The magnetic coils and electrodes were buried in the ground to reduce noise and ensure that the electrodes' orientations remained constant during surveying.Bentonite was used to minimize the contact resistance between the electrodes and the ground to achieve good electric signal quality during the measurement.The ground contact resistance was high due to the sand and resistive rocks in the study area; thus, bentonite helps to alleviate this problem.
The measured AMT time series were processed using the robust SSMT2000 processing code to convert the raw data to the frequency domain.MT-Editor software was then used to estimate the transfer function and, finally, the data were stored in EDI file format (a standard MT format) for each of the measured stations.Figure 5 shows sounding curves as plots of log apparent resistivity and phase versus the log of the period for all 10 stations after the processing.The AMT apparent resistivity curves of xy (red curves) and yx (blue curves) from all 10 stations in the study area show generally similar trends, indicating relatively higher resistivity values at short periods, with a decrease recorded up to values of around 0.001 s (1,000 Hz).The resistivity then increases again briefly and remains constant or undergoes a slow decrease before dropping sharply at some stations.High frequencies (short periods) reflect responses from shallow structures, whereas low frequencies (longer periods) reflect responses from deeper structures.However, the interpretation of the subsurface structure from these curves is potentially misleading because the penetration depth will vary for each station at a particular frequency (period) and the resistivity values are apparent (not true) resistivity values.Therefore, care must be taken, as these curves only provide a general description of the subsurface resistivity distribution.

Results and Discussion
The groundwater borehole cross-section illustrates the near-surface geology, hydro-stratigraphy, and lithology of Al-Jaww Plain (Figure 6).This cross-section reveals Quaternary alluvium deposits comprising sand and   gravel, with thicknesses of approximately 10-75 m.Below these alluvium deposits, the boreholes penetrate marl, unconsolidated sand and gravel, and clay, which are collectively interpreted as post-Fars Formation deposits of Pliocene age.The thickness of this unit increases southwestward.The Miocene Upper Fars Formation primarily consists of mudstone, marl, and limestone, and its thickness increases toward the northwest.Borehole GWP-007 records the maximum thickness of the Upper Fars Formation, exceeding 300 m (Figure 6).The Lower Fars Formation comprises marine sequences with interbedded shales, mudstones, and evaporitic rocks.Its thickness is over 200 m and increases toward the southwest.The Oligocene Asmari Formation is predominantly composed of limestone.Its shallowest occurrence is recorded in the eastern limb of the Jabal Hafit anticline (GWP-204), while its deepest occurrence is identified in borehole GWP-10.Thrusting of the Eocene Dammam and Rus formations, which comprise pre-Asmari-age limestone, occurs in the eastern part of Al-Jaww Plain near Jabal Malaqat.This results in the absence of Oligocene, Miocene, and Pliocene formations in boreholes GWP-246A and GWP-247A (Figure 6).The thrust fault associated with this deformation is also clearly visible in the seismic profile (Figure 7).Additionally, the borehole data indicate the presence of anticlines and synclines in Al-Jaww Plain.A synclinal fold is recorded in boreholes GWP-010 and GWP-007, whereas two anticlinal folds are identified in the southwest of the area (eastern limb of Jabal Hafit) and in boreholes GPW-011 and GPW-006.These folds are also evident in the seismic profile (Figure 7) and surface geological map (Figure 1).
Based on the borehole cross-section results, there are three interconnected aquifer types identified in Al-Jaww Plain.The first aquifer comprises Quaternary alluvium deposits, which behave as a near-surface unconfined aquifer.Recharged directly from the adjacent mountains during the rainy season, this aquifer represents the primary source of fresh groundwater in the southeastern area of Al-Ain.Due to its content of permeable sand and gravel deposits, this aquifer exhibits high hydraulic conductivity.This is the only aquifer observed in the eastern portion of Al-Jaww Plain due to thrust faulting.The second aquifer is associated with the area's Pliocene deposits and demonstrates connectivity with the upper Quaternary aquifer.The third aquifer is hosted in the Upper Fars Formation and is similarly interconnected with the Pliocene aquifer.Predominantly composed of mudstones, this aquifer exhibits moderate to low permeability and high salinity.In contrast, the Lower Fars Formation has very low permeability, making it unsuitable as an aquifer.This formation instead functions as the basal confining system for the hydrologic formations of Al-Jaww Plain.
Figure 7 displays uninterpreted and interpreted versions of seismic profile IQS11, illustrating the main structural and stratigraphic features of Al-Jaww Plain.As deep borehole data are unavailable for this region, the GWP borehole cross-section (Figure 6), which runs parallel to the seismic profile, was used to interpret the shallow sections of the profile.Additionally, interpretations published in previous studies (Ali et al., 2008(Ali et al., , 2009;;Bruno & Vesnaver, 2021;Cooper et al., 2014;Searle & Ali, 2009;Woodward, 1994) were used to aid the interpretation process.
The lowermost part of the seismic profile is interpreted as autochthonous shelf carbonates belonging to the Wasia and Thamama groups, while the middle section represents the Aruma foreland basin sequence and the Hawasina thrust sheet.The Aruma foreland basin sequence consists of the Lower Fiqa, Juwaiza, and Upper Fiqa formations.
The seismic profile reveals thrust faults, potentially associated with the obduction of allochthonous complexes, occurring at the contact between the shelf carbonates and the Lower Fiqa Formation (Figure 7).The upper part of the seismic profile corresponds to the Upper Cretaceous Simsima Formation and Tertiary formations including the Umm Er Radhuma, Rus, Dammam, Asmari, and Fars formations.The stratigraphy and structures observed within the Upper Tertiary formations are consistent with those identified in the groundwater borehole cross-section shown in Figure 6.
The seismic profile reveals several fold and thrust fault structures in Al-Jaww Plain.A synclinal fold is observed adjacent to the Jabal Hafit anticline.In the central portion of the profile, a broad anticlinal structure with gentle dips on both limbs is identified, which corresponds well with the findings from the GWP borehole cross-section (Figure 6).This anticline is adjoined by another synclinal fold featuring a significant thrust fault along its eastern limb.Movement along this thrust fault has caused uplift and erosion of the Upper Tertiary Fars Formation in the eastern part of Al-Jaww Plain.Our findings are consistent with those of Ali et al. (2008), who integrated gravity, magnetic, and seismic data to study the subsurface structure of the Al-Jaww Plain; their results showed that Al-Jaww Plain is underlain by a series of folds with axial traces sub-parallel to the axis of Jabal Hafit.
The 2D AMT model revealed three primary subsurface resistivity zones, as shown in Figure 8.These findings are consistent with the interpreted seismic profile and GWP borehole cross-section, as shown in Figures 6 and 7, respectively.The AMT model indicates the presence of a conductive shallow layer (referred to as Zone A) characterized by resistivity values less than 15 Ωm.Furthermore, the AMT resistivity cross-section also demonstrates a gradual increase in resistivity beneath the conductive shallow layer of Zone A. Zone B corresponds to a zone of intermediate resistivity (approximately 15-250 Ωm) beneath Zone A. At greater depths, the 2D inversion model also reveals a high-resistivity region (Zone C) characterized by resistivity values >250 Ωm.
The uppermost resistivity zone (Zone A), extending to a maximum depth of 1.5 km, encompasses the saturated Quaternary and Pliocene aquifers, in addition to the Upper Cretaceous Simsima and Tertiary aquifers.The Quaternary and Pliocene units in this zone represent various depositional environments, including alluvial, desert plain, sabkha, and aeolian sand deposits, as depicted in Figure 6.Notably, the Quaternary deposits within the Al-Jaww Plain hold the largest reserves of fresh to slightly saline groundwater in the eastern Abu Dhabi Emirate, with TDS values ranging from 500 to 6,000 mg/L (Bruno & Vesnaver, 2021).These aquifers are crucial for water resources in the region (Bruno & Vesnaver, 2021).The Upper Cretaceous Simsima and Tertiary formations primarily comprise porous and permeable carbonate units characterized as a low resistivity zone in the AMT (zone A, Figure 8).
Zone B, characterized by apparent resistivity values ranging from 15 to 250 Ωm in the AMT model, is interpreted as the Upper Cretaceous Aruma foreland basin sequence.This sequence comprises the Lower Fiqa, Juwaiza, and Upper Fiqa formations, as shown in the seismic profile in Figure 7.In contrast, Zone C, situated at depths exceeding 3 km in the eastern part of the study area, exhibits high apparent resistivity values (>250 Ωm).The composition of this zone is interpreted as units associated with the allochthonous unit of the Hawasina thrust sheet, which was emplaced during the Late Cretaceous obduction of the Semail ophiolite.This interpretation is consistent with the seismic profile in Figure 7, which indicates the presence of the Hawasina thrust sheet beneath the Simsima Formation in the eastern part of Al-Jaww Plain.
Due to the skin depth effect, the resolution of MT decreases at greater depths.Hence, we conducted a resolution test on the resistive body within Zone C to assess whether the recovered resistivity feature is well-constrained by the available data.We substituted the high resistive body in Zone C with a conductive structure (10 Ωm) and conducted a forward model to compare the model fit with the observed data.The RMS value increased when the conductive structure was used, indicating that the resistive structure in Zone C is crucial for an accurate model and is indeed well-constrained by the data.In addition, the 2D AMT model reveals the presence of an east-verging thrust fault within Zone A in the eastern Al-Jaww Plain.This thrust fault was likely formed in the late Oligocene-Miocene collision event between Central Iran and Arabia, which led to the closure of the Neo-Tethys Ocean.This thrust is also observed in the seismic profile (Figure 7) and is consistent with findings from gravity data analysis (Abdelmaksoud et al., 2023;Ali et al., 2008).In addition, Bruno and Vesnaver (2021) interpreted the subsurface structure beneath the shallow Quaternary deposits (Zone A) of Al-Jaww Plain as a detachment fold and thrust situated above the Fiqa Formation.
In the Al-Jaww Plain, the shallower Quaternary alluvium represents water-bearing lithostratigraphic units comprising a sequence of sand and gravel layers interbedded with mud layers.These sediments were deposited on the flood plains of rivers draining the Oman Mountains to the east (El Mahmoudi & Gabr, 2006).The primary contemporary water recharge sources of the Quaternary aquifer underlying the Al-Jaww Plain are groundwater underflow, rainfall infiltration from areas of elevated topography, and subsurface water flow through pores and fractures (Gabr et al., 2013).In addition, irrigation processes, fault zones carrying water from deeper saturated layers, and infiltration of water leakage from water transmission lines and urban areas further contribute to water recharge.The observed subsurface resistivity values in Zone A decrease toward the east Oman Mountains, which probably represent the area's main aquifer recharge source.
Overall, the results reported in this study indicate the presence of a Quaternary aquifer within a saturated layer that extends from the surface to a depth of 1.5 km.The area's geological structure, as identified in previous studies, plays a potentially significant role in determining the physical, geological, and hydrological properties of the study area's groundwater.The AMT study results presented here in the context of the findings of previous investigations provide an important basis for environmentalists, researchers, and decision-makers to conduct further investigations into the effects of regional geological and geophysical factors on groundwater occurrence and the environment in general.

Conclusions
The AMT data recorded in this study were inverted to construct a 2D electrical resistivity model representing the study area's subsurface resistivity from the ground surface to a depth of 5 km.The 2D resistivity inversion model was subsequently integrated with a seismic reflection profile and data obtained from 11 shallow groundwater boreholes.The resulting model reveals the presence of three distinct subsurface geo-electrical layers between the ground surface and a depth of 5 km.The first layer is a shallow zone characterized by low resistivity values (<15 Ωm).This layer represents the Quaternary and Pliocene aquifers, as well as the Upper Cretaceous Simsima and Tertiary aquifers, which occur between the surface and a maximum depth of 1.5 km.The underlying second layer exhibits moderate resistivity values (15-250 Ωm) and corresponds to the Upper Cretaceous Aruma foreland basin sequence.At depths exceeding 3 km and extending to the maximum modeled depth of 5 km in the eastern Al-Jaww Plain, a high-resistivity zone (>250 Ωm) is observed.This zone is attributed to the allochthonous unit of the Hawasina thrust sheet.
In future studies, we propose acquiring data from further AMT profiles to investigate the entirety of the study area to achieve a comprehensive characterization of the hydrogeological system of Al-Jaww Plain.Additionally, the application of a 3D resistivity model derived from AMT measurements, combined with low-frequency MT studies, would be beneficial in delineating subsurface structures associated with compressional tectonic events in the region.Overall, this integrated approach would help to achieve a more detailed understanding of the area's subsurface geological framework and aid in the identification and analysis of its key tectonic features.

Figure 1 .
Figure 1.Geology of Al-Jaww Plain(Ali et al., 2008) showing the locations of audio-magnetotelluric stations, seismic, and borehole data.The black dashed line is the Electrical Resistivity Tomography profile(El Mahmoudi & Gabr, 2006).The insert shows a regional map of the United Arab Emirates highlighting the study area (red star).The color bar shows elevation.Panel (a) is generated by GMT(Wessel et al., 2009).Note that MD-1 is the deep hydrocarbon exploration well used to depth convert the seismic profile (Figure7).

Figure 3 .
Figure 3. Phase sensitivity skew calculated for all audio-magnetotelluric (AMT) stations.The red line shows the threshold value of 0.3 (values of η > 0.3 are considered to represent 3D structure).The Phase Sensitive skew shows high values in the noisy region of the AMT deadband and at lower frequencies (longer periods > 1 s).However, in general, the data shows 1D and 2D structures.

Figure 4 .
Figure 4. Phase tensor maps plotted at different periods.

Figure 5 .
Figure 5. Audio-magnetotelluric sounding curves for each of the measurement stations.

Figure 7 .
Figure 7. (a) Uninterpreted seismic line IQS11 across Al-Jaww Plain (for location see Figure 1) and (b) interpreted seismic line IQS11, showing a series of folds and a high-angle thrust fault in the Tertiary section.This seismic line also illustrates the stratigraphic correlation of the Tertiary, Simsima, Fiqa, Hawasina thrust allochthon, and Mesozoic shelf carbonates.The Juwaiza Formation is interbedded between the Upper and Lower Fiqa Formations.The seismic profile was depth converted based on a velocity profile obtained from check shot data from a nearby deep exploration well (MD-1, see Figure 1).The horizontal arrow on top of the seismic profile shows extent of the 2D audio-magnetotelluric model presented in Figure 8.

Figure 6 .
Figure 6.Ground Water Program (GWP) borehole cross-section across Al-Jaww Plain.The locations of the GWP boreholes are shown in Figure 1.

Figure 8 .
Figure 8. 2D audio-magnetotelluric inversion results showing resistivity distribution across the E-W profile from the surface to a depth of 5 km.The direction of groundwater flow is from the east (western part of the Oman Mountains) to the west.