Groundwater Inspection and Aquifer Assessment Using Magnetotellurics and Magnetic Data at the Reclamation Area Around New Sphinx City, Egypt

,


Introduction
One of the most significant and ambitious projects in Egypt in recent periods is the reclamation and development of 1.50 million acres in the North Western Desert (WD).This project is facing major hurdles due to the limited availability of water resources in regions that are away from the Nile River.The area under study includes the Mostakbal Masr and Ganet Masr reclamation and development projects, which are situated in the southwestern and western regions of the New Sphinx City.It lies in the upper northeastern section of the WD.It covers approximately 1,115 Km 2 and is bordered by latitudes 29 ̊52.60 ̓ − 30 ̊10 ̓ N and longitudes 30 ̊ 31 ̓ − 30 ̊53 ̓ E (Fig. 1).
Numerous groundwater wells were dug across the investigated area.Thirteen of these wells (Table 1 and Fig. 2) provide information on their ground elevation, maximum depth, and water table, as well as any available information on water salinity with additional to 3 exploratory wells (WD 19-, WD 19-3, and E WD 19-1) within WD19 (El-Ahram) oil field.In Fig. 3, the generalized stratigraphic panel was presented, which showcases the rock units of the available groundwater wells within the area.
Generally, it was observed that the groundwater wells did not penetrate the Nubian Aquifer.Instead, the water was produced from the shallow aquifers belonging to Jurassic to Miocene rock units.The freshwater is at the uppermost  10-20 m below the water table and then altered into brackish to saline water with increasing depth.The water table ranged between 128 to 155 m from the topographiccontrolled ground surface.The key task of the current study is to investigate the deep groundwater aquifer (Nubian Aquifer) and its expected reserve, identifying the subsurface geological structural setting, and delineating the sedimentary cover thickness.This was accomplished by conducting a Land-Magnetic (LM) survey through 198 points with an average interval of approximately 2 km, furthermore, Nine Magnetotelluric (MT) stations were acquired during 2020 (Fig. 2).

Geological Setting
Hydrocarbon exploration played a great role in studying the North WD subsurface geology and tectonic style as well.Since the Paleozoic era, the North WD has been submitted to diverse tectonic frameworks, leading to the creation of numerous sedimentary basins, sub-basins, ridges, troughs and platforms (Khashaba et al., 2016).The investigated area is situated in the northeastern portion of the North WD.It contains three sedimentary basins; (Tiba-Natrun, Kattaniya, and El Gindi Basins) (Adel, et al., 2008).Tiba-Natrun basin is positioned in the northwestern corner of the area, while, the Kattaniya basin is placed in the central and northern region, and El Gindi basin is located in the southern part separated roughly by latitude 30 ̊ N.
Most portions of the northern WD are enveloped by a thin Miocene deposit that unconformably overlaps older stratigraphic sequence (Said, 1962).The stratigraphic section (Fig. 4) of the study area is belonging to the Paleozoic era to Recent period sedimentary sequence overlying the Precambrian basement ( Darwish et al., 2000;Adel et al., 2008).The exposed geological rock units as shown in Fig. 5 are mainly of the Miocene Gebel Khashab Formation rest over the Oligocene fluvial (Continental) sediments or basalt flows and unconformably overlayed by the Pliocene Sediments (CONOCO, 1987).
The subsurface Jurassic section of the study area was deposited in shallow marine to continental environment, which is differentiated into four rock units from the top to base: 1) the carbonates of Masajid Formation; 2) the siltstone and shale with high organic content of the Khatatba Formation; 3) the siltstone and sandstone of Wadi EL-Natrun and 4) Ras Qattara Formations (Adel et al., 2008).The maximum thickness of these Jurassic rock units reached 9430 ft (2874 m) (Jurassic Depocenter) at well T57-1 and pinching out to the south and are completely missing in the Wadi El Rayan platform (Darwish et al., 2000;Adel et al., 2008).The Lower Cretaceous sediments unconformably overlie the Jurassic section (Darwish et al., 2000).The area's structural framework (Figs. 6 and 7) is dealt with briefly by abundant authors (e.g., Wilcox et al., 1973;Moustafa, 1988;Abdel Khalek et al., 1989;Abdel Aal et al., 1990;Adel et al., 2008).The study area was influenced by three stages of deformation (Adel et al., 2008).
The first stage started with the initiation of the first Paleotethys followed by the Neotethys rifting (Bosworth and Tari, 2021) forming the early rifting phase (Mesozoic Rifting) during the Jurassic and Cretaceous periods.This rifting phase led to the opening of three basins.These are the NE-SW oriented Kattaniya basin, WNW-ESE oriented Tiba Natrun basin, and the NW-SE oriented EL-Gindi Sub-basin (Adel et al., 2008).The majority of faults created during the Jurassic exhibit an E-W to ENE-WSW orientation, predominantly stretching in a nearly north-south direction (Bosworth and Tari, 2021).
The second stage is the basin inversion in the Late Cretaceous-Early Tertiary interval.It caused the inversion of the Kattaniya basin forming positive structure inversion peaked in Santonian time but persisted moderately in the Early Tertiary and called Santonian event (Moustafa and Khalil, 1995;Guiraud and Bosworth, 1997;Guiraud, 1998;Adel et al., 2008;Bevan and Moustafa, 2012;Bosworth and Tari, 2021).In Contrast, the Tiba-Natrun Fault (Tiba-Natrun Basin) as well as the El-Gindi Fault (El-Gindi Sub-basin) were not affected by positive structural inversion but continued normal slip (basin subsidence) till the Miocene.The reason is that the trend of these faults is nearly orthogonal to the compression direction of the inversion (Adel et al., 2008).
The third stage is the late extension in Miocene age (Miocene extension) led to continued slip on the Tiba-Natrun and El-Gindi faults as well as formations of new NW-SE oriented normal faults (Adel et al., 2008).

Hydrogeological Conditions
Most of geologic formations in the North WD are considered as water bearing formations with different qualities and quantities.These aquifers have been investigated by different workers such as (Ezzat, 1974;Korany, 1975;Abdel Mogith et al., 2013;El-Sayed and Morsy, 2018).Base on their studies, the main aquifers include the Nubian Sandstone Aquifer (NSA), the Jurassic-Lower Cretaceous Aquifer (JLCA), the Upper Cretaceous-Eocene Limestone Aquifer (UCELA), the Oligocene Sand Aquifer (OSA), the Lower Miocene Moghra aquifer (LMMA), the Middle Miocene Marmarica Limestone Aquifer (MMMLA) and Pliocene to Recent Aquifers (PRA).Generally speaking, the abovementioned aquifers in North WD are structurally controlled and hydraulically connected (Hilmy et al., 1977;Abd El-Samie et al., 2006;El-Sayed et al., 2017;El-Sayed and Morsy, 2018).The NSA stands out as the primary aquifer formation in the WD of Egypt due to its substantial reserves of fresh water (El-Sayed and Morsy, 2018).The investigated area hydrogeological characteristics consists of a multiaquifer system differentiated into five main aquifers: the LMMA, the OSA, the UCELA, the JLCA, and the NSA.
The LMMA is composed mainly of sands, sandstone, siltstone, and shale.The groundwater occurs almost under free water table conditions (El-Sayed and Morsy, 2018).It is rested on the OSA sediments.
The OSA is consists mainly of sand & gravel layers with interbedded clay, and thin limestone bands at the bottom, and a thick layer of basalt at the top (Massoud et al., 2014).Within this study, we merged the LMMA and the OSA into a single aquifer known as the Oligo-Miocene Aquifer (OMA).These two aquifer systems are interconnected but are physically separated by a basaltic sheet.The groundwater table recorded in most drilled wells at depth ranged between 128 to 170 m from ground surface.Based on the well logging measurements and the lab tests of the produced groundwater, only the uppermost saturated layers with about 15 -20 m fresh water with low salinity ranges, but with increasing the depth, the salinity is increasing as well.
The UCELA is a confined aquifer composed mainly of limestone and shales at the upper layers and transformed into clastic layers composed mainly of sandstone, siltstone, limestone, and shale at its base.The UCELA sediments are unconformably underlies the LMMA and the OSA sediments and sometimes there is a connection between them.Also based on the well logging measurements and the lab tests of the produced groundwater from this aquifer, it has high salinity and with increasing the depth, the salinity is increasing as well.
The JLCA is a confined aquifer composed mainly of sandstone, shale, siltstone, and dolomite at the upper and transformed into limestone, shale, siltstone, and sandstone, at its base.Its sediments are unconformably underlies the UCELA sediments, and there are occasional linkages between them.Groundwater extracted from this aquifer often exhibits high salinity levels.
It is primarily composed mainly of sandstone, with limited proportions of limestone, siltstone, and shale.The sand grains are often consolidated by calcite or silica cement, which makes the rock more resistant to erosion and weathering.This composition provides the sandstone with good porosity and permeability, making it an important aquifer system in the region (Ghoubachi and El-Abd, 2016;Sherif and Sturchio, 2021;Salah et al., 2022;Araffa et al., 2023).

Land magnetic 2.3.1.1. Land magnetic data aquisition
During four-days campaigns in 2020, the total magnetic intensity field (F) through 198 stations were acquired using portable GEM GSM-19 Overhauser magnetometer to identify the subsurface structures controlling the water aquifers.Data measurements were done along (N-S and E-W) profiles using the available roads with station interval of about 2.0 km.Missalat Geomagnetic Observatory was served as base station for diurnal correction.It is situated around 60 km south of the surveyed area.

Land magnetic data processing and interpretation
Oasis Montaj software was utilized for both data processing and numerical 3D forward modeling.Correction for diurnal field variation in the measured stations was carried out using data from the Missalat Geomagnetic Observatory.The International Geomagnetic Reference Field (IGRF) was eliminated from all surveyed stations.Finally, we added the average IGRF value (one value) of the study area to all stations and got the corrected total magnetic intensity map (Fig. 9a).
The magnetic fields produced by geological features subject to distortion because of the inclination and declination angles of the Earth's magnetic field (EMF) (Araffa et al., 2017).In order to rectify this distortion, (Baranov, 1957) proposed the Reduction-To-the-Pole (RTP) method.This is accomplished by applying a mathematical procedure to a grid of values obtained from the the Total Magnetic Intensity (TMI) contour map (Araffa et al., 2017).The RTP of the TMI map (Fig. 9b) can be automatically calculated using (Oasis Montaj, 2007), where the parameters required for the calculation are the inclination (44.621 °), declination (4.565 °), magnetic field strength (43700 nT), and the sensor height (1.50 m) from the ground level.Subsequently, the power spectrum of the analytical signal was employed to distinguish between magnetic sources at different depths (Fig. 10a).The regional and residual maps were produced from the RTP data by implementing Low-Pass Filter (LPF) and High-Pass Filter (HPF).The residual Magnetic Anomaly (MA) map (Fig. 10b) depicts a range of closed anomalies characterized by different amplitudes, shapes, and trends, which may signify the consequences of structures present in the shallow rock units.Meanwhile, the regional MA map (Fig. 10c) displays the deep-seated closed anomalies, as explained by (Abdel Zaher et al., 2018).The estimated average depths of the delineated geological structures were 2.24 km for deep sources, and 0.668 km for shallower sources.According to Telford et al., (1990) derivatives have the tendency to accentuate shallow characteristics and sharpen the edges of anomalies.(Miller and Singh, 1994) suggest that the total horizontal derivative and tilt derivative can be valuable tools for identifying mineral exploration targets and mapping shallow basement structures.Tilt derivative, 1st order horizontal derivative (in the X and Y directions), and 1st order vertical derivative (in Z direction) carried out on the RTP map.The produced four maps (Figs. 11a,11b,11c,and 11d) were used to determine the contact location at zero contour to mapping the structural elements in the area.
El Dawi et al., (2004) employed the Euler deconvolution (ED) technique to quickly interpret potential field data, such as magnetic and gravity data.This method is especially effective for identifying boundaries and rapidly estimating depths.The present study used structure index values of 0, 0.5, 1.0, and 2.0, which were applied to the RTP map.The findings are presented in (Figs. 12a,12b,12c,and 12d).It was revealed that the optimal solution was obtained when the structure index equaled 0, and the depths of the causative source elements ranged from 250 m to over 2500 m.

3D Magnetic modeling
The 3D magnetic modeling is a computer-based method that generates a three-dimensional representation of the subsurface EMF.It involves making changes to the geometry and magnetic susceptibility of the layers through an interactive process.During the modeling, the shapes of the layers or bodies were manually adjusted, while the magnetic susceptibility was chosen both manually and automatically.At the final stage of the Modeling, automatic inversions of magnetic susceptibility within a chosen range of values were applied to enhance the agreement between the observed and modeled potential fields (Maystrenko et al., 2017).
The modeling process involves the creation of a 3D grid comprising numerous cells, with each cell representing a small volume of the subsurface.To define a model, a series of surface grids are stacked with specific distributions of remnant magnetization, and susceptibility assigned to the layer situated beneath each surface grid.Before conducting the modeling, a triangulation is carried out between the depth maps of the identified structures along a pre-defined initial grid to establish the 3D geometry of the model.
The Montaj GM-SYS 3D Modeling system was utilized to perform the 3D magnetic modeling.The construction of the model involves stacking multiple surface grids on top of one another, where each layer is characterized by predetermined distributions of susceptibility and remanent magnetization.The computations are carried out in the wave number domain and are guided by the algorithm presented in (Parker, 1972).
To construct the model, three grids were used: the observed RTP grid (Fig. 15a), the digital elevation model (DEM) grid (Fig. 15b), and the initial basement relief grid resulting from 2D modeling along 12 N-S and E-W profiles.These three grids were produced by gridding the data set into 500 m cell size.The model's geometry includes the coordinate system of WGS84/UTM zone 36N, with a total number of cells of X=67 and Y=60, a cell size of 500 m, origin X= 262000 m, origin Y=3309000 m, and Y-axis azimuth of 0°.The model parameters used include a basement magnetic susceptibility of 0.00775 cgs, EMF magnitude of 43700 nT, inclination of 44.621°, and declination of 4.565°.
After conducting a forward calculation with a number of Taylor terms equal to 7 and an FFT rectangular expansion percentage of 30%, magnetic structure inversion was applied using a maximum iteration number of 20.The misfit error grid (Fig. 15d) between the observed and calculated RTP (Fig. 15c) grids was evaluated, yielding a minimum value of -8.846 nT, a maximum value of 15.203 nT, an average value of -2.255 nT, and a standard deviation of 5.415 nT.The basement depth relief grid (Fig. 15e) resulting from the inversion ranges from -2250 to -4650 m.The 3D representation of the relief of the basement depth with DEM is depicted in Fig. 15f.The northwestern, southwestern, and eastern sectors of the area have the lowest basement depth relief, while the central, northeastern, and southeastern portions have the highest basement depth relief.

Magnetotelluric
The magnetotelluric method, also known as magnetotellurics (MT), is a passive exploration geophysical tool used to infer the subsurface electrical properties.It entails detecting natural fluctuations in the Earth's magnetic and electric fields in order to provide images of subsurface electrical features and their distribution at various depths (Cagniard, 1953;Wight et al., 1977;Vozoff, 1991).Simultaneous recordings of the time-varying magnetic and electric fields at a specific location are captured (Vozoff, 1991).MT technique is used for delineating the subsurface rocks electrical conductivity, which is principally dependent on the pore spaces interconnection and the resistivity of their fluid occupying (Arafa-Hamed et al., 2023).It utilizes concurrent measurements fluctuating horizontal electric fields ( & ) Below the surface, along with the horizontal ( & ) and vertical () magnetic fields (Aboud et al., 2023).At each frequency (ω), the measured components establish an interrelation that gives rise to a 2 × 2 complex impedance tensor (Z), as defined in equation ( 1).Typically, the scalar impedance elements are exhibited as apparent resistivity (ρa) and phase (φ) values, as shown in equation ( 2).This convention is commonly used and described by both (Chave and Jones, 2012;Aboud et al., 2023).
where μ0 is the free space magnetic permeability.

Magnetotelluric data aquisistion
In this work, the acquired magnetotelluric soundings have been measured using three magnetotelluric broad band (Metronix ADU-07e) instruments along nine stations during December 2020.Couple of Metronix MFS-06e broadband induction coil magnetometer sensors, which can measure a broad frequency range from 0.0001 Hz up to 10 kHz, has also been employed.Measurements of electric field variations were taken using non-polarized copper sulfate electrodes of local make, utilizing a dipole length of 100 m.The continuously recording time was ranged between 24 to 72 h.The MT phase response and apparent resistivity curves plotted versus periods or (frequency) for the observed nine MT stations, as depicted in Fig. 16, demonstrate that the XY and YX components exhibit nearly equivalent behavior across all periods, with similar trends observed for the majority of the curves.Additionally, the phase values consistently fall within the range of 0 to 90•.

Magnetotelluric data processing and interpretation
Fig. 17 depicts a 3D visualization of the iso-apparent resistivity maps at selected periods, along with the average apparent resistivity curve of the measured nine MT stations, which represents the study area.Upon qualitative examination, the iso-apparent resistivity maps can be divided into three distinct zones.The upper zone, spanning periods between 0.0009766 to 0.01563 seconds, is distinguished by relatively high apparent resistivity and corresponds to the shallow rock units in the dry zone.The second zone spans periods ranging from 0.0625 to 16 seconds and is distinguished by relatively low apparent resistivity values, representing the saturated sedimentary rock units.The third and final zone spans periods between 16 to 128 seconds and exhibits relatively high apparent resistivity values, corresponding to the crystalline basement.
The one-dimensional interpretation software ZONDMT1D is utilized for analyzing the data.This software utilizes Newton's method (More and Sorensen, 1982) to minimize the misfit between the calculated and observed curves, resulting in a more accurate representation of the subsurface structures.The process of inversion in geophysics may yield non-unique solutions as different models can produce identical calculated curves.To overcome this, prior information regarding the geoelectrical section of the study area, obtained from available well data, is utilized as the starting model.
The analysis yielded results in the manner of the true resistivity and thicknesses of the geoelectrical section at individually station displayed in Fig. 18.The misfit root mean square (RMS) values for the MT stations, namely MT-01 through were 3.80,5.70,3.80,4.40,4.60,6.10,7.30,6.90,and 3.60, respectively.The true resistivity and thickness data obtained were utilized to generate 2D geoelectrical sections, which are illustrated in Figs.19b through 21b, along with the corresponding total magnetic intensity reduced-to-the-pole profiles (Figs. 19a through 21a).Through the integration of magnetic 3D inversion, 2D geoelectrical sections, and available well data, interpreted geological crosssections were obtained and presented in Figs.19c through 21c.
The integrated cross-sections, in addition to the previous studies and available well data shows that the study area contains multi-aquifer system consists of four aquifers; the OMA, the UCELA, the JLCA, and the NSA as mentioned in Fig. 4. In this work the key target is to study the NSAS, which is divided into two major units (layers).
As presented in Table 2 and Fig. 22, depth of the upper unit top varied between approximately -1050 to -3100 m, while its bottom depth ranged from around -1670 to -3210 m below the mean sea level.Its average thickness is approximately 485 m.The lower unit had a top depth varied between approximately -2350 to -3360 m, and its bottom altitude was between approximately -3505 to -4245 m beneath the mean sea level.Its average thickness is approximately 1025 m.

Discussion
The combination of land magnetic and magnetotelluric measurements has gained significant importance in geophysical exploration, particularly for the assessment of subsurface structures, sedimentary rock thickness, geological features, and hydrogeological conditions.Additionally, the integration of these two methods can improve the resolution of the data, as magnetic data can help to constrain the depth of the subsurface structures that are inferred from MT data.Although many studies have investigated the northwestern desert of Egypt's groundwater aquifers, most of them have focused on the shallow aquifers.However, studying the NSA requires more effort due to its increased depth in this area.
The outcomes of this study deliver an important insight into the distribution and occurrence of the NSA, which has implications for various reclamation and development projects in Egypt, including Mostakbal Masr and Ganet Masr.The salinity of groundwater extracted from the drilled wells through these projects varied widely, where total dissolved solids (TDS) ranging from 1900 ppm in EL Yser Company Well-01 at the northeastern portion of the investigated area to 10634 ppm in Well-33/2 located in the southwestern region (see Table 1).This lateral variation in water salinity suggests that the upper groundwater aquifers are not interconnected, and the presence of thick shale/clay layers may contribute to increased water salinity because they contain several salts within their composition based on their marine depositional environment.
The results exhibit a high degree of consistency between the geophysical methods used and are in good agreement with the geological information obtained from shallow groundwater and deep exploratory wells located both around and within the study area.This includes structural trends, basement relief, and groundwater aquifers.The MT data results demonstrate that the NSA extends through the central zone of the area and is bounded by two inverted faults within the Kattaniya basin.It is gradually thinning towards the south and disappears completely at the Gindi basin.The NSA is believed to extend through the northern area in the Natrun and Tiba-Natrun basins, but MT measurements in this zone were limited due to restrictions imposed by the Egyptian armed forces.Therefore, to obtain more accurate information about the aquifer in this zone, further measurements are necessary.

Conclusions
The chief task of this paper is to explore the NSA, a deep groundwater aquifer.The study area includes Mostakbal Masr and Ganet Masr reclamation and development projects, located in the western and southwestern regions of the New Sphinx City, northeastern part of the Western Desert, Egypt.
The basement relief ranged from -2250 to -4650 m, while the NSA was found to extend through the central and northern regions.The aquifer thickness gradually decreases towards the south and eventually disappears affected by the Santonian event during the Late Cretaceous-Early Tertiary interval, which led to the inversion of the Kattaniya basin and the formation of positive structure.It is partitioned into two units: the upper and lower units, with an average thickness of approximately 500 m and 1000 m, respectively.The average depth of the top of the upper unit is approximately -2075 m, whereas the mean depth of the top of the lower unit is around -2855 m beneath mean sea level.
The results of this study have the potential to improve our comprehension of hydrogeological subsurface conditions in the NSA, where no previous studies have explored NSA within the study area.In addition, they can provide dependable data regarding the parameters and groundwater system associated with the NSA.The information obtained from this study is crucial in advancing sustainable development and expanding ongoing reclamation and agricultural initiatives and selecting and selecting optimal locations for digging deep groundwater wells within the area.This, in turn, facilitates the expansion and advancement of the agricultural field and the establishment of new urban communities.

Fig. 2 .Fig. 3 .
Fig. 2. Study area base map shows land magnetic survey stations, MT stations, groundwater wells, hydrocarbon exploration wells, and interpreted geological cross-sections

Fig. 6 .Fig. 7 .
Fig.6.Study area location map with the major structural setting, depositional basins, and geological cross section passing through the available hydrocarbon exploratory wells (modified afterAdel et al., 2008)

Fig. 8 .
Fig. 8. (a) NSA location map, and (b) Cross-sectional diagram provides a generalized view of the regional NSA, including the study area's projection, and depicts the inflow of deep fluids through faults into the aquifer (modified afterMohammed et al., 2022)

Fig. 16 .Fig. 17 .Fig. 22 .
Fig. 16.(a) The observed MT phase response and apparent resistivity data plotted versus periods or (frequency) for the nine MT stations.The XY components are represented by red curves, while the YX components are shown in blue curves, (b) Average phase response and apparent resistivity curves of the XY and YX components for the nine MT stations are illustrated with each station represented by a different color, where the color coding is as follows: MT-01 is represented by blue, MT-02 by green, MT-03 by red, MT-04 by cyan, MT-05 by yellow, MT-06 by magenta, MT-07 by orange, MT-08 by pink, and MT-09 by gold, (c) Average phase response and apparent resistivity curves from the all individual MT stations (gray) and their conceptual average (red)

Table 1 .
Available groundwater wells information

Table 2 .
The NSA upper and lower units (layers) parameters