Magnetic Signature of Gold Deposits: Example from Um Garayat Region, South Eastern Desert, Egypt

Magnetic methods of exploration have proved to be efficient and have potential in the gold mineralization industry. New magnetic processing technologies aid in improving the process of interpretation and gold opportunity identification. In this work, we show the possible application of combined digital magnetic filters to explore new gold mineralization localities with application to a well-known Au mineralization zone. Um Garayat (UG) region, southeastern desert, Egypt, is an ancient example of a potential area for gold mining. Modern analysis showed that other types of mineral concentrations are present. So, exploiting the magnetic signature of the area for future investment is of great interest. The old UG gold mine is characterized by volcanic and tectonic features such as faulting and folding that affect the arc sedimentary rock sequence of repeated deformation stages. A thorough geophysical effort has been carried out around the old gold mine in the UG area to explore the extension of mineralized ore deposits. A detailed geophysical survey using magnetics was carried out in this study together with the available aeromagnetic data. Field data sets on appropriate sites were measured, processed, and evaluated by suitable software. High magnetic anomalies were detected based on grid filter analysis and contact occurrence maps as marked as possible ore deposits after satisfying the geologic conditions for gold formation. An integrated understanding of attained results revealed that the new possible ore deposits are related directly to fault and fracture zones in the shape of lenses of variable thickness in this zone. Results show that newly detected mineral occurrences in the UG area are also controlled by major faults and hydrothermal solution enrichment along fault zones at a depth ranging from 20 to 70 m. Favorable fault/joint mineralized places were located. The relationship between the Au-quartz vein's strike direction compared with magnetic anomaly lineament analysis was studied. Search for new sources of Au and other mineral deposits in addition to quartz veins is needed as pockets of accumulated mineral-rich rock fragments are deposited in drainage wadis and fault/joint zones because of hydrothermal solution enrichment.


Introduction
The Nubian shield (NS) presents a potential source of mineralization in Egypt. Gold mineralization has perhaps been the most pursued activity for more than 6000 years in Egypt (Klemm & Klemm, 2013;Klemm et al., 2001), before the Old Kingdom and up to the present. Parts of the NS, a crustal block that runs more than 3000 km from Sinai and eastern Egypt to Ethiopia in the south, are the gold-bearing terranes that make up the Eastern Desert (ED). The Arabian Shield in the Arabian Peninsula, the eastern counterpart of the NS in Egypt, forms together with NS the Arabian-Nubian Shield (ANS), one of the biggest masses of young Neoproterozoic crust on Earth (Pease & Johnson, 2013). The ANS is host to several gold occurrences (Fig. 1), with hundreds of places being marked by trenches, adits, and dumps ( Fig. 2a) in the Eastern Desert. In alluvium, there are placer gold mines (Klemm et al., 2001). The bulk of these ancient mines were situated on quartz-carbonate veins that contained gold in metamorphic terranes, known as orogenic gold. At Sukari, such veins are now being mined in Egypt (Fig. 2), one of Egypt's most prolific gold mines. The Neoproterozoic Nubian Shield of Egypt's southeastern desert contains the Allaqi-Heiani belt (Zoheir & Klemm, 2007) (Fig. 1), a 250-km-long WNW-ESE Groves et al., 1984;Mekkawi et al., 2021;Rehman et al., 2019;Salem et al., 2013;Sultan et al., 2009); Gobashy et al., 2021aGobashy et al., , 2021bGobashy et al., 2022;Abdelazeem et al., 2019Abdelazeem et al., , 2021. Although harzburgite and dunite are serpentinized to create serpentinite, which may contain gold, this process also produces magnetite, which has a relatively high magnetic susceptibility (e.g., Hansen et al., 2005;Toft et al., 1990). The broad link between magnetic susceptibility and gold mineralization has only been briefly discussed in a few published studies. Modern geophysical inversion schemes allowed expressing the relation among subsurface magnetic susceptibility distribution, the observed surface magnetic anomalies, and the surface geology quantitatively. The magnetic tomographic sections increase the investigated zones to an unlimited dimension in magnetic exploration (Abdelazeem & Gobashy, 2016;Gobashy et al., 2021aGobashy et al., , 2021b. Magnetic inversion schemes are a member of the class of data-driven models that model the prospectivity of Au deposits using digital computing techniques. They can derive intricate spatial correlations between mineral occurrences and geoinformation that indicate deposits (Carranza, 2008(Carranza, , 2011Joly et al., 2015;Yousefi & Carranza, 2016). In the present work, we focus on the application of measured magnetic data on and around the potential Au mineralization zones in the UG area (Fig. 2b, c) to study the possible subsurface distribution of Au deposits on a regional scale (Area A) and around the old gold mine and wadi tributaries (Area B) and their relation to structural elements in the study area utilizing modern integrated grid filter analysis and available geologic data. Area B is contained in Area A; a detailed land magnetic survey is carried out in this area for further detailed interpretation. Area A is covered only by aeromagnetic surveys.

Geologic Setting
Geomorphologically, the study region is characterized by low, moderate to high topographic features ( Fig. 3a and b). Outcrops are either sharp or gentle slopes of highly weathered surfaces. Dunes, windblown areas, sand terrains, and playa deposits are frequent in the area. The wadis follow the major structural trends, and the main drainage pattern in the area is dendritic type, the upstream due north. Several wadis are in the area, including um Garayat, Allaqi, and Gebel UG mountain (250 m). Elevation ranges from 185 to 300 m above sea level, which is low relief topography, with ease of accessibility through Aswan asphalt road. Gold mineralization in the

Figure 4
The regional geology of the study area (Area A) and the Au occurrence locations Vol. 180, (2023) Magnetic Signature of Gold Deposits 1057 southeastern desert is associated mainly with a quartz vein system that emplaced granite syn-tectonically along the first deformation shear zone (El-Kazzaz & Taylor, 2001). The origin of auriferous metamorphic fluids, where the veins were formed along with fractures in the easily broken ductile shear zones, was related to deformation and metamorphism rather than epithermal origin. Genetically speaking, and as a criterial conclusion for future exploration in the area, while peak metamorphic conditions were noticeably M. Gobashy et al. Pure Appl. Geophys. greater at depth, up to 500-560°C, Zoheir et al. (2018) revealed that the volcano-sedimentary sequence and ophiolites in the Wadi Allaqi region had mostly metamorphosed under greenschist facies settings (Abd El-Naby & Frisch, 2002). Peak metamorphism coincided with compressive deformation, which was evident in the form of regional folding and sinistral shearing (Zoheir & Klemm, 2007). Additionally, they concluded that enhancing the vectoring of prospective zones would build on spectral and mineralogic characteristics suggestive of fluid concentration at structural junctions. Zones in the Wadi Allaqi area that satisfy the criteria for hydrothermal alteration along shear zones through the island arc metavolcanic/meta-volcaniclastic rocks may be highpotential sites for drilling operations (Zoheir et al., 2019). Metavolcanic, metasedimentary, and metagabbro rocks from the Neoproterozoic are particularly exposed in the area surrounding the UG gold mine. Metasediments include graphite schist and meta-mudstone; the graphite schist is covered in a stockwork of quartz veins and veinlets. According to Sabet et al. (1983), the metavolcanic rocks are made up of meta-andesites, meta-andesitic tuffs, metarhyolites, and meta-rhyolitic tuffs. These rocks were also invaded by andesitic, dioritic, and trachytic dykes as well as a number of quartz veins, veinlets, and lenses (Figs. 4 and 5). Ivanov (1988) and Oweiss and Khalid (1991) studied the rocks' petrography. Both concluded that Wadi Allaqi volcanics typically consist of intermediate to mafic basaltic andesite, andesite, and dacite rocks. According to El-Nisr (1997), the continental arc/margin setting and lowgrade metamorphism of the greenschist facies overlaid by severe hydrothermal changes have impacted these volcanic processes. These rock processes range from low K-tholeiite to calc-alkaline.

Gold Mineralization Sequence in the UG Area
The first of several events that led to the mineralization was the discovery of andesite-granodiorite porphyry in a subvolcanic setting. Rich in volatiles, the original magma chemistry ranged from moderate to acidic. Regional propylitization, which is comparable to the greenschist metamorphic facies, had an impact on the country rocks (Beane & Titley, 1981). Magma rich in H 2 O and CO 2 and deficient in S is indicated by the complete transition of mafic minerals into chlorite, magnetite, hematite, pyrite, sphene, rutile, calcite, epidote, and anatase during propylitization. There are three steps in the hydrothermal stage. Phase I started with apatite, tourmaline, and calcite-bearing quartz lenses and appeared after high-temperature silicification and pyrophillitization, which was the process of leaching alkalis. Minor quantities of pyrite, pyrrhotite, and chalcopyrite also appeared in the milky quartz veins and lenses after that. Alkali metasomatism, which was typified by extensive and intense sericitization, came next. Phase II shows the development of huge metasomatic quartz masses as a result of extensive pyritization, silicification II, and sericitization with copious orthoclase and albite and minimal hydrobiotite. Smaller quantities of local gold, au-tellurides, and proustite were found after that. The third stage of low-temperature silicification III is represented by quartz veins in the study area, which have a high native gold content, Au-tellurides, and silver minerals, and a low pyrite content. Calcite and pyrophyllite also formed during this period. Phase III represents the emergence of alunitization and kaolinitization, which continued to precipitate into the supergene stage. Covellite and chalcocite were created during the supergene stage, and nearly all sulfides, particularly pyrite, were changed into the common mineral limonite (Ivanov, 1988). In the vicinity of the auriferous quartz veins in the southern region, there are impregnations of chalcopyrite, covellite, chalcocite, and pyrite [Hussein (1990), Oweiss and Khalid (1991)]. Figure 6 a to e shows the different gold-bearing veins, excavations, mineshaft, and mining building in the area close to the mine.

Methodology
Combined aeromagnetic and land magnetics were used to investigate mineralized zones and to determine the characteristics of ore deposits, i.e., size, depth, thickness, and other physical properties to be evaluated in the study area. The aeromagnetic is used for regional reconnaissance studies (Area A), lineament analysis, and depth estimation of major Vol. 180, (2023) Magnetic Signature of Gold Deposits 1059 structures for controlling the area, while the land magnetic data are used for detailed analysis of the mine zone. A brief description of the used methods, measured data sets, type of instruments used, and data processing techniques will be given in the following paragraphs.
For the detailed land magnetic survey (Area B),[ 5000 magnetic stations were measured along random profiles to study an area of around 3000 m 9 3000 m in US wadi and its nearby heights. According to topography roughness, records were measured with a station separation of 5-20 m. The acquisition was carried out using two proton magnetometers: a fixed unit at the base station and the other moving along the measuring stations. The recorded point coordinates, elevation, and measurement time were noted using an attached GPS. These profiles are stacked together to form a total land magnetic intensity map (TMI). The map is reduced to a pole to remove the effect of skewness and then subjected to filters to reach the goal of the research. The aeromagnetic data are available from the EGPC (EGPC, 1983). Table 1 displays the survey's parameters. A flow diagram showing the processing sequence is given in Fig. 7.

Gradient Filters
Filtered magnetic maps can provide useful information. The gradients of the field are among these filters. In its most basic form, this contains vertical and horizontal gradients.
The majority of advanced filtering techniques rely on the vertical and horizontal (1st, 2nd, and 3rd) derivatives or gradients of magnetic anomalies, or both to delineate qualitatively features of the source bodies such as edges and centers, as well as the enhancement of low magnetic anomalies. The horizontal gradients are an effective edge detector; they are used in defining the position of edges of the source bodies. Shorter wavelength components of the field are amplified more strongly by the vertical gradients than longer wavelength components. The three first-order derivative components ( oT ox ; oT oy ; oT oz ), the three second-order diagonal components are calculated for the UG area using Oasis Montaj 8.3.3 software (Geosoft, 2015). Linear features are posted on the maps. Several trends can be obtained through rose diagrams and used for further processing.

Analytic Signal
The analytic signal or total derivative filter (AS) is applied to detect lineation and track magnetic discontinuities in the study area. The amplitude of the analytic signal is simply calculated as given in the form (Nabighian, 1972 and1984): where T is the magnetic field, and (x,y,z) are the Cartesian coordinates.

Tilt Angle Filter (TA)
Another popular modification with an enhanced output is provided by the TA filter (Gobashy et al., 2021a(Gobashy et al., , 2021bMiller & Singh, 1994;Salem et al., 2007Salem et al., , 2008Verduzco et al., 2004). It is given as: b Figure 6 UG mining occurrence, a gold-bearing quartz vein, b excavation of Qz veins, c mineshaft, d magnetic measurements over the shaft, and e historic mining building in the area Table 1 Parameters of the aeromagnetic data used and the survey specifications, eastern desert, Area II, Egypt. ( It is a contact strike detector that can be used effectively in complex geologic conditions.

Normalized Source Strength Transformation Filter
This filter is principally used as an efficient edge detection filter similar to the AS and tilt angle filter; however, it uses all the tensor elements. Hence, it is a more resolved and sharper anomaly map. NSS is derived from the magnetic gradient tensor (MGT). For a significant class of sources, it is not strongly reliant on magnetization direction, just marginally so. Therefore, it performs well in complicated structural areas. Its peak is directly above the magnetic source, and it is Flow diagram showing the processing sequence and methodologies used directly proportional to the source's strength. This trait is critical in mineral exploration as magnetic contacts are often used to locate mineralization zones. The angle between the magnetization and displacement vectors for compact sources that are well suited to being represented by a dipole may be computed from the eigenvalues. According to Blakely (1995), the magnetic field B produced by a magnetization distribution M may be expressed as follows: where £ r ð Þ is the potential. The location vectors of the observation point and the integration point, respectively, are denoted by r and r o , and C m = 10 -7 Henrym in SI units. Only five independent components constitute C; the magnetic gradient tensor, which may be diagonalized as: where And where eigenvalue k 2 is the intermediate eigenvalue, which has the least absolute value. Comparing the NSS to the AS, when geologic entities have remanent magnetism, the NSS can give more trustworthy information on the source geometry (Beiki et al., 2012). For the studied region, the NSS transformation map and the two complimentary maps k 2 and u are computed.

Magnetic Anomaly-Model Inversion
According to Bhattacharyya (1966), four factors control the calculated effect of a magnetic source, which are the bottom surface, top surface, magnetic susceptibility contrast, and observed magnetic anomaly value. Factors are calculated in a systematic forward modeling process over 66 profiles, and the anomaly is fitted to observed values. The simple inversion model is based on an assumed two-dimensional geologic model (depth, horizontal direction perpendicular to strike). The Y-axis is assumed to extend to infinity.

Advanced Grid Analysis
The grid analysis of the Centre for Exploration Targeting (CET) is an advanced tool that analyzes an image's texture to find regions of structural complexity. A detailed grid analysis is carried out to calculate the entropy, standard deviation, and structural complexity map from the RTP data. A heat map is produced that shows the future gold opportunity sites in the UG area. The structural complexity analysis is used to determine the likelihood of a gold deposit occurrence. This approach locates crossings, junctions, and changes of direction in the strike by determining magnetic discontinuities, identifying zones of discontinuity, and analyzing structural connections. It makes it easier to choose the locations that are thought to be promising (Holden et al., 2008(Holden et al., , 2010. Therefore, the approach entails the following steps:

Texture Analysis
The local neighborhood of each picture point is defined via texture analysis. The dispersion of grayscale pixel intensities is represented by the standard deviation (STD) filter (Holden et al., 2008).

Texture Ridge Detection
The texture analysis result is then subjected to texture ridge detection to locate the laterally continuous ridges (Kovesi, 1991(Kovesi, , 1997. Vol. 180, (2023) Magnetic Signature of Gold Deposits

Thinning of Texture Ridges
Phase symmetry-identified texture ridges are examined by line segment vectorization, which determines the line segments that constitute the ridges. To do this, a binary grid is created by thresholding the phase symmetry output, designating white for the ridges of the foreground texture and black for the ridges of the background texture (Lam, et al., 1992).

Euler Deconvolution
A semi-automatic Euler deconvolution approach was performed on the profile to quantitatively quantify the depth of the causal source of the  (Reid et al., 1990) is given as: where (x0, y0, z0) are the coordinates of the source, T (x, y, z) is the total intensity magnetic field in Cartesian coordinates, B is the regional (IGRF) field, and N is the structural index (Thompson, 1982). Euler depth solutions are estimated along the selected studied profiles.

Tomographic Magnetic Inversion
Two profiles are selected from the land magnetic survey and seven from the aeromagnetic map are used for a detailed tomographic magnetic inversion using Occam inversion (Constable et al., 1987). Results are 2D models for the distribution of the subsurface magnetic susceptibilities in the study area. In this technique, the 2D half-space is subdivided into a large number of cells or blocks with unknown magnetic susceptibilities. The relative values of these susceptibilities are obtained through magnetic inversion. In brief, the solution to the inverse problem is transformed into the Tikhonov parametric functional's minimization Pa (Portniaguine & Zhdanov, 1999): where m is unknown parameters (magnetic susceptibilities of each cell) and u(m) denotes a functional error estimated of the difference between the expected and observed (calculated) fields. The inequality restricting parameters are m L and m U , and s(m) is a function that helps to stabilize (stabilizer). These are based on the sample's magnetic susceptibility obtained in the research region that has been assessed.

Results and Interpretation
The reduced to pole aeromagnetic map (Fig. 8a) divides the study area (Area A) into zones of different magnetic intensities based on the different rock types in the area. Zone A represents high magnetic anomalies ([ 42,625 nT) concentrated at the NW(A1) north to Hariri gold occurrence, NNE(A2), NE(A5), NW(A4) east to Um Ashira occurrence, and A3 at the central area south Abu Swayel deposit. The intermediate magnetic anomalies (42,300 to \ 42,600 nT) correspond to C1 anomaly zones (light green), which correspond to the east of UG occurrence. Finally, the low magnetic anomalies B1 (\ 42300 nT) correspond to kaolinitic sandstone, and B2 corresponds to quaternary wadi alluvium. The qualitatively detected lineaments trend NE-SW and NW-SE. To examine the regional structural trends in the area, the data are upwarded to 3000 m, 6000 m, and 9000 m above sea level. This is shown in Fig. 8b, c, and d. The rose diagrams show that the dominant trends for the 3000 m upward map are NW, NNW, and SE trends, while for the 6000 m upward continued map NE-SW and NW-SE. Finally, the 9000 m upward continued map shows dominant NW-SE trends with minor NE-SW trends. This agrees with Marten (1986), as we previously mentioned, where the NW to NE represents the strike direction of the exposed mineralized zone and is found to be at the mine area shafts, whereas the regional structural trends are mostly NW with secondary N, NE, and E trends at UG region and Allaqi shear zone. Tectonically, this reflects different strong tectonic events that affected the area throughout geologic history.
The gradient maps, on the other hand, are very effective in edge enhancements. This can be seen in Fig. 9 where the dominant direction of the high magnitude values corresponding to edges or magnetic contacts are NW-SW, NE-SE, and NNE-SSW directions.
The analytic signal filter (Fig. 10a) also detects the edges or contacts in the study area. This filter utilizes only the three diagonal second-order gradients (Txx, Tyy, and Tzz) shown in Fig. 8. The tilt angle filter (TA) is attracted more to the E-W, NW-SE, and NE-SW directed lineaments. The zero contour line (Fig. 10b) directly indicates the trace of the magnetic contacts/faults and suture zones in the study area (Salem et al., 2008). These lineaments are in the E-W, NW-SE, and NE-SW directions. The normalized source strength (NSS) or l map is independent of the magnetization direction in the data, hence it is essential to test the data efficiency using a l filtered map. The results are very similar to the AS filter, where the l peaks correspond to contacts/faults; however, all magnetic tensor elements are used. Hence, the resolution is higher. Figure 11c shows the l map and the two complementary components, k 2 and u maps (Fig. 11a, b). Both can be used to estimate the depths of sources. Hilbig's analysis (1963) of the data shows that the calculated Konigsberger ratio (Q) is \ 1 (0.9876). Q provides a first-order indicator of remnant content. Since the remnant content is acceptable and Q is near 1 in the studied region, the results indicate that the RTP may still be utilized for additional magnetic inversion. The subsurface distribution of the magnetic susceptibilities of the rock units in the study area is examined through magnetic tomographic inversion. Figure 12 shows the results of this analysis along with seven selected profiles covering Area A. Figure 13 shows a quasi-3D view of the subsurface distribution of magnetic susceptibilities.
A more detailed tomographic analysis is conducted along two selected profiles P8 and P9 crossing at the UG mine. This is shown in Fig. 14. Figure 14a shows the location of both profiles, Fig. 14b is the inversion along P8, and Fig. 14c is the inversion along profile P9. The location of the UG mine is coincident with the contact zone between high and low inverted magnetic susceptibility zones. This might be related to a fracture/a large contact or suture zone as indicated also by normalized source strength and edge detection filters.

Magnetic Anomaly Model Inversion
Following Bhattacharyya (1966), 86 profiles are subjected to this simple modeling technique to detect the depth of magnetic sources. In all cases, the Y-axis is assumed to extend to infinity. Table 2 reviews some of the results of inverted anomalies; the depth of magnetic sources (ore body) ranges from 5 to 100 m, with width variation from 8 to 260 m. The location of selected anomalies for model inversion is shown in Fig. 15. Examples of the inverted models are shown in Fig. 16. The land magnetic survey in Area B is subjected to more detailed statistical filters that enhance the magnetic signal over the mineralized zones and identifies the magnetic texture over the known mineralization for further detection of opportunities in the study area. The CET grid analysis is applied to the corrected RTP land magnetic data (Fig. 17).
In this technique, a measure of the textural information within specified windows in the dataset is shown in (Fig. 17a). It depicts the statistical unpredictability of a local data set. Values with contours reflect the degree of randomness displayed by the texture in the area centered on each cell. Regions with strong statistical unpredictability are thought to have high entropy (pink color), whereas those with low statistical randomness have low entropy (blue color).
By locating symmetry axes that are closely connected to the periodicity of its spatial frequency, the phase symmetry (Fig. 17a) finds line-like features. The map clearly shows the linear and curvilinear features in the area which are in close agreement with the results of Euler and Tilt angle filters. Skeletonization (or line thinning) (Fig. 17b) is a morphologic operation that takes a binary grid and skeletonizes each foreground object by iteratively eroding its boundary cells until the object is only two cells wide. This process improves the resolution of the detected lineaments from the previous step.
The limits of the majority of magnetic anomalies are not clearly defined step edges but rather smoothly changing bands, whereas standard gradient-based edge detection approaches identify step edges. The phase congruency transform (Fig. 17b) is a contrast-b Figure 14 Tomographic inversion along profiles P8 (b) and P9 (c). The location map is shown in (a) invariant edge detection technique based on detecting the local spatial frequencies, much like the phase symmetry transform (Kovesi, 1991(Kovesi, , 1997. The map shows detected edges with general trends NE-SW and NW-SE. The high entropy sites are concentrated overlying the four corners of area B while the low entropy is represented at the SW and NE of the area. This indicates a possible underlying shearing or fracture zone with possible magnetization at the boundary corner of the area while the standard deviation (Fig. 17c) gives a rough idea of the regional heterogeneity in the data. At each grid point, it calculates  Vol. 180, (2023) Magnetic Signature of Gold Deposits 1073 the standard deviation of the data values within the neighborhoods. Significant features can differ greatly from the background signal. This is acknowledged at the locations where the entropy fluctuations occur ( Fig. 17c and d). Finally, the contact occurrence density (COD) (Fig. 17d) produces a heat map that highlights the abundance of structural contacts, including places where various structures converge and diverge as well as places where structures undergo major orientation changes. Similarly, the orientation entropy (OE) map (Fig. 17c) indicates areas of potential structural complexity. Locations of high structural complexity and low magnetic relief represent possible opportunities for future exploration of gold mineralization whereas high magnetic relief is suggested for magnetic mineralized types (e.g., serpentine). Figure 17d shows the possible localities of mineralization zones (marked as black stars).
On the other hand, the circular feature transform (CFT), central peak detection (CPD), amplitude contrast transform (ACT), and boundary tracing (PT) algorithms are sequentially used to see the feature boundaries throughout the porphyry identification process. This detection stage will select circularity responses above a user-given threshold value (this is taken as two grid cells as in the CET analysis). These resulting points are the detected centers of porphyry features. Figure 18a, b, and c shows such analyses that, in general, reflect the position of massive magnetized bodies (circular features) and/or concentrations of magnetized bodies (e.g., serpentines) or non-magnetized (low magnetic susceptibility) bodies. From the previous analyses of the data of areas A and B, a close view of mines or gold occurrences located in the areas is given: This shows that: The first gold occurrence (Um Ashira) is situated on a meta-sedimentary/old granitoid (microgranite) rock contact in a moderate-high magnetic area but with (80-100) 9 10 -5 magnetic susceptibility contrast to a [ 2 km depth on sides of the contact. Gold occurrence no. 2 (Haimur gold deposit) is situated on a meta-sedimentary/ophiolite (serpentine, talc, ankrite) rock contact in a moderatelow magnetic area with ranges of magnetic susceptibilities varying from (0 to -85) 9 10 -5 to [ 200 m depth on sides of the contact. Occurrence no. 3 (Hariary) is situated on a young granitoid (microgranite)/mafic-ultramafic intrusion (gabbro-dolerite) rock contact in a high RTP magnetic area with a remarkable l, AS, TDR edge anomaly, and (100) 9 10 -5 magnetic susceptibility with no susceptibility contrast to a [ 1800 m depth. Occurrence no. 4 (Nile Valley Block E mine) is situated on a young granitoid (tonalite)/mafic-ultramafic intrusion (gabbro-dolerite) rock contact, but with high magnetic susceptibility; contrast varies from (-10 to -100) 9 10 -5 to a [ 600 m depth. Occurrence no. 5 (Nile Valley Block A) is situated on a mafic-ultramafic intrusion (gabbro-dolerite)/old volcanic (intermediate to basic) rock contact, bounded by an RTP anomaly with (-10 to -100) 9 10 -5 magnetic susceptibility contrast to a[ 600 m depth. Occurrence no. 6 (Um Garayat) is situated within old (intermediate) volcanic subvolcanic group rock units, but with a not significant NSS, AS anomaly, a high RTP anomaly, and (-10 to -70) 9 10 -5 low magnetic susceptibility contrast to a [ 600 m depth. Occurrences 4, 5, and 6 are situated on the outer edge of a circular moderate-high magnetic anomaly area. Occurrence no. 7 (Atshani) is situated on a mafic-ultramafic intrusion (gabbrodolerite)/old volcanic sub-volcanic group (intermediate) rock contact, with a highly magnetized zone, a detected TDR contact boundary, and (0.0-100) 9 10 -5 magnetic susceptibility contrast to a [ 200 m depth. Occurrence no. 8 (Nekib) is situated on a young volcanic sub-volcanic group (rhyolite-microgranite)/old volcanic sub-volcanic group (schist and tuffs) rock contact on a remarkable and clear magnetic contact/faulted zone tracked by l and AS filter and (126-0) 9 10 -5 magnetic susceptibility contrast observed at 2300 m depth. The above results reveal and confirm that the majority of the studied gold occurrences are associated with contact/faulted zones (zero contours of the TA filter) with low to moderate RTP magnetic anomaly; this suggests a possible relation with tectonic contact zones (or shearing) of different rock units, the majority located on non-magnetic rock types. New suggested sites based on the previously mentioned criteria are shown on the COD or heat map (Fig. 17d).

Discussion and Conclusions
Magnetic imaging gave more detailed information about the subsurface structural conditions of the study area than the constructed surface geologic map of the same area especially in sedimentary and wadi deposit-covered locations. Geologically, Um Grayat is filled by Nile alluvium, wadi alluvium, windblown sands, and other superficial deposits of Pleistocene to Quaternary Hunting (1967) and El-Ramly (1973). An example of assured structural elements that coincide with RTP, AS, l, and TA maps shows that the Old Gold Mine of Um Garayat is located on the edges or contacts of different magnetized bodies. The dominant direction of these contacts is NE-SW. Regional trends interpreted from magnetic measurements emphasize also NW-SE and W-E trends. Shallow (residual) magnetic interpreted trends are N35°E ± 5°, N40°W ± 5°with dextral fault type, and W-E trend.
Site investigation shows that Au-quartz veins' strike recorded at the surface has a N (15°-30°) W compared to accompanied magnetic anomalies of NE-SW and N-S directions that are related to structural trends but it does not have magnetic properties (low magnetic susceptibilities).
Tomographic inversion together with grid filter analysis and normalized source strength transformation confirms that mineral occurrences are closely related to shear and fault/joint zones in the study area, i.e., hydrothermal alteration zones. The dimensiona of the mineralized zone are 10-70 m depth and 8-260 m width as depicted from drilling and geologic site investigation. This is confirmed also by the tilt angle analysis results. Magnetometry can precisely evaluate the mineralization zone for shallow and deep parts with a large degree of accuracy. The present study is an attempt to better understand, using magnetic data analysis, the gold genesis and distribution of the Arabian Nubian Shield and recommends an industry-supported reconnaissance and assay program for new discoveries in the southeastern desert of Egypt.

Acknowledgements
A project funded by the Science and Technology Development Fund (STDF), Egypt. Grant no. 25288 was used to carry out the land magnetic survey.
Author contributions Mekkawi M, supplied the study's funding, conception and design, acquisition of geophysical data. Gobashy M. M. supervised the analysis and interpretation of magnetic data, authoring the paper, critically revising it for the importance of intellectual content, and giving final approval of the version to be submitted. Ezz ElDin carried out the field measurement acquisition, processing and interpretation with the help and supervision of Arafa, S and M Gobashy and drafting the manuscript. The interpretation is part of Mr. Ezz Eldin's PhD thesis and project no. STDF 25288. Khalil MH: cooperated in the manuscript revision and preparation.

Funding
Open access funding provided by The Science, Technology & Innovation Funding Authority (STDF) in cooperation with The Egyptian Knowledge Bank (EKB). Land magnetic field measurements were supported financially by the Science and Technology Development Fund (STDF), Egypt, grant no. 25288.

Availability of Data and Material
Not available.

Code Availability
Not applicable.

Conflict of Interest Not applicable.
Ethical approval Not applicable.
Consent to Participate All authors are participating in this work.

Consent for Publication
All authors agree to publication in this journal.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.