Delineation of hydrothermal alteration zones associated with mineral deposits, using remote sensing and airborne geophysics data. A case study: El-Bakriya area, Central Eastern Desert, Egypt

El-Bakriya area is situated in the Central Eastern Desert of Egypt. It includes several mineral occurrences and/or deposits. Data from remote sensing and airborne geophysics (gamma-ray and magnetic) were combined to identify the hydrothermal alteration zones and structures related to these occurrences. Various processing of ASTER satellite images revealed fifteen zones of high probability for metallic mineralizations. Airborne gamma-ray spectrometric data clarify K-enrichment zones, which are correlated geologically with the distribution of calc-alkaline, alkali feldspar granites (Younger granites), and calc-alkaline quartzdiorites to granodiorites (Older granites), as well as basement metavolcanics and metasediments. An integrated hydrothermal alteration map was constructed, based on both ASTER remote sensing and airborne gamma-ray spectrometric data. This map delineates some metallic mineral occurrences and/or deposits, which are located in the detected hydrothermal alteration zones. Besides, it displays new potential zones for mineralization, such as: Gabal (G.) El-Shalul, G. El-Bakriya, G. Siwat El-Arsha and G. Umm Bisilla. The geologic and airborne magnetic maps demonstrate five structural lineament trends of E–W, ENE–WSW, NE–SW, NW–SE, and NNW–SSE directions, which might act as pathways to transport the hydrothermal solutions in the study area.


Introduction
The Eastern Desert (ED) of Egypt, where the study area lies, represents the northwestern part of the Arabian-Nubian Shield (ANS). The ANS has structural and economic importance, as it contains precious mineral deposits (e.g., gold, copper, zinc, silver, sulphides, * Ali Elsayed Ali Omar aliomarnma@gmail.com; aliomar@nma.org.eg 1 1 3 chromite, nickel, etc.), as well as petroleum and other resources (Al Shanti et al. 1978;Stern 1994;Blasband et al. 2000;Johnson et al. 2011;Abdel-Karim and El-Shafei 2018;Abdel-Karim 2021). El-Bakriya area, Central Eastern Desert, Egypt, is mainly composed of Precambrian igneous and metamorphic rocks, which are overlain by Upper Cretaceous sedimentary rocks. These rock exposures are traversed by several Wadis (Dry Valleys) filled by Quaternary sediments. The history of mining activity for exploring precious minerals in the study area started from Early Predynastic to Early Arab periods and extended to the twentieth century (El-Ramly et al. 1970;Botros 2004;Klemm and Klemm 2013).
Remote sensing and geophysical data are considered the first step in mineral exploration. They reduce the time and cost by detecting surface occurrences of mineralizations and their subsurface extensions as well as structures (Haldar 2018). The combination between remote sensing and airborne geophysical data is widely used in mineral exploration (e.g., Eldosouky et al. 2017Eldosouky et al. , 2021Aboelkhair et al. 2021;El-Qassas et al. 2021).
Recently, remote sensing has become one of the most important modern technologies applied to geological mapping and mineral exploration (Sekandari et al. 2020;Aita and Omar 2021;Arnous et al. 2021;Omar 2021). The different mineral deposits in the current study area are surrounded by haloes of hydrothermal alterations (Abd-Elmonem et al. 2000;Zoheir and Lehmann 2011;Salem et al. 2014;Khalil et al. 2016;Haggag and Abdelnasser 2021). These hydrothermal alteration zones are discriminated by spectral reflectance and absorption characteristics. Hydrothermal alteration mapping plays a vital and effective role in mineral exploration and is the most prevalent marker of mineralizations (Sabins 1999;Gabr et al. 2015;Pour et al. 2019;Traore et al. 2020;Omar 2021).
Gamma-ray spectrometric data is a good tool in mapping hydrothermal alteration zones related to mineralizations (Efimov 1978;Galbraith and Saunders 1983;Gnojek and Prichystal 1985). Besides, aeromagnetic survey is further useful in defining the boundaries of magnetic sources and structures, which are connected with mineralization (Ammar et al. 1983;Meshref 1990;Abuelnaga et al. 2019;Assran et al. 2019;Eldosouky et al. 2020).
The current study focuses on the use of multispectral satellite remote sensing data (ASTER and GDEM data), as well as airborne spectrometric and magnetic data, to delineate and map mineral alteration zones and structure lineaments in order to delineate the probable locations of new mineralized zones for development processes. The specific goals of this study are as follows: (1) Mapping the occurrences of hydrothermal alteration zones by applying many processing techniques, such as the optimum index factor (OIF), band ratios (BR), and feature-oriented principal component selection (FPCS), on ASTER remote sensing data; in addition to analysis of the K/eTh ratio, K-ternary image, and alteration F-parameter; and (2) Extract surface and subsurface structural lineaments, that may control the existence of mineralizations.
The study area is characterized by the presence of many important ore deposits including radioactive ones, such as: uranothorite, thorite, zircon, fergusonite, xenotime, and chromite, in addition to gold, which are associated with hydrothermal alteration zones (Sabet and Bondonosov 1984;Ghoneim 2014;Helmy et al. 2018;Hassan et al. 2021). In addition, arsenopyrite, pyrite, and trace quantities of chalcopyrite, sphalerite, tetrahedrite, pyrrhotite, galena, gersdorffite, and gold were found in quartz veins and surrounding wallrocks. They are associated with sericite (phyllic alteration), carbonate, chlorite (propylitic alteration) (Zoheir and Lehmann 2011). El-Baramiya region, contains several gold occurrences with gold-bearing quartz veins, most of which are located in the eastward-northeastward-facing El-Baramiya-Umm Salatit belt. El-Baramiya deposits are thought to represent a vein-type gold-arsenic mineralization, adjacent to chromite, magnesite, and antimony ores (Sabet and Bondonosov 1984). Uranothorite and thorite minerals are detected and hosted in alkali feldspar granite and diorite at G. El-Bakriya, associated with hydrothermal alteration mineral (HAM) zones (Hassan et al. 2021).
The auriferous veins at El-Baramiya strike in NE-SW and ENE-WSW directions. Au, Cu, and Cr mineralizations were found in three zones in EL-Baramiya: quartz vein, listwanite alteration, and graphite schist, which contain tectonized and highly-strained portions rich in sulphides, iron oxides (goethite), and malachite, as well as chlorite, talc, and calcite minerals (propylitic alteration) (Salem et al. 2014). In El-Baramiya, chromite, which is found in schistose serpentinites, appear as lenses and folded sheets up to 3 meters thick and 7 meters long (EI-Haddad and Khudeir 1989). The chromite, which was altered to ferrichromite in the study area, might have been caused by late origin of magmatic hydrothermal solutions. Yet, this alteration is also linked to fluid and, tectonic movements, as well as serpentinization. The matrix in the altered variations is mostly made up of chlorite around altered chromite, magnesite that fills fractures, and a trace of antigorite (Khudeir et al. 1992). Helmy et al. (2018) found that gold mineralization is associated with hydrothermal fluids and distributed along shear zones between granitoids and serpentinites, with a content of up to 3.922 ppm in W. El-Baramiya. Gold mineralization is strongly linked to hydrothermally and metasomatically altered regions of ultramafic metasedimentary rocks (averaging 1.0 m thickness), which are nearly fully serpentinized and afterwards changed into talccarbonate rock. The majority of auriferous veins cut graphite-and actinolite-schists, which are intimately connected with listvenite, talc, and a strongly ferruginous shear zone that extends in an E-W direction (El-Bedawi et al. 1983;Osman 1995). Harraz et al. (2012) interpreted the geochemical data from stream sediments in W. El-Baramiya region, using R-mode factor multivariate analytical technique. The application of this technique was successful in finding the binding of Ag-Au-As-Cu-Zn-Pb-Mo-W bearing minerals, indicating the possibility of mineralization in El-Baramiya gold resources. The district is underlain by graphite-schist, in contact with talc-carbonate in the central part of the study area, as well as serpentinized marble and Younger granites in its northwestern portion, which represent the most prospective locations for follow-up explorations.
Several important minerals were identified in G. Umm Salatit, located to the east of the study area, by Ghoneim (2014), such as: (1) Fergusonite, which is confined to muscovite granites, which contains mainly Y, Er, minor Ce-metals, U, Zr, Fe, Ca, Nb, and minimal quantities of Sn and W; (2) Sulphide minerals, such as (pyrite and galena) recorded also in muscovite granites; (3) Xenotime is found in trachyte; and 4) Chromite is found in serpentinites.

Materials and methods
3.1 ASTER data characteristics, pre-processing and processing techniques ASTER satellite sensors produce multispectral images with high spatial and spectral resolution. It senses the electromagnetic energy reflected and emitted from the surface of the earth and the atmosphere in 14 bands (Table 1). ASTER image level L1T 00,311,142,006,082,938, as one scene covers El-Bakriya area and date of pass (14-11-2006) is used in this study (Table 1). These data are radiometrically, terrain corrected, and rotated to a north-up UTM projection. Furthermore, in this paper, ASTER GDEM with a spatial resolution of 1.5 arc (about 30 m) was employed to extract and identify several geological structural systems (Pena and Abdesalam 2006;Aita and Omar 2021;Omar 2021).
Many pre-processing techniques were applied and required before processing of ASTER data, such as cross-talk correction, atmospheric correction, layer stacking and resampling of the 30-m spatial resolution SWIR region to resemble the 15-m spatial resolution VNIR region, so as to preserve the original digital values in the product image. These corrections normalize the data to a scene average spectral range and are appropriate for detecting mineral deposits in arid and semi-arid environments (Sheikhrahimi et al. 2019). Several software programs, including ENVI version 5, ArcGIS version 10.7, PCI Geomatica and Surfer 12, were used for preprocessing and processing ASTER data and ASTER GDEM in the current study.
Three processing techniques were applied to transform the ASTER satellite L1T format into an image, which include the optimum index factor (OIF), band ratios (BR), and feature-oriented principal component selection (FPCS). OIF technique is multivariate statistical information, which was used in this study to determine the best three-band combination from the VNIR and SWIR bands of the ASTER satellite image. The determined three-band 1 3 combination contains the most significant amount of significant information and the least amount of duplication. BR processing is an important technique for detecting, mapping and extracting of HAM, such as: argillic, phyllic, propylitic, and iron oxides (Sabins 1999;Inzana et al. 2003;Pour et al. 2018). In this study, BR of ASTER bands: 2/1 is used to detect iron oxides/hydroxides minerals (jasorite, hematite, and limonite); 5/8 is used to detect propylitic alteration (chlorite, epidote, and calcite); 4/6 is used to detect argillic alteration (kaolinite and alunite); and 7/6 is used to highlight the distribution of the probable sites of ore deposits associated with these alteration minerals (Mars and Rowan 2006;Pour and Hashim 2011;Van der Meer et al. 2012;Pour et al. 2018;Abubakar et al. 2019). FPCS is a technique for detecting halos of alteration in post-magma settings. It is used to minimize the dimensionality of correlated multispectral data and improve the spectral contrast between two locations. This approach highlights the spectral responses of specific objects by choosing four bands of ASTER satellite image data (Crosta and Moore 1989).

Processing of airborne geophysical data
The airborne magnetic and gamma-ray spectrometric survey, used in the present study, was conducted by Aero-Service Division (1984). It was carried out over a great part of the Eastern Desert of Egypt to assist in mineral, petroleum and groundwater exploration. The survey lines were flown directed in a NE-SW direction and at 1.0 km spacing. The tie lines were spaced 10 km and flown perpendicularly in a NW-SE direction. The ground clearance was 120 m above ground level (Aero-Service 1984). A Varian (V-85) proton precession magnetometer sensor and a high-sensitivity 256-channel gamma-ray spectrometer (model Hisens AGRS 3000F system) were used for data collection by Aero-Service. The spectrometric system has two detector arrays, the primary (terrestrial) of 50.3 L, and the secondary (atmospheric) of 8.4 L, and is composed of sodium iodide thallium activated crystals (Aero-Service 1984).
Gamma-ray spectrometric data were collected simultaneously, considering four energetic intervals. Since 238 U and 232 Th do not emit gamma radiation, the daughter products of their decay are used for gamma-ray quantification: 214 Bi and 208 Tl, respectively. The energy windows for each element were: 40 K (1.37-1.57 MeV), eU (1.66 1.86 MeV), eTh (2.41-2.81 MeV) and total count (0.41 2.81 MeV). Aero-Service company done the airborne gamma-ray spectrometric data processing according to the recommendations of IAEA (1976). Available data were corrected from Compton effect, flight effective height and background radiation removal from the aircraft, atmospheric radon gas and cosmic rays (Aero-Service 1984). Geosoft Oasis Montaj program is used in present study to gridding, processing and mapping the airborne geophysical data. Figure 2 displays the steps of methodology that were followed in the current study.
Airborne gamma-ray spectrometric data were utilized to describing the distribution and intensities of radioactive elements (K, eU and eTh) over the main lithologies using qualitative interpretation of K, eU, eTh and K-ternary composite image maps. Besides, the K/ eTh ratio map was constructed to determine where the relative concentrations of K relative to Th increased. Moreover, another indicator parameter, the alteration F-parameter is used to delineate the mineralized alteration zones in the investigated area. The most effective parameter of the three radioelements was suggested by Efimov (1978) as a so-called F-parameter (Gnojek and Prichystal 1985). It is defined as;

3
The F-parameter depends on two main significant relationships, the richness of uranium compared to eTh/K ratio and the richness of potassium compared to eTh/eU ratio of the rock.
Magnetic survey measures the magnitude and orientation of the Earth's magnetic field. Magnetic field at Earth's surface depends on fields generated in Earth's core, magnetic mineral content of surface materials, and remnant magnetization of surface rocks. Magnetic susceptibility is a physical parameter to which magnetic surveys are sensitive.
To overcome undesired distortion in the shapes, sizes, and locations of magnetic anomalies, due the effect of inclination and declination of the Earth's magnetic field, Baranov (1957) and, Baranov and Naudy (1964) proposed a mathematical approach known as reduction to the pole (RTP) to transfer the total magnetic data anywhere into magnetic anomalies as if measured at the magnetic pole. This process was performed by Aero- Service (1984).
In the present study, fast Fourier transformation (FFT) was applied on the RTP aeromagnetic data to produce the regional and residual magnetic component maps. First vertical derivative (FVD), horizontal gradient magnitude (HGM) and tilt derivative (TDR) techniques were also applied on the RTP aeromagnetic map to detect the structures that may play the main role in mineral exploration.
• The algorithm of the vertical derivative is given by Gunn (1975) as follows: where n is the order of the derivative.
• The horizontal gradient magnitude HGM (x, y) for the magnetic field M (x, y) is given by Cordell and Grauch (1985): where M x and M y are the two horizontal derivatives of the magnetic field ( M ) in x and y directions, respectively.
• The tilt derivative (TDR) technique was described by Miller and Singh (1994): where, VDR is the vertical derivative of the magnetic field, and HGM is its horizontal gradient magnitude.

Hydrothermal alteration mapping
The laboratory spectral signatures of the several HAM were selected from United States Geological Survey (USGS) Mineral Spectral Library (Clark et al. 1993). These spectral signatures were loaded in ENVI v.5 program. In VNIR-SWIR range, the HAM (phyllic, argillic, propylitic and ferrugination) were characterized ( Fig. 3) using the following features: (1) The phyllic-altered rocks typically contain sericite, which exhibits a prominent absorption feature at 2.20 μm (ASTER band 6), due to AL-O-H absorption, and a less (4) TDR = tan −1 VDR HGM Fig. 3 a Laboratory spectra and b ASTER endmember minerals spectra used in spectral classification methods (Crosta technique), using USGS spectral libraries for VNIR-SWIR spectral region. Solid coloured lines symbolize ASTER bands wavelength strong absorption feature at 2.33 μm (8th ASTER band), due to Fe-, Mg-O-H absorption; (2) The argillic-altered rocks are characterized by kaolinite and alunite, and these minerals exhibit AL-O-H absorption features at 2.20 and 2.17 μm (5th ASTER band), respectively (Abrams and Brown 1984;Spatz and Wilson 1995;; (3) Propyliticaltered rocks typically include chlorite, epidote and carbonates, commonly calcite, which exhibit an absorption feature at 2.33 μm due to Fe-O-H, Mg-O-H and CO3 vibrational bonds ; and (4) The ferrugination alteration minerals, which are due to secondary processes, typically include limonite, goethite, hematite, and jarosite. Goethite, hematite, and limonite have strong Fe 3+ absorption features at 0.97-0.83 (Fig. 3) and 0.48 μm, whereas jarosite has Fe-O-H absorption features at 0.94 and 2.27 μm (Hunt 1977). The following ASTER processing techniques were applied for mapping and detecting of outcrops of various rock units and their hydrothermal alterations.

Optimum index factor (OIF)
63 colour combinations (R-G-B) were created using permutations of ASTER satellite image (1, 2, 3, 4, 5, 6, 7, 8 and 9), except the thermal band. Table 2 displays the OIF calculation for the ASTER image's VNIR and SWIR bands. The order of bands in any band colour combination, like R-G-B has no influence on the values of OIF, which remain the same. The best colour combination (5, 4, and 1 bands) were used in this study, and displayed as RGB respectively (Fig. 4), according to OIF calculation. This composite is the best combination to identify the different types of alteration minerals (argillic, phyllic, ferrugination, and propylitic), in addition to delineating the outcrop of the various formations in the study area. The argillic and phyllic alteration minerals appear with brown to reddish tones, as a result of Al-O-H. The ferrugination and propylitic alteration minerals appear with green to pale green tones, as a result of Fe-, Mg-O-H. This variation in tones is due to differences in composition and sources of the different types of alteration minerals. Figure 5 shows the surface distribution of the different HAM: argillic, phyllic, propylitic, and ferruginations alteration minerals, which are driven from band 4/band 6, band 7/band 6, band 5/band 8, and band 2/band 1, respectively. According to the geological map of the study area ( Fig. 1), alteration zones can be classified as follow: the argillic alteration zone is hosted and associated with calc-alkaline granite, alkali feldspar granite, metamorphic, metavolcanic, and serpentinite. The phyllic alteration halo is hosted and associated with calc-alkaline granite, alkali feldspar granite, metamorphic, and limited parts of serpentinite. The propylitic alteration zone is hosted and concentrated in serpentinite and metavolcanic. The spatial distribution of ferruginous alteration zones (Jarosite, hematite, and goethite) are associated with metamorphic rocks, calc-alkaline granites, especially around G. El-Bakriya, W. El-Miyah, and G. Umm Bisilla, a small part of serpentinite southeast of W. El-Baramiya, and shale of Quseir Formation southwest the study area. GIS technologies were used to overlay numerous HAM anomalies (Fig. 5) into one map (Fig. 6a) to determine the link between gold mineralization and HAM. The gold mineralizations have a closer spatial link with the propylitic and argillic alteration zones than other alterations in the research region (Fig. 6a), being hosted in calc-alkaline granite, metasediments, metavolcanics and serpintinite. The BR technique alteration mineral maps (Figs. 5a-d) were integrated using GIS tools to delineate the alteration zones. These zones are thought to represent mineralized zone traps. The numerous alteration mineral layers, such as argillic, phylic, propylitic, and ferrugination, were assigned suitable weights (value 1), then the weight of each layer were algebraically summed, and the extracted map was classified into two categories: low and high probable zones of mineralization (Fig. 6b). Fourteen high-probability mineralized anomalies in the study area are distributed in several locations: around G. El-Shalul,

Feature-oriented principal component selection (FPCS)
In the current study, four bands were selected for FPCS to detect and map argillic (alunite, and kaolinite), phyllic (sericite, and illite), propylitic (chlorite, epidote, and calcite) and ferruginous (hematite, jarosite and goethite) alterations (Table 3). Phyllic alteration mineral mapping were performed using ASTER bands 3, 6, 4 and 9 in VNIR-SWIR spectral regions. This is according to the high reflectance at bands (3 and 6) and high absorption at bands (4 and 7) ( Fig. 3a and Table 3). Application of those selected bands in PCA process, was performed to create four PCs for VNIR-SWIR bands. Eigenvector values statistics of the selected VNIR-SWIR bands (Table 4) showed that the third component (PC3) illustrated a high eigenvector loadings in bands 4 and 7 (-0.687 and 0.710) respectively. The high loading is a positive loading, so, sericite distinguished by bright pixels in PC3. The thresholding process was performed on PC3 to omit the low concentration values of phyllic alteration, by using the equation (Mean + 2*Standard Deviation) ( Table 5). By applying this equation, the phyllic alteration areas appear as yellow colour (Fig. 7a).
For mapping the argillic alteration, two high reflectance ASTER bands (3, 6) and two high absorption ASTER bands (1, 4) were selected. The selected bands input in a PCA process were applied to produce four PCs for VNIR-SWIR bands. Eigenvector values statistics of input ASTER VNIR-SWIR bands (Table 4) revealed that the PC4 component is a good Fig. 6 Combination of the various alteration mineral mapping (a), and an integrated of the various alteration minerals (b) extracted from band ratios of ASTER VNIR-SWIR images of El-Bakriya area indicator to argillic alteration, due to high eigenvector loading at bands 4 and 6. Moreover, by applying the thresholding process (Table 5), argillic alteration areas appeared as green colour (Fig. 7b).
The propylitic alteration zone, which is the most common kind, is enriched in chlorite, epidote, albite, sericite, and calcite. This alteration does not mean the decomposition of biotite and amphibole only, but of most rock forming minerals and some accessories (depending on the original composition). Abedini et al. (2020) indicated from the petrographic studies that epidote and chlorite replaced amphibole and pyroxene during andesite host rock alteration. According to Damian (2003), the inner zone of the hydrothermal ore deposit is characterized by pervasive propylitic alteration, where, plagioclase crystals are replaced by an aggregation of sericite, chlorite, epidote, and calcite. Pyroxene crystals are replaced by carbonates and quartz.
Several investigations were conducted to map, identify, and detect the propylitic alteration minerals (chlorite, epidote, albite, sericite, and calcite) using ASTER remote sensing data in various regions across the world (Sabins 1999;Pour et al. 2018;Sheikhrahimi et al. 2019;Sekandari et al. 2020;Arnous et al. 2021). In this study, the PCA technique was applied on four bands (2, 4, 5, and 8 ASTER bands) to detect and map the epidote, calcite, and chlorite minerals, where these minerals have two high reflectance ASTER bands (4, 5) and two high absorption ASTER bands (2, 8) ( Fig. 3a and Table 3). Eigenvector values statistics of input ASTER VNIR-SWIR bands showed that the PC4 component is more contributing for these alteration types. The propylitic alteration appears as a bright pixel due to the fact that eigenvector loading in band 5 is a positive loading. The thresholding process was applied on PC4 to remove the low concentration values of the propylitic alteration areas (Table 4). The produced image (Fig. 7c) distinguished the propylitic alteration areas, which appear as purple colour.
For mapping the ferrugination alteration minerals, two high reflectance ASTER bands (4, 6) and two high absorption ASTER bands (3, 9) were selected. These four bands were used as input bands in the FPCS process to produce four PCS with their eigenvector values statistics (Table 4). PC3 component has high negative eigenvector loading for band 4 (−0.699), so this alteration is clear with darkness pixels. By multiplying all pixels of this alteration by (−1), the dark pixels were converted to bright pixels. Besides, by applying the thresholding process (Table 5), ferrugination alteration areas appear as a red colour (Fig. 7d).  The FPCS technique alteration mineral maps (Fig. 7) were integrated using GIS tools to determine the alteration zones. The numerous alteration mineral layers, such as phyllic, argillic, propylitic, and ferrugination, were reclassified and assigned suitable weights as the same in the BR technique maps. The extracted map was classified into two categories: low and high probable zones of mineralization (Fig. 8). The high probability of mineralization category was found to contain two types of alteration minerals, but the low probability of mineralization category contains one type. Nine high probability anomalies were defined and distributed in the study area (Nos. 1-9; Fig. 8). Anomalies 1, 3, 5, and 7 are hosted in calc-alkaline granites, anomalies 2 and 9 are associated with Taref Fm., anomaly 4 exists on metasediments, anomaly 6 is hosted in metavolcanics, and anomaly 8 is linked with serpintinite.

HAM modeling for detecting potential mineralization zones
The main target of this section is to display definite areas that contain more than one type of alteration minerals at one time. Wherever, the areas that contain more than one alteration mineral type are considered to be mineral traps in the study area. The integrated mineral maps produced from the enhanced remote sensing data (Figs. 6a and 8) were integrated using GIS tools to delineate the alteration zones in the research. An algebraic summation of the pre-estimated weights of these maps was performed. The obtained integrated alteration mineral predictive anomalies map ( Fig. 9) was divided into two categories: low and high probability of mineralization. Fifteen high probability anomalies were defined and distributed in the study area (Nos. 1-15; Fig. 9 and Table 6). The first anomaly is associated with metamorphic and serpentinite near G. El-Shalul. Anomalies numbers 2, 3, 5, 7, 9, 10, 11, 12, 13, and 14 are confined to calc-alkaline granites in several locations: around G. El-Bakriya, south G. Siwat Al-Arsha, around G. Umm Bisilla, north G. Siwat Al-Arsha, south W. El-Baramiya, between W. El-Miyah and W. El-Baramiya, southwest Daghbag, northwest and northeast G. El-Rukham, and north G. Umm Salatit, respectively. Anomalies numbers 8 and 15 are confined to serpentinite and distributed at G. Abu Marwa in the south of the study area, and south of G. Umm Salim, respectively. Anomaly number 4 is associated with metasediments and located at the north of Daghbag in the north of the study area. But, the anomaly number 6 is confined to Hammamat clastics and metavolcanics, restricted to east of G. El-Rukham, east of the study area. Finally, the gold mineralization and some associated minerals were associated with propylitic alteration in serpentinite hosted rocks in the study area north of W. El-Baramiya and near G. Umm Salatit. They were also associated with argillic, phyllic, and ferrugination alterations in Calc-alkaline granites host rock at G. El-Bakriya.  . 7 The resulted a PC3 from phyllic alteration minerals, b PC4 from argillic alteration minerals, c PC4 from propylitic alteration minerals and d PC3 from ferrugination alteration minerals, images using FPCS process for the input VNIR-SWIR spectral bands, El-Bakriya area, Central Eastern Desert, Egypt

Automatic lineament extraction
Lineaments, according to Gupta (1991), are simply linear or curvilinear edges that can be related to geological structures (faults, joints, and line weakness), geomorphological features (cliffs, terraces, and linear valleys), tonal contrast due to (vegetation, soil moisture, and rock compositions), and human activities and/or constructions (roads, tracks, buildings, mining, etc.). ASTER GDEM was used to extract structure lineaments in this investigation. Eight shaded relief images were created utilizing eight differing illumination directions (0°, 45°, 90°, 135°, 180°, 225°, 270°, and 315°) at a constant solar elevation of 30° in order to determine linear terrain features from the DEM (Fig. 10). These images were blended into a single shaded relief image using GIS overlay, yielding a single shaded relief image with multiple lighting directions (0°, 45°, 90°, 135°, 180°, 225°, 270°, and 315°) (Fig. 11). Shaded relief images were utilized as input data for the examined area's automatic lineament extraction procedure (Fig. 12). In addition, the lineament density map (Fig. 13) was produced from the lineament map using ArcGIS software v10.8. The produced density map (Fig. 13) was classified into five classes: very low density, low density, moderate density, high density, and very high density. The high and very high lineament density zones, which appear with violet and red colour respectively, are distributed in different regions of the study area especially at G. El-Shalul, G. El-Bakriya, G.

Radioelement concentrations maps
The spatial distributions of the three radioelement (K, eU, eTh) concentrations of El-Bakriya area are given on Fig. 14a-c. In this Figure, younger granites can be easily delineated due to their high concentrations of the three radioelements. Meanwhile, the remaining different geological units are characterized by relatively low values of natural radioelement abundances. The K, eU and eTh radioelement maps (Fig. 14a-c) show differences in the radiospectrometric levels, according to the differences in lithologic units.
The potassium (K) map (Fig. 14a) can be divided into three levels of K-concentrations. The first level (lowest) is characterized by values < 0.5% and is associated with serpentinites, as well as Quseir and Taref Formations. The second (intermediate) level ranges from 0.5% to 1.4% and is represented in parts of metagabbro, metavolcanics, metasediments, Hammamat sediments and Wadi sediments. The third (highest) level Fig. 9 An integrated alteration mineral predicative model anomaly map using remote sensing data processes of the El-Bakriya area, Central Eastern Desert, Egypt Table 6 The areas of high-potential mineralization due to hydrothermal alteration processes in the study area oscillates from 1.4% to 3.7% and is associated with metamorphic rocks, older granites, calc-alkaline granites and alkali-feldspar granites.
The equivalent uranium (eU) map (Fig. 14b) shows three characteristic levels of eU concentrations. The lowest level has eU contents diminishing to < 1.5 ppm and is connected with serpentinites, metagabbro, metavolcanics and metasediments. The intermediate level (1.5-2.5 ppm) is recorded over some parts of Quseir Formation, Taref Formation, Wadi sediments, Hammamat sediments and metamorphic rocks. The highest level possesses relatively high eU contents with values varying from 2.5 to 11.8 ppm, and is associated with older granites, calc-alkaline granites and alkali-feldspar granites. Figure 14c reflects the three concentration levels of equivalent thorium (eTh) map. The lowest level (less than 4.5 ppm) coincides with serpentinites, metagabbro, metavolcanics and metasediments. The intermediate level shows values that oscillate between 4.5 and 7.0 ppm and is correlated with Taref Formation, Wadi sediments, Hammamat sediments, Quseir Formation and metamorphic rocks. The highest level indicates concentration values exceeding 15.0 ppm eTh and is encountered over older granites, calc-alkaline granites and alkali-feldspar granites.

K/eTh ratio map
Thorium is rarely leached by post-magmatic processes and it does not generally accompany potassium during hydrothermal alteration processes. Therefore, the K/eTh ratio  relief images a, b, c, d, e, f, g and h) generated using eight contrasting illumination directions, 0°, 45°, 90°, 135°, 180°, 225°, 270° and 315° respectively, at a constant solar elevation 30o of El-Bakriya area, Central Eastern Desert, Egypt can reveal possible areas of hydrothermal alteration and mineralization (Galbraith and Saunders 1983). Mapping of the K/eTh ratio provided an excellent contrast between potassium related to alteration and anomalies associated with normal lithological variations (Fig. 14d). This ratio decreases over basic rocks and reaches to minimum values over Taref and Quseir Formation in the southwestern part of the study area, while the acidic rocks show the highest K/eTh ratio. This ratio increased with the increase of acidity, where the potassium content increases. The K/eTh ratio map (Fig. 14d) shows that, the zones affected by hydrothermal processes have high values of more than 0.30 of the K/eTh ratio. These zones are mainly associated with calc-alkaline granites, older granites, metavolcanics and metasediments.

Ternary image of K and K ratio data
The K-ternary image map is a composite colour image, consisting of the concentration distributions of K (in red), K/eTh (in green) and K/eU (in blue). The responses of white colour refer to areas that contain high K concentrations, while responses of black colour represent low K concentrations. Generally, the K-ternary image map gives a better picture of the geology. This can be seen in the good relationship between this Fig. 11 Combining eight shaded relief images at solar angels 0°, 45°, 90°, 135°, 180°, 225°, 270° and 315°, El-Bakriya area, Central Eastern Desert, Egypt 1 3 (Fig. 15a) and the geological map (Fig. 1). Areas with high K concentration responses are associated and largely confined to granitic and metamorphic rocks. The moderate to low-values K concentrations restricted to serpentinites, metavolcanics, metasediments, Hammamat clastics and metagabbro. In particular, low-values of K concentration are associated with Quseir Formation, Taref Formation, and some parts of serpentinites in the western, southwestern and southern parts of the study area (Fig. 15a). The variability of colour tones in the area under study reflects the felsic and basic intrusions and hydrothermal alterations.

Statistical treatment of the data
Processing of airborne gamma-ray spectrometric data of the studied area, includes: (1) Separation of the measurements of the three radioelements K (%), eU (ppm) and eTh (ppm), registered over every lithologic unit; (2) Determination of their characteristic statistics, such as minimum (Min.), maximum (Max.), arithmetic mean (X) and standard deviation (S). The results of the statistical analysis are shown on Table 7.
In the study area, twelve lithologic units were classified according to the geological map (Fig. 1). The radioelement concentrations recorded on these rock units were found to range Fig. 12 Surface lineaments automatically extracted from the shaded relief obtained from combining eight shaded relief images at solar angles 0°, 45°, 90°, 135°, 180°, 225°, 270° and 315° overlay on the density map, with directional analysis (rose diagram) of lineament frequency% and that of lineament lengths %, El-Bakriya area, Central Eastern Desert, Egypt from a minimum of 0.01% K for serpentinites, 0.1 ppm eU for serpentinites, metavolcanics and Wadi sediments, and 0.3 ppm eTh for serpentinites and Wadi sediments, to a maximum of 3.7% K, 11.8 ppm eU for calc-alkaline granites and 28.9 ppm eTh for alkali-feldspar granites (Table 7).

Alteration F-parameter
The F-parameter is very useful in identifying zones of potassium alterations associated with gold, copper and polymetallic mineralizations (Gnojek and Prichystal 1985;Abd El Nabi 2012). Efimov (1978), who determined the value of the F-parameter, showed that in common non-altered rocks, it can take values up to 1.2 or 1.3, and in altered rocks, it may be 2 or even 5, in exceptional cases 10. The F-parameter values (in %) for each rock unit were calculated, based on the F-parameter equation (No. 1). The results were summarized in Table 8, and shown on Fig. 15b as an alteration F-parameter colour map. As an illustration on Fig. 15b, values of the F-parameter, as estimated from airborne spectral gamma-ray data, vary from a minimum of 0.0 for Serpentinites, to a maximum of 2.54% for calc-alkaline granites (Table 8), showing the metasomatism of the potassic alteration associated with mineralization related to these granites.

Airborne magnetic data
The reduced to the north magnetic pole (RTP) map (Fig. 16a) reveals parallel positive and negative linear features. These features take a NNW-SSE direction and may be associated with faults, shear zones and buried dykes. KRS et al. (1973) and KRS (1977) referred to the existence of linear magnetic anomalies (normally or reversely polarized) that take a NNW-SSE direction (parallel to Red Sea axis) in the Southeastern Desert of Egypt. They are due to basic (ultrabasic) dykes, which lie near the surface and related to Red Sea rift system. The dykes contain considerable concentrations of ilmenite, ferrimagnetic minerals and sulphides (pyrite, pyrrhotite and partly chalcopyrite) (KRS 1977). The same pattern of linear magnetic anomalies is found in Saudi Arabia, and controlled the mineralizations (Dadet et al. 1970). The repetition of this phenomena along both sides of the Red Sea suggests that such phenomenon and mineralizations are closely related to the Red Sea rift system (Dadet et al. 1970).
The RTP map (Fig. 16a) displays also negative magnetic anomalies (blue colour), located in the southern and northwestern parts of the study area. These anomalies are associated on the surface, with Quseir Formation, metasediments, metavolcanics, older and calc-alkaline granites. Besides, this map shows high positive magnetic anomalies (pink colour) that have elongated and semi-circular shapes. These anomalies are situated at the southern, west-central and northeastern parts of the study area, around W. El-Baramiya, G. El-Bakriya and G. El-Hidilawi. They are structurally-controlled and mainly related to serpentinites and metagabbro (Fig. 16a). At the southwestern part of the study area, two high positive magnetic anomalies occur under Taref Formation, at shallow depth (Fig. 16a) and may be due to serpentinites, which may extend under this Formation. Most of the gold mines, mineralization occurrences, and/or deposits are associated with the contacts between high positive and negative magnetic anomalies in the study area. In addition, they are located on the ENE-WSW, NE-SW, NW-SE and NNW-SSE structural trends or their intersections (Fig. 16a).
The high-pass map (Fig. 16b) highlights the existence of the previously-mentioned negative and positive magnetic anomalies, but with small elongated and semi-circular closures, trending in the ENE-WSW, WNW-ESE and NNW-SSE directions. These anomalies appear more smooth and regional on the low-pass map (Fig. 16c). The existence of these anomalies (negatives and positives) on the RTP, high-and low-pass maps confirm that these anomalies have deep roots.
FVD, HGM and TDR filters were applied to the RTP aeromagnetic data (Figs. 16d,17a,b). These filters were used to delineate and emphasize the near-surface lineament structures, such as faults, contacts, dykes and shear zones, which may have main roles in mineral deposit distribution. The parallel positive and negative linear magnetic features that occur on the FVD map (Fig. 16d) are located in the central and northeastern parts of the study area, trend in a NNW-SSE direction. Moreover, this map displays the positive and negative magnetic anomalies with different sizes, amplitudes and frequencies, especially at the southern part of the study area (Fig. 16d).
The HGM map (Fig. 17a) describes high gradient values (pink) with amplitudes reaching 2.25 nT/m. The highest amplitude gradient values (more than 0.39 nT/m) may be related to different lithological and structural features (faults, contacts, dykes and shear zones), as shown on Fig. 17a, which shows two main trends in NNW-SSE and ENE-WSW directions.
The TDR map (Fig. 17b) is used to delineate the horizontal edges of the magnetic bodies or structures in the study area, where, the zero-contour line shows the horizontal positions of lateral changes of susceptibilities between negative and positive magnetic bodies (Fig. 17b). This map shows two main magnetic lineaments trends: NNW-SSE and ENE-WSW on the HGM map ( Fig. 17a and b).

Discussion
The integration of remotely sensed data with airborne spectrometric and magnetic data from the research region allows for the identification of suitable sites for mineral deposits (Eldosouky et al. 2017;Aboelkhair et al. 2021). Mineral favorability maps may be readily created using GIS by combining the findings of several investigations such as geological, geophysical, and geochemical research. GIS enables a more accurate forecast of mineralization potential by employing a strong weighting algorithm (Omar 2021;Arnous and Omar 2021).
Analysis and integration of the processed ASTER remote sensing and airborne geophysical (gamma-ray spectrometric and magnetic) data over the study area enabled us to infer accurately the boundaries of different rock units, hydrothermal alteration zones, and locations of mineralization. Besides, such data were utilized to detect the relations which might be present between the gamma-ray spectrometric anomalies and other known mineralizations, with mapped and interpreted structural trends. An integrated map (Fig. 18) was constructed by overlaying the ASTER multispectral satellite imagery processes (Figs. 4 and 9) and airborne gamma-ray spectrometric data (Figs. 14 and 15) to display the relation between mineralizations and hydrothermal alteration zones. The processed remote sensing data permitted the delineation of areas of high probabilities for mineralization, from 1 to 15. Besides, airborne gamma-ray spectrometric data illustrated the hydrothermal alteration zones (Fig. 18). Table 6 illustrated the relations between high probability mineralization category, alteration minerals and possible associated metallic minerals.
The other five areas of mineralization (Nos. 4,8,9,14 and 15;Fig. 18) correspond with moderate and low levels of gamma-ray spectrometric data (Figs. 14, 15b), shown as green to dark colours on the K-ternary image map (Fig. 15a). These areas are restricted to the contacts between Taref Formation with adjacent rocks, metamorphics, metasediments, metagabbro, serpentinites and calc-alkaline quartzdiorite to granodiorites ( Figs. 1 and 18). The processing and mapping of remote sensing data revealed that these areas are correlated with different types of alterations as: phyllic, argillic, ferrugination and propylitic (Figs. 7 and 9; Table 6). Au, Pb, Ta, La, Ni, Cu and Cr elements represent the possible metallic mineralizations that may occur associated with hydrothermal alteration areas (Table 6). The present study revealed that G. El-Shalul, G. El-Bakriya, G. Siwat El-Arsha and G. Umm Bisilla, may generally represent considerable and significant areas for metallic mineralizations.
The examination of Fig. 18 revealed that Cr, Ni, and Sb mineralizations are associated geologically with serpentinites, metasediments, and metavolcanics, located in the Fig. 18 Integration of alterations zones as extracted from remote sensing and airborne gamma-ray spectrometric data, El-Bakriya area, Central Eastern Desert, Egypt. OIF RGB-band 541 as background southern and southeastern parts of the study area. These mineralizations are connected with low radioactivity values (Figs. 14 and 15). The hydrothermal alterations delineated from remote sensing data demonstrate that Cr, Ni, and Sb elements are associated with ferrugination, and some propylitic and phyllic alterations (Figs. 7 and 9). This result agrees well with previous geological studies achieved in the study area by Zoheir and Lehmann (2011);Khalil et al. (2016).
The structural lineaments in the study area were extracted and interpreted from geologic, remote sensing, and airborne magnetic maps (Figs. 1,12,. The obtained results are presented in the form of rose diagrams (Figs. 12 and 19). The structural lineaments were mapped automatically by applying the hillshade process to ASTER GDEM data (Fig. 12). They demonstrate significant relationships between faults and mineralized zones, with a high density of lineaments around the mineralized regions (Fig. 13). The directional analysis of the automatically extracted lineament map and rose diagrams (Fig. 12), revealed the existence of five major faulting trends, in E-W, ENE-WSW, NE-SW, NW-SE, and NNE-SSW directions, in a descending order, with N-S direction, as a minor faulting trend. According to the geological map, gold mineralization and some associated elements, such as pb, Ni, Cr, and Sb are closely related to the high and very high structural lineament density categories (Fig. 13).
From the geophysical data , three trends (E-W, ENE-WSW, and NNW-SSE) are considered as representing fault zones. However, the ENE-WSW, NE-SW, NW-SE and NNW-SSE structural trends have a prime importance and are considered as significant targets for follow-up and field investigations for mineral occurrences. In general, the results derived from processing of remote sensing data match and agree well with those derived from geophysical data. A good agreement was found between these results with many previous field geological studies, carried out for the study area (such as El-Baramiya, Umm Salatit, Umm Salim, G. El-Bakriya, and Daghbag), which confirmed  . 16a); c High-pass (Fig. 16b); d Low-pass (Fig. 16c) maps of El-Bakriya area, Central Eastern Desert, Egypt the association of different mineralizations with these structural lineaments (Hagag and Abdelnasser 2021; Helmy et al. 2004;Kochine and Bassyuni 1968;Zoheir et al. 2019).

Conclusions
The analysis of the obtained results of the different ASTER remote sensing and airborne geophysical (gamma-ray spectrometric and magnetic) data revealed the following: 1-The ASTER remote sensing processing techniques (OIF, BR, and FPCS) indicated the existence of many hydrothermal alteration zones that were classified, according to their priorities, into fifteen high-potential mineralization zones. 2-The K anomalous zones are well defined along the K and K-ternary maps. Mapping of the K/eTh ratio is considered a useful tool to differentiate between potassium anomalies related to alterations and those associated with lithological variations. This map also defined the hydrothermal alteration zones, which are mainly associated with granitic and metamorphic rocks. The F-parameter map shows the locations of metasomatism of potassic alterations, associated with mineralizations that are related to granites. 3-A good agreement existed between the gamma-ray spectrometric findings, metallic locations and alteration zones, that were detected from the remote sensing data. In addition, the identification of new potential mineralization zones was achieved in the present study, e.g., G. El-Shalul, G. El-Bakriya, G. Siwat El-Arsha and G. Umm Bisilla. 4-The analyses of structural lineaments deduced from geological, ASTER RS, and airborne magnetic geophysical data revealed that the study area was affected by five significant sets of structural trends: E-W, ENE-WSW, NE-SW, NW-SE, and NNW-SSE. These structures played an important role in the transportation of hydrothermal solutions, loaded with minerals to the surface or near-surface to precipitate in the form of ores, in the study area. 5-It is recommended to carry out detailed geochemical and ground geophysical studies at the determined alteration zones that were detected from the present study and identify the known as well as the new mineralization occurrences and determine their grades and extensions.
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:// creat iveco mmons. org/ licen ses/ by/4. 0/.