Fusion of Multispectral Remote-Sensing Data through GIS-Based Overlay Method for Revealing Potential Areas of Hydrothermal Mineral Resources

: Revealing prospective locations of hydrothermal alteration zones (HAZs) is an important technique for mineral prospecting. In this study, we used multiple criteria inferred from Landsat-8 OLI, Sentinel-2, and ASTER data using a GIS-based weighted overlay multi-criteria decision analysis approach to build a model for the delineating of hydrothermal mineral deposits in the Khnaiguiyah district, Saudi Arabia. The utilized algorithms revealed argillic, phyllic, and propylitic alteration characteristics. The HAZs map resulted in the identiﬁcation of six zones based on their mineralization potential, providing a basis for potential hydrothermal mineral deposit assessment exploration, which was created by the fusion of mineral bands indicators designated very low, low, moderate, good, very good, and excellent and covers 31.36, 28.22, 20.49, 10.99, 6.35, and 2.59%. Based on their potential for hydrothermal mineral potentiality, the discovered zones match gossans related to sulﬁde mineral alteration zones, as demonstrated by previous studies.


Introduction
Remote sensing techniques have provided valuable tools for characterizing and delineating geological, structural, and lithological features that have aided in the identification of mineralization regions [1,2]. Because of its fine geospatial, radiometric, and spectral resolution, remotely sensed data provides significant information for mineral exploration. One of the main aims of remote sensing investigations is the delineation of hydrothermal alteration zones and the identification of the mineralogical signature [1,3,4]. Hydrothermal alteration zones (HAZs) and their grade must be characterized in order to identify possible mineral resource locations [1,2,5,6]. This is due to the fact that such a process is frequently linked to the economic concentration of base metals like Au, Cu, and Ag. Several studies were conducted using multispectral remotely sensed data to characterize the extent of the hydrothermally altered areas and to identify the minerals forming zones [3,[7][8][9][10][11][12][13][14][15].
Data from satellites can be used to detect new prospects prior to detailed and expensive ground research [4,15]. Landsat Operational Land Imager (OLI) and Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) images were used to process and analyze remote sensing multispectral datasets. Electromagnetic (EM) radiation reflected, transmitted, or backscattered from the Earth's surface is sensitive to remote sensing devices such as OLI and ASTER. With passive or active systems, remote sensing sensors can monitor wavelengths of EM radiation in the visible near-infrared and shortwave infrared (VIS/NIR/SWIR) to microwave. Landsat satellite image data have been utilized for lithologic mapping using image transformation techniques [7,8,[16][17][18].

Study Area
The present study is a part of Arabian Shield, Khnaiguiyah, Saudi Arabia. It extends between latitudes 24 •  The district of Khnaiguiyah is located at the eastern periphery of the Arabian Shield (Figure 1), which is the exposed Precambrian basement of the Arabian Plate. The Arabian Shield (ANS) is the northernmost extension of the East African Orogen [31,32] and consists of a collage of tectonostratigraphic terranes with ensialic and ensimatic arc affinities [33,34]. The convergence of East and West Gondwana caused the terrane amalgamation/accretion during the Pan-African event (780-600 Ma) [34]. The final suturing (680-610 Ma) of the ANS coincided with the development of the Nabitah fault zone, gneiss domes, and massive Molasse basins [35,36]. The NW-SE trending Najd fault system/Najd Orogeny (620-540 Ma) subsequently formed due to escape tectonics concomitant with the assembly of the Gondwana supercontinent [37,38]. By the Cambrian (541 Ma), the ANS was established as a stable juvenile continental block forming the northeastern margin of Gondwana [39]. Khnaiguiyah Zn-Cu deposits represent a substantial zinc resource and attracted considerable exploration efforts in the last three decades [40]. The four ore bodies of the Khnaiguiyah district comprise mineable reserves of up to 11 Mt averaging 7.41% Zn and 0.82% Cu [41]. Khnaiguiyah-type deposit features are consistent with metamorphism and deformation of volcanogenic massive sulfide (VMS) mineralization; the related stratiform Mn-rich units are mainly suggestive of a seafloor hydrothermal setting.
The Khnaiguiyah ores are hosted by the Shalahib Formation (1500 m thick), which is made up of felsic volcano-sedimentary rocks interlayered with carbonates [42]. The Shalahib formation predominantly comprises andesite and rhyolite volcanic rocks and associated pyroclastics and ignimbrites, and the whole sequence is affected by low-temperature greenschist-facies regional metamorphism. The Khnaiguiyah deposit lies within an area of 3 × 3 km. Four mineralized orebodies are interpreted hydrothermal mineral deposits containing Zinc and Copper that are hosted by strongly sheared and folded late Proterozoic medium to felsic volcanics/volcaniclastics. The shear zones, which are tens of meters thick, are oriented NS and dip 10 to 70 to the west. The hydrothermally altered rocks occur within discontinuous anastomosed bands 50 to 100 m wide and several Kilometers long and are regionally oriented along with the north-south regional foliation. Detailed analysis of surface and drill-core samples shows that the hydrothermal alteration zones and associated Zn-Cu-Fe-Mn mineralization are controlled by a shearing deformation phase that post-dated the first phase of regional folding. The hydrothermal alteration zone contains illite, kaolinite, quartz, albite, hematite, and calcite.
At this locality, the Precambrian basement is overlain by the basal conglomerates and cross-bedded red sandstone of the middle Cambrian Saq formation, which is in turn overlain by the Permo-Triassic shallow marine carbonates of the Khuff formation. Khnaiguiyah Zn-Cu deposits represent a substantial zinc resource and attracted considerable exploration efforts in the last three decades [40]. The four ore bodies of the Khnaiguiyah district comprise mineable reserves of up to 11 Mt averaging 7.41% Zn and 0.82% Cu [41]. Khnaiguiyah-type deposit features are consistent with metamorphism and deformation of volcanogenic massive sulfide (VMS) mineralization; the related stratiform Mn-rich units are mainly suggestive of a seafloor hydrothermal setting.
The Khnaiguiyah ores are hosted by the Shalahib Formation (1500 m thick), which is made up of felsic volcano-sedimentary rocks interlayered with carbonates [42]. The Shalahib formation predominantly comprises andesite and rhyolite volcanic rocks and associated pyroclastics and ignimbrites, and the whole sequence is affected by low-temperature greenschist-facies regional metamorphism. The Khnaiguiyah deposit lies within an area of 3 × 3 km. Four mineralized orebodies are interpreted hydrothermal mineral deposits containing Zinc and Copper that are hosted by strongly sheared and folded late Proterozoic medium to felsic volcanics/volcaniclastics. The shear zones, which are tens of meters thick, are oriented NS and dip 10 to 70 to the west. The hydrothermally altered rocks occur within discontinuous anastomosed bands 50 to 100 m wide and several Kilometers long and are regionally oriented along with the north-south regional foliation. Detailed analysis of surface and drill-core samples shows that the hydrothermal alteration zones and associated Zn-Cu-Fe-Mn mineralization are controlled by a shearing deformation phase that post-dated the first phase of regional folding. The hydrothermal alteration zone contains illite, kaolinite, quartz, albite, hematite, and calcite.
At this locality, the Precambrian basement is overlain by the basal conglomerates and cross-bedded red sandstone of the middle Cambrian Saq formation, which is in turn overlain by the Permo-Triassic shallow marine carbonates of the Khuff formation.

Data Used and Methods
This study used visible/infrared satellite-derived imagery to characterize mineralization associated with hydrothermal alteration zones. Landsat 8-OLI, Sentinel-2, and ASTER data ( Figure 2) were all employed to detect altered features and structural patterns. A comparison of these sensors is shown in Figure 2 [19].

Data Used and Methods
This study used visible/infrared satellite-derived imagery to characterize mineralization associated with hydrothermal alteration zones. Landsat 8-OLI, Sentinel-2, and ASTER data ( Figure 2) were all employed to detect altered features and structural patterns. A comparison of these sensors is shown in Figure 2 [19]. On 11 February 2013, the Landsat-8 (OLI) satellite was launched. Landsat-8 scene dimension is 85-km-cross-track-by-180-km-along-track. There are nine VIS/NIR and SWIR ranges reported, as well as two longwave thermal ranges. The pixel size of OLI channels was stated to be 30 m; however, TIRS has a spatial resolution of 100 m. The quantization level is 12-bit data that permits additional bits to be used to acquire optimal data, enabling the assessment of minor surface disturbances. Landsat-OLI scene (path/row 166/43; ID: LC08_L1TP_166043_20211209_20211215_01_T1) that was acquired on 9 December 2021.
NASA and METI (Japan's Ministry of Economic Trade and Industry) deployed AS-TER, an advanced multispectral satellite imaging system, onboard the TERRA spacecraft in December 1999. NASA's Land Processes Distributed Active Archive Centre provided the ASTER data (LP DAAC). ASTER data includes spectral ranges in the visible and nearinfrared (VNIR), shortwave infrared (SWIR), and thermal infrared (TIR): three bands (with 15 m spatial resolution) in the VNIR, six bands (with 30 m spatial resolution) in the SWIR, and five bands (with 90 m spatial resolution) in the TIR (TIR). In this investigation, the ASTER SWIR spectral bands (30 m spatial resolution) are used to measure between ~ 1.60 and 2.45 μm to allow discriminating between Al-OH, Fe, Mg-OH, H-O-H, and CO3 absorption features [43].
Preprocessing of the obtained ASTER scene (ASTER data ID: ASTB061106074035) included cross-talk correction and data orthorectification using the ENVI software application. Using ENVI v.5 software, the Log-residual (LR) technique was used to calibrate, normalize, and decrease noise from sensors and solar illumination in SWIR bands [2,14]. On ASTER data, this approach was used to remove atmospheric and topography impacts. As a result, the data became more reflective of the target area's composition and lithology. It was also possible to compare the retrieved endmembers of SWIR bands to reference spectra from the spectral library of the United States Geological Survey (USGS). This approach was used to reveal minerals using SWIR data [14,44].
On 23 June 2015, the Sentinel-2A satellite was launched, and the first data was taken a few days later. Sentinel-2 sensors gather data in the VIS/NIR, and SWIR, TIR wavelength ranges. These bands have a spatial resolution of 10-60 m. Sentinel-2 captures 13 bands in the VIS/NIR and SWIR spectrum. The VIS/NIR bands: blue B2 (490 nm), green B3 (560 nm), red B4 (665 nm), and infrared B8 (842 nm) have a 10 m pixel size, whilst the coastal On 11 February 2013, the Landsat-8 (OLI) satellite was launched. Landsat-8 scene dimension is 85-km-cross-track-by-180-km-along-track. There are nine VIS/NIR and SWIR ranges reported, as well as two longwave thermal ranges. The pixel size of OLI channels was stated to be 30 m; however, TIRS has a spatial resolution of 100 m. The quantization level is 12-bit data that permits additional bits to be used to acquire optimal data, enabling the assessment of minor surface disturbances. Landsat-OLI scene (path/row 166/43; ID: LC08_L1TP_166043_20211209_20211215_01_T1) that was acquired on 9 December 2021.
NASA and METI (Japan's Ministry of Economic Trade and Industry) deployed ASTER, an advanced multispectral satellite imaging system, onboard the TERRA spacecraft in December 1999. NASA's Land Processes Distributed Active Archive Centre provided the ASTER data (LP DAAC). ASTER data includes spectral ranges in the visible and near-infrared (VNIR), shortwave infrared (SWIR), and thermal infrared (TIR): three bands (with 15 m spatial resolution) in the VNIR, six bands (with 30 m spatial resolution) in the SWIR, and five bands (with 90 m spatial resolution) in the TIR (TIR). In this investigation, the ASTER SWIR spectral bands (30 m spatial resolution) are used to measure betweeñ 1.60 and 2.45 µm to allow discriminating between Al-OH, Fe, Mg-OH, H-O-H, and CO 3 absorption features [43].
Preprocessing of the obtained ASTER scene (ASTER data ID: ASTB061106074035) included cross-talk correction and data orthorectification using the ENVI software application. Using ENVI v.5 software, the Log-residual (LR) technique was used to calibrate, normalize, and decrease noise from sensors and solar illumination in SWIR bands [2,14]. On ASTER data, this approach was used to remove atmospheric and topography impacts. As a result, the data became more reflective of the target area's composition and lithology. It was also possible to compare the retrieved endmembers of SWIR bands to reference spectra from the spectral library of the United States Geological Survey (USGS). This approach was used to reveal minerals using SWIR data [14,44].
On 23 June 2015, the Sentinel-2A satellite was launched, and the first data was taken a few days later. Sentinel-2 sensors gather data in the VIS/NIR, and SWIR, TIR wavelength ranges. These bands have a spatial resolution of 10-60 m. Sentinel-2 captures 13 bands in the VIS/NIR and SWIR spectrum. The VIS/NIR bands: blue B2 (490 nm), green B3 (560 nm), red B4 (665 nm), and infrared B8 (842 nm) have a 10 m pixel size, whilst the coastal band B1 (443 nm) has a 60 m pixel geometry. The pixel sizes of the SWIR bands (B11: 1610 nm, B12: 2190 nm) are both 20 m. Sentinel-2 scene (S2A_MSIL1C_20211221T073331_N0301_R049_T38QNM_20211221T084355) is delivered as zip-compressed files in Sentinel's own SAFE format. The spectral bands are stored as jpg files in this SAFE file in three different geometric resolutions (10 m, 20 m, and 60 m. The jpg files of bands B2, B3, B4, and B8 with a spatial resolution of 10 m, and B11 and B12 with 20 m are stacked into a single GeoTIFF file of a uniform pixel size of 10 m. A subset of these data was conducted during preprocessing using SNAP software in order to minimize the computational time and the data. To analyze multispectral data, several methods have been used, including PCA, the utilization of band ratios, relative absorption band depth (RBD; 24), and mineral indices [11,12,22], as well as spectral analysis. Band ratios have been employed to investigate spectral differences between bands and to reduce topographic impacts [7,8,23,45]. The intensity of the hydrothermal activity can be used to reveal hydrothermal mineral assemblages [13,28]. Sub-pixel spectral classifications can thus be attributed to specific important hydrothermal minerals associated with propylitic (epidote, chlorite, calcite), phyllic (muscovite, sericite, illite), argillic (montmorillonite, kaolinite, dickite), and advanced argillic (alunite-pyrophyllite) alteration zones.
The band ratio is a transformation procedure for enhancing spectral differences in remote sensing data. It works by dividing pixels from one band by pixels from another band [46] and sometimes dividing bands of the numerator or/and denominator after mathematical calculation. The goal of this technique is to reveal the spectral characteristics of material so that variables on Earth's surface can be distinguished better [47]. Band ratios can be used to distinguish between soils, rock types, and land use effects [48][49][50][51]. The ENVI software and ArcGIS software packages v. 10.8 are utilized in the present study. The PCA process has been used to transform a large number of correlated spectral bands into a smaller number of uncorrelated spectral bands, which is a statistical approach used in image transformation. In the mapping of hydrothermal potential alteration zones, the selective "principal components" (PCs) technique has been frequently used [5,15]. Based on the eigenvectors of the selected bands, statistical parameters were examined to determine which PC image could be utilized to emphasize the particular minerals.
Spectral mapping was used to differentiate mixed pixels from unwanted pixels during the mineral extraction process. This enabled the mapping and identification of possible minerals based on the end member's spectral signature in comparison to those in the spectral library [5,14]. The MNF transformation [52] was used to derive a PPI that represented the input image's most spectrally pure pixels. This was utilized to detect endmembers using n-D visualization for mineral identifications based on spectral classifications. Using MNF and PPI, the n-D visualize viewer can locate, characterize, cluster (group), and pick the purest pixels (endmembers) in n-spaces. Each class indicated a mineral with a high absorption capacity.
Digital overlay approaches have been utilized to create integrative maps using Geographic Information System (GIS) technologies [28,29,53]. Predictive maps have also been created using knowledge-driven systems using weighted overlay analysis of ArcGIS that integrate multi-criteria decision-making based on expert judgment [30,54]. Each evidential image was reclassified into five classes using the Natural Breaks method; the class of high intensity of hydrothermal alteration is given "5" and the opposite given "1. As a result, the final prospective map can be created by combining several evidential maps [54].

Lithologic and Structural Characteristics
Goethite, hematite, and jarosite are examples of iron minerals that have diagnostic spectral characteristics near 0.43 m, 0.65 m, 0.85 m, and 0.93 m, which are close to Sentinel-2 band 1, band 4, band 8/8A, and band 9 [55]. Moreover, both hematite and jarosite exhibit reflectance characteristics near 0.72 m and 0.74 m, which are both close to Sentinel-2 band 6. Hematite also displays a distinguishing absorbance pattern at a wavelength of 0.88 m, which corresponds to Sentinel-2A band 8A. Thus, Sentinel-2 band ratios of 6/1, 6/8A, and (6 + 7)/8A were utilized to distinguish hematite + goethite, hematite + jarosite, and a mixture of iron-bearing minerals (see more information on iron mineral spectra in Ge et al. [55]) from felsic or sedimentary deposits in red from basement mafic to intermediate variations in cyan (See Figure 3a) as they contain ferromagnesian minerals [2]. Sentinel-2 ( Figure 3b) uses 11/8A, 3/4, and (6 + 7)/8A of ferric, ferrous, and a combination of iron-containing minerals to indicate likely areas rich in hydrothermal alteration in purple, mafic varieties in yellow, and sedimentary deposits in the northeast in red-orange, but vegetation in green. Images from the Sentinel-2 satellite (Figure 3c) 11/12, 11/8A, and (6 + 7)/8A displayed white-toned patches in HAZs [2]. The study's heights (Figure 3d) vary from 789 to 933 m above sea level, and the majority of the structural patterns are visible at these elevations. reflectance characteristics near 0.72 m and 0.74 m, which are both close to Sentinel-2 band 6. Hematite also displays a distinguishing absorbance pattern at a wavelength of 0.88 m, which corresponds to Sentinel-2A band 8A. Thus, Sentinel-2 band ratios of 6/1, 6/8A, and (6 + 7)/8A were utilized to distinguish hematite + goethite, hematite + jarosite, and a mixture of iron-bearing minerals (see more information on iron mineral spectra in Ge et al. [55]) from felsic or sedimentary deposits in red from basement mafic to intermediate variations in cyan (See Figure 3a) as they contain ferromagnesian minerals [2]. Sentinel-2 (Figure 3b) uses 11/8A, 3/4, and (6 + 7)/8A of ferric, ferrous, and a combination of iron-containing minerals to indicate likely areas rich in hydrothermal alteration in purple, mafic varieties in yellow, and sedimentary deposits in the northeast in red-orange, but vegetation in green. Images from the Sentinel-2 satellite (Figure 3c) 11/12, 11/8A, and (6 + 7)/8A displayed white-toned patches in HAZs [2]. The study's heights (Figure 3d) vary from 789 to 933 m above sea level, and the majority of the structural patterns are visible at these elevations.

Landsat-8
Band ratios are used to enhance hydrothermally altered zones and the oxidation zone that reflects the abundance of certain minerals. Band ratios 6/7, 6/2, and 6/5 * 4/5 [16] were Band ratios are used to enhance hydrothermally altered zones and the oxidation zone that reflects the abundance of certain minerals. Band ratios 6/7, 6/2, and 6/5 * 4/5 [16] were employed to improve the identification of rocks and minerals based on content mineralogy (Figure 4a,b). Band ratio 6/7 is susceptible to OH-bearing minerals, and band ratio 6/2 highlights rocks rich in FeO composition, so mafic igneous rocks have lower reflectance than other igneous rocks; and band 6/5 * 4/5 is useful to distinguish between mafic and non-mafic rocks based on their sensitivity to high Fe-bearing aluminosilicate concentration. Iron-bearing minerals and OH-bearing minerals are abundant in these areas. Sultan et al. [16] demonstrated that rationing in the 6/5 * 4/5 band is possible. Sultan et al. [16] use the sensitivity of the ratio to Fe-bearing aluminosilicates to distinguish mafic rocks (bluish color) from other rocks. Ramadan et al. [56] revealed the potential locations of hydrothermal mineral deposits in 6/7, 6/5, and 5 (Figure 4c,d). Using the band ratios 6/7, 4/2, and 5/6 in R, G, and B [7], felsic rocks are colored green, mafic rocks are colored blue, and areas of extensive hydrothermal alteration are colored light pink, yellow, and light red (Figure 4a,c,e, respectively). The 6/7 ratio emphasizes hydrothermal alteration and surface weathering oxides and hydroxides [23,57,58]. Clay minerals have a high Band 6 reflectance and a strong Band 7 absorption [10,58]. The 4/2 ratio is important for detecting iron oxide-bearing rocks (Figure 4e) due to considerable absorption in Band 2 and reflectance characteristics in Band 4 for iron oxides [7]. The felsic rocks in 4/2 appear in green color.

Sentinel-2
As shown in (Figure 5a), [59] offered three band ratios in R, G, and B: 11/12, 11/8, and 4/2. These band ratios were utilized in this research to distinguish the alteration zones. These ratios are proportional to the occurrence of OH-bearing minerals (11/12), iron oxides (4/2), and the band ratio 6/5, which is utilized to enhance the presence of ferrous oxides. Metavolcanics are colored green in this ratio, indicating high ferrous oxide content. Some wadi deposits have a purple tint due to the occurrence of clay minerals and ironbearing minerals in high concentrations. The locations of probable HAZs are revealed in a yellow color due to their high presence of clay and ferrous oxides. To improve the identification of rock units based on mineral content, ratio bands of 11/12, 11/2, and 11/8 * 4/8 [16] were used (Figure 5b,c). Band ratio 11/12 is amenable to To improve the identification of rock units based on mineral content, ratio bands of 11/12, 11/2, and 11/8 * 4/8 [16] were used (Figure 5b,c). Band ratio 11/12 is amenable to OH-bearing minerals; band ratio 11/2 is associated with the content of opaque minerals (e.g., FeO) in rocks, so mafic rock types have lesser reflectance than some other igneous rocks; and band 11/8 * 4/8 can be used to distinguish between mafic and non-mafic rocks based on their sensitivity to high Fe-bearing aluminosilicate concentration. Iron-bearing minerals and OH-bearing minerals are widespread in such regions.

ASTER
The OH-bearing minerals have reflectance at 1.656 µm (band 4). The argillic minerals (montmorillonite and kaolinite) contain absorption features at 2.205 (band 6), and kaolinite display double-shaped absorption features around 2.165 (band 5) and 2.205 (band 6), in contrast to phyllic minerals (muscovite and illite), which have a single deep absorption feature at 2.205 (band 6) ( Figure 6; from Mars and Rowan [3]). Propylitic minerals with a 2.335 m absorption characteristic. This most likely matches minerals like calcite and chlorite that contain CaCO3 or Mg-OH [3]. OH-bearing minerals; band ratio 11/2 is associated with the content of opaque minerals (e.g., FeO) in rocks, so mafic rock types have lesser reflectance than some other igneous rocks; and band 11/8 * 4/8 can be used to distinguish between mafic and non-mafic rocks based on their sensitivity to high Fe-bearing aluminosilicate concentration. Iron-bearing minerals and OH-bearing minerals are widespread in such regions.

ASTER
The OH-bearing minerals have reflectance at 1.656 μm (band 4). The argillic minerals (montmorillonite and kaolinite) contain absorption features at 2.205 (band 6), and kaolinite display double-shaped absorption features around 2.165 (band 5) and 2.205 (band 6), in contrast to phyllic minerals (muscovite and illite), which have a single deep absorption feature at 2.205 (band 6) ( Figure 6; from Mars and Rowan [3]). Propylitic minerals with a 2.335 m absorption characteristic. This most likely matches minerals like calcite and chlorite that contain CaCO3 or Mg-OH [3]. The ASTER band ratios 4/6, 4/5, and 4/7 boost argillic and sericitic alteration zones, respectively [6]. Furthermore, in these images, the ASTER-4/5 band ratio defines the advanced argillic alteration (e.g., alunite and dickite). As a result, the combination of band ratios 4/6, 4/5, and 4/7 in R, G, and B of ASTER is utilized to depict HAZs. Figure 7 depicts the classification of these combined ratios into five ranks of hydrothermal alteration, with the highest rank (0.60-0.64) in red, denoting locations with both argillic and sericitic regions of alteration. The ASTER band ratios 4/6, 4/5, and 4/7 boost argillic and sericitic alteration zones, respectively [6]. Furthermore, in these images, the ASTER-4/5 band ratio defines the advanced argillic alteration (e.g., alunite and dickite). As a result, the combination of band ratios 4/6, 4/5, and 4/7 in R, G, and B of ASTER is utilized to depict HAZs. Figure 7 depicts the classification of these combined ratios into five ranks of hydrothermal alteration, with the highest rank (0.60-0.64) in red, denoting locations with both argillic and sericitic regions of alteration. Figure 7c,d shows the results of merging band ratio images 4/6, 7/6, and (5 + 7)/6. Regions richer in white mica were identified using the band ratio of 7/6. Using such ratios, locations rich in Al-OH minerals are depicted in a white tone. The largest band ratios of 4/6 and 7/6 refer to the phyllic zone [60], and the values of (5 + 7)/6 are also true. The green-colored area has less hydrothermal alteration than the first, but pink-colored portions have the least amount of HAZs. Band ratios 4/6 in this composite highlight the white tone's alteration zones (Figure 7). Furthermore, band ratios of 7/6 were applied to determine the areas of an abundance of Al-OH minerals as white mica (muscovite) as revealed in white tone, which occupies the middle part of the present study area (Figure 7).   Figure 7c,d shows the results of merging band ratio images 4/6, 7/6, and (5 + 7)/6. Regions richer in white mica were identified using the band ratio of 7/6. Using such ratios, locations rich in Al-OH minerals are depicted in a white tone. The largest band ratios of 4/6 and 7/6 refer to the phyllic zone [60], and the values of (5 + 7)/6 are also true. The greencolored area has less hydrothermal alteration than the first, but pink-colored portions have the least amount of HAZs. Band ratios 4/6 in this composite highlight the white tone's alteration zones (Figure 7). Furthermore, band ratios of 7/6 were applied to determine the areas of an abundance of Al-OH minerals as white mica (muscovite) as revealed in white tone, which occupies the middle part of the present study area (Figure 7).
The contrast between hydrothermally altered zones was emphasized by the combination of band ratios (5 + 7)/6, (4 + 6)/5, and (7 + 9)/8 in R, G, and B, respectively ( Figure  8a). This allowed areas richer in phyllic, argillic [6,61], and propylitic minerals to be detected, respectively. In this combination, yellow areas revealed argillic and phyllic hydrothermal alterations. PCA was used to map areas of argillic hydrothermal alteration using ASTER bands 4, 5, and 6b. Table 1 shows the eigenvector values obtained using the specific bands (B4, B5, and B6) for the PCA method (Table 1)  The contrast between hydrothermally altered zones was emphasized by the combination of band ratios (5 + 7)/6, (4 + 6)/5, and (7 + 9)/8 in R, G, and B, respectively (Figure 8a). This allowed areas richer in phyllic, argillic [6,61], and propylitic minerals to be detected, respectively. In this combination, yellow areas revealed argillic and phyllic hydrothermal alterations. PCA was used to map areas of argillic hydrothermal alteration using ASTER bands 4, 5, and 6b. Table 1 shows the eigenvector values obtained using the specific bands (B4, B5, and B6) for the PCA method (Table 1). PC2 indicates a negative loading of band 4 (−0.780) and a positive loading of bands 5 (0.605) and 6 (= 0.157) according to eigenvector loadings. The locations of hydrothermal alteration are revealed in a white tone when negated (multiplied by −1) PC2 is displayed in greyscale (Figure 8b).   The band ratios OHI, KAI, and ALI were integrated into R, G, and B to distinguish between potential sites of argillic and phyllic alteration. The areas of increased hydrothermal alteration are congruent with structural features associated with granitic rocks, according to the classifications of these combined values. Each fraction's greatest value is denoted by a white tone. The minerals indices OHI, KLI, and ALI were displayed in R, G, and B, respectively (Figure 8c (Figure 6e). A two-dimensional (2D) scatter plot of band ratios (5 + 7)/6 (Al-OH content) vs. 5/7 (Al-OH composition) was used to explore ASTER Al-OH composition [62]. The diagram's extreme far bottom right side indicated regions with no Al-OH minerals, whereas the extreme top left side of the diagram revealed places with no Al-OH minerals. Cudahy et al. [62] found that plotting band ratios (5 + 7)/6 (Al-OH content) vs. 5/7 (Al-OH composition) clearly separated areas rich in Al-OH from those with no Al-OH, confirming prior findings. Higher values, indicating higher Al-OH concentrations, are highlighted in red and correspond to locations of significant hydrothermal alteration. As illustrated by the green in Figure 8e, this area was clearly delimited by lower Al-OH concentration and higher Mg-OH content.

XRD Analysis of Hydrothermal Alteration Zones
Samples were taken from the two main alteration zones (Figure 9), and 21 representative specimens were selected for XRD analysis at the National Research Center (Egypt). The results of the analyses revealed the presence of silica minerals (mostly quartz), gypsum, anhydrite, kaolinite, illite, clinochlore, and hematite, with a small percentage of microcline, calcite, and halite (Tables 2 and 3). Clay minerals (kaolinite, illite, sericite) in these zones are mainly the products of the decomposition of plagioclase feldspar, and the presence of ferric iron oxides and hydroxides (hematite, goethite) is related to the weathering of ferromagnesian minerals (mostly hornblende and clinopyroxene). The abundance of sulfates in the form of gypsum and anhydrite is highly indicative of the former presence of disseminated sulfide phases, which is compatible with these zones being either mature gossans or conduits within the feeder zones beneath massive sulfide mounds. crocline, calcite, and halite (Tables 2 and 3). Clay minerals (kaolinite, illite, sericite) in these zones are mainly the products of the decomposition of plagioclase feldspar, and the presence of ferric iron oxides and hydroxides (hematite, goethite) is related to the weathering of ferromagnesian minerals (mostly hornblende and clinopyroxene). The abundance of sulfates in the form of gypsum and anhydrite is highly indicative of the former presence of disseminated sulfide phases, which is compatible with these zones being either mature gossans or conduits within the feeder zones beneath massive sulfide mounds.  Table 2. B shows the samples location for Table 3.  Table 2. B shows the samples location for Table 3. Table 2. The results of XRD analysis; see Figure 9b for the location map.  Table 3. The results of XRD analysis; see Figure 8b for the location map.

Mineral Potential Map
The final potential location of mineralization was created by combining multi-criteria data. This allows for revealing the prospective areas of hydrothermal mineralization associated with hydrothermal alteration zones; thus, we use a variety of ways to emphasize alteration zones. The results of remote sensing analysis data were integrated using GIS approaches (band ratio, PC, mineral indices). These images were quantified and divided into various zones (different probability values). To obtain the prospective or promising map of mineral exploration, a succession of evidential maps is used. The approach of merging data in a GIS enabled the promotion and identification of the best exploration and mining locations. This aided exploration and the prediction of new mineralized zones. The recent innovative procedures that use recent digital technologies and creative geoinformation approaches have allowed for the detection of the optimal mineral resource area. Multiple datasets have been aggregated and integrated since the birth of the GIS to locate new mineralized zones [63] reliably.
Using a GIS-based overlay method, the likely locations of hydrothermal mineral deposits were revealed by combining several evidence hydrothermal alteration maps ( Figure 9). The mineral prospective map is divided into six groups (Figure 9a) based on their prospective for hydrothermal alteration amplitude: very low, low, moderate, good, very good, and excellent, and covers 31. 36, 28.22, 20.49, 10.99, 6.35, and 2.59 % percent of the research region, respectively. The red color indicates the most promising mineral deposit zone. The resulting map ( Figure 9) demonstrates a pattern of coherence in the hydrothermal-ore deposits found in the area's mines. Many sections of the potentially high zone, meanwhile, were restricted to wadi deposits and areas of sedimentary cover in the northeast of the research area.

Discussion
The use of various methodologies for three different sensors, including OLI, Sentinel-2, and ASTER data, clearly shows that hydrothermal alteration processes dominate the examined area. This is because identifying HAZs through fracture/fault zones is required when exploring mineral deposits that originated from hydrothermal processes [2,5,8]. As a result, the severity of the alteration can help determine where the ore body is located.
The relative band depth (5 + 7)/6 was efficiently adopted (Figures 7c and 8a) for excellent detection of Al-smectite, muscovite, sericite, and illite [2,62,65], and Al/Fe-OH minerals, such as muscovite, kaolinite, and jarosite [66]. Moreover, ASTER bands 5 + 7/6 'Al-OH content' vs. bands 5/7 (outside absorption) 'Al-OH composition' [62]. This diagram (Figure 8c) shows that the selected red pixels are rich in AL-OH, but the lowest ones are in green. The AL-OH area rich in minerals that are marked in red is consistent with the areas of high hydrothermal alteration.
Using GIS-based weighted overlay analysis to confirm the findings of band ratios and mineral indices acquired from Landsat-OLI, Sentinel-2, and ASTER data that revealed ironcontaining and Al-OH-carrying minerals revealed useful information regarding places rich in gossans. Such gossans that consist of limonite, goethite, hematite, malachite, and azurite reveal the existence of massive sulfide; porphyry and skarn deposits [64] are consistent with areas of high hydrothermal alteration intensity.

Conclusions
The Khnaiguiyah area, Saudi Arabia, is tested to delineate the area of probable mineral resources. The ability of multispectral remote sensing data to detect and characterize the hydrothermal alteration minerals is significant for mineral exploration. The present study used ASTER, Sentinel-2, and Landsat-OLI to identify potential areas of HAZs. The HAZs generated from these various multispectral sensors were combined through GIS to highlight the potential areas of HAZs. The highest grade of HAZs, which covers about 2.59 %, is compatible with areas of significant hydrothermal changes and has been verified with areas of gossans that revealed the presence of sulfide minerals.