Geochemistry of Red Soils in the Kas District of Antalya (Türkiye) Using Multivariate Statistical Approaches and GIS

: Red soil is formed by weathering due to the effect of bedrock and various ﬂuids in subtropical and tropical climate zones. The major and trace element geochemistry of red soil are important for understanding its environmental conditions and interpreting its genesis. This study aims to determine how red soil is formed in the study area, where magmatic rocks are located, and to determine their relationship with magmatic ﬂuids. In this context, geochemical analyses were conducted, and index values, spatial distributions, and multivariate statistics of the samples were calculated. Methods: A total of 49 red soil samples collected from the Sutlegen Village in the Kas district of Antalya were subjected to geochemical analysis. The following major elements were found in the red soil samples: Al 2 O 3 , SiO 2 , Fe 2 O 3 , TiO 2 , CaO, MgO, K 2 O, P 2 O 5 , and SO 3 . The Al 2 O 3 –SiO 2 – Fe 2 O 3 triangular diagram revealed that the samples underwent moderate laterization. Multivariate statistical analyses indicated that major element oxides of Al 2 O 3 , Fe 2 O 3 , and SiO 2 behaved differently from other components and revealed that red soil formations were associated with magmatic ﬂuids. The region is a potential area for economic bauxite resources. Detailed investigations should be conducted on mineral exploration for bauxite ore.


Introduction
About 15% (22 million km 2 ) of the continental crust of the earth consists of karstic lands [1,2].Reddish soils with a high clay content that have been usually formed in karstic environments are called red soil.Red soil is common in Southeast Asia, Mediterranean coasts, and North America [2].In Türkiye, red soil formations are observed in the Mediterranean, Aegean, and Marmara regions [3][4][5][6].
Several studies in the literature have examined the geochemical formations and model characterizations of red soil.The heavy metal enrichment in the red soil in western Sicily was associated with mineralogical compositions such as hematite, goethite, and magnetite [7].Hematite and lithiophorite minerals observed in the red soil samples collected from Eastern Piedmont in the US caused the enrichment of Fe 2 O 3 and MnO and their accumulation in the environment [8].It was determined that the red soil in the Guizhou province (China) was mainly composed of SiO 2 , Al 2 O 3 , and Fe 2 O 3 [1].The high Fe 2 O 3 concentration observed in the red soil in Saudi Arabia was associated with goethite and hematite minerals [9].The major element contents of the red soil in Jordan were found to be SiO 2 , Al 2 O 3 , TiO 2 , Fe 2 O 3 , and K 2 O [10].The genesis of the red soil in the Kucukkoras region in the Karaman province of Türkiye was attributed to the carbonate rocks (recrystallized limestone, cherty limestone, oolitic limestone), and the major elemental content of the red soil was found to contain Al, Ti, and Si while its trace elemental content was found to contain Ni, Mn, and Cr [4].The major elemental content of the red soil samples from the Ibradi region in Antalya (Türkiye) was found to contain SiO 2 , Al 2 O 3 , MgO, CaO, MnO, and Fe 2 O 3 , while the trace elemental content was found to contain V, Cr, Ni, Cu, Zn, Ga, Rb, Sr, Y, Zr, Nb, Mo, Ce, and Pb [6].It was found that the red soil of Ireland was composed of kaolinite and sepiolite minerals with rich silica content, having a lattice structure, and that their genesis was attributed to the basaltic rocks [11].The trace elements of Cr, As, Cd, Zn, As, and Bi were found to be enriched in the red soil associated with dolomitic parent rock on the Yunnan-Guizhou Plateau [12].
The genesis of red soil formations is controversial.The studies on the genesis of red soil formations have suggested three theories.The first theory states that autochthonous red soil was formed by the weathering of the carbonated or non-carbonated rocks in the region under acidic environmental conditions [13].
The study area, Sutlegen Village, is located within the borders of the Kas district of Antalya province (Figure 1).It has a boundary with the Demre district on the east and with the Elmali district on the north.It is located approximately 30 km from the harbor of Kas.In the literature, there are several studies on the geology of the region.However, no study has been conducted on the interpretation of the genesis and chemical properties of the red soil in the region by using various methods.In this context, the present study aims to obtain the chemical properties of the red soil formed in the region.Using the results of the chemical analysis, the geostatistical analyses, the spatial distributions, and the characterization of the formation model were conducted.
Minerals 2023, 12, x 2 of 31 from the Ibradi region in Antalya (Türkiye) was found to contain SiO2, Al2O3, MgO, CaO, MnO, and Fe2O3, while the trace elemental content was found to contain V, Cr, Ni, Cu, Zn, Ga, Rb, Sr, Y, Zr, Nb, Mo, Ce, and Pb [6].It was found that the red soil of Ireland was composed of kaolinite and sepiolite minerals with rich silica content, having a lattice structure, and that their genesis was attributed to the basaltic rocks [11].The trace elements of Cr, As, Cd, Zn, As, and Bi were found to be enriched in the red soil associated with dolomitic parent rock on the Yunnan-Guizhou Plateau [12].The genesis of red soil formations is controversial.The studies on the genesis of red soil formations have suggested three theories.The first theory states that autochthonous red soil was formed by the weathering of the carbonated or non-carbonated rocks in the region under acidic environmental conditions [13].
The study area, Sutlegen Village, is located within the borders of the Kas district of Antalya province (Figure 1).It has a boundary with the Demre district on the east and with the Elmali district on the north.It is located approximately 30 km from the harbor of Kas.In the literature, there are several studies on the geology of the region.However, no study has been conducted on the interpretation of the genesis and chemical properties of the red soil in the region by using various methods.In this context, the present study aims to obtain the chemical properties of the red soil formed in the region.Using the results of the chemical analysis, the geostatistical analyses, the spatial distributions, and the characterization of the formation model were conducted.

Study Area and Regional Geology
The study area in the Sutlegen Village is located in the Kas district of Antalya province, and it is bounded by the Katran Mountain in the north and Caybogazi Dam in the northeast.The study area is covered by the geologic map sheets of FETHIYE-P 23-a1 and FETHIYE-P 23-a2 with a scale of 1/25,000, produced by the General Directorate of Mineral Research and Exploration (MTA).The lithological units observed in the study area are mostly karstifiable carbonate rocks (Figure 2) [22].The lithological and stratigraphic characteristics of these rocks are as follows: (1) Beydaglari Formation (Kb): the formation, which is predominant in the study area, consists of karstifiable Liassic-Upper Cretaceous aged neritic limestones.(2) Gomuce Member (Tmsg): the unit, which consists of Burdigalian aged limestones, forms the bottom border of the Sinekci Formation.(3) Kibrisdere Member (Tmsk): the unit, which consists of Burdigalian aged clayey limestones, has fossil content.(4) Felenkdagi Conglomerate (Tmf): the unit, which was found to be Upper Langhian-Serravalian (?) aged, consists of gravel and conglomerates of various sizes and shapes.The unit also includes sandy limestone, sandstone, and mudstone.(5) Kasaba Formation (Tmka): the formation, which consists of different sedimentary intercalations of Upper Burdigalian-Lower Langhian aged units, consists of sandstone, conglomerate, claystone, and siltstone [22].(6) Caybogazi Member (Tmsc): the unit, which is covered by the Sinekci formation, is characterized by the Burdigalian aged clay stones.( 7) Susuzdag Formation (Tes): the unit, which was found to be Upper Lutetian-Priabonian aged, consists of limestones and recrystallized limestones with nummulite fossil content [22].The units that have the latest formation in the study area and that are not suitable for karstification are Quaternary-aged alluvium (Qal) and slope debris (Qym) [23].

Study Area and Regional Geology
The study area in the Sutlegen Village is located in the Kas district of Antalya province, and it is bounded by the Katran Mountain in the north and Caybogazi Dam in the northeast.The study area is covered by the geologic map sheets of FETHIYE-P 23-a1 and FETHIYE-P 23-a2 with a scale of 1/25,000, produced by the General Directorate of Mineral Research and Exploration (MTA).The lithological units observed in the study area are mostly karstifiable carbonate rocks (Figure 2) [22].The lithological and stratigraphic characteristics of these rocks are as follows: (1) Beydaglari Formation (Kb): the formation, which is predominant in the study area, consists of karstifiable Liassic-Upper Cretaceous aged neritic limestones.(2) Gomuce Member (Tmsg): the unit, which consists of Burdigalian aged limestones, forms the bottom border of the Sinekci Formation.(3) Kibrisdere Member (Tmsk): the unit, which consists of Burdigalian aged clayey limestones, has fossil content.(4) Felenkdagi Conglomerate (Tmf): the unit, which was found to be Upper Langhian-Serravalian (?) aged, consists of gravel and conglomerates of various sizes and shapes.The unit also includes sandy limestone, sandstone, and mudstone.( 5) Kasaba Formation (Tmka): the formation, which consists of different sedimentary intercalations of Upper Burdigalian-Lower Langhian aged units, consists of sandstone, conglomerate, claystone, and siltstone [22].(6) Caybogazi Member (Tmsc): the unit, which is covered by the Sinekci formation, is characterized by the Burdigalian aged clay stones.( 7) Susuzdag Formation (Tes): the unit, which was found to be Upper Lutetian-Priabonian aged, consists of limestones and recrystallized limestones with nummulite fossil content [22].The units that have the latest formation in the study area and that are not suitable for karstification are Quaternary-aged alluvium (Qal) and slope debris (Qym) [23].Regional geologic map of the study area (modified after in [22,23]).

Geochemical Analyses
The red soils formed due to the bauxite deposit in the study area are located close to the surface.The selected samples were collected from the surface outcrop considering the regional geology of the study area.A total of 49 red soil samples related to magmatic fluids Regional geologic map of the study area (modified after in [22,23]).

Geochemical Analyses
The red soils formed due to the bauxite deposit in the study area are located close to the surface.The selected samples were collected from the surface outcrop considering the regional geology of the study area.A total of 49 red soil samples related to magmatic fluids were collected from the Sutlegen Village, which is located in the Kas district of Antalya province (Türkiye).The collected samples were analyzed by the Mineral Deposits Laboratory in the Department of Geological Engineering at the Faculty of Engineering of Akdeniz University.The samples, which were taken out of plastic bags, were homogenized and ground to a size of <10 µm using the RM 200 mortar grinder (Retsch GmbH, Germany).Then, press-pastilles were produced using the press machine under a pressure of 30 MPa to have suitable dimensions, sizes, and smooth surfaces for the X-ray fluorescence spectrometry analysis [24].

Statistical Analyses
The obtained results were evaluated using the analytical method.The rare earth, trace, and major oxide element concentrations were assessed and interpreted using multivariate statistical methods.The following multivariate statistical analyses were considered appropriate for the study: correlation analysis, simple linear regression, factor analysis, and cluster analysis.The SPSS 23 statistical software was used to obtain these statistics.

Paleoclimatic Characteristics of the Red Soil Samples
The red soil formations present information on paleoenvironmental and paleoclimatic changes [2,12,25].In this context, index values calculated using the results of the geochemical analysis of the red soil samples are significant for the interpretation of the formation processes and examining the genesis.Therefore, numerous indices were computed.During the study, the Chemical Index of Alteration (CIA), Chemical Index of Weathering (CIW), Plagioclase Index of Alteration (PIA), and Silica to Alumina Ratio/Ruxton Ratio (R) were computed.
In order to understand the physical and chemical weathering conditions of the red soil that has been formed through the formation processes, the chemical index of alteration (CIA) is computed.The cations of Na + , K 2+ , and Ca 2+ in minerals such as feldspar, which is an important part of the weathering processes, tend to diminish in the environment [26,27].The CIA index is calculated using the values of the major element oxides of Al 2 O 3 , Na 2 O, K 2 O, and CaO (Equations ( 1) and ( 2)) to understand the chemical weathering processes [28].
CaO * = ml CaO − 10 3 Equations ( 3) and (4) are used to compute the chemical index of weathering (CIW) of the red soil samples to obtain information about weathering processes and the paleoclimatic behavior of the environment [29].
CaO * = ml CaO − 10 3 × mol P 2 O 5 (4) The plagioclase index of alteration (PIA) provides information about the mineral phases in the red soil samples, the source rock that played a role in the formation, and the paleoclimatic characteristics of the environment.The red soil is formed in clayey environments, where chemical weathering is at the highest levels, in the subtropical and tropical climatic zones (monsoon climate zones).The PIA index, which is computed for the environmental conditions of the clay, provides information about the extent to which the plagioclase mineral turns into clay minerals [26,30] (Equation ( 5)) [26].
The silica to alumina ratio/Ruxton ratio (R) presents information about the correlation of silica, which is one of the elemental compositions that is important in chemical weathering processes, with other elemental contents [31].The silica ratio, which decreases in chemical weathering processes, changes in direct proportion to the Ruxton ratio.The Ruxton ratio is expected to be >10 in rocks that have not undergone chemical weathering processes [30] (Equation ( 6)) [31].

Spatial Statistics
In the study, the locations of all collected samples were recorded using a Global Navigation Satellite System (Garmin Montana 680).The coordinates of all samples were collected in geographical coordinates (latitude and longitude) in the WGS84 datum.Then, all recorded samples and their corresponding measurements were combined into a dataset to obtain the spatial distribution of the samples.Next, this dataset was converted to vector points using ArcGIS 9 software.Finally, the spatial distribution of all samples and their corresponding measurements and computations were mapped using the Kriging interpolation method.In this process, the ordinary Kriging method with a spherical semivariogram model was used to generate the spatial distribution of collected samples [32].The distribution maps of the compounds of Al 2 O 3 , SiO 2 , Fe 2 O 3 , TiO 2 , CaO, MgO, and K 2 O, and elements of Mn and Rb, as well as the weathering indices of CIA, CIW, and PIA, were generated in raster format using the CoKriging interpolation method [33].In CoKriging interpolation, simple and prediction methods were used to generate the spatial distribution of multiple input data sets.

Results
The laterization, which plays an active role in the formation of red soil, begins with the depletion of the aluminosilicate minerals and alkaline group elements in the environment under low pH conditions [34][35][36].The degree of laterization reactions taking place under superficial conditions was determined by plotting the triangular diagram (modified after [37]), using the major elements of Al 2 O 3 , SiO 2 , and Fe 2 O 3 (Figure 3).
The triangular diagram of Al 2 O 3 -SiO 2 -Fe 2 O 3 was plotted to determine the degree of laterization of the red soil samples.Sample 30 and Sample 40 were found to have undergone a poor laterization, and Sample 19, Sample 20, and Sample 37 were found to have undergone a strong laterization, while all other red soil samples were found to have undergone moderate laterization.While Sample 30 and Sample 40 were found to have the lowest Al 2 O 3 concentration (45,000 ppm and 328,000 ppm), Sample 20 had the highest Al 2 O 3 concentration (538,000 ppm).In this context, a strong correlation was found between the degree of laterization and the Al 2 O 3 concentration.The heat map revealed that the increase in the degree of laterization in the region led to Al 2 O 3 , TiO 2 , and CaO enrichment, and SiO 2 and P 2 O 5 depletion in the environment.The laterization, which plays an active role in the formation of red soil, begins with the depletion of the aluminosilicate minerals and alkaline group elements in the environment under low pH conditions [34][35][36].The degree of laterization reactions taking place under superficial conditions was determined by plotting the triangular diagram (modified after [37]), using the major elements of Al2O3, SiO2, and Fe2O3 (Figure 3).The triangular diagram of Al2O3-SiO2-Fe2O3 was plotted to determine the degree of laterization of the red soil samples.Sample 30 and Sample 40 were found to have undergone a poor laterization, and Sample 19, Sample 20, and Sample 37 were found to have undergone a strong laterization, while all other red soil samples were found to have undergone moderate laterization.While Sample 30 and Sample 40 were found to have the lowest Al2O3 concentration (45,000 ppm and 328,000 ppm), Sample 20 had the highest Al2O3 concentration (538,000 ppm).In this context, a strong correlation was found between the degree of laterization and the Al2O3 concentration.The heat map revealed that the Table 1 presents the coordinates, which were prepared in a format suitable for the software running the geographical information system tools, the codes of the samples, and the results of the elemental analysis in ppm.Table 2 presents the chemical alteration index values of the red soil samples.
The results of spatial distribution were acquired from the Kriging interpolation technique utilizing the coordinates of the chemical analysis, their measurements, and computed indices:

Statistical Analyses 4.2.1. Descriptive Statistics
The results of the chemical analysis of the red soil samples, and the mean values ± standard errors of the major elements are listed in descending order as follows: Al 2 O 3 (45.40

Correlation Analysis
Red soil samples, which are formed in nature depending on environmental conditions and geological environment characteristics, basically consist of a combination of elemental compositions.Red soil samples were formed with the increase in major and trace elements such as Al 2 O 3 , Mn, Cr, Y, Ce, and Rb, as the concentrations of oxide compounds such as SiO 2 , CaO, SO 3 , K 2 O, P 2 O 5 , MgO, which are among the rock-forming components, decreased due to geological environment properties and their concentrations.An in-depth investigation of these elemental relationships is important for the determination of the genetic properties of the red soils in the region.Elemental relationships were extracted by Pearson correlation analysis.Elements that are enriched together in the formation environment and depleted in the environment with negative and positive correlations have been learned.It can be assumed that the data used in the study, of about 49 red soil samples, are normally distributed according to the central limit theorem [39][40][41][42][43][44][45][46][47][48][49][50][51].A correlation table was prepared using the equation of Pearson's correlation coefficient for the data assumed to be normally distributed (Table 3).The results of Pearson's correlation analysis are summarized as follows.There is a very strong positive correlation between K 2 O and MgO, which are the major elemental contents of the red soil samples (0.922 **).In addition, K 2 O and MgO have a moderate positive correlation with Br, which is one of the trace elements (0.790 **, 0.722 **).Sr, which is one of the trace elements, has a moderate positive correlation with P 2 O 5 , which is one of the major elements, and Y, which is one of the trace elements, (0.745 **, 0.773 **).There is a moderate positive correlation between Sn and Ga, which are trace elements, (0.707 **).There is a strong positive correlation between Ce and Cr, which are trace elements, (0.819 **).While Pb, which is a trace element, has a moderate positive correlation with Cr and Ce, which are also trace elements, and it has a strong positive correlation with Nb (0.765 **, 0.866 **, 0.780 **).There is a strong negative correlation between R (Ruxton ratio), which is one of the chemical weathering indices, and the major elemental content of Al 2 O 3 −0.888**).There is a strong positive correlation between the plagioclase index of alteration (PIA), which is one of the chemical weathering indices, and the major elemental content of CaO (0.946 **).There is a strong positive correlation between the chemical index of weathering (CIW) and the chemical index of alteration (CIA) (0.924 **).

Regression Analysis
The strong positive and negative correlations were revealed by the correlation analysis.Some correlations were also determined between the results of the elemental analysis of the red soil samples and the index values, which revealed the paleoclimatic characteristics.Therefore, the statistical explanation of the aforementioned correlations was also tested by using simple linear regression analysis.The scatter diagrams of the correlations were plotted under three models (Figure 7): (R-Al 2 O 3 ), (PIA-CaO), and (CIW-CIA).Regarding the chemical formation models of the red soil samples, R 2 (coefficient of determination), statistical errors, and coefficient predictions are given in Tables 4-6.

Regression Analysis
The strong positive and negative correlations were revealed by the correlation analysis.Some correlations were also determined between the results of the elemental analysis of the red soil samples and the index values, which revealed the paleoclimatic characteristics.Therefore, the statistical explanation of the aforementioned correlations was also tested by using simple linear regression analysis.The scatter diagrams of the correlations were plotted under three models (Figure 7): (R-Al2O3), (PIA-CaO), and (CIW-CIA).Regarding the chemical formation models of the red soil samples, R 2 (coefficient of determination), statistical errors, and coefficient predictions are given in Tables 4-6.Model 1 gives a linear negative correlation between Al 2 O 3 -R, while Model 2 gives a linear positive correlation between PIA-CaO, and Model 3 gives a linear positive correlation between CIA-CIW.The R 2 value of Model 1 is 0.78, the p-value of the model is 0, and the p-value of the b 0 is lower than 0.05.The R 2 value of Model 2 is 0.89, the p-value of the model is 0, and the p-value of the b 0 is lower than 0.05.The R 2 value of Model 3 is 0.85, the p-value of the model is 0, and the p-value of the b 0 is lower than 0.05.The explained variances in the mathematical models proposed for red soil formations are high.The probability value obtained from the ANOVA analysis of each model is 0. The parameter estimates of the mathematical equations designed for each model are below 0.05 [52][53][54][55].These models explain the chemical formation model at the significance level of 0.05.

Factor Analysis
Factor analysis was conducted to determine the variances in the correlations that were found to be statistically significant in the correlation table [56][57][58][59][60][61][62].The Kaiser-Meyer-Olkin test was conducted to evaluate the suitability of the data for factor analysis [63][64][65][66] (Table 7).The data were found to be suitable for factor analysis, and the sampling adequacy measure was found to be greater than 0.5 [67][68][69][70][71][72][73].The number of factors under which the data were collected was interpreted using the scree plot (Figure 8).The variances of these factors are given in Table 8.The scree plot reveals that the slope of the plot begins to disappear between three.According to the table that presents the variances in the factors, the data plained under two factors, with an eigenvalue greater than one.While Factor 1 54.32% of the total variance, Factor 1 and Factor 2 cumulatively explain 89.48% of variance.The cumulative variances in the geochemical data and indices were ca Principal component analysis (PCA) was conducted to learn the components o tained factors.The components and variances explained by each factor were foun lows: The geochemical compounds and the indices used for paleoclimatic interp that explained Factor 1 were found to be MgO, K2O, Rb, and PIA, and the explain ance was found to be 54.32%.The geochemical compounds and the indices used fo climatic interpretations that explained Factor 2 were found to be CIA, CIW, and the explained variance was found to be 35.16%(Table 9).As can be clearly see principal components' matrix, the major element of Al2O3 did not correlate with an (Figure 9).The scree plot reveals that the slope of the plot begins to disappear between two and three.According to the table that presents the variances in the factors, the data are explained under two factors, with an eigenvalue greater than one.While Factor 1 explains 54.32% of the total variance, Factor 1 and Factor 2 cumulatively explain 89.48% of the total variance.The cumulative variances in the geochemical data and indices were calculated.Principal component analysis (PCA) was conducted to learn the components of the obtained factors.The components and variances explained by each factor were found as follows: The geochemical compounds and the indices used for paleoclimatic interpretations that explained Factor 1 were found to be MgO, K 2 O, Rb, and PIA, and the explained variance was found to be 54.32%.The geochemical compounds and the indices used for paleoclimatic interpretations that explained Factor 2 were found to be CIA, CIW, and R, and the explained variance was found to be 35.16%(Table 9).As can be clearly seen in the principal components' matrix, the major element of Al 2 O 3 did not correlate with any factor (Figure 9).

Cluster Analysis
The red soil samples were classified according to their geochemical and paleoclimatic index properties by using hierarchical clustering analysis (Figure 10).The heat map method was used to understand the distribution of nominal variables with numerical variables [51].The distribution of the geochemical and paleoclimatic index properties of the red soil samples at different laterization degrees was obtained using the heat map (Figure 11).
The hierarchical clustering analysis revealed that the red soil samples were classified into three groups according to the geochemical and paleoclimatic index data.The major oxide elements of TiO 2 , CaO, MgO, K 2 O, P 2 O 5 , and SO 3 , the trace elements of Mn, Cr, Co, V, Sr, Ni, Zn, Ce, Y, La, Cu, Nb, Ba, Pb, Cl, As, Th, Ga, Hf, Sn, Rb, W, U, Au, Pt, Br, Sb, Ta, Hg, Cd, and Te, and the indices of PIA, CIA, CIW, and R constituted Group 1.The major oxide elements of Fe 2 O 3 and SiO 2 constituted Group 2. Al 2 O 3 , a major element, which behaved differently from other elements and weathering indices, constituted Group 3.   The hierarchical clustering analysis revealed that the red soil samples were classified into three groups according to the geochemical and paleoclimatic index data.The major oxide elements of TiO2, CaO, MgO, K2O, P2O5, and SO3, the trace elements of Mn, Cr, Co, V, Sr, Ni, Zn, Ce, Y, La, Cu, Nb, Ba, Pb, Cl, As, Th, Ga, Hf, Sn, Rb, W, U, Au, Pt, Br, Sb, Ta, Hg, Cd, and Te, and the indices of PIA, CIA, CIW, and R constituted Group 1.The major oxide elements of Fe2O3 and SiO2 constituted Group 2. Al2O3, a major element, which behaved differently from other elements and weathering indices, constituted Group 3.
The red soil that has undergone a weak laterization reaction has lower Al2O3, TiO2, and CaO contents.The red soil that has undergone a moderate laterization reaction has lower geochemical contents of TiO2 and Mn, and index values of PIA and R. The red soil that has undergone a high level of laterization reaction has lower geochemical contents of CaO, TiO2, and SiO2, and index values of PIA and R.

Discussion
Due to the high topographical variation in the land and the difficulty of access in the forest area, the sampling procedures were obtained from the best possible locations, and in a way that best represents the land.Comparisons of sample results in the international literature are briefly summarized.According to the results of the chemical analysis of the red soil samples collected from western Sicily, the major elemental contents were found to be Al 2 O 3 , Fe 2 O 3 , and TiO 2 , while the trace elemental contents were found to be Cu, Cr, Cd, Pb, Mn, Zn, and Ni.The values of these elements were observed in the following ranges: Cd, 0.85-1.9ppm; Cu, 30-37 ppm; Cr, 94-129 ppm; Mn, 86-2222 ppm; Ni, 49-68 ppm; Pb, 32-43 ppm; Zn, 109-166 ppm [8].The ranges of the above-mentioned elements observed in the red soil samples collected from the Sutlegen area are as follows: Cd, 0-56.2 ppm; Cu, 81.8-268 ppm; Cr, 335-892 ppm; Mn, 449-3730 ppm; Ni, 0-472 ppm; Pb, 59.6-133 ppm; Zn, 198-396 ppm.According to these values, the trace element concentrations of western Sicily red soil samples were found to be lower than those of the Sutlegen red soil samples.
The major elements of the red soil samples collected from the Guizhou Plateau in southern China included Al 2 O 3 , SiO 2 , Fe 2 O 3 , CaO, MgO, K 2 O, and Al 2 O 3 , and the trace elements included Sr, Y, Zr, Nb, Ba, Zn, Ni, Cu, and V.The chemical indices of alteration (CIA) values calculated using the major element contents of the red soil samples were found to range between 72 and 89 [1].According to the results of the geochemical analysis conducted for the red soil samples collected from Seydisehir in Konya (Türkiye), the CIA values were found to range between 85 and 97.According to the geochemical analysis results of Seydisehir red soil samples, the CIA values ranged between 90.30 and 100.On the other hand, the values of the chemical index of alteration (CIA) calculated for the Sutlegen red soil samples were found to range between 90.3 and 100.
The red soil samples collected from the Jubaila formation in Saudi Arabia were subjected to chemical analysis.According to the results of the chemical analysis, the major element contents were found to be SiO 2 , Al 2 O 3 , Fe 2 O 3 , CaO, MgO, Na 2 O, K 2 O, TiO 2 , MnO, and P 2 O 5 .The chemical index of alteration (CIA) values, calculated using the values of the major elements, were found to range between 25 and 84 [10].The CIA values calculated for the Sutlegen red soil samples were found to be higher than the CIA values calculated for the red soil samples collected from the Jubaila formation in Saudi Arabia.
According to the results of the chemical analysis of the red soil samples collected from the Chinese Tibetan Plateau, the concentrations of the major elements were found in the following order: SiO 2 > CaO > Fe 2 O 3 > Al 2 O 3 > K 2 O > MgO > TiO 2 > Na 2 O > P 2 O 5 > MnO [20].On the other hand, the concentrations of the major elements in the Sutlegen red soil samples were found in the following order: Al 2 O 3 > SiO 2 > Fe 2 O 3 > TiO 2 > CaO > MgO > K 2 O > P 2 O 5 > SO 3 .
The concentrations of the significant major and trace elements in the red soil samples collected from the Karaman province (Türkiye) were found to be in the following ranges: Si (90,735.05-211,715.1 ppm), Al (115,689.1-228,202.7 ppm), Fe, (69,663.87-118,134.8ppm), Mn (1161.69-1548.91ppm), Cr (166.4-254.8ppm), and Co (26.8-85.9)[4, [40][41][42][43].Considering the ranges of the above-mentioned major and trace element concentrations in the Sutlegen red soil samples, the concentrations of the Si and Mn elements were observed to be lower while the concentrations of the Al, Fe, Cr, and Co elements were observed to be higher.
In a study conducted on the red soil samples, which were lateritic weathering products, from Southwestern China, the CIA values were calculated using the results of the chemical analysis.The CIA values were found to range between 0.11 and 94.11 [44].The range of the CIA values of the Sutlegen red soil samples was found to be higher than that of the red soil samples from Southwestern China.Higher CIA values are associated with the intensity of the chemical weathering process.Therefore, the primary minerals in the soil turn into secondary minerals because of chemical alteration [28].The kaolinite samples, which consist of clay minerals, have a CIA value of 100.The closer the CIA value of the sample is to 100, the higher the kaolinite mineral enrichment is.The shales come in second place, and their CIA values range between 70 and 75.The illites, as one of the clay group minerals, come in third place, and their CIA value is 75.The CIA values of the granites and granodiorites range between 45 and 55 [26][27][28].
The triangular diagram (modified after [37]), showing the laterization degree of the red soil samples, revealed that they had weak, moderate, and strong laterization reactions.However, it is clearly seen that most of the samples underwent moderate laterization.
The abundance of Ca, Mg, and Sr was determined in the geochemical study carried out on the red clays formed on the river delta.Clays that have undergone the physical processes of weathering, sorting, dilution, and adsorption are characterized by red colors.The sediment transport along the delta has brought along geochemical differentiation [74].Red soils are heavily weathered soils.They contain secondary minerals with high Fe and Al content.However, soils that have not undergone high weathering and are characterized as tion.Therefore, it is thought that bauxite ore is present in the region.Detailed mineral exploration should be carried out in and around the study area.

Figure 1 .
Figure 1.The location map of the study area.Figure 1.The location map of the study area.

Figure 1 .
Figure 1.The location map of the study area.Figure 1.The location map of the study area.

Figure 5 .
Figure 5. Spatial distribution maps of the trace element concentrations of the red soil samples (a) Rb, (b) CIW, and (c) PIA.

Figure 8 .
Figure 8. Scree plot of the factor analysis.

Figure 10 .
Figure 10.Hierarchical clustering of the variables.Figure 10.Hierarchical clustering of the variables.

Figure 10 .
Figure 10.Hierarchical clustering of the variables.Figure 10.Hierarchical clustering of the variables.

Figure 11 .
Figure 11.Heat-map classification of the geochemical data and the paleoclimatic indices (a) Al 2 O 3 ; (b) CaO; (c) TiO 2 ; (d) SiO 2 ; (e) Fe 2 O 3 ; (f) Mn; (g) Co; (h) CIA; (i) CIW; (j) PIA; (k) R. The red soil that has undergone a weak laterization reaction has lower Al 2 O 3 , TiO 2 , and CaO contents.The red soil that has undergone a moderate laterization reaction has lower geochemical contents of TiO 2 and Mn, and index values of PIA and R. The red soil that has undergone a high level of laterization reaction has lower geochemical contents of CaO, TiO 2 , and SiO 2 , and index values of PIA and R.

Table 1 .
Geochemical analyses results of red soil samples (the major oxide elements (in %); trace elements and some rare earth elements (in ppm).

Table 2 .
The chemical alteration index values of the red soil samples.

Table 3 .
Pearson's correlation coefficient table according to the results of the chemical analysis and alteration indices. *−0.009

Table 4 .
Coefficient of determination (explained variance) table of the models.

Table 5 .
Statistical error (analysis of variance) table of the models.

Table 4 .
Coefficient of determination (explained variance) table of the models.

Table 5 .
Statistical error (analysis of variance) table of the models.

Table 6 .
Table of the coefficients of the mathematical models.
Extraction method: principal component analysis.a .Two components extracted.