A study on the mixed dielectric model of red clay Authors: Junhao Chen, Junli Nie, Yanling Feng, Xiangzhao Deng, Xijiang Zhang Published: 2026 ABSTRACT AbstractWhen ground-penetrating radar (GPR) detects red clay, the electric field induces polarization, causing corresponding characteristics in the dielectric constant and affecting GPR’s inversion of soil moisture content. This paper investigates the physical properties of Guiyang red clay, including its composition and liquid-plastic limits. Based on the electric double-layer structure on clay particle surfaces, bound water is classified into strongly and weakly bound water. Isothermal adsorption tests determined that the maximum strongly and weakly bound water contents of Guiyang red clay are 2.9% and 17.2%, respectively. Drawing on the unfrozen water model to calculate bound water content, a four-component dielectric mixing algorithm was applied to establish a comprehensive red clay dielectric mixing model integrating moisture content, texture, and frequency factors. Dielectric property tests obtained the complex permittivity of red clay at various moisture contents. Results verified that the mixing model well reflects their relationship during the bound water phase (i.e., moisture content θ within 0 ~ 17.2%). The strong correlation between predicted and measured values demonstrates the model’s feasibility and effectiveness. This provides guidance for the GPR-detectable soil moisture range, enabling more accurate inversion of soil moisture content.Keywords: Red clay, Electrical double layer, Bound water, Dielectric properties, Dielectric mixing modelSubject terms: Engineering, Environmental sciences, Materials science Introduction Ground-penetrating radar (GPR) is a non-destructive detection technology that transmits and receives high-frequency electromagnetic waves to characterize subsurface media. It is widely used for detecting soil moisture content. The primary theoretical basis for GPR detection lies in the contrast in dielectric constants. There exists a distinct relationship between the dielectric constant of soil and its moisture content, making the dielectric constant a critical parameter that causes wave reflection at media interfaces. When GPR is applied to red clay, the electric field induces polarization within the medium. This polarization caused the dielectric constant of red clay to exhibit corresponding characteristics in the reflected wave spectrum of the radar signal1, thereby enabling effective detection. Red clay was a special type of soil rich in hydrophilic clay minerals, making it highly prone to adsorbing and forming bound water2,3.It could be regarded as a porous medium composed of four components: air, dry soil, bound water, and free water4. Among these four components, the relative dielectric constant of bound water cannot be directly measured. Therefore, if a complex dielectric constant model of soil was to be constructed, bound water would be the most uncertain factor5,6. Most existing complex dielectric constant models for soil derive bound water content and dielectric constant through fitting algorithms. Jia Jinling7 utilized the concept of the simplex method to analyze and derive a complex transcendental equation that satisfies the testing of the complex dielectric constant of soil, then performed optimization iterations to solve the complex transcendental equation to obtain the complex dielectric constant of soil. For saline soil solutions, Gong Yan, Liu Baojiang, et al.8 determined the complex influence factors of ions in the soil solution and then directly used the Collie model to calculate the complex dielectric constant of the soil solution based on these factors. Liu Huagui, Zeng Jian, et al.9 determined the dielectric constant of soil by measuring the propagation time of electromagnetic waves along TDR probes. Boyarskii et al.10 measured the relaxation time of bound water on the surface of clay particles and constructed a dielectric model of bound water with the number of molecular layers of bound water as the independent variable. Wang and Schmugge11 proposed a linear combination of the dielectric constant values of bound water between those of free water and ice, and based on this, established a fitting expression for the dielectric constant of bound water. Some other models also used water content and exchangeable cations in soil to calculate the complex dielectric constant12. It was evident that existing algorithms for bound water rarely considered its fundamental cause—the electrochemical properties of soil13–15. Meanwhile, researchers generally agreed that in soil-water systems, bound water was primarily located on the surface of clay particles, serving as the cationic solution in the electric double layer of clay particle surfaces16,17. Based on the above discussion, the primary objective of this study is to establish a complex dielectric constant model for Guiyang red clay. The key to this model lies in the complex dielectric constant model of bound water and the bound water content model. The former attempts to be established using the electric double-layer theory of clay colloid chemistry, while the latter relies on research concerning the unfrozen water content in frozen soils. Experimental plan Basic physical properties of guiyang red clay To understand the relevant physical properties of Guiyang red clay, experiments such as XRD analysis and particle size testing were conducted. These tests provided basic physical properties of the red clay, including mineral composition, specific surface area, particle size, and porosity, while also supplying fundamental data for the establishment of a mixed dielectric model of the red clay. The red clay used in this experiment was collected from the Quaternary residual deposits on the campus of Guizhou University. It was primarily formed through the weathering residual and slope deposits of carbonate rocks (limestone and dolomite)18. Its main mineral components include quartz, kaolinite, muscovite, montmorillonite, and vermiculite. The specific mineral composition content is shown in Table 1. Through testing, the specific surface area (As) of the Guiyang red clay sample was determined to be 56.74 m²/g, the specific gravity (ρs) to be 2.64, the bulk density (ρb) to be 1.58 g/cm³, and the porosity (P) to be 0.402. Additionally, the liquid limit of the red clay sample was measured at 49.48%, the plastic limit at 35.56%, and the plasticity index at 13.92. The particle size analysis results are shown in Fig. 1. Table 1Mineral Composition of Guiyang Red Clay.Mineral CompositionQuartzKaoliniteMuscoviteMontmorilloniteVermiculiteProportion(%)57.523.917.60.50.5 Mineral Composition of Guiyang Red Clay. Particle Size Distribution Chart of Guiyang Red Clay. Analysis of Fig. 1 indicates that the particle size distribution of Guiyang red clay is heterogeneous, with particles smaller than 250 μm accounting for 72.39% of the total. This heterogeneity arose because mineral particles (< 20 μm) formed micro-aggregates (20–250 μm)19 under the influence of organic matter and metal ions. These micro-aggregates cause sand, silt, clay particles, and organic matter in the red clay to undergo cementation and cohesion, forming aggregates of varying sizes. These aggregates further organize and arrange themselves, ultimately constituting the macroscopic soil structure. Determination of bound water by isothermal adsorption experiment To accurately measure the bound water limits of Guiyang red clay, this study referred to the relevant research by Wang Pingquan20. Different saturated salt solutions (Table 2) were used to control the relative humidity environment. The saturated salt solutions were placed at the bottom of desiccators, and the dried soil samples were positioned in the upper part of the desiccators for adsorption (as shown in Fig. 2(a)). This method allowed for the measurement of the isothermal adsorption curve of Guiyang red clay and the analysis of its bound water range. Figure 2(b) shows the isothermal adsorption curve of Guiyang red clay. Table 2Saturated salt solutions and corresponding relative humidity.Saturated salt solutionLiBrLiCl·H2OCaCl2CO(NH2)2NaClKClZnSO4K2SO4H2ORelative Humidity (%)512315075.585.190.898100 Saturated salt solutions and corresponding relative humidity. Isothermal Adsorption Experiment: (a)Adsorption isotherm experiment experimental scenario; (b)Isothermal Adsorption Curve of Guiyang Red Clay. According to the characteristics of the Type IV isothermal adsorption curve21, the curve in Fig. 2(b) showed that the adsorption rate of water molecules was relatively low when the relative humidity was below 31%. This is due to the adsorption interaction between clay particles and water molecules, forming a relatively stable adsorbed water layer. This portion of bound water corresponds to strongly bound water. As shown in the figure, the maximum strongly bound water content is 2.9%. When the relative humidity exceeds 31%, the adsorption rate of water molecules increases significantly, primarily due to capillary water effects between soil particles. After the capillary water effects conclude, a plateau appears, where the stability of water molecules in the diffusion layer is weaker compared to strongly bound water. This portion of bound water corresponds to weakly bound water. Based on the figure, the maximum weakly bound water content is 17.2%. Establishment of a mixed dielectric model for red clay The red clay discussed in this study is a four-phase mixed medium, comprising dry soil particles, bound water, free water, and air. Among these, the dielectric constants of air and dry soil particles are fixed, while the dielectric constant of free water can be calculated using the Debye equation. The only uncertain parameter is the dielectric constant of bound water. The dielectric constant of bound water is closely related to its distance from the clay particle surface. Based on the electric double-layer structure on the clay surface, bound water can be classified into strongly bound water and weakly bound water. Strongly bound water refers to a very thin adsorbed solution layer on the clay particle surface, while weakly bound water exists outside the strongly bound water region but within the influence of the electric field generated by surface charges, forming the diffuse layer solution. Due to differences in the content and properties of these two types of bound water, their complex dielectric constants must be calculated separately. Electric double-layer theoretical model In essence, bound water is the electric double-layer solution on the surface of red clay particles. Therefore, we adopt the existing Stern-Gouy electric double-layer theory to describe the microstructure of clay particle surfaces and calculate the microscopic physical quantities of the soil17. According to electrochemical theory, when any solid comes into contact with a solution, an electric double-layer with opposite charges forms at the interface. This double-layer consists of surface charges and counterion solutions. Due to the incomplete lattice structure of clay particles, their surfaces carried a negative charge, while the counterions were positively charged cations [22]. Depending on the adsorption forces acting on the double-layer, the counterion solution can be further divided into an adsorption layer and a diffusion layer. Some ions are fixed to the clay particle surface through van der Waals forces, chemical bonds, and electrostatic attraction, forming the adsorption layer (also known as the Stern layer), which corresponds to the strongly bound water layer, as shown in Fig. 3. The remaining ions are dispersed in the solution by electrostatic forces, forming the diffusion layer (also known as the Gouy layer), which constitutes the weakly bound water layer. According to statistics from Olphen [23], the concentration of the cationic solution in the diffusion layer decreased exponentially with increasing distance from the surface. Electric Double-Layer System of Clay(Cited from Jin X’s research22. The total negative charge (σ) on the clay particle surface is the sum of the adsorption layer charge (σ₁) and the diffusion layer charge (σ₂): Theoretically, the charge quantity of the adsorption layer σ₁ can be calculated based on the number of cation adsorption sites and the corresponding interaction forces, expressed as follows: Where, is the number of adsorption sites per square centimeter on the surface of clay mineral particles;e = 1.6 × 10− 19 C, denoting the charge of a single electron=6.02 × 1023, the Avogadro constant;=18, the molecular weight of the solvent; is the concentration of the surrounding soil solution. is the adsorption layer potential(V);, representing the specific adsorption potential(kJ/mol); k=1.38 × 10− 23 J/K, the Boltzmann constant;is the thermodynamic temperature (K, Kelvin). The charge in the diffusion layer () is calculated based on the concentration of the surrounding soil solution () and the Stern potential () of the adsorption layer, using the following formula: In the equation, represents the average dielectric constant of the diffusion layer solution. When the surface charge density () and the soil solution concentration are known, the Stern potential () can be determined. All physical quantities in the electric double layer can be derived from the Stern potential. The thickness of the diffusion layer (d) can be expressed as: Expression for the relationship between ion concentration (n) in the diffusion layer and the distance from the surface (x): In the formula, the parameter α is defined as:. Unfrozen water model The Soil Freezing Characteristic Curve (SFCC) described the relationship between unfrozen water content and temperature [24]. As shown in Fig. 4, when the ambient temperature decreases (below 0 °C), soil water freezes sequentially based on its properties: first free water, followed by capillary water. After capillary water completely froze, bound water began to freeze [25,26]. According to the soil freezing characteristic curve, it can be observed that as the temperature decreases, the unfrozen water content decreases in a power-exponential manner. This pattern shares similarities with the exponential variation of ion concentration in the electric double-layer diffusion layer as the distance from the solid particle surface increases. Leveraging this characteristic, the ion concentration in the adsorption layer can be regarded as residual unfrozen water (i.e., strongly bound water ), while the ion concentration in the diffusion layer can be treated as ordinary unfrozen water (i.e., weakly bound water ) [27]. By drawing on the unfrozen water model, the strongly and weakly bound water content of red clay can be derived. Schematic Diagram of Unfrozen Water Content Variation with Temperature (after Xu Xuezu). According to the research by D.M. Anderson and A.R. Tice [28], the relationship between unfrozen water content and subzero temperature could be expressed by the following equation: where: Wu is the unfrozen water content (%);T is the subzero temperature (°C);C and D are constants determined by soil properties. Let W₀ represent the initial water content of the soil sample, and Tf denote its freezing point. Then: Establishing a mixed dielectric model for red clay (1)Strongly Bound Water Content Due to the electrostatic forces generated by the negative charges on clay particle surfaces, the diffusion layer near the particle surfaces (i.e., the solution layer containing ions adjacent to the particles) is essentially a cationic solution layer. The distribution of this ionic solution exhibits a unique structure, with higher ion concentrations closer to the particle surfaces and lower concentrations farther away. Therefore, the content of strongly and weakly bound water was related to their distance from the particle surfaces [29]. The relationship between the thickness of the water molecule layer (x) on the soil particle surface and the volumetric content of bound water (V) can be expressed as follows : Based on previous research findings, Verwey and Overbeek [29]and Olphen [30] suggested that the charge of strongly bound water is concentrated within approximately 5Å from the clay surface. Therefore, this study assumes an adsorption layer thickness of . Assuming the clay particle surface is an ideal plane and considering the linear relationship between specific surface area () and strongly bound water content, the maximum water content of strongly bound water () can be expressed as: where: is the specific surface area, the coefficient 0.9 is an adjustment parameter for the relationship between specific surface area and strongly bound water content [22], is the bulk density, is the adsorption layer thickness, approximately 5Å. For this red clay, substituting the values yields a maximum strongly bound water content of 2.99%, which aligns with the experimentally measured result. (2)Dielectric Constant of Strongly Bound Water Due to the non-uniform adsorption between strongly bound water and clay minerals, as well as the charged surface of red clay particles, the dielectric constant of strongly bound water () is not solely dependent on water content. According to assumptions by Dobson, Boyarskii D A [31], and others, the diameter of a bound water molecule was known to be approximately 2.5 Å or 2.8 Å. Therefore, the strongly bound water in red clay roughly corresponds to two layers of bound water molecules. Based on the conclusion by Wang and Schmugge [11], the dielectric constant of water on mineral surfaces was similar to that of ice. Hence, the dielectric constant of the first layer can be considered equal to that of ice () and conforms to the Debye equation, while the dielectric constant of the second layer of bound water molecules is more challenging to estimate. For calculation purposes, we assume the dielectric constant of the first layer equals that of ice (). The relaxation times of the first two layers of water molecules on the clay particle surface were known to be 5.0 × 10⁻¹⁰ seconds and 5.0 × 10⁻¹¹ seconds, respectively [22], both satisfying the Debye equation. Based on the two relaxation times of bound water molecules, the dielectric constant of the second layer of bound water is derived as 16 + j0.5, which is approximately five times that of the first layer. The complex dielectric constant of strongly bound water can be summarized as follows: where θ is the volumetric water content of the soil, and is the maximum strongly bound water content. The calculation results are shown in Fig. 5: Fitted Relationship Between Dielectric Constant of Strongly Bound Water and Soil Water Content. As shown in Fig. 5, when the soil water content is less than the maximum strongly bound water content, the dielectric constant of strongly bound water exhibits a linear positive correlation with soil water content. As the water content continues to increase and exceeds the maximum strongly bound water content, the dielectric constant of strongly bound water remains constant. Therefore, the fitted relationship between the dielectric constant of strongly bound water () and the volumetric water content (θ) for Guiyang red clay can be expressed as: (3)Weakly Bound Water Content By analogy with the unfrozen water model to calculate the weakly bound water content, it can be observed from Fig. 4 that the maximum weakly bound water content (=17.2%) can be regarded as approximately equal to the unfrozen water content at -1.5 °C. According to the unfrozen water model, the maximum volumetric content of weakly bound water is expressed as: where, , , ,, (absolute value of the subzero temperature), As is the specific surface area. Since the calculation result of the unfrozen water model is in terms of mass water content, it is necessary to multiply by the bulk density () to convert it to volumetric water content. Substituting the values yields a maximum volumetric bound water content of approximately 17.18%, with an absolute error of 0.02% compared to the experimental value, indicating the feasibility of the model. (4)Dielectric Constant of Weakly Bound Water According to experimental evidence from relevant researchers [32], the dielectric properties of unfrozen water were similar to those of saline solutions. Therefore, the dielectric constant of weakly bound water can be calculated using the Debye equation for saline solutions [33] (i.e., ). Based on this, in practical calculations, we use a sodium chloride (NaCl) solution of the same concentration to substitute for calculating the dielectric constant of weakly bound water. The expression for the dielectric constant of the sodium chloride solution is as follows: In the formula, f represents frequency; is the dielectric constant of the sodium chloride solution at the high-frequency limit, taken as 4.9; is the static dielectric constant of the sodium chloride solution; is the relaxation time of the saline solution; (imaginary unit); is the permittivity of vacuum, taken as 8.854 × 10⁻¹² F/m; is the effective conductivity of the saline solution, in S/m. In the calculation of this mixed model, the temperature is held constant. Therefore, , , and depend solely on the salt concentration. The thickness of the weakly bound water layer is approximately 20 Å. According to research by Xiao Jin et al., the average cation concentration in the soil is about 0.36 mol/L, which corresponds to a sodium chloride solution concentration of approximately S = 20.9‰. The calculated result for this concentration is 78.17. (5)Dielectric Constant of Bound Water The dielectric constant of bound water () is equal to the product of the dielectric constants of strongly bound water () and weakly bound water () and their respective volume fractions, expressed as Eq. (15). . The variation of the expression for the dielectric constant of bound water with soil moisture content can be fitted as shown in Fig. 6. Fitted Relationship Diagram Between Relative Dielectric Constant of Bound Water and Soil Moisture Content in Red Clay. This model represents the dielectric model expression for bound water in Guiyang red clay investigated in this study. (6)Mixed Dielectric Model of Red Clay. The red clay used in the experiment is a four-phase mixed medium, comprising dry soil particles, bound water, free water, and air, where bound water is further categorized into strongly bound water and weakly bound water. By applying a four-phase mixed dielectric algorithm, a mixed dielectric model for this red clay is established, with its dielectric constant expression given by: In the equation, the subscripts soil, air, ss, f, and b represent soil, air, dry soil particles, free water, and bound water, respectively. , where is the specific gravity, is the bulk density, and P is the porosity. , where θ is the volumetric water content. When , , and . When , , and , where is the maximum volumetric content of bound water. , and. α is a shape factor characterizing the geometric morphology of red clay particles. In the four-phase mixed dielectric model, α = 0.65. The dielectric properties of the components in red clay vary significantly. Similar to the dielectric constant of bound water, the final mixed dielectric model of red clay also exhibits distinct segmented characteristics, specifically manifested as: . In the equation, the subscripts soil, air, ss, f, and b represent soil, air, dry soil particles, free water, and bound water, respectively; P denotes porosity, and θ represents the volumetric water content of the soil. By applying the aforementioned model and substituting the relevant parameters, the fitted relationship for the mixed dielectric model of Guiyang red clay can be obtained, as shown in Fig. 7. The fitted relationship is expressed by Eq. (19): Fitted Curve of the Mixed Dielectric Model for Guiyang Red Clay. represents the relative dielectric constant of the red clay, and θ denotes the volumetric water content of the red clay. Dielectric testing validation of the mixed dielectric model This study employed the Agilent E4991A to test the dielectric properties of Guiyang red clay samples. Based on isothermal adsorption tests, three types of soil samples were selected with the following moisture states: dry soil, containing weakly bound water, and containing free water. The sample thicknesses were 0.66 mm, 0.58 mm, and 0.62 mm, respectively. The tests were conducted under room temperature conditions, with a frequency measurement range of 1 MHz to 1 GHz. Measurements were taken at linear equally-spaced frequency points. The experimental results are shown in Fig. 8. Dielectric spectrum of Guiyang red clay As shown in Fig. 8(a), within the frequency range of 1 MHz to 100 MHz, the real part of the complex dielectric constant for all three soil moisture states decreases exponentially rapidly as frequency increases. In the frequency range of 100 MHz to 500 MHz, the real part of the complex dielectric constant for all three soil types stabilizes with minimal variation. Beyond 500 MHz, the real part of the complex dielectric constant for red clay containing weakly bound water and free water exhibits a slow increasing trend, while that of dry soil shows a slow decreasing trend. Additionally, across the entire frequency range, the magnitude of the real part of the complex dielectric constant consistently follows the order: free water-containing sample > weakly bound water-containing sample > dry sample. As illustrated in Fig. 8(b), by comparing the imaginary part of the complex dielectric constant of dry soil and red clay at different moisture contents, the following observations can be made: In the frequency range of 1 MHz to 100 MHz, the imaginary part of the complex dielectric constant for red clay in all three moisture states decreases exponentially rapidly as frequency increases. In the frequency range of 100 MHz to 700 MHz, the imaginary part of the complex dielectric constant of red clay gradually stabilizes with minimal variation. Beyond 700 MHz, the imaginary part of the complex dielectric constant for red clay containing free water and weakly bound water exhibits a slow increasing trend, while that of dry soil shows a slow decreasing trend. Based on the above analysis, it can be concluded that the complex dielectric constant of red clay varies under different moisture content states. Additionally, as the frequency changes, both the real and imaginary parts of its complex dielectric constant also undergo alterations. To further validate the accuracy of the mixed dielectric model, we conducted isothermal adsorption tests again and obtained 12 groups of soil samples with different moisture contents. Dielectric property tests were performed on these samples to determine their dielectric constants.A comparison between the measured dielectric constant values of red clay and the predicted values from the model is shown in Fig. 9. It can be observed that for the first nine sets of data, corresponding to the stage where soil moisture content is in the bound water phase, the standard deviations between the measured and predicted values are relatively small, all below 3 (Analysis of the test results revealed that when the soil moisture content was 1.86% and 20.07%, the dielectric constant showed abnormal fluctuations. This phenomenon might be associated with differences in compaction during sample preparation, changes in test contact conditions, or fluctuations in instrument measurements. To prevent an impact on the research analysis, it was thus not included in the scope of the analysis). In contrast, the standard deviations for the last three sets of data are relatively larger, reaching up to 5.43. Comparison of dielectric constants in guiyang red clay. After excluding the two sets of abnormally varying data from the table, the predicted values calculated using the formula were fitted and compared with the measured values. As shown in Fig. 10, the coefficient of determination (R²) between the measured values and the model predictions is 0.9364, indicating a strong correlation between the experimental data and the simplified model. Further Pearson correlation analysis between the measured and predicted dielectric constants yields a Pearson correlation coefficient of 0.9204, with a p-value of approximately 0.0002, which is significantly lower than 0.05. This confirms a strong correlation between the two sets of values. Fitting of measured and predicted dielectric constants of Guiyang red clay. In summary, the mixed dielectric model for red clay proposed in this study is feasible and effective when the moisture content of the red clay is within the bound water phase, i.e., when the volumetric water content θ ranges from 0% to 17.2%. Within this range, the model provides reliable predictions. However, when the moisture content exceeds 17.2%, the model’s error increases significantly. As illustrated in Fig. 10, the dielectric constant predicted by the model for red clay under such conditions is higher than the experimentally measured values. Conclusion To improve the accuracy of ground-penetrating radar in detecting soil moisture content, this study focuses on Guiyang red clay and draws the following conclusions: (1) The properties and mineral composition of Guiyang red clay were determined through experiments. Based on the electric double-layer theoretical model of clay, the bound water in Guiyang red clay was classified into strongly bound water and weakly bound water. Isothermal adsorption tests revealed that the boundary between strongly and weakly bound water is 2.9%, while the boundary between bound water and free water is 17.2%. (2) The model is suitable for calculating the dielectric constant of red clay within the bound-water range. By using a four-phase mixed system for segmented calculations, the relationship expression of the mixed dielectric model for Guiyang red clay samples was derived based on the bound - water dielectric model grounded in the electric double-layer theory. Validation results demonstrate that, within the bound-water phase (i.e., when the moisture content θ ranges from 0% to 17.2%), the mixed dielectric model effectively describes the relationship between the dielectric constant and moisture content of red clay. Additionally, the predicted values show strong agreement with the measured values, with an R² of 0.9364, indicating a high correlation between the two. Author contributions J.C. performed conceptualization, methodology, investigation, and wrote the original draft. J.N. supervised the study, provided resources, reviewed and edited the manuscript, and handled funding acquisition. Y.F. contributed to methodology and wrote the original draft. X.D. and X.Z. reviewed and edited the manuscript, performed validation, and assisted with the experiments. All authors reviewed the manuscript. Funding Funding: This work is supported by National Natural Science Foundation of China(grant no.42264008), Deep Earth Probe and Mineral Resources Exploration-National Science and Technology Major Pro-ject(grant no.2024ZD1002202). Data availability The data from this study can be provided upon request with the author’s consent. Declarations Competing interests The authors declare no competing interests. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement The data from this study can be provided upon request with the author’s consent.