Mineralogical and Geochemical Evolution During Limestone Weathering and Pedogenesis in Shimen, Hunan Province, South China

Abstract

Understanding mineralogical transformations and elemental mobility during limestone weathering is critical for deciphering carbon cycling and critical zone evolution in karst terrains. This study investigates an in situ limestone weathering profile (12.6 m depth) in Shimen, Hunan Province, using integrated mineralogical (XRD, EPMA-EDS), elemental (XRF, ICP-MS), and Sr isotopic (MC-ICP-MS) analyses. Results reveal a two-stage pedogenic model: (1) Rapid dissolution of primary calcite (>95 wt% in bedrock to 1.1–48.5 wt% in soil) creates an abrupt bedrock–soil interface via volumetric collapse (>90%), accumulating acid-insoluble residues (quartz-dominated); (2) Subsequent weathering drives illitization of K-feldspar, trace element enrichment (e.g., Ni, Tl, Th τ up to 180) via illite adsorption, and radiogenic ⁸⁷Sr/⁸⁶Sr evolution (0.7076 in bedrock to 0.7292 in soil). Depth-dependent increases in chemical index of alteration (CIA: 6.79–79.96) and mass transfer coefficients confirm progressive weathering intensity. The profile acts as a net carbon source (58.5% depletion in soil inorganic carbon), highlighting significant CO₂ release during pedogenesis. These findings provide mechanistic insights into subtropical critical zone evolution and element cycling in carbonate-dominated systems.

1. Introduction

Weathering is the process by which rocks disintegrate at or near the Earth’s surface, involving interactions among multiple spheres, including the lithosphere, hydrosphere, atmosphere, and biosphere [1,2]. As one of the mechanisms for the redistribution of surficial materials on Earth, it constitutes a fundamental component of the Earth system, playing a pivotal role in global material cycles, climate regulation, and the evolution of landforms [1,3,4,5]. The essence of chemical weathering of rocks lies in a series of water–rock interaction processes (e.g., hydration, oxidation, and hydrolysis), which lead to the decomposition of primary minerals, the formation of secondary minerals, and the migration and redistribution of elements [3,6]. It is well established that the spatiotemporal distribution of temperature and precipitation defines climatic zones, and the observed variations in weathering intensity, chemical weathering rates, and soil types across different climatic zones provide direct evidence of climatic control over weathering processes [7,8].

The Shimen area in northern Hunan Province features a humid subtropical monsoon climate, where optimal hydrothermal conditions promote moderate chemical weathering intensity, making this region particularly suitable for studying weathering processes and their underlying mechanisms [9]. Limestone, a common sedimentary rock at Earth’s surface, primarily occurs in shallow-marine platforms and epicontinental basins [10]. Limestone is extensively distributed in the Shimen area, covering approximately one-third of the total region [11], with well-developed thick weathering profiles. These weathered strata not only provide an ideal natural archive for investigating mineralogical and geochemical evolution during weathering and pedogenesis but also hold significant implications for both geohazard prevention engineering and global carbon neutrality strategies [11,12].

In recent years, scholars both domestically and internationally have conducted extensive research on the weathering processes of limestone, yielding substantial scientific achievements. Based on comprehensive field observations, mineralogical and geochemical analyses of limestone weathering profiles, Wang et al. [13] proposed a two-stage weathering and pedogenesis model for carbonate rocks. Kirschbaum et al. [14] observed that limestone weathering profiles in karst regions generally has a nearly distinct soil–rock interface and is absent of the gradual weathering zones that are composed of weathering frontier, saprolite, and pedogenic front in most in situ rock weathering profiles. Liu et al. [15] conducted a systematic geochemical investigation of limestone-derived soil profiles across the Yunnan–Guizhou Plateau, Southwest China, demonstrating that pedogenetic processes integrate both in situ chemical alteration of weathering residues and allochthonous incorporation of detrital materials. Dabski et al. [16] pioneered an integrative approach combining geomorphometry, rock mechanics, and biogeochemical methodologies to elucidate, for the first time, the cross-scale evolutionary patterns of weathering rinds on limestone surfaces in the Alpine glacier retreat zone. It is noteworthy that although previous studies have comprehensively summarized the geochemical mechanisms governing limestone weathering and pedogenesis at regional scales, significant knowledge gaps persist regarding: (1) systematic characterization of the complete bedrock–soil profile weathering continuum, (2) mechanistic understanding of coupled mineral transformation and elemental mobility during weathering processes, and (3) quantitative assessment of carbon source/sink effects during bedrock-to-regolith conversion.

Rock weathering represents a fundamental geochemical process involving the mobilization and reorganization of carbon and other elements, manifesting distinct distribution patterns, evolutionary pathways, and kinetic differentials [1]. Consequently, investigating rock weathering constitutes a critical component in deciphering the biogeochemical behavior of carbon and associated elements during Earth surface processes. The weathering of Shimen limestone refers to the holistic process of physical and chemical alterations induced by atmospheric, aqueous, and biological interactions at or near the Earth’s surface, ultimately forming in situ unconsolidated deposits. However, systematic investigations into its chemical weathering mechanisms remain absent in the literature. Since Sr isotopes are generally not fractionated during surface processes and are thus considered an ideal tracer for material sources [17], this study integrates micro-mineralogical composition, major elements (including carbon content), trace elements, and Sr isotopic variations to investigate the weathering intensity and trends of the limestone weathering profile. By deciphering the mineral evolution sequence and its coupling mechanism with elemental migration and enrichment, this work provides new insights into the geochemical processes of limestone weathering. Additionally, the geochemical characteristics of regional soils can be significantly influenced by external inputs (e.g., atmospheric deposition, anthropogenic activities) and may not fully represent the original in situ weathering and pedogenic geochemical signals [15,18,19]. Previous studies have primarily focused on the petrogenesis of rocks from different periods, while investigations of soil trace elements were limited to simplistic sampling at the soil level without systematically integrating the weathering–pedogenesis continuum. The influence of complex environmental factors during rock weathering on soil development and evolution remains poorly constrained and warrants further exploration. Therefore, investigating mineralogical and geochemical characteristics at the rock–soil profile scale is essential to accurately elucidate the authentic geochemical processes of weathering and pedogenesis.

This study presents a comprehensive analysis of an in situ limestone weathering profile in the Shimen area, employing an integrated analytical approach utilizing XRD, EPMA-EDS, XRF, ICP-MS, and MC-ICP-MS to systematically investigate microscale mineralogical transformations, major and trace element geochemistry, and Sr isotope variations throughout the weathering sequence. The research elucidates the spatial distribution patterns of mineral phases, elemental concentrations, and Sr isotope ratios, while evaluating weathering intensity gradients, evolutionary trends, and the coupled relationships between mineral alteration processes, elemental migration/enrichment dynamics, and dissolution porosity effects. These findings provide fundamental insights into the geochemical processes governing limestone weathering and critical zone development in subtropical environments, contributing both theoretical advances in understanding carbonate weathering mechanisms and practical implications for ecosystem conservation in the Dongting Lake watershed, particularly regarding soil preservation and water quality management in this ecologically sensitive karst region.

2. Geological Setting and Sampling Description

The study area is located in Shimen County, Changde City, northern Hunan Province, and is tectonically situated in the transitional zone between the Yangtze Block and the Jiangnan Orogenic Belt (Figure 1a) [20,21]. Since the mid-Neoproterozoic, the region has undergone intense orogenic movements. Prolonged tectonic activity has generated abundant microfractures and fault zones within the rocks [22,23]. The regional stratigraphy is predominantly composed of the Early Paleozoic Silurian System, Late Paleozoic strata, Mesozoic Triassic System, and Cenozoic Quaternary System (Figure 1b). The Neoproterozoic Nanhua System can be divided into five lithologic members, which in ascending order are tuffaceous sandstone, silty slate, quartz sandstone, manganiferous dolomite, and tillitic mudstone [24]. No outcrops of the Sinian System are observed in the study area. The Cambrian System of the Early Paleozoic unconformably overlies the underlying Nanhua System at a high angle, comprising a black shale series with basal carbonaceous shale and top-layered limestone. The overlying Ordovician System consists of a graptolite-bearing assemblage, including argillaceous micritic limestone, calcareous shale, sandy shale, and siliceous rocks, exhibiting a conformable contact with the Cambrian. The Silurian System comprises a fossiliferous assemblage of carbonaceous shale, siltstone, grayish-green mudstone, and marl, yielding trilobites. It exhibits a conformable contact with the underlying Ordovician strata. The Devonian System of the Late Paleozoic consists of a suite of terrigenous clastic rocks, predominantly quartz sandstone, sandy shale, siltstone, and argillaceous dolomite. The Carboniferous System conformably overlies the Devonian and is characterized by carbonate-dominated deposits, mainly limestone, micritic limestone, and bioclastic limestone, with locally intercalated peloidal dolomite. The Permian System is largely absent in the study area, with only sporadic outcrops of limestone and dolomite [23]. The Triassic System is composed of limestone, marl, and minor argillaceous dolomite, with the sampled section located within this stratigraphic unit. The Jurassic System is predominantly absent in the study area, with only sporadic exposures of sandstone observed. The Cretaceous System consists of coarse-grained clastic deposits, dominated by conglomerates and pebbly sandstones. The Quaternary System displays two distinct soil units: a basal alluvial deposit overlain by an upper proluvial deposit. Alluvial soil, characterized by high limestone content and fertility, serves as a core region for wheat-corn rotation systems. It predominantly exhibits a silt loam texture with grayish-brown coloration, typically displaying blocky or granular structure. Proluvial soil is primarily utilized for pastureland and ecological conservation areas. It ranges from sandy loam to clay loam in texture, featuring a yellowish-brown surface layer and a pale yellowish-brown subsurface layer. Irregular sedimentary bedding and oriented gravel arrangements are commonly observed [22,23,25]. Additionally, the Shimen area is situated within a low-hilly karst landscape, representing a typical geomorphic unit in the subtropical carbonate rock regions of southern China. This terrain is characterized by gently undulating, rounded hills with an overall elevation of approximately 300 m above sea level. The mean annual temperature and precipitation are 16 °C and 1400 mm, respectively, with an average annual relative humidity of 83% and 1450 h of sunshine per year. The hill slopes typically exhibit convex profiles with gradients generally ranging from 25° to 35°. The sampling site for this study was located in a mid-backslope position. Prior to detailed sampling, extensive field investigations were conducted across multiple hill slopes within the study area. It was determined that the transition zone at the mid to lower backslope positions is the most common and accessible area, where complete and well-developed weathering profiles are consistently preserved [23,26,27]. The mild climate and excellent outcrop conditions have promoted extensive weathering profile development in the region, establishing it as an ideal natural laboratory for investigating chemical weathering processes in limestone.

To minimize the influence of allochthonous materials on the weathering profile, this study selected sampling sites at artificial slope outcrops within the residual zone. As shown in Figure 2, the limestone weathering profile exhibits a distinct vertical weathering sequence. Based on field observations of rock color, joint development, fabric, and roundness, the profile can be divided from bottom to top into three zones: (1) Bedrock layer: Grayish-white in color with sparse joints and fractures, exhibiting well-defined bedding and massive structure, and poor roundness; (2) Weathered layer: Dark gray with reddish-brown mottling, showing moderately developed joints and fractures, cataclastic structure, and intermediate roundness. Notably, nodular dissolution pores are observed at 10.4 m depth; (3) Soil layer: The soil layer exhibits a yellowish-brown appearance, with a soft consistency when moist and low hardness when dry, showing an absence of layering structure. It is characterized by a single-grain texture, with soil particles being moderately well-rounded. A small number of fine plant roots are observed within the layer, along with coarse clastic materials such as subangular to rounded rock fragments. The limestone weathering profile was sampled to a depth of 12.6 m, with a total of 50 samples collected, including 6 soil samples, 42 weathered rock samples, and 2 fresh bedrock samples. Following standard channel sampling methodology, samples were systematically collected from top to bottom across different weathering horizons. Approximately 1 kg of material was collected per sample, immediately sealed in zip-lock bags under ambient conditions, and transported to the laboratory. Detailed sampling depths and stratigraphic positions are illustrated in Figure 2.

3. Analytical Methods

3.1. Mineralogical Composition and Microscale Morphology

Mineralogical composition analyses for whole-rock were determined using an Empyrean DY1411 X-ray diffraction (XRD) at the State Key Laboratory for Critical Mineral Research and Exploration (SKLCMRE), Institute of Geochemistry, Chinese Academy of Sciences (IGCAS). The samples were first cleaned by removing extraneous materials (plant roots and rock fragments), then oven-dried at 60 °C for 24 h to ensure complete dehydration. The dried samples were mechanically crushed and ground to fine powder using an agate mortar. The homogenized powders were analyzed by XRD operating at a voltage of 40 kV and a current of 40 mA. XRD data were processed using JADE 6.0 software with Reference Intensity Ratio (RIR) for quantitative phase analysis of mineral compositions [28]. The analytical results are shown in Table S1. The mineral micromorphology analysis was conducted on representative thin sections prepared from each weathering horizon at the Wuhan Sample Solution Analytical Technology Co., Ltd., based on XRD analytical results. The employed instrument is a field emission electron probe (model JXA-iHP200F) manufactured by Japan Electronics (JEOL), featuring an integrated energy dispersive spectroscopy (EDS) system. High-resolution backscattered electron (BSE) imaging of mineral phases was performed using a JEOL JXA-iHP200F (Tokyo) electron probe microanalyzer (EPMA) operated at 20 kV accelerating voltage and 5 nA probe current. Qualitative chemical analysis of selected mineral grains was conducted using energy-dispersive X-ray spectroscopy for mineral identification. The EDS analyses were performed under operational conditions of 20 kV accelerating voltage, 20 nA beam current, and 1 μm probe diameter as explained in Li et al. [29].

3.2. Whole-Rock Geochemistry

The whole-rock major element compositions, carbon content, and trace element analyses were conducted at the Geological Experiment and Testing Center of Hunan Province. Major element concentrations were performed by X-ray fluorescence spectrometry (XRF) using a Rigaku RIX 2000 system (Tokyo, Japan) operating at 70 kV and 140 nA with Ka emission lines, achieving analytical uncertainties better than 2%. Loss on ignition (LOI) analysis was performed by heating pre-weighed samples in a muffle furnace at 1100 °C for 3 h until constant weight was achieved. The percentage of mass loss was calculated from the weight difference before and after heating after cooling to room temperature in a desiccator. The analytical procedures strictly followed the methodology described by Yang et al. [30], with detailed results presented in Table S2. The soil inorganic carbon (SIC) was determined using the volumetric method for soil carbonates, while soil organic carbon (SOC) was measured via the potassium dichromate oxidation method. Total carbon (TC) content was analyzed using a high-frequency infrared carbon–sulfur analyzer (SES-902, SAIENCE, Chengdu, China). Detailed experimental procedures followed the protocol established by Yu et al. [31], with all carbon fraction results presented in Table S2. Trace element concentrations were determined using a Thermo Fisher X2 series inductively coupled plasma mass spectrometer (ICP-MS), with analytical procedures following the protocol established by Qi et al. [32]. Analytical precision of most elements in the standard samples is typically 1–5%. Complete analytical results are presented in Table S3. Whole-rock Sr isotopic measurements were performed using a Thermo Fisher Scientific Neoma multi-collector inductively coupled plasma mass spectrometer (MC-ICP-MS) at the Wuhan Sample Solution Analytical Technology Co., Ltd., Wuhan, China, following the detailed analytical protocols established by Zhou et al. [33]. The complete dataset, including both reference materials and samples, is presented in Table S4.

3.3. Calculation of the Chemical Index of Alteration and Mass Transfer Coefficients

Considering the research objectives and available dataset, the chemical index of alteration (CIA) was employed to evaluate weathering intensity in the limestone from the Shimen area [5,34]. The CIA was calculated using the standard formula:

$$\mathrm{C}\mathrm{I}\mathrm{A}=\left[{\mathrm{A}\mathrm{l}}_2{\mathrm{O}}_3/{(\mathrm{A}\mathrm{l}}_2{\mathrm{O}}_3+{\mathrm{C}\mathrm{a}\mathrm{O}}^{*}+{\mathrm{N}\mathrm{a}}_2\mathrm{O}+{\mathrm{K}}_2\mathrm{O})\right]\times 100$$

(1)

where all components are expressed as molar proportions. CaO* denotes the content of silicate-bound CaO, calculated by subtracting carbonate-associated CaO [18,35,36].

The lowermost bedrock sample (24SM50) was selected as the parental reference material, with zirconium (Zr) serving as the immobile reference element for computing mass transfer coefficients (τ) of both major and trace elements to quantify their relative depletion or enrichment during chemical weathering processes [37,38]. The mass balance equations are formulated as:

$$\mathsf{\tau }={\displaystyle \frac{\raisebox{1ex}{${\mathrm{C}}_{\mathrm{X},\mathrm{S}}$}\!\left/ \!\raisebox{-1ex}{${\mathrm{C}}_{\mathrm{Z}\mathrm{r},\mathrm{S}}$}\right.}{\raisebox{1ex}{${\mathrm{C}}_{\mathrm{X},\mathrm{B}}$}\!\left/ \!\raisebox{-1ex}{${\mathrm{C}}_{\mathrm{Z}\mathrm{r},\mathrm{B}}$}\right.}}-1$$

(2)

In the given formula, C_X,S and C_Zr,S denote the concentrations of element X and Zr in the sample, while C_X,B and C_Zr,B represent their corresponding concentrations in the bedrock, with τX values >0, <0, and =0 indicating enrichment, depletion, and immobility of element X, respectively.

4. Results

4.1. Weathering Profile Mineral Composition

The XRD analysis results of the limestone weathering profile samples from the Shimen area are presented in Figure 3. The detrital mineral assemblage consists predominantly of quartz, plagioclase, K-feldspar, calcite, and dolomite, with illite being the sole clay mineral identified throughout the profile. Within the weathering profile, quartz content exhibits minimal oscillatory variations (2.1–4.9 wt%) throughout the bedrock and weathered layers but demonstrates a marked enrichment (up to 56.7 wt%) in the soil layer (Figure 3a). The plagioclase, K-feldspar, and illite contents exhibit a distinct stratigraphic partitioning, being virtually absent in the bedrock and weathered layers (<1 wt%) but showing abrupt enrichment in the soil layer (Figure 3b–d). Calcite maintains consistently high contents (~95 wt%) throughout both the bedrock and weathered layers but decreases abruptly in the soil layer (1.1–48.5 wt%; Figure 3e). Dolomite displays complex oscillatory behavior in content with depth through the weathering profile, exhibiting multiple fluctuations within a range of 0–6.6 wt% (Figure 3f). Notably, none of the minerals exhibit anomalous contents at dissolution pore locations (Figure 3).

EPMA-EDS analyses revealed the presence of euhedral prismatic to hexagonal quartz crystals in the bedrock layer of the limestone weathering profile, with grain sizes of 5–10 μm (Figure 4a). In the soil layer, quartz grains exhibit serrated to embayed dissolution textures (5–8 μm; Figure 4b), while plagioclase grains up to 10 μm with well-developed embayed dissolution features are simultaneously present (Figure 4c). Calcite is abundantly distributed in the bedrock layer, with most grains measuring several micrometers in diameter (Figure 4d), while dolomite exhibits characteristic rhombic structures reaching several micrometers in size (Figure 4e). Furthermore, the soil layer contains abundant poorly crystalline to amorphous iron-bearing minerals (Figure 4f).

4.2. Whole-Rock Major Elements in Weathering Profile

Figure 5 illustrates the variations of major elements in samples from the limestone weathering profile. Overall, the contents of SiO₂, Al₂O₃, and TFe₂O₃ show minor fluctuations in the bedrock and weathered layers with decreasing depth, followed by a sharp increase in the soil layer. Their respective concentration ranges are 1.40–62.00 wt%, 0.03–18.80 wt%, and 0.20–7.35 wt%. The MgO content exhibits irregular oscillations through the bedrock and weathered layers with decreasing depth, followed by a moderate increase in the soil layer, ranging from 0.49 to 1.44 wt%. CaO content shows minor fluctuations in the bedrock and weathered layers with decreasing depth, followed by a sharp depletion in the soil layer, ranging from 0.51 to 54.44 wt%. The contents of Na₂O, K₂O, and TiO₂ exhibit minor oscillations throughout the bedrock and weathered layers with decreasing depth, followed by significant enrichment in the soil layer, with concentration ranges of 0.08–0.36 wt%, 0.13–3.44 wt%, and 0.02–0.84 wt%, respectively. At the dissolution pore locations (10.4 m depth), relative depletion of SiO₂, Al₂O₃, TFe₂O₃, MgO, K₂O, and TiO₂ was observed, while CaO showed slight enrichment. Notably, the variation trends of SiO₂, CaO, Na₂O, and K₂O contents in the weathering profile correspond well with the abundance changes of quartz, calcite, plagioclase, and illite, respectively (Figure 3 and Figure 5), demonstrating a clear coupling between mineral transformations and elemental mobility during limestone weathering. The soil layer shows markedly high TFe₂O₃ concentrations (up to 7.35 wt%; Figure 5c), which correlates with both the absence of detectable iron-bearing minerals in XRD patterns and the EPMA-EDS observations of poorly crystalline to amorphous iron phases (Figure 3 and Figure 4f).

4.3. Whole-Rock Carbon and Weathering Indicators in Weathering Profile

Figure 6 illustrates variations in carbon and weathering indicators across the limestone weathering profile. The SIC content (0.00–58.50 wt%) shows minor fluctuations in bedrock and weathered layers with decreasing depth, followed by a sharp depletion in the soil layer. SOC content (0.04–1.57 wt%) exhibits minor oscillations through the bedrock and weathered layers, followed by a pronounced accumulation in the soil layer with decreasing depth. LOI values (9.04–43.08 wt%) show slight fluctuations in the bedrock and weathered layers but undergo a sharp decrease in the soil layer with decreasing depth. The CIA values (6.79–79.96) exhibit a progressive upward trend throughout the weathering profile, indicating intensifying chemical weathering toward the surface. At the dissolution pore locations (10.4 m depth), samples exhibit a significantly negative CIA anomaly. Furthermore, the depth-dependent variations of SIC content show strong correspondence with calcite abundance throughout the weathering profile (Figure 3e and Figure 6a).

4.4. Whole-Rock Trace Elements in Weathering Profile

Figure 7 illustrates trace element variations in the limestone weathering profile. Overall, Mo, Tl, and Ta exhibit similar concentration ranges (1–1.5 ppm) and demonstrate synchronous variation patterns with decreasing depth (Figure 7a). Th, Co, As, and Pb also display comparable concentration ranges (20–30 ppm) with synchronous depth-dependent variations (Figure 7b). Li, Ni, Cu, Cr, and Zn exhibit consistent concentration ranges (60–90 ppm) with synchronous depth-dependent variations (Figure 7c). Rb, V, and Ba display comparable concentration ranges (200–400 ppm) with covarying depth profiles (Figure 7d). Hf, Nb, and U (5–20 ppm) exhibit minor oscillations in bedrock and weathered layers, transitioning to progressive enrichment in the soil layer with decreasing depth (Figure 7e). Sr displays substantial concentration variability (100–4000 ppm), exhibiting an overall oscillatory decreasing trend throughout the weathering profile (Figure 7f). Notably, none of the trace elements exhibit significant concentration anomalies at dissolution pore locations (Figure 7).

4.5. Whole-Rock Sr Isotopes in Weathering Profile

The ⁸⁷Sr/⁸⁶Sr ratios of limestone weathering profile samples (Table S4) display remarkably high and variable values (0.707582–0.729201) across 50 analyzed samples. The ⁸⁷Sr/⁸⁶Sr ratios display a progressive enrichment trend throughout the weathering profile, with bedrock layer samples ranging from 0.707629 to 0.707664 (mean 0.707647), weathered layer samples from 0.707582 to 0.707802 (mean 0.707668), and soil layer samples exhibiting significantly more radiogenic values of 0.709230 to 0.729201 (mean 0.721773). Overall, the ⁸⁷Sr/⁸⁶Sr ratios exhibit a progressive up-profile enrichment trend with decreasing depth.

5. Discussion

5.1. Chemical Weathering Intensity and Trends

The weathering intensity of rock layers is closely linked to paleoclimatic conditions (e.g., humidity, temperature) and tectonic activity during weathering processes [39,40,41,42]. As shown in Figure 6, the CIA values of limestone samples range from 6.79 to 59.22, indicating a low degree of chemical weathering, while those of the soil samples range from 75.84 to 79.96, suggesting a moderate degree of chemical weathering [5,43].

During the incipient stage of rock weathering, the weathering reactions are governed by the reaction kinetics of a specific mineral within the rock. The weathering process of this mineral forms a positive feedback loop with the development of rock porosity and permeability, thereby exerting primary control over the formation of weathering profiles [44]. Through the limestone weathering profile (bedrock layer→weathered layer→soil layer), calcite content and sample porosity exhibit a transition from gradual reduction to abrupt decline (Figure 8a–c), demonstrating significant dissolution-alteration of this primary constituent mineral with progressive weathering [45,46]. The soil layer at different depths contains abundant illite pseudomorphs after K-feldspar (Figure 8d,e), along with observed K-feldspar grains exhibiting embayed dissolution textures (Figure 8f). These features collectively indicate relatively moderate mineral alteration during subsequent weathering stages, which correlates well with the intermediate CIA values recorded in the soil layer (Figure 6d).

As shown in Figure 9, the data points of limestone weathering profile samples plot along the evolutionary trends defined by Upper Continental Crust (UCC), North American Shale Composite (NASC), and Post-Archean Australian Shale (PAAS), demonstrating a characteristic continental weathering pattern [35,47]. The weathering trend of the samples shows near-parallel alignment with the A-CN axis (Figure 9), demonstrating preferential leaching of Ca and Na with concomitant Al enrichment during limestone weathering. K remains essentially stable during incipient weathering stages but shows slight depletion under moderate weathering conditions, consistent with the characteristic weathering sequence of early-stage Ca and Na removal followed by intermediate-stage K loss [48]. Under the influence of the five fundamental soil-forming factors (parent material, climate, time, biota, and topography), rocks undergo continuous weathering and soil formation, accompanied by progressive mineralogical transformations [5]. The dissolution of feldspars typically yields either illite or kaolinite [18,41,46]. Thermodynamic principles demonstrate that K-feldspar dissolution exhibits a higher Gibbs free energy change (ΔG) compared to plagioclase. Furthermore, for both feldspar types, the alteration pathway forming illite is thermodynamically favored over kaolinite formation, as evidenced by lower ΔG values [49]. Consequently, K-feldspar exhibits greater preservation potential than plagioclase during weathering processes, while feldspar dissolution preferentially leads to illite formation [18]. These thermodynamic principles are fully consistent with the mineralogical evolution observed in our weathering profile samples (Figure 3 and Figure 8d,e).

The Sr isotopic variations observed in our weathering profile provide further evidence for assessing the intensity and progression of limestone chemical weathering. Bedrock layer samples exhibit significantly higher Sr concentrations (761–4700 ppm) compared to the soil layer (62.4–428 ppm) (Table S3; Figure 7f), demonstrating substantial Sr mobilization during pedogenesis. The ⁸⁷Sr/⁸⁶Sr ratios of soil samples in our study profile range from 0.725574 to 0.729201 (excluding sample 24SM01; Table S4), which are significantly more radiogenic than those of acid-insoluble fractions from Loess Plateau loess [15,50] (0.718235–0.719673). This distinct isotopic signature, combined with the profile’s remote location from the Loess Plateau and its summit residuum, demonstrates that the Sr isotopic composition was not significantly affected by external sources (e.g., loess dust, atmospheric deposition, or anthropogenic inputs). These results therefore reliably reflect the in situ geochemical evolution of the weathering profile. Sr predominantly occurs in carbonate and silicate minerals through isomorphous substitution for Ca and Mg [51]. Carbonate minerals typically exhibit higher Sr concentrations but lower ⁸⁷Sr/⁸⁶Sr ratios compared to their silicate counterparts [15], while demonstrating greater weathering susceptibility [52]. Consequently, during progressive weathering, Sr hosted in carbonates is preferentially mobilized, whereas silicate-hosted Sr becomes enriched in residual phases. This mechanistic understanding aligns with the observed stratigraphic trend of increasing ⁸⁷Sr/⁸⁶Sr ratios toward the soil surface (Figure 10a), reflecting the cumulative effects of differential mineral weathering. The surface soil sample (24SM01) shows significantly depleted ⁸⁷Sr/⁸⁶Sr ratios (0.709230) due to combined humus complexation, surface denudation, and infiltration erosion effects (Figure 10a). In addition, the correlation between Sr content and ⁸⁷Sr/⁸⁶Sr ratios can also be used to characterize weathering processes in the profile [15]. In the soil layer, Sr exhibits a significant negative correlation with ⁸⁷Sr/⁸⁶Sr ratios (Figure 10b), further indicating that the increase in ⁸⁷Sr/⁸⁶Sr ratios in weathering products results from the leaching of Sr with lower ⁸⁷Sr/⁸⁶Sr values. Furthermore, the Rb/Sr ratio can be employed to investigate the mobilization characteristics of Sr during weathering processes. Rb can substitute for K in K-feldspar through isomorphous replacement, while Sr typically replaces Ca in calcite via isomorphic substitution [15,53,54]. Under comparable conditions, Rb-rich K-feldspar generally exhibits greater resistance to weathering than Sr-bearing calcite, with calcite typically possessing lower ⁸⁷Sr/⁸⁶Sr ratios than K-feldspar [15,55,56]. In the studied samples, the Rb/Sr ratio progressively increases with intensified weathering and shows a positive correlation with ⁸⁷Sr/⁸⁶Sr ratios (Figure 10c), demonstrating that the elevated ⁸⁷Sr/⁸⁶Sr ratios in weathering products result from the preferential dissolution of calcite with lower ⁸⁷Sr/⁸⁶Sr values.

5.2. Elemental Mobilization and Redistribution

The variation of chemical weathering intensity with depth in rock masses can be elucidated by comparing the mineralogical and elemental compositions between weathered samples and bedrock [57,58]. During the weathering of bedrock into soil, the leaching of mobile elements leads to a relative enrichment of immobile elements. Therefore, absolute element concentrations cannot accurately reflect the leaching or enrichment processes occurring during rock weathering. To eliminate this effect, an immobile reference element is typically selected to calculate the migration rate of target elements relative to the reference element, thereby characterizing the depletion or enrichment of minerals due to leaching, migration, and redeposition [18]. Previous studies have demonstrated that Zr is predominantly enriched in weathering-resistant accessory minerals and exhibits extremely low mobility, making it a commonly used immobile reference element for mass balance calculations in limestone systems [46,59,60].

Figure 11 illustrates the depth-dependent variations of τ values for both LOI and major elements in the studied weathering profile. These components can be categorized into four distinct groups based on their similar τ value evolution patterns during weathering, reflecting differential geochemical behaviors. All τ values of Na₂O and MgO fall within the range of 0–8 (Figure 11a), indicating only weak enrichment of these elements during limestone weathering. The extremely low abundances of plagioclase and dolomite in the bedrock (Figure 3b,f) suggest that this phenomenon results primarily from the dissolution of calcite during weathering [15]. The τ values (0–50) of SiO₂, TiO₂, K₂O, and TFe₂O₃ exhibit minimal variation throughout the bedrock and weathered layers, followed by a pronounced increase in the soil layer (Figure 11b), reflecting pedogenic enrichment of weathering-resistant quartz and rutile, authigenic illite formation, and oxidative precipitation of Fe²⁺ to Fe³⁺ oxides/hydroxides [5,18,61]. These geochemical signatures are corroborated by EPMA-observed mineralogical transformations (Figure 4 and Figure 8). Similarly, Al₂O₃ τ values (0–350) display subtle fluctuations in bedrock and weathered layers but dramatic enrichment in the soil layer (Figure 11c), unequivocally indicating clay mineral neoformation and associated Al sequestration [18,62] (Figure 3d and Figure 8d,e). The τ values (−1–7) of LOI and CaO display identical trends in the bedrock and weathered layers, with LOI slightly higher in the soil layer (Figure 11d). This implies that the decrease in LOI during weathering is primarily attributed to carbonate decomposition, while the higher LOI in the soil layer reflects increased crystalline water and organic matter [63,64]. Notably, negative τ anomalies of SiO₂, TFe₂O₃, K₂O, TiO₂, and Al₂O₃ at dissolution pores (Figure 11b,c) likely reflect element leaching by groundwater during limestone karstification [15,65], consistent with the observed low CIA values at these locations (Figure 6d).

As weathering and leaching progress, selective fractionation and enrichment of trace elements occur in the profile [66,67]. The depth-dependent variations in τ values of trace elements in the studied profile are illustrated in Figure 12. Based on similar τ value trends during weathering, the geochemical behavior of trace elements can be classified into six distinct categories. The τ values (0–12) of Co, Hf, and U exhibit irregular fluctuations throughout the weathering profile (Figure 12a), indicative of their secondary enrichment during limestone weathering. These elements predominantly occur in isomorphic substitutions within pyrite, zircon, and monazite—accessory minerals characterized by remarkable chemical stability and weathering resistance [61,68]. A distinct geochemical behavior is observed for Zn, Cu, Pb, Li, and Mo, which demonstrate progressively increasing τ values (0–27) with oscillatory upward trends along the weathering profile (Figure 12b), suggesting enhanced enrichment during weathering processes. In contrast, Sr displays a slight oscillatory decrease in τ values (−1–0.1) from bedrock to weathered layer, followed by dramatic depletion in the soil layer (Figure 12c). This pattern reflects Sr’s high geochemical mobility during pedogenesis, attributable to its isomorphic incorporation in calcite and other Ca-bearing minerals that undergo preferential dissolution during weathering [15]. Notably, Ba exhibits a subtle oscillatory increase in τ values (0–7) in the lower profile, with abrupt enrichment in the soil layer (Figure 12d). This pedogenic enrichment likely results from the precipitation of Ba²⁺ with sulfate ions (SO₄²⁻) generated through pyrite oxidation during weathering [69]. Similarly, the τ values (0–45) of Ta, Cr, and Nb exhibit a slight oscillatory increase from bedrock to weathered layer, followed by a sharp enrichment in the soil layer (Figure 12e). These elements are primarily hosted in highly refractory minerals such as chromite and Nb-Ta oxides [70,71]. The τ values (0–180) of Ni, Rb, V, Tl, Th, and As show a similar pattern of moderate oscillatory enrichment from bedrock to weathered layer, culminating in dramatic enrichment within the soil layer (Figure 12f). The systematic enrichment of most trace elements in the soil layer (Figure 12b,d–f) highlights the critical role of pedogenic processes in element redistribution [15,18]. Previous studies have well documented the strong adsorption capacity of clay minerals for trace elements [72,73,74]. In our study, the abundant presence of illite in soil samples (Figure 3d and Figure 8d,e) strongly suggests that illite adsorption represents the dominant mechanism controlling the exceptional enrichment of trace elements in the soil layer of this limestone weathering profile.

5.3. Pedogenesis Model of Weathering Shimen Limestone

The Shimen limestone weathering profile exhibits unique mineralogical and geochemical signatures while sharing evolutionary trends with crystalline rock weathering systems. From fresh bedrock to the lower soil layer, a sharp lithological discontinuity occurs at the rock–soil interface due to the extensive dissolution of calcite, the primary mineral in the protolith. Calcite is almost entirely depleted, while quartz, feldspar, and illite show proportional enrichment (Figure 3a–e). With decreasing soil depth, the contents of quartz and feldspar also decline (Figure 3a–c), reflecting mineral evolution patterns typical of crystalline rock weathering [74,75]. Since the limestone soil layer primarily forms through the residual accumulation of acid-insoluble residues and their subsequent weathering, its development is strongly influenced by the original composition of these insoluble components [76]. The acid-insoluble fraction in limestone mainly consists of relatively stable primary silicate minerals (e.g., quartz) and illite (Figure 3). Most unstable ferromagnesian dark silicate minerals are either altered or decomposed into more stable secondary phases before deposition. Even if a small fraction of dark minerals is transported to the sedimentary basin, they are likely transformed into more stable secondary minerals during diagenesis [15,77]. Consequently, the primary material (acid-insoluble residues) contributing to limestone soil formation predominantly comprises minerals that are stable under surficial conditions. This unique material composition leads to distinct pedogenic characteristics in the limestone weathering profile: when residual soils undergo further weathering, the process progresses relatively slowly, resulting in low weathering rates and weak geochemical differentiation throughout the profile. This is evidenced by minimal variation in the CIA values (Figure 6d), contrasting sharply with the pronounced weathering differentiation observed in crystalline rock profiles containing more unstable minerals [75,78]. In the studied limestone weathering profile, major element pairs (SiO₂-Al₂O₃, SiO₂-TFe₂O₃, TiO₂-Al₂O₃, and TiO₂-TFe₂O₃) in bedrock and weathered layers show highly clustered distributions. These data points, together with those from the soil layer, form a binary mixing trend in bivariate plots (Figure 13), indicative of a two-endmember linear distribution pattern.

The residual accumulation model for soil formation on carbonate bedrock, while theoretically established, has faced persistent challenges due to the scarcity of direct field evidence. Carbonate rocks typically contain minimal acid-insoluble residues (<5% in most South China karst regions), resulting in extreme volumetric reduction (>90%) during pedogenesis that completely obliterates protolith structures and transitional weathering zones—critical evidence required to verify in situ soil formation [15,65,76]. The Shimen limestone weathering profile provides conclusive field demonstration of autochthonous pedogenesis through preserved mineralogical and geochemical signatures. Systematic vertical variations in mineral assemblages (calcite dissolution and secondary clay formation) and consistent immobile element ratios (e.g., Ti/Nb, Zr/Hf) establish genetic continuity between bedrock and overlying soil, despite the absence of visible transitional layers. In typical South Chinese carbonate sequences with <5% insoluble residues, while primary structures are not preserved [15], diagnostic mineralogical indicators (e.g., authigenic feldspar-illite succession) and geochemical fingerprints (conservative element correlations) robustly demonstrate residual origins (Figure 9 and Figure 13).

Here, we propose a two-stage weathering model for limestone-derived soils (Figure 14). The first stage involves progressive accumulation of acid-insoluble residues from bedrock weathering to form residual soils, while the second stage features soil evolution processes analogous to those observed in non-carbonate rock weathering systems. In the Shimen limestone profile, rapid dissolution of primary calcite creates an abrupt rather than gradual transition during the initial stage, suggesting this phase can be geologically instantaneous [10,16]. The dominant mineralogical and geochemical signatures preserved in the Shimen profile actually represent the second weathering stage, exhibiting evolutionary patterns strikingly similar to those in crystalline rock weathering systems. This observation provides robust validation for the proposed two-stage pedogenic model.

5.4. Implications for Weathering Geochemistry in Carbonate Systems

Minerals exert significant control over the formation of limestone weathering profiles and the geochemical behavior of elements; consequently, mineral alteration processes are key to understanding limestone weathering [79]. The weathering of minerals is driven by their thermodynamic instability under surface conditions, constituting a spontaneous and irreversible thermodynamic process [4,49]. Because the reaction kinetics responsible for disrupting these mineral structures are relatively slow, these weathering-sensitive minerals are commonly observed undergoing alteration within surface environments [77,80]. It is widely acknowledged that limestone weathering initiates with the oxidative dissolution of calcite [10,77,79]. Based on the observed variations in mineral composition, the migration and redistribution of major and trace elements, and Sr isotope signatures within the weathering profile, we propose that the primary geochemical processes governing limestone weathering encompass: (1) the dissolution of calcite, (2) the decomposition of feldspar, and (3) the formation of illite. During the initial stage of weathering, dissolution of the soluble mineral calcite commences, resulting in the loosening of mineral cementation and an increase in porosity and pore size within the weathering profile [79]. In subsequent phases of weathering, alteration of the more resistant mineral K-feldspar leads to illite formation [5]. Additionally, water–rock interactions between CO₂-rich shallow groundwater and near-surface limestone generate nodular dissolution pores (Figure 2a).

Our results demonstrate that significant dissolution of calcite within the soil layer induces a pronounced decrease in SIC (Figure 3e and Figure 6a), thereby indicating that limestone weathering and pedogenesis in the Shimen area result in a carbon source effect. Throughout the weathering process, distinct mineral–element coupling relationships are observed, with illite playing a crucial role in the secondary enrichment of trace elements. Consequently, elucidating the geochemical processes governing limestone weathering is fundamental to understanding material transport and cycling within the Earth’s critical zone. Furthermore, we hypothesize that carbonate units exhibiting higher insoluble residue contents (>15%) retain discernible protolith features, including relict bedding structures and gradual mineralogical transitions observable in micromorphological analysis. This conceptual framework reconciles the theoretical model with empirical evidence, providing a mechanistic understanding of carbonate weathering processes across different geological settings. Thus, the Shimen weathering profile serves as a key reference for interpreting pedogenic origins in global karst terrains.

6. Conclusions

Based on a comprehensive mineralogical and geochemical investigation of an in situ limestone weathering profile in Shimen, Hunan, this study establishes a two-stage pedogenic model: initial rapid calcite dissolution creates an abrupt bedrock–soil interface through volumetric collapse (>90%), accumulating acid-insoluble residues (dominantly quartz); subsequent weathering of this residue drives progressive illitization of K-feldspar, trace element enrichment via illite adsorption, and radiogenic ⁸⁷Sr/⁸⁶Sr evolution. These processes, recorded by upward-increasing CIA (6.79–79.96) and elemental mass transfer coefficients, confirm in situ soil genesis despite minimal insoluble residues (<5%) and identify regional weathering as a net carbon source (58.5% SIC depletion), providing mechanistic insights into subtropical critical zone evolution and element cycling in karst terrains.