Introduction

Clay minerals are among the essential components of the solid soil phase, and their composition and relative content can affect many physical and chemical properties and the nutrient status of the soil1. The composition and relative content of clay minerals depend on the parent material, climate, and micro-topography2,3. In addition, clay mineral composition differs depending on the degree of pedogenesis4.

Clay minerals are formed through the neoformation and transformation of primary minerals derived from the weathering product of the parent material5. The mineral content and texture of parent materials can influence the formation of clay minerals6,7; therefore, the composition and characteristics of clay minerals are closely related to those of the parent materials. The pedogenic environment can also substantially influence the transformation of clay minerals through temperature and precipitation2. 1:1-type kaolinite is the dominant clay mineral in subtropical and tropical areas with high air temperature3,8,9,10. In contrast, 2:1-type clay minerals are readily detected in temperate and cool-temperate zones11,12. Clay mineral types constantly change as the soil weathers and develops. He et al.13 reported that young soils showed an initial illite-formation stage, and that the content of 2:1-type clay minerals decreased over time in the soil chronosequence. Several studies have indicated that there are significant differences in clay mineral composition and content among different soil genetic horizons and soil types14,15. Therefore, studies on the effects of different parent materials and pedogenic environments on the composition and content of clay minerals and their dynamic variation among different soil types are of theoretical importance. Several studies have focused on the influence of a single environmental factor, such as parent material or climate type, on the composition of clay minerals8,9,10,16. In some studies, authors have used only one classification system, such as the World Reference Base for Soil Resources (WRB)17, Soil Taxonomy (ST)18, or Chinese Soil Taxonomy (CST)19, to examine the variation in clay mineral composition among different soil types20,21,22,23. A few studies have investigated the variation in the composition of clay minerals derived from diverse parent materials (more than three parent materials) and the effects of comprehensive environmental factors (e.g., elevation and micro-topography) on clay minerals in the subtropics. However, the dynamic changes in clay minerals among different soil types based on multiple soil classification systems (CST, ST, and WRB) have received limited attention.

Owing to its diversity of parent materials, complex topography, and changeable climate, the subtropical region of China has diverse soil types24,25. Desilication and ferrallitisation are common in climates with high air temperatures and precipitation, and this results in the formation of a LAC-ferric horizon (lower-activity clay and free iron oxide rich) in Ferrosols. Clay migration is active in soils characterised by strong eluviation, and it occurs in soils originating from coarse parent materials. These soils easily develop an argic horizon (clay illuviation), which is necessary for Argosols formation. In areas with low air temperature, parent materials tend to be weakly weathered. The soils derived from these parent materials can form a cambic horizon (slight siallitisation), which is necessary for Cambosols formation. On steep slopes, Primosols are common due to the dynamic balance between soil formation and erosion25,26.

Hunan Province is home to an ideal series of soils that enable investigation of the effects of parent materials (more precisely, five parent materials) and topography (plains, hills, and mountains) on the composition of clay minerals under warm and humid subtropical climate. The parent rocks or materials in the upland soils of the Hunan Province mainly consist of Quaternary red clay (QRC), granite (GR), slate and shale (SS), sandstone (SDS), and limestone (LS)27. Thus, the pedogenic processes of upland soils vary widely because of the diversity of parent materials and the complex topography. It has been reported that the variation in clay mineral composition of Stagnic Anthrosols (paddy soil) is a consequence of diverse parent materials present in Hunan Province23; however, there have only been a few studies on the upland soils (natural soils) in this area. Therefore, in the present study, we selected 60 upland soils derived from five parent materials in the Hunan Province and analysed the effect of different parent materials, soil environments, diagnostic horizons, and soil types on clay mineral types and contents using X-ray diffraction. The goals of this study were to (i) investigate the effects of diverse parent materials and environmental factors (elevation and micro-topography) on soil clay mineral composition and (ii) explore the evolution of clay minerals during soil development and its implications for soil taxonomy.

Materials and methods

Study area and field sampling

This study was conducted in Hunan Province, south-central China (Fig. 1). Hunan Province is located in the middle reaches of the Yangtze River, bordering Dongtinghu Lake in the north. It covers a land area of 211,800 km2, characterised by the presence of hills and basins in the centre, mountains in the east, south, and west, and plains in the north. Hunan has a mid-subtropical, seasonally humid climate characterised by moderate air temperature, abundant precipitation, and clear differentiation between wet and dry seasons. The annual sunshine duration is 1300–1800 h, and the annual average air temperature is 16–18 °C. The area is frost-free 260–310 days a year, and the annual precipitation ranges from 1200 to 1700 mm27. Sixty sites were selected, covering elevations from 41 to 650 m a.s.l. (Table 1). Parent materials of the soils in the Hunan province mainly included granite (GR), slate and shale (SS), Quaternary red clay (QRC), limestone (LS), and sandstone (SDS). The formation period of GR consists of three geological times, namely the Caledonian, Indosinian, and Yanshanian ages. The primary minerals in GR include feldspar, quartz, mica, and amphibole. SS are formed from moderately metamorphic slate and sedimentary shale. The main slate rocks are argillaceous slate, siliceous slate, and silt slate. Shale is mainly composed of carbonaceous shale, silt shale, and sandy shale. The LS in this region is mainly composed of marine sedimentary carbonate rocks, and its stratigraphic chronology mainly runs from the Devonian to the Permian period. With respect to minerals, LS is mainly composed of calcite and often contains mixed penetrations (sand, clay, dolomite, and silica). Soils developed from QRC, which is formed from glaciofluvial deposits dissolved in the Quaternary interglacial period of the Cenozoic era, are usually sticky due to intense mineral weathering and heavy material illuviation. The SDS in the area mostly consists of marine sedimentary clastic rocks and mainly includes siltstone, sandstone (quartz sandstone, feldspar quartz sandstone, and argillaceous sandstone), and conglomerate. SDS contained minerals, including quartz sand, feldspar sand, and iron siliceous cement27.

Figure 1
figure 1

Study area and sampling sites.

Full size image
Table 1 Fundamental information about study sites and soil classification.
Full size table

We referred to The Manual of Soil Description and Sampling for collecting and describing the soil samples28. The soils were classified as Ferrosols, Argosols, Cambosols, or Primosols using the CST classification system; as Ultisols, Alfisols, Inceptisols, and Entisols using the ST; and as Acrisols, Lixisols, Alisols, Luvisols, Plinthosols, Umbrisols, Cambisols, and Leptosols according to WRB17,18,19. These soil classifications are consistent with those of previous studies that investigated correlations among the CST, ST, and WRB systems25,29. To avoid human interference and fully reflect the nature of the soil and the inheritance of parent materials, only the weathering B horizons in the soil profiles were sampled.

Soil mineralogy

The collected soil samples (clay fraction < 2 μm) were characterised using an X-ray diffractometer (Model D/Max–rA; Rigaku, Tokyo, Japan) with the following parameters: radiation, Ni-filtered CuKα; voltage, 40 kV; current, 40 mA. The diaphragm system was set as follows: divergence slit (DS) = anti-scatter slit (SS) = 1° and receiving slit (RS) = 0.3 mm. It measured from 3° to 30° 2θ at a scan rate of 2° 2θ min−1 and a step size of 0.02° 2θ. For preparing samples for orientation sampling, the clay fraction samples were air-dried (AD), saturated with ethylene–glycol (EG) at 70 °C for 3 h, and heated (HT) to 450 °C or 600 °C for 2.5 h. The X-ray patterns were analysed using XPowder, a software package for powder X-ray diffraction analysis. Kaolinite was identified based on the presence of 0.72 and 0.358 nm peaks after being AD and saturated with EG, and the 0.72 nm peak disappeared after heating to 600 °C. Then, illite was identified based on the presence of a 1.00 nm peak after being AD and saturated with EG, and the peak persisted after heating. Vermiculite and chlorite were distinguished by heat treatment. Shrinkage of the 1.42 nm peak to 1.00–1.03 nm after heating indicated the presence of vermiculite. I/S was identified based on the presence of a 1.45–1.54 nm peak, which slightly expanded after saturation with EG and shifted to 1.00 nm after heating. I/V was determined based on an unaltered D1 zone at 1.00–1.42 nm and appeared in the D3 zone at 0.50–0.47 nm; in the D2 zone, it shifted to < 1.0 nm after heating. The relative content of clay minerals was calculated based on the height of the diffraction peak30,31,32,33,34,35.

Soil physicochemical properties

Soil total K content was determined after digesting the samples with solid NaOH in a silver crucible at 450 °C for 15 min, and then gradually increasing the temperature. The concentration of K was determined using a flame photometer. Soil texture was determined using the pipette method and classified using the United States Department of Agriculture (USDA) classification standards. Soil organic carbon (SOC) was measured using the K2Cr2O7 wet oxidation method. Soil pH was measured using a glass electrode placed in a solution of soil and distilled water (1:2.5). The cation exchange capacity (CEC) and exchangeable Ca, Mg, K, and Na were determined using 1 M NH4OAc at pH 7. In each sample, the Fed was extracted using sodium dithionite-citrate. The total concentration of Fe, Al, and Si was determined by treating samples with Li2CO3-H3BO3 after heating at 900 °C for 0.5 h, and then, the solution was analysed by inductively coupled plasma atomic emission spectrometry (ICP-AES) (Table 2)36. CEC/clay was calculated using the following equation19:

Table 2 Physicochemical properties of upland soils derived from different parent materials in Hunan Province, China.
Full size table
$$\mathrm{CEC}/\mathrm{clay } \; \left({\text{cmol} \; \text{kg}}^{-1} \; \mathrm{ clay}\right)=\frac{\mathrm{CEC} \, (\mathrm{by }\;\; \text{NH}_{4}\,\mathrm{OAC }\,{pH}_{7}) \, ({\mathrm{cmol} \; \text{kg}}^{-1} \; \mathrm{ soil})}{\mathrm{clay} \, ({\mathrm{g} \; \text{kg}}^{-1})}\times 1000$$

Statistical analysis

The data were processed and analysed using SPSS 22.0 (IBM, Chicago, IL, USA). Before performing one-way ANOVA, the physicochemical properties of soils derived from different parent materials were tested for normal distribution and variance homogeneity using Kolmogorov–Smirnov and Levene test, respectively. Pearson’s correlation coefficient was used to test the correlation between clay minerals and the physical and chemical properties of the soil. The least significant difference test was used to test for significance. Significant differences are denoted by *P < 0.05, and extremely significant differences are indicated with **P < 0.01. All figures were created using Origin 9.1 software.

Results

Composition and relative content of clay minerals in different parent materials

The clay minerals in upland soils were composed of 1:1-type clay minerals kaolinite (0.71–0.73 nm, 0.35–0.36 nm), 2:1-type clay minerals illite (1.0–1.03 nm), vermiculite (1.42–1.49 nm shifting to 1.0–1.03 nm after heating), and mixed-layer minerals illite/vermiculite (I/V) (determined by the unchanged D1 zone at 1.00–1.42 nm, appeared in the D3 zone at 0.50–0.47 nm and shifted to < 1.0 nm in the D2 zone after heating), and illite/smectite (I/S) (1.45–1.66 nm shifted to 1.0–1.02 nm after heating) (Fig. 2). The I/S mixed-layer minerals were only revealed in the plinthic horizon derived from QRC.

Figure 2
figure 2

X-ray diffraction patterns of the representative upland soil profiles derived from different parent materials (K: kaolinite; I: illite; V: vermiculite; I/S: illite/smectite mixed-layer mineral; I/V: illite/vermiculite mixed-layer mineral; Q: quartz; AD: air-dried sample; EG: ethylene–glycol saturated sample; HT 450 °C: sample heated at 450 °C; HT 600 °C: sample heated at 600 °C; SS: slate and shale; LS: limestone; QRC: Quaternary red clay; GR: granite; SDS: sandstone).

Full size image

The relative content of clay minerals varied among upland soils derived from different parent materials (Fig. 3). In GR, the relative content of kaolinite was the highest (72%), and the content of illite and vermiculite were the lowest (12% and 3%, respectively). The kaolinite content (54%) in LS was slightly lower than that in GR, whereas the 2:1-type illite (20%) and vermiculite (16%) content in LS were higher than those in GR. The upland soils in LS contained a small amount of I/V mixed-layer minerals (10%). Kaolinite (39%), illite (28%), and I/V mixed-layer minerals (26%) were predominant in QRC, whereas vermiculite was less dominant (4%). In SS, the kaolinite and illite content varied substantially. The kaolinite content (43%) in SS was higher than that in QRC, but the content of illite (20%) and mixed-layer minerals (24%) was lower. In SS, the relative content of vermiculite (13%) was lower than that of other clay minerals. In SDS, the content of 1:1-type kaolinite (21%) was the lowest, and the content of 2:1-type illite (34%) and vermiculite (22%) were the highest. The content of I/V mixed-layer minerals (23%) in SDS was relatively higher than that in GR and LS.

Figure 3
figure 3

Relative content of clay minerals of studied upland soils derived from different parent materials (Kao: kaolinite; It: illite; V: vermiculite; I/S: illite/smectite mixed-layer mineral; I/V: illite/vermiculite mixed-layer mineral; GR: granite; LS: limestone; QRC: Quaternary red clay; SS: slate and shale; SDS: sandstone).

Full size image

Differences in clay mineral composition and relative content among diverse pedogenic environments

Correlation analysis revealed that clay mineral composition was correlated with elevation (Fig. 4). For QRC, the range of elevation was small, and the elevation difference was less than 70 m; therefore, no correlation analysis was conducted. Apart from QRC, the kaolinite content in other parent materials was positively correlated with elevation (SS: r = − 0.68, LS: r = − 0.66, GR: r = − 0.85, SDS: r = − 0.66), but an opposite trend was observed for 2:1-type illite (SS: r = 0.74, LS: r = 0.65, GR: r = 0.67, SDS: r = 0.60). Nevertheless, the magnitude of variation at distinct elevations differed among the parent materials. The change in the slope of clay minerals in LS (kKao = − 0.13, k2:1-type = 0.08) was higher than that in the other three parent materials. This indicated that with increasing elevation, the kaolinite content decreased, whereas the 2:1-type illite content increased. Compared to the clay minerals of other parent materials, the clay minerals of LS were more sensitive to variations in elevation.

Figure 4
figure 4

Relationships between relative content of clay minerals and elevation for the studied soils (a,b: slate and shale; c,d: limestone; e,f: granite; g,h: sandstone).

Full size image

In addition to the correlation between elevation and the composition and relative content of clay minerals, we also observed a correlation between topographic position and clay minerals (Fig. 5). By comparing the topographic position with the composition and content of clay minerals, we found that the samples with high quantities of mixed-layer minerals and 2:1-type clay minerals were mostly collected from the slope foot or toe, whereas soil samples with a high 1:1-type kaolinite content were mostly collected from the slope shoulder or crest.

Figure 5
figure 5

Relationships between relative contents of clay minerals and toposequence for the studied soils (Kao: kaolinite; It: illite; V: vermiculite; I/S: illite/smectite mixed-layer mineral; I/V: illite/vermiculite mixed-layer mineral).

Full size image

Differences in the composition and relative content of clay minerals at different diagnostic horizons and soil types (i.e., soil orders)

We identified 60 soil profiles with four diagnostic horizons based on the CST, i.e., LAC-ferric horizon, argic horizon, cambic horizon, and plinthic horizon19. The 60 soil profiles were classified as Ferrosols, Argosols, Cambosols, and Primosols in CST19; as Ultisols, Alfisols, Inceptisols, and Entisols in ST18; and as Acrisols, Lixisols, Alisols, Luvisols, Plinthosols, Umbrisols, Cambisols, and Leptosols in WRB17. The clay mineral composition was significantly different among the diagnostic horizons and soil types (i.e., soil orders) (Figs. 6, 7). The dominant clay mineral in the LAC-ferric horizon was 1:1-type kaolinite (63%), followed by illite (17%), I/V mixed-layer minerals (12%), and a small amount of vermiculite (8%). In the plinthic horizon, the relative content of mixed-layer minerals, including I/V (27%) and I/S (20%) mixed-layer minerals, was the highest (47%). The plinthic horizon also contained a high percentage of kaolinite (29%) and illite (23%) and a small amount of vermiculite (2%). The differences in clay mineral composition were negligible in the argic and cambic horizons.

Figure 6
figure 6

Relative content of clay minerals of different diagnostic horizons (Kao: kaolinite; It: illite; V: vermiculite; I/S: illite/smectite mixed-layer mineral; I/V: illite/vermiculite mixed-layer mineral).

Full size image
Figure 7
figure 7

Relative contents of clay minerals of different soil types (i.e., soil orders) (Kao: kaolinite; It: illite; V: vermiculite; I/S: illite/smectite mixed-layer mineral; I/V: illite/vermiculite mixed-layer mineral; CST: Chinese Soil Taxonomy; ST: Soil Taxonomy; WRB: World Reference Base for Soil Resources).

Full size image

Among all the soil types (i.e., soil orders), the content of 1:1-type kaolinite was the highest in Ferrosols (63%), followed by 2:1 clay minerals (I + V: 25%) and I/V mixed-layer minerals (12%). In Primosols, the content of 2:1-type clay mineral (I + V: 46%) and I/V mixed-layer mineral (36%) were the highest, and the content of 1:1-type clay mineral kaolinite (18%) was the lowest. The variation in clay mineral composition in Argosols and Cambosols was negligible. However, the content of mixed-layer minerals (I/V + I/S) was the highest in Plinthic Ali-Udic Cambosols (40%). Similar results were obtained using the ST and WRB classification systems. The content of 1:1-type kaolinite was the highest in Ultisols (51%) and Acrisols (63%), and the lowest in Entisols (18%) and Leptosols (17%). Mixed-layer minerals (I/V + I/S) were predominant in Plinthudults (47%) and Plinthosols (41%). This indicated that the clay mineral composition and relative content have a strong effect on Ferrosols, Ultisols, and Acrisols (high development degree); Primosols, Entisols, and Leptosols (low development degree); and Plinthic Ali-Udic Cambosols, Plinthudults, and Plinthosols (strong redox status).

Correlation between clay minerals and soil properties

Correlations between the composition and relative content of clay minerals with soil properties—such as SOC, CEC, soil texture, and pH—were determined (Fig. 8). Pearson’s correlation analysis showed that the content of 1:1-type kaolinite was positively correlated with the sand content (r = 0.27, P < 0.05), significantly and positively correlated with aluminium oxide and exchangeable Na (rAl = 0.53, P < 0.01; rNa = 0.34, P < 0.01), and significantly and negatively correlated with silt content, SOC, and silicon oxide (rsilt = − 0.53, P < 0.01; rSOC = − 0.43, P < 0.01; rSi = − 0.34, P < 0.01). The content of 2:1-type illite was negatively correlated with aluminium oxide and exchangeable Na (rAl = − 0.32, P < 0.05; rNa = − 0.30, P < 0.05) and significantly and positively correlated with silt content (r = 0.42, P < 0.01). Vermiculite content was positively correlated with SOC and pH (rSOC = 0.59, P < 0.01; rpH = 0.35, P < 0.01) and negatively correlated with potassium oxide and exchangeable Na (rK = − 0.37, P < 0.01; rNa = − 0.40, P < 0.01). The I/V mixed-layer mineral content was positively correlated with silicon oxide (r = 0.50, P < 0.01) and negatively correlated with aluminium oxide and iron oxide (rAl = − 0.53, P < 0.01; rFe = − 0.46, P < 0.01). Additionally, 1:1-type kaolinite was significantly negatively correlated with 2:1-type illite, vermiculite, and I/V mixed-layer minerals (rI = − 0.74, P < 0.01; rV = − 0.47, P < 0.01; rI/V = − 0.62, P < 0.01). Illite showed a positive relationship with I/V mixed-layer minerals (r = 0.28, P < 0.05).

Figure 8
figure 8

Correlativity of the clay minerals with some physico-chemical properties in studied soils.

Full size image

Pearson’s correlations between the clay minerals and CEC/clay and Fed are shown in Table 3. Kaolinite was negatively correlated with CEC/clay and positively and significantly correlated with Fed, whereas 2:1-type clay minerals (illite and vermiculite) were negatively correlated with CEC/clay and positively correlated with Fed.

Table 3 Correlation coefficients between the clay minerals and CEC/clay and Fed.
Full size table

Discussion

Effects of parent materials on the composition and relative content of clay minerals

The type of parent material and the degree to which it is weathered have a major influence on the neoformation and transformation of clay minerals. These factors can result in variations in clay mineral composition and content in the same area due to differences in the mineral composition and texture of parent materials7. GR is mainly composed of feldspar, quartz, biotite, and other primary minerals27, and in the present study, the content of sand and K2O in GR was significantly higher than that in the other four types of parent materials (Table 2). Under acidic and strong leaching conditions, feldspar intensely weathers and hydrolyses, and kaolinite can be mass produced by neoformation37. In addition, under conditions of high K content, vermiculation can be inhibited, and biotite can be oxidised and directly weathered to kaolinite38,39. Therefore, the kaolinite content in the GR group was generally high (Fig. 3). QRC is composed of Quaternary fluvioglacial sediments, and during development, it is subjected to both sedimentary-weathering soil-forming processes and the glacial-interglacial cycle40. Unstable factors in the sedimentary-weathering soil-forming process, strong cyclic climates, and environments that alternate between dry and wet seasons can promote illite alteration, a phenomenon that involves the transformation from single phase to mixed-layer minerals16,41,42. Thus, in the present study, the content of mixed-layer minerals in the QRC group was generally high.

Previous studies have shown that the plinthic horizon of QRC was formed on flat, low-lying terrains in a climate that frequently alternated between dry and wet25. Due to frequent fluctuations in groundwater, soil aggregates shrink when dry, a phenomenon that is conducive to the formation of cracks43. When the groundwater level rises, the soil is in a reducing state in which Fe3+ can be transformed into soluble and mobile Fe2+, thus promoting the leaching of Fe oxide from the soil. This process results in the formation of the plinthic horizon, which consists of uniform red soils with white veins and white spots44,45. Smectite is generally formed in tropical and subtropical areas with dry and wet seasons, flat terrain, and poor drainage46. In wet climates with poor drainage conditions, illite transforms into smectite by absorbing Mg2+, but in dry conditions, K is fixed in the layer of smectite and promotes the formation of I/S mixed-layer minerals5,47. Therefore, the environment is conducive to the transformation of illite to smectite, and the I/S mixed-layer minerals were found in the plinthic horizon.

In the present study, SDS was mainly composed of quartz sand, feldspar sand, and other iron and siliceous cements. The soils developed from SDS are characterised by high gravel content and low degree of development, and they occur at high altitudes (> 300 m)27. Therefore, the contents of 2:1-type clay minerals and mixed-layer minerals were the highest in SDS. LS is mainly composed of marine sedimentary carbonate rocks, and its stratigraphic chronology mainly runs from the Devonian to the Permian period27. The soils derived from LS are old and exhibit a high degree of weathering. Therefore, 1:1-type kaolinite was the main clay mineral in the soils derived from LS. The relative content of kaolinite in LS was lower than that in GR because of its high viscosity and weak leaching in LS.

The physical and chemical properties of soils developed from SS differ because the SS in the study area is composed of two kinds of parent rocks, i.e., low-grade metamorphic rock slate and sedimentary shale and as their components are relatively complex27. Leaching was strong in soils developed from sandy slate with a low pH and high silt content, and therefore, the content of 1:1-type kaolinite was high in these soils. However, leaching was weak in soils derived from clay shale with a high pH and high clay content, and therefore, the content of 2:1-type clay minerals and mixed-layer minerals was high in these soils48.

Effects of pedogenic environment on the composition and relative content of clay minerals

Warm and humid climatic conditions in subtropical regions promote soil weathering and development, and the eluviation and deposition of materials result in the development of acidic, fine-grained, strong-weathering soils characterised by desilication and ferrallitisation49,50. These environmental factors mainly contributed to the development of 1:1-type clay mineral kaolinite, and to a certain extent, 2:1-type illite and vermiculite. The mixed-layer minerals were dominated by I/V (Fig. 3). These results were consistent with those of studies on upland soil clay mineral characteristics in other regions in southern China8,9,10,51.

The regional pedogenic environments, especially the elevation and terrain, influence the transformation of clay minerals through temperature and precipitation or via the changes in hydrologic and thermal conditions. In high-altitude conditions with low temperatures, mica primary minerals can form a large amount of illite through depotassication in the mineral interlayer due to weak weathering52. In weak acidic and strong leaching environments, vermiculite can be easily formed when soil K and Mg are lost53,54,55. In high-altitude areas, the low temperatures result in decreased H4SiO40 and ferrallitisation activity, a phenomenon that can inhibit the conversion of 1:1-type kaolinite56. However, in low-altitude areas, transformation of 2:1-type clay minerals to kaolinite is promoted by higher temperatures that also increase ferrallitisation and H4SiO40 activity56,57,58. The transformation of clay minerals can also be affected by erosion in high-altitude areas. In areas at high altitudes or with steep topography, 1:1-type kaolinite is strongly eroded, whereas 2:1-type clay minerals are continuously created from parent materials59. In the present study, the content of 1:1-type kaolinite in soils derived from LS, GR, SS, and SDS decreased with increasing elevation, whereas the content of 2:1-type clay mineral showed an opposite trend (Fig. 4). These results were similar to those observed for low mountains and hills (< 1500 m)59,60, but different from those found for subalpine or alpine zones (> 1500 m)61. However, the varying gradient of clay mineral content with changing elevation differed in diverse climatic zones, even in the same low mountain/hill area. Based on the findings of previous studies on clay minerals in different climatic regions60,62, we established a series of regression models to analyse the influence of elevation on the composition and relative content of clay minerals (Table 4). In the present study, the decline in kaolinite content (\(\stackrel{-}{\mathrm{k}}\)= − 0.09) with elevation was greater than that recorded in tropical (k = − 0.03) and temperate oceanic regions (k = − 0.04)60,62. A possible reason for such a result is the relatively stable environmental and climatic conditions that exist in these two regions, which are responsible for the relatively slow material migration and mineral transformation in the soil.

Table 4 Results of the literature survey of correlation between clay mineral (kaolinite) and altitude in different climate areas.
Full size table

Nevertheless, the magnitude of variation of clay minerals content in different elevations varies with the type of parent materials. For LS, its weak alkalinity (pH: 5.5 ± 0.5), high clay content (clay: 650 ± 152 g kg−1), and slight weathering of minerals at high altitudes were conducive to the transformation of the primary mineral to illite via damouritisation and sericitisation52,63. At low altitudes, the high temperature and old stratigraphic chronology of LS (Devonian to Permian) will facilitate the transformation of 2:1-type clay minerals into 1:1-type kaolinite through neoformation or mineral degradation. Therefore, the sensitivity of clay minerals in LS with the variation of altitude was greater than that of the other parent materials.

Micro-topography can affect the transformation of clay minerals in the subsurface horizon by redistributing soil hydrological conditions in the upland soils of the Hunan Province (Fig. 5). The present study showed that a wide distribution of kaolinite in the slope shoulder or crest soils also indicates strong leaching conditions, whereas a high content of 2:1-type clay minerals (illite, vermiculite) and mixed-layer minerals in the slope foot or toe soils suggests frequent wet–dry cycles and weak leaching conditions. These findings were similar to those obtained by Fang et al.51, who reported that kaolinite was most frequently found in divergent-site surface soils, whereas 2:1-type clay minerals mainly existed in convergent-site (water-collecting) surface soils. However, the present study showed that the influence of micro-topographical features on the transformation of soil clay minerals could extend to the subsoil.

Dynamic changes in clay mineral composition among the different soil types (i.e., soil orders)

Soil classification systems, especially quantitative classification systems, such as the WRB, ST, and CST, are based on the theory of pedogenesis, diagnostic horizons, and diagnostic characteristics. Therefore, different soil types can often reflect different pedogenic processes and development stages17,18,19. Clay minerals develop from primary minerals that have undergone weathering and pedogenesis during soil development. Their composition and relative content can reflect the strength of soil weathering and changes in the soil-forming environment. As the degree of pedogenesis increases (i.e., the CEC/clay content decreases and Fed content increases), the kaolinite content increases, whereas the content of 2:1 clay minerals decreases (Table 3). In the ST system, 1:1 clay minerals were mostly found in ultisols and oxisols with a high degree of weathering, and 2:1 clay minerals (smectite) were mainly found in the Vertisols that shrink and expand. Conversely, vermiculite and illite were generally found in soil types with less intense weathering, such as Alfisols, Mollisols, and Aridisols64. Upon using the WRB, the 1:1-type kaolinite, which has low activity, was identified as the main clay mineral type in Ferralsols, whereas 2:1 clay minerals (such as chlorite, smectite, and vermiculite) were mostly found in Luvisols with a high activity65. In the present study, the content of 1:1-type kaolinite was the highest in Ferrosols (CST), Ultisols (ST), and Acrisols (WRB) with moderate ferrallitisation and strong weathering. In contrast, the content of 2:1 clay mineral was the highest in Primosols (CST), Entisols (ST), and Leptosols (WRB) with slight siallitisation and weak weathering, and the content of mixed-layer mineral was the highest in Plinthic Ali-Udic Cambosols (CST), Plinthudults (ST), and Plinthosols (WRB) with a strong redox status (Figs. 6, 9).

Figure 9
figure 9

Schematic representation of the conditions leading to evolution of the clay mineral composition in studied upland soils in Hunan, China (GR: granite; LS: limestone; QRC: Quaternary red clay; SS: slate and shale; SDS: sandstone; CST: Chinese Soil Taxonomy; ST: Soil Taxonomy; WRB: World Reference Base for Soil Resources).

Full size image

Conclusions

The clay mineral composition at the study sites is representative of the upland soils in Hunan Province. Clay minerals mainly included 1:1 kaolinite, which contains a certain amount of 2:1-type clay minerals such as illite and vermiculite and a small amount of mixed-layer minerals. Our results suggested that the parent material, regional pedogenic environment (elevation and micro-topography), and degree of pedogenesis (soil type) play important roles in the transformation of clay minerals. Kaolinite dominated the mineral distribution patterns in GR (sandy) and LS (old), whereas 2:1-type clay minerals were dominant in SDS (clastic). The content of mixed-layer minerals was high in QRC (glacial-interglacial cycles and wet-dry cycles). In low mountains and hills (< 1500 m), neoformation and transformation of clay minerals were determined by elevation and micro-topography. The influence of elevation on clay mineral transformation in the subtropical monsoon region is greater than that in tropical and temperate oceanic regions. Compared with the parent material, elevation showed a greater effect on the transformation of clay minerals in LS. Micro-topography can modulate the transformation of clay minerals in the subsurface horizon. The composition and relative content of clay minerals can efficiently indicate soil types with a high development degree (Ferrosols, Ultisols, and Acrisols), those with a low development degree (Primosols, Entisols, and Leptosols), and those with a strong redox status (Plinthic Ali-Udic Cambosols, Plinthudults, and Plinthosols).

The factors influencing clay mineral composition are complex and diverse. In the present study, only parent material, climate, and micro-topography were investigated, whereas the effects of vegetation type and pedogenic time were not explored. Future research should focus on these important factors to elucidate the mechanisms underlying clay mineral neoformation and transformation and to assess their significance in pedogenesis.