Dynamics of organic matter in soils formed on limestone and sandstone in the karst area in southwestern China: δ 13C and δ 15N approaches

To better understand biogeochemical processes associated with the dynamics of soil organic matter (SOM), the δ 13C and δ 15N values were analyzed in the SOM of limestone soil (developed on limestone) and sandstone soil (developed on sandstone), as well as in dominant plant species and litter samples collected on the studied soil profiles in karst areas of Southwest China. In general, the content of soil organic carbon (SOC) and soil organic nitrogen (SON) is highest in the surface soil layer and decreases downward in both types of soil profiles. In addition, the content of SOM (including both SOC and SON) in the limestone is much higher than in sandstone soil. The higher pH values, as well as the content of calcium, magnesium, and clay minerals, are probably the reasons for the higher SOM content in the limestone soil compared to the sandstone soil. In the SOM of the upper layers of yellow sandstone and limestone soil δ 13C values range from 2.6‰ to 6.3‰ and δ 15N values from 5.1‰ to 8.1‰, which is larger than in litters. Yellow sandstone soils have smaller differences between δ 13C and δ 15N values of SOM from litter and topsoil than the limestone soil. The δ 13C and δ 15N values of SOM increase with depth in yellow sandstone soil profiles, but not in the limestone soil profiles. An increase in the values of δ 13C and δ 15N with a decrease in the SOM content in the yellow sandstone soil significantly substantiates the decomposition of SOM, which causes higher values of δ 13C and δ 15N values in the deeper soil horizons. By contrast, the limestone soil has complex depth profiles of δ 13C and δ15N, which can be mainly attributed to land-use changes and organic and mineral interactions between soil accumulations.


Introduction
The world's soils are an integral part of the global carbon (C) cycle and as a source of atmospheric CO2 can release a large amount of C into the atmosphere every year [1]. SOC stocks in terrestrial ecosystems are almost three times greater than the storage of C in vegetation [2] and twice as large as the global atmospheric storage of C [3]. Due to the large amount stored in terrestrial ecosystems, a small change in SOC significantly affects the concentration of greenhouse gases in the atmosphere. Accordingly, concerns about global warming are increasing interest in quantifying the control of soil C dynamics and storage [4], [5]. Meanwhile, the issue of SOC is of growing concern as it is important for improving soil quality, sustaining food production and quality, and maintaining water quality [6]. Therefore, a better understanding of the mechanism and dynamics of SOC accumulation and storage is necessary to accurately quantify SOC stocks, study soil quality and productivity, and model the global C cycle.
Guizhou Province is located in the center of the karst zone of East Asia, which is one of the three largest karst areas in the world. The carbonate rocks outcrop in an area of 130000 km 2 , which is 73% of the total land area of the province [7]. Regional rocky desertification is a serious ecological problem in the subtropical karst areas of Southwest China, which makes the residents poorer and poorer due to the loss of soil and water [8]. Rocky desertification is a process of land degradation that occurs due to the combined effect caused by the imposition of irrational anthropological activities on the vulnerable karst eco-geo-environment background [9]. With the continuation of rocky desertification, soils in karst areas undergo degradation, which leads to chemical and physical (decomposition and erosion) loss of SOM. Therefore, to protect soil quality, a better understanding of how SOM is decomposed and lost during land degradation in the karst area is needed.
The stable isotopic composition of C is widely used to study biogeochemical processes in soils [10], [11], [12] and to assess the degree of decomposition of SOM [13], [14], [15]. The natural amount of 15 N is used to access the loss of nitrogen (N) and the nature of N mineralization [16], [17], [18], [19], [20], [21], [22] to compare the patterns of N uptake by plant species [23], as well as to determine the impact of land-use history on N cycles [24], [25], [26]. Accordingly, in recent decades, many researchers have widely used the methods of dual stable isotopes (both 13 C and 15 N) to study the dynamics of SOM [27], [28], [29].
There are two main types of soils formed respectively on carbonate rocks and sandstones in the karst area in the Guizhou Province of Southwest China. In this study, we investigated variations in SOC and SON content, as well as variations in their δ 13 C and δ 15 N values in limestone soil and sandstone soil. The main objectives of this study are to characterize the content of SOC and SON, as well as the distributions of their δ 13 C and δ 15 N values in limestone and sandstone soils, as well as to study the factors controlling variations in the content of SOC, SON, and their δ 13 C and δ 15 N values in different soils, and, finally, to achieve a better understanding the dynamics of SOM and soil formation processes in karst areas.

Study site
The investigated sites are located in Baiyi of Wudang and Wangjiazhai of Qingzhen, about 15 and 30 km, respectively, from Guiyang City (26°28′ 19′′ N, 106°49′ 18′′ E), the central region of Guizhou Province, China (Fig. 1), at an altitude of about 1300 m above sea level. The central region of Guizhou Province has a subtropical, monsoon, and humid climate, with an average annual temperature of 14.8°C and an average annual rainfall of 1100 mm. Five soil profiles were selected: three soil profiles were sampled in Wangjiazhai of Qingzhen, which include two soil profiles formed on limestone (profiles QZ-I and QZ-II) and one profile formed on sandstone (QZ-III); two soil profiles were sampled in Baiyi of Wudang, which include one soil profile formed on limestone (BY-II) and the other on sandstone (BY-I). Soils formed on limestone are called limestone soil, classified as calcic Cambisol, and soils formed on sandstone are called yellow sandstone soil, classified as Ferralsol according to the FAO-UNESCO soil classification system [30]. Limestone soils can be further classified as black limestone soils (profiles QZ-I and QZ-II) and yellow limestone soil (soil profile BY-II) depending on their color (or SOM content). The main types of forests are the Masson pine forest, mixed broadleaf forest, and shrub. Detailed features of each soil profile during sampling are summarized in Table 1.

Soil sampling and laboratory analysis
The litter horizons were removed before soil sampling. Soil samples were taken from the soil surface (0-5 cm depth), followed by sampling from the pit walls from a depth of 5 cm to a depth of about 100 cm in the lower horizon. Visible roots and organic debris were removed during sampling. Soil samples containing about 1-2 mg of SOC were placed in 15 mL centrifuge tubes, and tubes without soil were used as controls; 10 mL of 2 mol L −1 KCl solution was added to the test tubes and shaken for 1 hour at 25°C, and then centrifuged at 3000 r min −1 for 25 min, the supernatant was discarded, and then 10 mL of 0.1 mol L −1 HCl solution was added to the tubes and shaken for 24 hours at 25°C and then centrifuged at 3000 r min −1 for 25 min, washed until neutral with distilled water, centrifuged and dried at 60°C, then ground and kept for C and N isotopic analysis [31].
Before taking soil samples, litter was collected on an area of 4 m 2 for each profile. Samples of fresh foliage were also taken from the dominant tree, shrub, and grass species of each major vegetation type. Fresh foliage samples were cleaned with distilled water, and litter samples, separated from any adhering oil material, were dried at 60°C and finely grounded.
The pH value of the soil was measured using a pH electrode (Orion) in a ratio of 1:2.5 (m/v) soil to deionized water. The content of organic C and N was analyzed by combustion in an Elemental Analyzer (PE2400 II, USA) with an analytical accuracy of 0.1%.
For the analysis of stable isotopes, a sample of mass giving 0.5 mg C was placed in a quartz tube with CuO, and then the sample tube was evacuated and sealed with a flame. The organic C in the sample was oxidized to CO2 at 850°C for 5 hours. CO2 was collected and cryogenically purified in a vacuum extraction line, while the amount of CO2 was measured by the manometric method before being collected in a test tube with a burst seal for subsequent mass spectrometric analysis [32]. Stable C isotope ratios ( 13 C/ 12 C) were measured on a Finnigan MAT 252 gas isotope mass spectrometer after purification with liquid nitrogen. The nitrogen isotopes in SON were measured by the Xiao method [31]. Analytical accuracy, defined as the standard deviation obtained over 35 replicates of the IAEA C3 laboratory cellulose standard. IAEA-C3 (δ 13 C = −24.97‰, cellulose) was used as the standard for δ 13 C and the analytical accuracy was ± 0.1‰. Analysis of the potassium nitrate standard (MOR2386-01, 1.92‰) provided by Shoko Co., Ltd. (Tokyo, Japan), gave the mean (± S.D.) δ 15 Nair value 1.9 ± 0.2% (n = 5).
Isotope signatures are expressed by delta notation: δx (‰) =1000(Rsample/Rstandard − 1) where δx is the ratio of C or N isotopes in delta units relative to international standards (Pee Dee Belemnite for C and atmospheric N2 for N) and Rsample and Rstandard are the 13 C/ 12 C or 15 N/ 14 N ratios for samples and standards, respectively. All experimental analyzes were carried out at the State Key Laboratory of Environmental Geochemistry of the Chinese Academy of Sciences.

Carbon and nitrogen content in the soil
As shown in Fig. 3a, SOC usually decreases with depth and is significantly higher in limestone soil than in yellow sandstone soil. SOC and SON are unevenly distributed over soil profiles, with the largest accumulations of SOC and SON occurring in the upper 20 cm of the soil horizon. Surface soils of profile QZ-III (yellow sandstone soil) show the lowest SOC content (20.6 g kg −1 ), while surface soils of profile QZ-I (black limestone soil) show the highest SOC content (69.8 g kg −1 ). For limestone soils, the SOC content exceeds 4.0 g kg −1 , with a maximum of 69.8 g kg −1 at the surface, while for yellow sandstone soils, the SOC content is in the range of 2.0-51.2 g kg −1 . This result is consistent with that reported by Zhu and Liu [33].
The SON content, like the SOC content, is highest in the surface layer of the soil and lowest at a certain depth in the soil (Fig. 3b) and is generally higher in limestone soil than in sandstone soil. Soils from the two black limestone profiles show distinctly high SON contents, especially for soils deeper than 60 cm. Another yellow limestone soil and two yellow sandstone soils demonstrate different styles of SON profile in depth, first decreasing from the topsoil to a depth of about 30 cm and then gradually increasing toward the bottom of the profiles.   As shown in Fig. 4, the mass ratio of organic carbon to nitrogen (C/N) in surface soils ranges from 6 to 22, with the highest value found in the surface layer of the yellow sandstone soil profile (BY-I) and the lowest value found in the surface layer of the black limestone soil profile (QZ-II). Except for the surface layer of the BY-I profile, the surface layers of soils of the other soil profiles display a range of C/N mass ratios from 6 to 15 and a range of C/N mass ratios from 2 to 10 at a depth of 60 cm, which shows an overall decrease from the top to the depth of the profile. There is no clear difference between the mass C/N ratios of limestone soil and sandstone soil.

δ 13 C values of SOC
The yellow sandstone soils show different depth profiles of δ 13 CSOC from limestone soil profiles: the former shows a rapid increase in δ 13 CSOC from top to a depth of about 20 cm, and then maintains a constant or gradual increase downward, while the latter shows an overall increase in δ 13 CSOC within the first 20 cm from the soil surface, and then a decrease in δ 13 CSOC below about 20 cm to the depth at which the sample was taken (Fig. 5a). Maximum δ 13 CSOC values appear at depths of 25 and 15 cm, respectively, for the black limestone and yellow limestone soil profiles, maximum δ 13 CSOC values appear in soil samples at the bottom of sandstone soil profiles. The difference between the maximum and minimum δ 13 CSOC values ranges from 1.8‰ to 3.1‰ for the yellow sandstone soil profiles and from 1.3‰ to 1.4‰ for black limestone soil profiles. The profile of yellow limestone soil (BY-II) shows the largest difference (5.5‰) between the maximum and minimum δ 13 CSOC values.

Lithological control of SOM in soils
Heckman et al [4] recently conducted a systematic study of geological control of soil C cycling and microbial dynamics in temperate coniferous forests and demonstrated that the mechanisms of carbon stabilization differ in various soils formed from different parent rocks. They concluded that the change in the mechanism of carbon stabilization could be directly related to the control of parent material over soil pH and mineral complex. There is a clear difference in SOC and SON content between yellow sandstone and limestone soils, and their content in limestone soils is higher than in yellow sandstone soils. As shown in Fig. 6, black limestone soils have both high SOC and high pH values, while yellow sandstone soils have both low SOC and low pH values, indicating pH control of SOM in soils. Soil pH can be an important comprehensive control of SOC dynamics, although there is no direct correlation between pH and SOC [4]. Soil pH can play an important role in affecting SOM by controlling mineral and microbial variables [34], [35], as well as mineral dissolution and metal speciation [36]. For example, at higher soil pH, Fe is less soluble, and sorption reactions between the negatively charged surfaces of Fe-oxyhydroxide and SOC are preferable to the formation of metal-humus complexes [37]. High pH values of black limestone soils contribute to the preservation of mobile cations and traces of metals due to the sorption of minerals, which can stabilize SOC due to the formation of a calcium-humus complex and sorption of carbon [4]. On the other hand, yellow sandstone soils tend to have pH values below 5.0, therefore more calcium and magnesium are lost through leaching from the soil, and poor calcium enrichment by an organism leads to a lower SOM content than those in limestone soil.
As observed by many researchers [38], [39], there is a positive correlation between the SOC content and the clay content in the soil. According to Zhu and Liu [33], the limestone soils have high clay content and show a clear enrichment of clay from top to bottom soil. Clay minerals have a high specific surface area and carry a charge that allows them to bind and thus chemically stabilize organic matter. Clay aggregates also provide micropores for the physical protection of the SOM [40]. These two types of protective effects may be another important reason why The clay content in the soil is the main factor affecting the change in δ 13 C in SOM with depth in the profile [38]. Limestone soil profiles have a high clay content and show an obvious enrichment in clay from top to bottom soil [33]. The chemical and/or physical protection provided by clay minerals reduces the rate of decomposition of SOM and leads to the fractionation of a small C isotope. This may explain why the degree of 13 C enrichment in the SOM of limestone profiles is lower than in yellow soil profiles. At the same time, high content of clay in the soil can lead to a decrease in the supply of oxygen to the deep layers of the soil, so that most aerobes that can decompose SOM exist only in the upper layer of the soil. As a result, the corresponding soil depth at which the maximum 13 C values of SOM appears is distinctly small in limestone soil. The specific reasons require further study in the future.

Relationships of isotopic compositions between leaves of dominant plants, litters, and topsoil SOM
The chemical composition of plant resources is considered important for predicting the rate of decomposition of litter in forest soils [41], [42], [43]. However, much is unknown about how litter quality affects the amount and chemical nature of C stored in the long term as SOM [44]. The dual-isotope approach was chosen for separate tracking of C and N retention and transformation, as stable isotope markers ( 13 C, 15 N) combined with the SOM fractionation approach can be used to directly measure stabilization rates and pathways of specific substrates among defined SOM pools on shorter time scales [38], [45].
It was observed that the dominant plant species usually have the lowest δ 13 C values, followed by litter and finally the SOM of topsoil, which is consistent with previous studies [46], [47], [48]. However, we note that, in contrast to the δ 13 C value, the average δ 15 N values of plant litter are lower than those of the plant leaf and SOM of the upper soil layers for both types of soil profiles ( Table 2). The differences in the values of δ 13 C and δ 15 N between the dominant plant species and litter are much smaller than between the litter and SOM of the topsoil. SOM in the upper layer of soil of yellow sandstone and limestone soil profiles has a value of δ 13 C by 2.6-3.0‰ and 5.5-6.3‰, respectively, larger than litter. The same case is that SOM in topsoils of yellow sandstone and limestone soil profiles has a value of δ 15 N by 5.1-5.4‰ and 6.5-8.1‰, respectively, larger than litter. This indicates that the differences between the values of δ 13 C and δ 15 N of litter and topsoil SOM in yellow sandstone soil are much smaller than the differences between the litter and topsoil SOM in limestone soil.
Biological and biochemical transformations that occur during humification processes lead to the enrichment of 13 C in residual organic C in SOM [49]. For example, the value of δ 13 C of lignin is approximately 2.0-6.0‰ lower than in whole plant tissues, and therefore the decomposition of organic structures such as lignin in the soil may contribute to the enrichment of 13 C of SOM in the surface soils during the humification process as compared with the original plant material [49]. The most important processes are microbial respiration and fermentation, leading to an enrichment of 13 C in microbial biomass carbon compared to the emitted CO2. The second mechanism is mainly controlled by the rate of SOM turnover in the upper soil layer. The higher the turnover rate, the greater the increment [50].
As indicated above, in contrast to the δ 13 C value, the average δ 15 N value of plant litter is 1-3‰ lower than that of plant leaves for both types of soils. It is unclear why litters have clearly  15 N values than the current leaves of the plant since the decomposition of organic matter usually leads to enrichment with heavier isotopes in the residual materials. There are probably several reasons for the interpretation. First, the different plant species at the same area have different δ 15 N values; the difference in δ 15 N values can reach several ‰. The litter, if it consists of different plant tissues of different plant species, has on average a lower or different δ 15 N value than the current plant leaf. Second, at the initial stage of decomposition, the residual litter is enriched in 14 N (a decrease in δ 15 N from 2‰ to 3‰) [51]. Finally, exotic materials can enter the litter during their decomposition above the soil, as an example of this is that the introduction of atmospheric NH4 + into the litter reduces the value of δ 15 N of litter, because NH4 + is the main form of atmospheric deposition and it shows a value of δ 15 N on average −12.3 ± 6.7‰ around the sampling sites [52].
Like the δ 13 C values, the δ 15 N values increase from the litter to topsoil SOM by about 5.0‰ in yellow sandstone soil profiles and by about 7.0‰ in limestone soil profiles. One of the explanations for this enrichment in 15 N and 13 C due to SOM decay is that respiration preferentially uses organic matter enriched in 14 N and 12 C [46]. The differences in δ 13 C and δ 15 N values between litter and SOM in the topsoil of the limestone soil are greater, or, in other words, SOM in the topsoil of limestone soil profiles is more enriched in 13 C and 15 N than litter compared litter in yellow sandstone soil profiles. This suggests that the rate of decomposition or degree of decomposition of SOM in the topsoil of limestone soil profiles is higher than in topsoil of yellow sandstone, or that the SOM in the upper layer of limestone soil was derived from old vegetation different from the current vegetation cover. The latter interpretation is more reasonable since the degradation of SOM is usually faster or more intense in yellow sandstone soil with a lower pH than in limestone soil with a higher pH.

Degradation of SOM
Previous studies have shown that there are significant correlations between SOC and SON content, and these correlations are indicative of the main plant-derived N in soils [53]. As shown in Fig. 7a, the SOC and SON contents are largely correlated, which indicates that plants provide the main nitrogen supply to the soils of the karst area.
In all soil types, the content of SOC and SON decreases with increasing depth of soil profiles, which is consistent with what has been observed in many studies [53], [54], [55], [56], [57]. However, the extent of this reduction in SOC concentration varies between different soil profiles (Fig. 3a). Compared to SOC in the 0-20 cm soil horizon, the amount of SOC from 20 cm to the bottom depth, on average, decreases by 7.1% in limestone soil and by 31.5% in yellow sandstone soil. SOC and SON content is mainly concentrated in the surface layer of the soil, which can be attributed to more litter in the surface layers of the soil and more accumulation of organic matter. SOC stabilization is controlled by a combination of three interacting factors: chemical protection, physical protection, and inherent structural resistance to degradation [58]. SOC can be chemically protected by organo-mineral complexes with silt and clay particles, which limit the effects of enzymes that destroy SOC [59]. Physical protection occurs when SOC is contained in stable soil aggregates, limiting the diffusion of water and oxygen and reducing the physical access of degrading organisms [60]. In addition, depending on the type of microorganisms present in the soil, biochar, and the forms of lignin, tannins, and aliphatic substances, SOC is 13 more resistant to degradation and exhibits an inherently slower decomposition rate [61], [62]. These factors help to determine the overall biogeochemical stability of microbial and plant SOC, although the dominant mechanisms will vary depending on interactions between microbes, available substrates, and abiotic driving variables.
Significant correlations were found between δ 15 N and δ 13 C in yellow sandstone soil profiles but not in limestone soil profiles (Fig. 7b). The depth profiles of the δ 13 CSOC and δ 15 NSON in yellow sandstone soil and limestone soil vary in different styles: the former shows a rapid increase in δ 13 CSOC and δ 15 NSON values from topsoil to a depth of about 20-30 cm, and then remains almost constant or increases relatively slow downward, while the latter shows an increase from the topsoil to about 20 cm and then decrease downward, with the δ 13 CSOC values in the topsoil being almost the same or even higher than those of the soil at its deepest depth. The depth profiles of the δ 13 CSOC and δ 15 NSON values for yellow sandstone soils can be attributed to the decomposition of SOM by origin. In contrast, the depth profiles of the δ 13 CSOC values for limestone soils indicate a more complex origin or pedogenesis of limestone soil, which will be discussed in more detail later.
During the decomposition of SOM, kinetic fractionation of both 13 C/ 12 C and 15 N/ 14 N occurs in an open system, which leads to the concentration of 13 C and 15 N in the residual organic matter retained in the soil. The 13 C enrichment follows a Rayleigh distillation process, according to which the 13 C/ 12 C ratio is a function of the natural logarithm of the fraction of remaining SOC, resulting in the SOC being enriched with 13 C to 6‰ compared to the original biomass [63], [64]. Fig. 8 shows the variations in δ 13 CSOC and δ 15 NSON values with the natural logarithm of SOC and SON ( Fig. 8a and 8b) in yellow sandstone soil profiles, which suggests the Rayleigh distillation process responsible for the 13

Implications for land-use change in limestone soils
The distribution of limestone soils in the karst landform is not continuous and is uneven in the depth of the soil. The depth of the soil varies from an average depth of about 20 cm on the hillside to more than 1 m at the foot of the hill. In addition, the geochemical and pedological characteristics of limestone soil are heterogeneous, changing significantly even in a limited area. Sandstone soil profiles usually show a simple tendency to increase the depth of δ 13 C, but the depth profiles of δ 13 C in limestone soil are complex, with the higher value of δ 13 C in the topsoil than in deep soil. This phenomenon was also observed earlier by Zhu and Liu [33] for two limestone soil profiles. Accordingly, the pedogenesis of limestone soils distributed over a wide area in southwestern China is complex.
Ehleringer et al [14] and Balesdent et al (1993) [27] reviewed some hypotheses that have been developed over the past few decades to explain the commonly observed δ 13 CSOC profiles enriched with depth in well-drained mineral soils. In a discussion by Wynn et al [65], these hypotheses were grouped into 3 categories: (1) hypotheses involving the mixing of SOM with different 13 C content, (2) hypotheses involving the predominant decomposition of SOM components with different 13 C content, and (3) hypotheses concerning the kinetic fractionation of C isotopes during SOM maturation. The lowest values of δ 13 CSOC in the profiles of limestone soils are observed for the deepest soil, similar to the values in sandstone soil at the same depth. From bottom to top, the δ 13 CSOC values of the profiles of limestone soils increase and reach a maximum value at a soil depth of about 20 cm, and then decrease to the soil surface. From a depth of 20 cm and below, the profiles of limestone soils show a tendency to decrease in depth, which is most likely due to changes in vegetation types with increasing participation of C4 plants. From a depth of 20 cm and above, soil profiles are likely to receive more carbon from C3 plants than from C4 plants. Limestone soil profiles were collected from the slopes of the hills, which are suspected to have left the soil and been eroded. Soil erosion removes SOC-enriched surface soil and 13 C-enriched soil residue. As the vegetation recovers, the abandoned soil becomes rich in SOM, and the δ 13 CSOC value decreases due to an increase in the introduction of C3 plants. Thus, the depth profiles of δ 13 CSOC for limestone soil mainly indicate a history of land-use change, a history of deforestation, land cultivation, land abandonment, and vegetation restoration.

Conclusions
The amounts of SOC and SON are mainly concentrated in the surface layer of the soil, and decrease with increasing depth and appear in different amounts in different types of soils, with the SOM in limestone soil being significantly higher than in the yellow sandstone soil. The higher SOM content in limestone soil can be attributed to soil mineralogy, higher pH, and more Ca ions than in yellow sandstone soil.
The dominant plant species usually have the lowest δ 13 C values, followed by litter and, finally, the SOM of topsoil. However, in contrast to the δ 13 C value, the average δ 15 N values of the plant litter are lower than those of the plant leaves and SOM of the upper soil layer for both types of soil profiles. SOM in topsoils of yellow sandstone and limestone soil profiles has δ 13 C values, respectively, from 2.6‰ to 3.0‰ and from 5.5‰ to 6.3‰, and δ 15 N values, respectively, from 5.1‰ to 5.4‰ and from 6.5‰ to 8.1‰, which is more than in litters. The differences between the δ 13 C and δ 15 N values of the litter and SOM of topsoil in yellow sandstone soil are significantly less than in limestone soil.
The δ 13 C and δ 15 N values of SOM increase with depth in yellow sandstone soil profiles, which indicate that SOM degradation becomes more intense with increasing soil depth. The relationship that an increase in the values of δ 13 C and δ 15 N with a decrease in SOM content in the yellow sandstone soil significantly justifies the decomposition of SOM and, therefore, causes an increase in the values of δ 13 C and δ 15 N in deeper soil horizons of yellow sandstone soil profiles. However, the δ 13 C and δ 15 N values of SOM in limestone soil profiles have a complex tendency to change with increasing soil depth and SOM content, which can be mainly attributed to changes in land use and soil accumulation. Compared to yellow sandstone soil profiles, limestone soil profiles have less variation in δ 13 C and δ 15 N values, mainly due to soil pH value, clay content, and higher amounts of calcium and magnesium elements present in limestone soil profiles.