Responses of Soil Organic/Inorganic Carbon Concentrations in the Lower Yangtze River to Soil Development and Land Use

Abstract

Understanding the evolution and regulation of soil carbon (C) across different stages of geological development is essential for elucidating soil’s role in C storage and release processes. In this study, 1029 soil samples were collected from the alluvial layers of the lower Yangtze River. The chemical index of alteration (CIA) was employed to establish a gradient sequence of soil development, facilitating an investigation into the evolution of organic carbon (OC) and inorganic carbon (IC) in both surface and deep layers across various development stages, as well as their influencing factors. The results demonstrated that as soil develops, surface OC content increases significantly, while the deep layer exhibits no substantial changes. Notably, IC loss was particularly pronounced in surface soils, decreasing from 4.90 g/kg to 0.07 g/kg. Furthermore, the impacts of land use were more evident during the early stages of soil development. Paddy–dryland rotation (paddies) was found to enhance OC sequestration while maintaining IC levels comparable to those of drylands. Soil development directly influenced IC content, whereas its effect on OC content at different depths was primarily mediated by changes in the zirconium-to-rubidium (Zr/Rb) ratios and metal cation concentrations. This study highlights that OC accumulation during soil development predominantly occurs in surface layers, while IC leaching can also be detected at greater depths. At highly developed sites, paddies are recommended as beneficial for preserving C reserves.

1. Introduction

Over the last few centuries, atmospheric carbon dioxide (CO₂) concentrations have increased significantly due to factors such as population growth, urban expansion, and industrialization, making it a major driver of global climate change [1,2,3]. Soil represents the largest C reservoir in terrestrial ecosystems, storing approximately 2500 petagrams (Pg) of C (1 Pg = 10¹⁵ g), which exceeds the total C stored in the atmosphere and vegetation by 1.7 times [4]. Given the immense size of this C reservoir, even minor changes can have significant feedback effects on the global C budget and climate dynamics [5,6,7]. Therefore, exploring the mechanisms of change in the soil C pool and their driving factors is essential for developing effective soil management strategies and addressing climate change [8,9,10,11].

Soil is a heterogeneous spatiotemporal continuum, and numerous studies have demonstrated that various soil-forming factors—such as climate, biology, parent material, and topography—can directly alter the ratio and quantity of organic carbon (OC) and inorganic carbon (IC) [12,13,14], thereby indirectly affecting the storage and stability of the entire soil C reservoir [15]. However, the influence of time as a mediating and intensifying factor in soil formation has received relatively limited attention regarding its effects on soil element content and quality. Certain studies have found that soils in the early stages of development have extremely low organic matter due to a lack of vegetation cover, which enables them to sequester additional OC during their development [4]. Contradictory viewpoints suggest that these soils, characterized by a lower degree of weathering, exhibit higher initial calcium carbonate (CaCO₃) content, indicating elevated levels of IC [16]. Nonetheless, due to natural weathering processes (e.g., temperature, precipitation) and human activities (e.g., irrigation, fertilization), there are significant risks of leaching from these reservoirs. Therefore, understanding the evolution rates and trajectories of different C components in soil across developmental gradients is essential for dynamically adjusting soil C reservoir strategies.

Different agricultural land uses employ varied management practices, including types and quantities of fertilization, crop rotation systems, irrigation, and drainage [17,18]. These practices influence soil C storage by influencing the biogeochemical cycling processes within agricultural ecosystems [19]. The lower Yangtze River plain, characterized by flat terrain, abundant precipitation, and elevated temperatures, is one of China’s most important grain production areas [9]. Dryland and paddy–dryland rotation are the two primary types of agricultural land use in this region. Dryland farming relies on natural precipitation and primarily cultivates wheat and rapeseed, benefiting from good soil aeration conditions. In contrast, paddy fields utilize irrigation and flooding to meet the growth requirements of rice, creating a long-term hypoxic environment due to waterlogging [20]. These two types of agricultural land exhibit distinct soil properties. Understanding the relationship between land use and soil C content is crucial for formulating land management and planning policies aimed at promoting soil conservation, C reduction, and sustainable development [21]. This highlights the dynamic balance between human development and environmental protection. However, our understanding of the impacts of different agricultural land uses on soil C storage and quality remains limited, particularly regarding whether the processes of soil geological development enhance, maintain, or diminish the effects of land use.

The content of OC and IC in soil is influenced by various factors, including climate, topography, and human activities [22,23,24]. However, when studying evolution along prolonged soil development gradients, it is crucial to focus on the impact of specific factors. Owing to weathering and leaching, mineral metal elements in primary minerals (such as feldspar and mica), including calcium (Ca²⁺), iron (Fe³⁺), and aluminum (Al³⁺), may significantly decrease [25]. Rowley et al. [26] found that metal cations, due to their larger specific surface area and positive charges, are effective in stabilizing OC. However, Zhang et al. [27] indicated that excessively high concentrations of metal cations can restrict plant growth and hinder the pathways for organic matter input into the surface soil. Additionally, along soil development gradients, pH may serve as a key indicator of continuous change. In the initial stages of soil development, higher pH may negatively affect plant and microbial activity, leading to insufficient potential for OC replenishment. As soil develops, it is widely accepted that pH decreases, and acidic conditions may promote the dissolution of soil carbonates, resulting in IC loss [26]. Overall, the content of OC and IC in soil is the result of the combined effects of natural and human factors [28,29,30], making it essential to clarify the driving direction and intensity of environmental factors in the dynamic changes in OC and IC.

The main objective of this study is to assess how soil development and agricultural use alter soil OC and IC storage, and to provide a reference for C management in regions with similar alluvial soil development. We hypothesized that: (1) at the watershed scale, soil OC and IC storage will exhibit regular changes along the soil development gradient, although different layers may show varying evolutionary directions and rates; and (2) while OC content in paddies may be higher than that in drylands, the IC content may be lower. Additionally, the processes of soil geological development may enhance the comparison of soil C content between different land uses, based on the assumption that older lands have a longer land use history, resulting in more profound land use effects. In the present study, soil chronosequence emerges from the combined effects of: (1) initial pedogenic differentiation of alluvial material at the time of its deposition, and (2) subsequent pedogenic processes occurring post-deposition.

2. Materials and Methods

2.1. Study Area

This research was conducted in Anhui Province, China, situated in the lower Yangtze River Basin, spanning an area of approximately 20,200 km² (Figure 1). The area is geographically defined by latitudes 29°48′–32°03′ N and longitudes 116°04′–118°52′ E, with altitudes varying from 0 to 20 m above sea level. Characterized by a subtropical monsoon climate, the region experiences an average temperature of 15.8 °C and annual precipitation of 1170 mm.

This study employed a grid-based sampling strategy, utilizing ArcGIS 10.2 at a 4 × 4 km resolution, which initially pinpointed 1262 potential sampling locations. However, due to limitations imposed by land use (woodland, dryland, paddy–dryland rotation) and geographic barriers like rivers and urban areas, a total of 1029 soil samples were gathered during October and November 2003, following the harvest season. The sampling distribution comprised 163 samples from woodland (Populus euramevicana), 129 from dryland crops (wheat, Triticum aestivum; rapeseed, Brassica campestris), and 737 from paddy–dryland rotations (rice–wheat and rice–rape). Soil samples were obtained from both the surface layer (0–20 cm) and deeper layer (150–180 cm). Within a 10 m radius of each sampling point, five sub-samples were taken and combined into a single composite sample. These samples were then air-dried (after removing visible plant debris) and gently ground to pass through a 2 mm sieve.

2.2. Soil Samples and Chemical Analysis

The chemical index of alteration (CIA), proposed by Nesbitt and Young [31], serves as a measure to assess the extent of weathering in source regions. Weathering processes lead to the substantial leaching of alkaline metals, including Na, K, and Ca, resulting in the formation of clay minerals like montmorillonite, illite, and kaolinite. An elevated CIA value signifies a higher degree of sediment weathering. Our previous study in the lower Yangtze River Basin also confirmed a strong correlation between the CIA and soil age (Supplementary Figure S1) [16]. This study constructs relative developmental stages of various soils based on CIA calculations, defined by the formula:

$$\mathrm{CIA}=100\times \mathrm{n}\left({\mathrm{Al}}_2{\mathrm{O}}_3\right)/\left[\mathrm{n}\right({\mathrm{Al}}_2{\mathrm{O}}_3)+\mathrm{n}\left({\mathrm{CaO}}^{*}\right)+\mathrm{n}\left({\mathrm{Na}}_2\mathrm{O}\right)+\mathrm{n}({\mathrm{K}}_2\mathrm{O})]$$

(1)

In the CIA calculation, the content of each oxide is expressed in mole fractions. The term CaO^* denotes the calcium oxide in silicates, calculated by the formula:

$${\mathrm{CaO}}^{*}=\mathrm{CaO}-(10/3\times {\mathrm{P}}_2{\mathrm{O}}_5)$$

(2)

The zirconium/rubidium (Zr/Rb) ratio was utilized as a proxy for soil texture analysis. This method, effective in large dataset analyses [32], exploits the tendency of zirconium to concentrate in coarser particles and its association with heavy minerals, while rubidium is primarily found in clay minerals and micas, exhibiting limited mobility due to strong sorption [33]. Given the uniform Yangtze River alluvial parent material in the study area, the Zr/Rb ratio serves as a reliable indicator to reflect the soil texture.

Soil pH was measured using a glass electrode at a 1:2.5 soil-to-water ratio. Elemental content of Zr, Rb, Al³⁺, Ca²⁺, Fe³⁺, K⁺, Na⁺, Mg²⁺, and Si⁴⁺ was analyzed by X-ray fluorescence spectrometry (XFS) (XRF-1800, Shimadzu, Japan) on 5 g of air-dried soil samples ground to <200 mesh. Total carbon content was measured by dry combustion at 950 °C with an elemental analyzer (Elementar, Hanau, Germany), while soil OC content was determined using the Walkley–Black wet oxidation method [34]. Soil IC was calculated by subtracting OC from total carbon.

2.3. Data Collection

The digital elevation model (DEM) used in this study, characterized by a resolution of 30 × 30 m, was sourced from the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) global digital elevation model (GDEM), accessible through the platform of the Chinese Academy of Sciences (http://www.gscloud.cn, accessed on 1 October 2024). In addition, mean annual temperature, annual precipitation, and evaporation data were obtained from the Data Center for Resources and Environmental Sciences, under the Chinese Academy of Sciences (RESDC) (https://www.resdc.cn, accessed on 1 October 2024).

2.4. Factor Importance Identification for OC and IC

WE utilized partial least squares path modeling (PLS-PM) through the “plspm” R package (R version 4.1.0, R Development Core Team, 2019, Vienna, Austria) to investigate the influence of chemical index of alteration on the variability of soil OC and IC. PLS-PM serves as a multivariate statistical method that analyzes relationships among latent variables, which are indirectly measured by observable indicators [35]. In this framework, latent variables included pH, the Zr/Rb ratio, environmental factors, and metal oxides. Each latent variable was represented by one or more observed variables, selected based on rigorous screening criteria, including Cronbach’s alpha, Dillon–Goldstein rho, loadings, and cross-loadings [36]. After applying these criteria, the retained observed variables for analysis included temperature, precipitation, and evaporation for the environmental factor, while Al³⁺, Fe³⁺, and Ca²⁺ were considered for the metal oxide factor. The goodness of fit (GoF) index was utilized to assess the quality of the PLS-PM model, encompassing both the measurement and structural components. The GoF is computed as the geometric mean of the average communality and the average R² value.

To assess differences in soil OC and IC across land use types, PERMANOVA with 999 permutations was conducted, followed by LSD post hoc testing (p < 0.05). Analyses and visualizations were performed using various R packages, including “vioplot”, “vegan”, “nlme”, and “ggplot2”.

3. Results

3.1. Descriptive Statistics

The OC and IC content in surface and deep-layer soils across different land uses is presented in

Table 1. Statistical characteristics of soil organic carbon (SOC) and inorganic carbon (SIC) in surface and deep-layer soils across different land uses.

Land Use	Index	Surface Layer					Deep Layer
		Min	Max	Mean	SD	C.V.	Min	Max	Mean	SD	C.V.
Woodland	SOC	8.09	26.7	12.9 a	2.45	19.1	0.50	8.60	2.84 b	1.52	53.5
	SIC	0	7.14	0.34 b	0.86	254	0	7.70	2.31 a	0.93	40.3
Dryland	SOC	5.98	18.3	10.9 a	2.10	19.2	0.80	13.5	4.41 b	2.50	56.7
	SIC	0	9.20	1.75 b	2.11	121	0	8.30	3.43 a	1.89	55.1
Paddy–dryland rotation	SOC	5.71	25.8	12.8 a	2.33	18.3	0.30	22.8	3.35 b	2.61	77.8
	SIC	0	7.03	0.39 b	1.02	259	0	11.4	2.60 a	1.37	52.8

Note: SOC, soil organic carbon; SIC, soil inorganic carbon; SD: standard deviation; C.V.: coefficient of variation. Different lowercase letters indicate means are significantly different (p < 0.05) between surface and deep layers.

. Surface soils under woodland, dryland, and paddies exhibited OC concentrations of 12.9 ± 2.45 g/kg, 10.9 ± 2.10 g/kg, and 12.8 ± 2.33 g/kg (mean ± SE), respectively, while deep-layer soils exhibited significantly lower values of 2.84 ± 1.52 g/kg, 4.41 ± 2.50 g/kg, and 3.35 ± 2.61 g/kg (p < 0.05). The IC content in woodland, dryland, and paddies was 0.34 ± 0.86 g/kg, 1.75 ± 2.11 g/kg, and 0.39 ± 1.02 g/kg, respectively, whereas in deep-layer soils, the content was significantly higher, at 2.31 ± 0.93 g/kg, 3.43 ± 1.89 g/kg, and 2.60 ± 1.37 g/kg, respectively. The coefficient of variation (C.V.) for OC ranged between 18.3% and 77.8% across land uses in both soil layers, indicating moderate variation intensity. Deep-layer IC demonstrated moderate variability (C.V. = 40.3–55.1%), contrasting with strong variation observed in surface soils (C.V. = 121–259%).

3.2. Spatial Distribution of SOC and SIC in Surface and Deep Layers

The spatial distribution of OC and IC content in surface and deep layers was mapped using ordinary kriging (Figure 2). Notably, low OC content in surface soils is primarily concentrated in the newly formed alluvial plains along the river, exhibiting a radial distribution pattern extending outward from the center of the Yangtze River. In contrast, IC is less prevalent in surface soils, showing a distinct spatial distribution compared to OC, with high-value areas primarily located in the riverine region. Early alluvial soils farther from the river exhibit extremely low IC content, nearing disappearance. In the deep layer, both OC and IC are relatively uniformly distributed, with high-value areas scattered throughout, showing no significant spatial variability.

3.3. The Dynamics of SOC and SIC Content Along the Chronosequence

Along the soil development gradients, significant changes in the content of soil OC and IC in both surface and deep layers were observed (Figure 3). The OC content in surface soil shows a trend of gradual increase, peaking before slightly declining, specifically rising from 9.34 g/kg (C55) to 13.2 g/kg, then decreasing to 11.9 g/kg. The OC content in deep layer is significantly lower than that in surface soil (p < 0.05), ranging from 2.12 g/kg to 5.12 g/kg. Conversely, the IC content in deep layers is significantly higher than in surface soil, but both show a continuous decreasing trend with increasing soil age, which is opposite to the trend of OC. Specifically, as soil develops, the IC in surface and deep-layer soil decreases significantly from 4.90 and 6.54 g/kg (C55) to 0.07 and 2.17 g/kg, respectively.

3.4. SOC and SIC Content Dynamics over Different Land Uses Along the Chronosequence

Significant land use effects on soil OC and IC reserves were observed in both surface and deep layers (p < 0.05). As shown in Figure 4, at the same stage of soil development, except for the earliest C55 stage, the OC content in woodland and paddies was significantly higher than in dryland (p < 0.05). The distribution of IC in surface layers indicated that the content in dryland and paddies was significantly higher than in woodland during the early stages (p < 0.05). After the C70 stage, no significant differences were observed (p > 0.05). Before the C70 stage, in the deep layer, the OC and IC content in dryland was significantly higher than that in woodland and paddies (p < 0.05). As soil developed, no significant differences among land use types were noted (p > 0.05).

3.5. Factor Importance for SOC and SIC

This study utilized PLS-PM to investigate the direct and indirect effects of the chemical index of alteration and certain influencing factors on the concentrations of soil OC and IC in surface and deep layers (Figure 5). The path model accounts for 21% and 54% of the variance in surface soil OC and IC, respectively, and explains 25% and 49% of the variability in subsoil OC and IC. The increase in CIA significantly and directly facilitates the reduction of IC content across various soil layers (p < 0.05). During soil development, the decrease in metal oxides has distinct significant effects on surface OC and IC (p < 0.05) and serves as the key pathway influencing changes in subsoil OC and IC content. Additionally, the rise in the soil Zr/Rb ratio during soil development has a significant negative impact on surface OC while exerting a significant positive effect on IC (p < 0.05).

The relationships between surface- and deep-layer soil OC and IC content and various influencing factors are presented in

Table 2. Correlations between surface- and deep-layer soil organic carbon (OC), inorganic carbon (IC), and impact factors.

	SOC	SOC*	SIC	SIC*
CIA	0.20 *	−0.25 *	−0.66 *	−0.63 *
pH	−0.10	0.14	0.51 *	0.49 *
Ca2+	0.24 *	0.26 *	0.72 *	0.67 *
Al3+	0.23 *	0.37 *	0.12	−0.30 *
Mg2+	0.03	0.41 *	0.15	0.42 *
Fe3+	0.02	0.17	0.70 *	0.10
Si4+	−0.03	−0.44 *	0.10	−0.36 *
K+	0.03	0.36 *	0.01	0.10
Na+	0.03	0.10	0.10	0.24 *
Zr/Rb ratio	−0.17	−0.39 *	0.22 *	0.10
NDVI	0.07	0.15	−0.17	0.03
Evaporation	0.01	0.10	0.01	0.10
Precipitation	−0.05	0.04	0.02	0.10
Temperature	−0.10	−0.02	0.03	0.05
Elevation	0.05	0.02	0.03	−0.04

Note: SOC and SIC denote the content of soil organic carbon (SOC) and inorganic carbon (SIC) in the surface layer. SOC* and SIC* denote the content of soil organic carbon and inorganic carbon in the deep layer. CIA, chemical index of alteration. Asterisks (*) by numerical values indicate statistically significant correlations at the p < 0.05 level.

. Surface OC is primarily influenced by the concentrations of Ca²⁺ and Al³⁺. Mg²⁺, Si⁴⁺, and the Zr/Rb ratio are key factors affecting OC content in the deep layer. Ca²⁺ and CIA are the primary factors influencing IC content across different layers, while significant correlations were also observed between Fe³⁺ and surface IC, as well as between pH and deep-layer IC.

4. Discussion

4.1. The Effect of Pedogenic Stage Variables on Soil OC and IC Heterogeneity

Previous research has demonstrated that the process of soil development exerts a significant influence on soil properties [37,38,39]. In this study, the surface soil OC increased significantly by 49.9% prior to the C75 stage (p < 0.05) (Figure 3). On one hand, a higher developmental stage of the Yangtze River alluvium generally signifies a longer duration of soil formation and a more extended history of agricultural cultivation. This is accompanied by the decomposition of organic fertilizers and plant residues, resulting in a greater accumulation of organic matter in the surface soil. On the other hand, an increased degree of soil weathering enhances the proportion of finer particles, such as silt and clay, within the soil texture, thereby improving the physical protection of organic matter and subsequently reducing decomposition loss. Notably, the content of surface OC does not exhibit a continuous increase; rather, a slight decline is observed following the C75 stage. This finding aligns with the results of Su et al. [16], which indicate that the initial weathering products of parent material are primarily composed of amorphous minerals, exhibiting a strong capacity for organic matter adsorption. In contrast, during the later stages of weathering, these products predominantly transform into crystalline minerals, which demonstrate a reduced affinity for organic matter [40]. This may elucidate the observed slight decrease in OC content during the later stages. No significant change in OC in the deep layer was observed throughout the soil development chronosequence (p > 0.05), indicating that the transformation, decomposition, and turnover processes of OC are predominantly concentrated within the surface soil. Plant litter and organic fertilizers often accumulate in this surface layer [41]. Moreover, factors such as air and moisture contribute to reduced microbial abundance and activity in the deep layer. Consequently, the OC content in the deep layer not only remains lower than that in surface soils but also exhibits a high degree of stability.

The distribution of soil IC exhibits two distinctive characteristics along the soil development gradient. First, it shows a vertical distribution that is inversely related to that of soil OC, with significantly greater reserves found in the deep layer. This observation aligns with the findings of Du et al. [42]. Specifically, surface soils, which have higher OC and lower pH, create an acidic environment that accelerates the decomposition of soil carbonates [43]. Additionally, factors such as precipitation and irrigation may facilitate the movement of carbonates from the surface to the deep layer. Second, there is a consistent downward trend in IC throughout the stages of soil development. The content of soil IC in both surface and deep layers diminishes from 4.90 and 6.54 g/kg to 0.07 and 2.17 g/kg, respectively. This indicates that the loss and leaching of IC occur across the entire soil profile as soil development progresses. During weathering, primary minerals, such as carbonate minerals, undergo dissolution and transformation, releasing CO₂ and consequently leading to a reduction in IC [44]. The weathering reactions are often accompanied by the generation of acidic conditions, which further facilitate the conversion of carbonates into dissolved bicarbonate ions and CO₂. Additionally, agricultural practices, including irrigation and precipitation, continually promote the processes of desalinization and decarbonation, resulting in the loss of dissolved IC. Finally, the consumption of carbonates through plant uptake and microbial metabolic processes should not be overlooked [45].

4.2. Effects of Land Use on the Spatial Variation of SOC and SIC

It is well established that land use significantly influences the input and output processes of C by affecting the quantity of organic matter inputs and soil properties [46]. In this study, it was observed that at the same development stage, the surface soil OC content in woodland and paddies was significantly higher than that in dryland (p < 0.05). This can be attributed to differences in agricultural management practices and soil properties [47]. First, in the lower reaches of the Yangtze River, dryland farming predominantly cultivates crops such as wheat and rapeseed, leading to a substantial proportion of crop residues being removed from the ecosystem during harvest [21]. This significantly restricts the input of organic matter into the soil. Second, frequent tillage disrupts the structural integrity of soil aggregates, thereby reducing the capacity of soil particles to absorb OC. Additionally, this process accelerates the exposure of the tilled layer to the atmosphere, resulting in more rapid decomposition and consumption of organic matter [48].

In the deep layer, it was observed that prior to the C75 stage, dryland exhibited significantly higher OC content. This suggests that the cultivation and management practices employed in dryland farming can lead to the depletion of surface OC; however, deeper soils may retain greater amounts of OC due to long-term accumulation and minimal disturbance, thereby serving as a potential C pool. Notably, while there was abundant OC in the deep layer, it often originates from older plant residues, humus, or other complex organic materials. These substances possess higher molecular weights and intricate chemical structures, resulting in recalcitrant OC that is resistant to microbial decomposition, exhibiting enhanced chemical stability and proving challenging to utilize effectively [49].

The impact of land use on soil IC reserves primarily occurs during the early stages of soil development. Agricultural land (dryland and paddies) generally exhibits higher IC compared to woodland in both surface and deep layers. First, in the Yangtze River Basin, soil IC is predominantly derived from parent material carbonates and the remains of riverine biological skeletons, which are primarily adsorbed into finer soil particles. The notably lower Zr/Rb ratio (3.77) observed in agricultural lands compared to woodland (4.32) may partially explain the observed IC enrichment. Second, the substantial application of nitrogen fertilizers in agricultural lands facilitates changes in the carbonate precipitation and dissolution equilibrium through ammonification and nitrification processes [42]. Finally, the absence of external nutrient supplementation means that natural vegetation significantly consumes soil carbonates during its growth cycle.

4.3. Evolution of Soil Properties and Their Regulation on SOC and SIC

Soil OC and IC are influenced by various factors across different scales. Typically, on a global or national level, climatic factors play a predominant role [13,50], whereas on a regional or watershed scale, soil properties serve as the critical driving forces [51,52,53,54,55]. In the lower Yangtze River Basin, characterized by a uniform climate, gentle topography, and homogeneous parent materials, the degree of soil development (i.e., the time factor) might significantly influence the forms and reserves of C within the soil. Different developmental stages indicate that the soil has undergone varying processes of nutrient input and output. Pathway analyses demonstrate that the process of soil development can directly facilitate the depletion of IC, a loss that is evident throughout the vertical profile. Although soil development does not exert a direct influence on OC, the observed changes in other soil properties during this process become critical pathways for altering OC reserves.

With soil development, the content of metal cations significantly decreases due to factors such as physical and chemical weathering, leaching, and plant uptake, which in turn influence the concentrations of soil OC and IC (Figure 5). Pathway analyses indicate that metal cations exert divergent effects on surface- and deep-layer OC. While previous studies have identified a positive correlation between metal cations and OC due to the adsorption effect [21], the situation in the lower Yangtze River region, particularly in newly formed soils, presents a different narrative. Elevated levels of metal cations may restrict plant growth and microbial metabolic activity, thereby obstructing the pathways for organic matter input into the surface soil, which hinders the accumulation of surface OC. Conversely, in the deep layer, metal cations positively correlate with OC, representing the only significant pathway during soil development. This suggests that the accumulation of OC in deeper layers primarily relies on the adsorption effects of metal cations. Given a lower degree of weathering in the deep layer and the relatively smaller proportion of clay particles, the adsorption of metal ions is particularly crucial. Moreover, combined PLS-PM and chord diagram analyses reveal that metal cations, especially calcium ions (Ca²⁺), significantly promote the accumulation of IC. This finding aligns with the results of Zhang et al. [4], primarily because the IC in the alluvial deposits of the lower Yangtze River predominantly exists in the forms of CaCO₃ and Ca(H₂CO₃)₂.

Generally, soil with a higher proportion of fine particles demonstrates notable advantages in water retention, nutrient holding capacity, and the enhancement of microbial communities, thereby enabling the storage of greater amounts of OC. However, the influence of soil texture on IC remains a subject of ongoing debate. Kindler et al. [56] revealed that sandy soils, characterized by larger pore spaces, are more susceptible to leaching and subsequent loss of IC, whereas clay soils, exhibiting stronger capillary action, are more effective at retaining IC. In contrast, Zhang et al. [4] identified a negative correlation between IC and the content of clay and silt while observing a positive correlation with sand. This inconsistency may be attributed to anthropogenic agricultural practices, particularly irrigation in farmland, which could predominantly dictate the leaching of IC rather than being governed by soil texture.

4.4. Implications for C Storage in the Downstream of the Yangtze River

In the context of global climate warming, investigating changes in soil C reserves and composition under changing environmental conditions is essential [57,58]. In regions of intensive agricultural use, such as the lower Yangtze River Basin, soil development occurs alongside natural weathering processes and human activities. Different developmental stages exhibit distinct soil properties, leading to a gradual evolution in C sequestration. Given the critical role of soil C in climate regulation and the maintenance of ecosystem functions, it is imperative to adopt land management practices aimed at enhancing soil C reservoirs.

The surface layer of newly alluvial soils lacks organic matter inputs and currently has insufficient OC. Consequently, it exhibits significant potential for sequestering additional OC, a finding that has also been confirmed by previous studies. Deep layers are characterized by low biological activity and poor moisture-aeration dynamics, making it challenging to retain newly added OC effectively. Although they possess high C sequestration potential, exploiting this potential is difficult. Along the developmental gradient, soil IC shows a net loss in both surface and deep layers, particularly as surface soil approaches depletion in the later stages of development. Agricultural management practices, such as irrigation, can lead to the dissolution of carbonates, while fertilization—particularly with nitrogen fertilizers—can result in soil acidification. This process drives substantial migration of IC to other ecosystems, including the atmosphere and aquatic systems. As a critical component of the C pool, IC plays a vital role in regulating climate change and maintaining C balance within ecosystems. The depletion of IC has been widely reported in other ecosystems; however, research on agricultural ecosystems is relatively limited. The present study partially addresses this knowledge gap. To maintain land productivity and a net soil C surplus, more OC needs to be sequestered in more developed soils to offset the losses of IC. Compared to OC, IC is a more stable C pool, with turnover times spanning decades or even longer, demonstrating low sensitivity to land use changes [49]. This study also observed this characteristic: although the IC content in agricultural land is significantly higher than that in woodland, the difference between paddies and dryland is not significant. Therefore, at relatively consistent levels of IC, paddies can sequester more OC, making them a preferable land use option in areas with more developed soils.

5. Conclusions

This study employed a soil chronosequence approach to analyze the variations in OC and IC content in both surface and deep layers of alluvial soils developed from Yangtze River sediments, along with the influencing factors. We aimed to provide a reference for the scientific management of soil carbon pools. Our research indicates that the spatial variability of C distribution is greater in surface soil than in the deep layer. As soil develops, the OC content in the surface layer increases from 9.34 g/kg to a peak of 13.2 g/kg, demonstrating significant C sequestration potential. However, the substantial loss of IC during this process, particularly in the surface layer, emerges as a notable concern, with its content decreasing from 4.90 g/kg to 0.07 g/kg. In regions with advanced soil development, it is essential to compensate for IC loss by enhancing OC sequestration. Land use can reshape soil C structure, an effect that is particularly pronounced in the early stages of soil development. Under conditions where IC content is maintained comparable to that of drylands, paddy fields exhibit a significant advantage in OC sequestration. The changes in metal cations are identified as critical factors influencing OC and IC storage. In summary, this study provides a comparable dataset on the changes in soil OC and IC under intensive farming conditions in the lower reaches of the Yangtze River. It is anticipated that this research will offer insights into the dynamic adjustment of C management and land use-oriented management practices. Furthermore, the study provides valuable guidance for maintaining C balance and promoting the sustainable utilization of areas with similar alluvial development.