Cultivation Management Reshapes Soil Profile Configuration and Organic Carbon Sequestration: Evidence from a 45-Year Field Study

Abstract

Long-term human cultivation activities are the key factors of the vertical distribution and storage dynamics of soil organic carbon (SOC) in cropland. Based on a 45-year long-term field experiment, this study systematically compared SOC dynamics and carbon storage characteristics in soil profiles (0–200 cm) between cultivated land and adjacent natural forest. The findings reveal the hierarchical regulatory effects of tillage management on the soil carbon pool. The results show that: (1) Under both land use types, SOC content decreased exponentially with depth, but values in cultivated soils were 0.35–1.54% lower than in forest soils at each layer. SOC content in surface soil (0–78 cm) was significantly higher than in the subsoil (78–158 cm) and substratum layers (158–200 cm) (p < 0.01). At equivalent depths, SOC in cultivated land was significantly lower than in forest land (p < 0.01). Over 45 years, the SOC accumulation rate in the surface soil of cropland (0.07 g·kg⁻¹·yr⁻¹) was only half that of forest land (0.14 g·kg⁻¹·yr⁻¹). (2) The controls of soil physicochemical properties on SOC differed with land use: in forest soils, SOC correlated positively with clay content (r = 0.63, p < 0.01), whereas in cultivated soils, SOC was primarily regulated by total nitrogen (r = 0.94, p < 0.01) and sand content (r = 0.60, p < 0.01) and negatively correlated with bulk density (r = −0.55, p < 0.01) and pH value (r = −0.45, p < 0.05). (3) Long-term tillage significantly reshaped soil profile structure, thickening the plough layer from 20 cm to 78 cm. Surface carbon storage reached 20.76 t·ha⁻², an increase of 11.13 t·ha⁻² compared with forest soil (p < 0.01). However, storage decreased by 4.99 t·ha⁻² and 7.60 t·ha⁻² in the subsoil and substratum layers, respectively (p < 0.01). The SOC storage increment rate was 50.95 t·ha⁻²·yr⁻¹ higher than that of forest soil in the surface layer but 46.81 t·ha⁻²·yr⁻¹ and 11.12 t·ha⁻²·yr⁻¹ lower in deeper layers. These results confirm that cultivation alters soil structure and material cycling, enhancing carbon enrichment in surface soils while accelerating depletion of deeper carbon pools. This provides new insights into the vertical differentiation mechanisms of SOC under long-term agricultural management.

1. Introduction

SOC is of great significance in maintaining soil fertility, supporting plant growth, ensuring ecosystem stability, and mitigating climate change [1,2]. As a major reservoir of soil nutrients, SOC influences the soil’s physical, chemical, and biological properties, and is an essential component of the global carbon cycle. Its reserves and dynamics are vital for reducing greenhouse gas emissions and stabilizing the climate [3].

There are significant differences in the cycling rate and decomposability of SOC across soil depths. Compared with surface soils, deep SOC decomposes more slowly, has a longer residence time, and is therefore more stable in nature [4]. These depth-dependent differences are closely related to the overall stability of SOC pools. Although surface soils typically receive greater organic inputs, deep soils can store substantially larger amounts of organic carbon due to their greater volume of soil. In regions with relatively shallow soil profiles, where the distance from the soil surface to the underlying bedrock is limited [5], organic carbon stock is generally higher in surface soils than in deep soils. However, in regions characterized by thicker soil profiles, such as the Kanto Plain [6], organic carbon stock in deep soils far exceeds that in surface soils. Niu et al. found that in some regions, deep SOC accounts for more than 50% of the total stock. These findings indicate that when the depth is sufficiently large, deep soils may represent a dominant reservoir of organic carbon, surpassing surface soils in their contribution to long-term carbon storage [7].

Research on SOC stocks, both domestically and internationally, has largely been limited to surface soils, with relatively little attention given to deep SOC stocks below one meter. Different soil parent materials, such as brown soil and red soil, contain varying components of organic carbon, which in turn affect SOC stability [8,9]. Soil parent materials regulate the mineral composition, pH, and soil structure, thereby influencing the stability of deep SOC and its interactions with minerals [10].

Land use change strongly affects SOC dynamics. When natural ecosystems are converted to croplands, SOC typically declines due to reduced organic matter inputs and intensified cultivation, which accelerates SOC mineralization and decomposition [11]. Long-term cultivation is therefore a key factor shaping SOC dynamics in agricultural soils. By altering soil structure, affecting microbial activity, and regulating carbon inputs and outputs, cultivation significantly impacts the soil carbon pool. For example, conventional ploughing disrupts soil aggregates, enhancing organic matter decomposition and reducing SOC in the 0–20 cm layer by 15–25% within 20 years in intensively farmed Chinese black soils [1,12]. In a 5-year trial in the North China Plain, SOC storage in the top 30 cm ranged between 30.59 and 32.76 Mg·km⁻², with reduced tillage and no-tillage increasing annual SOC storage by 434.00 kg·km⁻²·yr⁻¹ and 204.00 kg·km⁻²·yr⁻¹, respectively, compared with conventional ploughing [13]. Most previous studies have focused on SOC dynamics across different land uses, but relatively few have systematically examined how long-term cultivation affects carbon pools across soil horizons, particularly the relationship between tillage-induced disturbance of soil depth configuration and SOC loss.

The black soil region of Northeast China has been managed under decentralized land use systems since the mid to late 1980s. In agricultural production, practices such as ridge sowing following stubble removal, ridging with small four-wheel tractors, and flat sowing after shallow rotary tillage using low-power machinery have been widely adopted. However, long-term reliance on small-scale machinery has resulted in pronounced soil degradation, including a shallower tillage layer, thickening of the plow pan, deterioration of soil structure, declines in soil organic matter quantity and quality, and intensified soil erosion [14]. Deep tillage practices, such as subsoiling and deep plowing, can break the plow pan, deepen the tillage layer, reduce soil bulk density, promote crop root penetration and the utilization of water and nutrients in deep soil, thereby increasing crop yields [15]. However, some studies have found that deep plowing can bring nutrient-poor subsoil to the surface, accelerate the mineralization of SOC, exacerbate nutrient depletion in the cultivated layer, and ultimately reduce soil water-holding capacity and crop productivity [16]. Huang et al. [17] conducted a comparative analysis of forest and farmland ecosystems and found that, over a 20-year period of vegetation restoration on cultivated land, increases in deep SOC storage were significantly greater in forest ecosystems than in agricultural systems. Moreover, SOC accumulation under natural restoration was higher than that under artificial restoration. Similarly, under the arid and cold ecological conditions of irrigated desert soil area in the Hexi Corridor, a 41-year long-term positioning experiment (1982–2022) examining dynamics of SOC within the 0–200 cm soil profile revealed significant differences among fertilization regimes in their capacity to enhance SOC content [18].

In this study, we compared long-term cultivated soils with nearby undisturbed forest soils, analyzing SOC content, storage, and their relationships with soil physicochemical properties. We also evaluated how long-term tillage reshapes soil carbon pools across horizons. The aim was to provide a new perspective on the impacts of cultivation on SOC distribution and storage, and to contribute to scientific guidance for soil carbon management and sustainable agricultural development.

2. Materials and Methods

2.1. Study Area

The study area located in the long-term experimental field of Shenyang Agricultural University, Shenyang City, Liaoning Province (41°50′2″ N, 123°34′00″ E), at an altitude of 72 m. This site lies in the core area of the southern Songliao Plain (Figure 1) and has a temperate, humid to semi-humid monsoon climate. The average annual precipitation ranges from 574 to 684 mm, the mean annual temperature is 7.0–8.1 °C, and the average annual sunshine duration is about 2373 h.

The predominant soil type is typical brown soil (moist leached soil), with loess or loess-like materials as its parent substrate. A long-term brown soil fertilization experiment has been conducted at this site since 1979, spanning 45 years. The cropping system is a corn-corn-soybean rotation. Each year, urea (46% N), superphosphate (12% P₂O₅), and potassium sulfate (50% K₂O) are applied at sowing, together with fully decomposed pig manure containing, on average, 119.6 g kg⁻¹ organic matter, 5.6 g kg⁻¹ N, 8.3 g kg⁻¹ P₂O₅, and 10.9 g kg⁻¹ K₂O. All fertilizers are broadcast as basal applications and incorporated into the soil to a depth of 20 cm using a rotary tiller. Sowing is carried out from April to May, and harvest occurs between September and October. After each harvest, crop residues are removed. As a control, a typical forest site was selected near the experimental plots (Figure 1). This forest lies on flat, stable terrain at the same slope position as the cropland and shares identical soil-forming factors except for land use. The vegetation is dominated by perennial oak trees, with a forest age of approximately 40 to 50 years.

2.2. Profile Description and Soil Sample Collection

According to the Manual of Field Soil Description and Sampling [19], the morphological characteristics of the profiles were carefully observed and described (

Table 1. Morphological characteristics of the investigated soil profiles.

Profile	Occurrence Layer	Depth (cm)	Color (Dry State)	Structure and Its Degree of Development	Soil Texture	Fixation (Dry)	Other Characteristics
Cultivated land 21-N001	Ap1	0–17	10YR 4/3	Moderately developed mesenchymal	Silty loam	Soft	A very small amount of iron-manganese nodules
	Ap2	17–28	10YR 4/4	Moderately developed thin sheet	Silty loam	Slightly hard	A very small amount of iron-manganese nodules
	ABr	28–78	10YR 5/4	Moderately developing mass	Silty loam	Slightly hard	A small amount of iron-manganese nodules
	Btrq1	78–111	10YR 5/4	Moderately developed middle ridge block	Silty loam	Hard	A small amount of clay film, iron-manganese nodules and silica neoformation powder
	Btrq2	111–133	10YR 4/4	Moderately developed middle ridge block	Silty loam	Hard	A moderate amount of clay film, a small amount of iron and manganese nodules and a large amount of silica neoformation powder
	Btrq3	133–158	10YR 4/3	Moderately developed middle ridge block	Silty loam	Very hard	A moderate amount of clay film, a small amount of iron and manganese nodules and a large amount of silica neoformation powder
	Cr	158–200	10YR 5/4	Moderately developed middle ridge block	Silty loam	Very hard	A small amount of iron-manganese nodules
Forest land 21-N006	Ahr	0–7	7.5YR 6/4	Moderately developing mass	Silty loam	soft	A small amount of iron-manganese nodules
	ABr	7–20	7.5YR 4/4	Moderately developing mass	Silty loam	soft	A small amount of iron-manganese nodules
	Br	20–62	10YR 4/3	Moderately developing mass	Silty loam	Slightly hard	A small amount of iron-manganese nodules
	Btrq1	62–85	10YR 5/4	Moderately developed small ridges	Silty loam	Hard	A small amount of iron-manganese nodules and a moderate amount of silica neoformation powder
	Btrq2	85–110	10YR 4/3	Moderately developed small masses	Silty loam	Hard	A small amount of iron and manganese nodules and a large amount of silica neoformation powder
	Btrq3	110–140	10YR 5/3	Moderately developed middle ridge block	Silty loam	Very hard	A small amount of iron and manganese nodules and a large amount of silica neoformation powder
	Cr	140–200	10YR 5/6	Moderately developed middle ridge block	Silty loam	Very hard	Moderate iron-manganese nodules

). Prior to sampling, approximately 5 cm of topsoil was removed with a rust-free steel shovel. Soil samples were then collected sequentially from the bottom to the top of each occurrence horizon across 28 profiles. For forest soil profiles, the surface litter layer of dead branches and fallen leaves was first cleared away (Figure 2).

2.3. Data Processing and Analysis

Soil bulk density was determined using the ring knife method [20,21]. Soil pH was determined using potentiometry [22]. Organic carbon and total nitrogen were analyzed using a TOC elemental analyzer [23]. Particle size distribution was measured with a laser particle size analyzer [24,25]. Experimental data were processed in Excel 2021, figures were plotted using Origin 2024, and Pearson correlation analyses were conducted in SPSS 27.0.

Soil organic carbon (SOC) storage was calculated by the stratification method [26], using the following equation:

$$\mathit{SO}{C}_s=\sum _{i=1}^n{C}_i{\rho }_i{T}_i(1{-\theta }_i)\times 10^{-1}$$

(1)

In the formula, SOC_s represents the organic carbon storage (t·ha⁻²) at the specified soil depth, C_i denotes the organic carbon content (g·kg⁻¹) of the i-th layer of soil, ρ_i stands for the bulk density of the i-th layer of soil (g·cm⁻³), T_i represents the thickness of the i-th layer of soil (cm), θ_i refers to gravel content (>2 mm) of the i-th soil layer (% by volume), and n signifies total number of soil layers included in the calculation. As the proportion of gravel (>2 mm) in the study soils was extremely low, this term was disregarded in the calculations.

A soil reconstruction model [27] was used to quantitatively estimate the gains or losses of selected components within 100 cm³ of weathered soil during soil formation. The general expression is:

$$\mathit{UVF} = {\displaystyle \frac{B{D}_w\times C_{\mathit{iw}}}{B{D}_{\mathit{pm}}\times C_{\mathit{ipm}}}}$$

(2)

$$D_{\mathit{jw}}=B{D}_w\times C_{\mathit{jw}}-\mathit{UVF}\times (B{D}_{\mathit{pm}}\times C_{\mathit{jpm}})$$

(3)

In the formula, BD_w and BD_pm represent the bulk density of the weathered layer and substratum, respectively; C_iw and C_ipm represent the concentration of an immobile reference component i in the weathered layer and substratum, respectively; C_jw and C_jpm represent the concentration of component j in the weathered layer and substratum, respectively; and D_jw represents gain or loss of component j in the weathered layer during soil formation. The rate of change in SOC content and SOC storage was expressed as the net gain or loss over the 45-year experimental period, normalized by time.

3. Results

3.1. Distribution of Soil Organic Carbon Content Under Different Land Use Patterns

Under different land use patterns in the study area, SOC content gradually decreased with increasing profile depth. At all depths, SOC content in cultivated land was consistently lower than in forest land (Figure 3). The mean SOC content in cultivated land was 3.54 g·kg⁻¹ (CV = 83.43%), while that in forest land was 3.35 g·kg⁻¹ (CV = 87.33%). The slightly higher variation in forest land indicates greater fluctuation in SOC content, but both land use types exhibited high-intensity spatial variation.

Analysis of variance indicated that SOC content in the surface soil layer of both land use types was significantly higher than that in the subsoil and substratum layers (p < 0.01). Furthermore, SOC content across all layers of forest land was significantly higher than that of cultivated land. Specifically, the average SOC contents in the surface, subsoil, and substratum layers of forest land were 7.38 g·kg⁻¹, 3.06 g·kg⁻¹, and 2.61 g·kg⁻¹, respectively. In contrast, the corresponding values for cultivated land were 5.41 g·kg⁻¹, 2.30 g·kg⁻¹ and 2.43 g·kg⁻¹, respectively.

Thus, cultivated land contained 1.97 g·kg⁻¹, 0.76 g·kg⁻¹ and 0.18 g·kg⁻¹ less SOC than forest land in the surface, subsoil, and substratum layers, respectively. The difference between land use types was smallest in the substratum layer, compared with the larger differences observed in the surface layer and the subsoil (Figure 4). Over the 45-year period, SOC content in the surface soil layer of forest land increased at a rate of 0.14 g·kg⁻¹·yr⁻¹, nearly twice that of cultivated land (0.07 g·kg⁻¹·yr⁻¹). In the subsoil, forest land showed a slight increase (0.01 g·kg⁻¹·yr⁻¹), whereas cultivated land exhibited a slight decline (−0.01 g·kg⁻¹·yr⁻¹). SOC content in the substratum layer of both land use types remained relatively stable.

3.2. Relationship Between Soil Organic Carbon Content and Physicochemical Properties

The results (

Table 2. Correlation coefficients between soil organic carbon content and soil physical and chemical properties.

Land Utilization Type	Bulk Density	pH	Total Nitrogen	Clay	Silt	Sand
Forest land	−0.55 **	−0.45 *	0.94 **	0.63 **	−0.75 **	0.60 **
Cultivated land	−0.78 **	−0.86 **	0.96 **	−0.42	−0.62 **	0.81 **

Note: ** p < 0.01 (extremely significant); * p < 0.05 (significant).

) show that SOC content in forest land was extremely significantly and positively correlated with total nitrogen, clay content and sand content (p < 0.01), while it was significantly and negatively correlated with bulk density, pH, and silt content. In cultivated land, SOC content was extremely significantly correlated with all physicochemical indicators except clay content (p < 0.01).

With increasing soil depth, bulk density (1.09–1.71 g·cm⁻³) generally showed a fluctuating upward trend. The bulk density of the 17–28 cm layer in cultivated land was significantly higher than in forest land (Figure 5), reflecting soil compaction and the influence of long-term ploughing. Soil pH values in both land uses were weakly acidic (5.2–6.5) and increased with depth, with mean values of 6.15 in forest land and 5.85 in cultivated land. The distribution pattern of total nitrogen followed the same trend as SOC, decreasing markedly with depth. Average values were higher in forest land (0.60 g·kg⁻¹) than in cultivated land (0.57 g·kg⁻¹). Clay content in both land uses increased from 0 to 100 cm, decreased slightly below 100 cm, and then increased again. Silt content first increased and then decreased with depth, while sand content declined significantly with increasing depth (Figure 5).

3.3. Distribution of Soil Organic Carbon Storage Under Different Land Use Patterns

The average SOC storage in cultivated land was 15.70 t·ha⁻² (CV = 35.16%), while that of forest land was 17.51 t·ha⁻² (CV = 35.73%). SOC storage in forest land fluctuated more strongly with depth compared to cultivated land. SOC storage increased with depth in forest land soils at 85–200 cm and in cultivated land soils at 133–200 cm (Figure 6).

Analysis of variance showed that in forest land, SOC storage in the substratum layer was significantly higher than in the topsoil and subsoil layers. In cultivated land, SOC storage in both the topsoil and substratum layers was significantly higher than in the subsoil layer. Across all soil layers, SOC storage differed extremely significantly between cultivated and forest land (Figure 7). Specifically, the average SOC storage in the topsoil was 9.63 t·ha⁻² in forest land and 20.76 t·ha⁻² in cultivated land, 11.13 t·ha⁻² higher in cultivated land. In the subsoil layer, SOC storage was 15.01 t·ha⁻² in forest land versus 10.02 t·ha⁻² in cultivated land, 4.99 t·ha⁻² lower in the latter. In the substratum layer, forest land storage (25.12 t·ha⁻²) exceeded that of cultivated land (17.52 t·ha⁻²) by 7.60 t·ha⁻². Over 45 years, SOC storage in forest land increased at rates of 30.26, 110.74, and 56.99 t·ha⁻²·yr⁻¹ in the surface, subsoil, and substratum layers, respectively. In cultivated land, the corresponding rates were 80.11 t·ha⁻²·yr⁻¹, 64.95 t·ha⁻²·yr⁻¹ and 45.87 t·ha⁻²·yr⁻¹, respectively.

4. Discussion

4.1. Variation in Soil Organic Carbon Content Under Different Land Use Patterns

SOC dynamics are mainly influenced by multiple factors, including climate, vegetation, soil physicochemical properties, and agricultural management. Among these, land use is a primary driver of SOC variation [28,29,30]. Differences in land use affect the vertical migration of SOC from surface to deeper layers, thereby altering its overall profile distribution [31]. In this study, SOC content under all land use patterns decreased progressively with depth.

Across all profile layers, SOC content in cultivated land was consistently lower than in forest land. This difference can be attributed largely to human disturbance [32]. Harvesting crops removes much of the above-ground biomass, reducing organic matter inputs and limiting SOC accumulation [33]. By contrast, forest ecosystems provide substantial organic matter inputs through litterfall, root turnover, and deadwood, all of which enhance SOC content. Deep-rooted trees also transport organic matter into subsoil horizons. Frequent tillage in cultivated soils further disrupts soil structure, increases aeration, and accelerates organic matter mineralization [34]. These processes, combined with shifts in microbial community composition, reduce SOC storage. Long-term monitoring over 45 years showed that SOC accumulation rates in forest land were significantly higher than in cultivated land, especially in the surface layer. SOC in the topsoil of forest land increased at 0.14 g·kg⁻¹·yr⁻¹—double the rate observed in cultivated land (0.07 g·kg⁻¹·yr⁻¹). Forest soils are less disturbed by human activities, maintaining a stable structure, rich microbial communities, and plant-microbe interactions that enhance carbon sequestration. In deeper layers, SOC in forest soils is further stabilized. On the one hand, vertical translocation delivers organic inputs downward; on the other hand, SOC decomposition is slower in subsoils [35]. These processes collectively explain why forest soils maintain higher SOC content than in cultivated land.

4.2. Influence of Soil Physical and Chemical Properties on Soil Organic Carbon

Because climate, topography, and substratum were similar across the study sites, differences in SOC were primarily driven by vegetation type, soil physicochemical properties, and land management. SOC content showed an extremely significant positive correlation with total nitrogen in both land uses (r = 0.96 for cultivated land; r = 0.94 for forest land). This indicates that higher SOC is consistently associated with higher nitrogen availability. SOC was negatively correlated with bulk density (r = −0.78 and −0.55 for cultivated and forest land, respectively), suggesting that higher SOC is associated with looser, less compact soils. This aligns with findings by Wu et al. [36], reinforcing the close linkage between SOC, nitrogen cycling, and soil structure. SOC was also significantly negatively correlated with pH (r = −0.86 and −0.45). Both forest and cultivated soils were weakly acidic (pH 5.2–6.5), conditions favorable for microbial activity, which enhances litter decomposition and SOC accumulation [37]. With respect to particle size, SOC in forest land was positively correlated with clay content (r = 0.63), consistent with previous studies [38]. However, no significant relationship was observed in cultivated soils. This may be due to long-term tillage, which alters soil structure and reduces organic inputs, thereby limiting the capacity of clay to adsorb and stabilize SOC [39,40].

Interestingly, unlike earlier reports showing positive relationships between SOC and silt, and negative relationships with sand [41,42], this study found extremely significant negative correlations between SOC and both silt and clay (r = −0.62 and −0.75), and positive correlations with sand (r = 0.81 and 0.60). These patterns may reflect long-term leaching and sediment redistribution, which altered particle-size composition within profiles. Strong leaching in surface layers reduces sand content, while deeper layers stabilize in texture. At the same time, organic matter inputs decline with depth, reducing SOC concentrations. This highlights that the influence of soil texture on SOC can vary significantly between regions, and its mechanisms require further investigation.

4.3. Changes in Soil Organic Carbon Storage Under Different Land Use Patterns

SOC storage is of critical significance in maintaining soil fertility, regulating carbon cycling, and mitigating climate change [43,44]. Different land use patterns significantly affect the vertical distribution and variability of SOC storage by altering vegetation residue inputs, management practices, and soil environmental conditions. In this study, SOC storage in forest soils at 85–200 cm and in cultivated soils at 133–200 cm increased with depth. This trend occurs because deeper soil layers experience less disturbance from human activities, organic matter decomposes more slowly, and root exudates and litter residues migrate downward through leaching, leading to a relatively stable carbon storage. In forest land, the substratum layer has the highest SOC storage, followed by the subsoil layer and topsoil layer. This reflects continuous carbon transport to deep soil via root secretions and decomposition of microbial and animal residues [45,46,47]. The substratum layer, being less disturbed, allows organic matter to decompose slowly and accumulate over time. In cultivated land, however, the topsoil layer stored the most SOC, followed by the substratum and subsoil layers. This pattern is largely due to the continuous application of organic fertilizers, which enhances organic matter input and SOC accumulation in surface layers. SOC storage is determined by SOC concentration, bulk density, and soil thickness [48]. In this study, soil thickness was a key factor influencing SOC variation. Long-term cultivation modified soil configuration, thickening the topsoil and creating marked differences in SOC distribution. As a result, cultivated soils showed carbon enrichment in the topsoil—where SOC storage was much higher than in forest soils—but carbon depletion in the subsoil and substratum layers, where SOC storage was lower than in forest soils. These findings clearly demonstrate the role of human cultivation in reshaping the vertical migration and redistribution of the soil carbon pool (Figure 8).

In the first year of reclamation, the topsoil thickness of cultivated land was similar to that of the forest land (about 20 cm). After 45 years of farming, the long-term application of tillage practices—such as ploughing, deep loosening, and fertilization—had two main effects: it enhanced the accumulation of surface organic matter, and it disrupted soil aggregates, leading to particle rearrangement and increased porosity. As a result, the cultivated topsoil expanded vertically, reaching 78 cm in thickness (Figure 8). The progressive thickening of the soil layer provided greater storage capacity for root growth and retention of dead roots, increased the activity and spatial extent of soil microorganisms, and created favorable conditions for SOC input, transformation, and storage. Over the 45-year period, SOC storage in both forest and cultivated soils increased across all layers, but at markedly different rates. Results showed that SOC storage in the topsoil of cultivated land was 11.13 t·ha⁻² higher than in forest land, with an accumulation rate 50.95 t·ha⁻²·yr⁻¹ greater. This difference can be attributed to the continuous addition of organic fertilizers in cultivated soils, which directly enhanced organic matter inputs, while the thickened topsoil provided greater capacity to retain and stabilize carbon. In addition, ploughing improved gas exchange between soil and atmosphere, stimulating microbial decomposition of organic matter and humus formation, further enhancing SOC accumulation in the cultivated topsoil. Although litter inputs occur in forest land, the topsoil is relatively compact, with low porosity. As a result, the decomposition of dead branches and fallen leaves is slow, and the thin soil layer has limited capacity to retain SOC. In addition, part of the organic matter is lost through surface runoff, while the SOC released by decomposition is prone to mineralization, ultimately constraining SOC accumulation in the forest topsoil. The subsoil layer extended from 20 to 140 cm in forest land, but from 78 to 158 cm in cultivated land. Continuous thickening of the cultivated topsoil pushed the initial boundary of the subsoil layer downward and reduced its thickness. This occurs because farming practices primarily affect the surface, leaving deeper layers less disturbed. As the topsoil expands, the upper boundary of the subsoil layer is forced deeper. Consequently, SOC storage in the cultivated subsoil layer was 4.99 t·ha⁻² lower than in forest land, and the accumulation rate was 46.81 t·ha⁻²·yr⁻¹ lower. This highlights the contrasting impacts of land use on the deep soil carbon pool. From a soil carbon cycle perspective, forest ecosystems maintain a continuous vertical migration of SOC via root secretions and persistent vegetation cover. Carbon gradually accumulates in deeper layers through root turnover and the long-term activity of microorganisms. In contrast, crop roots in cultivated soils are largely confined to the topsoil, resulting in limited carbon input to deeper horizons. The substratum layer occurred at 140–200 cm in forest land and 158–200 cm in cultivated land. Long-term cultivation altered the structure of the topsoil and subsoil layers, pushing the starting depth of the substratum layer downward, though its thickness remained largely unchanged. SOC storage in the cultivated substratum layer was 7.60 t·ha⁻² lower than in forest land, and its accumulation rate was 11.12 t·ha⁻²·yr⁻¹ lower. In forests, root systems continuously transport carbon into the substratum layer, whereas in cultivated soils, root-derived carbon inputs are concentrated in the plough layer. Moreover, mechanical tillage and compaction create a plough pan that restricts the downward movement of organic matter from fertilizers, further limiting SOC sequestration in deeper layers.

By optimizing tillage practices and implementing measures such as moderate deep loosening and reducing the frequency of soil plowing, it is possible to effectively maintain and enhance the carbon sequestration capacity of topsoil in agricultural fields while minimizing disturbance and damage to deeper soil carbon pools caused by anthropogenic tillage. Future research should focus on precisely quantifying the efficiency of soil organic carbon sequestration under different agricultural management practices, such as no-tillage and crop rotation. By integrating the historical context and long-term evolution of land use in the study area, it is essential to identify strategies that enhance soil carbon stocks while balancing soil ecological protection with the practical needs of agricultural production. Such efforts will provide actionable scientific guidance for the sustainable utilization and informed management of farmland soils in China’s black soil regions and other comparable brown soil areas nationwide.

5. Conclusions

(1) SOC content decreased with soil depth under both land uses, with lower levels in cultivated land compared to forest land. In all cases, topsoil contained significantly more SOC than the subsoil and parent layers. Over 45 years since land conversion, SOC in forest topsoil increased at twice the rate of cultivated land, while the forest subsoil layer showed a slight gain and cultivated land a slight loss. Substratum SOC remained stable.

(2) SOC was strongly and positively correlated with total nitrogen and negatively with bulk density and pH, in both land uses. Forest SOC also showed a strong positive correlation with clay, unlike cultivated soils. The relationships between SOC and silt or sand content differed from earlier studies, likely reflecting long-term leaching, sediment redistribution, and regional differences in soil formation and land management.

(3) Long-term tillage thickened the cultivated topsoil, leading to greater surface SOC storage (surface carbon enrichment) but reduced SOC storage in deeper layers (deep carbon loss). This shows that cultivation reshapes soil profile configuration, enhancing surface carbon but weakening deep carbon sequestration. In contrast, forest soils maintained a more balanced vertical distribution, with the substratum layer benefiting from sustained root-derived inputs.