The present study was carried out during 2022–23, using the soil samples collected from Horticultural Research Complex (HRC), Nagicherra, Agartala, Tripura (Latitude: 23.8116° N, Longitude: 91.3284° E), and the Indian Council of Agricultural Research (ICAR) – Research Complex for North–East Hill (RC NEH) Region, Lembucherra, Agartala, Tripura (Latitude: 23.9052° N, Longitude: 91.3147° E) (Figure 1). The study included two main categories of land use: Horticultural and Agricultural. Horticultural crops, situated within the HRC, included oil palm (Elaeis guineensis), rubber (Hevea brasiliensis), litchi (Litchi chinensis), citrus (Citrus sinensis), and guava (Psidium guajava L.) all, as well as ginger (Zingiber officinale). Uncultivated land (A) was used as a reference for comparison. At the ICAR–RC NEH Region, agricultural crops included rice (Oryza sativa)–fallow, vegetable cowpea (Vigna unguiculata)–rice (Oryza sativa)–maize (Zea mays L.), vegetable cowpea (V. unguiculata)–rice (O. sativa)–lentil (Lens culinaris), and vegetable cowpea (V. unguiculata)–rice (O. sativa)–mustard (Brassica juncea) with uncultivated (B) as a reference (Table 1). The soil in the study area was characterized by a pH ranging from highly acidic to slightly acidic (4.93–5.60), and elevated SOC levels between 8.95 and 16.6 g kg⁻¹. Bulk density (BD) was ranged from 1.15 g cm⁻³ to 1.40 g cm⁻³. Mean weight diameter (MWD) was highest under litchi (2.09 mm) and rubber (1.97 mm) plantations, whereas the lowest MWD was observed in uncultivated land uses (1.13–1.18 mm). Available nitrogen, phosphorus, and potassium ranged from 280 to 413 kg ha⁻¹, 10 to 19.7 kg ha⁻¹, and 119 to 185 kg ha⁻¹, respectively (Table 2).

Soil samples were collected following the method described by Dhyani and Tripathi (1999), taken at depth intervals of 0–15, 15–30, 30–60, 60–75, and 75–100 cm with the use of a core sampler. These samples were obtained from each fruit tree block and agricultural block, including the fallow (uncultivated) plot, between October and November 2022. Soil sampling was conducted using a core sampler, and the collected cores were transported to the laboratory, where the soil was carefully extracted. The collected soil was then dried naturally at ambient temperature, pulverized, and sifted through a 2–mm mesh for subsequent testing. To assess the microbial characteristics, portions of each soil sample were extracted and kept at a temperature of 4°C. Following pre incubation, these soil samples were adjusted to 50% of their maximum water holding capacity and used for soil microbial biomass carbon (MBC) estimation.

The physico–chemical properties of the samples were determined using the conventional methods detailed by Jackson (1973). Soil pH (1:2.5) and the contents of available nitrogen (N), phosphorus (P), and potassium (K) were determined using established standard methods, as outlined by Jackson (1973), Subbiah and Asija (1956), Bray and Kurtz (1945), and Hanway and Heidel (1952), respectively. The Microbial biomass carbon was measured using the chloroform fumigation technique outlined by Jenkinson and Powlson (1976). We calculated the MBC/TOC as the ratio of MBC to TOC, expressed as a percentage. The KMnO₄ oxidizable carbon (ROC) was determined through oxidation with 333 mM KMnO₄, following the method outlined by Blair et al. (1995). TOC was analyzed using dry combustion in a CHNS elemental analyzer (Euro Vector, EA3000). Soil organic carbon stocks (0–100 cm depth) were calculated using the TOC and BD values. The soil BD values were measured by the core method, as described by Allen et al. (1974). The particle size distribution and mean weight diameter were assessed using established techniques (Bouyoucos, 1962; Kemper and Chepil, 1965).

In 1995, Blair et al. (1995) proposed the carbon management index (CMI) to assess how land management changes affect SOM relative to a stable reference soil. CMI is calculated using the CPI and the LI. The CMI is calculated as follows:

Lability of C (L): L = C in fraction oxidized by KMnO 4 ( mg labile C g − 1 Soil ) C remaining unoxidized by KMnO 4 ( mg labile C g − 1 Soil )

To determine carbon stocks, measurements of TOC and bulk density BD were utilized. These estimations covered a soil depth of 100 cm. C stocks ( Mg ha − 1 ) = C TOC × BD × D

Where C_TOC is total organic carbon (g kg⁻¹); BD is the bulk density (g cm⁻³); D is the soil depth (cm).

Stratification ratios were obtained by comparing SOC stocks and fractions between two soil layers: the 0–15 cm layer and the 15–30 cm layer. Specifically, we divided the values from the top layer (0–15 cm) by the values from the layer below (15–30 cm), as described by Franzluebbers and Stuedemann (2002).

A one–way analysis of variance (ANOVA) was employed to compare the impact of various LUSs on soil physicochemical properties (0–15 cm). Two–way ANOVA was used to compare the impact of various LUSs and depth on TOC, ROC, Carbon stocks, including LI, CPI and CMI. All statistical analyses were conducted using R software (Version 2024.12.1 + 563; R Core Team, 2024). Data analysis was performed using the “agricolae” package, while the “metan” package (multi–environment trial analysis) was utilized to create the Pearson’s correlation coefficient matrix.

The readily oxidizable carbon (ROC) pool in soil is more responsive to management practices compared to recalcitrant carbon pool (non–labile) (Table 3; Figure 2). Readily oxidizable carbon content in litchi was 1.72 g kg⁻¹ followed by rubber (1.67 g kg⁻¹), oil palm (1.56 g kg⁻¹), citrus (1.36 g kg⁻¹) and guava (1.35 g kg⁻¹) which was significantly higher than the uncultivated (A) at 0.69 g kg⁻¹, followed by uncultivated (B) at 0.73 g kg⁻¹. Among the agricultural LUSs, highest mean ROC was observed in lentil LUS (1.30 g kg⁻¹). At the 0–15 cm depth, litchi had the highest ROC content of 2.44 g kg⁻¹, while uncultivated (A) recorded the lowest value at 0.87 g kg⁻¹. Agricultural LUSs recorded the lowest average ROC values compared to tree–based systems. The ROC content generally decreased with increasing depth across all treatments. At the 15–30 cm depth, litchi continued to show the highest value of 2.05 g kg⁻¹. This trend of decreasing ROC content with increasing soil depth was observed across LUSs, with the lowest values found at the 75–100 cm depth. The ROC/TOC ratio with maize, lentil, and mustard showing the highest values, while uncultivated lands recorded the lowest ratio. Rice had a moderately high ratio, whereas oil palm, litchi, citrus, guava, and rubber exhibited similar mid–range values.

Horticultural LUSs exhibited higher soil MBC levels compared to uncultivated (fallow) land (Figure 3). The highest MBC values were recorded in litchi (368 mg kg⁻¹), rubber (358 mg kg⁻¹), and oil palm (338 mg kg⁻¹), while the lowest values were observed in uncultivated uses. The microbial quotient (%) ranged from 1.45 to 1.72%, with the highest values observed in guava (1.72%) and rubber (1.70%) and while uncultivated land (1.45%). The results indicate that perennial plantations maintain higher microbial biomass and efficiency compared to uncultivated or annual cropping systems. The MBC exhibited a positive correlation with TOC (Figure 4), indicating that even minor variations in TOC could influence MBC under different land–use practices. The MBC/TOC ratio in soils under horticultural land–use systems was higher compared to agricultural systems (Figure 3).

Table 4 displays the TOC contents across different LUSs and soil depths at the study site. In all examined LUSs, both oxidizable carbon and TOC exhibited a notable decline with increasing soil depth. The highest averaged TOC content of all the depths was observed in litchi at 16.6 g kg⁻¹, followed by rubber at 15.6 g kg⁻¹ and oil palm at 15.3 g kg⁻¹. The lowest TOC content was recorded in the uncultivated A treatment. In all the LUSs, TOC accumulation was greater in the 0–15 cm layer, with a subsequent decrease as soil depth increased. At the 0–15 cm soil depth, litchi exhibited the highest TOC content of 22.2 g kg⁻¹, while uncultivated B had the lowest at 11.5 g kg⁻¹.

Carbon stocks (Figure 5) showed significant variation due to differences in the TOC content, BD and volume of the soil layer. At 0–15 cm depth, the highest carbon stock is observed in litchi (38.3 Mg ha⁻¹), rubber (37.8 Mg ha⁻¹), and oil palm (37.5 Mg ha⁻¹), followed by guava (32.3 Mg ha⁻¹), citrus (31.9 Mg ha⁻¹), and ginger (32.2 Mg ha⁻¹). The lowest values are recorded in uncultivated A (23.8 Mg ha⁻¹) and uncultivated B (24.1 Mg ha⁻¹). At 15–30 cm depth, litchi (35.1 Mg ha⁻¹), rubber (34.5 Mg ha⁻¹), and oil palm (33.5 Mg ha⁻¹) maintain higher carbon stocks. Moreover, the sampling depths (cm) also affected this variation as the 30–60 cm covered a high volume of soil compared to the other layers (0–15, 15–30, 60–75 and 75–100 cm). As a result, we observed higher SOC stock values in 30–60 cm soil depth. At 30–60 cm depth, the highest values are found in litchi (65.7 Mg ha⁻¹), rubber (62.0 Mg ha⁻¹), and oil palm (57.5 Mg ha⁻¹), while rice (41.7 Mg ha⁻¹), maize (44.3 Mg ha⁻¹), and mustard (43.2 Mg ha⁻¹) show lower stocks. Based on the stratification ratio (SR), approximately half of the TOC and SOC stocks were found within the surface soil layer (0–15 and 0–30 cm) (Figure 6). The SR values for TOC ranged from 1.14 to 1.22, while for SOC stocks, they varied between 1.07 and 1.20. Notably, these values were lower in uncultivated land compared to agricultural and horticultural systems, indicating differences in carbon distribution across land uses.

To study the influence of different land uses on the nature of SOC, particularly its lability, the lability index (LI) was computed (Figure 7). Rubber plantation consistently exhibited the highest LI across all depths, with values ranging from 1.54 at 0–15 cm to 1.45 at 75–100 cm. In contrast, rice consistently demonstrated the lowest LI across all depths, ranging from 1.38 at 0–15 cm to 0.98 at 75–100 cm. Other crops such as litchi, citrus, guava, oil palm, ginger, maize, lentil, and mustard showed intermediate LI values. Specifically, litchi ranged from 1.50 to 1.33, citrus from 1.52 to 1.25, guava from 1.48 to 1.18, oil palm from 1.47 to 1.27, ginger from 1.49 to 1.25, maize from 1.45 to 1.14, lentil from 1.41 to 1.13, and mustard from 1.44 to 1.10. Across all systems, the LI generally showed a decreasing trend with increasing soil depth.

Among the land use systems, litchi recorded the highest mean carbon pool index (CPI) across all depths (1.73), significantly greater than all other treatments (Figure 8). This was followed by rubber (1.62) and oil palm (1.59), indicating their relatively higher capacity to enhance soil carbon pools. Mustard showed moderate CPI values, while the lowest mean CPI was observed in rice (1.18), suggesting a limited contribution to carbon sequestration. CPI values consistently decreased with increasing soil depth, with the highest at 0–15 cm, followed by 15–30 cm, 30–60 cm, and the lowest at 75–100 cm (1.24), reflecting the reduced influence of surface land use practices in deeper soil layers. A significant interaction (A × B) between land–use systems and soil depths revealed differential CPI responses across profiles, with litchi maintaining higher CPI at all depths, while rice and ginger exhibited lower values.

The carbon management index (CMI) was significantly influenced by land–use systems, soil depths, and their interaction (Table 5). Among the land–use systems, litchi recorded the highest mean CMI value (245), followed by rubber (238) and oil palm (222). In contrast, rice showed the lowest mean CMI (143), indicating limited carbon management. Ginger and lentil also exhibited relatively low CMI values of 179 and 173, respectively, while guava and mustard showed moderate values of 192 and 165. CMI values exhibited a consistent decline with increasing soil depth: 225 at 0–15 cm, 213 at 15–30 cm, 198 at 30–60 cm, 174 at 60–75 cm, and 150 at 75–100 cm. Tree–based land use systems consistently displayed higher CMI values than agricultural systems, though all remained above 100, reflecting their potential in improving soil quality and carbon stabilization.

The Pearson’s correlation coefficients reveal (Figure 4) significant interrelationships among SOC, MWD, TOC, non-labile carbon (NLC), ROC, and C stocks, with coefficients ranging from 0.92 to 0.98 (p < 0.001). This suggests a robust, integrated carbon storage system where MWD and TOC support stable NLC and labile ROC fractions, enhancing C stocks. Moisture is identified as a key driver, exhibiting strong positive correlations with MBC (0.70, ***), SOC (0.75, ***), MWD (0.82, ***), TOC (0.84, ***), NLC (0.78, ***), ROC (0.81, ***), and C stocks (0.84, ***), underscoring its role in sustaining carbon pools. MBC further supports SOC (0.76, ***), MWD (0.84, ***), and other fractions, emphasizing microbial activity’s contribution. Bulk density shows negative correlations (−0.66 to −0.90, −) with carbon parameters, suggesting that soil compaction may impede carbon retention. pH displays weak, variable effects (0.23–0.45, ns-*), while CMI (0.45–0.61, −) is posited to enhance carbon management across the system. Readily oxidizable carbon and TOC exhibit a very strong positive correlation (Pearson’s coefficient = 0.97***), indicating their close association within the soil carbon pool (Figure 4).

In this study, the readily oxidizable carbon (ROC) pool showed greater responsiveness to management practices compared to the recalcitrant carbon pool (non-labile). Higher SOC levels in certain systems may be attributed to increased litter biomass on the soil surface and reduced oxidation in less disturbed soils (Pradeepa et al., 2018). Agricultural systems recorded the lowest average ROC values, primarily due to limited organic inputs from practices like crop residue removal, inadequate biomass incorporation, and conventional tillage (Wang et al., 2017; Thangavel et al., 2018; Pradeepa et al., 2018). Nutrient management treatments significantly impacted ROC levels within a short period, underlining the importance of proper nutrient management in maintaining soil carbon stocks. The correlation between ROC and TOC revealed a positive relationship, indicating their close association in the soil carbon pool. Land use type also plays a critical role, with forest systems contributing the largest carbon pools. As expected, plantation soils exhibited higher ROC levels due to higher organic inputs from tree litter and reduced disturbance. The mobilization and redistribution of soil organic components determine the qualitative and quantitative characteristics of SOM, emphasizing the influence of land use and management practices on soil carbon dynamics.

In tropical and subtropical soils, soil MBC is crucial as a readily available nutrient source for plants (Farooq et al., 2022). Horticultural LUSs exhibited higher soil MBC levels compared to uncultivated (fallow) land. It indicates that perennial plantations maintain higher microbial biomass and efficiency compared to uncultivated or annual cropping systems. The fluctuation and abundance of microbial populations beneath trees are influenced by litter quality, nutrient availability, and alterations in soil properties (Ramesh et al., 2013; Liu et al., 2021). Trees play a crucial role in maintaining favorable environmental conditions, such as higher moisture content and improved physical attributes. However, guava and citrus exhibited the lowest soil MBC among the studied tree plantations, likely due to limited carbon availability when compared to other plantations. As expected, the uncultivated soil showed minimal microbial activity, emphasizing the direct correlation between lower carbon input and reduced microbial presence (Thangavel et al., 2018). The higher MBC observed in orchards was attributed to increased soil moisture, nutrient availability, and SOC levels. Similarly, Li et al. (2021) observed that MBC is directly proportional to SOC amount across various agroecosystems. Shah et al. (2024) also emphasized that soil MBC demonstrated a significantly positive correlation with soil moisture, a trend that aligns with the findings of this study.

Variations in the microbial quotient (MBC/TOC ratio) reflect the input of organic matter into the soil, the efficiency of microbial integration, soil carbon losses, and the stabilization of organic carbon by soil minerals (Sparling, 1992). Soils exhibiting higher MBC/TOC values indicate improved quality. The greater MBC in horticultural soils could be a key factor contributing to the increased microbial quotient, indicating higher organic matter inputs and greater availability of SOM for microbial activity (Araujo et al., 2020; Yan et al., 2022; Ding et al., 2024). The MBC/TOC ratio remains unstable over time and is largely influenced by organic matter inputs (Anderson and Domsch, 1989). Moreover, this ratio tends to increase as SOM decomposes under microbial activity (Bohme and Bohme, 2006). Additionally, the microbial quotient is strongly influenced by carbon inputs from the rhizosphere, land use changes, and microbial biomass content (Lv et al., 2023).

Soil management practices and land use patterns play a crucial role in shaping SOC dynamics and the carbon flux from soil (Ahirwal et al., 2021; Swamy et al., 2023; Wang et al., 2023). In all examined LUSs, both oxidizable carbon and TOC exhibited a notable decline with increasing soil depth, likely attributable to reduced surface litter contributions in deeper layers (Jobbagy and Jackson, 2000; Meena et al., 2018). However, soil TOC content is highly dependent on both plant productivity and the amount of litter inputs (Cui et al., 2022; Manici et al., 2024). There is a clear link between SOC and TOC, indicating that they increase and decrease together. This insight helps us better understand soil carbon dynamics. Higher SOC in perennial crops was might be due to accumulation of leaf litter, root exudates and tree root biomass; poplar trees on an average add 3.5 Mg ha⁻¹ of leaf litter to the soil every year (Ralhan et al., 1996). Along with tree components, straw biomass from intercrops also adds some organic matter which forms the soil carbon (Marone et al., 2017). This decline in SOC (at lower depth) under agricultural LUSs based on crops like maize could be due to low root to shoot ratio as well as relatively shallow rooting depth (Zhao et al., 2018). The elevated TOC content in forestry and horticulture soils could be attributed to substantial yearly inputs of organic matter through leaf fall (Jangir et al., 2024), extensive biomass of root, and tree rhizodeposition (Li et al., 2024). It was thought that the bare soil and the breaking up of soil clumps due to the lack of vegetation, along with more erosion on slopes with heavy rainfall, might be the reasons f or the lowest TOC content in the control plot (Ramesh et al., 2015).

Carbon stock showed significant variation due to differences in the TOC content, BD and volume of the soil layer. At 15–30 cm depth, litchi, rubber, and oil palm maintain higher carbon stocks, due to the higher fresh or new carbon supply from the upper soil layer (Fontaine et al., 2007). Moreover, the sampling depths also affected this variation as the 30–60 cm covered a high volume of soil compared to the other. The correlation analysis (Figure 4) highlights a strong positive relationship between non–labile carbon and carbon stocks (dark blue, highly significant), indicating that as NLC increases, C stocks also rise significantly. The increased resistance of SOC to oxidation in deeper (lower) soil layers suggests a longer turnover time for SOC in sub–surface horizons (Gaudinski et al., 2000). Greater inputs of lignin and lipids from root litter likely enhanced the resistance of SOC in deeper soil horizons. Forests, above–ground inputs primarily comprise leaf and needle litter, with additional contributions from branches, bark, and fruits to the overall litter fall (Lorenz and Lal, 2005). In short, plantation crops showed a healthier carbon sink in the 0–100 cm depth compared to the control, indicating their potential for carbon sequestration. The type and amount of root exudates, along with the quality of plant and root litter (such as lignin and polyphenol content), influence the decomposition rate and SOC accumulation at different depths under various LUSs (Lal, 2009). A significant portion of carbon exists in stable forms (TOC–MBC–ROC), which are less susceptible to decomposition and crucial for long–term carbon storage, especially in tree–based systems. Tree species contribute to erosion control, organic matter accumulation, and improved nutrient cycling (Thangavel et al., 2018), thereby influencing TOC and SOC variations across soil depths. Additionally, root–derived compounds such as lignin and polyphenols play key roles in organic matter decomposition (Lal, 2009; Srinivasan et al., 2012). Studies have shown that undisturbed surface soils create favorable conditions for organic matter decomposition, leading to higher SOM accumulation (Leul et al., 2023). Variations in stratification ratios across SOC pools also highlight their interdependent nature (Franzluebbers and Stuedemann, 2002; Kurganova et al., 2021).

The observed decline in lability index (LI) with depth is likely due to the decreasing availability of labile carbon and the relative increase in more stable, recalcitrant carbon fractions in deeper soil layers (Liu et al., 2022). The higher LI values in cultivated soils, particularly in upper horizons, may be attributed to the decomposition of surface leaf litter, which produces dissolved organic matter that is then transported downslope through erosion and runoff (Reza et al., 2020). Interestingly, soils under litchi plantations exhibited a relatively stable LI across depths, which could be due to the continuous input of easily decomposable organic matter from litchi leaf litter (Lorenz and Lal, 2005). These findings highlight the role of vegetation type and land use in influencing the lability of SOC through variations in litter quality and soil management practices. Higher CPI values are indicative of greater SOC accumulation, which is associated with increased carbon inputs and reduced disturbance (Parihar et al., 2018). Land use systems like litchi, rubber, and oil palm likely contribute more organic matter (e.g., leaf litter, root biomass), promoting SOC buildup. Conversely, the lower CPI observed under rice and ginger may be due to factors such as residue removal, intensive tillage, or shorter rooting systems. According to Lal (2018), both carbon input levels and mineralization rates critically influence SOC dynamics. Moreover, conservation agriculture practices enhance carbon stabilization through the formation of soil aggregates, which physically protect SOC from microbial decomposition (Six et al., 2000). These findings underscore the importance of integrating land use strategies that minimize soil disturbance and maximize organic matter return for sustained carbon sequestration.

The high carbon management index (CMI) values under litchi, rubber, and oil palm suggest enhanced carbon inputs and a faster decomposition rate of organic residues, likely due to better litter quality and below–ground biomass dynamics (Table 5). In contrast, the lower CMI observed in rice, ginger, and lentil points to reduced organic matter input and possibly higher disturbance levels. Studies by Zhang et al. (2021) and Murindangabo et al. (2023) have shown that CMI, when combined with SOM lability indicators, offers a more sensitive tool for detecting changes in SOM status than total carbon stocks alone. Additionally, the decline in CMI with depth aligns with reduced root activity and organic matter input in subsoil layers. Positive correlations between CMI and carbon–related variables (Figure 4) further suggest that CMI is governed not only by the quantity of organic matter but also by its quality—including factors like the C/N ratio and chemical characteristics—which influence SOM lability and stabilization (Thangavel et al., 2018).