Analyzing the Trade-Offs between Soil Health Enhancement, Carbon Sequestration, and Productivity in Central India’s Black Soil through Conservation Agriculture

Abstract

The impact of conservation tillage (CST) practices on soil properties, carbon sequestration and yield sustainability over short, medium, and long durations remain insufficiently understood, especially in semiarid Central India. Therefore, our objective was to investigate the effects and optimal duration of CST adoption for enhancing soil properties, carbon sequestration, and sustainable yields. We conducted a study in farmers’ fields in the Akola district of Central India, where CST had been practised for 4 to 15 years, within a soybean + pigeon pea–chickpea cropping sequence. Our findings revealed significant (p < 0.05) improvements in soil physical properties with short-term CST practices (4 to 6 years), alongside increasing availability of nitrogen and phosphorus, with longer durations of CST implementation (10 to 15 years). The lowest soil organic carbon (SOC) was observed in conventional tillage (CT_y), while all CST practices increased SOC content over CT_y, ranging from 22.2 to 38.4%. Further, experimental soil dominated passive C pools (Cfrac3 + Cfrac4). Consequently, long-term CST practices facilitated positive C sequestration rates, contrasting with negative or minimal sequestration observed in CT_y and short-term CST treatments. However, compared to CST, CT_y demonstrated higher soybean equivalent yields and comparable chickpea equivalent yields mainly due to delayed germinations induced by lower soil temperatures in CST plots. We conclude that integrating site-specific characteristics, management practices, and regional climate conditions into conservation agriculture frameworks maximizes efficacy and ensures sustainable productivity. These findings help optimize agricultural practices considering potential yield losses or minimal changes despite implementing CST.

1. Introduction

Plough-based agricultural practices have significantly exacerbated several critical environmental issues, including accelerated soil erosion by water and wind, oxidation of soil organic matter (SOM), and the deterioration of soil structure and tilth [1]. Notably, historical events such as the Dust Bowl of the 1930s in the United States [2] and more recent instances in China since 2000, along with challenges such as excessive water withdrawal from aquifers like those in the Indo-Gangetic Plains of South Asia [3,4] underscore the detrimental impacts of plough-based farming methods on soil health and the environment. Consequently, there has been an increasing interest in developing plough-less agricultural methods, potentially reducing the impact on soil health and the environment [5,6].

In India, soil degradation poses a significant threat to sustainable agriculture, primarily due to harmful farming practices such as conventional tillage (CT), monoculture, and the removal of crop residues after harvest [7,8,9]. Furthermore, intensive agricultural methods often result in alterations to key soil health indicators, including soil structure, aggregation, porosity, strength, hydraulic conductivity (HC), infiltration, bulk density (BD), soil water content, soil organic carbon content (SOC), and microbial biomass and activity [10]. Healthy soil is essential for achieving optimal crop yields, particularly in favourable and extreme climatic conditions [11]. Soil health also plays a crucial role in crop adaptation and mitigation of the effects of climate change [11]. Therefore, optimizing tillage practices in Central India to prevent land degradation, improve soil health, maintain crop yields, and ensure ecosystem stability is imperative.

Conservation agriculture (CA) practices, which involve minimal tillage, maintaining permanent soil cover, and diversifying crops while optimizing the use of inorganic fertilizers, have increased SOM [12,13,14,15,16]. These practices have emerged as a sustainable production system that preserves soil health and ensures profitability [17]. Further, the negative environmental impacts of CT systems have prompted a shift towards conservation practices. These practices enhance soil properties and save time, energy, and water while protecting against water and wind erosion [18,19]. As a result, conservation tillage (CST) practices—such as reduced tillage, minimum tillage, and no-tillage—have become widely embraced as effective management techniques for enhancing soil health and SOM [20], maintaining productivity, and addressing climate change [1].

Despite the increasing recognition of CST in tropical regions, ongoing debates persist regarding its profitability and yield benefits compared to conventional practices. Numerous studies have questioned whether no-tillage effectively enhances soil carbon (C) levels and sustains crop productivity [21]. For instance, Ogle [22] found that in about 10% of the experimental treatments that they included in their meta-analysis, SOC did not increase. Moreover, several studies have reported decreased crop yields following no-till adoption [23,24,25,26]. Similarly, few studies indicate no change or even an increase in crop yields [27,28,29,30,31,32,33].

Semiarid regions worldwide face challenges due to significant fluctuations in rainfall and distribution, coupled with relatively less fertile soils. In the Vidarbha region of Central India, the intercropping of Soybean + Pigeon pea (in kharif) and Chickpea (in rabi) is a vital cropping system promoted as a climate-resilient pattern. However, the productivity of these systems falls short of their potential in the rainfed semiarid region. This shortfall can be attributed to the diverse topo-sequences of Black soils in Central India, characterized by poor tilth, limited drainability, and high concentrations of calcium or magnesium carbonates [34,35]. Hence, exploring the impacts of CST practices and residue retention on soil properties, C sequestration, and the productivity of cropping systems in this region is essential.

Numerous studies have examined the effects of CA on soil health, C dynamics, and crop productivity across different temporal scales worldwide. However, a significant gap exists in the literature regarding a comprehensive assessment of the short-term, medium-term, and long-term impacts of CA adoption on soil health, C pools, and C sequestration. Furthermore, research on CA in Central India’s Vertisol is limited. Hence, there is a critical need to research Central India’s Vertisol to enhance our understanding of the consequences of CA adoption. Therefore, we hypothesize that long-term adoption of CST practices could potentially lead to improved soil health, C sequestration, and sustainable crop yields. Therefore, in light of the preceding notion, the present study aims to address two specific questions related to CA adoption in Central India: (i) investigate the impact of CA practices on soil properties, C sequestration, and crop yields, and (ii) determine the optimal duration of CA adoption required to improve soil properties, C sequestration, and sustainable crop production.

2. Materials and Methods

2.1. Experimental Location and Climate

The experimental study was conducted between 2018–2019 and 2019–2020 in various villages across the Akola district of Maharashtra, India. These villages were selected for their representation of different farming conditions. Akola is in the sub-tropical region, with approximately 22°41′ N latitude and 77°02′ E longitude coordinates. The altitude of the area is approximately 304.42 m above sea level. The climate in Akola is semiarid, characterized by three main seasons. The summer months, from March to May, are hot and dry. This is followed by the warm and humid rainy season from June to October. Finally, the winter months, from November to February, are relatively mild and cold. Over the last fifteen years, the average annual precipitation in Akola has been recorded at 515.8 mm. Most rainfall occurs during the Southwest monsoon season, which lasts from June to October, contributing to a mean annual precipitation of 818.6 mm. The hottest month in Akola is May, with an average mean monthly maximum temperature of 42.6 °C. Conversely, the coldest month is December, with a mean monthly minimum temperature of 10.8 °C.

2.2. Treatment Details and Management Practices

We selected farmers’ fields within a 2–3 km radius based on the historical crop management practices observed in the area. For instance, farmers in these villages typically cultivate soybean and pigeon peas as intercrops during the kharif season and chickpeas during the rabi season. Most farmers in these villages follow the recommended fertilizer dosage provided by the university. The main variation among them lies in their duration of adopting CST practices. Some farmers have been practising CST and residue incorporation for up to 15 years, while others have been practising it for 12, 10, 8, 6, and 4 years, respectively. The duration of CST adoption (years) is relative to soil sampling in the second experimental year, i.e., 2019–2020. Residues were incorporated after the harvest of the previous crop into the next subsequent crop. This variation in CST adoption is the foundation for selecting farmers for our experimental investigation. In addition to these conservation practices, few farmers continued with CT methods. Consequently, selected farmers’ fields represent a different tillage practice. Further details about field management are given in

Table 1. Details of management practices.

Sites (Location)	Area (Acres)	Treatment Abbreviation	Treatment Details	Cropping Pattern
				kharif	rabi
Site1 20°35′16.0″ N 77°02′10.7″ E	2	CT_y	Conventional tillage every year: regular ploughing and harrowing each year	Soybean + Pigeon pea	Chickpea
Site 2 20°35′11.71″ N 77°02′13.57″ E	1.2	CST_4y	Conservation tillage, no ploughing and harrowing for 4 years, crop residue incorporated	Soybean + Pigeon pea	Chickpea
Site 3 20°35′12.46″ N 77°02′1.70″ E	1.5	CST_6y	Conservation tillage, no ploughing and harrowing for 6 years, crop residue incorporated	Soybean + Pigeon pea	Chickpea
Site 4 20°35′6.007″ N 77°02′1.222″ E	1.5	CST_8y	Conservation tillage, no ploughing and harrowing for 8 years, crop residue incorporated	Soybean + Pigeon pea	Chickpea
Site 5 20°35′14.68″ N 77°02′1.744″ E	2.5	CST_10y	Conservation tillage, no ploughing and harrowing for 10 years, crop residue incorporated	Soybean + Pigeon pea	Chickpea
Site 6 20°35′6.46″ N 77°01′54.59″ E	2.5	CST_12y	Conservation tillage, no ploughing and harrowing for 12 years, crop residue incorporated	Soybean + Pigeon pea	Chickpea
Site 7 20°35′8.977″ N 77°02′20.63″ E	2.0	CST_15y	Conservation tillage, no ploughing and harrowing for 15 years, crop residue incorporated	Soybean + Pigeon pea	Chickpea

.

2.3. Experimental Soil, Sample Collection and Analysis

According to the USDA soil classification, the soils at experimental sites are classified as Typic Haplustert (order: Vertisols). These soils are characterized as moderately deep to deep black-cracking clay soils. Due to farmers’ varying adoption periods of CST practices, ranging from 4 to 15 years, initial baseline data specific to each tillage practice were unavailable. To address this, we collected reference soil samples from adjacent non-cropped areas within the same experimental fields. These non-cropped areas had no history of tillage interventions and served as a baseline to estimate pre-cultivation soil conditions.

To complement these baseline samples, we also utilized historical data from earlier village site-specific fertility reports (https://www.pdkv.ac.in/?page_id=128; https://nbsslup.icar.gov.in/ accessed on 15 March 2019). These historical data were compared with current soil data collected from experimental fields under various CST durations. By analyzing SOC stocks and other soil properties from these non-cropped reference areas, we ensured that the baseline conditions aligned with expected ranges from previous studies. The non-cropped reference areas, which were not under forest or pasture management but remained uncultivated, provided a crucial point of comparison. This approach enabled us to evaluate the impact of CST practices on soil properties over time by establishing a reference for pre-cultivation conditions and understanding changes resulting from different CST durations. In general, the pH of the soil was 8.19, electrical conductivity (EC) 0.27 dSm⁻¹, SOC 6.37 g kg⁻¹, BD 1.40 Mg m⁻³, mean weight diameter (MWD) 0.68 mm, HC 0.71 cm hr⁻¹, available nitrogen (N) 193.90 kg ha⁻¹, phosphorus (P) 14.15 kg ha⁻¹, and potassium (K) 395.3 kg ha⁻¹.

Surface soil samples (0.15 m) were collected after the harvest of crops in the kharif and rabi seasons during 2018–2019 and 2019–2020. A 0.15 m depth was selected because it captures the topsoil layer where most biological activity and nutrient dynamics occur, which is crucial for the conservation agricultural based studies. Composite soil samples (5–6) were taken from each farmer’s field to account for spatial variability and provide a more accurate measurement of the soil properties. Soil sampling was performed using an auger and immediately transferred to the laboratory. These samples were air-dried in the shade and processed through a 2 mm diameter sieve after crushing.

The soil properties were analysed using standard procedures. Soil BD was assessed by the clod coating method [36] which involves coating undisturbed soil clods with a thin layer of a material like wax or resin. The volume of the clod is measured, usually by the water displacement method, and the BD is calculated by dividing the mass of the dried soil by its volume. HC was measured via the constant head method, which involves maintaining a constant water level in a soil column and measuring the flow rate of water through the soil [37]. MWD was determined using Yoder’s apparatus, which involves sieving a soil sample through a series of mesh screens to separate particles into different size fractions [38]. SOC was analysed using the wet oxidation method [39]. This method involves treating the soil with potassium dichromate and sulfuric acid and measuring the amount of dichromate reduced to determine C content. Available N was determined using the alkaline potassium permanganate method [40], which involves treating soil samples with potassium permanganate in an alkaline medium and measuring the N released. Available P was measured using 0.5M sodium bicarbonate at pH 8.5, in which phosphorus is extracted from the soil and quantified by colourimetric analysis [41]. Available K was analyzed using extractant neutral normal ammonium acetate at pH 7.0 and measuring its concentration on a flame photometer [42].

SOC fractions were analyzed using a modified Walkley and Black method, as outlined by Chan [43]. This approach involves measuring the oxidizability of organic C under various concentrations of sulfuric acid (H₂SO₄) solutions.

Cfrac1: Organic C oxidizable under a 6.0 mol L⁻¹ H₂SO₄ solution.

Cfrac1 represents the fraction of SOC that is readily oxidizable under a relatively mild acid concentration. This fraction is typically associated with more labile and easily decomposable organic C forms.

Cfrac2: Difference between the organic C oxidizable under 9.0 mol L⁻¹ H₂SO₄ and that oxidizable under 6.0 mol L⁻¹ H₂SO₄. This fraction usually corresponds to moderately labile organic C, which is less readily decomposable than Cfrac1 but more susceptible than Cfrac3.

Cfrac3: Difference between the organic C oxidizable under 12.0 mol L⁻¹ H₂SO₄ and that oxidizable under 9.0 mol L⁻¹ H₂SO₄. This fraction typically represents more stable organic C forms, which are less decomposable but still reactive under stronger acid conditions.

Cfrac4: Difference between the total SOC and the organic C oxidizable under 12.0 mol L⁻¹ H₂SO₄. Cfrac4 signifies the residual organic C that remains after treatment with 12.0 mol L⁻¹ H₂SO₄. This fraction includes the most stable and recalcitrant organic C forms, which are resistant to oxidation even under the strongest acid conditions.

These fractions help to differentiate between various levels of organic C stability and reactivity in the soil. By understanding the distribution of SOC across these fractions, one can gain insights into the turnover rates and stability of organic C in the soil.

The carbon management index (CMI) was calculated using the equations given by Venkatesh [44] based on oxidizable C fractions and TOC as described below:

$$LI=\left({\displaystyle \frac{Cfrac1}{total SOC}}\right)\times 3+\left({\displaystyle \frac{Cfrac2}{total SOC}}\right)\times 2+\left({\displaystyle \frac{Cfrac3}{total SOC}}\right)\times 1$$

(1)

$$CPI={\displaystyle \frac{sample total C \left(\mathrm{m}\mathrm{g} {\mathrm{k}\mathrm{g}}^{-1} \mathrm{s}\mathrm{o}\mathrm{i}\mathrm{l}\right)}{reference total C \left(\mathrm{m}\mathrm{g} {\mathrm{k}\mathrm{g}}^{-1} \mathrm{s}\mathrm{o}\mathrm{i}\mathrm{l}\right)}}$$

(2)

where reference total C is the total C content (mg kg⁻¹) of control plots [45].

$$CMI=CPI\times LI\times 100$$

(3)

SOC stock measurement:

To calculate SOC stock, we used the following formula based on Bhattacharyya [46]

$$SOC stock \left(\mathrm{M}\mathrm{g} \mathrm{C} {\mathrm{h}\mathrm{a}}^{-1}\right)=SOC \left(\%\right)\times {10}^2\times BD (\mathrm{M}\mathrm{g} {\mathrm{m}}^{-3})\times soil depth \left(\mathrm{m}\right)$$

(4)

where SOC (%) is the percentage of soil organic carbon content, measured from soil samples, BD (Mg m⁻³) is the bulk density of the soil, and soil depth (m) is the depth of the soil layer considered for the measurement (0.15 m).

• SOC Sequestration Measurement:

The amount of SOC sequestered was determined using the following calculation:

$$C sequestered (\mathrm{M}\mathrm{g} \mathrm{C} {\mathrm{h}\mathrm{a}}^{-1})=SO{C}_f-SO{C}_i$$

(5)

where SOC_f is SOC stock at the end of the study period (2020) and SOC_i is SOC stock of the baseline sample. Positive values indicate SOC gains (sequestration), while negative values indicate SOC losses.

• Equivalent Yield:

Equivalent yield is used to compare the yield of different crops by converting their yields into the yield of a single reference crop, often based on economic value or market price. It allows for a standardized comparison of different cropping systems.

$$Crop equivalent yield=\sum _{i=1}^n\left({\displaystyle \frac{Yi\times Pi}{Pr}}\right)$$

(6)

where Yi is the yield of the ith crop, Pi is the price of the ith crop, Pr is the price of the reference crop and n is the number of crops.

2.4. Biomass and Carbon Input

The annual biomass input in the experimental soil was quantified from soybeans, pigeon pea, and chickpeas. The annual biomass included leaf litter, stubbles and roots and rhizodeposition. Leaf litter and stubble biomass under the soybean + pigeon pea and chickpea cropping systems were measured within 1 × 1 m² plot areas and then extrapolated to hectares. C inputs from leaf litter and stubble biomass were estimated based on the C content in different plant parts. The C content in plant samples was determined using the ignition method [47]. Root biomass was measured immediately after crop harvesting using the core-sampling procedure [48]. Rhizodeposition of C from root turnover and exudates was assumed to be 10% of the harvestable above-ground biomass of the crop [49]. The C concentrations of leaf litter and stubble of soybean and pigeon pea ranged between 38–40% and 40–44%, respectively, while those of chickpea ranged between 49–50% and 38–40%. Weeds were managed by removal or herbicide application during crop growth; thus, C inputs from roots and rhizodeposition by weeds were not considered.

2.5. Statistical Analysis

Statistical analysis was carried out using one-way analysis of variance (ANOVA) [50]. Duncan’s multiple range tests (DMRT) were employed to compare the means across multiple treatments with a 95% confidence level. To examine the associations between various parameters, stepwise Pearson’s correlation and regression models were created using “R Studio, v. 3” [51]. Graphical representations were produced using Microsoft^® Excel^® 2021 MSO (Version 2408 Build 16.0.17928.20114) 64-bit, the “Plot function”, and the “ggplot2” package within R Studio.

3. Results

3.1. Biomass and Carbon Input in Soil

The total annual biomass input from soybean, pigeon pea and chickpea ranged between 2.28–2.83, 2.18–2.81 and 1.06–1.44 Mg ha⁻¹, respectively (

Table 2. Annual crop-mediated carbon input in soil.

Treatments	Biomass (Mg ha−1)				Carbon Input (Mg ha−1)		Total C Input
	Soybean	Pigeon Pea	Chickpea	Total
	LL + S	LL + S	LL + S		Soybean–Pigeon Pea	Chickpea
CT_y	2.28	2.18	1.06	5.52	1.95	0.5	2.45
CST_4y	2.59	2.46	1.19	6.24	2.21	0.56	2.78
CST_6y	2.61	2.51	1.22	6.34	2.24	0.58	2.82
CST_8y	2.67	2.57	1.28	6.52	2.29	0.61	2.9
CST_10y	2.69	2.62	1.35	6.66	2.32	0.64	2.96
CST_12y	2.71	2.73	1.4	6.84	2.38	0.66	3.04
CST_15y	2.83	2.81	1.44	7.08	2.46	0.68	3.14

LL: leaf litter; S: stubbles including roots and rhizodeposition.

). This annual biomass was used to calculate the respective C input. The total C input from soybean + pigeon pea intercropping ranged between 1.95 to 2.46 Mg ha⁻¹, while chickpea residue ranged between 0.5 to 0.68 Mg ha⁻¹. The total C input, including soybean + pigeon pea and chickpea, ranged between 2.45 Mg ha⁻¹ in CT_y to the highest of 3.14 Mg ha⁻¹ in CST_15y treatment. Further, all CST treatments had comparatively higher total C input than CT_y.

3.2. Soil Physicochemical Properties

Different tillage practices significantly affected the soil’s physicochemical properties (

Table 3. Effect of conventional and conservation tillage practices on soil physicochemical properties.

Treatments	BD (Mg m−3)	MWD (mm)	HC (cm h−1)	SOC (g kg−1)	Available N (kg ha−1)	Available P (kg ha−1)	Available K (kg ha−1)
CT_y	1.45 ± 0.01 a	0.63 ± 0.03 b	0.69 ± 0.04 b	5.13 ± 0.12 b	173 ± 6.7 b	9.61 ± 0.13 d	365.77 ± 14.3 b
CST_4y	1.40 ± 0.02 ab	0.68 ± 0.01 ab	0.71 ± 0.04 ab	6.27 ± 0.09 a	190 ± 8.9 ab	12.16 ± 0.75 cd	405.00 ± 17.2 a
CST_6y	1.38 ± 0.02 b	0.69 ± 0.02 ab	0.72 ± 0.02 ab	6.47 ± 0.15 a	196 ± 11.5 ab	14.14 ± 0.38 c	402.33 ± 19.2 a
CST_8y	1.38 ± 0.01 b	0.69 ± 0.02 ab	0.72 ± 0.00 ab	6.67 ± 0.03 a	198 ± 21.4 ab	17.73 ± 0.56 b	407.87 ± 12.4 a
CST_10y	1.37 ± 0.03 b	0.70 ± 0.01 ab	0.72 ± 0.01 ab	6.92 ± 0.06 a	203 ± 13.3 ab	17.82 ± 0.18 b	407.27 ± 11.4 a
CST_12y	1.36 ± 0.02 b	0.70 ± 0.02 ab	0.73 ± 0.03 a	7.10 ± 0.17 a	206 ± 8.8 a	18.35 ± 0.64 ab	410.57 ± 16.5 a
CST_15y	1.34 ± 0.01 b	0.71 ± 0.01 a	0.73 ± 0.02 a	7.13 ± 0.11 a	221 ± 3.9 a	20.30 ± 0.33 a	416.15 ± 9.8 a
LSD (p < 0.05)	0.06	0.07	0.03	0.96	32.2	2.6	30.29

Values are expressed as means ± standard deviation. In a column, values marked with the same letter are not significantly different from each other at p < 0.05, according to Duncan’s multiple range test (DMRT).

). Soil BD ranged from 1.34 to 1.45 Mg m⁻³. It was highest in CT_y treatment and reduced significantly in CST treatments, although the magnitude differed based on CST adoption years (4 to 15 years). However, the decrease in BD in CST treatments was statistically at par.

MWD ranged between the lowest of 0.63 mm in CT_y to the highest of 0.71 in CST_15y (Table 3). Following CST for 4 years (CST_4y), MWD improved significantly and was statistically equivalent for the rest of the CST treatments. Similar improvement was also observed for the HC of soil. HC was lowest in CT_y treatment and statistically equivalent in CST_4y, CST_6y, CST_8y and CST_10y treatments but significantly increased to 0.73 cm hr⁻¹ in CST_12y and CST_15y.

Available macro-nutrients in soil (N, P, and K) improved significantly following CST practices (Table 3). The available N, P and K in the present experimental soil ranged between 173–221, 9.61–20.30, and 365.77–416.15 kg ha⁻¹, respectively. The content of these nutrients was lowest under the CT_y treatment. Available N in CST treatment was higher by 9.8 to 27.7%, and the CST treatments were statistically at par. Similarly, available P and K in CST treatments were higher by 26.5–111% and 10–13.7%, respectively, compared to their respective CT_y treatments.

3.3. Soil Organic Carbon and Its Fractions

Different tillage practices showed a marked increase in SOC content ranging between 5.13 g kg⁻¹ to 7.13 g kg⁻¹ (Table 3). The lowest SOC was observed in CT_y treatment, while all CST treatments significantly increased SOC content over CT_y, ranging from 22.2 to 38.4%. Similarly, tillage practices significantly affected the C fractions of varying oxidizibility, i.e., Cfrac1, Cfrac2, Cfrac3, and Cfrac4 (

Table 4. Effect of conventional and conservation tillage practices on organic carbon fractions of varying oxidizibility.

Treatments	Cfrac1 (g kg−1)		Cfrac2 (g kg−1)		Cfrac3 (g kg−1)		Cfrac4 (g kg−1)
	Soybean	Chickpea	Soybean	Chickpea	Soybean	Chickpea	Soybean	Chickpea
CT_y	0.91 ± 0.03 e	0.28 ± 0.06 c	0.67 ± 0.05 c	0.46 ± 0.07 e	0.44 ± 0.04 e	0.56 ± 0.07 g	2.98 ± 0.04 d	2.01 ± 0.07 e
CST_4y	1.32 ± 0.01 d	0.67 ± 0.02 b	1.05 ± 0.05 b	1.02 ± 0.08 cd	0.83 ± 0.07 d	1.45 ± 0.06 d	3.51 ± 0.06 c	2.77 ± 0.05 d
CST_6y	1.36 ± 0.02 d	0.78 ± 0.06 b	1.10 ± 0.06 b	1.04 ± 0.06 cd	0.89 ± 0.06 cd	0.88 ± 0.01 f	4.51 ± 0.04 b	3.25 ± 0.06 cd
CST_8y	1.60 ± 0.03 c	0.93 ± 0.04 a	1.04 ± 0.04 b	1.14 ± 0.06 cd	0.86 ± 0.07 cd	1.17 ± 0.06 e	4.57 ± 0.08 b	3.71 ± 0.04 bc
CST_10y	1.78 ± 0.02 c	0.97 ± 0.06 a	1.14 ± 0.07 b	1.18 ± 0.04 c	1.01 ± 0.02 bc	2.02 ± 0.07 c	4.77 ± 0.01 a	4.18 ± 0.03 ab
CST_12y	2.00 ± 0.09 b	1.05 ± 0.08 a	1.15 ± 0.03 b	1.50 ± 0.04 b	1.29 ± 0.06 a	2.35 ± 0.05 b	4.82 ± 0.02 a	4.36 ± 0.05 ab
CST_15y	2.29 ± 0.04 a	1.06 ± 0.07 a	1.33 ± 0.07 a	1.74 ± 0.08 a	1.31 ± 0.07 a	2.80 ± 0.02 a	4.89 ± 0.06 a	4.43 ± 0.05 a
LSD (p < 0.05)	0.20	0.14	0.15	0.15	0.16	0.25	0.22	0.58

Values are expressed as means ± standard deviation. In a column, values marked with the same letter are not significantly different from each other at p < 0.05, according to Duncan’s multiple range test (DMRT).

). Irrespective of the treatments and cropping, the Cfrac1, Cfrac2, Cfrac3, and Cfrac4 accounted for around 15.9, 14.8, 17.2, and 52%, respectively, of TOC. In all the tillage treatments, a major portion of C (49–54%) was stabilized in cfrac4, a recalcitrant pool of SOC. In addition, following short-, medium- and long-term CST practices significantly improved the concentration of different C fractions compared to CT_y treatments. Specifically, after soybean and chickpea harvest, Cfrac1 was significantly highest under CST_15y treatment by 151 and 278%, respectively, compared to CT_y treatment. The corresponding increase in Cfrac2 was 98.5 and 278.3%, in cfrac3: 197.7 and 400%, and in Cfrac4 were 64.1 and 120.4%. Other CST treatments also had a significantly higher content of these C fractions over CT_y but significantly lower than CST_15y treatment.

3.4. SOC Stock, Sequestration and CMI

The SOC stock in the present experimental soil ranged between 11.27 and 13.91 Mg ha⁻¹ (Figure 1). The lowest SOC stock (11.27 Mg ha⁻¹) was observed in the CT_y treatment, while the CST treatments had significantly higher SOC stock, ranging from 13.22 to 13.91. Although the highest SOC stock (13.91 Mg ha⁻¹) was observed under the CST_15y treatment, it was statistically at par with the rest of the CST treatments. Compared to the CT_y treatment, the CST_15y treatment had 23.4% higher SOC stock.

The amount of C sequestered in the experimental soil varied between −0.11 to 0.6 Mg C ha⁻¹ (Figure 1). Long-term CT_y treatment depleted the SOC stock, resulting in negative C sequestration. Similarly, short-term CST treatment (CST_4y) also resulted in negative C sequestration (−0.06 Mg C ha⁻¹). In contrast, medium-term CST practices (6–8 years) were just able to maintain the initial SOC stock. However, significantly higher C sequestration (0.4–0.6 Mg C ha⁻¹) was observed in long-term CST practices (10–15 years).

The carbon management index (CMI) varied notably after soybean and chickpea harvests (Figure 2). After soybean harvest, CMI ranged between 118.6 and 136.07, with significantly lower values observed under the CST_6y treatment, followed by the CST_8y treatment. Conversely, higher CMI values were recorded under other CST treatments and the CT_y treatment, with a mean of around 135. However, the results differed after chickpea harvest, with significantly lower CMI values (52.46) observed under the CT_y treatment. In contrast, all other CST treatments exhibited notable increases in CMI values, with the highest values recorded in the CST_15y treatment, followed by the CST_12y treatment. Regardless of treatments, CMI values tended to be higher after the soybean harvest than after the chickpea harvest.

3.5. Crop Yields

Different tillage treatments significantly affected the soybean and chickpea equivalent yield ranging from 36.23 to 43.41 kg ha⁻¹ and 18.44 to 20.83 kg ha⁻¹, respectively (Figure 3). Soybean equivalent yield was lowest (~36.2 kg ha⁻¹) under CST_8y and CST_12y, while significantly highest yield (43.41 kg ha⁻¹) was observed under CT_y. Further, the soybean equivalent yield in CT_y treatment was statistically comparable to that in CST_4y and CST_15y. Regarding chickpea equivalent yield, the yield variation among treatments was comparatively smaller. The significantly lower yield was recorded in CST_10y, CST_12y and CST_15y, while the significantly highest yield was recorded in CT_y, which was statistically comparable to the yield obtained in CST_4, 6, and 8y treatments.

4. Discussion

Literature suggests that the impact of conservation-based agricultural practices on soil health, C pools and sequestration, and the sustainability of crop yields is determined by the interplay of several factors. These factors include soil type, environmental variables, and the nature of management being implemented. Different tillage practices, with or without residue management, affect soil properties differently.

4.1. Conservation Agriculture and Soil Properties

Our study revealed that the most significant improvement in soil physical properties was observed following short-term CST practices (CST_4 and 6 years), while improvement after that was not much more prominent. Soil BD in the plough layer decreased following the adoption of CST practices compared to CT. This reduction in BD is likely due to the higher SOM content in the soil, which enhances aggregation and consequently increases micro-pore volume [52]. This was further supported by a negative correlation between BD and SOC (Figure 4). Various studies have reported similar findings, indicating that soil BD decreases with increased SOC content [53,54]. However, some studies suggest that transitioning from CT to no-tillage and zero-tillage practices may increase soil BD [55,56,57] while others report no significant effect [58], attributing this to the lack of mechanical fracturing of the soil under no-tillage [59]. Notably, tillage operations disrupt surface compaction and temporarily increase soil porosity [60], leading to lower BD compared to no-till systems [55,57,61,62,63].

A thorough literature review by Strudley [64] revealed inconsistent findings regarding tillage effects on hydraulic properties, with variability observed across soil textures, climates, and specific management practices. Temporal and spatial variability often complicates the identification of long-term treatment effects. Following the adoption of CST practices in the present experimental soil, improvements in aggregation were observed, leading to enhanced continuity of soil pores and pore size distribution, thereby resulting in greater HC of the soil compared to CT practices. Likewise, Pagliai [65] and Abdollahi [66] propose that HC is generally lower with reduced tillage, though it may occasionally surpass levels observed with CT, as also noted in [67,68,69,70]. In contrast, Rücknagel [71] conducted seven field trials in Germany comparing HC values under CT and reduced tillage, finding similar HC values in five instances and significantly lower values in two cases with reduced tillage.

The formation and stability of soil aggregates are often linked to the amount of SOC, with MWD reported to increase with higher SOC levels. In our study, SOC was positively correlated with MWD (Figure 4), consistent with observations by Villamil [72]. Similarly, some studies found higher MWD in soil with CA due to minimal mechanical disturbance, and retention of residue on the surface, which increased SOC content and promoted the fractions of water-stable macro-aggregates [73,74,75]. Conversely, a decrease in aggregation with tillage may be attributed to frequent cycles of drying–rewetting and freezing–thawing, coupled with reduced residue addition, which increases macroaggregate susceptibility to physical disruption [76].

We found that available nutrients, particularly N and P, increased as the duration of CST implementation increased. CST reduces the SOM decomposition rate, which helps maintain higher SOC levels and retains other nutrients bound to SOM. The higher SOM content in CST systems serves as a reservoir for N, P, K, and other nutrients, replenishing plant-available supplies and mitigating losses due to soil erosion and plant uptake. Regular incorporation of crop residues in CST stabilizes the mineralization and immobilization cycles of SOM [77]. Dey [78] also reported a 9% increase in total N levels in CST-based rice–wheat systems compared to CT practices.

Additionally, the increased SOM in CST plots reduces the adsorption, precipitation, and fixation of applied soluble P by forming complexes or coatings on adsorption surfaces and creating soluble phosphate–humate complexes [79]. In contrast, CT redistributes surface-applied residue and soil P across various soil layers [80]. CST practices, however, tend to accumulate P at the soil surface, enhancing its availability due to limited soil mixing, which reduces fixation opportunities for applied P.

Crop residue left on the surface under CST also serves as a substantial source of organic P, which mineralizes over time, aligning P availability with crop requirements. Increased residue in CST plots may contribute to non-exchangeable K in the soil, improving the soil’s buffering capacity and increasing K bioavailability [77]. While short-term CST practices (CST for 4 years) showed a significant increase in K availability compared to CT, subsequent increments in medium and long-term CST practices were statistically similar to those observed in the 4-year CST period. These observations are consistent with the findings of Manono [81].

4.2. Conservation Agriculture and C Fractions and Sequestration

The labile C pools serve as an important food source for soil microbes, playing a vital role in regulating nutrient supply to sustain soil and crop productivity [43]. The relative prevalence of SOC fractions was Cfrac4 > Cfrac3 > Cfrac1 > Cfarc2. The magnitude of C in various oxidizable SOC fractions is influenced by the soil’s supply and availability of C sources; thus, adding organic sources increases the soil’s C content [82].

CST-based treatments provide a conducive environment for the microbial decomposition of SOM, leading to increased C content and a larger active C fraction that is essential for nutrient cycling [83]. The increased biomass input under CST plots, including root exudates, decomposing roots, and above-ground residues, potentially increases active fractions of SOC compared to CT treatments. However, increased microbial activity also enhances the stabilization and polymerization of the active C fraction into humus, resulting in a higher passive C pool in CST treatments, suggesting relatively more C sequestration under CST-based practices [84].

Our observations under a soybean + pigeon pea–chickpea cropping system of semiarid Vertisol revealed that about 69% of C was allocated to the passive pool, consistent with findings by Anantha [85] and Majumder [86], who reported 67% and 63% C stabilization in the passive C pool under different cropping systems. Interestingly, adding easily decomposable crop residues under CST effectively boosts active and passive C pools. Also, correlation analysis showed that most C pools were significantly associated with one another (Figure 4). All fractions of SOC significantly increased the soil’s capacity to sequester C, indicating that they were in dynamic equilibrium. Depletion or enrichment in any of the fractions would alter the equilibrium and affect the magnitude of the other fractions.

The implementation of CST practices helps stabilize SOC, in contrast to CT, where SOC is prone to instability and continuous oxidation due to tillage activities [87,88]. In CT systems, soil aggregates are mechanically disturbed and exposed to soil microorganisms, which can accelerate decomposition and reduce SOC as previously protected C is mineralized [59]. Other studies have similarly found that SOC levels are lower in CT systems compared to zero tillage practices [89,90].

Research suggests that the effectiveness of CST in sequestering C is significantly influenced by the duration of its application. For example, Al-Kaisi [26] noted that the short-term effects of no-tillage on soil C dynamics remain uncertain. Additionally, West and Post [91] observed that while no-tillage practices generally enhance soil C sequestration, noticeable improvements may take approximately 5–10 years to manifest. Similarly, Wiesmeier [92] indicated that after 11 years, no-tillage on average increased C stocks in fine-textured soils by 75.1–92.2% relative to CT, respectively. Likewise, Franzluebbers [93] observed minimal to no discernible rise in SOC content within the initial 2–5 years following the conversion from CT to CST. However, a notable increase often manifested between 5 to 10 years post-conversion. Generally, the short-term effects (≤10 years) of management practices on SOC are intricate [94] and subject to variation based on soil conditions such as texture, climate, biomass return, and the specific management techniques employed [26].

Consistent with these findings, our observations show that the adoption of short-term (4–6 years) CST-based practices resulted in negative (−0.06 Mg ha⁻¹) C sequestration, and medium-term (8 years) CST-based practices just maintained the antecedent level. In contrast, significantly higher amounts of C were sequestered under long-term CST_10, 12, and 15 years, ranging from 0.42 to 0.6 Mg ha⁻¹. Research by Lal [95] stated that a sequestration rate of 0.4–0.8 Pg C y⁻¹ could be achieved with the adoption of no-tillage, indicating that CST-based soils, being undisturbed, allow crop residues to accumulate on the soil surface, thus protecting pre-existing and stable SOM. These differences in C sequestrations are mainly because of the differences in the input of biomass C and soil moisture and temperature regimes [96,97].

These results suggest that long-term CST-based practices effectively increase soil C sequestration, which can help mitigate atmospheric CO₂ enrichment. Further, higher CMI values in CST-based treatments suggest that soil under CST practices may be of higher quality for demonstrating C rehabilitation than soil under other management, i.e., CT [98].

A relationship was plotted between annual C input in soil and the amount of C sequestered (Figure 5). The regression relationship was significant (y = 23x − 0.65; R² = 0.87; p < 0.05), explaining 87% variability in the amount of C sequestered. The relationship further revealed that around 2.82 Mg C ha⁻¹ year⁻¹ must be maintained in the experimental region to achieve zero change in SOC. This critical C input must be assimilated from plant parts or manures to keep SOC content balanced under the soybean + pigeon pea–chickpea rainfed production system in semiarid subtropical Vertisol. This rate of biomass C input is higher than that reported (1.10–1.14 Mg C ha⁻¹ year⁻¹) under sorghum, maize-blackgram, and groundnut-based CS under Vertisols in India [99,100,101] and (0.72 Mg C ha⁻¹ year⁻¹) under cotton + greengram intercropping system in semiarid Vertisol of Central India [53]. This is because the studies mentioned above were based on long-term nutrient management practices, and very minimal C was depleted in the control plot. However, our results align with Islam [102], who indicated that CT required higher C addition (8.60 and 9.13 Mg C ha⁻¹year⁻¹) to maintain antecedent level, whereas CST practices required less C addition (1.68 and 5.21 Mg C ha⁻¹year⁻¹) to maintain SOC level (zero change). Similarly, in rice–lentil rotation, Srinivasarao [101] reported that a minimum quantity of 2.5 Mg C ha⁻¹ year⁻¹ is required to maintain a SOC level for integrated nutrient management practice in the IGP.

4.3. Conservation Agriculture and Crop Yields

CA practices, such as no-till farming with residue retention and crop rotation, can impact crop yields depending on climate, soil type, management practices, and implementation duration. Research has demonstrated that CA can affect crop yields differently, resulting in improvements, reductions, or no significant change at all.

In our study, crop yields were generally lower under CST compared to CT. This observation is consistent with Salem [103], who found that zero tillage led to reductions in cob length, number of rows per cob, and grain yield by 18.8%, 15.8%, and 15.4%, respectively, compared to CT. Similarly, Afzalinia and Zabihi [55] reported that zero tillage decreased maize grain yield and yield components by 18.2% and 11.1%, respectively, in a short-term study compared to CT. Furthermore, a global meta-analysis conducted by Pittelkow [104], which analyzed 5463 paired yield observations from 610 studies across 48 crops and 63 countries, indicated that no-till practices have an overall negative impact on crop yields by 5.7%. Nevertheless, under specific conditions, no-till systems can yield equivalent or even superior yields compared to CT systems. Despite potential variations in initial yield impacts, the findings do not suggest that no-till consistently outperforms CT in the long run. The yield reduction with a no-tillage system is widely documented in wheat, cotton, sorghum, maize, peas, barley, rape, sunflower, and rye, in Australia [105], Burkina Faso [106], China [107], Japan [108], the Mediterranean [109], Mexico [73], the United States [110], Northwestern United States [111], Pacific Northwest United States [112], Western Great Plains [113], and Uzbekistan [114]. These studies highlight the causes of yield reductions, such as reduced early seedling growth, N and P deficiency, stubble removal, poor crop establishment, greater root disease pressure, partial crop residue retention, and not using complete systems. Nonetheless, when implemented correctly, no-tillage practices can sometimes result in better yields than CT, particularly in well-drained soils prone to water runoff and accelerated erosion [115,116,117].

Our experiment observed that soil temperatures within CST plots were 2–3 °C lower than those in CT plots after 45 days of seeding (data not presented). These reduced soil temperatures in the early stages of the growing season, attributed to the presence of crop residue, might potentially delay germination and hinder initial growth. This suggests that lower emergence rates under CST primarily affected the crop yields. Additionally, excessive residue shallowly mixed into the soil where seeds are placed results in a less consolidated seedbed, reducing crop emergence and yield [118]. Furthermore, long dry spells in semiarid tropics are also responsible for limiting yields, although the soil quality is improved.

4.4. SOC and Crop Yield Nexus under Conservation Agriculture

In our study, CST-based practices significantly increased soil C sequestration. However, the same trend was not observed for crop yields, as CST practices exhibited yields comparable/lower to those of CT (Figure 6). This discrepancy is also noted in several global studies on CA practices. For example, Sun [119] conducted a meta-analysis using data from 260 paired plots across 115 published studies. In regions with a harvest index below 40, combining crop residue retention and crop rotation in CA resulted in increased SOC sequestration and higher crop yields. However, no-till practices alone did not lead to yield improvements. The analysis highlighted that many regions, including parts of India, north-central Africa, and Australia, fall into this category. Arid regions, in particular, benefit the most from CA, achieving both enhanced SOC sequestration and improved crop yields compared to local CT. In contrast, more humid regions are likely to see SOC increases without significant yield gains, and certain colder regions might experience yield declines and potential soil C losses despite SOC gains. This emphasizes the need to consider regional climate and site-specific conditions when evaluating the benefits of CA.

5. Conclusions

The present study assessed the complexity of the relationship between conservation-based agricultural practices and soil health, carbon sequestration, and crop yields. We found that the most significant improvement in soil physical properties and available potassium was observed following short-term conservation tillage (CST) practices (CST_4 and 6 years). In contrast, available nutrients, particularly nitrogen and phosphorus, increased as the duration of CST implementation increased. Similarly, long-term CST practices resulted in positive carbon sequestration, while CT and short-term CST treatments showed negative or minimal C sequestration. Further, the study found that to maintain the antecedent SOC levels, around 2.82 Mg C ha⁻¹ year⁻¹ is needed in the present experimental soil, which was possible via adopting CST-based practices. This underscores the importance of sustained conservation practices for enhancing soil quality and long-term carbon sequestration benefits in swell–shrink soil of semiarid Central India. This was further confirmed by significantly higher CMI values under CST treatments compared to CT, indicating improved carbon management and soil quality. However, while long-term CST shows promise, its impact on crop yields varies. CT, for instance, yielded higher soybean equivalent yields and comparable chickpea equivalent yields to CST treatments, mainly due to delayed germinations induced by lower soil temperatures in CST plots. Therefore, our study underscores the significance of integrating site-specific characteristics, management practices, and regional climate conditions into CA frameworks when evaluating the potential benefits of adopting such practices. This comprehensive approach is crucial, as it acknowledges that certain regions may encounter yield losses or see no significant changes in crop yields despite implementing conservation measures.

6. Limitations of the Study

The primary limitation of this study is the execution in farmers’ fields, which introduces variability in management practices conducted by farmers and complicates the standardization of baseline data. We relied on reference samples from adjacent non-cropped areas and historical fertility reports, which may not fully represent pre-cultivation soil conditions. Additionally, the study’s findings are specific to a rainfed semi-arid environment, potentially limiting their generalizability to other regions with different climatic and soil conditions.