1. Introduction
As per estimates, the world population will be 9.7 billion by 2050 [
1]. The requirement of food is also supposed to rise by 56% by that time [
2]. One of the major goals of agriculture is to feed the burgeoning population and to attain food as well as nutritional security. The scope of the horizontal expansion of agriculture is limited due to land shortage and, therefore, an additional food demand must be met by vertical expansion, i.e., by increasing the cropping intensity and productivity of existing agricultural land. Rice-fallow is an important cropping system in South Asia covering an area of 22.3 M ha, and ~18.0 M ha of that area comes from India [
3]. Under that production system, only rice is taken as a rainfed crop in the rainy season depending on monsoonal rains, and the land remains fallow for the rest of the period. Hence, those areas have huge potential for increasing crop production and can play a pivotal role in regional food security. The uncertainty of monsoon rain is a serious issue in attaining a secured yield. The adoption of different resource conservation technologies (RCTs) could be a potential alternative in attaining the sustainable productivity and higher cropping intensity in the region [
4].
Rice (
Oryza sativa L.), largely cultivated by transplanting in rainfed conditions during the southeast monsoon season, is a staple food crop of the eastern region of India. Puddling, i.e., tillage in standing water, is extensively practiced for rice transplanting, which minimizes percolation losses, suppresses weed growth and improves plant nutrient availability [
5]. However, puddling is known to degrade the soil health by the formation of hardpan in the sub-soil, destroying soil aggregation and decreasing the population of beneficial microorganisms, thus reducing crop yields of subsequent post-rainy crops in rice-fallows [
6]. Puddled soil becomes very hard upon drying and restricts the root growth of succeeding crops, thereby limiting the utilization of the residual soil moistures [
7]. Ploughing, coupled with the removal of crop residue, rapidly depletes SOC pools in comparison to the natural ecosystem [
8].
India’s Eastern Hill and Plateau eco-region is a less developed region [
6] and, therefore, has a great scope for improvement in food production by the adoption of scientific agricultural practices. Over the generations, rice is generally grown in this region during the wet season (mid-June to October) in the lowlands. However, the rice production has now been shifted to terraced hill-slopes (medium land), where rice is grown as a monocrop rice-fallow system caused by increasing population pressure [
6]. Paddy yields in these regions are poor (<2.0 t ha
−1) and, due to the heavy dependence on rainfall, the crop is frequently prone to vagaries of the drought despite high annual precipitation (>1200 mm) [
9]. Declining soil health, irregularities in monsoonal rain, lack of irrigation facility, low residual soil moistures, prevalence of long-duration rice variety and free grazing/poor socio-economic conditions of farmers are major issues leading to the unsustainability of the rice-fallow system in the region [
10,
11,
12,
13]. The continuous cultivation of rainfed rice during the wet season followed by fallow during the winter and summer periods in this region without the use of conservation-effective measures resulted in accelerated soil erosion and depletion of the SOC [
14]. Therefore, immediate attention is required for the identification of suitable management and to devise a practical solution for improving soil quality, water productivity and sustainability of the rice-fallow system.
Conventional agriculture, which favors repeated tillage operation, intensive fertilizer application, water use and crop residue removal, has increased crop productivity by more than three times in the past century [
15]. However, use of the synthetic fertilizers and pesticides in conventional tillage has been responsible for degrading soil quality, eutrophicating water resources and causing biodiversity losses [
16]. On the contrary, CA depends on the principles of minimum soil disturbances, permanent soil covers of crop residue or covering crops and crop rotations [
17], which is often promoted as an environment-friendly alternative to the conventional tillage system [
18]. No-tillage (NT), a central component of CA, is one of the most cost-effective means to safeguard and recuperate the soil resources [
19]. Soil fertility in NT is reported to improve due to the reutilization of residue, increased microbial activity and a reduction in erosion due to the simultaneous decline in the requirement of synthetic fertilizers and, thus, circuitously protect the environment from radiative forces through the sequestration of carbon and nitrogen [
20]). NT also requires fewer inputs and had slightly lower or even comparable yields than the conventional tillage system [
17]. Time-based yield stability in NT is similar to that obtained under conventional tillage (CT) and, thus, a transition from CT to NT is possible without having yield penalties [
18].
Long-duration rice (150–160 day) varieties often delay the planting of winter crops, which aggravates the soil moisture and temperature stress, leading to reduced yields. To enhance crop yields and the yield stability of succeeding winter crops after rice, CA is presently being emphasized in rice-fallow systems [
21]. Energy and carbon budgeting of different CA-based production systems are extremely reliant on the type of tillage operation, crop residue and choice of crops. This calls for precise and accurate assessments of the energy and carbon budget of diversified CA-based cropping systems in the Eastern Hill and Plateau region. This will facilitate the designing of better energy- and carbon-efficient cropping systems in this region.
In the scenario of global climate change, low carbon-emitting and low energy-input farming systems are given paramount importance. Systematic studies on comparative evaluation of the different cropping systems in terms of carbon and energy intensiveness in diverse RCTs are not available for the Eastern Plateau and Hill region of India. The quantifications of the carbon pools, water and energy productivity and system productivities would be useful in the selection of better cropping systems and their management practices in the Eastern Plateau and Hill Region of India. In this context, the present study was conducted in a participatory mode for evaluating the feasibility of second and/or third crops in an extreme rice-fallow region and the impacts of RCTs on soil property. It has been reported that including short duration cultivars of pulses/oilseeds along with anchored residues could be a better alternative to reduce adverse effects of production systems on soil and the environment [
22]. Keeping this in view, the present investigation was aimed at the inclusion of a medium-duration rice variety followed by short duration oilseeds/pulses in the existing rice-fallow systems to evaluate their response to carbon footprints, water, energy and system productivity under different conventional and RCT-based management practices. The current investigation hypothesized that the intensification of the rice-fallow system using RCTs will lead to (1) an improvement in overall system productivity by including second and/or third crops; (2) an improvement in soil health vis-à-vis carbon dynamics; and (3) enhancement in water, energy and carbon efficiency.
2. Materials and Methods
2.1. Experimental Sites
A field trial was conducted at a farmer’s field in a participatory mode in Chene Village, Jharkhand, India (23°17′05″ N latitude and 85°25′59″ E longitude, altitude 648 m amsl) from 2015 to 2020 (
Figure 1). The experimental soil was alfisol with an acidic nature (pH = 5.13) and sandy loam in texture (67.5% sand, 20.3% silt and 12.2% clay). The soil organic carbon (SOC) content before experimentation was 4.1 g kg
−1 in 0–15 cm soil layer. The soil was low in available nitrogen (105.3 kg ha
−1) [
23], low in available phosphorus (4.82 kg ha
−1) [
24] and medium status in available potassium (190.4 kg ha
−1) [
25]. The experimental site is characterized by a hot and sub-humid climate with an average annual precipitation of 1350 mm, out of which 80–90% is received during the rainy cropping season (June to September).
2.2. Experiment Details
The study region has three main cropping seasons coinciding with rainy (July–October), winter (November–February) and summer (March–June) seasons. To match with these cropping seasons, the present experiment had five treatments as (1) fallow land (FL) in all three seasons (2), conventional practice with a rice-fallow–fallow cropping system (RF), (3) conventional practice with a rice–mustard–blackgram system (CP), (4) transplanted rice (TPR)–mustard–blackgram system under conservation agriculture practice (CA1) and (5) direct sown rice (DSR)–mustard–blackgram system under a conservation agriculture practice (CA2). In both the conservation agriculture (CA) treatments, all three principles of conservation agriculture, i.e., soil coverage with crop residue, crop rotation and minimum disturbance to soil with the adoption of zero tillage (ZT) practice were used in all the three crops.
As per conventional practices followed in the region, mustard and blackgram crops in CP were sown using broadcasting methods, while in both CA methods (CA1 and CA2), these crops were sown with ZT with residue retentions of the previous crop. Experiments were laid out in a randomized block design (RBD) with five replications (each plot size was 200 m
2). Treatment-specific details of various field operations are indicated in
Supplementary Table S1. Soil samples were collected from the fallow plot (FL) where no cultivation was performed. Observations from this fallow plot were taken as reference to facilitate the computation of indices related to carbon stock and carbon dynamics.
The medium-duration rice variety “Naveen” (130 days) was grown in all the treatments during wet season. About 21-day old seedlings were transplanted (2 seedlings per hill) in first fortnight of June with a plant spacing of 20 cm × 15 cm in RF, CP and CA1 treatments, while in CA2, having ZTDSR, rice was sown directly by ZT-fertilizer-cum-seed drill using seeding rates of 25 kg ha
−1 with a row-spacing of 20 cm. Date of paddy planting in CA2 was same as that of nursery sowing date for RF, CP and CA1. In CA1, having ZTTR treatment, plots were ponded with a water depth of 5 cm and then rice seedlings were transplanted with dibbler in the soft top layers of soil without puddling. Cropping calendar for the experimental years of 2015–2020 is presented in
Figure 2. In general, DSR was sown 21 days before transplanted rice depending on rainfall situation of particular year.
Recommended doses of N–P
2O
5–K
2O at 80–40–40 kg ha
−1 were applied to rice crop [
26]. Half dose of N and full doses of P
2O
5 and K
2O were applied as basal and remaining half dose of fertilizer N was top dressed in two equal splits at tillering and at panicle initiation stages. After harvesting of paddy, pre-emergence application of glyphosate (41% EC) was performed @ 2 l ha
−1 in CA1 and CA2 plots for managing weeds. Sowing of winter mustard (Pusa Mustard 30) having duration of 100–105 days was delayed to first fortnight of December mainly due to prevalence of prolonged high soil moisture after rice harvest. Recommended doses of N–P
2O
5–K
2O for mustard were 40–20–20 kg ha
−1, wherein half dose of N and full doses of P
2O
5 and K
2O applied as basal and remaining half dose of N was top dressed at 45 DAS. Summer crops of blackgram (TAU–1) having 60–65 day duration were sown in first week of April and recommended dose of fertilizer (20:40:20 kg ha
−1) was applied as basal dose. Applications of N, P
2O
5 and K
2O were performed through urea, diammonium phosphate (DAP) and muriate of potash (MOP), respectively. Before the start of each cropping season, farmyard manure (FYM) was applied at 5 Mg ha
−1 to each crop of rice, mustard and blackgram as basal. Accordingly, over the period of five years, 25 Mg ha
−1 of FYM was added in RF, while 75 Mg ha
−1 of FYM each was added to CP, CA1 and CA2 treatments.
2.3. Soil Sampling and Analysis
Composite soil samples (collected from seven randomly selected points in each plot) at depths of 0–0.15, 0.15–0.30, 0.30–0.45 and 0.45–0.60 m were collected during 2020 after harvest of summer crops. Soil samples were air-dried and passed through 2 mm sieve for analysis of soil organic carbon (SOC) and its fractions. Sub-samples of the collected soil were stored at 4 °C and were used later afresh (after no more than 24 h) for estimation of biological properties (soil microbial biomass carbon and dehydrogenase activity) of soil. A core sampler (5.0 cm diameter and 8.0 cm length) was used for the measurement of bulk density of soil of each layer of profiles from 0–0.15, 0.15–0.30, 0.30–0.45 and 0.45–0.60 m soil depth [
27].
2.4. Total Soil Organic Carbon
Total carbon (TC) was analyzed by carbon-hydrogen-nitrogen-sulfur (CHNS) analyzer (Elementar Vario EL III, Hanau, Germany) following dry combustion method [
28]. Total soil organic carbon (TSOC) [TC–inorganic carbon] stock at different soil depths was calculated by multiplying the respective SOC value with bulk density and depth of soil as given by [
29],
where TSOC
stock is TSOC stock (Mg ha
−1), SOC is soil organic carbon (g kg
−1),
ρ is soil bulk density (Mg m
−3) and
d is depth of soil layer in m.
The carbon retention efficiency (
CRE) under different treatments was calculated using the following formula [
30],
where
CRE is carbon retention efficiency (%) and
TSOCtreatment and
TSOCfallow represent total SOC stock (Mg ha
−1) from treatment plots (RF, CP, CA1, CA2) and fallow plot (FL), respectively. TCI is total carbon input expressed as Mg C ha
−1, while 0.60 stands for soil depth expressed in m.
The carbon sequestration rate (CSR) in different treatments was calculated using the following formula given by [
31],
where CSR is carbon sequestration rate (Mg ha
−1 year
−1) and
N is experimental year.
The residual carbon left in soil (
CR) and C build-up in soil (
CBS) in different treatments was estimated as follows [
32]:
where
CR is residual carbon left in soil (Mg ha
−1),
CBS is carbon build-up in soil (%) and
Ctreatment and
Cfallow (Mg ha
−1) represent carbon stock in the treatment plots (FP, CP, CA1, CA2) and fallow plot (FL), respectively.
2.5. Oxidizable Organic Carbon Fractions
Different fractions of TSOC were computed following the procedure of Walkley and Black method [
33] using 5, 10 and 20 mL of concentrated H
2SO
4 resulting in 3-acid aqueous solution ratios of 0.5:1, 1:1 and 2:1 that corresponded to 12, 18 and 24 N H
2SO
4. The C-oxidized by 24 N H
2SO
4 is equivalent to oxidizable C-obtained by standard [
34] method. The different fractions of TSOC were calculated as follows:
Very labile (VLC): organic carbon oxidizable by 12 N H2SO4;
Labile (LC): difference in organic carbon extracted between 18 and 12 N H2SO4;
Less labile (LLC): difference in organic carbon extracted between 24 and 18 N H2SO4;
Non-labile (NLC): difference between TSOC and organic carbon extracted with 24 N H2SO4.
2.6. Active and Passive Pools
Active pool (AP) of TSOC was estimated by adding VLC and LC, while passive pool (PP) represented the LLC and NLC.
2.7. Biological Properties of the Soil
The soil microbial biomass carbon (
SMBC) was determined by chloroform fumigation method [
35]. Soil microbial biomass carbon was calculated as follows,
where
SMBC is the soil microbial biomass carbon (µg g
−1) and
Fc is the organic carbon extracted from 0.5 M K
2SO
4 from fumigated soil (µg g
−1) minus organic carbon extracted from non-fumigated soil (µg g
−1). Dehydrogenase activity (DHA) in soil was estimated by the methodology outlined by [
36].
2.8. Crop Yield and System Rice Equivalent Yield (SREY)
Rice, mustard and blackgram were manually harvested, and biological yields from an area of 50 m
2 were recorded. Grain yield was recorded after threshing and kept at ~12% (
w/
w) moisture content following sun drying. In CP treatment, the crops were harvested close to the ground level. In CA1 and CA2 treatments, rice, mustard and blackgram were manually harvested leaving bottom one-third of plants as crop residue in the field. Yields of all non-rice crop were converted into rice equivalent yield (REY) for computing system productivity [
29].
where
REY is rice equivalent yield of non-rice crop (Mg ha
−1),
Yc is yield of non-rice crop (Mg ha
−1),
MSPc is minimum support price of non-rice crop in Indian rupees fixed by the Government of India (₹ Mg
−1) and
MSPr is the minimum support price of rice crop (₹ Mg
−1).
Where, SREY is system rice equivalent yield (Mg ha−1), Yr is yield of rice (Mg ha−1), REYw is rice equivalent yield of winter crop (Mg ha−1) and REYs is rice equivalent yield of summer crop (Mg ha−1).
2.9. Energy Budgeting
Energy budgeting of different treatments considered in this study comprised of input energy assessed from the farm input used in various operations and output energy generated in grain and straw yield. The treatment-wise input was supplied and field operations was recorded for the fifth year of experimentation (
Supplementary Table S2) when yield gain from CA systems started. In computing energy budgets, the amounts of inputs and field operations were multiplied with corresponding energy equivalents (
Supplementary Table S3). The output energy (OE) was computed by multiplying grain and stover yield with their corresponding energy equivalent. Various energy input indicators were computed in rice-fallow system as suggested by various researchers [
37,
38,
39]:
where,
NE is net energy (MJ ha
−1), output and input energy are expressed in MJ ha
−1,
EUE is energy use efficiency (%),
EP is energy productivity (kg MJ
−1),
Ye is economic yield of crop (kg ha
−1) and
SE is specific energy (MJ kg
−1); system productivity is expressed in kg ha
−1, EPr is energy profitability.
2.10. Carbon Budgeting
Effects of different production systems on the environment were evaluated in terms of carbon budgets. Carbon inflows and outflows of each production system were estimated by considering the type and quantity of different inputs consumed and outputs generated from the system in fifth year. Carbon flows were computed in units of C-footprint expressed in CO
2 equivalent per unit area and time (CO
2-e ha
−1) [
40]. The quantity of input or field operation was multiplied with corresponding emissions factors (
Supplementary Table S4) to get CO
2 emission in terms of kg CO
2-e ha
−1. The C-output (CO) was calculated by multiplying grain and straw/stover yield with average C-content of biomass (40% on dry weight basis) [
40]. The N
2O emission from applied fertilizer, manures/residues were estimated as suggested by [
41],
where N applied through fertilizers, manures, and crop residue; EF is emission factors for N
2O emission from N inputs and taken as 0.01 for Indian subcontinent, kg N
2O-N/kg N input [
41]; 44/28 is a coefficient converting N
2O-N to N
2O; 285 is global worming potential (GWP).
The seasonal emission of CH
4 from rice cultivation in CP, CA1 (ZTTPR) and CA2 (ZTDSR) were, 12.8, 12.8 and 5.6 kg CH
4 ha
−1 season
−1 [
42]. The values of CH
4 emission were converted to CO
2 equivalent by multiplying with GWP factor of 28. In our experiment, the mustard and blackgram were not in an anaerobic condition and only CO
2 and N
2O emissions were considered [
43].
We used six carbon-based indicators (C-footprint in spatial scales, C-footprint in yield scales, C- input, C-output, CSI, and C-efficiency) for comparative analysis of production systems in terms of environmental effects. The C-footprint (CF) represents overall C-intensiveness of different production systems and has been frequently studied to quantify environmental performance of production systems. Different components of carbon budgeting were calculated as follows [
4]:
where
CFs is C-footprints of a production system in spatial scale (CO
2-e kg ha
−1),
CEinputs is the CO
2 equivalent of carbon emissions from all the inputs used (CO
2-e kg ha
−1),
CEN2O is the C-emissions equivalent to N
2O emissions from the system (CO
2-e kg ha
−1),
CECH4 is the carbon emissions equivalent to CH
4 emissions from the system (CO
2-e kg ha
−1), CFy is the C-footprints in yield scales (CO
2-e kg Mg
−1),
CI is the C input to the system (kg C ha
−1),
CO is the C-output from the system (kg C ha
−1),
Bt is the total biomass produced by the system (kg ha
−1) and
CSI is the C-sustainability index and CE is the C-efficiency.
2.11. Irrigation Management
In the present study, rice was cultivated as purely rainfed crop. The furrow and flooding are most common methods of water application to mustard and blackgram crops cultivated in Eastern Indo-Gangetic plains (EIP) region and these conventional methods of irrigation were adopted in preceding winter and spring crops. In the case of mustard, two irrigations of 50 mm each were applied at critical stages viz. vegetative and pod formation, coinciding with 35–40 and 65–70 days after sowing (DAS) using furrow method of water application. At the time of sowing, the residual soil moisture left over after the harvest of rice was sufficient for the germination of mustard and only two irrigations were deemed necessary to meet the crop water requirement. In blackgram, three irrigations of 50 mm each were applied through flooding method at 5–10, 18–28 and 37–42 DAS. In order to apply 50 mm of irrigation, the duration of each irrigation event was decided based on pump discharge and amount of water to be applied.
2.12. Crop Water Use and Water Productivity
The actual crop water use was estimated using CROPWAT 8.0 model ([
44]. CROPWAT estimates the reference evapotranspiration (ET
0) using Penman–Monteith method and converts it to crop’s actual water use using mathematical functions based on climate, crop and soil parameters. The meteorological data viz. maximum and minimum temperature (°C) and daily rainfall were collected from the automatic weather station located nearby experiment sites, while relative humidity and sunshine hours were estimated using inbuilt functions available in CROPWAT. The soil parameters, i.e., field capacity and wilting point of soil in the experimental plots were determined using laboratory tests [
45]. The crop data pertaining to sowing date, crop duration and rooting depth as recorded during the period of experimentation were used in the CROPWAT simulations. Crop coefficient values (Kc) for initial to later stages of crop growth used in the modeling were taken from [
44].
Since rice was grown as rainfed crop, its water use was estimated using ‘no irrigation (rainfed)’ options available in CROPWAT. In rice, the water consumption was same in the case of rice-fallow (RF), CP and CA1, while the water consumption of rice in CA2 was estimated separately without considering nursery water requirements. The time and depth of water application for each irrigation in mustard and blackgram crops in CP were specified using the ‘Irrigation at user defined interval’ option available in CROPWAT. The crop water use under the condition of organic mulch was estimated using the adjusted crop coefficient (Kc) values. In case of organic mulches covering the soil surface, as was the case for CA1 and CA2 plots, Kc values for initial, mid and late seasons were reduced by 25%, 8% and 8%, respectively, as per the approach suggested by [
44]. The water productivity (WP) was worked out by dividing the total yield (Y) by total water use (CWU) by the crop during growing season. The system water productivity (SWP) was computed by dividing the REY of the whole cropping systems with the annual crop water use.
2.13. Statistical Analysis
The data pertaining to different indicators and attributes of production systems were analyzed using analysis of variance (ANOVA) for randomized block design (RBD). The significance of the treatment effect was determined using F-test. Principal component analysis (PCA) was performed using XLSTAT version 2021.1 (Addinsoft, Paris, France)
5. Conclusions
The present investigation was carried out to evaluate the response of different conventional and conservation agricultural practices in rice-fallow ecosystems of South Asia. The research outcomes clearly demonstrated that CA-based crop management practices can address the issues of deteriorating soil health, declining system productivity, water availability and greenhouse gas emissions. The present research concludes that conservational agricultural practices caused a significant improvement in TSOC both in labile and non-labile carbon fractions that led to an increase in carbon retention efficiency. The practice of conservational agriculture enhanced the very labile C in all of the soil depths and stabilized the SOC as a non-labile recalcitrant fraction in 0–0.30 m soil depth. This practice also showed a lower carbon footprint, higher carbon sustainability index and carbon efficiency.
The CA treatments were more water efficient and increased the crop water productivity over conventional practices. This has larger implications on the water-stressed Eastern Hill and Plateau region where water availability is a major concern. The modified CA-based cropping system increased the system productivity by 34–46% over the conventional rice-fallow system, highlighting the importance of including second and third crops in the rice-fallow eco-systems of South Asia. Such a modified CA-based system showed low energy consumption, higher energy use efficiency and improved energy productivity and energy profitability.
A cropping system having direct seeded rice (DSR)–mustard–blackgram under conservation agriculture was adjudged as the best practice in terms of C-sequestration, C-stabilization, biological activity with higher system productivity, energy use efficiency and reduced carbon footprints. Thus, the findings of our investigation could be helpful to researchers and policy makers for designing cleaner and eco-friendly production systems for rice-fallow in the Eastern Hill and Plateau region of eastern India and similar agro-ecotypes of South Asia.