Field Crops Research 291 (2023) 108791 Contents lists available at ScienceDirect Field Crops Research journal homepage: www.elsevier.com/locate/fcr Reduced tillage and crop diversification can improve productivity and profitability of rice-based rotations of the Eastern Gangetic Plains Muhammad Arshadul Hoque a, Mahesh K. Gathala c,*, Jagadish Timsina c,d, Md.A.T.M. Ziauddin b, Mosharraf Hossain b, Timothy J. Krupnik c a Farm Machinery and Post-Harvest Process Engineering Division, Bangladesh Agricultural Research Institute (BARI), Gazipur 1701, Bangladesh b Department of Farm Power and Machinery, Bangladesh Agricultural University, Mymensingh 2202, Bangladesh c International Maize and Wheat Improvement Center (CIMMYT), Sustainable Agrifood Systems, House 10/B, Road 53, Gulshan-2, Dhaka 1213, Bangladesh d Global Evergreening Alliance, Burwood East, Melbourne, VIC 3151, Australia A R T I C L E I N F O A B S T R A C T Keywords: Intensive rice (Oryza sativa)-based cropping systems in south Asia provide much of the calorie and protein re- Alternate tillage quirements of low to middle-income rural and urban populations. Intensive tillage practices demand more re- Calorie and protein yields sources, damage soil quality, and reduce crop yields and profit margins. Crop diversification along with Conservation agriculture conservation agriculture (CA)-based management practices may reduce external input use, improve resource-use Gross margin Labor use efficiency, and increase the productivity and profitability of intensive cropping systems. A field study was Relative yield change conducted on loamy soil in a sub-tropical climate in northern Bangladesh to evaluate the effects of three tillage Cropping systems diversification options and six rice-based cropping sequences on grain, calorie, and protein yields and gross margins (GM) for different crops and cropping sequences. The three tillage options were: (1) conservation agriculture (CA) with all crops in sequences untilled, (2) alternating tillage (AT) with the monsoon season rice crop tilled but winter season crops untilled, and (3) conventional tillage (CT) with all crops in sequences tilled. The six cropping se- quences were: rice-rice (R-R), rice-mung bean (Vigna radiata) (R-MB), rice-wheat (Triticum aestivum) (R-W), rice- maize (Zea mays) (R-M), rice-wheat-mung bean (R-W-MB), and rice-maize-mung bean (R-M-MB). Over three years of experimentation, the average monsoon rice yield was 8% lower for CA than CT, but the average winter crops yield was 13% higher for CA than CT. Systems rice equivalent yield (SREY) and systems calorie and protein yields were about 5%, 3% and 6%, respectively, higher under CA than CT; additionally, AT added approximately 1% more to these benefits. The systems productivity gain under CA and AT resulted in higher GM by 16% while reducing the labor and total production cost under CA than CT. The R-M rotation had higher SREY, calorie, protein yields, and GM by 24%, 26%, 66%, and 148%, respectively, than the predominantly practiced R-R rotation. The R-W-MB rotation had the highest SREY (30%) and second highest (118%) GM. Considering the combined effect of tillage and cropping system, CA with R-M rotation showed superior performance in terms of SREY, protein yield, and GM. The distribution of labor use and GM across rotations was grouped into four categories: R-W in low-low (low labor use and low GM), R-M in low-high (low labor use and high GM), R-W-MB and R-M-MB in high-high (high labor use and high GM) and R-R and R-MB in high-low (high labor use and low GM). In conclusion, CA performed better than CT in different winter crops and cropping systems but not in monsoon rice. Our results demonstrate the multiple benefits of partial and full CA-based tillage practices employed with appropriate crop diversification to achieve sustainable food security with greater calorie and protein intake while maximizing farm profitability of intensive rice-based rotational systems. Abbreviations: ANOVA, analysis of variance; AT, alternate tillage; BARC, Bangladesh Agricultural Research Centre; BARI, Bangladesh Agricultural Research Institute; CSISA, Cereal Systems Initiative for South Asia; CA, conservation agriculture; CT, conventional tillage; DFAT, Department of Foreign Affairs and Trade; EGP, Eastern Gangetic Plains; GM, gross margin; GR, gross return; REY, rice equivalent yield; R-M, rice-maize; R-M-MB, rice-maize-mung bean; R-R, rice-rice; R-W, rice-wheat; R-W-MB, rice-wheat-mung bean; SOC, soil organic carbon; SREY, system rice equivalent yield; SRFSI, Sustainable and Resilient Farming Systems Intensification; ST, strip tillage; TAFSSA, Transforming Agrifood Systems in South Asia; TVC, total variable cost; ZT, zero tillage. * Corresponding author. E-mail address: m.gathala@cgiar.org (M.K. Gathala). https://doi.org/10.1016/j.fcr.2022.108791 Received 8 June 2021; Received in revised form 18 October 2022; Accepted 9 December 2022 Available online 16 December 2022 0378-4290/© 2022 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). M.A. Hoque et al. F i e l d C r o p s R e s e a r c h 291 (2023) 108791 1. Introduction manual seeding or weeding in crops are generally performed by rural women, the degree and intensity of which depends on the region and Double cropping of monsoon rice with winter crops (maize, wheat, community; rice production is particularly arduous, time-consuming boro rice, pulses, oilseeds and vegetables) or rice-fallow in sequence is and resource-demanding (Chhetri et al., 2020; Gathala et al., 2021). commonly practised in south Asia (Timsina et al., 2010; Krupnik et al., There is a potential to eliminate these unwanted tasks and follow 2015). In many areas, especially in the alluvial Eastern Gangetic Plains CA-based mechanized crop production practices such as non-puddled (EGP) in Bangladesh, eastern India, and eastern Nepal, the third crop of rice transplanting and ZT/ST seeding. The intensive use of labor and mung bean or jute, or other short-duration crops may also be grown energy under conventional agronomic practices results in high produc- under the triple cropping systems (Timsina et al., 2013;Islam et al., tion costs; combined with low crop yields because of inefficient agro- 2019; Gathala et al., 2020a, 2020b). Among the various rotations, nomic management, this results in low gross margins (Gathala et al., rice-rice (R-R), rice-wheat (R-W), and rice-maize (R-M) together cover 2021). Low gross margins and laborious work experienced in agriculture approximately 75% of the net cultivable area in the EGP (Gathala et al., have forced rural male youth to work away from home to meet family 2020b). These crops contribute greatly to household food and nutrition household needs. Such rural youth outmigration further aggravates security, and livelihoods for low to middle-income rural and urban labor crises at crop peak seasons and the cost of hiring additional labor populations (Chauhan et al., 2011; Timsina et al., 2018; Islam et al., to support those families left behind on the farm and enable their agri- 2019). Due to the immense importance of these cereals and the un- cultural production to continue (FAO, 2018; Darbas et al., 2020; Gathala availability of new land for expanding cultivation in the et al., 2021). population-dense EGP, it is increasingly crucial to utilize existing dry A few studies (Krupnik et al., 2014; Islam et al., 2019) have evalu- season fallow land for multiple cropping. There is plenty of scope to ated the impacts of different tillage methods under double cropping intensify the cropping systems and practise triple cropping systems, systems, comparing the full suite of CA practices to AT, in which rice was especially by including short-duration pulses (mung bean), oilseeds puddled and winter crops were subjected to ST/ZT. However, there are (mustard), and fiber crop (jute) in the EGP (Krupnik et al., 2015; Islam also no robust studies comparing CA and AT with CT across various et al., 2019; Rashid et al., 2019; Gathala et al., 2020b). Despite their rice-based cropping systems; in particular, little work has been done to importance for human nutrition, most studies on cropping systems and assess protein and calorie yields in cropping systems under different tillage techniques in south Asia focus on yield improvements alone tillage practices. Further, the labor use and gross margins of the multiple without considering protein and calorie yields for food-insecure small- rice-based rotational systems in the EGP have not been examined. An holder farming households and rural communities (DeFries et al., 2015). improved understanding of the effects of tillage practices on crop pro- In the EGP, mainly rice-based cropping systems are practised with ductivity and profitability can help explain their performance and intensive tillage (puddling for rice and repetitive tillage in the ensuing identify the region’s most productive and profitable systems. Thus, the winter crops) and with the burning or complete removal of crop residues objective of this study was to evaluate CA and AT (i.e., CT in monsoon from the field. Concerns are increasing that intensive tillage practices rice and ST with residue retention in winter and spring crops) against combined with injudicious use of natural resources and external inputs CT, under six double and triple cropping systems in terms of grain yield, can negatively influence soil quality, potentially causing soil acidifica- grain protein and calorie yields, labor use, and gross margin of different tion and loss of soil organic carbon (SOC) (Jat et al., 2018, 2020). Re- crops and cropping systems. Such a study also provides the potential petitive tillage requires intensive water and energy use (Gathala et al., benefit of layering suitable crop rotations and tillage practices on loamy 2013, 2016, 2020a; Islam et al., 2019) and emits large amounts of soil in a sub-tropical environment in northern Bangladesh in the EGP, greenhouse gases (Ladha et al., 2015; Kumar et al., 2018; Gathala et al., representing a major rice-growing region. 2020b). More resource-efficient practices, including zero, reduced/strip tillage, full or partial retention of crop residue, and diversified crop 2. Materials and methods rotations, are widely advocated as options for decreasing the use of non-renewable resources, reversing soil degradation, and restoring soil 2.1. Site/soil description fertility, while reducing emissions (FAO, 2011; Pretty et al., 2018; Dixon et al., 2020; Gathala et al., 2020a). Conservation agriculture (CA) can From 2013–2016, a field study was conducted at the Regional facilitate improved crop establishment and timely sowing, maintain or Agricultural Research Station of the Bangladesh Agricultural Research increase yield, lower water and energy use, lower production cost and Institute (BARI) in the sub-tropical climate of Jamalpur (24◦ 56’ increase income, and improve the quality of soil, while improving sys- 30.68’’N lat., 89◦ 55’ 40.39’’E long., 21.6 m) in northern Bangladesh. tem resilience (Verhulst et al., 2010; Kumar et al., 2013, 2019; Gathala Jamalpur soils belong to AEZ 9 (Old Brahmaputra Floodplains), which et al., 2013, 2015, 2016, 2020b; Jat et al., 2014). Research in south Asia are characterized as raw alluvial, noncalcareous grey soils, known for has shown that CA or strip tillage (ST), with residue retention, typically poor drainage. AEZ 9 is part of the Brahmaputra-Jamuna Floodplains of results in greater yields and profits from non-rice crops compared to rice northern and eastern slopes of the Himalayas and has a catchment area (Erenstein and Laxmi, 2008; Gathala et al., 2015; Islam et al., 2019; of 583000 km2 that covers about 16,436 km2 area in Bangladesh. The Gathala et al., 2021). Farmers survey data also tend to confirm these soil was analysed at BARI for textural class, physical and chemical pa- findings (Keil et al., 2017; Akter et al., 2021). In conventional system of rameters, following the analytical standard procedure protocols. It was rice cultivation, rice seedlings are transplanted into repetitively tilled, classified as loam (sand, 47.3%; silt, 30.0%; and clay, 22.7% at 0–15 cm puddled, and flooded fields. However, seedlings can also be transplanted soil layer). Initial chemical analysis from 0 to 15 cm soil depth con- without puddling, which can save water, energy, labor, and overall ducted prior to the start of the experimentation in 2013 indicated that production cost for rice cultivation (Krupnik et al., 2014; Haque et al., the soil of the experimental site was slightly acidic-to-neutral (pH 6.2). It 2016; Hossen et al., 2018; Gathala et al., 2021). In practice, executing all had low soil organic matter (7.7 g kg− 1), very low total N (0.42 g kg− 1), the ‘required’ components of CA in rice-based systems where rice and low K and B (0.05 meq 100− 1 ml and 0.14 μg ml− 1), and medium S and succeeding crops are grown under contrasting environments is chal- Zn (10.58 μg ml− 1 and 0.80 μg ml− 1), but high P (12.8 μg ml− 1) and very lenging. There is an intermediate situation, however; farmers frequently high Cu, Fe and Mn (0.98, 36.0 and 3.7 μg ml− 1, respectively). make use of alternating tillage (AT), where monsoon rice is produced with full tillage and the dry winter season crops are established using 2.2. Weather information zero tillage (ZT) or ST, with or without residue retention (Krupnik et al., 2014; Keil et al., 2017). Weather data for the experimental period were monitored via the The most burdensome tasks of manual transplanting in rice and weather station installed at the experiment site. Total monthly rainfall in 2 M.A. Hoque et al. F i e l d C r o p s R e s e a r c h 291 (2023) 108791 each year varied considerably in amount and distribution. The total designated cropping rotations (each crop cycle was completed from July rainfall during the monsoon rice season (July–October) varied from 113 to June within a one-year annual crop calendar (Fig. 2)). In three years cm in 2014 to 83 cm in 2015 (Fig. 1). The maximum and minimum of experimentation, six and nine crops were grown under double crop- monthly mean temperatures during the rice season ranged from 33.0◦ to ping and triple cropping sequences, respectively. 34.7◦C and 22.6–28.4 ◦C, respectively, while monthly mean relative humidity varied from 75% to 88% across years. 2.3.1. Cropping systems The total rainfall during the winter ‘rabi’ season (November to April) All six cropping systems treatments had monsoon season rice in ranged from 10 cm in 2013–2014–31 cm in 2014–2015. The maximum common and differed in the winter (rabi) crops; two systems (rice-wheat monthly mean temperature ranged from 24.1◦ to 32.4◦C (2014–2015) to and rice-maize) were further intensified by including a short duration 23.5–36.4 ◦C (2013–2014), while the minimum monthly mean tem- mung bean in the spring season (March-June) as the third crop in one- perature ranged from 12.3◦ to 21.3◦C (2014–2015) to 12.2–28.7 ◦C in year crop cycle rotation (Fig. 2). The six cropping systems were 2013–2014. Temperatures were higher in April and May compared to monsoon aman rice–winter (boro) rice (R-R); aman rice-wheat (R-W); other months. The second year of experimentation (2014–2015) had aman rice–maize (R-M); aman rice–mung bean (R-MB); aman rice–- higher minimum and maximum monthly mean temperatures than the wheat–mung bean (R-W-MB); and aman rice–maize–mung bean (R-M- other two years, with the potential for heat stress and grain sterility. The MB). mean relative humidity remained below 70% in the drier months (March to May). 2.3.2. Three tillage options The three tillage treatments were conservation agriculture (CA), in 2.3. Experimental and treatment details which monsoon aman rice was transplanted without any preparatory tillage and puddling (wet tillage), while all other crops in rotation were The experiment was conducted in a split-plot design with four rep- sown under strip tillage (ST) without any prior tillage; alternate tillage lications, where six cropping systems were assigned to the main plots (AT) in which monsoon rice was transplanted in puddled soil (CT) while and with three tillage options in the subplots. The main plot size was the succeeding winter or spring crops were sown under ST (as with CA); 20 m × 20 m (400 m2) and the subplot size was 20 m × 6 m (120 m2). and CT, in which rice (both aman and boro) was transplanted in puddle The field experiment was conducted for three years, from monsoon (the soil and other winter crops were planted under intensive wet or dry ‘aman’ season) rice in 2013 to mung bean in the spring of 2016. In each tillage. annual crop cycle, two to three crops were grown in a sequence in In CT, tillage for land preparation for all crops and puddling for rice involved intensive traditional tillage with multiple 3–5 passes (i.e., 1–2 tillage operations to incorporate the leftover crop stubble prior to planting, followed by 2–3 operations including tillage and leveling for puddling and seeding). In AT, aman rice in all cropping systems was established under puddled condition, while for winter and spring season crops, seeds were sown into ~ 25 cm crop stubble, and crops were established under ST with residue retention. In CA, both monsoon and winter (boro) rice were established under non-puddled condition, while for other winter and spring crops, seeds were sown ~ 25 cm crop stubble and established under ST. Details of the cropping systems and tillage treatments are provided in Table S1. 2.4. Crop and residue management 2.4.1. Crop management Variety, seed rate, spacing, and fertilizer rates for each crop in each cropping system are presented in Table S2. The transplanting date was ≈ 15 July ( ± 3 days) for monsoon rice and ≈ 15 January ( ± 3 days) for winter boro rice; the sowing date was ≈ 15 November ( ± 3 days) for maize and wheat and ≈ 15 February ( ± 3 days) for mung bean in double cropping (R-MB) system and March end to April in triple crop- ping (≈ 25 March in R-W-MB and ≈ 25 April in R-M-MB, respectively) (Fig. 2) systems. The monsoon rice varieties under double- and triple- cropping systems were BR-11 and BINA DHAN-7, respectively. Other varieties were BRRI DHAN-29 for boro rice, BARI GOM-26 for wheat, NK 40 for maize, and BARI Mung-6 for mung bean. Fertilizers for each crop were applied according to rates recom- mended by the Bangladesh Agricultural Research Council (BARC), Fer- tilizer Recommendation Guide(FRG, 2018). Fertilizer rates for short-season transplanted aman season rice (T. aman), long-season T. aman, and boro rice were 68–22–25–30, 81–25–30–36, and 100–24–50–0 (N-P-K-S), respectively. Fertilizer rates for wheat and maize were 100–24–50–110 and 250–45–130–0 (N-P-K-S), respectively (Table S2). The same fertilizer rates, application, and management were followed across the systems and tillage practices in respective crops. For N, urea was applied at planting and 1–2 times as top-dressing. For P, K Fig. 1. Weather conditions during three years of experimentation in a sub- and S, diammonium phosphate, triple super phosphate, muriate of tropical environment, Jamalpur, Bangladesh (top: 2013–2014; middle: potash and gypsum were applied at planting, except for the second 2014–2015; bottom: 2015–2016). application of K to maize at tasseling. Weeds in the rice fields were 3 M.A. Hoque et al. F i e l d C r o p s R e s e a r c h 291 (2023) 108791 Fig. 2. Annual crop calendar for different cropping systems. R-R, R-W, and R-M indicate rice-rice, rice-wheat and rice-maize rotations, while R-MB, R-W-MB and R- M-BM indicate rice-mung bean, rice-wheat-mung bean, and rice-maize-mung bean rotations. controlled under CT and AT through puddling, tillage and manual more than 5 t ha− 1, whereas, in AT and CT, maize was harvested at weeding, but as part of CA, for rice and other crops, they were controlled ≈ 5 cm height in the R-M system. In triple cropping systems (R-W-MB by glyphosate and the pre-emergence herbicide Pretilachor. Due to the and R-M-MB), the winter crops (wheat and maize) were reaped at ≈ 25 absence of pests and diseases, there was no need for pesticide applica- and ≈ 40 cm, respectively, under CA and AT. In addition, whole stover tion (see Table S1 for details). of mungbean was retained in the field after final pod collection in CA Aman rice was grown as a rainfed crop, apart from about 180 mm of and AT, but this was removed in CT. During the three years of experi- irrigation water applied for puddling in each season. In boro rice, mentation, in CA under triple cropping systems (R-W-MB and R-M-MB), 1186–1795 mm of irrigation water was applied depending on rainfall on average annual crop biomass 8.0 and 11.3 t ha− 1, respectively, were amount and distribution across the seasons, where CA and AT required recycled; this was varied from 5.4 to 9.0 t ha− 1 in double cropping 400 and 200 mm less water than CT, respectively. Under CA and AT, systems (Table S3). flood irrigation was applied to achieve soft soil two to three days prior to seedling transplanting. After rice seedling transplanting, frequent irri- 2.5. Crop and systems grain, protein, and calorie productivity gation water was applied on a regular basis (3–4 days intervals) as in the conventional system to ensure proper crop establishment; then, irriga- 2.5.1. Yield sampling and measurements tion was applied by alternate wetting and drying. In each plot, one The procedure for rice and wheat grain and straw yield sampling was perforated PVC pipe (30 cm long, up to 20 cm deep) was set up. The soil similar. The samples for grain yield estimation were taken from the was then excavated from the pipe and daily water levels were measured. centre of the plots, from an area of 18 m2 in each plot. The harvested rice When the free-moving water level was more profound than 15 cm in any and wheat plants were bundled and left in the plots for thorough sun of the four replications, irrigation was applied to a flood depth of 7–8 cm drying, after which the grain and straw samples were weighed, followed above the soil surface, depending on the stage and height of the rice by threshing and weighing of the grain. Concurrently, grain moisture crop. In CT, however, full flood irrigation of at least 7–8 cm depth was readings were taken using a moisture meter (Draminski Co., Olsztyn, applied regularly for up to 30 days after transplanting to maintain Poland). Rice and wheat grain yields were adjusted to 14% moisture floodwater and then re-irrigated when hairline cracks were observed on content. The pre-weighed straw sub-samples were dried in an oven for the soil surface (Gathala et al., 2011). Irrigation was stopped two weeks 72 h at 70 ◦C to a constant weight. The samples were then weighed with before harvest. a precision balance. In maize, depending on seasonal rainfall, up to five irrigations at five Across all treatments, maize samples from each plot were taken from stages were applied with a total of 254–351 mm during the season: (1) at 5 m of 6 rows, with 1 m of row ends and each outside row excluded, the three-leaf stage (V3), (2) at the six-leaf stage (V6) after fertilizer was giving a harvest area of 18 m2. All the cobs from the sampling area were banded, (3) at V10 after banding, (4) at silking, and (5) at the grain harvested. Sampled cobs were thoroughly dried and weighed, after watery milking stage. In all tillage treatments in wheat, maximum of which they were shelled. A sub-sample of five representative cobs was four irrigations were applied: (1) at crown root initiation (CRI), (2) at collected for moisture determination. Maize yield was adjusted to 15.5% late tillering, (3) at flowering, and (3) at milking. On average, wheat moisture content. crop received 115–189 mm across treatments. Mung bean relied on soil- Mung bean samples for grain yield were taken from an 18 m2 area stored water and rainfall apart from one pre-sowing light irrigation. (6.0 m x 3.0 m). Due to the indeterminate growth habit of mung bean varieties, pods were harvested twice from each plot and straw yield 2.4.2. Crop residue management recorded after the second picking. The pods were sun-dried, weighed, At each crop harvest, leftover crop residue quantity was measured in threshed, and grains cleaned. Grain moisture was measured using a all treatment plots (Table S3). However, amount of crop residue retained moisture meter and yield was adjusted to 12% moisture content. In plots and soil surface coverage as mulch depended on crops, cropping sys- where neither pods nor grains had formed or matured, straw biomass tems, and tillage options. Aman rice was harvested at the height of was still recorded from a 2 m2 area. Straw biomass was weighed using a ≈ 25 cm in CA and AT and at ≈ 5 cm in CT. The succeeding winter crops field balance and sub-samples were dried in an oven for 72 h at 70 ◦C, were seeded/planted into about 3.0 t ha− 1 crop residue as a surface with dry weight recorded when constant weight was achieved. mulch/stubble in CA and AT. In CT, about 0.70 t ha− 1 crop residue was incorporated during land preparation. In R-W and R-R cropping systems, 2.5.2. Computation of rice equivalent yield wheat and boro rice were harvested at ≈ 25 cm height which was about The rice equivalent yield (REY) for the component crops of each 2.5–3.5 t ha− 1 crop residue retained as a mulch in CA plots, but in AT cropping system was calculated from the price in Bangladesh currency and CT, both crops were harvested at the height of ≈ 5 cm; this was ha− 1 (BDT ha− 1) as follows: incorporated before next aman rice. In CA, under the R-MB system, the whole mung bean residue (about 2.5 t ha− 1) was retained in the field. The maize stover was harvested at ≈ 40 cm height in CA, which was 4 M.A. Hoque et al. F i e l d C r o p s R e s e a r c h 291 (2023) 108791 ( ) / ( ) ( ) grainyield kgha− 1 ∗ priceof respectivecrop variety BDTkg− 1 grain calories of 345 Kcal, 344 Kcal, 344 Kcal and 351 Kcal per 100 g REY tha− 1 = price of BR 11rice(BDTkg− 1) ∗ 1000 were used for rice, wheat, maize and mung bean, respectively, to calculate grain protein (kg ha− 1) and grain calorie (Gcal ha− 1) equiva- To calculate REY, the prices of respective individual crop/cultivar lent yields. Although protein and caloric values may change following used in each cropping system in each year relative to the price of long- grain processing and cooking, we present them as an indicator of po- duration monsoon rice (cv. BR11). The respective crop of short-duration tential protein and energy availability from each treatment. System monsoon rice (cv. BINA DHAN-7), winter rice (BRRI Dhan-29), wheat, protein and calorie equivalent yields were obtained from the sum of all maize and mung bean prices, respectively and individual crop prices component crops’ protein or calorie yields for a cropping system. The (Table 1) were used in their respective seasons. The equivalent rice yield relative changes in SREY, system protein and calorie yields, and systems for each system (i.e., SREY, t ha− 1) was determined by summing the REY gross margins were compared for CA and AT against CT and for different of individual crops for each cropping system or tillage treatment. cropping systems against R-R system. 2.5.3. Computation of grain protein and grain calorie equivalent yields 2.6. Economic analysis In addition to REY, grain protein and calorie equivalent yields were also calculated using the grain protein and calorie conversion factors All inputs required to produce a crop and economically valuable reported in Bangladesh’s Food Composition Tables (Shaheen et al., outputs were recorded during each cropping season. The total variable 2013). Grain protein of 6.6 g, 11.2 g, 9.9 g and 23.7 g per 100 g and costs were considered for computing the total production cost. The variable costs included land use (land rent), labor use, tillage, machinery Table 1 use (for planting), seed, fertilizer, agro-chemicals, irrigation, weeding, Prices for various inputs and marketable outputs used during the three years of harvesting, and threshing. The human labor cost (based on 8-hour experiment in a sub-tropical environment, Jamalpur, Bangladesh. person-day) was accounted for land preparation, seeding/trans- planting, irrigation, fertilizer and agro-chemicals application, intercul- Input or output (prices in USDy) 2012–13 2013–14 2014–15 tural operations (such as weeding and bundling), harvesting, and Input category threshing and cleaning. The labor cost was computed using the labor Labor Wage (USD person- 3.10 3.10 3.10 day− 1) wage rate of the research station where the field experiment was con- Land use USD ha− 1 yr− 1 274 274 274 ducted. The irrigation cost was estimated as the total hours required for Seed Maize (USD kg− 1) 5.95 5.95 5.95 irrigation in each plot and then multiplied by the electricity unit (Kw Wheat (USD kg− 1) 0.42 0.42 0.42 h− 1) and per unit electricity charge. Similarly, the time (h) required by a Aman rice (USD kg− 1) 0.24 0.24 0.24 two- or four-wheel tractor-drawn machine to complete each field short variety Aman rice (USD kg− 1) 0.24 0.24 0.24 operation (such as tillage, seeding and threshing) was recorded and long variety expressed as h ha− 1; simultaneously the fuel consumption (L ha− 1) in Boro rice (USD kg− 1) 0.24 0.24 0.24 each plot for each operation was recorded. Gross returns (GR) were Mung bean seed (USD 0.89 0.89 0.89 kg− 1 computed by multiplying the grain and straw/stover yield of each crop ) Fertilizer Urea (USD kg− 1) 0.19 0.19 0.19 by the offered prices in the established local market, which varied from TSP (USD kg− 1) 0.26 0.26 0.26 season to season (Table 1). Gross margins (GM) were calculated as the MoP (USD kg− 1) 0.18 0.18 0.18 difference between GR and total variable cost (TVC). The systems GR, Gypsum (USD kg− 1) 0.18 0.18 0.18 TVC and GM were calculated by adding together the associated costs and ZnSO4 (USD kg− 1)) 2.14 2.14 2.14 benefits of the harvested crops within a crop calendar year. All input and Borax (USD kg− 1) 0.00 0.00 0.00 Irrigation Amount and application 0.89 0.89 0.89 output prices are presented in Table 1. fee (USD hr− 1) The distributions of labor use and gross margin under different Fuel Diesel ( USD L− 1) 0.81 0.81 0.81 cropping systems were plotted with scatter quadrate charts. Based on Threshing and Aman rice (USD t− 1 17.86 17.86 17.86 labor use and gross margin, four cropping system groups were desig- shelling grain) Boro rice (USD t− 1 17.86 17.86 17.86 nated: (1) low labor use and low gross margin (low-low); (2) low labor grain) use and high gross margin (low-high); (3) high labor use and high gross Wheat (USD t− 1 grain) 23.81 23.81 23.81 margin (high-high); and (4) high labor use and low gross margin (high- Maize (USD t− 1 grain) 5.95 5.95 5.95 low). Mung bean (USD t− 1 11.90 11.90 11.90 grain) Herbicide and Glyphosate (USD 10.48 10.48 10.48 2.7. Statistical analysis pesticide litre− 1) Afinity (USD litre− 1) 17.86 17.86 17.86 Before performing statistical analysis, the normality assumption of Pretilachor (USD 10.71 10.71 10.71 analysis of variance (ANOVA) was checked by Shapiro and Wilk (1965) litre− 1) Tilt (USD litre− 1) 14.88 14.88 14.88 using the JMP statistical software (V11 software, Buckinghamshire, UK). Furadon (USD kg− 1) 1.49 1.49 1.49 The test for homogeneity of variance was also performed using the Crop Bartlett’s test (Snedecor and Cochran, 1989). There was no need for data Maize Grain price (USD t− 1) 178.57 184.52 184.52 transformation as the normality assumption of ANOVA was fully met. Stover price (USD t− 1) 8.93 8.93 8.93 We followed the procedure to build statistical models for data analysis Cob price (USD t− 1) 5.95 5.95 5.95 Wheat Grain price (USD t− 1) 238.10 238.10 238.10 for split plot design in fixed plots, which is suggested for long-term ex- Straw price (USD t− 1) 5.95 5.95 5.95 periments with the complexity of cropping system: CS (main-plot), Aman rice (short Grain price (USD t− 1) 196.43 196.43 196.43 tillage; T (sub-plot) and year (Y), together with within-year repli- duration) Straw price (USD t− 1) 5.95 5.95 5.95 cation/block (R) (Onofri et al., 2016). We performed the serial corre- Aman rice (long Grain price (USD t− 1) 178.57 178.57 178.57 duration) Straw price (USD t− 1) 5.95 5.95 5.95 lation using a random-effects model, which accounted for the compound Boro rice Grain price (USD t− 1) 178.57 178.57 178.57 symmetry variance-covariance structure. The analysis progressed using Straw price (USD t− 1) 8.93 8.93 8.93 a complete three-factor analysis of variance (ANOVA). The treatments – Mung bean Grain price (USD t− 1) 714.29 714.29 714.29 ‘CS’ and ‘T′ - were fixed effects and were randomly allocated to plots. Stover price (USD t− 1) 5.95 5.95 5.95 ‘Year’ was a repeated factor; this was combined with the treatment † conversion rate: 1 USD = 84 BDT. model by introducing the term “Year + CSxY + TxY + CSxTxY”. 5 M.A. Hoque et al. F i e l d C r o p s R e s e a r c h 291 (2023) 108791 ‘Replication’ was a randomized unit, so we kept it under random effect bean REY. Winter rice (boro) REY under the R-R system was 15% higher by including replication interactions with all treatments. The final than the wheat REY, compared to the REYs under the R-W or R-W-MB model was tested using the JMP software. system, but winter crops REYs in the R-R and R-W systems were significantly lower (by 41.7% and 63.5%, respectively) than the maize Fixed effect: Y + CS + T + CSxY + TxY + CSxT + CSxTxY REY in the R-M and R-M-MB systems. Considering systems level REYs, Random effect: R + RxY + RxCS + RxCSxY + RxCSxT all six cropping systems followed the order from highest to lowest: R-W- MB ≥ R-M ≥ R-M-MB > R-R ≥ R-W ≥ R-MB. The R-W-MB system REY All variable means were compared using Tukey’s honest significant was 5%, 11%, 30%, 43% and 45% greater than those of R-M, R-M-MB, R- difference at p = 0.05, where significant treatment means were sepa- R, R-W and R-MB systems, respectively. The R-R system consistently had rated using alphabet letters. a higher system REY than the R-MB system. Among the tillage treat- ments, CT and AT had significantly higher (by 8.7% and 4.3%, respec- 3. Results tively) monsoon rice REYs than CA; however, under CA, compared to CT, the winter REY was 12% higher and system REY was 5% higher 3.1. Crops and cropping systems grain, calorie, and protein yields (Table 2). The cropping system by tillage interactions effect revealed that the ANOVA indicated that the grain, calorie, and protein yields did not REY of winter maize under CA and AT in R-M and R-M-MB rotations differ across years except for winter crops’ calorie and protein yields (8.8 t ha− 1) was the highest, followed by CT under the same rotations (Table 2). Individual cropping system and tillage effects on grain, pro- (7.7 t ha− 1). On the other hand, the REY of mung bean was lowest under tein, and calorie yields were influenced significantly. Still, cropping CT in R-MB, followed by CT in wheat under R-W and R-W-MB rotations system by tillage and cropping system by year interactions were sig- (Figure not shown). All cropping systems except R-R had the highest nificant for winter crop and cropping system yields, but not for monsoon winter crop REY under CA, followed by AT. The yield increases of winter rice. The year by tillage and year by cropping system by tillage in- crops under CA compared to CT ranged from 12% to 18%, with the teractions also did not affect yields significantly (Table 2). highest increase under R-W and the lowest under R-M and R-M-MB ro- tations. With AT, the yield increases of winter crops under different 3.1.1. Crops and cropping systems grain yields systems ranged from 11% to 14%. For R-R rotation, however, there was In this study, the REY of monsoon rice, winter crops, and cropping a 1% yield decrease under AT and a 4% decrease under CA. The R-M and systems remained unchanged by season. The monsoon REY of the short- R-W-MB rotations under CA and AT had the greatest systems-level yields duration rice variety (BINA DHAN − 7) was lower by about 8% in the (13.7–14.6 t ha− 1), while the R-W and R-MB systems, regardless of triple cropping systems (R-W-MB, R-M-MB) than the long-duration va- tillage treatment, had the lowest (9.4–10.2 t ha− 1) system yields (Fig. 3). riety (BR11) in the double cropping systems (R-R, R-M, R-W, R-MB). In all cropping systems except R-R, CA and AT had higher systems REY Considering the winter season crop REY, the R-M system produced the (by 6–9%) than CT. highest maize REY while the R-MB system produced the lowest mung Table 2 Effect of six cropping systems and three tillage options on rice equivalent grain yield, grain calorie, and grain protein for crops and cropping systems (2013–2016) in a sub-tropical environment, Jamalpur, Bangladesh. * and * * significance level at 0.01 and 0.05; † values in parenthesis are actual grain yields of respective crops; § values in parenthesis {} are rice equivalent yields of spring mung bean crop; Means followed by a common letter within a column are not significantly different by the HSD-test (Tukey’s honestly significant difference) at the 5% level of significance; cropping systems: R-R = rice-rice, R-W = rice-wheat, R-M = rice-maize, R-MB = rice-mungbean, R-W-MB = rice-wheat-mung bean, R-M- MB = rice-maize-mung bean; tillage and crop establishment treatments: CA = conservation agriculture, AT = alternate tillage, CT = conventional tillage. 6 M.A. Hoque et al. F i e l d C r o p s R e s e a r c h 291 (2023) 108791 Fig. 3. Interactive effects of cropping system and tillage options on system rice equivalent yields of grain (t ha− 1), system calorie (Gcal ha− 1), and system protein (kg ha− 1) in a sub- tropical environment, Jamalpur, Bangladesh. CA, AT, and CT indicate conservation agricul- ture, alternate, and conventional tillage, respectively. R-R, R-W, and R-M stand for rice- rice, rice-wheat and rice-maize rotations, while R-MB, R-W-MB and R-M-MB indicate rice-mung bean, rice-wheat-mung bean, and rice-maize- mungbean rotations. The shape of the legend in respective treatment combinations shows the data distribution to across years and replica- tions. The legends mean followed by a common letter horizontally are not significantly different by the HSD-test (Tukey’s honestly significant difference) at the 5% level of significance. 3.1.2. Crops and cropping systems grain protein and calorie equivalent systems, where CA provided higher calorie and protein yields, respec- yields tively, by 12.6% and 14.4% in winter crops and 3.0% and 6.4% in Analysis of variance of the main effects of crop system and tillage complete rotational systems (Table 2). showed a highly significant response to calorie and protein yields when The interaction effects of the cropping system by tillage showed the both seasons and cropping systems were taken into consideration highest grain calorie and protein yields (48 Gcal and 1208 kg ha− 1, (Table 2). The grain calorie and grain protein yields of winter crops were respectively) under CA and AT in the R-M system; the R-MB system had higher by 4.5% and 4.9% in Year 1, respectively than in Year 2 or 3; this the lowest calorie and protein yields (21 Gcal and 606 kg ha− 1, was due to better crop performance in Year 1. respectively) under all tillage treatments (Fig. 3). In both R-M and R-M- Both grain calorie and grain protein yields followed almost similar MB systems, the grain calorie and protein yields were significantly lower trends as in REY for both seasons and cropping systems. In the monsoon under CT than CA and AT. The increase in systems protein yield across rice season, the long-duration rice produced about 18.0% higher grain cropping systems (except for R-R) ranged 3–10% under CA and 8–10% calorie and protein yields than shorter-duration rice. In winter crop under AT compared to CT. season, maize had the highest calorie and protein yields and mung bean the lowest, with the former six times higher than the latter. Winter boro rice also produced 62% and 392% higher grain calories than wheat and 3.2. Crops and cropping systems economic analysis and labor use mung bean grains, respectively. Similarly, while maize yielded the highest protein yield, mung bean yielded the lowest; wheat and winter 3.2.1. Production cost and gross margin rice achieved statistically similar protein yield. At the cropping systems Production costs for both monsoon rice and different cropping sys- level, grain calorie and protein yields respectively followed the order tems were lowest in Year 3 compared to Years 1 and 2. (Table 3). In from highest to lowest: R-M > R-M-MB > R-R > R-W-MB = R-W > R- monsoon season, the production cost of the short-duration rice variety MB and R-M > R-M-MB > R-W-MB > R-W = R-R > R-MB. The R-M (BINA DHAN − 7) was lower by about 7.5% in the triple cropping sys- system consistently performed better in terms of both calorie and pro- tems (R-W-MB, R-M-MB) than that of the long-duration variety (BR11) tein yields due to high yield potential of winter maize (Table 2), together in the double cropping systems (R-R, R-M, R-W, R-MB) due to reduced with good protein content. The R-R system had the second highest cal- labor and associated inputs required for short duration cultivars orie yield and R-W-MB the second highest protein yield, the latter due to (Table 3). Of the winter crops, the winter boro rice had the highest the inclusion of the third crop in the system. production cost (USD 872 per ha− 1); this was 123% higher than the Tillage significantly influenced both grain calorie and protein yields lowest cost crop, mung bean (USD 391 per ha− 1). The second highest of all crops in both seasons and for all cropping systems. CT yielded 8.4% production cost was for maize in the R-M and R-M-MB systems; these and 7.8% greater calorie and protein yields in monsoon rice than in CA. costs were about 19% lower than for winter rice but 59% and 86% However, this trend was quite opposite in winter crops and cropping higher than for wheat and mung bean. Across cropping systems, the R- M-MB system was the most input intensive, with its production cost 7 M.A. Hoque et al. F i e l d C r o p s R e s e a r c h 291 (2023) 108791 Table 3 Effect of six cropping systems and three tillage options on rice equivalent labor use, production cost, and gross margin for crops and cropping systems (2013–2016) in a sub-tropical environment, Jamalpur, Bangladesh. Source Cost of production (USD ha− 1) Gross margin (USD ha− 1) Labor use (Person days ha− 1) Monsoon crop Winter crop System Monsoon crop Winter crop System Monsoon crop Winter crop System Year (Y) Year 1 650a 611a 1372a 276a 587a 881a 108a 81a 211a Year 2 650a 599a 1358a 238a 571a 827a 107ab 76a 205ab Year 3 596b 605a 1305b 289a 548a 862a 96b 77a 193b Cropping system (CS) R-R 652a 872a 1524b 253a 240c 494e 107a 138a 245a R-W 646a 460c 1106e 268a 399b 667d 105ab 39d 144e R-M 646a 722b 1368d 288a 939a 1227a 105ab 73c 178d R-MB 650a 391d 1041 f 287a 449b 736d 106a 105b 211c R-W-MB 600b 453c 1467c{414}§ 260a 475b 1079b{344}§ 100bc 36d 230b{94}§ R-M-MB 599b 733b 1563a 251a 909a 939c 99c 76c 209c Tillage option (T) CA 617c 604b 1326b 245b 626a 902a 100b 75c 195c AT 635b 614a 1356a 275a 595b 896a 104a 77b 202b CT 645a 597b 1352a 284a 485c 773b 106a 81a 212a Analysis of variance (p-values) Y 0.006 * * 0.292 0.002 * * 0.289 0.245 0.293 0.006 * * 0.089 0.003 * * CS < 0.001 * * < 0.001 * * < 0.001 * * 0.646 < 0.001 * * < 0.001 * * < 0.001 * * < 0.001 * * < 0.001 * * T < 0.001 * * < 0.001 * * < 0.001 * * 0.003 * * < 0.001 * * < 0.001 * * < 0.001 * * < 0.001 * * < 0.001 * * Y × CS 0.463 < 0.001 * * < 0.001 * * 0.199 < 0.001 * * 0.002 * * 0.338 0.003 * * < 0.001 * * Y × T 0.005 * * 0.365 0.007 * * 0.406 0.172 0.285 0.002 * * 0.039 * 0.264 CS × T 0.023 * < 0.001 * * < 0.001 * * 0.934 < 0.001 * * < 0.001 * * 0.012 * < 0.001 * * < 0.001 * * Y × CS × T 0.596 < 0.001 * * 0.133 0.999 0.675 0.934 0.209 < 0.001 * * 0.010 * * and * * significance level at 0.01 and 0.05; § values in parenthesis are rice equivalent of spring mung bean crop; Means followed by a common letter within a column are not significantly different by the HSD-test (Tukey’s honestly significant difference) at the 5% level of significance; R-R = rice-rice, R-W = rice-wheat, R-M = rice- maize, R-MB = rice-mungbean, R-W-MB = rice-wheat-mungbean, R-M-MB = rice-maize-mungbean; tillage and crop establishment treatments: CA = conservation agriculture, AT = alternate tillage, CT = conventional tillage. Fig. 4. Interactive effects of cropping system and tillage option on winter labor use and gross margin, and system labor use (person days ha− 1), production cost (USD ha− 1) and system gross margin (USD ha− 1) in a sub-tropical environment, Jamalpur, Bangladesh. CA, AT, and CT indicate conservation agriculture, alternate, and conventional tillage, respec- tively. R-R, R-W, and R-M indicate rice-rice, rice-wheat, and rice-maize rotations, while R- MB, R-W-MB, and R-M-BM indicate rice-mung bean, rice-wheat-mung bean, and rice-maize- mungbean rotations. The shape of the legend in the respective treatment combinations shows the data distribution across years and replica- tions. The legends mean followed by a common letter horizontally are not significantly different by the HSD-test (Tukey’s honestly significant difference) at the 5% level of significance. 8 M.A. Hoque et al. F i e l d C r o p s R e s e a r c h 291 (2023) 108791 higher than for R-R, R-W-MB, R-M, R-W and R-MB systems by about 3%, system by tillage showed the highest number of person-days ha− 1 7%, 14%, 41% and 50%, respectively. The introduction of mung bean in required for the R-R system with crops grown under CT or AT, while the the R-W and R-M cropping systems incurred an additional 22% cost, lowest number of person-days ha− 1 was required for the R-W system equaling approximately USD 278 ha− 1. with all tillage methods. In R-MB and R-W systems, the labor use didn’t In all three tillage options, the production cost of monsoon rice was make any difference among the three tillage options because of highest for CT and lowest for CA, whereas it was intermediate for AT. comparatively lower labor required for mung bean and wheat planting For winter crops, it was about USD 14 ha− 1 higher in AT than in CT or and land preparation in conventional system compared to other winter CA, while at the cropping systems level, it was about 2% lower in CA crops. Whereas winter maize and boro rice required more labor than CT or AT (Table 3). The cropping system’s interaction effect by respectively for dibbling and seedling transplanting, CA-based practices tillage showed that the highest production cost was for R-R and R-M-MB reduced effective labor requirement for R-M, R-M-MB, R-R, and R-W-MB systems grown under CT and lowest for R-MB with CA (Fig. 4). At sys- systems over CT (Fig. 4). tems level, the R-R system under CA had significantly lower production cost than under CT due to the elimination of puddling and less water 3.2.3. Distribution of gross margin and labor use requirement for crop establishment. The distribution of labor use and gross margin among six cropping In general, gross margins of all crops and cropping systems were systems is demonstrated in Fig. 5. The R-W rotation fell under the low- similar among years (Table 3). The gross margin for monsoon rice varied low group, with 100–190 person-days ha− 1 labor use and 250–900 USD from USD 251 to USD 288 ha− 1, but it did not differ among cropping ha− 1 gross margin. On the other hand, the R-M rotation fell under the systems. For winter crops, the highest gross margin was for maize in the low-high group, with labor use almost the same as for the low-low group R-M and R-M-MB rotations, followed by wheat and mung bean in the R- but with a higher gross margin (900–1550 USD ha− 1). The triple crop- W, R-W-MB and R-MB rotations. The lowest gross margin was for winter ping systems R-W-MB and R-M-MB were grouped under high-high, rice in the R-R rotation; this was 66% lower than the gross margin for indicating more labor intensive (190–280 person-days ha− 1) but with winter wheat in the R-W rotation. At the cropping systems level, the R-M high return (900–1550 USD ha− 1); R-M-MB also overlapped with the rotation had the highest gross margin; this was higher than for the R-R, high-low group. Finally, the R-R and R-MB rotations were grouped R-W, R-MB, R-M-MB and R-W-MB systems, by 149%, 84%, 67%, 31% under the high-low group, indicating that these are more labor intensive and 14%, respectively. The difference in gross margins between the (190–280 person-days ha− 1) with lower gross margin (250–900 USD highest (R-M) and the lowest (R-R) system was USD 733 per ha− 1. ha− 1) (Fig. 5). The CT achieved a gross margin of about USD 39 ha− 1 higher than CA in monsoon rice; however, CA had a higher gross margin than CT by USD 141 and USD 129 ha− 1 from winter crops and different cropping 3.3. Relative change in CA and AT compared to CT and various rotations systems, respectively. For both winter crops and different cropping compared to R-R rotation systems, AT also had a higher gross margin than CT, by USD 100 and USD 113 ha− 1, respectively. The interaction effects of cropping system The relative changes in SREY, system grain protein and calorie by tillage demonstrated that the R-M system practised with CA resulted yields, and system gross margin over CT and R-R rotation compared to in the highest gross margin. In contrast, the R-R system practised with alternative tillage options (CA and AT) and cropping systems are pre- CA resulted in the lowest gross margin (Fig. 4). The gross margins were sented in Fig. 6. The SREY increased under CA and AT by 4.6% and significantly higher in R-M, R-M-MB and R-W-MB systems by 16%, 32% 6.1%, respectively, compared to CT. These higher yields, in combination and 22%, respectively, under CA and AT than CT, but tillage options did with lower production costs under CA and AT, resulted in approximately not influence in the rest of the cropping systems (R-MB, R-W and R-R). 16% higher gross margin compared to CT. Regarding protein and calorie These results indicated that CA and AT practices are more profitable for yields, the increase was about 7% and 4%, respectively, under CA and the former than the latter systems. AT compared to CT. The change in SREY in the R-W-MB system compared to the R-R system was highest (30%) but negative (− 8%) 3.2.2. Labor use under the R-W system. There was a 24% increase in SREY under the R-M The main differences in labor use between different cropping systems system compared to the R-R system. In contrast, the increase was lower or tillage options are accounted for by differences in labor required for under the R-M-MB system due to the inability of farmers to harvest different crops under different tillage options in different seasons. The mung bean seed because of the shorter growing window between maize interaction effect of cropping system by tillage had a significant impact and monsoon rice (Fig. 2). A negative grain yield change (− 10%) and on labor use for both crop seasons and cropping systems (Table 3). negative protein yield change (− 14%) were also recorded for the R-MB Across the seasons, the number of person-days required for all crops and system compared to the R-R system. The highest difference in protein cropping systems was lower in Year 3 than in Year 1. For monsoon rice yield (~62%) was observed for the R-M and R-M-MB systems, followed grown under different cropping systems, the short-duration variety by the R-W-MB system (34%) compared to the R-R system. In terms of (grown under the triple-cropped system) required six person-days ha− 1 calorie yield, only the R-M and R-M-MB cropping systems had a positive less than the long-duration variety (grown under the double-cropped yield change, while the R-MB, R-W and R-W-MB systems showed system). Among the winter crops, the highest labor use (138 person- negative changes compared to the R-R system. Likewise, compared to days ha− 1) was required for winter rice cultivation, which was on the R-R system, all the alternative cropping systems had positive average was higher by 33, 63 and 100 person-days ha− 1 for mung bean, changes in gross margins, with highest (148%) under R-M and lowest maize and wheat, respectively. At the cropping systems level, the R-R (35%) under R-W. When we further deepened the analysis on the system had the highest labor use (245 person-days ha− 1), followed by R- combined effect of the tillage option by cropping system, SREY was W-MB (230 person-days ha− 1), and lowest for R-W (144 person-days higher by 55% and 42% in R-W-MB and R-M systems, respectively, with ha− 1). The intensification of the R-W to R-W-MB and R-M to R-M-MB CA compared to R-W system with CT (Fig. 3). In terms of system protein systems required 41 additional person-days ha− 1. and calorie yields, they were higher by 105% and 122% in the R-M The number of labor person-days ha− 1 under different tillage treat- system with CA compared to the R-MB system with CT. Likewise, the ments differed with crop seasons and the cropping system. The total highest and lowest labor use was associated with CA in R-R and R-W labor required for CA was fewer than for CT by 5, 6 and 17 person days systems, respectively, whereas the R-R system with CA used more than ha− 1 for monsoon rice, winter crops and cropping systems, respectively. 112 person days ha− 1 compared to the R-W system with CA. Similarly, AT also required fewer person-days ha− 1 than CT for different winter the highest gross margin (178%) was achieved under the R-M system crops and cropping systems. The interaction effect of the cropping with CA while over the lowest was for R-R system with CA (Fig. 4). 9 M.A. Hoque et al. F i e l d C r o p s R e s e a r c h 291 (2023) 108791 Fig. 5. Distribution of gross margin in respect of labor used for different cropping systems. The distribution is divided into four categories according to each system’s gross margin and labor used with respect to cropping systems: Low-Low, Low-High, High-High, and High-Low represent low labor use and low-profit margin, low labor use and high-profit margin, high labor use and high-profit margin, and high labor use and low-profit margin, respectively. R-R, R-W, and R-M indicate rice-rice, rice- wheat, and rice-maize rotations, while R-MB, R- W-MB, and R-M-BM indicate rice-mung bean, rice-wheat-mung bean, and rice-maize-mung bean rotations. 4. Discussion 4.1. Effects on crops and cropping systems grain, calorie, and protein productivity 4.1.1. Crops and cropping systems grain yield In this study, monsoon REY was lower by about 8% in the triple cropping systems than in the double cropping systems due to the use of the short-duration BINA DHAN-7 rice variety in the former. This variety has lower yield potential because of fewer grains per panicle and lower crop biomass than the long-duration variety BR-11 (Tiongco and Hos- sain, 2015). In winter, both maize and boro rice produced higher crop biomass and grain yield compared to wheat (Table S3). Their higher yield potential could be attributed to significantly higher number of growing degree days due to longer crop seasons, fewer cloudy days, and higher amount of solar radiation (Timsina et al., 2010; Gathala et al., Fig. 6. Relative change in system protein, system calorie, system rice equiva- 2016). Considering systems REY, all cropping systems followed the lent yield (SREY), and system gross margin over conventional tillage/R-R sys- tem over three years of study. CA, AT, and CT indicate conservation agriculture, order from highest to lowest: R-W-MB ≥ R-M ≥ R-M-MB > R-R ≥ R-W alternate, and conventional tillage, respectively. R-R, R-W, and R-M indicate ≥ R-MB. Although the relatively low-yielding, short-duration rice rice-rice, rice-wheat, and rice-maize rotations, while R-MB, R-W-MB, and R-M- cultivar was used in the R-W-MB rotation, the aggregate REY of this BM indicate rice-mung bean, rice-wheat-mung bean, and rice-maize-mung system was still higher due to the inclusion of mung bean in the rotation, bean rotations. which has a considerably higher market price (Table 1). The R-M system also produced higher system REY due to the use of the long-duration rice cultivar and the high potential productivity of the single cross maize hybrid (Timsina et al., 2010). The R-M-MB rotation did not produce a system yield higher than the R-M rotation despite the inclusion of short-duration mung bean because of its inability to set grains, as its grain-filling period coincided with the onset of the monsoon (Fig. 1). Consistent with our findings, Islam et al. (2019) also reported a higher 10 M.A. Hoque et al. F i e l d C r o p s R e s e a r c h 291 (2023) 108791 yield from the R-W-MB rotation compared to the R-R, R-W and R-M 2020; National Food and Nutrition Security Policy, 2020). Despite this, rotations in several locations of the EGP. However, in contrast to our the need for sustainable intensification and diversification of rice-based findings, Rashid et al. (2019) demonstrated that the R-M-MB rotation cropping systems under changing climate scenarios and ensuring the could be successfully grown under controlled conditions in clay loam food and nutrition security of the increasing population will continue to soils in high rainfall areas in research station in northwest Bangladesh, remain the major challenge in Bangladesh agriculture (FAO, 2020). Our though it would likely be difficult to disseminate to farmers’ field con- study can contribute to designing and developing the policy or strategies ditions. The R-R system consistently resulted in higher systems REY for achieving and sustaining food and nutritional security through sus- compared to the R-W or R-MB systems, but it requires intensive energy, tainable crop intensification and diversification (Gathala et al., 2020a; an excessive amount of water and labor, and entails a high cost of pro- Jat et al., 2020). In our study, the R-M system consistently performed duction (Gathala et al., 2015, 2016, 2020a). better in calorie and protein yields due to high yield of winter maize Traditionally, rice cultivation has been practiced under wet tillage (Table 2) and its high protein content (9.9%). The R-R system had the (puddled) conditions, creating soft anaerobic soil. Most rice varieties are second highest calorie yield and R-W-MB the second highest protein bred considering the suitability of seedling transplanting under wet yield. Higher calorie yield in the R-R system was due to the high-calorie tillage conditions. It is quite possible that such varieties do not perform content in rice grain (11.2%), while higher protein yield in the R-W-MB equally well under non-puddled (or zero tillage) conditions (Gathala system was due to the high protein content in mung bean (23.1%) et al., 2011; Chakraborty et al., 2017; Jat et al., 2019b). Our results also (Shaheen et al., 2013). Crop diversification, from R-R to R-M and confirmed that CT and AT (puddled for rice) had significantly higher (by R-W-MB systems, can provide balanced nutritional diets with higher 8.7% and 4.3%, respectively) monsoon REYs than CA among the tillage protein and calorie (Jat et al., 2020, 2018). treatments. On the other hand, CA resulted in higher REYs (12%) for Tillage significantly influenced the calorie and protein yields of all winter crops and different cropping systems than CT (Table 2). An in- crops in both seasons and of all cropping systems. In monsoon rice, CT crease in yield could be attributed to the continuous crop residue re- yielded 8.4% and 7.8% greater calorie and protein yields than CA. This tentions on the soil surface as mulch (Supplementary table 3), which trend was, however, quite the opposite for winter crops and cropping facilitates better growing conditions through more soil moisture con- systems, where CA had higher calorie and protein yields, respectively, servation, buffering surface soil temperature, and advancing the by 12.6% and 14.4% for winter crops and by 3% and 6.4% for cropping planting. These results are consistent with recent synthesis studies in the systems. These results agree with other findings in south Asia, where CA- EGP (Islam et al., 2019; Jat et al., 2020; Gathala et al., 2020b), where based management practices have resulted in 3.0–6.0% higher protein rice yields were reported to be either equal to, or lower under, CA than yield for different cropping systems (Jat et al., 2018; Jat et al., 2020). CT. However, these patterns were reversed for the succeeding dry season The higher calorie and protein yields under CA and AT reflected higher crops and rotational systems, with consistently higher yields under CA systems grain yield in these management practices. Adopting suitable than CT. cropping systems in combination with the right technologies can Recent synthesis studies with various crop rotations in south Asia improve the calorie and protein security of smallholder farmers in also demonstrate the potential of CA-based agronomic management for Bangladesh and south Asia (Islam et al., 2019; Gathala et al., 2020a; Jat increasing yields of many crops and cropping systems except for rice; in et al., 2020). The R-M system practiced with CA provided the highest these studies, the increment in grain yield varied depending on the system-level grain, calorie, and protein yields. Maize is not the direct layering of CA components over CT (Jat et al., 2020; Gathala et al., component of human diets in Bangladesh. However, this is a major 2020b). Singh et al. (2016) reported higher R-M system yields in source of poultry feed and chicken is one of the primary protein sources northwest India, and Islam et al. (2019) reported higher R-M and R-W of human diets in the country. Therefore, there is a scope to include and system yields in eastern India under ZT compared to CT. In the R-R promote maize in human diets by changing the consumers’ dietary system, however, CT provided a yield higher by 5% compared to AT and habits. Our findings confirm that adopting R-M and R-W-MB systems 8% compared to CA. Higher systems yield under the R-R rotation is practiced with CA with optimum resource use can improve the food and likely to be associated with soil organic matter build-up and nitrogen nutritional security of the increasing population of Bangladesh. Since accumulation under the fully soil-saturated condition (Sharma and De the R-M system requires higher amount of fertilizer compared to the R-R Datta, 1986). Consistent with these findings, Islam et al. (2019) also system (Supplementary Table 2), it might favor policy towards fertilizer reported higher systems yield under the R-R rotation compared to the subsidies and increase imports by the country. On the other hand, R-W or R-M rotation under CT compared to CA. However, Yadav et al. practicing the R-R system in the prime lands, where winter rice con- (2017) reported higher grain and biomass yields in eastern India for the sumes about 1700 mm ha− 1 water irrigated through groundwater R-R system under no-tillage with 30% residue retention compared to CT sources, would also require high energy for extraction and thus would with full residue incorporation. The rice yield reduction under CA was deplete the underground water. highly dependent on the soil type (texture), rainfall distribution, and land topography. The review of several studies established that heavier 4.2. Effect on crops and cropping systems profitability soils (clay to silty clay loam) had either lower or equal yield (− 6.1 to 2.3%) under no-till (non-puddled) system compared to conventional In intensive cropping systems of the EGP, it is essential to know how puddled system than lighter soils (sandy loam to coarse sand) where rice smallholders can maximize their farm profits with the efficient and yield reduction was about 10%. In the latter soils, the yield penalty was effective use of natural resources (land, water, energy, and labor). This much higher due to higher percolation rate and low nitrogen use effi- study demonstrated the effects of six cropping systems and three tillage ciency because of higher losses (Gathala et al., 2015; Chakraborty et al., options on systems productivity and profitability. The R-M rotation 2017; Chaki et al., 2020). In our study, rice yield under CA had either resulted in a higher gross margin (approximately 150%) than the R-R equal or lower yield penalty (<5%), but this was compensated by higher rotation due to the high yields of long-duration hybrid maize grown in yields of the succeeding dry season crops. This finding is consistent with winter and long duration monsoon rice. Further, maize cultivation Rashid et al. (2019) in northwest Bangladesh. required lower labor and production costs (due to less irrigation water application) compared to boro rice. The higher gross margin of the R-M 4.1.2. Crops and cropping systems grain protein and calorie equivalent system compared to the R-R system is consistent with other studies in the yields EGP (Jat et al., 2014, 2020; Gathala et al., 2016, 2021). A recent on-farm In the recent decade, Bangladesh has consistently shown the sus- study spread over several hundred farmers in three countries of the EGP tainable food production growth and has achieved food self-sufficiency demonstrated that the R-M rotation would be the most profitable or food security in terms of per capita calorie availability (Farukh et al., cropping system for smallholder farmers (Gathala et al., 2021). The 11 M.A. Hoque et al. F i e l d C r o p s R e s e a r c h 291 (2023) 108791 triple cropped R-M-MB rotation resulted in a lower gross margin than Simultaneously, it also promotes straw management behind the the double cropped R-M rotation due to additional production cost combine to distribute evenly in the field (presently all combines accu- required for mung bean cultivation. In addition, many farmers were mulate crop residue in a small strip), which creates ideal conditions for unable to harvest mung bean timely and achieve economic yield due to operating any seeding machines (Yadvinder-Singh et al., 2020; Singh its harvest time synchronizing with the monsoon rainfall. Although in- et al., 2021). It would also be good to encourage or incentivise farmers clusion of mung bean in the R-M rotation lowered the profit due to the willing to manage crop residue in their fields, which will contribute to a use of short-duration rice variety in rotation, such inclusion would be a clean environment, improve soil health, and conserve soil moisture by good option from the soil sustainability point of view as it has potential keeping mulch on the soil surface. to improve soil health (Gathala et al., 2015; Jat et al., 2018; Jat et al., 2019). The double cropped R-W and R-MB also resulted in approxi- 5. Conclusions mately 42% higher gross margin compared to the R-R rotation (Fig. 4). This resulted mainly from lower labor use and production cost of mung Increasing productivity and profitability and the long-term sustain- bean compared to winter rice, and their higher market rice. When the ability of cereal-based cropping systems remains a challenge for R-W system was intensified with the inclusion of mung bean after wheat, ensuring food and nutritional security of low to middle-income rural and it provided USD 247 ha− 1 profit over the R-W system. These results are urban populations in the EGP. This study showed that R-M and R-W-MB in conformity with Kumar et al. (2018), Gathala et al. (2020b), and systems could increase systems REYs and systems grain calorie and Gathala et al. (2021). An additional 16% benefit was achieved when all protein yields by 50%, 5% and 27%, respectively, while providing 133% five alternative cropping systems in the R-R system were layered with higher gross margins than the R-R system. An additional potential CA or AT management practices (Jat et al., 2020; Gathala et al., 2013, benefit could be harnessed when the five cropping systems (except the 2020b; Kumar et al., 2018, 2021). R-R system) studied here are grown using CA-based management The key criteria for selecting cropping systems in agriculture pro- practices. Our findings also indicated that the R-M system could result in duction systems are accessibility to labor and market, low labor wages, a higher gross margin with lower labor use than the R-R system. and higher gross margins (Emran et al., 2021). Labor accessibility in This study demonstrates that the application and adaptation of CA- farming systems is becoming a major challenge due to outmigration of based management practices can be beneficial for rice-based rotations agricultural labor from the rural agrarian communities, seeking alter- practiced on loamy soil of a research farm located in a subtropical native livelihood opportunities (Sugden et al., 2014; Gathala et al., environment of the EGP in south Asia. Although the research farm is 2021) though due to the recent COVID-19 impact, there is some push surrounded by farmers’ fields with similar climate and soil, crop man- backflow of labor in communities (Karim et al., 2020; Gatto et al., 2021; agement practices in research stations would differ from farmers’ fields Talukder et al., 2021). Our study compared and grouped the six crop- with natural conditions and different socio-economic situations. Our ping systems into four groups through scattering quadratic graphics findings offer a basket of technological interventions across the six considering the labor and gross margins obtained (Fig. 5). The analysis cropping systems for smallholders to adopt as either a part of the showed R-M as the best cropping system since it required low labor component of CA or full package of CA (i.e., AT) depending on farmers’ hours but provided high income (low-high group) while R-R and R-MB priority and risk bearing capacity. Long-term studies are recommended were worse systems since they required high number of labor hours and to see the changes in insect pests and weed dynamics; soil biological, provided low income (high-low group). Other four cropping systems fell chemical, and physical properties; and adaptation and mitigation to under low-low (low labor use-low income) and high-high (high labor climate change, etc., in farmers’ fields across soil types and climates. use-high income) groups indicating intermediate in preference. This Particularly, it is important to evaluate and disseminate the promising kind of analysis will be helpful for policy and development leaders cropping systems with CA or AT practices under farmers’ socio- wanting to prioritize the different cropping systems in different economic conditions in Bangladesh and the EGP while considering agro-ecologies while considering accessibility to labor and wages. These farmers’ varying risk-bearing and investment capacities. The latter results may also be helpful in the decision-making process of small considerations are likely to be particularly important where efforts to holder farmers regarding where to invest their available resources effi- extend these systems from research into real-world situations and ciently, i.e., family labor to work in farm, or work off-farm to be able to adoption among smallholder farmers are the development goals in the pay the labor wages and irrigation water sources, etc. EGP and south Asia as a whole. The above distribution patterns may change in future with the adoption of agricultural machines and farm mechanization or labor Declaration of Competing Interest dynamics. The Government of Bangladesh has recently been focusing on promoting agricultural mechanization throughout agricultural opera- The authors declare that they have no known competing financial tions from seed to seed, such as land preparation and planting, inter- interests or personal relationships that could have appeared to influence cultural operations, harvesting and threshing, and processing through a the work reported in this paper. mega subsidy scheme ((National Agricultural Mechanization Policy, 2020) Mechanization in agriculture overcomes several problems and Data availability makes farms more profitable. Simultaneously, this also offers to solve secondary problems such as residue burning after combine harvest, Data will be made available on request. over-tilling and puddling, over-exploitation of underground water, etc. (Shymsundar et al., 2019). As a result, in recent years, mechanized Acknowledgments harvesting has been increasing in Bangladesh. The Government of Bangladesh should put a favorable policy environment for promoting The authors gratefully acknowledge USAID/Bangladesh-funded CA-based sustainable intensification practices to mitigate the crop res- Cereal Systems Initiative for South Asia (CSISA-BD) and the CSISA idue burning problem soon (Shymsundar et al., 2019; Jat et al., 2020; Phase III project supported by USAID-Washington and the Bill and Gathala et al., 2020a). Immediate policy intervention is needed to ban Melinda Gates Foundation (BMGF). Additional support for field research crop residue burning. However, to manage the residues, an urgent need and technical for data processing was provided by the Sustainable and would be to access suitable machinery that could manage the crop res- Resilient Farming Systems Intensification in the Eastern Gangetic Plains idue in situ, like a Happy Seeder machinery developed and used in (SRFSI) project (CSE/2011/077) funded by ACIAR and DFAT, the One northwest India. This machine can plant/sow seed into crop residue of CGIAR Regional Integrated initiative Transforming Agrifood Systems in more than 10 t ha− 1 without any problem in a single pass. South Asia (TAFSSA; _https://www.cgiar.org/initiative/20- 12 M.A. Hoque et al. F i e l d C r o p s R e s e a r c h 291 (2023) 108791 transforming-agrifood-systems-in-south-asia-tafssa/) and the Islam, M.Z., Tiwari, T.P., McDonald, A.J., 2016. Productivity, profitability, and Bangladesh Agricultural Research Institute for supporting the lead au- energetics: a multi-criteria and multi-location assessment of farmers’ tillage and crop establishment options in intensively cultivated environments of South Asia. Field thor’s Ph.D. research.We would like to thank all funders who supported Crops Res 186, 32–46. this research through their contributions to the CGIAR Trust Fund: Gathala, M.K., Laing, A.M., Tiwari, T.P., Timsina, J., Islam, S., Bhattacharya, P.M., https://www.cgiar.org/funders/. We are thankful TP Tiwari for mana- Dhar, T., Ghosh, A., Sinha, A.K., Hossain, S., Hossain, I., Molla, S., Rashid, M., Kumar, S., Kumar, R., Chaudhary, B., Jha, S.K., Ghimire, P., Bastola, B., Chaubey, R., gerial and administrative support and Dr. Asaduzzaman for technical Gerard, B., 2020a. Energy-efficient, sustainable crop production practices benefit advice. The content and opinions expressed in this paper are those of the smallholder farmers and the environment across three countries in the Eastern authors and do not necessarily reflect the views of USAID, CGIAR, Gangetic Plains, South Asia. J. Clean. Prod. 246, 118982 https://doi.org/10.1016/j. BMGF, ACIAR or DFAT and shall not be used for commercial purposes. jclepro.2019.118982. Gathala, M.K., Laing, A.M., Tiwari, T.P., Timsina, J., Islam, S., Chowdhury, A.K., Chattopadhyay, C., Singh, A.K., Batt, B.P., Shrestha, R., Narendra, C.D.B., Gerard, B., Appendix A. Supporting information 2020b. Enabling smallholder farmers to increase productivity while achieving environmental and economic benefits: a meta-analysis of conservation agriculture- based sustainable intensification in the Eastern Gangetic Plains. Renew. Sustain. Supplementary data associated with this article can be found in the Energy Rev. 120, 109645. online version at doi:10.1016/j.fcr.2022.108791. Gathala, M.K., Laing, A.M., Tiwari, T.P., Timsina, J., Rola-Ruzben, F., Islam, S., Maharjan, S., Browm, P.R., Das, K.K., Pradhan, K., Chowdhury, A.K., Kumar, R., Datt, R., Anwar, M., Hossain, S., Kumar, U., Adhikari, S., Magar, D.B.T., Sapkota, B. References K., Shrestha, H.K., Islam, R., Rashid, M., Hossain, I., Hossain, A., Brown, B., Gerard, B., 2021. Improving smallholder farmers’ gross margins and labor-use Akter, S., Gathala, M.K., Timsina, J., Islam, S., Rahman, M., Hassan, M.K., Ghosh, A.K., efficiency across a range of cropping systems in the Eastern Gangetic Plains. World 2021. Adoption of conservation agriculture-based tillage practices in the rice-maize Dev. 138, 105266 https://doi.org/10.1016/j.worlddev.2020.105266. systems in Bangladesh. World Dev. Pers. 21, 100297 https://doi.org/10.1016/j. Gatto, M., Islam, A.H.M.S., 2021. Impacts of COVID-19 on rural livelihoods in wdp.2021.100297. Bangladesh: Evidence using panel data. PLoS ONE 16 (11), e0259264. https://doi. Chakraborty, D., Ladha, J.K., Rana, D.S., Jat, M.L., Gathala, M.K., Yadav, S., Rao, A.N., org/10.1371/journal.pone.0259264. Ramesha, M.S., Raman, A., 2017. A global analysis of alternative tillage and crop Haque, M.E., Bell, R.W., Islam, M.A., Rahman, M.A., 2016. Minimum tillage unpuddled establishment practices for economically and environmentally efficient rice transplanting: An alternative crop establishment strategy for rice in conservation production. Sci. Rep. 7 (1), 9342. https://doi.org/10.1038/s41598-017-09742-9. agriculture cropping systems. Field Crops Res 185, 31–39. Chauhan, B.S., Mahajan, G., Sardana, Timsina, J., Jat, M.L., 2011. Productivity and Hossen, M.A., Hossain, M.M., Haque, M.E., Bell, R.W., 2018. Transplanting into non- sustainability of the rice-wheat cropping system in the Indo-Gangetic Plains of puddled soils with a small-scale mechanical transplanter reduced fuel, labour and Indian sub-continent: problems, opportunities, and strategies. Adv. Agron. 117, irrigation water requirements for rice (Oryza sativa L.) establishment and increased 315–369. yield. Field Crops Res 225, 141–151. Chhetri, A.K., Regmi, P.P., Chanana, N., Aggarwal, P.K., 2020. Potential of climate-smart Islam, S., Gathala, M.K., Tiwari, T.P., Timsina, J., Laing, A.M., Maharjan, S., agriculture in reducing women farmers’ drudgery in high climatic risk areas. Clim. Chowdhury, A.K., Bhattacharya, P.M., Dhar, T., Mitra, B., Kumar, S., Srivastwa, P.K., Change 158, 29–42. https://doi.org/10.1007/s10584-018-2350-8. Dutta, S., Shrestha, R., Manandhar, S., Shrestha, S.R., Paneru, P., Siddique, Nur-E.- Darbas, T., Brown, P.R., Das, K.K., Datt, R., Kumar, R., Pradhan, K., Rola-Rubzen, M.F., Alam, Hossain, A., Islam, R., Ghosh, A.K., Rahman, M.T., Kumar, U., Rao, K.K., 2020. The feminization of agriculture on the Eastern Gangetic Plains: Implications Gerard, B., 2019. Conservation agriculture based sustainable intensification: for rural development. In: Lall, S., Kumar, N., Sinha-Kerkhoff, K. (Eds.), ’Bihar: Increasing yields and water productivity for smallholders of the Eastern Gangetic Crossing Boundaries’. Primus Books, New Delhi, pp. 97–126. Plain. Field Crops Res. 238, 1–17. DeFries, R., Fanzo, J., Remans, R., Palm, C., Wood, S.A., Anderman, T.L., 2105, Metrics Jat, H.S., Datta, A., Sharma, P.C., Kumar, V., Yadav, A.K., Choudhary, M., Choudhary, V., for land-scarce agriculture: Nutrient content must be better integrated into planning. Gathala, M.K., Jat, M.L., Yaduvanshi, N.P.S., Singh, G., McDonald, A., 2018. Science. 349: 238–240. Archives of agron. Soil Sci. 64 (4), 531–545. https://doi.org/10.1080/ Dixon, J., Rola-Rubzen, M.F., Timsina, J., Cummins, J., Tiwari, T.P., 2020. 03650340.2017.1359415. Socioeconomic impacts of Conservation Agriculture based Sustainable Jat, M.L., Gathala, M.K., Saharawat, Y.S., Ladha, J.K., Singh, Y., 2019b. Conservation Intensification (CASI) with particular reference to South Asia. In: Dang, Y., Dalal, R., agriculture in intensive rice-wheat rotation of Western Indo-Gangetic Plains: Effect Menzies, N.W. (Eds.), No-till Farming Systems for Sustainable Agriculture: on crop physiology, yield, water productivity and economic profitability. Int. J. Challenges and Opportunities. Springer. Environ. Sci. Nat. Res 18 (3), 555988. https://doi.org/10.19080/ Emran, S.-A., Krupnik, T.J., Aravindakshan, S., Kumar, V., Pittelkow, C.M., 2021. Factors IJESNR.2019.18.555988. contributing to farm-level productivity and household income generation in coastal Jat, M.L., Chakraborty, D., Ladha, J.K., Rana, D.S., Gathala, M.K., McDonald, A., Bangladesh’s rice-based farming systems. PLoS ONE 16 (9), e0256694. https://doi. Gerard, B., 2020. Conservation agriculture for sustainable intensification in South org/10.1371/journal.pone.0256694. Asia. Nat. Sustain 3, 336–347. https://doi.org/10.1038/s41893-020-0500-2. Erenstein, O., Laxmi, V., 2008. Zero tillage impacts in India’s rice–wheat systems: a Jat, R.K., Sapkota, T.B., Singh, R.G., Jat, M.L., Kumar, M., Gupta, R.K., 2014. Seven years review. Soil Res 100, 1–14. of conservation agriculture in a rice–wheat rotation of Eastern Gangetic Plains of FAO, 2011, Save and Grow: A Policymaker’s Guide to the Sustainable Intensification of South Asia: yield trends and economic profitability. Field Crops Res 164, 199–210. Smallholder Crop Production. United Nations Food and Agriculture Organization. Karim, M.R., Islam, M.T., Byomkesh, T., 2020. COVID-19′s impacts on migrant workers Rome, 102 p. from Bangladesh: In search of policy intervention. World Dev. 136, 105123 https:// FAO, 2018, The Linkages between Migration, Agriculture, Food Security and Rural doi.org/10.1016/j.worlddev.2020.105123. Development. Rome. 80pp. (http://www.fao.org/3/CA0922EN/CA0922EN.pdf). Keil, A., D’Souza, A., McDonald, A., 2017. Zero-tillage is a proven technology for Licence: CC BY-NC-SA 3.0 IGO. sustainable wheat intensification in the eastern indo-gangetic plains: what FAO, I.F.A.D., UNICEF, W.F.P., WHO, 2020, The State of Food Security and Nutrition in determines farmer awareness and adoption? Food Sec 9, 723–743. the World 2020. Transforming food systems for affordable healthy diets. Rome, FAO. Krupnik, T.J., Yasmin, S., Shahjahan, Md, McDonald, A.J., Hossain, K., Baksh, E., https://doi.org/10.4060/ca9692en. Hossain, F., Kurishi, A.S.M.A., Miah, A.A., Mamun, Md.A., Rahman, B.M.S., Farukh, M.U., Bashir, M.K., Hassan, S., Adil, S.A., Kragt, M.E., 2020. Mapping the food Gathala, M.K., 2014. Productivity and Farmers’ Perceptions of Rice-Maize System security studies in India, Pakistan and Bangladesh: Review of research priorities and Performance Under Conservation Agriculture, Mixed and Full Tillage, and Farmers’ gaps. Glob. Food Secur. 26, 100370. Practices in Rainfed and Water-Limited Environments of Southern Bangladesh. 6th FRG, 2018. Fertilizer Recommendation Guide-2018. Bangladesh Agricultural Research World Congress on Conservation Agriculture. Winnipeg, Canada. June 24. Council (BARC). In: Farmgate, Dhaka, 1215, p. 233. 〈http://www.bfa-fertilizer.or Krupnik, T.J., Ahmed, Z.U., Timsina, J., Shahjahan, M., Alanuzzaman, A.S.M., g/wp-content/uploads/2019/09/Fertilizer-Recommendation-Guide-2018-English. Azahar, K., Miah, A.B.M., Rahman, S., Gathala, M.K., McDonald, A., 2015. Forgoing pdf〉. the fallow in Bangladesh’s stress-prone coastal deltaic environments: Effect of Gathala, M.K., Ladha, J.K., Kumar, V., Saharawat, Y.S., Kumar, V., Sharma, P.K., 2011. sowing date, nitrogen, and genotype on wheat yield in farmers’ fields. Field Crops Effect of tillage and crop establishment methods on physical properties of a medium- Res 170, 7–20. textured soil under a seven-year rice− wheat rotation. Soil Sci. Soc. Am. J. 75, Kumar, V., Jat, H.S., Sharma, P.C., Balwinder-Singh, Gathala, M.K., Malik, R.K., 1851–1862. Kamboj, B.R., Yadav, A.K., Ladha, J.K., Raman, A., Sharma, D.K., McDonald, A., Gathala, M.K., Kumar, V., ShaR-Ma, P.C., Saharawat, Y.S., Jat, H.S., Singh, M., 2018. Can productivity and profitability be enhanced in intensively managed cereal Kumar, A., Jat, M.L., Humphreys, E., Sharma, D.K., Sharma, S., Ladha, J.K., 2013. systems while reducing the environmental footprint of production? Assessing Optimizing intensive cereal-based cropping systems addressing current and future sustainable intensification options in the breadbasket of India. Agric. Ecosys Environ. drivers of agricultural change in the north-western Indo-Gangetic Plains of India. 252, 132–147. https://doi.org/10.1016/j.agee.2017.10.006. Agric. Ecosyst. Environ. 177, 85–97. Kumar, V., Gathala, M.K., Saharawat, Y.S., Parihar, C.M., et al., 2019. Impact of tillage Gathala, M.K., Timsina, J., Islam, M.S., Rashid, M.H., Ghosh, A.K., Krupnik, T.J., and crop establishment methods on crop yields, profitability and soil physical Tiwari, T.P., McDonald, A., 2015. Conservation agriculture-based tillage and crop properties in riceewheat system of Indo- Gangetic Plains of India. Soil Use Manag 35, establishment options can maintain farmers’ yield and increase profits in South 303–313. https://doi.org/10.1111/sum.12473. Asia’s rice-maize systems: Evidence from Bangladesh. Field Crops Res 172, 85–98. Ladha, J.K., Rao, A.N., Raman, A., Padre, A.T., Dobermann, A., Gathala, M.K., et al., Gathala, M.K., Timsina, J., Md, Islam, S., Krupnik, T.J., Bose, T.R., Islam, N., Rahman, M. 2015. Agronomic improvements can make future cereal systems in South Asia far M., Hossain, M.I., Harun-AR-Rashid, Ghosh, A.K., Hasan, M.M.K., Khayer, M.A., 13 M.A. Hoque et al. F i e l d C r o p s R e s e a r c h 291 (2023) 108791 more productive and result in a lower environmental footprint. Glob. Change Biol. Sinha, A.K., Ghosh, A., Dhar, T., Bhattacharya, P.M., Mitra, B., Rakesh, S., Paneru, P., 01, e21. https://doi.org/10.1111/gcb.13143,1354e1013. Shrestha, S.R., Manandhar, S., Beura, K., Dutta, S., Pradhan, A.K., Rao, K.K., National Agricultural Mechanization Policy, 2020, Ministry of Agriculture, Government Hossain, A., Siddquie, H.N., Molla, M.S.H., Chaki, A.K., Gathala, M.K., Islam, M.S., of the People’s of Bangladesh. Pp 1–15. Dalal, R.C., Gaydon, D.S., Laing, A.M., Menzies, N.W., 2019. Trends in key soil National Food and Nutrition Security Policy, 2020, Ministry of Food, Government of parameters under conservation agriculture-based sustainable intensification farming Bangladesh. NFNSP https://fpmu.gov.bd/agridrupal/policy-and-planning- practices in the Eastern Ganga Alluvial Plains. In: Soil Res, 57. CSIRO Publishing,, frameworks. pp. 883–893. https://doi.org/10.1071/SR19162. Onofri, A., Seddaiu, G., Piepho, Hans-Peter, 2016. Long-term experiments with cropping Snedecor, G.W., Cochran, W.G., 1989, Statistical Methods, eighth edition. Iowa State systems: Case studies on data analysis. Eur. J. Agron. 77, 223–235. https://doi.org/ University Press. 10.1016/j.eja.2016.02.005. Sugden, F., Maskey, N., Clement, F., Ramesh, V., Philip, A., Rai, A., 2014. Agrarian stress Pretty, J.N., Benton, T.G., Bharucha, Z.P., Dicks, L.V., Flora, C.B., Godfray, H.C.J., et al., and climate change in the Eastern Gangetic Plains: Gendered vulnerability in a 2018. Global assessment of agricultural system redesign for sustainable stratified social formation. Glob. Environ. Change 29, 258–269. intensification. Nat. Sustain 1, 441–446. Talukder, B., vanLoon, G.W., Hipel, K.W., Orbinski, J., 2021. COVID-19′s implications on Rashid, M.H., Timsina, J., Islam, N., 2019. Effect of tillage and residue management on agri-food systems and human health in Bangladesh. Curr. Res. Environ. Sustain. 3, productivity, profitability and soil Fertility for the irrigated rice-maize-mung bean 100033 https://doi.org/10.1016/j.crsust.2021.100033. systems of South Asia. J. Crop Improv 33, 1–28. https://doi.org/10.1080/ Timsina, J., Jat, M.L., Majumdar, K., 2010. Rice-maize systems of South Asia: current 15427528.2019.1661056. status, future prospects and research priorities for nutrient management. Plant Soil Shaheen, N., Rahim, A.T.M., Mohiduzzaman, M., Banu, C.P., Bari, M.L., Tukun, A.B., 335, 65–82. Mannan, M., Bhattacharjee, L., Stadlmayr, B., 2013. Food Composition Table for Timsina, J., Singh, V.K., Majumdar, K., 2013. Potassium management in rice–maize Bangladesh (First edition). Institute of Nutrition and Food Science. Centre for systems in South Asia. J. Plant Nutr. Soil Sci. 176, 317–330. Advanced Research in Sciences, University of Dhaka,, Dhaka-1000, Bangladesh, Timsina, J., Wolf, J., Guilpart, N., van Bussel, L., Grassini, P., van Wart, J., Hossain, A., p. 201. Rashid, H., Islam, S., van Ittersum, M., 2018. Can Bangladesh produce enough Shapiro, S.S., Wilk, M.B., 1965. An analysis of variance test for normality (complete cereals to meet future demand? Agric. Syst. 163, 36–44. samples). Biometrika 52 (3/4), 591–611. Tiongco, M., Hossain, M., 2015. Adoption of modern varieties and rice varietal diversity Sharma, P.K., De Datta, S.K., 1986. Physical properties and processes of puddle rice soils. on household farms in Bangladesh. Harvest. Work. Pap. Number 22. 〈https://www. Adv. Soil Sci. 5, 139–178. harvestplus.org/sites/default/files/HarvestPlus_WorkingPaper22.pdf〉. Shymsundar, P., Springer, N.P., Tallis, H., Polasky, S., Jat, M.L., Sidhu, H.S., Verhulst, N., Govaerts, B., Verachtert, E., Castellanos-Navarrete, A., Mezzalama, M., Krishnapriya, P.P., Skiba, N., Ginn, W., Ahuja, V., Cummins, J., Datta, I., Wall, P., Deckers, J., Sayre, K.D., 2010. Conservation Agriculture, improving soil Dholakia, H.H., Dixon, J., Gerard, B., Gupta, R., Hellman, J., Jadhav, A., Jat, H.S., quality for sustainable production systems? In: Lal, R., Stewart, B.A. (Eds.), Advances Keil, A., Ladha, J.K., Lopez-Ridaura, S., Nandrajog, S.P., Paul, S., Ritter, A., in Soil Science: Food Security and Soil Quality. CRC Press, Boca Raton, FL, USA, Sharma, P.C., Singh, R., Singh, D., Somanathan, R., 2019. Field on fire: Alternatives pp. 137–208. to crop residue burning in India. Science 365, 536–538. https://doi.org/10.1126/ Yadav, G.S., Datta, R., Pathan, S.I., Lal, R., Meena, R.S., Babu, S., Das, A., Bhowmik, S.N., science.aaw4085. Datta, M., Saha, P., Mishra, P.K., 2017. Effects of conservation tillage and nutrient Singh, H.S., Singh, M., Lohan, S.K., Jat, R.K., Jat, M.L., 2021. Conservation agriculture management practices on soil fertility and productivity of rice (Oryza sativa L.)–rice machinery: Research advances and contributions towards “Make in India”. J. Agric. system in North Eastern Region of India. Sustainability 9, 1816. https://doi.org/ Phys. 21, 259–273. 10.3390/su9101816. Singh, V.K., Yadvinder-Singh, Dwivedi, B.S., Singh, Majumdar, K., Jat, M.L., Mishra, R. Yadvinder-Singh, Sidhu, H.S., Singh, Manpreet, Jat, H.S., Chhokar, R.S., Setia, R., Jat, M. P., Rani, M., 2016. Soil physical properties, yield trends and economics after five L., 2020. Conservation Agriculture and Scale of Appropriate Agricultural years of conservation agriculture-based rice-maize system in north-western India. Mechanization in Smallholder Systems. Manual, pp 180. Borlaug Institute for South Soil . Res. 155, 133–148. Asia (BISA). Int. Maize Wheat Improv. Cent. (CIMMYT). 14