The study was conducted on the Research Field of Bangladesh Agricultural University (BAU), Mymensingh (24˚43.407'N, 90˚26.22'E), Bangladesh. The region is under subtropical monsoon climate with a mean annual temperature of 26˚C, average annual rainfall of 1800 mm, and relative humidity of field belongs to a Non-calcareous Dark Grey Floodplain soil under the agro-ecological zone 9 (AEZ-9; Old Brahmaputra Floodplain soil) [16] and classified in the U.S. Soil Taxonomy as Aeric Haplaquept, moderately drained with silt loam texture and near neutral pH (6 to 6.5).

The experimental field has been managed under a conventionally ploughed system for many years with annual rice-rice sequence. The experiment was commenced with two soil disturbance levels (strip tillage, ST vs. conventional tillage, CT) and three N fertilizer rates: 60%, 100% and 140% of recommended N fertilizer (optimum dose, 100 kg N∙ha⁻¹) as urea under annual wheat (Triticum aestivum)-mung bean (Vigna radiata)-rice (Oryza sativa L) cropping sequence. The ST has been performed by a versatile multi-crop transplanter (VMP) which makes a 3 cm furrow after each 20 cm unploughed strip. In ST system, 30% residue (by height) of the previous crops for wheat and rice and 100% for mung bean were left on soil surface while in CT system same amount of residue was incorporated into soil by repeated ploughing using a power tiller. A split-plot experiment was established with three replications for each treatment combination. Tillage treatment was assigned to main plots and fertilizer to sub-plots. The size of each plot was 7 m × 7 m with a 50 cm buffer between plots. Total number of plots were (tillage x N rate x replications = 2 × 3 × 3) = 18.

Nitrogen fertilizer (urea) was applied in two equal splits at 7 and 35 days after sowing for wheat, as a single starter application for mung bean, and as three equal splits at 10, 30 and 50 days after transplanting for rice. Other nutrients, namely P (triple super phosphate), K (muriate of potash), S (gypsum), Zn (ZnSO₄∙H₂O), and B (H₃BO₃), were applied at 20 kg P, 60 kg K, 10 kg S, 2 kg Zn, and 1.5 kg B∙ha⁻¹ for wheat; 20 kg P, 30 kg K, and 10 kg S∙ha⁻¹ for mung bean; and 10 kg P, 30 kg K, 10 kg S, and 2 kg Zn∙ha⁻¹ for rice. Herbicides: Glyphosate (Roundup^®), a nonselective herbicide, was sprayed over the field at a rate of 1.85 kg∙ha⁻¹ at 2 to 3 days before the transplanting of rice seedlings or the planting of wheat and mung bean. In addition, Pretilachlor (Superhit^®, post emergence herbicide) was used at a rate of 450 g∙ha⁻¹ at 5 to 7 days after transplanting rice seedlings in ST. Insecticides: Brifar 5G was applied 50 days after planting of wheat, Diazinon was sprayed three times for mung bean (36, 48, and 59 days after planting), and Brifer 5G and Cidial 5G were applied to control rice insects. Irrigation was provided twice for wheat, once during the crown root initiation and again before the flowering stage, depending on soil moisture and crop water requirement. The rice fields were irrigated a day before the final land preparation and as and when necessary during the growing stage.

The field measurement of CH₄ and CO₂ was conducted during December to March after seven consecutive years of the same treatment application with the same cropping sequence. The CH₄ and CO₂ fluxes were measured using the static chamber method [17] beginning from the first irrigation water supply followed by the first split application of urea to until maturity of wheat. The soda glass chambers wrapped with a reflecting paper (40 cm × 40 cm wide and 40 cm high) and with stainless steel collars were inserted into the centre of each plot (7 m × 3.5 m) to a depth of 10 cm. Each collar had a neoprene seal which ensured an air-tight seal between the chamber lid and footing. At each sampling event lids were installed on the collars and gas samples were collected through the air tight rubber septa using a 20 ml polypropylene syringe equipped with a 25-gauge Luer lock needle at 30 min interval: at time 0, 30 min and 60 min. A 16 ml sample was collected from the headspace and injected into a pre-evacuated 12 ml vial (Labco Wycom Ltd.). Gas samples were collected from day 0 to 15 after each time urea application; on day 0, 1, 3, 5, 7, 10 and 15 and once a week until tillering stage of wheat.

Composite soil samples were collected adjacent to each gas sampling chamber using an auger and a core sampler to 0 - 15 cm depth and stored in sealable plastic bags in a cooler box. A portion of field-moist soil was processed after sieving through a 2 mm mesh to remove visible organic residues and analyzed for MBC and associated biological properties. The other portion of the field-moist soil was air-dried under shade at room temperature (~25˚C) for two weeks and processed (2 mm sieved) to analyze for selected physical and chemical properties. Soil bulk density was measured using core sampler (a metal ring with 5 cm inner diameter and 5 cm height) method [18]. Soil moisture content was measured by gravimetric method [19]. Soil organic carbon (SOC) was measured by wet oxidation method [20]. Soil total microbial biomass C (MBC) was determined by the CHCl₃ fumigation-extraction method [21] following:

MBC (mg/kg) = EC/kEC (1)

where EC is the amount of extracted organic C from CHCl₃ fumigated soil minus organic C extracted from non-fumigated soil, and kEC is the extraction efficiency (0.45).

Sample vials were sent to Teagasc Environment Research Center, Johnstown Castle, Co. Wexford, Ireland for CH₄ analysis. The gases were analyzed in a Varian 3800 gas chromatograph (CP-3800, Varian, Inc., Switzerland) equipped with electron capture detector using Argon (Ar) as the carrier gas. Gas flux was calculated from the change in headspace concentration over measured period using linearity: at time 0, 30 and 60 min. The Cumulative gas emissions were calculated by summing up all daily fluxes for the whole experimental period (35 days) and assuming that, where no daily data were available, the daily fluxes changed linearly.

where dGas in ppb to get concentration change over time; 10^x is recalculation, here 10⁻⁹V_chamber is volume of the chamber used; p is atmospheric pressure in Pa (100 is to convert Pa to hPa); MW is molecular weight of C; R is gas constant 8. 314 J∙mol⁻¹∙K⁻¹; T is temperature in Kelvin; 10^y is recalculation here 10⁶ (µg); A is area of the chamber.

A two-way analysis of variance (ANOVA) was performed using tillage and N rate as fixed variables and block as a random variable. The distribution of data for normality was checked before ANOVA. Data were statistically analyzed to ascertain the significant differences in main and interactions among tillage and C rate treatments. Post-hoc test was performed to separate differences between C rates using Tukey-Kamers multiple comparison Test. All statistical analyses were considered significant at p ≤ 0.05, unless otherwise mentioned. Pearson correlation analysis was performed to verify relationship between soil biophysicochemical properties and CH₄ and CO₂ fluxes. All the statistical analyses were performed on Statistix 10.

The highest CH₄ and CO₂ emission peak was observed on day 3 after urea application in both tillage systems except CO₂ flux in CT. Nitrogen fertilizer application rate significantly influenced mean and cumulative CH₄ and CO₂ fluxes. The CH₄ and CO₂ fluxes decreased in optimum dose of N fertilizer except ST for CH₄. The CO₂ emission significantly showed higher emission at minimum (60% of RD) fertilizer application in both tillage systems. Over the sampling period, the CH₄ emissions peak was observed on day 3 after the urea application in both split applications in either tillage, except 140RD in the second split in CT for CO₂ which showed the highest peak on day 2 (Figure 1(d)).

In the first split, the highest peak in CT for CH₄ was much higher than in ST, being 22.32 g C∙m⁻²∙d⁻¹ in 140RD in ST and 33.98 g C∙m⁻²∙d⁻¹ in 60RD in CT. In the second split, the CH₄ flux was little lower than in the first split (Figure 1(a)). Clearly, CH₄ emission peak in CT appeared a bit later and went down to the background level earlier than the ST which made the emission peak narrower than the ST. In both tillage systems, peak went down to the ground level on day 15 after the urea application. The peak in CT for CH₄ increased with the rate of N application being the highest in 140% RD and the lowest in 60% RD whereas in ST decreased with the rate of N application being the highest in 60RD and the lowest in 100RD (Figure 1(a), Figure 1(b)). Moreover, the peak in CT for CO₂ was unstable over the sampling period for both tillage systems (Figure 1(c), Figure 1(d)).

Figure 1. CH₄ flux (mean ± SE; n = 3) in three N application rates over time, day after urea application, in conventional (CT) and strip (ST) tillage systems; arrow shows the day of urea application.

Tillage had non-significant effect on both CO₂ and CH₄ emission. However, N levels showed significant variation on carbon dioxide and methane concentration. Among the interactions only N levels with tillage showed significant variation on carbon dioxide and methane. All other interactions were non-significant. The cumulative CO₂ emission significantly higher (606.52 kg C∙ha⁻¹) at 60RD level with ST which was statistically similar (585.95 kg C∙ha⁻¹) to CT with 60RD and 140RD (571.69 kg C∙ha⁻¹). In addition, for CH₄ emission we found that, higher GHG emissions originated mainly from the higher application rate of the N-fertilizer and 140RD observed significantly increase at ST which was similar to CT with 100RD and 140RD (Figure 2Figures 2(a)-(d)). Mean CH₄ emission significantly increased with increasing nitrogen doses. However, CO₂ emission decreased significantly in optimum dose of nitrogen fertilization (Figure 3Figure 3).

Figure 2. Cumulative emission of CH₄ and CO₂ flux (mean ± SE; n = 3) in three N application rates over time, day after urea application, in conventional (CT) and strip (ST) tillage systems.

Figure 3.Mean CH₄ and CO₂ emissions in three N applications rate under conventional (CT) and strip (ST) tillage systems; small letters show differences between N rates within a tillage treatment and capital letters show differences between two tillage treatments.

The measured soil biophysicochemical properties were significantly influenced by tillage and N fertilizer rates (Table 1) which is the effect of seven consecutive year’s continuous application of the same treatment with the same cropping sequence. The CT had significantly higher BD (p < 0.05) but lower moisture than ST. The ST significantly increased SOC, POXC, and MBC showing the mean values of 1.24%, 689.62 and 241.49 mg∙kg⁻¹ and 0.14% in ST and 1.01%, 548.14 and 185.2 mg∙kg⁻¹ in CT, respectively (Table 1). Considering the N rate, SOC and POC were significantly lower in 60% RD than in 100 (p < 0.01) and 140% RD (p < 0.01) where the latter two were similar to each other (p > 0.05). The MBC was significantly higher in 100% RD than in 60% RD (p < 0.05) and 140% (p < 0.05) where the later two were similar to each other (p > 0.05). Interaction effects of tillage and N rate were significant (p < 0.01) showing higher MBC in 100% RD in ST. The TN was significantly higher only in 140% RD than the other two N rates. No significant effects were found for soil moisture and basal respiration. The significant interactions between N and tillage were found only for MBC.

Table 1. Tillage and N rate effects on soil physicochemical properties (soil organic carbon (SOC), microbial biomass carbon (MBC), particulate organic carbon (POC), permanganate oxidisable carbon (POXC), basal respiration (BR), bulk density (BD), and soil moisture).

*p < 0.05; **p < 0.01; NS = non-significant.

Interrelation among CO₂ and CH₄ flux and all soil related controlling factors with their levels of significance are shown in Table 2. The CO₂ showed medium to high correlations with all the soil properties under study. There was significant negative correlation between CO₂ with soil labile carbons and microbial properties like POC, OC, POXC, BR, MBC and soil moisture. Significant negative correlations were recorded from CO₂ with soil physical property like BD (r² = 0.58). The correlation between CO₂ and other soil properties and the carbon pools ranged from 0.19 to 0.58. The lowest correlation was recorded between CO₂ and moisture (r² = 0.19). Considering CH₄ flux, we recorded all significant positive correlation with all the soil labile carbons and significant negative correlations with MBC, BR and BD. The positive correlation between CH₄ and the soil carbon pools ranged from

Table 2. Pearson correlation among CH₄, CO₂ fluxes and analyzed soil properties; soil organic carbon (SOC), microbial biomass carbon (MBC), particulate organic carbon (POC), permanganate oxidizeable carbon (POXC), basal respiration (BR), bulk density (BD) and soil moisture.

*p < 0.05; **p < 0.01; ***p < 0.001.

0.19 to 0.62. The strong positive correlation was found between CH₄ and POC (r² = 0.62) while the lowest correlation was recorded between CH₄ and BD (r² = 0.19).

In our study, there was no significant differences found between two tillages for both CH₄ and CO₂ emission. May be this is due to there were no differences regarding the size and activity of the methanotrophic community between tillage treatments as these bacteria affect CO₂ and CH₄ emission [22]. Tillage also affects gaseous diffusivity and the rate of supply of atmospheric CH₄ [23]. By contrast, CT improves macro porosity and maintains its continuity [24]. Carbon dioxide (CO₂) emissions caused by ST may be similar or slightly higher than those caused by CT if entire growing and fallow seasons are considered [25]. Reduced extracellular enzyme activities and fungal populations resulting from N fertilizer application resulted in decreased soil CO₂ emissions, our result also in line with this [26][27]. The increased levels of surface crop residues in NT probably serve as a barrier for CO₂ emissions from soil, decreasing the decomposition of crop residues because of reduced soil temperature and minimum soil-residue contact [28]. The inconsistent tillage effects on soil CO₂ fluxes suggest that tillage is not the only factor affecting CO₂ flux and that other factors are also involved. The differences between the soil CO₂ emissions under the tillage treatments may have been caused by variation in soil C mineralization.

The application rate of the fertilizer had the biggest impact on GHG emissions. Application of N fertilizer in our study increased CH₄ emissions from wheat field because of the promotion of wheat plant growth, providing additional C sources and emission pathways [29]. However, no significant effect of N fertilizer application on CH₄ emissions [30][31]. On the other hand the application of urea significantly decreased CH₄ emissions [32]. Results varied among studies because of the differences in soil texture or climate. These findings show that further study is needed to understand the functioning of these complex and dynamic systems., the annual mean fluxes of soil CH₄ uptake decreased with the amount of added N, with significant differences between different treatments (p < 0.05). Application of N fertilizer increases plant biomass production, stimulating soil biological activity, and consequently, CO₂ emission [33]. In addition, we observed no significant effect of N fertilizer application on cumulative CO₂ emissions, consistent with the results [34]. This finding may be due to the fact that CO₂ is reduced to CH₄ under anaerobic conditions, thus leading to significant differences in CH₄ emissions rather than in CO₂ emissions.

The results revealed that bulk density had significant negative effect on soil CO₂ flux. This relationship may be related to soil with lower bulk density levels having a greater ability to exchange air with the atmosphere due to its higher sand fraction. This study revealed that soil organic carbon content in bulk soils had a significant negative linear relationship with CO₂ flux. This indicated that abiotic and physical properties have an important role to play in soil gas exchange. This finding agrees with [35][36]. The bulk density exhibited a negative relationship with CO₂ emission [35]. There was significant negative correlation between CO₂ with soil labile carbons and microbial properties like POC, OC, POXC, BR, MBC and soil moisture. Significant positive correlations were recorded from CO₂ with soil physical property like BD. In this study, urea fertilizer was applied immediately after irrigation water supply to the dry soil. In addition, the significant and positive correlation between CO₂ flux and soil BD implies that higher BD increases soil anaerobiocity which decrease CO₂ further to C, consequently decreasing the CO₂ flux. Higher bulk density could also create barrier to diffuse up CO₂ to the surface with reduction of CO₂ to C. The CH₄ production may occur wherever organic matter is degraded under anaerobic conditions. It is generally accepted that CH₄ production, which is part of total C mineralization, is influenced by both the quality and quantity of organic matter present [23]. However, the relationship between CH₄ production and total C mineralization is highly variable. Soil organic matter is the main substrate for methanogens [23], while others believe that recent plant residues or fresh plant materials (either the current or the previous year’s production) are the main substrates [37].