Evaluating the Effects of Fertilizer Deep Placement on Greenhouse Gas Emissions and Nutrient Use Efficiency in Wet Direct-Seeded Rice During the Wet Season in Assam, India

Abstract

Mitigation of greenhouse gases (GHGs), improving nutrient-use efficiency (NUE), and maximizing yield in rainfed lowland rice cultivation poses significant challenges. To address this, a two-year field experiment (2020 and 2021) was conducted in Assam, India, to examine the impact of different fertilizer-management practices on grain yield, NUE, and GHGs in wet direct-seeded rice (Wet-DSR) during the kharif season. The experiment included eight treatments: control; farmer’s practice (30-18.4-36 kg N-P₂O₅-K₂O ha⁻¹); state recommended dose of fertilizer (RDF) @ 60-20-40 kg N-P₂O₅-K₂O ha⁻¹ with N applied in three splits @ 30-15-15 kg ha⁻¹ as basal, at active tillering (AT), and panicle initiation (PI); best fertilizer management practices (BMPs): 60-20-40 kg N-P₂O₅-K₂O ha⁻¹ with N applied in three equal splits as basal, at AT, and PI; and fertilizer deep placement (FDP) of 120%, 100%, 80%, and 60% N combined with 100% PK of RDF. The experiment was arranged out in a randomized complete block design with three replications for each treatment. The highest grain yield (4933 kg ha⁻¹) and straw yield (6520 kg ha⁻¹) were achieved with the deep placement of 120% N + 100% PK of RDF. FDP with 80% N + 100% PK reduced 38% N₂O emissions compared to AAU’s RDF and BMPs, where fertilizer was broadcasted. This is mainly due to the lower dose of nitrogen fertilizer and the application of fertilizer in a reduced zone of soil. When considering both productivity and environmental impact, applying 80% N with 100% PK through FDP was identified as the most effective practice.

1. Introduction

Climate change is considered one of the most crucial environmental issues of the 21st century, with a significant impact on a country’s food security and economy, especially in developing nations like India. Agriculture is both a victim and a contributor to climate change [1]. Rice, which is India’s most essential staple food crop, is expected to suffer significant yield reductions due to climate change [2]. Some estimates indicate that climate change may cause a loss of more than a quarter of rice production in most of India’s river basins [3]. In India, the agriculture sector is a major source of CH₄ and N₂O emissions. CH₄ emissions primarily come from rice cultivation and livestock rearing, while N₂O emissions mainly result from the application of nitrogenous fertilizers to agricultural soils. In 2016, the agriculture sector’s CO₂e emissions amounted to 407,821 Gg, accounting for roughly 14% of India’s total emissions. In the agricultural sector, rice cultivation contributed to 17.5% of GHG emissions in 2016, followed by fertilizer use on agricultural soils at 19.1% [4].

In Assam, rice cultivation occupies 2.35 million hectares out of the total 3.89 million hectares of cropped land, accounting for 93% of the total food grain output in the region [5]. Rice is grown in three different seasons, namely Ahu (autumn), kharif or Sali (monsoon), and Boro (summer), across various ecosystems such as deep water, rainfed, and irrigated conditions. The Sali or wet season is the most prominent rice season, receiving the highest level of precipitation (65%) among the three seasons. However, rice productivity during the wet (Sali) season is comparatively lower at 2055 kg ha⁻¹, compared to 2865 kg ha⁻¹ in the summer (Boro) season [6]. The main limiting factor in this season is the low nitrogen (N) use efficiency. N fertilizer, mainly urea, is typically broadcasted onto the soil surface as part of the conventional rice cultivation practice. Unfortunately, a significant portion of the applied fertilizer is lost due to various processes such as volatilization, denitrification, run-off, and leaching. Consequently, the efficiency of fertilizer use, particularly N-fertilizer use efficiency, is very low (30–35%) in rice cultivation [7]. To reduce N₂O emissions, improving N-use efficiency through appropriate crop management practices such as split application and fertilizer deep placement is crucial. FDP of urea or multi-nutrient (N, P, and K) briquettes is considered one of the most effective nutrient-management practices for enhancing these efficiencies in lowland rice fields.

FDP of urea with larger particle sizes enhances its use efficiency and provides environmental benefits by reducing runoff and N-volatilization losses. In urea deep placement (UDP), urea briquettes are placed approximately 7–10 cm below the soil surface, either manually or with a handheld applicator. This prolongs the availability of nitrogen to the plants compared to the conventional practice of urea application, resulting in a significant increase in nitrogen uptake and grain yield. Deep placement of NPK fertilizers offers additional advantages over UDP, including the simultaneous provision of all primary nutrients, elimination of extra labor required for broadcast application, and the incorporation of separate phosphorus and potassium during land preparation. Moreover, deep placement of NPK fertilizers reduces the loss of phosphorus and potassium through surface runoff, particularly during the wet season, thereby decreasing eutrophication in water bodies. As a result, deep placement of NPK fertilizer may increase nitrogen use efficiency and decrease environmental impacts [8]. Accurately estimating these emissions from paddy fields is crucial for the greenhouse gas (GHG) inventory and its implications on a national or regional level. With the aforementioned background, the present experiment was designed to investigate the effects of FDP on nitrogen use efficiency and GHG mitigation in wet direct-seeded rice during the wet season (Sali) in Assam.

2. Materials and Methods

2.1. Experiment Site

A 2-year research study was conducted at Assam Agricultural University, Jorhat, Assam (India) during the wet seasons of 2020–2021. The experiment site was situated at 26°43′ N latitude, 94°12′ E longitude, and an altitude of 86.6 m above mean sea level in the Upper Brahmaputra Valley Zone (UBVZ) of Assam. The total rainfall received during the crop growth period in the first year was 1098.8 mm, while the corresponding value for the second year (2021) was 650.8 mm. The mean weekly maximum and minimum temperatures during 2020 ranged from 23.90 to 34.20 °C and 8.20 to 26.40 °C, respectively. In 2021, the corresponding temperatures ranged from 24.39 to 35.59 °C and 8.48 to 25.83 °C, respectively (Figure 1). The soil at the site was sandy loam, characterized by an acidic reaction with a pH value of 5.5. It was medium in terms of available nitrogen (280.1 kg ha⁻¹) and available potassium (140.2 kg ha⁻¹), and low in terms of available phosphorus (21.3 kg ha⁻¹). The test variety used in the study was Ranjit-Sub1, a long-duration rice variety. Sowing was conducted manually at a seed rate of 60 kg/ha, with a row spacing of 20 cm.

2.2. Treatment Details and Observations

The experiment was conducted using a randomized block design with three replications and eight treatment combinations. The treatments were as follows: T₁: control (no fertilizer); T₂: farmers’ practice (30-18.4-36 kg N-P₂O₅-K₂O ha⁻¹); T₃: state recommended dose of fertilizer (RDF) @ 60-20-40 kg N-P₂O₅-K₂O ha⁻¹ with N in three splits @ 30-15-15 kg ha⁻¹ as basal, at active tillering (AT), and panicle initiation (PI); T₄: best-fertilizer management practices (BMPs): 60-20-40 kg N-P₂O₅-K₂O ha⁻¹ with N in three equal splits as basal, at AT, and PI; T₅: fertilizer deep placement (FDP) with 120% N + 100% PK of RDF; T₆: FDP with 100% NPK of RDF; T₇: FDP with 80% N + 100% PK of RDF; and T₈: FDP with 60% N + 100% PK of RDF. Synthetic fertilizers urea, diammonium phosphate (DAP), and a muriate of potash (MOP) were used to apply the nutrients according to the treatment requirements. In the farmers’ practice treatment (30-18.4-36 kg N-P₂O₅-K₂O ha⁻¹), half of the N dose, full dose of P₂O₅, and K₂O were applied as basal, and the remaining half of the N dose was applied at the active-tillering stage. In the FDP treatments, the fertilizers were applied as briquettes at a depth of 7–10 cm, while in other treatments, broadcasting was used. The briquettes, weighing 1 g each, were prepared with different compositions of fertilizers based on the recommended nutrient doses (60-20-40 kg N-P₂O₅-K₂O ha⁻¹), and applied between four hills at a depth of 7–10 cm using a briquette applicator 20 days after sowing (DAS).

Standard procedures were followed to record data on various growth and yield attributes, including effective tillers per square meter, filled grains per panicle, spikelets per panicle, panicle length, 1000-grain weight, grain yield, straw yield, and harvest index. The content of N, P, and K in rice grain and straw was estimated. Nitrogen use efficiency was assessed using partial factor productivity, agronomic efficiency, and recovery efficiencies. Phosphorus- and potassium-use efficiencies were quantified using recovery efficiencies. Partial-factor productivity for N (PFPN) was calculated as kg grain yield per kg N applied. N agronomic efficiency (AE) refers to the grain yield produced as a result of N application in the fertilized plot compared to the control plot. Recovery efficiency (RE) is the fertilized plot’s total N-P-K removal per unit of N-P-K applied compared to the control plot.

2.3. Gas Sampling and Measurement (Through Static Chamber)

The greenhouse gas emissions were measured using the static-chamber method to quantify the emissions of nitrous oxide (N₂O), and methane (CH₄). In each plot of the experiment, aluminum base plates measuring 50 cm in length, 40 cm in width, and 12 cm in height were placed on a 7 cm wide channel, with the rim inserted into the soil. These base plates were installed in two replicates and remained in place until rice harvesting. To measure N₂O and CH₄ emissions, a fabricated Perspex chamber (transparent) measuring 50 cm in length, 40 cm in width, and 80 cm in height was placed on the channel of the base plate, covering six rice hills. The channel was then filled with water to ensure an airtight seal between the base plate and the chamber, preventing the entry of ambient gases. Two battery-operated fans were attached to opposite sides of the chamber to promote air circulation and uniform dispersion. A rubber cap was fitted at the top of the chamber to allow airtight insertion of a syringe needle for gas sample collection. Before collecting the air samples, a fan was operated to homogenize the air inside the chamber. Gas samples were collected from each plot at 0-, 15-, and 30-min intervals using a syringe with a volume of 20 mL. The samples were collected in the morning hours (08:00–10:30 AM) as these emissions rates were considered representative of the entire day.

Throughout the growing period, gas samples for CH₄, and N₂O, were collected on a fortnightly basis. In addition, for the measurement of N₂O, additional sampling was conducted for three consecutive days following the application of nitrogen fertilizer. The collected gas samples were then injected into a gas chromatograph (Model No. Thermo Fisher, TRACE1110, Manufacturer: Thermo Fisher Scientific, Maharashtra, India) to detect the peaks of the different greenhouse gases. The fluxes of CH₄ and N₂O were calculated using the following formulas [9]:

$${\mathrm{CH}}_4 \mathrm{flux} (\mathrm{mg} {\mathrm{m}}^{-2} {\mathrm{hr}}^{-1}) = \frac{(∆{\mathrm{X}}_{\mathrm{CH}4} \times {\mathrm{EBV}}_{\mathrm{STP}} \times 16 \times 60 \times {10}^{-3})}{22.4 \times \mathrm{A} \times \mathrm{T}}$$

(1)

$${\mathrm{N}}_2\mathrm{O} \mathrm{flux} (\mathrm{mg} {\mathrm{m}}^{-2} {\mathrm{hr}}^{-1}) = \frac{(∆{\mathrm{X}}_{\mathrm{N}2\mathrm{O}} \times {\mathrm{EBV}}_{\mathrm{STP}} \times 44 \times 60 \times {10}^{-3})}{22.4 \times \mathrm{A} \times \mathrm{T}}$$

(2)

∆X_CH4 = Difference between CH₄ concentration (ppm) of initial (0 min) and final (30 min) sample

∆X_N2O = Difference between N₂O concentration (ppb) of initial (0 min) and final (30 min) sample

EBV_STP = Effective chamber volume at standard temperature and pressure (liter)

T = Time gap between initial and final sampling after placement of the chamber

A = Area occupied by the base plate (m²).

2.4. Estimates of Global Warming Potential (GWP), Greenhouse Gas Intensity (GHGI), and Carbon Equivalent Emissions (CEE)

CO₂ was obtained as the reference for GWP and the fluxes of CH₄ and N₂O were converted into CO₂ equivalents using their GWPs. As per the IPCC 2013 [10] factor, the global GWP was calculated as the following:

$${\mathrm{CH}}_4 \times 28 + {\mathrm{CO}}_2 + {\mathrm{N}}_2\mathrm{O} \times 265 (\mathrm{kg} {\mathrm{CO}}_{2-}\mathrm{equivalent} {\mathrm{ha}}^{-1})$$

GHGI was calculated by dividing GWP by grain yield [11] as follows:

$$\mathrm{GHGI} = \begin{array}{l}\mathrm{Global} \mathrm{Warming} Potential (GWP)\\ \mathrm{Grain} \mathrm{Yield}\end{array}$$

(3)

where GHGI is expressed in kg CO₂-equivalents kg⁻¹ of grain yield.

CEE of the treatment was calculated as follows:

CEE = GWP × 12/44 (kg ha⁻¹)

2.5. Economic Analysis

For each treatment, the cost of cultivation was estimated in USD per hectare. The gross return was determined by calculating the economic yield at the minimum support price (MSP) and the straw yield at the prevailing market rate, then adding the results. The net return was computed by deducting the cultivation cost from the gross return on a per-hectare basis.

$$Net return = gross return - cultivation cost$$

The benefit-cost ratio (B-C ratio) was calculated by dividing the net return by the entire cost of cultivation.

2.6. Statistical Analysis

The impact of different fertilizer management practices on growth parameters, yield attributes, yield, NPK uptake, cumulative CH₄, N₂O emissions, GWP, CEE, and GHGI was tested using combined analysis (pooled analysis) in GRAPES version 1.1.0. A comparison of the treatment means was conducted with a Tukey’s test at 5% [12].

3. Results

Effect of FDP on Growth, Yield, and NPK Uptake.

The fertilizer briquette with 120% N + 100% P and K resulted in the highest plant height (140 cm), number of effective tillers per square meter (318), LAI at 90 DAS (5.9), dry matter production at harvest (2116 g per square meter), number of spikelets per panicle (218), filled grains per panicle (198), panicle length (27.9 cm), and 1000-grain weight (20.2 g) compared to all other treatments (

Table 1. Effect of fertilizer management practices on growth parameters of wet direct-seeded Sali rice.

Treatment	Plant Height (cm) at Harvest	Effective Tillers (No. m−2)	LAI at 90 DAS	Dry Matter Production (g m−2) at Harvest
T1: Control	83 f	206 g	1.6 f	851 f
T2: FP with 30-18.4-36	95 e	227 f	2.2 ef	1096 e
T3: RDF as per AAU	115 c	264 d	3.5 cd	1509 c
T4: BMP	116 c	272 cd	3.7 c	1557 c
T5: FDP with 120% N + 100% PK	140 a	318 a	5.9 a	2116 a
T6: FDP with 100% NPK	129 b	298 b	4.9 b	1823 b
T7: FDP with 80% N + 100% PK	119 c	278 c	4.1 c	1644 c
T8: FDP with 60% N + 100% PK	105 d	245 e	2.7 de	1286 d
Treatment (T)	*	*	*	*
Year (Y)	NS	*	NS	*
Y * T interaction	NS	NS	NS	NS

BMP = Best fertilizer management practices, FP = Farmers’ practice, RDF = Recommended dose of fertilizer, FDP = fertilizer deep placement. NS = Nonsignificant; Treatment means in each column followed by the same letter are not significantly different (p > 0.05). Asterisk (*) stands for significant.

and

Table 2. Effects of fertilizer management practices on yield attributes and yield of wet direct-seeded Sali rice.

Treatment	Panicle Length (cm)	Filled Grains (No. Panicle−1)	Total Spikelets (No. Panicle−1)	1000-Grain Weight (g)	Grain Yield (kg ha−1)	Straw Yield (kg ha−1)	Harvest Index (%)
T1: Control	23.2 f	100 g	128 g	17.9 f	1970 g	3488 g	36.0 e
T2: FP with 30-18.4-36	24.3 e	117 f	144 f	19.0 e	2642 f	4163 f	38.8 d
T3: RDF as per AAU	26.4 c	152 d	177 d	19.4 c	3780 d	5273 d	41.8 bc
T4: BMP	26.5 c	155 cd	179 cd	19.5 c	3819 cd	5290 d	41.9 abc
T5: FDP with 120% N + 100% PK	27.9 a	198 a	218 a	20.2 a	4933 a	6520 a	43.1 a
T6: FDP with 100% NPK	27.2 b	179 b	200 b	19.9 b	4451 b	5988 b	42.7 ab
T7: FDP with 80% N + 100% PK	26.5 c	162 c	184 c	19.5 c	4002 c	5451 c	42.3 abc
T8: FDP with 60% N + 100% PK	25.7 d	135 r	161 e	19.2 d	3297 e	4727 e	41.1 c
Treatment (T)	*	*	*	*	*	*	*
Year (Y)	*	*	*	NS	*	*	NS
Y * T interaction	NS	NS	NS	NS	NS	NS	NS

NS = Nonsignificant; Treatment means in each column followed by the same letter are not significantly different (p > 0.05). Asterisk (*) stands for significant.

). Similarly, applying 120% N through fertilizer briquette with 100% P and K produced a higher yield (Table 2). However, using briquette with 100% N, P, and K (4451 kg ha⁻¹) resulted in a statistically higher grain yield (~18%) compared to the fertilizer doses recommended by AAU (3780 kg ha⁻¹). A similar trend was observed in straw yield. The highest harvest index was recorded in FDP with 120% N + 100% PK. However, there was no statistical difference in harvest index when the same level of NPK fertilizers was applied through deep placement of briquettes, although comparatively higher values were recorded compared to conventional methods. FDP with 120% N + 100% PK recorded higher N, P, and K uptake (

Table 3. Effect of fertilizer management practices on N, P, and K uptake of wet direct-seeded Sali rice.

Treatment	Uptake (kg ha−1)						Total Uptake (kg ha−1)
	Grain			Straw
	N	P	K	N	P	K	N	P	K
T1: Control	11.56 f	2.31 g	4.07 f	8.73 g	3.50 f	23.17 f	20.29 g	5.81 f	27.24 f
T2: FP with 30-18.4-36	16.90 e	3.15 f	5.79 e	11.05 f	4.27 e	28.07 e	27.95 f	7.43 e	33.86 e
T3: RDF as per AAU	27.63 c	4.94 e	8.60 d	17.57 d	5.53 d	36.57 d	45.19 d	10.47 d	45.17 d
T4: BMP	27.99 c	5.00 e	8.70 d	17.76 d	5.56 d	36.73 d	45.74 cd	10.55 d	45.43 d
T5: FDP with 120% N + 100% PK	36.48 a	8.21 a	13.41 a	26.87 a	7.75 a	52.49 a	63.35 a	15.97 a	65.89 a
T6: FDP with 100% NPK	32.71 b	7.40 b	11.95 b	22.95 b	7.11 b	47.76 b	55.65 b	14.52 b	59.70 b
T7: FDP with 80% N + 100% PK	29.38 c	6.60 c	10.62 c	18.65 c	6.39 c	43.31 c	48.02 c	12.99 c	53.92 c
T8: FDP with 60% N + 100% PK	21.72 d	5.42 d	8.60 d	15.04 e	5.55 d	37.21 d	36.76 e	10.97 d	45.80 d
Treatment (T)	*	*	*	*	*	*	*	*	*
Year (Y)	*	*	*	*	*	*	*	*	*
Y * T interaction	NS	NS	NS	NS	NS	NS	NS	NS	NS

NS = Nonsignificant; Treatment means in each column followed by the same letter are not significantly different (p > 0.05). Asterisk (*) stands for significant.

).

3.1. Efficiency Indices

N-use efficiency was assessed using partial factor productivity, agronomic efficiency, and recovery efficiencies. For P and K, only recovery efficiencies were considered. The highest agronomic efficiency for N was observed in the FDP treatment with 80% N + 100% PK, resulting in a 42.3 kg increase in grain yield per kg of N applied (

Table 4. Effect of fertilizer management practices on agronomic efficiency, partial-factor productivity of N, and recovery efficiency of N, P, and K of wet direct-seeded Sali rice.

Treatment	N Agronomic Efficiency (kg Grain Increase/kg N Applied)	N Partial Factor Productivity (kg Grain Yield Per kg N Applied)	Recovery Efficiency (kg Nutrient Taken Up/kg Nutrient Applied)
			N	P	K
T1: Control	-	-	-	-	-
T2: FP with 30-18.4-36	22.4	0.88	0.26	0.09	0.19
T3: RDF as per AAU	30.2	0.63	0.42	0.24	0.45
T4: BMP	30.8	0.64	0.43	0.24	0.46
T5: FDP with 120% N + 100% PK	41.2	0.69	0.60	0.51	0.97
T6: FDP with 100% NPK	41.4	0.74	0.59	0.44	0.81
T7: FDP with 80% N + 100% PK	42.3	0.83	0.58	0.36	0.67
T8: FDP with 60% N + 100% PK	36.9	0.92	0.46	0.26	0.46

). This treatment showed an agronomic efficiency 89% higher than the farmers’ practice. The highest partial factor productivity for N (0.92) was observed in the FDP treatment with 60% N + 100% PK (Table 4). However, the highest apparent recovery was observed in the FDP treatment with 120% N + 100% PK. This treatment also showed the highest apparent N-recovery efficiency (ANR) (0.60), apparent P-recovery efficiency (APR) (0.51), and apparent K-recovery efficiency (AKR) (0.97).

3.2. Green House Gas Emissions

Significant variations were observed in the emission rate values of CH₄ at different stages of the rice crop. The intensity of CH₄ emissions initially increased in all treatments studied until it peaked during the period of 49 DAS to 63 DAS (Figure 2). During this peak period, CH₄ emission rates ranged from 1.00 to 2.26 mg m⁻² h⁻¹ under different treatments (

Table 5. Effect of fertilizer management practices on CH₄ emission in wet direct-seeded Sali rice.

Treatment	CH4 Emission (mg m−2 h−1) at
	21 DAS	35 DAS	49 DAS	63 DAS	77 DAS	91 DAS	105 DAS	112 DAS
T1: Control	0.17 f	0.85 g	1.08 g	1.00 g	0.78 g	0.58 f	0.36 f	0.12 f
T2: FP with 30-18.4-36	0.21 e	1.05 f	1.35 f	1.25 f	0.97 f	0.72 e	0.45 e	0.15 e
T3: RDF as per AAU	0.28 c	1.40 d	1.79 d	1.66 d	1.29 d	0.96 c	0.59 c	0.20 c
T4: BMP	0.28 c	1.41 cd	1.80 d	1.67 d	1.30 d	0.97 c	0.60 c	0.21 c
T5: FDP with 120% N + 100% PK	0.35 a	1.78 a	2.26 a	2.10 a	1.64 a	1.21 a	0.75 a	0.26 a
T6: FDP with 100% NPK	0.32 b	1.62 b	2.06 b	1.92 b	1.49 b	1.11 b	0.68 b	0.24 b
T7: FDP with 80% N + 100% PK	0.29 c	1.46 c	1.87 c	1.74 c	1.35 c	1.00 c	0.62 c	0.21 c
T8: FDP with 60% N + 100% PK	0.25 d	1.25 e	1.59 e	1.47 e	1.15 e	0.85 d	0.53 d	0.18 d
Treatment (T)	*	*	*	*	*	*	*	*
Year (Y)	NS	*	*	*	*	*	NS	NS
Y * T interaction	NS	NS	NS	NS	NS	NS	NS	NS

NS = Non-significant; Treatment means in each column followed by the same letter are not significantly different (p > 0.05). Asterisk (*) stands for significant.

). The highest emissions were consistently observed in FDP with 120% N + 100% PK, followed by FDP with 100% NPK, except at 21 DAS. The impact of different fertilizer management practices on cumulative CH₄ emissions was significant (Table 5). The highest cumulative CH₄ emission (32.84 kg ha⁻¹ crop cycle⁻¹) was obtained in FDP with 120% N + 100% PK, followed by FDP with 100% NPK (Table 7). The lowest values of cumulative CH₄ emissions were observed in the control group (15.74 kg ha⁻¹ crop cycle⁻¹).

There were also significant variations in N₂O emissions due to different treatments. It is evident that the emission of N₂O from the rice field significantly varied with the stage of the rice crop. A detailed analysis (Figure 2) demonstrated that N₂O emissions in RDF as per AAU and BMP peaked at 7 DAS, 35 DAS, and 63 DAS. However, in FDP, only one N₂O emission peak was found at 21 DAS, 1 day after fertilizer application. A progressive increase in the rate of N₂O emission was observed in the unfertilized control treatment until it reached the emission peaks. Thereafter, the emission of N₂O gradually decreased until the harvesting stage of the rice crop (

Table 6. Effect of fertilizer management practices on N₂O flux in wet direct-seeded Sali rice.

Treatment	N2O Flux (µg m−2 h−1) at
	7 DAS	8 DAS	9 DAS	21 DAS	22 DAS	23 DAS	35 DAS	36 DAS	37 DAS	49 DAS	63 DAS	64 DAS	65 DAS	77 DAS	91 DAS	105 DAS	112 DAS
T1	1.89 d	1.94 e	2.12 f	2.94 e	3.24 e	3.43 d	22.25 f	22.45 f	22.75 f	35.91 f	47.76 e	47.73 e	47.78 h	20.55 g	13.69 g	7.40 e	4.45 d
T2	23.00 c	10.51 c	7.67 c	13.01 d	13.22 cd	13.31 ab	53.45 c	33.95 c	28.50 bc	54.78 b	60.87 b	61.08 b	61.38 c	43.29 b	25.83 b	18.81 a	8.69 b
T3	28.23 a	20.96 a	10.81 a	13.58 cd	13.90 cd	14.00 ab	62.44 b	39.84 b	29.95 ab	65.84 a	81.88 a	71.68 a	63.10 b	49.79 a	30.44 a	20.65 a	10.76 a
T4	25.28 b	14.22 b	8.56 b	13.35 cd	13.58 cd	13.68 ab	65.93 a	42.75 a	31.00 a	65.85 a	84.08 a	74.49 a	64.05 a	49.70 a	30.53 a	20.73 a	10.18 a
T5	1.94 d	1.97 d	2.17 d	20.19 a	16.84 a	14.87 a	25.46 d	26.00 d	26.60 cd	43.95 c	55.20 c	55.94 c	58.88 d	39.44 c	22.95 c	15.19 b	6.10 c
T6	1.90 d	1.96 de	2.17 d	17.95 b	15.79 ab	14.49 ab	24.22 e	24.46 e	26.51 cd	41.85 d	54.22 c	55.08 c	57.37 e	33.45 d	19.90 d	13.71 bc	5.52 c
T7	1.90 d	1.95 de	2.15 e	16.94 b	14.77 bc	12.96 b	20.41 g	23.35 ef	25.30 de	39.73 e	52.76 cd	52.76 cd	55.15 f	28.18 e	16.83 e	11.04 cd	4.49 d
T8	1.90 d	1.94 e	2.14 e	14.84 c	12.85 d	8.36 c	18.98 h	22.83 f	23.16 ef	37.85 ef	50.00 de	50.39 de	53.35 g	25.54 f	15.69 f	9.66 de	3.60 e
T	*	*	*	*	*	*	*	*	*	*	*	*	*	*	*	*	*
Y	*	*	*	*	*	*	*	NS	*	NS	*	*	*	*	*	*	*
Y * T	NS	NS	NS	NS	NS	NS	NS	NS	NS	NS	NS	NS	NS	NS	NS	NS	NS

T₁: Control; T₂: FP with 30-18.4-36; T₃: RDF as per AAU; T₄: BMP with 60-20-40; T₅: FDP with 120% N + 100% PK of RDF; T₆: FDP with 100% NPK; T₇: FDP with 80% N + 100% PK; T₈: FDP with 60% N + 100% PK. NS = Non-significant; Treatment means in each column followed by the same letter are not significantly different (p > 0.05). Asterisk (*) stands for significant.

and Figure 2). The effect of different fertilizer management practices on cumulative N₂O emissions was significant. The highest cumulative N₂O emissions (1.010 kg ha⁻¹ crop cycle⁻¹) were obtained in BMP, which was on par with state RDF as per AAU (1.004 kg ha⁻¹ crop cycle⁻¹). The lowest values were observed in the control group (0.502 kg ha⁻¹ crop cycle⁻¹) (

Table 7. Effect of fertilizer management practices on cumulative CH₄, and N₂O emissions, global warming potential (GWP), carbon equivalent emissions (CEE), and greenhouse gas intensity (GHGI) in wet direct-seeded Sali rice.

Treatment	CH4 Emission (kg ha−1 Crop Cycle−1)	N2O Flux (kg ha−1 Crop Cycle−1)	GWP (kg eq CO2 ha−1)	CEE (kg ha−1)	GHGI (kg eq CO2 kg of GY−1)
T1: Control	15.74 f	0.502 g	573.64 e	156.45 e	0.296 a
T2: FP with 30-18.4-36	19.65 e	0.857 b	777.11 d	211.94 d	0.295 a
T3: RDF as per AAU	25.95 c	1.004 a	992.58 b	270.70 b	0.262 b
T4: BMP	26.12 c	1.010 a	998.77 b	272..39 b	0.262 b
T5: FDP with 120% N + 100% PK	32.84 a	0.733 c	1113.38 a	303.65 a	0.227 d
T6: FDP with 100% NPK	29.98 b	0.681 d	1019.83 b	278.14 b	0.230 d
T7: FDP with 80% N + 100% PK	27.14 c	0.623 e	924.72 c	252.20 c	0.232 cd
T8: FDP with 60% N + 100% PK	23.03 d	0.575 f	797.20 d	217.42 d	0.242 c
Treatment (T)	*	*	*	*	*
Year (Y)	*	NS	*	*	NS
Y * T interaction	NS	NS	NS	NS	NS

NS = Nonsignificant; Treatment means in each column followed by the same letter are not significantly different (p > 0.05). Asterisk (*) stands for significant.

).

3.3. GWP, CEE, and GHGI

The term “global warming potential” (GWP) is used to compare the effects of different gases on global warming. A particular gas warms the Earth more than CO₂ does over that period if its GWP is bigger. The ratio of the time-integrated radiative forcing resulting from the immediate emission of one kilogram of a trace material to that of one kilogram of a reference gas is known as the global warming potential (GWP). The highest GWP and carbon emission efficiency (CEE) values were observed in the treatment with a fertilizer dose of 120% nitrogen (N) + 100% phosphorus (P) and potassium (K) (GWP-1113.38 kg eq CO₂ ha⁻¹; CEE- 303.65 kg ha⁻¹). This treatment was closely followed by the treatment with 100% NPK, which was found to be comparable to the best fertilizer management practice (BMP) and the recommended dose of fertilizer (RDF) recommended by the Assam Agricultural University (AAU) (Table 7). Notably, the lowest GWP and CEE values were observed in the control plot where no fertilizer was applied.

The greenhouse gas intensity (GHGI) correlates GWP with crop production. As with GWP, positive values reported as kg CO₂ equivalents per kilogram of rice grain produced indicate a net supply of GHGs to the atmosphere, whereas negative values indicate net GHG sinks in the soil. The highest GHGI was recorded in the control plot (0.296 kg eq CO₂ kg of grain yield⁻¹), which was statistically similar to the farmers’ practice (Table 7). In contrast, the lowest GHGI (0.227 kg eq CO₂ kg of grain yield⁻¹) was observed in the treatment with 120% N + 100% PK.

3.4. Available N, P₂O₅ and K₂O Content in Soil

The effect of different fertilizer management practices on soil available N (kg ha⁻¹), P₂O₅ (kg ha⁻¹), and K₂O (kg ha⁻¹) after harvest of the crop was significant (Figure 3). The highest available N, P₂O₅ (kg ha⁻¹), and K₂O (kg ha⁻¹) were obtained in FDP with 120% N and 100% PK. Regardless of nitrogen level, fertilizer application through deep placement of NPK briquettes resulted in comparatively higher available soil N, P₂O₅, and K₂O than conventional methods. The control plot recorded the lowest values.

3.5. Economics

Economics is an important parameter that decides the adoption levels of any newly developed farming technology by farmers. Any new technology should be technically and economically viable. Therefore, both the production of a crop and its cost of cultivation are equally important. Data on different components of economics (Figure 4) indicate that among the treatments, FDP with 120% N + 100% PK fetched the maximum mean gross and net returns with a mean B-C ratio. The minimum mean gross return of USD 675.11 ha⁻¹, net returns of USD 426.17 ha^−,1 and B-C ratio of 0.58 were in the control plots.

4. Discussion

4.1. Growth Parameters, Yield Attributes, and Yield

The variations in plant height and tiller count per square meter across treatments can be attributed to the slow release of nitrogen (N) from the briquettes for 65 days, which aligns with the plant’s demand [13]. This gradual release appears to have a positive effect on the number of tillers per square meter [14]. The application of NPK briquettes through deep placement resulted in a higher number of effective tillers compared to broadcasting prilled urea [15,16]. The deep placement of fertilizer enhances nutrient uptake, increases leaf area, and promotes higher dry matter accumulation, ultimately leading to increased grain yield [17]. An improvement in leaf area index (LAI) was observed with deep placement of N, as opposed to conventional application methods.

The increase in dry matter production with the application of 120% N + 100% PK through fertilizer deep placement (FDP) is likely due to the precise delivery of nutrients to the root zone, minimizing nutrient losses and promoting dry matter production, which is consistent with previous findings [18]. The enhanced chlorophyll content in leaves leads to a higher photosynthetic rate, providing sufficient photosynthates during grain development. The slow release of N over an extended period may stimulate spikelet formation per panicle, resulting in increased spikelet number, grain filling percentage, and 1000-grain weight in FDP treatments, thereby improving yield attributes compared to broadcasting, as reported elsewhere [15,16].

The application of 120% N through fertilizer briquettes combined with 100% P and K resulted in the highest yield (Table 2). Briquettes with 100% N, P, and K also led to a statistically significant 16% increase in grain yield (4451 kg ha⁻¹) compared to the fertilizer doses recommended by AAU (3780 kg ha⁻¹) and BMP (3819 kg ha⁻¹). This treatment also exhibited superior yield-contributing characteristics, such as a higher number of effective tillers per square meter (317), longer panicle length (27.9 cm), and higher counts of total and filled spikelets per panicle (218 and 198, respectively), as well as a higher 1000-grain weight (20.1 g), resulting in a greater grain yield (4933 kg ha⁻¹). This improvement is attributed to reduced nutrient losses [19], as the NPK briquettes provide a steady supply due to sustained release of N throughout the rice-growing season, and applied P and K are used more effectively [20]. Li et al. (2021) also reported higher grain yields with deep placement of N. FDP significantly increased grain yields compared to broadcasting during the wet season. The highest straw yield recorded under FDP with 120% N + 100% PK (6520 kg ha⁻¹) may be attributed to the sustained availability of nutrients throughout the crop growth period, enhancing leaf area, photosynthesis, and dry matter accumulation, which in turn supports vigorous crop growth and higher straw yield, consistent with previous findings [21].

4.2. Nutrient Content and Uptake

The application of FDP with 120% N + 100% PK resulted in higher uptake of nitrogen (N), phosphorus (P), and potassium (K) by rice, as shown in Table 3. In contrast, when N was broadcast applied in the form of prilled urea, the N uptake by rice was lower. This higher uptake observed with FDP can be attributed to the gradual release of N from the briquettes, which minimizes the release of ammonium into the soil-water system, thereby reducing losses through volatilization and denitrification. This mechanism improves N intake and assimilation by providing a prolonged supply of nutrients within the plant-water system. Previous studies have indicated that using briquettes significantly affects the N uptake of rice [22,23]. Furthermore, increased phosphorus (P) content and uptake were observed in the FDP treatments (Table 3). This could be due to the deep placement of P, which reduces interactions with soil and makes more P available to rice [24]. The reduced contact area with soil also decreases P fixation and enhances P availability. Nutrient interactions play a crucial role in the plant–soil system, as a deficiency of one nutrient might impair the absorption and utilization of others [25]. The synergistic action of N and P, when supplied locally, may enhance P uptake. Placing P near the root zone allows crop roots to more efficiently capture the applied P fertilizer, promoting lateral root growth and expanding the area for P uptake.

In addition, the deep placement of potassium (K) along with N and P enhances crop uptake by minimizing losses through runoff [26]. This synergistic interaction between P, N, and K, combined with deep placement, improves overall nutrient efficiency and crop performance.

4.3. Nutrient Use Efficiencies (NUE)

Agronomic efficiency, also known as N-use efficiency (NUE), depends on the synchronization between the nitrogen (N) requirements of crops and the available N supply [27]. It assesses the balance between the amount of N absorbed and utilized by the crop versus the amount lost. NUE specifically reflects the rice plant’s grain yield response to N fertilizer. Agronomic efficiency was highest under FDP with 80% N + 100% PK, resulting in an increase of 42.3 kg of grain per kg of N applied, representing an 89% improvement over traditional farmers’ practices (Table 4). This improvement demonstrates that deep placement of NPK briquettes significantly enhances N-use efficiency compared to broadcast applications of prilled urea. By placing urea deeper in the anaerobic soil layer, this method potentially reduces N concentrations in floodwater and the oxidized surface layer, leading to decreased N losses through denitrification, ammonia volatilization, and runoff, which in turn increases fertilizer N-use efficiency [15,16].

Partial-factor productivity for N (PFPN) quantifies the total economic outputs relative to all N sources used, including indigenous soil N and applied fertilizer N, expressed as kg of grain yield per kg of applied N [28]. The most notable PFPN was observed under FDP with 60% N + 100% PK, achieving a value of 0.92 (Table 4). Typically, PFPN is higher at lower N application rates and decreases as N levels increase, consistent with previous observations [29].

Apparent recovery, a primary indicator of nutrient uptake and utilization in rice, measures the efficiency of absorbed applied nutrients. The highest apparent recovery was observed under FDP with 120% N + 100% PK. This improved recovery in FDP treatments is likely due to the sustained availability of N in the form of NH₄⁺-N in the reduced soil layer, which extends the nutrient’s accessibility to the rice crop [26]. Importantly, deep placement significantly increased the recovery of applied P and K by 83% and 80%, respectively, compared to the broadcast application [16].

4.4. Greenhouse Gas Emissions

The study observed that methane (CH₄) emissions from the rice crops were initially low during the early growth stages, gradually increasing from maximum tillering to the panicle initiation stage (Figure 2). However, the emission rates began to decline following panicle development, with minimal emissions recorded during the crop’s maturity period. The CH₄ flux is known to be influenced by the above- and below-ground dynamics of plant biomass [30]. Consequently, the lower CH₄ emissions noted during the early stages of the rice crop in this study are likely due to reduced CH₄ transport from the soil to the atmosphere, attributable to lesser root and shoot growth during these stages. Additionally, it was observed that the CH₄ emission rate was lower in the initial phases of plant growth due to the rice plant’s reduced transport capacity, characterized by fewer tillers and smaller leaf area [30]. The highest methane emissions were recorded in treatments utilizing FDP with 120% N + 100% PK, decreasing with lower N doses in the fertilizer briquettes. Methane emissions were notably higher in plots treated with NPK briquettes compared to those treated with the recommended doses of fertilizer (RDF) by AAU and best fertilizer management practices (BMP) at equivalent N levels.

N₂O emissions were found to be sporadic and driven by specific events. Peaks in N₂O emissions were detected following the application of urea. However, the deep placement of fertilizer significantly reduced N₂O emissions throughout the rice-growing season compared to the broadcast application. This reduction is primarily because N₂O production predominantly occurs at the soil surface where microbial nitrification and denitrification are more intense compared to deeper soil layers. The deep placement of nitrogen in the reduced zone helps retain nitrogen in the form of NH₄⁺. This NH₄⁺-N then diffuses slowly from the reduced zone to the soil surface, minimizing the availability of inorganic nitrogen substrates in the biologically active zone. Consequently, the deep placement of fertilizer significantly lowers NH₄⁺-N levels in floodwater, effectively reducing nitrogen loss through NH₃ volatilization, nitrification, and denitrification [31]. In contrast, prilled urea provides readily available nitrogen, which facilitates nitrification and subsequent denitrification, leading to increased production and release of N₂O. These findings suggest that one-time deep placement of fertilizers could be an effective strategy for boosting rice output and nitrogen use efficiency (NUE) while simultaneously mitigating N₂O emissions. Given the concerns regarding greenhouse gas emissions and global warming, FDP emerges as a potentially preferable method for the wet season in Assam to reduce N₂O emissions under current climatic conditions.

The application of fertilizer briquettes with 120% N and 100% P and K yielded the highest global warming potential (GWP) among the treatments examined. This outcome can be attributed to the elevated levels of soil organic matter and saturation content, which are known to enhance CH₄ emissions and consequently increase the GWP. Despite the briquette application leading to lower nitrous oxide emissions compared to standard fertilizer practices (Table 7), the overall GWP was higher due to the aforementioned factors. Specifically, the briquette application resulted in 27% less N₂O emission compared to the recommended dosing of fertilizer (RDF) application practice recommended by AAU (Table 7). The carbon emission efficiency (CEE) exhibited a consistent pattern across all treatments (Table 7). The treatment utilizing briquettes with 120% N and 100% P and K displayed the lowest greenhouse gas intensity (GHGI) at 0.227 kg CO₂-equivalent per kg of grain. Conversely, the control treatment, which did not receive the optimized nutrient application, exhibited the highest GHGI at 0.296 kg CO₂-equivalent per kg of grain yield, mainly due to lower yield in the control plot. Fertilizer practices recommended by AAU and BMPs resulted in similar GHGI levels.

4.5. Available N, P₂O₅ and K₂O Content in Soil

The results demonstrate that FDP with 120%, 100%, and 80% N resulted in an increase of 3.47%, 2.90%, and 2.35% N above the initial available soil N. The control plot showed signs of soil N depletion. Deep fertilization improved the interaction between particles and soil, lowers N loss through runoff and other means, and increases the ultimate available soil N status, allowing more N to remain in the soil [32]. All FDP treatment plots showed higher available P₂O₅ and available K₂O buildup over the initial than the conventional method of fertilizer application. The final available soil K₂O and P₂O₅ status improved as a result of deep potassium and phosphorus placement.

4.6. Economic Analysis

The highest economic parameters (gross returns, net returns, and benefit-cost ratio) in FDP with 120% N + 100% PK might be due to high grain yield resulting from more availability of nutrients in time. The deep placement of fertilizer usually helps farmers best when they have little or no control over water management and timing of N application under rainfed wet season circumstances. Higher labor costs associated with the deep placement of fertilizer briquettes compared to the broadcast method of fertilizer application might be compensated by lower fertilizer costs and higher grain yields, resulting in higher economic returns [18]. The deep placement of NPK provides the added benefit of simultaneously supplying all primary nutrients, thereby ensuring balanced fertilization. This is particularly significant in countries such as India, where farmers frequently apply an excessive quantity of N fertilizers with insufficient P and K, and where one-time NPK applications save on labor. Finally, it may be concluded that the use of NPK briquettes over a broadcast of three different NPK fertilizers and its incorporation was economically viable and efficient for rice production in Assam condition during kharif season.

5. Conclusions

The findings of this study suggest that the most optimal approach to nutrient management for wet direct-seeded rice during the Sali/wet season is to apply 120% nitrogen (N) and 100% phosphorus-potassium (PK) of the recommended dose of fertilizer (RDF) via deep placement. This approach leads to the highest yield and NPK recovery efficiency. However, when considering factors such as agronomic efficiency, partial factor productivity, NPK recovery efficiency, global warming potential, and carbon emission equivalent resulting from the release of three greenhouse gases (CH₄, CO₂, and N₂O) from rice fields, it is advisable to apply 80% N with 100% PK of RDF through deep placement. A 20% reduction in nitrogen fertilizer dose without yield reduction has significant implications for food security, economics, and the environment, as well as GHG mitigation in the intensive rice production system in Assam, with the highest climate change vulnerability index among all Indian states. Reducing fertilizer doses will subsequently reduce the burden on the government for subsidies and import bills. This nutrient management practice is deemed the most suitable for Sali/wet season rice cultivation in Assam.