Responses of Methane Emission and Bacterial Community to Fertilizer Reduction Plus Organic Materials over the Course of an 85-Day Leaching Experiment

Abstract

Methane produced from paddy fields has a negative impact on global climate change. However, the role of soil bacterial community composition in mediating methane (CH₄) emission from waterlogged paddy soil using the column experiment is poorly known. In the present study, various fertilization treatments were adopted to investigate the effects of fertilizer reduction combined with organic materials (CK: control; CF: conventional fertilization; RF: 20% fertilizer reduction; RFWS: RF plus wheat straw amendment; RFRS: RF plus rapeseed shell amendment; RFAS: RF plus astragalus smicus amendment) on CH₄ emission and soil bacterial community during an 85-day leaching experiment. We hypothesized that the fertilizer reduction plus the organic materials could enrich the bacterial communities and increase CH₄ emission. The average CH₄ flux varied from 0.03 μg m⁻² h⁻¹ to 76.19 μg m⁻² h⁻¹ among all treatments in the nine sampling times, which may account for the experimental conditions such as air temperature, moisture, and anthropogenic factors. In addition, high-throughput sequencing was utilized to investigate the alteration of the soil bacterial community structure. It was revealed that the diversity and composition of the bacterial community in the topsoil amended with organic materials underwent significant shifts after the 85-day leaching experiment. Proteobacteria was identified as the dominant phylum of the soil bacteria, with an average proportion of 35.2%. For Firmicutes, the proportion of RFRS (11%) was higher than that in the CK (8%), RF (8%), RFWS (7%), RFAS (6%), and CF (5%) treatments. Additionally, Gammaproteobacteria and Alphaproteobateria were supposed to be the major class bacterial communities, with average proportions of 12.8% and 12.2%, respectively. For the RFWS treatment, the contribution of Alphaproteobateria was the highest among all the bacterial relative abundance. According to the correlation heatmap analysis, the top ten bacterial communities were positively related to soil microbial biomass carbon (MBC) and ammonia nitrogen (NH₄⁺-N) (p < 0.01). The findings also indicated that the RFRS treatment was the favorable management to alleviate CH₄ emission during an 85-day leaching experiment or possibly in paddy production. Collectively, these results predict that the impacts of different treatments on CH₄ production are strongly driven by soil microbial communities and soil properties, with soil bacteria being more prone to the crop residue degradation stage and more sensitive to soil properties. The discoveries presented in this study will be useful for assessing the efficacy and mechanisms of organic material amendments on CH₄ emissions in paddy soil.

1. Introduction

Rice (Oryza saliva) is one of the most important grain crops in the world and is considered a staple food by 40% of the world’s population. Meanwhile, previous investigations reported that the demand for rice will be expected to increase by 24% over the next 20 years [1,2]. As is well known, CH₄ ranks second to carbon dioxide (CO₂) in terms of global warming potential, with a 25-fold higher warming potential over a 100-year time span [3]. CH₄ emissions in agricultural activities in China amounted to 22,245 Gg in 2014, of which 40.1% originated from paddy fields [4]. Hence, to mitigate climate change, alternative agricultural practices need to be developed to curtail CH₄ emissions from rice paddies. Excessive utilization of chemical fertilizers and intensive cultivation can result in the loss of soil organic carbon (SOC) and the decline of nitrogen (N) use efficiency in croplands [5]. In recent years, scientists have utilized a series of agricultural practices such as irrigation regime, straw mulching and fertilizer reduction in rice production to mitigate methane emissions [6,7,8]. Crop residue incorporation, which is a potential alternative to agricultural management, has great positive environmental benefits, such as increasing soil fertility and crop yield [9], and decreasing chemical fertilizer application and air pollution from crop residue burning [10]. Additionally, straw incorporation combined with reduced chemical fertilizer can enhance N immobilization and decrease the soil oxygen and OC substrate sources for methanogens, resulting in an increase in CH₄ emission in rice paddies [11,12,13]. Previous observations indicated that the impact of straw return on soil microbial communities may also be attributed to differences in straw application rates, durations, fertilizer management, and other factors [14,15]. The chemical structure of different crop residues changes during decomposition, which in turn affects the decomposability of the remaining residues [16]. Therefore, the origins, types, and amounts of organic materials can play various roles in regulating soil characteristics and bacterial communities, thereby resulting in the difference in CH₄ emission rates in paddy soil. Consequently, further analysis is needed to understand how fertilizer reduction combined with organic materials on CH₄ emission and soil bacterial effects in long-term leaching experiments.

Generally, CH₄ emission from paddy soil involves three components: production, oxidation and transportation, which are dominated by soil microbial processes, including its production in flooded soil anaerobic conditions by methanogens and consumption (oxidation) in aerobic microsites by methanotrophs [17,18]. Previous studies also indicated methanogens are distributed among the seven orders of Euryarchaeota: Methanosarcinales, Methanobacteriales, Methanomicrobiales, Methanococcales, Methanopyrales, Methanocellales and Methanoplasmatales [19]. Therefore, key factors such as soil C/N and oxygen concentration, which regulate the ratio of methanogens to methanotrophs will ultimately determine the CH₄ emissions typically passing through a particular pathway from production in aerobic soil layers into the atmosphere [20]. Investigations indicated that CH₄ is produced by the decomposition of organic carbon in anaerobic soils or sediments [21]. The areas of bulk density, specific surface area and microhabitat of microbial organisms may be improved when straw returned to farmlands, resulting in the variations of CH₄ emissions fluxes. Straw incorporation with reduced chemical fertilizers also regulated the methanogenic and methanotrophic community structure. Previous investigations showed that straw retention and chemical fertilizer reduction changed methanogenic and methanotrophic community structure, meanwhile, increasing the relative abundance of Methyloaracoccus (a dominant genus of methanotrophs) from 0.02% to 61.8% [8,22]. Moreover, straw incorporation might accelerate the propagation of methanotrophs and improve CH₄ oxidation potential by influencing soil aeration and soil microorganism habitat [23]. Researchers also revealed that various carbon sources from organic fertilizer application might regulate CH₄ production by affecting the abundance of Methanosarcinales and Methanobacteriales [24,25]. Hereof, the effect of soil bacterial community on CH₄ production should be clarified in waterlogged paddy soil with fertilizer reduction plus organic material amendment.

However, the influences of diverse organic materials on soil bacterial communities and CH₄ emissions via soil column leaching simulation tests have received scant coverage. Herein, we postulated that the changes in soil physicochemical conditions (e.g., soil nutrients, aeration, and texture conditions) resulting from straw amendments might mediate the methane oxidation potential and community structure of methanotrophs in paddy field ecosystems. A long-term column leaching experiment was performed under fertilizer reduction in combination with various organic matters. The main objectives of the present study were to address the following questions: (1) Do fertilizer regimes stimulate flux? (2) How do fertilizer regimes change soil (0–10 cm) bacterial community and structure? (3) What are the potential microbial drivers of CH₄ emission over the course of an 85-day leaching experiment? The results obtained in the presented study may provide a new cognition on the effects of optimizing fertilization regimes on CH₄ emissions mechanisms and management strategies in rice production.

2. Materials and Methods

2.1. Design of Soil Column Leaching Experiment

Soil used in this study was taken from Xinji County, Hanzhong City, Shaanxi Province (33°0′16″ N, 108°48′44″ E), with a type of loggotrophic paddy soil and a texture of clayey soil. The conditions of irrigation and drainage were controlled by 10~30 mm thin water layer on the surface in the greening period, and the soil moisture in the root layer accounted for 60~80% of the field water capacity in the other growth periods. Organic materials including WS (Wheat Straw), RS (Rapeseed Shell) and AS (Astragalus Smicus) were also sampled, and then air-dried and crushed for further use. The total C, N, and P proportions of WS, RS, and AS were measured by Element Analyzer (Unicube, Elementar, Hanau, Germany) in dry weight (Table S1). Moreover, the materials and functional groups analysis of WS, RS and AS were employed by flourier transform infrared spectrometer (FTIR). The organic materials were dried to a constant weight at 60 °C and subsequently crushed using a 100-mesh sieve. Take a small amount (about 1–2 mg) of the sample and grind it into a fine powder using an agate mortar and pestle. Weigh out about 100 mg of potassium bromide (KBr) powder, which is infrared transparent and grind this in a separate agate mortar and pestle, and finally subjected to scanning at the specified wavelength of 4000–400 cm⁻¹. A resolution between 4 cm⁻ˡ and 8 cm⁻ˡ is sufficient for most applications and the number of scans can vary from 16 to 128, depending on the noise level that can be tolerated and the time available. The main functional groups, such as oxhydryl, carbonyl, alkyl, and sulfonyl were generally identified by the absorbance peaks at 3200–3600 cm⁻¹, 1650–1820 cm⁻¹, 2850–2960 cm⁻¹, and 1000–1400 cm⁻¹, respectively [26].

Soil column leaching experiments were adopted to explore the environmental implication of organic materials amendment in waterlogged paddy soil, in the Yangling, Shaanxi Province (34°15′ N, 108°34′ E). Soils in columns were collected in 0–40 cm soil layers every 10 cm (0–10, 10–20, 20–30, and 30–40 cm, respectively), which were dominated by a cultivation system of wheat–rice regime. The physicochemical characteristics of soils in the present study are shown in

Table 1. Physicochemical properties of soils used vertically (0–40 cm) in this study.

Depth (cm)	pH	SOM (g kg−1)	AK (mg kg−1)	AP (mg kg−1)	CEC (cmol kg−1)	TN (g kg−1)
10	7.38 ± 0.21 b	32.22 ± 0.44 a	130.40 ± 1.21 a	20.12 ± 0.61 a	18.60 ± 0.22 a	1.21 ± 0.10 a
20	7.59 ± 0.11 ab	24.58 ± 0.51 b	59.10 ± 4.21 b	13.80 ± 1.81 b	20.18 ± 2.81 a	1.24 ± 0.17 a
30	7.63 ± 0.02 a	10.97 ± 1.09 c	39.61 ± 1.10 c	7.55 ± 0.37 c	19.87 ± 3.61 a	0.39 ± 0.05 b
40	7.72 ± 0.03 a	6.53 ± 0.41 d	38.23 ± 1.11 c	5.66 ± 0.32 c	18.31 ± 1.31 a	0.26 ± 0.03 b

Notes: SOM: Soil organic matter; AP: Available potassium; AP: Available phosphorus; CEC: cation exchange capacity; TN: Total nitrogen. Data (means ± SD, n = 3) followed by different letters indicate significant differences among soil depths at p < 0.05.

. SOC contents were determined by the K₂Cr₂O₇ oxidation-FeSO₄ titration method. Soil samples were extracted by 1 M NH₄Ac four times, and then titrated with standard solution of hydrochloric acid (HCl), calculating the CEC value basied on dosage of HCl finally. Soil pH was determined with a pH meter (PB-10, Sartorius, Gottingen, Germany) in a soil:water solution (1:2.5, w:v). Total N was measured by an alkaline hydrolysis diffusion approach, available P was determined by a molybdenum blue colorimetric technique, and available K was quantified by flame photometry. The soil was extracted in 1 M NH₄Ac (pH 7.0), in accordance with the methods detailed in our previous study [27]. Additionally, the dissolved organic carbon (DOC) released from soil (0–10 cm) was determined by TOC analyzer (TOC-L CPH, Shimadzu, Kyoto, Japan) (Table S4) after leaching experiment. The excitation-emission matrices (EEMs) analysis of soil DOC samples was simultaneously processed by a fluorescent spectrometer (Lengguang Tech., F97, Shanghai, China), with emission (Em) wavelength 250–550 nm, excitation (Ex) wavelength 200–500 nm and increment of 5 nm, respectively. The bandwidths of Ex and Em were 10 nm, and the scanning speed was 6000 nm min⁻¹. Additionally, the fluorescent indices, including FI (the ratio of the fluorescence intensity at Em 450 nm and 500 nm, and at Ex 370 nm) [28], HIX (the ratio of the integral area over Em 435–480 nm to 300–345 nm, at Ex 254 nm) [29] and BIX (the ratio of fluorescence intensity at Em 380 nm and 430 nm, and at Ex 310 nm) [30] were also calculated in our work.

According to the bulk density of soil in the four layers, about 1000 g air-dried soil was filled in soil every 10 cm column (Figure 1B). Moreover, 0–10 cm soil layers were identified as different treatments (i.e., CK: control treatment without organic material and fertilizer; CF: conventional fertilization; RF: 20% fertilizer reduction compared with CF; RFWS: RF plus wheat straw amendment; RFRS: RF plus rapeseed shell amendment; RFAS: RF plus astragalus smicus amendment). Each treatment was repeated three times and eighteen soil columns were conducted in this study. The inputs of fertilizer application and organic materials were consistent with our previous study [31]. The amounts of WS, RS, and AS applications were 3.93 g, 2.36 g, and 2.36 g per column in topsoil (0–10 cm), which were converted by the amount of field fertilizer (Table S2). Approximately 1.5 L deionized water was used to stabilize column for three days and saturate soil before leaching experiment. The leaching experiment begins when the soil water content was 70% of field water capacity (FWC), and the approximate 1 cm high overlying water maintained over the course of a 70-day leaching experiment. The leachate of the four depths was artificially collected by latex tubing (10 cm length) and plastic bottle (25 mL) under natural conditions. Additionally, the DOC, nitrate nitrogen (NO₃-N), NH₄⁺-N, and total phosphorus (TP) concentrations have significant differences in different sampling depths. The surface water and leachate samples were collected after gas collection and the amounts of samples were approximately 15 mL for further detection every sampling time.

2.2. Collections and Detections of the CH₄ Fluxes

To facilitate the collection of gas samples, a cylindrical sampling chamber (20 cm × 10 cm) was utilized to sample methane gas on the 1st, 3rd, 6th, 10th, 15th, 25th, 40th, 60th, and 85th days (Figure 1C). Prior to gas sampling, the column was capped with the chamber, and the seam crossing was sealed with tape to avoid air leakage. A 50 mL syringe was employed to repeatedly mix the collected air for the zero-time sample. The other three gas samples were collected in the same manner every 10 min. Finally, the mixed gas samples were conserved in gas collection vessel for further detection. Gas chromatography (GC; Agilent 7890, Santa Clara, CA, USA), equipped with a flame ionization detector (FID) and and thermal conductivity detector (TCD), was employed to detect CH₄ concentration. The temperatures of the injection port, detector and oven were 250, 250 and 60 °C, respectively. Additionally, CH₄ fluxes were calculated by CH₄ accumulation from zero time to 1 h later air sampled. Beijing ZG Special Gases Science and Technology Corporation (Beijing, China) supplied the CH₄ standard gas with a concentration of 1.8 mg L⁻¹, used to calculate the integral area of gas volume fraction to sampled time. The GHG emission rates were calculated from the change in gas concentrations with time after the chamber was closed. The calculation of CH₄ fluxes followed the method described in the previous study [32]:

$$J={\displaystyle \frac{dc}{dt}}\ast {\displaystyle \frac{M}{V_0}}\ast {\displaystyle \frac{P}{P_0}}\ast {\displaystyle \frac{T_0}{T}}\ast H$$

where J represents the CH₄ (μg m⁻² h⁻¹) emission flux; dc/dt represents the slope of gas volume fraction to t (mm³ (m³ h)⁻¹); M is V is CH₄ molar mass; P is the atmospheric pressure of chamber (m²); T is the absolute temperature in the chamber (K); V₀, P₀, T₀ are the molar volume of gas (mL mol⁻¹), atmospheric pressure (Pa), absolute temperature (K), respectively, at standard condition; H is the height of chamber (m).

2.3. Topsoil DNA Extraction and High throughput Sequencing

The polymerase of TransGen TransStart Fastpfu DNA Polymerase AP221-02 was operated to amplify bacterial 16S rRNA genes following the primers [338F (ACTCCTACGGGAGGCAGCAG) and 806R (GGACTACHVGGGTWTCTAAT)] in V3-V4 hypervariable region [33]. QIA Quick Gel Extraction Kit (QIAGEN, Dusseldorf, Germany) and Real-Time PCR (ABI GeneAmp^® 9700, New York, NY, USA) were used to purify and quantify PCR products, respectively. Additionally, High throughput sequencing was performed on Illumina’s MiSeq-PE250 platform at the Allwegene Company (Beijing, China). The dataset was analyzed using QIIME and the sequences were clustered into operational taxonomic units (OTUs) at a similarity level of 97%. The Ribosomal Database Project (RDP) Classifier tool (NBTIC, Bethesda, ML, USA) was also employed to classify all the sequences into different taxonomic groups, which can calculate the abundances of OTUs and the parameters of diversities [34]. The raw sequences were deposited into the NCBI Sequence Read Archive (SRA) database with accession number PRJNA1065809.

2.4. Statistical Analysis

The original data were calculated by Microsoft Excel (Microsoft Excel 2019, Microsoft, Albuquerque, NM, USA). One-way analysis of variance (ANOVA) was regarded as being statistically significant in using SPSS 21.0 (SPSS Inc., Chicago, IL, USA). DUCAN test was used to determine significant differences among all treatments (p < 0.05). The correlations between various indices were assessed by Pearson’s correlation analysis (p < 0.05, p < 0.01 and p < 0.0001). Beta diversity of bacterial samples in different treatments was analyzed by principal Component Analysis (PCA)and Principal Coordinate Analysis (PCoA) which were conducted by R language packages coupled with broken-stick model based on Bray–Curits distances at OTUs level (Supplementary Materials). All figures in this study were drawn by Origin Pro (Origin 2021, Northampton, MA, USA).

3. Results

3.1. Characteristics of Soil and Organic Materials Used in This Study

The physicochemical properties of the soils used vertically (0–40 cm) in the present study are shown in Table 1. The SOM content obviously decreased with an increase in soil depth, showing a range of 6.53 g kg⁻¹ to 32.22 g kg⁻¹ (p < 0.05) (Table 1). Similarly, diverse origins of organic materials led to significant variations in the properties of WS, RS, and AS, encompassing nutrient contents, phase compositions, and main functional groups. The total C, N, and P proportions of WS, RS, and AS are shown in Supplementary Table S1, which indicates the C/N order of WS (47.87) > RS (43.07) > AS (24.83). The three organic materials mainly consist of carbon components rather than nitrogen and phosphorus, which can provide more carbon sources for microbial activity and regulate CH₄ emissions.

As a whole, the order of absorption peak intensity was WS > RS > AS at the same wavenumber. FTIR characteristics of AS, RS, and WS were shown in Figure 2, which indicated a similar trend in the main absorption peaks of 858 cm⁻¹, 1300 cm⁻¹, 1539 cm⁻¹, 1715 cm⁻¹, and 3009 cm⁻¹. The absorption peak at 858 cm⁻¹ pointed to the stretching vibration of -CHO in AS lower than that in WS and AS. The bending vibration O-H at 1300 cm⁻¹ showed the compounds of phenols, alcohols or carboxylic acids. The bending vibration N-H at 1539 cm⁻¹ illustrated the amine-group compound. The stretching vibration C=O at 1715 cm⁻¹ forecasted the compounds of ketones, aldehydes or carboxylic acids. The stretching vibration C-H at 1715 cm⁻¹ forecasted the alkene compound. The aromatic ketones and aromatic carboxylic acid compounds (benzophenone, benzoic) were considered the main substances in the three materials, which may provide nutrients for soil microorganisms during the decomposition of a long-term leaching experiment. Then, the activities and abundances of relative microorganisms can mediate the CH₄ emissions rates.

3.2. Changes of CH₄ Emissions over the Course of an 85-Day Leaching Experiment

The dynamic changes and the average emissions of the CH₄ fluxes are described in Figure 3. The emission of the CH₄ flux changed with a scope of 0.001 μg m⁻² h⁻¹~320.51 μg m⁻² h⁻¹ and the average CH₄ emission flux varied from 0.03 μg m⁻² h⁻¹ to 76.19 μg m⁻² h⁻¹ during the leaching experiment in the waterlogged paddy soil. One-way ANOVA and the DUCAN test (p < 0.05) were used to analyze the difference in CH₄ emission fluxes among all treatments over the course of an 85-day leaching experiment. A significant (p < 0.05) difference in the CH₄ emission flux occurred in RFWS (320.51 μg m⁻² h⁻¹) on May 2, meanwhile, the peak of average CH₄ emission flux also occurred here (76.19 μg m⁻² h⁻¹), which may be attributed to the abundant carbon released from WS during the leaching experiment. However, the CH₄ flux rate in the RFWS treatment (81.80 μg m⁻² h⁻¹) was lower than that in the RFAS treatment (121.01 μg m⁻² h⁻¹) on 2 July, which may account for the reduction of soil water or the destruction of anaerobic conditions.

3.3. Variations of the Soil Bacterial Community Structure

To reveal the variations of the bacterial community and structure in the topsoil plus organic materials and fertilizer after a long-term leaching test, high throughput sequencing was employed to distinguish the variations of the soil bacterial structure (Figure 4). Chao, ACE and Shanon analyses showed that organic material amendments with fertilizer reduction enriched the soil bacterial community diversities, which attributed to the favorable habitat for soil organisms. The Chao index is directly related to species richness, for which it provides a higher estimate compatible with the sample abundance data, and the Shannon index is a species diversity metric. For alpha diversity indices of Chao, the values of bacterial diversities showed that the RFWS (3398.27) treatment was higher than CK (2940.77), CF (3174.30), RFRS (3036.43), and RFAS (3153.34) (p < 0.05). The ACE values displayed an order of RFWS (3413.34) > RF (3297.59) > CF (3165.81) > RFAS (3153.79) > RFRS (3051.42) > CK (2941.83). When the Shannon analysis was adopted, the diversity values were RFWS (10.04) > RFAS (10.01) > RF (9.939) > CF (9.81) > RFRS (9.77) > CK (9.70). However, no difference existed in Simpson analysis among all treatments (Table S3).

The top ten dominant bacterial communities were identified in Figure 5A,B at the phylum and class levels, respectively. In this study, Proteobacteria, Actinobacteria, and Acidobacteria were regarded as the three major bacterial communities at the phylum level, with average proportions of 35.2%, 12.3%, and 15.3%, respectively. Among them, the relative distribution of Proteobacteria was significantly higher than that in all treatments, but no difference existed among the different treatments. For Firmicutes, the proportion of RFRS (11%) was higher than that in the CK (8%), RF (8%), RFWS (7%), RFAS (6%), and CF (5%) treatments. In addition, Gammaproteobacteria, Alphaproteobacteria, Deltaproteobacteria, and Acidoproteobacteria were considered to be the main bacterial communities at the class level, with average contributions of 12.8%, 12.2%, 10.0%, and 8.8%, respectively. Compared to CK (11%), the distribution of Gammaproteobacteria in the CF, RF, RFWS, RFRS, and RFAS amendments increased to 14%, 12%, 12%, 12%, and 14%, respectively. The contribution of Alphaproteobacteria was the highest in the RFWS treatments among all bacterial relative abundances, accounting for 15%. The abundance of Deltaproteobacteria ranged from 9% to 11%, which was higher than the distribution of Acidoproteobacteria in all treatments, ranging from 7% to 11%.

3.4. Correlation of the Bacterial Communities to Environmental Factors

To understand the effect of fertilizer reduction combined with organic materials on the structural composition of soil microorganisms, the differences in soil bacterial community structures were analyzed by PCoA (Figure 6A) and PCA (Figure 6B). The loadings for all variables involved in the PCA in different treatments are shown in the Supplementary Materials. PC1 and PC2 were obtained, with contributions of 14.7% on the first axis and 8.75% on the second axis in PCoA, which indicated a similar OTU composition between the RFWS and RFAS treatments. However, the CF and RFRS treatments showed similar variations in OUT compositions in PCoA, with the dispersion of data points in the four quadrants.

Conversely, the OTU compositions of the CK and RF treatments were similar, which demonstrated the fertilizer application rate was the main factor in bacterial community structures. Additionally, PC1 and PC2 were acquired by PCA, accounting for 11.8% and 9.67% of the differences in bacterial structure. The points of the RFWS, RFRS and RFAS treatments were located in the same quadrant above the main axis in PCA, which shows the semblable structural composition. However, compared to CF, the samples of the CK and RF treatments were located in the same quadrant below the main axis, which indicates the importance of the fertilizer reduction regime on soil microbial variations.

To explore the relationship between bacterial communities and environmental factors (Table S4), the changes of the top ten dominant soil bacterial communities and soil relative indices are shown by heatmap analysis at the phylum and class levels (Figure 7A,B). The correlation of soil bacterial communities and soil indices was analyzed using Origin Pro (Origin 2021, USA) and a p-value at 0.05, 0.01 and 0.001 levels, respectively. Overall, the indices of FI (Fluorescence intensity), HIX (Humification index) and α:β (ratio of new and old carbon), which represented soil dissolved organic carbon (DOC) spectral characteristics, had a highly significant correlation (p < 0.001). At the phylum level, Verrucomicrobia had a significant association with Armatimonadetes (p < 0.01), and the contribution of the Bacteroidetes community was significantly correlated with soil NH₄⁺ (p < 0.01), suggesting the crucial effect of soil characteristics on soil microbial structures. Regarding the class level, the relative abundance of Bacteroidia was strongly correlated with soil NH₄⁺ (p < 0.01). Additionally, the relative distribution of Actinobacteria showed a significant relationship with soil DOC and microbial biomass nitrogen (MBN) (p < 0.05), and the Bacilli community was significantly related to soil AP concentration.

4. Discussion

4.1. Fertilizer Reduction Plus Organic Materials Accelerate the CH₄ Emissions from the Flooded Paddy Soil

Based on the soil’s basic properties, the indices of AK, AP, CEC, and TN exhibit a similar variation trend where the upper-layer content is higher than that in the lower layer, which is attributed to the application of organic or inorganic fertilizers during long-term cultivation. The organic materials (WS, RS, and AS) used in this study are rich in carbon (C) and nitrogen (N), providing ample energy sources for soil microorganisms involved in CH₄ production. Previous studies have also demonstrated that variations in soil depth can significantly influence the distribution and availability of essential nutrients [35]. The high C and N contents in the organic materials could potentially stimulate specific microbial metabolic pathways involved in CH₄ production, warranting further investigation into the underlying mechanisms. The FITR analysis showed that WS contained more plentiful C=C and C=O groups than in RS and AS, which is consistent with the results of elements analysis.

Generally, the contents of carbon (C) and nitrogen (N) have significant influences on microbial metabolic pathways and CH₄ production. However, the effects of reduced chemical fertilizer combined with crop residues on soil bacterial community composition over the course of long-term leaching experiments need to be explored deeply. Previous studies reported that reduced chemical fertilizer combined with straw retention can enhance CH₄ emission by regulating the abundance of soil methanogens and methane-oxidising bacteria, which is consistent with the present results [36,37]. The exogenous organic matter can be biodegraded under both aerobic and anaerobic conditions, especially in the drying and wetting cycling process [38]. Moreover, it is widely agreed that straw amendments supply additional organic C substrate for methanogens and increase plentiful CH₄ emissions [39,40]. On the other hand, fertilization improves photosynthetic production and therefore provides more rhizodeposition C for methanogenesis, as well as the oxidization of rhizosphere soil for mathanotrophs [41]. The relative strategy can also facilitate soil oxygen transportation and enhance its available concentration [42]. The current results also indicated that the CH₄ emissions flux and the abundance of bacterial diversity significantly rose for the RFWS, RFRS, and RFAS treatments; however, there was no significant difference in the emission flux among the CK, CF, and RF treatments, which is in line with the published studies [39,43]. This phenomenon could be mainly attributed to the 85-day soil column leaching experiment that incorporated fresh straw into the waterlogged soil when the temperature was relatively high, as the CH₄ emissions flux peaked shortly after fertilization for 28 days (Figure 3) [44,45]. This was also likely due to the fact that the application of fresh crop residues to paddy soil accelerates CH₄ emission by regulating soil pH or enhancing anaerobic oxidation processes as an electron acceptor [46]. Therefore, to the best of our knowledge, the RFRS treatment should be recommended as the moderate strategy that mitigates CH₄ emission in rice paddies.

4.2. Potential Mechanism of CH₄ Emission under Fertilizer Reduction Plus Organic Materials

As is known, microbial communities play a crucial role in the decomposition of organic matter. Different microorganisms have specific enzymes and metabolic pathways for breaking down complex carbon compounds. This process can release substrates that are subsequently utilized for CH₄ production. On the other hand, certain microorganisms such as Methanotrophic bacteria can in return anaerobically oxidize CH₄, influencing the net emissions. Understanding these complex interactions is crucial for developing sustainable agricultural practices that balance crop productivity and environmental protection. It is well-known that N fertilization has a positive influence on paddy soil microbial communities, particularly in a waterlogged environment [47,48]. Nevertheless, the research on the response of soil bacterial community composition to fertilizer reduction combined with organic materials during the leaching simulation period is lacking. In this study, the top ten bacterial community compositions were identified among various treatments (Figure 5A,B), indicating that reducing chemical fertilizer plus using organic matter amendments can significantly regulate the main bacterial communities of methanogenesis after a long-term leaching experiment. As the biodegradable organic compounds decompose, the living conditions of the related organisms may deteriorate, and the favorable habitat of the related microorganisms may be destroyed due to the lack of fresh substrates or favorable soil conditions. The abundant carbohydrates, including starch and glucose, which may be released from organic matter, provide an energy substrate for the metabolism of most microorganisms [49]. Additionally, the redox potential decreases with the consumption of oxygen during straw decomposition, which leads to the growth of methanogens and inhibits the activity of methanotrophs [50]. The increase in degradable C causing the increase in CH₄ emissions due to straw incorporation is associated with reduced fertilizer [51]. With the leaching experiment process, the soil C/N, pH and redox significantly changed because of the reduction of water and nutrient contents, which may directly affect the activity and abundance of methanogens and methanotrophs. Future research could focus on unraveling the specific molecular mechanisms underlying the observed changes in microbial communities and their functional implications for soil processes.

Previous studies showed that organic amendments could significantly increase the soil microbial activity and microbial biomass [24,52]. In this study, fertilizer reduction plus organic material treatments increased the soil microbial C substrate, which could supply more C and N sources for methanogenic growth. In general, methanomicrobia and methanobacteria were considered to be the main compositions of the methanogenic archaeal community, shifting the basal metabolic activities in paddy soils [53]. The results were consistent with a previous study on the functional and structural responses of methanogenic archaeal communities with rice straw addition, which demonstrated that the hydrogenotrophic methanogenic pathway was in particular controlled by the availability of degradable OC [54]. PCoA and PCA analyses showed that the RFWS, RFRS and RFAS treatments had similar bacterial communities and OTU structural composition; however, relatively lower bacterial abundances were found in the CK, CF and RF treatments, which suggests the important role of organic C and moderate N sources on microorganism growth [55]. If there is an excess of one without a corresponding amount of the other, it can lead to nutrient imbalances that affect microbial growth efficiency [43]. Additionally, the form in which these nutrients are present (e.g., whether N is in an ammonium or nitrate form) can also modulate the impact on microbial growth. The correlation heatmap revealed that the distributions of Bacteroidetes and Bacteroidia were dramatically close to soil NH₄⁺ concentration, which was devoted to the degradation of crop residue or the growth of Methanogens [56]. Reporter indicated that hydrogenotrophic Methanocella and acetoclastic Methanosaeta, which can generate methane by using hydrogen as an electron donor and carbon dioxide as a carbon source, were considered the major methanogenic communities associated with CH₄ production and emission [57]. The major bacterial communities including Bacteroidia, Acidobacteriia, Actinobacteria and Proteobacteri obtained in this study were related to Methanocella, whose variations may mediate the CH₄ emission rate during the period of the leaching experiment. Moreover, factors such as the availability of hydrogen and carbon dioxide, as well as the overall environmental conditions (e.g., temperature, pH, and redox potential), can affect the growth and metabolic rate of the main bacterial communities relative to methanogenesis, thereby impacting CH₄ net emission. Altogether, our results showed that fertilizer reduction plus organic materials generated great changes in the CH₄ emission and the abundance of relative bacterial community composition in paddy soils during a long-term leaching experiment. However, the present experiment was isolated from field trials, and no plants participated in the leaching experiment process, which scarcely reflects the actual field conditions. Hence, further investigations of the fertilization regime’s impact on the CH₄ emission molecular mechanism should be conducted in a long-term field test.

4.3. Implication

The effects of fertilizer reduction combined with various crop residues on the variations of CH₄ emission and bacterial communities were investigated via an 85-day soil column leaching experiment.

It was demonstrated that the RFRS treatment significantly decreased and the RFWS regime treatment significantly increased the CH₄ emission from paddy soil. Simultaneously, this was primarily associated with the proliferation of Bacteroidetes and Bacteroidia, which were capable of better degrading various organics or mediating the composition of microbial substrates in waterlogged anaerobic conditions. Nevertheless, the biochemical processes of Methanomicrobia and Methanobacteria genes, which regulate CH₄ production, remain ambiguous. In future studies, the potential influence of the redox process on altering the CH₄ emissions mechanism by modifying soil-dissolved organic matter or the electron transport process should also be considered [58,59]. To reduce errors in experimental data and sampling, an automatic collection device should be adopted instead of manual operation, which can improve the consistency of the results in deep study. The release of ebullition fluxes that occurred during the duration of gas sampling should also be considered in the follow-up study. Further studies are also necessary to elucidate how different microbial communities are formed with different types of organic material amendments. Moreover, the present study only observed a slight variation in CH₄ emission with fertilizer reduction plus organic materials during the leaching process; therefore, more long-term field experiments should be carried out to verify the changes more accurately.

5. Conclusions

In this study, the ability of organic materials to balance soil fertility and regulate CH₄ emissions in paddy soil was demonstrated. The results showed that the CH₄ flux ranged from 0.001 μg m⁻² h⁻¹ to 320.51 μg m⁻² h⁻¹ using various plus organic material treatments over the course of an 85-day leaching experiment. RFWS, RFRS and RFAS showed higher CH₄ flux than that of the CK, RF and CF treatments, which illustrated the importance of N stimulation on the degradation of organic C from crop residue during the leaching experiment. Additionally, Bacteroidia, Acidobacteriia, Actinobacteria and Proteobacteria were confirmed to be the major bacterial community compositions at the phylum and the class analysis levels. Notably, Bacteroidetes and Bacteroidia were correlated with soil MBC, NH₄⁺-N, suggesting that the vital role of modest N should not be ignored when considering the CH₄ emission from waterlogged paddy soils. The results indicated that the application rates of chemical fertilizers and the choices of crop residues were equally important in rice cultivation.