Association between CH₄ uptake and N₂O emission in grassland depends on nitrogen inputs

Abstract

Increasing nitrogen (N) input has been recognized as one of the important factors influencing methane (CH₄) uptake and nitrous oxide (N₂O) emission in arid and semiarid grasslands. Numerous studies have examined the spatiotemporal variations of CH₄ and N₂O fluxes in various ecosystems; however, the variation of the interplay between CH₄ uptake and N₂O emission with increasing N has not yet been well understood. This study explored the relationship between CH₄ uptake and N₂O emission in a semiarid grassland in Inner Mongolia, northern China, under a gradient of 12-year N additions. We found a synergistic relationship at low-N levels, where CH₄ uptake and N₂O emission are positively correlated. Conversely, an antagonistic interaction emerged with a negative correlation between CH₄ uptake and N₂O emission observed at high-N levels, which was evidenced by a 33.62% decrease in CH₄ uptake and a 264.91% increase in N₂O emission. Further independent analysis, covering at least five N addition levels across grassland ecosystems in China, confirmed the general pattern: three of four cases showed a synergistic relationship at low-N levels and an antagonistic relationship at high-N levels. Given the increasing N deposition in the future, the dynamics between CH₄ uptake and N₂O emission are critical for understanding the impact of external N input on net greenhouse gas emission and consequent global climate change.

摘要

氮添加对草地生态系统CH₄吸收与N₂O排放耦合关系的影响

氮(N)输入增加是影响干旱和半干旱草地甲烷(CH₄)吸收和氧化亚氮(N₂O)排放的重要因素之一。已有大量研究探讨了不同生态系统中CH₄和N₂O通量的时空变化，但尚不清楚CH₄吸收和N₂O排放之间的耦合关系随着N输入的增加如何变化。本研究依托中国北部内蒙古半干旱草原地区开展了连续12年的N添加实验平台，探讨CH₄吸收与N₂O排放之间的耦合关系。研究结果表明，CH₄吸收和N₂O排放在低N添加条件下呈现正相关的协同关系；在高N添加条件下，呈现负相关的拮抗关系，具体表现为CH₄吸收减少了33.6%，N₂O排放增加了264.9%。中国草地生态系统至少5个N添加水平实验的独立分析证实了这一普遍规律：4个N添加实验中，有3个在低N水平上表现出协同关系，在高N水平上表现出拮抗关系。考虑到未来N沉降会增加，本研究中观察到的CH₄吸收与N₂O排放之间耦合关系的变化对于理解外源N输入对温室气体净排放及其对全球气候变化的影响至关重要。

INTRODUCTION

Methane (CH₄) and nitrous oxide (N₂O) are powerful greenhouse gases, with a global warming potential of 27.9 and 273 times more than carbon dioxide over a 100-year time period (Masson-Delmotte et al. 2021). Even minor fluctuations in the fluxes of CH₄ and N₂O can have substantial impacts on climate change. Semiarid grasslands are sinks and sources of atmospheric CH₄ and N₂O (Dijkstra et al. 2013; Shi et al. 2021). Nitrogen (N) increasing has been shown to suppress CH₄ uptake while enhancing N₂O emission in grasslands (Chen et al. 2021; Du et al. 2021; Shi et al. 2022; Stevens et al. 2022) with potentially significant implications for global climate change. An interesting pattern has been observed in semiarid grasslands, where peak CH₄ uptake coincides with the lowest N₂O emission (Mosier et al. 1996). However, this pattern is not consistent across all grassland ecosystems, particularly with global changes associated with increasing N.

N impacts on the rates of CH₄ uptake and N₂O emission in semiarid grasslands are intricately linked to essential global change processes (Xiao et al. 2023). However, the response of CH₄ uptake and N₂O emission to N addition is asynchronous (Ambus and Robertson 2006; Mori et al. 2005), which demonstrated as a complex synergistic or antagonistic relationship. For instance, in Northern Spruce Forests, increased atmospheric N deposition promoted both soil N₂O emission and soil CH₄ uptake, resulting in a synergistic relationship between the two gases (Maljanen et al. 2006). Conversely, in the Sapporo Forest, N addition inhibited CH₄ uptake while increasing N₂O emission, showing an antagonistic relationship (Kim et al. 2012). The interaction between CH₄ uptake and N₂O emission can be influenced by CH₄ oxidation, nitrification and denitrification processes, with ammonium (NH₄⁺) playing a crucial role (Blackmer et al. 1980; Bowman et al. 1995). NH₄⁺ competed with CH₄ for the active site of methane monooxygenase (MMO), thereby inhibiting CH₄ oxidation (Bodelier and Frenzel 1999; Holmes et al. 1995; Zhang et al. 2020). During nitrification, the oxidation of NH₄⁺ led to N₂O production (Levy-Booth et al. 2014). Meanwhile, denitrification converted nitrate (NO₃⁻) into N₂, with N₂O serving as an intermediate product (Kool et al. 2011; Long et al. 2017). Studies have demonstrated that low-N can enhance CH₄ uptake in temperate grasslands, whereas higher levels inhibit it (Peng et al. 2019; Zhang et al. 2023). This effect was mediated by a complex interplay of environmental and microbial factors (Bodelier and Laanbroek 2004; Shcherbak et al. 2014). Furthermore, N additions impacted N₂O emission by increasing soil inorganic N content, thereby modulating both nitrification and denitrification processes (Li et al. 2018). However, the underlying mechanisms governing the N-induced coupling of CH₄ uptake and N₂O emission remain unclear. To gain a comprehensive understanding of this relationship and enhance our predictive capabilities regarding CH₄ uptake and N₂O emission, an integrative research approach is required.

Although previous studies have explored the spatiotemporal dynamics of CH₄ uptake and N₂O emission in response to N addition in grasslands (Cai et al. 2009; Gu et al. 2019), these processes have often been assessed independently, neglecting their coupled dynamics (Chen et al. 2021; Dijkstra et al. 2013; Shi et al. 2021). Simulations of N deposition predominantly focused on examining the impacts of varying levels and frequencies of enrichment, often utilizing high dosages without intervals (Kim et al. 2012; Rees et al. 2004). The relationship between CH₄ uptake and N₂O emission is particularly significant given their opposing roles in the global climate system. For instance, CH₄ significantly contributes to short-term warming due to its high global warming potential, whereas N₂O posed a particular long-term concern due to its persistence and cumulative effect on the climate (Zhan et al. 2023). The interaction between these processes under multilevels of N over time remained uncertain. To address this knowledge gap, a comprehensive long-term study is necessary to capture the intricate complexity of these interactions with greater accuracy. It will improve our understanding of how N affected the coupling dynamics between CH₄ uptake and N₂O emission.

To elucidate shifts in the relationship between CH₄ uptake and N₂O emission and the underlying mechanisms, we measured the flux data in a field experiment and integrated the existing literature data. The field study utilized a long-term (initial from 2003) multilevel N addition experiment in a semiarid grassland, aiming to investigate the relationship between CH₄ uptake and N₂O emission across different N gradients. We hypothesized that (i) the coupling between CH₄ uptake and N₂O emission would exhibit either synergistic or antagonistic patterns; (ii) the coupling relationship between CH₄ uptake and N₂O emission would differ under low- and high-N levels. These hypotheses are built on the stimulatory effects of low-N on CH₄ uptake and the inhibitory effects of high-N (Peng et al. 2019), while both low- and high-N were found to promote N₂O emission (Gu et al. 2019; Guo et al. 2022). By strategically managing the coupling relationship dynamics in the future, we can potentially enhance CH₄ uptake and reduce N₂O emission in grassland ecosystems, thereby contributing to climate change mitigation and informing the development of effective grassland management strategies.

MATERIALS AND METHODS

Study site

This study was conducted in Duolun County (42°02′ N, 116°70ʹ E, 1324 m a.s.l.), a semiarid temperate steppe of Inner Mongolia, northern China. The average annual temperature is 2.1 °C, and the average annual rainfall is 385 mm. The landscape, located at elevations ranging from 1150 to 1800 m, is characterized by low foothills. According to the US Soil Taxonomy classification, the soil is identified as a Calcis-Orthic Aridisol, characterized by 20.30% ± 0.04% silt, 16.95% ± 0.01% clay and 62.75% ± 0.04% sand. The dominant plant species include Stipa Krylovii, Artemesia frigida, Potentilla acaulis, Cleistogenes squarrosa, Allium bidentatum and Agropyron cristatum.

Experimental design

The N fertilization experiment was established in 2003. Seven N treatments (1, 2, 4, 8, 16, 32 and 64 g N m⁻² y⁻¹) and a control (CK), each with eight replicates, were randomly allocated in the areas of homogenous block to avoid heterogeneity within individual plots. Each treatment comprised plots of 10 m × 15 m, separated by a 4-m wide buffer zone to ensure isolation between adjacent blocks. Nitrogen as urea was applied annually in mid-July at rates of 0, 1, 2, 4, 8, 16, 32 and 64 g N m⁻² y⁻¹. Since the local fertilization occurred during the growing season (May–September), all measurements and observations were conducted exclusively during this period.

Measurement of ecosystem CH₄ uptake and N₂O emission

Between 2013 and 2015 (the 10th to 12th consecutive year of N addition), we measured CH₄ uptake and N₂O emission using a static chamber technique (Livesley et al. 2009). Prior to annual data collection, three permanent stainless steel bases (50 cm × 50 cm × 12 cm), featuring a 3-cm deep water seal groove, were embedded into the soil to a depth of 12 cm following the methodology outlined by Zhang et al. (2012). A 50-cm high stainless steel chamber, coated with a heat and light insulating material, was securely placed over the base. Sealing was achieved by filling the groove with water. To ensure consistent air mixing within the chamber’s headspace, two electric fans were installed at the top of the chamber (Zhang et al. 2014). Gas samples were collected every 10 min for a duration of 30 min, using 60 mL syringes equipped with airtight stopcocks. Concurrently, the chamber’s air temperature, pressure (data from adjacent weather stations) and soil temperature using a Long Stem Thermometer 6310 and moisture (5–10 cm depth) using a portable soil moisture kit ML2x (ThetaKit, Delta-T Devices, Cambridge, UK) were monitored. The concentrations of CH₄ and N₂O in the collected samples were analyzed using a gas chromatograph (Agilent 7860, Agilent Technologies) and a specific mesh column equipped with a flame ionization detector for CH₄ and electron capture detection for N₂O, while maintaining the oven at a constant temperature. Nitrogen gas served as the carrier gas, flowing at a fixed rate. Fluxes of CH₄ and N₂O were calculated from a linear regression of concentrations from samples taken at 0, 10, 20 and 30 min after sealing the chamber (r² ≥ 0.95). Positive flux values, indicating CH₄ uptake and N₂O emission, were observed during weekly measurements from May to September from 2013 to 2015.

Measurements of soil properties

From May to September across the years 2013 to 2015, three soil cores were monthly extracted from each plot to a depth of 10 cm. These samples were homogenized and subsequently divided into two subsets for analysis. One subset was refrigerated at 4 °C for the determination of microbial biomass nitrogen (MBN), utilizing the chloroform fumigation–extraction method described by Liu et al. (2014). Concurrently, laboratory assessments were conducted to determine soil pH, NH₄⁺ and NO₃⁻ levels. Soil total nitrogen (STN) was measured using a CNS elemental analyzer (Variomax CNS Analyser, Elementar GmbH, Hanau, Germany), which analyzed the total nitrogen content of the soil samples. The second subset of soil samples was preserved at −20 °C for subsequent quantification of the pmoA gene using Quantitative Polymerase Chain Reaction (qPCR) with the primer pairs A189F/Mb601R (Cai et al. 2016). The polymerase chain reaction mixture for this process consisted of 10.0 µL of SYBR Premix Ex Taq (Takara), 0.5 µM of each primer, 2 µL of the DNA template and 7 µL of ddH₂O, and was analyzed on an ABI7500 Real-Time Detection System (Applied Biosystems, Inc., CA, USA).

Data selection and integration methods

Our analysis primarily focused on semiarid grasslands, with enhancements derived from integrating data from the literature, including four carefully chosen field observations. The selected field measurements covered at least one complete growing season. Since our study was aimed at developing a mechanistic understanding rather than quantifying the annual budget, we performed the measurements and analyses in the growing season only. The dataset encompassed diverse grassland types, spanning various N gradients and durations. The dataset comprised alpine grasslands from the Qinghai-Tibetan Plateau, Ergun’s forest grasslands, semiarid grasslands of Wei Yuan Town in You Yu County, Shanxi Province, and the Bayinbuluk grasslands of the Tianshan Mountains in Xinjiang. These locations were documented in studies conducted by Li et al. (2012), Peng et al. (2018, 2019), Sun (2018) and Ge et al. (2023).

Statistical analysis

Previous studies have primarily relied on correlation and regression coefficients to explore the relationship between CH₄ uptake and N₂O emission. However, direct comparison of the magnitudes of these coefficients was difficult, especially in cases where there was no significant relationship and the sample size is relatively small. Therefore, we introduced the coupling ratio (CR), defined as the logarithmic ratio, as a metric to assess the coupled variations between CH₄ uptake and N₂O emission. This approach allowed for a more accurate evaluation of the changes in their coupling relationship across varying N levels and years.

In this formula, we applied logarithmic transformations to both the numerator and the denominator to compress the range of larger CH₄ uptake and N₂O emission values. This addressed the scenario where the standard deviation of each group of data was approximately proportional to the square of its mean (Benoit 2011). By transforming the original data into a logarithmic scale (lgx), the variance became more uniform and multiplicative effects were converted into additive effects (Bartlett 1947). To enhance the formula’s generalizability and ensure its applicability across diverse datasets, we incorporated the maximum absolute value and the slack variable, denoted as ‘a’, from each data group. This addition ensured that both the numerator and the denominator remained positive, thereby upholding the validity of mathematical operations (Keene 1995).

The specific formula was as follows:

The mean value of the CR was determined for each level of N addition. The variable k represented the rate of change in the CR in response to N addition, where the subscript i designated the specific gradient of N addition. Our experimental design included eight different gradients, with i varying from 1 to 8 (i = 1, 2, …, 8). To determine the slope of the CR values across consecutive N addition levels, we used the following method:

Our study explored two potential scenarios: first, when both CH₄ uptake and N₂O emission increased concurrently, CR would remain constant within a certain range, exhibiting a relatively stable trend, consistent with findings by Mertler et al. (2021). Second, in the event of a decrease in CH₄ uptake and a simultaneous increase in N₂O emission, indicating an antagonistic relationship, CR would increase, indicated by a positive k value. A robust correlation typically underscored consistency or opposition in the relationship between variables, with stability in the direction and magnitude of change. Hypothesis testing was used to determine significant differences in CR values across different groups or gradients. Under low-N levels, the strength of the synergistic relationship was evaluated by assessing the consistency of CR through standard deviation, with a larger deviation indicating a more pronounced synergy (Mertler et al. 2021). Conversely, under high-N levels, CR exhibited a tendency toward higher values, with the intensification of the antagonistic relationship correlating with a greater rate of CR change or a larger k value (Bartlett 1947).

The coupling mechanism between CH₄ uptake and N₂O emission was investigated through structural equation modeling (SEM). Initially, a priori modeling was used to delineate the relationship between N addition and the processes of CH₄ uptake and N₂O emission in the soil environment, specifically accounting for varying N addition levels. This modeling approach was grounded in a thorough understanding of the underlying mechanisms. The pmoA gene, a component of MMO, played a crucial role in CH₄ oxidation, serving as a proxy for the methanotrophic activity (Wang et al. 2023a). NH₄⁺ served as a N source for the synthesis of MMO (Bodelier and Frenzel 1999), but also competed with CH₄ for MMO active sites (Acton and Baggs 2011; Zhang et al. 2020). Additionally, the nitrification process driven by NH₄⁺ resulted in the production of NO₃⁻ and N₂O (Deng et al. 2020; Gomez-Casanovas et al. 2016), further affecting the coupling between CH₄ uptake and N₂O emission. The NO₃⁻ participated in the denitrification process, which also produced N₂O (Long et al. 2017), and can affect methanotrophic activity (Whalen and Reeburgh 2000), influencing the CH₄–N₂O coupling. The protons generated during nitrification induced soil acidification, thereby decreasing pH and subsequently influencing the methanotrophic activity (Tian et al. 2016) and the abundance of the pmoA gene (Cai et al. 2016). Furthermore, soil acidification impacted the activity of nitrifying and denitrifying bacteria, which affected N₂O production (Prosser and Nicol 2012), ultimately altering the CH₄–N₂O coupling. Path coefficients were calculated to analyze the correlation between various factors. The logarithmic ratio and N gradient were plotted using RStudio (version 4.3.2). The correlation between the CR and variables such as pH, pmoA, NH₄⁺ and NO₃⁻ was analyzed using RStudio (version 4.3.2). The SEM analysis in this study was performed using SPSS 26.0 and AMOS 26.0.

RESULTS

Long-term N addition affected the relationship between CH₄ uptake and N₂O emission

During the growing season, CH₄ uptake exhibited a monthly variation, with the lowest value observed in May and the highest in July (Supplementary Table S1). The monthly variation of N₂O emission in response to N addition was not statistically significant (P > 0.05, Supplementary Table S1), with emission peaking in July and reaching its lowest in September (Supplementary Table S1). As N addition increased, CH₄ uptake initially increased, peaking at 4 g N m⁻² y⁻¹, and then decreased (Fig. 1a). Conversely, N₂O emission exhibited an exponential increase along the N gradient (Fig. 1b). At low-N addition levels (1–4 g N m⁻² y⁻¹), a highly significant positive correlation was observed between CH₄ uptake and N₂O emission (P < 0.01, Fig. 1c). However, at high-N addition levels (8–64 g N m⁻² y⁻¹), CH₄ uptake decreased by 33.62% (Supplementary Table S5), whereas N₂O emission increased by 264.91% (Supplementary Table S5). Notably, a significant negative correlation was observed between the two gases under these conditions (P < 0.05, Fig. 1c).

From 2013 to 2015, the coupling between CH₄ uptake and N₂O emission displayed significant interannual variation along the N addition gradient (P < 0.01, Supplementary Table S2, Fig. 2a). Within a specific range of low-N addition (1–4 g N m⁻² y⁻¹), the average value of the CR remained stable (Fig. 2c). However, the CR demonstrated significant interannual differences (P < 0.05, Supplementary Table S3). An analysis of standard deviation calculations revealed that the CR exhibited the highest similarity in 2015 (SD = 0.00264), followed by 2014 (SD = 0.00592), and the lowest similarity in 2013 (SD = 0.01331). This decreasing trend in standard deviation from 2013 to 2015 indicated an increasing similarity and strengthening of the synergistic relationship. Under high-N addition, the relationship between the CR and N followed a nonlinear pattern, gradually approaching a saturation state (Fig. 2b). The annual changes of the CR were 0.175 in 2013, 0.119 in 2014 and 0.088 in 2015. Compared with 2013, there was a 32% decrease in 2014 and a further 26.05% decrease in 2015 (Fig. 2d), suggesting a diminishing antagonistic relationship between CH₄ uptake and N₂O emission from 2013 to 2015.

Relationship between soil physical and chemical properties and the coupling between CH₄ uptake and N₂O emission

We further investigated the effects of N addition on soil physical and chemical properties. Our results indicated an increase in soil NH₄⁺ (Supplementary Fig. S1a), STN (Supplementary Fig. S1c), NO₃⁻ (Supplementary Fig. S1d) content and the copies of the pmoA gene (Supplementary Fig. S1e). However, a significant decrease in soil pH was observed (P < 0.05, Supplementary Fig. S1b). Interestingly, MBN showed a unimodal response to the N gradient, initially increasing at low-N levels, peaking at 8 g N m⁻² y⁻¹, and then rapidly declining, even falling below ambient levels (Supplementary Fig. S1f).

At low-N levels, NH₄⁺ was positively correlated with CH₄ uptake (R² = 0.215, P < 0.05, Supplementary Fig. S2a), while NO₃⁻ was positively correlated with N₂O emission (R² = 0.188, P < 0.05, Supplementary Fig. S2b). However, at high-N levels, NH₄⁺ exhibited a negative correlation with CH₄ uptake (R² = 0.02, P < 0.05, Supplementary Fig. S2c). Furthermore, a positive relationship was observed between N₂O emission and NH₄⁺ (R² = 0.256, P < 0.05, Supplementary Fig. S2d). NH₄⁺ was also positively correlated with NO₃⁻ (R² = 0.216, P < 0.05, Supplementary Fig. S2e). Notably, NO₃⁻ was negatively correlated with CH₄ uptake (R² = 0.163, P < 0.05, Supplementary Fig. S2f). As soil NH₄⁺ and NO₃⁻ concentrations increased and pH decreased, the regression slope of the correlation between CH₄ uptake and N₂O emission shifted from positive to negative (Supplementary Fig. S3a–c).

Under low-N conditions, the model demonstrated a significant negative influence of pH on CR (β = −0.22, Fig. 4a). Furthermore, a decrease in pH was found to significantly enhance the abundance of pmoA (β = −0.12, Fig. 4a), which in turn had a significant negative effect on CR (β = −0.19, Fig. 4a). Concurrently, NO₃⁻ exhibited a significant positive effect on CR (β = 0.18, Fig. 4a).

In contrast, under high-N conditions, the model demonstrated that NH₄⁺ had a significant positive influence on CR (β = 0.14, Fig. 4b). NH₄⁺ also had a substantial negative impact on pH (β = −0.42), resulting in a significant reduction in the abundance of pmoA (β = −0.45, Fig. 4b). Notably, despite its negative effect on pH, NH₄⁺ had a positive effect on pmoA abundance (β = 0.31), and this increased abundance significantly influenced CR (β = 0.51, Fig. 4b). Additionally, NO₃⁻ also exhibited a significant positive effect on CR (β = 0.15) and pmoA abundance (β = 0.18, Fig. 4b).

N-induced variation in CH₄ uptake and N₂O emission in other field experiments

Data collected from the semiarid grassland in Youyu, Shanxi, revealed a significant positive correlation (P < 0.05) between CH₄ uptake and N₂O emission across a N gradient ranging from 1 to 32 g N m⁻² y⁻¹ (Fig. 3a; Supplementary Table S3). The relationship between CH₄ uptake and N₂O emission in Bayinbuluk Grassland of Tianshan Mountain, Xinjiang, transitioned from synergistic to antagonistic with N increasing. A neutral N threshold of 9 g N m⁻² y⁻¹ was identified (Fig. 3b; Supplementary Table S3). In the meadow steppe of Erguna, a neutral N threshold of 10 g N m⁻² y⁻¹ was observed, followed by synergistic and antagonistic relationships between CH₄ uptake and N₂O emission (Supplementary Table S3) during 2020–2021. Notably, in 2021, CH₄ uptake increased significantly, while N₂O emission decreased (Fig. 3c). In a 3-year N addition experiment conducted in an alpine steppe on the Tibetan Plateau, a synergistic relationship was observed up to 4 g N m⁻² y⁻¹. However, when N exceeded 4 g N m⁻² y⁻¹, the relationship between CH₄ uptake and N₂O emission transitioned to antagonistic (Fig. 3d; Supplementary Table S3).

DISCUSSION

The impact of long-term N addition on the coupling between CH₄ uptake and N₂O emission

In this study, a positive coupling was observed between CH₄ uptake and N₂O emission at low-N levels, whereas a negative coupling was noted at high-N levels. These findings aligned with previous research conducted on the Tibetan Plateau (Peng et al. 2018, 2019). Specifically, in the Bayinbuluk grasslands of the Tianshan Mountains in Xinjiang, an N threshold of 9 g N m⁻² y⁻¹ was identified (Supplementary Table S4). This N threshold exceeded the one found in our studies, which was 4 g N m⁻² y⁻¹. Notably, our 3-year data revealed a decreasing antagonism and an emerging synergy with increasing N application (Fig. 2a). However, this trend was not observed in earlier studies that primarily focused on individual gas fluctuations (Chen et al. 2021; Li et al. 2012; Shi et al. 2022). It was important to clarify that the observed variation in the coupling with different N levels does not directly suggest that CH₄ influences N₂O or vice versa. Rather, these changes reflected the impaction of N on soil conditions, which in turn affected both processes (uptake and emission), indirectly modifying their coupling.

At low-N levels, N addition enhanced CH₄ uptake by stimulating the synthesis of MMO (Supplementary Fig. S1e), thus increasing CH₄ oxidation. Furthermore, low-N addition augmented N₂O emission by providing additional NO₃⁻ as substrates for denitrification (Supplementary Fig. S3b). A combined effect of CH₄ uptake and N₂O emission at low-N levels resulted in synergistic interaction. However, after 3 years of continuous N addition, CH₄ uptake gradually declined (Supplementary Table S5), which is consistent with previous findings (Chen et al. 2019; Shi et al. 2022). This decline may be attributed to soil acidification caused by N deposition, which inhibited methanotrophs and their CH₄ oxidize ability (Tian et al. 2016). Over the 3-year period, N₂O emission showed an increasing trend (Supplementary Table S5), likely due to enhanced nitrification and denitrification processes. Despite the decline in CH₄ uptake, N₂O emission remained high (Supplementary Table S5), thereby further strengthening the synergistic relationship (Fig. 2a). However, at low-N levels, CH₄ uptake consistently increased annually, while N₂O emission remained stable (Fig. 3d), indicating a gradual intensification of their synergistic interaction on the Tibetan Plateau (Fig. 3d). This suggested that there were variations in the coupling patterns between different grassland types.

At high-N levels, the competitive inhibition of MMO by elevated concentrations of NH₄⁺ and CH₄ reduced the oxidation of CH₄, thereby inhibiting CH₄ uptake (Supplementary Fig. S2c). Similar to the response observed at low-N levels, higher concentrations of NH₄⁺ and NO₃⁻ at high-N levels accelerated nitrification and denitrification processes (Supplementary Fig. S1a and c), resulting in greater N₂O emission (Supplementary Fig. S2d). The contrasting effect of inhibition–CH₄ uptake and promotion–N₂O emission under high-N levels created an antagonistic relationship. However, after 3 years of continuous N addition, the inhibition of CH₄ uptake was mitigated (Supplementary Table S5). Long-term N application induced soil acidification, modulating metal ion availability and stimulating MMO activity, which subsequently enhanced CH₄ oxidation (Zhang et al. 2023). Despite the increase in NH₄⁺ concentrations over 3 years of high-N inputs, N₂O emission decreased in 2015 (Supplementary Table S6). An unexpected decline in N₂O emission was also observed in semiarid grasslands of Shanxi under high-N conditions (Fig. 3a). Contrary to the prevailing notion that high-N levels consistently increase N₂O emission (Shcherbak et al. 2014), our findings indicated the existence of a saturation point, where continuous N inputs may inhibit microbial N₂O production processes or alter microbial community composition, reducing N₂O production. Furthermore, elevated N levels can increase NH₄⁺ uptake by plants, reducing the substrate availability for nitrifying and denitrifying pathways (Xu et al. 2012). Although the inhibition of CH₄ uptake was mitigated, the decrease in N₂O emission weakened the antagonistic relationship between these two processes (Fig. 2b and c). In contrast, CH₄ uptake increased annually at high-N levels in alpine meadows on the Tibetan Plateau. Similarly, N₂O emission also increased (Fig. 3d), resulting in an initial intensification and subsequent diminishment of the antagonistic relationship between these two processes (Supplementary Table S4). When analyzing the coupling between the CH₄ uptake and N₂O emission with N, the accumulated effect over time was crucial. The short-term effects may differ from long-term effects. Our study highlighted that the coupling may vary, fluctuating with prolonged N inputs, indicating a more dynamic and complex ecosystem nutrient input than previously recognized.

Microbial mechanisms and substrate dynamics influencing the coupling CH₄ uptake and N₂O emission

Our study aimed to elucidate the coupling dynamics between CH₄ uptake and N₂O emission at varying N levels, particularly examining how these interactions are influenced by substrate availability and microbial response mechanisms. Our findings revealed an intricate balance between soil acidification, methanotrophic activity and nitrification/denitrification processes, which jointly influence greenhouse gas fluxes. At low-N levels, soil acidification induced by N input enhanced the abundance of the pmoA gene, signifying an increase in methanotrophic activity (Fig. 4a). This increase in activity translated into greater CH₄ uptake. Furthermore, NO₃⁻ had a significant positive impact on CR, emphasizing its role in stimulating N₂O emission via the denitrification pathway (Fig. 4a). Consequently, a synergistic relationship was observed under low-N conditions, highlighting the intricate interaction between CH₄ oxidation and N₂O emission.

Conversely, at high-N levels, the coupling relationship between CH₄ uptake and N₂O emission shifted to an antagonistic state. Initially, NH₄⁺ promoted pmoA abundance and enhanced methanotrophic activity. However, the competition between NH₄⁺ and CH₄ subsequently reduced the efficiency of CH₄ uptake (Fig. 4b). The competition between NH₄⁺ and CH₄, coupled with the stimulatory effects of NH₄⁺ on CR, suggested that NH₄⁺ inhibited CH₄ uptake while simultaneously enhancing N₂O emission via nitrification (Fig. 4b). The stimulatory effect of NO₃⁻ on pmoA indicated that NO₃⁻ did not inhibit methanotrophic activity (Fig. 4b). Additionally, the stimulatory effect of NO₃⁻ on CR implied that the contribution of nitrification to N₂O emission increased, denitrification remained a significant contributor to sustaining high N₂O emission levels (Fig. 4b). We developed a conceptual model that integrating the findings of the SEM to elucidate the transitional patterns in the coupling relationship between CH₄ uptake and N₂O emission across varying N levels (Fig. 5). The model emphasized the pivotal roles of methanotrophs and ammonia-oxidizing bacteria in mediating the couplings. At low-N levels, NH₄⁺ was preferentially utilized for CH₄ oxidation enzyme synthesis, rather than nitrification, potentially due to the ecological niches and metabolic preferences of the respective microbial communities (Wang et al. 2023b). This reduced oxidative competition between NH₄⁺ and CH₄ facilitated more efficient CH₄ oxidation by methanotrophs (Fig. 5). However, N₂O emission was primarily driven by denitrification (Fig. 5). At high-N levels, the competitive inhibitory effect of NH₄⁺ on CH₄ uptake, consistent with the findings of this study, has been extensively documented (Zhang et al. 2020). Concurrently, as NH₄⁺ availability increased, denitrification remained robust, while nitrification intensified, resulting in elevated N₂O emission (Fig. 5). However, while methanotrophic activity intensified in acidic soils, CH₄ oxidation was diminished. Conversely, ammonia-oxidizing bacteria thrived in acidic soil conditions, increasing N₂O production. The activities of nitrifying and denitrifying bacteria, which determined the relative contributions of nitrification and denitrification to N₂O production, vary in accordance with soil N content (Han et al. 2024). For instance, in forest soils, the nitrification rate escalated by almost 100-fold with an increase in soil N content (Ollinger et al. 2002). Our conceptual model addressed the previously unresolved question concerning the relative importance of nitrification and denitrification in modulating N₂O emission across various N levels in grasslands.

Our study elucidated the microbial mechanisms that governed the N-induced transformation of the coupling between CH₄ uptake and N₂O emission. Further research is necessary to comprehensively understand the relationship between microbial functional groups and soil C and N flux dynamics. Future research should prioritize investigating the molecular-level impacts of N on the activity and community composition of methanotrophs and ammonia-oxidizing bacteria, aiming to gain a comprehensive understanding of the coupling between CH₄ uptake and N₂O emission in semiarid grasslands. Extending these analyses to a global scale to explore the coupling in diverse ecosystems and estimate greenhouse gas fluxes under global warming scenarios would be immensely valuable.

CONCLUSIONS

This study investigated the complex interplay between CH₄ uptake and N₂O emission in grassland ecosystems, particularly focusing on the effects of varying N levels. Our study differed from traditional research, which analyzed CH₄ and N₂O fluxes in isolation, by highlighting the coupled dynamics between these two gases, modulated by temporal variations in N. We identified NH₄⁺, NO₃⁻ and pH as crucial factors influencing this relationship, noting a significant shift at an N threshold of 4 g m⁻² y⁻¹. At low-N levels, the alleviation of N limitation in methanotrophs facilitated a synergistic interaction, resulting in enhanced CH₄ uptake. Despite lower pmoA copy numbers, the higher CH₄ uptake was attributed to increased MBN, leading to enhanced CH₄ monooxygenase synthesis and reduced NH₄⁺ competition. Conversely, at high-N levels, the increase in pmoA copy numbers did not correlate with higher CH₄ uptake primarily due to intense NH₄⁺ competition, which fostered substrate competition. The primary determinants of these dynamics are the dominance of denitrification at low-N levels and nitrification at high-N levels. Persistent low-N inputs may amplify variations in these interactions, whereas high-N inputs may attenuate them. These insights are crucial for predicting C and N cycling in the context of global change. However, caution should be exercised when extrapolating these results, and further research is imperative to validate the mechanisms elucidated in this study.

Supplementary Material

Supplementary material is available at Journal of Plant Ecology online.

Figure S1: N addition impacts on soil NH₄⁺ (a), pH (b), STN (c), NO₃⁻ (d), pmoA copies (e) and MBN (f).

Figure S2: The relationship between NH₄⁺, NO₃⁻ and CH₄ uptake and N₂O emission under different N treatments.

Figure S3: The relationship between NH₄⁺, NO₃⁻, pH and regression slope of CH₄ uptake and N₂O emission in the field experiment.

Table S1: Monthly variation of CH₄ and N₂O fluxes in response to N increase.

Table S2: ANOVA analysis of N variability in CR.

Table S3: ANOVA analysis of yearly variability in CR.

Table S4: Regression analysis of CH₄ uptake and N₂O emission across various locations and N addition levels.

Table S5: The impact of N additions and years on CH₄ uptake and N₂O emission.

Table S6: Fluxes of CH₄ uptake and N₂O emission, as well as fluctuations in NH₄⁺, NO₃⁻ and pH, between 2013 and 2015.

Funding

This study was partially supported by the National Natural Science Foundation of China (32271681) and by the Key Laboratory of Ecology and Environment in Minority Areas (Minzu University of China), National Ethnic Affairs Commission (KLEEMA202206).

Authors’ Contributions

Lihua Zhang devised the project, collected data, developed the research questions, and revised manuscript. Yiming Ren collected and analyzed data, developed tables and figures, written the draft manuscript. Yangxinzi Zao, Ying Zhao, Rui Su, Guowei Yang, Xiran Li, Jingru Kang, Yiyu Shi, Yuru Xie, Nannan Wang, Yunjiang Zuo, Kexin Li, Liyuan He, Xiaofeng Xu contributed to the manuscript editing.

Conflict of interest statement. The authors declare that they have no conflict of interest.

REFERENCES