Introduction

Methane (CH4) is an important greenhouse gas, contributing approximately 20% to the anthropogenic climate warming (IPCC 2013). Grasslands cover more than 20% of the Earth’s land surface and typically serve as a net sink for atmospheric CH4(Aronson & Helliker 2010; Chen et al. 2011; Dijkstra et al. 2013). However, this CH4 sink may be altered by the growing input of nutrients, particularly nitrogen (N) (Galloway et al. 2008), due to anthropogenic activities. While previous studies have reported various impacts of N (particularly NH4+) addition on the oxidation of atmospheric CH4 in soil (Bollag & Czlonkowski 1973; Liebig et al. 2008), it has become clear that N addition typically suppresses this oxidation process, with few exceptions (Aronson & Helliker 2010; Liu et al. 2017). This N-induced suppression of CH4 oxidation has been attributed to two mechanisms. First, CH4 and NH4+ compete for the same methane monooxygenase (MMO) enzyme (Gulledge et al. 2004), which can oxidize both CH4 (to CH3OH) and NH4+ (to NO2-) due to the similar molecular structure of CH4 and NH4+ (Dunfield & Knowles 1995; Gulledge et al. 2004). Second, the intermediates and end products (primarily nitrite) of methanotrophic NH4+ oxidation can be toxic to methanotrophic bacteria, explaining the inhibition of CH4 consumption under increased N inputs (Schnell & King 1994; Bodelier & Laanbroek 2004). According to the substrate competition mechanism, added NH4+reduces the amount of CH4 consumed by methanotrophic bacteria, based on enzyme kinetics (Davidson & Schimel 1995; Singh & Strong 2016), and inhibits the growth of methanotrophic bacteria as oxidizing NH4+ does not support CO2 fixation for cell growth (Bédard & Knowles 1989; Carlsen et al. 1991; Gulledge & Schimel 1998), further suppressing the CH4 oxidation (Gulledge & Schimel 1998).
While the negative impact of N addition is thus well understood, phosphorus (P) addition may alter this N-induced suppression on CH4 oxidation. In a recent study, P addition alone was reported to enhance CH4 oxidation (Veraart et al.2015). Given that P-fertilizer inputs and atmospheric P deposition are also increasing globally (Mahowald et al. 2008; Penuelas et al. 2012; Penuelas et al. 2013), investigating P impacts on CH4 uptake rate, and especially its interactions with the increasing N inputs, is important for better understanding the role that aerobic soils play in the global atmospheric CH4cycle.
Atmospheric deposition and fertilizer inputs of N and P have increased across the globe since the industrial revolution (Mahowald et al.2008; Blankinship et al. 2010; Penuelas et al. 2013; Brahney et al. 2015; Wang et al. 2017). Increased inputs of N may reduce the CH4 sink strength of grassland ecosystems, with potentially important feedbacks to the global climate system (Templer et al. 2012; Tian et al. 2015). The interactive effect of N + P additions on CH4 oxidation has been reported as neutral (Lund et al. 2009) or positive (Zhang et al. 2014), and P has been found to alleviate the N suppression on CH4 oxidation (Zhang et al. 2014); however, the mechanisms by which P alters the N suppression of atmospheric CH4 oxidation in soils remain elusive. We therefore integrated a 4-year field manipulation experiment, a meta-analysis, and an empirical modeling approach to investigate the effects of N + P additions on CH4 flux in grasslands and the underlying mechanisms.