Introduction

The enzyme N2O reductase (EC 1.7.99.6) plays a critical role in the regulation of soil N2O fluxes since it catalyzes the reduction of N2O to inert N2 (Simarmata et al., 1993; Nannipieri and Paul 2009; Richardson et al. 2009). Previous studies have shown that laboratory in vitro nitrate (NO3 −) amendment stimulates N2O reduction at low NO3 − concentration (5 mg N kg−1), but inhibits N2O reduction at high NO3 − concentrations (> 50 mg N kg−1) (Blackmer and Bremner 1978; Firestone et al. 1979; Senbayram et al. 2012).

Long-term excessive N fertilization has previously been reported to increase the N2O/(N2 + N2O) product ratio due to concurrent soil acidification (Qu et al. 2014). In addition to soil acidification, inhibition of NO3 − on N2O reductase activity may, theoretically, also reduce the transformation of N2O to N2, thereby increasing the N2O/(N2 + N2O) product ratio in soils (Blackmer and Bremner 1978). Although the inhibiting effects of NO3 − on N2O reduction were confirmed in vitro almost 40 years ago, it remains unclear how long-term N fertilization affects soil N2O reductase activity in situ and if the N2O/(N2 + N2O) product ratio is affected.

Determining potential N2O reductase activity and the N2O/(N2 + N2O) product ratio is a challenge due to the high background concentration of atmospheric N2 (Groffman et al. 2006). However, a system for the measurement of soil potential N2O reductase activity based on a direct N2 measurement method was previously developed (Qin et al. 2014). It was observed that N2O reductase activity tended to be lower in soils under a N fertilization rate of 400 kg N ha−1 year−1 than under a non-fertilization control (Qin et al. 2014). The current study demonstrates the effects of soil NO3 − concentration on the in situ temporal trends in potential N2O reductase activity and the N2O/(N2 + N2O) product ratio, in soil under differing N fertilization regimes, over a 2-year period. It was hypothesized that N fertilization would inhibit potential N2O reductase activity and result in a higher N2O/(N2 + N2O) product ratio.

Materials and methods

A long-term N fertilization field experiment has been conducted since 1997 at the Luancheng experimental station (37.90° N, 114.67° E), as previously described (Qin et al. 2014). In this current study, the 400 kg urea-N ha−1 year−1 (N400) and the non-fertilizer control (CK) treatments were compared. There were three replicates of each treatment. The cropping system comprises winter wheat (Triticum aestivum)/summer maize (Zea mays) in an annual double cropping rotation. A quarter of the annual fertilizer rate was applied at wheat sowing, another quarter at wheat jointing stage, and the remainder at the maize tasseling stage.

Each month for 2 years, five surface soil cores (0–20 cm) were randomly collected from each replicate, immediately mixed, and stored at 4 °C for further analyses. Soil collection was conducted at least 1 week after fertilizer application.

Soil potential N2O reductase activity was determined by the direct N2 measurement method (Qin et al. 2014). Field moist soil (10.0 g oven dry weight equivalent, 2-mm-sieved), with addition of 10 ml of 0.2 M sodium phosphate solution, was anaerobically incubated in a 120-ml flask for 1 h under optimal conditions. The N2 levels in the flask were analyzed by a robotized system to determine N2O reductase activity. The principles of this system have been previously described in detail (Molstad et al. 2007).

Soil N2 and N2O emissions were determined as follows: 20 g of dry weight equivalent of field moist soil were added into a 120-ml serum flask. The flask was sealed with a liquid polytetrafluoroethylene-coated butyl rubber septa and aluminum cap. Then, the headspace gas in the flask was repeatedly evacuated and filled with artificial N2-free air (79% helium and 21% oxygen), five times, with the pressure of the headspace gas adjusted to 101.3 kPa after the final fill. Then, the flasks were incubated at 25 °C for 5 days. A control treatment (flasks with autoclaved soil) was simultaneously incubated and used to evaluate the airtightness of the incubation system. The headspace N2O and N2 concentrations were determined at 24-h intervals using the robotized sampling and analyzing system (Qin et al. 2014). Soil N2O and N2 emission rates were calculated using the slopes of linear regressions derived from the changes in N2O or N2 concentration over time. Soil denitrification rate was calculated as the total of the N2O and N2 product rates.

All statistical analyses were performed using SPSS for Windows (Version 19.0, SPSS Inc., Chicago, IL, USA). Pearson’s correlation was used to investigate the correlations between soil denitrification rate and the N2O/(N2O + N2) product ratio and other environmental parameters.

Results and discussion

Compared with the CK, the N400 treatment significantly increased the denitrification rate and N2O/(N2 + N2O) product ratio (Figs. 1e and 2c). These results are consistent with those of a previous study showing that N fertilization increased denitrification rates at this site (Zhang et al. 2009). Soil denitrification rate and the N2O/(N2 + N2O) product ratio are known to be regulated by soil pH, water content, dissolved organic C (DOC), and NO3 − contents (De Rosa et al. 2016; Liu et al. 2017; Wang et al. 2017). In this study, no consistent effects of N fertilizer rates were observed on soil DOC, pH, and water content (Fig. 1b–d). It has been reported that excessive fertilization can result in a significant decrease in soil pH (Malhi et al. 1998). The lack of such an effect in the current study may result from the fact that the soil is calcareous soil with a high pH buffer capacity (Roelcke et al. 1996).

Fig. 1
figure 1

Monthly dynamics of soil NH4 + (a), water (b), pH (c), and DOC  concentration (d) and Denitrification rate (e). Data are presented as mean ± SE (n = 3). The results of multi-way ANOVA on the effects of fertilization, sampling date, and their interactions were listed in the middle–upper part of each graph. Asterisks and NS indicate statistical significance at p < 0.05 and non-significant difference, respectively

Full size image
Fig. 2
figure 2

Monthly dynamics of nitrate concentration (a), N2O reductase activity (b), and N2O/(N2O + N2) product ratio (c) in the soil under treatment of CK and N400. Data are presented as mean ± SE (n = 3). The results of multi-way ANOVA on the effects of fertilization, sampling date, and their interactions were listed in the middle–upper part of each graph. Asterisks and NS indicate statistical significance at p < 0.05 and non-significant difference, respectively

Full size image

The soil NO3 − concentrations under the N400 treatment (range from 17 to 58 mg NO3 −-N kg−1 soil) were significantly higher than those under the CK treatment (range from 5 to 11 mg NO3 −-N kg−1 soil) (Fig. 2b). This result is reasonable since a N fertilization rate of 400 kg N ha−1 year−1 has been reported to exceed the crop N requirement in this station (Qin et al. 2012). In vitro, NO3 − concentrations > 50 mg NO3 −-N kg−1 have been reported to inhibit soil N2O reductase activity (Blackmer and Bremner 1978; Firestone et al. 1979; Senbayram et al. 2012). In this study, the N2O reductase activity was significantly lower under N400 treatment than that under the CK treatment (Fig. 2a). The mean soil NO3 − concentration under the N400 treatment (30 mg NO3 −-N kg−1) was lower than the in vitro threshold (50 mg NO3 −-N kg−1 soil) previously reported to inhibit N2O reduction (Blackmer and Bremner 1978). However, extremely high NO3 − concentrations (> 100 mg NO3 −-N kg−1 soil) immediately following fertilization events have been previously reported at this site (Dong et al. 2009). The soil under the N400 in this study could have frequently suffered from high NO3 − concentrations a number of times during the 15-year field experiment. These results show for the first time that in situ long-term excessive N inputs supress N2O reductase activity.

Peaks in NO3 − concentration and the N2O/(N2 + N2O) product ratio and peaks in N2O reductase activity were synchronous following fertilization and irrigation events in the N400 treatment (Fig. 2). The N2O/(N2 + N2O) product ratio in the N400 treatment was positively correlated with the NO3 − concentration (Y = 1.52–0.95 X, where Y is N2O/(N2 + N2O) product ratio and X is NO3 − concentration) and the N2O reductase activity was negatively correlated with the NO3 − concentration (Y = 44.52–0.78 X, where Y is N2O reductase activity and X is NO3 − concentration) (Table 1). These long-term in situ results demonstrate the role soil NO3 − concentration plays in affecting soil N2O/(N2 + N2O) product ratios via regulating N2O reductase activity.

Table 1 Pearson correlations between soil N2O reductase activity (RA), N2O/(N2 + N2O) product ratio (PR), denitrification rate (DR), nitrate concentration (NO3 −), dissolved organic C concentration (DOC), water content (SWC), 5-cm soil temperature (ST), and ammonium concentration (NH4 +)
Full size table

This study confirms the previous in vitro observations: NO3 − addition inhibits N2O reductase activity and results in high N2O/(N2 + N2O) product ratio from soils (Blackmer and Bremner 1978; Firestone et al. 1979; Senbayram et al. 2012). Excessive N fertilization that promotes soil NO3 − accumulation will enhance N2O emissions not only by increasing total denitrification rates, but also by increasing the N2O/(N2O + N2) product ratio via inhibiting N2O reductase activity. These results indicate that strategies avoiding excess soil NO3 −, e.g., matching plant N demand and supply (N rate) and retaining NH4 + − N by using nitrification inhibitors, have the potential to mitigate N2O emissions from denitrification.