Effects of soil type and nitrate concentration on denitrification products (N₂O and N₂) under flooded conditions in laboratory microcosms

Abstract

Denitrification products nitrous oxide ((N₂O) and nitrogen (N₂)) were measured in three flooded soils (paddy soil from Vietnam, PV; mangrove soil from Vietnam, MV; paddy soil from Japan, PJ) with different nitrate (NO₃^–) concentrations. Closed incubation experiments were conducted in 100-mL bottles for 7 d at 25°C. Each bottle contained 2 g of air-dried soil and 25 mL solution with NO₃^– (concentration 0, 5 or 10 mg N L⁻¹) with or without acetylene (C₂H₂). The N₂O + N₂ emissions were estimated by the C₂H₂ inhibition method. Results showed that N₂O + N₂ emissions for 7 d were positively correlated with those of NO₃^– removal from solution with C₂H₂ (R² = 0.9872), indicating that most removed NO₃^– was transformed to N₂O and N₂ by denitrification. In PJ soil, N₂O and N₂ emissions were increased significantly (P < 0.05) by the addition of greater NO₃^– concentrations. However, N₂O and N₂ emissions from PV and MV soils were increased by the addition of 0 to 5 mg N L⁻¹, but not by 5 to 10 mg N L⁻¹. At 10 mg N L⁻¹, N₂ emissions for 7 d were greater in PJ soil (pH 7.0) than in PV (pH 5.8) or MV (pH 4.3) soils, while N₂O emissions were higher in PV and MV soils than in PJ soil. In MV soil, N₂O was the main product throughout the experiment. In conclusion, NO₃^– concentration and soil pH affected N₂O and N₂ emissions from three flooded soils.

Key words:

• acetylene inhibition method
• denitrification
• flooded soil
• nitrate removals
• nitrous oxide

INTRODUCTION

Denitrification is a microbial reaction in terrestrial and aquatic ecosystems in which nitrate (NO₃^–) is used as an alternative electron acceptor when oxygen is limited (Philippot et al. 2007). Denitrification is responsible for the emission of nitrous oxide (N₂O) as an intermediate product. Nitrous oxide is a greenhouse gas and contributes to depletion of the stratospheric ozone layer (Conrad 1996). The Intergovernmental Panel on Climate Change (IPCC) reported that direct emission of N₂O from agricultural soils has increased over the last few decades, in parallel with increasing use of fertilizers. In 2005, N₂O emission from agriculture soils accounted for 60% of global anthropogenic fluxes (Smith et al. 2007).

Denitrification requires a substrate of NO₃^– as an electron acceptor (Buresh and De Datta 1990; Weier et al. 1993; Cheng et al. 2004); available organic carbon (C) as an electron donor (Weier et al. 1993; Chiu et al. 2004); the absence of oxygen, which is related to soil moisture contents at > 60% water-filled pore space (WFPS; Weier et al. 1993); suitable soil temperature, which is optimal at 25°C (Canion et al. 2014); and suitable soil pH, which ranges from 3.8 to 8.2 (optimum around pH 7; Šimek and Cooper 2002).

Denitrification is expected to be the main process of nitrogen (N) transformation in areas where soil and sediments contain sufficient organic C under anaerobic conditions and receive water with a high NO₃^– concentration. These areas include flooded rice fields (Reddy and Patrick 1986; Buresh and De Datta 1990), riparian wetlands (Hefting et al. 2004; Van den Heuvel 2010) and mangrove forests (Chiu et al. 2004; Fernandes and Loka 2011). In such flooded soils, the main products of denitrification (N₂O or N₂) were dependent on environmental conditions. In flooded rice soils, N₂ was mainly produced by denitrification, because N₂O was further reduced to N₂ under anaerobic conditions (Huang et al. 2007; Iida et al. 2007). In contrast, in Tuvem mangrove sediments (Goa, India), incomplete denitrification was responsible for up to 43–93% of N₂O production (Fernandes et al. 2010).

To our knowledge, the factors affecting denitrification end products (N₂O and N₂) in flooded soils remain uncertain. We supposed that different physical and chemical properties of soil under the same flooded condition affect denitrification products. Our study was conducted in laboratory microcosms to investigate the emissions of N₂O and N₂ from three soils (two paddy soils and a mangrove soil) under flooded conditions with different NO₃^– concentrations (0, 5 or 10 mg N L⁻¹) in standing water. The objective of the study was to identify the controlling factors of N₂O and N₂ emissions from soils of different types under flooded conditions.

MATERIALS AND METHODS

Soil sampling sites

Soil samples were collected in May 2012 from lowlands, where soils are usually flooded: a paddy field in the Thanh Trung hamlet, Vietnam (PV; 16°31ʹ02”N, 107°30ʹ10”E); a mangrove forest located in the Huong Phong commune, Vietnam (MV; 16°33ʹ28”N, 107°36ʹ35”E); and a paddy field in Ushimado, Japan (PJ; 34°37ʹ22”N, 134°07ʹ40”E). The sampling sites were surrounded by agricultural fields. The PV, MV and PJ soils were classified according to the Food and Agriculture Organization (FAO) as Dystric Fluvisols, Gleyi-Salic Fluvisols and Dystric Gleysols, respectively. These flooded soils are usually under anaerobic conditions and rich in organic matter, which are favorable for denitrification. These areas would be expected to reduce the NO₃^– concentration of water entering from the agriculture fields by denitrification.

Soil properties

Soil samples were collected from the 0–20 cm layer and air-dried at room temperature, then passed through a 2-mm mesh sieve. All soil samples were subjected to measurements of pH and electrical conductivity (EC) at a soil:water ratio of 1:5 after shaking for 1 h at 175 rpm, using a digital pH meter (F-23, Horiba, Japan) and EC meter (DS-14, Horiba, Japan), respectively. Total carbon (TC) and total nitrogen (TN) contents of soil were determined using a CN Corder (MT-700, Yanaco, Japan). Mineral N was extracted with 2 M potassium chloride (KCl) solution (soil:solution ratio of 1:10). The ammonium (NH₄⁺-N) and NO₃^–-N contents in extracts were measured by spectrophotometry using a continuous flow analyzer (QuAAtro 2-HR, Bltec, Japan). Dissolved organic carbon (DOC) was measured with a total organic carbon (TOC) analyzer (TOC-5000A, Shimadzu, Japan) at a soil:water ratio of 1:10. Cation exchange capacity (CEC) was determined with neutral ammonium acetate (NH₄OAc; Chapman 1965). Soil texture was determined by the pipette method (Sarkar and Haldar 2005). Denitrifying bacteria in soils were estimated by the most probable number (MPN) method with Giltay’s medium (Ishikuri 2011).

Soils properties were as follows: TN, TC, DOC and CEC in the two paddy soils (PV and PJ) were higher than in the MV soil (Table 1). PJ soil was neutral while PV soil was moderately acid and MV soil was extremely acid. The density of denitrifying bacteria was the highest in PJ soil (2.7 × 10⁴ MPN g⁻¹), approximately 18 and 23 times higher than in PV and MV soils, respectively. Nitrate was not detected in all soil samples. The NH₄⁺-N content was highest in the PV soil at 30.1 mg N kg⁻¹, followed by those in MV and PJ soils.

Table 1 Selected properties of soils used.

Parameter	PV	MV	PJ
pH (1:5)	5.8	4.3	7.0
EC (mS m^−1)	5.9	77.0	7.1
Particle size distribution(%)
Clay	22	16	7
Silt	42	25	16
Sand	36	59	77
NO3^–-N (mg kg^−1)	n.d.	n.d.	n.d.
NH4^+-N (mg kg^−1)	30.1	12.3	1.4
DOC (mg kg^−1)	102.5	83.8	97.5
TC (g kg^−1)	10.4	6.3	7.9
TN (g kg^−1)	1.2	0.6	0.8
C:N ratio	8.9	9.9	10.4
CEC (cmolc kg^−1)	6.8	5.6	8.0
Density of denitrifying bacteria (MPN g^−1)	0.14 × 10^4	0.12 × 10^4	2.7 × 10^4

PV = paddy soil from Vietnam; MV = mangrove soil from Vietnam; PJ = paddy soil from Ushimado, Japan; n.d. = not detected. The density of denitrifying bacteria was measured by the most probable number method. EC, electrical conductivity; NH₄⁺-N, ammonium-nitrogen; NO₃^–-N, nitrate-nitrogen; DOC, dissolved organic carbon; TC, total carbon; TN, total nitrogen; CEC, cation exchange capacity.

Incubation experiment

Denitrification is a series of processes that transforms NO₃^– and nitrite (NO₂^–) to gaseous forms of NO, N₂O, or N₂. The acetylene inhibition method was used for estimating denitrification (Zaman et al. 2012). Because C₂H₂ inhibits the reduction of N₂O to N₂, we determined the denitrification rates as N₂O production.

To prepare the C₂H₂-saturated solutions, C₂H₂ gas was passed through three Pyrex gas-washing bottles which contained 0.5 M sulfuric acid (H₂SO₄), deionized water and a certain concentration of NO₃^– solution (0, 5 or 10 mg N L⁻¹), with a flow rate of 100 mL min⁻¹ for 30 min. For three soil samples, closed incubation experiments were conducted in a 100-mL bottle at 25°C for 7 d under flooded conditions. Each bottle contained 2 g of air-dried soil and 25 mL of NO₃^– solution (0, 5 and 10 mg N L⁻¹ in a solution are equivalent to 0, 63.4 and 126.5 mg N kg⁻¹ oven-dried soil, respectively) with or without C₂H₂. All treatments were conducted in triplicate. A 0.5-mL gas sample was collected after 3 h on day 0 and days 1, 3 and 7 and analyzed for N₂O using a gas chromatograph (GC-8A, Shimadzu, Japan) equipped with a ⁶³Ni electron capture detector. Differences in N₂O emissions during 7 d incubation with and without C₂H₂ were considered to be the amount of N₂ emission during denitrification. The N₂O and N₂ emissions were calculated by the equation below on the weight basis of air-dried soil (mg N kg⁻¹; Sun et al. 2012):

(1)

where V_N2O is the N₂O emission (mg N kg⁻¹), V₁ and V₂ are the N₂O emissions with and without C₂H₂, respectively, ρ is the density of N₂O (1.25 kg m⁻³), C_g is the concentration of N₂O in the gas phase (ppm_v), V_g is the volume of head space (m³), V_L is the volume of the liquid phase (m³), α is the Bunsen absorption coefficient (0.544 at 25°C), W is the dry soil weight (kg) and T is the temperature at sampling (25°C).

On day 7 of incubation, the butyl septum on the bottle was opened. The solution was taken out to analyze the NO₃^– concentration. Nitrate removal was obtained from the difference between NO₃^– concentrations on days 0 and 7.

Statistical analysis

Data was expressed as a mean ± standard deviation (SD). An analysis of variance (ANOVA) was performed to examine the effects of NO₃^– concentrations and soil types on N₂O and N₂ emissions at a significance level of 0.05. Linear regressions were analyzed to identify the relationship between N₂O + N₂ emissions and NO₃^– removal.

RESULTS AND DISCUSSION

Effect of NO₃^– concentrations on N₂O and N₂ emissions

At 25°C, N₂O and N₂ emissions in 63.4 and 126.5 mg N kg⁻¹ NO₃^– treatments were much greater than those in treatments without NO₃^– (Fig. 1). Addition of NO₃^– significantly enhanced denitrification and emissions of N₂O and N₂ (Weier et al. 1993; Cheng et al. 2004). The effect of NO₃^– concentrations (63.4 and 126.5 mg N kg⁻¹) on N₂O and N₂ emissions were significantly observed only in PJ soil (pH 7), while this effect was not clear in PV and MV soils (P > 0.05, Fig. 1). In PJ soil, N₂O and N₂ emissions increased with increasing concentration of NO₃^–. Denitrification rates depend more on the bacterial population (Philippot and Hallin 2005). The populations of denitrifying bacteria in PV and MV soils were smaller than in PJ soil (Table 1). Accordingly, PV and MV soils transformed less NO₃^– to N₂O and N₂ at higher NO₃^– concentrations than PJ soil (Fig. 1; Table 2). Nitrate removal from solutions with C₂H₂ were positively correlated with those of denitrification in three flooded soils (R² = 0.9872; Fig. 2). Most removed NO₃^–-N was transformed to N₂O and N₂ by denitrification within 7 d because the slope of the regression line was nearly equal to 1 (Fig. 2).

Table 2 Change of NO₃^– (mg N kg⁻¹)^† after 7 d incubation with acetylene at 25°C.

NO3^–	PV			MV			PJ
mg N kg^–1	Day 0	Day 7	NO3^– removal	Day 0	Day 7	NO3^– removal	Day 0	Day 7	NO3^– removal
0.0	0.0	0.0 ± 0.0	0.0	0.0	0.0 ± 0.0	0.0	0.0	0.0 ± 0.0	0.0
63.4	65.9	7.7 ± 8.1	58.2	66.5	62.8 ± 0.9	3.7	66.1	7.3 ± 4.4	58.8
126.5	131.4	78.9 ± 14.4	52.5	135.1	127.6 ± 0.4	7.5	132.0	35.9 ± 12.3	96.1

Data are the mean ± standard deviation, n = 3. ^†NO₃^– concentrations (mg N L^–1) in solutions were converted to NO₃^– (mg N kg^–1). Nitrate removal, difference in NO₃^– concentrations in solution between days 0 and 7; PV, paddy soil from Vietnam; MV, mangrove soil from Vietnam; PJ, paddy soil from Japan.

Figure 1 Accumulated nitrous oxide (N₂O) emissions (mg N kg^–1) in (a) PV, (b) MV, and (c) PJ soils and nitrogen (N₂) emissions (mg N kg^–1) in (d) PV, (e) MV, and (f) PJ soils during 7 d incubation at 25°C. Bars indicate standard deviation, n = 3. Note: N₂O emission (mg N kg^–1) was measured in the incubation bottle without acetylene (C₂H₂). N₂ emission (mg N kg^–1) was calculated by the differences in N₂O emissions during 7 d incubation with and without C₂H₂. PV, paddy soil from Vietnam; MV, mangrove soil from Vietnam; PJ, paddy soil from Japan.

Figure 2 Relationship between nitrate (NO₃^–) removal (x-axis) and denitrification (y-axis) in MV soil (triangles), PV soil (squares) and PJ soil (circles) on day 7 at 25°C. Bars indicate standard deviation, n = 3.

Effect of soil pH on N₂O and N₂ emissions

After 7 d incubation at 25°C with 126.5 mg N kg⁻¹, N₂O emissions in PV and MV soils (1.8 and 1.6 mg N kg⁻¹, respectively) were significantly higher than in PJ soil (0.4 mg N kg⁻¹; Fig. 1a–c). Nitrogen (N₂) emission in the 126.5 mg N kg⁻¹ treatment was highest in PJ soil (89.2 mg N kg⁻¹), followed by PV soil (53.7 mg N kg⁻¹). MV soil produced the lowest N₂ emissions regardless of NO₃^– concentrations, with a range from 0 to 1.1 mg N kg⁻¹ (Fig. 1d–f). Denitrification (N₂ + N₂O) was the highest in PJ soil (pH 7.0), followed by PV (pH 5.8) and MV (pH 4.3) soils.

In MV soil, N₂O was the main product of denitrification. In contrast, N₂ was the main product of denitrification in PJ and PV soils. The net N₂O production will be strongly affected by the rate of further reduction to N₂, and therefore the N₂O:N₂ ratio is an important parameter in determining N₂O production originating from denitrification (Van den Heuvel 2010). Results on day 7 showed that the N₂O:N₂ molar ratio was the highest in acidic MV soil (1.7), followed by PV soil (0.03) and PJ soil (0.005). The N₂O:N₂ ratio increased with the decrease in soil pH.

Previous studies have highlighted the effects of pH on denitrification, and on N₂O and N₂ emissions (Koskinen and Keeney 1982; Šimek and Cooper 2002; Čuhel and Šimek 2011; Van den Heuvel et al. 2011). Denitrification was low at pH 5 or below, and the optimum pH seemed to lie between 7 and 8 (reviewed in Šimek and Cooper 2002). In this study, soil pH was extremely acid, moderately acid and neutral for MV, PV and PJ soils, respectively. Probably due to the neutral pH, denitrification (N₂O + N₂) in PJ soil was higher than in PV and MV soils. Our study is consistent with previous studies in terms of the effect of pH on denitrification. Several laboratory experiments using pH-modified treatments and field experiments have confirmed that soil pH affects denitrification end products N₂O and N₂. Sun et al. (2012) showed that N₂O was the major component of gas products of denitrification in forest soils (acid soils), while N₂ was the dominant product in grassland soil (close to neutral soil). Koskinen and Keeney (1982) conducted a field experiment in which silt loam soil was maintained at different pH. They found that at the near-neutral pH of 6.9, N₂ was the main product, and a more acidic condition (pH 4.6 and 5.4) favored the emission of N₂O. In our study, higher N₂O emission by denitrification occurred in MV soil, because a small amount of N₂O was transformed to N₂. The mechanism of pH controlling denitrification products is explained by the sensitivity of N₂O reductase activity to pH. Under acidic pH, the activity and synthesis of N₂O reductase is inhibited, resulting in increased N₂O accumulation (Šimek and Cooper 2002). Soil pH can be used as a predictive parameter for denitrification and N₂O emission (Čuhel and Šimek 2011).

CONCLUSIONS

We investigated controlling factors of N₂O and N₂ emissions from three flooded soils of different types and NO₃^– concentrations. Results showed that denitrification (N₂O + N₂) in PJ soil was NO₃^– limited. In PV and MV soils, denitrification was increased by NO₃^– addition from 0 to 63.4 mg N kg⁻¹, but was not affected by that from 63.4 to 126.5 mg N kg⁻¹. Denitrification products (N₂O and N₂) varied among the studied soils. Denitrification (N₂O + N₂) was higher in the neutral soil (PJ) than that in the moderately acidic soil (PV) and the extremely acidic soil (MV). Nitrous oxide was the main product in MV (acidic soil) while N₂ was the dominant product in PJ and PV soils. We suggest that NO₃^– concentration and soil pH affected N₂O and N₂ emissions in three flooded soils.

ACKNOWLEDGMENTS

The authors wish to thank Mr. Tetsuo Idei in Ushimado, Japan, for permission to collect soil samples. This work was supported by the Japan Science and Technology Agency, Core Research of Evolutional Science and Technology (JST CREST). The first author gives thanks for the scholarships provided by the Okayama Gas Company and the Ministry of Education, Culture, Sports, Scientific and Technology, Japan (MEXT).