Suppression of nitrous oxide production by the herbicides glyphosate and propanil in soils supplied with organic matter

Abstract

Laboratory experiments were conducted to investigate the impact of two herbicides, a commercial formulation of glyphosate (Roundup) and propanil (DCPA), on nitrous oxide (N₂O) production and soil respiration in two different soils (Tyatkone and Miura) amended with rice straw and chitin. The N₂O production rates were very small in both control soils (maximum activity during a whole incubation period was 0.1 µg N kg⁻¹ of soil), but increased with the application of chitin (59 and 152 µg N kg⁻¹) and rice straw (26.6 and 6.2 µg N kg⁻¹) in Tyatkone and Miura soils, respectively. Throughout the 6-week incubation, the application of glyphosate and propanil suppressed cumulative N₂O production in both types of organic matter amended soils. In the rice straw amended Tyatkone soil, the application of glyphosate and propanil decreased N₂O production by 92 and 89%, respectively, and it decreased N₂O production by 20 and 79%, respectively, in the chitin amended Tyatkone soil. In the Miura soil amended with rice straw, the application of glyphosate and propanil decreased N₂O production by 50 and 65%, respectively, while it decreased N₂O production by 49 and 48%, respectively, in the soil amended with chitin. There were no significant differences in cumulative CO₂ evolution between the treatments. It might be suggested that the herbicides used in this study had no severely adverse effects on the overall soil microbial community. Collectively, glyphosate and propanil application can be used as a mitigation strategy to reduce N₂O production from agricultural soils with organic amendment.

Key words:

• chitin
• dentrification
• rice straw
• soil respiration

INTRODUCTION

Biological denitrification in soils is the major source of N₂O emission, especially under conditions of high C availability (Azam et al. 2002; De Wever et al. 2002; Stevens et al. 1997; Takakai et al. 2006). In Myanmar, farmers use available organic plant residues as well as N fertilizers to meet plant demands for nutrients, especially N. The benefits of organic amendments in sustainable agricultural farming are well documented and have been shown to improve soil physical, chemical and biological conditions (Hadas et al. 1996; Khaleel et al. 1981; Roberson et al. 1995; Wander et al. 1994). However, it is well known that the C/N ratio of organic materials is an important factor affecting N₂O production, and crop residues or manure with a low C/N ratio promotes N₂O emission (Cochran et al. 1997; Eichner 1990; Flessa and Beese 1995; Huang et al. 2004; Khalil et al. 2002; Lemke et al. 1999). Nitrous oxide is an important contributor to the greenhouse effect (Bouwman 1990) and N₂O emissions must be reduced to meet the Kyoto protocol (Mc Taggart et al. 2002).

Several researchers have already suggested possible strategies to minimize potentially harmful N₂O emissions in different agroecosystems: (1) by nitrification inhibitors (Bronson et al. 1992; Moiser et al. 1996), (2) by slow release fertilizers (Shoji et al. 1991), (3) by incorporation of crop residues with high C/N ratios (Huang et al. 2004), (4) by the timing of fertilizer and manure applications so that the crop makes maximum use of N (Weier et al. 1993), (5) by the surface application of liquid manure, not soil injection (Comfort et al. 1990; Flessa and Beese 2000; Paul and Beauchamp 1989). Some of these strategies are not currently being used because of limitations to their practicality or cost. For example, selecting the

Table 1 Physicochemical properties of Miura and Tyatkone soils

Soil sample	Soil type	Land use	pH (H2O)	NH4 (mg kg^−1)	NO3 (mg kg^−1)	Total C (g kg^−1)	Total N (g kg^−1)	C/N	MWHC (g g^−1)	Clay (%)	Silt (%)	Sand (%)	Texture
Miura	Andosol	Upland field	6.5	1.8	44.1	67.2	4.7	14.3	0.6	6.5	49.8	43.7	Silt Loam
Tyatkone	Nitosol	Upland field	6.8	0.9	29.3	6.6	0.6	11.0	0.4	11.7	17.6	70.7	Sandy Loam
MWHC, maximum water-holding capacity.

time of fertilizer or manure application is not always practical in many agricultural situations because of the precipitation patterns and a maturing crop canopy can make application difficult (Doerge et al. 1991). One of the controlling practices for decreasing N₂O emission has been the use of nitrification inhibitors with the fertilizer and manure. However, this technique does not always have the potential to produce increased crop yields, although it decreases N₂O production (Aulakh et al. 1984; Bundy and Bremner 1973).

While there is increasing concern that herbicides not only affect the target organisms (weeds), but also the microbial community present in the soil (Ahtiainen et al. 2003; Biederbeck et al. 1987; El Fantroussi et al. 1999; Priemé and Ekelund, 2001; Seghers et al. 2003), their potential effects on N₂O production are largely unknown. There are some contradictions in the impacts of agrochemicals on N₂O production. Soil fumigations with chloropicrin and methyl isothiocyanate stimulated N₂O production (Spokas and Wang 2003; Spokas et al. 2005, 2006). In contrast, Kinney et al. (2005) reported that the application of the herbicide prosulfuron and the fungicides mancozeb and chlorothalonil decreased N₂O emission, possibly because the chemicals inhibited nitrification and denitrification. Our previous study (Kyaw et al., unpublished data, 2006) demonstrated the potential of the herbicides glyphosate and propanil to suppress N₂O production through nitrification and denitrification. However, the combined effects of herbicide and organic matter amendments were not evaluated. Therefore, this study was conducted to investigate the impact of herbicides, commercial formulations of glyphosate and propanil, commonly used in Myanmar on N₂O production in two different soils following the application of rice straw and chitin as model substrates with high and low C/N ratios, respectively.

MATERIALS AND METHODS

Soils

The experiments were conducted using two types of soil. One soil was sampled from the plow layer (0–10 cm) at the Miura Experimental station, Kanagawa Prefecture Agricultural Center, Japan, and the other was taken from the research farm of Tyatkone, Department of Agricultural Research, Ministry of Agriculture and Irrigation, Myanmar. After sampling, soil samples were sieved through a 2 mm mesh and mixed well. Some of the physicochemical properties of the Miura and Tyatkone soils are described in Table 1.

Herbicides and organic matter

The herbicides used in the present study were glyphosate (N-[phosphonomethyl]glycine; as the commercial formulation Roundup, 41% active ingredient [a.i.], 59% water and surfactant, Nissan Chemical, Tokyo, Japan), Propanil (N-3,4-dichloropropionanilide; as the commercial formulation DCPA [emulsion], 35% a.i., 65% emulsion and organic solvents [xylene 12.5%, ethyl benzene 2.6%, calcium dodecylbenzene sulfonate 2.5% and poly[oxyethylene] 4.0%], Dow Agrosciences, Tokyo, Japan). Each herbicide was applied at field rate each 0.2 µL per 40 g (oven dry basis) (corresponding to 2 L a.i. ha⁻¹) of glyphosate and propanil. The application amount to the soil was calculated based on 10 cm depth and the specific gravity of soil is 1.0.

Chitin (Wako Pure Chemical Industries, Osaka, Japan) and finely ground rice straw were used as model compounds of organic matter (chitin: total C and N were 436 and 64.9 mg g⁻¹, respectively, C/N ratio 6.7; rice straw: total C and N were 364 and 11.1 mg g⁻¹, respectively, C/N ratio 32.8).

Herbicide and organic matter treatments

Soil moisture has been shown to be one of the most important factors controlling emissions of N₂O and NO (Akiyama et al. 2000; Dobbie et al. 1999; Hatano and Sawamoto 1997). In our previous report, total N₂O emission was markedly higher in the soil adjusted to 80% maximum water-holding capacity (MWHC) than in the soil adjusted to 60% MWHC, and we suggested that the higher N₂O emission was of denitrification origin (Kyaw et al., unpublished data, 2006). Therefore, the effect of two herbicides on N₂O emission through denitrification was assessed in soils adjusted to 80% MWHC following rice straw and chitin amendments. One kilogram of sieved soil (2 mm) was placed in a plastic container and distilled water was added to adjust the soil moisture to 55% of its MWHC and the soil was then pre-incubated for 7–10 days at 24 ± 1°C.

Forty grams (oven-dry basis) of the pre-incubated soil was placed in a 500-mL plastic bottle, supplied with KNO₃ (120 mg N kg⁻¹) and each organic matter (6 g kg⁻¹) and thoroughly mixed. The herbicides were added as an aqueous solution with a sufficient volume of water to adjust the final soil moisture to 80% MWHC. The treated bottles were capped with a double rubber stopper immediately after mixing and headspace gas samples were periodically taken, including just after mixing (0 h) during incubation at 24 ± 1°C. Soil samples that contained the fertilizer, but not the organic matter and herbicides were also prepared separately as controls. Three replicate bottles were prepared for each treatment. Water lost by evaporation was added periodically to maintain the initial water content.

Gas measurements

The rates of N₂O and CO₂ production were investigated periodically, including immediately after herbicide application (0 h). At every measurement, the lids of each bottle were changed from loosely covered aluminum foil to airtight double rubber stoppers. After 6–8 h of further incubation at 24°C in the dark, a gas sample was taken from the headspace. The concentration of CO₂ was measured by injecting 0.5 mL of the gas samples into a gas chromatograph equipped with a thermal conductivity detector (GC-8A, Shimazu, Kyoto, Japan) and a stainless steel column packed with Porapak-Q (80/100 mesh, 3 mm diameter and 2 m length). The detector temperature in the GC was set at 80°C. The bridge current and the column temperature were maintained at 140 mA and 60°C, respectively. Helium gas was supplied for the carrier gas at a flow rate of 40 mL min⁻¹. Concentration of N₂O by injecting 5 mL into a gas chromatograph equipped with an electron capture detector (GC-14A, Shimazu) and a stainless steel column (3 mm diameter and 2 m length), packed with Porapak-Q (80/100 mesh). The detector temperature and column temperature were set at 330 and 90°C, respectively, and Ar with 5% CH₄ mixture gas was supplied as a carrier gas at a flow rate of 23 mL min⁻¹ throughout the run. After each determination, all of the bottles were opened to ventilate the headspace and then the bottles were capped with aluminum foil and incubated again.

Statistics

Data were analyzed using standard anova procedures using Excel Statistics ver 1.1 (Social Survey Research Information Co., Ltd., Tokyo, Japan).

RESULTS

Soil respiration

An increase in soil respiration with rice straw began quickly at 12 h (half a day) after application to both Tyatkone and Miura soils. After this time it declined, while the CO₂ production rate started to increase in the soil amended with chitin 4 days after application (Table 2). Although the trends of the CO₂ production were a little different in both soils, the higher peak of CO₂ coincided with the peaks of N₂O within 15 days of incubation in both soils. Throughout the incubation period, most of the values in the rate of CO₂ production did not differ significantly with the application of herbicide, except for the application of glyphosate to the Miura soil amended with chitin.

Nitrous oxide emission

In the Tyatkone soil, the N₂O production rate of the soil receiving no organic matter amendment was less than 0.15 µg N kg⁻¹ h⁻¹ of soil, while that in the soil amended with rice straw was markedly higher in the early period (2 days after application), with a maximum peak of 26.6 µg N kg⁻¹ h⁻¹, and then sharply declined. With the application of the herbicides glyphosate and propanil, N₂O production was significantly (P < 0.05) inhibited in the soil amended with rice straw and the maximum production rate was not higher than 2.5 µg N kg⁻¹ h⁻¹. The increase in N₂O production following the amendment of chitin started 6 days after application and reached a maximum of 40.6 µg N kg⁻¹ h⁻¹ 15 days after application and then gradually decreased until the end of the incubation. Similarly, increased N₂O production by chitin was significantly (P < 0.05) inhibited by the application of glyphosate and propanil, particularly 5–19 days after incubation.

In the Miura soil, N₂O production was markedly higher in soils amended with rice straw and chitin, with maximum peaks of 6.2 and 87.6 µg N kg⁻¹ h⁻¹ 2 and 15 days after application, respectively, than in the soil without organic matter amendment (0.10 µg N kg⁻¹ h⁻¹). The application of glyphosate and propanil decreased N₂O production significantly (P < 0.05) in the chitin-amended soils (Table 3). In the soils amended with rice straw, the highest peaks of N₂O production were observed within 2 days after application. Increased N₂O production by rice straw was significantly reduced (P < 0.05) by glyphosate and propanil within 2 days of application. However, no significant difference in N₂O emission was observed again until the end of the incubation because of high variations within replicates.

DISCUSSION

The present study clearly demonstrated that the application of glyphosate and propanil reduced N₂O production from soils amended with rice straw and chitin. The N₂O production observed in this study is mainly derived from denitrification because the soils were incubated under wet conditions, 80% MWHC (corresponding to 73% of water-filled pore space), and very low amounts of

Table 2 Impact of herbicides on CO2 production in Tyatkone and Muira soils amended with organic matter

Treatments	Days after application
	0.5	2	3	4	5	6	11	15	19	26	31	35	40
Tyatkone soil
Control	9.8 a	7.3 a	6.5 a	38.6 b	7.3 a	5.5 a	4.1 a	3.6 a	3.3 a	2.7 a	3.1 a	0.8 a	0.5 a
RS	222.6 b	105.8 b	99.2 b	140.6 c	112.4 d	82.9 c	64.4 b	40.9 b	37.4 c	26.3 b	19.8 b	13.5 b	11.3 b
GPS + RS	214.3 b	85.7 b	87.2 b	107.4 c	92.0 d	65.7 bc	49.2 b	16.6 ab	21.2 b	23.9 b	17.0 b	12.4 b	8.9 b
PPN + RS	216.0 b	101.2 b	94.2 b	122.7 c	115.2 d	80.7 c	64.6 b	40.5 b	31.5 bc	27.4 b	19.7 b	16.5 c	16.1 c
Chitin	14.9 a	9.0 a	8.0 a	20.0 ab	29.0 b	44.5 b	290.7 d	146.1 c	99.5 e	43.7 c	29.6 c	25.3 d	16.3 c
GPS + Chitin	14.6 a	9.5 a	8.7 a	17.3 ab	34.4 c	64.9 bc	348.8 d	148.8 c	79.3 d	42.8 c	28.1 c	24.9 d	17.1 c
PPN + Chitin	18.0 a	10.4 a	9.0 a	14.2 a	25.1 b	29.4 b	142.1 c	223.1 d	135.2 f	59.4 d	30.1 c	24.7 d	18.1 c
Muira soil
Control	17.7 a	7.2 a	5.5 a	6.4 a	13.0 a	2.2 a	2.4 a	4.3 a	1.8 a	5.9 a	3.1 a	2.3 a	1.7 a
RS	189.8 b	112.5 b	86.2 b	79.2 c	172.5 b	40.4 b	24.5 a	10.0 b	17.0 b	20.7 b	16.6 c	11.8 b	11.4 b
GPS + RS	185.5 b	110.5 b	84.3 b	76.9 c	218.6 bc	41.0 b	25.8 a	18.6 b	16.5 b	23.7 c	16.6 c	11.8 b	10.2 b
PPN + RS	196.8 b	115.1 b	86.2 b	81.3 c	198.5 bc	43.0 b	24.7 a	16.5 ab	16.6 b	22.6 bc	16.1 c	9.5 b	8.9 b
Chitin	21.4 a	7.8 a	6.6 a	21.4 b	239.0 c	156.8 c	214.5 b	90.5 d	47.7 d	27.2 d	16.9 c	10.6 b	9.1 b
GPS + Chitin	24.2 a	8.1 a	7.2 a	23.1 b	322.0 d	163.4 c	234.2 b	64.4 c	41.1 c	23.2 bc	13.3 b	11.3 b	10.8 b
PPN + Chitin	24.9 a	9.0 a	7.3 a	23.5 b	259.9 cd	148.5 c	217.1 b	75.2 c	48.1 d	25.0 cd	15.1 bc	10.1 b	9.9 b
Values are means (n = 3; ug CO2-C g^−1 h^−1). Means in a column followed by the same letter are not significantly different using least significant difference (P < 0.05). GPS, glyphosate; PPN, propanil; RS, rice straw.

Table 3 Impact of herbicides on N2O production in Tyatkone and Muira soils amended with organic matter

Treatments	Days after application
	0.5	2	3	4	5	6	11	15	19	26	31	35	40
Tyatkone soil
Control	0.1 a	0.1 a	0.1 a	0.0 a	0.1 a	0.0 a	0.0 a	0.0 a	0.0 a	0.0 a	0.0 a	0.0 a	0.0 a
RS	19.7 b	26.6 b	11.5 b	3.3 b	1.1 b	0.9 bc	0.7 a	0.1 a	0.1 a	0.1 a	0.0 a	0.0 a	0.0 a
GPS + RS	2.3 c	1.3 c	0.5 a	0.5 a	0.2 a	0.2 a	0.1 a	0.1 a	0.1 a	0.0 a	0.0 a	0.0 a	0.0 a
PPN + RS	2.1 c	2.5 c	0.4 a	0.8 a	0.5 ab	0.3 a	0.1 a	0.1 a	0.1 a	0.0 a	0.0 a	0.0 a	0.0 a
Chitin	0.2 a	0.1 a	0.0 a	0.1 a	0.4 ab	0.9 c	18.3 b	40.6 d	25.6 d	3.3 b	0.3 a	0.0 a	0.0 a
GPS + Chitin	0.3 a	0.2 a	0.1 a	0.1 a	0.4 ab	1.5 bd	19.0 b	31.7 c	14.3 c	3.6 b	0.5 b	0.0 a	0.0 a
PPN + Chitin	0.1 a	0.1 a	0.1 a	0.1 a	0.1 a	0.1 a	1.8 a	8.1 b	7.8 b	0.8 b	0.3 a	0.0 a	0.0 a
Muira soil
Control	0.1 a	0.0 a	0.0 a	0.1 a	0.0 a	0.1 a	0.0 a	0.0 a	0.0 a	0.1 a	0.0 a	0.1 a	0.0 a
RS	5.9 b	6.2 c	1.9 b	0.8 a	0.1 a	0.1 a	0.0 a	0.0 a	0.0 a	0.0 a	0.1 a	0.1 a	0.0 a
GPS + RS	3.2 c	1.8 b	1.4 ab	0.5 a	0.1 a	0.1 a	0.2 a	0.0 a	0.1 a	0.1 a	0.0 a	0.0 a	0.0 a
PPN + RS	2.8 c	0.4 b	0.8 ab	0.6 a	0.1 a	0.2 a	0.2 a	0.0 a	0.1 a	0.1 a	0.0 a	0.0 a	0.0 a
Chitin	0.1 a	0.1 a	0.1 a	0.4 a	18.0 b	41.1 c	64.1 c	87.6 c	49.3 c	31.2 b	7.7 b	0.0 a	0.0 a
GPS + Chitin	0.1 a	0.1 a	0.1 a	0.8 a	13.9 b	18.0 b	48.8 b	39.1 b	22.3 b	9.0 a	1.8 a	0.0 a	0.0 a
PPN + Chitin	0.1 a	0.0 a	0.1 a	0.2 a	5.2 a	15.1 b	52.0 b	44.0 b	26.6 b	10.0 a	2.0 a	0.0 a	0.0 a
Values are means (n = 3; ug  N2O-N  g^−1 h^−1). Means in a column followed by the same letter are not significantly different using least significant difference (P < 0.05). GPS, glyphosate; PPN, propanil; RS, rice straw.

N₂O were produced from the soils without the organic amendments. Therefore, glyphosate and propanil may affect some of the denitrifying processes and, thus, suppress N₂O production. The influence of herbicides on N₂O production has received little attention. Muller et al. (1981) reported that glyphosate had no effect on denitrification in two soils when tested at a level of 2.6 kg ha⁻¹. Stratton and Stewart (1991) also investigated the effect of glyphosate on denitrification in forest soils and found that glyphosate at a field rate (1.7 kg a.i. ha⁻¹) had no significant effect on denitrification and had a stimulating effect when it was used at concentrations 100-fold higher than the field rate in a laboratory bioassay.

Nitrous oxide is an intermediate gas in the denitrification pathway (Zumft 1997). If N₂O production exceeds N₂O consumption, N₂O can be released from the site of activity (Firestone et al. 1980). Hence, lower N₂O emission in our study might result from a greater rate of N₂O reduction to N₂ or a lower rate of N₂O production from NO₃ reduction. Further investigation is necessary to verify the precise impact of herbicides on the denitrifying activity of soils.

The effects of glyphosate on N₂O production have not been tested in previous studies and additional organic substrates have not been used. However, denitrification is a dominant process for N₂O production, especially under conditions of high C availability (Azam et al. 2002; Stevens et al. 1997). Therefore, to the best of our knowledge, this is the first report to show a suppressive effect of glyphosate and propanil on N₂O production through denitrification.

Kinney et al. (2005) observed that the herbicide prosulfuron and the fungicides mancozeb and chlorothalonil inhibited N₂O production under different levels of moisture conditions at application rates of 0.02×, 1× and 10× the field rate (0.71 kg a.i ha⁻¹) in a soil microcosm experiment, and concluded that these chemicals affected nitrification and denitrification and, thereby, inhibited N₂O production. However in the report, detailed mechanisms were not discussed regarding the suppression of N₂O production by these agrochemicals.

Carbon availability has long been recognized to enhance denitrification in soils under anaerobic conditions (Burford and Bremner 1975; Sahrawat and Keeney 1986; Smith and Tiedje 1979). Huang et al. (2004) found that the incorporation of plant residues increased N₂O emissions and was negatively correlated with the C/N ratio in the plant residues. In this study, chitin (C/N ratio 6.7) induced significantly (P < 0.01) higher N₂O emission than rice straw (C/N ratio 32.8), especially in the Miura soil.

Soil respiration is considered to be an indicator of the response of the total soil population to herbicides. As most of the values of CO₂ production from the soils amended with rice straw and chitin were not significantly suppressed by the herbicide treatments in the incubation period (Table 2), our results suggest that glyphosate and propanil have no severe adverse affect on the overall soil microbial community. Our study demonstrated that the application of glyphosate and propanil could suppress increased N₂O production by organic amendments. However, further study is necessary to elucidate the suppression mechanisms and field-scale monitoring is an important next step.

ACKNOWLEDGMENTS

K. M. Kyaw sincerely thanks the Ministry of Education, Science, Sports and Culture of Japan for a scholarship. The authors also extend their thanks to Dr T. Okamoto for soil sampling and Professor M. Okazaki for invaluable suggestions. This study was partly supported by Tokyo University of Agriculture and Technology 21st Century Program (Evolution and Survival of Technology-based Civilization).