Effects of Fertilizers and Mushroom Residues on Soil N 2 O Emission Under Rice-Wheat Rotation in Chengdu Plain

Nitrous oxide (N2O) is a greenhouse gas, and agricultural landscapes are major sources of atmospheric N2O. In this study, different types and levels of fertilization were applied to rice and wheat plants, including full crop straw, pure synthetic fertilizer and spent agro-residues from mushroom cultivation. N2O flux measurements were performed once a week using gas static chromatography-chamber between 2008 and 2009. In order to find out the effect of MR application on N2O emission, soil samples, environment-monitoring factors including soil moisture and temperature and biomass were also determined. Soil nitrate and ammonia were measured in soil extracts. The results showed that the total N2O emission decreased to 19,066 kg ha in the rice stage and 45,312 kg ha in the wheat stage under the mushroom residue (MR) application. This observation indicated that MR application (22,656.40 kg for rice and 9,533.33 kg for wheat) induced a decrease of N2O emission by 62.52% and 67.55% as compared with fertilizer and straw application, which are 6328.77±740.81 g ha, 7310.60±279.73 g ha respectively. Therefore, MR application could be one of the most effective ways to reduce soil N2O emissions.


Introduction
Nitrous oxide (N 2 O) is an atmospheric trace gas contributing to global warming (IPCC, 1996).The major anthropogenic source of N 2 O has been attributed to agriculture, which is responsible for 50% of current anthropogenic emissions (IPCC, 2001).Furthermore, the ultimate cause of N 2 O emission is nitrogen (N) enrichment due to nitrogen fertilizer application tillage, crop residue incorporation and many other organic matter applications (Dobbie et al., 1999;IPCC, 2001;Zou et al., 2005;Miller et al., 2008;Xu et al., 2009).
Many studies reported that N 2 O emissions are increased with the application of N (IPCC, 2001;Hellebrand et al., 2003;Miller, 2008).Regarding tillage, N 2 O emissions from soil are higher under no-tillage (NT) than those under conventional-till (CT) (Skiba et al., 2002).However, other studies showed that the N 2 O emission is decreased under NT (Passionato et al., 2003).Similarly, Elmi et al. (2003) and Grandy et al. (2006) reported the similar level of N 2 O emission between CT and NT.Furthermore, there are also conflicting results for the effect of crop straw incorporation on N 2 O emissions.Crop straw can cause N loss through denitrification and N 2 O emissions (Baggs et al., 2000;Huang et al., 2004).However, Hao et al. (2001) and Yao et al. (2009) observed that the N 2 O emission is decreased when crop residues are retained.In addition, the positive effects of other agricultural organic matters (such as animal manure) on soil N 2 O emissions have been reported (Stevens & Laughlin, 2001;Khalil et al., 2002;Ding et al., 2007).Cattle manure reduces N 2 O emissions compared with ammonium sulfate (Velthof et al., 2002(Velthof et al., , 2003)).Field measurements in Northeast and North China showed that there is no significant difference in N 2 O emissions between organic manure and urea (Chen et al., 2002;Meng et al., 2005).However, few studies have described the effects of mushroom residues (MR) on soil N 2 O emissions.Relevant studies mainly focus on relationships between soil physic-chemical, chemical and biological properties in soil amended with mushroom residues.
MR is the leftover agro-residue from mushroom cultivation.It is a compost mixture of cereal straw, manure (poultry and/or cattle manure), calcium sulphate, soil, and residues of inorganic nutrients and pesticides (Williams et al., 2001).The residues contain not only abundant organic matter, nitrogen, phosphorus, potassium and other nutrients, but also a considerable amount of bacterial proteins (Chen, 2001;Li, 2003).In Chengdu Plain, the annual production of mushroom culture was 1.06 Mt in 2007, leading to a total of 0.28 Mt of fungal residues.Over 80% of fungal residues are disposed without recycling (Hu, 2008).This is not only a waste of resources, but also causes environmental problems, such as surface water pollution.Therefore, it is detrimental for mushroom cultivation, such as mycete and ácaro (Mi et al., 2005;Li & Cheng, 2008;Wang et al., 2008).MR recycling is one of the most effective ways to solve these problems.
In the present study, the optimal proportion of MR and synthetic fertilizers are determined for the analysis between different fertilization and N 2 O emission.This paper also presents the possible beneficial effects of MR application on soil N 2 O emissions by field experiments with a rice (Oryza sativa)-wheat (Triticumaestivum) rotation compared with a crop straw application.

Site Description and Materials
The experimental site was located in Hanchang County (latitude 30°27′ N, longitude 103°41′ E, altitude 483.7 m) in Chengdu Plain with a mean annual precipitation of about 1,098.2mm and a mean annual temperature of 16.0 o C. The soil at the experimental site is a Fe-accumuli-StagnicAnthrosols with long-term NT derived from grey alluvium of the MinjiangRiver.The basic physicochemical properties of the topsoil layer (0-10 cm) were listed in Table 1.The mushroom residues used in the experiment were produced in the standard cultivation house of one local edible fungi company.All the residues came from agaricus-bisporus which were fertilized by the mixture of cattle manure and rice straw.Before the experiment, the basic nutrient content of mushroom residues applied in rice and wheat season were measured and they were listed in Table 2.After the rice or wheat was harvested, the stubble height was less than 5 cm.

Experimental Design
Six different treatments were conducted in the investigation, including chemical fertilizer only (CF), chemical fertilizer with a full dose of rice or wheat straw returned (CFS), or a mixture of 50% MR and 50% chemical fertilizer.Table 3 shows six different kinds of fertilization applied in the field.The percentages of total N and CF in the MR were 50% and 100% respectively.The ratio was 0.5:1 (50% MR).When a full dose of mushroom cultivation residues was returned, the percentage of total N and that in the MR were both 100%.The ratio was 1:1 (100% MR); the percentage of total N and CF in the MR were 150% and 100% respectively.The ratio was 1.5:1 (150% MR).Each test was performed in triplicate.MR treatments were fertilized with the same dose of urea for N, single super-phosphate for P, and potassium chloride for K as for the CF treatments, and they were arranged in a randomized complete block design.The length and breadth of the plot was 5 m and 6 m respectively.
The soil N 2 O efflux under different fertilization was measured using the static chamber technique.The sampling boxes are cubical in shape and made by steel.In order to avoid temperature change inside the boxes caused by solar radiation, the outer flank of every box was wrapped up by sponge and then by a thin layer of silver paper.
Before the rice blooming stage, 45 kg ha -1 of K fertilizer was applied to the field surface.The remainder was applied to the field surface on the 4th day after the rice transplantation.In the wheat season, all fertilizers were applied to the surface before sowing.In the rice and wheat season, the major nutrient contents of mushroom cultivation residues returned to the field were 396.70 and 428.40 g per kg of organic matter, 16.20and 18.10 g per kg of N, 0.61 and 0.81 g per kg of P, and 2.47 and 2.99 g per kg of K, respectively.Concerning the law of growth cycle of wheat and rice, factors (temperature, moisture, NH 4 + and NO 3 -) affecting soil N 2 O emissions were monitored every 10 days during the wheat stage and once a week during the rice stage.
The top soil temperature at different depths (5 cm, 10 cm, 15 cm and 20 cm) both inside and outside of the gas chamber was measured by JM portable electronic thermometer.The measurement could be divided into two parts: routine determination and diurnal variation measurement.The coldest, the hottest and the most moderate soil temperatures in one day were measured in each plot at 7:00, 13:00 and 18:00 respectively in the routine determination.While in diurnal variation measurement, the soil temperature was observed every 2 hours between 7:00 and 9:00 during the growth stage of wheat and rice.
The top soil moisture was measured in each plot using the method of oven drying in aluminum-boxes.During the incubation, the initial soil moisture was measured and expressed as soil water content (%).

Soil NH 4
+ and NO 3 -contents were measured by a KCl extraction method as follows.Briefly, 5.00 g of moist soil (0-10 cm) was added to 50 ml of 1 M KCl solution, and the mixture was then vortexed for 1 h to fully extract NH 4 + and NO 3 -contents.Subsequently, the extract solution was analyzed by an auto analyzer (Foss, FIAstar5000, Danish).Soil NH 4 + and NO 3 -contents were expressed as follows:

N 2 O Flux Measurement
N 2 O flux was simultaneously measured using a static chamber.The plant density inside the flux chamber (0.5 m × 0.5 m × 0.5 m), covering four hills of rice and 30 hills of wheat in the field, was the same as that outside the chamber.In all plots, stainless steel bases were installed into the chambers before the rice transplantation and kept there until the next harvest season.In order to avoid soil disturbances during the sampling and measurements, removable wooden boardwalks (2 m of length) were set up at the beginning of the rice season.Gas samples were collected during a 10-day period post-fertilization with a 4-or 5-day interval.Moreover, gas samples were also collected during the next 4 months of rice growth with a 7-day interval.Four gas samples from each chamber were collected during the rice growth stage with a 5-min intervaland during the wheat growth stage with an 8-min intervalusing 60-ml vacuum vials.Due to the law of plant respiratory metabolisms, the photosynthesis of plant is strongest between 09:00 and 11:00AM and most of greenhouse gases are released during that time.Samples were collected between 09:00 and 11:00 AM on every sampling day.
The N 2 O concentration was measured using a gas chromatograph (GC) instrument (HEWLETT Packard 5890, seriesⅡ, USA) equipped with an electron capture detector (FID) through automatic injection (CA-5, Institute of Atmospheric Physics, China).Chromatography conditions were set up as follows: column temperature of 35 o C, injection temperature of 55 o C, and detection temperature of 200°C.A N 2 O peak was detected at a specific retention time.Before the sample analysis, the GC was calibrated with different dilutions of 10 mg kg -1 span N 2 O gas, and the chamber volume was then determined.Total N 2 O emissions were calculated as follows: Where D i is the number of days, F i is the measured efflux in the ith sampling interval, and n is the number of sampling intervals.

Calculations and Statistics
All data were analyzed using the General Linear Model of SPSS13.0.Independent variables are six different fertilization including CF, CFS, 50%MR, 100%MR, 150%MR, 200%MR while dependent variables are soil NO 3 -, soil NH 4 -, soil NO 3 -/NH 4 -,soil moisture, soil biomass and soil temperature.All non-normal data were log-transformed.Means were determined using the Uniform Minimum Variance Unbiased Estimators.Results were similar to the arithmetic means.Therefore, treatment means and standard errors presented in tables and figures were obtained from untransformed data.Comparisons of total N 2 O emissions were performed using the Least Significance Difference (LSD) test since comparisons are significant.Treatment means were compared using a protected LSD test.Pearson correlation coefficients were determined for soil NO 3 -, NH 4 + contents, temperature of surface soil and soil at 5 cm blow the surface, soil moisture and N 2 O emissions.P<0.05 was considered as statistically significant.
The effects of environmental conditions on N 2 O efflux were analyzed by a series of single and Multiple Regression Analyses (MRA).Multiple regression analysis has been widely used to model the cause-effect relationship between inputs and outputs and has predicting function.It can be generally expressed as where Y is a dependent variable (i.e., output variable), X 1, … , X n are independent or explanatory variables (i.e., input variables), θ 1 -θ p are regression parameters, ε is a random error, which is assumed to be normally distributed with zero mean and constant variance σ 2 , and f is a known function, which may be linear or nonlinear.If f is linear, then (3) becomes a multiple linear regression model and can be expressed as Y b b X +b X ⋯ b X ε (4) where b 0 is a constant and called intercept.Different functional forms decide different MRA models.
The regression parameters θ 1 -θ p or b 0 -b n are usually estimated using the least squares method (LSM), which can be expressed as an unconstrained optimization problem: where t = 1, … , T represent T different sample points.Once the regression parameters are determined, the corresponding regression model can be utilized for prediction.

Effect of MR Application on Seasonal N 2 O Emission
Average N 2 O emission ranged from -0.133 to 4.093 mg m 2 h -1 under CF, -0.079 to 2.061 mg m 2 h -1 under CFS, -0.173 to 3.261 mg m 2 h -1 under 50% MR, -0.310 to 1.125 mg m 2 h -1 under 100% MR, -0.173 to 0.793 mg m 2 h -1 under 150% MR, and -0.313 to 2.650 mg m 2 h -1 under 200% MR. Figure 1 shows that two N 2 O efflux peaks were observed after the fertilization during the winter wheat seeding stage and maturity stage, respectively.The first peak appeared due to the fertilization, and the second one was caused by the acute alteration of soil moisture.However, the soil moisture changed only during the first several months in the cold winter and during the wheat growth season.
Furthermore, two influx peaks during the tillering stage (around June 24 th , 2009) and flowering stage (around July 27 th , 2009) and two efflux peaks during the drained stage within the tillering stage (around November 9 th , 2008) as well as during the maturity stage (around May 4 th , 2009) were observed during the rice growing season though they were not very obvious (Figure 1).The first N 2 O efflux peak was observed after the fertilization, and the second one appeared on June 30 th when all field water was drained in order to prevent the rice from over-tillering until the heading stage.The third efflux peak occurred during the maturity stage after the drainage (Figure 1).Finally, the two influx peaks bothappeared during the flooding period.The correlation between the N 2 O efflux and topsoil NO 3 -/NH 4 + content was analysed.Data showed that except for the chemical fertilization treatment, other treatments were significantly different at the p < 0.05 level, and the 150% MR treatment was significant at the p < 0.01 level (Table 5).Therefore, although the N fertilizer was the factor affecting the soil nitrification and denitrification, a striking correlation could be found between the soil N 2 O efflux and soilat neither the rice stage nor the wheat stage.However, the N 2 O efflux was positively correlated with the NO 3 -/NH 4 + content.

Correlation Between N 2 O Efflux and Soil Moisture
The effects of environmental conditions on N 2 O efflux were investigated by a series of single and multiple regression analyses.Consistent with previous studies,no significant correlation between the soil moisture and N 2 O efflux was found for any treatment (Table 5).However, the two leaps of N 2 O efflux emerged at the period when soil was with high moisture level, which was also during the flooding time at the rice transplanting and wheat maturation stage (Figure 3).

Correlation Between N 2 O Efflux and Soil Temperature
Table 3 shows that the temperature of surface soil and soil at 5 cm blow the surface were both positively correlated with the N 2 O emission at the wheat stage.However, at the rice stage, a negative correlation was observed between the soil N 2 O emission and the temperature of surface soil or soil at 5 cm blow the surface.This difference was mainly resulted from different soil moistures between the wheat stage and rice stage.

Correlation Between N 2 O Efflux and Crop Biomass
The correlation between the soil N 2 O efflux and crop biomass during the wheat-rice rotation was analyzed.Data in the correlation matrix showed that the soil N 2 O efflux was significantly correlated with the crop fresh weight and dry weight, except for the dry weight under CFS and 200% MR (Table 5).* Correlation is significant at the 5% level, **correlation is significant at the 1% level.
Moisture value means 0-10 cm depth soil moisture.
The corretion between N 2 O emission and biomass, which countents fresh weight and dry weight.
p a means the correlation between N 2 O emission and soil surface temperature.
p b means the correlation between N 2 O emission and soil temperature 5cm under surface.

Effect of Soil Moisture, Temperature, Biomass and NH 4 + /NO 3 -Content on N 2 O Emission
Microbial denitrification and nitrification are responsible for the majority of N 2 O emissions under many soil environmental conditions (Fireston & Dowids, 1989), such as soil moisture, temperature and mineral fertilizer.A series of multiple regression analysis was used to investigate the effects of environmental conditions on soil N 2 O emissions.Data indicated that the soil temperature (especially the surface temperature), crop biomass and NO 3 -/NH 4 + content could affect the N 2 O emission, whereas the soil moisture could not (Table 5).Significant correlation between the soil moisture and N 2 O emission was not found in the study, which was consistent with the results of Jones et al. (2007) but different from Liu et al. (2007) and Beare et al. (2009).Liu et al. reported that favorable soil moisture can increase the N 2 O emission.
Other results showed that the N 2 O emissions are significant increased when the soil moisture is increased from 30% to 50%.Reinhard Well et al. (2008) and Liu et al. (2007) concluded that the nitrification greatly contributes to the N 2 O emission when the soil moisture is favorable.N 2 O is mainly derived from denitrification since nitrification can be depressed by higher soil moisture.However, significant correlation between the N 2 O emission and topsoil moisture was not found in this study.On the contrary, the three main N 2 O pulses, which occurred at the wheat maturity stage, the rice tillering stage and the maturity stage, occurred during the period when the soil moisture was acutely altered.
Nitrous oxide is produced from denitrification and nitrification processes in soils, and nitrification-denitrification activity was sensitive with temperature (Maag & Vinther, 1996).The daily N 2 O emission measured at each growth stage of rice or wheat was significantly associated with the soil surface temperature as well as the temperature of soil at 5 cm blow the surface.However, temperature played the opposite role in N 2 O emissions at wheat and rice stages.As the N 2 O efflux was positively correlated with the temperature at the wheat stage, it was decreased at the rice stage due to the flooding and dry environment.
Microbial denitrification and nitrification are responsible for the majority of N 2 O emissions in many soil environments.The elevated N 2 O emission induced by the addition of ammonia is mainly due to denitrification rather than nitrification (Clough et al., 2004).There wasn't any significant correlation between the N 2 O emission and soil NO 3 -or NH 4 + content respectively in the this study, which was consistent with results of previous study (Miller et al., 2008).However, there exists significant correlation between soil N 2 O emission and soil NH 4 + /NO 3 -.When taken together, the soil NO 3 -/NH 4 + content played an important role during the denitrification-nitrification period, at least under a wheat-rice rotation.

Effect of MR Application on Soil N 2 O Emission
Although MR consisted of many types of organic matters, its application decreased the soil N 2 O emission at both wheat and rice stages.Compared with mushroom residues, the crop residues also affected the N 2 O emission at both wheat and rice stages, which was consistent with previous study (Miller et al., 2008).Bacterial denitrification plays an important role in the global nitrification cycle, and it is a principal contributor of N 2 O to the atmosphere (Miller et al., 2008).Many studies have shown that the available C increases the microbial activity and O 2 consumption, leading to conditions favorable for denitrification.
Although higher N 2 O fluxes can resulted from the application of organic matters (such as manure, sewage, slurry and crop residue), the MR application still induces a reduction trend of soil N 2 O emission compared with the chemical fertilizer treatment (Christensen, 1983;Clayton et al., 1997;Scott et al., 2000).This difference can be caused by the following reasons.First, the organic matter within the mushroom cultivation residue contains a great deal of unavailable organic matter, which is retained after the mushroom cultivation.Therefore, the available organic matter in soil is not increased by the MR application.Second, N 2 O fluxes are decreased by the MR application mainly due to the decrease in [NO 3 -/NH 4 + ], which is positively correlated with the N 2 O emission.Finally, lower soil moisture is found in MR-treated plotsunder the same environmental condition.This is the direct reason for denitrification, which is more important for soil N 2 O effluxes than for nitrification.

Conclusions
MR application has a strong impact on both seasonal and total soil N 2 O emission.Results showed that average N 2 O emission ranged from -0.133 to 4.093 mg m 2 h -1 under CF, -0.079 to 2.061 mg m 2 h -1 under CFS, -0.173 to 3.261 mg m 2 h -1 under 50% MR, -0.310 to 1.125 mg m 2 h -1 under 100% MR, -0.173 to 0.793 mg m 2 h -1 under 150% MR, and -0.313 to 2.650 mg m 2 h -1 under 200% MR.The total N 2 O emission was decreased to 19,000 kg ha -1 in the rice stage and 45,000 kg ha -1 in the wheat stage under the MR application, respectively.This observation indicated that MR application (22,656.40kg for rice and 9,533.33kg for wheat) induced a decrease of N 2 O emission by 62.52% and 67.55% as compared with fertilizer and straw application, respectively.Therefore, MR application could be one of the most effective ways to reduce soil N 2 O emissions.
Seasonal dynamics in N 2 O emissions was largely regulated by moisture, cumulative temperature, soil NO 3 -/NH 4 + content and crop biomass status in the soil.It is found that the N 2 O efflux was positively correlated with the NO 3 -/NH 4 + content while no significant correlation between the soil moisture and N 2 O efflux was found for any treatment.Different soil moistures between the wheat stage and rice stage led to positive and negative correlations respectively between soil N 2 O emission and soil temperature.Data also showed that the soil N 2 O efflux was significantly correlated with the crop fresh weight and dry weight, except for the dry weight under CFS and 200% MR.
Based on previous study, this paper analyses the effect of mushroom residues on N 2 O emission in Chengdu Plain.sionissionmeasure the effect of the With the popularization of rice-(straw)-edible mushroom cycling mode in Chengdu Plain and in many other rural areas, how to measure effect of the whole cycling mode and its different stages on greenhouse gasses emission like N 2 O or C 2 O emission need to be explored.Its impact on air environment would be a good direction of research.As this paper simply concentrates on differences of N 2 O emission under different fertilization, how C and N circulate in the farmland ecosystem and how organic matter accumulates in soil need to be deepened in further study.

Figure 2 .
Figure 2. Weekly valure of topsoil NO /NH among the wheat-rice rotation

Figure 4 .
Figure 4. Amount of the top soil temperature from each treatment in wheat-rice rotation

Table 2 .
The nutrientcontent of mushroom residues applied in rice and wheat season

Table 3 .
The overview of different fertilization for field (kg ha -1 ) 2.3 Measurement of Soil Temperature, Moisture, Biomass Measurement and NH 4+ -N, NO 3 -N Determination

Table 4shows
2 O emission from plots with crop straw applications and chemical fertilizer was greater than that from plots fertilized with mineral fertilizer.Moreover, it was two times greater than that from plots with MR application, except for the 200% MR application.

Table 5 .
Correlation matrix of N 2 O emission with soil temperature, NO 3