Effects of land-use type and nitrogen addition on nitrous oxide and carbon dioxide production potentials in Japanese Andosols

Abstract

Land-use type and nitrogen (N) addition strongly affect nitrous oxide (N₂O) and carbon dioxide (CO₂) production, but the impacts of their interaction and the controlling factors remain unclear. The aim of this study was to evaluate the effect of both factors simultaneously on N₂O and CO₂ production and associated soil chemical and biological properties. Surface soils (0–10 cm) from three adjacent lands (apple orchard, grassland and deciduous forest) in central Japan were selected and incubated aerobically for 12 weeks with addition of 0, 30 or 150 kg N ha^–1 yr^–1. Land-use type had a significant (p < 0.001) impact on the cumulative N₂O and CO₂ production. Soils from the apple orchard had higher N₂O and CO₂ production potentials than those from the grassland and forest soils. Soil net N mineralization rate had a positive correlation with both soil N₂O and CO₂ production rates. Furthermore, the N₂O production rate was positively correlated with the CO₂ production rate. In the soils with no N addition, the dominant soil properties influencing N₂O production were found to be the ammonium-N content and the ratio of soil microbial biomass carbon to nitrogen (MBC/MBN), while those for CO₂ production were the content of nitrate-N and soluble organic carbon. N₂O production increased with the increase in added N doses for the three land-use types and depended on the status of the initial soil available N. The effect of N addition on CO₂ production varied with land use type; with the increase of N addition doses, it decreased for the apple orchard and forest soils but increased for the grassland soils. This difference might be due to the differences in microbial flora as indicated by the MBC/MBN ratio. Soil N mineralization was the major process controlling N₂O and CO₂ production in the examined soils under aerobic incubation conditions.

Key words:

• CO₂ production
• land-use type
• N addition
• N₂O production
• soil aerobic incubation.

INTRODUCTION

Global warming is unequivocal and may affect many ecosystems, and greenhouse gases (GHGs) are one of the dominating forces for these changes (IPCC 2007). Carbon dioxide (CO₂) is one of the most important GHGs, increasing from the pre-industrial level of 280 ppm to 379 ppm in 2005 (IPCC 2007). Nitrous oxide (N₂O), another important GHG, has 298 times higher global warming potential than CO₂ over a 100-year time span (Forster et al. 2007), also leading to depletion of the stratospheric ozone layer (Ravishankara et al. 2009).

Changing land-use types and associated management practices via changing land-cover (e.g., plant species) or agricultural activities (e.g., fertilizer, weeding) can have significant and long-lasting effects on soil pH, carbon (C) and nitrogen (N) contents, and soil microbial activities (Post and Mann 1990; Murty et al. 2002), and consequently can have a significant impact on GHG emission from the soil (Iqbal et al. 2008). Generally, forest soils have a large amount of annual soil organic matter (SOM) input. Grassland soils, affected by weeding or grazing, have a small amount of SOM. Orchard soils under long-term fertilization have a large amount of soil nutrients. Numerous studies have shown that land-use type is a critical controlling factor for GHG fluxes (Lin et al. 2010; Lang et al. 2011). However, the majority of the previous studies determining the fluxes of N₂O and CO₂ emissions from soils are usually based on single-location studies with one land-use type, such as paddy (Singla and Inubushi 2013), upland (Nagano et al. 2012), or forest soils (Kim et al. 2012). Thus, comparisons based on observations from adjacent ecosystems with different land management under the same climate and soil conditions can provide a more robust understanding of the specific effects of land-use types on GHG production.

Over the last century, reactive N emissions have increased three- to five-fold (Denman et al. 2007), and its emissions and deposition worldwide are projected to increase by 50–100% by 2030, with the largest increase occurring over East and South Asia (Reay et al. 2008). The total annual N deposition at 12 monitoring sites throughout Japan ranged from 3.1 to 18.2 kg N ha^–1 yr^–1 during 2003–2007 (Japanese Ministry of the Environment 2009). Increased N deposition could, in turn, increase N₂O emissions by enhancing N availability in soils for microbial processes (Bowden et al. 1991; Butterbach-Bahl et al. 1997). In the available reports, N deposition into soils has shown variable effects on soil CO₂ emissions, including an increase (Brumme and Beese 1992), a decrease (Janssens et al. 2010) and unchanged (Micks et al. 2004), but the controlling factors are still not clear. Understanding soil respiration under a N-abundant environment is critical because relatively small changes in respiration rates may dramatically alter atmospheric CO₂ concentrations as well as soil C sequestration rates (Bowden et al. 2004).

The objectives of this study were to investigate GHG production as affected by land-use types and N inputs together under aerobic experimental conditions, and to find factors influencing the responses of N₂O and CO₂ productions to these treatments. This study was conducted using soils exposed to three land-use types (apple orchard subjected to intensive agriculture, non-agricultural grassland, and unmanaged deciduous forest) located adjacent to each other to eliminate confounding effects such as climatic conditions and soil types.

MATERIALS AND METHODS

Study site and soil sampling

Soils (0–10 cm) were collected from the Environmental Horticulture and Forestry Experimental Farm of Chiba University (36°36'N, 139°00'E, 780–810 m a.s.l.) Gunma Prefecture, Central Japan, in May 2011. The mean annual temperature and precipitation in this area from 2007 to 2011 were 12.2°C and 1244 mm, respectively (Japan Meteorological Agency 2012).

Three adjacent lands (apple orchard, grassland, and deciduous broad-leaved forest: hereafter referred to as A, G and F, respectively) were selected as soil sampling sites. The A soil was treated with orchard fertilizer containing 12% N, 10% potassium (K), 6% phosphorus (P), 3% magnesium (Mg) and 0.4% boron (B) at ca. 400 kg ha^–1 once a year at the end of April (JA Higashinihon Kumiai Shiryou Ltd., Gunma, Japan). Weeding under the apple trees in A and at the G land-use type was done once a month during the growing season. The F soil, which was unmanaged, was dominated by Quercus serrata Murray, Castanea crenata Siebold et Zucc., and Magnolia obovata Thunb. Soils in the three lands were classified as Andosols (FAO/UNESCO Soil Taxonomy), which is the dominant surface soil in upland regions of Japan (Wada 1986). Collected soils were sieved to 2 mm to remove pebbles and plant debris.

Analysis of chemical and microbial properties of soils

Air-dried soil (10 g) was mixed with 25 mL and 50 mL distilled water respectively to measure soil pH and electrical conductivity (EC). Soil total carbon (TC) and total nitrogen (TN) were measured after combustion using a CN analyzer (MT-700, Yanaco Analytical Instruments Co., Kyoto, Japan). For soil inorganic N (SIN) measurements, 15 g fresh soil samples were extracted with 50 mL of 1 mol L^–1 potassium chloride. Soil nitrate nitrogen (NO₃^–-N) and ammonium nitrogen (NH₄⁺-N) concentrations were then analyzed colorimetrically, using a spectrophotometer (UV-1200V, Shimadzu Co., Kyoto, Japan), by the hydrazine reduction method (Kamphake et al. 1967; Hayashi et al. 1997) and Berthelot reaction (salicylate-nitroprusside-hypochlorite method, Searle 1984; Anderson and Ingram 1993), respectively. Soluble organic carbon (SOC) and soluble nitrogen (SN) were extracted from a fresh soil sample of 15 g with 50 mL of 0.5 mol L^–1 potassium sulfate. Concentrations of SOC and SN were then determined using a TOC analyzer (TOC-5000, Shimadzu Co.), and the hydrazine reduction method (Kamphake et al. 1967; Hayashi et al. 1997) following alkaline persulfate oxidation (Cabrera and Beare 1993), respectively.

Soil microbial biomass carbon (MBC) and nitrogen (MBN) were determined by the chloroform fumigation-extraction method (Brookes et al. 1985; Vance et al. 1987). Contents of organic C and total N in the fumigated and unfumigated soils were determined by the method described above for SOC and SN. A factor of 2.22 was used as efficiencies of both microbial biomass C and N extractions (Jenkinson et al. 2004).

Determination of N₂O and CO₂ production rates and N mineralization rate of soils

A laboratory incubation experiment was conducted to evaluate the effects of N addition on N₂O and CO₂ production potentials simultaneously in the three land-use types. Soils from each land-use type were treated without N addition (N0), with 3.3 mg N kg^–1 dry soil (d.s.) 4 weeks^–1 (equivalent to 30 kg N ha^–1 yr^–1) (N30), and with 16.5 mg N kg^–1 d.s. 4 weeks^–1 (equivalent to 150 kg N ha^–1 yr^–1) (N150). The ammonium nitrate (NH₄NO₃) solution was supplemented as a N source and it was applied equally in three split doses to simulate repeating N deposition in the fields with an interval of four weeks. The amounts of N applied in this study were about 3 and 15 times higher than the annual N deposition based on the bulk precipitation 30 km east of the sampling sites (Ohrui and Mitchell 1997).

Prior to incubation, the water-holding capacities of each soil were adjusted to 60%. A 15-g sample of each soil was transferred into a 100-mL glass bottle with nine replicates, and then all bottles were sealed with butyl rubber stoppers and incubated at 25°C in the dark for 12 weeks. Production rates of N₂O and CO₂ were measured weekly, and the stopper was removed from the bottle for 3 h of ventilation after each gas sampling. After four weeks incubation, three bottles were used to determine various soil properties, and then the remaining six replicates had the same doses of NH₄NO₃ added, and were further incubated in the same conditions. Three replicates among the six were used for the measurement of soil properties at the end of 8 weeks’ incubation. After this, the last three replicates were again treated with the same doses of NH₄NO₃ and kept in the same incubated conditions for measurements of GHG productions and soil properties.

Concentrations of N₂O and CO₂ were measured using gas chromatographs (GC-14B, Shimadzu Co.) equipped with an electron capture detector and thermal conductivity detector, respectively. Cumulative GHG production was calculated as the sum of the GHG production rate from each week. The amount of net N mineralization (NM) was determined using the equation:

(1)

Net N mineralization rate (NMr) was calculated using the equation:

(2)

Statistical analysis

The effects of land-use type and N treatment on the cumulative N₂O and CO₂ production potentials were examined using the multivariate analysis of variance (MANOVA). Tukey’s highly significant difference (HSD) test was used to identify differences among means, and Pearson’s correlation was applied to examine the relationships between the cumulative GHG productions and soil properties. Stepwise multiple regression analysis was conducted to find out the main factors influencing cumulative N₂O and CO₂ production potentials. All statistical analyses were performed using SPSS Statistics 20 (IBM, New York, USA).

RESULTS

Production rates of N₂O and CO₂ in soils

In the A soils, the temporal pattern of the N₂O production rate was characterized by the highest rate during the first week, decreased gradually to week 6, and then remained nearly constant during weeks 7–12 (Fig. 1a). The N₂O production rate in G and F soils fluctuated and remained nearly constant during the whole incubation period, except for G N150 during weeks 1–4 as well as F N150 in the first week (Fig. 1b and 1c).

Figure 1 Changes in nitrous oxide (N₂O) and carbon dioxide (CO₂) production in response to three levels of nitrogen (N) addition. Bars indicate the standard error (n = 3). A, G and F indicate apple orchard soil, grassland soil and forest soil, respectively. N0, N30 and N150 indicate N addition levels of 0, 30 and 150 kg N ha^–1 yr^–1, respectively. Arrows indicate the time of solution addition. N₂O-N, nitrous oxide-nitrogen; CO₂-C, carbon dioxide-carbon.

The production rate of CO₂ in the A soils was relatively high during weeks 1–6, after which it decreased and remained stable from weeks 7–12 (Fig. 1d). The CO₂ production for the G and F soils reached the highest rate during week 1 and then decreased gradually through week 12 (Fig. 1e and f). The CO₂ production rate tended to decrease with increasing N for the A and F soils (Fig. 1d and f), while it tended to increase for the G soils (Fig. 1e).

Cumulative production potentials of N₂O and CO₂ in soils

Land-use types and N treatments had significant impacts on the cumulative production of N₂O (p < 0.001) (Table 1 and Fig. 2a). In soils without N addition, the cumulative N₂O production in the A soils [0.86 mg nitrous oxide-nitrogen (N₂O-N) kg^–1 d.s. 12 weeks^–1] was 4.1 and 3.3 times higher than those in the G soils (0.21 mg N₂O-N kg^–1 d.s. 12 weeks^–1) and the F soils (0.26 mg N₂O-N kg^–1 d.s. 12 weeks^–1), respectively (Fig. 2a). The cumulative production of N₂O in all studied soils tended to increase with increasing N doses (Fig. 2a). The A soils significantly responded to N additions, being highest in the N150 treatment (p < 0.05) (Fig. 2a).

Table 1 General linear model analysis of the effects of land use type and nitrogen (N) addition treatment on cumulative nitrous oxide (N₂O) and carbon dioxide (CO₂) production during 12 weeks of incubation (n = 3)

Variable	Degree of freedom	Cumulative N2O production (mg N2O-N kg^−1 d.s.)			Cumulative CO2 production (g CO2-C kg^−1 d.s.)
		Type III sum of squares	F value	P value	Type III sum of squares	F value	P value
Land use (L)	2	3.21	296.41	0.001	3.68	39.9	0.001
N addition (N)	2	0.19	17.84	0.001	0.14	1.51	0.248
L*N	4	0.07	3.09	0.042	0.54	2.93	0.050
Error	18	0.1			0.83

d.s., dry soil. N₂O-N, nitrous oxide-nitrogen; CO₂-C, carbon dioxide-carbon.

Figure 2 Cumulative nitrous oxide (N₂O) and carbon dioxide (CO₂) production by soils collected from areas subjected to different land uses during 12 weeks of incubation under three levels of nitrogen (N) addition. Bars indicate the standard error (n = 3). Different letters indicate significant differences at p < 0.05 as determined by Tukey’s highly significant difference (HSD) test. A, apple orchard soil; G, grassland soil; F, forest soil; N0, N30 and N150 indicate N addition levels of 0, 30 and 150 kg N ha^–1 yr^–1, respectively. N₂O-N, nitrous oxide-nitrogen; CO₂-C, carbon dioxide-carbon.

Land-use types also showed a significant impact on the cumulative production of CO₂ (p < 0.001) (Table 1 and Fig. 2b). In soils subjected to the N0 treatment, the cumulative CO₂ production of the A soils [3.3 g carbon dioxide carbon (CO₂-C) kg^–1 d.s. 12 weeks^–1] was 1.6 and 1.3 times higher than that of the G soils (2.1 g CO₂-C kg^–1 d.s. 12 weeks^–1) and F soils (2.6 g CO₂-C kg^–1 d.s. 12 weeks^–1), respectively. On the other hand, N addition did not significantly affect the cumulative CO₂ production (p > 0.05) throughout the 12-week incubation period (Fig. 2b).

Relationships between N₂O and CO₂ production and soil properties

Table 2 shows the Pearson correlation coefficients between cumulative N₂O and CO₂ production (N0 group), and some of the soil properties measured before incubation. Cumulative N₂O and CO₂ production exhibited a significant positive correlation with SIN, SN and EC and a negative correlation with soil pH and TC (Table 2). The mean production rates of N₂O and CO₂ in each land-use type had a significant positive correlation with the mean net N mineralization rate for all treatments (Fig. 3a and b). Furthermore, a positive correlation between CO₂ and N₂O production rate was detected (R² = 0.692, p = 0.003, Fig. 3c), which is consistent with the study of Chen et al. (2012) who also observed a significant positive correlation between CO₂ flux and N₂O flux in situ. However, the key soil parameters affecting the both GHG productions were different. The stepwise multiple regression analyses showed that among all measured soil parameters, the concentrations of NH₄⁺-N and the MBC/MBN ratio significantly affected N₂O production, while significant parameters influencing the cumulative CO₂ production were NO₃^–-N content and SOC. The regression models for both cumulative productions were: N₂O = 0.577 + 0.167 NH₄⁺-N – 0.025 MBC/MBN (R² = 0.997, p < 0.001) and CO₂ = 1.18 + 0.059 NO₃^–-N + 0.003 SOC (R² = 0.923, p < 0.001), respectively.

Table 2 Pearson correlation coefficients between cumulative gas productions [nitrous oxide (N₂O) or carbon dioxide (CO₂)] from the soils without nitrogen (N) addition and initial soil properties under the three land use types (n = 9)

Variable	pH	EC	TC	TN	C/N ratio	NO3^−-N	NH4^+-N	SOC	SN	MBC	MBN
N2O	−0.928^**	0.972^**	−0.890^**	0.188	0.548	0.988^**	0.993^**	0.598	0.936^**	0.728^*	0.030
CO2	−0.795^*	0.816^**	−0.787^**	−0.413	0.203	0.886^**	0.866^**	0.838^**	0.817^**	0.407	0.405

^*and ^**indicate significant effects at p < 0.05 and p < 0.01, respectively. EC, electrical conductivity; TC, total carbon; TN, total nitrogen; C, carbon; N, nitrogen; NO₃^–-N, nitrate-nitrogen; NH₄⁺-N, ammonium-nitrogen; SOC, soluble organic carbon; SN, soluble nitrogen; MBC, microbial biomass carbon; MBN, microbial biomass nitrogen.

Figure 3 Correlation between net nitrogen (N) mineralization rate and carbon dioxide (CO₂) and nitrous oxide (N₂O) production rates with three levels of N addition under three land uses. A, apple orchard soil; G, grassland soil; F, forest soil; N0, N30 and N150 indicate N addition levels of 0, 30 and 150 kg N ha^–1 yr^–1, respectively. N₂O-N, nitrous oxide-nitrogen; CO₂-C, carbon dioxide-carbon.

DISCUSSION

Effect of land-use type on production potentials of N₂O and CO₂

In the present study, a standardized incubation temperature (25°C) and soil moisture condition (60% water-holding capacity) was used. These are considered to be effective in reducing the masking effects of soil temperature and water conditions on N₂O and CO₂ production, enabling the effects of other soil factors to be determined (Gödde and Conrad 2000). The reason for the largest N₂O production in orchard soils might be the high initial mineral N levels according to chronic fertilizer application in the fields, which is considered a key factor influencing N₂O emissions (Robertson et al. 2000; Mathieu et al. 2006). Land use types may directly alter soil properties, especially the biological properties (Merino et al. 2004), and thus indirectly affect GHG emission (Ishizuka et al. 2002). In turn, the correlation analysis between both cumulative GHG productions and the soil properties could become a carrier; this could indirectly reflect the impact of land-use types on both N₂O and CO₂ emissions. For example, chronic fertilizer use in the orchard fields resulted in significant decreases of soil pH and TC but an increase in soil EC (Table 3). Inubushi et al. (1996) suggested that at 60% water-holding capacities, ~90% of N₂O production was derived from nitrification. Hence, it was suggested that nitrification, the oxidation of ammonium to nitrate, should be a more dominant process of soil N₂O production in the present study. The stepwise multiple regression analysis also indicated that NH₄⁺-N concentration in soil was the key contributing factor, accounting for 98.3% variation to N₂O production. Martikainen and De Boer (1993) suggested, from an aerobic incubation study with forest soils, that N₂O production associated with soil nitrification would decrease when the pH is increased from 4 to 6. The present study agreed with this point, and a negative correlation between soil pH and cumulative N₂O production was observed among the three land-use types (Table 3). Under incubation conditions, soil CO₂ emissions are mainly caused by microbial respiration (Janssens et al. 2001), which is primarily controlled by the supply of readily decomposable SOM (Rustad et al. 2000). Land-use types could directly affect the concentrations of soil SIN and SOC (Table 3), and consequently affect soil CO₂ productions. The stepwise multiple regression analysis showed that soil NO₃^–-N and SOC content were the key contributing factors, accounting for 75.4% and 16.9% of the variations in CO₂ production, respectively.

Table 3 Chemical and microbial properties of initial soils at the three experimental sites

Land use	Apple orchard	Grassland	Forest
pH (H2O)	4.8 ± 0.3^b	5.5 ± 0.4^a	5.4 ± 0.2^a
EC (m S m^−1)	16.2 ± 2.9^a	7.6 ± 0.9^b	6.1 ± 0.8^b
TC (g kg^−1 d.s.)	93 ±13^c	124 ± 16^a	113 ± 11^b
TN (g kg^−1 d.s.)	8.5 ± 1.2^b	10.6 ± 1.6^a	7.2 ± 0.7^c
C/N ratio	11.0 ± 0.3^c	11.8 ± 0.3^b	15.8 ± 0.2^a
NO3^−-N (mg kg^−1 d.s.)	15.1 ± 4.8^a	3.5 ± 0.0^b	4.2 ± 0.8^b
NH4^+-N (mg kg^−1 d.s.)	4.1 ± 0.7^a	0.7 ± 0.1^b	0.7 ± 0.2^b
SOC (mg kg^−1 d.s.)	358 ± 57^a	198 ± 42^b	354 ± 32^a
SN (mg kg^−1 d.s.)	21 ± 6^a	12 ± 3^b	10 ± 1^b
MBC (mg kg^−1 d.s.)	737 ± 149^b	930 ± 97^ab	990 ± 121^a
MBN (mg kg^−1 d.s.)	51 ± 16^a	47 ± 6^a	59 ± 17^a
MBC/MBN ratio	16 ± 1.9^b	19.7 ± 1.5^a	17.1 ± 1.1^b

Different letters in the same column indicate a significant difference among the three land use types as determined by Tukey’s highly significant difference (HSD) test (p < 0.05, n = 3). H₂O, water; d.s., dry soil; EC, electrical conductivity; TC, total carbon; TN, total nitrogen; C, carbon; N, nitrogen; NO₃⁻-N, nitrate nitrogen; NH₄⁺-N, ammonium nitrogen; SOC, soluble organic carbon; SN, soluble nitrogen; MBC, microbial biomass carbon; MBN, microbial biomass nitrogen.

Intensive soil management has led to a considerable increase in the rates of GHG exchange between soils and the atmosphere (Bouwman 1990). Conversely, less intensive management improves biological properties (Emmerling et al. 2001), and conversion of agricultural land to grassland and forest usually results in considerable gains in soil microbial biomass and reductions in N₂O and CO₂ production (Paul et al. 2002). Agricultural soils often have higher CO₂ emissions than soils under native vegetation (Iqbal et al. 2010). Conversion from woodland to orchard could significantly increase soil CO₂ emissions (Iqbal et al. 2008) and CO₂ emissions from orchard soils have been found to be higher than those from forest soils (H. Liu et al. 2008). The present results also showed this as the apple orchard was subjected to intensive management and had higher soil cumulative N₂O and CO₂ production potentials than non-agricultural grassland and unmanaged forest (p < 0.001) (Fig. 2a and b). On the other hand, the orchard soils showed higher metabolic quotients (qCO₂) than the G and F soils (p < 0.05) (Table S1). This indicated that agricultural soils had substantial effects on soil microbial biomass C utilization efficiency under experimental conditions.

Combined effects of N addition and land-use type on the production potentials of N₂O and CO₂

Many studies, including the present one, have suggested that N addition is a key parameter that enhances the exchange rate of N₂O between soil and the atmosphere (Butterbach-Bahl et al. 1997; Ullah and Zinati 2006; Li et al. 2012). Liu and Greaver (2009) reported that the addition of N at rates of 10 to 562 kg N ha^–1 yr^–1 significantly increased soil N₂O emissions by an average of 216% across agricultural and non-agricultural ecosystems. Soil N₂O production was not only related to the rates of nitrification and denitrification, but also to soil N status and soil C/N ratio (Dise et al. 1998; Zhang et al. 2009). For the A soil with a lower soil C/N ratio (Table 3), N₂O production significantly increased with the N150 addition rate (p < 0.05) (Fig. 2a) but the G soil that also had a lower C/N ratio did not exhibit a significant increase with N addition (Fig. 2a). In addition, in the F soil with a higher soil C/N ratio (Table 3), N addition did not significantly affect N₂O production (Fig. 2a). SIN and SN are important sources of available N in the soil N pool (Denman et al. 2007), having large dynamics before and after incubation (Table 3 and Table S1). On the other hand, the present study observed positive correlations between the cumulative N₂O production and soil SIN and SN (Table 2), which is in accordance with Zhang et al. (2009). Considering these points, the conversion factor of N₂O from SIN as defined in Table 4 was calculated to explore the role of available N during N₂O production. In the N0 group, the conversion factor of N₂O/SIN in the A soil was a little lower than that in G and F soils, but in the N30 and N150 groups, the conversion factor in the A soil was significant higher than in G and F soils (Table 4). This suggested that N₂O production was more sensitively affected by N additions in the A soil having initial higher SIN concentrations (Table 3) than in G and F soils (Fig. 4). The added mineral N in the G and F soils, having lower initial concentrations of SIN and SN, was generally immobilized by microbes as their metabolic materials (Janssen 1996). Thus, the remaining N was not sufficient to increase soil N availability, and could not lead to the increases in N mineralization, nitrification or denitrification, and consequently it was not enough to influence the N₂O production rate (Bowden et al. 1991).

Table 4 Nitrous oxide (N₂O) conversion factor according to soil inorganic nitrogen (N) under three land-use types

Variable	N0	N30		N150
	CF	CF	NCF	CF	NCF
Apple orchard	4.56 ± 0.48^b	3.39 ± 0.18^a	1.20 ± 0.07^a	1.76 ± 0.16^a	0.70 ± 0.19^a
Grassland	5.00 ± 0.11^b	1.79 ± 0.18^b	0.41 ± 0.19^b	0.77 ± 0.08^b	0.40 ± 0.07^b
Forest	5.61 ± 0.18^a	1.85 ± 0.12^b	–0.02 ± 0.11^c	0.63 ± 0.03^b	0.13 ± 0.04^b

Different letters in the same column indicate a significant difference among the three land-use types as determined by Tukey’s highly significant difference (HSD) test (p < 0.05, n = 3).N0, without nitrogen addition; N30, with 9.9 mg N kg^–1 d.s. 12 weeks^–1 (equivalent to 30 kg N ha^–1 yr^–1); N150, with 16.5 mg N kg^–1 d.s. 4 weeks^–1 (equivalent to 150 kg N ha^–1 yr^–1). CF, conversion factor is calculated by cumulative N₂O production/(SIN + N addition); NCF, net N fertilizer conversion factor is calculated by cumulative N₂O production of (N treat – N0)/N addition.

Figure 4 Regression analyses between cumulative nitrous oxide (N₂O) production and nitrogen (N) addition. A, G and F indicate apple orchard soil, grassland soil and forest soil, respectively. N₂O-N, nitrous oxide-nitrogen.

In the A soil, during weeks 1–4, N₂O production rate was enhanced by N addition rates significantly (p < 0.05); the cumulative N₂O production amount in this period contributed almost 50% of that of the entire incubation period (Fig. 1a). In the present study, soil organic matter such as SOC, microbial biomass (e.g., MBC and MBN) and NH₄⁺-N decreased gradually with incubation time (Table 3 and Table S1), which was in accordance with the study of Lang et al. (2011). The first instance of N fertilization in the A soil could be rapidly increasing the available N for nitrifying and denitrifying microorganisms (Davidson et al. 2000; Nishina et al. 2009), and thus increasing N₂O production significantly. With the incubation time, interacting with N addition, N₂O production in the A soil did not increase significantly after the second and third N additions (Fig. 1a). These results indicated that the rates of organic matter decomposition, microbial activity and N mineralization were fast in the first six weeks of the laboratory incubation experiment; after this period, soil microbial activity and N transformation may have reached an equilibrium, so that N additions did not significantly increase N₂O production any more.

The effect of N addition on CO₂ emission was unknown in different land-use types and ecosystems. Li et al. (2012) reported that elevated N deposition led to a significant increase in the emission of CO₂ during the growing season in an alpine grassland. Litton et al. (2007) reported that increased N deposition could reduce allocation to fine roots and then reduce soil respiration rates (Janssens et al. 2010). Additionally, Ambus and Robertson (2006) showed that CO₂ efflux was not affected by increased N inputs in an unmanaged forest and grassland. In this study, laboratory incubation of root-free soil showed that heterotrophic respiration from the microbial community in the A and F soils was reduced with N additions (Fig. 1d and f), while it was increased in the G soil (Fig. 1e). Significant differences caused by short-term N additions were not observed in the aerobic incubation conditions (Fig. 2b). Similar dynamical trends were also found for qCO₂ (Table S1), indicating N addition had no significant effects on soil microbial biomass C utilization efficiency. N inputs can reduce the amount of soil microorganisms and their activities (Compton et al. 2004; Frey et al. 2004), and thus restrict CO₂ production from the A and F soils. Some studies considered that N additions would increase N mineralization rate and thus promote CO₂ emission (Hobbie 2000). Similar findings were found in the G soil in that N mineralization rate (Fig. 3b) increased with the N addition dose associated with higher CO₂ production. A nearly equal ratio of soil MBC/MBN from A and F soils (Table 3) may indicate their similar microbial flora, the rates of microbial N and C turnover and thus similar response of CO₂ to N addition (Fig. 2b). Therefore, further studies on soil microbial flora and the situ field conditions under the three land-use types are suggested.

CONCLUSION

Land-use type significantly influenced the cumulative production potentials of N₂O and CO₂ through its effects on soil biological and biochemical properties (p < 0.001). Apple orchard soils that were subjected to intensive agricultural techniques showed significantly increased production potentials of the two major GHGs relative to soils from a non-agricultural grassland and unmanaged mixed forest soils. Furthermore, the response of N₂O production to N addition was likely due to the soil available N status, especially at the early stages of N addition, and was relatively larger in apple orchard soil that received fertilizer N than in the other two land-use soils. Soil CO₂ production exhibited various responses to additional N, and increased in grassland soils but decreased in the apple orchard and forest soils. The soil N mineralization process, i.e., organic matter decomposition, accounted for the majority of N₂O and CO₂ production under the aerobic incubation conditions in all land-use soils.

SUPPLEMENTARY MATERIAL

Supplementary material for this article is available online from http://dx.doi.org/10.1080/00380768.2013.817938

ACKNOWLEDGMENTS

This study was supported in part by a Grant-in-Aid for Young Scientists (B) (No. 22780150). The first author thanks the Ministry of Education, Culture, Sports, Science and Technology, Japan, for providing financial support. The authors also thank Dr. A. Takenaka of the National Institute for Environmental Studies, Japan, for providing information regarding the dominant plant species in the forest. Finally, we thank Mr. J. Hara of Chiba University, Japan, for providing the field fertilization information.