Vertical distribution of denitrification end-products in

0 downloads 0 Views 2MB Size Report
and Regulations. IOP Publising Agrisfaoweb. http://agris.fao.org/agris-search/search. do?recordID=US201300114453 (Accessed October 2005). McCarty, G.W. ...
Science of the Total Environment 576 (2017) 462–471

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Vertical distribution of denitrification end-products in paddy soils Wei Zhou a,b, Longlong Xia a,b, Xiaoyuan Yan a,⁎ a b

State Key Laboratory of Soil and Sustainable Agriculture, Institute of Soil Science, Chinese Academy of Sciences, Nanjing 210008, China University of Chinese Academy of Sciences, Beijing 10049, China

H I G H L I G H T S

G R A P H I C A L

A B S T R A C T

• Indirect N2O emission via leachate was comparable to the IPCC default value. • The loss of N through denitrification can be underestimated if research is focused only on the surface paddy soil and the extent to which denitrification is underestimated may be affected by the application of organic carbon.

a r t i c l e

i n f o

Article history: Received 17 July 2016 Received in revised form 14 October 2016 Accepted 18 October 2016 Available online xxxx Editor: D. Barcelo Keywords: Denitrification Excess N2 Dissolved nitrogen oxides Soil profiles Paddy fields

⁎ Corresponding author. E-mail address: [email protected] (X. Yan).

http://dx.doi.org/10.1016/j.scitotenv.2016.10.135 0048-9697/© 2016 Elsevier B.V. All rights reserved.

a b s t r a c t Knowledge of denitrification process and its end-product at various depths of paddy soil is very important for our understanding of its role in mitigating reactive N and indirect nitrous oxide (N2O) emission. In this study, the end-products of denitrification were determined at four depths in a long-term field lysimeter experiment in southeast China over a rice season. Three treatments were included: (1) chemical fertilizer (NPK); (2) NPK plus pig manure (NPKM); and (3) NPK plus straw (NPKS). The concentration of dissolved N2O increased with soil depth across all treatments and the highest concentration of excess dinitrogen (N2) was observed at 0.2 m depth, as was the highest dissolved organic carbon (DOC) content. Denitrification reduced the amount of nitrate by 48–54% in the paddy soil profile, especially at 0.2 m depth (68–88%), whereas the lower reduction of NO− 3 (17–44%) in the subsoil (at 0.6 and 0.8 m depth) was accompanied by a higher concentration of NO− 3 . Our results demonstrated that DOC was the major limiting factor of denitrification in the subsoil. The application of pig manure markedly increased the amount of DOC in the surface soil, resulting in a high rate of denitrification, whereas the addition of straw had no effect on denitrification. The indirect emission factors for N2O (EF5-g, 0.001–0.006) were comparable with the default value (0.0025) reported by the Intergovernmental Panel on Climate Change. The low N2O production was probably caused by the complete reduction of N2O to N2, as reflected by the lower N2O/(N2O + N2) ratios in the paddy soil profile. Although the surface soil was identified as a hotspot for denitrification, a considerable amount of excess N2 was observed in the subsoil for all three treatments. We therefore conclude that the loss of N through denitrification may be significantly underestimated if only the surface soil is considered. © 2016 Elsevier B.V. All rights reserved.

W. Zhou et al. / Science of the Total Environment 576 (2017) 462–471

1. Introduction The contamination of groundwater by nitrate (NO− 3 ) is of global concern as a result of its impact on both the environment and human health (Galloway et al., 2008). High concentrations of NO− 3 in streams, lakes, and rivers increase the risk of eutrophication in surface waters (Stark and Richards, 2008). The contamination of surface water and groundwater with NO− 3 is common in watersheds dominated by agricultural activities, mainly as a result of diffuse pollution from intensive farming (Foster and Young, 1980; Townsend et al., 2003). The leaching of NO− 3 to groundwater below arable land in rice–wheat rotation systems has been reported in China (Zhu and Wen, 1992; Zhu and Chen, 2002). It has been recognized that a variety of processes such as denitrification and nitrate reduction to ammonium (DNRA) are involved in the nitrate dynamics, however the mitigation of excess NO− 3 in paddy soil generally relies on denitrification (Jarvis, 2000; Bernard et al., 2015; Shan et al., 2016). Denitrification is a multistep biological process producing nitrite (NO− 2 ), nitric oxide (NO), nitrous oxide (N2O), and dinitrogen (N2) from NO− 3 and mainly occurs in the anaerobic zone (Jahangir et al., 2013). Four requirements must be met to trigger denitrification. In addition to the presence of NO− 3 as the substrate, there needs to be an anaerobic environment, an electron donor, and microbes capable of denitrification. As NO− 3 and suitable microbes are ubiquitous in arable land, denitrification thus mainly depends on the existence of anaerobic conditions and the availability of electron donors (Seitzinger et al., 2006). Denitrification in the soil profile has been widely studied in wetland, grassland, and upland soils. The soil redox conditions and the presence of dissolved organic carbon (DOC) are the major controlling factors of denitrification in upland soil profiles (Sotomayor and Rice, 1996; Elmi et al., 2003; Elmi et al., 2005; Mathieu et al., 2006). For example, a high accumulation of NO− 3 in the soil profile has been reported in Chinese semi-humid croplands, where NO− 3 cannot be denitrified due to the presence of oxygen and a lack of carbon sources (Zhou et al., 2016a). However, in flooded environments, such as wetland soils and in groundwater, the rate of denitrification is mainly related to the amount of DOC (Hill and Cardaci, 2004; Xiong et al., 2006; Rivett et al., 2008; Minamikawa et al., 2010; Khalil and Richards, 2011). Many studies have observed the highest rates of denitrification in the upper soil horizon (0–0.2 m), the extent of which depends on the moisture content (Küstermann et al., 2010; Clément et al., 2002; Friedl et al., 2016). However, higher rates of denitrification in the subsoil (0.15–0.3 m) than in the surface soil (0–0.15 m) have also been reported (McCarty and Bremner, 1992; Paul and Zebarth, 1997). There is now increasing evidence of significant denitrification activity in subsoils (Hill and Cardaci, 2004; Khalil and Richards, 2011; Dixon et al., 2010) and subsoil denitrification has been suggested as an important mechanism for the removal of excess NO− 3 before it is leached to groundwater or discharged to the surface aquifer (Sotomayor and Rice, 1996; Fenton et al., 2009; Bernard-Jannin et al., 2016). Paddy soils are widely distributed across south and northeast China. Rice requires a 0.03–0.07 m deep layer of water during most of its growing season, which results in the development of a reducing layer under the root zone. The low, flat landform, loamy to clayey parent materials, and a long cultivation history of rice have led to the development of a unique soil profile in paddy soils that is conducive to denitrification (Li et al., 2014; Xing et al., 2002). Thus the denitrification process is different in paddy soil profiles from that in upland soils, where denitrification is limited due to the presence of oxygen. Complete denitrification in paddy soil profiles has been reported based on the observation of very low concentrations of N2O in leachates or the attenuation of nitrate during infiltration (Zhu et al., 2003; Xiong et al., 2006; Choi et al., 2013; Majumdar, 2013), though the evidence was insufficient because the concentration of N2 was not determined. As serious environmental problems are caused by high inputs of reactive N to agricultural land

463

in China, a knowledge of the factors controlling the denitrification process and the N2O/(N2 + N2O) ratio are crucial to our understanding of the extent of mitigation of NO− 3 via denitrification in paddy soil profiles (Zhu and Chen, 2002; Ju et al., 2009; Xia and Yan, 2012). Few studies have focused on denitrification in paddy soil profiles as a result of the methodological difficulties in the direct quantification of the end-products of denitrification. There are two known difficulties in the quantification of denitrification in soil profiles. First, the high background levels of atmospheric N2 make it difficult to determine the end-product of denitrification. Second, under field conditions, O2, NO− 3 , and DOC, factors that control denitrification in the deep soil layers, are diffused or leached from the upper soil layers. This differs from the results obtained by the incubation of soil cores, especially for paddy soils. Therefore it is unsuitable to investigate soil profile denitrification by soil core incubation of different layers. The lack of information about specific areas in paddy soil profiles with high or low rates of denitrification hinders the accurate quantification of the loss of N during the process as a whole (Groffman et al., 2006; Davidson and Seitzinger, 2006). N2O is an obligate intermediate in the denitrification process and N2 is the final product in this process. Thus the quantification of dissolved N2O and excess N2 can be used as a tool to investigate denitrification (Kana et al., 1994; Böhlke, 2002; Weymann et al., 2008). Membrane inlet mass spectrometry (MIMS) can be used to measure a large number of dissolved gases by auto-degassing water samples with almost no interference from the high background levels of atmospheric N2. It can thus be used to quantify denitrification in situ at low cost and has been used to study denitrification in water bodies, including groundwater and sediments (Weymann et al., 2008; Jahangir et al., 2014; Chen et al., 2014). In this study, MIMS was used to determine the N2 produced in profile from a flooded paddy soil. Although variety processes such as anaerobic oxidation of ammonia (Aanammox), chemodenitrification and Feammox (ammonium oxidation coupled to iron reduction) can contributed to N2 formation, denitrification was recognized as the major process to produce N2 in paddy soil (Hou et al., 2015; Zhu et al., 2011; Shan et al., 2016). As it is impossible to separate and to quantify the contribution of the N2 production from different processes in this study. Thus N2 production was not distinguished from denitrification and other processes in present study; instead, the net production of N2 was referred to as “denitrification” for simplicity (Li et al., 2014; Jahangir et al., 2014; Chen et al., 2014). We hypothesized that denitrification in the paddy soil profile might significantly reduce NO− 3 before it was leached to the groundwater or discharged into surface water bodies and that the N2O/(N2O + N2) ratios in the deeper layers would be affected by the availability of NO− 3 and DOC leached from the upper layers. The objectives of this study were (1) to investigate the distribution of denitrification in a paddy soil profile by measuring the concentrations of N2O and excess N2 at various vertical depths and (2) to link the denitrification end-product with biogeochemical parameters.

2. Materials and methods 2.1. Study sites The experiments were conducted from June 23 to October 23, 2015 in the long-term field lysimeter experiment at the Changshu Agroecological Experimental Station (31° 32′ N, 120° 41′E) of the Chinese Academy of Sciences in Jiangsu Province. The soil is classified as a hydromorphic paddy soil; the surface soil (0–20 cm) of the experimental plots was sampled on June 1, 2015 for the determination of physical and chemical properties and the results are listed in Table 1. The annual mean air temperature in this region is 25.6 °C and the annual mean precipitation is 1054 mm. The average groundwater table is below 0.5 m during the summer and below 0.8 m during the winter.

464

W. Zhou et al. / Science of the Total Environment 576 (2017) 462–471

2.2. Field treatment and management The long-term field lysimeters were built in 1998 and include the following three treatments: (1) NPK (chemical fertilizers N, P, and K); (2) NPKM (incorporation of fresh pig manure at 15 t season−1 plus chemical fertilizers); and (3) NPKS (incorporation of harvested rice/wheat straw at 2.25 t dry weight ha−1 plus chemical fertilizers). Each lysimeter plot has an area of 4 m2 (2 × 2 m) and a soil depth of 1 m. The side wall is 0.25 m above the soil surface. All of the treatments were laid out in a randomized block design with three replicate samples. The long-term experiments were conducted a rice-wheat rotation. In the rice season, the field lysimeters are typically continuously flooded for one month after the basal fertilizer. Then, an episode of midseason drainage for a week is imposed, followed by intermittent irrigation until the rice harvest. During the wheat season, the water supplies via precipitation only. In rice season, urea was applied at a rate of 300 kg N ha−1 for the NPK and NPKS plots and at a rate of 225 kg N ha−1 for the NPKM plots, with a split ratio of 4:3:3 for the basal fertilizer, tillering fertilizer, and panicle fertilizer, respectively. In the NPKM plots, pig manure containing 75 kg N ha−1 was applied as the basal fertilizer. The P and K fertilizers were applied as basal fertilizers at rates of 60 kg P2O5 ha−1 and 120 kg K2O ha−1, respectively. Rice (cultivar Nangeng 46) was transplanted on June 23, 2015 and harvested on November 11, 2015 in present study.

where X denotes the molar concentration of the parameters, XN2T represents the molar concentration of the total dissolved N2 in the water sample, XN2EA is N2 originating from “excess air”, i.e. dissolved gas components in excess of equilibrium and known subsurface gas sources entrapped in air bubbles during recharge, and XN2EQ is the molar concentration of dissolved N2 in equilibrium with the atmospheric concentration. To determine dissolved N2O, 5 mL sample was injected into an already vacuumed 18.5 mL glass vial (SVF-18.5 vacuum vials, NichidenRika Glass Co, Ltd., Japan) and ambient air was then allowed to flow into the vial, which was subsequently sealed and stored in a refrigerator at 4 °C for 24 h to equilibrate. About 10 mL gas sample was drawn from the vial for the determination of N2O using an Agilent 7890 gas chromatograph. The concentration of dissolved N2O was calculated using the equation reported by Weiss and Price (1980). + NO− 3 and NH4 were determined in filtered (cellulose acetate 0.45 μm) samples using a SmartChem 140 discrete auto-analyzer (Westco Scientific Instruments) and the DOC was determined using an N/C 3000 multi-analyzer (Analytik Jena AG, Germany). All samples were stored at 4 °C and analyzed within 7 days. 2.4. Estimation of emission factor of indirect N2O, reaction progress, and N2O mole fraction The emission factor of indirect N2O emission (EF5-g) from the leachate samples was estimated using the method reported by the IPCC (2006) and was calculated using Eq. (2):

2.3. Sampling and measurement Leachate collection boxes (0.1 × 0.15 × 0.1 m) were buried at depths of 0.2, 0.6, and 0.8 m in the middle of each lysimeter during building (Fig. 1). A stainless steel pipe (diameter 1 cm) was connected to each of the buried boxes and used to sample the leachates. Leachate samples were obtained from the buried boxes through a medical infusion tube (diameter 0.4 cm) connected to the stainless steel pipe. Surface water samples were obtained using a peristaltic pump (BT100M, Baoding Chuangrui Precision Pump Co. Ltd., China). Sampling was carried out once a day for a 10day period (from June 24 to July 2) after the application of the basal fertilizer and then twice per month until the rice was harvested. Triplicate samples were collected at a slow rate (50 mL min−1) to avoid the ebullition of dissolved gases during sampling. To determine the amount of dissolved N2 and Ar, samples were collected in 7-mL glass vials (Extainer, Labco Ltd., Buckinghamshire, UK) by slowly overflowing about 15 mL of excess water. A 100 μL volume of saturated HgCl2 solution was added to inhibit microbial activity and the glass vials were then immediately sealed with no headspace or bubbles. An additional 100 mL sample was collected in a plastic bottle to determine the amount of dissolved N2O, DOC, NO− 3 , and NH+ 4 . The temperature, pH, oxidation reduction potential (Eh), and dissolved oxygen (DO) of the surface water and leachates were measured in the field using a YSI Exo 1 Multiparameter probe. Dissolved N2 and Ar were determined using a high-precision MIMS (Bay Instruments, Eston, MD, USA). This instrument has previously been used to measure soil denitrification in paddy fields and the ability of river networks to remove nitrogen (Li et al., 2014; Zhao et al., 2015). Denitrified N2 expressed as excess N2 was estimated following the method described by Weymann et al. (2008). Xexcess N2 ¼ XN2 T −XN2 EA −XN2 EQ

ð1Þ

EF5−g ¼

N2 O NO3 −

ð2Þ

The reaction progress (RP) is the ratio between the amount of products and substrates (Weymann et al., 2008) and can be used to characterize the extent of NO− 3 elimination by denitrification (Böhlke, 2002). The RP was calculated using Eq. (3): RP ¼

TDN TDN þ NO3 −

ð3Þ

where the total denitrification (TDN) value is the sum of the denitrification end-products (N2O plus excess N2). The N2O mole fraction is the ratio of N2O to the TDN and was estimated using Eq. (4): N2 O mole fraction ¼

N2 O TDN

ð4Þ

EF5-g, RP and N2O mole fraction list in Table 2 were the mean values of every sample event over the whole rice season. 2.5. Statistical analysis The Kruskal–Wallis one-way analysis of variance (ANOVA) was used to determine the effects of the different treatments and the soil depth on the values for NO− 3 , N2O, excess N2, RP, EF5-g, and the N2O mole fraction. The differences among means were evaluated by a non-parametric multiple comparison test (Dunn's test p b 0.05). Spearman correlation analysis was used to investigate the relationship between environmental

Table 1 Properties of surface soil in the experiment plots. Treatment

pH (H2O)

Alkaline N (mg kg−1)

Total N (g kg−1)

Available P (mg kg−1)

Total P (g kg−1)

Available K (mg kg−1)

Total K (g kg−1)

Total organic carbon (g kg−1)

NPK NPKM NPKS

5.5 ± 0.1 6.6 ± 0.3 5.4 ± 0.1

149.2 ± 14.2 159.0 ± 18.6 128.8 ± 11.7

1.5 ± 0.2 2.0 ± 0.2 1.7 ± 0.1

28.1 ± 1.2 122.9 ± 23 40.4 ± 1.7

0.6 ± 0.1 2.0 ± 0.1 0.8 ± 0.1

130.5 ± 23.6 112.4 ± 8.4 143.7 ± 2.7

18.3 ± 1.9 16.8 ± 1.1 16.2 ± 2.7

15.4 ± 1.3 21.3 ± 2.1 17.6 ± 0.2

Data presented as mean ± standard error, n = 3.

W. Zhou et al. / Science of the Total Environment 576 (2017) 462–471

465

Fig. 1. Schematic diagram of a field lysimeter plot. Surface water samples were obtained using a peristaltic pump. Leachates were collected using a medical infusion tube connected to stainless steel pipes at three depths. The flow rate in each pipe was regulated by a valve.

factors and excess N2 and N2O. All statistical analyses were performed using SPSS version 19.0 (SPSS Inc., Chicago, IL, USA) and the graphs were created using SigmaPlot version 12.0 (Systat Software Inc., San Jose, CA, USA). 3. Results

between the NPK and NPKS treatments. Contrary to the NO_3, an obvious negative relationship between NH+ 4 concentration and soil depth was showed in all treatments (Fig. 2b). Irrespective of depth, mean NH+ 4 were 1.93, 1.29 and 2.56 mg N L−1 for the NPK, NPKM, and NPKS treatments, respectively. Significantly higher NH+ 4 concentration was observed in surface water in all treatments (p b 0.001), no significant difference was found among treatments.

3.1. Nitrate and ammonium concentrations in leachate and surface water samples

3.2. Dissolved N2O and excess N2

The nitrate concentration was significantly affected by the different fertilizer treatments and the soil depth (p b 0.001; Fig. 2a). An obvious increase in NO− 3 concentration with soil depth was observed in the NPKM and NPKS treatments. For the NPK treatment, the lowest NO− 3 concentration was found at 0.2 m depth. The maximum NO− 3 concentration was at 0.8 m depth for all treatments, indicating the accumulation − of NO− 3 in the subsoil. Irrespective of soil depth, the mean NO3 concen−1 for the NPK, NPKM, and trations were 0.94, 3.80, and 1.04 mg N L NPKS treatments, respectively. The highest NO− 3 concentration was observed with the NPKM treatments; no significant difference was found

The N2O concentration varied significantly between both the different treatments and at different depths (p b 0.001; Fig. 3a) and an obvious rising trend in N2O concentration with soil depth appeared in NPKM treatment. Irrespective of soil depth, the mean N2O concentrations over the rice season were 0.64, 5.80, and 0.90 μg N L−1 for the NPK, NPKM, and NPKS treatments, respectively, with the NPKM treatment showed significantly higher N2O concentration than the other two treatments. Temporal variations in the N2O concentration were remarkable across both the treatment and the soil depth (Fig. 4, left-hand panel). The N2O concentrations increased from June to August and then decreased

Table 2 Reaction progress (RP), N2O emission factor, and N2O mole fraction across the soil profiles. Treatment

Depth (m)

RPa

NPK

Surface water 0.2 0.6 0.8 Mean Surface water 0.2 0.6 0.8 Mean Surface water 0.2 0.6 0.8 Mean

0.69 0.88 0.18 0.44 0.54 0.72 0.78 0.30 0.17 0.48 0.68 0.73 0.41 0.26 0.51

NPKM

NPKS

EF5-g ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.22Aa 0.07Aa 0.29Ac 0.33Ab 0.36 0.26Aa 0.16Aa 0.40ABb 0.22Bb 0.38 0.29Aa 0.17Aa 0.33Bb 0.36ABb 0.35

0.0010 0.0035 0.0011 0.0021 0.0020 0.0022 0.0079 0.0096 0.0027 0.0058 0.0015 0.0022 0.0024 0.0012 0.0018

N2O mole fraction ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.0011Aa 0.0037Aa 0.0016Aa 0.0042Aa 0.0031 0.0040Aa 0.0250Aa 0.0214Aa 0.0038Aa 0.0171 0.0021Aa 0.0023Aa 0.0028Aa 0.0010Aa 0.0022

0.0005 0.0004 0.0152 0.0015 0.0026 0.0006 0.0004 0.0255 0.0157 0.0092 0.0208 0.0005 0.0072 0.0057 0.0082

± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.0006Aa 0.0004Aa 0.0305Aa 0.0013Aa 0.0111 0.0012Aa 0.0004Aa 0.0418Aa 0.0153Aa 0.0208 0.0770Aa 0.0003Aa 0.0097Aa 0.0134Aa 0.0382

Data presented as mean ± standard error, n = 3. Different capital letters indicate significant differences among treatments for the same depth, different lower case letter indicate significant differences among depths for the same treatment. a RP is the ratio between excess N2 + N2O and NO− 3 .

466

W. Zhou et al. / Science of the Total Environment 576 (2017) 462–471

+ Fig. 2. Concentration profiles for (a) NO− 3 and (b) NH4 at different depths for the three different fertilizer treatments averaged over the rice season. The dots, black solid center line, white solid line, box extent, and error bars denote the outliers, mean, median, 25th, 75th, 10th, and 90th percentiles, respectively.

until the rice was harvested. The concentration of N2O in surface water showed a smaller variation in all treatments. A remarkably higher temporal variability of N2O was observed at depths of 0.2 and 0.8 m for the NPK and NPKS treatments, whereas in the NPKM plots there was more temporal variability in the subsoil (0.6 and 0.8 m depth). In contrast with the N2O concentration profiles, the concentration of excess N2 decreased with soil depth and the highest values appeared at 0.2 m depth for all treatments (p b 0.001; Fig. 3b). Irrespective of soil depth, the excess N2 concentration ranged from 0.08 to 2.00, 0.51 to 2.94, and 0.11 to 1.72 mg N L−1, with corresponding mean concentrations of 0.93, 1.44, and 0.85 mg N L−1 for the NPK, NPKM, and NPKS treatments, respectively. The NPKM treatment had the highest value and the difference was not significant between the NPK and NPKS treatments. Excess N2 concentrations were remarkably variable over the rice season for all treatments at various depths (Fig. 4, right-hand panel), with an obvious increase that continued until August and then decreased until the rice was harvested. The excess N2 at 0.2 m depth in the NPKM treatment was different from that in the other treatments, with a high excess N2 concentration observed at the beginning of the rice season. 3.3. NO− 3 removal ability and N2O production Soil depth significantly affected the RP (p b 0.01), with lower values for the RP in the subsoil (Table 2). The NPKM treatment resulted in a lower RP than the other treatments, but the difference was not significant. The EF5-g value for the NPKM treatment was higher than for the other treatments (Table 2) but the difference was not significantly (p N 0.05). No significant difference was observed at different depths. The N2O mole fraction was higher for the NPKM treatment, but no significant difference was observed within treatments or at different depths. A significantly lower N2O mole fraction was seen at 0.2 m depth for all treatments.

4. Discussion 4.1. Nitrate and ammonium concentrations The high NO− 3 concentrations observed at deep layers indicated the accumulation of NO− 3 in the subsoil due to the high nitrogen fertilizer application and nitrification. However, the mean NO− 3 concentrations (0.67–3.80 mg L− 1) in these plots were lower than the legal limit of 10 mg N kg− 1 for drinking water. Studies previously carried out at same experimental site observed very low NO− 3 concentrations (b 1 mg L−1) in leachates during the rice season (Tian et al., 2007; Cao et al., 2014), suggesting that paddy soils can remove NO− 3 . A long-term study carried out at Rothamsted Experimental Station in the UK over 135 years showed that the amount of NO− 3 leaching resulting from treatment with farmyard manure was five times higher than that for treatment with N fertilizers (Powlson et al., 1989). Results from other lysimeter studies have also showed the greater risk of NO− 3 leaching from the application of pig slurry compared with chemical N fertilizers (Daudén et al., 2004; Yagüe and Quílez, 2015). The higher amount of NO− 3 leaching in the NPKM plot is probably because most of the N (up to 90%) in pig manure is mineral N or readily mineralizable organic N, which is prone to leaching, especially when the nitrogen input exceeds the demands of the crop (Di and Cameron, 2002). The method of application may also contribute to the high NO− 3 leaching in NPKM plots where the pig manure was applied as a basal fertilizer, resulting 45 kg N ha− 1 more N input than in the other two treatments at the early growth stage of the rice crop and thus more NO− 3 was leached as a result of the low demand for nutrients (Tian et al., 2007). The soil in present study is a permanent charge paddy soil and has a negative charge which limits NH+ 4 leachate; therefore, it is not surprise to observe the decreased NH+ 4 concentration with increased depth. The significantly high NH+ 4 concentration in the surface water may lead to high ammonia volatilization and would cause more nitrogen loss than denitrification as reported by Li et al. (2014).

3.4. Relationship between denitrification and environmental factors 4.2. Dissolved N2O and excess N2 Table 3 gives the Spearman correlation coefficients between the denitrification end-products and the soil-related controlling factor, together with their levels of significance. There was a significant positive correlation between N2O and NO− 3 , but a significant negative correlation was observed between N2O and DOC. The excess N2 concentration was significantly and positively correlated with DOC and temperature, and negatively correlated with NO− 3 . The DO and the Eh values had no significant effect on the concentration of excess N2. The soil depth showed a strong positive relationship with both N2O and NO− 3 and a negative relationship with the excess N2, DOC, DO, RP, and Eh values.

On average, the concentrations of N2O in this study were comparable with those previously measured during the rice season in south China (Xing et al., 2002; Xiong et al., 2006; Zhou et al., 2016b), but 10 times lower than the concentration observed in a Fluvisol paddy field in Japan (Minamikawa et al., 2010; Minamikawa et al., 2011a). Denitrification was found to be the main process responsible for the presence of N2O in paddy soils; the low DOC and the non-strict reductive conditions in the Fluvisol paddy field were not conducive to denitrification and thus resulted in a high N2O concentration (Xing et al.,

W. Zhou et al. / Science of the Total Environment 576 (2017) 462–471

467

Fig. 3. Concentration profiles for (a) N2O and (b) excess N2 at different depths for the three different fertilizer treatments averaged over the rice season. The dots, solid center line, white solid line, box extent, and error bars denote the outliers, mean, median, 25th, 75th, 10th, and 90th percentiles, respectively.

2002; Minamikawa et al., 2011b). Studies carried out in different types of soil also reported high levels of N 2O in the subsoil (Cates and Keeney, 1987; Hosen et al., 2000; Xing et al., 2002). The relatively high N2O concentration at deep layer can be explained by two factors. First, gas diffusion coefficients decrease with increasing depth and therefore it is difficult for the N2O produced at deep layer to diffuse into the atmosphere (Pingintha et al., 2010). Second, the decreasing amount of DOC with increasing soil depth impairs the denitrification ability of the soil, causing the accumulation of NO− 3

at deep layer (Kamewada, 2007; Zhou et al., 2016b). High concentrations of NO− 3 may also inhibit the reduction of N2O to N2 (Blackmer and Bremner, 1978; Cho and Mills, 1979; Dodla et al., 2008), which is demonstrated by the significant correlation between the concentrations of N 2O and NO − 3 in our study and others (Table 3; Zhu et al., 2003; Weymann et al., 2008). Lower concentrations of excess N2 with high concentrations of N2O and NO− 3 were observed at the subsoil for all treatments in present study confirmed the trend of decreased rates of denitrification with increasing soil depth.

Fig. 4. Temporal variability of N2O (left-hand panel) and excess N2 (right-hand panel) concentration profiles of for treatments with NPK, NPKM, and NPKS during the rice season. Error bars represent standard deviation. The null value was caused by midseason aeration or preparation for harvest.

468

W. Zhou et al. / Science of the Total Environment 576 (2017) 462–471

Table 3 Spearman correlation coefficients among all variables. Excess N2 Excess N2 N2 O NO− 3 NH+ 4 DOC DO EF5-g RP N2O mole fraction Temperature Eh pH Depth

1 –0.168⁎⁎ –0.387⁎⁎ 0.325⁎⁎ 0.644⁎⁎ ns

–0.071 0.279⁎⁎ 0.536⁎⁎ –0.420⁎⁎ 0.474⁎⁎ –0.150⁎⁎ –0.269⁎⁎ –0.569⁎⁎

N2 O 1 0.573⁎⁎ –0.108⁎ –0.303⁎⁎ –0.008⁎ 0.387⁎⁎ –0.440⁎⁎ 0.874⁎⁎ 0.053ns 0.034ns 0.149⁎⁎ 0.346⁎⁎

NO− 3

NH+ 4

1 –0.131⁎⁎ –0.452⁎⁎ 0.180⁎⁎ –0.473⁎⁎ –0.845⁎⁎ 0.508⁎⁎ –0.115⁎⁎ 0.323⁎⁎ 0.389⁎⁎ 0.331⁎⁎

1 0.484⁎⁎ 0.151⁎⁎ 0.050ns 0.234⁎⁎ –0.226⁎⁎ 0.262⁎⁎ 0.201⁎⁎ 0.372⁎⁎ –0.448⁎⁎

DOC

DO

1 –0.020 0.231⁎⁎ 0.400⁎⁎ –0.429⁎⁎ 0.481⁎⁎ –0.191⁎⁎ ns

0.020 –0.682⁎⁎

1 –0.260⁎⁎ –0.076ns –0.091⁎⁎ 0.027ns 0.470⁎⁎ 0.220⁎⁎ –0.339⁎⁎

EF5-g

1 0.302⁎⁎ 0.424⁎⁎ 0.219⁎⁎ –0.306⁎⁎ –0.284⁎⁎ –0.004

ns

RP

N2O mole fraction

Temperature

Eh

pH

Depth

1 –0.625⁎⁎ 0.242⁎⁎ –0.219⁎⁎ –0.099⁎ –0.271⁎⁎

1 –0.157⁎⁎ 0.010ns –0.013ns 0.349⁎⁎

1 –0.068ns –0.042ns –0.205⁎⁎

1 0.366⁎⁎ 0.205⁎⁎

1 0.047ns

1

ns

Not significant. ⁎ Significant at the 0.05 level. ⁎⁎ Significant at the 0.01 level.

The concentrations of excess N2 were comparable with those measured in agricultural rivers and groundwaters with concentrations of NO− 3 equal to those in this study (Jahangir et al., 2013; Chen et al., 2014). Significantly higher amounts of excess N2 were observed in the NPKM treatment, whereas no difference was observed between the NPK and NPK plots. As mentioned in the introduction, although excess N2 may origin from variety processes, denitrification was recognized as the main process for N2 production in paddy soil. Therefore, the higher excess N2 concentration indicated higher denitrification rate in the NPKM treatment. Numerous studies have confirmed that the addition of organic matter influences the rate of denitrification and the extent depends on the quantity and quality of the carbon source (Parkin and Meisinger, 1989; Ambus and Jensen, 1997; Dendooven et al., 1996; Dambreville et al., 2006). However, most of the rates in these previously reported experiments were determined by adding organic matter to collected soil samples and evaluating the immediate effects (Cai et al., 2001; Hashimoto and Niimi, 2001; Khalil and Richards, 2011). Our results showed that the long-term application of pig manure enhances denitrification in the paddy soil profile, whereas the addition of straw manure has no effect on denitrification. Pig manure was probably more effective in enhancing denitrification as a result of the higher amounts of readily decomposable carbon and nitrogen in the pig manure than in the straw manure (Kamewada, 2007). The long-term application of pig manure significantly enhanced soil total organic carbon, pH (Table 1) and surface water DOC (Fig. 5) which may due to the addition of more organic carbon and rapid microbial degradation of the manure (Schjønning et al., 1994) and these may also have contributed to the high rate of denitrification in the NPKM plots. With respect to depth, the highest amount of excess N2 was observed at 0.2 m depth in all treatments, indicating that the surface layer of the paddy soil was a hotspot for denitrification. This is consistent with the results of Elmi et al. (2003), who found maximum denitrification at a depth of 0.15–0.3 m under a conventional tillage system. Khalil and Richards (2011) measured N2O and N2 simultaneously in an He/O2 environment and found the highest rate of denitrification in the surface soil (0.15–0.25 m depth). Similarly, Well et al. (2003) observed high rates of denitrification in a number of water-saturated surface soils (0.15–0.35 m) using the 15NO− 3 tracer and/or acetyleneinhibition methods under field and laboratory conditions. These consistent results obtained by different methods highlight the feasibility of identifying specific areas with high or low denitrification potential through measuring the amount of excess N2. The underlying reasons for the high rate of denitrification in the surface soil may be the presence of sources with a higher total organic carbon and higher abundance of denitrifiers compared with the subsoil (Frey et al., 1999; Fierer et al., 2003). Moreover, fertilization will increase the concentration of NO− 3 in the surface soil, and thus increased denitrification.

Although the surface soil was identified to be the hotspot for denitrification, denitrification in the lower layers cannot be ignored. In this study, the concentrations of excess N2 at 0.6 and 0.8 m were 24–52% of the concentration at 0.2 m depth; the proportion was highest in the NPKM plot and lowest in the NKP plot. Similar results have also been reported by other researchers for different soil types, cropping systems, and different management regimes (McCarty and Bremner, 1992; Paul and Zebarth, 1997; Elmi et al., 2005). Based on the results from the present study, the loss of N from paddy soils through denitrification may have previously been underestimated because measurements were made using soil cores taken at 0.1 or 0.2 m depth (Paul and Zebarth, 1997; Arth et al., 1998; Li et al., 2014; Lan et al., 2015) and the extent to which denitrification may be underestimated can be affected by the application of organic carbon.

4.3. Indirect N2O emission factor and NO− 3 removal ability The N2O mole fractions at 0.2 m depth (0.0003–0.0005) were 10 times lower than in the subsoil (0.0015–0.0481) for all treatments, which was unsurprising given the high excess N2 in this layer (Fig. 3b). A low N2O mole fraction indicates that N2O had been further reduced to N2. Many previous studies have reported a low N2O mole

Fig. 5. Concentration profiles for DOC at different depths for the three different fertilizer treatments averaged over the rice season. The dots, black solid center line, white solid line, box extent, and error bars denote the outliers, mean, median, 25th, 75th, 10th, and 90th percentiles, respectively.

W. Zhou et al. / Science of the Total Environment 576 (2017) 462–471

fraction in soils where complete denitrification occurred (Weymann et al., 2008; Khalil and Richards, 2011; Jahangir et al., 2013). The mean values of EF5-g (0.001–0.006) were higher than the values reported by Xiong et al. (2006) (0.000006–0.0009), but much smaller than the results reported by Minamikawa et al. (2010) (0.0237–0.0257). Nevertheless, our results were comparable with the default value of 0.0025 reported by the IPCC (2007). As the supplies of O2 become more limited with increasing depths in paddy soils, N2O is more likely to be reduced to N2 under anaerobic conditions (Blicher-Mathiesen and Hoffmann, 1999; Xing et al., 2002). This might explain why the EF5-g value is low in rice-dominated agriculture. Other N2O production processes like nitrification and DNRA may also affect EF5-g and N2O mole fraction, however, it is not possible to separate the contributions from different processes in this study. The RP showed how much initial N was transformed to N2O and excess N2 in the soil profile. The high value for RP (0.68–0.88) in surface water and at 0.2 m depth can be considered as evidence for a hotspot for denitrification in the surface soil. The low value for RP (0.17–0.44) at deep layer (0.6–0.8 m) resulted in a higher concentration of NO− 3 . 4.4. Factors influencing denitrification of the soil profile Denitrification mainly depends on the presence of anaerobic conditions and the availability of electron donors in arable soils (Seitzinger et al., 2006). The Eh and DO values in all plots (Table 4) showed that strong reducing conditions were established in the paddy soil profile throughout the whole rice season. Therefore organic carbon was probably the major factor controlling denitrification in this paddy soil, as demonstrated by the strong positive relationship between the concentration of excess N2 and DOC. Carbon acts as an energy source as well as decreasing the levels of oxygen, eventually creating an anaerobic environment in the soil suitable for denitrification. A lack of organic carbon as an energy source for heterotrophic micro-organisms has been identified as the major factor limiting the rates of denitrification in aquifers (Devito et al., 2000; Pabich et al., 2001). In this study, the application of pig manure significantly enhanced the DOC (p b 0.01) in the surface soil, where high concentrations of excess N2 and low concentrations of dissolved N2O and NO− 3 were observed. By contrast, low concentrations of DOC (b 10 mg L−1) in the subsoil resulted in high concentrations of dissolved N2O and NO− 3 with a low concentration of excess N2 (Fig. 5). The DOC concentrations significantly decreased with depth for all treatments, suggesting that only small amounts of DOC moved down to the subsoil. This is in line with the results of McCarty and Bremner (1992), who reported that no DOC was leached from the surface soil into the subsoil during the decomposition of freshly added plant tissue. Kamewada (2007) reported that the long-term (20-year) application of organic material significantly increased the organic carbon content and denitrification enzyme activity in the plow layer, but no enhancement was detected in the subsoil. Subsoil denitrification was mainly driven by the leaching of organic C from the topsoil and may be strongly

469

limited by insufficient amounts of DOC (Weier et al., 1993; Brye et al., 2001). The situation is different in the surface soil because organic carbon is supplemented by crop residues or the application of organic materials (Hill and Cardaci, 2004; Kamewada, 2007; Yan et al., 2011). Furthermore, plant roots can exude small organic molecules, including sugars, amino acids, organic acids, and amides (Neff and Asner, 2001; Lu et al., 2002). These molecules influence the availability of soil nutrients both directly and indirectly by stimulating the activities of the microbial and fungal components of the soil biota (Rivett et al., 2008; Siemens et al., 2003). Thus maximum denitrification has been observed in the surface layer in the present study and by many others (Puckett and Hughes, 2005; Mayer et al., 2005; Jahangir et al., 2014). Although subsoil denitrification was limited in the paddy soil, in a similar manner to upland and wetland soils, low N2O/(N2O + N2) ratios and low NO− 3 concentrations were observed at subsoil, suggesting that the paddy soil profile created favorable environments for denitrification. 5. Conclusions The identification of areas of low or high denitrification in paddy soils is crucial for accurately quantifying nitrogen removal through denitrification and indirect N2O emissions to the atmosphere. Our results suggest that there is an obvious variability in denitrification and N2O production within the paddy soil that is significantly affected by the application of organic matter and the soil depth. The highest denitrification was found in the surface soil for all treatments and DOC was the major limiting factor for denitrification in the subsoil. The application of pig manure markedly increased the DOC in the surface soil, resulting in high rates of denitrification, whereas straw manure had no effect on denitrification. We observed high levels of NO− 3 reduction (48–54%) in the paddy soil profile, especially in the surface soil (68–88%) where denitrification was higher. Our results highlight the feasibility of identifying areas with high or low denitrification potential in paddy soil profile by measuring the concentration of excess N2. The emission factors for N2O (EF5-g) (0.001–0.006) were small and comparable with the default value (0.0025) of the IPCC (2007). Low N2O/(N2O + N2) ratios were observed across the paddy soil profile, indicating the complete reduction of N2O to N2 as a result of the anaerobic environment. Although the surface soil was identified as a hotspot for denitrification, considerable concentrations of excess N2 were also observed in the subsoil. We therefore conclude that the loss of N through denitrification can be underestimated if research is focused only on the surface soil and the extent to which denitrification is underestimated may be affected by the application of organic carbon. Ethical statement I certify that this manuscript is original and has not been published and will not be submitted elsewhere for publication while being considered by plant and soil. And the study is not split up into several parts to increase the quantity of submissions and submitted to various journals

Table 4 Mean values of temperature, DO, pH and Eh measured in surface water and leachate over the whole rice season. Treatment

Depth (m)

Temperature (°C)

DO (mg L−1)

pH

NPK

Surface water 0.2 0.6 0.8 Surface water 0.2 0.6 0.8 Surface water 0.2 0.6 0.8

24.8 24.9 23.8 23.6 25.2 24.7 24.1 23.7 25.0 24.9 24.1 24.0

5.5 0.6 0.9 0.7 3.0 0.5 0.7 1.2 5.0 0.7 0.7 0.8

6.8 6.4 6.6 6.5 6.7 6.4 6.5 6.6 6.7 6.3 6.5 6.5

NPKM

NPKS

Data presented as mean ± standard error, n = 3.

± ± ± ± ± ± ± ± ± ± ± ±

0.5 0.3 0.2 0.2 0.4 0.4 0.2 0.1 0.2 0.2 0.2 0.2

± 1.5 ± 0.1 ± 0.5 ± 0.4 ± 1.5 ± 0.1 ± 0.2 ± 1.0 ± 1.1 ± 0.4 ± 0.3 ± 0.5

Eh (mV) ± ± ± ± ± ± ± ± ± ± ± ±

0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.0 0.1 0.1 0.1

–22.6 –70.0 –56.9 –52.0 –49.8 –79.7 –65.9 –64.4 –25.3 –72.1 –60.8 –73.2

± 14.8 ± 13.2 ± 14.9 ± 16.3 ± 27.9 ± 25.8 ± 13.5 ± 14.2 ± 7.5 ± 16.8 ± 18.6 ± 12.8

470

W. Zhou et al. / Science of the Total Environment 576 (2017) 462–471

or to one journal over time. No data have been fabricated or manipulated (including images) to support my conclusions. No data, text, or theories by others are presented as if they were our own. The submission has been received explicitly from all co-authors. And authors whose names appear on the submission have contributed sufficiently to the scientific work and therefore share collective responsibility and accountability for the results. Acknowledgments This study was financially supported by the National Natural Science Foundation of China (No. 41425005, 41571130062). References Ambus, P., Jensen, E.S., 1997. Nitrogen mineralization and denitrification as influenced by crop residue particle size. Plant Soil 197, 261–270. Arth, I., Frenzel, P., Conrad, R., 1998. Denitrification coupled to nitrification in the rhizosphere of rice. Soil Biol. Biochem. 30, 509–515. Bernard, R.J., Mortazavi, B., Kleinhuizen, A.A., 2015. Dissimilatory nitrate reduction to ammonium (DNRA) seasonally dominates NO− 3 reduction pathways in an anthropogenically impacted sub-tropical coastal lagoon. Biogeochemistry 125, 47–64. Bernard-Jannin, L., Sun, X., Teissier, S., Sauvage, S., Sánchez-Pérez, J.M., 2016. Spatio-temporal analysis of factors controlling nitrate dynamics and potential denitrification hot spots and hot moments in groundwater of an alluvial floodplain. Ecol. Eng. http://dx. doi.org/10.1016/j.ecoleng.2015.12.031. Blackmer, A.M., Bremner, J.M., 1978. Inhibitory effect of nitrate on reduction of N2O to N2 by soil microorganisms. Soil Biol. Biochem. 10, 187–191. Blicher-Mathiesen, G., Hoffmann, C.C., 1999. Denitrification as a sink for dissolved nitrous oxide in a freshwater riparian fen. J. Environ. Qual. 28, 257–262. Böhlke, J.K., 2002. Groundwater recharge and agricultural contamination. Hydrogeol. J. 10, 153–179. Brye, K.R., Norman, J.M., Bundy, L.G., Gower, S.T., 2001. Nitrogen and carbon leaching in agroecosystems and their role in denitrification potential. J. Environ. Qual. 30, 58–70. Cai, Z.C., Laughlin, R.J., Stevens, R.J., 2001. Nitrous oxide and dinitrogen emissions from soil under different water regimes and straw amendment. Chemosphere 42, 113–121. Cao, Y.S., Tian, Y.H., Yin, B., Zhu, Z.L., 2014. Improving agronomic practices to reduce nitrate leaching from the rice–wheat rotation system. Agr. Ecosyst. Eviron. 195, 61–67. Cates, R.L., Keeney, D.R., 1987. Nitrous oxide production throughout the year from fertilized and manured maize fields. J. Environ. Qual. 16, 443–447. Chen, N.W., Wu, J.Z., Chen, Z.H., Lu, T., Wang, L.J., 2014. Spatial-temporal variation of dissolved N2 and denitrification in an agricultural river network, Southeast China. Agr. Ecosyst. Eviron. 189, 1–10. Cho, C.M., Mills, J.G., 1979. Kinetic formulation of the denitrification process in soil. Can. J. Soil Sci. 59, 249–257. Choi, B.Y., Yun, S.T., Kim, K., Kim, K.H., Lee, J.H., Han, J.S., 2013. A mesocosm study on biogeochemical role of rice paddy soils in controlling water chemistry and nitrate attenuation during infiltration. Ecol. Eng. 53, 89–99. Clément, J.C., Pinay, G., Marmonier, P., 2002. Seasonal dynamics of denitrification along topohydrosequences in three different riparian wetlands. J. Environ. Qual. 31, 1025–1037. Dambreville, C., Hénault, C., Bizouard, F., Morvan, T., Chaussod, R., Germon, J.C., 2006. Compared effects of long-term pig slurry applications and mineral fertilization on soil denitrification and its end products (N2O, N2). Biol. Fertil. Soils 42, 490–500. Daudén, A., Quílez, D., Vera, M.V., 2004. Pig slurry application and irrigation effects on nitrate leaching in Mediterranean soil lysimeters. J. Environ. Qual. 33, 2290–2295. Davidson, E.A., Seitzinger, S., 2006. The enigma of progress in denitrification research. Ecol. Appl. 16, 2057–2063. Dendooven, L., Splatt, P., Anderson, J.M., 1996. Denitrification in permanent pasture soil as affected by different forms of C substrate. Soil Biol. Biochem. 28, 141–149. Devito, K.J., Fitzgerald, D., Hill, A.R., Aravena, R., 2000. Nitrate dynamics in relation to lithology and hydrologic flow path in a river riparian zone. J. Environ. Qual. 29, 1075–1084. Di, H.J., Cameron, K.C., 2002. Nitrate leaching in temperate agroecosystems: sources, factors and mitigating strategies. Nuir. Cycl. Agroecosys. 64, 237–256. Dixon, E.R., Laughlin, R.J., Watson, C.J., Hatch, D.J., 2010. Evidence for the production of NO and N2O in two contrasting subsoils following the addition of synthetic cattle urine. Rapid Commun. Mass Spectrom. 24, 519–528. Dodla, S.K., Wang, J.J., DeLaune, R.D., Cook, R.L., 2008. Denitrification potential and its relation to organic carbon quality in three coastal wetland soils. Sci. Total Environ. 407, 471–480. Elmi, A.A., Madramootoo, C., Hamel, C., Liu, A., 2003. Denitrification and nitrous oxide to nitrous oxide plus dinitrogen ratios in the soil profile under three tillage systems. Biol. Fertil. Soils 38, 340–348. Elmi, A.A., Astatkie, T., Madramootoo, C., Gordon, R., Burton, D., 2005. Assessment of denitrification gaseous end-products in the soil profile under two water table management practices using repeated measures analysis. J. Environ. Qual. 34, 446–454. Fenton, O., Richards, K.G., Kirwan, L., Khalil, M.I., Healy, M.G., 2009. Factors affecting nitrate distribution in shallow groundwater under a beef farm in South Eastern Ireland. J. Environ. Manag. 90, 3135–3146.

Fierer, N., Schimel, J.P., Holden, P.A., 2003. Variations in microbial community composition through two soil depth profiles. Soil Biol. Biochem. 35, 167–176. Foster, S.S.D., Young, C.P., 1980. Groundwater contamination due to agricultural land-use practices in the United Kingdom. UNESCO. S R. Hydrol. 30, 268–282. Frey, S.D., Elliott, E.T., Paustian, K., 1999. Bacterial and fungal abundance and biomass in conventional and no-tillage agroecosystems along two climatic gradients. Soil Biol. Biochem. 31, 573–585. Friedl, J., Scheer, C., Rowlings, D.W., McIntosh, H.V., Strazzabosco, A., Warner, D.I., Grace, P.R., 2016. Denitrification losses from an intensively managed sub-tropical pasture– impact of soil moisture on the partitioning of N2 and N2O emissions. Soil Biol. Biochem. 92, 58–66. Galloway, J.N., Townsend, A.R., Erisman, J.W., Bekunda, M., Cai, Z.C., Freney, J.R., Martinelli, L.A., Seitzinger, S.P., Sutton, M.A., 2008. Transformation of the N: recent trends, questions and potential solutions. Science 320, 889–892. Groffman, P.M., Altabet, M.A., Böhlke, J.K., Butterbach-Bahl, K., David, M.B., Firestone, M.K., Voytek, M.A., 2006. Methods for measuring denitrification: diverse approaches to a difficult problem. Ecol. Appl. 16, 2091–2122. Hashimoto, T., Niimi, H., 2001. Seasonal and vertical changes in denitrification activity and denitrifying bacterial populations in surface and subsurface upland soils with slurry application. Soil Sci. Plant Nutr. 47, 503–510. Hill, A.R., Cardaci, M., 2004. Denitrification and organic carbon availability in riparian wetland soils and subsurface sediments. Soil Sci. Soc. Am. J. 68, 320–325. Hosen, Y., Tsuruta, H., Minami, K., 2000. Effects of the depth of NO and N2O productions in soil on their emission rates to the atmosphere: analysis by a simulation model. Nutr. Cycl. Agroecosyst. 57, 83–98. Hou, L.J., Zheng, Y.L., Liu, M., Li, X.F., Lin, X.B., Yin, G.Y., Jiang, X.F., 2015. Anaerobic ammonium oxidation and its contribution to nitrogen removal in China's coastal wetlands. Sci. Report. 5:15621. http://dx.doi.org/10.1038/srep15621. Intergovernmental Panel on Climate Change, 2006. IPCC Guidelines for National Greenhouse Gas Inventories. IPCC/IGES, Kanagawa, Japan. Jahangir, M.M.R., Johnston, P., Barrett, M., Khalil, M.I., Groffman, P.M., Boeckx, P., Richards, K.G., 2013. Denitrification and indirect N2O emissions in groundwater: hydrologic and biogeochemical influences. J. Contam. Hydrol. 152, 70–81. Jahangir, M.M.R., Minet, E.P., Johnston, P., Premrov, A., Coxon, C.E., Hackett, R., Richards, K.G., 2014. Mustard catch crop enhances denitrification in shallow groundwater beneath a spring barley field. Chemosphere 103, 234–239. Jarvis, S.C., 2000. Progress in studies of nitrate leaching from grassland soils. Soil Use Manag. 16, 152–156. Ju, X.T., Xing, G.X., Chen, X.P., Zhang, S.L., Zhang, L.J., Liu, X.J., Zhang, F.S., 2009. Reducing environmental risk by improving N management in intensive Chinese agricultural systems. Proc. Natl. Acad. Sci. U. S. A. 106, 3041–3046. Kamewada, K., 2007. Vertical distribution of denitrification activity in an andisol upland field and its relationship with dissolved organic carbon: effect of long-term organic matter application. Soil Sci. Plant Nutr. 53, 401–412. Kana, T.M., Darkangelo, C., Hunt, M.D., Oldham, J.B., Bennett, G.E., Cornwell, J.C., 1994. Membrane inlet mass spectrometer for rapid high-precision determination of N2, O2, and Ar in environmental water samples. Anal. Chem. 66, 4166–4170. Khalil, M.I., Richards, K.G., 2011. Denitrification enzyme activity and potential of subsoils under grazed grasslands assayed by membrane inlet mass spectrometer. Soil Biol. Biochem. 43, 1787–1797. Küstermann, B., Christen, O., Hülsbergen, K.J., 2010. Modelling nitrogen cycles of farming systems as basis of site-and farm-specific nitrogen management. Agr. Ecosyst. Eviron. 135, 70–80. Lan, T., Han, Y., Cai, Z.Q., 2015. Denitrification and its product composition in typical Chinese paddy soils. Biol. Fertil. Soils 51, 89–98. Li, X.B., Xia, L.L., Yan, X.Y., 2014. Application of membrane inlet mass spectrometry to directly quantify denitrification in flooded rice paddy soil. Biol. Fertil. Soils 50, 891–900. Lu, Y.H., Watanabe, A., Kimura, M., 2002. Contribution of plant-derived carbon to soil microbial biomass dynamics in a paddy rice microcosm. Biol. Fertil. Soils 36, 136–142. Majumdar, D., 2013. Biogeochemistry of N2O uptake and consumption in submerged soils and rice fields and implications in climate change. Crit. Rev. Environ. Sci. Technol. 43, 2653–2684. Mathieu, O., Lévêque, J., Hénault, C., Milloux, M.J., Bizouard, F., Andreux, F., 2006. Emissions and spatial variability of N2O, N2 and nitrous oxide mole fraction at the field scale, revealed with 15N isotopic techniques. Soil Biol. Biochem. 38, 941–951. Mayer, P.M., Reynolds, S.K., McCutchen, M.D., Canfield, T.J., 2005. Riparian Buffer Width, Vegetative Cover, and Nitrogen Removal Effectiveness: A Review of Current Science and Regulations. IOP Publising Agrisfaoweb. http://agris.fao.org/agris-search/search. do?recordID=US201300114453 (Accessed October 2005). McCarty, G.W., Bremner, J.M., 1992. Availability of organic carbon for denitrification of nitrate in subsoils. Biol. Fertil. Soils 14, 219–222. Minamikawa, K., Nishimura, S., Sawamoto, T., Nakajima, Y., Yagi, K., 2010. Annual emissions of dissolved CO2, CH4, and N2O in the subsurface drainage from three cropping systems. Glob. Chang. Biol. 16, 796–809. Minamikawa, K., Hayakawa, A., Nishimura, S., Akiyama, H., Yagi, K., 2011a. Comparison of indirect nitrous oxide emission through lysimeter drainage between an andosol upland field and a fluvisol paddy field. Soil Sci. Plant Nutr. 57, 843–854. Minamikawa, K., Nishimura, S., Nakajima, Y., Osaka, K.I., Sawamoto, T., Yagi, K., 2011b. Upward diffusion of nitrous oxide produced by denitrification near shallow groundwater table in the summer: a lysimeter experiment. Soil Sci. Plant Nutr. 57, 719–732. Neff, J.C., Asner, G.P., 2001. Dissolved organic carbon in terrestrial ecosystems: synthesis and a model. Ecosystems 4, 29–48. Pabich, W.J., Valiela, I., Hemond, H.F., 2001. Relationship between DOC concentration and vadose zone thickness and depth below water table in groundwater of Cape Cod, USA. Biogeochemistry 55, 247–268.

W. Zhou et al. / Science of the Total Environment 576 (2017) 462–471 Parkin, T.B., Meisinger, J.J., 1989. Denitrification below the crop rooting zone as influenced by surface tillage. J. Environ. Qual. 18, 12–16. Paul, J.W., Zebarth, B.J., 1997. Denitrification during the growing season following dairy cattle slurry and fertilizer application for silage corn. Can. J. Soil Sci. 77, 241–248. Pingintha, N., Leclerc, M.Y., Beasley, J.J., Zhang, G.S., Senthong, C., 2010. Assessment of the soil CO2 gradient method for soil CO2 efflux measurements: comparison of six models in the calculation of the relative gas diffusion coefficient. Tellus Ser. B Chem. Phys. Meteorol. 62, 47–58. Powlson, D.S., Poulton, P.R., Addiscott, T.M., McCann, D.S., 1989. Leaching of nitrate from soils receiving organic or inorganic fertilizers continuously for 135 years. In: Hansen, J.A.A., Henriksen, K. (Eds.), Nitrogen in Organic Wastes Applied to Soils. Academic Press, London, pp. 334–345. Puckett, L.J., Hughes, W.B., 2005. Transport and fate of nitrate and pesticides. J. Environ. Qual. 34, 2278–2292. Rivett, M.O., Buss, S.R., Morgan, P., Smith, J.W., Bemment, C.D., 2008. Nitrate attenuation in groundwater: a review of biogeochemical controlling processes. Water Res. 42, 4215–4232. Schjønning, P., Christensen, B.T., Carstensen, B., 1994. Physical and chemical properties of a sandy loam receiving animal manure, mineral fertilizer or no fertilizer for 90 years. Eur. J. Soil Sci. 45, 257–268. Seitzinger, S., Harrison, J.A., Böhlke, J.K., Bouwman, A.F., Lowrance, R., Peterson, B., Drecht, G.V., 2006. Denitrification across landscapes and waterscapes: a synthesis. Ecol. Appl. 16, 2064–2090. Shan, J., Zhao, X., Sheng, R., Xia, Y., Ti, C.P., Quan, X.F., Yan, X.Y., 2016. Dissimilatory nitrate reduction processes in typical Chinese paddy soils: rates, relative contributions, and influencing factors. Environ. Sci. Technol. 50, 9972–9980. Siemens, J., Haas, M., Kaupenjohann, M., 2003. Dissolved organic matter induced denitrification in subsoils and aquifers? Geoderma 113, 253–271. Sotomayor, D., Rice, C.W., 1996. Denitrification in soil profiles beneath grassland and cultivated soils. Soil Sci. Soc. Am. J. 60, 1822–1828. Stark, C.H., Richards, K.G., 2008. The Continuing Challenge of Agricultural Nitrogen Loss to the Environment in the Context of Global Change and Advancing Research. IOP Publishing Researchgateweb. https://www.researchgate.net/publication/263735523 (Accessed 30 March 2015). Tian, Y.H., Yin, B., Yang, L.Z., Yin, S.X., Zhu, Z.L., 2007. Nitrogen runoff and leaching losses during rice-wheat rotations in Taihu Lake region, China. Pedosphere 17, 445–456. Townsend, A.R., Howarth, R.W., Bazzaz, F.A., Booth, M.S., Cleveland, C.C., Collinge, S.K., Mallin, M.A., 2003. Human health effects of a changing global nitrogen cycle. Front. Ecol. Environ. 1, 240–246.

471

Weier, K.L., Doran, J.W., Power, J.F., Walters, D.T., 1993. Denitrification and the dinitrogen/ nitrous oxide ratio as affected by soil water, available carbon, and nitrate. Soil Sci. Soc. Am. J. 57, 66–72. Weiss, R.F., Price, B.A., 1980. Nitrous oxide solubility in water and seawater. Mar. Chem. 8, 347–359. Well, R., Augustin, J., Meyer, K., Myrold, D.D., 2003. Comparison of field and laboratory measurement of denitrification and N2O production in the saturated zone of hydromorphic soils. Soil Biol. Biochem. 35, 783–799. Weymann, D., Well, R., Flessa, H., Heide, C., Deurer, M., Meyer, K., Walther, W., 2008. Groundwater N2O emission factors of nitrate-contaminated aquifers as derived from denitrification progress and N2O accumulation. Biogeosciences 5, 1215–1226. Xia, Y.Q., Yan, X.Y., 2012. Ecologically optimal nitrogen application rates for rice cropping in the Taihu Lake region of China. Sustain. Sci. 7, 33–44. Xing, G.X., Cao, Y.C., Shi, S.L., Sun, G.Q., Du, L.J., Zhu, J.G., 2002. Denitrification in underground saturated soil in a rice paddy region. Soil Biol. Biochem. 34, 1593–1598. Xiong, Z.Q., Xing, G.X., Zhu, Z.L., 2006. Water dissolved nitrous oxide from paddy agroecosystem in China. Geoderma 136, 524–532. Yagüe, M.R., Quílez, D., 2015. Pig slurry residual effects on maize yields and nitrate leaching: a study in lysimeters. Agron. J. 107, 278–286. Yan, X.Y., Cai, Z.C., Wang, S.W., Smith, P., 2011. Direct measurement of soil organic carbon content change in the croplands of China. Glob. Chang. Biol. 17, 1487–1496. Zhao, Y.Q., Xia, Y.Q., Ti, C.P., Shan, J., Li, B.L., Xia, L.L., Yan, X.Y., 2015. Nitrogen removal capacity of the river network in a high nitrogen loading region. Environ. Sci. Technol. 49, 1427–1435. Zhou, J.Y., Gu, B.J., Schlesinger, W.H., Ju, X.T., 2016a. Significant accumulation of nitrate in Chinese semi-humid croplands. Sci. Report. 6:25088. http://dx.doi.org/10.1038/ srep25088. Zhou, Z.Q., Xu, X., Bi, Z.C., Li, L., Li, B., Xiong, Z.Q., 2016b. Soil concentration profiles and diffusion and emission of nitrous oxide influenced by the application of biochar in a rice-wheat annual rotation system. Environ. Sci. Pollut. Res. Int. 23, 7949–7961. Zhu, Z.L., Chen, D.L., 2002. Nitrogen fertilizer use in China–contributions to food production, impacts on the environment and best management strategies. Nutr. Cycl. Agroecosyst. 63, 117–127. Zhu, Z.L., Wen, Q.X., 1992. Nitrogen in Soils of China. JSST Press, Nanjing. Zhu, J.G., Liu, G., Han, Y., Zhang, Y.L., Xing, G.X., 2003. Nitrate distribution and denitrification in the saturated zone of paddy field under rice/wheat rotation. Chemosphere 50, 725–732. Zhu, G.B., Wang, S.Y., Wang, Y., Wang, C.X., Risgaard-Petersen, N., Jetten, M.S., Yin, C.Y., 2011. Anaerobic ammonia oxidation in a fertilized paddy soil. ISME J. 5, 1905–1912.