Annual N2O emissions from conventionally grazed ...

Science of the Total Environment 625 (2018) 885–899

Contents lists available at ScienceDirect

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

Annual N2O emissions from conventionally grazed typical alpine grass meadows in the eastern Qinghai–Tibetan Plateau Han Zhang a,b, Zhisheng Yao b, Kai Wang b, Xunhua Zheng b,c,⁎, Lei Ma b,c, Rui Wang b, Chunyan Liu b, Wei Zhang b, Bo Zhu d, Xiangyu Tang d, Zhenghua Hu a, Shenghui Han b a Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, College of Applied Meteorology, Nanjing University of Information Science and Technology, Nanjing 210044, PR China b State Key Laboratory of Atmospheric Boundary Layer Physics and Atmospheric Chemistry, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing 100029, PR China c College of Earth Science, University of Chinese Academy of Sciences, Beijing 100049, PR China d Key Laboratory of Mountain Surface Processes and Ecological Regulation, Institute of Mountain Hazards and Environment, Chinese Academy of Sciences, Chengdu 610041, PR 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

• Characteristics of annual N2O fluxes were clarified across three alpine meadows. • Air pollutant NO, which is rarely reported in alpine meadows, was observed yearly. • Up to 50% of annual N2O emissions were released in the non-growing season. • A linear dependence of the annual N2O emissions on the ANPP occurs significantly.

a r t i c l e

i n f o

Article history: Received 7 September 2017 Received in revised form 19 December 2017 Accepted 19 December 2017 Available online xxxx Editor: Elena Paoletti Keywords: Nitrous oxide emission Nitric oxide emission Alpine grass meadow Non-growing season Spring thaw Above ground net primary productivity

a b s t r a c t Annual nitrous oxide (N2O) emissions from high-altitude alpine meadow grasslands have not been effectively characterized because of the scarcity of whole-year measurements. The authors performed a year-round measurement of N2O fluxes from three conventionally grazed alpine meadows that represent the typical meadow landscape in the eastern Qinghai–Tibetan Plateau (QTP). The results showed that annual N2O emissions averaged 0.123 ± 0.053 (2SD, i.e., the double standard deviation indicating the 95% confidence interval) kg N ha−1 yr−1 across the three meadow sites. N2O flux pulses during the spring freezing-thawing period (FTP) were observed at only one site, indicating a large spatial variability in association with soil moisture differences. Approximately 34–57% (mean: 46%) of the annual N2O emissions occurred in the non-growing season, highlighting the substantial importance of accurate flux observations during this period. The simultaneous observations showed conservative, marginal nitric oxide (NO) fluxes of 0.058 ± 0.032 (2SD) kg N ha−1 yr−1. The N2O fluxes across the three field sites correlated negatively with the soil nitrate concentrations during the entire year-round period (P b 0.05). Furthermore, a significant joint regulatory effect of topsoil temperature and moisture on the N2O and NO fluxes was observed during the relatively warm periods. Based on the results of the present and previous studies, a simple extrapolation roughly estimated the annual total N2O emission from Chinese grasslands to be 73 ± 15 (2SD) Gg N yr−1 (1 Gg = 109 g). A linear dependence of the annual N2O fluxes on the aboveground net primary productivity (ANPP) was also found. This result may provide a simple approach for estimating the

⁎ Corresponding author at: State Key Laboratory of Atmospheric Boundary Layer Physics and Atmospheric Chemistry, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing 100029, PR China. E-mail address: [email protected] (X. Zheng).

https://doi.org/10.1016/j.scitotenv.2017.12.216 0048-9697/© 2017 Elsevier B.V. All rights reserved.

886

H. Zhang et al. / Science of the Total Environment 625 (2018) 885–899

N2O emission inventories of frigid alpine or temperate grasslands that are ungrazed either in the summer or year round. However, further confirmation of this relationship with a wider ANPP range is still needed in the future studies. © 2017 Elsevier B.V. All rights reserved.

1. Introduction Nitrous oxide (N2O) is a potent greenhouse gas and the single largest contributor to global stratospheric ozone depletion (Myhre et al., 2013; Ravishankara et al., 2009). Its atmospheric concentration is still increasing at a rate of approximately 0.3% per year, which is primarily as a consequence of terrestrial ecosystems (IPCC, 2013; Williams et al., 1992). Nitric oxide (NO) is a key precursor of tropospheric ozone, fine aerosol particles involved in haze pollution and a precursor of nitric acid formation in the atmosphere (e.g., Pilegaard, 2013). In fact, the exchanges of N2O and NO between terrestrial ecosystems and the atmosphere are closely interrelated, as both gases are produced in most cases by the same processes of nitrogen transformation (e.g., Firestone and Davidson, 1989). Therefore, it is necessary to investigate the emissions of both N2O and NO from terrestrial ecosystems via simultaneous field observations to improve the current understandings of nitrogen biogeochemistry. Grassland ecosystems are one of the most important terrestrial biome types, covering approximately 20% of the global land surface (Adams et al., 1990; Zhang et al., 2010). Global emissions of N2O from grasslands were estimated at 2.5 Tg N yr−1, accounting for approximately 18% of the global estimate of the total source flux (Lee et al., 1997). Generally, microbial nitrification (as an aerobic process) and denitrification (as an anaerobic process) have been identified as the major pathways for N2O production in soils (Firestone and Davidson, 1989). A multitude of interacting biotic (e.g., vegetation type, plant-microbe interaction) and abiotic (soil climate, physical and chemical properties) variables, which regulate the biological processes in soils, lead to the very variable and dynamic characteristics of soil N2O losses (Butterbach-Bahl et al., 2013). For example, Yao et al. (2010a) showed that soil moisture is a primary regulator of spatial variability in N2O emissions by measuring the N2O fluxes of soil cores from 30 representative field sites. In another instance, Abalos et al. (2014) carried out a greenhouse mesocosm experiment and found that plant species composition is a key driver of N2O emissions from grassland ecosystems. Meanwhile, numerous field measurements have shown that the N2O fluxes from grassland ecosystems are strongly influenced by vegetation species, soil physiochemical parameters, environmental factors and management regimes (e.g., Li et al., 2015; Wolf et al., 2010; Yang et al., 2015). Clearly, although the fluxes of this gas from grasslands have been investigated globally (e.g., Cowan et al., 2015; Mosier et al., 1991, 2002; Turner et al., 2008), the flux estimates are still highly uncertain due to the scarcity of data, and the limited observations do not reliably represent the known variability in soil types, soil properties and environmental conditions (Yao et al., 2010b). Therefore, further knowledge of N2O fluxes and their driving factors influencing this gas from various grassland types under different climate zones and management patterns is essential to reduce the uncertainties in nitrogen trace gas emission inventories of grassland ecosystems at the regional, national and global scales. Alpine meadows are the most productive grassland ecosystems in montane regions such as the eastern Qinghai–Tibetan Plateau (QTP). Thus, they serve to sustain the global market of grazing-oriented animal husbandry substantially. Accurate quantification and a thorough understanding of the regulatory factors of N2O and NO emissions from alpine meadow areas are essential for the development of environmental- and climate-friendly grazing-oriented animal husbandry in montane regions. The QTP, known as “the Third Pole” of the Earth, covers nearly

one-quarter of the land area of China (Chen et al., 2013a). It is one of the largest areas of alpine grasslands in the world. Alpine meadow is the dominant vegetation type and covers approximately 35% of the entire QTP land area (Cao et al., 2008). Due to vast expanse of the QTP and its high fragility to environmental disturbances, the alpine meadows in the plateau area are highly sensitive to climate change and human activity and consequently show pronounced feedbacks to both of these influences (Hu et al., 2010; Jiang et al., 2010; Li et al., 2015; Wu et al., 2010b; Zhang et al., 2014; Zheng et al., 2012). These feedbacks to changes in the climate and/or human activity may be shown, to some extent, by variations in N2O and NO emissions, as the fluxes of these gases from grasslands are mainly regulated by the environmental and management variables mentioned above. In addition to the scarcity of NO observations, previous in situ N2O measurements in alpine meadows in the QTP either focused on the growing season exclusively (Jiang et al., 2010; Zhang et al., 2014) or were conducted annually with a very coarse temporal (i.e., monthly or seasonally) resolution in the non-growing season (e.g., Li et al., 2015; Pei et al., 2003). The available studies with full-year observations performed in the grassland areas of other temperate regions indicated that emissions occurring during the nongrowing season, particularly the spring freezing–thawing period (FTP), significantly contribute to annual N2O emissions (Wagner-Riddle et al., 2007; Wolf et al., 2010; Yang et al., 2015). Based on the measured fluxes of nitrogen trace gases from seasonally snow-covered soils in a subalpine meadow, Filippa et al. (2009) indicated that microbially mediated emissions of gaseous nitrogen oxides could be a significant part of the nitrogen cycle during the winter. However, reliable full-year observations from high-altitude alpine meadows are scarce due to the harsh conditions for field operations during the cold winter season. This situation seriously hinders our understanding of annual N2O emissions and our insight into the nitrogen cycle within high-altitude alpine meadows. In the present study, the authors performed a study based on fullyear field measurements of N2O and NO fluxes from typical highaltitude meadows at three field sites in the eastern QTP. The goals of this study were to (i) characterize the full-year N2O emissions from typical alpine meadow ecosystems in the eastern QTP; (ii) evaluate the contributions from the non-growing season, particularly the spring FTP, to annual N2O emissions; and (iii) investigate the key regulatory factors on N2O and NO fluxes. These investigations were conducted to test the following hypotheses: (i) the non-growing season can contribute substantially to annual emissions of N2O from high-altitude alpine meadows; (ii) N2O fluxes are regulated by abiotic factors, such as soil temperature and other physical variables; and (iii) annual N2O emissions are related to the biotic variable aboveground net primary production. 2. Materials and methods 2.1. Descriptions of the selected field sites In this study, three adjacent field sites (see the geographic coordinates in Table 1) were selected in the northernmost terrain of Zoige County, Sichuan Province, China. The selected field sites are located in the source region of the Pai–Lung River, which is a subbranch of the upper Yangze River, and they represent typical high-altitude meadows of the eastern QTP. This region is subject to a frigid, humid monsoon climate. According to meteorological records from the Zoige

H. Zhang et al. / Science of the Total Environment 625 (2018) 885–899 Table 1 Some site features of the three investigated alpine grass meadows grazed periodically (MG). AP, annual precipitation (mm). AT, annual mean air temperature (°C). GWT, annual mean groundwater table (m). DPS, dominant plant species. ANPP, aboveground net primary productivity (g m−2 yr−1). ST, stocking rate (sheep unit ha−1 yr−1). ST, annual mean topsoil (5 cm depth) temperature (°C). Moisture, annual mean topsoil (0–6 cm depths) moisture content in water-filled pore space (%). Ammonium, nitrate and DOC, annual mean concentrations of soil (0–10 cm depths) ammonium, nitrate and dissolved organic carbon, respectively (mg N or C kg−1 dry soil). The value in a pair of parentheses indicate the standard error of four replicates. More details of the soil properties at each site are referred to the online supplementary materials (Table S1). Items

MG1

MG2

MG3

Latitude Longitude Altitude AP AT GWT DPS

34°01.1′N 102°43.9′E 3430 m ~617 3.7 b−4.5 Polygonum viviparum, Plantago asiatica, Gentiana macrophylla, Kobresia humilis

ANPP SR Texture Slope ST Moisture Ammonium Nitrate DOC

198 (7) 0.59 Silt loam ~8° 6.2 58 (11) 8.4 (1.3) 5.8 (1.0) 52.2 (5.4)

34°02.0′N 102°43.5′E 3326 m 617 2.7 −0.8 to −3.7 Deschampsia littoralis, Elymus nutans, Kobresia humilis 215 (4) 1.48 Silt loam ~11° 5.5 60 (13) 10.0 (1.4) 4.4 (0.6) 58.2 (7.1)

34°03.5′N 102°43.8′E 3231 m ~617 4.2 b−4.5 Kobresia humilis, Carex aridula 347 (26) 3.49 Silt loam b1° 6.8 40 (14) 18.4 (1.7) 1.7 (0.2) 142.9 (16.5)

Meteorological Station (33°34.8′ N, 102°58.2′ E, 3441 m a.s.l., approximately 80 km south of the selected field sites) during the period 1957–2000 (Huo et al., 2013), the annual mean temperature is 0.93 °C with the coldest weather occurring in January and the warmest occurring in July, and the annual rainfall averages at 648.4 mm, most of which falls during the growing season spanning from early to midApril through early to mid-October. The meadows grazed (MG) periodically at all of the selected field sites belong to the same Tibetan village, and they have been traditionally used as winter-spring/early summer pastures since early recorded history. The first site (hereinafter referred to as MG1) is located on the northwest-facing slope (with slope gradients ≈ 8°) of a mountain with a peak altitude of 3930 m. The vegetation of MG1 was dominated by Polygonum viviparum, Plantago asiatica, Gentiana macrophylla, and Kobresia humilis during the study period. The second site (hereinafter referred to as MG2) is located on a north-facing slope (with slope gradients ≈11°) and is approximately 1.5 km northeast of MG1. It neighbors a large piece of peat wetland at the slope base. The vegetation of MG2 was dominated by Deschampsia littoralis, Elymus nutans and Kobresia humilis. The third site (hereinafter referred to as MG3) is located on flat land (with slope gradients b1°) and is approximately 5 km to the north of MG1. The vegetation of MG3 was dominated by Kobresia humilis and Carex aridula. At each field site, the vegetation was spatially distributed homogeneously. Yak and Tibetan sheep grazed throughout all of the sites during the winter and late spring into the early summer, while grazing is forbidden during the remaining periods of the year. During our study period, grazing began on October 9 (MG3) or on November 29 (MG1 and MG2) and continued through December 20 in 2013 as well as during June 4 through July 7 and October 12 through November 11 in 2014 at all three sites. The stocking rates at MG1, MG2 and MG3 were 0.59, 1.48 and 3.49 sheep unit ha−1 yr−1, respectively. All of the aboveground biomass is usually used up during winter grazing. Thus, during the subsequent early summer grazing period, the livestock almost exclusively consume newly grown biomass. The vegetation community at all of the selected sites commonly appears in the alpine meadows of the eastern QTP, while the adopted grazing practices are

887

typical in this region. More details about the three sites are listed in Table 1. 2.2. Measurement of nitrous oxide fluxes The N2O fluxes at all sites were measured manually in situ during the full-year study period stated above using gas chromatograph-based static opaque chamber methods (Zhang et al., 2014; Zheng et al., 2008). Four spatially replicated locations (approximately 10–60 m apart from each other) were chosen randomly at each field site to account for special variations (e.g., differences in vegetation type, soil temperature and moisture). At each replicated location, one chamber base frame (length × width × height = 0.5 × 0.5 × 0.1 m3) made of stainless steel was inserted almost completely into the soil with its upper edge extending above the ground surface by ≤ 1 cm. To prevent possible influences on the N2O fluxes from soil and plant disturbances during the emplacement of the chamber bases, all of the replicated base frames were installed one week before the measurements were initiated. Each frame remained in place permanently unless it was moved to a new location nearby due to visible disturbances from vegetation and/ or the soil surface during the sampling period. Whenever there was snow cover, the snow on the base frame collar was carefully removed immediately prior to mounting a chamber onto the frame for gas sampling. The operation had to be carefully performed to avoid disturbances from the snow cover within and outside the base frame. A rubber strip was used to seal the connection joint between the base frame and the chamber to ensure tightness and avoid gas leakage. As soon as the gas from the enclosure was sampled, the chamber was removed from the base frame, and the snow cover on the base frame collar was carefully restored. The chambers were made of stainless steel sheets (with 0.9 mm thicknesses) and designed with length × width × height dimensions of 0.5 × 0.5 × 0.15 m3 for MG1 and MG2 and of 0.5 × 0.5 × 0.5 m3 for MG3 (i.e., the lower height was designed for shorter vegetation). Each chamber was wrapped with a layer of Styrofoam and aluminum foil to mitigate temperature increases inside the enclosures due to solar heating. Moreover, holes with diameters of 2 cm were placed in the top walls of the chambers that remained open while mounting the chamber onto the base frame to equilibrate the headspace air pressure. Then, a pressure balance tube (10 cm long and 1/4 in. for the inner diameter) that was designed according to Hutchinson and Mosier (1981) was fitted into this hole, which was open during the sucking of each gas sample but was closed during the sampling intervals to eliminate the effects of pressure imbalances during the sampling operations on the representativeness of the measured fluxes. The N2O fluxes at MG2 and MG3 were usually measured twice a week during the growing season and weekly during the non-growing season, and they were measured weekly year round at MG1. However, the measurement frequency was increased to once every 1 to 3 days during the spring FTP or after a rainfall event with precipitation N5 mm following a significant drought period. The gas sampling was normally performed between 9:00 and 11:00 local standard time (LST), when the gas flux can approximate the daily mean (e.g., Liu et al., 2010a). To measure the N2O flux, five air samples were taken from the headspace enclosure of a chamber at 10-min (MG1 and MG2) or 15-min (MG3) intervals using 60-mL polypropylene syringes. A longer sampling interval was adopted for the greater chambers heights at MG3 in order to maintain a similar precision for the measured fluxes among all of the sites. The N2O concentrations of the air samples were analyzed within 5 h after sampling at a temporary laboratory nearby, using a gas chromatograph (GC) instrument (HP7890A, Agilent Technologies, Santa Clara, CA, USA) equipped with an electron capture detector (ECD). For this, the DN-CO2 method was adopted (i.e., introducing a mixed gas with 10% carbon dioxide in pure dinitrogen (N2) in to the ECD cell as a buffering gas at a small mass flow rate of 1–3 mL min−1, and using N2 as the

888

H. Zhang et al. / Science of the Total Environment 625 (2018) 885–899

carrier gas to analyze N2O) as described in detail by Wang et al. (2010) and Zheng et al. (2008). The N2O concentrations were calibrated with a reference gas and corrected using an international standard organized by the World Meteorology Organization. The N2O fluxes were calculated as follows (Wang et al., 2013): (i) when there were less than five valid observations of N2O concentrations within a chamber enclosure or when there were five valid N2O concentrations but no detection of significant non-linearity, the linear model (Ct = a0 + a1t, where Ct is the measured concentration, a0 is the intercept, a1 is the slope of the fitting line, and t is the enclosure time) was applied; (ii) when a significant non-linearity with a higher determination coefficient than the linear fitting was detected for five observed N2O concentrations, the nonlinear model following Kroon et al. (2008) and Valente et al. (1995) (Ct = k1/ k2 + (C0 − k1/k2)∙exp(−k2t), where C0 is the N2O concentration at the beginning of the enclosure and k1 and k2 are the fitting parameters) was applied to obtain the initial rate of change of the gas concentrations. The adopted methods and procedures resulted in N2O flux detection limits of ±0.8 (for MG1 and MG2) and ±1.7 (for MG3) μg N m−2 h−1 on average (at the 95% confidence interval). 2.3. Auxiliary measurements To better understand the nitrogen biogeochemistry in association with the N2O emissions, the nitric oxide (NO) fluxes occurring simultaneously with N2O were also measured at sites MG2 and MG3. To measure the NO flux, two air samples (with volumes of 2.5–3.0 L each) were collected. The first sample was taken from the ambient air immediately after the chamber was mounted to the base frame, and the second sample was taken from the chamber headspace air following the last N2O sample collection (Mei et al., 2011). Each air sample was directly injected into an evacuated 5-L gas bag made of inert aluminumcoated plastic (Guangming Research & Design Institute of Chemical Industry, Dalian, China) using an air pump (flow rate: 2–3 L min−1; N86KNDC, KNF Neuberger, Inc., Freiburg, Germany) driven by 12-volt direct currency. The NO concentrations of the air samples were analyzed within 1 h after collection using a chemiluminescence-based NO–NO2– NOx analyzer (42i, Thermo Environmental Instruments Inc., USA) at the aforementioned temporary laboratory. The NO flux was derived from the concentration difference between the two samples and their sampling time interval following a simple linear change assumption (e.g., Mei et al., 2009). A dynamic dilution calibrator system (146i, Thermo Environmental Instruments Inc., USA) was used for bimonthly calibration of the analyzer. The methods and procedures for the air sampling, analysis and flux calculation followed the detailed description by Mei et al. (2009). It should be noted that the NO fluxes measured in this study represent only conservative magnitudes for the investigated alpine meadows. This is because the applied method with a linear change assumption might have significantly underestimated the NO fluxes by, e.g., 31% (ranging from 3% to 59% at the 95% confidence interval) compared with the non-linear method (Mei et al., 2009; Yao et al., 2015). Some auxiliary variables, which were expected to reflect the meteorological and soil conditions for the N2O production and emission, were also dynamically measured in addition to the NO fluxes. The meteorological variables include the precipitation, air temperature and air pressure, while the soil dynamical variables include the soil temperature, moisture content, and concentration of ammonium nitrate and dissolvable organic carbon (DOC). During the observational period, the groundwater table was monitored by using perforated polyvinyl chloride pipes, which were inserted into the soil (down to a 4.5 m depth) at all sites. Additionally, the aboveground net primary productivity (ANPP) was observed, and the soil properties were surveyed at each site. Hourly means of the air temperature and air pressure (at a height of 1.5 m), as well as the hourly precipitation, were recorded automatically at a meteorological station near MG2. The daily precipitation by

snowfall was measured manually with a heated rain gauge, while the snow cover depth was daily recorded. Temperature sensors (StowAway TidbiT v2, Onset Computer Co., Irvine, CA, USA) were used to record the soil temperatures continuously at a 5 cm depth at all of the sites during the entire observational period. While the gas samples were collected from chamber enclosures for the flux determinations, the soil temperature (5 cm depth) and topsoil volumetric water contents (θv, in %) (over depths of 0–6 cm) surrounding each chamber were manually measured. A digital thermometer (JM624, Tianjin JM Instrument Co. Ltd., China) and a portable frequency-domain reflector probe (MPM-160, Jiangsu RDS Technology Co. Ltd., China) were employed for the manual measurement of the soil temperature and moisture, respectively. When the soil was frozen, its water content (i.e., in the form of ice) was determined gravimetrically at depths over 0–6 cm. The gravimetric water content (θw, in %) was converted to θv following θv = BD · θw, wherein BD (in g cm−3) denotes the soil bulk density. The directly measured, or θw-originated, θv values were converted into water-filled pore space (WFPS, in %) according to the topsoil BD and a theoretical soil particle density (ρ = 2.65 g m−3) following WFPS = θv / (1 − BD / ρ). To measure the concentrations of ammonium, nitrate and DOC in the topsoil (over depths of 0–10 cm) at each field site, samples were taken weekly or biweekly (on a portion of the days when the gas fluxes were observed) with four spatial replicates (each near a replicated chamber location) using a soil auger with a 3-cm diameter. To collect samples at each replicated location, one auger of soil was collected. All of the fresh soil samples from the four spatial replicates were well mixed. Three subsamples taken from the resulting mixture were extracted with either 1 M potassium chloride solution to analyze the ammonium and nitrate or deionized water to determine the DOC. The subsample extracts were frozen in 50 mL polyethylene terephthalate bottles and stored at −18 °C for subsequent analysis with a continuous flow analyzer (San++ Continuous Flow Analyzer, Skalar Analytical B. V., the Netherlands). A more detailed description of the procedures can be found in Zhang et al. (2014). The aboveground standing biomass was measured to determine the ANPP. The measurements were conducted both early and late during the growing season and both immediately prior to and following each grazing period. To determine the gross livestock uptake and the newly grown biomass during the grazing periods at each site, sampling squares (50 × 50 cm2 each) with four replicates were erected to prevent yak and sheep from accessing to the area. The aboveground plants within each sampling square were harvested and then immediately ovendried at 105 °C for 30 min and then at 80 °C for 48 h before weighing. Soil samples from three replicates of the profiles (over depths of 0–100 cm) for each site were collected and air-dried. Then, their properties, including the soil organic carbon (SOC), total nitrogen (TN), pH (water), BD, and particle fractions of sand (0.02–2 mm), silt (0.002–0.02 mm) and clay (b 0.002 mm), were subsequently measured. The SOC and TN contents were analyzed using the potassium dichromate oxidation and Kjeldahl methods, respectively. 2.4. Statistical analysis The Excel package from Microsoft Office Standard 2010 (© 2010 Microsoft Corporation) was applied for the raw data organization and calculation processes. The SPSS Statistics Client 19.0 (SPSS Inc., Chicago, USA) and Origin 8.0 (OriginLab Ltd., Guangzhou, China) software packages were used for statistical analysis. A factor Analysis with SPSS 19.0 was applied to determine the correlation coefficient matrix for measured variables, including soil temperature; moisture; the concentrations of soil ammonium, nitrate and DOC; and the fluxes of N2O and NO. The correlation coefficients and their significance levels were determined using the standardized values (Z) of the directly measured data (X). The standardization followed the equation zik = (xik − xk)/σk, where k and i denote the ordinal numbers of the

H. Zhang et al. / Science of the Total Environment 625 (2018) 885–899

investigated variables and the measurements of the kth variable, respectively; zik and xik denote the elements of the arrays Z and X, respectively; and xk and σk are the mean and standard variance of the measurements of kth variable, respectively. Linear regression (Origin 8.0) was employed to identify the soil temperature effect on N2O and NO fluxes and the soil moisture influence on the logarithm of the molar ratio of the NO flux to N2O flux. Multiple linear regression (SPSS 19.0) was conducted to determine the combined effects of soil temperature and moisture on N2O and NO fluxes. In addition, linear regression (Origin 8.0) was applied to investigate the relationship between the annual N2O emissions and aboveground net primary productivity (ANPP). Whenever a “double standard deviation” (2SD) is specified for the error of an estimate, it indicates an uncertainty range at the 95% confidence interval. Otherwise, an error refers to a standard error of 3 or 4 replicates.

889

3. Results 3.1. Meteorological conditions, soil properties and variables, and other site features 3.1.1. Temperature definitions of the freezing-thawing period and nongrowing season According to the meteorological records near MG2 during the yearround observational period, the daily mean air temperature ranged between −13.2 and 16.4 °C with an average of 2.7 °C (Fig. 1a), which was approximately 2 °C warmer compared to the annual mean temperature (0.93 °C) of the multiple-year (1957–2000) observations at the meteorological station (Huo et al., 2013). The simultaneously recorded soil temperatures had an average of at 6.2 °C (ranging from − 5.5 to 18.1 °C), 5.5 °C (ranging from − 8.5 to 19.5 °C) and 6.8 °C (ranging from

NGS

GS

NGS

FTP MG1 MG2 MG3

Precipitation

40

10 30 0 20 -10 10

Snow cover (cm)

-20 0

0

5

100

10

80

15

60

Soil temperature ( °C)

40

WFPS (%)

Air temperature (°C)

20

50

Snow cover depth

Precipitation (mm d )

(a)

20 0

20 0 -20

(b)

Nov 11

Jan 11

Mar 11

May 11

Jul 11

Sep 11

Nov 11

Fig. 1. Daily precipitation and mean air temperature (a), snow cover depth and daily mean soil temperature (5 cm depth) and soil moisture (over depths of 0–6 cm) in the water-filled pore space (WFPS) (b) at the three field sites during the year-round observational period. The FTP, NGS and GS denote the freezing–thawing period in the early spring (Jan. 23 to Apr. 5 (MG1), Feb. 5 to Apr. 3 (MG2) and Feb. 5 to Mar. 13 (MG3)), the non-growing season (from Nov. 11, 2013 to Apr. 11 (MG1), 12 (MG3) or 13 (MG2), 2014 and from Oct. 12 (MG1 and MG2) or 21 (MG3) to Nov.11, 2014) and the growing season (from Apr. 13 (MG3), 14 (MG2) or 15 (MG1) to Oct. 11 (MG1 and MG2) or 20 (MG3)), respectively (see the text for their definitions). The legends apply for both panels. The standard deviations of the temperatures and the standard errors of the soil moisture are not shown.

890

H. Zhang et al. / Science of the Total Environment 625 (2018) 885–899

−8.4 to 22.8 °C) at MG1, MG2 and MG3, respectively (Fig. 1b). While a significant relationship between the air temperature (Ta, °C) and topsoil temperature (Ts, °C, 5 cm depth) was obtained with simultaneous measurements at MG2, i.e., Ta = (0.866 ± 0.034) Ts − (1.707 ± 0.354) (n = 289, r2 = 0.95, P b 0.001, with 2SD for parameter errors), the air temperatures at MG1 and MG3 (Fig. 1a) were estimated using this empirical function and the site-specific soil temperature measurements. The air temperature estimates at MG1 and MG3 showed annual average values of 3.7 and 4.2 °C, respectively (Table 1). The authors defined an FTP as the duration between the first and the last day with a daily mean soil temperature (5 cm depth) in the range of −2 to 2 °C. During the full-year period of this study, the FTPs occurring in the early spring lasted for approximately 6–8 weeks between January 23 and April 5 at MG1, February 5 and April 3 at MG2, and February 5 and March 13 at MG3 (Fig. 1). According to the measured or estimated air temperatures, the starting and ending dates of the growing season were determined as the period starting when the daily mean air temperature over 6–7 consecutive days was above 5 °C and ending when the daily mean air temperature over 6–7 consecutive days was below 5 °C (Lund et al., 2010; Tang and Arnone, 2013). In 2014, the growing season spanned from April 15 to October 11, April 14 to October 11 and April 13 to October 20 at MG1, MG2 and MG3, respectively (Fig. 1). The growing season length at MG3 was approximately 1 week longer than those at the other two sites.

d.s.)

40

DOC (mg C kg

d.s.) NO3 (mg N kg

60

d.s.) NH4+ (mg N kg

3.1.2. Precipitation and soil moisture The observed annual precipitation near MG2 amounted to 617 mm, 82% of which fell in the growing season (Fig. 1a). Both amount and distribution of the precipitation were normal in the year-round study period compared to the average values of previous long-term (1957–2000) records from the aforementioned meteorological station (Huo et al., 2013). Snowfall that mainly occurred in the non-growing season accounted for 18% of the annual precipitation, with a snow cover

depth of 0.5–11.0 cm (5.2 cm on average). Due to the small distances between the field sites, the annual precipitation amounts at MG1 and MG3 were assumed equivalent as the measured amount at MG2 (Table 1). The temporal variation in the WFPS at any of the sites is closely related to rainfall events and sometimes to freezing–thawing events (Fig. 1). During the spring FTPs, the WFPS peaked significantly at MG2 but increased slightly at MG1 and MG3. Over the year-round period, the WFPS ranged from 28% to 70% (mean: 58%), 22% to 83% (mean: 60%) and 8% to 63% (mean: 40%) at MG1, MG2 and MG3, respectively, and was significantly lower at MG3 (P b 0.01) than the other two sites. 3.1.3. Soil ammonium, nitrate and dissolvable organic carbon During the year-round period, the topsoil ammonium concentrations varied from 0.5 to 22.5 (mean: 8.8), 0.9 to 50.8 (mean: 11) and 1.7 to 54.4 (mean: 19.2) mg N kg− 1 dry soil (d.s.) at MG1, MG2 and MG3, respectively (Fig. 2a). The highest values at MG2 and MG3 appeared during the spring FTPs, while the concentrations at MG1 were at a constantly low level during the whole observation period. The topsoil nitrate concentrations varied from 0.3 to 16.4 (mean: 5.9), 0.4 to 13.9 (mean: 4.5) and 0.1 to 7.3 (mean: 1.7) mg N kg−1 d.s. at MG1, MG2 and MG3, respectively (Fig. 2b). At MG1, the DOC concentrations ranged from 21 to 157 (mean: 52) mg C kg−1 d.s. and were at a low, stable level during the year-round period, with only a slight increase during the FTP (Fig. 2c). At MG2, the topsoil DOC concentrations varied between 18 and 375 (mean: 58) mg C kg−1 d.s., and the highest values appeared during the spring FTP (up to a mean of 133 mg C kg−1 d.s. for the period) and remained at lower stable values during the remaining time of the year (Fig. 2c). At MG3, the DOC concentrations ranged from 28 to 578 (mean: 143) mg C kg− 1 d.s. (Fig. 2c). The concentrations at this site were stably lower in the summer and early autumn (July to September) compared with the other seasons (74 versus 171 mg C kg− 1 d.s. on average),

(a) Ammonium

20 0 20

(b) Nitrate

10 0 600

MG1 MG2 MG3

(c) DOC

400 200 0 Nov 11

Jan 11

Mar 11

May 11

Jul 11

Sep 11

Nov 11

− Fig. 2. Ammonium (NH+ 4 ), nitrate (NO3 ) and dissolved organic carbon (DOC) concentrations within the topsoil (over depths of 0–10 cm) at the three field sites. All of the data are the averages of three subsamples from the mixed soil samples collected from the individual sites, and the standard errors are not shown. The legends apply for all subfigures.

H. Zhang et al. / Science of the Total Environment 625 (2018) 885–899

and the annual maximum values appeared during the spring FTP (up to a mean of 362 mg C kg−1 d.s. for the period) (Fig. 2c). 3.1.4. Correlations among the soil chemical and physical variables Table 2 presents the correlations coefficient matrix for the daily measurements of the soil chemical and physical variables across the three field sites during the entire year-round observational period and under the conditions with soil temperature greater 2 °C at a 5 cm depth. As Table 2 demonstrates, not only when the topsoil was warm (N 2 °C) but also during the entire year-round period, the DOC concentrations had a highly significant (P b 0.001) positive correlation with the ammonium concentration, while they correlated negatively with the nitrate concentrations (P b 0.01). The DOC concentrations correlated negatively with soil temperature and moisture during the year-round period (P b 0.01). However, the negative correlations were not significant under the conditions with a topsoil temperature N2 °C. The nitrate concentrations showed a significantly positive correlation with the topsoil moisture when soil temperature was over 2 °C (P b 0.05) and a significant negative correlation with the soil temperature during the entire year-round period (P b 0.01). There was a highly significant negative correlation between the soil moisture contents and temperatures N 2 °C (P b 0.001). 3.1.5. Soil properties and other site features As listed in the online supplementary materials (Table S1) for the soil profiles at all sites, the topmost soil layers (0–10 cm) contained the highest SOC, with the largest carbon to nitrogen (C:N) ratios, in comparison with those of the deeper layers. The topsoil properties at the three sites show silt loam textures, pH (H2O) values of 6.5–7.1, SOC contents of 46–75 g C kg−1, TN contents of 3.9–5.7 g N kg−1, carbon to nitrogen ratios of 11.9–13.5, and BD values of 0.77–0.91 g cm−1. With an increase in the soil depth, the SOC and TN contents decreased rapidly at all sites. Detailed data for the major soil properties of the different layers are listed in Table S1 for potential future use by readers in model simulations involving the field sites of this study. The groundwater table varied temporally between 0.8 and 3.7 m below the soil surface at MG2 but was not detected at MG1 and MG3 at depths down to 4.5 m (Table 1).

Table 2 Correlation matrix for the daily nitrous oxide (N2O) and nitric oxide (NO) fluxes during the year-round period and the simultaneously measured soil variables across all of the + three field sites. [NO− 3 ], [NH4 ] and [DOC], soil (0–10 cm depths) concentrations of nitrate, ammonium and dissolved organic carbon, respectively. WFPS, soil (0–6 cm depths) moisture content in water-filled pore space. Ts, soil (0–5 cm depths) temperature. Symbols ⁎, ⁎⁎ and ⁎⁎⁎ represent P b 0.05, 0.01 and 0.001, respectively. For all Ts conditions and those of Ts N 2 °C, the correlation coefficients among NO and the soil variables resulted from 81 and 55 sets of observations, respectively. For all Ts conditions and those of Ts N 2 °C, the correlation coefficients among N2O and the soil variables were obtained from 119 and 84 sets of observations, respectively. NO

N2 O

[NO− 3 ]

[NH+ 4 ]

[DOC]

1 0.05 −0.12 −0.03 −0.04 −0.34⁎⁎ 0.32⁎⁎

1 −0.19⁎ 0.17 0.12 0.01 0.05

1 −0.08 −0.25⁎⁎ 0.18 −0.24⁎⁎

1 0.43⁎⁎⁎ −0.13 −0.07

1 −0.27⁎⁎ −0.34⁎⁎⁎

1 −0.06

1

For Ts N 2 °C NO 1 N2 O 0.09 [NO− −0.11 3 ] + [NH4 ] −0.02 [DOC] 0.02 WFPS −0.38⁎⁎ Ts 0.37⁎⁎

1 −0.24⁎ 0.07 0.12 −0.14 0.35⁎⁎

1 0.47⁎⁎⁎ −0.09 −0.20

1 −0.21 −0.15

1 −0.40⁎⁎⁎

1

For all Ts NO N2 O [NO− 3 ] [NH+ 4 ] [DOC] WFPS Ts

1 0.04 −0.30⁎⁎ 0.25⁎ −0.14

WFPS

Ts

891

The measured ANPP values at MG1, MG2 and MG3 were 198, 215 and 347 g m−2 yr−1 on average, respectively (Table 1). 3.2. Nitrous oxide fluxes At MG1, the valid N2O fluxes during the year-round observation period ranged from −0.2 to 2.2 μg N m−2 h−1. The observed N2O fluxes were constantly low and showed no obvious seasonal pattern (Fig. 3a). The annual cumulative N2O emission amount was 71 g N ha− 1 yr− 1 on average, of which the spring FTP and the nongrowing season accounted for approximately 14% and 57%, respectively (Table 3). At MG2, the valid N2O fluxes during the year-round period ranged from − 0.1 to 9.1 μg N m−2 h− 1, with a mean of 1.7 μg N m−2 h−1. Two peak N2O emission events were observed. One was associated with a freezing–thawing alternation during the spring FTP, while the other was triggered by heavy rainfall (N 10 mm d− 1) events in July and August (Fig. 3b). The measured daily fluxes resulted in an annual accumulated flux of 140 g N ha−1 yr−1 on average, of which the spring FTP and the non-growing season contributed approximately 36% and 57%, respectively (Table 3). At MG3, the valid fluxes ranged from − 0.3 to 7.9 μg N m− 2 h−1, with a mean of 2.0 μg N m−2 h−1. No significant pulse of N2O emission was detected during the spring FTP. However, the peak N2O emissions were observed in July and August and were associated with heavy rainfall events (Fig. 3d). The annual cumulative N2O emission amount at this site was 157 g N ha−1 yr−1 on average, of which 6% and 37% were contributed by the spring FTP and the non-growing season, respectively (Table 3). 3.3. Nitric oxide fluxes The conservative NO fluxes at MG2 throughout the year-round period were constantly marginal (Fig. 3c) and ranged from − 0.3 to 1.4 μg N m−2 h−1, with a mean of 0.2 μg N m−2 h−1. The annual cumulative NO emission amount at this site conservatively amounted to 21 g N ha− 1 yr− 1 on average, of which approximately 14% and 48% were contributed by the spring FTP and the growing season, respectively (Table 3). At MG3, the NO fluxes were stable and marginal during the nongrowing season (0.3 μg N m−2 h−1 on average), but they sporadically and slightly pulsed in July and August when drying–wetting alternations occurred due to rainfall events (Fig. 3e). The highest NO flux (25 μg N m−2 h−1) appeared at the end of July when a very slight rainfall wetted the soil. After August, the NO fluxes remained very low even following rainfall events. During the year-round period, the NO fluxes ranged from − 0.8 to 25 μg N m− 2 h− 1, with a mean of 1.3 μg N m−2 h− 1. The annual cumulative NO emission amount at this site was conservatively 94 g N ha−1 yr−1 on average. Clearly, NO fluxes in association with drying–wetting alternations during the growing season overwhelmingly dominated the NO emissions at this site. The contributions of the spring FTP and growing season to the annual cumulative NO flux were approximately 1% and 90%, respectively (Table 3). 3.4. Effects of soil environmental factors on the fluxes of both nitrogenous gases As Table 2 demonstrates, the N2O fluxes across the three field sites correlated negatively with the soil nitrate concentrations not only when the soil (5 cm depth) temperature was over 2 °C but also during the entire year-round period (P b 0.05). However, the determination coefficients of the linear regressions showed that this soil factor alone explained only 3.4–5.5% of the variances of the N2O fluxes. Soil temperatures warmer than 2 °C were correlated positively with the N2O fluxes across the three sites (P b 0.01) (Table 2). This temperature

892

H. Zhang et al. / Science of the Total Environment 625 (2018) 885–899

30

(a) MG1: N2O

20

10

N2O or NO flux (μg N m凟2 h凟1)

0 30

30

(b) MG2: N2O

20

20

10

10

0

0

30

30

(d) MG3: N2O

20

20

10

10

0 Nov 11

(c) MG2: NO

(e) MG3: NO

0 Feb 11

May 11

Aug 11

Nov 11 Nov 11

Feb 11

May 11

Aug 11

Nov 11

Date (2013凟 2014) Fig. 3. Nitrous oxide (N2O) fluxes at the three sites and nitric oxide (NO) fluxes from two alpine grass meadow sites during the year-round observational period. Vertical bars indicate the standard errors of four spatial replicates.

variable alone could explain 9–26% (mean: 17%) of the variances of the gas fluxes during the relatively warm periods of the observational period (Fig. 4b, d and f), whereas the correlation between the soil moisture and the N2O fluxes was insignificant (Table 2). Nevertheless, the combination of both factors improved the explanation of the variances of the N2O fluxes, explaining 17–42% (Table 4). As Table 2 shows, the NO fluxes across the three field sites had a significant positive correlation with the soil temperature and a negative correlation with moisture not only when the topsoil temperature was stably warmer than 2 °C but also during the entire year-round period

(P b 0.01). Either factor alone explained only approximately 10% of the variance of the NO fluxes across the two field sites. The combination of both soil environmental factors significantly improved the explanation of the gas fluxes variances, explaining 21–53% (P b 0.001) (Table 4). Table 2 shows no direct correlation between the N2 O and NO fluxes. However, they had a relationship that was significantly regulated by soil moisture. As Fig. 5 illustrates, the effects of soil moisture on the logarithm of the molar ratio of NO to N2O fluxes during the year-round period can be well described by a negative linear regression (P b 0.001).

Table 3 Cumulative emissions of nitrous oxide (N2O) and nitric oxide (NO) (in g N ha−1 per season, period or year) at the three alpine grass meadow sites. The value in a pair of parentheses indicates the standard error for four spatial replicates. Cumulative N2O emissions

MG1 MG2 MG3 Averagea Boundsb a b

Cumulative NO emissions

FTP

NGS

GS

Annual

FTP

NGS

GS

Annual

13 (4) 51 (29) 15 (8) 26 (21%) ±40

37 (9) 80 (29) 53 (16) 57 (46%) ±46

34 (5) 60 (24) 104 (27) 66 (54%) ±49

71 (10) 140 (35) 157 (16) 123 ±53

– 3 (0) 1 (1) 2 (3%) ±1

– 11 (1) 9 (2) 10 (17%) ±3

– 10 (1) 85 (22) 48 (83%) ±29

– 21 (2) 94 (24) 58 ±32

Given data are the means of the three field sites, with a percentage indicating the average contribution to the annual cumulative fluxes. Given errors indicate the uncertainty magnitudes at the 95% confidence interval. Definitions of the FTP, NGS and GS are found in the caption of Fig. 1.

H. Zhang et al. / Science of the Total Environment 625 (2018) 885–899

30

14 (a) The case without spring-thaw effect (MG1)

12 10 8

25

(b) For Ts > 2 °C at MG1

N2O

20

N2O: F = 0.06Ts + 0.15

NO

(n = 30, r2 = 0.15, P < 0.05)

15

6

10

4

N2O or NO flux (μg N m凟2 h凟1)

893

2

5

0

0

-2 14

-5 30 (c) The case with spring-thaw effect (MG2)

12 10

(d) For Ts > 2 °C at MG2

25

N2O: F = 0.09Ts + 0.38

20

8

(n = 54, r2 = 0.09, P < 0.05) NO: F = 0.34Ts 凟 0.10 (n = 54, r2 = 0.26, P < 0.001)

15

6

10

4 2

5

0

0

-2 14

-5 30 (e) The case without spring-thaw effect (MG3)

12 10

(f) For Ts > 2 °C at MG3

25

N2O: F = 0.20Ts + 0.05

20

8

(n = 60, r = 0.26, P < 0.001) NO: F = 0.34Ts 凟 1.82 2 (n = 60, r = 0.19, P < 0.001) 2

15

6

10

4 2

5

0

0

-2

-12

-8

-4

0

4

8

12

16

20

-5

0

4

8

12

16

20

Soil temperature (°C) Fig. 4. Effects of the soil temperature (Ts, 5 cm depth) on nitrous oxide (N2O) or nitric oxide (NO) fluxes (F). Measurement of NO fluxes was not performed at the MG1 site. The legends apply for all panels. Panels a, c and e are plotted against the full soil temperature ranges while b, d and f are plotted against soil temperatures N2 °C.

3.5. Relationship between annual N2O emissions and grassland vegetation productivity By synthesizing our data and those of other studies that took annual measurements for temperate or alpine grasslands that were ungrazed, either during the summer or year round, the authors obtained a significant simple relationship (Fig. 6) between annual N2O emissions (FN2O, in kg N ha−1 yr−1) and an ANPP no N 0.8 kg m−2 yr−1:

10 8

Ln (NO : N2O) = 凟 0.04WFPS + 0.81 (n = 133, r2 = 0.21, P < 0.001)

6

F N2 O ¼ 0:740  0:306 ANPP þ −0:045  0:112 n ¼ 12; r ¼ 0:70; Pb0:01

ð1Þ Table 4 Dependencies of the nitrous oxide (N2O) or nitric oxide (NO) fluxes (F, in μg N m−2 h−1) upon the soil (5 cm depth) temperature (Ts, in °C) that is stably warmer than 2 °C and the topsoil (0–6 cm depths) moisture in water-filled pore space (M, in %). Site and gas species MG1 MG2 MG3

N2O N2O NO N2O NO

r2

F = aTs + bM + c a

b

c

0.070 0.206 0.014 0.236 0.312

0.011 0.052 −0.013 0.051 −0.037

−0.56 −3.8 0.84 −2.4 −0.01

Ln (NO : N2O)

4 2

2 0 -2 -4 -6

n

P

-8 0 0.17 0.27 0.53 0.42 0.21

30 53 53 59 59

b0.05 b0.001 b0.001 b0.001 b0.001

10

20

30

40

50

60

70

80

90

Soil moisture (% WFPS) Fig. 5. Effects of the topsoil moisture content (over depths of 0–6 cm) in the water-filled pore space (WFPS) on the molar ratios of NO to N2O fluxes across both field sites during the year-round period.

H. Zhang et al. / Science of the Total Environment 625 (2018) 885–899

N2O emission (FN2O, kg N haˉ1 yrˉ1)

894

1.2 FN2O = (0.740f0.306)ANPP ˉ (0.045f0.112) (n = 12, r² = 0.70, P < 0.01)

1.0 0.8 0.6 0.4 0.2 0.0 -0.2 0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

Aboveground net primary productivity (ANPP, kg mˉ2 yrˉ1) Fig. 6. Relationship between annual nitrous oxide (N2O) emissions and the aboveground net primary productivity (ANPP) for grassland ecosystems that are ungrazed either annually or during the summer. The given symbols and bars are the means and standard errors of 3–4 spatial replicates. Filled black diamonds denote the data of this study, while the other symbols display the data from high-altitude alpine grass meadows (Zhang et al., 2014), typical semi-arid temperate steppes (Wolf et al., 2010), natural semi-arid temperate meadow steppes (Yang et al., 2015), an unfertilized artificial pastureland in a semi-arid temperate region (Baoling Mei, pers. comm.) and a shortgrass prairie in North America (Morgan et al., 2001; Mosier et al., 2002). The dashed lines for the parameter errors indicate the uncertainty ranges at the 95% confidence interval.

U F ¼ 0:306 ANPP þ 0:112

ð2Þ

According to the parameter errors given in Eq. (1), the magnitudes of the uncertainty at a 95% confidence interval (UF, in kg N ha−1 yr−1) are given by the ANPP values (Eq. (2)). The parameter errors of Eq. (1) are given as 2SD, which provide the uncertainty ranges at the 95% confidence interval (Fig. 6). 4. Discussion 4.1. Influencing factors on N2O and NO fluxes Most previous studies on alpine grasslands focused exclusively on the growing season (e.g., Hu et al., 2010; Jiang et al., 2010; Wei et al., 2014) and were thus unable to reflect the year-round temporal variations of N2O fluxes as they were affected by seasonally variable environmental conditions. Furthermore, the absence of simultaneously measured NO fluxes in most of the previous studies might have limited a better understanding of the nitrogen biogeochemistry responsible for the observed variable N2O fluxes. This study thus preliminarily attempted to fill this gap with simultaneous, year-round measurements of the fluxes of both gases at multiple alpine grassland sites. As has been reported in previous studies on different ecosystems (Chen et al., 2013b; Cui et al., 2012; Hall et al., 2008) and at laboratory conditions (Yao et al., 2010b), rainfall events after a dry period in a warm season are often followed by increased N2O and NO fluxes. In this study, pulsed NO fluxes at MG3, where the soil moisture contents were generally lower relative to the other two sites, were observed following soil rewetting by a light rainfall (b10 mm) after a dry period in the summer (Fig. 1 and Fig. 3e). In contrast, long or heavy rainfall events were followed by lower NO and increased N2O emissions (Fig. 1 and Fig. 3e). At MG2 (Fig. 1 and Fig. 3c), where drought stress on microbes might not have occurred because of the higher observed soil moisture, the NO fluxes showed no obvious response to the rainfall. Such different responses are in accordance with other studies (e.g., Liu et al., 2010a; Yamulki et al., 1997). A light rainfall following a drought period might have eliminated the drought stress on microbial communities, although the soil moisture remained relatively low. This situation could have stimulated NO production through nitrification and thus elevated the observed fluxes. A heavy or long rainfall event increased the soil

moisture while enhancing N2O and diminishing NO fluxes (Fig. 1 and Fig. 3c), which most likely restricted the diffusion of NO (from nitrification or denitrification) out of the soils, thereby favoring the transformation of NO to N2O via denitrification. Such differences could be supported by the significantly negative relationship of the molar ratios of NO to N2O fluxes against the soil moisture contents (Fig. 5). In cool (autumn) and/or cold (winter) seasons, similar to the reports of other studies (Chen et al., 2013b; Mei et al., 2011), the authors did not detect significant NO and N2O fluxes following precipitation events, even though they led to higher soil moistures. Such a result was attributable to low soil temperatures that suppress microbial processes for the production of NO and N2O (He et al., 2009). However, despite the low soil temperatures, N2O emission pulses were observed during the spring FTP at MG2 (Fig. 3b). Similar results have been reported by some previous studies (e.g., Cui et al., 2012; Fu et al., 2017; Wolf et al., 2010; Yang et al., 2015). Although elevated soil ammonium contents were detected during the spring FTP or following rewetting events (Figs. 1 and 2), the authors could not detect significant correlations between the ammonium concentrations and the NO or N2O fluxes during the full-year period or under the relatively warm conditions (Table 2). It is generally accepted that the NO and N2O production processes in soils are very complex and affected by multiple soil factors, e.g., soil ammonium and nitrate concentrations, temperature and moisture (Ludwig et al., 2001; Williams et al., 1992). Since these factors in wild fields are always changing, and since any one of them can be a limiting factor either occasionally or for long periods (Groffman et al., 2000; Skiba et al., 1997), it is difficult to identify a dominating factor and establish a highly significant relationship between one factor and the in situ measured fluxes of either gas within an annual period. Therefore, previous studies performed in growing seasons commonly reported the absence of a correlation between N2O fluxes and environmental variables under field conditions (e.g., Jiang et al., 2010; Pei et al., 2003; Wei et al., 2014). The influence of environmental variables on NO fluxes from alpine grassland ecosystems was not reported in the previous studies. In this study, the N2O fluxes across all three field sites during the fullyear period and periods with soil (5 cm depth) temperature warmer than 2 °C were found to significantly and negatively correlate with the soil nitrate concentrations (Table 2). The negative correlation may be ascribed (i) directly to the theoretical stepwise reduction of nitrate − (NO− 3 → NO2 → NO → N2O → N2) during denitrification (Hu et al., 2015), and (ii) indirectly to the intensified nitrate uptake by increased prosperous plant growth, as shown by the negative relationship trend (P b 0.10) between the annual mean values of the soil nitrate concentrations and ANPPs across the three meadows sites (adapted from Table 1). As the stepwise nitrate reduction implies, less intensive denitrification may allow for more nitrate in the soil. More abundant vegetation may stimulate N2O production by providing root exudates as carbon substrates for denitrification and nitrogen substrates derived from the priming effects of enhanced DOC (e.g., Kuyakov et al., 2000; Baxendale et al., 2014) for either nitrification or denitrification. These explanations are further supported by the significantly negative correlation between the concentrations of nitrate and DOC (Table 2) as substrates of the nitrogen transformation processes. In addition, a significant linear relationship between the N2O fluxes with the soil temperature observed during both the cool periods and warm non-FTPs (i.e., when the soil temperature at 5 cm depth was stably N2 °C) with the soil temperature could be established at all sites (Fig. 4, Table 2). Moreover, topsoil temperature and moisture had a significant joint regulatory effect on N2O fluxes when the soil temperature were N2 °C (Table 4), though the soil moisture alone showed no obvious effect (Table 2). Likely, the freezing–thawing effects on the N2O fluxes were eliminated if the topsoil temperature (5 cm depth) was stably higher than 2 °C; therefore, significant regulations by the temperature alone (Table 2) or by temperature with moisture (Table 4) could be detected. Another study indicated that usually changes in soil temperature and moisture may explain up to 95% of the temporal variations in field N2O fluxes (Butterbach-

H. Zhang et al. / Science of the Total Environment 625 (2018) 885–899

Bahl et al., 2013). In this study, however, the changes in these two factors only jointly explained 17–42% of the variations in the N2O fluxes (Table 4). Accordingly, the N2O emissions from the vast alpine meadows on the QTP may have the potential to produce a positive feedback on climate change, considering the increasing warming and wetting within this plateau (Chen et al., 2013a). In this study, the NO fluxes across the two sites (MG2 and MG3) during the full-year period and under the relatively warm conditions (with soil temperature at 5 cm depth higher than 2 °C) correlated significantly and positively with the temperature and negatively with the moisture (Fig. 4, Table 4), while they showed no significant correlations with ammonium, nitrate and DOC concentrations (Table 2). This finding indicates that, the NO emissions from the typical alpine grass meadows during the full-year period were primarily regulated by either soil temperature or moisture, both of which jointly explained 17% of the variance of the daily NO fluxes across the two field sites (P b 0.001). 4.2. Importance of the non-growing season and spring-thaw event on the emissions of either gas Enhanced N2O emissions have been detected during the spring FTP in boreal and temperate regions in previous investigations, including continental semi-arid steppe (Wolf et al., 2010; Yang et al., 2015), marsh (Song et al., 2008), forest (Goldberg et al., 2010), and agricultural systems (Cui et al., 2012). Previous studies have suggested that topsoil moisture has crucial effects on the emissions of N2O during freezing– thawing alternation periods due to the enhancement of denitrification with low oxygen availability (Teepe et al., 2000, 2004). This mechanism might account for the large variations in the N2O emissions during the spring FTP among the three field sites of this study, which showed a coefficient of variation (CV) of 81%. In this study, relatively higher N2O fluxes were observed during the spring FTP at MG2, but were hardly detectable at MG1 and MG3 (Fig. 3a and d). The flux pulses at MG2 might have been mainly due to the higher soil moisture contents during the FTP (63% versus 36–49% at the other two sites). The wetter topsoil at MG2 might have been attributed to the subsoil ice pan (the remaining ice from the winter because the groundwater table was much closer to the land surface), which could hamper the infiltration of water from thawing, as well as the lateral subflow of surface runoff due to thawing at the upper slope. The explanation that the soil moisture content was the primary driving factor of the spring-thaw N2O pulses can also be supported by other studies (Koponen and Martikainen, 2004; Teepe et al., 2004; Wolf et al., 2010; Yang et al., 2015), which indicate that a soil moisture of approximately 60% WFPS is conducive to higher N2O emissions during FTPs either under field conditions (Wolf et al., 2010; Yang et al., 2015) or laboratory conditions (Koponen and Martikainen, 2004; Teepe et al., 2004). Though an elevated soil mineral nitrogen content was observed during the FTP at all sites (Fig. 2a and b), it did not seem to be a restricting factor on the N2O emissions most likely due to the soil moisture limitations at MG1 and MG3. A few studies have indicated that the contribution from the nongrowing season, especially during the spring FTP, is very important for an assessment of annual N2O emissions from natural or managed ecosystems in temperate, boreal or frigid alpine regions (Röver et al., 1998; Wolf et al., 2010; Wu et al., 2010a). Röver et al. (1998) found that the three-month winter period accounts for 70% of the annual N2O release from temperate arable land. Wagner-Riddle et al. (2007) conducted a 5-year study on agricultural ecosystems in Ontario, Canada, and reported that the non-growing season accounts for 30–90% of the annual emissions mainly due to elevated N2O fluxes during soil thawing. Based on 5-year continuous automatic measurements with a high temporal resolution, Wu et al. (2010a) reported N2O pulses (Höglwald Forest, Germany) induced by thawing events, which contributed up to 73% of the annual N2O emissions. Filippa et al. (2009) nearly continuously measured N2O fluxes from a seasonally snow-covered subalpine meadow at Niwot Ridge, Colorado, and found that the winter

895

season accounted for 19–28% of the total annual N2O fluxes. For typical or meadow steppes under different stocking rates in semi-arid temperate regions, the spring FTP has been found to account for 45 ± 27% of the annual cumulative N2O fluxes (Wolf et al., 2010; Yang et al., 2015). In this study, the spring FTP and the entire non-growing season accounted for 9–36% (mean: 21%) and 38–57% (mean: 51%) of the annual N2O emissions, respectively, well supporting the aforementioned hypothesis. Accordingly, both this study and previous studies suggest that reliable measurements during not only the spring FTP but also the entire non-growing season are essential for accurate qualifications of annual N2O emissions for natural or managed terrestrial ecosystems in temperate, boreal or alpine regions. In this study, no NO emission pulse was detected during the spring FTP (Figs. 1 and 3). This agrees with the previous studies in a western European forest (Wu et al., 2010a), a wheat-maize cropping system in northern China (Cui et al., 2012), a subalpine meadow in western North America (Filippa et al., 2009) and an incubation of soil cores in laboratory (Koponen et al., 2006). Generally, NO production occurs within soils with good aeration as well as suitable temperatures, which are conditions that favor nitrification (Davidson, 1992). In this study, high soil water contents and/or low soil temperatures of approximately 0 °C inhibited the production of NO by decreasing the oxygen diffusion and NO transport from the soil to the atmosphere or by reducing the activities of enzymes for nitrification (Koponen et al., 2006; Menneer et al., 2005; Yao et al., 2010b). As a result, the NO fluxes during the spring FTP accounted for only 1–14% of the annual NO emissions (Table 3). However, approximately 10–52% of the annual NO emissions occurred during the non-growing season (Table 3). 4.3. Importance of nitrous oxide and nitric oxide emissions from grasslands The observations of annual N2O emissions from the three unfertilized high-altitude alpine meadow sites in the eastern QTP region revealed emissions of 0.123 ± 0.053 (2SD) kg N ha−1 yr− 1 (Table 3). These specific annual cumulative fluxes for the year-round period were slightly lower in magnitude than the annual fluxes of 0.18–0.27 kg N ha−1 yr− 1 from unfertilized high-altitude alpine grass or shrub meadows in the northeastern QTP that were exclusively grazed in winter (Fu et al., 2017; Zhang et al., 2014). Compared with other grassland types that were ungrazed either year round or in the summer, the annual N2O emissions observed in this this study fell in the range of 0.01–0.48 kg N ha− 1 yr− 1 from the unfertilized typical or meadow steppes of semi-arid temperate zones in the northern China (Wolf et al., 2010; Yang et al., 2015; Mei et al., pers. comm.) and were around the lower range limit of 0.13–0.44 kg N ha−1 yr−1 from the temperate shrub and shortgrass steppes in North America (Epstein et al., 1998; Matson et al., 1991; Mosier et al., 2002; Mummey et al., 1997). Including the data of this study, those reported in the literature and one observation provided by Mei B (pers. comm.), there have been a total of 29 observations of annual cumulative N2O emissions from different types of unfertilized temperate/montane grasslands in China (Table 5). These observations have an average of 0.22 ± 0.04 (2SD) kg N ha−1 yr−1. This error was propagated from the double standard deviations (Table 5) adapted from the individual standard errors and spatially replicated numbers presented by the reporters. Simply extrapolating this flux for a total area of 337 million ha (Zhang et al., 2010), the overall N2O emissions from the natural grasslands in China may be estimated at 73 ± 15 (2SD) Gg N yr−1 (1 Gg = 109 g). This value is in good agreement with the previous model estimates of 77 ± 13 and 71 Gg N yr−1 by Zhang et al., 2010 and Tian et al., 2011. Nevertheless, there is still significant uncertainty due to the scarcity of annual N2O observations for both desert steppes and alpine steppes. These two types collectively account for 39% and 51% of the grassland areas in Inner Mongolia (Han et al., 2008) and the QTP region (Hu et al., 2010; Wei et al., 2015), respectively. Therefore, further field studies are still needed in desert and alpine steppe regions to clarify this uncertainty.

896

H. Zhang et al. / Science of the Total Environment 625 (2018) 885–899

Table 5 Available observations of annual nitrous oxide emissions (kg N ha−1) from temperate grasslands throughout China. A, arid. AP, alpine. AT, artificially planted. N, natural community. SA, semi-arid. T, temperate. UF, unfertilized. LG, lightly grazed. MG, moderately grazed. HG, heavily grazed. SG, summer grazed. WG, winter grazed. 2SD, double standard deviations representing the uncertainty magnitude at the 95% confidence interval. Different values for the same grassland type with the same climate and management practices were measured in different years or at different sites. Grassland (climate/management)

Mean

2SD

Remarks

Meadow steppe (SA, T/UF, N, SG) Meadow steppe (SA, T/UF, N, WG) Meadow steppe (SA, T/UF, N, UG) Meadow steppe (SA, T/UF, AT, UG) Typical steppe (SA, T/UF, N, UG) Typical steppe (SA, T/UF, N, UG) Typical steppe (SA, T/UF, N, HG) Typical steppe (SA, T/UF, N, HG) Typical steppe (SA, T/UF, N, UG) Typical steppe (SA, T/UF, N, UG) Typical steppe (SA, T/UF, N, UG) Typical steppe (SA, T/UF, N, LG) Typical steppe (SA, T/UF, N, LG) Typical steppe (SA, T/UF, N, MG) Typical steppe (SA, T/UF, N, MG) Typical steppe (SA, T/UF, N, HG) Typical steppe (SA, T/UF, N, HG) Typical steppe (SA, T/UF, N, WG) Typical steppe (SA, T/UF, N, UG) Typical steppe (SA, T/UF, N, UG) Typical steppe (SA, T/UF, N, UG) Desert steppe (A, T/UF, N, UG) Desert steppe (A, T/UF, N, UG) Shrub meadow (T, AP/UF, N, WG) Shrub meadow (T, AP/UF, N, WG) Grass meadow (T, AP/UF, N, WG) Grass meadow (T, AP/UF, N, WG, early SG) Grass meadow (T, AP/UF, N, WG, early SG) Grass meadow (T, AP/UF, N, WG, early SG) Average

0.19 0.15 0.43 0.48 0.24 0.30 0.06 0.26 0.22 0.28 0.17 0.20 0.10 0.15 0.11 0.12 0.17 0.01 0.30 0.40 0.44 0.22 0.30 0.18 0.27 0.18 0.07 0.14 0.16 0.22

0.12 0.12 0.28 0.76 0.34 0.42 0.06 0.34 0.28 0.20 0.12 0.12 0.08 0.20 0.16 0.04 0.12 0.12 0.06 0.07 0.03 0.29 0.28 0.12

Yang et al., 2015

0.04 0.04 0.14 0.06 0.04

Zhang et al., 2014 This study

Mei B, pers. comm. Liu et al., 2010b

Wolf et al., 2010

Peng et al., 2011 Peng et al., 2011 Wang et al., 2011 Fu et al., 2017

This study presented conservative estimates of annual NO fluxes with an average of 0.021 to 0.094 kg N ha−1 yr−1 (Table 3). These magnitudes are similar to the NO flux (approximately 0.08 kg N ha−1 yr−1) measured in the laboratory for an alpine tundra soil (Yu et al., 2010). The measurements at MG2 and MG3 between June 3 and October 26 in 2014 (Fig. 3b and d) resulted in cumulative NO emissions of 0.009 to 0.077 kg N ha−1 on average, respectively, which account for 43% to 82%, respectively, of the annual cumulative fluxes. During the comparable period from June 3 to October 26 in 2013, a field study using the same method measured an average cumulative NO emission of 0.013 kg N ha−1 (adapted from Gao et al., 2016) from a similar alpine meadow in the eastern QTP, which could be extrapolated to an annual flux of 0.029 kg N ha−1 yr−1 by referring to the periodical contributions obtained in this study. There have been a few field measurements of NO emissions from grasslands in other regions. Martin et al., 1998 observed NO emissions of approximately 1.3 kg N ha−1 yr−1 from an unfertilized shortgrass steppe in North America. Tilsner et al., 2003 reported emissions of approximately 1.5 kg N ha−1 yr−1 from an extensively managed (unfertilized) meadow in northeastern Bavaria, Germany. Their annual NO fluxes were obviously much higher than those of this study, suggesting that the alpine meadows in the QTP may have a low potential for NO emissions. Low net NOx (i.e., NO + NO2) emissions (0.01–0.04 kg N ha−1 yr−1) measured using the widely accepted dynamic chamber technique have been reported for a typical semi-arid temperate steppe in Inner Mongolia, China (Holst et al., 2007). According to the canopy resistance values of NO2 (270–560 s m−1) for a meadow grassland (Plake et al., 2015) and the concentrations of this gas measured by Holst and his colleagues, the annual NO2 depositional flux may be ~0.10 kg N ha−1 yr−1. Then, the NO emission fluxes from the grassland investigated by Holst and his colleagues may be roughly estimated at 0.11–0.14 kg N ha− 1 yr− 1. These estimates are slightly

higher relative to the fluxes measured in the alpine meadows, which in turn were underestimated to some extent by this study due to the aforementioned method limitations. Nevertheless, the results from this study and Holst et al., 2007 and the results adapted from Gao et al. (2016) may indicate that either temperate high-altitude alpine meadows or typical temperate semi-arid steppes are marginal sources of atmospheric NO, with a conservative annual flux of 0.079 ± 0.043 (2SD) kg N ha−1 yr−1. 4.4. Relationship of annual N2O emissions with the aboveground net primary productivity It is generally accepted that seasonal N2O emissions can be predicted by the crop yield or biomass productivity of a cropping system (e.g., McSwiney and Robertson, 2005; Chen et al., 2008; Abdalla et al., 2010; Liu et al., 2012; Yao et al., 2013), because crop production depends on management and environmental factors such as nitrogen application, plowing practices, and weather conditions, and each of these factors can influence the production and emission of N2O gas (Chen et al., 2008). Similarly, the authors also found that the annual emissions of N2O from unfertilized temperate steppes and meadows across different regions that were subject to an ungrazing management regime either year round or in the summer, could be predicted by aboveground net primary productivity (ANPP) (Eq. (1)). In obtaining the relationship given by Eq. (1), all the cases (e.g., Du et al., 2006, 2008; Li et al., 2012) that applied a DN method that most likely introduced largely positive biases into the measured N2O fluxes (Fu et al., 2017; Zheng et al., 2008) were excluded. The DN method is a GC–ECD approach that uses pure N2 as the carrier gas, but does not involve ascarite or a buffering gas, with 5–10% methane or carbon dioxide in pure N2, for the N2O analysis (Wang et al., 2010; Zheng et al., 2008). This method has been proven to significantly overestimate measured N2O fluxes from croplands or grasslands. It especially imposes large positive biases on weak N2O sources when static opaque chambers are used in gas sampling (Fu et al., 2017; Zheng et al., 2008). The apparently positive relationship between the annual N2O fluxes and ANPPs (Fig. 6) probably resulted from (i) the direct stimulatory effects of soil moisture on both variables and (ii) the indirect stimulatory effects from plant growth on N2O production. Higher soil moisture due to greater amounts of precipitation in the previous growing season, additional snow in the winter, or a shallower groundwater table may not only stimulate N2O emissions during spring FTPs and thus higher annual N2O fluxes (e.g., Fu et al., 2017; Wolf et al., 2010) but also mitigate the water deficit for vegetation demands and thus facilitate higher ANPPs. At a typical semi-arid temperate steppe region, for instance, the topsoil moisture alone was found to account for 71% (P b 0.01) of the variations in the ANPPs subject to different stocking rates (adapted from Chen et al., 2011). In nitrogen-poor terrestrial ecosystems, such as natural temperate steppes and meadows, plant growth may indirectly stimulate N2O production and thus emission by root exudation of DOC. In addition, living root-originated DOC can promote the priming effects of organic nitrogen mineralization and thus provide nitrogen nutrient for further plant growth (e.g., Kuyakov et al., 2000; van der Krift and Berendse, 2001; Grigulis et al., 2013; Baxendale et al., 2014). Living root-originated DOC can also provide nitrogen substrates for microbial nitrification and denitrification whereby N2O is produced as a byproduct or intermediate product (e.g., Firestone and Davidson, 1989; Butterbach-Bahl et al., 2013) and directly provide carbon substrates for the heterotrophic microbial processes involved in N2O formation. This explanation is supported by the observations of this study (Table 1), which showed a linearly positive relationship between the annually averaged soil DOC concentrations and the ANPPs (P b 0.05). Eq. (1) may provide a simple, empirical approach to estimate annual N2O emissions from grassland ecosystems that are ungrazed either in the summer or year round and subject to temperate or frigid alpine climates. However, future studies are still needed to further confirm the

H. Zhang et al. / Science of the Total Environment 625 (2018) 885–899

link between N2O emissions and an ANPP range that is wider than that shown in Fig. 6. 5. Conclusions This study is based on year-round observations of nitrous oxide (N2O) and nitric oxide (NO) emissions from typical alpine grass meadows in the eastern Qinghai–Tibetan Plateau that are grazed conventionally/traditionally in the winter and from the late spring to early summer. The measured gas fluxes indicate a weak source of atmospheric N2O and are at a slightly lower level relative to other reports of N2O emissions from alpine meadow ecosystems. However, they fall within the wide range of annual fluxes from grasslands that are both subject to frigid alpine and temperate semi-arid or arid climates and different grazing practices. Despite large spatial variabilities in the pulsed N2O fluxes during the spring freezing–thawing period due to the spatial variability of soil moisture among the investigated sites, the nongrowing season (NGS) accounts for one-third to one-half of the annual cumulative emissions, highlighting the importance of accurate observations of N2O fluxes in the NGS. Simultaneous year-round observations indicate that high-altitude alpine grass meadows that are ungrazed in summer are marginal sources of atmospheric NO. Nitric oxide fluxes are constantly low most of the time. The fluxes that are stimulated by drying–wetting events in the growing season overwhelming dominate the annual cumulative NO emissions. Overall, the annual observations of N2O and NO emissions along with the soil characteristics and environmental factors can contribute to reducing the uncertainty regarding the annual budgets of N2O and NO emissions from regional grasslands with similar climate and management practices while improving the databases available for model calibration and evaluation endeavors. Annual cumulative N2O fluxes show a significant dependence on the aboveground net primary productivity (ANPP), which may provide a potential simple, empirical approach to estimate N2O emission inventories for grassland ecosystems subject to temperate or frigid alpine climates and ungrazed either in summer year round. Nevertheless, confirmation of this relationship is still needed for a wider ANPP range. In addition, further studies are needed to improve the accuracy in measuring NO emissions from alpine grasslands. Supplementary data to this article can be found online at https://doi. org/10.1016/j.scitotenv.2017.12.216. Acknowledgements This study was jointly financed by the Ministry of Science and Technology of China (grant numbers: 2016YFA0602303, 2012CB417100) and the National Natural Science Foundation of China (grant numbers: 41603075, 41375152, 41321064). Thanks are also due to Xiaolong Wang, Siqi Li, Yongfeng Fu, Fei Lin, Xiaoxia Hu, Lin Wang, Yan Liu, Ping Li, Lei Li, Yanqiang Wang and Qingqi Jia for their substantial help in laboratory sample analysis and field observation. References Abalos, D., De Deyn, G.B., Kuyper, T.W., van Groenigen, J.W., 2014. Plant species identity surpasses species richness as a key driver of N2O emissions from grassland. Glob. Chang. Biol. 20:265–275. https://doi.org/10.1111/gcb.12350. Abdalla, M., Jones, M., Ambus, P., Williams, M., 2010. Emissions of nitrous oxide from Irish arable soils: effects of tillage and reduced N input. Nutr. Cycl. Agroecosyst. 86:53–65. https://doi.org/10.1007/s10705-009-9273-8. Adams, J.M., Faure, H., Fauredenard, L., Mcglade, J.M., Woodward, F.I., 1990. Increases in terrestrial carbon storage from the Last Glacial Maximum to the present. Nature 348:711–714. https://doi.org/10.1038/348711a0. Baxendale, C., Orwin, K.H., Poly, F., Pommier, T., Bardgett, R.D., 2014. Are plant–soil feedback responses explained by plant traits? New Phytol. 204:408–423. https://doi.org/ 10.1111/nph.12915. Butterbach-Bahl, K., Baggs, E.M., Dannenmann, M., Kiese, R., Zechmeister-Boltenstern, S., 2013. Nitrous oxide emissions from soils: how well do we understand the processes and their controls? Philos. Trans. R. Soc. B 368, 20130122. https://doi.org/10.1098/ rstb.2013.0122.

897

Cao, G.M., Xu, X.L., Long, R.J., Wang, Q.L., Wang, C.T., Du, Y.G., Zhao, X.Q., 2008. Methane emissions by alpine plant communities in the Qinghai–Tibet Plateau. Biol. Lett. 4: 681–684. https://doi.org/10.1098/rsbl.2008.0373. Chen, S., Huang, Y., Zou, J., 2008. Relationship between nitrous oxide emission and winter wheat production. Biol. Fertil. Soils 44:985–989. https://doi.org/10.1007/s00374-0080284-4. Chen, W.W., Wolf, B., Zheng, X.H., Yao, Z.S., Butterbach-Bahl, K., Brüggemann, N., Liu, C.Y., Han, S.S., Han, X.G., 2011. Annual methane uptake by temperate semiarid steppes as regulated by stocking rates, aboveground plant biomass and topsoil air permeability. Glob. Chang. Biol. 17:2803–2816. https://doi.org/10.1111/ j.1365-2486.2011.02444.x. Chen, H., Zhu, Q.A., Peng, C.H., Wu, N., Wang, Y.F., Fang, X.Q., Gao, Y.H., Zhu, D., Yang, G., Tian, J.Q., Kang, X.M., Piao, S.L., Ouyang, H., Xiang, W.H., Luo, Z.B., Jiang, H., Song, X.Z., Zhang, Y., Yu, G.R., Zhao, X.Q., Gong, P., Yao, T.D., Wu, J.H., 2013a. The impacts of climate change and human activities on biogeochemical cycles on the Qinghai– Tibetan Plateau. Glob. Chang. Biol. 19:2940–2955. https://doi.org/10.1111/gcb.12277. Chen, W.W., Zheng, X.H., Chen, Q., Wolf, B., Butterbach-Bahl, K., Bruggemann, N., Lin, S., 2013b. Effects of increasing precipitation and nitrogen deposition on CH4 and N2O fluxes and ecosystem respiration in a degraded steppe in Inner Mongolia, China. Geoderma 192:335–340. https://doi.org/10.1016/j.geoderma.2012.08.018. Cowan, N.J., Norman, P., Famulari, D., Levy, P.E., Reay, D.S., Skiba, U.M., 2015. Spatial variability and hotspots of soil N2O fluxes from intensively grazed grassland. Biogeosciences 12:1585–1596. https://doi.org/10.5194/bg-12-1585-2015. Cui, F., Yan, G.X., Zhou, Z.X., Zheng, X.H., Deng, J., 2012. Annual emissions of nitrous oxide and nitric oxide from a wheat-maize cropping system on a silt loam calcareous soil in the North China Plain. Soil Biol. Biochem. 48:10–19. https://doi.org/10.1016/ j.soilbio.2012.01.007. Davidson, E.A., 1992. Sources of nitric oxide and nitrous oxide following wetting of dry soil. Soil Sci. Soc. Am. J. 56:95–102. https://doi.org/10.2136/ sssaj1992.03615995005600010015x. Du, R., Lu, D.R., Wang, G.C., 2006. Diurnal, seasonal, and inter-annual variations of N2O fluxes from native semi-arid grassland soils of inner Mongolia. Soil Biol. Biochem. 38:3474–3482. https://doi.org/10.1016/j.soilbio.2006.06.012. Du, Y., Cui, Y., Xu, X., Liang, D., Long, R., Cao, G., 2008. Nitrous oxide emissions from two alpine meadows in the Qinghai–Tibetan Plateau. Plant Soil 311:245–254. https:// doi.org/10.1007/s11104-008-9727-9. Epstein, H.E., Burke, I.C., Mosier, A.R., Hutchinson, G.L., 1998. Plant functional type effects on trace gas fluxes in the shortgrass steppe. Biogeochemistry 42:145–168. https:// doi.org/10.1023/a:1005959001235. Filippa, G., Freppaz, M., Williams, M.W., Helmig, D., Liptzin, D., Seok, B., Hall, B., Chowanski, K., 2009. Winter and summer nitrous oxide and nitrogen oxides fluxes from a seasonally snow-covered subalpine meadow at Niwot Ridge, Colorado. Biogeochemistry 95:131–149. https://doi.org/10.1007/s10533-009-9304-1. Firestone, M.K., Davidson, E.A., 1989. Microbiological basis of NO and N2O production and consumption in soil. In: Andreae, M.O., Schimel, D.S. (Eds.), Exchange of Trace Gases Between Terrestrial Ecosystems and the Atmosphere. John Wiley and Sons, Chichester, UK, pp. 7–21. Fu, Y.F., Lin, F., Hu, X.X., Liu, C.Y., Zheng, X.H., Zhang, W., Cao, G.M., 2018. Quantification of year-round methane and nitrous oxide fluxes in a typical alpine shrub meadow on the Qinghai–Tibetan Plateau. Agric. Ecosyst. Environ. 255:27–36. https://doi.org/ 10.1016/j.agee.2017.12.00. Gao, Y.H., Ma, X.X., Cooper, D.J., 2016. Short-term effect of nitrogen addition on nitric oxide emissions from an alpine meadow in the Tibetan Plateau. Environ. Sci. Pollut. R. 23:12474–12479. https://doi.org/10.1007/s11356-016-6763-5. Goldberg, S.D., Borken, W., Gebauer, G., 2010. N2O emission in a Norway spruce forest due to soil frost: concentration and isotope profiles shed a new light on an old story. Biogeochemistry 97:21–30. https://doi.org/10.1007/s10533-009-9294-z. Grigulis, K., Lavorel, S., Krainer, U., Legay, N., Baxendale, C., Dumont, M., Kastl, E., Arnoldi, C., Bardgett, R.D., Poly, F., Pommier, T., Schloter, M., Tappeiner, U., Bahn, M., Clément, J.-C., 2013. Relative contributions of plant traits and soil microbial properties to mountain grassland ecosystem services. J. Ecol. 101:47–57. https://doi.org/10.1111/ 1365-2745.12014. Groffman, P.M., Brumme, R., Butterbach-Bahl, K., Dobbie, K.E., Mosier, A.R., Ojima, D., Papen, H., Parton, W.J., Smith, K.A., Wagner-Riddle, C., 2000. Evaluating annual nitrous oxide fluxes at the ecosystem scale. Glob. Biogeochem. Cycles 14:1061–1070. https://doi.org/10.1029/1999gb001227. Hall, S.J., Huber, D., Grimm, N.B., 2008. Soil N2O and NO emissions from an arid, urban ecosystem. J. Geophys. Res.-Biogeo. 113:G01016. https://doi.org/10.1029/2007jg000523. Han, G., Hao, X., Zhao, M., Wang, M., Ellert, B.H., Willms, W., Wang, M., 2008. Effect of grazing intensity on carbon and nitrogen in soil and vegetation in a meadow steppe in Inner Mongolia. Agric. Ecosyst. Environ. 125:21–32. https://doi.org/10.1016/ j.agee.2007.11.009. He, F., Jiang, R., Chen, Q., Zhang, F., Su, F., 2009. Nitrous oxide emissions from an intensively managed greenhouse vegetable cropping system in northern China. Environ. Pollut. 157:1666–1672. https://doi.org/10.1016/j.envpol.2008.12.017. Holst, J., Liu, C., Brüggemann, N., Butterbach-Bahl, K., Zheng, X., Wang, Y., Han, S., Yao, Z., Yue, J., Han, X., 2007. Microbial N turnover and N-oxide (N2O/NO/NO2) fluxes in semi-arid grassland of Inner Mongolia. Ecosystems 10:623–634. https://doi.org/ 10.1007/s10021-007-9043-x. Hu, Y.G., Chang, X.F., Lin, X.W., Wang, Y.F., Wang, S.P., Duan, J.C., Zhang, Z.H., Yang, X.X., Luo, C.Y., Xu, G.P., Zhao, X.Q., 2010. Effects of warming and grazing on N2O fluxes in an alpine meadow ecosystem on the Tibetan plateau. Soil Biol. Biochem. 42: 944–952. https://doi.org/10.1016/j.soilbio.2010.02.011. Hu, H.W., Chen, D., He, J.Z., 2015. Microbial regulation of terrestrial nitrous oxide formation: understanding the biological pathways for prediction of emission rates. FEMS Microbiol. Rev. 39:729–749. https://doi.org/10.1093/femsre/fuv021.

898

H. Zhang et al. / Science of the Total Environment 625 (2018) 885–899

Huo, L.L., Chen, Z.K., Zou, Y.C., Lu, X.G., Guo, J.W., Tang, X.G., 2013. Effect of Zoige alpine wetland degradation on the density and fractions of soil organic carbon. Ecol. Eng. 51:287–295. https://doi.org/10.1016/j.ecoleng.2012.12.020. Hutchinson, G.L., Mosier, A.R., 1981. Improved soil cover method for field measurement of nitrous oxide fluxes. Soil Sci. Soc. Am. J. 45:311–316. https://doi.org/10.2136/ sssaj1981.03615995004500020017x. IPCC, 2013. Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. Jiang, C.M., Yu, G.R., Fang, H.J., Cao, G.M., Li, Y.N., 2010. Short-term effect of increasing nitrogen deposition on CO2, CH4 and N2O fluxes in an alpine meadow on the Qinghai– Tibetan Plateau, China. Atmos. Environ. 44:2920–2926. https://doi.org/10.1016/ j.atmosenv.2010.03.030. Koponen, H.T., Martikainen, P.J., 2004. Soil water content and freezing temperature affect freeze-thaw related N2O production in organic soil. Nutr. Cycl. Agroecosyst. 69: 213–219. https://doi.org/10.1023/b:fres.0000035172.37839.24. Koponen, H.T., Duran, C.E., Maljanen, M., Hytonen, J., Martikainen, P.J., 2006. Temperature responses of NO and N2O emissions from boreal organic soil. Soil Biol. Biochem. 38: 1779–1787. https://doi.org/10.1016/j.soilbio.2005.12.004. van der Krift, T.A.J., Berendse, F., 2001. The effect of plant species on soil nitrogen mineralization. J. Ecol. 89:555–561. https://doi.org/10.1046/j.0022-0477.2001.00580.x. Kroon, P.S., Hensen, A., van den Bulk, W.C.M., Jongejan, P.A.C., Vermeulen, A.T., 2008. The importance of reducing the systematic error due to non-linearity in N2O flux measurements by static chambers. Nutr. Cycl. Agroecosyst. 82:175–186. https://doi.org/ 10.1007/s10705-008-9179-x. Kuyakov, Y., Friedel, J.K., Stahr, K., 2000. Review of mechanisms and quantification of priming effects. Soil Biol. Biochem. 32:1485–1498. https://doi.org/10.1016/S00380717(00)00084-5. Lee, D.S., Bouwman, A.F., Asman, W.A.H., Dentener, F.J., Hoek, K.W.V.D., Olivier, J.G.J., 1997. Emissions of nitric oxide, nitrous oxide and ammonia from grasslands on a global scale. In: Jarvis, S.C., Pain, B.F. (Eds.), Gaseous Nitrogen Emissions From Grasslands. CAB International, Wallingford, pp. 353–371. Li, K.H., Gong, Y.M., Song, W., Lv, J.L., Chang, Y.H., Hu, Y.K., Tian, C.Y., Christie, P., Liu, X.J., 2012. No significant nitrous oxide emissions during spring thaw under grazing and nitrogen addition in an alpine grassland. Glob. Chang. Biol. 18:2546–2554. https:// doi.org/10.1111/j.1365-2486.2012.02704.x. Li, Y.Y., Dong, S.K., Liu, S.L., Zhou, H.K., Gao, Q.Z., Cao, G.M., Wang, X.X., Su, X.K., Zhang, Y., Tang, L., Zhao, H.D., Wu, X.Y., 2015. Seasonal changes of CO2, CH4 and N2O fluxes in different types of alpine grassland in the Qinghai–Tibetan Plateau of China. Soil Biol. Biochem. 80:306–314. https://doi.org/10.1016/j.soilbio.2014.10.026. Liu, C., Zheng, X., Zhou, Z., Han, S., Wang, Y., Wang, K., Liang, W., Li, M., Chen, D., Yang, Z., 2010a. Nitrous oxide and nitric oxide emissions from an irrigated cotton field in northern China. Plant Soil 332:123–134. https://doi.org/10.1007/s11104-009-0278-5. Liu, X.R., Dong, Y.S., Qi, Y.C., Li, S.G., 2010b. N2O fluxes from the native and grazed semiarid steppes and their driving factors in Inner Mongolia, China. Nutr. Cycl. Agroecosyst. 86:231–240. https://doi.org/10.1007/s10705-009-9287-2. Liu, C., Wang, K., Zheng, X., 2012. Responses of N2O and CH4 fluxes to fertilizer nitrogen addition rates in an irrigated wheat-maize cropping system in northern China. Biogeosciences 9:839–850. https://doi.org/10.5194/bg-9-839-2012. Ludwig, J., Meixner, F.X., Vogel, B., Forstner, J., 2001. Soil-air exchange of nitric oxide: an overview of processes, environmental vactors, and modeling studies. Biogeochemistry 52:225–257. https://doi.org/10.1023/a:1006424330555. Lund, M., Lafleur, P.M., Roulet, N.T., Lindroth, A., Christensen, T.R., Aurela, M., Chojnicki, B.H., Flanagan, L.B., Humphreys, E.R., Laurila, T., Oechel, W.C., Olejnik, J., Rinne, J., Schubert, P., Nilsson, M.B., 2010. Variability in exchange of CO2 across 12 northern peatland and tundra sites. Glob. Chang. Biol. 16:2436–2448. https://doi.org/ 10.1111/j.1365-2486.2009.02104.x. Martin, R.E., Scholes, M.C., Mosier, A.R., Ojima, D.S., Holland, E.A., Parton, W.J., 1998. Controls on annual emissions of nitric oxide from soils of the Colorado shortgrass steppe. Glob. Biogeochem. Cycles 12:81–91. https://doi.org/10.1029/97gb03501. Matson, P.A., Volkmann, C., Coppinger, K., Reiners, W.A., 1991. Annual nitrous oxide flux and soil nitrogen characteristics in sagebrush steppe ecosystems. Biogeochemistry 14:1–12. https://doi.org/10.1007/bf00000883. McSwiney, C.P., Robertson, G.P., 2005. Nonlinear response of N2O flux to incremental fertilizer addition in a continuous maize (Zea mays L.) cropping system. Glob. Chang. Biol. 11:1712–1719. https://doi.org/10.1111/j.1365-2486.2005.01040.x. Mei, B.L., Zheng, X.H., Xie, B.H., Dong, H.B., Zhou, Z.X., Wang, R., Deng, J., Cui, F., Tong, H.J., Zhu, J., 2009. Nitric oxide emissions from conventional vegetable fields in southeastern China. Atmos. Environ. 43:2762–2769. https://doi.org/10.1016/j.atmosenv.2009.02.040. Mei, B.L., Zheng, X.H., Xie, B.H., Dong, H.B., Yao, Z.S., Liu, C.Y., Zhou, Z.X., Wang, R., Deng, J., Zhu, J.G., 2011. Characteristics of multiple-year nitrous oxide emissions from conventional vegetable fields in southeastern China. J. Geophys. Res.-Atmos. 116. https:// doi.org/10.1029/2010jd015059. Menneer, J.C., Ledgard, S., McLay, C., Silvester, W., 2005. Animal treading stimulates denitrification in soil under pasture. Soil Biol. Biochem. 37:1625–1629. https://doi.org/ 10.1016/j.soilbio.2005.01.023. Morgan, J.A., Lecain, D.R., Mosier, A.R., Milchunas, D.G., 2001. Elevated CO2 enhances water relations and productivity and affects gas exchange in C3 and C4 grasses of the Colorado shortgrass steppe. Glob. Chang. Biol. 7:451–466. https://doi.org/ 10.1046/j.1365-2486.2001.00415.x. Mosier, A.R., Bronson, K., Schimel, D.S., Valentine, D.W., Parton, W.J., 1991. CH4 and N2O fluxes in native, fertilised and cultivated grasslands. Nature 350, 330–332. Mosier, A.R., Morgan, J.A., King, J.Y., LeCain, D., Milchunas, D.G., 2002. Soil-atmosphere exchange of CH4, CO2, NOx, and N2O in the Colorado shortgrass steppe under elevated CO2. Plant Soil 240:201–211. https://doi.org/10.1023/a:1015783801324.

Mummey, D.L., Smith, J.L., Bolton, H., 1997. Small-scale spatial and temporal variability of N2O flux from a shrub-steppe ecosystem. Soil Biol. Biochem. 29:1699–1706. https:// doi.org/10.1016/s0038-0717(97)00077-1. Myhre, G., Shindell, D., Bréon, F.-M., Collins, W., Fuglestvedt, J., Huang, J., Koch, D., Lamarque, J.-F., Lee, D., Mendoza, B., Nakajima, T., Robock, A., Stephens, G., Takemura, T., Zhang, H., 2013. Anthropogenic and natural radiative forcing. In: Stocker, T.F., Qin, D., Plattner, G.-K., Tignor, M., Allen, S.K., Boschung, J., Nauels, A., Xia, Y., Bex, V., Midgley, P.M. (Eds.), Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, pp. 659–740. Pei, Z., Ouyang, H., Zhou, C., Xu, X., 2003. Fluxes of CO2, CH4 and N2O from alpine grassland in the Tibetan Plateau. J. Grogr. Sci. 13:27–34. https://doi.org/10.1007/bf02873144. Peng, Q., Qi, Y.C., Dong, Y.S., Xiao, S.S., He, Y.T., 2011. Soil nitrous oxide emissions from a typical semiarid temperate steppe in inner Mongolia: effects of mineral nitrogen fertilizer levels and forms. Plant Soil 342:345–357. https://doi.org/10.1007/s11104-0100699-1. Pilegaard, K., 2013. Processes regulating nitric oxide emissions from soils. Philos. Trans. R. Soc. B 368, 20130126. https://doi.org/10.1098/rstb.2013.0126. Plake, D., Stella, P., Moravek, A., Mayer, J.C., Ammann, C., Held, A., Trebs, I., 2015. Comparison of ozone deposition measured with the dynamic chamber and the eddy covariance method. Agric. For. Meteorol. 206:97–112. https://doi.org/10.1016/ j.agrformet.2015.02.014. Ravishankara, A.R., Daniel, J.S., Portmann, R.W., 2009. Nitrous oxide (N2O): the dominant ozone-depleting substance emitted in the 21st century. Science 326:123–125. https://doi.org/10.1126/science.1176985. Röver, M., Heinemeyer, O., Kaiser, E.A., 1998. Microbial induced nitrous oxide emissions from an arable soil during winter. Soil Biol. Biochem. 30:1859–1865. https:// doi.org/10.1016/s0038-0717(98)00080-7. Skiba, U., Fowler, D., Smith, K.A., 1997. Nitric oxide emissions from agricultural soils in temperate and tropical climates: sources, controls and mitigation options. Nutr. Cycl. Agroecosyst. 48:139–153. https://doi.org/10.1023/a:1009734514983. Song, C.C., Zhang, J.B., Wang, Y.Y., Wang, Y.S., Zhao, Z.C., 2008. Emission of CO2, CH4 and N2O from freshwater marsh in northeast of China. J. Environ. Manag. 88:428–436. https://doi.org/10.1016/j.jenvman.2007.03.030. Tang, G.P., Arnone, J.A., 2013. Trends in surface air temperature and temperature extremes in the Great Basin during the 20th century from ground-based observations. J. Geophys. Res.-Atmos. 118:3579–3589. https://doi.org/10.1002/jgrd.50360. Teepe, R., Brumme, R., Beese, F., 2000. Nitrous oxide emissions from frozen soils under agricultural, fallow and forest land. Soil Biol. Biochem. 32:1807–1810. https://doi.org/ 10.1016/s0038-0717(00)00078-x. Teepe, R., Vor, A., Beese, F., Ludwig, B., 2004. Emissions of N2O from soils during cycles of freezing and thawing and the effects of soil water, texture and duration of freezing. Eur. J. Soil Sci. 55:357–365. https://doi.org/10.1111/j.1365-2389.2004.00602.x. Tian, H.Q., Xu, X.F., Lu, C.Q., Liu, M.L., Ren, W., Chen, G.S., Melillo, J., Liu, J.Y., 2011. Net exchanges of CO2, CH4, and N2O between China's terrestrial ecosystems and the atmosphere and their contributions to global climate warming. J. Geophys. Res.-Biogeo. 116. https://doi.org/10.1029/2010jg001393. Tilsner, J., Wrage, N., Lauf, J., Gebauer, G., 2003. Emission of gaseous nitrogen oxides from an extensively managed grassland in NE Bavaria, Germany. I. Annual budgets of N2O and NOx emissions. Biogeochemistry 63:229–247. https://doi.org/10.1023/a: 1023365432388. Turner, D.A., Chen, D., Galbally, I.E., Leuning, R., Edis, R.B., Li, Y., Kelly, K., Phillips, F., 2008. Spatial variability of nitrous oxide emissions from an Australian irrigated dairy pasture. Plant Soil 309:77–88. https://doi.org/10.1007/s11104-008-9639-8. Valente, R.J., Thornton, F.C., Williams, E.J., 1995. Field comparison of static and flowthrough chamber techniques for measurement of soil NO emission. J. Geophys. Res.-Atmos. 100:21147–21152. https://doi.org/10.1029/95JD01875. Wagner-Riddle, C., Furon, A., McLaughlin, N.L., Lee, I., Barbeau, J., Jayasundara, S., Parkin, G., Von Bertoldi, P., Warland, J., 2007. Intensive measurement of nitrous oxide emissions from a corn-soybean-wheat rotation under two contrasting management systems over 5 years. Glob. Chang. Biol. 13:1722–1736. https://doi.org/10.1111/j.13652486.2007.01388.x. Wang, Y.H., Wang, Y.S., Ling, H., 2010. A new carrier gas type for accurate measurement of N2O by GC-ECD. Adv. Atmos. Sci. 27:1322–1330. https://doi.org/10.1007/s00376010-9212-2. Wang, Z.W., Hao, X.Y., Shan, D., Han, G.D., Zhao, M.L., Willms, W.D., Wang, Z., Han, X., 2011. Influence of increasing temperature and nitrogen input on greenhouse gas emissions from a desert steppe soil in Inner Mongolia. Soil Sci. Plant Nutr. 57: 508–518. https://doi.org/10.1080/00380768.2011.591283. Wang, K., Zheng, X.H., Pihlatie, M., Vesala, T., Liu, C.Y., Haapanala, S., Mammarella, I., Rannik, U., Liu, H.Z., 2013. Comparison between static chamber and tunable diode laser-based eddy covariance techniques for measuring nitrous oxide fluxes from a cotton field. Agric. For. Meteorol. 171:9–19. https://doi.org/10.1016/j.agrformet.2012.11.009. Wei, D., Xu, R., Liu, Y.W., Wang, Y.H., Wang, Y.S., 2014. Three-year study of CO2 efflux and CH4/N2O fluxes at an alpine steppe site on the central Tibetan Plateau and their responses to simulated N deposition. Geoderma 232:88–96. https://doi.org/10.1016/ j.geoderma.2014.05.002. Wei, D., Xu, R., Tenzin, T., Wang, Y.S., Wang, Y.H., 2015. Considerable methane uptake by alpine grasslands despite the cold climate: in situ measurements on the central Tibetan Plateau, 2008–2013. Glob. Chang. Biol. 21:777–788. https://doi.org/10.1111/ gcb.12690. Williams, E.J., Hutchinson, G.L., Fehsenfeld, F.C., 1992. NOx and N2O emissions from soil. Glob. Biogeochem. Cycles 6:351–388. https://doi.org/10.1029/92GB02124. Wolf, B., Zheng, X.H., Brueggemann, N., Chen, W.W., Dannenmann, M., Han, X.G., Sutton, M.A., Wu, H.H., Yao, Z.S., Butterbach-Bahl, K., 2010. Grazing-induced reduction of

H. Zhang et al. / Science of the Total Environment 625 (2018) 885–899 natural nitrous oxide release from continental steppe. Nature 464:881–884. https:// doi.org/10.1038/nature08931. Wu, X., Brüuggemann, N., Gasche, R., Shen, Z.Y., Wolf, B., Butterbach-Bahl, K., 2010a. Environmental controls over soil-atmosphere exchange of N2O, NO, and CO2 in a temperate Norway spruce forest. Glob. Biogeochem. Cycles 24. https://doi.org/10.1029/ 2009gb003616. Wu, Y.B., Tan, H.C., Deng, Y.C., Wu, J., Xu, X.L., Wang, Y.F., Tang, Y.H., Higashi, T., Cui, X.Y., 2010b. Partitioning pattern of carbon flux in a Kobresia grassland on the Qinghai– Tibetan Plateau revealed by field 13C pulse-labeling. Glob. Chang. Biol. 16: 2322–2333. https://doi.org/10.1111/j.1365-2486.2009.02069.x. Yamulki, S., Harrison, R.M., Goulding, K.W.T., Webster, C.P., 1997. N2O, NO and NO2 fluxes from a grassland: effect of soil pH. Soil Biol. Biochem. 29:1199–1208. https://doi.org/ 10.1016/s0038-0717(97)00032-1. Yang, X.M., Chen, H.Q., Gong, Y.S., Zheng, X.H., Fan, M.S., Kuzyakov, Y., 2015. Nitrous oxide emissions from an agro-pastoral ecotone of northern China depending on land uses. Agric. Ecosyst. Environ. 213:241–251. https://doi.org/10.1016/j.agee.2015.08.011. Yao, Z., Wolf, B., Chen, W., Butterbach-Bahl, K., Brüggemann, N., Wiesmeier, M., Dannenmann, M., Blank, B., Zheng, X., 2010a. Spatial variability of N2O, CH4 and CO2 fluxes within the Xilin River catchment of Inner Mongolia, China: a soil core study. Plant Soil 331:341–359. https://doi.org/10.1007/s11104-009-0257-x. Yao, Z.S., Wu, X., Wolf, B., Dannenmann, M., Butterbach-Bahl, K., Brüggemann, N., Chen, W.W., Zheng, X.H., 2010b. Soil-atmosphere exchange potential of NO and N2O in different land use types of Inner Mongolia as affected by soil temperature, soil moisture, freeze-thaw, and drying-wetting events. J. Geophys. Res.-Atmos. 115:D17116. https://doi.org/10.1029/2009JD013528. Yao, Z., Zheng, X., Wang, R., Dong, H., Xie, B., Mei, B., Zhou, Z., Zhu, J., 2013. Greenhouse gas fluxes and NO release from a Chinese subtropical rice-winter wheat rotation system

899

under nitrogen fertilizer management. J. Geophys. Res.-Biogeo. 118:623–638. https:// doi.org/10.1002/jgrg.20061. Yao, Z., Wei, Y., Liu, C., Zheng, X., Xie, B., 2015. Organically fertilized tea plantation stimulates N2O emissions and lowers NO fluxes in subtropical China. Biogeosciences 12: 5915–5928. https://doi.org/10.5194/bg-12-5915-2015. Yu, J.B., Meixner, F.X., Sun, W.D., Mamtimin, B., Xia, C.H., Xie, W.J., 2010. Biogenic nitric oxide emission of mountain soils sampled from different vertical landscape zones in the Changbai Mountains, northeastern China. Environ. Sci. Technol. 44: 4122–4128. https://doi.org/10.1021/es100380m. Zhang, F., Qi, J., Li, F.M., Li, C.S., Li, C.B., 2010. Quantifying nitrous oxide emissions from Chinese grasslands with a process-based model. Biogeosciences 7:2039–2050. https:// doi.org/10.5194/bg-7-2039-2010. Zhang, W., Liu, C.Y., Zheng, X.H., Fu, Y.F., Hu, X.X., Cao, G.M., Butterbach-Bahl, K., 2014. The increasing distribution area of zokor mounds weaken greenhouse gas uptakes by alpine meadows in the Qinghai–Tibetan Plateau. Soil Biol. Biochem. 71:105–112. https://doi.org/10.1016/j.soilbio.2014.01.005. Zheng, X.H., Mei, B.L., Wang, Y.H., Xie, B.H., Wang, Y.S., Dong, H.B., Xu, H., Chen, G.X., Cai, Z.C., Yue, J., Gu, J.X., Su, F., Zou, J.W., Zhu, J.G., 2008. Quantification of N2O fluxes from soil-plant systems may be biased by the applied gas chromatograph methodology. Plant Soil 311:211–234. https://doi.org/10.1007/s11104-008-9673-6. Zheng, Y., Yang, W., Sun, X., Wang, S.P., Rui, Y.C., Luo, C.Y., Guo, L.D., 2012. Methanotrophic community structure and activity under warming and grazing of alpine meadow on the Tibetan Plateau. Appl. Microbiol. Biotechnol. 93:2193–2203. https://doi.org/ 10.1007/s00253-011-3535-5.