Potential methane production rates and its carbon isotopic composition from ornithogenic tundra soils in coastal Antarctic

BAO Tao, ZHU Renbin,*, BAI Bo & XU Hua

1 Institute of Polar Environment, School of Earth and Space Sciences, University of Science and Technology of China, Hefei 230026, China;

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

1 Introduction

Methane (CH4) is one of the important, chemically active greenhouse gases (GHGs), and its contribution to global greenhouse effect is about 15% at the one-hundred-year time scale[1-4]. The CH4emissions from Arctic terrestrial ecosystems, especially boreal wetlands, play a special role in the global carbon cycle due to the amplified warming of the region during the past few decades[5-6]. In maritime Antarctica,the coastal ice free areas contain some of the largest marine animal colonies on a global scale. Marine animal colonies,tundra vegetation and the interactions between them form a special terrestrial ecosystem[7]. Marine animals including penguins and other seabirds play an important role in the nutrient cycling of the ecosystems by transferring carbon and nitrogen from the marine to the terrestrial environment[7-9].The deposition of a large amount of penguin guano strongly influences the physical and chemical properties of local soils, and produces a kind of special soil named ornithogenic soil[7,10-12]. Recently, field observations have indicated that penguin colonies are a significant source for atmospheric CH4in maritime Antarctica[13-15]. Nevertheless, CH4production from fertile ornithogenic tundra soils are still poorly understood in coastal Antarctica. Tundra CH4emission is an integrated effect of the production, oxidation and transport of CH4in the soils. A better knowledge of these processes affecting CH4emission may provide more information about the response of tundra CH4emission to changing climate.

The technique of stable carbon isotopes has been proved to be a useful tool in studying the complicated processes of CH4production and emission[16-19]. CH4emitted from various sources may have different isotopic characteristics because of its isotopic fractionation effects. Isotope fractionation happens in all the major processes of CH4emission. The12C-substrate is preferentially utilized by methanogens for CH4production, and once formed,12CH4is consumed faster than13CH4by methanotrophs, and12CH4is transported faster than13CH4as well[20-23]. The δ13C-CH4produced through acetic acid fermentation ranged from −60‰ to −50‰, whereas the δ13C-CH4in the reduction of CO2/H2was below this range[24-25]. Generally the CH4, emitted from the combustion of fossil fuel, had relatively heavy isotopic compositions with the δ13C-CH4range of −50‰～−30‰, whereas CH4from biomass combustion had especially high13C-CH4(−28‰～−12‰). In addition, the relative contribution of acetate to CH4production (fac) and the fraction of CH4oxidized (fox) can be quantitatively estimated[19,26-27]using mass balance equations based on the measurements of δ13C in CH4, CO2and acetate,and of the isotope fractionation factors (αCO2/CH4, εacetate/CH4, αoxand εtransport). At present, studies on isotopes of soil CH4have been conducted in different ecosystems[25,28-31]. However, few data are available on the isotopes of CH4emitted from tundra soils in coastal Antarctica.

In this study, the ornithogenic tundra soil profiles were collected from four penguin colonies in coastal Antarctica during the 22nd Chinese National Antarctic Research Expedition (22nd CHINARE). Potential CH4production rates and its δ13C were measured based upon laboratory incubation experiments. Our objectives are (1) to measure potential CH4production rates from ornithogenic tundra soils and the carbon isotopic compositions of CH4emitted from the soils; (2) to discuss the factors affecting CH4production, consumption and its isotopic compositions.

2 Materials and methods

2.1 Study sites and microcosm sampling

Tundra ornithogenic soils were collected from the following four penguin colonies during the 22nd CHINARE.

The first site is located on Ardley Island, West Antarctica(62°13′S, 58°56′W), with 2.0 km length and 1.5 km width.This island was defined as an area of special scientific interest by the Scientific Committee of Antarctic Research (SCAR).It has one of the most important penguin colonies in maritime Antarctica[8]. Approximately 10200 penguin individuals colonized this island during the breeding season including Gentoo penguins (Pygoscelis papua), Adélie penguins (P.adeliae), and Chinstrap penguins (P. antarctica)[8,32]. Recently,CH4and N2O emissions from this penguin colony have been measured in situ by Zhu et al.[14-15]. In the active colony,continuous deposition of fresh guano and penguin trampling inhibits vegetation establishment, and soils are covered by layers of guano[8]. The ornithogenic soils had well expressed O (organic and A (accumulation) horizons, and were covered by thick continuous moss cover. One ornithogenic soil core(named AI) with a depth of 30 cm was collected at a poorly drained area about 100 m from the active penguin colony.

The second site is located in an emperor penguin colony at Prydz Bay, East Antarctica (69°22′S, 76°24′E). This colony is about 10 km from the Chinese Antarctic Zhongshan Station.This area has a typical cold, dry polar continental climate.It is one of the most important emperor penguin colonies in coastal Antarctica with about 10000 emperor penguins breeding here every year. One ornithogenic soil profile (named PB, about 25 cm depth) was collected from the colony. Only some algae were present on the surface of the ornithogenic soils at the sampling site[14].

The third site is located on Gardner Island (68°34′S,77°52′E) and the fourth site is located on Magnetic Island(68°32′S, 77°54′E), both in East Antarctica. These two islands are important Adélie penguins (P. adeliae) colonies. Two ornithogenic soil profiles (named GI and MI) were sampled from Gardner Island and Magnetic Island, respectively.Sparse algae grew on the ornithogenic tundra soils. All the areas above are covered by accumulated snow and ice during winter and the ornithogenic soils are frozen. Every summer the snow and ice melt, and soil freezing–thawing frequency increases considerably in these ice−free areas. The sampling sites are shown in Figure 1.

Ornithogenic soil profiles (about 20 cm depth) were collected from the different penguin colonies by in situ sectioning of the soil layer from top to bottom using a bamboo scoop during January and February 2006. Intact soil cores (6 cm inner diameter, about 30 cm depth) were obtained from penguin colonies by hammering 30 cm long PVC tubes into the soils and carefully digging the tubes out[8,33]. These samples were kept at −10ºC and transported to the laboratory in China for the incubation experiment.

2.2 Incubation experiment and measurements of potential CH4 production rates

Figure 1 Sampling sites for ornithogenic tundra soil profiles in penguin colonies of coastal Antarctica. (1) AI: penguin colony on Ardley Island, West Antarctica; (2) PB: emperor penguin colony at Prydz Bay, East Antarctica; (3) GI and MI: Adélie penguin colonies on Gardner Island and Magnetic Island, respectively, East Antarctica.

In the laboratory, all ornithogenic soil profiles were sectioned into three parts, and then mixed homogeneously. About 100 g (fresh weight) of soils were put into glass vessels(500 mL) for the incubation experiments. To investigate the potential ぼux of CH4in the field, we added 1 mL water to make up for water lost during the experiment in order to maintain field moisture. Additionally, to avoid the disturbance associated with thawing, which may lead to unusually high trace gas ぼuxes, the samples were completely thawed and then incubated in the dark at 4ºC, which is very close to local mean air temperature in the austral summer. The glass vessels with the soil samples were divided into two groups:the first group was incubated under mbient air conditions,to simulate the potential CH4fluxes from ornithogenic soils under local natural conditions. The second group was incubated under N2to establish anaerobic conditions,which was used to simulate the potential CH4fluxes under waterlogged−soil conditions from the effects of snowmelt water on the ornithogenic tundra soils. During the incubation, the headspace gases were collected every two hours, and then stored in 18 mL evacuated vials before analysis. After each gas sampling, the headspace gas was renewed by flushing with ambient air repeatedly or re-evacuating and re-flushing the glass vessel with N2five times to avoid CH4build-up in the headspace[34]. Two repetitions were made for each soil sample[35].

The CH4concentrations were determined by gas chromatography (Shimadzu GC−12A, Japan) equipped with a ぼame ionization detector (FID) and a molecular sieve 5A column (2.6 mm i.d. × 2.0 m) under N2(40 mL∙ min−1)as a carrier gas at 80ºC[36]. The standard gas for CH4was 8 ppmv. The variance coefficient for standard samples was within 0.1%–0.6% in 24 h. Potential CH4production rates were estimated from the changing rate of headspace CH4concentrations. The incubation time intervals were 24 h for the experimental samples. The cumulative ぼux was calculated by integrating the potential ぼuxes over the incubation period.

2.3 Measurement of δ13C-CH4

The carbon isotopic compositions of CH4were analyzed by using a continuous flow technique coupled to a Finnigan MAT−253 isotope ratio mass spectrometer (Thermo Finnigan,Bremen, Germany)[37]. The CO2in gas samples was directly analyzed while CH4in gas samples was converted into CO2and separated primarily on a PreCon (pre−concentration device). Then, the gas was piped into a GC equipped with a Pora PLOT Q column (25 m length; 0.35 mm i.d.) at 25°C under 2.0×105Pa for further separation. The separated gases were finally transferred into the mass spectrometer for δ13C determination. The reference and carrier gases used were CO2(99.999% purity and –23.73‰ δ13CPDB-value) and He(99.999% purity, 20 mL∙min–1). The precision of the repeated analysis was ±0.2‰ when 2.02 μL∙L–1CH4was injected. The Isotope ratios are presented as δ values, which are defined as: δx=[(Rsample/Rstandard) −1]×1000‰, where δxis the δ value of the heavy isotope x and R is the ratio of the heavy isotope(at%, atom percent) to the light isotope (at%).

2.4 General analyses of ornithogenic tundra soils

The ornithogenic soil samples were separated from soil profiles and mixed homogeneously for the general analyses.The pH was determined in distilled water and in a 1M KCl solution (soil: solution ratio 1:3). Total organic carbon (TOC)was analyzed from the dry soil by the potassium dichromate volumetric method with an analytic error of 2.5%[33]. Total nitrogen (TN), NH4+−N and NO3−−N were determined by an ion−selective electrode method with an analytical error of ＜5.0%[15]. Total phosphorus (TP) content was determined by ultraviolet visible spectrometry (UVS) with an analytical error of ＜2.0%. Sulphur (S) was analysed by the KI volume method after combustion in a SRJK−2 hightemperature furnace with an analytical error of ＜5.0%[33].Soil gravimetric moisture content (MC) was determined by drying the soil at 105 °C for 12 h. MCwas calculated as:MC= the weight of the lost water/dry soil weight×100%.The precision and accuracy of our results were monitored using reference materials (GBW07) in every batch of analysis. The measured values of the reference materials were in good agreement with the reference values, and the differences were within ±5%.

2.5 Data analysis

Statistical analysis was made with OriginPro 7.5 and Microsoft Excel 2007 for Window XP. Statistically significant differences in potential fluxes of CH4and δ13CCH4between incubation groups over each incubation period were assessed using a two-sample t-test. In all analyses where p＜0.05, the factor tested and the relationship were considered statistically significant[35,38].

3 Results and discussion

3.1 Physicochemical properties of ornithogenic tundra soils

As summarized in Table 1, there was a different soil moisture environment between these ornithogenic soils. Soil moisture in AI and PB was 38% and 76%, respectively. The ornithogenic soils were slightly acidic to neutral, with the pH values of 5.2−7.3. The TOC and TN contents in these samples were highly variable, ranging from 5.5%−14.7%for TOC, and from 0.49%−3.60% for TN. The TOC and TN contents in the ornithogenic soils from AI were one order of magnitude lower than those in other ornithogenic soils due to the deposition of less penguin guano. However,the AI soil had much higher C/N ratio (11.1) than other soils. The highest NH4+—N concentration in the soils was that from MI, followed by PB, and the lowest was AI. The NH4+—N concentration in the soils AI was two magnitudes lower than those in other ornithogenic soils. There was no great difference in NO3−—N concentrations between the ornithogenic soils. These soils in coastal Antarctica also had particularly high P (3.74—37.18 mg∙g−1) and S (1.65—2.38 mg∙g−1) contents (Table 1).

3.2 Potential CH4 production rates from ornithogenic tundra soils

As shown in Figure 2, the production rates of CH4from the ornithogenic soil MI increased gradually with the time both under ambient air or under N2incubation, and then reached a plateau after 15 h. Overall, the soil MI had almost the same CH4production rates under ambient air (mean 19.63±3.35 μg∙kg−1∙h−1) and under N2incubation (mean 20.32±5.40 μg∙kg−1∙h−1). The CH4cumulative emissions in the headspace showed a linear increase during the laboratory incubation (linear regression: positive slope with R2=0.99,p＜0.01 for the MI soils). Similarly, the CH4production rates from the GI soil rapidly increased to the highest level within 18 h, and then decreased to a low level under ambient air or under N2incubation. The GI soil produced low CH4emissions (mean 3.07±0.63 μg∙kg−1∙h−1) under ambient air while GI produced significant amounts of CH4emission under N2incubation (mean 6.10±1.95 μg∙kg−1∙h−1),indicating that anaerobic conditions might stimulate soil CH4production and emission because methanogen is a obligate anaerobe[39]. In addition, the fertile ornithogenic soils generally contained considerable organic matter due to the deposition of penguin guano, and this provided rich a matrix for soil CH4production[24]. The cumulative CH4emissions in the headspace from MI and GI both showed a linear increase during the ambient air or N2incubation, but the mean cumulative emissions from MI and GI were higher under N2incubation (0.20 mg C∙kg−1and 0.02 mg C∙kg−1, respectively)than under ambient air incubation (0.13 mg C∙kg−1and 0.006 mg C∙kg−1, respectively), further confirming that anaerobic conditions in the soils might increase CH4production rates.

Table 1 Physicochemical properties of ornithogenic soils at the sampling sites

Figure 2 The CH4 production rates and δ13C-CH4 from MI, GI, PB and AI ornithogenic tundra soils during ambient air and N2 incubations.Note: these are the mean values at each soil sample, and the vertical bars indicate standard errors of CH4 production rates and δ13C-CH4.

Similarly to GI, the CH4production rates from the AI soil rapidly increased to the highest level, and then decreased to a low level under both ambient air and under N2incubation.The AI soil produced almost equal CH4emission amounts under ambient air (mean 3.23±0.97 μg∙kg−1∙h−1) and under N2incubation (mean 3.35±0.96 μg∙kg−1∙h−1). The cumulative CH4emissions from AI were significantly lower than from other ornithogenic soils due to very low soil TOC and TN contents.

Contrary to the MI and GI soils from Adélie penguin colonies, the PB ornithogenic soil from an emperor penguin colony produced a significant amount of CH4emissions(17.47±3.74 μg∙kg−1∙h−1) under ambient air incubation, but negligible CH4emissions (0.70±0.28 μg∙kg−1∙h−1) under N2incubation. The cumulative CH4emissions were also much higher under ambient air (mean 0.11 mgC∙kg−1) than under N2incubation (mean 0.01 mgC∙kg−1). The PB ornithogenic soil had the highest soil moisture of all the samples, and soil water was highly saturated as shown in Table 1. Extremely anaerobic conditions under N2incubation could inhibit the CH4production or prevent CH4emissions escaping[40].Another possible reason was that the ornithogenic soils might emit more bioavailable organic carbon under aerobic conditions than under anaerobic conditions, leading to much higher CH4production rates under ambient air incubation[24].

It should be emphasized that water content is an important factor in determining soil microbial activity in Antarctic environments[41]. Therefore soil moisture is probably an important contributing factor to the higher production rates of CH4associated with ornithogenic tundra soils in coastal Antarctica. In our assessments, there was no need to saturate the ornithogenic soils to produce slurries, as in the methods used by Harris and Tibbles[42]and Cocks et al.[43], as our samples were at typical moisture contents for the different penguin colonies, and thus our results could reぼect potential CH4production rates under moisture conditions present in thefield.

3.3 The δ13C-CH4 emitted from ornithogenic tundra soils

The mean δ13C-CH4emitted from MI and GI were higher under N2incubation (mean −43.53±2.26‰ and−39.63±0.95‰, respectively) than under the ambient air incubation (mean −50.91±2.04‰ and −42.81±1.13‰,respectively), indicating that the anaerobic conditions were conducive to13C enrichment in the CH4(Figure 2 and Table 2). Similarly, the δ13C-CH4emitted from PB had significant differences under ambient air incubation and under N2incubation. The δ13C-CH4ranged from −32.9‰ to −41.1‰under N2incubation, whereas it ranged from −67.2‰ to−36.28‰ under ambient air incubation, further confirming that the CH4emitted from the ornithogenic soils could enrich13C under highly anaerobic conditions. The δ13C-CH4emitted from AI ranged from −35.0‰ to −29.0‰ under ambient air incubation, whereas it ranged from −37.8 ‰ to −33.6‰under N2incubation. The δ13C values of CH4emitted from AI were significantly higher than those from MI, GI and PB.Furthermore, the δ13C-CH4from AI was significantly higher under the ambient air incubation (mean −29.93±0.87‰) than under N2incubation (mean −35.88±0.76‰).

The CH4emitted from different sources might have different isotopic characteristics because of isotopic fractionation effects. Biogenic CH4generally has depleted13C (−65‰～−55‰) compared to the CH4emitted from the burning of fossil fuels (−50‰～−30‰)and biomass burning (−28‰～−12‰)[24]. In the soil systems, acetate fermentation and CO2/H2reduction are the two main mechanisms of CH4production. Generally CH4produced through CO2/H2reduction is poor in13C(δ13C=−60‰～−110‰), while CH4produced through acetic acid reduction is enriched in13C (δ13C=−50‰～−65‰)[44-47]. In addition, the CH4oxidation process can lead to13C enrichment[8,48]. In this study, the mean δ13C-CH4emitted from the MI, GI and PB soil profiles was 7.54‰,3.19‰ and 9.21‰ higher, respectively, under N2incubation than under ambient air incubation. This indicated that soil CH4production in MI, GI and PB might be predominantly due to acetate fermentation under N2incubation, whereas13C−depleted CH4emitted from MI, GI and PB might be predominantly produced through CO2/H2reduction under ambient air incubation. The13C-CH4enrichment under N2incubation confirmed that acetate fermentation might become an important source for atmospheric CH4in highly waterlogged tundra soils in coastal Antarctica.

For the AI ornithogenic soil, δ13C-CH4was 5.63‰higher under ambient air incubation than under N2incubation.The13C enrichment in CH4might be due to CH4oxidation by methane—oxidizing bacteria under ambient air incubation.In addition, soil moisture was much lower in AI than in other ornithogenic soils (Table 1), and such low soil moisture might lead to the formation of highly unsaturated soil. Generally unsaturated soils could provide more organic carbon that soil microorganisms need under aerobic conditions than under anaerobic conditions[24], and more organic carbon was transformed into13C, therefore the δ13C-CH4for AI was higher under ambient air incubation.

3.4 Correlations between δ13C-CH4 and CH4 production rates

Overall δ13C-CH4showed a significant negative correlation with soil CH4production rates under N2incubation (R2=0.41,p＜0.01) or under ambient air incubation (R2=0.50, p＜0.01).This indicated that δ13C-CH4had a close relationship withsoil CH4production rates, and decreased with increasing CH4production rates (Figure 3). For all the soil profiles, a significant negative correlation existed between δ13C-CH4and CH4concentration in the headspace under N2or ambient air incubation, further confirming that CH4emissions from the ornithogenic soils had an important effect on carbon isotopes of CH4(Figure 4). Quay et al.[49]also found the negative correlation between δ13C-CH4and CH4concentrations. The negative correlation might be due to CH4oxidation[8,48].Generally,12C-CH4had a faster reaction rate during the processes of CH4oxidation, with less12C occurring in the residual CH4, leading to enrichment in13C-CH4. The correlation between the CH4concentration and δ13C-CH4was more significant under ambient air incubation than under N2incubation, indicating that ambient air incubation might increase soil CH4oxidation compared with N2incubation.

Table 2 The mean δ13C-CH4 and CH4 production rate from Antarctic ornithogenic soils during the laboratory incubation experiments

Figure 3 Correlations between δ13C-CH4 and CH4 production rates from the ornithogenic tundra soil profiles in coastal Antarctica.

3.5 Effects of penguins on soil CH4 production rates and δ13C-CH4

As illustrated in Figure 5, potential CH4production from ornithogenic soils showed a significant positive correlation with total phosphorus (TP) and NH4+−N contents, pH and soil moisture (Mc), but a strong negative correlation with soil total sulfur (TS) content. On the contrary, the δ13CCH4showed a significant negative correlation with TP and NH4+−N contents, pH and Mc. The deposition of penguin guano strongly effects the physical and chemical properties of tundra soils via the influence of microbes[7]. These ornithogenic tundra soils had particularly high P and S levels(Table 1), two elements which were typical in penguin guano and could be used as indicators for the amount of penguin guano deposition in soils/sediments[50]. The correlations between potential CH4production and TP, TS contents from ornithogenic soils indicated that the amount of penguin guano deposition increased potential CH4production rates from tundra soils, but decreased the δ13C-CH4.

The CH4production rates from ornithogenic soils are correlated with the supply of extra methanogenic substrate by the deposition of penguin guano, and high CH4production might be the result of dissolved CH4, large microbial populations, rapid decomposition of guano, provision of labile organic compounds to methanogens, and anaerobic condition in the soils[51-52]. Similar results were also obtained in our previous studies[13,15,36,38,53]. Soil NH4+–N content may limit the capacity of the soil to take up CH4, as NH4+–N can inhibit the activity of methanotrophs[54-55]. Therefore, in this study high NH4+–N input from penguin guano might decrease soil CH4uptake, but greatly stimulate CH4production by methanogens in the ornithogenic soils. Favorable conditions for high CH4production are created by physical and chemical processes related to penguin activities: the input of penguin guano and penguin tramp[56]. The correlations between CH4production rates and soil TP, TS, Mc, pH and NH4+–N(Figure 5) further confirmed that soil physical and chemical processes associated with penguin activities were the predominant factors affecting CH4production and its δ13C from tundra soils in coastal Antarctica.

Figure 4 Relationships between the δ13C values and CH4 concentration in the headspace under ambient air and N2 conditions during the ornithogenic soil incubations.

Figure 5 Correlation between CH4 production rates, δ13C-CH4 and P, S, pH, NH4+−N and soil moisture in the ornithogenic tundra soils.Note: these are the mean values at each soil sample, and the vertical bars indicate standard errors of CH4 production rates and δ13C-CH4.If the regression is significant at p＜0.05, regression lines and R values are shown in the figure. The arrow indicates outlier data, which are excluded from the correlation.

Although the experiment was short-term and conducted at a stable temperature (4ºC), our results showed that ornithogenic tundra soils in penguin colonies do have high potential fluxes of CH4, indicating the importance of the CH4emissions from these colonies in Antarctic terrestrial ecosystems. In maritime Antarctica and the sub-Antarctic, the number of marine animals and their colonies is large. It has been estimated that about 5 million Adelie penguins, 3 million Chinstrap penguins and more than half a million emperor penguins are distributed around the Antarctic coast[57].Penguins play an important role in the nutrient cycling of Antarctic terrestrial ecosystems by transferring C and N from the marine to the terrestrial environment[7,9]. Penguin guano and fertile ornithogenic soils are the most important organic C and N reservoirs in the Antarctic terrestrial ecosystems[12,53].At present, data on CH4emissions from these colonies are still scarce. Much more work is needed in this region to characterize the CH4production and emissions from ornithogenic tundra soils in coastal Antarctica. The technique of stable carbon isotopes for CH4has been proved to be a useful tool in studying the processes of CH4production[16−18].The relative contribution of acetate to CH4production and the fraction of CH4oxidized can be quantitatively estimated[19,26-27]based on the measurements of δ13C in CH4, CO2and acetate, and of the isotope fractionation factors. However,investigations of the isotope fractionation factors for CH4production and emission from tundra soils have not been conducted in coastal Antarctica, and need to be further studied in the future.

4 Conclusions

The results are summarized as follows:

(1) The mean CH4production rates are highly variable in the ornithogenic soil profiles, ranging from 0.7 to 20.3 μg∙kg−1∙h−1. Our results showed that ornithogenic tundra soils had high potential production rates of CH4based upon experimental incubation under ambient air and under N2,indicating the importance of potential CH4emissions from ornithogenic tundra soils in Antarctic terrestrial ecosystems.

(2) The ornithogenic soil profiles MI, GI and PB had higher δ13C-CH4under N2conditions than under ambient air incubation. The average δ13C-CH4emitted from MI, GI and PB, respectively, was 7.54‰, 3.19‰ and 9.21‰ higher under anaerobic conditions than under aerobic conditions. Our results indicated that anaerobic conditions were conducive to the emissions of CH4enriched in13C, and acetic acid reduction under anaerobic conditions might be a predominant source for CH4production in ornithogenic soils.

(3) Overall, δ13C-CH4emitted from ornithogenic tundra soils showed a significant negative correlation with soil CH4production rates under N2incubation (R2=0.41, p＜0.01) or under ambient air incubation (R2=0.50, p＜0.01). The CH4emissions from the ornithogenic soils had an important effect on carbon isotopes of CH4in Antarctic tundra.

(4) Potential CH4production from ornithogenic soils showed a significant positive correlation with TP and NH4+–N contents, pH and Mc, but a strong negative correlation with soil total sulfur (TS) contents. On the contrary, the δ13CCH4showed a significant negative correlation with TP and NH4+–N contents, pH and Mc, indicating that the amount of penguin guano deposition increased potential CH4production rates from tundra soils, but decreased the δ13C-CH4.

1 Sun L G, Zhu R B, Xie Z Q, et al. Monitoring CH4concentrations on Fildes Peninsula in Antarctica. Prog Nat Sci, 2001, 11(9): 995–998 (in Chinese)

2 Wang Y S, Wang M X, Zheng X H, et al. Atmospheric CH4concentration and change in Beijing. Chin Sci Bull, 1994, 39(14):1306–1308 (in Chinese)

3 Etheridge D M, Pearman G I, Fraser P J. Changes in tropospheric methane between 1841 and 1978 from a high accumulation-rate Antarctic ice core. Tellus B, 1992, 44(4): 282–294

4 Chappellaz J, Barnola J M, Raynaud D, et al. Ice-core record of atmospheric methane over past 160,000 years. Nature, 1990,345(6271): 127–131

5 Christensen T R, Johansson T, Åkerman H J, et al. Thawing sub-arctic permafrost: effects on vegetation and methane emissions. Geophys Res Lett, 2004, 31(4): L04501

6 Kaufman D S, Schneider D P, McKay N P, et al. Recent warming reverses long-term Arctic cooling. Science, 2009, 325(5945): 1236–1239

7 Tatur A, Myrcha A. Ornithogenic ecosystems in the maritime Antarctic—formation, development and disintegration//Beyer L,Bölter M. Geoecology of Antarctic ice-free coastal landscapes. Berlin Heidelberg: Springer, 2002, 154: 161–184

8 Sun L G, Zhu R B, Yin X B, et al. A geochemical method for the reconstruction of the occupation history of a penguin colony in the maritime Antarctic. Polar Biol, 2004, 27(11): 670–678

9 Cannone N, Wagner D, Hubberten H W, et al. Biotic and abiotic factors influencing soil properties across a latitudinal gradient in Victoria Land, Antarctica. Geoderma, 2008, 144(1–2): 50–65

10 Ugolini F C. Antarctic soils and their ecology//Holdgate M W.Antarctic ecology, vol 2. London: Academic Press, 1970: 673–692

11 Michel R F M, Schaefer C E G R, Dias L E, et al. Ornithogenic Gelisols (Cryosols) from maritime Antarctica: Pedogenesis, vegetation and carbon studies. Soil Sci Soc Am J, 2006, 70(4): 1370–1376

12 Schaefer C E G R, Simas F N B, Gilkes R J, et al. Micromorphology and microchemistry of selected Cryosols from maritime Antarctica.Geoderma, 2008, 144(1–2): 104–115

13 Zhu R B, Sun J J, Liu Y S, et al. Potential ammonia emissions from penguin guano, ornithogenic soils and seal colony soils in coastal Antarctica: effects of freezing-thawing cycles and selected environmental variables. Antarct Sci, 2011, 23(1): 78–92

14 Zhu R B, Liu Y S, Xu H, et al. Methane emissions from three sea animal colonies in the maritime Antarctic. Atmos Environ, 2008,42(6): 1197–1205

15 Zhu R B, Liu Y S, Xu H, et al. Nitrous oxide emissions from sea animal colonies in the maritime Antarctic. Geophys Res Lett, 2008,35(9): L09807

16 Zhang G B, Ji Y, Ma J, et al. Intermittent irrigation changes production, oxidation, and emission of CH4in paddy fields determined with stable carbon isotope technique. Soil Biol Biochem, 2012, 52:108–116

17 Popp T J, Chanton J P, Whiting G J, et al. Methane stable isotope distribution at a Carex dominated fen in north central Alberta. Global Biogeochem Cycles, 1999, 13(4): 1063–1077

18 Conrad R, Klose M. Effect of potassium phosphate fertilization on production and emission of methane and its13C-stable isotope composition in rice microcosms. Soil Biol Biochem, 2005, 37(11):2099–2108

19 Bilek R S, Tyler S C, Sass R L, et al. Differences in CH4oxidation and pathways of production between rice cultivars deduced from measurements of CH4flux and δ13C of CH4and CO2. Global Biogeochem Cycles, 1999, 13(4): 1029–1044

20 Stendel M, Romanovsky V E, Christensen J H, et al. Using dynamical downscaling to close the gap between global change scenarios and local permafrost dynamics. Global Planet Change, 2007, 56(1–2):203–214

21 Wuebbles D J, Hayhoe K. Atmospheric methane and global change.Earth-Sci Rev, 2002, 57(3–4): 177–210

22 Wang R S, Huang G H, Liang Z B, et al. Advances in the research on sources and sinks of CH4and CH4oxidation (uptake) in soil. Chin J Appl Ecol, 2002, 13(12): 1707–1712 (in Chinese)

23 Stevens C M, Rust F E. The carbon isotopic composition of atmospheric methane. J Geophys Res, 1982, 87(C7): 4879–4882

24 Liu Y S. Temporal and spatial variations of greenhouse gas ぼuxes and their production mechanism in Antarctic Tundra. Ph. D. thesis, Hefei:University of Science and Technology of China, 2011 (in Chinese)

25 Kato T, Hirota M, Tang Y H, et al. Spatial variability of CH4and N2O fluxes in alpine ecosystems on the Qinghai-Tibetan Plateau. Atmos Environ, 2011, 45(31): 5632–5639

26 Krüger M, Eller G, Conrad R, et al. Seasonal variation in pathways of CH4production and in CH4oxidation in rice fields determined by stable carbon isotopes and specific inhibitors. Global Change Biol,2002, 8(3): 265–280

27 Tyler S C, Brailsford G W, Yagi K, et al. Seasonal variations in methane flux and δ13CH4values for rice paddies in Japan and their implications. Global Biogeochem Cycles, 1994, 8(1): 1–12

28 Cheng X L, Luo Y Q, Xu Q, et al. Seasonal variation in CH4emission and its13C-isotopic signature from Spartina alterniぼora and Scirpus mariqueter soils in an estuarine wetland. Plant Soil, 2010, 327(1):85–94

29 Lowe D C, Brenninkmeijer C A M, Brailsford G W, et al.Concentration and13C records of atmospheric methane in New Zealand and Antarctica: evidence for changes in methane sources. J Geophys Res, 1994, 99(D8): 16913–16925

30 Kelley C A, Dise N B, Martens C S. Temporal variations in the stable carbon isotopic composition of methane emitted from Minnesota peatlands. Global Biogeochem Cycles, 1992, 6(3): 263–269

31 Bellisario L M, Bubier J L, Moore T R, et al. Controls on CH4emissions from a northern peatland. Global Biogeochem Cycles,1999, 13(1): 81–91

32 Trivelpiece W Z, Trivelpiece S G, Volkman N J. Ecological segregation of Adelie, gentoo and chinstrap penguins at King George Island, Antarctica. Ecology, 1987, 68(2): 351–361

33 Sun L G, Liu X D, Yin X B, et al. A 1, 500-year record of Antarctic seal populations in response to climate change. Polar Biol, 2004,27(8): 495–501

34 Zhu R B, Liu Y S, Ma E D, et al. Greenhouse gas emissions from penguin guanos and ornithogenic soils in coastal Antarctica: effects of freezing–thawing cycles. Atmos Environ, 2009, 43(14): 2336–2347

35 Tilsner J, Wrage N, Lauf J, et al. Emission of gaseous nitrogen oxides from an extensively managed grassland in NE Bavaria, Germany. II.Stable isotope natural abundance of N2O. Biogeochemistry, 2003,63(3): 249–267

36 Sun L G, Zhu R B, Xie Z Q, et al. Emissions of nitrous oxide and methane from Antarctic tundra: role of penguin dropping deposition.Atmos Environ, 2002, 36(31): 4977–4982

37 Cao Y C, Sun G Q, Han Y, et al. Determination of nitrogen, carbon and oxygen stable isotope ratios in N2O, CH4, and CO2at natural abundance levels by mass spectrometer. Acta Pedolog Sin, 2008,45(2): 249–258 (in Chinese)

38 Zhu R B, Chen Q Q, Ding W, et al. Impact of seabird activity on nitrous oxide and methane ぼuxes from high Arctic tundra in Svalbard,Norway. J Geophys Res, 2012, 117(G4): G04015

39 Cai Z C, Xu H, Ma J. Methane and nitrous oxide emissions from ricebased ecosystems. Hefei: University of Science and Technology of China Press, 2009 (in Chinese)

40 Jones S K, Rees R M, Skiba U M, et al. Greenhouse gas emissions from a managed grassland. Global Planet Change, 2005, 47(2–4):201–211

41 Keeney D R, Nelson D W. Nitrogen-inorganic forms//Page A L,Miller R H, Keeney DR. Methods of soil analysis. Madison, WI,USA: American Society of Agronomy, 1982: 643–698

42 Harris J M, Tibbles B J. Factors affecting bacterial productivity in soils on isolated inland nunataks in continental Antarctica. Microb Ecol, 1997, 33(2): 106–123

43 Cocks M P, Newton I P, Stock W D. Bird effects on organic processes in soils from five microhabitats on a nunatak with and without breeding snow petrels in Dronning Maud Land, Antarctica. Polar Biol, 1998, 20(2): 112–120

44 Qu X, He P J, Laurent M, et al. Effect of temperature on methanogenic pathway during household waste anaerobic digestion by stable carbon isotopic signature of CH4. Environ Sci, 2008, 29(11): 3252–3257 (in Chinese)

45 Blair N E, Carter W D Jr. The carbon isotope biogeochemistry of acetate from a methanogenic marine sediment. Geochim Cosmochim Acta, 1992, 56(3): 1247–1258

46 Burke R A. Stable isotope ratios of microbial methane from shallow aquatic sediments. Abstracts of Papers of the American Chemical Society, 1988, 195: 108

47 Burke R A Jr, Martens C S, Sackett W M. Seasonal variations of D/H and13C/12C ratios of microbial methane in surface sediments. Nature,1988, 332(6167): 829–831

48 Barker J F, Fritz P. Carbon isotope fractionation during microbial methane oxidation. Nature, 1981, 293(5830): 289–291

49 Quay P D, King S L, Stutsman J, et al. Carbon isotopic composition of atmospheric CH4: fossil and biomass burning source strengths.Global Biogeochem Cycles, 1991, 5(1): 25–47

50 Sun L G, Xie Z Q, Zhao J L. Lacustrine deposit on Ardley Island:Penguin omithogenic sediment identification. Chin J Polar Res, 2000,12(2): 105–112 (in Chinese)

51 Sherlock R R, Sommer S G, Khan R Z, et al. Ammonia, methane and nitrous oxide emission from pig slurry applied to a pasture in New Zealand. J Environ Qual, 2002, 31(5): 1491–1501

52 Holter P. Methane emissions from Danish cattle dung pats in the field.Soil Biol Biochem, 1997, 29(1): 31–37

53 Simas F N B, Schaefer C E G R, Melo V F, et al. Ornithogenic Cryosols from maritime Antarctica: phosphatization as a soil forming process. Geoderma, 2007, 138(3–4): 191–203

54 Alam M S, Jia Z J. Inhibition of methane oxidation by nitrogenous fertilizers in a paddy soil. Front Microbiol, 2012, 3: 246

55 Merino A P, Pérez-Batallón, Macías F. Responses of soil organic matter and greenhouse gas fluxes to soil management and land use changes in a humid temperate region of southern Europe. Soil Biol Biochem, 2004, 36(6): 917–925

56 Zhu R B, Liu Y S, Xu H, et al. Marine animals significantly increase tundra N2O and CH4emissions in maritime Antarctica. J Geophys Res, 2013, 118(4): 1773–1792

57 Knox G A. The biology of the southern ocean. New York: Cambridge University Press, 1994