Plant and Soil

, Volume 327, Issue 1, pp 85–94

Seasonal variation in CH4 emission and its 13C-isotopic signature from Spartina alterniflora and Scirpus mariqueter soils in an estuarine wetland

Authors

    • Coastal Ecosystems Research Station of the Yangtze River Estuary, Ministry Education Key Laboratory for Biodiversity Science and Ecological Engineering, Institute of Biodiversity ScienceFudan University
    • Wuhan Botanical GardenThe Chinese Academy of Sciences
    • Department of Botany and MicrobiologyUniversity of Oklahoma
  • Yiqi Luo
    • Coastal Ecosystems Research Station of the Yangtze River Estuary, Ministry Education Key Laboratory for Biodiversity Science and Ecological Engineering, Institute of Biodiversity ScienceFudan University
    • Department of Botany and MicrobiologyUniversity of Oklahoma
  • Qing Xu
    • Institute of Forest Ecology, Environment and ProtectionChinese Academy of Forestry
  • Guanghui Lin
    • Key Laboratory of the Ministry of Education for Coastal and Wetland Ecosystems, and School of Life SciencesXiamen University
  • Quanfa Zhang
    • Wuhan Botanical GardenThe Chinese Academy of Sciences
  • Jiakuai Chen
    • Coastal Ecosystems Research Station of the Yangtze River Estuary, Ministry Education Key Laboratory for Biodiversity Science and Ecological Engineering, Institute of Biodiversity ScienceFudan University
    • Coastal Ecosystems Research Station of the Yangtze River Estuary, Ministry Education Key Laboratory for Biodiversity Science and Ecological Engineering, Institute of Biodiversity ScienceFudan University
Regular Article

DOI: 10.1007/s11104-009-0033-y

Cite this article as:
Cheng, X., Luo, Y., Xu, Q. et al. Plant Soil (2010) 327: 85. doi:10.1007/s11104-009-0033-y

Abstract

Although invasions by non-native species represent a major threat to biodiversity and ecosystem functioning, little attention has been paid to the potential impacts of these invasions on methane (CH4) emission and its 13C-CH4-isotope signature in salt marshes. An invasive perennial C4 grass Spartina alterniflora has spread rapidly along the east coast of China since its introduction from North America in 1979. Since its intentional introduction to the Jiuduansha Island in the Yangtze River estuary in 1997, S. alterniflora monocultures have become the dominant component of the Jiuduansha’s vegetation, where monocultures of the native plant Scirpus mariqueter (a C3 grass) used to dominate the vegetation for more than 30 years. We investigated seasonal variation in soil CH4 emission and its 13C-CH4-isotope signature from S. alterniflora and S. mariqueter marshes. The results obtained here show that S. alterniflora invasion increased soil CH4 emissions compared to native S. mariqueter, possibly resulting from great belowground biomass of S. alterniflora, which might have affected soil microenvironments and /or CH4 production pathways. CH4 emissions from soils in both marshes followed similar seasonal patterns in CH4 emissions that increased significantly from April to August and then decreased from August to October. CH4 emissions were positively correlated with soil temperature, but negatively correlated with soil moisture for both S. alterniflora and S. mariqueter soils (p < 0.05). The δ13C values of CH4 from S. alterniflora, and S. mariqueter soils ranged from -39.0‰ to -45.0‰, and -37.3‰ to -45.7‰, respectively, with the lowest δ13C values occurring in August in both marshes. Although the leaves, roots and soil organic matter of S. alterniflora had significantly higher δ13C values than those of S. mariqueter, S. alterniflora invasion did not significantly change the 13C- isotopic signature of soil emitted CH4 (p > 0.05). Generally, the CH4 emissions from both invasive S. alterniflora and native S. mariqueter soils in the salt marshes of Jiuduansha Island were very low (0.01–0.26 mg m-2 h-1), suggesting that S. alterniflora invasion along the east coast of China may not be a significant potential source of atmospheric CH4.

Keywords

CH4 emissionStable carbon isotopeSoil propertiesPlant invasionCoastal sediments

Introduction

Methane is the second most important greenhouse gas and its global warming potential is 25 times greater than CO2 on a mass basis (IPCC 2007; Dalal et al. 2008). Natural wetlands play an important role in contributing methane (CH4) to the atmosphere (Bartlett and Harriss 1993; Edwards et al. 2000). The CH4 produced from wetlands accounts for approximately 20% of the total atmospheric budget (Rodhe 1990; IPCC 2007). Previous studies have reported that CH4 emissions from wetlands result from the complex interactions between the processes of CH4 production, consumption, and transport, which are governed by various interrelated environmental factors, such as temperature, vegetation, soil properties, and microbial activities (e.g., Conrad 1999; Kutzbach et al. 2004; Laine et al. 2007).

Vegetation is one of the important factors that affect CH4 emissions from wetland ecosystems (e.g., Kutzbach et al. 2004; Minkkinen and Laine 2006). While an increasing number of studies has reported that vegetation affects gas emissions through regulating production and transport in wetland ecosystems worldwide (e.g., Van der Nat and Middelburg 2000; Cheng et al. 2007; Wilson et al. 2009), several studies have indicated that belowground parts of vegetation (decaying plant materials and fresh root exudates) could provide the substrates for methanogenesis (Whiting and Chanton 1993; Joabsson et al. 1999), which promote CH4 production (Whiting and Chanton 1993; Van der Nat and Middelburg 2000). Soil temperature can also change decomposition of soil organic matter through regulating soil respiration rates and microbial activities (Rustad et al. 2001; Laine et al. 2007), and can eventually regulate the amount and seasonal pattern of CH4 emission in wetlands (Flessa et al. 2008). Additionally, soil moisture influences CH4 fluxes directly by controlling the relative extent of oxic and anoxic environments within soils, and indirectly by affecting soil genesis and vegetation composition, which are important additional factors controlling CH4 fluxes (Kutzbach et al. 2004). Thus, the changes in the relative importance of these processes, for example, species composition caused by plant invasions induced by human activities, may change soil organic matter input and quality (Ehrenfeld et al. 2001; Ehrenfeld 2003), which can potentially lead to large changes in the magnitude of CH4 flux from wetland ecosystems.

Furthermore, numerous studies have reported that isotopic fractionation occurs during CH4 emissions from the submerged anoxic soil into the atmosphere (Chanton et al. 1997; Bilek et al. 1999; Fey et al. 2005; Venkiteswaran and Schiff 2005). Stable carbon isotope ratios (δ13C) of CH4 have proven to be a useful technique for partitioning individual atmospheric CH4 sources and sinks (Blair 1998; Edward et al. 2000), which is based on the fact that the CH4 production is almost exclusively derived from acetate fermentation and CO2 reduction (H2/CO2 pathway) in the anoxic wetland environments (Conrad 2005; Penning et al. 2005). Production of CH4 from the CO2 reduction pathway exhibits larger isotope fractionation than that from acetate fermentation because the 13C-CH4 produced by acetate fermentation is more enriched (-30 ~ -60‰) than that produced by CO2 reduction (-60 ~ -110‰) (Fey et al. 2005; Conrad 2005). Thus, this difference in carbon isotope fractionation can be used for partitioning C-flux sources in methanogenic environments (Conrad 2005). However, Penning et al. (2005) have demonstrated that the δ13C values of CH4 produced from either CO2 reduction or acetate fermentation have a characteristic signature, and the magnitude of fractionation is usually considerably different among natural ecosystems. Meanwhile, the pathway of CH4 production from either CO2 or acetate can vary seasonally under field conditions, but the causes of the variations are not fully understood (Fey et al. 2005).

Spartina alterniflora was transplanted into tidal marshes of the coastal zone in China in 1979 from its native range in the United States because of its great ability to rapidly colonize mudflats (Qin and Zhong 1992; Wang et al. 2006a). While its rapid growth has greatly helped to stabilize the tidal flats, the negative impacts have also been characterized by the displacement of native species (Chen et al. 2004) and the changes of biotic communities and ecosystem processes (Wang et al. 2006a; Liao et al. 2007, 2008; Li et al. 2009). For example, Jiuduansha is an estuarine island growing from constant deposition of sediments carried from the Yangtze River in China. Invasive S. alterniflora, a C4 plant, was introduced to the island in 1997 under the Green Recovery for Birds project in the Yangtze River estuary, where the native Scirpus mariqueter (a C3 plant) had dominated the island’s vegetation for over 30 years (Cheng et al. 2006). Previous studies have reported that S. alterniflora grows faster than the native species (Wang et al. 2006b) and its invasion has significantly changed soil organic matter (SOM) (Cheng et al. 2006), resulting in greater residue inputs to coastal marshes. Cheng et al. (2007) have compared trace gas emissions from S. alterniflora with those from a native Phragmites australis by establishing brackish marsh mesocosms to experimentally assess the effects of plant species, flooding status, and clipping on trace gas emissions. It is well known that the CH4 emissions via plant pathway contribute significantly to the total CH4 emission from wetland ecosystems (e.g., Schimel 1995; Van der Nat and Middelburg 2000; Chanton et al. 2002). Little effort, however, has been made to assess the impact of invasive plants on CH4 emissions and its 13C-isotopic signature from the soil associated with soil substrates in natural marsh ecosystems. We hypothesized that C4S. alterniflora invasion in the Jiudansha estuarine wetland significantly alters CH4 emissions and the 13C-isotopic signature because S. alterniflora differs from the native C3 plant in productivity, tissue chemistry, and may affect soil properties such as soil temperature and moisture differently. The aims of this study were to: (1) reveal the seasonal variations of CH4 emissions and the δ13C-CH4 signatures from the S. alterniflora, and S. mariqueter soils; (2) assess the effects of plant traits and environmental variables (soil temperature, soil moisture, and soil substrate) on CH4 emissions and its 13C-CH4-isotope signatures.

Materials and methods

Study area

This study was conducted in Shanghai Jiuduansha Wetland Nature Reserve in the Yangtze River estuary (31o03'–31o17'N, and 121 o46'–122 o15'E). Jiuduansha Island is an alluvial island that was formed from the sediments from the Yangtze River. The Island has an area of 425 km2 in 2003 and still continues to grow at about 70 m in radius per year (Chen 2003). Its climate is characterized by annual precipitation of 1,145 mm and annual mean temperature of 15.7°C, with monthly means of 27.3°C for July and 4.2°C for January (Chen 2003). Jiuduansha has developed as a stable island over a half century. The vegetation on the island consists of very few species, and the structure of the plant communities is relatively simple. The native species, S. mariqueter, dominated the salt marshes on this island until the invasive S. alterniflora was introduced in 1997 (Chen 2003).

Field sample collection

In April 2004, we selected two transects, 3 km long, along a transitional zone from S. alterniflora to S. mariqueter in the wetland (details given in Liao et al. 2007). Because S. alterniflora and S. mariqueter do not co-exist, both of the two species form their respective monocultures. In most of Jiuduansha wetland, S. alterniflora had replaced S. mariqueter over about 8 years since its intentional introduction in 1997, while S. mariqueter had dominated the salt marshes for over 30 years (Chen 2003, Cheng et al. 2006). We randomly selected six sampling sites on each transect, each of which measured 50 m wide and 200 m long. At each site, we collected several samples (up to 6) of litter, roots, and soil. One replicate (litter, roots, and soil) was taken from several randomly-selected separate samples at each site (Cheng et al. 2008). Soil samples were separated into two parts; one part was sealed in a soil tin and analyzed for gravimetric water content (percentage water, measured as g water/g dry soil × 100), the other was placed in a glass vial for isotopic analysis of soil organic carbon. All measurements were made four times from April to October (i.e. bimonthly) in 2004. We collected gas samples at the same time of day between 8:00 AM and 10:00 AM on each sampling occasion to minimize any effects of diurnal variation in emissions. We used the static closed chamber technique to measure CH4 emissions (Cheng et al. 2007). Briefly, we first inserted stainless-steel collars into the soil to a depth of 50 mm one day prior to each gas sampling. Every care was taken to minimize disturbance to the soil, particularly inside the chamber, during insertion. The small transparent acrylic resin chambers covering an area of 10 × 10 cm with a height of 10 cm were placed on the notched collars, and an airtight closure was ensured by water sealing during the measurements. Air inside the chambers was circulated with battery-driven fans during the measurement in order to ensure gas samples were well-mixed. The gas fluxes of CH4 in all plots were measured simultaneously. Generally, five gas samples of chamber air were taken using 25 ml polypropylene syringes at 0 min, 10 min, 20 min, 30 min and 40 min after enclosure. In order to equilibrate air pressure inside chamber, the same amount of air (25 ml) was pulled into the chamber through polypropylene syringes immediately after taking sample at 0 min, 10 min, 20 min and 30 min. Samples were injected into 10 ml pre-evacuated vials for laboratory analysis. The air temperature inside the chamber and soil temperature in the top soil layer (0–20 cm depth) was taken during each gas sample.

Laboratory analysis

The CH4 concentrations in the gas samples were analyzed by using gas chromatography (GC/FID Shimadzu 14 B), with a unibead-C column. Column, injection, and FID temperature were set at 100°C, 120°C, and 300°C, respectively, with a carrier gas flow rate of 65 ml min-1, and an injection volume of 0.5 ml (Towprayoon et al. 2005; Cheng et al. 2007). The CH4 fluxes were calculated by linear regression analysis of the change of gas concentration in the mixed chamber with time over a 40-minute period (Rolston 1986; Cheng et al. 2007).

Samples of roots, litter, and soil were dried at 50°C to constant weight and ground to pass through 20-mesh (0.85m m) sieves (Lin et al. 1999; Cheng et al. 2006). Subsamples of roots, litter, and soil were measured on an isotope ratio mass spectrometer (Thermo Finnigen, Delta-Plus, Flash, EA, 1112 Series, USA) for carbon isotopes. Urea and glycine were analyzed as a check on the accuracy and precision of isotopic ratios by the elemental composition analyzer. Precision for δ13C was ± 0.15 ‰.

Stable carbon isotope ratio (δ13C) in gas samples collected at initial (0), and final (40) minutes after enclosure was measured using a gas chromatograph combustion isotope ratio mass spectrometer (GCC-IRMS) system (Finnigan MAT model delta plus, Thermoquest, Bremen, Germany). The method requires only very small samples, namely several picolitres of the vapor (for principal operations see Brand (1996)). Briefly, the CH4 in the gas samples (10–400 µl) were first separated in a Hewlett Packard 6,890 gas chromatograph operating with a Pora Plot Q column (27.5 m length; 0.32 mm i.d.; 10 µm film thinkness; Chrompack, Frankfurt, Germany) at 25°C with He (99.996% purity; 2.6 ml min-1) as carrier gas. After conversion of CH4 to CO2 in the Finnigan Standard GC Combustion Interface III, the gases were transferred into the IRMS (Fey et al. 2005; Conrad and Claus 2005; Penning et al. 2005). The working standards were CO2 (99.998% purity; Messer-Griessheim, Düsseldorf, Germany) and methylstearate (Merck). The latter was intercalibrated at the Max-Planck-Institut für Biogeochemie, Jena, Germany (courtesy of Dr. W. Brand) against NBS22 (National Bureau of Standards 22) and expressed in the delta notation versus PDB (Pee Dee Belemnite) carbonate:
$$ \delta^13 {\text{C }} = 10^3 \left( {{{R_{\text{sa}} } \mathord{\left/ {\vphantom {{R_{\text{sa}} } {R_{\text{st}} }}} \right. } {R_{\text{st}} }} - 1} \right) $$
where R = 13C/12C of sample (sa) and standard (st), respectively. The precision of repeated analysis was ± 0.5‰ when 1.3 nmol CH4 was injected. The δ13C values from emitted CH4 were calculated by a mass balance model using both initial and final concentration and isotopic data (Chanton and Liptay 2000; Powelson et al. 2007).

Statistical analyses

The data analyses were performed using Stat Soft’s Statistica software for Windows (Version 6.0, StatSoft, Inc. 2001) and Microsoft Excel software for analysis of variance and t-test with a significance level of P < 0.05.

The δ13C values of plant matter and CH4, CH4 flux rates, and soil temperature and moisture reported here are the means of six measurements. Two-way ANOVA was performed to examine the differences in soil temperature, soil moisture, CH4 flux rates, the δ13C values of CH4 between the two marshes in relation to time. One-way ANOVA was further employed to examine the variation in soil temperature, soil moisture, CH4 flux rates, the δ13C values of CH4 with time within each marsh. The t-test was used to examine the differences in δ13C values of plant materials and soil organic matter between the two marshes.

Results

Organic matter, soil temperature and soil moisture

The plant matter of S. alterniflora, and S. mariqueter had the mean δ13C values of -13.0 ± 0.4 ‰, and - 27.4 ± 0.7 ‰, respectively (Table 1). Based on the δ13C values, vegetation in the S. alterniflora marsh consisted exclusively of C4 plant species while vegetation in the S. mariqueter marsh consisted exclusively of C3 plant species. The SOM in S. alterniflora, and S. mariqueter marshes had the mean δ13C values of - 23.5 ± 0.3 ‰, and - 24.3 ± 0.2 ‰, respectively (Table 1). Basically, the δ13C values of plant matter and SOM did not change significantly from April to October within each marsh (Table 1, p > 0.05).
Table 1

The values of δ13C (‰) of SOM, litter and roots in S. alterniflora and S. mariqueter marshes in Jiuduansha wetland in Yangtze River estuary, China

Marsh type

Month

SOM

Litter

Root

S. alterniflora

Apr

-23.4 ± 0.2a

-12.7 ± 0.3a

-13.2 ± 0.2a

Jun

-23.7 ± 0.2a

-13.5 ± 0.3a

-12.5 ± 0.3a

Aug

-23.8 ± 0.1a

-13.3 ± 0.2a

-12.2 ± 0.3a

Oct

-23.1 ± 0.2a

-13.4 ± 0.4a

-12.8 ± 0.3a

S. mariqueter

Apr

-24.1 ± 0.1b

-28.7 ± 0.3b

-26.6 ± 0.2b

Jun

-24.3 ± 0.2b

-27.5 ± 0.4b

-27.3 ± 0.3b

Aug

-24.6 ± 0.2b

-27.3 ± 0.3b

-26.8 ± 0.2b

Oct

-24.0 ± 0.1b

-27.9 ± 0.3b

-26.7 ± 0.2b

Values are means (n = 6) with standard errors. Different suffixes indicate significant differences between two marshes (t-test, p < 0.05)

Significant seasonal changes in soil temperature and moisture were recorded from April to October (Fig. 1), but soil temperature was not different between the two marshes during the same time period (Fig. 1a; Table 2; p > 0.05). In contrast, soil moisture in both marshes decreased from April to August and increased from August to October with greater soil moisture in S. alterniflora marsh compared to S. mariqueter marsh (Fig. 1b; Table 2; p < 0.05).
https://static-content.springer.com/image/art%3A10.1007%2Fs11104-009-0033-y/MediaObjects/11104_2009_33_Fig1_HTML.gif
Fig. 1

Seasonal variation of soil temperature and moisture in S. alterniflora and S. mariqueter marshes in Jiuduansha wetland in Yangtze River estuary, China. Error bars represent SE (standard error) of the means (n = 6)

Table 2

The p-values of two-way ANOVA for testing the effects of marsh type and sampling time on CH4 flux rates, δ13C of CH4, soil temperature, and soil moisture between S. alterniflora and S. mariqueter marshes in Jiuduansha wetland in Yangtze River estuary, China

Source of variation

CH4 Flux

13C-CH4

Soil T

Soil M

Plant community (PC)

0.037

0.761

0.49

0.043

Sampling month (SM)

<0.0005

<0.005

<0.005

<0.0005

PC × SM

0.049

0.813

0.45

0.091

Soil T: soil temperature; Soil M: soil moisture

$$ {\text{CH}}_4 \,{\text{flux rates and the}}^{13} {\text{C}} - {\text{CH}}_4 - {\text{isotopic signature}} $$
The CH4 flux rates from the two marshes followed similar seasonal patterns, and varied significantly from April to October, with the highest CH4 flux rate being observed in August and the lowest CH4 flux rate in October (Fig. 2a). The flux rates of CH4 from the S. alterniflora and S. mariqueter marshes ranged from 0.02 to 0.26 mg m-2 h-1, and from 0.01 to 0.23 mg m-2 h-1, respectively (Fig. 2a). The flux rate of CH4 emission from the S. alterniflora marsh was generally higher than that from the S. mariqueter marsh (Fig. 2a; Table 2; p < 0.05). Overall, CH4 emissions from S. alterniflora and S. mariqueter soils were positively correlated with soil temperature, but negatively correlated with soil moisture (p < 0.05) (Table 3).
https://static-content.springer.com/image/art%3A10.1007%2Fs11104-009-0033-y/MediaObjects/11104_2009_33_Fig2_HTML.gif
Fig. 2

Seasonal variation of CH4 emission rates and its 13C-isotopic signatures of S. alterniflora and S. mariqueter marshes in Jiuduansha wetland in Yangtze River estuary, China. Error bars represent SE (standard error) of the means (n = 6)

Table 3

Summary of regression analyses between CH4 emissions and soil properties of S. alterniflora and S. mariqueter marshes in Jiuduansha wetland in Yangtze River estuary, China

CH4 Emission

Species

Independent Variable

Regression Equation

R2 value

p value

 

S. alterniflora

Soil T

Y = 0.0088x - 0.08

0.33

0.049

 

Soil M

Y = -0.0142x + 0.29

0.48

0.013

S. mariqueter

Soil T

Y = 0.029x - 0.66

0.80

0.0001

 

Soil M

Y = -0.013x + 0.56

0.42

0.022

Soil T: soil temperature; Soil M: soil moisture

The δ13C values of CH4 from S. alterniflora, and S. mariqueter soils ranged from -39.0‰ to -45.0‰, and from -37.3‰ to -45.7‰, respectively, with the lowest δ13C values occurring in August for both marshes (Fig. 2b). Although the δ13C values of CH4 from S. alterniflora soil were higher in April and lower in October than those from S. mariqueter soil (Fig. 2b), two-way ANOVA showed no significant seasonal difference in the δ13C value of CH4 between the two marshes (Fig. 2b; p > 0.05).

Discussion

This study compared the effects of an invasive species (S. alterniflora) and a native species (S. mariqueter) on CH4 emissions and its 13C-CH4-isotopic signature in marshlands in the Yangtze River estuary, China. Our results obtained here showed that CH4 emission was greater from the S. alterniflora soil than that from the S. mariqueter soil (Fig. 2a). Numerous studies have reported that vegetation can foster CH4 production in anoxic wetland environments by contributing soil organic matter and/or acting as the pathway to allow CH4 to escape to the atmosphere without passing through the aerobic layer of sediments (e.g., Fey et al. 2005; Kutzbach et al. 2004). Our previous study has demonstrated that the invasive species S. alterniflora has greatly increased root biomass in the Jiuduansha estuarine wetland (Liao et al. 2007) and hence increased labile soil organic matter (Cheng et al. 2008) and the pathway to emit CH4 from soil (Cheng et al. 2007), which may result in a higher CH4 emission production from the S. alterniflora soil. Moreover, being consistent with the previous observations that the CH4 emission from salt marshes (e.g. Lovley and Phillips 1987; Roden and Wetzwel 1996; Liikanen et al. 2002) is relatively low compared with that from other wetlands like freshwater wetlands, and rice fields (e.g. Ding et al. 2003; Towprayoon et al. 2005), this study shows that the flux rates of CH4 in salt marsh ecosystems were quite low, ranging from 0.01 to 0.26 mg m-2 h-1 (Fig. 2a). There are two possible explanations for the very low CH4 emission rates from salt marsh soils. One explanation is that the low emission rates from salt marsh soils in this study could be partly due to omission of CH4 produced through the plant aerenchyma tissues (e.g., Ding et al. 2003; Cheng et al. 2007). Another explanation is that the low CH4 emissions in the salt marshes might reflect the fact that the greater amounts of HS- (or S2-) and Fe3- in the salt marsh soil limit CH4 production (e.g., Kostka et al. 2002; Neubauer et al. 2005) compared to other freshwater wetlands. These HS- (or S2-) and Fe3- reducers can outcompete methanogens for common substrates, thus, CH4 emissions in coastal wetlands are generally lower than those in freshwater ones (e.g., Kostka et al. 2002; Neubauer et al. 2005).

Although the δ13C value of S. alterniflora plant materials was significantly different from that of S. mariqueter (Table 1), the δ13C-CH4 signature did not differ significantly between S. alterniflora and S. mariqueter soils (Fig. 2b). Since short-term S. alterniflora invasion only contributed less than 10% to the soil labile organic C in the top 100 cm soil layer and caused slightly higher δ13C value of SOM compared with S. mariqueter (Cheng et al. 2008), we tentatively concluded that most of the emitted CH4 might have been produced from soil organic carbon rather than from root exudates and residues. Meanwhile, in salt marsh sediments containing rich labile organic carbon (Cheng et al. 2008), CH4 production occurs predominantly by the acetate fermentation pathway (Blair 1998; Edward et al. 2000), which is also the case in this study, reflected by δ13C-CH4 values. Generally, soils containing an abundance of labile organic carbon follow the acetate fermentation pathway while soils with more recalcitrant organic matter follow the CO2 reduction pathway (Edward et al. 2000). We found that the values of δ13C-CH4 were lower in August compared to those in other months within each marsh (Fig. 2b). The lower value of δ13C-CH4 is probably ascribed to reduced labile substrate for methanogenic bacteria (e.g., Kutzbach et al. 2004; Conrad and Claus 2005). Additionally, the δ13C values of surface emitted CH4 is regulated by CH4 oxidation, which tends to enrich 13C in residual CH4 (Krüger et al. 2002). The lowest δ13C values corresponded to the highest CH4 flux in August (Fig.2a vs. b), indicating that CH4 production was greater than CH4 oxidation.

Previous studies have reported that CH4 emissions from wetlands are affected greatly by both soil temperature and moisture (e.g., Christensen 1993; Moosavi and Crill 1998; Kutzbach et al. 2004; Laine et al. 2007). For example, several studies have indicated that CH4 production has a strong temperature response with reported Q10 values of 2.7–20.5 (e.g. Dunfield et al. 1993; Moosavi and Crill 1998). Our results showed that CH4 emissions increased from April to August and decreased from August to October (Fig.2a), and were positively correlated with temperature (Table 3; P < 0.05). This result is consistent with those of other studies in which CH4 emissions are correlated with soil temperature in wetland ecosystems (e.g., Christensen 1993; Nykǎnen et al. 1998). Kutzbach et al. (2004) have reported that the ratios between CH4 production and CH4 oxidation are controlled directly by soil moisture which regulates the relative extent of oxic and anoxic environments within soils. Our results showed that the CH4 emissions were negatively correlated with soil moisture in both S. alterniflora and S. mariqueter soils (Table 3; P < 0.05). Christensen et al. (2001) have indicated that the impact of soil water on CH4 emissions from wetlands depends on the water table status which can be envisaged as an on–off switch. When the water table falls below the soil surface, low soil moisture can drastically increase microbial CH4 oxidation, which reduces CH4 emissions (Kutzbach et al. 2004).

In summary, we found that S. alterniflora invasion on Jiuduansha Island increased CH4 emissions compared to the native S. mariqueter marsh. We assume that an increase in belowground biomass of S. alterniflora marsh changes soil microenvironments and/or the CH4 production pathway. However, S. alterniflora invasion did not significantly change 13C-CH4-isotopic signature. CH4 emissions from soils in both of two marshes followed similar seasonal patterns, and increased significantly from April to August and decreased from August to October. CH4 emissions from S. alterniflora and S. mariqueter soils were positively correlated with soil temperature (p < 0.05), but negatively correlated with soil moisture (P < 0.05). Overall, the CH4 emission from both S. alterniflora and S. mariqueter soils are relatively low compared to other types of wetlands, which is consistent with our previous results from the mesocosm study with S. alterniflora (Cheng et al. 2007). Although S. alterniflora invasion altered C and N dynamics in the salt marsh (Cheng et al. 2006, 2008; Liao et al. 2007), the results of this study, together with those of our early study (Cheng et al. 2007), tentatively suggest S. alterniflora invasion along the east coast of China may not be a significant potential source of atmospheric CH4. Nevertheless, we should acknowledge a limited number of samples in this study that may not sufficiently address the possible S. alterniflora invasion effects on CH4 emissions. More field studies and related modeling work are still needed to quantify the effects of S. alterniflora invasion on potential CH4 emissions and its feedbacks to the future climate change on larger scales.

Acknowledgements

This study was financially supported by National Basic Research Program of China (No. 2006CB403305), National Science Foundation of China (No. 30670330), and Ministry of Education of China (No. 20050246013). We thank Dr. Weixin Ding, Xiaoping Li and Leyi Li for lab analysis, and Chenghuan Wang, Zhichen Wang, Yongjian Gu, Haiqiang Guo and Jing Xie for their assistance with field sampling

Copyright information

© Springer Science+Business Media B.V. 2009