Aerobic Methane Emission from Plants in the Inner Mongolia Steppe

Nov 28, 2007 - derived directly from plant tissues under aerobic conditions. Because woody species are relatively rare in the Inner Mongolia steppe, a...
0 downloads 0 Views 270KB Size
Download PDF
Environ. Sci. Technol. 2008, 42, 62–68

Aerobic Methane Emission from Plants in the Inner Mongolia Steppe Z H I - P I N G W A N G , * ,†,‡ X I N G - G U O H A N , † G. GEOFF WANG,‡ YANG SONG,† AND J A Y G U L L E D G E §,| State Key Laboratory of Vegetation and Environmental Change, Institute of Botany, Chinese Academy of Sciences, Nanxincun 20, Xiangshan, Beijing 100093, China, Department of Forestry and Natural Resources, Clemson University, South Carolina 29634, Pew Center on Global Climate Change, 2101 Wilson Boulevard, Arlington, Virginia 22201, and Department of Zoology and Physiology, University of Wyoming, Laramie, Wyoming 82071

Received May 23, 2007. Revised manuscript received September 19, 2007. Accepted October 22, 2007.

Traditionally, methane (CH4) emission from terrestrial plants is thought to originate from belowground microbial metabolism under anaerobic conditions, with subsequent transport to the atmosphere through stems. However, a recent study reported aerobic CH4 emission from plants by an unrecognized process, a result that has since been questioned. We investigated CH4 emissions under aerobic conditions from aboveground tissues of 44 species indigenous to the temperate Inner Mongolia steppe. Ten herbaceous hydrophytes (wetlandadapted plants) were examined, two of whichsGlyceria spiculosa and Scirpus yagarasemitted CH4 from stems but not from detached leaves. Of 34 xerophytes (arid-adapted plants) examined, 7 out of 9 shrub species emitted CH4 from detached leaves but not stems, whereas none of 25 herbaceous xerophytes emitted CH4. The herbaceous hydrophyte, S. yagara, emitted highly 13C-depleted CH4, suggesting a microbial origin. Achillea frigida exhibited the highest CH4 emission rates among the shrubs and continuously emitted relatively 13Cenriched CH4 from detached leaves, indicating that CH4 was derived directly from plant tissues under aerobic conditions. Because woody species are relatively rare in the Inner Mongolia steppe, aerobic, plant-derived CH4 emission is probably negligible in this region. Our results may imply a larger role for aerobic CH4 production in upland ecosystems dominated by woody species or in ecosystems where woody encroachment is occurring as a result of global change.

Introduction Methane (CH4) is the second most important anthropogenic greenhouse gas in the atmosphere after carbon dioxide. The major sources and sinks of the global CH4 cycle are thought to have been identified, although they remain quantitatively uncertain (1). However, a recent report that plants produce and emit CH4 under aerobic conditions by an unknown mechanism raised the possibility that this process might be * Corresponding author phone: +86-10-62836547; fax: +86-1082591781; e-mail: [email protected], [email protected]. † Chinese Academy of Sciences. ‡ Clemson University. § Pew Center on Global Climate Change. | University of Wyoming. 62

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 1, 2008

a major unidentified source in the global budget (2). Keppler et al. (2) initially estimated aerobic CH4 emission from plants to be in the range of 62–236 Tg CH4 y-1, about 10–40% of the total annual source. Based on a variety of constraints, subsequent analyses produced estimates in the range of 10–69 Tg CH4 y-1 (1, 3, 4), lowering the potential source strength to 2–12% of the total. Such a contribution is smaller than the uncertainty surrounding larger CH4 sources, such as wetlands. If confirmed, however, it is large enough to refine the global CH4 budget by reducing uncertainties on larger terms (1). Beyond the initially large estimate of global aerobic CH4 emission from plants, the fundamental process of aerobic, plant-derived CH4 emission has been questioned because no chemical mechanism has been identified to account for the process (1, 5). Most (∼70%) CH4 from natural sources originates with strictly anaerobic microbial metabolism and no aerobic biochemical pathway is known for the production of CH4, nor is any eukaryotic organism known to produce CH4 other than in symbiosis with anaerobic methanogenic bacteria (6). Further, a recent, independent attempt to replicate the original result was unsuccessful. Dueck et al. (7) found no evidence for substantial aerobic CH4 emission in six species of hydroponically grown herbaceous plants using a 13C-labeling approach that should have been adept at detecting plant-derived CH4 emission. However, the original study (2) used well-established analytical methods for detecting CH4 and for identifying its sources based on 13C natural abundance signatures (2, 8). The study also did a credible job of ruling out enzymatic CH4 production, demonstrating a simple exponential response of CH4 production to increasing temperature from 30 to 70 °C in sterilized leaves without any sign of enzyme denaturation at higher temperatures. Some mysterious field observations could be explained by aerobic CH4 emission from plants. Two decades ago, anomalously high boundary-layer CH4 concentrations were detected in tropical savannahs during the wet season, when net consumption of atmospheric CH4 by savannah soils is maximal and known sources cannot account for the boundary layer accumulation (9). Clipping aboveground grass biomass decreased CH4 emission from savannah plots (10), an action that normally increases CH4 emission from grasses that transport CH4 from the soil to the atmosphere (11). Considering all factors, it is difficult to dismiss Keppler’s results. More independent studies are needed to understand aerobic CH4 emission by plants and assess the role, if any, this novel process plays in global CH4 cycling. Wetland plants, especially grasses and sedges, commonly transport microbially produced CH4 from wetlands soils to the atmosphere (11, 12). It is possible, however, that they also emit plant-derived, aerobically produced CH4 that is masked by soil-derived CH4. If so, the sources of wetland CH4 efflux may need to be reassessed. In general, upland soils are sinks for atmospheric CH4, but in situ observations are limited and the sink strength remains uncertain by a factor of 2. In the temperate Inner Mongolia steppe climate region, cool, dry conditions create a sharp contrast between small, low-lying wetlands populated by water-loving plants (hydrophytes), and expansive uplands populated by plants adapted to semiarid conditions (xerophytes) (13). The upland soils of the Inner Mongolia steppe consume CH4 as expected (14), but estimates of net ecosystem CH4 consumption extrapolated from chamber measurements may be anomalously high if xerophytes in this system emit plant-derived CH4 under aerobic conditions. Even if chamber measurements capture the correct net consumption rate, any aerobic 10.1021/es071224l CCC: $40.75

 2008 American Chemical Society

Published on Web 11/28/2007

CH4 production that may partially offset gross consumption has not been incorporated into the CH4 budget. Understanding the various sources and sinks is especially important from the standpoint of predicting consequences of ecosystem change for the CH4 cycle. Grasslands comprise approximately 40% of the world’s terrestrial surface, representing an important component of the global carbon cycle (15). The Inner Mongolia steppe constitutes an important component of the Eurasian temperate grasslands; the Xilin River basin represents a typical grassland-dominated area in this region (13). The Xilin basin harbors approximately 629 plant species from 291 genera and 74 families (16). To investigate CH4 emission by plants, we randomly sampled plant species around the Inner Mongolia Grassland Ecosystem Research Station, Chinese Academy of Sciences, located in the Xilin River basin. This study examined CH4 emission in closed-chamber laboratory incubations from fresh, aboveground biomass of 44 plant speciess10 hydrophytes and 34 xerophytessindigenous to the region. We investigated two main questions: (1) Which plants in the Inner Mongolia steppe, if any, produce plantderived CH4 under aerobic conditions? and (2) Are there differences in the sources of emitted CH4 among plant taxa or functional groups, such as hydrophytes and xerophytes or shrubs and herbs?

Materials and Methods Sampling and Site Description. Plants were sampled randomly in the grassland-dominated area around the Inner Mongolia Grassland Ecosystem Research Station (ca. 43°38′ N, 116°42′ E; 1187 m above sea level), operated by the Chinese Academy of Sciences. The climate is semiarid, temperate, and continental, with a mean annual temperature of ∼0.6 °C. The coldest monthly mean temperature is -21.4 °C in January, and the warmest is 18.5 °C in July. The mean annual precipitation is about 350 mm, with a rainy season between mid-June and mid-September. Approximately 10% of precipitation falls as snow during the winter. The growing season extends from late April to early October. Detailed descriptions of the Xilin River basin have been published elsewhere (13). A total of 44 plant species from this area were sampled randomly for CH4 emission. Ten hydrophytic species were sampled from low-lying areas and 34 xerophytic species were sampled from upland areas. Random sampling selected a variety of herbs, and shrubs, and C3 and C4 species were included (Table 1). All plants were sampled and transported to the laboratory intact with roots, stems, and leaves. Laboratory Incubation. CH4 emissions from fresh plants were examined in laboratory incubations during the summer of 2006. Fresh plants were sampled randomly during the early morning (6:00–7:00 a.m. local time) preceding each measurement event and transported whole (intact roots, stems, and leaves) to the laboratory immediately after sampling. Fresh, whole plants were washed in deionized water and air-dried for about 0.5 h. Plants were separated into stems and leaves just prior to analysis and either sealed immediately in gastight serum bottles or allowed to vent for 10 min prior to being sealed in serum bottles. A few grams of leaves or stems were placed into a 120-ml serum bottle; roots were always excluded from measurements to avoid soil and anaerobic microbiological contamination. Each serum bottle contained leaves or stems from a unique individual, providing triplicate replication for each species. The bottle was sealed immediately with a butyl rubber stopper (diameter 20 mm) and flushed with CH4-free air (200 mL min-1 for 3 min) from a compressed gas cylinder using “inlet–outlet” needles inserted through the stopper. Laboratory incubations were carried out for approximately 10–20 h in the dark at ambient temperature (20–22 °C). Most incubations were conducted in an initially CH4-free atmo-

sphere. Additional incubations were conducted with initial CH4 concentrations of 1.69 and 7.98 µL L-1. The initial CH4 concentration was achieved by flushing the headspace with air containing the desired CH4 concentration, and was measured immediately after purging to confirm the initial concentration. CH4 Flux Measurement. CH4 concentrations were analyzed at various time intervals using gas chromatography (GC). A 3-mL gas sample was withdrawn by syringe and immediately replaced by 3 mL of CH4-free air to maintain headspace pressure. Parallel blanks were used to account for dilution or leakage effects. The GC was a Hewlett-Packard 5890 Series II equipped with an injection loop, a flameionization detector operated at 200 °C and a 2-m stainless steel column packed with 13 XMS (60/80 mesh). The column oven temperature was 55 °C, and the carrier gas was N2 flowing at 30 mL min-1. A certified CH4 standard with a concentration of 7.98 µL L-1 (China National Research Center for Certified Reference Materials, Beijing) was used for calibration. The CH4 emission rate was calculated using the initial linear change of CH4 concentrations in the bottle at R2 g 0.9. At the end of each incubation, biomass was determined as oven-dried weight at 60 °C for 48 h. Discerning the Source of Emitted CH4. Three species with significant CH4 emission rates, including two hydrophytes (Glyceria spiculosa and Scirpus yagara) and one woody xerophyte (Achillea frigida), were studied further to examine the source of emitted CH4. The strategy for discerning the CH4 source included the following three elements: (1) Determine the anatomical source(s) of CH4 emission (stems vs leaves). If the source were microbial CH4 produced in the soil, most of the CH4 should be stored preferentially in the transport tissues of the stems. (2) Examine the dynamics of CH4 emission in whole and cut stems/leaves. If microbial CH4 is simply stored in transport tissues, physically cutting the tissues should hasten elimination of CH4 from the tissues. If, however, the tissues themselves are producing CH4 under aerobic conditions, both intact and cut tissues should emit CH4 continuously. (3) Measure and compare the δ13C of source tissues (stems or leaves) and the emitted CH4. By comparing the δ13C of each plant species to its emitted CH4, and the δ13C of CH4 emitted from different plant species, it might be possible to infer whether the CH4 originated from plant tissues or from microbial CH4. Detached stems and leaves were incubated separately as described above to determine the anatomical source(s) of CH4 emission. In parallel incubations, stems and leaves were cut into small sections to examine whether cutting would eliminate CH4 emission by rupturing storage voids. To measure 13C natural abundances in plant tissues and emitted CH4, fresh plants were washed to remove adhering soil particles using deionized water and air-dried for about 0.5 h. The plants were placed in 120-ml serum bottles and sealed with crimp caps containing silicon septa. Each bottle was purged immediately with CH4-free air for 5 min (200 mL min-1) and then incubated for 48 h in the dark at ambient temperature (20–22 °C). The bottles were then refrigerated (4 °C) for 2 weeks prior to δ13C measurement. Air samples were withdrawn from the headspace by syringes and the plants were oven-dried at 60 °C to a constant weight. The dried bulk biomass was ground to a fine powder (mesh number 100) using a mortar and pestle. The δ13C was determined using a mass spectrometer (Thermo Finnigan DELTAplusXP, Germany). The δ13C in bulk biomass was measured following combustion in a Thermo Finnigan PreCon into CO2 at 1000 °C. CO2 was purified using a vacuum line with cryogenic traps. The δ13C in emitted CH4 was measured following a CH4 prepurification using cryogenic traps followed by combustion to CO2. All δ13C values were reported relative to the international PDB standard. VOL. 42, NO. 1, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

63

TABLE 1. Emission Rates of CH4 from Plants (ng CH4 gdw-1 h-1)a ecotype

morphotype

photosynthesis

species

CH4 emission

family

hydrophyte

herb

C3

xerophyte

shrub

C3

herb

C4 C3

Glyceria spiculosa Scirpus yagara Agrostis gigantea Polygonum sicboldi meisn Polygonum amphibium Carex rostrata Carex dahurica Carex humida Juncus bufonius Juncus gracillimus Ulmus macracarpa Caragana microphylln Achillea frigida Achillea Intramongolica Achillea gmelinii Ceraloides arborescens Betula ovalifolia Salix gordejevii Kochia proslrata Melissitus ruthenica Taraxacum mongolicum Serratula centauroides Leymus chinensis Stipa grandis Achnatherum sibiricum Koeleria cristata Agropyron cristatum Poa attenuata Poa subfastigiata Bromus inermis Polygonum divaricatum Potenlilla bifurca Potenlilla tanacetifolia Potenlilla acaulis Thalictrum pclaloidcum Thalictrum supradecomposilum Allium ramosum Hemerocallis minor Allium bidentatum Salsola collina Agriophyllum laevis Agropyron cristatum Setaria viridis Achnatherum splendens

13.50 6.78,b 11.05 ND ND ND ND ND ND ND ND 0.55 1.40 3.09,b 3.39 0.85 1.22 2.29 ND ND 0.48 ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND

Gramineae Cyperaceae Gramineae Polygonaceae Polygonaceae Cyperaceae Cyperaceae Cyperaceae Juncaceae Juncaceae Ulmaceae Leguminosae Compositae Compositae Compositae Chenopodiaceae Betulaceae Salicaceae Chenopodiaceae Leguminosae Compositae Compositae Gramineae Gramineae Gramineae Gramineae Gramineae Gramineae Gramineae Gramineae Polygonaceae Rosaceae Rosaceae Rosaceae Ranunculaceae Ranunculaceae Liliaceae Liliaceae Liliaceae Chenopodiaceae Chenopodiaceae Gramineae Gramineae Gramineae

C4

a Fresh plants (intact leaves and stems) were incubated in the dark at room temperature (20–25 °C) for 20 h with initially CH4-free air. Data are averages of three replicates. ND indicates that no CH4 emission was detected or was too weak to quantify from linear change in CH4 concentration within 20 h. Unless stated otherwise, fresh plants were sampled during early July. b Measurement was repeated with fresh plants sampled during late August 2006.

Statistical Analysis. Statistical analysis was performed using the SAS program (17). Duncan’s multiple comparison tests were used to compare the variation in CH4 emission rates and the δ13C among groups at P < 0.05.

Results CH4 Emission from Fresh Plants. Detached stems and leaves of 44 plant species belonging to various functional groups (hydrophytes, xerophytes, herbs, shrubs, C3, and C4) were tested for CH4 emission (Table 1). Nine species emitted CH4, with rates ranging from 0.48 to 13.50 ng CH4 gdw (gram dry weight) -1 h-1 in the first half of July 2006. About 80% of the species tested emitted no detectable CH4. The proportion of species emitting CH4 was about the same (i.e., ∼20%) for most of the groups examined, including hydrophytes (emitting/total ) 2:10), xerophytes (7:34), C3 plants (8:38), and C4 plants (1:6). However, 78% of shrubs emitted CH4, whereas only 6% of herbs emitted CH4 (Table 1). All of the shrubs were xerophytes, but none of the more abundant xerophytic herbs emitted CH4. The highest CH4 emission rates (6.78–13.5 64

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 1, 2008

ng CH4 gdw-1 h-1) were in the two hydrophytes (G. spiculosa and S. yagara). Among the 7 xerophytes that emitted CH4, A. frigida exhibited the highest emission rates (3.09–3.39 ng CH4 gdw-1 h-1). CH4 Emission Dynamics and Anatomic Location. CH4 emissions from A. frigida (xerophytic shrub) and G. spiculosa and S. yagara (hydrophytic herbs) were examined in greater detail. A. frigida leaves emitted CH4 at a constant rate throughout the ∼10-h incubation, whereas A. frigida stems emitted CH4 at much lower rates (Figure 1a). When A. frigida leaves were cut into 2-mm segments and vented for 10 min prior to incubation, they still emitted CH4 continuously for 10 h, but at significantly higher rates than uncut leaves (Figure 1a). G. spiculosa emitted CH4 at high rates from stems but did not emit CH4 from detached leaves (Figure 1b). Whole G. spiculosa stems emitted CH4 as a pulse early in the incubation and then ceased to emit CH4 (Figure 1b), whereas whole S. Yagara stems (combined stem and sheath structure) emitted CH4 continuously during the incubation (Figure 1c). However, when G. spiculosa and S. yagara stems were cut

FIGURE 2. Natural-abundance carbon isotope ratios (δ13C) of plants and their emitted CH4. Different letters represent significant differences in δ13C between plant tissues and their own emitted CH4 (P < 0.05). Asterisk (*) represents significant δ13C difference in the CH4 emitted from A. frigida leaves and S. yagara stems (P < 0.05). Values are means ( 1 standard error (n ) 3). Fresh plants were sampled and incubations began on July 25.

FIGURE 1. Anatomical source of CH4 emission. (a) Achillea frigida: An average of 3.1 g of uncut and 3.8 g of cut leaves and 4.2 g of uncut and 5.1 g of cut stems. (b) Glyceria spiculosa: An average of 1.2 g of uncut and 1.7 g of cut leaves and 2.4 g of uncut and 2.7 g of cut stems. (c) Scirpus yagara: An average of 2.0 g of uncut and 2.6 g of cut stems. Samples were purged with CH4-free air, except for S. yagara samples, which were incubated with an initial background of ambient atmospheric CH4 (1.69 µL L-1). Plants were sampled and incubations began on the same day (July 17 for A. frigida and G. spiculosa; July 23 for S. yagara). Values are means ( 1 standard error (n ) 3). into 1-cm segments and allowed to vent for 10 min prior to incubation, neither of them emitted CH4 during the incubation (Figure 1b and c). 13C Natural-Abundance Carbon Isotope Ratios of Plants and Emitted CH4. Because both A. frigida and S. yagara emitted CH4 continuously from uncut tissues, we compared 13C natural abundance of their tissues and emitted CH (Figure 4 2). The δ13C values of A. frigida leaves and S. yagara stems were nearly identical at just under -27‰. The CH4 emitted from both species was significantly depleted in 13C as compared to the tissues from which it was emitted. However, the 13C abundance of CH4 emitted from the two species differed dramatically, with CH4 from S. yagara (-77.3‰), much more depleted than CH4 from A. frigida (-39.8‰). Effect of Ambient CH4 Concentration on CH4 Emission from A. frigida. To ensure that CH4 emission in A. frigida would still proceed in the presence of normal background CH4 concentrations, we tested the effect of different initial CH4 concentrations on the rate of CH4 emission from A. frigida leaves (Figure 3). There was no difference in CH4 emission rates with initial concentrations of 0 and 1.69 µL CH4 L-1. With an initial concentration of 7.98 µL CH4 L-1, emission

FIGURE 3. CH4 emission from A. frigida leaves incubated with various initial CH4 concentrations. An average (n ) 3) of 3.3, 2.9, and 3.0 g A. frigida leaves were incubated in initially 0 (CH4-free), 1.69, and 7.98 µL CH4 L-1, respectively, in the dark at 20–22 °C. (a) Average CH4 concentrations at each time point for all treatments. Standard errors are omitted to enhance readability. (b) The overall emission rate (average slope ( 1 standard error, n ) 3) from A. frigida leaves at each of the initial CH4 concentrations. Measurements began on July 19. rates were slower, although emission continued throughout the incubation, even at this elevated background concentration.

Discussion CH4 Emission in Different Functional Groups. Without regard to the CH4 source, most functional groups, including hydrophytes, xerophytes, C3 plants, and C4 plants, had similar proportions of species that emitted CH4sabout 20% in each category (Table 1). The exceptions were the herbaceous and shrub morphotypes. Herbs were the largest group, with 35 species, yet only two herb speciessboth hydrophytessemitted CH4. The majority of the herbs were sampled from uplands and none of those emitted CH4. In contrast, 78% of the shrubs, VOL. 42, NO. 1, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

65

all of which were sampled from uplands, emitted CH4 from detached leaves in the absence of roots and soil, which may harbor anaerobic bacteria, and under aerobic conditions. We cannot rule out that intact herbaceous plants would have exhibited significant CH4 emission under these conditions, as Keppler et al. (2) found that intact plants emitted CH4 1-2 orders of magnitude faster than detached leaves of the same species. However, given that our detection limit was apparently comparable to that of Keppler et al. (2), in order to evade detection the emission rates from herbs in our study must have been significantly lower than those observed by Keppler et al. (2). Hence, our results suggest that shrubs have higher potential than herbs for CH4 emission under aerobic conditions in the Inner Mongolia steppe, although examination of intact plants of each species is required to confirm this hypothesis. If confirmed, this dichotomy between woody and herbaceous plants in this ecosystem may offer a new direction of inquiry for seeking a mechanism for aerobic CH4 production in plants. CH4 Source. Only two herbs, both hydrophytes, emitted CH4. G. spiculosa and S. yagara apparently emitted CH4 only as a result of storing microbial CH4 in their stems prior to harvest, as detached leaves did not emit CH4 and CH4 emission was eliminated from stems by cutting them 10 min prior to incubation (Figure 1b and c). This conclusion is also consistent with the pulsed emission pattern observed for uncut G. spiculosa stems (Figure 1b), likely a consequence of the rapid escape of CH4 accumulated in lacunae and aerenchyma of stems. In contrast, uncut S. yagara stems emitted CH4 continuously throughout the incubation (Figure 1c). We suspect this contrast resulted from anatomical differences, given that that G. spiculosa is a Gramineae with “continuum void” stems, whereas S. yagara is a Cyperaceae with “intercellular void” stems, the latter being less “leaky.” CH4 emission from A. frigida, the shrub with the highest CH4 emission rates (Table 1), was localized mostly in the leaves and was continuous throughout the incubation period, implying a persistent CH4 source (Figure 1a). Cutting the leaves into small segments prior to incubation elevated the rate of CH4 emission rather than eliminating it (Figure 1a), suggesting that CH4 was not released from storage voids. Physical injury often causes elevated respiration associated with stress response, so elevated CH4 emission may have been connected to increased catabolic activity. However, Keppler et al. (2) found that intact plants emitted CH4 at much higher rates than dissected plants, suggesting that injury inhibits the process rather than stimulating it. A previous study of Inner Mongolia grassland plants found that the Compositae species with dense pubescence and waxy layer, such as A. frigida, emit volatile organic compounds (18), suggesting a possible anabolic source of CH4. However, Keppler et al. (2) reported that CH4 emission was nonenzymatic, calling into question the involvement of any metabolic pathway. The much greater 13C-depletion of CH4 from S. yagara compared to CH4 from A. frigida (Figure 2) implies different CH4 sources in the two species. Keppler et al. (2) proposed that the CH4 produced aerobically by plants entails a chemical reaction involving methoxyl groups in plant structural compounds. The major plant C1 (one-carbon unit) pool, which includes methoxyl groups from pectin and lignin, has a unique carbon isotope signature exceptionally depleted in 13C (8). Pectin-methoxyl 13C depletion compared to biomass 13C depletion (∆13C) is in the range of 21–45‰. Further conversion to gaseous CH4 could increase depletion a bit more. CH4 emitted from anoxic wetland soils as a result of microbial metabolism is also highly depleted in 13C (8). The 13C depletion of CH from S. yagara was large enough to be 4 derived from either anaerobic microbial metabolism or 66

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 1, 2008

pectin-methoxyl. However, the cutting experiment with S. yagara stems (Figure 1c) implicates soil-derived CH4 stored in the stems prior to harvesting the plants. According to Keppler et al. (8), the ∆13C for lignin-methoxyl compared to biomass was in the range of 11–16‰ for a variety of plants. The ∆13C of CH4 emitted from A. frigida was 13.1‰, ruling out an anaerobic microbial CH4 source and implicating ligninderived methoxyl as the best candidate source. Together, the results of the cutting experiment and the δ13C signatures observed indicate that CH4 emitted from S. yagara stems very likely originated from microbial metabolism in the soil, whereas CH4 emitted from A. frigida leaves very likely derived directly from plant compounds under aerobic conditions. The other six shrubs that emitted CH4 in this study exhibit key similarities to A. frigida, including being woody xerophytes, emitting CH4 primarily from leaves rather than stems, and growing in semiarid upland soils that do not accumulate CH4 that could become trapped in gas transport or storage tissues. It seems likely therefore that these six species share a common CH4 source with A. frigida. Examining the carbon isotope signatures in more of these plants will be useful for testing this hypothesis. Comparison with Previous Results. Two previous studies with divergent results have been published (2, 7). Keppler et al. (2) observed significant CH4 emission from plants under aerobic conditions, whereas Dueck et al. (7) found no evidence for substantial aerobic CH4 emission by plants. Our results showed that woody plants (shrubs/semishrubs) from the grasslands of Inner Mongolia had significant CH4 emission under aerobic conditions, whereas herbaceous species from the same area did not. Hence, our study concurs with Keppler et al. (2) that at least some plants emit CH4 under aerobic conditions, and with Dueck et al. (7) that perhaps not all plants do. The range of CH4 emission rates we measured was nearly the same as the range reported by Keppler et al. (2) for detached, fresh leaves (typically in the range of 0.2-3 ng gdw-1 h-1 at 30 °C). For whole plants, Keppler et al. (2) observed CH4 emission rates that were 1 or 2 orders of magnitude higher than those in detached leaves. Dueck et al. (7) only examined intact plants, but observed CH4 emission rates similar to those we and Keppler et al. (2) observed for detached leaves. However, because of the nature of the detection system Dueck et al. (7) used, the emission rates they observed were not statistically different from zero, and those authors concluded that plants did not emit CH4 at all in their final experiment specifically designed to boost CH4 production above their system’s detection limit. As a result, Dueck et al. (7) suggested that the high emission rates Keppler et al. (2) observed in intact plants could have resulted from the diffusion of residual CH4 from plant tissues or soil pore space into the incubation chamber, which was initially flushed to achieve a CH4-free atmosphere. We demonstrated that stems of hydrophytes can contain stored microbial CH4 (Figure 1b and c). CH4 from intact plants in the study of Keppler et al. (2) was more depleted of 13C than CH4 emitted from A. frigida in our study, but less depleted than CH4 emitted from S. yagara. (cf. Figure 2 in ref (2)), leaving the isotopic source signature in Keppler’s study somewhat ambiguous, as acknowledged by those authors. To test whether our results with A. frigida could have resulted from using CH4-free air in incubations, we compared the rate of CH4 emission from A. frigida leaves under three different initial CH4 concentrations (Figure 3). Emission rates were indistinguishable in CH4-free air and in ambient atmospheric CH4 (1.69 µL L-1), demonstrating that emission continues with a natural background CH4 concentration. With a background concentration more than four times higher than atmospheric CH4, emission rates slowed but continued linearly for the duration of the incubation, and were clearly

discernible above the elevated background CH4 concentration. These results indicate that CH4 was not following a concentration gradient into the headspace of our incubation bottles, nor were there roots or soil in the bottles to contribute anaerobic microbial contamination. Moreover, the large difference between δ13C values for CH4 emitted from S. yagara and from A. Frigida demonstrates that the source of CH4 was different for these two species under identical incubation conditions. Hence, the CH4 flux was likely not an artifact of the incubation procedure. A possible explanation for the negative results of Dueck et al. (7) is that all plants they examined were herbaceous, as we observed no plant-derived CH4 emission in any of the 35 herbaceous species we examined. However, Keppler et al. (2) did observe CH4 emission from detached leaves of herbaceous plants, including three of the same species examined by Dueck et al. (7). Hence, a discrepancy remains for CH4 emission from intact plants in those two studies. We did not examine intact plants and Dueck et al. (7) did not examine detached leaves. Moreover, each of the three studies discussed here employed different incubation and CH4 detection methods, all of which seem valid but not easily compared. Because of the stark differences in methods and plant origins among these three studies, it is difficult at this time to reconcile those results that differ. However, our results have much in common with those of Keppler et al. (2). Our results from A. frigida, in particular, provide independent confirmation of the basic discovery that the leaves of at least some plants emit CH4 derived directly from plant tissues by a yet unknown aerobic process, although our results do not address the higher rates Keppler et al. (2) observed from intact plants. Implications for Regional CH4 Budgets. Aerobic CH4 emission from plants may have implications for the global CH4 budget, but its significance is expected to vary regionally, primarily as a result of temperature dependence (1, 2). For instance, aerobic CH4 emission from plants may account for unexplained atmospheric CH4 anomalies observed seasonally in the tropics (1, 2, 9, 19). Although our results cannot provide quantitative information about the CH4 budget, they may have implications for the character of CH4 emissions in the Inner Mongolia steppe, where the mean annual temperature is near 0 °C, implying a minor role for aerobic CH4 emission from plants in comparison to the tropics, but suggesting a possible seasonal CH4 signal during summer months. In fact, Bergamaschi et al. (20) observed significant CH4 emissions via satellite retrieval and independent inversion modeling during the warm summer months between 40 and 60 °N, with the strongest signal localized over east Asia. The inversion model identified the dominant source as wetlands, with enteric fermentation and fossil hydrocarbons explaining the balance (see Figure 5 in ref (20)). In discerning the wetland source strength in high latitudes, the inversion model emphasized temperature control. If aerobic CH4 emission from plants is also controlled by temperature, these two signals would be confounded seasonally, so that part of the signal identified as wetland CH4 could originate from aerobic CH4 emission from plants (cf ref (1)), implying a possibly significant role for aerobic CH4 emission from plants in the regional CH4 budget during the summer in northern temperate grasslands. The Inner Mongolia steppe is strongly dominated by grasses (i.e., herbs), such as Leymus chinensis, Stipa grandis, and Agropyron cristatum, which exhibited no CH4 emission in our study. By comparison, most of the shrubs we examined emitted CH4, yet cover a small proportion of the steppe ecosystem. Hence, our results suggest that aerobic CH4 emission from plants has little potential to contribute a strong seasonal signal of CH4 emission simply because the plants that carry out the process are rare across the landscape,

particularly when compared to the much stronger source strength of small wetlands in the same landscape (13). This conclusion might be reversed if intact samples of the dominant grasses from this landscape emit CH4 at substantial rates, which is possible given the higher rates Keppler et al. (2) observed for intact plants compared to detached leaves. The shrubs, however, are probably too rare to contribute significantly to the regional CH4 budget even with higher emission rates from intact plants. Firm conclusions on these points await additional work examining intact plants and a formal census of in situ CH4 sources in the region. The implied lack of potential for aerobic CH4 emission in the Inner Mongolia steppe may differ from the situation in tropical savannahs of Venezuela, where anomalously high CH4 concentrations have long been observed in the boundary layer during the wet season, when net soil CH4 consumption, rather than emission, is maximal (9). Removal of savannah grasses during the wet season decreased the rate of CH4 emission, consistent with a plant-derived CH4 source (9, 10). The Venezuelan savannah example suggests that grasses would dominate aerobic CH4 emission to the atmosphere, if it occurs. On the other hand, our results imply that woody plants would contribute more plant-derived CH4 than herbs, which could have important implications for other ecosystems, such as forests, or tundra ecosystems experiencing woody encroachment/expansion because of global change (e.g., ref (21)). Aerobic CH4 emission by plants may also be speciesdependent. Most of the woody species we examined emitted CH4, but two did not (Table 1). Among randomly selected plant species in Brazil, Keppler found that some species emitted CH4 up to 4000 times faster than others (see ref (5)). At present, only a few dozen plant species have been tested. Clearly, more studies are needed to determine how widespread this trait is, and the role of phenology remains to be addressed as well. Detailed understanding of how local CH4 budgets will respond to future changes may therefore require accounting for aerobic CH4 emission from plants, which could vary depending on community composition, climatic zone, and phenology.

Acknowledgments This research was supported by an innovative group research grant (30521002) from the National Natural Science foundation of China, the state key basic research development program of China (2007CB106800), and a general program (30670402) from the National Natural Science Foundation of China. We greatly thank the anonymous referees and editors for their constructive comments that improved the manuscript.

Literature Cited (1) Butenhoff, C. L.; Khalil, M. A. K. Global methane emissions from terrestrial plants. Environ. Sci. Technol. 2007, 41 (11), 4032– 4037. (2) Keppler, F.; Hamilton, J. T. G.; Braβ, M.; Röckmann, T. Methane emissions from terrestrial plants under aerobic conditions. Nature 2006, 439, 187–191. (3) Kirschbaum, M. U. F.; Bruhn, D.; Etheridge, D. M.; Evans, J. R.; Farquhar, G. D.; Gifford, R. M.; Paul, K. I.; Winters, A. J. A comment on the quantitative significance of aerobic methane release by plants. Funct. Plant Biol. 2006, 33, 521–530. (4) Parsons, A. J.; Newton, P. C. D.; Clark, H.; Kelliher, F. M. Scaling methane emissions from vegetation. Trends Ecol. Evol. 2006, 21, 423–424. (5) Schiermeier, Q. The methane mystery. Nature 2006, 442, 730– 731. (6) Conrad, R. Soil microorganisms as controllers of atmospheric trace gases (H2, CO, CH4, OCS, N2O, and NO). Microbiol. Rev. 1996, 60, 609–640, 1996.. (7) Dueck, T. A.; deVisser, R.; Poorter, H.; Persijn, S.; Gorissen, A.; deVisser, W.; Schapendonk, A.; Verhagen, J.; Snel, J.; Harren, F. J. M.; Ngai, A. K. Y.; Verstappen, F.; Bouwmeester, H.; VOL. 42, NO. 1, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

67

(8)

(9) (10) (11) (12) (13)

(14)

68

9

Voesenek, L. A. C. J. No evidence for substantial aerobic methane emission by terrestrial plants: a 13C-labelling approach. New Phytol. 2007, 175, 29–35. Keppler, F.; Kalin, R. M.; Harper, D. B.; McRoberts, W. C.; Hamilton, J. T. G. Carbon isotope anomaly in the major plant C1 pool and its global biogeochemical implications. Biogeoscience 2004, 1, 123–131. Crutzen, P. J.; Sanhueza, E.; Brenninkmeijer, C. A. M. Methane production from mixed tropical savanna and forest vegetation in Venezuela. Atmos. Chem. Phys. Disc. 2006, 6, 3093–3097. Sanhueza, E.; Donoso, L. Methane emission from tropical savanna Trachypogon sp. grasses. Atmos. Chem. Phys. 2006, 6, 5315–5319. Schimel, J. P. Plant transport and methane production as controls on methane flux from arctic wet meadow tundra. Biogeochemistry 1995, 28, 183–200. Ding, W. X.; Cai, Z. C.; Tsuruta, H. Plant species effects on methane emissions from freshwater marshes. Atmos. Environ. 2005, 39, 3199–3207. Wang, Z. P.; Han, X. G.; Li, L. H.; Chen, Q. S.; Duan, Y.; Cheng, W. Methane emission from small wetlands and implications for semiarid region budgets. J. Geophys. Res 2005, 110, D13304; doi:13310.11029/12004JD005548. Wang, Y.; Xue, M.; Zheng, X.; Ji, B.; Du, R.; Wang, Y. Effects of environmental factors on N2O emission from and CH4 uptake by the typical grasslands in the Inner Mongolia. Chemosphere 2005, 58, 205–215.

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 1, 2008

(15) LeCain, D. R.; Morgan, J. A.; Schuman, G. E.; Reeder, J. D.; Hart, R. H. Carbon exchange and species composition of grazed pastures and exclosures in the shortgrass steppe of CO. Agric. Ecosyst. Environ. 2002, 93, 421–435. (16) Liu S. R.; Liu, Z. L. Outline of flora of the Xilin River basin, Inner Mongolia. In Inner Mongolia Grassland Ecosystem Research Station. Research on Grassland Ecosystem (No. 3); Science Press: Beijing, China, 1988; pp 227–268. (17) SAS. SAS/STATk User’s Guide Release 8.0; Cary, NC, 1999. (18) He, N. P.; Han, X. G.; Pan, Q. M. Variations in the volatile organic compound emission potential of plant functional groups in the temperate grassland vegetation of Inner Mongolia, China. J. Integr. Plant Biol. 2005, 47, 13–19. (19) Frankenberg, C.; Meirink, J. F.; vanWeele, M.; Platt, U.; Wagner, T. Assessing methane emissions from global space-borne observations. Science 2005, 308, 1010–1014. (20) Bergamaschi, P.; Frankenberg, C.; Meirink, J. F.; Krol, M.; Dentener, F.; Wagner, T.; Platt, U.; Kaplan, J. O.; Korner, S.; Heimann, M.; Dlugokencky, E. J.; Goede, A. Satellite chartography of atmospheric methane from SCIAMACHY on board ENVISAT: 2. Evaluation based on inverse model simulations. J. Geophys. Res 2007, 112, D02304; doi:02310.01029/02006 JD007268. (21) Kullman, L. Rapid recent range-margin rise of tree and shrub species in the Swedish Scandes. J. Ecol. 2002, 90, 68–77.

ES071224L