B53A-0917
Methane release and surface sediment environment in the East Siberian Arctic shelf.
Arctic shallow seabed sediments have received little attention with regard to their contribution to the global carbon budget and, particularly, the marine methane (or CH4) budget, because of their small areal extent (Archer, 2006; Shakhova and Semiletov, 2007). Furthermore, the low temperatures that characterize these sediments are not considered conducive to methanogenesis, though psychrophilic methanogens may exist (Rivkina et al., 2006). In case of the Siberian Arctic shelf, shallow sediments have not been considered a methane source to the hydrosphere or to the atmosphere because submarine (offshore) permafrost acts as a impermeable lid, preventing methane escape. However, recent data from the east Siberian Arctic shelf (ESAS) during 2003-2005, showed extreme CH4 super-saturation of surface waters (up to 4400% saturation), implying that high air-to-sea fluxes occur at times (Shakhova et al., 2005, 2007ab). Understanding the factors generating the high concentrations of methane in the waters of East Siberian shelf requires additional and detail research efforts. Thus in summer of 2005 we investigated the methane in ESAS surface sediment and sea water at 127 locations. Methane concentrations were determined on site with GC-FID head-space measurements. Spatial correlation among methane anomalies and fault zones has been found, while it was no correlation with sediment organic carbon content and mineralogical composition.
B53A-0918
Relic offshore permafrost and methane release over the East-Siberian Arctic shelf
Importance of huge pool of old carbon stored within off-shore permafrost in Siberian region is determined by gradual mobilization of old carbon during permafrost degradation and its incorporation into modern carbon cycle in form of methane. Acceleration of this process due to both natural and anthropogenic disturbance of coastal environment may not only enhance a positive feedback to a global warming, but also can potentially cause rapid or even abrupt climate change on Earth. Theoretically, during times of marine transgression, the sub-sea permafrost could reduce in thickness and develops nearly isothermal conditions close to the melting point. This might occur early on after the marine transgression (high heat flow), or it might take up to several thousand years (low heat flow). Due to the time lag existing between the maximum heat flow and maximum permafrost transformation, the most drastic changes in thermal regime of permafrost might occur not at a warmest time, (for example, at Holocene optimum), but further. Moreover, the key changes in permafrost properties might not be reaching a phase transition stages, but more likely reaching permeability for gas on a larger scale. As sub-sea permafrost does not necessarily represent a rocklike ice-bonded layer, but is sometimes ice free under negative temperatures as its salinity increases, this allows permeability for upward migration of gases, stored within permafrost and/or beneath it. Our recent study in the East-Siberian Arctic shelf (2003-2006) detected CH4 super-saturation of surface water in some areas up to 10,000 % above background level, implying that strong air-to-sea fluxes must occur at times. It leads to significant increase in atmospheric concentrations of methane above the sea surface – up to 8 ppm (latitude specific monthly mean concentrations is 1.85 ppm). Our first wintertime data (April 2007) shows extremely high CH4 concentrations (up to 5.7 µmol l-1) in the surface water beneath the sea ice. Being commensurable with concentrations, measured during the wintertime in thermokarst lakes of Siberian Lowland, these values represent the highest CH4 concentrations, observed in the Arctic Ocean, and are comparable to those registered over decaying gas hydrate fields in the Sea of Okhotsk. The vertical distribution of dissolved CH4, as well as the size and number of CH4 bubbles, trapped within the sea ice, strongly indicate ebullition as a mechanism of CH4 transfer to the water surface. The areas, which should be considered as favorable for opening potentially perennial pathways for methane escape after over 6,000 year inundation, are following: large rivers taliks; a meandering river system (paleo- valleys), running across the shelf; so-called geological disjunctives (fault zones, tectonically and seismically active areas); sedimentary basins; completely submerged, or transformed into sea lagoons, thaw lake taliks.
B53A-0919
Methane Ebullition During Simulated Lake Expansion and Permafrost Degradation
Methane, a potent greenhouse gas, is emitted by Arctic tundra and lakes. Ebullition, or bubbling, of methane from Arctic lakes has been shown to be a major transport mechanism from the sediment to the atmosphere, and ebullition rates are greatest near the edges of the lakes where active erosion is occurring. In regions of continuous permafrost, Arctic lakes have been expanding in recent decades, attributed to permafrost melting and development of thermokarst. Lake expansion occurs when the margins erode into water, supplying large amounts of organic rich material to the sediment-water interface. This allows carbon that was previously stored in the soil (active layer and permafrost) to become bioavailable and subject to decomposition. An increase in Arctic methane emissions as a result of permafrost thawing and lake expansion would constitute a positive feedback to Arctic warming. In order to better understand these processes, an experiment was initiated in July 2007 at the Barrow Environmental Observatory, Barrow, AK. Different layers of locally collected tundra soil were placed into incubation chambers at the bottom of a shallow (about 1 m deep) lake. Each experimental chamber consists of a bucket fixed underneath an inverted funnel, with a sampling port on top to capture and collect the emitted gases. Gas samples are analyzed for methane and carbon dioxide concentrations, as well as relevant isotopic compositions. Gas sampling has occurred at frequent intervals during the late summer and will continue through the early winter. Three replicates of each layer (active layer, seasonally frozen active layer and permafrost) were incubated, as well as an empty control chamber. An additional chamber containing thawed permafrost and cellulose-rich sawdust was placed for comparison, as cellulose is a major component of plant tissue and the fermentation of the cellulose should yield substrates for methanogenesis. Total production of methane versus organic carbon content of initial sample, kinetics of ebullition, and relative potential emissions from each tundra layer will be assessed.
B53A-0920
Reducing uncertainty in emission estimates of methane bubbling from arctic lakes
A recent first-order estimate suggests that arctic lakes are significant emitters of methane (24 ± 10 Tg CH4 yr-1) contributing as much as ~6% of global atmospheric CH4 sources annually. Emissions from arctic lakes are projected to increase as permafrost thaws in the Arctic, releasing tens of thousands of teragrams to the atmosphere in the form of bubbles. Emissions are thought to be particularly high from lakes influenced by permafrost degradation, a process that discharges labile organic matter to anaerobic lake bottoms, fueling biological methane production and emissions. Complete thaw of permafrost beneath lakes may destabilize deeper methane sources such as hydrate methane and natural gas, providing pathways for the release of sub- permafrost and/or intra-permafrost methane from biogenic or thermogenic sources. Alternatively, emission of deep methane to the atmosphere may occur independently from permafrost warming dynamics, for instance through geological fault activity. Despite the potential for large release of deep methane pools to the atmosphere, little is known about the occurrence, extent, and vulnerability of these methane sources. Here we present results of our efforts as part of the IPY to quantify, map, and project biological and geological methane emissions from arctic lakes in Alaska and Siberia in conjunction with permafrost degradation. This effort includes pioneering new methods of measuring methane bubbling (dominant mode of emissions) from lakes using field studies, geophysical measurements, isotope geochemistry, remote sensing Synthetic Aperture Radar analysis and the establishment of a Pan-Arctic Lake-Ice Methane Monitoring Network (PALIMMN).
B53A-0921
Patterns in methane emissions from arctic tundra: a comparison of chamber measurements over six growing seasons
Wet tundra ecosystems are well-known to be a significant source for atmospheric methane. With the predicted stronger effect of global climate change on arctic terrestrial ecosystems compared to lower-latitude ecosystems, there is a special obligation to study the natural diversity and the range of possible feedback effects on global climate that could arise from Arctic tundra ecosystems. Here we compare direct measurements of CH4 emissions from high arctic valley at Zackenberg, NE Greenland, studied by chamber techniques during summer seasons 1997, 1999-2000, and 2005-2007. The measurements were carried out at the same wetland site, nevertheless hydrology and plant composition of the site were found significantly changed between 2000 and 2005. First three years the measurements were provided by manual chamber technique using FID-GC as methane analyzer, in 2005 a permanent automatic chamber system with laser off-axis integrated-cavity output spectroscopy analyzer was installed instead. The study years had different climate patterns (temperature, water regime), and showed differences in seasonal CH4 fluxes. In this presentation we will propose an effort to obtain a detailed analysis of interannual differences in temperature and water regimes with resulting differences in methane emissions.
B53A-0922
Net Productivity and Reduced Carbon Flux at Different Timescales at a Subarctic Mire
While several environmental changes that appear linked to increased temperature and permafrost degradation have been noted in the panarctic, the concern is to what extent positive feedback mechanisms that may accelerate climate change are associated with these environmental and physical changes. Labile organic carbon, currently protected in frozen soils, may mobilize if soils are warmed and thawed and then transferred to the atmosphere as greenhouse gases. Detailed information about carbon cycling in perennially frozen carbon rich soils needs to be incorporated into coupled climate models. In this study, we account for the productivity and reduced carbon flux at the subarctic Stordalen Mire in northern Sweden. Methane (CH4) and total hydrocarbon (THC; CH4 and nonmethane hydrocarbons) fluxes were measured at three vegetation communities with different moisture and permafrost regimes; i) a dry palsa site underlain with permafrost, ii) a Sphagnum site with highly variable water table and iii) a wet Eriophorum site. The two latter sites are CH4 emitting environments. A long-term data record of five sequential growing seasons (2003-2007) acquired with an automatic chamber system allows for analyses at least three levels of temporal variability (diurnal, seasonal and annual). There was a clear relation between seasonal accumulated THC emissions and productivity (measured as NEE of CO2), when all vegetation communities were compared in the same analysis. At the Sphagnum site, ~2 gC m-2 was emitted as THC at the same time as ~40 gC m-2 was taken up as CO2, (days 173- 235, year 2003-2006). In comparison, a highly productive Eriophorum sp. community emitted ~9 gC m-2 (THC) while productivity was estimated to ~90 gC m-2 (CO2). Productivity differences between sites can be a measure of the THC emission potential, but the seasonal accumulated THC emissions within a site appeared to be largely independent of seasonal productivity at the same site. Short-term (hourly) NEE was related to hydrocarbon emission at the two CH4 emitting sites. A greater part of the carbon emitted as THC is likely to recently have been sequestered by the plants through photosynthesis and plant mediated CH4 transport during high CO2 metabolism will strengthen the short-term correlation between THC and CO2.
B53A-0923
Intensive studies on trace gas emission from a subarctic Swedish peatland.
Subarctic peatlands are one of the most sensitive ecosystems which react in a direct way to climate warming. Already now we can observe how formations underlain by permafrost are rapidly disintegrating and changing into a wetter hydrological state. These changes have triggered visible and substantial changes in the vegetation distribution and may represent an important feedback mechanism in changing climate through changes in the land-atmosphere exchanges of trace gases such as carbon dioxide (CO2) and methane (CH4).Within this study intensive and detailed flux measurements of both CH4 and CO2 were conducted during 2006 and 2007 over Stordalen mire, northernmost part of sub-Arctic Sweden (68° 20' N, 19° 03' E, alt. 351 m). Measurements were conducted with the use of two different techniques, automatic chambers and eddy correlation using a tunable diode laser. Landscape scale fluxes were documented by the eddy correlation system to be rather high, averaging 9 mg m-2 hr-1 during the peak season and these fluxes corresponded well with automatic chamber measurements in the wet minerotrophic parts of the peatland. It is the latter vegetation type that is expanding as the permafrost is melting. Apart detailed comparison between the results obtained with both techniques a full annual budget for the exchanges of radiative active trace gases between this mire and the atmosphere will be presented.
B53A-0924
Methane Dynamics in a Boreal Peatland: Combining Flux Measurements, Concentration Profiles in the Peat and Microbial Ecology at Different Spatial and Temporal Scales
Peatlands are one of the largest natural sources of methane to the atmosphere and methane is a significant component of the peatland carbon balance. We aimed to study the methane dynamics in a boreal, sedge- dominated fen, Siikaneva, by combining flux measurement using the eddy covariance (EC) technique and chambers, methane concentration measurements in the peat profile from 10 to 270 cm depth, and methane production potentials and microbial community composition at different depths. The temporal pattern in the fluxes was accurately detected with the EC method. The annual methane emission was 12.6 g CH4 m-2. The seasonal variation in the emissions was large, following the seasonality of the soil temperature. A small emission pulse was detected during the snow melt period. Despite the rather homogenous vegetation composition in the site, spatial variation in the fluxes was considerable and mostly controlled by water level and leaf area of sedges. The concentration profiles showed that most of the methane production took place at the depths of 20 and 50 cm, where most living roots are located. Concentrations at 10 cm were lower, indicating methane oxidation. Methane production potentials were highest at the depths of 20 and 10 cm. Production potentials and methanogenic communities varied across the site, especially at the depth of 20 cm. Vegetation was the major controller of the fluxes; plots in which the vegetation was removed showed very low fluxes. Occasionally, however episodic ebullition events were seen, when fluxes from the non-vegetated plots peaked, coinciding with the abrupt decrease in the concentrations in the peat. In conclusion, by combining different methods we can improve our understanding of the peatland methane dynamics.
B53A-0925
Seasonal Variation in Net Ecosystem Methane Flux in a Moderately-Rich Fen in Northern Alberta: Comparison Between Measurements and Model Calculations
We conducted measurements of seasonal variation in net ecosystem methane flux using the eddy covariance technique in a moderately-rich fen in northern Alberta, Canada during May-October 2007 as part of the Canadian Carbon Program (a follow-on program to the Fluxnet-Canada Research Network). Mid-day fluxes increased from near zero in May to a peak of approximately 4.3 mg m-2 h-1 in July and then declined through the rest of the growing season. There was a strong diurnal pattern in the concentration of methane (measured above the canopy at 10 m) and net methane flux, with low fluxes at night and peak fluxes occurring near mid-day. The diurnal pattern in concentration and flux had two primary causes. First, consistently low winds at night resulted in lack of turbulent exchange and a large build up in methane concentration below the eddy covariance sensors and the inlet for the methane analyzer. Second, model calculations suggested that stomatal closure at night limited methane flux through vascular plants. During the day open stomatal pores apparently allowed some methane flux through vascular plants thereby avoiding oxidation in aerobic layers near the surface of the peatland. The model calculations suggested that in July the net methane flux was a result of the difference between a maximum methane production rate of 6.3 mg m-2 h-1 and a maximum methane oxidation rate of 2.3 mg m-2 h-1 (both rates normalized to 10 degrees C), with 60 percent of methane production passing through vascular plants and being affected by stomatal control.
B53A-0926
Effects of Soil Warming and Drying on Methane Cycling in Northern Peatlands
Boreal peatlands contain a large portion of the earth's terrestrial organic carbon and may be particularly vulnerable to changes in climate. Temperature conditions in boreal regions are predicted to increase during the twenty-first century which may accelerate changes in soil microbial processes and plant community dynamics. Climate-driven changes in plant community composition might affect the pathways and rates of methanogenesis, the plant-mediated emission of methane and the scavenging of methane by methanotrophic bacteria. Climate change may also affect nutrient availability and cycling that indirectly affect methane cycling. To date, these feedbacks have not been incorporated into the carbon cycling components of climate models. We investigated the effects of soil warming and water-table manipulations on methane cycling in a field mesocosm experiment in northern Minnesota, USA. Large intact soil monoliths removed from a bog and fen received infrared warming treatments crossed with water-table treatments for six years. In years 5 and 6, concentrations, fluxes and isotopic compositions of methane were measured along with acetate, sulfate, ammonium, belowground net primary productivity and changes in N retention. Methane cycling is affected by changes in N availability associated with soil decomposition and through increased root productivity. An expansion of the rhizosphere of woody shrubs in bogs during the initial 4 yrs was associated with greater methane emission rates. We speculate that an increase in labile substrates associated with root exudates and enhanced plant transport may be factors contributing to the increase in methane emissions. Stable isotopic data from porewater support acetate fermentation as the principal pathway of methanogenesis in bog plots (mean ä13CH4 = -39.3 ‰). Under warm, wet conditions, the majority of the methane was isotopically heavy (mean ä13CH4 = -28.1 ‰), suggesting a predominance of methanotrophic activity throughout the soil system. Fen plots had lower porewater concentrations and emissions of methane than bog plots, despite much higher methane production potentials in fen peat. Increases in porewater ammonium may enhance methane oxidation under conditions of low water table. Our results illustrate the need for a more robust understanding of the multiple feedbacks between climate forcing and plant and microbial feedbacks in the response of northern peatlands to climate change.
B53A-0927
Ground Penetrating Radar Mapping of Spatially Continuous, Free-Phase Methane Trapping Layers in Glacial Lake Aggasiz Peatlands (GLAP), MN.
Ground Penetrating Radar (GPR) is a proven tool for non-invasive investigations of peatland stratigraphy due to the sensitivity of the method to minor variations in moisture content that coincide with vertical variations in peat fabric/structure. Detection of the interface between the peat and the mineral soil enables accurate (to about 25 cm) estimation of local peat thickness, while it is also possible to determine the internal stratigraphy of the Sphagnum peat mass. It has been previously postulated that woody deposits observed in cores through gassy peat may act as confining layers trapping free phase methane produced by methanogens. Ascending methane is assumed to be trapped as the wood layers are more structurally competent than the overlying peat fabric. Methane may be released from these pockets to the atmosphere during periods of abrupt atmospheric pressure changes. These conceptual models have been based on point source (peat core) data, leaving the spatial continuity of these confining layers unknown. We report on GPR measurements to investigate the spatial extent of such confining layers in the Glacial Lake Aggassiz Peatlands (GLAP). GPR data were collected from three sites in a 160 KM2 bog complex (1) the crest of the raised bog, (2) a midslope Sphagnum lawn (3) a fen water track on the lower slopes of the bog. Strong, laterally continuous and horizontal reflectors exist within the peat strata above the mineral soil interface at all three locations. At the fen site, the strongest reflector is between 1.8 – 2 m below the surface, whereas the Sphagnum lawn site contains a series of discontinuous reflectors at 2 m and 3 m below the surface. In contrast, The bog site is characterized by at several depths that are laterally continuous over tens of meters. The results imply that GPR could be used to non-invasively map likely methane accumulation hotspots if such layers indeed act to impede diffusive methane release to the atmosphere.
B53A-0928
Mapping Methane-Significant Dambo Wetlands in Central Uganda Using Remote Sensing and Topographic Data
Dambos are gently-sloping, seasonally-saturated valley floors that occupy up to 20 percent of the land surface of elevated plateaus in Eastern and Southern Africa. While they are likely candidates for contributing substantial methane fluxes to the atmosphere, their distribution and physical characteristics are poorly understood and there is almost no published data on their methane emissions. Dambo soil, vegetative, and hydrologic characteristics change along a gradual topographic gradient from occasionally-inundated margins to more frequently-inundated floors to perennially-inundated bottoms, and dambo methane production and emission likely varies along this gradient. We report on a study using multispectral remote sensing data and topographic attributes to map upland, margin, floor, and bottom classes of representative dambo wetlands in central Uganda. Two overlapping Systeme Pour l'Observation de la Terre (SPOT) 4 multispectral scenes, with an approximate spatial resolution of 20 meters and 2500 square kilometer extent, were acquired to coincide with the beginning and end of the January-to-March dry season in Central Uganda. Spectral indices and spectral mixture modeling fractions were calculated for both SPOT scenes. Vegetation phenological changes occurring in each dambo class were quantified by comparing remote sensing metrics in each SPOT scene. These metrics were combined with topographic data from the Shuttle Radar Topography Mission (SRTM) to classify the study area into upland, margin, floor, and bottom classes. Field data were used for training the classifier and assessing the accuracy of the final wetland classification. This map of methane-significant dambo wetlands is being employed to design the first systematic, regional scale field measurements of dambo methane emissions and will be used to model dynamic methane emissions in the study area. http://www.geog.utah.edu/~pdennison/dambos.html
B53A-0929
A First Estimate of African Tropical Dambo Wetland Spatial Extent, Saturation Seasonality and Duration, and Methane Emissions
Dambos constitute the largest geographic extent of seasonal wetlands in Central and Southern Africa, with potentially 2.3 million km2 of dambo wetlands throughout the African continent (United Nations Food and Agriculture Organization (FAO) estimate based upon the FAO soil map of the world). These indigenous African landscape features are seasonally saturated, grassy, channelless, gently sloping valley floors that occupy up to 20% of the gently undulating land surface on the elevated plateaus of Central and Southern Africa and can also be found scattered across the West African Sahel. With widespread occurrence, saturation and inundation for months during rainy seasons, an equatorial climate, and abundant soil organic matter, dambos likely play an important role in global atmospheric methane (CH4) dynamics. However, though a single unpublished measurement of dambo methane emissions has been reported, reliably estimating potential CH4 emissions for this type of wetland (amount, seasonality, and potential response to climate change) is difficult given a lack of reliable continent-wide data on dambo extent and annual duration of saturated conditions. Even where dambos have been well mapped (as in Zimbabwe), their methane-relevant components (margins, floors and bottoms) have not been distinguished. Moreover, static maps of dambo extent do not address significant variations in the annual duration of dambo saturation even within individual countries. To remedy this data gap, we report on a compilation of existing dambo data from the African continent. Using existing reports, unpublished data, paper maps, and available digital spatial datasets, we estimate: (1) the geographic extent of dambo-occupied landscapes, dambos, and dambo components (with a focus on the wettest dambo bottoms); (2) the duration of dambo saturation for different climatic regions; and (3) total month x km2 of annual dambo saturation. Applying the limited CH4 measurements available, we derive a first estimate of dambo CH4 annual emissions for the African continent.
B53A-0930
Sonar Determination of Environmental and Geologic Controls of Spatial-Temporal Variability in the Coal Oil Point Seep Field, California
Marine seepage of geologic methane, CH4, is a significant yet poorly understood contribution to atmospheric greenhouse gases, and is at least 20 times as potent as CO2. Marine seeps, excluding hydrates, are ~13% of natural emissions although there is significant uncertainty due the lack of published studies. Currently, sonar is the best method for seep emission quantification. Repeat sonar surveys spanning ten years were conducted in the Coal Oil Point (COP) marine hydrocarbon seep field in southern California. The field is one of the largest in the world emitting 105m3day-1 of CH4 from ~3 km2 of seafloor at depths from 2 to 90 m. Seepage arises from the Monterey Fm reservoir, which is overlain and capped by the Sisquoc Fm. Surveys sought to locate seeps, quantify emissions, relate seepage to geological structures, and identify spatial- temporal variability in the seep field to examine controls of and environmental factors. Sonars used were wide beam 3 kHz and 3-15 kHz. To estimate emissions, RMS sonar backscatter amplitude was calculated for a fixed depth window above the seafloor and normalized to the bottom return (termed J). Then, seepage above noise along track lines was quantified for J and gridded with ~40 m resolution. The noise level was calculated from the probability distribution of J for each track line. Emissions were derived from J using a lab calibration. Results suggest geology has a controlling influence on seep spatial distribution through features including fault planes, outcrops of the contact between the Monterey and Sisquoc Fms, and folded beds. The bed and fault plane interfaces provide pathways for bubbles to migrate from the Monterey Fm through and around the capping units. Emissions show significant variability on tidal to decadal time scales and decameter to km spatial scales, including the activation and deactivation of seepage areas. Environmental factors influence seepage- e.g., emissions inversely relate to tides- with variability occurring along pathways that appear to be geologically controlled. Emission uncertainty arises from the calibration function, which showed very poor dynamic range, and the inherent geometric limitations in wide beam sonar. Further, its low spatial resolution prevents correlation of seepage with smaller geologic structures. Recent studies show high frequency multibeam sonar mapping achieves spatial resolutions < 50 cm. This allows identification of individual bubble plumes, eliminating geometric uncertainty and improving dynamic range. http://www.bubbleology.com
B53A-0931
Methane uptake in forest and agro-ecosystems in Australia
Oxidation of methane by methanotrophic bacteria in aerated soils does provide a considerable global sink for greenhouse gases (-30 Tg CH4/yr). The form of land-use can have a significant impact on the methane uptake capacity of a soil. We investigated the sink strength for methane uptake of forest ecosystems and agro- ecosystems in Australia using automated measurement systems and manual chamber methods. Our results demonstrate large differences in the methane uptake capacity of Australian soils. Data from Western Australia showed that CH4 uptake rates increased with stand age of plantations and were greatest in an undisturbed native forest and lowest in an improved pasture. Measurements in differently aged forest ecosystems indicated that sites with the most recent fire disturbance had the lowest methane uptake rates. Generally, native forest ecosystems showed the greatest methane uptake rates (up to 130 kg CO2-e ha yr). Plantations (eucalyptus/pine) showed significantly lower methane uptake rates (around 15 kg CO2-e ha yr). Grazed pastures in Australia had the lowest uptake rates (6 kg CO2-e ha yr) and were occasional methane sources. The methane uptake rates of soils were only marginally influenced by environmental parameters over the course of a year. Between sites the methane uptake rates were not related to soil parameters such as soil bulk density. Experiments with excavated soil cores demonstrated that diffusivity of methane through the upper soil layer was the rate limiting step. Our results indicate that the community structure of methanotrophic bacteria and substrate diffusivity are the most important factor influencing methane uptake rates in soils. Disturbance events such as change of land-use or vegetation structure can have significant impacts on the capacity of soils to take up methane.
B53A-0932
Automated measurements of soil-atmosphere methane, carbon dioxide, and nitrous oxide fluxes in an undisturbed forest in the Brazilian Amazon
The soil-atmosphere fluxes of methane (CH4), carbon dioxide (CO2), and nitrous oxide (N2O) depend upon the interactions of biological processes of production and consumption and the physical transport of these gases. Soil temperature and moisture exert strong controls on the soil-atmosphere exchange of all three trace gas fluxes. We measured the soil-atmosphere flux of CH4, CO2, and N2O with an automated chamber system that was installed in April 2001 in the Tapajos National Forest, Para, Brazil. This is a mature forest on a clay Oxisol. The mean annual temperature is 25C with the diurnal range often exceeding the variability in the annual daily means. The mean annual precipitation is about 2000 mm per year with a distinct dry season from July to December. Eighteen aluminum chambers were installed in a 0.5 ha area close to the flux tower at the km 67 LBA site. Eight of these chambers were sampled about 5 times per day (closed 7% daily) and the other 10 chambers were sampled individually approximately once per day (closed 1.5% daily). As expected, the drained upland soils were most often a weak CH4 sink. Methane fluxes averaged -0.31 +/- 0.6 mg CH4 m-2 d-1 during the dry season and –0.08 +/- 0.26 mg CH4 m-2 d-1 during the wet season. Episodic positive fluxes of up to 18 mg CH4 m-2 d-1 were observed in the wet season following sustained rainfall events. For the 8 frequently sampled chambers, annual mean flux ranged between from 2.6 to 3.0 µmol CO2 m-2 s-1 over the 4-year period. Dry season fluxes averaged 2.3 to 2.7 micromol CO2 m-2 s-1 and the wet season averages ranged between 2.8 to 3.3 micromol CO2 m-2 s-1. Nitrous oxide fluxes ranged between 0.8 and 2.8 ng N-N2O cm-2 h-1 during the dry season and 0.02 to 0.7 ng N-N2O cm-2 h-1 during the dry season. Fluxes of CH4 showed no correlation with soil temperature or moisture on the diel scale. On the diel scale, soil CO2 flux was positively correlated with soil temperature and negatively correlated with soil moisture content during the dry season. Wet season fluxes of CO2 were also positively correlated with 5 cm soil temperature. Nitrous oxide fluxes correlated with 5 cm soil temperature over the diel cycle only during the wet season.
B53A-0933
Methanotrophic Bacteria Maintaining and Coexisting With a Methanogenic Consortium in Soils at Man-Made Natural gas Seeps at Petroleum Well Sites in Western Canada
Unwanted leakage of natural gas (>90% vol. CH4) in soil is ubiquitous at petroleum well sites in western Canada. Long term monitoring at several of those man-made seeps demonstrates that aerobic methanotrophy is the most important microbially mediated process in the soil. Bacterial oxidation of methane is severely impeded by soil freezing in winter and also in soils contaminated with liquid hydrocarbons. Soil gas compositions and carbon stable isotope data demonstrate that bacterial methanogenesis also occurs in deeper sections of the soil column at a number of the seeps. While methanogenic activity is common in soil contaminated with liquid hydrocarbons, it is also detected in soils where no evidence of oil contamination has been found (i.e., in "clean" soils). Methanogenic activity in "clean" soils occurs in the summer when soil temperature and moisture are comparatively high. Results also demonstrate that at this time of the year metanotrophic bacteria move upwards in the soil column whereby consuming nearly all available oxygen methanotrops are, thus, providing a habitat for the methane producing microorganisms. The concentrations of electron acceptors other than CO2 (HCO2) in soil moisture are very low or below detection, hence, indicating that methanogenesis is the energetically favorable terminal electron acceptor process in this environment. Conspicuous lack of soil biomass of methanotrophic origin evident from the low concentrations of soil organic matter (SOM) and the comparatively high carbon stable isotope composition of SOM, suggests that fermenting bacteria from the methanogenic consortium metabolize lysed cells and polysaccharide film left over from the demise of methanotrophic bacteria in winter, therefore practically feeding of the latter. Methane generated by the methanogens mixes with leaking gas and it is, sometimes, completely consumed by the methanotrophs. The seasonal interplay of methanogenic and methanotrophic bacteria at the man-made natural gas seeps in western Canada is a remarkable example of the ability of the soil microorganisms involved in the methane cycle to coexist and rapidly adapt to diverse environmental conditions.
B53A-0934
The Application Of Biofilter System For Reduction Of Methane Emissions From Modern Sanitary Landfills
Increased atmospheric concentrations of greenhouse gases (GHG) caused by anthropogenic activities has been related to global climate change. Methane, the second most important GHG after CO2, is 21 times more effective at trapping heat than CO2. Therefore, methane emission control is of utmost importance for global warming reduction. To minimize leachate production and protect groundwater resources, modern sanitary landfills are equipped with composite covers and gas collection systems. Methane from modern sanitary landfills is vented directly to the atmosphere, except for some of the largest landfills where it is recovered as energy and burned at the site. However, the efficiency of energy recovery systems in larger landfills is reduced as the amount of CH4 generated from landfill begins to decrease. In this study, the performance of a lab-scale model biofilter system was investigated to treat CH4 gas emitted from modern sanitary landfills by conducting batch and column experiments using landfill cover soil amended with earthworm cast as the filter bed medium. From the batch experiments to measure the influence of moisture content and temperature of the filter medium on CH4 removal capacity of a biofilter system, the optimum moisture content and temperature were found to be 10-15% by weight and 25-35°C, respectively. The column experiment was conducted to measure the influence of inlet CH4 concentration and CH4 loading rate on CH4 removal capacity of a biofilter system. As the inlet CH4 concentration decreased, the percentage of CH4 oxidized increased. Up to a CH4 loading rate of 2785 g CH4 m3 h- 1 (EBRT = 7.7 min), the CH4 removal efficiency of the biofilter was able to reach 100%. Based on the results of the study, the installation of a properly managed biofilter system should be capable of achieving a reduction in atmospheric CH4 emissions from modern sanitary landfills at low CH4 generation stage.
B53A-0935
Exploring the Relationship Between Wetland Methane Emissions and Net Ecosystem Productivity Using Experimental Shading and Labile Carbon Additions.
Methane (CH4) emissions from wetlands are positively correlated with net ecosystem productivity (NEP); however the relative importance of proposed controlling mechanisms remains poorly understood. The carbon supply hypothesis suggests that recent photosynthesis contributes labile carbon substrate to methanogenic habitats, resulting in higher CH4 emissions with increases in NEP. Plant gas transport is also hypothesized to be important for conducting gases between the soil and the atmosphere. High CH4 production rates often occur in saturated wetland soils where gas diffusion is extremely slow. The aerenchymous tissues of vascular wetland plants can serve as the primary pathway for CH4 emissions from the soil to the atmosphere, while also allowing CH4 to bypass more aerobic soil regions where CH4 oxidation could occur. Using a hypothesis-driven experimental approach, we established shading treatments in a Juncus- dominated wetland in the northern Colorado Front Range, and measured CH4 and CO2 fluxes with a static chamber technique. In the summer of 2007, the shading manipulations (45% and 65% shade) significantly reduced net ecosystem exchange (NEE; an approximation of NEP) and mean CH4 fluxes compared to control plots (p=0.02 and p=0.01, respectively). To test the carbon supply hypothesis, we injected a solution containing acetate (a primary methanogenic carbon source) to a depth of 20cm below the soil surface. Acetate additions stimulated CH4 emission rates across all plots by an average of 29.3% (p=0.01). However the strength of the CH4 emission response was not significantly related to plot treatment or NEE, indicating that reduced carbon supply could not explain the response to shading. We hypothesize that reduced plant gas transport was more important than labile carbon supply for driving the lower CH4 emission rates in shaded plots. The dry weight of above-ground biomass was lower in shaded plots (p=0.04), suggesting a possible link between plant gas transport capacity and the quantity of above-ground tissues. More work is necessary to understand the role of wetland vegetation communities and their gas transport properties as a mechanism to control patterns of CH4 emissions from wetlands.
B53A-0936
Quasi-continuous Measurements of Methane Fluxes Over a Managed Peatland in the Sacramento-San Joaquin Delta
Peatlands are composed of vegetation growing in water-logged, wetland environments and, in general, they are effective sinks of carbon because their saturated soils inhibit the decomposition of detritus. Because peatlands store so much carbon they have played a critical role in maintaining regulating and moderating climate and, consequently, they are expected to play a critical role in our future climate, especially if their water table drops and the organic peat is oxidized and released to the atmosphere as carbon dioxide. Fresh-water peatlands, with saturated soils and high carbon content, are also significant sources of carbon, in the form of methane, an important greenhouse gas. Most methane emission datasets and derived source-functions have been generated using chambers placed over wetlands, plants and soils. This methodology is useful and appropriate for studying mechanisms and spatial variability at small scales. However, it is less well suited for estimating annual fluxes or patterns over large areas. The eddy covariance method is capable of measuring methane efflux directly and for extended periods in a quasi-continuous manner. Unfortunately, this method has been under utilized, in part because of the reliance on expensive and labor-intensive tunable diode laser spectrometers. A new generation of infrared laser absorption spectrometers, which uses off-axis integrated cavity output spectroscopy, has recently become available (Los Gatos Research). The instrument can measure methane fluctuation at 10 Hz, hence it is suitable for continuous eddy covariance measurements of methane. Starting from April 2007, an eddy-covariance system, which measures CO2, CH4 and H2O directly and simultaneously, was installed to better estimate temporal and spatial patterns in methane efflux across the soil- plant-atmosphere interfaces of a managed peatland. The measurements were conducted at Sherman Island which is on the west side of the Sacramento-San Joaquin Delta and is on land managed by the California Department of Water Resources. Preliminary results show that the co-spectra of methane and vertical wind fluctuations are similar with the co- spectra of H2O, temperature and CO2. The fluxes were comparable with independent observations from chamber measurements. During the summer period, water table depth was 60 cm and CH4 fluxes were observed mainly during the night when the footprint was larger (due to increasing stability of the air) and advection from near wet fields were registered.
B53A-0937
Using Riverboat-Mounted Eddy Covariance for Direct Measurements of Air-water Gas Exchange in Amazonia
Gas evasion from Amazonian rivers and lakes to the atmosphere has been estimated to play an important role in the regional budget of carbon dioxide (Richey et al., 2002) and the global budget of methane (Melack et al., 2004). These flux estimates were calculated by combining remote sensing estimates of inundation area with water-side concentration gradients and gas transfer rates (piston velocities) estimated primarily from floating chamber measurements (footprint ~1 m2). The uncertainty in these fluxes was large, attributed primarily to uncertainty in the gas exchange parameterization. Direct measurements of the gas exchange coefficient are needed to improve the parameterizations in these environments, and therefore reduce the uncertainty in fluxes. The micrometeorological technique of eddy covariance is attractive since it is a direct measurement of gas exchange that samples over a much larger area than floating chambers, and is amenable to use from a moving platform. We present eddy covariance carbon dioxide exchange measurements made using a small riverboat in rivers and lakes in the central Amazon near Santarem, Para, Brazil. Water-side carbon dioxide concentration was measured in situ, and the gas exchange coefficient was calculated. We found the piston velocity at a site on the Amazon River to be similar to existing ocean-based parameterizations, whereas the piston velocity at a site on the Tapajos River was roughly a factor 5 higher. We hypothesize that the enhanced gas exchange at the Tapajos site was due to a shallow upwind fetch. Our results demonstrate the feasibility of boat-based eddy covariance on these rivers, and also the utility of a mobile platform to investigate spatial variability of gas exchange.
B53A-0938
The isotope signature of methane emitted from plant matter upon irradiation with UV light
New experiments show that dry and fresh leafs and other plant matter, as well as several structural plant components, emit methane upon irradiation with UV light and heating. We have determined the carbon and hydrogen isotopic composition of the produced methane in an experimental setup where plant matter is irradiated with UV light or heated in a quartz reactor. CH4 concentration, emission rate and isotopic composition are determined by measurements using an optical absorption technique (cavity ringdown spectroscopy), gas chromatography and isotope ratio mass spectrometry. We will present the source isotope signatures of the methane emitted from a range of natural plant materials. If the aerobic methane source from plants is indeed large, this has to be included in global isotope budgets.
B53A-0939
Aerobic Methane Generation From Plants (AMP)? Yes, Mostly!
In 2006, Keppler et al. (K) published an intriguing and revolutionary idea that aerobic methane is produced in plants (AMP) and released to the atmosphere. Their initial scaling calculations estimated the amount of AMP fluxing from living plants to range from 62-236 Tg/y and 1-7 Tg/y for plant litter. Houweling et al. (2006) (H) refined this flux to ca. 85 Tg/y PIH and 125 Tg/y present day. More recently, Dueck et al. (2007) (D) challenged the claim of AMP from intact plants. Their experiments cited "...No evidence for substantial aerobic methane emission by terrestrial plants..." (max. 0.4 ng/g h-1). Due to the significance of AMP in understanding present and palaeo-atmospheric budgets (e.g., Whiticar and Schaefer, 2007), we conducted a wide range of experiments to confirm or refute the existence and magnitude of AMP. For explanation, experiments of K were time-series batch samples measured by gas chromatography on purged and ambient samples, whereas D used continuous-flow cuvettes and measured by optical PAS with time series single injections. Our longer-term experiments with corn, wheat, tomato, red cedar, chestnut, moss and lichen (3-97 h, 32 °C) used a plant chamber, flow-through system with a GYRO, an optical spectrometer that enables continuous 1 Hz CH4 measurements with a precision of ca. 1 ppbv. We conducted over 100 chamber experiments on sterilized and non-sterilized (Cs-137 radiation) samples of: 1) intact living plants (IP), 2) fresh leaves (FL) and 3) dried leaves (DL); under both 1) high and 2) low light conditions (HL, LL), and with 1) ambient CH4 (AM, ca. 1.92 ppmv) and 2) purged methane (PM, 10 and 96 ppbv) levels. Our results demonstrate that IP-AMs have CH4 flux rates of 0.74-3.48 ng/g h-1. In contrast, IP-PMs show intense CH4 uptake rates of -28.5 to -57.9 ng/g h-1 (substantially different than K's reported emissions of 12-370 ng/g h-1 values). Our FL-AM-LL have CH4 flux rates of 0.36-2.05 ng/g h-1, whereas FL-AM-HL have significant CH4 generation of 0.27 to 12.7 ng/g h-1 (substantially higher than K's max of 3 ng/g h-1). FL-PM emissions are low (ca. 1 ng/g h-1). DL CH4 release is also low ranging from LL of 0.33 to HL of 3.37 ng/g h-1. Interestingly, our Cs-irradiated FL have increasingly higher CH4 emission rates with higher radiation dosages. We do not attempt to extrapolate our AMP laboratory experiments to global scales, nor make any physiological, biochemical or mechanistic claims. However at this point our work does indeed confirm that AMP is indeed operative and significant under certain conditions. The magnitude of our small scale, laboratory, AMP emission experiments is consistent with the earlier claims of K and H. We have, to some degree, emulated the experimental designs of both K and D. We remain intrigued by the findings, yet uncertain, if not puzzled, by the process and the discrepancies between groups.
B53A-0940
Aerobic Emission of Methane by Terrestrial Plant Material in Response to UV- irradiance
Recently, CH4 emission from living and dead plant tissues was demonstrated to occur under aerobic conditions. This work included some calculations to extrapolate the findings from the laboratory to the global scale and led various commentators to question the value of planting trees as a greenhouse mitigation option. These original findings have yet to be independently repeated and confirmed. Instead, the only other detailed study that has been published did not find any significant aerobic emissions of methane. We have set up experiments, in which we employ photo-acoustical and gas chromatographic methods to test the controversy regarding CH4 emission by plants. We will present data from very different types of plant material exposed to varying levels of UV-irradiance in order to test the hypothesis, that plant CH4 emission is a result of UV-mediated CH4 release. Further, we discuss the implications of our findings for an up scaling of CH4 production by terrestrial vegetation to a global level and thus potential reevaluation of the global CH budget.
B53A-0941
Evaluation of the Effects of Iron Oxides on Soil Reducing Conditions and Methane Generation in Cambodian Wetland Rice Fields
Atmospheric concentrations of methane have been steadily increasing over the last 100 years, which has given rise to research of wetland rice fields, recently identified as a major anthropomorphic source of methane. Establishment of experimental soil pots, cultivating an aromatic early variety rice strain in the Kean Svay District of Cambodia, have recently been carried out to evaluate methods to minimize methane release by promoting redox buffering by iron oxides. In the first series of experiments, iron oxides were added to the soils and the rate of change in reducing conditions and methanogenesis onset was monitored. In the second series of experiments, plots are subject to periodic drying cycles to promote rejuvenation of buffering iron oxides. Initial results indicate a delay in the onset of methanogenesis, and overall methane generation, in plots where initial iron oxides concentrations are elevated.
B53A-0942
Magnitude and Spatio-Temporal Variability of Methane Export From a Seasonally Stratified, Eutrophic Lake
Recent studies have suggested that significant amounts of methane can be released to the atmosphere from freshwater lakes, particularly through bubbling. Yet the relative importance of ebullition in the methane cycle of aquatic ecosystems is poorly understood, since it is a spatially and temporally heterogeneous process. The primary intent of this research is to measure the magnitude, rate and distribution of methane loss from lakes to the atmosphere due to both bubbling and diffusion, using the eutrophic Upper Mystic Lake in Massachusetts as a case study. Over the summer of 2007, under-water conical chambers were deployed to provide an estimate of bubbling fluxes, and the composition of the gases released in this process. Most of the bubbling occurred during short episodes with peak flux rates approaching 200 ml/m2/d, while average ebullition was approximately 30- 45 ml/m2/d. Spatial variability in fluxes was also observed - shallower locations had approximately 3 to 5 times less flux than deeper stations. The mixing ratio of methane present in the collected gas varied across stations and ranged from ~50% to 90%, except at some shallow locations where it was less than 10%. Concentrations of dissolved methane at the surface were less than 1uM leading to calculated rates of diffusive loss that are less than 20 ml/m2/d. This suggests that bubbling, rather than diffusion, is likely to be the more important pathway for methane release to the atmosphere.
B53A-0943
Factors Affecting Greenhouse Gas Emissions From Rice Agriculture
Experiments have shown that a few factors control the emissions of methane from rice fields. Among the most significant factors are water management and soil amendments. Continuous flooding and organic fertilizers result in the highest emissions of methane while intermittent flooding and use of nitrogen fertilizers produce more nitrous oxide. We measured fluxes of methane and nitrous oxide from tubs planted with rice grown in a greenhouse at Portland State University. We used classical factorial experimental design to calculate interactions between water management, nitrogen fertilizer application, and organic matter (chopped rice straw) for emission of methane and nitrous oxide. We will discuss the results of three years of experiments. This research was supported by the Office of Science (BER), US Department of Energy, Grant No. DE-FG02- 04ER63913.
B53A-0944
Methane Production Pathways and Concentration Profiles in Rice Cultivation
Rice agriculture is a significant source of global atmospheric methane (CH4) and a model system to study the mechanisms of CH4 production and emissions. We find that CH4 production occurred throughout the soil profile (0-20cm) and production via hydrogenotrophic pathway was dominant near the surface. Results of CH4 concentration and redox potential measurements indicate that CH4 production occurred throughout the soil profile. Straw amendment led to overall higher CH4 concentrations throughout the soil profile, and particularly soon after rice transplanting. Under continuous flooding, CH4 concentrations in the straw-added treatment increased as the rice season progressed and approached steady state, while in the control concentrations slowly continued increasing at lower concentrations throughout the season. CH4 concentrations decreased during intermittent flooding and resumed at much lower levels after re-flooding. Based on our measurements of ¦Ä13CH4 profiles and the fact that hydrogenotrophic methanogenesis leads to the production of lighter CH4 than acetate-dependent methanogenesis, we infer that the fraction of CH4 production due to acetate-dependent methanogenesis was higher at depth (5cm-20cm) than at the surface (0- 5cm). The contribution of acetate-dependent methanogenesis increased when straw was added. Surface root zone, defined by confining the whole rice root system to a 25¦Ìm mesh nylon bag, had heavier CH4 in the pore water than the bulk soil zone, which may indicate a greater role of oxidation near the roots. Hence stable isotope is important for studying CH4 production and secondary isotope fractionation processes such as CH4 oxidation. To understand the underlying processes, we need to study both concentration and the isotopic composition of CH4 in the soil profiles. This research was supported by US Department of Energy (No. DE-FG02-04ER63913).
B53A-0945
Analysis of Methane and Carbon Dioxide Fluxes in Forest Soils Using Concentration and Stable Carbon Isotope Ratio Gradients
The importance and mechanisms of methane/carbon dioxide exchange between forest soils and the atmosphere continue to be debated. We have conducted summer fieldwork in 2005 and 2006 to collect and analyse soil gas samples from forest locations in Quebec and British Columbia. Carbon isotope ratios and concentration soil profiles for both CH4 and CO2 are determined using GC-IRMS to clarify carbon fluxes within these forest soils. Initial observations of δ13C and concentration gradients of CO2 show a increase in [CO2] and decrease in δ13CO2 with soil depth. These trends are indicators of both net CO2 production in and efflux from these forest environments. In contrast, the CH4 data show increasing δ13CH4 and decreasing [CH4] with increasing soil depth. This indicates that the soils at both locations act as net sinks of atmospheric methane. Chamber flux measurements provide further support for this conclusion. They show a net accumulation of CO2 accompanied by a decrease in CH4 concentration with time in the chamber headspace. In addition, the CH4 is 13C-enriched while the CO2 is 13C-depleted in the chamber experiments. Variations present within the soil CH4 and CO2 profile indicate a non-uniform distribution of methane-oxidizing potential within the soil column. Furthermore, flux estimates using the chamber gas concentration data reveal large temporal and spatial variations in fluxes within the same location, potentially due to local heterogeneities in site characteristics.