Science of the Total Environment 505 (2015) 385–389
Contents lists available at ScienceDirect
Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv
Short Communication
Effect of thaw depth on fluxes of CO2 and CH4 in manipulated Arctic coastal tundra of Barrow, Alaska Yongwon Kim ⁎ International Arctic Research Center (IARC), University of Alaska Fairbanks (UAF), Fairbanks, AK 99775-7335, USA
H I G H L I G H T S • • • • •
Different production processes for CO2 and CH4 from manipulated Arctic coastal tundra CO2 emission in the drained section, from ecosystem respiration CH4 production in the inundated section, from methanogenesis Expansion of tundra lakes, implying enhanced CH4 release Disappearance of lakes, connected to stimulated CO2 emission
a r t i c l e
i n f o
Article history: Received 3 June 2014 Received in revised form 24 July 2014 Accepted 15 September 2014 Available online xxxx Editor: P. Kassomenos Keywords: Thaw depth CO2 and CH4 fluxes Manipulation Arctic coastal plain Tundra
a b s t r a c t Changes in CO2 and CH4 emissions represent one of the most significant consequences of drastic climate change in the Arctic, by way of thawing permafrost, a deepened active layer, and decline of thermokarst lakes in the Arctic. This study conducted flux-measurements of CO2 and CH4, as well as environmental factors such as temperature, moisture, and thaw depth, as part of a water table manipulation experiment in the Arctic coastal plain tundra of Barrow, Alaska during autumn. The manipulation treatment consisted of draining, controlling, and flooding treated sections by adjusting standing water. Inundation increased CH4 emission by a factor of 4.3 compared to non-flooded sections. This may be due to the decomposition of organic matter under a limited oxygen environment by saturated standing water. On the other hand, CO2 emission in the dry section was 3.9-fold higher than in others. CH4 emission tends to increase with deeper thaw depth, which strongly depends on the water table; however, CO2 emission is not related to thaw depth. Quotients of global warming potential (GWPCO2) (dry/control) and GWPCH4 (wet/control) increased by 464 and 148%, respectively, and GWPCH4 (dry/control) declined by 66%. This suggests that CO2 emission in a drained section is enhanced by soil and ecosystem respiration, and CH4 emission in a flooded area is likely stimulated under an anoxic environment by inundated standing water. The findings of this manipulation experiment during the autumn period demonstrate the different production processes of CO2 and CH4, as well as different global warming potentials, coupled with change in thaw depth. Thus the outcomes imply that the expansion of tundra lakes leads the enhancement of CH4 release, and the disappearance of the lakes causes the stimulated CO2 production in response to the Arctic climate change. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Methane (CH4) is a major greenhouse gas, with 25 times the global warming potential of CO2 (IPCC, 2007), and thermokarst lakes and wetlands represent a very large source reservoir of atmospheric CH4 in the pan-Arctic regions of 60°N to 75°N (IPCC, 2007; Walter et al., 2008). Recent Arctic warming has produced significant changes in permafrost degradation, expansion of the shrub community, and hydrological cycle, in association with the large amounts of soil carbon stored in permafrost soils of northern high-latitude (Tarnocai et al., 2009; Grosse ⁎ Corresponding author. Tel.: +1 907 474 2674; fax: +1 907 474 2679. E-mail address: [email protected].
http://dx.doi.org/10.1016/j.scitotenv.2014.09.046 0048-9697/© 2014 Elsevier B.V. All rights reserved.
et al., 2011). And this stored soil carbon pool in the Arctic is becoming much more vulnerable for release to the atmosphere as CO2 and CH4 due to ecological and hydrological changes (Walter et al., 2008; Zona et al., 2009; Sachs et al., 2010). CO2 and CH4 emitted from soil to the atmosphere are regulated by environmental factors such as soil temperature, soil moisture, water table, thaw depth, and oxygen availability. Mastepanov et al. (2008) found large CH4 releases from tundra during the onset of soil freezing in Greenland. I pose a similar question here: do CH4 emissions surge under the Arctic coastal tundra conditions of this water table manipulation experiment, which thaw depth is much deeper during autumn? To address this, autumn flux-measurements of CO2 and CH4, as well as environmental factors, were investigated from the manipulation station of
386
Y. Kim / Science of the Total Environment 505 (2015) 385–389
the Barrow Environmental Observatory (BEO) (Oechel, 2009; Zona et al., 2009; Sturtevant et al., 2012). This experiment was formed into three 300-m sections, including dry (drained), control (intermediate), and wet (flooded) tramlines, by controlling standing water. Strack and Waddington (2007) suggested a differential response of carbon cycling (e.g., CO2 and CH4) as well as different global warming potentials for the water table drawdown, resulting in enhanced CO2 emission and declined CH4 release, due to lower water table and warmer soil temperature. Thus, this study focuses on the effect of manipulation (potentially from thaw depth) on fluxes of CO2 and CH4 within Arctic coastal tundra during mid-September. 2. Materials and methods 2.1. Study site and flux-measurements This study site is situated on coastal tundra near Barrow, on the North Slope of Alaska (Fig. 1), and is characterized by poorly drained, wet meadow tundra. The seasonal maximum thickness of the active layer ranged from 30 to 90 cm (Bockheim et al., 1999). Average annual air temperature and precipitation for 30 years (1975 to 2004) were −11.8 °C and 108 mm, respectively, and snowpack melting and freezing began in early June and late September–early October. Flux-measurements were conducted along three 300-m tramlines (see Fig. 1) of the Biocomplexity Experiment (BE) station, which was configured for the water table manipulation of a) dry (drained: south; 71°17′00.7″N; 156°35′56.1″W); b) control (middle; 71°16′50.1″N; 156°35′46.8″W); and c) wet (flooded: north; 71°17′10.8″N; 156°36′10.7″W) treatments in the Barrow Environmental Observatory (BEO) station (Oechel, 2009; Zona et al., 2009; Sturtevant et al., 2012). The purposes for the water table manipulation experiment were to interpret changes in the vulnerable tundra ecosystem, such as alteration in the magnitude of permafrost degradation, active layer thickness, and emissions of CO2/CH4 in response to drastic climate changes in the Arctic (Oechel, 2009; Zona et al., 2009). Dominant plants were sedges, grasses, mosses, and dwarf shrubs such as Eriophorum vaginatum, Salix rotundifolia, Paper macounii, Alopecurus alpinus, Saxifraga caespitosa, Vaccinium vitus-idaea, Hippuris vulgaris, and Petasites frigidus (Raynolds et al., 2005).
Chamber-flux measurements of CO2 and CH4 were carried out on the soil surface at each of twelve locations along a 200-m tramline boardwalk within the three manipulation treatment sections in the autumn of 2007, during the onset of the surface water-freezing period, and when the boardwalk between 200 and 300 m line was submerged. The static chamber was formed of transparent polyethylene materials (50-cm diameter; 30-cm height), which consisted of a fan for merging inside-chamber air, a tube (0.3 cm ID, 0.5 cm OD) for air collection, and another tube for balancing between inside- and outside-chamber pressures. Air sampling was collected at intervals of 10 min for half an hour. Collected air samples were moved to pre-vacuum vials (10 mL), and then transferred to the laboratory for analyses of CO2 and CH4 concentrations. The analytical methods of CO2 and CH4 were conducted by Kim et al. (2007) in detail. The standard gases used in calibrations of CO2 and CH4 were reported in Kim et al., 2007. Precisions with calibrated ranges for CO2 and CH4 were usually less than 1% and 3%, respectively. Flux was calculated from the gradients of the relationship between gas concentration and sampling period with a correlation coefficient above 0.99 and a linear curve (Kim and Tanaka, 2003; Kim et al., 2007). The collected bubble air sample was analyzed for concentration and stable isotopes (δ13C and δD) of methane with a CH4 pre-concentration device and a continuous-flow gas chromatograph isotope ratio mass spectrometer (GC-IRMS) equipped with both a combustion and pyrolysis furnace. The precision of our system is 0.08‰ for δ13C and 2.2‰ for δD (1σ). δ13C and δD are expressed in δ notation compared to the international standard as follows: δ = (Rsample / Rstandard − 1) × 1000 (‰), where δ denotes 13C or D, and R is the molar ratio of heavy to light isotopes δ13C/δ12C or D/H from Vienna Pee Dee Belemnite (VPDB) and Vienna Standard Ocean Water (VSMOW), respectively. Because the collected bubble air sample may show higher CH4 concentration, the sample diluted to atmospheric level was analyzed. The system configuration resembles those previously reported by Rice et al. (2001), Miller et al. (2002), Yamada et al. (2003), Fisher et al. (2006), and Morimoto et al. (2006). The analytical method of stable isotopic methane was reported by Umezawa et al. (2009) in detail. Soil moisture and temperature were measured with a portable soil-moisture logger (HH2, DeltaT Devices, England) with probe (ML2, Delta-T Devices, England), and thermometer (Platinum Sensor, VWR Scientific Inc., U.S.A.), at 5-cm depth below the surface at each tramline, as were flux measurements. Temperatures of air at a 1-m height from the surface and of soil at 5-cm depth were monitored with a two-sensor thermometer (4 ChannelHOBO U-12, Onsetcomp Inc., USA). Thaw depth was measured with a 2 m long tile probe at an interval of 2 m at each tramline. In order to determine statistical significance, a t-test was used for two-sample comparison and multiple comparisons, respectively. An alpha level of 0.05 was used for all statistical procedures. 3. Results and discussion 3.1. Spatial patterns of environmental factors and fluxes of CO2 and CH4
Fig. 1. Aerial oblique photograph of the Biocomplexity Experiment station, looking north. Boardwalks (thin lines) and three 300-m tramlines (thick lines) are shown on the polygons, during the autumn period of 2007.
During mid-September of 2007, temperatures in air and soil had a diurnal pattern, as shown in Fig. 2a; furthermore, air temperature dropped below zero, with a newly formed icy water surface layer on 18 September 2007, trapping air bubbles in the wet treatment section (not shown). Sturtevant et al. (2012) simulated average thaw depths for all manipulation sections using ground heat flux measurements until 19 September, and based on the linear relationship between ground heat flux and thaw depth, after thaw depth was over 15 cm. This suggests that for declines in surface temperature below 2 °C, thawing ceases and refreezing starts from the bottom-up (Osterkamp and Romanovsky, 1997; Romanovsky and Osterkamp, 1997; Mastepanov et al., 2008). Thaw depth at three tramlines of N 100 m showed distinctly spatial patterns, as shown in Fig. 2b. Average thaw depths (average ± standard deviation) in dry, control, and wet treatment sections were 26.6 ± 2.7, 30.3 ± 4.1, and 35.5 ± 5.4 cm,
Y. Kim / Science of the Total Environment 505 (2015) 385–389
387
b)
a)
Fig. 2. Temporal variation of (a) temperatures in air and soil, and (b) spatial pattern of thaw depth in dry, control, and wet treatment sections during the autumn period of 2007.
respectively. There are significant differences (p = 0.268 and 0.074) between dry and wet sites, and control and wet sites; however, there is no significant difference (p b 0.0001) between dry and control sites at a confidence level of 95%. The effect of water table manipulation indicated a difference of 3–28 cm between dry and wet tramlines (Sturtevant et al., 2012). Spatial patterns of CO2 and CH4 fluxes are shown in Fig. 3, and include average CO2 fluxes of 3.81 ± 1.05, 0.82 ± 0.16, and 1.15 ± 0.21 gC m− 2 day−1, as well as average CH4 fluxes of 31.9 ± 9.4, 48.0 ± 16.3, and 71.7 ± 20.2 mgC m−2 day−1 in dry, control, and wet sections, respectively. CO2 emission at the wet site is higher than at the control site, suggesting the results of decomposition of acetate as the primary organic terminal product under the sub-oxic or anoxic environments, due to saturated standing water: CH3COOH → CO2 + CH4 in Alaska bogs and tundra wetlands (Duddleston et al., 2002; Ström et al., 2003) and/or for higher CO2 and CH4 emission by wet meadows in temperature/vegetation manipulation in Alaska wet-meadow and tussock tundra communities (Verville et al., 1998). CH4 fluxes showed much higher values when soil was submerged, as shown in Fig. 3. CH4 was released when surface soil temperature began to freeze, from late September to early October, ranging from 23.0 ± 4.9 to 301 ± 102 mgC m−2 day−1 in Greenland tundra (Mastepanov et al., 2008). CO2 and CH4 concentrations in trapped bubble samples were 5150 and 53,630 ppm, respectively, suggesting CH4 ebullition as a process of CH4 release within the wet treatment section (Walter et al., 2008). In addition, I calculated CH4 abundance from area and thickness, as well as the CH4 concentration in the bubbles, which was at least 2.57 gC m−2 day−1 based on the area of the circle with minor axis and the 12 h of the freezing period (see Fig. 2a). This value is much higher than the maximum CH4 emission of 1.98 gC m−2 day− 1 in Greenland during the onset of freezing (Mastepanov et al., 2008), and it suggests a large CH4 surge into the atmosphere due to the frequently freezing and thawing thin ice above the standing water surface during late September–early October (Sturtevant et al., 2012). The δ13C and
a) Dry
b) Control
δD of CH4 was − 73.1 and − 329‰ from bubble samples, respectively —values similar to those from ebullition events with ice koshkas: i.e., the Russian term for pockets of CH4-rich gas bubbles trapped in lake ice (− 75.5 to −67.6‰ of δ13C-CH4 and − 353 to −323‰ of δDCH4) in Alaska tundra lakes (Walter et al., 2008). Trapped-air CH4 bubbles during autumn were more δ13C-CH4 enriched than fresh point source bubbles (79.7 ± 3.0‰, n = 13: Walter et al., 2008). However, because δ13C-CO2 was not analyzed, the methane production pathway cannot be determined using the carbon fractionation factor (αc) between CO2 and CH4, for acetate fermentation and CO2 reduction. The αc values of the bubbles indicate the pathway of CH4 production, in which αc N 1.06 suggests that CH4 is produced mainly by CO2 reduction, and αc b 1.055 indicates production increasingly by acetate fermentation (see Fig. 5: Walter et al., 2008). 3.2. Response from CO2 and CH4 fluxes to environmental factors CO2 and CH4 fluxes showed no or much lower relationships to soil moisture before soil was entirely saturated, and with nearly constant soil temperature at 5-cm depth (not shown). On the other hand, the pattern of CH4 emission within control and wet sections represented a positively linear relationship with thaw depth, as shown in Fig. 4a, in which thaw depth elucidated 59 and 84% of the variability in CH4 emission when the water table was at or above the surface in the control and wet sections, respectively. Further, thaw depth is one of the most significant environmental factors in regulating higher CH4 emission (i.e., over 50 mgC m−2 day−1) in the control and wet manipulated sections, when soil moisture reached to be saturated (see Fig. 3b and c). The average water table in the wet manipulation section, which is a significant factor in controlling CH4 emission (Verville et al., 1998; Strack and Waddington, 2007; Zona et al., 2009), was significantly higher than those in the dry and control sections (Oechel, 2009). This water table drawdown manipulation experiment has also resulted in reduced CH4 emissions from the control sites (Moore and Roulet, 1993; Strack and
c) Wet
Fig. 3. Spatial variations in CH4 and CO2 fluxes in (a) dry, (b) control, and (c) wet treatments along each manipulation experiment tramline. Higher CH4 emission was produced when soil was saturated and showed deeper thaw depth.
388
Y. Kim / Science of the Total Environment 505 (2015) 385–389
b)
a)
Fig. 4. Responses from (a) CH4 and (b) CO2 fluxes to thaw depth (cm, TD), for which the equations show CH4 = 9.7 TD-260 (R2 = 0.59; p = 0.482) and CH4 = 11.3 TD-297 (R2 = 0.84; p = 0.003) under control (solid line) and wet (dashed) treatments, respectively, and CO2 = −0.62 TD + 18.8 (R2 = 0.28; p = 0.482) for dry treatment.
Waddington, 2007). Because this reduced the size of the anoxic zone and increased the oxic zone in the peat profile, the amount of CH4 produced has probably been reduced, while CH4 oxidation has increased, thereby diminishing CH4 emissions (Strack and Waddington, 2007). The response from CO2 flux in the dry section to thaw depth was lower (Fig. 4b), at which thaw depth explained 28% of CO2 variability, resulting from deeper thaw depth and the inundated manipulation of the section, as in control and wet sections. Zona et al. (2009) also demonstrated that CH4 emission tends to increase with a higher water table and subsequently deeper thaw depth in manipulation experiments within Alaskan Arctic tundra (see Fig. 6: Zona et al., 2009). The relationships between CH4 flux in these three manipulated sections as well as CO2 flux in the dry and control sections are shown in Fig. 5. CO2 flux in the dry site has a weakly linear relation to CH4 fluxes at the control site (Fig. 5a). Also, CO2 flux at the control site has linear relations to CH4 fluxes in both the dry and wet manipulated sites (Fig. 5b). Average CO2 flux in the control manipulation corresponded to 79 and 27% of the variability of average CH4 emission in the dry and wet sections, indicating that CO2 emissions in these two sections were enhanced with drier and wetter conditions (Verville et al., 1998), respectively. On the other hand, average CH4 emission showed increasing patterns with thaw depth and soil moisture, and indicated an increase of 50 and 123% in the control and wet sections compared to the dry treated section, respectively. Thus, this feature is thought likely to represent a difference in the production mechanisms of CO2 and CH4 under the three manipulation treated sections by this water table controlling experiment in the Alaska coastal plain tundra during autumn. To estimate the response from temperature dependence on CO2 and CH4 fluxes, the relationship was plotted, showing exponential curves of soil temperature at 5-cm depth from the equation (not shown): effluxes of CO2 and CH4 = β0 × eβ1 × T, where fluxes of CO2 and CH4 are the measured fluxes of CO2 (gC m−2 day−1) and CH4 (mgC m−2 day−1); T is soil temperature (°C); and β0 and β1 are constants. This exponential
a)
b)
relationship is commonly used to represent carbon flux as a function of temperature (Davidson et al., 1998; Davidson and Janssens, 2006; Kim et al., 2007, 2013). Q10 temperature coefficient values were calculated as in Davidson et al. (1998) and Kim et al. (2013): Q10 = e β1 × 10, where Q10 is a measure of the change in biochemical reaction rate at intervals of 10 °C and is based on Van't Hoff's empirical rule that a rate increase of two to three times occurs for every 10 °C rise in temperature (Lloyd and Taylor, 1994). Q10 values were 2.10 (dry), 2.06 (control), and 107 (wet) for CO2 flux; and 0.24 (dry), 0.01 (control), and 2.35 × 105 (wet) for CH4 flux, respectively. Interestingly, CH4 production rate in the flooded manipulation section is stimulated to a much greater extent than CO2 production within a narrower soil temperature range of 2.3 to 4.7 °C. This demonstrates the dominant contribution of methanogens to methane production under an anoxic environment of flooded sections during the autumn period. Further, a much higher methane concentration from trapped air bubbles may be related to the enhanced methanogenesis (Walter et al., 2008; Sturtevant et al., 2012). 4. Conclusions and implications This water table manipulation experiment resulted in distinct drained (dry), intermediate (control), and flooded (wet) treatment tramlines (Oechel, 2009; Zona et al., 2009), and the subsequent magnitude of CO2 and CH4 releases is likely thought to modulate for different thaw depths, depending on the water table, in the three sections. Higher CH4 emission occurred at each section when soil was submerged. Contrary to CH4 emission, however, CO2 emission was highest in the dry treatment section, compared to the control and wet sections. This suggests a different production mechanism for CO2 and CH4 under these water table manipulation sections. The isotopic discrimination of CH4 from trapped bubble air was similar to isotopic characteristics of ebullition events in Alaska tundra lakes (Walter et al., 2008). This study therefore found that the change in thaw depth under water table
c)
Fig. 5. Responses of (a) CH4 flux under control for CO2 flux under dry treatment, and of CH4 flux under (b) dry and (c) wet treatments for CO2 flux under control treatment.
Y. Kim / Science of the Total Environment 505 (2015) 385–389
manipulation affected production processes of CO2 and CH4, with thaw depth as a significant key in regulating CO2 and CH4, compared to other environmental factors during autumn. I have thusly figured the global warming potential (GWP) of these gases at these three particular treated sections throughout 61 days of September and October (autumn) by considering a 100-year timescale, and weighing the warming potential of CH4 at 25 times that of CO2 (IPCC, 2007). The quotients of GWPCO2 (dry/control) and GWPCH4 (wet/control) increased by 464 and 148%, respectively, and GWPCH4 (dry/control) declined by 66%. This suggests that CO2 emission from the drained section is thought to be enhanced further by soil and ecosystem respiration compared to the wet section, and CH4 emission in the wet area is likely thought to be stimulated under an anoxic environment by saturated standing water, in relation to other sections. On the other hand, CH4 emission in the drained area is constrained by relatively shallower thaw depth, which is CH4 oxidation (Strack and Waddington, 2007). Thus, the change in thaw depth in manipulated Arctic coastal tundra is important in modulating production mechanisms of CO2 and CH4 for the regional carbon balance, as is the potential for shallower thaw depth to enhance CO2 emission in the drained area and for deeper thaw depth to boost CH4 emission in the inundated area. Therefore, the expansion of tundra lakes implies the enhancement of CH4 emission (Walter et al., 2006); on the other hand, the shrinking and disappearing thermokarst lakes lead to the stimulated CO2 release (Yoshikawa and Hinzman, 2003; Smith et al., 2005) response to the recent changing climate in the Arctic. Acknowledgments This research was conducted under the JAMSTEC-IARC Collaboration Study with funding provided by the Japan Agency for Marine-Earth Science and Technology (JAMSTEC) under a grant to the International Arctic Research Center (IARC). The author thanks Dr. W. Oechel of San Diego State University for his research in the manipulation experiment of the BEO, and Mr. N. Bauer of the International Arctic Research Center (IARC) of the University of Alaska Fairbanks (UAF) for a review of the text as a professional English-language editor. Finally, I appreciate Dr. P. Kassomenos (Handling Editor) and an anonymous reviewer for beneficial comments. References Bockheim JG, Everett LR, Hinkel KM, Nelson FE, Barown J. Soil organic carbon storage and distribution in arctic tundra, Barrow Alaska. Soil Sci Soc Am J 1999;63:934–40. Davidson EA, Jassens IA. Temperature sensitivity of soil carbon decomposition and feedback to climate change. Nature 2006;440:165–73. Davidson EA, Belk E, Boone RD. Soil water content and temperature as independent or confounded factors controlling soil respiration in a temperate mixed hardwood forest. Global Change Biol 1998;4:217–27. Duddleston KN, Kinney MA, Kiene RP, Hines ME. Anaerobic microbial biogeochemistry in a northern bog: acetate as a dominant metallic end product. Global Biogeochem Cycles 2002;16:1063. http://dx.doi.org/10.1029/2001GB001402. Fisher R, Lowry D, Wilkin O, Sriskantharajah S, Nsbet EG. High-precision, automated stable isotope analysis of atmospheric methane and carbon dioxide using continuousflow isotope-ratio mass spectrometry. Rapid Commun Mass Spectrom 2006;20: 200–8. Grosse G, Harden J, Turetsky M, McGuire AD, Camill P, Tarnocai C, et al. Vulnerability of high-latitude soil organic carbon in North America to disturbance. J Geophys Res 2011;116:G00K06. http://dx.doi.org/10.1029/2010JG001507. Intergovernmental Panel on Climate Change (IPCC). Climate Change 2007: the physical science basis. New York, USA: Cambridge University Press; 2007. p. 996.
389
Kim Y, Tanaka N. Effect of forest fire on the fluxes of CO2, CH4, and N2O in boreal forest soils, interior Alaska. J Geophys Res 2003;108:8154. http://dx.doi.org/10.1029/ 2001JD000663. Kim Y, Ueyama M, Nakagawa F, Tsunogai U, Harazono Y, Tanaka N. Assessment of winter fluxes of CO2 and CH4 in boreal forest soils of central Alaska estimated by the profile method and the chamber method: a diagnosis of methane emission and implications for the regional carbon budget. Tellus 2007;59B:223–33. Kim Y, Kim SD, Enomoto H, Kushida K, Kondoh M, Uchida M. Latitudinal distribution of soil CO2 efflux and temperature along the Dalton Highway, Alaska. Polar Sci 2013; 7:162–73. Lloyd J, Taylor JA. On the temperature dependence of soil respiration. Funct Ecol 1994;8: 315–23. Mastepanov M, Sigsggard C, Dlugokencky EJ, Houweling S, Ström L, Tamstorf MP, et al. Large tundra methane burst during onset of freezing. Nature 2008;456:628–31. http://dx.doi.org/10.1038/nature07464. Miller JB, Mack KA, Dissly R, White JWC, Dlugokencky EJ, Tans PP. Development of analytical methods and measurements of 13C/12C in atmospheric CH4 from the NOAA Climate Monitoring and Diagnostic Laboratory Global Air Sampling Network. J Geophys Res 2002:107. http://dx.doi.org/10.1029/2001JD000630. Moore TR, Roulet NT. Methane flux: water table relations in Northern wetland. Geophys Res Lett 1993;20:587–90. Morimoto S, Aoki S, Nakazawa T, Yamanouchi T. Temporal variations of the carbon isotopic ratio of atmospheric methane observed at Ny Ålesund, Svalbard from 1996 to 2004. Geophys Res Lett 2006;33:L01807. http://dx.doi.org/10.1029/2005GL024648. Oechel WC. Ecosystem interactions. In: BEO subcommittee, Barrow Arctic Science Consortium for the Ukpeaġvik Iñupiat Corporation and North Slope Borough, editors. Barrow Environmental Observatory (Fifth annual report); 2009. p. 33–59. Osterkamp TE, Romanovsky VE. Freezing of the active layer on coastal plain of the Alaskan Arctic. Permafr Periglac Process 1997;8:23–44. Raynolds MK, Walker DA, Maier HA. Plant community-level mapping of arctic Alaska based on the Circumpolar Arctic Vegetation Map. Phytocoenologia 2005;35:821–48. http://dx.doi.org/10.11270340-269x/2005/0035-0821. Rice AL, Gotoh A, Ajie HO, Tyler SC. High-precision continuous-flow measurement of δ13C and δD of atmospheric CH4. Anal Chem 2001;73:4104–10. Romanovsky VE, Osterkamp TE. Thawing of the active layer on coastal plain of the Alaskan Arctic. Permafr Periglac Process 1997;8:1–22. Sachs T, Giebels M, Boike J, Kutzbach L. Environmental controls on CH4 emission from polygonal tundra on the microsite scale in the Lena river delta, Siberia. Global Change Biol 2010;16:3096–110. http://dx.doi.org/10.111/j.1365-2486.2010.02232.x. Smith LC, Sheng Y, MacDonal GM, Hinzman LD. Disappearing Arctic lakes. Science 2005. http://dx.doi.org/10.1126/science.1108142. Strack M, Waddington JM. Response of peatland carbon dioxide methane fluxes to a water table drawdown experiment. Global Biogeochem Cycles 2007;21:GB1007. http://dx.doi.org/10.1029/2006GB002715. Ström L, Ekberg A, Mastepanov M, Christensen TR. The effect of vascular plants on carbon turnover and methane emissions from a tundra wetland. Global Change Biol 2003;9: 1185–92. Sturtevant CS, Oechel WC, Zona D, Kim Y, Emerson CE. Soil moisture control over autumn season methane flux, Arctic Coastal Plain of Alaska. Biogeosciences 2012;9:1423–40. Tarnocai C, Canadell JG, Schuur EAG, Kuhry P, Mazhitova G, Zimov S. Soil organic carbon pools in the northern circumpolar permafrost region. Global Biogeochem Cycles 2009;23:GB2023. http://dx.doi.org/10.1029/2008GB003327. Umezawa T, Aoki S, Nakazawa T, Morimoto S. A high-precision measurement system for carbon and hydrogen isotopic ratio of atmospheric methane and its application to air samples collected in the Western Pacific region. J Meteorol Soc Jpn 2009;87:365–79. Verville JH, Hobbie SE, Chapin III FS, Hooper DU. Response of tundra CH4 and CO2 flux to manipulation of temperature and vegetation. Biogeochemistry 1998;41:215–35. Walter KM, Zimov SA, Chanton JP, Verbyla D, Chapin III FS. Methane bubbling from Siberian thaw lakes as a positive feedback to climate warming. Nature 2006;443. http:// dx.doi.org/10.1038/nature05040. Walter KM, Chaton JP, Chapin III FS, Schuur EAG, Zimov SA. Methane production and bubble emission from arctic lakes: isotopic implications for source pathways and ages. J Geophys Res 2008;113:G00A08. http://dx.doi.org/10.1029/2007JG000569. Yamada K, Ozaki Y, Nakagawa F, Tanaka M, Yoshida N. An improved method for measurement of the hydrogen isotope ratio of atmospheric methane and its application to a Japanese urban atmosphere. Atmos Environ 2003;37:1975–82. Yoshikawa K, Hinzman LD. Shrinking thermokarst ponds and groundwater dynamics in discontinuous permafrost. Permafr Periglac Process 2003;14(2):151–60. Zona D, Oechel WC, Kochendorfer J, Paw KT, Salyuk AN, Olivas PC, et al. Methane fluxes during the initiation of a large-scale water table manipulation experiment in the Alaskan Arctic tundra. Global Biogeochem Cycles 2009;23:GB2013. http://dx.doi. org/10.1029/2009GB003487.
Contents lists available at ScienceDirect
Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv
Short Communication
Effect of thaw depth on fluxes of CO2 and CH4 in manipulated Arctic coastal tundra of Barrow, Alaska Yongwon Kim ⁎ International Arctic Research Center (IARC), University of Alaska Fairbanks (UAF), Fairbanks, AK 99775-7335, USA
H I G H L I G H T S • • • • •
Different production processes for CO2 and CH4 from manipulated Arctic coastal tundra CO2 emission in the drained section, from ecosystem respiration CH4 production in the inundated section, from methanogenesis Expansion of tundra lakes, implying enhanced CH4 release Disappearance of lakes, connected to stimulated CO2 emission
a r t i c l e
i n f o
Article history: Received 3 June 2014 Received in revised form 24 July 2014 Accepted 15 September 2014 Available online xxxx Editor: P. Kassomenos Keywords: Thaw depth CO2 and CH4 fluxes Manipulation Arctic coastal plain Tundra
a b s t r a c t Changes in CO2 and CH4 emissions represent one of the most significant consequences of drastic climate change in the Arctic, by way of thawing permafrost, a deepened active layer, and decline of thermokarst lakes in the Arctic. This study conducted flux-measurements of CO2 and CH4, as well as environmental factors such as temperature, moisture, and thaw depth, as part of a water table manipulation experiment in the Arctic coastal plain tundra of Barrow, Alaska during autumn. The manipulation treatment consisted of draining, controlling, and flooding treated sections by adjusting standing water. Inundation increased CH4 emission by a factor of 4.3 compared to non-flooded sections. This may be due to the decomposition of organic matter under a limited oxygen environment by saturated standing water. On the other hand, CO2 emission in the dry section was 3.9-fold higher than in others. CH4 emission tends to increase with deeper thaw depth, which strongly depends on the water table; however, CO2 emission is not related to thaw depth. Quotients of global warming potential (GWPCO2) (dry/control) and GWPCH4 (wet/control) increased by 464 and 148%, respectively, and GWPCH4 (dry/control) declined by 66%. This suggests that CO2 emission in a drained section is enhanced by soil and ecosystem respiration, and CH4 emission in a flooded area is likely stimulated under an anoxic environment by inundated standing water. The findings of this manipulation experiment during the autumn period demonstrate the different production processes of CO2 and CH4, as well as different global warming potentials, coupled with change in thaw depth. Thus the outcomes imply that the expansion of tundra lakes leads the enhancement of CH4 release, and the disappearance of the lakes causes the stimulated CO2 production in response to the Arctic climate change. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Methane (CH4) is a major greenhouse gas, with 25 times the global warming potential of CO2 (IPCC, 2007), and thermokarst lakes and wetlands represent a very large source reservoir of atmospheric CH4 in the pan-Arctic regions of 60°N to 75°N (IPCC, 2007; Walter et al., 2008). Recent Arctic warming has produced significant changes in permafrost degradation, expansion of the shrub community, and hydrological cycle, in association with the large amounts of soil carbon stored in permafrost soils of northern high-latitude (Tarnocai et al., 2009; Grosse ⁎ Corresponding author. Tel.: +1 907 474 2674; fax: +1 907 474 2679. E-mail address: [email protected].
http://dx.doi.org/10.1016/j.scitotenv.2014.09.046 0048-9697/© 2014 Elsevier B.V. All rights reserved.
et al., 2011). And this stored soil carbon pool in the Arctic is becoming much more vulnerable for release to the atmosphere as CO2 and CH4 due to ecological and hydrological changes (Walter et al., 2008; Zona et al., 2009; Sachs et al., 2010). CO2 and CH4 emitted from soil to the atmosphere are regulated by environmental factors such as soil temperature, soil moisture, water table, thaw depth, and oxygen availability. Mastepanov et al. (2008) found large CH4 releases from tundra during the onset of soil freezing in Greenland. I pose a similar question here: do CH4 emissions surge under the Arctic coastal tundra conditions of this water table manipulation experiment, which thaw depth is much deeper during autumn? To address this, autumn flux-measurements of CO2 and CH4, as well as environmental factors, were investigated from the manipulation station of
386
Y. Kim / Science of the Total Environment 505 (2015) 385–389
the Barrow Environmental Observatory (BEO) (Oechel, 2009; Zona et al., 2009; Sturtevant et al., 2012). This experiment was formed into three 300-m sections, including dry (drained), control (intermediate), and wet (flooded) tramlines, by controlling standing water. Strack and Waddington (2007) suggested a differential response of carbon cycling (e.g., CO2 and CH4) as well as different global warming potentials for the water table drawdown, resulting in enhanced CO2 emission and declined CH4 release, due to lower water table and warmer soil temperature. Thus, this study focuses on the effect of manipulation (potentially from thaw depth) on fluxes of CO2 and CH4 within Arctic coastal tundra during mid-September. 2. Materials and methods 2.1. Study site and flux-measurements This study site is situated on coastal tundra near Barrow, on the North Slope of Alaska (Fig. 1), and is characterized by poorly drained, wet meadow tundra. The seasonal maximum thickness of the active layer ranged from 30 to 90 cm (Bockheim et al., 1999). Average annual air temperature and precipitation for 30 years (1975 to 2004) were −11.8 °C and 108 mm, respectively, and snowpack melting and freezing began in early June and late September–early October. Flux-measurements were conducted along three 300-m tramlines (see Fig. 1) of the Biocomplexity Experiment (BE) station, which was configured for the water table manipulation of a) dry (drained: south; 71°17′00.7″N; 156°35′56.1″W); b) control (middle; 71°16′50.1″N; 156°35′46.8″W); and c) wet (flooded: north; 71°17′10.8″N; 156°36′10.7″W) treatments in the Barrow Environmental Observatory (BEO) station (Oechel, 2009; Zona et al., 2009; Sturtevant et al., 2012). The purposes for the water table manipulation experiment were to interpret changes in the vulnerable tundra ecosystem, such as alteration in the magnitude of permafrost degradation, active layer thickness, and emissions of CO2/CH4 in response to drastic climate changes in the Arctic (Oechel, 2009; Zona et al., 2009). Dominant plants were sedges, grasses, mosses, and dwarf shrubs such as Eriophorum vaginatum, Salix rotundifolia, Paper macounii, Alopecurus alpinus, Saxifraga caespitosa, Vaccinium vitus-idaea, Hippuris vulgaris, and Petasites frigidus (Raynolds et al., 2005).
Chamber-flux measurements of CO2 and CH4 were carried out on the soil surface at each of twelve locations along a 200-m tramline boardwalk within the three manipulation treatment sections in the autumn of 2007, during the onset of the surface water-freezing period, and when the boardwalk between 200 and 300 m line was submerged. The static chamber was formed of transparent polyethylene materials (50-cm diameter; 30-cm height), which consisted of a fan for merging inside-chamber air, a tube (0.3 cm ID, 0.5 cm OD) for air collection, and another tube for balancing between inside- and outside-chamber pressures. Air sampling was collected at intervals of 10 min for half an hour. Collected air samples were moved to pre-vacuum vials (10 mL), and then transferred to the laboratory for analyses of CO2 and CH4 concentrations. The analytical methods of CO2 and CH4 were conducted by Kim et al. (2007) in detail. The standard gases used in calibrations of CO2 and CH4 were reported in Kim et al., 2007. Precisions with calibrated ranges for CO2 and CH4 were usually less than 1% and 3%, respectively. Flux was calculated from the gradients of the relationship between gas concentration and sampling period with a correlation coefficient above 0.99 and a linear curve (Kim and Tanaka, 2003; Kim et al., 2007). The collected bubble air sample was analyzed for concentration and stable isotopes (δ13C and δD) of methane with a CH4 pre-concentration device and a continuous-flow gas chromatograph isotope ratio mass spectrometer (GC-IRMS) equipped with both a combustion and pyrolysis furnace. The precision of our system is 0.08‰ for δ13C and 2.2‰ for δD (1σ). δ13C and δD are expressed in δ notation compared to the international standard as follows: δ = (Rsample / Rstandard − 1) × 1000 (‰), where δ denotes 13C or D, and R is the molar ratio of heavy to light isotopes δ13C/δ12C or D/H from Vienna Pee Dee Belemnite (VPDB) and Vienna Standard Ocean Water (VSMOW), respectively. Because the collected bubble air sample may show higher CH4 concentration, the sample diluted to atmospheric level was analyzed. The system configuration resembles those previously reported by Rice et al. (2001), Miller et al. (2002), Yamada et al. (2003), Fisher et al. (2006), and Morimoto et al. (2006). The analytical method of stable isotopic methane was reported by Umezawa et al. (2009) in detail. Soil moisture and temperature were measured with a portable soil-moisture logger (HH2, DeltaT Devices, England) with probe (ML2, Delta-T Devices, England), and thermometer (Platinum Sensor, VWR Scientific Inc., U.S.A.), at 5-cm depth below the surface at each tramline, as were flux measurements. Temperatures of air at a 1-m height from the surface and of soil at 5-cm depth were monitored with a two-sensor thermometer (4 ChannelHOBO U-12, Onsetcomp Inc., USA). Thaw depth was measured with a 2 m long tile probe at an interval of 2 m at each tramline. In order to determine statistical significance, a t-test was used for two-sample comparison and multiple comparisons, respectively. An alpha level of 0.05 was used for all statistical procedures. 3. Results and discussion 3.1. Spatial patterns of environmental factors and fluxes of CO2 and CH4
Fig. 1. Aerial oblique photograph of the Biocomplexity Experiment station, looking north. Boardwalks (thin lines) and three 300-m tramlines (thick lines) are shown on the polygons, during the autumn period of 2007.
During mid-September of 2007, temperatures in air and soil had a diurnal pattern, as shown in Fig. 2a; furthermore, air temperature dropped below zero, with a newly formed icy water surface layer on 18 September 2007, trapping air bubbles in the wet treatment section (not shown). Sturtevant et al. (2012) simulated average thaw depths for all manipulation sections using ground heat flux measurements until 19 September, and based on the linear relationship between ground heat flux and thaw depth, after thaw depth was over 15 cm. This suggests that for declines in surface temperature below 2 °C, thawing ceases and refreezing starts from the bottom-up (Osterkamp and Romanovsky, 1997; Romanovsky and Osterkamp, 1997; Mastepanov et al., 2008). Thaw depth at three tramlines of N 100 m showed distinctly spatial patterns, as shown in Fig. 2b. Average thaw depths (average ± standard deviation) in dry, control, and wet treatment sections were 26.6 ± 2.7, 30.3 ± 4.1, and 35.5 ± 5.4 cm,
Y. Kim / Science of the Total Environment 505 (2015) 385–389
387
b)
a)
Fig. 2. Temporal variation of (a) temperatures in air and soil, and (b) spatial pattern of thaw depth in dry, control, and wet treatment sections during the autumn period of 2007.
respectively. There are significant differences (p = 0.268 and 0.074) between dry and wet sites, and control and wet sites; however, there is no significant difference (p b 0.0001) between dry and control sites at a confidence level of 95%. The effect of water table manipulation indicated a difference of 3–28 cm between dry and wet tramlines (Sturtevant et al., 2012). Spatial patterns of CO2 and CH4 fluxes are shown in Fig. 3, and include average CO2 fluxes of 3.81 ± 1.05, 0.82 ± 0.16, and 1.15 ± 0.21 gC m− 2 day−1, as well as average CH4 fluxes of 31.9 ± 9.4, 48.0 ± 16.3, and 71.7 ± 20.2 mgC m−2 day−1 in dry, control, and wet sections, respectively. CO2 emission at the wet site is higher than at the control site, suggesting the results of decomposition of acetate as the primary organic terminal product under the sub-oxic or anoxic environments, due to saturated standing water: CH3COOH → CO2 + CH4 in Alaska bogs and tundra wetlands (Duddleston et al., 2002; Ström et al., 2003) and/or for higher CO2 and CH4 emission by wet meadows in temperature/vegetation manipulation in Alaska wet-meadow and tussock tundra communities (Verville et al., 1998). CH4 fluxes showed much higher values when soil was submerged, as shown in Fig. 3. CH4 was released when surface soil temperature began to freeze, from late September to early October, ranging from 23.0 ± 4.9 to 301 ± 102 mgC m−2 day−1 in Greenland tundra (Mastepanov et al., 2008). CO2 and CH4 concentrations in trapped bubble samples were 5150 and 53,630 ppm, respectively, suggesting CH4 ebullition as a process of CH4 release within the wet treatment section (Walter et al., 2008). In addition, I calculated CH4 abundance from area and thickness, as well as the CH4 concentration in the bubbles, which was at least 2.57 gC m−2 day−1 based on the area of the circle with minor axis and the 12 h of the freezing period (see Fig. 2a). This value is much higher than the maximum CH4 emission of 1.98 gC m−2 day− 1 in Greenland during the onset of freezing (Mastepanov et al., 2008), and it suggests a large CH4 surge into the atmosphere due to the frequently freezing and thawing thin ice above the standing water surface during late September–early October (Sturtevant et al., 2012). The δ13C and
a) Dry
b) Control
δD of CH4 was − 73.1 and − 329‰ from bubble samples, respectively —values similar to those from ebullition events with ice koshkas: i.e., the Russian term for pockets of CH4-rich gas bubbles trapped in lake ice (− 75.5 to −67.6‰ of δ13C-CH4 and − 353 to −323‰ of δDCH4) in Alaska tundra lakes (Walter et al., 2008). Trapped-air CH4 bubbles during autumn were more δ13C-CH4 enriched than fresh point source bubbles (79.7 ± 3.0‰, n = 13: Walter et al., 2008). However, because δ13C-CO2 was not analyzed, the methane production pathway cannot be determined using the carbon fractionation factor (αc) between CO2 and CH4, for acetate fermentation and CO2 reduction. The αc values of the bubbles indicate the pathway of CH4 production, in which αc N 1.06 suggests that CH4 is produced mainly by CO2 reduction, and αc b 1.055 indicates production increasingly by acetate fermentation (see Fig. 5: Walter et al., 2008). 3.2. Response from CO2 and CH4 fluxes to environmental factors CO2 and CH4 fluxes showed no or much lower relationships to soil moisture before soil was entirely saturated, and with nearly constant soil temperature at 5-cm depth (not shown). On the other hand, the pattern of CH4 emission within control and wet sections represented a positively linear relationship with thaw depth, as shown in Fig. 4a, in which thaw depth elucidated 59 and 84% of the variability in CH4 emission when the water table was at or above the surface in the control and wet sections, respectively. Further, thaw depth is one of the most significant environmental factors in regulating higher CH4 emission (i.e., over 50 mgC m−2 day−1) in the control and wet manipulated sections, when soil moisture reached to be saturated (see Fig. 3b and c). The average water table in the wet manipulation section, which is a significant factor in controlling CH4 emission (Verville et al., 1998; Strack and Waddington, 2007; Zona et al., 2009), was significantly higher than those in the dry and control sections (Oechel, 2009). This water table drawdown manipulation experiment has also resulted in reduced CH4 emissions from the control sites (Moore and Roulet, 1993; Strack and
c) Wet
Fig. 3. Spatial variations in CH4 and CO2 fluxes in (a) dry, (b) control, and (c) wet treatments along each manipulation experiment tramline. Higher CH4 emission was produced when soil was saturated and showed deeper thaw depth.
388
Y. Kim / Science of the Total Environment 505 (2015) 385–389
b)
a)
Fig. 4. Responses from (a) CH4 and (b) CO2 fluxes to thaw depth (cm, TD), for which the equations show CH4 = 9.7 TD-260 (R2 = 0.59; p = 0.482) and CH4 = 11.3 TD-297 (R2 = 0.84; p = 0.003) under control (solid line) and wet (dashed) treatments, respectively, and CO2 = −0.62 TD + 18.8 (R2 = 0.28; p = 0.482) for dry treatment.
Waddington, 2007). Because this reduced the size of the anoxic zone and increased the oxic zone in the peat profile, the amount of CH4 produced has probably been reduced, while CH4 oxidation has increased, thereby diminishing CH4 emissions (Strack and Waddington, 2007). The response from CO2 flux in the dry section to thaw depth was lower (Fig. 4b), at which thaw depth explained 28% of CO2 variability, resulting from deeper thaw depth and the inundated manipulation of the section, as in control and wet sections. Zona et al. (2009) also demonstrated that CH4 emission tends to increase with a higher water table and subsequently deeper thaw depth in manipulation experiments within Alaskan Arctic tundra (see Fig. 6: Zona et al., 2009). The relationships between CH4 flux in these three manipulated sections as well as CO2 flux in the dry and control sections are shown in Fig. 5. CO2 flux in the dry site has a weakly linear relation to CH4 fluxes at the control site (Fig. 5a). Also, CO2 flux at the control site has linear relations to CH4 fluxes in both the dry and wet manipulated sites (Fig. 5b). Average CO2 flux in the control manipulation corresponded to 79 and 27% of the variability of average CH4 emission in the dry and wet sections, indicating that CO2 emissions in these two sections were enhanced with drier and wetter conditions (Verville et al., 1998), respectively. On the other hand, average CH4 emission showed increasing patterns with thaw depth and soil moisture, and indicated an increase of 50 and 123% in the control and wet sections compared to the dry treated section, respectively. Thus, this feature is thought likely to represent a difference in the production mechanisms of CO2 and CH4 under the three manipulation treated sections by this water table controlling experiment in the Alaska coastal plain tundra during autumn. To estimate the response from temperature dependence on CO2 and CH4 fluxes, the relationship was plotted, showing exponential curves of soil temperature at 5-cm depth from the equation (not shown): effluxes of CO2 and CH4 = β0 × eβ1 × T, where fluxes of CO2 and CH4 are the measured fluxes of CO2 (gC m−2 day−1) and CH4 (mgC m−2 day−1); T is soil temperature (°C); and β0 and β1 are constants. This exponential
a)
b)
relationship is commonly used to represent carbon flux as a function of temperature (Davidson et al., 1998; Davidson and Janssens, 2006; Kim et al., 2007, 2013). Q10 temperature coefficient values were calculated as in Davidson et al. (1998) and Kim et al. (2013): Q10 = e β1 × 10, where Q10 is a measure of the change in biochemical reaction rate at intervals of 10 °C and is based on Van't Hoff's empirical rule that a rate increase of two to three times occurs for every 10 °C rise in temperature (Lloyd and Taylor, 1994). Q10 values were 2.10 (dry), 2.06 (control), and 107 (wet) for CO2 flux; and 0.24 (dry), 0.01 (control), and 2.35 × 105 (wet) for CH4 flux, respectively. Interestingly, CH4 production rate in the flooded manipulation section is stimulated to a much greater extent than CO2 production within a narrower soil temperature range of 2.3 to 4.7 °C. This demonstrates the dominant contribution of methanogens to methane production under an anoxic environment of flooded sections during the autumn period. Further, a much higher methane concentration from trapped air bubbles may be related to the enhanced methanogenesis (Walter et al., 2008; Sturtevant et al., 2012). 4. Conclusions and implications This water table manipulation experiment resulted in distinct drained (dry), intermediate (control), and flooded (wet) treatment tramlines (Oechel, 2009; Zona et al., 2009), and the subsequent magnitude of CO2 and CH4 releases is likely thought to modulate for different thaw depths, depending on the water table, in the three sections. Higher CH4 emission occurred at each section when soil was submerged. Contrary to CH4 emission, however, CO2 emission was highest in the dry treatment section, compared to the control and wet sections. This suggests a different production mechanism for CO2 and CH4 under these water table manipulation sections. The isotopic discrimination of CH4 from trapped bubble air was similar to isotopic characteristics of ebullition events in Alaska tundra lakes (Walter et al., 2008). This study therefore found that the change in thaw depth under water table
c)
Fig. 5. Responses of (a) CH4 flux under control for CO2 flux under dry treatment, and of CH4 flux under (b) dry and (c) wet treatments for CO2 flux under control treatment.
Y. Kim / Science of the Total Environment 505 (2015) 385–389
manipulation affected production processes of CO2 and CH4, with thaw depth as a significant key in regulating CO2 and CH4, compared to other environmental factors during autumn. I have thusly figured the global warming potential (GWP) of these gases at these three particular treated sections throughout 61 days of September and October (autumn) by considering a 100-year timescale, and weighing the warming potential of CH4 at 25 times that of CO2 (IPCC, 2007). The quotients of GWPCO2 (dry/control) and GWPCH4 (wet/control) increased by 464 and 148%, respectively, and GWPCH4 (dry/control) declined by 66%. This suggests that CO2 emission from the drained section is thought to be enhanced further by soil and ecosystem respiration compared to the wet section, and CH4 emission in the wet area is likely thought to be stimulated under an anoxic environment by saturated standing water, in relation to other sections. On the other hand, CH4 emission in the drained area is constrained by relatively shallower thaw depth, which is CH4 oxidation (Strack and Waddington, 2007). Thus, the change in thaw depth in manipulated Arctic coastal tundra is important in modulating production mechanisms of CO2 and CH4 for the regional carbon balance, as is the potential for shallower thaw depth to enhance CO2 emission in the drained area and for deeper thaw depth to boost CH4 emission in the inundated area. Therefore, the expansion of tundra lakes implies the enhancement of CH4 emission (Walter et al., 2006); on the other hand, the shrinking and disappearing thermokarst lakes lead to the stimulated CO2 release (Yoshikawa and Hinzman, 2003; Smith et al., 2005) response to the recent changing climate in the Arctic. Acknowledgments This research was conducted under the JAMSTEC-IARC Collaboration Study with funding provided by the Japan Agency for Marine-Earth Science and Technology (JAMSTEC) under a grant to the International Arctic Research Center (IARC). The author thanks Dr. W. Oechel of San Diego State University for his research in the manipulation experiment of the BEO, and Mr. N. Bauer of the International Arctic Research Center (IARC) of the University of Alaska Fairbanks (UAF) for a review of the text as a professional English-language editor. Finally, I appreciate Dr. P. Kassomenos (Handling Editor) and an anonymous reviewer for beneficial comments. References Bockheim JG, Everett LR, Hinkel KM, Nelson FE, Barown J. Soil organic carbon storage and distribution in arctic tundra, Barrow Alaska. Soil Sci Soc Am J 1999;63:934–40. Davidson EA, Jassens IA. Temperature sensitivity of soil carbon decomposition and feedback to climate change. Nature 2006;440:165–73. Davidson EA, Belk E, Boone RD. Soil water content and temperature as independent or confounded factors controlling soil respiration in a temperate mixed hardwood forest. Global Change Biol 1998;4:217–27. Duddleston KN, Kinney MA, Kiene RP, Hines ME. Anaerobic microbial biogeochemistry in a northern bog: acetate as a dominant metallic end product. Global Biogeochem Cycles 2002;16:1063. http://dx.doi.org/10.1029/2001GB001402. Fisher R, Lowry D, Wilkin O, Sriskantharajah S, Nsbet EG. High-precision, automated stable isotope analysis of atmospheric methane and carbon dioxide using continuousflow isotope-ratio mass spectrometry. Rapid Commun Mass Spectrom 2006;20: 200–8. Grosse G, Harden J, Turetsky M, McGuire AD, Camill P, Tarnocai C, et al. Vulnerability of high-latitude soil organic carbon in North America to disturbance. J Geophys Res 2011;116:G00K06. http://dx.doi.org/10.1029/2010JG001507. Intergovernmental Panel on Climate Change (IPCC). Climate Change 2007: the physical science basis. New York, USA: Cambridge University Press; 2007. p. 996.
389
Kim Y, Tanaka N. Effect of forest fire on the fluxes of CO2, CH4, and N2O in boreal forest soils, interior Alaska. J Geophys Res 2003;108:8154. http://dx.doi.org/10.1029/ 2001JD000663. Kim Y, Ueyama M, Nakagawa F, Tsunogai U, Harazono Y, Tanaka N. Assessment of winter fluxes of CO2 and CH4 in boreal forest soils of central Alaska estimated by the profile method and the chamber method: a diagnosis of methane emission and implications for the regional carbon budget. Tellus 2007;59B:223–33. Kim Y, Kim SD, Enomoto H, Kushida K, Kondoh M, Uchida M. Latitudinal distribution of soil CO2 efflux and temperature along the Dalton Highway, Alaska. Polar Sci 2013; 7:162–73. Lloyd J, Taylor JA. On the temperature dependence of soil respiration. Funct Ecol 1994;8: 315–23. Mastepanov M, Sigsggard C, Dlugokencky EJ, Houweling S, Ström L, Tamstorf MP, et al. Large tundra methane burst during onset of freezing. Nature 2008;456:628–31. http://dx.doi.org/10.1038/nature07464. Miller JB, Mack KA, Dissly R, White JWC, Dlugokencky EJ, Tans PP. Development of analytical methods and measurements of 13C/12C in atmospheric CH4 from the NOAA Climate Monitoring and Diagnostic Laboratory Global Air Sampling Network. J Geophys Res 2002:107. http://dx.doi.org/10.1029/2001JD000630. Moore TR, Roulet NT. Methane flux: water table relations in Northern wetland. Geophys Res Lett 1993;20:587–90. Morimoto S, Aoki S, Nakazawa T, Yamanouchi T. Temporal variations of the carbon isotopic ratio of atmospheric methane observed at Ny Ålesund, Svalbard from 1996 to 2004. Geophys Res Lett 2006;33:L01807. http://dx.doi.org/10.1029/2005GL024648. Oechel WC. Ecosystem interactions. In: BEO subcommittee, Barrow Arctic Science Consortium for the Ukpeaġvik Iñupiat Corporation and North Slope Borough, editors. Barrow Environmental Observatory (Fifth annual report); 2009. p. 33–59. Osterkamp TE, Romanovsky VE. Freezing of the active layer on coastal plain of the Alaskan Arctic. Permafr Periglac Process 1997;8:23–44. Raynolds MK, Walker DA, Maier HA. Plant community-level mapping of arctic Alaska based on the Circumpolar Arctic Vegetation Map. Phytocoenologia 2005;35:821–48. http://dx.doi.org/10.11270340-269x/2005/0035-0821. Rice AL, Gotoh A, Ajie HO, Tyler SC. High-precision continuous-flow measurement of δ13C and δD of atmospheric CH4. Anal Chem 2001;73:4104–10. Romanovsky VE, Osterkamp TE. Thawing of the active layer on coastal plain of the Alaskan Arctic. Permafr Periglac Process 1997;8:1–22. Sachs T, Giebels M, Boike J, Kutzbach L. Environmental controls on CH4 emission from polygonal tundra on the microsite scale in the Lena river delta, Siberia. Global Change Biol 2010;16:3096–110. http://dx.doi.org/10.111/j.1365-2486.2010.02232.x. Smith LC, Sheng Y, MacDonal GM, Hinzman LD. Disappearing Arctic lakes. Science 2005. http://dx.doi.org/10.1126/science.1108142. Strack M, Waddington JM. Response of peatland carbon dioxide methane fluxes to a water table drawdown experiment. Global Biogeochem Cycles 2007;21:GB1007. http://dx.doi.org/10.1029/2006GB002715. Ström L, Ekberg A, Mastepanov M, Christensen TR. The effect of vascular plants on carbon turnover and methane emissions from a tundra wetland. Global Change Biol 2003;9: 1185–92. Sturtevant CS, Oechel WC, Zona D, Kim Y, Emerson CE. Soil moisture control over autumn season methane flux, Arctic Coastal Plain of Alaska. Biogeosciences 2012;9:1423–40. Tarnocai C, Canadell JG, Schuur EAG, Kuhry P, Mazhitova G, Zimov S. Soil organic carbon pools in the northern circumpolar permafrost region. Global Biogeochem Cycles 2009;23:GB2023. http://dx.doi.org/10.1029/2008GB003327. Umezawa T, Aoki S, Nakazawa T, Morimoto S. A high-precision measurement system for carbon and hydrogen isotopic ratio of atmospheric methane and its application to air samples collected in the Western Pacific region. J Meteorol Soc Jpn 2009;87:365–79. Verville JH, Hobbie SE, Chapin III FS, Hooper DU. Response of tundra CH4 and CO2 flux to manipulation of temperature and vegetation. Biogeochemistry 1998;41:215–35. Walter KM, Zimov SA, Chanton JP, Verbyla D, Chapin III FS. Methane bubbling from Siberian thaw lakes as a positive feedback to climate warming. Nature 2006;443. http:// dx.doi.org/10.1038/nature05040. Walter KM, Chaton JP, Chapin III FS, Schuur EAG, Zimov SA. Methane production and bubble emission from arctic lakes: isotopic implications for source pathways and ages. J Geophys Res 2008;113:G00A08. http://dx.doi.org/10.1029/2007JG000569. Yamada K, Ozaki Y, Nakagawa F, Tanaka M, Yoshida N. An improved method for measurement of the hydrogen isotope ratio of atmospheric methane and its application to a Japanese urban atmosphere. Atmos Environ 2003;37:1975–82. Yoshikawa K, Hinzman LD. Shrinking thermokarst ponds and groundwater dynamics in discontinuous permafrost. Permafr Periglac Process 2003;14(2):151–60. Zona D, Oechel WC, Kochendorfer J, Paw KT, Salyuk AN, Olivas PC, et al. Methane fluxes during the initiation of a large-scale water table manipulation experiment in the Alaskan Arctic tundra. Global Biogeochem Cycles 2009;23:GB2013. http://dx.doi. org/10.1029/2009GB003487.
