This content has been downloaded from IOPscience Please scroll down to see the full text

Download details

IP Address 462217386

This content was downloaded on 17102013 at 0101

Please note that terms and conditions apply

Large methane emission upon spring thaw from natural wetlands in the northern permafrost

region

View the table of contents for this issue or go to the journal homepage for more

2012 Environ Res Lett 7 034009

(httpiopscienceioporg1748-932673034009)

Home Search Collections Journals About Contact us My IOPscience

IOP PUBLISHING ENVIRONMENTAL RESEARCH LETTERS

Environ Res Lett 7 (2012) 034009 (8pp) doi1010881748-932673034009

Large methane emission upon springthaw from natural wetlands in thenorthern permafrost region

Changchun Song14 Xiaofeng Xu234 Xiaoxin Sun1 Hanqin Tian3Li Sun1 Yuqing Miao1 Xianwei Wang1 and Yuedong Guo1

1 Northeast Institute of Geography and Agroecology Chinese Academy of Sciences Changchun130012 Peoplersquos Republic of China2 Climate Change Science Institute and Environmental Science Division Oak Ridge NationalLaboratory Oak Ridge TN 37831 USA3 International Center for Climate and Global Change Research and School of Forestry and WildlifeSciences Auburn University Auburn AL 36849 USA

E-mail songccmailneigaeaccn and xux4ornlgov

Received 21 May 2012Accepted for publication 25 June 2012Published 19 July 2012Online at stacksioporgERL7034009

AbstractThe permafrost carbonndashclimate feedback is one of the major mechanisms in controlling theclimatendashecosystem interactions in northern high latitudes Of this feedback methane (CH4) emissionfrom natural wetlands is critically important due to its high warming potential The freezendashthawtransition has been confirmed to play an important role in annual CH4 budget yet the magnitude of thiseffect is uncertain An intensive field campaign was carried out in the Sanjiang Plain Northeast Chinato estimate the CH4 emission in the spring freezendashthaw transition period The observation concludedthat a large CH4 source was caused by spring thaw the maximum hourly emission rate was486 g C mminus2 hminus1 more than three orders of the regularly observed CH4 emission rate in the growingseason In some sporadically observed lsquohot spotsrsquo the spring thawing effect contributed to a large CH4source of 313 plusmn 101 g C mminus2 which is approximately 80 of the previously calculated annual CH4emission in the same study area If our results are typical for natural wetlands in the NorthernHemisphere permafrost region we estimate a global CH4 source strength of 05ndash10 Tg C (1 Tg= 1012 g) caused by spring thaw in the Northern Hemisphere permafrost region in the year 2011Combining with available satellite and flask data a regional extrapolation reaches a temporal pattern ofCH4 emission during 2003ndash2009 which is consistent with recently observed changes in atmosphericCH4 concentration in the high latitudes This suggests that the CH4 emission upon spring thaw in thehigh latitudes might be enhanced by the projected climate warming These findings indicate that thespring thawing effect is an important mechanism in the permafrost carbonndashclimate feedback and needsto be incorporated in Earth system models

Keywords carbon-climate feedback methane natural wetland permafrost

S Online supplementary data available from stacksioporgERL7034009mmedia

Content from this work may be used under the termsof the Creative Commons Attribution-NonCommercial-

ShareAlike 30 licence Any further distribution of this work must maintainattribution to the author(s) and the title of the work journal citation and DOI

4 These authors have equal contributions to this research

1 Introduction

As one of the potent greenhouse gases methane (CH4) playsan important role in terrestrial ecosystemndashclimate interaction(Bousquet et al 2006 Keppler et al 2006 Mastepanov et al

11748-932612034009+08$3300 ccopy 2012 IOP Publishing Ltd Printed in the UK

Environ Res Lett 7 (2012) 034009 C Song et al

Figure 1 (A) The spatial distribution of the permafrost in the Northern Hemisphere (the red star indicates the location of the fieldcampaign) and (B) the distribution of the sites with identified CH4 emission pulses across the experimental field in the Sanjiang Plain MireStation (In total twenty sites with high CH4 emission pulses were identified S21 S22 and S23 are the observational sites for our growingseason monitoring reported in our previous publication (Song et al 2009) it should be noted that the distance and location of each site mightbe not accurate since most of the distances were estimated without instruments)

2008) especially under the changing global environment(Isaksen et al 2011 Walter et al 2006 Xu and Tian 2012Xu et al 2010) The recently observed increase in atmosphericCH4 concentration has drawn a large amount of attention toCH4 emission from natural wetlands in northern high latitudes(Dlugokencky et al 2009) The northern permafrost region isessential in global CH4 cycling especially over the trajectoryof global warming (Christensen et al 2004) because the densecarbon storage in permafrost soils is a potential source ofatmospheric CH4 (Christensen et al 2004 Koven et al 2011Schuur et al 2009) and is predicted to release more and moreCH4 to the atmosphere by 2100 (Anisimov 2007 Koven et al2011) even though the winter season had been previouslythought to be a period with a weak CH4 emission before anumber of field studies confirmed that the biological processesin winter directly make relatively large contributions to theannual CH4 budget (Campbell et al 2005 Hao et al 2006)

Given the isolation of the frozen surface in the permafrostregion a large portion of the produced CH4 cannot be releasedto the atmosphere immediately (Tokida et al 2007) Duringthe thawing season in the following year the release ofCH4 is expected to occur in a very short period formingan emission outburst (Christensen et al 2004) This effect isdefined as the spring thawing effect which might partiallycontribute to the recently observed increase in atmosphericCH4 concentration (Dlugokencky et al 2009 Tokida et al2007) Although the spring thawing effect on atmosphericCH4 has been recognized for a long time (Friborg et al1997) the magnitude of the spring thawing effect on CH4

emission remains far from certain (Friborg et al 1997 Tokidaet al 2007) This study was conducted to estimate the effectsof the spring thaw on CH4 release from natural wetlandsin the permafrost region in the Northern Hemisphere byusing a combination of field observations satellite data andflask data

Figure 2 Observed daily CH4 flux in the Sanjiang station (476N1335E) Northeast China (the red line shows the captured CH4pulse in the spring thawing period the black line shows thenormally observed CH4 flux in the study area four periods aredivided based on local meteorological conditions the fluxes areshown as daily average and hourly results are listed in table 1)

2 Data and methods

A combination of field observations NOAA (NationalOceanic and Atmospheric Administration) flask data(GLOBALVIEW-CH4) and a satellite-derived CH4 concen-tration dataset (SCIAMACHY Scanning Imaging AbsorptionSpectrometer for Atmospheric Chartography) were used inthis study

21 Field observations

We conducted an intensive field campaign in the SanjiangPlain Northeast China to monitor seasonal variations inCH4 emission from natural wetlands and estimate the CH4emission rate caused by spring thaw at site level (figures 12 and S3 available at stacksioporgERL7034009mmedia)The Sanjiang Plain is in the largest freshwater wetland area

2

Environ Res Lett 7 (2012) 034009 C Song et al

Table 1 Observed CH4 fluxes in the natural freshwater marshland in the Sanjiang Plain from 6 April to 11 April 2011 (The bold indicatesthe largest CH4 emission rate observed in this field campaign) Note in total 23 high CH4 emission pulses were measured More than onemeasurement might be carried out for one single site For example H1-1 and H1-2 are two samplings and measurements for one single CH4emission pulse

462011 Site H1-1 H1-2 H2 H3 H4 H5Observed flux (mg CH4-C mminus2 hminus1) 1937 27 11799 48346 180929 378590Duration (h) 2 2 25 3 2 2

472011 Site H9 H10-1 H10-2 H10-3 H12 H13Observed flux (mg CH4-C mminus2 hminus1) 3450 95034 485600 118017 1876 623

Duration (h) 15 15 15 35 15 15482011 Site H9-2 H16-1 H16-2 H16-3 H17 H18

Observed flux (mg CH4-C mminus2 hminus1) 422 17972 44259 464 12836 85601Duration (h) 24 24 24 3 2 2

492011 Site H19-1 H19-2Observed flux (mg CH4-C mminus2 hminus1) 968 969Duration (h) 1 1

4102011 Site H19-3 H19-4Observed flux (mg CH4-C mminus2 hminus1) 1337 989Duration (h) 1 1

4112011 Site H19-5Observed flux (mg CH4-C mminus2 hminus1) 322Duration (h) 1

in China there are approximately 10 400 km2 of naturalwetlands in this region (Song et al 2009) The study sitesare located at the Sanjiang Experimental Station of WetlandEcology Chinese Academy of Sciences (4735primeN 13331primeE)at an altitude representative of the natural freshwater wetlandin the Sanjiang Plain (56 m asl) This region is located atthe southern boundary of Northern Hemisphere permafrost(Qiu et al 2002) Since the CH4 emission pulses usuallyoccur in a very short time period (Tokida et al 2007)our previous observations had not captured this event Inthe year 2011 we increased the observational frequencyand broadened the monitoring area Fortunately the CH4emission pulses were captured (figures 1 and 2 table 1) Theobserved CH4 emission pulse can be viewed in our onlinesupporting video clip (video S1 available at stacksioporgERL7034009mmedia)

Across our sampling area twenty sites with high CH4emission pulses were identified (figure 1) of which thirteenwere measured at least once by using a static chamber method(Song et al 2009) We used 5ndash7 persons walking throughthe area to identify the potential bubbling spots after eachbubbling spot was identified we measured the flux rate iflabor resources allowed Measurements for several spots weremissed due to shortage of labor we recorded the locationsof these spots Twenty-three measurements were conductedfor CH4 emission pulses (table 1) There are fourteen highemission plots in a 30 mtimes 30 m study area the measurementsdefine CH4 emission from a 05 m times 05 m plot (figure 1 andtable 1) So we estimated that approximately 039 of the areahad high CH4 pulses This fractional coverage of high CH4emission was then used to estimate the spring thawing effectson CH4 emission at regional scale

The gas flux measurements were conducted by usingstatic dark chamber and gas chromatography techniques(Song et al 2009) The 05 m times 05 m times 05 m chamber

was put on the ground which was covered by 5ndash10 cm surfacewater as sealer A fan was installed in the sampling chamberto keep the air mixed during the gas sampling Since theemission rate was quite large the gas sampling process wascarried out four times with a 5 min interval rather than witha 10 min interval as in previous studies (Song et al 2009Xu et al 2005) The gas samples were stored in syringes forless than 12 h before being measured Gas chromatographywas used to measure the gas concentrations The gradient ofgas concentration during sampling was used to calculate thegas flux between the ecosystem and the atmosphere Positivevalues mean flux from the ecosystem to the atmosphereand negative values mean flux from the atmosphere to theecosystem (Song et al 2009) The mixing ratios of CH4 wereanalyzed with a modified gas chromatograph (Agilent 4890D)equipped with a flame ionization detector (FID) N2 was usedas the carrier gas with a flow rate of 30 ml minminus1 The CH4was separated with a 2 m stainless-steel 13 XMS column(6080 mesh) with an inner diameter of 2 mm CH4 wasdirectly measured by FID with an operation temperature of200 C The temperature for gas separation was 55 C Fluxeswere determined from the slope of the mixing ratio changesin four samples taken at 0 5 10 and 15 min after chamberclosure Sample sets were rejected unless they yielded a linearregression value of r2 greater than 09

The gas flux was calculated according to the followingequation from Song et al (2009)

F =dc

dt

M

V0

P

P0

T0

THc (1)

Here F is the CH4 flux (mg C mminus2 hminus1) dcdt is the slopeof the linear regression for the gas concentration gradientwith time (mol molminus1 hminus1) M is the molecular mass of eachgas (mg molminus1) P is the atmospheric pressure (Pa) T is theabsolute temperature during sampling (K) V0 T0 and P0 are

3

Environ Res Lett 7 (2012) 034009 C Song et al

the gas mole volume (m3 molminus1) absolute air temperature(K) and atmospheric pressure under standard conditions (Pa)respectively Hc is the height of the chamber during sampling(m)

22 Satellite and flask data

The satellite data (available at httpwwwsciamachyorg)were used to estimate the potential area with high CH4emission caused by spring thawing effects Combining theestimate of high CH4 emission area with the CH4 emissionrate derived from field observational data we provided alarge-scale estimation of the spring thawing effects on CH4emission from natural wetlands in the northern permafrostregion Flask data provided by NOAA were used to verify theincrease in atmospheric CH4 concentration over the study areaand to compare with the regional estimate of the CH4 emissionfrom natural wetlands in the northern permafrost regionThe GLOBALVIEW-CH4 Cooperative Atmospheric DataIntegration Project methane available at ftpftpcmdlnoaagovccgch4GLOBALVIEW was used in this study with thelast access on 16 June 2011 GLOBALVIEW-CH4 is a productof the Cooperative Atmospheric Data Integration ProjectWhile the project is coordinated and maintained by the CarbonCycle Greenhouse Gases Group of the National Oceanicand Atmospheric Administration Earth System ResearchLaboratory (NOAA ESRL) it is a cooperative effort amongthe many organizations and institutions making high-qualityatmospheric CH4 measurements In this study three siteslocated in the northern permafrost region were selected (tableS1 available at stacksioporgERL7034009mmedia) theannual CH4 concentrations measured at the three sites areshown in figure S4 (available at stacksioporgERL7034009mmedia)

23 Auxiliary data

The regional permafrost data were from the National Snowand Ice Data Center at httpnsidcorgdataggd318htmlThe circumpolar permafrost and ground ice data depict thedistribution and properties of permafrost and ground ice in theNorthern Hemisphere (20Nndash90N) The data set classifiespermafrost into four categories continuous discontinuoussporadic and isolated The fractions of permafrost in eachcategory are 90ndash100 50ndash90 10ndash50 lt10 Therelative abundance of ground ice in the upper 20 m isestimated in per cent volume (gt20 10ndash20 lt10 and0) (Brown et al 1998) The data set also contains the locationof subsea and relict permafrost The fractional distributionof natural wetlands was retrieved from and improved on thebasis of the data from Aselmann and Crutzen (1989) Finallywe generated the natural wetlands distribution in the northernpermafrost region by overlaying them in the same spatialdomain (figures S1 and S2 available at stacksioporgERL7034009mmedia)

24 Spatial extrapolation of spring thaw-induced CH4emission

We processed the SCIAMACHY (satellite-derived methaneconcentration in air volume) data to define the grids whichhave potential high CH4 emission (Bergamaschi et al 2009Frankenberg et al 2011 Schneising et al 2011) (figureS6 available at stacksioporgERL7034009mmedia) Dueto technical problems there are a lot of spatial gaps inthe CH4 concentration derived from the SCIAMACHYdata (figure S6 available at stacksioporgERL7034009mmedia) especially in the period of 2005ndash9 (Bergamaschiet al 2007 2009 Frankenberg et al 2011 Schneisinget al 2011) The percentage of the area with high CH4emission was calculated (figure S8 available at stacksioporgERL7034009mmedia) Then we combined the estimatedpercentage area of the lsquohot spotsrsquo of CH4 emission withthe spatial distribution of the natural wetlands across thenorthern permafrost region to estimate the spring thawingeffect on atmospheric CH4 concentration The criteria weused to identify potential high CH4 pulses based on satellitewere (1) the CH4 concentration in April is 100 ppb higherthan that in May (2) the estimated CH4 concentration inApril is higher than those in any other months (non-April)at the significance level of P = 005 The value of 100 ppbwas used because the uncertainties of the CH4 concentrationretrieved from the SCIAMACHY data were approximately50 ppb (Frankenberg et al 2011 Schneising et al 2011) sothe threshold of sufficient difference should be approximately100 ppb on the basis of normal distribution Based on thedelineation of the spatial coverage of the area with potentialhigh CH4 emission we further calculated the contributionof the spring thawing effect to the variations of atmosphericCH4 concentration (table S2 available at stacksioporgERL7034009mmedia)

25 Satellite data in capturing spring thaw-induced CH4emission

To evaluate the satellite data in capturing the spring thaweffects on CH4 emission we analyzed the basin-level CH4concentration derived from satellite data (figure S5 availableat stacksioporgERL7034009mmedia) The higher AprilCH4 concentration was observed for a few years in Ob Riverbasin Lena River basin Amur River basin and MackenzieRiver basin (figure S7 available at stacksioporgERL7034009mmedia) We used the 1CH4 (CH4 concentration inApril minus that in May) to quantify the magnitude of springthaw effects on CH4 emission After 2006 a strong impactis shown as the 1CH4 became larger and larger in Ob Riverbasin Amur River basin and Mackenzie River basin (figureS7 available at stacksioporgERL7034009mmedia)

3 Results and discussion

31 Observed high CH4 emission pulses

The field observations showed that the CH4 emission pulsesoccurred sporadically in some lsquohot spotsrsquo the hourly emission

4

Environ Res Lett 7 (2012) 034009 C Song et al

Figure 3 Air temperature and soil temperature from January 2010to April 2011 ((A) daily air and soil surface temperature over thetime period (B) temperature along the soil profile over the timeperiod the red rectangle indicates the time period of 6ndash8 April)

rate could be as high as 486 g C mminus2 hminus1 via bubblingemission (figure 2 table 1) which is almost three orders ofthe previously observed flux rate in the growing season (Songet al 2009) The CH4 emission pulse lasts for a very short timeperiod it could be a few hours for each outburst event andthree days for the study region The observed CH4 bubblingin the Sanjiang Plain lasted from 6 to 8 April and after thatthe CH4 emission rate decreased to the normal emission ratefor the spring season (figure 2) meanwhile the observed CH4outburst during 9ndash11 April was from one site only thus weassumed that the time period of observed CH4 emission pulseswas 6ndash8 April The total CH4 emission via bubbling from 6April to 8 April was summed to 313 plusmn 101 g C mminus2

In order to verify the effects of spring thaw on theobserved CH4 emission pulse we analyzed the temporalvariations of air and soil temperatures in the study area(figure 3) The temperature change along the soil profile (upto 1 m depth) showed that the surface 20 cm soil profile wascompletely frozen from mid-November 2010 to early April of2011 (figure 3) The soil profile completely thawed during 6ndash8April which exactly matches the time period of observed CH4emission pulse (figures 2 and 3) Therefore it is concludedthat the observed CH4 emission pulses were associated withthe spring thaw along the soil profile This mechanism forCH4 emission is very similar to that of nitrous oxide duringspring thaw as reported by a field study (Christensen andTiedje 1990) The CH4 produced beneath ice in winter isstored in the soil porositywater only a small portion of itis oxidized (Mer and Roger 2001 Roslev and King 1996)when the frozen surface is completely thawed in the followingspring the stored CH4 is released to the atmosphere in a veryshort time period normally in hours as observed in this studyGiven the large spatial heterogeneity of soil and vegetationthe soil thawing and associated CH4 emission pulse over thestudy area usually occur episodically and in some lsquohot spotsrsquoas observed (figure 1 and table 1)

The high CH4 emission rate in the spring thawing periodmight be caused by three reasons winter CH4 production(Dise 1992 Melloh and Crill 1996) lack of winter CH4

oxidation beneath the ice due to icersquos isolation of atmosphericoxygen (Melloh and Crill 1995) and the outburst of CH4emission (Tokida et al 2007) A number of studies have foundthat the CH4 emissions in the winter season are very important(Hao et al 2006 Song et al 2009) A few studies have reportedthat the winter season could contribute 232ndash462 to theannual budget of CH4 (Hao et al 2006) One study evenconcluded a 21 contribution of winter flux to annual CH4budget (Dise 1992) Based on the calculated temperaturesensitivity of CH4 emission in the same study area in ourprevious study (Song et al 2009) we estimated that the CH4emission in the winter season contributes 18 of the annualbudget After translating to unit emission the average CH4emission rate in winter is approximately 23 of that in thegrowing season in the Sanjiang Plain Relatively low CH4oxidation in winter also contributes to the observed CH4outbursts The frozen surface isolates atmospheric oxygen andimpedes oxidation of the produced soil CH4 Therefore alarge amount of CH4 is stored before release Finally the shorttime window allows the CH4 emission to form the observedpulses

32 Quantification of the spring thawing effect on CH4emission

For the reasons mentioned in section 31 high CH4 emissionpulses are expected To estimate the spring thawing effect onCH4 emission we used the following calculation to quantifythe spring thawing effect on CH4 emission

Fg = (1minus Oxidg)q0HLgQ10Tg10 (2)

Fw = (1minus Oxidw)q0HLwQ10Tw10 (3)

Fog =Fg

Dg(4)

Fow =Fw

Dw(5)

FSTE =Fow

Fog(6)

where Fg is the total CH4 emission in the growing seasonand Oxidg is the fraction of produced CH4 being oxidized inthe growing season q0 is the potential CH4 production rateat 0 C H is the depth of the soil profile for CH4 productionLg is the duration of CH4 production in the growing seasonQ10 is the temperature sensitivity of CH4 production Tg isthe average temperature in the soil profile during the growingseason Fw is the total CH4 emission in the winter seasonand Oxidw is the fraction of produced CH4 being oxidized inthe winter season Lw is the duration of CH4 production inthe winter season Tw is the average temperature in the soilprofile during the winter season Fog and Dg are the observedCH4 emission rate and duration of the CH4 release in thegrowing season Fow and Dw are the observed CH4 emissionrate and duration of the CH4 release in the winter season andFSTE dimensionless is the magnitude of spring effects onCH4 emission The values for all parameter are provided intable 2

5

Environ Res Lett 7 (2012) 034009 C Song et al

Table 2 Parameters used in the calculation of the potential CH4 emission in the spring thaw and their ecological meanings (All parametersare for the Sanjiang Plain Northeast China)

Parameter Value (unit) Ecological meaning Reference

Tw 027 (C) Mean temperature of the soil profile (0ndash50 cm) inthe winter season

This study

Tg 1616 (C) Mean temperature of the soil profile (0ndash50 cm) inthe growing season

This study

Q10 25 Temperature sensitivity of methane production inthe study site

Song et al (2009)

Oxidw 01 Fraction of produced CH4 in the soil porosityoxidized in the winter season

Mer and Roger (2001) Roslev and King (1996)

Oxidg 07 Fraction of produced CH4 in the soil porosityoxidized in the growing season

Conrad (1996) Mer and Roger (2001)

Dw 2ndash12 (h) Duration of the CH4 outburst during the springthawing period

This study

Lw 2904 (h) Length of the winter season This studyLg = Dg 3672 (h) Length and duration of CH4 production and

emission over the growing seasonThis study

It is reported that oxidation could consume up to 90of produced CH4 in the soil before releasing it to theatmosphere (Oremland and Culbertson 1992) A field studyin a continuously flooded rice paddy system concluded on a70 oxidation of produced CH4 (Mer and Roger 2001) anda study on wetland found a 72 oxidation of produced CH4(Freeman et al 2002) Thus we used 70 as the fraction ofCH4 oxidation in soils in the growing season It is reportedthat the oxidation of produced CH4 in soil is highly dependenton oxygen (King 1990) Meanwhile a field study reportedwinter oxidation as approximately 5ndash15 of that whichoccurs in summer (Chanton and Liptay 2000) therefore weused 10 as the fraction of CH4 oxidation in soilwater in thewinter season (table 2) and two hours as the normal durationof CH4 emission during the thawing period (table 1) Thetemperature sensitivity dependence of CH4 production wasset as Q10 = 25 as reported in our previous study (Song et al2009) The growing season at the Sanjiang Plain lasts from1 May to 30 September and the winter season the frozenseason referred in this study lasts from 1 December to 31March the following year Thus the length of the winter seasonis 121 days FSTE represents the relative magnitude of hourlyCH4 emission rate during the spring thaw compared to thatin the growing season The calculation suggests the springthawing effect could be as high as 167ndash1002 times The CH4emission rate during spring thawing could be more than threeorders of the growing season CH4 emission at its peak Thisis consistent with our field observations (table 1)

33 Spatial extrapolation of spring thaw-induced CH4emission

Combining these site observations with other data we furtherestimated the spring thawing effects on CH4 emission fromthe natural wetlands across the entire northern permafrostregion by using two methods (figures S1 and S2 availableat stacksioporgERL7034009mmedia) Firstly based onthe areal representativeness of each observed CH4 pulse inthe field campaign we estimate that approximately 039 ofthe natural wetland served as outlets for the stored CH4 in

Figure 4 Temporal variation of year-to-year change in observedCH4 concentration and calculated CH4 emission from the springthawing effect (CH4 change is defined as changes in atmosphericCH4 concentration at yearly time steps derived from NOAA flaskdata STE spring thawing effect)

spring across the monitoring area (figure 1 section 21) If themeasurements and the areal percentages are representativewe estimate that the spring thawing effect was 05ndash97 Tg Cof CH4 from natural wetlands across the northern permafrostregion in the year 2011 (table S2 available at stacksioporgERL7034009mmedia)

Meanwhile we combined the time-series column CH4concentration derived from SCIAMACHY data with fieldobservations to estimate the temporal variation of springthaw-induced CH4 emission in the northern permafrostregion (figure 4) The SCIAMACHY data were partiallyverified in capturing CH4 concentration variations inducedby spring thaw (section 25) The results showed that springthaw-induced CH4 emission increased from 2003 through2008 and slightly decreased in 2009 which is consistent withtemporal changes in CH4 concentration derived from NOAAflask data except for the year 2003 (figure 4) The increasein CH4 concentration in the year 2003 could be attributed tobiomass burning in Boreal regions of Asia and North America(Dlugokencky et al 2009 van der Werf et al 2006) Weacknowledge that the spring thaw-induced CH4 emission has

6

Environ Res Lett 7 (2012) 034009 C Song et al

existed for quite a long time since it was observed more than adecade ago (Christensen and Tiedje 1990 Christensen et al2004) This study confirmed that the spring thawing effecthas been enhanced recently which is probably associatedwith climate warming as pointed out by several recent studies(Dlugokencky et al 2009 Frankenberg et al 2011)

The contribution from the enhanced spring thawing effectto the recent increase in atmospheric CH4 concentrationmight be an important positive feedback of terrestrialecosystemndashclimate system caused by recent warming in theArctic region (Phelps et al 1998 Tokida et al 2007) It mightalso serve as the first trigger of permafrost carbonndashclimatefeedback if so the positive feedback might take place earlierthan previously expected

34 Uncertainty and research needs

We acknowledge that there are some uncertainties thatneed to be addressed in our further research effort Theextrapolation might cause large uncertainties in the regionalestimate since we assume constant CH4 emission over entirenorth permafrost region To improve spatial extrapolationof CH4 emission it is clearly necessary to developaccurate spatial data on the distribution of the hot spotsof CH4 emission from natural wetlands in the northernpermafrost region Furthermore all the field observationswere conducted during the daytime and nighttime CH4emission pulses were not covered Thus the current estimatemight underestimate the CH4 emission resulting from springthawing effects Yet we argue that this study captures theprimary CH4 outburst because nighttime usually has lowertemperature and thaw events should not occur as oftenas in daytime Meanwhile high-frequency observations areneeded to accurately estimate the CH4 emission pulsesThe comparison between satellite data and flask data showspartial verification of SCIAMACHYrsquos capacity in capturingvariations in atmospheric CH4 concentration driven byspring thaw effect-induced increase in CH4 concentration(figure 4) In addition integration with an atmospherictransport chemistry model would be an improvement forevaluating the spring thawing effect on atmospheric CH4concentration

4 Concluding remarks

This study shows the dramatic contribution of the springthawing effect to atmospheric CH4 variations caused by recentArctic warming The extremely high CH4 emission observedin this study confirmed that natural wetland might also emita large amount of CH4 via bubbling in the spring seasonin addition to the observed CH4 emission from thaw lakes(Walter et al 2006) This high CH4 emission enhances thepositive feedback through greenhouse gases in the Articregion (Chapin et al 2005) Thus Earth system models shouldtake this spring thawing effect into consideration to makemore accurate examinations of permafrost carbonndashclimatefeedback We anticipate that the estimated contribution ofthe spring thawing effect to atmospheric CH4 will stimulate

further research into the large-scale feedback from terrestrialecosystems to the climate system Given the projected changein the climate system in the 21st century the spring thawingeffect might get stronger and stronger serving as one of themost important mechanisms for permafrost carbonndashclimatefeedback To better understand this feedback therefore moreefforts are needed to investigate the processes or factorsresponsible for the spring thawing effect on CH4 emission

Acknowledgments

This work has been supported by the National Natural ScienceFoundation of China (41125001 40930527) National KeyBasic Research and Development projects (2009CB421103)NASA IDS program (NNG04GM39C) and NASA LCLUCprogram (NNX08AL73G S01) This research was sponsoredin part by the US Department of Energy Office of ScienceBiological and Environmental Research (BER) programand performed at Oak Ridge National Laboratory (ORNL)ORNL is managed by UT-Battelle LLC for the USDepartment of Energy under contract DE-AC05-00OR22725We thank Dr Oliver Schneising at the University ofBremen FB1 Germany for providing the satellite data andsome constructive comments The authors are grateful toDrs Wilfred M Post and Peter E Thornton at Oak RidgeNational Laboratory for reviewing this manuscript

Notice

The United States Government retains and the publisher byaccepting the article for publication acknowledges that theUnited States Government retains a non-exclusive paid-upirrevocable world-wide license to publish or reproduce thepublished form of this manuscript or allow others to do sofor United States Government purposes

Author contributions

CS XX and HT initiated the effort XS LS YM XWand YG conducted the field observations XX performed theoverall analysis of field flask and satellite data All authorscontributed to interpreting the results and writing the paper

ccopy US Government

References

Anisimov O A 2007 Potential feedback of thawing permafrost to theglobal climate system through methane emission Environ ResLett 2 045016

Aselmann I and Crutzen P J 1989 Global distribution of naturalfreshwater wetlands and rice paddies their net primaryproductiity seasonality and possible methane emissionsJ Atmos Chem 8 307ndash58

Bergamaschi P et al 2007 Satellite chartography of atmosphericmethane from SCIAMACHY on board ENVISAT 2Evaluation based on inverse model simulations J GeophysRes 112 D02304

Bergamaschi P et al 2009 Inverse modeling of global and regionalCH4 emissions using SCIAMACHY satellite retrievalsJ Geophys Res 114 D22301

7

Environ Res Lett 7 (2012) 034009 C Song et al

Bousquet P Ciais P Miller J Dlugokencky E Hauglustaine DPrigent C Van der Werf G Peylin P Brunke E andCarouge C 2006 Contribution of anthropogenic and naturalsources to atmospheric methane variability Nature 443 439ndash43

Brown J Ferrians O J Heginbottom J A and Melnikov E S 1998Circum-arctic map of permafrost and ground ice conditionsReport (Boulder CO National Snow and Ice DataCenterWorld Data Center for Glaciology)

Campbell J Mitchell M Groffman P Christenson L andHardy J 2005 Winter in northeastern North America a criticalperiod for ecological processes Front Ecol Environ 3 314ndash22

Chanton J and Liptay K 2000 Seasonal variation in methaneoxidation in a landfill cover soil as determined by an in situstable isotope technique Glob Biogeochem Cycles 14 51ndash60

Chapin F S et al 2005 Role of land-surface changes in Arcticsummer warming Science 310 657ndash760

Christensen S and Tiedje J M 1990 Brief and vigorous N2Oproduction by soil at spring thaw Eur J Soil Sci 41 1ndash4

Christensen T R Johansson T Akerman H J andMastepanov M 2004 Thawing sub-arctic permafrost effects onvegetation and methane emissions Geophys Res Lett31 L04501

Conrad R 1996 Soil microorganisms as controllers of atmospherictrace gases (H2 CO CH4 OCS N2O and NO) MicrobiolRev 60 609ndash40

Dise N B 1992 Winter fluxes of methane from Minnesota peatlandsBiogeochemistry 17 71ndash83

Dlugokencky E J et al 2009 Observational constraints on recentincreases in the atmospheric CH4 burden Geophys Res Lett36 L18803

Frankenberg C Aben I Bergamaschi P Dlugokencky E Jvan Hees R Houweling S van der Meer P Snel R andTol P 2011 Global column-averaged methane mixing ratiofrom 2003 to 2009 as derived from SCIAMACHY trends andvariability J Geophys Res 116 D04302

Freeman C Nevison G B Kang H Hughes S Reynolds B andHudson J A 2002 Contrasted effects of simulated drought onthe production and oxidation of methane in a mid-Waleswetland Soil Biol Biochem 34 61ndash7

Friborg T Christensen T R and Segaard H 1997 Rapid response ofgreenhouse gas emission to early spring thaw in a subarcticmire as shown by micrometeorologial techniques GeophysRes Lett 24 3061ndash4

Hao Q J Wang Y S Song C C and Huang Y 2006 Contribution ofwinter fluxes to the annual CH4 CO2 and N2O emissions fromfreshwater marshes in the Sanjiang Plain J Environ Sci-China18 270ndash5

Isaksen I S A Gauss M Myhre G Anthony K M W andRuppel C 2011 Strong atmospheric chemistry feedback toclimate warming from Arctic methane emissions GlobBiogeochem Cycles 25 GB2002

Keppler F Hamilton J T G Brass M and Rockmann T 2006Methane emissions from terrestrial plants under aerobicconditions Nature 439 187ndash91

King G M 1990 Dynamics and controls of methane oxidation in aDanish wetland sediment FEMS Microbiol Lett 74 309ndash23

Koven C D Ringeval B Friedlingstein P Ciais P Cadule PKhvorostyanov D Krinner G and Tarnocai C 2011 Permafrost

carbon-climate feedbacks accelerate global warming ProcNatl Acad Sci 108 14769ndash74

Mastepanov M Sigsgaard C Dlugokencky E J Houweling SStrom L Tamstorf M P and Christensen T R 2008 Large tundramethane burst during onset of freezing Nature 456 628ndash30

Melloh R A and Crill P M 1995 Winter methane dynamics beneathice and in snow in a temperate poor fen Hydrol Process9 947ndash56

Melloh R A and Crill P M 1996 Winter methane dynamics in atemperate peatland Glob Biogeochem Cycles 10 247ndash54

Mer J L and Roger P 2001 Production oxidation emission andconsumption of methane by soils a review Eur J Soil Biol37 25ndash50

Oremland R S and Culbertson C W 1992 Importance ofmethane-oxidizing bacteria in the methane budget as revealedby the use of a specific inhibitor Nature 356 421ndash3

Phelps A R Peterson K M and Jeffries M O 1998 Methane effluxfrom high-latitude lakes during spring ice melt J GeophysRes 103 29029ndash36

Qiu G Zhou Y Guo D and Wang Y 2002 Maps of GeocryologicalRegions and Classifications ed T Zhang (Boulder CONational Snow and Ice Data CenterWorld Data Center forGlaciology)

Roslev P and King G M 1996 Regulation of methane oxidation in afreshwater wetland by water table changes and anoxia FEMSMicrobiol Ecol 19 105ndash15

Schneising O Buchwitz M Reuter M Heymann JBovensmann H and Burrows J P 2011 Long-term analysis ofcarbon dioxide and methane column-averaged mole fractionsretrieved from SCIAMANHY Atmos Chem Phys 11 2863ndash80

Schuur E A Vogel J G Crummer K G Lee H Sickman J O andOsterkamp T E 2009 The effect of permafrost thaw on oldcarbon release and net carbon exchange from tundra Nature459 556ndash9

Song C Xu X Tian H and Wang Y 2009 Ecosystem-atmosphereexchange of CH4 and N2O and ecosystem respiration inwetlands in the Sanjiang Plain Northeastern China GlobChange Biol 15 692ndash705

Tokida T Mizoguchi M Miyazaki T Kagemoto A Nagata O andHatano R 2007 Episodic release of methane bubbles frompeatland during spring thaw Chemosphere 70 165ndash71

van der Werf G R Randerson J T Giglio L Collatz G JKasibhatla P S and Arellano A F 2006 Interannual variabilityin global biomass burning emissions from 1997 to 2004 AtmosChem Phys 6 3423ndash41

Walter K M Zimov S A Chanton J P Verbyla D andChapin F S 2006 Methane bubbling from Siberian thaw lakesas a positive feedback to climate warming Nature 443 71ndash5

Xu X and Tian H 2012 Methane exchange between marshland andthe atmosphere over China during 1949ndash2008 GlobBiogeochem Cycles 26 GB2006

Xu X Song C and Song X 2005 Linking of microorganisms to CO2CH4 and N2O dynamics in Calamagrostis angustifoliarhizosphere soil Acta Botan Sin 25 182ndash7

Xu X Tian H Zhang C Liu M Ren W Chen G Lu C andBruhwiler L 2010 Attribution of spatial and temporalvariations in terrestrial ecosystem methane flux over NorthAmerica Biogeosciences 7 3637ndash55

8