Net Ecosystem Carbon Fluxes From A Peatland In The Continuous Permafrost Zone, Greater Hinggan Mountains

Eddy covariance, as a micrometeorological technique, allows a non-invasivemeasurement of the exchange of CO₂between the atmosphere and vegetation, whichhas been extensively applied to measure terrestrial ecosystems carbon fluxes. As apart of wetland ecosystems carbon cycle research, methane flux observations atecosystem level were gradually improved. Closed-path eddy covariance （CPEC）technique was applied to measure CO₂, CH₄, H2O and energy fluxes in peatlandecosystem from2010to2012in the continuous permafrost zone of Greater HingganMountains, Northeast China. We discussed the theoretical problems on applyingCPEC technique to the long-term measurement of CO₂and CH₄fluxes. The temporalvariation and environmental controls on the net ecosystem CO₂exchange and CH₄fluxes of peatland were examined. Combining with the CO₂and CH₄fluxes duringthe snowpack period, the annual net carbon balance of the peatland was estimated.The main conclusions are:（a） The integral turbulent tests suggested that the quality of raw data obtained bythe CPEC instruments satisfied fluxes calculation. The cospectra and power spectraanalysis suggested that the effects of instruments including sensors separation, thedynamic frequency response of the FGGA and the sonic anemometer did not restrainthe high-frequency fluctuations. The energy balance closure of the peatland reached62%without consideration of soil and canopy heat indicated that it was valid forassessing the quality of fluxes. There was no significant difference between the rawCO₂and CH₄fluxes and fluxes corrected by double rotation suggested the peatlandtopography was suitable for fluxes observation. The raw flux should be corrected by WPL which could eliminate the effects of water vapor on flux calculation. A thresholdof friction velocity was set as0.1m/s according to analysis of the relationshipbetween nighttime NEE and friction velocity using average value test method. TheFSAM model was applied to do footprint analysis. The results indicated thecontribution of flux came from the peatland where we focused on.（b） The net ecosystem CO₂exchange （NEE） showed significant daily variationduring the growing season. The single-peaked patterns were observed in dailyvariation of NEE, which experiencing carbon release in nighttime and carbon uptakein daytime with the maximum usually reached between9:00and11:00. There was noobviously daily variation of CO₂emission during the non-growing season. Theseasonal patterns of daily summed NEE presented as single-peaked curves and themaximum usually appeared in July. During the measurement period of May toOctober; carbon release was occurred in May, September and October, carbon uptakewas shown in other months. Peatland NEE was controlled by PAR and daytime NEEfitted fairly well with the PAR in a rectangular hyperbolic function. During themeasurement periods, air temperature, VPD and relative humidity were other keyfactors, which affected peatland plant photosynthesis.（c） The seasonal patterns of peatland GPP and ER were also single-peaked curves,which increased firstly and then decreased with time. Peatland GPP was significantcorrelated with air and soil temperatures （R²was in the range of0.66to0.925）. Netradiation, soil volume water content and VPD were other important factorsinfluencing GPP. Ecosystem respiration （ER） had a strong relationship withtemperature, especially the soil temperature at10cm depth. The temperaturesensitivity coefficient （Q10） ranged from3.1to4.9. The following key factors werewater table and soil volume water content, which played certain roles in controllingpeatland ER.（d） The daily patterns of ecosystem CH₄fluxes were not clearly observed and theemission rates of CH₄in nighttime were larger than that in daytime. The peatlandecosystem CH₄fluxes showed slight seasonal variations which markedly varied indifferent measurement years. During the measurement periods, ecosystem CH₄fluxes were in the range of-1.95¹2.42mg CH₄m^-2d^-1. We found daily CH₄flux wascontrolled by net radiation and NEE during the growing season. Soil temperature,water table depth and active layer depth controlled on seasonal variation of ecosystemCH₄fluxes. Peatland GPP also could influence CH₄emission during the growingseason.（e） During the snowpack period, peatland acted as CO₂and CH₄source, andsnowpack gas fluxes showed significant seasonal patterns. The emission rates of CO₂and CH₄were ranged from6.86³3.29mg C m^-2d^-1and0.02⁰.15mg C m^-2d^-1,respectively. Snow depth and snow porosity were two key factors controlledsnowpack CO₂and CH₄fluxes, and air pressure affected CO₂emissions.（f） The peatland ecosystem acted as obvious CO₂sink and CH₄source in2011and2012. The annually accumulated NEE in2011and2012were-33.097and-29.633g Cm-2, and the accumulated CH₄were1.107and0.445g C m-2, respectively. Thus, theannually net carbon accumulation in two years was-31.99and-29.188g C m-2,respectively. The results indicated that the peatland ecosystem in the continuouspermafrost zone of Greater Hinggan Mountains acted as a weak carbon sink in thecurrent.（g） According to estimation, the peatland emitted12.718and11.071gC m^-2season^-1 during the winter （The average air temperature of five consecutive daysbelow0oC）, accounting for37.4and38.3%of the annual NEE in2011and2012.Winter CH₄emissions from the peatland were0.13and0.032gC m^-2season^-1in thetwo consecutive years. The narrow range of7-12%of the proportion of winter fluxesfrom annual CH₄emission found in the peatland ecosystem in the study area. Theseresults indicated the importance of winter carbon gases emission when assessedpeatland ecosystem annual net carbon balance in Northeast China.