黄河三角洲芦苇湿地生态系统碳、水热通量特征

Carbon,water and heat fluxes of a reed( Phragmites australis) wetland in the Yellow River Delta,China

利用涡度相关法对黄河三角洲芦苇湿地生态系统进行了连续两年的通量观测,对2009—2010年生长季芦苇湿地的净生态系统碳交换量(NEE),感热通量(Hs)和潜热通量(LE)数据进行了分析。结果表明,在日尺度上,芦苇湿地NEE日变化特征表现为两个CO2吸收高峰,分别出现在11:00和16:00左右,其特点是在午间出现了碳交换通量的降低。CO2吸收的日最大值在两个生长季出现的时间有所不同,分别出现在2009年7月(-0.30 mg CO2m-2s-1)和2010年6月(-0.37 mg CO2m-2s-1)。CO2排放的日最大值两个生长季均出现在9月,分别为0.19和0.25 mg CO2m-2s-1。Hs和LE的日动态均为单峰型,极值都出现在中午前后,生长季生态系统的能量消耗主要以潜热为主,且在日尺度上,热通量和NEE有显著的负相关关系。在季节尺度上,芦苇湿地生长季具有明显的碳汇功能,2009年生长季生态系统白天固定354.63 g CO2/m2,同时期夜间释放159.24 g CO2/m2,净CO2吸收量为-195.39 g CO2/m2。2009年整个生长季生态系统总初级生产力(GPP)为-651.13 g CO2/m2,生态系统呼吸(Re)为455.74 g CO2/m2,系统表现为碳汇。路径分析表明:光合有效辐射(PAR)显著影响NEE的日动态(R2=0.46—0.84),而NEE的季节动态主要受土壤温度的影响,降水和PAR的影响次之。

Wetlands are regarded as one of the largest ＇ unknowns＇ regarding future carbon （C） dynamics and greenhouse gas fluxes in the context of global change and climate policy-making. To understand the dynamics of carbon cycling of wetland ecosystem, we used an eddy covariance technique to measure the net ecosystem carbon dioxide （CO2 ） exchange （NEE, positive or negative values of NEE represent net losses or gains of C, respectively, for the ecosystem） , sensible heat flux （Hs） and latent heat flux （LE） between vegetation and the atmosphere at a reed wetland ecosystem in the Yellow River Delta during the periods of two growing seasons in 2009 and 2010. The total amount of rainfall in 2009 （571.4 mm） was higher than the annual average （551 mm）. In contrast, precipitation in 2010 （523.5 mm） was significantly lower than average. The results from 2-year eddy tower observations showed that there was a dual peak in diurnal pattern of NEE fluxes for the reed wetland, which occurred at about 11：00 and 16：00, respectively. There were two different temporal patterns for the maximum diurnal uptake values of CO2 in 2 years. The maximum diurnal uptake values of CO/were -0.30 mg CO2 m^-2 s^-1（July, 2009） and -0.37 mg CO2 m^-2 s^-1（June, 2010） ,respectively. The maximum diurnal emitting values of CO2 were 0.19 and 0.25 mg CO2 m^-2 s^-1 , respectively, and both occurred in September for 2 years. The diurnal patterns of Hs and LE were both single peaks, and their peak values both occurred at noon. The maximum latent heat flux was higher than the sensible heat flux, and the latent heat flux was the primary consumption component for net radiation during both two years. In diurnal scale, the heat fluxes were strongly negatively correlated to NEE fluxes （R2 1〉0.5, P〈0.0001 ）. On the seasonal scale, the reed wetland was a strong C sink during the growing season. In 2009, the wetland ecosystem fixed 354.63 g CO2 m-2 in daytime of the whole growing season, and meanwhile it released 159.24 g CO2/m2 in nighttime. Approximate 651.13 g CO2/m^2 was fixed by gross primary production （GPP） , and 455.74 g CO2/m^2 were released as ecosystem respiration （Ro）, which resulted in a strong sink of atmospheric CO2 with -195.39 g CO2/m^2 sequestered in 2009 growing season based on the observed data. Path analysis results showed that the fluctuation of diurnal NEE fluxes was closely related to the photosynthetic active radiation （PAR） （R2= 0.46-0.84）o However, soil temperature had the greatest effect on seasonal dynamics of ecosystem CO2 exchange during the growing seasons at the study site, higher than the contributions of the other environmental factors such as precipitation and PAR. Our results strongly suggested that the combination of temperature, precipitation and PAR, as well as phonological stage of vegetation, control the C dynamics of reed wetland ecosystem. Therefore, an accurate representation of these parameters is extremely valuable for developing accurate and predictive wetland C cycle models and for the success of forecasting carbon budgets of reed wetlands.