Multi-scale temporal variation in methane emission from an alpine peatland on the Eastern Qinghai-Tibetan Plateau and associated environmental controls

Most previous studies on methane (CH4) emissions from peatlands have focused on the boreal, subarctic, and tropical regions. Little is known about CH4 emissions from the alpine peatlands that are widely distributed in the eastern Qinghai-Tibetan Plateau (QTP), which are sensitive to climate change and human disturbance. To assess the magnitudes of daily and seasonal variations in CH4 flux in this area, and to identify the influential environmental factors, an eddy covariance (EC) tower with a LI-7700 open-path CH4 analyzer was established on Hongyuan Peatland. Total annual CH4 emission was 46.8 g CH(4)m(-2) in 2014, with emissions in the non-growing season accounting for 25% of the annual total. From May to September 2014, diurnal variation in CH4 emissions was observed, with CH4 fluxes varying between 0.15 and 0.25 mu molm(-2) s(-1). In contrast, during all other periods of 2014, no diurnal variation was observed, and CH4 flux varied below 0.05 mu mol m(-2) s(-1). A clear seasonal pattern in CH4 exchange with small surges in CH4 emission appeared in soil thawing and freezing seasons. Wavelet analysis was applied to the continuous CH4 flux time series to explore temporal variation in ecosystem CH4 exchange during the growing season. On daily timescales, changes in CH4 flux are in phase with changes in air temperature, and influenced by friction velocity and vapor pressure deficit (VPD). On weekly-to-monthly timescales, soil temperature can explain most of the variation in CH4 exchange. Though CH4 fluxes had no apparent correlation with water table level, fluxes were significantly (R-2 = 0.86) correlated with soil temperature measured at -25 cm depth, close to the water table level throughout the growing season, suggesting a synergistic effect of water table level and soil temperature on methane emission. This study highlights the need for continuous eddy covariance measurements and time series analysis approaches to adequately describe temporal variability in ecosystem CH4 exchange.