Environmental controls on ecosystem-scale CH4 emission from polygonal tundra in the Lena River Delta, Siberia

• Published in: Journal of Geophysical Research
• Main Authors: Sachs, T., Wille, C., Boike, J., Kutzbach, L.
• Format: Article in Journal/Newspaper
• Language: English
• Published: 2008
• Subjects: Arctic; lena river; permafrost; Tundra; Siberia
• Online Access: http://hdl.handle.net/11858/00-001M-0000-0018-2DAA-4

We present the first ecosystem-scale methane flux data from a northern Siberian tundra ecosystem covering the entire snow-free period from spring thaw until initial freeze-back. Eddy covariance measurements of methane emission were carried out from the beginning of June until the end of September in the southern central part of the Lena River Delta (72 degrees 22'N, 126 degrees 30'E). The study site is located in the zone of continuous permafrost and is characterized by Arctic continental climate with very low precipitation and a mean annual temperature of -14.7 degrees C. We found relatively low fluxes of on average 18.7 mg m(-2) d(-1), which we consider to be because of (1) extremely cold permafrost, (2) substrate limitation of the methanogenic archaea, and (3) a relatively high surface coverage of noninundated, moderately moist areas. Near-surface turbulence as measured by the eddy covariance system in 4 m above the ground surface was identified as the most important control on ecosystem-scale methane emission and explained about 60% of the variance in emissions, while soil temperature explained only 8%. In addition, atmospheric pressure was found to significantly improve an exponential model based on turbulence and soil temperature. Ebullition from waterlogged areas triggered by decreasing atmospheric pressure and near-surface turbulence is thought to be an important pathway that warrants more attention in future studies. The close coupling of methane fluxes and atmospheric parameters demonstrated here raises questions regarding the reliability of enclosure-based measurements, which inherently exclude these parameters.