Articles | Volume 10, issue 1
Research article
06 Jan 2017
Research article |  | 06 Jan 2017

A simple calculation algorithm to separate high-resolution CH4 flux measurements into ebullition- and diffusion-derived components

Mathias Hoffmann, Maximilian Schulz-Hanke, Juana Garcia Alba, Nicole Jurisch, Ulrike Hagemann, Torsten Sachs, Michael Sommer, and Jürgen Augustin

Abstract. Processes driving the production, transformation and transport of methane (CH4) in wetland ecosystems are highly complex. We present a simple calculation algorithm to separate open-water CH4 fluxes measured with automatic chambers into diffusion- and ebullition-derived components. This helps to reveal underlying dynamics, to identify potential environmental drivers and, thus, to calculate reliable CH4 emission estimates. The flux separation is based on identification of ebullition-related sudden concentration changes during single measurements. Therefore, a variable ebullition filter is applied, using the lower and upper quartile and the interquartile range (IQR). Automation of data processing is achieved by using an established R script, adjusted for the purpose of CH4 flux calculation. The algorithm was validated by performing a laboratory experiment and tested using flux measurement data (July to September 2013) from a former fen grassland site, which converted into a shallow lake as a result of rewetting. Ebullition and diffusion contributed equally (46 and 55 %) to total CH4 emissions, which is comparable to ratios given in the literature. Moreover, the separation algorithm revealed a concealed shift in the diurnal trend of diffusive fluxes throughout the measurement period. The water temperature gradient was identified as one of the major drivers of diffusive CH4 emissions, whereas no significant driver was found in the case of erratic CH4 ebullition events.

Short summary
Processes driving production and transport of CH4 in wetlands are complex. We present an algorithm to separate open-water automatic chamber CH4 fluxes into diffusion and ebullition. This helps to reveal dynamics, identify drivers and obtain reliable CH4 emissions. The algorithm is based on sudden concentration changes during single measurements. A variable filter is applied using a multiple of the interquartile range. The algorithm was verified for data of a rewetted former fen grassland site.