Process-oriented study on events of unusual winter time CO2 fluxes at the Bayelva site, Spitsbergen

Abstract:

Events of unusually strong CO₂ exchange between the soil and the atmosphere, in the form of both CO₂ emission to the atmosphere and CO₂ uptake by the ground, have been observed at several Arctic sites during the winter months. Currently there is no commonly accepted explanation for this phenomenon which is why its handling in CO₂ flux data sets varies largely between excluding the events as unphysical and proposing various physical explanations.

Here we present results from a detailed analysis of such events based on eddy covariance measurements recorded at the High Arctic Bayelva site on Svalbard during the winter period of 2015. Our research reveals that in total the events account for one third of the net carbon uptake at the site in 2015. We furthermore found the events to generally occur under conditions of strong turbulence induced by high wind speeds and simultaneous with large downward sensible heat fluxes and strong evaporation. While fluxes indicating an uptake of CO₂ by the soil are most likely caused by a technical issue of the open-path gas analyzer related to low level clouds or fog, no clear measurement issues could be identified for the time periods of strong apparent CO₂ release to the atmosphere. We therefore assessed the explanatory power of several physical mechanisms proposed in the literature. Our estimations reveal that the most frequently mentioned phenomenon of an episodic release of CO₂ from a reservoir within the snow pack or in the atmospheric layer below the measurement height alone is not able to explain the magnitude of the observed fluxes. In this context we furthermore discuss the suitability of the standard eddy covariance method to effectively exclude possible advection of CO₂ rich air to the study site from remote regions. Their non-negligible impact on the annual carbon budget reveals the potential of the CO₂ events to introduce an unexpected uncertainty to larger scale upscaling products of the current Arctic carbon cycle as well future climate projections. With our study we thus call attention to the importance of a thoughtful handling of seemingly unrealistic winter fluxes when processing or utilizing Arctic CO₂ flux datasets and encourage further investigations that specifically address this highly topical research question.