Impact of sea ice on air-sea CO2 exchange – A critical review of polar eddy covariance studies. (February 2022)

• Record Type: Journal Article
• Authors: Watts, Jennifer; Bell, Thomas G.; Anderson, Karen; Butterworth, Brian J.; Miller, Scott; Else, Brent; Shutler, Jamie
• Is Part Of: Progress in oceanography. Volume 201(2022)
• Publication Date: 2022-02
• Subjects: Air-sea gas exchange -- Gas transfer velocity -- Air-sea carbon-dioxide fluxes -- Eddy covariance -- Remote sensing -- Sea ice; Oceanography -- Periodicals; 551.4605
• DOI: 10.1016/j.pocean.2022.102741
• Languages: English
• ISSNs: 0079-6611

Abstract

Highlights: Differences in methods in published eddy covariance studies make results difficult to compare. Including ancillary datasets and their uncertainties will help to clarify study conclusions. Heterogeneous sea-ice environments should be a major focus for future eddy covariance studies. Abstract: Sparse in situ measurements and poor understanding of the impact of sea ice on air-sea gas exchange introduce large uncertainties to models of polar oceanic carbon uptake. The eddy covariance technique can be used to produce insightful air-sea gas exchange datasets in the presence of sea ice, but results differ between studies. We present a critical review of historical polar eddy covariance studies and can identify only five that present comparable flux datasets. Assessment of ancillary datasets, including sea-ice coverage and type and air-sea concentration gradient of carbon dioxide, used to interpret flux datasets (with a specific focus on their role in estimating and interpreting sea ice zone gas transfer velocities) identifies that standardised methodologies to characterise the flux footprint would be beneficial. In heterogeneous ice environments both ancillary data uncertainties and controls on gas exchange are notably complex. To address the poor understanding, we highlight how future efforts should focus on the collection of robust gas flux datasets within heterogeneous sea ice regions during key seasonal processes alongside consistent ancillary data with a full characterisation of their associated uncertainties.