Uncertainties in eddy covariance air–sea CO₂ flux measurements and implications for gas transfer velocity parameterisations

• Main Authors: Y. Dong, M. Yang, D. C. E. Bakker, V. Kitidis, T. G. Bell
• Format: Article
• Language: English
• Published: Copernicus Publications 2021-05-01
• Series: Atmospheric Chemistry and Physics
• Online Access: https://acp.copernicus.org/articles/21/8089/2021/acp-21-8089-2021.pdf

<p>Air–sea carbon dioxide (<span class="inline-formula">CO₂</span>) flux is often indirectly estimated by the bulk method using the air–sea difference in <span class="inline-formula">CO₂</span> fugacity (<span class="inline-formula">Δ<i>f</i></span><span class="inline-formula">CO₂</span>) and a parameterisation of the gas transfer velocity (<span class="inline-formula"><i>K</i></span>). Direct flux measurements by eddy covariance (EC) provide an independent reference for bulk flux estimates and are often used to study processes that drive <span class="inline-formula"><i>K</i></span>. However, inherent uncertainties in EC air–sea <span class="inline-formula">CO₂</span> flux measurements from ships have not been well quantified and may confound analyses of <span class="inline-formula"><i>K</i></span>. This paper evaluates the uncertainties in EC <span class="inline-formula">CO₂</span> fluxes from four cruises. Fluxes were measured with two state-of-the-art closed-path <span class="inline-formula">CO₂</span> analysers on two ships. The mean bias in the EC <span class="inline-formula">CO₂</span> flux is low, but the random error is relatively large over short timescales. The uncertainty (1 standard deviation) in hourly averaged EC air–sea <span class="inline-formula">CO₂</span> fluxes (cruise mean) ranges from 1.4 to 3.2 <span class="inline-formula"><math xmlns="http://www.w3.org/1998/Math/MathML" id="M15" display="inline" overflow="scroll" dspmath="mathml"><mrow class="unit"><mi mathvariant="normal">mmol</mi><mspace width="0.125em" linebreak="nobreak"/><msup><mi mathvariant="normal">m</mi><mrow><mo>-</mo><mn mathvariant="normal">2</mn></mrow></msup><mspace linebreak="nobreak" width="0.125em"/><msup><mi mathvariant="normal">d</mi><mrow><mo>-</mo><mn mathvariant="normal">1</mn></mrow></msup></mrow></math><span><svg:svg xmlns:svg="http://www.w3.org/2000/svg" width="67pt" height="13pt" class="svg-formula" dspmath="mathimg" md5hash="a4c66177a451915c5530d12ef0d4aec1"><svg:image xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="acp-21-8089-2021-ie00001.svg" width="67pt" height="13pt" src="acp-21-8089-2021-ie00001.png"/></svg:svg></span></span>. This corresponds to a relative uncertainty of <span class="inline-formula">∼</span> 20 % during two Arctic cruises that observed large <span class="inline-formula">CO₂</span> flux magnitude. The relative uncertainty was greater (<span class="inline-formula">∼</span> 50 %) when the <span class="inline-formula">CO₂</span> flux magnitude was small during two Atlantic cruises. Random uncertainty in the EC <span class="inline-formula">CO₂</span> flux is mostly caused by sampling error. Instrument noise is relatively unimportant. Random uncertainty in EC <span class="inline-formula">CO₂</span> fluxes can be reduced by averaging for longer. However, averaging for too long will result in the inclusion of more natural variability. Auto-covariance analysis of <span class="inline-formula">CO₂</span> fluxes suggests that the optimal timescale for averaging EC <span class="inline-formula">CO₂</span> flux measurements ranges from 1 to 3 h, which increases the mean signal-to-noise ratio of the four cruises to higher than 3. Applying an appropriate averaging timescale and suitable <span class="inline-formula">Δ<i>f</i></span><span class="inline-formula">CO₂</span> threshold (20 <span class="inline-formula">µatm</span>) to EC flux data enables an optimal analysis of <span class="inline-formula"><i>K</i></span>.</p>