Air-sea flux measurements over the Southern Ocean

• Main Author: Tupman, David James
• Other Authors: Brooks, I.
• Published: University of Leeds 2013
• Subjects: 551.46
• Online Access: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.617139

The Waves, Aerosol, and Gas Exchange Study (WAGES) collected 18 months of near-continuous and autonomous turbulent air-sea flux estimates from the research vessel RRS James Clark Ross. Supporting meteorological and sea-state measurements were also made, with the objective of improving air-sea flux parameterisations. Making turbulence measurements from a ship is technically challenging, due to bias caused by platform motion and airflow distortion. Typically, visual inspection of individual turbulence spectra is needed to quality control eddy covariance flux estimates; for WAGES the sheer volume of data motivated the development of an automated quality control method, to be performed on individual flux cospectra. The application of these tests allowed a robust relationship between the 10 m wind speed and the neutral drag coefficient to be developed, which had previously not been achieved with ship-based covariance measurements alone. This parameterisation is toward the higher end of the range of accepted values, and indicates some wind speed dependence of the Charnock parameter, rather than it being a constant. A detailed investigation of turbulent flow distortion was made; insights into the physics were gained, and a novel correction method for motion-correlated flow distortion was developed and validated. Two major modes of motion-correlated flow distortion of the turbulence were found: one correlated to the pitch, acknowledge in the literature; a second and more powerful mode correlated to the rate of change of the pitch, not acknowledged in any publication. The quality control and bias correction techniques developed for the momentum fluxes were transferred to a preliminary investigation of the sensible and latent heat fluxes. The uncertainty in the latent heat transfer coefficient was reduced considerably by use of the new techniques; however the sensible heat fluxes were dominated by noise, so discarded. The methods and corrections developed in this thesis could be used to reanalyse the turbulent flux measurements from many ship-based campaigns; improving our understanding of the physics of air-sea exchange without need for additional expensive measurements.