Each long-run leg is divided into 150-second (~9 km) sections ("runs"), over which the flight data is averaged and fluxes are derived. The distance 9 km was chosen as analysis from the Aerosol Cloud Coupling and Climate Interactions in the Arctic (ACCACIA) project showed this length to be optimal for sampling turbulence over sea ice (i.e. the minimum length required to sample the dominant turbulent length scales - the larger the run length the more likely the calculated fluxes are to be influenced by mesoscale circulations and broad changes in surface characteristics, and the fewer data points will be got)