In Situ Measurements of Size Resolved Sea Spray Aerosol Emissions with the Eddy Covariance Method at Svalbard (78.9°N)

Introduction The major source of the primary marine aerosol is breaking waves. The aerosol particles are emitted from the water surface into the atmosphere directly as droplets with the composition of seawater enriched with marine organic/inorganic compounds, bacteria and viruses. With climate change affecting temperature, wind, and ice, the key physical factors driving the sea spray emissions, the production of primary marine aerosols, is expected to change. In situ emission data from Arctic waters are required in order to validate or improve the existing sea spray parameterization for cold waters, Mårtensson et al. (2003). Such improvements or validations are important in order to estimate the over-all sea spray source in a changing climate and represent this large aerosol source in climate models (e.g. Struthers et al., 2011). During the Greenhouse Arctic Ocean and Climate Effect of Aerosols (GRACE) field campaign in the summer 2009 direct flux measurements were made from a 10 meter mast on the peninsula Brøggerhalvøya at Svalbard (78.938N, 11.3406E). Method The most direct method to quantify the fluxes is the eddy covariance method. The fluctuations in vertical wind (w'), horizontal wind (u'), temperature (T') and aerosol/gas concentration (c') are sampled in parallel and processed so that the covariance, , and equals the, momentum, sensible heat and aerosol/gas/latent heat fluxes. A Gill HS-50 ultrasonic anemometer measured the vertical and horizontal wind speed. The size resolved aerosol particle concentrations were measured from 0.25 to 2.5 µm diameter with two GRIMM 1.109 Optical Particle Counters (one heated up to 400oC), and the total aerosol particle concentrations with a TSI 3772 Condensational Particle Counter. CO2 and H2O concentration were measured by a Li-7500 Open Path Analyzer. Results For the whole campaign both power spectra and cospectra are analysed. Figure 1 shows examples of mean cospectras for June 12 for all 15 particle sizes together with the latent heat flux. The aerosol composition is non-volatile as the aerosol has been heated to 400oC. The atmosphere was near to neutral stratified with mean z/L=-0.06 (z is measurements height and L Obukhov length). All aerosol sizes and the latent heat show upward fluxes during this day. The response time for the OPC is one second and follows the curve from Eugster and Senn (1995) with one second damping. The latent heat spectra follow the formula from Kaimal et al. (1972) for high frequencies; the higher fluxes at low frequencies indicate an unstable atmosphere. The particle fluxes are corrected with the formula from Horst (1997). Analyses of the turbulence give important information about the instruments and measurement site to ensure good quality of the fluxes used for validation of parameteizations. From the ocean wind sector the upward fluxes totally dominated the aerosol fluxes in the OPC range, to a higher degree than any previous data set we have worked on. This is probably due to the low back ground aerosol concentrations, which minimize the aerosol deposition fluxes. Figure 1. Average of normalized cospectra for primary marine aerosols (red triangles) and latent heat (blue line). Green line shows damped spectra due to a response time of 1 second from Eugster and Senn, (1995). Black line with triangles shows neutral spectra formulae from Kaimal et al. (1972). Straight black line show sloop of -4/3. The averages are for June 12 (48 half hours for particles and 30 half hours for the latent heat); the mean wind speed was 7.1m/s. We would like to thank the staff at The Norwegian Polar Institute in Ny Ålesund for support and Leif Bäcklin and Kai Rosman at Stockholm University for help with constructions and set up of the equipment. This work was supported by the Swedish research council (VR) through the GRACE project. Eugster, W., Senn, W., 1995, Boundary-layer meteorology, doi:10.1007/BF00712375. Horst, T.W., 1997, Boundary-layer meteorology, 82, 219-233. Kaimal et al., 1972, Quart. J. Roy. Meteorol. Soc., 98,563-589. Mårtensson E. M., et al., 2003, J. Geophys. Res., doi:10.1029/2002JD002263. Struthers et al., 2011, Atmos. Chem. Phys, doi:10.5194/acp-11-3459-2011.