1Barcelona Supercomputing Center – Centro Nacional de Supercomputación, Barcelona, Spain
2NASA Goddard Institute for Space Studies, New York, USA
3Department of Applied Physics and Applied Math, Columbia University, New York, USA
4National Centers for Environmental Prediction, College Park, Maryland, USA
5Universitat Politècnica de Catalunya, Barcelona, Spain
Abstract. We investigate two of the major sources of uncertainty in the model estimation of the global distribution of sea-salt aerosol, i.e. the sensitivity to the emission parameterization and the influence of model resolution in coastal regions characterized by complex topography and/or steep orographic barriers where some observation sites are located. We evaluate a new sea-salt aerosol lifecycle module implemented within the online chemical transport model NMMB/BSC-CTM. Because of its multiscale core, the model is able to cover a wide range of scales. Global simulations using four state-of-the-art sea-salt emission schemes are evaluated against monthly-averaged aerosol optical depth (AOD) from selected AERONET Sun photometers, surface concentration measurements from the University of Miami's Ocean Aerosol Network and measurements from two NOAA/PMEL cruises (AEROINDOEX and ACE1). The model results are highly sensitive to the introduction of SST-dependent emissions and to the accounting of spume particles production. Depending on emission scheme, annual emissions range from 4312.9 Tg to 8979.7 Tg in the 2006. Sea-salt lifetime varies between 7.7 h and 12.0 h and the annual mean column mass load is between 5.9 Tg and 7.9 Tg. Observed coarse AOD monthly averages are reproduced with an overall correlation around 0.8 (a correlation of 0.6 is produced when applying the SST dependent scheme). Although monthly-averaged surface concentrations are overall in good agreement with the observations, there is a subset of coastal sites surrounded by complex topography where the global model overestimates by a factor of 2 or more. Using regional high-resolution simulations, we show that these large errors are mostly due to the global model's inability to capture local scale effects. In New Zeland, the increase in resolution produces a significant decrease of surface concentrations (up to 40%) – due to changes in the wind circulation and precipitation driven by the orographic barrier – which is in close agreement with surface concentration monthly climatologies measured by University of Miami stations in the region (Baring Head, Chatam Island and Inverncargill). The observed climatological precipitation in this area is well reproduced by the model at high resolution, while it is strongly underestimated when employing coarser scales. Our results outline that caution may be taken when evaluating and/or constraining coarse global sea-salt simulations with observations around coastal/orographic sites.