Uncertainties in the determination of global sub-micron marine organic matter emissions

Organic matter (OM) constitutes an important contribution to the composition of sub-micron sea-spray aerosol produced from biologically active waters. However, OM emission estimates vary by more than an order of magnitude. To estimate the uncertainties in the OM production estimates a sensitivity analysis has been performed in which various parameters have been varied. These include different sea-spray source functions, satellite-retrieved chlorophyll distributions, and a relationship correlating in situ organic mass measurements with satellite-retrieved chlorophyll data. The starting point was a baseline model from which the annual global emission of the water insoluble organic matter (WIOM) fraction in sea spray has been estimated to be 20.4 Tg. In this baseline the global WIOM emission is dominated by the contribution (80%) of the chlorophyll-poor regions (<0.3 mg m−3). Significant deviations from this estimate are introduced by the choice of the sea-spray source function and the assumed background organic mass fraction, each of which leads to an uncertainty of at least a factor of 2. In particular the chlorophyll-poor regions which dominate the WIOM contribution are strongly affected by the choice of the organic mass fraction parameterisation. The way the chlorophyll data are handled, such as different gap filling approaches, causes deviations in the OM emission that are in the order of 10% and is therefore of less importance. The present research indicates that special attention should be given to the low chlorophyll areas in e.g. the tropics, since there the parameterisations are most uncertain and at the same time these regions dominate total WIOM emissions.

► We model global emission of the organic matter (OM) fraction of sub-micron sea spray.
► The global source strength of the OM fraction is highly uncertain.
► Key uncertainties are the sea-spray source function and OM relationship.
► The tropical region was identified to be extra sensitive.
► We recommend a stronger observational basis to reduce the large uncertainties.