Droplet number uncertainties associated with CCN: An assessment using observations and a global model adjoint
We use the Global Modelling Initiative (GMI) chemical transport model\nwith a cloud droplet parameterisation adjoint to quantify the\nsensitivity of cloud droplet number concentration to uncertainties in\npredicting CCN concentrations. Published CCN closure uncertainties for\nsix different sets of simplifying compositional and mixing state\nassumptions are used as proxies for modelled CCN uncertainty arising\nfrom application of those scenarios. It is found that cloud droplet\nnumber concentrations (N-d) are fairly insensitive to the number\nconcentration (N-a) of aerosol which act as CCN over the continents\n(partial derivative lnN(d)/partial derivative lnN(a) similar to 10-30\n%), but the sensitivities exceed 70% in pristine regions such as the\nAlaskan Arctic and remote oceans. This means that CCN concentration\nuncertainties of 4-71% translate into only 1-23% uncertainty in cloud\ndroplet number, on average. Since most of the anthropogenic indirect\nforcing is concentrated over the continents, this work shows that the\napplication of Kohler theory and attendant simplifying assumptions in\nmodels is not a major source of uncertainty in predicting cloud droplet\nnumber or anthropogenic aerosol indirect forcing for the liquid,\nstratiform clouds simulated in these models. However, it does highlight\nthe sensitivity of some remote areas to pollution brought into the\nregion via long-range transport (e. g., biomass burning) or from\nseasonal biogenic sources (e. g., phytoplankton as a source of\ndimethylsulfide in the southern oceans). Since these transient processes\nare not captured well by the climatological emissions inventories\nemployed by current large-scale models, the uncertainties in\naerosol-cloud interactions during these events could be much larger than\nthose uncovered here. This finding motivates additional measurements in\nthese pristine regions, for which few observations exist, to quantify\nthe impact (and associated uncertainty) of transient aerosol processes\non cloud properties.