Effects of natural aerosols on climate

Natural aerosols are a key component of many biogeochemical cycles, they define the baseline from which the pre‒industrial to present‒day anthropogenic aerosol radiative forcing is calculated, and they dominate the net effect of all aerosols on the incoming solar radiation. However, their impacts on climate are complex, often nonlinear, and poorly understood; leading to large uncertainties.
Global model simulations are used in this thesis to define aerosol regions unperturbed by anthropogenic pollution. On a global annual mean, unperturbed aerosol regions cover 12% of the Earth (16% of the ocean surface and 2% of the land surface) with about 90% of unperturbed regions occurring in the Southern Hemisphere. In cloudy regions with a radiative forcing relative to 1750, results suggest that unperturbed aerosol conditions could still occur on a small number of days per month. However, these environments are mostly in the Southern Hemisphere, potentially limiting the usefulness in reducing Northern Hemisphere forcing uncertainty.
Clustering techniques were used to identify natural emissions regimes in the pre-industrial and present-day where biomass burning, biogenic volatile organic compounds, dimethyl sulphide, volcanic sulphur dioxide and sea spray emissions dominate the variance in cloud condensation nuclei concentrations. Regimes are generally located in regions close to each emission source, before significant mixing occurs within the atmosphere with other emission types. These regimes are ideal “natural laboratory” locations for field study of the impacts of each natural emission on aerosol behaviour.
When pre-industrial fire emissions from two global fire models are implemented in a global aerosol model, pre-industrial global mean cloud condensation nuclei concentrations increase by a factor 1.6-2.7 relative to the widely used AeroCom dataset. Higher pre-industrial aerosol concentrations cause a substantial reduction in the calculated global mean cloud albedo forcing of between 40 and 88 percent and a reduction in the direct radiative forcing of between 5 and 10 percent. When compared to twenty-eight other sources of uncertainty in our model, pre-industrial fire emissions are by far the single largest source of uncertainty in pre-industrial cloud condensation nuclei concentrations, and hence in our understanding of the magnitude of the historical radiative forcing due to anthropogenic aerosol emissions.