Global Evaluation of Aerosol Number Concentrations using the United Kingdom Earth System Model

Aerosols affect the global energy balance by directly scattering and absorbing radiation, and indirectly by their ability to act as cloud condensation nuclei which changes the microphysical properties of clouds. There is a need to understand the vertical distribution of aerosol particles globally, as this influences the cloud droplet number and therefore the indirect forcing. In this study, we evaluate global particle number concentrations using the United Kingdom Earth System Model (UKESM). The vertical aerosol distribution from the model is evaluated in the tropics, midlatitudes and high latitudes, and is compared to in-situ observations of aerosol size distributions from the NASA Atmospheric Tomography Mission (ATom). The particle number concentration measurements from the ATom campaign were obtained using two nucleation-mode aerosol size spectrometers, two ultra-high sensitivity aerosol spectrometers and a laser aerosol spectrometer (Brock et al 2019). The model overpredicts total particle number concentrations in the upper troposphere, mainly in the nucleation mode and underpredicts particles of all sizes in the boundary layer. Several sensitivity tests are performed to quantify the source of these biases and constrain the vertical aerosol distribution. Different atmospheric processes have a controlling influence on the particle number concentrations present in the atmosphere. Here we determine the importance of organic nucleation, free tropospheric nucleation rates, the magnitude of the coagulation sink, the scavenging of aerosol particles in a plume, wet and dry deposition rates, and the pH of cloud droplets. We also present several important technical improvements to the treatment of nucleation-mode microphysics in the model. Perturbing these atmospheric processes/parameters and improving the physical basis of the model helps us constrain the vertical aerosol distribution accurately and helps reduce the uncertainty in our estimation of the direct and indirect forcing.