Evaluate parameterizations of cloud droplet spectral dispersion by using worldwide aircraft data

Increased aerosol potentially impacts the cloud droplet spectrum, which in turn affects the aerosol-cloud interaction (ACI), known as the dispersion effect (DE). To consider DE in general circulation models (GCMs), many parameterizations have been proposed, but there are relatively few quantitative and global evaluations due to lacking suitable data and methodology. Additionally, both observations and numerical simulations confirmed that DE has opposite effects on ACI in aerosol- and updraft-limited regimes, but whether this effect can be reproduced by parameterizations has no clear conclusion until now. In this study, we used a liquid water content (LWC) binning method and worldwide data (China, Canada, Brazil, and Chile) to evaluate six dispersion parameterizations, namely Martin94, RLiu03, PengL03, Liu08, LiuLi15, and Zhang22. The LWC binning method ensures the difference between ACI values calculated by the cloud droplet number concentration and the effective radius is mainly caused by DE, which makes the quantitative calculation of DE possible. The results show that 1). DE has a weakening effect on ACI in the aerosol-limited regime, but an enhanced effect in the updraft-limited regime; 2). empirical parameterizations (Martin94, RLiu03, PengL03, and Liu08) can only show the weakening effect of DE on ACI, leading to an underestimation (-29% ~ -42%) for calculated ACI, especially for the updraft-limited regime; 3). both LiuLi15 and Zhang22 can reproduce the opposite effects of DE on ACI in different regimes, so we recommend giving priority to the LiuLi15 and the Zhang22 schemes when calculating DE in GCMs.