A mesoscale model for aerosol-cloud interaction studies WRF-PMCAMx-UF with insights to secondary ice production

The interaction between aerosols and clouds is a complex process and it causes large uncertainties in predicting the global climate. This interaction has been studied using chemical transport models (CTMs) as they simulate the distribution and composition of atmospheric aerosols. In this study, we developed a coupled version of the Weather Research and Forecasting (WRF) model with the PMCAMx-UF CTM (Skamarock et al., 2008; Patoulias et al. 2022). We did this by using prognostic cloud droplet number in the Morrison et al. 2009 cloud microphysics scheme of the WRF model. We calculated the prognostic cloud droplet number from the predicted aerosol fields of PMCAMx-UF using the Morales and Nenes 2014 activation scheme. In addition, we investigated the effects of prognostic cloud droplets to secondary ice production (SIP) in the WRF model. This involved the incorporation of various SIP processes, including Hallett-Mossop (HM), collisional fracturing and breakup (BR), droplet freezing and shattering (DS), and sublimational breakup of snow (SBS) and graupel (SBG), following the approaches outlined in Georgakaki et al. 2023. First we evaluated the impact of coupled WRF-PMCAMx-UF model with prognostic droplets to the same model with prescribed droplet number as well as the SEVIRI satellite observations. Secondly we evaluated the effects of adding SIP processes and prognostic droplets to non-SIP and prescribed droplet case and satellite observations. The results showed that using the combined model with prognostic droplets decreased the cloud droplet number concentration (CDNC) and liquid water content (LWC) when compared to the prescribed droplet simulation. This caused a more positive surface radiative forcing and thus a warming effect. In addition, the number of small particles decreased and large particle numbers increased when switching to prognostic droplets. Further, comparing to satellite observations, the prognostic droplet simulation performed better in terms of CDNC than the prescribed droplet simulation. Adding the SIP processes to the model increased the ice crystal number concentration (ICNC) as well as LWC in some areas. Compared to satellite observations, introducing SIP and prognostic droplets into the model performed slightly better in terms of CDNC as well as ice water path (IWP) than the non-SIP and prescribed droplet cases. Thus, a more realistic representation of CDNC as well as incorporation of SIP processes in the coupled model allows a more precise capture of evolving aerosol-cloud interactions in the atmosphere.