Abstract. We analyse the development of a radiation fog event and its gradual transition from optically thin fog in a stable boundary layer to well-mixed optically thick fog. A comparison of observations and a detailed large-eddy simulation demonstrate that aerosol growth and activation is the key process in determining the onset of adiabatic fog. Weak turbulence and low supersaturations lead to the growth of aerosol particles which can significantly affect the visibility but do not significantly interact with the long-wave radiation, allowing the atmosphere to remain stable. Only when a substantial fraction of the aerosol activates into cloud droplets can the fog interact with the radiation, becoming optically thick and well mixed. Modifications to the parameterisation of cloud droplet numbers in fog, resulting in lower and more realistic concentrations, are shown to give significant improvements to an NWP model, which initially struggled to accurately simulate the transition. Finally, the consequences of this work for common aerosol activation parameterisations used in climate models are discussed, demonstrating that many schemes are reliant on an artificial minimum value when activating aerosol in fog, and adjustment of this minimum can significantly affect the sensitivity of the climate system to aerosol radiative forcing.

In Part I of this two-part paper, a formulation was developed to treat fragmentation in ice–ice collisions. In the present Part II, the formulation is implemented in two microphysically advanced cloud models simulating a convective line observed over the U.S. high plains. One model is 2D with a spectral bin microphysics scheme. The other has a hybrid bin–two-moment bulk microphysics scheme in 3D. The case consists of cumulonimbus cells with cold cloud bases (near 0°C) in a dry troposphere. Only with breakup included in the simulation are aircraft observations of particles with maximum dimensions >0.2 mm in the storm adequately predicted by both models. In fact, breakup in ice–ice collisions is by far the most prolific process of ice initiation in the simulated clouds (95%–98% of all nonhomogeneous ice), apart from homogeneous freezing of droplets. Inclusion of breakup in the cloud-resolving model (CRM) simulations increased, by between about one and two orders of magnitude, the average concentration of ice between about 0° and −30°C. Most of the breakup is due to collisions of snow with graupel/hail. It is broadly consistent with the theoretical result in Part I about an explosive tendency for ice multiplication. Breakup in collisions of snow (crystals >~1 mm and aggregates) with denser graupel/hail was the main pathway for collisional breakup and initiated about 60%–90% of all ice particles not from homogeneous freezing, in the simulations by both models. Breakup is predicted to reduce accumulated surface precipitation in the simulated storm by about 20%–40%.

Abstract. We analyse the development of a radiation fog event and its gradual transition from optically thin fog in a stable boundary layer to well-mixed optically-thick fog. Comparison of observations and a detailed large-eddy simulation demonstrate that aerosol growth and activation is the key process in determining the onset of adiabatic fog. Weak turbulence and low supersaturations lead to the growth of aerosol particles which can significantly affect the visibility, but do not significantly interact with the long-wave radiation, allowing the atmosphere to remain stable. Only when a substantial fraction of the aerosol 5 become activated into cloud droplets can the fog interact with the radiation, becoming optically thick and well-mixed. Modifications to the parametrization of cloud droplet numbers in fog, resulting in lower and more realistic concentrations, are shown to give significant improvements to an NWP model, which initially struggled to accurately simulate the transition. Finally, consequences of this work for common aerosol activation parametrizations used in climate models are discussed, demonstrating that many schemes are reliant on an artificial minimum value when activating aerosol in fog, and adjustment of this minimum 10 can significantly affect the sensitivity of the climate system to aerosol radiative forcing.

Abstract. An intercomparison between 10 single-column (SCM) and 5 large-eddy simulation (LES) models is presented for a radiation fog case study inspired by the Local and Non-local Fog Experiment (LANFEX) field campaign. Seven of the SCMs represent single-column equivalents of operational numerical weather prediction (NWP) models, whilst three are research-grade SCMs designed for fog simulation, and the LESs are designed to reproduce in the best manner currently possible the underlying physical processes governing fog formation. The LES model results are of variable quality and do not provide a consistent baseline against which to compare the NWP models, particularly under high aerosol or cloud droplet number concentration (CDNC) conditions. The main SCM bias appears to be toward the overdevelopment of fog, i.e. fog which is too thick, although the inter-model variability is large. In reality there is a subtle balance between water lost to the surface and water condensed into fog, and the ability of a model to accurately simulate this process strongly determines the quality of its forecast. Some NWP SCMs do not represent fundamental components of this process (e.g. cloud droplet sedimentation) and therefore are naturally hampered in their ability to deliver accurate simulations. Finally, we show that modelled fog development is as sensitive to the shape of the cloud droplet size distribution, a rarely studied or modified part of the microphysical parameterisation, as it is to the underlying aerosol or CDNC.

Various improvements were made to a state‐of‐the‐art aerosol–cloud model and comparison of the model results with observations from field campaigns was performed. The strength of this aerosol–cloud model is in its ability to explicitly resolve all the known modes of heterogeneous cloud droplet activation and ice crystal nucleation. The model links cloud particle activation with the aerosol loading and chemistry of seven different aerosol species. These improvements to the model resulted in more accurate prediction especially of droplet and ice crystal number concentrations in the upper troposphere and enabled the model to directly sift the aerosol indirect effects based on the chemistry and concentration of the aerosols. In addition, continental and maritime cases were simulated for the purpose of validating the aerosol–cloud model and for investigating the critical microphysical and dynamical mechanisms of aerosol indirect effects from anthropogenic solute and solid aerosols, focusing mainly on glaciated clouds. The simulations showed that increased solute aerosols reduced cloud particle sizes by about 5 μm and inhibited warm rain processes. Cloud fractions and their optical thicknesses were increased quite substantially in both cases. Although liquid mixing ratios were boosted, there was however a substantial reduction of ice mixing ratios in the upper troposphere owing to the increase in snow production aloft. These results are detailed in the subsequent parts of this study.