The Microphysical Properties and Sensitivities of Marine Fog

Marine fog presents a modeling challenge. Accurate forecasts rely on understanding the unique behaviors of a variety of of marine fog types, but the marine environment is remote, and a lack of observations has contributed to a limited understanding. Cloud base lowering (CBL) in which radiative cooling at the top of a stratus cloud leads to the downward growth of cloud base and the formation of fog, is one of the more common ways that marine fog forms. However, relatively little work has been devoted to the influence of microphysics on the formation and evolution of CBL fog. This gap is addressed here by pairing multiple modeling experiments conducted using different models, microphysics schemes, and case setups with in situ observations. This dissertation provides insights into aspects of marine fog that can be used to inform future research and improve forecasts.Chapter 2 uses a marine fog event that occurred near Canada’s Grand Banks to investigate the sensitivity of simulated fog properties to six model parameters found primarily in the microphysics scheme. Analysis of model simulations shows that the shape parameter, which controls the relative width of the droplet size distribution, and the aerosol number concentration have the greatest impact on fog in terms of spatial extent, duration, and surface visibility. Additionally, we find that the influence of the shape parameter is expressed primarily through its effect on microphysical processes and not its effects on the radiative properties of clouds. Higher shape parameter and higher aerosol concentration, both of which reduce mean fall speed of droplets and/or suppress drizzle formation, lead to reduced visibility in fog but also delayed the onset of fog, shortened its lifetimes, and reduced its spatial extent. Chapter 3 employs a modeling experiment conducted on an idealized cloud base lowering fog case to build upon the results of Chapter 2, particularly the link between aerosol, mean fall speed, and the trade-off between the duration/extent and density of fog. We use a single-column model configured with bin microphysics to investigate the interplay among aerosols, microphysics, and CBL fog evolution under diverse meteorological conditions. We find that lower aerosol concentrations lead to earlier fog formation due to faster gravitational settling of larger droplets, which serves to flux moisture downward. Faster gravitational settling (among other mechanisms at low aerosol concentration) also suppresses entrainment at cloud top which aids in keeping the liquid water path high. However, faster gravitational settling also limits the fog water concentration through faster liquid deposition to the surface. It is these counteracting influences of gravitational settling that appear to cause both prolonged fog duration and suppressed fog water concentration. The relative strength of these counteracting influences depends on the environmental conditions.In Chapter 4, we return to the link between the assumed width of droplet size distributions in models and the behavior of simulated fog. We simulate the same idealized CBL fog case from Chapter 3, but with a bulk microphysics scheme to study the relationship between the shape parameter and fog properties. We once again find that higher shape parameter, which corresponds to a narrower droplet size distribution, suppresses fog formation but leads to lower minimum visibility during fog. We then pair this finding with an analysis of in situ observations of droplet size distributions for fog events that occurred on Sable Island to evaluate the implications of how observed fog droplet size distributions (DSDs) are parameterized in models. Most of the fog observations had bimodal DSDs. Conventional model parameterizations, which assume that the DSD follows a gamma PDF, do a poor job qualitatively replicating observed DSDs. Nonetheless, we find minimal overall bias between the mean fall speeds of the observed and approximated DSDs, but find that the collision rates of observed DSDs were better approximated with shape parameters four times greater than those calculated using relative dispersion. The results show that using relative dispersion from fog observations to parameterize DSD could result in models overestimating the duration of fog by up to 20%.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.