The Response of a Simulated Mesoscale Convective System to Increased Aerosol Pollution

November 19, 2015

Michal Clavner

Committee: Bill Cotton (advisor) Sue van den Heever (co-advisor) Thomas Birner Sonia Kreidenweis Steve Reising

Abstract

Mesoscale Convective Systems (MCS) are important contributors to rainfall in the High Plains of the United States as well as producers of severe weather such as hail, tornados and straight-line wind events known as derechos. It is therefore of interest to understand how different aerosols serving as cloud condensation nuclei (CCN) concentrations may impact these systems. This work focuses on the impacts of aerosols on the total precipitation amount, rates and spatial distribution of precipitation produced by an MCS as well as the characteristics of the derecho event. Past studies have shown that the impacts on MCS-produced precipitation to changes in aerosol concentration are strongly dependent on environmental conditions, primarily humidity and environmental wind shear. Changes in aerosol concentrations were found to alter MCS-precipitation production directly by modifying precipitation processes and indirectly by affecting the efficiency of the storm's self-propagation. Observational and numerical studies have been conducted examining the dynamics responsible for the generation of widespread convectively-induced windstorms, primarily focusing on environmental conditions and the MCS features that generate a derecho event. While the sensitivity of the formation of bow-echos, the radar signature associated with derecho events, to changes in microphysics have been examined, a study on a derecho-producing MCS characteristics to aerosol concentrations has not. In this study different aerosol concentrations and their effects on precipitation and a derecho produced by an MCS are examined by simulating a case study. The 8 May 2009 "Super-Derecho" was simulated using the Regional Atmospheric Modeling System (RAMS), a cloud-resolving model (CRM) with sophisticated aerosol and microphysical parameterizations. Three simulations were conducted. The first with no anthropogenic aerosols present in the simulation, the second with a moderate increase in aerosol concentrations by including anthropogenic emission. The third simulation contained the same aerosol distribution as in the second, however multiplied by a factor of 5 in order to represent a highly polluted scenario. In all three of the simulations aerosol concentrations were derived from the output of GEOS-Chem, a 3D chemical transport model. In the simulated MCS, the formation and propagation of the storm was not fundamentally modified by changes in the aerosol concentration, and the total MCS-produced precipitation was not significantly affected. However the precipitation distribution (convective vs stratiform) and derecho-strength surface wind characteristics did vary among the simulations. The more polluted simulations exhibited higher precipitation rates, higher bulk precipitation efficiency, a larger area with heavier convective precipitation and a smaller area with lighter stratiform precipitation. These differences arose because aerosol pollution enhanced precipitation in the convective region while suppressing precipitation from the stratiform-anvil. Higher aerosol concentrations led to invigoration of convective updrafts which elevated raindrop trajectories, which supported greater drop growth, and lofted more liquid cloud mass to higher levels. Both collision-coalescence and riming were enhanced in the convective regions in the more polluted simulation. The presence of greater aerosol concentrations in the free troposphere, as well as in the boundary layer, reduced both collision-coalescence and riming within the stratiform-anvil region. As a consequence, the more polluted simulations produced the smallest precipitation from the MCS stratiform-anvil region. In order to understand the impact of changes in aerosol concentrations on the derecho characteristics, the dynamical processes which produced the strong surface wind were determined by performing back-trajectory analysis during three periods of the simulated storm. A time dependent and nonlinear trend was found in the intensity of the derecho, determined by variations in both the area and strength of the surface winds of the simulated derecho to the increased aerosol concentrations which served as cloud condensation nuclei. During the formation period of the MCS, the non-linear trend was attributed to the microphysical impact of aerosol loading on the intensity of the cold-pool, primarily changes in melting rates. The aerosol loading impact on cold-pool intensity modified the balance between the horizontal vorticity generated by the cold-pool to the environment. The simulation with no anthropogenic aerosols exhibited the strongest cold-pool and rear inflow jet, while the simulation with the highest amount of anthropogenic aerosols exhibited the strongest convective downbursts. As the storm matured, the derecho winds were found to be associated with the formation of a mesovortex at the gust front and the non-linear trend in derecho intensity was attributed to a non-linear trend in mesovortex strength. The formation of a stronger mesovortex was found to increase the contribution of the derecho winds to an "up-down" downdraft trajectory following a convective downburst.