Nucleation Scavenging of Smoke Particles and Simulated Drop Size Distributions over Large Biomass Fires

Abstract

A prescribed burn experiment was conducted in Hardiman Township, Ontario, Canada in August 1987. The fire was of adequate intensity to force the formation of a cumulus cloud, and much of the smoke passed through this cloud. The smoke plume and the fire-driven cloud were observed using ground- and aircraft-based instrumentation. In this paper, we present calculations of the evolution of the smoke and drop size distributions within this cloud due to nucleation and condensation from a detailed microphysical entraining model. Initial conditions for the model simulations were specified using a three-dimensional warm-cloud dynamical model. We investigate the changes of parcel characteristics caused by the effect of entrainment, as well as the sensitivity of nucleation scavenging and drop number concentration to the different background environmental aerosol distributions. The fraction of smoke particles nucleated depends on the chemical nature of the particles, their size distribution, the quantity of smoke present at cloud base, the updraft velocity, and the environmental conditions. The predicted drop size distribution and the measured data above the fire are in good agreement for drop sizes that correspond to condensation on the measured smoke spectra.The results from this study can be used to develop a parameterization that relates the mass of smoke scavenged by nucleation to the updraft velocity and the initial smoke mass loading. We plan to extend this study to predict the effect of aerosols on drop spectra for use in climate models.