HR: 1340h
AN: A13A-0873    [Abstracts]
TI: A computationally efficient aerosol nucleation/condensation method: Pseudo-steady state gas phase sulfuric acid
AU: Adams, P J
EM: petera@andrew.cmu.edu
AF: Department of Civil and Environmental Engineering, Department of Engineering and Public Policy, Carnegie Mellon University, 5000 Forbes Ave., Pittsburgh, PA 15213, United States
AU: * Pierce, J R
EM: jrpierce@andrew.cmu.edu
AF: Department of Chemical Engineering, Carnegie Mellon University, 5000 Forbes Ave., Pittsburgh, PA 15213, United States
AB: Aerosol nucleation and condensation are two processes that compete for gas phase sulfuric acid when it is formed in the atmosphere. Without approximations, accurate numerical integration of aerosol microphysics requires time steps on the order of seconds or less when nucleation is occurring, significantly shorter than the time steps required by other processes governing aerosol microphysics. This computational burden makes the explicit numerical simulation burdensome in 3-dimensional atmospheric models. We have developed an efficient method for simulating nucleation/condensation by assuming that gas phase sulfuric acid is at a pseudo- steady state (PSS) concentration determined by chemical generation and its loss by nucleation and condensation. The time step for nucleation/condensation is then governed by 1) the change in the condensation sink, 2) the change in the generation rate of sulfuric acid, 3) the coagulation timescale, or 4) a master time step in the model. The PSS assumption fails only when the time for sulfuric acid to reach its PSS state concentration is not significantly shorter than the four timescales above. This may occur when the following conditions are met 1) the condensation sink is less than 10-3-10-4 s-1, 2) nucleation is not occurring, and 3) the gas phase production rate is changing. These conditions are not frequently met for a long period of time in the atmosphere. The PSS assumption allows for time step increases of two to three orders of magnitude during nucleation events in typical atmospheric conditions and has been shown to reduce the nucleation/condensation computational time by a factor of 10-100.
DE: 1610 Atmosphere (0315, 0325)
DE: 3311 Clouds and aerosols
SC: Atmospheric Sciences [A]
MN: 2007 Fall Meeting