AAAR 32nd Annual Conference
September 30 - October 4, 2013
Oregon Convention Center
Portland, Oregon, USA
Abstract View
Development and Application of a Particle Number Source Tagging Algorithm in an Aerosol Microphysics Model
DANIEL WESTERVELT, Jeffrey Pierce, Peter Adams, Carnegie Mellon University
Abstract Number: 386 Working Group: Source Apportionment
Abstract Aerosol nucleation is an important source of particle number in the atmosphere, but nucleated particles must overcome efficient coagulation and grow significantly to act as cloud condensation nuclei (CCN). Past work has tested the sensitivity of CCN concentrations to changes in nucleation rates, and results have been mixed. Some models have predicted large enhancements such that 60-70% of the CCN in the global boundary layer come from nucleation, whereas others found as little as 5% enhancement. These sensitivity studies are important but limited due to the nonlinear nature of aerosol microphysics models. In this work, we develop and apply a method within the TwO-Moment Aerosol Sectional microphysics algorithm (TOMAS) that would avoid these limitations by tracking or tagging the aerosol number size distribution as it evolves during condensation, coagulation, and nucleation. Tags can be assigned by their source process (e.g. nucleation or primary), species type (e.g. sea salt, black carbon, POA), or even source locations or sectors (e.g. cities, vehicles, industry). This is done in a “semi-offline” fashion, in which the core microphysical calculations still operate on total (untagged) number distributions to avoid computational burden. Instead, the rates of key microphysical processes are saved and passed to another module that updates the tagged distributions accordingly. The approach is similar to the Particle Source Apportionment Technology (PSAT) algorithm, developed for aerosol mass source apportionment in Wagstrom et al. (2008). With the number source tagging algorithm in place, we test coagulation, condensation and nucleation and quantify the amount of numerical diffusion present for each process. We apply our tagging method to observed nucleation events in a box model and calculate the nucleation contribution to CCN concentrations. Finally, we incorporate the algorithm into the GEOS-Chem-TOMAS global aerosol model, which will allow us to quantify the sources of CCN in a more definitive manner.