HR: 14:10h
AN: A13E-03    [Abstracts]
TI: CCN Predictions: Is Theory Sufficient for Indirect Forcing Calculations?
AU: * Sotiropoulou, R P
EM: rsot@eas.gatech.edu
AF: Georgia Institute of Technology, School of Earth and Atmospheric Sciences 311 Ferst Drive, NW, Atlanta, GA 30332 United States
AU: Medina, J
EM: jmedina@chbe.gatech.edu
AF: Georgia Institute of Technology, School of Chemical Engineering 311 Ferst Drive, NW, Atlanta, GA 30332 United States
AU: Nenes, A
EM: nenes@eas.gatech.edu
AF: Georgia Institute of Technology, School of Earth and Atmospheric Sciences 311 Ferst Drive, NW, Atlanta, GA 30332 United States
AU: Nenes, A
EM: nenes@eas.gatech.edu
AF: Georgia Institute of Technology, School of Chemical Engineering 311 Ferst Drive, NW, Atlanta, GA 30332 United States
AB: There are numerous studies that assess the ability to predict ambient cloud condensation nuclei (CCN) concentrations from aerosol size and composition measurements. All conclude that predictions often are significantly different from observations (e.g. Martin et al., 1994; Liu et al., 1996; Roberts et al., 2002; VanReken et al., 2003). This discrepancy may arise from many factors, the most common identified is the inability of current theory to fully describe CCN activity. Despite this "CCN closure" problem, predicting cloud droplet number concentration (CDNC) from observations of cloud updraft velocity and aerosol size/composition has proven to be remarkably successful, even for cases where CCN predictions are known to be significantly different from measurements (Snider and Brenguier, 2000; Snider et al., 2003; Conant et al., 2004). Since GCM assessments of the aerosol indirect effect require accurate predictions of cloud droplet number, field studies may suggest that a relatively large error in CCN concentration may not necessarily yield large errors in CDNC. This study focuses on quantitatively assessing the sensitivity of cloud droplet number to errors in predicted CCN concentrations. For this we use ground-based CCN and aerosol measurements obtained during the ICARTT campaign (July-August 2004) at the UNH Thompson Farm site to produce "observed" and "calculated" CCN spectra. These spectra are then introduced into a parameterization of cloud droplet formation (Nenes and Seinfeld, 2003), where errors in CCN concentration (i.e., difference between "observed" and "calculated" CCN spectra) can directly be related to resulting cloud droplet number errors. This exercise is repeated for all the CCN spectra in the dataset and for a wide range of cloud updraft velocity; this allows for determining conditions for which large errors in CCN concentration do not yield significant errors in cloud droplet number (and vice versa).
DE: 0305 Aerosols and particles (0345, 4801, 4906)
DE: 0320 Cloud physics and chemistry
DE: 3311 Clouds and aerosols
SC: Atmospheric Sciences [A]
MN: Fall Meeting 2005