Source attribution of aerosol size distributions and model evaluation using Whistler Mountain measurements and GEOS-Chem-TOMAS simulations
- 1Department of Atmospheric Science, Colorado State University, Fort Collins, CO, USA
- 2W.M. Keck Science Department, Scripps College, Claremont, CA, USA
- 3Air Quality Research Division, Atmospheric Science and Technology Directorate, Science and Technology Branch, Environment Canada, Downsview, ON, Canada
Abstract. Remote and free-tropospheric aerosols represent a large fraction of the climatic influence of aerosols; however, aerosol in these regions is less characterized than those polluted boundary layers. We evaluate aerosol size distributions predicted by the GEOS-Chem-TOMAS global chemical transport model with online aerosol microphysics using measurements from the peak of Whistler Mountain, British Columbia, Canada (2182 m a.s.l., hereafter referred to as Whistler Peak). We evaluate the model for predictions of aerosol number, size, and composition during periods of free-tropospheric (FT) and boundary-layer (BL) influence at "coarse" 4° × 5° and "nested" 0.5° × 0.667° resolutions by developing simple FT/BL filtering techniques. We find that using temperature as a proxy for upslope flow (BL influence) improved the model–measurement comparisons. The best threshold temperature was around 2 °C for the coarse simulations and around 6 °C for the nested simulations, with temperatures warmer than the threshold indicating boundary-layer air. Additionally, the site was increasingly likely to be in cloud when the measured relative humidity (RH) was above 90 %, so we do not compare the modeled and measured size distributions during these periods. With the inclusion of these temperature and RH filtering techniques, the model–measurement comparisons improved significantly. The slope of the regression for N80 (the total number of particles with particle diameter, Dp, > 80 nm) in the nested simulations increased from 0.09 to 0.65, R2 increased from 0.04 to 0.46, and log-mean bias improved from 0.95 to 0.07. We also perform simulations at the nested resolution without Asian anthropogenic emissions and without biomass-burning emissions to quantify the contribution of these sources to aerosols at Whistler Peak (through comparison with simulations with these emissions on). The long-range transport of Asian anthropogenic aerosol was found to be significant throughout all particle number concentrations, and increased N80 by more than 50 %, while decreasing the number of smaller particles because of suppression of new-particle formation and enhanced coagulation sink. Similarly, biomass burning influenced Whistler Peak during summer months, with an increase in N80 exceeding 5000 cm−3. Occasionally, Whistler Peak experienced N80 > 1000 cm−3 without significant influence from Asian anthropogenic or biomass-burning aerosol. Air masses were advected at low elevations through forested valleys during times when temperature and downwelling insolation were high, ideal conditions for formation of large sources of low-volatility biogenic secondary organic aerosol (SOA). This condensable material increased particle growth and hence N80. The low-cost filtering techniques and source apportionment used in this study can be used in other global models to give insight into the sources and processes that shape the aerosol at mountain sites, leading to a better understanding of mountain meteorology and chemistry.