In Situ Aerosol Size Distributions and Clear Column Radiative Closure During ACE-2

As part of the second Aerosol Characterization Experiment (ACE-2) during June and July of 1997, aerosol size distributions were measured on board the CIRPAS Pelican aircraft through the use of a DMA and two OPCS. During the campaign, the boundary layer aerosol typically possessed characteristics representative of a background marine aerosol or a continentally influenced aerosol, while the free tropospheric aerosol was characterized by the presence or absence of a Saharan dust layer. A range of radiative closure comparisons were made using the data obtained during vertical profiles flown on four missions. Of particular interest here are the comparisons made between the optical properties as determined through the use of measured aerosol size distributions and those measured directly by an airborne 14-wavelength sunphotometer and three nephelometers. Variations in the relative humidity associated with each of the direct measurements required consideration of the hygroscopic properties of the aerosol for size distribution based calculations. Simultaneous comparison with such a wide range of directly measured optical parameters not only offers evidence of the validity of the physicochemical description of the aerosol when closure is achieved, but also provides insight into potential sources of error when some or all of the comparisons result in disagreement. Agreement between the derived and directly measured optical properties varied for different measurements and for different cases. Averaged over the four case studies, the derived extinction coefficient at 525 nm exceeded that measured by the sunphotomoter by 2.5% in the clean boundary later, but underestimated measurements by 13% during pollution events. For measurements within the free troposphere, the mean derived extinction coefficient was 3.3% and 17% less than that measured by the sunphotometer during dusty and nondusty conditions, respectively. Likewise, averaged discrepancies between the derived and measured scattering coefficient were -9.6%, +4.7%, +17%, and -41% for measurements within the clean boundary layer, polluted boundary layer, free troposphere with a dust layer, and free troposphere without a dust layer, respectively. Each of these quantities, as well as the majority of the > 100 individual comparisons from which they were averaged, were within estimated uncertainties.