Analysis of tropospheric aerosol number density for aerosols of 0.2- to 3-1xm diameter: Central and northeastern Canada

. NASA's Atmospheric Boundary Layer Experiment conducted during the summer of 1990 focused on the distribution of trace species in central and northeastern Canada (altitudes <6 km) and the importance of surface sources/sinks, local emissions, distant transport, tropospheric/stratospheric exchange. Aircraft flights were based from North Bay, Ontario, and Goose Bay, Labrador, Canada. As part of the aircraft measurements, aerosol number density (0.2- to 3-/am diameter) was measured using an optical laser technique. Results show that summertime aerosol budgets of central and northeastern Canada can be significantly impacted by the transport of pollutants from distant source regions. Biomass burning in Alaska and western and central Canada exerts major influences on regional aerosol budgets. Urban emissions transported from the U.S./Canadian border regions are also important. Aerosol enhancements (mixed layer and free troposphere) were most prevalent in air with carbon monoxide mixing ratios > 110 parts per billion by volume (ppbv). When data were grouped as to the source of the air (5-day back trajectories) either north or south of the polar jet, aerosol number density in the mixed layer showed a tendency to be enhanced for air south of the jet relative to north of the jet. However, this difference was not observed for measurements at the higher altitudes (4 to 6 km). For some flights, mixed layer aerosol number densities were > 100 higher than free-tropospheric values (3- to 6-km altitude). The majority of the observed mixed layer enhancement was associated with transport of effluent-rich air into the Canadian regions. Aerosol emissions from natural Canadian ecosystems

The ABLE 3B instrument complement is described by Harriss et al. [this issue]. The in situ aerosol package was similar to that used during earlier ABLE missions (see, for example, Gregory et al. [1990Gregory et al. [ , 1992). An active scattering aerosol spectrometer probe (ASASP) (particle measurement system model ASASP-100) mounted external to the aircraft measured aerosol number density (#/cm 3) as a function of time (1-s frequency) and size diameter (0.12-to 3.12-/am diameter). The ASASP sizes aerosols into 15 bins that progressively increase in width (e.g., bin 1, 0.025/am wide and bin 15, 0.375 /am wide). As a result of an instrument malfunction, data measured in the two smallest bins of the instrument (0.12 to 0.2 /am) are not reported. The aerosol data are averaged over 10-s or l-min periods. The data have not been corrected for sampling uncertainties [e.g., Pinnick and Auvermann, 1979;Patterson et al., 1980;Garvey and Pinnick, 1983;Dye and Baumgardner, 1984;Pinnick et al., 1981], and the size bin calibration of the probe was performed by the manufacturer. Data measured during takeoff and landing and sampling within clouds are excluded from the analyses.

Experimental Techniques
The spheric data (potential temperatures of 310 ø to 320 ø ) of Figures 4a and 4b where the x axis scale has been expanded. As shown by this comparison, aerosol loadings above about 4 to 6 km are similar whether the air was from sources south or north of the polar jet. In fact, comparison of the total chemical signature (CO, nitrogen oxides, nonmethane hydrocarbons, fluorocarbons, etc.) for the same data groupings also showed little difference at the higher altitudes.

Specific Processes
The results from Figures 3 and 4 provide some information into the general nature of air that was observed to be enhanced in aerosol. The data of Figure 5

in combination with back trajectory analyses [Shipham et al., this issue] provide insight into the relative magnitude of specific processes which are important to aerosol loadings in the Canadian regions. Figure 5 shows data on a flight-by-flight basis.
The FT data correspond to measurements above 3-km altitude (same as Figure 2). Data above 3-km altitude are not available for flight 12. The ML data have been expanded to include measurements at altitudes <1 km as compared to <300 m used for Figure 2. This was necessary as some flights did not have sufficient sampling below 300-m altitude to allow box-and-whisker analyses. Figures 5a and 5b are box-and-whisker plots for the FT and ML results, respectively. Figure 5c plots the ratio of the median ML aerosol loading (Figure 5b) to the median FT loading (Figure 5a), i.e., (ML)/(FT). mass trajectories for flights 2, 3, 18, and 19 suggested polar source air at all altitudes with little modification during transport via biomass burning or urban pollution. As shown in Figures 5a and 5b, FT and ML aerosol number densities for these flights are among the lowest observed. In addition, the ratio of ML to FT loadings (Figure 5c) are also among the lowest (ratios (ML)/(FT)) for the individual flights ranged from 1.5 to 7; average of the ratio for the four flights was about 3). As a result and for ABLE 3B data only, we interpret (ML)/(FT) ratios below about 7 as suggesting that the increase in ML aerosol loading (compared to FT) is mostly the result of thermodynamic processes (higher air density and humidity of the ML samples compared to FT), mechanical processes (the presence of wind-blown surface elements in the mixed layer), and/or natural emissions from the underlying ecosystem. (ML)/(FT) ratios as large as a factor of 1.5 to 2 can be expected to occur due to differences in air density for the two altitude ranges.

Polar source, background air. As discussed by Shipham et al. [this issue] and Anderson et al. [this issue], the column of air (surface to about 6-km altitude) at any location and time was generally not from a single source but was often composed of mixtures of polar and mid-latitude air each influenced by different types of emissions. However, air
Canadian ecosystems. Mixed layer aerosol results for flights 2, 3, and 17 illustrate the aerosol production associated with natural Canadian ecosystems. ML data for these flights were mostly low-altitude (150 to 300 m) samplings over remote Hudson Bay lowlands (flights 2 and 3) or eastern Canadian wetland/forest (flight 17) ecosystems during periods in which the air was not significantly impacted by urban or biomass-burning emissions. As noted above, FT air for flights 2 and 3 was of polar origin. FT air for flight 17 was of split origin, consisting of mixtures of polar air similar to flights 2 and 3, north central Canadian air, and air exposed to urban emissions from the Great Lakes region. As shown in Figure

Concluding Remarks
Analyses of aerosol number density data (0.2-to 3-/am diameter) showed that the aerosol budgets of central and northeastern Canada are significantly impacted by the transport of pollutants from distant source regions. In particular, biomass burning in Alaska and western and central Canada exerts a major influence on regional aerosol budgets. Urban emissions transported from the U.S./Canadian border regions are also important.
In all cases, mixed layer aerosol number densities exceeded those of the free troposphere by factors of 1.3 to above 100. The highest mixed layer aerosol loadings (e.g., 100 to 200/cm 3) occurred in air influenced by distant pollution and emissions which were transported into the Canadian regions. When polar source air was relatively clean and devoid of influences from transported emissions, mixed layer number densities were in the range of about 5 to 10/cm 3 and mixed layer aerosol loadings were only about a factor of 3 higher than free-tropospheric values, of which a factor of 1.5 to 2 can be attributed to air density differences of the samples at the different altitudes.
When data were grouped according to the carbon monoxide levels measured in the air, the majority of aerosol enhancements occurred in air with CO >110 ppbv. When data were grouped as being measured in air which was from sources either south (mid-latitude) or north (polar) of the polar jet, the ratio of mixed layer to free-tropospheric number densities averaged about a factor of 100 (midlatitude) compared to factors of only 10 for air north of the jet. While mid-latitude mixed layer air showed a tendency to be enriched in aerosols (compared to polar air), both midlatitude and polar air showed similar aerosol loadings at the higher altitudes (e.g., 4 to 6 km).
The highest free-tropospheric aerosol loadings (3-to 6-km altitude) observed during ABLE 3B occurred in polar air which was influenced by Alaskan biomass-burning emissions injected into the air 4 to 5 days prior to sampling. Associated with the higher free-tropospheric loadings was a sizable increase in mixed layer loadings. However, mixed layer to free-tropospheric number density ratios were only about a factor of 4 (similar to polar background air) and thus the high mixed layer loadings were the result of aerosol-enriched, free-tropospheric air having been transported into the sampling region.