Combustion Aerosol, Entrainment and Clouds in the VOCALS Region

Extensive stratus clouds in the subtropical Southeast Pacific (SEP) respond to dynamic, thermodynamic, diurnal, precipitation, entrainment and microphysical influences that are all coupled to the land-ocean environment. As in all clouds, the nature and variability of aerosol effective as cloud condensation nuclei (CCN) exert a fundamental control over the number of cloud droplets that can form at available cloud supersaturation. Here we draw attention to the role of natural and combustion aerosol present in the region and their relation to clouds and/or pockets of open cells (POC's).

The NCAR C-130 flew 14 flights out of Arica, Chile and 140 hours in the VOCALS region between Oct. 15 and Nov. 15. These included 8 survey flights along 20S between 72W and 85W, 4 POC flights across a POC/cloudy region boundary and two N-S flights off the coast between about 20S and 30S latitude. All flights typically included legs about 10 min long near the surface, near cloud base, in cloud and above cloud with occasional profiles to about 4 km. POC flight legs were “stacked” above each other and were typically advected with the wind in an attempt to stay in the same airmass. The C-130 was equipped with extensive aerosol, gas, remote sensing and meteorological instruments but we will focus here on the relation between CO and various aerosol measurements. The latter reported here include cloud condensation nuclei (CCN), aerosol chemistry via AMS and condensation nuclei (CN), non-volatile CN at 360C, Black Carbon (BC) mass for a Single Particle Soot Photometer (SP2).

Our survey data along 20S revealed “patches” and “rivers” of elevated CO and combustion aerosol overlaying the inversion and often immediately above cloud level. Aerosol measurements indicate that a large number fraction of this aerosol has a size and composition making them effective CCN and their enhanced concentrations aloft can supply the MBL with CCN through entrainment. We find that air in the marine boundary layer (MBL) for low cloud regions and POC's consistently has lower carbon monoxide (CO) concentrations and associated aerosol compared to adjacent cloudy regions. As the lifetime of CO is insensitive to cloud processes and is on the order of 2 months, CO is conserved over the time scales associated with advection through the VOCALS region and provides a tracer for combustion derived aerosol.

Aerosol larger than 60 nm were commonly effective as CCN at less than 0.3% supersaturation (CCN0.3)in the region. Increases in aerosol number larger than this size tracked with increases in CO both above and below cloud. CCN0.3 in the clean MBL included natural sulfates and sea-salt aerosol and similar number concentrations were often seen above and below cloud while sizes aloft were often somewhat smaller. Once entrained into the MBL, growth through heterogeneous chemistry can also be expected to increase their size. POC regions also showed depletion of aerosol in CCN0.3 to values below 10 cm-3, presumably due to drizzle. These observations raise the question: "Can rivers or patches of advected combustion aerosol entrained into a drizzling stratus deck help stabilize clouds against drizzle removal and increase cloud persistence or cloud fraction?"

The larger scale meteorological regime in the SEP region influencing VOCALS is governed by anticyclonic flow linked to large scale subsidence. Analysis of these wind fields using the Weather and Research Forecast model for advection and divergence terms result in a residual that is interpreted as an entrainment velocity of about 0.4 cm/s along 20S (Rahn and Garreaud, 2010). These mean entrainment rates are probably higher in cloudy regions (say 0.5 cm s-1) and could mix in about 450 m of air from above the inversion in a day. Our observations suggest that the cloud free regions and POC regions, characterized by the lower CO values, have not entrained significant CO or pollution aerosol from aloft. However, adjacent cloudy regions with higher CO in the MBL must have entrained CO and aerosol from aloft. We hypothesize that more effective turbulent entrainment of pollution aerosol from above in cloudy regions can buffer depletion of MBL aerosol by drizzle when such aerosol is present. If clean FT air with lower CCN is entrained, the buffering against removal by drizzle may be much reduced and depletion of CCN in the MBL will be more rapid. Eventually, this depletion could possibly suppress cloud and drizzle formation and possibly lead to POC like conditions.

This process offers a means by which long range transport of FT aerosol can affect cloud properties in this and similar regions of extended stratus. While our data is consistent with this process, continued analysis and modeling is underway to challenge its effectivness. The effect of entrained pollution can be expected to maintain higher levels of CCN in MBL clouds compared to pollution free regions, thereby increasing cloud albedo relative to the case with clean air aloft. Perhaps more importantly, this resupply of CCN into the MBL would increase cloud persistence and lead to increased mean cloud fractions over the region. An increase in cloud fraction may be a far greater effective influence on regional cloud albedo (George and Wood, 2010). As most combustion sources are linked to anthropogenic activity, then one can ask whether combustion aerosol above extended stratus have changed over time (decades) and whether these extended cloud systems have properties that have been changed due to their presence. Due to the significance of these large scale cloud systems to global albedo, we will employ the VOCALS data set to challenge and explore this hypothesis and its significance.

References

George, R and R. Wood, Subseasonal variability of low cloud and radiative properties over the southeast Pacific Ocean, Atmos. Chem. Phys. Discuss., 9, 25275-25321, 2009

Rahn, D. and R. Garreaud, Marine boundary layer over the subtropical southeast Pacific during VOCALS-REx. Part II: Synoptic variability, submitted November 2009.