Modelling cloud processing of aerosol during the ACE-2 HILLCLOUD experiment

Citation
Mj. Flynn et al., Modelling cloud processing of aerosol during the ACE-2 HILLCLOUD experiment, TELLUS B, 52(2), 2000, pp. 779-800
Citations number
40
Categorie Soggetti
Earth Sciences
Journal title
TELLUS SERIES B-CHEMICAL AND PHYSICAL METEOROLOGY
ISSN journal
02806509 → ACNP
Volume
52
Issue
2
Year of publication
2000
Pages
779 - 800
Database
ISI
SICI code
0280-6509(200004)52:2<779:MCPOAD>2.0.ZU;2-4
Abstract
A numerical model has been used to simulate the conditions observed during the ACE-2 Hillcloud experiment and to study the processes which may be taki ng place. The model incorporates gas phase chemistry of sulphur and nitroge n compounds upstream of the cloud, and the interaction of aerosol, precurso r trace gases and oxidants within the cloud. Gas phase and aerosol inputs t o the model have been provided from measurements made in the field. Dynamic s of the air flow over the hill consisted of simple prescribed dynamics bas ed on wind speed measurements, and also for some cases modelled dynamics. I n this modelling study, it was found that during clean case studies particl es down to 40-55 nm diameter were activated to form cloud droplets, the tot al number of droplets formed ranging from 200 to 400 drops/cm(3) Significan t modification of the aerosol spectra due to cloud processing was observed. In polluted cases particles down to 65-80 nm diameter were activated to fo rm cloud droplets, the total number of droplets ranging from 800 to 2800 dr ops/cm(3). Modification of the aerosol spectra due to cloud processing was slight. In all cases, changes in the aerosol spectra were due to both the u ptake of HNO3, HCl, NH3 and SO2 from the gas phase, (the SO2 being oxidised to sulphate) and the repartitioning of species such as HNO3, HCl, and NH3 from larger particles onto smaller ones. Modelling results have been compar ed with observations made. Modelled droplet numbers are typically within 20 % of the best measured values. The mode of the droplet distribution typical ly around 10 20 mu m for clean cases and 4-8 mu m for polluted cases was fo und to be in good agreement with the measured values of 10-25 mu m for clea n cases, but not in such good agreement for polluted cases. Measurements of upwind and interstitial aerosol distributions showed that the smallest par ticles activated were 30 and 50 nm for clean and polluted cases respectivel y, slightly smaller than the model values quoted above. Measured upwind and downwind aerosol spectra showed similar modification to that predicted by the model in eight out of the eleven model runs carried out. Chemistry meas urements also give general evidence for both the uptake of species from the gas phase, and repartitioning of species from large particles onto smaller ones, though comparisons For individual cases are more difficult. From thi s modelling study, it can be concluded that in general, in the remote envir onment the exchange of hydrochloric acid, nitric acid and ammonia between a erosol particles and take up from the gas phase in the vicinity of cloud ma y be a very important mechanism in regulating the evolution of the aerosol spectrum. Further, the much more linear relationship between cloud droplet and accumulation mode aerosol number, which was observed in the measurement s made during the ACE-2 HILLCLOUD project is supported by these modelling r esults. The implications of this for the indirect effect will be explored i n future work.