Publications

An outline is presented here of the Po Valley Fog Experiment 1989, carried out within the EUROTRAC-GCE project. This experiment is a joint effort by several European research groups from 5 countries. The physical and chemical behaviour of the fog multiphase system was studied experimentally following the temporal evolution of the relevant chemical species in the different phases (gas, droplet, interstitial aerosol) and the evolution of micrometeorological and microphysical conditions, from the pre-fog situation through the whole fog evolution, to the post-fog period. Some general results, useful for describing the general features of the fog system, are presented here, while specific scientific questions on the different processes taking place within the system itself will be addressed in other companion papers of this same issue.

Simultaneous measurements of several non-volatile species in unscavenged aerosol particles and in fog droplets have revealed differences in partitioning for different chemical species. The average scavenged fraction of sulphate was 18 % and the corresponding fraction of elemental carbon was only 6 %. This suggests that the aerosol was externally mixed, and that the chemical mixture of the aerosol as a function of size is important in the context of nucleation scavenging. The measurements obtained could not distinguish between the two primary hypotheses for explaining the observed differences, (a) that the particles had the same size distribution and their chemical composition was the controlling factor, and (b) that the elemental carbon was associated with smaller particles than the sulphate, so that the difference in scavenging efficiency was controlled by the size distribution of the particles.

A Tandem Differential Mobility Analyser (TDMA) was used to study the hygroscopic growth of individual ambient aerosol particles in the Po Valley, Italy. The measurements were made during the GCE fog experiment in November 1989. During fog, the interstitial aerosol (D(p)(at ambient relative humidity) < 5 mum) was sampled. Two modes of particles with different hygroscopic growth were found for 0.030 mum < D(p)(dry) < 0.20 mum. On average, the fraction of particles in the two modes were almost equal. The mean growth factor at 85% r.h. was 1.44 +/- 0.14 for the more-hygroscopic mode and 1. 1 +/- 0.07 for the less-hygroscopic mode. The growth factors and the proportion of the particles that were less hygroscopic varied considerably from day to day, but no significant size dependence was seen. Comparison of growth factors for pure salt particles and the measured growth factors indicates that both hygroscopic modes contain a major insoluble part. The effect of the external mixing of hygroscopic properties on the activation of particles to fog droplets is discussed and the fraction of particles that were activated as a function of particle size is predicted. Comparison with the measured scavenging fraction as a function of particle size shows that the hygroscopic properties of the individual particle are as important as the particle size in determining if it will be activated in a fog.