Aerosol formation and subsequent particle growth in the ambient air have been frequently observed at the boreal forest site (SMEAR II), Southern Finland [Mäkelä et al., 1997; Kulmala et al., 1998]. The EU-funded project BIOFOR aims to a) determine formation mechanisms of aerosol particles in the boreal forest site, and b) verify emissions of secondary organic aerosols from the boreal forest site and quantify the amount of condensable vapors produced in photochemical reactions of biogenic volatile organic compounds (BVOC) leading to aerosol formation.

The BIOFOR project started in October 1997 and will continue until the end of 1999. During the project three international joint measurement campaigns have been carried out at the SMEAR II station. The BIOFOR Consortium consists of six laboratories: University of Helsinki Department of Physics (M. Kulmala, J.M. Mäkelä, P. Aalto, K. Hämeri, L. Pirjola; aerosol dynamics, modeling and measurements, inorganic gases, gas and aerosol fluxes), Fraunhofer-Institute for Atmospheric Environmental Research (W. Seidl; modelling of particulate phase and atmospheric chemistry, canopy model), Institute of Spectrochemistry and Applied Spectroscopy (T. Hoffmann; VOC profiles), Stockholm University Air Pollution Laboratory (H.-C. Hansson, R. Jansson, E.D. Nilsson (aerosol and gas measurements, boundary layer meteorology), University of Sunderland, Centre for Marine and Atmospheric Sciences (C.D. O'Dowd; aerosol measurements) and Finnish Meteorological Institute Air Quality Research (Y. Viisanen, R. Hillamo; trajectories, soundings, aerosol phase chemistry). Additionally, there are several individual researchers participating during the intensive campaigns: K. Bigg (single particle analysis), R. Weber (ultrafine aerosol measurements), J. Salm (ion measurements), and A. Laaksonen (nucleation modelling).

Figure 1. Monthly frequency of nucleation events during three years [J.M. Mäkelä and M. dal Maso, unpublished results].

At SMEAR II, continuous measurements of submicron aerosol number size distribution have been carried out since January 1996 (every 10 min). They show approximately 40 days per year with clearly detectable aerosol particle formation events. As seen in Figure 1, the most typical time for these events is March-April. Subsequent to the new particle formation, significant particle growth is usually observed. Almost 20% of the events will continue sufficiently long to produce particles with diameter over 80 nm, which can then become effective cloud condensation nuclei (CCN).

Figure 2. Typical number size distribution during the particle production and growth event. A sum of three log-normal size distributions (solid curve) has been numerically fitted into the measured spectrum (*). The characteristic numbers for the three modes (dashed curves)-i.e., number concentration, mean particle diameter and geometric standard deviation-for each mode are shown with the modes.

When the particle formation event occurs, the mode of the fresh particles appears into the measurement range. Figure 2 shows a typical aerosol size distribution spectrum measured using DMPS (Differential Mobility Particle Sizing) during the event. Here, three log-normal modes have been used to explain the spectrum's structure. The nucleation mode, which has already reached the size of 17.5 nm, practically dominates the spectrum with its high number concentration. Figure 3 shows the Figure 3 shows the evolution of the modal mean diameters of nucleation, Aitken, and accumulation mode particles during the particle production bursts, and subsequent growth. The fitting procedure, by which the modes have been obtained from the measurement data, has been presented by Mäkelä et al. [1999]. In Figure 3, an extended growth event is illustrated following a nucleation burst on July 20th 1996. For this event, particle growth from nucleation mode up to accumulation mode during the subsequent days is clearly observable.

Figure 3. Evolution of the mean diameters of nucleation (*), Aitken (+) and accumulation (o) mode particles during a period of five days in Hyytiälä. The values for modal mean diameters have been obtained by numerical fitting procedure.

Meteorological data including radiation is available for interpretation of weather conditions typically encountered during particle production events (see BIOFOR homepage). Vertical distributions ( 4­67 m) of different inorganic gases (SO2, NOx and O3) are also available.

During the BIOFOR project three campaigns have been performed, measuring hygroscopicity, composition and vertical profiles of particles. The atmospheric concentrations of organic and inorganic gases are also available during the campaign.

Measurements carried out so far demonstrate the capability of combining more detailed campaign measurements with continuous ones, e.g., the detailed chemical analysis will provide insight into composition of newly formed particles. However, the task to analyze chemical composition of nucleation mode particles is very difficult. On the other hand, the hygroscopicity of nucleation mode particles tells us their soluble fraction and changes in particle composition in situ. Some changes have already been seen during the growth process.

As a summary, two different cases can be compared (for more details see BIOFOR homepage):

a)	Nucleation day in spring, when 5,000­10,000 new particles cm-3 are usually formed coinciding with moderate hydrocarbon emissions from the forest. General meteorological conditions comprise northwest (polar) air trajectories, relatively low ambient temperatures, and large night and day temperature differences.
b)	Typical clear sunny summer day, with no nucleation observed; however, high hydrocarbon emissions from the forest are encountered. Ambient temperatures are typically high, temperature difference between day and night being relatively low.

From particle flux data, using the eddy covariance method [Buzorius et al., 1998], a small overall downward flux is usually observed. This clearly increases during nucleation events, with an exception of the cases when the surface wind was from direction of 220-250 (direction of the city Tampere and the Hyytiälä institute

ings); then a strong upward particle flux is observed.

As mentioned above, one of the specific scientific goals of the BIOFOR project is to investigate the connection between the observed nanometer particle formation events and the subsequent particle growth at the SMEAR station with the biogenic activity of the surrounding Scotch pine forest. From the emission and concentration measurements realized so far as well as from earlier studies, it is known that various monoterpenes are the dominant VOCs released from the conifers at the measuring site. Especially a-pinene and d-3-carene are emitted by Pinus sylvestris and both VOC species are known to produce a series of condensable products after photochemical oxidation [Griffin et al., 1999; Hoffmann et al., 1997]. In principle, the atmospheric concentrations of these aerosol forming products are controlled by the actual emission rate of the precursors (which depends on temperature, radiation and biological activity of the plants), the concentration of the oxidizing agents (ozone, OH- and NO3-radicals) and the mixing conditions above and within the forest. Thus, quantitative estimations on the contribution of biogenic VOC oxidation products to the particle phase relies on reliable measurements of a complex data set of different parameter and compounds. Consequently, emissions studies (branch enclosure technique) and VOC flux measurements using the gradient technique were carried out during the three BIOFOR campaigns. Although it is too early to finalize the results, it appears evident that especially the OH reaction of the biogenic VOCs is connected with the measured particle growth process. Maximum emission rates and fluxes of biogenic VOCs as well as a high photochemical production term during daylight conditions are very likely responsible for the observed simultaneous growth of the nucleation mode particles. However, whether the actual nucleation event is also related with low volatile organics from biogenic origin still has to be elucidated.

The modeling activities on BIOFOR [e.g., Pirjola, 1999] are being carried out concurrently in close conjunction with field data interpretation. To study aerosol formation, atmospheric chemistry, aerosol processes and meteorology should be combined, as has been done in the present project. Preliminary modeling of nucleation and growth events shows good agreement with experimental data; however, the exact nucleation mechanism remains unclear. The best guess so far is that ternary nucleation of water-ammonia-sulfuric acid is responsible for nucleation, and the condensation growth to detectable sizes is due to condensation of organic vapours. However, the results are very preliminary and more detailed data analysis and modeling activities are ongoing in order to elucidate the primary formation and growth processes.

References

BIOFOR homepage:

1. Buzorius, G., U. Rannik, J.M. Mäkelä, T. Vesala, and M. Kulmala, Vertical aerosol particle fluxes measured by eddy covaraince technique using condensational particle counter, J. Aerosol Science, 29, 157-171, 1998.
2. Griffin, R.J., D.R. Cocker III, R.C. Flagan, and J.H. Seinfeld, Organic aerosol formation from the oxidation of biogenic hydrocarbons, J. Geophys. Res., 104, 3555-3567, 1999.
3. Hoffmann, T., R. Bandur, U. Marggraf, and M. Linscheid, Molecular composition of organic aerosols formed in the alpha-pinene/ozone reaction: Implications for new particle formation processes, J. Geophys. Res., 103, 25569-25578, 1998.
4. Kulmala, M., A. Toivonen, J.M. Mäkelä, Analysis of the growth of nucleation mode particles observed in Boreal forest, Tellus, 50B, 449-462, 1998.
5. Mäkelä, J.M., P. Aalto, V. Jokinen, A. Nissinen, S. Palmroth, T. Markkanen, K. Seitsonen, H. Lihavainen, and M. Kulmala, Observations of ultrafine aerosol particle formation in boreal forest, Geophys. Res. Lett., 24, 1219-1222, 1997.
6. Mäkelä, J.M., I.K. Koponen, P. Aalto and M. Kulmala, One-year data of submicron size modes of tropospheric background aerosol in Southern Finland, J. Aerosol Sci. (submitted).
7. Pirjola, L., Effects of the increased UV radiation and biogenic VOC emissions on ultrafine sulfate aerosol formation, J. Aerosol Science, 30, 355-367, 1999.