EVALUATION AND FORECASTING OF THE ATMOSPHERIC CONCENTRATIONS OF ALLERGENIC POLLEN IN EUROPE Mikhail Sofiev 1 , Pilvi Siljamo 1 , Hanna Ranta 2 , Auli Rantio-Lehtimaki 2 1 Finnish Meteorological Institute, Helsinki, FInland 2 University of Turku, Aeronbiological Group, Turku, Finland 1 INTRODUCTION Diseases in the respiratory system due to aeroallergens, such as rhinitis and asthma, are major causes of a demand for increased healthcare, loss of productivity and an increased rate of morbidity. Pollenosis accounts for 12 - 45 % of overall allergy cases. The sensitisation to pollen allergens is increasing in most European regions. The adverse health effects of allergens can be reduced by pre- emptive medical measures. However, their planning requires reliable forecasts of high atmospheric pollen concentrations (Rantio-Lehtimäki, 1994), (Rantio-Lehtimäki & Matikainen 2002). There is convincing evidence that the long-range transport of pollen from remote regions can significantly modify pollinating seasons in many European regions. This is particularly important for Northern Europe - and especially for Finland, where the flowering takes place later in spring. This transport causes unforeseen and sudden increases of concentrations of pollen that can occur up to a month before the start of the local pollen season (Siljamo et al, 2004). The long-range transport can substantially increase the concentrations of allergenic pollen also during the local pollen seasons. However, the currently available pollen forecasts are based solely on local observations and do not consider the transport from other regions. At present, there is no modelling system in Europe that can simulate the pollen transport in the atmosphere. Further, there is no such model for evaluating the pollen emissions (including the pollinating season and the flowering characteristics of the relevant species) that would provide the input data for such atmospheric dispersion modelling. The current paper presents an on-going project of the Finnish Academy aiming at development of a numerical dispersion model for operational forecasting the atmospheric transport of natural pollen in Europe. Overall objectives of the project are:  to develop an integrated modelling system for simulating and forecasting the natural pollen emissions and transport at a European scale;  to evaluate the spatial distributions of pollen emissions and concentrations in Europe. 2 MATERIALS AND METHODS The birch pollen is the most important allergen with regard to atmospheric transport due to its ability to fly over large distances. There are two treelike birch species in Europe. Downy birch (Betula pubescens) is the most common in the northern part of Europe, while silver birch (Betula pendula) is dominating in more southern regions. Typical birch pollen grain has a size of 20-22 μm. It is fairly light (a full grain filled with protein material has a density of ~ 800 kg m -3 ), and approximately spherical. It is therefore a comparatively typical coarse aerosol particle, which behaviour in the atmosphere is more or less known. However, already the birch pollen is bigger and heavier than a “typical” regionally- dispersed aerosol, which raises a set of questions regarding the existence of the whole phenomenon. The three key questions to solve are: (i) whether the grain meets the assumptions behind all existing dispersion models, so that it can be treated as a “standard” (albeit coarse) atmospheric aerosol, (ii) what are the sources of pollen, their features and predictability by means of existing models; (iii) what are the features of such a pollutant, its physical and chemical transformations in the atmosphere and processes removing the grains from the air, which made it regionally- (or continentally- ) dispersing? There are two types of the European-wide observations that can be used for answering the above questions, as well as for the development, initialization and verification of the pollen transport model: phenological observations of the seasonal development of canopies, and measurements of the atmospheric pollen concentrations. The networks cover most of Europe in space, several decades in