Abstract : 1B.8
Statistical forecasting of air quality in the Inn valley

Esther Griesser, Johannes Vergeiner
Institute of Meteorology and Geophysics, University of Innsbruck, Austria

We present a model aiming on short time statistical forecasting of air pollution in the Austrian Inn- and lower Wipp Valley. The model was developed within the Alpine Space Project ALPNAP. It is based on multiple linear regression, conjoining a variety of informations: emission estimates (mainly from traffic and domestic heating), numerical weather prediction forecasts (ECMWF and ALADIN), climatological as well as actual air pollution values, and astronomical parameters. Forecasts are provided for nitrogen oxides (NO and NO2) and particulate matter on five locations within a horizontal distance of about 65 kilometers.

The changing influence of the relevant parameters on the resulting pollutant concentrations is studied for the march of the day as well as for each season. Also, the effect of using two different types of models - global circulation model ECMWF versus mesoscale model Aladin - is investigated.

In a further step, attempts are made on an objective classification of weather types significant for the development of air pollution within the concerned valleys. It is analysed whether such a segregation between weather types improves the score of the model.