A COMPARISON OF ODOUR COMPLAINT STATISTICS WITH MODEL CALCULATIONS OF ODOUR EPISODES

Abstract

The Austrian odour dispersion model (AODM) is a Gaussian model suitable for the prediction of ambient odour concentrations. Based on cumulative frequency distributions of calculated odour concentrations at receptor points, separation distances are obtained defined by odour impact criteria chosen as a combination of odour threshold and probability of threshold exceedance. At these separation distances, depending on the wind direction, the occurrence of odour sensation is analysed and compared with the well-known time pattern of the complaint statistics for odour. Here, the AODM is used to calculate separation distances for an odour threshold of 1 odour unit (OU) per cubic metre exceeded in 3% of the year. At a site in the Austrian North-alpine foreland, direction-dependent separation distances for a 1000 head pig unit (calculated on the basis of a two-year time series of meteorological data) lie between 99 m for northerly winds and 362 m for westerly winds. For these directiondependent separation distances we analysed when odour sensation can be expected in relation to meteorological parameters as well time of the day and year. For the main wind directions, West and East, odour sensation can be expected more often for higher wind velocities and a neutrally or stably stratified atmosphere around sunset. North and South winds show the typical diurnal variation of a local valley wind system with predominantly northerly daytime up-valley and southerly night-time down-valley winds. Odour sensation is therefore most likely around noon for North wind and during night time for South wind. This time pattern of the calculated odour sensation doesn't fit to the time pattern of the complaint statistics which shows complaints to occur predominantly in the afternoon and evening hours of the warm season when residents are outside. The presented comparison of odour complaint statistics with the calculated odour episodes is a helpful tool to find out when odour is perceived as most annoying. As a result, the evaluation of these values by the odour impact criteria should not only be based on statistical limits as it is done today but also by considering the annoying potential of odour due to the behaviour of the neighbours. Therefore the odour episode should be weighted by the time of the day and time of the year, as is done with the limit values for noise.