Author(s): Dekoninck Luc, Lercher Peter, Botteldooren Dick
Currently, traffic noise annoyance estimates are almost exclusively based on exposure-effect relationships derived from linking questionnaires to (simulated) LDEN noise maps. In particular in situations where local roads, main roads, highways, and railways contribute to annoyance, replacing LDEN with an alternative continuous indicator explaining the annoyance reported the citizens would be advantageous. The quality of the noise maps is directly related to the quality and the spatial resolution of the underlying traffic data. The uncertainties on the traffic data are rather high for low exposure roads while most of the population is living near these local roads. At the same time, peak levels and diurnal patterns may vary a lot amongst these exposure situations. Hence, performing noise measurement at each dwelling for quantifying exposure and linking it to annoyance surveys, could be an alternative, but that is even with cheap distributed monitoring networks not achievable. Spatially mapping of measurements and surveys might close this gap. To be able to feed the model with as many noise measurements as possible, we propose to use statistical attributes that are readily available in most measurement campaigns. Any existing noise measurement should allow to calculate the exposure indicator and feed the model. Features including the levels and diurnal aspects of LAeq, LA01, LA10, LA50, LA90, LA95, LAmax are therefore combined into a single Noise Quality Index (NQI). An NQI equation structure was derived based on the UK Noise Incidence dataset (1000 locations). This NQI is spatially mapped from the available measurements gathered in the past 15 years in various projects and consulting work (80-100) in Flanders on the Flemish Noise Annoyance questionnaire responses. Some preliminary results and its applicability in assessing noise impact of large infrastructure projects in Antwerp, Belgium and the alpine areas will be presented.
Name: Mr Luc Dekoninck