1 Characterisation factors for life cycle impact assessment of sound emissions 1 2 Cucurachi S. a* , Heijungs R . a 3 a Institute of Environmental Sciences (CML), Leiden University, P.O. Box 9518, 2300 RA Leiden, 4 The Netherlands 5 * Corresponding author. Tel.: +31 (0) 71 527 1478; e-mail: cucurachi@cml.leidenuniv.nl 6 This is an earlier version of the following article:, which has been published in final form at 7 19 http://onlinelibrary.wiley.com/doi/10.1002/ieam.269/abstract. 8 This document is the unedited Pre-print Version of a Submitted Work that was subsequently 9 accepted for publication in Science of the Total Environment, copyright © ELSEVIER after 10 peer review. To access the final edited and published work see: 11 http://www.sciencedirect.com/science/article/pii/S0048969713008619 12 (DOI: 10.1016/j.scitotenv.2013.07.080) 13 14 Abstract 15 Noise is a serious stressor affecting the health of millions of citizens. It has been suggested that 16 disturbance by noise is responsible for a substantial part of the damage to human health. However, no 17 recommended approach to address noise impacts was proposed by the handbook for life cycle 18 assessment (LCA) of the European Commission, nor are characterisation factors (CFs) and appropriate 19 inventory data available in commonly used databases. This contribution provides CFs to allow for the 20 quantification of noise impacts on human health in the LCA framework. Noise propagation standards 21 and international reports on acoustics and noise impacts were used to define the model parameters. 22 Spatial data was used to calculate spatially-defined CFs in the form of 10-by-10-km maps. The results 23 of this analysis were combined with data from the literature to select input data for representative 24 archetypal situations of emission (e.g. urban day with a frequency of 63 hertz, rural night at 8000 25 hertz, etc.). A total of 32 spatial and 216 archetypal CFs were produced to evaluate noise impacts at a 26 European level (i.e. EU27). The possibility of a user-defined characterisation factor was added to 27 support the possibility of portraying the situation of full availability of information, as well as a 28 highly-localised impact analysis. A Monte Carlo-based quantitative global sensitivity analysis method 29 was applied to evaluate the importance of the input factors in determining the variance of the output. 30 The factors produced are ready to be implemented in the available LCA databases and software. The 31 spatial approach and archetypal approach may be combined and selected according to the amount of 32 information available and the life cycle under study. The framework proposed and used for 33 calculations is flexible enough to be expanded to account for impacts on other target subjects than 34 humans and to other continents than Europe. 35