Cécile Nyffeler

and 4 more

   Lugrin R., Nyffeler C., and Ruas M.    Ecole Polytechnique Fédérale de Lausanne  Introduction In populated areas noise can be one of the major sources of pollution. If the presence of noise seems to go hand in hand with urban areas the introduction of vegetation could possibly reduce the discomfort it can cause to people and the possible effects it has on their health. The aim of this paper is therefore to study the correlation, which exists between greenness in urban areas and noise levels that can be measured both during day and night.As of now several studies have been published, which try to relate the presence of vegetation noise levels and the impact both have on the population. One work by Donald Aylor \cite{Aylor_1972} explored the way different types of vegetation hindered the effects of noise in the presence of different environmental factors (Wind, soil among others). After the observations, models were developed to further enlarge the scope of the study. Results showed vegetation could effectively reduce loudness. The effectiveness of this reduction relied heavily on the vegetation's structure: foliage, stems, ground structure.Another work \cite{Dzhambov_2015} aimed to link the presence of green areas with the attenuation of both psychological stress and physical health issues related to noise pollution. A survey was conducted correlating the presence of greenness and the perception of noise the population had. Results showed that the "Noise sensitivity" was clearly reduced when the access to green spaces was increased.For the present study the initial statement is that in areas with elevated greenness the noise decreases.  The area of study is the municipality of Vernier (Geneva, Switzerland). Day and night levels of noise were measured and a correlation was established with the green areas of the municipality based on satellite pictures in order to verify the hypothesis mentionned above.Data Several raster and vector layers around the municipality of Vernier were used.Four orthophotos cover the zone of interest. They are RVB images already georeferenced and have a ground resolution of 0.5 meter. They probably originate from the federal office of topography Swisstopo but this is not clearly mentioned in the data, likewise for the height model. This height model has a ground resolution of 1 meter.The noise data is provided by the Swiss noise databas \cite{nokey_f47c8}. The information is given in a raster file. It contains values of noise that were predicted by models and calculations based on noise sources data, traffic, urban fabric and terrain configuration  \cite{nokey_ecc2a}. Noise attenuation due to vegetation has not been taken into account to produce this databaseThe boundary of the municipality is stored in a shapefile as a polygon. It is projected according to the Swiss coordinates system SCR EPSG21781 as all the other georeferenced files used here.Methods  In order to obtain the results needed to do the study, both QGIS and GeoDa softwares were used following the respective "QGIS User Guide"\cite{nokey_13188} and "GeoDa User Guide" \cite{nokey_9ebde}. The data needed for the analysis was obtained from four RVB satellite images encompassing the municipality of Vernier, two raster layers with the sound levels around the municipality of interest, and one vector layer defining the boundaries of Vernier. All this content was imported to QGIS. A Virtual Raster Catalogue was then created to merge the four RVB images. Using the style properties of the catalog, both Red and Blue bands were removed so that only the green band remained visible. This allowed to keep only the reflectance information of the area, which is important to detect the vegetation. After that a 50x50m grid was created with its extent around the limits of the municipality of Vernier. With this grid all information contained in the green band raster and the day and night rasters was gathered in one single attribute table: levels of sound during day and night and of greenness in the area selected. This task was performed using the zonal statistical tool. Then, the main tool of analysis was created, namely selecting only the cells confined within the boundaries of the municipality to extract precisely the information of interest. This operation was made possible with the "Spatial Query" tool available in QGIS.In GeoDa this finer grid was exploited to extract both box maps and scatter plots. Those allow to establish comparisons between the different values of greenness and sound. They are also of interest to verify the spatial correlation between the levels of vegetation and noise, which allow for the verification of the hypothesis.Results The average values of each dataset for the commune of Vernier are shown in table \ref{350700}. There is quite a high range of values for the vegetation indicator, while the sound values for both day and night vary less. The data obtained during nighttime shows lower values as during daytime.
Nyffeler CécileEcole Polytechnique Fédérale de LausanneIntroductionRoad-traffic accidents represent the ninth cause of death worldwide (Murray and Lopez, 1997). It is thus important to get some insight on which societal group might be more exposed to such hazards. A clear association could be made between the probability of injuries by car crashes and the poverty of the person which was hit during the crash (Aguero-Valverde and Jovanis, 2006).  This finding was supported by several other studies.  Siddiqui et al. (2012) were indeed likewise able to affirm that lower median household incomes could be associated with higher road-traffic accidents probability. The aim of this paper is to judge if those findings are applicable to the communal level and to be able to state that the people living in poorer neighborhoods are indeed more vulnerable to car crashes than wealthier regions of the municipality. The commune of Vernier, Switzerland, was selected on the grounds that it is a highly contrasted municipality, and might thus be representative of what may happen at larger scales.Data  Several vector layers and text files were used in order to proceed to this analysis. The vector layers were all provided by the OpenData service of the Canton of Geneva SITG. Geographical point data containing the accident locations and housings addresses were used, as well as polygon layers defining the extent of the municipality zone and the inhabited areas.  The inhabited areas were characterized using a hectometric grid. The demographic data about the allowances were probably taken from the Swiss Federal Office of statistics FSO (not clearly mentioned in the data set).