Both scatterplots show a quite strong negative correlation between both independant and dependent variables. Namely it looks as ther more vulnerable people are the closer they live to gas stations and their pollution related areas. These result is according to hypothesis. It could be this category of people suffer from direct pollution emanating from both gas stations, as some housings are located within the 75 m rayon of influence mentionned in the paper (DIRE LEQUEL), and the polluted areas associated with them.
This comes as an interesting result showing that deprived people, which usually reside VOIR in areas located closer to road and therefore suffer from the air pollution that comes with it could face even aggravated health risk factors by also living near gas stations.
REGARDER EVENTUELLEMENT EN SELECTIONNANT DONNEES ETC...
Conclusion
The purpose of this short exercise was to use different tools to analyse geodata. It was about analysing correlations between noise and vegetation in the municipality of Vernier. This problematic is perturbed with other environmental factors as settlement form or population density \cite{Margaritis_2016} and \cite{Riedel_2013}. For such a study there are some issues that must be addressed as the importance to have appropriate data and to use the right information. Instead of taking the green band of an orthophoto, one would need to filter it or directly take a map of vegetation from Swisstopo. Moreover a map representing the real noise instead of a model of noise prediction is necessary to assess vegetation effect on noise dampening.
Contributions of the authors
All three author had to elaborate maps and statistical representation of data. Here we took the illustration and results obtained by MR. The report has been prepared by MR for the introduction and the methods. The part of data presentation has been written by RL. The results and discussion have been synthesised by CN. Finally the conclusion and references were done by RL.