Air pollution source identification from odor complaint data

Abstract

We try to identify the unknown source of air pollution caused by unknown substance. We do it using only complaints on odor submitted by population living in the ∼20 × 30-km area. We test all possible locations of pollution source on the grid with 1-km spacing covering the area. For each virtual location, we perform the dispersion modeling using AERMOD with meteorological data from three nearest stations. We assume a low, ambient temperature area source with unit emission rate. The space-time array of concentrations is correlated with the analogous array of complaints, and the best correlated source position is chosen. We performed extensive experimentation with data processing options to obtain the best spatially focused correlation hotspot. This included filtering the moments of time used for correlation, quantizing the complaint numbers and concentrations, and finding the best time lag between concentrations and complaints. The best lag was 3 h in total or 1 h added to propagation time from source to receptor. The resulting focus is 1–2 km wide, positioned exactly on an active landfill. We estimate this accuracy as very good, better than warranted by the quality of source data. This demonstrates the possibility of obtaining a good focus for the location of an unknown source even from very imperfect data. The technique applied can be useful in many situations where the inventory of air pollution sources is incomplete, especially in developing countries.

from #ORL-AlexandrosSfakianakis via simeraentaxei on Inoreader http://ift.tt/1RB0FSE
via IFTTT

from #Med Blogs by Alexandros G.Sfakianakis via Alexandros G.Sfakianakis on Inoreader http://ift.tt/1QH7f6K
via IFTTT