{"title":"PM2.5 FORECASTING IN THE MOST POLLUTED CITY IN SOUTH AMERICA","authors":"P. Perez, C. Menares, Camilo Ramirez","doi":"10.2495/AIR180181","DOIUrl":null,"url":null,"abstract":"According to a recent study of WHO, Coyhaique, a small city in the south of Chile is the most polluted city in South America. With 70,000 habitants, the reasons for the high PM2.5 concentrations in the city area during fall and winter are: topographic situation, stable atmospheric conditions and intense use of wood stoves for heating. During 2016, the 24h moving average exceeded 170 micrograms per cubic meter for 63 days. A neural network model that uses previous values of PM2.5, meteorological information and previous concentrations of NO2 and CO as input, which is trained with 2014 and 2015 data, is able to forecast 91% of these exceedances. This forecasting is very useful in order to alert the population and to motivate the authorities to take actions to control the emissions.","PeriodicalId":165416,"journal":{"name":"Air Pollution XXVI","volume":"68 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2018-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Air Pollution XXVI","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.2495/AIR180181","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 2
Abstract
According to a recent study of WHO, Coyhaique, a small city in the south of Chile is the most polluted city in South America. With 70,000 habitants, the reasons for the high PM2.5 concentrations in the city area during fall and winter are: topographic situation, stable atmospheric conditions and intense use of wood stoves for heating. During 2016, the 24h moving average exceeded 170 micrograms per cubic meter for 63 days. A neural network model that uses previous values of PM2.5, meteorological information and previous concentrations of NO2 and CO as input, which is trained with 2014 and 2015 data, is able to forecast 91% of these exceedances. This forecasting is very useful in order to alert the population and to motivate the authorities to take actions to control the emissions.