A semi-empirical method of PM-10 atmospheric pollution forecast at Santiago de Chile city

Santiago de Chile city, localized in the west side of the Andes Mountain (550 m.s.l.), South America, every year presents high levels of atmospheric pollution by PM-10 in autumn-winter periods. Therefore, forecast model research is a subject of high public interest in order to develop useful and appropriate tools for health impacts prevention on the population. In this work, we present a PM-10 atmospheric pollution semi-empirical forecast method developed froma singular behaviour observed on the relative humidity trends detected in Lo PradoHill meteorological station previous to critical episodes of urban atmospheric pollution by PM-10. This meteorological station is localized at the top of the coastal mountains (1035 m.s.l.), in the west side of the city. From systematic observations between 1998 and 2006 in this particular meteorological station, we have detected how the relative humidity follows a typical descendent behaviour when a type A synoptic weather condition arrives to central Chile and induces a PM-10 pre-emergency episode of urban atmospheric pollution. Our semi-empirical method is based on a sigmoidal behaviour of the relative humidity as a time function, observed in this meteorological station under a type A synoptic weather condition. We have confronted the present forecast method to different pre-emergency critical episodes occurred in the city between 1999 and 2006. Our results are amazing due to these pre-emergency episodes can be predicted at least by 34 hours before to the PM-10 pollution concentration reaches the maximum peak at the city. Contrarily to this specific station, the remainder eight stations localized in the surface of the city are not able to describe this particular phenomenon.