PM 10 Time Series Analysis Through Geostatistical Techniques

Particulate matter (PM) is an air pollutant comes from vehicular traffic, industrial activities and street dust, or from the atmosphere, by transformation of the gaseous emissions. In recent years the interest in the health effects of this pollutant have increased, since high concentration levels in urban area have been measured. Several studies suggest an association between fine particulate air pollution and the increase of the mortality rate. In particular, PM up to 10 micrometers in size (PM10) could cause negative health effects such as respiratory illness or cardiovascular problems. Hence, the analysis of temporal evolution of this pollutant could be useful in decision-making process for environmental policy. Typically, in time series analysis, the Box-Jenkins methodology is widely applied and the autocorrelation function (ACF) is used as a standard exploratory tool to identify the model structure . In this context, the use of geostatistical techniques could also be convenient, nevertheless these techniques are usually applied to analyze, through the variogram, spatial relationships among sample data measured at some locations in a domain and to predict the corresponding spatial phenomena.