On the Automatic Prediction of PM10 with in-situ measurements, satellite AOT retrievals and ancillary data

Daily monitoring of unhealthy particles suspended in the low troposphere is of major concern around the world, and ground- based measuring stations represent a reliable but still inadequate means for a full spatial coverage assessment. Advances in satellite sensors have provided new datasets and though less precise than in-situ observations, they can be combined altogether to enhance the prediction of paniculate matter. In this article we evaluate a methodology for automatic multi-variate estimation of PM₁₀ dry mass concentrations along with a comparison of three different cokriging estimators, which integrate ground measurements of PM₁₀, satellite MODIS-derived retrievals of aerosols optical thickness and further auxiliary data. Results highlight the need for further improvements and studies. The analysis employs the available data in 2007 over the Emilia Romagna region (Padana Plain, Northern Italy), where stagnant meteorological conditions further urge for a comprehensive air quality monitoring. Qualitative PM₁₀ full maps of Emilia Romagna are then automatically yielded on-line in a dynamic CIS environment for multi-temporal analysis on air quality.