Meytar Sorek-Hamer, Itai Kloog, David Broday
This research is developing in several directions: (a) Develop a model to predict concentrations of both PM fractions (PM2.5, PM10) in Israel, using the highly resolved satellite MAIAC-AOD data and different statistical models. We examine model calibration using the following variables: AOD, date, temperature and PBL height. We also study the relationships between hourly PM concentrations (i.e. at satellite overpass) and daily average PM concentrations, which were both used previously in models that associate satellite AOD products and ground PM levels. (b) Build a global database of ground PM data at the overpass hours of Terra (10:30) and Aqua (13:30) during the years 2005-2014. Data were already obtained for Israel, NE and SW USA, and few European countries (France, Italy, NE Spain, central Germany, etc.). (c) Analyse diurnal cycles in the different stations and compare daily average and hourly PM levels as well as to MAIAC/MODIS C06 AOD-based predicted PM. (d) Examine relationships between the PM2.5/PM10 ratio and the goodness of fit of the AOD-PM model. In particular, we are looking for variables/descriptors that can shed light on the large variability in model performance in different areas.