Validation of PM MAPPER Aerosol Optical Thickness Retrievals at 1x1Km2 of Spatial Resolution

The polar-orbiting MODerate resolution Imaging Spectrora- diometer (MODIS) on-board Terra and Aqua satellites is a key instru- ment for the daily monitoring of global aerosol properties over a large spectral range. Its aerosol retrieval algorithm is set to a size of 10×10 km2 of spatial resolution,...

Full description

Bibliographic Details
Main Authors: CAMPALANI, Piero, MAZZINI, Gianluca, T. N. T. Nguyen, S. Mantovani, M. Bottoni
Other Authors: Campalani, Piero, T. N. T., Nguyen, S., Mantovani, M., Bottoni, Mazzini, Gianluca
Format: Conference Object
Language:English
Published: IEEE 2011
Subjects:
Online Access:http://hdl.handle.net/11392/1515532
Description
Summary:The polar-orbiting MODerate resolution Imaging Spectrora- diometer (MODIS) on-board Terra and Aqua satellites is a key instru- ment for the daily monitoring of global aerosol properties over a large spectral range. Its aerosol retrieval algorithm is set to a size of 10×10 km2 of spatial resolution, and hence may not be adequate for detailed analy- sis at local scale. PM MAPPER is a software system capable of handling the multispectral data acquired by the MODIS sensors. It produces maps of Aerosol Optical Thickness (AOT) at increased spatial resolution up to 1×1 km2 , which are then available online in a GIS environment. This article shows the validation results of these products, obtained by com- parison with AOT measurements of several ground-based radiometers of the AErosol RObotic NETwork (AERONET) over Europe, for a period of 3 years (2007-2009). They show a good correlation between satellite prod- ucts and ground measurements. Different sizes of the spatio-temporal window to associate satellite and ground observations have been tested, and trends have been searched by tuning the comparisons for different years, seasons and land-cover classes. An optimal spatio-temporal win- dow for this kind of validation is also suggested. This could be used for other purposes as well, e.g. to perform improvements of AOT retrieval algorithm with machine learning techniques.