This study explores the use of satellite observations of Aerosol Optical Depth (AOD) to monitor temporal and spatial variations in particulate matter (PM2.5) in the Netherlands. AOD data can only be obtained during clear sky conditions, and are biased towards conditions of stagnant flow (southerly and easterly wind). It was shown that for such conditions, the AOD correlates reasonably well with PM2.5 (R2=0.6). This suggests that temporal variations of AOD can be used as a proxy for temporal variations in PM2.5.
The average PM2.5 concentration at Cabauw was 18 µg/m³. This confirms the relatively high background of PM2.5 in the Netherlands compared to rural areas in Europe. Satellite derived AOD maps from AATSR and MODIS show similar mean values (0.2-0.25) but observed features in the spatial distributions are distinctly different. This illustrates that the interpretation of these AOD maps in terms of PM is hampered because spatially dependent systematic errors in AOD can be of the same order of magnitude as the real spatial variations in AOD across the Netherlands.
Because of the uncertainties in current satellite data of AOD, it is not expected that better PM2.5 maps for the Netherlands can be constructed based on satellite data only. Satellite measurements of AOD can have added-value, particularly regarding the temporal variation of PM, when combined with atmospheric transport models and surface measurements of PM2.5.