WP3.3: Evaluation of satellite products
Quality assessment, in-situ and ground bases observations
vs Satellites
Assessment of the accuracy of aerosol satellite products (FM Breon, A. Vermeulen)
Aerosol products from a number of spaceborne platforms are made available to the scientific community through the ICARE web portal at the University of Lille. The data can be accessed via different web, ftp and internet tools.
Most products can be visualized and downloaded via a browse web interface.
- Aqua/MODIS (no link?) level-2 and level-3 products are downloaded on a daily basis from early 2003 to the present. All level-2 and level-3 aerosol products.
- Parasol level-2 (daily) and level-3 (monthly statistics) are processed in terms of aerosol products on a daily basis.
- Calipso daily measurements and aerosol products on a basis
- SEVIRI measurements are are processed in terms of aerosol
optical depth on a daily basis.
There is a multi-sensor interface to visualize and compare the aerosol products from various
instruments. A number of tools have been developed for data access and statistical analysis to help the user community to
read and analyse the data.
For more info visit the ICARE web site.
Comparison of CALIPSO backscatter lidar against ground observations of EARLINET (L. Mona, G. Papparlardo, C, Italy)
Earlinet measurements in coincidence with Calipso overpass have been extracted. The figure shows an example of Calipso subtrack (dotted line) together with the position of Earlinet stations (red and green dots). At present 25 lidar stations are involved in EARLINET, covering almost the whole European continent from Madrid to Belsk and from Andoya to Athens and other stations will still join this well established network. Out of 25, 15 EARLINET stations are Raman lidar stations and 7 are multiwavelength Raman lidars capable of providing extinction and backscatter profiles at different wavelengths.
Satellite backscatter lidar profile (right plot) versus EARLINET profile (left plot)
The main features of the two profiles from down-looking and up-looking lidars are very similar, with the signature of the PBL and a wide layer between PBL top and 7 km a.s.l. Nevertheless, there is a non-negligible difference in the upper part of the dust layer (6-7 km a.s.l.), at the center of the dust lofted layer, (2.5-4 km), and inside the PBL. Since in the case under investigation the distance between CALIPSO ground-track and CNR-IMAA is 43.1 km, obviously a comparison for the PBL region does not make sense. Differences in the free troposphere instead are to be ascribed mainly to the atmospheric variability and are to be investigated. To do it, wind profiles provided by the ECMWF operational forecast model is used: in the 2.5 – 5 km altitude wind speed and direction are almost constant (about 10.6 m/s, NW). This indicates that the free troposphere aerosol content observed by CALIPSO is transported by the wind over Potenza in about 1 hour. Considering air mass transport, the difference in the attenuated backscatter profiles in the 2.5-5km is significantly reduced, showing that at these altitudes transport mechanisms need to be accounted for.
Comparison of the Parasol products against measurements acquired by PHOTONS (FM Breon?)
We have started a comparison of the Parasol products against measurements acquired by PHOTONS in the framework of GEOmon. An example of the comparison is shown to the right. This figure uses 517 coincidences of Parasol aerosol optical depth measurements versus sunphotometer data. Here, the bias is 0.03 and the RMS is 0.08. Our objective is to provide similar plots and statistics for all satellite products archived at ICARE, and to distinguish various aerosol types and surface covers. This will allow the ranking of the various products in terms of coverage and accuracy.
