CALIPSO: 532nm extinction coefficient profile resolution
Posted: Wed Mar 25, 2026 2:07 am America/New_York
I would like to ask about the horizontal resolution of the extinction coefficient profile (532 nm) in level 2 CALIPSO product. I noted that in the algorithm theoretical basis document, it was mentioned that the HERA module is the part which is responsible for extracting extinction coefficient and optical thickness. The output of HERA is referred to as a ‘preliminary’ profile product, which is then passed through a ‘profile averaging engine’ before formulated into Aerosol profile product with a horizontal resolution of 40 km.
I tried to access 532nm extinction in the collection: “CALIPSO Lidar Level 2 Aerosol Profile, V5-00”, where the corresponding longitude and latitude array shows a separation of 5 km between successive points. And the 1st dimension of the array containing 532nm extinction is the same as that of profile lat/lon, and therefore it suggests that 532nm extinction profile is also being stored in 5 km pixels. Since it is expected to have a 40 km resolution, I suppose that the data of 532nm extinction is being repeated 8 times (oversampled) to fit in the 5 km pixels. However, the data does not seem to show 8 consecutive data points storing the same value. Unfortunately, there is not much mentioning of how this “profile averaging engine” works in technical documents or publications.
Can anyone help me with the interpretation? Lots of thanks.
I tried to access 532nm extinction in the collection: “CALIPSO Lidar Level 2 Aerosol Profile, V5-00”, where the corresponding longitude and latitude array shows a separation of 5 km between successive points. And the 1st dimension of the array containing 532nm extinction is the same as that of profile lat/lon, and therefore it suggests that 532nm extinction profile is also being stored in 5 km pixels. Since it is expected to have a 40 km resolution, I suppose that the data of 532nm extinction is being repeated 8 times (oversampled) to fit in the 5 km pixels. However, the data does not seem to show 8 consecutive data points storing the same value. Unfortunately, there is not much mentioning of how this “profile averaging engine” works in technical documents or publications.
Can anyone help me with the interpretation? Lots of thanks.