The quality of the additional profiles is slightly lower than the clear sky profiles but appears useful to support such applications as the retrieval of trace gas concentrations.

The number of available temperature profiles in the IASI L2 product will increase from the current 10-12% on a daily basis to over 50% of the IASI field of views. The additional profiles are retrieved with linear regression for fields of view which are considered 'clear enough' by the regression method itself, but did not pass the strict cloud detection scheme selecting the scenes, which are suitable for clear sky optimal estimation retrieval. The optimal estimation retrieval of these 10-12% clear fields of view remains unchanged in the new version of the processor. The two types of temperature profiles have different characteristics and can be discriminated with the help of the processing and quality flags within the products.

The most detailed information is available in FLG_FINCHC in the IASI combined sounding products in native format, where bit 15 is set if the temperature profile comes from optimal estimation and bit 1 is set if it comes from regression. In the near-real-time disseminated, BUFR encoded product, the flag FLG_QUAL can be used to distinguish the two types of temperature profiles as this flag is only set to a value different than zero, if an optimal estimation retrieval was successfully performed. For the best quality, only the optimal estimation retrieved temperature profiles should be considered.

The yield of water vapour profiles will increase in the same way as the temperature profiles, but in this case all profiles are obtained by regression.

The figures below provide an example of the coverage of atmospheric temperature at 500 hPa (in Kelvin) sensing time 20101117.10:00-20101118.10:00, with the current and upgraded product.

Figure 1 includes only profiles obtained by optimal estimation (FLG_QUAL>0).

Figure 2 shows all available profiles (optimal estimation and regression).

Figure 3 shows all available profiles optimal estimation and regression) but excluding fields of view with FRACTIONAL_CLOUD_COVER > 10%.

Figure 1: Only profiles obtained by optimal estimation (FLG_QUAL>0). (Click to enlarge)

Figure 2: All available profiles (optimal estimation and regression). (Click to enlarge)

Figure 3: All available profiles, excluding fields of view with FRACTIONAL_CLOUD_COVER > 10%. (Click to enlarge)

Back | Product and Service News