|
In a series of recent papers we have developed and tested a full physical retrieval scheme for the Infrared Atmospheric Sounder Interferometer or IASI, which is capable of simultaneously retrieving surface temperature, emissivity and atmospheric profiles of temperature, H2O, O3, HDO, CO2, N2O, CO, CH4, SO2, HNO3, NH3, OCS, CF4 atmospheric profiles. Until now, the performance of this scheme has concerned the column amount of gas species. In this paper, we will assess the capability of the methodology to retrieve the vertical profile of CO2. In this respect, the effectiveness of the method mostly relies on the degree of freedom (dof) of the retrieval. In principle, for an unconstrained Least Square approach, dof achieves its larger value and is equal to the number of layers, NL which is used to represent the vertical profile. Using Optimal Estimation we have dof ≤ NL and the actual value is determined by the background covariance matrix. In this study, we use a novel approach, which directly allows us to represent the CO2 profile with orthogonal eigenvectors or vertical modes that explain which fine-scale structures we can resolve with our retrieval. The number of vertical modes we can effectively resolve depends on the IASI information content for CO2. To check the capability of the scheme to retrieve the CO2 profile we have considered a retrieval exercise for a target area close to Mauna Loa and Cape Kumukahi validation stations in the Hawaii region. Five years of IASI data have been collected on such an area and processed according to the above methodology. It will be shown that IASI is capable of retrieving at most 2-3 information pieces or vertical modes for the CO2 vertical profile, which yield an accurate estimation of the CO2 total amount and the correct smooth shape expected for a long-lived species such as CO2 far from intense pollution sources.
|
