Recent advances in the random projection approach applied to the inversion of the full IASI (Infrared Atmospheric Sounder Interferometer) spectrum are exemplified with applications to the retrieval of greenhouse gases and pollutants. The case of simultaneous retrieval of CO, CO2, CH4, HNO3, NH3, OCS and N2O is exemplified in this study. Random projections provide a) an unified and coherent treatment of systematic and random errors; b) a compression tool, which can reduce the dimensionality of the data space; c) a noise model which is truly Gaussian therefore, making it possible to apply rigorously Optimal Estimation and derive the correct retrieval error; d) a simplified treatment of the inverse algebra to get the final solution. Many of the methodological aspects above have been shown in a recent paper by the authors (Serio et al. 2016, doi:10.1364/AO.55.006576) in dealing with the retrieval of surface parameters and atmospheric temperature, water vapour and ozone profiles. The present analysis is aiming at showing that we can fully exploit the compression capability of random projections to develop an inverse algorithm capable of dealing with the whole IASI spectrum, therefore with minimal loss of information content. The results is a final retrieval with improved precision and horizontal spatial and temporal resolution. Retrieved parameters and species include, surface temperature and emissivity (spectrum), Temperature, H2O, O3, HDO, CO2, N2O, CO, CH4, SO2, HNO3, NH3, OCS, CF4 atmospheric profiles. The effective vertical resolution of these profiles is depending on the degrees of freedom.
A retrieval case study has been set up, which includes thousands of observed spectra for a two-yearperiod(January2014toDecember2015)over sea surface in the Pacific Ocean close to the Mauna Loa (Hawaii) validation station. Results obtained until now (see e.g. Liuzzi et al 2016, doi:10.1016/j.jqsrt.2016.05.022) show that retrieval of gas species is obtained with unprecedented precision and spatial resolution, which open the way to study and understand regional patterns of greenhouse gases due to anthropogenic activities and/or natural events. In the present study, IASI retrieval for CO2, CH4 and N2O are compared to operational products from different satellites and instruments. These include AIRS (AtmosphericInfraredRadiometerSounder), GOSAT (Greenhouse Gases Observing Satellite), OCO-2 (Orbiting Carbon Observatory-2). The comparison shows that our retrieval methodology is by far superior to those of existing methods insisting on instruments, which covers the thermal infrared such as IASI, AIRS and GOSAT. Comparison with OCO-2 shows comparable results, but our methodology uses much less a-priori information than OCO-2.