SMOS L1 processor prototype: From digital counts to brightness temperatures

L1 processing of SMOS data will transform MIRAS instrument raw outputs into geolocated Brightness Temperatures, providing observation angles and additional parameters for the L2 Processor. The objectives of the L1 processor prototype (L1PP) activities were the definition of the product types and contents, the definition of the data processing models and the implementation and validation of the baseline processing algorithms. The L1 prototype is intended to be used in studies on the feasibility of the algorithms as well as on the advantages/disadvantages of other relevant proposed algorithms before they are taken to the SMOS operational chain. This paper will begin by presenting a very high level description of the L1 processing strategy and functional blocks of the processor (called processing units). A brief definition of the three L1 products types (L1a, L1b and L1c) will be presented, together with data-driven approach selected for the processor. The main focus will be on three aspects of the processing: the practical approach taken for the calibration of MIRAS data, implementing the in-orbit calibration Plan; the reconstruction algorithms chosen and their implementation, converting Calibrated Visibilities into Brightness Temperature Fourier components; and the geolocation algorithm, computing brightness temperatures over an Earth fixed grid along with observation angles, pixel size and accuracy.