Introduction, purpose and scope

This is the ATBD of the multi parameter retrieval. The method is based on the works of [Pedersen, 1991, Rückert et al., 2023, Scarlat, 2018, Scarlat et al., 2017, Scarlat et al., 2020] and will be described in this document. It is applied to CIMR L1b data which is resampled to a common footprint for all frequencies (we call L1R). It is using all CIMR frequency channels, namely 1.4, 6.9, 10.7, 18.7, and 36.5 GHz and ECMWF Analysis as input. The output of this multi parameter retrieval is in then resampled to the EASE2 grid for the Northern and Southern hemisphere. It is physically consistent and can be used in turn as a priori for the other retrievals as described in the different CIMR DEVALGO ATBDs.

This document describes the algorithm and processing steps of the L2R multi parameter retrieval product. The document is intended for the CIMR users and interested parties. It is not intended to replace a product user guide. The algorithm is implemented in the multi parameter retrieval algorithm which is distributed in the “algorithm” directory of this repository.

To this date, there are no instruments in operation which measure all frequencies at the same time as CIMR. Therefore, for the development of the algorithm, data from AMSR2 and SMOS were used as input. This has a direct impact on the retrieval performance as the collocation of the measurements is not fixed. For surface parameters, this might be less of an issue as the surface is not varying in so short time scales. For atmospheric parameters, however this is a major issue. While the retrieval relies on the idea of a fixed state observation, the current data for the development of the algorithm is coming from a potential divergent state. This is when AMSR2 and SMOS have different overflight times. This may lead to inconsistent results with higher residuals and may even prevent convergence of the retrieval. For this document daily means are used as input, which is only reliable for a stable atmospheric and surface state.