GRUAN PROCESSOR: UPDATES AND CURRENT STATUS WITHIN THE VICIRS PROJECT FOLLOW-UP

S. Gentile^1,2, V. Barlakas³, F. Carminati⁴, D. Cimini^1,2, F. De Angelis³, F. Di Paola¹, S. Larosa¹, F. Madonna⁵, V. Mattioli³, F. Romano¹, M. Viggiano¹, E. Ricciardelli¹

¹National Research Council – Institute of Methodologies for Environmental Analysis, Italy, Tito (PZ), ²Center of Excellence Telesensing of Environment and Model Prediction of Severe Events (CETEMPS), University of L’Aquila, 67100 L’Aquila, Italy, ³European Organisation for the Exploitation of Meteorological Satellites (EUMETSAT), ⁴Met Office UK , ⁵University of Salerno, Department of Physics, Italy

The GRUAN processor (Carminati et al., 2019) is a software developed in the framework of the European Horizon 2020 GAIA-CLIM project (http://www.gaia-clim.eu/page/gruan-processor) by the Met Office. The processor exploits the EUMETSAT NWP-SAF (Numerical Weather Prediction Satellite Application Facility) software RTTOV (Radiative Transfer for TOVS, Saunders et al., 2018) and the Radiance Simulator (Smith, 2017) to simulate brightness temperature (BT) from Numerical Weather Prediction (NWP) model profiles and radiosoundings from the GCOS Reference Upper-Air Network (GRUAN). The original GRUAN processor was based on RTTOV (v11.3) and was designed to work with GRUAN RS92. The software takes in input spatially and temporally collocated GRUAN and NWP profiles to generate two output files containing the BT simulated from the two profiles and the BT uncertainty simulated from GRUAN perturbed profile.

In the framework of the EUMETSAT study VICIRS (Development of VIcarious Calibration for MWI and ICI using RadioSoundings), a new branch of the GRUAN processor has been developed to work with the latest version of RTTOV (v13.2) and to support the new GRUAN RS41. The aim of the VICIRS study is to develop a vicarious calibration tool for the two conical-scanning radiometers Microwave Imager (MWI) and Ice Cloud Imager (ICI) aboard the Metop-SGB (Second Generation) satellites. Validation and vicarious calibration will be performed comparing clear-sky MWI/ICI BT observations with those simulated from radiosonde profiles using the GRUAN processor.

The follow-up activities on VICIRS study have focused on upgrading and expanding the radiosonde datasets, as well as supporting the initial feasibility planning for a potential Cal/Val campaign involving dedicated radiosonde launches during the MWI-ICI commissioning phase. The updated GRUAN processor (currently at v6.3.b.0.3) is part of the follow-up activities on VICIRS. The code has been updated to calculate the BT uncertainty considering the Jacobians from RS profiles and the uncertainty covariance matrices from the radiosounde T, q, P profiles and from the surface parameters T2m, Tskin, Q2m and Psurf. The original GRUAN processor propagated the uncertainties in radiance space via brute force perturbation assuming a complete correlation of the uncertainties at all levels. Furthermore, a rigorous estimate of the specific humidity uncertainty has been introduced from RH, T, P uncertainties. In addition, the tool has been updated to include Jacobians of T, Q, P profiles from NWP and Jacobians of T2m, Tskin, Q2m and Psurf parameters from NWP and the uncertainties associated to the NWP data are calculated from the ECMWF background error covariance matrix available in the NWP SAF 1D-Var package.

The GRUAN processor v6.3.b.0.3, being a part of the updated VICIRS tool, has been tested on the MWI/ICI L1B simulated observations collocated with the corresponding homogenized profiles available from the Radiosouding HARMonization (RHARM) dataset, as well as on a dataset consisting of match-ups between observations from the NASA Global Precipitation Measurement (GPM) Microwave Imager (GMI) and GRUAN/RHARM(v2) profiles.