M. Joana Fernandes (Universidade do Porto, Faculdade de Ciências, Portugal); Jérôme Bouffard (RHEA System SA, European Space Agency, France)

Sea level anomaly (SLA) data from two satellite altimeter missions are merged when the aim is the mapping and study of the ocean circulation and mesoscale processes. Generally, one of these missions is one of the so-called reference missions from the consortia NASA/CNES – TOPEX/Poseidon (T/P), Jason-1 (J1) or Jason-2 – and the other is the mission of the European Space Agency that operated simultaneously with the reference one (ERS-1/-2 or ENVISAT). The rationale for this data merging is that these missions provide a complementary sampling of the ocean, by repeating the observation of the Earth’s surface with different temporal and spatial scales. However, some past studies have shown that the combination of these two missions fail in reproducing the sea level particularly in high-variability regions, and some intense signals in the surface geostrophic velocity and eddy kinetic energy fields. Therefore, the simultaneous use of data from three satellite missions – adding one with an intermediate space-time sampling – has been proven to improve the estimation of these fields and the monitoring of their variability, when compared to in situ data (e.g. drifting buoys and tide gauges), while allowing the reduction of sampling errors. While this 3-mission configuration has been made available recently, with the combination of data from Jason-2, CryoSat-2 and SARAL/AltiKa, this can only be achieved during the 2000s by considering the simultaneously use of the Geosat-Follow-On mission (GFO). With a 17-day repeat cycle and an inter-track spacing of 163 km at the equator, GFO SLA data are available in RADS database with up-to-date orbit solutions and geophysical and/or atmospheric corrections, e.g. an updated Water Vapor Radiometer (WVR) correction, allowing a consistent sea level data set to be extracted from GFO, T/P and J1, and ERS-2 and ENVISAT.
In the frame of IDEAS+ CCN5, improved estimations of the across-track geostrophic currents from multi-satellite SLA, from 2002 onwards, over the North Western Mediterranean Sea, where the Northern Current (NC) develops, as well as the analysis of their intra-seasonal to inter-annual variability, are envisaged, targeting at a significant improvement in the quantification of the eddy and meandering structures of the NC. In the scope of this project, the University of Porto (UPorto) is responsible for providing a set of consistent GPD+ WTC for all missions used in the study, including GFO, for which this correction is being tuned for the first time.
The GPD+ wet tropospheric correction (WTC) equals the on-board MWR-derived correction whenever it exists and is valid; otherwise, new WTC estimates are computed from space-time objective analysis using a set of wet path delay observations. In general, this set includes the most recent and valid MWR observations for each altimetry mission, WTC measurements derived from coastal and island GNSS stations, and total column water vapour products from all available scanning imaging MWR (SI-MWR) on-board Remote Sensing missions. In order to improve consistency and long-term stability of the corrections, reducing the uncertainty in the long term sea level variation, the radiometer data sets have been inter-calibrated using the Special Sensor Microwave Imager/Sounder (SSM/IS) on board the Defense Meteorological Satellite Program (DMSP) satellite series as reference, due to their well-known stability and independent calibration.
Within the Sea Level CCI project, the Jason-1 WTC available in the Geophysical Data Records (GDR) Version E was used to inter-calibrate GFO MWR data, as well as other MWR corrections from contemporaneous missions. This correction no longer possesses the anomaly reported in the JMR Enhanced product, which affected the period defined by J1 cycles 228 to 279 that includes the period corresponding to the J1/J2 tandem mission, and induced a bias in the calibration of the J2 MWR-derived wet tropospheric correction, affecting mainly data points in the 0-50 km region from the coast.
The GPD+ GFO WTC spans the full period comprised between 2000 and 2008, and data gaps due to WVR invalidity periods (e.g., period of ~1 year corresponding to cycles 202 to 222) are no longer an issue. The validation of the new GFO WTC is performed by statistical analysis of SLA variance (e.g., by performing along-track and dual-mission crossover analyses). In the absence of GFO WVR measurements, the GPD+ WTC shows an improvement in the description of the small spatial scales, when compared to the European Centre for Medium-range Weather Forecasts (ECMWF) ReAnalysis (ERA) Interim model wet tropospheric delays.