Results of Towards consistent regional sea level budgets - OSTST_CORB Project

Luciana Fenoglio (University of Bonn, Germany)

CoAuthors

Bernd Uebbing (University of Bonn, Germany); Sophie Stolzenberger (University of Bonn, Germany); Roelof RIetbroek (University of Bonn, Germany); Jürgen Kusche (University of Bonn, Germany)

Event: 2020 Ocean Surface Topography Science Team Meeting (virtual)

Session: Salient results from the 2017-2020 OSTST PIs

Presentation type: Type Forum

The OSTST_CORB (Towards COnsistent Regional sea level Budgets) Project of the University of Bonn has addressed the use of low resolution model (LRM) and Synthetic Aperture Radar (SAR) radar altimetry mission data for studying coastal sea level change and establishing regional sea level budgets.

Salient results have been obtained improving the processing of the SAR mode altimeter data. RDSAR and unfocused SAR have been largely studied in the past four years. In coastal zone and in-land water, accuracy and precision have been significantly improved with the enhanced unfocused SAR processing SAMOSA+ and SAMOSA++ which measure up to 3 km from coast. For LRM the spatio-temporal altimetry retracker STAR allows to derive significantly improved results in coastal regions. Application to RDSAR data and comparison with SAMOSA+ retracked SAR mode data revealed a similar level of quality. SAR waveform and noise floor are different from those in LRM and the surface water slope is a new observable. The TUDaBo processor for RDSAR and open ocean SAR altimetry is publically available in GPOD. Open point is the further improvement of the SAR data precision, which depends on SWH and wave period (T02), by inclusion of the vertical motion of wave particles in SAR altimeter processing. First promising results are actually further investigated in an ESA project (HYDROCOASTAL CCN1) for implementation in TUDaBo.

The synergy with in-situ data and models of these new SAR data is largely superior to the previous LRM data. The monthly coastal variability from the SAR-SAM+ product agrees most favourably with high-resolution models (e.g. NEMO-WAM) opening the way to the combination and assimilation of altimetry and model data to reconstruct the 2D variability in the coastal ocean. The regional monthly coastal variability of model and altimeter data shows a stdd of 3.9 and correlation 0.90 for Sentinel-3A, which is about twice the maximum departure between the coastal variability of altimeter data products (SAR-SAM+ and RDSAR-TALES CryoSat-2, stdd 2.3 and correlation 0.96).

A validation of the newly retracked data was performed with in-situ, regional high resolution models and other mission altimeter data in the North-Eastern Atlantic. Finally, the sea level trends of the merged LRM and SAR altimetry time-series are consistent with the LRM trends over the complete altimeter interval 1993-2019.

The global fingerprint inversion has been significantly improved over the last few years. Individual components are estimated by fitting predefined patterns of steric or mass related sea level change to a combination of GRACE and radar altimetry data in order to estimate scaling factors for each component on a monthly basis. By connecting the estimated scaling factors with the known patterns, it is possible to reconstruct time series of global and regional total sea level change, as well as for individual mass and steric contributors. Today we are able to derive a global sea level budget which agrees with individual components independently estimated. It enables also the closure of the global sea level budget within about 0.1 mm/yr. The residual sea level signals mainly consist of so-far unmodelled eddy variations and changes in the global current system. Within the framework of the German research project "GROCE", we studied the impact of Greenland freshwater on the North Atlantic by performing ocean model simulations. For the North Eastern part of Greenland, the steric heights show a positive trend from 2009 onwards, which is in agreement with the output of the fingerprint inversion method.
 
Luciana Fenoglio
University of Bonn
Germany
fenoglio@geod.uni-bonn.de