Major evolution of Sentinel-3 Altimetry products

A new Processing Baseline, PB 2.61, introducing major enhancements to the Sentinel-3 Altimetry products, is planned to be released on 21 January 2020.

Last Updated

24 March 2022

Published on

17 January 2020

This release will increase the baseline collection (BC) of the products from version 003 to 004.

A full altimetry mission reprocessing (Sentinel-3A and 3B) will be available shortly, ensuring that the mission dataset is continuous and homogenous from beginning of mission up to now. The currently reprocessed dataset can already be accessed, see Sentinel-3 Altimetry: Reprocessing of Full Mission.

Figure 1: S3A Sea Surface Height Anomaly for May/2019, generated with the new Baseline collection. Data is filtered by SSHA < |0.75|m

The major changes introduced by the new Processing Baseline, PB 2.61, are the following:

Filtered Ionospheric correction
Correction of a software issue for 20Hz SWH
Update of Mean Sea Surface (MSS) to DTU18
Update of FES2014 Tide Model library to the latest version of the library
Improved Wind Model for very low/high wind speeds
Update of Characterisation of Sentinel-3 SRAL/MWR instruments:
- SRAL Antenna Aperture 3dB and Internal Path delays
- MWR Antenna Patterns

Detailed description of these changes

The Sea Surface Height Anomaly (SSHA) is now corrected for ionosphere delays using filtered ionospheric correction, still derived from the dual-frequency capability of the altimeter. This will reduce the noise in the SSHA. The filtered ionospheric correction is now available as a new variable iono_cor_alt_filtered_01_ku, the 'unfiltered' correction remains in the product.

The SSHA now filters out better sea-ice and land contamination. In trickier locations, users may need to recalculate the SSHA with unfiltered/modeled ionospheric correction.

The default MSS has been updated to the latest version, DTU18. The CNES/CLS15 MSS is still available in the product, but not applied to SSHA.

The tide library of FES2014 model was updated to the latest version, v2.9. The GOT 4.10 tide model is also available in the product, but not applied to SSHA.

For MWR, the antenna patterns used are now based on Sentinel-3A instead of Envisat. This improves the wet tropospheric correction close to the coast.

Figure 2: SSHA comparison between PB 2.45 (BC 003) and PB 2.61 (BC 004). No filtering has been applied to the data, just removed SSHA values larger than |0.5|m. The mean difference in the SSHA between these two baseline collections is of 3 mm, mostly in the zones above 60 degrees of latitude, North and South.

An anomaly in the fitting library was sporadically retrieving very low SWH (0 meters) at 20Hz. The full fitting library was updated and now this is corrected, it has an overall small effect on the other retracked parameters as range and sigma0.

Figure 3: SWH comparison between PB 2.45 (BC 003) and PB 2.61 (BC 004). Data has been filtered for open-ocean, excluded any sea-ice contamination, and limited to the -0.5 meters to 8 meters of SWH. The differences are more significant in low wave height areas, but the mean difference is of 5 cm in wave height.

The retrieved wind speed is now provided also for very low/high wind speeds, before it was truncated between 1.18 m/s and 28 m/s, this is no longer the case. For instance, the altimeter can now detect hurricane class speeds.

Figure 4: Wind Speed comparison between PB 2.45 (BC 003) and PB 2.61 (BC 004). Data has been filtered for open-ocean, excluded any sea-ice contamination, and limited to the 0 to 30 m/s. The mean difference is of 1 cm/s in wind speed.

Better characterisation of the Sentinel-3A and Sentinel-3B SRAL instruments have been implemented in this PB, making the geophysical retrievals of Sentinel-3A and Sentinel-3B more consistent. The Antenna Aperture 3dB and Internal Path delays were updated.

A better estimation of the Thermal Noise was implemented for SAR mode.

Several anomalies have been fixed, the most relevant being:

The PLRM SSHA is now computed using PLRM SSB instead of SAR.
Correction of EGM2008 Geoid Model present in the L2 product.
Satellite platform mispointing information is now used in L2 processing, relevant just for the very early days of the mission.

Full details about the corrected issues are available in the L1 Product Notice and L2 Product Notice.

The updated product specifications can be found here:
Product Data Format Specification - SRAL and MWR Level 1 products
Product Data Format Specification - SRAL and MWR Level 2 Marine products

Note: Due to the change of baseline collection, the filenames will have _004 instead of _003:
S3A_SR_2_WAT____20190812T020725_20190812T020952_20190812T035449_0147_048_074______MAR_O_NR_004.SEN3

Sample files of altimetry Level 2 products generated with the new Processing Baseline are available here:
ftp://ftp.eumetsat.int/pub/OPS/out/test-data/Test-data-for-External-Users/S-3_SRAL-MWR_L2_PB2.61_Test_Data/

They include ~1.5 days of Sentinel-3A and ~1.5 days of Sentinel-3B data in NTC (Non Time Critical) timeliness.

For more information, contact our User Service Helpdesk.