Abstract's details
Improvements and limitations of recent mean sea surface models: importance for Sentinel-3 and SWOT
CoAuthors
Event: 2018 Ocean Surface Topography Science Team Meeting
Session: The Geoid, Mean Sea Surfaces and Mean Dynamic Topography
Presentation type: Type Poster
Contribution: PDF file
Abstract:
Previous studies underlined the improved accuracy of the recent Mean Sea Surface (MSS) models. Despite the improvement, residual errors were shown to be significant for wavelengths shorter than ~100km. The MSS errors represent nearly 30% of the SLA variance along the Sentinel-3A tracks and are on the same order of magnitude as the instrumental noise floor of this altimeter (Pujol et al, AGU 2017).
In that context, reducing the MSS model errors at short wavelengths remains necessary to improve the quality of sea level anomalies along new repeat ground tracks. This is particularly important to take full benefit of the noise-reducing SARM technology (and in the future of SWOT).
In this study, we use 17 months of Sentinel-3A measurements to build a new local MSS (also known as mean profile) along the new ground track. We also combine this data with the gridded CNES/CLS model to ensure both large scale accuracy and multi-mission consistency. The result is a so-called hybrid mean profile.
The high precision of SARM data makes it possible to yield improve the smaller scales even with less than 2 years of data. The MSS-related error is reduced by 50% for critical wavelengths ranging from 30 to 100 km.
Furthermore, the so-called LR-RMC prototype processing is shown to be very attractive for future iterations of the HMP because the precision is improved, and the red-colored correlation is removed. Both features are critical to yield a trustworthy reference and to produce better sea level anomaly products at all scales.
In that context, reducing the MSS model errors at short wavelengths remains necessary to improve the quality of sea level anomalies along new repeat ground tracks. This is particularly important to take full benefit of the noise-reducing SARM technology (and in the future of SWOT).
In this study, we use 17 months of Sentinel-3A measurements to build a new local MSS (also known as mean profile) along the new ground track. We also combine this data with the gridded CNES/CLS model to ensure both large scale accuracy and multi-mission consistency. The result is a so-called hybrid mean profile.
The high precision of SARM data makes it possible to yield improve the smaller scales even with less than 2 years of data. The MSS-related error is reduced by 50% for critical wavelengths ranging from 30 to 100 km.
Furthermore, the so-called LR-RMC prototype processing is shown to be very attractive for future iterations of the HMP because the precision is improved, and the red-colored correlation is removed. Both features are critical to yield a trustworthy reference and to produce better sea level anomaly products at all scales.