An enhanced retrieval of the wet tropospheric correction for Sentinel-3 using dynamic inputs from ERA5

• Type: Article in International Scientific Journal
• Date: 2022

Authors

Vieira, T (Author) Other; Fernandes, MJ (Author) FCUP; Clara Lazaro (Author) FCUP

Journal

Title: Journal of Geodesy

Vol. 96

ISSN: 0949-7714

Publisher: Springer Nature

Indexing

ISI Web of Knowledge - 6 Citations

Scopus

Other information

Authenticus ID: P-00W-ENS

DOI: 10.1007/s00190-022-01622-z

Abstract (EN): Sentinel-3 (S3) satellites are equipped with microwave radiometers that perform brightness temperature (TB) measurements at 23.8 and 36.5 GHz to determine the wet tropospheric correction (WTC). The analysis of the two MWR-derived WTC present in S3 products, retrieved from three- and five-input neural network (NN) algorithms, suggest the need for their improvement. Focusing on the inputs and physical component, this paper aims at improving the WTC retrieval for open ocean, considering a suitable learning for S3 and a better accounting for the surface contribution to the WTC retrieval. Adopting a purely empirical approach, the learning database has been built using 1 year (2017) of valid S3A measurements, ERA5-derived WTC, and dynamic sea surface temperature (SST) from ERA5. The proposed approach is a similar NN with four inputs: TB at 23.8 and 36.5 GHz, altimeter backscattering coefficient and SST. Results show that the use of a dynamic SST, instead of static tables as currently adopted in S3, makes the fifth input (vertical temperature decrease) redundant. Comparisons with reference and independent WTC sources show that the WTC derived from this algorithm, when compared with those available in the S3 products, leads to a decrease in the RMS values of WTC differences, with respect to these independent WTC, by about 1 mm globally, that locally can reach almost 1 cm. This study proposes a new approach for the WTC of Sentinel-3 by considering a suitable set of dynamic inputs that better characterize the atmosphere, which is a significant enhancement over the current algorithms.

Language: English

Type (Professor's evaluation): Scientific

No. of pages: 19