Use of machine learning techniques to study the variability of the meridional volume, heat and salt transport at SAMBA line (34.5S)

• Grant number: 25/03785-6
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: May 01, 2025
• End date: April 30, 2026
• Field of knowledge: Physical Sciences and Mathematics - Oceanography - Physical Oceanography
• Principal Investigator: Edmo José Dias Campos
• Grantee: Arthur Horta Sandonato
• Host Institution: Instituto Oceanográfico (IO). Universidade de São Paulo (USP). São Paulo , SP, Brazil
• Associated research grant: 24/09961-8 - Variability of the Transports by the Deep Waters in the South Atlantic, AP.R

Abstract

The Atlantic Meridional Overturning Circulation (AMOC) plays a crucial role in the redistribution of heat, salt, and mass in the ocean, influencing global climate. One way to monitor its variability is through temperature profiles collected with Expendable Bathythermographs (XBTs) during transatlantic cruises of opportunity. However, the lack of simultaneous salinity measurements limits the estimation of meridional transports associated with the AMOC. In this project, we will explore machine learning techniques to estimate salinity profiles from XBT temperature profiles. Various methods will be tested and evaluated for their ability to reconstruct the salinity profiles. By combining these estimates, we will calculate meridional transports of volume, heat, and salt integrated along the basin, focusing on the AMOC upper-branch. The results will be compared with transport estimates derived from CPIES (Current and Pressure Inverted Echo Sounders) data from the SAMBA (South Atlantic Meridional Overturning Circulation Basin-wide Array) line, located at approximately 34.5°S, as well as from numerical model outputs. This comparison will allow us to assess the representativeness of the machine learning-based estimates. The analysis will include the interannual and interdecadal variability of the meridional transports. This study will contribute to the improvement of observational estimates of the AMOC, providing a more accurate diagnosis of its variability and its role in modulating global climate.