The Atlantic meridional overturning circulation (AMOC) is a system of currents that regulates the global climate system through its associated heat, freshwater, and carbon transport.

Variations in the AMOC are increasingly linked to climate impacts and coupled climate models suggest that it will weaken. However, programmes that directly monitor the AMOC have relatively short durations (two decades at most), and so it is unclear whether the AMOC has declined during the observational record. Direct observations are also too sparse to establish the spatial structure of the AMOC across the Atlantic. Therefore, it has become a scientific priority to determine whether AMOC strength and structure can be inferred through alternative ‘indirect’ means. To address this challenge, MEZCAL will bring together techniques combining satellite and autonomous ocean observations, machine learning approaches that build empirical relationships between the AMOC and observations, and adjoint-methods that connect atmospheric forcing with circulation in ocean models to deliver a new AMOC framework from indirect observations, which will:

  1. Extend the spatial coverage of the AMOC across the North Atlantic.
  2. Reconstruct the AMOC backwards in time from historical datasets and models.
  3. Make recommendations to ensure a sustainable future monitoring system.

By assembling the observational oceanography, artificial intelligence, and numerical modelling communities we will establish best practices for determining past, present, and future AMOC changes from indirect observations and create a step change in our understanding of the spatiotemporal structure of the AMOC, thus shaping the future of AMOC observing.

Aim

The aim of MEZCAL (Methods for Extending the horiZontal Coverage of the Amoc Latitudinally and retrospectively) is to address the challenges of the Atlantic Meridional Overturning Circulation (AMOC) current observing system and make a step change in our understanding of the AMOC’s spatial structure across the North Atlantic, as well as its long-term temporal evolution.

MEZCAL will deliver the following objectives:

  1. Design and implement a spatiotemporally complete AMOC framework from indirect observations using proven reconstruction techniques.
  2. Extend AMOC records backwards in time using machine learning and adjoint modelling methods.
  3. Make recommendations for an AMOC monitoring framework and enhanced observations to ensure a sustainable future AMOC monitoring system.
MEZCAL