Ocean currents play a key role in regulating regional climate by carrying heat away from the tropics and releasing it at higher latitudes. In the Atlantic Ocean, the transport of heat is northward throughout the basin and is governed by the Atlantic Meridional Overturning Circulation (AMOC), a basin-wide system in which warm surface waters move northward and cooler, deeper waters return southward. Because the AMOC affects weather patterns, sea levels and the pace of global warming, understanding how its heat transport varies over time is essential. Yet, directly observing this heat transport is difficult and expensive, so only a few latitudes have dedicated observing arrays.
As an alternative approach, it is also possible to estimate ocean heat transport indirectly using ocean heat content changes (i.e., heat stored in the water) together with surface heat fluxes (i.e., heat exchanged with the atmosphere). However, the measurements needed for this approach are patchy. For example, temperature and salinity profiles from Argo floats are sparse and unevenly spaced, leaving gaps in observing that introduce significant uncertainties. This limits the accuracy of heat transport estimates based on this approach.
To address this, EPOC scientists have developed a new approach that combines Argo data with satellite observations of sea level and ocean mass. Each dataset has its own spatial resolution and uncertainty structure, so the team merged them using a Bayesian hierarchical framework (a statistical modelling approach) that explicitly accounts for these differences. This allows us to leverage the dense sampling of satellites to compensate for the sparseness and irregular distribution of the hydrographic data and produce more accurate estimates of Atlantic heat transport at any latitude.
Applying this method, the team generated observation-based probabilistic estimates of meridional heat transport for the period 2004-2020 at 3-month-mean resolution across 12 latitudinal sections of the Atlantic Ocean between 65°N and 35° S. The resulting estimates closely match direct observations from the RAPID array at 26°N (Fig. 1), capturing both the magnitude and timing of variability, including the pronounced drop in 2010. Beyond 26°N, the method provides a detailed basin-wide view: heat transport is northward across the Atlantic, peaks near 26°N, declines toward higher northern and southern latitudes, and exhibits distinct regional variability. Notably, transport variability is coherent within, but not between, two bands of latitude: 35-16° N and 5° N-35° S (Fig. 2).
Overall, fusing satellite and in-situ data within a rigorous Bayesian framework yields a more physically consistent and accurate reconstruction of Atlantic meridional heat transport than approaches relying on hydrographic data alone. This provides a powerful new tool for monitoring changes in ocean heat redistribution, assessing AMOC variability, and improving our understanding of the ocean’s role in shaping the climate system.
Read the full article:
This new research is published in Ocean Science, available via open access
Calafat, F.M., Vallivattathillam, P. & Frajka-Williams, E. (2025) Estimates of Atlantic meridional heat transport from spatiotemporal fusion of Argo, altimetry, and gravimetry data. Ocean Science 21, https://doi.org/10.5194/os-21-2743-2025s