As most of the current global warming is absorbed by the oceans, upper ocean layers in particular are continously warming, but also deeper regions are affected. Consequently, not only sea level is rising - as the volume of warming water is expanding – but also biodiversity, sensitive ecosystems, and the sea ice extend are affected. Monitoring and predicting temperature changes in the oceans is therefore of great interest and is the subject of intensive research.

In their study, published in the journal Scientific Reports, scientists from GFZ section Earth System Modelling now present a method for monitoring ocean temperature changes from space. They use the fact that ocean tides generate an electromagnetic field that can be measured by modern satellite technology. Changes in this electromagnetic field allow to draw conclusions on water movements, and changes in oceanic heat and salinity budgets.

The scientists "trained" a so-called "feed forward" neural network with measured ocean temperature data from the years 1990 to 2015, and related simulations of tidal electromagnetic fields, measured by satellites. They were able to show that the trained neural network is capable of estimating temperature changes from variations of the magnetic tides.

Only recent satellite missions that investigate Earth’s magnetic field, like the Swarm mission, are able to measure the weak oceanic signals. In combination with latest advances in machine learning techniques, this allowed to train the neural network. In contrast to conventional observation methods, such as buoy measurements, oceanic magnetic signals provide the possibility to capture the ocean heat budget from sea surface to the deep oceans – even in regions where only very few local data exist so far.

Christopher Irrgang, first author of the study: "Observing the ocean heat budget is an essential component for quantifying the Earth’s changing climate. Our method can extend the currently available obseration techniques and thereby help to improve our understanding of climate change". (ak)

Original study: Irrgang, C., Saynisch, J., Thomas, M., 2019. Estimating ocean heat content from tidal magnetic satellite observations. Scientific Reports 9 (7893). DOI: s41598-019-44397-8