Global Ocean Heat Content Estimates from Swarm Satellite Magnetometer Observations

Currently, ocean generated electromagnetic signals which are emitted into space are recorded by the low-Earth-orbiting satellite mission Swarm. Observations of oceanic electromagnetic signals contain integrated information about water transports as well as about heat and salinity distribution. An artificial neural network was used to establish a respective nonlinear inversion scheme. We demonstrate how to infer ocean heat content estimates from magnetic signals of oceanic tidal origin. We show that the neural network can closely recover inter-annual and decadal heat content variations from simulated and observed tidal magnetic signals. Swarm based heat content estimates could complement the already existing in-situ measurements of upper ocean heat content and also allow insights into abyssal heat content, where in-situ data are still very scarce.