Meeting Documents

A Prototype for Remote Monitoring of Ocean Heat Content

Presentation title page: A Prototype for Remote Monitoring of Ocean Heat Content
Trossman, D.S. and Tyler, R. (2020)
Presented at: Ocean Sciences Meeting 2020
Presentation title page: A Prototype for Remote Monitoring of Ocean Heat Content


Ocean heat content (OHC) is a key climate variable that needs to be monitored to understand how sea level is changing, yet observing OHC remains a challenge. We propose a monitoring system comprised of satellite altimetry, gravimetry, and magnetometry to remotely sense OHC. It is first examined whether a depth integral of the ocean's electrical conductivity ("conductance"), which may be inferred from in situ methods and/or satellite magnetometers over the global ocean, could help monitor OHC. By examining the output of an ocean state estimate shown to agree well with observations (Estimating the Circulation & Climate of the Ocean, or ECCO), the fundamental limitations of using perfectly known ocean conductance to predict OHC are evaluated. It is found that the ocean's conductance and OHC fields are nonlinearly related but nevertheless highly correlated. A statistical framework tends to predict OHC more accurately than ocean salt content from ocean conductance in regions where conductivity is more sensitive to salinity than temperature. There is a trade-off between the accuracy of the predictions and the length of the time period over which changes in OHC can be detected in the ocean. OHC can be predicted most accurately in many poorly observed regions, such as those covered by sea ice. A statistical framework is further developed in which OHC is predicted using a combination of sea surface heights, bottom pressures, conductance, and/or bathymetry using ECCO. Because the statistical models included in this framework need to be trained and updated, the ECCO output is sampled like historical hydrographic transect observations, trained on these samples over various timescales, and validated on a global scale. It is found that yearly predictions minimize the mean squared error in OHC. We begin to assess whether changes in OHC will be detectable using imperfectly known observations by assessing covariances between the flow fields and the electromagnetic fields associated with the general ocean circulation in ECCO.
View Document (PDF, 144 KB) »