> I applied Bayesian data analysis to observations from submarine cables and tide gauges to infer the changes in the Florida Current transport at 27°N during 1909–2018.
These are important techniques for deriving a proxy for missing historical flow data. I have no confidence that the algorithms model a real-world trend but it seems to have promise as one component of a decent future model.
I followed qserasera's link to the Nature paper. The Methods [1] section says:
> I also use Florida Current transport from submarine telephone cables at 27∘N between West Palm Beach and Grand Bahama (Figs. 1b and 2). Using electromagnetic theory, one can estimate changes in the flow from voltages induced across the cable due to the transport of charged particles by the variable current.
> I applied Bayesian data analysis to observations from submarine cables and tide gauges to infer the changes in the Florida Current transport at 27°N during 1909–2018.
These are important techniques for deriving a proxy for missing historical flow data. I have no confidence that the algorithms model a real-world trend but it seems to have promise as one component of a decent future model.