Here, we test mutual-information (MI) Optimal flight paths for February 17, which maximizes the MI between the surface velocity measured on February 17 and either surface σT alone or surface σT + surface velocity on February 20. the flight paths that are 700 nmi long. In each case, the flight is allowed to travel in any of Nθ=10 different directions in Ns=11 segments between takeoff and landing. We also vary the number of points we sample along the flight paths.
When we target σT alone we optimize flights for information on the upper-layer eddies of the western wall. When we target σT and velocity we are interested in both the upper-layer eddies of the western wall and the full LC flow.
