> "ENSO cycles are inherently very disturbed, since they range from 2-7 years; periods no one can reliably predict yet."
Umm, all you are describing is the problem statement.
We all know that.
At some point are you planning on doing something other than punting on the problem?
> "Not bulk ENSO noise warned, but pervasive and itself chaotic sampling and sensor noise from ocean waves, clouds, temperature fluctuations, sea-bottom features, coastal irregularities, sensing array gaps, you name it."
ENSO exists only along the equatorial waveguide, creating stable standing waves with nodes that never change their spatial geometry. As with a topological insulator, the behavior appears to be stable against irregularities. Consider that every time that Tropical Instability Waves (TIW) arise during La Nina, the wavetrains *always* have a wavelength of 1200 kilometers.
Very few behaviors on this scale with such a large inertial mass will show statistical noise. It's one big standing wave, but with higher-order wavenumbers (via TIW) that constructively sharpen the peaks in the time-series
The model I have published is a comprehensive spatio-temporal model of standing wave behavior, separating the stable spatial nodes from nonlinear wave-breaking modes which create complex harmonics in the temporal domain.
The bottom-line is that the spatial pattern is easily predicted, and only the temporal pattern is challenging to those not aware of the LTE solution approach.