PaulP:"It's called a topological insulator because the flow is restricted to a surface, boundary, or interface instead of the bulk."
However, the Equatorial Kelvin Wave travels within the bulk of the Equatorial Belt between North and South Boundaries, which themselves could be accounted as insulating here. Metamaterial scientists tend to omit Galilean Invariance. Its great that geophysicists are applying concepts from Topological Metamaterials, but its a very early stage of cross-analysis. The conventional equations and existing data are also not fully definitive.
Of course ENSO and MJO interact quasi-harmonically, as any two interacting quasi-periodic systems would. We need not be "sad" when hardly anyone belabors the obvious.
Like weather prediction has slowly advanced, ENSO predictions will slowly improve, until predicting a few cycles ahead should be possible, but not much better. Lunisolar Tides must mostly play a marginal role compared to the dramatic daily and seasonal oscillations of Earth rotation and inclination, interacting with solar radiation, and the complex crustal geology of oceans, coasts, and continents.
Lunisolar noise in ENSO-QBO data needs to be accounted for in your Model. It can't be all or even mostly pure coherent forcing, but also some major degree of chaotic interference to identify and quantify. Your science will be all the better not to "punt" (arm-wave) the chaos aspect, but take it on with the rigor now possible. Once its accepted that Lunisolar inputs do have their place in the mix, accurate quantification can follow, even if strong prediction remains elusive.
A basic quantification is to compare the High Q of Lunisolar statistics to the Low Q of them in the thicket ENSO-QBO data. We could call this the non-dimensional Pukite Number.