The warming subsurface is a result of the downwelling phase of an eastward propagating oceanic kelvin wave, it’s not propagating into the middle latitudes. The tropical convection coupled to the broadening IPWP as well as the poleward transfer of previously deposited AAM (typical at this stage) is responsible for the +TNH/+PNA.
I went back to 1979, and found that out of 850mb wind, 200mb wind, sea-level pressure, OLR, SSTs, I made a custom index for subsurface temps at -150m.. I found that it was >150% correlation than all those things listed above, for the N. Pacific PNA pattern in now-time. So if the highest OLR anomalies could do is +60dm, the subsurface warm was +90dm.
Because the oceanic kelvin waves that produce said SSH anomalies are generally preceded well in advance by a restructuring of the IPWP, which is coupled with atmospheric circulation/convection.
The IPWP vacillations are even lower frequency than ENSO (multiyear to intradecadal for IPWP vs seasonal to multiyear for ENSO).
I've also found that sea-level height, or sea-level height anomalies has a very high correlation with the H5 pattern +time. Can you explain that one Phil?