Sorry, I wasn’t trying to suggest the KW itself refracts back westward upon reaching the EPAC. That wouldn’t make physical sense without some kind of barrier to restrict meridional evacuation. So the bathtub analogy is technically incorrect.
But when the OKW reaches the EPAC and diffracts into the two off-equator oceanic rossby waves, it aids equatorial upwelling behind the tail as water evacuates poleward, as I’m sure you know.
Hence the importance of a healthy follow-up OKW.
I see where you're coming from but the latter isn't true either. When a Kelvin wave reflects onto a boundary it returns as a rossby wave of the same sign. Meaning that these downwelling Rossby Waves created by reflection of the KW at the eastern boundary also suppress the thermocline. What actually enhances near-equatorial upwelling during El Nino events is the westward flowing surface currents which are in the same sense as earth's rotation generating cyclonic wind stress curl anomalies that augment Sverdrup transport, in effect increasing poleward export of water from the equator. Those westward propagating off-equatorial Rossby Waves actually do contribute to some extent to retaining heat in the Eq Pacific because their westward propagation contributes to anticyclonic wind stress curl that offsets the Equatorial discharge induced by the equatorial westerly current. In any case, I don't think we're having much issue generating more downwelling oceanic Kelvin Waves because another one has entered the fray as aforementioned, it will initially be met with destructive interference but when sub seasonal forcing realigns w/ the base state that's becoming more NINO like late in July and into August, this downwelling wave will likely intensify before reaching South America.
And does the +PMM actually aid the development of the niño, or is it largely a result of the low frequency state of circulation that favors warm pool discharges?
I’m more convinced of the latter. Too much off-equator convection can actually interfere with El Niño development, in the canonical sense. Some of the most prolific +PMMs like 1980 and 2014 were very anemic in the ENSO department.
I've discussed this at length numerous times, but the answer is yes literature shows that the PMM in both NWP simulations and observations contributes to the triggering of about 2/3rds of all El Nino events, is vital to the seasonal phase locking of ENSO, and when coupled with the negative phase of the SPMM increases the probability of central pacific/modoki El Nino. This year's +PMM regime is bigger than any that has ever been observed during the satellite era and the subsurface warming is way stronger than either 1980 & 2014 ever were, being more on par w/ moderate El Ninos like 2009 & 2002, which suggests that the former years may not be an adequate comparison here. Not every +PMM event will trigger an El Nino, however the probability of an El Nino is likely significantly higher when its coupled w/ a -AMO. 1980 and 2014 actually featured near or above normal SSTAs in the tropical Atlantic by this point in the summer, this year's SST anomalies have been rivaling the coldest years of the last -AMO regime in the 1980s & 1990s. Why does a cold tropical Atlantic matter for ENSO?
A colder Atlantic is certainly more favorable to El Nino for several reasons, cooler MDR SSTs means the easterly trades are stronger which favors air/mass more readily piling into South America, where a major anomalous upward branch of the Hadley Cell is juxtaposed during El Ninos, essentially meaning that a cold AMO superimposes itself to some extent onto the larger-scale NINO circulation. A colder Atlantic also means there's one less ocean basin that's competing for upward motion on the global scale, affording the Pacific a greater chance to steal the show per say. In addition, the cold phase of the AMO pulls the ITCZ in the Pacific equatorward, again favoring El Nino (Sung et al (2015)). A seiching mechanism related to the Atlantic Meridional Overturning Circulation has also been proposed but it's much more elusive.
Here's a few excerpts from literature I've sifted through in the past few months that provide verification of this cold AMO-NINO relationship.
"The results show that the tropical Atlantic warming associated with the positive AMO phase leads to a
westward displacement of the Pacific Walker circulation and a cooling of the tropical Pacific Ocean, thereby
inducing anomalous descending motion over the central tropical Pacific. The descending motion then
excites a stationary Rossby wave pattern that extends northward to produce a nearly barotropic anticyclone
over the North Pacific."
"The long-lasting cold surface conditions of North Atlantic, i.e., the negative phase of Atlantic Multidecadal Oscillation (AMO), can intensify the El Niño–Southern Oscillation through the enhanced air-sea coupling under the increased central-to-eastern tropical Paciﬁc mean sea surface temperature. However, the impact of warmer mean sea surface temperature (SST) is more efficient in the intensifying El Niño than La Niña, because of the nature of the exponential growth of atmospheric convection to SST change. Moreover, the farther eastward shift of the atmospheric convection during the negative AMO leads to the stronger El Niño due to the longer delayed negative feedback by oceanic waves. Therefore, the AMO mainly inﬂuences El Niño intensity rather than La Niña intensity."
"The SST anomalies in the El Niño years during AMO periods were stronger;"
"El Niño events were more signiﬁcantly enhanced than La Niña events during AMO periods. The asymmetric development of the ENSO during the different AMO phases is related to thewarmer and wetter basic state of the central-to-eastern tropical Paciﬁc, as shown in both the observations and the CM2.1 model output. Owing to the nonlinear properties in the SST-convection relationship, the warmer mean state of the SST in the central-to-eastern tropical Paciﬁc enhances the sensitivity ofatmospheric convection to the same SST anomaly"
Also given this year's +PMM is the strongest and one of the longest lived events that's ever been observed, this only gives me more confidence that this +PMM will trigger an El Nino this year.
You would be right that the current +PMM and -SPMM configuration does limit the maximum intensity of El Ninos especially in the canonical sense because the cold SE Pacific is a reflection of cumulative southern hemisphere mid-latitude RW activity and the trade wind anomalies in the subtropical SE Pacific propagate equatorward via WES feedback (described in Min (2015)) and affect the eastern portion of the equatorial basin only. However, the presence of this -SPMM doesn't limit the total probability of El Nino because it says nothing about the central Pacific and the dateline region which is in a favorable regime for warming right now. The base of the +PMM connects to the equator near the dateline, favoring westerly wind anomalies near the edge of the warm pool, and on its own, favors central Pacific El Nino. Like its counter part in the southern hemisphere, these westerly wind anomalies in a +PMM propagate towards the equator ultimately favoring El Nino, and the net ekman transport induced by the west - southwesterly wind anomalies generated in a +PMM transport water towards the eastern boundary region and the equator closer to the dateline, again favoring El Nino. The PMM in a general sense acts as effective red background noise "energizing" low frequency variance in the tropics, thereby exciting ENSO (see Lorenzo et al (2015) linked below).
All of the above evidence, literature, and current observations suggest that a weak-moderate central Pacific El Nino is most likely in 2018-19. Furthermore, the observational record in addition to the expected intensity of this El Nino (if one were to form) also argue that the probability of another El Nino in 2019-20 is higher than long-term average (~30-33%).
To close this post I'll reiterate that according to the ENS ONI, since 1865, 50-60% of first year-weak moderate El Ninos are followed immediately by another El Nino. In about 85% of those cases where another El Nino occurred in the following year, the 2nd El Nino was stronger than the first one. Keep this in mind going forward.
There's lots of literature on this topic, I'm only showing a few pieces I've read through.
"The majority of ENSO events in both nature and the coupled model are preceded by MM events."