After some big warmth, we finally saw some Autumn chill arrive over the past couple of days, which signifies the change of seasons. We are now approaching crunch time when it comes to compiling a winter forecast. In our Winter Forecast Webinar, one of the main topics discussed was the evolution of the La Nina and the effects it is already having on the atmospheric pattern. With La Nina conditions already having been established with regards to feedback in the atmospheric circulation, and the fact that ENSO events tend to peak in November, it would seem to be a given that La Nina would be at least somewhat or even perhaps a major driver this coming winter. And to some extent, we believe that is true. However, there is conflicting evidence regarding the future of the La Nina, and part of this will be evidenced by a pattern change to a +PNA and a trough in the East later this month. These factors will be discussed in detail in this article.
When analyzing the current state of ENSO — particularly as it pertains to trying to project its future — we often like to look at the subsurface rather than just the surface, since the subsurface can portend the future and does not have the same variability as the verbatim surface water temperature. It is also indicative of the entire ENSO circulation.
Anyway, we can see that there is clearly a strong and deep cold pool of anomalies below the surface in the eastern half of the Equatorial Pacific. These cooler waters can often be upwelled towards the surface if there is a stronger episode of trade winds, which we will discuss later. Additionally, the bottom panel shows the thermocline. Generally speaking, the waters are warmer in the western half of the Equatorial Pacific — towards Australia — than the eastern half of the Equatorial Pacific — towards Peru. This phenomenon is enhanced during La Nina, where stronger than usual trade winds “push” or “keep” the warmer pool of water in those locations, while cooler water gets upwelled towards the surface in the eastern half of the Equatorial Pacific. The same general idea exists at the subsurface, too, as generally the further in depth you go below the surface, the colder the water gets. This combination of factors results in positively sloped isotherms from west to east in the subsurface. When these positively sloped isotherms are steep, it is indicative of La Nina, and when this slope is much flatter, it is indicative of El Nino. It is clear that the thermocline is very steep, which could even portend a moderate La Nina. Notice the gradient where the thermocline is the steepest and as a result, how close the 17-22C waters are to the surface in the eastern half of the Equatorial Pacific.
Of course, not every strong cold pool gets upwelled to the surface, and usually it ends up being that only a part of this cold pool gets upwelled. But when looking at the trade winds for the next week or two, it seems pretty clear that a big increase in the trade winds will occur, which when combined with the big subsurface cold pool that was discussed above, could easily lead to a big drop in SSTs in the ENSO regions. This occurrence is probably a big reason why so many climate models have a big La Nina influence in their winter forecasts, despite the fact that the actual SSTs do not necessarily represent a La Nina just yet. But should these trade winds indeed upwell at least some of this cold pool, the atmosphere could easily shift towards a moderate La Nina. In fact, some indices that represent the ENSO to atmosphere connection, such as the SOI, show that we are full-blown La Nina. 30-day SOI values +7 or above indicate a La Nina, and right now we are around +9, and this is before this trade wind episode has truly started.
But it gets a bit more complicated than that. It is entirely possible that after this trade wind episode, the entire atmosphere will shift into a strong enough La Nina base state/equilibrium so that anything that happens after the trade wind episode will not have too much of an impact on the La Nina itself. However, it is also entirely possible that a change in the orientation of the tropical convection in shorter timescales, represented by the MJO, may help to halt the La Nina’s momentum. In fact, evidence of this is already starting to show on our forecast models in the medium range. Typically because of the strongest tropical convection being located further west in a La Nina, the corresponding latent-heat-release-induced ridging is near the Aleutians (and the amplitude of that ridging can be determined on other factors such as wave breaking and the Pacific Jet), with a downstream trough near the Western US and thus a downstream ridge in the Southeast. However, tropical convection related to the MJO will quickly be propagating well east of the usual La Nina locations and it should be able to maintain itself pretty strongly. What this helps to do is force a trough near the Aleutians and downstream ridging near the Western US. If this pattern is pronounced enough, it can also help lead to a big change in the trade winds and even reverse them, helping to yield westerly wind bursts that attempt to “slosh” the warmer waters further east.
We can see in the longer range CFS forecasts that after the initial trade wind surge, the trade winds are expected to significantly weaken and perhaps even reserve to westerly winds, coinciding with the pattern changes described above. Sometimes the CFS model overdoes the westerly wind bursts, but the idea at the very least of the trade winds significantly weakening is supported by the MJO propagating eastward. This means that it is likely that at some point later this month and to start November, the La Nina conditions in the atmosphere will be halted or even reverse somewhat.
The changes in the pattern at 500mb continue to move up on all ensemble guidance, with a strong ridge in the West leading to a large trough in the East, which may help start the Lake Effect Snow season. It certainly looks like it will be a chilly end to the month and start November, though the initial core of the colder air may be a bit west of the East Coast. While this is evidence that La Nina may not dominate, there is still a big question as far as what happens later in November. It’s possible that the initial La Nina equilibrium established may mean that the above pattern change is more temporary, and once the stronger MJO signal fades, we go back to a more La Nina base state. However, if the trade winds are not strong enough to truly establish a La Nina equilibrium, then the above pattern may be the beginnings of a longer term pattern with less La Nina emphasis, and instead just shades of it with more variability.
While it is too early to say whether the above pattern is indicative of a longer trend away from a La Nina, or a temporary blip in the midst of a La Nina equilibrium, a look at basic La Nina and -QBO analogs for November can give some hints, though not necessarily answers.
These analogs actually do show a ridge in the West and a trough in the East, despite the fact that these were La Nina years. Perhaps this indicates that it is typical for La Nina and -QBO years to have a tad more of an El Nino flavor in November, and that the longer term can still tilt more towards a La Nina. But this is a small sample of analogs, and 1989’s historic cold skews the picture somewhat. We are still a couple weeks away from getting a better idea regarding La Nina’s longer term effects. If the La Nina ends up trending a bit stronger from the trade winds, the above analogs may need to be changed to remove the negative neutral years and add the moderate La Nina years.
At the very least, however, the above pattern change towards more of an Aleutian Low combined with the -QBO may be able to perturb the stratospheric vortex somewhat, though it is way too early for any true disruptions. But some displacement of this vortex may forebode more blocking episodes later in the Fall and Winter.