Tuesday, May 28, 2019

ENSO Analysis and Outlook: Fall into Winter 2019

This post will analyze the current state of the El Nino-Southern Oscillation (ENSO) phenomenon as well as forecasts for its nature through fall 2019 and into the early winter months of 2019. It is critical to note right off the bat that this is *not* an ENSO forecast for the winter of 2019-2020, but the forecast period we will be going over will tread into November and December. Click on any image to enlarge it.

We will first define what the ENSO phenomenon is, and why we care about it.

Graphical depiction of the four different ENSO monitoring areas.
Source: Climate Prediction Center
The ENSO phenomenon, in a nutshell, is a primary driver of seasonal (and, through other shorter-term oscillations, weekly or even daily) weather patterns by way of sea surface temperature (SST) anomalies in the waters across the Equatorial Pacific. When these sea surface temperatures are above normal, we call it an 'El Nino' event. When these anomalies are below-normal, we call it a 'La Nina' event. While we monitor the entire Equatorial Pacific to analyze the ENSO phenomenon, there are four primary "zones" through which to observe. They are:


  • Nino 1+2. This is a small slice of the Pacific located between the Equator and the 10º South latitude line, extending from the far western tip of Peru to the 90º West longitude line.
  • Nino 3. This is a larger slice of the Equatorial Pacific which spans from 5º North to 5º South latitude lines, and from 90º West to 150º West longitude lines.
  • Nino 4. This is also a larger slice, and also extends between 5ºN and 5ºS on the latitude markers. For Nino 4, however, the space is spread by longitude from 150º West to about 160º East, crossing the dateline in the process.
  • Nino 3.4. This is the critical area to watch, and is typically viewed as the primary space with which to assess the state of the ENSO phenomenon. Spatially, it extends from 5ºN-5ºS latitudinally, and 120º West to 165º West longitudinally. 

Why do we break this space up into four different pieces rather than just average out the sea surface temperature anomalies and call it a day? A number of scientists with far more knowledge and research than I have come to determine that there can be more than one type of El Nino - where typically El Nino's bring warmer than normal waters to the eastern Pacific, an "El Nino Modoki" event brings warm waters to the western Pacific, and cooler waters to the eastern Pacific. This is not a trivial difference, but for our purposes here, we won't dive into that topic. For now, the key is understanding there are four different regions in which we monitor the ENSO phenomenon, with the Nino 3.4 region broadly being of most importance.

Let's view sea surface temperature anomalies over the Pacific now, with a focus on those regions that were just outlined.

Observed sea surface temperature anomalies on May 20th, centered over the Equatorial Pacific.
Source: NOAA
As of May 20th, sea surface temperature anomalies along the Equatorial Pacific were, on the whole, above normal. A solid swath of above-normal anomalies extended across Oceania to about the 130º West longitude line. From there to about the 110º West longitude line, however, SST anomalies were seen closer to zero, with a very small area of slightly below-normal temperature anomalies. We'll get in to why that's there in a little bit. By the 100º West line of longitude, however, solidly above-normal sea surface temperature anomalies return to the western tip of Peru.

Based on what was discussed earlier, this seems to point towards the presence of an El Nino event (the positive state of the ENSO phenomenon). To confirm or reject this, the Earth System Research Laboratory (ESRL) has composed a Multivariate ENSO Index (MEI) to quantify the state of the ENSO phenomenon, as shown below.

MEI, showing positive (El Nino) and negative (La Nina) changes to the ENSO phenomenon.
Source: ESRL
The MEI aims to determine if there is an El Nino in place (via positive index values), if there is a La Nina in place (via negative index values), or if there is a neutral-ENSO state (via index values equal to zero). You can identify a few extreme events on here, such as the strong El Nino in 1997 and the substantial La Nina event in 2010. Looking to the last several data points, it appears that there has been a general trend towards an El Nino event, but nothing particularly steady is in place.

I say nothing steady is in place because, as a general rule of thumb, an El Nino (La Nina) is present if sea surface temperature anomalies are above-normal by at least +0.5 degrees Celsius (below-normal by at least -0.5 degrees Celsius). If SST anomalies are positive but just barely so, it's technically a neutral-ENSO state, but clearly there's a better shot at an El Nino forming down the line. The same logic applies to SST anomalies that are negative but just barely so, with respect to a La Nina.

As of May 9th, the Climate Prediction Center continued its El Nino Advisory (click here for full briefing), which indicates that an El Nino event is ongoing. Indeed, the agency assigns a 70% probability of an El Nino continuing through the summer months, with a 55-60% chance of the El Nino persisting through the fall months. These probabilities may seem rather low given that the El Nino is actually occurring already, but as the rest of this post will show, it isn't that cut-and-dry with the ENSO phenomenon.

SST anomalies for each of the ENSO regions.
Source: CPC
Since we've already learned about the four different ENSO regions, it's time to apply that to observed data. Shown above are four panels of SST anomaly data over the past twelve months, with each panel corresponding to a different ENSO region. The top panel shows SST anomalies for the Nino 4 region; the second-from-top panel for the Nino 3.4 region; the second-from-bottom panel for the Nino 3 region; and the bottom panel for the Nino 1+2 region. In the aggregate, the data confirm that we are in an El Nino event, at least judging by sea surface temperature anomalies, with anomalies exceeding the +0.5º Celsius threshold in the Nino 4 and Nino 3.4 regions, with anomalies in the Nino 3 region right around that threshold. In contrast, the Nino 1+2 region has reversed to marginally-negative SST anomalies. This likely owes to the weak but broad area of slightly below-normal anomalies immediately southwest of the westernmost tip of Peru back on that observed SST anomaly graphic. Since it isn't a significant deviation, I don't see any major reason to raise concern over the difference in Nino 1+2 with the other three regions.

We are also able to look at sea temperature anomalies with varying depth along the Equator in the Pacific, as shown below.

Equatorial temperature anomalies (top) and observed nominal temperatures (bottom) as of May 18th.
Source: CPC
Viewing sea temperature anomalies along the Equator as a function of depth can prove massively beneficial to forecasting abilities, as it can enable the forecaster to identify an area of well-below-normal anomalies right below the surface that is eating away at above-normal SSTs on the surface. In this hypothetical, someone only viewing the surface map would think there's a solid El Nino in place, but the forecaster with the depth map as well can see that the El Nino is actually about to dissolve.

Turning back to actual data, the depth chart above shows a broad expanse of above-normal water temperatures extending from 140º East to about 120º West longitudinally, with the positive anomalies reaching a depth of almost 150 meters in the western portion of this swath. However, when reaching the 120º-100º West longitude area, well-below-normal temperature anomalies appear, and seem to be threatening the warmer anomalies located at the surface. We can view an animation of this depth map to see how these two opposing bodies of water have been interacting lately.

Sea temperature anomalies by depth, animated. Refresh the page if the animation stops looping.
Source: CPC
Indeed, when viewing the animation we are able to grasp the rather-dire situation the ongoing El Nino seems to be in. After covering almost the entire Equatorial Pacific with well-above-normal temperature anomalies from the surface to almost 150 meters down in March, cooler than normal waters have gradually grown between 150 meters and 250 meters below the surface since then and have materially weakened the formerly-stout positive temperature anomalies below the surface. Even more concerningly for the El Nino, the negative temperature anomalies appear to be growing and deepening east of the 120º West line of longitude, suggesting that the positive SST anomalies in that vicinity may be at risk of dissolving in coming weeks.

We have the ability to determine if this is likely to happen.

Equatorial Pacific upper-ocean (through 300 meters) heat anomalies over the last year.
Source of graphic: CPC
Source of annotations: Author
El Nino and La Nina events can be driven by Equatorial Kelvin Waves, and whether the wave moving eastward along the Equatorial Pacific is upwelling or downwelling. If that sentence made you raise an eyebrow, you're most likely not alone. I can assure you, though it's actually pretty simple to understand. Let's break it down.

The phrase 'Equatorial Kelvin Wave' seems intimidating, so for our purposes here all we need to know is that, from time to time, these Equatorial Kelvin waves develop in the western part of the Equatorial Pacific and gradually move eastward along the Equator. When they move eastward along the Equator, they can be either 'downwelling' or 'upwelling' waves.
Consider the explanation of a 'downwelling' Equatorial Kelvin wave as described by the NOAA (read the full article here):

"Normally, winds blow from east to west across the tropical Pacific, which piles up warm water in the western Pacific. A weakening of these winds starts the surface layer of water cascading eastward..."

In other words, if this wind pattern that blows winds from east to west breaks down, that warmer than normal water begins pushing eastward along the Equatorial Pacific. This anomalously warm water works its way eastward gradually and tends to sustain itself in the process. As a consequence, downwelling Equatorial Kelvin waves tend to be associated with El Nino events. You can see my annotations of downwelling Kelvin waves as solid lines on the above image.
On the flip side, an 'upwelling' Equatorial Kelvin wave can be thought of as the ocean waters trying to get itself a little more in balance in the wake of this very warm downwelling wave. Thus, an upwelling wave again features a Kelvin wave slowly progressing eastward, but this time it cools down the upper-ocean waters to a degree that upwelling Kelvin waves are generally associated more with La Nina events. In the above image, I've made an attempt to outline upwelling Kelvin waves by the dashed lines.

Given that we've had three clear downwelling Equatorial Kelvin waves traversing the Equatorial Pacific over the last year, as outlined on the above image, it's not necessarily a shock that we are in an El Nino at this time. In addition, we are able to use the above image to see that the emergence of cooler than normal water temperatures in the eastern equatorial Pacific appear to be the result of an upwelling equatorial Kelvin wave, as shown by the dashed line at around 110º West longitude.

I want to point out something that could endanger the El Nino by the time we reach late summer/early fall, however. Referring back to the above image, note how an area of cooler than normal water temperatures have developed between longitude lines 130º E and 160º E. I've circled this swath for two reasons: because it is a rather expansive area of cooler waters relative to previous below-normal anomalies in the other two more-apparent upwelling Kelvin wave episodes, and because it has brought about the strongest negative temperature anomalies in that portion of the equatorial Pacific in at least a year. The risk here is that this is the beginning of another upwelling equatorial Kelvin wave, which will move eastward with time and bring those below-normal temperature anomalies to the Nino regions. Should this occur (and it is not a certainty yet), it could endanger the El Nino, which is already in a more fragile position as a consequence of the existing sub-surface below-normal water temperature anomalies. We will have to monitor this in the coming weeks to see if it is indeed an upwelling Kelvin wave or merely an isolated area of cooler waters in the western equatorial Pacific.

As discussed, however, even if this is not an upwelling Kelvin wave beginning to form, there has been a material change in the ENSO phenomenon as of late that necessitates discussion.

Upper-ocean heat anomalies in degrees Celsius between 180º and 100º West longitude.
Source: CPC
The Climate Prediction Center has allowed us to compile those same anomalies shown in the previous image into a single graphic. This chart shows upper-ocean heat anomalies (in degrees Celsius) between the longitude lines of 180º and 100º West, from the surface to 300 meters down, if I recall correctly. In the presence of an El Nino, these anomalies should be positive, while a La Nina should bring these anomalies into negative territory.

During the month of May, we have seen a drastic shift in upper ocean heat anomalies, although in reality this shift began in mid-March. Indeed, after peaking at 1.5º C above-normal in the middle of March, those positive anomalies rapidly declined, to the point that they're now just barely in below-normal territory. The negative anomalies aren't strong enough to point to a La Nina, but the change from strongly positive values to marginally negative values is not a trivial one.

What does it mean? It's a good view of the evaporating above-normal temperature anomalies below the surface of the equatorial Pacific that we discussed earlier. The negative anomalies are likely a bit overdone and not reflective of the true nature of the ENSO phenomenon, given that there is a small yet significant area of below-normal anomalies around the 120º West line of longitude that is most likely distorting this graphic to the downside. As such, while the degradation in positive anomalies is accurate and noteworthy, the recent move into negative territory doesn't seem to be precise in my eyes.

Temperature anomalies between 2º North and 2º South latitude at the subsurface depths of 55 meters (left), 105 meters (center) and 155 meters (right).
Source: CPC
We are able to again see this material deterioration in the El Nino by looking at temperature anomalies along the Equatorial Pacific at three different depths: 55 meters, 105 meters and 155 meters below the surface. The 55-meter chart on the left shows the recent resurgence in below-normal temperature anomalies, erasing the above-normal anomalies that were in place as recently as late April. The 105-meter chart in the middle gives a better look at the underlying trend - indeed, steadfast positive temperature anomalies between 150º East and the dateline have deteriorated since late April, not to mention the elimination of well-above-normal anomalies centered around the 130º West line of longitude.
Perhaps most alarmingly for the viability of the El Nino, the positive temperature anomalies at a depth of 155 meters (right) have completely disappeared and have instead been replaced by marginally below-normal anomalies ever since late March. The lack of a solid underwater base for the El Nino does not bode well for its survival into fall and winter, especially if that aforementioned swath of colder than normal waters around the space between longitude lines 130º E and 160º E does turn out to be another upwelling Kelvin wave. Only time will tell, but this will certainly be something to watch as we move into the fall months.

We've analyzed a lot of observed data for the ENSO phenomenon, but scientists have put in a lot of hard work to create climate models that can anticipate the state of the phenomenon down the road. Let's take a look at these forecasts.

IRI/CPC suite of forecasts for SST anomalies in the Nino 3.4 region.
Source: IRI of Columbia University
There are two graphics of model guidance I want to go over. The first comes courtesy of Columbia University, and depicts a variety of weather models' forecast for sea surface temperature anomalies in the Nino 3.4 region from now until the January-February-March period of 2020. As stated at the start of this post, however we will only discuss forecasts going into November and December, as model guidance begins to diverge too much for my liking beyond that period.

Model guidance is in pretty good agreement on keeping the El Nino around through at least the August-September-October (ASO) period, with one or two outliers both to the upside and downside of the consensus. Beyond that period, however, divergence increases, although a general theme through the November-December-January period is that positive SST anomalies appear probable, especially relative to the potential for negative anomalies. It is worth making mention, however, of a cluster of models that prefer taking the Nino 3.4 SST anomalies into a level below +0.5º Celsius but above zero, a neutral-ENSO scenario. For now, though, we will side with guidance that prefers a weak El Nino through the fall months.

NMME suite of forecasts for SST anomalies in Nino region 3.4.
Source: CPC
The NMME suite incorporates many of the models used in the IRI/CPC suite, but is worth going over anyway because of its variance with the first contingent of models analyzed. While this group of models foresees the El Nino as likely to persist into August, it seems as though there is a higher chance of SST anomalies dipping into the 0.0º through +0.5º region, not high enough to merit an El Nino classification but again more likely to exhibit El Nino conditions as opposed to La Nina conditions. Beyond September, though guidance diverges too much to gather an accurate forecast.

To Summarize:
- An El Nino is currently in place, with an El Nino Advisory declared by the Climate Prediction Center.
- Recent sea temperature anomalies below the surface suggest the El Nino may be undergoing a material degradation, potentially posing a threat to the survival of the phenomenon through the fall and early winter.
- Despite the apparent threats to the El Nino, model guidance sees the El Nino continuing into the fall months and perhaps into early winter. Beyond then, however, guidance diverges too much to ascertain what will transpire into the heart of winter 2019-2020.

Andrew