Sunday, October 29, 2023

2023-2024 Winter Outlook

 Disclaimer

        While I am a meteorologist, seasonal forecasting is not my expertise. However, it is a interest I have had for many years. After nearly a decade of putting seasonal forecasting on the back burner, I have decided to get back into it. Seasonal forecasting is a different breed than your typical day-to-day forecasting as seasonal forecasting relies on trying to understand how the global atmospheric (and oceanic) circulations will evolve several weeks-to-months down the road. Studying historical weather patterns and comparing them to current conditions and projected evolution can provide significant clues several weeks out. As always, there are caveats as meteorology is not an exact science. There will be degrees of error and sometimes forecasts will be flat out wrong. One trait of a great forecaster is trying to understand what went wrong within the forecast and using this as a learning process. When it comes to forecasting, one of the most important aspects of a forecast is understanding the science behind it. If a forecast was correct, but the methodology behind the forecast was wrong, the forecast was not correct. Understanding what went wrong within a forecast is one of the best methods for a forecaster to grow and increase their skill.

    This outlook will contain numerous composites. Two main datasets were used:

  1. 20th Century Reanalysis Version 3 (20CRV3) 
  2. NCEP/NCAR R1
20CRV3 contains data beginning January 1836 through December 2015 and NCEP/NCAR R1 contains data beginning January 1948 through present. Given years prior to 1948 are being assessed, it was imperative to use the 20CRV3 dataset to capture this data. You'll also notice (especially within the GIF animations) differing climatological periods. When dealing with anomalies (departure from average) I felt it imperative to compare a year (or period) to it's own climatological period of record (which is typically 30-years) rather than the most current climatological period of record which is 1991-2020. It doesn't make sense to compare a period in the 1920's to averaged conditions between 1991-2020. Table 1 provides a breakdown of the climatological periods with the year range for that climatological period (the years that would use that climatological period):

Table 1

    Since data for the NCEP/NCAR R1 dataset does not begin until January 1948, I used the 20CRV3 dataset between 1900-1980 and NCEP/NCAR dataset for 1981-Present. This is because the earliest climatological period for the NCEP/NCAR dataset would be 1951-1980 given the dataset does not begin until 1948.

ENSO (EL Nino-Southern Oscillation)

Background

    The state of ENSO can play a significant role in the oceanic and atmospheric circulation patterns worldwide. When EL Nino or La Nina conditions are presented, established, and the ocean is strongly coupled with the atmosphere, these conditions can be significant drivers of the atmospheric circulation patterns. What are EL Nino and La Nina and what state are we in? The most common and widely used method of defining EL Nino or La Nina conditions is via the Oceanic Nino Index (ONI) created by the Climate Prediction Center (CPC) which focuses on sea-surface temperature anomalies (departure from average) within the equatorial Pacific. The equatorial Pacific is broken down into 4 different ENSO regions (Figure 1).

    Per the CPC's definition, an ENSO event is defined as EL Nino or La Nina when sea-surface temperature anomalies (SSTAs) between latitudes 120°W and 170°W and longitudes 5°S and 5°N (designated as ENSO Region 3.4) are at or above +0.5°C for a minimum of 5-consecutive trimonthly periods (EL Nino) or at or below -0.5°C for a minimum of 5-consecutive trimonthly periods (La Nina).

Figure 1: Courtesy of https://www.ncei.noaa.gov/monitoring-content/teleconnections/nino-regions.gif

    While the CPC's method is most widely used and accepted, it is important to understand that many other definitions are published and accepted. Earlier in 2022, Eric Webb introduced the Ensemble Oceanic Nino Index (ENS-ONI). For more information on Eric Webb please visit the following:

https://www.webberweather.com/bio.html

For more information on the ENS-ONI please visit the following:

https://www.webberweather.com/ensemble-oceanic-nino-index.html

A link to his paper further explaining the ENS-ONI can be found here:

https://rmets.onlinelibrary.wiley.com/doi/10.1002/joc.7535

    When creating a list of EL Nino and La Nina winters, I used both the ONI and ENS-ONI. The ONI only dates to 1950 while the ENS-ONI dates to 1850, however, my focus is from 1900-Present. While accuracy and validity of many meteorological variables decreases as one advances backward in time, it is still important to build upon a larger data set. 
 
    It should also be understood that the ONI and ENS-ONI are focusing solely on oceanic conditions (temperatures and temperature anomalies) within the equatorial Pacific. Due to the coverage of the Pacific Ocean, anomalous water temperatures within the basin can heavily influence atmospheric circulations, however, this is not always the case. There are two components to ENSO: oceanic component (sea-surface temperatures) and atmospheric component (air pressure, wind speed, wind direction). If the ocean-atmosphere are coupled this will strengthen the impact of the ENSO event and the ENSO event will play a significant role in atmospheric circulation patterns. If the two are not coupled, the role the ENSO event will have on the atmospheric circulation patterns will be lessened. 

    The Southern Oscillation Index (SOI), which is a measure of air pressure differences between Tahiti and Darwin, Australia and the Multivariate ENSO Index (MEI), which incorporates sea-level pressure, the zonal (west-east) and meridional (north-south) component of the surface wind, sea-surface temperatures, and outgoing longwave radiation (measure of clouds/precipitation) are two indices which can provide context into how coupled the ocean-atmosphere are. 

    In addition to defining ENSO state, the strength of an ENSO event, evolution of the ENSO event, and structure of the ENSO event (placement of greatest SSTAs) can play a critical role in the event's impact on atmospheric circulations. Strength and location of tropical forcing (rain/thunderstorms) can also be critical. 

Current ENSO State

    After three-consecutive years of La Nina conditions present within the equatorial Pacific, La Nina quickly faded during the winter of 2022-2023. Since the demise of La Nina, we've seen a rapid reversal of sea-surface temperatures within the equatorial Pacific with a transition from cooler-than-average waters to warmer-than-average waters (Figure 2).
Figure 2: Courtesy of https://vortex.plymouth.edu/myowxp/sfc/sst-a.html

    SSTAs within the four ENSO regions are also solidly above-average (Figure 3). After rapid warming through the summer months in all regions there has been some leveling off towards the end of the summer and early this fall. Note: ENSO Region 1.2 is the most volatile region and is prone to the greatest and quickest temperature changes. 

Figure 3: Courtesy of https://www.cpc.ncep.noaa.gov/products/analysis_monitoring/enso_update/ssta_c.gif

    Earlier this summer, the CPC announced the emergence of EL Nino. Based on the current SSTA configuration across the equatorial Pacific and the longevity of these anomalies, we will be looking at an EL Nino state for the winter of 2023-2024. Now that we've established the ENSO state, it is important to understand previous EL Nino events and how each played a role in atmospheric circulation patterns.

Evaluating Historical EL Nino Events

     Table 2 shows a list of all EL Nino winters (40 in total) combining the ONI and ENS-ONI. For this outlook, winter is defined as December-January-February-March (DJFM).

Table 2

    Figures 4a (left) and 4b (right) below portray 500mb height anomalies for all EL Nino winters combined. Two sets of composites were created as the reanalysis page allows up to a maximum of 20 individual years to be input and we have a total of 40. Two different data sets were also used: 20th Century Reanalysis Version 3 (20CRV3) and NCEP/NCAR R1. Data within 20CRV3 starts 1836 and ends in 2015. Data within NCEP/NCAR starts 1948 through the present. Given we're dealing with years dating prior to 1948, the 20CRV3 dataset was used to plot the first set of 20 EL Nino events (1902-1903 to 1963-1964) and the NCEP/NCAR dataset was used to plot the next set of 20 EL Nino events (1965-1966 to 2018-2019). A climatological period of 1900-1950 was used for the first set with a climatological period of 1960-2010 was used for the second set.

Figure 4a (left) Figure 4b (right): Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/

    We can draw three conclusions on what seems to be typical during an EL Nino winter with respect to the 500mb pattern across the northern hemisphere:
  1. Below-average heights (trough) around the Aleutians extending into the Gulf of Alaska. This indicates a deeper than usual Aleutian Low Pressure. 
  2. Above-average heights (ridging) within the Arctic domain extending into Canada and northern-tier of the United States. 
  3. Below-average heights (trough) across the southern half of the United States. 
    Note: You may notice the below-average heights across the southern half of the United States seem to be more pronounced in the first set of EL Nino winters. This may be due to the dataset being used (20CRV3 vs. NCEP/NCAR) and it may also be indicative of EL Nino events acting a bit more differently. There is some research indicative of this, however, that goes beyond the scope and purpose of this outlook. 

    If we look at each EL Nino event separately (Figure 5), we can see that a degree of spread exists between each event where these prominent features may vary in strength, structure, or even a complete reversal of these anomalies.

Figure 5: Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/

    In terms of temperatures, this pattern typically correlates to above-average temperatures across Canada and the northern-tier of the United States with below-average temperatures across the southern United States (Figure 6a and 6b). The 40 years were split into two composites like what was done with the 500mb height anomalies. 

    Note: You may also notice below-average temperatures seem to be more pronounced in the first set of 20 EL Nino winters. This also may be a product of the different datasets being used, EL Nino's behaving differently more recently, and also the legends are not equal.

Figure 6a (left) 6b (right): Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/

    Looking at each EL Nino event separately (Figure 7), you can see a degree of spread can exist well from one EL Nino event to the next.

Figure 7: Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/
    
    In terms of precipitation, EL Nino winters typically feature drier-than-average conditions across the Pacific-Northwest, northern Inter-mountain West, and within the Ohio and Tennessee Valley regions. Wetter-than-normal conditions are seen within California and along the southern United States (Figure 8a and 8b). The 40 EL Nino winters were split into two composite sets as well with a climatological period of 1895-2000 used. 
 
Figure 8a (left) 8b (right): Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/

       Note: The second set of 20 years seems to feature wetter-than-average conditions across California, Southwest, and Great Plains. While this is also partly due to the unequal legends, this could also be due to behaving more differently recently.

    Looking at precipitation anomalies for each individual EL Nino event (Figure 9), it's pretty clear significant variation can occur on an event-to-event basis.

Figure 9: Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/

    Mentioned in the background, the strength of an ENSO event (and in this case, EL Nino) can also be very important. A weaker event, especially when the ocean and atmosphere aren't coupled, may have little impact and influence on atmospheric circulation patterns whereas a stronger event, especially when the ocean and atmosphere are coupled, may have a significant influence on atmospheric circulation patterns. 
   
    The strength of ENSO events is characterized as weak, moderate, and strong with a super-strong designation for the most anomalous of events. One metric for characterizing the strength of an ENSO event is based on the SSTAs within ENSO Region 3.4. Table 3 provides the SSTA thresholds for each strength designation.
Table 3

    As Eric Webb points out on his website, there are also other metrics for classifying the strength of an ENSO event. One other metric uses the values of the ENS-ONI and ranks based on percentile. This can be read within the "ENSO Phase and Intensity" section on Eric Webb's website (linked above). For the purposes of this outlook, the CPC's method was used. 

    Using the thresholds presented in Table 3 and the list of EL Nino winter's from Table 2, Table 4 provides a breakdown of each EL Nino event by strength.

Table 4

    The criteria I used to define the strength was using the peak trimonthly ONI or ENS-ONI value during the event. There were several events in which the peak ONI was on the borderline of the strength designation thresholds. For those events, I added them into the weaker of the two phase categories but denoted it was borderline. There were also some events where the ONI or ENS-ONI differed slightly and they were categorized accordingly as well. Note the super-strong events of 1965-1966, 1972-1973, 1982-1983, 1997-1998, and 2015-2016.

    Like we did above with all EL Nino winters, we will look at 500mb height anomalies, temperature anomalies, and precipitation anomalies for weak EL Nino winters, moderate EL Nino winters, strong EL Nino winters, and super-strong EL Nino winters. Doing so can help us identify if we can find consistent signals depending on the strength of the event. 

    Given what was discussed within the disclaimer about the different datasets, two composites have to be made to illustrate what is typical of each strength in terms of 500mb height anomalies, temperature, and precipitation. One composite will consist of years which the 20CRV3 can incorporate and the other will consist of years with the NCEP/NCAR R1 can incorporate. 

    Figure 10a and 10b below illustrate a composite of 500mb height anomalies combining all weak EL Nino events. There are striking differences noted between figure 10a and 10b. Below-average heights are quite pronounced just south of the Aleutians and across the eastern third of the United States within figure 10a. These two features are noticeably absent in figure 10b. 

Figure 10a (left) 10b (right): Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/
    
This is something that will be discussed later in this outlook, but the height anomalies around the Aleutians are part of what is called the Pacific-North American (PNA) teleconnection index. As we'll see looking at each individual weak EL Nino event below (Figure 11), in these earlier years, there were some winters in which the below-average height anomalies (indicative of a much stronger-than-average Aleutian Low) were quite pronounced along with the below-average heights across the eastern United States. 

Figure 11: Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/

    Temperature anomalies associated with weak EL Nino winters (Figure 12a and 12b) typically feature below-average temperatures across the eastern third of the United States with above-average temperatures across the western United States. Since temperatures can be strongly correlated with the 500mb pattern, it's not a surprise to see the cool anomalies dampened in figure 12b which is coincident with the lack of below-average heights within this region.

Figure 12a (left) 12b (right): Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/

    Assessing each weak EL Nino individually (Figure 13) we can see a good portion of weak EL Nino winters featured above-average temperatures towards the west with below-average temperatures within the east, however, there are some events which don't fit this mold.

Figure 13: Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/
    
    Precipitation Anomalies during weak EL Nino winters (Figure 14) tend to feature drier-than-average conditions across the Pacific-Northwest, northern Intermountain West, and within the Tennessee Valley region with wetter-than-average conditions within the Southwest and Florida.

Figure 14: Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/

       Assessing each weak EL Nino individually (Figure 15) shows that there can be significant variation in precipitation anomalies for a given weak EL Nino winter with the greatest variation occurring within the western United States.

Figure 15: Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/

    Moving into the set of moderate EL Nino winters, the signal of below-average heights in the vicinity of the Aleutians into the Gulf of Alaska and across the eastern United States is much more pronounced than with weak EL Nino winters. (Figure 16a and 16b).

Figure 16a (left) and 16b (right): Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/

     Assessing the 500mb height anomalies for each moderate EL Nino individually (Figure 17) we see there is a clear signal for below-average heights within the southern and eastern United States, however, we also see some of the more recent moderate EL Nino's have deviated from this with more in the way of ridging (above-average height anomalies).

Figure 17: Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/

        Temperature anomalies during moderate EL Nino winters tend to be cooler-than-average across much of the United States, particularly the eastern third of the country (Figure 18a and 18b).

Figure 18a (left) 18b (right): Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/

    Precipitation during moderate EL Nino winters (Figure 19) tends to be below-average across the Pacific-Northwest, northern Inter-mountain West, Ohio/Tennessee Valley regions, and within the Northeast. Above-average precipitation tends to occur within California and within the far Southeast into the mid-Atlantic.

Figure 19: Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/

    However, assessing each moderate EL Nino event separately (Figure 20), we can see a great deal of spread can exist event-to-event, similar to that of weak EL Nino winters. 

Figure 20: Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/

    Getting into strong EL Nino winters, the 500mb pattern (Figure 21a and 21b) continues with the theme of below-average heights around the Aleutians and Gulf of Alaska indicating a stronger Aleutian Low with below-average heights across the southern United States. Again we note that this signal becomes a bit more muted in the more recent stretches of strong EL Nino episodes.

Figure 21a (left) 21b (right): Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/
    
    Assessing the 500mb height anomalies for each strong EL Nino winter individually (Figure 22) we can clearly see a major shift in the pattern over the United States between EL Nino events pre-1970 and post-1970 with strong EL Nino winters post-1970 featuring more in the way of ridging within the country.

Figure 22: Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/
    
    Given the significant differences within the 500mb pattern between strong EL Nino events prior to 1970 and post-1970, it's no surprise we see that reflected with temperature anomalies as well (Figure 23a and 23b).

Figure 23a (left) 23b (right): Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/

    Precipitation anomalies for strong EL Nino winters (Figure 24) are certainly a bit different than during weak or moderate EL Nino winters. While the northern Inter-mountain West and Ohio/Tennessee Valley regions are drier-than-average, the Pacific-Northwest is a bit on the wetter side with above-average precipitation with California and along the deep South,

Figure 24: Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/

    In Figure 25, assessing each strong EL Nino winter individually, we can see there can be significant variation in precipitation anomalies which falls in line with what we've seen during weak and moderate events.

Figure 25: Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/

    We can also separate strong EL Nino events into a separate category - super-strong as a way to distinguish the most anomalous of events. 

    The 500mb pattern during super-strong EL Nino winters (Figure 26) features below-average heights associated with a very strong Aleutian Low within the Gulf of Alaska with the below-average heights across the southern United States less pronounced while above-average heights across Canada into the northern-tier of the United States are much more pronounced.

Figure 26: Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/

    Assessing the 500mb height anomalies for each super-strong EL Nino winter individually (Figure 27), albeit a small sample-size, we generally see a muted pattern featuring below-average height anomalies across the southern United States with more pronounced above-average height anomalies across Canada and northern-tier of the United States. Note the winter of 2015-2016, almost absent of the below-average height anomalies across the southern United States.

Figure 27: Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/

    Temperatures associated with super-strong EL Nino winters (Figure 28) tend to be above-average across the northern-tier of the United States and with the lack of below-average heights within the southern United States, temperatures are fairly close to average.

Figure 28:Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/
    
Assessing each super-strong EL Nino winter individually (Figure 29) shows these events tend to feature above-average temperatures across most of the country with the exception being the super-strong EL Nino event of 1965-1966.

Figure 29:Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/
  
  In terms of precipitation (Figure 30), super-strong EL Nino winters tend to be wetter-than-average across the Pacific-Northwest (a huge change from other EL Nino strength classification), West, and the Northeast into the mid-Atlantic with somewhat drier-than-average conditions across the south (also a big change).


Figure 30: Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/
   
    When assessing precipitation anomalies for each individual super-strong EL Nino winter (Figure 31) we see there isn't a ton of deviation, however, we note the very dry winter of 1965-1966 which likely heavily skews the mean given the small data sample.
 

Figure 31: Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/

  ENSO Evolution 

    Now that we've exhaustively looked at historical EL Nino events (with a focus on the 500mb pattern, temperature anomalies, and precipitation anomalies we want to assess how this EL Nino has evolved thus far (with respect to sea-surface temperatures) and how the EL Nino projects to evolve. There are two types of EL Nino:

1) Canonical which is your typical EL Nino in which sea-surface temperatures warm off the Pacific side of the South American coast, or ENSO region 1.2 (Please revert back to Figure 1), and build westwards across the equatorial Pacific as easterly trade winds weaken.   

2) Modoki - This EL Nino event consists of the warmer water within the central Equatorial Pacific with colder waters east and west of these warmer anomalies. 

    Now only is how the EL Nino develops critical, but its evolution can be critical as well. For example, there are some canonical EL Nino events which transitioned to modoki events as the winter progressed. This is something that has to be kept in mind as we near closer to northern hemisphere winter. 
 
   As we assess ENSO evolution we are going to focus on the following:

1) The CPC's Oceanic Nino Index (ONI) which we will use as a guide in classifying the strength of this EL Nino event. 

2) The Southern Oscillation Index (SOI) - This product is widely regarded as a better indicator of the actual ENSO strength than the ONI alone. 

3) The Multivariate ENSO Index (MEI) - This product is a great tool to assess how coupled the ocean-atmosphere are. 

4) Evolution of sea-surface temperatures - This can tell us alot about the structure of the event. Are the warmest anomalies located in the eastern portion of the ENSO region (east-based event), do they build uniformly throughout the ENSO region (basin-wide event), or do they shift towards the western portion of the basin (west-based event)? Of course, we also need to keep in mind the modoki event.  

5) Depth of warmest temperatures - This can be a great indicator in assessing whether the ENSO event may continue to strengthen or not. 

6) Trade winds at the 850mb level - Across the northern hemisphere, trade winds blow east-to-west. This normal flow tends to yield upwelling of cooler waters off the Pacific side of South America and the easterly trade winds push the waters westward. However, when these trade winds weaken (or in some cases, reverse) waters can warm and this can lead to the development of EL Nino conditions. 

Oceanic Nino Index (ONI)

    Earlier within this outlook, we touched upon the ONI and Table 3 provided thresholds of sea-surface temperature anomalies for each strength designation. In Table 5 below, we look at the ONI going back to 2010. EL Nino events are highlighted in red with La Nina events in blue. While we don't see a "red" or EL Nino designation yet, this is because we have not had 5 consecutive trimonthly periods where the ONI has exceeded +0.5°C. This will occur once the ASO (August-September-October) value is calculated early in November. We do see, however, that JJA value was +1.1°C and the JAS value was +1.3°C. These values suggest we are currently dealing with a moderate EL Nino (referring to the thresholds provided in Table 3).

Table 5: Courtesy of https://origin.cpc.ncep.noaa.gov/products/analysis_monitoring/ensostuff/ONI_v5.php

   Southern Oscillation Index (SOI)

        As mentioned previously, the SOI is an index which measures the differences in sea-level pressure between Tahiti and Darwin, Australia. When SOI values are positive, this coincides with above-average sea-level pressure over Tahiti with below-average sea-level pressure over Darwin, Australia. Negative values of the SOI yield an opposite sign of pressure anomalies at each location. Given EL Nino's are typically associated with below-average sea-level pressure over Tahiti and above-average sea-level pressure over Darwin, during EL Nino episodes, the more negative the SOI the more coupled the ocean-atmosphere are said to be. Table 6 provides SOI values dating back to 2010. Figure 32a and 32b shows the average sea-level pressure anomalies in the vicinity of Tahiti and Australia for EL Nino events.

Figure 32a (left) 32b (right): Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/

Comparing to the defined EL Nino and La Nina events from Table 5, you can see the La Nina episodes matching up with prolonged periods of positive SOI values and EL Nino episodes matching up with prolonged periods of negative SOI values. The SOI values presented below were obtained from the following source: https://crudata.uea.ac.uk/cru/data/soi/

Table 6

    Since the demise of La Nina during the 2022-2023 winter, we note that SOI values have quickly transitioned from positive to negative, falling in line for what we would expect with a rapidly developing EL Nino. As mentioned above, the SOI is regarded as a better tool for evaluating the strength of an ENSO event over just the ONI itself. 

    Table 7 provides SOI values for each EL Nino event, looking at SOI values during the summer through early fall and then again during the late fall through early spring. 

Table 7

    Table 8 provides a breakdown of SOI values for the same periods as Table 7 for each weak, moderate, and strong EL Nino based on the ONI.

Table 8

Figure 33a - 33d look at sea-level pressure anomalies for weak EL Nino events during the summer through early fall (33a and 33b) and late fall through early spring (33c and 33d). 

Figure 33a (top left) 33b (top right) 33c (bottom left) 33d (bottom right): Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/

    Figure 34a - 34d look at sea-level pressure anomalies for weak EL Nino events during the summer through early fall (34a and 34b) and late fall through early spring (34c and 34d). 

Figure 34a (top left) 34b (top right) 34c (bottom left) 34d (bottom right): Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/

    Figure 35a - 35d look at sea-level pressure anomalies for weak EL Nino events during the summer through early fall (35a and 35b) and late fall through early spring (35c and 35d). 

Figure 35a (top left) 35b (top right) 35c (bottom left) 35d (bottom right): Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/

We can easily conclude the structure of sea-level pressure anomalies around Tahiti and Australia become much more apparent during moderate and strong EL Nino events with a stronger signal of below-average sea-level pressure anomalies around Tahiti and above-average sea-level pressure anomalies around Australia.  

    Next we will assess SOI values since the beginning of the year (Table 9). It should be noted when looking at current SOI values the Climate Prediction Center's (CPC) dataset was used. This is because the CPC has a more recent update of the SOI than the previous source provided above. It should be noted that each source has a slightly different calculation so if comparing the two datasets, the overall numbers may differ. The CPC data was obtained from the following source:
https://www.ncei.noaa.gov/access/monitoring/enso/soi

Table 9

If we look at sea-level pressure anomalies from June-September and thus far in the month of October (Figure 36a and 36b) we note the above-average sea-level pressure anomalies around Australia, however, sea-level pressure anomalies around Tahiti are relatively close to average (both during the summer and the month of October). 

Figure 36a (left) 36b (right): Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/ 

SOI Conclusion: At least thus far, while the ONI is indicative of a moderate EL Nino event, the structure of sea-level pressure anomalies associated with the SOI are not as representative of a moderate EL Nino. This is something to closely watch going through the next several weeks because if we don't see sea-level pressure anomalies respond to the oceanic conditions, the event will not be as strong advertised. 

Multivariate ENSO Index (MEI)

    The MEI incorporates 5 different variables (sea-level pressure, the zonal and meridional components of the surface wind, sea-surface temperature, and outgoing longwave radiation). Given this incorporates both oceanic and atmospheric components, the MEI is a great tool to assess how coupled the ocean-atmosphere are during a particular event. This can also be used as a metric for strength as if an event is strongly coupled, the event is more likely to be a significant driver in the atmospheric circulation. Positive values of the MEI correlate to EL Nino conditions with prolonged periods of positive values linked to EL Nino episodes. Negative values of the MEI correlate to La Nina conditions with prolonged periods of negative values linked to La Nina episodes. It should be noted MEI data only dates to 1979, thus we can't evaluate the MEI for ENSO events prior to 1979. 

    Table 10 provides MEI values dating back to 2010. Comparing to the defined EL Nino and La Nina events from Table 5, you can see prolonged periods of positive MEI values correlate to EL Nino events while prolonged periods of negative MEI values correlate to La Nina events. All MEI values below were obtained from the following source: https://psl.noaa.gov/enso/mei/

Table 10
    Table 11 provides MEI values for each EL Nino event, looking at MEI values during the late spring and early summer through fall and then again during the fall through late winter.

Table 11
    
    Table 12 provides a breakdown of MEI values for the same periods as Table 11 for each weak, moderate, and strong EL Nino based on the ONI.

Table 12

    Given EL Nino events are associated with positive MEI values, given that we know the MEI is a great tool to assess how coupled the ocean-atmosphere are, we can conclude the more positive the MEI value, the more coupled the ocean-atmosphere is. Values close to and above 1 represent and ocean-atmosphere coupling which is stronger while values closer to (or below) 0.50 represent a much weaker coupling. 

    Looking at MEI values since the beginning of the year (Table 13) we see the negative MEI values associated with the decaying La Nina during the end of winter with a rapid rise in the MEI through the summer with the latest reading at 0.6. 

Table 13

MEI Conclusion: The MEI has rapidly transitioned from negative to positive, which falls in line with the rapidly decaying La Nina and quick transition into EL Nino conditions. Should the MEI continue to rise over the next few months, this will indicate an ocean-atmosphere which is strongly coupled and enhance the impact the EL Nino will have on the atmospheric circulation. We will discuss the MEI a bit more below with some additional insight on where the MEI may trend. 

Evolution of Sea-surface Temperatures

    Earlier in this presentation of this outlook, we briefly looked at the evolution of sea-surface temperature anomalies over the past year (Figure 2). We are going to go more in depth regarding the evolution of sea-surface temperatures, assess the depth of the warmer waters, and look at 850mb trade winds. We will use this assessment, combined with incorporating the ONI, SOI, and MEI to gauge a basis of where this EL Nino may head strength wise as we move into the winter. 

    Figure 37 looks at sea-surface temperatures within the equatorial Pacific from late July towards mid October with Figure 38 looking at sea-surface temperature anomalies during the same period. We can note the following:

1. Very warm sea-surface temperature anomalies spanning the equatorial Pacific (consistent with EL Nino).

2.  A westward expansion of well above-average sea-surface temperature anomalies into the equatorial Pacific Ocean. 

3. A decrease and reduction of the more extreme sea-surface temperature anomalies with the greatest decrease within the Nino 1.2 region.

Figure 37: Courtesy of https://www.cpc.ncep.noaa.gov/products/analysis_monitoring/enso_update/sstanim.shtml
Figure 38: Courtesy of https://www.cpc.ncep.noaa.gov/products/analysis_monitoring/enso_update/sstanim.shtml

        Figure 39 looks at the depth of sea-surface temperatures across the equatorial Pacific with Figure 40 looking at the anomalies at different depths. We note the following:

1.The depth of the warmer waters across the eastern equatorial Pacific are very shallow with the greatest depth of warmer waters out towards the central and western Pacific. Now - this isn't totally uncommon, especially since the eastern equatorial Pacific typically features the upwelling of cooler waters due to the cool ocean current off the South American coast. 

2. Moving into the Fall, we note a significant decrease in the depth of the warm anomalies within the eastern equatorial Pacific. 

Figure 39: Courtesy of https://www.cpc.ncep.noaa.gov/products/analysis_monitoring/enso_update/wkxzteq.shtml

Figure 40: Courtesy of https://www.cpc.ncep.noaa.gov/products/analysis_monitoring/enso_update/wkxzteq.shtml

    Next, we'll look at 850mb zonal wind anomalies across the equatorial Pacific (Figure 41). Blues and purples correlate to easterly and stronger easterly trade winds with reds and pinks correlating to weak easterlies or even westerlies getting towards the pinks. We note a recent strong westerly wind burst between 150°W and 150°E. This recent westerly wind burst could help to re-ignite the EL Nino a bit over the next week. We also note the forecast for another burst of westerly winds (though not as intense) moving towards the end of the month. 

Figure 41: Courtesy of http://mikeventrice.weebly.com/hovmollers.html

ENSO Evolution Conclusion: While we saw a rapid transition from La Nina to EL Nino and the two most recent trimonthly value of the ONI are indicative of a moderate EL Nino, the SOI tells us that this event is not as strong as indicated (thus far), however, the MEI indicates the ocean-atmosphere are likely to become coupled which will enhance the influence of the EL Nino event. The assessment of the sea-surface temperature evolution tells us there isn't a whole lot in place for this EL Nino event to strengthen any further, however, this recent westerly wind burst may give the EL Nino a bit of a push. Based on all of this, I am anticipating this EL Nino will end up as a strong EL Nino (per ONI criteria), however, unless the sea-level pressure anomalies around Tahiti and Australia become more reflective of a strong EL Nino, this EL Nino may act more like a moderate EL Nino event. 

Tropical Forcing

    In addition to a particular ENSO event, strength, and structure, tropical forcing can play a critical role in the atmospheric circulation and can sometimes be a better discriminator than ENSO itself. One measure of tropical forcing is by assessing outgoing longwave radiation (OLR). From the National Center for Atmospheric Research (NCAR), OLR is a measure of the amount of energy emitted to space by Earth's surface, oceans, and atmosphere. As such, it is a critical component of the Earth's radiation budget. OLR values are often used as a proxy for convection in tropical and subtropical regions since cloud top temperatures are an indicator of cloud height (National Center for Atmospheric Research Staff (Eds). Last modified 2022-09-09 "The Climate Data Guide: Outgoing Longwave Radiation (OLR): HIRS." Retrieved from https://climatedataguide.ucar.edu/climate-data/outgoing-longwave-radiation-olr-hirs on 2023-10-22). Essentially, the location and strength of convection (rain and thunderstorms) can play a critical role in the atmospheric circulation. 

    The schematic below (Figure 42) presents the Pacific Walker Cell Circulation pattern during ENSO neutral conditions, EL Nino conditions, and La Nina conditions. 

Figure 42: Courtesy of https://www.climate.gov/news-features/blogs/enso/walker-circulation-ensos-atmospheric-buddy

    We can see that during EL Nino conditions, convection is typically shifted east of 180° while during ENSO Neutral or La Nina conditions convection is typically west. As shown in Figure 43, the placement of convection (indicated by the darker blues) can differ from EL Nino event-to-event as well as how robust the degree of convection can be.
    
Figure 43: Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/ 

    After assessing OLR anomalies for each individual EL Nino event. EL Nino events were also broken down based on where the tropical forcing (indicated by the convection) was located (Table 14). Given what we see with the schematic presented in Figure 42 which shows an eastward shift in convection during EL Nino events, it makes sense the list of events where OLR anomalies are east of the dateline is the largest while the list of events where OLR is focused west of the dateline is least (which is more typical of ENSO Neutral and La Nina conditions).

Table 14

    Figures 44-46 show averaged OLR anomalies for December-March broken down into being focused West of the Dateline (Figure 44), Centered Around the Dateline (Figure 45), and East of the Dateline (Figure 46). 

    
Figure 44: Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/ 

Figure 45: Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/ 

Figure 46: Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/ 

    Figures 47-48 look at 500mb height anomalies for EL Nino events where tropical forcing is focused west of the dateline (Figure 47a, Figure 47b, Figure 48). As the length of this outlook is already beginning to rival The Odyssey I am only going to post composites looking at 500mb. 
  
Figure 47a (left) 47b (right): Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/ 

Figure 48: Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/ 

    Figures 49-50 look at 500mb height anomalies for EL Nino events where tropical forcing is centered around the dateline (Figure 49a, Figure 49b, Figure 50).

Figure 49a (left) 49b (right): Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/ 

Figure 50: Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/ 
    
Figures 51-52 look at 500mb height anomalies for EL Nino events where tropical forcing is east of the dateline (Figure 51a, Figure 51b, Figure 52).

Figure 51a (left) 51b (right): Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/ 

Figure 52: Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/ 

Table 15 provides a breakdown of where the tropical forcing is located for each EL Nino strength:

Table 15

    What is the current state of tropical forcing? Figure 53a (left) looks at OLR anomalies across the equatorial Pacific since November of 2022 with figure 53b (right) looking at OLR anomalies from a global perspective with the date centered on October 22, 2023. 

Figure 53: Courtesy of https://www.cpc.ncep.noaa.gov/products/precip/CWlink/MJO/enso.shtml#current

    Tropical Forcing Conclusion: Dating back to last fall, the greatest degree of convection and tropical forcing has been predominately west of the dateline, even with the emergence of EL Nino, which as we know from above, is not very common during EL Nino episodes. It is very possible we could see an eastward shift with the tropical forcing as we move towards winter, but for now tropical forcing has been consistently west of the dateline. 

Pacific Decadal Oscillation (PDO)

    Our next assessment revolves around the PDO. The PDO is often described as a long-lived EL Nino-like pattern of climate variability (Zhang et al. 1997). While ENSO episodes may only last several months to a few years, the state of the PDO (positive or negative) can be dominant for as long as 20-30 years with some intra-year variability due to influences of ENSO state. Figure 54 provides the typical sea-surface anomaly patterns associated with each phase of the PDO. Given the structure of sea-surface temperature anomalies within the equatorial Pacific, there is a close connection between PDO and ENSO with EL Nino events more likely to occur during warm phases of the PDO and La Nina events more likely to occur during cool phases of the PDO, however, as we'll see further down that is not always the case!
Figure 54: Courtesy of http://research.jisao.washington.edu/pdo/

Where do we stand with the PDO? Looking at the structure of sea-surface temperature anomalies within the Pacific (Figure 55) and comparing to that of Figure 54, we see a resemblance to the cool phase or a negative PDO given the highly anomalous warm sea-surface temperatures spanning the north Pacific, however, we notice over the past month these warm anomalies have started to fade. Given this, it appears the negative PDO is rapidly eroding.
Figure 55: Courtesy of https://vortex.plymouth.edu/myowxp/sfc/sst-a.html

    Table 16 provides a breakdown of PDO values for the months of December - March and averaged for the period during EL Nino Winter's. It is pretty clear majority of EL Nino events are associated with the warm phase of the PDO, but that is not always the case. The PDO data was obtained from the following source: https://www.ncei.noaa.gov/access/monitoring/pdo/

Table 17
    
What is the importance of the PDO? Figure 56 looks at 500mb height anomalies for some of the most positive PDO winter's on record (Figure 56a) and most negative PDO winter's on record (Figure 56b). We can see  positive PDO's typically result in a stronger Aleutian Low with a much weaker Aleutian Low during negative PDO's. Given what we know about EL Nino's influence on the Aleutian Low, the combination of EL Nino and positive PDO can result in a very strong Aleutian Low.

Figure 56a (left) and 56b (right): Image provided by the NOAA-ESRL Physical Sciences Laboratory, Boulder Colorado from their Web site at https://psl.noaa.gov/ 

Pacific-North American Teleconnection Pattern (PNA)

    While the PDO is a long-lived pattern of climate variability, the PNA is more on the shorter scale in terms of fluctuations between the positive phase and negative phase. The PNA is one of the most prominent modes of low-frequency variability in the Northern Hemisphere extratropics  and consists of a pattern of air pressure anomalies at four locations over the Pacific Ocean and North America (Climate Prediction Center). Figure 57 shows the average 500mb pattern for the positive phase and negative phase of the PNA. 

Figure 57: Courtesy of https://www.climate.gov/news-features/understanding-climate/climate-variability-pacific-north-american-pattern

        If we take a look at the positive phase of the PNA, we notice striking similarities to the 500mb pattern associated with a positive PDO and EL Nino conditions with the deep Aleutian Low. Table 18 provides PNA values for December-March and averaged over these months for all Nino's. Note that data for the PNA only goes back to 1950. PNA data was obtained from the following source: https://www.ncei.noaa.gov/access/monitoring/pna/

Table 18

PDO/PNA Conclusion: Given the recent trends regarding sea-surface temperatures within the Pacific Ocean and a focus on the North Pacific where we've seen weakening warm anomalies, the negative PDO seems to be rapidly weakening. It is very possible we see a PDO signal which isn't overly strong this winter and we see the PDO continue to become less negative and perhaps become a bit more reflective of a positive PDO. 

    For the PNA pattern, we can likely expect a PNA which is more positive than negative moving through the winter, however, given the PNA can go through short-term fluctuations we may see periods where the PNA is positive (perhaps even quite positive at times, especially if the degree of ocean-atmosphere coupling becomes strong) and periods where the PNA is more neutral or even slightly negative. The PNA will be a big player in how active the weather pattern is across the United States this winter.

Quasi Biennial Oscillation (QBO) and Influences on the Stratosphere and Polar Vortex (PV) 

    One of our final aspects we will discuss is the QBO and its influence on the stratosphere and the polar vortex (both the stratospheric polar vortex and tropospheric polar vortex). The pattern configuration across the Arctic can be a significant player in the weather conditions across the United States, particularly the northern portion of the country. The QBO measures zonal winds within the stratosphere which circulate the equator. These winds alternate between westerly and easterly and start at about 10mb and migrate downwards before dissipating at 80mb. As these winds descend through the stratosphere they are replaced by winds of the opposite direction. Descending westerly winds within the stratosphere are known to enhance, or strengthen the polar vortex while descending easterly winds increase the potential for sudden stratospheric warming (SSW's) events and a weaker polar vortex. This can increase the potential for blocking patterns across the high latitudes of the Arctic, favoring shots of colder, Arctic air to become displaced into the northern United States. 

    Table 19 provides QBO values since the beginning of 2022. QBO data was obtained from the following source: https://acd-ext.gsfc.nasa.gov/Data_services/met/qbo/QBO_Singapore_Uvals_GSFC.txt

Table 19

    QBO Conclusion:
Since winter of 2022-2023, we have seen a reversal in the QBO with the westerly phase of the QBO diminishing and a developing easterly phase with descending easterly winds. This could have significant implications on the Arctic domain this winter, particularly on the polar vortex. The polar vortex is a low pressure system which exists 24/7/365 and is strongest in winter and weakest during the summer. As mentioned above, the easterly phase tends to result in a weaker polar vortex. given we have an established easterly QBO within the stratosphere (which likely has not yet peaked), this could make it difficult for the polar vortex to become established and strong during the winter. This would increase the likelihood for polar vortex displacement and potentially a greater frequency of Arctic intrusions of colder air into the northern United States. 

Modoki EL Nino 

    Before we wrap up this winter outlook, there are a few things to discuss which will be done briefly as this is already very long. In terms of EL Nino, structure, and evolution one thing to watch is whether this EL Nino transitions into a modoki EL Nino. Typically, EL Nino events evolve within ENSO regions 1.2 (reference Figure 1) and build west towards the central equatorial Pacific. A modoki EL Nino is one in which the warm anomalies occur within the central equatorial Pacific and are flanked by cooler water temperatures on east side of the warm anomalies. Modoki EL Nino events have seem to become increasingly common since the 1980's and it is extremely interesting as this is about the time we've seen a significant flip in how EL Nino seems to impact the atmospheric pattern. Table 20 provides a list of modoki EL Nino Winter's. This does not include any transition winters.

Table 20

    There are other considerations which should be addressed as well such as solar activity and volcanic activity which can have an impact on how the global circulation evolves during the winter and have impacts on the stratosphere and polar vortex. My knowledge in this regard is quite limited and as such, should these factors play a significant role my forecast would be wrong (even if it appeared I was fairly close). 

2023-2024 Winter Outlook

    Ideally, I would like to do a month-to-month breakdown as well, however, time is becoming of the essence and I will never be able to complete this outlook. So I will just focus on the winter as a whole. Essentially, I think there will be increased likelihood for intrusions of Arctic air east of the Rocky Mountains with an active jet stream across the southern United States and up along the East Coast. What has been a remarkably wet summer in the Northeast United States will likely persist through the winter and this should result in increased potential for wintry events. 

    While many winters over the past 10-15 years have been dominated by the atmospheric configuration across the Pacific, this winter will likely be made by how the configuration evolves across the Arctic. If blocking patterns really struggle to materialize this winter, this forecast would be garbage and we would see the likelihood of above-average temperatures for much of the country, however, the extreme southern states, coastal mid-Atlantic, and coastal Northeast would likely remain active with above-average precipitation. 

Below, is what I am thinking for winter as a whole. Figure 58 provides temperature forecast for December-March and areas which are most likely to feature above-average temperatures, near-average temperatures, and below-average temperatures for the winter as a whole. Figure 59 provides precipitation forecasts for December-March and areas which are most likely to feature above-average precipitation and below-average precipitation. 

EL Nino Peak Strength: 

I am expecting the EL Nino to peak as a strong event per the ONI, however, based on the SOI I would anticipate this EL Nino acts more like a moderate EL Nino...and perhaps even a lower end moderate unless we see some significant reflections within the SOI. 

Figure 58

Figure 59










1 comment: