FREE NEWSLETTER: Econintersect sends a nightly newsletter highlighting news events of the day, and providing a summary of new articles posted on the website. Econintersect will not sell or pass your email address to others per our privacy policy. You can cancel this subscription at any time by selecting the unsubscribing link in the footer of each email.

Page II: Weather and Climate Report - Last updated Novermber 26, 2017

Written by

This Page contains expanded versions of graphics that auto-update plus scientific information that was first published in a weekly version of the GEI Weather and Climate Report.

In general, text information that updates is not updated on this Page but can be found by looking at the last five weeks of the weekly version of the Weather and Climate Report and the index to those is

Click here for Historical Weekly Weather Post Listing

The cycle through the month is that the Monday Report after the Third Thursday contains the analysis of the NOAA Seasonal Update. the Monday after the last day of the month includes the analysis of the NOAA update of the early version of the following month. Australia reports every two weeks so their update report is either in the current week's edition of the GEI Weather and Climate Report on the previouis weeks edition. On weeks when NOAA is not releasing major new updates, there is often a science theme in the GEI Weather and Climate Report for that week. 

We are going to this new approach as trying to maintain this page up to date on information that changes but does not auto-update has become very difficult. So look here for information that auto-updates and scientific discussions which appeared in prior editions. For current discussions of seasonal outlooks etc, please refer to the five most recent GEI Reports and you will find that information there.


A. Worldwide Weather: Current and Three-Month Outlooks: 15 Month Outlooks

B. Factors Impacting the Outlook

1. Very High Frequency (short-term) Cycles

2. Medium Frequency Cycles such as ENSO and IOD

3. Low Frequency Cycles such as PDO, AMO, IOBD, EATS.

C. Computer Models and Methodologies

D. Reserved for a Future Topic

Note: Graphics in this report are auto-updated by NOAA (National Oceanic and Atmospheric Administration and the various weather organizations within NOAA as it is a complex organization) and other sources.  The commentary from Econointersect is updated once or twice a week and may lag the graphics updates.

A. Worldwide Weather: Current and Three-Month Outlooks: 15 Month Outlooks

U.S. Weather

Current Through Seven Day U.S. Weather Maps.

Let's start by looking at the enhance infrared satellite imagery.  This graphic updates every six hours. The time used is ZULU which in some places is called Greenwich Mean Time (GMT) or UTC.  Here is a ZULU to normal time converter. Where I live I just subtract six hours (not worrying about the one hour that Daylight Savings or Standard Time changes that formula) which means that 10 pm my time shows up as 4 am ZULU with tomorrow's date so one has to be careful in how to interpret the time stamp.

IR Western Sector

Let's take a look at the weather fronts. Here is a national animation of weather front and precipitation forecasts with four six hour updates and a second day of two 12 hour updates the second of which is intended to provide coverage out to 60 hours. Beyond 48 hours, additional maps are available at the link provided above.

current highs and lows

The explanation for the coding used in these maps, i.e. the full legend, can be found here although it includes some symbols that are no longer shown in the graphic because they are implement by color coding.

And here is pretty much the same information with interpretation and a focus on tropical storms.  It does not cover as wide an area e.g. it does not cover the Western Pacific or the Atlantic far east of the U.S.  

Eastern Pacific Tropical Storms

There is always the possibility that the outer circulation can enter the circulation of a High or Low over CONUS and provide moisture to that feature. This is a useful diagram attributable to Kelvinsong.

Hurricane structure

Here is a broader view of projected tropical hazards and benefits over an approximately two week period. There are two views. One is more focused on the Pacific and one that includes the Indian Ocean and covers Asia more completely. Both graphics auto-update on Fridays. More information can be found here. The discussion at that link there may update on Tuesdays. I am not sure. Looking at the first graphic, there is no indication of sustained Monsoonal activity in the U.S. The wet area is the Gulf of Mexico. As I have indicated, we seem to be cycling through wet and dry periods caused by tropical storms off the coast of Mexico which are in some cases providing a flow of moist air into CONUS which impacts the Southeast and sometimes a much wider area. 

Tropical Hazards

This graphic covers a larger part of the world. You see more activity over by Asia. 

Tropical Hazards

Here is a better look at the Western Pacific.

Western Pacific Tropical Activity

Sometimes it is useful to take a look at the location of the Jet Stream or Jet Streams.

Moisture imagery is also a very useful thing to monitor as generally speaking you can not have precipitaiton without moisture and pretty much you needs clouds present or soon to be formed.

Water Vapor Imagery

Here is a Navy version

GOES15 Southwet

Not all moisture converts to precipitation. Based on my Weather Modification research, for the Rocky Mountain States the generally accepted numbers are that about 20% of moisture forms clouds and about 30% of the water in clouds falls as precipitation. So (0.2)X(0.3) is 0.06 or about 6% of the moisture passing over these states ends up as precipitation. I have no idea what the numbers are for states that do not have mountains. One thing is for sure, there is usually a lot of moisture in the atmosphere and that is not going to change due to Global Warming or at least it is not going to decline.

It is useful to understand what is going on with the jet stream.

Current Jet Stream

And sometimes the Jet Stream forecast is revealing. This is the forecast out five days.

Jet Stream Five Days Out

To see it in animation, click here. At the time this article was published the animation shows a tendency for there to be a Southern Branch of the Polar Jet Stream in addition to the usual Northern Branch. It can bring storms further south than usual.

This longer animation shows how the jet stream is crossing the Pacific and when it reaches the U.S. West Coast is going every which way.

Here is a very flexible computer graphic. You can adjust what is being displayed by clicking on "earth" adjusting the parameters and then clicking again on "earth" to remove the menu. Right now it is set up to show the 500 hPa wind patterns which is the main way of looking at synoptic weather patterns. .

U.S.  3 Day to  7 Day Forecasts

You can enlarge the below daily (days 3 - 7) weather maps by clicking here.

Or individually and thus larger by clicking Three Day or  Four Day or Five Day or Six Day or Seven Day 

Short term forecasts

And below is a another view which highlights the highs and the lows re air pressure which in theory it should be the same as the Day-3 day or  6-day map above but it may not be identical because it may have been issued slightly earlier or later than the above maps and may have been prepared by a different meteorological team. It should be pretty close and is easy to read.  

Here is the forecasted pattern of Highs and Lows three days out.

Day 3 highs and lows

And here is six days out:

Day 6 Weather Forecast

This next map is the mid-atmosphere 7-Day chart rather than the surface highs and lows and weather features.  In some cases it provides a clearer less confusing picture as it shows the major pressure gradients.

7 Day 500 MB Geopotential Forecast

Because "Thickness Lines" are shown by those green lines on this graphic it is a good place to define "Thickness" and its uses. You can find a full style explanation (thorough) at that link or just remember that Thickness measures the virtual temperature plus moisture content) of the lower atmosphere and is very useful especially in the winter at identifying areas prone to snow and in the summer areas which are going to be hot and humid. Here is a U.S. style explanation of "Thickness" by Jeff Haby who is a valuable source of Haby Hints for anyone who wants an explanation of a meteorological term. The thickness lines do not yet indicate winter conditions which might be thickness levels below 540 for most areas. The above link is is an explanation for the U.K. The levels for CONUS might be slightly different. Obviously these thickness lines do not tell you about mountain peaks. The particular definition of "thickness" on this graphic may not be the best way to define the snow line and this is discussed in the link provided but it is I believe the older method and gives a first approximation which can be further refined as per the discussion in the link. This graphic indicates normal west to east movement of air masses for the most part except along the coasts.

The following table is useful but designed for Europe. I was not able to find the corresponding information for the U.S. but it would not be drastically different. In the U.S., the last zero is dropped off the Thickness Level so where it says 5640 that would show as 564 in the U.S. Also remember the temperatures shown are in Centigrade. So 580 would correspond to about 80F in full sunshine during the summer. That is why I say the 582 and 576 thickness levels are a bit unusual for mid to late October. 

Thickness Table

(6 - 10 and 8 to 14 day) Forecasts

Note: This is a forecast for six days from when it is issued.

When I provide comments, they apply to the forecast issued on the most recent Monday. Please note these maps auto update daily and on weekends they are generally issued without review or discussion by the NOAA meteorological team.   

1.  Temperature Forecast for days 6 - 10:


2.  6 - 10 Day Precipitation Forecast:

precipitation 610

And now what they call Week Two which is days 8 - 14. Notice the overlap of days 8, 9, and 10. This reflects the difficulty of precisely timing the movement of highs (ridges) and lows (troughs) which is the primary mechanism for the development of these forecasts especially during the winter when these patterns tend to move from west to east.

3.  The 8 - 14 Day (Second Week) Temperature Forecast:

Temperature forecast 814

4.  The 8- 14 Day Precipitation Forecast:

Precipitation Forecast 814

In theory the week one forecast each week should be what was forecast as week two the prior week. You can track the ability of NOAA to predict things by looking at the prior week's forecast.

At this point I think it might be useful to mention the different procedures used by NOAA to make the 6 - 14 day Outlooks and the Monthly, Seasonal and Long-Term Outlooks. The shorter the outlook, the more NOAA depends on current weather patterns and the models basically extrapolate out the current weather patterns and to a large extent rely on the analogs that I discuss below. High frequency cycles such as the MJO, PNA, OA, NAO are a factor and you can learn more about them elsewhere in this Report. When similar weather patterns occurred in the past, what did the next 6 - 10 days look like? That is the basis for using analogs and the driver of weather forecasting models.  So the short-term Outlooks do not depend very much on assumptions about medium- and low- frequency (slow changing) cycles such as ENSO and the Pacific i.e. the phase of the PDO. To a large extent NOAA ignores the AMO in their forecasting.  Those medium- and low-frequency cycles might be reflected in the analogs and I try to tease them out each Monday in my analysis of the Analogs issued with the the 6 -10 Day Outlook for that Monday.

The further out you go, the less the Outlook depends on current weather patterns and the more it depends on medium- and low frequency-cycles and statistical trends. So the preliminary Outlook for the following month issued with the seasonal outlook may actually use a different methodology in its preparation than the updated Outlook for the following month issued on the last day of the preceding month because the updated Outlook for the following month will have 14 days of the Outlook created by a more reliable method and is less dependent on general assumptions. I do not want to make too much about this as often there is not much difference between the preliminary Outlook for the following month and the updated Outlook for the following month issued on the last day of the month. But sometimes it changes and the above explanation is part of the reason for that. It is a sensible approach.

Current (or upcoming) Month Forecast.

Current and three month outlook

Long Range Forecasts


Long range temperature outlooks


Long range Precipitation Outlook

If you want larger versions of each map (temperature and precipitation) you can find them here. And each of those maps can be clicked on to further enlarge them.
The analysis can be found in the weekly Report of the GEI Weather and Climate Report folloing the updating of these map which can be found here.

These folks look at this from a very detailed perspective.

Recent El Nino Impacts

Weather Pattern Impacts (very likely related to El Nino)

Recent Impacts of Weather Mostly El Nino but possibly Also PDO and AMO Impacts.

Below are snapshots of 30 Day temperature and precipitation departures over the life of this El Nino. The end date of the 30 day period is shown in the graphic.  It is a way of seeing how the impacts of this El Nino of unfolded.

June 15, 2015 30 Day Temperature and Precipitation Departures.

July 13, 2015 30 Day Temperature and Preciptiation Departures

August 10 2015 30 Day Temperature and Precipitation Departures

Sept 5, 2015 30 Day Temperature and Precipitation Departures

Oct 3, 2015 30 day Temperature and Precipitation Departures

30 day Temperature and Precipitation Departures

30 Day Temperature and Precipitation Departures as of November 14

Dec 14, 2015 30 temperature and precipitation departures

30 Day Temperature and Precipitation Departures as of November 14

Dec 21, 2015 30 day temperature and precipitation departures.

Dec 21, 2015 30 day temperature and precipitation departures.

January 4, 2016 30 Day Temperature and Precipitation Departures

January 4, 2016 30 Day Temperature and Precipitation Departures

Jan 16, 2016 30 day Temp and Precip Departures.

Recent Impacts of Weather Mostly El Nino but possibly Also PDO and AMO Impacts.

Below are snapshots of 30 Day temperature and precipitation departures over the life of this El Nino. The end date of the 30 day period is shown in the graphic.  It is a way of seeing how the impacts of this El Nino have unfolded.

June 15, 2015 30 Day Temperature and Precipitation Departures.

July 13, 2015 30 Day Temperature and Preciptiation Departures

August 10 2015 30 Day Temperature and Precipitation Departures

Sept 5, 2015 30 Day Temperature and Precipitation Departures

Oct 3, 2015 30 day Temperature and Precipitation Departures

30 day Temperature and Precipitation Departures

November 30, 2015 30 day temperature and precipitation departures.

January 4, 2016 30 Day Temperature and Precipitation Departures

February 1, 2016 30 Day Temperature and Weather Departures.

Feb  29, 2016 temperature and Precipitation Departures.

March 7, 2016 30 Day Temperature and Precipitation Departures

April  2016 20 day temperature and precipitation departures.

Remember this is a 30 day average and last week I used a different graphic so this can not be compared to last week but is best compared with last month. The La Nina pattern persists for much of the West with respect to both precipitation and temperature but is a normal El Nino for the Mississippi Valley in March. Northern California was wet but it is hard to say if that looks like El Nino or La Nina. This is one strange El Nino and for the 2nd or 3rd strongest in modern history it is a mystery that has not been given adequate attention. 

Lets take a look at the combined results for the first three months of 2016: January, February and March.

April 4, 2016 90 Day Temperature and Precipitation Departures.

Well that does not look like an El Nino pattern to me but more like a La Nina pattern for precipitation and just plain warm pretty much everywhere which is neither an El Nino nor a La Nina Pattern.

And here is the April (30 day) graphic.

May 2, 2016 30 Day Temperature and Weather Departures

We have seen a gradual change in April to a more typical El Nino pattern. The Northwest is dry and the Southwest is a bit wetter than normal. One area along the Southern California western Arizona border had very good  El Nino precipitation. The lee side of the Rockies for some reason were wet all the way to Canada and probably into Canada but not shown in this graphic. It certainly has remained dry in Mexico. The Temperature Pattern has been very close to a typical El Nino pattern in April.

I realize this is a lot of graphics but one needs to look at the history of an event to assess it. As you can see, so far we are not having the expected El Nino Impacts in CONUS.

Here is the ninety day picture for February, March, and April.

May 2, 2016, 90 Day Temperature and Precipitation Departures.

The precipitation pattern still looks more like a La Nina than an El Nino. It has been warmer than climatology everywhere.

And here is the graphic from last week which added a week and removed the seven earliest days so it is a 30 day analysis through May 7, 2016

May 9, 2016 30 Day Temperature and Precipitation Departures.

It looked a lot more like a traditional El Nino pattern the last 30 days. You see that in both the precipitation and temperature.

May 16, 2016 30 Day Temperature and Precipitation Departures

But less so with one week added. The dry belt south of the Southwest has moved a bit north. We do not hear a lot about Mexico but they must be hurting.

And here is the latest 30 Day Temperature and Precipitation Departure (same as anomalies) graphic.

May 23 30 Day Temperature and Precipitation Departures



Here is what May looked like:

May 30, 2016 30 Day Tempeature and Precipitation Departures

The final days of this El Nino behaved like an El Nino. Quite interesting.

But looking at a longer time period in this 90 days or approximately three months.  

May 30, 2016 90 Day Tempeature and Precipitation Departures

Looking at the three months (March - May), it certainly at least with respect to precipitation was more like a La Nina event than an El Nino event. Except for Texas. Northern California caught up and Northeast Mexico also. But Arizona, New Mexico and Southern California did not participate in this El Nino in 2016 although they did in the Fall of 2015. Variability is the norm

And then we started to track June.

Here is the 30 day period through June 25.

June 27, 2016, 30 Day Temperature and Precipitation Departures

It has sure been dry. I has also sure been warm.

Comparison with the Three Strongest El Ninos

I thought this comparison of the three strongest El Nino's since 1950 with the current El Nino would be interesting. The red numbers are El Nino values; the blue are La Nina Values. I am just show the ONI values for the three strongest El Ninos and the current one. 

comparison of 72/82/97 and current El Nino

Obviously we do not know how this one will play out but I wanted to see if there were any clues as to which of the other three Super-El Ninos this one might most resemble.

What I see is that this one had an ONI of 0.5 or higher two months earlier than the 1997/1998 El Nino. But it has strengthened a lot faster than the 1982/1983 El Nino. Notice the 1972/1973 El Nino peaked in OND as this one most likely will also. But the PDO stayed negative during the entire duration of the 1972/1973 El Nino and the AMO also was negative. The 1982/1983 El Nino occurred during PDO+/AMO- conditions which are very favorable for the development of a traditional El Nino. The 1997/1998 El Nino occurred during PDO+ and AMO+ conditions but it was true PDO+ and not induced by the El Nino which most likely is the case with this El Nino. Also the AMO was far more positive at that time and now is close to neutral but may be reading positive during recent weeks.

So I conclude that we do not have a good model. The NOAA analogs are totally inconsistent with regards to suggesting that current conditions are reminiscent of past conditions. So we have little to go on. And yet NOAA and JAMSTEC are certain they know what the impacts will be. It is good to have confidence even if your basis for that confidence may make no sense.

Perhaps I should have shown 2014 also as it was close be recording as an El Nino which was not the case for the other three which were not immediately preceded by what was almost an El Nino. That is another reason why I have basically zero faith in the NOAA and JAMSTEC forecasts. To me this is uncharted territory. 

A helpful person provided me with this graphic. The maps are (from left to right and row to row downward) in declining strength of the El Nino.

Recent El Ninos

You can get a larger image here. You probably also get a larger image by just clicking on the graphic. It may required doing that twice.

I will now attempt to dissect this set of data a bit. My choice of the three areas to categorize was arbitrary. The analysis could be done state by state and actually in some cases the impact was in a part of a state. I picked three areas that are often discussed re the impact of an El Nino. I could have added Arizona and I could have added the Northwest above California and other parts of CONUS as well. Where I thought the impact of a particular El Nino was small or highly variable within the regional category, I did not include it in that category.

The breakdown of El Ninos into Traditional and El Nino is based on the data I have obtained from various articles so there can be some debate whether I have gotten it correct for every single El Nino. The 1953/1954 El Nino is not even recorded as an El Nino by the Japanese but I put it into the Traditional Column for no good reason, but I had to put it somewhere. I wanted to see if the condition of the two major ocean cycles was important. The condition of the PDO and AMO in the table below is my interpretation as the Index varies from month to month and thus is not always of the same sign throughout every El Nino, so it is my interpretation of the predominant condition of the PDO and AMO for each particular El Nino.

This is a first attempt at this sort of analysis so it is not professional paper grade and has not been peer reviewed. I am open to all comments on the analysis.  The maps are above so anyone can do their own analysis.

Okay, here is what I came up with.


   Traditional El Ninos

PDO/ AMO           El Nino Modoki PDO/ AMO       
Events where California was wet






+    +

+    -

-     -

-    +

N    -








+    -

N   +

+   N

-    -

-    -

N   -

+   +

Events where California was dry




+    -

-    +

+    -






-    -

+   -

+   -

-    -

-   +

Events where the Southeast was wet




+    +

+    -

-     -








-   -

N  +

+  N

+   -

-   -

N   -

-   +

Events where the Southeast was dry





+    -

-    +

+    -

+    -




+    -

-     -

+   +

Events where New Mexico was wet



+   +

+   +





+    -

N   +

+    -

+   +

Events where New Mexico was dry





-    +

+    -

-    +

N   -




+   -

-   -

-   -


And then I tabulated the above. I have highlight in large bold type those data elements I considered most significant.

  No of Traditional Events Number  PDO + or Neutral Number AMO - or Neutral

No of Modoki Events

Number PDO + Neutral Number AMO - or Neutral
California Wet


3 3


5 5
California Dry


2 2




Southeast Wet 3 2 2


4 5
Southeast Dry 4 3 3


3 2
NM Wet


2 0 4


NM Dry




3 1


Here are my conclusions:

1. El Nino is significantly overrated as being useful in predicting winter precipitation patterns for CONUS other than Traditional El Ninos for California. For California, Modokis where the PDO is not Positive or Neutral have a slight dry bias.
2. For the Southeast, Modokis are more likely to be wet than Traditional El Ninos.
3. For New Mexico, only strong Traditional El Ninos are wet; the weak ones are dry.  El Ninos that make New Mexico dry tend to be associated with AMO - conditions which to me was a surprise.
4. Further analysis might reveal additional patterns. It may be that the condition of the AMO and PDO are really what is most significant.

How about La Nina?

Importance of the Strength of the La Nina

In many ways the ENSO Cycle has impacts like the PDO Cycle. So to some extent the intensity of a La Nina determines the impacts. That is why it is important to think about how intense this La Nina might be. Here is a graphic that shows the impacts of recent intense La Nina's on CONUS.

La Nina Impacts used in May 9, 2016 Report.

It is useful to look at the Phases of the PDO and AMO during these three months of each of the these major La Ninas.

Phase of Pacific and Atlantic Cycles Dec to Feb
1988 - 1989 - -
1998 - 1999 - +
1999 - 2000 - -
2007 - 2008 - +
2010 - 2011 - +


Curiously or not so curiously all five of these extreme La Nina's occurred with the PDO in a Negative State. It seems like the PDO will be Positive this coming winter. So it is very likely that the impacts of a La Nina would be significantly less than what is shown here but with the same direction i.e. the dry areas would be dry but less so and the wet areas would be wet but less so.
The Monitoring of the PDO thus becomes critical re forecasting the Winter of 2016/2017

Let us not Forget the Atlantic.

I call your attention to the Atlantic and the warm area called the MDR or Main Development Area. 

Expected Hurricane Season

It has not worked out as forecast as the MDR is above average SSTs not below average SST's.

Daily SST Anomaly

In some fields when certain assumptions or forecasts fail to materialize, one begins to question the whole forecast. Apparently NOAA does not work that way. I think it should.


Compariso with 1997/1998 El Nino

And I think it is useful to take a look at the 1997 second half of the year impacts of the 1997/1998 Super El Nino.

1997 Precipitation Impacts of 1997/1998 El Nino

Notice the impacts on CONUS precipitation might have been different than one might have expected. This graphic and the OLR graphic are very related. This measures precipitation the OLR measures clouds which are fairly well correlated.

El Nino in the News

El Nino may impact Brazil. This is one of many articles on this topic. Brazil is a large nation with different climate zones and a complex pattern of where crops are grown. So the exact impacts by crop are difficult to predict.

2 typhoons over Pacific: 1 could strike Japan, clip Philippines, Taiwan: CNN read more here

2015 US Fall Forecast: Fire Danger to Worsen in West; September May Yield Tropical Impact in South  Accuweather  here  [Editors Note: Accurate but Quite an Interesting Twist]

The major impact of El Nno so far has been on Indonesia and here is a distressing article on that.

Asia-Pacific Region: El Niño Snapshot - August 2015  more here.

I found this map from that article which was produced by the UN to be quite interesting. It illustrates how in some cases the impacts are organized by layers that relate to the distance from the Equator. Every El Nino is different and I am not sure if the Phillippines are suffering from drought right now. 

Asia - Pacific Forecast from UN

Forecast of a 2015 “Godzilla El Niño” Threatens the Global Coffee Community more here.

El Niño Could Be Among Strongest on Record, Raising Risk of Floods Read more here.

As El Nino looms, cloud seeding gets tailwind. Read more here.

This article on Arctic Ice Melt is interesting but to me the graphic is even more interesting.

Alaska Climate History used in May 23, 2016 Report.

It shows the impact of the unusual strength and positioning of the Aleutian Low. looking at the 1972/1973 El Nino and the 1982/1982 El Nino and the 1997/1998 El Nino one sees some very interesting differences. I do not see much of a trend in the high temperature peaks with this winter being an outlier and probably due to the position and strength of the Aleutian Low. One may see a trend in the low temperatures but this may be related to the PDO.

India: first drought followed by strong Monsoon.

Heat makes precipitation.

Adaptation to El Nino in Indonesia  Not a new paper but very interesting.

As was previously suggested here, the near El Nino Modoki of 2014/2015 gave this El Nino a Head Start.

Full Paper

The Fort McMurray Disaster

2016 Weather Disasters (to date)

The Hot and Wet in India article might fit in this category. It is hard to separate out Global Warming from normal climate cycles unless you are a  Climate Change Alarmist in which case it is all very clear to you or a CC Denier in which case it also is totally clear to you. Since I do not fit in either category I am left with the very difficult task of trying to sort it out. It is a bit like trying to sort out every medical condition from aging. We know that aging increases the odds of adverse medical conditions. But not all adverse medical conditions are attributable to age. This can become a problem what a doctor tells you are too young to have the symptoms you report or alternatively one dismisses certain symptoms as being simply age catching up with you when there is something not age related or only marginally age related going on. It is better to look at the evidence than start with a predetermined answer.

How Weather Shocks can Cause Severe Food  Impacts  I have not read this article yet  but it looks very interesting. 

Central U.S. Dousing this past weekend.

Oklahoma City Rainfall Record Broken

Houston Texas Flooding

Global Warming in the News

The first article in the El Nino News Section could also fall in this category.

Hot and Wet in India

Attitudes Towards Warming Assessed







Worldwide Precipitation Outlook

The Australian Queensland Bureau of Meteorology uses the SOI to make a worldwide precipitation forecast and here it is. It is interesting because it is based on just one variable: the air pressure differential between Tahiti and Darwin Australia.


Queensland Australia analysis of impact of consistantly low SOI values.

You can read about it here  and find updated maps here as they do not autoupdate. I have not researched the skill level of this tool and the SOI is mostly a tropical index, but I include it because it is interesting.

Here are the low-level wind anomalies. This graphic is not as compact as the graphic provided by the weekly NOAA ENSO Report (more white space) but this version auto-updates so you will always have the latest version of this Hovmoeller. As you can see, the wind gust of a several weeks ago at 160E is over and a subsequent less intense wind gust at 160W to 140W has also played out. There may be some activity starting in the Western Pacific that is just starting to show up on this graphic. Take a look over at 130E to 140E which by chance happens to be Darwin Australia one pole of the SOI Index. I think this is too far west to impact the current El Nino. We can also see some activity at 160E which is more significant and needs to be monitored.

Low Level Wlind Anomalies

Here is another graphic that is less compact than the prettied up version published by NOAA on Mondays but which has the advantage of auto-updating. You can see how the convection pattern (really cloud tops has in May shifted to the East from a Date Line (180) Modoki pattern to a 170W to 120W Traditional/Canonical El Nino Pattern. But the signs of an El Nino are getting quite faint and shifting to the west.

OLR Anomalies Along the Equator

Let us now take a look at the progress of the Kevin wave which is the key to the situation.

And here is one more way of looking at the situation. I like this Hovmoeller a lot and I have now been able to find a version that autoupdates but is not prettied up.  I will take the auto-update feature.   You can see the Kelvin Wave that got started in February which started this Warm Event. There have been earlier such events that proved to be not very strong. But if you look at the bottom of the Hovmoeller which represents the current situation, you can see that this latest Kelvin Wave is moving to the East fairly rapidly and we will see the impact of that on declining ONI estimates fairly soon. The main impact of this Kelvin Wave is already East of 170 West the westernmost extension of the Nino 3.4 region.  It has moved throught 30 degrees of longitude in about six weeks so perhaps next week in my weekly report I will attempt to projects its progression in the months ahead.

Kelvin Waves Auto-updates

You can see below in the graphic which shows temperature along the Equator as a function of depth, both the magnitude of the anomalies and their size. You can now see where 2C (anomaly) water is impacting the area where the ONI is measured i.e. 170W to 120W. The 2C anomaly now extends to about 140W and the blips visible further to the west are no longer evident. The subsurface warm water appears to be making its way to the surface to some extent.  This will be apparent when we discuss the TAO/TRITON graphic and my crude estimation of the ONI value that that I develop from that graphic.

The big issue is where will the +6C anomaly water go as it reaches the beaches of Ecuador? To the extent it surfaces it can create convection and impact the Walker Circulation which could then provide positive feedback to this El Nino. But that warm water might tend to go north or south or both. That is part of the phase out process for an El Nino and that is where we are in the life of this El Nino. It is peaking and will soon begin its decline.

Subsurface Heat Anomalies

The bottom half of the graphic is not that useful in terms of tracking the progress of this Warm Event as it simply shows the thermocline between warm and cool water which pretty much looks like this as shown here during a Warm Event except now the cooler water is not making it to the surface to the east along the coast of Equator. 

In the upper graphic, notice the boundary of the 1.5C plus water temperature anomaly (which is now the 1.0C plus water temperature anomaly) is now close to 170W and moving towards the East. That is why I believe the ONI will soon peak and begin to decline. We shall see. There could be another Kelvin Wave forming and there is the issue of how the Walker Circulation might extend the life of this Warm Event. In this regard you might want to read the following post. A key graphic from that post which is a standard graphic is below.

Walker Circulation

The problem with the above graphic is that it represents the ideal case. You can see the uniform nature of the cells/loops which are areas of rising air and convection (precipitation) and areas where the warm air subsides. Where the air is rising that is an area of low pressure and where the air is subsiding those are areas of high pressure. Things always have to equalize. But the pattern is not always as shown. I believe that in the 1997 El Nino there was a giant Kelvin Circulation that extended from the Pacific to the Indian Ocean not the two cells shown above. So the general case and the variations are important. That is why it is important that we differentiate between Modokis and Traditional El Ninos which have different Walker circulations. This is a good article on the 1997 Super El Nino. I am not in any way suggesting that this El Nino is at all like the 1997/98 El Nino but simply pointing out that the details of each Warm Event are different and that we can not just group them all together and expect to make sense of things.

For my own amusement, I thought I would recalculate the ONI again as I have been doing recently. To refine my calculation I have totally changed my approach and rather than having the anomalies be the way I organized the data, I have divided the 170W to 120W ONI measuring area into five subregions and have mentally integrated what I see below and recorded that in the table I have constructed. Then I take the average of the anomalies I estimated for each of the five subregions. So now I am applying more subjectivity but it should produce a better estimate. Notice the boundary of the 1.5C plus water anomaly is now close to 170W and moving towards the East. That is why I believe the ONI will soon peak and begin to decline. But all the Meteorological Agencies in the World believe the ONI will continue higher. Could I possibly be correct? We shall see. It is not clear the extent to which the exact value of the ONI is useful in forecasting weather. There is a correlation between the strength of a Warm or Cool Event and the weather impacts. But it is not clear that the ONI or any other single index is able to be the independent variable in the forecasting model.

Current SST and wind anomalies

Here is another way of looking at it:

.SST Anomalies Hovmoeller

This Hovmoeller shows a lot of useful information. I could copy it into MSPaint and draw some lines on it but then it would not autoupdate so I do not wish to do that. But take a lot at 140E 160E, 165E, 175E, 120W and 90W. Remember reading from top to bottom one is reading the earlier times to the more current times. So you can see how this Warm Event started at 140E, has moved to 160E and then to 165E and lately you can see continued movement towards 175E, which it has now reached, but very slowly. You can also see that the entire Equator is warm. You can especially see the impact east of 90W where the Kelvin Wave is crashing into Ecuador. Also more warmer water has expanded towards 120 W. The eastern progress has been slower than I had anticipated. Leaving aside the SOI issue which until this week was no longer consistent with an El Nino, but has come to life perhaps just temporarily, this is clearly an El Nino type sea-surface temperature (SST) pattern right now. But to me it seems to be a pattern that will play out as it does not appear to be going to be reinforced. It may well play out more slowly that I have anticipated. But it is playing out.

You will not see the ONI decline until the warm water over at 175E has moved to 170W. Until then, the ONI could easily continue to rise but probably not by very much although some models are predicting it will peak at about 2.0. Once the warm surface water no longer extends west of 170W the ONI should begin to decline. I hesitate to offer a new estimate of when that will happen as I have not had the time to play those animations enough times to make an estimate. Plus it depends on the way the subsurface pod of warm water behaves and I do not know enough to predict how that will work its way through. The size of the deep water anomaly may have a bearing on the rate of movement to the east which is slower than what I read in the literature as being more usual. It may also impact where it rises. It has to go somewhere or mix or come to the surface and experience evaporative cooling. The MJO now appears to have some impact but probably small.  We will watch the progress of this Warm Event together because I have provided the graphics that will allow us to do that.   

I thought I would show what things look like this year compared with the most recent very powerful El Nino namely the 1996-1997-1998 period.  Notice the SOI went very negative early in 1997. That is why I have concluded that if there is to be an El Nino it is likely to be a 2015/2016 El Nino not the 2014/1015 El Nino that NOAA has been advertising

1907 SOI

And here is the current reporting of the SOI.

2013-2014-2015 SOI

View from Japan as Compared to the View from NOAA.

NOAA issued an updated Seasonal Outlook on October 15. The JAMSTEC (Japanese) analysis is also available. Let us compare them.

First let's take look at the ONI forecasts which is the most widely available measure of the strength of an El Nino (or La Nina) and is the deviation of the sea surface temperature from climatology (normal) in a small part of the Eastern Pacific along the Equator considered to be the best place to assess the strength of an ENSO phase whether it be El Nino, ENSO Neutral, or La Nina.

First the NOAA forecast of conditions in the Nino 3.4 area which defines the ONI Index:

OMG there are two different versions of the model results. There is this one:

ENSO Forecast

And below is another version (and the version below is the one that NOAA shows in their weekly ENSO Report so I am assuming that it is the one that they want us to use) .  I am not sure of the specifics of the difference but the results are modified in some way presumably to make them more accurate. I think it has to do with the base used to establish "Climatology" which is usually defined as conditions from 1981 - 2010 (not sure why it says 1982 - 2010).  But the Equator has been experiencing a warming trend thus a thirty-year average may overstate warm anomalies. So, to me, reading the legend in the graphic below It looks like the base was adjusted to 1999 - 2000 reflecting the steady rise of temperatures in the Pacific. It is possible that there is also an adjustment for the variation in the model results for the multiple members of the ensemble but I do not know how that would work given that they still refer to the mean of the different forecasts of the members of the ensemble. So I can not tell you exactly why the model results below differ from the model results above.

It would be good if NOAA decided which approach they wanted to use and presented only one version of this graphic. But I understand their dilemma.The below graphic is probably the most useful and will agree with what is officially recorded but the above version is more consistent with what the other meteorological agencies report. I prefer the below as it is more realistic.They both show this El Nino peaks in October or November of 2015.

CFSv2 forecast

And here is the JAMSTEC forecast of the ONI.

ONI Forecast Nov 1, 2055

Here is the discussion

Nov. 20, 2015

Prediction from 1st Nov., 2015

ENSO forecast:

The SINTEX-F model predicts that the current strong El Niño will reach its peak in late boreal fall and start to decay after boreal winter. The tropical Pacific will return to a neutral state by 2016 boreal summer and then will turn to a La Niña in following boreal winter.

Indian Ocean forecast:

The model predictions suggest that the positive Indian Ocean Dipole (IOD) will decay in late boreal fall. Subsequently, a basin-wide warming will be seen in boreal winter and spring in response to the tropical Pacific El Niño.

Regional forecast:

In boreal winter, as a seasonally averaged view, most parts of the globe (including Japan) will experience a warmer-than-normal condition, while northern Europe, northeastern Russia, southeastern China, and central/eastern U.S. will experience a colder-than-normal condition. All those may be partly related to the on-going strong El Niño.

According to the seasonally averaged rainfall prediction in boreal winter, eastern Australia, southern Africa, Brazil, and Southeast Asia will experience a drier-than-normal condition, while Europe and U.S (particularly the western and southeastern parts) will experience a wetter-than-normal condition. The strong El Niño and the California Nino could be partly responsible for those. The SINTEX-F model predicts that Japan will experience a wetter and warmer-than-normal condition in boreal winter.

But then there was this bonus discussion which is well worth reading. They do some bragging about the forecast model but it may be well deserved. 

Now I will compare the two forecasts using three time periods: Short Term, Medium Range, and Long Term 

These maps present anomalies or deviations from the usual climatology. So if I forget to say warmer than climatology etc, that is just me not being as careful with terminology as I should be. Everything in the following discussion is related to climatology. So wet means wetter than climatology and dry means dryer than climatology etc. 

SHORT TERM (December - January - February) JAMSTEC did not publish NDJ information so I am working with DJF. So we are not comparing the forecasts for November. .

Sea Surface Temperatures.

Starting with the NOAA results (and notice that they show Africa twice which can be confusing but allows Africa to be more easily considered in the context of the oceans that surround it):

Short Range SST Anomalies NOAA

And now the JAMSTEC Sea Surface Temperature Anomaly.

DJF SST JAMSTEC from October 1, 2015

They look fairly similar if you can get by the garish colors used by the NOAA graphic artists which certainly emphasizes the warm water off the coast of Mexico that is creating these storms off the west coast of Mexico.  I find the JAMSTEC graphics easier to look at. I am curious about the cold anomaly that NOAA shows for the Equator in the Atlantic. There also appears to be a difference with respect to the waters off of Japan with JAMSTEC projecting cool and NOAA warm. NOAA also projects both an area of cool anomaly in the South Pacific and further south a strong warm anomaly. That is an important difference.

Now let's take a look at the Temperature forecasts for that period starting with NOAA:

Near-term  three month temperature forecast

And then JAMSTEC:

JAMSTEC NDJ 2015 Temperature from October t 1 SST

For CONUS, JAMSTEC is forecasting that the colder than climatology area will cover a significantly larger area. NOAA does not cover the rest of the world. In the JAMSTEC map (and the discussion above which was provided by JAMSTEC) you can see that most of the world is projected to be warmer than climatology with the U.K and Scandinavia, Eastern Siberia, the Highlands of Southwest China and a small part of Argentina being the few exceptions. Given that there is Global Warming and the forecasted values are being compared to a base that may not fully reflect the warming which has occurred, any cold anomalies are especially significant.

Now let us look at Precipitation again starting with the NOAA map.

SON 2015 NOAA Precipitation Outlook Issued on Aug 20, 2015

And here is the JAMSTEC Precipitation Map.

NDJ 2015 Precipitation Forecast based on September 1 Model Runs

The NOAA forecast and JAMSTEC forecast are pretty similar for that period with regards to CONUS but JAMSTEC extends the wet area into the Northwest but the outlook for Alaska differs between the two forecasts. Interesting features of the JAMSTEC map outside of CONUS include dry conditions in Central and Northern Brazil, Central America, parts of Canada, parts of Australia and New Zealand, Iran, Afghanistan, South Africa and the Mediterranean. The dry area for India is no longer shown.

MIDRANGE (March - April - May)

Starting with Sea Surface Temperature Anomalies and beginning with NOAA:

Mid Range  NOAA SST Anomaly Forecast

And then JAMSTEC:

JAMSTEC MAM 2015/2015 SST from October 1, 2015

JAMSTEC is still showing El Nino into 2016. But it is looking a bit more Modoki-ish in terms of the severance of the connection of the warm water with the South American Coastline. Notice the less intense coloration off the Coast of Ecuador and Peru in the JAMSTEC map. But the AMO is more Negative and the PDO is more clearly defined in the JAMSTEC map. They seem to have different opinions on the ocean temperature anomalies off of the coast of West Africa but NOAA has greatly reduced their cool anomaly off of West Africa.

Now let us take a look at Temperature and again first NOAA:

Mid-Range NOAA Temperature Outlook

And then JAMSTEC for the same three months

MAM 2015 Temperature Anomaly based on October 1 Model Run

JAMSTEC now is more aggressive at forecasting widespread colder than climatology conditions for CONUS as compared to NOAA. JAMSTEC is showing cold for parts of Europe. Both JAMSTEC and NOAA show Alaska as still being warmer than climatology which suggests the Aleutian Low will be strong and positioned to direct warmer air over Alaska.

Now Precipitation and again staring with NOAA:

Mid-range NOAA Precipitation Forecast

And then JAMSTEC.

MAM 2015/2015Precipitation Forecast based on October 1 2015 Model Run

JAMSTEC is showing more of a west/east split on precipitation for CONUS while NOAA is more north/south. Brazil is still dry, the rest of South America is wet. Southern Africa is still dry. Western Europe is dry while further east it is wet. Indochina is dry.

LONG RANGE (June - July - August)

First NOAA: And at this point in time this is the furthest out they go and their maps through Apr-May-Jun still shows the El Nino in place. 

NOAA Long Range SST Anomaly

And then JAMSTEC:

JAMSTEC Long Range from October 1, 2015

JAMSTEC is well along to ENSO Neutral in their forecast perhaps pausing and resembling an El Nino Modoki Type II on the way. But the warm water off of the West Coast of North America continues so that it is not clear which of the two will have the most impact on North American weather. JAMSTEC is still showing PDO+ even as the pattern shifts to La Nina. That could be signifying a Climate Change in the Pacific to PDO+ and possibly also AMO- but it is too soon to tell.

Now let us look at Temperature and again starting with NOAA

 2016 NOAA Long Range Temperature Outlook Issued on Oct 15, 2015

And then JAMSTEC

JAMSTEC Long Range from October 1 SST

There is a total difference for the pattern in CONUS with NOAA being warm and JAMSTEC cool extending up into parts of Canada. Again it is a warm Alaska. But in the JAMSTEC maps, Europe remains cool as well as parts of Argentina and far Eastern Siberia. China is mixed.

And finally the Precipitation forecast and again starting with NOAA:

NOAA Long Range Precipitation

And then JAMSTEC.

Predicted MAM Long Range Precipitation based on October 1, 2016 Model Run

For CONUS JAMSTEC is still showing a lot of wetter areas especially in the East. Europe is a bit dry. Indonesia is wet rather than dry as was the case earlier as in this time frame the El Nino transforms into a La Nina.


The main conclusions relate to temperature and precipitation. I can only compare the two for CONUS and Alaska, but I can also comment on what JAMSTEC is projecting for the rest of the World which I have done for the three time frames: near-term, medium range, and long term. The ONI forecasts are similar but the translation into air temperature and precipitation for CONUS and Alaska are somewhat different NOAA versus JAMSTEC especially further out in time. A few general conclusions are that:

  • Both NOAA and JAMSTEC forecast a La Nina starting it the Summer of 2016.
  • The JAMSTEC forecast is forecasting a wider deviation from climatology for CONUS than NOAA: Wetter and Cool
  • The JAMSTEC forecast predicts dry conditions for Europe, Australia, Brazil. Outside the U.S. El Nino is mainly a drought event. 

Looking Ahead to the Winter of 2016/2017


Repeating from the NOAA Discussion:


It is not a perfect procedure but I took a look at the ONI values for the Oct - Nov - Dec three-month period starting in 1950. I used an ONI of 0.5 or higher as an indication that it was a warm event and possibly an El Nino although it takes more than one three-month period of an ONI of 0.5 or higher to define an El Nino. Similarly, I took -0.5 or more negative to indicate La Nina conditions during that three-month period. Then I tabulated the number of times that there was a +0.5 or greater in one year followed by a -0.5 or more negative in the following year.

Year ONI Year ONI Year ONI Year ONI Year ONI Year ONI
1950 -0.7 1960     0 1970 -0.9 1980  0.1 1990  0.4 2000 -0.8
1951  0.7 1961 -0.2 1971 -0.9 1981 -0.1 1991  1.2 2001 -0.3
1952  0.2 1962 -0.3 1972  2.0 1982  2.1 1992 -0.1 2002  1.3
1952  0.8 1963  1.2 1973 -1.9 1983 -0.8 1993  0.1 2003  0.4
1954  -0.5 1964 -0.8 1974 -0.7 1984 -0.9 1994  0.9 2004  0.7
1955 -1.6 1965  1.8 1975 -1.5 1985 -0.2 1995 -1.0 2005 -0.4
1956 -0.5 1966 -0.1 1976  0.8 1986  1.0 1996 -0.4 2006  0.9
1957  1.3 1967 -0.4 1977  0.8 1987  1.2 1997  2.3 2007 -1.2
1958  0.6 1968  0.6 1978 -0.1 1988 -1.7 1998 -1.3 2008 -0.5
1959 -0.1 1969  0.8 1979  0.5 1989 -0.2 1999 -1.4 2009  1.2


My tabulation was nine times "no" and ten times "yes". On two occasions there were two years of 0.5 or more then followed by a year with -0.5 or less. Not shown is the value for 2010 which was -1.3 which is one of the ten times that there was this year to year dramatic change. So I concluded that there is a 50% chance of a La Nina in the winter of 2016/2017 on a statistical basis alone. However my general feeling is that the winter of 2016/2017 is most likely to be ENSO Neutral.

To examine this question more carefully, I prepared the below table where I show what I know about the ten years where the ONI changed dramatically from positive to negative. I show what I know about the warm event which in most cases was an El Nino of some sort and I show the PDO and AMO with the sign/phase in the year shown followed after a comma by the sign/phase in the following year. It is not conclusive but I do not see the current pattern of PDO+ and AMO+ or Neutral represented in this table except in 1997 and probably also in 1987. I do not expect the PDO to be negative next year so that kind of invalidates the 1997 case as being predictive. 1987 was a Modoki and this now is a pretty much traditional El Nno so I do not believe the 1987 case is particularly predictive. Thus I conclude that the winter of 2016/2017 being a La Nina is less than 50%. If you knew nothing you might assign a probability of 25% for a La Nina year. Given that this winter will be an El Nino, you might assign a probability of 33% to the following year being a La Nina. It is like a box of one red, one blue and two white balls. If you draw from that box, the probability of a red ball is 25%. If you remove the blue ball because you just had an El Nino winter, the Markovian probability of a red ball increases to 33%. I think this is how Australia looks at things. But ENSO is not a four-year cycle but more like a 5 to 7 year cycle. So we could try to be more fancy but the results may not be better. 

Reversal Years
Year Comments PDO AMO
1953   -,+ +.-
1963 Modoki Type I -,- -.-
1969 Modoki Type II which started in 1968 +,- -.-
1972 Traditional El Nino +,- -.-
1982 Traditional El Nino +,+ -.-
1987 Modoki Type1 +,- N.-
1994 Modoki -,+ -.+
1997 Traditional El Nino +,- +.+
2006 Traditional El Nino -,- +.+
2009 Modoki Type II N,- +.+

I have no data on the 1953/54 El Nino and I do not know why but it appears not to have been recognized by Japan. So I do not know if we are dealing with half the events being Modokis are more than half. The reason that might be important is that a Modoki is closer to being a La Nina than a traditional/canonical El Nino as the warm water is not as far east. So it might be easier for a Modoki to convert to a La Nina except some Modokis transform into traditional El Ninos which just happened. The 2014/2015 Warm Event was probably best described as a Near Modoki Type II that has now transformed itself into a traditional but late in the season El Nino. Obviously we have more to learn. 



B1. Very High Frequent Cycles (NAO, AO, PNA, MJO, Aleutian Low)

Air Pressure and Wind Related Cycles.

Air Pressure Cycles.

Arctic Oscillation (AO) and North Atlantic Oscillation (NAO). Basically both the AO and the NAO relate to the differential between mid-latitude and high latitude atmospheric pressure. The NAO is a bit easier to define as there are two semi-permanent features: the Icelandic Low and the Azores High sometimes called the Bermuda High. But still there are at least three different ways of measuring the NAO. I am not sure how and exactly where the AO is measured, but in both the AO and NAO a positive index indicates the differential between the mid latitude highs and the Arctic air pressure are larger than usual and a negative index means the opposite. The following table is my interpretation of the impacts of the AO and NAO but one needs to understand that these are two atmospheric factors (and thus somewhat independent of sea surface temperatures SST but that is not totally clear) which interact with many other factors so please view the following table as potential tendencies rather than a hard and fast prediction tool and also a work in progress.

Winter Impacts (Summer is different) AO+ AO- NAO+ NAO-
Mid-Latitude Jet Stream Stronger Weaker Stronger Weaker
Weather Variability Lower Higher Lower Higher
Alaska, Scotland, Scandinavia Wet Dry Wet and Warm Dry and Cold
Greenland and Newfoundland Warmer Colder    
U.S. East Coast Warmer Cold spells in the Midwest   Cold spells as far south as Florida
Western United States and the Mediterranean Dry Wet   Storms track further south
Southern Europe and North Africa      

Wet and Stormy

Here is a NOAA graphic that I modified (made horizontal rather than vertical) to show the impact of the Arctic Oscillation. The North Atlantic Oscillation's impact is similar but further east. It is not clear if these are two separate patterns or two parts of the same pattern as they are not perfectly correlated but they do represent somewhat the same situation.

Arctic Oscillation

Without having yet performed a careful analysis, it would appear that a negative AO or NAO or both present a negative factor for the economies of the nations impacted.

The current value of the NAO Index can be found here. The AO here.

Are the AO and NAO and ENSO Related?

Aleutian Low

Both the Tropics and the Aleutian Low have a major bearing on the weather of the Lower 48 States. Here is a good reference on the topic and this is Part II of that article. The Aleutian Low  S.N. Rodionov,  N.A. Bond,  J.E. Overland

Let us start with a more conventional map. Welcome to Beringia. Learn more here.  Also, all of Beringia is on the North American tectonic plate, as possibly is most of Japan although that is a bit controversial.

Beringia US Park Service

Now that you understand the part of the World I am talking about, the following climate-oriented graphic is of interest. The Aleutian Low is a semi-permanent weather feature most prominent in the winter. This graphic represents the mean (average) pattern of the Aleutian Low.

Aleutian Low Normal Position

Notice the ridge (the circle in the lower right of the graphic) off the West Coast which IMO is the RRR (Ridiculously Resilient Ridge). Also remember that low pressure areas (such as the Aleutian Low) in the Northern Hemisphere rotate counter-clockwise so the position of the Aleutian Low determines:

  • How cold it is in Siberia
  • How cold it is in Alaska
  • Where the storm track enters Canada or the Lower 48
  • The increase or decrease of the ice pack separate from any secular trend due to Global Warming

My major interest in this weekly report is to examine how the Aleutian Low impacts weather in the Lower 48. I am by no means an expert on this subject but I am doing my best to provide excerpts from the above two articles (which I believe are two of the best articles on this subject) and to provide some explanation that might help the reader understand what is in those two articles. It is a difficult topic and at this point not completely understood.

Compare the above which is the average condition of the Aleutian Low to the Day 3 Forecast.

Day 3 highs and lows

I apologize for not being able to find a current map: this is the Day 3 forecast but none of the current maps that I could find showed this large an area. Time flies so a Day 3 forecast works just fine. This graphic auto-updates so you will have to related the current (Day 3 conditions) at the time you read this information with the information provided below.

Here is some more information that I have extracted from the two linked articles (both from the same authors). They are perhaps difficult a bit difficult to read but you can find slightly more readable versions of these graphics at the above links. These are low resolution graphics and unless I had decided to wait and write the author and ask if he would send me some higher resolution versions, we have to work with these and I do think they are adequate for our discussion.

Let us start by considering what is considered the stronger Aleutian Lows which means lower Sea Level Pressure at the center and perhaps some other factors such as the area with low Sea Level Pressure (SLP). Notice the storm track.

Aleutian Low Strong Pattern

And then there are the years with a weak Aleutian Low. Notice the split low with two parts both generally weaker than a single low-pressure system and notice the quite prominent high pressure ridge off the coast of the U.S. Lower 48 States. Also notice the storm track which is interrupted and then reforms.

Aleutian Low Weak Pattern
Another way of looking at things is not to focus on the sea level pressure but the location of the center of the Aleutian Low. This is done by looking at individual months rather than conditions over the four-month winter. The approach was pioneered by Overland and Pease (1982) and confirmed in the above two papers. There is an overlap of authors so I conclude that it is a hardy band of dedicated researchers that have elected to focus on Beringia. A number of different types of situations were looked at but the two with the most explanatory ability are called W1 and C1 both of which explain a substantial amount of the variation in warm versus cold winter months in Beringia. But I am using the results to better understand the weather downstream of the Aleutian Low. Type W1 is defined as an Aleutian Low with a single center located north of 51N and between 156W and 173W. Type C1 is a split Aleutian Low with a western center located south of 52N and an eastern center positioned no further east than 140W. This type of analysis is done by month not by the winter season. The analysis is provided in the next two graphics. 

First the information for the W1 Type

Aleutian Low W1

Notice the Aleutian Low is shifted way to the west and the RRR is very pronounced. Also notice the storm track and the absence of significant storm activity one the eastern side of the Pacific in Beringia. Also, with W1 you can see how Bering Strait ice would be melted. I am not going to bore you, but there are special challenges when working with spherical coordinates versus Cartesian coordinates. In theory I am a mathematician and I have to say the discussion of this in the papers is fascinating but not relevant to our discussion.

Then the C1 Type.
Aleutian Low C1

Notice the split Aleutian Low and the really pronounced RRR. Also notice the storm track.

The following table relates to Figures 3 and 4 i.e. the winter analysis not the W1/C1 analysis.

Year Strength (inverse of NP Index Value) ENSO PDO AMO
Analysis of Strength of the Aleutian Low
1950 Low La Nina - +
1951 Low La Nina - +
1952 Low El Nino - +
1955 Low La Nina - +
1956 Low La Nina - +
1961 High Neutral + +
1969 Low El Nino Modoki II - -
1970 High El Nino + -
1971 Low La Nina N -
1972 Low La Nina - -
1977 High El Nino + -
1978 High El Nino + -
1981 High Neutral + -
1982 Low Neutral + -
1983 High El Nino + -
1986 High Neutral + -
1987 High El Nino + -
1988 High El Nino Modoki II + -
1989 Low La Nina - -
2003 High El Nino Modoki I + +

In this table I did not use a smoothing algorithm for the AMO and PDO but instead simply recorded their condition as per their associated index for the four- month winter season. I am assuming and fairly certain that the years provided by the author are the year starting in January of the Nov/Dec/Jan/Feb period used to categorize the Aleutian Low as being one of the ten strongest or weakest since 1944 which was a climate shift in the Pacific i.e. the Pacific Decadal Oscillation changed signs.

Looking at the table I was able to confirm the authors contention that during this period of time, the years of strong Aleutian Lows (very low surface air pressure) tend to be El Nino years with a positive PDO and the years of weak Aleutian Lows (surface air pressure greater than average) tend to be La Nina years with a negative PDO. The authors did not report on any consideration of the AMO but the above table which I created suggests that the AMO is not a factor.

Curiously this relationship did not apply prior to 1922 so that remains a mystery.

Aleutian Low Correlations with NP and PDO

Also of interest, since 1900 the first year in the above graphic, the NP Index appears to be trending lower meaning the Aleutian Lows are getting stronger. Is this Global Warming? The length of the trend suggests that it might be but still one can not explain the lack of the more recent correlation between the NP and PDO during the period 1900 to 1922. Also curiously there appears to be no current measurements of the NP index or ALP index at least that I could find so one wonders if NOAA has access to this data in their computer models.

Japan Bonin High

This article might explain it. From the Summary.


The Bonin high is a subtropical anticyclone that is predominant near Japan in the summer. This anticyclone is associated with an equivalent-barotropic structure, often extending throughout the entire troposphere. Although the equivalent-barotropic structure of The bonin high has been known for years among synopticians because of its importance to the summer climate in east Asia, there are few dynamical explanations for such a structure. The present paper attempts to provide a formation mechanism for the deep ridge near Japan. We propose a new hypothesis that this equivalent-barotropic ridge near Japan is formed as a result of the propagation of stationary Rossby waves along the Asian jet in the upper troposphere (‘the Silk Road pattern’).  First, the monthly mean climatology is examined in order to demonstrate this hypothesis.  It is shown that the enhanced Asian jet in August is favourable for the propagation of stationary Rossby waves and that the regions of descent over the Eastern Mediterranean Sea and The Aral Sea act as two major wave sources.  Second, a primitive-equation model is used to simulate the climatology of August.  The model successfully simulates the Bonin high with an equivalent-barotropic structure. The upper-tropospheric ridge is found to be enhanced by a height anomaly of more than 80 m at 200 hPa, when a wave packet arrives.  Sensitivity experiments are conducted to show that the removal of the diabatic cooling over the Asian jet suppresses the Silk Road pattern and formation of an equivalent-barotropic Ridge near Japan, while the removal of the diabatic heating in the western Pacific does not."  [Editor's Note: This is another of the atmospheric features that determine climate although this feature may not be as robust as some others and is mainly a factor during the summer.]

We have talked before about semipermanent Highs and Lows but one can never talk about it enough as these are so very important. Here is a K - 12 write up that keeps it simple.

CONUS during the winter of 2015/2016  has been most impacted in the West by the Aleutian Low and the Pacific High and in the East, the Icelandic Low and Bermuda High have also played a role. I want to focus on the Aleutian Low and the Pacific High which some call the North Pacific High and I call the Eastern Pacific Subtropical High (EPSH) because there is another High at the western end of the Pacific (actually two: the WPSH and Summer Okhotsk High).

Here is a good water vapor image of the Aleutian Low.

North Carolina State Image of Aleutian Low.

Pacific High

The Pacific High is located off the west coast of North America. During the summer months, it is located just off the coast of California. Winds blow clockwise around a high pressure system. This keeps the western coast of the United States relatively dry during the summer especially compared to locations along the East Coast. During the winter, the Pacific High moves farther south. This allows winds to blow into the west coast and the polar front to move south into the United States. This brings storms into the coast making the winter season wet across the western United States.

During the winter 0f 2015/2016 the Aleutian Low has been very strong but generally centered further west and north than is usual for an El Nino. At the same time, the Pacific High - which derisively is referred to as the Ridiculously Resilient Ridge (RRR) - has been further north than one would expect and thus valiantly protected the Western Coast of CONUS from Pacific storms. Thus we had a northerly displaced El Nino impact.

Of great interest is the topic of whether the changes in these features are cyclical (for periods of time one way and then for periods of time the opposite way) or secular (gradually changing in a particular direction) i.e. due to Global Warming. It is difficult to tell (and even more difficult to know if one is being scammed by those who do the studies) since ... oh well, no need to pull back the curtain and reveal how science is really conducted. It is better to maintain the illusion that scientists are objective and fully qualified to perform their research. Most are and if I provide links to articles or discuss articles, and if I do not say differently, you can assume that I have concluded that these are real scientific papers not commissioned papers and they have been written by qualified researchers. 

A way of looking at things is that if you see a change in the Eastern  Pacific you wonder if there has been a corresponding change in the Western Pacific. Here are two links to essentially the same paper on this so if you are interested please click here or here. One or the other may load and print more easily on your computer. I notice when I read that paper and the list of references that they seemed to be talking about the 1976 - 77 Pacific Climate Shift. So I conclude this is all probably mostly part of the PDO Cycle. A way to verify that would be to see if the changes discussed in this article (two links to the essentially the same article) have reversed now that the PDO seems to be shifting again to the same phase as what occurred in 1976-1977. That phase reversed in 1998.


Monsoons are seasonal changes in the prevailing wind direction caused by the changing differential between land and ocean temperatures during various seasons and there are usually associated changes in the location and strength of High and Low Pressures Systems. 

North American Monsoon (NAM)

What is the North American Monsoon. There are many resources for understand the North American Monsoon but I found these graphics from this University of Arizona report to be very useful.

The below graphic show how CONUS winter and summer storm tracks differ. In the winter there are usually two storm tracks as shown in this graphic. One is the Mid-Latitude Jet Stream and the other is the occasional Southern Branch of that same Jet Stream. The Southern Branch is on again off again and that is why the Southwest tends to be semi-arid. This graphic represents some sort of average conditions and does not take into account such factors as the phases of the PDO, AMO, ENSO and other factors which  impact the location and behavior of the Jet Stream. In the summer, the storm track is further north so the Southwest rarely has precipitation delivered by the prevailing westerlies. Instead the differential heating of the land versus the oceans creates low pressure and convection in Mexico and north of that process there are various high pressure systems which when well positioned draw the moist subtropical air into CONUS. I was glad to see that this graphic showed the influence of the Bermuda High on the process and we discussed that last summer and we will discuss it gain soon but probably not next week as next week is when NOAA releases their updated Seasonal Outlook. The Bermuda High also called the Azores High or NASH has a seasonal migration which brings it closer to CONUS in the summer but it also has its one 60 year cycle so we will talk about that also.

Summer Versus Winter

How does it come on?

Just as the above graphic relates to an average year, the below also applies to an average year and shows the seasonal evolution from June to July. In June, the High Pressure system called the Monsoonal Ridge is not usually far enough north for the anticyclonic wind pattern to draw the moist tropical moisture up into Arizona and New Mexico. This tends to not happen on a regular basis until July.  Notice the potential to impact a very large number of states. It is not just the Southwest which is impacted. It is a very complicated process as there are two mountain ranges in Mexico which impact things and there are two bodies of water which provide the moisture: the Gulf of California and the Gulf of Mexico.
One can see how the IPCC CC Models have great difficulty in putting this all together to assess what the ultimate impact will be,  I do not blame them but I do blame politicians who draw assumptions which are premature and academics who write reports when they know full well that there are many studies that arrive at different conclusions. When the answer is "we do not know for sure", that is a proper answer. We have a better idea of how the process worked historically during cool periods (megadroughts) than we know how it will work during periods that are warmer than the historical record.

North American Monsoon Development June to July

Remember, these graphics are simplifications.

The North American Monsoon affects Mexico and the the southern tier of the U.S. mostly Arizona and New Mexico but it can impact states to the north and to the Northeast of Arizona and New Mexico thus it has a significant impact in the summer over the entire Southwest and even the Plains States. 

And in term of how El Nino and Pacific and other factors impact the North American Monsoon (NAM), this from a presentation on the Albuquerque NM Weather Service web site may be helpful.

Monsoon Graphics Andy Church

Click here for a larger Image.

This article on the role of sea surface temperatures in the Gulf of California re the initiation of the NAM in Arizona may also be of interest. This article may be related.

And here is what we were looking for in June as a precursor to the development of the North American Monsoon. A nice big high pressure center in Northern Mexico heating up the land.

NAM June GeoPotential

Seasonal Distribution of Precipitation during the Monsoon


Similar patterns are characteristic of other monsoons around the world which are generally based on the land warming and thus impacting the direction of prevailing winds.

Re Understanding the Southwest Monsoon this may may come in handy

Sierra Madre Mexico

Some may find this article on Northern Hemisphere Monsoons to be of interest.

Multidecadal to multicentury scale collapses of Northern Hemisphere monsoons over the past millennium Yemane Asmerom and Victor J. Polyak Department of Earth and Planetary Sciences, University of New Mexico, Albuquerque, NM 87131; Jessica B. T. Rasmussen Leander Independent School District, Leander, TX 78646; Stephen J. Burns Department of Geosciences, University of Massachusetts, Amherst, MA 01003 and Matthew Lachniet Department of Geoscience, University of Nevada, Las Vegas, NV 89154

The abstract is of interest.

Late Holocene climate in western North America was punctuated by periods of extended aridity called megadroughts. These droughts have been linked to cool eastern tropical Pacific sea surface temperatures (SSTs). Here, we show both short-term and long-term climate variability over the last 1,500 y from annual band thickness and stable isotope speleothem data. Several megadroughts are evident,including a multicentury one, AD 1350 to  1650, herein referred to as Super Drought, which corresponds to the coldest period of the Little Ice Age. Synchronicity between southwestern North American, Chinese, and West African monsoon precipitation suggests the mega-droughts were hemispheric in scale. Northern Hemisphere monsoon strength over the last millennium is positively correlated with Northern Hemisphere temperature and North Atlantic SST. The mega-droughts are associated with cooler than average SST and Northern Hemisphere temperatures. Furthermore, the megadroughts, including the Super Drought, coincide with solar insolation minima, suggesting that solar forcing of sea surface and atmospheric temperatures may generate variations in the strength of Northern Hemisphere monsoons. Our findings seem to suggest stronger (wetter) Northern Hemisphere monsoons with increased warming.

This may be new information to some. Warmer is wetter, cooler is drier.

The methodology of this research is interesting. Stalagmite BC-11 was collected in 2004 from Bat Cave, a room in Carlsbad Cavern, New Mexico. There was a second stalagmite labeled BC2 assessed also but since it had been broken off (circa 1923 AD and may have been compromised) was considered to be a bit less reliable and the thickness of the layers were calibrated to tree ring data. The thickness of the bands was used to estimate the annual precipitation which in that area is mostly Monsoonal. Apparently the research team were able to determine the source of the precipitation i.e. the Pacific (mostly winter) versus Gulf of California and Gulf of Mexico (mostly Monsoon).

You might  find this graphic of interest.

Asmerom Bat Cave Study

What is shown here are periods of drought shown in oranges and browns (which I gather are supposed to look gray) and the long period where the Monsoon was essentially absent. The authors call this the Hiatus. The Period 1350 AD to 1650 AD is of most interest. This corresponds with what is called the Little Ice Age (LIA).  In 1680 there was in New Mexico what is called the Pueblo Revolt.  My house is located either next to or perhaps on the site of a key part of that revolt. Like with most historical events, there are many causes but changes in climate were probably involved both in creating the large population in this area (similar to the current level of population here) and making life very difficult for this large population. We are quite familiar with the more or less 60 year cycles commonly called the PDO and AMO. But are there longer cycles? Another piece of the puzzle might be the Sunspot Cycles which are described in this paper.

There is much that is interesting in the stalagmite (I keep them separate by associating the "g" in stalagmite with growing from the ground and the "c" in stalactite as growing from the ceiling) analysis paper. It provides a basis for thinking about droughts that are longer than can be attributed to phases of the AMO and PDO. Such droughts may well tend to occur Worldwide (bad news re thinking that food deficits in one area can be made up elsewhere) and they tend to be associated with cold SST's not warm SST's. Talk about "an inconvenient truth". I doubt that the political establishment offers refunds for the dissemination of incorrect information.

The IPCC Climate Change models seem to agree that in the future all the Monsoons will be wetter except the North American Monsoon. The North American Monsoon is an outlier relative to the IPCC AR5 WGI Report. It is projected to be weaker. This paper raises questions on that re the historical record. But the North American Monsoon is a strange Monsoon as the actually monsoon is located in the Sonoran Area of Mexico. The U.S. is just on the edge of this monsoon. So the Monsoon itself could be stronger but the Impact on the U.S. Southwest might not be stronger. However this study suggests that in the Guadalupe Mountains of Southeast New Mexico, the North American Monsoon was stronger when it was warmer. This actually makes sense when you realize how monsoons work. But the Guadalupe Mountains are not the entire Southwest. So by itself it does not tell the full picture but it is pretty suggestive. It is also important to recognize that the North American Monsoon, sometimes called the Southwest Monsoon, or the Arizona Monsoon, impacts many CONUS states not just the states in the Southwest. So there are many variables including start and end dates, extent of impact beyond the Southwest, and the split of impact between Arizona and New Mexico.

Other than Tropical Storms, the major feature impacting CONUS in the Summer is the Southwest Monsoon or North American Monsoon if you prefer.

Here is a very good paper that explain the U.S. Monsoon.

A METHOD FOR DEFINING MONSOON ONSET AND DEMISE IN THE SOUTHWESTERN USA ANDREW W. ELLIS, ERINANNE M. SAFFELL and TIMOTHY W. HAWKINS Office of Climatology, Department of Geography, Arizona State University, Tempe, AZ 85287-0104, USA

The method employed by the Phoenix NWSFO [Editors' Note. See below i.e. this method is no longer officially used] in determining the onset and demise of the annual monsoon season locally is the only full attempt at such a declaration. The NWSFO in Tucson only declares the onset, using a similar method to that used at Phoenix, whereas other NWSFOs make no such declarations regarding the  monsoon  season.  The  beginning  of  the  monsoon  season  at  Phoenix  or  at  Tucson  is  defined  as  the  first day of the first occurrence of three successive days during the summer months characterized by mean daily dew-point  temperatures  of  55°F  (12.8°C)  and  54°F  (12.2°C)  or  greater  respectively.  Any  subsequent  day exhibiting  a  mean  dew-point temperature equal to  or  greater than  the  threshold  is  classified  as a ‘monsoon day’ up through to the ending date of the monsoon season. The end date at Phoenix is determined in retrospect by a forecaster upon the review of daily moisture conditions and atmospheric circulation patterns nearing the fall season. Using this method, the monsoon season at Phoenix traditionally begins on 7 July (Table I), but it has begun as early as 19 June and as late as 25 July. Similarly, the season typically begins at Tucson on 3  July  (Table I),  but  it  has begun  as early  17  June  and  as late  as 25  July.  The  season  at  Phoenix  typically stretches across a 70 day period through to 14 September (earliest end date: 19  August; latest end date: 10 October), but it has been as short as 34 days and has persisted for as long as 104 days (Table I). Between the beginning  and end  of the  monsoon,  the mean  number  of monsoon  days  is  55,  with  a  minimum  of 27 days and a maximum of 99 days.


Dew Points and the North American Monsoon

Notwithstanding the above, Phoenix and Tucson NWS have moved away from formally predicting the start of the monsoon by the meeting of particular criteria. Now the Monsoon in Arizona starts on June 15. I am not sure of the reason for this downgrade in precision but I assume it means less phone calls from the media. When dealing with the media, KISS. The Phoenix NWS in their daily technical discussion is quite detailed in discussing the onset of the NAM. So the level of analysis has not declined just the way things are described officially

An oddity about the recording of Monsoons is discussed in part in this paper. The follow are my comment on their analysis.

The purging of the June 5 Start in 1972 of the Monsoon in Phoenix is interesting. Prior to this year's El Nino. the 1972/73 El Nino was the third strongest ever and then ONLY STRONG EL NINO to occur with AMO-. I have presented the evidence before and will present it again in the near future and it is available on Page II of this report that AMO- is correlated with the annual migration of the Bermuda Low and is correlated with strong Monsoons. So it is not a surprise that 1972 produced a very early Monsoon.

Here is a good recent paper on the evolution of the NAM. Mechanisms for the Onset and Evolution of North American Monsoon Ehsan Erfani and David L. Mitchell  Desert Research Institute, Reno, Nevada  University of Nevada, Reno, Nevada

It focuses on Sea Surface Temperatures (SST's) and is not a surprise. The reason for the study is interesting. NOAA was having difficulties forecasting precipitation during a Monsoon. It does raise questions about the logic of concluding that Global Warming will reduce the strength of Monsoons. The paper concludes that 29C in the Northern Gulf of California is necessary to have this Monsoon. 27.5C is the general estimate of what causes Convection but apparently the Marine Boundary Level requires 29C for what is called CIN (temperature inversion) to be surmounted. But there are other factors. One has to be creative to figure out how more convection leads to a drier Planet.

Monsoon Model Bias

Monsoon References

Kenneth Grantz Literature Review



Special Topic of Most Influence on 2105 Weather Worldwide - The ENSO Cycle - Currently the now probable appearance of an El Nino - with some hints that the level of optimism might be a bit overstated.

I guess we should start by describing what ENSO is. Basically the El Nino Southern Oscillation is no more or less than:

  • The location of the warm water in the Tropical  Pacific i.e. collecting to the west or to the east. 
  • The resulting pattern of convection and precipitation in the Pacific called the Walker Cycle and 
  • Any interactions between the movement of warm water and the impact of these movements on the prevailing Easterlies along the Equator which themselves have an impact on the movement back and forth of the warm water. 

When the warm water is shifted to the West we have what is called La Nina Conditions. When the warm water is shifted to the East we have what is called El Nino Conditions and anything in between is called ENSO Neutral.  Although the Walker Circulation as it applies to the tropics in the Pacific and is usually measured by what is called the Southern Oscillation (SOI) Index is important for confirming the phases of ENSO and sustained values of -8 are associated with El Nino Conditions (and it is at that level now), the approach most used to assess ENSO conditions is to measure the Sea Surface Temperatures in four areas of the tropical pacific shown in the below graphic.

El Nino Zones

Statistics show that the temperature anomaly in the area labeled NINO 3.4 above has the highest correlation with the weather impacts we observe worldwide with the La Nina and El Nino phases of ENSO so this is the most common measure used and is called the Ocean Nino Index or ONI. More information is available here. One very interesting statement in that document is

"During the remainder of the year a larger SST anomaly, up to +1.5°C in November-December-January, is required in order to reach the threshold to support persistent deep convection in that region."

This graphic shows the hstorical evolution of the STT Anomalites in the four measurement areas.

SST Anomalies

This is interesting to me given the excitement when the ONI exceeds 0.5. The ONI is not the only measurement of interest and some say it is not the best but it is the one most widely utilized.

The first graphic below shows the most recent runs of the tool used to predict which phase of ENSO (El Nino Southern Oscillation), El Nino or La Nina, is likely in the coming months. A word on how to read this graphic. Notice there are a number of different model runs shown on that graphic and color coded. It is important to remember when looking at the forecasts of these models that getting above the 0.5 line does not make an El Nino Event but essentially creates what is called El Nino Conditions. One has to average that level for five consecutive overlapping three-month periods for an El Nino to be declared. So we are looking  ahead and seeing positive signs of an El Nino coming but are not nearly able to make a definitive forecast at his point in time especially in terms of the strength of the event. Notice the ensemble mean is now substantially below 1.0 especially for the early winter months (eyeballing it it seems like a SST of 0.8 which is barely an El Nino)  meaning that even those committed to there being an El Nino this winter are predicting a very weak one. Also notice the movement of the forecast mean higher into this coming summer which conflicts with the the discussion recently released by NOAA. 

ENSO Forecast

This second graphic is the projected sea surface temperatures.  It provides a more global view of many cycles not just ENSO.  The current information strongly suggests that next winter will have some characteristics associated with El Nino conditions worldwide. Notice the red area off the coast of Ecuador. That is the signature of El Nino. The warm water returning from the Western Pacific to the Coast of Ecuador. Notice this has now been updated to be a Nov/Dec/Jan map i.e. this Winter in the Northern Hemisphere. Of interest this graphic unlike the first graphic has been statistically adjusted by screens to improve the calibration of the data shown. Notice the projected continued warm water off the coast of Alaska a characteristic of a Modoki Type II and possibly a shift to a positive phase of the PDO.  But keep in mind that this graphic is beyond the Spring Prediction Barrier so one has to keep that in mind as the models may well change as we move beyond Spring.

Along the Equator, by this point in time (Jan  - Mar 2016) the most significant part of the warm anomaly along the Equator is predicted by NOAA to still be in El Nino mode i.e. shifted way to the east.

SST Out Months

Notice the higher latitude anomalously warm water to some extent in the northern Pacific and to a greater extent in the eastern Pacific. To me this PDO+ pattern is more significant than the projected anomalies along the Equator.

The other maps in that series can be found here.

This shows the current condition of Sea Surface Temperatures

This version auto-updates so you can see the evolution.

Daily SST Anomaly

To put this into animation click here. It goes fast and shows SST from two weeks ago up until I guess today or yesterday.  

Here are two others in that series a one month and three month average of the prior SST data.

Monthly SST Anomaly

And here is the three-month. 

Seasonal SST Anomalies as of May 25, 2015Here it is as an animation showing the evolution of the recent sea surface temperature anomalies. It shows the movements better than the static graphics but you have to pay close attention and perhaps watch it over and over again. I know I do.

Pacific SST Anomalies Animation


Here is a version that auto-updates and starts at an earlier time and goes through steps three days at a time. But you have to watch closely. Watch the El Niño dissipate and turn into a Modoki Type II.  But is now  in front of our eyes developing more characteristics of a Traditional El Nino. The Gulf of Mexico is interesting also.

SST Anomaly Animation


Here is another useful too and it updates DAILY.  It  provides a visual representation of the Tropical Pacific both surface temperature and winds absolute and anomalies. One can see what is called the "Cold Tongue" that extends into the Tropical Equator off of Ecuador. This needs to be warmed up for there to be a Traditional El Nino and that might be happening right now.


Current SST and wind anomalies

This graphic is currently showing the favorable conditions of reduced Easterlies (see the arrows in the anomalies graphic pointing to the east) and the warm water anomaly off the coast of Peru which is supporting the development of a possible El Nino. You can find a larger version here.

This animation may provide some additional insight but it may, when you look at it, not have progressed to the current time frame. Note the dates as they flash by. They are weekly averages so you only get a more recent frame each week. This graphic shows the absolute temperatures not the anomalies. You can see the tendency for warm water to pool in the Western Pacific which in the extreme is La Nina Conditions. When it returns to the East you get what is called an El Nino. If the warm water is mostly near the Date Line it is a Modoki.

Equatorial SST Animation

Here is a total Pacific view. Unlike the above graphic which shows the absolute temperatures, this shows the anomalies compared to average conditions. N

Pacific SST weekly Anomalies

The following map may be helpful in understanding how ENSO works.

OSU Origins Graphic

In the above, you can see what is called the "cold tongue" where the current coming up from Antarctica is redirected out to sea as it runs into Peru.

Humbold Current

You can read more about it here courtesy of Wikipedia.

View from Australia

El Nino

Australia POAMA ENSO model run

The discussion can be found here.

IOD (Indian Ocean Dipole)


The graphic comes with only a very short discussion which can be found here.

The interrelationship between the IOD and El Nino is complicated and not fully understood.

July 27, 2015 History of ONI Index


The Oceanic Niño Index (ONI) Index can be found here.

Many data resources on ENSO can be found here

 A good resource on Tropical Meteorology can be found here

The Multivariate ENSO Index (MEI) can be found here.  Some believe it is a better Index in that it covers more variables than the ONI Index.

Information on Modoki El Nino a variant of El Nino can be found here.  An even more technical analysis can be found here

The tradition Southern Oscillation Index or SOI Index which basically describes the pattern of convection along the Equator in the Pacific can be found here.

The formula for calculating the SOI is:

                 [ Pdiff - Pdiffav ]
        SOI = 10 -------------------


Pdiff   =   (average Tahiti MSLP for the month) - (average Darwin MSLP for the month),
Pdiffav   =   long term average of Pdiff for the month in question, and
SD(Pdiff)   =   long term standard deviation of Pdiff for the month in question.

Some additional information on the Troup SOI can be found here.

Strangely although -8 (El Nino) and +8 (La Nina) is the official guidelines for using the SOI levels to identify ENSO phases they sometimes use -6 and+6 for internal use within Australia as indicated in this poster.

Australia use of SOI

But actually the traditional SOI Index maintained by the Australia Bureau of Meteorology is off the Equator a bit. A different index NOAA Equatorial Southern Oscillation Index maintained by NOAA is right on the Equator and it can be found here.

SOI and the Release of Ocean Heat into the Atmosphere

I thought it would be useful to review this paper by Chris R. de Freitas and John D. McLean. You can find the full paper here

I am just going to present two figures (Figure 1 and Figure 3) from that article and discuss them. Notice on the Y Axis in both graphics the SOI goes from very negative numbers (more El Nino-ish) down to more positive numbers (more La Nina-ish).

SOI and MGT deFreitas

Basically what this is showing is that the Mean Global Temperature (MGT) tends to track the Southern Oscillation Index (SOI). This makes sense because the SOI is a measure of how concentrated the Warm Pool in the Western Pacific is. When it is thicker but concentrated in a smaller area, the opportunity for evaporation and convection is less than when the Warm Pool is spread out and thinner but covers a larger area. This is the ENSO Cycle and the SOI is one of the two Indices that measure the state of the ENSO Cycle although NOAA tends to ignore it unless it confirms their wild ideas about what is an El Nino and what is not. Evaporation is the way ocean heat is transferred into the atmosphere. So the Tropical Pacific works like a battery absorbing heat from sunshine and periodically releasing heat.

This from Wikipedia might provide some additional useful information.

"For example, when water evaporates, energy is transferred from a water molecule to an air molecule that contains less water vapor than its surroundings. Because energy is required for the water molecule to overcome the forces of attraction between water particles, the transition from water to vapor requires an input of energy and causes a temperature drop in the water molecule's surroundings.

If the vapor then condenses to a liquid on a surface, then the vapor's latent energy absorbed during evaporation is released as the liquid's sensible heat onto the surface.

Thus not only does convection transfer energy from the ocean to the atmosphere, the movement of water vapor by wind transfers the location of where this happens to where the water vapor created by evaporatlon is reconverted to latent heat as cloud droplets releasing sensible heat i.e. heat that can be felt and measured by a thermometer.

But what you see in the above graphic is a discontinuity right after 1995 when there was a substantial increase in temperature. The Graph on the right (and please notice the slightly different scale on the Y Axis) shows that the relationship between the SOI and MGT resumed after a few years. Those few years are generally considered to be a Climate Shift in the Pacific and was triggered or at least occurred at the same time as the Mega-El Nino of 1997/1998.

I do not know if anyone has an explanation as to why this relationship that normally holds did not hold during a few years in the mid-nineties. But it provides a pretty good explanation of what is called the Hiatus in the increase in MGT. Elsewhere in this Report, I provide a pretty convincing case that we will soon have another Climate Shift in the Pacific and perhaps that also will come with a Mega El Nino perhaps next year or perhaps without such an event. It might again result in a stepwise increase in MGT. I would suggest that attempting to figure out why the 1997/1998 was so effective at releasing stored ocean heat, would be useful. It also would be useful if NOAA learned what a Modoki was. That is important as an El Nino Modoki does not provide as much surface area for heat transfer to the atmosphere as a Traditional/Canonical El Nino.

El Nino Modoki

Not all El Nino's are alike and in recent years the distinction between traditional or canonical (true to form) El Ninos and Central Pacific El Nino's often called El Nino Modoki has become increasingly clear. In Japan they are forecasting an El Nino morphing into an El Nino Modoki. This becomes important because the impacts of each type of El Nino on weather worldwide is very different. 


This is illustrated in the following maps:

Starting with surface temperature:

Traditional El Nino  El Nino Modoki

Modoki Temperature Panel -2

Notice for the US, the Modoki results in an extreme division of the warmer versus cooler temperatures with the East very much colder than climatology. You can see the differences in Asia and other places.

And now considering precipitation

Traditional El Nino  El Nino Modoki

Modoki Precipitation Panel

Notice that for the U.S., the Modoki results in a drier West much like a La Nina.

You can find these graphics and more information here and here.

And below is a breakdown of precipitation anomalies from climatology by important parts of the world very impacted by the Modoki form of El Nino. You can find more information on this here.

Ashoc Geographical Breakdown

Again the Modoki has almost the opposite impact as a Traditional El Nino in many parts of the World. It is most extreme for CONUS and Japan with the Modoki being dry and the Traditional El Nino wet.


So there is a big difference in the differential impact on weather yet BOTH WILL REGISTER AS AN EL NINO ON THE ONI INDEX.

Some think a Modoki is some sort of strange thing. To me it looks like a Modoki is simply an El Nino that did not fully make it to Ecuador or after reaching Ecuador has taken its sweet time getting all the way back to the Western Pacific Warm Pool. From a weather perspective, with a Modoki you end up with two Walker Cells. So you have a lot of precipitation in the Central Pacific. Notice how the impact on Japan and the Western U.S. varies between a Canonical El Nino and an El Nino Modoki.

Most believe the El Ninos of 2002/2003 and 2004/2005  were actually of the Modoki variety and you can see that in the winter precipitation data for New Mexico and Rio Grande Stream Flow. The below may be a Bob Tisdale compilation but it is based on the Ashok formula which is generally recognized as the way of identifying a Modoki.

The Ashok formula is EMI = [SSTA]A-0.5x[SSTA]B-0.5x[SSTA]C where the A anomaly is calculated in the region 165E -140W,10S-10N, B is 110W-70W, 15S-5N, and C is 125E-145E,10S-20N.

Ashok Analysis

The 2009/2010 El Nino was also a Modoki,  So you see what you read in the news may not be very useful. I do not believe that many TV and Radio station weather people (many of whom are not even meteorologists but simply actors) understand the difference between these two types of El Nino's. It is fairly recent knowledge. 

First of all I want to start with the original defining of a Modoki which was publicized in a report by Ashok et al. in 2007. You can read the full report here. The definition has evolved a bit over time as Type I and Type II Modokis were later defined but I think this author is working with Ashok's original definition.

"Using observed data sets mainly for the period 1979–2005, we find that anomalous warming events different from conventional El Nino events occur in the central  equatorial Pacific. This unique warming in the central equatorial Pacific associated with a horseshoe pattern is flanked by a colder sea surface temperature anomaly (SSTA) on both sides along the equator. Empirical orthogonal function (EOF) analysis of monthly tropical Pacific SSTA shows that these events are represented by the second mode that explains 12% of the variance. Since a majority of such events are not part of El Niño evolution, the phenomenon is named as El Nino Modoki (pseudo-El Nino Nino) (‘‘Modoki’’ is a classical Japanese word, which means ‘‘a similar but different thing’’). The El Niño Modoki involves ocean-atmosphere coupled processes which include a unique tripolar sea level pressure pattern during the evolution, analogous to the Southern Oscillation in the case of El Niño. Hence the total entity is named as El Nino–Southern Oscillation (ENSO) Modoki. The ENSO Modoki events significantly influence the temperature and precipitation over many parts of the globe. Depending on the season, the impacts over regions such as the Far East including Japan, New Zealand, western coast of United States, etc., are opposite to those of the conventional ENSO. The difference maps between the two periods of 1979–2004 and 1958–1978 for various oceanic/atmospheric variables suggest that the recent weakening of equatorial easterlies related to weakened zonal sea surface temperature gradient led to more flattening of the thermocline. This appears to be a cause of more frequent and persistent occurrence of the ENSO Modoki event during recent decades."

Citation: Ashok, K., S. K. Behera, S. A. Rao, H. Weng, and T. Yamagata (2007), El Nino ̃o Modoki and its possible teleconnection,J. Geophys. Res.112 , C11007, doi:10.1029/2006JC003798.

 Now let us look at the impact of Modoki or not Modoki on convection.

Fei Zie Figure 8

As I mentioned these charts were prepared for a different purpose namely to examine when convection was so powerful that the clouds came close to entering the Stratosphere. But this also shows at least partially where the strongest convection activity was in the case of 8d for traditional/canonical El Nino events and in the case of 8c where convection was during Modoki events. If nothing else, this tells you about cloud height.

As you can clearly see, the pattern is very different between 8c and 8d. But in 8a you see that the sea surface temperature anomalies were almost identical.

To make things even more complicated, some believe there are two kinds of El Nino Modoki's and this is described here. Among other things, the referenced article provides the details on how the following breakdown was developed. Based on these authors, we have since 1950 had sixteen El Ninos broken down as follows:

Traditional or Canonical El Nino Modoki Type I Modoki Type II
1951/1952 1963/1964 1968/1969
1965/1966 1987/1988 1979/1980
1972/1973 1990/1991 1991/1992
1976/1977 2002/2003 1992/1993
1982/1983   2004/2005
1997/1998   2009/2010

Although both the researchers work for NOAA and were based in Miami Florida when they did this research, they seem to have used the Chinese System of identifying an El Nino i.e. focusing on Nino 3.0 rather than Nino 3.4. Nino 3.0 is the eastern part of Nino 3.4. This may explain the slight difference between their characterizations and that of Bob Tisdale and NOAA in that these Chinese authors do not seem to recognize 2006/2007 as any kind of El Nino. It barely met the ONI criteria of five consecutive overlapping three month periods with the ONI being 0.5 or higher and in the first period it was exactly 0.5 so it may very well have not recorded as an El Nino when Nino 3.0 was used as the criteria rather than Nino 3.4 which is a bit further west.

That does not seem to be a good explanation to me but this was "barely" an El Nino and any slight change in criteria could rule it out. Tisdale categorized it as a canonical El Nino and when I look at the precipitation situation in New Mexico and the values of the PDO Index it seems to me that this was more of a relief between two La Nina events, one of which was quite strong, rather than an El Nino so I think the characterization in the above table is probably more on target than the NOAA characterization although it is a borderline case. 

But the general conclusion is that since the PDO turned negative in 1998 we may not have had a traditional or what is called technically a canonical El Nino. I have not yet matched the above categorization of Modoki I and II to precipitation patterns in the U.S. The referenced authors focused on the Western Pacific mainly China.

It tells me that we have to be very careful about how we relate to something called an El Nino since there are at least three varieties and each has a different impact on Global Weather.

Let us talk a bit more about the work of Chunzai Wang and Xin Wang. From the abstract of their article.


Based on the opposite influence on rainfall in southern China during boreal fall, this paper classifies El Niño Modoki into two groups: El Niño Modoki I and II which show different origins and patterns of SST anomalies. The warm SST anomalies originate in the equatorial central Pacific and subtropical northeastern Pacific for El Niño Modoki I and II, respectively. Thus, El Niño Modoki I shows a symmetric SST anomaly distribution about the equator with the maximum warming in the equatorial central Pacific, whereas El Niño Modoki II displays an asymmetric distribution with the warm SST anomalies extending from the northeastern Pacific to equatorial central Pacific. Additionally, the warm SST anomalies in the equatorial central Pacific extend further westward for El Niño Modoki II than El Niño Modoki I. Similar to canonical El Niño, El Niño Modoki I is associated with an anomalous anticyclone in the Philippine Sea which induces southwesterly wind anomalies along the south coast of China and carries the moisture for increasing rainfall in southern China. For El Niño Modoki II, an anomalous cyclone resides east of the Philippines, associated with northerly wind anomalies and a decrease in rainfall in southern China. Canonical El Niño and El Niño Modoki I are associated with a westward extension of the western North Pacific subtropical high (WNPSH), whereas El Niño Modoki II shifts the WNPSH eastward. Differing from canonical El Niño and El Niño Modoki I, El Niño Modoki II corresponds to northwesterly anomalies of the typhoon steering flow which are unfavorable for typhoons to make landfall in China."

Modoki I and II Seasonal Patterns

Here are the cyclone tracks for the three flavors of El Nino


Modoki impact on Western Pacific Storm Tracks.

As you can see:

  • Traditional El Ninos generate cyclones that develop further east than Modoki cyclones
  • Traditional El Nino cyclones tend to complete their journey further north than Modoki cyclones
  • Modoki Type II cyclones have the best chance of providing precipitation to the far west of the Tropical Pacific. Modoki Type I cyclones have a better chance of providing precipitation to Southern China.
  • Not shown in this graphic (I could have provided another set of graphics to show this) is the situation in the Philippine Sea. Modoki Type I tends to create an anomalous High Pressure System and Modoki Type II a Low Pressure System there. This changes the wind directions for the entire area.

A very good reference to the literature on ENSO can be found here.  It is an amazing document and the discussion on Global Warming is very interesting. There is no consensus on how Global Warming has or will impact ENSO although many think it does, But they disagree on how!

All of this is very important for many reasons which technically boil down to how the overall behavior of the Pacific and Tropical Pacific are interrelated. Currently we decompose the Pacific into the Pacific north of the Equator which we describe by one index called the PDO and the Tropical Pacific which we describe by one index called the ONI.  But from what I read (and I have in this article provided the links) it probably is less than fully productive to look at things the way we have been looking at things and I anticipate that more people will be looking at things differently.

Comparing El Nino to La Nina

ENSO Typical Winter Circulation US

The first two graphics, and there are many versions of these graphics but the above are very colorful, show how the Jet Stream tends to be further south during an El Nino (not this past winter). Notice the Aleutian Low in the El Nino graphic. The Blocking High in the La Nina pattern is reminiscent of summer weather but it is further off shore than a typical summer so it allows Pacific storms to enter the U.S. via the Northwest. 

Now we take a look at the Walker Circulation which is where the warm water causes evaporation and convection (cloud formation) and where these clouds tend to drop their precipitation which creates downwards air movements. The colors also reveal the location of warmer than climatology and cooler than climatology Sea Surface Temperatures (SST) which are used to define the pattern. Pay special attention to the water along the Equator.

La Nina Walker Circulation

It is difficult to see the difference between the La Nina (above) and Neutral conditions (below) but you should pay attention to the more pronounced (darker - thicker arrows) convection zone in the El Nino state as compared to the Neutral state and similarly the stronger subsidence zone in the La Nina state. Subsidence tends to warm and dry air masses.

ENSO Neutral

It is a lot easier to see the different between Neutral and El Nino as the pattern tends to reverse with Convection Zones converting to Subsidence (drying) Zones. This winter the Convection Zone was further west than usual and I have mentioned that almost every week.

Walker Circulation

But these are theoretical constructs and are only a guide to trying to understanding what is going on. And of course these graphics do not address Modokis. They are addressed on Page II of this report which can be accessed by clicking on Medium Frequency Cycles such as ENSO and IOD

Focus on the Indo-Pacific Warm Pool

The below is the typical situation of the Indo-Pacific Warm Pool. [Click for Source] This  graphic is not showing temperature anomalies but actual temperatures. The water colored red needs to move to the area between 170W and 120W to have an El Nino. That would shift the precipitation pattern since warm water evaporates more readily than cooler water. That is how ENSO works. it changes the location along and near the Equator where water is evaporating to form clouds and where the precipitation takes place and thus the locations where warm air is subsiding and drying. That pattern at any point in time is called the Walker Circulation. In general, the Walker Circulation differs dramatically for La Nina versus El Nino. I have presented those graphics showing the pattern many times and am not doing so tonight to keep this report at a manageable size.

Western Pacific Warm Pool

I do not want to go into a full discussion of ENSO dynamics this evening but the below graphic explains a lot. You see the prevailing Easterly Trade Winds which slacken off during an El Nino. The Trade Winds skim the warm water off the surface and it accumulates in the Western Pacific and to some extent in the Indian Ocean. The "Thermocline" is pretty interesting. It is the dividing line between the near surface temperature gradient and a steep discontinuity to much cooler water. The angle shown is steeper for La Nina and flatter during ENSO Neutral and may even point downwards during a strong El Nino when warm water accumulates off the coast of Ecuador. Notice the actual elevation of the water in the Western Pacific along the Equator may be as much as a meter higher than off the coast of Ecuador during La Nina. Warm water is less dense and the Easterlies cause water to build up in the Western Pacific. ENSO is like a battery. La Nina charges it. The discharge process is called El Nino. When viewed as a battery it should raise a red flag when an El Nino is projected soon after a very powerful El Nino occurred. The has not been sufficient time for the ENSO battery to recharge. The most recent La Nina was minimal at best.

ENSO Dynamics
More information is available here.

Queensland Australia also publishes graphics on how the Walker Circulation is related to the SOI and they are very interesting.

Queensland interpretation of SOI on Walker Circulation

Obviously they are mostly interested in the impact on Australia but the Walker Circulation is a worldwide process and the rightmost part of these graphics relate to North and South America.

El Nino Impacts

New NOAA Tool

I do not know when this was made available but I just noticed it which you can click on here. They say it shows the impact of ENSO but really it is set up to show the impact of El Nino with the instructions to reverse the sign for the impact of La Nina which is an approximation but is probably reasonable. The major problem I have with this is that the regression analysis did not take into account the strength of the ENSO event or the type of event i.e. for El Nino: Traditional, Modoki Type I, Modoli Type II. So it averages all El Nino events together. It is a clever graphic however as you can indicate which months you are interested in.

North Carolina has a very good State Climatologist so that is a good place to seek information on the impact of a possible El Nino. The following charts were all prepared by NOAA but I found them here so I am using them from the North Carolina Climatologist's web site.

This chart is interesting:

Typical El Nino

Notice all the red off the coast of Ecuador and Peru. The current forecast map looks a lot like this but not quite the amount of warm water off of Ecuador and Peru. So that is the difference between what is currently projected and a truly powerful El Nino.

This chart shows the impacts for December through February.

Winter El Nino Impacts

And here is the information for June through August. As you can see, El Nino (as is La Nina) is a worldwide event.

El Nino June through August

These charts do not show the impact on ocean fisheries which is another important part of the story. Wet or dry; warm or cold can have an impact related to the difference from the norm. Crops that like moisture usually thrive during wet periods but too much moisture can also be harmful and certain crops prefer it to be relatively dry. So one has to look at the impacts region by region. And of course there is the risk of flooding in some places.

Some have raised issues about the coffee crop and in the U.S. the Florida citrus harvest is often impacted by El Nino.  Every area highlighted on this map is likely to be impacted either negatively or positively by an El Nino event if it occurs.

Here is another set of graphics from British Met which can be found here.  It shows impacts for both El Nino and La Nina and is broken down by precipitation and by temperature. There are many ways to do this analysis and show the results. 

El Nino Precipitation

El NinoTemperature

La Nina Precipitation

La Nina Temperature

It is very difficult for researchers to isolate internal variability (ENSO, decadal and multi-decadal ocean cycles) from the secular trend in climate due to anthropogenic forcings. That is one reason why much can be learned from observing the current impacts of El Nino,  La Nina and the longer ocean cycles. To some extent, climate change resembles a continual and strengthening El Nino but there is a limit to the usefulness of that analogy since El Nino does not impact the globe in exactly the same way as anthropogenic climate change. But there are a lot of similarities.

Also every instance of a phase of a climate cycle is unique. So the above represents the typical impacts. 

Focusing on the U.S.

How is  El Nino likely to  impact the U.S.?  This first map shows the impact on temperature as compared to ENSO Neutral Conditions.

Temperature Deviation from Mean

It would be colder in the south and warmer in the north but it is not a big difference. But is it enough to put the Florida citrus crop in danger? How might natural gas prices be impacted?

This map shows the impact on precipitation.

Precipitation Deviation from Mean

It is wetter in the West and in the Southeast.

And again these are not big differences but when you compare the El Nino data not to the mean but to La Nina conditions, it starts to be more significant.

Spring and Summer Impacts of El Nino

First let us take a look at the impact of an El Nino on Spring weather in the North Atlantic and Northern Europe Region.

"Delayed ENSO impact on spring precipitation over North/Atlantic European Region." Full article by Ivan Here Buick and Fred Scarfskin can be found here.

ABSTRACT The delayed impact of winter sea-surface temperature (SST) anomalies in tropical Pacific on spring precipitation over the North Atlantic/European (NATE) region is examined using both measured and modeled data for the period 1901–2002. In an AMI-type Atmospheric General Circulation Model (ACM) ensemble, the observed delayed spring precipitation response in Europe to winter ENSO-related SST anomalies is well reproduced. A series of targeted ACM/coupled GC experiments are performed to further investigate the mechanisms for this delayed influence. It is found that late winter ENSO SST anomalies lead to the well-documented Crosby wave train arching from the Pacific into the Atlantic region. A positive (negative) ENSO event leads to a quasi-barotropic trough (ridge) in the North Atlantic region. The resulting wind and cloud changes cause anomalies in the surface heat fluxes that result in negative (positive) SST anomalies in the central North Atlantic and anomalies of the opposite sign further to the south. The SST anomalies persist into spring and the atmospheric response to these anomalies is an extension of the ENSO-induced trough (ridge) into the European region, leading to enhanced (reduced) moisture flux and low-level convergence (divergence) and thus positive (negative) precipitation anomalies. Although the signal is overall relatively weak (correlation coefficients of European spring rainfall with winter ENSO SSTS of about 0.3), a proper representation of the outlined mechanism in seasonal forecasting systems may lead to improved seasonal predictions.

So what they are saying is that El Nino can lead to a wetter, cooler Spring for Northern Europe. If you read the paper you will see that there are a lot of nuances but of course the major problem is that we have not had an El Nino Winter. Nevertheless, it is reasonable to consider the increased odds, probably small, of some degree of a cooler, wetter spring in the North Atlantic/European Region.

But what if it is a Modoki and not a Traditional El Nino. This is the Abstract of an article that can be found here.

Contrasting Impacts of Two Types of ENSO on the Boreal Spring Hadley Circulation: Juan Fen and Japing Li

State Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, China

The possible influences of two types of ENSO [i.e., the canonical ENSO and ENSO Modoki (EM)] on Hadley circulation (HC) during the boreal spring are investigated during 1979–2010. El Niño events are featured with a symmetric pattern in equatorial zonal-mean sea surface temperature anomalies (SSA), with a maximum around the equator. In contrast, the zonal-mean SSA associated with El Niño Modoki events shows an asymmetric structure with a maximum around 10°N. The contrasting underlying thermal structures corresponding with ENSO and EM have opposite impacts on the simultaneous HC. In El Niño years, a symmetric anomalous meridian circulation is seen, with enhanced rising around the equator and anomalous descent at about 15°N and 20°S. In contrast, an asymmetric equatorial meridian circulation is observed for El Niño Modoki years, with anomalous ascent around 10°N and descent at about 10°S and 20°N. The contrasting meridian circulation anomalies within ENSO and EM are caused by their different meridian SSA structure. This result is theoretically explained, indicating that anomalous meridian circulation is subject to the meridian SSA gradient. Moreover, the observed results are reproduced in numerical experiments driven by anomalous warming in the eastern and central Pacific Ocean. Thus, the authors conclude that the anomalous HC linked to ENSO and EM is induced by the accompanying meridian gradient in zonal-mean SSA.

I did not have access to the full paper but I interpret the abstract to suggest that convection zones may shift five to ten degrees of latitude north. I do not understand the full implication of this but given that there are three major circulations in each hemisphere: Hadley, Ferrell, Polar each of which nominally spans 30 degrees, a shift of the Hadley Circulation by 5 to 10 degrees north, is likely to have a significant impact which I have referred to in the past as a rotation. 

You have seen this before [Author D. Windrim]

Earrth' Major Circulation Patterns

We tend to think of Planet Earth as divided into six zones: three in the Northern Hemisphere and three in the Southern Hemisphere.  It all starts at the Equator where warm air rises. That air cannot travel into outer space so it moves toward one or the other of the Poles. At some point, the air cools and subsides creating what is called the Subtropical Ridge. It is another way of describing the Hadley Circulation. But the location of this Subtropical Ridge moves to some extent as part of its seasonal cycle but it is also influenced by the phase of the ENSO Cycle. Since we are in an El Nino Phase now, which will for sure last through the summer, there is a tendency for the subsidence process to occur closer to the Equator than usual. We will see how this might impact our weather this summer. It is most likely to impact the Southwest but it should also have impacts in the Southeast as well.

Here is some background information on the Subtropical Ridge.

And let us not forget about the East Coast and here is one more.useful article. But these are of most interest during a La Nina Cool Event. But it is kind of two sides of the same coin.

This from a recent forecast from the Albuquerque NWS.



The point is simply that the location of certain semipermanent features such as the Eastern Pacific Subtropical High (sometimes called the North Pacific High or the RRR) and the Bermuda High (Sometimes called NASH or the Azores High) impact our weather. The Bermuda High has a 60 year cycle namely the AMO which is gradually increasing the western extension of the Bermuda High year by year at this point in its cycle. Not totally sure about the movement of the Eastern Pacific Subtropical High but you can learn a lot from this slide presentation. And this graphic tells a story.

Yu, Lu, and Kim El Nino Location

Basically this is saying that with PDO Negative, Modokis are more likely and the Hadley Cell is stronger. But right now we have something much more like a traditional Eastern Pacific El Nino and we may have PDO Positive but for sure the Pacific Subtropical High is shifted to the East. Thus we have the Pacific High shifted or extending to the East and the Atlantic High extending to the West which puts two Highs close to each other. That is not the usual way things happen as there normally is H - L - H - L - H etc. not H - H or H - H - H. When you have adjacent clockwise systems, they do not get along well together. Thus the interesting excerpts above from the Thursday Technical Discussion released by the Albuquerque Office of the NWS. Another way of putting it is "we are not sure what will happen to the moisture from Hurricane Carlos".  Also there may be a tendency for the two highs to tend to rotate around each other. I think that would most likely cause the Pacific High to move around the Bermuda High. I think I have detected what I have called "rotation" in the past. I am not talking about anything dramatic but something subtle.

We are talking about a Spring and Summer pattern that might show up frequently over the next few decades with the AMO shifting into its Neutral and then Negative Phase and be more exaggerated in El Nino years. Wouldn't it be nice if water planners understood this sort of thing?

Worldwide Impacts

For these first six graphics, Winter is on the Left and Spring and Summer are on the right.

First we will look at the mid-atmosphere I'm geopolitical height anomalies as this provides a good way of determining weather patterns.

500mb Geopotential Height Anomalies World

There is quite a bit of change from Nov - Mar impacts to May - Sep impacts and these are anomalies relative to 1981 - 2010 climatology so it does not simply reflect the seasonal change. Less land area is impacted in the Spring and Summer than in the Winter. In general it appears to be mostly a muting of impact in the Northern Hemisphere but in the North Atlantic it is the opposite impact which we have discussed earlier as a delayed telecommuting between the Pacific and the Atlantic during a warm event in the Pacific. There appears to be an enhancement of impact in the Southern Hemisphere which is not at all strange as Spring and Summer in the Northern Hemisphere is Fall and Winter in the Southern Hemisphere and El Nino is primarily a Fall and Winter event. Australia seems to be an exception.

Next we will look at Temperature Anomalies.

World El Nino Temperature Anomaly Winter Vs Summer

It looks like the Poles see the most variation of anomalies between Winter and Spring/Summer which again is simply the reality that the seasons are reversed in the Southern Hemisphere as compared to the Northern Hemisphere. In general the impacts of El Nino are muted in the Spring and Summer in the Northern Hemisphere. But there is a tendency for the Northern Hemisphere to be slightly cooler during an El Nino Spring and Summer. Although we will look at the U.S. in more detail you can see the slight impact on the U.S. here. I will be coming back to that later.

Next we will look at precipitation.

World Precipitation Anomaly El Nino Winter Vs Summer

And again the impacts are muted in the Spring and Summer except along the Equator. I was surprised by the very low level of impact in the Nov - Mar graphic on the left for the U.S. More on that later.

Focus on U.S. Impacts

Now we will zero in on the U.S. and consider impacts with higher resolution graphics.

500 MB Pressure US El Nino Winter vs Summer

The major impact for CONUS appears to be somewhat lower 500m geopolitical heights in the Southwest. On the other hand British Columbia seems to have the opposite occur. And there is a tendency towards lower pressure in the North Atlantic which we discussed earlier in this week's report.

Now we will look at temperature.

U.S. Temperature Anomaly El Nino Winter vs Summer

The major impact appears to be in the North Central states which might be a decree cooler in the Spring and Summer during El Nino Conditions. There appears to be a rather small area in Southern Nevada and Southern California that might be one degree warmer. I was perplexed at first when I noticed that the graphic on the right does not appear to agree with the worldwide graphic presented earlier. But then I realized that the earlier graphic was based on anomalies relative to 1981 - 2010 and this graphic uses 1971 - 2000 as the base climatology. This convinces me that the impacts of the Pacific (PDO) and Atlantic (AMO) may in Spring and Summer overwhelm the impacts of El Nino.

Now let us take a look at precipitation.

UI.S. Precipitation Anomaly El Nino Winter vs Summer

This is the most interesting of the graphics, a bit wetter in the Center of CONUS and quite a bit drier on parts of the East Coast. But none of this shows up in the earlier worldwide graphic. Again it seems to me that the base climatology which is different in the worldwide graphics (all of this information comes from the same website) overwhelms the impact of El Nino especially in Spring and Summer. An increase of five inches of precipitation in a small area in the Mid-west and a decrease of five inches of precipitation in coastal Alabama may seem like a large impact but even if valid it applies to two to four climate divisions i.e. a very small area. Most of the impacts shown for the center of the U.S. in this graphic are of one inch magnitude wetter or two inches of magnitude drier and summer especially tends to be wet in that part of the U.S.. So I am not sure that these anomalies are meaningful. This of course raises the question of why NOAA thought it was useful to even raise the question of whether this warm event might ultimately be declared an El Nino when the impacts on Spring and Summer are minimal at best.

El Nino versus La Nina in the Spring and Summer

Now let us shift our approach and compare El Nino Spring and Summer Impacts to La Nina Spring and Summer Impacts. These are likely to be larger as we are comparing what should be two opposite impacts in many cases. In these two graphics, El Nino is on the left and La Nina is on the right.

First we can look at temperature.

Temperature Anomaly El Nino vs La Nina

At first glance there are no surprises here. With El Nino the West is a bit cooler and with La Nina the Plains States and Great Lakes extending into the Northeast is a bit warmer. The information on El Nino should be consistent with what was shown earlier and is consistent with the worldwide graphic which uses the same base climatology.

Now we look at precipitation.

Precipitation Anomaly El Nino vs La Nina

And here surprisingly we see little impact of El Nino but La Nina is clearly slightly drier for much of the Eastern part of CONUS. That was a little bit of a surprise but suggests the delayed impact of La Nina in the following season further east similar to how El Nino might telecom impacts further east on a delayed basis as described in one of the papers I reviewed earlier in this report.

You can find some interesting graphics here that differentiate between weak, moderate and strong El Nino. It also has separated out Spring impacts for Spring and Summer combined as I have done above.It is a private website so I can not just lift their graphics so I have provided you with a link to Jan Null's website which is called Golden Gate Weather Services. 

General Conclusion: El Nino is not a factor in the U.S. in the Spring or Summer and NOAA and some in the Media have wasted a lot of people's time.

California Niño/Niña. It seems that Baja California has its own cycle of changes in the upwelling of cold water and that this may be somewhat independent of the ENSO process but be instead specific to Baja California and Southern California. More information on this can be found here.

California NINO Information

The map to the left shows the geographic area we are talking about and the chart on the right shows: (a) Lead-lag correlation coefficients between the July-September California Niño/Niña indices and the 3-month-running mean California Niño/Niña (CA, grey filled bar), along-shore surface wind (ASW, dark open bar) and upwelling (UWI, blue line) indices from 1982-2011. The ASW is positive equatorward, and UWI positive upward.

This is a difficult graphic to interpret but basically it is showing that July to September conditions in the summer impacts wind and upwelling into the following winter and this may not be connected to the ENSO Cycle but is a separate (intrinsic) phenomenon.

A couple of thoughts about this research:

A. It is interesting how the Japanese found this pattern rather than some professor at a California University. Does this tell us anything about the lack of focus by the U.S. on U.S. climate other than projects funded under the rubric of Climate Change?

B. There more than likely are other similar micro climates like this. Micro is relative as this pattern may impact a fairly large area.

Comparing Current El Nino to 1997/1998 El Nino

First a graphic provided by Andrew Church of the Aluquerque NWS Office.

Graphic from Andrew Church Feb  2016 Spring Analysis.


Let us discuss perhaps the most famous El Nino ever (although the 1982/1983 El Nino is famous also). I have been able to include the links to the NOAA graphics in the following NOAA Discussion (so you can click on the Figures indicated in the article) but not the links that move you between sections of the article as GEI does not support those commands and they complicate reading the article. You can read the article in the original by going to  the El Nino of 1997/1998 . The purpose of this discussion is to compare and contrast the current El Nino with the World Champion 1997/1998 El Nino. The text in the report is the complete article from NOAA. I have selectively added some of their graphics into the article here so you do not have to click on those to see them.  Also, I have added some graphics from the current El Nino for comparison and some of my comments as [Editor's Notes].

As mentioned above, my analysis is still a work in progress. I may do some additional work on it during the week and present a more complete analysis in my August 17 Weather and Climate Report or at a later time. Let's face it. Every El Nino is different so it is not easy to compare one to another. But the comparison may be informative  and helpful to those who can benefit from additional insight into how this El Nino may evolve. And it is important to remember that for CONUS, the major impacts are likely during Fall so the data I am presenting is prior to the full impact of this El Nino, not a retrospective. Not everyone will want to wade through this detailed and highly technical analysis so they can scroll down to the section on current weather conditions and forecasts. .

From what I see so far, It suggests that this El Nino is not going to match the 1997/1998 El Nino but it is closer to it than one might think. I suspect the Faux El Nino of 2014/2015 discharged some of the energy in the ENSO battery or we might be having a record matching or beating El Nino right now. Also, and not much discussed, this El Nino is arriving early and thus may have lesser impacts on the Northern Hemisphere. One thing I have learned from attempting this comparison is that El Nino is very much about the Southern Hemisphere and not much information is presented on that in the NOAA Weekly ENSO Update.

From the CPC.NCEP.NOAA Report on this subject, with the link provided above, Very strong 1997-98 Pacific warm episode (El Niño)

1. Overview

The global climate during 1997 was affected by one of the strongest Pacific warm episodes on record. These warm episode conditions developed rapidly in March, with strong ENSO conditions persisting from May through the end of the year (and subsequently well into 1998) (Figs. 20, 21 ). During this episode the abnormally warm SSTs which covered the eastern half of the equatorial Pacific (Fig. 21a) were comparable in magnitude and areal extent to the famed 1982/83 El Niño (Fig. 20a). These 1997 warm episode conditions were accompanied by a strong negative phase of the Southern Oscillation, with the equatorial Southern Oscillation Index (SOI) also comparable in magnitude to that observed during 1982/83 (Fig. 20b). Also evident since April was a markedly reduced strength of the low-level (850-hPa) equatorial easterly winds across the eastern tropical Pacific (Fig. 21c). At times these anomalies indicated a complete disappearance of the easterlies across the entire eastern Pacific, along with a complete collapse of the normal equatorial Walker circulation. These anomalies were also comparable to those observed during the 1982/83 El Niño (Fig. 20c).

The above conditions were associated with a dramatic alteration of the global pattern of tropical rainfall and deep tropical convection, as indicated by above-normal rainfall across the eastern half of the tropical Pacific and by significantly below-normal convection across Indonesia and the western equatorial Pacific (Fig. 22). The combined zonal extent of these rainfall anomalies covered a distance more than one-half the circumference of the earth.

Selected impacts associated with these warm episode conditions included 1) excessive rainfall across the eastern half of the tropical Pacific, 2) significantly below-normal rainfall and drought across Indonesia and the western tropical Pacific, 3) below-normal hurricane activity over the North Atlantic during August-October, with a simultaneously expanded region of conditions favorable for tropical cyclone formation over the eastern subtropical North Pacific 4) excessive rainfall and flooding in equatorial eastern Africa during October-December 5) a dramatic eastward extension of the South Pacific jet stream to well east of the date line during June-December which resulted in enhanced storminess across southeastern South America and central Chile and abnormally dry conditions across the Amazon Basin, central America, the Caribbean Sea and the subtropical North Atlantic throughout the period.

2. Evolution of the 1997 El Nino

The year began with a continuation of weak cold episode conditions in the tropical Pacific during December 1996-February 1997 (DJF 1996/97). These conditions included a well-defined tongue of abnormally cold SSTs extending across the eastern tropical Pacific (Fig. 23b), with equatorial SSTs greater than 28°C confined to the area west of the date line (Fig. 23a). SSTs were slightly below normal in the Niño 1+2 region (Fig. 24a), the Niño 3 region (Fig. 24b) and the Niño 3.4 region (Fig. 24c), and substantially below the 28°C threshold for convection (Gadgil et al. 1984) in all three regions

[Editors Note: I have added a graphic which has two sets of information. The left-hand figure is the long-run history of the Nino 3.4 reading which in the U.S. is the major way of tracking the phases of ENSO i.e. El Nino versus Neutral versus La Nina. It goes back to 1950 so you can easily see the 1997/1998 El Nino there followed by a long La Nina and then a series of weak El Nino's. It is easy to follow that graphic and you can see how rapid the buildup was in the Nino 3.4 reading which is often referred to as the ONI the Ocean Nino Index. On the right is more recent information but not just for the ONI which is the second row down but for four Nino Indices representing different areas along the Tropical Pacific and they go from west to east with the Nino 1 + 2 being a little south of the Equator to cover Equador and Peru. You can see that it has not been as dramatic a rise although one always has to be careful when reading a graph to be aware of the way the X and Y axis are structured. There was the Faux El Nino in the winter of 2014/2015 which was a false start and probably drained some of the energy from this warm event which otherwise would probably be even stronger and rival the 1997/1998 El Nino. You can see that until recently it has been mostly a Central Pacific warm event (Nino 4) which some call a Modoki. We clearly are in an El Nino now but the Nino 1+2 area SST anomalies appear to have peaked and leveled off. There is a lot of information in this combined graphic. It does not auto-update but NOAA updates this information weekly in their ENSO Report so it is avaiable.]

August 10 2015 NINO HISTORY

The colder-than-normal conditions were accompanied by a strongly sloping equatorial oceanic thermocline (Fig. 25a), with increased thermocline depths across the western Pacific and reduced depths over the extreme eastern Pacific [The center of the thermocline is approximated by the 20°C isotherm]. These variations in thermocline depth were accompanied by abnormally warm ocean temperatures in the western and central tropical Pacific between 50-200 m depth and by abnormally cold water in the eastern tropical Pacific between the surface and 50-100 m depth. During this period, the atmosphere featured 1) a positive phase of the Southern Oscillation, with below-normal sea level pressure (SLP) across the western tropical Pacific (Fig. 21b), 2) a broad area of slightly enhanced low-level easterlies across the central tropical Pacific (Figs. 21c, 26a ), 3) enhanced tropical convection over Indonesia and the western tropical Pacific [indicated by negative values of anomalous Outgoing Longwave Radiation (OLR)] and suppressed convection in the vicinity of the date line (Fig. 26a), and 4) westerly wind anomalies at upper levels across the eastern tropical Pacific (Fig. 27a). Collectively, these conditions reflected an enhanced equatorial Walker circulation and a continued coupling between the positive phase of the Southern Oscillation and below-normal SSTs across the eastern tropical Pacific.

In contrast, March and April featured an extremely rapid transition to one of the strongest warm episodes of the century. SSTs increased nearly 1.5°C over the normal annual cycle in the Niño 1+2 region during March (Fig. 24a), and nearly 1.0°-1.5°C over the annual cycle in the Niño 3, Niño 3.4 and Niño 4 regions.  In the Niño 4 region this increase occurred during a two-week period and was greater than the entire annual cycle of SST for that region. A second period of very rapid SST increases in the east-central Pacific then occurred during April, as SSTs in both the Niño 3 and Niño 3.4 regions climbed an additional 1°C over that expected from the normal annual cycle. Thus, by mid-April SSTs exceeded 28°C across the central and east-central equatorial Pacific (Figs. 24b-d ), with values averaging 1°-3°C above normal in all four Niño regions. Area-averaged SSTs in the Niño 3, Niño 3.4 and Niño 4 regions then remained nearly constant at values greater than 28°C throughout the remainder of the year. This warming reflected a nearly complete elimination of the annual cycle in SSTs across most of the equatorial Pacific, which is normally characterized by a peak in temperatures during March-April and a minimum during September-October.

For the MAM season as a whole, mean SSTs greater than 29°C extended to east of the date line, and values greater than 28°C extended to approximately 160°W (Fig. 23c). These temperatures averaged 0.5°-2.0°C above normal across the central and east-central tropical Pacific (Fig. 23d). This warming was accompanied by increased depths of the oceanic thermocline everywhere east of the date line (Fig. 25b), and by a flattening of the thermocline across the region. In the eastern tropical Pacific, this suppressed thermocline reflected substantially reduced oceanic upwelling in association with a weakening of the low-level equatorial easterly winds (westerly wind anomalies) (Fig. 26b ). These conditions were accompanied by suppressed convection throughout Indonesia and enhanced convection in the vicinity of the date line, which is opposite to the pattern observed the previous season.

Editor's Note: Although labeled Outgoing Longwave Radiation Anomalies the OLR (description from NOAA: Negative (Positive) OLR are indicative of enhanced (suppressed) convection and hence more (less) cloud coverage typical of El Niño (La Niña) episodes. More (Less) convective activity in the central and eastern equatorial Pacific implies higher (lower), colder (warmer) cloud tops, which emit much less (more) infrared radiation into space) is a surrogate for precipitation with blue being wet and the red shades being dry. So in MAM we had the suppressed convection in the Western Pacific and increased precipitation near the Date Line. But we have not yet had the 1997/1998 shift in precipitation to the Eastern Pacific. In fact there has not been much chance since May. You can also look back on this Hovmoeller and see the Faux El Nino of 2014/2015 and see why I never concluded that it was real]. 

OLR Anomalies Along the Equator

It may be useful to compare the current OLR Anomalies with the 1997/1998 Super El Nino.

1997 El Nino OLR History

It is now August and this El Nino is a bit early but the OLR pattern has not shifted to the East as in did with the 1997/1998 El Nino. It still could happen, but so far it has not. There has been almost no impact.

During JJA, SSTs remained very warm throughout the entire eastern half of the tropical Pacific, with the 29°C isotherm expanding eastward to approximately 150°W, and the 28°C isotherm extending eastward to approximately 125°W (Fig. 23e ). These extremely warm waters are highly abnormal for that time of year (Fig. 23f), a period normally characterized by a marked decrease in SSTs across the eastern tropical Pacific. As a result, SST anomalies increased substantially throughout the region, and exceeded 3°- 4°C between 130°W and the west coast of South America (Fig. 23f). [Editor's Note we are currently running about 2.6C and I do not think it will increase but it might. So this El Nino is just under the 1997/1998 El Nino on that metric.]. This increase in anomalies was accompanied by a further flattening of the oceanic thermocline across the eastern Pacific (Fig. 25c), as the 20°C isotherm dropped to more than 100 m depth and ocean temperatures increased to more than 7°C above normal between 50-125 m depth. [Editor's Note: It this case it has been 6C but mostly 5C. However one has to keep in mind the base keeps on being adjusted by NOAA due to a trend that may be Global Warming or may be the PDO but the main point is that anomalies and absolute values are not the same thing.  So the subsurface water temperature is slightly higher than the anomaly indicates.]

[Editor's Note: I incorporated Figure 25 from the NOAA report directly in the discussion because of its importance so you do not need to call up that link.]

1997 SST

[Editor's note: for comparison purposes this is the current subsurface Thermocline situation

Subsurface Heat Anomalies

[Editor's Note: One of the things I notice is that our current conditions correspond to the Sep-Oct-Nov conditions in 1997 with just a slight difference in intensity.

It is difficult to read the NOAA graphic for 1997 but it looks like an anomaly of 7C and the most extreme we have had is 6C over a smaller area of the subsurface. At this time of the year in 1997 the warm anomaly covered a much larger extent of the Equator. There was no sign of an Upwelling Wave to signal the end of  the Kelvin Wave Activity. I have sensed that we have been in the final stages of this El Nino and this analysis makes me more confident that what I have sensed is indeed correct. At one point it seemed that his was an El Nino that arrived late but that was because we were considering the Faux El Nino of 2014 as a warm event that was slow in getting started. Now I am thinking what we have is a warm event that is peaking early. I am also thinking that the weather services may not be sufficiently taking that into account in their long-term forecasts]

The JJA period also featured an increasingly negative phase of the Southern Oscillation, and a further decrease in the strength of the 850-hPa easterly winds (3-6 m s-1 below normal) across most of the central and eastern tropical Pacific (Fig. 26c ). These conditions were accompanied by increased tropical convection (Fig. 26c) and rainfall across the entire eastern half of the Pacific, and by decreased rainfall across the western tropical Pacific and Indonesia. These changes in tropical convection reflected 1) a pronounced eastward extension of the primary area of tropical convection to well east of the date line, and at times an actual shift of the main region of tropical convection to the eastern half of the tropical Pacific (not shown), and 2) a strengthening and equatorward shift of the intertropical convergence zone (ITCZ) in the Northern Hemisphere (not shown). [Editor's Note: I thought I would show what things look like this year compared with the very powerful 1996-1997-1998 El Nino.  Notice the SOI went very negative early in 1997. Also notice it went positive the following Spring.

1907 SOI

And here is the current reporting of the SOI

2013-2014-2015 SOI \

The 1997 El Nino came on suddenly. This one has had a gradual build up re the increasingly negative SOI ] Also observed during JJA was the development of an anomalous upper-level anticyclonic circulation in the Southern Hemisphere subtropics between the date line and 90°W (Fig. 27c). This feature is a recurring aspect of the winter hemisphere circulation during strong warm episodes (Arkin 1982), and reflects several important changes in the flow occurring in the Tropics, the subtropics and the extratropics. In the Tropics, the equatorward flank of the circulation anomaly reflects anomalous upper-level easterly flow across the eastern Pacific, and thus comprises important structural elements of the much weaker-than-normal equatorial Walker circulation observed during the period. In the subtropics, above-normal heights (not shown) accompanying the anticyclonic circulation anomaly reflect an eastward extension of the mean subtropical ridge to well east of the date line, in response to the increase in tropical convection and deep tropospheric heating across the eastern tropical Pacific. In the extratropics, the anomalous anticyclonic circulation is an integral component of the coupling process between changes in tropical convection and changes in the wintertime jet stream across the South Pacific. The Southern Hemisphere circulation was also characterized by recurring high-latitude blocking over the high latitudes of the eastern South Pacific, a feature typical of strong warm episode conditions (Karoly, 1989).Strong warm episode conditions continued during SON (Figs. 23g, h), with SSTs greater than 28°C extending eastward from Indonesia to 125°W and greater than 29°C extending eastward to approximately 140°W (Fig. 23g). The normal cold-tongue that typically occupies the eastern half of the tropical Pacific at this time of the year was notably absent, consistent with the collapse of the normal annual cycle in SSTs throughout the region (Figs. 24b, c). A nearly isothermal temperature structure was also observed from the surface to 150 m depth, with ocean temperatures exceeding 9°C above normal at 50-150 m depth in the eastern Pacific (Fig. 25d).

Also during SON, El Niño-related enhanced rainfall and heavy tropical convection developed across equatorial eastern Africa. This rainfall was associated with low-level easterly wind anomalies across the tropical Indian Ocean (Fig. 26d), and with a continuation of extremely suppressed convection throughout Indonesia. Also observed was a continuation of enhanced upper-level westerlies and an extended jet stream across the subtropical South Pacific (Fig. 27d), resulting in continued heavy precipitation across Chile and southeastern South America. Elsewhere, above-normal rainfall and increased storminess developed across the Gulf Coast of the United States (Fig. 22), in association with enhanced upper-level westerlies across the southern tier of the country.

3. Equatorial Walker Circulation

Over the equatorial Pacific the divergent component of the atmospheric circulation is intimately related to the distribution of tropical convection, which in turn is an integral part of the still larger Southern Oscillation (Bjerknes 1969). This divergent circulation is often partitioned into its zonal and meridional components, respectively called the equatorial Walker circulation and the tropical Hadley circulation. The equatorial Walker circulation is characterized by ascending motion over Indonesia and the western tropical Pacific, and descending motion over the east-central equatorial Pacific, with upper-level westerly (low-level easterly) flow completing the "direct" circulation cell. Following Halpert and Bell (1997), we illustrate the equatorial Walker circulation using pressure-longitude plots of the vector field whose horizontal component is the divergent zonal wind and whose vertical component is the scaled pressure vertical velocity. The pressure vertical velocity was subjectively scaled to give a sense of the relative vertical motion in the equatorial plane. The seasonal mean equatorial Walker circulation and anomalies during 1997, along with the accompanying seasonal relative humidity anomalies, are shown in Fig. 28.

During DJF 1996/97, a well-defined equatorial Walker circulation was present (Fig. 28a), with ascending motion over the western tropical Pacific, descending motion over the eastern Pacific and a circulation center near 170°W. These conditions reflected a slight strengthening and an overall westward shift of the circulation center compared to normal (Fig. 28b), consistent with weak cold episode conditions and a positive phase of the Southern Oscillation. These conditions dissipated rapidly during MAM 1997, as a near-normal strength and location of the Walker circulation prevailed (Figs. 28c, d).

By JJA 1997, ascending motion and deep tropical convection encompassed the tropical Pacific between 140°E and 120°W (Fig. 28e ), while no well-defined pattern of vertical motion was evident over Indonesia. This anomalous vertical motion field (Fig. 28f) reflected a nearly complete disappearance of the equatorial Walker circulation. The pattern was also accompanied by enhanced relative humidity everywhere east of the date line, and by reduced relative humidity across the western tropical Pacific and Indonesia. These conditions strengthened during SON 1997, with the equatorial Walker circulation again nearly absent (Figs. 28g, h).

4. South Pacific jet stream during July-September 1997

In both the Northern and Southern Hemisphere, the extratropical wintertime jet stream over the western and central Pacific is intimately related to the distribution of tropical convection across Indonesia and the tropical Pacific. Thus, the interannual variability of these jet streams is strongly influenced by the ENSO. During strong El Niño conditions the wintertime jet stream extends eastward to well east of the date line, and over the eastern Pacific is shifted well equatorward from normal. These changes in the jet stream reflect a deep baroclinic jet structure often extending across the entire Pacific Basin, along with a pronounced eastward shift of the normal jet exit region to well east of the date line. These conditions then contribute to enhanced storminess and above-normal precipitation at lower latitudes of both North and South America.

During 1997, the South Pacific jet stream was particularly impacted during July-September by the ongoing strong El Niño conditions, while the primary impacts on the North Pacific jet stream did not occur until early 1998. Thus, this analysis focuses on the wintertime South Pacific jet stream, which extended across the entire South Pacific and brought enhanced storminess and above-normal precipitation throughout Chile and southeastern South America [Editor's Note" This longer current animation shows how the Jet Stream is crossing the Pacific. Perhaps the followin graphic which includes a discussion of how the Polar Jet Stream is usually impacted by ENZO will be helpful]

ENSO Jetstream

[Editors's Note: I have not seen this happening. In general the Jet Stream has been further north rather than further south. This may change in the coming months.]

The core of the South Pacific jet stream (approximated by wind speeds greater than 50 m s-1) during July-September is typically located between 22.5°-32.5°S and extends eastward from eastern Australia to approximately 150°W (Fig. 29a ). The jet entrance region is normally located over eastern Australia, and is characterized by a local maximum in along-stream increases in geostrophic wind speed. Additional characteristics of the entrance region include confluent geostrophic flow at upper levels and a strong poleward component of the horizontal ageostrophic flow directed toward lower geopotential height. This ageostrophic flow is one component of the thermodynamically direct, transverse ageostrophic circulation typical of any midlatitude jet entrance region (Palmen and Newton 1969, sections 1.5 and 8.3; Hoskins et al. 1978, Keyser and Shapiro 1986), and produces the required westerly momentum and kinetic energy increases that air parcels experience as they approach the jet core.

Farther downstream, the jet exit region is normally found between the date line and approximately 125°W, and is characterized by a local maximum in along-stream decreases in geostrophic wind speed. Characteristic features of this exit region include diffluent geostrophic flow at upper levels and a strong equatorward component of the ageostrophic flow directed toward higher geopotential height. This ageostrophic flow is one component of the required thermodynamically indirect, transverse ageostrophic circulation typical of any midlatitude jet exit region, and produces the required westerly momentum and kinetic energy decreases that air parcels experience as they exit the jet.

The July-September 1997 period featured an eastward extension of the jet stream across the entire South Pacific (Fig. 29b), and an extension of the jet core to 105°W (nearly 45° east of normal). This extension was accompanied by a pronounced eastward shift in the regions of along-stream decreases in geostrophic wind speed and strong diffluent geostrophic flow, and by a nearly complete elimination of these features in the vicinity of the climatological mean jet exit region. Collectively, these conditions reflected an eastward shift in the location of the jet exit region to between 130°-90°W. This dramatic structural change in the jet stream was accompanied by a dynamically consistent eastward shift in the primary region of equatorward-directed ageostrophic flow at upper levels to the observed jet exit region, indicating a corresponding shift in the entire thermodynamically indirect, transverse ageostrophic circulation that characterizes the jet exit region.

This jet extension and eastward shift of the jet exit region were intimately related to an eastward extension of the subtropical ridge to well east of the date line, which is identified by a well-defined anticyclonic circulation anomaly across the entire eastern subtropical South Pacific during JJA and SON (Figs. 27c, d). Additional aspects of this link between the two features are revealed by examining their common attributes. One common feature is the region of enhanced westerlies over the eastern South Pacific, which comprises both the poleward flank of the anticyclonic circulation anomaly and the extended South Pacific jet stream (compare Figs. 27c, and 29b, c). Two additional common features are the poleward flow and equatorward flow along the western and eastern flanks of the anticyclonic anomaly, respectively, which contain important dynamical information regarding links between the anomalous subtropical ridge and changes in the jet entrance and exit regions.

[Editor's Note: Here is a better look at the current Western Pacific.]

Western Pacific Tropical Activity

[Editor's Note: But this is the North Pacific not the South Pacific. Below is the South Pacific but I am not sufficiently familiar with that graphic to fit it into the discussion but it is amusing to see low pressure areas spinning clockwise. But I was expecting to see a giant anti-cyclone and I do not see it. I do see westerlies. I have more work to do!]

Southern Pacific

The anomalous poleward flow comprises several important structural changes occurring in the exit region of the climatological mean Pacific jet. First, it contributes to anomalous geostrophic confluence throughout the region (Fig. 29c), which also coincides with the entrance region of the anomalous westerly wind maximum. Second, it comprises a dynamically consistent pattern of anomalous ageostrophic flow at upper-levels, directed toward lower geopotential heights at an angle nearly orthogonal to the jet axis. This ageostrophic flow reflects an anomalous thermodynamically direct, transverse ageostrophic circulation, and results in abnormally strong Lagrangian increases in kinetic energy throughout the region (Fig. 30a). In this particular case, both the rotational (Fig. 30b) and divergent (Fig. 30c) components of the ageostrophic flow contributed strongly to these kinetic energy tendencies. Collectively, these anomalies are consistent with an almost complete elimination of the normal jet exit region in the vicinity of the date line, and with a reduced strength of its attendant transverse ageostrophic circulation.

A similar examination indicates that the equatorward flow along the eastern flank of the anticyclonic circulation anomaly comprises important structural and dynamical features of the observed jet exit region. For example, this equatorward flow contributes to geostrophic diffluence in the observed jet exit region (Fig. 30c), an area which also coincides with the exit region of the anomalous westerly wind maximum. The equatorward flow also comprises a coherent pattern of ageostrophic flow directed toward higher geopotential heights at upper levels, at an angle nearly orthogonal to the jet axis. This ageostrophic component of the flow reflects the well-defined thermodynamically indirect, transverse ageostrophic circulation previously noted in the jet exit region, and results in Lagrangian decreases in kinetic energy throughout the area (Fig. 30a). In this case, the rotational component of the ageostrophic flow contributes more to these kinetic energy tendencies (Fig. 30b) than does the divergent component (Fig. 30c).

Thus, in this case the anomalous poleward and equatorward flow found respectively along the western and eastern flanks of the anticyclonic circulation anomaly are strongly linked to jet dynamical processes through El Niño-related changes in the subtropical ridge. These flow features also highlight the jet-like character of the anomalous westerly wind maximum found along the poleward flank of the anticyclonic circulation anomaly.

Past History of ENSO Events.

History of ENSO Events

With respect to relating analog dates to ENSO Events, the following table might be useful. In most cases this table will allow the reader to draw appropriate conclusions from NOAA supplied analogs. If the analogs are not associated with an El Nino or La Nina they probably are not as easily interpreted. Remember, an analog is indicating a similarity to a weather pattern in the past. So if the analogs are not associated with a prior El Nino or prior La Nina the computer models are not likely to generate a forecast that is consistent with an El Nino or a La Nina.

El Ninos and La Ninas


  Start Finish Max ONI PDO AMO Start Finish Max ONI PDO AMO
            DJF 1950 JJA  1950 -1.5 - N
T MJJ  1951 DJF  1952 1.2 - +          
  JFM 1953 JFM 1954 0.8 - + AMJ 1954 ASO 1956 -1.7 - +
M MAM 1957 JJA  1958 1.8 + -          
M OND 1958 FMA 1959 0.6 + -          
M MJJ  1963 JFM 1964 1.4 - - AMJ 1964 DJF  1965 -0.8 - -
M AMJ 1965 MAM 1966 2.0 - -       - -
M SON 1968 AMJ 1969 1.1 - -          
T JAS  1969 DJF  1970 0.9 N - JJA  1970 DJF  1972 -1.4 - -
T AMJ 1972 FMA 1973 2.1 - - AMJ 1973 JJA  1974 -2.0 - -
            SON 1974 MAM 1976 -1.7 - -
T ASO 1976 JFM 1977 0.9 + -          
M ASO 1977 DJF  1978 0.8 N            
M SON 1979 JFM 1980 0.6 + -          
T MAM 1982 MJJ  1983 2.2 + - ASO 1983 DJF  1984 -1.0 + -
            SON 1984 JAS  1985 -1.1 + -
M ASO 1986 JFM 1988 1.7 + - AMJ 1988 AMJ 1969 -.18    
M AMJ 1991 MJJ  1992 1.7 + -          
M ASO 1994 FMA 1995 1.1 - - JAS 1995 FMA 1996 -1.0 + +
T AMJ 1997 AMJ 1998 2.4 + + JJA  1998 JFM 2001 -1.7 - +
M MJJ  2002 JFM 2003 1.3 + N          
M JJA  2004 JFM 2005 0.7 + + OND 2005 FMA 2006 -0.8    
T ASO 2006 DJF  2007 0.9 - + JJA  2007 MJJ  2008 -1.6 - +
            OND 2008 FMA 2009 -0.8    
M JJA  2009 FMA 2010 1.6 N + MJJ 2010 AMJ 2011 -1.7 + +
            JJA 2011 FMA 2012 -1.1 - +
T OND 2014 AMJ 2016 2.6 + N JAS 2016 NDJ 2016 -0.7 + +

ONI Recent History

November 6-, 2017 Revised Historical ONI Readins.

Note: Without fanfare the base climatology was changed from ERSST.V4 to ERSST.V5 

The JAS reading was adjusted to -0.2 which remains an ENSO Neutral Reading and the new ASO reading of -0.4 is also an ENSO Neutral Reading. The full history of the ONI readings can be found here. The MEI index readings can be found here.

e Observations:

  1. To be recorded as an El Nino or a La Nina, there must be five consecutive overlapping three-month periods where the ONI value is either 0.5 or greater (El Nino) or -0.5 or less (La Nina). But in some cases the warm or cold event lasts a lot longer than five consecutive overlapping three-month periods: perhaps as long as three years.
  2. Generally El Nino and La Nina take turns. In many cases an El Nino is followed by a La Nina and in turn the La Nina is followed by an El Nino etc. I have tried to organize the data in the below table to show this.
  3. But sometimes there is a short break between two El Ninos or two La Nina's and one wonders if that was really two events or a single event where the criteria forced there to be a declaration of two or three of the same phase of ENSO in a series when it really should have been considered a single event.
  4. The maximum ONI reading may not be representative of the strength of an El Nino or La Nina over its life since it might be a two or three year event with a short period of time with very high positive or negative ONI.
  5. And of course it is important to remember that although the ONI is measuring the temperature anomaly in an area of the Equatorial Pacific that is considered to be the correct place to measure ENSO events (in Asia the exact location is slightly different), in reality what the ONI is telling us is simply whether the Equatorial Warm Pool is to the west over by Indonesia, in the Central Pacific only (a Modoki), or extending from the Date Line all the way to the coast of Ecuador and Peru.  ENSO is really about the geographical distribution of warm water along the Equatorial Pacific.
  6. The prevailing Easterlies and other factors tend to move the warm surface water to the west creating the La Nina condition. When the pool of warm water builds up and if there is a slowdown of the Easterlies or if there is a wind burst that starts a Kelvin Wave, the warm water moves to the east. If it only gets to the Date Line, that is called a Modoki. If it makes it all the way to Ecuador and Peru that is a Traditional El Nino and if the warm water is pretty much evenly distributed, that is called ENSO Neutral. When warm water is spread out, there is more convection which causes more clouds and clouds further east and the process of evaporation to cause convection cools the water. So although the process is not regular and thus not a true cycle it is a process that must proceed from one extreme to the other and can not stay in any one state indefinitely.

Thus El Nino, El Nino Modoki and La Nina as well as ENSO Neutral should be considered normal events and part of the rhythm of Nature and not sensationalized. And yet each phase has positive and negative impacts on different parts of the World or at least half of the World and probably much more than half.

Analysis of the Above ONI Data

Now let us analyze that data. You can follow what I have done as all I have done to get started is look in the above table for five or more consecutive three-month periods where the ONI was 0.5 or greater (El Nino) or -0.5 or less (La Nina) as that is the NOAA definition of an El Nino or La Nina but in theory the synchronization with the Atmosphere must also be present but for now let's ignore that as I have no way of easily going back to see how the atmosphere and ocean were relating.

  El Ninos La Ninas
  Start Finish Max ONI PDO AMO Start Finish Max ONI PDO AMO
            DJF 1950 J FM 1951 -1.4 - N
T   JJA 1951   DJF 1952 0.9 - +          
   DJF 1953   DJF 1954 0.8 - + AMJ 1954  AMJ 1956 -1.6 - +
M MAM 1957   JJA 1958 1.7 + -          
M SON 1958  JFM 1959 0.6 + -          
M   JJA 1963  JFM 1964 1.2 - - AMJ 1964  DJF 1965 -0.8 - -
M  MJJ 1965 MAM 1966 1.8 - - NDJ 1967 MAM 1968 -0.8 - -
M OND 1968   MJJ 1969 1.0 - -          
T  JAS 1969   DJF 1970 0.8 N -  JJA 1970  DJF 1972 -1.3 - -
T AMJ 1972  FMA 1973 2.0 - - MJJ 1973  JJA 1974 -1.9 - -
            SON 1974 FMA 1976 -1.6 - -
T ASO 1976  JFM 1977 0.8 + -          
M ASO 1977   DJF 1978 0.8 N -          
M SON 1979  JFM 1980 0.6 + -          
T MAM 1982

 MJJ 1983 

2.1 + i SON 1984 MJJ 1985 -1.1 + -
M ASO 1986  JFM 1988 1.6 + - AMJ 1988 AMJ 1989 -1.8 - -
M MJJ 1991    JJA 1992 1.6 + -          
M SON 1994

 FMA 1995

1.0 - - JAS 1995 FMA 1996 -1.0 + +
T AMJ 1997   AMJ 1998 2.3 + +  JJA 1998 FMA 2001 -1.6 - +
M  MJJ 2002   JFM 2003 1.3 + N          
M  JJA 2004 MAM 2005 0.7 + +          
T ASO 2006   DJF 2007 1.0 - + JAS 2007  MJJ 2008 -1.4 - +
M JJA 2009 MAM 2010 1.3 N +  JJA 2010 MAM 2011 -1.4 + +
            JAS 2011 FMA 2012 -0.9 - +
T MAM 2015 NA 1.0 + N          


I need to review this table a couple of times before I draw too many conclusions as I may have made some errors mostly in terms of categorizing the period of time during an ENSO event as being PDO and AMO positive or negative. Those indices vary by month so it is tough to determine what the condition was for the duration of the ENSO event. In the left hand column for El Nino's I have designated them as Traditional (T) or Modoki (M). The literature on which ENSO events were a Modoki is not always clear and since it is a Japanese concept, the ENSO events which were considered by JAMSTEC to have been an El Nino are not 100% the same as those designated by NOAA. But the correlation is pretty close.

By my count (and over the next week I will check my work) we have had 22 El Nino Events and 14 La Nina Events since 1950 so that is about 36 events in 65 years. So ENSO events are not rare. About 15 of those El Nino events were Modokis which could be argued are closer to ENSO Neutral than being an El Nino so if one eliminates some percentage of the El Nino Modokis, you end up with perhaps about an equal number of El Ninos and La Ninas. I will look at the ocean conditions associated with these ENSO events during the week. Some believe the ocean conditions, especially the PDO, impact the ratio of El Ninos to La Ninas. Others believe that Global Warming may be the cause of so many Modokis. The three most powerful El Ninos were all traditional El Ninos as is the current one. But the current El Nino is quite different from any of the prior three champions so drawing conclusions about it is not straightforward.

Here is a list of ENSO events since 1950. The same information is shown earlier but in a different format. Notice the ONI values for a La Nina never exceed absolute value of 2.0. These divisions are arbitrary but convenient. It is not clear that impacts are proportional to the ONI. But it provides a general idea of how many El Ninos and La Ninas there have been and how they measured up on the ONI scale. Most ENSO events are weak and have little impact on weather. There have been 21 ENSO events since 1950 which have been other than weak. That is about a third of the 65 year period. Given that the most recent El Nino was a "wimp" it seems that we have not had a strong ENSO event yet this Century.

El Nino   La Nina
Weak   Moderate   Strong   Very Strong   Weak   Moderate   Strong
L-0.5  to -1   -1  to -1.5   -1.5 to -2   -2 or more negative   0.5 to 1   1 to 1.5   1.5 to 2
























































































So far I have not been able to find a comparable El Nino to compare with our current El Nino which makes it very difficult to really have an opinion as to how this is going to play out.

1982-1983 El Nino Impacts. Source

La Nina Impacts

This may be of less interest right now but a general description can be found here and the December through February impacts can be found here and the June through August impacts can be found here.

Information on different varieties of La Nina can be found here and here and here.

Additional information on La Nina can be found here. The parallel discussion of El Nino can be found here.

Focusing on India and Australia.

Both India and Australia are bordered by the Indian Ocean but Australia is bordered by both the Indian and Pacific Oceans. How does his impact their climate?

World Map

Think about currents

Worlds Currents.


ndian Ocean Dipole (IOD)

The IOD is defined as the difference between SST anomaly in a western area (60E-80E,10S-10N) and an eastern area (90E-110E,10S-0S). 

IOD Boxes

The Australian Bureau of Meteorology has their own proprietary Model (POAMA) for forecasting the IOD.

POAMA IOD Issued July 5, 2015

It comes with only a very short discussion which can be found here.

To what extent are the ENSO and IOD related? That is discussed here

And how does the IOD impact Australia? This is discussed here.

And there are variations of IOD. So if one wants to really delve into the subject you could read this article.

Indian Ocean Dipole

"The Indian Ocean Dipole (IOD) is a coupled ocean-atmosphere phenomenon in the Indian Ocean. It is normally characterized by anomalous cooling of SST in the south eastern equatorial Indian Ocean and anomalous warming of SST in the western equatorial Indian Ocean. Associated with these changes the normal convection situated over the eastern Indian Ocean warm pool shifts to the west and brings heavy rainfall over the east Africa and severe droughts/forest fires over the Indonesian region."

Thus it is easy to remember how to interpret the IOD. "+" means warm to the west and "-" means warm to the east.

Here is a discussion of how El Nino Modoli impacts the IOD.

Different impacts of various El Nino events on the Indian Ocean Dipole

Xin Wang and Chunzai  Wang

Focusing on China and Japan.

This discussion of how the impacts of El Nino and El Nino Modoki impact China and Japan can be found here.



Much information on a variety of such cycles can be found here.

Attempting to Extrapolate to the Impact of Ocean Cycles Worldwide  - a First Attempt at Putting it all Together.

This table is a first attempt at putting it all together. It is just that: a first attempt. Anyone with additional information please provide it to me by sending it to the GEI Publishers. 

Geographical Area PDO+ PDO- AMO+ AMO- Comments
Western US

Wet. High ratio of El Ninos to La Ninas

Dry. High ratio of La Ninas to El Ninos Amplifies PDO -

Amplifies PDO +

The Arctic Oscillation (A) mediates the impact of the PDO on the Northern Tier and other parts of the U.S.
Eastern US Dry Wet

Cold Winter

Dry and Warm Summer

Warm Winter

Wet and Cold Summer

AMO Negative tends to correlate but lag by several years the North American Oscillation (NAO)  which means a colder Europe and cold air intrusions into the Center of the U.S. and the North East.

AMO Positive is the opposite but the correlation with the NAO is lower.

Europe NA NA

Cold Winter

Wet and Warm Summer

Warm Winter

Dry and Cold Summer

South America West Coast impacted more Often by El Nino East Coast impacted more often by La Nina Moisture further North Wet Summer Note when the North Atlantic is Warm the South Atlantic is Cool and the Intertropical Convergence Zone (ITCZ) shifts to the south.  
West Africa NA NA Wet Summer Dry Summer Wet West African summers tend to increase hurricane activity in the U.S.
Australia Dry Wet NA NA PDO Impacts the Indian Ocean Dipole IOD so that impacts both Australia and India but probably in the opposite way as India and Australia are in different hemispheres.
India Dry Wet NA NA
Comments Arctic Oscillation (AO) very important mediator of PDO impacts in the US especially the Northrn Tier. North American Oscillation (NAO) very important mediator of AMO impacts especially for the Central and Norther Tier of the U.S. and Europe  

Attempting to put this into a two-dimensional matrix may not have been a good idea as there are not only impacts among the ocean cycles shown across the top of this matrix but also interactions with the air pressure cycles (which themselves may be related to Ocean Cycles.) So look at this matrix as just a first attempt at showing the interactions. I will be refining it.  There may be conflicts not only in the literature but also due to the interaction of so many different forcings including climate change so again this is just a starting point to think about how these different cycles impact our climate. NA perhaps should have been NS as meaning not significant. All the cycles impact everything but some have more impact than others and that is what I am attempting to show.

Atlantic Multidecadal Oscillation (AMO) Index can be found here.

It is useful to understand the Atlantic Meridional Overturning Circulation (AMOC) which if you ignore the wind component is called the Thermohaline Circulation since changes in salinity as the water moves north plays a big role. You can read about that here.

Here is the abstract from a recent paper on the AMO.  I do not believe that either this or this or this is new information but they are interesting. This is the Abstract of the McCarthy et al article.

"Decadal variability is a notable feature of the Atlantic Ocean and the climate of the regions it influences. Prominently, this is manifested in the Atlantic Multidecadal Oscillation (AMO) in sea surface temperatures. Positive (negative) phases of the AMO coincide with warmer (colder) North Atlantic sea surface temperatures. The AMO is linked with decadal climate fluctuations, such as Indian and Sahel rainfall, European summer precipitation,  Atlantic hurricanes and variations in global temperatures. It is widely believed that ocean circulation drives the phase changes of the AMO by controlling ocean heat content. However, there are no direct observations of ocean circulation of sufficient length to support this, leading to questions about whether the AMO is controlled from another source. Here we provide observational evidence of the widely hypothesized link between ocean circulation and the AMO. We take a new approach, using sea level along the east coast of the United States to estimate ocean circulation on decadal timescales. We show that ocean circulation responds to the first mode of Atlantic atmospheric forcing, the North Atlantic Oscillation, through circulation changes between the subtropical and subpolar gyres—the intergyre region. These circulation changes affect the decadal evolution of North Atlantic heat content and, consequently, the phases of the AMO. The Atlantic overturning circulation is declining and the AMO is moving to a negative phase. This may offer a brief respite from the persistent rise of global temperatures, but in the coupled system we describe, there are compensating effects. In this case, the negative AMO is associated with a continued acceleration of sea-level rise along the northeast coast of the United States."

Another way at looking at things is to think about the models of variablity in the North Atlantic. This paper by Iris Grossmann and Philip J. Klotzbach can be found here.

This is a good time to discuss the Atlantic Multidecadal Oscillation or AMO. It has a big impact on CONUS summer weather especially precipitation and we will be discussing this more in the weeks ahead. This graphic is from the paper Variations in North American Summer Precipitation Driven by the Atlantic Multidecadal Oscillation QI HU,SONG FENG, AND ROBERT J. OGLESBY.  Here is the link. We will be discussing this and related papers a lot more in the coming weeks.


You can see that the patterns are kind of the opposite of each other. One expects the AMO to be moving towards neutral but it has not just yet. The Index itself is measured from 0 to 60N, 7.5W to 75W. Here is the AMO Index. It is a bit more difficult to pin down but there is a related cycle: the North Atlantic Oscillation (NAO). Most pay attention to the high-frequency seasonal variations in this oscillation but there is a low-frequency component also. Here is a good reference link. The Scandinavians pay more attention to this cycle for obvious reasons.

Pacific Decadal Index (PDO) can be found here. The original research paper describing the PDO (even before it was named) can be found here. This is an index that covers both the north and south Pacific and is called the Interdecadal Pacific Oscillation Index.

At this point it might be useful to discuss the PDO. The below shows the different pattern of where the surface water is warm and where it is cool in the Pacific during two most discussed phase of the PDO. The graphic on the left is PDO+ and notice how well it conforms to where the models say we will be in the second quarter of 2015. Actually that is about where we are today. It is very different from conditions since mid 1998.

PDO Positive ------------------------------------------------- PDO Negative

PDO Phase Graphics

The seminal work on the impact of the PDO and AMO on U.S. climate can be found here. And here is a later version but I do not have a link that shows it in color but I believe the maps have not changed from the earlier version. 

The key maps are shown below:

-McCabe Drought AMO/PDO

Drought frequency (in percent of years) for positive and negative regimes of the PDO and AMO. (A) Positive PDO, negative AMO. (B) Negative PDO, negative AMO. (C) Positive PDO, positive AMO. (D) Negative PDO, positive AMO.

As of the time that I have last updated this document, September 29, 2014, the PDO Index has become positive but slowly declining back towards normal and the AMO Index flirted with going negative but is currently again positive i.e. the situation is currently PDO + (warm)/ AMO + (warm) or consistent with the C regime in the above McCabe et al graphic.  Clearly this is a situation that requires close monitoring. The current values of the PDO and AMO may not be nearly as significant as values that generally remain the same sign for a lengthy period of time. 

Here is another approach that is worldwide in nature but only focuses on the Pacific Ocean.

Currently an issue of great importance is whether or not the pattern of warmer areas versus colder areas in the Pacific has changed. Such changes are often referred to as a Pacific Climate Shift. There is some discussion that we may be either in a Pacific Climate Shift now or that one may be imminent.

Recently I referenced this article and this article which address this subject but did not discuss them. I thought I would need to discuss both but it turns out that the first link covers both the 1977 - 1998 Positive Phase of the PDO and the 1947 - 1976 Negative Phase of the PDO. So I will focus on just that first paper: "Tropical Pacific Forcing of a 1998–1999 Climate Shift:  Observational Analysis and Climate Model Results for the Boreal Spring Season  Bradfield Lyon,  Anthony G. Barnston, David G. DeWitt". What is very unusual is that these authors are addressing the Northern Hemisphere Spring/Southern Hemisphere Fall while most climate researchers focus on winter or summer. So this is unusual and thus to me especially interesting. 

One complication in presenting this material is that I am most comfortable characterizing the condition of the Pacific using a particular index the Pacific Decadal Oscillation or PDO. The PDO refers to the Northern Hemisphere but there is a related index called the Interdecadal Pacific Oscillation (IPO) that covers both Hemispheres. As I discuss the Lyon et al. paper you will see that the authors do not rely on any of these indices to aggregate data for their graphics since basically their methodology recreates them. I am using the terms PDO and IPO in my discussion as do the authors in their article to provide a frame of reference that fits with other information I have presented.

Most of the findings in that paper that I want to discuss show up in the below graphic. Unfortunately, unless you have a color printer you need to view this on the screen. It is a very simple graphic really. The authors have removed the impact of Global Warming and the ENSO cycle. So they really are attempting to capture the impact on precipitation of the changes from one Pacific Ocean configuration to another. For purposes of relating the discussion to the U.S. I have taken the liberty in my comments to refer to the periods of time they are comparing as being periods which are considered to have been PDO Positive or PDO Negative. For certain readers a reference to the IPO would be perhaps more recognizable but the PDO and IPO are highly correlated. In the graphic the authors just display their data in terms of periods of time but those periods can be described by the PDO index and (as well as the IPO index) I have done that.

With respect to the below graphic,

Figure 4a compares precipitation since the PDO went Negative to the period of time (1977 - 1998) when it was Positive.

Figure 4b compares 1977-1998 with 1947-1976 the prior negative configuration of the Pacific.

So in both cases the goal is to compare a period of PDO positive with a period of PDO negative and since it is looking at the entire world, the IPO is used not the PDO. The IPO covers the entire Pacific not just the Northern Hemisphere. Again, the authors discuss the IPO but do not use it as part of their methodology as their methodology is the methodology used to define the IPO.

Figure 4c shows the part of the results which are statistically significant at the 10% level.

A value of this analysis is that it not unreasonable to anticipate the reversal of the colors on the chart 4c if the PDO has indeed gone positive. Recently, there have been very high readings of this index but I am not at all convinced that the climate shift has taken place and in fact I think it is too soon and that the recent high readings are a result of the near El Nino Modoki Type II in the Pacific. But that is speculation on my part since I think that most people who are knowledgeable about oceans cycles are fairly convinced that they are easier to identify after they occur than to predict. But a switch to PDO+ will occur sooner rather than later. Later in this report, there is a discussion where I discuss the logical conclusion as to when the PDO will shift based on the work of some other researchers. I have also made my own assessment of when the Pacific will shift. My analysis is purely statistical and I arrive at an estimate of 5 - 20 years with perhaps 10 to 15 years being the most likely time for the next peak of the PDO Index. Some believe the shift has already begun.

It is important to remember that the PDO and IPO are oscillations not regular cycles and most likely represent three or more separate phenomena. The Aleutian Low may have its own cycle. The Kuroshio-Oyashio Extension (KOE) may exhibit cyclical behavior. ENSO is a well known cycle. Some of these cycles may be impacted by the AMO. There is a South Atlantic version of the AMO also. So the PDO and IPO may be a combination of cycles with different wavelengths. So a change in the Pacific may take many different forms and occur in an irregular manner. The below graphic shows the differences over a multi-year period. One could view the analysis by year and see how it is impacted by the ebb and flow of the situation in the Pacific but it would be difficult to interpret that amount of data so this approach is easier to interpret but we should recognize that the conditions in the Pacific are not binary but vary and not in a totally regular way.

Climate Shift LyonFigure 4

The following is from the Summary and Conclusions Section of their paper:

"Observationally-based analyses and reanalysis products have been used to document a multidecadal shift in Pacific SSTs in 1998–1999 and associated atmospheric changes that are akin to the shift of 1976–1977. Emphasis is on the 3-month season of MAM, which has received only modest attention from previous investigators examining related multidecadal SST variations. The motivation for the study was to examine in greater detail the abrupt decline in MAM East African rainfall reported by LD that occurred in 1999 in order to view those results in a global domain and on multidecadal timescales. An EOF analysis of the residual anomalies of SSTs since 1900, determined by removing the (simultaneous) effect of ENSO and the global warming signal from the data by linear regression, revealed a loading pattern in the Pacific generally similar to the PDO. The associated PC time series shows a shift in 1999 indicative of, among other features, a change in the background state of Pacific SSTs towards cooler than average conditions in the east-central tropics. A composite difference of the full SST field in fact shows an average cooling of over 0.5 for the period 1999–2012 relative to 1977–1998 in that region. The PC time series of the residual SST EOF reveals earlier shifts in 1925, 1946 and 1976, which are all consistent with previously reported shifts in the PDO (e.g., Mantua et al. 1997)

As expected, the precipitation response to the 1999 shift in SST is large scale, with statistically significant changes in MAM season totals for the post-1998 period seen in multiple locations including drying over East Africa and central-southwest Asia, coastal regions of southeastern China, parts of northeastern Australia and the southwestern US. Drier conditions are also seen over the central Indian Ocean and the east-central Pacific Ocean. Wetter conditions include a zonally elongated band across the northern Indian Ocean extending from the eastern Arabian Sea eastward across southern India to the Philippines, the western tropical Pacific and northern South America. A southwestward displacement of the SPCZ is also identified.

Atmospheric circulation changes during MAM at 850 hPa [Editor's note the following table may help in relating to Air Pressure levels. 850 hPa is generally used to look at frontal movements, 500 hPa (which is pretty much the average for half the air being below and half above this level of air pressure) is often considered a good way to look at steering currents and the area of about 250 hPa is where the Jet Stream tends to be most pronounced. These are all generalizations but useful and provide a way compare the findings of different researchers] associated with the recent shift are observed

air pressure and altitude

in both hemispheres and include an anomalous anticyclonic circulation southeast of the Aleutians [Editors Note: Is this the RRR?], enhanced easterlies in the equatorial Pacific and subtropics in both hemispheres and anomalous westerlies across the northern Indian Ocean. This latter feature, also identified by LD, is consistent with an earlier onset of the South Asian monsoon in recent years (Kajikawa and Wang 2012; Xiang and Wang 2013 ), with Xiang and Wang (2013) suggesting it is forced by Rossby waves emanating from the anomalous rainfall in the western tropical Pacific. Near-surface wind changes in the tropical Pacific are generally similar to those reported by Hu et al. (2013). The PC time series associated with a joint EOF analysis of the anomalous 850 hPa wind components in 20CR going back to 1930 indicates a similar circulation pattern was present during the previous Pacific cool phase over the northern Indian Ocean, further supporting the notion of a decadal modulation of South Asian summer monsoon onset. At 200 hPa, anomalous stationary wave features included a PNA-like pattern in the northern hemisphere, with suggested wave propagation from the east-central Pacific region into the southern hemisphere as well. An anomalous wave train emanating from the western tropical Pacific into both hemispheres is also identified, most likely forced by the anomalous rainfall and associated diabatic heating in the western tropical Pacific. The PC time series associated with the EOF analyses of both the 850 hPa wind field and 200 hPa anomalous stationary waves both show an abrupt and statistically significant shift in 1999 in R2. [Editor's note: R2 refers to a particular data series]

Model simulations using the ECHAM4.5 atmospheric general circulation model forced with globally observed SSTs (‘‘full simulation’’) generally capture most of the observed precipitation and atmospheric circulation features associated with the recent shift. While both the full and POGA simulations tended to show greater interannual variability associated with ENSO than in observations during the post-1998 period, based on an EOF analysis both sets of model experiments also produced a statistically significant shift during 1999 that is consistent with observations. POGA runs of the ECHAM4.5 forced with observed SSTs only in the tropical Pacific were also able to capture most of the main features associated with the recent shift and generated more realistic precipitation changes over the northern Indian Ocean and southern Africa compared with the full simulations. This result points to the fundamental role played by the tropical Pacific in forcing the observed atmospheric conditions, which is consistent with the conclusions drawn by previous studies examining the 1976–1977 shift (e.g., Graham 1994; Trenberth and Hurrell 1994; Deser et al. 2004).

Overall this study has documented a recent shift in Pacific SSTs, its connection with multidecadal variability and its influence on regional climate changes around the globe during the MAM season. While the SST loading pattern identified in the EOF analysis of the residual SST field (i.e., after linearly removing ENSO and the global warming signal) resembles the behavior of the PDO, emphasis is not placed here on strict definitions of low frequency patterns of Pacific SSTs. The results presented would also map onto the interdecadal Pacific Oscillation (IPO) as described by Power et al. (1999), for example. And regardless of the analysis method used to identify it, we do not address the issue of causality of the observed, multidecadal variability and the recent associated shift. Causality is a fundamental question that goes beyond the scope of this study, with multiple hypotheses put forward in the scientific literature. On the one hand, a null hypothesis positing that such variability can be explained as red noise processes alone [Editor's Note: I believe this is the way climatologists describe Brownian Noise i.e. randomness - in a very loose way], including the reddening of the ENSO signal (Newman et al. 2003) has been advanced, while alternatively, various physical mechanisms have also been suggested, many contained in the recent review by Liu (2012). From a practical perspective, however, there is useful information in recognizing that the recent shift has occurred, having consequences for regional precipitation variability in several regions of the globe.

As the original motivation for the study was to examine in greater detail the abrupt decline in MAM East African rainfall in 1999, here the observational evidence, climate model full simulations and POGA run experiments all support the conclusion that this recent decline is directly related to multidecadal Pacific SST variability, with the forcing coming from the tropical Pacific. More generally, further work is needed to elucidate the physical mechanisms at work in generating the regional climate variations that have occurred around the globe during the post-1998 period."


How Might we Predict the Future State of the PDO and AMO and other Ocean Cycles?

A new paper is out. Unfortunately I can only access the abstract. Some with subscriptions may find the full paper here.

Klöwer et al. (2014) Atlantic meridional overturning circulation and the prediction of North Atlantic sea surface temperature.

• North Atlantic sea surface temperature exhibits high decadal predictability potential.
• Model bias hinders exploiting the decadal predictability potential.
• An innovative method was developed to overcome some of the bias problem.
• North Atlantic sea surface temperature will stay anomalously warm until about 2030.

The Atlantic Meridional Overturning Circulation (AMOC), a major current system in the Atlantic Ocean, is thought to be an important driver of climate variability, both regionally and globally and on a large range of time scales from decadal to centennial and even longer. Measurements to monitor the AMOC strength have only started in 2004, which is too short to investigate its link to long-term climate variability. Here the surface heat flux-driven part of the AMOC during 1900–2010 is reconstructed from the history of the North Atlantic Oscillation, the most energetic mode of internal atmospheric variability in the Atlantic sector. The decadal variations of the AMOC obtained in that way are shown to precede the observed decadal variations in basin-wide North Atlantic sea surface temperature (SST), known as the Atlantic Multidecadal Oscillation (AMO) which strongly impacts societally important quantities such as Atlantic hurricane activity and Sahel rainfall. The future evolution of the AMO is forecast using the AMOC reconstructed up to 2010. The present warm phase of the AMO is predicted to continue until the end of the next decade, but with a negative tendency.

The conclusions of this paper are more or less in line with the recent IPCC AR5 WG1 Report which failed to detect appreciable slowing of the North Atlantic Meridional Overturning Circulation (NAMOC). Most believe that this means that we have not yet seen the lengthening of the AMO. At this point it is useful to mention that most refer to these ocean cycles as oscillations as they are not as regular as the shape of a true cycle but the choice of terminology is a matter of style as the "business cycle" is also not very regular but we use the term cycle to describe it. I believe that if we become more comfortable that the AMO and PDO are regular occurrences they will increasingly be referred to as cycles.

And then there is this paper.

On the observed relationship between the Pacific Decadal Oscillation and the Atlantic Multi-decadal Oscillation Department of Atmospheric and Oceanic Sciences, University of Wisconsin-Madison, Madison, USA; Laboratory of Climate, Ocean-Atmosphere Studies, Department of Atmospheric and Oceanic Science, Peking University, Beijing, China

Journal of Oceanography (Impact Factor: 1.46). 01/2011; 67(1):27-35. DOI: 10.1007/s10872-011-0003-x

ABSTRACT We studied the relationship between the dominant patterns of sea surface temperature (SST) variability in the North Pacific
and the North Atlantic. The patterns are known as the Pacific Decadal Oscillation (PDO) and the Atlantic Multi-decadal Oscillation (AMO). In the analysis we used two different observational data sets for SST. Due to the high degree of serial correlation in the PDO and AMO time series, various tests were carried out to assess the significance of the correlations. The results demonstrated that the correlations are significant when the PDO leads the AMO by 1 year and when the AMO leads the PDO by 11–12years. The possible physical processes involved are discussed, along with their potential implication for decadal prediction."

Another cycle of interest is the East Asian Surface Air Temperature (EATS).  This paper is of interest.

And here is a similar paper published in 2007.

A third paper brings the East Asian Surface Air Temperatures (EATS) into the equation. It is not just about the U.S. This paper can be found here.

From the Abstract:

"The authors analyzed the lead-lag connection of the Atlantic Multidecadal Oscillation (AMO) with East Asian surface air temperatures (EATs) using instrumental records, and compared the results with the Pacific Decadal Oscillation (PDO). The maximum correlation was found when EATs led the AMO by five to seven years (with a correlation coefficient of 0.72, whereas the correlation coefficient was −0.91 when the AMO led EATs by 24–28 years). This is different from the PDO, which mostly correlated with EATs when the PDO led EATs by 13–15 years (with a correlation coefficient of 0.67, whereas the correlation coefficient was −0.76 when EATs led the PDO by 24–26 years). The PDO led the AMO by 19–21 years (with a correlation coefficient of 0.71, whereas the correlation coefficient was −0.84 when the AMO led the PDO by 16–18 years). These results support a previous understanding that EATs positively correlate with the AMO, and imply that the observed East Asian warming trend may have been slowing down since the early 2010s."

Also of interest.

EATS Table 1 Revised

The top of this Figure 1 show the three ocean cycles. It may not be easy to find another such graphic. The bottom shows the results of the correlation of the various lags. I added to the Legend at the top (the bold type) to make it clearer and hopefully I did this correctly. I will provide what I did to the Author and if I got it wrong I assume I will be so informed. This is easier to undertand by looking at the following table.

EATs Lag Table 1

I said it would be easier. I did not say it would be easy. But you can see that these are pretty hefty correlation coefficients. So one has a fighting chance at figuring out what the climate is going to be like over the next 100 years by region. The major problem is the limited amount of data that was available to do this correlation analysis and you can see that from Figure 1 above. It is the reason that the leads and lags computed in this paper for the AMO and PDO different slightly from those computed in the papers I presented two weeks ago. It is also quite possibly the reason that the IPCC has not yet included these low frequency ocean cycles in their climate models. There really is not enough data to do so. Thus the IPCC climate models are essentially useless for predicting climate for periods of time less than 100 years and yet many do so and then are surprised when the projected and observed levels of temperature and precipitation diverge. If you do not include ocean cycles in a forecasting model it will not incorporated "internal variability".

Now I attempt to put the information from the three papers together and this is what I came up with. This table relates the AMO to the PDO. So "Lead" means the AMO reaches a peak or minimum and begins to trend in the other direction in advance of the PDO beginning to change its direction (if it is positive begin to become less positive and then become negative or vice versa)."Lag" means the AMO chances direction later than the PDO which could be expressed as the PDO leads the AMO. So yes this is a difficult concept to grasp when looking at a table of numbers. For me it is more understandable when I look at a graph of the time series of the amplitudes of the surface temperature anomalies of the AMO and PDO. 

AMO relationship to the PDO re Leads and Lags Lead Lag

Sum of Absolute Values of the Lead and Lag

Twice the number in the prior column
Orgeville and Peltier -13 +17 30 60
Wu, Liu, Zhang and Delworth* 12 to 14 12 to 13 26 52
Li and Luo -(16 to 18) 19 to 21 37 74

* This paper decomposes the PDO into a high Frequency 20 year cycle and a low frequency presumably 60 year cycle. The above data is for the Low-Frequency Cycle and in this paper the authors also report a high correlation for the PDO lagging the AMO by 1 - 3 years. Although I show a positive correlation for this paper in the above table in regards to the AMO leading the PDO it may well be that I have not read the paper carefully enough and that the correct sign of that correlation is also negative. So I would not conclude that this paper is inconsistent with the other two. 

In all cases where both low frequency and high frequency cycles were studied and reported on, I have tabulated the results with respect to the low- frequency long component of the cycle. Note that similar decompositions into sub-cycles within the most familiar duration of these cycles are done by most researchers and not just for the PDO but many believe the PDO has both a high frequency approximately 20 year cycle (possibly ENSO related) and a low frequency approximately 60 year cycle. This is possibly the reason the correlation coefficients are not totally impressive.

To better understand "leads" and "lags" I now present another graphic from the Li et al paper.

AMO and PDO shifted back by 17 years

Here the authors plotted the AMO versus the PDO which was shifted back by 17 years. Notice when you do this the AMO and PDO are always in a different phase. On the other hand if you had moved the PDO ahead by 20 years they would be in the same phase. Thus the sign of the correlation coefficient may be a function of how the relationship is described e.g. if A leads B by N years, B lags A by the same N years. So the choice of terminology may be somewhat arbitrary.

Much more important if the two cycles are 74 years in duration (as seems to be indicated by the calculations performed by the Li and Luo paper), or if you prefer, 17 years is not very different than 20 years, it would appear that the four conditions:

  • AMO+/PDO+
  • AMO+/PDO-
  • AMO-/PDO+
  • AMO-/PDO-

occur approximately an equal percentage of the time i.e. 25% for each combination.This is a very important conclusion, if correct, and I believe it is supported by all three papers. This makes the McCabe et al (and I do not want to not mention Julio Betancourt who was instrumental in that work) analyis of drought probability shown above that much more valuable.

Predicting the State of Ocean Cycles Going Forward and the Impact on the Climate of the Lower 48 States.

The paper I want to discuss can be found here.

Joint statistical-dynamical approach to decadal prediction
of East Asian surface air temperature
LUO FeiFei & LI ShuangLin

First I want to begin with some information from this paper which explains the three most important ocean cycles. 

"The Atlantic Multidecadal Oscillation (AMO) is a leading fluctuation pattern of SST in the North Atlantic region, which has been linked to the Meridional Overturning Circulation (MOC) in many models (Delworth and Mann, 2000; Enfield et al., 2001; Sutton and Hodson, 2005, 2007; Knight et al., 2005). The AMO has a period of 65 years and an amplitude of 0.4°C in instrumental records (Delworth and Mann, 2000; Enfield et al., 2001; Sutton and Hodson, 2005). Following earlier AMO definitions (Enfield et al., 2001; Wang et al., 2009), the AMO index was defined as the annual averaged low-frequency SST anomaly (SSTA) in the North Atlantic basin (0°-60°N, 75°-7.5°W). The temporal evolution of AMO index (dashed line in Figure 1(a)) is characterized by two cold phases from the early-1900s to the late-1920s and from the mid-1960s to the 1990s, and two warm phases from the 1930s to the 1960s and from the mid-1990s until now. These are in agreement with previous studies (Delworth and Mann, 2000; Enfield et al., 2001; Sutton and Hodson, 2005, 2007; Knight et al., 2005; Wang et al., 2009). Figure 2(a) shows that the AMO explains approximately 50% of the internal decadal variance of SSTs over the North Atlantic. In particular, the proportions of variance are greater than 70% over the tropical and eastern North Atlantic. Besides, the proportion is approximately 50% in the western tropical Pacific.

Editor's Note: I need to understand this better but I believe that although the IPO covers both the North Pacific and South Pacific, the main difference is that the PDO excludes the tropics. Thus the IPO is more influenced by the ENSO Cycle. It thus might be a better index although it certainly was not used by McCabe et al so that needs to be considered. On the other hand, depending on the smoothing algorithm used, the IPO might be very similar to the PDO. In this paper only the low frequency part of the cycle is utilized so that means that ENSO has probably been removed. At any rate I am using the authors' analysis of the IPO as a surrogate for the PDO. 

The Inter-decadal Pacific Oscillation (IPO) is the decadal climate variability in the leading Pacific SST pattern. During the positive phase, the IPO is marked mainly by cool anomalies over the mid-high latitudes of the South and North Pacific, while it is warm along the west coast of the United States and the tropical Pacific (Power et al., 1999; Folland et al., 1999). The situation is opposite in the negative phase. The IPO index is defined as the time series of the first empirical orthogonal function (EOF) of SSTAs in the Pacific basin (60°S-60°N, 120°E-80°W) (Power et al., 1999; Folland et al., 1999). The first mode accounts for 43% of Pacific SSTs. The IPO (dashed line in Figure 1(b)) experiences one cold period from the mid-1940s to the mid-1970s, and two warm periods from the 1920s to 1940s and from the 1970s to 1990s. The warm phase of the IPO shifts to a cold phase at the beginning of the 21st century. This is consistent with previous results (Folland et al., 1999).The IPO is the most important in the tropical mid-eastern Pacific, explaining 60% of the variance. In the mid-high latitudes of the Pacific, the proportion is approximately 50% ).

The Indian Ocean Basin-wide Decadal (IOBD) pattern is characterized by a uniform basin-wide warming or cooling in the Indian Ocean on the decadal timescale (Allan et al., 1995; Li et al., 2012). The IOBD pattern is defined as the leading EOF of SSTAs over the Indian Ocean (20°S-25°N, 35°-120°E), which explains 70% of the total variance. From the dashed line in Figure 1(c), it can be seen that the IOBD pattern has a warm period from the 1870s to the late 1880s, and subsequently moves into a cold phase with a weak amplitude in the 1890s until the 1950s. The IOBD pattern remains in a cold phase from the 1950s to 1980s, and thereafter shifts to a warm phase. Figure 2(c) shows that the IOBD pattern is more important over the tropical Indian Ocean and western Pacific (eastern Philippine Sea) and explains approximately 60% of the variance. Accordingly, the AMO, IPO, and IOBD pattern represent the leading mode of the internal decadal variance of SST in the North Atlantic, Pacific, and Indian Ocean, respectively. The sum of the proportions of variance of the three modes is greater than 90% over most of the North Atlantic, tropical mid-eastern Pacific, Indian Ocean, and parts of the western Pacific."

Notice that each of the three ocean cycles tend to also influence sea surface temperatures (SST) in the other two oceans.  Thus it is reasonable to conclude that the combination of the phases of the three ocean cycles impact the climate of various parts of the World.

Recently I Ialked about North America. Although the climate of North American may be influenced by all three ocean cycles as far as I know only the impact of the North Atlantic and the North Pacific has been carefully studied. So that is what I will focus on in this weekly report. But we should not forget that some of the variance observed by McCabe et al in their work might have been explained by the IOBD if it had been included. 

The following Graphic from the Luo and Li paper is very interesting.

ShaungLinLi Ocean Cycle Decadal Projections

Now I have modified the above to line up some key dates in the graphics namely the beginning of decades starting from the one we are in and looking forward. The decades are numbered across the top as "1", "2", "3", "4".

ShaungLinLi Ocean Cycle Projections 10 Year Increments with Numbers

The seminal work on the impact of the PDO and AMO on U.S. climate can be found here. The key maps which were presented earlier are repeated below:

 McCabe Maps modified to include the subtitles

You may have to squint but the drought probabilities are shown on the map and also indicated by the color coding with shades of red indicating higher than 25% of the years are drought years (25% or less of average precipitation for that area) and shades of blue indicating less than 25% of the years are drought years. Thus drought is defined as the condition that occurs 25% of the time and this ties in nicely with two combinations of AMO and PDO phases i.e. four combinations.

In reality there are more than four combinations since for example AMO positive can cover years where the index is just barely positive to the years where the index is most positive. That may be possible to deal with using a software package but working with maps averaging the years where the indicated combination occurred makes for a more convenient way to display information but does not reflect that there is a continuum of combinations of two indices.

So combining the work of McCabe et al with the work of Lou and Li , what can we say about these beginning years in the decade which started in 2014 and subsequent decades out into our future?

1. 2010

This was clearly AMO+/PDO-. This is McCabe Condition D and full explains the approximately twice a century severe drought in the Southwest and also in the Great Lakes area.  

2. 2020

This might be PDO+/AMO+ to AMO Neutral. This might be somewhere between McCabe A and McCabe C but closer to McCabe C. McCabe C is associated with extreme drought in the Northern Tier west of the Great Lakes. McCabe C also is associated with a high probability of drought in the South Atlantic Coast states extending into the Mid-West.

3. 2030

PDO-/AMO Neutral. This might be somewhere between McCabe B and McCabe D. Note the IPO (which I am using as a surrogate for the PDO) is forecast by Lou and Li to be less negative at its low point presumably due to the combination of subcycles with different amplitudes within the IPO. Also notice the significant difference between McCabe B and McCabe D. This illustrates the importance of the Atlantic SST's

4. 2040

PDO+/AMO Neutral. This might be somewhere between McCabe A and McCabe C.  McCabe C seems to flip drought in the West from south to north. So McCabe A and McCabe B both spare the Southwest but the condition of the Atlantic may have a big impact on the Northern Tier west of the Great Lakes.  McCabe C also is associated with a high probability of drought in the South Atlantic Coast states extending into the Mid-West.

5. And dare we project out to say 2050? This might be PDO?/AMO-. That might result in conditions intermediate to McCabe A and McCabe B. McCabe B would raise the risk for drought in the Southeast including moderate probability of drought in Florida and small areas of extreme drought in other places.  


  • There is no intention of suggesting that the drought probability for a particular year in a particular geographic area can be predicted accurately with this methodology. It is intended to provided a reasonable scenario of how these probabilities might vary over time. This can be very useful for water planning. 
  • The IPO is not the PDO but they are highly correlated.
  • The authors show signficant confidence intervals around their projections. i.e. the timing of the peaks and valleys of these cycles could fall in the shaded area of the graphic rather than exactly as shown.
  • Other authors might project these cycles slightly differently
  • I believe the authors have taken Global Warming into account. But RCP 4.5 may be considered by some as being too conservative. 
  • The McCabe et all maps are maps of drought frequency. I am assumng they are highly inversely correlated with amounts of precipitation.

I have not tried to interpret the implications for each state in the Lower 48. The McCabe et al analysis is a mathematical analysis based on historical data. Underlying the data are climate patterns determined by mountains and interactions with Canada and Mexico. If I better understood all of these patterns I would feel more confident in making predictions state by state. I am hoping that my experiment with combining the emerging information on ocean cycles with the work of McCabe et al and similar will inspire others to undertake more work in this area. 

Now let us focus on the AMO.

This is a good time to discuss the Atlantic Decadal Oscillation or AMO. It has big impact on CONUS summer weather especially precipitation and we will be disussing this more in the weeks ahead. This is from the paper Variations in North American Summer Precipitation Driven by the Atlantic Multidecadal Oscillation QI HU,SONG FENG, AND ROBERT J. OGLESBY discussed below. But here is the link if you want to reference it now.


You can  see that the patterns are kind of the opposite of each other. One expects the AMO to be moving towards neutral but is has not just yet. The Index itself is measured from 0–60N, 7.5–75W.  Here is the AMO Index.

First this paper focusing on winter impacts:

Imprint of the Atlantic multi-decadal oscillation and Pacific decadal oscillation on southwestern US climate: past, present, and future Petr Chylek, Manvendra K. Dubey, Glen Lesins, Jiangnan Li, Nicolas Hengartner


The surface air temperature increase in the southwestern United States was much larger during the last few decades than the increase in the global mean. While the global temperature increased by about 0.5° C from 1975 to 2000, the southwestern US temperature increased by about 2° C. If such an enhanced warming persisted for the next few decades, the southwestern US would suffer devastating consequences. To identify major drivers of southwestern climate change we perform a multiple-linear regression of the past 100 years of the southwestern US temperature and precipitation. We find that in the early twentieth century the warming was dominated by a positive phase of the Atlantic multi-decadal oscillation (AMO) with minor contributions from increasing solar irradiance and concentration of greenhouse gases. The late twentieth century warming was about equally influenced by increasing concentration of atmospheric greenhouse gases (GHGs) and a positive phase of the AMO. The current southwestern US drought is associated with a near maximum AMO index occurring nearly simultaneously with a minimum in the Pacific decadal oscillation (PDO) index. A similar situation occurred in mid-1950s when precipitation reached its minimum within the instrumental records. If future atmospheric concentrations of GHGs increase according to the IPCC scenarios (Solomon et al. in Climate change 2007: working group I. The Physical Science Basis, Cambridge, 996 pp, 2007), climate models project a fast rate of southwestern warming accompanied by devastating droughts (Seager et al. in Science 316:1181–1184, 2007; Williams et al. in Nat Clim Chang, 2012). However, the current climate models have not been able to predict the behavior of the AMO and PDO indices. The regression model does support the climate models (CMIP3 and CMIP5 AOGCMs) projections of a much warmer and drier southwestern US only if the AMO changes its 1,000 years cyclic behavior and instead continues to rise close to its 1975–2000 rate. If the AMO continues its quasi-cyclic behavior the US SW temperature should remain stable and the precipitation should significantly increase during the next few decades."

Special Note: This paper by Chylek et al. explains a common error in many climate studies including a recent analysis performed by the U.S. Bureau of Reclamation titled Upper Rio Grande Climate Assessment. It is likely that his sort of error is very prevalent in many reports because the researchers are not sufficiently familiar with what the IPCC calls internal variability. Recent focus on what has been called the Warming Hiatus has called attention for the need of researchers to become familiar with the impact of the major ocean cycles on climate or alternatively at least detrend the temperature data.

The Chylek et al paper is focused primarily on winter climate in the Southwest and to me it in many ways is a redo of the McCabe et al analysis but I do not believe the authors see it that way based on their reference list. I do believe their findings are similar which does not come as a surprise.

There are two theories on what controls summer precipitation. This paper is a good explanation of how summer precipitation is controlled by the Atlantic and in particular the AMO.

Variations in North American Summer Precipitation Driven by the Atlantic Multidecadal Oscillation QI HU,SONG FENG, AND ROBERT J. OGLESBY

"ABSTRACT (Excerpt)

During the warm phase, the North Atlantic subtropical high pressure system (NASH) (Editors Note: many refer to NASH as either the Bermuda High or the Azores High) weakens, and the North American continent is much less influenced by it. A massive body of warm air develops over the heated land in North America from June–August, associated with high temperature and low pressure anomalies in the lower troposphere and high pressure anomalies in the upper troposphere. In contrast,during the cold phase of the AMO, the North American continent, particularly to the west, is much more influenced by an enhanced NASH. Cooler temperatures and high pressure anomalies prevail in the lower troposphere, and a frontal zone forms in the upper troposphere. These different circulation anomalies further induce a three-cell circulation anomaly pattern over North America in the warm and cold phases of the AMO. In particular, during the cold phase, the three-cell circulation anomaly pattern features a broad region of anomalous low-level southerly flow from the Gulf of Mexico into the U.S. Great Plains. Superimposed with an upper-troposphere front, more frequent summertime storms develop and excess precipitation occurs over most of North America. A nearly reversed condition occurs during the warm phase of the AMO, yielding drier conditions in North America. This new understanding provides a foundation for further study and better prediction of the variations of North American summer precipitation, especially when modulated by other multidecadal variations—for example, the Pacific decadal oscillation and interannual variations associated with the ENSO and the Arctic Oscillation."

Some of the graphics from this paper are extremely interesting so I am showing them. You can enlarge them here and here.  The way to look at this graphic is that the situation for the AMO being in its warm phase is shown on the left and the situation for the AMO being in its cool phase is shown on the right. The top graphic represents the upper troposphere and the bottom graphic the lower troposphere.

Variations in Summer Precipitation Hu et al

It is amazing to me that the patterns in the upper and lower troposphere can be so different but the easiest explanation perhaps is that the stronger/larger/and 20 degrees of longitude further west Bermuda High (also known as NASH, Azores High) associated with AMO- results in the jet stream being shifted further north and creating a counter-clockwise/cyclonic curvature which is conducive to storm formation.

The AMO has been positive since the mid nineties and peaked around 2010 or a bit earlier and is now pretty much neutral but we will be in AMO negative soon and for a long time. It might be useful to better understand how the Atlantic Multidecadal Oscillation presents itself and the following table presents some information which comes from an excellent paper by Iris Grossmann and Philip J. Klotzbach which can be found here.

It is only one paper so it may not be correct but it certainly is provocative. Between the two papers it points out the need to be able to either:

A. Forecast the AMO and PDO or

B. Develop a set of reasonable scenarios for how they might vary over time.

I have done that a little earlier in this report so therefor: I DO NOT KNOW HOW TO MAKE IT ANY CLEARER. THE AMO WILL BE BECOMING LESS POSITIVE AND SHIFT TO AMO NEGATIVE AND THE PDO WILL SOON BECOME PDO POSITIVE IF IT HAS NOT ALREADY DONE SO. The AMO has been positive since the mid nineties and peaked around 2010 or a bit earlier and is now pretty much neutral but we will be in AMO negative soon and for a long time. It might be useful to better understand how the Atlantic Multidecadal Oscillation presents itself and the following table presents some information which comes from the  Iris Grossmann and Philip J. Klotzbach paper.

The AMO and its Manifestations
NAO   Positive + Negative -
AMO   Negative - Positive +
Years (will vary among researchers)   1990-1925, 1971-1994 1875-1899, 1926-1970, 1995-Present
sub-Polar Area*   Cold Warm
West Atlantic* 20N - 45N Warm Cold
East Atlantic* 0 - 30N Cold Warm

* These locations are really associated with the NAO rather than the AMO but I think they are fairly similar and I will continue to research this. But the key point is that both the AMO and NAO tend to show up as a tripole i.e. three different areas of temperature anomalies. At any rate it is more complicated than some presentations that describe it as a uniform warming or cooling of the Atlantic. There is a third cycle called the Atlantic Meridional Mode AMM which relates mostly to the differential strength of the Trade Winds north and south of the Equator. One can understand why Climate Change models have difficulty incorporating these ocean and air pressure cycles. Of course that means these models are not very reliable for forecasting intervals shorter than 100 years.

East Asian Surface Temperature Oscillation (EATs)

Now that I have introduced the topic of the East Asian Surface Temperature Cycle, this paper is of interest.

Is El Nino and the AO or NAO correlated to any significant extent?

Beginning with this article.  From the Abstract

"We have applied a multiresolution cross-spectral analysis technique to resolve the temporal relationship between the NAO and ENSO. The study shows significant coherence between NAO and Nino3 SST in about 70% of the warm ENSO events from 1900 to 1995, of which 33% and 37% are associated with a 5-to 6-year period (El) and a 2-to 4-year period (E2) oscillation terms in the spectral decomposition, respectively. The dominant teleconnection pattern associated with changes in the mean atmospheric circulation during the initial winter of a typical E1 and E2 events is the positive phase of the Pacific/North American (PNA) pattern. Non-coherence Between the NAO and ENSO occurs during relatively weak Nino3 SST anomaly, with a teleconnection pattern which shows a strong negative phase of the NAO and a pattern which resembles a weak eastward shifted negative phase of the PNA pattern."

And from an article in the American Meteorological Society: 

The dynamical mechanism for the late-winter teleconnection between El Niño–Southern Oscillation (ENSO) and the North Atlantic Oscillation (NAO) is examined using the output from a 2000-yr integration of a coupled general circulation model (GCM). The coupled model captures many salient features of the observed behavior of both ENSO and NAO, as well as their impact on the surface climate in late winter. Both the observational and model data indicate more occurrences of negative phase of NAO in late winter during El Niño events, and positive NAO in La Niña episodes.

The potential role of high-frequency transient eddies in the above teleconnection is diagnosed. During El Niño winters, the intensified transient disturbances along the equatorward-shifted North Pacific storm track extend their influences farther downstream. The eddy-induced negative height tendencies are found to be more coherent and stronger over North Atlantic than that over North Pacific. These negative height tendencies over the North Atlantic are coincident with the southern lobe of NAO, and thus favor more occurrences of negative NAO events.

During those El Niño winters with relatively strong SST warming in eastern equatorial Pacific, the eastward extension of eddy activity is reinforced by the enhanced near-surface baroclinicity over the subtropical eastern Pacific. This flow environment supports a stronger linkage between the Pacific and Atlantic storm tracks, and is more conducive to a negative NAO phase.

These model results are supported by a parallel analysis of various observational datasets. It is further demonstrated that these transient eddy effects can be reproduced in atmospheric GCM integrations subjected to ENSO-related SST forcing in the tropical Pacific.

So it would appear that there may be some correlation of an El Nino and negative NAO with the direction of causality being from the ENSO process to the occurence of a negative NAO. It does seem that the development of the El Nino and its intensity and location will impact the intensity and duration of a possible negative NAO but lagged from the initiation of an El Nino or La Nina.That may be important for the U.S. East Coast and Northern Europe

C. Computer Models

I have been asked why computer models are not perfect. Here are two good reasons:

1. The model is incomplete. This can occur for two reasons:

1. the people designing the model are unaware of all the factors influencing the outcome so one or more factors are not included in the model.

2. the people designing the model are aware of other factors but either the data to properly include them is unavailable or the cost to collect the data is deemed to be larger than the benefit or the funding is simply unavailable. The Earth is a large Planet and the atmosphere and oceans have three dimensions. So depending on the desired degree of resolution, that is a lot of data. The following from the History of Numerical Weather Forecast is a pretty good introduction to weather forecasting models:

Until the end of the 19th century, weather prediction was entirely subjective and based on empirical rules, with only limited understanding of the physical mechanisms behind weather processes. In 1901 Cleveland Abbe, founder of the United States Weather Bureau, proposed that the atmosphere is governed by the same principles of thermodynamics and hydrodynamics that were studied in the previous century. In 1904, Vilhelm Bjerknes derived a two-step procedure for model-based weather forecasting. First, a diagnostic step is used to process data to generate initial conditions, which are then advanced in time by a prognostic step that solves the initial value problem. He also identified seven variables that defined the state of the atmosphere at a given point: pressure, temperature, density, humidity, and the three components of the flow velocity vector. Bjerknes pointed out that equations based on mass continuity, conservation of momentum, the first and second laws of thermodynamics, and the ideal gas law could be used to estimate the state of the atmosphere in the future through numerical methods. With the exception of the second law of thermodynamics, these equations form the basis of the primitive equations used in present-day weather models.

This article also might be of interest in terms of background on the history of the development of weather forecasting models.

Of course there has been much progress since then but you get the idea that one has to be able to describe the current conditions of the atmosphere (and land and ocean surfaces) to be able to make forecasts.

One approach to deal with lack of data is to extrapolate missing data using equations that estimate the missing data based on the actual observations. That provides a way to initialize a model. An improvement on that is to vary the estimates of the initial conditions (consistent with the distribution of variations in the real world and there is the rub) and run the models to see how sensitive the models are to small changes in the estimates of the initial conditions. This is one version of "ensembles" a fancy word for a collection of related things. You can have an ensemble of different initial conditional assumptions or an ensemble of slightly different models and the "spread" of the solutions is assumed to be a measure of the confidence one should have in the forecasts. I leave it to the reader to decide if that approach is consistent with how they arrive at confidence about the future. I am not saying that there is a better approach, but I am saying that if the model is wrong and "sensitivity" analysis indicates that small changes in assumptions or internal formulae for parametrization do not cast doubt on the forecast, that forecast will still be less than useful if the model itself is incomplete or just plain wrong.

Sufficient computational capacity is also important. Here is a blog post on that topic, which is a little out of date, but purports to describe the situation in the U.S. It is possible that some of the issues discussed in that blog post have been addressed since it was written, but my experience is that other nations lead the U.S. If you do not like the situation, talk to Congress. NOAA needs to be reorganized, streamlined, given new leadership and then provided adequate funding. Part of the situation is a result of the decision to provide opportunities for the private sector and not have the Federal solution so comprehensive that there was no room for the private sector to add value. It is a delicate balancing act.

2. The model is based on historical data but the situation has changed.

Thus the model continues to interpret the inputs as they should be interpreted consistent with the situation at the time the data was collected. Hence my conclusion that computer models essentially "predict the past not the future" which could be reworded to be less inflammatory as models "predict the future in the same way they would have predicted a similar situation in the past and thus their prediction of the future is essentially the same as predicting the past".

The reason for the above is that the way you build a model is by collecting information on the results that have occurred in the past when certain conditions applied. A real good example would be cyclones where we have theoretical and observed data on how tropical waves evolve into cyclones. But what if many of the factors in the lifecycle of cyclones change over time due to Climate Change. Will our models still be as effective as in the past?

From a mathematician's perspective, a relationship is valid only if the domain of samples used in the creation of the model includes the values used in the running of the model. From a practical perspective, you have no way to calibrate and test such a model if you have no data for situations faced that have never been faced before. We run into this problem with Climate Change models. The domain of data does not include carbon dioxide concentrations of 400 PPM or average temperatures greater than what we had for the last thirty years.

So the models are informed guesses as to how things might be in the future. The model designers attempt to calibrate their models against historical data but we do not have historical data where carbon dioxide levels are as high as today nor average temperatures including ocean temperatures are as high as today. Efforts to use paleo data are actually quite laughable since you normally do not for improve an accuracy problem by deciding to use a more approximate set of data.


For weather forecasting models, the incompleteness is the bigger problem. For Climate Change models, their use beyond the envelop of where the results are able to be calibrated is the larger problem.

D. Reserved for a Future Topic


Click here for a list of Sig Silber's Weather Posts

>>>>> Scroll down to view and make comments <<<<<<

Make a Comment

Econintersect wants your comments, data and opinion on the articles posted. You can also comment using Facebook directly using he comment block below.

Econintersect Weather

Print this page or create a PDF file of this page
Print Friendly and PDF

The growing use of ad blocking software is creating a shortfall in covering our fixed expenses. Please consider a donation to Econintersect to allow continuing output of quality and balanced financial and economic news and analysis.

Keep up with economic news using our dynamic economic newspapers with the largest international coverage on the internet
Asia / Pacific
Middle East / Africa
USA Government



Analysis Blog
News Blog
Investing Blog
Opinion Blog
Precious Metals Blog
Markets Blog
Video of the Day


Asia / Pacific
Middle East / Africa
USA Government

RSS Feeds / Social Media

Combined Econintersect Feed

Free Newsletter

Marketplace - Books & More

Economic Forecast

Content Contribution



  Top Economics Site Contributor TalkMarkets Contributor Finance Blogs Free PageRank Checker Active Search Results Google+

This Web Page by Steven Hansen ---- Copyright 2010 - 2017 Econintersect LLC - all rights reserved