NOAA: El Niño to Help Steer U.S. Winter Weather

October 15, 2009

Information concerning how our local winter around Northern Indiana, Southwest Lower Michigan and Northwest Ohio have been impacted in the past during El Nino winters can be found further down on this page.  

El Niño in the central and eastern equatorial Pacific Ocean is expected to be a dominant climate factor that will influence the December through February winter weather in the United States, according to the 2009 Winter Outlook released on October 15th, 2009 by NOAA’s Climate Prediction Center (CPC). Such seasonal outlooks are part of NOAA’s suite of climate services and are relayed through local National Weather Service offices.

El Nino is defined by warmer than normal sea surface temperatures (SSTs) in the Pacific Ocean near the equator. These above normal SST's must exist for a period of several months and be at least 0.5C above normal. When these SST's (officially called the Oceanic Nino Index, or ONI) are generally above 1.5, the El Nino is considered to be a strong event.

“We expect El Niño to strengthen and persist through the winter months, providing clues as to what the weather will be like during the period,” says Mike Halpert, deputy director of the Climate Prediction Center – a division of the National Weather Service. “Warmer ocean water in the equatorial Pacific shifts the patterns of tropical rainfall that in turn change the strength and position of the jetstream and storms over the Pacific Ocean and the U.S.”


       Winter Outlook - Temperature           


              Winter Outlook - Precipitation

 Figure 1. CPC Winter Outlook (Dec-Feb). For more information on how to interpret and understand the probabilities on these maps, click here.

“Other climate factors are also likely to play a role in the winter weather at times across the country,” added Halpert. “Some of these factors, such as the North Atlantic Oscillation are difficult to predict more than one to two weeks in advance. The NAO adds uncertainty to the forecast in the Northeast and Mid-Atlantic portions of the country.”

Highlights of the U.S. Winter Outlook (December through February) include:

  • Warmer-than-average temperatures are favored across much of the western and central U.S., especially in the north-central states from Montana to Wisconsin. Though temperatures may average warmer than usual, periodic outbreaks of cold air are still possible.
  • Drier-than-average conditions are expected in the Pacific Northwest and the Ohio and Tennessee River Valleys.
  • Northeast: Equal chances for above-, near-, or below-normal temperatures and precipitation. Winter weather in this region is often driven not by El Niño but by weather patterns over the northern Atlantic Ocean and Arctic, such as the North Atlantic Oscillation. These patterns are often more short-term, and are generally predictable only a week or so in advance.

This seasonal outlook does not predict where and when snowstorms may hit or total seasonal snowfall accumulations. Snow forecasts are dependent upon winter storms, which are generally not predictable more than several days in advance. Snowfall accumulations can also be highly variable and localized.

NOAA understands and predicts changes in the Earth's environment, from the depths of the ocean to the surface of the sun, and conserves and manages our coastal and marine resources. Visit   

  • To see what the probabilities for above normal, near normal and below normal temperatures are for 10 select locations across Northern Indiana, Northwest Ohio, and Southwest Lower Michigan, visit our local Three-Month Temperature Outlook maps on our climate page. These maps give downscaled probabilities from CPC which show the chances of exceeding the average temperature during 3-month periods. These maps are updated about once per month when new outlooks are issued by CPC.

What Have Past El Nino Winters Brought to Our Local Area?

Local Climatological Data and Trends at South Bend and Fort Wayne Indiana, representing most of our area

It is always interesting to look at temperature, precipitation and snowfall data at individual local sites to see if we can spot any unusual data or trends that might have different impacts than those indicated in the large scale outlooks. South Bend (SBN) and Fort Wayne (FWA) Indiana are considered Local Climatological Data (LCD) sites and have a long history of weather records as well as complete sets of normal records. "Normals" are 30 year averages that indicate what temperatures, precipitation and snowfall should typically be on any given day or month of the year based on what has previously happened. We can compare different years to the normals to see how those years varied from what would typically be expected.

We begin by looking at the large scale pattern we are currently in. The 2008-2009 winter season was characterized by a "near La Nina" pattern as the criteria for an actual La Nina was just not quite reached. However, sea surface water temperatures in the Pacific were below normal and showed La Nina qualities, but just fell short of meeting the length of time needed to be classified officially as a La Nina (For information on how a La Nina or El Nino are classified, click here). The summer months of 2009 were also quite cool, with the average temperature for this past summer season below 70F across much of our area. Now we are quite confident that an El Nino will be ongoing during the 2009-2010 winter season (conditions in the Pacific are currently meeting El Nino requirements, but this must persist for several months before officially classified as an El Nino). 

Based on this current pattern of going from a La Nina type pattern to an El Nino pattern, we looked for past years since 1950 when such a pattern existed. We call these "Analog" years, because they represent a pattern similar to what we are currently experiencing. Scientists use analog years to see if there are any similarities in the weather, and use that information to help shape future forecasts. A large number of analog years can give better clues to patterns compared to just a few analog years.

Since 1950, there have been 18 El Nino episodes (1951-52, 1957-58, 1963-64, 1965-66, 1968-69, 1969-70, 1972-73, 1976-77, 1977-78, 1982-83, 1986-87, 1987-88, 1991-92, 1994-95, 1997-98, 2002-03, 2004-05, 2006-07).  Eleven of these El Nino episodes were preceded by winters with La Nina or near La Nina conditions (1951-52, 1957-58, 1963-64, 1965-66, 1968-69, 1972-73, 1976-77, 1986-87, 1997-98, 2002-03, 2006-07).  

These 11 years will serve as our "analog" years since they are similar to the pattern we have been in and the pattern we are expecting this winter. In addition to this data, we also looked at years within our analogs when summers between the La Nina and El Nino episodes were very cool with an average summer temperature of less than 70F. This included 5 of the 11 analog years at South Bend (1961-62, 1963-64, 1965-66, 1972-73, 1997-98), but only 3 of the 11 analog years at Fort Wayne (1965-66, 1972-73, 1997-98).  

These analog years represent a very small data sample and therefore the results are by no means conclusive or indicative of any trends with scientific validity. However, it is interesting to look at some of the trends and see what the averages indicate.  

Monthly Averages During All El Nino Episodes

Figure 2 (below) shows temperature, precipitation and snowfall data from South Bend and Fort Wayne during all El Nino winters since 1950. This data shows trends of having fall temperatures being near or below normal followed by a rather warm signal in December. January and February show split temperature signals between South bend and Fort Wayne with South Bend trending slightly warmer than normal and Fort Wayne slightly below normal. 

As for precipitation and snowfall, the late fall months tend to have above normal precipitation and below normal snowfall at Fort Wayne while South Bend averages above normal precipitation and above normal snowfall. The winter begins with similar precipitation trends in December, but snowfall shows a definite trend toward below normal values. We begin to see a shift in precipitation patterns in January and February with below normal averages, especially in February where average departures are more than one half inch below normal. Snowfall trends reflect a similar pattern, though there is a slight nudge above normal at South Bend in January. Looking into the early spring month of March, we tend to see above normal temperatures with near normal precipitation, but snowfall averages above normal.

We must remember that these are averages of 18 different El Nino Episodes ranging from weak to strong. Some weak to moderate El Nino winters can be cold with higher than normal precipitation and snowfall while strong El Nino winters are usually warmer with less precipitation and snowfall. Certain winters with much above normal or much below normal conditions can greatly affect the averaging, thus some inconsistencies can show up. The averages below generally represent conditions similar to the winter outlook issued by the CPC for our area. This indicates higher probabilities for below normal precipitation, but equal chances that temperatures could average above, below or near normal.  

 Average Monthly Temperatures during El Nino Episodes


Figure 2. Temperature, Precipitation and Snowfall departures from normal during all El Nino Episodes since 1950.

 Monthly Averages During Analog Years 

Since we are exiting a near La Nina winter and about to enter an El Nino winter, we wanted to see if there were any significant differences in the El Nino winters in this type of pattern. Figure 3 (below) shows this data for our 11 analog years, as well as analog years when the El Nino was weak or moderate in intensity (ONI <=1.5). Departures for all El Nino's are also shown for comparison. Once again we must caution that the sample size of these analog years is relatively small and therefore not as scientifically sound as a larger data set. This is especially true with respect to the analogs with ONI <=1.5 with only 6 events to look at. 

Some interesting trends can be seen in this data. Beginning with temperatures, we see a very cool late fall pattern at both South Bend and Fort Wayne with the analog years supporting or even enhancing the signal. The first official month of winter, December, shows the 11 analog years being above normal in temperature but not to the degree that we see when averaging all 18 El Nino events. The trend is reversed during our 6 analog events with a distinct cold average at both sites. Interestingly, this colder than normal trend continues in January and February, with the strongest negative departure occurring in our 6 analog years with weak to moderate El Nino's. One explanation of this significant change is likely due to the fact that strong El Nino's make our winter months so much warmer than normal that they significantly weight the average because they are so warm. More of these weak to moderate events are needed to increase the sample size and add validity to this data. Finally, the beginning of spring sees all data sets showing above normal temperatures during all episodes.

Precipitation trends in our analog sample years also show differing trends compared to all El Nino years. Significant departures from above normal in all El Nino events to below normal in our analog data can be seen in the late fall months. The exception is at South Bend in November when the trend in all episodes is for above normal precipitation. The largest negative departure is seen at Fort Wayne during our 6 weak to moderate El Nino events. As we get into the winter months, precipitation remains above normal for December for all episodes at both sites. January data is generally clustered around normal with only Fort Wayne seeing a jump to above normal in our weak to moderate El Nino analog years. The end of winter signals a definite dry bias with significantly below normal precipitation in the month of February. These departures are so strong that they can offset the above normal departures in the early winter and help make the entire winter average below normal, which agrees with the CPC outlook. The beginning of spring sees a reversal from slightly below normal precipitation in all El Nino's to significantly above normal in our weak to moderate events.

The last graph in figure 3 shows snowfall departures. The late fall is characterized by significantly above normal snowfall at South Bend. November departures are above normal during all El Nino events, but climb to over 1.5 inches in our 11 analog years, and rise to over 3.0 inches above normal in our weak to moderate El Nino's that follow La Nina winters. This same above normal feature can be seen in December at South Bend as well, though snowfall departures are below normal in our other two samples. These trends are reversed in January and especially February as the departures become more negative as the winter progresses. Fort Wayne shows a distinct below normal trend in all winter months, with a curious exception in February during our 6 weak to moderate El Nino events. A closer look at the data reveals that 5 of the 6 February's in our analog data set had above normal snowfall. The greatest being 6.7 inches above normal in 2003. The lone below normal month was February 1969 with a departure of 4.8 inches below normal. The beginning of spring shows a definite shift toward above normal snowfall in all data sets at both sites.


Temperature Departures During El Nino and Analog Years

Precipitation Departures During El Nino and Analog Years

Snowfall Departures During El Nino and Analog Years

Figure 3. Temperature, Precipitation and snowfall departures from normal during El Nino years, analog years, and analog years when weak or moderate El Nino occurred.

Years with a Cool Summer Between a La Nina and El Nino

The last set of data we will examine is our smallest data set yet. We looked at our 11 analog years of going from a La Nina pattern to an El Nino and looked for years when the summer was also cool, with an average temperature of less than 70F. This pattern would nearly match our current year as the average summer temperature at both South Bend and Fort Wayne was below 70F.

Figure 4 (below) shows the departures during our 11 analog years along with departures when we had a cool summer within them. Remember, there are only 5 years at SBN and 3 at FWA since 1950 that meet this criteria, so no scientific conclusions can be reached. However, some interesting shifts in the data have occurred. Temperatures are generally below normal in the late fall, though South Bend shows an interesting spike in October of above normal. The winter months are characterized by a significant jump to above normal temperatures, which is also carried into the early spring.

Precipitation trends toward below normal values in the late fall. Again, a curious positive anomaly can be seen at Fort Wayne in October. With only 3 years of data, this certainly is nothing significant. December and January trends are varied but again we see the dominate signal in February of below normal conditions, equaling or exceeding the December and January combined departures. The above normal precipitation trend is more evident in March during these limited events. Snowfall data indicates similar patterns to our 11 analog years with a trend toward even bigger anomalies. 


Figure 4. Temperature, precipitation and snowfall departures during cool summers between La Nina and El Nino winters.


For the most part, the analog data sets tend to support the overall forecast from CPC for our area, with a few interesting exceptions. South Bend and Fort Wayne temperatures during the winter months (Dec-Feb) tend to average above normal in all 18 El Nino and 11 analog episodes. The exception in our data was during our 6 analog years with a weak to moderate El Nino. The trend in these years, which is similar to our current year, showed significantly colder than normal winters as all three months in these years averaged below normal temperatures. The CPC outlook for our area this winter calls for equal chances of above normal, below normal or near normal temperatures. Our data generally supports this, but one could argue that we may see below normal temperatures based on our 6 analog years. However, as we have been mentioning throughout this article, this sample size is quite small and scientifically does not represent a large enough data set to be statistically valid. Our even smaller data set for cool summers within our analog years reverts back to above normal temperatures through the winter. This should show the variability and reinforce the idea that many other variables help control our weather patterns in the winter.

In all of our data sets, precipitation for the winter months averaged below normal. However, this was mainly because the average for February was so large that it offset the trends of December, which was above normal for all data sets, and January, which was right around normal or above normal. So while the CPC outlook calls for the winter as a whole to be below normal, our data shows we could see periods of above normal precipitation, especially through the first two thirds of winter.

Past research here at NWS Northern Indiana in conjunction with Purdue University examined an El Nino winter where precipitation was nearly 200 percent above normal, far opposite of the expected below normal outlook. The conclusions found that a single winter storm system in January produced enough precipitation to offset the remainder of the winter, which was below normal. This study also showed that our region is very close to the polar jet stream which tends to stay just to our north in El Nino winters. However, if this jet buckles southward in conjunction with a strong and moisture laden low pressure system, we could see abundant and above normal precipitation, and possibly snowfall if temperatures are cold enough. This is why it can be very difficult to forecast a winter weather outlook for our part of the country in El Nino winters.

Finally, snowfall departures varied early in the season between South Bend and Fort Wayne, but trended toward below normal by the middle and end of the winter months. South Bend showed an increase in snowfall in November and December while Fort Wayne remained below normal in these months. Once again, a strong signal in February tended to bring down the averages and offset any positive anomalies from the first part of the winter.

So what does all of this data and information mean? Basically it supports the CPC winter outlook for our area with trends indicating that temperatures could be above or below normal while precipitation certainly trends toward below normal, thanks mainly to a very dry February in many years. This data should be viewed as simply an interesting look at past years and trends, with no strong statistical significance at this time. Analog years simply use past data to give one idea of what future weather may hold in store. There are many variables though that can change with the weather patterns so no two years will ever be exactly alike. We will have to wait and see what the 2009-2010 winter really holds in store for our area.

We hope you have enjoyed this in depth look at some interesting weather data and trends during El Nino years. Feel free to send your comments or questions to Sam Lashley, NWS Northern Indiana Climate Focal Point, care of

Sam Lashley 
NWS Northern Indiana Office is the U.S. government's official web portal to all federal, state and local government web resources and services.