As this current El Nino continues to evolve, it helps to visualize how these changes in the tropical Pacific ocean temperatures drive the changes in the hemispheric circulation which impact our winter weather across the US.
How do we define an El Nino? The Climate Prediction Center uses the ONI (Oceanic Nino Index) to define El Nino and La Nina events. Their definition for El Nino/La Nina using the ONI is as follows:
An El Nino event is defined when the three month running mean of (ERSST.v4) sea surface temperature anomalies in the Niño 3.4 region climb above the threshold of 0.5o for five consecutive overlapping three month periods.
In other words, we take the average ocean temperature anomaly in the 3.4 region over a 3 month period, say September through January. Take the next five month overlapping periods--October through December and November through January--(the first two for example) and find these averages. That is your five consecutive, 3 month running mean. If this period's average is at or above +0.5 degrees, we classify this as an El Nino. The ERSST.v4 dataset goes back to 1850. Using this methodology, the top 10 strongest E Ninos are easy to find. Hat tip to Eric Webb for compiling the ERSSTv4 data.
The top ten El Ninos are as follows: 1877-78, 1888-89, 1982-83, 1997-98, 1896-97, 1902-03, 1905-06, 1940-41, 1957-58, 1972-73 What do the sea surface temperatures for each of these El Ninos look like? I created an animation of each El Nino's six, three month overlapping period starting with the September-November period ending with the February-April period. Each El Nino evolves differently. The position of the equatorial warm water, position and strength of the warm/cool pools in the PDO region are different with some, similar to others. Here is each sea surface temperature anomaly animation from the list above.
1877-78 El Nino
1888-89 El Nino
1982-83
1997-98
1896-97
1902-03
1905-06
1940-41
1957-58
1972-73
How does the atmosphere respond in each of these cases at the 300mB level? More on this later.
What were the winter temperatures like in each of these strong (both eastern and central) El Ninos?
One glaring similarity is that each of these El Nino events (1888-89, 1940-41 and 1957-58) average temperatures across the eastern US stayed close or slightly above average in December and in January. Then in February, the bottom falls out and temperatures drop significantly below average across a large area of the east. If you look at the Pacific sea surface temperature configuration of these winters (eyeballing the ENSO region, Eastern Pacific/west coast/Gulf of Alaska regions and Pacific Meridional Mode area), they closely resemble this year!
You have probably heard about the BIG EL NINO or SUPER EL NINO or some phrase like that. If not then you will soon enough. Eventually EL NINO will trend on social media by the end of September. You heard it here first.
But hold on. Doesn't Big El Nino mean mild winter? Not necessarily. See the differences between the last BIG El Nino in July 1997 and July 2015. Also, notice the differences in the North Atlantic ocean temperatures? This is why the winter 2015-16 outlook is not an "El Nino shoo-in". Not all El Ninos are alike. (See my post on the different El Ninos and what they mean for our weather)
When will we know enough to make an educated assessment on the upcoming winter? Probably by the end of September. Stay tuned...
The persistent pattern featuring frequent rain across the Ohio Valley ramped up into high gear the second half of June. Most points from Columbia, Missouri east into the Mid Atlantic states received between 100 and 400% MORE rainfall than normal this last month.
Parts of northern Ohio had between 10 and 15 inches as a stalled front oscillated back and forth triggering clusters of rain and storms. Cleveland's monthly total was 3rd most in 145 years, most since 1972. Ft Wayne, Indiana set their all-time June rainfall record with 11.98". It broke the record for the wettest month set in July of 1986!
Last summer's wet region was centered in the heart of the corn belt.
This summer's wet areas have shifted east into the Ohio Valley including Pennsylvania, Maryland and portions of New Jersey.
Compare the past two summer to the very warm summer of 2012. Dry conditions prevailed with the development of a flash drought across parts of the US. I gave a talk at the Ohio State Weather Symposium on the causes and the conditions that feed that summer's dry pattern.
The average temperatures for July and August in 2012 were certainly influenced by the lack of rainfall. Notice the location of the temperature anomalies.
Temperature anomalies in 2014 were noticeably cooler over the wetter ground in the Corn Belt.
I emphasize that the rainfall or lack thereof was not the primary driver of the patterns in either 2012 or 2014. It was an enhancer. June temperatures were at or slightly below normal.
Record high temperatures across the Midwest were markedly lower in 2014 vs 2012.
Big question, how will the June rainfall in the Ohio Valley influence the temperatures in the upcoming weeks?
More than likely, we should see a dampening of long stretches of heat in the mid-west, corn belt and Ohio Valley. July could end up with temperatures at or below average from St. Louis to Cleveland.
The newest updated from the MEI (Multivariate Index) website run by Klaus Wolter shows a significant increase over the last month of +0.61 to +1.57 as of April/May which rates the El Nino in the strong category using a ranking method. If we look at the top April/May rankings according to Mr. Wolter then then take away the El Nino events that were weakening as of April/May, we are left with the 5 years below. ALL OF THESE YEARS STAYED AT EL NINO LEVELS THROUGH THE END OF THE YEAR.
One important note: Only 1998, 1993, 1992, 1987 and 1983 had HIGHER MEI levels in April/May. 1987 was a only building El Nino. The rest were fading or close to it.
The El Nino is very apparent in the current sea surface temperature anomaly map in the last week of May.
The 2014 El Nino (although not official using the ONI) was considered weak by many meteorologists including myself. If you assume a weak El Nino for a time last year, this year's much stronger event would be a 2nd year El Nino. Expanding the possible El Nino matches using the MEI TIME SERIES for "double El Nino events" we found two such events: Late 1991 through 1993 and 1986/87. A 2nd year El Nino has the advantage of an ocean already primed from the previous one.
It's not an exact match but comparing these two years to 2014-15 gives us an idea of a possible El Nino outcome. Remember that this year's El Nino is developing much earlier. Most El Nino's develop in late summer and fall.
By examining the MEI, assuming a building El Nino, a double El Nino event (last year was weak) and a very high April/May MEI, a good match so far is the 1986-87 event.
What were the bi-monthly overlapping sea surface temperature anomalies starting in April through the end of 1987? Notice the cooler water in the western Pacific near Australia which is necessary for the development of a strong El Nino according to the Australian Bureau of Meteorology.
Watch how the lower temperature areas warm by the end of the year (lower two maps below) and the warmest regions start to cool a bit as the El Nino started to fade by year's end.
How about this year's event? The water off the coast of Australia (circled in red) is starting to cool comparing April 1st to may 27th but not at the level of the 1986-87 event...yet.
This El Nino continues to strengthen much earlier than other El Ninos. We continue to watch the water temperatures in the western Pacific near Australia as they cool. How cool will they get? Will the El Nino peak in late Fall then taper off?
If we took all of the rain this past weekend across northern Ohio and calculated the volume in gallons, how much would we have?
First, I eyeballed the area across northern Ohio that received roughly 3/4" of rain by late Sunday morning. The area is 16,000 square kilometers (6200 square miles)...
The area that received roughly 1.5" of rain was around 2000 square miles...
The smaller area that received at least 2.5" wasa little under 400 square miles.
Convert the "square miles" to "square feet" for each area:
6300 square miles = 173 BILLION square feet (3/4" rain)
2000 square miles = 56 BILLION square feet (1.5" rain)
400 square miles = 11 BILLION square feet (2.5" rain)
First, I calculated the volume of water for the area that received 3/4" of rain:
In order to not double count volumes, I subtracted the starting 3/4" from the second 1.5" to get the HEIGHT for the volume calculation of the second area
VOLUME 2 = 56 BILLION square feet x (1.5" - 3/4") rainfall or 0.0625 ft =
roughly 3.5 BILLION cubic feet
I subtracted second 1.5" from the third 2.5" to get the HEIGHT from the volume of the third area.
VOLUME 3 = 11 BILLION square feet x (2.5" - 1.5") rainfall or 0.083 ft =
roughly 886 MILLION cubic feet
Add the volumes together, we get: ~15.2 BILLION cubic feet
Convert CUBIC FEET to GALLONS gives us: ~114 BILLION GALLONS
Its Memorial Day here at the station. I remember my dad, grandfather, uncle, brother-in-law who served in our Armed Forces over the generations.
I've been watching a few showers popping up on radar. Nothing heavy, nothing widespread. One helpful tool we use in determining where and when precipitation develops is the newly operational HRRR model (short for High Resolution Rapid Refresh). In essence, this model assimilates 3 km radar data into it calculation each hour. So rather than wait a full 6 hours until the next set of model outputs, the HRRR runs a NEW output every 60 minutes taking into account the ew radar information.
This morning, I ran a little experiment to see how accurate the HRRR was in determining the light showers it was showing for 9AM. The NWS Cleveland radar was down for repairs so this was a perfect opportunity to see how this would affect the output.
Here is the HRRR Output each hour starting with the 5z (1am) run and concluding with the 11z (7am) run. So the 5z run is 8 hours from 9am. The 11z run is 2 hours from 9am. So each successive model run brings us closer to the 9am target time.
A light shower west of Cleveland...
By the 8z run, the HRRR started to shift the light shower further west into the main cluster of rain along the Ohio/Indiana border. By 11z, it had merged with the western rain area.
Here is what the actual radar looked like at 9AM using the TDWR radar in Grafton since the NWS radar was still down as of this writing. Hardly any precipitation!
Thinking outloud... * The HRRR overestimated the development * Unknown whether Cleveland radar downtime had an affect * Unknown whether or not TDWR radar data from the FAA is incorporated into HRRR model.
Last year at this time (Spring 2014) the talk of a "Super El Nino" was at the forefront of weather circles. The resultant El Nino later in the year surely didn't fit the hype for a variety of reasons. This year's close El Nino cousin is a little different. I believe the "hype" is more warranted this go-around.
Here is a great El Nino animation from NASA showing its evolution.
Using the ONI (Oceanic Nino Index), I logged all El Ninos since the early 1900s. The ONI is a 3 month overlapping sea surface temperature anomaly index of the Nino 3.4 region, the area used in defining the ENS state.
In determining how different this evolving El Nino is compared to past events, I used the actual NON-OVERLAPPING monthly Nino 3.4 sea surface temperatures anomalies for each early stage El Nino event starting in March and continuing through December. My goal was to see any slight variation in ocean temperature that was smoothed by the 3-month overlapping ONI. Note: Many of these El Nino events continued into the following year not shown here.
The first chart below shows El Nino events from 1982 to early 2015.
Some key points: This year's event has started MUCH FASTER and earlier than the historic El Nino events in 1997 and 1982. The 1986-87 event peaked in summer then slowly dropped by fall after a similar start.
The current El Nino is well ahead of all of the events from the early 1950s through the late 1970s. Some of these El Ninos were weak. 1965 and 1972 was strong.
The El Ninos of the late 1920s, 30s and 40s were late bloomers showing little sign of El Nino until mid fall.
The early 20th century El Ninos had more erratic evolutions. Even the strong El Nino of 1918 struggled until fall/early winter.
Of all 23 El Ninos I charted, only the El Nino of 1905 had WARMER ocean temperatures in May compared to this year's event. That year peaked at 1.5 in September.
So in summary...
* El Nino headlines need historical perspective. Don't be quick to compare this event to the 1997 event per the model projections.
* The 2015 El Nino is building faster than any El Nino since 1905
* The rapid rise in ENSO 3.4 ocean temperatures this early in spring historically usually means the El Nino will sustain itself through the summer. The major El Ninos of 1997, 82 and 72 started off slower. A big start doesn't necessarily mean 2015 will be another 1997 or 1982. In fact, the warmth is more centralized. The warmth in '82 and '97 sloshed eastward by summer.
Warmth so far this year has two lobes, one central near dateline and another eastern closer to South America.
* The moderate El Ninos (SST 3/4 between 1.0 and 1.5) in 1987, 72, 65, 57, 30, 25, 23, 18, 05 and 1902 all sustained El Nino status through December per the ENSO 3/4 ocean temperatures.
What does all of this mean for the summer, fall and upcoming winter?
* The next 2-3 months will be critical in determining how this El Nino will impact our (US) weather this summer and especially fall.early winter. The position of the warmth will be a big factor.
* How much cool water near Australia relative to the ENSO 4 and 3/4 region temperatures will be HUGE in driving the westerly wind bursts necessary in sloshing more warmth to the east keeping the El Nino machine going.
Next week, we'll dive more into why the COOL WATER NEAR AUSTRALIA is so critical in the development of El Nino
