El Nino Rains: Comparing 2015 so far to other beasts of the past

The fever pitch of El Nino headlines and articles isn’t slowing down. In fact, it’s only growing:

El Nino Fever Pitch

It comes mostly with speculation, preparation, and wonder. Some of the headlines are ridiculous (just go back through photos on my Twitter timeline). Most are reasonable. But it’s news. And it’s clearly a high impact weather phenomenon.

So it begs the question: How’s it doing? Is El Nino behaving like it’s supposed to? Specifically with regard to rainfall. Remember, this El Nino event is a strong one. In fact, it’s the strongest since 1997, and in some respects, it may be the strongest on record. So we have a pretty clear set of analogs to look at and see if this year fits the mold.

Judging by the temperature forecast for the next 10-15 days, the pattern seems to be mostly behaving like a strong El Nino should in December. How about rain? Let’s look back at other actual strong El Nino events and see where 2015-16 is falling relative to those events, specifically in terms of rainfall.

A caveat here: Remember, it’s early. Winter is a marathon, not a sprint. Because something looks one way on December 8th, doesn’t mean one month from now we’re going to be in the same boat. But there are a few interesting nuggets to pull out of this, and I think it’s mostly a good idea to give people some perspective on certain story lines.

Just as a note, my definition for “strong” El Nino events was using a blend from Jan Null’s list post-1950 and Klaus Wolter’s list pre-1950, as well as some “artistic liberty.” Not everyone agrees on the perfect definition of what a strong El Nino is, but hopefully I captured the majority. I’m tracking rain from July 1-June 30.

Southern California

When people think of El Nino, many think of SoCal. You know, mudslides in Malibu, etc. I believe the mayhem of 1997-98 and subsequent personification by Chris Farley has set a level of expectations.

Thus far, 2015 is behaving pretty much about where other strong El Nino events have done in Los Angeles:

Los Angeles Rainfall

Every strong El Nino event back to the 1800s has had normal or above normal rainfall in Los Angeles. Thus far, thanks mostly to a freak wet event in early autumn, Los Angeles is indeed above normal. It is worth noting, that 2015-16 wouldn’t be the only strong El Nino year to see a “freak” early autumn rainfall event followed by a prolonged period of dryness into December. So if people are wondering where the rain is, wait a few more weeks. We are entering the ramp up period. If things don’t start picking up by late month, then we can start to worry.

Northern California

Northern California is a conundrum during strong El Nino events. Historically, San Francisco is split between above and below normal rainfall, so it’s tough to say too much at this early stage. That said, it is worth nothing, that as of right now, this El Nino is on the lower end of the envelope in terms of rainfall in strong El Nino events in the Bay Area:

San Francisco Rainfall

Like Los Angeles, it’s still early. The ramp should start soon if we’re going to go above normal, so again, if things don’t perk up by late December, it might be time to plan on a normal or drier than normal winter at best.


This is where it gets interesting. For the most part, the El Nino signal in the Pacific Northwest is somewhat mixed. You get dry years and you get wet years. Usually, however, you don’t have a super wet year when there’s a strong El Nino.

In Portland (where the data is only available back to the 1957-58 El Nino) no strong El Nino has been wetter than normal. Seattle is mixed from slightly above normal to way below normal. So far in Seattle? It’s the wettest strong El Nino to this point on record, and it’s not even close. This autumn has been incredibly wet so far.

Seattle Rainfall

If we keep up at this rate (and the current forecast implies this), this will quickly become one the wettest, if not the wettest strong El Nino event on record in Seattle.

Great Basin/Rockies

In Denver, the bend is toward a wetter than normal outcome in strong El Nino years, but it is mixed. So far, Denver has been mid-pack for strong El Nino events.

Denver Rainfall.png

Salt Lake City is a bit more mixed, split almost evenly between wetter than normal and drier than normal (82-83 was a beast there). So far, we’re doing middle of the pack there also.

SLC Rainfall

It will be interesting to follow this as we go through the cold season. I’ll likely be tracking this, as it’s good to establish where this El Nino event falls. For everyone, it will probably be different. But thankfully we have a useful sample in a number of places to utilize.


Can the Packers-Niners be the Coldest Game Ever?

Sensationalism reigns supreme in today’s 24/7 news cycle world. If you aren’t the loudest or boldest, why even bother, right?

Today’s Drudge Report has a headline reading:


Bleacher Report agrees: http://bleacherreport.com/articles/1909338-packers-49ers-playoff-game-might-be-coldest-nfl-game-ever

The San Francisco Chronicle concurs also: http://blog.sfgate.com/stew/2014/01/02/49ers-playoff-game-in-green-bay-could-be-coldest-ever/

There’s no question that Sunday’s Packers/Niners game is going to be painfully, brutally cold, and it will most likely crack the list of 10 coldest games in NFL history.

According to NFL.com, the coldest game in history was the Ice Bowl on December 31, 1967, also at Lambeau of course, with a temperature of -13°. The second coldest was 1/10/82 at Cincinnati between the Bengals and the Chargers, with temperature of -9°. And #3 all-time was Chiefs/Colts in Kansas City on 1/7/96, with a temperature of -6°.

So for it to be the coldest ever, it would need to be -13° during the game. For it to crack the top three, we’d need to have a temp of -6° during the game. Can we do it? Let’s examine the evidence.

The game is scheduled to kick off at 3:40 PM local time. Weather models and data are frequently in Greenwich Mean Time or Zulu (Z) time. Accordingly, this game is slated to kick off at 21:40Z. Assuming an average NFL games lasts roughly 3 hours, we’ll examine temperatures in the 21Z to 00Z timeframe.

First off, the bane of most weather forecasters is MOS (model output statistics). It takes weather models, applies statistical algorithms, and produces a temperature, precip, cloud cover, etc. forecast based on the model and history. Sometimes it works nicely, but a lot of times it can be less than stellar. Anyway, looking at the two primary models’ MOS forecasts from this morning:


I have highlighted the key times. The GFS model (top) is showing temperatures falling from -6° to -10° during the game. The NAM model however disagrees substantially and is quite a bit warmer, with temperatures dropping from +6° to +2° during the game. Worth noting: If you look at the far right numbers (-25° top and -21° bottom), that indicates the the models are only four degrees apart for the Monday morning low temperature in Green Bay. So the issue is timing. The GFS is faster bringing in the Polar cold to Green Bay than the NAM model, which is a few hours slower.

A higher-resolution version of the NAM model is run also. It goes out 60 hours (vs. 84 hours like the normal resolution version the MOS output is derived from). Looking at that from the PSU E-Wall:


The map above is for 00z (or right around the end of the game). It shows temperatures just approaching 0° west of Green Bay (I circled the Green Bay area). So add another to the slower/”milder” camp. Incidentally, the European model (considered the gold standard of weather models) is somewhat between these two, with temps dropping to somewhere in the -2° to -5° range during the game.

The National Weather Service forecast office in Green Bay is on the colder side of things. Their forecast is below. They have temps falling to -6° to -8° or so, with wind chills of -25°.


And the last thing I will show you is what the SREF, or short-range ensemble forecast spread in temperatures is for Sunday afternoon. This is a suite of a bunch of short-range models, all with different parameters to create a cluster or consensus forecast.


The black line in the “middle” represents the mean of the various models averaged together. That mean forecast is for temps to drop from roughly -6° to -9° during the game. Note that leading into the gametime period, it does appear the model consensus splits into two camps…a milder one and a cold straight through one. The bigger cluster heading into the game window appears to be near the mean or just milder. So the mean is probably a fair representation of this model.

So, with all this being said, will Sunday’s Packers/Niners Wild Card matchup be the coldest NFL game in history? Most likely not. However, there is at least a reasonably small chance it could be if things come together perfectly. The GFS model does tend to rush certain air masses and weather systems, so I would be apt to lean warmer than it, but colder than the NAM model. This would put me at something like -3° to -1° at kickoff and maybe -7° to -5° at the end of regulation.

Either way, it should be a memorably cold game that’s something of a throwback or relic to old school football.