Saturday, November 24, 2018

Q&A: Sunday Storm

So, anyone have any suggestions for a new hobby?  This one is killing me.

I'm going to try to anticipate and answer your questions about the Sunday/Sunday night storm.  Here we go...

Q: How much snow we gonna get?

A: Wow, you skipped right past the foreplay and went right in...

Q: Blah blah blah. What do the models say?

A: Here's the graphic:


Q: Why are they all different?

A: If they were all the same, we'd only have one model and a bunch of models would be out of a job. Like Melania.

The models are starting to agree on a couple of things:
  • A later-arriving, slower-moving storm.  Slower pace means snow could pile up where it does fall, and in those spots, have a greater impact on Monday morning commute.
  • A more compact precipitation field.  Oddly, the storm tracks on most of these models are very favorable for heavier snows in southern Wisconsin, but the snow gradients are really tight.
Q: Wait, your prediction isn't on there. Chicken?

A:  It's at the bottom of the post. 

Q:  Okay, which model should we trust?

A:  The correct one.  Alternatively, NAM.

Q:  What could still change?

A:  High pressure to the southeast could kick our storm farther northwest, or the track could flatten out and go completely zonal, or equatorial sea surface temperatures could suddenly cool, or a negative North Atlantic Oscillation could...

Q:  You lost me.

A:  No worries, I don't even understand negative NAO.

But seriously, the slightest movement to the north in the storm track, or a wider precipitation field, will make a big difference.

Q:  Exactly when will snow start and end in my backyard?

A:  I have no idea where your backyard is.  Ask me on Facebook.

Q:  Will schools be closed Monday?

A:  Probably not in Dane County.  Possibly southern Green.  More likely to the east and southeast. 

Also, wasn't that Melania joke funny?

Q:  Seriously, how much snow?

A:  Okay, here's the completed chart.

 
Q:  Shit!

A:  Disclaimer: still subject to change, snow lovers.  30 miles could turn this forecast upside down.

Q:  So, where should I go if I really want to see a shit-ton of snow?

A:  Elgin, Illinois would be a good bet.  No I-Pass?  Try Lake Geneva.
 
Q:  Will it be windy and cold too?

A:  First windy, then cold. 

Q:  Cool. When's the next storm after that?

A:  In December.

I'll be the first to let you know if anything changes!!

Thursday, November 22, 2018

Sunday Snow Update: Models Align and Go Big

Either the weather models are really onto something, or they all went out and got shitfaced together last night.

Regardless, now would be a really good time to assess your Sunday travel plans.  More on that in a bit.

We're 3 days out from what will likely be our first significant snowstorm of the season.  This is when we expect to start seeing some model consistency, and the first runs of the day are giving us exactly that.  Track and intensity will still come into sharper focus the closer we get, but models now are at least similar in their interpretations of this event.

Here's how NWS is currently assessing the path of the low.  I mentioned in my post yesterday that the "ideal" track for heavy snow in southern Wisconsin is through central Illinois and into lower Michigan.  The NWS path cone gives us that.

(Click any image to embiggen it)


Here's an animation of the Euro's track:


And Euro snow totals:


GFS totals:


GFS-FV3 has consistently been really big with totals:


And the GDPS (Canadian):


Meanwhile, Accuweather is at 5.1" and the Weather Channel ranging 6-10".  Accuweather, for reasons even Accuweather can't explain, is frequently 2-4 degrees warmer than every other forecast on the planet.  And 2-4 degrees would make a huge difference with a storm like this.

All of these totals will change as track and intensity becomes clearer.  But this level of model agreement is enlightening 72 hours out.

My expectation is that by Noon Friday, NWS out of Milwaukee will have issued a Winter Storm Watch for our area.  That watch could come even earlier because of holiday weekend travel impacts.  Also tomorrow, we'll start to see the short-range high resolution models, which will be helpful in assessing storm track.

If you are traveling home on Sunday, and home is in the southern half of Wisconsin, I'd suggest getting home by 2pm or so.  In the meantime, fill up your gas can and test your snowblower.  And be sure to tease your kids with the possibility of school closings Monday morning.

One more note.  We haven't discussed wind yet.  I would expect a fair amount of wind out of this storm (from the northeast), making travel even more treacherous.  White-out conditions aren't out of the question.

Feel free to share if you have friends or family traveling Sunday.  Have a great Thanksgiving.  I'm around to answer any questions.

~Scott


Wednesday, November 21, 2018

Significant Snow Possible Sunday/Monday

Normally, I'd wait another 24-48 hours before posting about a storm that's a good 4 days away, but (1) I know a lot of folks will be traveling Sunday as they wrap up Thanksgiving weekend, and (2) there has been some model consistency on this storm for the last couple of days, which increases confidence.

Here's the quick take, and then we'll get into some detail:
  • There will be a storm in our part of the country late this weekend;
  • Current thinking on timing is late Sunday/Monday;
  • It will bring rain to some, snow to others, and a rain-to-snow changeover for others;
  • Because of the path of the storm, there should be abundant moisture, which could result in heavy (6"-plus) snowfall for some;
  • Timing, storm track and accumulations will come into better focus sometime on Friday.
The two main mid-range weather models, GFS and Euro, have been literally alternating their interpretations of this storm.  The GFS model runs every 6 hours; the Euro every 12 hours.  During one run, GFS will show heavy snow in Wisconsin and Euro will show rain...then in the next run, GFS goes light snow and Euro goes heavy.  Since Monday, they've vacillated between 0" and 20" (GFS FV3) of snow in our general area.  This is pretty normal at this range.

Here's the set-up:

Low pressure comes on shore in the northwest early Friday.  You see it on the surface map west of Oregon:


In the middle of the country, a separate system will bring us rain Friday night and Saturday morning.

The second system moves southeast toward Utah and Colorado 24 hours later:


After dipping into the southern plains (and grabbing significant moisture from the Gulf), the low starts to head northeast, and is centered in Missouri Sunday morning:


 These tracks will wobble and change as we get closer to the actual event.  There are three key variables I'm watching that will greatly impact the type of precipitation we get, and how heavy it is:
  1. How far south will the low get before it tracks northeast?  If it gets into northern Texas, it can pick up significant Gulf moisture. 
  2. How much cold gets wrapped to the southeast with the passage of the first low on Saturday? This will help determine our precipitation type, and the timing of a rain-to-snow changeover on Sunday.
  3. Where will the low track once it starts heading northeast?  If you scroll back up to the Sunday morning surface map, you can see high pressure in North Carolina.  This ridge, if it sets up, could serve as a "block" for our low, sending it slightly west and bringing us mostly rain.  Remember that a low cannot "move" a high; the high always wins when it comes to steering.  But if that high pressure is weak, or sets up father south or east, that allows the low to track through central Illinois and into lower Michigan, which would be the "ideal" storm track if you're hoping for heavy snow.  
Storm track is vital because the heaviest snow will fall the northwest and west of the low, as its counter-clockwise circulation pulls cold air in behind it.  The low position shown below (0z GFS-FV3 model; showing 7pm Sunday) is ideal for heavy snow in southern Wisconsin.  So it's no wonder that this is the model that's currently throwing down 20" in Janesville.


As we get closer to the weekend, shorter-range mesoscale models will come into play, so we'll be able to see how NAM, WRF and the Canadian model are interpreting track and intensity.  It's at that point that the National Weather Service will determine if we have enough confidence and model consistency to issue a Winter Storm Watch.

I'll be updating this storm throughout the weekend, and will probably have predictions for you on Friday afternoon.

In the meantime, let me know if you have any questions!

~Scott

Sunday, November 4, 2018

Dueling Models and Our First Snow

GFS and the Euro just can't play nicely together.  Our two main weather models are fighting again, and if you're a fan of winter, you're rooting for the Euro to win.  If you don't, you're pulling for GFS.

If you've been keeping up with Rippeology, you know that my forecasts have favored the potential for early season snow and cold as we wait patiently for El Nino to set up.  In my Winter Outlook on October 14th, I discussed early season snow pack in Canada, and how that might impact our November.  In last week's post (pertaining to the current November 3-10 period), I showed you how cold and slow-moving systems could set us up for snow in the near future.

Those predictions look like they may verify.  It's snowing in northern Wisconsin this morning; some spots will see 3-5" by late afternoon.  And looking ahead 7-10 days, it's quite possible (likely even) that southern Wisconsin could see its first accumulating snow.

Before we discuss that, let's take a quick look at the short term:

  • Today's rain will amount to over an inch in spots and be accompanied by some fog.
  • Another system will quickly replace today's low. Look for rain to redevelop again Monday night as low pressure deepens to our southeast. 
  • For election day on Tuesday, that low will also create a fair amount of wind. Here's noon Tuesday wind as that low quickly scoots north of the Hudson Valley.  Be prepared for blustery conditions as you head out to vote.
(Remember you can click on any image to enlarge).


Once we get through Tuesday, things start to get interesting.  Reinforcing shots of Canadian air will bring us our coldest temperatures of the season.  But this is where the GFS and Euro models start to throw down different solutions.

Take a look at 6:00 Friday evening.  Note where GFS puts low pressure, off the east coast:

 
 Now look where the Euro puts the low:

 
As a result of the closer proximity of the low (remember, winds rotate counter-clockwise around a low), here's how the Euro sees surface temps at 6am Saturday morning...


And how GFS sees it...


So, the models disagree by 8 degrees in Madison.  But that 8 degrees makes a difference.  If the Euro is right, the cold pocket settles long enough to sustain snow.  If GFS is right, temps easily rebound to more seasonal levels and snow becomes less likely.

Watch how the Euro spills cold air into our area over the next 10 days...


With that cold air in place, Euro sees two chances for snow: 1) a slight chance Thursday night into early Friday, and 2) a better chance next Sunday night into Monday.  Again, GFS disagrees.

Here's 10-day total snowfall from GFS...


And from the Euro...


So, which model should we believe when they can't both be right (but could both be wrong)?  The Euro tends to do better in mid-range, and seems to handle winter storm tracks better.  But we're in a transitional window right now, fall to winter.  And seasonal transitions are really hard to call because temperature is SO important.  That's why that 8 degree difference caught my eye.

I still favor snow for south-central Wisconsin over the next 10 days.  How much remains to be seen.  But rest assured I'll be watching it closely and keeping you posted!

~Scott

Saturday, October 27, 2018

Wintry Weather in Week 2?

GFS isn't showing it.  The Euro isn't showing it.  But it looks to me like conditions may be right for some wintry weather in our area starting next weekend.

Let's start with the surface map for Friday morning, November 2nd:


That low to our south looks to be a slow mover on Friday and Saturday.  Depending on how it ultimately tracks, it could produce some snow in the upper Midwest on Friday night. 

There are plenty of indications that we'll have a sufficient cold pocket in place to support snow over a 7-10 day period, starting with that low on Friday.

There's the Euro ensemble throwing down cold...


GFS showing anomalous cold in the middle part of the country...


The CFS weekly with troughiness in the middle of the country for the 11/3-10 period...


And the day 10 GFS showing a clear path for low pressure systems out of the Gulf of Alaska to make their way to the center of the country...


All of these indicators of cold and energy would suggest the possibility of wintry weather.  It's far too early to get more specific than that, but now you know what I'll be watching for over the next several days!

Sunday, October 14, 2018

Winter 2018-19 Outlook: 3 Winters?

Hey. Whaddup?

It's been over a year since I blogged on Rippeology.  Last year's winter forecasting was done almost entirely on Facebook, and for those of you who followed along, thank you.  I was very happy with the way last year's forecasts verified.

So, apparently I post a lot about weather.

Before we take a look into the Winter 2018-19 crystal ball, a few notes for those of you who might be new to Rippeology:
  1. I'm not a meteorologist. I'm a self-taught weather hobbyist who has geeked out over this stuff since I read my first weather book at age 7. I don't suck at it, but I wouldn't plan your March wedding around this forecast either.
  2. Like almost all of my weather forecasts, this one is specific to southern Wisconsin. If you live somewhere else and need some prognosticating, feel free to ask me on Facebook.
  3. In the event you find it stupid that anyone would barf up a winter outlook this early in the season, let me say this: there's a high degree of confidence in the guidance already available. I could wait another month, but with the exception of some clearer definition around long-range patterns (LRC, for example), I don't expect much to change. So a forecast today would be pretty similar to one I might issue closer to Thanksgiving.
  4. Finally, the impatient among you are welcome to scroll all the way down to the end of the post for the summary forecast. That, however, would make you a bit of an asshole, considering I put a lot of work into this. So if you do that, don't tell me you did. 

All that being said, let's get to it.

First, let's set the stage with some basic information:
  • Average snowfall in Madison is about 50" per year.
  • Last year, Madison got about 44" of snow. About half of that fell from mid January to mid February. Follow the green line (click any image to enlarge):

  • Last year, we experienced a couple of pretty cold periods: from December 23rd to January 6th; from January 13th to 17th; and from February 2nd to 12th. Also unseasonably cold in early April.
Any winter outlook should start, in my opinion, with a look at ENSO trends, or El Nino Southern Oscillation. You've heard of El Nino (warmer than average sea surface temperatures in the eastern equatorial Pacific Ocean) and La Nina (cooler than average temps in that same region). Those temperatures impact global weather, sometimes mildly, sometimes extremely. Here's a typical Nino winter in North America:



Generally speaking, El Nino results in a somewhat warmer and less snowy winter for us in Wisconsin. La Nina generally produces the opposite. Remember the winter of 2007-08 and our record 102" of snow? That was a moderate Nina year.

But these are not absolute predictors.

Last winter, weak La Nina conditions were present and our snowfall was just a bit under average.  This year, the Climate Prediction Center estimates a ~70% chance of El Nino conditions developing in the Nov/Dec/Jan/Feb period.


In fact, most models predict sea surface temperatures between 0.5 and 1.5 degrees (Celsius) above normal. We can categorize that as a weak-to-moderate El Nino.


Knowing that, we can look back at Madison winters with similar moderate Nino conditions. It happened in the winters of 2002-03, 2009-10, and 2014-15.  Here's Madison snowfall for those 3 winters:
  • 2002-03: 29" of snow
  • 2009-10: 52" of snow
  • 2014-15: 34" of snow
So what the hell happened in 2009-10?  An 18" blizzard on December 9, 2009. Video:
 
Take that freakish storm out of the equation, and all 3 of these weak-to-moderate Nino winters produced about 40% less snowfall than our 50" average.
So, if ENSO modeling was all we had, we'd be looking at a better than 50/50 chance of below average snowfall for Madison this winter. But there's more to consider.

For instance, we know that snow cover in Canada is already running way above normal right now.


Why is that important?  Because snow pack further cools the Canadian air that will continue to spill south throughout the fall. The temperatures we've been experiencing in October have been much below normal, due in part to the snow-cooled air that follows the cold fronts coming through southern Wisconsin. If that continues (and there's no reason to believe it won't), the colder air interacting with surface lows in our area is likely to produce frozen precipitation a little earlier than we might otherwise expect it.

That, and the fact that El Nino won't fully set up until mid-winter, is a hint of early snow in our neck of the woods.

What else we got?  Well, we have models up the wazoo. Meteorologist Ben Noll publishes them on his website (click on "snow maps" in the upper right).
  • The ECMWF (Euro) model thinks we'll have below average snowfall every month except, possibly, December.
  • The UKMET model thinks we'll have above average snowfall January through March, with February being especially snowy. But the British are prone to hyperbole.
  • The MeteoFrance model agrees about a snowy February, but thinks every other month will be below normal.
  • As far as temperatures go, the CFS monthlies have us above average every month except February.
See a trend there? With the exception of the Euro, models are currently pointing to a colder, snowier February.

So which models do we trust?  All of them combined, but none of them individually.  What I really look for in these long-range models is some level of agreement.  And we have that to some degree.

Let's get to it then.

SUMMARY FORECAST:

OMG, did you actually scroll all the way down here without reading the rest?

OK then.  Here's where I think we are.  We get 3 phases of winter:

November and December: average snowfall (about 18-20"); average temperatures.

January to mid-February: below average snowfall; slightly below average temperatures until we start to see Nino impact.

Mid-February to early April: slightly above average snowfall; slightly above average temperatures.

All totaled, I think we finish the winter 10-25% under our average snowfall, and while there will be occasional cold spells, I don't see a prolonged deep freeze for us.

STUFF YOU'LL ASK ME:

Snow for opening weekend of gun season?  Probably not, or at least not much.

A white Thanksgiving?  The RRWT guidance indicates a cold snap November 21-26, so a snowy Thanksgiving weekend isn't out of the question.

A white Christmas?  Likely.

Best time to take a warm weather vacation?  Last two weeks of February.

Now, do me a favor if you would.  Go back to Facebook and share my original post linking to this blog.  The more people following Rippeology, the more fun I have doing weather for you guys.

Thanks for reading!

~Scott