A different blizzard track

Looking at the National Weather Service's hazards map for today, it's clear where the story is:

Another powerful snowstorm forecast to hit the northeast.  Let's run through a quick overview of what the current model forecasts are doing with this storm.

Here's the GFS sea-level pressure (black contours) and relative humidity at the surface (colors; not as important here...) starting with a 24 hour forecast for tomorrow morning at 12Z.  I get these graphics from Weather Underground's Wundermap utility, which not only lets you zoom in on various areas but also is the only place to get full-resolution ECMWF images for free...

We can see that there's a 1000mb low over northeastern Lake Huron and central Ontario.  Note that this is very different from the first blizzard at the end of January, which moved up the coast from the south.  This sort of storm that comes out of the west-northwest is commonly referred to as a "clipper" system, as they tend to be rather fast moving.

Twelve hours later on Saturday evening, the low is forecast to basically be centered over New Jersey.

It hasn't deepened too much, but the storm is about to move out over the northern Atlantic.  Here's a map of the current sea-surface temperatures off the east coast from NOAA/NESDIS:

You can see right near the coast the temperatures are around 4-5 Celsius (39-41 Fahrenheit) but they rapidly warm into the 15-20 Celsius (59-68 Fahrenheit) range the further you get from the coast.  There's a sharp gradient of temperature there, and that's exactly what a storm like this can feed on to grow.  Storms form as the atmosphere's response to temperature and pressure imbalances.  The atmosphere wants to try and "smooth out" the temperature as much as possible.  Therefore, when a developing storm meets an region like this (a "baroclinic zone" if we want to have a technical term), it tends to deepen quickly, and that's exactly what we see.  Here's the forecast for Sunday morning at 12 Z:

An explosion of contours!  The low has deepend to below 976 mb as it feed on the baroclinic zone off the coast.  In just 12 hours the low has deepened nearly 24 mb---clearly meeting the 24mb in 24 hours deepening rate we use for "bombogenesis".  This deepening tightens the pressure gradients and will increase the winds.  Here's the GFS wind forecast for Sunday afternoon:

Strong, 50 kt. winds forecast over Cape Cod and along the new England coast.   The winds along New England and the Mid-Atlantic states at this point have strong northerly-northwesterly components (offshore winds) and they will be bringing in much colder air at the low levels.  The flow above, however, is wrapping around the low from the east over Nova Scotia and into New England, bringing moister, oceanic air over this low-level cold air and setting the stage for snow.  Here's the GFS precipitation forecast for Sunday afternoon:

Heavy precipitation along the New England coast.  The ECMWF forecast has slightly less precipitation at this time, but the position of the low is very similar:

All in all, it will be another blizzard for the New England coast...

Stumbling about in the snow

Let's talk a bit about the forecasting of this blizzard event in the northeast yesterday and today.  A lot of people are writing about what went into this forecast and why New York and Philadelphia did not get nearly as much snow as was forecast (thought Boston and points east did!).  The common theme?  Forecasters didn't adequately communicate the uncertainty regarding this forecast.

Just how uncertain was the forecast?  Here's an example of the predicted snowfall accumulation at JFK airport in New York City from several of our model systems---the NAM (the reds and greys), the GFS (the blues) and the Rapid Refresh (green) starting at 12Z yesterday (yesterday morning).

The numbers drop over time after reaching their peak because this forecast factors in melting and compaction of the snow which will decrease its depth over time.  You'll notice that all of the higher forecasts (20-34") are all red, orange or grey---these estimates all were based on forecasts from the NAM model.  On the other hand, the lower forecasts (6-12") are all based on the GFS (or RAP) forecasts.  In total, this is a lot of uncertainty---as few as 6" or as much as 34".  6" is a normal, everyday winter snow.  36" will keep a city shut down for days.  And this is the morning before the event, when leaders and other interested parties have to make decisions about committing resources, closing things down, etc.  What is a forecaster to do?

It turns out that the GFS had the low pressure center moving further east.  Here's the 30 hour GFS surface pressure forecast (the black contours) from the 12Z run, so this this a forecast for where the low would be around the middle of the day today:

Note the center of low pressure is forecast to be southeast of Martha's Vineyard, with a longitude east of Cape Cod.  This GFS forecast kept the heavier snow bands east over Long Island, but not over New York City, hence the lower snow totals that we saw above.  Here's the exact same forecast, but from the ECMWF model (thanks to Cliff Mass for grabbing these):

Here the low pressure center is further to the west---centered straight south of Cape Cod at this time.  The ECMWF had snow bands further west as well---over New York City.  The NAM happened to agree more with the ECMWF as to the position of the low.

From reading the National Weather Service forecast discussions, this seemed to be what led to the decision to go with the higher snow totals.  This is from the 630 AM EST forecast discussion yesterday at the New York Weather Service Office:

"THE 00Z ECMWF REMAINED VERY CONSISTENT WITH ITS
PREVIOUS RUNS. IT DID LOWER QPF SLIGHTLY...BUT THE OVERALL EVOLUTION
OF THE SYSTEM REMAINED SIMILAR TO ITS 12Z RUN. DID NOT WANT TO MAKE
DRASTIC CHANGES TO THE EXPECTED SNOW AMOUNTS AND HEADLINES WITH
JUST ONE MODEL CYCLE. IN FACT...THE LATEST 06Z NAM HAS COME INTO
CLOSER AGREEMENT WITH THE 00Z ECMWF."

Earlier in the discussion, though, they suggest that the spread in the models, particularly at such short forecast lead times, was more than they would like.  And again in the 1:34 PM EST discussion after the 12Z models shown above had come in:

"THE STORM APPEARS ON TRACK. THE 12Z GFS IS AN OUTLIER FROM THE 12Z
NAM AND 12Z ECMWF. HAVE UPPED SNOWFALL TOTALS ACCORDINGLY WITH
MOST AREAS IN THE 20-30 RANGE FOR THIS EVENT."

So that was the logic.  And, in the past, this logic has served us well.  The European model (the ECMWF) has long been considered the "gold standard" of our weather prediction, usually surpassing the US GFS model in terms of forecast skill.  Indeed, it was the ECMWF's superior performance to the GFS in predicting the path of Hurricane Sandy that recently led to a large investment from Congress to improve the GFS.

What's more, even by 00Z yesterday evening, it was hard to distinguish which model is doing better.  The 12-hour forecasts from the GFS, ECMWF and NAM had the low in virtually the same place at that time, so it was hard to determine which way it would go.  It's also challenging when the low is off the coast, as it was here, to get an accurate fix on what's going on...there are limited surface observations over the ocean, so we have to rely on satellite data which can be tricky to interpret.

So what ended up happening?  The low actually moved to the east, as the GFS had been predicting.  Here's a map put together by @anthonywx showing the analyzed position of the low today at 15Z (the black L) and where the ECMWF forecast it would be 3 hours later (the red L, same as the position shown in the maps above):

So the GFS solution ended up playing out---the low was further east as was a lot of the heavy snow.  This didn't stop Boston and Massachusetts from getting heavy snow, but it definitely pulled things away from New York and Philadelphia.

Note the difference in the position of the low---120 miles.  At a quick glance, as far as a 30-hour model forecast goes, that really isn't too bad.  It just so happened that there was a strong gradient in snowfall on the western edge of this storm, at that gradient happened to pass through the largest city in the country.

There was a lot of uncertainty, but forecasters used knowledge that has served them well in the past to refine their forecast in support of decision makers.  Personally, I think that this is a good case of "better safe than sorry".  Would giving the public better estimates of the actual forecast uncertainty (which was large in this case) have helped people make better decisions?  Should we have put up graphics that said the snowfall could be anywhere from 6"-36" in New York City?  What was the actual forecast uncertainty? Again, I think that it was right of the forecasters to use their experience to try and refine that number, even if it may not have been the best refinement here.  Could an automated system have done better?  Much will be debated in the days to come.

And, for what it's worth, the GFS brings another Nor'easter through the area next Monday.  Who (or what) will we trust then?

A quick note on the southern plains blizzard

Heavy rain is falling over central Oklahoma, southern Kansas and north Texas right now as a powerful surface low is deepening over the Texas Panhandle.  Here's the latest map of surface observations from the area.

There's a lot going on with this map, so I brought it into Powerpoint and have done a quick analysis of what I feel are the key features going on here.

This is a classic pattern--you can see that a large area of moderate to heavy snow has developed to the northwest of the surface low where winds are out of the north at 30-35 knots sustained.  Definitely a winter storm.  As this low deepens and moves eastward, points further east are also going to start feeling the winter punch.

So what is driving the development of this cyclone?  Notice that I've drawn a cold front on the eastern side of the map heading off to the northeast.  This cold front is associated with another surface low pressure center way up over central Quebec in Canada.  You can see that in this surface map over the entire continental US from this afternoon.

This cold front is associated with (like all fronts) a baroclinic zone--a zone where there is a large horizontal gradient in temperature.  Along this cold frontal boundary (which runs from northern Michgan down through the Chicago area and back toward the developing surface low in Texas), there is indeed a temperature gradient--from the low 50s in central Illinois back down to the low 30s in Iowa.  That colder air to the north behind the front is associated with a strong high pressure center, here analyzed over the Dakotas.

Remember that surface cyclones in mid-latitudes (like those we see over us) rely on horizontal temperature gradients as their source of strength.  They feed off of these baroclinic zones.  All they need is something to get them going, some sort of lifting mechanism to help support pressure falls at the surface.  And--what do you know--we have a shortwave trough aloft that's moving out of the desert southwest today (as per this 18Z RUC analysis this afternoon at 500mb):

So we have a shortwave trough aloft moving over a low-level baroclinic zone (that pre-existing cold front associated with the weakening low in Canada).  Excellent ingredients to spin up a cyclone.  Notice that there is a small, cyclonically curved jet streak on the southern and eastern side of this low aloft.  The exit regions of such jet streaks are favorable places for upward vertical motion, which will help lower the pressure at the surface.  Notice that the exit region of this jet streak is right over the location where the surface low is deepening over north Texas.  This is not a coincidence...

All of the major models are in pretty good agreement that the surface low will move eastward over Oklahoma over the next 24 hours (at least, the GFS, ECMWF and the NAM).  And, fortunately for our forecasters, that looks to be what is happening.  Here's a map of the previous 3-hour pressure changes.  You can see that there's a concentrated area of pressure falls in central and eastern Oklahoma.  Since the pressure will tend to fall as a low pressure center approaches, this gives us a good indication that the low is headed that way:

Going back to the surface map above over then entire United States, you can see that the high pressure over the Dakotas associated with the cold air behind that dying cold front is helping to generate a strong pressure gradient between the high pressure center and the low pressure center in Texas.  This strong pressure gradient is helping to really speed up those winds on the northern side of the low, contributing to the blizzard-like conditions.

So, we have a big weather event on our hands.  People in Oklahoma, Kansas and Texas should heed all blizzard and winter weather warnings and avoid travel if at all possible.  I'll probably post again tomorrow on this storm as it continues to move east into the Mississippi Valley.  A complex forecast for precipitation type is setting up across the upper midwest as a variety of factors come together to make forecasting difficult.  It will be interesting to watch.

The Dry Slot--Blizzard Killer?

Well, it's around midnight central time now--and the snow is still falling for many people.  But it ended rather soon for many as well--too soon, given some of the forecasts.  However, earlier Tuesday it was a very, very impressive looking storm--

Fig 1 -- Radar mosaic from 2008Z, February 1st, 2011.

This storm has a very well-defined structure.  The center of the low pressure at this point (20Z) was in southeastern Missouri.  This is slightly faster and further north than our forecast yesterday led us to believe--here was Monday's 36-hour forecast of the surface pressure field at 00Z Wednesday--four hours after the above image.

Fig 2 -- GFS 36 hour forecast of surface temperature (shaded) and mean sea level pressure (contoured) for 00Z, February  2nd, 2011.

Note how it had the low pressure center in nearly the right place (maybe a bit too far southeast) but...it had it there four hours later than it actually got there.  So this storm was moving faster and slightly further north than models anticipated.

Is it really that big of a deal?  The models generally got it right after all.  However, this small timing and track difference has huge implications for what areas expect to see the heaviest snowfall, what areas will see ice instead of snow, and what areas will see the snow end sooner than expected.

A lot of it has to do with the position of something commonly called the "dry slot".  This is a standard feature found in most well-defined mid-latitude cyclones.  So what is it?

Mid-latitude cyclones are often described as having "belts" of air moving around them.  Think of this like "jets" of air being moved around the low pressure center.  As we know, air wants to move from high to low pressure and air spins counter-clockwise around areas of lower pressure in the northern hemisphere.  We can track air from different regions as it is drawn toward and around the cyclone.  I'm going to focus on two different kinds of air--warm, moist air from the south (think Gulf of Mexico) and cooler, drier air from the northwest (think the high plains or central Canada).  How will these different kinds of air be drawn into the low?

Fig 3 -- Schematic of different "conveyor belts" of air moving into a classic mid-latitude cyclone.

The above schematic shows what typically happens.  Warm, moist air to the south is advected northward to the east of the low.  In fact, the advance of this warm air helps define and drive the warm front to the east of the low.  However, the air already in place gets colder the further north we go.  So as this warm air moves northward, it also begins rising over the colder air already in place.  The closer to the low pressure center the air gets, the more it gets "caught up" (there are more technical details behind this...) in the circulation and gets "wrapped around" to the backside of the low. Because it rises up off the ground more, this air tends to move much more freely and can be somewhat easily brought around to the back side (western side) of the low.

In contrast, the cool, drier air to the northwest isn't necessarily warmer than the surrounding air.  In fact, it's being brought south toward much warmer air.  Because it's cooler than its surroundings as it heads south, this air doesn't rise much and tends to stay near the surface where there's lots of friction (among other things).  As such, this air doesn't move as freely and, while it still somewhat gets caught up in the low's circulation, it can't be as freely advected around the low.  So it usually kind of just gets drawn into the center of the low.  We can see that in the tan arrow of the diagram above.

Of course, to get any kind of precipitation, be it snow, rain, freezing rain, ice, whatever--you need moisture.  The air in the "warm conveyor belt" has lots of moisture.  The colder, drier air moving in from the northwest does not.  As such, we only really see precipitation associated with the warm conveyor belt.

Fig 4 -- Schematic of precipitation areas in association with "conveyor belts".

The area where there is much drier air being advected around and into the low is relatively devoid of precipitation because there is no moisture there.  This area is what is called the "dry slot."

The above diagram should make a whole lot of sense--we see this all the time in our cyclones.  Look how this diagram matches up to the radar image above:

Fig 5 -- Overlay of "conveyor belt" diagram with the above radar image.

This explains that comma-shaped pattern to the precipitation--the warm conveyor belt feeds the areas of precipitation.  Dry air being wrapped into the low from the northwest/southwest, in contrast, forms that "slot" where there is no precipitation--the dry slot.

Now we can explain why subtle changes in the track forecast for this storm are so important.  Take the area around Saint Louis.  Previous forecasts had called for 12-15 inches of snow in the Saint Louis metro area on Tuesday.  They were based on the more easterly track of the low pressure center models had been hinting at.  This would bring that heavy "wrap-around" precipitation on the back side of the low right through Saint Louis and give them an extended period of snow well into tonight.

But...the low was slightly further north and faster than expected--so instead, much of Saint Louis and points south got into the dry slot, which they were not necessarily expecting.  I've heard spotter reports of only 4-5 inches that have fallen instead of the forecast 15.  In contrast, areas north of Saint Louis like around Quincy Illinois, did receive some 10-15 inches of snow--they stayed in that wrap-around precipitation while the dry slot stayed to the south and east.  This is why small changes in forecast position can mean the difference between a good forecast and a bust.  Of course, Saint Louis is still seeing blizzard-level winds, so the blowing and drifting of that snow is creating huge problems.  The blizzard warning is by all means warranted and it is still a very dangerous situation out there...

One additional note about dry slots.  As the leading edge of this low-level drier air advects northward along with the surface low, enhanced convergence can often form near the surface.  Think of this kind of like a sort of dryline like we see during severe weather episodes in the southern plains.  Convergence along the edge of this dry air mass mixing along can be enough to fire off some convection.  Here's a radar image from late Tuesday evening as the leading edge of the dry slot began pushing into southern Michigan:

Fig 6 -- KIWX 0.5 degree base reflectivity from 0706Z, Feb. 2, 2011.

Note how the precipitation in the area I circled seems more cellular--less like the widespread swaths you see up north?  That is convection forming along the leading edge of the dry slot.  Often these are the best places to find that awesome weather phenomenon that is thundersnow.  In fact, as the dry slot approached Chicago earlier Tuesday evening, thundersnow was indeed reported there.  Pretty amazing stuff...

Anyhow, this storm continues to move east--tomorrow is going to be a hectic day for much of the northeast and quite the headache in the central US as recovery from the storm begins.

Two short waves, one big storm, Chicago blizzard?

It's been a week or so since I last posted on here.  The annual meeting of the American Meteorological Society was here in Seattle last week and needless to say I was very busy during that time.  But now that's done and my schedule can resume something more normal...

Today's National Weather Service summary map looks pretty spectacular:

Fig 1 -- NWS watches and warnings as of 2300Z, January 30, 2011.

The darker purple from Montana down through the high plains and into southern Wisconsin is all winter weather advisories.  The pinks through much of Missouri and eastern Kansas, Nebraska and Oklahoma are winter storm watches.  And what about that bright green blob surrounding Chicago?  That's a blizzard watch for the Chicago CWA.  Milwaukee is also issuing blizzard watches at this time.  I quote the following from Milwaukee's latest Winter Weather advisory:

"NEEDLESS TO SAY THIS COULD BE A HISTORIC BLIZZARD CAPABLE OF

PARALYZING PARTS OF SOUTHEAST WISCONSIN."

Pretty powerful words.  So what's the setup for this?

Right now, there is a highly amplified ridge over the west coast with broad scale troughing aloft across the eastern half of the country:

Fig 2 -- GFS 12Z analysis of 500 mb geopotential heights and vorticity on Jan 30, 2011.  From the HOOT website.

Note two individual shortwaves are evident in the west embedded in the larger-scale flow.  One is the nearly cutoff shortwave in southern Oregon and northern California.  The other is further north on the British Columbia/Alberta border in Canada.  This northern shortwave is located in the region we usually see "Alberta Clipper" type storms form.  It's somewhat unusual to have two shortwaves stacked on top of each other like that. Latest model forecasts show these two shortwaves merging together over the plains by late Tuesday:

Fgi 3 -- GFS 60 hour forecast of 500 mb geopotential height and vorticity valid 00Z, Wednesday, Feb. 2, 2011.  From the HOOT website.

Cold air in association with the northern shortwave is forecast to spill down into the plains as that shortwave moves south.  We can see a blob of colder temperatures moving down across the northern plains at 850mb on Monday:

Fig 4 -- GFS 30 hour forecast of 850mb temperatures and geopotential heights valid 18Z, Monday, Jan. 31, 2011.  From the HOOT website.  

Of course, there's already very warm air to the south in Texas and along the Gulf Coast.  As the cold air moves south, this is going to increase the temperature gradient across the central part of the country.  You can already see how quickly the 850mb temperature changes from Nebraska into Oklahoma on the image above.  This increasing temperature gradient means two things:

1. Frontogenesis is going on at the low-levels--a cold front is most likely developing from Iowa down through Oklahoma at the time above.
2. Based on those thermal wind arguments (yes, those again...), the increasing temperature gradients below translate to strengthening winds aloft parallel to the forming front.

Do we see winds increasing aloft in this area?  We do indeed see a jet streak strengthening in that area--and all the way back down into west Texas.

Fig 5 -- GFS 36 hour forecast of 300mb winds and geopotential height at 00Z, Tuesday, Feb 1, 2011.  From the HOOT website.

Quite the dynamic pattern here.  Not only can we see those two shortwaves starting to merge together, but look at how amplified that ridge along the west coast has become.  It even has a north-south oriented jet streak!  That's pretty rare.  We also have two jet streaks that are roughly east-west oriented--one somewhat associated with the exit region of the northern trough and the other associated with the exit region of the southern trough.  If we look at these in terms of our four-quadrant jet streak model:

Fig 6 -- Same as figure 5 but annotated with the four-quadrant model showing regions of convergence and divergence associated with jet streaks.

We see that divergence regions of both of these jet streaks almost "coincide" over the same region.  This is a form of a phenomenon known as "jet streak coupling", where the combined effect of multiple jet streaks can really enhance the divergence (or convergence) aloft.  Of course, with this huge amount of divergence going on aloft, the pressure at the surface is going to fall rapidly:

Fig 6 -- GFS 60 hour forecast of surface (2m) temperature, MSLP and wind barbs valid 00Z, Wednesday, Feb 2, 2011.  From the HOOT website.

There's a strong surface low forecast by the GFS to move into southern Illinois and continue trekking northeastward into Michigan.  Look at the tight pressure gradient forming across the upper midwest and into the southern Plains.  This translates to very strong northeasterly winds across the upper midwest on Wednesday.  So if this model forecast were to verify, blizzard-level winds would indeed be possible on Wednesday across northern Illinois and southern Wisconsin.  The reasoning behind this blizzard watch begins to make sense.

But what about snowfall amounts?  How cold will it get?  We'll take a look at the models again tomorrow sometime and see how things are continuing to shape up.  For now, just be on the lookout for a strong winter storm during the middle of this week.