March slopfest coming?

Winter just refuses to let up even as the days lengthen and we head into March. The average high for tomorrow in DCA (Reagan Nat'l Airport) is 51. The forecast high is only 27. The is all because of a persistent flow of arctic air from Canada. And of course, with any mention of cold, we also need to mention the polar vortex. It is true that this is what is contributing to the cold. This low pressure, which usually sits over the north pole, has part of it sitting over Canada. The farther south is dives, the farther south the cold air dives with it.
Where the Polar vortex is position will determine how cold we get. If it is positioned to the west, the cold will be to the west and we will be under the influence of the subtropical ridge, which will pump warm, SE winds into the area. The polar vortex to the east, say Ontario, Canada, will better as this will allow the cold air to spill south into the area.
Models have been flip-flopping between There are two model camps, one wants to keep the PV to the west, therefore keeping it warmer here and more rain. This was the 00z euro run, which brought temps up into the 40s and 50s for out area. The most recent trend has been to shift south, allowing the cold air to rush in. The GFS was on the forefront of this trend, advertising an extreme cold air damming event. It progged temps to fall below freezing as the precip arrived, continuing to drop through the 20s on Monday and bottom out in the teens on Monday night, as the precip is pulling away. This would create a buffet of rain, freezing rain, sleet, and snow. P-type and the amounts of each precip varied between runs.

This is a Skew T chart...

It represents the vertical column of the atmosphere at any given point. This Skew T is for Silver Spring on Monday, March 3, at 7:00AM.

Now how to read them:

The bottom of the chart the ground level, the top is about 16000 meters up, or 10 miles.

The red line is in temperature so in this case, the surface temperature will be between 0 and -10 c

The dark green line that is roughly parallel to the red line is the dewpoint. It is not always following the red line, and when it does, it means that the air is saturated. If it curves sharply to the left, it means that there is dry air at that level.

The 850 temp is commonly used to mark the rain snow line. If the air is below freezing between the ground and the 850 line, you can assume that there will be snow. If there is a warm layer in between, there could be freezing rain or sleet. If the surface is above freezing, it will likely be rain.

To the right are some weird symbols. Those are wind barbs, denoting wind speed and directions at different altitudes.

This Skew T shows surface temps are well below freezing, in the mid 20s, but the 850 temps are still slightly above freezing. This would indicate a warm nose aloft from southerly flow, resulting in melting of snowflakes. Judging by the depth of the cold layer, it will be sleeting at this time. Notice that within the red circle, the bottom most 2 wind barbs are pointing to the bottom left. That indicates NE winds, this would reinforce the cold wedge and that is why we are below freezing at the surface. This is classic cold air damming. the only thing unusual is that it is during March. Notice the other wind barbs are pointing in the opposite direction, resulting in warm air overrunning the cold an producing precipitation.

Previously, the Euro had been the polar opposite of the GFS, keeping the PV in western Canada, and therefore, skimping on the cold air and letting the surface low ride to our north, keeping us in the warm sector for most of the run. Since the Euro has the best reputation, this also cast doubt on the GFS solution. The Euro also caught on in its last run, bringing the low center much farther south than before, and giving us a good hit of snow and  sleet with very little liquid precip. The operational run was also backed by the ensemble runs. These are lower resolution runs of the same model algorithm using slightly different data to account for any data errors. If the ensembles come out to be similar to the operational, it indicates that the solution is sound, there are no data errors that cause calculation errors to add up, significantly affecting the solution. It essentially verifies the solution. There are no graphics because they require payment for their data. all info is from other people with subscriptions, not from any personal observations.

The Canadian model gives a warm start to the storm before bringing it colder and giving us a good thump of snow on the back end. The Canadian has been flipping between cold and warm-ish solutions and never showed any all rain solutions.

Hour 106 Monday, 3AM

Hour 120, Monday, 7:00PM

Pitfalls:

1. Model limitations. Garbage in garbage out is the saying. The disturbances that will form this storm is still over the pacific and Canada. This means they have not been sampled enough. the models are taking wild guesses at what the storms are like.  If they guess wrong, then the outcome will be really wrong. A simple example: 2^32=4,294,967,296 2.001^4,364,222,021.24. That's a 10 million difference when you only change the number by .001. Imagine these computers doing billions of calculations a second for hours to run these models. A small error really does add up. It should take the storms another day or two to reach the US, where there are more data collection points so it will take another 2 days until we get a firm handle on what might happen. 

2. The PV position at the time of the storm is still unresolved. although the Euro was colder in its last run, it still has the PV farther west. The time in which it gets here will be critical, too late and we will just get rain with a little bit of snow. Stay in western Canada and we get warm rain. We need it to get here earlier.

3. The models still disagree on the magnitude and how the event will unfold. The GFS is calling for a strong first part, and then a weak second part. The Euro wants to keep it all weak. The Canadian wants to bring in the second part stronger and colder with a weak and warm first part. All these scenarios are still in play

4. Timeframe: we are still 3 days out. still plenty of time to shift favorably or non favorably. The 00z and 12z Euro runs were worlds apart. It only took 12 hrs to go from a non-even to a possibly major event. It could take the same time to go back too. we are also out of range of many of the short range, hi-res models. those would help a lot in pinning down the freezing line.

The bottom line is there is the potential of a major winter storm on Monday and Tuesday of next week. the precip types and amounts we get are not resolved yet and it will be a decent time before they are.

Thanks for reading!

-Alex

Geek Maps 2/25/14

This is the models page... It is a selection of short range, hi res models, which would have the best handles on small disturbances like this. That being said, they can also mess up. This is a model fail:

so is this:

We are talking about the second kind. If you were expecting anything sexy, get out... well the Euro snow maps for 10 days out can be quite beautiful. But those soon become model fails because of the law of snowstorms a week+ out which states that the Euro should have a beautiful Miller a track snowstorm, giving us a foot+ of snow in model la la land, only to turn it into a messy or dead storm as the time progresses.

Back to out little snow event. This snow band will be very narrow and small shifts can affect snow totals. All models give us between .1 and .25 inches of liquid. This would translate to between 1 to 2.5 inches of snow using the average 10:1 ratio. However, upper level temperatures will be cold enough to support 15:1 ratios, translating to 1.5-3 inches of snow. Snow should start in the early morning hours and progress through the morning commute before tapering off near 11:00 AM. At this range, we have entered nowcasting.

HRRR

 NAM

RGEM

RAP

All of these models paint the most precip across or just north of DC, aka MoCo, therefore, we have a 75% chance of a delay if all goes expected. A concern is that the storm's moisture gets wrung out over the mountains. Another concern is the late start of the snow. Radar will not look very impressive at 4 AM even 5 AM. The snowbank is modeled to form right over us just like today's band. The MoCo peoples might look at the radar and be like pshhhh, all clear, no delay. One thing that may be in our favor was today's overperformance. This may make the county a little more trigger happy and call the delay anyway.

HRRR simulated radar for tomorrow

Thanks for reading!

-Alex

Today was a surprise...

Today's snow was quite a surprise. It was not caught well by any of the models save for some short range models such as the NAM 4km Nest and the Canadian RGEM models. Global models did not foresee any snow for today. 1-3 inches of snow fell in the immediate DC metro, with less elsewhere, quite a interesting bullseye. It seems that more snow falls elsewhere most of the time. 2.8 inches fell at DCA, bringing the seasonal total to 18.3 inches at DCA.

Total snowfall today

The NWS has issued a winter weather advisory for tonight into tomorrow morning for snow with 1-3 inches possible.

THE NATIONAL WEATHER SERVICE IN BALTIMORE MD/WASHINGTON HAS
ISSUED A WINTER WEATHER ADVISORY FOR SNOW...WHICH IS IN EFFECT
FROM 1 AM TO NOON EST WEDNESDAY.

* PRECIPITATION TYPE...SNOW.

* ACCUMULATIONS...1 TO 3 INCHES.

* TIMING...SPREADING EAST BETWEEN 2 AND 4AM. SNOW WILL END BY NOON.

* TEMPERATURES...MID TO UPPER 20S.

* WINDS...LIGHT AND VARIABLE...BECOMING WEST AROUND 10 TO 15 MPH
  BY NOON.

* IMPACTS...SNOW ACCUMULATION WILL AFFECT THE WEDNESDAY MORNING
  COMMUTE. UNTREATED ROADWAYS WILL BE SLICK.

Official Snow forecast

These are the NWS forecasts around MoCo.

Location	Min	Likely	Max	0"	0.1-1"	1-2"	2-4"	4-8"	8-12"	12-18"	>18"
Burtonsville, MD	0	2	4	2%	26%	35%	37%	0%	0%	0%	0%
Chevy Chase, MD	0	2	4	3%	27%	32%	38%	0%	0%	0%	0%
Damascus, MD	0	2	3	4%	28%	45%	23%	0%	0%	0%	0%
Germantown, MD	0	2	4	3%	27%	37%	33%	0%	0%	0%	0%
Poolesville, MD	0	2	3	5%	28%	40%	27%	0%	0%	0%	0%
Rockville, MD	0	2	4	3%	27%	33%	37%	0%	0%	0%	0%
Silver Spring, MD	0	2	4	3%	27%	32%	38%	0%	0%	0%	0%
Takoma Park, MD	0	2	4	3%	26%	30%	41%	0%	0%	0%	0%

The WPC seems quite excited about this clipper.

DAY 1...

...MID-ATLANTIC/NORTHEAST...

ANOTHER STRONG...FAST MOVING SHORT WAVE WILL TRAVERSE ACROSS THE
MID-ATLANTIC EARLY HOURS ON WEDNESDAY AS THE JET STREAK
STRENGTHENING ACROSS THE REGION.  THIS WILL REINFORCE ANOTHER COLD
FRONT TO PASS THROUGH...AND WITH ENOUGH LIFT AND MOISTURE IN PLACE
A SECOND ROUND OF SNOW FOR THE MID-ATLANTIC IS EXPECTED.  THE
ECMWF/GFS AND EVEN THE NAM/SREF HINTED AT A SLIGHTLY ENHANCED QPF
BULLSEYE OVER THE DELMARVA REGION...BETWEEN 0.10-0.20 INCHES. 
KEPT THE ACCUMULATIONS IN THE MID-ATLANTIC AT 1"...BUT THERE IS
CERTAINLY A CHANCE TO SEE LOCALLY HEAVIER AMOUNTS. 
HOWEVER...SINCE THESE HIGHER TOTALS WILL BE LOCALIZED...DECIDED TO
KEEP OUT THE 4 INCHES OR MORE PROBABILITIES. 

For those who want hope...

This map basically shows the high end scenario of this storm. Although this is more likely than a low end bust, it is still quite unlikely. The highest amounts are more likely to the east since they will have the influence of a developing coastal low.
There are 2 ways of achieving higher than forecast amounts. The first , and traditional way is to have more precipitation than expected. This is a very dry system, a clipper. However, the atmosphere over our region has been sorta primed by the snow today. The dew points are higher, cutting down on snow lost to evaporation. In addition, this storm will have more upper level support, with a jet streak (local maxima of jet stream winds) over the area, this will allow some banding to occur and enhance snowfall rates. This may result in overachieving totals. This was the case today. The jet streak was right over the DC area so the snow fell over us. There is no guarantee that the bullseye will be over the same area again. The second way, is to have higher snow to liquid ratios. Upper level temperatures will be very cold, allowing snowflakes with open structures to form. This will mean there is more air in the snow but it would accumulate more. The snow will be light and fluffy.

We have a decent chance of a 2hr delay, maybe even a day off if the storm way over performs by a lot. Need to go do hw... Maybe I will come out with model outputs later but those are all about the same with .1-.2 QPF...

Thanks for reading!

-Alex

Snow? It's almost March...

Our pleasant week of 50s and 60s has been rudely shoved out by a cold front that passed this morning. Spring has been put on hold for at least another week or two. It seems that the polar vortex has dropped down for possibly its final visit of the season. We will be entering a cold (for March) and possibly stormy pattern, with multiple chances at snow although climatology dictates that the snow chances rapidly diminish as we get into March.

First up is a new experimental forecast with near 100% accuracy

Disclaimer: This is intended as satire and should not be taken seriously.

Second, the February 12-13 Storm was rated a 3, a major storm, on the The Northeast Snowfall Impact Scale (NESIS). Do you remember this storm? Yes, it was only 2 weeks ago. all that's left of it are a few dirty snow piles and fun memories. High instability over the region allowed heavy snow to develop. The "Deathband" that set up over the DC area on the night of Feb 12, giving us more than a foot of snow, far exceeding forecasts. A radar loop of that band can be found here. Dulles got 13.3 inches, BWI got 11.5, and Reagan National got a grand total of 7.0 inches of snow. Snow totals can be found here.

From the NOAA site: "NESIS scores are a function of the area affected by the snowstorm, the amount of snow, and the number of people living in the path of the storm. The aerial distribution of snowfall and population information are combined in an equation that calculates a NESIS score which varies from around one for smaller storms to over ten for extreme storms. The raw score is then converted into one of the five NESIS categories. The largest NESIS values result from storms producing heavy snowfall over large areas that include major metropolitan centers."

For more information about the NESIS scale and how past storms have scored on this scale, visit the NOAA page.

Third, is an analysis of the possible snow in the next week or so. It is almost March, meteorological spring. Before March, we don't seem to have much of a change of getting anything more than an inch or two of snow. 

The first threat will be on Tuesday night through Wednesday. some models also indicate the possibility of snow showers during Tuesday with highs in the mid to upper 30s. Up to an inch is expected though the NWS is even less enthusiastic than that, forecasting less than an inch across the area. If the storm does over perform, it will be to the north and east as the coastal low gets it's act together more quickly than expected.

The polar energy is too far displaced north to interact with the surface low to the south and create a storm.

More likely than not, this storm will fail to give us anything more than an inch. The rest of the week will be sunny, with highs in the low to mid 30s, unseasonably cold for this time of the year when highs average highs are approaching 50.

The second threat is on Friday night. this has the potential for a bit more, maybe 1-2  inches, only because it is farther out and there is less certainty. It certainly could also go out to sea without as much as a flurry. The storm is not likely to trend higher in snow totals. What seems like a bleak outlook is not so bleak. We are not missing storms because it is too warm, after all, we are approaching spring, but because there are too many disturbances and storms are getting pushed out too fast befor they have the opportunity to intensify. This reasons to believe that the last one of the pattern will have to most potential, with nothing to keep it moving along. This is the March 3-4 storm. The models agree to a storm around then, but do not agree with the track of precip type. But with a strong high to the north, we could squeeze a few inches out before, a changeover to sleet and freezing rain. Of course, this storm is 8 days out and we have seen many storms that have looked like blockbusters in this timeframe turn out to be duds. Case in point, the Friday non-event, which was modeled a week ago tby the Euro to be another major snowstorm.

If conditions warrant, I will come out with more snow forecasts, but as of now, it seems as though this may be the last Snow analysis post of the season. Maybe we can do a sever weather analysis, but that seems less exciting in my opinion. Maybe a cherry blossom and banana blossom watch... Anyways... As always... 

Thanks for reading! 

-Alex