You can't use the evening model runs alone to make a forecast but they can support a trend. And the trend is snow for the weekend.
Apparently, the National Weather Service is coming around to the idea as well. At 6 PM, they revised the forecast for our area to include a 50% chance of snow for Saturday night and Sunday.
So...game on. I'll be checking out the late-night runs tonight. Still 48 hours or so from the event - plenty of time for things to change for better or worse.
Early call would be: a decent chance of a "significant" winter storm Saturday night into Sunday. Storm track makes all the difference, as usual. Temperatures should easily support snow.
Thursday, December 16, 2010
This is What I Mean
Forecast for the weekend from the National Weather Service:
According to them, there's not enough model agreement to start worrying people about snow. I tend to agree.
But...
From the SAME National Weather Service, the precipitation forecast for Sunday:
Forget the models, even the humans aren't agreeing.
So we wait and watch. The evening models are running now but they're hardly ever used operationally (i.e. in service of an official forecast). Next big model runs late tonight - we'll see how they go.
Just a thought: the above map, if the forecast verifies, would put us well into the 5-inch range for snowfall on Sunday. Not the biblical snow of last year - but respectable all the same.
.SATURDAY...MOSTLY CLOUDY. HIGHS IN THE MID 30S. WEST WINDS AROUND 5 MPH...BECOMING SOUTHWEST AROUND 5 MPH IN THE AFTERNOON. .SATURDAY NIGHT...CLOUDY. LOWS IN THE LOWER 20S. .SUNDAY...PARTLY SUNNY. HIGHS IN THE MID 30S. .SUNDAY NIGHT...PARTLY CLOUDY. LOWS IN THE LOWER 20S.
According to them, there's not enough model agreement to start worrying people about snow. I tend to agree.
But...
From the SAME National Weather Service, the precipitation forecast for Sunday:
Forget the models, even the humans aren't agreeing.
So we wait and watch. The evening models are running now but they're hardly ever used operationally (i.e. in service of an official forecast). Next big model runs late tonight - we'll see how they go.
Just a thought: the above map, if the forecast verifies, would put us well into the 5-inch range for snowfall on Sunday. Not the biblical snow of last year - but respectable all the same.
Snow Falling Now, 1-3 inches, Models All Over the Place for the Weekend
Quick update. Snow has started falling in the DC Metro. If you're not seeing it, it's because you're in an office with no windows. Go outside!
Mostly light snow with occasional moderate bursts has been falling for a couple of hours now. Expect it to continue throughout the afternoon before winding up in the evening - maybe around 6 or 7.
The models still don't know what to do with the weekend storm. I've seen the local National Weather Service forecast add and remove snow chances for Saturday night about 5 times in the last 4 days. Almost all of the models show a low forming near the Gulf, almost all of them show it moving to the Carolinas and then some kind of model kegger breaks out and they all lose their minds. At least two show measurable snow for DC and the trend in the last two model runs is for a storm that is closer to us.
All we can do is watch. More to come when I know it.
Mostly light snow with occasional moderate bursts has been falling for a couple of hours now. Expect it to continue throughout the afternoon before winding up in the evening - maybe around 6 or 7.
The models still don't know what to do with the weekend storm. I've seen the local National Weather Service forecast add and remove snow chances for Saturday night about 5 times in the last 4 days. Almost all of the models show a low forming near the Gulf, almost all of them show it moving to the Carolinas and then some kind of model kegger breaks out and they all lose their minds. At least two show measurable snow for DC and the trend in the last two model runs is for a storm that is closer to us.
All we can do is watch. More to come when I know it.
Wednesday, December 15, 2010
Thursday Snow More Likely, Weekend Snow Less Likely
Quick post just to update the situation. Storm tomorrow (Thursday) looks to track far enough north so that we get some snow from it. 1-2 inches at the most.
The storm this weekend refuses to phase on 2 of the 4 model sets that I usually use. This leads to a solution where the developing low moves off the North Carolina coast and directly out to sea. Little or nothing for us.
Two other models - including the GFS - still have a coastal storm with somewhat more phasing present. Still need more model runs to determine the final story for the weekend.
Stay tuned.
The storm this weekend refuses to phase on 2 of the 4 model sets that I usually use. This leads to a solution where the developing low moves off the North Carolina coast and directly out to sea. Little or nothing for us.
Two other models - including the GFS - still have a coastal storm with somewhat more phasing present. Still need more model runs to determine the final story for the weekend.
Stay tuned.
Monday, December 13, 2010
Two Shots at Snow: Thursday and Late Weekend
Well, it's certainly cold enough to snow but we'll have to wait a bit for any moisture or possible storms to line themselves up. Still, the atmosphere is getting busy and there's plenty of energy to fuel storms that do form. Right now, in addition to some flurries and snow showers on Tuesday that shouldn't amount to much, we have two shots at significant snow in the next 7 days. The first is Thursday the second one is late in the weekend. That second storm is the one to watch closely.
But the storm Thursday could make a difference in the track of that second potential storm this weekend.
So let's talk about Thursday first.
A weak storm is on the models for Thursday moving northeastward from the southern U.S. and sliding just to the south of our area. The models have been fairly consistent with the track of the storm being to our south but a little inconsistent in how they describe the associated precipitation field.
Then the GFS and NAM runs late this afternoon showed a slightly more northern track which placed us in the precipitation area. According to these model, the atmospheric temperatures would be cold enough to support snow and indications are for about 1-2 inches of snowfall.
Still, the other models don't agree. They show accumulating snow for southern Virginia and North Carolina only.
I'm wary of trusting one model run - but here's the GFS:
As you can see, the track of the center is predicted by this run of the GFS to be more like along the VA/NC border instead of through central NC where it had been forecast earlier in the week. It bears watching but will almost certainly be a low-impact storm even if it does take a more northern track. You can see in the map above that over land, the maximum precipitation is between 0.10 and 0.25 inches which if we assume a 10:1 ratio (the rule of thumb) means about 1-2 inches.
Here's the crazy thing though: if you love snow (like I do), you should probably hope this doesn't happen.
I can see all of you snow lovers out there now:
Bear with me. Let's talk about the weekend storm potential first and then we'll come back to this crazy "if you love snow, you may not want it to snow Thursday" idea.
Long-range models have had a fairly large coastal storm for December 18-20 in our area for some time now. Ironic since, if it does happen, it will be right around one year to the day of the first huge snow in 2009. A big storm on the anniversary of the largest one-day December snowfall in Washington, DC history.
Anyway, when you're watching long-range models a few weeks out, here's the most important thing to look for until about a week before the predicted event: is the storm still there with each consecutive model run?
As far as that goes: so far so good.
Now that we're inside of a week away, we also start to look at storm track. Here's where the models are, to use technical terms, all over the damn place.
For now, we'll take the GFS as a fairly reliable long-range model. Here's what is shows for Saturday evening:
A developing storm moving from the northern Gulf of Mexico and across the southeastern US, emerging just off the coast of South Carolina. To the northwest, a solid push of cold air that shows the freezing line to our south. According to the GFS, still some 120 hours away from this forecast time, whatever is falling Saturday night will be snow.
6 hours later the GFS has the storm intensifying rapidly. Snow really gets going especially in central VA which looks to be the "bullseye" for this run of the GFS. Up here in the DC area we have less snow as the storm slides along the North Carolina coast - more eastward than northward.
6 hours more and the low is moving off the coast throwing some precipitation back into our area but the motion of the storm suggests lighter amounts than anything we saw on the 19th of December last year:
The east-northeast motion combined with the absence of any "bombing" or rapid intensification indicates that the GFS is predicting that the jet streams will not phase. I think I wrote about phasing last year but if not: here's a quick description:
Flow patterns in the upper atmosphere occur in sequences of waves. These waves can and do interact with one another sometimes in a way that is destructive, sometimes in a way that amplifies the original wave. When waves combine to form a larger wave, it is called "phasing". In meteorology, especially in the winter, we look for phasing of the northern and southern branches of the jet stream which tends to amplify existing storms. It's a tricky thing with respect to timing and it doesn't happen all that often, relatively speaking.
If the phase doesn't happen, you get a weak storm like the one predicted for Thursday. If it happens too early, a storm like the one above goes too close to the coast and we end up with rain. If it happens too late (like in the model runs above), it goes out to sea before it can get really wound up.
All that said: it's a good model run for us (remember "snow = good"). This far out, details like storm track and phasing are several runs away from being decided. We have the storm on the models still - that's one victory. The storm moves up the coast - that's another.
Now. What's bad about the Thursday storm?
Because of a complicated series of atmospheric interactions that have to do with something called the "Polar Vortex", the probability of an out-to sea solution for the December 19th storm (like the models show above) probably increases the further north the Thursday storm tracks. So if you want the potential for a big storm on Sunday, you want that Thursday storm to be as far to our south as possible.
Lots still to be worked out. We'll be closer to a solution for the Thursday storm tomorrow. I'll be keeping an eye on things!
Wouldn't a monster storm on the anniversary of a monster storm be cool?
Just to jog your memory: 12/19/2009:
But the storm Thursday could make a difference in the track of that second potential storm this weekend.
So let's talk about Thursday first.
A weak storm is on the models for Thursday moving northeastward from the southern U.S. and sliding just to the south of our area. The models have been fairly consistent with the track of the storm being to our south but a little inconsistent in how they describe the associated precipitation field.
Then the GFS and NAM runs late this afternoon showed a slightly more northern track which placed us in the precipitation area. According to these model, the atmospheric temperatures would be cold enough to support snow and indications are for about 1-2 inches of snowfall.
Still, the other models don't agree. They show accumulating snow for southern Virginia and North Carolina only.
I'm wary of trusting one model run - but here's the GFS:
As you can see, the track of the center is predicted by this run of the GFS to be more like along the VA/NC border instead of through central NC where it had been forecast earlier in the week. It bears watching but will almost certainly be a low-impact storm even if it does take a more northern track. You can see in the map above that over land, the maximum precipitation is between 0.10 and 0.25 inches which if we assume a 10:1 ratio (the rule of thumb) means about 1-2 inches.
Here's the crazy thing though: if you love snow (like I do), you should probably hope this doesn't happen.
I can see all of you snow lovers out there now:
 |
| "Crazy fool's done lost his freakin' mind!" |
Bear with me. Let's talk about the weekend storm potential first and then we'll come back to this crazy "if you love snow, you may not want it to snow Thursday" idea.
Long-range models have had a fairly large coastal storm for December 18-20 in our area for some time now. Ironic since, if it does happen, it will be right around one year to the day of the first huge snow in 2009. A big storm on the anniversary of the largest one-day December snowfall in Washington, DC history.
Anyway, when you're watching long-range models a few weeks out, here's the most important thing to look for until about a week before the predicted event: is the storm still there with each consecutive model run?
As far as that goes: so far so good.
Now that we're inside of a week away, we also start to look at storm track. Here's where the models are, to use technical terms, all over the damn place.
For now, we'll take the GFS as a fairly reliable long-range model. Here's what is shows for Saturday evening:
A developing storm moving from the northern Gulf of Mexico and across the southeastern US, emerging just off the coast of South Carolina. To the northwest, a solid push of cold air that shows the freezing line to our south. According to the GFS, still some 120 hours away from this forecast time, whatever is falling Saturday night will be snow.
6 hours later the GFS has the storm intensifying rapidly. Snow really gets going especially in central VA which looks to be the "bullseye" for this run of the GFS. Up here in the DC area we have less snow as the storm slides along the North Carolina coast - more eastward than northward.
6 hours more and the low is moving off the coast throwing some precipitation back into our area but the motion of the storm suggests lighter amounts than anything we saw on the 19th of December last year:
The east-northeast motion combined with the absence of any "bombing" or rapid intensification indicates that the GFS is predicting that the jet streams will not phase. I think I wrote about phasing last year but if not: here's a quick description:
Flow patterns in the upper atmosphere occur in sequences of waves. These waves can and do interact with one another sometimes in a way that is destructive, sometimes in a way that amplifies the original wave. When waves combine to form a larger wave, it is called "phasing". In meteorology, especially in the winter, we look for phasing of the northern and southern branches of the jet stream which tends to amplify existing storms. It's a tricky thing with respect to timing and it doesn't happen all that often, relatively speaking.
If the phase doesn't happen, you get a weak storm like the one predicted for Thursday. If it happens too early, a storm like the one above goes too close to the coast and we end up with rain. If it happens too late (like in the model runs above), it goes out to sea before it can get really wound up.
All that said: it's a good model run for us (remember "snow = good"). This far out, details like storm track and phasing are several runs away from being decided. We have the storm on the models still - that's one victory. The storm moves up the coast - that's another.
Now. What's bad about the Thursday storm?
Because of a complicated series of atmospheric interactions that have to do with something called the "Polar Vortex", the probability of an out-to sea solution for the December 19th storm (like the models show above) probably increases the further north the Thursday storm tracks. So if you want the potential for a big storm on Sunday, you want that Thursday storm to be as far to our south as possible.
Lots still to be worked out. We'll be closer to a solution for the Thursday storm tomorrow. I'll be keeping an eye on things!
Wouldn't a monster storm on the anniversary of a monster storm be cool?
Just to jog your memory: 12/19/2009:
Friday, December 10, 2010
Weekend Storm Likely to be Rain (Primarily) Then VERY Cold Next Week
After some model wavering, it seems a pretty good bet that except for relatively short periods at the beginning and the end, the Saturday-Sunday storm will be primarily rain.
Most of the models show a pretty strong low pressure system moving north from the south central US on Saturday. Clouds will increase rapidly Saturday afternoon but I think the precipitation holds off until evening. The temperatures will be warmer than we've seen, mid-40's, but they'll drop off in the evening - especially when the precipitation starts.
Because of that, I wouldn't be surprised if the precipitation started off as a bit of sleet which would change to rain as the night progressed and the low pressure center to our west draws warm air from the south over us as it moves northward. Temperatures overnight Saturday should stay in the mid-30's because of this storm track which means the second-worst thing winter can bring: cold rain.
Rain continues most of the day Sunday and because of the strength of the cold air behind the storm, I think it probably gets in here quickly enough to change our precipitation over to snow before it ends Sunday evening/night. It will be close but this is some serious cold air and it's going to be rushing in - so any precipitation that's left will change over.
Temperatures on Monday and Tuesday will spend the majority of the day in the 20's with Tuesday being the coldest day of the week, I think. Maybe not out of the 20's at all. By Wednesday we "warm" into the mid-30's.
Next chance for precipitation looks like late next week. Too soon to tell how much cold air is still around and how much moisture that system has.
Updates as I see 'em but for now, count on more rain than anything else from Saturday night through Sunday.
Most of the models show a pretty strong low pressure system moving north from the south central US on Saturday. Clouds will increase rapidly Saturday afternoon but I think the precipitation holds off until evening. The temperatures will be warmer than we've seen, mid-40's, but they'll drop off in the evening - especially when the precipitation starts.
Because of that, I wouldn't be surprised if the precipitation started off as a bit of sleet which would change to rain as the night progressed and the low pressure center to our west draws warm air from the south over us as it moves northward. Temperatures overnight Saturday should stay in the mid-30's because of this storm track which means the second-worst thing winter can bring: cold rain.
Rain continues most of the day Sunday and because of the strength of the cold air behind the storm, I think it probably gets in here quickly enough to change our precipitation over to snow before it ends Sunday evening/night. It will be close but this is some serious cold air and it's going to be rushing in - so any precipitation that's left will change over.
Temperatures on Monday and Tuesday will spend the majority of the day in the 20's with Tuesday being the coldest day of the week, I think. Maybe not out of the 20's at all. By Wednesday we "warm" into the mid-30's.
Next chance for precipitation looks like late next week. Too soon to tell how much cold air is still around and how much moisture that system has.
Updates as I see 'em but for now, count on more rain than anything else from Saturday night through Sunday.
Monday, December 6, 2010
Outlook for Winter 2010 and a Storm (probably rain) on Sunday
Welcome back to the blog! As crazy as it is during the Holidays, I'll try to start posting regularly again. Hopefully there will be something to post about!
Everybody wants to know the same thing: will this be a winter like 2009? Will it even snow? How much?
Those things are difficult to predict and long-range forecasting is really only slightly more accurate than throwing darts at index cards on the wall with snowfall totals and dates. But it is possible to make an educated guess and large-scale climate conditions make it a bit easier.
Almost everyone has heard of El Niño and has heard it blamed for everything from huge storms in the western United States to a string of car thefts in Kansas City. In reality, El Niño is defined as "warmer-than-normal sea surface temperatures in the western Pacific Ocean". This warm water causes a succession of atmospheric responses that affect the weather in the United States.
The exact opposite condition is known as La Niña. The cooler-than-normal sea surface temperatures in the western Pacific Ocean induce a low pressure zone there. Since everything in the atmosphere is interconnected, this unusually low pressure has side effects throughout North and South America.
Currently, there is a moderate La Niña occurring and it's forecast to strengthen a bit (which is an odd turn of phrase since it actually means that sea surface temperatures in the western Pacific will get a little colder) during the months of November-January.
So, what is the typical winter experience for the United States during a La Niña of the strength we're expecting?
In terms of precipitation:
First precipitation:
Green regions labeled "A" indicate above-average precipitation. The gradient lines are percentage probability of occurrence. Orange regions labeled "B" indicate below-average precipitation.
Again, keeping in mind this is an average, this forecast map suggests that more storms will pass to our west than to our east. If you like snow, you want coastal storms that pass offshore or move up the coast. Storms that pass to our west do two things in the winter: they bring rain or they bring ice. Both of those suck.
Now temperature:
Again, look at the trends. Washington isn't quite in the "above normal" 30% area but it's close. Generally, the forecast is for normal to slightly-above normal temperatures. That, in combination with the first map, suggests a winter where snow events are less frequent than last year and where many storms are deflected to the west of us.
This all adds up to the likelihood of a winter with normal to slightly below-normal snowfall in the DC area and temperatures that are normal to just above-normal.
About Sunday. A fairly large storm is expected to develop but most models are following the large-scale trends and suggest that the storm will pass to our west. This will probably lead to a very rainy and cold Sunday. It bears watching. If the track is further to the east, it could get much more interesting.
If this were 2009, I'd be excited about the possibility but, for now, it seems like this one is going to be wet and not frozen.
We'll see how the models trend as the week goes on.
Everybody wants to know the same thing: will this be a winter like 2009? Will it even snow? How much?
Those things are difficult to predict and long-range forecasting is really only slightly more accurate than throwing darts at index cards on the wall with snowfall totals and dates. But it is possible to make an educated guess and large-scale climate conditions make it a bit easier.
Almost everyone has heard of El Niño and has heard it blamed for everything from huge storms in the western United States to a string of car thefts in Kansas City. In reality, El Niño is defined as "warmer-than-normal sea surface temperatures in the western Pacific Ocean". This warm water causes a succession of atmospheric responses that affect the weather in the United States.
The exact opposite condition is known as La Niña. The cooler-than-normal sea surface temperatures in the western Pacific Ocean induce a low pressure zone there. Since everything in the atmosphere is interconnected, this unusually low pressure has side effects throughout North and South America.
Currently, there is a moderate La Niña occurring and it's forecast to strengthen a bit (which is an odd turn of phrase since it actually means that sea surface temperatures in the western Pacific will get a little colder) during the months of November-January.
So, what is the typical winter experience for the United States during a La Niña of the strength we're expecting?
In terms of precipitation:
- Above-average precipitation in the Pacific Northwest and Northern Rockies
- Above-average precipitation in the Ohio Valley
- Below-average precipitation in the south-central and southeastern U.S.
- "Normal" everywhere else
- Below-average temperatures in the coastal sections of the western states
- Above-average temperatures in the central and southern states
- "Normal" everywhere else
First precipitation:
Green regions labeled "A" indicate above-average precipitation. The gradient lines are percentage probability of occurrence. Orange regions labeled "B" indicate below-average precipitation.
Again, keeping in mind this is an average, this forecast map suggests that more storms will pass to our west than to our east. If you like snow, you want coastal storms that pass offshore or move up the coast. Storms that pass to our west do two things in the winter: they bring rain or they bring ice. Both of those suck.
Now temperature:
Again, look at the trends. Washington isn't quite in the "above normal" 30% area but it's close. Generally, the forecast is for normal to slightly-above normal temperatures. That, in combination with the first map, suggests a winter where snow events are less frequent than last year and where many storms are deflected to the west of us.
This all adds up to the likelihood of a winter with normal to slightly below-normal snowfall in the DC area and temperatures that are normal to just above-normal.
About Sunday. A fairly large storm is expected to develop but most models are following the large-scale trends and suggest that the storm will pass to our west. This will probably lead to a very rainy and cold Sunday. It bears watching. If the track is further to the east, it could get much more interesting.
If this were 2009, I'd be excited about the possibility but, for now, it seems like this one is going to be wet and not frozen.
We'll see how the models trend as the week goes on.
Subscribe to:
Posts (Atom)