Wednesday's Winter Wallop

in

We are 48 hours away from yet another wintry headache across the Valley. From a forecasting standpoint, this one is somewhat "easier" than the system that shifted south on us 8 days ago. The storm track with this one is not exactly "locked in" yet but we have enough of a consensus among the various computer models that we can start giving you more detail than we could a few days ago.Precipitation will arrive before sunrise Wednesday morning in the form of RAIN, or at least a mix of rain and wet snow. Roads will be wet pretty much everywhere until 9:00 on average (perhaps turning slick a bit earlier than that in northern Trumbull, and somewhat later from Route 30 south). Notice the projected temperatures locally at 8:00 Wednesday morning: gfs_t2m_b_ohio_21This shows readings still near freezing, and I could see where it is even a few degrees higher than this at 8:00. Especially from Route 30 south.Then, a changeover to heavy, wet snow will occur from north to south. Travel conditions will likely deteriorate pretty quickly. It is March, the snow is falling during the day, the sun angle is much higher than a couple of months ago, and it will be very MILD right before this storm. It takes a HARD thump of snow at this time of the year under these conditions to get the roads to get slick quickly. But I do think that will happen.The vertical velocity field early in the afternoon shows strong upward motion in the atmosphere around the region. Faster rising air=heavier precipitation rates: gfsvvSnow will taper off very late in the afternoon and early in the evening. The afternoon/evening "rush" will be a mess.How much snow?? A few forecasting challenges here.1) Final snow totals will depend a lot on how quickly the changeover to snow occurs at the start of the storm Wednesday morning. The more moisture "wasted" on rain or a mix, the less snow accumulation. The changeover will happen last and the atmosphere will cool slowest from roughly Route 30 southward. So I think snow totals will be appreciably less in places such as East Liverpool and East Palestine compared to our far northern communities.2) The snow-to-liquid ratios will be changing. At first, the ratios will be low, perhaps 7:1 (typical is 10:1). This means the snow will be wet and will not accumulate as efficiently AT FIRST. The ratios will quickly get higher though, perhaps 12:1 or so by the end of the afternoon. So it will take less liquid to produce more snow toward the end of the storm.3) The storm will be strengthening as it pulls east of the Valley, leading to the possibility of some very hefty snow totals in northeastern Trumbull and northern Mercer counties. Especially in Mercer, heavy snow may persist well into the evening.Here's the SREF model for around the airport in southeastern Trumbull County: Screen Shot 2014-03-10 at 8Out of the 21 "members" of this model, each represented by a colored line, there is a range from about 0" to 15". Throwing out the extremes and focusing on where most of the lines are clustered, a reasonable range there is 5-8". The mean, the dark black line, is about 6" or so.Other models, such as the GFS, European and Canadian generally jive with this idea (there are differences, yes, but they are not HUGE at this stage of the game).We will put our "official" snow forecast out this afternoon. But I suspect it will look something like this: 2-4" southeastern Columbiana, 4-8" most everywhere else in viewing area, with perhaps an area of 8"+ in northeast Trumbull and northern Mercer. Stay tuned!!!Thanks for reading!Eric

BLOG: Big Storm Next Week? Or Not?

in

Happy Friday!You may or may not have heard rumors and/or have seen images being shared on Facebook about a potential HUGE storm next week. Some have blaring headlines that speculate that if there were to be a storm, it could be comparable to the Superstorm of 1993.I am not going to dive too deep into the meteorology yet today, but will show you a few things demonstrating why the uncertainty at this stage of the game is quite high.The pattern IS ripe for a storm of some variety. But, will everything come together for a BIG storm? Or will the pieces not line up and all we have is "small potatoes". That's the big question. IF there were to be a sizable storm of some kind here, it would be in the Wednesday-Wednesday night time frame. Here's what the latest models say:Canadian model says "meh": canwedIt has a wave of low pressure moving through the Great Lakes and then strengthening some off the coast of Maine. Minimal impacts for the Valley.The GFS model is a bit more interesting: gfswedIt has a storm tracking south of the Ohio River and by Wednesday evening (shown here) it is getting set to head off the coast. This track is one that can bring us a light to moderate snowfall. Perhaps on the order of a few/several inches.The rumors of a BIG storm are courtesy of the European model. If you are new to this, the European model is *generally* regarded as superior to most other models, for many reasons. BUT, it has it's fair share of "busts" too. Here's the European look for Wednesday evening: eurowedA much healthier looking system, and farther north than the GFS. In fact, this run has it far enough north that if it were to be exactly right we would have some concern for mixed precipitation in parts of the Valley (as opposed to all snow).So, pretty big differences between these 3 models right? A Meteorologist's Instant Headache.Why is the European model idea more than just a little credible? Two reasons. 1) It's the European, which again...in a general sense is superior to the other medium range models. 2) The "operational" run, shown here, has quite a bit of support from the European "ensembles".Wha?? Again if you are new to this blog you may be wondering what in the world I am talking about. Here's the quick and dirty answer: The "operational" run is the high-resolution version of a model that is widely available and seen and used by many.  Ensembles are lower-resolution versions of the the same model, run multiple times with the initial meteorological conditions changed somewhat each time. This is done to account for the fact that the data fed into any ONE run of a model is never going to be perfect. Changing that data a bit multiple times helps account for those inevitable errors. Sometimes, these small changes in the initial data result in BIG changes to the model "solution", or what it does with all the weather systems on the map. Sometimes, the changes are small. When the changes are small, the confidence in that solution, or forecast, is higher.ANYWAY (still with me?), there are over 50 members of the European ensemble model. The mean, or average of all of these is not tooooo far from the idea of the operational model. So, confidence in that operational solution is higher than average.For Youngstown, right now the operational model (blue) gives us 7 inches of snow. The ensemble mean (green) has 4 inches: Screen Shot 2014-03-07 at 8.53.18 AMKeep in mind that the model output will count ice/sleet as snow, so if the model is "seeing" mixed precipitation, the snow numbers will be inflated to a certain degree.So, bottom line 5 days out: Next week will start mild but there is potential for a Winter Storm Wednesday-Wednesday night. Right now it's nothing more than potential. I think odds favor accumulating snow in the Valley, but does that mean one inch or much more? No idea yet.As always, will keep you updated! Have a good weekend and thanks for reading. Eric  

BLOG: A Look At Where We've Been, Where We Are Going

in

With my attention focused on the weekend (non)storm, I did not get a chance to do a recap of Meteorological Winter over the weekend. Meteorological Winter is the 3 coldest months of the year (Dec-Feb). This is different than Astronomical Winter, which is based on the Earth's position in it's orbit around the Sun and runs from about Dec 21/22-March 21/22 each year.Of course, the headline of this winter is "IT'S BEEN COLD AND SNOWY!!". But let's take a deep dive into the numbers to see HOW cold and snowy it has been.First, an overview of how each day of the season went temperature-wise: wintertempsYou can see just how often we were below-average, especially in January and February.Here's the snowfall graph for Dec-Feb and how this year compares to average and the snowiest winter ever (just 3 years ago): wintersnowSpeaking of snow, it of course was one of the snowiest December-February periods on record here. The list:snowhist This of course only tells part of the story, as snow did occur in November and early this March. How we stack up when you take the entire "snow season" into account: snowoctnowNotice the top 6 are ALL recent years! WHY is that? I am not sure but it's interesting and merits some investigating!Back to temperatures. For Meteorological Winter, this will go down as the 10th coldest season on record:temphist December was actually a touch above average. If you just look at January-February...janfebWe have had many below-zero low temperatures. You can add 1 to our total this year as we were of course below zero this morning: belowzeroAnd high temperatures have been below freezing a LOT: highsAdd one more to this year, as it will only be in the 20s this afternoon.  Put another way...since our first sub-freezing high on November 24, 55% of days have featured a high below freezing in Youngstown.Is there relief in sight? We will bust into an early Spring?No.While it is unlikely that the kind of harsh, record-breaking cold that we have had over the last 24 hours will return...temperatures should generally be below average for the foreseeable future. There will be the occasional mild day, but they will be few and far between. The latest GFS Ensemble model temperature outlook for days 8-16:816 After that, the CFS (Climate Forecast System) outlook isn't great either. March 18-28:1929 March 23-April 2: 243 Wish I had better news! But thanks for reading anyway.Eric

BLOG: What happened to the BIG storm??

in

On Wednesday afternoon, I posted this on social media:"-Putting numbers in a forecast more than a few days out is a recipe for flip-flopping and changing forecasts with each new run of the models. 4-8" becomes 6-10" becomes 3-6" becomes 6-10", etc etc. We would rather you hear as few numbers as possible so that our message is clear. The closer we are to an event, the less likely it is that the forecast will have to change much. "In some ways, I didn't follow my own credo. This storm was coming on a weekend and by Thursday, I felt it was time to give people a firm idea of what to expect. We should have waited until Friday.Our initial forecast was for 5-10". We are getting 2-5". So, what happened? Why are we getting less snow?This storm was supposed to come in 2 "waves". The first one, the one we are getting this morning, followed by a second, heavier wave tonight and Monday morning. Here's this morning's radar:natrad The reason for the lower amounts is that the 2nd wave is not going to get pulled northeast like originally expected.A look at the features at 18,000 feet tells the story a bit better. 500Essentially, the "northern feature", a trough of low pressure moving through the Great Lakes, was supposed exhibit some "pull" on the southern feature...lifting it somewhat north. BUT, the northern feature ended up moving too fast to give the southern one a lift. The southern feature missed it's ride, basically. So, it is moving more due east. Getting the interactions of these things right was key to getting the forecast right. The southern feature did not get onto the mainland US until yesterday, so the models had a hard time figuring this out. And therefore, humans were fooled.Our forecast could have been worse! Thursday morning we were staring at models that looked like this: bigeuroYup, that's the European model, usually one of our better models, predicting a FOOT of snow in parts of the Valley.Some well-known national weather companies had us getting 6-12" and for a time and I saw a map that had the Valley in a "10"+" band for a while. So our 5-10" forecast was actually sort of conservative.The lesson for forecasters? In very, VERY complex situations like this one...it's best to wait until you are within 48 hours of an event before making a snow accumulation forecast. It is tough to abide by this, especially when it is coming on a weekend. And it's 2014, so social media "buzz" builds to a fever pitch, even a WEEK or more ahead of a possible storm. There is pressure to make a statement, to get your forecast "out there".There is a reason why we are vague and non-committal when viewers, family members, neighbors, pets, etc. start asking us "how much you thinking" about a storm that's more than a couple or few days away.  It's not because we ENJOY being coy. It's because weather forecasting is complicated. Things change. The atmosphere is so complex it boggles the mind. It's because these types of things can and do (and always will) happen.Just know this, I  (along with Jess and Mike) are on it. When it's obvious changes are needed, we start making them. Bad forecasts don't stay bad for long. Take a look back at my tweets, Facebook postings, etc. My train of thought is on there.Winter can end any time now!Eric

BLOG: Snow Storm Update

in

Good morning all,If you followed my social media posts and/or watched me on TV last evening, you noticed I was becoming concerned that the storm is not going to behave as originally expected. That trend continued overnight.So, what changed?After the first wave of snow Saturday night/Sunday morning, a second surge of moisture was supposed to ride up the front from southwest to northeast, bringing out heaviest snow Sunday night and Monday morning.  BUT, instead of coming northeast...this second surge will track DUE east and miss us to the south. gfsSo now expect the WORST of the storm to be Sunday morning and afternoon, with snow tapering off and ending by the end of the afternoon. Simulated radar at 9:00am: 9amBy 1:00pm, the snow should ALREADY be winding down north of I-80:1pmIt will snow HARD for a time and travel will be slow Sunday. That's the bad news. The good news? Monday morning should be just fine for school/work.Total accumulations across the region: 2-5 inches.More updates on Facebook and Twitter when I have some time today and tonight.Eric

BLOG: Heavy Snow Coming, But How Much?

in

Good morning all,It's a windy and cold Thursday and we have to watch for some beefy snow showers and squalls this morning. This will be followed by subzero temperatures tonight. Stay up on the latest short term weather by checking out the social media outlets (mine and Jess's).This post will focus on the Sunday-Monday snow threat. We plan on looking at the latest models before putting accumulations in our official forecast this afternoon, but this post will give you a peek "behind the curtain".First of all, it's important to note that the disturbance largely responsible for the snow threat is still over the Pacific Ocean. Computer models will be more trustworthy once the system is over land, since there will be more data available for them to analyze. Here's an animation showing the system crashing into California and then coming east. Click to animate:output_f74bhhThe colors represent the "vorticity" or spin in the atmosphere at about 18,000 feet. Spinning air parcels at that level aid in air rising and clouds/precipitation forming.Snow will get going late Saturday night and continue at varying intensity Sunday. There will be times Sunday that it is not snowing that hard at all. We are still too far out to pinpoint when snow is likely to be lighter/heavier during Sunday.We are pretty confident the worst of this storm will be Sunday NIGHT and early Monday. Here's a few computer model snapshots at 1am Monday morning:GFS: gfsEuropean: euroCanadian: canThe Canadian looks a bit different than the other two; snow is lighter at this time step on the Canadian. This is an outlier at the moment.One thing we look at when figuring out how hard precipitation will be falling is the "vertical velocity" at about 10,000 feet. It shows how quickly the air is rising. The faster it is rising, the heavier the precipitation is likely to be. Here's the GFS vertical velocity map at 1am Monday:vvlIt has the best upward motion south of Rt. 30, but still looks decent for most of the Valley.How Much Snow?? That is the question.Let's see what the current models say. First, the GFS:gfssnowA general 6-10" range for the WFMJ viewing area.The Canadian:  cansnowLooks like a range of 5-12", with the heaviest south.The European: eurosnowYikes! A solid 11-12" for the region. You are likely to see this map or something similar on social media outlets today.BUT! A word of caution. This is the "operational", or "deterministic" run of the European.  There are 51 "members" of the European Ensemble Model. What's an ensemble model? A lower resolution version of the "deterministic" model is run 51 times, with each "member" having slightly different initial conditions (pressure, temperature, humidity, etc.) This proves useful because we simply cannot model the atmosphere perfectly. Running a model 51 times, allowing for "wiggle room" in the initial conditions gives us more of a range of possibilities.Anyway, the mean, or average snow accumulation of those 51 members is significantly lower than the operational...which makes the 11-12" idea harder to believe. Have a look:euro ensThe mean is 5-6".So, what's the bottom line?? I think the low end of the snowfall range will be around 5". What's the top end? Not sure yet. Probably somewhere between 8-10". More to come this afternoon!Thanks for reading!Eric

BLOG: The Never Ending Winter

in

This morning was another reminder that sometimes a small amount of snow can be WORSE for road conditions than a few inches....especially if it comes at the wrong time of day. While some places got an inch or so, most locations "only" got a coating, but it came at rush hour and roads got slick in a hurry.The radar at 9:15am showed the last of the hefty snow bands in Lawrence County, with flurries elsewhere:915rad It sure was nice to see bare ground before this morning's snow. The snow cover map does not reflect today's snow and shows a nice zone of bare ground across the region:depth For the rest of the work week, we will be "nickled and dimed" by a few weak systems that can, as this morning's did, drop small accumulations of snow. Another one will cruise through Wednesday morning and another Thursday. Snowfall amounts with each will generally be under an inch, but watch for "sneaky" bands like we had this morning that can drop locally a bit more.The BIG story this week? Yup, the COLD. Our 7-day forecast:CaptureWe are forecasting the 2nd coldest stretch of weather for these dates since 1930 in Youngstown. Here's the list of coldest February 25-March 3 periods:coldest Plan on 3 straight mornings with Wind Chills below zero. Here's a computer Wind Chill forecast for daybreak Wednesday through Friday mornings:Wednesday:wcwedsThursday:wcthursFriday:wcfridayThe longer range? It looks ugly through most of the first half of March. The 8-16 day outlook off the Climate Forecast System model shows temperatures much below average for March 4-12:816 We have some forecasting challenges ahead for the weekend and early next week. There is likely to be a storm that impacts the Valley sometime in the Sunday-Monday time frame. How big of a storm, what type of precipitation it will bring, if snow....how much...all these things are unclear right now.If you are seeing web sites/Facebook pages that have scary maps with dramatic headlines and/or are providing SPECIFIC snowfall forecasts a week or more out...STAY AWAY! Stay far, far away. Thanks for reading!Eric