"Super-storm" "Frankenstorm" Sandy was the result of a strong hurricane transitioning into an extra-tropical (or "post-tropical") system just as it was being captured by a trough in the upper levels of the atmosphere. The overall mechanics of the storm was incredibly well-predicted and well-modeled, as we were talking about the possibilities (we actually first mentioned it on Sunday, October 21st, but waited for a few more ticks on models before bringing it out into the open on Monday and Tuesday. The Euro (ECMWF) proved true to its moniker as the most accurate forecast model in the medium range, being the first to drag the storm inland instead of out to sea like 99% of all tropical systems that have ever approached that area. When it came closer to 'game-time', but still with days remaining to prepare, the GFS, NAM, and CMC all converged on the same solution-- and the ball was set to roll. The meteorology required to pull it all off is truly unique, and it is a testament to the power of modern super-computer processing to crank out these models based on billions of calculations and variable simulations.
The image below shows the climatological history of every tropical system that came within 200-miles of Sandy's eye, and yet still made landfall on the US coast at at least Tropical Storm strength.
Notice how there aren't many of them (since 1850 no less!), and ALL of them are located far to the west (with the exception of a single relatively weak tropical storm that wobbled around in the Atlantic before hitting New England.
Simply put: It just doesn't happen, and you are all party to what might have been the only time you'll witness such a thing in your lifetime.
This post will concentrate on the local impacts of this storm, as the entire story of Sandy seems limitless, and the billions of dollars in damages makes it a rival with the top US natural disasters of all time. Let's take another look at the timeline and how it all played out:
Here was our forecast timeline that we put out before the storm. The actual landfall point was in Cape May county, New Jersey (my family has become very fond of North Wildwood, NJ, so we know the area well). Landfall itself occurred during our live blog chat (that was very well attended Thank You All). This location would be predicted by a blend of the NAM and the Euro, while the GFS was farther north at the T-minus 24-hours mark. A few blog participants wondered what possible effects the slightly earlier arrival of Sandy may have on the forecast itself-- and it seems clear now that even very small changes made a difference.
Here was the predicted snowfall accumulation map put up on early Sunday morning, almost 48-hours before the event:
This became largely the basis for the final snowfall forecast we went with on WSAZ-TV in the hours before the storm.
As a matter of comparison, here are the other projections, both with the models, and with what the National Weather Service came out with less than 24 hours before the flakes fell in the lowlands Tuesday morning (click on each image for a larger one):
|NAM - Modeled Snowfall Projection - "Frankenstorm"||GFS - Modeled Snowfall Projection - "Frankenstorm"||NWS - Modeled Snowfall Projection - "Frankenstorm"|
As a forecaster, seeing these kinds of contrasts and these crazy numbers from a crazy storm really makes it stressful to get something accurate and useful to the viewers. I could have just went with 1-8" or something like that, but that's not as useful forecast. That being said, there were tremendous differences within small increments of elevation (as expected). In downtown Huntington and Charleston, not a lot of snow actually fell, and what did fall melted during the same day. However, in the hills and ridgetops within city limits had a drastically different story.
Regardless, seeing an approaching weather pattern that is going to hinge on such lines, it's important to have a good handle on climatology. Records in the tri-state area go back more than 100 years, and in that time, the most amount of snow that ever fell in the entire month of October in Charleston, WV was 2.8". Essentially, I ended up posturing myself at all-time record to twice the all-time record snowfall amounts. That should be an incredibly rare forecast-- and indeed this was an incredibly rare storm.
A huge amount of you tuned in to our expanded storm tracking blog page (thank you again), and we hope you found it helpful. As you all know, there are still quite a bit of after-effects, including power outages and gradually clearing high elevation roadways, so you can still monitor the scene up there.
The best ball-park figure for snowfall totals would probably be from the National Operational Hydrologic Remote Sensing Center (yes, I am out of breath on that one). This uses satellite technology combined with crazy math algorithms to get at snowfall amounts from above. Here's what it has ascertained for this event in the tri-state:
|NOHRSC - Snow Depth - October 31st, 2012|
Notice the return algorithm only divides up the snowfall categories into near exponential scale, which doesn't do a whole lot of justice above the 4" mark. However, you can clearly see the swath modeled by the GFS (though it had it too deep, was going to miss the landfall location, and didn't lock in the correct northern mountain snows we were going to see).
I'll list some snowfall numbers we had in to the station below (averaged amounts from multiple viewer reports from the same communities). The "big board" numbers are still coming in-- having no power and no communications can be a bother to get reports out. Usually the less you hear, the worse it is, when it comes to a large-scale natural disaster. These totals may well be updated for posterity sake:
|The Big Board|
|New Hope, WV||36"|
* Also a new all-time monthly snowfall record for October (shattering the old 15.5" record from 1918).
|Central West Virginia
||Southern West Virginia
|Charleston (airport)||10"*||McArthur||2-4"||Belfry||2-4"||Delbarton (Mingo Central H.S.)||14"|
|Charleston (hills)||6-8"||Gallipolis||2-4"||Warfield||1-2"||Horsepen Mountain
|East Pea Ridge||2-4"||Middleport||0.5"|
* The 10.1" measured at Yeager airport during this storm is the new all-time snow record for the month of October, incredibly shattering the previous record of just 2.8". For records that go back more than 100 years, that is a large amount of standard deviations from average.
All the places that got less than 4" largely saw their snow melted by the end of the day. This was expected, as we were talking about that violent hair's breadth during that critical Tuesday morning commute where the difference was going to be between heavy cold rain and a 1-2"/hour snowfall accumulation. It seems the battle line was between about 900 and 1200 feet, where snowfall accumulations at the higher end were double and triple that of the lower range.
If you have your own snowfall reports you wish to add to the mix, please feel free to do so in the comments section. Your pictures are welcome as well.
Speaking of pictures, WSAZ had hundreds of pictures sent in with your viewer experience during this historic storm, and we are (again) very thankful. Feel free to view them below-- enjoy! :-)
Click on the following link and see them all on their own page!
One picture in particular caught my eye, and I wanted to share this one with you:
This was sent in by Jo Huffman, as they crossed the 30" mark in snowfall over in Mount Nebo this morning. A few things come to mind, which can represent some final thoughts:
1. This snow is not the fluffy stuff you can clean with a broom. On that picnic table alone, there are more than 500-pounds of snow. As of this post, we've heard of at least 8 partial roof collapses out in Nicholas County. Flat roofs are the most vulnerable obviously. Any rain that falls into this snow is just going to make it heavier. Also, make sure you take frequent breaks shoveling this snow. It is an amazing workout, and it the exertion can sneak up on you in ways it rarely does when you're at the gym. We've had reports of deaths in West Virginia from folks succumbing to cardiac arrest while shoveling.
2. Water is blue. That should come as no surprise when at the ocean, but it can be when photographing snow. Your camera often looks for "white" wherever it's pointed, and if you don't manually "white-balance" your camera you're going to end up with a blue image. In this case though, it's even more difficult. There was so much snow with such a high water content, that the snow literally was blue.
3. This snow will melt, and where it has been melting it has been doing so quickly. In the coming days, morning temperatures will be sinking lower as skies clear out (mainly Friday/Saturday mornings). Radiational cooling can/will be enhanced by any remaining snowpack on the ground. The meltwater during the day can ice up at night, causing otherwise unforseen slick spots on driveways, walkways, parking lots, and even the streets themselves. Be careful.