The posts in this forum are NOT official forecast and should not be used as such. They are just the opinion of the poster and may or may not be backed by sound meteorological data. They are NOT endorsed by any professional institution or 
I write an email wx update for WREL (~510 subscribers, a few here
). Generally this is read by folks just interested in learning about pending and svr wx; it's not a weenie-forum as it's pretty basic. (No, I'm not hinting for more readers...it would bore most here)
This bust was so widespread it has bothered me all day; how could so many be so wrong so close to an event? Here are a couple thoughts...I'd welcome yours as I think this event has shown a growing, serious problem. I haven't included the weather content, just the commentary.
------------------------------------------------------------------------------
DISCUSSION - PARTLY PERSONAL: Typically, these segments are separate however a departure is in order. Obviously, most "official" forecasts and the last Update were atrocious with the latest Mid-Atlantic event. Long time readers know there's usually touch of humor in the Partly Personal section; however today I'm in no mood for fun and games. If you know Bruce Springsteen music, you'll understand I feel I've been dancing in the dark. This discussion is a commentary about why the recent storm was a near total bust; hoping the public gleans some understanding of both the weather and forecast techniques.
Here's a snip from last Sunday's Update: "Early Thursday, snow from the Northern Stream migrates from the Ohio Valley into the Northeast while rain moves off the Atlantic coast south of Virginia...both look like a miss for Virginia. So you can see while far from quiet, there is no big event anticipated this week."
Here's a snip from the last Update: "Snow should expand over Virginia by dawn Thursday...generally south of VA I-64 from Staunton to Norfolk. Totals under 3 inches from I-64 to southern VA. Extreme southern VA counties may exceed 3 to 4 inches of snow."
How can a forecast written over 72 hours before a storm be more accurate than one written 12 hours ahead of the same storm? The answer is computer error and human error. Meteorologists and weather enthusiasts utilize computer guidance (called computer models). Essentially, these models project what the weather may be up to 15 days in advance. As data is presented, it's up to humans to analyze the models and craft a forecast. So far, so good.
Last weekend the models were in general agreement this storm would develop, travel off the southeast coast providing mostly rain south of the VA/NC border. I'll devote at least one hour looking at weather maps and charts prior to writing an Update in an attempt to determine if the models accurately see real-world weather. On Sunday this all made sense and I was comfortable with the ideas presented. Remember, that's over three days before the storm.
On Monday, the same models began to show the storm moving into Virginia, temps drop and heavy precipitation is transported from the Gulf. I disagreed and WREL's on-air forecasts stuck with the idea this is a "miss" for the local area. Tuesday, computer guidance continued showing an expanding snow event over NC and southern Virginia. I didn't agree but grew concerned the models saw something that people were not. "Official" forecasts began to change and after several conversations with other forecasters, confidence in my original ideas on Sunday dropped. Wednesday, the models stuck with snow and cold and "official" forecasts changed from light rain to snow. But real-world weather (and similar previous storms) made me think this was not right and the snow would not materialize. By Wednesday afternoon, the storm was less than 18 hours away from the Mid-Atlantic and every forecast had gone to colder and snow. NWS Warnings, Watches and Advisories were hoisted and a snow storm was the talk of the region. Before sending the last Update, maps and charts were again reviewed and while not sold on this outcome, I finally went along. All the TV people were ramped up for snow, The Weather Channel had "SNOW/RAIN" displayed for much of the region...in short, the train had left the station. So I ran and jumped aboard. I was wrong.
The problem here is Blind Faith. When I was a kid, if a cute girl batted her eyes at me, I'd take the bait and follow along (Gentleman: tell the truth, you did the same thing!). Most times, there was nothing there and I'd think I was led on. It was taken on Blind Faith the young lady was actually interested in me. Real-world experience would ultimately prove the contrary. After a couple times through that, one learns what's going on. Apparently, I have forgotten this lesson. Far too many forecasters take on Blind Faith what computer models tell us. If the models flirt, we get hooked... doesn't matter what real-world weather is telling us. Like the cute girl, we confuse flirting with fact. I have resisted this desire to mimic each computer model but this week I took the bait and that stings.
Most of the public will never know this; they'll simply conclude it was another terrible forecast. But the problem is still there because too many who do this will continue to spew out whatever the latest computer model thinks will happen. Did you know there are some TV stations and newspapers displaying upcoming weather based solely on computer models? I am worried about the weather community worshiping these models to the point there is no human intervention. Late Wednesday night we could see this was going bad very fast... but it was too late. Every "official" forecast, most meteorologists and I already pulled the trigger and you awoke to weather very unlike what you were told the night before.
Good things come from bad mistakes if there's an effort to learn. Many of those I've talked with today are genuinely stunned by how many of us blew it. If each forecaster now dining on humble pie takes time to see what is happening, the public will be better served. Mistakes are forgiven; a forecast based solely on computers is laziness and that is unforgivable.
This opine is shared because, for whatever reason, you have a more than passing interest in weather. It's presumed you're interested in knowing why a forecast can be so far flung from reality. Mistakes will again be made but I hope this sheds light on why so many were so wrong. Do I feel like I was sucker punched today? Yes. When I was a dopey kid, I took flirtatious bait until learning that's all it was. The computer models winked and away I went. This week reminded me of youthful error. Lesson learned the hard way this week but as The Who sang: "Don't Get Fooled Again. No No."
Thanks for listening.
That's the way it looks from here.
