Weather Models: Can I Have Some Syrup With That Waffle?Last Edited:
Yes! It’s finally the weekend. This means we get to sit back, relax and enjoy ourselves with our friends and family. Ahhh. It’s a very important holiday weekend. For those of you who celebrate Christmas, Merry Christmas to you. If you don’t. I still hope you have a fantastic weekend. If you’re reading this and work prevents you from being with your family, thank you for your service.
Alright – on to the topic at hand. Models, modelology and weather apps. All things I despise. To the untrained eye it looks like things bounce around way too often and those of us who are tasked with weather forecasting don’t know what we’re talking about. It can’t be farther from the truth. Unfortunately in the world of technology everything is at your fingertips and you can get any information within a split second – including the weather forecast.
What could possibly be wrong with that? Most apps don’t have any human input to decipher what’s right and wrong. They are updated on average every 6 hours to reflect what the latest model data shows. Most are based off the GFS (run 4x per day). This sometimes changes the forecast multiple times a day. For example: Yesterday each run of the GFS changed the snow totals for the end of this coming week. We started the day off with 3-6” then 6-12” then 1-3” and finished with 10-15”. Ridiculous. People who don’t know what’s actually going on behind the scenes then think some weather forecaster has no clue what’s going on. Not the case. They’re updated with bad information.
Now that I called out weather apps, let me move on to a different subject. General use of models. I’m not the biggest fan. We’ve seen their epic fails many times in the past, as recently as the bad forecast for November and December on the models specifically calibrated for long range climate output. Most of that information was junk. They don’t take every variable into consideration. That’s the issue. The way I forecast weather is different. I look at what has happened in the PAST and determine what could happen in the future based on that data. It is MUCH more realistic to look at a set of conditions under a specific pattern that has HAPPENED and forecast weather for a similar pattern occurring today. Now, there ARE tweaks that need to be made but overall it’s a better system.
Some forecasters (and by no fault of their own) are comfortable JUST looking at model data. Like the weather apps, these forecasts flip flop throughout the day – of WAFFLE – hence the title of this blog. This is much more common in forecasters who have little experience or purely do it for social media. The issue with this method is this: Maps are posted that get thousands of likes and shares MANY days in advance that most likely won’t come to fruition. Example:
1 Run shows an absolute historic storm:
1 run later there’s something completely different:
This toys with our emotions and chips at viewers’ faith in those of us in the meteorological community. I like waiting until just a few days before an event to post maps or discussions because let’s face it – even WITH all the advancements in technology, meteorology is STILL very fickle and in-exact. We need to focus on getting the front half of the 7 day forecast correct before going into too much detail on something well off in the future.
I don’t own the weather… and neither does anyone else. Everyone has the right to post whatever they want on facebook. I’m not trying to stifle anyone in anyway, but when these maps get posted it DOES make our jobs harder and I’m sure most in my field would agree.
In regards to storm chances next weekend? I’m onboard. Onboard for exactly what? Not sure yet. All I know is that the pattern SCREAMS snow in the Mid-Atlantic and that’s good enough for me. The one thing I AM 100% certain of? The cold air moves in and stays with us well through the first week of the new year. Honest to goodness COLD weather. Finally. Here’s a look at what the 5k foot temperatures look like.
Beautiful. Enjoy your weekend!