Calling All Monday Morning Arm Chair QuarterbacksDecember 31st, 2012 at 3:30 pm by Jonathan Conder under Weather
I questioned on whether or not to talk about this topic, or whether if I should have to defend my forecast. First, if I miss or “bust” the forecast I am going to be the first one to admit it and I have always held myself accountable, giving an apology and explanation to the WANE viewers. But, I have a really hard time sitting on the sidelines, letting people say “you got it wrong,” when in all reality, I feel we nailed the Friday Night snow forecast 24 hours out.
What I am talking about here is the snow forecast on Friday for early Saturday morning. Here is the link to my initial forecast Friday afternoon, which I also posted on the the weather blog. That evening, I even updated the forecast, bringing the band of +3″ slightly more to the west. Here is the exact image from the 2nd blog update that evening:
Now, compare this to the official amounts from the National Weather Service and area weather spotters:
With the exception of southeastern Blackford county seeing 6″ of snow, I say that’s a pretty precise and accurate forecast. (Only off by an inch on the high range of the forecast.)
So, why are people so “angry” on social media? I am not going to call out any specific people here, but feel free to visit our WANE FB feed from Friday night and Saturday Morning. Are people angry because they hate snow? Are they angry because of the “looming fiscal cliff?” Or is this what social media has become, a place to be negative and vent? That being said, I do want to thank the WANE FB fans who did step in and support our brand and our forecast.
Again, I hope this post didn’t waste your time and honestly, I am not trying to stroke my ego, I just feel people have to realize that we can get it right and forecasting has come a long way.
This is from a good friend of mine and one heck of a forecaster back in Oklahoma and his reference to the Christmas Forecast in OKC:
For those upset about the weather forecast, let me tell you a little secret…
100 years ago the best forecast was what is currently happening down the road in the next town over a telegraph or radio.
50 years ago the best forecast was a few hours.
1980s gave us the 2 day forecast with considerable error using the best computers.
1990s gave us the 5 day forecast with considerable error as super computers really kicked in.
2000s gave us the 7 day with error, but significantly improved the 2 and 5 day as computer processing speed went through the roof.
2010s we’re stretching to 15 day trends, with 7 days reasonably close, 5 days darn good, and 2 days almost perfect in general terms.
What we have not and cannot do is estimate the exact amount of precipitation over your house and what type it will be. The best we can do is estimate over a broad region. We move from science to art/craft when hyper-local amounts come in. You put it on a grid and average it out and spread it around. A statistical means applied to an unscientific method. The main reason for this limitation is money. There’s not enough of it to build the vast array of measuring stations that are needed to input that data into the models. The model will take any error fed in and magnify it. In addition to the numerous locations, timing is key. We run the models 4 times a day but only 2 of those 6 am/pm have most of the data available to input. We would need to launch soundings 4x a day and put them in a very dense network rather than 1 or 2 a state. Finally each model applies a slightly different combination of physics, because we don’t know everything we need to know about the atmosphere to pick the perfect mathematical formula.
Just keep all of this in mind for the next storm…we’ve come a long way and produced quite a spoiled generation as a result.
Alright, I digress. Thanks for your time and have a Happy New Year!