Sunday, January 25, 2015

We Need To Talk

Back in November I issued my winter forecast. You can refer to that here (

This is a quote directly from my winter forecast blog entry:

 "All indices I've looked at suggests increased precipitation amounts for the eastern US. More precipitation could mean more intense storm systems than last year and increases the likelihood for a block bluster snow event measured in feet rather than inches. Also could mean heavier rain events as well on the warm side of a low pressure system. Unfortunately for the south west I don't see much in the way of relief for the record breaking drought.  "

I spent a month, few couple hours each day to produce my winter forecast, and up until now I thought I did something wrong. I've gone over, over, and over again to find if I made a mistake somewhere. I even did some quick research on the strong -QBO phase, which can be found here:

and the second entry with results here: (

My conclusion after looking into this because some meteorologists were bringing this up and I wasn't going to take their word for it:

"My conclusion, the strong -QBO has minor influence on our winters and is only a percentages of the larger picture in a winter forecast. It alone cannot be used to determine how winter will behave. I made this decision after seeing very little in the way of linkage between this winter and the other winters with similar QBO indices. A strong -QBO can produce both snowy/snow less winters as seen in the data. I would love to use the majority of the 1970's winters as evidence but there are a few multi-decadal oscillations that are in totally opposite regimes today than 30+ years ago. Going off the data from this last decade it looks like we still have a strong possibility for a snowy winter though not at cold as last, which is still in line with my winter forecast. "


I've spent years learning long range forecasting and trying to improve with every mistake I've made. However, this winter forecast, is not going to bust. The product I put out was 100% from me, and I didn't let any other winter forecast or meteorologists get into my head. I ran the data, did the math, and produced it from scratch. I even did audits in December and still came up with the same results, despite the "canceled winter" worries from the Weather Weenie community.

Up until this point my forecast hasn't been to bad, except for the lack of snow. I did note that we could see a lot of precipitation this winter and that we could end up on the warm of a lot of systems. That seemed to be what happened most of the time. Things are changing...

With 31.8" so far last year and 22.1" this year we are about 10" behind last year, but in a week from today we could surpass that number from last, with multiple storm chances and ample amounts of cold air.

First system: Monday - Tuesday

Okay, this one is a big deal. The Euro has been consistent with it's track and total snowfall while the other models are all over the place and messy. Yes, I like the Euro solution because I consider it the most reliable model in the world.

And yes, that's two to three feet of snow over the area. To make matters worse temperatures will be in the low 20's so the liquid to snow ratio will be more like 15:1 instead of the 10:1. These numbers on the Euro (if the liquid equivalent is correct) are off by a third cause they are based on 10:1 ratios. 

I do need to point out the other models are still showing a major snowstorm but along the lines of 10"-15". 




These models are showing about half the snowfall as the EURO. All I can say is that there is a lot of data to be processed here and perhaps the EURO is the only one with enough computer power, or maybe it's wrong. Right now I'm not sure. Either way a major snow storm is on the way. I'll post my snow map late tonight or tomorrow morning. 

On top of this it looks like another storm is possible by the end of the week. 

Beyond this system it looks cold with more storms on the horizon. Let's take it one system at a time though. Check back for more later. 

No comments: