Various models - various runs - various data

Last winter, news outlets seemed to "discover" weather models when writing poorly educated articles for their various parent companies...

Since I'm getting excited about model runs, I thought I would do a dual post.

1.) - to show off some of the new runs

2.) - to talk about what "model runs" really tell us when talking about weather forecasting

---------


Various countries around the world (or groups of countries) have weather systems (massive computers) in place that run programs daily, after we plug all of the real/live weather recordings into the blanks.

Through years and years of research and understanding, these computer programs have been coded to do the best we (or they) possibly can at forecasting what will happen in the future...

Based on the results that we see - it's REALLY freakin hard to know exactly what will happen... Two of the best are the GFS (Global Forecast System- american) and the European Model (ECMWF - European Centre for Medium Range Weather Forecasts), although they have their good days and bad days. --- Canada actually has one to! The CMC-GDPS (Canadian Meteorological Centre)

People who are "into" this kind of thing then monitor the differences between the models, and take note when they actually agree...


A prime example of when they disagree would be hurricane Sandy... The Euro was saying (days in advance) that the hurricane was going to hit the USA, whereas the GFS (days in advance) was saying it was going out to sea... You can guess who won...

The best of the best weather forecasters use their own skill, as well as comparing the "model data" - to make forecasts. They can actually read/understand situations that confuse the models, and predict what will actually happen. It's darn cool!

An alternate example is what we have here, where the models actually AGREE (well in advance) that something interesting could happen... We still don't know for sure, but having both the GFS and the Euro hint at a major Christmas storm, several days in advance, greatly increases the chance of the storm actually occurring.

SO! Back to the main story -


Each individual model run (twice a day for the Euro, 4 per day for the GFS) is run to define exact numbers/weather events. The results are what I often post here... BUT - that is NOT A FORECAST

I repeat, it is NOT a forecast

It should be treated as a glimpse of a "probable event"... Each time the model is run, something different gets spit out... When news broadcasters focus on a single model run, it's totally inaccurate and poor journalism.

With our "potential storm" - we can get a rough idea of when (Dec 24/25) the thing may peak in intensity, and a decent idea that it will be somewhere near Ontario when it does fully "bomb out" - but the exact details are far from certain.

As things start to make more sense, I'll post some details on my thoughts (guesses) as to any potential rare birds.. For now, I present several NEW model runs (from the GFS and Euro), that have all been spit out since my post yesterday

You can see how much it changes each time. Forecasters have a good idea that something is going to happen near Christmas, but exact details of wind direction, snow, rain, total precip, etc are extremely difficult to know in advance... Should be fun though!

Now check these out:

Day 6 and 7 from NOAA's NWS (Weather Prediction Centre) - showing an official FORECASTS

In my experience, the formal forecast is often weaker than the model runs, likely due to the potential that the event may not happen at all. It's rare, but sometimes the models were just plain "wrong" - and storms suddenly disappear from their runs. Check back for more info in the coming days!

(Not only for the birds, but maybe a good heads up that the driving will be rough over the peak days of the holiday season)