The Wednesday afternoon radar shows moderate to heavy snow over the northern third of Minnesota into southwestern portions of western Ontario Canada. Most of central and northern Wisconsin is seeing steady rain or thunderstorms and significant showers and thunderstorms some of which are becoming severe have developed over central and northwestern Arkansas …the southern half of Missouri …and the far eastern portions of Oklahoma. FAST EAST LOADING RADAR
The midday 12z Wednesday operational or regular GFS model came out and it showed a much drier looking pattern for the 6-10day over the Plains and the Midwest when compared to the previous model runs. This was also the case with the operational regular 11 to 15 day GFS model.
All weather models have bad runs OR what we call flip flops in the weather biz. One the reasons why the GFS is constantly losing in terms of its accuracy and performance to the European model is because the GFS model – for a lot of different reasons -seems to have more these bad model runs where it comes up with these goofy/ weird solutions.
One way to get around this is by taking a look of the operational run and comparing it to the model ENSEMBLE. Typically the operational run should be close to very close to the ensemble mean. But as you can see in this image below that is NOT the case as the GFS ensemble in the 6-10 DAY is significantly wetter than the operational run.
Now that being said let’s be clear…-the 6 to 10 day forecast IS going to be drier than the NEXT 5 DAYS.
But it will not be as dry dry as what the operational or regular GFS model was showing.
The Canadian 6-10 day ensemble is very close to the GFS ensemble 6-10 day
Next let’s take a look of the 11 to 15 day and compare the operational run to the GFS ensemble.
Again we can see that the in GFS ensemble it is significantly wetter than the operational run.
And again the 111-5 day Canadian Model ensemble closely matches the GFS 11-15 day ensemble.
SUMMARY the result of this is that it is pretty clear that the operational GFS in the 6 to 10 day and in the 11 to 15 day seems to be a bad model run. Those who think that the
pattern is suddenly going to towards a more sustained a dry interval in early and mid MAY 2017 might want to exercise some caution here.
If you missed last week’s webinar, you may VIEW A RECORDING.