First to let you know that for some reason the Geniuses at Facebook are blocking my ability to create new posts and new comments on my own Facebook page. I think it is just a temporary technical glitch but it is taking a while to resolve which is why I have not been able to make any updates since early on Thursday morning.

Here we are at 3 PM Thursday afternoon and it is clear that the weather models have once again completely screwed up. It is not just that MY forecast for the snow in the northern Shenandoah Valley ….the Eastern portion to WV ….central and eastern MD .. southeast PA that did not verify. It is clear after you look at all the different various sources out there both… NWS TWC accu wx or other private meteorologists or TV meteorologists… that absolutely none of the forecast have verified outside of a few isolated locations.

NWS in the central and northern portions of the Shenandoah Valley had anywhere from 3 to as much as 8 inches of snow. That simply did not happen.

The severe icing anticipated along the NC VA border into the northern locations of northwest NC In locations such as Greensboro Winston-Salem Wilkesboro and Hickory …has not developed. Temperature stayed above 32 degrees as late as 3 or 4 am. Thursday. Yes the temps did drop to 31 degrees but the precipitation as you can tell from the radar has come to an end.

To the best of my knowledge this is now the fourth time since the pattern turned active in the middle of January that the winter forecast and winter storm watches from the Washington DC Roanoke & Wakefield offices have busted. and yes my forecast have busted too. Let me show you what I’m talking about

Let’s take a look regarding how bad even the short range weather models have handled this event which is symptomatic of all the big winter weather events for the Middle Atlantic region we have had so far this season.

This image is the 1PM THURSDAY . Let’s compare this to the 6Z short range models which generally come out between 3 and 4 am. The idea here is that getting the next 12 hours right should be fairly easy and at least have some connection to reality when you look at conditions at 1 on Thursday afternoon.

But as you will see almost all of the short-range models except for one completely busted only 12 hours after they initialized. Not three days later. Not 5 days . But just 12 hours.

Here is the operational GFS model from 6z valid for 1PM Thursday — only 12 hours later from when this model was initialized .When you compare with a model showing here to the axle one radar it’s stunning how bad the GFS 12 hour forecast actually is.

Here is the 6Z rgem — short-range high-resolution Canadian model which normally is a pretty good model. But here 6z RGEM Is unbelievably bad. It has a large area of significant freezing rain covering the western half of NC into the Richmond metro area and a huge area of heavy sleet in all of the Shenandoah Valley …into SC BAL / northern VA then into the northern Delmarva / southern NJ. . But as the 1PM radar shows that that is not the case at all.

Here is the high resolution supposedly improved version of the GFS model known as the GFS fv3 or GFS 16 valid for 1pPM Thursday. Again It is way off from reality. There is no band of moderate or even Light freezing rain in Eastern Virginia into the Richmond area or in southern Maryland at 1 pm.

Here is the early morning 0z EURO European model valid at 18 hours 1 on Thursday. Note that the model was showing an area of light to moderate freezing rain in the VA Piedmont into northern VA a DC Metro Baltimore and northern DEL. And none of that is remotely close to being correct based upon the 1 PM OBS/ radar. Notice the widespread light to moderate rain covering all of central & eastern North Carolina Southeast Virginia and the Delmarva. but that does not match the actual surface conditions or the 1 p.m. radar at all.

What is surprisingly accurate again is the 6z NAM 3km model valid for 1 p.m. This model seems to have an accurate depiction of what’ is going on in the Middle Atlantic region with the diminishing freezing rain and rain … sleet in NJ and snow in northern NJ and NYC.

Indeed over the past several events it has been the 3KM NAM model which has consistently shown the least amount of snow ….the weakest kind of weather systems… and shortest duration of the event Meanwhile the the Canadian GFS and the European models have consistently shown significantly more snow and ice and a longer duration event.

And finally here is the 8Z HRRR model also valid for 1 PM and this model is also pretty accurate to what’ is actually going on at the surface and on the radar.

Obviously the question then becomes what is going on with these short-range models that they cannot even get the next 12 hours correct? Why are they off this much just trying to figure out the next 12 hours? I have seen this sort of collapse in the short range weather models before — and it turns out that cause is often the LA NINA.

On the other hand the global and short-range models have done a much better job in places like PA NJ NYC … New England as well as the Ohio Valley ….the Midwest and the Plain states. In those areas the severe cold DID show up as forecasted days in advance and in those locations the snow did fall as the models were indicating. For example …the various models showed significant snow for the past several days was going to fall late Wednesday night into Thursday morning in portions of Tennessee and Kentucky and that is exactly what happened. There was no breakdown of the weather models in these areas .

Why this keeps happening in Virginia Maryland Delaware and to a lesser degree in Northwest North Carolina is a mystery to me. But I will keep thinking and looking at the data to try and figure out what the answer is.



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
DT Wxrisk

DT Wxrisk

Meteorologist ... Atheist.. Dyslexic ..Baseball.. Fat tail distributions ..Good Judgement Projection… Black Swans/ Taleb …Choas / non Linear Dynamics… ENTP