Joe B. brings brutal cold and more Snow to Texas....

[tano68]

Lastly, if the NWS forecasted like JB we would be in a state of panic all the time. In fact we would probably do nothing but sit at home waiting for this week's distaster to strike. The ramifications of NWS forecasts extend well beyond the computer screen. Extreme statements and forecasts at long range time periods are ill advised.

[sertorius]

Jeff:

Excellent point-NWS has responsibility beyond cpmprhension when you think about it-the NWS out of Topeka does a great job-even this morning, they said we have little conf. in the forcast due to differing temp. solutions thru out the week-but I think thier call of 40's is reasonable given the pattern and our location. Also, the models have done horrible the past 2 weeks for my area at least-just now is the gfs comming back to its extreme arctice out break-the NWS can only forcast off the data they are given esp. in long range.

[Portastorm]

A few comments:

First, the snows of christmas Eve in S TX were historical. You must go back to 1895 to get totals anywhere close to those seen with the 2004 event. It is extremely unlikely that any of S TX or SE TX will see snowfall totals of 8-13 inches again within our lifetimes. The chances of it happening on Christmas are even smaller and really near 0.

It is true that light amounts of snow are a little more common. Many believe snow is rare since we have not had it this far south since 1989, however a look back shows several snow events from the 50' to the 70's (1-6 inches) with the 90's snow drought.

Now looking at next week. Cold air will be coming south, now I don't see it as record breaking, but highs near freezing as far south as the coast will be possible. The whole deal with the moisture and winter precip. is new in the model data with the GFS really latching on today. However, I have questions with the profiles along the Gulf coast and it would appear most "frozen" precip will be inland around Dallas. Freezing rain and icing however would be possible as far soth as Victoria to Lake Charles do to a shallow cold air mass. This is not the same pattern as Christmas, as the trough is directed more at the SE US and not the S plains.

We will see how this all plays out over the next few days

Lastly, if the NWS forecasted like JB we would be in a state of panic all the time. In fact we would probably do nothing but sit at home waiting for this week's distaster to strike. The ramifications of NWS forecasts extend well beyond the computer screen. Extreme statements and forecasts at long range time periods are ill advised.

I will agree that "extreme statements and forecasts at long range time periods are ill advised." However, I also think that when you are forecasting you should take into account ALL of the tools are your disposal. Unless the NWS has a mandate to its forecasters that they cannot use anything but the GFS, then these folks should also be looking at other medium range models as they forecast 5-7 days out.

In my job I am expected to use all of the information I am privy to in helping advise management on making decisions. I'm sure it is the same with most of us. But for anyone who reads forecast discussions on a regular basis (and I do), it is clear that the majority of NWS forecasters in Texas go with the GFS and nothing else. Well, the GFS is a computer model and is subject to error like any other model. And in certain patterns it is known to perform poorly. So, with all of these biases in mind, why not consider all the facts at hand? I think that is what we're saying. I don't think any of us are taking the NWS to task for not forecasting a snowstorm or something.

Hey, I don't know ... maybe they're too busy in a shift to examine all of that data and I would understand. But if they're not, and they're not bothering to look at anything beyond the GFS, well then it is pure laziness.

As for JB, he is no doubt prone to being dramatic. But given his performance history this winter, I would safely go with his thoughts. He's been right a lot more than wrong. Last winter he was lousy. But in the last 3 months he's had the hot hand.

[cctxhurricanewatcher]

Jeff,


Portstorm said much of what I would have said about the NWS and using other models. For example, when I was really getting intrested in weather observation when I was a teenager back in the 1980, I recall listening to the "old school" forecaters from NWS Brownsville on my Radio Shack cube weather radio about the potential cold outbreaks. I never recall them blowing of the threat of a severe cold snap like the "New School" forecasters do now. If the GFS says it's going east, by golly that is what they put in their AFD's. The Old School guys never would jump the gun like that based on on or two model runs from a model that is known to be less the reliable with cold in Texas.

[jeff]

I work closely with NWS and I know the amount of work required on a normal sunny day...marine, aviation, and public forecast. Toss in severe weather and that work load triples. Say ice and snow a week or so out and you will be dealing with none stop public phone calls for a week, and then the media gets wind of if and they want to set up in the waiting room of the office hounding every forecaster as they walk in the door. Then local OEM's get involve and start requesting conference calls, TXDOT wants updates, ect, ect, ect. It is still a week away yet everyone wants to know right now what is going to happen in 7 days. The models do their usual flip flop and the forecast changes, then everyone wants to know why. This is why the NWS uses broad statements and are very cautious about what they release publicly. During a tropical situation you can multiply this by 100 for public phone calls, media interviews, conference calls ect.

The public holds forecasters accountable for what they say and the forecast they produce. You bomb a forecast (and it happens) you will hear about it. Everyone thinks that weather predict is straight forward, well it is not, in fact it is nowhere near easy (I think most on here realize this). Try forecasting a training rainfall event (location and amounts) 12 hours out!!!!

[Portastorm]

And just so fair is fair, I'm going to give a real "thumbs up" sign to the NWS Texas guys for the afternoon discussions I read. All of them say -- and I will paraphrase -- "hey, it's going to get very cold late this next week. How cold we're not sure yet but cold enough that it could be the season's coldest."

That's all I'm asking. Not a 7-day-out forecast for "25 and snow in Austin" on Saturday, Jan. 15th. Just a good, solid, sensible heads up and none of this "modified air mass" or "Pacific front" silliness that we saw yesterday and the day before.

[amawea]

I have been reading J.B's forecast for at least 4 years. I firmly believe he is the best long range winter weather forecaster out there.
I also believe that the nws can't stand it that a private weather service might out do them. Heck. As far as I can see the NWS totally relies on the models. Sheesh. Where is the human element in correcting the model bull crap. At least J.B. sticks his neck out. And btw. he nailed the Texas coast snow event 10 days out.
amawea