psuhoffman

We’ve lost some old regulars in here too the last few years. Hardly ever see zwytz or Ian anymore. But we picked up some new ones and we had more to spare maybe. Hopefully it’s not a larger trend.

I have no idea I don’t ever look at it.

Progressive bias so tuck that in and amp it to 980!

Why is the philly sub so dead? I know many years ago on the old forums when there would be one big storm thread there were some good Philly area Mets and contributors. What happened? Just random chance or did something go down I don’t know about?

2.4” at DCA

better pattern = better probabilities.

Try not to put so much effort into it.

You are exaggerating. In that 6 year period there was 44.4”. In this 6 year period even if we don’t get another flake this winter there has been 69.2”! The only way you have a case is using a weird specific number of years you cherry picked but most importantly you include this year and assume we don’t get 2.4” more this season. Odds are we do and then what? And if we don’t then yes this was the worst 4 years stretch. But not the worst 1 or 2 or 3 or 5 year stretch. What does it mean anyways? This has just been similar to every other snow drought we have at least once every decade except the 1960s.

Ensemble forecasting has increased accuracy with general longwave patterns out to day 10 or so...but the problems really come when people try to pin down details (like the exact location of a low) in the 5-10 day period.

I think the first wave goes north...but a trailing wave on the front is a possibility

You shouldn't be using single op runs at long range for synoptic scale details. If you are...that is user error not model error. But I am not talking LONG range...forecasts were atrocious even at 48 hours before we had the aid of NWP. HUGE busts at 24-48 hours are MUCH less frequent now compared to any time in the past. You are always focused on what the limitations of guidance are. And there are limitations...we cannot model the atmosphere perfectly even with computers. But the fact is they are helping to make forecasts much better than they would be without NWP...and that is the bar. Not are they perfect...but do they make things better...and they do.

cause I know how much you all crave it

problem is there will likely be a few days of dead space behind it also... there could maybe be a little NS vort diving in behind if it ends up not phasing that could bring a clipper type snow. That kind of thing wouldnt show up at range...but Ive seen hints that is possible. But other then a fluke type thing like that...after the weekend we have to wait for the dump of cold getting ejected out of AK. As that presses down into the midwest there will be the natural see saw effect on the longwave pattern and we will ridge here so that first wave will most likely go to our north...maybe so far to our north we don't even get any precip from it. It's after...later that next week that things could get interesting starting with the chance at a trailing wave along the front and then see if the STJ wants to play nice after that.

It wont...but no matter what happens this weekend was always and still is a very low probability proposition. After that we ridge out for a couple days as the initial dump of the AK cold airmass comes into the midwest. So barring some crazy good luck with this weekend thing...we likely are going to have to hope the very good pattern look after Feb 5th is correct and we can score something. Frankly odds wise our chances of snow remain about the same from Feb 1 through March 10th...then go down very dramatically after that. We still have some time to at least save something out of this dreadful winter.

In the past when you would look at a setup that is 48 hours away...and you break down our odds of this or that based on the barometric pressure... you are using the guidance. Otherwise you would have no freaking idea exactly what our barometric pressure would be in 48 hours...or what the pressure at Pittsburgh would be...or anywhere else. Your tools and application of very sound methodology are great...but they would have absolutely no predictive worth beyond about 24 hours without NWP because its impossible to extrapolate the atmosphere very far out in time without the aid of computers. Some of the best minds used to try back before NWP and it was mostly a disaster of busted forecasts.

Thanks for rewording my sentence.

you might be trending towards a meltdown if we don't get some snow soon

Are you assuming they won’t show higher odds when we actually do get a good pattern?

Yea but if we look at the pattern for the whole 15 day period that map covers...it is about "normal" The first 7 days are god awful for snow...with some lottery ticket super phase bomb scenario being the only chance, then after that a warmup as a trough dumps into the central US initially...then it gets good day 10-15...but only 5 of the 15 days have a good pattern...the other 10 are crap...so overall that washes out to about an "average" chance of 3" and since for that period you probably get 3" about 50% of the time...that seems about right...I would say that probability fits the pattern I am looking at pretty good. I think those probability maps have actually been about what they should be given the patterns...so not sure what all the fuss is about.

They have looked very similar a LOT lately because we have spent a LOT of time in the same general pattern (SUCK) lately. But we have had times when we are above 80% on those probability maps (other then when there is a storm like right in front of us and its 100% but that is like DUH) but those were typically periods when we were legitimately about to get slammed like in 2014 or from range before January 2016. I also think there were some pretty good looking probability maps before the run we had in Late Feb into Mar 2015. I also remember seeing some pretty good probability maps when the blocking started in March 2018.. and guess what...all those were times where we actually DID snow. But like I pointed out above 80% of the time in the winter we do NOT get 3" of snow in a 15 day period so if those maps are close to accurate...80% of the time they SHOULDNT show a high probability of snow. So yes MOST of the time they will look pretty sucky because most of the time our snow results will be pretty sucky. The rare times when a legit good pattern is coming you will see those probability maps increase. Look at the one I posted from last night. If we start to see maps like that for multiple runs (with above 50% prob) on the GEFS/GEPS/EPS then we know our odds are improving and that is useful. ETA: I am not saying a snow mean or probability map is the best way to analyze the pattern but those maps have SOME limited usefulness in terms of getting a general sense of what the guidance thinks our odds are in the next 2 weeks. Thats all...

They are serving a purpose though...they have been telling us our chances of snow are not good...and guess what...they havent been. When/if the pattern gets better those maps will show a better probability. Just because its depressing to see that our chances suck everyday doesn't make the information useless. However... @Weather Will I have noticed that over the last couple weeks...when we get the rare good run of a model somehow you forget to post that...its not fair to post all the dreadful ones and then not post when something good pops up...like last nights 0z GEFS run which was littered with hits day 10-15. I know 6z backed off but if we start to see more runs like this it could be a sign that our prospects are improving..especially since the look coincides with a better h5 pattern being progged.

This isnt a new problem... from Dec 1996 to March 2002 DC only had 2 snowy weeks in 6 years... lat January 2000 and March 1999. Other than that the entire 6 year period was a snow wasteland. There are even worse periods of futility than that.

The background base state has SUCKED since mid December... so all of these "threats" have been like buying a powerball ticket then getting excited for the drawing... I will say this...the pattern coming up in February is actually better. It's not the best pattern ever but its legitimately less hostile to snow. We have to see if that continues to progress closer and doesn't "degrade" but if that look ends up real then we could get some shots that aren't full court buzzer beater odds.

They actually do have "SOME" usefulness but the issue is they highlight how much we suck at snow and no one wants that thrown in their face EVERYDAY. But they do show what our chances are with respect to "climo". Problem is...when the mean shows about 30% chance of snow over 15 days...especially when it all comes in a 5 day chunk...that is actually an above normal chance of 3" of snow for DC. I broke down the period from Dec 1 to Mar 15 into roughly 15 day chunks... I basically split each month in half... doing that gives us 210 roughly 15 day periods...and DC got 3" of snow 42 times in those 210 periods...so that is about 20% of the time. So when we have a 30% chance of 3" of snow that simply tells us we have a slightly better than normal chance at 3" of snow. When we see those means get up above 50% that is REALLY good compared to avg. Problem is our average is no snow...80% of the time we go a 15 day period in winter without 3" of snow so...that is just normal. But we do sometimes see that mean near 0% and that tells us its almost a no hope shut out the lights pattern the next 2 weeks... and we get a lot of those. Here is the breakdown in terms of DC odds of seeing 3" of snow during a period in the last 30 years Dec 1-15 13% Dec 16-31 6% Jan 1-15 13% Jan 16-31 33% Feb 1-14 27% Feb 15-28/29 23% Mar 1-15 23%