January 2022 Obs/Disco

[weathafella]

I told you weenies to be patient to 1/5 and beyond.  If that gets delayed we may want to start looking at futility numbers....

[ORH_wxman]

32 minutes ago, Torch Tiger said:

3 bad winters, 3 slightly below to near avg. winters, 2 good winters.  (not factoring in temperature departures).  Not terrible. Were records were lower down near/in the city rather than the airport too?

Garbage decade. I think only the 1930s was worse. Median was 56.3 and mean is about 58”. Only 2 above average. Even the 1980s had 4 above average winters (3 if we count 87-88 as average since it was barely above avg). 

[CoastalWx]

Canadian sipping the egg nog still with a 949 low? Lol. 

[Baroclinic Zone]

35 minutes ago, weathafella said:

I told you weenies to be patient to 1/5 and beyond.  If that gets delayed we may want to start looking at futility numbers....

A week behind schedule it would appear

[Sey-Mour Snow]

20 minutes ago, CoastalWx said:

Canadian sipping the egg nog still with a 949 low? Lol. 

 

[TauntonBlizzard2013]

You almost have to tip your cap to that Canadian solution for next weekend. The first system in god knows how long that doesn’t get shredded to pieces, and it’s going to be heavy heavy rain for 75% of the posters on this forum.

 

[WxWatcher007]

9 minutes ago, TauntonBlizzard2013 said:

You almost have to tip your cap to that Canadian solution for next weekend. The first system in god knows how long that doesn’t get shredded to pieces, and it’s going to be heavy heavy rain for 75% of the posters on this forum.

 

That's how we know it's real.  

[DavisStraight]

59 minutes ago, CoastalWx said:

Canadian sipping the egg nog still with a 949 low? Lol. 

gets down to 942 in Maine

[weathafella]

1 hour ago, Baroclinic Zone said:

A week behind schedule it would appear

Yes.   Let’s hope the train doesn’t get canceled.

[CoastalWx]

4 minutes ago, DavisStraight said:

gets down to 942 in Maine

 

40 minutes ago, Sey-Mour Snow said:

 

LOL I didn’t see that.   that’s awesome 

[MJO812]

[MJO812]

Interesting euro.  2 lows like the other models but the 2nd low is alot colder than the other models

[rimetree]

How about we get the Euro to comprise with itself and get those two lows into one over the benchmark.

[40/70 Benchmark]

3 hours ago, ORH_wxman said:

The 1930s through the early 1950s were pretty much dogshit for BOS snow (and most of New England)…there were a few good years in there like 1947-48…but the bad ones vastly outweighed them. 

But when the skies cleared out at night, it plummeted to 13 instead of 18

[MJO812]

Eps

[moneypitmike]

1 hour ago, MJO812 said:

Eps

Let me know if it (and others) are showing anythingn on Friday night.

[MJO812]

Another gfs solution

[Go Kart Mozart]

So, you are teaching a math class with 51 knuckleheads and one bright student.  You give them a tough problem.  Is the mean answer of the 51 knuckleheads really going to better than the bright student's answer?

Another way to phrase this....does statistical analysis really show that an ensemble mean is better than the operational, and if so, at what time range?  I recall the pros on this site saying that four days-out is the magic cutoff....is that based on analysis mythology?

[weathafella]

42 minutes ago, Go Kart Mozart said:

So, you are teaching a math class with 51 knuckleheads and one bright student.  You give them a tough problem.  Is the mean answer of the 51 knuckleheads really going to better than the bright student's answer?

Another way to phrase this....does statistical analysis really show that an ensemble mean is better than the operational, and if so, at what time range?  I recall the pros on this site saying that four days-out is the magic cutoff....is that based on analysis mythology?

Except ensembles aren’t humans with “known knuckleheads”. Another way of putting it in 1953 as a kid is your healthiest friend died despite being vaccinated polio.  Are the polio vaccines bad?

[Go Kart Mozart]

1 hour ago, weathafella said:

Except ensembles aren’t humans with “known knuckleheads”. Another way of putting it in 1953 as a kid is your healthiest friend died despite being vaccinated polio.  Are the polio vaccines bad?

 

I am not clear on your vaccine analogy.   So Dr. Salk and his team have settled on the "optimal mixture" for the vaccine (the operational).  If 51 kids are given the Salk vaccine, and each kid receives a slightly perturbed mixture (an ensemble member), will those 51 kids do better, statistically, than the one kid receiving the optimal mixture?

I suppose forecasting really can't be compared to vaccines anyway, as vaccination errors don't compound over time in the manner of prediction errors.  I am assuming the benefit of ensembles is to smooth out the "compounding error chaos".  I am sure they have value after X days...I just wonder if we are not prone to something like the Golden Mean Fallacy, where we overrate the ensembles by viewing the average answer as a good answer.

I suspect Tip will chime in on this...if so, please keep it under 50,000 words.

 

 

[UnitedWx]

8 minutes ago, Go Kart Mozart said:

 

I am not clear on your vaccine analogy.   So Dr. Salk and his team have settled on the "optimal mixture" for the vaccine (the operational).  If 51 kids are given the Salk vaccine, and each kid receives a slightly perturbed mixture (an ensemble member), will those 51 kids do better, statistically, than the one kid receiving the optimal mixture?

I suppose forecasting really can't be compared to vaccines anyway, as vaccination errors don't compound over time in the manner of prediction errors.  I am assuming the benefit of ensembles is to smooth out the "compounding error chaos".  I am sure they have value after X days...I just wonder if we are not prone to something like the Golden Mean Fallacy, where we overrate the ensembles by viewing the average answer as a good answer.

I suspect Tip will chime in on this...if so, please keep it under 50,000 words.

 

 

I love Tips 50,000 word descriptions, but prefer the 25,000 version

[weathafella]

14 minutes ago, Go Kart Mozart said:

 

I am not clear on your vaccine analogy.   So Dr. Salk and his team have settled on the "optimal mixture" for the vaccine (the operational).  If 51 kids are given the Salk vaccine, and each kid receives a slightly perturbed mixture (an ensemble member), will those 51 kids do better, statistically, than the one kid receiving the optimal mixture?

I suppose forecasting really can't be compared to vaccines anyway, as vaccination errors don't compound over time in the manner of prediction errors.  I am assuming the benefit of ensembles is to smooth out the "compounding error chaos".  I am sure they have value after X days...I just wonder if we are not prone to something like the Golden Mean Fallacy, where we overrate the ensembles by viewing the average answer as a good answer.

I suspect Tip will chime in on this...if so, please keep it under 50,000 words.

 

 

Yeah the polio was a bad example.   But the idea of perturbations is to see how far slight changes take the guidance.  Since these slight errors are often seen on initialization then perhaps we should get a large sample.   To me, if out of 51 you have 2 extreme outliers maybe they should be discarded.   Statistically the median is a better representation of the ensemble product but I am not aware of median in any of the ens products.

I know the operational models skill beyond d5 is pretty poor so a mix of ensembles should give more robust ideas but either way, it’s rare when the op locks in 7 days out and holds.   We used to see it on the euro but I can’t remember the last time.

 

[HIPPYVALLEY]

This may be inevitable but hopefully flips the script.

 

[OceanStWx]

2 hours ago, Go Kart Mozart said:

 

I am not clear on your vaccine analogy.   So Dr. Salk and his team have settled on the "optimal mixture" for the vaccine (the operational).  If 51 kids are given the Salk vaccine, and each kid receives a slightly perturbed mixture (an ensemble member), will those 51 kids do better, statistically, than the one kid receiving the optimal mixture?

I suppose forecasting really can't be compared to vaccines anyway, as vaccination errors don't compound over time in the manner of prediction errors.  I am assuming the benefit of ensembles is to smooth out the "compounding error chaos".  I am sure they have value after X days...I just wonder if we are not prone to something like the Golden Mean Fallacy, where we overrate the ensembles by viewing the average answer as a good answer.

I suspect Tip will chime in on this...if so, please keep it under 50,000 words.

Ideally a good ensemble will make sure that the ultimate result falls within the ensemble forecast goalposts, not smooth out the errors. You actually want those errors to compound and lead to different solutions because we know the model is going to have those errors from the observed atmosphere anyway. 

2 hours ago, weathafella said:

Yeah the polio was a bad example.   But the idea of perturbations is to see how far slight changes take the guidance.  Since these slight errors are often seen on initialization then perhaps we should get a large sample.   To me, if out of 51 you have 2 extreme outliers maybe they should be discarded.   Statistically the median is a better representation of the ensemble product but I am not aware of median in any of the ens products.

I know the operational models skill beyond d5 is pretty poor so a mix of ensembles should give more robust ideas but either way, it’s rare when the op locks in 7 days out and holds.   We used to see it on the euro but I can’t remember the last time.

Median heights? No, but there are median values for other variables. I also think some of the clustering analysis gets closer to median solutions too. 

[MJO812]

6z gfs is weenie 

[Snowcrazed71]

For some reason.. You can't post pictures ( won't allow anything over 1.95 MB ) which is pretty much everything? Lol

[MJO812]

26 minutes ago, Snowcrazed71 said:

For some reason.. You can't post pictures ( won't allow anything over 1.95 MB ) which is pretty much everything? Lol

Have to delete your attachments 

[HoarfrostHubb]

1 hour ago, Snowcrazed71 said:

For some reason.. You can't post pictures ( won't allow anything over 1.95 MB ) which is pretty much everything? Lol

You can (on an iPhone at least) choose the size of the pic you are adding.  You can choose a low or medium sized file 

[40/70 Benchmark]

2 hours ago, MJO812 said:

6z gfs is weenie 

At like day 11.....have you noticed a pattern?

[Baroclinic Zone]

Just now, 40/70 Benchmark said:

At like day 11.....have you noticed a pattern?

You're melting.