What Went Wrong in Winter 23-24/Base State/Will It Ever Snow Again??

[WEATHER53]

From page 47

First-The numerical models will have enough accuracy that errors will no longer be significant 

Second-accuracy and resolution producing crisp, clear and complete  pictures of the atmosphere at all times 

Third-By 2025 the power of computers will have increased 100,000X and will clearly resolve the weather for the (DC) area.

These writers  and contributors were not noobs nor  foolishly  hopeful.  They were up to date and on target with their current situation and where things were headed.

What happened?

Well one thing we see clearly right here-vigorous defense of status  quo. Almost like we are helpless if we try another route.

Probably  one more entry of where the best and brightest of  the time validly thought this science was headed.  What happened????

[Ji]

4 hours ago, WEATHER53 said:

From page 47

First-The numerical models will have enough accuracy that errors will no longer be significant 

Second-accuracy and resolution producing crisp, clear and complete  pictures of the atmosphere at all times 

Third-By 2025 the power of computers will have increased 100,000X and will clearly resolve the weather for the (DC) area.

These writers  and contributors were not noobs nor  foolishly  hopeful.  They were up to date and on target with their current situation and where things were headed.

What happened?

Well one thing we see clearly right here-vigorous defense of status  quo. Almost like we are helpless if we try another route.

Probably  one more entry of where the best and brightest of  the time validly thought this science was headed.  What happened????

the models have been predicting a cold 2nd half of march now since Mid February...and it looks like we are on track. Thats pretty impressive dude!

[WxUSAF]

 

[Terpeast]

35 minutes ago, WxUSAF said:

 

Next winter the sfc temp anomaly may look pretty much like this map if the pacific doesn't change and we get a +qbo nina. Could be historically warm and snowless for the immediate east coast up to coastal SNE. 

At some point, we've gotta get a dip that breaks the climo trend. Don't know when that's gonna be.

[pazzo83]

On 3/7/2024 at 12:28 PM, WEATHER53 said:

See you dismiss as “hopeful” 

You are not qualified to do that.  You were about 15 when he wrote it and are not a met  nor atmospheric scientist. You are able to provide great detailed discussions with very little forecasting . You are most certainly and ardent model hugger and defender of the failing status quo 

The abilities you assess us as “not having” are largely because they have not been sought due to defenders  of. and monetary concerns about, the preservation of the current operating methods 

Howard you need to take a break from here.

[WEATHER53]

13 hours ago, pazzo83 said:

Howard you need to take a break from here.

When I seek such advice, please render it

For now cling to your wishes and dependencies and I will continue to not engage in denial. 
 

I’ll add the Fact that you concerns be addressed Not to someone who posts real info 3-5 times per day but rather those averaging 40+ and more Every Day. 
 

[pazzo83]

1 hour ago, WEATHER53 said:

When I seek such advice, please render it

For now cling to your wishes and dependencies and I will continue to not engage in denial. 
 

I’ll add the Fact that you concerns be addressed Not to someone who posts real info 3-5 times per day but rather those averaging 40+ and more Every Day. 
 

I'll dish out advice when I see fit.  Your posts have been nasty and condescending on here for quite some time.  Take some time off.

[cbmclean]

On 3/8/2024 at 7:47 PM, WxUSAF said:

 

A fascinating graphic: some musings:

1. The east has clearly been hit harder than the west.  This actually gives me a bit of hope.  Yeah there is nasty background warming but perhaps there really is a multidecade cycle that happens to be "favoring" eastern NA for warmth.  

2. On the other hand, the eastern CONUS is on average lower in elevation than the west which suggests that it is less able to rely on elevation for cold and is this more dependent on cold air advection from the source regions.  So it makes sense that combined trends of warmer source region and increased pacific influence are hurting the east worse.

3. Man 1998-2000 looks like it was an ugly period.

[MillvilleWx]

1 hour ago, cbmclean said:

A fascinating graphic: some musings:

1. The east has clearly been hit harder than the west.  This actually gives me a bit of hope.  Yeah there is nasty background warming but perhaps there really is a multidecade cycle that happens to be "favoring" eastern NA for warmth.  

2. On the other hand, the eastern CONUS is on average lower in elevation than the west which suggests that it is less able to rely on elevation for cold and is this more dependent on cold air advection from the source regions.  So it makes sense that combined trends of warmer source region and increased pacific influence are hurting the east worse.

3. Man 1998-2000 looks like it was an ugly period.

Outside the come from behind victory in Jan 2000, there was no winter to be found in these parts. I was really young, but I remember basically getting out early all 3 years from a lack of snow days used. It was a rough stretch. Add in the March 2001 fail and dismal 2002 and you got yourself a 5 year stretch of rough winters. 2003 is when it broke, and it broke big. There’s been terrible streaks of futility throughout history in these parts. We’re on a cooler, that’s for sure. 

[Jebman]

51 minutes ago, MillvilleWx said:

Outside the come from behind victory in Jan 2000, there was no winter to be found in these parts. I was really young, but I remember basically getting out early all 3 years from a lack of snow days used. It was a rough stretch. Add in the March 2001 fail and dismal 2002 and you got yourself a 5 year stretch of rough winters. 2003 is when it broke, and it broke big. There’s been terrible streaks of futility throughout history in these parts. We’re on a cooler, that’s for sure. 

It's getting so cool, folks are heading out west to Vail to ski.

[uncletim]

On 3/8/2024 at 7:47 PM, WxUSAF said:

 

Would love to see this go back a couple of decades further. Growing up in Faifax County in the sixties and early seventies, I learned to ice skate and never was on a rink until my last couple of years in high school. Ponds froze for at least a handful of weeks most winters. The other reason I would like to see it go back further is to see how those late seventies winters look compared to earlier. Was that period as anomalous in context of the previous couple of decades?

[psuhoffman]

7 hours ago, cbmclean said:

A fascinating graphic: some musings:

1. The east has clearly been hit harder than the west.  This actually gives me a bit of hope.  Yeah there is nasty background warming but perhaps there really is a multidecade cycle that happens to be "favoring" eastern NA for warmth.  

2. On the other hand, the eastern CONUS is on average lower in elevation than the west which suggests that it is less able to rely on elevation for cold and is this more dependent on cold air advection from the source regions.  So it makes sense that combined trends of warmer source region and increased pacific influence are hurting the east worse.

3. Man 1998-2000 looks like it was an ugly period.

Unfortunately I think the fact the boundary layer is warming faster is the simplest most likely culprit of that. Same reason the more elevated parts of our region haven’t suffered as much the last 8 years, and I mean compared to average, obviously they will always get more snow.  But places with some elevation in our region are closer to avg over the last 10 years than 95. 
 

Also you’re adding heat from the gulf. That’s probably some of it too. But neither of those factors is going to change. 

[cbmclean]

5 hours ago, psuhoffman said:

Also you’re adding heat from the gulf. That’s probably some of it too. But neither of those factors is going to change. 

I am hopeful that some of the gulf warmth is due to the eternal SER associated with our --------PDO episode.  IF (big IF) that backs off in a +PDO hopefully the gulf warmth can back off some.

[psuhoffman]

49 minutes ago, cbmclean said:

I am hopeful that some of the gulf warmth is due to the eternal SER associated with our --------PDO episode.  IF (big IF) that backs off in a +PDO hopefully the gulf warmth can back off some.

That map included the last +PDO cycle 

[cbmclean]

3 hours ago, psuhoffman said:

That map included the last +PDO cycle 

Good point.

[psuhoffman]

@WEATHER53 here is a discussion that has nothing to do with the models... you should chime in, I would appreciate your input.  @Terpeast@WxUSAF @brooklynwx99would also love your thoughts as always!  

I have done some initial analysis of trends and what went wrong with my analog identification process for this winter, as well as 2020 when I also drastically over predicted snowfall.  This is in its initial stages but I think a few adjustments to my composite score needs to be made.  

When I first developed my methodology I backwards applied it for 20 years and found it to predict snowfall about 65% of the time.  This was scoring with a "much above, slightly above, near normal, below, much below" five category verification.  And when it missed it was often only by one category.  Very rarely was it 2 or more categories off.   However, since 2016 it seems to have absolutely no correlation to snowfall.  My analog scoring system is no longer effective.  I have started to analyze what seems to have changed.

First of all ENSO needs to be dropped down some, it would still be one of the more significant factors but not as dominant in the score calculation as it was.  The QBO seems to have almost no correlation to snowfall since 2016.  The issue is that the QBO is correlated to snowfall secondarily.  It's primary correlation is to the AO and SPV.  Those correlations are still strong...the issue is that the AO and SPV do not seem to correlate to snowfall here nearly as much as they did prior to 2016.  This break is really screwing up my analog method.  Another factor is I need to increase the PDO score.  But not just the raw number.  I think I was already weighting the PDO raw numeric correctly, but that I have found within the PDO there are smaller scale cycles of extreme periods of negative, positive, and neutral periods.  And a neutral period coming out of an extreme negative is not the same as one coming from a positive for example.  I need to factor where in the PDO cycle we are as well as the raw number.  Lastly, I need to change now I adjust for warming.  The method of simply decreasing the mean snowfall of the analog mean by 15% is no longer sufficient.  First of all it seems it may be closer to 20% or more lately.  But the other issue is I need to get more detailed and score on a curve.  1965 would need to be adjusted way more than 2015 for example.  But more so I need to look at how the snow came.  For example...1966.  It's one of the few very snowy winters that was still high on my new analog methodology I will explain soon, but most of that snow evaporates if I adjust the snowfall with my new method.  There were 3 significant MECS level snows that winter.  But 2 of them came with temps marginal near or above freezing.  Those need to be adjusted more than a winter like 2014 or 2015 when the snow came with true arctic air around.  Plus 1966 needs to be adjusted more due to the time period.  Those adjustments would almost totally eliminate 2 of the 3 major snowstorms.  That leaves 1966 as only near normal in snowfall and all the snow coming from one cold period in mid January...starting to sound familiar?  

Anyways...this is still early in my tinkering stages but if I drop the enso score about 25% and add my new PDO cycle method into the equation as an equal factor to the enso, drop the QBO by 50%, and apply my more detailed method for snowfall adjustment across seasons, I get these new analogs in hindsight for this past winter.  

1972-73, 1953-54, 2009-10, 1965-66, 1951-52, 2006-07

Using my new methodology, decreasing the snowfall with a more detailed methodology, and dropping the one OBVIOUS outlier of 2010, the new mean snowfall prediction for this winter would have been 11".  This new method would have nailed it.  And using this method for the post 2016 period it seems to have about the same success that my old method had prior to 2016.  These changes seem to adjust for the new normal or current cycle if you will.  Take your pick.  

Interestingly...I have no idea what the hell 2010 is all about.  It still scores very very high.  It came during the middle of an equally awful PDO cycle.  Think about how bad 2007 to 2013 was without 2010.  If you adjust for the longer scale rate of snowfall decline hidden within the cyclical variance it was just about as bad as this current 8 year period, with the exception of 2010 being in the middle to break it up.  But imagine if instead of what we got in 2010 we had a 10" winter that year...we would remember that period just as bad as this one.  What happened in 2010 needs more examination.  

Lastly...looking at this PDO mini cycles, I think we are due for a flip in about 1-2 years.  The PDO cycles are shortening for some reason.  This current deeply negative one started in 2020.  The deeply negative PDO cycles typically last about 5-7 years, but again are trending shorter.  What is making this one seem worse is we had a couple of pretty bad dud years leading into this period which is partly bad luck and partly due to the fact bad years are becoming statistically more likely any given season.  But either way, this current DEEPLY negative PDO shouldn't last too much longer.  Doesn't mean the PDO suddenly becomes awesome, we could have a neutral period following this.  Or we could see a strongly positive AO offset a better PDO.  But many of our awful PDO cycles DID have a great snowfall period bookending them on both sides so its possible we see a snowier period soon.  But I wouldn't bank on it until we see signs the PDO cycle is flipping.  

 

[Jebman]

Okay. Got it. From now on I will use the PSU Metric in my calculations for upcoming winters. Wow man you are already sussing the changes out. This is going on my KBE board and on Science Weather.

[stormy]

21 hours ago, psuhoffman said:

@WEATHER53 here is a discussion that has nothing to do with the models... you should chime in, I would appreciate your input.  @Terpeast@WxUSAF @brooklynwx99would also love your thoughts as always!  

I have done some initial analysis of trends and what went wrong with my analog identification process for this winter, as well as 2020 when I also drastically over predicted snowfall.  This is in its initial stages but I think a few adjustments to my composite score needs to be made.  

When I first developed my methodology I backwards applied it for 20 years and found it to predict snowfall about 65% of the time.  This was scoring with a "much above, slightly above, near normal, below, much below" five category verification.  And when it missed it was often only by one category.  Very rarely was it 2 or more categories off.   However, since 2016 it seems to have absolutely no correlation to snowfall.  My analog scoring system is no longer effective.  I have started to analyze what seems to have changed.

First of all ENSO needs to be dropped down some, it would still be one of the more significant factors but not as dominant in the score calculation as it was.  The QBO seems to have almost no correlation to snowfall since 2016.  The issue is that the QBO is correlated to snowfall secondarily.  It's primary correlation is to the AO and SPV.  Those correlations are still strong...the issue is that the AO and SPV do not seem to correlate to snowfall here nearly as much as they did prior to 2016.  This break is really screwing up my analog method.  Another factor is I need to increase the PDO score.  But not just the raw number.  I think I was already weighting the PDO raw numeric correctly, but that I have found within the PDO there are smaller scale cycles of extreme periods of negative, positive, and neutral periods.  And a neutral period coming out of an extreme negative is not the same as one coming from a positive for example.  I need to factor where in the PDO cycle we are as well as the raw number.  Lastly, I need to change now I adjust for warming.  The method of simply decreasing the mean snowfall of the analog mean by 15% is no longer sufficient.  First of all it seems it may be closer to 20% or more lately.  But the other issue is I need to get more detailed and score on a curve.  1965 would need to be adjusted way more than 2015 for example.  But more so I need to look at how the snow came.  For example...1966.  It's one of the few very snowy winters that was still high on my new analog methodology I will explain soon, but most of that snow evaporates if I adjust the snowfall with my new method.  There were 3 significant MECS level snows that winter.  But 2 of them came with temps marginal near or above freezing.  Those need to be adjusted more than a winter like 2014 or 2015 when the snow came with true arctic air around.  Plus 1966 needs to be adjusted more due to the time period.  Those adjustments would almost totally eliminate 2 of the 3 major snowstorms.  That leaves 1966 as only near normal in snowfall and all the snow coming from one cold period in mid January...starting to sound familiar?  

Anyways...this is still early in my tinkering stages but if I drop the enso score about 25% and add my new PDO cycle method into the equation as an equal factor to the enso, drop the QBO by 50%, and apply my more detailed method for snowfall adjustment across seasons, I get these new analogs in hindsight for this past winter.  

1972-73, 1953-54, 2009-10, 1965-66, 1951-52, 2006-07

Using my new methodology, decreasing the snowfall with a more detailed methodology, and dropping the one OBVIOUS outlier of 2010, the new mean snowfall prediction for this winter would have been 11".  This new method would have nailed it.  And using this method for the post 2016 period it seems to have about the same success that my old method had prior to 2016.  These changes seem to adjust for the new normal or current cycle if you will.  Take your pick.  

Interestingly...I have no idea what the hell 2010 is all about.  It still scores very very high.  It came during the middle of an equally awful PDO cycle.  Think about how bad 2007 to 2013 was without 2010.  If you adjust for the longer scale rate of snowfall decline hidden within the cyclical variance it was just about as bad as this current 8 year period, with the exception of 2010 being in the middle to break it up.  But imagine if instead of what we got in 2010 we had a 10" winter that year...we would remember that period just as bad as this one.  What happened in 2010 needs more examination.  

Lastly...looking at this PDO mini cycles, I think we are due for a flip in about 1-2 years.  The PDO cycles are shortening for some reason.  This current deeply negative one started in 2020.  The deeply negative PDO cycles typically last about 5-7 years, but again are trending shorter.  What is making this one seem worse is we had a couple of pretty bad dud years leading into this period which is partly bad luck and partly due to the fact bad years are becoming statistically more likely any given season.  But either way, this current DEEPLY negative PDO shouldn't last too much longer.  Doesn't mean the PDO suddenly becomes awesome, we could have a neutral period following this.  Or we could see a strongly positive AO offset a better PDO.  But many of our awful PDO cycles DID have a great snowfall period bookending them on both sides so its possible we see a snowier period soon.  But I wouldn't bank on it until we see signs the PDO cycle is flipping.  

 

You seem to be confident that a PDO flip will make a difference. I'm not convinced.

During the past 100 years, other factors have resulted in snow droughts for 10 years or more.

Please give me examples of how a positive PDO enhanced snowfall.

[pazzo83]

oh man stormy's not convinced - time to shut it down.

let's ignore the fact that psu has repeatedly provided examples of the positive correlation between a positive PDO and regional snowfall here - you are aware that the site has a search function, right?

[stormy]

17 hours ago, stormy said:

You seem to be confident that a PDO flip will make a difference. I'm not convinced.

During the past 100 years, other factors have resulted in snow droughts for 10 years or more.

Please give me examples of how a positive PDO enhanced snowfall.

Regardless of your theory, I'm not convinced a positive PDO will be the magical elixir for snowfall because the 50's and 60's both witnessed a mostly negative PDO with a brief positive from about 1958 - 1962.  Under mostly negative conditions as we have witnessed since 2020, the 50's witnessed below normal snowfall and the 60's witnessed above normal snowfall.  I don't find any significant impact because of the PDO phase.  

[WEATHER53]

47 minutes ago, stormy said:

Regardless of your theory, I'm not convinced a positive PDO will be the magical elixir for snowfall because the 50's and 60's both witnessed a mostly negative PDO with a brief positive from about 1958 - 1962.  Under mostly negative conditions as we have witnessed since 2020, the 50's witnessed below normal snowfall and the 60's witnessed above normal snowfall.  I don't find any significant impact because of the PDO phase.  

Most index assertions that have blossomed in the last 10 years are theoretical 

Why has the mid Atlantic switched from about 50% As and 50%Bs to about 80% Bs.

 

[cbmclean]

54 minutes ago, stormy said:

Regardless of your theory, I'm not convinced a positive PDO will be the magical elixir for snowfall because the 50's and 60's both witnessed a mostly negative PDO with a brief positive from about 1958 - 1962.  Under mostly negative conditions as we have witnessed since 2020, the 50's witnessed below normal snowfall and the 60's witnessed above normal snowfall.  I don't find any significant impact because of the PDO phase.  

I don't think anyone has been touting the PDO flip as a magic elixir, at least not in any of the serious posts I have been seeing.  The general theory (and hope) is that some of the suck we have been seeing in the last ~8 years is due to a prolonged period of very negative PDO and that when that relaxes we may see some improvement.  I don't think many are expecting it to go back to "normal".  It just MAY not suck so bad.

@psuhoffman has been the person I have talk most about this topic and he has made it VERY clear that he believes that at at best a +PDO will only mitigate the suck and that there are non-cyclical factors leading to an ever-worsening base state, which is my own belief as well. 

What impact, if any will a +PDO have?  TBD.

[stormy]

17 minutes ago, cbmclean said:

I don't think anyone has been touting the PDO flip as a magic elixir, at least not in any of the serious posts I have been seeing.  The general theory (and hope) is that some of the suck we have been seeing in the last ~8 years is due to a prolonged period of very negative PDO and that when that relaxes we may see some improvement.  I don't think many are expecting it to go back to "normal".  It just MAY not suck so bad.

@psuhoffman has been the person I have talk most about this topic and he has made it VERY clear that he believes that at at best a +PDO will only mitigate the suck and that there are non-cyclical factors leading to an ever-worsening base state, which is my own belief as well. 

What impact, if any will a +PDO have?  TBD.

 The current "deeply negative" PDO began in 2020. That is only 4 years ago. Prior to that it was moderate/strongly positive.  During nearly all of this time we have been well below normal snowfall whether or not it was cold enough to snow.  A lot of this has been dependent on timing and track of precipitation bearing systems.

[psuhoffman]

3 hours ago, stormy said:

 The current "deeply negative" PDO began in 2020. That is only 4 years ago. Prior to that it was moderate/strongly positive.  During nearly all of this time we have been well below normal snowfall whether or not it was cold enough to snow.  A lot of this has been dependent on timing and track of precipitation bearing systems.

I am going to address this and your previous post here.  

From 2014 to 2018 was a positive PDO spike and taken as a whole was a very snowy period.  From 2007 to 2013 the PDO was negative and other than 2010 those years all sucked.  

There is more to the PDO than the raw numbers.  It is also NOT a magic bullet operating in a vacuum.  But if you look at the last 5 deeply negative PDO periods, they all sucked major ass wrt snowfall here.  1949-1957, 1971-1976, 1989-1992, 2007-2013 and 2020-2024 were the last 5 deeply negative PDO periods and they all were incredibly low snowfall here.  

But the PDO is not the end all wrt snowfall.  We can get a +PDO low snowfall season if other factors are not good.  A +AO can offset a +PDO, for example.  And we have in the past been able to get a snowy winter in a -PDO, but those mostly came during weak or moderately -PDO periods not strongly -PDO cycles.  Look at the 1960s for example...and the snowiest seasons came early in the 60s when the PDO was positive...yes it remained snowy into the -PDO cycle later but it was a weak to moderate -PDO cycle NOT a deeply negative one.  Once the PDO went deeply negative in the early 70's the snowfall stopped!  

There is a matter of degrees to this.  No one factor makes up 100% of the equation wrt our snowfall.  But what the last 75 years suggests is that when the PDO goes into a deeply negative cycle we are in big trouble and it tends to suck.  A deeply negative PDO cycle seems to overwhelm the rest of the pattern and it's very hard for us to get much snow regardless of what those other factors are doing during these ---PDO periods.  

When the PDO improves does it mean we suddenly get a ton of snow?  No.  We might get a +AO season, in which case the PDO won't matter.  The truth is we are south of where it reliably snows and so we need multiple factors to line up in order to get snow.  But one of those factors is the PDO and it would be very helpful if it would get out of the suck ass phase that it is in now.  

[psuhoffman]

@stormy it has rarely been cold enough to snow the last 5 years.  Yes it gets cold enough once in a while each winter...but when you spend 80% of the winter "too warm" that is not bad luck when most storms come along when it's too warm...that is just math. 

[psuhoffman]

4 hours ago, WEATHER53 said:

Most index assertions that have blossomed in the last 10 years are theoretical 

Why has the mid Atlantic switched from about 50% As and 50%Bs to about 80% Bs.

 

What criteria do you use to identify analogs for your seasonal forecast?  

[WEATHER53]

21 hours ago, psuhoffman said:

What criteria do you use to identify analogs for your seasonal forecast?  

You must stop responding to every single post I make, 95% of which are not directed to you

its stalking

its creepy

its not going to drive me away 

You are out of line. I have asked for you to cut it out and now I’m telling You-Cut it out.

Moderators-the history is right in  this thread with support from many other threads. This guy is obsessive. Take care of it please and thank you 

[Ji]

1 hour ago, WEATHER53 said:

You must stop responding to every single post I make, 95% of which are not directed to you

its stalking

its creepy

its not going to drive me away 

You are out of line. I have asked for you to cut it out and now I’m telling You-Cut it out.

Moderators-the history is right in  this thread with support from many other threads. This guy is obsessive. Take care of it please and thank you 

dont show Howard this chart

[pazzo83]

4 hours ago, WEATHER53 said:

You must stop responding to every single post I make, 95% of which are not directed to you

its stalking

its creepy

its not going to drive me away 

You are out of line. I have asked for you to cut it out and now I’m telling You-Cut it out.

Moderators-the history is right in  this thread with support from many other threads. This guy is obsessive. Take care of it please and thank you 

yeah pretty typical response here when you are completely unable to answer the question.

[psuhoffman]

4 hours ago, WEATHER53 said:

You must stop responding to every single post I make, 95% of which are not directed to you

its stalking

its creepy

its not going to drive me away 

You are out of line. I have asked for you to cut it out and now I’m telling You-Cut it out.

Moderators-the history is right in  this thread with support from many other threads. This guy is obsessive. Take care of it please and thank you 

How is asking you a simple question “trying to drive you away”?  You’ve criticized the use of certain indices commonly used to identify analogs. So isn’t it a fair question then how you identify analogs?  You say you want non model focused discussions but 3 times recently when I attempted to engage with you in exactly that way by asking a benign question, you got weirdly hostile.  
 

As for stalking…