The Seasonal Snowfall Futility Markers

[donsutherland1]

22 minutes ago, WinterFire said:

@donsutherland1 did this analysis for DCA recently in another forum. I won’t repost without permission but it presented a very clear decline in seasonal snowfall combined with a stark increase in average winter temp and shorter cold season length. Perhaps he’d be willing to re-share here for reference?

Here it is:

[cbmclean]

22 minutes ago, donsutherland1 said:

Here it is:

Thanks for the analysis Don.  

[DarkSharkWX]

id be interested to see this for IAD or Damascus(even though there isn't much data for Damascus). i'd expect the same downward trend, but maybe its decreasing less/more, etc.

[Terpeast]

3 minutes ago, DarkSharkWX said:

id be interested to see this for IAD or Damascus(even though there isn't much data for Damascus). i'd expect the same downward trend, but maybe its decreasing less/more, etc.

Stats from Ocean City MD or coastal delaware would also be interesting to see. They’ve done a lot better lately. 

[cbmclean]

1 hour ago, donsutherland1 said:

Here it is:

By the way, what is the definition of "Season Length"?  First frost to last frost?

[donsutherland1]

50 minutes ago, cbmclean said:

By the way, what is the definition of "Season Length"?  First frost to last frost?

First measurable snow to last measurable snow (snow season).

[cbmclean]

I guess the idea to move statistics discussion here didn't work out.  Stats haters can't blame me.

[psuhoffman]

1 hour ago, WesternFringe said:

@psuhoffman

When we grade students (you are in education like me, right?), do we use the median or the mean?  Why?  Do we use GPA (Grade Point Average) or GMA (Grade Median Average)?  Why?  We use the mean because it better represents all of a student's scores, not just the one grade in the middle.  Now, I am fully aware that outliers can skew the mean, but with an n of 136, that isn't an issue.  We worry about outliers skewing the mean when we have a low n (< 18), not with an n over 100.  This is why educators are usually required to have a certain minimum threshold for number of grades in determining a student's final grade.  I know no statistician that would be worried about skewing the mean with an n of 136, especially when none of the numbers in the data set are more than 4 times the mean.  If we had outliers of 200" of snow in a year when the mean is 15", then maybe.  Don't we wish.  But these annual snowfall outliers count, and should be represented in the data.  By using the median, one essentially takes all of the best years out of the data and reduces the n precipitously. This is a bad idea, especially with snow! lol

I am not saying that snowfall isn't decreasing for DC, but I am saying it has not been a drastic decrease (especially in our lifetimes).  This is the data set I used:

https://www.weather.gov/media/lwx/climate/dcasnow.pdf

I ran another linear regression starting with 1942, as you did.  The slope of that line is -0.058" and the standard error is -0.054".  The R squared is .01, which means 99% of the variability from year to year is random and not do to the independent variable, which in this case is time (presumably the climate changing over time).  

So, yes, DC is seeing .058" inches less of snow per year since 1942.  Most people wouldn't be able to notice the difference of .058" of snow per year.  If you can, you have superhuman perception.  99% (R squared) of the change in annual snowfall from 1942 is explained by randomness, and not due to the passage of time.  

The slope of the line goes down and up depending on the starting point, of course.  When I choose different starting points closer to present (to match the ages of folks in here and thus their winter memories), the slope of the line can be literally almost 0, or can actually be positive.  

In summary, I concede that the slope of the line is negative, and always have in my posts.  However, I also have learned that the human brain constantly looks for patterns and trends, even when they are non-existent or super small.  That is why I like stats and means and cold hard numbers.  Hats off to you sir if you can notice a decline of hundredths of an inch per year where 99% of the change is due to randomness, or statistical noise. 

No one can tell the future, but if I had to guess we have some monster years coming up in the El Nino years which will make the slope flatter and closer to zero. The sky isn't falling, we can still do snow, and good years lie ahead.  When 99% of the variability is random statistical noise, this tells me we are not in some inevitable downward spiral regarding snowfall.

Again, I stick by my conclusion that although annual DC snowfall is falling (no pun intended), the change is very small, not cataclysmic, and 99% due to randomness.    

Moving this here so as not to clutter up the main thread any further.  

Counter points...  

1) Using grades might not be the best example since my district enforces a minimum grade of 50 so as not to skew students grades with 0s lol.  But lets not get into that please.  Thought I would start with some levity. 

2) The claim that outliers don't skew the mean is predicated on the mean of the whole dataset.  But we are comparing the change in mean over time.  To do that you have to chunk the mean into smaller portions.  Climo uses a 30 year mean.  If we remove just 4 years out of those 30 the current mean for DCA is only 9".  Those 4 years are most definitely skewing the mean of the current 30 year period...and by doing that is also skewing the slope of the trendline.  

3) This depends what you are trying to tease out of the data.  The mean tells you over a period of time how much total snowfall is likely to fall proportional to each year.  But it does not tell you how much snow is most likely to fall in any given year.  The truth is over the last 30 years that 9" number (what you get if you remove the 4 super outliers) is way closer to what you should expect going into any given year than the 14" mean number with which was only reached 26% of the time.  In any given season there was a 74% chance you were getting less snow than that.  If you are trying to calculate how much snow you should expect in any given season, the mean is not the best tool.

4)  Maybe we are trying to get at different things.  What I am getting at are in any given season the odds of getting a specific amount of snow are dropping.  The odds of getting 8", 10", 12", 15" and so on...going into any season are going down.  And going down a LOT.  The odds of getting a 4", 6", 8" storm are all dropping.  That is what I am focused on.  That going into any specific season the chances of it being crap are going up.  The mean does not capture that fact as well.  

5) You keep focusing on what the change is per year.  Over time very small changes add up to be significant.

6) In a vacuum its correct to say 99% is attributed to random chance because the math you are using doesn't know any corroborating evidence.  But we do know some additional facts.  The fact that it is getting warmer for instance.  If you apply logic to the fact that at the same time the slope of snowfall is decreasing correlates to an increase in temperature the amount we should attribute to "random chance" goes down.  

7) Lastly I am not making this about human caused climate change.  Maybe this is cyclical.  Maybe its not human caused.  Maybe at some unknown point in the future the trend reverses.  Those are all true.  But that is not what I am getting at.  I am not living at some random unknown point in the future.  I don't care what the chances of snow will be is 2323 I care what the chances of snow are right now in 2022.  For the sake of this argument I don't care if the warming is man made or natural it is still impacting my snowfall right now just the same.  And since we don't know when or if the trend will reverse I don't even care about that.  What my analysis was showing is that our chances of snow in any given season right now are significantly lower than they were during recent recorded history.  It's not even a predictive thing.  I am not making a statement about what snow probabilities will be 50 years from now.  I have no idea.  I am making a statement on what they are RIGHT NOW. 

 

Your analysis isn't wrong.  Everything you say is 100% true.  But you are focusing on different metrics that me and teasing different things out of the data.  You seem to be more focused on predictive measures of snowfall over any long period of time with statistical certainty.   That is a very different concept than trying to calculate what the probability of a given amount of snow in one specific season is right now and how that probability has changed over the last 100 years.  We are looking at very very different things.  

[WinterFire]

11 minutes ago, psuhoffman said:

5) You keep focusing on what the change is per year.  Over time very small changes add up to be significant.

 

This part. I’m not sure if you’re meaning to imply this, @WesternFringe, but focusing on an annual average change of -.06” makes it seem like your argument is that DC winters have only seen a decrease of .06” from 1941 to today. But that’s not what a slope of -.06 implies—rather, it’s indicating that the annual snowfall from 1941 has decreased by a total of something like 4.5”. Which is noticeable, I’d argue. So yes, between last year and this year, most people would not notice a decrease of .06”. But over the last 30 years, we do notice a decrease of nearly 2” (which incidentally matches relatively closely with the 30 year average in the 00s of 15.4”/year compared to the 20s average of 13.9”/year). 

[WesternFringe]

1 hour ago, WinterFire said:

This part. I’m not sure if you’re meaning to imply this, @WesternFringe, but focusing on an annual average change of -.06” makes it seem like your argument is that DC winters have only seen a decrease of .06” from 1941 to today. But that’s not what a slope of -.06 implies—rather, it’s indicating that the annual snowfall from 1941 has decreased by a total of something like 4.5”. Which is noticeable, I’d argue. So yes, between last year and this year, most people would not notice a decrease of .06”. But over the last 30 years, we do notice a decrease of nearly 2” (which incidentally matches relatively closely with the 30 year average in the 00s of 15.4”/year compared to the 20s average of 13.9”/year). 

@psuhoffman

I am not meaning to imply that the change has only been .06” since 1942.  I know what the slope means, and I think most folks, including PSU, know what I meant.  I do think it is difficult for most people (even snow weenies like us) to notice a 1.5” change in annual snow over a 30 year period.  I barely notice a 1.5” change between 2 consecutive years.

PSU, you make valid points, and everything you say is true as well.  Indeed, we are looking at very different metrics of the same data.  I don’t care what the snowfall is in 100 years from now either.  I do care about looking at the data and yes the cycles and using it to form what my best guess is for the next 40 years.  DC is definitely in a bad cycle overall in the latest couple of years.  Like I said, I had 28.5” last year, which is about 4 inches above my mean.  Therefore, I have a different perspective than a lot of the DC centered folks in here.

Our strongest area of agreement is when you conceded that it all might be cyclical.   Looking at the R squared, I think it is and that the data show it.  I don’t agree that we can discount the R squared because we know there is a (small) warming trend.  The R squared should respond accordingly if this were playing a huge role in drastically cutting our snowfall totals.

Admittedly, some of the other time periods I ran a regression for had the R squared at more like 94%, meaning 6% was not due to random noise.  That’s actually a decent signal that something other than natural variability is at play and that warming plays a part.

Our approaches to the numbers and our desires for future snowfall are different.  You seem to be more focused on the chances for any given year of getting x amount of snow.  I am definitely more focused on how many inches total we get in the next 10, 20, and up to 40 years (if I live that long!).  I could care less if it happens in big bunches or whether it is spread evenly amongst the years.  In fact, I would probably vote for larger events and bigger years mixed in with dismal years than getting 15” every year like clockwork.  Different strokes for different folks.

ETA: Trust me though, I will be watching the R squared like a hawk.  If DC continues to have 5 year runs with relatively little snow without bonanza years mixed in, the R squared will rise quickly and I will be worried. Or throw in the towel or move north to New England or Upstate NY where I grew up!  Lol 

Imo, I just don’t think the data warrant any panicking yet.  I think it is primarily cycles and noise at this point in time.

 

 

[WesternFringe]

There is a psychological phenomenon known as the primacy and recency effect.  People tend to remember the beginning and end of a list of numbers better than the numbers in the middle.  Same with words. 

It also applies to memories.  People tend to remember their childhoods and early memories better than what comes in between.  They tend to remember the last few years better than what is in the middle, too.

A bunch of people in here seem to remember childhood years full of snow and then remember/note the last 5 or so have sucked for their location.  I think this creates misinterpretations and develops biases.

For me, looking at means over long periods of time helps to balance out and distinguish my perception from reality.

[pazzo83]

4 hours ago, WesternFringe said:

@psuhoffman

I am not meaning to imply that the change has only been .06” since 1942.  I know what the slope means, and I think most folks, including PSU, know what I meant.  I do think it is difficult for most people (even snow weenies like us) to notice a 1.5” change in annual snow over a 30 year period.  I barely notice a 1.5” change between 2 consecutive years.

PSU, you make valid points, and everything you say is true as well.  Indeed, we are looking at very different metrics of the same data.  I don’t care what the snowfall is in 100 years from now either.  I do care about looking at the data and yes the cycles and using it to form what my best guess is for the next 40 years.  DC is definitely in a bad cycle overall in the latest couple of years.  Like I said, I had 28.5” last year, which is about 4 inches above my mean.  Therefore, I have a different perspective than a lot of the DC centered folks in here.

Our strongest area of agreement is when you conceded that it all might be cyclical.   Looking at the R squared, I think it is and that the data show it.  I don’t agree that we can discount the R squared because we know there is a (small) warming trend.  The R squared should respond accordingly if this were playing a huge role in drastically cutting our snowfall totals.

Admittedly, some of the other time periods I ran a regression for had the R squared at more like 94%, meaning 6% was not due to random noise.  That’s actually a decent signal that something other than natural variability is at play and that warming plays a part.

Our approaches to the numbers and our desires for future snowfall are different.  You seem to be more focused on the chances for any given year of getting x amount of snow.  I am definitely more focused on how many inches total we get in the next 10, 20, and up to 40 years (if I live that long!).  I could care less if it happens in big bunches or whether it is spread evenly amongst the years.  In fact, I would probably vote for larger events and bigger years mixed in with dismal years than getting 15” every year like clockwork.  Different strokes for different folks.

ETA: Trust me though, I will be watching the R squared like a hawk.  If DC continues to have 5 year runs with relatively little snow without bonanza years mixed in, the R squared will rise quickly and I will be worried. Or throw in the towel or move north to New England or Upstate NY where I grew up!  Lol 

Imo, I just don’t think the data warrant any panicking yet.  I think it is primarily cycles and noise at this point in time.

 

 

isn't your independent variable here time, so that is what you are using to try to explain the variance in your observed data (snowfall)?  Let's ignore the fact that running a simple linear regression on a time series, especially when there is a cyclical component here (thus the observations are not all IID), is problematic.  But if you are trying to see if a warming base state would be used, you'd think you'd include something reflecting that in your analysis (and not just time).

[cbmclean]

1 hour ago, pazzo83 said:

isn't your independent variable here time, so that is what you are using to try to explain the variance in your observed data (snowfall)?  Let's ignore the fact that running a simple linear regression on a time series, especially when there is a cyclical component here (thus the observations are not all IID), is problematic.  But if you are trying to see if a warming base state would be used, you'd think you'd include something reflecting that in your analysis (and not just time).

Starting off with a linear regression is not unreasonable even if we suspect that the observations are not IID.  I'm not completely sold on the existence of cycles, at least on a decadal scale.  Purely random data can show lots of clumps suggestive of patters which are not really there.  It would be very interesting to put the data through some sort of Fourier analysis to see if there any real periodicity, but I don't have the tools for that.

[WesternFringe]

2 hours ago, pazzo83 said:

isn't your independent variable here time, so that is what you are using to try to explain the variance in your observed data (snowfall)?  Let's ignore the fact that running a simple linear regression on a time series, especially when there is a cyclical component here (thus the observations are not all IID), is problematic.  But if you are trying to see if a warming base state would be used, you'd think you'd include something reflecting that in your analysis (and not just time).

If it were warming quickly or significantly enough to affect the snowfall totals (dependent variable) over time (independent variable), then it would show up as a part of the R squared 

[fujiwara79]

1 hour ago, WesternFringe said:

If it were warming quickly or significantly enough to affect the snowfall totals (dependent variable) over time (independent variable), then it would show up as a part of the R squared 

The 30-year snowfall average at DCA has decreased by 20% since 1941.  But there has also been a significant increase in the 30-year wintertime precipitation average.  A decrease in wintertime snow, juxtaposed with an increase in wintertime precipitation, is strong evidence that there probably are more "33 and rain" events lately - and that it's not just our imagination.

Do us a favor and check what percentage of each winter's (for simplicity, use DJF) precipitation falls as frozen, and show us how that has varied over time.  You'll find that the percentage of precipitation that falls as frozen has markedly decreased.  Bigly.   

[WesternFringe]

10 hours ago, fujiwara79 said:

The 30-year snowfall average at DCA has decreased by 20% since 1941.  But there has also been a significant increase in the 30-year wintertime precipitation average.  A decrease in wintertime snow, juxtaposed with an increase in wintertime precipitation, is strong evidence that there probably are more "33 and rain" events lately - and that it's not just our imagination.

Do us a favor and check what percentage of each winter's (for simplicity, use DJF) precipitation falls as frozen, and show us how that has varied over time.  You'll find that the percentage of precipitation that falls as frozen has markedly decreased.  Bigly.   

It sounds like you have already done this analysis since you have come to a conclusion, so why would you ask me to ‘do us a favor’ and do it?

Either you have done it already and it would be wasting my time, or you are talking smack like you know the results without having done the analysis.  Which is it?

[87storms]

I think the median is important for the sake of trying to find out what the 50th percentile is.  If you like snow and don't need the big dog storms, then you'd obviously want a higher median.  Just running a box and whisker plot on the DCA data since 1887 (I know, it's not ideal) shows the IQR to be from 8" to 24".

 

[87storms]

On 12/28/2022 at 8:59 PM, donsutherland1 said:

Here it is:

This is interesting, but I do think we're too cyclical of a snow town for 30 year averages.  I grew up in the 80s which definitely jaded my view, but it had really solid winters and made me think snow was typical lol.  Chopping off the 70s and 80s is really going to make it look worse than it is, though there are clear trends in the wrong direction.  Just going by the eye test, losing the small/minor events is probably the biggest issue we've been facing.

[WesternFringe]

30 minutes ago, 87storms said:

I think the median is important for the sake of trying to find out what the 50th percentile is.  If you like snow and don't need the big dog storms, then you'd obviously want a higher median.  Just running a box and whisker plot on the DCA data since 1887 (I know, it's not ideal) shows the IQR to be from 8" to 24".

 

So, 50% of years DC gets 8” to 24”, 25% of years are more and 25% less.  I hear you that box and whisker plots have limited usefulness, but it is still informative to look at the data in as many ways as possible.

[WesternFringe]

13 hours ago, fujiwara79 said:

The 30-year snowfall average at DCA has decreased by 20% since 1941.  But there has also been a significant increase in the 30-year wintertime precipitation average.  A decrease in wintertime snow, juxtaposed with an increase in wintertime precipitation, is strong evidence that there probably are more "33 and rain" events lately - and that it's not just our imagination.

Do us a favor and check what percentage of each winter's (for simplicity, use DJF) precipitation falls as frozen, and show us how that has varied over time.  You'll find that the percentage of precipitation that falls as frozen has markedly decreased.  Bigly.   

Here is precipitation for DC by month and by season and by year since the 1870s.  I don’t have time to run the numbers right now (especially if you already have), but a quick look tells me there is no obvious increasing trend for winter precip in DC.  If anything, it looks to me like there used to be more in the late 1800s and throughout much of the 1900s when compared to the last 30 years.

I will run the numbers later, but again, if you’ve done this analysis already, by all means share it with us.

https://www.weather.gov/media/lwx/climate/dcaprecip.pdf

eta: in fact, I would be willing to bet good money that DC winter precip and snowfall are highly and positively correlated, rather than negatively correlated as you suggest.

[George BM]

@psuhoffman

Don't know whether you saw this yet... but there's something potentially "eye-opening?" I saw on twitter (yes I know don't shoot me yet) about mid-latitude (30-60N) SSTs over the last nearly century and a half for Nov-Dec.

[North Balti Zen]

Bumping to this page to keep an eye on the futility marks (as this is becoming potentially more and more relevant)

Quote

 

BWI:

.7 - '49/'50

1.2 - '72/'73

1.8 - 2011/12

1.8 - 2019/20

2.3 - 2001/02

DCA

.1 - '97/'98

.1 - '72/'73

.6 - 2019/20

2.0 - 2011/12

2.2 - '75/'76

Dulles:

2.2 - '72/'73

2.6 - 2001/02

2.9 - 2019/20

3.7 - 2011/12

4.4 - '80/'81

 

 

[North Balti Zen]

That's 2 absolute dead ratters in the last 10 years - if we continue to take this header into the non-snow frozen ground - would be 3 dead ratters in 11 years. And a few notably fallow years just off this list as well. 

[George BM]

18 minutes ago, psuhoffman said:

But seriously the 12z euro had a similar thing. Perfect SW pass, sub 540, all rain from what “should” have been a 1-3” snow.  Those things add up and I’m noting them way too frequently. 

There is always “higher” pressure to our east as a wave approaches. Calling that a high pressure is a stretch and if a very brief 5mph regional south wind (it’s not like it’s a screaming fetch from the gulf) is too much in January with sub 540 thicknesses and a track to our south please explain to me what the right setup is for what used to be a typical small to moderate snow other than the obvious “we need arctic air” which is and always has been an exceptionally rare event. 

What I was referencing earlier.

 

[Terpeast]

8 minutes ago, George BM said:

What I was referencing earlier.

 

I’ve been wondering this but did not want to say anything until now. 

Maybe unscientific conjecture… but if mid-latitude SSTs warm faster than tropical SSTs, are we going to be in a permanent “La Nina” background state, even as tropical SSTs are just as warm as historical neutral or even El Ninos from way in the past? 

Scary thought. 

[cbmclean]

1 hour ago, Terpeast said:

I’ve been wondering this but did not want to say anything until now. 

Maybe unscientific conjecture… but if mid-latitude SSTs warm faster than tropical SSTs, are we going to be in a permanent “La Nina” background state, even as tropical SSTs are just as warm as historical neutral or even El Ninos from way in the past? 

Scary thought. 

From the Wikipedia article on the Pleistocene (I know it's Wikipedia but thought it was interesting nonetheless):

According to Mark Lynas (through collected data), the Pleistocene's overall climate could be characterised as a continuous El Niño with trade winds in the south Pacific weakening or heading east, warm air rising near Peru, warm water spreading from the west Pacific and the Indian Ocean to the east Pacific, and other El Niño markers.

So if the ice age was a continuous El Nino the new warm future will be...?

[psuhoffman]

12 minutes ago, Its a Breeze said:

What we want doesn't match the geography of our area...

99% of our storms are crazy rare things happening at the right time.

for 60 years DC averaged 4 snowfalls a season of 1" or greater.  For the past 20 years that has dropped to 2.9 and over the last 10 it has dropped to 2.4.  Furthermore, its worse when you dig into that.  It used to be rare to get a season with less than 3 snowfalls.  For a long time 3-5 snowfalls per year was the common and anything above and below that was rare.  Recently we still get the rare season with 6 or 7 snowfalls but what is happening is in the rest of the seasons suddenly getting only 1 or 2 has become common.  What used to be rare, getting less than 3 snowfalls in a whole year, now is happening regularly. 

 

It was never easy to get snow.  But it was easier.  And I have noted recently that we seem to need increasingly anomalous events (crazy combinations of arctic air with uncommon storm tracks, ridiculous wave spacing, perfect timing) the kinds of things that just aren't ever going to be a repeatable staple to get reliable snow regularly.  The most common way to get a snowfall in prior eras was simple...get a wave to slide under us and have enough cold air in place.  Lately...unless we get a string of anomalous events the cold part is simply missing.  My point was I would like to see us get snow in a way that is repeatable.  Simple.  Easy.  Anything that takes a congruence of several unlikely events all going our way to get snow is unlikely to repeat very often and doesn't really give hope we have broken out of this funk we are in.  

[psuhoffman]

On 1/2/2023 at 8:29 PM, Terpeast said:

I’ve been wondering this but did not want to say anything until now. 

Maybe unscientific conjecture… but if mid-latitude SSTs warm faster than tropical SSTs, are we going to be in a permanent “La Nina” background state, even as tropical SSTs are just as warm as historical neutral or even El Ninos from way in the past? 

Scary thought. 

 

19 minutes ago, WxUSAF said:

A strong 50-50 low providing timely confluence and cold air is the most obvious path to a majority frozen event for us. I think 6z gfs teased another path, but one with a lot more difficulty. Guidance has waffled on whether any northern stream energy phases into our big southern stream shortwave. 6z gfs phases it in somewhat late and in a sloppy manner. IF you can get that phase timed just right with just the right low location, it could supply some cold air. But that’s obviously a Hail Mary sort of option. 

 

1 hour ago, CAPE said:

Oh you mean this isn't good?

 

1 hour ago, Weather Will said:

It’s too soon to give up.  6Z GEFS is an improvement over 0Z.

I had wanted to try to avoid this general topic completely but its become obvious that is impossible...but I will keep my part of this discussion to this thread so as not to annoy those that would rather not think about this in the main thread.  

@Terpeast I see this as all related.  A lot of this is just conjecture at this time simply because we don't yet have a long enough period of data to prove with statistical significance but some very bright minds along with well designed simulations have theorized that there is a causality between AGW and several of these chronic pacific issues we have been having for a majority of our winter seasons recently from the expansion of the Hadley Cell to some of the persistent SST issues in the tropical Indian and Pacific, to the prevalence of a la nina base state.   Then, add in that on top of the longwave pattern issues created by these problems in the Pac, regardless of that at any given moment in time warm anomalies over the NH land masses are outnumbering cold about 70-30.  So even without those issues simply by pure chance our odds of having cold over us are going to be more difficult.  And lastly...there is conjecture that the SST changes also related to warming in the Atlantic are partly aiding the prevalence of the WAR (which is also a feedback issue from the pac longwave pattern as well!) and why its been difficult to get or sustain a 50/50 feature.  Basically, there are a lot of interrelated factors caused by warming, but every single one of them seems to be the opposite of what we would want.  The only factor related to warming that could work in our favor is the increase in precipitation patterns during the cold season in general.  And I do think we have seen evidence of this in the rare cases where we get a cold season but that benefit only matters when its cold which is becoming increasingly rare.  

@CAPE and @WxUSAF both of your analysis of the issues regarding the specific details we need (50/50 or perfect phasing) are 100% accurate.  My point is just, at some point it feels hopeless if we need some super anomalous string of events to go exactly perfectly to get snow as if we are living in coastal SC or something.   My point the last few days when I voiced my concern with this setup has been the snow solutions seemed to be relying on something I don't see as very likely and even if it did happen is certainly not something we can expect to become a reliable repetitive way to get snow.   Needing some mid latitude system to phase and bomb into a super cyclone and lead to the inception of a TPV in exactly the perfect location at exactly the perfect time is not the way I want to have to roll here.  

Yes a 50/50 is a feature in just about every one of our 20" snowstorms.  Because to get a storm that big in the DC/Balt area we need some pretty major resistance to the WAA necessary.  We are too far south to be obliterated by a CCB like NYC north...the only way we get a HECS here is if a majority of that snow comes from a strong WAA feed ahead of the wave and for that to stay all snow and produce prolific totals we need a 50/50.  But I am not lamenting that this isn't going to be a 20" storm.  What happened to getting a 6" snow from a messy storm where the mountains and get 20" from a setup like this.  When I examined every 4"+ snow at BWI years ago that made up a lot of our events.  Frankly very very very few of our storms were because we lucked into the absolutely perfect conditions, and when that happened thats when we got those 20" storms.  But most of the snow events were messy.  What happened to getting 3" at DCA and 5" at BWI and 6" at IAD from a messy storm where we mix but the big snow is to our NW.  That used to be the typical outcome of a system that tracked to our south but everything wasn't "perfect".  Lately it seems we need perfect to get any snow.  

Basically...the storms like these below...

Hi    Low   QPF    Snowfall

36-30    1.13     4.6

37-30       .77     4.1

45-33      .86      6.9

34-31      .92      3.0

37-33      .92     3.4

39-32      .72      3.0

43-33      1.09    5.5

39-31       1.04     6.9

39-32       .67       3.4

33-30       1.63      6.4

38-33        1.03       3.3

I pulled all of these were from DCA from my past snowfall study.  Look at the temps and precip...these were all obviously mix storms...and I remember the H5 from most of these it was because there was no 50/50 and a marginal "pac puke" as we call it lately, airmass to work with...but they were good tracks during mid winter and so we at least got some snow from them.  And my area did great from all these storms.  I think they were all 12" plus up here.  But now...look at many of the permutations in the 3 ensembles.  They have perfect track systems with rain to Montreal!   Another good example is 1998.  I bring that up because that year is showing up in current analogs a lot.  And yea that year SUCKED for DC snow so its easy to say "well this happened before".  But I was at Penn State that year and we got crushed with snowstorm after snowstorm all winter.  Even my area here had about 20" that season because all those perfect track rainstorms in DC were a messy mix with a few inches of snow up here.  Places with elevation to the NW of the cities actually got quite a bit of snow that winter.  But most of the permutations of this are rain even in those places with a perfect track, and we've had several of these over the last few years.  That is my bigger concern.  We can't seem to get any frozen event, even a flawed mixed type, unless everything is perfect or some incredibly unlikely string of dominoes all fall exactly the right way.  

And lastly, yea the pac is the problem.  But guess what the PAC is what is upstream from us, its HUGE, its the largest heat source on the planet, and other than the rare times when there is cross polar flow (and that isn't even really a good longwave pattern to get a system under us!) we will be dealing with an airmass that has significant influence from the PAC.   The airmass this week isn't pure tropical pac puke.  There is quite a mix of flow from the Yukon area mixed in with mid latitude pacific air.  The problem is its torching even in the Yukon, and the mid latitude pac air is +5 to +10 also!  And both of those facts don't seem to be a right now at this moment bad luck kinda of problem.  They seem to be a permanent base state status quo lately.  

@Weather Will I am not giving up, there is still a path to get snow here.  The phased bomb 50/50 scenario could come back.  Or Wxusaf's scenario of a perfect phase could happen.   But what is depressing to me is how difficult that is and how that is never going to be a reliable way for us to get snow.  Look at the ensembles.  They are very good simulations based on the current conditions.  They are realistic possible permutations.  And the vast majority of the permutations show a rain solution even with a perfect track and even in places to our NW.  Even if we get lucky and we get some super bomb to cause a TPV to form exactly where we need it and this ends up snow...that doesn't change the fact that it was a realistic scenario that a storm could take an absolutely perfect track on what is statistically the coldest week of the year in DC, and it would be nothing but rain all the way to Canada!  That is alarming in a larger 30k foot view kinda way to me.  I am not focused on the specifics of this event, that is a bigger problem imo.  

[psuhoffman]

PS: I just feel like the micro excuses are wearing out at this point.  Its been a string of overly specific reasons/excuses why a pretty good synoptic setup failed.  "December climo sucks", the pacific was bad, no HL blocking, the WAR...but what they all have in common is no cold air.  And frankly we've failed with exact opposite pacific patterns recently so I am tired of the "its the pacific" excuse.  The pacific is always going to be there upstream of us so if we can't snow because the pacific is too warm thats never going away.  

[Maestrobjwa]

Just now, psuhoffman said:

PS: I just feel like the micro excuses are wearing out at this point.  Its been a string of overly specific reasons/excuses why a pretty good synoptic setup failed.  "December climo sucks", the pacific was bad, no HL blocking, the WAR...but what they all have in common is no cold air.  And frankly we've failed with exact opposite pacific patterns recently so I am tired of the "its the pacific" excuse.  The pacific is always going to be there upstream of us so if we can't snow because the pacific is too warm thats never going away.  

What about last Jan? We had cold then, didn't we? And I can remember other cold Januarys since 2016 as well...and a few where the nina just didn't deliver favorable tracks and we were cold enough but missed the precip. So I can see one example (Feb 2020) where you can say warm...(maybe another Feb 2021), but I'm not seeing the warm Januarys you're seeing here. (Again, last year around this time we had a couple threats! And Feb just did what nina Febs do, imo)