• Hello, please take a minute to check out our awesome content, contributed by the wonderful members of our community. We hope you'll add your own thoughts and opinions by making a free account!

Wintry Winter 2018-19 Discussion

Yep DEeeeeTeeeee is all in
af0673a14b31974bbe200237d231a847.png
3af1466e991bd3439dd5f5050fd1fbc7.png



Sent from my iPhone using Tapatalk
 
Must be missing something on which is better for snow, but for Raleigh mod and strong Nino's avg the most. Yes, strong nina is actually the highest of all of them due to 99/00 and small sample size but who is wishing for a strong nina for our best snow chances.

Average all nino's, RDU averages 7.2" snow with a median of 7.4. Average all ninas for RDU and it's 7.0" with a median of 5.9".

Used these years. I didn't get into warm/cool neutrals just straight declared enso. This goes back 70 years which is a big enough sample size.
https://ggweather.com/enso/oni.htm

Averages:
Weak Nino: 6.48"
Mod Nino: 7.72"
Strong Nino: 7.88"
Weak Nina: 6.83"
Mod Nina: 5.83"
Strong Nina: 8"
 
Last edited:
Kylo said:
Must be missing something on which is better for snow, but for Raleigh mod and strong Nino's avg the most. Yes, strong nina is actually the highest of all of them due to 99/00 and small sample size but who is wishing for a strong nina for our best snow chances.

Average all nino's, RDU averages 7.2" snow with a median of 7.4. Average all ninas for RDU and it's 7.0" with a median of 5.9".

Used these years. I didn't get into warm/cool neutrals just straight declared enso. This goes back 70 years which is a big enough sample size.
https://ggweather.com/enso/oni.htm

Averages:
Weak Nino: 6.48"
Mod Nino: 7.72"
Strong Nino: 7.88"
Weak Nina: 6.83"
Mod Nina: 5.83"
Strong Nina: 8"
Click to expand...
Interesting. This is what I got for Raleigh
0f79324fb5da9ce0ea46c8ea53c8506e.jpg



Sent from my iPhone using Tapatalk
 
Don’t know if this means anything but the deer are going crazy here in middle Georgia. Their moving all over the place the bucks are out more than they’ve ever been this season. Maybe their sensing they need to gain more fat for a rough winter?
 
Round Oak Weather- Zeb said:
Don’t know if this means anything but the deer are going crazy here in middle Georgia. Their moving all over the place the bucks are out more than they’ve ever been this season. Maybe their sensing they need to gain more fat for a rough winter?
Click to expand...
My corn must be working... I’m ready!
 
Kylo said:
Must be missing something on which is better for snow, but for Raleigh mod and strong Nino's avg the most. Yes, strong nina is actually the highest of all of them due to 99/00 and small sample size but who is wishing for a strong nina for our best snow chances.

Average all nino's, RDU averages 7.2" snow with a median of 7.4. Average all ninas for RDU and it's 7.0" with a median of 5.9".

Used these years. I didn't get into warm/cool neutrals just straight declared enso. This goes back 70 years which is a big enough sample size.
https://ggweather.com/enso/oni.htm

Averages:
Weak Nino: 6.48"
Mod Nino: 7.72"
Strong Nino: 7.88"
Weak Nina: 6.83"
Mod Nina: 5.83"
Strong Nina: 8"
Click to expand...

70 years is nowhere near enough samples when you split them up into several bins, sample size of around 30 for each ENSO bin is what you ideally want to shoot for. You have to use pre-1950 ENSO data like Jon did to get anywhere close to that. I have found that the state as a whole averages slightly more snow in El Ninos than La Ninas and that the most detectable and likely significant signal in terms of ENSO-related snowfall is an increase over climatologically favored areas of the western and northwestern piedmont as well as the mountains during El Ninos while there's little significant change in areas like RDU. Obviously when you look at this spatially, this also means RDU and the surrounding eastern piedmont tends to do better relative to those climatologically favored areas in La Ninas but still tends to receive generally similar amounts of snow.
https://www.webberweather.com/ensemble-oceanic-nino-index.html
 
Does anyone know what the ENSO state back in December of ‘63 was? Doing some reading this morning and see Memphis had one hell of a pre Christmas winter storm that year.
 
Webberweather53 said:
70 years is nowhere near enough samples when you split them up into several bins, sample size of around 30 for each ENSO bin is what you ideally want to shoot for. You have to use pre-1950 ENSO data like Jon did to get anywhere close to that. I have found that the state as a whole averages slightly more snow in El Ninos than La Ninas and that the most detectable and likely significant signal in terms of ENSO-related snowfall is an increase over climatologically favored areas of the western and northwestern piedmont as well as the mountains during El Ninos while there's little significant change in areas like RDU. Obviously when you look at this spatially, this also means RDU and the surrounding eastern piedmont tends to do better relative to those climatologically favored areas in La Ninas but still tends to receive generally similar amounts of snow.
https://www.webberweather.com/ensemble-oceanic-nino-index.html
Click to expand...

Agree on sample size, why I think the median should be considered. Post 1950, Nino's has a higher median, more 7"+ seasons and more 10" + seasons.. The numbers Jon had for the super/strong nina's are skewed by 2 winters. The average is the highest but the median is 3". I would rather have blocking over choice of enso state. Either way we need to be lucky too.
 
Kylo said:
Agree on sample size, why I think the median should be considered. Post 1950, Nino's has a higher median, more 7"+ seasons and more 10" + seasons.. The numbers Jon had for the super/strong nina's are skewed by 2 winters. The average is the highest but the median is 3". I would rather have blocking over choice of enso state. Either way we need to be lucky too.
Click to expand...

I see your point but it is also a bit flawed (there’s no perfect way to do this with such a small sample size)... Why are you doing post 1950 when data to the 1800s is available? Post 1950 would get rid of a lot of La Niña data which have several large snows and a lot of data in general, which would help your chart.

Median could help to get realistic real world values so I more or less agree there, but these sample sizes have many years that have 0 or Trace snowfall years. So median would artificially skew toward 0...and while it’s more realistic (as we are closer to 0 given any year) it’s still flawed.


When assessing whether or not larger snows are more common, as you mention 7” and 10”+ years are in Niños, you’re also running into a sample size issue...as large snow years are uncommon at RDU.

For instance, since 1889, all of Raleigh’s 10”+ snowfall years, are as follows:

Strong and Super Nino: 4
Moderate Nino: 2
Weak Nino: 2
Strong and Super Niña: 3
Moderate Niña: 2
Weak Niña: 5

So by this logic you would still rank Weak Nina’s at top for large snowfall years, but obviously you see the problem here....sample size is about as low as it gets. 5 years out of 129 years isn’t exactly any different than 2 years out of 129 years. So your conclusion on large snowfall years would tell you all years have equal chance, and that’s because of rarity.



Sent from my iPhone using Tapatalk
 
Last edited:
Kylo said:
Agree on sample size, why I think the median should be considered. Post 1950, Nino's has a higher median, more 7"+ seasons and more 10" + seasons.. The numbers Jon had for the super/strong nina's are skewed by 2 winters. The average is the highest but the median is 3". I would rather have blocking over choice of enso state. Either way we need to be lucky too.
Click to expand...
If GCMs were of sufficient spatial and temporal resolution to resolve mesoscale snow features we probably could get a good idea of what the response from ENSO is right over RDU. I think either way that it’s inconclusive but the data shows percentile wise relative to the rest of central and eastern NC individual storms and winters during La Niña or cold neutral ENSO tend to bullseye the Triangle area. However, in El Ninos, there’s a discernible signal for big snowfall gradients between the Triad and Triangle because we get more & stronger storms that dump more precipitation over the area while also pushing the rain-snow line a little further inland. I expect to see the latter more than likely pan out this winter overall but it’s far from guaranteed
 
  • Like
Reactions: Jon
Webberweather53 said:
If GCMs were of sufficient spatial and temporal resolution to resolve mesoscale snow features we probably could get a good idea of what the response from ENSO is right over RDU. I think either way that it’s inconclusive but the data shows percentile wise relative to the rest of central and eastern NC individual storms and winters during La Niña or cold neutral ENSO tend to bullseye the Triangle area. However, in El Ninos, there’s a discernible signal for big snowfall gradients between the Triad and Triangle because we get more & stronger storms that dump more precipitation over the area while also pushing the rain-snow line a little further inland. I expect to see the latter more than likely pan out this winter overall but it’s far from guaranteed
Click to expand...

I bet if I crunched the same numbers for GSO we’d see the Niños pop to the top. I thought about doing it, but haven’t had the time. That’s the way it goes though lol. So many ideas but no time to do it!

I’m hopeful for this season especially for February. But we have to remind ourselves how rare big snows are, small and trace events are our norm by a long shot.

I did find that in Weak Nino years (n=15), February has by far the most 1”+ snowfalls and winters in these years seem backloaded as it also beats out Jan. I think with moderate, the events were more uniform for Jan and Feb.


Sent from my iPhone using Tapatalk
 
Jon said:
I see your point but it is also a bit flawed (there’s no perfect way to do this with such a small sample size)... Why are you doing post 1950 when data to the 1800s is available? Post 1950 would get rid of a lot of La Niña data which have several large snows and a lot of data in general, which would help your chart.

Median could help to get realistic real world values so I more or less agree there, but these sample sizes have many years that have 0 or Trace snowfall years. So median would artificially skew toward 0...and while it’s more realistic (as we are closer to 0 given any year) it’s still flawed.


When assessing whether or not larger snows are more common, as you mention 7” and 10”+ years are in Niños, you’re also running into a sample size issue...as large snow years are uncommon at RDU.

For instance, since 1889, all of Raleigh’s 10”+ snowfall years, are as follows:

Strong and Super Nino: 4
Moderate Nino: 2
Weak Nino: 2
Strong and Super Niña: 3
Moderate Niña: 2
Weak Niña: 5

So by this logic you would still rank Weak Nina’s at top for large snowfall years, but obviously you see the problem here....sample size is about as low as it gets. 5 years out of 129 years isn’t exactly any different than 2 years out of 129 years. So your conclusion on large snowfall years would tell you all years have equal chance, and that’s because of rarity.



Sent from my iPhone using Tapatalk
Click to expand...

I assumed 70 years is a reasonable length period and it has most readily available data. Curious that post 1950 leans toward nino but the pre 1950 looks to lean decidedly towards Niña based on your stats. Have to wonder why.

If we delved deeper into this I would imagine there’s more to just ENSO state that drives snowfall for RDU. Great research though, always enjoy weather stats.
 
Kylo said:
I assumed 70 years is a reasonable length period and it has most readily available data. Curious that post 1950 leans toward nino but the pre 1950 looks to lean decidedly towards Niña based on your stats. Have to wonder why.

If we delved deeper into this I would imagine there’s more to just ENSO state that drives snowfall for RDU. Great research though, always enjoy weather stats.
Click to expand...

Yeah winter forecasting is insanely tough for that reason, ENSO is only a small piece of the puzzle. Even when controlling for dozens of other factors it’s still hard to nail seasonal forecasts , but I’d say ENSO is a very large puzzle piece.


Sent from my iPhone using Tapatalk
 
Great seeing all the mets going with above average snow forecasts here.
 
Jon said:
I bet if I crunched the same numbers for GSO we’d see the Niños pop to the top. I thought about doing it, but haven’t had the time. That’s the way it goes though lol. So many ideas but no time to do it!

I’m hopeful for this season especially for February. But we have to remind ourselves how rare big snows are, small and trace events are our norm by a long shot.

I did find that in Weak Nino years (n=15), February has by far the most 1”+ snowfalls and winters in these years seem backloaded as it also beats out Jan. I think with moderate, the events were more uniform for Jan and Feb.


Sent from my iPhone using Tapatalk
Click to expand...
If you do decide to look into Greensboro’s snow data just a fair warning that a lot of it is missing before 1930! Thankfully I was able to either find the original publications or provide reasonable estimates and I’d be more than happy to fork that over because NWS NOWData is missing a number of storms between 1903-1930. I was able to extend Greensboro’s (& everyone else’s) record back to 1895-96 based on the aforementioned original publications from NCEI
 
Anybody concerned abou the rapidly rising QBO screwing everything up?
 
You must log in or register to reply here.
Back
Top