Jump to content
  • Member Statistics

    17,608
    Total Members
    7,904
    Most Online
    Vesuvius
    Newest Member
    Vesuvius
    Joined

January 2015 Pattern Discussion


CapturedNature

Recommended Posts

I totally agree, but the -QBO and solar maximum were always two signals complementary to the SAI suggesting a SSW was likely in the second half of winter.

He is saying that the SSW that took place last week should take a couple weeks to propagate and will therefore flip the AO negative by the last week of January. From there, it should stay predominantly negative for the following 4-6 weeks. That's what needs to happen for the SAI to not "fail" as a meaningful prediction tool IMO. The final AO value for DJF is important too, but if it misses the mark because the timing was a couple weeks late and the AO stays negative into March, it doesn't invalidate the research.

In fact, even if we get a raging +AO for the rest of winter, it doesn't mean the premise is faulty...rather, it means that it is an incomplete picture and we don't know what we are missing (but we would have a really good place to start looking). Imagine if Einstein threw out all his theories because someone observed something in the universe that didn't line up perfectly.

 

Well that's what I was saying the other day. If indeed it fails (which I am NOT saying), then it's back to the drawing board. That's how science evolves. 

 

I do agree that if we see a flip in the near term, that would be a good thing for his work, but SSWs are possible regardless of SAI. Perhaps the state of the atmosphere is such that it responds quicker to any tropospheric heat flux thanks to the SAI? Maybe. I also think that if something like a trough in NW Asia can screw up what was an incredible SAI response, perhaps we need to reevaluate a few things regarding the SAI. The atmosphere is one big puzzle. It could mean the SAI is a smaller piece. Or, this could be the one winter out of the next 7 or 8 where the signal was very weak. Sample size as always., is important. 

Link to comment
Share on other sites

  • Replies 1.1k
  • Created
  • Last Reply

Well that's what I was saying the other day. If indeed it fails (which I am NOT saying), then it's back to the drawing board. That's how science evolves. 

 

I do agree that if we see a flip in the near term, that would be a good thing for his work, but SSWs are possible regardless of SAI. Perhaps the state of the atmosphere is such that it responds quicker to any tropospheric heat flux thanks to the SAI? Maybe. I also think that if something like a trough in NW Asia can screw up what was an incredible SAI response, perhaps we need to reevaluate a few things regarding the SAI. The atmosphere is one big puzzle. It could mean the SAI is a smaller piece. Or, this could be the one winter out of the next 7 or 8 where the signal was very weak. Sample size as always., is important. 

I wouldn"t say back to the drawing board though with 90% meaning 1 out of ten doesn"t work out and this could be 1.

Link to comment
Share on other sites

Well that's what I was saying the other day. If indeed it fails (which I am NOT saying), then it's back to the drawing board. That's how science evolves.

I do agree that if we see a flip in the near term, that would be a good thing for his work, but SSWs are possible regardless of SAI. Perhaps the state of the atmosphere is such that it responds quicker to any tropospheric heat flux thanks to the SAI? Maybe. I also think that if something like a trough in NW Asia can screw up what was an incredible SAI response, perhaps we need to reevaluate a few things regarding the SAI. The atmosphere is one big puzzle. It could mean the SAI is a smaller piece. Or, this could be the one winter out of the next 7 or 8 where the signal was very weak. Sample size as always., is important.

Exactly. But I think Eurasian snow cover advance is more likely than not to be a real phenomenon that increases the likelihood of a -AO, is all I'm getting at (as opposed to "the SAI took a dump this year" which would suggest that is a slam dunk and not a probabilistic forecasting tool). But if it is someday proven to be synoptically invalid, then that's science.

Link to comment
Share on other sites

The SAI is going to fail about 2 out of 10 times if we assume the correlations are accurate...which they probably aren't. But that is something to keep in mind. It is not a 100% success rate. My WAG is that the SAI is probably closer to 60-70% than 80%. The correlation was lower(more like 70) going back to 1972 versus 1997. So the recent success is probably just a run of good luck.

Link to comment
Share on other sites

I wouldn"t say back to the drawing board though with 90% meaning 1 out of ten doesn"t work out and this could be 1.

 

Right it could be the one time, but I guess my point is that the signal was overwhelming. If it were to fail, the fact that we had such a strong signal may need to be looked at...even for this one time.  

Link to comment
Share on other sites

I wouldn"t say back to the drawing board though with 90% meaning 1 out of ten doesn"t work out and this could be 1.

Yeah, that expression usually implies starting from scratch. It would more be that 2014-15 would need to be studied in detail to see where the theory needs to be modified or added to.
Link to comment
Share on other sites

Exactly. But I think Eurasian snow cover advance is more likely than not to be a real phenomenon that increases the likelihood of a -AO, is all I'm getting at (as opposed to "the SAI took a dump this year" which would suggest that is a slam dunk and not a probabilistic forecasting tool). But if it is someday proven to be synoptically invalid, then that's science.

 

Agree. See Will's comment too regarding sample size. 

Link to comment
Share on other sites

Right it could be the one time, but I guess my point is that the signal was overwhelming. If it were to fail, the fact that we had such a strong signal may need to be looked at...even for this one time.

Which could simply indicate that we are not measuring the signal correctly.
Link to comment
Share on other sites

Yeah the very strong signal and having the complete opposite happen is a pretty horrific miss. Not all misses are created equal.

This would definitely need to be looked over with hindsight after the season is over if we finish with a strong +AO

Link to comment
Share on other sites

The SAI is going to fail about 2 out of 10 times if we assume the correlations are accurate...which they probably aren't. But that is something to keep in mind. It is not a 100% success rate. My WAG is that the SAI is probably closer to 60-70% than 80%. The correlation was lower(more like 70) going back to 1972 versus 1997. So the recent success is probably just a run of good luck.

Could be good luck, or it could be something else about the state of the atmosphere. The two strongest waves will still interfere with each other.

Link to comment
Share on other sites

Could be good luck, or it could be something else about the state of the atmosphere. The two strongest waves will still interfere with each other.

Given the weak sample size since 1997, the statistician in me leans toward random noise in the positive direction. It doesn't mean it is useless, just that the extremely high correlation since 1997 might be a product of that variance. I prefer looking at the longer period.

But as the science evolves, we will obviously learn more. This branch of the science is still in its infancy.

Link to comment
Share on other sites

Yeah the very strong signal and having the complete opposite happen is a pretty horrific miss. Not all misses are created equal.

This would definitely need to be looked over with hindsight after the season is over if we finish with a strong +AO

Only if you see it as deterministic and not probabilistic. A 80% correlation doesn't have to mean an 80% correct theory...it can also be a measure of the influence of the metric on the atmospheric pattern.

Link to comment
Share on other sites

Only if you see it as deterministic and not probabilistic. A 80% correlation doesn't have to mean an 80% correct theory...it can also be a measure of the influence of the metric on the atmospheric pattern.

A year like this year would put a serious dent in that interpretation of it. If the signal is that powerful of a variable, then that said variable predicts one of the most negative AOs on record, it would be almost impossible for us to get something opposite of that if that weighting of the variable was actually correct.

We'd have to see that other 20% weight be so obscenely overwhelming to an 80% weight that is already predicting a monster -AO...so overwhelming in fact that not only the monster -AO doesn't occur, but it actually ends up positive.

This tells me there is a lot more to it. Perhaps there are years when the SAI just doesn't matter as much...but the key would obviously be to pick out which years those are before they happen. What type of base state do we need to be in for the SAI to have a weaker influence.

Link to comment
Share on other sites

Like a ssw-induced AO regime change?

Well you think that would help, even just to get some ridge bridging into north Greenland.  I think we saw this last year.  We had a very good but not great snow year, however we had several events with very high hourly snow rates.   But I think it will take more than an SSW.  There is something going on with the NAO lately that I don't think we understand.  On Scott's loop yesterday, there 2 other semi-permanent features that I noticed besides the Western Ridge...they were troughs over Europe and just on the other side of the north pole.  Maybe one of them needs to shift?

Link to comment
Share on other sites

The overall pattern in the 11-15 day and beyond features a pretty big block in the AK and Bering Sea region. Technical differences arise in whether or not a weak trough will form in between higher heights in the Bering Sea and the west coast, and whether or not the PNA tries to remain +.  IMO it is not a dry look at all right now. As usual, a rather +NAO will mean the chance for cutters are there, but it could be an interesting time starting next week. The EC ensemble is most bullish with the cold as it has the best blocking orientation.  Even if we lose some west coast riding at some point, the Bering Sea block is noteworthy and keeping the cold locked into Canada.  Someting that anomalous this far out is very impressive.

 

I'd urge caution here, the models including the enbsembles are extremely volatile of late. The EPS were dropping the major hammer from Wed aftn 12z through Thu aftn 12z, then last nights run had a remarkable shift warmer for the run (given its a 50 member average)..Some change images for ya 00z versus 12z

 

post-402-0-02577100-1421428943_thumb.png

post-402-0-63002900-1421428951_thumb.png

post-402-0-77571200-1421428960_thumb.png

 

I totally agree, but the -QBO and solar maximum were always two signals complementary to the SAI suggesting a SSW was likely in the second half of winter.

He is saying that the SSW that took place last week should take a couple weeks to propagate and will therefore flip the AO negative by the last week of January. From there, it should stay predominantly negative for the following 4-6 weeks. That's what needs to happen for the SAI to not "fail" as a meaningful prediction tool IMO. The final AO value for DJF is important too, but if it misses the mark because the timing was a couple weeks late and the AO stays negative into March, it doesn't invalidate the research.

In fact, even if we get a raging +AO for the rest of winter, it doesn't mean the premise is faulty...rather, it means that it is an incomplete picture and we don't know what we are missing (but we would have a really good place to start looking). Imagine if Einstein threw out all his theories because someone observed something in the universe that didn't line up perfectly.

 

We technically didnt have a SSW last week...Now it was a very solid and brief split, and another shot to weaken the NAM/Polar vortex state and that has implications going forward here, but we shall see.

Link to comment
Share on other sites

I'd urge caution here, the models including the enbsembles are extremely volatile of late. The EPS were dropping the major hammer from Wed aftn 12z through Thu aftn 12z, then last nights run had a remarkable shift warmer for the run (given its a 50 member average)..Some change images for ya 00z versus 12z

 

attachicon.gifeps1.PNG

attachicon.gifeps2.PNG

attachicon.gifeps3.PNG

 

 

We technically didnt have a SSW last week...Now it was a very solid and brief split, and another shot to weaken the NAM/Polar vortex state and that has implications going forward here, but we shall see.

The idea that it looks to be a stormy, snowy pattern certainly is up for debate. Last night's data certainly didn't make one feel warm inside if they want a lot of snow over the next 2 weeks

Link to comment
Share on other sites

The idea that it looks to be a stormy, snowy pattern certainly is up for debate. Last night's data certainly didn't make one feel warm inside if they want a lot of snow over the next 2 weeks

 

I do agree with what Will/Scott have said here over all, cold air involved and an active pattern gives us a good chances in the medium range...but cutters are a concern

The EC ensembles had basically ball-shriveling cold on some of those runs. So going a bit warmer than that is still going to be cold. It might actually help make it more active.

 

That is true that it shifted from ball shriveling to just colder than normal, but I've never seen a run to run shift as extreme as this morning on the EPS ...Following from A gas-weighted HDD standpoint as I have for the last 3 years, that run lost more HDDs than any run Io can remember. My caution recommendations are purely from a volatility standpoint on the runs, whose to say it doesnt keep trending warmer or shift back extremely colder next run? 

 

I think the general idea of not being dry stands. It's a pattern that can offer quick moving coastals and cutters. Just accept it. Hopefully we can cut down on the cutters.

Link to comment
Share on other sites

 Agree those differential maps show a nice snowstorm, what the problem?

 

 

Well to be fair, you don't want the underlying trend to have a weakening ridge out west all the time...because eventually we won't be starting from such a steroid ridge baseline, which means a trend less amped would hurt us a lot more than it does on the 00z EC ensemble.

Link to comment
Share on other sites

Well to be fair, you don't want the underlying trend to have a weakening ridge out west all the time...because eventually we won't be starting from such a steroid ridge baseline, which means a trend less amped would hurt us a lot more than it does on the 00z EC ensemble.

one run is not a trend though, GEFS stay solid

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.

×
×
  • Create New...