Teleconnections: A Technical Discussion - Page 29 - Teleconnections, Atmosphere, Arctic, Climate Change - 33andrain Jump to content
Snowy Hibbo

Welcome to the Teleconnections thread!

Check out some of the other driver threads and resources here:

Arctic thread

Stratosphere thread

Weather research portal

Message added by Snowy Hibbo

Recommended Posts

10 minutes ago, Isotherm said:

The following is not meant to be taken as an identical comparison or expectation of future outcomes, but merely an example of the timing of progression possible this year.

Below, I have screen shots of December 1984 and December 1985 at Central Park. A major stratospheric split event occurred on January 2nd, 1985 - easterly QBO winter. Very similar timing to this year's zonal wind reversal yesterday.

 

Note the blowtorch in late December (December was +6.3 in NYC) with record highs. NYC recorded anomalous warmth on the day of the SSW event on January 1st-2nd, much like this year. Subsequently, the temperature diminution initiated approximately 1 week following the SSW event -- January 8th. Beyond that point, temperatures were cold to very cold (the departure was -3.8 for January). Keep in mind every year has its own idiosyncrasies. 

 

The point here is the interesting correlation between SSW occurrence and contemporaneous record warmth, with alterations about 1 week following for SPV-split event in an EQBO year. 

 

2pqkgzt.png

 

314ujpl.png

 

Amazing looking at that chart on December 27th 4.8” of snow fell and two days later the high was 70 degrees. Wow!  If that was early April no big deal but late December that was incredible 

  • Like 2

Share this post


Link to post
Share on other sites

Posted some of this in the general January thread, but it's fast-moving and got no responses.  Thought I'd try again here.

 

I don't really follow the long range forecasting/MJO stuff much - are there published correlations between the various phases (and their amplitudes) and things like temperature and snowfall for the Philly-NYC area (and in general)?  Also, how accurate are the forecasts of the MJO phases/amplitudes beyond about 10-15 days?  Deterministic numerical model forecasting for specific days tends to become quite inaccurate beyond about Day 7-8 (there's a reason the NWS doesn't make specific forecasts beyond then, unlike click-bait AccuWeather), due to the chaotic nature of the weather, but I've seen pattern forecasting out to a month or so being done pretty decently by some of the folks around here and elsewhere, so presumably there is some skill to the MJO forecasts beyond Day 10-15, since the long-range folks appear to rely heavily on them in their long-range pattern forecasts. 

 

I imagine one could ask the same questions about long range forecasts for the other dozen or more indices and their correlations to sensible weather outcomes in the long range in specific locations, too.  I would also imagine that folks who are really serious about long-range forecasting are probably doing Fourier style sensitivity analyses of a variety of combinations of indices vs. hindcast outcomes to find correlations and then using those (plus seasonal analogs) to formulate long range forecasts.  I'm sure the easy answer would be for someone to say, something like don't be lazy, read the links, lol.  Unfortunately, that's far more work than I want to put into this right now and I'm guessing some of the experts could summarize it far more quickly than I'd figure it out (or maybe I need a tutor, lol - but it's been 30+ years since I took grad level chem E courses in thermo, transport, p-chem, modeling, calc, etc., so I've probably forgotten 80% of it, sadly).  I get the sense, though, that I'm not alone in half getting much of this, but not enough to make complete sense of it all.  Thanks, in advance for any light shed on the topic.  

Share this post


Link to post
Share on other sites

The last couple post on the subject have been phenomenal keep up the great work guys

  • Like 3

Share this post


Link to post
Share on other sites

I think these discussions are fantastic. Thanks so much for this information. I'm very new to learning about the AAM, so I had a few questions:

 

As I read through the posts, it seems that +AAM leads to a colder pattern for the East U.S and W. Europe. Yet, I have also read posts that say +AAM extends the Pacific jet. For the Eastern U.S, this fast Pacific jet has been bringing us zonal flow and milder Pacific air. In part because of this extended jet, we haven't had much snow all winter. For those of us in the East U.S, does a +AAM lead to mild winter weather (because of the extension of the Pacific jet)? If not, why is a +AAM favorable when it extends the Pacific jet and sends Pacific air into the U.S.?  

Thanks 

  • Like 3

Share this post


Link to post
Share on other sites

Sorry to interrupt your discussion about the GWO and its implications, but I can't really understand you, unless I have some more clarity on the value of the data. 

 

The fundamentals of the theory of AAM are pretty much clear to me, but when trying to determine the value of AAM and GWO for the weather, I am hindered by (what appear to be) contradictionary data from the relevant plots. 

 

First of all when I look at the vertical and zonal integral relative AAM, the value is between +1 and +2. 

1040858343_IntegralrelativeAAM.PNG.abe07b75a91acd99495db9953d7ed03f.PNG

But looking at the CFSv2 forecast, the actual value of Global Relative AAM is about 0,5.

1915693630_AAMforecastCFS.PNG.49c5e9089392d5930f329a4b584de288.PNG

 

The bias corrected version shows a even bigger deviation, its value is about -0,5. 

Is this a deviation in the data or is there any real difference between the vertical and zonal integral relative AAM and the Global Relative AAM, which I don't understand?

 

The MJO is a very useful tool for understanding the weather and for forecasting the weather, because the data are clear and unambigious. When considering the actual AAM on the other hand, it can be either positive or negative (depending on a bias corrected version or not), which makes a great difference for the phase of the GWO. The GWO can be either in phase 8 or 1 and will move to phase 5 or three in the coming week (http://atlas.niu.edu/gwo/).

 

I would like to have some clarity in this matter. Perhaps it would be helpful to have a link to the next plot of the GWO, often used at this forum. Maybe it gives me the fixed data I need to have a reliable reference.

 

993781971_GWOrelative.PNG.6ab5a403a7c8a7a29a9b9c956ccba2f0.PNG

 

Share this post


Link to post
Share on other sites

I agree with this, how can models have different initialisation points to start with?

Share this post


Link to post
Share on other sites
5 hours ago, Snowy Hibbo said:

G'day @Plapsel @Tee47

 

The charts have different intialisation points, because of model bias. The WDT/GSDM Solutions charts are based on R1/RNL observational data, run from NOAA. The GEFS data has bias that isn't even corrected by this version, and is being tweaked with constantly I believe. The bias was positive a few months back, and now it is rather negative. I believe this is a process of trial and error. Similar with the CFS data, but it is closer to the RNL data (less biased). These initialisations are based on a model's interpretation of the atmosphere (more to it than clicking in observational data). The RNL data is I would suggest closer to realtime  obs data in nature, so probably the best one to base your Forecast off.

 

Look here for the latest GWO plot:

http://gsdmsolutions.com/~gsdm/clim/gcm/gwo_90d.gif

User: gsdm01

Password: gu3st#1

 

I would base your thinking of the GSDM/GWO on that particular plot.

 

@WxBillMoHill

 

The best way to look for positive North American Mountain Torque is to see if there are highs to the East of the Rockies (and lows to the West of the Rockies). A descending high over the Rockies from Western Canada into the jetstream towards the Eastern US is often a +NAMT making situation.

 

What a +NAMT does to the jetstream is that it strengthens it coming out through the Eastern US and into the Atlantic. This amplifies the pattern, and can eventually bring in colder air if other factors are in place. 

 

@ru848789

 

Sorry for such a long wait for a response.

 

There are no published correlations to MJO and/or GWO, and snowfall/temps for the Philly/NYC area. It's hard to do statistical analysis with phase charts, unless you are talking about singular events like SSWs. I have done a few statistical analyses for various regions and the GLAAM (among other climate predictors). It's a very slow process manually.

 

As for MJO verification, I posted that chart a few posts above. At 15 days, it is pretty poor. EPS does an okay job in verification stats, but the reality is you have to see the pattern for yourself, and look at other factors that influence the MJO orbit.

 

As for wider scale verification, I think it is quite hard to do statistical analysis to that kind of extent, given all of our time in the weather domain is usually limited. There's verification plots out there, for example like the MJO plot, but the list isn't very big at all. I may eventually get some verification/statistical plots for the Eastern US (NYC/Philly) into a proposed project on the forum, but that is way down the track.

 

Thanks for taking the time to address my questions.  Figured verification scores for various phases/indices might not be that great and that there might not be readily available correlations of these parameter to sensible weather, either globally or regionally, but thought I'd still ask.  Also, one more question while I'm at it.  Has anyone published anything in the seasonal winter forecasting arena which has been as "well accepted" as Dr. Gray's methodology for seasonal tropical forecasting?  Thanks!

Share this post


Link to post
Share on other sites

How can I get 500 hpa and 850 hpa 40 day average anomalies?

Share this post


Link to post
Share on other sites

×
×
  • Create New...