Recretos

I extracted the IOBW data and made some linear correlations with it. The positive phase looks very similar to the Glosea5 and POAMA forecasts. IOBW certainly is also one of the strong forcings in the seasonal model output.

Thats somewhat true. We also have to note that seasonal models are generally less effective at trop/strat coupling, except the ECM SEAS or perhaps GloSEA5. CFSv2 is currently forecasting a weaker vortex with actually some heat intrusions into the main circulation, but the darn model wouldn't know how to handle an SSW even if it modelled one.

Well, I dont think so, since I am also running (plotting) the POAMA model, from the Australian BOM office. Their model has a configuration wit h very low amp strat signal, so I use it mainly as a model to evaluate trop forcings. It is interesting, because it was portaying the same pattern since early summer, and its not until now that Glosea5 totally bowed down to it, despite having significantly supreme stratospheric forcing.

Agreed to a certain extent.

The concept is definitely interesting. Thanks for sharing. Tho there are a few things that could be further discussed or researched. But lets leave that for now. This tho: Actually, 2015 would have to go against only 4 or 5 years of data, since that is your actual sample size for your theory. It is true that the time span you looked at is that long, but from all those years, you have just about 4 or 5 that actually prove your theory, which is totally fine, but I dont think you can count other years. This: Are you sure you have done your research, because 68/69 might not have a major SSW but had major strat influences And: How did you actually prove that none of that either hindered or helped the outcome? Since those are some major climate drivers, which should not be disproved so lightly and making them inferior to your theory. Just a hint, since they are among the actual drivers that actually make or break your theory in reality. The summer AO index is not driving the weather literally, but its being driven along the way, in a very roughly sense of course. Because in theory it could be that the climate configuration is what made those winters and the correlation with the AO is purely situational, since some of the drivers or forcings have much better long term (decadal) correlations for winter. To make an actual 65 data correlation between the July AO and DJF 500mb heights, this is what comes out: There is a general signal present, showing the positive AO having a correlation to a positive winter NAO, but the correlation itself is statistically not really significant at all. Tho it is true this is a 3-month mean correlated with one month AO. I averaged the Dec, Jan and Feb 500mb correlations with July AO. The strongest signal was tho present in February, but that is a month with a lot of stratospheric mumbo-jumbos, which could affect the outcome more than the general correlation with the July AO. There is also a possibility of autocorrelation in a sense, since sometimes its the winter patterns that tends to affect the circulation down the line, including spring/summer, especially in a 65 year data sample. it is also true, that you have taken specific years with similar values and you got a decent correlation, which supports your theory, but my data sample does show that there is a lack of general correlation between July AO and winter AO/NAO, tho it is true that some years have a much better correlation, but given that it is a very specific one, we must not talk about a general summer AO influence/correlation, but more of a situational. So in the end, what I am trying to say, is that if your theory would actually go against 65 years of data, its on very weak legs, but as I said, the basis your your theory is very specific, so that changes it a bit. it would also help if you would also research what exactly is that affects the outcome of your theory, or why is it that summer AO correlates with winter AO atmospheric dynamics-wise, to further support it or make it even as a forecasting tool.

CFSv2 has a generally weaker vortex, displaced towards Iceland to Scandinavia region.

The band of brothers is forecasting the further slow establishment of the re-emerging polar vortex. And now we have the upgraded GEFS, which is now configuration-wise the best free ensemble forecasting system for the stratosphere and otherwise.

As far as the Atlantic discussions go, there is one thing to consider on a more longer term. The negative AMO phase is supposed to be induced also from below as much as from above if i remember correctly (writing this in a hurry). Comparing the Atlantic sub-surface temperatures, in this case at 150m, from August back to January, it actually shows that north Atlantic subsurface waters have further cooled, with the exception of the warm gulf stream which is expected to warm seasonally. And comparing August sub-surface temperatures to the 2009-2014 August mean, we actually get the more classic looking -AMO "pendant" or horseshoe. Data is from GLOSEA5 analysis on a 0.25° grid. The 09-14 August mean is derived from the ECMWF ORAS4 reanalysis.

Not sure if this comment is legit or sarcastic.

Looking at further correlations, the positive AMO signal seems to be still present.

Goni also looks nicely organised by HWRF.

Atsani also looks cool in 3D by HWRF data

Foundations of the new polar vortex are starting to appear over the pole in the FI.

Not only stratosphere and hurricanes, I also do 3D maps of ElNino. And sea surface height.

Well, so far, the strat seems to be doing the opposite of what is supposed to be the main signal from a positive ENSO3.4.

Judging by the correlations, and considering that the ENSO region is strongly positive, it looks interesting, tho the factor is not that strong.

I began test plotting tropical storms in 3D. Here is an example for Soudelor, from 2km forecast HWRF data. And an analysis plot from 0.25 GFS grid, to give you an idea for comparison how detailed does GFS see these storms.

All in all, it means nothing. Its just a graphical example of the model output, a.k.a the grid resolution.

Well I'm sorry, but like I said in the seasonal thread, I just cant see a March 2013 repeat. Despite monthly JMA mean having some ideas, but I am not really buying them. So I wont be completely offtopic, here is one of the new models I added into my "strat suite". Its none other than FIM9. The other one is FIM9 ZEUS. Next generation non-hydrostatic global model with an icosahedral grid. Too bad the main dynamics are over, so I cant really test it. But still, there is always the next season. Model resolution is 13km and grid resolution is 0.125°, which is the same as the ECMWF operational full resolution grid. The FU berlin seems to be using 0.5° or even 1°ECMWF grids, probably to cut on the data amount. The FIM9 and FIM9 ZEUS data both come in the 13km/0.125° grid resolution, and with the temporal resolution of 1 hour, which is quite phenomenal. Data access is through NOAA GSD HIWPP. I have noticed that it (FIM9) is being added to the WxBell model suite so I gave it a shot myself and it looks really good in high resolution. Regards p.s. Besides, march 2013 was an ongoing SSW response pattern. So good luck with that this year.

Who even came up with the idea of a March 2013 repeat?

Considering most people here are from UK, you should most certainly also look at AMO, AMM and AWP.

Always when in a hurry. You know, I dont have the time to always do deep analysis and reanalysis. Ok if it makes someone feel better, here is the JMA monthly ensemble forecast, which is closer to March 2013 pattern.

March 2013 repeat? I dont think so. At all. The March 2015 pattern looks to be more like the complete opposite of March 2013.

Not sure where to put this, but since it is related to seasonal forecasting since it is SST analysis, I will put it in here. My new test products. The text needs editing, etc, but the main point is on the graphic.

GFS Full resolution model terrain and actual grid (calculation points).