Winter 2014-2015 Thread

[mitchnick]

Hey Mitch....does the CFS show a frigid winter yet? Don't worry, sooner or later it will.

Bob, Mitch....we need a Dec 5 forecast every few days just to get the juices flowing.

I'm going to hold to that, and will check every day just in case too!

I can't believe that October will be here in <2 weeks....this late summer vacation has got my head all screwed up

he!!, Garrett County will be seeing flakes falling in 4-6 weeks I'm guessing

[Deck Pic]

I think we need to be aware of a faux -PDO where all the warm anomalies are shoved west of the dateline toward Japan...that can be a good pattern for us....a cold pattern

[wxmeddler]

New Zone Forecast areas for LWX. Goes into effect Dec 3. Montgomery, Howard, Hartford and Louden are now split. Also another two zones created for the skyline drive areas.

 http://nws.noaa.gov/om/public/zones-lwx.pdf

[mitchnick]

while everyone eagerly awaits Siberian snows in October, I found this link I saved from last year where Judah Cohen used his new model that takes into account the much vaunted Siberian snow cover along with a few other variables

you can see what his model predicted for the eastern US last year based upon it......oooops!

http://www.nsf.gov/news/special_reports/autumnwinter/predicts.jsp

 

just when you thought it was going to get easy come late October!

[wildweatherman179]

Lol. I love it when stuff like that happens.

[hosj III]

GEFS generally depicts Aleutian troughing by early October.

 

 

 

 

Also, what are we looking for with regards to snow cover advancement? I'd imagine we want a -AO with troughiness in Siberia.

[Eskimo Joe]

New Zone Forecast areas for LWX. Goes into effect Dec 3. Montgomery, Howard, Hartford and Louden are now split. Also another two zones created for the skyline drive areas.

 http://nws.noaa.gov/om/public/zones-lwx.pdf

Was on that project.  The split has been needed for years, and the snow event from Feb 12-14 of this year is a perfect example.  Nearly 2 ft. of snow fell in Damascus, while only 10 in. was recorded in Takoma Park.  

[uncle W]

DCA's snowiest winters...30" or more...similar look to NYC's snowiest winters but some different years mixed in...

[mattie g]

Cold + precip = snowy winters. Got it.

[hosj III]

Cold + precip = snowy winters. Got it.

Before I saw those pictures I figured we did best in warm+dry. I guess I have a lot to learn.

[CAPE]

Hey Mitch....does the CFS show a frigid winter yet? Don't worry, sooner or later it will.

Bob, Mitch....we need a Dec 5 forecast every few days just to get the juices flowing.

JAMSTEC has a cold snowy winter Has had it for a couple months now. Moderate Nino, cold central/eastern US, and above normal precip along the east coast. Book it.

[mitchnick]

JAMSTEC has a cold snowy winter Has had it for a couple months now. Moderate Nino, cold central/eastern US, and above normal precip along the east coast. Book it.

September forecast is not out yet, is it?

the JAMSTEC website is only updated through August from what I can see

[CAPE]

September forecast is not out yet, is it?

the JAMSTEC website is only updated through August from what I can see

No its not. I hadn't looked at it lately so I went to the website yesterday and was a bit surprised to still see the August forecast.

[AlaskaETC]

Andrew from the Weather Center blog isn't a big fan of the JAMSTEC forecast, doesn't think it will fully verify:

 

http://theweathercentre.blogspot.com/2014/09/jamstec-model-predicting-harsh-winter.html

[CAPE]

Andrew from the Weather Center blog isn't a big fan of the JAMSTEC forecast, doesn't think it will fully verify:

 

http://theweathercentre.blogspot.com/2014/09/jamstec-model-predicting-harsh-winter.html

I don't think anyone would expect any long term/climate model guidance to fully verify. If it holds when the Sept guidance comes out that would be 3 straight months with a favorable outlook for eastern US winter. 

[T. August]

Andrew from the Weather Center blog isn't a big fan of the JAMSTEC forecast, doesn't think it will fully verify:

 

http://theweathercentre.blogspot.com/2014/09/jamstec-model-predicting-harsh-winter.html

JB would disagree.

[GaWx]

Andrew from the Weather Center blog isn't a big fan of the JAMSTEC forecast, doesn't think it will fully verify:

 

http://theweathercentre.blogspot.com/2014/09/jamstec-model-predicting-harsh-winter.html

 

1) With all due respect to Andrew, he's clearly mistaken on how he's assessing the strength of the Nino on this 8/1/14 JAMSTEC Nino 3.4 forecast map for DJF. It actually is showing a weak El Nino. From Andrew in his assessment of what this JAMSTEC run's DJF map shows for 3.4: "First off, we see a moderate to borderline-strong El Nino evolving in the Pacific. SST anomalies of 1.2 to 1.5 degrees Celsius above normal might be a bit too strong for this winter, as I'm currently expecting only a weak to possibly moderate El Nino."

 

 What's wrong with what Andrew said about this JAMSTEC run? He's saying that it is predicting a +1.2 to +1.5 for DJF. His mistake is that it is not predicting a +1.2 to +1.5 and isn't even close. Based on just my eyeballing, it is predicting only ~+0.9. How do I come up with ~+0.9? I'll show here:

 

 When looking at the map, which I've attached below, keep in mind that Nino 3.4 covers 120W-170W between 5N and 5S. The warmest anomaly is a very tiny area near the equator in a portion of 120-125W. That means that tiny area is very likely only barely into the +1.2 to +1.5 range per the legend (the 5th pink area as you head up). So, essentially the warmest anomaly of any part of 3.4 is only near +1.2 and over only a very tiny % of 3.4. I estimate it covers <1% of 3.4, but I'll call it 1%. The next warmest area is +0.9 to +1.2. Even that covers <50% of 3.4..I estimate 45%. The rough average of that area is likely within +1.00 to +1.05...call it +1.025. The next warmest area is +0.6 to +0.9, which I estimate covers 50% of 3.4 and would likely average within +0.75 to +0.80...call it +0.775. Finally, roughly 4% of 3.4 is covered by the +0.3 to +0.6 category. I estimate that area is averaging near +0.55. So, here's the rough weighted average:

 

 (0.01 x +1.2) + (0.45 x +1.025) + (0.50 x +0.775) + (0.04 x +0.55) = +0.88. So, my rough estimate for JAMSTEK's 8/1/14 DJF map prediction for 3.4 is only +0.88...so, say +0.9. i.e., in weak Nino territory:

 

 

2) Related to the above, JAMSTEC also has a graph that shows the month by month predictions for 3.4. (see below) that is also from August. Per eyeballing, the ensemble mean (red line) is predicting ~+0.80 for D, ~+0.90 for J, and ~+0.95 for F. So, DJF is again ~+0.88..again call it ~+0.9. which is within weak Nino territory and again well below the mentioned +1.2 to +1.5 range:

 

 

**Edit: Aside: It shows near +0.85 for Sep., which while still doable, seems likely to verify too warm. Also, it shows nearly +0.5 for August. August was actually only ~+0.20. So, IF there is any bias, I'd say that this JAMSTEC DJF prediction is biased too warm in 3.4.

 

3) Using similar techniques of estimation and looking at the map, I'm conservatively estimating that this DJF map is calling for a PDO of +0.60 (i.e., it could even be showing somewhat higher). This is not chickenfeed  as it would be near the 75th percentile of highest DJF PDO's since 1900 when using the U of Wash. table. So, if we get a +0.60 averaged DJF PDO along with a weak Nino, then a solidly cold E US winter as per this same JAMSTEC run would actually be a VERY realistic possibility. The other crucial item would be a DJF averaged -NAO.

[millwx]

No its not. I hadn't looked at it lately so I went to the website yesterday and was a bit surprised to still see the August forecast.

 

I *think* the JAMSTEC updates on the 20th.  But may not run automatically (so, didn't run on Saturday).  So, I suspect the September update will take place on Monday (it's already Monday AM in Japan... so, any time now).  However, I'm guessing this because I think the last update was Aug 20th.  And I don't think it's like the IRI site (3rd Thursday), because Aug 20th was the 3rd Wednesday... 3rd Wed of Sep was last week.  So, my best guess is, JAMSTEC is updated today.  If not, it should be early this week.  Same with the GEOS5.

[millwx]

1) With all due respect to Andrew, he's clearly mistaken on how he's assessing the strength of the Nino on this 8/1/14 JAMSTEC Nino 3.4 forecast map for DJF. It actually is showing a weak El Nino. From Andrew in his assessment of what this JAMSTEC run's DJF map shows for 3.4: "First off, we see a moderate to borderline-strong El Nino evolving in the Pacific. SST anomalies of 1.2 to 1.5 degrees Celsius above normal might be a bit too strong for this winter, as I'm currently expecting only a weak to possibly moderate El Nino."

 

 What's wrong with what Andrew said about this JAMSTEC run? He's saying that it is predicting a +1.2 to +1.5 for DJF. His mistake is that it is not predicting a +1.2 to +1.5 and isn't even close. Based on just my eyeballing, it is predicting only ~+0.9. How do I come up with ~+0.9? I'll show here:

 

 When looking at the map, which I've attached below, keep in mind that Nino 3.4 covers 120W-170W between 5N and 5S. The warmest anomaly is a very tiny area near the equator in a portion of 120-125W. That means that tiny area is very likely only barely into the +1.2 to +1.5 range per the legend (the 5th pink area as you head up). So, essentially the warmest anomaly of any part of 3.4 is only near +1.2 and over only a very tiny % of 3.4. I estimate it covers <1% of 3.4, but I'll call it 1%. The next warmest area is +0.9 to +1.2. Even that covers <50% of 3.4..I estimate 45%. The rough average of that area is likely within +1.00 to +1.05...call it +1.025. The next warmest area is +0.6 to +0.9, which I estimate covers 50% of 3.4 and would likely average within +0.75 to +0.80...call it +0.775. Finally, roughly 4% of 3.4 is covered by the +0.3 to +0.6 category. I estimate that area is averaging near +0.55. So, here's the rough weighted average:

 

 (0.01 x +1.2) + (0.45 x +1.025) + (0.50 x +0.775) + (0.04 x +0.55) = +0.88. So, my rough estimate for JAMSTEK's 8/1/14 DJF map prediction for 3.4 is only +0.88...so, say +0.9. i.e., in weak Nino territory:

 

JAMSTEC080114Nino34predictMap.gif

 

2) Related to the above, JAMSTEC also has a graph that shows the month by month predictions for 3.4. (see below) that is also from August. Per eyeballing, the ensemble mean (red line) is predicting ~+0.80 for D, ~+0.90 for J, and ~+0.95 for F. So, DJF is again ~+0.88..again call it ~+0.9. which is within weak Nino territory and again well below the mentioned +1.2 to +1.5 range:

 

JAMSTECAugust2014Nino34predictGraph.gif

 

**Edit: Aside: It shows near +0.85 for Sep., which while still doable, seems likely to verify too warm. Also, it shows nearly +0.5 for August. August was actually only ~+0.20. So, IF there is any bias, I'd say that this JAMSTEC DJF prediction is biased too warm in 3.4.

 

3) Using similar techniques of estimation and looking at the map, I'm conservatively estimating that this DJF map is calling for a PDO of +0.60 (i.e., it could even be showing somewhat higher). This is not chickenfeed  as it would be near the 75th percentile of highest DJF PDO's since 1900 when using the U of Wash. table. So, if we get a +0.60 averaged DJF PDO along with a weak Nino, then a solidly cold E US winter as per this same JAMSTEC run would actually be a VERY realistic possibility. The other crucial item would be a DJF averaged -NAO.

 

I don't mean to dog on Andrew, as I don't even know him, but that's not a "misinterpretation" of the JAMSTEC Nino forecast... that's a GROSS misinterpretation!  There isn't even a +1.2 contour on the map in the equatorial Pacific (ok, one tiny spot at 120W).  I wouldn't be so mean and negative if it was a simple misreading (e.g., like the first red shade is +0.3 to +0.6C... which would be an easy mistake to make).  But if that were the case, while Nino3 might be moderate to borderline strong (which, I admit, wouldn't be good), the "official" Nino reading (Nino3.4) would still barely be into moderate range... +1.1C, *maybe* +1.2C, at most.  And, again, that would be if you misread the map by 0.3C.  So, REALLY, it's about +0.8 to +0.9C... as you correctly noted.  So, Andrew's misreading of the JAMSTEC is pretty horrific... he's off by basically one full Nino category (about 0.5C).  It'd be fine if he were saying he thought the Nino forecast were wrong... that HE thinks it'll be a moderate to borderline strong (though I'd disagree with him, that'd be fine... that's his opinion).  But that doesn't appear to be what he's saying here... from what you're saying (I haven't actually read his forecast myself), he's actually stating that the JAMSTEC is predicting a mod/strong Nino... which is not even in the ballpark of accurate.

[GaWx]

I don't mean to dog on Andrew, as I don't even know him, but that's not a "misinterpretation" of the JAMSTEC Nino forecast... that's a GROSS misinterpretation! There isn't even a +1.2 contour on the map in the equatorial Pacific (ok, one tiny spot at 120W). I wouldn't be so mean and negative if it was a simple misreading (e.g., like the first red shade is +0.3 to +0.6C... which would be an easy mistake to make). But if that were the case, while Nino3 might be moderate to borderline strong (which, I admit, wouldn't be good), the "official" Nino reading (Nino3.4) would still barely be into moderate range... +1.1C, *maybe* +1.2C, at most. And, again, that would be if you misread the map by 0.3C. So, REALLY, it's about +0.8 to +0.9C... as you correctly noted. So, Andrew's misreading of the JAMSTEC is pretty horrific... he's off by basically one full Nino category (about 0.5C). It'd be fine if he were saying he thought the Nino forecast were wrong... that HE thinks it'll be a moderate to borderline strong (though I'd disagree with him, that'd be fine... that's his opinion). But that doesn't appear to be what he's saying here... from what you're saying (I haven't actually read his forecast myself), he's actually stating that the JAMSTEC is predicting a mod/strong Nino... which is not even in the ballpark of accurate.

millwx,

Andrew is saying that this August 1st version of JAMSTEC is predicting +1.2 to +1.5 for DJF in Niño 3.4. He's considering that moderate to borderline strong (which would be an appropriate description if it were really predicting +1.2 to +1.5). He also said he's expecting a weak to possibly moderate El Niño this winter. So, the irony here is that this JAMSTEC is in reality predicting what Andrew, himself, is predicting though he doesn't realize that due to his misinterpretation!

Edit and aside: Looking more closely at JAMSTEC's DJF temperature prediction, it actually looks like it may very well not be cold enough in the E US if we get the weak Niño/+PDO combo it is predicting along with a solid DJF averaged NAO. The coldest it shows for the E 1/3 of the US is -0.9 C. Whereas that's chilly, that's nowhere near as cold as I think it would get in a good portion of the E US if the aforementioned combo does materialize for DJF. Actually, looking around the world, the coldest shown is only down to the -1.2 to -1.5 range near DFW, TX! For a coldest world anomaly, that seems awfully tame. So, I'm assuming it is being conservative and the most important thing is that the US would be the coldest country anomalywise in the world by a good margin if the map's general pattern were to verify well.

[millwx]

millwx,

Andrew is saying that this August 1st version of JAMSTEC is predicting +1.2 to +1.5 for DJF in Niño 3.4. He's considering that moderate to borderline strong (which would be an appropriate description if it were really predicting +1.2 to +1.5). He also said he's expecting a weak to possibly moderate El Niño this winter. So, the irony here is that this JAMSTEC is in reality predicting what Andrew, himself, is predicting though he doesn't realize that due to his misinterpretation!

Edit and aside: Looking more closely at JAMSTEC's DJF temperature prediction, it actually looks like it may very well not be cold enough in the E US if we get the weak Niño/+PDO combo it is predicting along with a solid DJF averaged NAO. The coldest it shows for the E 1/3 of the US is -0.9 C. Whereas that's chilly, that's nowhere near as cold as I think it would get in a good portion of the E US if the aforementioned combo does materialize for DJF. Actually, looking around the world, the coldest shown is only down to the -1.2 to -1.5 range near DFW, TX! For a coldest world anomaly, that seems awfully tame. So, I'm assuming it is being conservative and the most important thing is that the US would be the coldest country anomalywise in the world by a good margin if the map's general pattern were to verify well.

 

GaWx,

 

An Ensemble model (as the JAMSTEC is) will almost-by-definition always underestimate the peak anomalies (perhaps dramatically)... even our normal 15 day ensemble models.  So, ignore the absolute anomalies.  Just focus on the relative strengths.  Why will the Ensembles underestimate peak anomalies?  Two reasons.  First, if reality supports truly intense anomalies, they may be near the upper bounds of what is even atmospherically reasonably plausible (i.e., there is some reasonable bound on the temperatures... like D.C. will never see a low temperature of 30 below zero).  As such, in such a case, the spread in the Ensemble members will be such that few will overestimate the anomalies while many will underestimate them... putting the ensemble mean too weak.  But even if we don't have an extreme case like this... the second problem is, even if every member nailed the anomaly (let's just, for example, say the peak seasonal anomaly is predicted to be -3F), there will likely be no -3F contour on the ensemble mean map.  Why?  Because the placement will be slightly different on all maps (maybe a LOT different on some of them).  The result is that the ensemble mean will overestimate the EXPANSE of the significant anomalies, but underestimate the CORE intensity of the anomalies.

 

In the case of the August JAMSTEC run, that probably means the Rockies, Midwest, Lakes, Northeast and N Mid-Atlantic will likely be closer to normal (but still leaning below - which, for snow lovers is probably still just fine in those areas... don't need massive cold anomalies there).  On the flip side, it's probably indicating brutal temps for TX, the S Plains, Lower MS Valley, TN Valley and *probably* the Southeast and lower Mid-Atlantic (definitely still colder than normal in the SE/lower Mid-Atl ...I just say "probably" on the severity, because it's tough to interpret given what I just said about how an ensemble model will hand anomalies; that area is solidly in the "Belows", but outside of the most severe contours).

 

On a side note, I thought the JAMSTEC would get updated today (see my comment up-thread).  Guess not.  It's already evening there and still no update.  It was updated on the 20th of August.  And I think the July update was also on the 20th.  I hadn't followed it too closely prior to that.  So, I assumed they update on the 20th.  When it didn't come out on Saturday, I assumed, since it only comes out once a month, they run it (or at least do the data distribution) manually... so, it doesn't get done on the weekend.  So....... I presumed today would be the day.  No such luck, apparently.  Still not out there.  The GEOS5 (NASA model) also still not updated (I think it gets updated about now too).  Those are the only two "primary" monthly-run seasonal forecasts not updated for September yet.  I'm getting impatient... there's less than 10 days left in the month!

[clskinsfan]

Was on that project.  The split has been needed for years, and the snow event from Feb 12-14 of this year is a perfect example.  Nearly 2 ft. of snow fell in Damascus, while only 10 in. was recorded in Takoma Park.  

 

I agree. This was a great move. It has been needed for years. And from the maps it looks like they got the cutoff correct. At least for VA. I cant speak for MD.

[Bob Chill]

3.4 continues to improve. Enso regions as a whole look ok. However, good luck finding an enso analog that looks similar in the Pac north of the eq (especially the eastern half). 04 is the closest match as a whole but far from perfect. 

 

 

 

 

 

I would expect the region north of enso and s of HI to cool as we approach met winter (mostly a guess based on previous warm enso events). 

 

I'm also not really buying into the ne pac warm pool supporting a -epo again this winter just because it exists today. That area can easily cool off within a month or 2 (and has been slowly this month)  if the pattern is hostile. 

[WxUSAF]

North Atlantic has really cooled over the last 10 days.  Is that mostly from Edouard?  Can kind of see his path around 30N/50W.  

[Deck Pic]

3.4 continues to improve. Enso regions as a whole look ok. However, good luck finding an enso analog that looks similar in the Pac north of the eq (especially the eastern half). 04 is the closest match as a whole but far from perfect. 

 

 

 

 

 

I would expect the region north of enso and s of HI to cool as we approach met winter (mostly a guess based on previous warm enso events). 

 

I'm also not really buying into the ne pac warm pool supporting a -epo again this winter just because it exists today. That area can easily cool off within a month or 2 (and has been slowly this month)  if the pattern is hostile. 

 

meh...looks pretty stagnant..most of the warm anomalies are west of the dateline..not sure it matters much...

[AlaskaETC]

ENSO thread mentioned that the trade winds are kicking up again, which may make things more complicated. I would hate to see this as the "El Nino that Never Was." I am probably worrying too much.

[Deck Pic]

ENSO thread mentioned that the trade winds are kicking up again, which may make things more complicated. I would hate to see this as the "El Nino that Never Was." I am probably worrying too much.

 

we are in better shape than 2012..I think if Nino is going to die, we will see it happen over the next 2-3 weeks...that said we want to make sure we maintain at the least, and then see a rise soon...we should know soon what is up soon..in 2012 right when we should have had momentum, it died...and our chances were over...though it did get nino-ish for a time..

 

[Bob Chill]

meh...looks pretty stagnant..most of the warm anomalies are west of the dateline..not sure it matters much...

 

The whole region has improved slowly through the last 3 weeks. Definitely not great but we aren't losing ground at least. The basin as a whole looks better than where we started this month. Where we go from here is tough to say. I would guess sideways. 

 

[mitchnick]

we are in better shape than 2012..I think if Nino is going to die, we will see it happen over the next 2-3 weeks...that said we want to make sure we maintain at the least, and then see a rise soon...we should know soon what is up soon..in 2012 right when we should have had momentum, it died...and our chances were over...though it did get nino-ish for a time..

 

weekly numbers have 3.4 remaining at +.5C and 4 rising from +.7C to +.8C

1&2 and 3 were the same as last week

stagnant it is

[mitchnick]

some interesting dates/years on that map

http://www.cpc.ncep.noaa.gov/products/predictions/short_range/tools/gifs/500hgt_comp_sup814.gif