Medium Range Discussion Winter 2011-12

[Ji]

With respect to the ongoing La Niña, the ENSO Region 3.4 anomalies warmed to -0.6°C for the week centered around 2/15 vs. -1.0°C for the prior week. The warming is likely a response to several recent bouts of the SOI's being < 0. Overall, the La Niña should, in general, continue to fade and the recent warming suggests that neutral ENSO conditions could arrive at some point in March. The recent development should not make a material difference in the posted idea concerning March (Message #1197).

any idea of Enso for next season?

[donsutherland1]

any idea of Enso for next season?

Not yet. Most of the guidance I've seen shows either neutral or weak El Niño conditions going into the winter. However, the model skill level from this far out is very poor. For example, the CFSv2's error approaches 1.0°C.

[Quixotic1]

Not yet. Most of the guidance I've seen shows either neutral or weak El Niño conditions going into the winter. However, the model skill level from this far out is very poor. For example, the CFSv2's error approaches 1.0°C.

OT, but haven't noticed a new ENSO thread. I saw the CPC/IRI probabilistic forecast of ENSO conditions up til November. Seemed to be calling for about even chances for any of Nino, Nina or Neutral.

http://iri.columbia.edu/climate/ENSO/currentinfo/figure1.html

Of the five previous episodes of multi-year Ninas, four of them ended with a Nino the following winter. The one that didn't was 2001-2002, which ended as slightly negative-neutral. With the models trending positive, a Nino might be the best bet. Course, that may be wishful thinking on my part.

[donsutherland1]

With the models trending positive, a Nino might be the best bet. Course, that may be wishful thinking on my part.

I hope that will be the case. My guess is that by late spring/early summer, there will be a reasonable model consensus.

[donsutherland1]

Some prior runs have teased warm-weary Easterners about significant cold (a few runs were truly extreme, with one featuring single-digit lows as far south as Washington, DC) and snowfall opportunities during the first week in March. However, the composite anomalies I came up with showed that the cold would likely be focused in the West (Pac NW) and a large portion of Canada. The day 11 objective analogs rolled forward 2 days (to be centered on March 4) also focus the cold away from the eastern CONUS.

With the GFS ensembles showing the AO (currently +1.145 and rising) reaching between +2 and +3, along with a somewhat negative PNA, one probably should be wary of cold model runs in the East. Instead, the cold will likely be dumped into the West (possibly the Plains, too).

Moreover, in a somewhat cruel irony, the 1985 analog is finally popping up for the first week in March. Unfortunately, it's not the one most here would welcome. Based on the forecast AO and PNA, the March 8-10, 1985 period is showing up. That timeframe featured conditions reasonably similar to the temperature anomalies shown above.

In sum, new data since I posted my March 1-7, 2012 thoughts (#1183 in this thread) have increased my confidence in that idea. Cold shots in the East will likely continue to follow the winter pattern: quick-hitting and not too severe. The Pacific Northwest will likely feature more sustained cold.

[am19psu]

I'm definitely concerned about busting the period post March 7. There is quite a large difference between the Roundy MJO guidance (which shows the MJO zipping through the phase space, getting into P7 by March 11 and P1 by the 18th) and the Euro Weekly Ensemble (which shows the MJO in P5 and P6 at those same periods, respectively). The Roundy guidance would bring cold back to the East much more quickly than the Euro, which shows very warm temps through its Week 3 and Week 4 periods.

[Ginx snewx]

Some prior runs have teased warm-weary Easterners about significant cold (a few runs were truly extreme, with one featuring single-digit lows as far south as Washington, DC) and snowfall opportunities during the first week in March. However, the composite anomalies I came up with showed that the cold would likely be focused in the West (Pac NW) and a large portion of Canada. The day 11 objective analogs rolled forward 2 days (to be centered on March 4) also focus the cold away from the eastern CONUS.

With the GFS ensembles showing the AO (currently +1.145 and rising) reaching between +2 and +3, along with a somewhat negative PNA, one probably should be wary of cold model runs in the East. Instead, the cold will likely be dumped into the West (possibly the Plains, too).

Moreover, in a somewhat cruel irony, the 1985 analog is finally popping up for the first week in March. Unfortunately, it's not the one most here would welcome. Based on the forecast AO and PNA, the March 8-10, 1985 period is showing up. That timeframe featured conditions reasonably similar to the temperature anomalies shown above.

In sum, new data since I posted my March 1-7, 2012 thoughts (#1183 in this thread) have increased my confidence in that idea. Cold shots in the East will likely continue to follow the winter pattern: quick-hitting and not too severe. The Pacific Northwest will likely feature more sustained cold.

See a lot of 56s here, now that would be unreal!

[donsutherland1]

See a lot of 56s here, now that would be unreal!

Ginx,

Unfortunately, the composite temperature anomalies for the 1956 dates shown are very warm in the East. The composite is not very different from the composite anomalies I posted in Message #1183 in this thread.

[donsutherland1]

I'm definitely concerned about busting the period post March 7. There is quite a large difference between the Roundy MJO guidance (which shows the MJO zipping through the phase space, getting into P7 by March 11 and P1 by the 18th) and the Euro Weekly Ensemble (which shows the MJO in P5 and P6 at those same periods, respectively). The Roundy guidance would bring cold back to the East much more quickly than the Euro, which shows very warm temps through its Week 3 and Week 4 periods.

I agree. The difference is quite striking. However, it appears that the MJO is now progressing more slowly through Phase 1, so I lean against the Roundy guidance at this time. I favor a rate of progression somewhere between the faster Canadian and slower ECMWF ensembles.

[usedtobe]

I agree. The difference is quite striking. However, it appears that the MJO is now progressing more slowly through Phase 1, so I lean against the Roundy guidance at this time. I favor a rate of progression somewhere between the faster Canadian and slower ECMWF ensembles.

Don,

I'm with you about going with a compromise between the Canadian and Ecmwf plus even when we went through phases 8-2 last time it just wasn't that cold.

[Ginx snewx]

Ginx,

Unfortunately, the composite temperature anomalies for the 1956 dates shown are very warm in the East. The composite is not very different from the composite anomalies I posted in Message #1183 in this thread.

It was the roll forward that interests me Don.

[Ottawa Blizzard]

Some prior runs have teased warm-weary Easterners about significant cold (a few runs were truly extreme, with one featuring single-digit lows as far south as Washington, DC) and snowfall opportunities during the first week in March. However, the composite anomalies I came up with showed that the cold would likely be focused in the West (Pac NW) and a large portion of Canada. The day 11 objective analogs rolled forward 2 days (to be centered on March 4) also focus the cold away from the eastern CONUS.

With the GFS ensembles showing the AO (currently +1.145 and rising) reaching between +2 and +3, along with a somewhat negative PNA, one probably should be wary of cold model runs in the East. Instead, the cold will likely be dumped into the West (possibly the Plains, too).

Moreover, in a somewhat cruel irony, the 1985 analog is finally popping up for the first week in March. Unfortunately, it's not the one most here would welcome. Based on the forecast AO and PNA, the March 8-10, 1985 period is showing up. That timeframe featured conditions reasonably similar to the temperature anomalies shown above.

In sum, new data since I posted my March 1-7, 2012 thoughts (#1183 in this thread) have increased my confidence in that idea. Cold shots in the East will likely continue to follow the winter pattern: quick-hitting and not too severe. The Pacific Northwest will likely feature more sustained cold.

Don, March 4, 1985 had a pretty big snowstorm (about 10") in both Toronto and Ottawa, while March 1956 wasn't too bad for snowfall in both of those cities.

[Snowstorms]

Don, March 4, 1985 had a pretty big snowstorm (about 10") in both Toronto and Ottawa, while March 1956 wasn't too bad for snowfall in both of those cities.

We were lucky to be on the cold side of the gradient. With strong temperature gradients as depicted by Don's maps, there's quite the gradient which is a good recipe for not only stronger storms but more severe wx as climo favors severe wx starting to develop across the south.

This pattern is a typical Nina pattern which is FAR better than what we've been seeing which is very unusual for a Nina regime, esp a second year Nina.

[donsutherland1]

Don, March 4, 1985 had a pretty big snowstorm (about 10") in both Toronto and Ottawa, while March 1956 wasn't too bad for snowfall in both of those cities.

The February 24-25 period offers Toronto an opportunity to pick up accumulating snow. Right now, my initial thinking is something in the 7.5 cm-15.0 cm (3"-6") range appears reasonably likely. There may be some additional opportunities afterward.

Along the East Coast, I'm less optimistic despite the 2/22 12z ECMWF's 192-hour solution and, to a lesser extent, the 12z ECMWF Ensembles solution. First, only one KU snowstorm occurred during La Niña conditions and a positive AO (January 2000 event). The blockbuster March 1956 La Niña snowstorm had an AO-/PNA+ combination. Typically, with AO+/PNA- combinations (what is forecast for the 192-hour timeframe) and ENSO Region 3.4 anomalies < 0, snowstorms have a more moderate impact (March 2009 is an example) or focus on New England (April Fool's Day Storm of 1997 and mid-March 2007 storm). In short, if there is a winter weather event around 192 hours, my guess is that it would far more likely be a relatively localized event/near miss/changeover case than a KU-type one.

[uncle W]

The February 24-25 period offers Toronto an opportunity to pick up accumulating snow. Right now, my initial thinking is something in the 7.5 cm-15.0 cm (3"-6") range appears reasonably likely. There may be some additional opportunities afterward.

Along the East Coast, I'm less optimistic despite the 2/22 12z ECMWF's 192-hour solution and, to a lesser extent, the 12z ECMWF Ensembles solution. First, only one KU snowstorm occurred during La Niña conditions and a positive AO (January 2000 event). The blockbuster March 1956 La Niña snowstorm had an AO-/PNA+ combination. Typically, with AO+/PNA- combinations (what is forecast for the 192-hour timeframe) and ENSO Region 3.4 anomalies < 0, snowstorms have a more moderate impact (March 2009 is an example) or focus on New England (April Fool's Day Storm of 1997 and mid-March 2007 storm). In short, if there is a winter weather event around 192 hours, my guess is that it would far more likely be a relatively localized event/near miss/changeover case than a KU-type one.

first off Don, great work...I agree about no KU events unless we get a similar storm to the October 29th event...if we see any snowfalls they might be like March snowfalls from 1955, 1965, 1968, 1976, 1989, 1999...3-5" range...two to three day cold period if we are lucky...without cold air it can't snow...This winter will end up with the least amount of days with a minimum of 32 or lower in NYC...The record will be crushed if March is warm...

[donsutherland1]

first off Don, great work...I agree about no KU events unless we get a similar storm to the October 29th event...if we see any snowfalls they might be like March snowfalls from 1955, 1965, 1968, 1976, 1989, 1999...3-5" range...two to three day cold period if we are lucky...without cold air it can't snow...This winter will end up with the least amount of days with a minimum of 32 or lower in NYC...The record will be crushed if March is warm...

Thanks Uncle W. I agree about the fewest days with minimum readings of 32°F or below in NYC. Already, some of what I'm looking at hints that the general milder than normal readings likely in the first week in February will persist beyond the first week of the month.

[donsutherland1]

Some evening thoughts:

1. The idea of a 7.5 cm-15 cm (3"-6") snowfall for Toronto still appears reasonable for the February 24-25 timeframe. There is some potential for somewhat more snowfall. Should such a snowfall materialize, it would be Toronto's biggest snowfall of the winter. Additional opportunities for snowfall appear possible afterward.

2. The idea that Toronto would likely not see any -20°C (-4°F) days in February (Message #808 in this thread) appears increasingly probable. To date, the month's lowest reading is -13.9°C (7.0°F) on February 11.

3. The month is likely to end on a colder than normal note in the Pacific Northwest as the PNA (currently +0.341) heads negative in coming days. The generally colder than normal weather should persist through at least the first week in March. At least one strong shot of cold with the potential to send the temperature into the 20s in Seattle appears possible over the next 2 weeks. There may also be some opportunities for snowfall.

4. The NAEFS is showing above normal temperatures across the eastern third of the U.S. for the March 1-7 timeframe. With the NAEFS, objective analogs, and composite pattern based on expected teleconnections all showing warmth in that part of North America, any cold shots will likely be short-lived and rather moderate. Severe cold outbreaks that have shown up in the extended range on some runs of the GFS (not the 2/23 0z run) appear unlikely given the increasingly strong AO+ (currently +2.275 and rising) that is now getting established. Similarly, guidance hinting at KU-type East Coast snowstorms is likely to be incorrect. No such storm has occurred with an AO of +2.4 or above.

[ORH_wxman]

Similarly, guidance hinting at KU-type East Coast snowstorms is likely to be incorrect. No such storm has occurred with an AO of +2.4 or above.

I agree Don...the more Nina-like pattern developing doesn't support a big KU type storm in the east...esp with a +AO....they mostly look like SWFEs like 2/24 and now 2/29-3/1 is trending that way which makes sense in this pattern...someone to the north of the Nina gradient will probably get some decent snow, but that gradient looks to develop over NY State and New England rather than the large cities...though BOS could conceivably pull out a few minor to moderate events if things break right.

[donsutherland1]

I agree, Will, about SWFEs being much more likely than a KU-type storm.

[winterymix]

I agree, Will, about SWFEs being much more likely than a KU-type storm.

G-morning, Don.

Way down here, south of the Mason-Dixon Line, we (with brown lawns) probably have a 90% to 95% likelihood of enjoying accumulating snow ten months from now, at the soonest. Before we turn all attention to Severe and then to Tropical seasons, I have a question for you:

Where would you turn for broad-brush guidance about the 2012-2013 Winter season?

If we look at Klaus Wolter's work:

http://www.esrl.noaa.gov/psd/enso/mei/

and scroll down to this chart:

Discussion and comparison of recent conditions with historic La Niña events

it suggests to me that the index will slowly trend toward the mean during the next few months. Based upon analogs of previous La ninas of the current strength, can you make an educated guess about the state of the MEI for the 2012-2013 winter? Wolter calls for a 40% continuation of La nina for next winter.

[donsutherland1]

I have a question for you:

Where would you turn for broad-brush guidance about the 2012-2013 Winter season?

If we look at Klaus Wolter's work:

http://www.esrl.noaa.gov/psd/enso/mei/

and scroll down to this chart:

Discussion and comparison of recent conditions with historic La Niña events

it suggests to me that the index will slowly trend toward the mean during the next few months. Based upon analogs of previous La ninas of the current strength, can you make an educated guess about the state of the MEI for the 2012-2013 winter? Wolter calls for a 40% continuation of La nina for next winter.

Winterymix,

Even a guess at the Winter 2012-13 ENSO is risky at this point. Model verification is poor from this point in time. For example, the mean error in the CFSv2 forecast is around 1.0°C for the Region 3.4 anomaly. Usually by late spring/early summer, the verification scores are better.

One might have some early insight from the trends in the SOI through May. If the April and May figures remain solidly positive, that could be an indicator that Winter 2012-13 would feature a third consecutive La Niña. If not, odds would probably favor a Neutral or El Niño winter.

For me, it's too soon to take a guess concerning Winter 2012-13's ENSO, much less the overall outcome. However, a neutral ENSO might provide the worst possible outcome for the Mid-Atlantic region.

[Isotherm]

Speaking of third year ENSO following double ninas, there were several such cases since 1950. Bolded is what happened in the third year.

1949-51: Nina

1951-52: Weak El Nino

1954-56: Nina

1956-57: Weak Nina

1970-72: Nina

1972-73: Strong El Nino

1973-75: Nina

1975-76: Strong Nina

1998-2000: Nina

2000-01: Weak Nina

2010-12: Nina

2012-13: ?

We've essentially seen the full spectrum of possibilities, from strong Nina to strong Nino for the third year following a double Nina.

US temp composite of those 3rd year events (not very useful in itself, but interesting to look at nonetheless).

All of those years except 2000-01 were warm/snow starved for most of the East (i.e., NYC recorded < 25" in those years save for 2000-01).

Obviously it's a small sample size, and ENSO state was widely different from year to year. NAO/AO values tended to be positive in the warm winters, whereas only 2000-01 featured a neg NAO of the 5 seasons mentioned. 1951-52 was easily the worst weak el nino (snow wise and cold wise) for much of the Eastern US since 1950. Generally weak ninos are good winters for the Northeast, but there are exceptions. Based on 3rd year seasons alone, the data is pessimistic concerning a much better winter next year in the East; however, as noted, the sample size is much too small to make a conclusion. If we can get a weak Nino going, that'd be the most desirable scenario for the entire Eastern seaboard. If we remain in a LA NINA or neutral ENSO, chances are the Southeast and Mid atlantic will see another bad winter.

[HM]

I agree to go a bit slower with the MJO than Roundy here. What becomes interesting is then there would be a phase 4-6 potential during mid-March. This isn't quite the blowtorch signal as it is in January, esp. if you can continue to press the forcing eastward and allow the GWO to orbit from 3-4 to 5-6. This would shorten the waves and allow for a cyclone to break along the East Coast. The more time the MJO delays its arrival to these phases, the better potential for this and possibly a late-season wet snowstorm for the Northeast.

[HM]

I agree to go a bit slower with the MJO than Roundy here. What becomes interesting is then there would be a phase 4-6 potential during mid-March. This isn't quite the blowtorch signal as it is in January, esp. if you can continue to press the forcing eastward and allow the GWO to orbit from 3-4 to 5-6. This would shorten the waves and allow for a cyclone to break along the East Coast. The more time the MJO delays its arrival to these phases, the better potential for this and possibly a late-season wet snowstorm for the Northeast.

When I say this isn't quite the blowtorch signal, I was talking strictly along the East Coast, especially later in the month.

[Isotherm]

I agree to go a bit slower with the MJO than Roundy here. What becomes interesting is then there would be a phase 4-6 potential during mid-March. This isn't quite the blowtorch signal as it is in January, esp. if you can continue to press the forcing eastward and allow the GWO to orbit from 3-4 to 5-6. This would shorten the waves and allow for a cyclone to break along the East Coast. The more time the MJO delays its arrival to these phases, the better potential for this and possibly a late-season wet snowstorm for the Northeast.

Good points. March is quite a bit different than NDJF in terms of sensible wx response to the MJO in the CONUS. Phase 4-5 are coolish but certainly far from torch in the Northeast. But the problem is the MJO seems to be one its own, at least for now, in terms of favorable teleconnectors. Phase 2-3 right now should be a colder one across the Eastern US, but in fact its warmer than normal due to the low height field across the nern latitudes once again. As long as we have a positive AO and NAO persisting into mid March, I don't have much hope for significant snow in I-95. I think the interior of the Northeast may be in for quite a ride though over the coming month.

Euro ensembles indicate phase 5/6 by March 15-20th. If the MJO wave continues strong, and if we can get the AO/NAO to change, the end of March or early April might offer something when MJO circulates into more conducive phases again. A lot of IFs there though.

[donsutherland1]

In parts of the Southeast, February is ending in memorable fashion. Today, a number of cities tied or set new record high temperatures. Some highlights included:

Charleston, SC: 81° (tied daily record)

Charlotte: 76° (tied daily record)

Savannah: 84° (set new daily record)

Shreveport: 87° (set new daily record) -- just missed monthly record of 89°

[QVectorman]

I'm definitely concerned about busting the period post March 7. There is quite a large difference between the Roundy MJO guidance (which shows the MJO zipping through the phase space, getting into P7 by March 11 and P1 by the 18th) and the Euro Weekly Ensemble (which shows the MJO in P5 and P6 at those same periods, respectively). The Roundy guidance would bring cold back to the East much more quickly than the Euro, which shows very warm temps through its Week 3 and Week 4 periods.

Adam can you briefly explain why Roundy's PS diagram is so dampened when it comes to the SD's or direct me to some literature. I assume it's because he uses the EEOF PCs....which I have no clue what that it. Have been looking at it for the past few weeks and his definition of 1 SD is completely different and larger than the the normal PS diagram's definition of 1 SD. On his the MJO hardly ever makes it out of the COD. Thanks!

[am19psu]

Adam can you briefly explain why Roundy's PS diagram is so dampened when it comes to the SD's or direct me to some literature. I assume it's because he uses the EEOF PCs....which I have no clue what that it. Have been looking at it for the past few weeks and his definition of 1 SD is completely different and larger than the the normal PS diagram's definition of 1 SD. On his the MJO hardly ever makes it out of the COD. Thanks!

Yeah, it just has to do with the way that the EEOFs are constructed. If you're not familiar with higher level stats, I don't really know how to explain EEOFs. Basically think of them as the atmospheric patterns that explain the most variance in the atmosphere while being linearly independent of each other. They explain a lot, but not all of the variance, of the OLR pattern, so it makes sense that their magnitude should be lower (even much lower) than what actually occurs.

FWIW, I wouldn't use the PS products. The Hovmollers are much more instructive.

[Bob Chill]

Don, I think you saw my post in the MA subforum about spring temps using analogs based on current enso. It was a cold march look and after putting it together I started thinking about other factors because it was pretty obvious that it was likely wrong. I made a new post based on similar NAO analogs and disregarded enso completely. March temp map looked pretty much like what everybody is calling for.

That led me to some additional questions about things I don't have the knowledge to understand. Looking back on this winter there are a couple of things that stood out to me. On paper, the mid-Jan to mid-Feb timeframe had a -ao/+pna look. It gives the appearance that we did have the much anticipated "mini-winter" period in the east but we all know that is not the case. I'm still working on the data analysis about the nao/ao relationship but it's pretty uncommon to have such an anomalous -AO and have the NAO remain positive. It's more common when the AO is -1.5 or higher but once you get down to -2.0 and lower, the NAO typically in negative territory.

I only understand MJO basics so this is a simple observation. We did finally get a good run through 8-1-2 but the typical cold east coast response didn't happen. It appears that the stubborn +NAO trumped any chance at delivering the goods here.

This may have been touched on already in this thread, so point me there if it has but I'm interested in understanding what the basic root cause for having such a stubborn +NAO this winter. Unless I'm missing something else that is obvious, it seems to me that the NAO trumped everything this year and for lack of a better term, it cancelled our winter.

I found some decent NAO analogs. The December temp composite is a pretty good match but Jan-Feb show normal to slightly above in the east so they aren't all that great. These are the years that seemed to match up well with the NAO:

1982-83

1993-94

1994-95

1999-00

2004-05

[donsutherland1]

Bob,

Three quick things:

1. ENSO, the teleconnections (AO, PNA, NAO, EPO), MJO, etc. all have an impact in shaping the pattern. Sometimes one or a few have a disproportionate influence. December 2011 was a case in point where the extreme AO+ regime overwhelmed all the other factors.

2. Persistence in one of the factors can sometimes compensate for a lack of magnitude. For example, the MJO remained locked in Phases 4-6 through much of December and January and part of February, often in the so-called "circle of death." The earlie AO+ regime also amplified the warmth. As a result, the outcome has been a warm one in the East despite the PNA+/AO- period setup that existed from late January into mid-February.

3. There can be a lag before a factor exerts influence. The MJO had raced through Phases 8 and 1 in late February. The atmosphere did not yet have time to respond. In the meantime, the AO rose to > +2. As a result, the outcome was a warm one in the East during the PNA+/AO- period setup that existed from late January into mid-February and has remained, to date, generally warm despite the more favorable MJO phases.

Anyway, those are some thoughts as to what might have occurred.