The Seasonal Forecast Thread

[Roger J Smith]

Just to be clear, I did my sunspot-temperature analysis based on the generally accepted peak years which can be found in any publication about solar activity and I am aware of the conventional sunspot number system and even how it is derived.

The exact timing of cycles and peaks can be anyone's subjective opinion and so anyone would be free to analyze the data in any system they wanted to use, within reason. My analysis is not subjective, it is based on the published peak years and that is the only significance that it has (if any), if there may be a more significant periodicity more approximately aligned to the sunspot cycle that is still waiting for anyone else to demonstrate and explain.

The average period in quiet solar intervals seems to lengthen to 12-14 years (the Maunder was too inactive to be very certain but the Dalton displayed an average of 14.5 years from 1801 to 1830) and the average period in active periods is only slightly over 10 years. For example, from 1917 to 2001 it was 10.5, from 1917 to 1989 it was 10.3 and from 1917 to 1957 it was exactly 10 and the only peak year that didn't end in 7 in that run was 1928.

From 1837 to 1870 it was 10.75 years, from 1718 to 1787 it was a shade under 10 years. The period 1761 to 1778 was exceptionally "fast" with peaks in 1761, 1769 and 1778. However, as implied, these exact peak years are approximate within a 3-5 year active peak period and it's really frequently a case of a nearly flat-top peak where you could choose any one of three consecutive years for your peak.

Given how quiet the Sun has been recently, the next "peak" may be delayed well into the next decade and could perhaps be as late as 2026 although most authorities are predicting 2024.

 

[Guest]

Maybe the sun is trying to save Earth by cooling down a bit? hehe, sorry.

[Roger J Smith]

This is a short validation report on my summer forecast published in this thread on 16th of May. 

The temperature forecast was generally good and somewhat conservative. I gave a CET of 15.9 overall, and reality was 16.4 C.

The summer was expected to rank 79th warmest out of 358 all time, and ranked 39th. Of the more recent summers (since 1949) the prediction ranked 19th out of 68, but reality was 12th warmest.

The monthly estimates were 14.8 (0.5 below actual), 16.8 (0.1 below) and 16.2 (0.8 below) for an average monthly error of 0.5 C, in each case somewhere between 1981-2010 normal values and what was actually recorded. The forecast began to falter more towards the end of August which was a very warm spell for most. 

Comments about the anomaly pattern being such that northern areas would not be as relatively cool compared to southern areas probably did not show much skill, the higher anomalies were in the southeast but it's probably true that parts of the southwest came in relatively closer to average. 

Predicted warm spells ran earlier than what actually happened, if I had gone with a two week lag it would have verified. Also, I said that the pattern might cough up one very warm day like 2015 did, and that certainly happened (record high 23.6 on 20th July).

The predictive discussion of precip appears to be fairly good, perhaps some readers could help me out on this, but I went with the idea of a wet summer in parts of the south trending to dry in parts of Scotland. From what I've seen and read on line, the actual patterns were more complicated but there were certainly very wet spells in the south and as mentioned in the discussion, in northern France which saw severe flooding in June. Mid-July and late August were given as particularly wet possible periods, not sure if these stood out against other candidates or not.

The prediction was based on a muted blocking pattern with occasional weak zonal periods, and this is about how the circulation behaved, there were frequent spells of southerly winds making their best impact in counties east of a Hampshire to south Yorkshire line, and I was probably expecting perhaps more southeasterly component at times, than what actually happened, but the weak zonal periods that did develop were about what I had in mind. 

In North America, what actually happened was a widespread warm anomaly with a trend to cooler anomalies in the west towards the end of the season, and a displacement of the core of heat from the west-central states early in the season to the east coast by August. The hurricane seasonal forecast is ongoing and seems en route to some validity with one hurricane landfall already, an event not seen for several years of general sparse tropical activity. 

There was even a comment in the forecast that if it had a bias it was conservative and the expected error was towards warmer temperatures. So on that basis, I am reasonably pleased with the outcome of this forecast. 

Working on a winter forecast, I figure the autumn will continue to be generally mild and occasionally stormy, at least to the last third of November (nudge, nudge, wink, wink). 

In the longer term, some years with summers of very similar mean CET to 2016 include 1775, 1783, and 1794. All of the winters that followed those three were exceptionally cold. So this is a good sign perhaps, not sure what my research is going to tell me during the coming month, back in October to reveal my 2016-17 winter forecast.

[BLAST FROM THE PAST]

Thanks Roger.  For me the autumn pattern is set up for repeats of warm to very warm bursts and generally more settled than not but with one or two big interruptions from...although I think main talking point will be the warmth.

Re winter, early thoughts are for start of winter as I think a dramatic switch in Dec  A general cool /cold stormy NW'ly flow month.  I'm loving this late warmth....golf pilgrimage is starting as I rapidly approach retirement 

 

 

BFTP 

[chrisbell-nottheweatherman]

May I enquire as to how the following may affect seasonal forecasting over the next few months?

http://www.sciencemag.org/news/2016/09/unprecedented-disruption-atmospheres-pacemaker-foretells-wet-winter-europe

[Roger J Smith]

Winter 2016-17 outlook

My research model is indicating that the first half of the winter is likely to be colder than normal, and the second half milder. CET forecasts follow:

OCT 12.8 (very mild)

NOV 6.0 (slightly cooler than recent averages)

DEC 3.8 (somewhat colder than normal)

JAN 5.5 (rather mild, second half may be near 7.5)

FEB 4.5 (average).

MAR 8.0 (mild/warm)

In terms of the general circulation, I am expecting rather strong blocking to develop through November after perhaps a rather stormy first half, and spells of wintry cold may develop from late November through mid-December in particular, then this block will rapidly decay around New Years to yield a much milder zonal and sometimes stormy pattern in January. That will then relax to rather bland patterns in February and a southerly type of blocking by March. 

Best chance of snowfall would obviously be mid-December, and with a full moon / northern max event 14-15 Dec, that's when I would expect to see possible snowfall events. I think that snow would be followed by the coldest weather of the season then some gradually milder conditions through the Christmas-New Years period. 

The most likely time for stormy conditions to develop would be late December, mid January and late January around the times of the new moons and mid-January full moon. This theory is partly established by frequency studies but also a second unrelated factor happens to overlap this winter with the lunar events. 

February may turn out rather dry if this overall scenario is valid, with storminess more likely in the Mediterranean and Balkans. 

In North America, the winter is likely to be quite cold in general, with heavy snowfalls likely for the east coast of the United States and also in the Midwest states. 

All forecasts are based on numerical index values from research factors and there is an assumption that teleconnections will be incorporated. However, with the Pacific in a sort of weak La Nina or neutral pattern, no strong indications are available there. In the North Atlantic, the tendency has been for a cold pool to be focused more to the west-central longitudes and this at least can reinforce both the blocking scenario and the active phase mid-winter. I would expect that SST anomalies around Britain might fluctuate more as a response than a driver of these trends. Solar activity of course is rather low but I only see correlations over long-period averages rather than year to year. 

Confidence is somewhat lower in this outlook than some others I have published, for the reason that the blended index values are in two camps during both the colder and milder episodes. What I'm saying there is that the cold may be more extreme than indicated, and the mild spell could be delayed as a result. I have given some rather conservative values for the temperature trend as a result, the Dec 3.8 could turn out as low as 2.5 if snow cover persists. 

(re the previous post and question, I will have to study this more before commenting)

[BLAST FROM THE PAST]

Hi Roger

For me is the CET......Dec could be pretty extreme for some but the CET may not reflect this....same as the rest of winter IMO

 

BFTP

[damianslaw]

Interesting reading Roger.. I'm a bit of a believer that mother nature balances herself out, i.e. very warm or very wet periods one year are likely to be preceded or followed by very cold or very dry periods. Nov 09 and Nov 10 good examples, an exceptionally wet very mild Nov followed by a very cold second half November. March 12 very warm and dry , followed by the very cold March of 2013.

So with this logic the exceptionally very wet and very mild Nov and Dec last year, may be followed by the complete opposite! i.e. extremes follow extremes, and last Nov and more so December around here was exceptionally extreme, as extreme as it can be for rainfall and mild weather, I really wouldn't be surprised to see something much colder and drier to end this year.

 

[Easily sled]

Accuweather winter forecast is out for Europe - not pretty reading for the UK and Ireland. Hopefully it doesn't go to plan....

http://www.accuweather.com/en/weather-news/europe-winter-forecast-2016-2017-stormy-uk-mild-wet-france-germany/60378723

 

 

[BLAST FROM THE PAST]

I think one can say looking at recent winters that more of the same is coming.  Well if more of the same of last winter comes this time one must ask what is controlling our seasons.  We probably shouldn't experience the same winter.....but I suspect we are going to see some sharp contrasts across our shores again.  

 

BFTP

[Cheshire Freeze]

4 hours ago, Aljoca said:

Accuweather winter forecast is out for Europe - not pretty reading for the UK and Ireland. Hopefully it doesn't go to plan....

http://www.accuweather.com/en/weather-news/europe-winter-forecast-2016-2017-stormy-uk-mild-wet-france-germany/60378723

 

 

'A stormy weather pattern will set in before wintertime with active weather during the second half of October across Ireland and the United Kingdom.'

A quote from that piece.

Right well we'll see how that one goes because if that goes awry I won't have much confidence in the rest of the forecast period!

[Walsall Wood Snow]

On 06/10/2016 at 20:52, BLAST FROM THE PAST said:

I think one can say looking at recent winters that more of the same is coming.  Well if more of the same of last winter comes this time one must ask what is controlling our seasons.  We probably shouldn't experience the same winter.....but I suspect we are going to see some sharp contrasts across our shores again.  

 

BFTP

When you say recent Winters are you referring only to the most recent 3 Winters, or say are the Winters from 2008 to 2013 included in that statement? I hope so because that at least would give us a range of options going forward, whereas if you're only referring to the Winters we've had so far since 2013 the prospects looks pretty dire from a coldies perspective. You've long been the eternal optimist though when it comes to cold Winter weather so the idea of you losing faith is a sobering one to say the least.

[BLAST FROM THE PAST]

WWS I'm musing.....I think there'll be decent shots no fear!!

 

BFTP

[Roger J Smith]

I promised to comment on the article that was linked, in my own case it doesn't affect my outlook at all at this stage, I am keeping my eyes open for learned discussion of this topic over the winter and a post-mortem on it. 

As my forecast includes spells of blocking and zonal flow, I think the only way the findings discussed in the article might affect my thinking would be to swing the balance so that one spell was more significant than the other. This is not really a finding that I can easily plug into my methodology, it's a bit too esoteric. If somebody found that the Sun was running at 80% of normal output, that I could plug in. 

It may be a very significant thing or this year's version of the October failproof predictive index (how's that looking now?). 

One factoid worth mentioning is that we're about to have the biggest mid-October windstorm on the west coast of North America since the 12 October 1962 "Columbus Day storm" of great renown. Once again, it's a remnant of a Pacific tropical system being reborn in fast westerly flow near the Oregon coast. I shouldn't tease people like this, finding an analogue in autumn 1962 is like being invited into the blonde's flat (remember that image?). 

[BLAST FROM THE PAST]

And the end result of getting nowhere after the invite.  

 

BFTP

[Summer Sun]

On 06/10/2016 at 21:15, CreweCold said:

'A stormy weather pattern will set in before wintertime with active weather during the second half of October across Ireland and the United Kingdom.'

A quote from that piece.

Right well we'll see how that one goes because if that goes awry I won't have much confidence in the rest of the forecast period!

 

 

Doesn't look stormy for the 2nd half of the month a very quiet Atlantic so far as we near the halfway point

Edited October 14, 2016 by Summer Sun

[chrisbell-nottheweatherman]

12 hours ago, Roger J Smith said:

I promised to comment on the article that was linked, in my own case it doesn't affect my outlook at all at this stage, I am keeping my eyes open for learned discussion of this topic over the winter and a post-mortem on it. 

As my forecast includes spells of blocking and zonal flow, I think the only way the findings discussed in the article might affect my thinking would be to swing the balance so that one spell was more significant than the other. This is not really a finding that I can easily plug into my methodology, it's a bit too esoteric. If somebody found that the Sun was running at 80% of normal output, that I could plug in. 

It may be a very significant thing or this year's version of the October failproof predictive index (how's that looking now?). 

One factoid worth mentioning is that we're about to have the biggest mid-October windstorm on the west coast of North America since the 12 October 1962 "Columbus Day storm" of great renown. Once again, it's a remnant of a Pacific tropical system being reborn in fast westerly flow near the Oregon coast. I shouldn't tease people like this, finding an analogue in autumn 1962 is like being invited into the blonde's flat (remember that image?). 

I've thanked Roger privately, but thought I'd post here as well to say that I thank him for his response.  It sounds quite reasonable to me to say that we cannot predict what effect it might have, and the Guardian headline ("The die is cast for a wet and stormy winter") that made me read to Science magazine article seems a massive over-simplification.

[Roger J Smith]

14 hours ago, BLAST FROM THE PAST said:

And the end result of getting nowhere after the invite.  

 

BFTP

Made myself a key though ... 

[BLAST FROM THE PAST]

1 hour ago, Roger J Smith said:

Made myself a key though ... 

You've had that for a while me thinks ;-)

 

BFTP

[The PIT]

On 06/10/2016 at 16:34, Aljoca said:

Accuweather winter forecast is out for Europe - not pretty reading for the UK and Ireland. Hopefully it doesn't go to plan....

http://www.accuweather.com/en/weather-news/europe-winter-forecast-2016-2017-stormy-uk-mild-wet-france-germany/60378723

 

 

Already wrong so can be slung in the recycle bin.

[Dancerwithwings]

1 hour ago, The PIT said:

Already wrong so can be slung in the recycle bin.

"Stormy second half to October" yeah right! Mainly an easterly for most of this month. Umbrella's not needed.

Edited October 22, 2016 by Dancerwithwings

[chrisbell-nottheweatherman]

On 22/10/2016 at 13:55, The PIT said:

Already wrong so can be slung in the recycle bin.

Not sure about the recycling bin, but it isn't going the way they thought as yet...

[Roger J Smith]

Reasoning behind a winter forecast may be fairly insulated from October outcomes, one could compare it to a game vs the warmup portion, more or less irrelevant once stronger counter-forces are put in play. Once November or December go bad though, this creates growing doubt about how the heart of the winter season (Jan-Feb in most instances) might turn out. 

There's a general blocked and low-energy feel to things that cannot help but create some optimism for this coming winter (in Europe and eastern North America, feeling somewhat doomed to the mild side of any blocking here). 

[LomondSnowstorm]

Hi folks, been a while since I've been on here. I've linked to my winter forecast here:

http://lomondsnowstorm.blogspot.se

On the tablet at the moment but I'll  try and copy it over to this post when I get  chance. 

[Ravelin]

26 minutes ago, LomondSnowstorm said:

Hi folks, been a while since I've been on here. I've linked to my winter forecast here:

http://lomondsnowstorm.blogspot.se

On the tablet at the moment but I'll  try and copy it over to this post when I get  chance. 

Thanks, very interesting and a forecast I'd not be upset about if it comes to pass. Just go and sort out Feb though.