Winter 2022-23 - General Discussion

My first stab at this challenge has been to identify best analogues based on (a) recent trends and (b) presumed external drivers. That has yielded a collection of mostly mid-table winters that do not stand out for either mild or cold outcomes, but do include perhaps a few more cold ones than mild ones. From all that, and bearing in mind the recent steady upward trend in winter mean temperatures, the most logical call would be a slightly milder winter relative to 1991-2020 "normals" but with a good chance of some significant cold spells. The analogue set also looks a bit drier than average suggesting a combination of anticyclonic influences and blocking when colder spells do develop. I could see a pattern of cold being at both ends of those analogue winters and for January to trend milder than average.

For what it's worth, the three years judged to be running closest to 2022 were 1868, 1911 and 1995. These are listed in chronological order rather than implying a rank order, their similarities were more or less equal. Of those, 1868 was followed by the mildest (CET) winter on record, while 1911 had a fairly cold winter (Feb 1912 was well below average) and 1995 had a rather average sort of winter with mixed intervals. I am watching how these three analogues fare during the autumn, but so far all three of them had similar outcomes to this year in September (started warm, turned coolish mid-month). Also 1825 had broad similarities, and by Dalton era standards, the following winter was not particularly cold.

I then looked at North American data to see if there was any reason to favour one of these analogues, but in fact all of them were quite similar to this past summer (they all produced significant heat in July and August after rather bland Junes). The winters that followed in eastern North America were near average for 1868-69 (especially if the surrounding 30-year average was used to assess), very mild to very cold in 1911-1912, and snowy with some cold spells for 1995-96. (1825-26 had less data available, what there was seemed rather ordinary compared to some other years in that decade).

Anyway, I did read through this thread before commenting and found that the general consensus here for whatever reason seemed to be similar to what this analogue technique has suggested, there are more voters for relatively mild and dry than other outcomes in the thread, some hopes for limited cold making an appearance, and doubts about this being a 2013-14 type mild and wet stormy winter. I don't see any reason to dissent from this at this early stage. Will return to the question in mid-October which is my usual time to hazard a final guess.

That's exactly it, as much as I like the snow, with bills going up and no sign of coming down anytime soon, I'd rather have a boring mild winter with no storms till March, although i did think of getting a wind mill to generate power but if no Storms happen then that's a waste of money

How much does the temperature outside actually impact heating bills? Heating is required every day in Ireland from at least Nov -> Mar generally anyway. Does whether it's 4 degrees or 9 degrees outside really change much? Genuine question here.

I'm surprised there hasn't been more public awareness about energy saving leading up to winter, you'd be surprised how many people don't know the basics about reducing their consumption with minimal impact to their lifestyle.

we have just about the mildest winters in Europe as it is so with the exception of the elderly who have difficulty keeping warm I can’t see much hardship for the healthy this winter.

The people living in Athens for example will feel the cold more than us.

Sure is borderline yeah, particularly remember that January day in the Dublin mountains earlier this year, when everything above around 250 metres was a winter wonderland (knee-deep snow in places) with almost blizzard conditions on the summit of Two Rock Mountain, yet there was absolutely nothing in the city and suburbs. If only it was 2 degrees colder!

I think eventually somebody will stumble upon a method that works reasonably well, of course some claim to be there or near there now, I think some workers in this field are somewhat better than random guesswork but I haven't seen any real evidence of any near-perfect skill at this task, if such a person existed they would probably be scoring a lot higher than even the best contest entrants (and that time scale is presumably easier to crack being basically half a month that requires a forecast and half that already has something of a reliable forecast).

Then I wonder if that were to happen, what would become of the weather enthusiast field in general? There's almost nobody regularly posting on the astronomy forum in part because predictions there almost always come to pass.

When I say "works reasonably well" or "near perfect skill" I am probably talking about fairly small tolerances, the complexity of weather outcomes is always going to create a challenge to hit perfect accuracy and reasonable skill would probably look something like the 24-48 hour time range looks now.

My own search for answers has certainly convinced me of one thing, any reliable method will be fairly complicated and will need a component of response to short-term human-caused factors that might interfere with the reliability of anything purely statistical. Maybe that would look like this -- if you had a reliable model that showed you a +0.3 anomaly, you might adjust that to +0.7 if you thought part of the input came from cases with significantly less ambient greenhouse gas. But I am fairly sure it would work in a more complex way than that. All of which is rather frustrating because you cannot necessarily tell whether a statistical approach is as meaningless as it might appear, if the results need to be reconfigured to match the new atmospheric chemistry.

Progress towards an outcome could be relatively quick or it might be a case of working through certain stages. Since we don't know what philosophy will actually produce this outcome, we can't really say in advance whether progress might be all at once or in stages. My hunch is that it would be in stages. Something might begin to separate itself from other techniques and study of its diminishing errors could make the last stages easier to attain. I suspect this could happen in this century. When I first started doing this sort of research, we did not all have personal computers, access to all the data that is now possible, and the speed of calculation that, if nothing else, can rule out non-working paradigms in a matter of hours rather than months or years. The same level of intellect applied in today's situation is almost orders of magnitude better supported, so if a result could be had, the chances of it happening will continue to improve (the growing numbers of people interested also tends to up the chances). Once upon a time, only a handful of people were looking for the planet Neptune, and two of them succeeded independently in finding it. Imagine if ten thousand people are looking for this kind of solution. As I said to a relative who didn't venture to contradict me, they can't all be as stupid as me.