Anyone who has read the newspapers lately can’t have failed to notice this winter’s weather is in the headlines. Justification for claims of a ‘big freeze’ has come from sources as diverse as the plucky Bewick Swan settling into the comfort of the WWT reserve at Slimbridge, Gloucestershire earlier than ever before, to the strong El Niño and cool North Atlantic Ocean.
But what can we genuinely say about prospects for the coming winter, and what is the influence from phenomena like El Niño? Jeff Knight, from the Met Office Monthly to Decadal Prediction team explains.
In the Met Office we produce outlooks for the UK weather as a whole over three monthly periods. These outlooks are not forecasts in the conventional sense, although they are still made using computer prediction models. While a forecast might say ‘it will rain tomorrow’, the chaotic nature of the atmosphere beyond a few days ahead leads to growing forecast uncertainty, making it meaningless to try to make the same kind of forecast for a day in three months’ time.
Fortunately, atmospheric chaos is only part of the story and, when we consider the broad characteristics of the weather over a three month period, we can see influences from a range of global climate factors that we can endeavor to predict. While the unpredictable part means there is always a range of possible outcomes, the part we can try to predict allows us the opportunity to identify which types of weather are more likely than others. As a result, our outlooks are more useful for professionals who need to assess risk, such as contingency planners, than the public generally. Our current outlook covers the period from November to January.
So what are the global drivers that might influence our weather this winter?
El Niño is the biggest news story currently in global climate. This episodic warming of the waters of the equatorial Pacific Ocean occurs every few years – the last event happened in 2009-10. This ocean warming covers an area about 1,000 km wide and 13,000 km long, stretching along the equator from the South American coast to the West Pacific. El Niño events release a vast quantity of oceanic heat into the atmosphere so it is not surprising that El Niño has effects on weather across the globe.
This year’s El Niño started to grow in April and it has now become a strong, mature event similar to the landmark 1997-8 event. Typically, growth will peak around the end of the year and decline during the first half of the following year. We have already seen its effect on global weather systems: this summer’s Indian monsoon rainfall fell to drought levels and very hot, dry conditions in Indonesia have contributed to widespread forest fires.
Currently, the outlook for El Niño is for further growth over the next two months. Events are often ranked in terms of sea surface temperatures in Central Pacific, and by this measure, this year’s El Niño is more likely than not to become the strongest on record. Temperatures further east near to South America are likely to be not quite as exceptional as in 1997-8. No two El Niños are identical and even very similar events have slightly different characteristics.
What does El Niño imply for the UK this winter?
Unlike some parts of the world, the effect of El Niño on Europe is relatively subtle. In El Niño years there is a tendency for early winter to be warmer and wetter than usual and late winter to be colder and drier. Despite this, it is just one of the factors that influence our winters, so other influences can overwhelm this signal – it is relatively straightforward, for example, to find years where these general trends were not followed.
What about the Atlantic Ocean?
Closer to home, sea surface temperatures to the west of the UK have been notably lower-than-average in recent months. While it is true the westerly winds that we typically get in winter would have to pass over this region, it is unlikely that this will directly have a strong bearing on expected temperatures. This is because temperatures at this time of year are strongly affected by the direction of the wind. Eastern Europe and Scandinavia are 10-20°C colder than the Atlantic Ocean in winter, so our weather will depend much more on how often winds blow in from the north and east than whether the Atlantic is 1-2°C cooler than usual.
More broadly within the North Atlantic Ocean, sub-tropical temperatures to the south of this cool region are widely above average. This combination results in an increased north-south temperature gradient, which is expected to provide greater impetus for Atlantic depressions. For the UK, this would favour relatively mild, unsettled weather conditions.
Our weather is also affected by changes in the stratosphere
European winters are also sensitive to what is happening in the stratosphere, the layer of the atmosphere between 10 and 50 km up that lies above the weather. The equatorial stratosphere is home to the Quasi-Biennial Oscillation (QBO), a cycle that sees winds switch from easterly to westerly and back roughly every 27 months. First noted by Met Office scientists over 40 years ago, the link with European winter weather has stood the test of time. This year, the QBO is in a westerly phase, which implies an increased chance of a mild and wet winter at the surface.
A considerable part of the year-to-year differences between UK winters is related to the occurrence of sudden stratospheric warmings (SSWs). In these events, the polar stratospheric vortex – the fast moving circulation of stratospheric air that whirls around the North Pole in winter – abruptly breaks down. They occur one winter in two on average, and events are most common in January or February. In the majority of cases SSWs lead to the establishment of cold easterly flow at the surface across Europe and the UK. The last SSW was in January 2013, and this event contributed to the cold late winter and early spring in that year.
Whether we get an SSW or not depends on a number of influences, such as El Niño and the QBO. Currently our models suggest an increased likelihood of an SSW from January onwards. If this were to happen, its effects would not be felt much before the end of our November to January outlook period. At the moment, therefore, this is still a long way off, and we consider this suggestion to be tentative.
So what can we expect in the UK this winter?
Most of the global drivers discussed above tend to increase the chances of westerly weather patterns during our November to January outlook period. Our numerical prediction model, being sensitive to these drivers, also predicts a higher-than-normal chance of westerly conditions. This results in an outlook for an increased chance of milder- and wetter-than-usual conditions, and a decreased chance of colder and drier conditions, for the UK. Our outlook also indicates an increase in the risk of windy or even stormy weather.
It should be noted that these shifts in probability do not rule out the less favoured types of weather completely. Also, a general tendency for one type of weather over the three months as a whole does not preclude shorter spells of other types of weather.
Finally, there are hints that the outlook might be rather different in the late winter, with an increased risk of cold weather developing. Nevertheless, it is currently too early to be confident about this signal.
So to sum up, we have no idea what the weather is going to be, so our forecast will be right whatever happens.
I just hope our contingency planners are not going to base any plans on this vague prediction otherwise they could end up seriously embarrassed.
As it says in this post, noone can say what the weather is going to be because of the fundamental physics of chaos.
At least these guys have worked out what can be said, even if its more limited. They don’t revel in being totally in the dark, as you seem to do.
I think you’ve missed the point. This forecast is supposed to assist contingency planners. A look at previous similar forecasts have proved to be anything but helpful to contingency planners. In fact a few years ago our local council publically blamed the Met Office forecasts for the lack of salt when a harsh winter was predicted to be milder than average.
Not at all. If you have to make these plans then you’d be mad to bet your shirt on it going one way or another. You just shift your position appropriately in response to the altered likelihood, just like planners make adjustments in response to all sorts of information of varying degrees of certainty. Its no different.
“If you had to make these plans then you’d be mad to bet your shirt on it going one way or the other”
Precisely, and that’s exactly why these vague forecasts are useless for contingency planning.
False strawman – a proper planner won’t ever bet their shirt one way or another for any reason.
Exactly, and that’s why these vague forecasts are useless to the contingency planners they’re aimed at.
Very interesting and astonishingly complex.
There was an amateur weather forecasters from Thirsk called Bill Foggitt who sadly died in 2004 . He made weather predictions based on observing the behaviour of birds, moles, flies, pine cones and seaweed and such like.
The Met Office long range forecasts reminds me of the value of forecasts that he made – very little!
Yes we have moved on from seaweed and pine cones, the Met Office now look to sophisticated science such as El Niño-Southern Oscillation, sudden stratospheric warming and of course the Quasi-Biennial Oscillation, but they are of no more use than noticing that the rooks are flying in circles round the village church.
It’s not that they are always wrong, occasionally they could be correct, but because they are so unspecific as regards the severity of the event, areas affected, and of course the most important thing when will they occur, to me they can be of little value.
Once again, it says in this thing that the forecasts are for specialists who can use it rather than curious amateurs like you an I who will always want more detail. Just because we can see the limitations doesn’t mean its useless for everyone.
I’m just happy that they share their thinking even though the product is not really for us. It is always good to hear what knowledgeable people have to say.
Exactly what do you suggest the average local authority or “specialist” as you call them should take from this forecast which would assist them in their contingency plans?
I worked for the Met Office for over 30 years as an observer and programmer and although I am not expert on all things meteorological, I have always had an interest in weather and climate and would describe myself has a couple of notches up from a “curious amateur”. Have a look at some of the many subjects that I regularly cover in this blog as a curious amateur and hopefully you will get the idea. As for forecasters, I have worked with some good ones, but I have also come across many that are poor and don’t give a damn for weather!
Oh dear, I have upset you by inadvertently demeaning your expertise. I’m sorry – I am aware of your blog and its very interesting. All I meant to imply was that our interest is more about our own local and extreme weather rather than what the Met Office is offering here.
As far as I know the Met Office forecast for contingency planners for November till January has not been issued yet.
Having now read the post, I see they are now considering as far ahead as January (I thought they would do that in early November).
Click to access A3_plots-temp-NDJ_v2.pdf
Click to access A3_plots-precip-NDJ_v2.pdf
Can you tell what temperatures will be like in January let alone for a particular week in January from any of those graphs? It’s just as I said, no better than pine cones and a complete waste of time.
Why can’t we accept the fact and stop wasting time and money chasing the holy grail that is seasonal forecasting?
The best that we will ever get are very broad brush trends of what a coming season will bring and if those graphs are anything to go by they have very little value.
If you verify NWP forecasts like I do, you will find there is very little or no precision after T+120, and until they crack that problem (which to me is more pressing and even more elusive problem than even seasonal forecasting) they shouldn’t bother publishing what they find, because it illustrates more what they don’t know than what they do.
It would be great to know that but why is this your measure of success? You are right, if conditions for each week this winter could be accurately predicted it would have immense value. They can’t, but according to this product some lesser level of skill is available, with proportionately lower value. What is not clear is why you assert there is no value.
How do you know what people out there are able to use this level of capability? There must be some large businesses e.g. transport with nationwide interests that are sensitive to how severe the winter is on average. Large companies have to take positions about the future based on all sorts of uncertain information – I don’t see how using a seasonal forecast is any different.
I agree. Our ability to predict weather ahead of a few days is so poor that it is absolutely pointless to try and make predictions about a season. It bugs me when we see silly front page news stories like “Britian set for arctic freeze!!!” it’s an opportunity for the media to sell their papers, and nothing more. I find myself actually insulted by it. Obviously the media think I’m too stupid to know that these predictions are useless, and that I’ll be so taken in by their nonsense that I’ll waste my hard earned money on their paper. Not going to happen.
Still, I must admit I found this article interesting. It’s clear that we can’t make any meaningful predictions from it, but I was intrigued to read about the various phenomena that affects our weather, and the relative strengths of influence of these phenomena.
Regarding the problem of it highlighting what we don’t know more than what we do, I would personally regard a scenario in which we at least know what we don’t know as a good starting point 🙂
Whether correct or not, the Met Office are suggesting that overall the period up to January is more likely to be milder than average than colder than average. They are not saying ‘nothing’ nor saying ‘everything’ (in fact that would still be the case if they suggested the most likely scenario was temperatures – or rainfall – very close to average). But they cannot be dogmatic that far ahead – nor can any scientist – though some amateurs issue public forecasts attempting to do so (usually they predict freezes and seldom if ever predict mild or ‘normal’ conditions).
Reassuring. To read some of the less reliable newspapers you’d think we will be experiencing an ice age this winter.
Interesting to hear you are reassured. These newspaper articles always looked daft to me, but then I’m a weather anorak. I think its a bit worrying if they are actually misleading (or worse frightening) people who aren’t as nerdy about this stuff.
It would be nice to get a response from a local authority about how useful they found the long range forecast if they made use of it. I suppose you wouldn’t get this kind of information in a freedom of information request.
I think that may be right, unfortunately. Local authorities are required to respond to FOI, but the request would have to be something like ‘what info do you hold on this subject?’. If they haven’t written anything down about how useful they found the long range forecast, which seems reasonably probable to me, then all you’ll get is ‘we don’t hold anything on that’, and you will never know whether they did or not.