Winter Forecast 2016/17

Background Factors

Solar flux – below 100 consistently, close to minimum levels, generally supresses jet activity or at least allows other factors to override.

QBO - ?, was going to flip to negative state but for strong El Nino, currently neutral/slightly positive at 30-50mb states, hard to account for so perhaps worth looking at neutral years. However, generally La Nina years favour early disruption over late, and downwelling of positive anomalies potentially aid this later on.

ENSO – weak La Nina winter, particularly weak, though may have influence later on.

Stratosphere – possibility of vortex split early November – incredibly early, should consider possibility of very cold/ blocked November/December as a result.

October AO – very negative, potential for impact on first half of winter

Atlantic SSTS – area from 45-60 degrees north, 15-45 degrees west has had positive temperature anomaly compared to very negative for the last few winters, some linkage between this and the NAO suggestive of enhanced probability of negative NAO. This, coupled with Met Office expectation (backed up by recent paper) and current vortex disruption gives reasonably high confidence in a negative NAO pattern to predominate for at least the first half of winter.

Arctic Sea Ice – once again near record low extent, correlates with negative AO in winter

Snow Advance Index – close to record positive levels (again), also correlates with negative AO/enhanced blocking

Models – generally were more towards a blocked outlook, though latest GCM looks a bit mild for northern hemisphere in general, though overall quite similar to the standard ‘cold mid Atlantic warm everything else’ output which works most of the time other than in NH winter.

Conclusion – ENSO/QBO/Solar signals all relatively weak, little forcing for either enhancement or particular weakening of jet from these other than perhaps some strengthening later as a result of WQBO/LN. All other signals point towards enhanced blocking, particularly early winter. 2010/11 as an analogue year sticks out – La Nina, early splitting of the vortex, similar SST profile…

 Methodology

Early winter analogue years: 1961, 1967, 1997, 2001, 1956, 1963, 2011, 1980, 1969, 1988, 1988, 1982, 2010, 1981

Late/full winter analogues: 1961, 1961, 1967, 1967, 1997, 1997, 2013, 2001, 1956, 1985, 1981, 2011, 1962

As usual I selected two overlapping sets of composite years based on different characteristics, the first for November/December and the second for January/February. I felt this was particularly important given the early disruption of the stratospheric vortex and the substantially negative Arctic Oscillation, which could have a significant impact in the near term in particular. The early years are more heavily weighted towards years with those characteristics, whereas the latter gives more weighting to the relatively low solar activity, weak La Nina and emerging positive QBO.

Analogue temperature mean (degrees Celsius) and anomaly wrt 1948-2015 average

 November   December   January      February

4.285714	2.1	2.076923	2.984615
-0.5113	-0.8522	-0.33203	0.580138

Rainfall (mm)

Nov           Dec        Jan     Feb                                                                   

160	151.6	122	135.47
6.294	-11.5	-36.1	21.662

So a fairly cold signal for November (no surprises there) but moreso December. The temperature anomaly for Europe suggests the cold will be more strongly concentrated across northwestern Europe with some relative warmth further south/east, which happily looks a lot like the forecast temperatures for the first ten days:

Forecast

November

NH Height Anomaly for November Analogue Years

European Temperature Anomaly for November Analogue Years

After the initial plunge of cold air into Europe this week the pattern looks likely to realign towards something perhaps similar to what the later stages of the GFS are showing, with low pressure generally centred towards or just north of the UK, blocking around Greenland and over Russia, and polar maritime incursions into Scotland and at times further south/east, but with most of the continent avoiding prolonged cold (for now). On the face of it, this might seem a familiar pattern to the one we saw for much of the last three winters, with storms coming in from the Atlantic bringing colder air in behind, but given the underlying weakness of the vortex along with the lessening of the cold anomalies in the Atlantic and reduced solar activity I would expect these troughs to be slower moving, giving potential for frost and, at times, snow to low levels. However, while a prolonged cold spell is a greater risk this November than in any other since 2010, I wouldn’t expect anything on the same scale as that, at least not until the end of the month, and in the south of England temperatures could end up around average overall. Rainfall looks to be around or above average, greater the further west (and, increasingly as the month goes on, further south), you go, but with sunshine also probably a bit above average.

December

December NH Analogue Mean Height Anomaly

December Analogue Temperature Anomaly

Though confidence in this is naturally lower, the likelihood is that December will be a cold month across northern and western Europe, and the potential for a spell of severe cold with significant snowfall looks to be significantly greater than normal. While pinning down the specifics of this is a near impossible task, given the signal for a negative NAO tendency blocking is most likely to be centred towards Greenland, with the slow moving trough of November sinking southeastwards into continental Europe bringing deep cold from the Arctic. This seems most likely to occur sometime between the end of November and the middle of December, with the transition most likely to take place gradually and messily, with slow moving frontal systems bringing a mix of rain, sleet and back edge snow before eventually pushing far enough southeast to allow Arctic air to push in. This would obviously bring the potential for snowfall to all areas, though particularly the further northeast given the source of the airmass, but if the blocking is sufficient to allow this pattern to persist for more than a week large scale frontal snow affecting large areas would likely develop. However, as I’ve said this is somewhat speculative to say the least – there’s also a fairly strong signal for either a battleground setup, with a messy mix of rain and snow, or a dry but still cold spell. I would expect the month to feature elements of each of these setups, more likely the former than the latter, but in general a cold month with a good chance of some snow, moreso the further north you are. Interestingly, all of the analogue years did actually feature fairly widespread white Christmases, either in terms of snow falling or lying snow, so I feel the annual white Christmas speculation has some decent grounding this time. Either way, the Christmas period is likely to see either a reinvigoration or a continuation or the cold spell, with more potential for snowfall, though also the prospect of a messy battleground setup with the transition from snow to rain and back happening infuriatingly often.

Overall: the coldest scenario is a watered down version of December 2010, with a mean temperature of 0-1C for Scotland, 1-2.5C in the CET zone, mildest scenario being something like December 2012 or December 2000 with one relatively short cold spell, a period of dry fine weather and a lot of cold rain CET 3.5-4.5C, Scotland mean temperature around 2-3C. The variance on precipitation is very high for this month, with the average a bit below normal but not by enough to have any confidence. I suspect above average precipitation is likely for the south of England either way but elsewhere, with maybe as much snow (or more) than rain, it might actually be a slightly drier than average December for the first time in years.

 January

January Analogue Height Anomaly

January Analogue Temperature Anomaly

This also has the potential to be a cold month, though subtly different characteristically to December. The third act of the ‘slow moving trough hanging around Europe’ sees it sliding ever further south, hinting that the source of the cold is more likely to become ever further east… This brings another month with the possibility of widespread snowfall, this time extending the risk south and east. For Scotland, while there’s likely to be further periods of snow, particularly for the east coast, there’s also the prospect of a dry month, with high pressure fairly close by for most of it. However, this doesn’t tell the whole story – while the pressure map looks incredibly good for those who like snow, the anomalies aren’t especially strong, and the temperature average looks is only marginally below average. That’s because, as is often the case with La Nina, our old friend the polar vortex is going to come roaring back, accompanied by the jet. The likely endpoint for the cold is a slowly sinking high, bringing gradually increasing temperatures on its western flank before gradually transitioning towards a more typical mild wet and windy regime. The order of this isn’t set by any means though (although as things go I’m pretty confident overall in the general trend of the winter) – it’s possible that, as in some of the stronger analogue years, there’s a mild though fairly dry spell towards the beginning or middle of the month with an easterly near the end, but either way the broader picture is trending away from sledge weather and towards a banal normality.

Overall: coldest scenario, something like a slightly less severe January 2010, a good amount of snow and cold but also quite a lot of sunshine and only a gradual and stunted transition to mild. Mildest: two extreme versions of this – either a dry month with some really mild southerlies at times or, if the jet ramps up a bit early, the easterly signal could turn out to be another January 2014 style mess, with the frontal zone further north, the snow ending up stuck here in Sweden (which I’d be fine with, FYI) and a month of cold rain, also messing up the ‘dry’ bit of the forecast. On balance, though temperatures likely to be a bit below average overall, precipitation below average, some snow but also the chance of some relative warmth.

February

February Analogue Height Anomaly

February Analogue Temperature Anomaly

February in this case is the slap in the face of months for snow lovers, sore but reassuring because at least you know that the rest of the forecast wasn’t a dream (or in this case, a wildly overoptimistic projection from a fellow snow fanatic). It’s also reassuring because it confirms both my slightly educated ‘hunch’ about this winter and much of the consensus. The pressure pattern has, as you can see, completely flipped – suddenly the blocking is gone from Greenland to be replaced by deep blue, which, of course, means rain. Of all the forecast maps I’ve produced, this one looks the absolute worst for cold weather in northwestern Europe, not even close to something salvageable. If I’m wrong I’d be delighted, but February looks like a wet month and probably quite a mild month.

Overall: coldest scenario, well the second snowiest February on record in Scotland is conspicuously on this list and it also featured a La Nina, but the QBO was trending negative at the time and c’mon, you’ve already had two months of snow what more do you want? Mildest: high pressure hangs around the Azores and it’s time to get out the barbecue, except not really because it’s Scotland in February and it’ll probably start raining soon so why the hell did you think that was a good idea?

I’m going to add in more details soon, like a summary for each of the factors considered and the mean CET for the analogue months. Thanks for reading, it’s been a pleasure writing this as always and I hope it comes off (except for February of course)!