We released our Winter 2018/19 seasonal forecast back on Tuesday, November 27th 2018 here. Now that the season is over and the weather stats are in, we believe that it is very important to evaluate how good the forecast performed with what actually occurred. Before we get onto looking at the month by month breakdown, let’s assess the teleconnections we mentioned in the forecast with how they progressed through Winter 2018/19.
North Atlantic Oscillation (NAO)
The North Atlantic sea surface temperature anomaly (SSTA) profile for much of 2018 was conducive to a positive NAO which gave us the indication that there might give an increased chance of a mild, wet Winter. Daily NAO data from NOAA reveals that whilst the NAO was not as positive as most years back to Winter 2013/14 (Winter 2012/13 was the most recent Winter that had a negative NAO), it was still a positive NAO season and it generally stayed positive for the most part with the exception of the beginning of January when a mid-Atlantic ridge tended to set up just out to the west of us allowing a northerly flow of air into Europe resulting in some very heavy snowfalls though the UK was on the periphery of this cold air. 2018/19 was not a wet Winter in the UK, as a positive NAO would usually suggest but it was a mild season. This is an unusual combination for Winter in the UK although we had a similar mild and dry Winter as recent as 2016/17. Overall, we think that the NAO was well forecast although maybe should have focused on it a bit more when deciding the actual Winter forecast.
One of the biggest speculations in the lead up to Winter 2018/19 was the possibility of a Sudden Stratospheric Warming (SSW) event sometime in December 2018 as the Climate Forecast System (CFS) model frequently forecasted one to occur around that period. A SSW event did indeed occur at the start of January 2019 as shown by the 60N zonal mean zonal winds reversing to easterly (below 0 m/s) in the chart below at 10hPa in the stratosphere.
However, unlike the four most recent SSW events (January 2009, February 2010, January 2013 and February 2018), the January 2019 SSW event failed to propagate down into the troposphere at 60N with the NAO showing little sign of going negative even into February. Yes, there was a cold spell in the second half of January continuing into the first few days of February but the depth of cold during this spell wouldn’t be too out of place in an average Winter as the air was polar maritime sourced (from the North Atlantic) than from the Arctic.
Like we said in the forecast, there is no guarantees of cold from a SSW event for the UK, just an increased chance via northern blocking forming. This time around, this did not occur.
Quasi-Biennial Oscillation (QBO)
The QBO was forecast to descend into westerly through Winter 2018-19 which gave mixed results historically as to be expected but overall, there was an increased chance of a mild, wet Winter. Again it was a mainly dry Winter but mild or very mild. We mentioned in the forecast that there tends to be a time lag on the impact from a descending QBO phase and one thing which could have broken the effects from the westerly QBO was the onset of the SSW event. However, the SSW event failed to propagate and this did not end up happening. There was a strengthening of the Polar Vortex in the latter part of the Winter which has continued into March with a very cool stratosphere in comparison to early January 2019 when the SSW event happened.
El Niño-Southern Oscillation (ENSO)
Winter 2018/19 was forecasted to have weak to moderate El Niño conditions by several models and agencies and some even forecasted El Niño Modoki (a central based El Niño) type conditions in the equatorial Pacific. El Niño historically gives an increased chance of positive NAO conditions earlier in the Winter whilst negative NAO conditions later in the Winter but no guarantee. We’re still currently in an El Niño phase although an official El Niño hasn’t been designated just yet as we’ve had four tri-monthly periods (up to Dec-Jan-Feb) with an anomaly of at least +0.5°C above average in the equatorial Pacific. The threshold for a designated El Niño is five tri-monthly periods with anomalies successfully equalling or exceeding the +0.5°C above average criteria. Whilst the atmosphere is likely starting to couple up with the oceans right now as the Southern Oscillation Index (SOI) is going negative, this was not the case through the Winter season. This has been a very weak El Niño event during 2018/19 as the atmosphere and oceans have been out of sync with one another through the season. The SOI tended to be up and down with the atmosphere favouring La Niña conditions but the ocean favoured El Niño. This resulted in some conflict and the failure of the forecasted El Niño Modoki. Therefore, the atmosphere favoured La Niña conditions to occur through the second half of Winter 2018-19 – La Niña favours positive NAO historically in the second half of the Winter (though this is not the case always). ENSO forecast for 2018/19 was not a great one although it does look like we’ll be getting the designated El Niño in the end.
Month by month breakdown
Now that the teleconnections are out of the way, it’s time to talk about our month by month verdict or breakdown we did in the seasonal forecast.
We forecasted a close to average December in terms of temperature if favouring the milder side of average due to factors like ENSO. December 2018 was overall a very mild month and featured similar temperatures to December 2016, though not as mild as the phenomenal month of December 2015. We felt December would be a rather cooler month than what actually happened due to the possibility of nightime frosts when high pressure builds in. However, there was limited frost actually as skies tended to be on the dull side with parts of the west recording a very dull December. December was expected to be on the wetter than average side which verified for the west of the UK but it was a drier month in the east of the country. Overall, we feel that the December forecast went down pretty well though underestimated the temperature.
The January forecast was for a close to average month again in terms of temperature and favouring the colder side of average as a result of potential frost and blocking. With the blocking in mind, we thought a dry January was likely. Indeed, this was probably the highlight of the seasonal forecast as it was near perfect to how January actually evolved. It was the driest January in the UK since 2006 generally and temperatures were close to average, relatively mild in the north and relatively cool in the south.
February was where the forecast went very wrong though as we forecasted a wet and cold month with the Atlantic tending to fight against blocking to the north delivering the possibility of some snowy spells. Whilst the start of the month recorded some significant outbreaks of snow in the south of the country, it wasn’t a snowy month by any means and was generally dry. It was also exceptionally mild, record mild for the UK in terms of mean maximum temperature although just beaten by February 1998 overall for mean temperature due to mean minimum temperatures tending to be lower.
At the end of the day, we don’t think it was a bad seasonal forecast especially given that it was a compromise based on conflicting signals although there’s no hiding from the fact that February’s forecast was a disaster.
What did you think of the Metcast Winter 2018/19 forecast? Feel free to let us know your thoughts in the comments or on our Twitter page.