Met Office Winter seasonal forecast – must try harder

Normally seasonal forecasts from the Met Office are couched in so many vague terms and indecipherable graphs that you can’t work out exactly what they are trying to say, but this time they made the dreadful mistake of being a little more specific especially with regard to temperature, when they talked about it in the Met Office Blog.

Temperature

Basically the seasonal forecast said that at the start of the meteorological winter (December, January and March) it looked “more likely than usual to be cold“, with a “heightened risk of wintry weather during December and into January“, but for the second half of winter it said that from mid-January to the end of February “there appears to be a shift towards less risk of cold conditions“.

Figure 1 – courtesy of the Met Office Blog

Well they were certainly wrong about their forecast of temperatures for December and January, the three Winter months ended up being either mild or very mild across most of the UK, as you can see from the mean temperature anomaly charts (fig 2).

Figure 2 – Courtesy of the Met Office

A colder spells did occur from the 20-27th of January, but it only really affected the southeast of England, when cold continental and clear skies allowed a series of moderate or severe frosts. Apart from the Buchan like cold spell early in the month, February was generally very mild, so maybe they got that bit of the outlook correct.

Figure 3

The promise of “more frequent northerly and easterly winds” didn’t materialise either, that’s not because it wasn’t anticyclonic or not blocked, because at times it was, it just meant that the anticyclone or the block wasn’t quite in the right place, as this LWT analysis shows (fig 4).

Figure 4 – Data courtesy of the CRU

Precipitation

I would have liked to comment on what the seasonal outlook said about precipitation for this Winter, but despite a good rummage around the Met Office web site I can’t seem to find the PDF. I can find the latest seasonal forecast for this Spring, but not the one for last Winter, how odd. That may be down to me, but I have noticed from experience, that it’s wise to take a screenshot, or download a copy of a PDF, because these things can disappear right before your eyes.

Conclusion

I do believe that the Met Office should be more transparent about their seasonal forecasts. Why not keep a web page that list all seasonal forecasts of recent years, which also included a simple verification table of just how the forecast did, nothing too overly elaborate or statistical, just green ticks and red crosses would suffice, then at least I wouldn’t have to all this effort to do these post mortems.

I remain skeptical about the Met Office’s longer range decadal forecasts, when they still have such a long way to go to crack the seasonal forecasting problem that’s proving so elusive. I don’t know what went so drastically wrong with this particular Winter forecast, perhaps it was too much reliance on the NAO or the QBO indices that let them down, but I do know that if the public are ever to place any value in seasonal forecasting, then it’s back to the drawing board for Doctor Jeff Knight and Professor Adam Scaife I’m afraid, and a “must try harder”.

About xmetman

An ex-metman passionate about all things to do with weather, climate and clouds
Bookmark the permalink.

3 Responses to Met Office Winter seasonal forecast – must try harder

  1. xmetman says:

    I can’t see any live data from it as yet.
    No good for you though – coverage only up to 52 deg N lat!
    I always swear by Blitzortung when it comes to SFERICs
    http://en.blitzortung.org/live_lightning_maps.php?map=10

  2. 1saveenergy says:

    Bruce have you seen this –
    March 6, 2017 Flashy First Images Arrive from GOES-16 Lightning Mapper.
    http://www.goes-r.gov/mission/firstGLM_images.html

    The first lightning detector in a geostationary orbit, the Geostationary Lightning Mapper (GLM), is transmitting data never before available to forecasters.
    http://www.goes-r.gov/spacesegment/glm.html

  3. 1saveenergy says:

    Since they got their £97million super computer, they seem to get 97% of long-term forecasts wrong.

    ( They use similar Al Gore-ithms to accurately forecast next century’s weather/climate…there’s a comfort. )

    They did better in the old days with a Sinclair ZX81 & a strip of seaweed.(:-))