Over at the Talkshop, Tim Channon has had a look at the Met Office’s recent outlooks ahead of ex hurricaine Gonzales which sadly killed two in London*, recording gusts of 108mph in the Cairngorms, 88mph in Oban and 60-70mph in many parts. It caused wide disruption but nothing like the storms of last October and winter.
Image the Weather Network
Tim has collated some of the outlooks and said this which caught my eye having followed Gonzalo fairly closely
Concentrating on the South East, London area. See them start to realise what is coming. Meantime I could see the remains of an Atlantic hurricane was forecast by the American [models] to pass over the UK today… forecast a week or so ago. The track was marked.
Indeed, the American GFS model had picked out the track quite well whilst the Met Office again appeared less certain. Xmetman has an animation of the GFS run from four days ahead and runs in six hour increments. The following image is a screengrab for 1200 BST so that the charts are comparable.
This was the Met Office from five days ahead (image copyrights as per captions)
three days ahead
The way the track moves south can also be seen in the evolution of the yellow warnings taken between Saturday and Tuesday.
This is a few frames I captured from the BBC weather forecast on Monday evening
And this is the surface analysis chart from meteocentre for what actually happened at 1100 UTC (1200 BST) on Tuesday.
The GFS performed well as did Piers** with a more intense southerly track affecting more areas. I’m not sure about the Met Office’s performance in their bread and butter area – the 5 day forecast. Considering the range of options presented by the models a wider warning covering the South may have been more useful at an earlier date. Even if Gonzales was mostly standard Autumn fare for the British Isles and hardly out of the ordinary, if a warning or watch is to be put in place it can be refined as zero hour approaches. Most of the public are not weather nerds, they quickly glance at a forecast and that’s about it. They don’t often look at the radar to see what is incoming, so the futility in trying to use a fine brush to fill in detail when a broad one will suffice cannot be overestimated. A warning for London at 0947 after the morning commute is over and the children are at school may not be the best way to go about getting a message out.
* It has been reported that due to a falling tree the police had cordoned off the road before 58-year-old Teresita Sison died but not the pavement. A sad confluence of events.
The Daily Mail also has some great images from Gonzalo and the shelf cloud seen at Clacton-on-Sea is caputured beautifully by reader Samantha Batts
** Piers said yesterday “it is nevertheless true that for storm long range predictions and especially these ‘EndGame’ warnings /corrections to standard models we have a VERY high success rate.”
I hope to do a further piece on how to use WeatherAction R periods during severe weather events shortly. It is also note worthy the geomagnetic instability and earth facing coronal holes preceeding Gonzalo.