Ooops!

[xmetman]

Not quite going to plan this morning in the BBC Meteogroup forecast. It will be interesting to see this lot heads. 

Looks like another third overcast day for our part of Devon, whatever happened to that good start?

0600 - 0800 UTC 9 June 2018

[xmetman]

The CB cluster certainly seems to have a good northerly component to its track, but whatever model the BBC are using whisks it quickly away to the west and dissipates it, we shall see.

[xmetman]

Alina is still relying on taped forecasts for both the 1130 and 1158 broadcasts.

When speaking about the showers over Scotland she reassuringly adds "don't take these blobs of blue and greens too literally":-)

But the showers are still coming northward from France and have just about reached the coast of Hampshire and the Isle of Wight.

I wonder what went wrong with the T+06 frame of the 0600 model run?

And just which model is it?

[xmetman]

Well the UKMO model is even worse than the one being used by MeteoGroup. 

By 1400 they have the shower over Cherbourg dissipating in situ.

What I can't understand is that the radar indicates that they have a good N'NE component to them - surely the model can't have got the steering flow that wrong?

[Smartie]

Satellite imagery shows the showers over France are triggering and evolving alone multiple boundaries associated with a tropopause anomaly over Biscay/western France. Cells are moving northwards into the Channel but also initiating on the arcs of boundaries to the east and evolving with a more NEly component. The UKV domain was expanded to cope better with these sorts of situations but it has issues with upscale growth and organisation of convection.

[xmetman]

Thanks Smartie. 

Showery light rain has now spread across a large part of central southern England.

Alina is still sticking with the out-of-date NWP which has no rain at all.

The question is why doesn't she just swap to a rainfall radar animation?

Well the obvious answer to that is that she can't because it would show how out of line the NWP is!

[Smartie]

Brest 12Z sonde  failed above 500 hPa but but steering winds are SEly veering Sly

Trappes 12Z winds are Sly veering SSWly.

Both ascents have a large warm nose but this has obviously been easily overcome by heating and lift and convergence associated with the vortex over Biscay. Bulk Richardson Number is lowish so storms are likely to be of an organised, supercell flavour. Rather dry, inverted-V profiles suggest risk of large hail and strong convective downdraughts.

[Len W]

Yes Bruce. It seems there is no honesty in presenting the whole situation on the Met Office forecast.

The presenter would lose their job, or at least be severely reprimanded, if they said the model output looks wrong as shown by these observations (radar).

There has to be blind faith in the model output.

:-(

Len

[Smartie]

You are a cynical lot!

The echoes over the Channel are likely returns from hydrometeors aloft as they are some distance from radars- I don't know how much rain is reaching the surface. The model output is likely derived surface PPN rate (not derived reflectivity) so can't be directly compared with the radar product.

I would think the major source of uncertainty is the depiction of the tropopause vortex for which there seems to be a large spread in the global models. Add uncertainties in the handling of convection in the mesoscale models and you've got a tricky situation requiring mesoscale analysis and nowcasting.

I doubt if the presenters have time to do this properly, even if they are trained in the intricacies, as they are always having to prepare the next TV/radio forecast/party piece to camera.

[xmetman]

I suppose we are!!!

It wasn't the amount of rain so much as showers weren't forecast at all by the model for southern England.

It was apparent as early as T+03 that the 0600 model run was mishandling them.

The showers that the model was forecasing moved westward and not northward.

Both the UKMO and the ECMWF models were as bad as each other.

We do have a state of the art lightning location and weather radar system - why not use them especially in showery/thundery weather - and mores so when the model is out of whack.

[Len W]

What me cynical?

With my reputation.

[John Hall]

Some rain definitely reached the surface, as on the first day of the Hampshire v Surrey Championship game being played at Southampton only 53 overs were bowled out of a scheduled 96.

Not having checked the forecast this morning, I was surprised when here in Cranleigh a beautiful sunny morning was quite suddenly replaced by full cloud cover at about 2 pm. It subsequently became quite gloomy, though we haven't had any rain so far.

[xmetman]

This feature was TOTALLY missed by the two main NWP models and not picked up by the BBC weather presenters at all. You could have forecast the track quite easily by 0900 and seen that as a feature it would extend bodily northward by simple extrapolation.I don't believe how they could have possibly missed it and they obviously received no guidance from Meteogroup headquarters. The Met Office alternative was just as useless.