Norman Lynagh wrote:

Freddie wrote:

On Monday, 25 February 2019 07:27:54 UTC, Norman Lynagh wrote:
The met Office forecast for the Peak District issued at 0410 this
morning is a bit odd. For 0600 the weather was forecast to be
sunny despite the fact that sunrise wasn't until 0705. The
forecast temperature for 300m a.s.l. was 11°. In Tideswell, at
300m, the actual temperature at 0600 was -0.4°.

The forecast temperature at 300m for 0900 is also 11°. This also
seems unlikely as the current temp(at 0725) is -0.9°.

I can't explain the "sunny before sunrise" bit, but the temperature
forecast is probably one for the "free air" (or maybe hilltop)
temperature taken from the model forecast, as there is a very low
inversion this morning. I'm guessing you are in a valley of sorts?
At Leek, not so far from you, the temperature was 8.3 at 0700z, with
a dew point of -3.

It's happened again this morning. The Peak District forecast issued at
0343 this morning predicted a temperature of 13° at 300m at 0600
today. Actual temperatures at 0600 we

Tideswell (303m): +0.4°
Buxton (335m): +0.6°
Wardlow (299m): +2.8°
Curbar (251m): +2.6°
Burbage (298m): +2.9°

I raised this issue with the Met Office yesterday. They confirmed that
the temperature predictions in these forecasts are straight out of the
model and that the model isn't handling the low-level inversion very
well. Their response included the following

-------------------------------------------------------------
Of course, the problem with automated data is that it does have the
inevitable downside of performing especially badly in some very
specific weather situations, and causing the errors that you have
rightly mentioned.
-------------------------------------------------------------

My view has always been that raw model output is a tool for
experienced meteorologists to use but is not end-user material. The
models are just not good enough for that and this present situation
is a perfect example of that. It does the reputation of the Met
Office no good at all.

Sorry to keep banging on about this but I think it's pretty appalling.

The Peak District forecast issued at 1718 this evening predicts a
temperature of 9° at 300m at midnight tonight. In Tideswell it's
currently (1930z) 5.6° and falling quite quickly. At 300-350m across
the Peak District it's mostly in the range 5-8° at present. I don't
see any reason why it shouldn't continue to fall and by midnight it
looks certain to be a long way below the predicted 9°. Whether this is
model output or a human product I am not impressed. Anyone with half a
brain in their head could have deduced that the temp would fall away
quickly tonight with clear skies, no wind and a low dew point. If this
is the best that the model can do then it would be best to stop issuing
the product.

If the temp recovers and ends up at 9° at midnight then I'll have to
eat a large slice of humble pie!

--
Norman Lynagh
Tideswell, Derbyshire
303m a.s.l.
https://peakdistrictweather.org
twitter: @TideswellWeathr