BBC nags Met Office for horseracing form book

Posted: March 29, 2013 by tchannon in climate, Forecasting, humour, Incompetence, media, Natural Variation, weather

Over on Talkshop Suggestions Gray quotes nicely so here it is


By Roger Harrabin

Last spring’s forecast has been obtained by BBC News under Freedom of Information.

The Met Office three-monthly outlook at the end of March stated: “The forecast for average UK rainfall slightly favours drier than average conditions for April-May-June, and slightly favours April being the driest of the three months.”

A soul-searching Met Office analysis later confessed: “Given that April was the wettest since detailed records began in 1910 and the April-May-June quarter was also the wettest, this advice was not helpful.”

In a note to the government chief scientist, the Met Office chief scientist Prof Julia Slingo explains the difficulty of constructing long-distance forecasts, given the UK’s position at the far edge of dominant world weather systems.

She says last year’s calculations were not actually wrong because they were probabilistic.

The Met Office explained it this way: “The probabilistic forecast can be considered as somewhat like a form guide for a horse race.”

Something does not compute, the BBC using FOI with the end effect of mocking the Met Office?

The story is covered various other places, Bishop Hill being one worth a read.

In the same interview, Julia slingo came out with this gem:

I think it was quite right that we looked at this particular forecast last year because in March we were facing really very serious pressures on water resources – a major drought that had been going on for nearly a couple of years – and I felt when I looked at the seasonal forecast at that time that I would be not being fair to the Government if I didn’t emphasise the fact that we did see a slightly enhanced risk of the drought continuing.
Likewise, I did also emphasise that there was also quite a chance that April would also be wetter than normal but in the context of where we were at that particular point as a country I felt it was right to emphasise the risk of dry conditions continuing as a precautionary principle.

Slingo’s wet phase.
For goodness sake Slingo either scram or grow up although back to school might be a good idea, oh, school doesn’t get it yet either.

My take is that natural world physics is teaching an engineering lesson, that natural variability is much wider and operates on longer timescales than the likes of Slingo want to take on board. Low rainfall, high rainfall and every other shade of grey, sunshine, and snow white.

I can’t answer for her brain but I can for the weather. Most of the weather and climate is not random as accepted and taught in statistics. This has been known for years. See here and here where there are many papers and links to investigate.

The name here is Hurst-Kolmogorov, the way the kind of random fluctuations show in weather. Koutsoyiannis has even derived a more correct statistical formula which not only give the gauss answer but are generalised if the Hurst exponent is not 0.5, which is the normal random case. (ranges from 0 which would be a completely cyclic behaviour to 1 which is highly random periodic)

Rainfall in the UK, temperature, probably most things have an exponent around 0.7, check the first of the “here” links, where I show this is the case with CET. This characteristic has the periodic behaviour such as spells of wet, dry, cold, hot, windy, calm, familiar to all Brits.

A twist though, came across a CSIRO tome which claimed indirectly Hurst does not apply, on what I spotted is flawed justification. I must drag that out and write it up. Recent floods in Oz. odd that, so they don’t get periods of dry and periods of wet? Except in politics.

This stuff seems to link in to log-log behaviour, maximum entropy and few other way out things which seem to be so but can’t be proven as such. I suppose that makes it deniable. No-one can actually know.

The AGW community are in strong denial of this, logical given that admission of there being much greater natural variation would make a mess of their doomsaying. FGS it’s weather, get on with it.

Rant by Tim Channon, co-mod
(I usually don’t, looks bad, yet boy could I say some things including not so politely)

  1. What puzzles me as much as anything else is that the Met’s 3-month outlooks are readily available here.

    These hold the current ones, the older ones are usually in the Met Archives.

    I have been regularly looking at these and using them for two years. So why does Harrabin need to FOI?

    Sounds like he has been badly off the ball.

  2. No Julia, it wasn’t the Precautionary Principle, it was you covering your large bureaucratic back,

  3. mitigatedsceptic says:

    Am I correct in thinking that probabilistic forecasts can never be wrong? If so, it seems to me that Met Office is now beyond reproach whatever the mismatch between forecast and the real event.
    I wonder if this fuzzy approach could be used in corporate reporting too?

  4. oldbrew says:

    See Josh’s comment at Bishop Hill – Mar 29, 2013 at 9:53 AM

    Or just…place your Betts (sic)

  5. J Martin says:

    The BBC using FOIA as if they were sceptics ?!

    They were happy to go to court spending tens of thousands of pounds to prevent themselves from embarrassment, which they ended up suffering anyway as the information was found to be on the internet.

    At the risk of being disingenuous to the BBC, is this a case of double standards, or are cracks beginning to show, are some of the eco loons that control the BBC beginning to lose their grip on power ?

  6. Roger Andrews says:


    With Hurst exponents in the 0.7 range we can expect that a wetter or warmer than average month would likely be followed by another wetter or warmer than average month. In the Met Office graph shown below this works eight out of twelve times, which I think is a better success rate than the Met Office has achieved.

  7. tchannon says:

    There is an old heuristic that the best prediction of weather is the same tomorrow as today and also that must be beaten to do prediction.

  8. Roger Andrews says:

    Reminds me of the time I was driving around with my cousin in the pouring rain during one of my infrequent trips back to England. “This is the worst weather we’ve had since yesterday”, he said.

  9. Paul Vaughan says:

    This Slingo appears to be quite a dangerously misguided & misleading character. I suspect that moving forward many (including those close to her) will be dealing with her using tactics that are normally reserved for dealing with the worst intransigents. Possibly a good time to preemptively take the early retirement package Julia! Or maybe try at least showing crystal clear, genuine appreciation & respect for nature …no tricks — everyone’s watching.

  10. tchannon says:

    Dated 2010 a blog article by autonomous mind ripping the Julia character apart makes highly amusing reading in 2013

    “Just stop the lies right now, OK? The forecasts were rubbish, not the communication of them.”

  11. Ulric Lyons says:

    My solar based forecast for April 2012 was deterministic, it was guaranteed to be a colder wetter month.

  12. lgl says:

    “given the UK’s position at the far edge of dominant world weather systems.”

    Ah I see, the problem is they still have not discovered that the dominant ‘weather system’ in the UK during winter is the NAO. Not that it would help since it’s not possible to forecast the NAO either …

  13. Ulric Lyons says:

    lgl says:
    “..the dominant ‘weather system’ in the UK during winter is the NAO.”

    Can be, but not always, the AO is more of a decisive factor this month.

    “..since it’s not possible to forecast the NAO either …”

    Yes it can be forecast by predicting the solar conditions.

  14. tchannon says:

    Edge of a globe?

  15. lgl says:

    Then show us the correlation between solar conditions and NAO, say last 50 years.

  16. Ulric Lyons says:


    You could have a cursory look by comparing monthly NAO with plasma speed variations, paying particular attention to the winter values:

    Ideally it would be checked at a weekly scale, even at a monthly scale 50yrs worth is too long for one graph.