Bishop Hill A copy of that cold winter forecast
Jan 21, 2011 Climate: MetOffice When the kerfuffle over the Met Office's winter forecast blew up, I wrote to the Quarmby team to see if they had actually received a copy of the Met Office's cold-winter forecast, which was apparently sent to the Cabinet Office. It is alleged that the forecast should have provided sufficient warning to the government machine to ensure that everyone was ready for what happened in December.
Today, rather later than I expected, the Quarmby team have responded and have helpfully provided a copy of the forecast:
Met Office Initial Assessment of Risk for Winter 2010/11
This covers the months of November, December and January 2010/11, this will be updated monthly through the winter and so probabilities will change.
Temperature
3 in 10 chance of a mild start
3 in 10 chance of an average start
4 in 10 chance of a cold start
Precipitation
3 in 10 chance of a wet start
3 in 10 chance of an average start
4 in 10 chance of a dry start
Summary: There is an increased risk for a cold and wintry start to the winter season.
Looking further ahead beyond this assessment there are some indications of an increased risk of a mild end to the winter season.
Reader Comments (79)
A few impressions of this forecast:
1- it is useless for planning purposes.
2- the emphasis of 'risk' regarding all aspects of weather, unless colloquial, indicates a fixation on normal weather being dangerous, which is odd for people who allegedly understand weather
3- it is a bureaucrat's CYA paper, not a serious piece of work.
The first thing that springs to my mind is that it must be a comfortable fence
So what it is really saying is that there is a 60% chance it will be average or warmer.
Note that they say cold as being temperatures below average. The Dec temperatures were not cold and just below average. They were very, very cold and way below average.
Is this the best the Met Office can do. It's useless besides being wrong.
That counts as a forecast of the devastating start to this winter? Millions were supposed to have been spent based on that?
I wouldn't have bothered to buy a woolly hat.
Your Bishopness could you perhaps compare and contrast the official winter forecast from mmm let me see .... Piers Corbyn? Joe Bastardi? and see where we get biggest bang for our buck.
Wow! - all that modelling expertise and we get a probability out of 10 (Simples!) followed by a cringingly awful, hopeful even, suggestion that "there could be a mild end to the winter season"
That would be Spring then would it?
I don't know how much they charged but I can provide a 10p coin for a fee of £250.
The ConDems are looking for cuts aren't they??!!
6/4 odds on for a less than colder than average start to winter. Better get the barbeque out of the shed then.
National public weather forecast service costing how many millions a year! Utter disgrace, there has to be grounds for reform if this is the best that they can do.
The summary is priceless: "There is an increased risk for a cold and wintry start to the winter season." - or this winter will be wintry!!
LB, that's unfair. Of all possible winter types, the probs were forecast as follows:
mild & wet .09
mild & avg .09
mild & dry .12
avg and wet .09
avg and avg .09
avg and dry .12
cold and wet .12
cold and avg .12
cold and dry .16
So they forecast the highest probably correctly, even if 84% of scenarios were otherwise.
"...increased risk of a mild end ..."
Since when has mild weather been a risk?
Summarising the forecast: there is a risk of weather but we have no idea what it is actually going to do.
The Quarmby reported stated: -
“The Met Office gave ‘early indications of the onset of a cold spell from late November’ at the end of October”
Met Office says:-
Summary: There is an increased risk for a cold and wintry start to the winter season.
Government says:-
Nowt! And whilst it pains me I can’t blame them, because whatever the “Met Office Initial Assessment of Risk for Winter 2010/11” is supposed to relate, remains a mystery.
“increased risk” increased over what? Do they mean 4 in 10 cold is an increase over 3 in 10 mild?
Reminds me of the old infamous accounting story – “Whilst the information you have given me maybe correct it is of no effing use whatsoever.”
Boy would we be in real deep if they did not have all that computing power. Maybe when they get their new one we will be safe knowing that it will be 3.75 in 10?
Oh my god 33 million plus running costs and that's it? a 'could be ' 'maybe' finding great! I think I will stick to chicken bones and tarot cards possibly more accurate!!
This is very typical of the UK Met Off. It is useless except for covering one's posteria (arse). What on earth does it mean? 70% chance it might not be mild, 70% it might not be average, 60% chance it might not be cold. Reinterpreted. Don't know, haven't got a clue, need a billion pound computer and 500 extra people. OR 70% chance of cold start, and 60% chance it might be cold. Oh right, I'll just go buy £10 billion of snow clearing equipment. Harabin your a clown along with your best friends the UK Gov, the UK met off and UK broadcasting company.
Re: steveta_uk
I thought all that white stuff falling from the sky made it either "cold and wet" or "cold and average" which means they gave equal probability to:
mild & dry .12
avg and dry .12
cold and wet .12
cold and avg .12
By MET office standards, i guess that makes the global warming forecast for the next 100 years "warmer than average".
Tut! You guys just don’t understand percentages.
They were 40% sorta right.
30% not a million miles away.
30% negatively correct.
And 100% useless.
But in 50 years it will definitely be hot, based on the same computer programs that can't exceed 40% probability for a cold winter.
Time to put UK weather forecasting out to competitive tender
Page 4 item 7.
National Grid don't get told.
http://www.nationalgrid.com/NR/rdonlyres/9721EF19-2BA8-4DBD-880D-90406603C176/43423/WinterOutlook2010_11.pdf
And they want a multi-million pound computer to generate this sort of crap? Josh, give them a Ouija Board in the next cartoon you do about the Met Office
bernie
"The summary is priceless: "There is an increased risk for a cold and wintry start to the winter season." - or this winter will be wintry!!"
Not only that - but there is an increased risk that the early spring will be early springy!
One assumes that 'wintry' has to do with precipitation since 'cold' only has to do with temperature. Note the risk is given for cold AND wintry, i.e. higher than normal precipitation coupled with lower than normal temperatures.
Now, let's look at the figures. On the face of them, there is a 60% chance that it will NOT have a cold start, and a 70% chance that it will NOT have a wet start. If these were independent variables then we'd end up with the probability of 88% that it will NOT start 'cold and wintry'. That doesn't exactly make me think that cold and wintry is very likely. We're knocking on 90% here so that's well above 'likely' and practically into 'very likely' territory in IPCC-speak that we shan't be seeing a colder AND wintrier start than normal.
If we give equal probabilities to all outcomes (i.e. assuming the Met Office did no better than throwing darts blindly at a dartboard, or used a random number generator to predict the weather) then we could give 33.33% probability to all three outcomes: cold, average, mild; and wet, average, dry. So the probability that the winter will NOT start 'cold and wet' if the individual outcomes are random and equal probability of turning up would be 88.89%.
Is someone trying to tell me that because 88% likelihood (according to Met Office super-dooper prediction) is less than 88.89% likelihood by throwing dice that the Met Office can say with a straight face "There is an increased risk for a cold and wintry start to the winter season". What, less than 1% difference in risk compared to sheer randomness enables them to say that?
Pull the other one.
There is only the three temperature options, mild, average or cold and they have selected to tell us that it will be none of the above but if anything there is a slightly less chance of it not being cold.
Has nobody got any gonads within the department. Look at the reports that the independant forecasters gave us strong, positive statements that their reputations stand or fall by. Not it could be or possibly or 'increased risk of' it's IT WILL BE A COLD AND SEVERE START TO WINTER.
My advice to the MET office is to split off the climate scientists into their own section, office space available in the Falklands, and get back to doing what you know best
That forecast seems to be the kind of stuff you could come up without any analysis required at all.
If one assumes a Gaussian distribution to winter temperatures, then in fag-packet terms, you'll get roughly equal chances of a cold, average or warm winter.
As they've given chances out of 10 then 3/10, 3/10 and 4/10 are effectively amount to a shrug and saying "we don't know".
Peter:
Yep, I can just hear Sir Humphrey saying these things with that supercilious grin on his face as if he has revealed some deep truth that is hidden from the rest of us.
This news is so predictable, MET office as of old, it predicted anything could happen and anything did.
Here's my forecast of what the forecast was likely to be back on Jan 11th
Never seen a MET office forecast yet that was specific enough to be a true we will nail our colours to the mast forecast. The most forceful its likely to have been is ‘The is a small but possibly larger chance of the winter being colder than normal’ . I would be extremely suprised if it was any more definate than that, unless they added a note along the lines of ‘And Peirs says it will be really bad and he’s a better forcaster that us’ LOL
If I was in the Cabinet office I would be spitting tacks, spin spin spin.
If I was in the cabinet office I would be getting the P45's organised.
does anybody have a copy of a contemporaneous forecast from either Bastardi or Corbyn?
For a while we received forecasts in Canada that the daily temperature would be "seasonal". Nobody know aht hat meant. I can also remember the summer daily forecasts that low would be in the teens (Celcius) and the high would be in the 20s. Thus the day could either be quite cool (20, 10) or quite hot (29,19) and th weather forecast would be correct. Things have improved since then.
And the taxpayers have to fork out for this CYA nonsense?
Re: Terrys
All that white stuff falling from the sky made it very white and slippery, but didn't actually amount to much precipitation in total volume - it was a dryer than average December.
From the Met:
December was, however, generally drier and sunnier than normal. Less than a third of the average rainfall was recorded over most of Wales, western England and western Scotland. Provisionally, it was the driest December over the UK since 1963 and the third driest in the 100-year series. It was a very sunny month in the west and north, with over twice the normal amount and, provisionally, Northern Ireland and western Scotland enjoyed their sunniest December on record. In contrast, it was a dull month in south-east England with around 60% of average sunshine.
That one had me laughing. They call that a forecast?
The Met Office is too ridiculous to make jokes about. But Josh will do the business again.
How can I compete with such a forecast - it is just too funny all by itself... 'an increased risk of a mild end...'
Ring the alarm bells now!
Bish,
You made this up... right? Oh yeah, I forgot, you can't make this stuff up.
Ouija board ? Nah, they use a dart board...
Josh you should just put the forecast into a cartoon as it stands with a caption of "Honest it's not one of mine!"
Who wants to be the original forecast was:
.33333 warmer
.33333 same
.33333 cooler
Then someone noticed when they rounded things off, they got:
.3 warmer
.3 same
.3 cooler
Which didn't add up to 100%, so they rounded the last item up to get:
.3 warmer
.3 same
.4 cooler
In other words, the reason they got a higher chance of it being cold was simply due to rounding. They had to round one of the items, and by chance it was the last item in the list.
So on the strength of "forecasting" that there was a roughly equal chance of the start to winter being average, mild or cold, they spin this through Harrabin as having warned the Government about the second coldest December in 400 years?
And what's even worse, they must have thought no-one would bother to check what it was that they actually did forecast (despite the manifest existence, in their eyes, of a massive, oil-funded denial machine)?
I'm struggling for appropriate words here. "Stupid" and "arrogant" seem to fit the bill. Perhaps unprecedentedly arrogant and stupid, in fact.
Is that the complete forecast provided to the Quarmby team?
It says This covers the months of November, December and January 2010/11.
But then it talks only about what the "start" of winter might be like and then what the "end" of the winter season might be like. Nothing about the period between the start and the end.
Anyway, what is the "start" of winter? The day of 21 December?
Yes, I too think it's a leg pull.
ge0050
Thanks, that made me laugh out loud. Not least because I strongly suspect it's true.
@ AngusPangus: I'm struggling for appropriate words here. "Stupid" and "arrogant" seem to fit the bill.
Unequivocally stupid and arrogant?
The best you can do with this Met Office assessment is to conclude that temperatures were likely to be average or colder for December (70%) rather than milder or average (60%). It suggests a forecast of slightly below average temperatures.
I bet that is what the government and government agencies took it to be.
What happened was the coldest December on record with temperatures 5C below average.
What has not been explained is how the Met Office probability map, based on their own models, predicted a 60-80% chance of above average winter temperatures changed to an assessment that the government took to indicate was a forecast of slightly below average temperatures.
It suggests that this Met Office assessment was an aggregate of seasonal forecasts from other metereological agencies.
Finally, we can take it that Roger Harabin was knowingly spinning a falsehood of a Met Office prediction of an extremely cold start to this winter.
"Looking further ahead beyond this assessment there are some indications of an increased risk of a mild end to the winter season." Increased risk compared to WHAT? Equal probabilities of all outcomes, what you'd get by throwing dice?
Something like this then:
4 in 10 chance of a mild end
3 in 10 chance of an average end
3 in 10 chance of a cold end
So given the start figures:
3 in 10 chance of a mild start
3 in 10 chance of an average start
4 in 10 chance of a cold start
In other words, 60% 'chance' that winter won't start cold, but an increased 'risk' that it will, and 60% 'chance' that winter won't end mild, but an increased 'risk' that it will.
So now we can play this any way we like, and pretend we're better than throwing dice. Check it out for yourselves: if it starts cold the MO say "well, we said there was an increased risk [40%] of it starting cold". If it doesn't start cold the MO say "well, we said it was less likely than not [40%] to start cold". If it starts average the MO say "well, we said it was less likely than not [40%] that it would start cold, and less likely than not [30%] that it would start mild".
If it ends mild the MO say "well, we said there was an increased risk [40%] of it ending mild". If it doesn't end mild the MO say "well, we said that it was less likely than not [40%] to be ending mild". And if it ends average the MO say "well, we said it was less likely than not [30%] that it would end cold, and less likely than not [40%] that it would end mild".
So, you really can have your cake and eat it. Whatever happens you can con people that you are 'skillful'. Think astrology.
Mmm, they got it right with a cold start, though completely missed out on the intensity of the cold, and, from where I'm sitting, they are right (thus far) about a mild end to it. Where they seemed to have gone wrong is in precipitation. Am I reading that right?
No sorry. I really must concentrate. I got that wrong. Greater chance of mild to average and a greater chance of average to wet. Must engage brain before posting.
And in the end this is not far off, 'it might be mild or cold and it might be dry or wet'.
@ Martin A
or perhaps "robustly" arrogant and stupid...
Is this what you get for £70 million a year?