Published on:

The Met Office keeps getting 3-month forecasts wrong on the warm side

I wrote the following op-ed in The Times (behind a paywall) on 2
July.

As I cowered in my parked car in a street in Newcastle last
Thursday, nearly deafened by hail on the roof of the car, thunder
from the black sky and shrieking girls from the doorway of a
school, a dim recollection swam into my mind. After inching back
home slowly, through the flooded streets, I googled to refresh the
memory. On 23 March this year, the Meteorological Office issued the following prediction:

“The forecast for average UK rainfall slightly favours
drier-than-average conditions for April-May-June as a whole, and
also slightly favours April being the driest of the 3 months. With
this forecast, the water resources situation in southern, eastern
and central England is likely to deteriorate further during the
April-May-June period.”

That went well, didn’t it? April-May-June was the wettest ever
in England, though not in Britain. According to the private
forecaster MeteoGroup, June was probably the wettest in England and
Wales since 1860, the dullest since 1909 and the coldest since
1991. The water resources situation, far from deteriorating, is a
cup that overfloweth.

The Met Office’s track record of short-range (5-day) forecasting
is, in my experience, very good and getting better, but its
longer-range predictions have often been not just badly wrong, but
consistently biased on the warm, dry side.  In 2007, it
wrongly forecast a warm summer. In 2008 it wrongly forecast a mild
winter. In 2009, it said “the chances of getting the barbecue out
are much higher than last year” but the summer was a washout. Also
that year it said that the trend towards milder winters was likely
to continue, whereupon a savage winter followed.

Chagrined, it said it would give up seasonal forecasting, but
continued to produce much the same information in three-month
forecasts. In October 2010 it saw “a very much smaller chance of
average or below-average temperatures” in the coming winter shortly
before the coldest December for 100 years. These mispredictions
were not without consequence. The under-preparedness of airports
and councils for the big freezes at the beginning and end of 2010
was directly related to the forecasts they had sought.

Now look at the curriculum vitae of the chairman of the Met
Office, Mr Robert Napier. He is also chairman of the Green Fiscal
Commission and the World Conservation Monitoring Centre, director
of the Carbon Disclosure Project, the Alliance of Religions and
Conservation and the Climate Group. He is so high up in the church
of global warming, he is a carbon cardinal. I am sure he is a man
of great integrity, but given this list you have to wonder if one
of the organizations he chairs does not occasionally – and perhaps
unconsciously – aim to please him with warm long-range
forecasts.

Of course, these days the narrative has changed and we are
usually told to expect more extreme weather events as a result of
climate change, rather than a warming trend per se. If June was
indeed the wettest since 1860, that is extreme, but with only a few
centuries of data, records are bound to be broken from time to time
— and it is not much of an extreme that fails to beat a 152-year
record.

Likewise, in November 2009 when torrential rain swept away the
bridge at Workington, a Cumbrian rain gauge recorded the greatest
rainfall in any 24 hour period since British records began — a
total of 316mm (12.5in). Astonishing: till you read that it did not break the “day” rainfall
record, which is measured from 9am to 9am and which is still held
by Martinstown in Dorset on 18 July 1955 – 279mm. That’s before
global warming was supposed to have shown up.

What, in other words, was so special about the climate in 1860
or 1955 that it too produced extreme events? The truth is that for
all the talk of climate change, a trend of half a degree of warming
in half a century is still very much less relevant to airports,
wedding planners or breeding birds than the random and occasionally
extreme variation that is bound to show up in some years with or
without man-made climate change. As they say in physics, the noise
is greater than the signal. It certainly was last Thursday.

 

Update 1: The Times pubpished a letter from Bob Ward calling my
article “rather silly”. Since his letter confirmed the accuracy of
my main point — that no trend can be discerned in the signal

“While he is right that we cannot yet detect the
signal of climate change within the relatively
small datasets of extreme weather events in the UK…”

and contradicted no other of my arguments, I thought his letter
was “rather silly”.

 

Update 2: Paul Homewood has some good graphs here, showing how trend-free June’s weather in
England still is.

here is one of his charts.

By Matt Ridley | Tagged:  rational-optimist  the-times