Published on:

Variability matters more than trend

This is a version of an article I published in The Times on 27
March:

 

The east wind could cut tungsten; the daffodils are weeks
behind; the first chiffchaffs are late. It’s a cold spring and the
two things everybody seems to agree upon are that there’s something
weird about the weather, and it’s our fault. Both are almost
certainly wrong.

 

On weird weather, it is true that the contrast with last year’s
warm March is striking, as is the difference between the incessant
rain of the last twelve months and the long drought that preceded
it in most of England. In the last year, America’s had a heatwave,
a superstorm and now a bitterly cold spring. Australia has just had
an “angry summer”. And so on.

 

The government’s retiring chief scientist, Sir John Beddington,
claimed this week that “we are seeing more variability”. Is he
right? On the whole, no. Forget the anecdotes and examine the
data.

 

Start with America. Professor Roger Pielke of the University of
Colorado has documented that floods, hurricanes, tornadoes and
east-coast winter storms have shown no increase since the 1950s,
while droughts have shown a slight decrease. The only thing that
has changed is the financial damage done by storms, but as he drily
remarks “The actual reason for the increasing number of damaging
tropical storms has to do with the reporting of damages.”

 

What about elsewhere in the world? There has been no trend in
tropical cyclone intensity or frequency worldwide at all. The
Intergovernmental Panel on Climate Change itself, though heavily
infiltrated by environmentalists in recent years, stated in a
recent special report on climate extremes that over the coming two
to three decades “signals are relatively small compared to natural
climate variability” (as Matthew Parris pointed out last week,
don’t you hate this habit of making forecasts in the present
tense?), and that “even the sign of projected changes in some
climate extremes over this time frame is [sic] uncertain”.
Translated: the weather is just as likely to become less extreme as
more extreme.

 

So why is everybody convinced otherwise? Partly because they
have been listening too much to the big insurance companies, which
have a vested interest in bidding up our anxiety, as Dr Pielke’s
remark reminds us. Also it seems even government chief scientists
suffer from what psychologists call “availability heuristic” – when
people judge the probability of events from how easy it is to think
of examples. Here’s an instance: “I cannot recall such a cold
March”, says a man with dim memories of 1963 as he reads a Met
Office report that March 2013 is the coldest in Britain since…1963.
Do you remember March 1963? I don’t.

 

So next time some pub bore tells you this cold month is caused
by the extensive melting of Arctic sea ice last summer, ask him if
the same thing happened in 1983. In any case, even if there were
evidence for changing weather, blaming every weather event on
climate change is lazy at best and dishonest at worst. As I wrote
in these pages during the cold December of 2010, “It’s not climate
change. It’s weather: just a cold snap.”

 

Not that politicians took my advice. Mayor Michael Bloomberg of
New York and Governor Chris Christie of New Jersey were quick to
blame “superstorm” Sandy on climate change not least because it was
a convenient way to deflect blame for the disastrous power and fuel
shortages that afflicted communities for so long after the
storm.

 

There was perhaps an echo of this last month when Lord Smith,
chairman of the Environment Agency, incautiously blamed recent
flooding on “increasing instances” of “convective rain, which sits
in one place and just dumps itself in a deluge over a long period
of time.” Not only did this sound to annoyed flood victims and
their MPs as an excuse; it drew a sharp rebuke from the veteran
weather forecaster Bill Giles, who said “There is nothing new about
convective rain. Perhaps next time he should get a meteorologist to
check his answers.”

 

As we have just experienced in a year, the variability of
weather matters far more than the trend in climate. Noise matters
more than signal. The world grew about half a degree warmer in the
last third of the twentieth century, but the difference between
March 2012 and March 2013 was almost ten times as great as that
(7.7C vs 3.1C).

 

Suppose we get another half a degree of warming in the next
three decades – and we may well do, even though there’s been no net
global warming now for 16 years and the latest peer-reviewed
studies (from James Annan of the Japan Agency for Marine Earth
Science and Technology, Magne Aldrin of the Norwegian
Computing Center and Michael Schlesinger of the University of
Illinois among others) all confirm that the climate is less
sensitive to carbon dioxide than the IPCC has been saying. That
warming would be one-tenth as much change over 30 years as we just
experienced between two versions of March in consecutive years. It
would be equivalent to moving house to a new village 250 feet lower
down a hill (temperature changes by 0.65C per 100 metres of
altitude), or a couple of counties to the south.

 

It would also be about one-quarter as much warming as you would
experience if you moved from rural Surrey to central London. Nearly
all the stories of recent years about how much earlier flowers have
been blooming came from within the massive urban heat island that
is London: Kew for example has experienced more local warming than
it has global warming. There is much less evidence of changing
seasons from rural Britain. My diary records of the date I first
heard first chiffchaff sing since 1986 show lots of variation, but
no trend at all.

 

In the heyday of climate-change mania of a few years ago we were
all told to plant cacti and drought resistant lawn grass. Yet we’ve
just had a run of three out of four hard winters and they killed
off the eucalypts and Wollemi pines my father planted in his
arboretum. It’s a racing certainty that you will still have to plan
for occasional hard winters and you won’t have to tell your
children what snow is.

 

The lesson that weather matters more than climate is not just a
bit of fun. Airports and our councils forgot to plan for snow a few
years ago, because they were more focused on the trend than the
variability. In 2010 Brisbane disastrously overfilled a dam because
it expected drought to return; the dam could not absorb a flood
when it came.

 

Climate is what you expect; weather is what you get. Which is
presumably what the first chiffchaffs will be soon be saying to
themselves as they desperately search the barren tree branches for
frozen insects.

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