Wired - Matt Reynolds
Until now scientists have been cagey about linking extreme weather
events such as this summer's heatwave to climate change. An emerging
field is changing all that
 |
| Patrik Stollarz/AFP/Getty Images |
The summer of 2018 has not been a normal summer. Throughout June and July
an extended heatwave set record-breaking high temperatures across the northern hemisphere.
In Japan, more than
22,000 people were taken to hospital
with heat stroke as the country recorded its highest-ever temperature
of 41.1 degrees Celsius. In California, Portugal and as far north as the
Arctic Circle huge wildfires, encouraged by months of unusually dry
conditions, followed the searing heat.
For years,
climatologists asked to explain these kind of extreme events have fallen
back on a well-worn phrase. “It’s impossible to attribute a single
weather event to climate change,” the refrain goes. And they’re right.
Weather is by its very nature unpredictable – extreme events will always
happen in one place or another, regardless of global temperature
levels, and they’re not necessarily tied to one particular cause.
For Friederike Otto, deputy director of the
Environmental Change Institute
(ECI) at the University of Oxford, this response has its drawbacks. “If
scientists don’t answer, someone else gives an answer and it’s usually
people who aren’t interested in the size and have their own agenda,” she
says. Instead, Otto wondered if scientists could start saying whether
climate change had made certain extreme weather events more or less likely.
Now
Otto is right at the heart of a growing scientific movement called
extreme event attribution. Her aim? To be able to point to an extreme
weather event and use climate modelling to say whether that same event
would have been more or less likely to happen in a world where humans
hadn’t caused global temperatures to rise by a whole degree over the
last 120 years.
Up until a few years ago, it wasn’t
possible to draw that link with any degree of accuracy, Otto says. But
in 2004, Pete Stott at the UK Met Office
published a paper in the scientific journal
Nature
showing that climate change had at least doubled the risk of the 2003
European heatwave that killed tens of thousands of people.
Twelve years
later the Bulletin of the American Meteorological Society
dedicated an entire issue
to the new field of extreme event attribution. In the introduction, its
editors argued that it was now possible to detect the effects of
climate change on some events with high confidence. “That was really the
first time we could say that we can attribute events to anthropogenic
climate change,” Otto says.
In late 2014, Otto helped set up the
World Weather Attribution
(WWA) initiative – a collaboration between the ECI, the
Netherlands-based Red Cross Red Crescent Climate Centre and Royal
Netherlands Meteorological Institute. The aim of the project wasn’t just
to draw a link between extreme events and climate change, but to
provide this analysis in real-time so they’d have answers while the
extreme weather event was actually happening.
In July, while most of Europe was still sweltering under unusually high temperatures, Otto released her
analysis of this year’s heatwave.
She looked at temperature measurements in seven locations in Ireland,
Denmark, the Netherlands, Norway, Sweden and Finland and used different
climate models to estimate how likely it was that those same
temperatures would occur in a world without climate change.
To
do this, Otto ran hundreds of simulations on a pair of climate models.
These are the same kind of models used for weather reports that take
into account variables like rainfall, wind, temperature and air
pressure. The only variable changed in the two models is the
concentration of greenhouse gases in the atmosphere, as these are main
drivers of climate change.
One model represents our present-day
atmosphere while the other models the kind of world we’d be living in if
the concentration of greenhouse gases hadn’t ramped up post-1900. By
simulating hundreds of years of weather using these two models, Otto and
her colleagues are able to compare what the weather might be like in a
world with or without climate change.
When it comes
to the northern European heatwave, the analysis was unequivocal. Climate
change caused by humans has made these kinds of temperatures at least
twice as likely to happen. In other words, if we were living in a world
where humans hadn’t heated up the planet by a whole degree, this summer
heatwave would have been half as likely.
In
Copenhagen, Denmark, which saw temperatures of just below 31 degrees
Celsius, climate change made the likelihood of those temperatures
occurring five times higher. In Linköping, Sweden, it was six times
higher. This is weather attribution in action, in real-time. “What
climate change can do, and in the case of this heatwave what it has
done, is change the likelihood of an event occuring,” she says.
Although
the language of likelihood and chance may sound imprecise, this is a
great deal more certain than anything researchers would have been able
to identify a decade. Otto compares her analyses to studies into smoking
and cancer. Although it’s impossible to definitively say that an
individual case of lung cancer was caused by smoking, we can draw a
direct link between smoking and the likelihood of someone developing
cancer.
So far, Otto and her colleagues have analysed a raft of extreme weather events including
extreme rainfall in Japan in early July 2018 that killed 200 people and the
unusually cold winter across North America
last year. And the analyses don’t always turn up a link between climate
change and extreme weather events. In a study into the 2015 Ethiopian
drought that affected nearly 10 million people across the country, Otto
and her colleagues found no influence of climate change.
And finding examples where climate change isn’t
behind an extreme weather event is just as worthwhile as finding cases
where it is. Otto’s hope is that individuals and governments use her
analyses to futureproof themselves against the impact of climate change.
“One of the main motivations behind this project is that climate change
is already happening today and it’s having an impact on our day to day
life,” she says. In the UK, extreme event attribution has already been
used to help assess the risk of flooding in local areas, and Otto
expects this kind of analysis to be used more widely as the field
becomes more established.
“For a large part of the
world it’s still a very new science,” she says. But this emerging field
could help governments to start making their decisions on what might
happen on the future instead of thinking about what has happened in the
past. “If you have a changing climate and you only look at the past you
will not get the right answer,” Otto says.
At the moment the WWA’s
analyses compare a world with no warming to a world with one degree of
warming, but Otto also runs models that look at how the weather will
change if the world warms by a further one degree, as is projected to
happen by the end of this century. If that happens, the temperatures in
Copenhagen we saw this summer will be four times more likely to happen
in the future.
And it’s these extreme weather events
that we should be paying attention to – not just the headline figures of
global temperature increases. “Global mean temperature doesn’t kill
anyone,” Otto says. “It’s extreme events that kill people.”
Links
