One of the things I shared with my father was a passion for weather forecasts. This, for him, went well beyond watching Michael Fish or Ian McCaskill pointing to ridges of high pressure or weak fronts on big blue maps; when we stayed in the far northeast of Scotland for summer holidays, my father would phone the nearest weather station every morning, hoping for a rare sunny forecast for the more beautiful but rainswept northwest. I see it now as a sign of his hopeful, never-say-die spirit, and it worked; I’ll never forget the day we set out at the crack of dawn and reached the bird sanctuary of Handa Island by mid-morning, blessed by a cloudless sky and an Atlantic not grey but turquoise.
It was also a sign of how far ahead you could forecast weather with any accuracy in those days: probably not more than 24 hours with any confidence. The increased accuracy of short- and medium-range weather forecasting, helped by ever more sophisticated satellite data, seems to me one of the great unsung success stories of our time.
Nowadays forecasting is pretty accurate up to five, and not too bad up to 10 days, ahead – an enormous advance on those early, last-minute times. This has relevance not just for weather nuts like my father and me, but also for whole sectors of the economy, including transport, energy, agriculture and construction. The economic value of weather forecasting to European economies is estimated in the tens of billions of euros.
My interest in all this has become somewhat less amateurish since I started a research project on the ethical implications of Earth observation – or how our new-found ability to look at the Earth from space might encourage us to treat it with more care. I have become a little obsessed with satellites, as well as with weather; they seem to me an astonishing manifestation of human ingenuity, but they do not render us either omnipotent or omniscient. In fact for all their observational power, they can remind us of the limits of our knowledge. Sure, short- and medium-range weather forecasting has become vastly more accurate; but, as Alain Ratier, the director-general of Eumetsat (the leading European organisation providing satellite weather and climate data) explained to me recently, we still have no way of accurately forecasting weather more than a fortnight ahead. “Beyond 15 days,” said Ratier, “you cannot predict the details of weather. You have only noise.”
That may be of little consolation to anyone planning an outdoor event a month hence, or wondering when this winter’s icy grip, extended far into the season formerly known as spring, will ever be released. It could also deliver a healthy dose of scepticism, reminding us that so-called long-range weather forecasting is little more than guesswork based on statistical averages (remember the famous barbecue summer predicted in the UK in 2009, which turned into a damp squib?).
Should it also cast doubt on the science of climate change? Climate is after all nothing more or less than the sum total of weather over decades, or as Ratier puts it, “The probabilistic distribution of all weathers.” If we can’t predict the weather more than a fortnight ahead, how we can talk confidently about climate trends?
Ratier strikes me as a relatively cool and cautious man, but he is not a climate sceptic. Scientists (and journalists) should not be lured into making knee-jerk, unverifiable claims relating particular weather events to climate change, he believes, not least because they play into the hands of the well-funded climate-sceptic lobby, which employs scientists to point out the flaws in their arguments and thus by extension spread general doubt. But neither should a scientist, or a journalist, be tempted into the position of Pontius Pilate, wringing his or her hands, looking skywards and asking: “What is truth?”
For all his circumspection about ascribing every storm or extreme weather event to climate change, Ratier is quietly confident about the progress of the relatively new science of the attribution of recent climate change; he calls the work being done in this area “promising, interesting, careful, thorough.”
Understanding of atmospheric physics and chemistry, observations of actually occurring changes, and computer-based climate models which factor in (and out) anthropogenic greenhouse gas emissions combine to demonstrate that events such as the 2003 European heatwave and the Australian floods of the last three years are both extremely unlikely to have occurred without increased human emissions and much more likely to occur in the future.
The huge uncertainties and variables involved in weather and climate are daunting; but our knowledge has increased and will increase further. Who is wiser, the gambler who hopes the dice will fall right, or the person, like my father, who takes advantage of the little that we know to plan, just a little, ahead? The very worst thing to rely on might be what Auden called “our lack of faith, our dishonest mood of denial.”
More columns at www.ft.com/eyres