Stay tuned: The 365-day outlook is next.
Sitting in his office in Elephant and Castle, a seedy suburb of south London, Piers Corbyn admits he prefers to do his calculations by hand. It's not a matter of "religious opposition" to using computers; he simply gets a better feel for how his calculations are evolving.
As he explains his life's work to me, Corbyn is soft-spoken and patient, though he still looks every inch the mad professor: salt-and-pepper beard, tangled hair, wire-frame glasses, loud necktie, crumpled flannel pants, drip-dry shirt, and a manic gleam in his eye. He's surrounded by stacks of scientific journals and cardboard boxes jammed with books. The trash is overflowing. Charts and maps are taped to the walls. His desk is barely visible beneath heaps of paper.
Negotiating the clutter, he reaches into his briefcase and pulls out a tattered plastic folder filled with outlined maps of northern Europe, each covered with scribbled comments and diagrams in black ink. Admittedly, Corbyn never actually says "now if my calculations are correct." But if indeed they are, the conventional approach to weather forecasting - one that involves enormously powerful and expensive supercomputers, and is the basis for nearly every weather report produced - will suddenly look rather ridiculous. If I'm to believe Corbyn, his scrawls represent something conventional science says cannot exist: a detailed weather forecast that reaches nearly a year into the future.
How detailed? One convert says that Corbyn gave him a perfect morning-to-midnight description of his wedding day. He could have gotten something like that on the telly the night before, of course. But Corbyn made the forecast six months ahead. Most of Corbyn's work, done through his company, Weather Action, is not quite so refined. But his look-ahead maps do show fronts and pressure systems, and his forecasts come complete with detailed estimates on temperature, precipitation, wind, and sunshine levels.
To his critics, Corbyn's a heretic, an unscientific crank, perhaps even a fraud. They point to the secret formula by which he prognosticates, for one. The 11 months' worth of scribbled forecasts he shows me were made using a proprietary method he says correlates solar activity with weather on Earth. But to forecast that far ahead would mean foretelling the movement of inherently chaotic weather systems, and that, contend his detractors, defies the laws of physics.
Another sign he's not your everyday weatherman: the conspicuously displayed photocopy of a check for £2,291 hanging on the wall. Unique among meteorologists, Corbyn bets on his forecasts. Unusual among bettors of any stripe, he wins regularly. The check on the wall is a payout from London bookmaker William Hill on one of their monthly bets.
While there are questions about why his method works that Corbyn can't or won't yet answer, he claims the method does indeed work. And he says his critics - most in government jobs with the Meteorological Office in the UK - are narrow-minded and hypocritical.
For Corbyn, 51, all this is a lot more than an intellectual exercise. He has his eye both on writing a new chapter in meteorological science and on grabbing a piece of an international business worth an estimated $2 billion a year. The World Meteorological Organization figures that accurate weather reports already benefit the global economy to the tune of $40 billion annually. Forecasts of tropical cyclones, for example, enable oil companies to reduce delays in drilling operations, which can cost up to $250,000 per rig per day. Predicting the path of a hurricane can cut days off factory shutdowns, which cost the US economy an estimated $15 million a day. Our own National Weather Service found that the airline industry alone saves close to half a billion dollars annually by using detailed forecasts: Every avoided flight-cancellation saves $40,000, and every avoided flight-diversion saves $150,000. (Interestingly, the travel industry is not so sanguine about the potential benefits of long-term forecasts, because they could adversely affect bookings.)
In addition to government-funded organizations, there are also a growing number of private companies that produce forecasts, often focusing on specific industries - aviation, shipping, petroleum, insurance, and so on. But since they all rely on conventional meteorological techniques, none of them can offer detailed long-range predictions. If indeed Corbyn's so-called Solar Weather Technique really works, it means his forecasts are months beyond anything else out there.
In the pantheon of scientists who have pursued the perfect weather forecast, Lewis Fry Richardson holds a hallowed place. He was a meteorological visionary little honored in his own time, and it's not surprising to hear Corbyn claim kinship with a man who waited decades for his ultimate vindication. "It is interesting how extremely narrow-minded the traditional meteorologists are," he says. "They talk about their history, they praise [Richardson] for standing out against the stream. And then they say: 'Who are these upstarts, these people who don't know anything about meteorology? The sun ... what a ridiculous idea.'"
A physicist and mathematician, Richardson set himself the task of coming up with a numerical weather-forecasting model, and began his work in 1913. When World War I broke out the next year, he looked for a way to be of service that didn't violate his pacifist beliefs and eventually became a driver for a Quaker ambulance unit working with the French army. He took his weather work with him, and later recounted that he sat on a heap of hay in a cold barn during his off hours and labored through thousands of paper-and-pencil calculations. His goal was to work out what the weather ought to have been for a six-hour period in central Germany on May 20, 1910 (a Norwegian meteorologist having already published the detailed statistics for that date and place which Richardson needed as his initial data).
Richardson carved the map of northern Europe into cells and repeatedly applied equations to predict how the conditions in each would change over time, both at the surface and at several levels in the atmosphere. Eventually, he came up with his hindcast. It's easy to picture Richardson as an eccentric - his scholarly activities and bookish appearance earned him the sobriquet "the Professor" among the troops - and to imagine him rubbing his hands as he consulted the historical record to see how his backward prognostication checked out. And it's just as easy to see his brow knitting when he discovered that, as his obituary by the Royal Society of London put it, "the results of the computation were strikingly at variance with the observed facts." Not the first weatherman to sift through the wreckage of a bad prediction, he quickly produced a plausible explanation for the error: The initial data he used, not his calculations, were flawed. If the method of data collection could be improved, which would entail repositioning Europe's weather stations to fall in line with his grid pattern, maybe it would be possible to devise an accurate forecast.
Richardson proposed dividing the globe into thousands of cells, capturing current meteorological data from them, running all the information through his mathematical formulas, and then predicting the weather. The idea, however, had one gargantuan obstacle to overcome: The process of collecting the required data and formulating a forecast was spectacularly time-consuming. Only by dramatically speeding up the process would it be possible to prepare a forecast that was not out of date before it was finished. Richardson had an answer for this problem, too - one that might be called massively parallel processing. He proposed setting up 64,000 people to work together in a vast installation to formulate global weather forecasts in real time.
Preposterous.
But after World War II, nearly a quarter century after Richardson outlined the wild scheme in his 1922 volume Weather Prediction by Numerical Process, something curious happened. Computing pioneer John von Neumann saw that the first digital computers, built to simulate the physics of nuclear weapons, could also be used to model the weather. By the time Richardson died in 1953, just short of his 72nd birthday, the University of Pennsylvania's ENIAC had run his equations - and they worked. Today, weather forecasts are produced using pretty much the method Richardson described. The world's fifth most powerful forecasting computer, a Silicon Graphics/Cray T3E900, resides in Bracknell, England, in a meteorological office building named for Richardson. Its 880 DEC Alpha microprocessors, running at 450 MHz, work together much the same way Richardson imagined his 64,000 human weather computers would. Richardson might have blown his first forecast, but his long-range outlook on the future of the science was dead on.
Whether or not Corbyn and Richardson are cut from the same renegade cloth, there is one important distinction between the two scientists. Richardson published his work and left it to others to prove him right. Corbyn refuses to reveal the full details of his theory. In fact, he will not tell anyone - not even his own employees - exactly how he makes his forecasts, though he insists that the rules are safely locked away in case he gets hit by a bus. Not that it was always so. Corbyn was a schoolboy meteorologist who, back in the early '60s, had a penchant for writing up his experiments.
Corbyn says that the day he was born, March 10, 1947, nature put on a spectacular show. One of the largest sunspots ever recorded had appeared on the sun. And western England, where he was born, saw a sudden end to a miserable winter. "It was called the Great Sunspot of 1947 - it's in books, and I've seen pictures. And I was also born on the day of a thaw. It had been a tremendously cold winter up to then and that day ... there were gales and a heavy rain and the thaw just starting to set in. It's all true - my father told me."
That's the only known portent in a life in which sunspots and weather have become entwined. The stirrings of his scientific interest were modest: "I was asking my father lots of questions, the way children do. Like, how does cement set? I was very inquiring - my father always said that. I'd cover everything, ask questions that there weren't really answers for." He also built his own observation station. He had some thermometers, a homemade rain gauge, and, to measure wind speed, an anemometer cobbled together from pieces of copper sheet, some curtain rod, and part of a bicycle wheel. In 1962, when he was 15, he started keeping daily records. "I measured that extraordinary winter in 1962 to 1963, when it reached minus 17 degrees," he remembers.
At 18, Corbyn entered London's Imperial College of Science, Technology, and Medicine, and at 20 he was first published in Weather, a journal of the Royal Meteorological Society. He wrote articles about how he built a brine-filled plastic barometer that extended 30 feet up the side of a house and about a design for an electrical thermometer that tracked average daily temperatures. Evincing a taste for public life mirrored in his brother, Jeremy - now a Labour Party member of Parliament - he was elected president of the college student union. By the time he left the college with a degree in physics, he had participated in anti-Vietnam protests - "What was the chant, 'Hey, hey, LBJ, how many kids did you kill today?'" - and became involved in what he now calls "revolutionary radical groups."
He returned to science in 1979, going to Queen Mary College just outside London to pursue a master's in astrophysics. As part of the course's basic inquiry into star formation and development and the structure of the sun, he became interested in sunspots. And in doing that, he says, he remembered a book his Aunt Mary had given him when he was a child - a book that included a brief reference to the possibility of a link between solar activity and terrestrial weather.
Speculation over such a link, and the possibility that it influences life on Earth in unexpected ways, goes back nearly two centuries. In 1801, British astronomer William Herschel noticed that the price of wheat fluctuated with the number of sunspots, which in turn seemed to coincide with warmer weather. In 1894, English astronomer E. W. Maunder published evidence of a seven-decade decline in the number of sunspots, from 1645 to 1715, that paralleled part of a prolonged period of cold in Europe now called the Little Ice Age. Studies of the sun and solar activity continued, and beginning in the late 1950s, an ever more ambitious series of space probes confirmed basic theories - the solar wind, the existence of Earth's protective magnetosphere. If there's a lesson to be learned from a brief review of the study of solar-terrestrial interaction, it's how recent the bulk of our detailed knowledge is and how much remains to be investigated.
Corbyn's idea for measuring sunspots was to use weather as a kind of spot meter. "I thought I should treat Earth's weather as a detector of particles from the sun," he says. To do that, he got hold of historical records of solar activity and the corresponding terrestrial weather. He noticed right away that there did indeed seem to be evidence of a link between the two and that his spot-meter idea ought to work. Then, Corbyn says, he realized that this process might also work the other way around: that by watching solar activity - some aspects of which can be predicted months and years in advance - he might be able to predict the weather months in advance too. "And I thought, 'To hell with the sunspots, the weather is more important!'"
So in the early '80s he got busy, working with sunspot figures and weather records. By December 1986, he had scribbled a tentative forecast that he carried with him during a visit to his mother's home in Wiltshire. He had forecast a sudden cold snap in January. But over the next couple weeks, the weather was unusually mild - the temperature didn't even dip below freezing for a week. Corbyn had almost forgotten about his prediction when he left his mother's thatched cottage early in the morning on January 12 to bicycle the six miles to the train station.
"It was about 6:45, and there was a light dusting of snow, and a bitter wind," he recalls. "I was going up this hill, and there was frost forming on my beard. The sun was just rising, and I thought, 'Shit, this is cold.'" The temperature was minus 12 degrees Celsius. Corbyn's predicted cold snap had materialized. "I thought, 'So it's working! There must be something to it.'" And with that he set out to develop his new weather-forecasting technique with renewed enthusiasm.
The weather forecasts most of us see every day are produced by supercomputers churning through mountains of raw data to build a three-dimensional picture of the atmosphere and then, using mathematical equations, to project how that picture will change a day, three days, or a week into the future. As anyone who has paid even passing attention knows, the further into the future, the less reliable the forecast.
The main reason is that the awesome computing power brought to bear on forecasting works with flawed data. Despite a sophisticated network of weather stations and buoys scattered over Earth's surface, despite an international weather air force of planes, balloons, and satellites, despite a global network that keeps up a constant flow of readings from all of the above, the initial data is imperfect. The main problem is that the reporting stations are spaced irregularly, and for large parts of the globe - a sizable chunk of the Pacific, for instance - readings don't yet exist. To account for this, the data must be mathematically massaged and missing values filled in by what amounts to educated guesswork. Finally, let's not forget Nature's unescapable monkey wrench: chaos. Weather evolves by the rules of chaos - that is, by mostly indecipherable rules. Tiny, ephemeral changes in temperature or barometric pressure can cause huge variations for which the models cannot account.
The UK Meteorological Office, armed with its monster Silicon Graphics/Cray machine, is a good example of how forecast reliability deteriorates over time. The office cranks out highly accurate pictures of the next day's weather: 86 percent of the time, it predicts the temperature to within 2 degrees Celsius and can tell you within a few hours when it will start raining. Three-day forecasts hit about 70 percent on target - they are as accurate today as one-day forecasts were 25 years ago. Five-day outlooks are substantially less reliable. The weather folks can improve their odds for five-day forecasts by comparing overlapping results from supercomputers in different countries - when several agree, it's a good sign. But, says Ewen McCallum, head of forecasting at the Met Office, "no matter what model you've got, after day seven the drop-off of accuracy is very fast. Most scientists believe there is a limit to deterministic forecasting."
Don't try telling that to Mark Bailey, a special markets manager for the Yorkshire Electricity Group. The utility company was one of Corbyn's early customers after he went into the long-range-prognostication business in 1990; it uses the service as part of its effort to anticipate winter demand for power.
"I would put the success rate of Corbyn's forecasts in excess of 70 percent," says Bailey. That sounds pretty good for a general outlook done months in advance. But Bailey is still floored by a far more personal example of what he regards as Corbyn's forecast effectiveness. He had become so convinced that Corbyn was reliable that in early 1995, when he was making plans for his wedding the following September, he asked Corbyn to predict the weather for the big day.
In March, six months before the event, Corbyn "predicted that on Saturday we'd have a fine sunny morning, becoming hazy in the afternoon, and rain in the late evening into the night." When the day came, Bailey recalls, "it was an excellent morning, a bright afternoon, and when we came out of the reception at one o'clock in the morning, it was pouring with rain. He got it spot-on."
Most of Corbyn's long-range prognostications are more general, but still come accompanied with forecast maps that include specific locations of fronts and pressure centers. The outlooks often include boldly specific language weeks ahead of an event: "Deep thundery depressions - probably tropical storm remnants or after-effects - cross the British Isles from the west and south bringing thunderstorms with torrential rain in places and generally much cooler conditions." Compare that with the 30-day outlook - not "forecast" - for December from the National Weather Service's Climate Prediction Center: The only clear statement was that La Niña, the cooling of surface sea temperatures in the tropical Pacific, made things very hard for the mathematical models. It might be dry or wet, cold or not-so-cold, depending on your location.
Corbyn and company make the most out of it when Weather Action's forecasts hit close to the bull's-eye. The service's Web site (www.weatheraction.com) flashed a message for weeks after an April 1998 forecast for a storm near the end of October bore fruit. But some of Corbyn's claimed successes leave plenty of room for interpretation about just how useful they actually were. Weather Action's monthly bulletin for last October, for instance, announced that "hurricanes or intense hurricanes are predicted likely to hit US coasts or the Caribbean ... The high danger time windows for formation of those storms which are likely to head for land are 6-9 Oct. and 20-23 Oct." As it happened, two hurricanes, Lisa and Mitch, did come to life close to the predicted times. But the first, Lisa, became an official hurricane - a minimal one, at that - much later than Weather Action predicted and never threatened the US or the Caribbean. Mitch was a storm of historic violence, but it largely blew itself out before its landfall - in Central America, and the disaster it triggered was caused by incessant rainfall after it stalled. That reality didn't stop Weather Action from declaring in its November bulletin, "The two major Atlantic tropical storms in October were both predicted by Weather Action."
With his first paying customers signed up, Corbyn continued to refine his method. It wasn't until 1995, after nearly five years and with several new clients coming aboard, that he took on two partners and formed Weather Action. Customers now include Britain's Coca-Cola bottler, agrochemical giants Monsanto and Hydro Agri, and a number of petrochemical companies and utilities. British Gas Services uses Weather Action's forecasts to decide how many emergency repair crews to keep on standby. Larger customers (Weather Action commercial director Melvyn Wallace says big accounts number "in the dozens, not the hundreds") may pay an annual fee of £24,000 (about $40,000) and upward for Weather Action's continually updated forecasting service and deeper analysis of climate trends. Most services cost anywhere between £480 to £6,000 (about $800 to $10,000) a year. One of the company's most widely used products - short-, middle-, and long-range regional forecasts by fax and phone - draws "hundreds" of calls from farmers, gardeners, and other customers every day from the spring through the end of summer.
When promoting the film Bean, PolyGram relied on Corbyn's forecasts to identify rainy weekends in advance and book extra television ads (people are more likely to go to the movies on a drab, rainy weekend). Filmmakers use Weather Action's forecasts when planning location shoots, because a delayed shoot can cost upwards of $160,000 a day. Rick Brownlow, a specialist insurance broker, offers discounts for film and advertising production companies willing to arrange their shooting schedules around Corbyn's predictions. "This is something the industry has been crying out for," says Brownlow. "For a feature film, people start scheduling up to 12 months in advance. Using traditional methods, nobody would have a clue about what the weather would be like."
Weather Action went public in October 1997 on London's Alternative Investment Market, and it was initially valued at £6 million (though its market cap had fallen to £4.25 million by late fall 1998). The company employs 16 people, including five full-time weather forecasters and five people doing background research on how solar activity interacts with the weather. Revenue for '97 was about £250,000, a figure Corbyn says grew markedly in '98 (by comparison, the Met Office's commercial arm employs 137 people and in 1997 brought in £25 million).
It's clear that some businesses, large and small, are willing to pay good money to get Corbyn's picture of the weather future. But Weather Action faces an uphill struggle convincing new clients that its long-range forecasting methods really work. This is due largely to an outcry from mainstream meteorologists who apparently fear that the company's tendency toward self-promotion, coupled with lack of clarity about its methods, is giving commercial forecasting a bad name.
"Site-specific forecasting for specific weather events a season ahead is scientific mumbo-jumbo," says the Met Office's McCallum, one of Corbyn's most outspoken critics. "I could make a forecast for the month after next, but I wouldn't expect you to pay money for it."
So how does Corbyn do what he does?
His Solar Weather Technique combines statistical analysis of over a century of historical weather patterns with clues derived from solar observations. He uses what he calls Weather Action Indicators, "a system of rules that tells you where to look in the past for relevant data, for similar weather patterns." He has spent years fine-tuning these rules - taking into consideration such factors as the relative orientation of the sun and Earth, and the direction of stratospheric winds - to increase the accuracy and detail of his forecasts.
Each day, Corbyn obtains the latest available solar observations - from US government Web sites, "courtesy of the American taxpayer" - and calculates a set of parameters predicting how the Weather Action Indicators will vary over the next few months or year. He then looks for similar states that have occurred in the past, and stitches together "replays" of individual weeks of weather patterns. He also looks for what he calls "red spikes" - sudden changes in Weather Action Indicators. "Those are very important for timing," he says.
Corbyn is not entirely alone in at least considering whether the sun plays a part in the weather, beyond its long-understood role as a heat and light source. Most government-supported science tends toward a cautious view - allowing, for instance, that the influence of solar activity on the weather is not completely understood and deserves study. More striking is research published in 1997 by Eigil Friis-Christensen and Henrik Svensmark of the Danish Space Research Institute. Their work suggests that the intensity of cosmic particle radiation - a factor known to be regulated by solar activity - plays a role in Earth's cloud cover. High levels of solar activity, they say, correspond with relatively low cosmic radiation, and low radiation levels appear to be associated with a thinning cloud cover and thus higher global temperatures. Scientists at CERN, the European particle physics laboratory, have proposed an experiment to test the Danish theory.
Among weather folk, though, the attitude of Tim Stockdale toward solar influences appears typical. Stockdale, a scientist at the European Centre for Medium-Range Weather Forecasts in Reading, England, concedes that fluctuations in the output of the sun influence the atmosphere. But, he says, "the variations are very small, too small to have any real impact on a three-month time scale, let alone on a day-to-day basis." Besides, Stockdale adds, Corbyn's method "goes against the laws of physics and the mathematical description of the atmosphere as we understand it."
Much of Corbyn's problem with other scientists arises from his continued reluctance to publish the details of his method in peer-reviewed journals. "For 10 years the scientific community has asked him to publish the scientific basis for his forecasts," says Philip Eden, a former council member of the Royal Meteorological Society who has written several books on Britain's weather and pens a weekly column on the topic for The Sunday Telegraph. "All we want is for him to describe his methods, just as any reputable scientist would, and allow his results to be analyzed objectively. I would love him to have discovered the key to long-range forecasting. But nothing I have seen has given me any indication that he's got it."
Stockdale, too, says Corbyn will not be taken seriously by the scientific community unless he agrees to play by its rules. "This could be a hugely important piece of science," he says. "We understand that he might have reasons for wanting to keep his methods secret, but he should have to demonstrate them convincingly."
"Who is to say he is wrong?" counters Allen Webb, director of the UK arm of Oceanroutes, the world's largest private forecasting agency. "We've been amazed at the attitude that the Met Office takes. Their stance is based on the fact that Mr. Corbyn has not published any papers, but then he has started up a commercial operation, so why should he? Factions of the industry have said he must be wrong because they don't agree with him. It's a very closed-minded approach."
It must be said, the situation is very different from that of 1922, when Richardson published his results: There was no private forecasting industry back then and he had little to lose by revealing his methods. But if Corbyn were to give away his secret, he'd also be giving away his company's key asset. "We keep our techniques under wraps," says Melvyn Wallace, Weather Action's commercial director, "because we want to have the market advantage until we've built up the company."
For his part, Corbyn promises that his critics (and competitors, for that matter) will have their chance to see what's under the hood - eventually. "The idea that something isn't science because it hasn't been published is complete nonsense," he argues. "I would estimate about 60 percent of world science is totally secret - it's run by the military. You can't say the stealth bomber or British atom bomb weren't science because they were secret. We do intend to publish and we will choose the time. But there are plenty of questions that we still haven't understood as to why what we do works. If we publish now, there would be as many questions as there would be answers."
While the world waits, some have begun trying to assess Corbyn's forecasts. Two analyses by Dennis Wheeler, a researcher at the University of Sunderland, conclude that it's highly unlikely Corbyn's "hits" - gales, storms, and cold spells predicted as much as two years in advance - were the result of chance. Not exactly the kind of claim you can put in an ad - "Really, folks, it's not luck!" Still, Wheeler puts the odds at "several hundred to one against getting Corbyn's levels of success."
Wheeler's research was commissioned by a group of insurance companies looking to get the jump on storms and other natural disasters that might mean big payouts to clients. Wheeler won't reveal the names of the firms involved, citing competitive advantage - if they admit to using Corbyn's forecasts, their rivals could start using them too. But Wheeler is writing an academic paper based on Corbyn's work, which he intends to submit to Meteorological Applications, a theoretical journal published by the Met Office.
In October, further endorsement for Corbyn's Solar Weather Technique came with the announcement of a partnership with Reuters, which will flash Weather Action's 30-day forecasts across trading screens in the commodities and energy markets. Robert Fish, a product manager at Reuters, points out that he would not be making Weather Action's forecasts available if he thought they were inaccurate. The same month, Weather Action also signed a deal with the Information Service, one of the largest suppliers of premium-rate telephone-based information in Britain, to provide 14-day forecasts by fax, along with specialist services for skiers and gardeners.
Even though hard evidence for a sun-weather link would bolster Corbyn's case, it would only imply that a weather-forecasting system based on solar observations might be feasible. It would not prove that Corbyn's particular set of rules actually work, and he would still have to convince the skeptics that it really is possible to relate specific solar events to specific weather events. When asked what he thinks of Corbyn, the Danish institute's Svensmark entertains the possibility that a new approach could succeed. "It's impossible to do a scientific check" he says, "but my guess is that he's using correlations of some kind to make empirical rules, which might be working."
Perhaps the key to assessing the debate is this: Simple common sense says that much more is unknown than known about the forces that drive the weather. It's equally certain that scientists investigating the workings of climate - what we've taken to be a phenomenon happening entirely inside our own little cocoon in the universe - will uncover more and more ways in which outside forces push us and pull us.
Corbyn's conviction that the sun's behavior is the driving force behind earthly weather patterns is linked to another deeply held belief: The sun and radiation from outer space play a far more important role than the burning of fossil fuels in any global warming that might be taking place. The purveyors of the principal theory of global warming, he says, have sold the world a bill of goods. "If you piss in a lake, the level will go up," says Corbyn. "But it wouldn't be an important factor. [Human contribution to carbon dioxide levels] is not as minuscule as that, but it's not important."
Eventually, Corbyn hopes that his work will lead to the rise of a new meteorology, combining old-fashioned supercomputing with newfangled solar factors. He believes his techniques for forecasting will prove as influential in the 21st century as Lewis Fry Richardson's numerical methods have proved in the 20th. Consider that Richardson devised his theories using pencil and paper in a freezing barn on the battlefields of France and that they now form the basis of a multibillion-dollar industry. Suddenly the idea of another revolutionary weather-forecasting technique emerging from Piers Corbyn's shambolic south London office doesn't seem so ridiculous after all.
And who knows? Maybe in a couple of generations, Piers Corbyn will have a building named after him too.
PLUS
