Each year, the Moscone Center, a cavernous underground complex of halls a couple of blocks from the heritage cable cars and thrusting skyscrapers of downtown San Francisco, welcomes tens of thousands of scientists from around the world. From the study of the sun’s surface to the interior of the earth’s core, from the weather on Mars to our own world’s ocean currents, this huge meeting covers the latest advances in the geophysical sciences. It’s quite a circus.

The wide subterranean lobbies bustle with activity. People queue up in front of coffee stations and hurry in and out of restrooms, old friends gather in sociable huddles and collaborators sit together at tables peering at laptops. In odd corners, smartly dressed young scientists, hopeful of impressing future employers, silently rehearse their upcoming presentations.

But it is through the double doors in the meeting rooms themselves that the real action of the week takes place. In a host of halls across the complex, postgraduate students, postdoctoral researchers, tenured academics and emeritus professors present their latest research findings from microphoned platforms to serried ranks of chairs. Delegates who are not presenting at that time can pick and choose from the multiple sessions taking place in parallel. They can dip in and out, grazing for the scientifically most interesting titbits, or settle into a session for the duration, spend the time with a community of specialists and catch up with all their latest progress. The scale of the centre and the grandeur of the rooms means there are usually enough seats for everyone.

Even before our session in December 1996 began, every spot was taken. The theme – the detection and attribution of climate change – had until recently been rather obscure and pursued by only a handful of researchers. But this year it had been thrust into the limelight. Right at the start of my career in climate science, I was going to be presenting my early results to a huge crowd. It felt like a very intimidating initiation into my new field of research.

At the start of the year, the Intergovernmental Panel on Climate Change (IPCC) had published its latest report.³ Previously, the United Nations body charged with assessing the latest scientific understanding had not been able to say whether past warming was human-induced or natural in origin. Now, it had come to a very different conclusion, that ‘the balance of evidence suggests a discernible human influence on global climate’. It was a conclusion that would have major repercussions.

A carefully considered summation of the latest findings by the small band of researchers who made up the International Detection and Attribution Group (which had been founded only the year before), the ‘discernible human influence’ statement prompted a bitter controversy. For the first time, the scientific consensus was that the finger of blame for climate change pointed firmly towards emissions of fossil fuels. This was not what the fossil fuel industry and many lawmakers in the US Congress wanted to hear. In recent months, they had tried to dispute the IPCC’s findings in a controversy that had reached the pages of national newspapers and been aired in Congressional hearings. The two leading protagonists in that dispute were due to speak at today’s session: Ben Santer, the climate scientist who had taken the leading role in crafting the ‘discernible human influence’ statement, and Patrick J. Michaels, State Climatologist of Virginia and vocal critic of the IPCC.

With that in mind, I shouldn’t have been surprised to see so many people here. But it still came as a shock. For the first time it struck home that this research field of mine was of interest to an awful lot of people.

I had arrived in good time with my two colleagues from Britain: Simon Tett, my mentor from the Met Office Hadley Centre, and Myles Allen, his energetic collaborator from the University of Oxford, and we had found a place near the front next to an aisle. Even before the talks had begun, people had lined up against the back wall and along the huge sliding partitions that separated our hall from the much quieter ones on either side. Late arrivals were leaving again through the thick double doors behind us, looking disappointed. There was an excited hubbub of chatter and some of the other speakers were standing around at the front getting a feel for the atmosphere.

Despite the crowds, ours was a regular session of the conference that featured many technical presentations that provided incremental advances to scientific knowledge. My talk was going to present one of those advances. After just four months working on the subject of climate change, I could hardly have expected to have already made an earth-shattering discovery. But the work I was about to present was relevant to the high-stakes research question that had drawn in the crowds. I hoped people would find my findings interesting and worthwhile. Most of all, with so many eyes trained on me, I hoped I wouldn’t make a fool of myself.

A hush had descended, the lights had dimmed, and the chair was inviting the first speaker up to the raised podium to begin the session. Not a word of what they said went in, nor did any of the colourful pictures projected on to the giant screen in front of me make any sense. Instead, I was consumed with apprehension as to how I would appear to all these hundreds of people, once I too had climbed the steps, clipped on my microphone and begun to talk. The time raced by. The audience clapped my predecessor. And then, without even being aware of how I had got there, I found myself standing on the same spot that the previous speaker had just vacated.

I dared to look down at the dimly lit sea of faces in front of me, took a deep breath and began to speak. Using data crunched by the Hadley Centre climate model, I told them, I had compared trends in surface temperatures with those expected from natural climate oscillations. Warming at the earth’s surface, I had found, was now outside the envelope of temperatures that could be explained by natural processes.⁴ To detect changes in climate, you had to look at the data over twenty years or more. Over a decade, temperatures could cool, even in the presence of greenhouse warming. With climate change it was important to look at the long-term picture. When you did, the data showed that recent warming was highly significant.

In a flash, my allotted time was up. People clapped politely and I was free to return to the safe anonymity of my seat in the vast crowd. A career hurdle overcome, I was at last capable of listening to the other speakers. Soon we would come to the main attraction. But first we would hear from other members of the supporting cast, including those of my slightly more experienced British colleagues, Simon and Myles. It was their boyish enthusiasm that had infected me with a strong passion for my newfound area of research. They had taught me a lot, these peers of mine, since I joined the Met Office Hadley Centre in July. They had been investigating the causes of climate change for over a year. I had several years’ experience in other branches of atmospheric science, including researching the environmental consequences of the Chernobyl nuclear accident and the depletion of ozone by destructive chemicals. But I was new to this particular specialism and I had much still to learn. Compared with me, my two colleagues were veterans.

Simon rushed through his latest results with eager excitement. He had studied atmospheric temperatures high above the ground as measured over recent decades by weather balloons. He found that how they were changing could best be explained by taking account of human activities. His results confirmed previous expectations, that greenhouse gas emissions should warm climate, affecting temperatures not just at the surface but aloft as well.⁵

Myles was up next and looked totally at home in front of a large and expectant audience as he expounded his new idea. He wanted to improve our understanding of climate change by developing a new method for working out exactly how much warming was caused by human activities. Air pollution in past years had obscured the sky and shielded the earth from some of the sun’s rays, holding back warming from the increasing greenhouse gases in the air. Exactly how much was still uncertain. Myles had developed a set of detailed equations to work it all out. His engaging style was attractive even though I couldn’t see many people following his complicated mathematics.⁶

There were other speakers from other institutes. Like ours, they were technical addenda to the main business that had filled up this hall to overflowing. It was normal business during a technical session of a scientific congress. But unusually, rather than presenting to a handful of interested specialists, this time we had found ourselves presenting to the massed ranks of the world’s geophysicists.

At last, we had arrived at the promised showdown when the two principals would be invited to make their respective cases. Patrick J. Michaels strode confidently around the stage, eschewing the mathematical equations and colourful illustrations that had featured in previous talks. Instead he talked to a sequence of bullet-pointed position statements projected on to the giant screen behind him. Global warming was not a problem, he claimed.⁷ Natural processes caused much larger changes in climate than any human activities could produce. The IPCC had become politically compromised.

And then it got personal. According to Michaels, the work of Ben Santer was fundamentally flawed. Michaels claimed that the changes Ben had attributed to human activities in a paper recently published in the prestigious scientific journal Nature were in fact entirely natural. Temperatures in the lower part of the atmosphere had warmed differently in the southern hemisphere than the northern hemisphere, a feature that could not be explained by human causes according to Michaels.⁸ Not just that, Ben had been instrumental in distorting the latest IPCC report, including its conclusion that there was ‘a discernible human influence on global climate’, for political ends. The effect of this political misuse of science, he claimed, was to deceive policymakers and the public into falsely believing human activities were causing global warming thereby promoting unnecessary restraints on economic growth that would destroy world economies.⁹

It was a slick presentation, easy to follow and clear in its conclusions. It was also now clear what all the fuss was about, why every seat was taken, and why an expectant stillness had fallen about the audience as we waited for Ben’s response. Given the recent history of the climate change issue, I too was eager to hear what he was going to say.

The reality of the earth’s greenhouse gas effect had been established back in the nineteenth century. And concentrations of carbon dioxide in the atmosphere, the main greenhouse gas associated with human-induced emissions, had risen steadily since monitoring began in the late 1950s. But widespread awareness of climate change as a global issue did not emerge until the late 1980s with a growing concern that atmospheric temperatures were also starting to rise. The United Nations established the Intergovernmental Panel on Climate Change in 1988 to assess scientific understanding of the issue. The same year, British prime minister Margaret Thatcher gave a speech at the Royal Society, in which she said that humanity had ‘unwittingly begun a massive experiment with the system of this planet itself’, and two years later she opened the new Hadley Centre for Climate Prediction and Research at the Met Office.¹⁰ Then, in 1992, the United Nations held the Earth Summit in Rio de Janeiro at which nations agreed to stabilize greenhouse gas concentrations in the atmosphere ‘at a level that would prevent dangerous anthropogenic (human-induced) interference with the climate system’.¹¹

With the potential for global action on climate change, lobby groups started to form to fight back against possible regulation. Generously funded by the oil, auto and coal industries, groups like the Global Climate Coalition, founded in 1989 by ExxonMobil and the American Petroleum Institute, supported a small group of contrarian scientists – of which Michaels was one – to attack the science of climate change.¹² For many years, climate science had received little attention outside of research labs and specialist conferences. By the mid 1990s, all that had changed.

It was the emerging field of detection and attribution of climate change that was bearing the brunt of this new attention. In past years, the focus of climate research had been on trying to predict global warming by the end of the twenty-first century, a distant prospect of no immediate concern to many people. But now, thanks to the research carried out by Ben and other members of the International Detection and Attribution Group, the IPCC had concluded that climate had already, by the mid 1990s, changed significantly. The threat posed by emissions of greenhouse gases was no longer a distant one requiring action in years to come. Instead, we had shown, it was a present one, requiring action now.

The world was facing a huge decision, whether to cut back its use of fossil fuels and, in doing so, radically change its entire means of energy generation. That decision rested on the headline conclusion from the IPCC that humans were influencing climate. In claiming that this finding was fundamentally flawed, Michaels was disputing not just Ben’s scientific insights and integrity but also the entire basis on which many nations were now arguing climate change should be urgently addressed.

Having met him a few days before at his lab nearby, the claim that Ben was part of an international conspiracy to fraudulently mislead the American public seemed ridiculously far-fetched. But this audience knew, like I did, that the scientific claims Ben was making were not beyond reasonable challenge. Identifying the signal of human-induced climate change amid the large natural variations of our constantly varying climate pushed at the limits of current climate science. Observations were limited and climate models were still in the early stages of development. The crowd of sceptical geophysicists here in the hall were not going to accept claims about the causes of climate change, whichever way they went, without convincing evidence.

Ben’s presentation style was very different from Michaels. Rather than roam about the stage like Michaels had done, Ben sheltered behind the lectern while he earnestly and laboriously laid out his case by means of a sequence of detailed illustrated slides. He first explained how atmospheric temperatures had altered over the last twenty-five years, as observ...