Chapter 1
Born in a War
The balance of evidence suggests that there is a discernible human influence on climate.
âThe contentious sentence in âSummary for Policy Makers,â IPCC Second Assessment Report (1995)
It is November 27, 1995, several years before my colleagues and I published our âhockey stickâ study. Bill Clinton has been president for nearly three years. The Dow Jones Industrial Average just passed the 5,000 mark for the first time. The TV series E.R., created by novelist Michael Crichton, is the top-rated show on television.
In Madrid, Spain, the Intergovernmental Panel on Climate Change (IPCC) is holding the final plenary meeting for the Second Assessment Report, the purpose of which is to summarize the consensus among scientists regarding the extent of humanityâs impact on Earthâs climate. At a nearly identical latitude on the other side of the Atlantic, I am working on my Ph.D. dissertation in New Haven, Connecticut. I am oblivious to what is taking place in Madrid; Iâm just trying to finish up my research in time to defend my dissertation the coming spring and begin a career as a professional climate researcher.
My research at the time focused on the importance of natural variabilityâthat is, the role of nature, not manâin explaining changes in Earthâs climate. The one scientific article I had submitted for publication that touched on the topic of human-caused climate change would, ironically, a few months later be hailed by those who contest the proposition that humans play a significant role in observed climate changes. (That article simply demonstrated a relatively minor inconsistency between theoretical climate model predictions and actual climate observations.)1 I was especially interested in the role that natural oscillations in the climate system might have played in observed changes in climate during the modern observational era of the nineteenth and twentieth centuries. My first-ever article analyzing climate proxy records had just been published a week earlier.2 In that article, my coauthors and I showed that these natural oscillations persist over many centuries and might be more important than many scientists had acknowledged in explaining certain modern climate trends. This work, in another twist of irony, would also be celebrated by contrarians in the climate change debate, who were ostensibly unaware that both natural and human influences on climate can and almost certainly do coexist. I myself did not doubt that humans were changing the climate; the extent of evidence was already significant. I had simply chosen to focus in my research on the issue of natural climate variability.
Meanwhile, back in Madrid at the IPCC plenary, a fierce argument had broken out between the scientists crafting the report and government delegates representing Saudi Arabia, Kuwait, and some other major oil exporting nations that profit greatly from societal dependence on fossil fuel energy and, according to the New York Times, had âmade common cause with American industry lobbyists to try to weaken the conclusionsâ of the report.3 The tussle was over whether one could state with confidence that human-caused climate change was already observable. The scientists argued that âthe balance of evidence suggests an appreciable human influence on climateâ because only when human impacts were included could the rise in temperature over the past century be accounted for. The Saudi delegate complained that the word appreciable was too strong. He demanded weaker wording.
For two whole days, the scientists haggled with the Saudi delegate over this single word in the âSummary for Policy Makers.â They debated, by one estimate, nearly thirty different alternatives before IPCC chair Bert Bolin finally found a word that both sides could accept: âthe balance of evidence suggests a discernible human influence on climate.â The term discernible established a middle ground by suggesting that climate change was indeed detectable, as the scientists argued, while acknowledging that humanityâs precise role in that change and its magnitude were still subject to disputeâa concession that no doubt pleased the Saudi delegate. This sentence would go on to become famous or, in some circles, infamous. The fact that two entire days at the final plenary were devoted to debating a single word in the reportâs summary gives you some idea of how contentious the debate over the reality of human-caused climate change had become by 1995.
Why did the scientists care so much about the wording? What would be the harm, after all, if the wording were weakened a bit? I suppose it comes down to how deeply scientists care about getting things right. Details matter, and we argue passionately with each other about them. We donât suffer perceived inaccuracies lightly, and more than anything else, we donât like being misrepresented. The fact that the science in this case might have deep real-world consequences only amplified these natural inclinations.
Among the scientists who fought hard against any watering down of the reportâs key conclusion was Ben Santer, a climate specialist who works for the Department of Energyâs Lawrence Livermore National Laboratory in California. The recipient of a coveted McArthur âgeniusâ award in recognition of his groundbreaking contributions to our understanding of climate change, Santer was a primary author on a series of important papers establishing the human role in observed climate change. As such, Santer was in a better position than anyoneâand certainly than a bureaucrat with a political agendaâto assess the level of scientific confidence in concluding that human activity was changing the climate.
As it happens, I had met Santer for the first time a little more than a year earlier, in July 1994, at a two-week workshop on climate science at the National Center for Atmospheric Research. I was attending the workshop as a graduate student invitee, and Santer was one of the invited speakers. I asked him a question about certain details of his analysis following his presentation. His response came across as a bit defensive, as if he perceived my question as an attack. Only later would I understand why.
Santerâs work on climate change detection, unbeknownst to me, had been under increasing attack from contrarians in the climate change debate. In 1994, for example, his findings regarding the match between observed and model-predicted surface temperature changes was criticized4 by Patrick Michaels, a University of Virginia climate scientist who edited the World Climate Report,5 a newsletter with fossil fuel industry funding6 that featured criticisms of mainstream climate change research.
The attacks against Santer were ratcheted up dramatically following the November 1995 IPCC plenary. In February 1996, for example, S. Fred Singer, the founder of the Science and Environmental Projection Project and a recipient over the years of substantial fossil fuel fund-ing,7 published a letter attacking Santer in the journal Science.8 Singer disputed the IPCC finding that model predictions matched the observed warming and claimedâwronglyâthat the observations instead showed cooling. Singer went further. He claimed that inclusion of Santerâs work in the report violated IPCC rules because the work hadnât yet been published. In fact, the IPCC rules did not require a work cited to be published at the time of the report; if it did, the lag time involved in getting a publication to print would essentially render the report obsolete on arrival. The IPCC requirement was simply that the work be available to reviewers upon request, which Santerâs work was. Moreover, a substantial component of the research in question had been published.
Meanwhile, the Global Climate Coalition (GCC), a group also funded by the fossil fuel industry,9 circulated a report to Washington, D.C. insiders accusing Santer of abusing the peer review system and of âpolitical tamperingâ and âscientific cleansingââa charge that was especially distasteful given that Santer had lost relatives in Nazi Germany.10 The purported basis of these allegations? At the request of the IPCC leadership after the Madrid plenary, Santer, as lead author on an IPCC chapter, had removed a redundant summary so that his chapterâs structure would conform to that of the other chapters, all of which had summaries only at the beginning.
A few months later, Frederick Seitz, the founding chairman of another industry-funded group,11 the George C. Marshall Institute, published an op-ed in the Wall Street Journal repeating the same charges. While the paperâs editors did eventually publish Santerâs rejoinder, they in effect neutered his response by editing his words beyond recognition and removing the names of the forty colleagues who had cosigned the letter, thus leaving Journal readers with the misleading impression that Santer stood alone in his defense against the specious charges.12
With the help of sympathetic media outlets such as the Wall Street Journal, climate change deniers were able to spread false charges about Santer faster than heâor his colleaguesâcould possibly hope to refute them. The practice of isolating someone like Santer to make an example of an individual scientistâwhat I call the âSerengeti strategyââis a tried-and-true tactic of the climate change denial campaign. The climate change deniers isolate individual scientists just as predators on the Serengeti Plain of Africa hunt their prey: picking off vulnerable individuals from the rest of the herd.
The Santer episode encapsulates the toxic and incendiary environment that existed, largely unbeknownst to me, at the time that I was finishing my Ph.D. and preparing to enter the world of climate research. Little did I know that similar attacks might be made against me just a few years hence, when my work, like Santerâs, would be featured as a major pillar of evidence by the IPCC.
Tricks and Treats
It is late December 1974. Iâve just turned nine, and, as usual, my family and I are celebrating my birthday with relatives in Philadelphia. For more than a year now, Iâve been pestering my Uncle Paulâan artist and successful entrepreneur to whom Iâd always looked for wisdom on all matters of lifeâto explain what it means to go faster than the speed of light. I was intrigued by such âgee whizââbut ultimately scientificâconcepts. For my birthday that year, Uncle Paul had given me a copy of a popular novel considered inspirational at the time, with an inscription indicating that it would answer my questions. I enjoyed the book, though to this day I canât figure out what it had to do with warp speed, time travel, or any related topics. But I know that already by that age I was fascinated with the world of science.
Math and science were the subjects that had always come most easily to me; perhaps having a father who was a college math professor had something to do with it. In high school, when other kids were partying on Friday nights, I was hanging out with my computer buddies writing programs to solve challenging problems. In fall 1983, after having seen the movie War Games, I became determined to write a self-learning tic-tac-toe computer program, just as in the movie, a program that could learn from its mistakes, a rudimentary type of artificial intelligence. The movie carried a thinly veiled lesson about the futility of global thermonuclear war: There can be no winner in a tic-tac-toe game expertly played; if neither player makes a mistake, the game will always result in a tie. Perhaps if the computerâin the movie, it had seized control of Americaâs missile program and was preparing to launch a massive nuclear attackâcould be brought to understand this paradox, it could recognize the futility of nuclear war. For me at the time, however, it was just an interesting and challenging computer problem to tackle.
Machine learning of this sort was in principle relatively straightforward. The real challenge was in how to go about constructing an algorithmâa set of operations or calculations, here in the form of a computer programâto solve the problem as efficiently and elegantly as possible. I had the computer play itself, just like in the movie. That was the easy part. In the beginning, I simply had it make random moves every turn. When it lost to itself, however, I would store both the final and previous configurations of the tic-tac-toe board in a âblacklistââmoves that would no longer be available to the computer.13 The blacklist was used to ensure that the computer, while it continued to make random moves, would not make the same losing move again; in this way it would gradually âlearnâ how to play tic-tac-toe.
In practice, it might take a very long time for the computer to become skilled enough to avoid losing because there are so many possible sequences of moves, and the program gets slower and slower as it has to scan an increasingly long list of disallowed moves before each turn. But I discovered a âtrickââthe term scientists and mathematicians often use to denote a clever shortcut to solving a vexing problemâto get the computer program to learn much faster. The trick was to exploit the concept of symmetry. A tic-tac-toe game is the same no matter how you rotate the board, whether you flip it vertically or horizontally, or whether you switch the role of Xs and Os. When you take that symmetry into account, there are actually many fewer truly unique board configurations and many fewer losing moves that need to be stored in a blacklist. Now I could get the computer to become unbeatable in tic-tac-toe far more readily. The adrenaline rush, for the scientist, comes from finding tricks that make a problem easier to crack. Thatâand eating pizza with my friendsâwas my idea of a fun Friday night.
A Random Walk
A year later, in August 1984, as Ronald Reagan was completing his first term in office and Michael Jacksonâs Thriller was the top-selling record album, I headed off to college at the University of California, Berkeley (UC Berkeley). In part, I must confess, I was looking to get away from the harsh Amherst, Massachusetts, winters Iâd endured for the first eighteen years of life, but I was also attracted by the schoolâs reputation as one of the worldâs leading scientific research institutions. I chose to major in physics, with a second major in applied math. The summer following my freshman year, I began doing research in theoretical physics that emphasized the computational approaches to problem solving I so enjoyed.
The project I was working on bore some resemblance to the tic-tac-toe problem that had captivated me in high school; it too involved the concept of randomness. The project employed what is known as a Monte Carlo method, named for its resemblance to a casino gameâa method that I would make use of years later in my climate research. Much as gamblers in Monacoâs famous casino town engage in random rolls of the dice in hope of monetary reward, scientists generate random numbers on a computer in hope of simulating processes in nature that have a random component.
One example is the molecular interactions that govern the behavior of a solid or liquid. While the fluctuations of the individual molecules are random in nature, external conditionsâthe ambient temperature, in particularâinfluence the collective behavior of the molecules. The warmer the temperature, for example, the more energetic the random fluctuations. Thus low temperatures favor relatively ordered states (e.g., ice crystals), while high temperatures favor relatively disordered states (e.g., water vapor). Shifts between these states are typically abrupt. There is a critical temperature at which the system, when warmed, will suddenly undergo a phase transition from the ordered state to the disordered state, or vice versa in the case of cooling. One can explore phase transitions by representing the interactions between molecules in a computer model simulation, generating random molecular perturbations in the model to mimic the real-world random fluctuations of molecules.
I was using this type of Monte Carlo approach to investigate the theoretical behavior of liquid crystalsâthe materials used in liquid crystal displays (LCDs) employed in laptops, TVs, and digital watches. My research was aimed at determining how the critical temperature of the transition between the ordered and disordered phases of liquid crystals might vary under different conditions.14 My adviser was a theoretical physical chemist named Tony Haymet, who much laterâcoincidentally enoughâwent on to direct one of the worldâs premier climate research institutions, the Scripps Institution for Oceanography at the University of California, San Diego.
When Tony left UC Berkeley a couple of years after my arrival, I continued my undergraduate research with Didier de Fontaine, a professor of materials science studying the properties of an exciting new materialâa high-temperature superconductor. A superconductor is a material that conducts an electric current with no resistance, a property with profound real-world applications such as in the operation of super-fast bullet trains. Conventional (metallic) superconducting materials need to be cooled nearly to a temperature of absolute zero, making them expensive to maintain. In the mid-1980s, scientists discovered that certain ceramic materials had a remarkable property; they super-conducted at much higher temperatures, above even the temperature of liquid nitrogen (a very inexpensive coolant).
When I joined de Fontaine and his group in late 1987, they had been working for months to model the behavior of just such a material: yttrium barium copper oxide (YBCO...