Skip to main content

A New Book Manages to Get Climate Science Badly Wrong

In Unsettled, Steven Koonin deploys that highly misleading label to falsely suggest that we don’t understand the risks well enough to take action

Hurricane Irma striking in Miami bending palm trees in the storm

Steven Koonin, a former undersecretary for science of the Department of Energy in the Obama administration, but more recently considered for an advisory post to Scott Pruitt when he was administrator of the Environmental Protection Agency, has published a new book. Released on May 4 and entitled Unsettled: What Climate Science Tells Us, What It Doesn’t, and Why It Matters, its major theme is that the science about the Earth’s climate is anything but settled. He argues that pundits and politicians and most of the population who feel otherwise are victims of what he has publicly called “consensus science.”

Koonin is wrong on both counts. The science is stronger than ever around findings that speak to the likelihood and consequences of climate impacts, and has been growing stronger for decades. In the early days of research, the uncertainty was wide; but with each subsequent step that uncertainty has narrowed or become better understood. This is how science works, and in the case of climate, the early indications detected and attributed in the 1980s and 1990s, have come true, over and over again and sooner than anticipated.

This is not to say that uncertainty is being eliminated, but decision makers have become more comfortable dealing with the inevitable residuals. They are using the best and most honest science to inform prospective investments in abatement (reducing greenhouse gas emissions to diminish the estimated likelihoods of dangerous climate change impacts) and adaptation (reducing vulnerabilities to diminish their current and projected consequences).


On supporting science journalism

If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.


Koonin’s intervention into the debate about what to do about climate risks seems to be designed to subvert this progress in all respects by making distracting, irrelevant, misguided, misleading and unqualified statements about supposed uncertainties that he thinks scientists have buried under the rug. Here, I consider a few early statements in his own words. They are taken verbatim from his introductory pages so he must want the reader to see them as relevant take-home findings from the entire book. They are evaluated briefly in their proper context, supported by findings documented in the latest report of the Intergovernmental Panel on Climate Change. It is important to note that Koonin recognizes this source in his discussion of assessments, and even covers the foundations of the confidence and likelihood language embedded in its findings (specific references from the IPCC report are presented in brackets).

Two such statements by Koonin followed the simple preamble “For example, both the literature and government reports that summarize and assess the state of climate science say clearly that…”:

  • “Heat waves in the US are now no more common than they were in 1900, and that the warmest temperatures in the US have not risen in the past fifty years.” (Italics in the original.) This is a questionable statement depending on the definition of “heat wave”, and so it is really uninformative. Heat waves are poor indicators of heat stress. Whether or not they are becoming more frequent, they have clearly become hotter and longer over the past few decades while populations have grown more vulnerable in large measure because they are, on average, older [Section 19.6.2.1]. Moreover, during these longer extreme heat events, it is nighttime temperatures that are increasing most. As a result, people never get relief from insufferable heat and more of them are at risk of dying.

  • “The warmest temperatures in the US have not risen in the past fifty years.” According to what measure? Highest annual global averages? Absolutely not. That the planet is has warmed since the industrial revolution is unequivocal with more than 30 percent of that warming having occurred over the last 25 years, and the hottest annual temperatures in that history have followed suit [Section SPM.1].

Here are a few more statements from Koonin’s first two pages under the introduction that “Here are three more that might surprise you, drawn from recently published research or the latest assessments of climate science published by the US government and the UN”:

  • “Greenland’s ice sheet isn’t shrinking any more rapidly today than it was eighty years ago.” For a risk-based approach to climate discussions about what we “should do,” this statement is irrelevant. It is the future that worries us. Observations from 11 satellite missions monitoring the Arctic and Antarctic show that ice sheets are losing mass six times faster than they were in the 1990s. Is this the beginning of a new trend? Perhaps. The settled state of the science for those who have adopted a risk management approach is that this is a high-risk possibility (huge consequences) that should be taken seriously and examined more completely. This is even more important because, even without those contributions to the historical trend that is accelerating, rising sea levels will continue to exaggerate coastal exposure by dramatically shrinking the return times of all variety of storms [Section 19.6.2.1]; that is, 1-in-100 year storms become 1-in-50 year events, and 1-in-50 year storms become 1-in-10 year events and eventually nearly annual facts of life.

  • “The net economic impact of human-induced climate change will be minimal through at least the end of this century.” It is unconscionable to make a statement like this, and not just because the adjective “minimal” is not at all informative. It is unsupportable without qualification because aggregate estimates are so woefully incomplete [Section 19.6.3.5]. Nonetheless, Swiss Re recently released a big report on climate change saying that insurance companies are underinsuring against rising climate risks that are rising now and projected to continue to do so over the near term. Despite the uncertainty, they see an imminent source of risk, and are not waiting until projections of the end of the century clear up to respond.

The first of these misdirection statements about Greenland is even more troubling because the rise in global mean sea level has accelerated. This is widely known despite claims to the contrary in Chapter 8 which is described in the introduction as a “levelheaded look at sea levels, which have been rising over the past many millennia.” Koonin continues: “We’ll untangle what we really know about human influences on the current rate of rise (about one foot per century) and explain why it’s very hard to believe that surging seas will drown the coasts any time soon.”

The trouble is that while seas have risen eight to nine inches since 1880, more than 30 percent of that increase has occurred during the last two decades: 30 percent of the historical record over the past 14 percent of the time series. This is why rising sea levels are expected with very high confidence to exaggerate coastal exposure and economic consequences [Section 19.6.2.1].

His teaser for Chapter 7 is an equally troubling misdirection. He promises to highlight “some points likely to surprise anyone who follows the news—for instance, that the global area burned by fires each year has declined by 25 percent since observations began in 1998.” Global statistics are meaningless in this context. Wildfires (if that is what he is talking about) are local events whose regional patterns of intensity and frequency fit well into risk-based calibrations because they are increasing in many locations. Take, for example, the 2020 experience. Record wildfires were seen across the western United States, Siberia, Indonesia and Australia (extending from 2019) to name a few major locations.

Take a more specific example. From August through October of 2020, California suffered through what became the largest wildfire in California history. It was accompanied by the third, fourth, fifth and sixth largest conflagrations in the state’s history; and all five of them were still burning on October 3. Their incredible intensity and coincidence can only be explained by the confluence of four climate change consequences that have been attributed to climate changes so far: record numbers of nighttime dry lightning strikes during a long and record-setting drought, a record-setting heat wave extending from July through August, a decade of bark-beetle infestation that killed 85 percent of the trees across enormous tracks of forests, and long-term warming that has extended the fire season by 75 days.

So, what is the takeaway message? Regardless of what Koonin has written in his new book, the science is clear, and the consensus is incredibly wide. Scientists are generating and reporting data with more and more specificity about climate impacts and surrounding uncertainties all the time. This is particularly true with regard to the exaggerated natural, social and economic risks associated with climate extremes—the low-probability, high-consequence events that are such a vital part of effective risk management. This is not an unsettled state of affairs. It is living inside a moving picture of what is happening portrayed with sharper clarity and more detail with every new peer-reviewed paper.

The author benefited from conversations with Henry Jacoby, Richard Richel and Benjamin Santer in preparing this essay.

This is an opinion and analysis article.

Gary Yohe is the Huffington Foundation Professor of Economics and Environmental Studies, Emeritus, at Wesleyan University in Connecticut. He served as convening lead author for multiple chapters and the Synthesis Report for the IPCC from 1990 through 2014 and was vice-chair of the Third U.S. National Climate Assessment.

More by Gary Yohe