Not so fast, says Nathan Myhrvold, one of America’s most versatile applied scientists and inventors. “While I hope to see affordable fusion reactors in my lifetime,” he says, “I’m not holding my breath.” (He’s 63.) The machine at the National Ignition Facility was designed to make “at most a few bright flashes of raw energy a day. That’s it. Not to capture the energy, not to turn it into electricity.” He says it’s “sort of like those spectacular gasoline explosions you see in the movies when they blow up a car.” It’s a “long, long road” from there to an internal-combustion engine that “converts thousands of controlled gasoline explosions into useful energy every second.”

Make no mistake—Mr. Myhrvold is concerned about climate change. But he’s a scientific realist who thinks the shibboleths on the subject—embodied in such documents as the Paris Agreement of 2016—are misbegotten. Mankind isn’t capable of reducing emissions enough to keep temperatures from rising unacceptably.

He laments that policy makers largely scorn geoengineering—human interventions in the Earth’s natural systems to thwart or neutralize climate change. Such interventions could make a difference by following two broad approaches: solar-radiation management, “which seeks to reflect sunlight back into space,” and direct air capture, which means “sucking carbon-dioxide from the sky.” While these methods may “sound crazy,” he says, “they could work.” But research of this kind is actively discouraged.

The case for geoengineering begins with the argument that the alternative is insufficient. Data from rigorous long-term CO2 measurements around the world show that “despite all the coal plants shut down, all the electric vehicles sold, all the solar and wind power deployed, all the people now working from home rather than commuting,” the concentration of CO2 continues to rise “just about as fast as it has for the past 40 years, and faster than it did in the 1960s and ’70s.” He underscores the point with a rhetorical question: “Do you see the effect on emissions of the Great Recession of 2008 and 2009, or of the global pandemic shutdown? Nor do I.”

“If the projections are true, and I think they are, I think we’re already screwed,” he says. If we stop “all human-done” CO2 emissions today, climate models suggest the Earth’s climate would continue to get warmer. “It would keep going up for almost 100 years. It would take 145 years before it eventually went down to the same temperature it is today. That’s just the physics of the problem.” 

Geoengineering is about “deliberately trying to reduce climate change.” Excess CO2 traps a little less than 1% of heat from the sun, “so if we could make the sun 1% dimmer, we could shut off climate change.” When Mount Pinatubo, a volcano in the Philippines, erupted in 1991, it lowered world-wide temperatures by 1 degree Celsius for about 18 months. Human-emitted particulate pollution has historically offset about 20% of human-emitted CO2. “Ironically,” he says, “the Clean Air Act made our air better but hurt climate change.”

The simplest solar-radiation management scheme, Mr. Myhrvold says, “is to emit particles in the stratosphere to mimic Mount Pinatubo. We invented a particularly elegant way to do this with balloons and a pipe to the sky.” By “we,” he means Intellectual Ventures, the company Mr. Myhrvold founded in 2000 after leaving Microsoft, where he spent 13 years and rose to the position of chief technology officer. Intellectual Ventures “creates, incubates and commercializes” new inventions.

“Marine cloud brightening” is another solar-related intervention. “The idea is to increase the number and size of low clouds that form over the oceans so that more incoming sunlight bounces back into space instead of heating the ocean.” Scientists have proposed a variety of ways to do this. One, which Mr. Myhrvold’s company has explored, is to outfit ships with equipment to spray seawater into the air as they traverse the ocean. “The salt particles can serve as nuclei for water vapor to condense into droplets, thus forming clouds.”

Intellectual Ventures has also researched an idea for draining energy from hurricanes—“not a climate intervention per se, but a way to deal with the more intense cyclonic storms that some models project are likely to occur in a warmer world.” Low-cost floating tubes would harness wave action to increase the mixing of the ocean’s warm top layer—the source of heat that powers hurricanes—with cooler deep water. “Our simulations found that if enough of these were deployed in the path of the hurricane,” he says, “it could knock a storm down by a category or two.” But it would take thousands of tubes, so would be a challenge to deploy. “The real stumper was not technological but practical. Who would pay for the system and manage it?”

Direct air capture, the other major form of geo-intervention, may prove harder to do than solar-energy capture. Mr. Myhrvold says there are “three big issues” with extracting CO2 from the sky. The first is that CO2 is only 400 parts per million, or 0.04%, of the air. “So you need to work hard to get it—i.e., move a lot of air.” The second: CO2 dissolves in seawater. About half of human-emitted CO2 is in the ocean, and it is in equilibrium with the atmosphere. “If we withdraw CO2 from the air, the system re-equilibrates. So if we pull two units out of the air, we remove one unit net, because one unit comes out of the ocean.” The third issue is where to put the CO2: “It is very hard to store that much gas. If you reduce it to solid carbon, it would be immense.”

Mr. Myhrvold has met with several “very enthusiastic, very brilliant” entrepreneurs who have technologies either to remove carbon dioxide directly from the atmosphere or to make Earth a lot less hot. He recently spent time with a startup that has “a clever idea” for direct air capture. “At one point in the meeting I asked a question that I ask everyone in geoengineering: ‘You’re acting as if the world wants to solve the problem of climate change. Why is that?’ ” He explains his cynicism: “There are activists who oppose funding or experiments. There’s no evidence I can see that many of the people involved in the climate debate want a solution.”

The enthusiastic reaction to the experiment at Livermore Lab reinforces that point. Compared with that fusion reaction, Mr. Myhrvold sees far more promise in conventional nuclear power, or fission. “But there is a crazy notion of only liking some flavors of carbon-free power. Renewables are OK, but nuclear is not.”

Mr. Myhrvold’s scientific interests range beyond geoengineering to include computing, paleontology, astronomy and “cutting-edge” cooking. He is a co-author of “Modernist Cuisine: The Art and Science of Cooking” (2011), a five-volume cookbook that explains the science of food preparation and the latest kitchen technologies.

Prodigies, it is said, retain a childlike character into older age, and Mr. Myhrvold is no different. He went to college at 14, graduating from UCLA four years later with both a bachelor’s degree in math and a master’s in geophysics and space physics. Words tumble out at a boyish pace when he speaks, especially when excited. He’s easily distracted, as evident when a staffer brings in a slice of corn bread baked on the premises in a vast kitchen used for food experiments. “This is made with 100% maltitol,” he says as he cuts himself a bite. Maltitol is a sugar alcohol—sweet but edible by diabetics.

Mr. Myhrvold has four degrees—after UCLA, he went on to earn a master’s in mathematical economics and a doctorate in applied math from Princeton. He plays these down, joking that his friends used to say he was “trying to have more degrees than a thermometer.” But he lights up when he says he has a Grand Diplôme from the École de Cuisine La Varenne, a culinary school. He’s particularly proud of his 1979 fellowship from the Hertz Foundation, which gives out a handful of cash awards annually to promising young applied scientists. In December the foundation endowed a new fellowship in Mr. Myhrvold’s honor.

Geoengineering would appear to be the application of science par excellence. But along with the activists who “don’t want a technical solution to climate change,” he says, “there’s a second set of people who may not have that ideology, but have a more realpolitik sort of view.” This group—which comprises most Western governments—want “people to shut up” about interventions. He likens this approach to “The Little Engine That Could,” a children’s story about a small blue engine that pulled an entire train over a hill, inch by inch, through the sheer force of its will. “Opponents worry that once you have geoengineering, people won’t make sacrifices to cut emissions. They want a sword of Damocles hanging over humanity as a means to force us to follow their ideology.”

Mr. Myhrvold uses an analogy he describes as “horrible in some ways.” When the AIDS epidemic hit, some people saw it as punishment from God. “Their attitude was, ‘This is what you get if you indulge in the practices which God doesn’t approve of.’ ” In climate change, he says, this moralistic attitude takes the following form: “I don’t like aspects of our society, I don’t like technology, I don’t like capitalism, and this is nature’s retribution. And so we have to change the way we live.” Such beliefs “have become a very powerful disincentive, particularly for academic researchers.”

Things may have to get worse before they can get better. He believes the opposition will have to cease as the world experiences “very severe climate consequences” and demand rises for “an intervention to fix the problem.” He calls it “insane” not to fund such research and invokes another analogy: heart surgery and stents.

“You could imagine a world in which cardiology doesn’t exist because the medical profession said, ‘You fat bastards. You did it to yourselves. We’re not going to help you.’ ”

Mr. Varadarajan, a Journal contributor, is a fellow at the American Enterprise Institute and at New York University Law School’s Classical Liberal Institute.