Solar panels adorn the roof of my Colorado home. I helped build two large solar farms. I felt sure I was helping mitigate global warming, and it felt good! And as a happy additional benefit, Colorado’s net-metering meant my electric bill for the sunny summer months was $0. A win for the climate and a win for me. Or so I thought.

My journey began when, as a concerned scientist, I decided to understand what it would take to clean up America’s fossil-fueled electric power systems. Wind and sunshine are free and plentiful. It seemed to make perfect sense to put them to work cleaning up dirty and polluting power plants. After installing my solar panels several decades ago, I really got into it. I studied the science, met with utility and power executives, talked with leading academics, and pored over the data from already operating clean energy systems (solar, wind, and nuclear power). I also studied promising new technologies: Allam cycle gas, gas turbines with carbon capture, fusion, and hot rock geothermal. But as I got further into the science, the results became harder to hear; they began to deeply challenge many of my cherished beliefs.

The math for solar panels explains why. Solar panels turn sunlight into electricity, and they are—by far—the cheapest way to generate electricity. Replacing dirty coal with clean solar sounds terrific. But the problem is that word “replace.” A coal plant generates electricity 24 hours a day, seven days a week, rain or shine, all year long. Solar panels, on the other hand, suffer through nights, clouds, dust, snow, and weak winter sun, all of which impact a solar panel’s ability to generate electricity. In fact, and on an annual basis, the average North American solar farm generates meaningful power less than 20 percent of the time.

I’d never really thought about that. Obviously, power for 20 percent of the time can’t replace 24-hour dispatchable (whenever needed) baseload power. To cover those huge night and dark-day gaps, and to keep the lights always on, something else is needed—a lot of something else. In Colorado and elsewhere, most of that something else, most of those gaps, are filled with power from coal or gas plants. And that is the problem. Although solar farms have allowed fossil-fueled plants to cut back and save emissions on sunny days, they haven’t been able to replace them 24/7 and obviously cannot do so by themselves.

The result is that to fill the power gap, utilities employing solar and wind farms still need to keep their fossil-fueled generators running to provide power when the sun isn’t shining or the wind not blowing.

Many of us have hoped that batteries could someday fill the gaps by storing and saving energy that could be distributed when needed to fill the gaps. Unfortunately, there is no known technology that can efficiently and economically store the incredible amount of energy required for the length of time required. In winter, for example, clouds can cover most of the continent for a week or more, shutting down solar production (the output from solar farms drops by 75 to 90 percent when clouds cover the sun). Batteries can store enough power for a few hours, but storing enough electricity in batteries to power the nation for a cloudy week is simply not feasible. And as we electrify everything, the demand and the gap will grow exponentially.

Hydrogen? That doesn’t really work either. Unlike natural gas, great quantities of free hydrogen have not been found in nature. Clean, green hydrogen must be manufactured by electrolyzing water—an energy-intensive process. The gas then must be compressed, cooled, and sent to storage. To get the energy back out, it must be sent to a power plant where it will be burned in gas turbines driving electric generators. The process is hugely inefficient. Two thirds of the energy used to make the hydrogen is lost. We can’t afford to triple the number of solar farms to make up for hydrogen’s inefficiency.

But suppose if by some magic there were enough efficient storage to power the entire nation for a cloudy week. When the sun came back out, we’d have to recharge all that storage—put as much power back in as we took out. That means that for the next sunny week, all our solar power would be recharging storage—leaving nothing available to power the economy. In order to also do that, therefore, we’d need twice as many solar farms; one to run the economy, another the same size to recharge storage.

And then what happens if the clouds come back before recharging is complete? In short, and it’s something rarely considered, filling the gap with storage would require more than doubling the number of solar farms, doubling the cost, land use, and resources. Frankly, to meet our 2050 net zero goals, that much storage and renewable overbuild is simply not a realistic option. We need something else.

In short, and as much as I wanted it to be otherwise, when it comes to replacing reliable, weather-independent, 24/7 coal or gas plants, the intermittency of solar and wind farms simply doesn’t fill the need. Some early adopters of intermittent renewable energy like California and Germany are seeing the problem already. It comes in the form of grid instability and blackouts. As a result, California is cutting new solar way back, keeping gas plants running, and postponing the planned shutdown of the Diablo Canyon nuclear plant. California businesses and wealthy homeowners are buying diesel generators.

Strangely, the media hasn’t yet gotten the message. Daily media messaging tells us that renewables are the solution. It reinforces our hopes and distracts us from doing what’s really needed. Indeed, America is incurring many hundreds of billions of dollars of public debt for clean energy. And there is great pressure to use most of that money almost exclusively to build solar and wind farms, farms that, by themselves, won’t allow us to close fossil plants.

But please don’t get me wrong. Solar and wind, when properly integrated, can, and will, make a significant contribution to carbon reduction. But the backbone of any robust system must be reliable 24/7 (baseload) power. And the problem is much bigger than just replacing existing fossil plants. America’s reindustrialization, along with the electrification of transportation, heating, and industry, will double or triple electrical demand before 2050. It will take massive new energy generation to meet that challenge. Our nation still needs to come up with a plan to meet that projected increase in baseload demand with clean 24/7 energy.

But back to climate. Records show that 2023 was the hottest year in the last 125,000 years. If our nation fails to build workable solutions that can truly replace fossil fuels, emissions will keep rising, the planet will get hotter, life will get harder, and our grandchildren will rightly blame us.

If we really want to mitigate climate change, it’s time to put to work technologies whose physics and math say they actually can supply the massive baseload power to replace those fossil plants and meet the new demand. Fortunately, solutions exist.

The Solution

As hard as it may be to hear (and it took me a long time to accept this), nuclear energy is the only solution capable of replacing fossil fuels at scale. It has already been done. After the 1973 oil crisis, France began replacing almost all its fossil plants with new nuclear plants. The French finished the job in less than 15 years. Since then, those plants have been providing more than 70 percent of France’s electricity (most of the rest is hydropower). Today, excluding hydropower, France has the most reliable, cleanest, and cheapest electricity in Europe. It’s also the safest—no notable accidents. To meet the expected increase in demand, France has just decided to build at least seven more nuclear plants.
Because it is such an obvious solution that works, more countries are beginning to follow France’s example. Allow me to get a little wonky. There are about 440 nuclear power plants operating in 32 countries today. There are another 60 reactors under construction across the world, and still another 110 planned. Some countries are expanding their existing nuclear fleet—like China, which is bringing online dozens of new reactors (China is bringing more reactors online than any other country). Others, like Poland, are new to nuclear but moving forward aggressively, trying to free themselves from reliance on Russian gas. Now aware that nuclear is essential, more than 30 newcomer countries are considering, planning, or starting nuclear power programs. Twenty-two nations at COP28 in Dubai last December signed a pledge committing to tripling nuclear power. Russia’s response is to try to dominate nuclear energy. By offering finance, Russia has become the biggest exporter of nuclear reactors around the world, with an order book estimated to be worth about $200 billion.

Where is the United States in all this? Going the other way. Almost five decades ago public opposition driven by fear and oil company pressure effectively halted the planned conversion to nuclear of most of our coal and oil-fueled generators. We still get 20 percent of our electricity (and most of our non-hydropower clean electricity) from the 94 nuclear reactors that are still running. Unfortunately, many of them will soon be due for retirement. Replacements? In the last 30 years we’ve built only two new reactors, and no more are planned.

It’s worth looking at why and how this happened. Following WWII, I believe well-intentioned scientists inflated the danger of radiation in order to scare the world away from using—or even testing—atomic weapons. And as so often happens, there have been unintended consequences. Reacting to those public fears, the U.S. government set radiation limits for nuclear power plants far stricter than necessary for safety. Limits were set more than 100 times stricter than anything that has ever been shown to cause human harm, 50 times stricter than a full CT scan, even more than 20 times stricter than the high but harmless natural background radiation (radiation from the Earth) in some parts of the world. If we viewed highway risks the same way, we’d mandate a maximum speed of 1 mph It would unquestionably save lives – but at a huge cost to society. We’ve effectively restricted nuclear to 1 mph when the data says we could go 15 times faster (15 mph) with almost no increase in risk.

To get a better balance, perhaps it would help if we used the safety data from actual experience. In some parts of the world, populations have lived for millennia with background radiation as much as 100 times greater than America’s strict standard, with no elevated cancer or health problems. Navy sailors live and sleep for months next to naval reactors. Then there is the data from the world’s hundreds of power reactors that are not under America’s strict safety interpretations. The safety of those other reactors is every bit as good as the American reactors.

In fact, and despite the myths, the safety record of nuclear is unmatched by any other energy source. In the last 70 years, for example, radiation from nuclear accidents has caused fewer than 200 deaths (all from Chernobyl). By contrast, the UN claims that fossil fuel pollution causes some 8 million deaths each year! Even the manufacture and installation of solar panels causes more deaths per unit of delivered power. In practice, and over its lifetime, nuclear power is safer, more reliable, cheaper, and less environmentally damaging than any other electric power system. And it is also the cleanest, generating no CO₂ and producing less toxic waste than even solar. I would argue that by being overstrict with nuclear, we forced ourselves to use more oil, coal, and gas, which created much greater harm to our health, society, and economy.

So while my solar panels can certainly make a contribution, they can’t replace that dirty CO₂-producing coal plant that provides my electricity when the sun isn’t shining. The only thing that could do that before 2050 is a new nuclear power plant. If the United States is to meet the anticipated doubling of electricity demand and its climate goals, the American people must demand that their leaders include a significant buildout of new nuclear power.

The rest of the world is already doing so. The Koreans build large U.S.-designed nuclear plants in under five years and for less than the cost of a coal plant. American innovators are developing a new generation of smaller, cheaper, and even safer, small modular reactors (SMRs) that could see commercial operation this decade. On the other hand, if America fails to launch an accelerated nuclear buildout, we’ll not only keep damaging the climate, we’ll also be handing our adversaries an unopposed opportunity to move an energy-hungry world away from us and into their orbit.