CO2 is simply not a danger to human health, and it’s actively good for crops

Donald Trump has revoked the official doctrine that carbon dioxide is a danger to human health. This removes the justification for the federal government treating it as a pollutant. An anti-scientific lurch that will kill people – or a welcome return to common sense? What does the science actually say?

Carbon dioxide can kill you. In August 1986 Lake Nyos in Cameroon suddenly released around 200,000 tonnes of the gas in a natural but rare phenomenon known as a limnic eruption. Being heavier than air, it settled in an invisible cloud over the surrounding land and asphyxiated 1,746 people and over 3,000 livestock. But that was concentrated CO2. To leap from that to saying that the average concentration of carbon dioxide in the air – currently just over 0.04 per cent, up from just under 0.03 per cent a century ago – is a danger to human health would be clearly idiotic. All sorts of things are poisonous when concentrated but harmless when extremely dilute.

The argument in favour of the “endangerment finding” made by Barack Obama’s Environmental Protection Agency (EPA) in 2009 was not that carbon dioxide is toxic but that it is causing and will continue to cause global warming. That in turn will cause things like sea level rise and more heatwaves and heavier rainfall and possibly other problems, which will in turn endanger human lives, if not now then at some point in the future. Ergo, carbon dioxide is itself a dangerous substance.

Even on those terms, it is a fairly flimsy argument. As the independent meteorologist Chris Martz explains, the Clean Air Act does not classify carbon dioxide as an “air pollutant.” Obama had a choice: he could have tried to persuade Congress to amend the Act or he could have tasked the EPA to “find” evidence that carbon dioxide is a danger to public health. He chose the latter course. Clearly this was no less political a decision than Donald Trump’s decision to reverse the finding.

Defenders of the endangerment finding argue that, whatever the history of the decision, there exists evidence that increasing carbon dioxide in the atmosphere could flood and drown people in the future or give them heat stroke. They are right. But they must also accept there exists evidence that such warming might bring benefits and save lives. You cannot look at one and ignore the other.

Take the simplest measure of all: how many people die in cold weather and how many die in hot weather. Worldwide, about nine times as many die of cold as heat. This is true even in fairly hot countries like Greece and India. Is this ratio likely to change much in the future? Possibly not, because global warming happens more in the north in winter and at night, less in summer, in the tropics or in the daytime. So deaths from heat may rise only slowly while deaths from cold may fall quite fast.

Suppose though that we do experience significant net harms from warming in 75 years’ time. More heatwaves, more flooding, perhaps worse storms (though so far there is no upward trend in the frequency or intensity of storms and a downward one in droughts).

Those future threats will affect richer people. On current trends – according to the economic models used by climate scientists – the average person will be at least four times as rich in 2100 as they are today, probably more in poor countries in Africa where incomes are growing more rapidly. Posterity will be better able to afford protection against the effects of climate change than we are: flood defences, air conditioning and so forth.

This is already happening. Deaths from droughts, floods and storms have dramatically declined to record low levels in recent decades as people have acquired better forecasts, better shelter and better transport. Hurricanes like Melissa last year just don’t kill that many people any more: she killed about 100. This will be even more true in the future. Helping rich people tomorrow by punishing poor people today is the very opposite of compassionate or rational.

But all these effects are dwarfed by another, far bigger trend that is almost entirely positive, not negative. This is the well-established phenomenon of global greening. Carbon dioxide is plant food. Trees, grasses, crops, algae, corals, seaweeds, phytoplankton all depend on the gas for growth. Increasing its concentration in the air from 0.03 per cent to 0.04 per cent has had a directly measurable effect on the growth rate of all plants.

This has shown up in higher crop yields, faster tree growth and a general greening of the planet in all ecosystems, measurable by satellites, from the Arctic tundra to the tropical rainforest. In the words of the author of one key paper on the subject, in 30 years the world has added the equivalent of “a green continent about two times the size of mainland USA”. And that was published 10 years ago so the area of extra greenery is even bigger today.

How much is this worth to human health, let alone wildlife? Ross McKitrick of the University of Guelph in Canada and his colleagues have done a series of detailed calculations based on thousands of different studies and concluded that the effect of global greening, especially on crops, makes carbon dioxide into a net benefit to humankind, far outstripping any negative effects from global warming. Moreover, this net benefit will continue till at least 2050, even if warming accelerates somewhat.

In 2017, there was an attempt to make this good news go away (climate scientists hate good news), in the form of a paper in Nature arguing that crop yields would soon fall, not rise, so the net benefits of carbon dioxide were a myth. This gave the Biden administration a renewed excuse to argue that carbon dioxide should be treated as a net danger to humanity.

McKitrick got hold of the underlying data and found a massive mistake so obvious as to suggest negligence if not worse. In no less than 360 of the case studies included in the paper, the effect of global greening on crop yield had been simply left out. When he put it in, the results were reversed: carbon dioxide went from doing net harm in the future to having a net benefit on the crop yields of wheat, maize, millet, sorghum, rice and soybean, right up to an implausible 4C of future warming and beyond.

McKitrick’s responding paper pointing this out was published last year. It supports the case that carbon dioxide cannot reasonably be called a net danger to human health, even on a climate change basis; it is more likely a net beneficiary. The defenders of the endangerment finding can argue that such a calculation is based on forecasts that might be wrong but that would be a foot-shooting expedition because the same is true of all their papers too.

In truth, the debate about the endangerment finding is a political one and always has been. Scientists chose to interpret the evidence in a way that supported a political preference. They cannot now complain when others do the same.