An international team of scientists has discovered a significant error in the climate models used to predict future warming. The models overestimate a natural process called nitrogen fixation by about 50% — and this has knock-on effects for how much CO₂ plants can absorb.
Here's how it works: microorganisms in soil convert atmospheric nitrogen into forms plants can use. Plants need this nitrogen to grow, and plant growth is what absorbs CO₂ from the atmosphere. This process — where rising CO₂ stimulates more plant growth, creating a natural carbon "sink" — is called the CO₂ fertilization effect.
Because the models assumed more nitrogen was available than actually is, they predicted more plant growth and therefore more CO₂ absorption. The researchers calculated that the fertilization effect is approximately 11% lower than models predicted.
This is complicated by agriculture: human farming activities have increased nitrogen fixation by 75% over 20 years through industrial fertilizers. This partly compensates for the natural overestimate, but fertilizer use creates its own problems — nitrogen runoff pollutes waterways, and nitrogen compounds like nitrous oxide are powerful greenhouse gases.
The study, published in the Proceedings of the National Academy of Sciences (PNAS), demonstrates how one wrong number in a model can cascade through the entire system. However, the researchers note that climate models contain hundreds of interacting variables. Correcting one error doesn't necessarily mean overall predictions are wrong — other errors might compensate. The key takeaway is that the natural world may absorb less of our emissions than we hoped.