水稻养活数十亿人口——但其在助长气候变化中的作用正在增大

Rice feeds more than half the world. From terraced paddies in Southeast Asia to irrigated fields in China and India, it underpins daily meals for billions of people.

But the same flooded soils that help rice thrive also create ideal conditions for microbes that release climate-warming gases.

In a new study, our team of environment and agriculture scientists found that greenhouse gas emissions from rice paddies have nearly doubled globally since the 1960s, averaging about 1.1 billion tons of carbon dioxide-equivalent emissions per year in the 2010s. That’s roughly equal to the annual emissions of 239 million cars.

This makes rice-growing the largest emissions source in agriculture outside of livestock, and rice demand is expected to keep rising.

Farmers have ways to reduce their rice crops’ emissions without lowering their yields. If every grower used the best currently available “climate-smart” options, we found that global rice emissions could be reduced by about 10% by midcentury. However, greater reductions are needed to slow climate change, which would require developing additional, more effective strategies.

Why rice emissions have increased

Rice emissions have risen for two reasons: the expansion of rice cultivation area and the intensification of management practices.

Just over half of the global increase is from the expansion of rice-growing areas. In Africa, for example, the rice-growing area has roughly doubled since the 1960s, helping drive a twofold rise in methane emissions in the region.

At the same time, rice farmers are using more fertilizers and organic amendments, such as straw and manure, planting more productive rice varieties and growing the plants closer together. The result is more rice but also more greenhouse gas emissions.

We found that one practice in particular – leaving rice stalks in the field after harvest and then plowing them into the soil to improve soil fertility – was responsible for about 18% of rice’s increase in overall net emissions since the 1960s. The reason: It increases the organic matter in the soil, which microbes then decompose, creating more methane emissions.

Rising global temperatures further accelerate microbial activity in the soils, meaning even more emissions.

Fertilizer is another major contributor to emissions. Use of synthetic nitrogen increased by about 76% after 2000, boosting nitrous oxide – another powerful greenhouse gas. It contributed about 9% of the increase in total global net emissions from human activities.

Irrigation practices also affect emissions. In the past, irrigated rice paddies were kept flooded throughout the growing season, resulting in constant greenhouse gas emissions produced by microbes that thrive in the wet environment. Over the past two decades, however, more farmers have used intermittent flooding – draining their fields periodically.

This change has lowered methane emissions compared with keeping the paddies continuously flooded. However, we found a slight increase in nitrogen oxide emissions as soils cycled between wet and dry, which induces microbes to transform nitrogen in organic matter into nitrogen oxide gases, particularly nitrous oxide.

Climate impact of rice production

Putting a full climate price tag on rice production is harder than measuring one greenhouse gas at a time.

Rice paddies emit methane and nitrous oxide from wet or flooded soils. They also remove carbon dioxide from the atmosphere as rice grows, and they lose carbon from their soils between crop seasons.

A credible global estimate requires consistently accounting for different gases and soil carbon changes, as well as the uncertainty involved in tracking data across space and time.

To do that, we combined three approaches:

An ecosystem computer model allowed us to simulate crop growth, water conditions and soil processes to estimate changes in methane, nitrous oxide and soil carbon together.

An artificial intelligence-powered machine learning model improved estimates where measurements were sparse to cover all rice regions in the world.

And a meta-analysis of more than 1,200 field experiment sites provided direct evidence of how practices such as irrigation, fertilizer use and management of crop residue affect emissions.

Together, they allowed us to quantify emissions from 1961 to 2020, determine what drove those emissions, and test the potential of mitigation techniques under future climate conditions.

What works and doesn’t for climate mitigation

There are ways to reduce emissions from rice production without sacrificing yield.

Our study found that reducing fertilizer use and residue applications, managing irrigation to allow dry periods in between flooded ones and reducing tillage could, together, reduce global greenhouse gas emissions from rice by about 10% by midcentury.

We were surprised to find that replacing chemical fertilizers with more organic choices is not always better from a greenhouse gas perspective, although it is valued in organic farming.

Maintaining moderate amounts of straw and other crop residue in the field can help boost soil fertility, but too much can increase methane emissions and accelerate the loss of carbon from the soil. Another option is to convert part of the residue into biochar – burning it under low-oxygen conditions before mixing it into flooded soils. Biochar can help stabilize soil carbon and reduce methane emissions.

Improving water management can be a powerful tool for reducing emissions. Periodically draining fields reduces methane production, though it may slightly raise nitrous oxide emissions. This strategy is particularly effective in regions with reliable irrigation infrastructure, including large parts of Asia.

Managing fertilizer use is also an effective mitigation strategy, particularly in highly fertilized systems, including parts of China and South Asia. Excess nitrogen increases nitrous oxide without a clear increase in crop yields and increases water pollution. Reducing overapplication of nitrogen reduces emissions and water pollution, and it saves farmers money in the process.

The effects of tilling, the practice of plowing the soil between crop seasons, have large regional differences. Reducing tilling is often promoted as climate-friendly, but we found that it does not always minimize net emissions in flooded systems. In rice fields in temperate zones, including much of the U.S. and China, cooler conditions can limit methane production, allowing the soil carbon benefits of reduced tilling to outweigh the methane risk. In warmer, persistently flooded systems, however, low-oxygen conditions can boost microbial activity, increasing methane production and accelerating soil carbon loss.

Overall, we found that no single practice works everywhere. Each region will need to assess the most effective practices for reducing emissions.

A climate ceiling for rice production

The bottom line is both hopeful and sobering: Targeted sets of optimized practices can deliver meaningful emission reductions without losing rice yields, but the total global possible reduction is modest.

To reduce emissions further will require better guidance to help farmers determine the best levels of organic amendments, such as straw or biochar, and new approaches that can reduce emissions without undermining rice production.

Hanqin Tian receives funding from US Department of Agriculture, US National Science Foundation, and Andrew Carnegie Fellowship Program.

Pep Canadell receives funding from the Australian National Environmental Science Program-Climate Systems Hub.

Shufen (Susan) Pan receives funding from U. S. National Science Foundation

Jingting Zhang does not work for, consult, own shares in or receive funding from any company or organization that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.