An Ecosystem on Steroids

In a report in Nature, Mark Friedl and Josh Gray write that one factor contributing to the widening seasonal swing in CO2 levels is increased production of four leading crops—corn, wheat, rice, and soybeans. Photo credit: Leo Papandreou/Flickr

Each year in the Northern Hemisphere, levels of atmospheric carbon dioxide (CO2) drop in the summer as plants “inhale” and temporarily store carbon, and then climb again as they “exhale” and release carbon as they decompose after their growing season. Over the past 50 years, the size of this seasonal swing has increased by as much as half, for reasons that aren’t fully understood. Now a team of researchers led by Boston University scientists has shown that changes in production of corn and three other major crops—wheat, rice, and soybeans—are the source of as much as one quarter of the increase in this seasonal carbon cycle.

“In the Northern Hemisphere, there is a strong seasonal cycle of vegetation,” says Mark Friedl, a College of Arts & Sciences (CAS) professor of earth & environment and senior author of a study about the research published in November 2014 in Nature. “Something is changing about this cycle; the ecosystems are becoming more productive, pulling in more atmospheric carbon during the summer and releasing more during the dormant period.”

Most of this annual change is attributed to the effects of higher temperatures driven by climate change—including longer growing seasons, greater uptake of carbon by vegetation, and the “greening” of higher latitudes with more vegetation. “But that’s not the whole story,” says Josh Gray, a CAS research assistant professor of earth & environment and lead author on the paper. “We’ve put humans and croplands into the story.”

The scientists gathered global production statistics for four leading crops—corn, wheat, rice, and soybeans—that represent about 64 percent of all calories consumed worldwide. They found that production of these crops in the Northern Hemisphere above the tropics has more than doubled since 1961, a gain that translates to about a billion metric tons of carbon captured and released each year.

These croplands are “ecosystems on steroids,” says Gray, noting that they occupy about six percent of the vegetated land area in the Northern Hemisphere, but are responsible for up to a quarter of the total increase in seasonal carbon exchange of atmospheric CO2, and possibly more.

The increase in crop production doesn’t have a significant impact on global terrestrial carbon uptake and release, since essentially all carbon in the harvested crops is released each year. However, understanding the effects of agricultural production, the researchers say, helps to improve understanding of how terrestrial ecosystems are tied to the global climate, especially in terms of how natural ecosystems might buffer rising levels of CO2 in the future.

Funded primarily though programs supported by the National Aeronautics and Space Administration (NASA) and the National Science Foundation (NSF), the inspiration for the study came during a dinner conversation in Boston last year among the BU scientists, who use remote sensing to study changes on the earth’s surface, and atmospheric chemist Eric Kort of the University of Michigan in Ann Arbor.

Kort was co-author of a 2013 paper by Heather Graven of Imperial College London and colleagues that demonstrated the overall shift in seasonal CO2 levels, but did not look in detail at agricultural production. At the dinner, Kort suggested that contributions from crops could help to explain the magnitude of the shift.

“We thought, ‘Somebody should do the math,’” recalls Gray. “I did some quick calculations on a napkin to see if crops could account for enough carbon to impact the seasonality appreciably. It seemed like they could, and the next several weeks were a very exciting flurry of missteps and recalculations as we honed in on a preliminary model/number.” Gray and Friedl worked on the initial model with Stephen Frolking, a biogeochemist at the University of New Hampshire at Durham.

“We needed additional expertise and data, so we built our final team and refined the model,” says Gray. “I spent the next several months trying to prove that we were wrong, but couldn’t, so we wrote it up and submitted it to Nature.”
In addition to Gray, Friedl, and Frolking, the final team also included Christopher Kucharik of the University of Wisconsin-Madison, Navin Ramankutty (then at McGill University and now at the University of British Columbia in Vancouver), and Deepak Ray of the University of Minnesota Institute on the Environment. This team used data on Land Cover and Phenology from the NASA Moderate Resolution Imaging Spectrometer (MODIS) Land Cover and Phenology along with many other measurements and statistical products in this study.

The work highlighted the extraordinary increases in crop production in recent decades. “It’s a remarkable story of what we’ve done in agriculture in general,” says Friedl. “And in particular corn, which is one crop that’s just exploded. Corn alone accounts for two-thirds of the crop contribution to the increased seasonal exchange in carbon, and nearly 90 percent of that is produced in the Midwestern United States and China.”

“Over the last 50 years, the area of croplands in the Northern Hemisphere has been relatively stable, but production has intensified enormously,” he adds. “The fact that such a small land area can actually affect the composition of the atmosphere is an amazing fingerprint of human activity on the planet.”