Elevated carbon dioxide emissions from human activities increase the uptake of carbon by plants but may decrease storage in soil.
An international team led by Lawrence Livermore National Laboratory (LLNL) scientists synthesized 108 elevated carbon dioxide (CO2) experiments in various ecosystems to find out how much carbon is absorbed by plants and soil. The research appears in the journal Nature.
The terrestrial biosphere takes up about 30 percent of CO2 emitted by human activities each year. Whether this land carbon sink can be sustained in the future partly depends on the extent to which elevated CO2 (eCO2) may increase carbon storage in plants and soils. While plant growth often increases in eCO2 experiments, individual studies have observed the amount of carbon in the soil can increase, remain unchanged or even decline.
“The variations in soil carbon accrual under elevated atmospheric CO2 remain poorly understood and this contributes to uncertainties in climate projections,” said César Terrer, a LLNL Lawrence Fellow and lead author of the paper. “We needed to take a look at the effect of elevated atmospheric carbon dioxide on plant biomass and carbon pools through a synthesis of real-world experiments and apply state-of-the-art statistical methods to find out what drives variation in soil carbon storage across the world.”
The team found that that when plant biomass is strongly stimulated by eCO2, soil carbon storage declines; conversely, where biomass is weakly stimulated, soil carbon accumulates.
“This unexpected trade-off appears related to plant nutrient acquisition strategies, because enhanced biomass requires mining the soil for nutrients, which decreases the amount of soil carbon buildup,” Terrer said.
This results in a negative relationship between above-ground plant biomass and soil carbon, in contrast to the classic paradigm in soil ecology and models, which suggests plant biomass and soil carbon increase in tandem.
The results indicate that doubling atmospheric CO2 causes a stronger increase in soil carbon stocks in grasslands (8 percent) than forests (about 2 percent), even though plant biomass in grasslands responded less strongly (9 percent) than in forests (23 percent).
“Ecosystem models don’t accurately reproduce this negative relationship between plant biomass and soil carbon accumulation, which implies that our projections of future soil carbon may need to be revised,” said LLNL scientist and co-author Jennifer Pett-Ridge.
Other institutions contributing to the work include Stanford University; Indiana University; Northern Arizona University; Universitat Autònoma de Barcelona, Spain; Oak Ridge National Laboratory; University of Exeter, UK; University of California, Berkeley; Lawrence Berkeley National Laboratory; ETH, Zurich; University of Minnesota; Western Sydney University, Australia; University of Cambridge; University of Oxford; Washington State University; Jet Propulsion Laboratory; University of California, Los Angeles; and University of Antwerp, Belgium.
The research was funded by LLNL’s Lawrence Fellow program and the Laboratory Directed Research and Development program.
stark8 [at] llnl.gov