The Earth’s terrestrial ecosystems – its lands, forests, jungles and so on – are crucial “sinks” for atmospheric carbon, holding nearly 30 percent of our annual CO2 emissions as they breathe in the carbon-rich gas and convert it into biomass growth. However, widespread land-use change and deforestation has introduced severe uncertainty when estimating future carbon storage in terrestrial ecosystems. Now, supercomputers at the San Diego Supercomputing Center (SDSC) have stepped in to help close that analytic gap.
Benjamin Sleeter, lead author of the study, is a research geographer with the US Geological Survey (USGS). “Since 1850, land-use change has added nearly half as much carbon to the atmosphere as fossil fuel emissions and has exerted a dominant influence on the storage of carbon in terrestrial ecosystems,” Sleeter explained. So the study set about a task: assessing the effects of land-use change in California’s forests, shrublands, grasslands and soils in terms of their CO2 sequestration.
The researchers used the land use and carbon scenario simulator (LUCAS) to project changes in those Californian ecosystems’ carbon balance under a range of scenarios through the year 2100. To accomplish that, they turned to SDSC’s Comet supercomputer, an Intel Haswell-based system with 1,944 standard compute nodes and 72 Nvidia-based GPU nodes (divided between K80s and P100s), delivering 2.76 peak petaflops.
The team developed 32 scenarios to test, eventually deciding to analyze in-depth the four that covered the most ground. “Access to Comet allowed us to perform a rigorous uncertainty assessment by running four of the 32 scenarios many times,” Sleeter said. After testing those four scenarios on Comet, the researchers obtained fairly uniform results.
“We discovered that no matter how we ran the calculations, carbon dioxide will decline by nearly 10 percent within the next 80 years,” Sleeter said, “unless we work with policy-makers to promote both reduced land development and global climate action.”
Of course, Sleeter qualified the results to a degree. “Our framework didn’t include variability in key parameters, such as changes in vegetation type, which may result from the coupled effects of climate change and high-severity fire,” he explained. Furthermore, he suggested that if all 32 scenarios had been tested, at least a few of them would have varied more significantly.
Still, Sleeter is hopeful that this research lays groundwork for future examination of this crucial climate question. “Our research,” he said, “presented the foundation to help assess how and where to implement changes to the way we manage our natural landscapes in our state and beyond.”
About the research
The research discussed in this article was published as “Effects of 21st-century climate, land use, and disturbances on ecosystem carbon balance in California” in volume 25, issue 10 of Global Change Biology. It was written by Benjamin M. Sleeter, David C. Marvin, D. Richard Cameron, Paul C. Selmants, A. LeRoy Westerling, Jason Kreitler, Colin J. Daniel, Jinxun Liu and Tamara S. Wilson. The full research article can be accessed here.