The contribution of semi-arid ecosystems to interannual global carbon cycle variability

Annual carbon uptake by terrestrial ecosystems is on average equal to about 25% of emissions from anthropogenic fossil fuels and net land cover change. Large year-to-year variability in the terrestrial carbon sink influences the atmospheric CO₂ growth rate with the underlying mechanisms of variability poorly constrained and thus the evolution of future land carbon uptake unclear. The exceptionally large land carbon sink in the year 2011, almost 40% of anthropogenic emissions, provided an opportunity to investigate this year-to-year variability using a variety of carbon cycle observation techniques, including a terrestrial biogeochemical model, an atmospheric inversion, and remote sensing data. We found that the global land sink anomaly was driven mainly by semi-arid vegetation activity in the Southern Hemisphere, with almost 60 percent of carbon uptake attributed to Australian ecosystems, where prevalent La Niña conditions caused up to six consecutive seasons of increased precipitation. Since 1981, vegetation expansion over Australia was found to drive a four-fold increase in the sensitivity of continental net carbon uptake to precipitation. These combined results suggest that the higher-turnover rates of carbon pools in semi-arid biomes are an increasingly important driver of global carbon cycle inter-annual variability with implications for the paradigm that tropical rainforests drive carbon cycle variability at inter-annual timescales. More research in semi-arid regions is needed to identify mechanisms of carbon turnover at inter-annual scales and to determine the causes, and their possible interactions, in driving vegetation expansion over longer time scales.