Two Aspects of Decadal ENSO Variability Modulating the Long‐Term Global Carbon Cycle

The El Niño–Southern Oscillation (ENSO) drives variations in terrestrial carbon fluxes by affecting the terrestrial ecosystem via atmospheric teleconnections and thus plays an important role in interannual variability of the global carbon cycle. However, we lack such knowledge on decadal time scales, that is, how the carbon cycle can be affected by decadal variations of ENSO characteristics. Here we examine how, and by how much, decadal ENSO variability affects decadal variability of the global carbon cycle by analyzing a 1,801‐year preindustrial control simulation. We identify two different aspects, together explaining ~36% of the decadal variations in the global carbon cycle. First, climate variations induced by decadal ENSO‐like variability regulate terrestrial carbon flux and hence atmospheric CO2 on decadal time scales. Second, decadal changes in the asymmetrical response of the terrestrial ecosystem, resulting from decadal modulation of ENSO amplitude and asymmetry, rectify the background mean state, thereby generating decadal variability of land carbon fluxes. Plain Language Summary The El Niño–Southern Oscillation (ENSO) is an important driver of year‐to‐year variation of the global carbon cycle due to its impacts on the global climate variability. For example, most parts of the tropical land experience drought during El Niño events, and therefore rainforests and savanna regions do not capture well carbon dioxide compared to normal years because a high temperature and a lack of precipitation during El Niño events lead to less photosynthesis over the tropics. This is a well‐known feature in year‐to‐year variation, but not in decadal time scales due to a lack of long‐term observations. Here we examine how, and by how much, decadal ENSO variability affects decadal variation in the global carbon cycle by analyzing a 1,801‐year Earth System simulation. We found that two different aspects of decadal ENSO variability, associated with decadal changes in the tropical Pacific Ocean and asymmetric characteristics between El Niño and La Niña, drive decadal change in the terrestrial carbon fluxes. As a result, these two aspects together can explain ~36% of the decadal variability in the global carbon cycle. Key Points About 36% of decadal variations in global NBP can be explained by two aspects of decadal ENSO variability First, decadal ENSO‐like variability induces interdecadal changes in terrestrial carbon fluxes via atmospheric teleconnections Second, decada