Assessing the Decadal Predictability of Land and Ocean Carbon Uptake. Issue 5 (15th March 2018)

• Record Type: Journal Article
• Authors: Séférian, Roland; Berthet, Sarah; Chevallier, Matthieu
• Is Part Of: Geophysical research letters. Volume 45:Issue 5(2018)
• Page Start: 2455
• Page End: 2466
• Publication Date: 2018-03-15
• Subjects: decadal predictability -- carbon fluxes -- internal variability -- Earth system model; Geophysics -- Periodicals; Planets -- Periodicals; Lunar geology -- Periodicals; 550
• DOI: 10.1002/2017GL076092
• Languages: English
• ISSNs: 0094-8276

Abstract

Abstract: The decadal predictability of carbon fluxes has been examined over continents and oceans using a "perfect model" approach based on a 400 year preindustrial simulation and five 10‐member ensembles from the Centre National de Recherches Météorologiques‐Earth System Model version 1. From these experiments, we find that the global land uptake and ocean carbon uptake are potentially predictable by up to six years, with a median predictability horizon of four years. Predictability of global carbon uptake is prominently driven by the ocean's predictability. The difference in predictability between ocean and land carbon fluxes stems from the relative capability of ocean or land to generate low‐frequency fluctuations in carbon flux. Indeed, ocean carbon fluxes display low‐frequency variability that emerges from the year‐to‐year variability in the North Atlantic, the North Pacific, and the Southern Ocean. The Southern Ocean carbon uptake can be predicted up to six years in advance and explains most of the global carbon uptake predictability. Key Points: Global carbon fluxes can be predicted up to six years in advance Predictability of global carbon flux is driven by ocean, not by land The Southern Ocean supports most of the global carbon flux predictability