Snow damage strongly reduces the strength of the carbon sink in a primary subtropical evergreen broadleaved forest

A primary subtropical evergreen broadleaved forest in southwest China experienced a particularly extreme snowfall event during January 2015. The 2015 event enabled the quantification of the impact of the extreme meteorological event on the forest carbon balance. We analyzed five years of continuous measurements of CO2 exchange across the biosphere/atmosphere interface in the forest using an eddy covariance technique. We quantified how exposure to an extreme meteorological event affected ecosystem processes that determine gross primary productivity (GPP) and ecosystem respiration (Reco), and thus annual net carbon (C) sequestration. The forest canopy was severely damaged by the heavy snow, and the leaf area index (LAI) decreased significantly from January to July 2015. GPP, net ecosystem exchange (NEE), and Reco all sharply decreased in 2015 after the heavy snow. On average, a strong decrease of 544 g C m−2 year−1 in annual NEE in 2015 was associated with a decrease of 829 g C m−2 year−1 in annual GPP and a decrease of 285 g C m−2 year−1 in annual Reco. Overall, annual net C uptake in 2015 was reduced by 76% compared to the mean C uptake of the previous four years. A sharp increase in carbon uptake was also observed in 2016, indicating that long-term, continuous measurements should be carried out to evaluate the overall response to the disturbance.