Carbon fluxes from a temperate rainforest site in southern South America reveal a very sensitive sink

• Main Authors: Jorge F. Perez‐Quezada, Juan L. Celis‐Diez, Carla E. Brito, Aurora Gaxiola, Mariela Nuñez‐Avila, Francisco I. Pugnaire, Juan J. Armesto
• Format: Article
• Language: English
• Published: Wiley 2018-04-01
• Series: Ecosphere
• Subjects: AMERIFLUX; Chile; Chiloé Island; eddy flux; evergreen; FLUXNET
• Online Access: https://doi.org/10.1002/ecs2.2193

Abstract Ecosystems where carbon fluxes are being monitored on a global scale are strongly biased toward temperate Northern Hemisphere latitudes. However, forest and moorland ecosystems in the Southern Hemisphere may contribute significantly to the global and regional C balance and are affected by different climate systems. Here, we present the first data from an old‐growth forest representative of temperate, broad‐leaved rainforests from southern South America. Carbon fluxes monitored over two years using the eddy covariance technique showed that this rainforest acts as an annual sink (−238 ± 31 g C/m2). However, there were significant pulses of carbon emission associated with dry episodes during the summer months (i.e., peak of the growing season) and periods of significant carbon fixation during the cold austral winter, indicating that the carbon balance in this forest is very sensitive to climate fluctuations. The carbon fixation surges in winter seem to be related to the mild temperatures recorded during this period of the year under the prevailing oceanic climate. Winter carbon gain was more relevant in determining the annual carbon balance than summer pulse emissions. Regarding the annual carbon balance, this southern forest resembles the patterns observed in montane tropical forests more than the behavior of narrow‐leaved evergreen temperate forests from the Northern Hemisphere. These patterns make this southern forest type relevant to understanding the mechanisms and thresholds that control ecosystem shifts from carbon sinks and sources and will provide key data to improve global dynamic vegetation models.