The vulnerability of the Amazon forest to drier climate: results from an individual-based ecosystem model (ED-2.2)

Friday, 19 December 2014: 11:05 AM
Marcos Longo1,2, Ryan G Knox3, Naomi Marcil Levine4, Luciana F Alves5, Damien Bonal6, Matthew Hayek7, Scott R Saleska8, Kenia T Wiedemann7, Paul R Moorcroft9 and Steven C Wofsy7, (1)EMBRAPA Brazilian Agricultural Research Corportation, Campinas, Brazil, (2)Harvard University, Earth and Planetary Sciences, Cambridge, MA, United States, (3)Lawrence Berkeley National Laboratory, Berkeley, CA, United States, (4)University of Southern California, Los Angeles, CA, United States, (5)IAPAR Agronomic Institute of Brazil, Campinas, Brazil, (6)INRA Institut National de la Recherche Agronomique, UMR EEF 1137, Paris Cedex 07, France, (7)Harvard University, Cambridge, MA, United States, (8)University of Arizona, Tucson, AZ, United States, (9)Harvard Univ, Cambridge, MA, United States
The Amazon forest is a major component of the Earth System, however its structure, dynamics and functioning could be severely disrupted if droughts became more frequent. The vulnerability of different Amazonian areas to droughts was evaluated using the Ecosystem Demography model (ED-2.2). We simulated forest dynamics under a suite of rainfall regime scenarios for two sites in the Eastern Amazon with strong rainfall seasonality but very different total rainfall. While the model simulations predicted nearly no response in mortality or primary productivity at Paracou (wettest site), the model suggests that even moderate shifts of the mean rainfall could cause significant biomass loss at Tapajos (driest site), particularly amongst the largest trees, pioneers, and evergreens. The simulations that produced the largest losses were those in which the return period of extreme droughts became shorter than one decade hence not allowing
sufficient time for recovery. By defining vulnerability as the change in rainfall regime needed to reduce the return period to dangerous levels, we identified the most vulnerable areas to be near the forest Southern and Eastern limits, but also over large areas in Eastern Amazon.