HR: 17:50h
AN: GC44A-10    [Abstracts]
TI: Amazon collapse in the next century: exploring the sensitivity to climate and model formulation uncertainties
AU: * Booth, B
EM: ben.booth@metoffice.gov.uk
AF: Met Office Hadley Centre, FitzRoy Road, Exeter, EX1 3PB, United Kingdom
AU: Collins, M
EM: matthew.collins@metoffice.gov.uk
AF: Met Office Hadley Centre, FitzRoy Road, Exeter, EX1 3PB, United Kingdom
AU: Harris, G
EM: glen.harris@metoffice.gov.uk
AF: Met Office Hadley Centre, FitzRoy Road, Exeter, EX1 3PB, United Kingdom
AU: Chris, H
EM: chg@ceh.ac.uk
AF: CEH, Wallingford, Oxford, OX, United Kingdom
AU: Jones, C
EM: chris.jones@metoffice.gov.uk
AF: Met Office Hadley Centre, FitzRoy Road, Exeter, EX1 3PB, United Kingdom
AB: A number of recent studies have highlighted the risk of abrupt dieback of the Amazon Rain Forest as the result of climate changes over the next century. The recent 2005 Amazon drought brought wider acceptance of the idea that that climate drivers will play a significant role in future rain forest stability, yet that stability is still subject to considerable degree of uncertainty. We present a study which seeks to explore some of the underlying uncertainties both in the climate drivers of dieback and in the terrestrial land surface formulation used in GCMs. We adopt a perturbed physics approach which forms part of a wider project which is covered in an accompanying abstract submitted to the multi-model ensembles session. We first couple the same interactive land surface model to a number of different versions of the Hadley Centre atmosphere-ocean model that exhibit a wide range of different physical climate responses in the future. The rainforest extent is shown to collapse in all model cases but the timing of the collapse is dependent on the magnitude of the climate drivers. In the second part, we explore uncertainties in the terrestrial land surface model using the perturbed physics ensemble approach, perturbing uncertain parameters which have an important role in the vegetation and soil response. Contrasting the two approaches enables a greater understanding of the relative importance of climatic and land surface model uncertainties in Amazon dieback.
DE: 3337 Global climate models (1626, 4928)
DE: 4901 Abrupt/rapid climate change (1605)
SC: Global Environmental Change [GC]
MN: 2007 Fall Meeting