Investigating the feedbacks between CO2, vegetation and the AMOC in a coupled climate model
Investigating the feedbacks between CO2, vegetation and the AMOC in a coupled climate model
复制标题
DOI:
10.1007/s00382-019-04634-2
复制
发表时间:
2019-09-01
期刊:
影响因子:
4.6
通讯作者:
Singarayer, Joy
中科院分区:
文献类型:
--
作者:
Armstrong, Edward;Valdes, Paul;Singarayer, Joy
The Atlantic Meridional Overturning Circulation (AMOC) is an important component of the climate system, however its sensitivity to the terrestrial biosphere has been largely overlooked. Here the HadCM3 coupled climate model is run for millennial timescales to investigate the feedbacks between vegetation and the AMOC at increasing CO2. The impact of agricultural conversion (termed land-use change; LUC) and the role of the simulated 'background' vegetation (termed land cover change; LCC) are investigated. LUC cools climate in regions of high crop fraction due to increased albedo. LCC is shown to evolve at higher CO2, with a northward migration of the tree line in the Northern Hemisphere and dieback of the Amazon. This generally acts to enhance the impact of climate change primarily due to albedo changes. Density in the Greenland-Iceland-Norwegian (GIN) Seas is crucial in driving the AMOC. Increasing CO2 decreases regional sea surface density, reducing convection and weakening the AMOC. The inclusion of LCC is shown to be responsible for a significant proportion of this weakening; reflecting the amplification effect it has on climate change. This acts to decrease the surface density in the GIN Seas. At elevated CO2 (1400 ppm) the inclusion of dynamic vegetation is shown to drive a reduction in AMOC strength from 6 to 20%. Despite the cooling effect of LUC, the impact on the AMOC is shown to be small reflecting minimal impact it has on GIN Sea density. These results indicate the importance of including dynamic vegetation in future AMOC studies using HadCM3, but LUC may be insignificant. In the context of other climate models however, the importance of vegetation is likely to be overshadowed by other systemic model biases.