Investigation of the Impact of a Heinrich-Event-like Abrupt Event Superimposed Onto the RCP 8.5 Scenario
Abstract:The CMIP5/IPCC (AR5) projections of climate change showed the temperatures should increase by between 1.3 °C and 4.4 °C by 2100 and the sea level rise between 26 cm and 82 cm on average. Superimposed on these climatic and hydrologic trends associated with the RCP 8.5 most pessimistic scenario, it is important to investigate the possible effect of an abrupt event like an ice-sheet surge.
This is first justified by the recent collapse of the Larsen B ice shelf which illustrates that a non-linear response of the cryosphere may occur in a warming world. Furthermore, in glacial periods, ice-sheets have been unstable and huge surges of icebergs did occur and deeply modified the climate. The aim of this presentation is to show the hypothetical climatic consequences of such a Heinrich-type event on future climate change.
To achieve this goal, 3 scenarios of rapid ice-sheet deglaciation have been designed, corresponding to an additional sea level rise of around 3 m :
- S1 corresponds to a contribution from Greenland only;
- S2 from West-Antarctica only;
- S3 from both ice-sheets.
We use the global atmosphere-ocean general circulation model (OAGCM) IPSL-CM5A-LR. The freshwater perturbation is applied near the ice-sheet(s) during 20 years from 2050 to 2070 during an RCP8.5 perturbation. The RCP8.5 scenario is then continued until 2100 (without freshwater).
For these types of scenarios, previous experiments for paleoclimatic cases help us diagnose the most vulnerable areas. The North Atlantic and the collapse of thermohaline circulation is one key issue, but far field teleconnections with the Asian monsoon will also be explored. We will investigate the timing and amplitude of the climate impacts due to the perturbations, which may be quite different in our 3 scenarios.
These abrupt events consequences can be important for the populations and the consideration of rapid changes should improve the reliability of IPCC predictions.