PP44A-06
Transient earth system model simulations as age-scale generators for paleo proxy data?

Thursday, 17 December 2015: 17:15
2012 (Moscone West)
Axel Timmermann, IPRC, University of Hawaii at Manoa, Honolulu, HI, United States, Mona Stockhecke, University of Minnesota Duluth, Duluth, MN, United States, Tobias Friedrich, University of Hawaii at Manoa, Honolulu, HI, United States and Laurie Menviel, University of New South Wales, Climate Change Research Centre, Sydney, NSW, Australia
Abstract:
Generating age models for paleo proxy data can be extremely difficult. Oftentimes assumptions are made which are based on hypothetical relationships between climate and orbital forcings. Whether these relationships (expressed in terms of correlation models) are physically justified and whether they are stationary is testable using transient climate model simulations.

Several standard methods to generate age models for paleo-proxy data are scrutinized here, such as orbital tuning and synchronization to benthic stacks. To overcome some of the fundamental weaknesses of these methods we propose to use transient paleo climate model simulations to derive dynamically and physically consistent age models for paleo-proxy data. We illustrate this suggestion using

1) millennial-scale climate variations during MIS

2) orbital-scale climate variability during the past ~800 ka

Ad 1) A physically forced MIS3 global hindcast model simulation with an earth system model, designed to match the reconstructed North Atlantic SST variability, can be used to determine the relative timing of different climate and biogeochemical variables at various locations with respect to an initial absolute reference timescale (GICC05 in our case). Corresponding leads and lags are a result of the physical equations of the climate system – not of oversimplisitic statistical assumptions (such as wiggle matching). The key assumption for this approach is that global patterns of Dansgaard-Oeschger variability are caused by centennial to millennial-scale AMOC variability.

Ad 2) A transient earth system model simulation of the past ~800 ka is forced with observed greenhouse gas variations (on EDC3), orbital and estimated ice-sheet forcing. Simulated rainfall variations over the Eastern Mediterranean are compared with hydroclimate reconstructions from Lake Van. The simulated rainfall agrees well with the hydroclimate reconstruction (on the MoSto27 timescale) for the first 200 ka. Following this we demonstrate how the simulated rainfall variations from 201 to 600 ka can then be used to develop an independent age model for the Lake Van sediment core. Tie points from 40Ar/39Ar measurements and geomagnetics can then be used for verification of the simulation-based age model.

<< Previous Abstract | Next Abstract >>