Greenland ice sheet model parameters constrained using simulations of the Eemian Interglacial

• Main Authors: Robinson, A., Calov, R., Ganopolski, A.
• Format: Article in Journal/Newspaper
• Language: English
• Published: München : European Geopyhsical Union 2011
• Subjects: ddc:550; climate variation; Eemianglacial-interglacial cycle; GRIP; ice sheet; isotopic analysis; mass balance; paleoclimateproxy climate record; regional climate; Greenland; Ice Sheet
• Online Access: https://oa.tib.eu/renate/handle/123456789/622; https://doi.org/10.34657/988

Using a new approach to force an ice sheet model, we performed an ensemble of simulations of the Greenland Ice Sheet evolution during the last two glacial cycles, with emphasis on the Eemian Interglacial. This ensemble was generated by perturbing four key parameters in the coupled regional climate-ice sheet model and by introducing additional uncertainty in the prescribed "background" climate change. The sensitivity of the surface melt model to climate change was determined to be the dominant driver of ice sheet instability, as reflected by simulated ice sheet loss during the Eemian Interglacial period. To eliminate unrealistic parameter combinations, constraints from present-day and paleo information were applied. The constraints include (i) the diagnosed present-day surface mass balance partition between surface melting and ice discharge at the margin, (ii) the modeled present-day elevation at GRIP; and (iii) the modeled elevation reduction at GRIP during the Eemian. Using these three constraints, a total of 360 simulations with 90 different model realizations were filtered down to 46 simulations and 20 model realizations considered valid. The paleo constraint eliminated more sensitive melt parameter values, in agreement with the surface mass balance partition assumption. The constrained simulations resulted in a range of Eemian ice loss of 0.4–4.4 m sea level equivalent, with a more likely range of about 3.7–4.4 m sea level if the GRIP δ18O isotope record can be considered an accurate proxy for the precipitation-weighted annual mean temperatures.