Dating ice flow change near the flow divide at Roosevelt Island, Antarctica, by using a thermomechanical model to predict radar stratigraphy

• Published in: Journal of Geophysical Research
• Main Authors: Martin, Carlos, Hindmarsh, Richard C.A., Navarro, Francisco J.
• Format: Article in Journal/Newspaper
• Language: unknown
• Published: American Geophysical Union 2006
• Subjects: Glaciology; Antarctic; West Antarctic Ice Sheet; Ross Ice Shelf; Conway; Roosevelt Island; Antarc*; Antarctica; Ice Sheet; Ice Shelf
• Online Access: http://nora.nerc.ac.uk/id/eprint/81/

Radar-detected internal layering contained in some ice divides shows upwarped arches termed Raymond bumps. The distribution of their amplitude with height can date the onset of divide flow, reflecting changes in the basin structure of the ice sheet. The distribution depends on rheology, surface geometry, accumulation rate, and temperature. Conway et al. (1999) used an isothermal parameterized ice flow model to estimate a date of 3200 years B.P., with no error estimate, for the onset of divide flow in Roosevelt Island, Ross Ice Shelf, which they associated with grounding line retreat of the West Antarctic Ice Sheet. No other retreat dating exists for an area of the Ross Ice Shelf distant from geological exposures. Employing a full thermomechanically coupled transient model, we use a direct search to determine which ice rheology best fits the observed bump distribution and surface profile, estimating the sensitivity of the dating to model parameters and exploring possible reasons for the observed asymmetry of the surface profile. Our main results are as follows: (1) A best estimate of the date of grounding line retreat near Roosevelt Island is 3000 years B.P., bounded by 2300 and 4200 years B.P. (2) Standard rheology (power n = 3) can only match the observed bump distribution for unrealistic isothermal models. (3) For thermomechanically coupled models a high power (n = 4) in Glen's law is required. (4) Low-power rheologies and wind scouring cannot produce correct bump amplitude distributions. (5) Asymmetric accumulation cannot explain the observed asymmetry of the surface profile. (6) Modeling isochrones in flanking regions also indicates changes in flow around 3000 years B.P.