American Geophysical Union, Journal of Geophysical Research, F2(114), 2009
DOI: 10.1029/2008jf001025
Full text: Download
Radar layer geometry in divide areas is strongly influenced by the operation of the Raymond effect, which causes upwarping of the layers as a consequence of the nonlinear rheology of ice. The detailed geometry of these layers is known to store a record of change in the cryosphere, of local thinning, and of the age of formation of the divide and has been surmised to provide information about lateral motion of divides. Such lateral motion can be caused by changes in flanking ice streams, and the divide area thereby contains a record of ice stream dynamics. It has also been suggested that a large perturbation of divide position will obliterate the cumulative effects of the operation of the Raymond mechanism, leading to the disappearance of Raymond bumps. Since the Raymond effect has a strong influence on the age‐depth relation in ice cores, knowledge of whether its operation is localized (leading to strong bump formation) or distributed is crucial in the interpretation of ice cores. The detailed evolution of ice divide radar layer geometry remains poorly understood. Employing a full thermomechanically coupled transient model, we qualitatively explore the effect of divide migration on radar layer geometry. Certain qualitative features emerge which can be used to infer history of cryosphere change, in particular, in areas distant from the usual sites of geological dating. There remains uncertainty about the influence of sliding on the operation of the Raymond effect. Under certain conditions, the existence of sliding can damp or eliminate the operation of the Raymond effect. If this is generally true, then dating of ice divides may simply be a date for the freezing of the divide bottom. We show that sliding does not necessarily eliminate the formation of bumps. Dates of divide formation are likely to be dates for the location of the ridge at a particular spot. Raymond bump evolution is weakened by flow along the ridge. We explore quantitatively the strength of this effect, using a scaling analysis to show that the weakening can efficiently be described by one parameter, the ratio of along‐ridge slope to a measure of the across‐divide curvature.