The role of ice stream dynamics in deglaciation

• Published in: Journal of Geophysical Research: Earth Surface
• Main Authors: Robel, Alexander A., Tziperman, Eli
• Format: Article in Journal/Newspaper
• Language: unknown
• Published: American Geophysical Union 2016
• Subjects: ice streams; deglaciation; ice sheet dynamics; Laurentide; glacial cycles; Ice Sheet
• Online Access: https://doi.org/10.1002/2016JF003937

Since the mid-Pleistocene transition, deglaciation has occurred only after ice sheets have grown large while experiencing several precession and obliquity cycles, indicating that large ice sheets are more sensitive to Milankovitch forcing than small ice sheets are. Observations and model simulations suggest that the development of ice streams in the Laurentide Ice Sheet played an as yet unknown role in deglaciations. In this study, we propose a mechanism by which ice streams may enhance deglaciation and render large ice sheets more sensitive to Milankovitch forcing. We use an idealized configuration of the Parallel Ice Sheet Model that permits the formation of ice streams. When the ice sheet is large and ice streams are sufficiently developed, an upward shift in equilibrium line altitude, commensurate with Milankovitch forcing, results in rapid deglaciation, while the same shift applied to an ice sheet without fully formed ice streams results in continued ice sheet growth or slower deglaciation. Rapid deglaciation in ice sheets with significant streaming behavior is caused by ice stream acceleration and the attendant enhancement of calving and surface melting at low elevations. Ice stream acceleration is ultimately the result of steepening of the ice surface and increased driving stresses in ice stream onset zones, which come about due to the dependence of surface mass balance on elevation. These ice sheet simulations match the broad features of geomorphological observations and add ice stream dynamics that are missing from previous model studies of deglaciation. © 2016 American Geophysical Union. Issue online: 13 September 2016; Version of record online: 31 August 2016; Accepted manuscript online: 9 August 2016; Manuscript Accepted: 4 August 2016; Manuscript Revised: 18 July 2016; Manuscript Received: 3 May 2016. This work has been supported by the NSF grant AGS-1303604. E.T. thanks the Weizmann Institute for its hospitality during parts of this work. A.R. has been supported by the NSF Graduate Research ...