School of Computing and Information Science

Published: 01/2013

Abstract:

Introduction:
A wide range of evidence shows that the current distribution of ice on the surface is anomalous, and that the Amazonian period was characterized by a variety of non-polar ice-related deposits ranging from high-latitude mantles, to mid-latitude lobate debris aprons, lineated valley fill, concentric crater fill, and pedestal craters, to low-latitude tropical mountain glaciers [1-3]. General circulation models (GCM) and glacial flow models illustrate the orbital parameter and atmospheric/surface conditions under which periods of glaciation are favored [e.g., 4-5], and the resulting patterns of accumulation of snow and the flow of ice [6-8]. Geological observations and impact crater size/frequency distribution data strongly suggest that during the Late Amazonian, a significant part of the mid-to-high latitudes in both hemispheres was covered by regional snow and ice deposits (preserved today beneath pedestal craters [9-11]) and that local depressions (primarily impact craters) were the sites of significant ice accumulation, and preservation beneath a residual debris cover (concentric crater fill (CCF) [12-13]). Pedestal crater (Pd) heights show that a significant amount of snow and ice accumulated in the mid-to-high latitudes during these periods (regionally the mean height is ~50 m, but values up to 160 m are seen in Utopia [14]). Based on previous work [15] we have concluded that at typical mid-latitude Martian temperatures (215 K) with typical flat inter-crater terrain (~1o) considerable ice thickness (800-1000 m) is required to initiate flow.

Based on this, we ruled out the likelihood that the Pd layer was the last phase of a thick persistent ice sheet that reached a configuration capable of supporting significant flow. Instead, we feel the Pds measure a transient, relatively thin, ice-rich layer that deformed as it covered and flowed into the crater depressions. Evidence exists for the latter case in that Pd, perched craters, and excess-ejecta craters, described in detail in [9-11], relate to the impacts into an ice-rich layer that is at most a few hundred meters thick. Each type reflects differing degrees of penetration, followed by complete  sublimation of any un-armored regions of the ice complex. Repeated deposition and removal of this thin layer is suggested in that an ~80 m height difference is observed between two Pds 20 km apart. In addition, 30 have superimposed Pds. For this to occur, the first Pd would have to form, the entire ice-rich layer outside the armored zone would have to be removed, a second ice-rich layer would have to re-form, and the superimposed Pd could then be emplaced. Clearly this requires that there be multiple episodes of ice-rich layer cover.

GCM results [4-5] predict ice accumulations as high as 10 mm/a during periods of high obliquity exactly in the mid-high latitudes where these craters occur. With such high accumulation rates, coupled with the steady 100 Ka beat of the Laskar et al. orbital solutions [13], one must ask why the transient layer never gets any thicker than the 50-160 m suggested by the Pds, since GCM results suggest this layer could form in as little as 20 Ka, a time much shorter than the duration of the obliquity excursions. Estimates of the volume of the Pd-defined layer [9-12] are close to the known volumes of the polar caps [16-17] that are the source of moisture for the high-obliquity mid-high latitude precipitation. We have suggested [15] that the transient layer is “supply-limited” in that when the cap is exhausted, the source is removed, and deposition ceases even if the obliquity is still high. We examine here the likelihood that this can be the case. How thick a transient layer is possible, given the constraints on the total amount of water available?

Citation:

J. L. Fastook and J. W. Head. Amazonian non-polar glaciation: Supply-limited glacial history and the role of ice sequestration. Lunar and Planetary Science XXXXIV, #1256, 2013.