The Role of Thermal Stresses in Rock Weathering and Sediment Production in a Polar Desert: A Study of Surface Erosion from Mullins Glacier, Antarctica

Tuesday, 16 December 2014
Jennifer L Lamp1, David R Marchant1, Sean L Mackay1 and James W Head III2, (1)Boston University, Boston, MA, United States, (2)Brown University, Providence, RI, United States
In this study, we examine the physical weathering of Ferrar dolerite clasts along a multi-million-year-old soil chronosequence on Mullins Glacier, a debris-covered glacier in the McMurdo Dry Valleys (MDV), Antarctica. Collected morphological field data show that: (1) with increasing distance from the headwall, clasts become more rounded and deeply buried by surrounding sediment; (2) clasts with exposure histories > ~20,000 years exhibit disintegration via flaking of mm-scale surface fragments; (3) these flakes increase in thickness with distance, with an overall average of 1.8 mm. Coupling the field data with improved chronological control for surface ages, we estimate an erosion rate of ~9 cm Ma-1 via flaking, which represents ~ 60% of the currently assumed maximum erosion rate for the region. We test thermal fatigue as a mechanism for flake detachment by collecting high-frequency temperature data for rock surfaces and at depth on multiple clasts, as well as meteorological data (air temperature, relative humidity, wind speed/direction, solar intensity) during the austral summer. Vertical temperature gradients across flakes surpassed 8°C during the 28-day study interval, and rates of surface temperature change exceeded 5°C min-1. The latter value greatly exceeds the accepted value for producing thermal fracture in igneous rocks. Dolerite samples were also collected to determine rock surface albedo, near-surface geochemistry, and mechanical properties; these data, in combination with the acquired field data, were used to create a model of internal temperature and thermally induced stresses in a typical clast. Overall, our results demonstrate that the production of altered rinds modifies thermal properties at the rock surface and may help facilitate fracture at the interface between altered and unaltered material. Additional sediment analyses show that the detached flakes add to the surrounding regolith, increasing in abundance with inferred soil age. This process modifies clast shape and promotes self-burial, providing a negative feedback to further erosion. Our measurements imply that the detachment of altered material on Mullins Glacier represents a dynamic equilibrium process that may have important implications for rates of landscape evolution in ice-free polar deserts like the MDV.