Inferring Ice Stream Properties
Overview
Many important properties of glaciers are unobservable but necessary to constrain to model current ice dynamics and predict future changes accurately. Inverse methods are used to infer these unobservable parameters from observable quantities. In particular, these methods are used to estimate ice viscosity and basal properties - key properties that dictate the forces that resist flow in ice streams - from surface velocity datasets. My work focuses on evaluating the accuracy of these methods and using them to estimate ice softness and ice rheology in rapidly-deforming regions of ice streams (in particular, the margins of ice streams).
Figure 1
I show using synthetic tests that we can infer both ice viscosity (a function of the flow-rate parameter such that large values of the flow-rate parameter translate to low ice viscosity) and basal slipperiness (a representation of how ‘slippery’ the bed is, or the lack of resistance the bed provides to the ice moving over it) if strain-rates are included in the regularization. Synthetic observed velocity and the synthetic ‘true’ parameter distributions are found in the first two rows. The misfit in the first row is a metric for the ‘convergence’ of the method. The results of the inversion are shown in the bottom two rows, in which we show that our method accurately reproduces the spatial variability of the two parameters.
Figure 2
We apply this new approach to estimate the flow-rate parameter (A) and basal slipperiness (the inverse of basal drag) in Bindschadler and MacAyeal Ice Streams in West Antarctica. We find that ice is soft in the margins (high flow-rate parameter) and that the ice streams flow over very soft sediment (very low basal drag/very high slipperiness).
Related Publications
[1] Ranganathan, M., Minchew, B., Meyer, C., & Gudmundsson, G. (2021). A new approach to inferring basal drag and ice rheology in ice streams, with applications to West Antarctic Ice Streams.Journal of Glaciology, 67(262), 229-242. doi:10.1017/jog.2020.95