The sensitivity of flowline models of tidewater glaciers to parameter uncertainty

Abstract

Abstract. Depth-integrated (1-D) flowline models have been widely used to simulate fast-flowing tidewater glaciers and predict change because the continuous grounding line tracking, high horizontal resolution, and physically based calving criterion that are essential to realistic modeling of tidewater glaciers can easily be incorporated into the models while maintaining high computational efficiency. As with all models, the values for parameters describing ice rheology and basal friction must be assumed and/or tuned based on observations. For prognostic studies, these parameters are typically tuned so that the glacier matches observed thickness and speeds at an initial state, to which a perturbation is applied. While it is well know that ice flow models are sensitive to these parameters, the sensitivity of tidewater glacier models has not been systematically investigated. Here we investigate the sensitivity of such flowline models of outlet glacier dynamics to uncertainty in three key parameters that influence a glacier's resistive stress components. We find that, within typical observational uncertainty, similar initial (i.e., steady-state) glacier configurations can be produced with substantially different combinations of parameter values, leading to differing transient responses after a perturbation is applied. In cases where the glacier is initially grounded near flotation across a basal over-deepening, as typically observed for rapidly changing glaciers, these differences can be dramatic owing to the threshold of stability imposed by the flotation criterion. The simulated transient response is particularly sensitive to the parameterization of ice rheology: differences in ice temperature of ∼ 2 • C can determine whether the glaciers thin to flotation and retreat unstably or remain grounded on a marine shoal. Due to the highly non-linear dependence of tidewater glaciers on model parameters, we recommend that their predictions are accompanied by sensitivity tests that take parameter uncertainty into account.

Understanding controls on rapid ice‐stream retreat during the last deglaciation of Marguerite Bay, Antarctica, using a numerical model

“…Likewise, as found by Enderlin et al . [], the model dependency upon parameters, for example, in calculating shear stresses, may influence the speed of grounding‐line retreat, but we do not believe these model limitations will substantially affect the spatial pattern of retreat slowdowns or accelerations. However, further investigations using three‐dimensional higher‐order models that fully resolve lateral shear stresses [e.g., Gudmundsson et al ., ; Pattyn et al ., ] will help to refine our understanding of the impact that topographic width has upon ice‐stream grounding‐line retreat rates.…”

Impact of Fjord Geometry on Grounding Line Stability

“…Grounding line retreat in flowline-type models has been suggested to depend on parameter choices (Enderlin et al, 2013b), as well as the calving law employed (Haseloff and Sergienko, 2017;Schoof et al, 2017). The crevasse-water-depth criterion we use assumes that calving is a result of longitudinal stretching (Benn et al, 2007).…”

Impact of frontal ablation on the ice thickness estimation of marine-terminating glaciers in Alaska

“…One method presented by Farinotti et al (2009) and successfully applied several times since then (e.g. Morlighem et al, 2011;Huss and Farinotti, 2012;Clarke et al, 2013;Maussion et al, 2019a) combines ice flow dynamics and mass conservation principles to constrain mass fluxes through given glacier cross sections.…”