A Bayesian ice thickness estimation model for large-scale applications

Mauro A Werder,Matthias Huss,Frank Paul,Daniel Farinotti,Amaury Dehecq

doi:10.1017/jog.2019.93

Abstract

AbstractAccurate estimations of ice thickness and volume are indispensable for ice flow modelling, hydrological forecasts and sea-level rise projections. We present a new ice thickness estimation model based on a mass-conserving forward model and a Bayesian inversion scheme. The forward model calculates flux in an elevation-band flow-line model, and translates this into ice thickness and surface ice speed using a shallow ice formulation. Both ice thickness and speed are then extrapolated to the map plane. The model assimilates observations of ice thickness and speed using a Bayesian scheme implemented with a Markov chain Monte Carlo method, which calculates estimates of ice thickness and their error. We illustrate the model's capabilities by applying it to a mountain glacier, validate the model using 733 glaciers from four regions with ice thickness measurements, and demonstrate that the model can be used for large-scale studies by fitting it to over 30 000 glaciers from five regions. The results show that the model performs best when a few thickness observations are available; that the proposed scheme by which parameter-knowledge from a set of glaciers is transferred to others works but has room for improvements; and that the inferred regional ice volumes are consistent with recent estimates.

Highlights

Glacier volume and ice thickness distribution are highly important for various aspects of glaciology, as well as for related sciences, such as hydrology (e.g. IPCC, 2013; Bahr and others, 2015)
We present the results of our model for three different setups: (i) we revisit the test glacier Unteraar, used in the Methods section, to illustrate various aspects of the model; (ii) we calibrate and validate the model using glaciers with ice thickness data from four Randolph Glacier Inventory (RGI) regions, with a focus on our method for regional-scale application; and (iii) we demonstrate the large-scale applicability of the BITE-model by applying it to 30 000 glaciers in five RGI regions
The Unteraar Glacier test-case is taken from the Intercomparison eXperiment (ITMIX) project (Farinotti and others, 2017) which provides the following data: outline and surface digital elevation models (DEMs), mass balance

Summary

Introduction

Glacier volume and ice thickness distribution are highly important for various aspects of glaciology, as well as for related sciences, such as hydrology (e.g. IPCC, 2013; Bahr and others, 2015). It is imperative to be able to infer ice volume and thickness distribution of large glacier samples from readily available data, such as glacier inventories and digital elevation models (DEMs). The development of approaches capable of estimating ice thickness has recently become a relevant branch of glaciological research and various models of different complexity and input data requirements have been proposed (see Farinotti and others, 2017, for a review). Considerable differences were found in the skill of the individual models to reproduce point ice thickness observations but even the best approaches were subject to large errors. This indicates a pressing need to further develop the corresponding approaches

Objectives

Methods

Results

Discussion

Conclusion