Articles | Volume 14, issue 9
Research article
15 Sep 2020
Research article |  | 15 Sep 2020

Bayesian calibration of firn densification models

Vincent Verjans, Amber A. Leeson, Christopher Nemeth, C. Max Stevens, Peter Kuipers Munneke, Brice Noël, and Jan Melchior van Wessem

Data sets

Surface mass balance and snow depth on sea ice working group (SUMup) snow density subdataset, Greenland and Antarctica, 1950-2018 L. Koenig and L. Montgomery

Continuous density log of icecore BER11C95_25 S. Gerland and F. Wilhelms

Density of firn core DML96C07_39 F. Wilhelms

Sulfate-Based Volcanic Record from South Pole Ice Core J. Cole-Dai

Rapid ablation zone expansion amplifies north Greenland mass loss: modelled (RACMO2) and observed (MODIS) data sets B. P. Y. Noël

Dominion Range, Newall Glacier - Core and Snowpit Chemistry Data P. A. Mayewski, W. B. Lyons, G. Zielinski, M. Twickler, S. Whitlow, J. Dibb, P. Grootes, K. Taylor, P. Y. Whung, L. Fosberry, C. Wake, and K. Welch

The Community Firn Model C. M. Stevens

Short summary
Ice sheets are covered by a firn layer, which is the transition stage between fresh snow and ice. Accurate modelling of firn density properties is important in many glaciological aspects. Current models show disagreements, are mostly calibrated to match specific observations of firn density and lack thorough uncertainty analysis. We use a novel calibration method for firn models based on a Bayesian statistical framework, which results in improved model accuracy and in uncertainty evaluation.