Preprints
https://doi.org/10.5194/tcd-8-5511-2014
https://doi.org/10.5194/tcd-8-5511-2014

  28 Oct 2014

28 Oct 2014

Review status: this preprint was under review for the journal TC. A revision for further review has not been submitted.

Model calibration for ice sheets and glaciers dynamics: a general theory of inverse problems in glaciology

M. Giudici1,2,3, F. Baratelli1,4, A. Comunian1, C. Vassena1, and L. Cattaneo1,2 M. Giudici et al.
  • 1Università degli Studi di Milano, Dipartimento di Scienze della Terra "A. Desio", via Cicognara 7, 20129 Milano, Italy
  • 2CNR-IDPA (Consiglio Nazionale delle Ricerche, Istituto per la Dinamica dei Processi Ambientali), via Mario Bianco 9, 20131 Milano, Italy
  • 3CINFAI (Consorzio Interuniversitario Nazionale per la Fisica delle Atmosfere e delle Idrosfere), Piazza Niccolò Mauruzi 17, 62029 Tolentino (MC), Italy
  • 4MINES ParisTech, Centre de Géosciences, 35 Rue Saint-Honoré, 77305 Fontainebleau, France

Abstract. Numerical modelling of the dynamic evolution of ice sheets and glaciers requires the solution of discrete equations which are based on physical principles (e.g. conservation of mass, linear momentum and energy) and phenomenological constitutive laws (e.g. Glen's and Fourier's laws). These equations must be accompanied by information on the forcing term and by initial and boundary conditions (IBCs) on ice velocity, stress and temperature; on the other hand the constitutive laws involve many physical parameters, some of which depend on the ice thermodynamical state. The proper forecast of the dynamics of ice sheets and glaciers requires a precise knowledge of several quantities which appear in the IBCs, in the forcing terms and in the phenomenological laws. As these quantities cannot be easily measured at the study scale in the field, they are often obtained through model calibration by solving an inverse problem (IP). The objective of this paper is to provide a thorough and rigorous conceptual framework for IPs in cryospheric studies and in particular: to clarify the role of experimental and monitoring data to determine the calibration targets and the values of the parameters that can be considered to be fixed; to define and characterise identifiability, a property related to the solution to the forward problem; to study well-posedness in a correct way, without confusing instability with ill-conditioning or with the properties of the method applied to compute a solution; to cast sensitivity analysis in a general framework and to differentiate between the computation of local sensitivity indicators with a one-at-a-time approach and first-order sensitivity indicators that consider the whole possible variability of the model parameters. The conceptual framework and the relevant properties are illustrated by means of a simple numerical example of isothermal ice flow, based on the shallow-ice approximation.

M. Giudici et al.

 
Status: closed (peer review stopped)
Status: closed (peer review stopped)
AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment
Printer-friendly Version - Printer-friendly version Supplement - Supplement
 
Status: closed (peer review stopped)
Status: closed (peer review stopped)
AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment
Printer-friendly Version - Printer-friendly version Supplement - Supplement

M. Giudici et al.

M. Giudici et al.

Viewed

Total article views: 1,414 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
849 460 105 1,414 78 104
  • HTML: 849
  • PDF: 460
  • XML: 105
  • Total: 1,414
  • BibTeX: 78
  • EndNote: 104
Views and downloads (calculated since 28 Oct 2014)
Cumulative views and downloads (calculated since 28 Oct 2014)

Cited

Saved

Latest update: 22 Apr 2021
Download
Short summary
Inverse problems (IPs) are fundamental to calibrate numerical models, but are not yet widely diffused in glaciology, even if some theroretical developments and excellent applications were published. This paper provides a conceptual framework for IP in cryospheric sciences, which permits to clearly define and discuss the properties of IP. The examined topics include: the role of data and parameters: the link of ill-posedness with ill-conditioning and identifiability; sensitivity analysis.