01 Dec 2022
01 Dec 2022
Status: this preprint is currently under review for the journal TC.

Observed and modeled Greenland firn properties (1980–2020)

Megan Thompson-Munson1, Nander Wever1, C. Max Stevens2, Jan T. M. Lenaerts1, and Brooke Medley2,3 Megan Thompson-Munson et al.
  • 1Department of Atmospheric and Oceanic Sciences, University of Colorado Boulder, Boulder, CO, USA
  • 2Earth System Science Interdisciplinary Center, University of Maryland College Park, College Park, MD, USA
  • 3NASA Goddard Space Flight Center, Greenbelt, MD, USA

Abstract. The Greenland Ice Sheet's (GrIS) firn layer buffers the ice sheet's contribution to sea level rise by storing meltwater in its pore space. However, available pore space and meltwater retention capability is lost due to ablation of the firn layer and refreezing of meltwater as near-surface ice slabs in the firn. Understanding how firn properties respond to climate is important for constraining the GrIS's future contribution to sea level rise in a warming climate. Observations of firn density provide detailed information about firn properties, but they are spatially and temporally limited. Here we use two firn models, the physics-based SNOWPACK model and the semi-empirical Community Firn Model (CFM) to quantify firn properties across the GrIS from 1980 through 2020. We use an identical forcing (MERRA-2 atmospheric reanalysis) for SNOWPACK and the CFM in order to isolate model differences. To evaluate the models, we compare simulated firn properties, including firn air content (FAC), to measurements from the SUMup dataset of snow and firn density. Both models perform well, though their performance is hindered by meltwater percolation and the spatial resolution of the atmospheric forcing. In the full ice-sheet simulations, the spatially-integrated FAC (i.e., air volume in the firn) for the upper 100 m is 34,645 km3 from SNOWPACK and 28,581 km3 from the CFM. The discrepancy in the magnitude of the modeled FAC stems from differences in densification with depth and variations in the models' treatment of atmospheric input. In more recent years (2005–2020), both models simulate substantial depletion of pore space. During this period, the spatially-integrated FAC across the entire GrIS decreases by 2.8 % and 1.2 % in SNOWPACK and the CFM, respectively. The differing magnitudes of the 2005–2020 spatially-integrated FAC of -66.6 km3 yr-1 in SNOWPACK and -17.4 km3 yr-1 in the CFM demonstrate how model differences propagate throughout the FAC record. Over the full modeled record (1980–2020), SNOWPACK simulates a loss of pore space equivalent to 3 mm of sea level rise buffering, while the CFM simulates a loss of 1 mm. The greatest depletion in FAC is along the margins, and especially along the western margin where observations and models show the formation of near-surface, low-permeability ice slabs that inhibit meltwater storage.

Megan Thompson-Munson et al.

Status: final response (author comments only)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse

Megan Thompson-Munson et al.

Megan Thompson-Munson et al.


Total article views: 379 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
272 98 9 379 3 4
  • HTML: 272
  • PDF: 98
  • XML: 9
  • Total: 379
  • BibTeX: 3
  • EndNote: 4
Views and downloads (calculated since 01 Dec 2022)
Cumulative views and downloads (calculated since 01 Dec 2022)

Viewed (geographical distribution)

Total article views: 367 (including HTML, PDF, and XML) Thereof 367 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
Latest update: 06 Feb 2023
Short summary
To better understand the Greenland Ice Sheet’s firn layer and its ability to buffer sea level rise by storing meltwater, we analyze firn density observations and output from two firn models. We find that both models, one physics-based and one semi-empirical, simulate realistic density and firn air content when compared to observations. The models differ in their representation of firn air content, highlighting the uncertainty in physical processes and the paucity of deep firn measurements.