In situ measurements of meltwater flow through snow and firn in the accumulation zone of the SW Greenland Ice Sheet

Abstract

Abstract. The Greenland Ice Sheet is losing mass, part of which is caused by increasing runoff. The location of the runoff limit, the highest elevation from which meltwater finds its way off the ice sheet, plays an important role in the surface mass balance of the ice sheet. The recently observed rise in runoff area might be related to an increasing amount of refreezing: ice layer development in the firn hinders vertical percolation and promotes lateral runoff. To investigate meltwater flow near the runoff limit in the accumulation zone on the southwest Greenland Ice Sheet, we carried out in situ measurements of hydrological processes and properties of firn and snow. The hydraulic conductivity of icy firn in pre-melt conditions measured using a portable lysimeter ranges from 0.17 to 12.8 m hr-1, with flow predominantly occurring through preferential flow fingers. Lateral flow velocities of meltwater on top of the near-surface ice slab at the peak of the melt season measured by salt dilution- and tracer experiments range from 1.3 to 15.1 m hr-1. With these lateral flow velocities the distance between the slush limit, the highest elevation where liquid water is visible on the ice sheet surface, and the runoff limit could be up to 4 km in regions where near-surface ice slabs are present. These measurements are a first step towards an integrated set of hydrological properties of firn on the SW Greenland Ice Sheet, and show evidence that meltwater runoff might occur from elevations above the visible runoff area.

In situ measurements of meltwater flow through snow and firn in the accumulation zone of the SW Greenland Ice Sheet

“…The Kozeny-Carman equation relates a medium's permeability to pressure drop and fluid viscosity for laminar flow through a packed bed of solids, and when combined with Darcy's law, it can be used to predict permeability (Kozeny, 1927;Carman, 1937;Bear, 1972):…”

Daily variations in Western Greenland slush limits, 2000–2021

“…While not measured directly by Holmes (1955), their study indicated that lateral runoff in white slush reached at least 4.2 km (2.6 miles) above the highest visible slush fields. Clerx and others (2022) directly measured lateral flow velocity inside the snow, in a shallow layer of meltwater on top of the ice slab. Multiplying measured velocities with a rough estimate of days per melt season where water can flow yields a value similar to Holmes (1955).…”

A thermodynamic nonequilibrium model for preferential infiltration and refreezing of melt in snow

“…At the macroscopic scale, refrozen melt structures have been readily observed in the field at the scale of meters to kilometers (Humphrey et al, 2012;Lazzaro et al, 2015;Culberg et al, 2021). Horizontal refrozen structures are low permeability regions in snowpack that hinder downward percolation, promote lateral runoff (Culberg et al, 2021;Clerx et al, 2022), and play an important role in snow hydrology.…”

“…While the generation of melt at snow surface can be relatively uniform in space, meltwater infiltration through the underlying snowpack is known to be highly heterogeneous in nature, forming (1) vertical preferential flow pathways that channelize meltwater (e.g., ice pipes) and (2) lateral flow pathways guided by horizontal low permeability zones (e.g., capillary barriers or ice lenses). Both types of preferential pathways have been observed in the field directly or indirectly (Campbell et al, 2006;Humphrey et al, 2012;Kinar & Pomeroy, 2015;Culberg et al, 2021;Clerx et al, 2022), but systematic investigation and mechanistic understanding of these phenomena are lacking. In particular, laboratory experiments in 3D samples (Waldner et al, 2004;Katsushima et al, 2013;Avanzi et al, 2016) have shown the percolation of meltwater into 3D snowpack/column to be intrinsically unstable, analogous to gravity-driven water infiltration through dry soil (Glass et al, 1989;Selker et al, 1992;Glass & Nicholl, 1996).…”

“…Additionally, when melt interacts with subfreezing snowpack, it can readily refreeze as ice and decrease local snow porosity. This refreezing process reduces the effective infiltration rate by both consuming liquid water available for transport and by lowering the hydraulic conductivity of snow, which hinders vertical percolation and promotes lateral runoff (Culberg et al, 2021;Clerx et al, 2022). The resulting heterogeneous porosity structures, such as ice pipes or ice lenses, play an important role in snow hydrology and geohazard assessment but the mechanism of their formation remains poorly understood.…”