We document the density and hydrologic properties of bare, ablating ice in a
mid-elevation (1215 m a.s.l.) supraglacial internally drained catchment in
the Kangerlussuaq sector of the western Greenland ice sheet. We find
low-density (0.43–0.91 g cm
Each summer a vast hydrologic network of lakes and rivers forms on the surface of the western Greenland ice sheet ablation zone in response to surface melting (Chu, 2014; Smith et al., 2015). Evidence suggests that most or all of this water is efficiently delivered via supraglacial rivers to moulins, crevasses, and, ultimately, to proglacial rivers and surrounding oceans (van As et al., 2017; Colgan et al., 2011; Lindbäck et al., 2015; Rennermalm et al., 2013; Smith et al., 2015). The assumption of efficient meltwater delivery is reflected in regional climate and surface mass balance models for Greenland that instantaneously credit ablation zone surface runoff to the ocean with no physical representation of hydrologic processes taking place on the bare ice surface (Smith et al., 2017). Field studies and satellite remote sensing, however, have found evidence of substantial meltwater runoff delays on daily to monthly timescales in the ablation zone (van As et al., 2017; Karlstrom and Yang, 2016; Koenig et al., 2015; Lindbäck et al., 2015; Overeem et al., 2015; Rennermalm et al., 2013; Smith et al., 2017). Similar runoff delays are observed on valley glaciers elsewhere (Karlstrom et al., 2014; Munro, 1990), inferred to relate to the presence of a degraded, porous “weathering crust” (Müller and Keeler, 1969) on the bare ice surface of glaciers and ice sheets that stores meltwater, delaying its delivery to supraglacial channels via porous subsurface flow (Irvine-Fynn et al., 2011; Karlstrom et al., 2014; Munro, 2011). The porous weathering crust may also provide a locus for internal and/or surficial refreezing of meltwater (Hoffman et al., 2014; Paterson, 1972; Willis et al., 2002). Together, hydrologic processes in the weathering crust are similar to those of meltwater transport, storage, and refreezing in snow and firn (Cox et al., 2015; Forster et al., 2014; Harper et al., 2012; Machguth et al., 2016). The presence of weathering crust in Greenland, however, has gone largely undocumented, and little is known about its effect on hydrologic processes in the bare ice ablation zone, where > 85 % of ice sheet surface meltwater runoff is generated (Machguth et al., 2016).
Weathering crusts are fractured, disintegrated, or “rotten” ice layers that form during the melt season on the thermally transient surface of ablating glaciers (Fountain and Walder, 1998; Irvine-Fynn et al., 2011; Müller and Keeler, 1969). In temperate ice, liquid water exists within an interconnected network of meltwater veins (Lliboutry, 1996; Mader, 1992; Nye and Frank, 1973). When glacier ice is exposed to water, these veins coarsen to the order of tenths of a millimetre in diameter, a process referred to as “rotting” (Nye, 1991). On bare ice surfaces exposed to solar radiation, this action is intensified by the transmission and absorption of solar radiation through the upper few metres of ice (Cook et al., 2016; Fountain and Walder, 1998; Irvine-Fynn and Edwards, 2014). Subsurface radiative heating enhances melting along ice grain edges, further coarsening vein networks and disaggregating ice crystals, creating a layer of porous ice typically < 2 m thick (Fig. 1) (Irvine-Fynn et al., 2011; Müller and Keeler, 1969).
Weathering crust formation reflects a balance between the vertical depth of
subsurface melting and the rate of ice surface lowering. This balance evolves in response to spatio-temporal changes in the surface energy balance
during the melt season (Müller and Keeler, 1969).
During clear-sky conditions, solar radiative heating promotes development
and deepening of the weathering crust. The depth of subsurface melting is
typically limited to < 2 m by the exponential attenuation of
radiative heating with depth (Brandt and
Warren, 1993). Conversely, during exceptionally warm, windy, or cloudy
conditions when surface melt rates are enhanced relative to subsurface
melting, the weathering crust may decay or be rapidly removed (Müller and Keeler, 1969). As the weathering crust
develops, a shallow, depth-limited aquifer may establish in the near-surface
porous ice (Fig. 1) (Irvine-Fynn and Edwards, 2014). At
vertical depths where meltwater drains through the permeable weathering
crust to seeps and supraglacial channels, the near-surface ice density is
reduced with no detectable change in glacier surface height (Hoffman et al., 2014; Müller and Keeler,
1969). Consequently, weathering crust ice density exhibits a characteristic
non-linear increase from a very low-density (< 0.5 g cm
Weathering crust is often enhanced by cryoconite, the biologically active dark sediment that preferentially absorbs solar radiation, locally enhancing melt in quasi-cylindrical holes that deepen into the weathering crust and brighten the ice surface relative to dispersed or uniform debris-covered ice (Bøggild et al., 2010). Cryoconite holes are coupled to the weathering crust via porous subsurface water exchange (Cook et al., 2016) and surface flow that redistributes sediments, nutrients, and microbial cells between holes and the ice surface, potentially controlling their distribution and ecological structure (Edwards et al., 2011; Hodson et al., 2007; Hotaling et al., 2017; Takeuchi et al., 2000). Weathering crust hydrology, therefore, exerts a dynamic control on the “photic zone” where solar radiation, liquid water, nutrients, and air provide habitat for a rich microbial community within the upper few metres of an ablating glacier (Fig. 1) (Irvine-Fynn and Edwards, 2014). Physical controls on these ecohydrological interactions have only recently been explored and remain poorly understood, especially the seasonal evolution of depth-variable ice density, permeability, water storage, and microbial mobility (Cook et al., 2016; Irvine-Fynn et al., 2012; Stevens et al., 2018).
Despite the hydrological and ecological implications of weathering crust for supraglacial processes, no studies have described the physical structure or documented subsurface meltwater storage for the Greenland ice sheet weathering crust. When present, the weathering crust could provide a temporary storage reservoir, thus modulating meltwater delivery to supraglacial channels, crevasses, and englacial hydrologic systems (Karlstrom et al., 2014). In addition, because mass may be removed from the weathering crust without detectable change in glacier surface height, the growth and decay of the crust may confound estimates of sub-seasonal surface mass balance made from ice surface elevation change or surface energy balance models that neglect its presence (van den Broeke et al., 2008; Munro, 1990). Weathering crust structure and hydrologic storage is therefore an important but understudied component of the Greenland ice sheet bare ice ablation zone. The purpose of this study is to describe the physical structure and hydrologic storage of the weathering crust in a mid-elevation Greenland ice sheet supraglacial catchment. We provide an initial set of measurements of near-surface ice density, porosity, water saturation, and water table height, and use these data to estimate meltwater storage within the weathering crust. To illustrate the implications of our findings, we extrapolate this storage estimate across the study catchment for comparison with proglacial meltwater runoff volumes. Finally, we discuss broader implications of the findings for ablation zone hydrology and surface mass balance processes to guide future work.
The data presented in this study were collected during a 6–14 July 2016
field campaign in the middle ablation zone (67.049
Ortho-rectified image mosaic of the study area at 6 cm ground
resolution from RGB camera imagery collected 10 July 2016 on board a
quadcopter drone. Background 30 m Landsat image collected same day. Shallow
ice cores extracted at 80 m intervals (blue circles) along the 800 m
transect provide ice density measurements to depths of 1.1 m, with two
additional shallow ice cores extracted to 1.8 m depth at interval 1. Insets
(below) show the 63.1 km
At 80 m intervals along the 800 m transect (Fig. 2), shallow ice cores
0.9–1.1 m deep were collected with a 7.25 cm diameter Kovacs Mark III
coring system (
At six sites (cores 1, 2, 4, 5, 9, and 10), the upper 14–30 cm of ice
lacked sufficient cohesion for intact removal with the coring system. To
obtain density measurements for this material, ice samples were removed
adjacent to the core sites with a Snowmetrics©(
Density measurement uncertainty cannot be quantified with known accuracy as
each ice core segment was unique in size and shape. Based on visual
inspection, we consider 1.5 cm (
The porosity of the near-surface ice was examined to determine the liquid
meltwater storage capacity of the study area weathering crust. In theory,
the total porosity of a solid material is the ratio of pore space volume to
total volume and is calculated from the ratio of measured density to pure
material density (Dingman, 2002):
To measure
To estimate
At 8 m intervals along the 800 m transect, the presence/absence of liquid water saturation within the weathering crust, the depth of cryoconite holes, and the depth to water within cryoconite holes were measured with respect to the ice sheet surface. First, the presence/absence of liquid water saturation was assessed by drilling a 1 m deep hole into the weathering crust with a 5 cm diameter Kovacs auger. The drilled holes were monitored for liquid water refilling within 30 min as an indication of subsurface water saturation. Second, the nearest cryoconite hole within a 1 m radius of each measurement interval was identified and the total depth of each hole and the depth to water in each hole below the surface were measured. The height of water in each hole is calculated as the difference between the depth of the hole and the depth to water. The depth to water in the holes is used as an estimate of the depth to liquid water saturation (i.e. the water table height). Absence of cryoconite holes was noted if none were present within a 1 m radius of the 8 m measurement interval.
As an additional qualitative check on the weathering crust structure, a Snowmetrics ©steel pointed depth probe was forced downward adjacent to each 1 m drilled hole until impenetrable ice was encountered. The expectation was that these measurements would approximate the depth to the shoulder of the subsurface density profile, roughly corresponding to the depth of rotten unsaturated ice as per Fig. 1 in Müller and Keeler, (1969). Initial field observations confirmed the upper few tens of centimetres of ice was composed of weakly-bonded, coarse-grained ice that was easily removed with a flat bladed shovel and penetrated with the depth probe. The depth probe measurements are used as a qualitative description of the weathering crust structure in Sect. 3.3.
The total volumetric water storage
Finally, for illustrative purposes we scale our
Throughout the study area, the ice sheet surface was characterized by a layer
of coarse-grained, weakly-bonded ice, a few tens of centimetres thick
(Fig. 3). Bulk
Subsurface
Subsurface measured ice density (
The source of this density variability likely corresponds to core
stratigraphy. While coring, alternating weak and resistant layers were
qualitatively observed based on the resistance to downward motion. This
structure was confirmed by the presence of alternating layers of
coarse-grained (> 1 cm), weathered ice and clear, solid ice
lenses in all cores. The ice lenses were readily identified in the core
stratigraphy and removed intact from the granular, friable ice between
lenses (Fig. 5). The ice lenses contained visible
closed air bubbles trapped in clear solid ice. Subtle evidence of internal
melting along coarse grain edges was visible in some ice lenses, but most
were solid with minimal or no apparent evidence of weathering. Densities of
these lenses were not measured in the field but based on their solid
structure are estimated to be in the range of typical glacier ice densities
(e.g. 0.83–0.90 g cm
Previous analyses of weathering crusts have not reported ice structure,
therefore the pattern we find of alternating coarse-grained, weathered ice
and clear, solid ice cannot be compared to previous studies (e.g. Hoffman et
al., 2014; Müller and Keeler, 1969; Schuster, 2001). Though refrozen
meltwater lenses are found in firn at elevations above the study area (Cox et
al., 2015; Machguth et al., 2016), refrozen meltwater lenses are unlikely in
a bare ice, ablating weathering crust (Schuster, 2001). Rather, the observed
stratigraphy likely reflects differential weathering of the underlying
structural ice fabric (Hudleston, 2015). Surface expression of differential
weathering is visible as contrasting dark and light areas along the transect
(Fig. 2), similar to kilometre-scale foliated bands associated with
outcropping of stratified impurities in the study region (Wientjes et al.,
2012). At the scale of the shallow ice cores, stratified distributions of
crystal size and shape, bubble elongation and distribution, and impurity
content with depth could each influence rates of subsurface radiative
heating (Brandt and Warren, 1993; Liston et al., 1999) and hence could
promote differential weathering of centimetre-scale foliated ice layers at
depth (Hudleston, 2015). Meltwater advection along micro-seams and cracks, or
along foliated planes with enhanced permeability (Wakahama et al., 1973)
could provide an additional differential heat source at depth, either via
enhanced rotting of temperate ice (Nye, 1991) or, if transported to cold ice,
via meltwater refreezing. The ice lenses, then, may represent structural
resistance to weathering, and/or result from heterogeneity in subsurface
flow paths that promote differential weathering of subsurface ice. We would
thus expect lenses to be localized features, which helps explain the lack of
consistent stratigraphy among cores. Mechanism aside, the
Effective porosity
Linear relationship (
The ice surface topography along the study transect was highly variable
across short spatial scales (< 10 m) (Fig. 7). Qualitatively, the
surface was characterized by hummocks and hollows separated by shallow rills
(often flowing) and pitted cryoconite deposits. Water-filled cryoconite holes
were ubiquitous across the study area surface, though variability in
cryoconite hole water levels and spatial coverage was observed. For example,
at 14 of the 100 measurement locations no cryoconite holes were present
within the nominal 1 m observation radius and at nine locations all cryoconite
holes within the 1 m radius were dry at the time of observation. At the
remaining 77 locations cryoconite holes contained measurable water levels.
Cryoconite holes were 25.2
In 83 of 100 drilled 1 m holes, water from surrounding ice refilled the hole within the nominal 30 min post-drilling observation period. Refilling rates were not systematically measured but were observed to vary from nearly instantaneous refilling before the auger was removed, to relatively slow (and incomplete) refilling over the 30 min observation period, suggesting substantial variability over short spatial scales. In addition to this rapid refilling and the widespread presence of water-filled cryoconite holes, all but one of the ten shallow ice core boreholes were observed filling with water during the post-drilling period, though the equilibrium height of water in these holes was not measured. Collectively, these measurements suggest the ice was saturated across the entire 800 m transect to a depth of at least 1 m, albeit with substantial spatial variability in refilling rates.
Based on these observations, we characterize the near-surface ice as composed
of two continuous layers with varying thicknesses. The upper layer consisted
of low-density (0.33–0.56 g cm
The vertical structure of the higher density, saturated ice was highly variable, consisting of alternating layers of coarse-grained, porous ice and clear, solid ice lenses. The thickness of the saturated ice layer could not be definitively determined with the drilling equipment. However, at two locations shallow ice cores 1.8 m deep were extracted. The densities of these cores were not measured, but at both sites the ice cores consisted of coarse-grained, porous ice alternating with clear, solid ice lenses across their entire depth. There were no qualitative differences between the ice in these cores and the ice presented in Fig. 4 and Fig. 5. At one of these two sites, weathered ice persisted to 1.8 m depth. At the other site, a 20 cm thick segment of solid ice was found between 1.6 and 1.8 m, possibly marking the transition to cold, solid, impermeable ice at this location.
Averaged across the 94 cm mean depth of the 10 shallow ice cores,
Shallow ice core depth, mean core density, mean core porosity, and
specific water storage depth (
Meteorological records of
Given the transient nature of the weathering crust, it is important to place
these findings in a seasonal context. Antecedent meteorology such as the
timing of snowmelt, rainfall, and prevalence of shortwave radiation, would
each influence weathering crust growth and decay. Albedo data recorded at the
KAN-M automatic weather station (AWS) indicate the spring snow cover melted
out on
Further, MAR data suggests conditions during summer 2016 favoured weathering
crust growth in the study region. These include below average cloud cover
and rainfall, and above average downward shortwave radiation (e.g. compare
to 2000–2016 period, Figs. 1–4 in Tedstone et al., 2017). These
meteorological conditions suggest that the presence of a well-developed
weathering crust in the study area at the time of observation is not
surprising, though inferring a likely thickness is not possible without a
physical model for weathering crust development. The AWS data presented in
Fig. 8 provide context for our study, but a
detailed investigation of weathering crust formation is well beyond our
scope here. Nevertheless, the > 1.6 m thickness of weathered ice
we find is perhaps surprising given the ephemeral snow cover and
We have presented measurements of near-surface ice density which, to our knowledge, provides the first characterization of the structure and hydrologic storage of a bare ice weathering crust in the Greenland ice sheet ablation zone. These data suggest 14–18 cm of liquid meltwater was stored within porous, low-density ice at the time of observation, and that substantial subsurface melting may occur in the Greenland ice sheet bare ice ablation zone. Together, these findings suggest hydrologic processes in the bare ice ablation zone are affected by porous ice, and that surface lowering measurements may not accurately quantify total mass loss during periods of weathering crust growth and decay in the Greenland ice sheet ablation zone.
Water storage in the weathering crust has been reported (Irvine-Fynn et al., 2011; Larson, 1978) but is generally not considered a significant component of water storage in supraglacial environments, owing to its transient nature (Fountain and Walder, 1998; Jansson et al., 2003; Müller and Keeler, 1969). While more work is required to determine the spatial extent and seasonal evolution of the conditions found in this investigation, our documentation of a saturated weathering crust storing up to 18 cm of liquid meltwater supports the possibility of a substantial transient reservoir in Greenland's bare ice ablation zone, consistent with observations of weathering crust for supraglacial environments worldwide (Irvine-Fynn, 2008; Larson, 1978; Munro, 1990). Though a snapshot characterization, the weathering crust structure presented in Fig. 7 is consistent with conceptual models of the near-surface weathering crust–cryoconite hole hydrologic system (e.g. Fig. 1; Irvine-Fynn and Edwards, 2014; Müller and Keeler, 1969) and confirms this system is present in the Greenland ice sheet ablation zone. The ubiquity of water-filled cryoconite holes, the rapid refilling of drilled holes with liquid water, and the excavation of saturated ice cores to depths > 1.6 m suggests the study area weathering crust acts as a depth-limited aquifer (Irvine-Fynn et al., 2011), storing meltwater in the seasonally-temperate near-surface ice and likely delaying the delivery of meltwater to supraglacial streams and rivers via saturated subsurface flow (Irvine-Fynn et al., 2011; Karlstrom et al., 2014; Munro, 2011).
In addition to meltwater storage, we describe the structure of the weathering crust. We find a pattern of porous, granular ice alternating with solid ice lenses in the upper 1–2 m of weathering crust in the study area, rather than a homogeneous rotten near-surface ice layer (e.g. Müller and Keeler, 1969). Given the rapidly ablating ice surface prior to the study, we posit the solid ice lenses are emergent structural features, as refrozen meltwater is unlikely in an ablating weathering crust (Schuster, 2001). Though beyond the scope of the data collected in this study, we hypothesize two mechanisms to explain the observed stratigraphy. First, stratified distributions of crystal size and shape, bubble elongation and distribution, and impurity content with depth could influence rates of subsurface radiative heating (Brandt and Warren, 1993; Liston et al., 1999). The ice lenses may then represent optically transparent ice layers with larger crystal size, lower air bubble content, or lower impurity content. The optical properties of these layers may reduce absorption of shortwave radiation, substantially reducing internal melting relative to optically opaque layers. Second, meltwater advection along micro seams, cracks, or foliated ice layers with enhanced permeability may promote differential melting via sensible and frictional heat transfer (Hambrey, 1977; Hambrey and Lawson, 2000; Wakahama et al., 1973). Therefore, underlying structural features such as foliation, cracks, and fractures caused by thermal expansion (Sanderson, 1978) may be accentuated by differential radiative heating, enhanced “rotting” by meltwater along preferential flow paths, or heating due to meltwater refreezing. Together, these suggest weathering crust formation in the study area may be more complicated than previous descriptions of a process driven solely by solar radiative heating (Hoffman et al., 2014; Müller and Keeler, 1969), and suggest meltwater dynamics and ice structure may be important controls on weathering crust development.
Though we interpret the lenses as structural features, there is evidence that internal refreezing of meltwater occurs in weathering crust on the Dry Valley glaciers in Antarctica (Hoffman et al., 2014). Although the climatic context is different, this raises the possibility of meltwater refreezing within the weathering crust ice matrix in Greenland. If so, refreezing would represent a heat source within near-surface ice, and a possible sink for meltwater retention (Pfeffer et al., 1991). Though detailed energy balance studies suggest internal refreezing is negligible in near-surface porous ice on alpine glaciers in the Canadian Rockies (Paterson, 1972; Schuster, 2001), such analyses have not been performed for the Greenland ice sheet ablation zone. Regardless of internal refreezing at depth, we frequently observed night-time refreezing of meltwater at the surface of cryoconite holes and water tracks in the study area (Fig. 9), though the magnitude of this refreezing was not studied. In addition to careful observation of subsurface ice structure, future work should determine if internally refrozen meltwater occurs within weathering crust in the Greenland ice sheet ablation zone, especially during seasonal transitions from temperate to cold near-surface ice.
Night-time refreezing of meltwater at the surface of
While extrapolating these local scale findings to broader areas of the
Greenland ice sheet is not justified presently, it is illustrative to
consider the potential meltwater storage volume of the weathering crust in
our study catchment. For example, if we assume our shallow ice core data are
broadly representative of conditions across its 63 km
Our findings of low-density, saturated weathering crust in the Greenland ice sheet ablation zone have at least three implications for Greenland ice sheet surface mass balance (SMB). First, subsurface meltwater generation within the weathering crust does not materially lower the ice surface (Braithwaite et al., 1998; Müller and Keeler, 1969; Munro, 2011) Lateral drainage of internal meltwater through the permeable weathering crust to supraglacial channels reduces weathering crust ice density, by removing mass with no detectable change in surface height. As a result, mass change during periods of weathering crust development may be underpredicted or, during periods of weathering crust removal, overpredicted, if determined solely from ice surface elevation changes (Braithwaite et al., 1998; LaChapelle, 1959; Müller and Keeler, 1969). In the Kangerlussuaq region of the southwest Greenland ice sheet ablation zone, penetration of shortwave radiation into near-surface ice is estimated to generate 20–30 % of total summertime melt, suggesting ice surface elevation change measurements may not be reliable for short-term model validation in this region unless subsurface melt is accounted for (van den Broeke et al., 2008; Munro, 1990).
Second, the timing, magnitude, and location of meltwater delivery to the
englacial system is powerfully altered by surface hydrologic processes
operating on the Greenland ice sheet bare ice surface
(Smith et al., 2017). In
addition to catchment size and shape, transient water storage in the
weathering crust has been inferred to attenuate the timing of meltwater
delivery to englacial and proglacial hydrologic systems
(Karlstrom et al., 2014; Munro,
2011). Typical flow velocities of 0.4–2.6 m s
Finally, the weathering crust provides a substrate for retention of impurities, cryoconite, and microbial communities that influence the Greenland ice sheet ablation zone surface albedo (Bøggild et al., 2010; Lutz et al., 2014; Ryan et al., 2017; Yallop et al., 2012). Cryoconite deposits locally enhance melt, forming quasi-cylindrical melt holes that deepen into the weathering crust (Gribbon, 1979), likely reducing their direct effect on mesoscale ice albedo patterns in southwest Greenland (Ryan et al., 2018; Tedstone et al., 2017). Conversely, interstitial water within the weathering crust, such as that documented in this study, provides abundant habitat for microalgae and cyanobacteria (Irvine-Fynn and Edwards, 2014), which reduce ice surface albedo (Yallop et al., 2012). Subsurface water exchange may further redistribute soluble impurities and microbes between the permeable weathering crust and cryoconite holes (Cook et al., 2016), while channel invasion of cryoconite holes during periods of weathering crust removal may disperse cryoconite sediments and microbes across the ice surface (Hodson et al., 2007; Takeuchi et al., 2000). Thus, while it has not been confirmed, weathering crust hydrology, in addition to its growth and removal, could modulate the distribution of impurities and microbial communities on the Greenland ice sheet ablation zone surface, and hence could influence surface albedo patterns.
Underpinning each of these implications of weathering crust, however, is the transient nature of its growth and decay. Our study provides a snapshot characterization of what appears to be a deeply developed weathering crust, approximately midway through a summer characterized by below average cloud cover, albedo, rainfall, and spring snow depth, earlier than average snow disappearance, and above average downward shortwave radiation (e.g. compare to Fig. 1–4 in Tedstone et al., 2017). These conditions suggest abundant time for weathering crust development, and lack of conditions conducive to its removal or decay. Interannual variability in these conditions is substantial, and the conditions we document may not be representative of normal conditions. The net seasonal effect of weathering crust processes on Greenland ice sheet ablation zone hydrology and mass balance remains poorly understood and should form the basis for future work.
This study suggests the presence of a water-saturated weathering crust at least
1 m thick on the bare ice surface of the Greenland ice sheet ablation zone.
The observed characteristics of this weathering crust are similar to those
described for supraglacial environments worldwide (Cook et al., 2016; Hoffman
et al., 2014; Irvine-Fynn and Edwards, 2014; Karlstrom et al., 2014; Larson,
1978; Müller and Keeler, 1969; Munro, 2011). Namely, the weathering crust
acts as a depth-limited aquifer (Irvine-Fynn et al., 2011), storing liquid
meltwater and likely slowing its transport to supraglacial streams via porous
subsurface flow (Cook et al., 2016; Karlstrom et al., 2014). Our empirical
relationship (
The data used in this investigation are archived in the
PANGAEA open access data repository at
MGC, LCS, and AKR designed the experiment. MGC, CM, AKR, LHP, JR, and SC collected the field data. JR assisted with unmanned aerial system image processing. MGC performed the data analysis. MGC wrote the manuscript with contributions from all authors.
The authors declare that they have no conflict of interest.
This project was funded by the NASA Cryosphere Program grant NNX14AH93G
(P.I. Laurence C. Smith) managed by Thomas
P. Wagner. We thank Robert Hawley of Dartmouth University for the generous
lending of the shallow ice corer. We thank Polar Field Services for their
field support, Charlie Kershner (George Mason University), Brandon Overstreet
(University of Wyoming), Sasha Leidman (Rutgers University), and Rohi
Muthyala (Rutgers University) for their field work assistance. Data from the
Programme for Monitoring of the Greenland Ice Sheet (PROMICE) and the
Greenland Analogue Project (GAP) were provided by the Geological Survey of
Denmark and Greenland (GEUS) at