Dark ice dynamics of the south-west Greenland Ice Sheet

Abstract

Abstract. Runoff from the Greenland Ice Sheet (GrIS) has increased in recent years due largely to changes in atmospheric circulation and atmospheric warming. Albedo reductions resulting from these changes have amplified surface melting. Some of the largest declines in GrIS albedo have occurred in the ablation zone of the south-west sector and are associated with the development of "dark" ice surfaces. Field observations at local scales reveal that a variety of lightabsorbing impurities (LAIs) can be present on the surface, ranging from inorganic particulates to cryoconite materials and ice algae. Meanwhile, satellite observations show that the areal extent of dark ice has varied significantly between recent successive melt seasons. However, the processes that drive such large interannual variability in dark ice extent remain essentially unconstrained. At present we are therefore unable to project how the albedo of bare ice sectors of the GrIS will evolve in the future, causing uncertainty in the projected sea level contribution from the GrIS over the coming decades.Here we use MODIS satellite imagery to examine dark ice dynamics on the south-west GrIS each year from 2000 to 2016. We quantify dark ice in terms of its annual extent, duration, intensity and timing of first appearance. Not only does dark ice extent vary significantly between years but so too does its duration (from 0 to > 80 % of June-JulyAugust, JJA), intensity and the timing of its first appearance. Comparison of dark ice dynamics with potential meteorological drivers from the regional climate model MAR reveals that the JJA sensible heat flux, the number of positive minimum-air-temperature days and the timing of bare ice appearance are significant interannual synoptic controls.We use these findings to identify the surface processes which are most likely to explain recent dark ice dynamics. We suggest that whilst the spatial distribution of dark ice is best explained by outcropping of particulates from ablating ice, these particulates alone do not drive dark ice dynamics. Instead, they may enable the growth of pigmented ice algal assemblages which cause visible surface darkening, but only when the climatological prerequisites of liquid meltwater presence and sufficient photosynthetically active radiation fluxes are met. Further field studies are required to fully constrain the processes by which ice algae growth proceeds and the apparent dependency of algae growth on melt-out particulates.

Incorporating modelled subglacial hydrology into inversions for basal drag

“…The southern boundary is further south relative to Bougamont et al (2014) but north of the southern boundary in de Fleurian et al (2016). The eastern boundary was selected to extend up ice of the GPS stations (Tedstone and Neinow, 2017). The western boundary is the ice margin.…”

“…The end of the winter season is assumed to be day 120 of the year (30 April). Although the exact day is somewhat arbitrary, this day was selected as it is shortly before surface runoff begins in the study area and shortly before GPS records in the study site show enhanced motion (Tedstone and Neinow, 2017). (Fig.…”

Modelling seasonal meltwater forcing of the velocity of the Greenland Ice Sheet

“…These three years serve as analogues for summers with with average, elevated, and extreme melt intensity respectively (following ). (Tedstone and Neinow, 2017). Purple diamonds show the locations of automatic weather stations (van de Wal et al, 2015).…”

“…A key validation dataset in the Russell Glacier area is GPS surface velocity measurements for -2012(Tedstone and Neinow, 2017. A time series of hourly and daily averaged surface speeds are provided in the dataset.…”

“…The Russell Glacier area is selected as a study site to take advantage of the numerous observations available. These observations include radar flight lines constraining bed topography (Morlighem et al, 2015), meteorological data constraining climatic input (Noël et al, 2015), and GPS data (Tedstone and Neinow, 2017) which 5 provide a calibration and validation data set for model output.…”