An official website of the United States government
Englacial drainages of supraglacial lakes on the Greenland Ice Sheet serve as a full thickness surface-to-bed meltwater delivery system. Downstream impacts of these meltwater transport events on ice stability are not well understood. Furthermore there does not yet exist comprehensive identification of these events, which limit inference and stability modeling. To close this gap, we have compiled this dataset of lake drainage events which comprises of labeled machine learning-ready imagery stacks for supraglacial lakes from the 2019 summer season in central west Greenland. Each lake is labeled by its drainage class: no drainage, englacial drainage, lateral drainage, or crevasse drainage. The raw labels are contained in the .csv file, and time series satellite (Copernicus Sentinel-2) imagery stacks for each lake in the .nc files. This dataset builds upon the foundational work of Dunmire et al., 2025 to enable deep learning computer vision model training for classification of lake drainages based on transient physical spatial features. Dataset DOI: doi.org/10.25740/sf350xp4038
more »
« less
Central West Greenland Supraglacial Lake Drainage Classification Dataset (2019 Summer)
Englacial drainages of supraglacial lakes on the Greenland Ice Sheet serve as a full thickness surface-to-bed meltwater delivery system. Downstream impacts of these meltwater transport events on ice stability are not well understood. Furthermore there does not yet exist comprehensive identification of these events, which limit inference and stability modeling. To close this gap, we have compiled this dataset of lake drainage events which comprises of labeled machine learning-ready imagery stacks for supraglacial lakes from the 2019 summer season in central west Greenland. Each lake is labeled by its drainage class: no drainage, englacial drainage, lateral drainage, or crevasse drainage. The raw labels are contained in the .csv file, and time series satellite (Copernicus Sentinel-2) imagery stacks for each lake in the .nc files. This dataset builds upon the foundational work of Dunmire et al., 2025 to enable deep learning computer vision model training for classification of lake drainages based on transient physical spatial features. Dataset DOI: doi.org/10.25740/sf350xp4038
more »
« less
- Award ID(s):
- 2344690
- PAR ID:
- 10669209
- Publisher / Repository:
- Stanford Digital Repository
- Date Published:
- Subject(s) / Keyword(s):
- Greenland Supraglacial lakes Deep learning Computer vision
- Format(s):
- Medium: X
- Institution:
- Stanford University
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Supraglacial lakes form on the surface of the Greenland Ice Sheet during the summer months and can directly impact ice sheet mass balance by removing mass via drainage and runoff or indirectly impact mass balance by influencing ice sheet dynamics. Here, we utilize the growing inventory of optical and microwave satellite imagery to automatically determine the fate of Greenland-wide supraglacial lakes during 2018 and 2019, a cool and warm melt season respectively. We use a machine learning time series classification approach to categorize lakes into four different categories: lakes that 1) refreeze, 2) rapidly drain, 3) slowly drain, and 4) become buried lakes at the end of the melt season. We find that during the warmer 2019, not only was the number of lake drainage events higher than in 2018, but also the proportion of lakes that drained was greater. By investigating mean lake depths for these four categories, we show that drained lakes were, on average, 22% deeper than lakes that refroze or became buried lakes. Interestingly, drained lakes had approximately the same maximum depth in 2018 and 2019; however, lakes that did not drain were 29% deeper in 2018, a cooler year. Our unique two-year dataset describing the fate of every Greenland supraglacial lake provides novel insight into lake drainage and refreeze in a relatively warm and cool year, which may be increasingly relevant in a warming climate.more » « less
-
Supraglacial lakes on the Greenland Ice Sheet drain through physically distinct pathways: hydrofracture, moulins, lateral stream routing, and crevasse-fields. Each drainage mechanism carries unique implications for ice sheet dynamics. Existing automated classifications reduce each lake's drainage behavior to a time-series of scalar values representing the observed water surface-area and classify each lake based on drainage rate (e.g., rapid vs. slow). This scalar reduction conflates physically different drainage mechanisms, which can only be determined through consideration of full spatio-temporal tracking. Here we introduce a human-benchmarked, machine learning-ready benchmark dataset that pairs full Sentinel-2 multispectral satellite imagery time series with human-expert-labels assigned for N=1679 supraglacial lakes in the central-west basin of the Greenland Ice Sheet during the 2018 (n=679) and 2019 (n=1000) melt seasons. The dataset is formatted as per-lake CF-1.8 NetCDF files each containing: six Sentinel-2 reflectance bands at 10 meter spatial resolution and daily cadence over the 153 day melt season (1 May to 30 September); a per-pixel binary cloud mask; co-registered lake water masks (both static and dynamic); and the human-assigned drainage classification labels. We accompany the dataset with a baseline deep learning classifier, demonstrating the utility of the dataset both in deep learning workflows and in extending lake drainage classification from rate-based to mechanism-based. The dataset is released through the Stanford Digital Repository under a CC BY 4.0 license, and the accompanying open-source sat-tile-stack preprocessing software under an MIT license.more » « less
-
Supraglacial lakes have been observed to drain within hours of each other, leading to the hypothesis that stress transmission following one drainage may be sufficient to induce hydro‐fracture‐driven drainages of other nearby lakes. However, available observations characterizing drainage‐induced stress perturbations have been insufficient to evaluate this hypothesis. Here, we use ice‐sheet surface‐displacement observations from a dense global positioning system array deployed in the Greenland Ice Sheet ablation zone to investigate elastic stress transmission between three neighboring supraglacial lake basins. We find that drainage of a central lake can place neighboring basins in either tensional or compressional stress relative to their hydro‐fracture scarp orientations, either promoting or inhibiting hydro‐fracture initiation beneath those lakes. For two lakes located within our array that drain close in time, we identify tensional surface stresses caused by ice‐sheet uplift due to basal‐cavity opening as the physical explanation for these lakes' temporally clustered hydro‐fracture‐driven drainages and frequent triggering behavior. However, lake‐drainage‐induced stresses in the up‐flowline direction remain low beyond the margins of the drained lakes. This short stress‐coupling length scale is consistent with idealized lake‐drainage scenarios for a range of lake volumes and ice‐sheet thicknesses. Thus, on elastic timescales, our observations and idealized‐model results support a stress‐transmission hypothesis for inducing hydro‐fracture‐driven drainage of lakes located within the region of basal cavity opening produced by the initial drainage, but refute this hypothesis for distal lakes.more » « less
-
Drainage of supraglacial lakes via hydro-fracture is widely argued to be a mechanism for destabilization of grounded ice sheets under climate-warming scenarios because it may accelerate surface meltwater access to the ice-sheet bed. Progress in interrogating this hypothesis has been hindered by the lack of regional observations of hydro-fracture event occurrence, and the lack of observations of regional ice-sheet response to hydro-fracture events. Here, we remedy both deficiencies using a 22-station Global Navigation Satellite System array to discern inter-lake, hydro-fracture-event triggering potential between lakes spanning the mid-to-upper Greenland Ice Sheet ablation zone. In four separate instances, multiple lake hydro-fracture events occur close in time at similar elevations; meanwhile, strain rates across higher-elevation lake basins are unperturbed. Our findings support a simple model for the inland progression of surface-to-bed meltwater pathways beneath lakes: pathway initiation migrates alongside advancing surface melt, but is not accelerated by drainage activity at distant, lower-elevation lakes.more » « less
