README2.md

Content structure

Inventory

subglacial_lake_event_inventory_37locations.gpkg: GeoPackage file in the NSIDC Sea Ice Polar Stereographic North projection (EPSG:3413) showing the locations of 37 active lakes. The attribute field no corresponds to the no. of the lake presented in the paper. (See the Conversion table below.)

Workflows

[LocationName]/analysis*.ipynb: Workflow for each [LocationName] (see conversion table below). These Jupyter notebooks start by downloading ArcticDEM and ICESat-2 data and end with the visualization of the lake elevation change. To execute analysis*.ipynb, one needs the following auxiliary files which are included in the data set:

• ArcticDEM_Strip_Index_s2s041_*.gpkg: Subset of the ArcticDEM strip index, containing the strip outlines within a single RGI region.
• RGI2000-v7.0-G-*.zip: RGI 7.0 glacier outlines in a single RGI region.
• [LocationName]/refgeo-[LocationName].gpkg: A rectangle extent at the lake area, defining where to process the data.
• [LocationName]/[LocationName]-off-ice-exclusion.gpkg: Polygons that indicate non-stable ground. ICESat-2 points falling within the polygons are not used for aligning the DEMs.

Conversion table between location name and lake number

Lake number	Location name
1	MilneTerminus
2	Milne
3-a	MilneCentral (left)
3-b	MilneCentral (top)
3-c	MilneCentral (right)
4-a	OttoN (bottom)
4-b	OttoN (top)
5	IcebergE
6	AgassizSWMinor1
7	TaggartLake
8	MittieNE
9	YelvertonNW
10	UnnamedFW
11	UnnamedHSE
12	Antoinette
13	AgassizSWMinor2
14	MittieMain
15	ClydeRiverSW
16	PennyN
17	PennyW
18	SvenHedin
19	MittieMinor
20	OttoNW
21	Otto
22	Disraeli
23	UnnamedF
24	Airdrop
25	LakeToubergE
26	GoodFriday (top)
27	GoodFriday (bottom)
28	Dlberville
29	AgassizSWMain (top)
30	AgassizSWMain (bottom)
31	JolliffeNArm
32	MittieNW
33	BylotNE
34	ClydeRiverW

Results

1. [LocationName]/[LocationName]_15m_hydroevent-evmd-bitmask_[Suffix].tif: Main results for each glacier. Different suffixes mean:

   • slope: dh/dt of the linear model; $a_1$ in Eq. (1)
   • duration: total length of $t$ of the linear model
   • dh: maximum short-term dh of the sigmoid model; see Eq. (4)
   • t: timing of the maximum short-term dh of sigmoid model
   • gp-dh: maximum short-term dh of the GP model
   • gp-t: timing of the maximum short-term dh of the GP model

2. [LocationName]/[LocationName]_15m_hydroevent-evmd-bitmask_viz*.png: Visualization of the results.

3. [LocationName]/[LocationName]_15m_median_topo.tif Median topography of all the available DEM strips.

4. [LocationName]/deminput_processed.csv: List of the ArcticDEM strip DEMs used in the analysis. The third column represents the 1-sigma uncertainty in m.

5. [LocationName]/ref_elevs.csv: reference elevations extracted from the ICESat-2 ATL-6 data; for DEm alignment.

6. [LocationName]/refgeo-[LocationName]_15m.tif: Reference grid geometry. All the resulting geotiffs have the same image extent and size as this image.