README file for 'Automated mapping of Antarctic surface meltwater' This code relates to the following paper, which describes the methodology in detail: Tuckett, P. A., Ely, J. C., Sole, A. J., Lea, J. M., Livingstone, S. J., Jones, J. M., & van Wessem, J. M. (2021). Automated mapping of the seasonal evolution of surface meltwater and its links to climate on the Amery Ice Shelf, Antarctica. The Cryosphere. 1-35. Please contact Peter Tuckett (University of Sheffield) for more information - patuckett1@sheffield.ac.uk Software Requirements: Google Earth Engine (requires a user account) - copy the contents of each .txt file into a new GEE script (https://code.earthengine.google.com), and save Matlab FILE DETAILS: Create_cloud_free_ice_mask.txt - Run this script first to create the cloud-free ice mask which is required during the lake mapping script - Requires an asset (shapefile) to define the area over which to create the ice mask (same as used for mapping script) - Run the script in Google Earth Engine - Exports the ice mask to your assets folder within Earth Engine Amery_GEE_lake_mapping.txt - Lake mapping script to run in Google Earth Engine - Ensure that 'icemaskPath' is set to match the location of the ice mask within assets - Specify the date range and temporal resolution (time window length) under 'SPECIFY CONFIG VARIABLES' - The script maps surface water for the given region and date range, and attaches visibility metadata - Exports a GEOJSON file to google drive - set the output location and file names at the end of the script - Download the GEOJSON file(s) from Google Drive, ready to be used in the Matlab post-processing script Post_processing_antarctic_lakes.m - Post-processing script to be run in Matlab once GEOJSON outputs have been created from GEE - Requires some functions that may need to be downloaded from Mathworks - File pathways/directory structure will likely need editing by the user - The script is setup to create three versions of shapefile outputs: Output folder names follow a similar format to the geojson files. For each year, three output folders are created, based on the three stages of the Matlab script: 1) Individual_tiles (raw shapefiles for each individual tile, per time window) 2) Combined (Merges tiles, to create a single shapefile per time window) 3) Filtered_shapefiles (Lake polygons are unioned over tile boundaries, lake centroids are recalculated, an area threshold is applied and metadata is assigned, including visibility scores)