Melt pond detection with Sentinel-2 and Landsat-8 in the Beaufort Gyre using Google Earth Engine

Melt pond evolution in the Arctic is an important factor in the sea-ice albedo balance and subsequent polar climate system.The spectral signature of melt ponds on Arctic ice floes is distinguishable from surrounding sea ice features in optical satellite imagery.The melt pond fraction (MPF) can be derived using a support vector machine algorithm with a class accuracy of 96%.In this paper, we provide a JavaScript-Python workflow through Google Earth Engine and Collaboratory to extract and compare the MPF from the Sentinel-2 and Landsat-8 multi-spectral instruments.We then assess the results of the onset-melt conditions in 2017 and 2020 respectively.