Articles | Volume 15, issue 12
The Cryosphere, 15, 5639–5658, 2021
https://doi.org/10.5194/tc-15-5639-2021
The Cryosphere, 15, 5639–5658, 2021
https://doi.org/10.5194/tc-15-5639-2021

Research article 13 Dec 2021

Research article | 13 Dec 2021

Improving surface melt estimation over the Antarctic Ice Sheet using deep learning: a proof of concept over the Larsen Ice Shelf

Zhongyang Hu et al.

Data sets

MCD43A3 MODIS/Terra+Aqua BRDF/Albedo Daily L3 Global - 500 m V006 C. Schaaf and Z. Wang https://doi.org/10.5067/MODIS/MCD43A3.006

MOD09GA MODIS/Terra Surface Reflectance Daily L2G Global 1 km and 500 m SIN Grid V006 E. Vermote and R. Wolfe https://doi.org/10.5067/MODIS/MOD09GA.006

High-resolution meteorological observations, Surface Energy Balance components and miscellaneous data from 10 AWS and one staffed station in Antarctica C. Jakobs, C. Reijmer, M. R. van den Broeke, P. Smeets, and G. König-Langlo https://doi.org/10.1594/PANGAEA.910473

Ice and Climate: Regional modelling IMAU https://www.projects.science.uu.nl/iceclimate/models/antarctica.php#2-1

Improving surface melt estimation over the Antarctic Ice Sheet using deep learning: a proof of concept over the Larsen Ice Shelf Z. Hu, P. Kuipers Munneke, S. Lhermitte, M. Izeboud, and M. van den Broeke https://doi.org/10.5281/zenodo.5769661

Download
Short summary
Antarctica is shrinking, and part of the mass loss is caused by higher temperatures leading to more snowmelt. We use computer models to estimate the amount of melt, but this can be inaccurate – specifically in the areas with the most melt. This is because the model cannot account for small, darker areas like rocks or darker ice. Thus, we trained a computer using artificial intelligence and satellite images that showed these darker areas. The model computed an improved estimate of melt.