New PROTECT publication!

The article “Improving surface melt estimation over the Antarctic Ice Sheet using deep learning: a proof of concept over the Larsen Ice Shelf” was published as a PROTECT publication.

Read here the main results:

  • Accurately estimating surface melt volume of the Antarctic Ice Sheet (AIS) is challenging and has hitherto relied on climate modelling, or on observations from satellite remote sensing. Each of these methods has its limitations, especially in regions with high surface melt.
  • Here we develop a framework correcting the model-observation mismatch of surface melt in the AIS with a deep learning model, which utilizes inputs from the physically based model, the regional atmospheric climate model version 2.3p2, and remote sensing albedo observations.
Temporal changes in the corrected regional atmospheric climate model version 2.3p2 (RACMO2) surface melt using the deep multilayer perceptron (MLP) model at automatic weather station (AWS) 18, (a) for all albedo differences (Δα) and (b) for a subset between November 2015 and April 2016 along with the augmented AWS observations for comparison.
  • As a result, our study demonstrates the opportunity to improve surface melt simulations using deep learning combined with satellite albedo observations.

Link to the paper:

Be the first to comment

Leave a Reply

Your email address will not be published.