Ayobami Ogunmolasuyi , Colin R. Meyer, Ian Mcdowell, Megan Thompson-Munson, Ian Baker

Understanding firn densification is essential for interpreting ice core
records, predicting ice sheet mass balance, elevation changes, and future
sea-level rise. Current models of firn densification on the Antarctic Ice
Sheet (AIS) are semi-empirical, complex, and rely on sparse climatic data
and surface density observations. In this work, we introduce a deep
learning technique to study firn densification on the AIS. Our model,
evaluated on six density cores, shows an average root mean square error
(RMSE) of 39 kg m-3 and explains 98% of the variance (r2 = 0.98). We
use the model to generate surface density and the depths to the 550 kg
m -3 and 830 kg m -3 density horizons across the AIS to assess spatial
variability. Comparisons with observations and the Herron and Langway
(1980) model at six locations with different climate conditions
demonstrate that FirnLearn more accurately predicts density profiles in the second stage of densification and complete density profiles without
direct surface density observations. This work establishes deep learning
as a promising tool for understanding firn processes and advancing a
more universally applicable firn model.