.png){kind=icon}
.png){kind=icon}
Return
Predicting Délı̨nę ice road surface temperatures on Great Bear Lake in the Northwest Territories, Canada using long short-term memory networks and their relationship to ice melt processes
.png){kind=icon}
PRE
AI
I
M
H
DOI：10.1016/j.coldregions.2026.104963.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Long Short-Term Memory (LSTM) Recurrent Neural Network (RNN) models were developed to predict surface temperatures on an ice road in the Northwest Territories (NWT), Canada. • The mean absolute errors (MAEs) in the predictions of surface temperatures at two- and seven-day lead times versus the observed were around 4.8 °C for the test sets during the seasonal ice road opening and closure periods. • The predicted surface temperatures are linked to the melt and breakup of the ice on a river adjacent to the ice road and along the ice road itself. • The models presented in this paper could be used to enhance safer ice road operations by providing predictions of ice road surface quality degradation and melt onset and providing critical input to predictions of ice road flexural strength, bearing capacity, and elastic modulus.
Keywords:
LSTM
ice road
surface temperature prediction
ice melt
neural networks
Journal
IF:
3.8
Papers:
3.6K
Citations:
1.2W
Organization
Cited Papers