A Transfer Learning Approach for Automatic Mapping of Retrogressive Thaw Slumps (RTSs) in the Western Canadian Arctic

Retrogressive thaw slumps (RTSs) are thermokarst landforms that develop on slopes in permafrost regions when thawing permafrost causes the land surface to collapse. RTSs are an indicator of climate change and pose a threat to human infrastructure and ecosystems in the affected areas. As the availabi...

Full description

Bibliographic Details
Main Author: Lin, Yiwen
Other Authors: Knudby, Anders
Format: Thesis
Language:English
Published: Université d'Ottawa / University of Ottawa 2022
Subjects:
Retrogressive Thaw Slumps
CNN
transfer learning
Canadian Arctic
Arctic
Banks Island
Climate change
permafrost
Thermokarst
Victoria Island
Online Access:http://hdl.handle.net/10393/44369
https://doi.org/10.20381/ruor-28580
Description
Summary:Retrogressive thaw slumps (RTSs) are thermokarst landforms that develop on slopes in permafrost regions when thawing permafrost causes the land surface to collapse. RTSs are an indicator of climate change and pose a threat to human infrastructure and ecosystems in the affected areas. As the availability of ready-to-use high-resolution satellite imagery increases, automatic RTS mapping is being explored with deep learning methods. We employed a pre-trained Mask-RCNN model to automatically map RTSs on Banks Island and Victoria Island in the western Canadian Arctic, where there is extensive RTS activity. We tested the model with different settings, including image band combinations, backbones, and backbone trainable layers, and performed hyper-parameter tuning and determined the optimal learning rate, momentum, and decay rate for each of the model settings. Our final model successfully mapped most of the RTSs in our test sites, with F1 scores ranging from 0.61 to 0.79. Our study demonstrates that transfer learning from a pre-trained Mask-RCNN model is an effective approach that has the potential to be applied for RTS mapping across the Canadian Arctic.

Similar Items