Abstract
The continuous and precise mapping of glacier calving fronts is essential for monitoring and understanding rapid glacier changes in Antarctica and Greenland, which have the potential for significant sea level rise within the current century. This effort has been mostly restricted to the slow and painstaking manual digitalization of the calving front positions in thousands of satellite imagery products. Here, we have developed a machine learning toolkit to automatically detect glacier calving front margins in satellite imagery. The toolkit is based on semantic image segmentation using Convolutional Neural Networks (CNN) with a modified U-Net architecture to isolate the calving fronts from satellite images after having been trained with a dataset of images and their corresponding manually-determined calving fronts. As a case study we train our neural network on a varied set of Landsat images with lowered resolutions from Jakobshavn, Sverdrup, and Kangerlussuaq glaciers, Greenland and test the results on images from Helheim glacier, Greenland to evaluate the performance of the approach. The neural network is able to identify the calving front in new images with a mean deviation of 96.3 m from the true fronts, equivalent to 1.97 pixels on average, while the corresponding error for manually-determined fronts on the same resolution images is 92.5 m (1.89 pixels). We find that the trained neural network significantly outperforms common edge detection techniques, and can be used to continuously map out calving-ice fronts with a variety of data products.
Highlights
In recent decades, tidewater glaciers discharging ice from the Greenland Ice Sheet have been thinning, speeding up and retreating inland [1,2,3,4,5,6]
The U-Net architecture used by Ronneberger et al [25] used a training set that consisted of 30 512 × 512 pixel images
In order to test the ability of the neural network to predict calving fronts beyond the training set for different glacier geometries, we test the trained network on images of Helheim glacier, whose geometry is unknown to the Neural Network (NN) during training
Summary
Tidewater glaciers discharging ice from the Greenland Ice Sheet have been thinning, speeding up and retreating inland [1,2,3,4,5,6]. Because the manual ice front digitization process requires a considerable time investment, most current records of calving front retreat are limited to only a few ice front positions per glacier per year, if any. This shortage of data poses a challenge to seasonal analyses of calving glaciers (e.g., [8,9,10,11]), yet seasonal factors may be critical to understanding the pattern of long term retreat of Greenland’s glaciers [12] or to understand for instance the level above which a glacier may be pushed out of balance compared to its state of seasonal, natural variability [6]
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.