CALFIN: Calving front dataset for East/West Greenland, 1972-2019 ...

• Main Authors: Cheng, Daniel, Hayes, Wayne, Larour, Eric
• Format: Dataset
• Language: English
• Published: Dryad 2020
• Subjects: calving front; glacier; Shapefile; Neural networks; FOS Earth and related environmental sciences; Greenland; Hayes; Kangerlussuaq; greenlandic; Jakobshavn; Rink Isbrae; Upernavik
• Online Access: https://dx.doi.org/10.7280/d1fh5d; https://datadryad.org/stash/dataset/doi:10.7280/D1FH5D

We present Calving Front Machine (CALFIN), an automated method for extracting calving fronts from satellite images of marine-terminating glaciers. The results use Landsat imagery from 1972 to 2019 to generate 22,678 calving front lines across 66 Greenlandic glaciers. The method uses deep learning, and builds on existing work by Mohajerani et al., Zhang et al., and Baumhoer et al. Additional post-processing techniques allow for accurate segmentation of imagery into Shapefile outputs. This method is uniquely robust to the impact of clouds, illumination differences, ice mélange, and Landsat-7 Scan Line Corrector errors. CALFIN provides improvements on the current state of the art. A model inter-comparison is performed to evaluate performance against existing methodologies. CALFIN's ability to generalize to SAR imagery is also evaluated. CALFIN's fronts are often indistinguishable from manually-curated fronts, deviating by 2.25 pixels (86.76 meters) from the true front on a diverse set of 162 testing images. The ... : We collect our source data from Landsat NIR band images spanning from 1972 to 2019. We provide 1997 manually-masked calving fronts, which we use for training, validating, and testing our automated algorithm. We also provide over 17912 automatically generated fronts, along with estimated mean errors calculated for each basin. This dataset contains a total of 19909 calving fronts for Helheim, Kangerlussuaq, Kong Oscar, Hayes, Rink Isbrae, Upernavik, Jakobshavn, Petermann, Kangiata Nunaata, and 62 other nearby glaciers along East/West Greenland. We process our source data by utilizing deep learning, in the form of the Google DeeplabV3+ Xception derived CALFIN Neural Network. Additional post-processing techniques allow our method to achieve accurate and useful segmentation of raw Landsat subsets into Shapefile outputs. ...