Search for: All records

The objectives of this paper are to investigate the trade-offs between a physically constrained neural network and a deep, convolutional neural network and to design a combined ML approach (“VarioCNN”). Our solution is provided in the framework of a cyberinfrastructure that includes a newly designed ML software, GEOCLASS-image (v1.0), modern high-resolution satellite image data sets (Maxar WorldView data), and instructions/descriptions that may facilitate solving similar spatial classification problems. Combining the advantages of the physically-driven connectionist-geostatistical classification method with those of an efficient CNN, VarioCNN provides a means for rapid and efficient extraction of complex geophysical information from submeter resolution satellite imagery. A retraining loop overcomes the difficulties of creating a labeled training data set. Computational analyses and developments are centered on a specific, but generalizable, geophysical problem: The classification of crevasse types that form during the surge of a glacier system. A surge is a glacial catastrophe, an acceleration of a glacier to typically 100–200 times its normal velocity. GEOCLASS-image is applied to study the current (2016-2024) surge in the Negribreen Glacier System, Svalbard. The geophysical result is a description of the structural evolution and expansion of the surge, based on crevasse types that capture ice deformation in six simplified classes. 

As climate warms and the transition from a perennial to a seasonal Arctic sea-ice cover is imminent, understanding melt ponding is central to understanding changes in the new Arctic. National Aeronautics and Space Administration (NASA)’s Ice, Cloud and land Elevation Satellite (ICESat-2) has the capacity to provide measurements and monitoring of the onset of melt in the Arctic and on melt progression. Yet ponds are currently not identified on the ICESat-2 standard sea-ice products, in which only a single surface is determined. The objective of this article is to introduce a mathematical algorithm that facilitates automated detection of melt ponds in the ICESat-2 Advanced Topographic Laser Altimeter System (ATLAS) data, retrieval of two surface heights, pond surface and bottom, and measurements of depth and width of melt ponds. With ATLAS, ICESat-2 carries the first spaceborne multibeam micropulse photon-counting laser altimeter system, operating at 532-nm frequency. ATLAS data are recorded as clouds of discrete photon points. The Density-Dimension Algorithm for bifurcating sea-ice reflectors (DDA-bifurcate-seaice) is an autoadaptive algorithm that solves the problem of pond detection near the 0.7-m nominal along-track spacing of ATLAS data, utilizing the radial basis function for calculation of a density field and a threshold function that automatically adapts to changes in the background, apparent surface reflectance, and some instrument effects. The DDA-bifurcate-seaice is applied to large ICESat-2 datasets from the 2019 and 2020 melt seasons in the multiyear Arctic sea-ice region. Results are evaluated by comparison with those from a manually forced algorithm.