Exact information on the calving front positions of marine- or lake-terminating glaciers is a fundamental glacier variable for analyzing ongoing glacier change processes and assessing other variables like frontal ablation rates. In recent years, researchers started implementing algorithms that can automatically detect the calving fronts on satellite imagery. Most studies use optical images, as calving fronts are often easy to distinguish in these images due to the sufficient spatial resolution and the presence of different spectral bands, allowing the separation of ice features. However, detecting calving fronts on synthetic aperture radar (SAR) images is highly desirable, as SAR images can also be acquired during the polar night and are independent of weather conditions (e.g., cloud cover), facilitating year-round monitoring worldwide. In this paper, we present a benchmark dataset (Gourmelon et al., 2022 b) of SAR images from multiple regions of the globe with corresponding manually defined labels providing information on the position of the calving front (https://doi.org/10.1594/PANGAEA.940950). With this dataset, different approaches for the detection of glacier calving fronts can be implemented, tested, and their performance fairly compared so that the most effective approach can be determined. The dataset consists of 681 samples, making it large enough to train deep learning segmentation models. It is the first dataset to provide long-term glacier calving front information from multi-mission data. As the dataset includes glaciers from Antarctica, Greenland, and Alaska, the wide applicability of models trained and tested on this dataset is ensured. The test set is independent of the training set so that the generalization capabilities of the models can be evaluated. We provide two sets of labels: one binary segmentation label to discern the calving front from the background, and one label for multi-class segmentation of different landscape classes. Unlike other calving front datasets, the presented dataset contains not only the labels but also the corresponding preprocessed and geo-referenced SAR images as PNG files. The ease of access to the dataset will allow scientists from other fields, such as data science, to contribute their expertise. With this benchmark dataset, we enable comparability between different front detection algorithms and improve the reproducibility of front detection studies. Moreover, we present one baseline model for each kind of label type. Both models are based on the U-Net, one of the most popular deep learning segmentation architectures. In the following two post-processing procedures, the segmentation results are converted into 1-pixel-wide front delineations. By providing both types of labels, both approaches can be used to address the problem. To assess the performance of different models, we suggest first reviewing the segmentation results using the recall, precision, F₁ score, and the Jaccard index. Second, the front delineation can be evaluated by calculating the mean distance error to the labeled front. The presented vanilla models provide a baseline of 150 m ± 24 m mean distance error for the Mapple Glacier in Antarctica and 840 m ± 84 m for the Columbia Glacier in Alaska, which has a more complex calving front, consisting of multiple sections, compared with a laterally well constrained, single calving front of Mapple Glacier.

Glaciers across the globe react to the changing climate. Monitoring the transformation of glaciers is essential for projecting their contribution to global mean sea level rise. The delineation of glacier-calving fronts is an important part of the satellite-based monitoring process. This work presents a calving-front extraction method based on the deep learning framework nnU-Net, which stands for no new U-Net. The framework automates the training of a popular neural network, called U-Net, designed for segmentation tasks. Our presented method marks the calving front in synthetic aperture radar (SAR) images of glaciers. The images are taken by six different sensor systems. A benchmark dataset for calving-front extraction is used for training and evaluation. The dataset contains two labels for each image. One label denotes a classic image segmentation into different zones (glacier, ocean, rock, and no information available). The other label marks the edge between the glacier and the ocean, i.e., the calving front. In this work, the nnU-Net is modified to predict both labels simultaneously. In the field of machine learning, the prediction of multiple labels is referred to as multi-task learning (MTL). The resulting predictions of both labels benefit from simultaneous optimization. For further testing of the capabilities of MTL, two different network architectures are compared, and an additional task, the segmentation of the glacier outline, is added to the training. In the end, we show that fusing the label of the calving front and the zone label is the most efficient way to optimize both tasks with no significant accuracy reduction compared to the MTL neural-network architectures. The automatic detection of the calving front with an nnU-Net trained on fused labels improves from the baseline mean distance error (MDE) of 753±76 to 541±84 m. The scripts for our experiments are published on GitHub (https://github.com/ho11laqe/nnUNet_calvingfront_detection, last access: 20 November 2023). An easy-access version is published on Hugging Face (https://huggingface.co/spaces/ho11laqe/nnUNet_calvingfront_detection, last access: 20 November 2023).

Knowledge on changes in glacier calving front positions is important for assessing the status of glaciers. Remote sensing imagery provides the ideal database for monitoring calving front positions, however, it is not feasible to perform this task manually for all calving glaciers globally due to time-constraints. Deep learning-based methods have shown great potential for glacier calving front delineation from optical and radar satellite imagery. The calving front is represented as a single thin line between the ocean and the glacier, which makes the task vulnerable to inaccurate predictions. The limited availability of annotated glacier imagery leads to a lack of data diversity (not all possible combinations of different weather conditions, terminus shapes, sensors, etc. are present in the data), which exacerbates the difficulty of accurate segmentation. In this paper, we propose Attention-Multi-hooking-Deep-supervision HookNet (AMD-HookNet), a novel glacier calving front segmentation framework for synthetic aperture radar (SAR) images. The proposed method aims to enhance the feature representation capability through multiple information interactions between low-resolution and high-resolution inputs based on a two-branch U-Net. The attention mechanism, integrated into the two branch U-Net, aims to interact between the corresponding coarse and fine-grained feature maps. This allows the network to automatically adjust feature relationships, resulting in accurate pixel-classification predictions. Extensive experiments and comparisons on the challenging glacier segmentation benchmark dataset CaFFe show that our AMD-HookNet achieves a mean distance error of 438 m to the ground truth outperforming the current state of the art by 42 %, which validates its effectiveness.

Position changes of glacier calving fronts are important indicators for evaluating the health of ice sheet outlet glaciers and changes in ice dynamics. However, manual delineation of calving fronts in remote sensing imagery is a time-consuming task, resulting in potential large costs. Deep learning-based methods have made remarkable progress in automatically segmenting and delineating glacier calving fronts from remote sensing imagery. The relatively few remote sensing images and the limited geometric changes for glacier observations both reduce the diversity of the data and exacerbate the difficulty of accurate segmentation. Here we describe a novel automatic method for detecting glacier calving fronts in synthetic aperture radar (SAR) images, termed HookFormer. Our approach processes high-resolution (target) and low-resolution (context) inputs with a unified Transformer architecture. The global-local tokens from the context and the target branches are integrated purely by the proposed cross-attention mechanism and cross-interaction module to complement and enhance each other. Moreover, we redesign the HookFormer architecture based on the CNN model AMD-HookNet aiming to improve computational efficiency while achieving significant performance gains with only half of the model parameters/FLOPs. We conduct an in-depth analysis and make extensive comparisons based on the challenging glacier segmentation benchmark dataset CaFFe. As the first pure Transformer approach, HookFormer sets a new state of the art with a mean distance error of 353 m to the ground truth, outperforming the baseline, Swin-Unet, and AMD-HookNet by 53 %, 39 %, and 19 %, respectively.

With an increasing need for secured water supply, a better understanding of the water consumption behavior is beneficial. This can be achieved through end-use classification, i.e., identifying end-uses such as toilets, showers or dishwashers from water consumption data. Previously, both supervised and unsupervised machine learning (ML) techniques are employed, demonstrating accurate classification results on particular datasets. However, a comprehensive comparison of ML techniques on a common dataset is still missing. Hence, in this study, we are aiming at a quantitative evaluation of various ML techniques on a common dataset. For this purpose, a stochastic water consumption simulation tool with high capability to model the real-world water consumption pattern is applied to generate residential data. Subsequently, unsupervised clustering methods, such as dynamic time warping, k-means, DBSCAN, OPTICS and Hough transform, are compared to supervised methods based on SVM. The quantitative results demonstrate that supervised approaches are capable to classify common residential end-uses (toilet, shower, faucet, dishwasher, washing machine, bathtub and mixed water-uses) with accuracies up to 0.99, whereas unsupervised methods fail to detect those consumption categories. In conclusion, clustering techniques alone are not suitable to separate end-use categories fully automatically. Hence, accurate labels are essential for the end-use classification of water events, where crowdsourcing and citizen science approaches pose feasible solutions for this purpose.

The measurement of ice thickness is of great importance for the accurate estimation of glacier volume and the delineation of bedrock topography. In particular, this is a crucial factor in forecasting the future evolution of glaciers in the context of a changing climate. In order to derive the ice thickness, the travel time of electromagnetic waves in radargrams acquired by radio-echo sounding (RES) systems is analyzed. This can only be achieved by identifying the ice surface and underlying ice bottom in corresponding radargrams. Manually identifying these two reflection horizons in RES data is a laborious and time-consuming process. Consequently, scientists are attempting to automate this task through the use of techniques such as deep learning. Such automation can significantly reduce the time between a field campaign and the calculation of the glacier's ice thickness distribution. In this paper, we present the first benchmark dataset for delineating the ice surface and bottom boundaries in RES data to facilitate standardized comparisons of deep learning models in the future. The “IceAnatomy” dataset comprises radargrams and the corresponding manual picks, amounting to a total of over 45 000 km of observations. The RES data originate from three sources: FAU (Friedrich-Alexander-Universität Erlangen-Nürnberg, Institute of Geography), CReSIS (Center for Remote Sensing and Integrated Systems), and AWI (Alfred Wegener Institute, Helmholtz Centre for Polar and Marine Research). The dataset comprises different RES systems as well as different pre-processing methods. In addition, the data were acquired over a large range of geographical and glaciological settings, featuring different thermal regimes present in Antarctica and the Southern Patagonian Ice Field. This diversity ensures that the models' behaviors can be analyzed in different scenarios. We define a standardized train–test split for each source in the dataset. This allows us to introduce not only a baseline model trained on the entire training set (the “omni”-model), but also three source-specific baseline models. The source-specific models are trained exclusively on the subset of the training data acquired by the specified source. The baseline models provide an initial benchmark against which subsequent models can be compared. The source-specific models demonstrate more accurate results than the omni-model. For the FAU, CReSIS, and AWI test sets, the source-specific models achieve mean meter errors of 2.1, 23.1, and 4.9 m for the ice surface and 9.1, 78.2, and 29.3 m for the ice bottom. In relation to the mean measured ice thickness of the test set, these errors equate to 1.2 %, 3.1 %, and 0.3 % for the ice surface and 4.9 %, 10.4 %, and 1.5 % for the ice bottom. The dataset and implementation are available at https://doi.org/10.5281/zenodo.14036897 (Dreier et al., 2024) and https://doi.org/10.5281/zenodo.14038570 (Dreier, 2024).