With the ongoing effects of global warming, Antarctic supraglacial meltwater plays a critical role in ice-sheet instability. Spaceborne remote sensing provides a viable solution to detect surface melt in these vast and remote regions. Innovative deep learning image segmentation methods have the potential to improve supraglacial lake (SGL) mapping by adding spatial awareness to the mapping process, as currently established methods only consider the spectral variety of multispectral satellite data (Sentinel-1 and Landsat-8) on a per-pixel basis. In this study, we adapt a U-net image segmentation method to detect SGLs in Sentinel-2 data and quantify its performance in comparison to established mapping approaches in Antarctica. At a 90% probability threshold, the proposed workflow achieves a producer’s accuracy, user’s accuracy, and an F1 score of 0.915, 0.912, and 0.913, respectively. We compare the classification results to those of an established Random Forest classifier, and find that the U-net achieves better results in 13 of the 16 validation scenes. In a relative comparison with five established SGL products, the U-net approach maps, on average, the largest lake area, although there is disagreement between the datasets due to different data sources, cloud cover, and misclassification of some clouds as lakes. The main advantages of the developed approach are its capabilities to map lakes under thin clouds and ice, resulting in more continuous lake areas than conventional approaches, and a reduced number of input features required, which will be advantageous in future large-scale SGL mapping efforts.

Marine heat waves (MHWs) are prolonged periods of elevated ocean temperatures that can devastate marine ecosystems, fisheries, and coastal communities. Skillfully predicting these events with sufficient lead time is crucial for mitigating their adverse effects. This study presents a probabilistic subseasonal MHW forecast tool using a U-Net-based neural network architecture, with a focus on the Northern Indian Ocean and the Arabian Sea. The model was trained using sea surface temperature and sea surface height reanalysis data. The U-Net-based forecast tool demonstrated significant predictive skill up to 10 weeks in advance across various deterministic and probabilistic skill metrics. The model outperformed persistence and climatology-based benchmarks, especially in the tropical warm pool. Future applications of explainable artificial intelligence (XAI) methods have the potential to identify the sources of predictive skill, inform understanding of underlying dynamics, and improve dynamic subseasonal to seasonal forecast models.

The logistics, costs, and capacity needed to complete extensive archaeological pedestrian surveys to inventory cultural resources present challenges to public land managers. To address these issues, we developed a workflow combining lidar-derived imagery and deep learning (DL) models tailored for cultural resource management (CRM) programs on public lands. It combines Python scripts that fine-tune models to recognize archaeological features in lidar-derived imagery with denoising QGIS steps that improve the predictions’ performance and applicability. We present this workflow through an applied case study focused on detecting historic agricultural terraces in the Piedmont National Wildlife Refuge, Georgia, USA. For this project, we fine-tuned pretrained U-Net models to teach them to recognize agricultural terraces in imagery, identified the parameter settings that led to the highest recall for detecting terraces, and used those settings to train models on incremental dataset sizes, which allowed us to identify the minimum training size necessary to obtain satisfying models. Results present effective models that can detect most terraces even when trained on small datasets. This study provides a robust methodology that requires basic proficiencies in Python coding but expands DL applications in federal CRM by advancing methods in lidar and machine learning for archaeological inventorying, monitoring, and preservation.

In this paper, an improved U-net welding engineering drawing segmentation model is proposed for the automatic segmentation and extraction of sheet metal engineering drawings in the process of mechanical manufacturing, to improve the cutting efficiency of sheet metal parts. To construct a high-precision segmentation model for sheet metal engineering drawings, this paper proposes a U-net jump structure with an attention mechanism based on the Convolutional Attention Module (CBAM) attention mechanism. At the same time, this paper also designs an encoder jump structure with vertical double pooling convolution, which fuses the features after maximum pooling+convolution of the high-dimensional encoder with the features after average pooling+convolution of the low-dimensional encoder. The method in this paper not only improves the global semantic feature extraction ability of the model but also reduces the dimensionality difference between the low-dimensional encoder and the high-dimensional decoder. Using Vgg16 as the backbone network, experiments verify that the IoU, mAP, and Accu indices of this paper’s method in the welding engineering drawing dataset segmentation task are 84.72%, 86.84%, and 99.42%, respectively, which are 22.10, 19.09 and 0.05 percentage points higher compared to the traditional U-net model, and it has a relatively excellent value in engineering applications.