The development of artificial intelligence makes it possible to rapidly segment landslides. However, there are still some challenges in landslide segmentation based on remote sensing images, such as low segmentation accuracy, caused by similar features, inhomogeneous features, and blurred boundaries. To address these issues, we propose a novel deep learning model called AST-UNet in this paper. This model is based on structure of SwinUNet, attaching a channel Attention and spatial intersection (CASI) module as a parallel branch of the encoder, and a spatial detail enhancement (SDE) module in the skip connection. Specifically, (1) the spatial intersection module expands the spatial attention range, alleviating noise in the image and enhances the continuity of landslides in segmentation results; (2) the channel attention module refines the spatial attention weights by feature modeling in the channel dimension, improving the model’s ability to differentiate targets that closely resemble landslides; and (3) the spatial detail enhancement module increases the accuracy for landslide boundaries by strengthening the attention of the decoder to detailed features. We use the landslide data from the area of Luding, Sichuan to conduct experiments. The comparative analyses with state-of-the-art (SOTA) models, including FCN, UNet, DeepLab V3+, TransFuse, TranUNet, and SwinUNet, prove the superiority of our AST-UNet for landslide segmentation. The generalization of our model is also verified in the experiments. The proposed AST-UNet obtains an F1-score of 90.14%, mIoU of 83.45%, foreground IoU of 70.81%, and Hausdorff distance of 3.73, respectively, on the experimental datasets.

The Himalayan, Karakoram, and Hindu Kush (HKH-TMHA) are the three main mountain ranges in the high-mountain Asia region, covering the China–Pakistan Economic Corridor (CPEC). In this study, we identified glacial lakes in the HKH-TMHA region based on multitemporal Landsat images taken from 1990 to 2020. We analyzed the spatial distribution and evolution of glacial lakes in the HKH-TMHA region from the perspective of their elevation, size, and terrain aspect; then, we described their temporal changes. The results showed that approximately 84.56% of the glacial lakes (84.1% of the total lake area) were located at elevations between 4000 m and 5500 m, and glacial lakes with areas ranging from 0.01–0.5 km² accounted for approximately 95.21% of the number and 63.01% of the total area of glacial lakes. The number (38.64%) and area (58.83%) of south-facing glacial lakes were largest in HKH-TMHA and expanded significantly over time. There were 5835 (664.84 ± 89.72 km²) glacial lakes in 1990; from 1990 to 2020, the number of glacial lakes in the HKH-TMHA region increased by 5974 (408.93 km²) in total; and the annual average increase in the area of glacial lakes reached 13.63 km² (11.15%). In 2020, the total number of glacial lake reached to 9673 (899.66 ± 120.63 km²). In addition, most glacial lakes were located in the Eastern Himalayan, China, and the Indus Basin. Based on the precipitation and temperature analyses performed in our study area, we found inconsistent climate characteristics and changes in the three mountain ranges. In general, the daily precipitation (temperature) increased by 1.0766 mm (1.0311 °C), 0.8544 mm (0.8346 °C), and 0.8245 mm (−0.1042 °C) on the yearly, summer, and winter scales, respectively. Glacial melting and climate change are common contributors to glacial lake expansion. The investigation of glacial lakes in this region can provide basic supporting data for research on glacial lake-related disasters, land cover, and climate change in the high-mountain Asia region.

Xie, Z.; Zhang, H.; Zhao, X.; Du, Z.; Xiang, L., and Wang, W., 0000. Assessment of heavy metal contamination and wetland management in a newly created coastal natural reserve, China. The investigation of the content and distribution of heavy metals in coastal soils is useful for improving our understanding of biogeochemical cycles and their potential feedback to global environmental change. The accumulation of heavy metals in the Yellow River Delta (YRD) was investigated using 43 sampling sites to determine the concentrations and vertical distribution of heavy metals. Factor analysis, contamination factors, and the geoaccumulation index were applied to evaluate the contamination condition. The contamination factors and geoaccumulation index clearly indicated that the coastal ecosystems are still in their pristine state with respect to metal pollution. Factor loadings reveal that the first principal component was strongly and positively related to arsenic, chromium, copper, nickel, lead, and zinc,...

The coal mining industry is developing automated and intelligent coal mining processes. Accurate determination of the geological conditions of working faces is an important prerequisite for automated mining. The use of machine learning to extract comprehensive attributes from seismic data and the application of that data to determine the coal strata thickness has become an important area of research in recent years. Conventional coal strata thickness interpretation methods do not meet the application requirements of mines. Determining the coal strata thickness with machine learning solves this problem to a large extent, especially for issues of exploration accuracy. In this study, we use seismic exploration data from the Xingdong coal mine, with the 1225 working face as the research object, and we apply seismic multiattribute machine learning to determine the coal strata thickness. First, through optimal selection, we perform seismic multiattribute extraction and optimal multiparameter selection by selecting the seismic attributes with good responses to the coal strata thickness and extracting training samples. Second, we optimise the model through a trial-and-error method and use machine learning for training. Finally, we illustrate the advantages of this method using actual data. We compare the results of the proposed model with results based on a single attribute, The results show that application of seismic multiattribute machine learning to determine coal strata thickness meets the requirements of geological inspection and has a good application performance and practical significance in complex areas.

Surface ice/snow is a vital resource and is sensitive to climate change in many parts of the world. The accurate and timely measurement of the spatial distribution of ice/snow is critical for managing water resources. Object-oriented and pixel-oriented methods often have some limitations due to the image segmentation scale, the determination of the optimal threshold and background heterogeneity. Therefore, this study proposes a method for automatically extracting large-scale surface ice/snow from Landsat series images, which takes advantage of the combination of image segmentation, the watershed algorithm and a series of ice/snow indices. We tested our novel method in three different regions in the Karakoram Mountains, and the experimental results show that the produced ice/snow map obtained a user’s accuracy greater than 90%, a producer’s accuracy greater than 97%, an overall accuracy greater than 98% and a kappa coefficient greater than 0.93. Comparing the extraction results under segmentation scales of 10, 15, 20 and 25, the user’s accuracy and producer’s accuracy from the proposed method are very similar, which indicates that the proposed method is more reliable and stable for extracting ice/snow objects than the object-oriented method. Due to the different reflectivity values in the near-infrared band in the snow and water categories, the normalized difference forest snow index (NDFSI) is suitable for Landsat TM and ETM+ images. This study can serve as a reliable, scientific reference for rapidly and accurately extracting ice/snow objects.

Collapse columns are complex geological anomalies that present safety hazards in coal mines. In practice, the boundaries and internal features are difficult to finely characterize. In this study, we use the Jeffrey divergence to reveal the spectral disparity when the channel wave seismic ray path encounters a collapse column. We use the forward mode and applications for verification. We find that the centroid peak frequency shifts toward lower frequencies, the frequency band narrows and the rays through a collapse column cause attenuation of energy. When a seismic ray path encounters a large collapse column, no channel wave appears in the seismic record, the dispersion curve velocity is greater than the velocity of a normal coal seam, the centroid peak frequency drops and the frequency band shrinks. When a seismic ray passes through the boundary of a collapse column, the dispersion curve shows two parallel velocity phases. When a seismic ray path passes through a small collapse column, the decrease in the centroid frequency becomes the only identifying feature. The proposed channel wave spectral disparity characteristic and velocity tomography imaging method can effectively identify small collapse columns and improve the accuracy and precision of the location of collapse column boundaries, with a position deviation of less than 10 m.