Search

Coastal land cover classification is a significant yet challenging task in remote sensing because of the complex and fragmented nature of coastal landscapes. However, availability of multitemporal and multisensor remote sensing data provides opportunities to improve classification accuracy. Meanwhile, rapid development of deep learning has achieved astonishing results in computer vision tasks and has also been a popular topic in the field of remote sensing. Nevertheless, designing an effective and concise deep learning model for coastal land cover classification remains problematic. To tackle this issue, we propose a multibranch convolutional neural network (MBCNN) for the fusion of multitemporal and multisensor Sentinel data to improve coastal land cover classification accuracy. The proposed model leverages a series of deformable convolutional neural networks to extract representative features from a single-source dataset. Extracted features are aggregated through an adaptive feature fusion module to predict final land cover categories. Experimental results indicate that the proposed MBCNN shows good performance, with an overall accuracy of 93.78% and a Kappa coefficient of 0.9297. Inclusion of multitemporal data improves accuracy by an average of 6.85%, while multisensor data contributes to 3.24% of accuracy increase. Additionally, the featured fusion module in this study also increases accuracy by about 2% when compared with the feature-stacking method. Results demonstrate that the proposed method can effectively mine and fuse multitemporal and multisource Sentinel data, which improves coastal land cover classification accuracy.

Although metallic mineral resources are most important in the economy of Mongolia, mining activities with improper management may result in the pollution of stream waters, posing a threat to aquatic ecosystems and humans. In this study, aiming to evaluate potential impacts of metallic mining activities on the quality of a transboundary river (Selenge) in central northern Mongolia, we performed hydrochemical investigations of rivers (Tuul, Khangal, Orkhon, Haraa, and Selenge). Hydrochemical analysis of river waters indicates that, while major dissolved ions originate from natural weathering (especially, dissolution of carbonate minerals) within watersheds, they are also influenced by mining activities. The water quality problem arising from very high turbidity is one of the major environmental concerns and is caused by suspended particles (mainly, sediment and soil particles) from diverse erosion processes, including erosion of river banks along the meandering river system, erosion of soils owing to overgrazing by livestock, and erosion by human activities, such as mining and agriculture. In particular, after passing through the Zaamar gold mining area, due to the disturbance of sediments and soils by placer gold mining, the Tuul River water becomes very turbid (up to 742 Nephelometric Turbidity Unit (NTU)). The Zaamar area is also the contamination source of the Tuul and Orkhon rivers by Al, Fe, and Mn, especially during the mining season. The hydrochemistry of the Khangal River is influenced by heavy metal (especially, Mn, Al, Cd, and As)-loaded mine drainage that originates from a huge tailing dam of the Erdenet porphyry Cu-Mo mine, as evidenced by δS values of dissolved sulfate (0.2 to 3.8 ‰). These two contaminated rivers (Tuul and Khangal) merge into the Orkhon River that flows to the Selenge River near the boundary between Mongolia and Russia and then eventually flows into Lake Baikal. Because water quality problems due to mining can be critical, mining activities in central northern Mongolia should be carefully managed to minimize the transboundary movement of aquatic contaminants (in particular, turbidity, dissolved organic carbon, Fe and Al) from mining activities. [ABSTRACT FROM AUTHOR]