Due to a lack of data and practical models, few studies have extracted tailings pond margins in large areas. In addition, there is no public dataset of tailings ponds available for relevant research. This study proposed a new deep learning-based framework for extracting tailings pond margins from high spatial resolution (HSR) remote sensing images by combining You Only Look Once (YOLO) v4 and the random forest algorithm. At the same time, we created an open source tailings pond dataset based on HSR remote sensing images. Taking Tongling city as the study area, the proposed model can detect tailings pond locations with high accuracy and efficiency from a large HSR remote sensing image (precision = 99.6%, recall = 89.9%, mean average precision = 89.7%). An optimal random forest model and morphological processing were utilized to further extract accurate tailings pond margins from the target areas. The final map of the entire study area was obtained with high accuracy. Compared with the random forest algorithm, the total extraction time was reduced by nearly 99%. This study can be beneficial to mine monitoring and ecological environmental governance.

针对华北地区尾矿库自动提取问题，将基于深度学习的SSD目标检测模型应用于遥感图像尾矿库提取。首先标记华北地区2 000个样本，随机挑选1 500个作为训练样本，剩余样本作为测试样本，验证模型的检测精度。分析卷积层对应感受野与图像中尾矿库尺寸关系，发现原始SSD模型漏检误检大型尾矿库。改进SSD模型结构，提出增加额外卷积层的策略，提高对大型尾矿库目标的检测精度。实验表明，在置信度阈值为0.3时，改进的SSD模型相比原始模型，检测精确率提高10.0%，召回率提高14.4%，提高了大型尾矿库检测精度。验证了基于深度学习的SSD目标检测模型自动提取尾矿库的可行性以及改进算法的有效性。

In order to automatically extract tailing ponds in North China, the SSD target detection model based on deep learning is applied to extract tailing ponds from remote sensing images. Firstly, 2 000 samples in North China were labeled as the foundation of database. 1 500 samples were randomly selected as training samples, and the remaining samples were used as test samples to verify the detection accuracy of the model. By using the original SSD model, the targets of large tailing ponds can not be detected accurately. In this work the relationship between the corresponding receptive field of convolution layer and the size of tailing reservoir in image was analyzed. Moreover, in order to improve the detection accuracy of the model for large-scale tailing pond targets, we modified the structure of SSD model by introducing an extra convolution layer. Experiments show that, compared with the original SSD model, the modified SSD model improves the detection accuracy by 10% and the recall rate by 14.4% at 0.3 confidence level. The detection accuracy for large tailing ponds is also improved. In this work, the feasibility of automatic extraction of tailing reservoir based on deep learning SSD model and the effectiveness of the modified algorithm were verified.

The timely and accurate mapping and monitoring of mine tailings dams is crucial to the improvement of management practices by decision makers and to the prevention of disasters caused by failures of these dams. Due to the complex topography, varying geomorphological characteristics, and the diversity of ore types and mining activities, as well as the range of scales and production processes involved, as they appear in remote sensing imagery, tailings dams vary in terms of their scale, color, shape, and surrounding background. The application of high-resolution satellite imagery for automatic detection of tailings dams at large spatial scales has been barely reported. In this study, a target detection method based on deep learning was developed for identifying the locations of tailings ponds and obtaining their geographical distribution from high-resolution satellite imagery automatically. Training samples were produced based on the characteristics of tailings ponds in satellite images. According to the sample characteristics, the Single Shot Multibox Detector (SSD) model was fine-tuned during model training. The results showed that a detection accuracy of 90.2% and a recall rate of 88.7% could be obtained. Based on the optimized SSD model, 2221 tailing ponds were extracted from Gaofen-1 high resolution imagery in the Jing–Jin–Ji region in northern China. In this region, the majority of tailings ponds are located at high altitudes in remote mountainous areas. At the city level, the tailings ponds were found to be located mainly in Chengde, Tangshan, and Zhangjiakou. The results prove that the deep learning method is very effective at detecting complex land-cover features from remote sensing images.

Tailings pond engineering is a complex and extensive system with many risk factors that can trigger a dam failure. It is important to clarify the evolutionary relationships among the factors and to enhance effective management to reduce the risk of dam failure. In this paper, an effective and reliable method for analyzing the evolution of tailings pond dam failure risk by combining DEMATEL and MISM is proposed. Firstly, 35 risk factors affecting tailings pond failure were summarized. An index system for evaluating the imfluence factors of dam failure was constructed from four aspects: personnel, management, environment, and system. Secondly, the Decision-making Trial and Evaluation Laboratory (DEMATEL) was used to study the influence relationships among the factors, for analyzing and identifying the key causal factors. Subsequently, the Modified Interpretative Structural Model Method (MISM) was used to classify the cause factors into five levels of influence as well as to determine the degree of integrated influence between the risk factors. Finally, an evolutionary model of tailings pond dam failure risk was constructed based on the results of the analysis. The results of the study indicated the followings: 1) System risk accounted for 58.58% of the total weight, while personnel risk accounts for 15.51%. To maintain the stability of the tailings pond system, personnel risk should not be neglected in addition to focusing on systemic factors. 2) Rainfall intensity was an essential causal factor. Focusing on rainfall intensity and taking appropriate measures effectively reduced the risk of dam failure. The height of the dam and the depth of the seepage line accounted for a large proportion of the causal factors, making it possible to control the height of the dam and accurately monitor the depth of the seepage line to improve the stability of the dam. 3) In the tailings pond dam risk evolution model, there were 30 factors with higher mutability and correlation, which played a transitional role in risk transfer. A risk factor transfer network diagram was established for this purpose as a diagnostic map. The research results can provide new methods and ideas for tailings dam failure risk analysis research and practice.

Seepage of oil sand process-affected waters (OSPW) from tailings ponds into surface waters is a common concern in the minable oil sands region of northeast Alberta. Research on seepage has been extensive, but few comprehensive treatments evaluating all aspects relevant to the phenomenon are available. In this work, the current information relevant for understanding the state of seepage from tailings ponds was reviewed. The information suggests the infiltration of OSPW into groundwater occurs near some ponds. OSPW may also be present in sediments beneath the Athabasca River adjacent to one pond, but there are no clear observations of OSPW in the river water. Similarly, most water samples from tributaries also show no evidence of OSPW, but these observations are limited by the lack of systematic, systemic, and repeated surveys, missing baseline data, standard analytical approaches, and reference materials. Waters naturally influenced by bitumen, discharge of saline groundwaters, and dilution also potentially affect the consolidation of information and certainty of any conclusions. Despite these challenges, some data suggest OSPW may be present in two tributaries of the Athabasca River adjacent to tailings ponds: McLean Creek and Lower Beaver River. Irrespective of the possible source(s), constituents of OSPW often affect organisms exposed in laboratories, but research in all but one study suggests the concentrations of organics in the surface water bodies assessed are below the standard toxicological effect thresholds for these compounds. In contrast, many samples of groundwater, irrespective of source, likely affect biota. Biomonitoring of surface waters suggests generic responses to stressors, but the influence of natural phenomena and occasionally nutrient enrichment are often suggested by data. In summary, valuable research has been done on seepage. The data suggest infiltration into groundwater is common, seepage into surface waters is not, and anthropogenic biological impacts are not likely. (C) 2019 Elsevier B.V.