RESUMEN
In waste rock piles, the leaching process involved in acid rock drainage is mainly controlled by water flow. This paper (Part 2) investigates the effects of heterogeneities on the water flow patterns by applying probability density functions to hydrogeological properties. In this study, a piecewise constant distribution is proposed to describe the permeability inside waste rock piles, which reflects the effect of both finer and coarser pores. Compared with uniform water flow obtained from traditional homogeneous modeling, various water flow patterns and their pathways inside waste rock piles can be simulated by the proposed model. In addition, the leaching process is also investigated by coupling the calculated water flow with the geochemical reaction based on the water film model proposed in part 1. For demonstration, these models are integrated and applied to the full-scale waste rock pile at Equity Silver mine in British Columbia, Canada. Because the iron loading is highly correlated to the acidity at this site, it is found that the fluctuation of annual lime consumption for neutralization at this site can be well predicted by the integrated model. In addition, the results indicate that waste rock piles with different spatial patterns of permeability distribution, but with the same probability density function, may have different water flow patterns and spatial distributions of iron concentrations inside the pile. However, the total water flow discharge rate and iron loading profiles from the pile are almost the same on the temporal scale.
Asunto(s)
Contaminantes del Agua , Colombia Británica , Modelos Teóricos , Fenómenos Físicos , Agua , Contaminantes del Agua/análisisRESUMEN
In this paper, a machine learning algorithm based on artificial neural network architecture investigates the correlation between drainage chemistries in seepage water and ambient weather conditions around waste rock piles. The proposed neural network consists of a long short-term memory unit and a fully connected neural network which uses sequenced input to consider current and previous weather impact on the drainage chemistries. A 20-year (1998-2017) monitoring database obtained from the full-scale waste rock pile of the Equity Silver mine in BC, Canada is used for validating the proposed approach. The neural network is trained based on total precipitation and mean temperature as input and the acidity as output. The results indicate that the calculated acidity from the trained neural network matches with that measured in the field well. In addition, the accuracy of calculated acidity can be further increased by adding a time tag of acidity measurement date into the input layer. This refined approach can capture the long-term evolution and dynamics of hydrogeochemical and biochemical properties inside the waste rock piles.
Asunto(s)
Contaminantes del Agua , Canadá , Redes Neurales de la Computación , Agua , Contaminantes del Agua/análisis , Tiempo (Meteorología)RESUMEN
Geochemical reactions taking place at the rock surface and pore water interface, and rapid preferential water flow through waste rock piles are identified as two primary steps for acid rock drainage (ARD) and metal leaching (ML) processes. This paper (Part I) develops a water film model to describe the interactions among sulphide minerals, pore water and oxygen, which considers the reactive surface areas as the primary sites to capture geochemical reactions including sulphide oxidation and neutralization reactions, and also considers acid and metal ion storage in pore water. In addition, the proposed water film model is further coupled with a pile-scale mass transport model to investigate a specific case of the main waste rock pile at the Equity Silver mine, Canada. Overall, the simulated profile of oxygen concentration matches the historical monitoring data. The modeling results revealed potential controlling mechanisms for ARD generation inside the waste rock pile and provided insights into the impact of an engineered cover on the waste rock pile.
Asunto(s)
Contaminantes del Agua , Agua , Canadá , Drenaje , Modelos TeóricosRESUMEN
Acid rock drainage (ARD) is a major environmental problem that poses significant environmental risks during and after mining activities. A new methodology for environmental risk assessment based on probability bounds and a geochemical speciation model (PHREEQC) is presented. The methodology provides conservative and non-conservative ways of estimating risk of heavy metals posed to selected endpoints probabilistically, while propagating data and parameter uncertainties throughout the risk assessment steps. The methodology is demonstrated at a minesite located in British Columbia, Canada. The result of the methodology for the case study minesite shows the fate-and-transport of heavy metals is well simulated in the mine environment. In addition, the results of risk characterization for the case study show that there is risk due to transport of heavy metals into the environment.
Asunto(s)
Residuos Industriales , Metales Pesados/toxicidad , Minería , Modelos Teóricos , Contaminantes Químicos del Agua/toxicidad , Animales , Colombia Británica , Lagos , Metales Pesados/análisis , Oncorhynchus , Perciformes , Probabilidad , Medición de Riesgo/métodos , Incertidumbre , Contaminantes Químicos del Agua/análisisRESUMEN
Acid rock drainage (ARD) is a major pollution problem globally that has adversely impacted the environment. Identification and quantification of uncertainties are integral parts of ARD assessment and risk mitigation, however previous studies on predicting ARD drainage chemistry have not fully addressed issues of uncertainties. In this study, artificial neural networks (ANN) and support vector machine (SVM) are used for the prediction of ARD drainage chemistry and their predictive uncertainties are quantified using probability bounds analysis. Furthermore, the predictions of ANN and SVM are integrated using four aggregation methods to improve their individual predictions. The results of this study showed that ANN performed better than SVM in enveloping the observed concentrations. In addition, integrating the prediction of ANN and SVM using the aggregation methods improved the predictions of individual techniques.
Asunto(s)
Inteligencia Artificial , Monitoreo del Ambiente/métodos , Modelos Químicos , Redes Neurales de la Computación , Algoritmos , Probabilidad , IncertidumbreRESUMEN
The selection of remedial alternatives for mine sites is a complex task because it involves multiple criteria and often with conflicting objectives. However, an existing framework used to select remedial alternatives lacks multicriteria decision analysis (MCDA) aids and does not consider uncertainty in the selection of alternatives. The objective of this paper is to improve the existing framework by introducing deterministic and probabilistic MCDA methods. The Preference Ranking Organization Method for Enrichment Evaluation (PROMETHEE) methods have been implemented in this study. The MCDA analysis involves processing inputs to the PROMETHEE methods that are identifying the alternatives, defining the criteria, defining the criteria weights using analytical hierarchical process (AHP), defining the probability distribution of criteria weights, and conducting Monte Carlo Simulation (MCS); running the PROMETHEE methods using these inputs; and conducting a sensitivity analysis. A case study was presented to demonstrate the improved framework at a mine site. The results showed that the improved framework provides a reliable way of selecting remedial alternatives as well as quantifying the impact of different criteria on selecting alternatives.
Asunto(s)
Toma de Decisiones , Restauración y Remediación Ambiental/métodos , Minería , Modelos Teóricos , Método de Montecarlo , Medición de Riesgo , Sensibilidad y EspecificidadRESUMEN
Acid mine drainage (AMD) is a global problem that may have serious human health and environmental implications. Laboratory and field tests are commonly used for predicting AMD, however, this is challenging since its formation varies from site-to-site for a number of reasons. Furthermore, these tests are often conducted at small-scale over a short period of time. Subsequently, extrapolation of these results into large-scale setting of mine sites introduce huge uncertainties for decision-makers. This study presents machine learning techniques to develop models to predict AMD quality using historical monitoring data of a mine site. The machine learning techniques explored in this study include artificial neural networks (ANN), support vector machine with polynomial (SVM-Poly) and radial base function (SVM-RBF) kernels, model tree (M5P), and K-nearest neighbors (K-NN). Input variables (physico-chemical parameters) that influence drainage dynamics are identified and used to develop models to predict copper concentrations. For these selected techniques, the predictive accuracy and uncertainty were evaluated based on different statistical measures. The results showed that SVM-Poly performed best, followed by the SVM-RBF, ANN, M5P, and KNN techniques. Overall, this study demonstrates that the machine learning techniques are promising tools for predicting AMD quality.