Time filter

Source Type

Betrie G.D.,University of Sao Paulo | Sadiq R.,University of Sao Paulo | Morin K.A.,Minesite Drainage Assessment Group | Tesfamariam S.,University of Sao Paulo
Science of the Total Environment | Year: 2014

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. © 2014 Elsevier B.V.


Betrie G.D.,University of Sao Paulo | Tesfamariam S.,University of Sao Paulo | Morin K.A.,Minesite Drainage Assessment Group | Sadiq R.,University of Sao Paulo
Environmental Monitoring and Assessment | Year: 2013

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. © 2012 Springer Science+Business Media B.V.


Betrie G.D.,Athabasca University | Sadiq R.,University of British Columbia | Morin K.A.,Minesite Drainage Assessment Group | Tesfamariam S.,University of British Columbia
Environmental Technology and Innovation | Year: 2015

Acid rock drainage (ARD) is a major environmental problem that poses serious ecological risks during and after mining activities. To minimize the ecological risks and to lower remediation costs in the mining industry, ecological risk assessment is highly important in various phases of mining. In this study, a methodology for ecological risk assessment using probability bounds is presented. The methodology is demonstrated with a case study at a mine site. A fugacity-based model was employed to conduct the exposure characterization. Median lethal concentrations from toxicity studies were used to derive predicted no-effect concentrations (PNEC) and to characterize the effect of copper and zinc on the receptors. Probabilistic risk-quotient and overlaps between distributions of exposure and effect were employed to characterize risk. Data and parameter uncertainties in exposure, effect, and risk characterizations were propagated and quantified using a probability bounds approach. The exposure modeling results showed that the predicted concentrations of copper and zinc slightly exceeded the observed concentrations. The results of the effect characterization showed that the derived effect concentrations for copper and zinc are acceptable compared with guideline values. The risk characterization result indicated that a high probability of ecological risk may exist due to metals that are transported into a nearby lake. Moreover, the results showed that the methodology handles uncertainties due to imprecision and randomness in an integrated manner. © 2015 Elsevier B.V.


Betrie G.D.,University of British Columbia | Sadiq R.,University of British Columbia | Morin K.A.,Minesite Drainage Assessment Group | Tesfamariam S.,University of British Columbia
Journal of Environmental Management | Year: 2013

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. © 2013 Elsevier Ltd.


Betrie G.D.,University of British Columbia | Sadiq R.,University of British Columbia | Nichol C.,University of British Columbia | Morin K.A.,Minesite Drainage Assessment Group | Tesfamariam S.,University of British Columbia
Journal of Hazardous Materials | Year: 2016

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. © 2015.


Betrie G.D.,University of British Columbia | Sadiq R.,University of British Columbia | Mohn K.A.,Minesite Drainage Assessment Group | Tesfamariam S.,University of British Columbia
Proceedings, Annual Conference - Canadian Society for Civil Engineering | Year: 2012

Fate and transport models have extensively been used to predict distribution of toxic substances in the multi-media environment. In mining industry, predictive models are commonly used to evaluate performance of mitigation measures and estimate remediation costs during different phases of a mine lifecycle. These models are often used in deterministic form; however the probabilistic analysis through Monte Carlo analysis simulations is also popular to describe parameter uncertainties. In pre-mine phase, where data and information that characterize a mine site are scarce, some parameters of fate and transport models can be best described as random and can subjectively be defined as fuzzy variable. This paper presents a fuzzy-probabilistic approach to propagate parameters uncertainties throughout the modeling process. An aquivalence-based fate and transport model was developed for a mine site. This model was integrated with fuzzy-probabilistic algorithm to predict the distribution of copper concentrations in soil and groundwater. The prediction results showed the distribution of copper concentrations in groundwater, the associated prediction uncertainties and sources of uncertainty.


PubMed | University of British Columbia and Minesite Drainage Assessment Group
Type: | Journal: Journal of hazardous materials | Year: 2015

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.

Loading Minesite Drainage Assessment Group collaborators
Loading Minesite Drainage Assessment Group collaborators