Abstract:

In the mining industry, screening water quality model predictions with provincial and federal guidelines is standard practice for assessing environmental risks to the aquatic environment and human health. Traditional (manual) screening methods are inefficient and error-prone, leading to increased costs and less effective assessments. There are limited options for automated screening systems that can be customized to site-specific requirements.

This study introduces an automatic screening and visualization system, developed using open-source programming language (R language and related packages) and piloted on public databases published by the British Columbia Government. This system aims to eliminate human error and provide consistent, accurate results. A comprehensive dataset, consisting of piecewise functions of exposure and toxicity modifying factors (ETMFs) was analyzed to establish conditional expressions for automated screening. These expressions were instrumental in accurately calculating concentration limits for various water contaminants. Model projections, sourced from GoldSim® – an industry-standard application for water quality modelling of mine sites - and ETMF functions were integrated, enabling a coherent interpretation of model data. By applying logic vectors, the system helps to verify compliance with established water quality guidelines. Tested extensively with real project data, the system showed notable efficiency improvements over traditional screening, processing data at an exponentially faster rate than manual methods. Its accuracy was validated for several common water quality parameters. A simplified online version of the screening tool, complemented by an interactive visualization dashboard, is being deployed publicly. This system can simplify the evaluation of complex water quality data, enhances decision-making accessibility and improves transparency.

Menu

Jonathan Love