Open Collections

Abstract

Management of both contact water and non-contact water on mining projects is critical for planning, operations, and closure/reclamation. This paper will describe the characteristics of the two most commonly used platforms for water balance modelling (Microsoft Excel and GoldSim), and the difference between a deterministic model and a stochastic model. It will review the various components of a site-wide model including the climate module, diversion, and storage of non-contact water, collection, storage, and use of contact water, facility operations (heap leach facilities, tailings storage, dry stack operations, etc.), miscellaneous water uses (dust control, truck wash facilities, potable water, etc.), water treatment facilities, and potential sources of makeup water (well fields, surface water sources, pit dewatering, treated water, etc.). Model calibration procedures are described for those projects with an operating history and available monitoring data. Stochastic models are intended to quantify risk (generally the risk of having too much water and spilling to the environment, or the risk of having too little water to maintain operations). They can be developed in either the Excel or the GoldSim platform. In transforming a deterministic model to a stochastic model, key input parameters must be converted to frequency distributions. Recommended distribution types are described. Also described in this paper are the procedures used in a Markov Chain analysis of changes in state for the prediction of future daily precipitation realizations in the Richardson weather generation model (WGEN) in the GoldSim platform. Finally, it is important during the development of a water balance model and during the planning and permitting phase of a project, to recognize that it is likely to be transformed into a tool for operations and, ultimately, for the transition into closure and reclamation.