Predictions of whether and when a waste rock pile may start to generate acidic water, and how long a pile may release elevated concentrations of metals to the environment, are related on a fundamental level to an understanding of fluid flow within a waste rock pile. Our analysis of the hydrogeological properties of waste rock piles is based on a synthesis of data from four mine sites; Myra Falls, B.C., Island Copper, B.C., Elkview Mine, B.C., and Golden Sunlight Mine, Montana. The emphasis in this study is on pile hydrostratigraphy and the textural properties of the rock mass, spatial and temporal variations in water content within a waste rock pile, temperature profiles within a waste rock pile and their response to the infiltration of water following a rainfall event, and the large-scale hydrogeologic characterization of a waste rock pile inferred from outflow hydrographs recorded in toe drains.
Four hydrostratigraphic models are proposed to characterize waste rock piles; they differ depending upon material types and construction methods. This framework differentiates between porous flow in finer sandy gravel materials and channelized flow in coarser materials. The models are non-segregated coarse-grained rock piles that transmit water rapidly to the base of the pile, non-segregated fine-grained rock piles that are likely to contain a basal saturated zone, segregated rock piles that contain a fine-grained crest zone that may not permit the passage of significant quantities of water; and layered, segregated dumps that contain a finer-grained crest and sandy gravel layers parallel to the face of the rock pile.
Volumetric water content is an important characteristic of a waste rock pile. It appears to be closely associated with the textural properties of the pile, and it can be used to scan the pile hydrostratigraphy. For a given waste rock pile, at each depth, values of the water content appear stable throughout the year. For the data from Golden Sunlight Mine, attempts to monitor matric potential using heat dissipation sensors, and to correlate changes in matric potential with rainfall events, were generally not successful. Temperature appears to be one of the more reliable parameters to use for tracing the movement of water in those regions of the pile that are reactive and generating heat. The fluctuation of the water table in response to infiltration is affected by the permeability structure of the pile and location within the pile. The permeability structure of the pile is the spatial distribution of permeability values within the different regions of the pile. The data we examine is suggestive of rapid infiltration of water through waste rock piles, although sampling frequencies were not adequate to develop precise estimates of fluid velocities.
A methodology is presented, based on kinematic wave theory that relates the outflow hydrograph recorded in toe drains to large-scale parameters characterizing the hydraulic conductivity structure of the waste rock pile. The outflow in response to an infiltration event is treated as an integration of the outflows from different channel groups within the pile. Water transfer from the channels to the finer-grained matrix is taken account of in the analysis. Application to the Island Copper data set suggests that the approach holds promise as a means of characterizing large-scale flow processes in a waste rock pile. Further work is warranted to improve the model in its representation of channelized flow, and to apply the method to rainfall events at a number of different sites to gain insight to the relationship between hydrostratigraphy, and flow responses.
The most significant limitation of the existing database is that no single site provided a complete data record of the important parameters required to characterize the hydrologic behavior of a waste rock pile, and the frequency of sampling was often insufficient for our purposes. In our opinion, to better understand the hydrology of a waste rock pile, the following measurements should be given priority: water content and temperature profiles through the unsaturated zone, water table elevation, volumetric discharge at toe drains, and rainfall and air temperature. Workplans are presented for three types of monitoring studies; a pile assembly study, a pile monitoring study, and a pile disassembly study. It may be advantageous to link these workplans to operations at a low-grade stockpile. It is important to coordinate the suite of measurements made prior to and during the disassembly of a pile.
