Diavik Waste Rock Project: Simulation of the geochemical evolution of a large test pile using a scaled temperature and sulfide-content dependent reactive transport model.

The Diavik Waste Rock Project (DWRP) project included four principal components focused on the development of techniques for assessing the environmental impacts of waste rock at mine sites. These components were small-volume laboratory experiments, intermediate- and large-volume field experiments, and assessment of the operational-scale waste-rock stockpiles, which facilitated characterization of waste-rock weathering at different scales. The heavily instrumented large-scale field experiments (test piles) were constructed to replicate, as closely as practicable, the temperature, water flow, and gas transport regimes of a waste-rock pile that is exposed to annual freezing and thawing cycles and to facilitate characterization of the long-term weathering of a low-sulfide waste rock. An integrated conceptual model of sulfide-bearing waste-rock weathering, developed at the small scale, was applied to assess the capacity of the conceptual model to capture the geochemical evolution of the waste rock at the large field-scale test-pile experiment. The integrated conceptual model was implemented using reactive transport code MIN3P, taking into account scale-dependent mechanisms. The test-pile mineralogy was similar to the small-scale laboratory experiments and included low-sulfide waste rock with an S content of 0.053 wt% (primarily pyrrhotite). The flow regime of the test pile was simulated using parameters measured as part of other DWRP investigations, including temporally variable infiltration estimates that represented the measured precipitation events at the site. The temporally and spatially variable temperature of the test pile was interpolated from values measured using instrumentation installed at the beginning of the experiment and was included in the simulation to refine the temperature dependence of the geochemical reactions. To allow continuous, multi-year simulation, freezing was also simulated to represent the conditions experienced at the test-pile experiment. Normalized root mean square error analysis of the large-scale field experiment simulation results indicated most parameters compare well to measured daily mass flux (i.e., the fraction of the range of annual values encompassed in the residual was less than 0.5 for SO, Fe, Ni, Si, Ca, K, Mg, Na, and pH and 1.0 or less for all parameters except Cu). The method of using an integrated conceptual model developed from the results of humidity cell experiments to implement a mechanistic approach for assessing the primary geochemical processes of waste-rock weathering on a large scale was shown to provide reasonable results; however, the results are specific to the study site and the approach requires application to various sites under different geological and climatological conditions to facilitate further refinement.