Abstract
Seepage modelling for a conventional tailings storage facility (TSF), “dry” stack, waste rock dumps (WRD), heap leach facilities (HLF), and combined facilities (co-disposal) is typically performed in 1D and 2D to predict seepage into the groundwater system, which is then used to determine the need for geosynthetic liners (or other materials with low permeability that provides a hydraulic barrier). 3D simulations are seldom performed due to the complexity of the model and the computing load. The modelling presented herein relies on Hydrus-1D for modelling the seepage through the mine waste. Hydrus-1D is an industry standard one-dimensional time-dependent code that can simulate water and solute transport in a variably saturated porous medium. Hydrus-1D can also be coupled with the geochemical model PHREEQC for water quality predictions from a reactive media. PHREEQC is a frequently used standalone model for predicting water quality from mine waste installations.
The concept for this modelling tool is to discretize in time and in space the deposition of the mine waste to enable 1D modelling for each of the individual blocks. The 1D blocks also enable modelling of stacking material as the mine waste facility is constructed. The discretization of the time-dependent 3D objects, the assembling and stacking of the Hydrus-1D simulations and the data compilation was automated using Python, which relied on Autodesk® Dynamo® to discretize the 3D objects, and Hydrus-1D for the seepage modelling.
Autodesk® Dynamo® is used to discretize the variable 3D geometry of the TSF into independent columns at different time steps. Each column is then assembled into multiple stacks representing the stacking of mine waste over time as the waste facility is gradually constructed. Python is then used to read the inputs, run the Hydrus simulations for each discretized block and stacks, and processes the results.
This paper limits the discussion to seepage modelling, but as mentioned above, this computing tool can also accommodate water quality predictions via the coupled PHREEQC model available in Hydrus-1D.
Authors:
Michel Noël | SRK Consulting, Canada
Imanol Vega | SRK Consulting, Canada
Felipe Lopez Rivarola | SRK Consulting, Argentina
Ignacio Ezama | SRK Consulting, Argentina
Murray McGregor | SRK Consulting, UK
Ariel Terlisky | SRK Consulting, Argentina
Click here to view other presentations at this event.
