Article title

By Hugo Melo

Modelling Landform Evolution

Authors

Author 1

Author 2

Author 3

Author 4

Landform evolution models (LEMs) are used to estimate potential surface water runoff and predict resulting erosion and deposition processes on a landform scale. When applied to mining landforms, predicting landform erosion can be useful to inform closure design and to predict post-closure performance. The use of LEMs to simulate the post-closure evolution of water-shedding mining landforms, such as waste rock dumps, is well-documented. However, tailings storage facilities (TSFs) are unique landforms in that they often have contributing upstream catchments, long and flat (tailings beach) slopes consisting of spatially variable material types, and can be closed systems (i.e. do not release water). These factors all need to be accounted for when using an LEM to simulate a TSF. 

SRK took on landform evolution modelling to simulate the evolution of a valley-fill TSF over a 1,000-year post-closure period. The client’s project was located in an arid region of Australia. The key objectives of the LEM included assessing whether water erosion of land surfaces could have the following effects:

  • Compromise the water storage capacity of the closed TSF;
  • Result in entrainment of tailings in water discharged to the environment; and
  • Result in embankment failure or significant erosion or undercutting of the TSF spillway.

Landform evolution modelling was undertaken using the CAESAR-Lisflood (CL) model to simulate surface water runoff and to predict the resulting erosion and deposition from single storm events and from cumulative, long-term simulations.

The CL model combines a topographic evolution model (CAESAR) with a high-quality hydraulic model (Lisflood). Recent developments in CL modelling have significantly improved the run speed and computational efficiency of the model. This particular project took advantage of these advances, incorporating the use of digital elevation models with different spatial resolutions, sensitivity analyses of key model parameters, long-term simulations using 1,000 years of sub-daily rainfall data, and the assessment of alternative design options. The model considered a number of different scenarios including: tailings capping alternatives, a detailed embankment-only model, different project stages, and various climatic scenarios.

The results indicated that the water storage capacity of the TSF is expected to be progressively reduced due to the sediments that are eroded and deposited within the pond area; however, not to the extent that could result in water (and potentially entrained tailings) being discharged to the environment. The LEM was also used to identify areas where management options could be used to improve or mitigate long-term landform instability, and to develop recommendations for future works to calibrate and verify the model.