4.3 Waste rock dumps

Waste rock typically emerges from an open pit in a relatively dry condition. Once placed in a surface dump, it is exposed to wetting and degradation influenced by:

  • the magnitude and intensity of rainfall
  • rainfall infiltration and evaporation losses
  • the height and slopes of the waste rock dump (WRD)
  • the nature of the waste rock, including its chemical and physical characteristics.

4.3.1 Geochemistry of surface waste rock dumps

Monitoring must verify the geochemical characteristics and model assumptions that guide waste rock dumping plans in order to continue to protect adjacent water values throughout the life of the operation. Contaminant load predictions made during exploration can be checked and adjusted and more complete operational datasets compiled to help plan water-quality and other control measures as part of rehabilitation and to determine whether covers are required (see Appendix 2).

Where soil or benign waste rock covers are found to be needed to reduce future risk from mine drainage, cover designs should integrate what has been learned from monitoring waste rock hydrology and geochemistry, and the covers should be monitored for erosional stability and performance so that covers designed for new landforms can continually improve.

In some instances, metal-tolerant plants (metallophytes) may be useful for reducing soil concentrations of particular metals via phytoremediation (Baker et al. 2010). A proactive leading practice approach to phytoremediation would be to investigate the presence and unique characteristics of site-specific metallophytes during pre-mining ecological surveys to identify species suitable for remediation and to run operational research trials to evaluate their performance. The monitoring data can be used to inform the choice of rehabilitation species mix and establishment methods when integrated with other aspects of rehabilitation planning.

4.3.2 Water monitoring of surface waste rock dumps

On wetting-up by rainfall infiltration, WRDs have the potential to generate base seepage into the underlying foundation and to topographic low points around the toe of the dump. The seepage is likely to be contaminated if the dump contains reactive waste rock. As the dump wets up, the amount and rate of base seepage increases, and the trigger rainfall necessary to initiate base seepage and the time lag before base seepage occurs following triggering both decrease.

In order to design control measures to manage WRD base seepage and potential contamination and assess the performance of those measures, monitoring is needed before, during and after the construction of dumps to enable the prediction of their hydrological and geochemical behaviour.

The quantity and rate of production and the quality and fate of surface run-off and base seepage from a surface WRD are all important in assessing the extent of potential environmental impacts. The balance between rainfall infiltration and run-off depends on the geometry and construction of the dump, the physical nature of the waste rock and the rainfall regime (climatic conditions in the monsoonal tropics are very different from those in semi-arid or temperate zones).

Seepage to the foundation often results in groundwater mounding beneath and around the WRD. The extent of mounding should be monitored by means of borehole piezometers. Since it is very difficult to install piezometers directly beneath an existing WRD, they are typically located around and just outside the WRD footprint. Piezometer installation involves placing a screen at the depth of interest down a borehole and sealing the borehole above and below that depth. Water levels in the borehole may be monitored manually using a down-hole dip meter or electronically using permanently installed pneumatic or vibrating wire piezometer tips connected to the surface by electronic cables. Borehole water sampling should be used to monitor groundwater quality.

The rate of the surface expression of WRD base seepage at topographic low points around the toe of the dump should be monitored using V-notch weirs and samples collected for water quality monitoring.

In view of the relative difficulty of obtaining representative and reliable direct measurements of rainfall infiltration into WRDs and base seepage from them using localised lysimeters,5 monitoring should preferably be directed at understanding the overall water balance and wetting-up over time for the dump. Automated weather stations installed on WRDs provide primary input data for the water balance. The stations should be equipped with metrological sensors, including solar irradiance and evaporation pans, so that actual evaporation can be calculated and estimates can be made of rainfall infiltration and run-off. The volume of surface run-off should be measured in flumes located in run-off drains designed to capture the bulk of the run-off to provide this component of the overall water balance and provide a cross-check of the infiltration estimates.

Following the closure of a surface WRD, it is necessary to monitor rehabilitation to assess whether closure performance targets have been met. These include targets related to surface stability, such as erosion due to rainfall run-off; soil erodibility; rock mulch/vegetation covers; the water quality of seepage and run-off; dust generation by wind; the performance and stability of drainage works and vegetation establishment; and the sustainability of post-mining land uses. Further details are in the leading practice handbooks Preventing acid and metalliferous drainage (DIIS 2016d), Mine closure (DIIS 2016e) and Mine rehabilitation (DIIS 2016g).

Collapsed - Case study: Erosion monitoring for stable landforms

The key requirement in monitoring erosion is to ensure that the data obtained provides the specific information needed by the site. In some instances, it may be sufficient to demonstrate that erosion rates are declining. In others, there may be greater concern about potential off-site impacts.

Minara Resources Ltd operates the Murrin nickel operation in the north of the Western Australian goldfields. Initial rehabilitation works conducted at the site on constructed landforms showed good vegetation establishment but high rates of erosion.

Consequently, the site engaged expert consultants to design landforms with lower erosion potential. The Water Erosion Prediction Project (WEPP) model was chosen to provide erosion simulations for design purposes. This model requires complex soil erodibility data and a range of assumptions about landscape condition and performance. For that reason, there was considerable interest in obtaining erosion data from constructed landforms to refine the modelling and generate even more cost-effective landform designs.

Therefore, the erosion monitoring objectives were:

  • to demonstrate that erosion rates were consistent with site targets
  • to enable validation and more precise calibration of the erosion modelling used in landform design at the site, thereby enabling continuous improvement in the design process.

For a range of designed concave slopes, measurements of rill frequency and volume were used to estimate cumulative erosion on landforms constructed in 2004 and 2005. Those measurements were compared with predictions of erosion based on the original design simulations. Actual erosion potential for the periods of interest was assessed by using data on actual rainfall to provide a comparison against predicted long-term averages. Calculated erosion potential for the periods of interest was found to be considerably higher than the predicted long-term average, illustrating the importance of considering rainfall records when assessing measurements of erosion.

In general, cumulative erosion measured in late 2008 showed good agreement with calculated erosion potential. Of great value was data collected in situations where flow patterns, soil condition and/or landform construction clearly did not match the assumptions used in the initial design. That data was used in evaluating the accuracy of the initial design assumptions.

In one or two cases, the observed variations led to slight changes in construction and rehabilitation methods, rather than refinement of the modelling process. In general, the observations made during the measurement of rill volume were extremely useful, demonstrating that data without associated interpretation or qualitative observation and verification is of significantly reduced value.

LANDFORM

LOCATION

POTENTIAL CUMULATIVE EROSION SINCE

CONSTRUCTION

MEASURED CUMULATIVE EROSION SINCE

CONSTRUCTION

2/3 Upper slope (not corner) 37.4 28.3
Lower slope (below tree debris) 37.4 31.9
7/2 concave Upper slope 37.4 0
Lower slope 37.4 0
7/2 back Upper slope (not corner) 30 30.1
9/4 west* Upper slope (30 m from crest) 100-150 102.5
Upper slope (20 m from crest) 100-150 156.6

* Landform not constructed to specification and expected to exceed design erosion rates.

The erosion monitoring undertaken at the site has provided:

  • validation of the landform design process used
  • confidence in the stability of existing landforms that have been constructed to specification
  • refinement and improvement in the design process.

This has led to changes to landform design, including the elimination of flow-concentrating structures such as berms, more effective containment of run-off on the tops of the landforms, and the use of computer simulations of run-off and erosion to develop lower gradient concave slopes.

Subsequently, assessment of erosion relative to model predictions has been carried out for landforms on other mine sites, and the data was reported by Howard & Roddy (2012). The level of agreement between predicted and observed cumulative erosion was extremely strong, provided the initial simulations used experimentally measured erodibility parameters.

Worker taking erosion measurements in the field

Reference:

Howard, E.J. and Roddy, B.P. (2012b)

Monitoring seepage as well as surface run-off water quality and volume is also crucial for understanding risks to wildlife, grazing cattle or sheep and nearby communities. Animals often interact with or drink from seepages or soaks at the toe of WRDs, seepage channels and containment ponds. Risks to wildlife are a function of the extent of interaction, species behaviour and water quality. Wildlife monitoring may be required, complemented by monitoring seepage chemistry, to gain an understanding of ecosystem sensitivity to key water-quality parameters.

It is also necessary to monitor sediment or seepage interception dams to ensure that they capture mine water (not clean water, which should be diverted around disturbed areas) and have sufficient capacity over the life of a project to perform as required. Water quality in streams and natural water bodies down- gradient from mine run-off and seepage areas also requires monitoring to assess downstream risks to aquatic fauna and flora. Stream conditions and the diversity and abundance of biota change considerably through the seasons, so seasonal monitoring programs may need to be implemented. Remote sensing can be used to detect changes over time at a landscape level (see Section 4.14.2).

Ecotoxicology evaluations enable the aquatic impacts to be assessed. Effective monitoring can also distinguish between chronic and acute impacts and help to evaluate the performance of landforms and water management systems.

If wetland filters are used as a treatment of run-off water with low-level contamination and suspended solids or for tertiary treatment of water discharged from a water treatment plant, they need to be monitored to ensure that they can manage and treat water at the rates required (for example, taking into account variation in rainfall) and that the water they release meets water-quality discharge requirements for the site. The concentrations of metals that accumulate through time should also be monitored to provide an indication of whether or not the wetland may ultimately need to be either cleaned out and re-established or remediated in the future.

4.3.3 Monitoring of in-pit waste rock dumps

For an in-pit WRD, the monitoring-related issues are essentially the same as for a surface WRD. However, if a partially backfilled pit will be actively or passively flooded at closure, there are specific monitoring requirements that will need to be addressed to ensure that final water quality in the flooded pit meets closure targets.

Footnotes

5 The limitation of lysimeters is that they only give spot values of a highly variable parameter and need to be extremely well designed, constructed and maintained.

Share this Page