Mine closure needs better water-quality prediction
Understanding post-closure water quality (Image source: Adobe Stock)
Mine closure is often discussed as something that happens at the end of a mine’s life — in reality, many of the most important closure decisions are made much earlier; while the mine is still being designed, financed, operated, and assessed. Water is one of the clearest examples, writes Shameer Hareeparsad, director: geochemistry, WSP in Africa
Once underground workings begin to flood after closure, groundwater starts interacting with the materials left behind. Those interactions can influence water quality for decades. They may determine whether closure plans are credible, whether regulators have confidence in the proposed management measures, and whether surrounding groundwater systems are protected over the long-term.
For underground mines that use paste backfill, this presents a specific question: how will the backfilled material behave once it is saturated after closure?
Paste backfill has an important role in modern underground mining. It can support excavations, improve ground stability, allow mined-out voids to be filled with mineralised residues, and reduce the amount of tailings placed on surface. These are significant benefits. However, backfill should not only be assessed for how well it can be placed underground or how it performs structurally during operations. It also needs to be understood as a geochemical material that may interact with groundwater once the mine has closed.
Understanding the geochemical source
That requires a different kind of prediction. One of the practical tools used in this kind of prediction is a geochemical source term. It refers to the estimated quality and loading of water that may be released from a material into the receiving environment. Put more simply, it helps answer a practical question: what could this material contribute to groundwater, and under what conditions?
Post-closure water quality must be predicted using the best available evidence, while still recognising the uncertainty involved in modelling systems that will evolve underground over long periods.
Underground nickel mine assessment
In a recent underground nickel mine assessment, the paste aggregate fill included sulphide-bearing tailings, crushed waste rock, process residues, and a slag-based cement binder with an additional binder aid. From a geochemical perspective, this is not an inert blend. The tailings contain sulphide minerals that could generate acidity under certain conditions. At the same time, the binder contributes alkalinity, which can help buffer that acidity once the system becomes saturated.
The question, therefore, was not whether the material was “safe” or “unsafe”. The more relevant question was how the material was likely to behave as groundwater returned to the stopes after mining ceased.
To answer this, the assessment considered two main mechanisms by which solutes could move from the backfill into groundwater. The first is advection, where water movement carries dissolved constituents through pore spaces. The second is diffusion, in which constituents move more slowly due to concentration differences between the backfill and the surrounding water.
During the early flooding period, advective movement can be more significant because hydraulic gradients remain, and more soluble constituents may be flushed from the material. As the system becomes fully saturated and those gradients weaken, diffusion is expected to become more important over the longer term.
Laboratory testing provided the basis for this assessment. Modified triaxial permeability testing was used to estimate advective porewater quality, while the Leaching Environmental Assessment Framework method was used to derive diffusion coefficients. These results were then used to inform groundwater modelling and closure planning.
The results showed a material with low hydraulic conductivity, which limits bulk water movement through the backfill. The long-term porewater chemistry was predicted to remain strongly alkaline and dominated by calcium and chloride. Under those conditions, most metals are expected to remain at low concentrations in solution.
The early flushing period is different. The upper-case scenario, which represents a cautious view of what could happen during initial saturation, showed short-lived increases in certain constituents, including calcium, chloride, sodium, potassium, strontium, and copper. This is important because closure planning must account not only for the eventual stable condition, but also for the transitional period when a flooded underground system is still adjusting.
That does not mean the early-stage results should be treated as a forecast of permanent water quality. In the modelling, those upper-case conditions were deliberately conservative. The assessment applied cautious assumptions, including high mass-transfer rates and the exclusion of some secondary mineral reactions that could reduce dissolved concentrations.
Why conservative modelling matters
This approach is intended to avoid understating the potential loading to groundwater.
That precaution is essential. In mine closure, overconfidence is expensive. If predictions are overly optimistic, the consequences may not become visible for years, when treatment requirements, environmental impacts, or liability concerns are far harder to manage. A conservative model does not remove uncertainty, but it gives project teams, regulators, and other decision-makers a more defensible starting point.
This is also why site-specific testing matters. It is tempting to borrow assumptions from other operations or to rely on generic expectations about how paste backfill should behave. That is seldom good enough. The mineralogy, binder chemistry, groundwater setting, saturation rate, and underground geometry all influence how a backfilled stope may perform after closure.
Why site-specific testing matters
No laboratory programme can perfectly reproduce the underground environment over the course of decades. Temperature, groundwater chemistry, hydraulic gradients, mineral reactivity, and stope-scale flow paths will always introduce uncertainty. But a structured approach can narrow that uncertainty and make it easier to refine predictions as monitoring data becomes available during operations and after mining has ceased.
Planning for uncertainty
The scrutiny around closure plans is also changing. Regulators, financiers, communities, and internal governance teams increasingly expect mining companies to show not only that a closure plan exists, but that it is based on credible evidence and can be adapted as conditions change.
Paste backfill can play an important role in more responsible underground mining and closure design. But its value depends on understanding both its structural and geochemical performance. A material that works well operationally still needs to be assessed for how it may behave once the mine is no longer active and groundwater conditions begin to recover.
Bringing water quality into closure planning
The broader lesson is that post-closure water quality should not be treated as something to discover after flooding has occurred. It should be part of the design conversation from the outset.
Better prediction will not eliminate every closure risk. Mining systems are too complex for that. But it can help the industry make more informed decisions, identify where monitoring is needed, and demonstrate that long-term water protection has been considered before the mine reaches the end of its life.
That is where closure planning needs to move: not towards false certainty, but towards evidence strong enough to support responsible decisions in conditions that will always carry some uncertainty.
Read more:
FLS highlights lifecycle approach to African mining operations
Mine dewatering: integrated pump technology
Rethinking South Africa's wastewater infrastructure
