Progress continues in TSF design and monitoring, but challenges persist

It has been remarkable to witness over the years the amount of focus placed on Tailings Storage Facilities (TSFs) brought on by several significant failures that captured the attention of mainstream media. The same level of focus could be felt being shifted within the mining industry.

TSFs are where, in most cases, the bulk of the processed ore will end-up. It is generally spent, processed material with little to no economic value, and it must be managed to remain safe for people and the environment throughout its operational life and in perpetuity after closure.

The challenge with this framework  is that the risks associated with tailings mismanagement can have disastrous consequences such as the 2019 Brumadinho TSF failure in Brazil, which was captured on camera while it occurred, and closer to home in South Africa, the Jagersfontein TSF failure in 2022.

TSF failures have motivated for new standards, such as the Global Industry Standard for Tailings Management (GISTM), which is mandatory for member mines of the International Council on Mining and Metals (ICMM) to comply with. However, it is encouraged for non-members to also align with. It is worth noting that the intention for this standard is not to replace any regulatory requirements that may be applicable to a TSF in any country, and more to provide a robust standard were gaps in regulation, or local tailings practices could exist. The overarching principle of the standard remains to encourage a culture of “Zero-Harm” with tailings management, which it promotes through its 77 Principles covered under 6 key topics. What is important to draw from GISTM is the protection of project affects persons and communities, and impetus on the full cycle of a TSFs life – from conceptualisation to post-closure.

Advancements in Investigation and Monitoring Methods

Exciting areas of advancement have been in the investigation methods and in the monitoring of tailings dams, most of which were facilitated by innovations in technology and analysis methodologies as well as  telecommunication expansion.

From the design of TSFs, we have seen the adoption of “undrained” soil models becoming more widely adopted, as well as the testing methodologies for determining the potential for tailings “liquefaction”. This is a phenomenon of rapid strength reduction caused by the increase in pore-water pressure that cannot dissipate quickly enough, and by the nature of soil material which tends to “contract” while shearing which causes the tailings mass volume to decrease thus further increasing the pore-water pressure.

The liquefied tailings, if not correctly designed for with adequate confinement, could result on flow slide from the TSF with near water-like behaviour under triggering events, such as seen from the dramatic footage captured at Brumadinho. Applying undrained behaviour to soils that are susceptible to undrained shear properties where they could be applicable, and where there is potential for liquefaction, has allowed better insight into potential failure mechanisms and critical slip surfaces.

Encouragingly, geotechnical laboratories in South Africa have seen a significant increase in capacity to test tailings material for its Critical State, as well specialist contractors with improved capabilities with in-situ testing, especially with Cone Penetration testing (CPTu).

The past several years has seen an improvement in the efficiency and accuracy with local geotechnical laboratories increasing their capacity with specialised testing equipment, including Triaxial Compression, Direct Simple Shear, and Rowe Cell apparatus. This has significantly reduced the turnaround time and cost of these tests, and improvement in the quality of test results leading to fewer test repeats and overall improved project efficiency.

This has been especially useful for Critical State test work which requires many tests under varying conditions.

Increased Data Analysis Capabilities

Advancements in the testing methodologies, and the equipment used have improved how operational and closed TSFs are investigated. This includes advancements in the following simultaneous testing with CPTu:

• Shear wave velocity testing (Seismic) for better characterization of the in-situ tailings small strain stiffness (Gmax) and dynamic response under cyclic loading.
• Resistivity testing, which can with accuracy determine the degree of saturation of tailings material, and locate the water table in a tailings stack, and identify potentially liquefiable zones of tailings.

The application of these methods has been well-supported by increased computing power, that can support highly detailed and complex analyses more efficiently than before. However, more recently, the potential to include artificial intelligence and machine learning in streamlining the analysis of data and classification of tailings material in almost live-time. And even collate Critical State test work and derive solutions, identify outliers, all with far superior efficiency than human effort. While caution, and the human touch is still strongly advocated in these cases, a more widespread advantage to machine learning applications is its use in TSF monitoring, which may include:

• Optimising a monitoring plan based on the interpretation from field data, and slope stability assessments, including instrumentation type, and location
• Live analysis, and interpretation of monitoring telemetry data
• Development and updating of trigger limits.

These tools should be used in support of the engineer’s judgement, and never to replace it, and therefore a good understanding, and close oversight with the use of any technology is still required.

Compliance Still a Challenge

As much as technology, understanding, and methodologies have been developed that align with the principles of GISTM adoption of these, and even complete compliance remains a challenge for many mines. In the context of African operations, challenges to complete adoption are often affected by:

• Skill shortages and retention
• Capital investment
• Accessibility
• Telecommunications
• Limited historical hydroclimatic and seismic monitoring record.

Complete GISTM adoption requires a plethora of skills, the availability of which is limited in Africa, which has also been victim to skills emigration to other parts of the world, and has placed greater reliance on the import of skills from more developed parts of the world or left knowledge gaps in key positions in the tailings lifecycle. This impact extends to the capacity and capabilities of tailings consultancies in Africa.

In addition telecommunication connection and physical access remain a challenge in Africa, where many of the mining developments are in rural, under-developed areas. Implementing TSF live monitoring systems rely on access to some data capturing and upload system, usually reliant on stable internet access, or direct access to a server. Furthermore, electronic and mechanical instrumentation require calibration by the instrument manufacturer. The remoteness of many mining operations also limits the availability of reliable historical hydroclimatic (rainfall, evaporation, and flood), and seismic monitoring data is highly limited. These are critical components to the development of a TSF design basis – a GISTM requirement. There is heavy reliance on satellite monitoring, and climate models that are calibrated on limited data.

It is important to remain practical in designing TSFs and implementing monitoring plans when considering site-specific challenges and constraints, and engineering solutions around these – this philosophy is at the core of Entail’s service delivery.