Why Tailings Risk Is a Governance Problem, Not an Engineering One

Mining companies have spent hundreds of millions of dollars upgrading tailings technology since the Brumadinho disaster, yet the global standard that emerged from that catastrophe says almost nothing about which technology to use. It is almost entirely about how organisations make decisions.

Tailings storage facilities hold the largest volumes of mining waste on earth, and their failure history reveals a recurring pattern. The geotechnical design was often sound; the system surrounding it was not. Cost pressures overrode monitoring decisions, hydrology and geotechnics were evaluated in separate silos, and warnings were recorded but never escalated.

The disasters were organisational before they were physical. That insight is now embedded in regulation, investor expectations, and the governance architecture leading miners are racing to conform with.

This piece explains what systems thinking actually means in a tailings context, why the Global Industry Standard on Tailings Management is built the way it is, and what that structure tells you about which companies have genuinely absorbed the lesson, and which are still treating tailings as an engineering problem rather than an enterprise-wide risk.

Why tailings failures are rarely just engineering failures

Here is the uncomfortable pattern. Facilities that passed technical review, that were signed off by qualified engineers against accepted design criteria, still failed. If the problem were purely geotechnical, that would not happen with the regularity the record shows.

The record of tailings failure patterns across modern mining history shows that geotechnical design errors are rarely the primary cause; the more common thread is a breakdown in how disciplines communicate, how warnings are recorded, and how cost pressures interact with safety decisions over time.

The Brumadinho/Feijão dam collapse in Brazil in January 2019 killed hundreds of people and became the reference point for the entire industry. Post-incident commentary did not primarily blame the dam’s physical design. It pointed to over-reliance on company-internal engineering assessments without sufficiently independent review, insufficient monitoring, and the absence of effective emergency planning.

Five years earlier, the Mount Polley failure in Canada in 2014 told a similar story. Analysts traced it to cost pressures, design compromises, and a weak integration of hydrology and geotechnics, disciplines that were each assessed in isolation rather than against each other.

The common thread is change management. A decision made in one part of the operation quietly rewrites the assumptions another part was relying on. A throughput change in processing alters the tailings stream chemistry, which alters deposition behaviour, which interacts with the water balance in ways the geotechnical team never modelled.

The failure pathways tend to cluster around a handful of discipline interactions:

• Processing to deposition: changes in grind size or thickening alter how tailings settle and consolidate, shifting the physical behaviour of the stored material.
• Hydrology and geotechnics: water balance and structural stability are evaluated separately, so a change in one is never tested against the other.
• Monitoring to escalation: instrumentation records warning signs, but the organisational path to act on them is broken or too slow.
• Operations to closure: operational decisions are made without reference to closure and post-closure obligations, storing up long-term liabilities.

A June 2026 subscriber workshop reported by The Intelligent Miner, with input from Trevor Sparks, Brian Ayres, Karen Chovan, and Tabatha Chavez Matus, framed the core lever plainly: better tailings outcomes depend on coordinating decisions, knowledge, and expertise across the whole mining system, not on optimising any single discipline.

For you as an investor, this reframes the question entirely. The useful signal is not “what technology does this company use?” It is “how does this company make decisions across disciplines, and what happens when those decisions pull against each other?” A company’s incident history, and crucially its willingness to disclose near-misses, reveals more about systemic risk than any technology claim.

What GISTM actually requires, and why the architecture looks the way it does

If the failures were organisational, the response had to be organisational too. That is the logic behind the Global Industry Standard on Tailings Management, and once you see the failure patterns, each requirement starts to look less like bureaucracy and more like a direct countermeasure.

GISTM was published in August 2020 through a collaboration between the International Council on Mining and Metals (ICMM), the UN Environment Programme (UNEP), and the UN Principles for Responsible Investment (PRI). It contains 77 requirements and covers the full facility lifecycle, from site selection through to post-closure.

The standard is built primarily as a governance instrument. It requires board-adopted policy, clear senior accountability, independent technical review boards, public disclosure, and emergency preparedness. Each of those maps onto a specific failure mode seen at Brumadinho and Mount Polley.

ICMM members committed to full conformance for all extreme and very-high consequence facilities by 5 August 2023, and for all remaining facilities by 5 August 2025. The conformance picture that emerged tells you where the industry actually sits.

The ICMM Tailings Progress Report 2025 covered 836 tailings facilities across member companies. As of the August 2025 deadline, 67% (558 facilities) were in full conformance and 33% (278 facilities) were in partial conformance.

That 33% is the number worth sitting with. A third of facilities among the most committed operators in the world had not reached full conformance by their own deadline. For you, that gap is not something to accept at the aggregate company level; it is something to probe facility by facility, because partial conformance can hide meaningful risk inside an otherwise reassuring headline.

From voluntary reporting to independent certification: the GTMI shift

Before January 2025, conformance was largely self-declared and internally audited. That recreated exactly the information asymmetry that characterised pre-Brumadinho risk assessments: the people reporting the risk were the same people exposed to the consequences of reporting it honestly.

The launch of the Global Tailings Management Institute (GTMI) in January 2025, supported by ICMM, UNEP, and PRI, changed that. It provides a structured, independent third-party auditing and certification framework against GISTM.

Independent tailings certification through the GTMI framework requires facilities to demonstrate integrated documentation, internal audit results, and external review board findings as a package, rather than submitting each element separately, which is why the certification timeline at Cerro Corona ran across multiple internal and external audit cycles.

The Gold Fields Cerro Corona Mine shows what that looks like in practice. Achieving certification, delivered with technical support from Stantec and reported on 25 September 2026, required identifying data gaps, building technical documentation, running internal audits, commissioning external audits, and completing independent technical review as integrated activities rather than a sequence of boxes to tick.

For you, the distinction matters. Independently certified conformance can be read differently from uncertified self-reporting, and ESG rating frameworks increasingly embed that difference.

How leading miners are putting systems integration into practice

Principle only gets you so far. The more useful test is what systems integration looks like when a real company has to document it, and the record shows a genuine spread of progress even among the largest and most committed operators.

Vale, the company at the centre of Brumadinho, now reports GISTM implementation across 37 tailings storage facilities in its Iron Ore Solutions business and 13 in Vale Base Metals. The Base Metals disclosures report 100% compliance with all 77 GISTM requirements across those 13 facilities, verified through external consultancy audits finalised in January 2026 and reported in executive summaries through to June 2026.

Teck reports conformance across 26 of its 27 facilities as of September 2026. Freeport-McMoRan announced on 29 October 2025 that it had implemented GISTM across all applicable facilities by its August 2025 target, verified by an independent third party.

BHP‘s 2026 Tailings Storage Facility Policy Statement frames implementation as spanning “all assets throughout the life cycle of TSFs, from design to post-closure,” positioning tailings as an enterprise risk rather than a project line item.

Not every operator is at the same point. Sibanye-Stillwater reported on 19 August 2026 full conformance for its Very High and Extreme consequence facilities, with validation and closure of outstanding projects scheduled through 2026. Anglo Asian Mining has obtained an initial Digbee ESG rating and committed to adopt GISTM by the end of 2026, as noted in September 2026.

The variation in disclosure granularity tells you as much as the conformance figures. A company reporting at the requirement level, all 77 requirements per facility, is demonstrating a meaningfully different internal governance posture from one reporting a single aggregate number. The depth of the disclosure is itself evidence of the depth of the system behind it.

The shift toward evidence-based safety outcomes reflects a broader move across the sector away from documenting compliance activities and toward demonstrating measurable results, a distinction that shows up in how companies like Vale now report against each of the 77 GISTM requirements rather than at a portfolio level.

The investor’s systems-thinking checklist for tailings due diligence

You now have the theory and the live examples. What follows that groundwork is the practical part: a structured way to interrogate a company’s tailings governance that you can apply to any mining disclosure you pick up.

Treat this as a diagnostic of organisational capability, not a compliance scorecard. The underlying question is simple. Does the company manage tailings as an enterprise-wide risk, or as a project-level engineering problem? The first three questions reveal governance architecture; the last three reveal operational integration.

Governance-architecture questions

1. Is there a board-approved tailings policy that references GISTM, with clear accountability at senior executive level?
2. Does the company publicly list all facilities with their consequence classifications, active or closed status, and GISTM conformance level?
3. Are independent technical review boards in place, and are external audits or GTMI certifications disclosed?

Operational-integration questions

1. Does disclosure cover closure and post-closure planning, not only active operations?
2. Are recent capital and operational decisions, such as switching to filtered or dry-stack tailings, presented alongside governance and monitoring enhancements rather than as standalone technology claims?
3. Is there a transparent record of tailings incidents and near-misses, with evidence that the organisation actually learned from them?

ESG rating agencies now treat tailings as a risk spanning environmental, social, and governance dimensions rather than a narrow technical compliance matter. Anglo Asian Mining‘s Digbee ESG rating is an example of a third-party framework explicitly folding tailings governance into its assessment.

The core principle for your due diligence: tailings risk is a systemic organisational risk, not a technical compliance matter. The governance architecture GISTM requires is the evidence of whether a company treats it that way.

One gap limits how far any of this can take you. There is no consolidated global count of all active tailings facilities across every company and jurisdiction, and no consolidated global annual failure-rate data post-2020. That means GISTM disclosure quality is currently the best available proxy for systemic risk. For you, the absence of facility-level reporting with consequence classifications is not a minor gap; it signals that a company may not have the internal systems to manage tailings risk at the level GISTM demands.

Red flags that signal siloed tailings management

A handful of warning signs tend to cluster where tailings are still managed in isolation. Read each as an organisational signal rather than a one-off compliance miss:

• No board-approved tailings policy that references GISTM.
• No facility-level consequence classification disclosed.
• No independent technical review board and no GTMI certification.
• No transparent incident or near-miss disclosure.
• Technology claims presented without any accompanying governance or monitoring context.

Why governance architecture, not technology, is now the investor signal

Step back and the picture settles into a clearer vantage point. Technology is necessary, but it is not the variable that separates one tailings risk profile from another. The distinguishing factor is the quality of the governance and systems integration wrapped around that technology.

The technology-first view deserves a fair hearing, because it is not wrong. Filtered tailings, paste tailings, and dry-stack storage genuinely reduce physical risk by cutting water content and improving stability, and ICMM notes that companies including Newmont and Freeport-McMoRan are developing new technological applications for safer storage.

But the institutional response that followed Brumadinho, GISTM and GTMI, reflects a consensus that technology without governance integration is not enough. The standard is built around decisions, accountability, and review, not around a preferred engineering method.

The technology-versus-governance framing is a false choice. Safer technology and integrated governance are not alternatives; the consensus position embedded in GISTM is that physical design only holds when an organisation is built to make, review, and escalate decisions across disciplines.

For you, GISTM conformance and governance disclosure quality are the closest thing to a credit rating for tailings risk, and they deserve the same analytical seriousness. As GTMI matures and more facilities move through independent certification, the information environment will improve, shifting the picture from self-declared to independently verified. In the interim, with no global failure-rate data to lean on, disclosure quality and governance architecture remain the clearest signal available.

For readers wanting to understand the broader context in which GISTM disclosure sits, our full explainer on ESG reporting reform in mining examines how investor pressure and regulatory convergence are reshaping what mining companies must disclose, and where tailings fits within that wider accountability shift.

The investor who evaluates tailings as an enterprise-wide systems risk rather than a technical line item is using the same framing that ICMM, UNEP, PRI, and the world’s largest miners now use. That is simply engaging with the risk where it actually lives.

This article is for informational purposes only and should not be considered financial advice. Investors should conduct their own research and consult with financial professionals before making investment decisions.