Two erroneous statistics that are most widely cited in the tailings impoundment space are: 1) there are an estimated 3,500 tailings storage facilities (“TSFs”) worldwide, or a modified version that there are 3,500 active TSFs, and 2) tailings dams fail at a rate 10 times greater than water dams based on that estimate.  Both numbers are incorrect and have no basis in fact.  Unfortunately, the 3,500 estimate  in its original context was a pre internet effort to develop a global count as the introduction to a paper on why tailings dams fail. Only three governments provided estimates and having no other basis for estimation, just doubled the total of those three. The author never expected or intended for that number to become so widely used especially into the era of the internet . 

A more accurate present world estimate of 29,000-35,000, developed by WMTF is presented in the table below where the principal documentation is direct or “cited with documentation” sourced from actual official counts by nation.  Authoritative documentation by nation totals 19,214 which represents 60% of the total based on data world mineral production .

While its possible , as WMTF has done, to develop a meaningful summary of the present state of the world portfolio, the nation level data is essential to pin pointing risk and accurately mapping failure history. Gaps in national disclosure and impacts on interpreting failure history are apparent in the above map prepared by young journalist Marco Ranocchiari for an article . This map makes it clear that Australia has the lowest failure footprint among all nations with inventory disclosure. It clearly highlights the Philippines and Brazil has having a very high magnitude profile relative to the size of inventory.


From our examination of over 8,000 records WMTF estimates that 50%-60% (14,500-17,500 ) are active, 25% -30% each inactive and abandoned (each 7,250-.8,250). As explained in our detailed presentation of the relevance of operating status to a risk centered portfolio disclosures, there are not enough data points among the 8,000 records we examined to firmly estimate distribution within any given national or company profile by operational status.


We see more and more investors and work by leaders in the mining and tailings space citing and using the WMTF estimate which we presented by invitation at three investor webinars in February 2019. These data are revised since then from additional counts. We are glad to see other independent estimates which reconcile with ours.

Research papers and NGO writings and websites trying to call attention to the threats and risks of poorly designed and/or poorly managed TSF’s that cite the 3,500 estimate, even as and estimate of “active TSF’s”, grossly undemine their own purpose by minimizing the magnitude of TSF risks as a global issue in every major mineral-producing nation, especially those nations that have been long engaged in sourcing raw minerals.

It is important to appreciate that an accurate one-time count such as that attempted by the Church of England (COE) tailings intervention program has very limited value except for certain “point in time” global assessments such as WMTF is now undertaking to assess risk and the costs of de risking the present world portfolio.


Additionally, even as a “present state count, the COE approach of mining company survey does not produce a representative or useable data set. The COE effort resulted in details on only 875 useable records out of 1806 reported after duplicates, errors and “closed facilities” were removed. records ( about 4% of the world count and 8% of world volume). Moreover the COE, as transmitted to us in November 2019 by Elaine Baker, is definitely not representative of the world portfolio in distribion by nation, operating status, hazard potential, size, height, or age when examine in the context of 8,000 records in other authoritative portfolios.

In general, a survey of mining companies does not include , by definition, abandoned facilities which represent an estimated 20% of the total world portfolio in 2020. Only disclosures by the licensing local authorities would include abandoned facilities which can also pose a risk of failure.


As a portfolio the COE dataset is of larger facilities with a higher ratio of active to inactive facilities, and a much larger rate of growth next 5 years as compared to the world total portfolio. These differences have important implications for world portfolio risk assessment. For example, it means that catastrophic failures among the COE portfolio miners will tend to have greater impacts on world mineral supply and be of greater magnitude as compared with expected failures in the entire world portfolio.

The attempted unpublished world compilation by Dr. Paulina Concha Laraurri at the Columbia Water Center cited in the table below is a much more representative “point in time” world compilation because it mainly utilized known official national disclosures accounting for nearly 3,000 structures or about 14% of total world portfolio. Together with 2019 disclosures by the U.S. Army Corps of Engineers (1635) and Brazil (769), the Laraurri World compilation enabled both an assessment of the COE disclosures and some important conclusions about the present State of World Mine Tailings (2020).


We will never have a meaningful global census of TSF’s until every nation publicly discloses, annually or with reasonable frequency, a risk-centered account of all TSF’s, ever permitted (active and inactive, abandoned and closed).  A simple descriptive list with latitude and longitude, ownership and the basic ICOLD UNEP Bulletin 121 descriptors is of little analytic or risk assessment value to investors, to communities within a potential or to actual mining affected area, or to regulators.

Moreover, the Bulletin 121 descriptors have not been shown to be reliable predictors of risk or failures.  A census by survey and voluntary  response by some international entity not not charged with licensing and oversight is not meaningful especially if it does not include key risk indicators.

To its credit, the COE survey of miners in its instiutional portfolio did include the risk-centered parameters we encouraged as a first cut map of TSF status related investor risk.

The inclusion of planned depositions and hazard potential in the COE Tailings survey added a great deal to our understanding of the utlization patterns within the existing world portfolio .

To be meaningful as a portfolio risk assessment tool a national TSF disclosure needs to include risk based descriptors  such as NP/AP ratios, granulometry( particle size distribution and soil classification), inherent geological risk, susceptibility to natural disasters, and descriptors of down stream risks in the event of failure It is also extremely important to include performance and operational details on leaks and emissions, raises and malfunctions since last disclosure and to place any given portfolio in a context of global, national mineral production data.

At present only the U.S. Army Ccorp of engineers and Brazil are including hazard potential in national disclosures even though it seems apparent from Dr. Paulina Concha Laraurri’s work at the Columbia Water Center that her scores on potential severity of impact from failure actually correlate extremely well with severity of failure. Dr Laraurri’s approach also affords an excellent tool for prioritizing stability assessment and de risking programs at the national or mine portfolio level as well as affording the possibility of a total potential risk score for a given country, investor, or mine company portfolio.

For our present objective of presenting an accurate snapshot of the state of the entire world mine tailings portfolio as of 2020, a well documented estimate of the size of the present world TSF portfolio is relevant and useful in the context of our ongoing work to map and track risk in the present world portfolio . 

We welcome all contributions and corrections with documentation especially on the large mineral-producing nations which do not provide public disclosure on TSF’s.

Please submit any corrections or additions to Lindsay Newland Bowker

We welcome and need your financial support as well for our continued development work on a meaningful risk focused global assessment of the World Tailings Portflio.  Our budget 2021 2023 for further development of the Global Tailings Census is $40,000 . You may ear mark your USA tax deductible contribution to this project by check payable to World Mine Tailings Failures. Mailing address John Z.Steed,Esq., Treasurer, World Mine Tailings Failures 36 Main Street PO Box 368 Blue Hill,Maine04614..  or by wire transfer contact Our US non proft id is 833883531. We are chartered in the State of Maine. The code for this project is PRED01.


The funds will be used mainly to combine the COE disclosures, the ACOE disclosures, Laraurri compilation and the WMTF World Census ( a life of facility structure) into a single data set and to continue to develop and update nation counts. The aim is to develop and analyze more actuarially useful data on at least the top 10 producers in each mineral space.

We have several separate existing data sets of different dates and formats . All of the country disclosures are facility specific, not just world totals. We will, however, continue to search for and add summary data to the summary table by nation.

We are also happy to do presentations to your organization. As a presenter in the context of your organized and managed webinar our minimum honorarium is $1,000 for a 5-10 minute presentaion including a power point with 10 minutes of Q&A.  contact Lindsay Newland Bowker to discuss. Payment , as above.