CONTACT: Lindsay Newland Bowker 207 367 firstname.lastname@example.org
DATE: October 1, 2016
According this 2014 report by the Brazilian Government, Brazil has 633 tailings dams, the overwhelming majority built after enactment of the 2010 statutory overhaul. 144 are large enough to cause catastrophic social, environmental economic and public liability loss in the event of failure . In this report, Brazil classifies only 32 of the 633 facilities as high risk by its procedures for analyzing risk. Those procedures however are more like the standard hazard classification which speaks simplistically and narrowly to the issue of how much harm a given outflow could cause. They do not recognize whether what is built and operating can be reasonably expected to attain environmental standards. For example it does not recognize the different propensity for failure for a given height and capacity configuration based on ARD, rate of raise, or dam design, drainage system, capacity to separate silt from sands , proportion of fines, dryness through the stack and other tailings specific characteristics not shared by water dams.
All will recall that the July 2015 inspection report prepared by VOG BR gave a satisfactory rating to the Fundao. As VOGBR explained in a briefly posted letter that is excerpted at length at our wordpress, the report was a required annual compliance report whose form and content is dictated by the regulations and law.
VOG BR wrote that Samarco did not allow them to take any independent measurements or visit the site. Samarco, they said, gave them the data and contractually specified the analysis to be used. This same level of “annual safety inspection” pursuant to law has presumably attended oversight of all of Brazils large dams
Photo From valeaguamais.wordpress.com
Taking a more comprehensive, forensically informed view of risk and evaluating Brazils legal framework, its obviously low level of technical competence in mining, we believe all 144 large tailings facilities in Brazil should be considered at significant risk of failure and a program of independent multi-disciplinary risk assessment be undertaken for all 144. All upstream dams higher than 50 m should receive top priority attention.. The immediate priority should be on the 15 mega dams of extreme height and size relative to the world’s tailings dams. Vale’s mega dam at the Tapira Complex, already exceeding permitted capacity and in process of further expansion and already evidencing cracks and other signs of strain has been widely identified as of particular concern. Our reasons for considering all 144 large tailings dams at risk are:
(1) the 2010 statute reform did not address any of the root causes of catastrophic failure focusing instead on the structure of criminal and civil liability. Under that law the fatal flaws in the Fundao were not detected or addressed until the worst failure in recorded world history actually occurred
(2) Vale controls 114 of 633 total TSF facilities and accounts for a very significant part of all of Brazils iron ore production generating 200 million tonnes of ore per year more than 40% of that in tailings and wastes in just the Iron Quadarangle area.. Vale aim to increase throughput in the 19 Quadrangle by 50% by 2030. Its own 2013 self assessment of the safer alternative of paste thickening and dry stack concluded most of its presently generated tailings did not meet technical criteria for the safe alternatives of paste thickening and dry stack. Only 4 of its 114 TSF’s nationwide , 4%, met the criteria for dry stack or paste thickening. Vale succeeded in making Brazil back down from post Fundao failure reforms that would have put controls on height and rate of rise ( per globally recognized best practice) and banned upstream construction for large dams. (Eisenhammer) Vale, 50% owner of the Samarco vetted the major expansion and ok’d proceeding even though its own consultant flagged the absence of adequate capacity on the Fundao and the absence of any plan or space on site to provide capacity . Vale’s inventory of TSF’s includes many, if not most, of these very large tailings dams.
(3) there is no mining competence within any of the permit issuing entities or within the state of Brazil so even if the legal framework addressed root causes of catastrophic failure there is no technical capacity within government to determine whether environmental and conservation standards are actually attainable . Virtually all of the 144 large tailings dams and the 15 mega talings dams were permitted with no technical competence on the part of regulatory oversight
(4) Minas Gerais alone, a major center of mining in Brazil, .seems both inept and a victim of complete regulatory capture (as was and is the case in Brtish Columbia) . Rio Pomba had two prior serious events (2003 and 2006) before the final very serious failure on 2007. Minas Gerais is responsible for 3 of the world’s 18 very serous failures 1/1/1996-12/31/2015, 17%) and 3 of the worlds 20 serious failures post 1996,15%. We don’t need details on Minas Gerais’ total ore throughput to know that this is out of scale with Minas Gerais’ rank in global ore throughput..
(5) Brazil, since passage of its 2010 statute has embarked on a massive improperly regulated expansion of metals production sowing the seeds for catastrophic failure (http://www.ibram.org.br/sites/1400/1457/00000294.pdf There were only 241 tailings dams in Brazil in 2011, 641 by 2012.( see Figure 9 of cited report)
Herculano Tailings Dam Failure Minas Gerais
To Brazils credit and pursuant to the 2010 statutory overhaul, Brazil here has at least made some account of its TSF’s and reported out a few useful publicly available stats. Every nation should publicly report and monitor the condition of all permitted large TSF dams as this Brazil report does. It’s major flaw, however is that it focuses on a common set of measurements amongst all types of dams and neglects to address key elements of risk and function that are use specific. For example, no discussion of upstream construction and rate of raise in tailings dams.
As you may know ICOLD, the International Committee on large dams does not keep track of tailings dams for fear their terrible safety record would reflect adversely on public perception of water dam safety (Kiernan 2016) Brazil here at least counts and gives basic stats on its tailings dams. and we can, because of our deep immersion in the forensics of the worlds mining catastrophes make some inference from these data on the likely massive public liability already formed in these 144 large TSF’s.
It is our advocacy that Tailings Dams require their own risk assessment criteria and their own relevant stats and legal framework. Our advocacy is that every permitting district should have an independent multi -disciplinary Mine Review Board separately appointed and mandated in statute similar to Buildings Departments and the Inspection process for construction of large buildings. They should have the power to act against or in lieu of permitting districts that are not applying existing authority, a reality that is itself a root cause of failure ( BC Auditor Generals’ Report 2016)
This report applies the same size categories to tailings dams as to other dams using the ICOLD definitions where a large dam is defined as a height greater than 15m or a capacity greater than 3 million cubic meters.
In this case the criteria that define a large water dam and those that define a large talings dam are reasonable and fit with the forensics of tailings dam failure. Tailings dams of 3 M cubic meters are capable of catastrophic failure which Dr Chambers & I define as a release > than 1M cubic meters.( Bowker Chambers 2015, Bowker Chambers 2016) In a total failure 30%-50% of contents is generally released.( Rico 2010 and forensics)
In general as a risk analysts and as an anlayst in general, I discourage comparisons between water dams ( purpose built to completion to a known range of flow with materials rigidly meeting specific specifications, retaining only water and build as you go and mostly earthen from materials at hand, retaining various toxic materials and variability in the structural integrity of materials available for each raise. Obviously they are not meaningfully comparable..
We have to reiterate that the meaningless manufactured stat that tailings dams fail at a rate 10 times water dams should be permanently retired and never again cited. by any responsible researcher, journalist or spokes person.
Each type of retention dam should be assessed for its own distinct inherent engineering and failure risks based both on contents and means methods and materials of construction.
To discourage comparisons with water dams it would be better if water dams were separately reported and monitored on standards suitable for them and mine tailings ( and other type of waste and waste dams eg those retaining pregnant leach solutions) were reported and monitored on their own terms for risks specific to the dam type.)
Ignoring the comparisons by dam type in this report ( and resisting any urge to cite them on a comparison basis) here are some of the stats in this 2014 Brazilian government report.
per figure 4:
519 of the 663 Brazilian tailings dams are not “large dams” by definition and therefore also not capable of catastrophic failure as defined in Bowker Chambers 2015 and Bowker Chambers 2016.
144 of the 663 are large dams.
of these 15 have a dam capacity of greater than 75 M cubic meters ( 7 of these > 200 M cubic meter capacity ( see Figure 10 as well)
The distribution of Brazils tailings dams by height is show at figure 12 on page 29
The Fundao was 74 M high at failure and had a capacity of 62 million cubic meters. Mt Polley was 40 m high at failure and had a capacity of 75 million cubic meters ( Chambers Bowker Failure Data Base 2015, revised 2016)
As we learned lately BHP did not even have a geophysicist on staff before the man made catastrophic failure of the Fundao and had no informed risk management policy on the many TSF’s at its mines all over the world.. Samarco had an independent Tailings Review Panel but the effect of that on loss prevention and control is apparent. It s not an exaggeration to consider all 144 large dams in Brazil as at risk .
Dr. Chambers & I know from our deep dive into the forensics of all catastrophic TSF failures recorded history ( Bowker Chambers 2015, Bowker Chambers 2016) that the greatly expanded volume of ore production needed to attain a unit of finished metal as grades fall precipitoulsly worldwide is accommodated mostly by pushing existing TSF’s beyond their design capacity in height, volume, years of useful life and foot print. Chuquicamata , for example, still planning to use its same 43 SqKM 1985 TSF for its current massive underground expansion. At Samarco it was known the Fundao did not have capacity for the planned expansion. Vale who control 144 of Brazils tailings dams ( see Quadra 4 p 51) signed off on that. All knew the Fundao never had its two key engineering requirements for safe operation from the beginning . All knew there was no identified vetted plan for expanding onsite capacity to adequate and soundly manage the volume and rate of tailings deposition that the expansion would require. That happens everywhere. We rarely see a specific focused review of tailings capacity for these big throughput expansions and investors , before Fundao, never thought to ask these questions.
Dr. Chambers and I have been saying for two years now hat the records tells us the first and most urgent regulatory and public attention must be directed to the already accumulated but not yet manifest catastrophic failure in existing permitted operating facilities.( Bowker Chambers 2015, Bowker Chambers 2016)
It is clear that this cannot be a self evaluation and it is clear that Minas Gerais and most permitting districts do not have the competence to either oversee or conduct these re evaluations.
These re evaluations have to reach well beyond the usual routine proximate cause of failure. The world record of failure is now perfectly clear that root causes of catastrophic failure which lay behind these incompetent or just wrong engineering choices cited as “proximate cause” ( Bowker Chambers 2016)
Brazils risk classification and its overhaul of mining law miss what have been identified as key root causes of catastrophic failure( Bowker Chambers 2016)
Brazil needs to stop the clock on its aggressive mine expansion agenda, assess the liabilities they have allowed to accumulate in this massive improperly regulated expansion post 2010 and form a strategy for unwinding that liability. Unfortunately it seems that present law des not provide any response until the destruction happens. They need new law now to specifically address what they will find when they undertake a more meaningful risk assessment than is provided in this 2014 perfunctory report.
Blight, Geoffrey (2010). “Management & Operational Background to Three Tailings Dams Failures in South Africa.” Chapter 42, Slope Stability in Surface Mining, ed. Hustrulid, W.A., McCarter, Kim, Van Zyl, Dirk, Society for Mining Metallurgy and Exploration, e-book, 2010.
Blight, G.E. and A.B. Fourie (2004). “A Review of Catastrophic Flow Failures of Deposits of Mine Waste and Municipal Refuse.” Proceedings International Workshop, “Occurrence and Mechanisms of Flow-like Landslides in Natural Slopes and Earthfills.” Sorrento, 19-36, Picarello (ed), Patron, Bologna
Bowker, L.N. and Chambers, D.M. (2016).” Root Causes of Tailings Dam Overtopping: The Economics of Risk & Consequence” Proceedings of “Solutions 16”, (September 2016 ) ( presently in compilation by Solutions16 meanwhile pre publication copy available from the authors)
Bowker, L.N. and Chambers, D.M. (2015). “The Risk, Public Liability, & Economics of Tailings Storage Facility Failures.” July 21, 2015, http://www.csp2.org/technical-reports, (June 14, 2016).
Chambers D.M. and Bowker L.N.(2016) Tailings Dam Failures 1915-2015 (excel Free publicly downladadble) www.csp2.org
Chambers, D.M. (2014). “Options For Defining Environmental “GO/NO-GO” Zones For Mines.”January2, 2014 http://www.csp2.org/technical-reports, (June 14, 2016). Commissioned Directed By Bowker Associates) http://www.csp2.org/files/reports/Go-NoGo%20Zones%20-%20Chambers%202Jan14.pdf
Eisenhammer, S and Nogueira, M (2016). “Brazil Mining Dam Reforms Unsettle Companies, Do Little for Safety.” Reuters, May 11, 2016
Kiernan, Paul (2016b). “Engineers Say Brazilian Disaster Shows World-Wide Danger from Hoover Dam-Size Earthen Structures Holding ‘Tailings’ Waste.” Wall Street Journal (WSJ), April 5, 2016
Mt Polley Expert Panel (2015). “Report on Mount Polley Tailings Storage Facility Breach.” Independent Expert Engineering Investigation and Review Panel, Province of British Columbia, January 30, 2015.
ADDITIONAL BACKGROUND ON MINING IN BRAZIL
http://www.e-mj.com/features/850-brazil-mining.html?showall=1#.WAUkSWYzUjU Excellent and through Engineering and Minig Journal 2011 article on the economics of mining, economic policy, economic and political history of mining in Brazil( n English)
This thesis which also combines analysis of water dam and tailings dam failures has good text and photos of all of Brazils failed dams (in portuguese)http://www.smarh.eng.ufmg.br/defesas/656M.PDF
Mining.Com’s 2013 profile of 10 largest foreign investments in Brazil Mines..No details on tailings..just thumbnail sketches Anglo America is prominent on the list http://www.mining-technology.com/features/featurethe-10-biggest-foreign-owned-mines-in-brazil-4415407/2
Vale’s History In Its own words .http://www.vale.com/Documents/valehistorybook10.pdf
Vale’s Self Assessment of Possibility of Using Safer Sounder Tailings Storage http://www.paste2013.com/wp-content/uploads/2013/07/Mendes-et-al-Geotechnical-parameters-of-iron-ore-tailings-from-the-Quadrilatero-Ferrifero-after-different-treatments-and-ore-processing.pdf
Lindsay Newland Bowker, CPCU, ARM Environmental Risk Manager
Science & Research In The Public Interest
15 Cove Meadow Rd.
Stonington, Maine 04681
207 367 5145