ORE2-Tailings allow benchmark of tailings dams probability of failure

Today we discuss how ORE2-Tailings allow benchmark of tailings dams probability of failure.

The considered mine for this case history has a portfolio of 15 dams.

The figure below shows in the vertical axis the annual Probability of failure (pf) results under the form of yellow vertical bars. The extreme of the bars represent the minimum and maximum estimates of the probability of failure for each dam. Hence, the length of the bars measures the uncertainty on the evaluation of the probability at the time of the assessment and with the data available at that time.

The horizontal axis shows the dams of a real life sample dam portfolio. The four left green bars correspond to benchmarking values, namely: Mount Polley and Samarco annual probability of failure evaluated with the ORE2-Tailings methodology, the min-max values of the worldwide portfolio based on historic records and finally the values obtained by a Doctoral Thesis at UBC which attempted a theoretical estimate of the annual probability of failure of standard and dewatered tailings. The vertical axis bears the annual probability of failure. For each structure a yellow bar depicts the Uncertainty related to the probability estimate.

The horizontal axis shows the names of the dams in the dam portfolio.

ORE2-Tailings allow benchmark of tailings dams probability of failure

The horizontal axis shows four benchmark values, namely:

• Mount Polley and Samarco annual pf evaluated with ORE2-Tailings 2.0 considering data available to us BEFORE the respecive failures,
• the min-max values of the world-wide portfolio based on historic records, materialized by the two horizontal lines crossing the graph, and finally
• the values obtained by a Thesis at UBC which attempted a theoretical estimate of the annual pf of standard and dewatered tailings.

Results of a ORE2-Tailings case history

ORE2-Tailings allow tailings dams benchmark of probability of failure which show that in the considered case there are dams below the historic benchmark.

Some dams overlap the benchmarks limits and some are above the upper limit. In this portfolio all dams have significantly lower probabilities of failure than Mount Polley or Samarco. We estimated those after the failures, but using data available before the catastrophes.

Of course, additional studies and information, additional monitoring would allow to narrow the uncertainties (length of the yellow bars). If data are acquired through classic monitoring of Space Observation, they can be seamlessly integrated in the ORE2-Tailings analysis and deliver an updated probability of failure for each dam.

• Effects of mitigation and other actions and inaction
• Mitigation implemented on dams will push the respective bars down. Finally,
• Long term lack of maintenance, climate change effects will tend to push the bars up.

The combination of the graph above with the cost of consequences functions for each dam gives the risks. Consequences will be derived performing dam break analyses and then business interruption, direct & indirect losses, social and legal impacts, etc.

Like we have stated already, cost of consequences are multidimensional additive functions.

Closing remarks

It is time for mining companies, governmental agencies to benefit from the results of a ORE2-Tailings deployment.

The architecture of the ORE2-Tailings database allows to drill through the data. That ultimately may reveal trends and patterns that cannot be detected when using other methods.

Users start from day one with a “lesson learned”, “historic view” of hundred years using data gathered through various sources. Additionally, thanks to the architecture and specific methodology, ORE2-Tailings will allow each user to build their own historic and lesson learned database.

Tagged with: benchmark of tailings dams, tailings dams probability of failure

Category: Consequences, Probabilities, Risk analysis, Risk management, Uncategorized