15.5 Probability of Failure of the Portfolio’s Dams
243
Table 15.9 (continued)
1.7
6.74 * 10
−6
1.14 * 10 −2
1.83
1.17 * 10
−5
1.84 * 10
−2
2.37
1.97 * 10
−4
5.96 * 10 −2
4
Refer to Tables 11.1 and
11.2 in Sect. 11.1.2: FoS
average 1.17
f
evaluations
3
n/a
n/a
3.2
n/a
n/a
3.3
n/a
n/a
3.8
n/a
n/a
used in p
6.60 * 10
−4
to
2.06 * 10
−3
1.16 * 10
−3
to
3.52 * 10
−3
7.00 * 10
−3
to
1.65 * 10
−2
1.94 * 10
−1
2.30 * 10
−1
2.50 * 10
−1
3.80 * 10
−1
dams. However, the SLM method is a much more deterministic than ORE2_Tailings
in the sense that the range of probability of failure is narrower. Also, for a few cases
the SLM method gives completely different values, as it lacks critical data input
specific to tailings dams.
Finally, it is easy to verify that the annual probability of pseudo-static failure—i.e.,
the pseudo-static p f divided by the return—is in the particular case of this portfolio
always either lower than or of the same order of magnitude as the p f ESA (see
Sect. 11.2.3 for some caveats).
Thus, in Fig. 15.6 we have only included the data of ESA with the probability of
failure as a function of a FoS. Figure 15.6 shows a comparison between the SLM
and ORE2_Tailings results for the probability of failure as a function of the FoS
determined by the engineers (Table 15.6) for each dam of the portfolio.
As already stated earlier regarding Dams 1A and 1B, it is worth noting that the
SLM method does not contain any provision for water balance or freeboard, as it
is a general slope methodology, while ORE2_Tailings includes those factors and
many more, dam-specific ones. ORE2_Tailings compute those probabilities mainly
because of the short, relatively widely-spaced boreholes performed during the investigations. The same is true for for Dam 1B (see Sect. 15.5.1 for more details).
In the case of Dam 2, where there are few tailings dam-specific features, the
correspondence of SLM with ORE2_Tailings is very good.
For Dam 3: the SLM method does not consider liquefaction, the divergence with
ORE2_Tailings here is again significant (see Sect. 15.5.3 for more details).
For Dam 4: ORE2_Tailings penalizes the probability of failure as it is characterized by the paucity of data and ignorance on many aspects of its history.
The SLM method does not look at specific tailings dams deficiencies.
243
Table 15.9 (continued)
1.7
6.74 * 10
−6
1.14 * 10 −2
1.83
1.17 * 10
−5
1.84 * 10
−2
2.37
1.97 * 10
−4
5.96 * 10 −2
4
Refer to Tables 11.1 and
11.2 in Sect. 11.1.2: FoS
average 1.17
f
evaluations
3
n/a
n/a
3.2
n/a
n/a
3.3
n/a
n/a
3.8
n/a
n/a
used in p
6.60 * 10
−4
to
2.06 * 10
−3
1.16 * 10
−3
to
3.52 * 10
−3
7.00 * 10
−3
to
1.65 * 10
−2
1.94 * 10
−1
2.30 * 10
−1
2.50 * 10
−1
3.80 * 10
−1
dams. However, the SLM method is a much more deterministic than ORE2_Tailings
in the sense that the range of probability of failure is narrower. Also, for a few cases
the SLM method gives completely different values, as it lacks critical data input
specific to tailings dams.
Finally, it is easy to verify that the annual probability of pseudo-static failure—i.e.,
the pseudo-static p f divided by the return—is in the particular case of this portfolio
always either lower than or of the same order of magnitude as the p f ESA (see
Sect. 11.2.3 for some caveats).
Thus, in Fig. 15.6 we have only included the data of ESA with the probability of
failure as a function of a FoS. Figure 15.6 shows a comparison between the SLM
and ORE2_Tailings results for the probability of failure as a function of the FoS
determined by the engineers (Table 15.6) for each dam of the portfolio.
As already stated earlier regarding Dams 1A and 1B, it is worth noting that the
SLM method does not contain any provision for water balance or freeboard, as it
is a general slope methodology, while ORE2_Tailings includes those factors and
many more, dam-specific ones. ORE2_Tailings compute those probabilities mainly
because of the short, relatively widely-spaced boreholes performed during the investigations. The same is true for for Dam 1B (see Sect. 15.5.1 for more details).
In the case of Dam 2, where there are few tailings dam-specific features, the
correspondence of SLM with ORE2_Tailings is very good.
For Dam 3: the SLM method does not consider liquefaction, the divergence with
ORE2_Tailings here is again significant (see Sect. 15.5.3 for more details).
For Dam 4: ORE2_Tailings penalizes the probability of failure as it is characterized by the paucity of data and ignorance on many aspects of its history.
The SLM method does not look at specific tailings dams deficiencies.