180
12 Consequences
12.2.3 Outflow Volumes
The other input parameter is the outflow volume, for which many empirical relations
exist, unfortunately each one of them proven wrong to some extent by actual failures,
for various reasons.
Various researchers have attempted to define the volume of the outflow based on
failures history and easy to determine dam characteristics. None of them is satisfactory, but they offer a “first stab” at evaluating runout volumes.
Rico et al. (2008) calculated the Volume of the runout VF using the total
impounded volume (VT) in Mm
3 as in Eq. (12.5)
VF = 0.354 ∗ VT
1.01 R
2
= 0.86
(12.5)
and the outflow run-out distance travelled by the tailings in km (Dmax) is obtained
using VF and the dam’s height (in meters) at the time of failure (H) as in Eq. (12.6)
DMAX = 1.61 ∗ (H ∗ VF)
0.66 R
2
= 0.57
(12.6)
Many analysts/engineers directly use such regression equations in a deterministic way to specify exposure. However, as site conditions vary significantly there is
considerable uncertainty that needs to be quantified. Proof of that is the very low
correlation calculated for Eq. (12.6). We will note that Eq. (12.5) basically states
that 1/3 of the impoundment volume will spill (Tailings and water) through a dam
breach.
Azam and Li (2010) analyzed 218 tailings accidents. Let’s note that they divided
that number of accidents by the number of “mines”—over 18,401—to publish an
accident “frequency” of 1.2% for the hundred years of history they analyze, a quite
absurd value considering that the number of mines does not equate the number
of tailings or active tailings. Nevertheless they agree with our results (Oboni and
Oboni 2013) that the frequency failures peaked in the 1970s–1980s and then declined
afterwards, i.e., into the 2000s. Azam and Li stated that usually about one-fifth of the
contained volume is released. However, they stated that the vast majority of releases
were of unknown volume (see Fig. 6 of their referenced publication).
Case studies of past failures show the total outflow volume will often be less than
the total storage above the bottom elevation of the breach because some portion of
the tailings remains stable in the TSF under their own static strength. This is different
than a water reservoir, where most or all of the stored water above the bottom of the
breach elevation will out flow.
None of the above accounts for liquefaction (see Sect. 9.3) and Rico et al. (2008)
rightly pointed out that some of the parameters contributing to the uncertainty in the
predictions include sediment load, fluid behaviour (depending on the type of failure),
topography, the presence of obstacles stopping the flow, and the proportion of water
stored in the tailings dam (linked to meteorological events or not).
12 Consequences
12.2.3 Outflow Volumes
The other input parameter is the outflow volume, for which many empirical relations
exist, unfortunately each one of them proven wrong to some extent by actual failures,
for various reasons.
Various researchers have attempted to define the volume of the outflow based on
failures history and easy to determine dam characteristics. None of them is satisfactory, but they offer a “first stab” at evaluating runout volumes.
Rico et al. (2008) calculated the Volume of the runout VF using the total
impounded volume (VT) in Mm
3 as in Eq. (12.5)
VF = 0.354 ∗ VT
1.01 R
2
= 0.86
(12.5)
and the outflow run-out distance travelled by the tailings in km (Dmax) is obtained
using VF and the dam’s height (in meters) at the time of failure (H) as in Eq. (12.6)
DMAX = 1.61 ∗ (H ∗ VF)
0.66 R
2
= 0.57
(12.6)
Many analysts/engineers directly use such regression equations in a deterministic way to specify exposure. However, as site conditions vary significantly there is
considerable uncertainty that needs to be quantified. Proof of that is the very low
correlation calculated for Eq. (12.6). We will note that Eq. (12.5) basically states
that 1/3 of the impoundment volume will spill (Tailings and water) through a dam
breach.
Azam and Li (2010) analyzed 218 tailings accidents. Let’s note that they divided
that number of accidents by the number of “mines”—over 18,401—to publish an
accident “frequency” of 1.2% for the hundred years of history they analyze, a quite
absurd value considering that the number of mines does not equate the number
of tailings or active tailings. Nevertheless they agree with our results (Oboni and
Oboni 2013) that the frequency failures peaked in the 1970s–1980s and then declined
afterwards, i.e., into the 2000s. Azam and Li stated that usually about one-fifth of the
contained volume is released. However, they stated that the vast majority of releases
were of unknown volume (see Fig. 6 of their referenced publication).
Case studies of past failures show the total outflow volume will often be less than
the total storage above the bottom elevation of the breach because some portion of
the tailings remains stable in the TSF under their own static strength. This is different
than a water reservoir, where most or all of the stored water above the bottom of the
breach elevation will out flow.
None of the above accounts for liquefaction (see Sect. 9.3) and Rico et al. (2008)
rightly pointed out that some of the parameters contributing to the uncertainty in the
predictions include sediment load, fluid behaviour (depending on the type of failure),
topography, the presence of obstacles stopping the flow, and the proportion of water
stored in the tailings dam (linked to meteorological events or not).