15.11 Risk Communications
261
• perceives mismanagement, incompetence, and lack of conscientiousness as the
central issues in accidents (Bouder et al. 2007).
Moreover, academic and popular literature suggest agreement that the public’s
distrust of industry has developed over the past half-century as a result of repeated
failures to provide adequate and/or accurate risk information to the public and, in
some cases, blatant or perceived conflict of interest (Oboni et al. 2013).
At the other end of the spectrum of inaccurate information we find sensationalism, especially if unsupported by facts, and the result of biased interpretation
of incomplete databases or “gut feeling” (see Sect. 15.8). This is obviously not a
recommendable path when communicating risks. Considering that today modern
dams, especially the largest ones, are better investigated, designed, built and managed/monitored that any older structure, based on our experience, we see their probability of failure decrease towards a value close to those of hydraulic structures of
the same caliber. The portfolio studied in this chapter is an eloquent demonstration
of this statement.
Just as a reminder, many catastrophic failures in the last 45 years occurred for
medium-height dams, storing base and precious metals tailings. Samarco was an
exception with its 100 m height, but its overall quality was poor, as highlighted by
forensic studies, with design changes and an extremely high average raise per annum.
The recently failed Corrego dam (near Bernardinho, Brazil) had the longest life to
failure of the group (43 years), Samarco and Baia Mare the shortest. In all these
failed structures, investigations, design, construction, operations and management
lead them far astray from what would be considered excellent quality in a large dam
today.
This leads to us recommend avoiding any sensationalist statements about risks,
as the interplay of probability and consequences necessarily means neither that risks
of a large dam are always bigger nor that smaller dams have more tolerable failures.
Additionally, and to confirm the above, we unambiguously showed in (Caldwell
et al. 2015) that the graph relating volume released to lives lost does not show any
correlation. The same would be true of the dam height or the run-out distance or
outflow volume with the same conclusions!
In summary, explaining what is planned, is now being done, and is foreseen to
mitigate risks is at least as important as explaining how small those risks may be. The
public may be much more receptive if mitigative plans come first and the statement
“trust us, risks are small” is avoided, as we pointed out in Sect. 1.1. In conclusion:
• Simplistic statements equating bigger dams with bigger risks are to be left aside
as they can easily be disproven and do not help any constructive dialogue.
• As structures become larger, their probability of failure should go down towards
the values commonly accepted for major hydro-dams. That reduction necessitates
action on all aspects of the dam investigation, design, construction, monitoring,
inspections and management, as all contribute to the chance of failure. Increasing
the FoS does not clearly reflect any of the above necessary improvements.
• Population increase, land use shifts, and environmental constraints increase the
consequences of failures and therefore tend to increase risks. Again, the probability
261
• perceives mismanagement, incompetence, and lack of conscientiousness as the
central issues in accidents (Bouder et al. 2007).
Moreover, academic and popular literature suggest agreement that the public’s
distrust of industry has developed over the past half-century as a result of repeated
failures to provide adequate and/or accurate risk information to the public and, in
some cases, blatant or perceived conflict of interest (Oboni et al. 2013).
At the other end of the spectrum of inaccurate information we find sensationalism, especially if unsupported by facts, and the result of biased interpretation
of incomplete databases or “gut feeling” (see Sect. 15.8). This is obviously not a
recommendable path when communicating risks. Considering that today modern
dams, especially the largest ones, are better investigated, designed, built and managed/monitored that any older structure, based on our experience, we see their probability of failure decrease towards a value close to those of hydraulic structures of
the same caliber. The portfolio studied in this chapter is an eloquent demonstration
of this statement.
Just as a reminder, many catastrophic failures in the last 45 years occurred for
medium-height dams, storing base and precious metals tailings. Samarco was an
exception with its 100 m height, but its overall quality was poor, as highlighted by
forensic studies, with design changes and an extremely high average raise per annum.
The recently failed Corrego dam (near Bernardinho, Brazil) had the longest life to
failure of the group (43 years), Samarco and Baia Mare the shortest. In all these
failed structures, investigations, design, construction, operations and management
lead them far astray from what would be considered excellent quality in a large dam
today.
This leads to us recommend avoiding any sensationalist statements about risks,
as the interplay of probability and consequences necessarily means neither that risks
of a large dam are always bigger nor that smaller dams have more tolerable failures.
Additionally, and to confirm the above, we unambiguously showed in (Caldwell
et al. 2015) that the graph relating volume released to lives lost does not show any
correlation. The same would be true of the dam height or the run-out distance or
outflow volume with the same conclusions!
In summary, explaining what is planned, is now being done, and is foreseen to
mitigate risks is at least as important as explaining how small those risks may be. The
public may be much more receptive if mitigative plans come first and the statement
“trust us, risks are small” is avoided, as we pointed out in Sect. 1.1. In conclusion:
• Simplistic statements equating bigger dams with bigger risks are to be left aside
as they can easily be disproven and do not help any constructive dialogue.
• As structures become larger, their probability of failure should go down towards
the values commonly accepted for major hydro-dams. That reduction necessitates
action on all aspects of the dam investigation, design, construction, monitoring,
inspections and management, as all contribute to the chance of failure. Increasing
the FoS does not clearly reflect any of the above necessary improvements.
• Population increase, land use shifts, and environmental constraints increase the
consequences of failures and therefore tend to increase risks. Again, the probability