Evolution of geotechnical risk analysis in North American practice - Risk and Reliability in Geotechnical Engineering - page 468

12.6 enVIronMental reMeDIatIon (1980-1995)
The beginning date of 1980 is the year Superfund was passed by the U.S. Congress
(P.L. 96-510, 42 U.S.C., December 11, 1980). Superfund —more formally known as the
Comprehensive Environmental Response, Compensation, and Liability Act of 1980
(CERCLA)—authorized U.S. Government agencies to recover natural resource damages
caused by releases of hazardous substances. This led to a spate of site investigations and
plans for site remediation across the country and to an equally large number of lawsuits and
enforcement actions. The ending date, 1995, is the year of expiration of the Superfund Tax
on oil and chemical companies for cleaning up abandoned sites.
The Superfund era led to at least two innovative developments in geotechnical risk analy-
sis: a focus on spatial sampling methodologies for characterizing subsurface soil properties,
especially those properties associated with chemicals rather than engineering mechanics, and
the creation of a literature on probabilistic geohydrology models by which to forecast fate
and transport through groundwater and to characterize the uncertainty in these forecasts.
Spatial sampling methodologies received considerable advancement during the environmen-
tal remediation (ER) period especially under the support of the U.S. Environmental Protection
Agency (USEPA). The agency mandates health-risk analysis for engineering options based on
risk standards, for example, for municipal landfills and for Superfund cleanups. This created
a need for statistically valid assessments of site properties and to guidance on sampling plans
and statistical data analysis (USEPA 1991, 2002). A number of textbooks appeared during
this period on statistical methods of environmental site characterizations (Ott 1995). This
period also saw a dramatic increase in the use of “geostatistics” for soils sampling.
Geohydrology was another important area of development in the ER era, leading to the
evolution of a suite of probabilistic tools for modeling subsurface flows, along with several
textbooks on stochastic approaches (Marsily 1986; Kitanidis 1997; Rubin 2003).
12.7 DaM SaFetY (1986-ongoIng)
Prior to the failure of the Teton Dam in 1976, little attention was paid to probabilistic char-
acterizations of embankment reliability or to quantitative risk analysis of dam safety. With
the passage of the Reclamation Safety of Dams Act (P.L. 95-578) in 1978, and Executive
Order 12148 requiring U.S. federal agencies to implement federal guidelines for dam safety,
the US Bureau of Reclamation (USBR) began a concerted effort to address dam safety, and
eventually, through the lens of risk analysis. The act was amended in 1984, 2000, 2002, and
2004; the program continues to operate. The starting date for this period is the year that
USBR began publicly reporting on its emerging risk analysis approach.
In the mid-1980s (Parrett 1986), the USBR risk cadre pioneered the use of risk analysis in
dam safety in the United States, championing a number of methodological approaches that
would come to characterize dam safety evaluations: event tree analysis, subjective probabil-
ity assignment, quantified loss-of-life modeling, and tolerable risk criteria, among others.
Publication of the Public Protection Guidelines in 1997 was a benchmark in this progres-
sion (USBR 1997). BCHydro and Ontario Power Generation working through the Dam
Safety Interest Group with USBR were also early promoters of risk analysis for dam safety
(Salmon and Hartford 1995), as was ANCOLD that created a working group on risk analy-
sis in 1987 (McDonald 1995), and the Dam Safety Committee of the Government of New
South Wales promulgated risk analysis approaches starting in the 1990s.
Dam safety continues to be a test bed for developments in geotechnical risk analysis.
Following Hurricane Katrina in 2005, the U.S. Army Corps of Engineers (USACE) converted