Earthquake Hazard Annual Exceedance Probabilities

DOE-STD-1020-2002

C.3.1

Earthquake Hazard Annual Exceedance Probabilities

Historically, non-Federal Government General Use and Essential or Low Hazard

facilities located in California, Nevada, and Washington have been designed for the seismic

hazard defined in the Uniform Building Code. Other regions of the U.S. have used the UBC

seismic hazard definition, other building code requirements, or have ignored seismic design.

Past UBC seismic provisions (1985 and earlier) are based upon the largest earthquake intensity

that has occurred in a given region during about the past 200 years. These provisions do not

consider the probability of occurrence of such an earthquake and thus do not make any explicit

use of a probabilistic seismic hazard analysis. However, within the last 20 years there have been

developments in building codes in which the seismic hazard provisions are based upon a

consistent annual probability of exceedance for all regions of the U.S. In 1978, ATC-3 provided

probabilistic-based seismic hazard provisions (Ref. C-1). From the ATC-3 provisions, changes

to the UBC (Ref. C-2) and the development of the National Earthquake Hazards Reduction

Program (NEHRP, Ref. C-3) have resulted. A probabilistic-based seismic zone map was

incorporated into the UBC beginning with the 1988 edition. Canada and the U.S. Department of

Defense have adopted this approach (Refs. C-4 and C-5). The suggested annual frequency of

exceedance for the design seismic hazard level differs somewhat between proposed codes, but all

lie in the range of 10-2 to 10-3. For instance, UBC (Ref. C-2), ATC-3 (Ref. C-1), and NEHRP

(Ref. C-3) have suggested that the design seismic hazard level should have about a 10 percent

frequency of exceedance level in 50 years which corresponds to an annual exceedance frequency

of about 2x10-3. The Canadian building code used 1x10-2 as the annual exceedance level for

their design seismic hazard definition. The Department of Defense (DOD) tri-services seismic

design provisions for essential buildings (Ref. C-5) suggests a dual level for the design seismic

hazard. Facilities should remain essentially elastic for seismic hazard with about a 50 percent

frequency of exceedance in 50 years or about a 1x10-2 annual exceedance frequency, and they

should not fail for a seismic hazard which has about a 10 percent frequency of exceedance in 100

years or about 1x10-3 annual exceedance frequency. Recently the IBC 2000 has adopted use of

USGS maps for 2% exceedence probability in 50 years based on NEHRP 1997 provisions.

These are being incorporated in this standard for PC-1 and PC-2 facilities.

On the other hand, nuclear power plants are designed so that safety systems do not fail

if subjected to a safe shutdown earthquake (SSE). The SSE generally represents the expected

ground motion at the site either from the largest historic earthquake within the tectonic province

within which the site is located or from an assessment of the maximum earthquake potential of

the appropriate tectonic structure or capable fault closest to the site. The key point is that this is

a deterministic definition of the design SSE. Recent probabilistic hazard studies (e.g., Ref. C-6)

have indicated that for nuclear plants in the eastern U.S., the design SSE level generally

corresponds to an estimated annual frequency of exceedance of between 0.1x10-4 and 10x10-4 as

is illustrated in Figure C- 4. The probability level of SSE design spectra (between 5 and 10 hz) at

the 69 eastern U.S. nuclear power plants considered by Ref. C-6 fall within the above stated

range. Figure C-4 also demonstrates that for 2/3 of these plants the SSE spectra corresponds to

probabilities between about 0.4x10-4 and 2.5x10-4. Hence, the specified hazard probability level

C-14