Research Activities


A Methodology for Estimating the Risk of Post-Earthquake Hazardous Materials Release:

Pilot Application to the County of Los Angeles

by H.A. Seligson, R.T. Eguchi and K.J. Tierney

This article presents results from a study performed by NCEER (in conjunction with funding from the National Science Foundation) to develop a methodology to assess the risk of earthquake-induced hazardous materials release and associated population impacts. The methodology combines seismic hazard analyses with fragility modeling of facilities handling hazardous materials, and data on airborne toxic releases, to estimate the potential population exposure to post-earthquake release. A demonstration application has been performed for the County of Los Angeles. Full documentation will be presented in a forthcoming NCEER technical report. Questions and comments should be directed to Hope Seligson, EQE International, at (714)833-3303; or email: has@eqe.com.

While recent earthquake disasters (1994 Northridge and 1995 Kobe) have produced few documented occurrences of hazardous materials release, failures in previous events, such as the release of chlorine gas from a chlorine repackaging facility in the 1987 Whittier Narrows earthquake (FEMA, 1987), indicate that even one such occurrence may place significant demands on limited emergency resources. The objective of this project was to develop a methodology that would enable local jurisdictions to determine the magnitude of the problem and identify areas most susceptible to earthquake-induced hazardous materials release. Reporting and permitting requirements may facilitate implementation of such a model; extensive chemical inventories are often collected locally, usually by the Fire Department. With Los Angeles County selected as the demonstration area, this assessment was limited to facilities using ammonia and/or chlorine. The chemical inventory utilized for this study was obtained from a survey conducted by the South Coast Air Quality Management District (1985).

The generalized methodology as developed for this study is presented in figure 1. The five major steps include:

The results enable local emergency managers to prepare for and mitigate potential earthquake-induced releases. This paper will concentrate on the facility modeling and vulnerability aspects of the methodology, since seismic hazard analysis is a well-documented topic.

Hazardous materials in urban areas are widespread and diverse. This pilot application opted to examine potential impacts of airborne releases of anhydrous ammonia and chlorine. Inventory requirements for the methodology include location, quantity of material stored, and classification as either processing and/or storage facilities. Twenty-two facilities were included in the pilot study, storing between 4 and 1000 tons of chlorine, and as much as 206 tons of ammonia. Facility locations are shown in figure 2; facility usage and on-site storage data are given in table 1.

Table 1 :Chemical Facility Use and Storage - Los Angeles County
Facility Facility Type Chemical storage
Chlorine
Storage
Ammonia
Storage
Ammonia
Processing
Chlorine
(Tons)
Ammonia
(Tons)
1
2
3
4
5
x
x
x
x
x

x
x
x
x
4
32
8
12
180
40
57
26
206
6
7
8
9
10
x
x
x
x




x

x
5
10
450
5

15


26
11
12
13
14
15
x
x
x
x
x

x


x



x
454
1000
25
20
270

14

15
1
16
17
18
19
20
x
x

x
x


x



x
90
48

10
6


26
10
21
22
x x
x

24 2
100
Total 19 6 7 2653
Tons
538 Tons

Two "generic" facility models were developed to allow for generalized application of fragility models; a chemical processing model (including storage vessels, reactors, piping and a separator/regenerator), and a storage model (consisting of storage vessels and piping only). A review of typical facility design resulted in the representation of the processing components with "typical" vessels and piping (a 20 ton reactor, a 1 ton separator vessel, and 3-inch diameter piping), while the make-up of the storage components was determined by typical vessel sizes and the reported amount of on-site storage. For example, a processing facility storing 4 tons of chlorine and 40 tons of ammonia would be assumed to have a 20 ton reactor, a 1 ton separator, two 2-ton chlorine storage vessels, and one 50-ton ammonia storage vessel. This allowed for development of a limited number of fragility models to represent the possible range of components.

Damage probability matrices for the various components were developed through the use of expert judgment. Results indicate that the most vulnerable components were horizontal (large quantity) storage vessels and reactor vessels. Critical failure modes for these vessels would likely involve failure of connecting piping, rather than failure of the vessels themselves.

The size and shape of the area exposed to a potential earthquake-induced hazardous materials release were estimated through a chemical dispersion analysis. The results provide an estimate of the zone of vulnerability, or area in which specific health criteria may be exceeded for a given release and meteorological condition. Selected health criteria were based on the Emergency Response Planning Guidelines (ERPG) of the American Industrial Hygiene Association (1988). For this study, ERPG-3 - "the maximum airborne concentration below which it is believed that nearly all individuals could be exposed for up to one hour without experiencing or developing life-threatening effects" - was chosen. This exposure level is 20 ppm for chlorine and 1000 ppm for ammonia.

Three release types were simulated; an instantaneous release from a catastrophic storage vessel failure, a continuous release from a piping failure, and a finite duration release from failed piping. For each release mode, a zone of vulnerability or hazard footprint was determined. As a conservative estimate, the composite maximum width and length were taken to represent a generalized footprint for each release mode. Although the hazard footprints were sometimes irregular, varying from tear-drop shape to circular, the hazardous materials plumes were modeled as ellipses for the purpose of this study. A probabilistic model was developed to determine the likelihood that a given population center would be within a hazardous material plume. Derivation of the probability model is documented in the forthcoming report.

Population data from the 1980 census, aggregated by enumeration district, were utilized for this study. For each enumeration district, a population count is associated with a representative geographic point location. Within Los Angeles County, approximately 7.5 million residents were associated with 6,500 enumeration districts, with an average population of about 1,150 people.

Regional analysis was accomplished through development of a computer program - Program "PLUME." Based on estimated ground shaking intensity at each hazardous materials facility, PLUME calculated the probability of failure in each failure mode for each facility component. Dimensions of the resultant plume were estimated. For each population center, the likelihood of being within a given plume was determined. Individual probabilities were aggregated to determine the probability of hazardous materials exposure at ERPG 3 for each population center.

For the 22 selected sources of ammonia and chlorine within Los Angeles County, the following results have been estimated:

The results of this study highlight the potential hazard posed by storage of large quantities of chlorine and ammonia in areas expected to suffer strong ground shaking. Chlorine may be stored in vessels as large as 90-ton rail cars, whose failure plumes can extend over seven miles. The identification of chlorine storage as a major threat enables users to address the risk by concentrating efforts on improving performance of existing vessels, developing smaller, safer vessels, or perhaps relocating major storage facilities.

References

AIHA (1988), "Emergency Response Planning Guidelines," American Industrial Hygiene Association Emergency Response Planning Guideline Committee, Akron, Ohio.

FEMA (1987), "The Los Angeles-Whittier Narrows Earthquake of October 1, 1987; Federal/State Hazard Mitigation Survey Team Report," prepared by the Federal Emergency Management Agency, Region IX; California Governor's Office of Emergency Services, Southern California Earthquake Preparedness Project and Planning Division.

South Coast Air Quality Management District (1985), "Information on Firms Surveyed by AQMD in 1985."

Click here to return to the Table of Contents


| Home | NCEER | Information Assistance | Publications | Databases, Software |
| Events | New! | Links | Data Resources |