• Military

• WMD Menu                           

• Introduction
• Systems
• Facilities
• Agencies
• Operations
• Countries
• Hot Documents
• News
• Reports
• Policy
• Budget
• Congress
• Links

• Intelligence
• Homeland Security
• Space

D.5 MULTIPLE-BUILDING EVENT

This section addresses the potential releases and consequences of a situation involving multiple source terms (both radiological and chemical) stemming from a single event affecting the combined areas of LLNL Livermore site and SNL, Livermore. As described in Appendix I, an earthquake with a return period of 5000 years was postulated as the initiator for this accident scenario. Buildings expected to survive the 0.8g earthquake were examined for survivability during a }0.9g earthquake. As a rough comparison the January 24 and January 27, 1980, Livermore earthquakes, recorded as 5.4 and 5.6 Richter Magnitude events, generated maximum measured peak ground accelerations of 0.26g at a distance of 18km from the epicenter. The October 17, 1989 Loma Prieta earthquake, recorded as a 7.1 Richter Magnitude event, generated maximum measured peak ground accelerations of 0.68g at a distance of 7km from the epicenter.

D.5.1 Building Selection and Assumptions

The selection processes described for radiological buildings and sources in section D.2.1 and for chemical buildings and sources in section D.3.1 are also the basis for buildings selected for seismic analysis as defined in Appendix I. The results of the analysis presented in Appendix I, Table I-4 are summarized here in Table D.5-1. In all cases, buildings were evaluated based on a 0.8g earthquake which is the design basis for a high hazard facility. Those buildings not meeting analysis criteria were assumed to fail containment and were candidates for possible releases. Buildings that maintained containment during the 0.8g earthquake were analyzed for failures due to a 0.9g earthquake. All buildings analyzed for the 0.9g event are expected to maintain containment with the exception of the Building 968 equipment room. This new damage does not result in any additional release of tritium.

In addition to those buildings identified as having a potential release initiated by an earthquake, all buildings identified for accident analysis were also subjected to seismic analysis. In some instances, the postulated scenario cannot be initiated by a seismic event, and the locations and associated releases were not considered as part of the multiple-building event scenario. One example of this type of scenario is the Building 493 uranium fire. For this scenario a large source of fuel is required to oxidize and disperse the uranium. During the multiple-building event, such a large fuel source is not present; therefore, that location has been eliminated from consideration.

For each location chosen for seismic analysis, specific equipment associated with postulated releases was evaluated. As an example, the overhead crane in Building 625 was found to be in danger of falling off its tracks during a severe earthquake; therefore, the accident scenario for this location involves the overhead crane. In cases where the structure of a building is likely to be damaged, no credit is taken for fire protection, ventilation, or other mitigating features in limiting the amount of radioactive materials or chemicals released.

For the cases analyzed, a secondary fire was eliminated from consideration because of the installation of seismic shutoff valves throughout the natural gas pipeline system and the limited amounts of combustible and flammable materials in the evaluated areas. This does not mean that an earthquake of this severity will not cause major fires at the various facilities. After the 1989 Loma Prieta earthquake, many fires burned uncontrolled in the city of San Francisco due to the failure of natural gas pipelines. The major cause for failure of the pipelines was the nature of the ground (landfill) in the affected areas. Specific information concerning the seismic stability of the area surrounding LLNL Livermore site and SNL, Livermore is contained in Appendix K. While fires may result from an earthquake such as that postulated for the initiating event in this section, the number and magnitude of the fires would not be expected to be as severe as those experienced in 1989. The fires would generally be expected to involve offices and administrative areas where fire loadings are higher than in rated buildings and where fire suppression capabilities are generally not as extensive.

Table D.5-1 Potentials for Loss of Confinement Due to Postulated 0.8g and 0.9g Earthquakes

^a Seismic documentation inadequate for complete extrapolation to current criteria.
^b Future use only; seismic analysis not conducted.
^c Material not affected by earthquake.

D.5.2 Description of Potential Releases Following a 0.8g Earthquake

This section provides a general description of the radiological and chemical releases that may occur as a direct result of a 0.8g earthquake (Appendix I). Scenarios and consequences are discussed in general terms only. For specific information concerning individual scenarios, refer to the referenced sections.

D.5.2.1 Radiological Releases

The buildings selected for seismic analysis are those identified during the selection process (section D.2.1). For some releases, the only initiator postulated is an earthquake; for others, the release cannot be initiated by an earthquake. Brief descriptions of the radiological releases are presented in this section.

A release of americium-241 was postulated during an earthquake from the unhardened portion of Building 251, Heavy Element Facility (section D.2.8.5). Seismic analysis has shown that this portion of the building would not survive a 0.8g earthquake. The dominant exposure pathway, inhalation, would result in a 70-year CEDE of 0.14 rem at the western site boundary and a 70-year collective CEDE of 426 person-rem for the exposed population in the western sector. The hardened portion was examined for damage due to a }0.9g earthquake and was found to survive.

Building 298, Fusion Target Fabrication Facility (section D.2.8.10), is part of the proposed action for the LLNL Livermore site. In support of Inertial Confinement Fusion (ICF) tritium research, tritium handling and target gas filling will be undertaken. A 5 g tritium research/target fabrication facility is proposed. Although the inventory of tritium in the building would be less, the entire administrative limit of 5 g tritium was assumed to be at risk for the purpose of this accident analysis. This at-risk quantity is assumed to be released at ground level. The release would result in a 70-year CEDE of 0.11 rem at the western boundary and a 70-year collective CEDE of 420 person-rem for the exposed population in the western sector.

A tritium release is also postulated from Building 331, Hydrogen Research Facility (section D.2.8.4). This facility is not expected to withstand a 0.8g earthquake. The release would be expected to involve 3.5 g of tritium gas. The release would result in a 70-year CEDE of 0.060 rem at the western boundary and a 70-year collective CEDE of 290 person-rem for the exposed population in the western sector.

Seismic review has shown that Building 332, the Plutonium Facility, would survive the postulated 0.8g earthquake and retain it's safety features which contain radioactive materials. This facility was examined for damage due to a }0.9g earthquake and was found to survive.

Building 391, NOVA-Upgrade/National Ignition Facility (section D.2.8.10), is part of the proposed action for the LLNL Livermore site. The facility will be used to advance the understanding and demonstrate the scientific feasibility of inertial confinement fusion. Fuel pellets containing tritium will be provided by Building 298 or from an offsite source. The inventory limit for this building is 5 g of tritium. An accidental release from Building 391 would be bounded by that from Building 298 with respect to site boundary close.

A fire that oxidizes 5000 kg of natural uranium was postulated for Building 493, Separator Support Facility (section D.2.8.7). This scenario requires a large source of fuel and cannot be initiated by an earthquake.

A fire involving a drum of TRU waste was postulated for Building 612, Waste Handling and Storage Facility (section D.2.8.6). Seismic review has shown that this facility would survive a 0.8g earthquake. Since the facility will survive, no release is modeled for an earthquake initiator. This facility was examined for damage due to a }0.9g earthquake and was found to survive.

Building 625 is not expected to survive the postulated earthquake. A release involving 8 drums of radioactive wastes with 6 curies of americium-241 was postulated (section D.2.8.8). The dominant exposure pathway, inhalation, would result in a 70-year CEDE of 0.27 rem at the western site boundary or 70-year collective CEDE of 4033 person-rem for the exposed population in the western sector.

A tritium release is again postulated from Building 968, Tritium Research Laboratory (section D.2.8.2). This facility is not expected to withstand a 0.8g earthquake. The release from this facility would be expected to involve 50 g of tritium gas. The release would result in a 70-year CEDE of 0.92 rem at the western boundary and a 70-year collective CEDE of 4217 person-rem for the exposed population in the western sector.

The bounding release scenario would occur under existing conditions. Significant tritium inventory reductions are planned for Buildings 968 and 331. These reductions would be completed before inventory increases occurred in Buildings 298 and 391. Therefore, the total radioactive material released as a result of the postulated 0.8g earthquake from Buildings 968, 331, 251, and 625, would produce a collective population dose of 9000 person-rem (Table D.5-2). For the general population of 1,417,586, this seismic event would result in 5 cases of fatal cancer and a total health detriment (as defined in section D.2.9) of 7 in the western sector.

Most of the smaller radionuclide sources that are not treated in the above scenarios are solid sealed sources that will not be released as part of a multiple-building event. However, two additional smaller tritium sources, 160 Ci and 200 Ci, were not explicitly treated but can be shown to contribute a negligible additional amount to the radiological health risk, as follows: The tritium source (200 Ci) in Building 231 is only 400 m from a nearest site boundary and gives a resultant dose of approximately 8×10^-5 rem per tritium Ci. These unaccounted-for additional tritium sources therefore cannot contribute more than approximately 3×10^-2 rem, which is negligible in relation to the calculated doses shown previously.

Table D.5-2 Collective Population Dose for the Multiple Building Event 70-Year Doses

D.5.2.2 Chemical Releases

The buildings selected for seismic analysis are those identified during the selection process (section D.3.1). For some releases, the only initiator postulated is an earthquake; for others, the release cannot be initiated by an earthquake. Brief descriptions of the chemical releases are presented in this section. For all the facilities described here, the 0.8g earthquake is a beyond-design-basis event.

An ammonia release from Building 131, Engineering (section D.3.3.3), could be caused by damage to the building or equipment in the Engineering Records area during an earthquake. The chemical plume would be expected to travel downwind above the Emergency Response Planning Guidelines level 3 (710 mg/m³) approximately 0.3 km. No deaths or life-threatening injuries would be expected from this release.

Building 141 would be expected to fail during a 0.8g earthquake. An assessment of the operations at this location suggests that only liquid wastes would be released. This liquid waste would contaminate asphalt and drainage areas with copper and other metal finishing wastes. These wastes would pose no health hazard and could easily be removed.

A hydrogen chloride release was postulated from Building 151, Nuclear Chemistry, during an earthquake (section D.3.3.6). The chemical plume from this release would remain above the Emergency Response Planning Guidelines level 3 (152 mg/m³) out to approximately 340 m from the point of release. No deaths or life-threatening injuries would be expected due to the resulting exposures.

Building 166 uses arsine, a highly toxic and carcinogenic chemical, to grow experimental semiconductor crystals. Building 166 is expected to survive the 0.8g earthquake and does not contribute to the materials released. This building is also expected to survive during a }0.9g earthquake.

Building 322, the Plating Shop, uses strong acids and cyanide solutions in support of metal finishing operations for various facilities at the LLNL Livermore site. Building 322 is not expected to survive the postulated 0.8g earthquake, and release of hydrogen cyanide was postulated (section D.3.3.8). The resulting chemical plume would be expected to travel downwind above the Immediately-Dangerous-to-Life-or-Health level (54 mg/m³) out to less than 0.6 km beyond the site boundary. Deaths would be expected to result from this release.

A sulfuric acid spill was postulated for Building 514, Tank Farm. The postulated release could not be initiated by a seismic event. This area is expected to survive a 0.9g earthquake.

A handling accident at Building 518, Industrial Gas Yard, was analyzed in section D.3.3.1. Assuming maximum consequences, only nontoxic gases could be released from this facility during an earthquake.

Building 612, Waste Handling and Storage, would be expected to survive a 0.8g earthquake. This building was examined for damage due to a }0.9g earthquake and was found to survive.

Building 693, Liquid Waste Storage, is a new facility for storing nonradioactive liquid wastes. Assuming maximum consequences, no release could be postulated from this facility. However, the fire department would be expected to provide around-the-clock support at the facility until the fire protection systems could be repaired.

Building 614, Waste Storage, was examined for damage due to a }0.9g earthquake. The areas that could release significant quantities of toxic chemicals were found to survive.

The postulated earthquake could result in release of toxic chemicals from several locations as described above. Due to the nature of the chemicals and their associated health effects, it would be possible for deaths to occur beyond the site boundary but not farther than 600 m from the boundary. There is no scientific basis for predicting the combined health effects of exposure to this combination of chemicals. The combined health effects may be worse than the effects due to the exposure to a single chemical.

It is acknowledged that some smaller additional releases of similar materials from other buildings or areas may occur as a result of this event; however, the additional contribution to health effects risks from the smaller releases would be insignificant. Most of the other locations use chemicals in much smaller quantities which when spilled or otherwise released would only affect persons in the same laboratory or room.

In addition, the method used for combining the individual bounding releases is extremely conservative. Each individual scenario assumes that on the day of the earthquake the maximum allowable quantity of material is in the building and in the worst at-risk position. The probability of this occurring for even one scenario is extremely low. The probability of this occurring for all scenarios is clearly much lower. Consequently, the elimination of additional smaller releases from consideration in the multiple-building event scenario is more than offset by the conservatism of assuming that the bounding scenario releases occur concurrently and are directly additive.