Cover Image

Not for Sale



View/Hide Left Panel
Click for next page ( 67


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 66
APPENDIX C Estimating the Consequence Term in the Risk Metric Equation Human-Health Consequences The human-health aspect of the consequence term is quantified by estimating the number of individuals that could receive permanent health effects from a release as well as the severity of envi- ronmental consequences. A facility with processes that fall under 40 CFR Part 68 Subpart G, the facilities Risk Management Plan (RMP), must make the plan available to local emergency response personnel and use the toxic endpoints defined in 40 CFR 68.22 to identify the number of off-site individuals who could be exposed to the listed hazards. When assessing the consequences along a transport corridor, the techniques used to estimate the population that is potentially exposed to the release for a fixed facility could also be used for a release along the transport route. Since the release could occur at any point along the route, the number of individuals that could be exposed can be estimated by overlaying the threat area--defined using the endpoints listed in 40 CFR 68.22 or the ERPG-2/TEEL-s endpoints--onto the average (or worst-case) residential population den- sity along the route. In order to establish the residential population that could be affected, an "impact distance" must be selected. Clearly, the distance would vary according to the type of haz- ard involved in an incident. For example, the Emergency Response Guidebook (ERG 2008) lists a number of protection distances for responders that are based on different materials. Although the distances are not specifically correlated to population exposure, they provide reasonable distances that can be used for this Guide. Eight hundred meters (approximately one-half mile) has been selected because it represents a distance that encompasses the endpoints of the great majority of hazmat releases and related events. The residential population could be estimated by either using a GIS with population and route layers, where the average population density can be calculated by using the average population den- sity within 800 meters from the hazmat route or fixed facility, or by dividing the region's residen- tial population by the region's land area to obtain an average per unit of land area density. Using normal atmospheric dispersion parameters in the ALOHA (2007) dispersion model and using the PROBIT (probability estimates) equations for estimating fatalities (CCPS 2000), the plume area inside the 2 percent fatality line for a release from a bulk shipment of 20,000 kilograms of chlorine in 8 minutes results in a plume area of 0.65 km2. A corresponding scenario for ammonia, which is not a dense gas, results in an area of 0.04 km2, and for acrolein, another toxic heavy gas, an area of 2.3 km2. One of the reasons why the acrolein area is so much greater is because once released, it forms a pool that takes much longer-- 30 minutes--to evaporate. Since the probability of a fatality for acrolein using the PROBIT coef- ficients is a function of the concentration times the exposure time, the longer exposure time results in more fatalities. For chlorine and ammonia, the probability of a fatality is a function of the concentration squared times the exposure time, which makes the concentration a more C-1

OCR for page 66
C-2 A Guide for Assessing Community Emergency Response Needs and Capabilities for Hazardous Materials Releases determining parameter in the PROBIT equation. The 50 percent contour obtained from ALOHA using average meteorological conditions is considered to provide a reasonable estimate of fatalities if no effort were made to protect the population from the release. Once the area inside the contour is estimated, the area can be expressed in individuals/km2 area and multiplied by the population density to estimate the number of fatalities. The number of fatalities can then be translated into a consequence measure using the CARVER scale (0 = 1 to 10 fatalities; 1 = 11 to 100 fatalities; etc.). ALOHA is the logical tool to estimate impact areas, as it was designed to be an emergency response planning tool. It has a library that contains the properties of numerous hazmat as well sev- eral commonly used endpoints. Regarding the toxic endpoints, the ALOHA library provides sev- eral choices. The default concentrations endpoints are the 60-minute Acute Exposure Guideline Levels [AEGL (60 min)]. However, you can select ERPG or TEEL concentrations. The ERPG con- centrations are developed and formally adopted by the American Industrial Hygiene Association (AIHA). The TEELs (Temporary Emergency Exposure Limit) values have been developed by The Subcommittee on Consequence Assessment and Protective Actions (SCAPA) that is part of the Comprehensive Emergency Management System funded by the U.S. Department of Energy/ National Nuclear Security Administration. ALOHA also estimates the damage area for three other potential concerns for emergency response planners: fires, BLEVEs, and the plume area where flash fires are possible because the lower flammable limit might be exceeded. The extent of the impact is judged by determining the number of people within the impact zones that may receive a significant exposure. Where sheltering-in-place is a feasible option, the fraction of the people normally outside or inside with the windows open can be used to reduce the num- ber of people estimated to receive significant exposures, which is typically only a small per- centage of the potentially exposed population. It is reasonable to assume that everyone within the hazard zone will be affected for fires and explosions/BLEVEs. In addition, if the toxic release is preceded by a fire or explosion that caused structural damage, the effectiveness of the structures to shelter people might be compromised. Since the goal is to use the CARVER scale, whatever measure used must be translated into fatal- ities. If one of the ERPG or AEGL limits is used as an estimate of fatalities, a large number of fatal- ities would be projected. These limits assume everyone within the contour, say the EPRG-2 contour, will be a fatality. In actuality, while people exposed at the ERPG-2 level might become dis- orientated after a 30-minute exposure, many will be able to walk from the plume prior to becom- ing disorientated. PROBIT curve for 50 percent fatality will result in a much smaller number. Even this curve will result in a large number of fatalities, but is considered more realistic given the assumption that no protective measures are taken. The resulting estimate of fatalities for three pos- tulated rail accidents in the region are shown in Table C-1. On the CARVER scale, these three acci- dents have [C] values of 2, 5, and 3, respectively. Environmental Consequences For the environmental damage estimates, the goal will be to determine, for each type of haz- ard, which economic damage category is appropriate. For most, it will be Level 1, less than a mil- lion dollars. One way to estimate the environmental damages for hazards that have the capability of totally destroying a structure--flash fires, fires, and BLEVEs or explosions--is to use ALOHA to estimate the damage radius and then use Table C-2 to estimate which level of loss is appro- priate. A fire, if not prevented from spreading, can involve nearby structures and do extensive damage. An explosion or BLEVE can do a lot of structural damage, resulting in replacement of

OCR for page 66
Estimating the Consequence Term in the Risk Metric Equation C-3 Table C-1. Estimated fatalities from postulated railroad accidents. Damage Quantity of Radius or Damaged Population Facility or Material Radii for Area Consequence Hazard Route Ellipse kg meters acres people Railroad s BLEVE (Ethylene Oxide) 40,000 80 1.24 2 Railroad s Toxic Gas (Chlorine) (L) 20,000 1630.23 3,057 Railroad s Toxic Gas (Chlorine) (S) 20,000 44.53 83 (S) small release; (L) large release. the structure as part of the damage estimate. The damage radius is estimated using models like ALOHA, and then the value of houses or businesses that might be within the damage radius is then used to estimate the costs. Table C-2 was developed for another project to estimate eco- nomic losses on a per-acre basis when the structures or habitat are essentially destroyed. The esti- mates were developed predominately for security risks, but should be equally valid for addressing safety risks. The endpoints for damage to structures should use the ALOHA endpoints for flash fires, explosions/BLEVEs, and the extent of flammable gas clouds. Since the structures would not be entirely destroyed, it might be appropriate to use 10 percent of the replacement value. This would represent replacement of windows and repair of minor structural damage. Land contamination impacts are a concern if the released material kills plants and trees or forms a particulate that is deposited on the ground. For example, if arsine (AsH3) is released into the atmosphere, it will react to form arsenic oxide (As2O3) that will deposit on the ground. You can obtain an estimate of the ground concentration at any location by multiplying the maximum airborne concentration in mg/m3 by the duration of the release in seconds and the particle de- position velocity expressed in meters/sec. Preventing human exposure by confiscating crops or decontaminating land or buildings would result in the greatest costs. It would be very conserva- tive to assume the same area used for estimating population impacts experienced some damage from the release event. However, since the damage would not be complete, it would be unrea- sonable to use the values in the second part of Table C-2 for land and aquatic contamination. The extent of the land impacts is also sensitive to the type of hazard. Ammonia will do a lot of damage to a wetland because of its aquatic toxicity, but it is a beneficial fertilizer on farmland. Again, 10 percent of the land and aquatic land contamination numbers shown in Table C-2 might be a reasonable estimate. The emergency response planning organization should use Table C-2 as a guide and adjust the level of impacts as appropriate for their region. It would be reasonable to increase or decrease the estimates shown in Table C-2 based on the cost of living in the region. For example, if the scenario of concern is a security risk to a structure of national significance, commonly termed Table C-2. Estimated per-acre values. Area Type Residential Commercial Industrial Land Use Farm Land Wetland Rural $ 150,000 $ 1.2 million $ 2.4 million Fallow $ 200 $ 50,000 Suburban $ 1.2 million $ 12 million $ 24 million Low-value crop $ 1,000 $ 100,000 Urban $ 8 million $ 50 million $ 80 million High-value crop $ 400,000 $ 400,000

OCR for page 66
C-4 A Guide for Assessing Community Emergency Response Needs and Capabilities for Hazardous Materials Releases Table C-3. Estimated infrastructure damage from BLEVEs and releases on railroad. Number of Quantity Typical Type Infra- Land Area Infrastructure Facility or of Infrastructure structure Hazard Damaged Units in Route Material Damaged Damage Hazard Zone kg acres # $ Residential / Railroad s BLEVE (Ethylene Oxide) 40,000 1 2 300,000 Rural Residential / Railroad s Toxic Gas (Chlorine) (L) 20,000 0.92 2 300,000 Rural Residential / Railroad s Toxic Gas (Chlorine) (S) 20,000 0.09 0 0 Rural (S) small release; (L) large release. "iconic structures," whose replacement value might be much higher than any number in the above table, then a higher damage estimate might be used. Just to show the extent of possible damages, the failure of the I-35 Bridge in Minneapolis was estimated to result in economic costs to the community that exceeded $300 million. Clearly this would have been classified as a "very high" economic cost, one much higher than any value shown in Table C-3. The next damage estimate is for land damage. This estimate is shown in Table C-4. The pos- sible land values for the region are shown in Table C-2 depending on the type of land damaged. Consequence measures for each of the accidents considering population impacts and infra- structure and land damage from the postulated accidents are shown in Table C-5; the maximum impact from each of the consequence measures should be carried forward. The same technique would be used for all the accident scenarios in the region. The results of this assessment are shown in Table C-6. Table C-4. Estimated damage to land values from postulated railroad accidents in region. Quantity Land Area Typical Type of Environmental Facility or of Material Damaged Land Damaged Consequence Hazard Route kg acres $ Railroad s BLEVE (Ethylene Oxide) 40,000 1 Low Crop Value 1,000 Railroad s Toxic Gas (Chlorine) (L) 20,000 0.92 Low Crop Value 917 Railroad s Toxic Gas (Chlorine) (S) 20,000 0.09 Low Crop Value 92 (S) small release; (L) large release. Table C-5. Consequence measures for postulated accidents on rail line s. Population Infrastructure Consequence Maximum Facility or Consequence Damage Environmental Consequences Hazard Route [C] [C] [C] [C] Railroad s BLEVE (Ethylene Oxide) 2 1 1 2 Railroad s Toxic Gas (Chlorine) (L) 5 1 1 5 Railroad s Toxic Gas (Chlorine) (S) 3 1 1 3 (S) small release; (L) large release.

OCR for page 66
Estimating the Consequence Term in the Risk Metric Equation C-5 Table C-6. Risk portfolio--H, V, and C terms. Vulner- Consequence Hazard [H] Capability Response Risk ability [C]* [ERC] Time [RTF] Metric Facility or Route Description [V] Pop. Env. Facility Z Fire (ethylene) 3 4 2 Roads x, y Fire (gasoline) 3 2 1 Facility Z Explosion (ethylene) 2 3 2 Railroad s BLEVE (ethylene) 4 2 1 Facility Z Toxic Gas (chlorine) (L) 3 4 1 Facility Z Toxic Gas (chlorine) (S) 4 2 2 Railroad s Toxic Gas (chlorine) (L) 3 5 1 Railroad s Toxic Gas (chlorine) (S) 4 3 1 Roads x, w Toxic Gas (ammonia) (L) 1 4 2 Roads x, w Toxic Gas (ammonia) (S) 2 2 1 Roads x, u Toxic Liquid (37% HCl) (L) 2 2 2 Roads x, u Toxic Liquid (37% HCl) (S) 3 1 1 *The maximum of the consequence values (population and environmental) are used in the Risk Metric calculation. (S) small release; (L) large release.