1.2. Response Phases

1.2.1. PHASES 2A&B

• Notification: The incident occurred in the early morning hours. The initial response was led by the local volunteer fire department and the county Sheriff’s Office. The state EOC was notified of the incident. The railroad company reported the release of chlorine to the NRC. The state Department of Health and Environmental Control (DHEC), State Law Enforcement Division, County Emergency Services, Federal Railroad Administration, and the EPA were notified via a NRC report number. EPA’s Regional Duty On-Scene Coordinator dispatched Superfund Technical Assessment and Response Team (START) contractors to the local command post. EPA established an EOC and coordinated an information stream to appropriate agencies. EPA activated a UC and a PIO. National Response Team¹³² were informed and activated to assist Regional NRT and the designated Federal On-Scene Coordinator (FOSC) in managing this incident. An initial evacuation of a 0.5-mile radius was ordered and persons located outside the evacuation area were advised to shelter-in-place.

• First Response: Personnel from the local volunteer fire department, the county Sheriff’s Office, State DHEC, State Law Enforcement Division, County Emergency Services, Federal Railroad Administration and railroad company representatives were at the scene. These entities made the initial assessments of the conditions in and around the derailment scene. Mutual aid assets from other local agencies were on scene or en route.¹³³

EPA collected all available data from local responders and initiated air monitoring in support of the response effort for protection of response workers as well as the general public. Given the magnitude and hazardous nature of the incident, EPA requested additional resources from Atlanta, including EPA staff, contractors, the ATSDR, and a team from the U.S. Coast Guard (USCG) National Strike Force (NSF Strike Team) that has Level A PPE capabilities.

START personnel in Level B protection entered the hot area to monitor chlorine levels using single point monitors. Based on the chlorine levels and other factors (e.g., wind patterns, unknown integrity of the railcars), federal, state, and local officials decided that the evacuation area should be expanded to a 1.0- mile radius, a shelter-in-place area with a 2.0-mile radius should be established, and a mandatory curfew instituted.

Initial reports indicated that several fatalities and numerous injuries resulted from the release. Further investigations subsequently determined that nine persons, including the engineer, died from exposure to the chlorine gas. Eight of the deceased were found in areas lower than the crash due to the gas dissemination into low-lying areas.

Persons who were potentially exposed to chlorine were sent to local hospitals for decontamination and follow-up care. Unofficial reports indicated that approximately 250 people were admitted to the hospital emergency room for treatment.

The Governor issued an executive order declaring a state of emergency for the county and schools and businesses were closed.

The local hospital established two triage units to handle the patient overflow. Exposed individuals were told to report to decontamination units set up at a local school. Two tents were established for decontamination and medical attention.

The FBI and National Transportation Safety Board began investigations of the derailment site as the railroad company began initiating operations to remove undamaged railcars and stabilize the damaged railcars. Concurrently, search and rescue operations were conducted by local officials.

1.2.2. PHASE 2C

• Characterization: UC tasked the Planning Section Chief to develop an IAP, a SAP, a HASP and an AAMP. EPA and USCG, with state and local assets, deployed for sampling and monitoring to define the chlorine gas plume and monitor air quality at locations in the surrounding community.

START and NSF Strike Team set up Area Rae™ chlorine monitors and established data monitoring centers; the operation was conducted in Level B. Chlorine monitors were placed around the area. Checkpoints were set up where responders entering the area of the incident were required to check in and out. These checkpoints were used to warn responders of the latest monitoring results and current activities in the area of the incident. A web-based server was created by a contractor for use in posting air monitoring data.

• Remediation (Cleanup): The site was managed by a task force which coordinated all sampling and analysis, decontamination, and health and safety issues. The sampling plan was modified as needed to include samples to verify decontamination efficacy, iteratively, during the remediation process. Representatives from EPA, CDC, potential UCG participation, FEMA, the state, the railroad company, and other stakeholders were selected to participate in the Technical Working Group.

EPA used a model (Areal Locations of Hazardous Atmospheres) to predict the chlorine gas plume movement. EPA also oversaw application of lime slurry to the incident scene area to minimize chlorine releases from the soils near the damaged chlorine tanker rail cars. The state investigated the scope of the fish kill and disposed of the fish carcasses. An EPA veterinarian mobilized to the site to assist local animal control officials with animal care issues in both the exclusion area and evacuated areas. The railroad company decontaminated and removed the railcars.

• Clearance: UC was tasked to develop the Clearance Sampling Plan using the clearance decision criteria agreed upon by technical SMEs. The detection limits were evaluated as sufficient to meet agreed upon clearance decision criteria. Sampling and analysis continued to verify decontamination efficacy and the area was cleared by UC.

EPA, NSF Strike Team, and START provided additional assistance to local officials in monitoring and evaluating conditions in buildings impacted by the chlorine release. One entry was conducted to recover a computer processing unit that contained critical information to support the restart process for the impacted factory. No elevated chlorine levels were detected.

• Phase 3 Restoration/Recovery: The IC/UC, closely coordinating with the local agencies, cleared the residential and business areas for re-occupancy and communicated information to the community. All air and surface concentrations were below clearance decision criteria. Local and state Departments of Environmental Protection (DEPs) continued ambient air monitoring to ensure levels were below health-based exposure guidelines established by technical SMEs. The PIO continued to be in place to address public concerns.

The evacuation order was lifted. EPA conducted air monitoring for detection of chlorine and wipe sampling in homes prior to re-entry (testing surfaces for pH). Multiple buildings on site were entered with no detections noted on chlorine-specific monitoring equipment.

EPA, ATSDR, and all state agencies established a work group to begin working on a residential reoccupation plan for implementation at the time the evacuation was lifted. The group decided to produce a fact sheet that would be distributed to the public when they returned to their homes. The fact sheet described precautions that residents needed to take and addressed concerns regarding the effects that the chlorine may have had on the environment.

EPA investigated and monitored impacts of the release of chlorine and diesel fuel to the surface water bodies. State DHEC continued the investigation of streams and water bodies near the derailment, including the fish kill observed near the incident.

1.2.3. APPLICABLE TOXICITY VALUES FOR DECISION-MAKING

The example provided in Table B-2 demonstrates an outdoor release of chlorine gas. Detections at either the odor threshold or the detection limit of the handheld detectors were approximately 1 ppm, which is comparable to a 4-hour AEGL-2 as noted. The odor threshold is between 0.01 and 0.05 ppm, which is equivalent to the 10-minute AEGL-1. This level of detection prompted an evacuation within 2 miles of the release. Although no specific health-based criteria exist for surface wipe data, public health representatives based their decisions regarding clearance of potentially impacted areas on surface wipe data for pH and on the odor threshold. Detailed descriptions and references for exposure guidelines can be found in Appendix A. Figure 8 shows a comparison of exposure guidelines for chlorine gas.

Table B-2: Example Scenario 1-—Large-Scale Chlorine Release

Phases 2a&b (0-12 hours, Day 1)

Phase	Health effects of concern	Activities and Decisions	Criteria used for decision-making	Notes
Chemical release early morning from large tanker rail car Unknowns at this phase Activities included: determine types of chemicals, locations, source amounts Controls to minimize further responder/public exposure, deaths or severe injuries	Immediate/near-term (acute) effects: Deaths reported Upper respiratory distress Breathing difficulties, burning eyes General concerns and panic Some cases of psychosomatic illnesses Approximately 250 injuries reported with many hospitalized	Emergency Response workers: Air monitoring hot area during operations to control/remove source Spikes indicated by field instrumentation required responders to evacuate; Level B PPE General public: Local officials instituted shelter-in-place. People were told to shut off heater/AC system, close windows. Public advisories issued via radio and television. People within 2 miles evacuated.	Liquid/gas odors Any “positive” (not quantified) hits with gross level field identification tests Visible liquid/gas >DOT (Orange Book) large spill protection distances (day) Chlorine = 1.5 miles (night 4.6 miles) –>local decision-makers used 2 miles	Decision-makers: Local emergency and environmental health departments

Phases 2b&c

Phase	Health effects of concern	Activities and Decisions	Criteria used for decision-making	Notes
Later Day 1 to Day 6 Activities included: control/remove source, decontaminate surrounding soil with lime slurry, and initiate clearance sampling in a phased approach. Multi-agency fact sheets under development Re-occupancy of the 2-mile area.	Immediate/near-term (acute) effects: Odor/mild irritation (minimal to none reported) Psychosomatic illnesses continue to be reported along with general public concerns.	Response workers: Air monitoring continued for hot area during operations to control source General public: Air monitoring for limited search and rescue operations Phased re-occupancy strategy: Modeling results indicated area of concern was 600 yards; supported by air monitoring; Modified evacuation area of 0.5 mile. Mandatory sampling of targeted facilities (schools, businesses) and voluntary sampling of homes for re-entry of areas outside evacuation area	Field detection >0.1 ppm required action (highest level was 3.7 ppm) MiniRae™ and AreaRae™ or equivalent used134 “Non-detects” for chlorine with field instrumentation (and no odors/visible contaminant) Surface wiping for unusual pH Surface water sampling (pH) for ecological assessment Fish kill due to chlorine release and runoff Due to fish kill, problem limited to fish carcasses	Decision-makers: Local emergency and environmental health departments together with federal agency support: HHS- ATSDR for re- occupancy fact sheets, EPA for environmental sampling

Consequence Management Phase (recovery, reentry)

Phase	Health effects of concern	Activities and Decisions	Criteria used for decision-making	Notes
Day 6 to Day 14 Phased re-occupancy ongoing and final removal of tank source By Day 14 all curfews lifted, facilities open and operations restored Multi-agency fact sheets distributed (ATSDR and state/local health departments) for re- occupancy concerns Federal (EPA) and state/local sampling teams Mandatory sampling in hot area and 600 yards out. Additional residents requested sampling: Over 750 homes businesses sampled	Immediate/near-term (acute) effects: None Permanent/chronic injury/illness: None Public concern will be high and required continuous: public assurances Pets: Four pets (2 dogs, 2 cats) were determined to have died from chlorine exposure (determined to have been outside in direct plume). Assurances to public required regarding pet health concerns.	Response workers: Limited/no PPE required End response work General public: Resumed use/re- occupancy (unrestricted) of all buildings and residences	Air monitoring is primary form of clearance sampling for the volatile/non- persistent chemicals, and targeted/surface sampling to support findings: “Non-detects” for chlorine with field instrumentation (and no odors/visible contaminant) Surface wiping for unusual pH Waste disposal for tanks, decontaminated soil, and fish carcasses Re-occupancy guidelines include recommendations to replace air filters, open windows and circulate air, throw out unprotected food (regular waste disposal), and flushing water systems for 2 min. Additional home inspections required to evaluate potential damage to critical infrastructure (electronics/ wiring, phone lines) damage from corrosive chlorine gas	Decision-makers: Local emergency and environmental health departments and local veterinarians Offered to sample residences beyond mandatory sampling areas; decision was supported by HHS, ATSDR and EPA. Multi-agency fact sheets developed and distributed

Graphic