2010-738. Chemical Facility Anti-Terrorism Standards

AGENCY:

Department of Homeland Security, National Protection and Programs Directorate.

ACTION:

Request for comments; withdrawal.

SUMMARY:

The Department of Homeland Security (DHS or the Department) invites public comment on issues related to certain regulatory provisions in the Chemical Facility Anti-Terrorism Standards (CFATS) that apply to facilities that store gasoline in aboveground storage tanks. In addition, we are withdrawing the version of this document published in the Federal Register, at 75 FR 1552, on January 12, 2010, because the footnotes in that document were misplaced. This document supersedes the January 12 document.

DATES:

Written comments must be submitted on or before March 16, 2010.

ADDRESSES:

You may submit comments, identified by docket number DHS-2009-0141, by one of the following methods:

• Federal eRulemaking Portal: http://www.regulations.gov. Follow the online instructions for submitting comments.
• Mail: U.S. Department of Homeland Security, National Protection and Programs Directorate, Office of Infrastructure Protection, Infrastructure Security Compliance Division, Mail Stop 8100, Washington, DC 20528.

FOR FURTHER INFORMATION CONTACT:

Todd Klessman, Office of Infrastructure Protection, Infrastructure Security Compliance Division, Mail Stop 8100, Washington, DC 20528, telephone number [phone].

SUPPLEMENTARY INFORMATION:

Abbreviations and Terms Used in This Document

ASP—Alternative Security Program

CFATS—Chemical Facility Anti-Terrorism Standards

COI—Chemical(s) of Interest

CVI—Chemical-Terrorism Vulnerability Information

DHS—Department of Homeland Security

EPA—Environmental Protection Agency

RMP—Risk Management Program

SSP—Site Security Plan

STQ—Screening Threshold Quantity

SVA—Security Vulnerability Assessment

VCE—Vapor Cloud Explosion

I. Comments Invited

A. In General

DHS invites interested persons to submit written comments, data, or views. For each comment, please identify the document number and agency name for this notice. DHS encourages commenters to provide their names and addresses. You may submit comments and materials electronically or by mail as provided under the ADDRESSES section. DHS will file in the public docket all comments received by DHS, except for comments containing confidential information, sensitive information, or Chemical-terrorism Vulnerability Information (CVI) as defined in 6 CFR 27.400(b).

B. Handling of Confidential and Sensitive Information and Chemical-Terrorism Vulnerability Information (CVI)

Do not submit comments that include trade secrets, confidential commercial information, Chemical-terrorism Vulnerability Information (CVI) or other sensitive information to the public docket. Please submit such comments separate from other non-sensitive comments regarding this notice. Specifically, please mark any confidential or sensitive comments as containing such information and submit them by mail to the individual listed in the FOR FURTHER INFORMATION CONTACT section. Any comments containing CVI should be marked and handled in accordance with the requirements of 6 CFR 27.400(f).

DHS will not place any confidential or sensitive comments in the public docket; rather, DHS will handle them in accordance with applicable safeguards and restrictions on access. See, e.g., 6 CFR 27.400. See also the DHS CVI Procedural Manual (“Safeguarding Information Designated as CVI,” September 2008, located on the DHS Web site at http://www.dhs.gov/​chemicalsecurity). DHS will hold any such comments in a separate file to which the public does not have access, and place a note in the public docket that DHS has received such materials from the commenter.

C. Reviewing Comments in the Docket

For access to the docket to read the public comments received and relevant background documents referred to in this notice, go to http://www.regulations.gov.

II. Background

A. Chemical Facility Security Rulemaking

Section 550 of the Homeland Security Appropriations Act of 2007 (Pub. L. 109-295, Oct. 2006) required the Department to issue, within six months, interim final regulations for the security of chemical facilities that, “in the Secretary's discretion, present high levels of security risk.” Under that authority, the Department promulgated the Chemical Facility Anti-Terrorism Standards, 6 CFR part 27 (CFATS), on April 9, 2007. See 72 FR 17688.

The CFATS interim final rule sought public comment on Appendix A, a tentative list of over 300 chemicals of interest (COI) with the potential to create significant human life or health consequences if released, stolen or, diverted, or sabotaged. Section 27.200(b)(2) of the CFATS regulation requires any chemical facility that possesses any COI at or above the applicable screening threshold quantity (STQ) specified in Appendix A to complete and submit an online data collection (the Top-Screen) to DHS. The Department uses the facility's Top-Screen and, where applicable, other available information to perform a preliminary assessment of the facility's capacity to cause significant adverse consequences if targeted for a terrorist attack.^[1] DHS uses that preliminary consequence assessment to make an initial high-risk determination for the facility. See 6 CFR 27.200-27.210.

The Department assigns each facility that is initially determined to be high-risk to a preliminary risk-based tier level (Tiers 1-4, with Tier 1 representing the highest risk) and notifies the facility that it must submit a Security Vulnerability Assessment (SVA) to DHS. The Department uses the SVA to make a final high-risk and tiering determination. Only those facilities that are finally determined to be high-risk are subject to the full scope of the regulations and required to submit, for DHS approval, Site Security Plans (SSPs) or Alternative Security Programs (ASPs) that satisfy the risk-based performance standards specified in the CFATS regulations. See 6 CFR 27.220-27.225.

DHS issued the final Appendix A on November 20, 2007. See 72 FR 65396. The November 2007 rule clarified that chemicals of interest listed in Appendix A due to potential risks related to “release” are classified as Release-Explosives, Release-Flammables, or Release-Toxics, according to the type of potential harm they may cause. See 72 FR 65397. In response to comments on the tentative Appendix A, DHS also added provisions to CFATS to clarify under what circumstances ^[2] and in what manner facilities must calculate the quantities of certain types of COI under Appendix A to determine if they are required to submit Top-Screens. See 72 FR 65397-65398.

B. Special Provisions for Counting COI in Mixtures

Among other clarifications made in November 2007, DHS added § 27.203, which instructs facilities on when and how to calculate the STQ for certain types of chemicals of interest. With respect to chemicals in gasoline, § 27.203(b)(1)(v) requires facilities to count release-flammable COI (such as butane and pentane) contained

in gasoline, diesel, kerosene or jet fuel (including fuels that have flammability hazard ratings of 1,2, 3 or 4, as determined by using National Fire Protection Association (NFPA) [standard] 704 * * * ) stored in aboveground tank farms, including tank farms that are part of pipeline systems.

In response to comments requesting that DHS clarify whether and how facilities should count COI in mixtures when calculating whether a facility meets or exceeds the applicable STQ under Appendix A, the November 2007 rule also added § 27.204. That section specifies how to calculate the amount of Release-Toxic, Release-Flammable and Release-Explosive COI (as well as Theft-COI) in chemical mixtures. See 72 FR 65399, 65416. In particular, § 27.204(a)(2) (the “flammable mixtures rule”) clarified how to calculate the quantity of Release-Flammable COI contained in chemical mixtures, including gasoline ^[3] and the other fuels specified in § 27.203(b)(1)(v), for purposes of Appendix A.

The CFATS flammable mixtures rule generally parallels the rules previously adopted by the U.S. Environmental Protection Agency (EPA) under the Clean Air Act's Risk Management Program (RMP) for counting—or excluding—flammable chemicals contained in mixtures that may be inadvertently or accidentally released.^[4] See 72 FR 65402. As explained in the preamble to the November 2007 rule, however, given the different purposes, scope, and applicability of CFATS and the EPA RMP rules, there are several important differences between the CFATS and RMP mixture regulations. See 72 FR 65398-65399, 65401-65402.

One such difference is that the CFATS flammable mixtures rule requires that Release-Flammable COI (such as butane or pentane) contained in gasoline (and other fuels specified in § 27.203(b)(1)(v)) must be counted under Appendix A, even though EPA does not count the flammable chemicals in gasoline under the terms of the RMP “mixtures rule,” 42 CFR 68.115(b)(2).^[5] See 72 FR 65399 and n. 8. The November 2007 rule explained that, while EPA's RMP rules are premised solely on accidental releases of chemicals, the COI in these flammable mixtures, including gasoline, should be counted under Appendix A because of the potential consequences to human life or health of an intentional terrorist attack. See 72 FR 65399.

C. Implementation of CFATS

Over 36,000 facilities have submitted Top-Screens to DHS and about 6,500 of those facilities were preliminarily determined by DHS to be high-risk and required to submit SVAs. DHS is now in the process of notifying those facilities that, based on review of their SVAs, DHS has finally determined to be high-risk and thus required to submit SSPs. On June 29, 2009, DHS issued final high-risk notifications to 10 aboveground gasoline storage tank facilities (i.e., terminals).^[6] Subsequently, DHS extended the SSP due dates for those facilities to allow the Department to coordinate further actions regarding terminals as a group. This extension is indefinite, pending the Department's consideration of certain technical issues and questions raised during the initial high-risk determination process for those facilities, as discussed below.

III. Issues Raised by the Gasoline Terminals Industry

A. Petition From International Liquid Terminals Association

Soon after promulgation of the November 2007 Appendix A final rule, several trade associations representing gasoline terminals raised both technical and procedural issues related to the applicability of Appendix A and the Top-Screen requirement to those facilities. Procedurally, those associations claimed that DHS did not provide advance notice and opportunity to comment on the provisions of §§ 27.203 and 27.204 related to aboveground fuel storage facilities that DHS added to CFATS in November 2007. Technically, the industry associations claimed that DHS had overestimated the potential consequences of a terrorist attack on gasoline terminals by relying on a model that calculates the impacts of a “vapor cloud explosion” from release of flammable liquids from aboveground storage tanks, which the industry asserted is unrealistic for gasoline terminals.

On May 13, 2009, the International Liquid Terminals Association (ILTA) submitted a petition to DHS under the Administrative Procedure Act requesting that DHS exempt gasoline from CFATS and remove all references to gasoline terminals from § 27.203(b)(1)(v) and the CFATS flammable mixtures rule (§ 27.204(a)(2)).^[7] Through this notice, DHS invites comments on certain technical issues related to the applicability of CFATS to gasoline terminals.

B. Modeling of Potential Consequences From Aboveground Gasoline Storage Tanks

In deciding to add provisions to CFATS for counting chemicals of interest in aboveground gasoline storage tanks, DHS considered several possible methods for modeling the potential consequences of terrorist incidents directed at such facilities—i.e., the vapor cloud explosion (VCE) model and the “pool fire” model.

1. Modified VCE Model for Gasoline Terminals

In essence, a VCE model calculates the maximum distance at which a vapor cloud produced by release of flammable chemicals would be harmful or lethal to persons in or near the cloud (the “distance to endpoint”), based on the amount of flammable liquid chemical available, the estimated amount of the liquid that would convert to vapor, and the distance the vapor cloud could spread before becoming too “lean” to explode when exposed to an ignition source.

Since EPA had already developed a VCE model for estimating the consequences of accidental releases of flammable chemicals, including flammable mixtures, under the RMP regulations, DHS used the EPA VCE model as a starting point for modeling potential VCE consequences for all Release-Flammable COIs, including those at gasoline terminals.^[8] DHS modified the EPA VCE model, however, to account for certain differences between gasoline and other flammable liquids mixtures, as explained below. DHS believes the modified VCE model reflects a plausible worst-case scenario for terminals and is an appropriate tool for assessing the potential consequences of a terrorist attack against gasoline terminals.

Specifically, DHS refined the EPA VCE model to provide an even more plausible estimate of the potential consequences of a terrorist attack on gasoline terminals in particular. While EPA's VCE model assumes that (up to) ten percent of a given amount of a flammable liquid will participate in the explosion (the “yield factor”), DHS assumes that only one percent of gasoline will participate, based on gasoline's combustion properties and its storage at ambient conditions.^[9] This modification results in a reduction of the potential consequences calculated by the model, as compared to EPA's model, and appears to be consistent with the consequences from prior vapor cloud explosions involving gasoline, as discussed below. Therefore, the modified VCE model allows DHS to reasonably estimate the number of plausible worst-case casualties resulting from a successful attack on a gasoline terminal.

DHS understands that the formation of a gasoline vapor cloud with the potential to cause significant harm to human life and health requires that a number of natural and man-made circumstances combine in a certain way, and that accidental gasoline vapor cloud explosions are therefore uncommon. DHS has determined, however, that those necessary conditions are more likely to exist in the event of an intentional terrorist incident than in the context of an accident, and thus, that it is reasonable and within the Secretary's discretion under Section 550 to apply the modified VCE model to gasoline terminals. See generally 72 FR 65399.

For example, in 2005 (long after EPA excluded gasoline from the RMP rule, see n. 5, supra), a vapor cloud explosion resulting from an unintentional overflow of a gasoline storage tank at the Buncefield Oil Storage Depot in Hertfordshire, UK caused significant injuries and other damage. Several gasoline storage trade associations have asserted that the combination of specific circumstances resulting in the Buncefield incident—e.g., accidental but prolonged and undetected overflow of the tank, failure of detection devices, congestion from nearby obstacles, weather conditions favoring accumulation rather than dispersal of the vapor cloud ^[10] —are so rare that DHS should disregard the possibility of such explosions at gasoline terminals.^[11]

DHS has concluded, however, that a terrorist seeking to cause such an explosion could target a facility where the necessary physical conditions exist (or are likely to occur at some point in time). In order to maximize the consequences of the explosion, such a terrorist could attempt to cause gasoline to leak or overflow from the targeted tank(s) in such a way as to make formation of a vapor cloud more likely than it would be in an accident like the Buncefield explosion.

Nonetheless, DHS invites public comment on the modified VCE model and on any alternatives to the specific modification made by DHS to the yield factor in the model.

2. “Pool Fire” Models

DHS considered other options for evaluating the potential consequences of a release from such facilities. Specifically, DHS considered an existing model that calculates the potential consequences from the radiated heat of a “pool fire” caused by ignition of liquid gasoline suddenly released from one or more aboveground tanks, but that implicitly assumes the pool fire is confined within dikes or other secondary containment surrounding the tank(s).^[12] The gasoline industry asserts that this “contained pool fire” scenario is more realistic for terrorist incidents involving gasoline terminals (e.g., attacks using explosive devices or weapons) than the VCE scenario. The industry also asserts that the potential consequences of such contained pool fires do not warrant subjecting terminals to any CFATS requirements.

DHS did not rely on the “contained pool fire” scenario, however, because any model that assumes the effectiveness of secondary containment does not represent a plausible, worst-case terrorist scenario, since an adversary seeking to maximize the consequences of attacking a terminal would also attempt to breach the secondary containment.^[13]

DHS is currently considering, however, and seeks comments on, whether it is feasible to refine existing models or develop a new model for uncontained pool fires (i.e., where the contents of one or more gasoline storage tanks escape from secondary containment),^[14] so that such a model could be used for future consequence assessments for gasoline terminals—in lieu of or in addition to the modified VCE model.

IV. Issues for Commenters

Comments that will provide the most assistance to DHS should address the following issues and questions. Commenters should include explanations and relevant supporting materials with their comments whenever possible.

a. Comments on the inclusion of 6 CFR 27.203(b)(1)(v) (counting of Release-COI in gasoline, diesel, kerosene, or jet fuel in aboveground storage tanks) and 6 CFR 27.204(a)(2) (the flammable mixtures rule), as they apply to gasoline terminals.

b. Comments on the applicability of the modified VCE model to gasoline terminals, including: Whether the reduction of the vapor yield for gasoline from ten percent (as in EPA's VCE model) to one percent reasonably reflects the potential consequences for a vapor cloud explosion from gasoline (as compared to other liquid flammable chemicals); and whether a different yield factor adjustment might better reflect the potential consequences for a vapor cloud explosion from gasoline.

c. Comments on whether a reasonable model exists or should be developed for future use that would allow DHS to estimate the plausible worst-case consequences of an uncontained pool fire resulting from a successful attack on gasoline terminals.

Dated: January 12, 2010.

Rand Beers,

Under Secretary for National Protection and Programs.

Footnotes

1.  In this notice, the terms “consequence” or “consequentiality” refer to the potential adverse effects on human life or health from a successful terrorist incident at a chemical facility. See generally 72 FR 17696, 17700-17701. DHS also has authority to determine that a facility is high-risk based on potential consequences to national security or critical economic assets. See 6 CFR 27.105; 72 FR 17700-17701.

2.  Among other things, the November 2007 rule provided additional criteria related to the physical state (liquid, gas, or solid), concentration levels, and forms of packaging applicable to various chemicals of interest that must be counted under Appendix A.

3.  There is no single chemical composition for the mixture typically called “gasoline,” which varies in content and blending components from company to company, region to region, and season to season. All formulations of gasoline, however, contain a significant percentage of certain release-flammable chemicals (e.g., pentane, butane) and typically have a National Fire Protection Association (NFPA) flammability hazard rating of 3.

4.  See 40 CFR Part 68.

5.  EPA's exclusion of flammable chemicals in gasoline from the RMP rules was mandated by the Chemical Safety, Information, Site Security and Fuels Regulatory Relief Act, Pub. L. 106-40. Cf. 72 FR 65410 (EPA RMP program excludes flammable fuels). In addition, EPA agreed to delete gasoline from the original version of the RMP mixture rule, which had included gasoline, in settlement of litigation with the gasoline industry. See 63 FR 640 (Jan. 6, 1998). The RMP exclusion for gasoline and other flammable fuels was codified by EPA at 40 CFR 68.126.

6.  This notice will refer to all facilities with aboveground gasoline storage tanks, including facilities (such as petroleum refineries) that may possess other chemicals that trigger the Top-Screen requirement, as “gasoline terminals” or “terminals.” Approximately 4,000 terminals submitted Top-Screens and DHS initially identified 405 of those facilities as high-risk.

7.  The ILTA petition is included in the public docket for this notice and available for review at http://www.regulations.gov.

8.  EPA's VCE model is available in Appendix C of EPA's “RMP Guidance for Offsite Consequence Analysis” (April 1999) at http://www.epa.gov/​OEM/​docs/​chem/​oca-all.pdf.

9.  See Letter dated December 10, 2008, from Sue Armstrong, DHS, to Robin Rorick, American Petroleum Institute, et al., which is available in the public docket for this notice.

10.  The ignition of such a vapor cloud, and the resulting explosion, would be relatively easy to cause once the other circumstances were in place.

11.  See “Buncefield Major Incident Investigation Board: The Buncefield Incident,” 11 December 2005 Final Report (2008), available at http://www.buncefieldinvestigation.gov.uk/​reports. DHS does not believe that it is necessary or appropriate to detail all the circumstances of that incident, or to respond to every facet of the gasoline terminals industry analyses of those circumstances, in this notice.

12.  The pool fire model is described in EPA's “RMP Guidance for Offsite Consequence Analysis” (April 1999) at http://www.epa.gov/​OEM/​docs/​chem/​oca-all.pdf. As is true for the VCE model, EPA's RMP pool fire model reflects assumptions that may be appropriate for worst-case accidental release scenarios but that are not necessarily appropriate for plausible, worst-case intentional release scenarios.

13.  See letter dated December 10, 2008, from Sue Armstrong, DHS, to Robin Rorick, American Petroleum Institute, et al., available in the public docket for this notice. The mitigating effects, if any, of secondary containment may be taken into account, however, during the Department's determination as to whether a covered facility's Site Security Plan satisfies the CFATS risk-based performance standards.

14.  Models currently available for calculating the consequences of an uncontained pool fire include assumptions that may be appropriate for releases from certain small sources (e.g., a gasoline tank truck) but that are not realistic or appropriate for worst-case modeling of large-scale releases (e.g., a sudden release from an aboveground gasoline storage tank). For example, the current EPA RMP model assumes that the surface upon which the gasoline has been released is perfectly flat and non-permeable. See EPA's “RMP Guidance for Offsite Consequence Analysis” (April 1999) at http://www.epa.gov/​OEM/​docs/​chem/​oca-all.pdf.