Licensing and Safety Requirements for Launch

DEPARTMENT OF TRANSPORTATION

Federal Aviation Administration

14 CFR Parts 413, 415, and 417

[Docket No. FAA-2000-7953; Notice No. 02-12]

RIN 2120-AG37

Licensing and Safety Requirements for Launch

AGENCY:

Federal Aviation Administration (FAA), DOT.

ACTION:

Supplemental notice of proposed rulemaking (SNPRM).

SUMMARY:

The Federal Aviation Administration (FAA) is amending an earlier proposal to amend the commercial space transportation regulations governing licensing and safety requirements for launch. The FAA takes this action to propose certain changes, respond to comments on the earlier proposal, and clarify assumptions underlying the costs analysis associated with the original proposal. The intended effect of this action is to allay commenters' concerns that the costs of launching from a federal launch range will increase as a result of this rulemaking.

DATES:

Send your comments on or before October 28, 2002. The FAA will host a public meeting in Washington, DC at 800 Independence Avenue, SW., on September 6, 2002 from 8:30 a.m. to 4 p.m.

ADDRESSES:

Address your comments to the Docket Management System, U.S. Department of Transportation, Room Plaza 401, 400 Seventh Street, SW., Washington, DC 20590-0001. You may also submit and review comments through the Internet at

http://dms.dot.gov.

FOR FURTHER INFORMATION CONTACT:

For technical information: Michael Dook, (202) 385-4707. For legal information: Laura Montgomery, (202) 267-3150. If you would like to present a statement at the public meeting, or if you have questions about the logistics of the meeting, contact Brenda Parker, (202) 385-4713 before August 23, 2002.

SUPPLEMENTARY INFORMATION:

I. Comments Invited

II. Background

III. Changes to October 2000 Proposal

A. Grandfathering

B. Risk Limit for Each Hazard

C. Debris Thresholds for Use in Flight Safety Analysis

IV. Issues of Concern to Commenters

A. Authority and Need for Rulemaking

B. Cost Impacts on Licensed Launches from Federal Launch Ranges

C. FAA and Air Force Process for Relief from Common Launch Safety Requirements

V. Section-by-Section Analysis of the SNPRM

VI. Procedural Matters

I. Comments Invited

You may participate in this rulemaking by submitting written data, views, or arguments. We also invite comments relating to the environmental, energy, federalism, or economic impact that might result from adopting the proposals in this document. Substantive comments should be accompanied by cost estimates. Comments must identify the regulatory docket number and be submitted in duplicate to the DOT Rules Docket address specified above.

You may also present comments at the public meeting. The FAA will prepare an agenda of speakers, which will be available at the meeting. If we receive your request after the date specified above, your name may not appear on the written agenda. To accommodate as many speakers as possible, the amount of time allocated to each speaker may be less than the amount of time requested. Persons requiring audiovisual equipment should notify the FAA when requesting to be placed on the agenda.

All comments received, as well as a report summarizing each substantive public contact with FAA personnel concerning this proposed rulemaking, will be filed in the docket. You may review the public docket containing comments to these proposed regulations in person in the Dockets Office between 9:00 a.m. and 5:00 p.m., Monday through Friday, except Federal holidays. The DOT Rules Dockets Office is on the plaza level of the NASSIF Building at the Department of Transportation at the above address. We will consider all comments received on or before the closing date before taking action on this proposed rulemaking. Late-filed comments will be considered to the extent practicable, and consistent with statutory deadlines. The proposals in this document may be changed in light of the comments received.

Commenters wishing the FAA to acknowledge receipt of their comments submitted in response to this document must include a pre-addressed, stamped postcard with those comments on which the following statement is made: “Comments to Docket No. FAA-2000-7953.” The postcard will be date stamped and mailed to the commenter.

Public Meeting Procedures

The FAA will present a description of the SNPRM at the public meeting. The FAA will use the following procedures to facilitate the meeting:

(1) The meeting is designed to give interested parties an overview of the contents of the SNPRM to facilitate the public comment process. Therefore, the meeting will be informal and non-adversarial. No individual will be subject to cross-examination by any other participant; however, FAA representatives may ask questions to clarify a statement and to ensure a complete and accurate record. Participants will also have the opportunity to ask questions about the SNPRM.

(2) There will be no admission fee or other charge to attend or to participate in the meeting. The meeting will be open to all persons who are scheduled to present statements or who register between 8:30 a.m. and 9 a.m. on the day of the meeting. While we will make every effort to accommodate all persons wishing to participate, admission will be subject to availability of space in the meeting room. The meeting may adjourn early if scheduled speakers complete their statements in less time than is scheduled for the meeting.

(3) Speakers may be limited to a 10-minute statement. If possible, we will notify speakers if additional time is available.

(4) We will try to accommodate all speakers. If the available time does not permit this, we will generally schedule speakers on a first-come-first-served basis. However, we reserve the right to exclude some speakers if necessary to present a balance of viewpoints and issues.

(5) Sign and oral interpretation can be available at the meeting, as well as an assistive listening device, if requested at least 10 calendar days before the meeting.

(6) Representatives of the FAA will chair the meeting. A panel of FAA personnel involved in this proposal will be present.

(7) We will make a transcript of the meeting using a court reporter. We will include in the public docket a transcript of the meeting and any material accepted by the FAA representatives during the meeting. Any person who is interested in buying a copy of the transcript should contact the court reporter directly. Additional transcript purchase information will be available at the meeting.

(8) The FAA will review and consider all material presented by participants at the meeting. Position papers or material presenting views or arguments related to the SNPRM may be accepted at the discretion of the presiding officer and subsequently placed in the public

docket. We request that persons participating in the meeting provide six copies of all materials presented for distribution to the FAA representatives. You may provide other copies to the audience at your discretion.

(9) Statements made by FAA representatives are intended to facilitate discussion of the issues or to clarify issues. Any statement made during the meeting by an FAA representative is not intended to be, and should not be construed as, an official position of the FAA.

Availability of SNPRM

You can get an electronic copy of this SNPRM using the Internet through the FAA's web page at

http://www.faa.gov/avr/arm/nprm/nprm.htm

or the Government Printing Office's web page at

http://www.access.gpo.gov/su_docs/aces/aces140.html

.

You can also get a copy by submitting a request to the Federal Aviation Administration, Office of Rulemaking, ARM-1, 800 Independence Avenue SW., Washington, DC 20591, or by calling (202) 267-9680. Make sure to identify the amendment number or docket number of this SNPRM.

II. Background

Under existing regulations, the FAA evaluates, on an individual basis, a launch operator seeking an FAA license to launch from a non-federal launch site. A non-federal launch site is not located at a federal launch range. We issue a safety approval when we determine that the launch demonstrates an equivalent level of safety to that provided by a launch from a federal launch range.

See

14 CFR part 415, subpart F for more details. For a licensed launch operator launching from a federal launch range, 14 CFR part 415, subpart C applies. For launch from a federal launch range, the FAA issues a safety approval if an applicant satisfies subpart C and has contracted with a federal launch range for safety-related launch services and property whose provision and use are within the experience of the federal launch range. 14 CFR 415.31.

On October 25, 2000, the FAA proposed licensing and safety requirements for the conduct of a launch.

Licensing and Safety Requirements for Launch; Notice of Proposed Rulemaking,

65 FR 63921 (Oct. 25, 2000) (“October 2000 NPRM” or “NPRM”). The FAA proposed requirements for obtaining a license for a launch from a non-federal launch site. The proposed requirements for obtaining a license would not, however, apply to any launch from a non-federal launch site where a federal launch range performed the safety functions. For this type of launch, the licensing requirements of 14 CFR part 415, subpart C apply. The FAA proposes no revisions to subpart C of part 415.

The October 2000 NPRM also proposed to codify the safety requirements that a launch operator must satisfy to protect the public from the hazards of launch. The safety requirements would apply to all licensed launches of expendable launch vehicles, whether from a federal launch range or a non-federal launch site.

The FAA received comments to the original proposal on April 23, 2001.

1

Comments on the October 2000 NPRM generally fall into three categories: comments that caused the FAA to propose changes to the NPRM here; comments that did not cause changes, but did cause the FAA to address commenters' concerns in this preamble; and comments that the FAA is still considering and will address in the final rule. The next two sections of this preamble address the first two categories of comments. Interested readers should also see the section-by-section analysis portion later in this preamble for a description of the specific changes. The changes to the October 2000 NPRM proposed in this SNPRM include addressing how and when the proposed regulations would apply to pre-existing launch systems, changes to the measure of acceptable risk, and changes to the debris thresholds that would be used in flight safety analysis. The FAA is, through this supplemental notice of proposed rulemaking (“SNPRM”), also revising and reorganizing its proposed regulations regarding flight safety analysis. The FAA is still reviewing and considering the many technical comments and suggestions, which will be addressed in the final rule.

1

Aircraft Owners and Pilots Association,

Apr. 13, 2001; The Boeing Company, Int'l Launch Services, Lockheed Martin Corporation, Orbital Sciences Corporation, and Sea Launch Company (the “Joint Commenters”) in

Consolidated Industry Response to FAA NPRM, Licensing and Safety Requirements for Launch, October 25, 2000,

Vol.s 1 and 2 (Apr. 23, 2000) (“JC Vol. I” and “JC Vol. II”); Comments, Hugh Q. Cook, (Mar. 13, 2001);

Comments to Licensing and Safety Requirements for Launch; Notice of Proposed rulemaking October 25, 2000,

Kistler Aerospace Corporation, (Apr. 23, 2001); Letter from Tom Marsh, Lockheed Martin Corporation, (Apr. 6, 2001);

Comments on DOT NPRM Licensing and Safety Requirements for Launch, Docket No. FAA-2000-7953,

Lou Gomez, NMOSC (undated); Orbital Sciences Corporation (Apr. 23, 2001); Sea Launch Company, L.L.C (Apr. 20, 2001);

XCOR Aerospace Comments in Response to FAA Notice of Proposed Rulemaking on Licensing and Safety Requirements for Launch

(undated) (“XCOR Comments”). Under separate cover, a number of commenters filed cost impact assessments:

Boeing Proprietary Cost Impact Analysis in Response to NPRM on Licensing and Safety Requirements (Docket No. FAA-2000-7953),

(April 20, 2001) (“Boeing Costs”);

Lockheed Martin Cost Impact Analysis

(“Lockheed Cost Estimates”) (proprietary);

Orbital NPRM Cost Impact Assessment,

Orbital Sciences Corporation (Apr. 23, 2001)(“Orbital Cost Impact Assessment”) (proprietary); Sea Launch Company, L.L.C. (Apr. 20, 2001) (“Sea Launch Costs”) (proprietary).

Since 1998

2

, the FAA and the Air Force ranges have been working together to achieve common safety standards that may be universally applied to licensed and government launches. The FAA anticipates that for licensed launches that are conducted at federal launch ranges, the ranges will continue to implement these requirements. As explained in past rulemakings, the FAA conducts a baseline assessment of the adequacy of the federal launch ranges to determine whether the FAA may rely on the safety requirements of the ranges and on their implementation of those requirements.

3

The FAA's baseline assessments document the capabilities, safety program, standards and policies of each federal launch range. The FAA recognizes, of course, that the federal launch ranges of the Department of Defense and National Aeronautics and Space Administration have their own missions separate from the support of commercial or otherwise licensed launches. Accordingly, the FAA proposes to codify the ranges' safety requirements to fulfill, in part, the FAA's own responsibilities for safety. Codification identifies those requirements upon which the FAA relies for licensed launch operators to achieve safety, and, in the unlikely

event that either of the ranges can no longer provide support on a non-interference basis for commercial launch, ensures that a launch operator is informed of the safety requirements with which it must comply. Because the different ranges experience different meteorological, geographical and population environments, the ranges do not always implement their requirements in the same manner. The FAA attempted, in the NPRM, to identify the underlying intent shared by the ranges' safety requirements, and then presented those principles in the NPRM, in a more generally applicable and abstract form, which may be unfamiliar to those accustomed to launching from a particular range.

2

In recognition of the efforts of the FAA and the ranges to achieve common safety standards, an interagency working group led by the Office of Science and Technology Policy and the National Security Council of the White House recommended, among other things, that the FAA and the U.S. Air Force “continue their cooperative development of common safety requirements to be applied to government and commercial launches at federal and non-federal launch sites.” White House Office of Science and Technology Policy and National Security Council,

The Future Management and Use of the Space Launch Bases and Ranges,

38 (Feb. 8, 2000). At the same time, the working group recommended that the FAA and the U.S. Air Force formalize their respective responsibilities for the safety of space launches through a memorandum of agreement.

Id.

at 39. The report urged that the federal ranges retain current responsibilities for the safety of government activities, and retain safety of commercial flight activities at the Eastern and Western Ranges. On January 16, 2001, the FAA Administrator and the Assistant Secretary of the Air Force entered into a Memorandum of Agreement Between Department of the Air Force and Federal Aviation Administration on Safety for Space Transportation and Range Activities. A copy of the MOA is available on AST's Web site (

http://ast.faa.gov)

.

3

See Commercial Space Transportation Licensing Regulations,

64 FR 19586, 19596-97 (Apr. 21, 1999).

III. Changes to October 2000 Proposal

A. Grandfathering

Although the proposed requirements are derived from existing range requirements, there are, for any number of different reasons, launch vehicles and launch operators who would not comply with the requirements as proposed in the NPRM. For example, in the NPRM, the FAA noted that there might be instances where the ranges had granted waivers to the requirements of Eastern and Western Range 127-1, Range Safety Requirements (“EWR 127-1”). NPRM, 65 FR 63941. Additionally, the FAA recognizes that there are launch operators operating under older versions of EWR 127-1 who would not meet current federal range standards or, therefore, the proposed FAA requirements. In the NPRM, the FAA noted that launch operators might experience cost impacts from bringing their operations into compliance with the proposed requirements, and requested comments on the FAA's plan not to “grandfather” such noncompliances.

The FAA received comments suggesting that, in addition to existing waivers, other candidates for grandfathering exist. JC Vol. I at 9. The comments noted that the ranges grandfather sub-systems on launch vehicles that become non-compliant when the ranges implement new safety requirements. Additionally, comments called the FAA's attention to the ranges' “tailoring” process, by which a range determines whether a launch operator's proposed alternative, although not compliant with the letter of the range requirements, nonetheless meets the intent behind the requirement. Commenters urged the FAA to accept existing tailoring agreements. For all these scenarios, including waivers, tailoring and existing range grandfathering arrangements, launch operators urged that the FAA “grandfather” current launch systems. Launch operators urged cost and range practice as the reasons for grandfathering. The FAA is considering adopting some of the suggestions contained in the comments to this rulemaking, but requests additional comment and information in light of the considerations discussed below.

1. Applicability and Effective Dates of Requirements

Commenting launch operators requested that the FAA provide more detail regarding how and whether grandfathering would work. The FAA specifies an effective date for each rule promulgated. There are a number of options for determining an effective date. A rule might apply, for example, to all launches that took place after a certain date, regardless of when the launch vehicle was designed or built. Usually, for such a decision an agency would provide a fairly lengthy lead-time. Alternatively, a rule might apply to all launch vehicle components manufactured after a certain date. Again, a lengthy lead-time might be necessary to allow a licensee to incorporate any changes into its design and subsequently manufactured hardware. Finally, in accordance with Department of Transportation and FAA usage, the FAA's proposed regulatory requirements will not employ the term “grandfather,” but will, instead, describe how and when part 417 would or would not apply.

For a meets intent certification or noncompliance to qualify under the FAA's proposed version of grandfathering, the federal range approval of such relief from a safety requirement would have to exist as of the effective date of proposed part 417. The FAA intends to allow sufficient time between the issuance of the final rule and the date that part 417 would become effective for federal ranges to make decisions on pending requests for relief that might be in work at the time a final FAA rule is issued. For launches from Air Force ranges, the Air Force and the FAA intend to have the joint relief process, discussed in section IV.C of this supplemental notice, in place prior to the effective date of part 417. This will allow for a smooth transition from pre-existing Air Force relief approvals that would qualify for the FAA's proposed version of grandfathering, to the joint process that will be used to resolve future requests for relief from launch safety requirements.

2. Range Approach to Implementing new Safety Requirements

At the Air Force's launch ranges, EWR 127-1 governs. The Air Force's range safety organizations periodically update these requirements, and determine the extent to which those updates will affect existing launch vehicles and systems. Commenting launch operators noted that “the existence of such new requirements does not necessarily make an existing system unsafe or expose the public to greater safety risks.” JC Vol. I at 9. EWR 127-1 recognizes this, and grandfathers and maintains the approvals of previously approved systems unless the Chief of Safety or the launch operator determines one of the following:

a. Existing programs make major modifications or include the use of currently approved components, systems, or subsystems in new application (through tailoring if desire[d]) Exception: Previously approved existing components, systems, or sub-systems that do not increase the risks, do not degrade safety, or can survive new environments [that] are equivalent to or lower [less severe] than the originally approved qualification levels shall be honored and do not have to meet new requirements [do not have to be upgraded] as long as data and analyses show that the criteria have been met.

b. The Range User has determined that it is economically and technically feasible to incorporate new requirements into the system.

c. The system has been or will be modified to the extent safety approvals no longer apply.

Note:

Risk and hazard analyses developed jointly by Range Safety and the Range User shall be used to determine applicability of the safety approvals.

d. A previously unforeseen or newly discovered safety hazard exists that is deemed by either Range Safety or the Range User to be significant enough to warrant the change.

e. The system does not meet the requirements existing when the system was originally accepted.

Note:

This category includes systems that were previously approved, but when obtaining the approval, the noncompliances to the original requirement were not identified.

f. A system or procedure is modified and a new requirement reveals that a significant risk exists.

g. Accident and incident investigations and reports may dictate compliance with the document.

EWR 127-1,

Appendix 1C, 1C.1.4, 1-35 (Dec. 31, 1999).

As review of the above range exceptions shows, a host of possibilities may trigger a requirement for a launch operator to change its launch vehicle or systems to conform to the latest safety requirements. These possibilities may be divided into two general conditions: where a launch operator is implementing other changes to its launch vehicle, and where the safety considerations are so overriding that a change is required. Accordingly, although grandfathering may be automatic under the range regime, grandfathering is not unlimited.

The issue of grandfathering highlights how the Air Force has successfully dealt with the issue of providing for appropriate public safety while taking into consideration the issues of cost, schedule, and mission assurance. The FAA recognizes that there are parallels that can be drawn between the Air Force's approach to ensuring public safety, including the use of grandfathering, and the FAA's regulatory focus on ensuring public safety without placing undue burden on the launch industry. Since publishing the NPRM, the FAA has considered further the Air Force's approach to grandfathering and how the Air Force has successfully implemented its grandfathering policies to ensure public safety without placing undue burden on the launch industry. Upon the urging of the commenters, the FAA proposes to adopt a similar approach to determining when non-compliance with a particular requirement may be permitted to continue.

3. Applicability of Proposed Requirements to Pre-Existing Range Meets Intent Certifications

Under this SNPRM, proposed section 417.1(b) would permit a launch operator not to have to demonstrate an equivalent level of safety to the FAA for certain range “meets intent” determinations if the launch operator was licensed by the FAA and launched from a federal range. In the NPRM the FAA, while proposing not to grandfather noncompliances with the proposed requirements, was silent with respect to how it would treat meets intent certifications. This meant that all launch operators would be required to satisfy all the FAA's proposed launch safety requirements once those requirements went into effect. To satisfy a requirement, a launch operator would have to meet the requirement as stated in the FAA's proposed regulations or demonstrate that an alternative approach provided an equivalent level of safety. For existing launch vehicles operating from federal ranges, the federal range safety organizations have granted “meets intent certifications” for substitutes preferred by the launch operators to some of the current range safety requirements. Because the current federal range safety requirements provide the basis for the FAA's proposed requirements, any grant by a federal launch range of a meets intent certification creates the possibility that the launch operator would not necessarily comply in a literal sense with a proposed FAA requirement.

The federal ranges have granted meets intent certifications when they found that a launch operator's proposed approach, although literally non-compliant with a requirement, complied with the overall intent of the requirement. To obtain meets intent approval from a federal range, a launch operator's proposed substitute has to maintain an equivalent level of safety despite not meeting the exact requirement.

EWR 127-1

at 1-vii (Dec. 31, 1999). For all intents and purposes, a range safety meets intent certification constitutes one form of the FAA's equivalent level of safety. Additionally, a federal range's tailoring of launch safety requirements for specific launch vehicle programs often includes meets intent certifications that apply to a launch vehicle program on a permanent basis.

The FAA now proposes through section 417.1(b) that a launch operator would not need to demonstrate an equivalent level of safety to the FAA for satisfying an FAA requirement for a licensed launch from a federal range, if two conditions were met. The first condition would be that the launch operator would have to have a license from the FAA to launch from the federal launch range and the license would have to be in effect as of the effective date of part 417. This is reasonable because, to date, through its baseline assessments, the FAA has relied on the federal range determinations that a particular substitute to a range requirement met the intent of that same requirement. In the context of meets intent certifications, the FAA sees no need to revisit or second-guess that past reliance. Under this SNPRM, the possessor of “meets intent certification” could continue to rely on the range's determination, where a future or different licensee could not. Additionally, even the same licensee would not be able to rely on a pre-existing meets intent certification for any other vehicle or application other than the one for which it was originally granted.

Thus, the second condition would be for the launch operator to have a written pre-existing “meets intent certification” for the requirement from the federal launch range from which the launch will take place, or a substitute that the same range approved during tailoring of the range safety requirements for that launch operator. This proposal is consistent with the ranges' own approach to “grandfathering.” Under current practice, range grandfathering applies only at one launch site.

See

Appendix 1C, 1C.1.4 a (permitting grandfathering unless a currently approved component, system or subsystem is to be used in a “new application”). If a launch operator has launched a vehicle from one range and proposes to launch from a different range, the other range will review the substitution for acceptability.

Review due to a change in launch site is necessary because different conditions at different launch sites may dictate different decisions. If, for example, not performing an environmental test is acceptable at one range, different environments at a different launch site may require that the test be conducted. Environmental factors such as salt, fog and temperature may vary from site to site, as may the potential for extreme environments, such as earthquakes on the west coast and hurricanes on the east coast, thus changing the need for and requirements governing component testing. Similarly, with a change in trajectory profile brought about by launching from a different site, vibrations could occur at different times of flight. The ranges see a need to address and consider these changes and determine whether a substitution acceptable at one launch site is acceptable at another. The FAA agrees with this reasoning and proposes to maintain this practice.

Under this SNPRM, the “meets intent certification” would have to exist as of the effective date of part 417 and the duration of the “meets intent certification” would have to include the licensed launch in question. If a pre-existing meets intent certification did not apply to a future licensed launch, the launch operator would have to demonstrate an equivalent level of safety to the FAA. For example, the ranges have granted some launch operators meets intent certifications that allowed them to fly without a flight termination system on an upper stage of their launch vehicles. Such range approvals are highly dependent on launch specific conditions and do not necessarily apply outside of certain launch azimuths. The FAA recognizes, however, that even for a meets intent certification granted only for a specific launch there may be a possibility that

the reasons that merited grant of a meets intent certification will apply again and the FAA will be able to find an equivalent level of safety. However, just as the ranges reserve the right to make that determination for a different set of circumstances, so, too, will the FAA. For future FAA-licensed launches from federal ranges, launch specific decisions such as these will be handled through a coordinated FAA and federal range review process as discussed in section IV.C of this SNPRM.

4. Pre-existing Range Waivers and Non-Compliances That Satisfy Range Grandfathering Practices

Under proposed section 417.1(b)(1) of this SNPRM, the FAA would not apply a requirement of proposed part 417 to a licensed launch if the launch operator is currently licensed by the FAA to launch from a federal range, and if the range has either previously approved a waiver for the requirement or if the noncompliance is in accordance with federal range “grandfathering” practices. Unlike a meets intent certification where a launch operator satisfies a requirement through an alternative that provides an equivalent level of safety, a launch operator at a federal range might not satisfy a current range safety requirement and, therefore, would not satisfy one of the FAA's proposed launch safety requirements. A federal range may have approved such non-compliances as specific waivers or the non-compliance may have resulted from the launch vehicle program being initiated under an earlier version of the range safety requirements and being subject to Air Force grandfathering policies.

In the NPRM the FAA proposed not to grandfather non-compliances, but requested public comments on the issue. Upon consideration of input from industry and the federal range safety organizations, the FAA now believes that it would be appropriate to provide a form of grandfathering that is nearly identical to the Air Force's grandfathering policy. The FAA's version of grandfathering, namely, partially limiting the reach of its requirements, would apply to federal range waivers and other noncompliances that have been grandfathered by a federal range. Since the NPRM was published, the FAA has considered further how grandfathering is implemented in current practice at the federal ranges, including recognizing that there is a degree of safety assurance that can be derived from the demonstrated flight history of an existing vehicle.

The FAA now proposes to permit, with some exceptions, that a requirement of this part would not apply to a licensed launch from a federal range, if certain conditions were met. These conditions would be the same as those the FAA is proposing for pre-existing meets intent certifications, as discussed above. The first condition would be that the launch operator would have to have a license from the FAA to launch from the federal launch range and the license would have to be in effect as of the effective date of proposed part 417. A launch operator who had a launch license on the day that part 417 became effective would satisfy this condition. Although the possessor of the waiver will be able to rely on the range determination, a future or different licensee will not. Additionally, the same licensee would not be able to rely on a pre-existing waiver for any vehicle or application other than the one for which it was originally granted.

The second condition would be that the launch operator, as of the effective date of proposed part 417, had, for that requirement, a written waiver from the federal launch range, or a pre-existing noncompliance that satisfied the federal launch range grandfathering criteria. The FAA intends this provision to encompass noncompliances regardless of the avenue through which they arise. In the first instance, a range may grant a waiver. In the second, a range may have approved a launch vehicle or system under requirements in place some time previously. Although the range requirements may change, a launch operator is not always required to upgrade the launch vehicle or system as discussed above. This provision would apply to both forms of pre-existing non-compliance.

The condition that a range approval be in writing would apply to range waivers.

See EWR 127-1

at 1-38, Appendix IC, IC.2.4 (describing required range approvals). For a launch vehicle that has been grandfathered, the range maintains a version of the range safety requirements that apply to the vehicle. These are the requirements that are “tailored for that vehicle.” For any new safety requirement that the range determines must apply to an existing launch vehicle, the range will update the tailored set of range safety requirements.

Just as with the FAA's proposed approach to pre-existing meets intent certifications, the FAA would condition not applying a requirement for a licensed launch on an existing non-compliance being already approved for the licensed launch in question. If the range approval of a pre-existing non-compliance did not apply to a future licensed launch, the launch operator would have to meet the requirement as written or demonstrate an equivalent level of safety to the FAA and the Air Force in the joint relief process discussed in section IV.C of this notice. Because waivers are granted for situations where an equivalent level of safety is not achieved, the FAA considers it even more important than with pre-existing meets intent certifications that the FAA review the acceptability of a waiver when there are differences from the circumstances that warranted grant of the waiver in the first place. As with the meets intent certification, the FAA recognizes that the reasons for a waiver may exist again. However, just as the ranges reserve the right to make that determination for a different set of circumstances, so, too, will the FAA.

5. Limits to Grandfathering

As discussed previously, range grandfathering is not necessarily guaranteed under current practice at the federal ranges. Depending on the criticality of an issue and, given time and opportunity, a federal launch range will strive to bring a launch operator's vehicle and operations into compliance with current safety requirements. Accordingly, the FAA proposes to codify that practice as well in proposed section 417.1(b)(2).

Like the ranges, even if the launch operator were to satisfy the conditions of proposed section 417.1(b)(1) for a specific requirement of proposed part 417, the FAA proposes that a launch operator must comply with proposed part 417, including by providing a demonstration of an equivalent level of safety, whenever the launch operator makes modifications that affect the launch vehicle's operation or safety characteristics. As with the Air Force's current practice, proposed § 417.1(b)(2) would require a launch operator to upgrade if the FAA or the launch operator determined that a previously unforeseen or newly discovered safety hazard existed that was a source of significant risk to public safety, or if a federal range previously accepted a component, system, or subsystem, but did not identify a noncompliance to an original federal range requirement. In the past, this meant that a launch operator making a major change to its launch vehicle had to upgrade the launch vehicle to satisfy current safety requirements. For example, modifications made to a launch vehicle to allow the use of strap-on solid rocket boosters where none were originally

approved would be considered major modifications that could affect the vehicle's operation and safety characteristics. As a result, many aspects of the original flight termination system would have to be upgraded to comply with the most current requirements. This change would have the effect of codifying the federal launch ranges' current practice.

The FAA also proposes, as under current practice, that a launch operator bring its launch vehicle or launch into compliance with a requirement when it uses the launch vehicle or a component, system, or subsystem in a new application. A new application may include launching the vehicle from a new launch site or using a safety component on a different stage of the vehicle other than the stage for which it was originally approved.

6. Grandfathering of a Launch Vehicle Program at an Air Force Range

The FAA recognizes that the Air Force and licensed launch operators at Air Force ranges often consider a launch vehicle program as a whole grandfathered. The FAA's proposed grandfathering provisions would govern the applicability of individual safety requirements. As is current practice in implementing the Air Force's requirements, the FAA's proposed requirements may be applied to a launch vehicle program such that all aspects of the existing program are grandfathered without the need to upgrade to satisfy the safety requirements of proposed part 417. The Air Force and the FAA are involved in an extensive effort to identify and maintain common launch safety requirements through an interagency group consisting of both Air Force and FAA personnel, called the Common Standards Working Group.

4

The Common Standards Working Group worked to ensure that the FAA's proposed requirements are consistent with the Air Force's grandfathering requirements and can be implemented without duplication of effort. A launch vehicle program that is fully compliant with the Air Force's grandfathering requirements could be fully compliant under the FAA's proposed requirements. This would be possible in the event that all the non-compliances or meets intent certifications for a particular launch vehicle satisfied the FAA's proposed criteria.

4

The Common Standards Working Group consists of, in addition to FAA representatives, Air Force representatives from Air Force Space Command, the Air Force Space and Missile Center, Air Force Safety Center, safety personnel from both the Eastern and Western Ranges, and each of their contractors working in support of this joint effort.

B. Risk Limit for Each Hazard

1. Changes to NPRM Proposal

In proposed section 417.107 of the NPRM, the FAA proposed to aggregate the risks attributable to all mission hazards and set a cap on the total mission risk of all hazards at an expected average casualty of 30 × 10

−6

. The FAA received comments in opposition to this proposal from the public, and addressed the concerns with the other members of the Common Standards Working Group. The changes proposed here constitute the results of the consensus reached between the FAA and the U.S. Air Force through the Common Standards Working Group. In summary, the FAA, with the agreement of the U. S. Air Force, now proposes through this rulemaking to adopt the current practice at the 45th Space Wing and to set a cap on the risk presented by each hazard. Because of the differences in underlying assumptions and methodologies for assessing the risk of each hazard, the FAA will not require or consider a limit on the total mission risk created by all the hazards of launch. For any given launch, the risk attributable to the whole mission tends to arise out of one hazard. Accordingly, as a general matter, the FAA still expects the aggregated risk of most launches to remain near an E

c

of 30 × 10

−6

.

In the NPRM, the FAA proposed to require that an aggregate of the hazards created by a particular launch not exceed an E

c

of 30 × 10

−6

.

NPRM,

65 FR 63921, 63981 (proposed section 417.107(b)). This meant that a launch operator would have had to account for all hazards, including, but not limited to, the risks associated with debris, toxic releases and far field blast overpressure. The FAA proposed this limit after consultations with Air Force safety personnel at the 30th and 45th Space Wings. Both wings were receptive to this approach because it supported a theoretical goal of launch risk management, which is to quantify all hazards in a single, normalized risk measure. As noted in the NPRM, the 30th Space Wing found that one hazard typically served as the source of the risk attributable to a mission.

NPRM,

65 FR 63921, 63936. Conditions that are conducive to driving up the risk associated with one hazard usually make another hazard less significant. Accordingly, representatives of the 30th Space Wing advised that launch availability would not be jeopardized at Vandenberg Air Force Base with a total mission risk cap of 30 × 10

−6

. Thus, although the 30th Space Wing advised that it did not, in practice, set a ceiling for aggregate risk at 30 × 10

−6

, launches from Vandenberg could meet the standard.

As discussed in the NPRM, the experience of the 45th Space Wing differed. The current practice of the Eastern Range, as described in the NPRM, was to cap two hazards, debris and far field blast overpressure, at an E

c

of less than or equal to 30 × 10

−6

.

NPRM,

65 FR 63921, 63936. Although the Eastern Range estimates that it accepts a risk at an E

c

of 233 × 10

-6

for the risk attributable to a launch's potential toxic releases, its analysis does not account for a variety of factors that may reduce risk but are difficult to quantify. A review of licensed launches between September 4, 1997, and August 23, 2000, shows that only two out of 39 licensed launches took place with an E

c

for toxic releases in excess of 30 × 10

−6

.

Eastern Range Aggregate Risk Study,

RTI Int'l (Oct. 2, 2001). One occurred on May 4, 1999, with an E

c

for toxics of 57 × 10

−6

for the launch of a Delta III. The other occurred on July 10, 1999, with an E

c

for toxics of 114 × 10

−6

for a Delta II launch vehicle. Because all indications pointed to the ability of Western Range launches to continue to satisfy an aggregated risk criteria, and because the Eastern Range stated that most of the higher toxic risk numbers applied only to federal government launches, such as the Shuttle and Titan vehicles

5

, both ranges and the FAA agreed to propose the aggregated mission risk cap in the October 2000 NPRM.

5

The Air Force advises the FAA that it will accommodate this discrepancy to the common standards through its own granfathering or waiver process.

The FAA received comments opposed to aggregating mission risk. Launch operators commenting on the October 2000 NPRM stated they expect the E

c

values from downrange debris risk

alone

to be close to or surpass the 30 × 10

−6

criteria with flight azimuths entailing African or European overflight. JC Vol. I at 8 (emphasis in original);

accord

Boeing Cost Impact at 2. The launch operators therefore believed that a single, collective E

c

at the proposed level would restrict launch availability and cause launch delays, both of which increase launch costs.

6

6

The FAA would like to clarify a misunderstanding on the part of the launch operators commenting about how risk is calculated. In the Joint Comments, the launch operators argue that “[t]he fact is, that the actual public risk can only be realized at one given point in the launch timeline. If a launch vehicle is terminated during up-range flight, there is no threat to the down-range public. Conversely, by the time down-range public is potentially endangered, the up-range public is clear of risk.” JC Vol. I at 9. Risk calculations must assess the risk for the entire launch. When making risk calculations to determine whether the pubic risk criterion is satisfied for a launch, risk is not calculated during the launch but before the flight takes place and accumulated for all stages of flight. The risk calculation must account for all stages of flight if it is to be used to determine whether flight should be initiated, which is the intended use of the public risk criterion. The mutual exclusivity of failure scenarios has long been recognized and appropriately accounted for in the risk analyses performed at the Air Force ranges. When calculating risk, one of the important variables, namely, the probability of the launch vehicle's failure (Pf), is proportioned as a failure rate over each phase of flight so that there is some mathematical accounting for the fact that a launch vehicle can only fail once during flight.

In light of the concerns raised by launch operators, the FAA again revisited current practice at the ranges through consultations with the Common Standards Working Group. The working group explored in detail the philosophies and limits behind current risk assessment approaches and what was proposed in the NPRM. Air Force current requirements permit different aggregation practices.

See

EWR 127-1, 1-41, Appendix 1D, 1D.1b (“The overall risk levels

may or may not

be an additive value that includes risks resulting from debris, toxic and blast overpressure exposures.” (Emphasis added))(cited in

NPRM,

65 FR at 63936). The current practices at each of the two ranges remain as described in the NPRM. Results of the study conducted in 2001 indicated that there were only a few commercial launches in the past five years that would not have satisfied the aggregation criteria. Having explored a number of alternatives, the FAA now proposes to codify a less restrictive practice of not aggregating risks as proposed by the Common Standards Working Group.

Although the Common Standards Working Group agrees that a risk assessment that determines the total risk due to all hazards associated with a single launch would be an ideal approach, the group also agrees that there are a number of reasons not to codify such an approach at this time. The Common Standards Working group proposes separate risk criteria for each hazard because it is current practice for the 45th Space Wing, the range from which the majority of commercial launches take place, and because it reflects the disparate approaches to and abilities in modeling the risks of each hazard. Currently, the differences between the hazards create differences in how to measure the risks attributable to each of those hazards. A risk measure accounts for a number of things, including the probability of the undesired event occurring (usually related to the launch vehicle's probability of failure), the characteristics of the hazard, and the characteristics of any exposed populations. At this most general level, both ranges assess risk to account for each of these factors. When it comes to addressing each hazard, however, differences arise. Although the models of the two Air Force ranges tend to account for similar factors, the input to those models differs at each coast.

Because the FAA and the Air Force intend for their methodologies to account for the same factors, such as serious injury, population and the like, the Common Standards Working Group had to review the current practice underlying the risk assessment for each hazard. That review demonstrates how difficult it is to normalize among hazards.

Population characteristics are, at the most abstract level, treated similarly in that the methodologies and models attempt to describe the location or other attributes of an exposed population in a reasonably conservative manner. But what constitutes a reasonably conservative estimate for one hazard may differ for another hazard, which makes assessing each hazard through a separate inquiry a reasonable exercise. For example, when assessing the risks posed by far field blast overpressure, the conservative approach, in the absence of data detailing true locations, would be to assume all the population is located inside buildings and thus exposed to the danger of flying glass. When assessing the risk posed by a release of toxic substances, on the other hand, the conservative approach would be to assume that at least a portion of the exposed population was outdoors, thus increasing the likelihood of harm from the release. The characteristics of a population relevant to an assessment will also vary depending on the hazard at issue. For example, age will play a role in whether a person is harmed by a toxic release: a toxic exposure that fails to injure a healthy adult may seriously injure an infant or the infirm. Age is a much less important parameter for penetration injuries due to flying glass shards. Accordingly, age characteristics may be necessary for one assessment but not another.

In analyzing how a particular hazard may cause an injury, the elements of the risk assessments also diverge. Each hazard causes a different kind and degree of serious injury, so that employing separate methodologies and models to address each is reasonable for purposes of analyzing what harms a person. For example, inert debris causes injuries of penetration, blunt trauma or crushing. Explosive debris may cause knockdown and blast injuries, including, for example, “blast lung,” gastrointestinal blast injury, damage to the inner ear, and eardrum rupture. Air blast loading caused by far field blast overpressure may break windows and pose a threat of laceration to building occupants or those nearby. Toxic releases may result in damage to the respiratory system, skin, and eyes.

These different injuries are produced by different causes and the thresholds and measures for serious injury from each hazard will vary. For inert debris, risk assessments tend to account for such characteristics as the mass of the debris, the impact velocity of the debris, debris orientation or the projected area of the debris or a combination of any of these characteristics. The threat posed by a gaseous toxic release is generally characterized by the concentration levels, described in parts per million, and the duration of exposure. An assessment of the far field blast overpressure risk will account for a variety of window characteristics, including window types, fragment sizes, velocities, distances propelled, or impacts per unit area.

The result of this review is that it is reasonable to perform separate risk assessments and employ separate criteria because of the difficulty in normalizing risk across all the different hazards. The current models for estimating risk used at the Air Force ranges represent the state of the art. Nonetheless, current techniques still cannot aggregate the risk across all hazards in a consistent manner without introducing additional uncertainty. This is due to differences in how the hazards are modeled and the nature and quantification of the serious injuries that result from each hazard.

2. Alternatives Considered

The Common Standards Working Group explored a number of alternatives before settling on the proposal described above. Those alternatives and their benefits and drawbacks are discussed here. The Common Standards Working Group sought to identify risk assessment procedures that would best protect the general public and reflect current practice without unduly burdening the launch community. In doing so, the working group considered several options both individually and in combination. Chief among the concepts considered were various forms of risk aggregation and risk accumulation. Aggregation requires the risk assessment to combine and limit the total risk

associated with the three main hazard categories. Aggregation would ensure that a single risk measure capped the combined risk due to the three main hazard categories. Accumulation combines the risk in the launch area with risk incurred downrange. The group also considered options related to increasing the maximum allowable expected casualty level and imposing different expected casualty limits on new and mature vehicles.

In addition, the Common Standards Working Group considered a third option that would have required the same risk assessment as the original aggregation and accumulation option outlined in the NPRM. The only difference between the two proposals would have been an increase in the maximum allowable E

c

value under this option. Aggregating and accumulating with an increased E

c

limit could have prevented the risk assessment from becoming overly conservative by adjusting the acceptable risk criterion. However, the main difficulty with this option would have been that choosing a new expected casualty limit would have been difficult to justify in the absence of historical data on which to base it. This difficulty could be mitigated, however, through a focused scientific study dedicated to logically determining an expected casualty limit. In fact, the Department of Defense's Range Commander's Council has previously conducted a similar study that could be used as a baseline for any future research.

A fourth option would have required a launch operator to aggregate risks across the three main categories of hazards without accumulating the flight risks incurred in the launch area with those incurred downrange. The result would have been two separate casualty expectation values for each licensed mission. One value would have represented the aggregate risk in the launch area while the other would have represented the risk downrange. In a departure from the current practice as outlined in EWR 127-1, this option would have imposed individual caps on aggregate risk in both areas but would not have imposed a total hazard cap on any single launch. This option may have had less of an impact on launch operators than the NPRM proposal to aggregate, but would have recognized the different methods used to calculate launch area hazards compared to downrange hazards. These differences include variations in the nature of necessary data and the fidelity of the analyses. Such variations reflect the fact that the ranges typically are not concerned with toxic releases or distant focusing of blast overpressure downrange because most or all of the fuel on board the vehicle would have been consumed en route, or lost on reentry due to the break up and dispersion of liquid fuels. Also, data regarding meteorological conditions tends to be unavailable for most downrange far field blast overpressure concerns. As a result, downrange risk would consist almost entirely of the debris risk, whereas launch area risks would also include overpressures and toxic releases. However, the underlying premise of this option is flawed by the fact that separating launch area risks from downrange risks is contrary to pure risk assessment philosophy in that it considers a launch in discrete parts instead of as a single continuous event. For missions involving multiple distinct periods of population overflight, assessing the risk to each region of overflight separately could result in missions with a very high expected casualty even though the mission met the risk criteria for each overflight area. In other words, such an approach would mask the true risk of the whole mission. Another disadvantage is that, like with other proposals in favor of aggregation, it might be difficult to define and calculate a consistent methodology that normalized the effects of each of the hazards. This particular disadvantage arises from the fact that the same expected casualty value may reflect two different things when applied to two different hazard categories. For example, an E

c

of 30 × 10

−6

for toxic releases means something different than 30 × 10

−6

for debris because, in most cases, more people would have to be exposed to a toxic release to inflict the same number of casualties as a debris impact. Similarly, the potential for fatalities is much higher for a launch with an E

c

of 30 × 10

−6

for debris than an E

c

of 30 × 10

−6

for a toxic release due to the nature of the two different hazards. In other words, with debris hazards, a higher percentage of the casualties are fatalities than with toxic hazards. The final and crucial shortcoming of this option is the difficulty in distinguishing between where the launch area ends and the downrange segment begins. This question might not be critical for a coastal range where the physical boundary between land and sea makes for a logical divider. However, no such physical partition exists for an inland launch site.

Under a fifth option, a launch operator would have been required to aggregate overall risks into a single maximum E

c

while also capping the maximum allowable risk associated with any one hazard category. Since this option would not have required accumulation, a risk assessment would have required six separate E

c

calculations for each licensed launch. Launch operators would have needed to calculate an E

c

value for each of the three hazard categories for the launch area and an E

c

value for each of the three hazard categories for the downrange portion of the launch resulting in a total of six E

c

values. This plan would have required each of the six E

c

values to meet the individual cap while requiring the sum of the six values to meet the total allowable aggregate E

c

value. The major benefit of this option would have been the ability to recognize the differences between the three main hazard categories while still capping the maximum allowable overall risk level. Unfortunately, not accumulating risks could lead to problems in defining the point in flight where the launch area ends and the downrange segment begins as discussed under the previous option.

The risk assessment proposed under a sixth option would have been very similar to those outlined in the preceding paragraph in that it would have aggregated overall risks into a single maximum E

c

, as well as capping the risk of each hazard separately; however, the cap on the maximum allowable risk associated with any one hazard category would have been on the accumulation of launch area and downrange risks for each hazard. This option would have effectively reduced the number of separate expected casualty values from six to three. This option would not have offered any significant benefit over the other options considered and involves the shortcomings associated with aggregation.

Under a seventh option, one set of risk criteria would have been developed for new vehicles while a separate set would have been developed for mature vehicles. This option would have allowed the FAA and the launch operators to recognize the role that operational experience with a particular launch system plays in reducing the level of uncertainty involved in calculating the risk associated with launching a particular vehicle. However, the differences between new and mature vehicles are already addressed under current practice by accounting for the demonstrated reliability of different launch vehicles. Currently, there are no accepted definitions for new and mature launch vehicles.

In summary, the FAA proposes to adopt the Common Standards Working Group determination that, for the reasons discussed above, risk should be limited by hazard. The FAA would limit the risk permitted for debris, far field blast overpressure and toxic release to an E

c

of 30 × 10

−6

for each hazard rather than an E

c

of 30 × 10

−6

for a total of all three hazards as proposed in the NPRM.

C. Debris Thresholds for Use in Flight Safety Analysis

Based on comments received, the FAA is proposing different thresholds for inert and explosive debris from those proposed in the October NPRM. The October 2000 NPRM would have required that certain probability analyses account for debris with a ballistic coefficient of three or greater. Under 417.107(c) of this SNPRM, the probability analyses would have to account for debris with a kinetic energy of 11 ft-lbs or greater at impact. For explosive debris, such as solid propellant fragments that will explode upon impact, the FAA is changing its proposal from 3.0 psi blast overpressure to blast overpressure of 1.0 psi or greater. The proposed debris thresholds would be applied when demonstrating that a launch satisfies the risk criteria for collective and individual risk of casualties to the public and the criteria for probability of impact for ships and aircraft.

In proposing requirements governing the calculations that are part of a launch operator's demonstration of compliance with the public risk criteria, the FAA's intent is to protect against casualties, the proposed definition in section 417.3 of the NPRM of which is “death or serious injury.” Not all pieces of debris have the potential to be lethal or cause a person a serious injury. Accordingly, the FAA does not intend that a probability analysis account for all debris, only that which has the potential to cause serious injury or death.

In proposed sections 417.225 and 417.227 and appendices A and B of the NPRM, the FAA proposed a methodology for conducting a debris risk analysis and analyses for defining hazard areas used to ensure compliance with the individual risk and ship and aircraft impact criteria.

See NPRM,

65 FR 64017, 14 CFR 417.225 and 227 and appendixes A and B (proposed). The NPRM proposed that these analyses account for debris with a ballistic coefficient of 3.0 or more, and the analysis would have had to account for a 3.0-psi blast overpressure radius and projected debris effects for all potentially explosive debris. At the time the NPRM was drafted, the FAA believed that these thresholds were consistent with the FAA's definition of casualty, but would not be as conservative as any such thresholds currently used at the federal ranges. However, Air Force members of the Common Standards Working Group raised the concern that any analysis that was limited to these thresholds would not account for significant potential casualties, particularly serious injuries that could result from launch vehicle debris. The FAA has come to agree with the Air Force's concern and has been working with the Air Force as part of the Common Standards Working Group and have identified appropriate thresholds for debris.

The Common Standards Working Group is continuing to explore what measures of concern are most appropriate for distinguishing casualty due to launch vehicle accidents. Improvements in modeling may provide room for better measures of what inert or explosive debris might cause a casualty. Recent models suggest that a change in the proposed measure for inert debris from ballistic coefficient to kinetic energy would be appropriate. Overpressure remains the most appropriate casualty measure for explosive debris; however, a change in the pressure level that presents a hazard would be appropriate. The FAA is proposing new thresholds that reflect the latest thresholds for inert and explosive debris that are being considered by the Common Standards Working Group. The FAA specifically requests comments on the debris thresholds proposed in this SNPRM, including any proposals for alternative approaches to estimating casualties.

The FAA is proposing that a launch operator's demonstration of compliance with the public risk criteria incorporate one of two approaches when applying the proposed thresholds for inert and explosive debris. The more sophisticated of the two approaches, and the one which would result in the more accurate casualty estimate, would require the use of probabilistic human vulnerability models. These models account for the probability of casualty to any person exposed to the threshold levels or greater for inert and explosive debris. The simpler of the two approaches would count all members of the public exposed to the threshold levels or greater as casualties. The simpler approach would result in a relatively conservative casualty estimation, which may be sufficient for a launch operator, depending on the specifics of a proposed launch. Any probabilistic casualty model used for a launch would have to be approved by the FAA during the licensing process or, if the launch is from a federal range, accepted as part of the FAA's baseline assessment of the federal launch range, as is current practice.

Probabilistic human vulnerability models estimate the likelihood of a casualty as a function of specific parameters that describe the contact with the hazard. The parameters may include kinetic energy, kinetic energy per unit area, overpressure, or toxic concentration. Probabilistic human vulnerability models possess greater fidelity than analysis approaches that employ simple conservative assumptions, such as counting every person exposed to the debris thresholds or greater as a casualty. These models possess greater fidelity because they typically account for the variability in how debris may harm different people such as infants, adults or the elderly to account for age, body weight and physical health. Probabilistic human vulnerability models also account for the variability associated with different injury mechanisms such as blunt trauma, crushing and penetration, as well as the variability of response associated with different parts of the body and body positions, such as whether a person is standing, sitting or supine. These models may account for the variability associated with fragment shape, weight and density and the inherent mathematical uncertainties associated with any probabilistic analysis. A human vulnerability model that reasonably accounts for these factors will produce more accurate casualty estimations than would the use of simple conservative assumptions. Accordingly, the use of a probabilistic human vulnerability model may prove to increase launch availability without jeopardizing public safety.

It must be noted that there are expenses associated with employing probabilistic human vulnerability models that can be avoided if the specifics of a proposed launch allow the use of a simple conservative approach. These models may possess significant development costs, including the highly specialized and knowledgeable personnel that would be involved. Such models would typically require more detailed input data. For example, in addition to knowing the number of people in a given area, the input to a probabilistic human vulnerability model could require statistics on the physical characteristics of the people and whether they are expected to be in the open or sheltered, and if sheltered, the characteristics of the shelters. A launch operator would have to weigh the costs associated with developing and using a

probabilistic human vulnerability model against the potential for increased launch availability.

Some of the probabilistic human vulnerability models currently used by the Air Force use the Abbreviated Injury Scale (AIS) of the Association for the Advancement of Automotive Medicine to define casualties, and to distinguish between serious injuries and those of lesser severity. The AIS is an anatomical scoring system that provides a means of ranking the severity of an injury and is widely used by emergency medical personnel. Within the AIS system, injuries are ranked on a scale of 1 to 6, with 1 being a minor injury, 2 moderate, 3 serious, 4 severe, 5 critical, and 6 a non-survivable injury. A scaling committee monitors the AIS evolution. A review of the current Air Force models found that they count an injury that qualifies as AIS Level 3, 4, 5, or 6 as a casualty. The Common Standards Working Group has recommended that any future casualty models used to satisfy Air Force and FAA requirements incorporate AIS Level 3 or greater as the standard for distinguishing casualties from injuries of lessor severity. When using the AIS for the purpose of casualty modeling, any injury that, due to its severity, qualifies as AIS Level 3, 4, 5, or 6 would be counted as a casualty. The FAA agrees that the use of AIS Level 3 or greater is appropriate for describing a medical condition sufficiently to allow modeling of casualties for purposes of determining whether a launch satisfies the public risk criteria.

The FAA recognizes that the 45th Space Wing conducts risk assessment of debris with a kinetic energy of less than 11 ft-lbs for blunt trauma on occasion, but the FAA does not currently plan to codify that practice. The circumstances surrounding that approach currently appear unique to the 45th Space Wing and constitute a response to the crowds of visitors that the Eastern Range must protect for launches. Numerous debris pieces with expected impact kinetic energies of less than 11 ft-lbs may significantly contribute to the risk of a launch when population density is sufficiently high. Also, the criterion of 11 ft-lbs of expected kinetic energy at impact does not ensure protection from serious injuries due to potential penetration wounds. For the time being, however, the FAA will not address this issue. The Common Standards Working Group considered a proposal for a threshold level near 40 ft-lb/in

2

to protect against serious penetration injuries from inert debris impacts. However, the Common Standards Working Group needs more time to evaluate an appropriate debris characteristic to protect against serious penetration injuries. The FAA invites public comments on this subject.

1. Inert debris

This SNPRM reflects two changes to the debris measure proposed in the NPRM: a change of the parameter measured to establish the probability of a casualty due to debris from ballistic coefficient to kinetic energy and a possible increase in conservatism, depending on the characteristics of a debris piece, of the threshold from a ballistic coefficient of three to a kinetic energy of 11 ft-lbs. The FAA proposed, throughout the NPRM, using ballistic coefficient as a metric for human vulnerability to estimate risk from inert debris impacts. Comments received from the Air Force and its contractor, ACTA Inc., as part of the Common Standards Working Group highlighted the pitfalls of relying on that metric. These comments have persuaded the FAA that defining hazardous debris as all pieces with a ballistic coefficient (often referred to as beta) of three or greater may fail to adequately protect the public in some cases. The FAA is now changing its proposal to use kinetic energy as the metric for estimating risk to the public from inert debris at a threshold level of 11 ft-lbs.

Specifying ballistic coefficient as a criterion ignores many important factors. The velocity of a debris piece at impact is an important factor in establishing whether an injury would result, but the terminal velocity of a debris piece at impact can vary significantly depending on the altitude at impact and its ballistic coefficient. Therefore, using ballistic coefficient as a casualty measure for inert debris would not indicate the velocity of impacting debris. Additionally, a debris fragment's ballistic coefficient does not indicate its mass, which is another important factor in establishing injury potential due to impact. A heavy fragment with a large area may be lethal, even though its ballistic coefficient is less than three. Similarly, a light fragment with a small area may be harmless even though its ballistic coefficient is greater than three. For example, consider a 30 pound debris piece, such as a rocket motor case fragment, that behaves like a tumbling plate, with an aerodynamic reference area of 11 square feet and a subsonic drag coefficient of 0.9. This piece has a ballistic coefficient of about three. The terminal velocity for this piece is about 50 feet per second, or 34 miles per hour. This piece would have a kinetic energy of about 1,164 ft-lbs at impact. The NPRM asserts that “a ballistic coefficient of three correlates approximately to a hazardous debris piece possessing 58 ft-lbs of kinetic energy.”

NPRM,

65 FR 63935. The above example shows, however, that the kinetic energy of debris with a beta of three can be significantly greater than 58 foot-pounds. Accordingly, it is appropriate to consider other factors for determining whether a fragment would produce a casualty.

Inert launch vehicle debris of concern to the FAA typically threatens humans primarily from blunt trauma due to nearly vertical impact. The debris piece's potential to cause a serious injury upon impact with a person depends primarily on the mass and shape of the debris and the velocity at which it impacts. Because kinetic energy on impact accounts for these three factors, the FAA believes it to be the appropriate metric for gauging the potential for blunt trauma.

Recently published human vulnerability model results examined by the Common Standards Working Group suggest that for the general public, a kinetic energy of 11 ft-lbs at impact would be a reasonable threshold level for any analysis intending to account for virtually all serious injuries from blunt trauma. When applied as a threshold, 11 ft-lbs would represent the kinetic energy level for debris that could, depending on the specifics of an impact with a person, cause a casualty. As an example, 11 ft-lbs at impact corresponds to a one-quarter inch thick square aluminum plate with an edge length of about two inches and a weight of about 1.5 ounces impacting at a velocity of approximately 60 mph.

One must note that not every impact of debris at 11 ft-lbs or greater will necessarily result in a casualty. The probability of casualty due to such an impact is further dependent on a number of other factors specific to the debris and the impact scenario. Probabilistic human vulnerability models are often used to account for these other factors, and an analysis that employs these models will produce a more realistic casualty estimate than a deterministic analysis that counts all expected impacts of 11 ft-lbs or greater as casualties.

The choice of 11 ft-lbs as a threshold also has practical benefits. The FAA realizes that there is no standard threshold currently in use, and the human vulnerability models used at the federal ranges today may vary depending on the launch vehicle and other factors. The Air Force members of the Common Standards Working Group have indicated that the models currently used at Air Force ranges satisfy the

proposed 11-ft-lb threshold. For example, the debris model used for a Atlas IIAS launch from Cape Canaveral Air Force Station accounts for inert debris with kinetic energy at impact greater than or equal to 7 ft-lbs. A standard threshold would facilitate the development and application of more standardized models with associated efficiencies. For these reasons, the FAA is proposing to use kinetic energy as the metric for estimating the risk of casualties due to blunt trauma from inert debris impacts at a threshold level of 11 ft-lbs.

This SNPRM would require any risk analysis for blunt trauma due to launch vehicle debris to account for all potential debris with 11 ft-lbs or greater of kinetic energy at impact. The analysis would apply the relatively sophisticated approach using probabilistic models to assess the probability of casualty due to any debris with kinetic energy at impact of 11 ft-lbs or greater, or it could apply a more simple approach where each expected impact of a person with kinetic energy of 11 ft-lbs or greater would be counted as a casualty.

2. Explosive Debris

In sections 417.225 and 417.227 of the October 2000 NPRM, the FAA proposed that a flight safety analysis, a flight hazard area analysis, and a debris risk analysis had to account for a 3.0-psi blast overpressure radius or greater and projected debris effects for all potentially explosive debris. Explosive debris is debris with the potential to explode upon surface impact. At the time the NPRM was drafted, the FAA believed that this threshold was consistent with the FAA's definition of casualty and would not be more conservative than any such thresholds currently used at the federal ranges. However, comments received from the Air Force and its contractor, ACTA Inc., as part of the Common Standards Working Group indicated that there is a significant potential for casualties at blast pressures below 3.0 psi. The FAA has reviewed this issue with the Common Standards Working Group and now proposes to reduce its threshold for explosive debris to 1.0 psi.

Many factors complicate the determination of threshold blast loads from explosive debris that could cause serious injury. These factors include the substantial difference in vulnerability of people in the open and people in buildings, the substantial variability of protection afforded by various building types, the complex nature of blast wave propagation through groups of buildings or hilly terrain, the potential for far field window breakage due to atmospheric focusing of a blast wave under special conditions, and the general lack of data on casualty-blast load relationships for occupants of various building types. In addition to the direct effect that blast overpressure can have on a person, blast may cause serious injury by breaking glass that may strike a person, by blowing people down, or by collapsing a structure with people in or near it.

People in the open are generally less vulnerable to serious injury from blast loads than occupants of typical buildings, particularly if ear damage is discounted as a serious injury. However, persons standing in the open can be seriously injured as a result of being blown-down by overpressure. Blow-down potential is a function of both blast overpressure and impulse. For an explosive yield of 10,000 pounds TNT, the threshold for serious injury due to blow-down for a 70-kg person is near 1.4 psi.

The FAA recognizes that blast thresholds used currently at federal ranges may vary depending on the analysis being performed and the specifics associated with the people and property being protected. The October 2000 NPRM's proposal to address the risk associated with 3.0-psi overpressure would have addressed risks only to someone standing outside in the open, a typical assumption for overflight risk analysis. The ranges pointed out that this failed to account for risks to persons in or near a building or other structures. Glass can break at 1.0 psi—or even less—which means that a person in a building is at risk from flying glass shards or other secondary hazards and may be more at risk than a person in the open. The current practice at the ranges accounts for such secondary hazards of explosive debris. The Department of Defense Explosive Safety Board (DDESB) approves the siting of buildings that may be subject to approximately 1 psi over pressure level in the event of an accident. Additionally, the Air Force launch ranges use 1.0 psi to determine a hit to ships for probability of impact calculations. Accordingly, the Common Standards Working Group has reviewed the casualty models and analysis processes used at the Air Force ranges and concluded that the use of 1.0 psi as a threshold for explosive debris would be consistent overall with current practice at those ranges and in the explosive safety community at large.

Although the FAA is proposing overpressure as a threshold parameter, blast effects on humans, especially building occupants, are generally sensitive to the positive phase impulse, as well as the peak overpressure, of a blast load. For example, an explosion with a 50,000-lb TNT equivalent from a launch accident would produce on the order of a 1% probability of serious injury for occupants of typical buildings in the United States located at the 1.0-psi overpressure radius from the source of the blast. However, a more typical explosion (1000-lb TNT equivalent) from a launch accident would produce less than a 0.01% probability of serious injury in the same circumstances. It is important to note that these estimates account for the probability of serious injury due to broken glass shards propelled by the blast and assumes the occupants are equally likely to be anywhere in the building. The difference in the probability of serious injury in the two examples is primarily due to the greater impulse of a large explosion compared to one with a lesser yield. However, the probability of serious injury in both cases at the 1.0-psi overpressure radius is relatively small. Most typical impacts of explosive launch vehicle debris would result in small yields, far below a 50,000-lb TNT equivalent; therefore using a 1.0-psi peak incident overpressure level as a threshold in a simple explosive overpressure vulnerability model would, the FAA believes, capture any overpressure which would cause serious injury while at the same time account for the role played by the impulse of the blast as well.

When applying the 1.0-psi threshold, any probability analysis would have to account for a 1.0-psi blast overpressure radius for all potentially explosive impacting debris. The analysis may apply a relatively sophisticated approach that uses probabilistic models to determine casualty due to any blast overpressures of 1.0-psi or greater or apply a simpler approach that counts all people within the 1.0-psi overpressure radius as a casualty. When using the simple approach, the peak incident overpressure would be computed with the Kingery-Bulmash relationship, without regard to sheltering, reflections, or atmospheric effects. For persons located in buildings, the peak incident overpressure would be computed at the shortest distance between the building and the blast source. A person would be considered a casualty when located anywhere in a building subjected to peak incident overpressure equal to or greater than 1.0 psi.

The FAA anticipates that launch operators launching smaller vehicles, such as Pegasus Taurus, will be able to take advantage of the simple approach. Launch operators conducting launches of larger vehicles would likely resort to use of probabilistic models. The FAA

requests comments on the proposed debris thresholds and their application, which allows for both simple and sophisticated analysis methods. Because the FAA considers the proposed debris thresholds and their application to be consistent with current practices at the federal ranges it does not anticipate cost impacts, but requests comments on this point.

IV. Issues of Concern to Commenters

A. Authority and Need for Rulemaking

Some commenters questioned the FAA's authority to conduct this rulemaking, and whether it was consistent with Congressional intent. They also questioned its necessity. The FAA has the authority to conduct this rulemaking,

7

and codification of the safety requirements is necessary. The statute and the legislative history support the proposed codification of launch safety requirements. The rulemaking is necessary to identify genuine and universal safety requirements, which includes identifying and codifying the intent behind existing range safety requirements. Currently, federal requirements consist of a mix of safety and mission requirements. Some are available readily to the public. Others are typically only in the possession of range analysts. This rulemaking identifies those requirements with which a launch operator must comply under current practice. The FAA intends that streamlined performance requirements offer the same high level of safety and the flexibility of current practice. Finally, the FAA is concerned that adopting the suggestion to only apply proposed part 417 to non-federal launch sites could result in confusion regarding safety requirements at the federal ranges. This discussion describes the reasons for the FAA's position that it has the authority to conduct this rulemaking, that the rulemaking is consistent with Congressional intent, and that it is necessary for public safety.

7

Accord

JC Vol. I at iii (“the FAA has the flexibility under the CSLA to develop and issue its own rules”), Lockheed at 2, 5.

1. Authority for Rulemaking

The Joint Commenters assert that the FAA's regulation of launch safety is not statutorily mandated, and is inconsistent with the Act's “finding that private sector launch and associated services should be regulated only to the extent necessary to protect, among other things, the public health and safety.” JC Vol. I at ii. In support of this argument, the commenters point to the FAA's authority to accept the assistance of other executive agencies in carrying out the Act, the Air Force's comprehensive safety requirements and the safety record achieved at the ranges. JC Vol. I at ii; Lockheed at 6. Lockheed Martin and other commenters suggest that the rulemaking is inconsistent with Congressional intent, as embodied in legislative history, to streamline the licensing process. JC Vol. I at iii; Lockheed at 6.

Congress found that the FAA should “only to the extent necessary, regulate * * * launches, reentries and services to ensure compliance with international obligations of the United States and to protect the public health and safety, safety of property, national security and foreign policy interests of the United States.” 49 U.S.C. 70101(a)(7). This rulemaking would identify and codify regulations containing the standards that protect public safety. Congress also found that the provision of launch services would be “facilitated by stable, minimal, and appropriate regulatory guidelines that are fairly and expeditiously applied.” 49 U.S.C. 70101(a)(6).

The commenters acknowledge that the FAA has the authority under 49 U.S.C. 70101-70121 (referred to as “Chapter 701” or “the Act”) to issue safety regulations. JC Vol. I at iii;

accord

Lockheed at 2, 5. Accordingly, the commenters' position that the rulemaking fails to satisfy the Act appears to be based on the belief that the FAA's rulemaking may somehow be inconsistent with Congressional intent. As a preliminary matter, the FAA notes that intent becomes a matter of significance to statutory interpretation only when the statute itself is unclear. The Act is not unclear.

Chapter 701 authorizes the Department of Transportation and thus the FAA, through delegations, to oversee, license and regulate commercial launch and reentry activities and the operation of launch and reentry sites as carried out by U.S. citizens or within the United States. 49 U.S.C. 70103, 70104, 70105. The Act directs the FAA to exercise this responsibility consistent with public health and safety, safety of property, and the national security and foreign policy interests of the United States. 49 U.S.C. 70105.

2. Congressional Intent

Despite the commenters' claims to the contrary, review of legislative history shows that the FAA's rulemaking would satisfy Congressional intent. Review of the commenters' proposed interpretation of Congressional intent shows that Congress did not attempt to foreclose this rulemaking. Instead, some of the comments take legislative history out of context and argue that observations offered for a different day apply to the current situation. The comments attempt to portray Congressional intent as opposing a rulemaking—such as this—that codifies safety requirements. As explained below, the FAA does not share this interpretation.

Even if intent were an issue, the best expression of Congressional intent is contained in the language of the Act itself. This meaning may be discerned by analyzing the design of the statute as a whole. The Act itself specifically created a civilian regulatory regime for safety. Congress in 1984 neither foresaw nor forbade the conduct of this rulemaking. Instead, Congress gave the FAA responsibility for safety and authority to conduct rulemakings. Where Congress intended to bar duplication of responsibilities in the Act, it did so explicitly.

See, e.g.,

49 U.S.C. 70117(b); S. Rep. No. 98-656, 15 (1984)(explaining that because regulatory regimes for communications satellites and land remote sensing satellites already exist, a duplicative process would be unnecessary). The regulatory regime for launch safety is that of the FAA. Had Congress viewed the Air Force's safety oversight as sufficient to require no codification of safety standards, Congress could have done so as explicitly as it ensured against duplication of the roles of the Federal Communications Commission and the National Oceanic and Atmospheric Administration.

8

Moreover, Congress could have failed to vest safety responsibility in the FAA. Congress did neither of these things.

8

That the FAA may seek the assistance of the head of another executive agency does not accomplish nearly as much as the commenters suggest. Given the FAA's continued reliance on the federal launch ranges, now and for the foreseeable future, it is certainly a statutory provision of which the FAA is aware, but not one that stands in the way of the FAA identifying safety standards through rulemaking.

Lockheed Martin separately urges reliance on a Senate report that accompanied passage of the original Commercial Space Launch Act to support its claim that this rulemaking runs counter to Congressional intent. Lockheed at 6. The cited legislative history does not go as far as Lockheed recommends. Lockheed states, that “Congress stated unambiguously that the Act, and implementation of the Act, should reduce the regulatory burden for commercial launch operators and that the authority of * * * the

Secretary* * * to issue additional requirements and regulations must conform with the Congress' expressed desire to streamline the licensing process for commercial launch * * *.” Lockheed at 6. The FAA first notes that what Lockheed cites in support of its assertion is not the language of the statute itself, but the regulatory impact statement of the Senate Report. S. Rep. No. 656, 98th Cong., 2d Sess., 5 (1984),

reprinted in

1984 U.S.C.C.A.N. 5328, 5332. More significant, however, is the fact that Lockheed has added a word, the word “must,” to the cited language, thereby changing the meaning of the statement from one of description to one of admonition. Accordingly, the Senate report does not have the meaning that Lockheed would ascribe to it. Instead, in discussing the new authority conferred upon the Secretary, the report notes that the Secretary's authority “to issue additional requirements and regulations conforms with the Committee's desire to streamline the commercial launch and launch operations process and to facilitate compliance with the required regulations.” Sen. Rep. No. 656 at 5. A better interpretation is that the Committee thought that the new authority streamlined the existing situation.

Indeed, the situation at that time was a difficult one for a launch operator. Prior to passage of the Act, a launch operator, for example, had to obtain an export license under the International Traffic in Arms Regulations. Sen. Rep. No. 656 at 37. This was why the legislation gave the Secretary “exclusive licensing authority” for commercial launch. Sen. Rep. No. 656 at 5, 37. The FAA's interpretation is also more consistent than Lockheed's with the Committee's other statement to the effect that “the legislation would provide for a more stable regulatory environment than that which currently exists.* * *” Sen. Rep. No. 656 at 6. The regulatory environment that existed at the time would have required a launch operator to satisfy the requirements of numerous federal agencies.

9

9

Contemporaneous and historical accounts describe the regulatory environment with which a launch operator had to comply as consisting of 18 federal agencies and 22 federal statutes. Kay, W.D., “Space Policy Redefined: The Reagan Administration and the Commercialization of Space,”

Business and Economic History,

237-247 (Fall 1998); “Industry Observer,”

Av. Week & Space Technology,

15 (Oct. 22, 1984).

Likewise, although Lockheed does accurately describe Congressional encouragement to avoid duplicative and unnecessary regulation, (Lockheed at 6 (citing Sen. Rep. No. 656 at 3, 19)), the FAA's work with the Air Force in achieving common standards is designed to attain that very goal. In summary, the history at the time indicates, and the actual words used by the Committee demonstrate that Congress intended to streamline the existing regulatory process, not to argue against the possible future codification of safety requirements.

3. Necessity for This Rulemaking

Although some commenters assert that this rulemaking is not necessary to protect public safety, Chapter 701 directs the FAA to regulate to the extent necessary to protect public safety. The FAA believes that if a launch operator is to be expected to satisfy safety requirements, those requirements must be clear, open and published. In the October 2000 NPRM, the FAA announced that it considered the range safety requirements necessary because they were the requirements with which the ranges had achieved their level of safety. The FAA continues to find that the proposed requirements are necessary to achieving safety. The following discussion provides the reasons for the FAA's position.

Launch operators should achieve the same level of safety, regardless of whether they launch from a federal launch range or a non-federal launch site. Safety standards should be common between the FAA and the ranges. Most significantly, the FAA must identify the standards by which it judges safety; and, having identified those standards, the FAA must provide full disclosure that those standards apply at both federal launch ranges and at non-federal launch sites. Not only has the FAA identified its own proposed standards, in doing so, it has provided the additional benefit of identifying what the federal launch ranges themselves in fact require, and the standards they impose on launch operators through their own internal requirements.

a. Genuine and Universal Safety Requirements

Different federal launch ranges have implemented different approaches to achieving the same safety goals. The FAA proposes to codify the intent behind these different requirements where possible. In the interest of achieving universal applicability, namely, requirements that can apply regardless of differences in geography, mission, meteorological conditions and other factors, the FAA worked with the ranges to identify the underlying intent. Additionally, some of the range requirements documents require a launch operator to provide data that the range, in turn, subjects to standards contained in internal range documents. The internal standards are available upon request and provide greater insight into the intent behind particular information or safety requirements. This rulemaking would codify those as well.

Although, generally, Lockheed Martin maintains that the proposed requirements are new and different from EWR 127-1, Lockheed Martin stated that it would object as well to the proposed requirements, even if it thought that the FAA could succeed in codifying the Air Force requirements, on the grounds that those requirements are not the “real, ultimate requirements” of public safety, which the Air Force is able to accept through “tailoring.”

10

Lockheed at 3. The FAA's intent, however, has been to determine what those “real, ultimate requirements” are, so that they may be shared and codified as performance standards.

11

For example, the standards governing the creation of impact limit lines are not contained in EWR 127-1, but may be found instead in a flight safety analysis handbook,

Flight Control and Analysis General Reference Handbook,

RTI Rep. No. RTI/6762/03-02F (Apr. 24, 1997). This rulemaking attempts to unveil those requirements. Indeed, the Administrative Procedure Act directs that an agency's requirements be public. 5 U.S.C. 552(a)(1)(D).

10

“Tailoring,” as explained by EWR 127-1, permits the preparation of an individually “tailored” requirements document to ensure that only applicable or alternative equivalent requirements are levied upon a launch vehicle program.” EWR 127-1, 1-21, 1.6.3 (Oct. 31, 1997).

11

An unintended consequence of translating some of the details of EWR 127-1 into performance requirements has been to appear to create new requirements.

See, e.g.,

discussion of surveillance requirements, IV.B. Additionally, as described in the NPRM and elsewhere here, the FAA has proposed more detailed requirements to serve as a roadmap for what the FAA considers demonstrates satisfaction of those performance requirements, and against which alternatives might be measured.

The FAA's requirements may appear different from EWR 127-1 because they attempt to capture both the written requirements of EWR 127-1 and how the ranges have implemented those requirements. The FAA, aware of the safety expertise resident at the federal launch ranges, consulted with the ranges and reviewed the ranges' own requirements, as embodied in the EWR 127-1 and in NASA's Range Safety Manual for Goddard Space Flight Center (GSFC)/Wallops Flight Facility (WFF), RSM-93. Range safety personnel advised the FAA that not all of their requirements were enforced in a standardized manner because the ranges

had granted waivers, deviations and “meets intent certifications” to launch operators in response to the requests of the launch operators for relief. The ranges have also used “tailoring.” Typically, this involves not imposing requirements that do not apply, and rewriting any requirement where the intent of the requirement is satisfied through other means.

EWR 127-1,

Appendix 1A, 1-23 (Dec. 31, 1997).

The FAA is building in similar flexibility by recognizing where the ranges have been willing to grant relief and incorporating those determinations into the requirements as proposed through this rulemaking so that particular non-compliances would no longer require waivers. For example, the lot acceptance and qualification test requirements for percussion activated device (PAD) primer charges used in a flight termination system that were proposed in the FAA's October 2000 NPRM (proposed 14 CFR E417.31) are relaxed in comparison to the Air Force's current version of EWR 127-1. The NPRM proposes to reduce the number of units to be tested and to reduce the types of tests to be conducted. These proposed changes from current Air Force requirements are based on lessons learned over the past few years and earlier decisions made by Air Force range safety to waive or tailor such requirements for existing launch vehicle systems. One launch operator that currently launches from Air Force ranges, having seen the proposed PAD requirements in the FAA NPRM has since approached the Air Force with a request to apply the FAA requirements to its launch vehicle. These improvements and others identified during the development of the October 2000 NPRM are now being incorporated into the Air Force's new Space Command manual that will replace 127-1. Thus, in many ways, particularly with respect to the particulars of the flight safety system requirements, the FAA believes that this rulemaking may provide a more comprehensive and streamlined version of the ranges' own requirements.

During the discussions between the ranges and the FAA regarding safety requirements for non-federal launch sites, the FAA attempted to identify the common underpinnings of the range requirements to achieve more universal applicability, particularly in the area of flight safety analysis. Flight safety analyses that the Air Force ranges apply on each coast are directed toward each coast's geography, meteorological conditions, and mission profiles. As the FAA worked to make the range requirements more general so that they might apply wherever a launch took place, the question arose as to why the safety requirements for licensed launch operators should differ from site to site. No good reason was evident. Moreover, with the goal of achieving universal applicability of as many of the requirements as possible by identifying the common intent underlying different approaches to similar safety questions, permitting different standards seemed unnecessary.

In the course of these discussions, the ranges and the FAA saw a number of benefits to having common standards. Common standards would provide launch operators certainty in planning. Common standards would permit a body of expertise to support those standards. In the unlikely event that the Air Force ever pulled back from its oversight of commercial activity, a step the Air Force has contemplated within past years, standards will already be in place for FAA licensed launches from a federal range. Also, it might be difficult to justify imposing different standards of safety on licensed launch operators based merely on whether the launch took place from a non-federal launch site or from a federal launch range.

In summary, the applicability of part 417 to all licensed launches, regardless of their launch location is necessary. Universality ensures a single standard of safety. Publication of the requirements currently in place permits a launch operator to know and plan for the requirements with which it must comply. The comments' suggestion that part 417 only apply to non-federal launch sites is based on a misperception that the FAA has proposed “significant changes,” in the form of new, more conservative requirements, JC Vol. I at 8, 12, to a proven process, when, to the contrary, this rulemaking only identifies and proposes to codify the intent underlying existing requirements in a performance standard format.

12

This is not to say that there were no problems with the regulations proposed in the October 2000 NPRM. The commenters identified certain areas of the FAA's proposed regulatory text that might be interpreted as more conservative than current practice at the federal ranges. This was not the FAA's intent and the FAA is working to make the appropriate adjustments, some of which are presented in this SNPRM.

12

The presence of design requirements shows what the FAA proposes to find acceptable. Launch operators should note that the opportunity to provide a clear and convincing demonstration of an equivalent level of safety is embedded in each design oriented requirement.

See also

NPRM, 65 FR 63940-41 (discussing reasons for acceptability of Sea Launch's comparable flight safety system).

b. Identification of Standards and Resulting Application

Commenters' suggestion that the FAA refrain from applying part 417 to launch from a federal launch range does not address the need to identify safety standards, fails to recognize that this exercise has identified those standards, and falls prey to the law of unintended consequences. Having identified its standards, the FAA does not believe that it would be helpful to claim that they do not apply. The logic of how the FAA evaluates the acceptability of the federal launch ranges should alleviate concerns over any seeming duplication between the FAA and the Air Force. The Joint Commenters proposed that the FAA apply part 417 only to non-federal launch sites. For the FAA to agree that part 417 would only apply at non-federal launch sites would, however, be confusing at best and misleading at worst.

Part 417 would contain the standards by which the FAA would assess the adequacy of both a licensee and a federal launch range. The FAA assesses a launch operator through the licensing process and a federal launch range through a baseline assessment. Because the FAA obtained the standards in part 417 from the federal launch ranges own standards and practices, the FAA, of course, anticipates that the federal launch ranges will satisfy proposed part 417. Nonetheless, whether through changes in Air Force or NASA policy or because of the failure of a range safety system, it is conceivable that some element of range safety might not satisfy the ranges' own current requirements. In fact, the ranges advise that they may, from time to time, waive requirements for their own equipment, and a launch operator may remain unaware of this waiver.

Even if the FAA acquiesced in the commenters' proposal and declared that part 417 only applied at non-federal launch sites, it would still have to use some set of standards against which to measure the continued adequacy of the federal launch ranges whenever the FAA updated its baseline assessments. Those standards would be found in part 417. Accordingly, to say that part 417 did not “apply” at the federal launch ranges might confuse some into thinking that part 417 had no applicability whatsoever, even in the baseline assessment context. Others might believe that the FAA was misleading them regarding the applicability of part 417 at federal launch ranges given that the FAA would assess the adequacy of

the ranges against part 417. The FAA does not consider it advisable to create such confusion. None of the points raised by the comments address this fundamental issue, and the FAA invites the public to take this additional opportunity to present alternatives that take this consideration into account. It is one that the FAA does not believe it can ignore, but recognizes that those with a different perspective may be able to offer insights currently unavailable to the FAA.

Because the range safety requirements are part of how the ranges have achieved their high level of safety, the FAA considers those requirements necessary for continuing to achieve that same level of safety for FAA-licensed launches at both non-federal launch sites and federal launch ranges. The FAA and the commenters take away different lessons from the past safety records. Although the Joint Commenters point to the safety record of the past as justification for not requiring further regulation, the FAA looks to the safety record of the past and attributes that successful record, in some measure, to the launch safety requirements themselves. Accordingly, when the FAA began its own attempt to codify requirements that would govern launch safety at non-federal launch sites, it looked first to the ranges' own requirements and the FAA has continually worked with the Air Force to ensure that in the future the two agencies' requirements are consistent and do not conflict.

c. Implementation

Other comments received in response to the NPRM include concerns about how the FAA would implement the proposed requirements at the federal launch ranges, whether the FAA would grant waivers as readily as the ranges, and whether FAA oversight would result in reduced flexibility, both in meeting the intent rather than the letter of the requirement and in terms of operational flexibility. Because the Act directs the FAA to encourage, facilitate and promote private sector launches, 49 U.S.C. 70103(b), the Joint Commenters indicate that the FAA should streamline its licensing and regulating regime by continuing to rely on the ranges for the implementation of launch safety requirements. JC Vol. I at ii.

One of the reasons the commenters argue that this rulemaking is not necessary is because they fear that the FAA's identification of the safety standards would constitute duplication of oversight. This is not so much a concern regarding the necessity of having safety standards as a concern with their implementation. The comments recommend that a single entity be responsible for the safety of licensed launches.

A review of what the FAA proposed in the NPRM should allay these concerns. Of first and foremost importance, the commenters should note that the FAA intends no duplication of oversight. The proposed standards themselves, which were derived from range requirements and practices, will apply to all licensed launches, regardless of the location of the launch site. Applicability of standards is different, however, from duplication of oversight. Oversight means inspection, monitoring and otherwise checking whether a licensee is in compliance with the requirements of the Act, the FAA's regulations and its license. As the FAA noted in the October 2000 NPRM, the FAA does not now and does not intend through this rulemaking to duplicate the work, evaluation, inspection and monitoring conducted by the federal launch ranges. NPRM, 65 FR 63924. The FAA relies on its baseline assessments of the ranges, and those baseline assessments have found the ranges safety requirements acceptable. NPRM, 65 FR 63924. Likewise, the FAA has found acceptable the ranges' implementation of those requirements. There are situations, however, where the ranges may, for reasons of their own, change their support for licensed launches. In such a case, the launch operator would likely have to perform its safety work itself. Also, as noted, if “a documented range safety service has changed significantly or has experienced a recent failure” the burden of demonstrating safety at a range shifts to the launch operator. NPRM, 65 FR 63924. The FAA sees little change from current practice in this regard.

The FAA does not agree that this rulemaking will result in loss of flexibility. The NPRM would allow for flexibility through the use of performance requirements, where appropriate. The FAA worked extensively with federal range safety personnel to develop common launch safety requirements that refine and adapt many of the current federal range standards into performance requirements.

For each specific safety issue, the NPRM may contain different levels of performance requirements as needed to respond to the complexity of space launch systems and the potential for negative consequences to public safety. For example, a flight termination system is one of the most critical systems on a launch vehicle for ensuring public safety. Hence, to ensure flight termination system reliability the NPRM contains comprehensive design and test performance requirements for the systems, components and piece parts. Also, the FAA does not attempt to mandate requirements related to achieving the success of the mission, and will permit the launch operator to accept its own risks on that score, where there is no impact on public safety. For example, where safety is ensured by the working of the flight safety system, the NPRM calls for a launch operator to provide for launch vehicle tracking without specifying detailed requirements to ensure reliable tracking. Aside from some general performance requirements, the reliability of the tracking system is left to the launch operator with the understanding that if all tracking data is lost during flight the flight termination system will be used to destroy the vehicle. For a licensed launch from a federal range, the launch operator typically relies on the range to provide reliable launch vehicle tracking. The FAA's proposed requirements do not dictate a change from such practices.

In addition to the use of performance requirements, the FAA proposes to allow flexibility by permitting a license or a license modification applicant to demonstrate an equivalent level of safety for a proposed alternative approach. Although the proposed regulations would provide the requirements with which a licensee must comply, the FAA anticipated that a launch operator might wish to employ alternative means of achieving an equivalent level of safety. In that case, if a launch operator clearly and convincingly demonstrated an equivalent level of safety, the FAA would accept the alternative. Once accepted, an alternative approach would become part of the terms of the license, and the FAA would consider making the substitute available for the benefit of others through the advisory circular process or some other means. The FAA has also demonstrated its flexibility with the licensing of launches such as those of Sea Launch, where there are a number of aspects that do not conform to current practice at U.S. launch ranges. Also, the FAA recognizes that the NPRM represents only a version of current practice: the safety methods used at the U.S. ranges often differ from one another. The FAA has worked with the federal range organizations to develop common launch safety requirements that present a more generalized description of the current practices at the ranges. Where there may

be differences between the methodologies defined in the NPRM and those used at a federal range, the current practices at the federal ranges typically do provide an equivalent level of safety to the NPRM.

The Joint Commenters expressed concern that if the NPRM were implemented as drafted, launch operators on federal ranges would have to demonstrate compliance with two sets of requirements overseen and administered by two separate and independent government agencies. The commenters believe that this would be cumbersome and inevitably would lead to costly and duplicative safety efforts with no appreciable increase in public safety. The FAA is continuing to work with the federal ranges to eliminate these concerns. Under current regulations, the FAA issues a license to an applicant proposing to launch from a federal launch range if the applicant satisfies the requirements of part 415, subpart C, of the licensing regulations and has contracted with the federal launch range for the provision of safety-related launch services and property, as long as the safety related launch services and proposed use of property are within the experience of the federal launch range. The NPRM does not propose to change this overall approach. The FAA does not duplicate analyses performed by the federal launch ranges or routinely review those analyses during the launch safety review. Instead, the FAA relies on its knowledge of the range processes as documented in the FAA's baseline assessments. The FAA's baseline assessments document each federal launch range's capabilities, safety program, standards and policies. The January 16, 2001 Memorandum of Agreement between the FAA and the U.S. Air Force explains the roles and responsibilities of the Air Force and the FAA for overseeing safety of commercial space launch and reentry.

The Joint Commenters expressed doubt that the Air Force and the FAA would be able to work together in an efficient way toward a common goal. The commenters indicated that if the FAA NPRM were implemented, it would result in competing safety requirements at the Air Force ranges. These concerns are unfounded. The Air Force and the FAA remain committed to the partnership outlined in the MOA and to ensuring that competing safety requirements do not exist. The MOA calls for developing common launch safety requirements and for coordinating the common requirements. The Common Standards Working Group is continuing to participate in developing the FAA's final rule and a revised Air Force range safety requirements document. The common standards will be contained in the Code of Federal Regulations and Air Force documents. FAA rules appear in the Code of Federal Regulations. The Air Force range safety requirements, which must address a broader range of issues, will encompass the same common launch safety requirements as well as other issues unrelated to launch safety. When the final Air Force and FAA documents are in place, a licensed launch operator at an Air Force range, in day-to-day practice would only need to work from the Air Force's range safety document so long as the FAA's launch safety requirements are contained there as well. This would be no different from the process in place for licensed launch operators today. The FAA and the Air Force are also working under the MOA to develop processes for implementing the common launch safety requirements together, including coordinated review and disposition of requests for relief from common requirements, as explained in section IV.C of this discussion. Although part 417 would contain the legal requirements with which a licensee must comply, when launching at a federal range, a licensed launch operator's primary day-to-day interface would continue to be the federal range. A unified launch safety community that includes FAA representatives will address any issues that may arise to ensure that all federal launch range and FAA licensing concerns are addressed.

B. Cost Impacts on Licensed Launches From Federal Launch Ranges

Comments in response to the October 2000 NPRM indicate that the launch industry has concerns about how the proposed rule would work, and how the FAA and the Air Force work together. The concerns have led to a perception that this rulemaking will result in significantly increased costs for the launch operators. To address some of these concerns, the FAA is proposing changes to the October 2000 NPRM in this SNPRM, as described earlier in this preamble. The FAA also hopes to clarify some of these issues. Some possible cost impacts identified by the commenters have led the FAA to revisit whether its proposed requirements actually captured current practice. The majority of the concerns underlying the costs the launch operators attribute to this rulemaking are, however, unfounded. The following discussion explains why.

1. Commenters believed some of the proposed requirements were new. Commenters may not be fully familiar with the precise nature of the safety services the ranges provide, and thus believe that some of the proposed requirements in the October 2000 NPRM are new, but, in fact, those requirements are already in place. Similarly, launch operators believe that a number of the more abstract expressions of different range requirements are new. Instead, a number of them are the FAA's proposed attempt to describe the common standards underlying different approaches taken at different federal launch ranges.

2. The launch operators believe that this rulemaking changes their legal responsibility for safety. They are, however, already responsible for safety under the statute and their licenses, and they would not be required to duplicate the work of the federal ranges as a result of this rulemaking.

3. Some of the commenters think that the more onerous requirements governing how to obtain a license apply to federal range launches. The licensing requirements proposed in this rulemaking, however, would apply to an applicant obtaining a license to launch from a non-federal launch site.

4. Commenters expressed concern over a loss of flexibility. These concerns should be allayed by the FAA's proposal to permit a demonstration of an equivalent level of safety, the grandfathering proposal and waiver coordination.

5. Although not a concern raised by the commenters, the FAA requests comment on the neighboring launch operator issue addressed below.

All this is not to say that the comments lack merit. There are a number of instances where the FAA wishes to make changes based upon the comments received. To determine whether it captures current practice, the FAA will revisit the issues raised by such comments. Some changes have already been proposed through this SNPRM, and the FAA requests views on whether the commenters still assign costs to these matters.

13

As one example, commenters attributed an array of costs to the FAA's original proposal not to grandfather. If the launch operators satisfy the FAA's proposed conditions, these same launch operators may be eligible for the FAA's

version of “grandfathering” and need no longer anticipate costs associated with making changes in their operations.

13

See

Boeing Costs at 2, 3, 4 (first and second comments), 9 (first comment), 11 (fifth comment), 12 (first and second comments), 22 (second comment), 23 (fourth comment), 24 (first and sixth comments), 25 (first and second comments), 27 (second comment), 28 (first comment), 29 (first and third comments), 33 (second and third comments), 37 (first and third comments), 40 (all comments); Lockheed Costs Estimates 2, 19 and 26; Orbital Cost Impact Assessment at 6 (item 2 regarding aggregation, items 4, 5 and 7); Sea Launch Costs at 23, 24 (second comment labeled a, b and c).

In addressing these cost issues, the FAA found several comments that it does not understand. Because this SNPRM provides an opportunity through its additional comment period to obtain clarity, the FAA urges those commenters who provided the cited comments to assist the FAA in better understanding their differences.

14

14

See

Boeing at 10 (fifth comment); 23 (second, third and fourth comments); 24 (second comment); 27 (first comment); 28 (second comment); Orbital Cost Impact Assessment at 6 (items 3b, 9 and 13-16); Sea Launch Costs at 2 (first and second comments), 7, 10 (first comment), 11, 18-19, 22, 36.

1. “New” Requirements

Some launch operators attributed costs to their launches from federal launch ranges in the belief that the FAA proposed new requirements that the launch operators would not be able to satisfy. The confusion appears to stem from several sources, including the FAA's more generalized description of different range practices, and unfamiliar requirements contained in Air Force handbooks. For instance, in the NPRM, the FAA proposed a number of requirements that attempted to reconcile the different approaches of the Eastern and Western Ranges and thus restated the requirements in a more abstract or generalized fashion. Additionally, the comments appear to indicate a lack of familiarity with some of the particulars of the range's analyses requirements and existing FAA requirements. The last category of seemingly new requirements appears to consist, to the best of the FAA's ability to interpret them, of misreadings of the proposal.

Commenters attributed a number of costs to generalized expressions of different range practices. For example, in the NPRM, proposed sections 417.113(b)(2), 417.121(f), 417.225, and appendix C, 417.5(g), (h) and (i) would determine whether downrange surveillance was needed on the day of launch. To protect ship traffic down range of the launch area, the FAA proposed that a launch operator identify where its launch vehicle's stages or other planned ejected debris would impact, determine the corresponding hazard area or areas

15

, use statistical ship density data to determine whether the launch operator needed to survey the downrange hazard areas for ships, and if downrange surveillance was necessary, determine whether risks at the time of flight required that the launch operator wait until any ships departed from downrange ship hazard areas before initiating flight.

See

14 CFR 417.107(b)(3), 417.121, 417.225, and appendix C, C417.5(g) (proposed), 65 FR 63931 (discussion accompanying proposed regulations). A launch operator would be permitted to initiate flight only if the collective probability of impacting any ship in the downrange hazard areas with planned debris would be less than or equal to 1 x 10

−5

. 65 FR 63931. If a launch operator demonstrated, using statistical ship density data and the formula provided in the NPRM, that the collective ship-hit probability in the downrange flight hazard areas was less than or equal to 1 x 10

−5

, the launch operator would not have to survey the downrange hazard areas on the day of flight.

Id.

In their comments, launch operators expressed concern over this proposed standard.

15

For both ships and aircraft, the FAA proposed in the NPRM and proposes in appendix A of this SNPRM section A417.23(k) and (1) that an impact hazard area for ships down range of the launch site would consist of an area centered on the planned impact point and defined by the larger of the three-sigma impact dispersion ellipse or an ellipse with the same semi-major and semi-minor axis ratio as the impact dispersion,w here, if a ship were located on the boundary of the ellipse, the probability of hitting the ship would be less than or equal to 1 × 10

−5

. Each aircraft hazard area downrange of the launch site would encompass an air space region, from an altitude of 60,000 feet to impact on the Earth's surface, that contains the larger of the three-sigma drag impact dispersion or an ellipse with the same semi-major and semi-minor axis ratio as the impact dispersion,w here, if an aircraft were located on the boundary of the ellipse the probability of hitting the aircraft would be less than or equal to 1 x 10

−8

.

Commenters claimed that the proposed requirement was new and would mean that launch operators would have to survey downrange impact areas for launches from the Eastern Range. JC Vol. II at 50, 83;

see

JC Vol. I at 8. The FAA does not agree with either of these assertions. When preparing the NPRM, the FAA consulted extensively with both the Eastern and Western Ranges to ensure that the FAA would capture current requirements. The FAA also considered its own experience with the launches of Sea Launch. As far as the FAA is aware, the overwhelming majority of licensed launches conducted from federal launch ranges today would satisfy the FAA's proposed requirements without having to survey downrange hazard areas located in broad ocean waters.

The Joint Commenters stated that if the FAA considers the surveillance efforts of the federal launch ranges sufficient, then the FAA should not change or add the requirements. JC Vol. II at 50. According to the commenters, surveillance of multiple downrange impact hazard areas for a single launch could require multiple aircraft. JC Vol. II at 50. Mechanical problems on the surveillance craft and weather could require a scrub of the launch with resulting cost impacts.

Currently, a range surveys its launch area (which correlates to the FAA's proposed flight hazard area) for the presence of ships and aircraft prior to launch. The ranges do not typically survey downrange stage impact areas located in broad ocean waters. This does not, however, mean that the proposed requirement is new or that the ranges would not currently survey downrange impact areas if it were determined necessary to protect the public.

16

To the contrary, both the Eastern and Western Ranges have advised the FAA that range analysts have addressed the issue. The ranges have not needed to survey downrange impact areas because of the low density of ship traffic and the nature of the traffic, in broad ocean waters, where spent stages currently land. For example, unlike the recreational craft closer to shore, much of the shipping downrange for a typical launch from Cape Canaveral is commercial in nature and the ranges anticipate that those ships monitor the notices to mariners that advise of the presence of hazard areas. However, if a stage impact area proved to be located near a greater density of ship traffic that did not monitor notices to mariners as closely as commercial shipping pilots do, a range could well require surveillance at that stage impact hazard area. Downrange hazard area surveillance is often performed for launches from Wallops Flight Facility. These launches typically involve small rockets with downrange stage impacts that are relatively close to shore where there are significant numbers of pleasure craft and fishing vessels. The FAA proposes to formalize the analysis process that the ranges have been implementing, and would establish a proposed formula and threshold for determining when surveillance of down range impact areas would be necessary. The FAA believes that typical orbital launches from the federal launch ranges meet the FAA's proposed criteria, and that downrange surveillance would continue not to be required for typical launches from those ranges. The comments to the NPRM indicate that the launch operators believe the contrary. Accordingly, the FAA requests that, through the comment period, the launch

operators share the reasoning underlying their conclusion.

16

The commenters' assertion,

see

JC Vol. II at 83, that the ranges do not conduct downrange surveillance for reasons of impracticality is not consistent with what the ranges have advised the FAA. The range do not, in most cases, conduct downrange surveillance because a safety analysis shows that it is not currently necessary.

After discussion with some of the launch operators, the FAA believes that the launch operators did not recognize that the FAA, to identify requirements that can be applied to the majority of licensed launch activity, wherever it might occur, was merely articulating a more generalized, abstract version of what the ranges are already doing in order to identify the underlying intent. Accordingly, where some of the commenters attributed costs to this requirement,

17

the FAA does not, either for launches from a federal launch range or from a non-federal launch site. The surveillance issue constitutes one example of the tendency to characterize as new what were, in fact, generalized expressions of different range requirements. The commenters attributed other costs on the basis of this misconception as well.

18

17

See, e.g.,

LM Cost Impact Analysis at 3, 13, 23, 26 (proprietary).

18

See

Lockheed Cost Estimates 5 and 7; Orbital Cost Items 2, 3, 5 and 8; Sea Launch Costs at 15-16, 22.

Additionally, the comments appear to assume that many of the ranges' own internal requirements, when proposed in the NPRM, were new. A range conducts its own flight safety analyses based upon raw data provided by a launch operator. Because the launch operators may only be familiar with the data that they themselves provide the ranges, they worried that the standards that the FAA identified were new.

19

In fact, the federal ranges have been performing the analyses and satisfying these requirements on behalf of the launch operators under current practice.

19

See

Boeing Costs at 14 (first comment), 15, 16, 17 (first comment), 18, 38 (first comment); Lockheed Cost Estimates 11 and 13; Orbital Cost Impact Assessment at 6 (items 1 and 2a).

The FAA has grouped remaining concerns regarding proposed requirements that are only seemingly new into two categories. The one category consists of comments that attribute costs to existing FAA requirements.

20

The other category consists of comments that attribute costs where the commenter misread the proposed requirement.

21

20

See

Boeing Costs at 6 (first, second and third comments), 12 (third comment), 30 (second comment); Lockheed Cost Estimate 6; Sea Launch Costs at 1 (first and second comment), 4-5 (comments labeled a, j, k, n) 7 (first comment), 8 (first, second and third comments), 10 (first comment), 13 (second comment), 17 (comment labeled a) and 20.

21

See

Boeing Costs at 19 (fourth comment), 29 (fourth comment), 34 (fifth comment), 37 (second comment) and 38 (second comment); Sea Launch Costs at 2 (second comment), 5 (comments labeled 1 and m), 7 (second comment), 9 (first comment), 21 (first full comment).

2. No Change in Responsibility

As a separate issue, commenting launch operators stated that this rulemaking would change their responsibility for safety, and thus increase their costs. This was not an issue that the FAA addressed in the NPRM because the FAA already considers a launch operator responsible for safety under the statute, the regulations and its launch license.

See

14 CFR 415.71. The FAA recognizes, however, that this comment may arise from a belief that the launch operator must use its own employees, rather than continue to rely on the services provided by a federal launch range.

22

If that is the case, the FAA believes that it can set that concern to rest. Under existing 14 CFR 415.31, the FAA grants a safety approval to a launch operator proposing to launch from a federal launch range if the applicant satisfies the requirements of subpart C and has contracted with the range for the provision of safety related services. The Commercial Space Operations Support Agreement and its annex constitutes such a contract. The FAA is proposing to codify the safety requirements of the range and anticipates that the ranges will continue to satisfy those requirements. Nonetheless, to ensure that there is no remaining confusion on this score, the FAA is revising its current proposal to include a provision in proposed 14 CFR 417.203(d) that if a launch operator has contracted with a federal launch range for the provision of any flight safety analysis for a licensed launch, and the FAA has assessed the range and found that the range's analysis methods satisfy the requirements of this subpart, the FAA will treat the federal launch range's analysis as that of the launch operator. For any such analysis, the launch operator need not provide the FAA any further demonstration of compliance. The FAA hopes that this clarifies that licensed launch operators may continue their existing arrangements with the federal launch ranges, and that the primary interface for a launch operator launching from a federal launch range remains the range.

22

See, e.g.,

Boeing Costs at 1, 20, 30 (first comment), 38 (first comment); Lockheed Martin Estimate 8 (attributing costs to requirement that launch operator conduct flight safety analyses now provided by the range); Orbital Cost Impact Assessment at 6 (Items 2 and 10: attributing costs to dual safety approval submittals and shif to FAA oversight).

3. Operational or Licensing Changes

Commenting launch operators raised concerns grounded in the notion that the October 2000 NPRM would result in large changes for licensed launch operators operating at federal launch ranges. Specifically, they feared that the requirements for obtaining and maintaining a license would change. JC Vol. I at 3. The FAA requests that in light of the following discussion, the launch operators revisit whether they should ascribe costs to these perceived changes.

On the basis of information provided by the comments, it appears to the FAA that some commenters assigned costs to what they saw as proposed changes in maintaining license compliance if they launched from a federal launch range.

23

Many of these purported costs arise out of the belief that the proposed requirements would subject a launch operator at a range to dual administrative requirements. In the NPRM, however, the FAA proposed that the administrative requirements for submitting material to the FAA contained in part 417 applied in total to all licensed launches from a non-federal launch site.

NPRM,

65 FR 63977 (proposed 14 CFR 417.1). Accordingly, unless a range changed its processes, the FAA does not anticipate that this rulemaking would require a launch operator launching from a federal range to demonstrate satisfaction of a part 417 requirement twice. Other costs in this category of concern appear to arise out of the launch operators' fear that the federal ranges will not obtain a satisfactory baseline assessment from the FAA for one requirement or another. Given that the FAA proposes these requirements in coordination with the Air Force through the Common Standards Working Group, the FAA has every reason to expect that the federal ranges will continue to satisfy the requirements.

23

Boeing Costs at 1 (second comment), 5 (all comments), 7 (all comments), 8 (all comments), 9 (second, third and fourth comments), 10 (first, second and fourth comments), 11 (first and fourth comments), 12 (second comment), 13 (first, second, third and fourth comment), 14 (second comment), 15, 16, 17 (first second and third comments), 18, 19 (first, second and third comments), 21, 22, 23 (first comment), 26 (second and third comments) 27 (third comment), 28 (first and third comment), 30 (second comment), 31 (first and second comment) and 38 (first comment); Lockheed Cost Estimates 3, 4, 9, 10, 11, 12, 13, 14, 20, 23, 24, and 25(b).

Similarly, commenters assigned costs to a perceived change in the requirements for obtaining a license to launch from a federal launch range. Commenting launch operators, apparently referring to proposed 14 CFR part 415, subpart F, contended that the new requirements for obtaining a license would be unduly burdensome and unwieldy. JC Vol. I at 10-11. They believe they will be required to

demonstrate compliance with two sets of requirements when launching from a federal range. JC Vol. I at 3. The FAA can, however, reassure launch operators who launch from federal launch ranges that proposed subpart F would not apply to them. Existing part 415, subpart C (Safety Review and Approval for Launch from a Federal Launch Range), which governs safety reviews for launch license applications from a federal launch range, will continue to apply. Proposed subpart F, (Safety Review and Approval for Launch of an Expendable Launch Vehicle from a Non-Federal Launch Site), applies to license applications for launch from outside of a federal launch range.

See

NPRM, 65 FR 63944, 63965 (proposed section 415.101 and accompanying discussion). Indeed, as stated in the NPRM, not only would proposed subpart F not apply to a license governing a launch from a federal launch range, but “the proposed regulations for obtaining a license would not * * * apply to any launch from a non-federal launch site where a federal launch range performs the safety functions.”

Id.

at 63922.

In the event that the Joint Commenters meant to warn that proposed subpart F would be unduly burdensome for obtaining a license for launch from a non-federal launch site, the FAA notes that, for such launches, it must require the same level of safety at non-federal launch sites as the ranges have achieved in the operation of their federal launch sites. Accordingly, information demonstrating that the current standards, as proposed in part 417, are satisfied is necessary.

4. Flexibility and Performance and Design Requirements

Commenters claimed costs on account of a perceived loss of flexibility.

24

The Joint Commenters stated that the October 2000 NPRM contained additional detailed design and testing requirements that will increase operating costs for all launch programs. Promulgating new requirements is not the FAA's intent, and should not be the effect of the FAA's final rule. Instead, the FAA's provision of a route for a launch operator to demonstrate an equivalent level of safety for a proposed alternative, willingness to grandfather and coordination on a waiver process should demonstrate that the FAA will be flexible.

24

Boeing Costs at 25 (third comment), 26 (fourth comment), 29 (third and fourth comments), 34 (first comment), 35 (fourth and seventh comments); Lockheed Cost Estimate 21; Oribital Cost Impact Assessment at 6 (items 1, 2a, 5, 6, 11 and 12); Sea Launch Costs at 2 (first and second comments), 5 (comments labeled 1 and m), 8 (first and second comments), 16 (lightning), 22 (alternate flight safety system), 26-35, 38, 40-42.

The commenters believe that the regulatory language used in the NPRM would reduce flexibility in implementing the requirements and that the FAA has changed standards that are currently goals and presented them as hard requirements. The FAA recognized early in the development of the NPRM that it was not always possible to adopt the range safety standards as written in current federal range safety documents because regulations must contain only that which is actually required.

EWR 127-1

contains both guidance and requirements. Recommended FAA approaches may appear in guidance documents, such as FAA advisory circulars. Alternatives may be approved through the licensing process.

When faced with a current standard that was in the form of a goal or preferred approach, the FAA, in coordination with federal range personnel, often had to either rewrite the standard as a performance requirement that described the intent of the original goal or omit it from the NPRM if it was determined to be unnecessary. For example, the federal launch ranges have a reliability goal of a minimum of 0.999 at the 95% confidence level for the flight termination system onboard a launch vehicle. Such a goal does not directly translate into a regulatory requirement for which compliance must be demonstrated. A 0.999 reliability at a 95% confidence level can be demonstrated only through a large number of launches or tests of the complete system while exposed to flight environments. The FAA worked with the federal ranges to understand the intent of the goal and how it has actually been implemented. As a result, the FAA's proposed regulations would require each flight termination system and command control system to have a reliability

design

of 0.999 at a confidence level of 95% to be demonstrated through an analysis of the design. The FAA is not proposing that this reliability be demonstrated through testing because it is not always practical to require the thousands of system level tests necessary to demonstrate compliance with the confidence level. Instead, the FAA is proposing an approach that has been developed in close coordination with the federal launch ranges, and that incorporates performance oriented design requirements for components coupled with comprehensive qualification and acceptance testing of components and preflight confidence tests of the entire system. The design and test requirements together with the required reliability analysis should ensure the reliability of the flight termination system.

In their discussion on the highly detailed requirements of the NPRM, the Joint Commenters referenced the FAA's licensing of Sea Launch and stated their belief that if Sea Launch had sought FAA approval under a regulatory regime as set forth in the NPRM, the process would have been far slower and more expensive for the launch operator. JC Vol. I at 7. The FAA disagrees. In licensing Sea Launch, the FAA used the current range safety requirements as the basis for evaluating the safety of the proposed launch vehicle and operations: the same requirements used as the basis for the October 2000 NPRM. It was during the evaluation of Sea Launch that the FAA developed various approaches for allowing flexibility in implementing specific requirements, including demonstrating an equivalent level of safety. These requirements and provisions for flexibility were refined and included in the NPRM. The FAA's conclusion was that Sea Launch could satisfy the requirements in the NPRM with no greater effort than was expended during its initial licensing. In effect, Sea Launch was held to the FAA's current requirements. Published requirements, however, with an appropriate level of detail should provide for a consistent, open and fair licensing process for all launch operators.

5. Neighboring Launch Operators

The FAA has learned that each Air Force launch range treats a portion of the public differently. For a launch conducted by a licensed launch operator, the FAA considers other launch operators at a launch site members of “the public.” Historically the Eastern Range and the Western Range did not consider anyone who operated at the range to be a member of the public. For approximately the past five years, however, the Eastern Range has been applying the FAA definition of the public when calculating the public risk associated with a licensed launch. At the Western Range other launch operators are not counted to ascertain their contribution to the collective risk to the general public. Some few personnel of other launch operators, at the request of those launch operators, are subjected to a higher level of risk than the rest of the public, which may include allowing them inside impact limit lines or hazard areas during the flight of a launch vehicle.

For the FAA, this approach has both safety and financial responsibility

implications. A launch operator may face issues surrounding launch availability and possible increases in insurance premiums. Although the FAA currently proposes no changes from its current practice, the FAA wishes to bring this issue to the attention of the public to obtain comments regarding the impact of the current approach and possible alternatives. The FAA notes that it is willing to entertain alternatives and implementation proposals. The issue is discussed in greater depth below.

In addition to placing the general public at risk, a launch operator's activities may place its neighbors at risk. Different launch operators are each others' neighbors at a single launch site. When, for example, launch operator A launches from one launch pad, adjacent launch operator B may be located within the impact limit lines or a hazard area created by launch operator A's launch. Nonetheless, for reasons of safety, security, or mission assurance, launch operator B may wish to keep some of its personnel working at the second launch point, even during the hazardous activities, including the flight of launch operator A's launch vehicle. Launch operator B's pressure vessels may require tending. Launch operator B may need to maintain the security of the site. Launch operator B may be interested in meeting a tight schedule. Typically, because the location exposes people to greater risk, the range will require the neighboring launch operator to train, shelter and otherwise attempt to protect its people from the increased risks.

The launch operators in this example are engaged in activities in support of separate launches and do not contract with each other for the launch that is about to take place. For these reasons, the FAA treats them as “the public” with respect to each other.

25

In existing 14 CFR part 420, which governs licensing and safety requirements for the operation of a launch site, the FAA defines the “public” as “people and property that are not involved in supporting a licensed launch, and includes those people and property that may be located within the boundary of a launch site, * * *, and any other launch operator and its personnel.” 14 CFR 420.5. In the October 2000 NPRM at § 417.3, the FAA proposed a similar definition for “public safety” as the safety of other launch operators and their personnel.

25

Although the FAA does not regulate or oversee the safety of the workers of a licensee, the workers of a neighboring launch operator are members of the public and the FAA has always intended that they be protected as such.

Likewise, for determining financial responsibility under existing 14 CFR part 440, the FAA treats other launch operators and their personnel as third parties. A licensed launch operator does not sign cross waivers with neighboring launch operators,

see generally

14 CFR 440.17, and the personnel of neighboring launch operators are treated as third parties in the maximum probable loss analysis that determines the amount of financial responsibility a licensee must shoulder. 14 CFR 440.3(15). The FAA, when calculating the maximum probable loss that may occur to members of the public, requires that a licensee demonstrate financial responsibility for those members of the public who have a chance of being harmed on the order of 1 × 10

−7

or more.

See

14 CFR 440.3(11)(i). This means that if any personnel of launch operator B are within the contours of an area where there is chance of an individual being harmed of 1 × 10

−7

or more, the FAA will assess the contribution of those individuals to the final financial responsibility determination.

The 30th Space Wing takes a different approach. At the Western Range, the 30th Space Wing relies on the definitions in EWR 127-1 to treat certain identified personnel of neighboring launch operators as not being members of the public, or, in the parlance of EWR 127-1, as “Wing-essential.” EWR 127-1 defines “mission-essential” and “non-essential” personnel, and, by implication, Wing-essential personnel. For the first two categories, different levels of risk, protection and exposure are available. In the portion relevant to this discussion, EWR 127-1 defines mission-essential personnel as “those persons necessary to successfully and safely complete a hazardous or launch operation and whose absence would jeopardize the completion of the operation.” EWR 127-1 at 1-vii (Dec. 31, 1999). This category includes, among others, “persons specifically authorized by the Wing Commander to perform scheduled activities.”

Id.

The ranges have a different mission than that of the FAA. Being military installations, they include within their mission not only the successful launch of a given launch vehicle, but the continued operations of other vehicles and programs deemed essential to the mission of the wing by the Wing Commander. These activities include, for example, support of commercial launches, launch of national need payloads, strategic weapons testing, warfighter support, payload processing and other activities that promote the function of the range as a whole.

“Non-essential” personnel, on the other hand, are persons who are not otherwise mission or Wing-essential, and include the general public, visitors, members of the media, and “any persons who can be excluded from Safety Clearance Zones with no effect on the operation or parallel operations.”

EWR 127-1

at 1-viii. EWR 127-1 does not contain a definition for Wing-essential, but the 30th Space Wing interprets the mention of Wing-essential personnel in the two definitions to permit a category of persons to be treated as mission-essential for purposes of calculating risk and requiring sheltering. This category may include personnel of neighboring launch operators who are present to perform safety, security or other tasks necessary to continue that second launch operator's operations at the launch site, but does not include anyone performing routine administrative, maintenance, or janitorial functions. Under the interpretation of the 30th Space Wing, when an employee of launch operator B is present within the impact limit lines or, albeit very infrequently, a hazard area for launch operator A's launch, that employee must be sheltered, and is included in a higher risk threshold.

See EWR 127-1

at 1-12, 1.4d (Oct. 31, 1997). In contrast to the permissible E

c

of 30 × 10

−6

for the general public, the workers of the launch operator conducting the launch may be exposed to a higher risk of 300 × 10

−6

. Based on information from the 30th Space Wing, there may be, for a given licensed launch at the Western Range, over 100 people who are members of the public under the FAA's definitions, but who the FAA has not identified as such in its financial responsibility determinations due to the differences in definitions.

At the Eastern Range, the 45th Space Wing treats other launch operators as members of the public when calculating public risk due to a licensed launch. The Eastern Range may permit the personnel of neighboring launch operators to remain within the impact limit lines or the flight hazard area in approved hardened structures for a launch. The Eastern Range, when assessing collective risk to the public, counts the neighboring launch operator's personnel as members of the public. In other words, the presence of too many of such people may produce an E

c

in excess of 30 × 10

−6

. Accordingly, their numbers are limited for that reason.

The FAA and the Air Force now confront the question of whether to

continue the FAA and 45th Space Wing approach, or to adopt a variation on the approach of the 30th Space Wing. The Air Force intends to standardize these approaches at its ranges. The former is current practice for the bulk of licensed launches, but the latter was the practice at both ranges prior to the adoption by the 45th Space Wing of the FAA's definition of “the public,” and may provide greater operational flexibility, both for the launch operator conducting the launch and for the neighboring launch operator who wants to continue operations during the hazardous activities of the first launch operator. Greater operational flexibility may come with a price, however. Although the FAA could, through rulemaking, permit some members of the public to be exposed to greater risk than others, especially if they are protected,

26

the FAA must point out that the launch operator conducting the launch would have to demonstrate sufficient financial responsibility under part 440 to protect financially against loss to those members of the public. In other words, where a neighboring launch operator's personnel are exposed to risk sufficient to trigger a requirement for financial responsibility coverage, the insurance premiums of the launch operator who is about to launch may increase. Conversely, that first launch operator may find the increased flexibility in its own operations worth the potential increase in premiums.

27

26

The Eastern and Western Ranges advice that risk assessments account for any sheltering of the neighboring launch operator's personnel.

27

The FAA notes that it has not been aware, in the course of conducting its maximum probable loss analyses in accordance with 14 CFR part 440, that some of the personnel identified as mission essential at the ranges were, in fact, what the FAA considers members of the public, and should therefore have been considered at the 10

−7

threshold instead of the 10

−5

threshold. Because of this possible confusion, the FAA may not have addressed third parties who should have been considered in financial responsibility determinations for licensed launches from both the Eastern and the Western Range. If the FAA determines that their presence requires an increase in the financial responsibility for which a licensee must prepare, that increase would be mandated by existing requirements and would be a decision that was independent of this rulemaking.

The FAA and the Common Standards Working Group intend to explore this issue further so as to ensure a common approach. Before the FAA conducts any rulemaking on this issue, the FAA requests comments on the public's experience with the impacts of the two approaches that have been in practice to date. Are there cost impacts associated with either approach? Do the benefits of one outweigh the advantages of the other? Do concerns for worker safety of the neighboring launch operator suggest that no one other than the participants in that launch be allowed in the areas of greater risk? In other words, even with the benefits of increased operational flexibility, would launch operator B not want its employees exposed to greater risk than the general public? Additionally, implementation raises issues. Were the FAA and the ranges to adopt the Western Range's approach, the ranges could oversee and coordinate the presence of different launch operators and their personnel. At a launch site operated by a licensed launch site operator, the FAA already requires that a launch site operator schedule its customers. 14 CFR 420.55. However, the launch site operator does not assess risk under current requirements. The FAA requests comments on the advisability of imposing such a requirement on a launch site operator.

C. FAA and Air Force Process for Relief From Common Launch Safety Requirements

Launch operators commenting on the October 2000 NPRM expressed concern for problems they believe will arise if both the Air Force and the FAA oversee the safety of launches from Air Force ranges. JC Vol. I at 1; Lockheed at 3. In response, the Air Force and the FAA have established a permanent safety working group to develop common launch safety standards and implementation processes. This working group has drafted a process for coordinated review of requests for relief from launch safety requirements as well as tailoring of requirements for future programs. This process is outlined in a draft

Memorandum of Understanding (MOU) between Air Force Space Command and the FAA Office of the Associate Administrator for Commercial Space Transportation for Resolving Requests for Relief from Common Launch Safety Requirements.

The MOU will provide for Air Force and FAA coordination on issues that may arise for a specific launch. For day-to-day operations at an Air Force range, the Air Force will remain the primary point of contact for the launch operators. For a licensed launch, when a request for relief from a common requirement is made to either agency, each agency will ensure notification of the other, and the two agencies will coordinate activities with the launch operator to ensure an efficient and timely resolution.

The draft coordination process contains provisions to address issues “prior to day of launch,” when there is time to coordinate and formally document the resolution of an issue before launch, and “day-of-launch” (flight minus 24 hours, often called “real-time”) coordination on issues that arise, albeit infrequently, during a launch countdown prior to flight. The Air Force and the FAA will also jointly participate with launch operators in tailoring of common launch safety requirements during the development of launch vehicle systems to be used for licensed launches from Air Force ranges. The coordination process between the Air Force and the FAA will provide for sharing of data to avoid duplication of effort. This coordination will allow for joint resolution of issues regarding common launch safety requirements while ensuring that both agencies' requirements and concerns are addressed without placing undue burden on launch operators. A copy of the draft Air Force/FAA MOU is available on AST's Web site at

http://ast.faa.gov.

The agencies will continue to administer their own waiver processes. In conjunction with the Air Force/FAA Common Standards Working Group, the two agencies addressed whether the FAA could baseline the Air Force's waiver process. The group determined that the FAA, once its requirements became final, could not baseline the Air Force's waiver process. The FAA cannot delegate its responsibility for safety. The FAA has the authority to waive its own requirements. 49 U.S.C. 70105(c)(3). As the January 2001 Safety MOA between the FAA and the Air Force recognized, neither agency may waive the requirements of the other. Although Chapter 701 allows another agency to assist the FAA, and the FAA plans to continue to accept the assistance of the Air For

This text is long and has been trimmed here. Open the source document for the complete record.