Risk-Based Alternative to Pressure Testing Older Hazardous Liquid and Carbon Dioxide Pipelines Rule

DEPARTMENT OF TRANSPORTATION

Research and Special Programs Administration

49 CFR Part 195

[Docket No. PS-144; Amdt. 195-65]

RIN 2137-AC 78

Risk-Based Alternative to Pressure Testing Older Hazardous Liquid

and Carbon Dioxide Pipelines Rule

AGENCY: Research and Special Programs Administration (RSPA), DOT.

ACTION: Final rule.

-----------------------------------------------------------------------

SUMMARY: This final rule allows operators of older hazardous liquid and

carbon dioxide pipelines to elect a risk-based alternative in lieu of

the existing rule. The existing rule requires the hydrostatic pressure

testing of certain older pipelines. The risk-based alternative would

allow operators to elect an approach to evaluating the integrity of

these lines that takes into account individual risk factors. This would

allow operators to focus resources on higher risk pipelines and effect

a greater reduction in the overall risk from pipeline accidents.

DATE: This final rule takes effect November 4, 1998.

FOR FURTHER INFORMATION CONTACT: Mike Israni, (202) 366-4571, or e-

mail: [email protected], regarding the subject matter of this

final rule, or Dockets Unit (202) 366-4046, for copies of this final

rule document or other material in the docket.

SUPPLEMENTARY INFORMATION:

Background

On June 7, 1994, RSPA published a final rule, ``Pressure Testing

Older Hazardous Liquid and Carbon Dioxide Pipelines,'' (Amdt. 195-51;

59 FR 29379) to ensure that certain older pipelines have an adequate

safety margin between their maximum operating pressure and test

pressure. This safety margin is to be provided by pressure testing

according to part 195 standards or operation at 80 percent or less of a

qualified prior test or operating pressure. The pipelines covered by

the rule are steel interstate pipelines constructed before January 8,

1971, steel interstate offshore gathering lines constructed before

August 1, 1977, or steel intrastate pipelines constructed before

October 21, 1985, that transport hazardous liquids subject to part 195.

Also covered are steel carbon dioxide pipelines constructed before July

12, 1991, subject to part 195.

On June 23, 1995, the American Petroleum Institute (API) filed a

petition on behalf of many liquid pipeline operators that proposed a

risk-based alternative to the required pressure testing rule. API

indicated that its proposal would allow operators to focus resources on

higher risk pipelines and to effect a greater reduction in the overall

risk from pipeline accidents.

In order to determine whether the API proposal had merit, RSPA held

a public meeting on March 25, 1996. On May 8 and November 7, 1996, and

on May 17, 1997, RSPA briefed the Technical Hazardous Liquid Pipeline

Safety

[[Page 59476]]

Standards Committee (THLPSSC) on the API proposal and steps taken by

RSPA to develop a proposed rule. As discussed in more detail below,

RSPA finds considerable merit in a risk-based approach to pressure

testing of older hazardous liquid pipelines. It provides accelerated

testing of electric resistance welded (ERW) pipe, incorporates the use

of new technology, and provides for continuing internal inspection of

older pipelines through a pigging program. RSPA has been working

actively with the pipeline industry to develop a risk management

framework for pipeline regulations. The API proposal is consistent with

the risk assessment and management approach to safety. The API proposal

provides an opportunity to pilot a risk-based approach in a rulemaking

forum. Accordingly, this final rule requires a risk-based alternative

to the pressure testing rule that has been modeled after the API

proposal.

RSPA has extended time for compliance with the pressure testing

rule in order to allow completion of this final rule on a risk-based

alternative. The deadline for complying with Sec. 195.302(c)(1) is

extended to December 7, 1998. The deadline for complying with

Sec. 195.302(c)(2)(i) is extended to December 7, 2000. The deadline for

complying with Sec. 195.302(c)(2)(ii) is extended to December 7, 2003.

(62 FR 54591; October 21, 1997).

Major Features of Risk-Based Alternative

The risk-based alternative to the rule requiring the pressure

testing of older pipelines has six main features:

1. Highest Priority Is Given to the Highest Risk Facilities; Lowest

Risk Facilities Are Excepted From Additional Measures

Pre-1970 electric resistance welded (ERW) and lapwelded pipelines

susceptible to longitudinal seam failures exhibit the highest potential

risk because of their combination of probability of failure and

potential for larger volume releases as evidenced by historical

records. Pressure testing is the only available technology for

verifying the integrity of pre-1970 ERW and lapwelded pipelines,

because it can detect the type of seam failures endemic to some ERW and

all lapwelded pipe. This risk-based alternative requires accelerated

testing of pre-1970 ERW and lapwelded pipe susceptible to longitudinal

seam failure in certain locations (risk classification C and B) where

people and environment might be significantly affected. However, in

locations (risk classification A) where consequences to the public or

environment are less significant, the risk-based alternative allows

delayed testing for pre-1970 ERW and lapwelded pipe susceptible to

longitudinal failure and allows the operator to determine the need for

pressure testing of other types of pipe.

2. Consequence Factors Such as Location (Population and Environment),

Product Type, and Release Potential Are Taken Into Consideration When

Setting Testing Priorities

This risk-based alternative takes into account the most significant

variables that may impact the severity of a release, i.e., location

with respect to populated and environmentally sensitive \1\ areas, the

nature of the product transported, and the potential volume of product

release. Historically, a very small percentage of releases adversely

impacted public safety and environment. By taking these potential

consequences into consideration in the timing of tests, an operator's

resources will be more effectively applied to reduce risks.

---------------------------------------------------------------------------

\1\ ``Environmentally sensitive areas'' is not currently

defined, but operators are encouraged to use their best judgment in

applying this factor. This factor may be defined in future

rulemaking.

---------------------------------------------------------------------------

3. Best Available Technology Is Applied To Verify Pipeline Integrity

The risk-based alternative encourages the use of the most effective

means to ensure pipeline integrity. This rule utilizes the strength of

two primary technologies--pressure testing and magnetic flux leakage/

ultrasonic internal inspection devices. Each technology provides

testing advantages in particular circumstances. This rule allows the

operator to evaluate the pipeline risk considerations and to choose the

most appropriate technology.

4. Timing of Tests Is Based on Risk

Considering the probability and consequence factors, the risk-based

rule increases the priority of a limited amount of pre-1970 ERW and all

lapwelded pipelines and maintains the three-year timing for risk

classification B and C lines which represent the highest risk to people

and environment. Pipelines with lower risks (risk classification A) are

allowed a longer testing schedule or are eliminated (non-high risk pre-

1970 ERW pipelines) from a mandatory testing requirement. Nothing in

this rule precludes an operator from accelerating these schedules based

on their pipeline operating and maintenance history.

5. Reduces Test Water Requirements

This rule would allow operators options that require less test

water and generate less water requiring treatment.

6. Provides an Opportunity To Reduce Operating Costs and Maintain the

Necessary Margins of Safety by Applying the Risk-Based Concept

Acceptance and implementation of this rule provides an opportunity

to pilot a risk-based approach to regulation. OPS anticipates increased

use of risk-based approaches in future rulemakings.

Proposed Rule

RSPA published an NPRM (63 FR 5918; February 5, 1998), proposing to

add a new section to Part 195 entitled ``Risk-based alternative to

pressure testing.'' NPRM also proposed that existing Sec. 195.303

``Test pressure'', and Sec. 195.304 ``Testing of components'' would be

renumbered as Sec. 195.304 and Sec. 195.305 respectively. The comment

period closed April 6, 1998. Commenters included an industry

association, two pipeline operating companies and a safety consultant.

Advisory Committee Review

On May 6, 1998, RSPA submitted the proposed rule and regulatory

evaluation to the Technical Hazardous Liquid Pipeline Safety Standards

Committee (THLPSSC). Each proposed hazardous liquid pipeline safety

standard must be submitted to the THLPSSC for Committee's view as to

its technical feasibility, reasonableness, cost-effectiveness, and

practicability. At the meeting, the THLPSSC declined to approve the

proposed rule and unanimously requested that ``environmentally

sensitive areas'' be included within the consequence factors for

setting testing priorities. Some members argued that not including an

environmental factor at this time would result in many segments of

pipeline remaining untested for many more years. The Committee asked

that the proposed rule be resubmitted for consideration through a mail

ballot. On May 12, 1998, RSPA sent letter ballots to THLPSSC members to

vote on revised language to be included in the final rule. RSPA

received 10 of 12 ballots. All 10 members voted to approve the proposed

rule provided the revised language was included. The THLPSSC also

recommended discussion in the preamble to the final rule of the need to

include consideration of environmentally sensitive areas even before a

clear definition of the term is developed.

[[Page 59477]]

RSPA did not include an environmental factor in the proposed rule

because of the lack of agreement on a definition. Following public

briefings on the progress of the rulemaking at the THLPSSC meetings in

November 1996 and May 1997, API objected to inclusion of an

environmental factor as premature in light of the ongoing rulemaking to

define unusually sensitive areas (USAs). At that time, RSPA intended to

include an interim definition that could later be replaced, if

appropriate, by the definition of USAs.

Although we do not necessarily agree that a definition of USAs

should be the sole basis for inclusion of an environmental factor for a

risk-based alternative to pressure testing, we recognized in the

proposed rule the difficulties of defining an environmental factor

before the USA definition is formulated. The difficulty in articulating

a factor was made very apparent by THLPSSC members at the May 1997

meeting. One member argued that the environmental factor under

consideration for the proposed rule was inadequate; two other members

challenged that argument. Discussions with the members and API

following that meeting indicated little chance of agreement on a

definition prior to definition of USAs. Based on the discussion at the

THLPSSC on May 6, 1998, it appears that there is broad agreement that

environmentally sensitive areas will be considered by the industry even

in the absence of a definition. Accordingly, we are following the

advice of the THLPSSC and including environmentally sensitive areas

within the consequence factors in this final rule. We recognize that we

may need to revisit this issue once we have defined ``unusually

sensitive areas.''

The Final Rule

The new Sec. 195.303 ``Risk-based alternative to pressure testing''

would allow an operator of older hazardous liquid and carbon dioxide

pipeline to elect an approach to evaluating the integrity of lines that

takes into account individual risk factors. This alternative

establishes test priorities based on the inherent risk of a given

pipeline segment. Each pipeline is assigned a risk classification based

on several indicators. In assigning a risk classification to a given

pipeline segment, the first step is to determine whether or not the

segment contains pre-1970 ERW and lap-weld pipe susceptible to

longitudinal seam failures. Certain pre-1970 ERW and lap-weld pipeline

segments are susceptible to longitudinal seam failures. An operator

must consider the seam-related leak history of the pipe and pipe

manufacturing information as available, which may include the pipe

steel's mechanical properties, including fracture toughness; the

manufacturing process and controls related to seam properties,

including whether the ERW process was high-frequency or low-frequency,

whether the weld seam was heat treated, whether the seam was inspected,

the test pressure and duration during mill hydrotest; the quality

control of the steel-making process; and other factors pertinent to

seam properties and quality.

The next step is to determine the pipeline segment's proximity to

populated and environmentally sensitive areas (Location).

``Environmentally sensitive areas'' is not currently defined. However,

we expect operators to use their best judgment in applying this factor.

Some good examples of areas which would be environmentally sensitive

are waters used for drinking and fishing. This environmental factor may

be defined in a future rulemaking.

The risk classification of a segment is also adjusted based on the

pipeline failure history, the product transported, and the volume

potentially releasable in a failure. Additional guidance for use of the

alternative is provided in a new appendix B.

The pipeline failure history, denoted in the final rule as

``Probability of Failure Indicator,'' is an important factor. The

history of past failures (types of failures, number of failures, sizes

of releases, etc.) plays an important role in determining the chances

of future occurrences for a particular pipeline system. Therefore, it

has been included as risk factor in the matrix for determining the risk

classification. In the final rule the probability of failure indicator

is considered ``high risk'' if the pipeline segment has experienced

more than three failures in last 10 years due to time-dependent defects

(due to corrosion, gouges, or problems developed during manufacture,

construction or operation, etc.). Pipeline operators should make an

appropriate investigation of spills to determine whether they are due

to time-dependent defects. An operator's determination should be based

on sound engineering judgment and be documented. In addition, the final

rule provides compliance dates and recordkeeping requirements for those

operators who elect the risk-based alternative to pressure testing of

older hazardous liquid and carbon dioxide pipelines.

RSPA believes this rule will provide the pipeline industry with the

flexibility to elect alternative technology for evaluating pipeline

integrity without sacrificing safety.

Discussion of Comments

RSPA received four comments in response to the NPRM. Commenters

included one industry association (API), two pipeline operating

companies, and a safety consultant. Three commenters including API

expressed strong support, but one commenter (a safety consultant)

opposed issuing this risk-based rule.

Performance measures--In the proposed rule, RSPA sought comment and

information on how to measure the performance of this risk-based

alternative to determine effectiveness, particularly in comparison with

the pressure test rule. RSPA received no comment. RSPA plans to examine

the future performance of those pipeline segments that are pressure

tested and compare it to the future performance of pipeline segments

that are internally inspected or that are not tested at all.

Failure history--In the proposed rule, RSPA sought comment on

excluding insignificant failures from the failure history risk factor.

RSPA also sought comment on whether the failure should be quantified or

if only a reportable incident should be considered.

One operator commented that only Department Of Transportation (DOT)

reportable incidents be included. API commented that spills, regardless

of whether reportable or not, should be included in the risk-based

alternative engineering evaluation process by the operator making its

own engineering judgment. The judgment should be documented and

applied, when appropriate, to the failure history risk factor. API

believes that proper documentation removes subjective judgments during

agency audits/evaluations of the use of the risk-based alternative.

One commenter asked whether third party damage resulting in the

immediate release of product would be considered a time-dependent

defect in Table 6.

RSPA agrees that proper documentation would clarify the validity of

decisions about whether spills are related to time-dependent defects or

are truly insignificant during agency evaluation of the use of the

risk-based alternative. This also eliminates need for failures to be

quantified. Third party damage resulting in the immediate release of

product does not constitute a time-dependent defect. Time-dependent

defects are defects that result in spills due to corrosion, gouges, or

problems developed during manufacture, construction or operation, etc.

This is already covered in subnote 2 in Table 6

[[Page 59478]]

of Appendix B. Therefore, no changes have been made to Table 6.

Opposition to issuing the risk-based rule--One commenter (a safety

consultant) opposed issuing this rule. Commenter argued that this rule

might have been more meritorious had it been proposed after the results

were in on the risk management demonstration projects. This commenter

said that the notice published in the Federal Register on November 15,

1996 (61 FR 58605) states that the demonstration projects will test

whether allowing operators the flexibility to allocate safety resources

through risk management is an effective way to improve safety,

environmental protection, and reliability. They will also provide data

on how to administer risk management as a permanent feature of the

Federal pipeline safety program if risk management proves to be viable

regulation alternative. Therefore, this commenter said this rulemaking

should be delayed until the completion of the risk management

demonstration projects. This commenter also contended that the purpose

of the API petition requesting the risk-based alternative was to

reduce, or delay, the economic burden on pipeline companies as a result

of the requirements of the final rule for pressure testing published by

RSPA on June 7, 1994, (59 FR 29379).

RSPA disagrees that this rule should be delayed until completion of

the risk management demonstration projects. The Accountable Pipeline

Safety and Partnership Act of 1996 (Pub. L. 104-304, Oct. 12, 1996)

that establishes the Risk Management Demonstration Program contemplates

a limited number of projects. RSPA will approve no more than ten (10).

Currently, none of projects being considered addresses the pressure

testing of older pipelines that are impacted by the June 1994 pressure

test rule. The Demonstration Program is looking at whole set of

activities rather than focusing on an individual regulation. Also,

delay until completion of the projects would unreasonably delay

addressing issues of older hazardous liquid pipelines. These pipelines

include high risk ERW pipelines.

The risk-based approach to older pipelines provides an opportunity

to pilot a risk-based approach in a rulemaking forum as opposed to a

demonstration project forum. RSPA believes this rule will provide the

pipeline industry with the flexibility to elect alternative technology

for evaluating pipeline integrity without sacrificing safety.

Proposed Sec. 195.303(b)(4)(ii)--API suggested that this paragraph

be revised to clarify that up to three time-dependent failures in 10

years would be low-risk. The proposed rule inadvertently limited the

low risk assignation to two failures. This is inconsistent with the

proposed Table 6. We agree and have revised this paragraph to be

consistent with Table 6.

Proposed Sec. 195.303(c): API said that the last sentence in the

text of Sec. 195.303(c) should be clarified so that operators

understand that for those segments that fall under Risk Classification

A ``no additional measures'' refers to no additional measures under

this subpart (i.e. subpart E--Pressure Testing). API said that the last

sentence as proposed appears to be broader. We have revised this

section for clarity as recommended by the API.

Proposed Sec. 195.303(g): API said that the text of Sec. 195.303(g)

should be clarified so that operators understand that pressure testing

under the risk-based alternative, like the existing final rule, would

be a one-time test. The review of risk classifications should be

required only for those pipeline segments that have not yet been tested

under Sec. 195.303(a) or Sec. 195.303(c). We agree and have clarified

the wording.

Proposed Sec. 195.303(i): API said that requiring operators to give

a written notification and get approval from the Administrator before

discontinuing from this program, should be eliminated from this

rulemaking. Adding that this section is confusing, contradictory and

results in a different standard of care for the risk-based alternative

compared with the existing final rule. API said that operators should

have flexibility to elect test portions and change plans of their

system using the existing final rule and portions of their systems

under the risk based alternative. The intent of Sec. 195.303(i)

requirement is to avoid operators switching from one testing program to

another, causing delays in testing. Eliminating this requirement may

make it difficult to enforce the regulatory deadlines. Requirements in

this rule does not prevent an operator from choosing pressure testing

for some segments and risk-based alternative for the remaining segments

of a pipeline. Therefore, this section is retained.

Do previous in-line inspections on pipeline systems constitute

compliance? API and one commenter requested that RSPA should allow

previous in-line inspections and subsequent maintenance of a pipeline

documented by company records as in compliance with this rule. RSPA

will accept previous in-line inspections on pipeline conducted in the

five years prior to the effective date of this final rule provided that

anomalies found by previous smart pig runs have been repaired and

pipeline has been maintained. RSPA will not accept older in-line

inspections for the following reasons: (1) Technology keeps changing

rapidly and internal inspection devices have greatly improved in recent

years, (2) older internal inspection devices probably did not provide

adequate data, (3) new corrosion or other defects may have developed

since last in-line inspection.

Appendix B Table 1--API suggested that term ``pipeline system'' be

changed to ``pipeline segment'' in Footnote 1 to Table 1, for clarity

and agreement with the intent of the risk-based rule. We agree.

Additional Clarifying Guidance for both Operators and Inspectors--A

number of operators (via API) offered suggestions for ways of making

the rule more understandable, including rearranging the tables in the

appendix, making the tables more explicit or providing flow charts that

visually clarify the decision-making paths. RSPA realizes that a

flowchart or decision tree with a couple of examples could aid the

operators. However, the need to avoid further delay in addressing the

issues of older hazardous liquid pipelines makes it impossible for RSPA

to prepare such additional aids to implementation at this stage.

Nothing precludes API with the help of its members from developing a

flowchart and perhaps a few examples on how to apply this risk-based

rule for its members.

V. Rulemaking Analyses

Executive Order 12866 and DOT Regulatory Policies and Procedures

This final rule is a significant regulatory action under Executive

Order 12866. Therefore, this rule was reviewed by the Office of

Management and Budget. In addition, this final rule is significant

under DOT's regulatory policies and procedures (44 FR 11034; February

26, 1979) because it is the first explicitly risk-based approach to

rulemaking final by the Office of Pipeline Safety. A copy of the

regulatory evaluation to this rule is also available in the docket

office for review.

This section summarizes the conclusions of the regulatory

evaluation. RSPA's pressure testing final rule was published on June 7,

1994 (59 FR 29379) along with a regulatory evaluation which found that

the rule had a positive net benefit to the public, i.e., the benefits

of the rule exceeded the cost (Present value costs of the earlier

proposal were estimated to be between $134-$179 million in 1997 dollars

while the present value benefits were

[[Page 59479]]

estimated as $230-$283 million). RSPA believes that the risk-based

alternative maintains the necessary margins of safety, therefore, the

benefits of this alternative should be similar to the benefits of the

earlier proposal. The present value costs for the risk-based

alternative are estimated to be between $88.4-$98.4 million for reasons

described below. The final rule allows the use of alternative

technology (smart pigs) for evaluating pipeline integrity. On average

smart pig testing is less expensive than pressure testing by $2,650/

mile. In some cases smart pig technology provides more information

about pipeline anomalies than pressure testing. The risk-based

alternative would reduce the total amount of test water, which should

lower the waste treatment costs and generate less hazardous waste. The

risk-based alternative would allow operators to forgo testing where

pipelines have low operating pressures, transport non-volatile product,

operate in rural and environmentally non-sensitive areas, and have good

records on pipeline failure history.

This risk-based approach is an ongoing process. RSPA believes that

the risk-based alternative maintains the necessary margins of safety

for the public and environment. Moreover, RSPA concludes that this

alternative has the potential for positive improvements for the

environment while reducing operating costs by allowing operators to

elect those test methods most appropriate to the circumstances of each

pipeline.

Regulatory Flexibility Act

The regulatory flexibility analysis of the earlier final rule

concluded that it would not have a significant impact on a substantial

number of small entities. RSPA believes that because this regulation

offers an alternative to operators that could reduce the less than

significant impact of the earlier regulation even further, this rule

does not have a significant impact on a substantial number of small

entities. Based on the facts available about the anticipated impact of

this rulemaking action, I certify pursuant to Section 605 of the

Regulatory Flexibility Act (5 U.S.C. 605) that the action will not have

a significant economic impact on a substantial number of small

entities.

RSPA, in the proposed rule, had requested comments from small

entities which might be impacted by this rule. We received no comments.

This supports our earlier conclusion that this rule will have no

significant impact on a substantial number of small entities.

Executive Order 12612

This rule will not have substantial direct effect on states, on the

relationship between the Federal Government and the states, or on the

distribution of power and responsibilities among the various levels of

government. Therefore, in accordance with E.O. 12612 (52 FR 41685;

October 30, 1987), RSPA has determined that this final rule does not

have sufficient federalism implications to warrant preparation of a

Federalism Assessment.

Executive Order 13084

This rule has been analyzed in accordance with the principles and

criteria contained in Executive Order 13084 (``Consultation and

Coordination with Indian Tribal Governments''). Because this rule would

not significantly or uniquely affect the communities of the Indian

tribal governments, the funding and consultation requirements of this

Executive Order do not apply.

Unfunded Mandates

This rule does not impose unfunded mandates under the Unfunded

Mandates Reform Act of 1995. It does not result in costs of $100

million or more to either State, local, or tribal governments, in the

aggregate, or to the private sector, and is the least burdensome

alternative that achieves the objective of the rule.

Paperwork Reduction Act

This rule does not substantially modify the paperwork burden on

pipeline operators. Under the current pressure testing regulations

operators are required to have testing plans, schedules, and records.

The risk-based alternative would require the same or equivalent plans,

schedules, and records for either pressure testing or internal

inspection. Therefore, there is no additional paperwork required.

Operators who choose the risk-based alternative will be required to

have records that the pipeline segment which is not being tested

qualifies for the risk-based alternative. According to conversations

between OPS and the pipeline industry some of this information is

already available in the form of drawings or plans that can be found

either in operators' Facility Response Plans required by the Oil

Pollution Act of 1990 (OPA 90) or in emergency response plans required

by RSPA.

Operators will be required to periodically review the pipelines

that qualify for the risk-based alternative to ensure that they still

qualify. OPS believes that operators can conduct this review as part of

their normal procedures.

Because of the above analysis, OPS does not believe that operators

will have any additional paperwork burden because of this alternative,

and therefore no separate paperwork submission is required.

National Environmental Policy Act

RSPA has analyzed this action for purposes of the National

Environmental Policy Act (42 U.S.C. 4321 et seq.) and has determined

that this action would not significantly affect the quality of the

human environment. An Environmental Assessment and a Finding of No

Significant Impact are in the docket.

List of Subjects in 49 CFR Part 195

Anhydrous ammonia, Carbon dioxide, Petroleum, Pipeline safety,

Reporting and recordkeeping requirements.

In consideration of the foregoing, RSPA amends part 195 of title 49

of the Code of Federal Regulations as follows:

PART 195--[AMENDED]

1. The authority citation for part 195 continues to read as

follows:

Authority: 49 U.S.C. 60102, 60104, 60108, and 60109; and 49 CFR

1.53.

2. Section 195.302 is amended by adding a new paragraph (b)(4) to

read as follows:

Sec. 195.302 General requirements.

* * * * *

(b) * * *

(4) Those portions of older hazardous liquid and carbon dioxide

pipelines for which an operator has elected the risk-based alternative

under Sec. 195.303 and which are not required to be tested based on the

risk-based criteria.

* * * * *

3. Section 195.302(a) is amended by removing cross-reference

``Sec. 195.304(b)'' and adding cross-reference ``Sec. 195.305(b)''.

4. In paragraph (c) of Sec. 195.302, the introductory text is

revised to read as follows:

Sec. 195.302 General requirements.

* * * * *

(c) Except for pipelines that transport HVL onshore, low-stress

pipelines, and pipelines covered under Sec. 195.303, the following

compliance deadlines apply to pipelines under paragraphs (b)(1) and

(b)(2)(i) of this section that have not been pressure tested under this

subpart:

* * * * *

[[Page 59480]]

Secs. 195.303 and 195.304 [Redesignated as Secs. 195.304 and 195.305]

5. Section 195.303 Test pressure. and Sec. 195.304 Testing of

components. are redesignated as Sec. 195.304 Test pressure. and

Sec. 195.305 Testing of components.

6. Part 195 is amended by adding a new Sec. 195.303 to read as

follows:

Sec. 195.303 Risk-based alternative to pressure testing older

hazardous liquid and carbon dioxide pipelines.

(a) An operator may elect to follow a program for testing a

pipeline on risk-based criteria as an alternative to the pressure

testing in Sec. 195.302(b)(1)(i)-(iii) and Sec. 195.302(b)(2)(i) of

this subpart. Appendix B provides guidance on how this program will

work. An operator electing such a program shall assign a risk

classification to each pipeline segment according to the indicators

described in paragraph (b) of this section as follows:

(1) Risk Classification A if the location indicator is ranked as

low or medium risk, the product and volume indicators are ranked as low

risk, and the probability of failure indicator is ranked as low risk;

(2) Risk Classification C if the location indicator is ranked as

high risk; or

(3) Risk Classification B.

(b) An operator shall evaluate each pipeline segment in the program

according to the following indicators of risk:

(1) The location indicator is--

(i) High risk if an area is non-rural or environmentally sensitive

\1\; or

(ii) Medium risk; or

(iii) Low risk if an area is not high or medium risk.

(2) The product indicator is 1

---------------------------------------------------------------------------

\1\ (See Appendix B, Table C).

---------------------------------------------------------------------------

(i) High risk if the product transported is highly toxic or is both

highly volatile and flammable;

(ii) Medium risk if the product transported is flammable with a

flashpoint of less than 100 deg. F, but not highly volatile; or

(iii) Low risk if the product transported is not high or medium

risk.

(3) The volume indicator is--

(i) High risk if the line is at least 18 inches in nominal

diameter;

(ii) Medium risk if the line is at least 10 inches, but less than

18 inches, in nominal diameter; or

(iii) Low risk if the line is not high or medium risk.

(4) The probability of failure indicator is--

(i) High risk if the segment has experienced more than three

failures in the last 10 years due to time-dependent defects (e.g.,

corrosion, gouges, or problems developed during manufacture,

construction or operation, etc.); or

(ii) Low risk if the segment has experienced three failures or less

in the last 10 years due to time-dependent defects.

(c) The program under paragraph (a) of this section shall provide

for pressure testing for a segment constructed of electric resistance-

welded (ERW) pipe and lapwelded pipe manufactured prior to 1970

susceptible to longitudinal seam failures as determined through

paragraph (d) of this section. The timing of such pressure test may be

determined based on risk classifications discussed under paragraph (b)

of this section. For other segments, the program may provide for use of

a magnetic flux leakage or ultrasonic internal inspection survey as an

alternative to pressure testing and, in the case of such segments in

Risk Classification A, may provide for no additional measures under

this subpart.

(d) All pre-1970 ERW pipe and lapwelded pipe is deemed susceptible

to longitudinal seam failures unless an engineering analysis shows

otherwise. In conducting an engineering analysis an operator must

consider the seam-related leak history of the pipe and pipe

manufacturing information as available, which may include the pipe

steel's mechanical properties, including fracture toughness; the

manufacturing process and controls related to seam properties,

including whether the ERW process was high-frequency or low-frequency,

whether the weld seam was heat treated, whether the seam was inspected,

the test pressure and duration during mill hydrotest; the quality

control of the steel-making process; and other factors pertinent to

seam properties and quality.

(e) Pressure testing done under this section must be conducted in

accordance with this subpart. Except for segments in Risk

Classification B which are not constructed with pre-1970 ERW pipe,

water must be the test medium.

(f) An operator electing to follow a program under paragraph (a)

must develop plans that include the method of testing and a schedule

for the testing by December 7, 1998. The compliance deadlines for

completion of testing are as shown in the table below:

Table.--Sec. 195.303--Test Deadlines

------------------------------------------------------------------------

Risk

Pipeline segment classification Test deadline

------------------------------------------------------------------------

Pre-1970 Pipe susceptible to C or B 12/7/2000.

longitudinal seam failures A 12/7/2002.

[defined in Sec. 195.303(c) &

(d)].

All Other Pipeline Segments..... C 12/7/2002.

B 12/7/2004.

A Additional testing

not required.

------------------------------------------------------------------------

(g) An operator must review the risk classifications for those

pipeline segments which have not yet been tested under paragraph (a) of

this section or otherwise inspected under paragraph (c) of this section

at intervals not to exceed 15 months. If the risk classification of an

untested or uninspected segment changes, an operator must take

appropriate action within two years, or establish the maximum operating

pressure under Sec. 195.406(a)(5).

(h) An operator must maintain records establishing compliance with

this section, including records verifying the risk classifications, the

plans and schedule for testing, the conduct of the testing, and the

review of the risk classifications.

(i) An operator may discontinue a program under this section only

after written notification to the Administrator and approval, if

needed, of a schedule for pressure testing.

Sec. 195.406 [Amended]

7. Section 195.406(a)(4) is amended by removing cross-reference

``Sec. 195.304'' and adding cross-reference ``Sec. 195.305''

8. A new Appendix B is added to part 195 to read as follows:

Appendix B--Risk-Based Alternative to Pressure Testing Older

Hazardous Liquid and Carbon Dioxide Pipelines

Risk-Based Alternative

This Appendix provides guidance on how a risk-based alternative

to pressure testing older hazardous liquid and carbon dioxide

pipelines rule allowed by Sec. 195.303 will work. This risk-based

alternative establishes test priorities for older pipelines, not

previously pressure tested, based on the inherent risk of a given

pipeline segment. The first step is to determine the classification

based on the type of pipe or on the pipeline segment's proximity to

populated or environmentally sensitive area. Secondly, the

classifications must be adjusted based on the pipeline failure

history, product transported, and the release volume potential.

[[Page 59481]]

Tables 2-6 give definitions of risk classification A, B, and C

facilities. For the purposes of this rule, pipeline segments

containing high risk electric resistance-welded pipe (ERW pipe) and

lapwelded pipe manufactured prior to 1970 and considered a risk

classification C or B facility shall be treated as the top priority

for testing because of the higher risk associated with the

susceptibility of this pipe to longitudinal seam failures.

In all cases, operators shall annually, at intervals not to

exceed 15 months, review their facilities to reassess the

classification and shall take appropriate action within two years or

operate the pipeline system at a lower pressure. Pipeline failures,

changes in the characteristics of the pipeline route, or changes in

service should all trigger a reassessment of the originally

classification.

Table 1 explains different levels of test requirements depending

on the inherent risk of a given pipeline segment. The overall risk

classification is determined based on the type of pipe involved, the

facility's location, the product transported, the relative volume of

flow and pipeline failure history as determined from Tables 2-6.

Table 1. Test Requirements--Mainline Segments Outside of Terminals, Stations, and Tank Farms

----------------------------------------------------------------------------------------------------------------

Pipeline segment Risk classification Test deadline \1\ Test medium

----------------------------------------------------------------------------------------------------------------

Pre-1970 Pipeline Segments C or B 12/7/2000 \3\................... Water only.

susceptible to longitudinal seam A 12/7/2002 \3\................... Water only.

failures \2\.

All Other Pipeline Segments...... C 12/7/2002 \4\................... Water only.

B 12/7/2004 \4\................... Water/Liq.\5\

A Additional pressure testing not

required.

----------------------------------------------------------------------------------------------------------------

\1\ If operational experience indicates a history of past failures for a particular pipeline segment, failure

causes (time-dependent defects due to corrosion, construction, manufacture, or transmission problems, etc.)

shall be reviewed in determining risk classification (See Table 6) and the timing of the pressure test should

be accelerated.

\2\ All pre-1970 ERW pipeline segments may not require testing. In determining which ERW pipeline segments

should be included in this category, an operator must consider the seam-related leak history of the pipe and

pipe manufacturing information as available, which may include the pipe steel's mechanical properties,

including fracture toughness; the manufacturing process and controls related to seam properties, including

whether the ERW process was high-frequency or low-frequency, whether the weld seam was heat treated, whether

the seam was inspected, the test pressure and duration during mill hydrotest; the quality control of the steel-

making process; and other factors pertinent to seam properties and quality.

\3\ For those pipeline operators with extensive mileage of pre-1970 ERW pipe, any waiver requests for timing

relief should be supported by an assessment of hazards in accordance with location, product, volume, and

probability of failure considerations consistent with Tables 3, 4, 5, and 6.

\4\ A magnetic flux leakage or ultrasonic internal inspection survey may be utilized as an alternative to

pressure testing where leak history and operating experience do not indicate leaks caused by longitudinal

cracks or seam failures.

\5\ Pressure tests utilizing a hydrocarbon liquid may be conducted, but only with a liquid which does not

vaporize rapidly.

Using LOCATION, PRODUCT, VOLUME, and FAILURE HISTORY

``Indicators'' from Tables 3, 4, 5, and 6 respectively, the overall

risk classification of a given pipeline or pipeline segment can be

established from Table 2. The LOCATION Indicator is the primary

factor which determines overall risk, with the PRODUCT, VOLUME, and

PROBABILITY OF FAILURE Indicators used to adjust to a higher or

lower overall risk classification per the following table.

Table 2.--Risk Classification

----------------------------------------------------------------------------------------------------------------

Hazard location Product/volume Probability of failure

Risk classification indicator indicator indicator

----------------------------------------------------------------------------------------------------------------

A.............................. L or M................. L/L................... L.

B.............................. Not A or C Risk Classification

C.............................. H...................... Any................... Any.

----------------------------------------------------------------------------------------------------------------

H=High M=Moderate L=Low.

Note: For Location, Product, Volume, and Probability of Failure Indicators, see Tables 3, 4, 5, and 6.

Table 3 is used to establish the LOCATION Indicator used in

Table 2. Based on the population and environment characteristics

associated with a pipeline facility's location, a LOCATION Indicator

of H, M or L is selected.

Table 3.--Location Indicators--Pipeline Segments

----------------------------------------------------------------------------------------------------------------

Indicator Population \1\ Environment \2\

----------------------------------------------------------------------------------------------------------------

H...................................... Non-rural areas................................ Environmentally

sensitive \2\ areas.

M ............................................... ......................

L...................................... Rural areas.................................... Not environmentally

sensitive \2\ areas.

----------------------------------------------------------------------------------------------------------------

\1\ The effects of potential vapor migration should be considered for pipeline segments transporting highly

volatile or toxic products.

\2\ We expect operators to use their best judgment in applying this factor.

Tables 4, 5 and 6 are used to establish the PRODUCT, VOLUME, and

PROBABILITY OF FAILURE Indicators respectively, in Table 2. The

PRODUCT Indicator is selected from Table 4 as H, M, or L based on

the acute and chronic hazards associated with the product

transported. The VOLUME Indicator is selected from Table 5 as H, M,

or L based on the nominal diameter of the pipeline. The Probability

of Failure Indicator is selected from Table 6.

[[Page 59482]]

Table 4.--Product Indicators

------------------------------------------------------------------------

Indicator Considerations Product examples

------------------------------------------------------------------------

H........................... (Highly volatile and (Propane, butane,

flammable). Natural Gas Liquid

(NGL), ammonia)

Highly toxic........ (Benzene, high

Hydrogen Sulfide

content crude

oils).

M........................... Flammable--flashpoin (Gasoline, JP4, low

t 18''.

M................................. 10''-16'' nominal diameters.

L................................. 8'' nominal diameter.

------------------------------------------------------------------------

H=High M=Moderate L=Low.

Table 6 is used to establish the PROBABILITY OF FAILURE

Indicator used in Table 2. The ``Probability of Failure'' Indicator

is selected from Table 6 as H or L.

Table 6.--Probability of Failure Indicators

[in each haz. location]

------------------------------------------------------------------------

Failure history (time-dependent

Indicator defects) \2\

------------------------------------------------------------------------

H \1\............................. >Three spills in last 10 years.

L................................. Three spills in last 10

years.

------------------------------------------------------------------------

H=High L=Low.

\1\ Pipeline segments with greater than three product spills in the last

10 years should be reviewed for failure causes as described in subnote

\2\. The pipeline operator should make an appropriate investigation

and reach a decision based on sound engineering judgment, and be able

to demonstrate the basis of the decision.

\2\ Time-Dependent Defects are defects that result in spills due to

corrosion, gouges, or problems developed during manufacture,

construction or operation, etc.

Issued in Washington, DC, on October 26, 1998.

Kelley S. Coyner,

Administrator, Research and Special Programs Administration.

[FR Doc. 98-29242 Filed 11-3-98; 8:45 am]

BILLING CODE 4910-60-P