Emergency Flow Restricting Devices/Leak Detection Systems

DEPARTMENT OF TRANSPORTATION

Research and Special Programs Administration

49 CFR Part 195

[Docket No. PS-133; Notice 1]

RIN 2137--AC 39

Emergency Flow Restricting Devices/Leak Detection Systems

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

ACTION: Advance notice of proposed rulemaking.

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

SUMMARY: In 1991, the Department issued a report on emergency flow

restricting devices (EFRDs) that proposed seeking public input on the

placement of EFRDs at certain locations on hazardous liquid pipelines.

The Pipeline Safety Act of 1992 mandated that the Department issue

regulations prescribing the circumstances under which operators must

use EFRDs and other equipment used to detect and locate pipeline

ruptures on hazardous liquid pipelines. The regulations are to be

issued following a survey and assessment of the effectiveness of such

equipment. This advance notice of proposed rulemaking (ANPRM) poses a

series of questions in order to solicit public input for the survey

process.

DATES: Interested persons are invited to submit written comments in

duplicate by April 19, 1994. Late-filed comments will be considered to

the extent practicable. Interested persons should submit as part of

their written comments all the material that is considered relevant to

any statement of fact or argument made.

ADDRESSES: Send comments in duplicate to the Dockets Unit, Room 8421,

Research and Special Programs Administration, U.S. Department of

Transportation, 400 Seventh Street SW., Washington, DC 20590. Identify

the docket and notice numbers stated in the heading of this advance

notice. All comments and materials cited in this document will be

available in the docket for inspection and copying in room 8421 between

8 a.m. and 4 p.m. each working day. Visitors are admitted to DOT

headquarters building through the southwest quadrant at Seventh and E

Streets. Commenters may request copies of the questions in a format

which can be filled out and returned to the RSPA. Requests should be

made to Lloyd W. Ulrich, Office of Pipeline Safety, room 2335, 400

Seventh Street SW., Washington, DC 20590, telephone (202) 366-4556 or

FAX (202) 366-4566.

FOR FURTHER INFORMATION CONTACT: Lloyd W. Ulrich, (202) 366-4556,

regarding the subject matter of this advance notice, or Dockets Unit,

(202) 366-5046, for copies of this advance notice or other material in

the docket.

SUPPLEMENTARY INFORMATION:

Background

The RSPA has been concerned for some time with the issue of more

rapid leak detection on hazardous liquid pipelines, and the optimum

placement of EFRDs to limit commodity release after the location of the

release in the hazardous liquid pipeline has been identified.

Section 203 of the Hazardous Liquid Pipeline Safety Act (codified

at 49 U.S.C. app. Sec. 2002(n)) as amended by the Pipeline Safety Act

of 1992 (the 1992 Act) (Pub. L. 102-508) mandated the Secretary, within

two years of enactment, to conduct a survey and assess the

effectiveness of emergency flow restricting devices (EFRDs) and other

procedures, systems, and equipment used to detect and locate hazardous

liquid pipeline ruptures and minimize product releases from hazardous

liquid pipeline facilities. The 1992 amendments further mandated that

the Secretary issue regulations within two years of completion of the

survey and assessment. These regulations would prescribe the

circumstances under which operators of hazardous liquid pipelines would

use EFRDs and other procedures, systems, and equipment to detect and

locate pipeline ruptures and minimize product release from pipeline

facilities. The Secretary has delegated this authority to the Research

and Special Programs Administration (RSPA) (See 49 CFR 1.53).

Also, the Department's March 1991 report titled ``Emergency Flow

Restricting Devices Study'' contained proposals that we seek public

input on the placement of EFRDs in urban areas, at water crossings, at

other critical areas affected by commodity release, and areas in close

proximity to the public outside of urban areas.

This ANPRM solicits public input for the survey process mandated by

the 1992 Act as well as the proposals from the Department's 1991 EFRD

study. The ANPRM requests information and data by posing a series of

questions. This approach is utilized rather than conducting a

traditional research survey of a selected number of respondents in

order to obtain a broader base of data and to accelerate the regulatory

process.

Notice on Highly Volatile Liquids--1978

In 1978, the RSPA issued an NPRM (43 FR 39402; September 5, 1978)

proposing requirements intended to limit spillage from hazardous liquid

pipelines carrying highly volatile liquids (HVL)1 in inhabited

areas by requiring installation of remotely controlled valves

(RCVs)2 or automatically controlled valves (ACVs).3 This

proposal was later withdrawn (46 FR 2130; January 8, 1981) because

hazardous liquid pipeline industry studies demonstrated that placement

of closely spaced valves over the full length of an HVL pipeline was

not a reasonable method of reducing the effects of an accident.

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

\1\The term ``HVL'' is defined in 49 CFR 195.2 as a hazardous

liquid which will form a vapor cloud when released to the atmosphere

and which has a vapor pressure exceeding 276 kPa (40 psia) at

37.8 deg.C (100 deg.F). The commodities included in the term ``HVL''

are LPG, anhydrous ammonia, and certain natural gas liquids.

\2\An RCV is any valve which is operated from a location remote

from where the valve is installed. The location is usually at the

pipeline control or dispatching center. The linkage between the

pipeline control center and the RCV may be by fiber optics,

microwave, telephone lines, or satellite.

\3\An ACV is any valve which automatically closes in response to

a rate of pressure drop or flow rate in the pipeline which exceeds a

preset level.

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

Mounds View Accident

A July 8, 1986, accident on a gasoline pipeline focused interest on

EFRDs. The accident, caused by a ruptured pipe seam on a gasoline

pipeline in Mounds View, Minnesota, resulted in two deaths, one injury,

and property damage well in excess of $1,000,000. The accident was

exacerbated by backflow or draining from the pipeline after the

manually operated valves on either side of the ruptured section were

closed. The spill ignited approximately 20 minutes after the rupture.

It took the pipeline operator over 1 hour and 40 minutes from the time

of the rupture to isolate the ruptured section. Since this accident,

the pipeline company installed a computerized leak-detection system and

RCVs on either side of Mounds View (a distance of about 5.7 miles).

Advance Notice on Certain Safety Proposals--1987

In 1987, as a result of the same accident, the RSPA again addressed

RCVs and ACVs in an ANPRM (52 FR 4361; February 11, 1987). This ANPRM

invited public comment on the merit of certain safety proposals

advanced by Congress, the Minnesota Commission on Pipeline Safety, and

the National Transportation Safety Board. One safety proposal was to

convert shutoff valves required by the pipeline safety regulations on

existing pipelines to RCVs or ACVs, and require similar valves on new

pipeline construction.

Both gas and hazardous liquid pipeline operators indicated that

neither RCVs nor ACVs were installed as shutoff valves as standard

practice. They indicated that RCVs and ACVs had little effect to

mitigate the extent of the spill because often, especially in populated

areas on gas pipelines, ignition occurred before either type of valve

could shut down a pipeline.

The specific concern of false closure of ACVs was identified in

these comments. There was substantial agreement by both gas and

hazardous liquid pipeline operators that ACVs should not be used as

EFRDs because of their unreliability. This unreliability was due to the

inability of ACV sensors to distinguish between a leak and normal

operating fluctuations. Pipeline operators indicated numerous

documented cases of unintended closures of ACVs. A false closure of an

ACV on a hazardous liquid pipeline can cause an immediate pressure

buildup or surge which may result in a pipeline rupture.

On September 23, 1987, the ANPRM was discussed at the joint meeting

of the RSPA's Technical Pipeline Safety Standards Committee and the

Technical Hazardous Liquid Pipeline Safety Standards Committee. (Both

technical committees were established by the Secretary of

Transportation to advise the Department on the technical feasibility,

reasonableness, and practicability of all proposed gas and hazardous

liquid pipeline safety standards and all amendments to existing

standards.) The committees recommended that the Department study the

selective use of RCVs and ACVs.

Emergency Flow Restricting Devices Study--1991

Section 305 of the Pipeline Safety Reauthorization Act of 1988

(Public Law 100-561), enacted on October 31, 1988, directed a study of

the safety, cost, feasibility, and effectiveness of requiring gas and

hazardous liquid pipeline operators to install EFRDs in existing and

future pipeline systems in varying circumstances and locations.

In March 1991, in response to this Congressional mandate, the

Department issued the study titled ``Emergency Flow Restricting Devices

Study.'' One of the conclusions in the study was that RCVs and check

valves4 are the only feasible EFRDs. Another conclusion was that

requiring the retrofitting of all existing manually operated valves to

RCVs on hazardous liquid pipelines in urban locations, as well as new

valves in urban areas appeared to be cost effective. Still another

conclusion in the study was that for an RCV to be effective, a modern

supervisory control and data acquisition (SCADA) system with a well-

designed leak detection subsystem was necessary to reduce spills from

hazardous liquid pipelines. The study found that there was no

significant benefit from installing EFRDs on gas transmission

pipelines.

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

\4\Check valves are valves that permit fluid to flow freely in

one direction and contain a mechanism to automatically prevent flow

in the other direction.

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

SCADA systems utilize computer technology to analyze data (e.g.,

pressure, temperature, and delivery flow rates) that are continuously

gathered from remote locations on the pipeline. Computer analysis of

this data is used to assist in day-to-day operating decisions on the

pipeline and to provide input for real-time models of the pipeline

operation which can identify and locate leaks.

SCADA-based leak detection subsystems are composed of hardware and

software programs that employ a real-time modelling procedure to

compare the current operational conditions of a segment of pipe to an

``ideal'' operating state. This ideal state is sometimes recalibrated

during operations to accommodate variations in conditions (e.g.,

temperature or pressure fluctuations in the pipe that occur due to

changes in the materials in transport or external environmental

conditions). An ``alarm'' is sent to a central operator when the

software model detects a condition that is ``substantially'' different

from the idealized state. What makes the condition ``substantially''

different, thereby triggering the alarm, is determined by the model

designer and the conditions imposed on the model, as well as by the

amount of data available on the ``ideal'' state and its normal

operational variability.

An RCV can operate without a SCADA system installed. However, for

an RCV to be used effectively in reducing a spill, the dispatcher must

be able to determine that a pipeline failure has occurred, identify the

location of the failure, and then quickly initiate closure of the

valve. Accomplishing these actions in a timely manner requires the

installation of a SCADA system including a well-designed leak detection

subsystem. The extensive pollution which resulted from a 1988 pipeline

failure in Maries County, Missouri, to be discussed later in this

ANPRM, might have been avoided if a leak detection subsystem had been

installed with the SCADA system allowing operator personnel to detect

the leak.

It is clear from the RSPA's analysis of information and data

obtained in conducting the March 1991 EFRD study, that spillage from a

pipeline failure can be significantly reduced by RCVs only where a

modern SCADA system is equipped with a well-designed leak detection

subsystem. The type and sophistication of the control system, installed

as part of an existing SCADA system, depends on the age of the control

system.

The March 1991 EFRD study contained a number of proposals to

address the issue of EFRDs. One of the proposals was that the

Department conduct a research study on whether SCADA systems, including

well-designed leak detection subsystems, should be required on

hazardous liquid pipelines in order to enhance the safe operation of

the pipelines. Enhanced safety requirements would include provision for

more rapid response following accidents, including valve spacing

criteria and initiating the closure of RCVs. This study is presently

being conducted by the Volpe National Transportation System Center

(VNTSC) and is discussed later in this ANPRM.

Another proposal from this study was for the RSPA to issue a notice

of proposed rulemaking proposing to require, on hazardous liquid

pipelines with SCADA systems installed, that existing manually operated

main line block valves5 in urban areas be retrofitted to make them

RCVs and install RCVs when new valves are installed in urban areas.

This ANPRM seeks data on valves located in urban areas.

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

\5\A valve which provides a positive shut off of commodity flow

both upstream and downstream of the valve is generally known as a

``block valve'' because it blocks the flow in the pipeline.

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

Other proposals in the study suggested public input on whether the

hazardous liquid pipeline safety regulations in 49 CFR part 195 should

be revised to require valve spacing criteria for EFRDs at the following

locations: (1) Where the valves could most effectively reduce the

likelihood of the escaping liquid entering the water at water crossings

that are more than 100-feet wide, and on either side of a reservoir

holding water for human consumption; (2) At other critical areas

affected by commodity release; and (3) At specific locations outside of

urban areas on hazardous liquid pipelines in proximity to the public.

Some of the questions posed in this ANPRM are designed to provide

data on which the RSPA will decide on a further course of action

concerning the proposed placement of EFRDs in these locations.

Past Data Collection

To broaden the data base for the March 1991 EFRD study, the RSPA

solicited information from the public, including gas and hazardous

liquid pipeline operators and equipment manufacturers, through a

Federal Register notice (54 FR 20945; May 15, 1989). A series of 15

questions addressed a number of EFRD/leak detection-related issues

including SCADA technology, establishing a maximum allowable spill

value, and criteria for valve spacing. The 72 responses to the notice

are contained in Docket PS-104; Notice 1 and are available for review

in the Docket Unit, room 8421.

The notice included a series of questions about leak detection

subsystems which are part of operators' SCADA systems. The responses to

the questions indicated SCADA systems are becoming more sophisticated

and leak detection subsystems are becoming more common on hazardous

liquid pipelines. The sensitivity of leak detection subsystems on

hazardous liquid pipelines was reported to range from 0.5 percent of

flow to 5 percent of flow over a 1- to 2-hour period. Once a leak is

suspected, the time for the dispatcher to respond by closing valves

ranges from a few minutes for an RCV to an average of about 2 hours for

manually operated valves.

Commenters to the notice were also asked to discuss the advantages

and disadvantages of establishing a valve spacing requirement based on

a maximum spill criterion. Two advantages cited by commenters to

support establishing a maximum spill criterion on a hazardous liquid

pipeline were: (1) Reduction in the exposure to the public of the

possible hazard created by a spill and (2) improved contingency plans

since the plans could be based on a spill of a set volume. However,

commenters cited more disadvantages than advantages.

Most hazardous liquid pipeline operators opposed setting a maximum

spill criterion. They indicated a maximum acceptable spill would vary

widely along the length of any pipeline depending on spill location.

Establishing one criterion for all pipelines would not account for the

variables at each spill location. Commenters indicated that more

important than establishing an arbitrary spill limit is the need to

consider the line profile, drainage gradient, length, and diameter of

the hazardous liquid pipeline, susceptibility to outside force damage,

population density, and potential hazards to public safety and the

environment. Commenters indicated a spill limit would require more

valves, particularly on large diameter hazardous liquid pipelines. This

would increase the opportunity for inadvertent valve closure, leakage

from the valve itself, and vandalism. One commenter stated that

protection of the public and environment is related more to exposure of

the hazardous liquid pipeline to the public and response time in

detecting and responding to a failure, than to setting a limit on the

volume of product released. This commenter stated: ``The most effective

means for mitigating potential pipeline failure hazards is prompt leak

identification, rapid pipeline shutdown, and immediate dispatch of

response crews to the failure site.''

Commenters were asked if the spacing of RCVs and ACVs is determined

by a maximum spill from the hazardous liquid pipeline, what should that

maximum spill value be? None of the commenters provided a maximum spill

value. Hazardous liquid pipeline operators reiterated that the

information provided in the responses to the valve spacing question

should be used in the context of spill mitigation rather than to

establish a single maximum spill criterion. One commenter stated that,

in addition to pipe diameter, terrain, and the pipeline's route near or

in urban areas, the RSPA should consider the probability of failure,

magnitude of the leak, and consequences of the leak in establishing a

maximum spill criterion.

December 24, 1988, Failure in Maries County, Missouri

The legislative history for the 1992 Act cites a December 24, 1988,

failure in Maries County, Missouri to demonstrate the need of adequate

leak detection equipment. The failure resulted in a crude oil spill of

approximately 20,554 barrels (863,268 gallons). The cause was the

abrupt change in pressure and fluid flow from the switching of flowing,

low density crude oil from one pipeline into another containing a

substantially heavier oil.

Crude oil released entered a tributary of the Gasconade River, the

Gasconade River, the Missouri River, and eventually the Mississippi

River near St. Louis, Missouri. In order to control the contamination

from the large volume of crude oil released, it was necessary to shut

down several water companies along these rivers and a brewery in St.

Louis.

Failure of pipeline personnel at the dispatching station to

recognize that a rupture had occurred and to shut down the pipeline

greatly increased the volume of crude oil spilled.

The Gasconade River and its tributary, into which the crude oil

spill first entered, were bracketed by manually operated block valves.

The RSPA estimates that the installation of a check valve would have

prevented drainage from the 5 mile of pipe on either side of the river,

thereby substantially reducing the size of the spill. Also, the

installation of a leak detection subsystem on the SCADA system would

probably have substantially reduced the size of the spill.

Report From the National Institute of Standards and Technology

The legislative history of the 1992 Act also cites a July 1989

report from the National Institute of Standards and Technology (NIST),

U.S. Department of Commerce (Report Number NISTIR 89-4136) which

resulted from an investigation of the Maries County, Missouri pipeline

failure. In the report, titled ``An Assessment of the Performance of

Older ERW Pipelines'', NIST found that the installation of EFRDs could

significantly reduce the damage from pipeline failures and recommended

that they be installed in ``critical risk locations.''

Current SCADA Study by the VNTSC

In May 1992, the RSPA commenced a research study with the VNTSC to

analyze SCADA systems and computer-generated leak detection systems.

The purpose of the research study is to determine the feasibility and

costs of requiring pipeline operators to install a SCADA system

including a leak detection subsystem, and determine what impediments

exist or what system improvements are needed to minimize the time it

takes SCADA systems to detect and locate leaks, and make

recommendations to resolve these difficulties. As mentioned previously,

this new initiative is based on findings from the Department's March

1991 EFRD study concerning RCVs. These valves maximize the value of

SCADA-based leak detection systems by helping to mitigate damages from

detected leaks.

The first phase of this study included a literature search on the

subject, on-site interviews with seven pipeline operators, interviews

with five equipment vendors, and development of a mathematical model

describing optimal valve spacing for given annual pipeline failure

rates per mile and costs, and a method to evaluate alternative leak

detection system performance characteristics to reduce pipeline spill

volumes.

Every pipeline operator surveyed by the contractor used some sort

of SCADA system. Most operators had at least one computerized leak

detection system, either one purchased from a vendor, custom designed

by the operator, or a combination of the two systems. All operators

interviewed believed that the condition of high false alarm rates was a

major drawback to the installation and operation of leak detection

systems. The problem occurs due to the required trade-off between the

threshold volume sensitivity of the leak detection system and the

resulting false alarm rate when this sensitivity is too high. All the

operators interviewed emphasized that the most critical link in leak

detection was the interface between the system itself and the pipeline

dispatcher, and that there was no substitute for a highly competent

pipeline dispatcher.

The VNTSC is drafting a report on the first phase of the study.

Once the report is completed, a copy will be placed in the docket to

this rulemaking.

Regulatory Analysis and Notices

A. Impact Assessment

This ANPRM is not considered a significant regulatory action under

section 3(f) of Executive Order 12866 and was not reviewed by the

Office of Management and Budget. The ANPRM is not considered

significant under the Regulatory Policies and Procedures of the

Department of Transportation (44 FR 11034).

B. Regulatory Flexibility Act

This ANPRM would not have a significant economic impact on a

substantial number of small entities (i.e., small businesses,

governmental jurisdictions, and non-for-profit organizations) under the

criteria of the Regulatory Flexibility Act. This ANPRM would apply to

operators of hazardous liquid pipelines, all of whom are large

businesses. Therefore, I certify that this ANPRM will not, if

promulgated, have a significant economic impact on a substantial number

of small entities. This certification is subject to modification as a

result of a review of comments received in response to this ANPRM.

C. Federalism Assessment

The ANPRM has been analyzed in accordance with the principles and

criteria in Executive Order 12612 (``Federalism''), and does not have

sufficient federalism impacts to warrant the preparation of a

federalism assessment.

D. Paperwork Reduction Act

There are no new information collection requirements in this ANPRM.

Questions

The RSPA is issuing this ANPRM to solicit data from the public

through a series of questions as the means of conducting the survey

mandated in the 1992 Act. The response from the public to these

questions will aid in developing proposals on what circumstances and

criteria operators must install EFRDs and other equipment to limit

product release from hazardous liquid pipelines. The failures discussed

above suggest that releases can be reduced when EFRDs and well-designed

leak detection systems are installed on hazardous liquid pipelines.

Assessing the data received from the questions in the ANPRM should

accelerate the rulemaking process required by the 1992 Act. The data

gathered by this ANPRM, the findings from earlier reports on the

subject of EFRDs, including the Department's March 1991 EFRD study, and

the work accomplished so far in the SCADA contract with the VNTSC could

form the basis for any notice of proposed rulemaking concerning the

proposed placement of EFRDs and criteria for leak detection systems.

The RSPA is considering a systems approach to reducing spills from

hazardous liquid pipelines. The system involved includes equipment,

personnel, software and procedures to accomplish three tasks: (1)

Detect that a failure and resultant spill has occurred; (2) Identify

the location of the spill; and (3) Shut the pipeline down in order to

reduce the amount of the spill. The first two tasks involve

computerized leak detection systems, while the third task involves the

installation of EFRDs.

Many of the following questions are directed to the operators of

hazardous liquid pipelines. They relate to pipeline system operational

data in addition to the physical location of pipeline facilities in

relation to geographical and topographical features which can only be

obtained from pipeline operators. However, the RSPA solicits comments

to questions which do not involve data on a particular hazardous liquid

pipeline from other members of the public including State agencies,

trade associations, and environmental organizations, both private and

public. The RSPA believes that State pipeline safety agencies can

contribute significantly to this rulemaking because of the States'

unique experience with regulating intrastate hazardous liquid pipelines

and as the Department's agent on interstate hazardous liquid pipelines.

Questions 18 and 19 are directed to the nonregulated public. These

commenters are requested also to suggest additional questions,

including clarification questions, which may emerge from reviewing this

ANPRM.

To aid in analysis of the responses, commenters are requested to

respond using the same numbering system which is used in this ANPRM.

SCADA-based Leak Detection System Sensitivity and Reliability

The RSPA needs data on which to base decisions on what should be

proposed for SCADA-based leak detection systems. The RSPA is starting

from the premise that most if not all hazardous liquid pipeline

operators have installed a SCADA system which is used for the everyday

efficient operation of the pipeline. The SCADA study by the VNTSC has,

so far, found this to be true. (Commenters are requested to indicate if

this premise is true.) The RSPA must decide whether to propose: (1) A

specific type of leak detection system; (2) whether to propose

requiring certain criteria which would embody the attributes of all of

the presently recognized computerized leak detection systems; (3) a

combination of (1) and (2); or (4) some other leak detection system

requirement which at present is unknown to the RSPA but which may

emerge from comments to this ANPRM.

The questions are intended to obtain responses which relate to

operational data that a hazardous liquid operator has concerning the

SCADA-based leak detection system installed on its pipeline system

including the sensitivity and reliability of that system.

Questions 1 through 6 primarily relate to the experience on a

segment of the operator's hazardous liquid pipeline system which is

covered by a SCADA-based leak detection system. If the operator has

segments of its hazardous liquid pipeline system covered by more than

one SCADA-based leak detection system, please submit responses to the

series of questions 1 through 6 for each segment of the covered

pipeline system. For instance, a SCADA-based leak detection system may

be installed on a 400 mile segment of an interstate pipeline in Texas

and another SCADA-based leak detection system on a 200 mile segment in

Virginia. The RSPA requests a separate set of responses for each

segment, not aggregate responses for all of the SCADA-based leak

detection systems for all parts of the operator's pipeline system.

Several topics will be addressed in the set of questions below.

These are: (1) The method(s) of leak detection in use on the segment

described in the data submission; (2) leak detection alarms which occur

at the hazardous liquid pipeline systems operating center; (3) the leak

detection and SCADA system availability; and (4) the actual performance

of leak detection systems in identifying and locating leaks on an

operational hazardous liquid pipeline.

If the operator does not presently have this data, we encourage the

operator to gather the data for at least one month and then submit it

to the RSPA. System alarms history should be provided to the RSPA as a

log and may be submitted either as a computer printout or on a diskette

using standard ASCII format as long as the segment identifying

information is clearly noted on the data. Experimental (or simulation-

based) data may be provided as well as operational data which only

reflect actual operational experience.

The leak detection performance data should be provided as a log and

may be submitted either as a computer printout or on a diskette using

standard ASCII format. Historical performance data gathered during

developmental phases such as system installation and modification also

should be submitted.

It would be helpful if commenters group data for each different

data collection time period or pipeline segment, so that all data

(questions 1-6) relates to only one specific segment and time period.

For the purposes of these questions, a pipeline segment is defined as

that part of the pipeline between two points where the product can be

contained, such as between two pressure pump stations, between a

pressure pump station and a terminal, between a pressure pump station

and a valve, or between two valves.

Question 1: Provide the following general information about the

segment of hazardous liquid pipeline to which the series of questions

1-6 relate:

1.1 Pipeline segment length description covered in this data

submission:

1.1.1 Starting point (mile post or survey station no.)________

1.1.2 Ending point (mile post or survey station no.)________

1.1.3 Length of segment (miles)________

1.2 Pipeline nominal diameter (in.)________

1.3 Number of pumping stations on segment? ________

1.4 Number of injection points on segment?________

1.5 Number of delivery points on segment?________

1.6 Commodity(s) transported during this data history________

1.7 Nominal flow rate (bbls/day) ________

1.8 Beginning date covered by this data history (MM/DD/YY)________

1.9 Ending date of this data history (MM/DD/YY) ________

Question 2: Classify the leak detection system(s) installed on this

pipeline segment (check each that applies and answer questions 4

through 6 for each system checked).

2.1 Mass balance________

2.2 Pressure wave ________

2.3 External hydrocarbon sensor________

2.4 Other (specify) ________

Question 3: For each leak detection system checked in Question 2,

check whether the system was supplied by an independent vendor or was

the system developed within your company.

3.1 For the system in Question 2.1?

3.1.1 Vendor (name) ________

3.1.2 Internal company developed________

3.2 For the system in Question 2.2?

3.2.1 Vendor (name) ________

3.2.2 Internal company developed ________

3.3 For the system in Question 2.3?

3.3.1 Vendor (name) ________

3.3.2 Internal company developed ________

3.4 For the system in Question 2.4?

3.4.1 Vendor (name)________

3.4.2 Internal company developed ________

Question 4: For the alarm history, leak detection system

availability history, and performance data of the leak detection system

submitted, include answers to the following:

4.1 For the time period reported, at what threshold volume was the

leak detection system set to alarm (including any error bandwidth that

is incorporated into that amount) (bbls.)?________

4.2 At that volume how long should detection take (mins.)? ________

4.3 What was the average detection time for that volume

(mins.)?________

4.4 For each alarm during the time period reported in your response to

Question 4, include the following data:

4.4.1 Alarm Initiated (MM/DD/YR & hours & minutes in military

time6)________

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

\6\``Military time'' is using a 24 hour clock. For instance,

4:00 pm = 1600 hours or 5:15 pm = 1715 hours.

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

4.4.2 Alarm Cleared (MM/DD/YR & hours & minutes in military

time)________

4.4.3 The length of time it took to identify the cause of the

alarm (if not equal to the difference between the initiation and

cleared time)(mins.)?________

4.5 For each alarm, was the alarm attributed to one of these causes

(Y/N)?

4.5.1 A leak ________ (If ``yes'', go to 4.6)

4.5.2 An operational change ________

4.5.3 Data errors (associated with telemetry fluctuations)

________

4.5.4 Component failure (hardware or telecommunications) ________

4.5.5 Human error (e.g., failure to adjust the leak detection

software system to commodity-specific parameters) ________

4.5.6 Other (specify) ________

4.5.7 Undetermined ________

4.6 If a leak was detected--

4.6.1 What was the cause of the leak (check)?

4.6.1.1 Corrosion? ________

4.6.1.2 Failed pipe or pipe seam? ________

4.6.1.3 Outside force damage by other than natural forces?

________

4.6.1.4 Outside force damage by natural forces? ________

4.6.1.5 Malfunction of control or relief equipment? ________

4.6.1.6 Operator error? ________

4.6.1.7 Other (specify) ________

4.6.2 Was the leak on pipe originally installed on the pipeline

segment (Y/N)? ________

4.6.3 What year was the pipe originally installed (year)? ________

4.6.4 If the answer to 4.6.2 was ``no'', what year was the pipe

replaced or modified (year)? ________

4.6.5 What action did you take?

4.6.5.1 Shut pipeline down (Y/N) ________

4.6.5.2 Shut down leak detection system (Y/N) ________

4.6.5.3 Left pipeline and leak detection systems running,

conducted visual inspection (Y/N) ________

4.6.5.4 Other (specify) ________

Question 5: For leak detection system availability (SCADA-based or

non-SCADA-based), include answers to the following:

5.1 For each instance of leak detection system unavailability reported

during the time period, include the following data:

5.1.1 Was this a complete shutdown of the SCADA/leak detection

system (Y/N)? ________ (If ``no'', go to question 5.3)

5.2 If ``yes'', answer the following (check all that apply):

5.2.1 Date and time the system stopped running (MM/DD/YR & hours &

minutes in military time) ________

5.2.2 Date and time the system resumed (MM/DD/YR & hours & minutes

in military time) ________

5.2.3 Was the shutdown attributed to one of these causes (Y/N)?

5.2.3.1 Dispatcher decision? ________

5.2.3.2 Input failure (telemetry/telecomm error)? ________

5.2.3.3 Software failure of the SCADA system? ________

5.2.3.4 Software failure of the leak detection system? ________

5.2.3.5 Software failure of both? ________

5.2.3.6 Hardware failure of the SCADA system? ________

5.2.3.7 Hardware failure of the leak detection system? ________

5.2.3.8 Hardware failure of both? ________

5.2.3.9 Undetermined ________

5.3 If the leak detection system itself did not completely shut down,

did it issue an alarm (Y/N)? ________

5.4 If an alarm was issued, was the problem attributed to any of the

following (Y/N)?

5.4.1 Dispatcher decision? ________

5.4.2 Input failure (telemetry/telecomm error)? ________

5.4.3 Software failure of the SCADA system? ________

5.4.4 Software failure of the leak detection system? ________

5.4.5 Software failure of both? ________

5.4.6 Hardware failure of the SCADA system? ________

5.4.7 Hardware failure of the leak detection system? ________

5.4.8 Hardware failure of both? ________

5.4.9 Undetermined ________

Question 6: Answer the following on leak detection system

performance:

6.1 What was the circumstance(s) under which data was collected for

this segment and time period?

6.1.1 System development (Y/N) ________

6.1.2 Leak detection system pre-operational demonstrations on a

segment of operational pipeline (define segment length) (Y/N) ________

6.1.3 Existing system modification/testing (Y/N) ________

6.1.4 Actual system operation (Y/N) ________

6.1.5 Other (specify) ________

6.2 For each leak detected by the system during the time period,

include the following data:

6.2.1 The specific detection threshold volume (include any error

bandwidth that is incorporated in that amount)(bbls.) ________

6.2.2 Pipeline length between leak detection measuring devices in

the pipeline segment on which leak occurred (miles) ________

6.2.3 Commodity transported at the time of the alarm ________

6.2.4 Flow rate at the time of the alarm (bbls/hr) ________

6.2.5 Estimated (or actual) leak volume (bbls.) ________

6.2.6 Estimated (or actual if known) size of hole or rupture (in.)

________

6.2.7 Estimated7 (or actual) date and time leak occurred (MM/

DD/YR & hours & minutes in military time) ________

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

\7\Data provided from simulations should include the simulated

leak start and end times, however, actual leak start times are not

expected from operational data since there is a lag between the

actual leak start and when it is detected.

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

6.2.8 Date and Time leak detected (MM/DD/YR & hours & minutes in

military time) ________

6.2.9 Date and Time leak located (MM/DD/YR & hours & minutes in

military time) ________

6.2.10 Location of leak as indicated by leak detection system

(mile post or survey station no.) ________

6.2.10.1 Was a leak detection/SCADA system alarm issued (Y/N)?

________

6.2.10.2 If ``yes'', the date and time alarm issued (MM/DD/YR &

hours & minutes in military time) ________

6.2.10.3 If ``yes'', the date and time alarm cleared (MM/DD/YR &

hours & minutes in military time) ________

6.2.10.4 Dispatcher response (check all that apply):

6.2.10.4.1 Pipeline shutdown ________

6.2.10.4.2 Leak detection system shutdown only ________

6.2.10.4.3 Contacted pipeline personnel to check for operational

or system explanations for alarm (other than a leak) ________

6.2.10.4.4 Dispatched personnel to approximate leak location

________

6.2.10.4.5 Other (specify) ________

6.2.11 Actual location of leak as determined by field observation

(mile post or survey station no.) ________

Placement of EFRDs at water crossings, locations affected by

commodity release, and rural areas where the public is in proximity The

request for information notice, documented in the Department's March

1991 EFRD study, asked ``Where should RCVs and ACVs be placed and

why?'' The response provided a number of specific locations, e.g.,

locations where possible ground movement might occur, and densely

populated locations, such as near a school or hospital, near an office

building or factory, or near a shopping center. River crossings were

also specified by some commenters. Although conventional wisdom would

seem to suggest installing RCVs at these locations, the RSPA presently

has no data which supports requiring the installation of EFRDs at these

locations. The number of these locations is unknown, but one of the

following questions will solicit data on the number of such areas which

might be affected by a pipeline release.

Likewise, the number of failures which have resulted in water

pollution is unknown because the Department does not require the

occurrence of pollution to be identified on the hazardous liquid

pipeline accident report. However, the RSPA knows from research for

developing the interim final rule for onshore oil pipeline response

plans (58 FR 244, January 5, 1993) under the Oil Pollution Act of 1990,

Public Law No. 101-380, 104 Stat. 484, (OPA 90) that of the

approximately 2,700 oil pipeline spills reported each year to the

Environmental Protection Agency (EPA), about half affect water. The

accident effects which would be reduced by the installation of EFRDs

are related more to pollution than safety. Once the hazardous liquid

mixes with water, the likelihood of a fire or explosion is reduced

considerably. Two of the following questions address the issue of water

pollution as a result of pipeline failures.

The accidents which occurred in Mounds View, Minnesota and Maries

County, Missouri demonstrate that an assessment by the RSPA should be

conducted concerning the installation of EFRDs at specific locations

along hazardous liquid pipelines where the pipelines are in proximity

to the public in rural areas, bodies of water (particularly bodies of

water containing drinking water intakes), and other critical locations

affected by commodity release. This position is supported by the NIST

report (discussed above) which recommended installation of EFRDs in

critical risk locations to significantly limit the extent of damage if

a failure occurs. Critical risk locations are defined in the NIST

report as locations where the risk to public safety, property, and the

environment is great.

The RSPA is also asking questions to gather data regarding

locations where valves are presently required by the regulations to

protect bodies of water (49 CFR 195.260 (e) and (f)). The regulations

in 49 CFR 195.260 currently require valves to be placed at locations

along a hazardous liquid pipeline that will minimize damage or

pollution from accidental discharges\8\ (Section 195.260 (c)), on

either side of a water crossing that is more than 100-feet wide

(Section 195.260(e)), and on either side of a reservoir holding water

for human consumption (Section 195.260(f)). The American Petroleum

Institute has indicated that most of these main line block valves are

manually operated. The current regulations do not require that they be

EFRDs. The financial impact on the regulated industry, if the RSPA were

to require the installation of RCVs, is unknown so cost data will be

obtained through responses to questions set forth below.

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

\8\DOT does not know on what factors hazardous liquid pipeline

operators base their judgment on where and how far apart to place

such valves.

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

These questions apply only to hazardous liquid pipelines with leak

detection systems since the March 1991 EFRD study concluded RCVs are

only effective where leak detection systems are installed. In addition,

the questions address valve spacing, particularly whether EFRDs should

be installed at other critical locations. Such critical locations would

be identified during the rulemaking process after a review of data on

water pollution from hazardous liquid pipeline spills collected by

agencies, such as the EPA, and as a result of the RSPA implementation

of OPA 90. Placement of a check valve on one side of a location and an

RCV on the other side could reduce the number of valves that would

require remote control capability.

The location of valves would be proposed, as a result of this

assessment, only to effectively reduce the amount of liquid entering a

body of water depending on the terrain, and would not necessarily be

immediately located on either side of the water crossing. For instance,

valves placed immediately on either side of a water crossing would be

effective in reducing pollution only if the failure was in the water

crossing, a rare occurrence according to anecdotal evidence. Valves

would only be effective in locations where the hazardous liquid

pipeline operator has a SCADA-based leak detection system so that each

RCV can be closed soon after a failure is detected.

Past Leak History

Question 7: For the period 1983 through 1992 (10 years), how many

failures have occurred on your pipeline system resulting in product

release entering a body of water?\9\________

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

\9\``Body of water'' includes, but is not limited to, creeks,

streams, rivers, tributaries to rivers, lakes, reservoirs, and

waters which are used for recreation.

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

Question 8: Provide the following information for each failure in

Question 7:

8.1 Date of the failure (MM/YY)________

8.2 What was the cause of the release (check)?

8.2.1 Corrosion?________

8.2.2 Failed pipe or pipe seam?________

8.2.3 Outside force damage by other than natural forces?________

8.2.4 Outside force damage by natural forces?________

8.2.5 Malfunction of control or relief equipment?________

8.2.6 Operator error?________

8.2.7 Other (specify)________

8.3 Answer the following concerning the commodity--

8.3.1 At the time of the release what was the total volume of

commodity between valves immediately upstream and downstream of the

release location (bbls.)?________

8.3.2 How much commodity was released (bbls.)?________

8.4 How far away from the point of release were the valves referred to

in 8.3.1 (miles)?________

8.4.1 Upstream from the release?________Type of valve (check)?

8.4.1.1 manual________

8.4.1.2 Check valve________

8.4.1.3 RCV________

8.4.1.4 ACV________

8.4.2 Downstream from the release?________Type of valve (check)?

8.4.2.1 manual________

8.4.2.2 Check valve________

8.4.2.3 RCV________

8.4.2.4 ACV________

8.5 Were these valves installed to comply with 49 CFR 195.260 (e) or

(f)\10\ (Y/N)?________

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

\10\49 CFR 195.260 (e) requires valves to be installed on each

side of a water crossing that is more than 100 feet wide from high-

water mark to high-water mark. 49 CFR 195.260 (f) requires valves to

be installed on each side of a reservoir holding water for human

consumption.

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

8.6 Were these valves EFRDs (RCVs or check valves)(Y/N)?________

8.7 Was there a leak detection system or systems operating on the

pipeline which experienced the release (Y/N)?________(specify each

type):

8.7.1 Pressure wave front monitoring?________

8.7.2 Volume monitoring?________

8.7.3 Other (specify)?________

8.8 Indicate for each activity the time to shut down the pipeline

8.8.1 Detection time (mins.)?________

8.8.2 Shutdown time including shutdown of pumping stations and

isolation of the pipeline section (mins)?________

8.8.3 Time to drain the pipeline section (mins.)?________

8.8.4 Total time to shut down pipeline from the time release was

detected to completion of drainage from the section involved

(mins.)?________

8.9 If the pipeline section did not contain EFRDs, would the

installation of EFRDs have reduced the shutdown time and/or amount of

the release (Y/N)?________

8.9.1 If ``yes'', by how much for each (estimate)?

8.9.1.1 The total shutdown time of

8.8.4 (mins.)?________

8.9.1.2 Amount of the release (bbls.)?________

8.9.2 If ``no'', why not?________________________

8.10 What was the total estimated cost of the release including--

8.10.1 Cost of repair or replacement of pipeline facility?

$________

8.10.2 Cost of product lost? $________

8.10.3 Cost attributed to loss of use of the pipeline? $________

8.10.4 Cost of damage to property other than the pipeline?

$________

8.10.5 Cost of bodily harm and/or loss of life (For analytical

purposes, loss of life is valued at $2,500,000 and significant bodily

harm reported per Section 195.50(e) is valued at $450,000)? $________

8.10.6 Cost of environmental clean-up (whether or not paid by the

operator)? $________

8.10.7 Estimated cost of damage to the environment, i.e., natural

resource damage, assessed by a court or State agency (exclusive of

clean-up cost)? $________

8.10.8 Cost of litigation? $________

8.10.9 Other costs? (specify) $________

8.10.10 Total cost? $________

8.11 How far from the release did the commodity enter a body of water

(miles)?________

8.12 Were there other areas of risk, other than a body of water,

affected by the release (Y/N):

8.12.1 Urban area?________ If yes, distance from release

(miles)________

8.12.2 Rural area in proximity to population?________ If yes,

distance from release (miles)________

8.12.3 Other (specify)________ Distance from release

(miles)________

8.13 Were there one or more public water intakes affected by the

release (Y/N)?________

8.13.1 If the answer to question 8.13 was ``yes'', how

many?________

8.13.2 For each response to 8.13.1, approximately how far was the

public water intake downstream from where the release entered the body

of water (miles)?________

8.13.3 Did high river flow due to flooding affect the release

reaching the water intake(s)(Y/N)?________

Valves Installed Per 49 CFR 195.260 (e) & (f)

As stated previously, 49 CFR 195.260 (e) and (f) require valves to

be placed on hazardous liquid pipelines in order to protect water.\11\

The March 1991 EFRD study proposed that the RSPA obtain public comment

on whether the valves at these locations should be EFRDs. The next

series of questions are posed to obtain data to make that decision.

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

\11\See footnote 10.

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

For this series of questions, operators are requested to provide

data only on hazardous liquid pipelines which have leak detection

systems installed, since the study concluded RCVs are effective only on

hazardous liquid pipelines with leak detection systems.

Question 9: How many locations have valves installed to comply with

49 CFR 195.260 (e) and (f)?________

9.1 How many locations have two block valves installed?________

9.2 How many locations have one block valve and one check valve

installed?________

9.3 How many locations have two block valves and one check valve

installed?________

9.4 Are any of the block valves RCVs (Y/N)?________

9.4.1 If ``yes'', how many locations reported in 9.1 are

RCVs?________

9.4.2 If ``yes'', how many locations reported in 9.2 are

RCVs?________

9.5 How many block valves which are not RCVs are installed to comply

with 49 CFR 195.260 (e) and (f)?________

Question 10: Estimated cost data to convert the block valves

reported in 9.4 to RCVs is requested. Report data to questions 10.1-

10.3 for each valve size (diameter) in your pipeline system:

10.1 What valve diameter does this series of questions pertain

(in.)?________

10.2 How many block valves are installed of this diameter?________

10.3 What is the total estimated cost to convert all of these valves

to RCVs? ________

10.3.1 Installation cost (material & labor) $________

10.3.2 Communication system cost $________

10.3.3 Other installation costs (specify) $________

10.3.4 Total installation cost $________

10.3.5 Annual operating cost $________

10.3.6 Annual maintenance cost $________

10.3.7 Other annual costs (specify) $________

10.3.8 Total annual costs $________

Question 11: Estimated cost data for installing new RCVs on your

pipeline system is requested. Report data to questions 11.1-11.6 for

each valve size (diameter) in your pipeline system:

11.1 What valve diameter does this series of questions pertain

(in.)?________

11.2 Cost of a manually operated block valve $________

11.3 Cost of an equivalent RCV $________

11.4 Communication cost $________

11.5 Other installation costs (specify) $________

11.6 Total installation costs $________

Question 12: What factors should the RSPA use in determining when a

manually operated valve should be converted to a RCV in order to reduce

the effects to bodies of water in case of a release?________

Locations Affected by Commodity Release

Question 13: Would a release from your pipeline affect the

following locations (answer ``yes'' or ``no'' and provide rationale for

your answer)?

13.1 Wetlands as defined in 40 CFR 230.3?12________

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

\1\240 CFR 230.3(t) states: The term wetlands means those areas

that are inundated or saturated by surface or ground water at a

frequency and duration sufficient to support, and that under normal

circumstances do support, a prevalence of vegetation typically

adapted for life in saturated soil conditions. Wetlands generally

include swamps, marshes, bogs and similar areas.

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

13.2 Critical habitat for endangered/threatened species?________

13.3 National/State parks?________

13.4 Marine sanctuaries?________

13.5 Federal wilderness areas?________

13.6 Coastal Zone Management Act designated areas?________

13.7 National monuments?________

13.8 National seashore/lakeshore recreational areas?________

13.9 National preserves?________

13.10 National wildlife refuges?________

13.11 National conservation areas? ________

13.12 Hatcheries? ________

13.13 Waterfowl management areas? ________

13.14 Public drinking water intakes? ________

13.15 Other areas (specify)? ________

Rural Areas13 Affected

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

\1\3As defined in 49 CFR 195.2: ``Rural area means outside the

limits of any incorporated of unincorporated city, town, village, or

any other designated residential or commercial area such as a

subdivision, a business or shopping center, or community

development.''

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

Question 14: How many of the following areas would be affected by a

release from your pipeline (answer 14.1-14.8 with the number of areas)?

14.1 Areas where it would take more than two hours to reach and close

a block valve once the location of a release is identified? ________

14.2 Areas of possible ground movement including areas of: known

seismic risk, slope instability, landslide, and mine subsidence?

________

14.3 Schools? ________

14.4 Hospitals? ________

14.5 Closely spaced individual dwellings (defined as areas similar to

class 2 locations as defined in 49 CFR 192.5(c)14)? ________

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

\1\449 CFR 192.5(c) states: A Class 2 location is any class

location unit that has more than 10 but less than 46 buildings

intended for human occupancy. (A ``class location unit'' is defined

in the regulations in 49 CFR 192.5(a) as an area that extends 220

yards on either side of the centerline of any continuous 1-mile

length of pipeline.)

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

14.6 Office buildings? ________

14.7 Factories or plants, such as power plants? ________

Valves in Urban Areas

The March 1991 EFRD study concluded that it was feasible from a

benefit to cost standpoint to retrofit existing manually operated block

valves on hazardous liquid pipelines located in urban areas to RCVs and

to install RCVs when installing new valves in urban areas. An urban

area is one which is not a rural area15. A proposal to require

RCVs in urban areas would apply to hazardous liquid pipelines which

have installed leak detection systems, as the Department found in the

March 1991 study that RCVs are effective only where leak detection

systems are installed.

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

\1\5See footnote #13.

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

The RSPA wants to establish a data base on manually operated block

valves located on hazardous liquid pipelines in urban areas to validate

the conclusions made in the March 1991 study.

Hazardous liquid pipeline operators are requested to respond to

Questions 15-17 for pipelines in their systems which are located in

urban areas and on which a leak detection system is installed:

Question 15: For your pipeline system, report the number of

manually operated block valves that are installed in urban areas.

Report data to questions 15.1-15.4 for each pipeline nominal diameter

located in urban areas in your pipeline system.

15.1 What nominal pipeline diameter does this series of questions

pertain (in.)? ________

15.2 How many block valves are installed of this nominal diameter?

________

15.3 For the total reported by nominal diameter, how many valves are

installed to limit release of the commodity transported? ________

15.4 For each total reported by nominal diameter, how many are

installed for pipeline maintenance purposes? ________

Question 16: For the period 1983 through 1992 (10 years), how many

failures have occurred on your pipeline system in urban areas? ________

Question 17: Provide the following information for each failure in

Question 16:

17.1 Date of the failure (MM/YY)________

17.2 What was the cause of the release (check)?

17.2.1 Corrosion? ________

17.2.2 Failed pipe or pipe seam? ________

17.2.3 Outside force damage by other than natural forces? ________

17.2.4 Outside force damage by natural forces? ________

17.2.5 Malfunction of control or relief equipment? ________

17.2.6 Operator error? ________

17.2.7 Other? (specify) ________

17.3 How much product was released (bbls.)?________

17.4 How far away from the point of release were there block or check

valves located on either side of the release (miles)

17.4.1 Upstream from the release?________

17.4.2 Downstream from the release?________

17.5 What was the total estimated cost of the release including--

17.5.1 Cost of repair or replacement of pipeline facility?

$________

17.5.2 Cost of product lost? $________

17.5.3 Cost attributed to loss of use of the pipeline? $________

17.5.4 Cost of damage to property other than the pipeline?

$________

17.5.5 Cost of bodily harm and/or loss of life ( For analytical

purposes, loss of life is valued at $2,500,000 and bodily harm reported

per Section 195.50(e) is valued at $450,000)? $________

17.5.6 Cost of environmental clean-up (whether or not paid by the

operator)? $________

17.5.7 Estimated cost of damage to the environment, i.e., natural

resource damage, assessed by a court or State agency (exclusive of

clean-up cost)? $________

17.5.8 Cost of litigation? $________

17.5.9 Other costs? (specify) $________

17.5.10 Total cost? $________

Questions for the Nonregulated Public

The preceding 17 questions relate to the gathering of pipeline

system operational data which can be answered only by hazardous liquid

pipeline operators. However, as stated earlier, the RSPA is also

soliciting comments and ideas from the nonregulated public including

State agencies, trade associations, and environmental organizations.

The following 2 questions are directed to these members of the public.

Question 18: The RSPA is attempting to determine which critical

locations should be protected from hazardous liquid pipeline releases

by the installation of EFRDs. From the locations listed below, please

provide a ranking by probability with a ranking of ``1'' representing

the location which poses the greatest probability of combined safety

and environmental risk to the public. (Questions 18.10-18.12 are left

blank for the commenter to specify locations of risk not listed in

questions 18.1-18.9.)

18.1 Locations where valves are required to be placed by 49 CFR

195.260 (e) and (f) on hazardous liquid pipelines in order to protect

water?16________

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

\1\6See footnote #10.

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

18.2 Locations where it would take more than two hours to reach and

close a block valve once the location of a release is

identified?________

18.3 Locations of possible ground movement including areas of: known

seismic risk, slope instability, landslide, and mine

subsidence?________

18.4 Schools?________

18.5 Hospitals?________

18.6 Closely spaced individual dwellings (defined as areas similar to

class 2 locations as defined in 49 CFR 192.5(c)?)17________

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

\1\7See footnote #14.

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

18.7 Shopping malls and similar locations?________

18.8 Office buildings?________

18.9 Factories or plants, such as power plants?________

18.10 Other location (specify)________

18.11 Other location (specify)________

18.12 Other location (specify)________

Question 19: From the locations listed below, please provide a

ranking of consequences from a hazardous liquid pipeline release with a

ranking of ``1'' representing the location that would result in the

greatest combined public safety and environmental consequences from a

release of hazardous liquid from a pipeline. (Questions 19.10-19.12 are

left blank for the commenter to rank the benefits for the risk

locations specify in questions 18.10-18.12.)

19.1 Locations where valves are required to be placed by 49 CFR

195.260 (e) and (f) on hazardous liquid pipelines in order to protect

water?18________

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

\1\8See footnote #10.

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

19.2 Locations where it would take more than two hours to reach and

close a block valve once the location of a release is

identified?________

19.3 Locations of possible ground movement including areas of: known

seismic risk, slope instability, landslide, and mine

subsidence?________

19.4 Schools?________

19.5 Hospitals?________

19.6 Closely spaced individual dwellings (defined as areas similar to

class 2 locations as defined in 49 CFR 192.5(c)?)19________

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

\1\9See footnote #14.

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

19.7 Shopping malls and similar locations?________

19.8 Office buildings?________

19.9 Factories or plants, such as power plants?________

19.10 Other location (specify)________

19.11 Other location (specify)________

19.12 Other location (specify)________

Issued in Washington, DC on January 12, 1994.

George W. Tenley, Jr.,

Associate Administrator for Pipeline Safety.

[FR Doc. 94-1177 Filed 1-18-94; 8:45 am]

BILLING CODE 4910-60-P