oEPA
           United States
           Environmental Protection
           Agency
Office of Solid Waste
and Emergency Response
Washington DC 20460
                                 June 1988
           Final
           Report
Review of
Emergency
Systems

Report to Congress

Section 305(b) Title
Superfund Amendments
and Reauthorization Act of
1986
                                        w

-------

-------
                              EXECUTIVE SUMMARY
     In 1984, an accidental release of methyl isocyanate from, a chemical
plant in Bhopal, India, created a cloud of toxic vapors that left over 2,000
people in the surrounding community dead and several thousand more injured.
This tragedy created worldwide concern about the potential for similar
accidents elsewhere.  In the United States, the concern intensified when, not
long afterward, a large release of aldicarb oxime occurred at a facility in
Institute, West Virginia, fortunately without loss of life.

     These incidents clearly demonstrated the potential for disastrous
accidental releases of the chemicals that have become part of modern life.
The result was a new urgency in efforts to establish national programs to
address chemical accidents.  Under the Air Toxics Strategy, the Administrator
of the U.S. Environmental Protection Agency (EPA) committed the Agency to the
prompt development of a program to foster community planning and preparation
for serious releases.  The result, EPA's Chemical Emergency Preparedness
Program (CEPP), was launched nationally in November 1985 as a voluntary
effort.  In 1986, Congress mandated many facets of CEPP in the Emergency
Planning and Community Right-to-Know Act, Title III of the Superfund
Amendments and Reauthorization Act (SARA).

     Title III creates a structure for emergency planning efforts.  The
legislation's requirements ensure that State and local governments will have
entities in place to receive information from industry and to coordinate
planning with industry.  These entities, the State Emergency Response
Commissions (SERCs) and Local Emergency Planning Committees (LEPCs), bring
together representatives of all the important elements, including
environmentalists, health organizations, and public officials.  The SERCs and
LEPCs provide a forum for a continuing dialogue on accident prevention and
community preparedness.

     Section 305(b) of Title III required EPA to conduct a "review of
emergency systems for monitoring, detecting, and preventing releases of
extremely hazardous substances at representative domestic facilities that
produce, use, or store" these substances and to report to Congress on the
findings from the review and the Agency's recommendations for initiatives to
develop or improve emergency systems.  EPA submitted an interim report in May
1987.  This final report to Congress fulfills the requirements of Section
305(b).
APPROACH

     EPA developed its approach to the review in consultation with the
States, with professional and trade associations, and with environmental
groups.  An EPA workgroup, with the participation of the Federal Emergency
Management Agency (FEMA), collected information for the review from four
sources:

     •    Literature on previous and ongoing research.

-------
                                     -11-
     •    Literature on previous and ongoing research.

     •    A questionnaire to obtain qualitative information from a wide
          range of domestic facilities.

     •    Another questionnaire on public alert systems sent to
          communities with jurisdiction over the surveyed facilities.

     •    Site visits to a limited number of the surveyed facilities.

     The review focused on 21 chemicals chosen from[the SARA Section 302(a)
list of extremely hazardous substances.  These 21 chemicals were selected to
provide a sample of chemicals representative of the list as a whole.  Only
facilities that handled one or more of the 21 chemicals in certain quantities
received questionnaires.                            ;

     After evaluating the information obtained from!these sources, EPA
convened panels of technical experts to review, preliminary findings and
provide additional information.  The Agency then presented its findings and
recommendations to the States for comment.  An early version of this report
was given a limited review by outside parties, including States, industry,
trade associations, other Federal agencies, environmentalists, and the expert
panel chairmen.
FINDINGS

     In general, EPA's findings are as follows:

     Prevention of accidental releases requires a holistic approach that
     integrates technologies, procedures, and management practices.

     Prevention of chemical accidents requires a comprehensive, integrated
approach that takes into account the hazards of the chemicals involved, the
hazards of the process, the capabilities of the facility personnel, and the
potential impact on the community.  Sustaining a comprehensive approach
depends on management's commitment to install, maintain, and update
appropriate technologies, as well as to provide training for the operators.
Active company involvement with the local community and with industry and
professional groups also reflects management's commitment to safety.
                                                    i
     "State of the art" for technologies and techniques to monitor,
     detect, and prevent accidental releases cannot be defined
     generlcally.

     The review indicates that no single technology;is most effective in
every situation.  Site-, process-, and chemical-specific factors dictate the
choice of technology or technique as do the operating procedures and
management practices employed at a facility.  For public alert systems, the
effectiveness of a technology depends on the size and characteristics of the
population and on the management practices.  There is, therefore, no "best"
method.

-------
                                    -111-
     The larger chemical producers appear more aware of potential
     hazards and of methods to prevent releases.

     In general, the larger chemical producers appear to have a better
awareness of hazards, and use more methods to address them than do smaller
producers and chemical storage facilities, distributors, and repackagers.
The larger companies are also more actively involved in industry and
professional groups and in community programs.  This finding indicates the
need to enhance the use of existing mechanisms, such as trade associations,
professional groups, and the SERCs and LEPCs, to disseminate to smaller
companies information on available technologies and techniques.

     Some technologies and techniques need further research and
     development before they will be technically feasible and cost-
     effective.

     The review identified a number of technologies that require research and
development, as well as additional data needs:  (1)  Inexpensive, reliable,
chemical-specific detectors for combustible gases, solids, and corrosives;
(2)  Laser-based, remote sensing systems;  (3)  Mitigation technologies for
liquid, liquid/vapor, and solid releases; and (4) Data on source
characterization, human health effects, equipment failure rates, and human
error rates.

     The review indicated that few facilities have installed perimeter alert
systems.   Those facilities that have investigated such systems reported that
more reliable and inexpensive perimeter monitors need to be developed for the
systems to be technically feasible and cost-effective.

     Improved communications are needed in most phases of the public
     alert systems.

     The review indicated that substantial improvements in public alert could
be obtained for little cost by establishing effective decision-making and
communications procedures between facilities and communities, and between
officials and the community.  The LEPGs established under Title III should
contribute greatly to this end.
RECOMMENDATIONS

     Industry, Federal, State, and local authorities all have roles to play
in preventing the release of hazardous substances.  Because chemical
facilities are complex and require site-specific safety assessment and
contingency planning, there should be close collaboration between industry
and the community.  Title III mechanisms such as the LEPCs should enhance
prevention by increasing public awareness, public participation, and public
expectations for safe operations.  The LEPCs should also play an important
role in improving local emergency response capabilities.  The States, through
Title Ill's SERCs and State agencies, have an active role in overseeing
facilities.

-------
                                     -iv-          ;

     Industry must assume the primary responsibility for preventing accidents
and ensuring the safety of its workers and the public health of the
surrounding community.  Industry, with the assistance of professional and
trade associations, should also take the lead in conducting research on
prevention technologies and in disseminating information.  The Federal
government should act as a catalyst, identifying problem areas and providing
technical assistance when needed.

     Specifically, EPA recommends:        ""'"'!"

     •    Continuing the expert panels convened "to review this study and
          expanding them to include professional and erivironmeTital
          groups.  These panels will provide advice to EPA in a number
          of technical areas.
          Strengthening the LEPCs and SERCs by
          assistance and guidance.
providing technical
          Conducting further studies on the causes
          and ways to prevent them.
    of chemical accidents
     •    Encouraging industry and professional and trade associations
          to conduct research that will develop and refine cost-
          effective, reliable prevention, detectionj monitoring,'and
          public alert methods,  procedures, and devices.. Appropriate
          means for transferring the results- of the research should also
          be developed.

     •    Developing guidance, where needed, on such topics as the use
          of in-place sheltering, hazard'assessment techniques,
          successful management techniques, and prevention methods.

     •    Working with other agencies to develop training to enhance
          Federal and State staff expertise in conducting chemical
          process safety audits.     '          '    \-
                                                   i
     •    Working with international organizationsjto draw upon their
          expertise.                               [•    -  • < •'   •''• "

     •    Encouraging the study of management techniques that can be
          applied to the management of risk and disseminating the
          results to the industry, especially to smaller facilities.

     •    Encouraging the development of standard procedures and
          protocols for public alert decision-making and information
          collection to expedite the notification of the public in a
          chemical emergency.

     Besides the general recommendations that apply across the several areas
considered in this review, EPA also recommends the :following:

-------
                                -v-

     In management:                                    ;

     Management should be trained in hazard evaluation and accident
     prevention.                                     •.•-...   \ •

     In prevention:     ..-'.-         -            •.-••:...-:'

     Emphasis should be placed on pre-release prevention rather
     than on post-release measures.      .

     The terminology used in hazard evaluation should be  -
     standardized.

     Mitigation technologies for large-scale releases should be
     evaluated.
•    In monitoring and detection:

     The emphasis should be on developing systems that are
     relatively inexpensive and that do not require frequent
     calibration or trained technicians to operate.

     The application and interpretation of air dispersion models in
     real-time situations should be done with great caution, and
     then only by people aware of the model's limitations and
     assumptions.

     Remote sensing systems, such as chemical-specific laser
     systems, should,be developed for more general use.

•    In public alert:

     Warning message protocols for English and non-English speaking
     populations should be developed at the local level.

     Notification and alert equipment must be maintained and tested
     regularly.

     Public warning.technologies in high-risk and densely populated
     areas should.be improved.       :  ... . .

     Studies of public response  to warnings in chemical emergencies
     should be conducted to improve warning systems.      .

Each of these recommendations is elaborated in the following chapters.

-------

-------
                                  CONTENTS
EXECUTIVE SUMMARY




1.   INTRODUCTION




     1.1  SARA Title III:  Background	    1




     1.2  The Statutory Requirement for a Review of Emergency Systems .  .    1




     1.3  Organization of this Report	    2




2.   APPROACH




     2.1  Scope	    4




     2.2  Methodology of the Review	    5




     2.3  Research and Literature Review  	    6




     2.4  Selection of Chemicals	    6




     2.5  Facility Selection  	    7




     2.6  Facility and Community Surveys  	    9




     2.7  Site Visits	10




     2.8  Analysis	10




     2.9  Limitations of the Data	10




3.  GENERAL FINDINGS AND RECOMMENDATIONS




     3.1  Findings	12




     3.2  Recommendations	   14




4.  MANAGEMENT




     4.1  Management Practices   	   16




          4.1.1  Key Elements in Management Programs	16




          4.1.2  Safety Management Organization  	   18




          4.1.3  Training    	18

-------
          4.1.4  Maintenance	  19

          4.1.5  Accident Investigation	j.  .  .  .	19
                                                    i
          4.1.6  Interactions with Outside Organizations   	  ,  .  19

     4.2  Development and Dissemination of Management Techniques  ....  19

     4.3  Recommendations	  .  .  .  ;  .  .  .  . .  .  .  .  .  .  21

5.  PREVENTION SYSTEMS      ,

     5.1  Hazard Evaluation	22

     5.2  Pre-Release Technologies  ...................  24

          5.2.1  Siting and Facility Design .....  	  26

          5.2.2  Process Design	26

          5.2.3  Process Control and Monitoring  .  . ;	27

          5.2.4  Protection	27
                                                    i ..
     5.3  Post-Release Mitigation	•.'1.  .  .  .  .......  28
                                                    i

     5.4  Recommendations 	  ....;........  29

6.  MONITORING AND DETECTION SYSTEMS                ' -•'   "                     '.

     6.1  Detection Technologies                '   • >

          6.1.1  Observation by Operator  .......... 	  30

          6.1.2  Multi-Detector Systems . .  .  ... |. *..........  31
                                                    i
          6.1.3  Fixed Detector/Multi-Sampling Point; Systems   	   32

          6.1.4  Remote Sensing Systems ............  	  33

     6.2  Perimeter Monitors	  33

     6.3  Field Monitoring Systems              ,   ;    .

          6.3.1  Portable Monitoring Instruments  .............  35

          6.3.2  Field Laboratory Instruments  . ........ 	  35

     6.4  Dispersion Modeling 	 ................  36

     6.5  Recommendations	'..-...- i  ......  37

-------
    PUBLIC ALERT SYSTEMS




     7.1  Technologies  .  .  .  . .  . .  .  .  .  .  ... .




          7.1.1  Communications Technologies  . .   . .




          7,1.2  Alert Technologies .......   . .




          7.1.3  Notification Technologies  .  . .   . .




          7.1.4  Alert and Notification Technologies




          7.1.5  Warning Systems in Use .  .  .  ". .   . .




     7.2  Facility/Community Interactions ,  ....   . • ;




     7.3  System Effectiveness  .  . .  .<-• .  •.  .  . .   . .




     7.4  Recommendations	   . .
                                                              38




                                                              40




                                                              40




                                                              41




                                                              41




                                                              42




                                                              43




                                                              45




                                                              46
Appendices




Appendix 1 -•




Appendix 2 -•




Appendix 3 -•




Appendix 4 -•




Appendix .5 - •




Appendix 6 - •




Appendix 7 - •




Appendix 8 - •




Appendix 9 - •




Appendix 10




Appendix 11  •




Appendix 12




Appendix 13




Appendix 14
 Glossary of Terms/Acronyms       '        ,




 Survey Responses




 Hazard Evaluation Procedures          ,  .




 Survey Responses on Prevention




 Prevention and Protection Measures




 Monitoring and Detection




 Public Alert;        .          .     •  -.      ,




 Expert Panel  Summary          ,




•„ Facility Site Visits                    ;




•- EPA's Accidental Release Information Program




•- International Activities   .      • .-




•- Facility Questionnaire     -




•- Community Questionnaire




•- Bibliography             .             • ••-•

-------

-------
                              1.  INTRODUCTION
     This report presents the findings and recommendations of the U.S.
Environmental Protection Agency's (EPA) review of emergency systems for
monitoring, detecting, and preventing accidental releases of extremely
hazardous substances to the environment, and of systems for alerting the
public to such releases.  EPA is submitting this report to Congress in
fulfillment of Section 305(b) of Title III of the Superfund Amendments and
Reauthorization Act of 1986 (SARA) .
1.1  SARA. TITLE III:  BACKGROUND

     In 1986, Congress passed Title III of SARA, called the "Emergency
Planning and Community Right-to-Know Act."  This legislation requires
industry to share information on hazardous chemicals and releases and
requires States and local communities to prepare to deal with releases.   One
of the driving forces behind Title III was a heightened awareness of the
potential for chemical releases, caused by a succession of major accidents at
chemical facilities.  The 1984 accidental release of methyl isocyanate from a
chemical plant in Bhopal, India, created a cloud of toxic vapors that left
over 2,000 people in the surrounding community dead.  This tragedy created
worldwide concern about the potential for similar accidents elsewhere.  In
the United States, the concern was intensified when, not long afterwards, a
large release of aldicarb oxime occurred at a chemical facility in Institute,
West Virginia.  Although no lives were lost at Institute, the facility had no
way of automatically notifying the community of the release and the
notification given was not timely.  These incidents clearly demonstrated the
potential for disastrous accidental releases of the chemicals that have
become a part of modern life; they also made the public aware that releases
of chemicals from the facilities where they are manufactured, processed,
used, or stored not only can occur, but do occur, even in this country.

     The result was a new urgency in efforts to establish national programs
to address chemical emergencies.  Under the Air Toxics Strategy, the
Administrator of EPA committed the Agency to the prompt development of a
program to foster community planning and preparation for serious releases of
extremely hazardous substances from local chemical facilities.  The result,
EPA's Chemical Emergency Preparedness Program  (CEPP), was launched nationally
in November 1985 as a voluntary program.  Congress incorporated many of the
facets of the CEPP  in Title III.  The requirements of Title III ensure that
State and local governments will have mechanisms in place to receive
information from and to coordinate emergency planning with industry.


1.2  THE STATUTORY  REQUIREMENT FOR A REVIEW OF EMERGENCY SYSTEMS

     Section  305(b) of SARA includes the requirement that EPA conduct a
"review of emergency systems for monitoring, detecting, and preventing
releases of  extremely hazardous substances at  representative domestic
facilities that produce, use, or store extremely hazardous substances."  EPA
was  required to report its interim findings in May 1987  and to include
recommendations  in  a final report.  EPA submitted its Interim Report  in May
1987;  this document is the final report required by Section 305(b).

-------
                                      -2-

     Section 305(b) specifies that the report based on the review shall
include findings not only on the status of current technological capabilities
to monitor, detect, and prevent releases, but also on the status of devices,
systems, and procedures for, providing timely and effective public warning of
an accidental release.  In addition, the report must address the technical'
and economic feasibility of perimeter monitoring systems for detecting
releases from facilities.  Recommendations are required on (1) improving
devices, systems, and procedures for alerting the public in the event of an
accidental release, and (2) initiatives to support the development of new or
improved technologies or systems to facilitate the monitoring, detection, and
prevention of releases.

     The requirement for this review suggests that the public will be alerted
more promptly and effectively when a release occurs if adequate public alert
devices, systems, and procedures are in place.  It also acknowledges that the
prevention, monitoring, and detection of accidental releases must rely not
only on the technologies in place at facilities that produce, use, or store
extremely hazardous substances, but also on management support and effective
management practices.  By providing .information about current technological
capabilities at domestic facilities for preventing releases and alerting the
public, this review serves as an important adjunct to the emergency planning
and community right-to-know provisions of Title III.
1.3  ORGANIZATION OF THIS REPORT

     Following this introductory chapter, this report contains six chapters.

     «    Chapter 2:  Approach is a discussion of EPA's approach to
          conducting the Section 305(b) review.  It describes the scope
          of the review and discusses the methods EPA used to select
          chemicals and facilities on which to focus the study.

     •    Chapter 3: General Findings and Recommendations discusses the
          general findings of this review and provides an overview of
          the principal recommendations.   Recommendations on specific
          issues are presented in subsequent chapters.

     •    Chapter 4: Management discusses the importance of management
          in preventing of releases and outlines the, components of a
          management system.  The management practices discussed support
          the specific technologies and procedures covered in the
          remaining chapters.

     •    Chapter 5:  Prevention Systems  provides a discussion of hazard
          evaluation, prevention/protection technologies, and post-
          release mitigation.

     •    Chapter 6:  Monitoring and Detection Systems discusses the
          findings on the technologies available to monitor and detect
          releases within the facility and at its perimeter as well as
          models to predict dispersion of released substances in the
          air.

-------
                                     -3-

     B    Chapter 7:   Public Alert Systems reviews technologies and
          procedures in use to alert the public to emergencies.            ' ,

     These .chapters present the technologies,  techniques, and practices
related to monitoring, detection, and prevention o:f accidental releases of
extremely hazardous substances, and to alerting the public should a release
occur.  More detailed technical information is included in the appendices to
this report, which begin with a glossary in which the technical terms and
acronyms used in this report are, defined.  Additional technical background
materials are available in the EPA reference library.                       .

-------
                                      -4-

                                 2.  APPROACH


      This  chapter explains  EPA's approach  to the  Section 305(b) review of
 emergency  systems.   It begins with a  discussion of  the scope of the review,
 then describes  the methods  used to collect and analyze the data.  The Interim
 Report to  Congress under  Section 305(b) provides  greater detail on certain
 aspects of the  approach.


 2.1  SCOPE

      EPA and the  Federal  Emergency Management Agency  (FEMA), which took the  '
 lead responsibility for the review of public alert  systems, defined the scope
 of  the review in  a manner intended to support sound recommendations on the
 areas specified by Congress in  Section 305(b).

      Section 305(b)  required EPA to review systems  for monitoring, detecting,
 and preventing  releases of  extremely  hazardous substances.  "Extremely
 hazardous  substances"  are the specific chemicals  on the list referred to in
 SARA Section 302(a), which  was  published in the Federal Register on November
 17,  1986 (51 FR 41570), and revised on April 22,  1987 (52 FR 13378) and on
 February 25,  1988 (53  FR  5574).

      In this report, the  phrase "detection system"  refers primarily to
 technologies for  detecting  leaks or catastrophic  releases within process and
 storage areas.  Basically the same  technologies are used for detecting
 releases at  the facility  fenceline  or perimeter.  The phrase "monitoring
 system" refers  to methodologies and instrumentation used to .perform sampling
 and analyses in a community potentially affected by a release to determine
 whether, and to what extent, the released  chemical  is present in the
 environment.

      The phrase "prevention systems"  refers broadly:to any technology or
 management practice  that  aids in preventing releases of extremely hazardous  '
 substances.   Prevention systems include not only  technologies and procedures
 that prevent the  loss  of  a  chemical from containment in the system in which
 it  is manufactured,  processed,  used,  or stored, but;also any technology or
 procedure  that  can be  used  to mitigate a release.  Accordingly, the
 "prevention  systems" EPA  reviewed  include hazard evaluation techniques,
 monitoring,  control, and  back-up systems in chemical processes, and
 techniques for  minimizing the dispersion or spread of a release beyond the
 release point.

      The statute  also  required  a review of the status of public alert devices
 or  systems, which include any automatic technology or communications link to
 warn adjacent communities of a  significant release.  Because the way in which
 a public alert  technology is used can be at least as important as the
 technology itself, EPA and  FEMA also  considered the procedures being used for
 communications  between the  facilities and local officials and between those
 officials  and the public.

      Section 305(b)  stated  that EPA "may select representative extremely
hazardous  substances from the substances on the list referred to in Section
 302(a) for the  purposes of  this review."  EPA chose to focus the study on 21

-------
                                     -5-

listed chemicals.  Section 2.4 of this chapter describes the method EPA used
to select the chemicals.

     EPA focused this review on sudden, accidental releases -- the types of
releases that can be considered emergencies -- rather than on routine
emissions.  Similarly, the review focused on releases that involve
significant quantities of acutely toxic chemicals, rather than chronically
toxic chemicals.  Chemicals are listed as "extremely hazardous substances"
under Section 302(a) because of their acute toxicity, that is, their ability
to cause significant adverse effects on human health with brief, one-time
exposure.  The emergency systems on which this review focused are those able
to monitor, detect, or prevent releases that could cause effects from short-
term exposures, rather than those systems intended primarily for routine or
low-level releases that would endanger human health and the environment only
over an extended period.

     EPA concentrated the review on facilities that produce, use, or store
the 21 extremely hazardous substances that were selected.  Only facilities
believed to be handling one or more of these 21 substances in certain
quantities were included in the surveys and site visits.

     The review considered the prevention of releases to any environmental
medium -- air,, land, surface water, ground water  -- but emphasized releases
to air because they present the greatest potential for immediate and
widespread harm to human health and the environment.

     One final aspect of the scope defined for this review deserves emphasis.
EPA considered management practices and operating procedures to be integral
components of  the emergency systems and technological capabilities the review
was required to assess.  The commitment of corporate and facility managers to
accident prevention is, in fact, the underpinning of effective accident
prevention.  The installation of the most advanced technologies is an
ineffectual safeguard against accidents unless those technologies are
properly operated and maintained in a workplace where safety is explicitly
and implicitly valued.  Accordingly, this review  addressed management
practices and  procedures.
 2.2  METHODOLOGY OF THE REVIEW

     EPA's  approach to this  review involved the  following  steps:

     •    A review of previous research,  studies,  documentation,  and
          ongoing  research to  identify available technologies,  systems,
          procedures, and methodologies;

     •    The selection of 21  chemicals on which to focus  the review;

     •    The selection of a sample of domestic  facilities believed to
          be handling the 21 substances at levels above the Section
          302(a) threshold planning quantity (TPQ).  The purpose  was  to
          collect  data to determine which of the available technologies
          are being used;

-------
                                      -6-

      •    A survey of the sample of facilities,  using a written
           questionnaire;

      •    A parallel survey of the  communities  surrounding the  selected
           facilities,  again using a written questionnaire;

      •    Site visits  and interviews at a limited number of the surveyed
           facilities to supplement  the information from the
           questionnaires;

      •    An analysis  of  the data obtained through the preceding steps
           to generate  findings;               .      ;                     .    .

      •    A review of  the findings  by a panel of technical  experts; and

      •    Consultations with selected States  on  the findings and
           recommendations.

      Throughout the study the Agency consulted with States,  industry,  ' .  •.
professional organizations,  trade associations,  and environmental groups.  ;•.,
(The  States  that participated in the review are  California, Delaware,
Illinois,  Maryland,  New Jersey,  Ohio, Tennessee,  and West Virginia.)  A
public meeting on  the  Agency's proposed approach for the study  as well as on
the interim  findings was held on April 14,  1987.   The first versions of the
questionnaires were circulated to both the  States  and professional groups for
comment.   The facility questionnaire was  then pretested at  a few facilities
suggested  by the Chlorine  Institute before being revised and sent to the full
sample of  facilities.   Once  the  initial analysis  of the data was complete,
EPA convened expert panels to review the  findings before again meeting with
representative States  to discuss  the findings and proposed  recommendations.
Drafts of  the report were  circulated to certain  States and  key professional
groups for their comments, as well  as to  the  chairmen of,the expert panels.  <


2.3  RESEARCH AND LITERATORE REVIEW

     EPA conducted  a literature search to identify  available technologies and
techniques.   Included were reports  on ongoing prevention projects, industry
research,  symposia,  and past studies on the full range of emergency systems.

     To conduct  the  literature search, a number of  engineering data bases
were searched to provide coverage of the world's significant engineering and
technological  literature.  Journals, publications of engineering societies
and organizations,   technical reports, monographs, and publications from
conferences were searched for relevant citations.  Books as well as
manufacturers' literature on specific technologies Were also reviewed.   The
literature search provided a basis for defining what technologies are
available and for identifying areas of ongoing research.  A bibliography of
the literature reviewed is included as Appendix 14 to this report.
2.4  SELECTION OF CHEMICALS

     Congress stated that EPA could select "representative" chemicals from
the Section 302(a) list of extremely hazardous substances for the purpose of

-------
                                     -7-

this review.  As noted, EPA selected 21 chemicals.'  These chemicals are
listed in Exhibit 2-1.  The selection process, which was described in detail
in the Interim Report, was designed to yield a sample of chemicals
representative of a broad range of hazards common to the chemical industry.
Accordingly, the monitoring, detection, prevention, and public alert systems
used by the facilities handling these chemicals should be representative of
the emergency systems used by many facilities that handle extremely hazardous
substances.

     EPA chose nine chemicals because of their large production volumes,
widely acknowledged potential hazards, involvement in past plant and
transportation accidents, and generally recognized special handling
procedures and controls.  These chemicals -- ammonia, chlorine, hydrocyanic
acid, hydrogen fluoride, hydrogen sulfide, methyl isocyanate, phosgene,
sulfur dioxide, and sulfur trioxide -- provide a wide range of reactivity,
flammability, and corrosivity hazards.  The remaining 12 chemicals were
selected from the list of extremely hazardous substances, using a random
number generation technique within certain guidelines.  Specifically, the
selection process was designed to yield a mix of physical and chemical
characteristics, as indicated in Exhibit 2-1.  EPA initially selected 20
chemicals.  However, when the Agency could'not identify a significant number
of facilities that handled the two liquids with medium vapor pressure -- •
hydrazine and tetraethyltin -- the Agency selected a twenty-first chemical
with similar characteristics, benzotrichloride, to increase the number of
potential respondents in that category.
2.5  FACILITY SELECTION

     In consultation with  State agencies, trade and professional
associations, and public interest  groups, EPA  identified data bases listing
facilities  that were believed  to manufacture,  process, use, or store the 21
selected chemicals.  Published reference  sources  for the chemical industry
were also collected. The facilities  identified from these sources served as
the frame from which the sample of facilities  to  survey was selected;
Several conditions  guided  the  selection of  the facilities to survey:

     •    An individual facility could be surveyed for only one
          chemical".

     •    Up to two facilities of  a  company handling the same chemical
          could be  surveyed, but only if  there were known to be sig-
          nificant  differences between the  two with respect to age, type
          of process,  or quantity  of the  chemical handled.

     •    When one  facility had to be selected from several handling a
          chemical  at  levels above the TPQ, the decision was based on
          which facility seemed most important to include for the
          purposes  of  the  survey.  The factors considered included the
          number  of other  facilities handling  the chemical, the need to
          include major uses and major production processes for each
          chemical, and the need to  include diversity  in the sample with
          respect  to location, age,  and size.

-------
                                      -8-
                                 EXHIBIT 2-1

             SAMPLE OF EXTREMELY HAZARDOUS  SUBSTANCES FOR REVIEW
   Chemical Name
                              Form
 Vapor
Pressure^
 TPQ
 (Ibs)2
Acrylonitrile
Ammonia*
Benzenearsonic acid
Benzotrichloride
Chlorine*
Chloroacetic acid
Furan
Hydrocyanic acid*
Hydrazine
Hydrogen fluoride*
Hydrogen sulfide*
Mechlorethamine
Methiocarb
Methyl bromide
Methyl isocyanate*
Phosgene*
Sodium Azide
Sulfur dioxide*
Sulfur trioxide*
Tetraethyltin
Trichloroacetyl Chloride
                              Liquid
                              Gas
                              Solid
                              Liquid
                              Gas
                              Solid
                              Liquid
                              Gas
                              Liquid
                              Gas
                              Gas
                              Liquid
                              Solid
                              Gas
                              Liquid
                              Gas
                              Solid
                              Gas
                              Solid
                              Liquid
                              Liquid
 High
 High
 Low
 Medium
 High
 Low
 High
 High
 Medium
 High
 High
 Low
 Low
 High
 High
 High
 Low
 High
 High
 Medium
 Low
10000
  500
  . 10
  100
  100
  100
  500
  100
 1000
  100
  500
   10
  500
 1000
  500
   10
  500
  500
  100
  100
 , 500
1.   Vapor pressure ranges:
                         High   =  > 100 mm of mercury
                         Medium =  1 < vp < 100 mm of mercury
                         Low    =  < 1 mm of mercury
2.
Threshold planning quantity (TPQ), the quantity of an extremely
hazardous substance present at a facility that will trigger the
planning requirements of SARA Section 302.  The TPQs listed above
for solids apply when the substance is molten, a fine powder, or in
solution; otherwise the TPQ for solids is 10,000 Ibs.

Denotes one of the nine chemicals selected because of high production
volumes, recognized hazards, involvement in past accidents, and
recognized need for special handling procedures.

-------
                                     -9-

     When fewer than 100 facilities were identified as handling a chemical,
all facilities were surveyed unless they were disqualified by the above
conditions.  More than 100 facilities were identified for only four
chemicals: ammonia, chlorine, hydrogen sulfide, and sulfur dioxide.
Additional considerations -- use of the chemical, size of the chemical
operations, age of the facility, release history, location, and the
reliability of the data bases through which the facilities were identified -
were used to select a sample of under 100 facilities for each of these four
chemicals.
2.6  FACILITY AND COMMUNITY SURVEYS

     In the fall of 1987, EPA mailed questionnaires to 522 facilities
selected as described above.  The questionnaire, which is included in this
report as Appendix 12, requested detailed information on:

     •    Major unit processes and operations at the facility,
          including:

               Potential process and chemical hazards.

               An engineering drawing of the process highlighting the
               potential release points.

               A description of relevant equipment and personnel.

               An inventory of the quantities of extremely hazardous
               substances used at the facility.

               A description of equipment to prevent, detect, and
               monitor releases.

     •    The types of procedures used to prevent, detect, and mitigate
          accidental releases and any potential hazards identified,
          including techniques available to model the dispersion of
          released substances, and spill control techniques and
          technologies.

     •    Relevant management practices, including reporting practices,
          hazard evaluation programs, plans for upgrading prevention
          technologies and procedures, management organization relating
          to safety and to accident prevention, and operator training
          programs.

     •    Public alert equipment and technologies as well as
          administrative practices, concentrating on notification of the
          local community and on activities (such as emergency
          simulation exercises) conducted with the community.

     Separate questionnaires developed in conjunction with FEMA were mailed
to 277 communities that have jurisdiction over one or more of the selected
facilities.  Because the Local Emergency Planning Committees (LEPCs)
established under Title III were just being formed, the surveys were mailed
to the head of the local agency responsible for emergency planning, with the

-------
                                     -10-

instruction that the questionnaire should be completed by the.person in
charge of emergency planning with the facility.  The community questionnaire,
which is included as Appendix 13 of this report, requested information on:
                                                                            i
     •    Community emergency planning organization and staffing,
          designated contact points between the facility and the
          community, and primary and back-up communication equipment
          used by the facility to notify the community.

     •    Procedures used in an emergency, communication equipment used
          to notify the public, information needed;from the facility,
          and time needed to make decisions in an emergency situation.

     •    Specific land use and populations (residential, industrial,
          commercial, institutional) within one and five miles of the
          facility.                                '

     •    Coordination with other emergency response organizations.

     Approximately one-third of the facilities producing, handling,  or
storing the 21 chemicals above the TPQ, and half,of the communities returned
completed questionnaires.  For details on the respondents, see Appendix 2.


2.7  SITE VISITS

     To confirm and clarify the survey information, EPA conducted seven site
visits to facilities that produce, use, or handle phosgene, chlorine,
acrylonitrile, sulfur dioxide, sulfur trioxide, and hydrogen fluoride.
Trained representatives from EPA and States verified the information on the
questionnaire and discussed management approaches in detail with company
staff.  The site visits provided an opportunity to obtain in-depth
information to augment the data from the survey.  The reports from the site
visits are summarized in Appendix 9.


2.8  ANALYSIS                                      ;

     After the survey questionnaires were re turned j EPA and FEMA evaluation ;
teams analyzed the data from the surveys as well as the information gained
from the literature search.  The evaluation teams were composed of EPA and
FEMA staff.  The evaluation teams focused on certain questions:  what
technologies and procedures are available, which are being used, and why
facilities and communities made the choices they have.  EPA then convened
five panels of experts to review the findings and to provide supplemental
information where possible.  The five panels covered hazard assessment and
prevention, monitoring and detection, mitigation,  management, and public
alert.  The panels included prominent representatives from industry, State
and other government agencies, and academia.


2.9  LIMITATIONS OF THE DATA

     EPA designed the surveys to provide a broad "snapshot" of current
practices and technologies in use.  The survey samples Were not designed to


-------
                                     -11-

yield statistically significant results.  Consequently, the results are not
to be interpreted as statistically significant.  The findings from the
questionnaires reflect conditions at the facilities and communities that
returned questionnaires, not necessarily conditions at all facilities in the
original sample or in the domestic industry as a whole.

     The holistic nature of prevention and response systems creates
difficulties in interpreting the survey findings.  A single fact, or a set of
facts,  such as the presence and use of particular prevention technologies,
does not provide sufficient information to judge the degree of prevention
provided at a particular facility.  Conversely, the absence of certain
technologies may not mean that the facility is doing less than it should to
prevent releases, because, for any facility, many alternative ways to prevent
releases may be available.  The absence of certain critical practices, such
as preventive maintenance, however, can be taken as an indication of
insufficient emphasis on accident prevention.

     While the information collected in this review must be interpreted-with
an awareness of its limitations, EPA believes the data provide insight into
current accident prevention practices.  In particular, the data indicate
areas of possible concern and areas where additional information, research,
guidance, and study would be useful.

-------
                                     -12-

                   3.  GENERAL FINDINGS AND RECOMMENDATIONS


     This chapter discusses  the overall  findings and recommendations of the
review.  Findings and recommendations  that relate to specific types of
emergency response systems are discussed in subsequent chapters.


3.1  FINDINGS

     Prevention of accidental releases requires a holistic approach that
     integrates technologies, procedures, and management practices.

     Prevention of accidental releases requires a comprehensive, integrated
approach that takes  into account the hazards of the chemicals involved, the
hazards of the  process, the  capabilities of the facility personnel, and the
potential impact on  the community.  Prevention should be considered at every
stage of development and operation --  siting, layout, design, redesign,
retrofit, and shutdown -- and prevention measures should be built into the
facility.   In  general, it is more cost-effective to prevent accidents than
to attempt to mitigate releases when they occur.  For some releases,  !
especially unconfined vapor  releases,  mitigation may not be technically
feasible.

     A comprehensive approach to safety  is dependent on management's
commitment to the safe operation of the  facility.  This commitment includes
the willingness to install,  maintain,  constantly review, and update the
appropriate technologies where needed.  The development of training and
refresher training programs  as. well as maintenance programs is essential to
ensure that the facility is  operated safely.  Management commitment is also
reflected in extensive company interactions with the local communities and
with industry and professional groups.

     "State of  the art" for  the technologies and techniques used to
     monitor, detect, and prevent accidental releases cannot be defined
     generically.

     Overall, the Section 305(b) review  indicated that there is no single
method or technology that works best in every situation.  In each area
considered in this review, the determination of what constitutes a "state-of-
the-art" technology  for a particular facility depends on the individual
circumstances of the facility -- its location and layout,  its process, the
chemicals handled, and the hazards associated with the specific chemicals and
processes.  For example, within process design, the number of possible
combinations of equipment is so great  that only an overall examination of the
design in a particular setting can reveal whether the design is appropriate..

     This finding also applies to public alert technologies.   While the
number of technologies is limited, the appropriateness of any particular
system will depend on the characteristics of the community;  for example,  the
notification system  needed by a small  community at spme distance from a
facility will be very different from the system needed in urban areas.

     Therefore, attempting to specify  a "state-of-the-art" technology or
technique or management approach for all facilities is neither feasible nor

-------
                                     -13-

desirable.  Each facility must be considered individually, and appropriate
technologies must be determined in conjunction with operating procedures and
management practices.  In addition, labelling a particular technique or
technology as "state-of-the-art" may lead facilities to use it, perhaps in
inappropriate circumstances, and make no efforts to move beyond it.

     The larger chemical producers appear more aware of potential
     hazards and of methods to prevent releases.

     In1 general, the larger chemical producers appear.to have a better
awareness,' of hazards . and risk management and implement more release
prevention techniques and technologies than do the smaller chemical
producers, users, handlers, and distributors.  This does not mean that some
smaller•companies do not have adequate systems for preventing releases, but
rather that they appear less aware of the potential for,significant releases
and the associated prevention methodologies available.   It is generally
accepted that the smaller companies may not have adequate resources to
install the latest prevention technologies; some may consider their
inventories insufficient to warrant using the technologies.  Many of these
companies'appear to lack an understanding of formal hazard evaluation
techniques.  The management commitment of the smaller companies, as
demonstrated by participation in industry and community programs,  also
appears to be lower than that of the larger producers.

     Some of the technologies and techniques need further research and
     development before they will be technically feasible and cost-
     effective.  '

     The review identified several areas where technologies either do not
exist, are unproven for large-scale use, or are still prohibitively expensive
for most facilities.  For example, it is clear that release detection systems
do not exist for some chemicals;  of those that do exist, many are not viewed
as reliable or cost-effective.  Some of the data required for air dispersion
modeling and probabilistic risk assessments need to be improved.  As
explained in subsequent chapters, additional research is'needed in these
areas.                                          .        •

     Congress specifically directed EPA to consider the technical and
economic feasibility of perimeter monitoring systems.  The review indicated
such systems will require additional development before they will be cost-
effective and sufficiently reliable to warrant widespread use.

     Improved communications are needed in most phases of the public
     alert process.

     The review indicated that substantial improvements in public alert could
be obtained for little cost, by establishing effective decision-making and
communications procedures between facilities and public officials, and
between officials and the community.  The planning entities established by
Title III, notably the LEPCs, should go far toward solving this problem by
fostering dialogue between industry and communities.  While sophisticated
technologies for warning do exist, unless they are easy to install, maintain,
and finance, their use would not necessarily improve the public alert
process.

-------
                                     -14-
3.2  RECOMMENDATIONS
     Industry, Federal, State, and local authorities all have a role to play
to prevent accidental releases of hazardous chemicals.   Because chemical
plants are complex and require site-specific safety assessment and
contingency planning, there needs to be a .close collaboration between
industry and the community with an understanding that the common goal is the
prevention of chemical accidents;  •    •           j .

     Industry should take the primary responsibility for preventing accidents
and ensuring the safety of its workers and the public health of the
surrounding communities. , Local and State governments are responsible for
public protection and oversight of the process of preparedness, prevention,
and mitigation.                                                 '
                                               • -  i
     The Federal government should act as a catalyst, identifying problems,
and providing guidance and technical assistance to State'and local
governments and to industry.  Assistance to industry.should include (1)
collecting information on hazardous substances and their effects, (2)
informing manufacturers of successful emergency response techniques, and (3)
encouraging industry, trade and professional organizations, and,States and
localities to develop standards and guidelines.        ,     ; .

     Therefore, all levels of government, industry, professional and trade
organizations, and environmental groups must work closely together to:

     •    Develop and share information on the causes of accidents, on
          hazard evaluation techniques, and on prevention methods;

     •    Identify needs and sponsor research to improve the application
          of hazard evaluation and technologies for monitoring,
          detecting, and preventing releases of hazardous chemicals;
                                                  i         .
     •    Establish a process for exchanging and sharing information
          nationally and internationally in these 'areas with the
          industrial community, with an emphasis on smaller chemical
          companies, users, and distributors, as well as with the
          general public;' and                            ..

     •    Conduct further studies in specific areas such as the
          effectiveness of various public warning procedures.

     To this end, EPA commits to:                 \

     m    Continue the dialogue with industry, States,  professional and
          trade organizations, and environmentalists through the
          expansion and continuation of-the technical panels convened
          for the Section 305(b) study.  These panels will deal with
          issues regarding hazard evaluation and .prevention,, monitoring
          and detection, public alert, and'management support.  They
          will be asked to refine the discussions in this.report,
          identifying gaps in methods and technologies, and identifying
          an effective means of disseminating information to those who
          need it.                                ;

-------
                           -15-

Work to strengthen the State Emergency Response Commissions
(SERCs) and LEPCs by providing technical assistance, guidance,
and information on chemical hazards, causes of accidents,
prevention methods, and effective procedures and technologies
for notifying the public of such accidents.  This information
can be used in their dialogue with industry, not only to aid a
community's emergency planning effort and efforts to assess,
understand, and manage risk at the local level, but also to
stimulate discussion of ways to prevent accidents.

Conduct further study on the causes of accidents and ways to
prevent them through such mechanisms as EPA's Accidental    ;
Release Information Program, which focuses on the causes of
chemical accidents and prevention methods and is currently
being pilot-tested.  This information must be shared with the
SERCs, LEPCs, industry, professional and trade associations,.
and'the public.  The studies should address the rationale for
using specific technologies and techniques, the effects of
liability on industry practices,  and other important issues.

Encourage industry and professional and trade associations to
continue to improve the state of practice and conduct research
that will develop and refine cost-effective, reliable
prevention, monitoring, detection, and public alert methods,
procedures, and devices.  Encourage a continuation of the
research at the Department of Energy (DOE) test facility on
the behavior of dense gas releases.   Suitable means for
transferring the results of the research to users should also
be developed.

Develop guidance, where warranted, in areas such as the use of
in-place sheltering, hazard assessment, successful management
approaches, appropriate use of real-time dispersion modeling,
and methods for dealing with hazards.

Enhance Federal and State expertise in the field of chemical
process safety by working with other agencies to develop
training for the conduct of facility audits in this area.  The
Agency will continue to investigate accidental releases of
hazardous chemicals in cooperation with other agencies such as
the Occupational Health and Safety Adminstratioii (OSHA),  the
Nuclear Regulatory Commission (NRG), and FEMA.

Work with international organizations to identify and draw
upon their expertise in the causes of accidents and to share
information on hazardous chemicals,  modeling, and the causes
of accidents and ways to prevent them.   The' recent agreements
reached at the High Level Conference on Accidents Involving
Hazardous Substances under the auspices of the Organization
for Economic Cooperation and Development laid the foundation
for this work.        .

-------
                                     -16-  ,

                                4.   MANAGEMENT
     The commitment of management to accident prevention, mitigation, and
preparedness is essential.  Without such commitment, installation of the most
advanced technologies will be an expensive, but ineffectual safeguard for
preventing serious injury, death, or environmental damage.

     While accidents can occur in well-managed facilities, the lack of
management commitment can lead to disaster.  Accident investigations have
discovered incidents where the releases occurred because managers ordered
operators to use equipment in ways that directly conflicted with posted
procedures.  Part of management's responsibility is to train workers to
operate equipment properly.  The lack of management commitment to safety is
evident in the cases where releases have been the result of inadequately
trained operators.  The best equipment can be extremely dangerous in the
hands of untrained workers.
4.1  MANAGEMENT PRACTICES
                                                   i
     The ultimate responsibility for the safe design, operation, and
maintenance of a facility rests with management.  A comprehensive management
approach to accident prevention begins with the choice of the site,
equipment, process, and design, and continues throughout the operation of the
facility.  A comprehensive approach must include up-to-date operating
procedures, training programs, regular maintenance schedules,  the use of
recognized hazard evaluation methods,  and an active involvement with the
community.                         •            .<.....

     This section outlines key elements in a comprehensive program, then
discuss some of these in greater detail in light of the findings of this
review.
     4.1.1  Key Elements in Management Programs

     Several organizations, both professional and
programs to define good management practices for
of facilities, and for production, processing, use
substances.  The components of such programs include
following items:
 industrial,  have outlined
the construction and design
   and storage of hazardous
     among others, the
          Capital project review and design process review should be
          conducted to analyze all designs for safety problems before
          they are approved; this can include preliminary hazard
          assessments as well as pre-start-up safety inspections.  Human
          factors, such as the ease of operating equipment, should be
          considered in the design of the system.  '

          Management must recognize that compliance with standards and
          codes of industry, associations, and laws of governments is a
          minimum, and the intent of these standards must be applied on
          a case-by-case basis.  Compliance with standards alone does
          not ensure a safe operation.             !

-------
     9.
     10.
                           -17-

 Process  safety  information should be documented to provide
 identification  of  the hazards, a description of key design
 data,  and technical specification of each step of the path to
,a  safe operation.  This document should include rules and
 procedures for  a safe operation.

 Accountability  of  personnel  involved in the operation of the
 facility should be defined.

 Risk management should include identification and evaluation
 of potential hazards using valid hazard evaluation techniques.
 These  must be regularly scheduled hazard evaluations involving
 key personnel,  not just weekly or monthly inspections.

 All process and facility changes should be evaluated for their
 impact on safety;  each process or facility change should be
 subjected to the same rigorous review as would be applied to a
 new process.  Authorization  for change should include new
 operating procedures, training, and maintenance schedules.

 Process  and equipment integrity should be checked by periodic
 testing  and inspection; this includes quality assurance.

 Prompt investigation of all  serious and potentially serious
 incidents should determine the cause(s) of the incident.  The
 investigators should make recommendations for corrective
 actions.

 Training,  including testing  and refresher courses, should be
 provided to all workers.

 Audits of key elements should be conducted by technically
 qualified personnel; deficiencies and corrective actions
 should be documented.
     11.  The 'facility should have methods to enhance its knowledge of
          process safety; that is, the facility should have a method to
          increase knowledge about the processes and equipment.

     12.  Emergency procedures should be in place to respond to  a
          release and to contact the local community.  The facility
          should also have ongoing programs of communications with the
          community.

     The required level of detail in the above elements needed to create a
safe operation will vary with the chemicals, the size of the facility, and
the complexity of the process.  In addition, the overall effectiveness of a
management system that contains all of the above elements will depend on the
commitment of the managers to comply with both the letter and the spirit of
each of the elements and any other elements that may be needed.

-------
                                     -18-         |

     4.1.2  Safety Management: Organization        i     '

     A variety of ways exist in which to organize:a facility to ensure
chemical safety.  Some facilities establish separate safety departments;
others designate safety officers; still others delegate the responsibility to
a number of departments.  Some facilities that responded to the questionnaire
stated that safety was everyone's job.  Because it is management's
commitment, rather than its structure, that ensures safety, there is no
"best" organizational structure.

     The expert panel on management reviewed the facility questionnaires to
assess qualitatively the degree of management awareness and commitment to
safety.  The panel looked at the reports on training, accident investigation,
appreciation of the hazards involved, management response to these hazards,
and management involvement.  For about a third of!the respondents, the expert
panel felt the management approach seemed consistent with "setting goals and
measuring achievements," two key elements of quality management in any area,
including safety.                                 ;

     About a quarter of the respondents did not seem to appreciate the risks
involved, although in some cases, those risks may!have been small.
Therefore, the panel was uncomfortable with the degree of management  .
awareness of risk for a significant minority of responding facilities.

     For many facilities, especially among the smaller companies, safety
appears to mean worker safety rather than loss prevention.  When asked about
protection systems, these facilities listed items such as respirators rather
than technologies such as scrubbers.  For these facilities, management must
be made aware, through training and other means, of the possibility of major
releases and the importance of management in preventing such releases.

     4.1.3  Training

     For safe operation and maintenance of a facility, employees must be
trained in operating procedures related to their particular work, safety
measures for handling the specific chemicals, fire and accident procedures,
and emergency response measures.  While 135 of the 146 facilities responding
to the questionnaire train their employees, internal, formal training
programs appear to be limited to the larger facilities.  It should not be
concluded that less formal training is necessarily any less effective.  In
small operations, a one-on-one training format may be the only feasible
method and could be more effective.                  '
                                                  \r  .                       .
     All the reporting facilities that handle the:more hazardous air toxics,
such as phosgene, hydrogen cyanide, and methyl isocyanate, provide training
and equipment to address emergencies.  Where fire protection is relevant
because of flammable chemicals in use, virtually every respondent offers
training programs and regularly scheduled drills.  Only 15 of the respondents
reported training in hazard evaluation/loss prevention; one of the 15, a
chlor-alkali facility has developed an in-house training program to teach
hazard evaluation techniques to management and operating .personnel.

     The findings from the survey are supported by the information EPA has
collected through its Accidental Release Information Program (ARIP), which
has established a data base on the causes of accidents as well as on actions

-------
                                     -19-

taken to prevent recurrences.  An overwhelming number of the ARIP respondents
cited training of operators as a release prevention measure.

     4.1.4  Maintenance

     While preventive maintenance is always preferred, it seems that
preventive maintenance for critical elements is not universally or
consistently applied at facilities that handle, use, or store hazardous
chemicals.  More than a third of the facilities that returned questionnaires
reported that they have no preventive maintenance program.  In addition, many
of those that do preventive maintenance evidently do it for a limited range
of process equipment and not on a scheduled basis.  While a few facilities
reported using techniques such as ultrasonic inspections to find "hot spots"
in their pipelines, other facilities stated that pipelines do not need
inspections.

     4.1.5  Accident: Investigation

     Investigation of accidents and near misses can provide important
information, on flaws in process design or operation and should produce clear
recommendations for improvements.  This appears to be generally understood in
the chemical industry although a number of facilities reported that they
investigated only those accidents in which people were injured' or a discharge
permit was violated.

     Most of the facilities that responded to the questionnaire use formal
accident investigation procedures.  Almost all reported that they had
modified equipment or operations following the investigations and that they'
perceived improvements from the changes.  Because of apparent liability
concerns, industry is reluctant to share accident investigation information,
which could help prevent accidents at other facilities.

     4.1.6  Interactions with Outside Organizations

     Participation in the Chemical Manufacturers Association's Community
Awareness and Emergency Response Program (CAER) or related industry or public
programs reflects the facility management's outward commitment to safety.
Almost all of the chemical producers that responded to the survey reported
involvement in programs such as CAER or local cooperative agreements with
other companies.  On the other hand, the survey indicated that users and
distributors are more likely not to participate in industry programs.  The
same pattern was evident in answers to questions on involvement with LEPCs
and local communities:  chemical producers showed a higher relative
involvement in the more sophisticated emergency coordination and exercise
participation activities;   users, storers,  and distributors reported
relatively little coordination.


4.2  DEVELOPMENT AND DISSEMINATION OF MANAGEMENT TECHNIQUES

     Since Bhopal, .the large chemical producers,  trade associations, and  " :•
professional societies most directly involved in the safety of chemical
production processes have demonstrably increased their efforts to address
large-scale problems of chemical safety.  Many of the larger companies have
conducted intensive internal reviews of technologies and management practices

-------
                                     -20-

and corrected problems.  These companies as well as the trade associations
have attended meetings and symposia on the subject of risk management and
safety practices.  This type of meeting is important because, whereas CAER
deals with external relations, these gatherings deal with internal issues of
risk management.  The effect of the Bhopal tragedy1 and other accidents has
been less marked on smaller chemical producers and' on distributors and
handlers.  In reviewing attendance at prevention symposia it appears that
these companies are far less likely to participate: in professional
organizations and public meetings on safety.      |

     Scientific management techniques are studied within the fields known as
operations research and management sciences.  Since the 1940s techniques have
been developed in many areas potentially related to chemical safety,
including production scheduling, inventory control, management information
systems, and decision-making under conditions of risk and uncertainty.
However, professionals in management science have paid scant attention to the
issue of managing loss prevention in the chemical industry.  In a recent 18-
month period, fewer than one percent of the presentations at meetings and
symposia on management dealt directly or indirectly with chemical safety.

     Areas where the work of management science could be extended to more
facilities handling hazardous chemicals include:  i

     •    The organization and control of production processes - - risks
          of chemical releases should be factored into the design and
          daily operating decisions;

     •    Inventory control -- the adverse effects' of accidental
          releases should be reflected in the cost; equation;

     •    Quality control and quality assurance --! situations that can
          lead to a release are fundamentally similar to situations that
          can lead to an inadequate product;      i

     •    Facility location, design of distribution systems, and the
          layout of the facility -- while long focused on balancing
          costs, these are also relevant to preventing releases; and

     •    Use of decision analysis tools -- these Approaches can be used
          to consider the possibility of accidents' in the plans and
          operations.

     Some managers who responded to the survey did; not have a clear
appreciation of the hazards in their operations.  Studies on decision-making
practices of managers in general have found that:
                                                  I
     •    Most managers ignore the least probable options, no matter how
          large the consequences (positive or negative) ; and

     •    Most do not consider the possibility of several improbable
          options occurring at once, and generally; focus only on the
          seriousness of the outcome.  As a result; of this narrow view,
          the managers do not look at the low-probability, high-risk
          events.

-------
                                     -21-

     This finding is in line with the attitude that "it can't happen here."
The challenge is to overcome this attitude.  The chemical disasters of the
past several years show the necessity of considering the worst possible case;
Most disastrous accidents would not have been considered credible before they
occurred.
4.3  RECOMMENDATIONS
          The message on chemical safety and its management- must be
          disseminated to users, handlers, and producers of chemicals,
          especially the smaller facilities.

          Management should be trained in the area of hazard evaluation
          and accident prevention.

          Wider study of management techniques should be encouraged;
          successful management approaches and models should be made
          available to chemical producers, users, and handlers,
          especially to smaller companies.

          Mechanisms should be developed to make users, handlers, and
          distributors more aware of risks and risk management.  Title
          III mechanisms are appropriate for this because they compel
          the facility to examine their operation and convince the local
          community that they are operating in a safe manner.

          Guidance on training and maintenance should be developed.

          Mechanisms should be developed to collect, analyze, and
          disseminate accident investigation information.  Industry
          should be encouraged to share this information.  This practice
          has been limited in recent years by liability concerns.

          EPA must work closely with OSHA and other agencies such as NRG
          to foster prevention initiatives.  These can include
          participation in chemical process safety audits and the
          development of guidance and guidelines.

-------
                                     -22-

                            5.   PREVENTION SYSTEMS
     This chapter discusses the techniques and technologies available and in
use for the prevention and mitigation of accidental releases of extremely
hazardous substances.  These technologies and techniques can be divided into,
three categories:  hazard evaluation techniques, pre-release prevention
techniques and technologies, and mitigation technologies.
          .                                                           '
     Whild" this chapter focuses on technologies and techniques, prevention
does not depend on a single piece of equipment or a single technique.
Prevention must be part of a comprehensive, integrated system that considers
the hazards of the chemicals involved; the hazards 'of the process, the
hazards to the community, and the capabilities of facility personnel. , None
of the elements should be considered in isolation nor should any single
technical solution be considered a complete solution to a particular problem.
Each change in a facility, process, or procedure will have multiple effects
that must be assessed in the context of the entire operation.

     The systems covered in this chapter include technologies such as process
control instrumentation, specific designs of process vessels and the process,
line, as well as techniques for identifying hazards and critical elements in
a process.  Prevention systems also include factors such as facility siting
and layout, and management procedures such as training programs.  Because a
wide variety of technologies exist, especially in the pre-release prevention
area, this review does not focus on particular technologies, but instead
presents an overview of prevention systems based oii information collected
through the literature search and supplemented by the responses to the
facility survey.
5.1  HAZARD EVALUATION                             ;

     Hazard evaluation techniques are formal procedures employed to identify
potential risks that could lead to an accidental release and, in some cases,
to evaluate the probability of an accident and its (potential effects.   Hazard
evaluation procedures provide a means of identifying needed changes in
process design, operation, or monitoring.  For process lines, hazard
evaluation methods can help identify the critical e.lements that may need to
be modified to incorporate redundant or back-up systems.

     The American Institute of Chemical Engineers (AIChE) has published a
report, Guidelines for Hazard Evaluation Procedures.  which identifies eleven
formal, qualitative, recognized techniques and describes the capabilities of
each as well as the situations in which they are appropriate.  These eleven
techniques identify the hazards at a facility or in a process.  The
applicability of any particular technique depends on the size and complexity
of the facility and on the level of risk involved, j The eleven techniques are
described in Appendix 3.  Exhibit 5-1 shows the applicability of these
techniques in a thorough hazard evaluation process j

     Another type of hazard evaluation is quantitative risk assessment, the
most highly detailed of which is probabilistic risk assessment (PRA) ,  used in
the nuclear power industry.  PRAs use generally recognized methods, such as
fault tree analysis, to quantify the probability of a release, and use source

-------
                                                                    Exhibit 5-1
                                                  Hazard Evaluation  Procedures
                                                                            HAZARD EVALUATION PROCEDURES
    Identify Deviations From
              Good Practice

            Identify Hazards
      Estimate "Worst Case"
              Consequences

     Identify Opportunities to
      Reduce Consequences

  Identify Accident-Initiating
                     Events

     Estimate Probabilities of
            Initiating Events

     Identify Opportunities to
      Reduce Probabilities of
            initiating Events
      Identify Accident Event
             Sequences and
              Consequences
     Estimate Probabilities of
           Event Sequences

      Estimate Magnitude of
     Consequences of Event
                 Sequences
     Identify Opportunities to
  Reduce Probabilities and/or
Conseq. of Event Sequences
                               Process/
                               System
                              Checklists
Safety
Review
 Relative
 Ranking
Dow&Mond
PreBmtnaiy
 Hazard
 Analysis
"What IT
Method
            Failure Mode*
  Hazard and    Effects and      Fault         Event       Case        Human
OperatibiHty Study   CriUcaltty       Tree         Tree     Consequence       Error
              Analysis      Analysis       Analysis     Analysis       Analysis
                                         Primary Purpose    KXXJ Provides Context Only

                                         Secondary Purpose  {»«H Primary Purpose for Previously Recognized Hazards
                              Source: The Center for Chemical Process Safety, American Institute of Chemical Engineers.  Guidelines for Hazard Evaluation Procedures.
                              Prepared by Battelle Columbus Division. New York, New York, 1985.
S83185-lb

-------
                                     -24-           .

term models, dispersion models, and health effect models to predict effects
on the populations exposed.  Probabilistic risk assessment is the most costly
and time-consuming of the hazard evaluation procedures.

     PRAs are not recommended for every facility.  Many operations do not
need a quantitative method such as a PRA to identify hazards thoroughly
enough to take steps to address those hazards.  For facilities with complex
processes, PRAs can be useful for comparing the effects of using different
technologies, but in their current state, the quantification may be
meaningful only for qualitatively assessing the relative level of risk; that
is, the PRA can be used to compare the level of risk between two
technologies, but not to define an absolute level of risk.
                                                   f
     The accuracy of a PRA depends on the availability of reliable data on
the actual causes of accidents and on equipment failure rates and human error
rates.  These data are not, at present, sufficiently reliable.  Ideally, data
on equipment failure rates should be specific to the facility, process, and
operating procedures; at this time, however, most of the data are derived
from data bases that are less specific.  Despite these limitations,
quantitative methods such as PRAs can help define where data are needed and
can be used to examine the effects of data gaps on the analysis of risks.
The AIChE plans to publish guidelines this year on quantitative risk
assessment procedures and on obtaining process equipment reliability data.

     The facility questionnaire responses indicate that ongoing efforts to
inform chemical facilities about hazard evaluation jtechniques need to be
expanded.  Although almost all of the respondents stated that they use hazard
evaluation methods, only half are using AIChE-recognized techniques.  The
other facilities listed less formal techniques or techniques not generally
considered to be hazard evaluations (such as hazard committees and safety
training).  Some facilities listed recognized techniques such as safety
audits and checklists, but stated that they conducted them daily, weekly,  or
monthly, which indicates that they are simply routine safety inspections
rather than formal hazard evaluations.  In general, many facilities seemed to
equate hazard evaluation with routine safety measures.   This finding
indicates that, beyond disseminating information on accepted techniques, the1
terminology used in this area should be standardized.
5.2  PRE-RELEASE TECHNOlJOGIES

     Pre-release prevention or loss prevention technologies are designed to !
reduce the probability that the primary containment of the chemical will be
breached.  Pre-release prevention includes such factors as facility siting
and layout, as well as prevention technologies.  The overall prevention
process is diagrammed in Exhibit 5-2.

     The use of particular technologies is site-, process-, and chemical-   ;
specific.  The technologies can be either intrinsic -- integral to the design
of the facility and process line --or extrinsic -- supplemental to the     ,
process technologies in place.  While the use of appropriate technologies is
essential, the safe operation of a facility is the result of a complex and
interlocking set of factors in which not only technology, but also management
practices, play critical roles.  The best technologies available will not be

-------
                                                         Exhibit 5-2

                                                   Prevention Process
                                                                                                          Unscheduled
                                                                                                             Hazard
                                                                                                           Evaluation
                                                                                                         Change Design
                                                                                                           Operation
                                                                                                           Scheduled
                                                                                                             Hazard
                                                                                                           Evaluation
S8442-2a

-------
                                     -26-

adequate if they are not safely maintained and operated, which results from
an overall management commitment to safety.        ;

     5.2.1  Siting and Facility Design

     As was discussed in the Interim Report, the siting of a facility and its
layout can affect the degree of risk that a release of extremely hazardous
substance may pose to the surrounding community.  Since the majority of the
facilities that responded to the questionnaire were at least 15 years old,
survey respondents were usually unable to provide information on the criteria
used during the initial stages of the facility's development.

     Residential or commercial development that occurs after a facility is
built can increase potential risks.  Some of the facilities that responded to
the survey identified process changes that were macle because of changes in
the local community.

     As stated in the Interim Report, a number of widely accepted codes,
standards, and recommendations provide minimum safety standards for equipment
design, procedures, and systems.  It should be stressed that these standards
and codes are minimums.  Management must understand the degree of protection
provided by the standards; for the most part, meeting the minimum safety
requirements is not sufficient for facilities handling extremely hazardous
substances.

     5.2.2  Process Design.                         j

     The number of possible approaches to limit accident potential is so
great that the ability of a system to prevent releases can only be appraised
through inspection by experienced technical personnel or by a detailed hazard
evaluation.  However, regardless of the specific equipment used in a
particular process line, it is important that, for critical elements, the
process line include redundant systems, backups, and interlocks (an automatic
shutdown of a process if a critical piece of safety equipment becomes
inoperative).  For example, emergency power and cooling systems can prevent
releases if the primary systems fail.  EPA's ARIP data base indicates that
such emergency backup systems may not be widely perceived as necessary; very
few ARIP respondents reported installing such systems after a release.
Appendix 10 provides a summary of ARIP's initial findings on release
prevention and post-release measures.              ;

     Process design changes that may serve to prevent releases include the
substitution of less hazardous chemicals, reduction of the severity of
process condition (e.g., temperature and pressure), reduction of process
complexity, and improved operation and safety procedures and'training.  Many
of the surveyed facilities cited the installation of hardware within the
process as an action taken to minimize the frequency and severity of
emergency releases.  An ammonia facility upgraded flares, decreased its
inventory, eliminated pressurized transport, and installed pipeline isolation
valves; as a result, its reportable releases decreased from 12 per year in
1982 to one in 1987.  A phosgene facility upgraded!its process control system
from analog control to computerized, direct digital control with interlocking
safety/shutdown logic; the facility noted a significant reduction in
accidental releases.                               i

-------
                                     -27-

     A few facilities such as the ammonia facility reported the reduction of
inventories of hazardous chemicals; one reported modifying its process to
completely eliminate the storage of one chemical.  Among facilities handling
extremely hazardous substances, there appears to be a trend toward smaller
storage facilities.

     5.2.3  Process Control and Monitoring

     Process control and detection capabilities are necessarily dependent on
the chemical and the complexity of the process.  (The monitoring and
detection devices referred to here include the instrumentation and other
controls on the process line itself, rather than detection devices at the
perimeter of the process area.)  The process hazards most commonly identified
by respondents are overpressurization, runaway (exothermic) chemical
reactions, fire and explosions, and overfilling vessels.  Many of the
controls and detectors used to monitor these hazards are the standard
equipment used to control pressure, temperature,  and flow.  In addition,
although not directly related to process monitoring, corrosion monitoring
(using probes or ultrasound testing) and vibration detection are important in
the prevention of accidental releases.  Appendix 5 provides examples of
prevention and protection methods for the chemicals considered in this
review.

     While the facilities responding to the questionnaire cited a number of
areas in which they use process monitors,  many of these appear to be designed
for routine monitoring rather than specifically for detecting accidental
releases.   Routine process monitors can be very effective as leak detectors,
if properly placed and maintained.  For example,  a sudden pressure drop may
indicate a vapor leak .in the system and can be the first warning of a
problem, even before air monitors detect a cloud.  Of the 67 facilities that
reported using leak detectors, most were handling gases or highly volatile
substances.

     Although methods for detecting a disabled control device and back-up
systems may not be the primary focus of prevention, they are important
because the operators need to know when control devices, which are a front-
line defense, are not functioning.  Most of the respondents have some way of
detecting disabled controls, with monitors, operator observation, and, routine
inspection being most commonly used.  Fewer than half of the respondents',
however, appeared to recognize the need for redundant controls.  For example,
a chlorine packaging facility handling large amounts of chlorine in a
moderately populated area reported that it has no back-up systems for
pressure sensors used for process monitoring; the facility relies on routine
inspection to detect disabled control devices.

     5.2.4  Protection

     Pre-release protection is the application of equipment, systems, and
procedures to capture, neutralize, or destroy a toxic chemical before it. is
released,to the environment.  The types of pre-release protection measures
needed at,a facility depend on the specific hazardous substances present and
the surrounding environment that their release could affect.  The measures
can be applied only when the released chemical is confined, when it can be
transported into properly sized control systems,  or when circumstances allow
remedial actions to be taken.

-------
                                     -28-          I

     Pre-release protection measures applicable to potential releases to
surface soil and water from storage tanks and process equipment include
maintaining back-up pumps or  secondary structures to remove or contain
discharging liquids, possibly followed by treatment of the substance.  For
reactive or soluble gases and vapors, two common protection devices are
emergency scrubbers and flares.  Scrubbers employed as a protection measure
usually use water or aqueous  mixtures with specific scrubbing reagents.
Chemicals such as ammonia, chlorine, hydrogen cyanide, hydrogen fluoride,
hydrogen sulfide, methyl isocyanate, phosgene, sulfur dioxide, and sulfur,,
trioxide are readily amenable to scrubbing.  Flares are an appropriate
measure for flammable gases and vapors, but generally not an appropriate
measure for streams producing highly toxic or acidic, corrosive combustion
products.

     The responses to the questionnaire  indicate that scrubbers are used by a
significant number of facilities.  Flares are not ajs widely used as
scrubbers.  Whether scrubbers and flares are an appropriate protection
method, however, depends on the process  involved.  ,It was difficult to
determine from the survey responses whether the facilities not using them
fail to do so because they are not appropriate or because the facilities do
not recognize their potential usefulness.  Only the largest facilities have
stand-by scrubbers to be used for accidents.  Other facilities have scrubbers
available to mitigate routine releases;  these scrubbers may be able to
mitigate accidental releases.  In general, flares are likely to be found only
at large facilities or facilities handling large volumes where the subject
chemical would be flared in a mixture with other flammable gases and vapors.
Flares appear to be used particularly by facilities! handling acrylonitrile,
ammonia, hydrogen cyanide, hydrogen sulfide, and phosgene.
5.3  POST-RELEASE MITIGATION                      :

     Post-release mitigation measures are used when a loss of containment has
occurred, but while the extremely hazardous substance is still within the
facility perimeter.  They serve to reduce the extent,to which the substance
migrates off-site by reducing the quantity or concentration of the substance
to which people would be exposed.  Effective mitigation applications are very
specific to site, location,  substance properties, process characteristics,
and scale of operation. Uniformity and consistency Jin the application of any
one mitigation technique is, therefore, neither expected nor desired.

     Mitigation measures can be applied to some atmospheric releases of vapor
and gases as well as to liquid spills to soil and surface water.  However,
mitigation is effective and  viable only for some release scenarios'; in many
situations, mitigation systems are not technically feasible.

     The unconfined nature of large vapor releases and the short response
times renders mitigation of  such releases difficult.  If the release results
from a total containment failure with an instantaneous release of all
material (as opposed to a gradual release), the chemical cloud usually moves
beyond the perimeter before  any mitigation measures' can be activated.  For
such scenarios, concentration on ways to prevent the release is vital.
                                                   !
     Although water sprays and curtains are used, their effectiveness for
vapor releases has not been  demonstrated on a large1 commercial scale.  To

-------
                                     -29-

design and evaluate a mitigation measure requires full-scale testing with
varying meteorological conditions; this testing is very costly, requiring
expert personnel, instrumentation, data analysis capability, and a special
site (for example, the DOE testing site in Nevada).

     Liquid releases can be large and instantaneous as well, as was
demonstrated in the recent fuel oil release into the Monongahela River in
Pennsylvania.  In many cases, however, where the released chemical is a
liquid.with a low vapor pressure, properly designed dikes, pits, and dikes in
combination with measures such as foams that limit volatility can contain a
release.

     Two-phase releases, where both liquid and vapor result, require a
combination of liquid phase and vapor phase techniques, applied
simultaneously.  Because the vapor phase techniques are of  limited usefulness
in many cases, one goal of two-phase mitigation is to prevent the chemical
from vaporizing.  For unconfined, two-phase releases, the effectiveness of
the technologies -- foams, absorbents, reactants, physical  covers, and water
sprays/fogs/curtains --is unproven for large-scale releases.

     Solid releases are usually considered less of a hazard to the public
because solid particles require greater energy to become airborne and do not
have the same natural buoyancy as gases and vapors.  Water  sprays, fogs,  and
curtains may be an option for these releases, but the technology for
mitigating these releases is relatively undeveloped.
5.4  RECOMMENDATIONS
          For extremely hazardous substances, emphasis and resources
          should be placed on pre-release prevention rather than on
          post-release measures.

          Information on prevention as a comprehensive, holistic
          approach should be disseminated to smaller companies.

          Information on the applicability and limitations of hazard
          evaluation techniques should be widely disseminated and the
          terminology in the hazard evaluation area should be
          standardized.

          A mechanism should be developed to improve the data on
          equipment failure rates and human error  rates.  The government
          should work with standard-setting professional organizations
          to improve the understanding of the data.

          Research should be conducted on mitigation technologies for
          liquid, two-phase, and solid releases.   Mitigation
          technologies for large-scale releases should also be
          evaluated.

          Information on the design and effectiveness of protection
          devices, focusing on operation under emergency conditions,
          should be developed and disseminated.

-------
                                     -30-           !
                                                   •I
                     6.   MONITORING AND DETECTION SYSTEMS


     This chapter discusses detection systems used to detect releases in the
process areas of a facility, and monitors used outside the facility, either
at the perimeter or in the community.  In addition, the chapter discusses air
dispersion models that have been developed to predict the movement and
concentration of releases of extremely hazardous substances.
                                                   i                 •
     Detection systems are integral to a facility's1 emergency response      ;
capabilities.  By using these systems, releases can, in some cases, be
detected sufficiently early to prevent catastrophic results.  There are four:
basic process area detection systems:  observation by operator, multi-
detector systems, fixed detector/multi-sampling point systems, and remote
sensing systems.                                   :

     Monitoring of ambient concentrations is used to determine the impact of
a release and the distances to which local residents must be evacuated or
within which residents must be asked to remain indoors.  Monitoring is also
used to determine when concentrations have decreased sufficiently to allow
re-entry into the area.  Congress specifically directed the Agency to examine
the technical and economic feasibility of perimeter alert systems,  detection
systems deployed around the perimeter of a facility.  The review also looked
at two systems used in the field:  portable monitoring instruments and field
laboratories.

     This chapter provides an overview of the current status of technological
capabilities of monitoring and detection systems and discusses the general
factors affecting the decisions to install the various technologies.
Emphasis is given to systems used to detect and monitor the movement of a
plume formed by chemicals that are released into the air.   Releases to
surface and ground water, which pose less immediate threats to public health
and the environment, are monitored by long-established techniques;  monitoring
and detection of these releases were covered in the Interim Report.  This
chapter also discusses some of the general limitations of the technologies
and of the use of existing dispersion models.      t


6.1  DETECTION TECHNOLOGIES                        i    '  '
                                                   i
     6.1.1  Observation by Operator                '

     The primary method industry uses to detect releases is observation by
process operators and other personnel; observation includes monitoring of
instruments as well as seeing, hearing, or smelling, a release.  At many
plants, personnel are required to conduct inspections on a regular basis
using checklists.  In some plants, video camera surveillance augments the
inspections.

     The primary reason for relying upon observation is that personnel
present at the facility may detect a release before it becomes serious.
Experienced, trained personnel can recognize a release situation before the
release would trigger any instrument.  The main disadvantage is that this
approach may result in short-term exposures of humans to the released
chemical.  Relying on observation by operators may also place unrealistic

-------
                                     -31-

demands on human performance.  However, observation continues to be the prime
method because, for some chemicals such as ammonia and chlorine,  the human
senses of smell, taste, and sight are much more sensitive than any instrument
now.available and are also more reliable.

     For chemicals such as phosgene that do not have an appreciable or
disagreeable odor or do not cause irritation prior to reaching concentrations
of concern, an instrumentation-based detection system is needed to prevent
releases from remaining undetected until it is too late to initiate
protective action.  For some chemicals such as hydrogen sulfide,  which has a
strong odor even at safe levels, instrument-based detection systems are also
needed because the chemical destroys the sense of smell as it reaches acute
levels.  In all cases, facilities should consider detection systems to
reinforce observation.

     6.1.2  Multi-Detector Systems

     Multi-detector systems involve the placement of a number of real-time
detectors at selected locations throughout a process area.  When one set of
detectors senses that chemical concentrations have exceeded a set level, an
electrical signal triggers a visual or audio alarm.  There are two types, pf
multi-detector systems:  general parameter devices (e.g., hydrocarbons,,
combustible gas detectors) and compound-specific devices.  The general
parameter detectors respond to a wide variety of chemical compounds and are
used in areas where it is unlikely that any compounds other than those of
concern will interfere with detection.

     Compound-specific devices produce electrical signals that are directly
related to the concentration of a single compound.  There can, however, be
interfering compounds present that will either cause a response or inhibit
the response of the device to the target compound.  Therefore, this type of
device is used only where there is little chance of encountering interfering
compounds.  (See Appendix 6 for a description of specific detection devices.)

     Multi-detector systems provide 24-hour detector availability.  One
disadvantage of the detectors is that a relatively large number of detectors
may be needed to ensure that a plume of released chemical does not slip by
the detectors.  Another disadvantage is that many of these detectors are
relatively insensitive and must be placed fairly close to the potential
release point, thus limiting the size of the area they can cover.
Additionally, they require regular maintenance and calibration.    '

     Among the facilities responding to the questionnaire, those that use
process,area detection systems are generally using chemical-specific systems.
The decision to install such a system depends on the perceived reliability of
the system and its cost-effectiveness, both of which are site-specific.;, A
small.facility in a remote area may decide that the costs of the system are
not justified because the result of a release will be negligible.  The same.
size facility handling the same chemical in an urban location may decide a
process area detection system is needed because of the potential consequences
of an accidental release.                                           -..-.•

     In general, to most surveyed facilities cost-effective means
inexpensive, and reliable means dependable and easy to use.  The more -costly
the systems, the less likely facilities are to install them.  Reliability is

-------
                                     -32-

usually seen in terms of the need for calibration and for monitoring of
instrumentation by skilled technicians.  For example, some chlorine detectors
are calibrated at the factory and hold their calibration for months.
Combustible gas detectors are affected by ambient conditions such as
temperature and humidity and need continual adjustment.  Another factor, in
reliability is the ability of the detector to function without false alarms.
Some detectors can be triggered by other chemicals and others simply go off
periodically for no discernible reason.  Whatever the cause of the false
alarms, any significant number of them can cause people to ignore or disable
the alarms and, therefore, render the detection system useless.
                                                   I                         !
     For a number of chemicals, particularly chemicals released in solid
form, no satisfactory detection systems exist.  None of the facilities
handling solids reported using detectors.  In part,; they may not use
detection systems because many large particulates do not migrate very far
and, therefore, do not usually create emergency conditions.  An additional
problem for particulate releases is that the available detection systems are
costly and unreliable; most systems cannot distinguish between benign dust
and particles of hazardous substances.

     Highly corrosive chemicals, such as hydrofluoric acid, destroy detectors
fairly rapidly.  The materials that resist corrosion, such as stainless
steel, do not make good detectors.  For some chemickls, one possible solution
in this area is to use the destruction of glass and plastic by the eorrosive$
as a. detection system.  The rapid clouding of the glass would be the
indicator of the release.  For this method to be feasible, however, the
detectors would have to be very inexpensive because each detector could be
used only once.
                                                   !
                                                   r
     For some chemicals considered, particularly ammonia, the facility
questionnaire responses indicated a lack of awareness of the risks and of
available detection systems.  Ammonia detectors are reliable, but only four
of the 25 facilities that use the chemical use detebtors for air releases.
Some facilities that use ammonia as a refrigerant do not appear to recognize
the potential risk from releases of ammonia vapor. ' Because these facilities
frequently contain large quantities of ammonia and are often unattended at
night, information on the availability and capabilities of existing detection
systems should be provided to them.                '

     6.1.3  Fixed-Detector/Multi-Sampling Point Systems
                                                   s                 • •
     Fixed-detector/multi-sampling point systems use a single detectpr or
instrument.  Samples from several process points are pumped to the instrument
and introduced in rotation for analysis.  This type of system is most often
used when an expensive or complicated instrument or1 method of analysis such
as gas chromatography is needed.  The cost and complexity of the systems make
it impractical to install a detector at each sampling point.  These systems,
usually installed to detect releases of organics, and also used to monitor
releases of inorganic toxic gases, are becoming more common.
                                                   i
                                                   t
     The sophisticated instruments used in fixed detector/multi-sampling
point systems can be highly sensitive.  The trade-off for the high
sensitivity is the loss of continuous monitoring.  Because individual sample
lines are analyzed in rotation, each sampling point! is not monitored
continuously.  For example, an individual analysis for each point often takes

-------
                                     -33-

1.5 minutes; if 20 points are monitored in sequence, each point is then
checked only every half hour, which would make it possible for a plume to
slip by the array undetected.  Additional research and, development needs,to
be done to develop systems that can sample each point at much higher
frequencies.  As with multi-detector systems, fixed-detector systems should
be designed to require little calibration and maintenance.

     6.1.4  Remote Sensing Systems

     Remote sensing systems rely on a fixed-station detector, which receives
optical signals from a released substance.  A detector is placed within a
potential release area and pointed at a light source.  The detector measures
the intensity of the light.  When a chemical plume crosses the light beam,•
part of the light is absorbed or scattered by the molecules of the chemical,
diminishing the intensity of the light reaching the detector.  The detector
then triggers an alarm.

     While a few of these systems have been custom-designed for use in the
detection of volatile chemicals, most commercially available systems have
been designed to monitor dust levels in mines and concrete plants.  These
systems are relatively undeveloped for general use and are very expensive.
Laser-based systems, now in the initial stages of development, hold the
potential of providing an effective detection system because, in theory,
laser systems could scan a facility for a specific chemical every 20 seconds.
6.2  PERIMETER MONITORS

     Perimeter alert monitors are detection devices placed at the perimeter
of a facility to detect releases that are moving into the community;  they
serve both to indicate that a release has occurred and to identify the
direction.  The latter is important in determining which areas may need to be
notified first to take protective actions.

     Perimeter monitors are usually fixed-point detectors and are basically
the same as process area detection systems, except that they must have a
higher degree of sensitivity because of the lower chemical concentrations
that result from dispersion.  Only eight of the facilities that returned,
questionnaires are using perimeter monitoring systems.

     Facilities that have considered perimeter monitoring systems and decided
against installation cite a number of reasons.  A large facility would
require 
-------
                                     -34-          :




                                   EXHIBIT 6-1
                                                  f

               COST ESTIMATES FOR PERIMETER MONITORING SYSTEMS
                              Estimated Cost per Monitoring Station
                         Fluorescent
                         S02 Analyzer
              Photo-electric
               Tape Sensor-*-
                   Electrolytic
                 Chlorine Detector
Equipment Cost             $29,000

Installation Cost           $6,500

Operating/Maintenance       $8,400
Cost (per year)

Annualized Cost2           $13,000
                  $5,000

                  $4,000

                  $7,500
                       I

                  $8,700
                      $5,000

                      $6,000

                      $1,600


                      $3,000
                              Estimated Total Annual System Cost
                         Fluorescent
                         S02 Analyzer
              Photo-electric
               Tape Sensor*
                   Electrolytic
                 Chlorine Detector
Estimated Annual
System Cost for
Large Facility^

Estimated Annual
System Cost for
Small Facility4
$550,000
$130,000
$370,000
 $87,000!
$126,000
 $30,000
1.   Photo-electric tape sensors can be used to detect chlorine, phosgene,
     hydrogen sulfide and ammonia.                :

2.   Annualized cost is the annualized equipment and installation cost using
     a discount rate of 5 percent added to the yearly operating and
     maintenance cost.  A system operating life of;10 years is assumed.

3.   A large facility is assumed to have an area of 640 acres (one
     square mile), approximately 21,000 feet of linear perimeter, and 42
     monitoring stations placed every 500 linear perimeter feet.
                                                  |                  _  ,
4.   A small facility is assumed to have an area of 30 acres,
     approximately 5,000 feet of linear perimeter,  and 10 monitoring  ,
     stations placed every 500 linear perimeter feet.

-------
                                     -35-

and the topography of a facility may make it difficult to install perimeter
systems.  Respondents listed a number of benefits in having the detection
system installed in the process area rather than on the perimeter.  With a
process area system, the chance of detecting the release early and of
preventing greater releases is substantially higher.  By the time a release
has reached the perimeter, it has already reached the community.  Process
area systems are also more cost-effective because far fewer detectors are
needed to provide adequate coverage.

     One respondent reported an extensive investigation of a laser perimeter
alert system.  The facility identified at least three companies that are
actively marketing laser systems to monitor for chemicals such as hydrogen
sulfide, ammonia, sulfur dioxide, hydrogen fluoride, and chlorine.  The
facility found that the laser systems it investigated could provide real-time
measurements of a chemical plume as it crosses the laser's path.  Those
systems had maximum monitoring distances ranging between 100 meters and 5
kilometers, with measurement sensitivity in the parts per million range.  The
respondent evaluated a prototype system and decided against it for several
reasons:  the technology was not considered reliable; the laser may pose
vision safety problems; and the system is currently not economically feasible
for a large facility.
6.3  FIELD MONITORING SYSTEMS

     6.3.1  Portable Monitoring Instruments

     Portable monitoring instruments, used outside the perimeter of the
facility, are real-time, direct-reading instruments that one person can
easily carry and operate.  They are available as both general parameter and
compound-specific detectors.  The advantages of these instruments lie in
their portability and real-time readout.  Disadvantages derive from their
limited sensitivity, from the difficulty of calibration, and from the
possible exposure of monitoring personnel.

     These systems are in use throughout the industry, in some cases to meet
OSHA monitoring requirements.  They are also used in some communities in
emergency release conditions.  In all cases, their usefulness is dependent on
having an operator who knows how to use the instrument and how to interpret
the data properly.

     Robots that can examine an accident scene are being investigated for use
in situations where the chemicals are extremely toxic.  Robots have been used
in highway accidents when the chemicals involved were either too toxic to
risk human exposure or were likely to explode.

     6.3.2  Field Laboratory Instruments

     Field laboratory instruments are mobile lab units transported by van or
bus.  They can produce accurate, precise data on concentrations in air in the
sub-parts-per-million range.  The main advantage of field lab instruments is
that they can quickly provide accurate data; their main disadvantage is that
they are not widely available, must be transported promptly to the accident
site, and require calibration at the site.  For a field laboratory to be
useful in an emergency situation, it must be maintained in a constant state

-------
                                     -36-

of readiness.  In addition, the instruments must be calibrated for the
specific chemicals in the release, which requires that the people operating
the lab know what those chemicals are.
6.4  DISPERSION MODELING

     Dispersion models are computer-based simulation programs that predict
the direction and concentration of a contaminant plume.  Available models
vary from general dispersion models to highly sophisticated, real-time,
proprietary models.

     At present, dispersion models are best used as training and planning
tools.  For these purposes, the simple models are adequate.  The facility and
community can model situations and vary factors such as the direction of the
plume, the wind speed, and the time of day.  With these various release
scenarios the emergency planners can develop training exercises to respond to
the different situations that would arise from the different scenarios.

     Although half the responding facilities use some air dispersion model, a
number indicated that in real-time emergency situations, the applicability of
even the most sophisticated models is questionable.;  As with any model, the
reliability of the results depends on whether all the variables that could
affect the outcome are included and on the accuracy of the data used.  Most
models are based on Gaussian dispersion techniques and may include dense gas
dispersion algorithms.  However, the understanding of dense gas dispersion is
relatively new and these gases may not behave in a Gaussian manner.  (See
Appendix 6 for a description of the models in use at responding facilities.)

     The models currently in use do not adequately consider a number of key
elements that could dramatically alter the release 'scenario.  These factors
include the dispersive characteristics of the chemicals, source strength,
micrometeorology (i.e., local weather conditions), and heat generation (most
chemical plants generate large amounts of heat, which disrupts the general
airflow).  Source strength (size, physical state, and rate of a release) is
the hardest of these to determine in an emergency and is the most critical in
all models.

     The greater the reliance on model output for decision-making, the
greater the need for validation and calibration of the model for the specific
site.  These validation exercises, however, are extremely expensive.  They
require data on source strength and release parameters, meteorological data,
etc., over a sufficient period of time to cover the range of conditions that
could occur.  While the accuracy of the data is not critical for training
purposes, in an emergency, if each necessary factor were off by ten percent
(which is not unusual) , the final prediction could be off by 50 to 100      >
percent.  None of the models available can yet react to changes quickly and
accurately enough to generate more than an estimate of the release.

     The level of sophistication of models has increased rapidly in recent
years and is likely to continue.  Current models need to be validated and
then calibrated for use at specific sites.  In addition, data on source
strength calculations and acute health effects need to be improved.  The
latter information is necessary to interpret the model.  The current models
are useful when used by an expert who realizes their limitations, but can be

-------
                                     -37-

misinterpreted by someone not familiar with the underlying assumptions.   The
AICHE has recently published Guidelines for Use of Vapor Cloud Dispersion
Models.  describing the applicability and limitations of these models.


6.5  RECOMMENDATIONS

     *    Industry should be encouraged to conduct the research needed
          to improve the reliability and cost-effectiveness of detection
          systems.  The emphasis should be on developing systems that
          are relatively inexpensive and that do not require frequent
          calibration or trained technicians to operate.  Remote sensing
          systems (e.g., chemical-specific, laser-based systems) should
          be developed for more general use.

     •    Reliable, inexpensive, throw-away detection systems should be
          developed for corrosives.

     •    Dispersion models should be used in real-time situations only
          with great caution and then only by people aware of their
          limitations.

     •    Information on the capabilities and limitations of models
          should be further developed and disseminated.

     •    Further research should be conducted on source strength data
          and acute human health effects data.  (International
          organizations are currently doing research, as are domestic
          companies and Federal agencies such as EPA, DOE, and the
          Department of Transportation.)  This research is critical for
          understanding and overcoming the limitations of dispersion
          models.                                                      	

-------
                                     -38-

                           7.   PUBLIC ALERT  SYSTEMS
     Public alert and notification is a multi-phased process that begins with
the detection of the release at the facility.  If the facility determines
that the release, could migrate and affect the public, the facility contacts
the local authorities.  These authorities must then decide if and how to
alert the public,  and what actions to recommend.  The components, of these
phases must be carefully integrated into alert and notification procedures
that recognize the need for timely and effective communication.  Exhibit 7-1
illustrates the public alert process.  In an emergency situation, timing is
critical; in rapid-onset events, the release could reach the public very
quickly.  Therefore, the public alert process must be prepared in advance to
operate without delays.

     The public response to a warning is a component that cannot be
guaranteed.  In many cases, hearing a warning is, in and of itself,
insufficient to induce people to take action.  A variety of factors that
relate to the nature of the warning, the characteristics of the receiver, and
the process of confirmation affect response.  Public education is a key part
of the public alert and notification process because it prepares people to  .-
understand what to do when a warning occurs.

     This chapter provides an overview of the technologies and techniques
available for alert and notification.  Most of the information on procedures
and technologies actually in use is derived from the community survey and
from the expert panel.  The data on the first phase of the public alert
process is from the facility questionnaire.


7.1  TECHNOLOGIES

     Public alert systems include two types of equipment:   (1) communication
technologies that facilities and communities use to exchange information and
(2) technologies for  alerting and notifying the public.  The public includes
residential, business  and commercial, institutional, and transient
populations.

     Public warning systems can provide an  alert,, a notification, or both.

     •    An alert system provides a signal  that something  out of the
          ordinary is occurring that requires people to seek more
          information.  Examples of alert technologies include sirens or
          alarms.

     «    Notification is the process by which people are provided a
          warning message and information.   Examples of notification
          technologies include  emergency broadcast systems  (BBS), radio,
          television,  and cable override.

     •    An alert/notification system  serves both purposes.   Examples
          of dual  systems  include  tone-alert radios, telephone dialing
          systems,  loudspeakers, and public address  systems.


-------
                             FACILITY
                                                       Exhibit 7-1

                                                 Public Alert Process
                                                                           LOCAL OFFICIALS
                                                                                                        PUBLIC
                                                                                    Should
                                                                                   Public be
                                                                                   Warned
Yes
S8442-la

-------
                                     -40-

     Some systems, depending on how they are used, may not fall into precise
categories.  For example, helicopters equipped with loudspeakers are a dual
system, but in reality typically do not provide notification because few
people hear the broadcast message.

     7.1.1  Connunlcations Technologies

     The initial step of the public alert process is the facility's
notification of the appropriate community officials.  The primary channels
are commercial telephones and two-way radios.  While these represent the most
common forms of communication systems, experts do not consider them highly
reliable.  Telephones may fail (sometimes from the same event that caused the
chemical accident) or may be busy.  Two-way radios often operate at different
frequencies, are found inoperable, or are difficult to use because of heavy
traffic on the appropriate frequency.

     The communication systems designed to overcome such problems --
dedicated telephone lines (a separate line not linked with commercial
traffic), 911 telephone systems, dedicated radios, pagers, and special alarm
systems -- are not commonly used in the communities surveyed.  The main
communications links, therefore, are the ones that frequently cause warning
failures.  This is not to say failure is certain, merely that advanced
communications equipment exists in relatively few situations.

     7.1.2  Alert Technologies

     Sirens/Alarms.  Although it may be expensive to install and maintain the
systems, sirens and alarms can provide a relatively rapid alert to most
potentially threatened populations.  (See Appendix 7 for cost estimates of
alert systems.)  A few types of sirens have public address capabilities, but
most only sound a noise, which limits their utility.  While some communities
have tried, in essence, to code alarms, so that a wavering tone means
something different than short blasts, the public rarely differentiates among
different alarm signals.  Other problems that constrain the use of sirens and
alarms are false alarms due to technical failures, equipment failures in
emergencies, maintenance problems, coverage problems (particularly in adverse
weather), difficulties in propagating sounds into buildings, and public
indifference to sirens.

     Modulated power lines.  Warning systems exist that use alterations of
the cycle-per-second frequency of regular electric power lines to activate a
warning light, or a buzzer or siren.  Many of the advantages of tone-alert
systems, discussed below, hold for this type of warning device.  However,
modulated power line technology is relatively expensive to install, test, and
maintain and it cannot be used if electrical systems fail.

     Aircraft.  In special cases, airplane and helicopters can be used as
part of the warning process.  Low-flying aircraft can carry sirens or
bull-horns to provide an alert or warning message to remote populations or to
people who cannot b'e reached by normal communication methods.  However, this
method requires access to aircraft and sound systems that can broadcast a
message over the noise of the aircraft.

-------
                                     -41-

      7.1.3  Notification Technologies

      Radio.   Radio is  often used to  disseminate warning  information because
 it can reach a large number of people  during  non-sleeping hours.  Certain
 radio stations have been designated  Emergency Broadcast  Stations as part of
 the National Warning System.   Other  radio  stations broadcast warnings  in most
 emergency situations as  well.   Prearranged plans accelerate the speed with
 which a radio warning  can be issued.

      One disadvantage  of radio is that a broadcast covers areas not at- risk.
 Second,  all  information  must be conveyed verbally.  Third, radio reaches only
 a  small .portion of the population during night-time hours.  Fourth, because
 most  stations are  privately operated,  problems can arise in priorities
 regarding warning  broadcasts.

      Television.   TV stations  can warn the public by interrupting normal
 programming  or by  displaying scrolled  text on the bottom of the screen.  TV
 reaches a large number of people,  particularly in the evening hours.  Like
 radio,  it is  of little use during sleeping hours.  TV is particularly good
 for warnings  of slowly developing events.   One major advantage of TV is.the
 ability to use graphic information such as maps or diagrams In the .warning.

      Cable Override.   In many  areas people have cable TV, which means .that
 local stations play a  lesser role in reaching the public.  As a result,
 systems  have  been  developed to broadcast a message of local applicability
 over  all  cable channels.   Thus,  a person in Cheyenne, Wyoming, watching a
 Chicago- station or a movie channel, could  still receive a warning of a
 Wyoming chemical emergency.  The same  advantages and disadvantages of
 conventional  broadcast TV apply.,                                   ;

      7.1.4 Alert  and  Notification Technologies
     Personal Notification.  Personal notification involves having emergency
personnel go door-to-door or to groups of people to deliver a personal
warning message.  This warning mechanism can be used in sparsely populated
areas, in areas with a large seasonal or diurnal population (such as a
recreation area), or in areas that are not covered by electronic warning
capabilities.  The chief advantage of personal contact is that people are
more willing to respond to a warning because they are more likely to be.lieve
that a danger exists.  The disadvantage is that it is time-consuming to
implement and may require the commitment of many vehicles and personnel.

     Loudspeakers/Publie Address Systems.  Existing public address (PA)
systems can be used to notify people in areas that are" covered by such
systems.  Often schools, hospitals, prisons, nursing homes,  sports arenas,
theaters, or shopping centers have PA systems.  In addition, portable
loudspeakers in vehicles can be used to warn nearby populations and
populations who have no other means of receiving the warning.   They are
particularly useful during night-time hours when many people are asleep.
Their chief disadvantages are that it is often difficult for people to hear a
warning broadcast from a moving vehicle and it is difficult for people to
confirm the warning, particularly if they only heard a part of it.       ~

     Tone-alert Radio/Pagers.   Tone-alert radios are specialized warning
devices that can be remotely activated.  Upon receipt of a code broadcast

-------
                                     -42-

frora a radio transmitter, the radio emits a tone and broadcasts a prerecorded
or read message.  The radio receivers operate on normal electrical power;
some have battery backups.  The advantages of the tone-alert systems include
a quick dissemination time, the combination of an alerting signal with
specialized messages, and round-the-clock availability.  Disadvantages
include maintenance problems, availability during power failures, and limited
utility for warning people outdoors.

     Telephone - Automatic Dialers.  Two types of automatic dialers currently
exist:  switching and computerized dialing equipment.  Switching technology
has been developed that is capable of simultaneously calling hundreds to
thousands of exchanges using automatic switching equipment.  Some systems
will automatically hang up phones in use and block out incoming calls during
the transmission of the emergency message.  These systems play prerecorded
messages, which can be updated fairly quickly, or broadcast messages,
providing timely information.

     The chief advantage of telephone warning systems is the ability to
disseminate a message quickly to people at home.  Automatic dialing systems,
however, are expensive and primarily limited in their use for that reason.
(See Exhibit A 7-7 in Appendix 7.)  Automatic telephone systems are currently
used chiefly within an interorganizational network such as emergency response
personnel or institutional facilities at risk.  Recent developments make this
an attractive option for small communities or for areas of a community where
a prompt warning is needed.                                          .

     7.1.5  Warning Systems in Use

     Warning systems in use can be characterized as follows:

     1.   Enhanced systems that use sirens and some form of specialized
          alerting such as tone-alerts.                                     ;

     2.   Siren-based systems that rely on sirens for alert with use of
          media-based notification.

     3.   Ad hoc systems that rely on media, EBS, and door-to-door or
          route alert.

     Enhanced systems are capable of fast alert and fast notification.
Siren-based systems have the potential for fast alert  (based on coverage)
with notification being more problematic.  Ad hoc systems  require more time;
to implement and reach the public with a message.

     According to the responses to the community questionnaire, ad hoc
methods are the predominant means to warn people in close  proximity to      ,
chemical facilities.  Siren-based systems are used in a third of the
communities that responded to the questionnaire.  A few use an enhanced
system involving both sirens and tone-alert radios.  All of these systems
would be effective in an emergency with a 3- to 4-hour lead time or to
support a precautionary response.  In a rapid moving event, however, the
majority of systems, including siren-based systems, are unlikely to provide
an effective warning.                               .

-------
                                     -43-

     For institutional populations, communities are more likely to use
tone-alert based systems, as well as the telephone.  A lesser reliance is
placed on ad hoc or sirens.  Nevertheless, conventional systems are used in
about one-half of the communities.   A few of the communities have no
provisions to warn institutional populations.  Transient populations such as
tourists are largely ignored as special populations requiring warning.  The
majority of the communities rely on ad hoc or siren-based systems to warn
transients.

     Few communities use the most sophisticated communications equipment or
warning system technologies.  It is clear that some communities do not need
such equipment because the risk does not justify the expense:  In other
communities the differences are more critical because the potential risks are
greater.  Overall, the ability of the majority of systems to provide a timely
alert and notification is questionable, particularly in a rapid-onset event.


7.2  FACILITY/COMMUNITY INTERACTIONS

     Because, in many cases, the facility and the public officials will have
very little time to act in a real emergency, there must be a dialogue between
them beforehand to ensure that, should a release occur, both sides can move
quickly.  Specific individuals at the facility and in the community should be
designated as contact points.  The process is even more effective if each of
the individual contact people has a designated backup.   The communities and
facilities that responded to their respective surveys expressed a similar
level of knowledge about whom to contact:   approximately three-quarters could
name either the person or the position of the contact point.

     Once a community's point of contact receives an alert, a notification
process is set in motion within the community.   Seldom does the person
receiving the notification have, or take on, the authority to issue a warning
to the public.  The efficiency of the community response to the initial
notification will determine the ability to provide an effective and timely
warning.

     A major factor determining the efficiency of the process is knowledge of
what to do following the alert.  Such knowledge may be reflected by a well-
articulated description of the steps to be taken or by the existence of a
standard operating procedure to follow after the notification.  Uncertain
knowledge, or, even worse, no knowledge of what to do will delay or impede an
effective warning.  About half the communities described clear procedures and
half described vague procedures.  Because almost two-thirds of the
communities indicated they have a written warning plan or procedure, it
appears some of the communities may not fully understand the contents of the
plan or were poor in communicating the contents.

     The facility survey addressed the decision-making procedures at
facilities for determining when a release justified notification and for
notifying communities about the release.  The majority of facilities
described rather vague procedures ("we'd call city hall") or did not describe
procedures.  Roughly one-third of the facilities responding to the survey
have clear or standard procedures for decision-making and communications.

-------
                                     -44-                                   ;

     Public officials need to know more than just that an emergency exists to
make a timely warning decision;  they must have specific details^on the
nature of the incident.  The warning to the public must contain clear,
concise instructions on appropriate protective actions.   Section 304 of Title
III requires that facilities provide information to the community when they
notify the community of a release.  The information required includes the
name of the chemical released, an estimate of the quantity released, the time
and duration of the release, the medium or media into which the chemical was
released, acute or chronic health risks and advice about medical attention
for exposed individuals, precautions that should be taken, and the name and
number of someone to contact for further information.

     The survey indicates that many communities do not know what information ,
to ask for in an emergency.  The most common and frequently cited item of
information needed by the community is the type(s) of chemical released or :
involved.  The next most frequently requested information is the size or
amount of material released and the risk to human health.  Relatively few
responding communities indicated a need for information on plume or release;
location, speed of dispersion, potential pathways, or protective action
recommendations.  Still fewer indicated the need for information on facility
response -- what the facility was doing to control the event or whether
community assistance was needed.  Many, however, expressed a general need for
information on what happened at the facility.

     The next stage of the alert process is for the community to reach a
decision about whether to issue a warning to the public.  This may not be a
single decision but a series of decisions regarding precautionary warnings,:
warnings to take protective actions, and warnings that the situation is not,
hazardous or that the emergency has ended.

     Clearly defined procedures can lead to more timely and effective
decisions, provided the appropriate officials are familiar with the plans. ;
About a third of the communities surveyed have clear or standard procedures
for making a warning decision.  The remainder specified rather vague
procedures and, in a few cases, no procedure.  The lack of written or at
least clearly defined procedures in the majority of the communities can be ;
interpreted as another constraint to effective and timely public alert, which
increases the probability of a warning system failure or a delay in issuing
the alert.

     Protective action recommendations are an important part of the        ;
notification procedure; the public expects guidance on what to do, not merely
notification of the danger.  Overall, the use of sheltering as a protective
action strategy is not widely perceived as a viable option when compared to
evacuation.  A large number of communities have an evacuation-only
philosophy, a lesser number a shelter-only policy.  Such policies reduce the
problems in decision-making, but may increase the threat to the public
because evacuation may unnecessarily expose people to chemicals from which ;
they could be protected by remaining indoors.  In addition, in rapid-moving
events, evacuation may not be a reasonable option.  Research is needed to
develop guidelines on when sheltering in-place should be recommended.

-------
                                     -45-
7.3  SYSTEM EFFECTIVENESS

     Because the credibility of a warning message is crucial in the
determination of public response, the source of the warning information is a
critical element of the warning process.   Emergency warnings are more likely
to be effective when multiple sources of warning are indicated, and when
local authorities, political as well as technical, are associated with the
warning message.  Among the communities that responded to the survey, there
was a tendency to overidentify emergency managers and political or management
positions as sources of warning and to underrepresent technical information
sources.

     Preparedness is achieved partly by having plans and procedures, and
partly by "priming" the system.  Priming includes testing equipment and
public education.  Pre-planned messages are also a sign of preparedness.  The
respondents test warning equipment on a fairly routine basis, with most
reporting such tests weekly or monthly.  Some communities even reported
testing warning equipment and procedures daily, but a few reported testing
warning equipment and procedures less often than monthly and some reported no
testing at all.  Written protocols for communications with the general
public, via the EBS or the media, have been prepared in a third of the
communities in the survey, and protocols for institutional facilities are
available in less than one community in five. Few communities have protocols
for foreign-language populations.

     Although little evidence exists that public education makes a difference
as to whether a warning system will function effectively, most respondents
agreed that it contributes to an effective response.  The majority of
communities have either no public information program in place, or a poorly
developed one.

     Most of the communities report that the facility has provided
information describing the hazardous chemicals used at the facility.  Half of
the responding communities reported that their coordination with the chemical
facility includes joint participation in exercises.  About a third of the
communities reported that they had only initial contact or no contact with
the facility in question.

     From the facility perspective, the local fire departments, local
emergency planning or civil defense offices, and local law enforcement
agencies are the three major types of community organizations involved with
chemical emergency alert and notification procedures.  Facilities have varied
levels of interaction with each of the three groups.  The majority of the
facilities that have participated in exercises with such groups reported
having either one or two such events in the past two years.  All of  the
facilities indicated that they had at least some coordination with local fire
departments.

     With respect to management practices, few communities have
well-developed plans and procedures to guide emergency response.  Notably
lacking are capabilities to make decisions.  Both lack of procedures and,
more basically, knowledge about what information is needed to make a
decision, suggest major problems with issuing a timely warning.  Also lacking
are pre-planned warning messages and public information programs.

-------
                                     -46-
7.4  RECOMMENDATIONS

     As the SERCs and LEPCs established under Title III become more active,
they should provide local forums for discussing the types of notification and
alert problems identified by the study and go far toward solving them. .  By
requiring industry to participate fully in the emergency planning process and
to share information on the hazards of the chemicals present at their
facilities, Title III promotes a dialogue that can identify where procedures;
and equipment are inadequate and strengthen the communication links between
the facility and the local community.

     The law specifically requires emergency plans that, among other things,
designate facility and community emergency coordinators and lay out
procedures for providing timely, reliable, and effective notification.  These
plans are to be tested with exercises and updated.  It is important for the
Federal government to continue to support and strengthen the capabilities of
the SERCs and LEPCs by providing technical guidance and assistance in these
areas.

     The improvement of public alert systems is feasible without the
development of new technologies.  The problem of disseminating existing
technology and knowledge is greater at present than the problems created by
the lack of appropriate technology.  Unless new technologies lead to low-cost
equipment that can rapidly alert and notify the public and that can be easily
installed and maintained, further technological advances would only increase
the gap between practices and the available technologies.

     At the local level the feasibility of improvement depends on two
factors.  The first is the dissemination of information on low-cost or
no-cost improvements.  Major improvements in management practices and
procedures can be achieved without major expenditures.  The second is the
availability of funds for improved communication equipment and warning system
equipment.  It is unlikely that all communities have the funds to install new
communication devices or new warning systems.  Improvements in these areas
will require assistance to the communities or cost sharing.  Currently,
improving management practices and developing better decision-making
procedures within both the facilities and the communities appear more
critical than improving technology.  The most sophisticated equipment is
relatively useless if it is not used properly.

     The following recommendations are made:

     •    Standard operating procedures for initial response to alerts,
          warning decision-making, and protective-action recommendations
          should be developed at the local level.  Computerized
          emergency planning and management systems and decision aids
          should be refined and adopted where appropriate.

     •    Local officials should have relevant chemical profiles at hand
          and a means to interpret the data, such as assistance from
          local AIChE chapters that have been established to assist
          emergency response programs.

-------
                           -47-

A standardized information protocol to guide the community in
seeking appropriate information from the facility during the
initial notification should be developed.

Warning message protocols for both English-speaking, and where
needed, non-English-speaking populations should be developed
at the local level.

:The working relationships among personnel at the facilities
and officials within the community emergency response
structure should be improved.  Communities should be
encouraged to conduct frequent exercises.

Public information programs should be strengthened; SERCs and
LEPCs could participate in this effort.

Notification and alert equipment must be maintained and tested
regularly.

Communications equipment within community emergency response
organizations should be updated where feasible.  Communication
technologies between facilities and communities should be
improved to include additional back-up capabilities.

Public warning technologies in high-risk and densely populated
areas should be improved.  Communication links between
communities and institutions such as hospitals should be
improved.

Studies of public response to warnings in chemical emergencies
should be conducted to improve warning systems.

Better information should be developed on sheltering versus
evacuation as a protective action.

-------

-------
Appendix 1
Glossary of Terms/Acronyms

-------

-------
                              GLOSSARY OF TEEMS
Accidental Release.  An unexpected discharge or emission into the
environment, possibly involving a fire or explosion, resulting from
operational errors, improper maintenance, or equipment failure in the course
of industrial activity.

Audit (Process Safety Audit).  An inspection of a plant/process unit,
drawings, procedures, emergency plan and/or management systems, etc., usually
by an off-site team to assess plant safety.

Containment/Control.  A system to which toxic emissions from safety release
discharges are routed to be controlled.

Control System.  A system designed to maintain automatically all controlled
process variables within a prescribed range.

Dump Tanks.  Standby, empty tanks for transfer of chemicals from a
process/storage unit that may be leaking/ruptured or to relieve pressure.

Exothermic.  A term used to characterize the evolution of heat.  Specifically
refers to chemical reactions from which heat is produced.

Extremely Hazardous Substances.  Substances appearing on the list referred to
in Section 302(a) of SARA and published in the Federal Register on November
17, 1986 (51 FR 41570, as revised on April 22, 1987, 52 FR 13378 and Feb. 25,
1988, 53 FR 5574).  The list is composed of acutely toxic chemicals that
might pose an acute or chronic hazard to a community upon release.

Facility.  A location, with, one or more structures, at which a process or set
of processes produce, refine, use, or repackage chemicals, or a location
where a large enough inventory of chemicals are used or stored so that a
significant accidental release of a toxic chemical is possible.

Hazard.  A characteristic of the chemical/system/plant/process that
represents a potential for an accident.

Hazard Analysis Procedures:  Techniques and procedures used to identify
undesired events that lead to the realization of a hazard, the analysis of
the mechanisms by which the undesired events could occur, and usually the
estimation of the extent, magnitude, and likelihood of any harmful effects.

Hazard Evaluation Procedures:  Qualitative and quantitative procedures for
determining what failures or series of events could result in accidental
releases of extremely hazardous substances and their probability of
occurrence.  These procedures should include hazard identification and hazard
analysis procedures.

Hazard Identification Procedures:  Qualitative techniques and procedures used
to survey a plant or process to identify what equipment and procedure
failures or series of events could result in an accidental release.

Monitoring and Detection Systems:  (a)  Technologies ranging from process
instrumentation and in-plant detection devices to perimeter alert devices for

                                    A 1-1

-------
anticipating emergency or upset conditions, identifying constituent chemicals
in releases, providing data on the chemical composition of releases, and
determining concentrations of chemicals in releases; and (b) Modeling
technologies for determining the magnitude and direction of hazards posed by
releases.                                                         	

Perimeter Alert System:   A system or array of detectors placed at the
perimeter of a facility to detect releases of chemicals.                 •

Post-Release Mitigation Systems.  Systems and techniques applied,to an
extremely hazardous substance after the loss of containment has occurred, but
while the substance is still within the plant boundaries, to reduce the
possibility of off-site migration.

Pre-Release Prevention Systems.  Systems and procedures designed to reduce
the probability that the primary containment of an extremely hazardous
substance will be breached.                     ..   :              ...-.-.  :   ,

Pre-Release Protection Systems.  Specific pre-release control techniques that
either contain, destroy, or reduce the quantity of the substance prior...to its
release to the environment.

Prevention Systems.  Any technology or management practice that aids in
preventing accidental releases.  Prevention systems include pre-release."-  ,
prevention systems, pre-release protection systems, post-release mitigation
systems, as well as hazard evaluation techniques.

Primary Containment.  The containment provided by the piping, vessels, and
machinery used in a facility for handling chemicals under normal operating
conditions.

Probability.  An expression of the likelihood of an event or event sequence
occurring.

Process.  The sequence of physical and chemical operations for the
production, refining, repackaging, or storage of chemicals.

Process Machinery.  Process equipment, such as pumps, compressors, heaters,
or agitators, that would not be categorized as piping and,vessels.

Public Alert Systems.  Equipment, technologies, and procedures for providing
timely and effective public warning of an accidental release as well as for
informing the' public of precautionary measures.

Qualitative Evaluation.  An assessment of the risk pf an accidental release
in relative terms, the result of the assessment being a verbal description of
the risk.

Quantitative Evaluation.  An assessment of the risk of an accidental release
in numerical terms, the end result being a number that reflects risk, such as
faults per year.

Reactivity.  The ability of one chemical to undergo a chemical reaction with,
another chemical.  Reactivity of one chemical is always measured in reference
to the potential for reaction with itself or with another chemical.         ;

                                    A 1-2

-------
Real-Time. 'The actual time during which an emergency or event is occurring.

Redundancy.  For control systems, redundancy is the presence of a second
piece of control equipment where only one would be required.  The second
piece of equipment is installed to act as a backup in the event that the
primary piece of equipment fails.

Review (Process Safety Review).  An inspection, for the purpose of assessing
safety, of a plant/process unit, drawings, procedures, emergency plans, and
management systems, usually by an on-site team and usually problem-solving in
nature. "•..-.••'•'•.•                                                  • •"  '<-*

Risk.  A measure of potential economic loss or human injury in terms of the '
probability of the loss or injury occurring and the magnitude of the loss or
injury if it occurs.

Routine Release:  A process emission, such as atmospheric venting, designed
into a process of operation to maintain operational control.

Secondary Containment.  Process equipment specifically designed to contain
material that has breached primary containment before the material is
released to the environment and becomes an accidental release.

Shortstops.  A chemical that when added to chemical reactants will stop any
further reactions.
                                     A 1-3

-------
                                  ACRONYMS






AIChE -- American Institute of Chemical Engineers.




ARIP --  Accidental Release Information Program.




CABR --  Community Awareness and Emergency Response Program.




DOE --  U.S. Department of Energy.




DOT --  U.S. Department of Transportation.




BBS  --  Emergency Broadcast System.




EOC  --  Emergency Operations Center




EPA  --  U.S. Environmental Protection Agency.      !




FEMA  --  Federal Emergency Management Agency.




LEPC --  Local Emergency Planning Committees.




OSHA -- Occupational Safety and Health Administration.




ERA. --  Probabilistic Risk Assessment.




SARA  --  Superfund Amendment and Reauthorization Act of 1986,




SERG  --  State Emergency Response Commission.




TPQ  --  Threshold Planning Quantity.
                                    A 1-4

-------
Appendix 2
Survey Responses

-------

-------
                               SURVEY RESPONSES
Facility Survey

     As of February 16, 1988, 146 of the 522 facilities sent surveys had
returned completed questionnaires; because one facility returned two
questionnaires, for different chemicals, the number of completed
questionnaires is 147.  Two facilities returned incomplete questionnaires;
four other facilities returned completed questionnaires marked as
confidential business information.  No information from these six
questionnaires is included in the data base.

     Of the 372 questionnaires that were not completed, 72 facilities did not
complete the questionnaire because they no longer use the chemicals being
considered or use them in quantities below the threshold planning quantity
(TPQ).  Another 55 facilities, while they were listed in the data bases, had
either gone out of business or EPA was unable to contact them by mail or by
follow-up calls.

     Exhibit A 2-1 provides a breakdown of the facilities surveyed by
chemical, and the response rate.  The responses are concentrated in a
relatively small number of the chemicals, which is partly a function of the
normal density of facilities throughout the U.S. using the various chemicals.

     The respondents included large chemical manufacturers,  small chemical
manufacturers, refineries, food manufacturers, chemical distributors, batch
process plants, and water treatment plants.  The smallest plant that
responded had six full-time employees.   Exhibit A 2-2 is a summary of
responses by facility size (number of employees).

     As can be seen from Exhibit A 2-3, the responding facilities are
concentrated in the Atlantic Coast,  Great Lakes, and Gulf Coast regions, with
comparatively few facilities in New England and the Midwestern and Western
regions of the U.S.   This distribution is,  in part, a function of the normal
density of industrial facilities in the various regions.

     Exhibits A 2-4 and A 2-5 summarize the distribution with respect to age
of the facilities, with both the date of the construction and the date of the
latest modification.   Although the majority of respondents were older
facilities,  most indicated recent modifications.
                                    A 2-1

-------
                                  EXHIBIT A 2-1

                          SURVEY RESPONSES BY CHEMICAL
                               Number of Facilities
Chemical Name
Requested to
  Complete
Questionnaire
 Could Not
Be Contacted
    Were
Not Eligible
Number of,
Completed
Questionnaires
Acrylonitrile
Ammonia
Benzenearsonic Acid
Benzotrichloride
Chlorine
Chloroacetic Acid
Furan
Hydrazine
Hydrogen Cyanide
Hydrogen Fluoride
Hydrogen Sulfide
Mechlorethamine
Methiocarb
Methyl Bromide
Methyl Isocyanate
Phosgene
Sodium Azide
Sulfur Dioxide
Sulfur Trioxide
Tetraethyl Tin
Trichloroacetyl Chloride
23
86
3
4
94
34
5
9
15
36
56
2
12
18
6
19
20
52
17
3
8
1
12
0
0
2
7
0
0
3
5
6
0
0
3
1
2
3
7
2
0
1
5
2
1
1
4
; 7
0
1
2
6
7
2
5
4
1
: 2
6
' 7
4
0
5
7
25
0
2
38 V
7
1
4
( -• H»
3
7 • ,- • '
20
0
1 .'•':' .
6 ; '
2 ; . '
6
1 •. .- '
13 ;-;
4
0 ;
0
Totals
       522
       55
    72
    147'
     One facility completed two questionnaires, one for chlorine, one for
     ammonia.

                                    A 2-2

-------
            Number of Facilities
     ooooooooooo
   en
   o
 3
 Q-
 (D
   CO

   CD
o


1 8
CO O
   CO
   01
   o
   o
                                               CO




                                               CD
                                               CD

                                               (0
   m

   a
0) g

0) £
Tl

fi>
O
                                               CD

-------
                                       Exhibit A 2-3
                           Facility Responses by Location
LEGEND
Region 1
Region 2
Region 3
Region 4
Region 5
Region 6
Region 7
Region 8
Region 9
Region 10
CT, ME, MA, NH, Rl, VT
NJ,NY
DE, MD, PA, VA WV
AL, FL, GA, KY, MS, NC, SC, TN
IL,IN,MI,MN,OH,W1
AR, LA, NM, OK, TX
1A, KS, MO, NE.
CO,MT,ND,SD,UT,WY
AZ, CA, HI, NV
AK, ID, OR, WA
S84112-2a

-------
                   Number of Facilities
             ooooooooooo
g
i
           CO
           a
           o
CO
o>
IVD
           CO
           o>
           rp
           CO


           -vl
        o
        s.
        CD
           CO
           CO
I

00
           CO
00
00
                                                         o
                                                         flfl
                                                         JD
                                              o
                                              o

                                              CO
                                              — c
                                              o  s
                                              3  >

                                              Q  •?
                                                         o
                                                         o
                                                         a
                                                         O

-------
Number of Facilities
                               o
                               s>
                               «•*•
                               

                               a-

                               Tl

                               O
                               O
                               (D
                               (0
                               CO
                               
-------
Community Survey

     FEMA sent 277 questionnaires to communities identified as having
jurisdiction over one or more of the facilities selected for the facility
survey.  Of these, 142 communities returned completed questionnaires.
Twenty-three returned the survey because the facility in question no longer
was covered by the jurisdiction or no longer existed.  Of the communities
that responded, about 40 were adjacent to one of the 152 facilities that
returned the facility survey.  Exhibit A 2-6 shows the facility response by
geographic location.

     The number of households in an area, the population distribution, and
the population density are relevant aspects of the warning problem.  The
chemical facilities selected for the survey are located in fairly densely
populated areas.  An average of about 4,400 people reside within a 1 mile
radius of the facilities.  In one case, 35,000 people are estimated to live
within 1 mile.  On average, the population density is about 1,370 people per
square mile assuming a uniform distribution, but because about 30 percent of
the area within 1 mile of the facilities is in residential use, the areas
that are populated actually have a greater density.  Within 5 miles of the
facilities, the average population is 42,600 with an average density of 550
people per square mile.  The maximum population within 5 miles of a facility
is estimated at 600,000.  In several cases, however, the population within 5
miles is close to zero.  Overall, however, the population density within 1
mile of the plant is much greater than within 5 miles.  Exhibit A 6-7 gives
the distribution of community responses by size of the community.

     Another important characteristic is population fluctuations.  These can
occur on a diurnal cycle; for example, a large number of commuters may be
present during the day, but not at night.  Conversely, suburban areas may
have their peak populations at night.  Weekly fluctuations may occur on
certain days as in the case of a recreation area which is crowded on
weekends.  Finally, seasonal variation can be significant.  Diurnal
population fluctuations are reported at roughly one-half of the sites.
Weekly population fluctuations are more uncommon, occurring in 29 percent of
the communities surveyed.  Seasonal population fluctuations are reported in
only 15 percent of the communities..

     A third important population characteristic involves special population
subgroups.   Institutional populations include people in schools,  hospitals,
correctional facilities, nursing homes and other facilities in which people
cannot fully care for themselves.  They require special warning
considerations because they require assistance and, hence, a longer time to
respond to an emergency.  About 83 percent of the communities have
institutionalized populations within 5 miles of the facilities.  Several have
facilities in numbers sufficiently large that respondents would report that
there are many or numerous but not report the actual number, while others may
have only a few.  Schools are the most common facility (64 percent of the
communities) reported, followed by hospitals and nursing homes (41 and 45
percent, respectively).  Twenty-five percent of the communities report having
prisons or jails.  All such institutional facilities require special and
rapid warning consideration in order to give them the extra time needed to
respond.
                                    A 2-7

-------
                                  Exhibit A 2-6
                     Community Responses by Location
LEGEND
Region 1
Region 2
Region 3
Region 4
Region 5
Region 6
Region 7
Region 8
Region 9
Region 10
CT, ME, MA, NH, Rl, VT
NJ,NY
DE, MD, PA, VA WV
AL, FL, GA, KY, MS, NC, SC, TW
IL, IN, Ml, MN, OH, Wl
AR, LA, NM, OK, TX
IA, KS, MO, NE
CO,MT,ND,SD,UT,WY
AZ, CA, HI, NV
AK, ID, OR, WA
S84112-la

-------
                                          Exhibit A 2-7
                          Distribution of Sample Communities
    Population
    Category

                 < 1000
                                    10      15      20      25

                                     Number of Communities
S84112-la

-------

-------
Appendix 3
Hazard Evaluation Procedures

-------

-------
                     AIChE HAZARD EVALUATION PROCEDURES
Process/system checklists contain information concerning specific items in
process design, construction, and operation and provide a means for
standardized evaluation of plant hazards.   They are a convenient means of
communicating the minimally acceptable level of hazard evaluation that is
required for any job.

Safety review is also referred to as a Process Safety Review, a Loss
Prevention Review, or a Process Review.  It is intended to identify plant
conditions or operating procedures that could lead to an accident.  A safety
review involves inspections of plant equipment and procedures by a team of
specialists.

Dow and Mond Index provides a method for relative ranking of the risks in a
chemical process plant.  The method assigns penalties to process materials
and conditions that can contribute to an accident.  Credits are assigned to
plant safety procedures that can mitigate the effects of an accident.

Preliminary Hazard Analysis is intended for use only in the preliminary phase
of plant development for cases where past experience provides little or no
insight into potential safety problems.  The analysis focuses on the
hazardous substances and on major plant elements.

What If Analysis is used to consider carefully the result of unexpected
events that would produce adverse consequences.  The method involves the
examination of possible deviations from intended design, construction,
modification, or operating procedures of a process.  It requires a basic
understanding of what is intended and the ability to combine or synthesize
possible deviations that would cause an undesired result.

Hazard and Operability Studies are conducted by teams that brainstorm to
identify "hazards and operability problems by searching for deviations from
design intents.  The team considers the causes and consequences of a
deviation.

Failure Modes. Effects, and Criticality Analysis is a tabulation of the
system/plant equipment, their failure modes, each failure mode's effect on
the system/plant, and a criticality ranking for each failure mode.  The
failure mode is a description of how equipment fails.

Fault Tree Analysis is a deductive technique that focuses on one particular
accident event and provides a method for determining causes of the event.
The fault tree is a graphic model that displays the various combinations of
equipment faults and failures that can result in the accident event.

Event Tree Analysis considers operator response or safety system response to
an initiating event in determining accident outcome.  The results are
accident sequences.

Cause-Consequence Analysis is a blend of fault tree and event tree analysis.
The cause-consequence diagram displays the interrelationships between
accident outcomes and their basic causes.
                                    A 3-1

-------
Human Error Analysis is a systematic evaluation of the factors that influence
the performance of human operators, maintenance staff, technicians, and other
personnel.  It will identify error-likely situations that can cause an
accident.
                                    A 3-2

-------
                                EXHIBIT A 3-1

  NUMBER OF FACILITIES USING AIChE-REGOGNIZED HAZARD ASSESSMENT TECHNIQUES
Techniques
Facilities'
"What If" Analysis

Hazard and Operability Studies

Fault Tree Analysis

Failure Modes,  Effects,  and
Criticality Analysis

Preliminary Hazard Analysis

Event Tree Analysis

Human Error Analysis

Dow and Mond Hazard Indices

Cause-Consequence Analysis
    42

    36

    14

     8


     8

     4

     1

     1

     1
*  Many facilities listed more than one method.  A total of 73 of the 146
facilities listed at least one AIChE-recognized method.
                                    A 3-3

-------

-------
Appendix 4
Survey Responses on Prevention

-------

-------
                                EXHIBIT A 4-1




                       FACILITIES CITING SCRUBBERS

Chemical
Ammonia
Chlorine
Hydrogen Cyanide
Hydrogen Fluoride
Hydrogen Sulfide
Sulfur Dioxide
Facilities Citing
Facilities Responding Scrubbers
25 5
37 18
3 2
7 3
21 7
13 10

                                EXHIBIT A 4-2




              METHODS OF DETECTING DISABLED CONTROL DEVICES
Method of Detection
Number of Facilities Citing
Monitors




Routine Inspection




Operator Observation




Alarms




Routine Testing
           40




           45




           42




           36




           15
                                  A 4-1

-------
                                EXHIBIT A 4-3




             SPECIFIC MONITORS REPORTED  FOR RELEASE DETECTION
Monitors
Number of Facilities Citing
In-Stack



Ambient Air



Sewers



Surface/waste water



Ground Water
            18




            52




            12




            33




             4
                                   A 4-2

-------
Appendix 5
Prevention and Protection Measures

-------

-------
                                 APPENDIX 5
     This appendix consists of two exhibits.  Exhibit A 5-1 identifies
general prevention/protection measures for a broad range of hazards
associated with a variety of hazardous chemicals.  Exhibit A 5-2 identifies
additional measures that apply specifically to the 21 chemicals used as a
focus for this review.  The measures identified are general in scope and not
intended to be all inclusive.  The appropriateness of any measure is
facility- and process-specific.
                                    A 5-1

-------
                                EXHIBIT A 5-1         ;

                  EXAMPLES OF PREVENTION/PROTECTION MEASURES
                    THAT APPLY TO MANY HAZARDOUS CHEMICALS
Hazard Area
Prevention/Protection Measures
Fire/explos ion
Equipment failure:
     general, vessel
     storage,
     intermediate
     tanks  and  lines
Pipes  and valves
 Seals  and fusible plugs
Isolate equipment from ignition or fire sources
and flammable materials; inert gas padding to
prevent air/oxygen to prevent air/oxygen or water
levels; explosion-proof electrical equipment;
proper grounding for static discharges; protect
wire and cable feed for icritical controls from
fire damage; fireproof control valves; heat
shielding of critical equipment and structure;
automatic sprinkler systems, deluge nozzles, foam
application; non-sparking tools and equipment;
temperature controls, interlocks, alarms; "fail-
safe"  (e.g., a valve closes to shut off flow when
power  is lost) controllers and equipment;
explosive gas, heat, flame monitors, and alarms;
periodic drills, testing,inspection of fire      ;
equipment; written fire response procedures; vent
flame  arresters, conservation vent; blast doors.

Ensure proper design; materials on construction;
equipment hydrostatic and pressure testing and
inspection procedures;  leak checks; preventive
maintenance; corrosion, erosion monitoring;
proper installation; design for vibration, thermal
expansion; secondary containment; berms; diking;
emergency drain pumps,  foam injection.

In addition to general  items: excess flow valves,
flange and valve seal guards; double piping; non-
flanged, non-threaded (all welded) pipe; valve
testing.

In addition to general  items: purge pump seals  to
prevent escape of material; use double mechanical
seals; test and inspect! fusible plugs; install
containment around fusible plugged equipment.
                                     A 5-2

-------
Hazard Area
Prevention/Protection Measures
Overpressure
Pumps,  critical
     mechanical
     equipment
Process changes
Loss of heating/
     cooling, flow,
     inerting, power,
     air, mixing
Overfilling
Contamination, runaway
     reaction
Human error
In addition to general items: proper relief vent
design; vents free from restrictions; system
operating procedures; alarms and interlocks for
pressure and level;  redundant controls;  vent
relieved vapors to scrubbers/flares to prevent
release to atmosphere.

In addition to general items:  standby backup
equipment; regular testing to ensure operability
of backup equipment; additional preventive
maintenance for critical devices.

Hazard analysis prior to change; written
"experimental operating conditions" procedures
that document non-standard operation until changes
are complete and standard; increased inspection
and diligence during changeover.

Temperature, pressure, flow monitors; alarms;.
interlocks; backup pumps, generators, instrument
air cylinders, heating, cooling systems; fail-safe
equipment; monitor cooling water for process
contaminants; non-mechanical spargers to maintain
mixing.

Vessel level monitors, alarms, interlocks; diking,
secondary containment; operating procedures;
neutralization of spills; use of absorbent
materials, foams.

Contaminant monitoring, alarms, interlocks;
quenching systems; reactor dumping vessels; deluge
systems; specific operating parameters,  controls;
standard operating conditions; isolation of
possible contamination sources (blanking);
backflow prevention  (checkvalves).

Training and retraining; testing; job safety
audits; j ob performance checks; standard j ob
procedures; checklists; corrective actions;
redundant backups (e.g., "dead-man"  switch);
training on process  simulators; safety policies,
programs, procedures.
                                    A 5-3

-------
Hazard Area
Prevention/Protection Measures
External impacts:
     lightning,
     flood, tornado,
     earthquake,
     collision with
     vehicles or
     other equipment
Isolate pipes,  vessels, etc.  from high traffic  :
areas; install barricades;  install proper grounds
and lightning rods;  emergency shutdown procedures
during dangerous situations;  flood walls and    '
drains for high water; earthquake-designed
equipment footing; wind protection.
Note:  Use of any of these measures should result from detailed hazards
analyses of specific operations.  A variety of prevention measures may be
used to establish an overall prevention strategy for!a specific operation.
                                    A 5-4

-------
                                               EXHIBIT A 5 - 2

                  EXAMPLES OF SPECIFIC EEEVEHTIOH/EEOTECTION MEASURES FCR THE 21 CHEMICALS
Chemical
                             Problem Area
                                                                     Comments
Ammonia
                             Atmospheric releases from processes,
                             relief discharges.
Emergency vent scrubbing to remove
ammonia.
                             Liquid choked vapor lines;
                             overpressurization.
Ensure proper heating for vaporization
control; temperature monitoring with
interlocks and shutdowns; excess flow
control with restricted orifices.
                             Corrosive to copper and galvanized
                             surfaces.
Ensure proper materials of construction;
monitor corrosion.
                             Reactivity
Mixing several chemicals can cause severe
fire hazards and explosions.  Ammonia in
container exposed to heat of fire may
explode. Avoid contamination with
backflow prevention, isolation; cool fire
exposed containers.
Chlorine
                             Extremely corrosive in presence of
                             water.
Moisture monitoring; ensure proper
materials of construction; corrosion
monitoring; startup and maintenance
procedures to ensure dry conditions; pad
gas dryers and monitors; prevent
backflow.
                             Reactivity
Reacts vigorously with most metals at
high temperature; control temperature,
materials of construction, avoid copper
and aluminum for high temperature
applications or potential; incompatible
with plastics and rubber; mixture with
fuels may cause explosion; hydrogen and
chlorine mixtures are exploded by almost
any form of energy; avoid contact with
fuels and hydrogen.
                             Liquid choked vapor lines;
                             overpres surization.
Ensure proper materials of construction;
monitor corrosion.
                             Atmospheric releases from processes,
                             relief discharges.
Emergency vent scrubbing to remove
chlorine.
                                                    A 5 - 5

-------
Chemical
                             Problem Area
                                                                     Comments
Hydrogen cyanide
Runaway reactions from excess ECN
Redundant flow control; flow limits by
line or orifice size; flow interlocks;
concentration monitoring and interlock.
                             Atmospheric release from processes,
                             relief discharges
                                        Emergency vent scrubbing to remove HCN.
                             Fire hazards
                                        Vapor may explode if ignited in a
                                        confined area, may become unstable and
                                        subject to explosion it stored for
                                        extended time or exposed to high
                                        temperature and pressure; see general
                                        fire hazard precautions.  Unstabilized
                                        material may polymerize spontaneously
                                        with explosive violence, explosion
                                        potential severe when exposed to heat or
                                        oxidizers, may form explosive mixtures
                                        with air; ensure isolation from
                                        oxidizers, maintain storage stability;
                                        keep containers cool in fire situations;
                                        use;nitrogen padding for vessels.
Hydrogen fluoride
                             Water contamination
                                        Prevent backflow; continuous moisture
                                        monitoring; corrective action procedures
                                        for wet feedstocks.
                             Atmospheric release  from processes,
                             relief discharges

                             Reactivity
                                        Emergency vent scrubbing to remove HF.
                                        Will attack glass, concrete, and certain
                                        metals, e.g. those containing silica;
                                        attacks natural rubber, leather, many
                                        organic materials; may generate flammable
                                        H2 gas in contact with some metals; watch
                                        material of construction; monitor for H2.
                                        When heated, highly corrosive fumes
                                        emitted, corrosive actions on metals may
                                        generate H2; keep vessels cool if exposed
                                        to heat or flame.
                                                    A 5-6

-------
Chemical
                             Problem Area
                                                                     Comments
Hydrogen sulfide
                             Smell sensitization
                                        If odor'used to detect leaks,  consider
                                        monitoring equipment;  workers  may become
                                        desensitized to odor.
                             Fire/explosion
                                        Avoid ignition sources in contained
                                        areas.   Forms explosive mixtures with air
                                        over a wide range;  reacts explosively
                                        with several halogenated organics,
                                        inorganics; avoid contact with these and
                                        eliminate air in process systems; use
                                        nitrogen padding.
Methyl isocyanate
Reactivity
Highly reactive in water, acid, alkali,
amine, iron, tin, copper, thier salts,
other catalysts; watch material of
construction, moisture, contaminants.
                             Polymerization
                                        May exothermically autopolymerize;  heat
                                        produced raised pressure; may rupture
                                        containers; continually sample
                                        containers, monitor quality.  Control
                                        temperature, install redundant controls,
                                        alarms, backups.
Phosgene
Reactivity
Decomposes with water (not vigorously);
control moisture content; use padding to
prevent moist air inlets.  Reacts
violently with aluminum, isopropyl
alcohol, other organics; ensure proper
materials of construction, avoid
contamination, backflow; isolate phosgene
processing streams.
                             Atmospheric release from processes,
                             relief discharges
                                        Emergency vent scrubbing to remove
                                        phosgene.
 Sulfur dioxide
                             Reactivity
                             Atmospheric release from processes,
                             relief discharges
                                        Explosive with several chemicals,
                                        incompatible with aluminum and some
                                        metals. Reacts with water to form toxic
                                        and corrosive fumes; avoid contact with
                                        incompatibilities; ensure materials of
                                        construction; maintain dry process
                                        systems.

                                        Emergency vent scrubbing to remove SO,.
                                                    A  5-7

-------
Chemical
                             Problem Area
                                                                     Comments
Sulfur trioxida
                             Reactivity
                                        Combines with,, water explosively to form
                                        sulfuric acid;  may ignite combustible
                                        materials; explosive increase in vapor
                                        pressure when alpha form melts, flammable
                                        poisonous gases may form in tanks and
                                        hopper cars; on exposure to air, absorbs
                                        moisture, forms dense white fumes.  Avoid
                                        contact with moisture; keep containers
                                        cool to control melting; use redundant
                                        controls; control ignition sources around
                                        tank vents; use proper vent system to
                                        control vapor concentrations.
Acrylonitrile
Reactivity/Explosion
Attacks copper and, in high
concentrations, aluminum. Avoid contact
with strong acids, amines, alkalis,
strong oxidizers, copper and copper
alloys, aluminum.  Ensure proper
materials of construction; ensure
replacement equipment contains no copper.
                              Self-polymerization in presence of
                              alkali,  heat,  light,  lack of oxygen
                                        Avoid exposure to light, strong bases;
                                        ensure proper temperature monitoring and
                                        control, with backup; frequent sampling
                                        andicorrective action procedures;
                                        inhibitor addition and control; air
                                        padding.
Benzonearaonic  acid
Beating/Decomposition
Emits poisonous fumes of arsenic when
heated to decomposition; ensure proper
temperature control with backups; sample
for 'material integrity; keep vessels cool
in fire and heating situations.
B onzotrichloride
                              Reactivity
                                        May react violently with water;
                                        hydrolysis in water forms corrosive
                                        benzoic  and hydrochloric acids; fire may
                                        produce  irritating or poisonous gases;  :
                                        flammable/poisonous gases may accumulate
                                        in tanks and hopper cars; materials may
                                        ignite combustibles;  Monitor for
                                        moisture; watch materials of
                                        construction; corrosion control; control
                                        ignition sources around tank vents; use
                                        proper vent system to control vapor
                                        concentrations; use dry chemical or foam
                                        for fires; keep containers  cool.
                                                     A 5-8

-------
Chemical
                             Problem Area
                                                                     Comments
Chloroacetic acid
                             Equipment pluggage
                                        Flow sensors with alarms; protection
                                        pumps to prevent overheating or
                                        overpressure in loss of flow; steam or
                                        electric trace lines to keep material
                                        molten.
                             Heating/fire
                                        When heated to decomposition,  emits
                                        highly toxic chlorine and phosgene.
                                        Water may cause frothing if it gets below
                                        surface and turns to steam;
                                        Flammable/poisonous gases may accumulate
                                        in tanks and hopper cars; some gases may
                                        ignite combustibles; use redundant
                                        cooling to ensure temperature control;
                                        apply water fog gently to make foam to
                                        extinguish fire.
Fur an
                             Reactivity
                                        Very dangerous when exposed to heat or
                                        flame; may form unstable peroxides on
                                        exposure to air; contact with acids can
                                        start a violent reaction; avoid contact
                                        with strong acids and oxidizing agents;
                                        use redundant temperature sensors; proper
                                        mixing and backup; redundant cooling
                                        systems; use nitrogen pad storage
                                        vessels; ensure air free system.
Hydrazine
Thermal decomposition or exothermic
reaction.
Avoid contact with copper, cobalt,
molybdenum, and iron oxides 
-------
Chemical
                             Problem Area
                                                                     Conxnents
Mothiocnrb
Fire/explosion
Dust may be explosive; avoid dusty
conditions and potential ignition
sources.
Methyl bromide
Heating
Sodium azide
Reactivity/Stability
When heated to decomposition, emits    ;
highly toxic fumes.  Use redundant
temperature control and monitoring to
ensure heating to less than
decomposition.

Stable unless in contact with acid; avoid
contact with acid solutions; forms
explosion-sensitive materials with some
metals such as lead, silver, mercury,
copper.  Check material of construction.
Tatraethyltin
Reactivity
Trichloroacetyl chloride     Reactivity
Avoid contact with strong oxidizers; upon
decomposition, emits acrid smoke and
fumes; isolate process from strong
oxidizers, control heat, use backup
cooling and temperature control.

May react violently with water.  Avoid
water entry into process streams use
foam, dry chemical, for fire control.
                                                  A 5 - 10

-------
Appendix 6
Monitoring and Detection

-------

-------
               COST ESTIMATES FOR PERIMETER MONITORING SYSTEMS
     Perimeter monitoring systems were defined for the purposes of cost
estimation as integrated detector networks in place at a facility fenceline
to provide continuous, real-time measurement of releases of specific
chemicals from the facility.  Industry representatives and equipment vendors
indicated that chemical detectors or monitoring stations for systems in use
are placed between 250 and 1,000 feet apart along the facility perimeter,
depending on the local topography, local meteorology, distribution of
potential release sources and receptors,  and other factors.   For the purposes
of cost estimation, it was assumed that detectors are placed 500 feet apart
along the facility perimeter.

     Equipment, installation, and operating/maintenance costs were estimated
for three types of commercially available chemical detectors: fluorescent
sulfur dioxide analyzers, photo-electric tape sensors, and electrolytic
chlorine detectors.  Photo-electric tape sensors can be used to detect
chlorine, phosgene, hydrogen sulfide, and ammonia.  Cost estimates are based
on information obtained from the Section 305(b) questionnaires and site
visits,  as well as follow-up contacts with industry representatives and
equipment vendors. The estimates were prepared for small and large
facilities.  A large facility is. assumed to have an area of 640 acres (one
square mile) and approximately 21,000 feet of linear perimeter, requiring a
total of 42 monitoring stations.  A small facility is assumed to have an area
of 30 acres and approximately 5,000 feet of linear perimeter, requiring a
total of 10 monitoring stations.

     The annualized costs of the perimeter monitoring systems were calculated
by discounting the total equipment and installation cost of the systems using
a discount rate of 5 percent and adding the annual operating and maintenance
cost.  A system operating life of 10 years was assumed, and no economies of
scale were assumed for systems with large numbers of detectors.  A salary
plus overhead rate of $50,000 per year was assumed for a technician to
operate and maintain the monitoring system.


[Note: The comments in this appendix should not be interpreted to imply
endorsement of any specific detection device or model by EPA.]
                                    A 6-1

-------
                                EXHIBIT A 6-1

       NUMBER OF PROCESS AREA AND PERIMETER MONITORING SYSTEMS REPORTED
Chemical Name

Acrylonitrile
Ammonia
Number of Completed
Questionnaires
7
25
Benzenearsonic acid 0
Benzotrichloride
Chlorine*
Chloroacetic acid
Furan
Hydrocyanic acid
Hydrazine
Hydrogen fluoride
Hydrogen sulfide
Mechlorethamine
Methiocarb
Methyl bromide
Methyl isocyanate
Phosgene
Sodium azide
Sulfur dioxide**
Sulfur trioxide
Tetraethyltin
2
38
7
1
4
3 .
7
20
0
1
6
2
6
1
13
4
0
Trichloroacetyl Chloride 0
Total Reported
147
Process Area
Monitors
5
4 ;
0
0 >
19
0
0
3
0
0
13
0
0
0
1
5 !
0
3
1
0
o ;
54
Perimeter
Monitors
0
0
0
0
3
0
0
0
0
0
2
0
0
0
0
0
0
3
0 :
0
0
8 i
*
One facility reporting ammonia also provided information on their
process area and perimeter chlorine monitoring systems.

Several facilities reporting sulfur dioxide had from one to four sulfur
dioxide monitors located off-site in the surrounding communities.  These

                               A 6-2          !

-------
                                EXHIBIT  A 6-2

                     MAJOR METHODS USED BY GAS ANALYZERS
METHOD
                      DESCRIPTION
Golorimetric
Electrochemical
Spectrometric
Chromatographic
Indicator tubes
Reaction between toxic gas and chemically sensitive
solution or coating causes measurable color change.

Chemical or physical properties of toxic gas changes the
electrical parameters of an input sensor.
Electrochemical measurements include conductivity,
potentiometric,  and coulometric.

Light beam is altered by contaminant gases.
Spectrometric methods include infrared, Raman, microwave,
and laser-induced detection atmospheric radiation
(LIDAR).

Instrument separates and measures gas concentration using
various detectors including thermal conductivity, flame
ionization, flame photometric (^S, S02) nitrogen-
phosphorus, electron capture (Cl2, HF), and
photoionization.

Toxic gas pumped through the detector tube produces a
stain length proportional to the toxic gas concentration.
                                    A 6-3

-------
                                                                              6—3


                                                                     OF DEXBCXXGK DEVICES
Instrument
                                Description
                                                              Gases
                                                              Benefits/Capability
                                                                                                                            Limitations
Photovac 10SSO
(Foxboro Corp.)
Model PI 101 Hazardous Waste
Detector
(HNU System Inc.)
Model 128 Century Organic
Vapor Analyzer
(Foxboro Corp.)
Detector Tubes
(Draeger)
Combustible Gas Indicator
Automated gas chromatograph
with photoionization
detector; direct gas
sampling.
Automated gas chromatograph
with photoionization and
electron capture detectors;
direct gas sampling.
Either in total organic or
gas chromatograph mode
(flame ionization
detector).
Chemical stains the
chemical sensitive
substance in tube.
                                Caloric potential;  measures
                                increased heat on platinum
                                filament.
Low volatile organics and
inorganics such as
phosphine, arsine, ammonia,
nitric oxide, benzene,
hydrogen sulfide.

Organic and inorganic
vapors.
Non-methane compounds in
total organic mode;
separates most organic
components in GC mode.
Chlorine, hydrogen
chloride, hydrogen cyanide,
hydrogen sulfide, methyl
bromide, methylene
chloride, sulfur dioxide,
etc.

Combustible gases.
Self-contained, portable
unit; ppb for hydrogen
sulfide, benzene.
Battery operated and
portable.  0.1-2000 ppm
detection range; accuracy
within 4 percent at Sppm.
0.2 ppm detection limit.
Easy to use; rapid results;
inexpensive^-  .  .-  . .   -  .
                                                              Self-contained,  battery
                                                              "operated.
                               Short term puff can
                               saturate; sensitive to
                               weather changes; requires
                               experienced analyst to
                               operate and interpret data.

                               Hot gas-specific; weather
                               sensitive; experienced
                               analyst required; short
                               puff can saturate; frequent
                               calibration required.

                               Not gas specific for
                               organics; calibration
                               required; experienced
                               analyst needed; must be
                               within inches of source in
                               total organic mode;
                               saturated by short puff.

                               Each tube identifies only
                               one gas; operator must know
                               which gas has been
                               released; subject to
                               interference; low accuracy.
                               Filament damaged by
                               silicons, halides,
                               tetraethyl lead; invalid
                               readings under oxygen
                               deficient conditions; must
                               be calibrated before use.
MIRAN IBX Infrared
Spe ctrophotometer
(Foxboro Corp.)
Air analyzer using a single
beam spectrophotometer;
infrared spectrometric.
Over 100 gases.
Self-contained, battery
operated; designed to
quantify 1 or 2 components;
0-100 ppm.
                               Expensive.  Subject to
                               interference by water vapor
                               and COg; requires
                               experienced operator.

-------
Instrument
                                Description
                                                               Gases
                                                              Benefits/Capability
Limitations
555 Continuous Color
Analyzer (CEA)
Colorimetric; reagent
solutions change color in
presence of gases.
ci,, HCI, m
al?
:_, SO,, HCH, et
O £1
0-100 ppm for most gases;
a-100 ppb for HCH; self-
contained, battery
operated, portable;
accuracy within 1 percent.
Each gas requires
modification to system;
requires skilled operator;
subject to interference.
HS-7 Unit
SC-7 Unit (GasTech)
Electrochemical sensor.
H2S, N02, 'S02, C12, HCI.
                                                              Calibrated only once a
                                                              year;  less susceptible to
                                                              corrosion and
                                                              contamination;  continuous
                                                              or intermittent
                                                              measurements;  self-
                                                              diagnostic capability;  0-
                                                              100 ppm range;  accuracy
                                                              within 3 percent.
Subject to interference;
each gas requires separate
unit.
PPM — parts per million
PPB — parts per billion

-------
                                                                     EXHIBIT A 6-4
                                                  AIR DISEEESICR MODELS CITED BY FACHITY UiSiUUlUllS
System/Model Kama
Capability
Description/Requirements
                                                                                                       Cost Estimate
                                                                                                                                     Facilities
System Approach for             Real-time
Emergency Response (SAFER)
Texas Episodic Model (1EM)      Planning
Industrial Source Complex       Planning
Short Term (ISCST)
DEGADIS                         Planning
Complex^ Hazardous Aix   - •• -    -Real-time
Release Model (CHARM)
Puff                            Planning
Texas Climatological Model      Planning
(TCM)
Hazard Assessment System for    Planning
Toxic Emissions (HASTE)
                         Employs standard Gaussian dispersion
                         modeling equations; can be linked for real-
                         time meteorological data; can model various
                         types of release scenarios

                         A Gaussian plume model; focuses on short-
                         term releases of nonreactive pollutants;
                         accepts meteorological data; operates on
                         IBM PC.

                         Uses Gaussian dispersion equations to model
                         chemical releases primarily for disaster
                         analysis.  Operates on PC with
                         meteorological data inputs.

                         Developed for Coast Guard to predict
                         contaminant movement for heavier than air
                         gases; capable of modelling instantaneous
                         and continuous ground level releases.

                         Includes- chemical data base and map
                         editors; capable of mapping concentration
                         isopleths; allows real-time meteorological
                         data input.

                         Applicable only to five neutral or buoyant
                         gases (Gaussian puff dispersion); designed
                         to incorporate type of short-term release,
                         emission rate, facility characteristics,
                         and weather data.

                         A olimatological steady-state Gaussian
                         plume model for determining long-term
                         average pollutant concentrations of non-
                         reactive pollutants.

                         Intended for planning; includes models and
                         graphic presentation; no weather inputs.
                                               About $75,000;  varies with
                                               choice of hardware and
                                               needs of facility
                                               $80
                                               $100-150
                                                                                                       Public Domain
                                               $9,500-15,000
                                               $200-450
                                               $100-150
                                               $78,600-basic system;
                                               $50,000-added options
                                                                                                                                        20
                                                                                                                                        10

-------
System/Model Name
Capability
Description/Requirements
                                                                                                        Cost Estimate
                                                                                                                                      Facilities
Computer Assisted Management    Planning
Emergency Operations (CAMEO)
Emergency Information System    Planning
(EIS)
Meteorological Information      Real-time
and Dispersion Assessment
System (MIDAS)
World Bank Hazard Analysis
(WHAZAN)
Energy Analysts Hazard
Analysis Package (EAHAP)
Other Models/Systems
Planning
Planning
Provides the following: chemical               Public Domain
information, response information, air
modelling, mapping, response resources,
inventory, emergency contacts, facility
information, route information, population
information, vulnerability zone
designations, and release records; requires
Apple computer.

Capability includes data base records of       $4,000
chemicals, facility characteristics,
transportation routes, vulnerability zones,
geographical layout.

A plume trajectory model; calculates the       $55,000-100,000
impact of releases under routine or
accident conditions; covers 14 scenarios
using features of Gaussian model equations
and heavy gas dispersion algorithms
(DEGADIS)

Models chemical dispersion and spill           $1,000
behavior; 13 mathematical models assess
data base of 30 hazardous substances to
predict effect.

Contains 3 models; OOMS model for              $75,000
pressurized releases; DEGADIS for heavier
than air gases; Gaussian for neutral or
buoyant gases;  covers 7 release scenarios
and uses weather data manually entered;
runs on IBM PC.
                                                                                                                                          20
   Several facilities use more than one model.

-------

-------
Appendix 7
Public Alert

-------

-------
                               LIST OF EXHIBITS


A 7-1  PRIMARY WARNING TECHNOLOGIES (PERCENT OF COMMUNITIES)

A 7-2  SECONDARY WARNING TECHNOLOGIES  (PERCENT OF COMMUNITIES)

A 7-3  COST ESTIMATES: A WARNING SYSTEM OF THREE SIRENS

A 7-4  COST ESTIMATES: A WARNING SYSTEM OF 50 SIRENS

A 7-5  COST ESTIMATES: A WARNING SYSTEM OF 1900 RESIDENTIAL GRADE TONE ALERT
          RECEIVERS

A 7-6  COST ESTIMATES: A WARNING SYSTEM OF 10 COMMERCIAL GRADE TONE ALERT
          RECEIVERS

A 7-7  COST ESTIMATES:  AUTOMATIC TELEPHONE DIALERS

A 7-8  SUMMARY OF INITIAL AND ANNUAL COSTS FOR FIVE ALTERNATIVE WARNING
          CONFIGURATIONS IN A 10-MILE  SQUARE EMERGENCY PLANNING ZONE

A 7-9  COST ESTIMATES:  WARNING SIRENS AND RESIDENTIAL GRADE TONE ALERT
          RECEIVERS

A 7-10 COST ESTIMATES:  WARNING SIRENS AND AUTOMATIC TELEPHONE DIALERS

A 7-11 COST BREAKDOWN FOR FIVE ALTERNATIVE SYSTEM CONFIGURATIONS

-------

-------
                                     Exhibit A 7-1

                       PRIMARY WARNING TECHNOLOGIES
                                  % of communities

Permanent
sirens
e-alert radio
sphone ring-
lown system
Fixed loud-
akers/public
address
Emergency
least system
Door to door
table sirens/
speakers on
vehicles
TV/radio
ible override
Commercial
telephone
Public within
1 mile of
facility
f f f •" fff
ff f ffff f *
-;' ;'43L4'" '
12.5
: ;• ^ -'"' •

10.3
, '
48^
30.1
.-:-.W;
44.1
•*•* ^^rtt -£t •" •'•
4K*\f K\f '•
14.0
Public 1 to 5
miles from
facility
- .-^
11.8
: «#.

10.3
' ?'»
19.9
•* "" x ' '
49.3
" :: &?
10.3
                                                     Institutional
                                                   facilities within
                                                      5 miles
 Transient
populations
   Two-way radio
  Air or helicopter
           Other
S84042-la
                                        A 7-1

-------
                                     Exhibit A 7-2

                     SECONDARY WARNING TECHNOLOGIES
                                  % of communities
                    Public within
                      1 mile of
                      facility
Public 1 to 5
 miles from
   facility
 Institutional
facilities within
   5 miles
 Transient
populations
       Permanent
           sirens


  Tone-alert radio
  Telephone ring-
    down system

      Fixed loud-
  speakers/public
         address

      Emergency
 broadcast system


     Door to door

  Portable sirens/
  loudspeakers on
         vehicles

         TV/radio
    Cable override


      Commercial
       telephone


   Two-way radio


  Air or helicopter


           Other
'.*?s
3.7
O J> "'* 'i
4*«4» ' ' f{f ,
7.4
1 ^ ' x- ';
20.6 C
48.5
36,0 ' i
36.0
24>3 .{ ':
24.3
* "* /••'
16.2
..^
\~:~. 3,7 '- '
2.2
X- <•* "" ^ -"'v ff
7.4
•i^i '
49.5
\ ,; i-44,1 ^, ^
33.8
l?:^'
23.5
U;"!!*'-'-
16.2
:^ ^ * -•
• f "" •* .•
: . --K-
3.7
-'.;->;:':' ;
6.6
'*' ' '*i*i'* ""''"
•^ ^ , ^fmtmk £v f /^
41.9
\/35.3'>'
30.9
' ' S,
19.9
123
13.2
' 5 ""
2.2
- ' ' ;6,6 "/;
„ / ' •• ,
1.5
X
5.1
' , ''' * ' ''
26.5
;' 26.5 ;"'
19.9
'?
ii.0'-'"'i|
jf
t $
10.3
f
:'.**''••
9.6
',''' " §
S84042-la
                                        A 7-2

-------
                          EXHIBIT A 7-3

          COST ESTIMATES:  A WARNING SYSTEM OF 3 SIRENS
 Device and Unit Costs
  Total
Costs ($)
122 dB: 3 @ $10,000
Site Selection
Installation
Power Hook-up
Administration
   One-Time Costs
Annual Maintenance
Annual Power Cost
Annual Testing
Annual Administration
    Annual Costs
 30,000
    252
   ,080
   ,580
   ,440
4,
2,
I,
 38,352
  1,035
    216
    600
    360
  2,211
   One-Time Cost:
   Annual Costs:
$38,352
$ 2,211
                              A 7-3

-------
                                EXHIBIT A 7-4

               COST ESTIMATES:  A WARNING SYSTEM OF 50 SIRENS
       Device and Unit Costs
  Total
Costs ($)
       122 dB: 50 @ $10,000
       Site Selection
       Installation
       Power Hook-up
       Administration
          One-Time Costs
       Annual Maintenance
       Annual Power Cost
       Annual Testing
       Annual Administration
          Annual Costs
  500,000
    4,200
   68,000
   43,000
   24,000
  639,200
   17,250
    3,600
   10,000
    6,000
   36,850
          SUBSYSTEM COSTS

       One-Time Cost:
       Total Annual Costs:

          TOTAL SYSTEM COST

       Total One-Time Cost*
       Total Annual Costs:**
 $ 41,832
 $  2,651
 $681,032
 $ 39,501
*    Does not include cost of transmitters
**   For 1 year only.
                                    A 7-4

-------
                                EXHIBIT A 7-5

            COST ESTIMATES:  A WARNING SYSTEM OF 1900 RESIDENTIAL
                         GRADE TONE ALERT RECEIVERS
       Device and Unit Costs
  Total
 Costs ($)
    Residential TAR: 1900 @ $55
    Installation
    Administration
         One-Time Costs
    Annual Maintenance •
    Annual Power Cost
    Annual Testing
    Annual Administration
         Annual Costs

         SUBSYSTEM COSTS

    One-Time Cost
    Annual Costs:

         TOTAL SYSTEM COST

    Total One-Time Costs:*
    Total Annual Costs:**
 104,500
  15,675
   5,700
 125,875
  20,900
   6,650
   1,900
  14,250
  43,700
$ 41,832
$  2,651
$167,707
$ 46,351
*    Does not include cost of encoder,
**   For 1 year only.
                                    A-7-5

-------
                                EXHIBIT A 7-6

               COST ESTIMATES:  WARNING  SYSTEM OF 10 COMMERCIAL
                         GRADE TONE ALERT RECEIVERS
       Device and Unit Costs
  Total
Costs ($)
       Commercial TAR: 10 @ $300
       Installation
       Admini s trat ion
           One-Time Costs
       Annual Maintenance
       Annual Power Cost
       Annual Testing
       Annual Administration
           Annual Costs
  ;3,000
  |   450
      30
   3,480
     300
      35
      30
      75
  :   440
       One-Time Cost:*
       Annual Costs:**
$
$
3,480
  440
*    Does not include cost of encoder.
**   For 1 year only.
                                    A 7-6

-------
                            EXHIBIT A 7-7

    COST ESTIMATES:   AUTOMATIC TELEPHONE DIALERS (USING THE BUTLER
                  NATIONAL ADAS VI:   $20,000/UNIT)
   Device and Unit Costs
 Total
Costs- ($)
11 ATD @ $20,000
Installation
Data Entry
Administration
   One-Time Costs
   Annual Maintenance
   Annual Power Cost
   Annual Testing
   Annual Administration
   Annual Line Charge
      Annual Cost
 220,000
   4.
   1,
   1,
400
320
320
 227,040
   2,200
     110
   2,200
   1,320
  31,680
  37,510
   SUBSYSTEM COSTS

One-Time Cost:
Annual Costs:

   TOTAL SYSTEM COST

Total One-Time Cost:*
Total Annual Costs:**
$ 41,832
$  2,651
$268,872
$ 40,161
 Does not include cost of conversion to touch-tone
  telephone.
 For 1 year only.
                                A 7-7

-------
                                EXHIBIT A 7-8     j

                   SUMMARY OF INITIAL AND ANNUAL CQSTS FOR
                 5 ALTERNATIVE WARNING SYSTEM CONFIGORATIONS
                IN A 10-MILE  SQUARE EMERGENCY PLANNING ZONE*
         Alternative System
           Configuration
       Cost,  $**
Initial           Annual***
         Outdoor Warning Sirens

         Residential Grade
         Tone Alert Receivers

         Automatic Telephone
         Dialers

         Outdoor Warning Sirens
         Residential Grade and
         Tone Alert Receivers

         Automatic Telephone
         Dialers
681,100
167,800
268,900
174,100
234,600
39,500
46,300
40,100
23,300
24,900
*    All configurations include the subsystem costs.
**   Costs are rounded to the nearest $100.
***  For 1 year only.
                                    A 7-8

-------
                                                EXHIBIT A 7-9
COST ESTIMATES: HfiBHH

Siren Costs
Total
Device and Unit Costs 'Costs ($)
122 dB: 7 @ $10,000 70,000

Site Selection 588
Installation 9,520
Power Hook-Up 6,020
Administration 3.360
One-Xime Costs 89,488
Annual Maintenance 2,415
Annual Power Cost 504
Annual Testing 1,400
Annual Administration 840
Annual Costs 5,159







*G SIEEHS AHD RESIDENTIAL GRADE TOTE ALERT BECEIVERS

Residential Grade Tone Alert Receiver Costs
Total
Device and Unit Costs Costs ($)
Residential TAR: 675 37,125
@ $55
Installation 5,569
Administration 2.025
One-Time Costs 44,719
Annual Maintenance 7,425
Annual Power Cost 2,363
Annual Testing 675
Annual Administration 5.063
Annual Costs 15,526


SUBSYSTEM COSTS
One-Time Cost: $ 41.832
Annual Costs: $ 2,651
TOTAL SYSTEM COST
Total One-Time Cost: $ 176.039*
Total Annual Cost: $ 23.336**

*   Does not include cost of transmitters  or encoder.
**  For one year only.
                                               A 7-9

-------
                                               EXHIBIT A 7-10

                      COST ESTIMATES:  WASHING SIBEHS AND AUTOMATIC TELEPHONE DIAIEBS

Siren Costs
Total
Dovico and Unit Costs Costs ($)
122 dB: 7 6 $10,000 70,000
Site Selection 588
Installation 9,520
Power Hook-Up 6,020
Administration 3.360
Ona-lime Costs 89,488
Annual Maintenance '2,415
Annual Power Cost 504
Annual Testing 1,400
Annual Administration 840
Annual Costs 5,159





Automatic Telephone Dialer Costs
Total
Device and Unit Costs Costs- ($)
:
5 AID @ $20,000 100,000
Installation 2,000
'
Data Entry 600
Administration 600
One-Time Costs 103,200
Annual Maintenance 1,000
Annual Power Cost 50 •
Annual Testing 1,000
Annual Administration 600
Annual Line Charse . 14,400
Annual Cost 17,050
SUBSYSTEM COSTS
One-Time Cost: $ 41,832
Annual Costs: $ 2.651
TOTAL SYSTEM COST
Total One-Time Cost: $ 234,520*
Total Annual Cost: $ 24,860**

*   Does not include cost of transmitters or conversion to touch-tone  telephone.
**  For one year only.
                                                A 7-10

-------

COSt BREAKDOWN FOR
EXHIBIT A 7-11
FIVE ALTERNATIVE SYSTEM CONFIGURATIONS

Alternative
System
Configuration Quantity
Outdoor Warning Sirens 50
Residential Grade
Tone Alert Receivers 1,900
Automatic Telephone Dialers 11
Outdoor Warning Sirens 7
Residential Grade
Tone Alert Receivers 675

Outdoor Warning Sirens 7
Automatic Telephone Dialers 5

Subsystem Common to All Alternatives
Outdoor Warning Sirens 3
Commercial Grade
Tone Alert Receivers 10

Hardware Costs, $* Total Costs, $*

Unit Total Initial Annual**
10,000 500,000 639,200 36,900

55 104,500 125,900 43,700
20,000 220,000 227,000 37,500
10,000 70,000 89,500 5,200

55 37.100 44.700 15.500
107,000 132,200 20,700
10,000 70,000 89,500 5,200
20,000 100.000 103.200 17.100
170,000 192,700 22,300

10,000 30,000 38,400 2,200

300 3.000 3,500 400
33,000 41,900 2,600
*   Costs  are rounded to the nearest $100.
**  For 1  year only.
                                              A 7-11

-------

-------
Appendix 8
Expert Panel Summary

-------

-------
                     SUMMARY OF EXPERT PANEL DISCUSSIONS
Management: Panel

     The management panel conducted a qualitative assessment of the returned
facility questionnaires.  Recognizing that the questionnaires could not
describe in full the hazards encountered at the facilities, the panel
concluded that:

     •    About a third of the facilities appeared to be aware of the
          hazards and were employing good management practices to manage
          those hazards.

     •    Another third described substantial activity in the areas of
          training, accident investigation, and risk identification.

     •    About a quarter of the responses suggested that the inherent
          risks were not fully appreciated.  Some of these risks may be
          minor.

     The panel recommended a further study of the issue of managing process
safety.  Such a study should recognize that the entire management system must
be evaluated and that the ability to differentiate between appropriate and
inappropriate management involvement requires experience and understanding of
the technical-managerial relationship.

     The panel recommended for consideration the programs that AIChE and the
Organization Resources Counselors have developed that outline good management
practices for the manufacturing, handling, use, and storage of hazardous
chemicals.

Prevention Panel

     The panel on prevention made a number of recommendations,  especially for
research and technical transfer needs.

     In the area of hazard evaluation techniques, the panel recommended that
the techniques described in the AIChE guidance document be considered the
"recognized, acceptable" methods.  The panel strongly opposed any suggestion
of a hierarchy of techniques and specifically stated that quantitative
methods should not be considered preferable in all circumstances.  The panel
recommended that:

     •    A number organizations including AIChE and EPA could enhance
          their efforts to disseminate guidance on hazard evaluation,
          especially to smaller companies, which do not appear to
          recognize the need for hazard evaluations.

     •    Some organization should assume a central role to standardize
          hazard evaluation terminology.

     •    Source strength data and acute health effects data should be
          bolstered for use in hazard evaluations.
                                    A 8-1

-------
     In the area of pre-release prevention and protection, the panel
suggested that reducing risk in a facility is achieved primarily through. t
redundant systems, interlocks, and management policies.  Appropriate
combinations of individual technologies are chemical- and process-specific,
and, therefore, it is impossible to describe a generic best available
technology for prevention, other than the best combination of the chemical
engineering, materials engineering, and process controls that are applicable
to a given situation.  Given a quality design, the operating, maintenance,
and management procedures are critical to safe operations.

Mitigation Panel

     The panel stated that mitigating releases is generally less cost-
effective than preventing releases.  Because mitigation technologies are
site-, process-, and chemical-specific, uniformity or consistency of
application should not expected or desired.  Mitigation approaches should be
part of the overall risk management'process and shpuld not be considered in '
isolation.  Because mitigation equipment is used rarely, it is critical that
management practices ensure that the equipment is always operational.

     The panel identified areas for further research:

     •    For releases directed to mitigation devices, such as
          scrubbers, flares, and stacks, improved design methods and
          approaches for handling large bursts are needed.

     •    For unconfined liquid/vapor releases, large scale
          demonstrations of the effectiveness of foams, water
          sprays/fogs/curtains,, absorbents, reactants, and physical
          covers should be conducted, including demonstrations in
          adverse conditions.

     •    For releases of solids, the magnitude of the problem needs to
          be defined to determine if developing mitigation techniques is
          warranted.

     The panel also suggested that water sprays/fogs/curtains should be
considered for unconfined vapor releases, but that the nature of these
releases makes them difficult to mitigate.  Therefore, the emphasis should be
on prevention.

Monitoring and Detection Panel

     The panel concluded that where chemical-specific, reliable, operable,
cost-effective detectors are available, at least some facilities are using
them.  Where detectors are seen as unreliable and not cost effective, they
will not be used.  The panel suggested further study to determine why certain
systems are not being used and to investigate management's commitment.

     The panel stated that significant advances can be made where no
technology exists, and that for some existing technologies need to be
modified to improve their reliable and cost-effectiveness.

     The panel suggested that perimeter alert systems may not be useful for
smaller releases.  They also stated that, in general, it is better to spend

                                    A 8-2                              '     .

-------
money preventing accidents than to spend money on systems to monitor releases
after accidents.        '                                          "' "  "''*-'

     For models, the panel stated that real-time models must be validated and
calibrated for specific sites.  In general, the panel emphasized the need for
good data to obtain good predictions.

Public Alert Panel

     The panel indicated a number of areas where improved planning is needed:
identification of contact points, notification of special groups such as
institutions and transient populations, and cross-jurisdictional contacts.
The panel felt the estimated public warning times given by respondents to the
community survey pointed to problems in providing effective warning in rapid-
onset events.

     The panel recommended that communities develop independent technical
expertise, rather than relying on the facility.  The facilities should'
broaden their contacts with the community to include the media, health
facilities, and emergency response units in their planning efforts.

     The panel recommended that future research focus on:

     •    An analysis of alternative guidelines for the use of
          evacuation and sheltering.

     •    An analysis of response to actual chemical emergencies in
          light of the existence of planning bodies and activities
          mandated by Title III.

     •    'An analysis of the warning process with attention to the
          effect of interpersonal communication and interaction.

     •    An analysis of the efficiency of public education programs in
          the warning process.
                                    A 8-3

-------

-------
Appendix 9
Facility Site Visits

-------

-------
                        SUMMARY OF FACILITY SITE VISITS


     To  confirm and clarify reported information, EPA conducted seven site
visits at  facilities that had returned questionnaires.  A team of
Headquarters  and Regional EPA personnel,  trained by the AIChE in process
chemical safety hazards,  met with company staff at each facility and toured
the process area.   State  EPA personnel also participated in some of the
visits.  These  visits provided an opportunity to gain insight into
management's  operations and decision-making processes.

     The team visited:

     •     Two acrylonitrile plants,  a producer in Texas and a user
           (plastics manufacturing)  in Illinois;

     •     Two phosgene plants,  a  specialty chemical liquid phosgene
           producer/marketer in Texas,  and a plastics manufacturer in,
           Indiana that produces and consumes phosgene onsite;

     •     A liquid sulfur dioxide and gaseous sulfur trioxide plant in
           New Jersey producing sulfuric acid;

     n     A regional drinking water chlorination operation; and

     •     A hydrofluoric  acid distributor in California.

     Several  different factqrs  were considered in selecting facilities:

     •     Unusual  comments  on the questionnaires, such an indication of
           unique  systems  or methods;

     •     Differences between facilities  handling several hazardous
           substances  and  those  handling a single chemical;

     •     The geographic  distribution;

     •     The potential impact  of a release, based on the chemical's
           toxicity,  quantity stored or used, the population,  and past
           releases;  and

     •     The number of employees  and the  age of the facility.

     Except in  one  case,  the  site visits  generally confirmed the information
provided in the questionnaire.  This is not unexpected since the facilities
agreed to  the site visit.   Most differences occurred because the respondent
either did not  fully understand a  question or gave an incomplete answer.   The
one exception was the hydrofluoric acid importer who answered the
questionnaire concerning warehouse  distribution operations,  but was
apparently unfamiliar with  actual daily operations and did not verify the
information with the warehouse personnel.

     The visits clarified, expanded, and emphasized certain areas of the
survey data that were necessary to understand the methods of operation and
management practices.  For example, the sulfuric acid producer indicated  on

                                    A 9-1

-------
the questionnaire it was uncertain whether an effective warning could be
given to the community in !an emergency.  Discussions revealed that the
management had only minimal contact with the local response agencies because
of coordination problems, and, therefore, had not conducted any drills or
training exercises.  The liquid phosgene producer stated it was the largest
merchant/marketer of liquid phosgene in the country.  However, due to the
corporate management's perceived risks in transporting liquid phosgene by
rail, it will withdraw from the business in 1989.  Instead, it will make the
phosgene-derived intermediate chemical onsite and ship this much less
hazardous chemical to users.                                    -~ " ''

     The site visits also revealed some unique technologies and practices.
The acrylonitrile producer in Texas has a unique mitigation technology in
place at the barge loading area.  A continuous bubble barrier is located
across the entrance to the loading area which creates a water level one-inch
higher than the surrounding water level.  Thus, any surface or subsurface
spills are contained and prevented from entering the bayou.  No spill has
ever spread beyond the barrier.

     The phosgene producer in Indiana reported the use of containment around
certain process equipment, but the visit showed the technology is used much
more extensively.  Any system that could contain gaseous or liquid chlorine
under pressure has secondary containment.  This includes storage vessels,
pipes (through the use of an annulus), and flanges.,  Furthermore,' the air
evacuated from within the containment is continuously monitored for phosgene
before being vented to continuously operated scrubbers.  Also, by producing
and consuming all phosgene onsite, transportation, unloading, and handling
hazards are eliminated.

     In the water supply operation, chlorine is transferred from 90-ton
railroad tank cars into 17-ton tank trucks for distribution to five regional
plants.  Although these transfer operations are near a lightly populated
area, special excess-flow control and check valves are used to prevent
overfilling the trucks.  Should these fail, special piping directs the •
chlorine into a vaporizer and water chlorinator.  Tjie chlorinated water is
disposed in the underground tunnel used to transport the Colorado River water
to the treatment site.                             ;
                                                   i
     Overall, the site visits were very successful.;  Most companies expressed
their appreciation at being able to show their commitment to environmental
protection and safe operation, not only for their employees, but also for
surrounding communities.
                                    A 9-2

-------
                Chemical Unloading Facility, Metropolitan Water
             District of Southern California,  Riverside,  California

     The 3-acre facility serves as the central liquid chlorine transfer point
for five water treatment plants and reservoirs in the greater Los Angeles area.
Maximum chlorine inventory is 600,000 pounds.  Chlorine is transferred only
during daylight hours.  Four fill-time people are employed.  The facility is
subject to earthquakes and lies in the flight path of March Air Force Base.

     Hazard and risk assessments are performed by Facility personnel and
include definitions of likely accident scenarios based on industry incidents"
and Chlorine Institute information.  Specific site safety needs are reviewed
regularly and updated.  Management's commitment to safety and training is the
cornerstone in reducing risk to employees and the community.  Standardized
chlorine handling procedures have been developed and employees receive
intensive initial training and refresher training on them.  The procedures
stress safety and leak detection measures and are used throughout the Greater
Los Angeles areas' plants, reducing the chances of human error.

     The greatest potential for a large chlorine release exists at the
unloading dock where three 90-ton railroad tank cars may be located at one
time.  Chlorine arrives by rail, transferred to 17- or 19-ton tank trucks or 1-
ton cylinders and is then shipped to the various water treatment plants.   Both
the rail cars and trucks are dedicated to chlorine service and are equipped
with special valves and excess-flow control devices to prevent accidental
releases.  Chlorine is transferred using a pressure differential between the
rail car and the receiving tank.  Moisture monitors shut down the process if
moisture is detected in the compressed air during transfer.  Any excess
chlorine from the transfer process is dissolved in water in a chlorinator.  The
chlorine solution is then injected into the Val Verde Tunnel which carries
Colorado River water to various locations in southern California.  Plant
management believes the system's check valves, fault alarms, purging lines, and
other built-in safety equipment make the facility inherently safe.  The
Facility was also designed to withstand major earthquakes.

     Chlorine gas detectors provide continuous air monitoring at three
locations:  the transfer shelter, chlorinator room, and cylinder storage area.
Audible and visual alarms are activated at the facility as well as at the
central water district communications center.

     The Facility has an emergency response plan which includes emergency phone
numbers and safety data sheets.  Employees are trained to take immediate
emergency action to mitigate small leaks following rigid safety procedures.  If
a major leak occurred, local fire and police departments would take the lead in
notifying the community as described in the Riverside County Emergency Response
Plan.  Facility safety and technical representatives would be available to
assist local officials.  The facility is investigating, computerized air
dispersion models, but has concerns about their accuracy and utility during
emergencies.

     The metropolitan water district has participated in several emergency
exercises with local agencies,  including an eight-hour command post exercise
involving over fifty State and  local personnel.  The district has  also
conducted earthquake exercises  to assess communications strategies and
problems.

                                     A 9-3

-------
                 Borg-Warner Chemicals, Inc., Ottawa,! Illinois
                                                     i
     The Borg-Warner plant, which uses acrylonitrile to make plastics, occupies
about 80 acres of a 238 acre tract and employs about 440 people.  The plant is
subject to tornados and lies within the New Madrid Seismic Zone, but has not
experienced any damage in the past.                  '                   ^

     A consultant performed a hazard/risk assessment which emphasized
consequences based on meteorological conditions rather than on the probability
of accidents.  Dispersion modeling was done to determine which weather-1 •  •••
conditions favored a "worst-case" scenario.  Fire hazards, flashback potential;
boiling liquid expanding vapor explosions, toxicity, and combustion products
were also examined.  Risk was not quantified numerically because failure rate
data were unavailable for some equipment.  Annual internal and external risk ••
management audits by plant personnel use a "what if" approach.

     Borg-Warner's pre-release prevention practices do not depend solely on the
nature of the facility, its processes, and products, but are also based on
government and industry reports, common sense, and experience.  Their       '
preventive engineering practices are specific to each chemical process and to
the potential sources of a release, and are designed to confine releases within
the facility's boundaries.  Chemical inventories are jcept at an absolute
minimum for safety purposes and to reduce costs.  Dump tanks are used to stop '
runaway reactions.  Surge capacity is built into the processes to prevent '• • •
releases from plant upsets.                          j

     Closed circuit television cameras continuously monitor the facility's
grounds for safety and detection purposes.  Visual inspections are used to
detect malfunctioning equipment.  Backup or override systems are not used.
Borg-Warner considered the installation of perimeter monitoring devices and
concluded that the complexity and cost of such a system was not warranted based
on the risks associated with their facilities.  General hydrocarbon monitoring
is done at the source.                               I
                                                     i
                                                     i
     Borg-Warner has no designated emergency control 'center, although the main
gate guardhouse would be the most likely place to establish one.  The fire
brigade leader has the authority to notify the community during business and
non-business hours.  The site manager, assistant manager, and building manager
serve as emergency advisory groups.  Plant alarm codes, backed by radios and
telephones, designate areas of the emergency during a release event.
Firefighting foams are available to mitigate fires and large spills.  The plant
does not have a hazardous materials response team.   "•
                                                     \
     The LEPC depends extensively on Borg-Warner's technical expertise.  Plant
personnel and resources are available to assist public officials in planning
and training as well as during an emergency.  Chemical quantity and release
rate, combined with meteorological conditions, are the criteria used to
determine when the local community is to be notified.'  The County Sheriff's
office is the primary notification point.  No public warning systems exist.
                                     A 9-4

-------
                  F&S  Distributing Company, Vernqn,  California

     The F&S Distribuging Company stores about 300 fifty-five gallon drums of
hydrogen fluoride solutipn (100,000 pounds) for Westco Chemicals on a 2.5 acre
site.  The drums remain sealed while in warehouse storage awaiting wholesale
distribution.  F&S employs about 10 people.  The facility is subject to
earthquakes, although no damage has been caused in the past.

     No formal hazard or risk assessment has been done.  F&S has the material
safety data sheets (MSDS) on hydrogen fluoride, but no additional information
on handling and safety procedures.

     Spills could occur during the loading or unloading of the drums as the
result of drops or forklift procedures.  No written plans exist for procedures
to be followed in case of a spill.  Soda ash is available for neutralizing
spills, but no training program exists on how to mitigate a spill or use safety
equipment.  The drums are stored on a flat concrete floor without berms.  F&S
hired a consultant to provide hazardous materials driver-training to their
employees.  One problem is that the training information and MSDSs are in
English, but most of their employees are hispanic.

     The warehouse has a fire alarm that can be heard at a nearby fire station
and is also electronically tied into the station.  The local fire department
has a catalogue of materials stored at the warehouse and makes quarterly fire
safety inspections.   No practice drills or coordination with local officials
has been done.
                                     A 9-5

-------
                    Monsanto Chemical Company, Alvin, Texas

     Acrylonitrile is produced, with a maximum storage capacity of 34.8 million
pounds.  The facility occupies 500 acres of a 3000 acre tract.  Monsanto
employs about 875 people.  Although the plant is subject to hurricanes and is
above the 100-year flood plain, they have recently experienced two 500-year
floods.                              '                \
                                                     i
                                                     i
     The HAZOP method is used by corporate personnel in all projects exceeding
$2 million.  Corporate safety also uses a "what if" audit to determine what
quantities of spilled materials would go beyond the plant boundaries.  Plant
management does not consider acrylonitrile to be acutely hazardous; therefore,
a specific hazard analysis on the acrylonitrile production units and storage
area has not been done.

     Redundant interlock systems are used to isolate specific process units to
prevent runaway reactions and to ensure safe shutdowns.   Main controls and
instruments are hard-wired into the control panels and not tied into the
computer software; thus, loss of computerized control' does not create unsafe
conditions.  Maintenance programs include regular inspecting of vessels,
ultrasonic thickness testing of tanks, tab testing for corrosion rates, and x-
ray, dye penetrant, and metallurgical testing of welds.   The entrance to the
barge loading area on the bayou has a continuous air-bubble barrier that raises
the surface level of the water about one inch, preventing any spills from
escaping into the bayou.                             :                         ;

     Monsanto developed their own automatic continuous air monitoring system
for acrylonitrile since, at that time, commercial systems were not available.
The system is a multi-point chromatograph which uses |53 sensing heads plant-
wide.  One system is located in the storage area and bne in the production
area.  No perimeter monitors are used.        .       |

     The guard station at the main gate serves as the! emergency response
center.  A mobile van, fully equipped with radios and' emergency response gear,
is the alternate center.  An internal radio information system with a siren
alerts employees.  The plant maintains two trained, self-sufficient emergency
response teams; a medical emergency team and a fire control/hazardous chemical
spill team.  High-density foam and hazardous materials foams are available to
mitigate fires and spills.  Monsanto has developed and verified spill models
in-house.  These are' available on the main-frame computer at corporate
headquarters; plant personnel have dial-up access.  They have investigated the
SAFER system but feel its accuracy is vulnerable to input data which may not be
known in an emergency.  Corporate headquarters has several models for off-line
analysis to determine local impact of releases, but they would not be used in
real-time.

     Monsanto, in cooperation with two other local chemical companies, has
installed a warning horn to alert local community residents and conducted
emergency response drills with the community.  Company representatives serve oh
the local LEPC.
                                     A 9-6

-------
                         PPG  Industries,  LaPorte,  Texas

     Liquid phosgene is produced for sale and used on-site as an intermediate
for specialty chemicals.  Maximum phosgene inventory is 50,000 pounds.
Although this plant is the largest merchant/marketer of liquid phosgene in the
country (40 million pounds per year), they will stop transporting liquid
phosgene by rail in 1989 due to corporate management's concern about the
transportation risks involved.  The facility occupies 30 acres of a 515 acre
tract.  PPG employs about 122 people.  The plant is subject to hurricanes, but
has never,experienced any flooding or damage.

     Process reliability studies are done by either a corporate-level
reliability engineer or a contractor, using a fault-tree analysis.  Informal
"what if" hazard analyses are also done by plant management during process
engineering reviews.  PPG considers liquid phosgene more hazardous than gaseous
phosgene.  Therefore, plant practice is to transfer only gaseous phosgene
within the plant.  If liquid phosgene is needed, it is vaporized, transferred,
then recondensed to liquid.

     When PPG purchased the plant in 1979, Pyrex glass pipe was used throughout
the plant to prevent corrosion from acidic products.  PPG recently completed a
$1 million project to replace most of the Pyrex with teflon-lined steel pipe,
preventing recurring leak problems.  Main controls are both computerized and
hard-wired so that emergency shut-down can be done either through the computer
or manually.  The computer system and field instruments use a non-interruptible
power supply in case commercial power is lost.  Critical pumps and pollution
control devices have diesel power backup.                              ,

     Four phosgene-sensitive tape monitors are used in the process areas:  one
8-point monitor in the production section and two 12-point and one 8-point in
three other phosgene areas.  PPG has three 4-point perimeter monitoring units
on their fence lines.  PPG will replace these units with 20 single-point
monitors, costing about $185,000.

     PPG's laboratory building near the main gate or the north-side guard shack
serves as the emergency response center.  The plant manager or shift supervisor
has the authority to respond in case of an emergency.  The plant has a SAFER  •
computer modeling system to aid in responses.   Although originally purchased to
help predict plume movement,  the system has evolved into a training and
planning tool to educate employees on the consequences of releases to the
surrounding area.

     PPG participates in a mutual-aid radio net set up by local industry which
would be used to alert and seek assistance from one another.  The plant manager
also serves on the local LEPC.  A siren and voice communications public alert
system was recently installed in LaPorte and has not been fully tested.  PPG
participates with local community agencies in mock disaster drills.  For in-
plant notification, they have a voice-synthesized public address system which
allows employees to report emergencies without removing protective equipment.
                                     A 9-7

-------
                 General Electric Company, Mt. Vernoii, Indiana

     Chlorine and phosgene are made and used on-site; in the production of high-
density plastics.  Chlorine and phosgene storage is about 250,000 pounds and
10,000 pounds, respectively.  The facility occupies 300 acres of a 1000 acre
tract.  About 3600 people are employed.  The plant is subject-to tornados and
is located in a region with an earthquake history.

     A contractor performed a probabilistic risk assessment on the GE facility
to define likely accident scenarios, assign event probabilities,,.and model
consequences for the surrounding population.  The analyses cost about $100,000.
The plant's management has set the "acceptable" risk level at 10 deaths per
million years.  All GE plants are standardized toward the same level of safety,
regardless of locale.  GE reported spending over $60 million in design and
engineering modifications to reduce risk.           (
                                                    t
     All systems that contain gaseous or liquid phosgene or chlorine under
pressure have secondary containment features.   All pipes carrying phosgene or
chlorine have an annular design with dry nitrogen sweep gas in the outer
annular space.  All tanks, compressors, and unit processes containing chlorine
or phosgene are housed in containment buildings operated under vacuum.  The
nitrogen sweep gas and containment buildings are continuously monitored for
chlorine or phosgene leaks; alarms automatically sound in the control rooms.
The chlorine and phosgene facilities have separate emergency destruction
systems (sodium hydroxide packed-tower scrubbers) which operate continuously to
neutralize any releases.  The nitrogen sweep gas and the sweep air from the
vacuum system are also vented through continuously-operated packed-tower
scrubbers which are independent of the emergency destruction system.  The
plant's structures are reinforced to withstand high winds and earthquakes.

     Chlorine electrodes are used to detect free chlprine in the production
area.  Infrared detectors are used to detect phosgene in on-line processes and
in the scrubber exhaust.  Phosgene-sensitive tape monitors with photocell
sensors operate continuously at 11 perimeter stations.

     GE has two emergency response centers, one of which is always manned.  A
mid-level manager is assigned to the center on a rotating basis and has the
primary authority to respond in case of an emergency.  A SAFER computer
modeling system is used for emergency responses in both centers.  GE feels that
SAFER allows real-time predictions.  GE has pre-modejLed some release scenarios
and stored them in the computer's memory.           j

     GE's plant alert system is integrated with the bommunity's.  The Local
Emergency Planning Committee (LEPC) includes several GE employees and GE
provides technical support.  No-notice emergency drills are conducted about
quarterly to test the emergency response centers.
                                     A 9-8

-------
                 Essex Industrial Chemicals, Newark, New Jersey

     Gaseous sulfur dioxide and sulfur trioxide are produced to make sulfuric
acid and oleum, plus liquid sulfur dioxide  for sale.  The maximum sulfur
trioxide inventory (as 65 percent oleum) is 130,000 pounds; maximum liquid
sulfur dioxide inventory is 800,000 pounds.  About 125 people are employed on
the 15 acre facility.  While the plant is not subject to natural disasters, it
lies in the flight path of Newark International Airport.

     Essex uses a "what if" hazards evaluation developed in-house which
involves corporate and plant staff, product supervisors, operators, mechanics,
maintenance personnel, and process engineers.  Essex does not use mathematical
probabilities to define risk because they feel such numbers are ambiguous.
They rely on their 30 years of operating experience and knowledge to determine
accident possibilities.

     Pre-release prevention is obtained through design engineering, operation
and maintenance.  The stainless-steel sulfur dioxide converter was the first of
its kind installed in the country, representing the latest in converter
technology.  Corrosion is reduced by operating above the dew point of sulfuric
acid.  During the annual maintenance shutdown, the converter, heat exchangers,
and piping are visually checked for corrosion and questionable areas are
ultrasonically'tested for thickness.  For 1988, Essex has estimated that about
34 percent of their capital budget will be allotted for environmental projects,
including retirement of process components to ensure against potential spills.

     A photometric sulfur dioxide analyzer is installed in the stack to monitor
routine emissions.  However, due to maintainability and reliability problems,
Essex rely on the Reich iodine titration test as the primary indicator of
sulfur dioxide emissions.  No perimeter monitors are installed because they
feel that the presence of other sulfurous compounds in the area would cause
interferences and false alarms.

     Essex has reviewed the SAFER system but is concerned about its accuracy
and its price.  They are generally suspicious of computerized dispersion models
and their predictions, as many hidden assumptions are usually made.  They have
a simple Gaussian dispersion model in a LOTUS spreadsheet format to predict
releases from their stack,  but do not have a computer model at the plant.

     The administration office, opposite the main plant, serves as the
headquarters for emergency response.  An in-plant voice paging system is the
primary notification method for employees,  with telephones as backup.   The
plant manager or. the shift supervisor has the authority to notify the state and
local officials in an emergency.  Public officials would be notified by
telephone.

     Plant management has experienced communications problems with the Newark
LEPC.  Management has made only introductory contact with the local response
agencies and has not conducted any drills with them.  The Newark fire
department and their HAZMAT team have toured the plant.   Plant management feels
they could alert and evacuate their own personnel, but are concerned about the
efficiency of the responding organizations since they have not coordinated or
participated with them in drills.
                                     A 9-9

-------

-------
Appendix 10
EPA's Accidental Release Information
Program

                              F

-------

-------
                 EPA'S ACCIDENTAL RELEASE INFORMATION PROGRAM
     In  1987, EPA instituted the Accidental Release Information Program
 (ARIP).  The purpose of this program is to focus corporate management
 attention on potential chemical accidental problems, stimulating them to take
 initiatives to prevent accidents.  In addition, ARIP is designed to establish
 a national data base that contains detailed information about the causes and
 circumstances of chemical accidents, as well as about the actions taken by
 industry to prevent and prepare for such releases.  Several currently
 available data bases contain records of accidental releases that have
 occurred, but none has complete and reliable details of the circumstances
 surrounding accidental releases.  In addition, not much is known, aside from
 the Section 305(b) review, about current industry prevention practices.

     Facilities that have received an ARIP questionnaire were selected from
 the National Response Center (NRG) data base, which is operated by the
 Transportation Systems Center of the Department of Transportation.  In
 accordance with CERCLA requirements, facilities must report to the NRG all
 releases of hazardous substances that exceed the Reportable Quantity (RQ)
 established for that chemical.  Facilities that met one or more of the
 following three criteria are targeted for ARIP letters.

     (1)  Multiple Releases - The facility experienced two or more
          releases that exceeded the Reportable Quantity within a 12-
          month period;

     (2)  Exceed a Multiple of RO Release - The facility experienced a
          release of 1,000 pounds or more of a chemical with an RQ of 1,
          10, or 100 pounds or a release of 10,000 pounds or more of a
          chemical with an RQ of 1,000 or 5,000 pounds;

     (3)  Death/Injury - The facility experienced a release that caused
          death or injury.

     EPA used a two-phase approach to gather the necessary accidental release
 information.  First, the Agency conducted a pilot test of the ARIP
 questionnaire in EPA Region IV in early 1987.  Facilities who met the
 criteria were sent a questionnaire along with a letter requesting their
 completion of the questionnaire for each release - - pursuant to several
 statutory authorities,  including CERCLA section 104(e).  Based on the results
 of the pilot test,  EPA revised the letter and the questionnaire and adopted
 other survey administration changes suggested by the pilot study.   In the
 second phase of ARIP information-gathering approach,  the questionnaire was
 distributed nationally to facilities that met the criteria.   By April 1988,
 103 questionnaires had been returned.   Because the sample size is limited,
 the results are not intended to be statistically valid or representative of
 all facilities with releases.   The Agency is currently undertaking an
 evaluation of ARIP.   It plans to refine the questionnaire and decide whether
 to pursue a regulatory approach to obtain the necessary information.

     As the information from ARIP is received,  it is entered into a data
base.   EPA will be evaluating these data and producing reports summarizing
 the causes of chemical accidents,  effective and ineffective techniques to
prevent a release,  and recommendations for prevention programs to decrease

                                   A 10-1

-------
the likelihood of accidental releases.  The information data base and report
findings will be appropriately disseminated to support other on-going
government efforts addressing accidental releases and to raise the level of
awareness in industry and in the community concerning the need for
preparedness and prevention.  The report and data base also will be made
available to state and local governments through the SERCs and LEPCs to
support emergency response services and contingency planners.

     Industry will be encouraged to review important findings and conclusions
to learn about effective industry-wide practices.  Headquarters will provide
the findings of the report to professional organizations (AIChE, ASSE), trade
associations (API, GMA) and public interest groups (when discussing guidelines
and technologies.                                  |

     The following tables present response data to four key questions in the
ARIP survey.  The survey asked what types of release prevention measures the
facilities used (Exhibit A 10-1), what immediate activities were taken in
response to a release (Exhibit A 10-2) , what short-Item cleanup activities
were implemented to restore or improve release prevention (Exhibit A 10-3),
and what additional, long-term measures are planned to prevent future
releases (A 10-4).  Each exhibit provides a frequency count indicating the
number of times a response was given, and a percentage indicating the
percentage of surveys that included the specific responses.
                                    A 10-2

-------
                                Exhibit A 10-1

                         ARIP QUESTIONNAIRE RESPONSES
                         RELEASE PREVENTION PRACTICES
Survey Question Response
Frequency
           Percent of
           103 Surveys
Operator monitoring and inspection program
Instrument for operation monitoring and warning
Release prevention equipment
Preventive maintenance
Containment
Standard Operating Procedures
Training
Release control plan
Equipment installation checks
Emergency equipment available for
 release mitigation
Release detection equipment or warning system
Divert release to wastewater treatment plant
None
Other
21
14
 5
 8
37
31
66
27
 1
23

 4
 3
 3
18
                    20
                    14
                     5
                     8
                    36
                    30
                    64
                    26
                     1
                    22

                     4
                     3
                     3
                    17
 Note:    Surveys permitted more than one response.
                                   A 10-3

-------
                               Exhibit A 10-2

                        ARIP QUESTIONNAIRE RESPONSES
                         IMMEDIATE RESPONSE ACTIVITY
Survey Question Response
Reduce system pressure
Apply spray scrubber
Shut or open valves
Isolate feeds
Transfer contents from failed equipment
Dilute or neutralize release
Activate fire fighting equipment
Containment
Plant/process shutdown
Divert release to wastewater treatment plant
Remove, dispose, treat contaminated soil
None
Other
Frequency
i
16
11
26
20
! 17
1 29
6
1 29
: 19
8
: 12
1
13
Percent of
103 Surveys
16
11
25
19
17
28
6
28
18
8
12
1
13
fcNote:     Surveys permitted more than one response.
                                   A 10-4

-------
                                Exhibit A 10-3

                        ARIP QUESTIONNAIRE RESPONSES
                  POST RELEASE  CHANGES IN PREVENTION METHODS
Survey Question Response
Frequency
Percent of
103 Surveys
Monitoring equipment: install/repair/replace           14
Process equipment: inspection                          15
Process equipment: upgrade/refine                      50
Training                                               13
Review/change standard operating procedures            24
Change equipment settings                               1
Change preventive maintenance procedures                4
Repair/install/expand/improve containment              18
Expand capacity                                         2
Install emergency backup systems                        4
Review/change monitoring procedures                     6
Better labeling of process equipment                    3
Check for similar problems and upgrade equipment        4
Install pad to prevent soil contamination               3
None                                                    3
Other                                                  17
                    14
                    15
                    49
                    13
                    23
                     1
                     4
                    17
                     2
                     4
                     6
                     3
                     4
                     3
                     3
                    17
 Note:    Surveys permitted more than one response.
                                   A 10-5

-------
                                Exhibit A 10-4      ,

                        ARIP QUESTIONNAIRE RESPONSES
                    ADDITIONAL PREVENTION METHODS PLANNED
Survey Question Response
Frequency
                                                           .*
Percent of
103 Surveys
Regular equipment inspections
Regular review of operating procedures
Preventive maintenance
Evaluate/install backup equipment
Evaluate/install monitoring equipment
Install/change containment
Refine/upgrade SOPs
Expand operator training
Equipment upgrade
Participation in audits/seminars
Change equipment settings
Refine/improve process
Check for similar and upgrade equipment
None
Other
     16
      3
     10
      3
      6
      6
     14
     16
     16
      4
      1
      4
      4
     20
     16
    16
     3
    10
     3
     6
     6
    14
    16
    16
     4
     1
     4
     4
    19
    16
*Note:    Surveys permitted more than one response.
                                    A 10-6

-------
Appendix 11
International Activities

-------

-------
                       PREVENTION OF CHEMICAL ACCIDENTS

                           INTERNATIONAL ACTIVITIES
BILATERAL

     The U.S. is currently working with Mexico and Canada on a number of
prevention activities.  In January 1988, the U.S. and Mexico signed a Joint
Contingency Plan for Accidental Chemical Releases Along the Border.  As part
of the planning process, 28 cities along the border were identified as areas
in which principal planning activities must take place.  The Plan calls for
the development of plans for the 14 so-called Sister Cities on either side of
the inland border.  Among preventive measures in these plans are: (1)
identification of hazards; (2) exchange of information on chemicals; (3)
development of inventories, and (4) recognition of safety practices.  These
activities also include providing training and technical assistance to local
communities on appropriate mitigation and cleanup procedures as well as on
protection of workers onsite.

     The U.S. and Canada are developing a joint contingency plan similar to
that with Mexico to deal with chemical accidents.
MULTILATERAL

     The U.S. has been involved over the past year in intensive efforts at
the international level to focus more attention on chemical incidents.  In
February 1988,  the Fourth High Level meeting of the Organization for Economic
Cooperation and Development (OECD) member nations adopted a set of goals to
prevent and respond to accidents involving hazardous substances.  They are:

     1.   To limit the frequency and severity of accidents through
          better measures to prevent releases of hazardous substances;

     2.   To prevent adverse effects from accidents through better land
          use planning; and

     3.   To mitigate the consequences of accidents through the
          development of adequate emergency plans and measures.

     The OECD Environment Committee recognized the need for a continuing
forum that would provide the opportunity for member nations to continue to
exchange information on a wide variety of policy and technical issues and to
share this information with developing countries through the United Nations
Environment Program (UNEP).   Specific issues the forum will address include:

     1.   Consideration of broad policies established or required in the
          field;

     2.   Explanation of individual country program initiatives of
          cross-nation utility and interest; and

     3.   A wide variety of specific technical subj ects such as
                                   A  11-1

-------
          a.   Dispersion models for dense gas.

          b.   Information management methodology.

          c.   New techniques in response and cleanup.

          d.   Networking of accident data bases.

          e.   Technology for prevention and mitigation.

          f.   Response technical assistance.

          g.   Risk assessment and effectiveness in planning and in
               management practices.               :

     There will be approximately five meetings of this forum over the next
year, beginning in June 1988.

     Member countries were encouraged to promote appropriate legislative,
regulatory, or administrative measures required to meet these goals.  Several
areas were suggested for consideration in developing adequate safety and
prevention measures that include: (1) identifying hazardous installations
that require special preventive measures; (2) establishing a licensing or
permitting system for installations; (3) evaluating risk assessments or
safety studies of hazardous installations; (4) conducting an effective
inspection program; and (5) ensuring appropriate zoning and land-use policies
for new installations handling hazardous chemicals.!

     UNEP has developed a data base on chemicals that could serve as a      .
central point for countries to access.  The U.S. is also supporting the
development of guidance on planning for chemical emergencies.  This document,
entitled the Handbook on Awareness and Preparation for Emergencies at the
Local Level (APELL) is expected to be published in the fall of 1988.
                                    A 11-2

-------
Appendix 12
Facility Questionnaire

-------

-------
                                                        Form Approved
                                                        OMB Number 2050-0079
             SURVEY OF MONITORING,  DETECTION,  RELEASE PREVENTION,
         AND PUBLIC ALERT SYSTEMS FOR EXTREMELY HAZARDOUS SUBSTANCES
     Please see the instructions for clarification of these questions.
(Instructions are provided for all but the .self-explanatory questions).
you require additional clarification for any question, please use the
telephone number provided on page 3 of the questionnaire.
                                            If
Facility Name:
Facility Location:
                     No.     Street
          City
County
State
Zip
Parent Company, if any:

Nature of Business: 	

SIC Code(s): 	
Dun & Bradstreet ID#:

-------
                                     -2-
                  SEGTION 1:   EXTREMELY HAZARDOUS  SUBSTANCES
1.1
Does this facility use, store, process, produce, manufacture,  import,  or
package any of the extremely hazardous substances listed below in
quantities larger than the threshold planning quantity (TPQ)?
                                             YES
                                                 NO
1.2
     If "YES," please complete the entire questionnaire.

     If "NO," please provide the information requested in Question 1.2,
     sign, and return pages 1, 2, and 3 of the questionnaire to the
     address indicated on page 3 of the questionnaire.

Indicate the maximum quantity present at any one time of each substance
listed below that is handled at your facility in quantities greater than
the TPQ.
     NAME
Acrylonitrile
Ammonia
Benzenearsonic Acid
Benzotrichloride
Chlorine
Chloroacetic Acid
Furan
Hydrazine
Hydrocyanic Acid
Hydrogen Fluoride
Hydrogen Sulfide
Mechlorethamine
Methiocarb
Methyl Bromide
Methyl Isocyanate
Phosgene
Sodium Azide
Sulfur Dioxide
Sulfur Trioxide
Tetraethyltin
Trichloroacetyl chloride
                         CAS No.

                        107-13-1
                        7664-41-7
                        98-05-5
                        98-07-7
                        7782-50-5
                        79-11-8
                        110-00-9
                        302-01-2
                        74-90-8
                        7664-39-3
                        7783-06-4
                        51-75-2
                        2032-65-7
                        74-83-9
                        624-83-9
                        75-44-5
                        26628-22-8
                        7446-'09-5
                        7446-11-9
                        597-64-8
                        76-02-8
  TPQ
 (Ibs)

10,000
   500
  1  10*
   100
  ,100
  ;ioo*
  ;soo
 1,000
   100
  ;100
   500
    10
  :500*
 1,000
  ;500
  ,  10
  ;soo
   500
  iioo
  '100
   500
   MAXIMUM
QUANTITY (Ibs)
     TPQ for these solids is 10,000 pounds if the solid is not in powdered,
     solution, or molten form, or does not meet certain reactivity criteria
     (see 40 CFR 355.30).                         :

-------
                                     -3-
1.3  Please provide the following information for the person at your facility
     we may contact regarding your response to this questionnaire.
          NAME of CONTACT PERSON:
          TITLE:
          ADDRESS:
          PHONE:
     THIS QUESTIONNAIRE IS SUBMITTED BY:
          TITLE:
DATE:
RETURN TO:
          U.S. Environmental Protection Agency
          Section 305(b) Survey
          P.O. Box 2734
          Fairfax, Virginia  22031

     IF YOU HAVE ANY QUESTIONS CONCERNING THE QUESTIONNAIRE, PLEASE FEEL FREE
     TO TELEPHONE (703) 934-3950.

-------
                                     -4-
                      SECTION 2:  FACILITY INFORMATION
2.1  Total number of employees at your facility:
2.2  Number of production and/or other workers directly involved
     the designated substance:	
with
2.3  What is the longitude and latitude of your facility?

          Longitude:    Degrees	        Minutes	    Seconds	

          Latitude:     Degrees	        Minutes	    Seconds	

2.4  Please provide a facility sketch (see sample and further    directions
     in instructions).
2.5  What is the fenceline area of this facility (in acres)?
                                                   i
2.6  What is the approximate average population density within a
     radius of your facility?                      j
2-mile
2.7  What is the average distance from the process and storage areas for the
     designated substance to your property line?

2.8  What factors were used to establish the above1given boundaries between
     process and storage areas, fenceline, and property line?

2.9  Does your facility have a safety/loss prevention officer?

                              YES  	      NO	

                              NAME:  	'

2.10 Does your facility have a department that deals solely with safety/loss
     prevention?                                   '
                              YES  	      NO	  !

2.11 How many individuals at your facility are working full-time specifically
     on safety/loss prevention issues?	;	L_
     Part time?	

2.12 Does your facility report to a  corporate-wide safety  officer or
     department?                                   :
                              YES	     NO	      Not Applicable	
 2.13  Please  attach  a  simple  organizational  chart  for  your  facility showing
      management  responsibility for  safety/loss prevention.

-------
                                     -5-
2.14 Please briefly describe training required at your facility for all
     levels of employees in the table provided below.  Indicate the frequency
     of retraining (see instructions for sample).
             TRAINING
     STAFF
MGMT
          FREQUENCY OF RETRAINING
          STAFF               MGMT
2.15 Please describe management activities related to safety/loss prevention
     in the table provided below (see instructions for clarification and
     sample).
          MANAGEMENT ACTIVITIES
                                                         FREQUENCY
2.16 Does your facility participate in the Community Awareness and Emergency
     Response (CAER) program of the Chemical Manufacturers Association?
               YES
NO
OTHER PROGRAM
          If "OTHER," please describe below.
2.17 What procedures do you have in place, or have you used, to investigate
     chemical accidents that occur at your facility, specifically those
     involving the designated substance?

-------
                                     -6-
2.18 What do you do with chemical accident investigation findings?
     does your facility report these findings?
                                   To whom
2.19 Describe changes in your facility or in your operations during the last
     5 years to reduce the potential for, or the frequency and severity of,
     accidental releases of the designated substance.
2.20 Comment on the effectiveness of the changes indicated above.
2.21 Is your facility unusually susceptible to natural disasters (flood,
     severe storms, volcano, hurricane, tornado, earthquakes, etc.)?
                    YES
NO
          If "YES," describe the particular natural disaster to which you are
          susceptible and measures you have taken  to protect your facility.

-------
                                     -7-
                       SECTION 3:  HAZARD ASSESSMENTS
3.1  For the designated substance,  describe briefly the methodologies,  if
     any, by which your facility identifies safety hazards.
3.2  How often are these methodologies employed?
3,3  Which of the methodologies identified in Question 3.1 has your facility
     used within the past year?
3.4  How pften are these methodologies updated?
3.5  Which individuals or organizations perform assessments using the
     above-mentioned methodologies at your facility?   Please include
     contractors and corporate personnel from off-site locations (if
     applicable) and their titles,
3.6  What conditions or events would trigger an unscheduled assessment or
     analysis using any of the methodologies identified in Question 3.1?

-------
                                     -8-
3.7  Describe actions taken after a potential hazard is identified.
3.8  When operational and process changes are made, including changes in both
     hardware and operating conditions,  how are the safety aspects evaluated?

-------
                                     -9-
           SECTION 4:  PRE-RELEASE PREVENTION, PROCESS MONITORING,
                               AND MITIGATION
4.1  Provide a simple flow diagram of the process(es) for manufacturing,
     processing, use, disposal, and/or handling of the designated substance.
     Show the flow or movement of the substance from its creation or entrance
     to your facility until its disposal or exit.  (Please see the
     instructions for an example.)

4.2  When was this process or handling system constructed?     	.

4.3  When did the most recent major upgrade occur?             	
4.4  Is the process line you diagrammed above dedicated to the manufacture,
     processing, or use of the designated substance?
                    YES
NO
4.5  Please describe any changes/modifications to your process or facility
     that were instituted as a result of significant changes in the facility
     neighborhood.

-------
                                     -10-
4.6  On the chart below list the codes for the process and chemical hazards
     you have identified for the process(es) previously diagrammed (Question
     4.1).= List codes for the release prevention oi: process control
     equipment used because of those hazards,  process monitors,  and
     techniques used to mitigate any releases that might occur.   (Please see
     INSTRUCTIONS for an example and further explanation of information
     requested.)
Equipment
  Name
Location
 Number
Process
Hazard
 Code
Chemical
 Hazard
  Code
Control    Monitor  Mitigation
 Code	Code	Code

-------
                                     -11-
4.7  On'the chart below, list each control code you identified in Question
     4.61and briefly describe the rationale for the choice of this particular
     prevention or control device (please see instructions for an example).
Control -
 Code
Rationale
4.8  How do you detect when control equipment is disabled or "down," and what
     systems do you have for backup or override in the event of a
     malfunction?
4.9  On the chart below, list each pre-release monitor code you identified in
     Question 4.6 and briefly describe the rationale for the use and choice
     of this particular technology.
Monitor
 Code
Rationale
4.10 How do you detect when this monitoring equipment is disabled or "down,"
     and what systems do you have for backup or override in the event of a
     malfunction?

-------
                                     -12-
4.11 On the chart below, list each mitigation code [you identified in Question
     4.6 and briefly describe the rationale for the use and choice of this
     particular technique or technology (please.see instructions for an
     example).
Mitigation
   Code
             Rationale
                      »          ,
4.12 How do you detect when mitigation equipment is disabled or "down," and
     what systems do you have for backup or override in the event of a
     malfunction?                                  !
4.13 On the chart below, show the frequency of inspection, audit, testing,
     and maintenance for each control and monitor you identified in Question
     4.6.
    Monitor
or Control Code
Frequency of:
Inspection
Audit    Maintenance    Testing
 4.14  Describe  the preventive  maintenance  (including frequency)  for process
      equipment and  storage units  used by  your facility.

-------
                                     -13-
4.15 Please attach any information on ':coritrol/detection/monit6ring/mitigatidn
     devices,  approaches, or techniques 'used at'your facility  that are
     innovfiPtive or state-of-the-art.  '"-                  -
4.16 Does your facility have any concerns regarding safety/loss prevention
     that have not been addressed  in any of the  above questions?  Please
     explain below.

-------
                                     -14-
      SECTION 5:  AIR, WATER, GROUND WATER RELEASE MONITORING/DETECTION
5.1  On the chart below, indicate which ambient air, in-stack, waste water
     sewerj storm sewer, surface water, or ground-water monitoring devices
     you use to detect accidental releases^ and describe the rationale for
     using those devices.
 Detector/
Monitor Code
                  Rationale
5.2  How do you detect when this equipment is disabled or "down," and what
     systems do you have for backup or override in. the event of a
     malfunction?
5.3  On the chart below, for each detector or monitor you identified in
     Question 5.1, show the frequency of inspection, auditing, testing, and
     maintenance.
 Detector/
Monitor Code
Frequency of:
Inspection
Audit
Maintenance
Testing

-------
                                     -15-
5.4  What mitigation techniques does your facility use (or would it use)  if
     an accidental release has contaminated the ground water?
5.5  Describe any perimeter monitoring devices your facility uses for the
     designated substance.  Please reference the submitted site sketch
     (Question 2.4).
5.6  If your facility has considered the installation of perimeter monitoring
     devices, or has installed such devices, please comment on your
     assessment of the economic feasibility of establishing, maintaining, and
     operating these devices.
5.7  Describe any
     substance.
off-site monitors your facility uses for the designated
5.8  Describe the reliability or limitations of any perimeter or off-site
     monitoring systems used by your facility and the rationale for
     selection.

-------
                                     -16-
                             SECTION 6:   MODELING
     Please describe any models used at your  facility  to predict the movement
of plumes of released chemicals_ through .environmental  media  (see instructions
for clarification) ,  For each model described,  indijcate whether the model
predicts chemical movement in (1) air, (2) ground water, or  (3) surface
water.  Also indicate whether the model is-used for  (a) response or (b)
planning.                                     •

-------
                                     -17-
                           SECTION 7;   PUBLIC ALERT
7.1  How has your facility's emergency organization coordinated with each of
     the following agencies or organizations in the past several years?
     (Circle appropriate numbers below for.each agency/organization)
Circle
agencies
that exist
in your
area
No Coor-
dination
at all
Initial
intro-
ductory
contact
onlv
Developed
emergency
response
nlans with
On -go ing
coordinated
emergency
effort
Participated
in emergency
exercises
with*
A.  Fire/rescue
    dept.           0
B.  Other internal
    facility
    emergency org.  0
C.  Local/county
    police          0
D.  Local emergency
    management/
    civil defense   0
E.  State
    police          0
F.  Industry
    mutual aid
    assoc.          0
G.  State emergency
    management      0
H.  Hospital
    emergency
    rooms           0
I.  State EPA       0
J.  Others
    (incl.
    schools, nur-
    sing homes, etc.)
                    0
                    0
                    0
1
1

1
1

1
1
1.
1
1
1
1
2

2


2

2


2

2
2
2
2
2
2
3

3


3

3


3

3
3
3
3
3
3
4

4


4

4


4

4
4
4
4
4
4
    For each agency/organization that you hold coordinated exercises with,
    please.use the space under the right-hand column to enter the number of
    times in the last 2 years that you have held such exercises.

-------
                                     -18-
7.2  Describe the criteria used to determine when, if at any time, the local
     community is to be notified by your facility of a chemical release at
     your facility.                                 I
7.3  Describe the procedure your facility follows in notifying the local
     community once the above criteria are met.     ;
7.4  Who at the facility makes the decision to notify the community?  Is there
     a. back-up person (identify by title)?          '•
7.5  Who, if different from the above, makes the decision to notify the
     community during non-business hours in the event of an emergency?
7.6  Who, if different from the above, actually notifies the community?
7.7  Describe procedures used to  inform and update the  following groups during
     emergency situations.                             .    .   .

           (a)  Personnel at Facility        .     .   :        '      '   ;

-------
                                     -19-
          (b)  Local Community
7.8
7.9
Are the above described notification procedures, including all methods of
communication, tested during safety audits or emergency exercises?
                    ALL
                        SOME
NONE
          Please explain.
Based on past experience, if any, please estimate the time required to
notify the community or the population within a 5-mile radius of your
facility of an accidental release.
7.10 Indicate the personnel and other resources available from your facility
     to assist public officials
Personnel/Resource
                                 In planning,
                                 preparedness
                                 and training
                        In an emergency
Expertise or tech assistance
Fire brigade/dept.
Emergency team
HazMat team
Decontamination team
Medical personnel
Decontamination equipment
Monitoring equipment
Protective equipment
Containment equipment
Cleanup equipment
Other 	-  .
Other 	
Other	
7.11 What do you consider to be  the major problems  and constraints to getting
     a timely and effective warning to  the public around your facility in the
     event of an emergency  (please attach additional sheets, if necessary)?

-------
                                     -20-
7.12 What is your overall assessment of your facility's capability to  provide
     an alert in the event of an emergency?

     a)  Within the facility

     	It is certain that an effective warning would be made.
     	It is somewhat certain that an effective warning would be made.
     	It is somewhat uncertain that an effective warning would be made.   .
     	It is highly uncertain that an effective warning would be made.

     b)  To the community

     	It is certain that an effective warning would be made.
     	It is somewhat certain that an effective warning would be made.
     	It is somewhat uncertain that an effective; warning would be made.
     	It is highly uncertain that an effective warning would be made.

-------
                                     •21-
    SECTION 8:  NEW PREVENTION, CONTROL, DETECTION, MONITORING, OR RELEASE
                           MITIGATION TECHNOLOGIES
8.1  Describe any programs you conduct at your facility or within your company
     for the development/testing of new technologies for:

          (a)  Detectors
          (b)  Monitors
          (c)  Alarm/Public Alert Systems
          (d)  Control Devices
          (e)  Hazard Identification/Evaluation
          (f)  Intrinsically Safer Chemicals/Processes
          (g)  Pre-release Prevention
          (h)  Pre-release Protection
          (i)  Post-release Mitigation
          (j)  Reactive Chemicals
          (k)  Storage and Transfer of Hazardous Materials
          (1)  Human Factors
          (m)  Modeling of Releases (Source Strength and Dispersion)
          (n)  Process Measurement
          (o)  Other

-------
                                     -22-
8.2  Please describe any problems or weaknesses you have noticed or
     experienced with:

          -(a)  The control devices specified in Questions! 4.6 and 4.7.
          (b)  The detector devices specified in Question 5.1.
          (c)   The monitoring devices specified in Sections 4 and 5.
          (d)  The mitigation devices or procedures>specified in Questions
               4.6, 4.11, and 5.4.
          (e)  The models specified in Section 6.
          (f)  Alarm/public alert systems.
          (g)  Other
8.3  Are there any specific areas of research that you believe should receive
     greater support and attention by industry, government, or both?  Please
     explain.                                      ..;..
8.4  Were there any hazards (Question 4.6) for which control, detection,
     monitoring, or mitigation techniques are either outdated, outmoded, or
     non-existent?  Please explain.

          (a)  Detectors
          (b)  Monitors
          (c)  Alarm/Public Alert Systems
          (d)  Control Devices

-------
                                     -23-


          (e)  Hazard Identification/Evaluation


          £f)  Intrinsically Safer Chemicals/Processes


          (g)  Pre-release Prevention


          (h),  Pre-release Protection


      •••••• (i)  Post-release Mitigation                ,


          (j)  Reactive Chemicals


          (k)  Storage and Transfer  of Hazardous Materials


          (1)  Human Factors


          (m)  Modeling of Releases  (Source Strength and Dispersion)
 .•'•••>• •  ' .-.I' '.,'.-. ; " . -  •• .  .  .•  .-..-.:• r •••,-.-•

          (n)  Process Measurement


          (o)  Other


8.5  Your comments are welcome.   Please feel free to make them below.
WC0028Q.mom

-------

-------
Appendix 13
Community Questionnaire

-------

-------
        SECTION 305(b) TITLE III
       Superfund  Amendments and
       Reauthorization Act of 1986
   COMMUNITY SURVEY TO REVIEW
PUBLIC ALERT AND WARNING SYSTEMS
 FOR CHEMICAL PLANT EMERGENCIES
       (Form Approved; OMB# 2050-0079)
   U.S. ENVIRONMENTAL PROTECTION AGENCY

   FEDERAL EMERGENCY MANAGEMENT AGENCY

      OAK RIDGE NATIONAL LABORATORY
              December 1987

-------
INTRODUCTION
In the letter introducing this questionnaire, a facility or company was identified, that
according to EPA records, is located within your community.
       Is this facility or company in your community and jurisdiction? (circle response)

      1 YES, IN COMMUNITY AND JURISDICTION; proceed to next page.

      2 NO, IN COMMUNITY BUT NOT UNDER OUR JURISDICTION;
            Please give this questionnaire to the agency and person with emergency
            planning responsibilities for this facility to complete.

      3 NO, NOT IN COMMUNITY.
            Is the facility or company in a nearby community? (circle response)
            1  Yes; if yes, which community?	;	
               Who do we contact in that community about emergency planning for a
               chemical emergency? Name	,	
                                  Position.
            2  No; Your Community Name:
            Please return this questionnaire in the envelope provided to:

            Oak Ridge National Laboratory
            PO Box X
            Oak Ridge, TN 37831

            Attn.  John H. Sorensen, MS  206;  Bldg.  4500N

-------
 INSTRUCTIONS
 In some of the following questions we refer to "the facility". With respect to your
 community this; refers to the facility identified in the cover letter. Please answer the
 questions only in reference to that facility. Most of the questions do not require specialized
 instructions. When a question does require instructions* they are provided with the
 question.
 A. Emergency Planning                                          :
 Q A-l Does your community have an Emergency Operations Center (HOC)? (circle
       number of your answer)
       1  YES, a permanent dedicated EOC that is maintained in operating condition.
       2  YES, a permanent EOC that is also used for other purposes.
       3  YES, a temporary EOC that is established when the need arises.
       4  NO, an EOC does not exist.
       5  OTHER ARRANGEMENT; explain    .                    	
 Q A-2 Is there an alternative or backup EOC? (circle number)
       1  YES
•- ••	2   NO       	
 Q A-3 About how many personnel (FTEs or Full time equivalents) work on emergency
      planning in the community?
       	FTEs (if none, write 0)
 Q A-4 About what percent of their time is devoted to chemical emergencies?
 Q A-5 Does someone in the community have a formal responsibility for planning for a
       chemical emergency? (circle number)
       1  YES  ,
       2   NO
 Q A-6 Does the community have an emergency plan? (circle number)
       1  YES
       2   NO
 Q A-7 Does it have a section on chemical emergencies? (circle number)
       1   NO
       2   YES; If Yes, in what year was it adopted?
            19

-------
QA-8 Is there a special plan or annex for {his chemical facility? (circle number)
      1  YES                                         !           .
      2   NO
Q A-9 About how many emergency personnel would be available to respond to an
      accident at a chemical facility, (write, number in each category; if none, write 0)  .
      	FULL TIME PAID EMERGENCY PERSONNEL (35 hours per week or more)
      	PART TIME PAID EMERGENCY PERSONNEL (less than 35 hours per week)
      	VOLUNTEER EMERGENCY PERSONNEL
      	PERSONNEL FROM NON-EMERGENCY AGENCIES OR DEPARTMENTS
      	PERSONNEL AVAILABLE FROM MUTAL AID AGREEMENTS
      	 OTHER PERSONNEL; explain            *	.
B. Communications               .                    I
Q B-l Who at the facility has primary responsibility for initially notifying the local
      community about an emergency at the facility? (Enter name and positions of all; if
      no one or if person is unknown, circle appropriate response below)
      posrnoN(S):                             NAME(S)   •
      1  NO ONE
      2  UNKNOWN
Q B-2 Who would notify the community if those persons were not available? (Enter name
      and positions of all; if no one or if person is unknown, circle appropriate response
      below)
      POSITION(S):
NAME(S)
      1  NO ONE
      2  UNKNOWN

-------
Q B-3 Who has the designated official responsibility for receiving an alert from the
      chemical facility? (Enter name and positions of all; if no one or if person is
      unknown, circle appropriate response below)
       POSITIONS):
NAME(S)
       1   NO ONE

     :'2   UNKNOWN
Q.B-4 Who would they notify if that person was not available? (Enter name and positions
      of all; if no one or if person is unknown, circle appropriate response below)
       POSmON(S):
NAME(S)
       1   NO ONE

 -•     2   UNKNOWN        •        •                .

Q B-5  Are there any days of the week and times of day that it would be difficult for the
       community to receive an alert  (either not issued or not received)?  (circle number)

  .   .  1.  NO

       2   YES ...  please describe when and why: (indicate time periods by day and

                  hours and the reasons each period presents difficulties)
Q B-6 What communication equipment would the facility use to notify the community of a
      chemical accident (primary and back-up)?

      PRIMARY EQUIPMENT:

-------
      BACK-UP EQUIPMENT:
Q B-7 What procedure do you follow when the initial warning is received?
Q B-8 Please describe the communication equipment within your HOC. (such as
     commercial telephone; 911 telephone; dedicated telephone; automatic ring-down
     system; manual alarm; automatic alarm, radio; computer link; or other)
Q B-9 How often is this communications equipment routinely tested? (circle number)

      1  YEARLY

      2  SEMI-ANNUALY                              '

      3  MONTHLY

      4  WEEKLY                                     \

      5  DAILY                   .;      ,                  .  *          ..;

      6  NEVER
                                                      :         -' '   *
Q B-10 Please describe any mobile communications resources available to the community.

-------
Q B-l 1 What information does the community need from the chemical facility in an
       emergency notification to make a decision to warn the public?
Q B-l2 The emergency warning system to alert and inform the public in case of a chemical
      emergency at the facility is primarily comprised of (circle all that apply):
      1   FIXED (permanently installed) MECHANICAL SIRENS
      2   FIXED ELECTRONIC SIRENS •
      3   FIXED HORNS, BELLS OR WHITTLES
      4   FIXED LOUDSPEAKERS/PUBLIC ADDRESS
      5   FIXED FLASHING LIGHTS / STROBES
      6   PORTABLE LOUDSPEAKERS/PUBLIC ADDRESS
      7   PORTABLE SIRENS/WHISTLES
      8   NOA A WEATHER RADIO
      9   EMERGENCY BROADCAST RADIO STATION
      10  TONE ALERT RADIOS'.." ,...'   /
      11  RADIO PAGERS
      12  AUTOMATED TELEPHONE DIALERS
      13  OTHER; specify	,	-
Q B-13  Is there a person in the community with the responsibility of maintaining
       communications with the facility during an emergency? (circle number)
      1 NO
     2 YES; Please indicate the name and position of all:
      POSITIONS):
... NAME(S)

-------
8
C. Emergency Decision  Making
                                                          !   " ''-  ' ..' . '. •
Q C-l Who has the legal authority to activate an emergency warning or public alert system
      in your community?                       ,            ,

       POSITIONCS):                                      [NAME'S)'
Q C-2 Does anyone in the community have the assigned responsibility to make the
      decision to warn the public in the event of a chemical accident? (circle number)
      iNO                                              ;    •  •.   .
      2  YES; same as the person(s) with legal authority in previous question
      3  YES; different than the person(s) with legal authority in previous question;
          indicate the name and position of all:
       POSITIONCS):
NAME(S)
Q C-3 Please describe the process for making the decision to warn the public after
      receiving an initial alert from the facility?
Q C-4 Many different types of emergencies can occur at a chemical facility. One type is a
      very fast release of hazardous materials posing a clear threat ;to public safety.  A
      second type is a slowly developing problem with a potential !for a release;

      a. What is the minimum number of people that would have to be involved in
      making the decision to warn the public? (please record the tbtal number of people
      playing an active role in making the decision to warn the public)
              . PEOPLE IN A FAST-MOVING EMERGENCY  |	

               PEOPLEJN A SLOWLY DEVELOPING EMERGENCY

-------
 Q C-5. Once you have received an initial alert from the facility about how long would it
       take to mobilize the necessary people that make the decision once an initial alert is
    .-. , .received and what is the basis for your estimate? (please estimate the range as a
       minimum or fastest possible time and the most likely time in hours -H and minutes
       M; if less than 1 hour write a 0 before the H and specify minutes only)
       IN A FAST-MOVING EMERGENCY:
            MINIMUM:
 H __ M;"   MOST LIKELY:
      ' IN A SLOWLY DEVELOPING EMERGENCY:

            MINIMUM:	H	M;     MOST LIKELY:.

       BASIS FOR ESTIMATE:
                         .H	M;


                         .H	M;
Q C-6 Is there a written procedure for making the decision to issue a public warning?
     • (circle number)                        -   	      	  <•-•   •
      1  NO-  - -    -	.

   .   2  YES; please attach a copy of the procedure	

Q C-7 How long would it take to make a decision  to notify the public? (please estimate the
      range as a minimum and most likely time in hours -H and minutes -M; if less than
      1 hour write a 0 before the H and specify minutes only)

      IN A FAST-MOVING EMERGENCY-
           MINIMUM:
:H
.M;    MOST LIKELY:
      IN A SLOWLY DEVELOPING EMERGENCY:
           - MINIMUM: ^H	M;     MOST LIKELY:
.H
                                H
.M;
                               .M;
Q C-8 there are a number of ways to protect the health and safety of people from a
      release of hazardous chemicals (protective actions).

      a. What protective actions would be considered for recommendation in a chemical
       emergency for the general population?

-------
10
      b. What protective actions would be considered for recommendation in a chemical
          emergency for institutional facilities (such as hospitals, schools, prisons, or
          nursing homes)?                      ,            ;         ,
Q C-9 Does your community have a written procedure for making a decision about what
       protective actions to recommend or order? (circle number)      •-••••

       1  NO                                                         ,

      2  YES; please attach a copy of the procedure            ;

Q C-10 Does your community have a written plan or procedure for, issuing an
       alert/warning? (circle number)                     ,   '   ,
       1  NO          '     -..•'...-      '          j..-  •	

      2  YES; please attach a copy of the procedure           . 1.
Q C-ll  Does your community have a written warning/alert plan or procedure
        specific to the facility? (circle number)     .......         ;

       i  NO           .,     ,,.,,.,..  ,,   .    .  ;,..  ...„
       2  YES; please attach a copy of the procedure
Q C-12 What effort, if any, has your community made to provide the general public with
       information about chemical hazards and emergency response? (Please describe and
       attach any relevant information describing these efforts)   i

-------
                                                                         11
D; Populations  at  Risk
Q D-l About what percent of the land use within one mile of the facility is in the following
      categories?
      	_%  OPEN SPACE
      	%  INDUSTRIAL (wholesale; manufacturing)
              _% COMMERCIAL (retail; offices)
              .% SUBURBAN RESIDENTIAL (single family)
              .% URBAN RESIDENTIAL (multi-family)
              _% OTHER	      .	..(specify)
Q D-2 About what percent of the land use within five miles of the facility is in the
      following categories? ?
      ___%  OPEN SPACE
      	%  INDUSTRIAL (wholesale; manufacturing)
              .% COMMERCIAL (retail; offices)
              _% SUBURBAN RESIDENTIAL (single family)
              _% URBAN RESIDENTIAL (multi-family)
              _% OTHER                       	(specify)
Q D-3  Approximately how many people live within 1 mile of the facility? (if none write 0)
       	PEOPLE
Q D-4  Approximately how many people live within 5 miles of the facility? (if none write
  ••   -  0).. ' ;   •'   •         . -...-,..- .,:••.-...
       	PEOPLE
 Q D-5 Are there significant fluctuations in the size of the population (such as workers,
       tourists, or visitors) in any of the area within 5 miles of the facility?
......  .a. During the day pr night ? (circle number)
	   1 NO	
 	2 YES; please describe the circumstances, timing of the fluctuation, and number of
              people involved

-------
12
       b. During different seasons? (circle number)

       1  NO
       2  YES; please describe the circumstances, timing of the fluctuation, and number of
              people involved
       c. During weekends?  (circle number)
1 iNW
2 YES; please describe the circumstances, timing of the flu
people involved





ctuation, and number of
• •

• • "f - _ . • ,


Q D-6 Are there institutional populations within 5 mile of the facility  (eg. schools,
       hospitals, nursing homes, correctional facilities)? (circle number)

       1  NO
       2  YES; please list below and provide an estimate of population in each
       Name of Facility                Type                  !          Population

-------
                                                                             13
E. Public Alert and Warning

Q E-l Please indicate the primary and secondary methods, if any, for warning the following
      populations within 5 miles of the facility. Primary methods are those systems that are most
      likely to be used in an emergency. Secondary methods are backup options that are
      currently available for use if primary methods fail. (Please read the entire list of warning
      options. Then place a P for primary and a S for secondary in the appropriate boxes for each
      population group. Leave cells blank if warning method would not be used.)
permanent sirens
tone-alert
radio
telephone
ringdown
system
fixed loudspeakers/
public address
emergency
broadcast
system
door to door
portable sirens/
loudspeakers
on vehicles
TV/
radio
Cable
Override
Commercial
Telephone
Two-way
radio
Air or
helicopter
Other
Institutional
Public withinl mile Public 1 to 5 facilities Transient
of facility miles from facility within 5 miles populations




















































       Please specify other,

-------
14
Q E-2 How often are the warning equipment and procedures routinely tested? (circle u
      number)

      1  YEARLY                                    ..;....'
      2  SEMI-ANNUALY                                ;

      3  MONTHLY

      4  WEEKLY                                       '
      5  DAILY                                    "  '  1
      6  NEVER                                         I
                                                        I      ,
Q E-3 Is there a special warning system(s) in your community for another type of hazard?
      (eg. civil defense outdoor sirens, nuclear power plant system, flash flood warning
      system; circle number)                                '.

      1  NO                                            ;

      2  YES; please describe                              :
Q E-4 Please estimate how long it would take to notify each population group and briefly
      describe the basis for that estimate (check unknown if you cannot estimate)

a. Residents within 1 mile of facility:                          i

      	HOURS 	MINUTES  ___ UNKNOWN

      BASIS FOR ESTIMATE:

-------
b. Residents 1 to 5 miles of facility:
      	HOURS 	MINUTES
      BASIS FOR ESTIMATE:
UNKNOWN
                                                                   15
c. Institutional populations listed in question D-6:
      Facility Name
      	,	HOURS
      	HOURS
      	HOURS
      	HOURS
      	HOURS
    Warning Time
  .MINUTES  	UNKNOWN
               _UNKNOWN
               _UNKNOWN
                UNKNOWN
 MINUTES
 MINUTES
 MINUTES
. MINUTES
                UNKNOWN
      BASIS FOR ESTIMATES (include a description of who at the facility the estimate refers
      to; eg. an administrator or the population at the facility):
d. Fluctuating (transient) populations listed in question D-5:
      Group Name                      Warning Time
HOURS
HOURS
HOURS
HOURS
HOURS
MINUTES
MINUTES
MINUTES
MINUTES
MINUTES
UNKNOWN
UNKNOWN
UNKNOWN
UNKNOWN
UNKNOWN
      BASIS FOR ESTIMATE:

-------
16
F. Warning Content
Q F-l Which of the following authorities (or equivalents) would be identified in the
      warning as the source of the warning information ? (circle all that apply or no one)
      1  MAYOR                                 "!.••
      2  COUNTY EXECUTIVE
      3  CIVIL DEFENSE OR EMERGENCY OFFICIAL
      4  CITY OR COUNTY MANAGER                i
      5  PUBLIC HEALTH OFFICIAL                  !
      6  ENGINEER/SCIENTIST FROM FACILITY
      7  ENGINEER/SCIENTIST FROM GOVERNMENT  j
      8  POLICE CHIEF                            !
      9   SHERIFF                                  \
      10 FIRE CHIEF                               i
      11 CHEMICAL FACILITY MANAGER
      12 STATE OFFICIAL                           \
      13 NO ONE WOULD BE IDENTIFIED             '
      14 OTHER (specify) 	
Q F-2 Do you have a written message protocols (For example Emergency Broadcast
      Station or EBS messages) for communicating with the general public in an
      emergency? (circle number)
      1 NO; please describe the information you would provide in the message.
      2 YES; please attach a copy of the protocols

-------
                                                                             17
Q F-3 Do you have a written message protocols (For example Emergency Broadcast
       Station or EBS messages) for communicating with the institutional faci]iti?ft in an
       emergency? (circle number)

       1  NO; please describe the information you would provide in the message.
       2 YES; please attach a copy of the protocols.
Q F-4 Do you have warning messages for non-English speaking populations?
      (circle number)

       1   YES, for the following language(s):
       2 NO, do not have non-English population(s).

       3 NO, have not developed.

-------
18
G. Coordination                                            j
Q G-l How has your community's emergency organization coordinated emergency
       planning for a chemical accident with each of the following agencies/organizations
       in the past several years? (check each category that applies to each group)
1. Chemical
facility
2. Other local
agencies
3. Local relief
(red cross)
4. State CD
5. State Police
6. Other
communities
7. FEMA
8. Hospitals
9. State EPA
10. US EPA
11. Media
                 Does not   No        Initial/      Developed    On-going   Participate
                 apply in    Contact    Introduc-   Emergency   Coordinated in Emer-
                 this        At All     tory Contact Response
                 community           Only       Plans With
Emergency  gency Exer-
Effort       cises With
                     Check all that apply for each Agency/Organization
Others
(including institutions such as
schools nursing homes etc.)

-------
                                                                           19
Q G-2 How many times in die last 2 years you participated in emergency exercises on
      chemical accidents? (if none write 0)

      __ EXERCISES

Q G-3 Has your community used information from the Chemical Manufacturers
      Association CAER Program (Community Awareness/Emergency Response)?
      (circle number)

      1  YES

      2  NO

Q G-4 What entity has the state designated as the Local Emergency Planning District in
      your area to develop an emergency plan for chemical accidents? (please
      describe)
Q G-5  Has the membership of this committee been appointed? (circle number)

       1  NO

       2  YES; if yes, how is your community represented on this
         committee?
Q G-6  Has the facility provided the community with information describing the hazardous
       chemicals used at the facility? (circle number)

       1   NO

       2  YES; if yes, what have they provided?

-------
20
Q G-7 Indicate the resources that you currently have, resources you have received from the
      chemical facility, resources that would be available from the chemical facility,
      resources that you could receive from a mutual aid agreement, or resources from
       another source to assist you in a chemical emergency.(check all that apply)
                      Currently   Provided
                      have in     by the
                      community chemical
                                 facility
Available    Have mutual  Have from
in an        aid agreement another
emergency   (with another source
from facility community)
Expertise or
 technical assistance

Fire team
Emergency team
HazMat team

Decontamination
 team
Medical personnel

Decontamination
 equipment
Monitoring
 equipment
Protective
 equipment

Other	
H.  COMPUTER USE                                     '•
                                                           \  -
Q H-1  Does your community use a computer in emergency planning, that is, in preparing for an
       emergency? (circle number)                            ;

       1  NO                                              ;

       2  YES; What is it used for?
       What type of computer?

-------
                                                                    21
Q H-2 Does your community use a computer in emergency management, that is, in responding to
      an emergency? (circle number)
      1 NO (if no to both Q's H-1 and H-2, go to question 7 below)
      2 YES;  What is it used for ?
      What type of computer?
Q H-3 How many people directly involved in emergency management in your community
     use computers for the each of the following activities?
      WORD PROCESSING
      SPREAD SHEETS
      DATA BASE MANAGEMENT
      COMMUNICATIONS
      & DATA ACCESS
      SPECIAL EMERGENCY
      PLANNING FUNCTIONS
      SPECIAL EMERGENCY
      MANAGEMENT FUNCTIONS
      OTHER APPLICATIONS
      PLEASE SPECIFY
. PEOPLE
. PEOPLE
. PEOPLE

. PEOPLE

. PEOPLE

. PEOPLE


 PEOPLE

-------
22
QH-4 FOR EACH COMPUTER APPLICATION: How often do emergency personnel
        in the community (including yourself) use each of the following computer
        applications? (Circle Appropriate Number)              ;




Word
Processing
Data Base &
Resource
Management
Spread Sheet
& Budgeting
Communication
& Data Access
1
DAILY



1


1

1

1
2
WEEKLY



2


2

2

2
3
MONTHLY



3


3

3

3
4
A FEW
TIMES
A YEAR

4


4

4

4
5
ONCE A
YEAR OR
LESS

i 5 '


.. : ,5
i
5 '•'
1
"• 5
6
NEVER



6


6

6

6
Special Emer-
gency Planning
Functions       1

Special Emer-
gency Manage-
ment Functions  1
Q H-5 If you use software designed specifically for emergency planning functions please
       describe the programs and their use? (circle number)

       1  DO NOT USE

      2  USE; please complete.

      Software                        Use                ;

-------
                                                                            23
Q H-6 If you use software designed specifically for emergency management functions please
       describe the programs and their use? (circle number)
       1  DO NOT USE
       2  USE; please complete.
       Software
Use
Q H-7 Do you use a model for predicting the dispersion of chemicals? (circle number)
      1  NO
      2  YES; please describe the model and its uses.
I.  Overall  Assessment
Q1-1  What do you consider to be the weakest link in the sequence of tasks that are
       involved in getting a timely and effective warning to the public around the chemical
       facility?
       Why?

-------
24
Q1-2 Overall what is your assessment of the capability to provide a timely warning to the
      public within 5 miles of the facility in the event of a serious emergency? (circle
      number)

      1  IT IS HIGHLY CERTAIN THAT AN EFFECTIVE WARNING WOULD BE MADE
      2  IT IS SOMEWHAT CERTAIN THAT AN EFFECTIVE WARNING WOULD BE MADE
      3  IT IS SOMEWHAT UNCERTAIN THAT AN EFFECTIVE WARNING WOULD BE MADE
      4  IT IS HIGHLY UNCERTAIN THAT AN EFFECTIVE WARNING WOULD BE MADE
Q1-3 Are there other facilities in your community that require emergency
      plans because they have hazardous chemicals? (circle number)
      1 NO; skip to question 6
      2 YES;  if yes, about how many facilities?
      	FACILITIES IN THE COMMUNITY            ;
Q1-4 On the whole how do the emergency planning efforts of the facility
      identified in the cover letter compare to the other chemical facilities in
      your community? (circle number)

      1 MUCH BETTER THAN OTHERS
      2 SOMEWHAT BETTER THAN OTHERS
      3 ABOUT THE SAME AS OTHERS
      4 SOMEWHAT POORER THAN OTHERS
      5 MUCH POORER THAN OTHERS
Q1-5 If an emergency occurred at the facility how does the community's ability to issue a
      timely warning to the public around the facility compare with "the ability to issue a
      warning around other facilities? (circle number)          :

      1 SIGNIFICANTLY BETTER THAN FOR OTHERS   '!'
      2 SLIGHTLY BETTER THAN FOR OTHERS
      3 ABOUT THE SAME AS FOR OTHERS
      4 SLIGHTLY POORER THAN FOR OTHERS         i
      5 SIGNIFICANTLY POORER THAN FOR OTHERS

-------
                                                                  25
Q1-6 The following are areas in which the acquisition of new resources or improvements
      of existing capabilities could enhance preparedness for chemical emergencies in
      your community.  If you could obtain these over the next several years which
      would you want first, second, third and so forth? Please rank these areas in the
      order which you feel they are needed by your community. (Please read the entire
      list. Place a 1 by the area that is needed first  Next place a 2 by the area needed
      second. Continue until all areas are ranked.)
          COMMUNICATIONS EQUIPMENT
          COMPUTER WITH EMERGENCY MANAGEMENT SYSTEM
          DECONTAMINATION EQUIPMENT
          FUNDING FOR A PLANNER
          FUNDING FOR SENDING STAFF TO TRAINING
          FUNDING TO PREPARE AN EMERGENCY PLAN
          MEDICAL EQUIPMENT
          MONITORING EQUIPMENT
          PROTECTIVE CLOTHING
          PUBLIC ALERT/WARNING EQUIPMENT
          RESPIRATORY PROTECTION EQUIPMENT
Q1-7 Advances in knowledge and technology can improve the basis for emergency
      preparedness. The following are areas in which improvements could enhance
      preparedness for chemical emergencies.  If you could obtain these over the next
      several years which would you want first, second, third and so forth? Please rank
      these areas in the order which you feel they are needed by your community. (Please
      read the entire list. Place a 1 by the area that is needed first Next place a 2 by the
      area needed second. Continue until all areas are ranked.)
          IMPROVE COMMUNICATIONS TECHNOLOGIES
          IMPROVE COMPUTERIZED DISPERSION MODELS
          IMPROVE DECISION SUPPORT SYSTEMS
          IMPROVE INFORMATION HOT-LINES
          IMPROVE KNOWLEDGE ON PROTECTIVE ACTION EFFECTIVENESS
          IMPROVE KNOWLEDGE ON THE TOXICITY OF CHEMICALS
          IMPROVE MONITORING TECHNOLOGIES
          IMPROVE PROTECTIVE EQUIPMENT
          IMPROVE PUBLIC ALERT/WARNING TECHNOLOGIES
          IMPROVE TECHNICAL PLANNING GUIDES
          IMPROVE TRAINING PROGRAMS/COURSES

-------
26
Q1-8 Has any emergency occurred in the last 5 years that has resulted in a public warning
      in your community?
      1  NO                                            '
      2  YES; please briefly describe the incident(s) and how the public were warned
How many people in total assisted in answering this questionnaire?
      	PEOPLE

About how long did it take to complete?
      	HOURS
Who in the community can we contact for additional information if necessary?
      NAME:
      ADDRESS:
      PHONE:
Have you included  the  following, plans, procedures, and protocols
including:
      Decision making procedures  for  public warning decision (Q C-5)
      Decision making procedures  for  selecting  protective  actions  (Q C-13)
      Written warning/alert plan and procedures (Q C-14)
      Warning/alert plan and procedures for facility (Q C-15)
      Warning message protocols (Q F-2 and F-3)       j

-------
Appendix 14
Bibliography

-------

-------
                                  MANAGEMENT

Alderfer, Clayton P. and H. Bierman, Jr.  Choices with Risk-: Beyond the Mean
and Variance,  Journal of Business, 43. 341-353, 1970.

Bowman, Edward H.  A Risk-Return Paradox for Strategic Management, Sloan
Management Review, 21. 17-31, 1980.

Bowman, Edward H.  Risk Seeking by Troubled Firms, Sloan Management review,
23, 33-42, 1982.

Boykin, Raymond F., R.A. Freeman, and R.R. Levary.  Risk Assessment in a
Chemical Storage Facility, Management Science, 30. 512-517, 1984.

Budescu, David V. and T. ₯allsten, Consistency of Interpretation of
Probabilistic Phrases, Organizational Behavior and Human Decision Processes,
36. 391-405, 1985.

Cox, R.A., and D.H. Slater. State-of-the-Art in Risk Assessment of Chemical
Plants in Europe, in Low Probability/High Consequence Risk Analysis: Issues,
Methods, and Case Studies, Ray A. Waller and Vincent T. Covello, Eds. Plenum
Press, New York, 1984.

Fischoff, Baruch, S. Lichtenstein, P. Slovic, S.L. Derby, and R. Keeny.
Acceptable Risk, Cambridge University Press, New York, 1981.

Kahneman, Daniel and A. Tversky.  Variants of Uncertainty, Cognition, 11. 143-
157, 1982.

Kunreuther, Howard. Limited Knowledge and Insurance Protection, Public Policy,
24. 227-261, 1976.

Laughhunn, D. J., J.W. Payne, and R. L. Crum.  Management Risk Preferences for
Below Target Returns, Management Science, 26. 1238-1249, 1980.

March, James G. and Z. Shapira.  Management Perspectives on Risk and Risk
Taking, Management Science, vol. 33, No. 11. 1404-1418, 1987.

Slovic, Paul, B. Fischhoff, and S. Lichtenstein, Behavioral Decision Theory,
Ann. Rev. Psychology, 28. 1-39, 1977.

-------


-------
                                  PREVENTION
GENERAL

Lees, F.P.  Loss Prevention in the Process Industries.  London:  Butterworths;
1980.
HAZARD EVALUATION

Battelle Columbus Division.  Guidelines for Hazardous Evaluation Procedures.
The Center for. Chemical Plant Safety.  New York:  American Institute for
Chemical Engineers; 1985.

Kletz, T.A.  Eliminating Potential Process Hazards.  Chemical Engineering.
April 1, 1985.

Ozog, H.  Hazard Identification, Analysis and Control.  Chemical Engineering.
February 18, 1985.


PREVENTION REFERENCE MANUALS

Davis, D.S. and G.B. DeWolf.  Technical Support to the South Coast Air Quality
Management District Toxic Chemical Accidental Air Releases.  U.S.
Environmental Protection Agency.  EPA-600/8-87-040 (NTIS PB87 232 260), 1987.

Davis, D.S., G.B. DeWolf, and J.D. Quass.  Prevention Reference Manual -
User's Guide, Overview for Controlling Accidental Releases of Air Toxics.
U.S. Environmental Protection Agency.  EPA-600/8-87-028 (NTIS PB87 232 112),
1987.

Davis, D.S., G.B. DeWolf, and J.D. Quass.  Prevention Reference Manual -
Control Technology, Volume 1:  Prevention & Protection Measures for
Controlling Accidental Releases of Air Toxics.  U.S. Environmental Protection
Agency.  EPA-600/8-87-039a (NTIS PB87 228 656), 1987.

Davis, D.S., G.B. DeWolf, and J.D. Quass.  Prevention Reference Manual:
Chemical Specific, Volumes 1-12.  U.S. Environmental Protection Agency.  EPA-
600/8-87-034a-l, 1987.


PROCESS DESIGN CONSIDERATIONS

American Insurance Association.  Hazard Survey of the Chemical and Allied
Industries.  Engineering and Safety Services, New York, 1979.

Anyakora, S.N., G.F.M. Engel, and F.P. Lees.  Some Data on the Reliability of
Instruments in the Chemical Plant Environment.  The Chemical Engineer, Ragby,
England.  Number 225, 1971.

Bartosiak, G.  Guide to Thermocouples.  Instruments and Control Systems,
Chilton Co., Radnor, Pennsylvania, November 1978.

-------
Bradford, Mike and David G. Durrett.  Avoiding Common Mistakes in Sizing ,
Distillation Safety Valves.  Chemical Engineering, July 9, 1984.        .    ,

Considine, D.M.  (ed.).  Process Instruments and Controls Handbook, McGraw-
Hill, New York,  1985.                               !
                                                    i
Courty, P.H., J.P. Artie, A. Converse, P. Mikitinko,; and A. Sugier.  C^-Cg
Alcohols From Syngas.  Hydrocarbon Processing, November 1984.

Fauske, Haus K.  Emergency Relief System Design.  Chemical Engineering
Progress, August 1985.                              !  "        ,         ,

Fitt, J.S.  The  Process Engineering of Pressure Relief and Slowdown, Systems,
Loss Prevention  and Safety Promotion in the Process Industries.  First
International Loss Prevention Symposium, Amsterdam, Holland, 1974.
                                                    i        ,   '•  '    '     -
Floar, P.O.  Programmable Controllers Directory, First Edition, Technical Data
Base Corporation, Conroe, Texas, 1985.
                                                    F         (           ' •  i'  1
Flow Measurement Handbook.  Omega Engineering Corporation, Stamford,    .    .
Connecticut, 1985.                                  i

Green, D.W. (ed.).  Perry's Chemical Engineers' Handbook, (Sixth Edition),
McGraw-Hill, New York, 1984.

Groggins, P.H. (ed.).  Unit Processes in Organic Synthesis, Fourth Edition,
McGraw-Hill, New York, 1952.

Guide for Explosion Venting.  National Fire Protection Association, Quincy,
Massachusetts, Vol. 68, 1978.

Herrick, E.G., J.A. King, R.P. Ouellette, and P.N. Cheremisinoff.  Unit  . •
Process Guide to Organic Chemical Industries.  Ann Arbor Publishers, Ann
Arbor, Michigan, 1979.                              i

Hix, A.H.  Safety and Instrumentation Systems in Loss Prevention, Volume 6.
American Institute of Chemical Engineers, New York, 1972.
                                                    t    .    •        -,•'.'/'.
Kletz, T.A.  Make Plants Inherently Safe.  Hydrocarbon Processing, September
1985.

Kremers, J.  Avoid Water Hammer.  Hydrocarbon Processing, March 1983.
                                                    i
Lawley, Herbert  G. and Trevor A. Kletz.  High-Pressu're-Trip Systems for Vessel
Protection.  Chemical Engineering, May 12, 1985.    |                   : ,

Liptak, B.C.  Costs of Process Instruments.  Chemical Engineering, September
7, 1970.                                            I

Liptak, B.C.  Costs of Viscosity, Weight, and Analytical Instruments.
Chemical Engineering, September 21, 1970.           '                    „-

-------
Liptak, B.C.  Safety Instruments.and Control Value Costs.  Chemical
Engineering, November 2, 1970. '

Liptak, B.C. arid V. Christa (ed'.) .  -Instrument Engineers' Handbook, Revised
Edition.  Chilton Book Company, Radnor, Pennsylvania, 1982.

Papa, D.M.  How Back Pressure Affects Safety Relief Valve.  Hydrocarbon
Processing, May 1983.             :

Peters, M.S. and. K.Dj Timmerhaus.  Plant Design and Economics for Chemical
Engineers, Third Edition, McGraw-Hill, New York, 1980.                 :

Pressure and Strain Measurement Handbook and Encyclopedia.  Omega Engineering
Corporation, Stamford, Connecticut, 1985.

Rase, H.F.  Piping Design for Process Plants.  John Wiley and Sons, Inc., New
York, 1963..

Rasmussen, E.J.  Alarm and Shutdown Devices Protect Process Equipment,
Chemical Engineering, May 12, 1975.

Recommended Practice for the Design and Installation of Pressure Relieving
Systems in Refineries:  Part I -  Design; Part II - Installation.  American
Petroleum Institute, Washington,  D.C., API RP 520, 1977.

The Richardsen Rapid Construction Cost Estimating System, Volumes 1-4:   ''.
Richardsen Engineering Services,  Incorporated, San Marcos, California, 1986.

Scully, William A.  Safety-Relief-Valve Malfunctions:  Symptoms, Causes and
Cures.  Chemical Engineering.  August 10, 1981.

Shinsky, F.G.  Process Control Systems.  McGraw-Hill, New York, 1979.     '    '•

Temperature Measurement Handbook and Encyclopedia.  Omega Engineering-''
Corporation, Stamford, Connecticut, 1985.

U.S. Nuclear Regulatory Commission.  Reactor Safety Study. ' National Technical
Information Service, WASH-1400 (NUREG 75/014), October 1975.

Weir, E.D., G.W. Gravenstine, and T.F. Hoppe.  Thermal Runaways:  Problems
with Agitation.  Plant/Operations Progress.  American Institute of Chemical
Engineers, New York, July 1986.

Wilmot, D.A., and A.P. Leong.  Another Way to Detect Agitation.  Loss5'
Prevention.  Volume 11, American Institute of Chemical Engineers, New York,
1977.
PHYSICAL PLANT DESIGN CONSIDERATIONS

Failure Analysis and Prevention.  Metals Handbook; 8th Edition.  Volume 10.
American Society for Metals, Metals Park, Ohio, 1975.

-------
Fontana, M.G., and N.D. Greene.  Corrosion Engineering.  McGraw-Hill,  New
York, 1978.
Guide for Inspection of Refinery Equipment.  American Petroleum Institute.
Washington, B.C., December 1982.

Guidelines to the Selection and Use of Corrosion Resistant Metals and Alloys.
CS79-10.  Pfaudler Company, Rochester, New York, 1979.    ,

Kletz, T.A.  Plant Layout and Location:  Methods for Taking Hazardous
Occurrences into Account in Loss Prevention.  American Institute of Chemical
Engineers, New York, Vol. 13, 1980.             '    ,

Kletz, T.A.  What Went Wrong.  Gulf Publishing Company, Houston, 1985.

Lewis, D.J.  The Mond Fire, Explosion and Toxicity Index Applied to Plant
Layout and Spacing in Loss Prevention.  American Institute of Chemical
Engineers, New York, Vol. 13, 1980.

McNaughton, K.J.  Selecting Materials for Process Equipment.  Materials
Engineering I.  McGraw-Hill, New York, 1980,

Munson, R.E.  Safety Considerations for Layout and Design of Processes Roused
Indoors in Loss Prevention.  American Institute of Chemical Engineers, New
York, Vol. 13, 1980.

National Electric Code.  National Fire Protection Association, Quincy,
Massachusetts, 1984.

NFPA 30, Flammable and Combustible Liquids Code, 1984.  National, Fire
Protection Association,.Quincy, Massachusetts,  1984.

Piping Component Standards for Refinery Service, Std.  593-609.  American
Petroleum Institute, Washington, D.C.               !          ,
                                                    f
Proceedings of the International Corrosion Forum.  4 volumes.  National
Association of Engineers, Boston, March 1985.

Spec 12D, Specification  for Field Welded Tanks  for Storage of Production
Liquids, 9th Edition.  American Petroleum Institute, Washington, D. C., .January
1982.                                               ;

Spec 12f, Specification  for Shop Welded Tanks for Storage of Production
Liquids, 8th Edition.  American Petroleum Institute> Washington, D.C., January
1982.                                               ;

Std 620, Recommended Rules for Design and Construction of Large, Welded; Low
Pressure Storage Tanks,  7th Edition.  American  Petroleum Institute,
Washington, D.C., January 1982.      ,     ,          ;               .   • <•

Tomfohode, J.H.  Design  for Process Safety.  Hydrocarbon Processing,  December
1985.               .     .......      ! ,     .  .   -   •  : ,.	
                                   .-'.:,      •             '      '     ; '* "
Vhlig.  I.H.  Corrosion Handbook.  John Wiley and Sons, New York, 1948.

-------
PROCEDURES AND PRACTICES  '                       .                        .

Burk, A.F.  Operating Procedures and Review.  Paper presented  at  the Chemical
Manufacturers' Association Process Safety Management Workshop, Arlington,
Virginia, May 7-8, 1985.

Conrad, J.  Total Plant Safety Audit.  Chemical Engineering.   May 14, 1983.

Kubias, P.O.  Technical Safety Audit.  Paper presented at the  Chemical
Manufacturers' Association Process Safety Management Workshop, Arlington,
Virginia, May 7-8, 1985.                                        ;      .   .

Stus, T.F. ' On Writing Operating Instructions.  Chemical Engineering;       :
November 26, 1984.
PROTECTION TECHNOLOGIES

Evans, F.L. Jr.'  Equipment Design Handbook for Refineries and Chemical Plants.
Volume 2.  Gulf Publishing, Houston, 1974. •                               ,,

Joseph D., et al.  Evaluation of the Efficiency of Industrial Flares:
Background-Experimental Design-Facility.  EPA-600/2-83-070 (NTIS PB83-
263723), August 1983.

Kent, G.R.  Practical Design of Flare Stacks.  Hydrocarbon Processing and
Petroleum Refiner, August 1964.

Pohl, J.H. and N.R. Soelberg.  Evaluation of the Efficiency of Industrial
Flares:  Flare Head Design and Gas Composition.  EPA-600/2-85-106 (NTIS PB86-
100559), September 1985.

Product Line Reference Catalog.  National Air Oil Burner Company, Inc.,
Philadelphia, Pennsylvania, 1984.

Straitz, J.F. III.  Flaring with Maximum Energy Conservation.  Pollution
Engineering, Volume 12, February 1980.

Tan, S.H.  Simplified Flare System Sizing.  Hydrocarbon Processing,  October
1967.


LEAK PLUGGING                        '.    '                .  '

The Chlorine Institute, Inc.   Chlorine  Institute Emergency Kit "A" for 100-Ib
+ 150-lbv Chlorine Cylinders.  Washington, D.C.,  1986.                '

The Chlorine Institute, Inc.   Chlorine  Institute Emergency Kit "B" for"'""
Chlorine Ton Containers.  Washington, D.C.,  1981.

The Chlorine Institute, Inc.   Chlorine  Institute Emergency Kit "C" for
Chlorine Tank Cars and Tank Trucks.   Washington,  D.C.,  1981.

-------
Cook, R.L. and R.W. Melvold.  Development of a. Foam >Plugging Device for
Hazardous Chemical Discharges.  Proceedings of the 1980 National Conference on
Control of Hazardous Materials Spills.  Vanderbilt University, Nashville,
Tennessee, 1980.

Mitchell, R.C., C.L. Hamermesh, and J.V. Lecce.  Feasibility of Plastic Foam
Plugs for Sealing Leaking Chemical Containers.  U.S. Environmental Protection
Agency.  EPA-R2-73-251.  May 1973.

Schelling, Carter.  Reducing Repair Downtime with Pipe Freezing.  Plant
Engineering, July 24, 1986.

Vrolyk, J.J., R.C. Mitchell, and R.W. Melvold.  Prototype Systems for Plugging
Leaks in Ruptured Containers.  U.S. Environmental Protection Agency,  EPA-
600/2-76-300.  1976.

Zajic, J.E. and W.A. Himmelman.  Highly Hazardous Materials Spills and
Emergency Planning.  Marcel Dekker, Inc., New York, 1978.


SECONDARY CONTAINMENT                               ;

Aarts, J.J. and D.M. Morrison.  Refrigerated Storage Tank•Retainment.Walls.
CEP Technical Manual, Ammonia Plant Safety and Related Facilities, Volume 23.
American Institute of Chemical Engineers, New York, 1983.

Feind, K.  Reducing Vapor Loss in Ammonia Tank Spills.. CEP Technical Manual,
Ammonia Plant Safety and Related Facilities, Volume 17.  American Institute of
Chemical Engineers, New York, 1975.                 '

Hendriks, N.A.  Safety Wall Systems for Ammonia Storage Protection.  CEP
Technical Manual, Ammonia Plant Safety and Related Facilities, Volume 21.
American Institute of Chemical Engineers, New York, 1979.

Johnson, D.W. and J.R. Welker.  Diked-In Storage Areas, Revisited'.  Chemical
Engineering, July 13, 1978.


SPRAY SYSTEMS

Beresford, T.C.  The Use of Water Spray Monitors and'Fan Sprays.for.Dispersion
Gas Leakages.  I. Chem. E. Symposium Proceedings.  The Containment and
Dispersion of Gases by Water Sprays, Manchester, England, 1981.

Deaves, D.M.  Experimental and Computational Assessment of Full-Scale Water
Sprays Barriers for Dispersing Dense Gases.  4th International Loss Prevention
Symposium.  Harrogate, England, September 1983.     ;

Eggleston, L.A., W.R. Herrera, and M.D. Pish.  Water Spray to Reduce. Vapor
Cloud Spray in Loss Prevention, Volume 10, American:Institute of Chemical
Engineers, New York, 1976.                          :

-------
Emblem, K. and O.K. Madsen.  Full Scale Test of a Water Curtain in Operation.
5th Loss Prevention Symposium, Cannes, France, 1986.

Experiments with Chlorine.  Ministry of Social Affairs, Voorburg,  Netherlands,
1975.

Greiner,'M:L.  Emergency.Response Procedures for Anhydrous Ammonia Vapor
Release.  Ammonia Plant Safety and Related Facilities.  Volume 24.  CEP
Technical Manual, American Institute of Chemical Engineers, New York,  1984.

Harris, N.C. , The Control of Vapor Emissions from Liquified Gas Spillages.
3rd Loss Prevention Symposium, Basle, Switzerland, 1980.

Harris, N.C.  The Design'of Effective Water Sprays - What We Need to Know.   I.
Chem. E. Symposium Proceedings.  The Containment and Dispersion of Gases by
Water Sprays, Manchester,  England, 1981.

Kirk, R.E. and D.F. Othmer.  Encyclopedia of Chemical Technology.   Third
Edition, John Wiley & Sons, New York, 1980.

Martinsen, W.E. and S.P. Muhlenkanp.  Disperse LNG Vapors'with Water.
Hydrocarbon Processing, July 1977.

McQuaid, J.  The Design of Water-Spray Barriers for Chemical Plants.  2nd
International Loss Prevention Symposium. . Heidelberg, West Germany, September
1977.

McQuaid, J. and R.D. Fitzpatrick.  The Uses and Limitations of Water-Spray
Barriers.  I. Chem. E. Symposium Proceedings.  The Containment and Dispersion
of Gases by Water Sprays,  Manchester, England, 1981.

Moodie, K.  Experimental Assessment of Full-Scale Water Spray Barriers for
Dispersing Dense Gases.  I. Chem. E. Symposium Proceedings.  The Containment
and Dispersion of Gases by Water Sprays, Manchester, England, 1981.

Moodie, K.  The Use of Water Spray Barriers to Disperse Spills of Heavy Gases.
Plant/Operations Progress, American Institute of Chemical Engineers, New York,
October 1985.

Moore, P.A.C. and W.D. Rees.  Forced Dispersion of Gases by Water and Steam.
I. Chem. E. Symposium Proceeding.  The Containment and Dispersion of Gases by
Water Sprays.  Manchester, England, 1981.

Prugh, R.W.  Mitigation of Vapor Cloud Hazards.  Part II.  Limiting the
Quantity Released and Countermeasures for Releases.  Plant/Operations
Progress, July 1986.

Smith, J.M. and van Doom, M.  Water Sprays in Confined Applications:   Mixing
and Releasing from Enclosed Spaces.  I. Chem. E. Symposium Proceedings.  The
Containment and Dispersion of Gases by Water Sprays, Manchester, England,
1981.

Vincent, G.C., et al.  Hydrocarbon Mist Explosions - Part II, Prevention by

-------
Water Fog in Loss Prevention, Volume 10, American Institute of Chemical
Engineers, New York, 1976.

Watts, J.W.  Effects of Water Spray on Unconfined Flammable Gas in Loss
Prevention, Volume 10, American Institute of Chemical Engineers, New York,
1976.

Zalosh, R.G.  Dispersal of LNG Vapor Clouds with Watier'Spray Curtains.  I.
Chem. E. Symposium Proceedings.  The Containment and Dispersion of Gases by-
Water Sprays, Manchester, England, 1981,                              -   ,


STEAM CURTAINS                                      j

Barker, G.F., T.A. Kletz, and H.A. Knight.  Olefin Plant Safety During the
Last 15 Years in Loss Prevention, Volume 11, American Institute of Chemical
Engineers, New York, 1977.

Bockmann, T., G.H. Ingebrigtsen, and T. Hakstad.  Safety Design .of the
Ethylene Plant in Loss Prevention, Volume -14, American Institute of'Chemical
Engineers, New York, 1981.

Cairney, E.M. and A.L. 'Cude.  The Safe Disposal of Large Clouds of Flammable
Heavy Vapours.  Institute of Chemical. Engineers Symposium Series No.  34, .
London, England, 1971.

Rulkens, P.F.M., et al.  The Application of Gas Curtains for Diluting .
Flammable Gas Clouds to Prevent Their Ignition.  4th International Symposium
on Loss Prevention, Harrogate, England, 1983.

Seifert, H., B. Maurer, and H. Giesbrecht.  Steam Curtains - Effectiveness and
Electrostatic Hazards.  4th International Symposium on Loss Prevention,
Harrogate, England, 1983.                                 	    .

Simpson, H.G.  The ICI-Vapout Barrier - A Means of Containing and Dispersing
Leakages of Flammable Vapour.  Power and Works Engineering-, May 8, 1974.


FOAM SYSTEMS                            ,            :               '      '

Chandnani, M.F.  Design Fundamentals.   Fire Protection Manual for Hydrocarbon
Processing Plans, Volume 1, Third Edition, Houston, 1985.

Evans, M.L. and H.A. Carroll.  Handbook for Using Foams to Control Vapors from
Hazardous Spills.  EPA report 600/8-86/019, July 1986.

Gillespie, P.J. and L.R. DiMaio.  How Foam Can Protect HPI Plants.  Fire
Protection Manual for Hydrocarbon Process Plants.  Volume 2, Houston, 1981.

Gross, S.S. and R.H. Hiltz.  Evaluation of Foams for Mitigating Air Pollution
from Hazardous Spills.  EPA Report 600/2-82-029, March 1982.
                                            .  .  . ..           .   .       , j.
Hiltz, R.H.  The Potential of Aqueous.Foams to Mitigate the Vapor Hazard from
Released Volatile Chemicals.  International Symposium on Preventing Major

-------
 Chemical Accidents.  American Institute  of Chemical Engineers, New York,
 February 1978.                                                     .
VAPOR DISPERSION MODELING

Bardley, G.I. andR.J. Carpenter.  The  Simulation of Dense Vapour Cloud
Dispersion Using Wind Tunnels and Water Flumes.  4th International Symposium
on Loss Prevention and Safety Promotion in  the Process Industries. The
Institution of Chemical Engineers, Symposium Series No. 80, London, England
1983.

Cheah, S.C., J.W. Cleaver, and A. Millward.  The Physical Modelling of Heavy
Gas Plumes in a Water Channel.  4th International Symposium on Loss Prevention
and Safety Promotion in the Process Industries.  The Institution of Chemical
Engineers, Symposium Series No. 80, London, England, 1983.

Havens, J.A.  Evaluation of 3-D Hydrodynamic Computer Models for Prediction of
LNG Vapor Dispersion in the Atmosphere.  Gas Research Institute, Chicago.
Contact No. 5083-252-0788, Annual Report, March 1984 - February 1985.

Mackison, F.W., R.S. Stricoff (eds.).   NIOSH/OSHA Pocket Guide to Chemical
Hazards.  U.S. Government Printing Office,  DHHS (NIOS) Publication No. 78-
210, Washington, D.C., 1980.

McNaughton, D.J., G.G. Worley, and P.M. Bodner.  Evaluating Emergency Response
Models for the,Chemical Industry.  Chemical Engineering Progress, January
1987.                                  '.  •   . ...   - •

McNaughton,,D.J., A.A. Marshall, P.M. Bodner, and G.G. Worley.  Evaluation and
Assessment of Models for Emergency Response Planning,  prepared for Chemical
Manufacturing Association by TRC Environmental Consultants, Inc., 1986.

Van Olden, A.P.   On the Spreading of a Heavy Gas Released Near the Ground.-
1st International Loss Symposium, The Hague, Netherlands, 1974.

Van Olden, A.P.   A New Bulk Model for Dense Gas Dispersion:  Two-Dimensional
Spread in Still Air.  I.U.T.A.M. Symposium  on Atmospheric Dispersion of .-Heavy
Gases and Small Particles.  Delft University of Technology.  The Netherlands,
1983.                     ;            -...-•          -.     ,

Wighus, R.  Simulation of Propane Dispersion from Spills in a Petrochemical
Plant Using a Water Tunnel Technique.  4th  International Symposium on Loss
Prevention and Safety Promotion in the Process Industries..  The. Institution of
Chemical Engineers, Symposium Series No. 80, London, England,  1983.

       **: *"•                        !                F
PLANT SITING AND LAYOUT

McDavid,  Catherine.  "Prevention of Environmental and Public Health .Damage
from Major Industrial Accidents through Innovative Siting Approaches" in
Avoiding, and Managing Environmental Damage  from, Major Industrial Accidents.

-------
Air Pollution Control Association Conference, Vancouver, Canada, November
1985.                                              i

Mecklenburgh, J.C.  Plant Layout.  Leonard Hill, London, 1973.

Miller, C.F., F.A. Leone, and W. Bryan.  Liquified Gaseous Fuels Spill Test
Facility:  Site Evaluation Study.  Lawrence Livermore National Laboratory,
Berkley, CA, February 1985.
                         '           .                                •     %V-*.' s" ' •
Ridner, R.M. and S. Wachtell, "Establishment of Design Criteria for Safe...
Processing of Hazardous Materials," in Loss Prevention, Volume 7.  AIChE, New
York, 1973.     ,                                   I                '

RPC, Inc.  Siting Industrial Facilities on the Texas Coast.  Texas Coastal
Management Program, Texas General Land Office, Austin, Texas, September 15,
1978.                                  -           • |
                                                   '       .    •.          ' \'.,
Shrivastava, Paul.  "The Accident at Union Carbide Plant in Bhopal:  A Case
Study in Avoiding and Managing Environmental Damage from Major Industrial ,
Accidents."  Air Pollution and Control Association Conference, Vancouver,
Canada, November 1985.
                                                   i                          '
                                                   "i  '
METEOROLOGICAL INSTRUMENTATION                     !

Ambient Monitoring Guidelines for Prevention of Significant Deterioration .
(PSD).  U.S. Environmental Protection Agency Publication No. EPA-450/4-80-
012, November 1980.                                ,   .
                                                   i
Boykin, Raymond F. and Mardyros Kazarians.  Quantitative Risk Assessment for
Chemical Operations.  Pickard, Lowe, and Garrick, Inc.  International
Symposium on Preventing Major Chemical Accidents, Washington, D.C., 1987.

Guideline on Air Quality Models  (Revised).  U.S. Environmental Protection,
Agency Publication No. EPA-450/2-78-027R, July 1986.

Quality Assurance Handbook for Air Pollution Measurement Systems, Volume IV:
Meteorological Measurements.  U.S. Environmental Protection Agency Publication
No. EPA-600/4-82-060, February 1983.               ;            .
                                                   |
Strimaitis, David, G. Doffngole, and A. Bass.  On-Site Meteorological  ,,
Instrumentation Requirements to Characterize Diffusion from Point Sources.
EPA-600/9-81-020, C-l-C-5.

-------
                          MONITORING AND DETECTION
Adrain, R.S. et al. The Measurement of SC>2 in Power Station Plumes with
Differential LIDAR.   Opt; and Quantum Electron (GB), Vol. 11, No. 3.  pp.
253-64, May, 1979.  .

Ahmed, S.A. et al. Remote Monitoring of Gaseous Pollutants by Differential
Absorption Laser Techniques. EPA-600/2-80'049, p. 212. Feb, 1980.

Alden, M. Applications of Laser Techniques for Combustion Studies, DE
83751402/XPS.  Lund Inst. of Tech. (Sweden), FYSIKA Institutionen, Report No.
LRAP-22, 158 pp.> 1984.

Alden, M. Remote Measurement of Atmospheric Mercury using Differential
Absorption LIDAR. Opt. Lett. (USA), Vol. 7, No. 5.  pp. 221-3, May,      1982.

Altmanri, J. et al. Remote Measurement of Atmospheric N20 with a DF Laser
LIDAR, Applied Optics v!9 n20  p. 3453-3457, Oct. 15, 1980.

Ancellet, G. et al. LIDAR Measurements of Atmospheric Pollutants (S02, 63) DE
85751349/XPS. International LASER LIDAR Conf., Aix-En-Provence, France,
August, 1984.                                                       ,

Anlauf, K. G. et al. Comparison of Three Methods for Measurement of
Atmospheric Nitric Acid ggand Aerosol Nitrate and Ammonium.  Atmospheric
Environment v!9 n2  p. 325-333, 1985.

Bakalyar, D.M. et al. Shot-Noise Limited Detection of OH Using the Technique
of Laser Induced Fluorescence, N84-20844/6/XPS. Oakland Univ., Rochester, MI,
Dept. of Physics., Report No. NAS 1.26:175433, 30 p., 1984.

Baranbv, P.A. et al.  Dye Laser for LIDAR Ozonometry. ZH. PRIKL. SPEKTROSK.
(USSR), Vol. 36, No. 4.  pp. 574-7, April,  1982.

Baranov,- P. A. et al. Parameter Stability in a Dif f erentical-Absorption LIDAR
for^Monitoring Nitrogen Dioxide. Journal of Applied Spectroscopy v41 n2 p.
895-898. Aug. 1984.                                                    -

Baumgartner, R.A. et al.  Continuously Tunable IR LIDAR with Applications to
Remote Measurements of S02 and CH^.. Appl. Opt. (USA), Vol. 17, No. 22.  pp.
3555-61, November, 1978.

Bethke. G. W. et al. Mobile LIDAR System Developments and Operating
Procedures.  General Electric Co.  EPA-600/2-79-197, November, 1979.

Breinig, A. et al.  Data Collection and Analysis for HCL, S02, and N02
Imission - Emission Differential Absorption LIDARS. DE86 901106/XPS.  A.
Breinig etal. Workshop On Differential Absorption LIDAR (DIAL) Data Collection
and Analysis Techniques.  Virginia Beach, VA, USA, 9 p., November, 1985.

-------
Bordonali, C. et al.  Remote Sensing of Gaseous Pollutants with Laser
Ann.1st. Super Sanita (Italy), Vol. 19, No. 4.  pp. ;569-77, 1983.

Browell, E.V. LIDAR Measurements of; Tropospheric Gases.  Optical Engineering
v21 nl,  p. 128-132. Jan rFeb. 1982.                ;  ;                      -

Burch, D.E. and Gryvnak, D. A.Remote Monitoring of Nitic Oxide by Gas-filter
Correlation Techniques.  EPA-600/2-76-277 November, 1967.    ,

Byer, R.L. et al. Remote Measurements of Trace Species in the Troposphere.
A1AA J. (USA), Vol. 20, No. 3.  pp. 395-403, March, 1982.

Camagni, P. et al. Optical Remote Sensing of Air Pollution.  Lectures of a
Course. (Editors) ISPRA, Italy; Elsevier, Amsterdam,| Netherlands, VIII,  pp.
422, 1984.                                          :

Camagni, P. et al. Study of Stack Emissions by Combination of LIDAR and
Correlation Spectrometer.  NUOVO CIMENTO C  (Italy), Vol. 4C, Ser. 1, No. 3.
pp. 359-71, May-June, 1981.       ,               '.•;,:.

Capitini, R. et al. DIAL LIDAR Measurements of Atmospheric Pollutants (S02,
03) During the Fos-Berre 83 Experiment. 12  International Laser Radar
Conference Aix-en-Provence, France,  p.'6  Report-No;. CEA-CONF-7-384 Aug. 13
1984.                                       .        ;       ;           ;
                                            :'"!«'
Capitini, R. LIDAR Teledetection and Measurement of Minority Atmospheric
Constituents.  Bull. Inf. Sci.'and Tech. (France), No. 230-231, 101-8,
July-October, 1978.            •      •               •            ,
                                                    I            :          '
Corio, W. et al. The Remote Measurement of Atmospheric Pollution using Laser
Techniques. ALTA FREQ.  (Italy),.Vol. 48, No. 3.  pp. 119-29, March, 1979.

Edner, H.    Gas-Correlation LIDAR (Air Pollution Remote Sensing). Opt. Lett.
(USA), Vol. 9 No. 11.   pp. 493-5, November, 1984.

Edner, H. LIDAR Techniques for Atmospheric  NO Monitoring, DE 83751403/XPS.
Lund Inst. of Tech. (Sweden) FYSIKA Institutionen, Report No. LRAP-21, 21 p..,
1983.

Edner, H. et al. UV LIDAR Techniques for Atmospheric NO Monitoring.     .•
Tehniska Hoegskolan, Lund (Sweden).  Fysiska Institutionen .1983 .  p.. 21, Report
No. LRAP-21                                         !

£ngoyan, T.M. et al.Determination of N02 Concentration in the Atmosphere
by Differential-Absorption LIDAR-. Journal of Applied ;Spectroscopyv39 nlrJuly,'
1983.  p. 809-813                                 , I                	

Features of Processing  LIDAR Signals in Measuring Gaseous Impurities in the
Atmosphere. Optoelectron/Instrum, Data Process/nl p. 83-87. 1984.
        ;        •                       ;-          ••        /I---   ,'-''"'yt  -.'•
Force, A.P. et al. Laser Remote Sensing of  Atmospheric Ammonia Using a,
C02 LIDAR System. Appl. Opt.  (USA), Vol. 24, No. 17J  pp. 2837-41, September,
1985.  ,                          •    ,,                        .•..--..

-------
Fredricksson, K. Mobile,LIDAR System for Environmental Probing. Applied Optics
v20 n24 December 15, 1981.  p. 4181-4189

Fredricksson, K; et al. Pollution Monitoring using Nd:YAG Based LIDAR
Systems.  Optical and Laser Remote Sensing.  Publ: Springer-Verlag, Berlin,
Germany, F.R.  p. 148-155. 1983.

Fredricksson, K.A.  et al. Evaluation of the DIAL Technique for Studies on N02
Using a Mobile LIDAR System.  Applied Optics v23 n9  p. 1403-1411, May 1,
1984. ,    :.                       .      . ;

Fredricksson, K.A. at al. LIDAR Measurements on Sulphur Dioxide Plumes, DE
85750231/XPS.  Lund. Inst. of Tech. (Sweden), FYSIKA INSTITUTIONEN, Report No.
LRAP-32, 42 p., October, 1983.                                       ,

Gage, K.S. et al. Advances In Remote Sensing of the Atmosphere. Rev. Geophys.
and Space Phys. (USA),  Vol. 21, No. 5.  pp. 955-65, June, 1983.

Galletti, E.  et al. Gas Correlation LIDAR for Methane Detection.  Thirteenth
International Laser Radar Conference. Report No. N-87-10263 NASA-CP-2431, Aug,
1984.                              .-.-.:,-•.•

Godlevskii,  A.P. et al. A Spectrochemical LIDAR for Analyzing the Elemental
Composition of the Atmospheric Aerosol.  ZH. PRIKL. SPEKTROSK (USSR), Vol. 39,
No. 5.  pp.  734-40, November, 1983.

Grant, W.B.     LIDAR Systems for Atmospheric Pollution Monitoring.  W.B, Grant
OSA/IEEE Conference on Laser and Electro-Optical Systems (Digest of Technical
Papers)  76,  1978.

Griggs,. M. Atmospheric Remote Sensing.  Proceedings of the Society of
Photo-Optical Instrumentation Engineers, Vol. 91.  Methods for Atmosphere
94-5, 1976.                        '     ,      :

Hardesty, R.M. Measurement of Range-Resolved Water Vapor Concentration by
Coherent C02 Differential Absorption LIDAR .PB84-236892/XPS.   National Oceanic
and Atmospheric Administration, Boulder, CO, Report No. NOAA-TM-ERL-WPL-118,
269 p.,  March, 1984.

Harris,  G. et al. Measurement of NOg and HN03 in Diesel Exhaust Gas by"Diode
Laser Absorption Spectrometry.  Environmental Science and Technology V21 n3 p.
299-304, March, 1986.

Hawley;J.G.    Plume Model Validation and Development Project:  Differential
Absorption LIDAR Measurements of S02 in Moderately Complex Terrain.  Final
Report.  SRI International, Menlo Park, CA (USA), p. 82 Report No. EPRI-EA-3758
May, 1985.

Hawley,  J.G.  et al. Ground-Based Ultraviolet Differential Absorption LIDAR
(DIAL) System and Measurements. Springer Series in Optical Sciences v39  p.
128-137. 1983.                         :                               :

Hawley,  J.G.  et al. Mobile Differential Absorption LIDAR (DIAL) for Range
Resolved Measurements of S02, 03, and N02-   A1AA Pap. AlAA Aerosp. Sci.  Meet.
19th, St. Louis, MO, p. 6, 1981.

-------
Hawley, J.G. et al.  Plume Model Validation and Development Project:  Plains
Site LIDAR Measurements of Sulfur Dioxide.  Final Report.  SRI International,
Menlo Park, CA (USA), p. 137 Report No. EPRI-EA-307J2, December, 1983.

Hawley, J.G. et al. Ground-Based Ultraviolet Differential Absorption LIDAR
(DIAL) System and Measurements.  Optical and Laser Remote Sensing. SRI
International, Menlo Park, CA.  Publ:  'Springer-Verlag, Berlin, Germany, F.R,
p. 128-137, 1983.

Heaps, W.S. Measurement of Hydroxyl Radical in the Upper Atmosphere using
LIDAR from the Space Shuttle. Applied  Optics v!9 n2  p. 243-249. 1980.

Heaps, W.S. et al. Balloon Borne LIDAR Measurements of Stratospheric Hydroxyl
Radical. Journal of Geophysical Research v88 nC9  p. 5281-5289. 1983.

Heaps, W.S. et al.  Stratospheric Ozone and Hydroxyl Radical Measurements by
Balloon-Borne LIDAR.  Applied Optics v21 n!2  p. 22,65-2274. June 15, 1982.

Henrich, H.J. et al. Hydrogen Chloride Measurements with a DAS LIDAR in the
Marine Atmosphere. International Geoscience and Remote Sensing Symposium
Strasbourg, France 27 p. 10. August, 1984.

Hertel, W.R.   Detection of Atmospheric Carbon Monoxide from a Shuttle-Borne
LIDAR.  AD-A164271/9/XPS. Air Force Inst. of Tech., Wright-Patterson AFB, OH,
School of Eng., Report No. AFITGEPPH-83D-4, 137 p., December, 1983.

Hill, J.C. et al.  Time-Resolved Measurement of Vehicle Sulfate and Methane
Emissions with Tunable Diode Lasers. SAE Preprints n 800510 for Meet p. 8,
Feb. 25-29, 1980.

Houston, J.D, et al. Gas Dispersion Measurements using a Mobile Raman LIDAR
System. Thirteenth International Laser Radar Conference.   Report No.
N-87-10263; NASA-CP-2431. Aug. 1986.

Houston, J. D. et al. Raman LIDAR System for Methane Gas Concentration
Measurements. Optech Inc., Downsview,  Ont., Canada.  Appl. Opt. (USA) Vol. 25,
No. 13.  p. 2115-21, 1 July,  1986

Hull, R.J. et al. Coherent Laser Radar Remote Sensing, AD-A139448/5/XPS.
Mass. Inst. of Tech. Lexington, Lincoln Lab., Final Report, 64 p  September,
1983.

Johnson, D.W. et al. Duguay Test Grid  LIDAR Project.   Workshop On  Remote
Sensing for Chemical Defense.  Myrtle  Beach, S.C., USA, p. 4  Report No.
PNL-SA-13143; CONF-8504142-1. April, 1985.

Kibler, J.F. et al.  Air Pollution Model  Parameter Estimation using Simulated
LIDAR Data. A1AA J.  (USA), Vol. 15 No. 10.  pp.  1381-4,  October,  1977.

Killinger, D.K. et al.  Laser  Remote  Sensing of Atmospheric Pollutants.  Final
Report AD-A 159531/3/XPS.  Mass Inst.  of  Tech.,  Lexington, Lincoln  Lab.,  Proj.
No.  1900,  65 p., September,  1984.

-------
Kokkinos, D.S. LIDAR System Using Polarization Discrimination Techniques for
Molecular air Pollution Monitoring.  Publ. City Univ. of New York, N.Y., N.Y.,
p. 218 Thesis (Ph.D.) 1984.

Lahmann, W. et al.  State-of-the-Art DAS LIDAR for S(>2 and N02. International
Geoscience and Remote Sensing Symposium, Strasbourg, France. P. 7. August, 27
1984.
Laser Remote Sensing of Environmental and Climatic Parameters in the
Troposphere.  TIB/B87-00621/XPS.  Technical Fair and Conference on Ultrahigh
Frequency Technology:  Microwave Technology and Optoelectronics (MIOP 86)
Wiesbaden (Germany F.R.) 24 p., June, 1986.

Levine, J.S. et al. Development of Tunable High Pressure CC>2 Laser for LIDAR
Measurements of Pollutants and Wind Velocities - October, 1977 through
January, 1979. NASA Contractor Reports n3274, p.23 April. 1980.

Lo, Wayne. Turnable Diode Laser Development and Spectroscopy Applications.  GM
Research Lab., Warren, MI, USA.  Proceedings of SPIE - The International
Society for Optical Engineering v438, Turnable Diode Laser Dev. and Spectrosc.
Appl., San Diego, CA, USA, August 25-26, 1983.  Publ. by SPIE, Bellingham, WA,
USA, p. 203, 1983.

Loda, R.T. Remote Detection of Polynuclear Aromatic Hydrocarbons Using
Laser-Induced Fluorescence, AD-A152294/5/XPS. Naval Weapons Center, China
Lake, CA, Report No. NWC-TP-6587, 39 p., August, 1984.

Lunqvist, S. Air Pollution Monitoring with a Q-Switched C02 Laser LIDAR using
Heterodyne Detection. Appl. Opt. (USA), Vol. 20, No. 14.  pp. 2534-8, July,
1981.

Lundqvist, S. et al. Air Pollution Monitoring with a Q-Switched C02 Laser
LIDAR using Heterodyne Detection. Applied Optics v20 n!4, p.2534-2538, July
15, 1981.

Magee, T.J.  GSFC Ground-Based LIDAR for the Measurement of Stratospheric
Ozone.  Thirteenth International Laser Radar Conf.  Report No. N-87-10263;
NASA-CP-2431, August, 1986.

Mantion, J.B. et al. Distant Detection of a Gaseous Contaminant by Means of
a Raman LIDAR. Tec. Regul. and Mando Autom. (Spain), Vol. 8, No.  54,  December
1975.

Marichev, V.N. et al. Optimization of Spectial Measurements in LIDAR Sounding
of Atmospheric Gases by the Method of Differential Absorption.   1ZV. VUZ F1Z
(USSR), Vol. 28, No. 3.  pp. 47-51; Inst. of Atmos. Opt., Acad. of Sci.,  USSR,
Date:  March, 1985.

Megie, G. et al.  Ozone and Water Vapor Monitoring using a Ground-Based LIDAR
System.  Optical and Laser Remote Sensing.   Springer-Verlag, Berlin,  Germany,
p.223-228. F.R. 1983.

Megie, G. et al. Ozone and Water Vapor Monitoring Using a Ground-Based LIDAR
System. Springer Series in Optical Sciences v39,  p. 223-228. 1983.

-------
Megie, G.J. et al. LIDAR Measurements of Ozone Vertical Profiles.  Appl.  Opt.
(USA) Vol. 24, No. 21.  pp. 3454-63, November, 1985.                        •

Morradian, A. Application of Laser Technology to Atmospheric Monitoring.
Proceedings of the 8th Materials Research Symposium,  pp. 277-86, 1977.

Menyuk, N. et al. Laser Remote Sensing of Hydrazine MMH and UDMH Using a
Differential-Absorption C02 LIDAR. Applied Optics v21 n!2 P.2275-2286. June
15, 1982.                                                                .

Menyuk, N. et al. Remote Sensing of NO using a Differential Absorption LIDAR.
Applied Optics v!9 n!9 p. 3282-3286, October, 1980.

Molecular Studies with Turnable Diode Lasers.  Laser Focus (Littleton Mass)
V21 n6 June, 1985.  p. 136-142

Molina, L.T. et al. FTIR - Spectrometer-Determined Absorption Coefficients of
Seven Hydrazine Fuel Gases:  Implications for Laser Remote Sensing.  Applied
Optics v23 n21 November 1, 1984

Murray, E.R. Remote Measurement of Gases using Discretely Tunable Infrared
Lasers.  Proc. of the Soc. of Photo-Optical Instrumentation Engineers, Vol.
95.  Modern Utilization Of 96-104, 1976.

Murray, E.R.  Remote Measurement of Gases Using Differential-Absorption LIDAR.
Opt. Eng. (USA), Vol. 17, No. 1.  pp. 30-8, Jan.-Feb., 1978.

Murray, E.R.   Remote Measurement of Ethylene using C02
Differential-Absorption LIDAR.  Appl. Opt. (USA), Vol. 17, No. 5.  pp. 814-17,
March, 1978.

Murray, E.R. et al. Remote Measurement of HCL, 014, and N02 using a
Single-Ended Chemical-Laser LIDAR System. Appl. Opt. (USA), Vol. 15, No.  12.
pp. 3140-8, December, 1976.

Nadler, S. et al. Tunable Diode Laser Measurements of Formaldehyde Foreign-Gas
Broadening Parameters and Line Strengths in the 9-11 MUm Region. Appl. Opt.
(USA) Vol. 26 No. 9.  p. 1641-6, 1 May, 1987
                                                    i                  '
Paulson, M.R. LIDAR Measurements Indicating Atmospheric Inhomogeneities.
AD-A166 440/8/XPS. Naval Ocean Systems Center, San DpLego, CA, Report No.  NO
SCTD-867, p.15  February, 1986.

Plume Model Validation and Development Project:  Differential Absorption LIDAR
Measurements of S02 in Moderately Complex Terrain. Electr. Power Res. Rep.
EPRI EA 3758, p.96. May,1985.                                            .

Poultney, S.K. Quantitative Remote Measurements of Pollutants from Stationary
Sources using Raman LIDAR.  Appl. Opt. (USA), Vol. 16, No. 12,  pp. 3180-2,
December, 1977.                                                       .

Pournag, J.C. et al. Application of Lasers to the Remote Measurement of
Atmospheric Gases.  Utilization of Raman Resonance Diffusion. AGARD Cpnf. ,
Proceedings. No. 300, Special Topics in Optical Propagation.  28th Mtg. of the
ELE 12/1-11/81.

-------
Reid, J. et al. High Sensitivity Detection of Trace Gases at Atmospheric
Pressure using Turnable Diode Lasers. Optical and Quantum Electronics v!7 nl
p.31-39.  January, 1985.

Remote and In-Situ Detection of Atmospheric Trace Gases-Infrared Spectroscopy
for Ammonia. Electr. Power Res. Inst. Rep/EPRI EA 4370, p.192.  December,
1985.
Rosen, H. et al. Remote Detection of Pollutants using Raman Scattering.  Appl.
Opt,  (USA), Vol. 14, No. 11.  pp. 2703-6, November, 1975.

Sachse, G.E. et al.Fast-Response, High-Precision Carbon Monoxide Sensor using
a Tunable Diode Laser Absorption Technique. J. Geophys. Res. (USA) Vol. 92 No.
D2.  p. 2071-81 20 February, 1987

Schiff, H.I. et al.Tunable Diode Laser Systems for Measuring Trace Gases in
Tropospheric Air.  H.I. Schiff etal., Environmental Science and Technology v!7
n8  p. 352A-364A. 1983.

Sochor, V, Optical Methods of Detecting Air Pollution.  JEMNA MECH. and OPT.
(Czechoslovakia), Vol. 23, No. 6.  pp. 169 -73 •, June, 1978.

Selmr, F. et al., J. Geophys. Measurement of Gas Phase Hydrogen Peroxide in
Air by Tunable Diode Laser Absorption Spectroscopy. Res. (USA) Vol. 91 No. D5.
p. 5371-8 30 April, 1986

Shibata, T. et al. Excimer LIDAR Measurements of Ozone.  International Council
of Scientific Unions MAP Symposium Kyoto, Japan 26 November, 1984,  Report No!
MAP-HANDBK-8, December, 1985.

Sperling, R.B,  Evaluation of the Correlation Spectrometer as an Area SOo
Monitor.  EPA-600/2-75-077, October, 1975.

Staehr. W. et al. State-of-the-Art DAS LIDAR for S02 and N02 DE 85752203/XPS.
International Geoscience and Remote Sensing Symposium, Strasbourg, France, p.
7. Report No. GKSS-84E50, August, 1984.

Stewart, R. et al. Development of a Pulsed 9.5 MU m LIDAR for Regional Scale
03 Measurement.  Optical Engineering v!9 n4 p.503-507. July-August, 1980.

Sugimoto, N. et al. Remote Measurement of N02 Profile in a Stack Plume by a
Differential Absorption LIDAR.  OYO BUTURI (Japan), Vol. 50, No. 9.  pp.
923-8, September, 1981.

System for the Depth-Resolved Remote Measurement of the Gaseous Atmospheric
Pollutants N02 and S02 by the Differential Absorption and Backscattering
Techniques. DE/86751968/XPS, Report 6KSS-85E48, 126 p., 1985.

Takeuchi, N. et al. Detectivity Estimation of the DAS LIDAR for N02.  Appl.
Opt.  (USA), Vol. 17, No. 17.  pp. 2734-8, September, 1978.

Takeuchi, N. et al. Wide-Area Air Pollution Measurement by the NIES Large
LIDAR.  Springer Series in Optical Sciences v39   p. 364-373,  1983.

-------
Uchino, 0. et al.  Observation_ of Stratospheric Vertical Ozone Distribution by
a XeCL LIDAR.  Journal of Geophysical Research v88 nC9  p. 5273-5280, June 20,
1983.                                               ;

Uthe, E.E.  Airborne C0£ DIAL (Differential Absorption LIDAR) Measurement of
Atmospheric Tracer Gas Concentration Distributions.   SRI International, Menlo
Park, CA  (USA) p. 8 Report No. AD-A-172683/5/XAB. August 1, 1986.

Uthe, E.E. Application of Surface Based and Airborne: LIDAR Systems for
Environmental Monitoring. Journal of the Air Pollution Control Association,
Vol. 33, No. 12 (December),  p. 1149, 1983.         :

Uthe, E.E. et al.  Airborne LIDAR Mapping of SFg Concentration Distributions
for Transport and Diffusion Studies.  Sponsored by Nuclear Regulatory
Commission, Wash., B.C., Office of Nuclear Regulatory, NUREG/CR-4698/XPS, 39
p., August, 1986.

Vainer, Yu Raman. LIDAR for Analysis of Industrial Atmospheric Pollutants.
KVANTOVAYA ELEKTRON, MOSKVA (USSR), Vol. 6, No. 3.  pp. 494-9, March, 1979.

Walega, J.G. et al. Comparison of a Chemiluminescent and a Turnable Diode
Laser Absorption Technique for the Measurement of Nitrogen Oxide, Nitrogen
Dioxide and Nitric Acid.  Environmental Science and Technology v!8 nil p. 823-
826. November, 1984.

Wall, D.L. New Developments in Tunable Diode Laser Sources Relevant to
Infrared Fiberoptics Technology. Proc. SPIE Int. Soc. Opt. Eng.  (USA) vol.
618.  p. 199-203, 1986.

Wank, R.H. Detection of Atmospheric Carbon Dioxide from a Shuttle-Borne LIDAR.
AD-A164308/9/XPS.  Air Force Inst. of Tech., Wright-Patterson AFB, OH, School
of Eng., Report No. AFITGEPH82D-25, 133 p., December, 1982.

Wang, R.H. Detection of Atmospheric Carbon Dioxide from a Shuttle-Borne LIDAR.
Air Force Inst. of Tech. Wright-Patterson AFB, OH  (USA).  School of
Engineering, p. 133  Report No. AFITGEPH-82D-25.  December 1982.

Weitkamp, C. Remote Measurement of Tropospheric Parameters by LIDAR.
Workshop on Experiences with the Application of Advanced Air Pollution
Assessment Methods and Monitoring Techniques.  Lindau, Bodensee, F.R. Germany,
p. 16. October 2, 1985.

Weitkamp, C. Remote Measurement of Trophospheric Parameters by LIDAR.
DE86901107/XPS.  Workshop on Experiences with the Application of Advanced Air
Pollution Assessment Methods and Monitoring Techniques. 16 p., Lindau,
Bodensee, F.R. Germany, October, 1985.
                                                    i'
Weitkamp, C. et al. Hydrogen Chloride Measurements with a DAS LIDAR in the
Marine Atmosphere. 752202/XPS. International Geoscience and Remote Sensing
Symposium, Strasbourg, France,  Report No. GKSS-84E49, 10 p. August, 1984.

Weitkamp, C. et al. Propagation of Hydrogen Chloride  in Plumes of Incineration
Ships:  Preliminary Results of the Measurement Campaign with the Research Ship
TABASIS DE 83751305/XPS. Report No. GKSS-83E10, 101 p., 1983.

-------
Zaromb, S.  Field Evaluation of a System for the Remote Sensing of Gaseous Air
Pollutants. Final Report, Grant No. R800834 USEPA.  Research Triangle Park,
N.C. August 10, 1973.

-------

-------
                               WARNING SYSTEMS


Albert, M. B. and L. Segaloff. "Task Silence:  the post-midnight alarm and
evacuation of four communities affected by an ammonia gas release."
Philadelphia: Institute for Cooperative Research, University of Pittsburgh
(unpublished paper cited in Quarantelli, 1980), 1962.

Anderson, L. , J. Keaton, T. Saarinen and W. Wells II. The Utah Landslides,
Debris Flows and Floods of May and June 1983. Washington, DC: National Academy
Press, 1984.

Anderson, W. "Tsunami warning in Crescent City, California and Hilo, Hawaii,"
in The Great Alaska Earthquake of 1964:  Human Ecology, Committee on the
Alaska Earthquake of the National Research Council (ed.). Washington, DC:
National Academy of Sciences, 1970.

Anderson, W. "Disaster warning and communication processes in two
communities," The Journal of Communication 19 (June): 92-104, 1969.

Anderson, W. "Seismic sea wave warning in Crescent City, California and Hilo,
Hawaii," Working Paper No. 11.  Columbus, OH:  Disaster Research Center, Ohio
State University (cited in Quarantelli, 1980), 1966.

Anderson, W. "Crescent City revisited:  a comparison of public warning
procedures used in the 1964 and 1965 emergencies," Research Note No. 11.
Columbus, OH:  Disaster Research Center, Ohio State University (cited in
Quarantelli, 1980), 1965.

Anderson, W. "The Baldwin Hills, California dam disaster," Research Note No.
5.  Columbus, OH:  Disaster Research Center, Ohio State University (cited in
Quarantelli, 1980), 1964.

Baker, E. J. "Hurricane Elena: Preparedness and response in Florida," Florida
Policy Review (Winter): 17-23, 1986.

Baker, E. J. "Public response to hurricane probability forecasts," NOAA TM NWS
FCST 29.  Washington, DC:  National Weather Service, 1984a.

Baker, E. J. "Deciding whether to evacuate a beach community during a
hurricane threat," (Unpublished Manuscript), Florida State University,
Tallahassee, Florida, 1984b.

Baker, E. J. "Coping with hurricane evacuation difficulties," pp.  13-18, in E.
J. Baker (ed.) "Hurricanes and Coastal Storms," Florida Sea Grant College
Report No. 33, 1980.

Baker, E. J. "Predicting response to hurricane warnings: A reanalysis of data
from four studies," Mass Emergencies 4: 9-24, 1979

Baker, E. J. and T. M.  Carter. "The Role of information in public response to
warnings," Paper presented at the American Meteorology Society Conference
(Draft), 1984.

-------
Baker, E. J., J. C. Brigham, J. A. Paredes, and D.D. Smith. Social Impact of
Hurricane Eloise on Panama City, Florida, Florida Sea Grant Report.
Gainesville:  University of Florida, Sea Grant Program, 1976.

Bartlett, G.S..P. S. Houts, L.K. Byrnes and R.W. Miller. "The near disaster at
Three Mile Island," International Journal of Mass Emergencies and Disasters 1
(March): 19-42, 1983.

Barton, A. Communities in Disaster. Garden City, NY: Doubleday,  1969.

Bates, F. L., et al. The Social and Psychological Consequences of a Natural
Disaster, National Research Council Disaster Study 18. Washington, DC:
National Academy of Sciences, 1963.

Blum, R. and B. Klass. A Study of Public Response to Disaster Warnings.   Metro
Park, CA:  Standford Research Institute (cited in Quarantelli, 1980),  1956.

Boek, W. and J. Boek. "An exploratory study of reactions to an impending
disaster."  Albany, NY:  New York State Department of Health (paper cited in  '.
Quarantelli, 1980), 1956.

Bolin, R. "Disaster characteristics and psychological impacts," pp.  3-28 in B.
Sowder (ed.), Disasters and Mental Health: Selected Contemporary Perspectives.
Rockville, MD:  National Institute of Mental Health,!U.S.  Department of Health
and Human Services, 1985.

Bonk, W., R. Lachman and M. Tatsuoka. A Report of Human Behavior During the
Tsunami of May 23, 1960.  Hilo, HI:  Hawaiian Academy of Science, 1960.

Brunn, S., J. Johnson and D. Ziegler. Final Report on a Social Survey of Three
Mile Island Residents.  East Lansing, MI:  Department of Geography,  Michigan
State University, 1979.                              ,

Burton, I.  The Mississauga Evacuation, Final ReportI  Toronto:  Institute for
Environmental Studies, University of Toronto, 1981.

Carter, T. M. Probability of Hurricanes/Tropical Storms: A Users Guide for
Local Decision-Makers.  Silver Spring, MD:  NationaliWeather Service,  1983.

Carter, T. M., S. Kendall and J. P. Clark. "Household response to warnings,"
International Journal of Mass Emergencies and Disasters 1  (March): 95-104,
1983.

Carter i T. M. , J. P. Clark, and R. Leik. "Organizational and household
response to hurricane warnings in the local community," paper presented at the
AAAS Meetings, Houston, TX. 1979.

Chemical Manufacturing Association. Emergency Warning Systems. Washington, DC:
CMA, 1987.                           •                i
                                                     |   - :.    .  .
Chiu, A., et al.  Hurricane Iwa, Hawaii, November 23 j 1982. Washington,  DC:
National Academy of Sciences, 1983.              '  '  '.

-------
Christensen, L. and C. E. Ruch. "The effect of social influence on response to
hurricane warnings," Disasters 4 (2): 205-210, 1978.

Christensen, L. and C. E. Ruch. "Assessment of brochures and radio and
television on hurricane awareness," Mass Emergencies 3: 209-216, 1980.

Clark, J. and M. Carter. "Response to hurricane warnings as a process:
determinants of household behavior," pp. 19-24, in E. J. Baker (ed.)
Hurricanes and Coastal Storms, Florida Sea Grant College Report No. 33, 1980.


Clifford, R. A. The Rio Grande Flood:  A Comparative Study of Border
Communities, National Research Disaster Study #17.  Washington, DC:  National
Academy of Sciences, 1956.

Cochrane, H.,  E. Gruntfest, M. Stokes, H. Burgess, G. Burgess and L.
Steinback. Flash Flood on the Big Thompson: A Case Study. Denver, CO: Western
Governors' Policy Office for Policy Research, 1979.

Committee on Natural Disasters. Hurricanes Iwa, Alicia and Diana - Common
Themes.  Washington, DC: National Academy Press, 1985.

Committee on Socioeconomic Effects of Earthquake Predictions. A Program of
Studies on the Socioeconomic Effects of Earthquake Predictions.  Washington,
D.C.:  National Academy of Science, 1978.

Cutter, S."Emergency preparedness and planning for nuclear power plant
accidents," Applied Geography 4: 235-245, 1984.

Cutter, S. and K. Barnes. "Evacuation behavior at Three Mile Island,"
Disasters 6: 116-124, 1982.

Danzig, E. R., P. W. Thayer and L. R. Galanter. The Effects of a Threatening
Rumor on a Disaster-Stricken Community, Disaster Study #10. Disaster Research
Group.  Washington, DC:  National Academy of Sciences, 1958.

Davenport, S.  "Human response to hurricanes in Texas:  Two studies," Working
Paper No. 34.   Boulder, CO: Institute of Behavior Science, University of
Colorado, 1978.

Diggory, J. C.  "Some consequences of proximity to a disease threat,"
Sociometry 19 (March):47-53, 1956.

Dillman, D., M. Schwalbe, and J. Short. "Communication behavior and social
impacts following the May,  18, 1980 eruption of Mt St. Helens," in Mt. St.
Helens: Three Years Later,  Proceedings of a Conference.  Pullman, Washington,
Washington State University, 1983.

Drabek, T. E.  Human System Response to Disaster: An Inventory of Sociological
Findings.  New York, NY: Springer Velag, 1986.

Drabek, T. E.  "Shall we leave?  A study of family reactions when disaster
strikes," Emergency Management Review 1 (Fall): 25-29, 1983.

-------
Drabek, T. E. "Social processes in disaster:  Family evacuation," Social
Problems 16 (Winter):  336-349, 1969.

Drabek, T. E. and K. Boggs. "Families in disaster:  Reactions and relatives,"
Journal of Marriage and the Family 30 (August):  443-451, 1968.

Drabek, T. E. and J. S. Stephenson, III. "When disaster strikes," Journal of
Applied Social Psychology 1 (2): 187-203, 1971.     !

Drabek, T. E. , H. L. Tamminga, T. S. Kilijanek and G. R. Adams. Managing
Multiorganizational Emergency Responses:  Emergent Search and Rescue Networks
in Natural Disasters and Remote Area Settings.  Boulder, CO:  Institute of
Behavioral Science, University of Colorado, 1981.

Dynes, R. Organized Behavior in Disasters.  Lexington, MA: D. C. Heath, 1970.
                                                    i
Dynes, R. et al. Report of the Emergency Preparedness and Response Task Force,
Staff Report to. the President's Commission on the Accident at Three Mile
Island.  Washington, DC: U.S. Government Printing Office, 1979.

Erikson, K. T. Everything In Its Path.  New York:  Simon and Schuster, 1976.

Fisher, D. "Planning for large-scale accidents:  Learning from the Three Mile
Island," Energy 6: 93-108, 1981.

Fitzpatrick, J. and J. Waxman. "The March, 1972 Louisville, Kentucky chlorine
leak threat and evacuation," Working Paper #44.  Columbus, OH: Disaster
Research Center, Ohio State University  (cited in Quarantelli, 1980), 1972.

Flood Loss Reduction Associates. Technical Guidance Manual for Local Flood
Warning and Preparedness Programs.  New York: State Department of
Environmental Conservation, 1984a.

Flood Loss Reduction Associates. Prototype Local Flood Warning Plan.  New
York: State Department of Environmental Conservation, 1984b.

Flood Loss Reduction Associates. Prototype Local Flood Warning and
Preparedness Program.  New York: State Department of Environmental
Conservation, 1984c.

Flynn, C.  "Reaction of local residents to the accident at Three Mile Island,"
pp. 49-63 in Accident at Three Mile Island, edited by D. Sills and others.
Boulder, CO: Westview Press, 1982.

Flynn, C. Three Mile Island Telephone Survey:  A Preliminary Report.
Washington, DC: U.S. Nuclear Regulatory Commission, 1979.

Flynn, C. B. and J. A. Chalmers. The Social and Economic Effects of the
Accident at Three Mile Island.  Tempe, AZ: Mountain iWest Research, Inc. with
Social Impacts Research, Inc. 1980.            -     ,    .

Forrest, T. R. Hurricane Betsy, 1965; A Selective Analysis of Organizational
Response in the New Orleans Area, The Disaster Research Center Historical and

-------
Comparative Disaster Series, No. 5.  Columbus, OH: Disaster Research Center,
The Ohio State University, 1979.                               ......

Foster, H. Disaster Planning: The Preservation of Life and Property. ' New
York: Springer-Verlag, 1980.

Fowlkes, M. and P. Miller, Love Canal:  The Social Construction of Disaster.
Washington, DC: FEMA, 1982.

Foxworthy, B. and M. Hill. Volcanic Eruptions of 1980 at Mt. St. Helens, The
First 100 Days, Geological Survey Professional Paper 1249. Washington, DC: U.
S. Government Printing Office, 1982.

French, J., R. Ing, S. Von Allmen and R. Wood. "Mortality from flash floods: A
Review of National Weather Service Reports, 1969-81," Public Health Reports 98
(6) (Nov.-Dec.): 584-588, 1983.

Fritz, C. "Disasters," pp. 651-694 in Merton and Nisbet (eds.), Contemporary
Social Problems.  New York: Harcourt, 1961.

Fritz, C.  "Disasters compared in six American communities," Human
Organization 16 (Summer): 6-9, 1961.

Fritz, C. E. and E. S. Marks. "The NORC studies of human behavior in
disaster," Journal of Social Issues 10 (3): 26-41, 1954.

Fritz, C. E. and H. B. Williams. "The human being in disasters:  A  research
perspective," The Annals of the American Academy of Political and Social
Science 309 (January): 42-51, 1957.

Fritz, C. E. and J. H. Mathewson. Convergence Behavior in Disasters, National
Research Council Disaster Study #9.  Washington, DC:  National Academy of
Sciences, 1957.

Graham, W. J. and C. A. Brown. The Lawn Lake Dam Failure:  A Description of
the Major Flooding Events and an Evaluation of the Warning Process.  Denver:
Bureau of Reclamation, 1983.                                           ...

Gray, J. "Characteristic patterns of and variations in community response to
acute chemical emergencies," Journal of Hazardous Materials 4: 357-365, 1981a.

Gray, J.  Three Case Studies of Organized Responses to Chemical Disasters,
Misc. Report # 29.  Columbus, OH:  Disaster Research Center, Ohio State
University, 1981b.

Greene, M., R. Perry and M. Lindell. "The March 1980 eruptions of Mt. St.
Helens:  Citizen perceptions of volcano threat," Disasters 5 (No. 1): 49-66,
1981.                                                                    ,

Gruntfest, E.  "What people did during the Big Thompson Flood," Working Paper
32.  Boulder, CO:   Institute of Behavioral Science, University of Colorado,
1977.                                                                  :•:....

-------
Gruntfest, E., T. Downing and G. White. "Big Thompson flood exposes need for
better flood reaction system to save lives," Civil Engineering (February)
72-73, 1978.

Haas, J. and P. Trainer. "Effectiveness of the tsunami warning system in
selected coastal towns in Alaska."  Rome:  Proceedings of the Fifth World
Conference on Earthquake Engineering, 1974.

Hannigan, J. A. and R. A. Kueneman. "Anticipating flood emergencies: A case-<
study of a Canadian disaster subculture," pp. 129-146 in Disasters: Theory and
Research, E. L. Quarantelli (ed.).  Beverly Hills, CA: Sage, 1978.

Hannson, R, D. Noulles, and S. Bellovich. "Knowledge, warning and stress,"
Environment and Behavior 14 (2) 171-185, 1982.       ,

Hart, G., L. Escalante, W. Petak, C. Pinkham, E. Swartz and M. Wurtele. The
Los Angeles, California, Tornado of March 1, 1983. Washington, DC: National
Academy Press, 1985.

Hodge, D., V. Sharp and Marion Marts. "Contemporary responses to volcanism:
Case studies from the Cascades and Hawaii," pp. 221-248 in Volcanic Activity
and Human Ecology, P. D. Sheets and D. K. Grayson (eds.).  New York:  Academic
Press, 1979.

Houts, P. et al. "The protective action decision model applied to evacuation
during the TMI crisis," Mass Emergencies 2(1): 27-39; 1984.

Button, J. R. "The differential distribution of death in disaster: A test of
theoretical propositions," Mass Emergencies 1 (October): 261-266, 1976.

Janis, I. Psychological Stress. John Wiley and Sons, New York, 1958.

Janis, I. L. and L. Mann. "Emergency decision makingf  A theoretical analysis
of responses to disaster warnings," Journal of Human;Stress 3 (June): 35-48,
1977.                                                i

Johnson, J. "Planning for spontaneous evacuation during a radiological
emergency," Nuclear Safety 25: 186-193, 1983.

Johnson, J. and D. Ziegler.), "A spatial analysis of evacuation intentions at
the Shoreham Nuclear Power Station," pp. 279-301 in M. Pasqualetti and K.
Pijawka  (eds.), Nuclear Power:  Assessing and Managing Hazardous Technology.
Boulder: Westview Press, 1984a.
                                                     i
Johnson, J. and R. Ziegler. "Distinguishing human responses to radiological
emergencies," Economic Geography 59: 386- 402, 1984b.

Keast, D. N. et al. Procedures for Analyzing the Effectiveness of Siren
Systems  for Alerting the Public. Repoort No. NUREG/CR-4227.  Richland, WA:
Battelle, 1982.                                      :

Keating, J., et al. "Emergency evacuations during fires:  Psychological
Considerations," in R. F. Kidd and M. J. Saks (eds.) Advances in Applied

-------
Social Psychology (Vol. 2).  Lawrence Erlbaum Associates, Hillsdale, New
Jersey, 83-100, 1983.

Keating, J. And E. Loftus. "Vocal alarm systems for high rise buildings; A
case study," Mass Emergencies 2: 25-34, 1977.

Killian, L. M. A Study of Response to the Houston, Texas Fireworks Explosion.
National Academy of Sciences/National Research Council Disaster Study 2.
Washington, DC:  National Academy of Sciences, 1956.

Killian, L. Evacuation of Panama City for Hurricane Florence. Washington, DC:
National Academy of Science (cited in Quarantelli, 1980), 1954.

Killian, L. "The significance of multi-group membership in disasters,"
American Journal of Sociology (January): 309-314, 1952.

Kutak, R. I. "The sociology of crises:  The Louisville flood of 1937," Social
Forces 17: 66-72, 1938.

Lachman, R. and W. Bonk. "Behavior and beliefs during the recent volcanic
eruption at Kapoho," Science 133: 1405-09, 1960.

Lachman, R., M. Tatsuoka and W. Bonk. "Human behavior during the tsunami of
May, 1960," Science 133 (May 5): 1405-1409, 1961.

Leik, R. K., T. M. Carter, J. P. Clark et al. Community Response to Natural
Hazard Warnings:  Final Report.  Minneapolis, MN:  University of Minnesota,
1981.

Leik, R. K. , S. A. Leik, K. Ekker and G. A. Gifford. Under the Threat of Mount
St. Helens:  A Study of Chronic Family Stress.  Minneapolis, MN:  Family Study
Center, University of Minnesota, 1982.

Lindell,  M. and R. Perry. "Nuclear power plant emergency warnings: How would
the public respond?," Nuclear News (February): 49-57, 1983.

Lindell, M., P. Bolton, R. Perry, G. Stoetzel, J. Martin and C. Flynn.
Planning Concepts and Decision Criteria for Sheltering and Evacuation in a
Nuclear Power Plant Emergency.  Bethesda, MD: Atomic Industrial Forum, 1985.

Liverman, D. and J. Wilson. "The Mississauga  train derailment and evacuation,
10-16 November, 1979," The Canadian Geographer 25: 365-375, 1981.

Louisiana Department of Public Safety. Southeast  Louisiana Hurricane Study,
Evacuation Behavioral Survey, Final Report.   Baton Rouge: Office of Emergency
Preparedness,  1984.

McLuckie, B. F,  "The warning system in disaster situations: A selective
analysis,"  The Ohio  State University: Disaster Research Center, 1970.

Mack, R. W.  and G. W. Baker. The Occasion  Instant, National Academy of
Sciences/National Research Council Disaster  Study #15.  Washington, DC:
National, Academy  of  Sciences, 1961.

-------
Mileti, D. "Role conflict and abandonment in emergency workers," Emergency
Management Review 2: 20-22, 1985.                                            ,

Mileti, D. Natural Hazard Warning Systems in the United States. Boulder, CQ: .;
Institute of Behavioral Science, University of Colorado, 1975.

Mileti, D. S. and E. M. Beck. "Communication in crisis:  Explaining evacuation
symbolically," Communication Research 2 (January): 24-49, 1975.

Mileti, D. and J. Sorensen. "Planning and Implementing Warning Systems," in M.
Lystad (ed.), Mental Health Care In Mass Emergencies: Theory and Practice. New
York: Brunner/Mazel Psychological Stress Series (In Press).

Mileti, D. S. and J. Sorensen. "Why people take precautions against natural
disasters" in N. Weinstein (ed.) Taking Care: Why People Take Precautions.
Cambridge: Cambridge University Press, 1987.

Mileti, D., T. Drabek, and J. Haas. Human Systems in: Extreme Environments.
Boulder, CO:  Institute of Behavioral Science, The University of Colorado,
1975.                                                                     .

Mileti, D. S., J. R. Button and J. H. Sorensen. Earthquake Prediction Response
and Options for Public Policy.  Boulder, CO:  Institute of• Behavioral Science,
The University of Colorado, 1981.                                  ..-••.

Mileti, D. S., J. Sorensen, and W. Bogard. Evacuation Decision-Making:
Process and Uncertainty, ORNL TM-9692. Oak Ridge,TN: Oak Ridge National
Laboratory, 1985.

Moore, H. E.  Tornadoes Over Texas. Austin, TX: University of Texas Press,
1958.

Moore, H. E. with the collaboration of F.  L. Bates, J. P. Alston, M. M.
Fuller, M. V. Layman, D. L. Mischer and M. M. White. ; And the Winds Blew.  :
Austin, TX:  The Hogg Foundation for Mental Health, University of Texas, 1964.

Moore, H. E., F. L. Bates, M. V. Layman and V. J. Parenton. Before the Wind:
A Study of Response to Hurricane Carla, National Academy of Sciences/National
Research Council Disaster Study #19.  Washington, DC: National Academy of.
Sciences,  1963.

Moore, W. , E. Cook, R. Gooch and C. Nordin. The Austin, Texas, Flood of May
24-25, 1981. Washington, DC: National Academy Press. National Oceanic and
Atmospheric Agency (1981), The Disastrous Southern California and Central
Arizona Floods, Flash Floods and Mudslides of February 1980. Silver Spring,
MD: National Weather Service, 1982.
                                                    i     ,             •
Nehnevajsa, J. "Western Pennsylvania: Some issues in warning the population .-..
under emergency conditions." Pittsburgh, PA:  Center for Social and Urban
Research, University of Pittsburgh, 1985.

Nigg, J. M. "Ethnicity and response to warnings," Paper presented at fche
National Hurricane Conference, New Orleans, LA, May 1, 1985. 1985.

-------
Nigg, J. M. "Awareness and behavior:  Public response to prediction
awareness," pp. 71-96, Perspectives on Increasing Hazard Awareness,  Thomas F.
Saarinen (ed.).  Boulder, GO:  Institute of Behavioral Science, University of
Colorado, 1982.

Nylens, L. and 0. Hultaker. "Communications in disaster situations," The
Police Chief (June) 28-34, 1987.

Panel on the Public Policy Implications of Earthquake Prediction. Earthquake
Prediction and Public Policy.  Washington, DC:  National Academy of Sciencesj
1975.

Perry, R. W. Comprehensive Emergency Management: Evacuating Threatened
Populations. Greenwich,  CT: JAI Press, Inc, 1985.

Perry, Ronald W. "Population evacuation in volcanic eruptions, floods and
nuclear power plant accidents:   Some elementary comparisons."  Journal of
Community Psychology 11 (January): 36-47, 1983.

Perry, R. W. "Evacuation decision-making in natural disasters," Mass
Emergencies 4 (March): 25-38, 1979a.

Perry, R. W. "Incentives for evacuation in natural disaster-Research based
community emergency planning,"  Journal of the American Planning Association 45
(4): 440-447, 1979b.

Perry, R. W. and M. R. Greene.  Citizen Response to Volcanic Eruptions: The
Case of Mount St. Helens.  New York:  Irvington Publishers, Inc, 1983.

Perry, R. W. and M. R. Greene.  "Emergency management in volcano hazards:  The
May 18, 1980 eruption of Mt. St. Helens," The Environmental Professional 4  ,>
(Fall):  340-350, 1982a.

Perry, R. W. and M. Greene. "The Role of ethnicity in the emergency decision-
making process, " Sociological Inquiry 52 (Fall): 309-334, 1982b.

Perry, R. W. and M. Lindell. "Predisaster planning to promote compliance with
evacuation warnings," pp. 44-49 in E. J. Baker (ed.) Hurricanes and Coastal
Storms, Florida Sea Grant College Report 33, 1980.

Perry, R. W. and A. Mushkatel.  Minority Citizens in Disaster. Athens, GA:
University of Georgia Press, 1986.        .

Perry, R. W. and A. Mushkatel.  Disaster Management: Warning Response and
Community Relocation. Westport, CT: Quorum Books, 1984.

Perry, R. W. and J. M. Nigg. "Emergency management strategies for
communicating hazard information," Public Administration Review 45:72-77,
1985.                                                               ,

Perry, R. W. , M. R. Greene and M. K. Lindell. "Enhancing evacuation warning
compliance:  Suggestions for emergency planning," Disasters 4 (4): 433-449,
1980.        •'.-..               ,                                    :

-------
Perry, R. W., M. K. Lindell and M. R. Greene. "Crisis communications: Ethnic
differentials in interpreting and acting on disaster warnings," Social
Behavior and Personality 10 (1): 97-104, 1982a.

Perry, R. W. , M. Lindell and M. R. Greene. "Threat perception and public
response to volcano hazard," The Journal of Social Psychology 116: 199-204,
1982b.
                                                    i
Perry, R. W. , M. K. Lindell and M. R. Greene. Evacuation Planning in Emergency
Management. Lexington, MA: Lexington Books, 1981

Pinellas County Department of Civil Emergency Services. Hurricane Elena
Critique. Pinellas County, FL. ecology and Geography,  1986.

Quarantelli, E. L. "Research findings in organizational behavior in disasters
and their applicability in developing countries," Preliminary Paper #107.
Newark, DE: Disaster Research Center, University of Delaware, 1986.

Quarantelli, E. L. "Social problems of relocation of disaster-at-risk
communities: Some questions and some problems," Preliminary Paper #100.
Newark, DE: Disaster Research Center, University of Delaware, 1985a.

Quarantelli, E. L. "Social support systems: Some behavioral patterns in the
context of mass evacuation activities," pp. 122-136 in B. Sowder (ed.),
Disasters and Mental Health: Selected Contemporary Perspectives. Rockville,
MD: National Institute of Mental Health, U.S. Department of Health and Human
Services, 1985b.

Quarantelli, E. L. "Chemical disaster preparedness at the local community
level," Journal of Hazardous Materials 8: 239-249, 1984a.

Quarantelli, E. L. "Perceptions and reactions to emergency warnings of sudden
hazards," Ekestics 309 (Nov.-Dec.), 1984b.

Quarantelli, E. L. Evacuation Behavior: Case Study of the Taft, Louisiana
Chemical Tank Explosion Incident.  Columbus, Ohio:  Disaster Research Center,
Ohio State University, 1983.                        !

Quarantelli, E. L. Transportation Accidents Involving Hazardous Chemicals
Versus Those Involving Dangerous Nuclear Materials, Misc. Report #30.
Columbus, OH:  Disaster Research Center, Ohio State University, 1982a.

Quarantelli, E. L. Sociobehavioral Responses to Chemical Hazards: Preparations
for and Responses to Acute Chemical Emergencies at the Local Community Level,
Final Report to the National Science Foundation.  Columbus, OH: Disaster
Research Center, Ohio State University, 1981a.

Quarantelli, E. L. The Reality of Local Community Chemical Disaster
Preparedness:  Three Case Studies, Misc. Report # 28,  Columbus, OH:'Disaster
Research Center, Ohio State University, 1981b.      !

Quarantelli, E. L. Evacuation Behavior and Problems:  Findings and
Implications From the Research Literature.  Columbus,  OH: Disaster Research
Center, Ohio State University, 1980.

-------
Quarantelli, E. L. "Images of withdrawal'behavior in disasters: Some basic
misconceptions," Social Problems 8: 68-79, 1960.

Quarantelli, E. L. "The behavior of panic participants," Sociology and Social
Research 41: 187-194, 1957.

Quarantelli, E. L. "The nature and conditions of panic," American Journal of
Sociology 60: 267-275, 1954.

Quarantelli, E. L. "Structural factors in the minimization of role conflict: A
reexamination of the significance of multiple group membership in disaster,"
(preliminary paper 49). Columbus, OH: Disaster Research Center, Ohio State
University.                                                                 '

Rayner, J. Hurricane Barbara: A Study of the Evacuation of Ocean City,
Maryland, August, 1953. Washington, DC: Committee on Disaster Studies,
National Academy of Science (cited in Quarantelli, 1980), 1953.               '

Rogers, G. Presidentially Directed Relocation Compliance Attitudes,
DCPA-01-77-Ca-0218.  Pittsburgh, PA:  Pittsburgh University, 1980.

Rogers, G. and J. Nehnevajsa. "Warning human populations of technological
hazard," Proceeding of the ANS Topical Meeting on Radiological Accidents
(CONF-860932). Oak Ridge National Laboratory, Oak Ridge, Tennessee, 357-362,
1987.             	

Rogers, G. and J. Nehnevajsa. Behavior and Attitudes Under Crisis Conditions.
Washington, DC: FEMA, 1984.

Ruch, C. and L. Christensen. Hurricane Message Enhancement. College Station,
TX:  Sea Grant College Program, Texas A&M University, 1981.

Saarinen, T. and J. Sell. Warning and Response to the Mt.'St. Helens Eruption.
Albany, NY: SUNY Press, 1985.

Saarinen, T., V. Baker, R. Durreriberger and T. Maddock. The Tucson, Arizona,
Flood of October 1983. Washington, DC: National Academy Press, 1984.

Savage, R., J. Baker, J. Golden, A. Kareem and B. R. Manning. Hurricane
Alicia, Galveston and Houston, Texas, August 17-18, 1983. Washington, DC:
National Academy of Sciences.  Scarilon, T. J, (1978) "Media coverage of
crisis: Better than reported, worse than necessary," Journalism Quarterly 55:
69-72, 1984.                        •              :

Scanloh; T. J. and A. Frizzell. "Old theories don't apply; Implications of
communication in crisis," Disasters'3: 315-319, 1979.

Scanlon, T. J., J. Jefferson and D.-Sproat. The Port Alice Slide,  (EPC Field
Report 76/1). Ottawa, Ontario: Emergency Planning Canada, 1976.

Scanlon, T. J., M. Padgham and W. Star. The Peel Regional Police Force and the
Mississauga Evacuation: How a Police Force Reacted'to a Major Chemical
Emergency. Ottawa, Ontario:• Canadian Police College,1980.

-------
Schaffer, R. and E. Cook. Human Response to Hurricane Celia, 50 pp.  College
Station, TX:  Environmental Quality Program, Texas A&M University (cited in
Quarantelli, 1980) 1972.

Segaloff, L.  Task Sirocco:  Community Reaction to an Accidental Chlorine
Exposure, 42 pp. Philadelphia, PA:  Institute for Copperative Research,
University of Pennsylvania (cited in Quarantelli, 1980), 1961.

Simpson, R. H.  "Will coastal residents reach safe shelter in time," pp.     r
25-27, in E. J. Baker (ed.), Hurricanes and Coastal Storms, Florida Sea Grant
College Report No. 33, 1980.
                                                    I                  , .
Simpson, R. H. and H. Riehl. The Hurricane and Its Impact.  Baton Rouge, LA:
Louisiana State University Press, 1981.

Simpson, R. H. , B. Hayden, M. Garstang, and H. Massie. "Timing of hurricane
emergency actions," Environmental Management 9: 61-70, 1985.

Sims, J. H. and D. D. Baumann. "The tornado threat: Coping styles of the North
and South," Science 176:  1386-92, 1972.
                                                    i
Sorensen, J. "Warning systems in the 1985 Cheyenne flash flood," In E.
Gruntfest (ed.) What We have Learned Since the Big Thompson Flood.  Institute
of Behavioral Science, University of Colorado, Boulder, 1987.

Sorensen, J. H. Evaluating the effectiveness of warning systems for nuclear
power plant emergencies: Criteria and Application,"pp. 259- 277 in M.
Pasqualetti and K. Pijawka (eds.) Nuclear Power: Assessing and Managing
Hazardous Technologies. Boulder: Westview Press, 1984a.

Sorensen, J. H.  "Public confidence in local management officials:
Organizational credibility and emergency behavior," Proceedings of a
Conference on Evacuation Risks in Nuclear Power Plant Emergencies. Penn State
University, 1984b.

Sorensen, J. H. Evaluation of Emergency Warning System at Ft. St. Vrain
Nuclear Power Plant, ORNL/TM-8171. Oak Ridge, TN: Oak Ridge National  .
Laboratory, 1982.                                   '

Sorensen, J. H. "Emergency response to Mt. St. Helens' eruption: March 20 to
April 10, 1980," Working Paper 43.  Boulder, CO:  Boulder Institute of
Behavioral Science, University of Colorado, 1981.   |
                                                    i
Sorensen, J. H. and D. S. Mileti (in press), "Decision-making uncertainties in
emergency warning system organizations," International Journal of Mass
Emergencies and Disasters  (March, 1987).

Sorensen, J and D. Mileti. "Programs to Encourage Precautionary Behavior in
Natural Disasters:  Review and Evaluation," in N. Welinstein (ed.), Taking
Care:  Why People Take Precautipns.  Cambridge:  Cambridge University Press,
1987, pp 208-230.

-------
Sorensen, J. H. and P. J. Gersmehl. "Volcanic hazard warning system:
Persistence and transferability," Environmental Management 4 (March): 125-136,
1980.

Sorensen, J. H.,  D. S. Miletl, and E. D. Copenhaver. "Inter and
intraorganizational cohesion in emergencies," International Journal of Mass
Emergencies and Disasters  3(3): 27-52, 1985.

Sorensen, J. H.,  E. D. Copenhaver, D. S. Mileti and M. V. Adler.
Organizational Interface in Reactor Emergency Planning and Responses.
NUREG/CR-3524, ORNL-6010. Oak Ridge, TN: Oak Ridge National Laboratory, 1984.

Stallings, R. "Evacuation behavior at Three Mile Island," International
Journal of Mass Emergencies and Disasters 2: 11-26, 1984.

Strope, W. E. , J. R. Devaney and J. Nehnevajsa. "Importance of preparatory
measures in disaster evacuation," Mass Emergencies 2: 1-17, 1977.

Tanzos, T .C. et al. Warning Systems Technical Review and Warning System
Evaluation Guidelines. McLean, VA: P.C. Vorhee, 1983.

Taylor, J. B., L. A. Zurcher and W. H. Key. Tornado:  A Community Responds to
Disaster.  Seattle, WA:  University of Washington Press, 1970.

Tierney, K. A Primer for Preparedness for Acute Chemical Emergencies, Mono.
No. 14.  Columbus, OH:  Disaster Research Center, Ohio State University, 1980.

Towers, D.',-et al. Evaluation of Prompt Alerting Systems at Four Nuclear Power
Stations. (NUREG CR-2655).  Pacific Northwest Laboratory, Richland, Washington,
1982.                 . .

Treadwell, N. Hurricane Carla - September 3-14, 1961, Office of Civil Defense
Region 5, Denton, TX:  U.S. Government Printing Office (cited in Quarantelli,
1980), 1962.

Turner, R. H. "Waiting for disaster:  Changing reactions to earthquake
forecasts in Southern California," International Journal of Mass Emergencies
and Disasters 1:   307-334,  1983.

Turner, R. H., J.. M. Nigg,  D. H. Paz and B. S. Young Earthquake Threat:  The
Human Response in Southern California.  Los Angeles, CA: Institute for Social
Science Research, University of California, Los Angeles, 1979.

USDOC. Johnstown Pennsylvania Flash Flood of July 19-20, 1977, Natural
Disaster Survey Report 77-1. Rockville, MD: National Oceanic and Atmospheric
Administration, 1977.

USDOT  (U.S. Department of Transportation). Hazardous Materials Emergency
Response Guidebook. Washington, DC: U.S. Department of Transportation, 1984.

USEPA. Chemical Emergency Preparedness Program Interim Guidance. Washington,
DC: EPA, 1985.

-------
USFEMA. Guide For Evaluation of Alert and Notification Systems for Nuclear
Power Plants;  FEMA Rep-10  Washington, DC:  FEMA, 1985..

USFEMA. A Guide to Hurricane.Preparedness Planning for State and Local
Officials, CPG2-16.  Washington, DC:  FEMA, 1984.                     .

USFEMA. Preparing for Hurricanes and Coastal Flooding, A Handbook for Local
Officials, Publication 50.  Washington, DC:  Federal Emergency Management
Agency, 1983.                                       ,        •               ,
                                                    i
USFEMA. planning Guidance for the preparation of the Federal Radiological
Emergency Response Plan, Final Draft #4.  Washington, DC:  Government Printing
Office, 1982.               ,  ,                                        .   ,,

USFEMA. U.S. Crisis Relocation Planning, P-P-7.  Washington, DC:  FEMA,  1981.

Vogt, B. M. and J. H. Sorensen. Evacuation in Emergencies: An Annotated Guide
to Research, ORNL/TM-10277. Oak Ridge, TN: Oak Ridge Natipnal Laboratory,1986,

Wallace, A. F. Tornado in Worcester, National Academy of Sciences/National
Research Council Disaster Study #3.  Washington, DC:  National Academy of
Sciences, 1956.                                     i
                                                    i
Weller, J. "Response to tsunami warnings:  The March, 1964 Prince Williams
Sound earthquake," Working Paper #15.  Columbus, OH:  Disaster Research
Center, Ohio State University (cited in Quarantelli; 1980), 1967.

Wendell, M. "Legal aspects of flood warning and evacuation," pp. 28-31,  in E.
J. Baker (ed.) Hurricanes and Coastal Storms, Florida Sea Grant College Report
No. 33, 1980.                                       i
                                                    !
Wenger, D. E. "Mass media and disasters," Preliminary Paper #98. Newark, DE:
Disaster Research Center, University of Delaware, 1985a.

Wenger, D. E. "Collective behavior and disaster research," Preliminary Paper
#104. Newark, DE: Disaster Research Center, University of Delaware, 1985b.

Wenger, D. E. "Local and national media coverage of;disaster: A content
analysis of the print media's treatment of disaster'myths," Preliminary Paper
#96. Newark, DE: Disaster Research Center, University of Delaware, 1986.

Wilkenson, K. and P. Ross. Citizens Response to Warnings of Hurricane Camille.
Report No. 35.  College Station, MS: Mississippi State University, Social
Science Research Center, 1970.

Williams, H. B. "Human factors in the warning and response systems," pp.
79-104 in G. Grosser et al. (ed.) The Threat of Impending Disaster. Cambridge
MA: MIT Press, 1964.                                ; ' "

Windham, G., E. Posey, P. Ross, and B. Spencer. Reaction to Storm Threat
During Hurricane Eloise.  Report No. 51.  State College, MS:  Mississippi
State University, Social Science Research Center, 1977.

-------
Withey, S. B. "Reaction to uncertain threat,"  pp.  93-123 in G.  Baker and D.
Chapman (eds.) Man and Society in Disaster.  New York:  Basis Books,  1962.

Worth, M. F. and B. F. McLuckie.  Get to High Ground!   The Warning Process in
the Colorado Floods June 1965,  Disaster Research Center Historical and
Comparative Disasters Series.   Columbus,  OH:   Disaster Research Center, The
Ohio State University, 1977.

Young, M. "Role of extended  family in a disaster," Human Relations 7: 383-391,
1954.

Yutzy, D."Aesop 1964:  Contingencies affecting the issuing of public disaster
warnings at Crescent City, California,"  Research Note #4. Columbus, OH:
Disaster Research Center, Ohio State University (cited in Quarantelli, 1980),
1964.

Ziegler, D. and J. Johnson.  "Evacuation behavior in response to nuclear power
plant accidents," Professional Geographer 36:207-215,  1984.
                                                *U.S. GOVERNMENT PRINTING OFFICE: 1989-617-003/84317

-------

-------