United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park, NC 27711
EPA-454/R-93-002
(Revises EPA-450/4-91-007)
May 1993
Air
EPA
GUIDANCE ON THE APPLICATION
OF REFINED DISPERSION
MODELS FOR HAZARDOUS/TOXIC
AIR RELEASES
-------
EPA-454/R-93-002
Guidance on the Application of
Refined Dispersion Models for
Hazardous/Toxic Air Releases
U S Environmental Protection Agency
--
Chicago,
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Technical Support Division
Research Triangle Park, NC 27711
May 1993
-------
This report has been reviewed by the Office of Air Quality
Planning and Standards, U.S. Environmental Protection Agency, and
has been approved for publication. Any mention of trade names
and commercial products is not intended to constitute endorsement
or recommendation for use.
EPA-454/R-93-002
-------
ACKNOWLEDGEMENTS
This report was prepared by Radian Corporation for the United States
Environmental Protection Agency under contract number 68-DO-0125 with Mr.
Jawad S. Touma as the Work Assignment Manager. He is on assignment to the
U.S. EPA from the National Oceanic and Atmospheric Administration,
U.S. Department of Commerce. Providing assistance were Messrs. Jim Yohn and
Doug Blewitt, Amoco Corporation (SLAB and HGSYSTEM model application and
interpretation), Drs. Jerry Gait and Roy Overstreet, National Oceanic and
Atmospheric Administration (ALOHA model application and interpretation), Dr.
David Yeh, South Coast Air Quality Management District (general review), and
Dr David Guinnup, EPA (general review).
iii
-------
PREFACE
This document supersedes "Guidance on the Application of Refined
Dispersion Models for Air Toxic Releases," EPA-450/4-91-007.
-------
TABLE OF CONTENTS
ACKNOWLEDGEMENTS '
PREFACE v
1.0 INTRODUCTION 1-1
2.0 SCENARIO AND CHEMICAL SELECTION 2-1
2.1 Selection of Chemicals for Release Modeling 2-1
2.2 Description of Chemical Release Classes 2-2
2.2.1 Two-phase Gas Release (Choked) 2-3
2.2.2 Two-phase Gas Release (Unchoked) 2-4
2.2.3 Two-phase Pressurized Liquid Release 2-5
2.2.4 Two-phase Refrigerated Liquid Release 2-6
2.2.5 Single-phase Gas Release (Choked) 2-6
2.2.6 Single-phase Gas Release (Unchoked) 2-7
2.2.7 Single-phase Liquid Release
(High Volatility) 2-8
2.2.8 Single-phase Liquid Release
(Low Volatility) 2-9
2.2.9 Additional Release Class Considerations 2-9
2.3 Selection of Release Scenarios for Modeling 2-14
2.4 Verification of Release Scenarios in the Review of
Modeling Submittals 2-20
3.0 MODELS 3-1
3.1 ADAM 3-1
3.2 ALOHA . . : 3-2
3.3 DEGADIS 3-3
3.4 HGSYSTEM 3-4
3.5 SLAB 3-5
4.0 MODEL INPUT 4-1
4.1 Observable Data 4-2
• 4.1.1 Vessel Shape and Dimensions 4-4
4.1.2 Hole Location and Orientation 4-4
4.1.3 Diking 4-5
4.1.4 Surface Description 4-5
4.1.5 Solar Radiation '. 4-5
4.2 Chemical Data Requirements 4-6
4.3 Release Class 4-9
4.4 Continuous or Instantaneous Release Categories 4-13
4.4.1 Non-source-term Model 4-14
4.4.2 Source-term Model 4-15
4.5 Release-class-specific Calculations 4-16
4.5.1 Two-phase Releases (Gas and Liquid) . • 4-22
4.5.2 Single-phase Gas 4-22
4.5.3 Single-phase, High Volatility Liquid 4-23
vii
-------
TABLE OF CONTENTS (Continued)
4.5.4 Single-phase, Low Volatility Liquid 4-23
4.6 Determination for Choked Flow for Gas Releases 4-23
4.7 Emission Rate 4-24
4.7.1 Two-phase Gas Release (Choked) 4-25
4.7.2 Two-phase Gas Release (Unchoked) 4-26
4.7.3 Two-phase Pressurized Liquid 4-27
4.7.4 Two-phase Refrigerated Liquid 4-29
4.7.5 Single-phase Gas Release (Choked) 4-30
4.7.6 Single-phase Gas Release (Unchoked) 4-30
4.7.7 Single-phase Liquid Release
(High Volatility) 4-31
4.7.8 Single-phase Liquid Release
(Low Volatility) 4-33
4.8 Release Temperature 4-34
4.8.1 Two-phase Gas Release (Choked) 4-35
4.8.2 Two-phase Gas Release (Unchoked) 4-37
4.8.3 Two-phase Pressurized Liquid 4-37
4.8.4 Two-phase Refrigerated Liquid 4-37
4.8.5 Single-phase Gas Release (Choked) 4-38
4.8.6 Single-phase Gas Release (Unchoked) 4-38
4.8.7 Single-phase Liquid Release
(High Volatility) 4-39
4.8.8 Single-phase Liquid Release
(Low Volatility) 4-39
4.9 Vapor Fraction 4-40
4.9.1 Two-phase Gas Release (Choked) 4-41
4.9.2 Two-phase Gas Release (Unchoked) 4-41
4.9.3 Two-phase Pressurized Liquid 4-42
4.9.4 Two-phase Refrigerated Liquid 4-43
4.10 Initial Concentration 4-43
4.10.1 Single-phase Gas Release 4-45
4.10.2 Single-phase Liquid Release
(Low Volatility) 4-46
4.11 Density 4-47
4.12 Release Diameter or Area 4-50
4.12.1 Low-momentum Release 4-52
4.12.2 High-momentum Release 4-52
4.13 Release Buoyancy 4-53
4.13.1 Continuous Release 4-53
4.13.2 Instantaneous Release .... 4-53
4.14 Release Height 4-54
4.15 Ground Surface Temperature 4-55
4.16 Averaging Time 4-55
4.17 Meteorology 4-58
4.17.1 Wind Speed and Direction 4-58
4.17.2 Stability Class 4-59
4.17.3 Surface Roughness Length 4-60
viii
-------
TABLE OF CONTENTS (Continued)
4.17.4 Wind Speed at 10 m Altitude 4-62
4.17.5 Ambient Temperature, Relative Humidity,
and Pressure 4-64
4.18 Output Definition 4-64
5.0 MODEL INPUT DEVELOPMENT FOR RELEASE CLASS 5-1
5.1 Two-phase Gas Release (Choked) Example 5-2
5.1.1 Observable Data 5-4
5.1.2 Chemical Data Requirements . . . . - 5-4
5.1.3 Release Class 5-4
5.1.4 Continuous or Instantaneous Release
Categories 5-8
5.1.5 * Release-class-specific Calculations 5-9
5.1.6 Determination of Choked Flow for Gas
Releases 5-9
5.1.7 Emission Rate 5-10
5.1.8 Release Temperature 5-12
5.1.9 Vapor Fraction 5-12
5.1.10 Initial Concentration 5-13
5.1.11 Density 5-13
5.1.12 Release Diameter or Area 5-16
5.1.13 Release Buoyancy 5-17
5.1.14 Release Height 5-18
5.1.15 Ground Surface Temperature 5-18
5.1.16 Averaging Time 5-18
5.1.17 Meteorology 5-19
5.1.18 Output Definition 5-20
5.2 Two-phase Gas Release (Unchoked) Example 5-20
5.2.1 Observable Data 5-21
5.2.2 Chemical Data Requirements 5-21
5.2.3 Release Class 5-21
5.2.4 Continuous or Instantaneous Release
Categories 5-28
5.2.5 Release-class-specific Calculations 5-28
5.2.6 Determination of Choked Flow for Gas
Releases 5-28
5.2.7 Emission Rate 5-29
5.2.8 Release Temperature 5-30
5.2.9 Vapor Fraction 5-30
5.2.10 Initial Concentration . 5-31
5.2.11 Density 5-31
5.2.12 Release Diameter or Area 5-34
5.2.13 Release Buoyancy 5-35
5.2.14 Release Height 5-36
5.2.15 Ground Surface Temperature 5-36
5.2.16 Averaging Time 5-36
5.2.17 Meteorology 5-37
ix
-------
TABLE OF CONTENTS (Continued)
5.2.18
5.3
5.4
5.5
— ' • fc * i w ,
Two -phase
5.3.1
5.3.2
5.3.3
5.3.4
5.3.5
5.3.6
5.3.7
5.3.8
5.3.9
5.3.10
5.3.11
5 . 3 . 12
5.3.13
5.3.14
5 . 3 . 15
5.3.16
5.3.17
5.3.18
Two -phase
5.4.1
5.4.2
5.4.3
5.4.4
5.4.5
5.4.6
5.4.7
5.4.8
5.4.9
5.4.10
5.4.11
5.4.12
5.4.13
5.4.14
5.4.15
5.4.16 '
5.4.17
5.4.18
Pressurized Liquid Example , ,
Observable Data
Chemical Data Requirements
Release Class
Continuous or Instantaneous Release
Categories
Release-class-specific Calculations
Determination of Choked Flow for Gas
Releases
Emission Rate.
Release Temperature
Vapor Fraction
Initial Concentration
Density
Release Diameter or Area
Release Buoyancy
Release Height
Ground Surface Temperature
Averaging Time
Meteorology
Output Definition
Refrigerated Liquid F/*an»ple
Observable Data
Chemical Data Requirements
Release Class
Continuous or Instantaneous Release
Categories
Release-class-specific Calculations
Determination of Choked Flow for Gas
Releases
Emission Rate.
Release Temperature
Vapor Fraction
Initial Concentration
Density
Release Diameter or Area
Release Buoyancy . .
Release Height
Ground Surface Temperature
Averaging Time
Meteorology
Output Definition
Single-phase GJ»? Release (footed) Example
5.5.1
5.5.2
Observable Data
Chemical Data Requirements
. . 5-38
. . 5-43
. . 5-43
. . 5-43
. .5-46
. . 5-47
. . 5-47
. . 5-47
. . 5-48
. . 5-48
. . 5-48
. . 5-49
. . 5-51
. . 5-52
. . 5-53
. . 5-53
. . 5-53
. . 5-54
. . 5-55
. . 5-55
. . 5-56
. . 5-56
. . 5-56
. . 5-62
. . 5-63
. . 5-63
. . 5-63
. . 5-64
. . 5-64
. . 5-65
. . 5-65
. . 5-68
. . 5-68
. . 5-69
. . 5-69
. . 5-70
. . 5-70
. . 5-71
. . 5-71
. . 5-72
. . 5-77
5.5.3 Release Class
-------
TABLE OF CONTENTS (Continued)
5.5.4 Continuous or Instantaneous Release
Categories 5-78
5.5.5 Release-class-specific Calculations 5-78
5.5.6 Determination of Choked Flow for Gas
Releases 5-78
5.5.7 Emission Rate 5-79
5.5.8 Release Temperature 5-80
5.5.9 Vapor Fraction 5-80
5.5.10 Initial Concentration 5-80
5.5.11 Density 5-81
5.5.12 Release Diameter or Area 5-83
5.5.13 Release Buoyancy 5-85
5.5.14 Release Height 5-87
5.5.15 Ground Surface Temperature 5-87
5.5.16 Averaging Time 5-87
5.5.17 Meteorology 5-87
5.5.18 Output Definition 5-88
5.6 Single-phase Gas Release (Unchoked) Example 5-89
5.6.1 Observable Data . . .• 5-91
5.6.2 Chemical Data Requirements 5-91
5.6.3 Release Class 5-96
5.6.4 Continuous or Instantaneous Release
Categories 5-96
5.6.5 Release-class-specific Calculations 5-97
5.6.6 • Determination of Choked Flow for Gas
Releases 5-97
5.6.7 Emission Rate 5-98
5.6.8 Release Temperature 5-98
5.6.9 Vapor Fraction 5-98
5.6.10 Initial Concentration 5-98
5.6.11 Density 5-99
5.6.12 Release Diameter or Area 5-102
5.6.13 Release Buoyancy 5-103
5.6.14 Release Height . 5-103
5.6.15 Ground Surface Temperature 5-104
5.6.16 Averaging Time 5-104
5.6.17 Meteorology 5-104
5.6.18 Output Definition 5-105
5.7 Single-phase Liquid Release (High Volatility)
Example 5-105
5.7.1 Observable Data 5-106
5.7.2 Chemical Data Requirements 5-106
5.7.3 Release Class 5-111
5.7.4 Continuous or Instantaneous Release
Categories 5-111
5.7.5 Release-class-specific Calculations 5-112
XI
-------
TABLE OF CONTENTS (Continued)
5.7.6 Determination of Choked Flow for Gas
Releases 5-112
5.7.7 Emission Rate 5-113
5.7.8 Release Temperature 5-113
5.7.9 Vapor Fraction 5-114
5.7.10 Initial Concentration 5-114
5.7.11 Density 5-114
5.7.12 Release Diameter or Area 5-116
5.7.13 Release Buoyancy 5-117
5.7.14 Release Height 5-118
5.7.15 Ground Surface Temperature 5-118
5.7.16 Averaging Time 5-118
5.7.17 Meteorology 5-119
5.7.18 Output Definition 5-120
5.8 Single-phase Liquid Release (Low Volatility)
Example 5-120
5.8.1 Observable Data 5-122
5.8.2 Chemical Data Requirements 5-122
5.8.3 Release Class 5-122
5.8.4 Continuous or Instantaneous Release
Categories 5-126
5.8.5 Release-class-specific Calculations 5-127
5.8.6 Determination of Choked Flow for Gas
Releases 5-127
• 5.8.7 Emission Rate " * 5-127
5.8.8 Release Temperature 5-131
5.8.9 Vapor Fraction 5-131
5.8.10 Initial Concentration 5-131
5.8.11 Density 5-134
5.8.12 Release Diameter or Area 5-137
5.8.13 Release Buoyancy 5-137
5.8.14 Release Height 5-138
5.8.15 Ground Surface Temperature 5-138
5.8.16 Averaging.Time 5-138
5.8.17 Meteorology 5-139
5.8.18 Output Definition 5-140
6.0 MODEL INPUT USED 6-1
6.1 ADAM 6-2
6.2 ALOHA 6-23
6.3 DEGADIS • 6-33
6.4 HGSYSTEM 6-38
6.5 SLAB 6-66
7.0 MODEL OUTPUT "7-1
7.1 ADAM 7-2
7.2 ALOHA 7-6
xii
-------
TABLE OF CONTENTS (Continued)
7.3 DEGADIS .
7.4 HGSYSTEM
7.5 SLAB . .
8.0 DETERMINING INPUT FOR MODELING "WORST CASE" IMPACTS 8-1
8.1 Model Input 8-1
8.1.1 Exit Velocity, Emission Rate, and Jets 8-3
8.1.2 Release Temperature 8-4
8.1.3 Release Diameter 8-4
8.1.4 Release Height - 8-4
8.1.5 Ground Temperature 8-4
8.1.6 Meteorology 8-5
8.2 Simulation Mechanics 8-6
8.3 Using this Document for Off-site Consequence Analysis . . . .8-14
9.0 REFERENCES 9-1
APPENDIX A: Relationships Between Selected SI Units A-l
APPENDIX B: Triple Point Diagrams, Flash Diagrams, and
Physical Property Tables B-l
Xlll
-------
LIST OF FIGURES
2-1 Two-phase Gas Release 2-4
2-2 Two-phase Liquid Release 2-5
2-3 Single-phase Gas Release 2-7
2-4 Single-phase High Volatility Liquid Release 2-8
2-5 Single-phase Low Volatility Liquid Release 2-9
2-6 Example Release Scenario 2-10
2-7 Triple Point Diagram for Anhydrous Ammonia 2-24
2-8 Method for Determination of Pre-release Conditions
Where One Condition is Known 2-25
2-9 Method for Determination of Pre-release Conditions
Where Two Conditions are Known 2-26
2-10 Flash Diagram for Chlorine 2-28
4-1 Determining Initial Release Class 4-11
4-2 Two-phase Input 4-17
4-3 Gas to Gas Release Input 4-18
4-4 High-volatility Release Input (Not Two-phase) 4-19
4-5 Low-volatility Release Input 4-20
4-6 Check Input and Complete 4-21
5-1 Two-phase Gas Release 5.3
5-2 Two-phase Pressurized Liquid 5-39
5-3 Two-phase Refrigerated Liquid 5-57
5-4 Single-phase Gas Release (Choked) 5-73
5-5 Single-phase Gas Release (Unchoked) 5-90
5-6 Single-phase Liquid Release (High Volatility)- 5-107
5-7 Single-phase Liquid Release (Low Volatility) 5-121
•
xv
-------
LIST OF FIGURES (Continued)
Page
7-1 Typical ADAM Calculation Results 7-3
7-2 Footprint Window 7-7
7-3 Concentration versus Time Plot 7-8
7-4 Dose versus Time Plot 7-9
8-1 Dispersion Zone Schematic 8-2
xvi
-------
LIST OF TABLES
Page
2-1 Possible Release Classes Resulting From
Typical Storage Conditions 2-12
2-2 Storage Release Scenarios and Post-Release
Characteristics of Selected Chemicals 2-15
2-3 Descriptive Scenarios for Modeling 2-17
2-4 Typical Storage Quantities for Selected Chemicals
{Anhydrous) ' . . . . 2-19
4-1 Representative Values of Surface Roughness for a
Uniform Distribution of Selected Types of Ground Cover 4-61
4-2 Values for the Exponent in the Wind Profile Curve as a
Function of Stability Class and Surface Roughness 4-63
5-1 Input Summary for the Two-phase Choked-Gas
Release Example 5-5
5-2 Observable Data Summary for the Two-phase Choked Gas
Release Example 5-7
5-3 Input Summary for the Two-phase Unchoked Gas Release
Example 5-22
5-4 Observable Data Summary for the Two-phase Unchoked
Gas Release Example 5-24
5-5 Input Summary for the Two-phase Pressurized Liquid
Release Summary 5-41
5-6 Observable Data Summary for the Two-phase Pressurized
Liquid Release Example 5-44
5-7 Input Summary for the Two-phase Refrigerated Liquid
Release Example 5-58
5-8 Observable Data Summary for Che. Two-phase Refrigerated
Liquid Release Example 5-60
5-9 Input Summary for the Single-phase Choked Gas
Release Example 5-74
5-10 Observable Data Summary for the Single-phase Choked
Gas Release Example 5-76
xvii
-------
LIST OF TABLES (Continued)
Page
5-11 Input Summary for the Single-phase Unchoked
Gas Release Example 5-92
5-12 Observable Data Summary for the Single-phase Unchoked
Gas Release 5-94
5-13 Input Summary for the High-volatility, Single-phase
Liquid Release Example 5-108
5-14 Observable Data summary for the High-volatility,
Single-phase Liquid Release Example 5-110
5-15 Input Summary for the Low-Volatility Liquid
Release Example . 5-123
5-16 Observable Data Summary for the Low-volatility Liquid
Release Example 5-125
6-1 Input Parameters for the Two-phase Ethylene Oxide Gas
Release (Choked) 6-3
6-2 Input Parameters for the Two-phase Ethylene Oxide Gas
Release (Unchoked) 6-4
6-3 Input Parameters for the Two-phase Pressurized
Liquid Chlorine Release 6-5
6-4 Input Parameters for the Two-phase Pressurized
Sulfur Dioxide Release 6-6
6-5 Input Parameters for the Single-phase Anhydrous Hydrogen
Fluoride Gas Release (Choked) 6-7
6-6 Input Parameters for the Single-phase Liquified
Ethylene Oxide Release 6-8
6-7 Text Summary for Two-phase Gas Release (Choked) 6-24
6-8 Text Summary for Two-phase Gas Release (Unchoked) 6-25
6-9 Text Summary for Two-phase Pressurized Liquid 6-26
6-10 Text Summary for Two-phase Refrigerated Liquid 6-27
6-11 Text Summary for Single-phase Gas Release -(Choked) . . f 6-28
6-12 Text Summary for Single-phase Gas Release (Unchoked) 6-29
xviii
-------
LIST OF TABLES (Continued)
Page
6-13 Text Summary for Single-phase Liquid 'Release
(High Volatility) 6-30
6-14 Text Summary for Single-phase Liquid Release
(Low Volatility) 6-31
6-15 DEGADIS Model Input Parameters 6-34
6-16 DEGADIS Input 6-37
6-17 Summary of the HGSYSTEM Models Used 6-39
6-18 Input Parameters Used in the HEGADASS Model to Simulate
Release Class 1 6-43
6-19 Input Parameters Used in the HEGADASS Model to Simulate
Release Class 2 6-44
6-20 Concatenated Input File for Release Class 3 6-46
6-21 PLUME Model Input for Release Class 4 6-47
6-22 PGPLUME Link File Created by the PLUME Model for
Release Class 4 6-48
6-23 PGPLUME Input for Release Class 4 6-49
6-24 HEGADAST Input Parameters for Release Class 5 6-52
6-25 PLUME Model Input for Release Class 6 6-56
6-26 PGPLUME Link File Created by the PLUME Model for
Release Class 6 6-58
6-27 PGPlume Link File Created by the PLUME Model for
Release Class 6 6-59
6-28 PLUME Input for Release 7 6-60
6-29 HEGADASS Link File Generated by PLUME for
Release Class 7 6-61
6-30 HEGADASS Input File for Release Class 7 6-63
6-31 HEGADASS Input File for Release Class 8 6-65
6-32 SLAB Model Input Parameters 6-67
xix
-------
LIST OF TABLES (Continued)
Page
6-33 SLAB Model Input Used for the Eight Release Classes 6-69
7-1 Example Data Output File from the ADAM Model 7-5
7-2 Example Output from the Jet Sub-model of DEGADIS
JETRLU/DEGADIS v2.1 7-12
7-3 Example Output from the Ground Level Dispersion Sub-model
of DEGADIS.0 7-14
7-4 Partial Listing of a Steady-state Simulation 7-24
7-5 Partial Listing of a Transient Simulation 7-25
7-6 Partial Listing of the Output from DEG4 7-27
7-7 Example HEGADASS Output 7-31
7-8 Example Output of HEGADASS Showing Volume of Concentrations . . . 7-36
7-9 Example of HEGADAST Output 7-38
7-10 Example Output of HEGADAST Showing volume of Concentrations . , . 7-44
7-11 Input File for HTPOST for Scenario 5 7-48
7-12 Example PLUME Output 7-50
7-13 Example PGPLUME Output 7-54
7-14 Example SLAB Output 7-56
8-1 Example of a Methodological Data File to Identify
"Worst Case" Impacts 8-8
8-2 HEGADASS Input File to Identify "Worst Case"
Meteorological Conditions 8-9
8-3 Concatenated Input File for Scenario 3 8-10
8-4 DOS Batch File to Identify "Worst Case" Meteorology
for Case 3 8-11
8-5 Identification of "Worst Case" Meteorology for Scenario 3 .... 8-13
xx
-------
SECTION 1
INTRODUCTION
The "Guideline on Air Quality Models (Revised) "(1) provides.guidance on
the use of air quality modeling techniques for a variety of pollutants.
Although many of the models in the guideline have been used in simulating
releases of hazardous air pollutants, it has been recognized that there is a
need to provide models that specifically address impacts of hazardous air
•
pollutants.
To meet this need, EPA published "Workbook of Screening Techniques for
Assessing Impacts of Toxic Air Pollutants (Revised) "(2) and developed the
TSCREEN model<3). The workbook provides a logical approach to the selection
and use of appropriate screening techniques for estimating emission rates and
ambient concentrations resulting from eighteen different types of release
scenarios. TSCREEN, a model for screening toxic air pollutant concentrations,
is an IBM-personal computer model that implements the methods described in the
workbook. EPA co-sponsored the development of the DEGADIS refined dense gas
model'*0 and conducted a statistical model evaluation study of several dense
gas models using three experimental programs055. In 1990, EPA provided
general guidance considerations on determining input and applying several
dense gas models that are in the public domain(6). EPA has recently published
a guidance document that illustrates the possible range of different kinds of
accidental releases of hazardous air pollutants that might take place at
Superfund sites and demonstrates how atmospheric dispersion models, including
dense gas models, should be applied(7).
Title III of the Clean Air Act Amendments (CAAA) of 1990 lists many
chemicals as hazardous air pollutants and requires establishing regulations to
prevent the accidental release and to minimize the consequence of any such
•
1-1
-------
releases. With the large number of potential release scenarios chac are
associated with these chemicals, there is a need for a systematic approach for
applying mathematical models to estimate impact from these potential release
scenarios. Because many chemicals may form dense gas clouds upon release, and
refined mathematical models that can simulate these releases are complex,
particular attention is paid to models can address these types of releases.
This document provides general guidance on characterizing hazardous air
pollutant releases and shows how to apply appropriate dispersion models. It
supersedes the document "Guidance on the Application of Refined Dispersion
Models for Air Toxic Releases"(6) published in 1991. Specifically, this
document:
• Helps determine likely or reasonable storage conditions for
specific chemicals for which a release might occur;
• Helps determine release classes (e.g., liquid or gaseous phase) of
a hazardous release;
• Defines the steps to be taken when determining if a release should
be considered a dense gas release (and thus require the use of a
model capable of such a simulation);
• Defines the methods used to determine the input variables used by
commonly-used refined models in the public domain;
• Points out the implications and effects of various choices for
input information;
• Shows, by example, Che calculation of Che input variables used by
the models;
• Describes the outputs available from the models; and
• Discusses how to determine the input that gives the "worst-case"
impact conditions.
Section 2 Introduces Che term "release class" and defines the eight
release classes used in this document. General release scenarios which may
occur, and lead to the release classes, are tabulated. A method is described
1-2
-------
for the determination of likely or reasonable storage conditions for specific
chemicals.
Section 3 describes the models used in this document, which include:
• ADAM;
• ALOHA;
• DEGADIS;
• HGSYSTEM; and
• SLAB.
Section 4 provides a method to determine a release class from observable
parameters. This section also describes the methods to calculate model input
parameters and how to determine if a release should be considered a dense
release.
Section 5 gives examples of how inputs are calculated for each of the
•
release classes using the methods described in Section 4. Section 6 documents
which of the inputs calculated in Section 5 are used by each model and some of
the implications and effects of various choices for input. Section 7
describes how to determine specific output parameters from the models.
Section 8 discusses how to determine input that yields the "worst case"
impact. Appendix A discusses the relationship between some of the SI units
(Le Systeme International d'Unites). This document gives all calculation
results in SI units. Appendix B contains data associated with the chemicals
considered in this project.
The information contained in this document, especially sections 6, 7,
and 8, was obtained directly from the developers of the models or, when those
individuals were not available, from others who are intimately familiar with
that model. Specifically, the. model consultants performed the following:
• Selection of final model input;
• Running the model;
1-3
-------
Providing direction on how to interpret the output; and
Reviewing and contributing to the section of the report on
determining input for modeling the "worst case" impacts.
1-4
-------
SECTION 2
SCENARIO AND CHEMICAL SELECTION
This section describes the chemicals, release classes, and release
scenarios addressed by the models in this study. Section 2.1 describes the
chemical selection criteria. Section 2.2 introduces and defines the chemical
"release classes" under which the model input guidance is categorized in
Sections 4 and 5. Section 2.3 summarizes the example "release scenarios"
which- were chosen for analysis. Section 2.4 provides guidance on how to use
model input to make an initial release class determination.
2.1 SELECTION OF CHEMICALS FOR RELEASE MODELING
The following chemicals were chosen for the modeling study:
• Ammonia (NH3) , anhydrous and aqueous;
• • Chlorine (C12);
• Hydrogen Fluoride (HF), anhydrous and aqueous;
• Hydrogen Chloride (HC1), anhydrous and aqueous;
• Ethylene Oxide (CH2CH20) ; and
• Sulfur Dioxide (S02).
These chemicals were selected because each one meets the criteria
established for the modeling study: 1) they are all commonly used chemicals
that may form dense gas clouds when released; and 2) they all present acute
toxicity or flammability hazards and are, therefore, of particular interest to
regulatory agencies. In addition, the six chemicals together represent all of
the five basic release classes (identified and described in the following
section) when released under certain conditions.
2-1
-------
2.2 DESCRIPTION OF CHEMICAL RELEASE CLASSES
The release class defines the physical state of the chemical as it
leaves containment and the manner in which it enters the atmosphere to form a
vapor cloud. Depending on the physical properties (vapor pressure or boiling
point) and pre-release storage conditions, hazardous chemicals can be emitted
from a container as a liquid, a vapor, or both (a two-phase release).
Released liquids may form a vapor cloud through volatilization. A liquid can
be volatilized either completely or partially as it is released, forming a
vapor cloud or a vapor and droplet mixture. Conversely, vapors may partially
or completely condense to form liquid droplets when released. The degree of
condensation, if it occurs, will again depend on the pre-release conditions
and physical properties of the material. Condensed vapor may fall to the
ground to form a pool which, in turn, volatilizes to the atmosphere.
For purposes of defining dense gas model input, hazardous material
releases have been separat-ed into eight general categories of release classes.
A separate set of calculations and considerations for defining model input is
required for each release class. The methods and procedures for determining
the release class and establishing necessary model input are defined in
Section 4.0. The release class categories are listed and described in
Sections 2.2.1 through 2.2.8. Section 2.2.9 discusses additional release
class considerations, and includes an example of a hazardous material release
that falls into multiple release classes.
Vapor release rates may be limited due to the "choked" flow phenomenon
which develops when containment pressures are relatively high. Choked flow
refers to the velocity limit at which a material may be released and depends
on the" containment pressure and the physical properties of the material.
Since the release velocity is limited under choked flow conditions, the
volumetric flow rate through a given hole size is also limited, and is
independent of containment pressure. However, the mass emission rate will
vary with containment pressure above the choked flow limit, since the
material's density varies with pressure.
2-2
-------
2.2.1 Two-phase Gas Release (Choked)
A two-phase gas release is a release of a gas that partially condenses
to form two phases (gas and liquid). Conversely, a two-phase liquid release
begins as a liquid and partially vaporizes to form two phases. In terms of
post-release characteristics, two-phase releases are similar regardless of
whether the material is initially a gas or liquid. However, the distinction
between two-phase gas and liquid releases is important, since the model input
calculations (primarily to determine the emission rate) are different for the
two cases.
A two-phase gas release is a release of a material that is normally a
gas under ambient conditions. The material may be stored either as a
compressed gas or as a liquified gas; however, a two-phase gas release (as
opposed to a two-phase liquid release) from liquified gas storage must come
from the vapor space within the storage container. A two-phase gas release
occurs due to cooling of the gas when it expands as the material's pressure
drops upon release. The resulting adiabatic expansion subcools some or all of
the material (subcooling refers to cooling to a temperature below a material's
boiling point). Thus, a liquid phase is formed immediately after the release.
Figure 2-1 illustrates the two-phase gas release class. As the figure
shows, there are several possible fates for the liquid formed in the gas. The
liquid may "rain out" from the vapor/liquid release and form a liquid pool
which will evaporate. The condensed liquid may also form an "aerosol", or
finely dispersed mist, which is transported downwind in the vapor cloud where
it ultimately vaporizes. The formation of a liquid pool, aerosol, or
combination of the two is dependent upon the material and its release
characteristics.
In a cwo-phase choked gas release, containment pressures are
sufficiently high as to cause both condensation upon depressurization and
attainment of choked flow conditions. At containment pressures above the
2-3
-------
Liquid Pod
Figure 2-1. Two-phase Gas Release
choked flow threshold, the release velocity (and, therefore, the volumetric
release rate through a given hole size) is independent of containment
pressure. However, the mass emission rate will vary with containment pressure
above the choked flow limit since the material's density varies with pressure.
2.2.2 Two-phase Gas Release (-Unchoked)
The condicions defining a cwo-phase unchoked gas release are similar to
chose chat define a cwo-phase choked gas release. In a two-phase unchoked gas
release, containment pressures are sufficiently high to cause condensation
upon depressurization, but not so high as to result in choked flow conditions.
Model input calculations, particularly those for determining the
emission rate, are different for choked and unchoked gas release conditions.
At pressures below the choked limitation, the release velocity (and therefore,
the volumetric release rate through a given hole size) is a function of the
containment pressure. At pressures above the choked flow limitation, the
release velocity through a given hole size is independent of containment
pressure.
•
2-4
-------
2.2.3 Two-phase Pressurized Liquid Release
A two-phase liquid release is a release of a liquid that partially
vaporizes to form two phases (liquid and gas). A "two phase pressurized
liquid" release is a release of a liquid phase material that is a gas at
ambient conditions. The material is held in the liquid phase via pressure in
the containment vessel. Upon release, the liquid pressure immediately drops
to ambient pressure, resulting in sudden vaporization of some or all of the
liquid. This phenomenon is termed "adiabatic flash". The flash fraction
defines che fraction of liquid which flashes to vapor, and the extent of
flashing depends upon the storage temperature and physical properties of the
material (e.g., the a ambient boiling temperature and the heat of vaporization
of the material).
Figure 2-2 illustrates a two-phase liquid release under conditions where
flashing occurs. As in the case with two-phase gas releases, the liquid
portion of the release may either form an aerosol, a short-lived liquid pool,
or both. The tendency for aerosol formation is greater when the material is
stored under pressure than when the material is refrigerated (described in the
following section) since the release energy tends to be relatively high for
pressurized releases.
Ptmhli baud foot
Figure 2-2. Two-phase Liquid Release
2-5
-------
2.2.4 Two-phase Refrigerated Liquid Release
A two-phase refrigerated liquid release is a release, from a containment
vessel in which the material is held in the liquid phase via refrigeration.
The material exists as a vapor at ambient temperature, but is stored at a
temperature below its boiling point (i.e., it is stored as a subcooled
liquid). If flashing occurs, the release is usually mildly flashing, since
the release temperature is normally just above the material's boiling point.
Therefore, the flashed fraction in a two-phase refrigerated liquid release is
usually much less than that of a two-phase pressurized liquid release.
Regarding the de-termination of model input, the primary distinction
between two-phase refrigerated liquid release and a two-phase pressurized
liquid release involves the calculation of the emission rate, which is based
on the difference between the containment pressure and the ambient pressure.
Since refrigerated liquids are stored at ambient pressure, containment
pressure is determined by the mass of liquid above the release location (i.e.,
the height of the liquid level). This force is typically much smaller than
that of the pressurized case, and therefore usually results in a lower release
rate.
The released material from a two-phase refrigerated liquid release may
form an aerosol and/or a liquid pool. If a pool is formed, it will volatilize
as it absorbs heat from the surroundings and the spill surface. As the pool
evaporates it will cool and the evaporation rate should decrease. The
tendency for aerosol formation from a refrigerated liquid may be minimal since
the release energy is typically low.
2.2.5 Single-phase Gas Release (Choked)
A single-phase gas release is a release of a material that is a gas at
ambient conditions and remains a-gas throughout the release (i.e., no liquid
phase develops). The material may be stored as either a compressed or a
liquified gas. A single-phase gas release from liquified gas storage must
come from the vapor space within the containment vessel.
2-6
-------
Figure 2-3 illustrates a single-phase gas release. A gas release will
not condense if its pressure is too low to result in "subcooling" the material
during the release (i.e., to a temperature below its boiling point). Choked
flow develops if release pressures are sufficiently high to reach the choking
velocity limitation (refer to Section 2.2.1).
It should be noted that the development of choked velocities and the
condensation of gas releases is highly dependent upon the physical properties
of the released material. Section 2.2.2 described conditions where release
pressures were such that condensation occurred but choked flow did not
develop. For a single-phase, choked gas release, the situation is reversed.
That is, the pressures are such that no condensation occurs but choked flow
does. Since these phenomena are chemical-property dependent, one material may
exhibit the "condensation but no choked flow" combination, whereas another may
exhibit the "choked flow but no condensation" combination.
GwOnJy
Figure 2-3. Single-phase Gas Release
2.2.6 Single-phase Gas Release (Unchoked)
Conditions resulting in a single-phase unchoked gas release are similar
to those that result in single-phase choked gas releases. That is,
containment pressures are below those that cause either condensation or
choked-flow conditions during the release. -
As discussed in Section 2.2.2, emission rate calculations for defining
model input are different for choked and unchoked gas release conditions. At
pressures below the choked limitation, the release velocity (and thus the
2-7
-------
volumetric release rate through a given hole size) is a function of the
containment pressure. For choked gas releases, the release velocity is
independent of the containment pressure.
2.2.7 Single-phase Liquid Release (High Volatility)
A single-phase release of a high volatility liquid is a release of a
material that, although a gas at ambient conditions, is stored in the liquid
phase through refrigeration to a temperature below its boiling point (the
material is "subcooled"). The release is a single-phase release if its
release temperature is at or below the normal boiling point of the material,
so that no flashing occurs.
Figure 2-4 illustrates a single-phase high volatility liquid release.
The released liquid will form a pool which will volatilize as it absorbs heat
from the surroundings and the spill surface. The temperature of the pool will
be reduced as material evaporates. As previously discussed, the tendency for
aerosol formation from a refrigerated liquid may be minimal since the release
energy is typically relatively low. When calculating the atmospheric emission
rate (the rate of vapor cloud formation) for a single-phase high volatility
liquid, it is assumed that immediate volatilization of the material occurs
(for model input calculations, see Section 4.0).
Figure 2-4. Single-phase High Volatility Liquid Release
-------
2.2.8 Single-phase Liquid Release (Low Volatility)
A singl'e-phase release of a low volatility liquid is a release of a
material that is a liquid at ambient conditions. Neither refrigeration nor
pressurization is needed to hold the material in liquid form. Figure 2-5
shows the single-phase low volatility liquid release.
A low volatility liquid will, form a liquid pool upon release. The rate
of ejection of mass from the container into the pool depends on the release
size and height of che liquid level above the hole in the containment vessel.
The atmospheric emission rate (the rate of vapor cloud formation) is only
dependent on evaporation. Therefore, the area of the pool, the ambient
temperature and wind speed, and'the volatility (vapor pressure) of the
released liquid determine the atmospheric emission rate.
Figure 2-5. Single-phase Low Volatility Liquid Release
To determine the rate of atmospheric emission (rate of vapor cloud
formation) of a low volatility liquid, the calculated liquid ejection rate
must be compared to the atmospheric emission (volatilization) rate to
establish the limiting condition. Section 4.0 explains how this comparison is
made to determine Che atmospheric emission rate for use as model input.
2.2.9 Additional Release Class Considerations
A given "release scenario" may simultaneously or sequentially produce a
release that fits more than one of the release classes described in the
preceding sections. To define the term "release scenario", and to illustrate
2-9
-------
how multiple release classes may develop, the following example scenario is
provided.
A hazardous chemical which is a dense vapor at ambient conditions is
stored as a liquid under pressure. A pipe leading from the bottom of
the tank is severed, and the pressurized liquid flows toward the open
pipe. The pressure of the liquid drops as it flows through the pipe,
resulting in vaporization of part of the liquid within the pipe
(adiabatic flash) before it reaches the point where it is released to
the atmosphere. Once all of the liquid within the vessel is released,
the remaining compressed vapor is expelled through the release opening.
Figure 2-6 illustrates this scenario. As shown, this type of release
scenario initially results in a two-phase pressurized liquid release. The
liquid fraction may form an aerosol, a short-lived liquid pool, or a
combination of the two. If all of the liquid fraction is assumed to form
aerosol, the atmospheric emission rate (vapor cloud formation rate) is assumed
to be equal to Che mass emission rate of both the liquid and vapor phases. If
all of the liquid fraction is assumed to'form a pool, the emission rate would
be a combination of the rate of the vapor release rate and the rate at which
the pool evaporates. If partial aerosol formation is assumed, a combination
of the two calculations would be used.
Pressurized
Liquified
Gas
POM to*
Aaratx Oroowi
Figure 2-6. Example Release Scenario
2-10
-------
It should be noted that, for the purposes of this modeling study, the
rate of vapor cloud formation from a two-phase pressurized liquid release is
assumed to be equal to the sura of the vapor and liquid release rates from the
vessel. The basis for this assumption is that any pool that forms will
evaporate very quickly, and the assumption of immediate vaporization results
in a maximum vapor cloud generation rate which tends to produce conservative
modeling results.
A second distinct release class results when all of the liquid is
released from the vessel. The remaining compressed vapor will be expelled
from the tank under either two-phase or single-phase gas release conditions,
which may be either choked or unchoked. The specific gas release class and
rate is determined by the pressure of the gas within the vessel. Under actual
conditions, the tank pressure will decrease as the gas is expelled. For
purposes of this study, however, the initial pressure is used to calculate the
maximum release rate, which is assumed to be constant for the duration of the
release. A maximum release rate tends to produce conservative modeling
results.
A dense gas modeling analysis of scenarios involving multiple release
classes may require combining results from multiple modeling runs. Where
multiple release classes occur simultaneously, the vapor cloud generation
rates from each class may be added to determine model input. Sequential or
overlapping release classes may need to be considered individually and in
combination to determine the most reasonable approach for interpreting
results.
Table 2-1 shows the possible release classes which can result from
common storage configurations. The table summarizes the criteria for
determining the initial release class, and lists potential subsequent release
classes chat might develop for each-storage situation. Mos-t release scenarios
chosen for this study are based on common storage configurations.
2-11
-------
TABLE 2-1. POSSIBLE RELEASE CLASSES RESULTING FROM TYPICAL STORAGE CONDITIONS
Storage Condicion
Initial Release Class
Release Class Considerations
Liquified Gas
(Pressurized)
Liquified Gas
(Pressurized)
Liquified Gas
(Refrigerated)
Liquified Gas
(Refrigerated)
Two-phase Pressurized Liquid
(hole below liquid level)
Single- or Two-phase Gas
(hole above liquid level)
Single-phase, High Volatility
Two-phase Liquid
(hole below liquid level)
Liquid phase of release may form aerosol droplets
and/or liquid pool, which will vaporize very
quickly. Vapor release will result after liquid
level falls below height of hole, or after all
liquid is released from the vessel. Vapor release
could be two-phase or single-phase, choked or
unchoked, depending on release pressure.
Release will be two-phase if condensation occurs,
choked or unchoked. Subsequent release classes
could involve transition from two-phase to
single-phase, and/or choked to unchoked if vessel
pressure drops over time. While liquid is in
vessel, however, pressure will remain relatively
constant unless substantial liquid cooling results
from vaporization in vessel.
Liquid will remain single-phase if release
temperature is below boiling point. Liquid will
form^ short-lived pool which evaporates as heat is
absorbed from surroundings. Vapor release will
occur after material liquid level drops below hole.
Pressure of vapor release will depend on liquid
temperature in vessel, but will normally be below
condensation and choked flow pressures.
Two-phase release occurs if release temperature is
above material's boiling point. Release is normally
mildly flashing since release temperature is
normally near boiling temperature. Possible
subsequent release classes are same as for two-phase
liquid release from pressurized vessel.
-------
TABLE 2-1
(CONTINUED)
Storage Condition
Initial Release Class
Release Class Considerations
N)
Liquified Gas
(Refrigerated)
Single-phase Gas
(hole above liquid level)
Compressed Gas
Single- or Two-phase Gas
Liquid
(Chemical is liquid
at ambient
temperature)
Single-phase, Low Volatility
Liquid
Since storage condition dictates that material is
cooled to below' boiling point, vapor pressure of
liquid will be less than one atmosphere. Thus,
vapor will be released relatively slowly as single-
phase gas (as long as liquid remains cooled).
Pressures will most likely be below minimum required
for either condensation or choked flow conditions.
No subsequent release classes will develop.
Condensation will depend on material properties and
storage pressure. Flow could be choked or unchoked.
Subsequent release classes are same as for single-
or two-phase gas release from pressurized liquified
gas vessel. Since no liquid is in container, the
vessel pressure will continuously drop during
release.
Released liquid will form pool that will volatilize
at a rate depending on material's vapor pressure.
Minor vapor release from vessel will occur after
liquid is expelled.
-------
2.3 SELECTION OF RELEASE SCENARIOS FOR MODELING
A total of eight release scenarios covering each of the release classes
described in Section 2.2 were assembled for the purposes of this study. Most
of the scenarios were selected from a larger general list, which covers most
of the release class possibilities for each selected material in a typical
storage condition.
Table 2-2 lists the eight release classes covered in the modeling
analysis, as well as general release scenarios which could produce each of the
release classes for a given chemical. Also included in Table 2-2 are the
normal boiling temperatures of the each material, and a summary of the post-
release state and dense gas vapor cloud hazards for each material.
Several "in-process" scenarios were selected since not all release
classes could be covered using reasonable storage conditions with materials in
vapor-liquid equilibrium (a condition which exists for nearly all storage
conditions). In addition, a stack release was selected to produce a suitable
example of a waste gas stream which is a combination of a hazardous material,
air, and water vapor. This scenario was selected primarily to illustrate the
determination of release density for a typical stack release.
Table 2-3 summarizes the release scenarios selected as examples for
discussion in this document. A description of the release scenario is
provided, along with the release class each scenario represents. The ehtylene
oxide saturated vapor phase pipeline release scenario is used for two release
classes (two-phase gas release, choked and unchoked) . Although some of the
selected release scenarios could produce multiple release classes, only one
class was addressed for each. Each of the eight release classes described in
Section 2.2 was included to provide an example of model input determination
for each class.
2-14
-------
TABLE 2-2. STORAGE RELEASE SCENARIOS AND POST-RELEASE CHARACTERISTICS OF SELECTED CHEMICALS
Release Class
Aniuon la. Anhydrous
and Aqueous
Chlorine
Hydrogen fluoride,
Anhydrous and
Aqueous
Hydrogen Chloride,
Anhydrous and
Aqueous
Ethylene Oxide
Sulfur Dioxide
Boiling Point (for
Anhydrous
Material)
Single-phase Gas
(Choked or
Unchoked)
-33.4*C (28.1*F)
1. Bole in NH,
(anhydrous)
pressurised liquid
storage vessel
above liquid
level,
-34.6*C (-30.3*P)
1. Hole in liquid
storage vessel
above liquid
level.
19.*'C (66.9*P)
1. Hole In HP
(anhydrous)
pressurised liquid
storage vessel
above liquid
level.
-85*C (-121*P)
Hole in
pressurised HC1
(anhydrous)
storage vessel.
10.6'C (50.1'P)
1. Bole in liquid
storage vessel
above liquid
level,
-10'C (1**P)
1. Hole in liquid
storage vessel
above liquid line,
Two-phase Gas
(Choked or
Unchoked)
Two-phase
Refrigerated
Liquid
^ Single-phase,
!_. High Volatility
t-n Liquid
''
Two-phase
Pressurised Liquid
Single-phase Low
Volatility Liquid
.
or
2 Bole in NH,
(anhydrous) vapor
storage vessel.
Bole in NH,
(anhydrous)
refrigerated
liquid storage
vessel .
Bole in NH,
(anhydrous)
refrigerated
liquid storage
vessel.
Bole in NH,
(anhydrous)
pressurized liquid
storage vessel
below liquid
level . Also may
form a pool.
Hole in NH,
(aqueous) tank
below liquid
level. Liquid
pool formed.
2. Bole In vapor
storage vessel.
N/A
N/A
Hole In
pressurized liquid
storage vessel
below liquid
level . Also may
form a liquid
pool.
N/A
or
2. Hole In BP
(anhydrous) vapor
storage vessel.
Hole In HF N/A
(anhydrous)
refrigerated
liquid storage
vessel.
Hole In HP N/A
(anhydrous)
refrigerated
liquid storage
vessel.
Hole In HF N/A
( anhydrous )
pressurised liquid
storage vessel
below liquid line.
Also may form
liquid pool.
Hole in HF Bole in HP
(aqueous) tank (aqueous) vessel
below liquid below liquid
level. Liquid level. Liquid
pool formed. pool formed.
2. Bole in vapor
storage vassal.
N/A
N/A
Hole in EO
pressurised liquid
storsge vessel
below liquid
level.
N/A
2. Hole In vapor
storage vessel.
N/A
' N/A
Hole in
pressurised liquid
storage vessel
below liquid line.
Also may form
liquid pool.
N/A
-------
TABLE 2-2. (CONTINUED)
Release Class
Single-phase High
Volatility Liquid
Ammonia , Anhydrous
and Aqueous Chlorine
Uol. In NH, N/A
(anhydrous)
refrigerated
liquid storage
v*ss*l .
Bydrogen fluoride,
Anhydrous and
Aqueous
Bole In BF
(anhydrous)
refrigerated
liquid storage
vessel.
Hydrogen Chloride,
Anhydrous and
• Aqueous Ethylene Oxide
N/A >N/A
Sulfur Dioxide
N/A
Post-Release State
Ga», normally
lighter than air.
Can be heavier
than air due to
aerosol effect)
liquid droplets
and cooling fram s
flash. Also, a
brief-lived
boiling liquid
ralease Is
possible.
Gas (2.5 tines
heavier than air)
plus aerosol.
Also, a brief-
lived boiling
liquid ralease is
possible.
Gas which is
lighter than air
but could be
heavier than air
due to self-
association of BF
at low P, 1, and
due to aerosol
effect. Also, a
boiling liquid
release is
possible.
Gas (1.3 times
heavier than air).
Aqueous HC1 liquid
Gas (1.5 times
heavier than air).
Gas (2.2 times
heavier than air).
I
*-•
o\
Post-Release
Basards and
Chain es
Aqueous MB, liquid
Toxic, flammable
(but high minimum
Ignition temp.--
1562*F and narrow
fl tamable
concentration
range, 16-25X
volume).
Reactive and
corrosive when
combined with
water.
Toxic, but not
flammable.
Corrosive when
combined with
moisture (HC1
formation).
Aqueous HF liquid
Toxic,
hygroscopic; heat
evolves and forms
corrosive
hydrofluoric acid
•when combined with
water.
Toxic, non-
flammable, highly
corrosive.
Toxic, highly
flammable.
Toxic.
Honcombustlble.
Combines with
water to form
sulfurous acid
(H.SO,). Also may
be reduced by Ba in
the presence of
water to form R,S.
N/A indicates that releue class does not occur for typical storage conditions of the material.
-------
TABLE 2-3. DESCRIPTIVE SCENARIOS FOR MODELING
Chemical
Scenario Description
Illustration
Chlorine
Tubing from bottom of pressurized, ambient temperature
vessel storing liquified chlorlng Is broken off. This
produces Initial two-phase pressurlted liquid release.
Chlonne Ton Conlaincr
Gas Hductor
0,Gu
CI2 Liquid
I
\
==«
I'ully Developed
Two-ltiase Mow
I ii|ind luluclor
Hydrogen Fluoride,
anhydrous
Rupture disk breaks and relieves material from vapor
•pace of pressurized, ambient temperature tank storing
liquid HF. This produces an Initial single-phase gas
release (choked)
Tramlcr Line
Kupiure
Disk
Clinked V.ipor I'liasc
I IF Release ^
Hydrogen Fluoride Liquid
Hydrogen Chloride,
anhydrous
Water flow Is lost to tower absorbing HC1 from vent
stream. This results In single-phase (as release
(unchoked).
Vapor I'liise
llydiogcn Chloride Kdeasc
with Air and Water Vapor
I lydrogcn Chlonde_
& Air
Tower Water Supply
AbsorpUon Tower
-------
TABLE 2-3. (CONTINUED)
Chemical
Hydrogen Chloride,
aqueous
Ethylene Oxide
to
i
t-*
Co
Scenario Description
Illustration
Hole develops In side of atmospheric pressure tank
storing 30 ut.X hydrochloric acid In liquid phase.
results In a single-phase, low volatility liquid
release.
This
Hole develops In refrigerated vessel storing liquid EO
at atmospheric pressure. Since hole Is below the liquid
level, the Initial celease Is a single-phase, high
volatility liquid release.
Saturated 30% Hydrochloric
Acid at Ambient Temperature
Low Volatility
Spill Contained
in Dike
Liquid Spill
Refrigerated
Eihylene Oxide
Refrigerated Liquid Release
(Immediate Vaporization)
Ethylene Oxide
Bole develops In saturated vapor phase pipeline. This
leak results In two-phase gas release. The release flow
rate may be choked or unchoked. Thl* scenario Is used
In two release class examples.
Bhylcni; Oxide Pipeline
Eihylene Oxide Vapor ^
at Constant 1 cmncraumr
Two Pli.ise ti.is Hclc.isc
Sulfur Dioxide
A hole develops In a pipe from the bottom of a vessel
storing liquid S02. Initial release Is a two-phase
refrigerated liquid release.
Refrigerated Sulfur Dioxide Liquid I Two Phase
/ Refrigerated
J Liquid Release
^/ i«»
JL
To
Process
-------
The selection of release scenarios for hazardous material release impact
modeling for an actual modeling submittal is not a trivial matter. Scenarios
are most frequently selected as a result of a formal "hazard analysis" study.
One of the goals of a hazard analysis is to produce a set of release scenarios
meeting predetermined criteria for probability of their occurrence. The
process of combining scenario probability (which may be determined
qualitatively or quantitatively) with consequence analysis results (determined
by applying impact modeling results) is termed "risk assessment". The final
result provides useful qualitative and/or quantitative data on which emergency
planning or other risk management decisions (e.g. whether to invest in risk
reduction measures) are based.
2.4 VERIFICATION OF RELEASE SCENARIOS IN THE REVIEW OF MODELING SUBMITTALS
The starting point for any modeling submittal is the release scenario(s)
chosen for modeling. The level of detail used to describe the release
scenario(s) in the submittal may vary widely depending on specific regulatory
requirements. At a minimum, the storage, processing, or transportation
conditions necessary to define all model input should be included in the
submittal to allow for complete model input verification.
The purpose of this section is to outline some procedures that can be
used to check and verify pre-release conditions defined in scenario
descriptions. Also, this section includes some basic guidance for determining
an initial release class. The final determination of release class should,
however, be made using the procedures outlined in Section 4.3.
Table 2-4 provides a listing of typically-encountered storage conditions
for each of the chemicals chosen for this modeling study. For releases
involving storage of any of the chemicals, the storage phases and volumes can
2-19
-------
TABLE 2-4. TYPICAL STORAGE QUANTITIES FOR SELECTED CHEMICALS (ANHYDROUS)
Storage
Cylinders
Ton Containers
Single Unit
Tanks
Tank Trucks
Barges
Fixed Storage
Vessels
Ammonia'"'
Stored
Quantity Phase*
100. 150 Lp
Ib
2000 Ib Lp
26.5 tons Lp
15-20 Lp
tons
bOO, 1100 Lp
tons
25+ tons Lp or
Chlorine'9'
Quantity
100, 150
Ib
2000 Ib
16, 30,
55, 85,
90 tons
15-20
tons
600, 1100
tons
25+ tons
Stored
Phase*
LP
LP
^
LP
LP
LP
Sulfur Dioxide'10'
Quantity
100, 150
Ib
2000 Ib
20, 30.
40. 50
tons
15-20 tons
600. 1100
tons
25+ tons
Stored
Phase*
LP
LP
LP
LP
*
LP
Ethylene
Quantity
176 Ib
72.5 tons
(20,000
gallons)
25 tons
Oxide'1"
Stored
Phase*
LP
LP
LP
Hydrogen
Fluoride'12'
Stored
Quantity Phase*
150 Ib Lp
20-90 Lp
tons
20-90 Lp
tons
Hydrogen
Chloride'11'
Stored
Quantity Phase*
150 Ib G
NJ
O
* Scored Phase Types:
Lp - Liquid phabc held by pressurization
LR - Liquid phase held by refrigeration
G - Gas
-------
be checked against Table 2-4. While the list does not include all conceivable
storage configurations, it will allow reviewers of modeling submittals to
determine if the described condition should be questioned or whether it is
within the range of normal conditions.
The best indicator of the types of release classes that can be expected
from a given release scenario is the boiling point of the chemical. The
boiling point determines both the material's pre-release state and the
potential release class. A material with a boiling point above ambient
temperature may be stored or processed as a liquid without the aid of
pressurization or refrigeration. Conversely, a material with a boiling point
below ambient temperature is a gas that must be compressed or cooled to be
held in the liquid phase. For pressurized liquid releases, a simple check for
flashing can be made by comparing the boiling point of the material to ambient
temperature. If the ambient temperature is higher than the boiling point,
there will be some flashing during the release.
The boiling point of the material can be used to make an initial
determination of the reasonableness of modeling submittals. For instance, if
a submittal is based on a scenario resulting in a vapor release of benzene
(normal boiling point 80.1'C or 176.1°F) from the bottom of a process vessel,
the pre-release condition is not possible unless the vessel is heated above
benzene's boiling point or the vessel operates under substantial vacuum. The
boiling points of the chemicals selected for the modeling study are given in
Table 2-2.
The boiling point of a material represents the equilibrium temperature
at which its vapor pressure is equal to atmospheric pressure. The equilibrium
relationship between all three phases (liquid, gas, and solid) at any given
pressure and temperature is conveniently represented by a triple-point
diagram. For storage scenarios, which almost always involve liquid-vapor
equilibrium conditions, the use of the triple-point diagram provides a simple
and accurate method for verifying the reasonableness of pre-release
conditions. Triple-point diagrams for each chemical selected for the modeling
study are included in Appendix B.
2-21
-------
The use of triple-point diagrams to check pre-release conditions is best
illustrated by the following example.
Model input is calculated based on pro-release conditions for anhydrous
ammonia stored as a liquid under pressure. The storage vessel is
isolated from the process at the time of the release and, therefore, the
liquid ammonia is in equilibrium with its vapor at ambient temperature
(specified as 30°C, or 86°F). A hole develops in the vapor space of the
vessel resulting in a single-phase gas release. The emission rate, as
input to the model, is calculated using a containment pressure of 152
kPa absolute (1.5 atmospheres).
Figure 2-7 shows the triple-point diagram for anhydrous ammonia. To
check the reasonableness of the pre-release conditions, a point is located on
the liquid-gas equilibrium line at the specified release temperature of 30°C.
Figure 2-7 shows that the corresponding equilibrium pressure is 11 atm, which
is well above the 1.5 atmospheres used to calculate the vapor emission rate in
the example. Since vapor release emission rates increase with increasing
storage pressure, the example analysis results in an underprediction of the
emission rate, and most likely an underestimation of vapor cloud impacts.
The triple-point diagram represents equilibrium conditions at which
materials can exist in multiple phases. Conversely, the diagram can be used
to show where only a single phase is possible. If the pressure-temperature
combination given in the above example were accurate, the triple-point diagram
shows that the stored ammonia would exist exclusively as a vapor. Thus, the
diagrams can be used to check for the proper material phase, as well as for
consistency between pressure and temperature for liquid/vapor equilibrium
conditions.
To verify the pre-release conditions of a material in vapor-liquid
equilibrium, any two of the following conditions must be known: 1) the
temperature, 2) the pressure, and 3) the phase of the material. With two
conditions known, the third condition can be readily determined using Che
triple-point diagram as in the preceding example. When only one condition is
known, the parameter which is known can be used to intuitively estimate the
pre-release state using the method given in Figure 2-8. Figures 2-8 and 2-9
2-22
-------
to
/
B
£
a
s
120
110
100
90
80
70
50
40
30
20
10
-10
Solid
Crlllckl Point
(I32.6°C, 111.31 aim)
Liquid
Gas
Triple Polnl Region
1.9
AMMONIA
OtUII or Triple Polnl Rcglo
-200
-.100
100 200
Temperature (°C)
300
400
500
Figure 2-7. Triple Point Diagram for Anhydrous Ammonia
-------
ro
START
with known stored chemical
How many
conditions (phase.
T. P) are known?
If pressure is known
Assume material is a
compressed gas at
ambient T.
Determine or
estimate storage
quantity and use
ideal gas law
(P-nRT/V)tofmdP
Assume storage is al
ambient T. Find P
from vapor-liquid
curve on triple point
diagram.
Material is likely a
refrigerated liquified
gas with P< 14.7
psia.
If boiling point is > ambient T,
material is a low volatility
liquid (use vapor-liquid line on
triple point diagram). If boiling
point is < ambient T, material is
either a compressed gas (ideal
gas law) or a pressurized liquid
(vapor-liquid line). Storage mass
and volume could be used to
determine phase by comparing
to liquid and vapor densities.
Material is either a
refrigerated liquified
gas or a low volatility
liquid (phase is liquid).
T can be found from the
vapor-liquid line on Ihe
triple point diagram.
Material is either a
compressed gas (ideal
gas law) or a
pressurized liquid
(vapor-liquid line).
Storage mass and
volume could be used
to determine |4iase by
comparing to liquid
and vapor densities.
Figure 2-b. Method for Determination of Pre-Release Conditions Where One Condition is Known
-------
10
ui
Which variable must be found? I
-Phi
-Tempcraturi
Gas
Material is a •
compressed gas.
Determine or estimate
storage quantity and
use ideal gas law
(P=nRT/V)tofindP
Material is a
refrigerated liquified
gas wilhP< 14.7psia.
Exact P can be
determined from the
vapor-liquid line on
the triple point
diagram.
^
"^^ <
r
Material is a compressed
gas. Determine or
estimate storage quantity
and use ideal gas law
(T-PV/nR)tofindT.
Material is either a
liquified gas under
pressure or a low
volatility liquid.. P can be
determined from the
v apor-liquid line on the
triple point diagram.
Material is either a
refrigerated liquified
gas or a low volatility
liquid. T can be
determined from the
vapor-liquid line on the
triple point diagram.
Material is a liquified
gas under pressure. T
can be determined from
the vapor-liquid line on
the triple point
diagram.
Find point on triple
point diagram
corresponding to known
P, T. Region will
indicate the phase.
Figure 2-9. Method for Determination of Pre-Release Conditions Where Two Conditions are Known
-------
Include the logic used to check pre-release conditions when only partial
information is available in a modeling submittal.
The triple-point diagram can be used to determine unknown parameters
only when the material is in liquid-vapor equilibrium. For materials that are
stored only in the gas phase (not in equilibrium with its liquid), the exact
pre-release conditions can be defined only if the stored quantity is known.
The assumption that the stored quantity is equal to the capacity of the vessel
will most likely result in a conservative emission rate estimate. For
compressed gases, the assumption of maximum storage quantity results in
maximum storage pressure, which produces the maximum initial emission rate. A
maximum quantity assumption tends to produce more conservative vapor cloud
impacts since the release duration will also tend to be maximized.
Figures 2-8 and 2-9 provide a method to define or intuitively estimate pre-
release conditions for compressed gasses, as well as for materials in vapor-
liquid equilibrium.
Figures 2-8 and 2-9 apply to conditions where the material is contained
in the gas phase or under liquid-gas equilibrium conditions. Where in-process
release scenarios are specified, the material may or may not be at equilibrium
conditions just before the release. Where non-equilibrium releases are
specified, the pressure and temperature are usually measured or can be
estimated from the design limits of the vessel. For example, a pressure
relief valve is designed to open when the pressure reaches a predetermined
value. If the consequences of a release are Co be investigated, the pressure
of the release can be assumed to be that predetermined value.
Some components of the release class can be readily determined after the
pre-release conditions have been verified. As previously described, the
material's boiling point provides an indication of the types of release
classes that can be expected to result from a given release scenario. Further
determination of the release class, specifically, the degree of flashing of a
pressurized liquid release, can also be verified. A "flash diagram" may be
used to determine the approximate fraction of liquid that will flash to vapor
during a release. Use of the flash diagram is best illustrated through the
2-26
-------
use of the following example.
A modeling submittal specifies a release of chlorine from a failure at
the bottom of a pressurized liquid storage vessel. The scenario
description states that an emergency containment/sump system is in place
that removes 90 percent by mass of the released liquid chlorine before
it vaporizes from a pool which presumably has formed. The storage
temperature is specified as 25°C (77°F).
Figure 2-10 is the "flash diagram" for chlorine showing the degree of
immediate vaporization of liquid chlorine as it is released from containment.
According to the specified release temperature, (25°C), 20% of the liquid
chlorine flashes as it is released from the vessel. The flashed material
immediately forms a vapor cloud, rather than forming a liquid pool. The
flashed fraction determined from Figure 2-10 exceeds the 10% assumed in the
example scenario as the amount ultimately entering the atmosphere as a dense
gas vapor cloud. The assumption in the example scenario thus results in a
substantial underprediction of the atmospheric emission rate of chlorine and,
most likely, an underprediction of the downwind vapor cloud impacts.
-50
50
10t 150 200 250 300 350
Tcmpmtm CO
Figure 2-10. Flash Diagram for Chlorine
The procedures outlined in this section provide a method to estimate
pre-release conditions for a given release scenario and a simple technique to
2-27
-------
SECTION 3
MODELS
This section contains short descriptions of the five models illustrated
in this document. Those models are:
• ADAM;
• ALOHA;
• DEGADIS;
• HGSYSTEM; and
• SLAB.
It is assumed that, before a model is exercised, the user will become
familiar with the chosen model through use of technical references and user's
guides written for that model. This document is not meant to be a replacement
for a given model's documentation. Descriptions are provided here simply to
allow prospective users to learn the general features of those models with
which they may be unfamiliar.
3.1 ADAM<13)
ADAM is a PC-based dispersion model. It allows the user to specify the
source-term right after a release or select from a menu of source-term
options. However, the source-term must be either an instantaneous release or
a steady-state continuous release; the model cannot handle a limited-duration
release. Also, ADAM is not intended for elevated releases. Furthermore, ADAM
does contain an algorithm for dealing with releases from jets, it treats the
jet as being on the ground, and horizontally aligned with the wind.
ADAM has a database of chemicals, from which the user must select the
chemical input for the model run. The database contains eight chemicals. It
3-1
-------
is possible, however, to increase the database to include more than the
original eight chemicals. Other input the user provides are the concentration
contour of interest, and the averaging time.
ADAM will output a single contour showing the location of the gas cloud
at the specified concentration level, along with a table showing the amount of
time travelled by the eloud, its velocity, and the width of the contour. ADAM
will also provide peak concentrations as a function of distance and, if the
release is instantaneous, a peak dose may be calculated as a function of
distance for the specified averaging time. In both cases, the information is
calculated for the centerline of the contour. ADAM is unable to determine a
time-history of concentration at specific points.
3.2 ALOHA(14)
ALOHA is a PC-based, menu-driven model, which employs a graphical user
interface (GUI). There are five menus from which the user may select input
for a model run. These menus are: "Chemical," "Site Data," "Atmospheric,"
"Source," and "Computational." The release type may be selected from the
"Source" menu. There are three options: .1) pool evaporation; 2) liquid or
gas leaks from a tank; or 3) a gas leak from a ruptured pipe. The user does
not have to enter the release type if he or she knows the amount of chemical
entering the atmosphere as a gas, along with other information, such as the
temperature of the gas.
If the chemical desired is in the database, only the name need be
specified; otherwise, all the data necessary can be input from the "Chemical"
menu.
Output from the model includes a text summary, a plot of the "footprint"
of concentrations that are greater than a user-specif-ied concentration, time-
series plots of concentration and dose at a particular location, and a time-
series plot of the release site.
3-2
-------
ALOHA also has a version of DEGADIS imbedded in it. ALOHA's formulation
is almost identical to the stand-alone DEGADIS. However, some of the
numerical techniques to solve the conservation equations are not the same.
Also, the two models' output formats are quite different. The "official" name
of the ALOHA dense gas model is "ALOHA-DEGADIS."
3.3 DEGADIS
(4)
DEGADIS is a PC-based model that can be applied to a wide range of
releases including: gases and aerosols; continuous, instantaneous, finite-
duration, and time-variant releases; ground-level, low-momentum area releases,
and ground-level or elevated upwardly-directed stack jet releases.
For aerosol releases, the user must characterize the density of the
release, because the model is not capable of doing so itself. The model only
simulates one set of meteorological conditions and does not accept any type of
real-time meteorological data. Data, including chemical properties, must be
input interactively, or from a. file, for each run, since there is no chemical
database. Terrain is assumed to be flat and unobstructed. Input required
includes: emission rate; release area and release duration; chemical
characteristics; stack parameters; and standard meteorological data. Receptor
input includes: desired averaging time; above-ground height of receptors; and
maximum distance between receptors.
The model automatically writes an output file; no graphical
representations are provided. The contents of the file are: input data;
plume centerline elevation, mole fraction, concentration, density, and
temperature at each downwind distance; sigma y and sigma z values at each
specified downwind distance; off-center line distances for two specified
concentrations at a user-specified receptor height at each specified downwind
distance; and concentration vs. time histories for finite-duration releases.
3-3
-------
3.4 HGSYSTEM(15)
HGSYSTEM is a PC-based system consisting of seven distinct models, each
of which represents a different type of release. 'The seven models are:
1) spillage of HF from a vessel (HFSPILL); 2) spreading/evaporation from a
liquid pool (EVAP); 3) flashing of HF (HFFLASH); 4) jet flow, near-field
dispersion of HF (HFPLUME); 5) jet flow, near-field dispersion of an ideal gas
(PLUME); 6) ground-level heavy gas dispersion (HEGADIS); and 7) elevated
passive dispersion (PGPLUME). In some cases, the models may be used together
in sequence. That is, the output from one model may be used as the input to
another model. Because several models may be involved with any particular
scenario, modeling possibilities can be complex.
HGSYSTEM contains scripts that guide the user through the process. The
model involving pool evaporation (number 2 above) requires the most input
concerning properties of a liquid release. HF modeling (models 3 and 4)
requires less input since the chemical information already exists in the model
itself. Modeling of a jet flow, near-field dispersion (number 5 above)
requires a description of the release and the ambient conditions. Finally,
modeling of elevated passive dispersion (number 7) requires input from a near-
field model, along with selected data on ambient conditions and dispersion.
For each of the seven models, detailed input is required on the thermal
properties of che gas, its initial concentration and areal extent, and a
complete description of ambient conditions and site characteristics. HEGADIS,
which is the heavy gas atmospheric dispersion model, should be implemented
after a source-term model has been specified. There are three choices: 1)
steady-state; 2) finite-duration; or 3) transient. For the jet model, a jet
may be elevated and inclined in the vertical; however, it is assumed that the
jet points "downwind."
For output PGPLUME generates tabular data for distances downwind of the
source, whereby the mole-concentration is reported in a vertical cross-section
of the plume that is normal to the transport direction. Being a steady-state
3-4
-------
or finite duration release model, no time-histories are reported. HEGADIS
uses two postprocessing algorithms to calculate output. The algorithm,
HSPOST, (used for steady-state releases) is used to determine concentrations
at any point (x,y,z). It also reports concentrations, geometrical parameters,
and temperature for points which are along the ground in line with the axis of
the plume. This postprocessor may also calculate tabular or plotted results
for releases of finite duration. HTPOST (used for transient releases)
averages the time-series concentrations described from the transient version
of HEGADIS. This postprocessor can output data at a. particular time, or the
user may specify a particular point and obtain a time-series of concentration.
3.5 SLAB(16)
SLAB is a PC-based dispersion model of denser-than-air releases. It can
be applied to four types of releases including: a ground level evaporating
pool, an elevated horizontal jet, a stack or elevated vertical jet, and an
instantaneous volume source. All sources except the evaporating pool may be
characterized as aerosols. The model can simulate multiple sets of
meteorological conditions in a single run, but it does not accept any type of
real time meteorological data. Data is input directly into the model from an
external file. Input data include source type, source properties, spill
properties, field properties, and standard meteorological parameters. There
is no chemical database, but some chemical properties are available in the
user's guide.
The model does not generate any type of graphical output, but rather
automatically sends the tabular results to a printer file. These results
include: input data; instantaneous spatially-average cloud parameters; time-
averaged cloud parameters; and time-averaged concentration values at the plume
centerline and at five off-centerline distances at four user-specified heights
and at the height of the plume.
3-5
-------
SECTION 4
MODEL INPUT
This section provides an overview of the input required by hazardous air
pollutant dispersion models. The overview includes descriptions of the
decision-making process and calculations required for determining model input.
*
Most of the input for the models was developed using techniques described in
the EPA Workbook'25.
This section is organized to lead the user from a physical description
of the release (called observable data here) to input parameters which are
calculated.for the various models. Section 4.1 describes the observable data
at the release site that are required to perform the input calculations.
Section 4.2 lists the data required to define a chemical for a model. It also
describes methods which can be used to approximate the data for a chemical
mixture. Section 4.3 details the methods used to determine the basic release
class of a given scenario. Section 4.4 discusses the differences between
continuous and instantaneous releases.
Section 4.5 presents a series of flow charts which illustrate the
approach used Co calculate various input for the models. The techniques used
depend on the release class of a given scenario. Sections 4.6 through 4.13
provide calculation methods for the following input:
• Choked/unchoked flow (4.6);
• Emission rate (4.7);
• Release temperature (4.8);
• Vapor fraction (4.9);
• Initial concentration (4.10);
• Density (4.11);
• Release diameter or area (4.12); and
• Release buoyancy (4.13).
4-1
-------
Sections 4.14 through 4.17 discuss other input considerations which
require little, if-any, calculation. This input is: release height, ground
temperature,.averaging time, and meteorological data. The final section,
4.18, briefly describes model output concerns.
When a scenario is defined, the first thing to do is determine the
release class that best represents the scenario. To determine the release
class, a series of calculations may be required. The determination of a
release class may be performed through a process of assumption and
calculation. For example, in determining whether a release is two-phase or
not, the assumption is made that it is single phase. Calculations of release
parameters are performed for the single-phase release class. The results of
the calculations are then checked for consistency. If the consistency checks
fail, the release must be a two-phase class. Since input parameters must be
calculated before the release class can be determined, this section is
organized by the calculated parameters (e.g. the emission,rate), with release
class sub-headings.
The order of the parameters presented in Sections 4.6 through 4.13 is
not necessarily consistent with the order of calculation. Each release class -
requires a specific order of calculation. If a prior calculation is required,
it will be noted in the section. For instance, the emission rate calculation
for the "two-phase gas release (unchoked)" in Section 4.7.2 requires the prior
calculation of a change in enthalpy (AH) and release density(prei) . These
parameters are listed at the beginning of Section 4.7.2, along with the
sections in which those calculations are detailed.
4.1 OBSERVABLE DATA
The initial step in preparing the input required for any air release
model is to gather the observable data. Observable data are the physical
descriptions of a release before and at the time of the release. As far as
input is concerned, there are two types of models: source-term and non-source-
term. • '
4-2
-------
Source-term models can estimate the conditions of the gas state of a
release entering the atmosphere. If the release is a liquid, the source-term
model calculates the rate of emission of the liquid from the container and the
rate of the gas and aerosol injection into the atmosphere. If the release is
a gas, the source-term model can calculate the injection of the gas into the
atmosphere, along with any aerosol that may form due to expansion cooling. As
•
a gas de-pressurizes it cools. If the cooling is sufficient, the gas may
condense into droplets. Besides emission rate, parameters such as
temperature, initial concentration, initial density, etc., are also calculated
by source-term models.
Models which cannot calculate release conditions based on storage
conditions are called non-source-term models. Models rely on the user to
supply the release condition information. The user, therefore, must perform
these calculations outside of the model. The model then uses a dispersion
model to predict impacts based on the user-calculated input.
Both source-term and non-source-term models require observable data to
perform the calculations. Thus, the only difference is whether the user or
•
the model carries out the.calculations of input concerning release conditions.
For both the non-source-term and source-term models, the following information
is required:
• Chemical;
• Storage temperature;
• Storage pressure;
• Hole size;
• Depth of liquid in container above hole;
• Maximum pool size; and
• Meteorolo'gical conditions.
THe ADAM, ALOHA, and HGSYSTEM (HF only) models contain source-term
models. Therefore, many input values are calculated internally by these
models.
4-3
-------
The calculations in this section are designed, for the most part, to
provide input to non-source-term models. However, some calculations may still
be required for some of the source-term models. For example, a model may
still require the emitted gas temperature for some release classes.
4.1.1 Vessel Shape and Dimensions
In the models described in this document, three general shapes of
vessels are allowed. These are a horizontal cylinder, a vertical cylinder,
and a sphere. The ADAM model, however, assumes only, vertical cylinders. If a
spherical vessel is involved, it can be conservatively modeled by assuming a
vertical cylinder with a diameter equal to the sphere diameter.
4.1.2 Hole Location and Orientation
Besides hole size, a source-term model requires information on the
height of the hole both with respect to the ground and with respect to the
vessel bottom. The height above ground determines the height at which the gas
escaping the vessel will enter the atmosphere. The height of the hole from
the vessel bottom determines the amount of liquid that can be released and the
pressure due to the liquid depth above the hole location.
A release may be from a pipe connected to a vessel. Some resistance to
flow may occur due to the condition and size of the pipe.
The vertical orientation of the hole will determine if a release is a
vertical, horizontal, or intermediate release. The azimuthal orientation of
the hole will determine the initial direction of a release. Generally, higher
concentrations can be expected downwind if a jet release is in the same
direction as che ambient wind. The larger the vector difference in a jet
velocity and the ambient wind velocity, the more mixing will occur. For
example, a jet escaping directly into the wind will create more turbulence
than a jet released in the same direction as the wind. This is especially
true if the wind speed is comparable to the jet exit speed.
4-4
-------
4.1.3 Diking
If the release is a. liquid, the liquid may spread as the release
continues. If there are dikes or terrain features in the area, the pool that
forms may be restricted to a maximum size. This maximum size would limit the
surface area of the spill available for evaporation into the atmosphere. Gen-
erally, the larger the surface area of the pool, the larger the total emission
rate. This parameter is also used in the Workbook calculation for a low
volatility release. In the Workbook, the total area of the pool is required.
Source-term models may take intb account the shape (circular, rectangular,
etc.) of the diked area.
4.1.4 Surface Description
For liquid pools heat is transferred from the surface on which the spill
occurred. At a minimum, the ground temperature is required for the
calculation of the heat transfer rate. A more complete calculation would
allow the specification of other parameters describing the surface. Examples
of such parameters are: thermal conductivity, specific heat, and porosity.
4.1.5 Solar Radiation
Another factor in determining heat transfer to a pool of liquid is solar
radiation. Direct insolation onto the pool surface can be a significant
source of energy. Rather than asking for the value directly, a location
(latitude and longitude) and time of day may be specified. The model then can
estimate the location of the sun with respect to the spill location, and
hence, a thermal radiation value. Sometimes a cloud transmittance factor is
also requested to account for attenuation of the sunlight by the clouds.
Although, strictly speaking, this parameter should be included in the
meteorological section, it is almost solely used in the source-term
calculation. The other meteorological parameters affect the transport and
dispersion of the emission in addition to the source-term calculation.
4-5
-------
4.2 CHEMICAL DATA REQUIREMENTS
Normally, data for some chemicals are supplied by the model. If a.
chemical is not already in the model's database, it must be added. The most
often required chemical data are:
• Molecular weight;
• Heat capacity;
• Boiling point;
• Heat of vaporization;
• Liquid density;
• Vapor pressure;
• Critical values (temperature, pressure, and volume); and
• Enthalpy.
Values are available for many pure substances in sources listing many of
the parameters describing a chemical. There are also techniques available for
estimating physical and chemical properties for uncommon pure substances.
Some of the sources are:
Perry's Chemical Engineers' Handbook, 6th ed,, edited by Robert H. Perry
and Don W. Green, McGraw-Hill, Inc., New York, New York, 1984 (data and
estimation techniques);
The Properties of Gases & Liquids, 4th ed., Reid, Prausnitz, and Poling,
McGraw-Hill, Inc., New York, New York, 1987 (data and estimation
techniques);
Data Compilation Tables of Properties of Pure Compounds, T.E. Daubert
and R.P. Danner, Design Institute for Physical Property Data, American
Institute of Chemical Engineers, New York, New York, 1985 (data); and
CRC Handbook of Chemistry and Physics, 71st ed. , edited by David R.
Lide, PhD, CRC Press, Inc. Boca Raton, Florida, 1990 (data).
Most of the available data are for pure chemicals. Often a chemical
mixture is the substance of concern. Models require a single set of chemical
data for the material released. A chemical mixture is treated as a pure
chemical by developing the parameters of a fictitious or "pseudo-chemical."
For a gas or liquid mixture, the parameters describing the gas or liquid p"hase
4-6
-------
of the pseudo-chemical are made up from using the parameters of the individual
components of the mixture and their molar or mass fractions. The first
pseudo-chemical parameter calculated is usually the molecular weight of the
mixture. The mixture molecular weight is calculated using either mole
fractions (fA) or mass fractions (w^ of the individual component i along with
each component's molecular weight (M^. The formula for calculating molecular
weight (M) of a mixture using the mole fraction is:
M = £ f ^
c
The formula for calculating molecular weight of a mixture using the mass
fraction is:
c
The relationship between the mass and molar fractions is:
In general, the parameters for a mixture (P) are calculated with the use
of one of the following equations:
The first equation applies if the parameter is in units of inverse moles
[or has no mole or mass units (such as temperature)]. The second equation
applies when the parameter is in units of inverse mass. Vapor pressure is,
however, a special case. The vapor pressure of a mixture is the sum of the
vapor pressures (in equilibrium with the liquid mixture) of the individual
4-7
-------
components. The boiling point is defined as the temperature when the mixture
vapor pressure is equal to 1 atmosphere.
The gas-phase mixture in equilibrium with the liquid-phase mixture
probably will not have the same proportions of components as the liquid-phase
mixture. The pseudo-chemical parameters for the two phases should be kept
separate. The liquid-phase parameters influence the emission rate but have
little or no effect on the dispersion of the gas phase after it leaves the
source.
For these pseudo-chemicals there are two methods of generating the
required thermodynamic data. In both methods, the data describing the liquid
mixture would represent the weighted average of the mixture, as described
above. For example, the boiling point of a mixture could be given as the
temperature at which the sum of the partial pressures of the constituents is
equal to 1 atmosphere.
The first approach for calculating thermodynamic data for the vapor
state is to use the weighted average scheme described above. When determining
the concentration of a specific species in the vapor mixture, the
model-calculated concentration needs to be modified by multiplying it by the
molar fraction (ft).
In the second method, the data describing the vapor state must be
specific to the chemical of interest. The vapor mixture in equilibrium with
the liquid mixture will be made up of many, if not all, of the components in
the liquid. The relative concentrations (partial pressures) of the vapor
mixtures will be different than those of the liquid mixture. The vapor
pressure data would represent the vapor pressure of only a single constituent
in equilibrium with the liquid mixture.
A pure substance may have varying parameters describing the substance.
An example is hydrogen fluoride. At high concentrations, the individual
molecules can form oligomers (polymers which contain two, three, or four
4-8
-------
monomers). The dynamic effect is to increase the molecular weight of the
release. It may be desirable to have different data sets for a compound such
as hydrogen fluoride. One set of parameters could be used for low .
concentration releases, the other for pure releases. From previous work(17),
there is sufficient data to create a table relating the apparent molecular
weight of hydrogen fluoride to the temperature and concentration of the
hydrogen fluoride gas. Such a table is included in the database for hydrogen
fluoride in Appendix B. As that table shows, in pure concentration (1,000,000
ppra) hydrogen fluoride at 20°C has an apparent molecular weight of 51.6.
Above 70 "C, however, hydrogen fluoride is present as a monomer at all
concentrations giving a molecular weight of 20.
4.3 RELEASE CLASS
Probably the most important part of determining the impacts from a
release is determining the manner in which the release occurs. The release
classes covered in this document are:
• Two-phase gas (choked or unchoked gas released as gas and liquid);
• Two-phase liquid (liquid released as gas and liquid);
• Single-phase gas (choked or unchoked gas released as gas);
• • Single-phase, high-volatility liquid (liquid released as gas); and
• Single-phase, low-volatility liquid (liquid evaporating as gas).
A release class is a categorization of a release scenario based on the
chemical's physical storage conditions (gas or liquid) and initial
vaporization (single- or two-phase) during the release from storage. The
release class and chemical determine the parameters required for a model to
carry out a simulation. Chemicals may be stored as a gas or as a liquid. A
chemical stored as a gas under pressure may enter the atmosphere in both
liquid and gaseous form. Liquid droplets may form due to cooling from
expansion after a release. Such a suspended phase mixture is called an
aerosol. A chemical with a boiling point below the ambient temperature that
is stored either under pressure at the ambient temperature or refrigerated to
a temperature below its boiling point may also form an aerosol upon release.
4-9
-------
If no aerosol forms when such a liquid is released, it would be considered a
high-volatility liquid. A chemical with a boiling point above the ambient
temperature that is released will most likely form a liquid pool, except when
the chemical is stored under high pressure (the high pressure release may emit
the liquid as "droplets). This pool will evaporate into the atmosphere.
Chemicals with lower volatility or vapor pressure will evaporate more slowly
than chemicals with a higher volatility, given that all other conditions are
the same.
The question of aerosol formation and fate is not yet clearly answered.
For screening purposes it is suggested for two phase releases that all the
liquid in the release be assumed to be entrained with the vapor phase as an
aerosol. As mentioned above, aerosols can form due to cooling. They can also
form when a liquid comes out at high pressure, by splashing or impacting on
surrounding objects, or comes from superheated storage.
The steps in determining the release class are given in Figure 4-1.
Before the determination can be made, the storage phase must be known. A flow
diagram (Figure 2-8) given in Section 2.4 which can be used to determine the
storage phase. As mentioned in the introduction to this section, determining
the release class using the flow diagram in Figure 4-1 requires some
calculations. These calculations are described in the sections referenced in
parentheses in Figure 4-1. The same calculations required to determine the
release class may also be used as model Input for the selected release class.
Flow charts outlining the required calculations for each release class are
presented with the discussion of each release class.
In Figure 4-1, the initial question is the storage state (gas or liquid
which can be determined from Figure 2-8).
For material stored in the gas phase, you must determine if there is
choked flow. The calculation for determining if there is choked flow is
described in. Section 4.6. A choked flow can limit the emission rate.
4-10
-------
Tref=Reference Temperature
Pref = Reference Pressure
Tc = Critical Temperature
vapor
fraction from
Appendix B flash
diagram <0?
Calculate the emission
temperature (4.S.2) and
vapor fraction of emission
(4.9.3)
Yes
Yes
Is the vapor mass
fraction <1 andX)?
the boiling point
below ambient
temperature?
Low volatility spill
(Fig. 4-5)
T
r
Calculate temperature at
choked flow. Call it Tref.
Choke pressure Pref.
(4.7.1)
' No
Calculate discharge
temperature. Call it Tref.
Ambient pressure Pref.
(4-8.2)
Two phase, release
(Fig. 4-2)
Figure 4-1. Determining Initial Release Class
4-11
-------
Next, a reference temperature (Tr.f) and reference pressure (pref),
indicative of the conditions at the hole must be determined. The method of
determining the reference temperature and pressure depends on whether the flow
is choked or not. If the flow is choked, then the pressure (p*) and
temperature (T*) at choke conditions must be determined. This calculation is
described in Section 4.7.1. The values of p* and T» could then be set to
reference values pref and Traf. If the flow is unchoked, the reference
pressure, pref, is equal to the ambient pressure (pa). The reference
temperature, Tref, is set to the unchoked release temperature (Trei) calculated
in Section 4.8.2.
Once the reference temperature -and reference pressure have been
determined, they.can be used to determine if the release is two-phase (i.e.,
whether any condensation occurs during the release). This is done by
performing two checks to determine if the release is single-phase. If both
checks prove negative, the release is two-phase. The first check is to
compare the reference temperature to the critical temperature of the chemical.
If Ttef > T0, the release is single phase. If not, the second check is be
performed. In this check, if the vapor pressure of the chemical (from the
data in Appendix B) at Tref is greater than pref, then the release is single-
phase. Otherwise, the release is two-phase.
For material stored in the liquid phase, a preliminary check may be
carried out as indicated in Figure 4-1. This preliminary check is a graphical
determination of whether the release may result in flashing (a two-phase
release). The equation used to determine the flash fraction diagrams in
Appendix B is based on the assumption that the temperature after the emitted
chemical is out of the container is equal to the boiling point temperature.
If there is still uncertainty as co whether a flash occurs after
referring to the flash diagram in Appendix B, a calculation for the flash
fraction must be carried out. The flash fraction (Frel) calculation is
described in Section 4.9.3. This calculation is the same whether the release
4-12
-------
is a pressurized or a refrigerated liquid. When a flash is occurring, the
release temperature (Ttel) is the bbiling point.
If no flash occurs (Frel - 0), then the boiling point of the chemical
determines whether the release is of a high or low volatility chemical. If
the boiling point is lower than the ambient temperature, the release should be
considered one of high volatility. If the boiling point is higher than the
ambient temperature, assume a low volatility release.
4.4 CONTINUOUS OR INSTANTANEOUS RELEASE CATEGORIES
For a source-term model, the words "instantaneous" and "continuous"
normally describe the manner in which the chemical gets out of the container.
For a non-source-term model, the words "instantaneous" and "continuous"
describe the manner in which the chemical enters the atmosphere. In some
models the emission rate (either from the container or into the atmosphere)
can be time varying.
All of the models allow the user to specify whether the release is
instantaneous or continuous. However, as stated above, these terms are
interpreted differently depending on whether a source-term model is being used
or not. If a specific model provides guidance on the determination of whether
a release is continuous or instantaneous, that guidance should be followed.
Some of the models also allow a finite duration release. A finite
duration release is one that only exists for a limited period of time. That
is, a finite duration release is neither instantaneous, nor does it go on
without ceasing. For example, if a relief valve opened and the entire
contents of the vessel escaped in 15 minutes, a finite duration release of 15
minutes would have occurred. If a model allows a finite duration release, the
actual time of duration (rather than specifying an instantaneous or continuous
release) should always be used.
4-13
-------
Most models do not take into account mitigation efforts during a release
(for example, if an automatic or human-activated shutoff valve is turned on
during a release). In this case, the release duration would be limited to the
time from the beginning of the release to the closing of the valve. If the
actual release time is significantly different from the model computed release
time, the release may be better simulated by assuming it is instantaneous or
by using a finite release duration model.
When a gas is released from a container, it goes from the container
immediately into the air. An aerosol may form, but it also is positioned in
the air. A liquid release, on the other hand, can actually be two releases.
One is the release from the container. If the liquid entirely flashes or all
•
the liquid is suspended as an aerosol, that is the only release. But, if the
liquid forms a pool on the ground, a second release would occur involving
release of the chemical from the liquid state to the vapor state and its
injection into the atmosphere.
4.4.1 Non-source-term Model
To decide whether the release should be considered continuous or instan-
taneous two factors need to be known: 1) the release duration, and 2) the
location of the nearest receptors or the maximum distance to the lowest
concentration of interest. To determine the category, if a concentration of
interest is specified, the release first needs to be simulated as a continuous
release. Then the travel time to the maximum distance of the concentration of
interest is compared to the actual release duration. If only-receptor
locations are of interest, their distance from the source can be substituted
for the maximum distance of the concentration of interest.
In practice, a continuous release initially should be assumed. The time
required for the material to reach a specific downwind distance is then
compared to the actual emission duration. If the emission duration is longer
than the time it took for the material to reach a downwind point, the release
4-14
-------
may be considered continuous. Otherwise, the release should be considered
instantaneous. The travel time (ttrav) to some site at a downwind distance (X)
from the release site can be approximated using this formula:
t = 2X
ttrav —
where: u -.ambient wind speed.
If the model allows a finite duration release, the release duration
should-also be compared to the desired averaging time. If the release
*
duration is less than the averaging time of concern, then a transient model
should be used to reflect average concentrations more accurately. For
example, if the release duration is 5 minutes and the averaging time of
concern is 15 minutes, it is important to use a finite duration model to
simulate the proper average concentrations.
4.4.2 Source-term Model
Source-term models normally use the words "continuous" and
"instantaneous" to describe how the material escapes from confinement. If the
model has guidance on determining whether a release is continuous or
instantaneous, it should be foil-owed. If such guidance is not available, the
following is provided.
For a source-term model, the general rule of thumb is if the chemical
escapes in less than one minute, it is assumed to be an instantaneous release.
If the release is of longer duration, another approach is used, which is
related to the emission rates from the two releases that a liquid goes
through. If the emission rate from the container is much faster than the rate
of evaporation from the formed liquid pool, the emission from the container
can be assumed to be instantaneous. In this case, a pool forms, and the
container emission just makes the pool deeper. The limiting emission for
entry of the gas phase into the atmosphere is the evaporation.
4-15
-------
If the emission from the container is such that the pool formed has an
evaporation rate that is the same as the rate of emission from the container,
the release should be considered continuous. In this case, if the emission
rate from the container increases, the size of the pool would increase.
4.5 RELEASE-CLASS-SPECIFIC CALCULATIONS
•
Each release class requires a series of calculations to determine the
initial emission conditions:
• Rate;
• Temperature;
• Vapor/liquid fraction;
• Concentration (air, water vapor, and chemical); and
• Density.
The method for determining each value, even the order of the
calculation, changes with the release class. In addition, a specific
calculation may be needed in determining the release class and used as model
input. This section describes the overall flow of calculating the input for
each release class. The following sections describe the input calculations in
more detail for each of the release classes. The Workbook(4), from which
these calculation techniques were taken, presents the calculations in detail
by release class.
The overall approach for determining model input is to first' use
Figure 4-1 to determine the release class. Second, use the appropriate flow
chart (Figures 4-2 through 4-5) for the release class-dependent calculations.
When the release-class-dependent calculations have been completed, the common
input requirements, presented in flow chart form in Figure 4-6, should be
determined.
It is assumed that, before the calculations listed in the flow charts
for the individual release classes are completed, the approach given in
Figure 4-1 has been used to determine the release class. Note that some of
4-16
-------
Calculate
vapor
fraction
(4.9.1)
Calculate
emission
temperature
(4.8.2)
Calculate
vapor fraction
(4.9.2)
Calculate density
(no air or water
vapor)
(4.11)
Calculate
emission rate
(4.7.2)
c
Start with storage condition
(Fig. 2-8)
ition I
Gas
.1.
t
Liquid by
pressure
1
Refrigerated
liquid
Check input and
complete
(Fig. 4-6)
Calculate
emission
temperature
(4.8.3)
Calculate
vapor fraction
(4.9.3)
Calculate density
(no air or water
vapor)
(4.11)
1
Calculate
emission
temperature
(4.8.4)
Calculate density
(no air or water
vapor)
(4.11)
Figure 4-2. Two-Phase Input
-------
Calculate
emission rate
(4.7.5)
Calculate emission
temperature
(4.8.5)
Set air and water
vapor fractions
at release
.(4.10.1)
Calculate density
(4.11)
Calculate
emission
temperature
(4.8.6)
Set air and water
vapor fractions
at release
(4.10.1)
Check input and
complete
(Fig. 4-6)
Figure 4-3. Gas to Gas Release Input
4-18
-------
Start
(Fig. 4-1)
Calculate
emission rate
(4.7.7)
Assume emission
temperature is equal
to boiling point
Assume air and water
vapor fractions are
zero
Calculate density
from gas formula
(4-11)
Check input and
complete
(Fig. 4-6)
Figure 4-4. High-Volatilicy Release Input (Not Two-Phase)
4-19
-------
Calculate pool
evaporation rate
(4.7.8)
Start
(Fig. 4-1)
Calculate
maximum
emission rate
(4.7.7)
Is pool
evaporation
greater than
Is maximum
pool size-
known?
Calculate pool
size from
maximum
emission rate
(4.7.8)
1
V
Set emission rate
to pool
evaporation rate
Calculate vapor
pressure of chemical
and set initial
concentration
(4.10.2)
Set initial water
vapor and air
concentrations to
same ratio as
ambient
(4-10)
Calculate density
from gas formula
(4.11)
Check input and
complete
(Fig. 4-6)
Figure 4-5. Low-Volatility Release Input
4-20
-------
Calculate
emission
dimension
(4.12)
Calculate dense-gas
criteria
(4.13.2)
Calculate dense-
criteria
(4.13.1)
Stop! Use a
non-dense gas
model
Continuous or
Instantaneous
emission?
Determine
ground surface
temperature
(4.15)
Select release
height
(4.14)
select averaging
lime
(4.16)
— >
Determine
meteorology
(4.17)
Done
Figure 4-6. Check Input and Complete
-------
the calculations required for a specific release cLass may have already been
performed when using Figure 4-1.
4.5.1 Two-phase Releases (Gas and Liquid)
A flow diagram showing the steps required to calculate the input for the
two-phase releases is shown in Figure 4-2.
The gas release becomes two-phase through condensation of the material
due to cooling by expansion. The flow may be choked, which limits the
emission rate and the amount of condensation that can occur.
Liquids with boiling points lower than the ambient temperature are
stored either under pressure or are refrigerated to a temperature that is
lower than their individual boiling points. In many cases, the liquid is
stored in a container in which there is a combination of pressure and
refrigeration. In those cases, it is suggested that the calculations be
performed for the pressurized case. The pressurized case may have
aerosolization occurring because of jetting at the exit. Refrigeration-only
releases would not tend to have aerosolization. The presence of aerosols tend
to keep the emitted cloud cooler longer than if there are no aerosol. The
cloud should be at boiling point until all the aerosols have evaporated.
4.5.2 Single-phase Gas
A flow diagram showing the steps required to calculate model input to
simulate a non-condensing gas release is shown in Figure 4-3. The release may
exhibit choked flow. As the flow diagram shows, different paths of cal-
culation must be followed for choked flow and unchoked flow. Although the
same quantities are calculated for both the choked and unchoked flows, the
methods used for the calculations are different.
4-22
-------
4.5.3 Single-phase. High Volatility Liquid
Figure 4-4 shows a flow diagram for calculating the input for a release
of a highly volatile liquid. This is the simplest of the flow charts, with no
decision branches. The released material is assumed to be in a pure vapor
state. The density can be calculated using the perfect gas law.
4.5.4 Single -phase. Low Volatility Liquid
Figure 4-5 shows the flow diagram for calculating input for a
low-volatility liquid release. In this case, the release rate into the
atmosphere is controlled by the pool size. The pool size, in turn, may be
limited by the rate at which the liquid exits the container or by containment
from a dike or bund.
4.6 DETERMINATION OF CHOKED FLOW FOR GAS RELEASES
Gases may be stored at high enough pressures that, if the gas is
escaping through a hole rather than the container failing, the gas escape
_ •
velocity may reach the speed of sound. However, the escape velocity can not
be greater than the speed of sound no matter how high the pressure under which
the material is stored. The pressure at the exit point of the gas (p*) is
calculated by:
where: 7 - the ratio of the specific heat at constant pressure to that at
constant volume; and
ps - storage pressure (Pa) .
If p* > pa, then the flow should be considered choked. If p* < pa> then the
flow is not choked. For the calculations in this document, the units of p*
4-23
-------
should be Pascals (Pa). The standard atmosphere at sea level has a pressure
of 101325 Pa.
Whether a flow is choked or not is important for determining which sets
of equations to use for the calculation of emission rate and temperature. In
the following sections, the gas release calculations will be divided into
those for critical or choked flow and those for subcritical, or unchoked,
flow.
In the case of gases from stacks, the calculations given here are not
required. The determination of whether a flow is choked or not assumes that a
storage pressure is known, but for a stack^release, this may not be the case.
Stack release volume rates may be governed more by a release temperature or
mechanical device (such as exhaust fans). In those cases, a storage pressure
is not specified. Normally, the emission rate is known through measurement,
assumed based on design limitations, or estimated from emission factors.
Temperature, if not known, can be estimated as the ambient temperature.
4.7 EMISSION RATE
„ •
The emission rate input to the model is often known when calculating
impacts from a release that occurred in the past. The amount of mass that was
lost from the system and the duration of the release are normally known or
calculable. In planning or in a real-time release, the emission rate normally
requires a calculation using the design criteria of the container involved in
the release.
The emission rate is not always calculated before the other parameters
•
discussed in this section. The flow charts of input calculations given for
the various release classes can be used as a guide Co the proper order of
calculation. Where a prior calculation is required, i-t is noted along with
the section in which to find that calculation.
4-24
-------
4.7.1 Two-phase Gas Release (Choked)
Prior calculations required:-
p* - critical pressure (Pa) (4.6).
Flow chart reference: Figure 4-2
A series of calculations is required to determine the conditions at
choked flow. The parameters describing the conditions can then be used to
calculate the emission rate.
The first of these parameters is the temperature (T.) corresponding to
the critical pressure. It can be calculated from:
p. = !Ol325Pa
where: M - molecular weight (kg/kmol);
X - heat of vaporization at Tb (J/kg);
R - gas constant - 8314 J/kmol °K;
Tb - normal boiling point (°K); and
T. - temperature at choked conditions (°K).
The vapor fraction at choked flow conditions (F*) can be estimated from:
T
F = 1 + _
' MA.
MCp Id -i - R lid ^S
P IT.J \P,
where: Cp = gas heat capacity at Ts (J/kg °K); and
Ts = storage temperature (°K).
The change in enthalpy (AH*) from storage conditions to the choked flow
conditions is given by:
4-25
-------
AH. = Cp(Ts - T.) + A(l - F.)
The last choked flow parameter required before the emission rate can be
calculated is the density at choked flow conditions (p»). It is given by:
P. =
-1
- liquid density at Tb (kg/m3).
With the choked flow parameters, the emission rate can be calculated as;
:•= A0p,
2 (0.85)
f AH' )
(D
p / /
i
2
where: E -
f -
emission rate (kg/s);
area of hole (m2) ;
0.0045, estimated (dimensionless);
pipe length (m); and
pipe diameter (m).
4.7.2 Two-phase Gas Release (Unchoked)
Prior calculations required:
AH
change in enthalpy between storage and release conditions
(J/kg) (4.9.2); and
density at release temperature (kg/m3) (4.11).
The intermediate calculations required for the emission rate of an
unchoked gas flow to be calculated are very similar to the calculations
required for the choked flow. In the case of choked flow, the calculations
4-26
-------
are carried out at the choked conditions. For the unchoked case, the calcu-
lations are carried out at the release temperature and ambient pressure.
The emission rate (E) is calculated from:
i
E = AoP
oHrel
2(0.85)
AH
\ •
where: E - emission rate (kg/s);
*
A0 - area of hole (m2);
f - 0.0045, estimated (dimensionless);
Lp = pipe length (m); and
Dp - pipe diameter (m).
4.7.3 Two-phase Pressurized Liquid
Prior calculations required: None.
Flow chart reference: Figure 4-2
The emission rate for a pressurized liquid can take into account a
length of pipe through which a release may occur. If the pipe length is zero
(the emission is directly from a tank), the emission rate will become that
calculated from the standard orifice equation for incompressible flow. If the
pipe length is greater than the length of pipe required to establish an
equilibrium flow, a different equation (presented below in this section) is
required. The longer a pipe the more the resistance to the flow of the
chemical is encountered. By ignoring the pipe length the emission rate will
be maximized.
The expression for the emission rate from a system with a pipe shorter
than the pipe length required for equilibrium flow (Lp/Le < 1) is:
where: E - emission rate (kg/s);
4-27
-------
A0 - area of hole (m2) ;
M - molecular weight (kg/kmol) ;
X - heat of vaporization at Tb (J/kg) ;
R. - gas constant - 8314 J/kmol °K;
2(ps -P.)P! C
Ts - storage temperature (°K);
ps - storage pressure (Pa) ;
Cpl— liquid heat capacity at Ts ;
pa - ambient pressure (Pa) ;
PL - liquid density at Tb;
Tb - normal boiling point (°K);
C - discharge coefficient, 0.6 (dimensionless);
Lp - pipe length (m) ; and
Le - pipe length required for equilibrium flow (m) , assumed 0.1 m.
If Lp/Le > 1, the expression becomes:
where: F - a pipe friction factor given by
F2 = _
.2 - 1
I DP ;
where: f - 0.0015, estimated (dimensionless); and
D_ - pipe diameter (m).
4-28
-------
4.7.4 Two-phase Refrigerated Liquid
Prior calculations required: None.
Flow chart reference: Figure 4-2
The emission rate for a refrigerated liquid can be calculated from:
i
E = A,
2C2(ps -
RT
where: E - emission rate (kg/s);
A0 - area of hole (m2);
M - molecular weight (kg/kmol);
A - heat of vaporization at Tb (J/kg);
R - gas constant - 8314 J/kmol "K;
•
Ts — storage temperature (°K);
ps - storage pressure (Pa);
Cpl = liquid heat capacity at Ts;
Pi - liquid density at Tb;
Tb - normal boiling point (8K);
C - discharge coefficient, 0.6 (dimensionless);
psv = the vapor pressure of the chemical at Ts (Pa) ;
psv = l01325Pa exp!
AM/ i
R T»,
and
a pipe friction factor given by
r»2 _ 1
where: f - 0.0015, estimated (dimensionless);
Lp = pipe length (m); and
D_ - pipe diameter (m).
4-29
-------
4.7.5 Single-phase Gas Release (Choked)
Prior calculations required: None.
Flow chart reference: Figure 4-3
The estimate for the emission rate can be calculated from:
E = CA
PsPsY
where: E - emission rate (kg/s);
A0 - area of hole (m2) ;
ps - storage pressure (Pa);
ps - storage density (kg/ra3);
C - discharge coefficient, 0.75 (dimensionless); and
7 - the ratio of the specific heat at constant pressure to that at
constant volume.
4.7.6 Single-phase Gas Release (Unchoked)
Prior calculations required: None.
Flow chart reference: Figure 4-3
The emission rate estimate for unchoked gas flow that is not condensing
can be calculated .from:
E = KYA0[2Ps(ps -Pa)]
where: E = emission race (kg/s);
A0 - area of hole(m2);
ps - storage density (kg/m3);
ps = storage pressure (Pa) ;
pa » ambient pressure (Pa);
4-30
-------
and
y = 1 - [2s - E*| (0.41 + 0.35 p4)
PsY
A0 / AI; and
C = 0.62;
/32 - A0 /
A! - cross -sectional area of gas container in direction of flow.
4 . 7 '. 7 Single-phase Liquid Release (High Volatility)
Prior calculations required: None.
Flow chart reference: Figures 4-4 and 4-5
A high volatility liquid can have a range of release rates of its gas
phase into the air. The largest injection rate would occur if the liquid
vaporizes immediately upon release. However, if the liquid is refrigerated, a
pool of the liquid may form. In that case the vapor emission would be from a
combination of the vaporization of the liquid occurring straight from the hole
and evaporation from the boiling liquid pool. Since the calculations in this
document and the Workbook are intended to be done using only a hand- calculator
and the purpose of this document is to provide screening or conservative
calculations for input to models, the most conservative case, which is
immediate vaporization of the liquid, is assumed.
Two steps are required to calculate the emission rate for a high
volatility liquid. The first part of the calculation is to estimate the
pressure at the hole from which the liquid is being released. This is done
using the following formulas:
where: ph - pressure at the hole (Pa);
4-31
-------
ph = max(pa/ps) +
pa - ambient pressure (Pa);
ps - storage density (kg/m3);
g - acceleration due to gravity (9.806 m/s2) ; and
H! - vertical distance between hole and the top of liquid level (m)
ps - the vapor pressure of the chemical at Ts (Pa);
ps = 101325Pa exp
M - molecular weight (kg/kmol);
A - heat of vaporization at Tb (J/kg);
R - gas constant - 8314 J/kmol °K;
Ts - storage temperature (°K); and
Tb - normal boiling point (°K).
With the value of the pressure at the hole known, the emission rate can
be estimated by:
i
"2
E -KA^p,^ -p.)]
where: A.Q — area of hole (m2) ;
K =
and C — 0.65 (dimensionless).
' 0Z - A0 / AI; and
Ax — cross-sectional area of gas container in direction of flow.
4-32
-------
4.7.8 Single-phase Liquid Release (Low Volatility)
Prior calculations required:
E - emission rate of liquid from hole (kg/s) (4.7.7); and
Trel - temperature of the liquid after release (°K).
Flow chart reference: Figure 4-5
The calculation of the emission rate for the low volatility case begins
just the same as the high volatility case, by calculating the liquid release
rate from the container. In the high volatility case, it is assumed that the
liquid vaporizes as it comes out. However, for the low volatility case,
another emission rate from the evaporating pool (Epool) is calculated and
compared to the container emission rate.
The emission rate from a pool can be estimated as:
Epool = 6.94xlO"7(l + 0.0043(T2 - 273 .15) *2)Ur0-75ApM—
Pvh
where: Epool - pool emission rate (kg/s);
Ur - ambient wind speed at 10 m altitude (m/s);
Ap - pool size (m2) ;
M. = molecular weight (kg/kmol) ;
T M T f\y T f0V^i '
12 = J-rel or J-a (• K' •
Ta • - ambient temperature (°K);
[T2-273.15]* - 0 if [T2-273.15] < 0
- [T2-273.15] if [T2-273.15] > 0;
pvh = vapor pressure of hydrazine at T2 (Pa)
pvh = exp
76.8580 - 72j,5'2 - 8.22ln(T2) + 0.0061557T2
T2
4-33
-------
pv - vapor pressure of chemical at T2 (Pa)
pv = 10l325Pa exp
A - heat of vaporization at Tb (J/kg);
Tb - normal boiling point (°K); and
R - gas constant - 8314 J/kmol °K.
Note that T2 can be either of two values. The value of Epooi should be
calculated fpr each value of T2. The higher Epool should be used in the
subsequent calculations. Normally the higher value of T2 will provide the
highest pool emission rate. However, this may not always be the case since
the vapor pressure of hydrazine is used as a scaling factor and enters as an
inverse.
If the value of Ep0ol < E, then the pool's evaporation determines the
rate at which material enters the atmosphere. In that case, Epool and Ap
should be used in calculating the emission rate and source size.
If Epool > E, then the container discharge determines the rate at which
the chemical enters the air. As the chemical is released, a pool is formed
that is smaller than Ap. The evaporation from the formed pool is equal to E.
A new pool size should be calculated using the equation for Epool but setting
the value of Epool equal to E. This new pool size should be used as the source
size.
4.8 RELEASE TEMPERATURE
Source-term models require the storage temperature of the chemical just
before release. The source-term model calculates Che change in temperature
after the release. Non-source-term models require the emission temperature
after depressurization has occurred. An external calculation is required for
determining the depressurized temperature.
4-34
-------
If the material release is colder than the ambient air, its density may
be increased enough that it acts as a dense gas, even if its molecular weight
is less than air's molecular weight. If the material released is warmer than
the ambient air, its density may be less than that of the ambient air and it
will act as a buoyant release. As the release moves downwind, entrainment of
the ambient air and contact of the plume with a surface can change the plume
temperature. Entrainment with ambient air makes the plume temperature
approach that of the ambient air. Contact with a surface makes the plume
temperature approach that of the surface.
There are many effects that temperature can have on a release. For
example, in an actual release, a pressurized gas was initially escaping
through a small hole. The cooling due to expansion was so great that not only
did water vapor in the air condense, it froze. The hole became clogged with
ice, reducing the flow rate. As the flow rate decreased, the ice melted.
Some of the liquid water reacted with the escaping material which, in this
case, formed an acid. The acid dissolved some of the container wall which
made the hole larger. The release rate increased, the cooling increased, and
a cycle of large emissions and low emissions began. This is an extreme case,
and probably no model will ever take into account such detailed effects.
However, it is important to realize some of the effects that temperature can
have, not only on the dispersion, but on the source itself.
4.8.1 Two-phase Gas Release (Choked)
Prior calculations required:
Frel - flash fraction of chemical of interest (dimensionless)
(4.9.1) (as a check);
T,, - temperature at choked conditions (°K) (4.7.1); and
F* - mass vapor fraction at choked conditions (dimensionless)
(4.7.1).
Flow chart reference: Figure 4-2
4-35
-------
Note that in this case, a two-step solution may be required. The
release temperature is calculated here using the value of Frel calculated in
Section 4.9.1. The first time calculating release temperature assume that
Frei < 1. If the subsequent calculation in Section 4.9.1 indicates that Frel
is equal to 1, recalculate the release temperature.
If Frel < 1, the release temperature can be estimated using the
Glausius-Clapeyron equation:
Pa = 101325Pa _„, B , ^
T"l.
where: Trel - release temperature (°K);
M - molecular weight (kg/kmol);
A - heat of vaporization at Tb (J/kg);
R - gas constant - 8314 J/kmol °K;
Tb - normal boiling point (°K); and
pa - ambient pressure (Pa).
If Fral a 1, it should be set to 1 for later calculations, and the
following equation should be used for the release temperature:
~ *<1 - F.)
where: A =» heat of vaporization at Tb (J/kg); and
Cpl- liquid heat capacity at Ts (J/kg °K).
Note that the value of Cpl is to be at Ts. The data in Appendix B is
for Cpl at a fixed temperature. Using the value of Cpl in the Appendix rather
than correcting for temperature, considering other uncertainties in the
screening approach, will not add significant error to the results.
4-36
-------
4.8.2 Two-phase Gas Release CUnchoked')
Prior calculations required: None..
Flow chart reference: Figure 4-2
The release temperature can be estimated using the Glausius-Clapeyron
equation:
pa = !Ol325Pa exp
-K-\lb J-rel,
where: Trel - release temperature (°K);
M - molecular weight (kg/kmol);
A - heat of vaporization at Tb (J/kg);
R - gas constant - 8314 J/kmol °K;
Tb - normal boiling point (°K); and
pa - ambient pressure (Pa).
4.8.3 Two-phase Pressurized Liquid
Prior calculations required: None.
Flow chart reference: Figure 4-2
For this release class, the same calculation is used for release
temperature as the calculation given for "Two-Phase Gas Release (Unchoked)."
See Section 4.8.2.
4.8.4 Two-phase Refrigerated Liquid
Prior calculations required: None.
Flow chart reference: Figure 4-2
4-37
-------
For this release class, the same calculation is used for release
temperature as the calculation given for "Two-Phase Gas Release (Unchoked)."
See Section 4.8.2.
4.8.5 Single-phase Gas Release (Choked)
Prior calculations required: None.
Flow chart reference: Figure 4-3
- The temperature is calculated based on the assumption that the change
occurs in two steps. The first step is when the release goes to choked flow
conditions (adiabatic and reversible temperature change). The second occurs
between the choked flow conditions and ambient pressure conditions (adiabatic
but not reversible). The equation to use in estimating the discharge
temperature is:
Trei =Ts[l - 0.85(JLI^)]
where: Trel - release temperature (°K);
Ts - storage temperature (°K); and
7 «» the ratio of the specific heat at constant pressure to that at
constant volume.
4.8.6 Single-phase Gas Release (Unchoked)
Prior calculations required:
E - emission rate (kg/s) (4.7.2).
Flow chart reference: Figure 4-3
4-38
-------
The calculation of the release temperature for the unchoked case depends
on the rate of emission, which is less than that in the choked flow case. The
equation to use is:
_ 2Tg
Trel " [i + J(i + 4aT7T]
where :
a . 1
and E - emission rate (kg/s);
A0 — area of hole(mz);
pa - ambient pressure (Pa);
M - molecular weight (kg/kmol);
R - gas constant - 8314 J/kmol °K; and
Cp - gas heat capacity at Ts (J/kg °K) .
4.8.7 Single-phase Liquid Release (High Volatility)
Prior calculations required: None.
Flow chart reference: Figure 4-4
This release class is based on the assumption that, as soon as the
liquid leaves the container, it vaporizes. The release temperature can be
assumed to be the normal boiling point of the chemical.
4.8.8 Single-phase Liquid Release (Low Volatility)
Prior calculations required: None.
Flow chart reference: Figure 4-5
4-39
-------
This release class is based on the assumption that the spilled material
evaporates from a pool. The normal boiling point of the chemical is higher
than the ambient temperature. The release temperature into the air can be
assumed to be the same as the ambient temperature or the storage temperature,
whichever is lower. This enhances any heavier-than-air effects on the
dispersion.
•
In reality, the temperature from an evaporating pool may be less than
the ambient temperature because of the cooling effect due to evaporation.
Such a temperature calculation requires more than a simple formula; it
requires a complete model for an evaporating liquid. Many models assume that
the temperature is the same as the ambient temperature in calculating the
emission rate. This is a conservative assumption, since higher temperatures
lead to higher emission predictions. In summary, for a conservative approach,
assume that the liquid is evaporating due to a high surface temperature but
entering the atmosphere at a low temperature (the lower of ambient or pool
temperature).
Low volatility liquid releases should always be continuous or of finite
duration. Evaporation is the primary method in which the material enters the
air., Unless the release is very small, evaporation may take a while. If a
low volatility release is small enough to evaporate quickly into the
atmosphere, it is probably small enough to ignore.
4.9 VAPOR FRACTION
The vapor fraction of a release is always a required calculation for a
two-phase release class. For single-phase releases, it is assumed that all
the material enters the air in the vapor phase because: 1) the material is
stored in the vapor state and cooling due to expansion is insufficient to
cause condensation; 2) the release is of a high volatility liquid that
vaporizes instantly upon release to the environment; or 3) the release is from
a low volatility liquid emitted to the air through evaporation.
4-40
-------
Since the two-phase releases are the only classes exhibiting a mixture
of liquid and vapor fractions , they are the only ones for which calculations
are presented here. For all other cases, the vapor fraction of the release
(Fral) should be set to 1; that is, only vapor is assumed to enter the
atmosphere .
4.9.1 Two-phase Gas Release (Choked)
Prior calculations required:
F* - vapor fraction at choked flow conditions (dimensionless)
(4.7.1);
*
T. - temperature at choked flow conditions (8K) (4.7.1); and
Trel - release temperature (°K) (4.8.1).
Flow chart reference: Figure 4-2
The vapor fraction after decompression (Frel) is given, in this case,
as:
where: Cpi= liquid heat capacity at Ts (J/kg °K) ; and
A - heat of vaporization at Tb (J/kg) .
If Frei > !. set it equal to 1 and correct the release temperature (Trel)
calculation in Section 4.8.1.
4.9.2 Two-phase Gas Release (Unchoked)
Prior calculations required:
Trel - release temperature (°K) (4.8.2).
4-41
-------
Flow chart reference: Figure 4-2
The vapor fraction after decompression (Frel) for unchoked flow is
estimated from:
MrMC>.lrt^iJ-R1Itl:,
where: M = molecular weight (kg/kmol);
A - heat of vaporization at Tb (J/kg);
R - gas constant - 8314 J/kmol °K;
Ts = storage temperature (°K);
Cp - gas heat capacity at Ts (J/kg °K);
ps — storage pressure (Pa); and
pa = ambient pressure (Pa).
This value is also used to calculate the change in enthalpy from storage
conditions to ambient conditions (AH). The value for the change in enthalpy
is given by:
AH = Cpl(Ts -Trel) + A.(l - Frel)
4.9.3 Two-.phase_Pr_es_surized Liquid
Prior calculations required:
Trel - release temperature (°K) (4.8.2).
Flow chart reference: Figure 4-2
4-42
-------
The vapor fraction after release (Frel) is given as:
F _ Cpl(Ts -Trel)
rel A.
*
where: A - heat of vaporization at Tb (J/kg);
Ts = storage temperature (°K); and
Cpl= liquid heat capacity at Ts (J/kg °K) .
4.9.4 Two-phase Refrigerated Liquid
Prior calculations required: None.
Flow chart reference: Figure 4-2
This release class uses the same calculation for vapor fraction as the
calculation given for "Two-Phase Pressurized Liquid" in Section 4.9.3.
4.10 INITIAL CONCENTRATION
The initial concentration sets an upper limit for the maximum con-
centration that may be seen downwind since the maximum concentration
(considering all altitudes) cannot increase with downwind distance. If the
initial concentration is low enough, it may also mean that the emission need
not be treated as a denser-than-air release. The concentrations of emitted
chemical, air, and water vapor are used in determining the density of the
release. The concentrations are defined as:
fa = molar fraction of air;
fw - molar fraction of water vapor; and
f, - molar fraction of chemical of interest.
4-43
-------
The only release classes where the values of fa, fw, and f± need to be
calculated are the single-phase gas release and the low volatility liquid
release. In all the other classes, the release is near the boiling point of
the chemical. Either a gas cools to near its boiling point and condensation
occurs or a liquid is vaporizing or flashing. For these near-boiling-point
releases, the conservative assumption is that f A is 1 and fa and fw are both
zero. This, in fact, is the implicit assumption in most models.
In single phase gas or low volatility liquid releases the chemical of
interest is in the gas phase and, in general, could be made up of multiple
species. If the chemical of interest is a mixture, its data description
should be created as described in Section 4.2. Knowing any two of the molar
concentrations of the three constituents determines the third because:
f.+fw+fi = 1
Measurement of the individual constituents would normally be required
for determining the molar fractions. The water vapor fraction is the easiest
to estimate or measure. If the temperature (T) of the release is known (in
Kelvin) and it is less than about 370 °K, the saturated molar fraction of the
water vapor (es) at 1 atmosphere of pressure is estimated as:
Iog10es = 6 .3994--
The value of es gives the maximum molar fraction of water vapor. At 298 °K,
es is about 0.03. For releases near moderate atmospheric temperatures, the
molar fraction of water vapor is small. However, in energy content
calculations, the amount of water vapor present can be significant because of
the greater neat capacity of water vapor versus dry air. If the relative
humidity is also known, the molar fraction of water vapor can be calculated
as:
f = { RH } e
LW ^ 10CT s
where: RH - relative humidity in percent. .
4-44
-------
It should be noted that these equations for es and fw are only strictly
valid for a water vapor and air mixture. Adding a chemical to the mix can
introduce error to the estimate. The amount of error and the direction of the
error would depend on the chemical. However, for input screening purposes,
these equations are adequate.
In reality, if a release has a jet phase where the emission is expand-
ing, some air will be entrained. The calculation to estimate the amount of
ambient air entrained is beyond the scope of this data screening study. When
there is uncertainty, the assumption should be made to make fi as close to 1
as possible.
4.10.1 Single-phase Gas Release
Prior calculations required: None.
. Flow chart reference: Figure 4-3
If the release is from a vessel containing only the chemical of inter-
est, the release should be treated as pure. This means that the value of fi
would be 1 and the other fractions would be zero. In most pressurized con-
tainers, this would be the case. The special case is that of a stack release.
A stack release is the most likely release to have all three consti-
tuents present. If a liquid water cloud is visibly being emitted from the
stack, then fw can be assumed to be equal to es. When water is used as a
scrubbing media, the emission is often water saturated at the release point.
The value for fa or fA must be determined from measurement or deduced from
information about the release. An example of such a deduction is given below.
.The conservative assumption is that fa is equal to zero; i.e. that the stack
release is of chemical undiluted by air.
An example of a deduction from release information would be the case
where a release of a known amount of chemical occurred over a known amount of
4-45
-------
time inside a building. The ventilation system would carry the spilled
material out to a vent. If the volume flow rate of the ventilation system
(V^) (from design documents) and the volume emission rate of the chemical of
interest (Vr) are known, then the value of fi could be estimated from:
f - Vri
1 " V~
rv
The value of fw would be assumed to be consistent with the humidity in the
vessel. In other words, the values of fw and fa would be prorated over
(1 - fi) at the same relative proportions that were in the air within the
building. If the molar fractions of water vapor and air inside the building
were fw' and fa' , respectively, the values for fw and fa would be calculated
by:
fw = fi(l - f^
and
f. - fad - fi)
4.10.2 Single-phase Liquid Release (Low Volatility)
Prior calculations required:
Trel - release temperature (°K) (4.8.8).
Flow chart reference: Figure 4-5
This release class is for a pool of evaporating liquid. The chemical
has a boiling point higher than the ambient temperature. The initial molar
k-46
-------
concentration of the chemical that is directly over a pool of evaporating
liquid can be estimated by:
f = *Z
1 Pa
where: pv - vapor pressure of chemical at Trel (Pa); and
pa - ambient pressure (Pa).
For pools of liquid that are not boiling, the value of fi is less than 1. The
rest of the emission can be assumed to include air and water vapor in the same
proportions as those found in the ambient air, unless some process is going on
that alters these proportions. For example, if a water spray were in use
during a spill, the value of fw could be assumed to be equal to es, the
saturation mole fraction of water.
4.11 DENSITY
Prior calculations required:
Trel - • release temperature (°K) (4.8);
Frel - initial mass vapor"fraction of chemical of interest
(dimensionless) (4.9);
fx - initial molar fraction of chemical of interest (dimen-
sionless) (4.10);
fa - initial molar fraction of air (dimensionless) (4.10); and
fw " initial molar fraction of water vapor (dimensionless)
(4.10).
Flow chart reference: Figure 4-6
Among the models used in this study, the DEGADIS model requires the user
to input the release density. The specific use of this value in DEGADIS is
described below. Cloud density is also used in determining whether a release
•
4-47
-------
is to be treated as a dense gas, as described in Section 4.13. The molar
fractions of each of the constituents respectively are fa, fw, and fA for air,
water vapor, and chemical of interest. Descriptions and relationships of
these molar fractions are given in Section 4.10. In calculating the density
of the initial release, the partitioning of the cloud into air, water vapor,
and the chemical of interest should be taken into account . Also, the
chemical of interest should be allowed to be in both liquid and gas phases.
The chemical of interest may be in the form of an aerosol (liquid droplets and
vapor phase chemical). The initial partitioning of liquid and gas phases of
the chemical of interest is determined by calculating the flash fraction,
described in Section 4.9. The fraction of a two-phase release in the gas
state is given as Frel. The density for each constituent is required to
calculate the overall density of the release. The density of the chemical of
interest can be determined by:
PI ;
where the density of the gas phase can be calculated assuming it is a perfect
gas.
The density would be given by:
p = P*Mi
Hg RTrei
where: pa - ambient pressure (Pa); and
R - gas constant - 8314 J/kmol °K.
The density of the liquid phase for specific chemicals is available in
Appendix B.
4-48
-------
The densities of the air and the water vapor in the cloud can be
determined by also assuming a perfect gas. These densities (kg/m3) are:
PaMa
O =
Ka RT
w RTe
where: Ma - molecular weight of air (kg/kmol) ; and
M,, - molecular weight of water (kg/kmol) .
Once the individual 'component densities have been determined, the
overall density (prei) of the released cloud can be determined by calculating:
. MtPw
where: MT - mean molecular weight of all material released (air, water vapor,
and chemical of interest) (kg/kmol) and given by the formula:
Note that if fa and fw are both zero (a release of pure chemical) , then
fi is equal to 1 . In that case, prel is equal to pi.
The DEGADIS model requires density triplets when aerosol is present in
the release. Each triplet consists of:
• Mole fraction of chemical of interest (f^ ;
• Mass concentration of chemical of interest (pi) ; and
• Mass density of the mixture (pmix) .
In the DEGADIS model, the values for fa and fw are based on the relative
proportion of these constituents in the ambient air. • The -assumption is made
4-49
-------
that all air and water vapor come directly from the atmosphere and that none
come from the chemical source. The DEGADIS model would prefer a schedule of
triplets for various air and chemical mixtures. Since the temperature, liquid
content, vapor content, and chemical concentrations vary as air is entrained
into the release, it is not a trivial calculation to create such a schedule.
A simpler approach is to provide the model with only two triplets in the
schedule. The first triplet would be for cases where only air is present in
the release (fa is 1 and fx and fw are both.zero). This triplet would be:
0.0, 0.0, pa. Because there is no contaminant in this triplet, the pa entered
here should be evaluated at the ambient temperature Ta, 'rather than the
release temperature Te. . The second triplet would be for cases where there is
no air present in the release (fi is 1 and fa and fw are both zero) . The
second triplet would then become: 1.0, prei, prei- DEGADIS will linearly
interpolate the air contaminant mixture for all chemical concentrations.
4.12 RELEASE DIAMETER OR AREA
Source-term models normally require the hole size, stack diameter, or
pool area of a release. Non-source-term models require a source size that
represents the size of the release after any expansion has taken place. For
the source-term model, the input requirement is fulfilled from the list of
observables. For the non-source term model, some calculations may be
required.
In the case of a liquid pool evaporating, the release area is simply the
area of the pool. This is because the emission velocity into the atmosphere
is low and there is no expansion of the vapor coming from the liquid surface.
In the case of a jet release, some expansion can take place, and a calculation
should be done co provide an estimate of the expansion effect.
There are two methods that can be used to estimate the post-expansion
dimension of a release. One can be used in cases with low release momentum.
The other can be used in the case of-high momentum releases such as jets.
4-50
-------
If the release is a choked gas the diameter of the hole used in non-
source term models should be changed. In this case the high momentum release
formula should be used, but the value of ps should be replaced with the
density at choked conditions (p«) . If the release is unchoked and the exit
velocity is of the same order or less than the ambient wind velocity, the low
momentum formula should be used. For a continuous release, if the exit
•
velocity is unknown, the smaller diameter calculated from the two formulae
should be used. Using the smaller diameter will minimize the release buoyancy
criteria discussed in Section 4.13.1. The smaller the release diameter, the
more likely that the release will behave as a denser- than-air release.
If the exit velocity (u) is unknown, but the emission rate (E) , density
in the container (ps) , and hole or emission area (A) are known, an exit
velocity can be estimated. The formula to use is:
With an approximation, the exit velocity can be compared to the ambi-ent wind
speed. The exit velocity for evaporating pools is negligible and need not be
estimated.
Instantaneous releases normally are not associated with an exit velocity
and the low momentum formula should be used. If the released volume of an
instantaneous release is known, the release diameter can be assumed to be the
spherical or cylindrical equivalent diameter of the released volume. If the
release is at ground level, assume the cloud formed is cylindrical (with base
on ground) with the height equal to the diameter. If the release is not at
•
ground level assume the cloud is shaped like a sphere.
4-51
-------
4.12.1 Low-momentum Release
Prior calculations required:
E - emission rate (kg/s) (4.7); and
prel - release density (kg/m3) (4.11).
•
Flow chart reference: Figure 4-6
The diameter of a source after expansion can be estimated as:
D _ 21 E
urel -
where: Drel - diameter (m); and
Ur - ambient wind speed (m/s).
4.12.2 High-momentum Release
Prior calculations required:
prel - release density (kg/m3) (4.11).
Flow chart reference: Figure 4-6
The diameter after expansion can be estimated as:
where: DreL - diameter (m);
Ds - hole or stack diameter (m); and
ps - density at hole (kg/m3) .
4-52
-------
4.13 RELEASE BUOYANCY
The models discussed in this document handle dense gas releases . If a
release can be considered not to be denser than air, standard passive
dispersion modeling or a model capable of handling neutrally buoyant releases
should be applied. To determine whether a release should be considered denser
than air, a comparison is done of the Richardson number describing the release
to a selected value. The formulation of the Richardson number, and thus the
calculation, depends on whether the release is instantaneous or continuous.
4.13.1 Continuous Release
•
Prior calculations required:
E - emission rate (kg/s) (4.7);
Pr»i ' release density (kg/m3) (4.11); and
Dr-1 - release diameter (m) (4.12).
The criteria (Cp)for determining whether a continuous release should be
considered denser than air is if:
C
f9fP«.l -
Orel I Pa
where: Ur - ambient wind speed (m/s);
g - acceleration due to gravity (9.806 m/s2); and
Pa - air density (kg/m3).
4.13.2 Instantaneous Release
Prior calculations required:
E - emission rate (kg/s) (4.7); "
pr-1 - release density (kg/m3) (4.11); and
•
4-53
-------
The criteria (Cp) for determining whether an instantaneous release
should be considered denser than air is if:
C
p
Ur2 I Pa
> 0.2
where: Ur - ambient wind speed (m/s);
g - acceleration due to gravity (9.806 m/s2);
Et - total amount of material released (kg) ;
pa - air density (kg/m3) .
The value of Et must be calculated from the emission rate (E) and
duration of the release as:
Et = E At
where: At - release duration (s) .
The value of Et can also be calculated if the volume of the vessel is known
by:
where: Vt — total volume of the vessel (m3) ; and
p^ =• storage density (kg/m3) .
4.14 RELEASE HEIGHT
The release height represents the altitude above ground that the
emission enters the atmosphere. For liquid releases with evaporating pools,
this height should be that of the liquid- air interface, which is typically at
ground level.
4-54
-------
4.15 GROUND SURFACE TEMPERATURE
An appropriate input value for ground surface temperature should be that
from routine on-site measurements. If such data are not available, ground
surface temperature may be approximated to be equal to the ambient temperature
measured at the standard height for most applications. Experience to date
indicates little, if any, sensitivity in predicted model concentrations due to
changes in this parameter.
4.16 AVERAGING TIME
Hazardous chemical releases are often of short duration, and the concen-
trations of interest are short-term averages. Typical concerns from a
hazardous air release are the maximum short-term concentration and the maximum
dose. Many hazardous air pollutant models are designed to provide
concentration predictions for unit averaging times ranging from 1 second to 1
hour. By contrast, regulatory models for most criteria pollutants have a
basic averaging time of 1 hour for concentration estimates.
Refined dense gas models can typically provide concentration estimates
for user-specified averaging times of 1 hour or less. Due to entrainment with
ambient air, a dense gas release typically does not remain significantly dense
for travel times longer than 1 hour. Meteorological input, however, is often
based on an hourly average. To determine the appropriate averaging time, the
user should examine the release duration and the concentration levels of
concern (keeping health or other reference levels in mind).
If a concentration exposure to population is of concern, the averaging
time should be determined from the concentration limit selected. For example,
Short Term Exposure Limits (STEL) have an implicit time interval of 15
minutes, the Immediately Dangerous to Life and Health limit (IDLH) has an
implicit interval of 30 minutes, and the Emergency Response Planning Guideline
(ERPG) concentrations have a 60-minute implied exposure time. If flammability
or possibility of explosion is of concern, the averaging time should be on the
4-55
-------
order of seconds since an ignition can occur when the instantaneous
concentration is within the flammable limits of the chemical.
Defining an appropriate averaging time for dense gas modeling applica-
tions is complicated. Large averaging times allow for more plume meander, and
therefore, lower average concentrations. Thus, a 5-minute ensemble average
concentration is generally less then a 1-minute ensemble average since more
meander can generally occur in 5 minutes than in 1 minute. Thus, some
researchers use much shorter averaging times (i.e., 5-10 seconds) than the
release duration, especially when comparing model predictions with field
measurements when these field measurements are made with shorter averaging
times. However, meteorological input (i.e., stability classification) used in
conjunction with these models is based on a longer duration, usually 60
minutes.
The time scales relevant to determining the averaging time to be used in
the models can be designated as:
thaz -« averaging time associated with the hazard being assessed;
trel - duration of the contaminant release into the atmosphere; and
ttrav - travel time from the release site to a receptor.
If ta is the averaging time that represents the effect of plume meander, then
the largest recommended ta is the minimum of thaz, trel, and ttrav (for steady-
state or transient releases); smaller values of ta can be chosen to compare
with field observations or to obtain a conservative estimate of downwind
impacts. If ta t is the averaging time that accounts for transient effects
(not pertinent for steady-state releases), ta t is recommended to be thaz.
When the averaging time used is Less than the requested averaging time,
the output must be converted to represent the requested averaging time. To
extrapolate from a reported average concentration to the requested averaging
time, multiply the reported average concentration by the ratio of the actual
averaging time to the requested averaging time. This means that to find a
4-56
-------
concentration averaged over the requested time, a higher concentration may
need to be found in the output. For example, if the release duration is 500
seconds and the requested averaging time is 1000 seconds, the averaging time
of 500 seconds should be used. To convert the concentrations reported in the
output to concentrations for the requested averaging time, each reported
concentration must be multiplied by (500 s/1000 s) or 0.5. This means that if
a 1000 second average concentration of 1 ppm is of interest, the output will
have to be searched for a concentration of (1 ppm/0.5) or 2 ppm.
An alternative to average concentrations being used to determine impact
is the use of dose. Dose is an indication of total exposure at a point. It
is usually expressed in units of ppm-minutes. The simplest dose (D)
calculation is given by:
t
/•
to
where:C — concentration (ppm);
t - time (min); and
tfl = time the release began (min).
Some expressions for dose give an increased weighting to the chemical by
raising the concentration to some power N (change C to GN in the above
equation). In the simple dose calculation, N has che value of one. When N is
not equal to one, dose should actually be referred to as toxic load or some
other name since the units would no longer be ppm-minutes. The values of N
for the chemicals in this document are(18>:
Species N
Ammonia 2
Chlorine 2
Ethylene Oxide • Not available
4-57
-------
Species N
Hydrogen chloride 1
Hydrogen fluoride 1
Sulfur dioxide 1
4.17 METEOROLOGY
Much of the input discussed above is only weakly dependent on the
meteorological conditions. Meteorology is most important to the transport and
dispersion of material after it has been released. Many releases can be
considered as having two independent input streams: one is the description of
the source term, and the other is the meteorological conditions at the time of
and following the release.
4.17.1 Wind Speed and Direction
Wind speed is used in determining: 1) plume rise, 2) plume dilution,
and 3) mass transfer in evaporation models. In very light winds, dense gases
tend to form "pancake-shaped" clouds near the source, and the dense cloud may
not be very deep until further downwind. At higher wind speeds, the rate of
air mixing is increased (more energy is added to the mixture), and the maximum
concentrations can decrease. For releases from liquid pool spills, high wind
speed increases the rate of evaporation, and thus, the plume source strength.
However, high wind speed also results in more dilution due to increased
entrainment of outside air, which can lead to a lowering of maximum
concentrations.
Concentration estimates predicted by air dispersion models decrease as
the wind speed increases. However, mosc dense gas models are less sensitive
to this increase in wind speed at close-in distances, where gravity effects
are dominant. Further downwind, the behavior of dense gas models and non-
dense gas models is more similar.
4-58
-------
If the user wishes to assess the hazard of a past event, wind speed
should be obtained from on-site measurements usually made at the standard 10 m
level height. The wind speed at release height is frequently adjusted inter-
nally by the model using a power law equation. Guidance for on-site meteoro-
logical data collection is available in an EPA document'19'. On-site
meteorological data are normally averaged over a 1-hour interval.
The wind direction is used to approximate the direction of transport of
the plume. The variability of the direction of transport over a period of
time is a major factor in estimating ground-level concentrations averaged over
that time period, i.e., plume meandering. For response analyses, wind direc-
tion should be estimated from on-site or nearby measurements. For planning
analyses, wind direction should be chosen to maximize potential off-site
impacts.
The wind speed and direction used can have large effects on jet
releases. In the extreme case, if a high velocity horizontal release is in
the direction that the ambient wind is blowing toward and the wind speed is
comparable to the release velocity, the release would have a relative minimum
of mixing with the ambient air.
4.17.2 Stability Class
Stability conditions are typically assessed by means of the Pasquill-
Gifford (PG) stability categories where Category A represents extremely
unstable conditions and Category F represents moderately stable conditions.
The modeling guideline(1) recommends several methods for determining the PG
stability category. For example, in urban areas the stability class is often
assumed to be at least one class level less stable than for rural areas. As
guidance documents continue to be created, new methods may become available.
For a given wind speed, stable atmospheric conditions provide smaller
levels of atmospheric turbulence than unstable conditions. The influence of
atmospheric stability on the dispersion of a dense gas (as a result of altered
4-59
-------
levels of ambient turbulence) may be similar to that for neutrally buoyant
releases, but may also be much less. As in the case of wind speed, various
models exhibit different sensitivity levels in concentration prediction due to
changes in stability class(20).
For planning analyses, where there is a need to obtain conservative
estimates from models, it may be necessary to make a number of refined model
simulations using various stability and wind speed combinations. If a release
is at ground level and is not an upwardly directed jet, the F stability class
will most likely lead to the largest concentration impacts downwind. In all
other cases, multiple simulations with varying stability class will be needed.
•
For near-field impacts, dense gas and jet releases will be only weakly
sensitive to stability class. As the release becomes neutrally buoyant or a
non-jet, the plume is more influenced by the atmospheric conditions such as
stability class.
Also see Section 8.0, "Modeling 'Worst Case'," for some discussion on
stability effects for the specific models.
4.17.3 Surface Roughness Length
In principle, the surface roughness length is a measure of the roughness
of a surface over which a fluid is flowing. For a homogeneous surface, ics
value is sometimes approximated as l/10th of the average height of the surface
irregularity. When the landscape contains any obstructions (i.e.,
nonhomogeneous), an effective roughness length must be determined. Typical
values of surface roughness are shown in Table 4-1.
The overall effects of increasing surface roughness are to retard che
horizontal, buoyancy-induced spreading of the plume or cloud and to enhance
the mixing between plume and environment as a result of the ambient and plume
turbulence(21). As in the case of wind speed and stability class, various
models exhibit different capabilities to incorporate the effect of surface
4-60
-------
TABLE 4-1. REPRESENTATIVE VALUES OF SURFACE ROUGHNESS FOR A UNIFORM
DISTRIBUTION OF SELECTED TYPES OF GROUND COVER<21)
Surface Surface Roughness (m)
Ice 0.00001
Snow 0.00005 to 0.0001
Sand 0.0003
Soils 0.001 to 0.01
Short grass 0.003 to 0.01
Long grass 0.04 to 0.10
Agricultural crops 0.04 to 0.20
4-61.
-------
roughness in predicted concentrations(22>. Because releases of a. dense gas may
occur in industrial settings where the presence- of a wide variety of structure
heights and shapes is common, some users have input large surface roughness
values into a model to account for the presence of such obstacles. However,
these large surface roughness values can significantly decrease modeled
concentrations. For low or ground-level releases, increasing the surface
roughness value by a factor of 10 may result in concentration reductions by
about a factor of 2<22). The use of "enhanced" roughness values for simulating
the effects of obstacles in an industrial setting has yet to be thoroughly
tested and justified. For planning analyses where there is a need for
obtaining conservative estimates from these models, the use of a surface
roughness value characteristic of the smallest roughness element in the
vicinity of the release is probably most appropriate. If no information is
available on the surface roughness, a value of 0.01 m is suggested.
4.17.4 Wind Speed at 10 m Altitude
The stability class and surface roughness length can also alter the
vertical wind profile assumed in a model. The wind speed at any height can be
estimated by(23):
where: u = wind speed at height z;
ux — wind speed measured at height z^\ and
p - stability and surface roughness related exponent.
Some example values of p are given in Table 4-2.
The stability class determination requires the wind speed at 10 meters
altitude. However, in the above formula the stability class is used to
calculate the 10 meter altitude wind. When the 10 meter wind speed
calculation is complete, the stability class should be confirmed.
4-62
-------
TABLE 4-2. VALUES FOR THE EXPONENT IN THE WIND PROFILE CURVE AS A FUNCTION OF
STABILITY CLASS AND SURFACE ROUGHNESS (z0)(23)
Stability/z0
A
B
C
D
E
F
0.01 m
0.05
0.06
0.06
0.12
0.34
0.53
0.10 m
0.08
• 0.09
0.11
0.16
0.32
0.54
1.00 m
0.17
0.17
0.20
0.27
0.38
0.61
3.00 m
0.27
0.28
0.31
0.37
0.47
0.69
4-63
-------
4.17.5 Ambient Temperature. Relative Humidity, and Pressure
To assess the consequences of a past event, input parameters should be
used that are representative of the conditions at the time of the release. An
appropriate input value for each of these parameters should be obtained from
on-site measurements at the time of release. Guidance on methods for
•
collecting these data are presented in the meteorological data guidance
document5193. If on-site data are not available for these parameters,
observations from nearby National Weather Service (NWS) stations may be used
instead.
These parameters can have some marked effects on hazardous releases.
For instance, if the ambient temperature is above or below the boiling point
of a chemical, it will determine whether the liquid may flash, boil, or
evaporate from the wind. Similarly, the relative humidity can affect the
dynamics of elevated jet releases^155. Varying the humidity can result in
altering the plume height above the ground. Models are not normally as
sensitive to the pressure. Since the pressure is used to determine release
criteria (two-phase versus single-phase releases), some variations in
simulations could be expected for a spill in-Denver as opposed to a spill at
sea level.
4.18 OUTPUT DEFINITION
When running a model, a set of output parameters of interest are
normally specified. The basic parameters are concentrations of interest and
receptor locations. The ERPGs for each chemical are listed in Appendix B.
Some STEL and IDLH values are also listed in the Appendix. The receptor
locations need to be determined on a case-by-case basis. Normally all
receptors are at ground level. The distance from the release site is
determined by whether on-site or off-site (or both) impacts are being studied.
4-64
-------
A number of outputs may be desired. The outputs that are most often
requested and those available from the models in this study are discussed in
Section 7.
4-65
-------
SECTION 5
MODEL INPUT DEVELOPMENT FOR RELEASE CLASS EXAMPLES
In the previous section methods for determining input to hazardous air
pollutant simulation models were provided. In this section, examples of the
eight release classes described in Section 2 are presented to illustrate model
input requirements. For each example, a release scenario is presented
followed by a description of how the input values are developed. Although in
reality there may be more than one release class associated with a single
release scenario, only one release class is covered in this section for each
release scenario. If a scenario is made up of more than one release class,
each class should be modeled separately.
Not all of the input presented here for a given release class are used
by every model. Section 6.0 of this document gives, for each model, the input
actually used in each release class.
Two key assumptions have been made for all examples presented in this
section. First, it is assumed in all the scenarios that the release takes
place in the direction with the wind. A release against the wind would have
more turbulence and mixing, which would reduce maximum concentrations
downwind. Second, the release is assumed to occur at or near the ground,
which keeps the emission closer to the ground than does a vertical release.
The examples presented in this section are organized by release class.
As explained in Section 4, calculation of some release parameters is required
before a release class determination can be made. Determining a release class
is often an iterative procedure. A release class is assumed, calculations are
carried out, and a self-consistency check is made. If the self-consistency
check fails the actual release class must be different than the one assumed.
5-1
-------
The release class determination for each example is described in each
subsection 5.X.3 (where X is 1 through 8).
The calculations performed in the examples presented in this section are
discussed in detail in Section 4. The subsections in this Section are related
to the subsections in Section 4 such that the calculation performed in
subsection 5.X.N is described in subsection 4.N.X. For example, emission rate
calculations are described in Section 4.7. The calculation for the emission
rate in the first example release scenario is given in subsection 5.1.7 and
the corresponding calculation technique is described in subsection 4.7.1.
5.1 TWO-PHASE GAS RELEASE (CHOKED) EXAMPLE
The scenario for this example is as follows. A 3-inch pipe delivers
saturated vapor-phase ethylene oxide from a storage area to a process area.
The temperature and pressure within the pipe are 86.9°F (303.67 °K) and 2.0
atmospheres absolute (202,650 Pa), respectively. The pipe is supported on a
pipe rack 12 ft (3.66 m) above grade level. The pipe's length for this
scenario is assumed to be zero. (The longer a pipe is the greater the
resistance in the pipe to the flow of the chemical inside. Thus, by ignoring
the pipe length, the emission rate will be maximized.)
A leak develops in the flange of the pipe with a size equivalent to a
round hole of 0.5 inches (0.0127 m) in diameter. The release of vapor-phase
ethylene oxide continues until the pipe flow is manually isolated, 8 minutes
after its inception. The pressure and temperature within the pipe remain
constant throughout the duration of the release. Ethylene oxide is in its
vapor phase at the release point (See Section 5.1.3 for the determination).
The material flows through the release opening under critical (choked)
conditions (also shown in Section 5.1.3). Upon cooling due to expansion, part:
of the material condenses as it enters the atmosphere, thereby forming a two-
phase release. Figure 5-1 provides a simple sketch of the release.
5-2
-------
Ethylene Oxide Pipeline
-
Ethylene Oxide Vapor at
Constant Temperature & Pressure
Two Phase Gas Release
Figure 5-1. Two-Phase Gas Release (Choked)
5-3
-------
The meteorological conditions at the time of release are:
• Northeast wind at 12 mph;
• Temperature at 57.2°F;
• Relative Humidity of 62%;
• Pressure of 1 atm (sea level);
• 3/8 cloud cover;
• Early-afternoon release; and
• Measurement height of 15 ft.
The nearest public boundary is 200 m from the release point.
A summary of the model input is provided in Table 5-1. Each of the
following sections describes how these input values were developed.
5.1.1 Observable Data
The observable data input were listed in the scenario description.
Table 5-2 summarizes this information.
5.1.2 Chemical Data Requirements
The data for ethylene oxide is in Appendix B.
5.1.3 Release Class
This section describes the calculations required to determine the
release class for this example. The flow diagram in Figure 4-1 is used as a
guide.
Note that some assumptions made in the process of determining the
release class conflict with the description of the release scenario. Also
note that many of the calculations required to determine the release class are
also required for other model input. Rather than presenting all these
calculations on one section, reference is made to the appropriate section for
each calculation.
5-4
-------
TABLE 5-1. INPUT SUMMARY FOR THE TWO-PHASE CHOKED GAS RELEASE EXAMPLE
Chemical Data (5.1.2)
From Appendix B
Release Type (5.1.3)
Two-phase choked flow
Continuous or Instantaneous Release Categories (5.1.4)
Continuous
Emission Rate (5.1.7)
0.06340 kg/s
Release Temperature (5.1.8)
283.85 "K (Tb)
Vapor Fraction (5.1.9)
Fr.i ~ 0.9867
Initial Concentration (5.1.10)
fa - 0.0
fw - 0.0
fi - 1.0
Density (5.1.11)
prel - 1.916 kg/m3
pa' - 1.220 kg/m3 (at ambient temperature)
Release Diameter or Area (5.1;12)
Diameter - 0.01347 m (choke point)
Diameter - 0.01725 m (expanded)
Release Buoyancy (5.1.13)
C^, - 2.435 < 6 (use dense gas model)
Release Height (5.1.14)
3.66 m
Ground Surface Temperature (5.1.15)
Same as ambient
Averaging Time (5.1.16)
5 seconds
5-5
-------
TABLE 5-1. (CONTINUED)
Meteorology (5.1.17)
Wind Speed and Direction
45° <§ 12 mph (5.37 m/s)
Measurement height - 15 ft (4.57 m)
Speed @ 10 m - 5.63 m/s
Stability Class
C
Surface Roughness Length
ZO - 0.01 m
Ambient Temperature, Relative Humidity, and Pressure
Temperature - 57.2"F (287.52 °K)
Relative Humidity - 62%
Pressure - 1 atm (101325 Pa)
Output Definition (5.1.18)
LEL (3%) and UEL (100%) concentrations
Minimum distance of concern - 200 m
Observable Data (5.1.1)
See Table 5-2
5-6
-------
TABLE 5-2. OBSERVABLE DATA SUMMARY FOR THE TWO PHASE CHOKED GAS RELEASE
EXAMPLE
Release Description
Species: Ethylene oxide
Container: Horizontal pipe
Diameter: 3 inches (0.0762 m)
Temperature: &7°F (303.67 °K)
Pressure: 2.0 atm (absolute) (2.0265 x 105 Pa)
Hole site:
Inner diameter: 0.5 inch (0.0127 m)
Hole area: - 1.267x10'* m2
Height of hole: 12 ft (3.66 m)
Duration: 8 mirf
Nearest boundary: 656.2 ft (200 m)
Meteorology:
Temperature: 57.2°F (287.52 °K)
Relative Humidity: 62%
Pressure: 1 atm (101325 Pa)
Wind Speed: 12 mph (5.37 m/s)
Wind Direction: Northeast (45°)
Measurement hgt: 15 ft (4.57 m)
Surface Roughness: 0.01 m
Cloud cover: 3/8
Time: Early-afternoon
5-7
-------
The ethylene oxide in the pipe is in gas phase. To select the release
class, it must be determined whether the release is choked or not and whether
the gas cools enough for condensation to occur.
The release flow is choked (see Section 5.1.6) when the pressure at the
exit (p«) is calculated as being higher than the ambient pressure (pa). In
determining the.release class, first it is assumed that the release is single
phase. The value of the temperature at choked conditions (T») is estimated by
the equation:
= '(3
-------
example is 74 seconds, This advection time was calculated using a wind speed
of 12 mph (5.37 m/s) and a distance of 200 meters. Since the emission
duration is longer than the time required to reach the downwind point of
interest, this release is considered continuous.
For a source -term model, the value of td is also sufficiently long to
treat this release as a continuous release for most receptors.
5.1.5 Release - class - specific Calculations
•
The flow chart for the calculations of input for a choked two -phase gas
release is shown in Figures 4-2 and 4-6.
5.1.6 Determination of Choked Flow for Gas Releases
To determine whether the flow is choked or not, (see Section 4.6) the
exit pressure (p*) must be estimated using the formula:
where: 7 - the ratio of the specific heat at constant pressure to that at
constant volume (1077.96 J/kg °K / 888. 96 J/kg °K) - 1.2126.
/ ~
p, = (2.0265xl05Pa) - ± -
\12126 + l
Since p, > pa (1.01325xl03 Pa), flow is choked.
5-9
-------
5.1.7 Emission Rate
As described in Section 4.7.1, a number of parameters are required to
calculate the emission rate for this release class. These are the temperature
(T,) , vapor fraction (F*), change in enthalpy (AH*), and density (p*) at choke
flow. The first of these is calculated by solving:
p.=101325Pa exp
J_
T.
or
1.1389xl05Pa = 101325Pa
8314J/kmol°K
1
283.85°K
which yields a T* of 287.01 °K.
The value for F, is estimated from:
or
F = 1
MC.
- R
F. = 1
287.01° K
(44.053kg/kmol)(5.6900xl05J/kg)
(44.053kg/kmol)(1077.96J/kg°K)
- (83141/kmorK) In
f2.Q265xloW
,1.1389xl05Pa,
= 0.9758
5-10
-------
The change in enthalpy from storage conditions to choke conditions is
calculated by:
AH, = CpCTs - TJ
or
AH, = (1077.96J/kg°K)(303.67°K - 287.01°K)
+ (5.6900xl05J/kg)(l - 0.9758)
= 3.1729xl04J/kg
The emission density at choke flow is given as;
P. =
RT
1 -
Pi
-i
or
P, =
(0.9758)f (8314J/kmol0K)(287.Ql°K) -
( (1.1389xl05Pa)(44.053kg/kmol) t
; f1 ~ a9758 \
i882.67kg/m3|
= 2.1546kg/m3
-i
Finally, the emission rate for choked flow is:
= A o
2(0.85)
f AH« )
\
f 4fL ^
' + "D7JJ
i
2
5-11
-------
Thus , using a zero-length pipe (L*,) to give a conservative estimate;
E = (1.267x10 -4m2)(2.1546kg/m3)
3.1729xl04J/kg
2(0.85)
+ 4(0.0045)(0.0m)>|
0.0762m Jj.
= 0.06340kg/s
5.1.8 Release Temperature
The calculation of release temperature for this release class is
described in Section 4.8.1. The first attempt at calculating the release
temperature should be with the assumption that the value of Frel is less
than 1. If the value of Frel later is calculated to be greater than 1, the
value of Tral must be modified. The release temperature is calculated from:
pa=101325Pa exp
XMf 1
1
thus,
101325Pa=1013251>a
8314J/kmol°K
1
283.85°K T
k *
Since the pressure in this example is 1 atm, the release temperature is the
boiling point. Thus, Tr8l is 283.85 °K.
5.1.9 Vapor Fraction
From Section 4.9.1, the vapor fraction can be calculated from:
5-12
-------
thus,
F = 09758 + (1971.56J/kg°K)(287.01°K - 283.85°K)
** ' 5.6900xl05J/kg
= 0.9867
Since the value of Frel is between zero and one, recalculation of Trei is not
required.
5.1.10 Initial Concentration
Since this is a two-phase flow, it is safe to assume that the release is
of the liquid and vapor states of the chemical in equilibrium. Since.the
release temperature is at the boiling point, the vapor pressure is 1 atm.
This means that in this release only ethylene oxide is emitted. Thus, under
this assumption, the concentration levels (described in Section 4.10) are:
f. - 0.0
fw » 0.0
f = 1.0
5.1.11 Density
The release starts as aerosol and vapor of ethylene oxide. The water
vapor molar fraction is assumed to be zero. The density of the water vapor
component of the atmosphere is also assumed to be negligible. As explained in
Section 4.11, when aerosol is present, the DEGADIS model will require two
triplets, one describing pure air and one describing the aerosol and vapor
mix. The density of air (including water vapor) at the ambient temperature is
5-13
-------
required to be able to provide one of the triplets. Assuming that air behaves
as a perfect gas:
RT
a
= (101325Pa)(28.9kg/kmol)
(8314J/kmol°K)(287.520K)
= 1225kg/m3
Similarly, the water vapor density is given as:
_PaMw. (101325Pa)(18.Q2kg/kmol)
Pw RTa "(8314J/kmol°K)(287.52°K)
= 0.7638 kg/m3
The mole fractions of ambient water vapor (fw') and air (fa') are given as:
/ fwn "*"-T£
£ = Ml IQ\ T« /
w oo
oo
= 0.01019
fax = 1 - S^ = 1 - 0.01019 = 0.9898
The apparent molecular weight of the air and water vapor of the ambient
atmosphere is:
5-14
-------
= (0.9898)(28.9kg/kmol) + (0.01019)(18.02kg/kmol)
= 28.79kg/kmol
Finally, the ambient density pa' is given as:
\-l
f'tjr f'lsr
P! =
(0.9898)(28.9kg/kmol)
(0.01019)(18.02kg/kmol)
1
^(28.79kg/kmol)(U25kg/m3) (28.79kg/kmol)(0.7638kg/m3)J
= 1220kg/m3
For the second triplet, the density of the aerosol/vapor mixture (p^ at
the initial portion of the release is required. To calculate pit the
densities of the vapor- (pg) and liquid (p±) must be known. The temperature
used in the calculation is the release temperature, which in this case, is the
boiling point.of ethylene oxide. Assuming a perfect gas, ps is:
=
P
M . (101325Pa)(44.Q53kg/kmol)
8 RTe (8314J/kmol°K)(283.85°K)
= 1.891 kg/m3
The liquid density at the boiling point is given in Appendix B as 882.67
kg/m3. The overall aerosol/vapor mixture density then, from Section 4.11, is:
Pi =
Pi
5-15
-------
or
/ 0.9867 ^ (1-0.9867) V1
1 (l.891kg/m3 882.70kg/m3j
= 1.916kg/m3
As stated in Section 5.1.10, the release is of pure ethylene oxide.
That is, both fa and fw are zero. Therefore, the release density (prel) is
equal to the ethylene oxide density (p±) .
5.1.12 Release Diameter or Area
The release diameter can be calculated using the high momentum formula
(Section 4.12.2) for the expansion size. Since this release is choked, there
are two release diameters to be considered; which one is to be used depends on
the model. The first release diameter considered is that of the choke
conditions. This diameter will replace the actual hole size in models
requiring input on hole size. If the actual hole size were used, a calculated
exit velocity would be greater than sonic speed. The choke point release
diameter from the high momentum formula is:
Prel
= 0.0127m
= 0.01347m
2.1546kg/m3
1.916kg/m3
The second release diameter is calculated using the storage density
rather than the choke density. This release diameter is used in models
requiring the after-expansion size of a release. To calculate this release
5-16
-------
diameter, the gas density inside the container (ps) is required. Assuming the
perfect gas law:
= PsM; _ (2.0265xl05Pa)(44.053kg/kmol)
Ps RTS" (8314J/kmol°K)(303.67°K)
= 3.536 kg/m3
The expanded release diameter is then:
%
o.oi;
0.01'
A
Prd
!7m
725m
3.536kg/m3
1.916kg/m3
5.1.13 Release Buoyancy
The release buoyancy is used to determine if a dense gas model should be
used, rather than a neutral or positive buoyancy model. If prel is less than
or equal to pa, the release is not to be treated as a dense gas release. In
this case, prel is 1.916 kg/m3 and p3' is 1.220 kg/m3. Since prel is greater
than pa', the denser-than-air criteria should be calculated Co determine
whether this should be treated as a dense gas or not.
The formula in Section 4.13.1 is used to calculate the denser-than-air
criterion, since this release is being considered a continuous release
(Section 5.1.4). The larger Drel from Section 5.1.12 will be used to maximize
the criterion. The criterion (Cp) is given by:
5-17
-------
u.
rg(E/Pre,)fPrel - p y
= (5.37m/s)
\~tr
' (9.806m/s2)(0.06340kg/si/1.916kg/m3)
0.01725m
1.916kg/m3 - 1.220kg/m:
1.220kg/m3
/j
= 2.435
Since Cp < 6, a dense gas model should be used.
5.1.14 Release Height
The flange with the leak is 12 ft (3.66 m) above grade. The gas and
liquid droplets that form are both assumed to enter the atmosphere at that
altitude.
5.1.15 Ground Surface Temperature
No direct information on the ground surface temperature was given.
Therefore, the ground surface temperature is assumed to be equal to the
ambient temperature, i.e., 87°F or 303.67 "K.
5.1.16 Averaging Time
The averaging time is specified as 5 seconds to make the model-predicted
concentrations comparable to the Lower and Upper Explosion Limits.
5-18
-------
5.1.17 Meteorology
Wind Speed and Direction
The wind speed and direction are either assumed to be available from on-
site meteorological equipment or are average conditions at the site. The wind
speed is given as 12 mph (5.37 m/s). .The wind direction is given as northeast
(45°). The measurement height is given as 15 ft (4.57 m). To determine the
greatest specific impact, multiple wind speeds may be required in conjunction
with multiple stability classes.
Stability Class
The stability class is not given explicitly. However, a stability class
can be estimated from the information provided using the method described in
the Workbook, Section 3.1.2. In this example, there is 3/8 cloud coverage and
the release occurs in the early afternoon. These conditions indicate that
insolation is strong. The stability class estimated in the Workbook Table 3-2
for a wind speed greater than 5.37 m/s at 10 m (assuming wind speed increases
with height) during a day with strong insolation would be C.
Surface Roughness Length
In keeping with che value suggested to be used for planning, the surface
roughness is assumed to be 0.01 m.
Wind Speed at 10 m Altitude
Given the previous estimates of surface roughness, stability class, and
wind speed at 15 ft, the wind speed at 10 meters can be estimated using the
equation in 4.17.3:
5-19
-------
,4.57m
= 5.63m/s
This wind speed at 10 meters altitude still indicates a C stability class.
Ambient Temperature. Relative Humidity, and Pressure
The ambient temperature, relative humidity, and pressure are all assumed
to be available from on-site equipment. -See the observable data, Table 5-2.
5.1.18 Output De f ini t i on
The concentrations to output should be for the LEL (3%) and UEL (100%).
The actual impact values to be predicted are listed in Section 7.0. The
specified point mentioned in Section 7.0 is assumed to be the site boundary,
200 m directly downwind of the release.
5.2 TWO-PHASE GAS RELEASE (UNCHOKED) EXAMPLE
In this example, the 3 inch ethylene oxide pipeline described in the
previous example is now operating at a pressure of 1.58 atmospheres absolute
(160,094 Pa) and the ethylene oxide within the pipe is a saturated vapor at
71.6°F (295.16 °K).
A leak develops in the flange of the pipe with a size equivalent to a
round hole of 0.5 inches (0.0127 m) in diameter. As in the previous example,
che release is 12 ft (3.66 m) above grade level. The release of vapor-phase
ethylene oxide continues until the pipe flow is manually isolated, 8 minutes
after its inception. The pressure and temperature within the pipe remain
constant throughout the duration of the release. As shown below, the ethylene
oxide is in its vapor phase at the release point. The flow of material
5-20
-------
through the opening is under subcritical (unchoked) conditions. As the gas
expands upon release, it cools, causing part of the material to condense as it
enters the atmosphere, thereby forming a two-phase release situation. Since
the release configuration is identical to that of the previous example,
Figure 5-1 also represents this release scenario.
•
The meteorological conditions at the time of release are:
• Southwest wind at 6 mph;
• Temperature at 74°F;
• Relative Humidity of 37%;
• Pressure of 1 atm (sea level);
• 5/8 cloud cover;
• Early-morning release; and
• Measurement height of 20 ft.
The nearest public boundary is 300 meters from the release point.
Model input is summarized in Table 5-3. The following sections describe
how specific input values were calculated.
5.2.1 Observable Data
The observable data input was provided in the scenario description.
These data are summarized in Table 5-4.
5.2.2 Chemical Data Requirements
The data for ethylene oxide is in Appendix B.
5.2.3 Release Class
This section describes the calculations required to determine the
release class for this example. The flow diagram in Figure 4-1 was used as a
guide.
5-21
-------
TABLE 5-3. INPUT SUMMARY FOR THE TWO PHASE UNCHOKED GAS RELEASE EXAMPLE
Chemical Data (5.2.2)
From Appendix B
Release Type (5.2.3)
Two-phase unchoked flow
Continuous or Instantaneous Release Categories (5.2.4)
Continuous
Emission Rate (5.2.7)
0.0503 kg/s
Release Temperature (5.2.8)
283.85 *K (Tb)
Vapor Fraction (5.2.9)
Fr-1 - 0.9779
Initial Concentration (5.2.10)
fa - 0.0
fw - 0.0
fi - 1.0
Density (5.2.11)
pr-1 - 1.9336 kg/m3
pa' - 1.183 kg/m3 (at ambient temperature)
Release Diameter or Area (5.2.12)
Diameter - 0.01548 m
Release Buoyancy (5.2.13)
Cp - 1.226 < 6 (use dense gas model)
Release Height (5.2.14)
3.66 m
Ground Surface Temperature (5.2.15)
Same as ambient
Averaging Time (5.2.16)
5 seconds
5-22
-------
TABLE 5-3. (CONTINUED)
Meteorology (5.2.17)
Wind Speed and Direction
225° (§2.68 m/s
Measurement height - 20 ft (6.10 m)
Speed @ 10 m - 5.63 m/s
Stability Class
C
Surface Roughness Length
ZO - 0.01 m
Ambient Temperature, Relative Humidity, and Pressure
Temperature - 74°F (296.48 °K)
Relative Humidity - 37%
Pressure - 1 atm (101325 Pa)
Output Definition (5.2.18)
LEL (3%) and UEL (100%) concentrations
Minimum distance of concern — 300 m
Observable Data (5.2.1)
See Table 5-4
5-23
-------
TABLE 5-4. OBSERVABLE DATA SUMMARY FOR THE TWO PHASE UNCHOKED GAS RELEASE
EXAMPLE
Release Description
Species: Ethylene oxide
Container: Horizontal pipe
Diameter: 3 inches (0.0762 m)
Temperature: 71.6°F (295.16 °K)
Pressure: 1.58 atm (absolute) (1.6009 x 10s Pa)
Hole site:
Inner diameter:
Hole area:
Height of hole:
Duration:
0.5 inch (0.0127 m)
1.267x10'* m2
12 ft (3.66 m)
8 min
.Nearest boundary: 984 ft (300 m)
Meteorology:
Temperature:
Relative Humidity:
Pressure:
Wind Speed:
Wind Direction:
Measurement hgt:
Surface Roughness:
Cloud cover:
Time:
74°F (296.48 °K)
37%
1 atm (101325 Pa)
6 mph (2.68 m/s)
Southwest (225°)
20 ft (6.10 m)
0.01 m
5/8
Early-morning
5-24
-------
Note that some assumptions made in the process of determining the
release class conflict with the description of the release scenario in 5.2.1.
Also note that many of the calculations required to determine release class
are also required for other model input. Rather than presenting all these
calculations in one section, reference is made to the appropriate section for
each calculation.
The ethylene oxide in the pipe is in the gas phase. To select the
release class, it must be determined whether the release is choked or not and
whether the gas cools enough for condensation to occur.
The release flow is unchoked (as shown in Section 5.2.6) when the
pressure at the exit (p*) is calculated as being lower than the ambient
pressure (pa). In the determination of the release class, it is first assumed
that the release is single phase. A discharge temperature (Trel) is required
so that the vapor pressure at that temperature can be compared to the ambient
temperature. The equation for an unchoked single phase gas release is given
in Section 4.8.6 as:
2T
S
rci
[1 + /(I + 4aTs)]
where:
a =
The emission rate is required to calculate the temperature. The
emission rate calculation is described in Section 4.7.6. The expressions
required for the emission rate are:
5-25
-------
E = KYAJ2ps(ps - pa)]*
K-
P - P
Y = 1 - 152 IS I (0.41 + 0.35
PSY
The ft term is the square root of the ratio of the hole size to the area
containing the volume of the ethylene oxide. The hole size in this example is
1.267x10"* m2. The surface area of the pipe containg the volume of ethylene
oxide is the length of the pipe multiplied by the circumference of the pipe.
The length of the pipe is not given in the scenario description. In the first
release class example (Section 5.1), it was assumed that the pipe was zero in
length which resulted in maximizing the emission rate. From Figure 5-1 it is
apparent that the pipe is much longer than it is wide. Thus the surface area
of the pipe is much larger that the area of the hole, which leads to the /8
term being assumed to be zero. Using the constants in Section 4.7.6, the
value of 7 (1.2126) from Section 5.2.6, and ethylene oxide data from Appendix
B, this leads to the following:
K = C « 0.62
Y = 1 - I?! 5-10.41
PSY
» I . (1.6009xl05Pa - 1.01325x10 Wj Q41
( (1.6009xl05Pa) (12126)
= 0.8759
5-26
-------
E = KYA.pp.fe. - Pa)]2
= (0.62)(0.8759)(1.267xlO-4m2)[2(2.873kg/m3)(1.6009xl05Pa - 1.01325xl05Pa
=0.0400kg/s
( (0.0400kg/s)(8314J/kmorK)
2(1077.96J/kg°K) ( (1.01325xl05Pa)(44.053kg/kmol)(1.267xlO '4m2)
=1.604xlO-4°K-1
2T
T = s
•"•id
2(295.16° K)
[l + ^(1 + 4(1.604x10-4°K-1)(295.16°K))]
= 282.4°K
The value of Trel is less Chan the critical temperature of ethylene oxide
(469.15 8K) given in Appendix B. Trel is also less than the normal boiling
point of ethylene oxide (283.85 °K). The ethylene oxide is at Trel at
atmospheric pressure. This means that the ethylene oxide vapor pressure is
less than the ambient pressure. From che flow chart in Figure 4-1, the
release class therefore must be a two-phase gas release. Use the flow chart
in Figure 4-2 to calculate the rest of the input.
5-27
-------
5.2.4 Continuous or Instantaneous Release Categories
In this example, the emission duration (td) is given as 8 minutes. This
is the amount of time that elapsed before a valve was closed or the release
was mitigated in some other manner.
For a non-source-term model, the value of td must be compared either to
advection time to a receptor or to the maximum downwind distance reached by a
concentration of interest. Using the formula in Section 4.4 for ttrav, and
assuming a wind speed of 6 mph (2.68 m/s) and a distance of 300 meters, the
advection time is calculated to be 223.8 seconds. Since the emission duration
is longer than the time required to reach the downwind point of interest, the
release is considered continuous.
For a source-term model, the value of td is also sufficient to treat
this release as a continuous release for most receptors.
5.2.5 Release-class-specific Calculations
•The flow chart of the calculations required for input for an unchoked
two-phase gas release is shown in Figures 4-2 and 4-6.
5.2.6 Determination of Choked Flow for Gas Releases
As discussed in Section 4.6, to determine whether the flow is choked or
not, the exit pressure (p*) must be estimated using the formula:
In this example, the exit pressure (p.) is:
5-28
-------
p, = (1.6009xl05Pa)(
V-2126 + I)
. 1J2126
1(1.2126 - 1)
= 8.997xl04Pa
where: j - the ratio of the specific heat at constant pressure to that at
constant volume (1077.96 J/kg °K / 888.96 J/kg °K) - 1.2126.
Since p* < pa (1.01325xl05 Pa), flow is unchoked.
5.2.7 Emission Rate
As indicated in Section 4.7.2, to calculate the emission rate, two
parameters must be known. These are the change in enthalpy during the release
(AH, Section 5.2.9) and the release density (prel, Section 5.2.11). Once
these two parameters are known, the emission rate can be calculated using the
following expression:
E = A
2(0.85)
f AH }
(l . 4fl-p)
U Dp jj
1
2
or, in this example, where the pipe length is zero:
E = (1.267x10
2(0.85)
2.4767xlQ4J/kg
4(0.0045)(0.0m)
D
= 0.0503kg/s
5-29
-------
5.2.8 Release Temperature
As discussed in Section 4.8.2, the release temperature (Trel)is
calculated from the Glausius-Clapeyron equation:
prel=101325Pa exp
_L
T
1
or, in this example,
101325Pa = 101325Pa exp
(5.6900xl05J/kg)(44.053kg/kmol)
8314J/kmol°K
1
283.85°K T
Since the pressure after release is one atmosphere, the release temperature is
the boiling point. The fact that prel is equal to the atmospheric pressure
(101325 Pa) means that the exponent must be zero. It is only zero if Trel is
equal to Tb. Thus, Trel is 283.85 °K. Otherwise, the equation would have to
be re-written solving for Tral.
5.2.9 Vapor Fraction
As discussed in Section 4.9.2, the vapor fraction can be calculated
from:
T
MI
MC.lnlqrH -Rtal —
TtdJ IP.,
or, in this example,
5-30
-------
283.85°K
(44.053kg/kmol)(5.6900xl05J/kg)
(44.053kg/kmol)(1077.96J/kg°K)
J295.16°r
- (8314J/kmol°K)
\283.85°K,
1.6009xl05Pa
1.01325xl05Paj
= 0.9779
Now that the values of Fral and Tj.^ are known, the change of enthalpy in
the release can also be calculated. This change in enthalpy (AH) is given as:
AH = Cp(Ts -
= (1077.96J/kg°K)(295.16°K - 283.85°K)
+ (5.6900xl05J/kg)(l - 0.9779)
= 2.4767xl04J/kg
5.2.10 Initial Concentration
Since this release is a two-phase flow, it is safe to assume that all
material being emitted is ethylene oxide. Therefore, it follows that the
concentration values (described in Section 4.10) are:
fa - 0.0
fw - 0.0
fx -1.0
5.2.11 Density
The release starts as a mixture of aerosol and vapor. The water vapor
molar fraction is assumed to be zero. The density of the water vapor
5-31
-------
component of the atmosphere is also assumed to be negligible. As explained in
Section 4.11, when aerosol is present, the DEGADIS model will require two
triplets, one describing pure air and one describing the aerosol and vapor
mixture. The density of air at ambient temperature (pa) is required to
calculate one of the triplets. Assuming the air behaves as a perfect gas:
= PaMa_ (101325Pa)(28.9kg/kmol)
Pa RTa"(8314J/kmol°K)(296.48°K)
= 1.188 kg/m3
Similarly, the water vapor density (pw)is given as:
•- P>MW- (101325Pa)(18.02kg/kmol)
Pw ~ RTa ~(8314J/kmol°K)(296.48°K)
= 0.7407 kg/m3
The mole fractions of ambient water vapor (fw') and air (fa') are given as:
f' . *» IQ\ T* ;
w Uooj
/ 07 \ /6J994 - 23g ]
_ [ 3/ ] JQ\ 296.48-Kj
"Uooj
= 0.01074
and
fa = 1 - f^ = 1 - 0.01074 = 0.9893
The apparent molecular weight (M'r)of the air and water vapor of the ambient
atmosphere is:
5-32
-------
= (0.9893)(28.9kg/kmol) + (0.01074)(18.02kg/kmol)
= 28.78kg/kmol
Finally, the ambient density (pa') is given as:
'W
Pa =
or
/ = / (0.9893)(28.9kg/kmol) + (0.01074)(18.02kg/kmol) r1
Pa (k(28.78kg/kmol)(1.188kg/m3) (28.78kg/kmol)(0.7407kg/m3)J
= 1.183kg/m3
The density of the aerosol/vapor mixture at the initial portion of the
release is required to calculate the second triplet. To calculate the density
of the aerosol/vapor mixture (p^ , the densities of the vapor and liquid must
be known. The temperature used in the calculation is the release temperature.
In this example, the release temperature is the boiling point of ethylene
oxide. Assuming a perfect gas, the density of the gas state (pg) is:
_ PaMi _ (101325Pa)(44.053kg/kmol)
P* RTe (8314J/kmol°K)(283.85°K)
= 1.891 kg/m3
The liquid density at the boiling point is given in Appendix B as 882.70
kg/m3. The formula for calculating the overall aerosol/vapor mixture density
is (from Section 4.11):
5-33
-------
Pi
or
J 0.9779 . d-0.9779) Vi=l
U.891kg/m3 882.70kg/m3J
Since the release is of pure ethylene oxide (fa and fw are both zero) ,
the release density (ptel) is equal to that of the ethylene oxide density (p^) .
5.2.12 Release Diameter or Area
The release diameter is calculated as discussed in Section 4.12. Since
this is not a choked flow release, an estimate is required for exit velocity.
To estimate exit velocity, the formula requires the gas density inside the
vessel (ps) . Assuming the perfect gas law yields:
_ _ (1.6Q09xlQ5Pa)(44.053kg/kmol)
Ps~ RTS (8314J/kmol0K)(295.160K)
= 2.874kg/m3
The exit velocity can then be estimated as :
O.Q5Q3kg/s
(1.267x10 -4m2)(2.874kg/m3)
138m/s
5-34
-------
With such a high emission velocity, the release diameter can be
calculated using the high momentum formula given in Section 4.12.2. The
release diameter from the high momentum formula is then:
= 0.0127m
= 0.01548m
2.874kg/m3
1.9336kg/m-
5.2.13 Release Buoyancy
The release buoyancy is used to determine if a dense gas model should be
used rather than a neutral or positive buoyancy model. If prel is less than
or equal to pa,, the release is not to be treated as a dense gas release. In
this case, prel is 1.934 kg/m3 and pa' is 1.183 kg/m3. Since prel is greater
than pa', the next step is to apply the denser-than-air criterion to determine
of whether this release should be treated as a dense gas or not.
The calculation of the denser-than-air criterion should use the formula
given in Section 4.13.1, since this release is considered a continuous release
(Section 5.2.4). The criterion (Cp) is calculated as:
u.
'rel
5-35
-------
Cc = (2.68m/s)
(9.8Q6m/s2)(0.0503kg/s/1.9336kg/m3)
0.01548m
1.9336kg/m3 - 1.183kg/m-
1.183kg/m3
= 1226
Since Cp < 6, a dense gas model should be used.
5.2.14 Release Height
The flange with the leak is 12 ft (3.66 m) above grade. Both the gas
and liquid droplets that form are both assumed to enter the atmosphere at that
altitude.
5.2.15 Ground Surface Temperature
There is no direct information on the ground surface temperature.
Therefore, the ground surface temperature is assumed to be equal to the
ambient temperature, which is given as 74°F or 295.16 °K.
5.2.16 Averaging Time
The averaging time is specified as 5 seconds. This averaging time was
selected to make the model-predicted concentrations comparable to the Lower
and Upper Explosion Limits.
5-36
-------
5.2.17 Meteorology
Wind Speed and Direction
The wind speed and direction are either assumed to be available from on-
site meteorological equipment or else represent average conditions at the
•
site. The wind speed is given as 6 mph (2.68 m/s). The wind direction is
given as southwest (225°). The measurement height is given as 12 ft (6.10 m).
To determine the greatest specific impact, multiple wind speeds may be
required in conjunction with multiple stability classes.
Stability Class
The stability class is not given explicitly; however, it can be
estimated from the information provided using the method described in the
Workbook, Section 3.1.2. In this example, there is 5/8 cloud coverage, and
the release occurs in the early morning. These conditions indicate that the
insolation is slight. The stability class indicated in Table 3-2 of the
Workbook for a wind speed between 2 to 5 m/s at 10 m during a day with slight
insolation is "C." The measurement height in this example is a little over 6
meters.
Surface Roughness Length
In keeping with the suggested value to be used for modeling done for
planning purposes, the surface roughness is assumed to be 0.01 m.
Wind Speed at 10 m Altitude
The wind speed at 10 meters can be estimated using Che surface
roughness, stability class, and wind speed at 6.1 meters. Using the formula
5-37
-------
given in Section 4.17.3 and assuming a stability class of "C," the predicted
wind speed at 10 meters (u) is:
u
= 2.76m/s
which falls within the wind speed limits.for "C" stability class "C"
conditions.
Ambient Temperature. Relative Humidity, and Pressure
The ambient temperature, relative humidity, and pressure are all assumed
to be available from on-site equipment. See the observable data, presented in
Table 5-4.
5.2.18 -Output Definition
The appropriate concentrations to be specified for generating output are
the LEL (Lower Explosive Limit) and UEL (Upper Explosive Limit) concentrations
given in the ethylene oxide data table provided in Appendix B. The actual
impact values to be predicted are listed in Section 7.0.
5.3 TWO-PHASE PRESSURIZED LIQUID EXAMPLE
In this example, chlorine is used in the biological control of cooling
water at a chemical plant. A one-ton container is used to deliver chlorine to
the cooling tower basin (see Figure 5-2). This container is a pressurized,
atmospheric temperature vessel (70°F) containing both liquid- and vapor-phase
chlorine. The vapor pressure (6.86 atmospheres absolute) of chlorine forces
it through a connection to the lower valve on the ton container into the
5-38
-------
Chlorine Ton Container
with Internals
Gas Eductor
C^Gas
Fully Developed
Two-Phase How
Liquid Eductor
Figure 5-2. Two-Phase Pressurized Liquid
5-39
-------
delivery header. The storage pressure can be determined from the triple point
diagram for chlorine provided in Appendix B. A one-meter long, 3/8 inch
diameter section of flexible tubing is used to deliver chlorine to the
delivery header. This tubing is accidentally severed at the delivery header
during routine maintenance. The tubing is two meters above the ground level.
At the time of the accident, the vessel contains 500 Ibs of chlorine in
the liquid phase. As liquid chlorine is released, its pressure drops and some
of the liquid flashes to vapor (adiabatically) within the tubing. The exis-
tence of flashing (two-phase flow) can be determined by the flash diagram for
chlorine given in Appendix B. The chlorine is released horizontally under
fully developed, two-phase flow conditions. Since the release is somewhat
remote from the vessel, the cooling effect of vaporization within the tubing
does not result in cooling of the chlorine within the vessel.
The two-phase release continues until the liquid level within the
container drops below the level of the passage to the liquid eductor (exit
pipe). At this level, 30 Ibs of chlorine remains within the vessel in the
liquid phase. After the liquid level has reached the level of the liquid
eductor, chlorine is emitted as a vapor for the duration of the release. A
schematic of the vessel and tubing dimensions is given in Figure 5-2.
The meteorological conditions at the time of release are:
• South wind at 10 mph;
• Temperature at 70°F;
• Relative humidity of 50%;
• Pressure of 1 atm (sea level);
• 4/8 cloud cover;
• Mid-morning release; and
• Measurement height of 10 m.
The nearest public boundary is assumed to be 100 meters from the release
site. Table 5-5 presents a summary of the model input. The following
sections describe how the input values were developed.
5-40
-------
TABLE 5-5. INPUT SUMMARY FOR THE TWO-PHASE PRESSURIZED LIQUID RELEASE EXAMPLE
Chemical Data (5.3.2)
From Appendix B
Release Type (5.3.3)
Two-phase
Continuous or Instantaneous Release Categories (5.3.4)
Continuous
Emission Rate (5.3.7)
0.3170 kg/s
Release Temperature (5.3.8)
239.09 °K (Tb)
Vapor Fraction (5.3.9)
Fr-1 - 0.1779
Initial Concentration (5.3.10)
f. - 0.0
fw - 0.0
fi - 1.0
Density (5.3.11)
PM1 - 20.11 kg/m3
pa' - 1.191 kg/m3 (at ambient temperature)
Release Diameter or Area (5.3.12)
Diameter - 0.06201 m
Release Buoyancy (5.3.13)
Cp - 1.311 < 6 (use dense gas model)
Release Height (5.3.14)
2 m
Ground Surface Temperature (5.3.15)
Same as ambient
Averaging Time (5.3.16)
15 minutes
5-41
-------
TABLE 5-5. (CONTINUED)
Meteorology (5.3.17)
Wind Speed and Direction
180° @ 10 mph (4.47 m/s)
Measurement height — 10 m
Speed @ 10 m - 4.47 m/s
Stability Class
C
Surface Roughness Length
ZO - 0.01 m
Ambient Temperature, Relative Humidity, and Pressure
Temperature - 70°F (294.3 °K)
Relative Humidity - 50%
Pressure - 1 atm (101325 Pa)
Output Definition (5.3.18)
STEL
Minimum distance of concern - 100 m
Observable Data (5.3.1)
See Table 5-6
5-42
-------
5.3.1 Observable Data
Observable data were given in the description of the release scenario,
A recap of the information in the scenario description is given in Table 5-6.
For modeling done for planning purposes, the direction that the hole is facing
should be assumed to be in the same direction as the wind. This ensures that
maximum ground-level impacts of the released material are predicted by the
model.
5.3.2 Chemical Data Requirements
The chemical data for pure chlorine are in Appendix B.
5.3.3 Release Class
This section describes the calculations required to determine the
release class for this example. The flow diagram in Figure 4-1 is used as a
guide. Many of the calculations required to determine the release class are
also required for other model input. Rather than presenting all calculations
in one section, reference is made to the appropriate section for each
calculation.
In this scenario, there are three releases. The first release is the
two-phase flow of Che liquid escaping the container. The second release is
the constant-rate release of the remaining liquid in the container as it
vaporizes. The third release is the decreasing-rate, gas-phase release from
the container that occurs as the pressure in the container falls to 1 atm.
Only the first release is a two-phase flow, so only it will be described in
this section. The second and third release classes that occur are gas
releases with varying emission rates.
The initial release class was determined by using the flow diagram in
Figure 4-1. The chlorine is stored as a liquid. As the flash diagram in
5-43
-------
TABLE 5-6. OBSERVABLE DATA SUMMARY FOR THE TWO PHASE PRESSURIZED LIQUID
RELEASE EXAMPLE
Release Description
Species: Chlorine
Container: Horizontal cylinder
Diameter:
Length:
Volume:
Total liquid:
Liquid below
liquid eductor:
Temperature:
Pressure:
30 inches (0.762 m)
81.5 inches (2.07 m)
33.34 ft3 (0.944 m3)
500 Ibs
30 Ibs
70°F (294.3 °K)
6.86 atm {absolute) (6.95 x 10s Pa)
Hole site: Connected tubing
Length: 3.281 ft (1 m)
Outer diameter: 3/8 inch (0.009525 m)
Inner diameter: 0.2770 inch (0.007036 m)
Height of tube: 6.562 ft (2 m)
Nearest boundary: 328.1 ft (100 m)
Meteorology:
Temperature:
Relative Humidity:
Pressure:
Wind Speed:
Wind Direction:
Measurement hgt:
Surface Roughness:
Cloud cover:
Time:
70°F (294.3 °K)
50%
1 atm (101325 Pa)
10 mph (4.47 m/s)
South (180°)
32.81 ft (10 m)
0.01 m
4/8
Mid-morning
5-44
-------
Appendix B indicates, when chlorine is stored as a liquid flashing may occur.
Therefore, the value of Frel is required to make the final release class
decision. But, in order to calculate Frel, the value of Trel is required.
This value is obtained by solving the equation described in Section 4.8.2:
pa = 101325Pa exp
3LM[ 1 _ J_
P T T
K I1!) ire
or, for this example:
pa = 101325Pa exp
(2.878xlQ5J/kg)(70.914kg/kmol)
8314J/kmol°K
239.09°K
Setting pa to 1 atm (101325 Pa) leads to a Trel of 239.09 °K (the boiling
point) •
After Trel has been calculated, the value for Frel (see Section 4.9.3)
can be obtained by:
CT - T
= (927.13J/kg°K) (294.3°K - 239.09°K)
2.878xl05J/kg
= 0.1779
Since Frel is between zero and one, according to the flow chart shown in
Figure 4-1, this release is two-phase. Now the flow chart shown in Figure 4-2
can be used to determine the model input.
5-45
-------
5.3.4 Continuous or Instantaneous Release Categories
In this example, the emission rate is 0.3170 kg/s (from Section 5.3.8).
The duration of the two-phase portion of the release is however long it takes
for the liquid to escape the container. There are 500 Ibs of liquid in the
container, but 30 Ibs still remain after the liquid has stopped being
released. Thus, the duration of the release (td)"is:
=673s
=11.2minutes
For a non-source-term model, the value of td must be compared to advec-
tion time to a receptor or maximum downwind distance reached by a concentra-
tion of interest. Using the formula given in Section 4.4 for ttrav, and
assuming a wind speed of 10 mph (4.47 m/s) and a distance of 100 meters, the
advection time is 44.7 seconds. .Since the emission duration is longer than
the time it takes the material to reach the downwind point of interest, the
release is considered continuous.
For a source-term model, the value of td is sufficiently long to treat
this release as a continuous release for most receptors.
Note that, initially, it should be assumed that the release is
continuous until an emission duration (calculated from the emission rate) can
•
be compared to travel times to the receptors of interest, as described in
Section 4.3.
5-46
-------
5.3.5 Release-class-specific Calculations
The flow chart of the calculations of input for a two-phase pressurized
liquid release is shown in Figures 4-2 and 4-6.
5.3.6 Determination of Choked Flow for Gas Releases
This determination is not required since this is a liquid release.
5.3.7 Emission Rate
As described in Section 4.7.3, to calculate the emission rate, the first
task is to compare the ratio of Lp/Le to 1. In this example, the value of Lp
is 1 m and the value of Le is 0.1 m. Since the ratio is 10, a pipe friction
factor (F) must be calculated. This is given by:
2 _ 1
F2 =
DP
1
L , (4X0.0015)(lmy
( 0.007036m
= 0.5397
F = 0.7347
The emission rate can then be calculated by the equation:
E = A0F1
5-47
-------
E = (3.888x10 -5m2)(0.7347)
f(2.878xl05J/kg)(70.91kg/kmol)(6.951xlQ5Pa)>|
{ (8314J/kmol°K)(294.3°K)2 J
( 294.3°K j
927.13J/kg°K
= 0.3170 kg/s
5.3.8 Release Temperature
•
The value of Tr.L was already calculated in Section 5.3.3 in determining
the release class. Its value is 239.09 °K.
5.3.9 Vapor Fraction
The value of Frel was already calculated in Section 5.3.3 in determining
the release class. Its value is 0.1779.
5.3.10 Initial Concentration
Since this is a two-phase flow, it is safe to assume that only chlorine
is being emitted. Based on this assumption the concentrations (described in
Section 4.10) are:
fa - 0.0
fw - 0.0
f, > 1.0
5-48
-------
5.3.11 Density
The release starts as chlorine aerosol and vapor. The water vapor molar
fraction is assumed to be zero. The density of the water vapor component of
the atmosphere is also assumed to be negligible. As explained in Section
4.11, when aerosol is present, the DEGADIS model will require two triplets,
one describing pure air and one describing the chlorine aerosol and vapor
mixture. For one of the triplets, the density of air at the ambient
temperature is required. Assuming that the air behaves as a perfect gas:
P .*£
Ha RTa
_ (101325Pa)(28.9kg/kmol)
(8314J/kmol°K)(294.3°K)
= 1.197 kg/m3
The water vapor density is given as:
_ (101325Pa)(18.02kg/kmol)
(8314J/kmol°K)(294.30K)
= 0.7462 kg/m3
The mole fractions of ambient water vapor (fw') and air (far) are given as:
2353)
100
io
/6.3994-235
\ T-
5-49
-------
= 0.01268
and
fa' = 1 - f£ = 1 - 0.01268 = 0.9873
The apparent molecular weight of the air and water vapor of the ambient
atmosphere is:
w
= (0.9873)(28.9kg/kmol) + (0.01268)(18.02kg/kmol)
= 28.76kg/kmol
1
The ambient density pa' is given as:
Pa =
= / (0.9873)(28.9kg/kmol)
{ (28.76kg/kmol)(1.197kg/m3)
+ (0-Q1268)(18.Q2kg/kmol)
(28.76kg/kmol)(0.7462kg/m3)
= 1.191 kg/m3
For the second triplet, the density of the chlorine aerosol/vapor
mixture at the initial portion of the release' is required. To calculate the
density of the aerosol/vapor mixture (p^ , the densities of the chlorine vapor
and liquid must be known. The temperature used in this calculation is the
release temperature. In this example, the release temperature is the boiling
5-50
-------
point of chlorine. Assuming a perfect gas, the density of the chlorine gas
state (pg) is:
P -
8
= (101325Pa)(70.914kg/kmol)
(8314J/kmol°K)(239.09°K)
= 3.615 kg/m3
The chlorine liquid density at the boiling point is given in Appendix B as
1562 kg/m3. Using the formula given in Section 4.11, the overall chlorine
density is:
pg PI
J 0.1779 +(1 -0.1779) V1
1 (3.615kg/m3 1562kg/m3J
= 20.11kg/m3
Since the release is of pure chlorine, the release density (prei) is
equal to the chlorine density (pL) .
5.3.12 Release Diameter or Area
With a hole diameter of 0.2770 inches . (0.007936 meters), the hole area
is 4.946xlO"5 m2. Since the storage phase is liquid, the storage density (ps)
•
is that of the liquid, 1562 kg/m3 (from Appendix B) . The exit velocity can
then be estimated (from Section 4.12) as:
5-51
-------
0.3170kg/s
(4.964x10 -5m2)(1562kg/m3)
= 4.09m/s
With such a high emission velocity, the release diameter can be
calculated using the high momentum formula given in Section 4.12.2, which
requires the density inside the vessel.
Using the high momentum formula, the release diameter is:
"S
- 000"
A
Prel
7036m
1562kg/m3
20.11kg/m3
= 0.06201m
f •
5.3.13 Release Buoyancy
The release buoyancy is used to determine if a dense gas model should be
used rather Chan a neutral or positive buoyancy model. The first step is to
compare prel to pa' . If prei is less than or equal to p3' , the release should
not be treated as a dense gas release. In this case, prel is 20.11 kg/m3 and
pa is 1.191 kg/m3. Since prel is greater than pa', the next step it to apply
the denser-than-air criterion.
•
Since this is being considered a continuous release (Section 5.3.4"), the
calculation of the denser-than-air criterion should use the formula given in
Section 4.13.1. The criterion (Cp) is given by:
5-52
-------
P»| - Pa
Pa
= (4.47m/s)
(9.806m/s2)(0.3170kg/s/20.11kg/m3)
0.06201m
20.11kg/m3 - U91kg/m-
1.191kg/m3
= 1.311
Since Cp < 6, a dense gas model should be used.
5.3.14 Release Height
The tubing that has been severed is 2 meters above the ground. Since
the released material immediately flashes and all liquid is assumed to be
suspended as an aerosol, the release height is assumed to be the same as the
height where the liquid leaves the container system, i.e. 2 meters.
5.3.15 Ground Surface Temperature
No direct information is given on the ground surface temperature.
Therefore, the ground surface temperature is assumed to be equal to the
ambient temperature, which is 70°F or 294.1 °K.
5.3.16 Averaging Time
The averaging time is specified as 15 minutes, to make the model
predicted concentrations comparable to the STEL (Short Term Exposure Limit)
concentration.
5-53
-------
5.3.17 Meteorology
Wind Speed and Direction
The wind speed and direction either are assumed to be available from on-
site meteorological equipment or else represent average conditions at the
site. The wind speed is given in this example as 10 mph (4.47 m/s). The wind
direction is given as south (180°). The measurement height is given as 10
meters. To determine the greatest specific impact, multiple wind speeds may
be required in conjunction with multiple stability classes.
Stability Class
The stability class is not given explicitly; however, a class can be
estimated from the information provided. Using the method described in the
Workbook, Section 3.1.2. In this example, there is 4/8 cloud coverage, and
the release is in the mid-morning, indicating that the insolation is moderate.
According to Table 3-3 of the Workbook, the stability class for a. wind speed
of 4.47 m/s at 10 m during a day with moderate insolation is estimated as "B"
or "C." For this simulation, the stability class of "C" will be used.
Surface Roughness Length
In keeping with the value suggested for use in modeling done for
planning purposes, the surface roughness is assumed to be 0.01 m.
Wind Speed at 10 m Altitude
Since the wind speed at the 10 meters height is given, it is not
necessary to estimate this value estimated using stability class and surface
roughness.
5-54
-------
Ambient Temperature. Relative Humidity, and Pressure
The ambient temperature, relative humidity, and pressure are all assumed
to be available from on-site equipment. See the observable data, Table 5-6.
5.3.18 Output Definition
The appropriate concentration to be specified when generating output is
the STEL (Short Term Exposure Limit) concentration given in the chlorine data
table in Appendix B. The actual impact values to be predicted are listed in
Section 7.0.
5.4 TWO-PHASE REFRIGERATED LIQUID EXAMPLE
In this example, a process hold-up vessel contains liquid sulfur
dioxide, which is cooled to a temperature below its saturation pressure. The
storage temperature is 122°F (50°C, 323.15 °K) and the storage pressure is 15
atmospheres absolute .(1.52 x 106 Pa). Referring to the sulfur dioxide triple
point diagram contained in Appendix B, this combination of temperature and
pressure place sulfur dioxide in the "liquid only" region of the diagram.
A hole develops in the 2-inch (0.05 m) diameter pipe that transports the
subcooled sulfur dioxide from the storage vessel to a process area. The hole
is 0.5 inches (1.3 x 10"2 m) in diameter, and is located at a point 6.6 ft
(2.0 m) downstream of the vessel. At the time of the release, the vessel
contains 14,520 Ib (6,586 kg) of liquid sulfur dioxide. The feed to the
vessel is isolated at the time of the release, however, the release proceeds
until all of the sulfur dioxide originally contained in the vessel is
depleted. Since the release is remote from the vessel, the temperature and
pressure remain constant within the vessel throughout the release.
The sulfur dioxide is released as a liquid. Since it is subcooled and
pressurized, however, it vaporizes very quickly upon the release. The release
is therefore treated as a gas release with an emission rate equal to that of
5-55
-------
the liquid release rate from the vessel. A schematic ol the re~U - r,e configu-
ration is provided in Figure 5-3.
The meteorological information is:
• Northeast wind at 7 mph;
• Temperature at 64.4°F;
• Relative Humidity of 42%;
• Pressure of 1 atm (sea level);
• Stability class assumed to be "E"; and
• Measurement height of 6 m.
The nearest public boundary is 80 meters from the release point.
A summary listing of the model input is provided in Table 5-7. The
following sections describe how the input values were developed.
5.4.1 Observable Data
The observable data were provided in the scenario description.
Table 5-8 summarizes the information provided in the scenario description.
For planning purposes, it should be assumed that the direction that the hole
is facing is the same direction in which the wind is blowing. This will
ensure that ground level impacts of the released material are maximized.
5.4.2 Chemical Data Requirements
The chemical data for sulfur dioxide are provided in Appendix B.
5.4.3 Release Class
This section describes the calculations required to determine the
release class for this example. The flow diagram in Figure 4-1 is used as a
guide. Many of the calculations required to determine the release class are
also required for other model input. Rather than presenting all calculations
5-56
-------
From
Storage
Refrigerated Sulfur Dioxide Liquid
Two Phase
Refrigerated
Liquid Release
To
Process
Figure 5-3. Two-Phase Refrigerated Liquid
5-57
-------
TABLE 5-7. INPUT SUMMARY FOR THE TWO-PHASE REFRIGERATED LIQUID RELEASE
EXAMPLE
Chemical Data (5.4.2)
From Appendix B
Release Type (5.4.3)
Two-phase
*
Continuous or Instantaneous Release Categories (5.4.4)
Continuous
Emission Rate (5.4.7)
4.154 kg/s
Release Temperature (5.4.8)
263.15 °K (Tb)
Vapor Fraction (5.4.9)
FEBL - 0.2140
Initial Concentration (5.4.10)
fa - 0.0
fw - 0.0
ft - 1.0
Density (5.4.11)
Pr.L - 13.76 kg/m3
pa' - 1.206 kg/m3 (at ambient temperature)
Release Diameter or Area (5.4.12)
Diameter - 0.1308 m
Release Buoyancy (5.4.13)
Cp - 0.5068 < 6 (use dense gas model)
Release Height (5.4.14)
0.3049 m
Ground Surface Temperature (5.4.15)
Same as ambient
Averaging Time (5.4.16)
15 minutes
5-58
-------
TABLE 5-7. (CONTINUED)
Meteorology (5.4.17)
Wind Speed and Direction
45" @ 7 mph (3.13 m/s)
Measurement height -6m
Speed @ 10 m - 3.72 m/s
Stability Class
E
Surface Roughness Length
ZO - 0.01 m
Ambient Temperature, Relative Humidity, and Pressure
Temperature - 64.4°F (291.15 °K)
Relative Humidity - 42%
Pressure - 1 atm (101325 Pa)
Output Definition (5.4.18)
STEL
Minimum distance of concern — 80 m
Observable Data (5.4.1)
See Table 5-8
5-59
-------
TABLE 5-8. OBSERVABLE DATA SUMMARY FOR THE TWO PHASE
RELEASE EXAMPLE
Release Description
Species: Sulfur dioxide
Container: Horizontal cylinder
Diameter: 4 ft (1.220 m)
Length: 12 ft (3.659 m)
Volume: 150.8 ft3 (4.277 m3)
Total S02: 14520 Ibs (6586 kg)
Temperature: 122 °F (323.15 °K)
Pressure: 15 atm (absolute) (1.52 x 106 Pa)
Hole site: Connected tubing
",-;D IVOUID
Length:
Inner diameter:
Area:
Height of tube:
6.562 ft (2 m)
0.50 inch (0.0127 m)
1.267 x 10'* m •
1 ft (0.3049 m)
Nearest boundary: 262.5 ft (80 m)
Meteorology:
Temperature:
Relative Humidity:
Pressure:
Wind Speed:
Wind Direction:
Measurement hgt:
Surface Roughness:
Stability Class:
64.4°F (291.15 °K)
42%
1 atm (101325 Pa)
7 mph (3.13 m/s)
NE (45°)
19.68 ft (6 m)
0.01 m
5-60
-------
in one section, reference is made to the appropriate section for each
calculation.
In this scenario, there are two releases. The first release is the
two-phase flow of liquid escaping from the container. The second release is
the decreasing-rate gas-phase release from the container that occurs as the
pressure in the container falls to 1 atm. All of the liquid escapes from the
container. Only the first release is a two-phase flow, so only it is
discussed in this section. The second release class is a gas release with a
varying emission rate. This second release class needs to be modeled
separately.
The initial release class was determined by using the flow diagram shown
in Figure 4-1. The value of Frel is required to make the final decision
regarding the release class. But in order to calculate Frel, the value of Trel
is required. The value for Trel is obtained by solving the equation described
in Section 4.8.2:
pa = 101325Pa exp
A.M
1
or
pa = 101325Pa exp
(3.8874x10 5J/kg)(64.06kg/kmol)
1
8314J/kmol°K
1
263.15°K
Setting pa to 1 atm (101325 Pa) leads to a Trel of 263.15 °K (the boiling
point, Tb).
After Trel has been calculated, the value for Frel (see Section 4.9.3)
can be obtained by:
5-61
-------
Cpl(Ts - Trel)
rci *\
^»
= (1386.32J/kg°K) (323.15°K - 263.15°K)
3.8874xl05J/kg
= 0.2140
Since Frel is between zero and one, according to the flow chart in Figure 4-1,
this release is two-phase. Now you can use the flow chart in Figure 4-2 to
determine the model input.
5.4.4 Continuous or Instantaneous Release Categories
In this example, the emission rate is 4.154 kg/s (from Section 5.4.8).
The duration of the two-phase portion of the release is however long it takes
for the liquid to escape the container. There are 6586 kg of liquid in the
container and all the liquid is lost. Thus, the duration of the release (td)
is:
t = 6586kg
4.154^
s
= 1585 s
= 26.4 minutes
For a non-source-term model, the value of td must be compared to
advection time to a receptor or maximum downwind distance reached by a
concentration of interest. Using the formula given in Section 4.4 for ttrav,
and assuming a wind speed of 7 mph (3.13 m/s) and a distance of 80 meters, the
advection time is 51.1 seconds. Since the emission duration is longer than
the time it takes the material to reach the downwind point of interest, the
release is considered continuous.
5-62
-------
For a source-term model, the value of td is sufficiently long to treat
this as a continuous release for most receptors.
5.4.5 Release-class-specific Calculations
The flow chart of the calculations of input for a two-phase refrigerated
liquid release is shown in Figures 4-2 and 4-6.
5.4.6 Determination of Choked Flow for Gas Releases
*
This determination is not required since this is a liquid release.
5.4.7 Emission Rate
As indicated in Section 4.7.4, the parameters psv and F must be
calculated before the emission rate can be estimated. The calculations for
these parameters are:
p= 101325Pa exp
R
1 1
= 101325Pa exp
( 1 1
3.8874xl05J/kg)(64.06kg/kmol)
8314J/kmol°K
323.15°K
= 8.386xl05 Pa
and
5-63
-------
F2 =
1 +
D_
1
F =
L + (4)(0.001S)(2m))
( 0.0508 j
0.8089
0.8994
With these parameters, the emission rate can be calculated as:
E = A.
RT
S
= (1.267x10 -4m2)[(2)(0.6)2((1.5199xl06Pa) - (8.386xl05Pa))
(1460.4kg/m3)
(0.8089)
(1386.32J/kg°K)(323.15°K)
f (3.8874xl05J/kg)(64.06kg/kmol)(1.5199xl06Pa) V
( (8314J/kmol°K)(323.15°K) J
= 4.154 kg/s
5.4.8 Release Temperature
The value of Trel was previously calculated in Section 5.4.3 in
determining the release class. Its value is 263.15 °K.
5.4.9 Vapor Fraction
The value of Frel was previously calculated in Section 5.4.3 in
determining the release class. Its value is 0.2140.
5-64
-------
5.4.10 Initial Concentration
Since this is a two-phase flow, it is safe to assume that all that is
being emitted is sulfur dioxide. Based on this assumption the initial
concentrations (described in Section 4.10) are:
f, =0.0
fw - 0.0
fi - 1.0
5.4.11 Density
The release starts as sulfur dioxide aerosol and vapor. The water vapor
molar fraction is assumed to be zero. The density of the water vapor compo-
nent of the atmosphere is also assumed to be negligible. As explained in
Section 4.11, when aerosol is present, the DECADIS model requires two
triplets, one describing pure air and one describing the sulfur dioxide
aerosol/vapor mixture. The density of air at ambient temperature is required
for one of the triplets. Assuming that the air behaves as a perfect gas:
p -
Pa
(101325Pa)(28.9kg/kmol)
(8314J/kmol0 K)(291.15° K)
= 1.210 kg/m3
Similarly, the water vapor density is given as:
5-65
-------
=
Pw
= (101325Pa)(18.02kg/kmol)
(8314J/kmorK)(291.15°K)
= 0.7543 kg/m3
The mole fractions of ambient water vapor (fw') and air (fa') are given as:
- 2353\
- M 1( -)
UooJ
/ 42 \ 6.3994 -
= _Z±_ I 1Q\ 291.15'K
UooJ
= 0.00873
fa' = 1 - fi = 1 - 0.00873 = 0.9913
The apparent molecular weight of the air and water vapor of the ambient
atmosphere is :
= (0.9913)(28.9kg/kmol) + (0.00873)(18.02kg/kmol)
= 28.81 kg/kmol
Finally, the ambient density pA' is given as:
Pa =
\-l
,
w
/ »r/
W
f (Q.9913)(28.9kg/kmol) + (0.00873)(18.02kg/kmc
((28.81kg/kmol)(1.210kg/m3) (28.81kg/kmol)(0.7543kg
U06kg/m3
5-66
-------
The density of the sulfur dioxide aerosol/vapor mixture at the initial
portion of the release is required for the second triplet. To calculate the
density of the aerosol/vapor mixture (PJ^) , the densities of the sulfur dioxide
vapor and liquid must be known. The temperature used in this calculation is
the release temperature. In this example, the release temperature is the
boiling point of sulfur dioxide. Assuming a perfect gas, the density of the
sulfur dioxide gas state (pg) is:
= (101325Pa)(64.06kg/kmol)
(8314J/kmorK)(263.15°K)
= 2.9668 kg/m3
The sulfur dioxide liquid density at the boiling point is given in Appendix B
as 1460.4 kg/m3. The overall sulfur dioxide density is:
g
or
Pg P,
= ( °-2140 + (1-0-2140) V1
1 l,2.9668kg/m3 1460.4kg/m3J
= 13.761 kg/m3
Since the release is of pure sulfur dioxide, the release density (prel)
is equal to the sulfur dioxide density (p^ .
5-67
-------
5.4.12 Release Diameter or Area
Since the storage phase is liquid, the storage density (ps) is that of
the liquid, or 1460.4 kg/m3 (from Appendix B) . The exit velocity can then be
estimated, as discussed in Section 4.12, as:
u =
4.154kg/s
(1.267x10 -
= 22.5m/s ------
With such a high emission velocity, the release diameter can be
calculated using the high momentum formula given in Section 4.12.2. This
formula requires that you know the density inside the vessel.
Using the high momentum formula, the release diameter is:
= 0.0127 m
= 0.1308 m
146ft4
13.761 kg/m
5.4.13 Release Buoyancy
The release buoyancy is used to determine if a dense gas model should be
used rather than a neutral or positive buoyancy model. The first step is to
compare prel to pA' . If prel is less than or equal to pa', the release should
5-68
-------
not be treated as a dense gas release. In this example, prel is 13.76 kg/m3
and pa' is 1.206 kg/m3. Since prel is greater than pa', the next step is to
apply the denser-than-air criterion.
Since this is being considered a continuous release (Section 5.4.4), the
calculation of the denser-than-air criterion should use the formula given in
Section 4.13.1, . The criterion (Cp) is given by:
CP =
= [ (9.806m/s2)(4.154kg/s/13.76kg/m3)
[ 0.1308m
13.76kg/m3 - 1.206kg/Dr
1.206kg/m3
= 0.5068
Since Cp < 6, a dense gas model should be used.
5.4.14 Release Heieht
The tubing that has been severed is 1 foot (0.3049 m) above the ground.
Since the released material immediately flashes and all liquid is assumed to
be suspended as an aerosol, the release height is assumed to be the same as
the height where the liquid leaves the container system, i.e. 0.3049 meter.
5.4.15 Ground Surface Temperature
No direct information is given on the ground surface temperature.
Therefore, the ground surface temperature is assumed to be equal to the
ambient temperature, which is 291.15 °K.
5-69
-------
5.4.16 Averaging Time
The averaging time is specified as 15 minutes. This is to make the
model predicted concentrations comparable to the STEL (Short Term Exposure
Limit) concentration.
•
5.4.17 Meteorology
Wind Speed and Direction
The wind speed and direction either are assumed to be available from on-
site meteorological equipment or else represent average conditions at the
site. The wind speed is given in this example as 7 mph (3.13 m/s) . The wind
direction is given as northeast (45°). The measurement height is given as
6 meters. To determine the greatest specific impact, multiple wind speeds may
be required in conjunction with multiple stability classes.
Stability Class
It is assumed that this simulation is for planning purposes. Thus, the
stability class is set to "E".
Surface Roughness Length
In keeping with the value suggested for use in modeling done for
planning purposes, the surface roughness is assumed to be 0.01 m.
Wind Speed at 10 m Altitude
With the surface roughness, stability class, and wind speed at 6 meters,
the wind speed at 10 meters can be estimated. Using the formula given in
5-70
-------
Section 4.17.3 and assuming a stability class of "E," the predicted wind speed
at 10 meters (u) is:
/i i'a.v./«,<
= (3.13m/s)
I 6m;
= 3.72m/s
Ambient Temperature. Relative Humidity, and Pressure
The ambient temperature, relative humidity, and pressure are all assumed
to be available from on-site equipment. See the observable data, Table 5-8.
5.4.18 Output Definition
The appropriate concentration to be specified when generating output is
the STEL (Short Term'Exposure Limit) concentration given in the sulfur dioxide
data table in Appendix B. The actual impact values to be predicted are listed
in Section 7.0.
5.5 SINGLE-PHASE GAS RELEASE (CHOKED) EXAMPLE
In this example, an anhydrous hydrogen fluoride storage vessel contains
liquid- and vapor-phase hydrogen fluoride. The vessel is being filled from a
rail car through a flexible transfer line when an overpressure/overtemperature
condition develops within the vessel, resulting in the actuation of a rupture
disk within the vapor space of the vessel. The rupture disk is then piped to
a remote area through 13.1 ft (4 m) of pipe with 2.5 inches inside diameter.
At the time of the release, the hydrogen fluoride within the vessel is
at the rupture disk set pressure, which is well above the saturation pressure
of hydrogen fluoride. Immediately upon actuation of the rupture disk, the
5-71
-------
pressure within the vessel drops to its release pressure, whic' ^.01
atmospheres absolute (2.04 x 105 Pa), and the liquid within the \essel remains
subcooled at 165°F (347.0 °K). The vessel temperature and pressure conditions
remain constant throughout the release. However, the pipe is heated to
increase the loft of the effluent. The temperature inside the pipe is 365 °K,
which is the temperature of the release into the air. The release is isolated
45 seconds from the time of the rupture disk actuation. During the release,
hydrogen fluoride is emitted as a vapor only, with the vapor flow at critical
(choked) conditions within the pipe. Figure 5-4 provides a schematic of the
release.
The meteorological conditions at the time of the release are:
• Northwest wind at 14 mph;
• Temperature at 71.6°F;
• Relative Humidity of 45%;
• Pressure of 1 atm (sea level);
• 7/8 cloud cover;
• Early-evening release; and
• Measurement height of 20 ft.
The nearest public boundary is 200 meters from the release point.
Table 5-9 summarizes the model input. The following sections describe how
these input values were developed.
5.5.1 Observable Data
The observable data were discussed in the scenario description, and are
summarized in Table 5-10. For modeling that is done for planning purposes, it
should be assumed that the direction the hole is facing is the same direction
in which the wind is blowing. Most relief valves point vertically, but
assuming a horizontal position will ensure maximum ground level impacts of the
released material.
5-72
-------
Transfer Line
Rupture
Disk
Choked Vapor Phase
HF Release x^.
Hydrogen Fluoride Liquid
Figure 5-4. Single-Phase Gas Release (Choked)
5-73
-------
TABLE 5-9. INPUT SUMMARY FOR THE SINGLE PHASE CHOKED GAS RELEASE EXAMPLE
Chemical Data (5.5.2)
From Appendix B
Release Type (5.5.3)
Single Phase Choked
Continuous or Instantaneous Release Categories (5.5.4)
Ins tantaneous
Emission Rate (5.5.7)
0.8513 kg/s
Release Temperature (5.5.8)
313.6 "K
Vapor Fraction (5.5.9)
Frel - 1.00 (no flash, but all mass emitted as vapor)
Initial Concentration (5.5.10)
fa - 0.0
fw - 0.0
fi - 1.0
Density (5.5.11)
pr-1 - 1.3602 kg/m3
pa — 1.188 kg/m3 (at ambient temperature)
Release Diameter or Area (5.5.12)
Diameter - 4.155 m (instantaneous)
Diameter - 0.06649 m (continuous, choke conditions)
Diameter — 0.0635 m (continuous, expanded)
Release Buoyancy (5.5.13)
Cp - 0.332 > 0.2 (use dense gas model) (instantaneous)
Cp - 2.638 < 6.0 (use dense gas model) (continuous)
Release Height (5.5.14)
12 ft (3.66 m)
Ground Surface Temperature (5.5.15)
Same as ambient
Averaging Time (5.5.16)
15 minutes
5-74
-------
TABLE 5-9. (CONTINUED)
Meteorology (5.5.17)
Wind Speed and Direction
315° @ 14 mph (6.26 m/s)
Measurement height - 20 ft (6.1 m)
Speed @ 10 m - 6.64 m/s
•
Stability Class
D
Surface Roughness Length
ZO -.0.01 m
Ambient Temperature, Relative Humidity, and Pressure
Temperature - 228C (295.15 °K)
Relative Humility - 45% .
Pressure - 1 atm (101325 Pa)
Output Definition (5.5.18)
STEL
Minimum distance of concern - 200 m
Observable Data (5.5.1)
See Table 5-10
5-75
-------
TABLE 5-10. OBSERVABLE DATA SUMMARY FOR THE SINGLE PHASE CHOKED
EXAMPLE
Release Description
Species: Hydrogen Fluoride
Container: Horizontal cylinder
Diameter: 4.9 ft (1.5 m)
Length: 16.4 ft (5 m) Tangent to Tangent
Volume: 187.0 ft3 (5.30 m3)
Total HF: 500 Ibs (2268 kg)
Temperature: 347 °K
Pressure: 2.01 atm (absolute) (2.0397 x 105 Pa)
Hole site: Connected tubing
v, RELEASE
Length:
Inner diameter:
Area:
Height of tube:
Temperature:
Nearest boundary:
Meteorology:
Temperature:
Relative Humidity:
Pressure:
Wind Speed:
Wind Direction:
Measurement hgt:
Cloud cover:
Surface Roughness:
Time:
13.1 ft (4 m)
2.5 inches (0.0635 m)
3.167xlO'3 m2
12 ft (3.66 m)
365 8K
200 m
71.6°F (295.15 °K)
45%
1 atm (101325 Pa)
14 mph (6.26 m/s)
Northwest (315°)
"20.0 ft (6.10 m)
7/8
0.01 m
Early evening
5-76
-------
5.5.2 Chemical Data Requirements
The data for pure HF is in Appendix B.
5.5.3 Release Class
This section describes the calculations required to determine the
release class for this example. The flow diagram in Figure 4-1 is used as a
guide. Note that some of the assumptions made in the process of determining
the release class conflict with the description of the release scenario.
Also, many of the calculations required to determine the release class are
also required for other model input. Rather than presenting all calculations
in one section, reference is made to the appropriate section for each
calculation.
The release is a pure gas release. It must be determined whether the
release is choked or unchoked. As shown in Section 5.5.6, it can be
determined that the pressure at the exit (p*) is greater than the ambient
pressure, which implies that the release is choked. To determine the release
class, it first must be assumed made that the release is single phase. The
calculations will be checked at the end to ensure that this assumption is
valid. The storage temperature just before the release is 365 °K. This value
is assumed to be the storage temperature (Ts) .
The value of T* can then be estimated by the equation:
= (365.0° K)
u 1.3976 + 1
= 304.5° K
The value of T« is less than the critical temperature of hydrogen fluoride,
which is given in Appendix B. So the vapor pressure at T* must now be
5-77
-------
compared Co p*. Using the vapor pressure equation in the Appendix, at
temperature T*=304.5o K, the vapor pressure is 1.524xl05 Pa. According to the
flow chart in Figure 4-1, since this vapor pressure is greater than p*, this
release class is a single-phase gas release. The flow chart in Figure 4-3 can
now be used to determine the model input.
5.5.4 Continuous or Instantaneous Release Categories
The duration of the release (td) is given as only lasting 45 seconds.
For a non-source-term model, the value of td must be compared either to
advection time to a receptor or to the time it takes a concentration of
interest to reach a maximum downwind distance. Using the formula in
Section 4.4 for ttrav, and assuming a wind speed of 14 raph (6.26 m/s) and a
distance of 200 meters, the advection time is calculated to be 63.9 seconds.
This emission duration is shorter than the time required to reach the downwind
point of interest. Therefore, the release should be considered either
instantaneous or of finite duration.
For a source-term model, the value of td is short enough to treat this
as an instantaneous or finite duration release for most receptors.
5.5.5 Release-class-specific Calculations
The flow chart showing the calculations of input for a choked
single-phase gas release is presented in Figures 4-3 and 4-6.
5.5.6 Determination of Choked Flow for Gas Releases
Section 4.6 explains how to determine whether a release is choked or
unchoked. The pressure at the exit point of the gas (p,) is given by:
5-78
-------
.](*-!)
, 1.3976
= 2.0397xl05Pa| ,
U-3976+1J
= 1.0784xl05Pa
where: 7 = Cp / Cv = 1455.57 J/kg °K / 1041.49 J/kg °K = 1.3976.
Since p* > pa - 101325 Pa, the flow in this release is choked.
5.5.7 Emission Rate
Knowing that the release is classified as a single-phase choked gas
release, we can determine its emission rate, using the information provided in
Section 4.7.5.
E = CA
PsPsY
2
Y + 1.
(T-l)
and
PSM
Ps RT
= (2.0397xl05Pa)(20.006kg/kmol)
(8314J/kmol°K)(365°K)
= 1.3447 kg/m3
5-79
-------
So,
E = (0.75)(3.167xlO
(2.0397xl05Pa)(1.3447kg/m 3)1.3976
^ 1.3976+1.
(1.3976+1)
| (1.3976-1)
= 0.8513kg/s
5.5.8 Release Temperature
We can determine the release temperature based on the information
provided in Section 4.8.5.
Trel = TJ1 - 0.85 Y ~
= (365° K) 1 - 0.851
= 313.6°K
Y + i;j
1.3976 - 1
1.3976 + 1
5.5.9 Vapor Fraction
This is a single-phase release; therefore, all of the emission is in the
gas phase.
5.5.10 Initial Concentration
Since this is a single-phase release of pure HF, the initial
concentrations (described in Section 4.10) are:
fa = 0.0
fw =0.0
f, = 1.0
5-80
-------
5.5.11 Density
As discussed in Section 4.11, hydrogen fluoride can associate into
oligomers at high concentrations and temperatures lower than about 50°C(17).
Inside the vessel, this was not a problem because the temperature was kept at
about 92°C. However, when released, the temperature is 313.6 °K (40°C), so
self association can be important. The self association can be considered in
the density calculation by treating the hydrogen fluoride as having an
apparent molecular weight (Mi'). Pure HF at 313.6 °K would have an apparent
molecular weight of about 25.3 kg/kmol(17). Since the apparent molecular
weight increases rapidly with decreasing temperature and a rapid temperature
reduction would be expected immediately upon release, a higher molecular
weight will be assumed by using the value at a temperature about 10 °C lower
than the storage temperature. This gives an apparent molecular weight of
about 35 kg/kmol. The temperature and, therefore, the apparent molecular
weight, is about what it was at choked flow conditions, T*. This value of the
apparent molecular weight, 35 kg/kmol, should be used in calculating density
effects. Since the apparent molecular weight changes as the release is
dispersed, the DEGADIS model should be supplied a pair of triplets created
using prel and pa (calculated below).
The release is entirely in the gas phase of a pure component. The •
release density (prel) is equal to the density of the component (prel) in the
gas phase. The density is then:
_ paM/
Prel Pg RT
KIrel
= (101325Pa)(35kg/kmol)
(8314J/kmorK)(313.6°K)
= 1.3602kg/m3
The ambient air density is also required for buoyancy effects. The dry air
component of the atmosphere is given by the perfect gas law as:
5-81
-------
p =
Pa
= (101325Pa)(28.9kg/kmol)
(83 14J/kmol° K)(295. 15° K)
= 1.193kg/m3
Similarly, the water vapor density is given as:
o
Pw
= (101325Pa)(18.02kg/kmol)
(8314J/kmorK)(295.15°K)
= 0.7441kg/m3
The mole fractions of ambient water vapor (fw') and air (fa') are given as;
ItiaaA 2353 \
(6.3994 - - 1
T>
oo
2353
/
\
6.3994 -
295.15°K
oo
= 0.0120
f ' = 1 - f i = 1 - 0.0120 = 0.9880
w
The apparent molecular weight of the air and water vapor of the ambient
atmosphere is:
5-82
-------
Mj = f.X
= (0.9880)(28.9kg/kmol) + (0.00120)(18.02kg/kmol)
= 28.77kg/kmol
Finally, the ambient density (pa') is given as:
\-l
Pa =
= f (0.988Q)(28.9kg/kmol)
((28.77kg/kmol)(1.193kg/m3)
+ (0.0120)(18.Q2kg/kmol) V1
(28.77kg/kmol)(0.7441kg/m3) J
= 1.188kg/m3
5.5.12 Release Diameter or Area
Since the emission is assumed to be instantaneous, the size of the
release can be estimated from the volume required to hold the emitted mass at
prel. The amount of mass released (Et) can be calculated from the emission
rate (E) and duration of the release as:
Et = E At
= (0.8513kg/s)(45s)
= 38.31 kg
The volume required (V) is then:
5-83
-------
P.d
= 38.31kg
L3602kg/m3
= 28.16 m3
As'suming that the cloud is in the form of a cylinder with a radius equal
to the cloud depth, the diameter can be calculated as:
= 4.155 m
If it is assumed that the release is not instantaneous, the pseudo hole
diameter needs to be calculated. Since this release is choked, there are two
release diameters to be considered. The one that is to be used depends on the
model being used. The first release diameter to consider is that which
represents the choke condition. This diameter replaces the actual hole size
in those models that require hole size. If the actual hole size were used,
the calculated exit velocity would be greater than sonic. The choke point
release diameter from the high momentum formula is then:
\ Prel
5-84
-------
Dre, = 0.0635m
\
1.491kg/m3
1.360kg/m3
= 0.06649m
where pv, is the density at the choked conditions and is given by:
= P»Mi _ (1.0784xl05Pa)(35kg/kmol)
P* RT. (8314J/kmol°K)(304.5°K)
= 1.491 kg/m3
The second release diameter is calculated using the storage density (ps)
rather than the choke density. This diameter is used in those models that
require the after-expansion size of a release. However, because of the
difference in molecular weight of the material in the storage state (20.006)
from when it is released (35), the release density (prel) is larger than ps.
This would lead to a diameter smaller than the actual diameter. Therefore, in
this case, the actual diameter should be used.
5.5.13 Release Buoyancy
From the information provided in Section 4.13.2, the criterion for
determining whether an instantaneous release should be considered denser-than-
air is if:
g(E/Pre,)1/3
Prel ~ Pa
Pa
1
> 0.2
The value of Et must be calculated from the emission rate (E) and the
duration of the release such that:
5-85
-------
Et = E At
= (0.8513kg/s)(45s)
= 38.31 kg
The buoyancy criterion is then:
C =
9.806m/s2(38.31kg/1.3602kg/m3)1/3
(6.26m/s)2
1.3602kg/m3 - 1.188kg/m3)'
(1.188kg/m3)
= 0.332
Since Cp > 0.2, a dense gas model should be used for this example.
If it is assumed that the release is not instantaneous, the buoyancy
criterion would be given by:
u.
~ Pa
D
rel
= (6.26m/s)
Pa
(9.806m/s2)(0.8513kg/s/1.360kg/m3)
0.06649m
1.360kg/m3 - 1.188kg/nr
1.188kg/m3
= 2.638
Since C„ < 6, a dense gas model should be used.
5-86
-------
5.5.14 Release Height
The rupture disk is located at the top of the vessel at a height of 12
ft (3.66 m).
5.5.15 Ground Surface Temperature
No direct information was given on the ground surface temperature.
Therefore, the ground surface temperature is assumed to be equal to the
ambient temperature, which is 22°C (295.15 °K).
5.5.16 Averaging Time
The averaging time is specified as 15 minutes in order to make the
model-predicted concentrations comparable to the STEL (Short Term Exposure
Limit) concentrations.
5.5.17 Meteorology
Wind -Speed and Direction
The wind speed and direction are assumed either to be available from on-
site meteorological equipment or else are representative of average conditions
at the site. The wind speed in this example is 14 mph (6.26 m/s). The wind
direction is northwest (315°). The measurement height is given as 20 ft (6.10
m). To determine the greatest specific impact, multiple wind speeds may be
required in conjunction with multiple stability classes.
•
Stability Class
The stability class is not given explicitly. However, a class can be
estimated from the information provided using the method described in the
Workbook, Section 3.1.2. In this example, there is 7/8 cloud coverage and the
release is in the mid-morning, indicating that the insolation is slight.
5-87
-------
Based on Table 3-2 of the Workbook, the stability class for a wind speed of
6.26 m/s during the day with slight insolation would be "D."
Surface Roughness Length
In keeping with the value suggested for modeling done for planning
purposes, the surface roughness is assumed to be 0.01 meters.
Wind Speed at 10 m Altitude
With the surface roughness, stability class, and wind speed at 6.1
meters, the wind speed at 10 meters can be estimated. Using the formula in
Section 4.17.3 and assuming a stability class of "D," the predicted wind speed
at 10 meters (u) is:
= (6.26m/s) -
\6.lm)
= 6.64m/s
which falls within the wind speed limits for "D" stability.
Ambient Temperature, Relative Humidity, and Pressure
The ambient temperature, relative humidity, and pressure are all assumed
to be available from on-site equipment. See the observable data, Table 5-10.
•
5.5.18 Output Definition
The appropriate concentration level to be used for generating output is
the STEL (Sho'rt Term Exposure Limit) concentration given in the hydrogen
5-88
-------
fluoride data table in Appendix B. The actual impact values to be predicted
are listed in Section 7.0.
5.6 SINGLE-PHASE GAS RELEASE (UNCHOKED) EXAMPLE
In this example, an absorption tower is employed to recover hydrogen
chloride vapor from a waste gas stream in a process facility. Vapor phase
hydrogen chloride and air are fed to the tower, which uses water as an
absorption media to remove all of the hydrogen chloride during normal
operations. A pump failure in the water supply system to the tower results in
a temporary decrease in the capacity of the water supply to the tower. The
resulting decrease in absorption efficiency results in the release of 4,250
actual cfm (2.01 m3/s) of a mixture of hydrogen chloride, water vapor, and air
for twelve minutes before shut off.
The stack temperature (at the point of release) is 120°F (322.0 °K). At
this temperature, the emission is saturated with water vapor (12.4% water).
The volume (molar) percentages of hydrogen chloride and air are 72.9% and
14.7%, respectively. A mist eliminator removes all of the entrained liquid
phase materials at the top of the stack, and only vapor is released to the
atmosphere. The stack is 34 inches (0.86 m) in diameter and 28 ft (8.54 m)
tall. A schematic of the release is provided in Figure 5-5.
The meteorological conditions at the time of release are:
• East wind at 2.2 mph;
• Temperature of 64°F;
• Relative Humidity of 36%;
• Pressure of 1 atm (sea level);
• 3/8 cloud cover;
• Early-morning release; and
• Measurement height of 10 m.
The nearest public boundary is 525 ft from the release point.
5-89
-------
Vapor Phase
Hydrogen Chloride Release
with Air and Water Vapor
Hydrogen Chloride & Air:
Tower Water Supply
Absorption Tower
Figure 5-5. Single-Phase Gas Release (Unchoked)
5-90
-------
Table 5-11 summarizes the model input for this example. The following
sections describe how the input values were developed.
5.6.1 Observable Data
Observable data were presented in the scenario description and are
summarized in Table 5-12.
5.6.2 Chemical Data Requirements
The data for pure hydrogen chloride are in Appendix B. However, this
emission is a mixture of hydrogen chloride, water, and air, all in the gas
phase. Therefore, the emission may be treated as a pseudo chemical consisting
of this mixture. Some models can handle the mixture internally by specifying
the initial concentrations of each of the components. These initial
concentrations are calculated in Section 5.6.10. In other models, it is
assumed that the emission is always of a pure chemical and so the emission
must be of the pseudo chemical.
Since this release contains only the gas phase, the only parameters
required to describe the pseudo chemical are the apparent molecular weight and
the gas-phase specific heat. The method for calculation of the parameters for
a pseudo chemical is given in Section 4.2.
The apparent molecular weight is calculated in Section 5.6.11 using the
molar fractions of the components, which are given in Section 5.6.10. The
specific heats are given in units of J/kg °K, which are mass units, so the
calculation needs mass fractions. The relationship between mass fraction and
molar fraction is given.by:
WjM = f.Mj
5-91
-------
TABLE 5-11. INPUT SUMMARY FOR THE SINGLE PHASE UNCHOKED GAS RELEASE EXAMPLE
Chemical Data (5.6.2)
From Appendix B
Release Type (5.6.3)
Single Phase Unchoked
Continuous or Instantaneous Release Categories (5.6.4)
Continuous
Emission Rate (5.6.7)
2.5151 kg/s
Release Temperature (5.6.8)
322.04 °K
•
Vapor Fraction (5.6.9)
Frel - 1.00 (no flash, but all mass emitted as vapor)
Initial Concentration (5.6.10)
fa - 0.1467
fw - 0.1238
ft - 0.7295
Density (5.6.11)
Prel - 1.2513 kg/m3
pa - 1.208 kg/m3 (at ambient temperature)
Release Diameter or Area (5.6.12)
Diameter - 0.8636 m
Release Buoyancy (5.6.13)
Cp — 1.048 < 6 (use dense gas model)
Release Height (5.6.14)
8.54 m
Ground Surface Temperature (5.6.15)
Same as ambient
Averaging Time (5.6.16)
30 minutes
5-92
-------
TABLE 5-11. (CONTINUED)
Meteorology (5.6.17)
Wind Speed and Direction
90° <§ 2.2 mph (0.98 m/s)
Measurement height - 10 m
Speed @ 10 m - 0.98 m/s
Stability Class
B
Surface Roughness Length
ZO - 0.01 m
Ambient Temperature, Relative Humidity, and Pressure
Temperature - 64°F (290.9 °K)
Relative Humidity - 36%
Pressure - 1 atm (101325 Pa)
Output Definition (5.6.18)
IDLH
Minimum distance of concern - 525 ft (160.1 m)
Observable Data (5.6.1)
See Table 5-12
5-93
-------
TABLE 5-12. OBSERVABLE DATA SUMMARY FOR THE SINGLE PHASE UNCHOKED GAS RELEASE
EXAMPLE
Release Description
Species: Hydrogen chloride
Container: Stack
Diameter:
Height:
Volume rate:
Temperature:
34 in (0.8636 m)
28 ft (8.54 m)
4250 ft3/min (2.01 m3/s)
120°F (322.04 °K)
Nearest boundary: 525 ft (160.1 m)
Meteorology:
Temperature:
Relative Humidity:
Pressure:
Wind Speed:
Wind Direction:
Measurement hgt:
Surface Roughness:
Cloud cover:
Time:
64°F (290.9 °K)
36%
1 atm (101325 Pa)
2.2 mph (0.98 m/s)
East (90°)
32.81 ft (10 m)
0.01 m
3/8
Early morning
5-94
-------
so, for this example,
= (0.1467)(28.9kg/kmol)
(33.07kgflonol)
= 0.1282
M
= (0.1238)(18.02kg/kmol)
(33.07kg/kmol)
= 0.06746
WHd
M
= (0.7295)(36.46kg/kmol)
(33.07kg/kmol)
= 0.8043
The specific heat is then calculated as:
5-95
-------
= (0.1282)(1004J/kg°K)
+ (0.06746)(4180J/kg°K)
+ (0.8043)(799.81J/kg°K)
=1054J/kg°K
5.6.3 Release Class
The release class is determined following the flow chart in Figure 4-
1. First, since this is a stored gas, we must determine whether the release
is choked or unchoked. In Section 5.6.6, it is determined that the exit
velocity is much smaller than sonic velocity. Since choke flow only occurs at
or near sonic velocity, the flow in this release is unchoked.
The next step in determining if the release is two phase or not is to
compare the vapor pressure of the HCl at the release temperature to the
release pressure. The release pressure is 1 atm. The release temperature is
given as 322 °K. The boiling-point of pure HCl is 188.1 °K, which means the
vapor pressure should be well in excess of 1 atm at 322 °K. According to the
flow chart shown in Figure 4-1, since the HCl vapor pressure is greater than
the release pressure, the release is single phase. Thus, it is concluded that
this release scenario should be treated as an unchoked single-phase gas
release. The flow chart in Figure 4-3 should be used to determine the model
input.
5.6.4 Continuous or Instantaneous Release Categories
The duration of the release (td) is given as 12 minutes.
For a non-source-term model, the value of td must be compared either to
the advection time to a receptor or to the time it takes a concentration of
interest to reach maximum downwind distance. Using the formula in Section 4.4
5-96
-------
for ttrav, and assuming a wind speed of 2.2 mph (0.98 m/s) and a distance of
525 ft (160.1 m) , the advection time is calculated as 327 seconds. Since the
emission duration is longer than the time to reach the downwind point of
interest, the release is considered continuous.
For a source- term model, the value of td is also sufficiently long to
treat this as a continuous release for most receptors .
5.6.5 Release - class - specific Calculations
The flow chart showing the calculations of input for an unchoked
single-phase gas release is given in Figures 4-3 and 4-6.
5.6.6 Determination of Choked Flow for Gas Releases
We can determine whether or not the flow is choked by calculating the
exit velocity. Since this release is a stack release, the formula using
pressure cannot be used. However, the exit velocity (ve) may be calculated
from the stack area (AQ) and the volumetric release rate (Vf ) . So,
= 2.01m3/s
0.586m2
= 3.43 m/s
Since va is very much less than sonic velocity, we can consider the
stack emission to be unchoked; choked flow occurs when the exit velocity is at
or near sonic velocity.
5-97
-------
5.6.7 Emission Rate
The total mass emission can be calculated from the specified volumetric
release rate (Vf) and the release density (pral). Therefore,
E_ \T
~ Vf Pld
= (2.01m3/s)(12513kg/m3)
= 2.5151 kg/s
5.6.8 Release Temperature
The release temperature (Trel) at the stack exit is specified as 120°F
(322.04 °K). It is a measured value for this example.
5.6.9 Vapor Fraction
The emission is completely in the vapor phase.
5.6.10 Initial Concentration
From Section 4.10, the water vapor fraction can be estimated if the
temperature is known. Since the release temperature (Trel) is 322.04 °K, the
saturated molar fraction of the water vapor (es) at 1 atm of pressure is:
2353
= 0.1238
The fractions of HC1 and air emitted during the release are specified as
0.7295 and fa - 0.1467, respectively.
5-98
-------
5.6.11 Density
The densities are calculated using the formulas presented in
Section 4.11. Since this release is completely in vapor phase, the release
density can be calculated assuming an ideal gas.
Pw=
Trel - 322.04 °K
Mi - 36.46 kg/kmol
M* - 18.02 kg/kmol
Ma - 28.9 kg/kmol
Therefore,
= (101325Pa)(36.46kg/kmol)
1 (8314J/kmol°K)(322.040K)
= 1.3798 kg/m3
5-99
-------
Pw
= (101325Pa)(18.02kg/kmol)
(8314J/kmol°K)(322.04°K)
= 0.6819 kg/m3
= (101325Pa) (28.9kg/kmol)
Pa (8314J/kmol°K) (322.04°K)
= 1.0937 kg/m3
The density at release (prei) can be calculated by:
.1-1
where :
Therefore ,
MiP.
= faMa H- fwMw + f;M.
= (0.1467)(28.91^/kmol) + (0.1238)(18.02kg/kmol)
+ (0.7295)(36.46kg/kmol)
= 33.07 kg/kmol
Ptd "
(0.1467)(28.9kg/kmol)
(33.07kg/kmol)(1.0937kg/m3)
(0.1238)(18.02kg/kmol)
(33.07kg/kmol)(0.6810kg/m3)
(0.7295)(36.46kg/kmol) I
(33.07kg/kmol)(1.3798kg/m3)
5-100
-------
= 1.2513 kg/m3
The density of the ambient air also needs to calculated so that buoyancy
can be determined. The ambient air density can be derived from the perfect
gas law at the ambient temperature, 290.9 °K:
p =
RTa
(83Tl4J/kmor1C)(290.90K)
,_ - 1 30.00736
= l^llkg/m3
= 0.9926
Similarly, the water vapor density is given as:
P.MW
Pw =
RTa
= (101325Pa)(18.02kg/kmol)
(8314J/kmol°K)(290.9°K)
= 0.7549kg/m3
The mole fractions of ambient water vapor (fw') and air (fa') are given as:
l\
oo
= 0.00736
The apparent molecular weight of the air and water vapor of the ambient
atmosphere is:
5-101
-------
M{ = f.X + fX
= (0.9926)(28.9kg/kmol) + (0.00736)(18.02kg/kmol)
= 28.82 kg/kmol
Finally, the ambient density pa' is given as:
= f (0.9926)(28.9kg/kmol)
( (28.82kg/kmol)(1.211kg/m3)
+ (0.00736)(18.02kg/kmol) V1
(28.82kg/kmol)(0.7549kg/m3) j
= 1.208kg/m3
5.6.12 Release Diameter or Area
The temperature inside the stack is the same as the temperature at the
exit. The pressure is atmospheric both inside and outside the stack. Because
this is an unchoked stack release of a gas, the density can be calculated from
the perfect gas law. Since the in-stack and at-exit temperatures are the
same, the density inside the stack (ps) is the same as the density of the
release (prel) .
The exit velocity is calculated in Section 5.6.6-as 3.43 m/s. With such
a high emission velocity, the release diameter can be calculated using the
high momentum formula in Section 4.12.2. However, the formula modifies the
actual hole size by the square root of the ratio of the release density inside
the stack to the release density outside the stack. Since these two densities
are equal, the release diameter (Drel) is equal to the actual diameter (Ds) .
The formula for the release diameter requires the density inside the vessel.
5-102
-------
The area of interest is the stack, which has a diameter of 34 inches
(0.8636 m).
5.6.13 Release Buoyancy
The release buoyancy is used to determine if a dense gas model should be
used rather than a neutral or positive buoyancy model. The first step is to
compare prel to pa' . If prel is less than or equal to pa' , the release is not
to be treated as a dense gas release. In this case, prel is 1.2513 kg/m3 and
pa' is 1.208 kg/m3. Since prel is greater than pa' , the next step is to
calculate the denser- than-air criterion to determine whether this release
should be treated as a dense release gas or not.
The calculation of the denser- than- air criterion should use the formula
in Section 4.13.1, since this is being considered a continuous release
(Section 5.6.4). The criterion (Cp) is given by:
c u-
0.8636m
1.2513kg/m3 - 1.208kg/m:
1.208kg/m3
= 1.048
Since Cp < 6, a dense gas model should be used.
5.6.14 Release Height
The stack height is 28 ft (8.54 m), which is where the emission enters
the atmosphere.
5-103
-------
5.6.15 Ground Surface Temperature
No direct information is directly given on the ground surface
temperature. Therefore, the ground surface temperature is assumed to be equal
to the ambient temperature, which is 64°F (290.9 °K).
5.6.16 Averaging Time
The averaging time is specified as 30 minutes to make the model
predicted concentrations comparable to the IDLH (Immediately Dangerous to Life
and Health) concentration.
5.6.17 Meteorology
Wind Speed and Direction
The wind speed and direction either are assumed to be available from on-
site meteorological equipment or else represent average conditions at the
site. In this example, the wind speed is 2.2 mph (0.98 m/s). The wind
direction is east (90°). The measurement height is given as 32.81 ft, or 10
m. To determine the greatest specific impact, multiple wind speeds may be
required in conjunction with multiple stability classes.
Stability Class
The stability class is not explicitly given. However, the stability
class can be estimated from the information provided using the method
described in the Workbook, Section 3.1.2. In this example, there is 3/8 cloud
coverage and the release takes place in the early morning, indicating that
insolation is slight. Based on Table 3-2 in the Workbook, the stability class
estimated for a wind speed to be 0.98 m/s at 10 m during the day with slight
insolation would be "B."
5-104
-------
Surface Roughness Length
In keeping with the value suggested for modeling done for planning
purposes, the surface roughness is assumed to be 0.01 m.
Wind Speed at 10 m Altitude
Since the wind speed at 10 meters height is given, it does not need to
be estimated using stability class and surface roughness.
Ambient Temperature. Relative Humidity, and Pressure
The ambient temperature, relative humidity, and pressure are all assumed
to be available from on-site equipment. See the observable data, Table 5-12.
5.6.18 Output Definition
The appropriate concentration level to use to generate output should be
the STEL (Short Term Exposure Limit) concentration given in the hydrogen
chloride data table in Appendix B. The actual impact values to be predicted
are listed in Section 7.0.
5.7 SINGLE-PHASE LIQUID RELEASE (HIGH VOLATILITY) EXAMPLE
In this example, ethylene oxide is stored in its liquid phase within a
vessel by maintaining the ethylene oxide at a temperature just below its
boiling temperature. A 0.25 inch hole develops near the bottom of the tank.
At the time of the release, the ambient pressure is 1 atm, and the ethylene
oxide within the vessel is at its saturation pressure of 1 atm and is also at
its normal boiling temperature of 51.3°F (283.85 °K). The liquid level within
the 3.5 meter diameter tank is 9.2 ft (2.8 m) above grade level. The hole is
located 0.15 ft (0.5 m) above grade level.
5-105
-------
Liquid ethylene oxide is released under the column head pressure of
vessel contents until the liquid level reaches that of the hole. The ambient
temperature at the time of the release is 82.4"F (301.15 °K) . Since the boil-
ing point of ethylene oxide (equal to the release temperature of 283.85 °K) is
below the ambient temperature, the liquid vaporizes very rapidly as it leaves
the vessel. The release is therefore treated as a vapor phase release at a
rate equaling the liquid release rate from the vessel. A schematic of the
release is provide in Figure 5-6.
The meteorological conditions at the time of release are:
• North wind at 2.0 m/s;
• Temperature at 28°C;
• Relative Humidity of 50%;
• Pressure of 1 atm (sea level);
• Stability class assumed to be "E"; and
• Measurement height of 10 m.
The nearest public boundary is 100 meters from the release point.
Table 5-13 summarizes the model input for this example. The following
sections describe how the input values were developed.
5.7.1 Observable Data
The observable data was given in the scenario description. Table 5-14
summarizes this information. For modeling done for planning purposes, the
direction that the hole is facing should be assumed to be in the same
direction in which the wind is blowing. This will ensure maximum transport of
the released material.
5/7.2 Chemical Data Requirements
The chemical data for pure ethylene oxide is given in Appendix B.
5-106
-------
Refrigerated
Ethylene Oxide
Refrigerated Liquid Release
(Immediate Vaporization)
Figure 5-6. Single-Phase Liquid Release (High Volatility)
5-107
-------
TABLE 5-13. INPUT SUMMARY FOR THE HIGH-VOLATILITY, SINGLE PHASE LIQUID RELEASE
EXAMPLE
Chemical Data (5.7.2)
From Appendix B
Release Type (5.7.3)
Two-phase
Continuous or Instantaneous Release Categories (5.7.4)
Continuous
Emission Rate (5.7.7)
0.1220 kg/s
Release Temperature (5.7.8)
283.85 °K (Tb)
Vapor Fraction (5.7.9)
Frel — 0.0 (from flash, but assumed to boil/evaporate immediately)
Initial Concentration (5.7.10)
fa - 0.0
fw - 0.0
ft - 1.0
Density (5.7.11)
Pret - 1.891 kg/m3
pa - 1.162 kg/m3 (at ambient temperature)
Release Diameter or Area (5.7.12)
Diameter - 0.1372 m
Release Buoyancy (5.7.13)
Cp - 1.403 < 6 (use dense gas model)
Release Height (5.7.14)
0.5 m
Ground Surface Temperature (5.7.15)
Same as ambient
Averaging Time (5.7.16)
30 minutes
5-108
-------
TABLE 5-13. (CONTINUED)
Meteorology (5.7.17)
Wind Speed and Direction
0° @ 2 m/s
Measurement height — 10 m
Speed @ 10 m - 2 m/s
Stability Class
E
Surface Roughness Length
ZO - 0.01 m
Ambient Temperature, Relative Humidity, and Pressure
Temperature - 301 °K
Relative Humidity - 50%
Pressure - 1 atm (101325 Pa)
Output Definition (5.7.18)
IDLH
Minimum distance of concern - 100 m
Observable Data (5.7.1)
See Table 5-14
5-109
-------
TABLE 5-14. OBSERVABLE DATA SUMMARY FOR THE HIGH-VOLATILITY, SINGLE PHASE
LIQUID RELEASE EXAMPLE
Release Description
Species: Ethylene oxide
Container: Vertical cylinder
Diameter: 3.5m
Height:
Volume:
Total EO: 23780 kg
Temperature: 283.85 °K (Tb)
Pressure: 1 atm (absolute) (101325 Pa)
Hole site: In vessel
Liquid above hole:
Inner diameter:
Area:
Height of hole:
2.3 m
0.25 inch (6.35 x 10'3 m)
3.167 x 10"5
0.5 m
m
Nearest boundary: 328.1 ft (100 m)
Meteorology:
Temperature:
Relative Humidity:
Pressure:
Wind Speed:
Wind Direction:
Measurement hgt:
Surface Roughness:
Stability Class:
28°C (301 °K)
50%
1 atm (101325 Pa)
2 m/s
North (0°)
10 m
0.01 m
E
5-110
-------
5.7.3 Release Class
In this scenario, only one release occurs. The liquid pours out of the
container until the liquid level is at the hole level. Since the liquid is
stored at its boiling point, the gas pressure inside the container is at
one atmosphere. No significant gas release would be expected since the inside
and outside pressures are the same, unless the refrigeration stopped. In that
case, the remaining liquid may boil and the resulting vapor would escape.
The release class is determined using the flow diagram in Figure 4-1.
The ethylene oxide is stored as a liquid. The flash diagram in Appendix B
indicates that flashing will not occur since the storage temperature is equal
to the boiling point and the flash diagram curve begins its upward branch just
at the boiling point.
Since no flash will occur, the next step is to decide if the boiling
point is above or below the ambient temperature. The boiling point is 283.85
°K. The ambient temperature is 301 °K. Since the boiling point is below the
ambient temperature, the material released will boil/evaporate as soon as it
reaches the atmosphere. From the flow chart in Figure 4-1, this scenario is
defined as a high-volatility liquid release. The flow chart in Figure 4-5
should be used to determine the model input.
5.7.4 Continuous or Instantaneous Release Categories
In this example, the emission rate is 0.1220 kg/s (from Section 5.7.8).
The duration of the release is how long it takes for the liquid to escape the
container. There is 23,780 kg liquid in the container at a depth of 2.8 m.
The hole is at a height of 0.5 m above the container bottom. This means that
the fraction of the liquid that remains in the container after the liquid
stops being released is 0.5/2.8. The duration of the release (td) is then
calculated as:
5-111
-------
t = (1 - (0.5m/2.8m)))23780kg
OJ
td = 160111s
• = 44.5 hours
For a non-source-term model, the value of td must be compared either to
the advection time to a receptor or to the time it takes the concentration of
interest to reach the maximum downwind distance. Using the formula in
Section 4.4 for ttrav, and assuming a wind speed of 2 m/s and a distance of
100 meters, the advection time is calculated to be 327 seconds. Since the
emission duration is longer than the time required to reach the downwind point
of interest, the release is considered continuous.
For a source-term model, the value of td is also sufficiently long to
treat this as a continuous release for most receptors.
5.7.5 Release-class-specific Calculations
The flow chart showing the calculations of input for a single-phase high
volatility liquid release is presented in Figures 4-4 and 4-6.
5.7.6 Determination of Choked Flow for Gas Releases
This determination is not required since this scenario involves a high
volatility liquid release.
5-112
-------
5.7.7 Emission Rate
From Section 4.7.7, to calculate the emission rate, the pressure at the
exit hole first must be calculated. This is done by the equation:
or, in this example,
Ph = 1.01325xl05Pa + (882.70kg/m3)(9.806m/s2)(2.3m)
a
= 1212xl05Pa
Since the storage pressure (psv) is the same as the ambient pressure (pa) , 1
atm, the maximum is also 1 atm.
The term ft (as defined in Section 4.7.6) approaches zero in this release
from a tank since the area of the hole is very much smaller than the area of
the tank. This means that K is equal to C (0.65). The equation in
Section 4.7.7 can then be written as;
^
•c - r* A. n ~ /"•» _ ~ YI 2
or
E = (0.65)(3.167xlO-5m2)
[2(882.70kg/m3)((1.2123xl05Pa) - (1.01325xl05Pa))]2
= 0.1220 kg/s
5.7.8 Release Temperature
For this release class, it is assumed that, as soon as the liquid leaves
the container, it vaporizes. Therefore, the release temperature can be
assumed to be the normal boiling point of the chemical.
5-113
-------
5.7.9 Vapor Fraction
The value of Frel was previously determined to be zero in the
determination of the release class (Section 5.7.3). However, it is also
assumed that, after the liquid gets into the atmosphere, it will
boil/evaporate immediately without forming a pool or liquid droplets, since
the boiling point is lower than the ambient temperature and storage pressure
is near ambient pressure.
5.7.10 Initial Concentration
Since this is a high volatility flow, it is safe to assume that only
ethylene oxide is being emitted. Based on this assumption the initial
concentrations (described in Section 4.10) are:
fa - 0.0
£w - 0.0
fi - 1.0
5.7.11 Density
The densities required can be calculated using the formulas discussed in
Section 4.11. The release starts as pure ethylene oxide gas. The water vapor
molar fraction is assumed to be zero. The density of the water vapor
component of the atmosphere is also assumed to be negligible. The density of
air at the ambient temperature is required in order to calculate density
effects. Assuming that the air behaves as a perfect gas:
Pa " RTa
5-114
-------
= (101325Pa)(28.9kg/kmol)
P* (8314J/kmol°K)(301.15°K)
= 1.170kg/m3
Similarly, the water vapor density is given as:
P.M,
Pw =
RT
a
= (101325Pa)(18.02kg/kmol)
(8314J/kmorK)(301.15°K)
= 0.7293kg/m3
The mole .fractions of ambient water vapor (fwf) and air (fa') are given as:
63994 - 2353
\iooj
= 0.01927
f' = 1 - f;
Aa A Lw
= 1 - 0.01927
= 0.9807
The apparent molecular weight of the air and water vapor of the ambient
atmosphere is:
5-115
-------
= (0.9807)(28.9kg/kmol) + (0.01927)(18.02kg/kmol)
= 28.69kg/kmol
Finally, the ambient density pa' is given as:
p/ =
= / (0.9807)(28.9kg/kmol)
( (28.69kg/kmol)(1.170kg/m3)
+ (0.01927)(18.02kg/kmol) V1
(28.69kg/kmol)(0.7293kg/m3) J
= U62kg/m3
Because the ethylene oxide is assumed to vaporize instantaneously upon
release, the release density can be calculated using the perfect gas law:
= (101325Pa)(44.053kg/kmol)
(8314J/kmol°K)(283.85°K)
= 1.8914kg/m3
5.7.12 Release Diameter or Area
With a hole diameter of 0.2770 in (0.007936 m) , the hole area is
3.167xlO~5 m2. Since the storage phase is liquid, the storage density (ps) is
the same as that of the liquid, or 883 kg/m3 (from Appendix B) . The exit
velocity (from Section 4.12) can then be estimated as:
5-116
-------
u =
Q.1220kg/s
(3.167x10
= 4363m/s
With such a high emission velocity, the release diameter can be
calculated using the high momentum formula given in Section 4.1,2.2. The
release diameter calculated using the high momentum formula is:
= 0.00635m
N
883kg/m
= 0.1372m
5.7.13 Release Buoyancy
The release buoyancy is used to determine if a dense gas model should be
used rather than a neutral or positive buoyancy model. The first step is to
compare prel to pa' . If prel is less than or equal to pa' , the release should
not be treated as a dense gas release. In this case, prel is 1.8914 kg/m3 and
pa' is 1.162 kg/m3. Since pXBi is greater than pa', the next step is to apply
the denser-than-air criterion.
The calculation of the denser-than-air criterion should be based on the
formula given in Section 4.13.1, since this is being considered a continuous
release (Section 5.7.4). The criterion (Cp) is given by:
5-117
-------
cp =
id I "a
= (20m/s)
(9.806nVs2)(0.1220kg/^1.8914Jcg/m3)
0.1372m
- U62kg/m-
U62kg/m3 '
= 1.403
Since Cp < 6, a dense gas model should be used.
5 . 7 . 14 Release Height
The hole in the tank is 0.5 meters above the ground. Since the released
liquid is assumed to vaporize immediately upon release, the release height is
assumed to be the same as the height where the liquid leaves the container
system, or 0.5 meters above the ground.
5.7.15 Ground Surface Temperature
No information on the ground surface temperature is given. Therefore,
the ground surface temperature is assumed to be equal to the ambient
temperature, 301.15 °K.
5.7.16 Averaging Time
The averaging time is specified as 30 minutes to make the
model-predicted concentrations comparable to the IDLH (Immediately Dangerous
to Life and Health) concentrations.
5-118
-------
5.7.17 Meteorology
Wind Speed and Direction
The wind speed and direction are assumed either to be available from
on-site meteorological equipment or representative of average conditions at
the site. In this example, the wind'speed is given as 2 m/s. The wind
direction is north (0°). The measurement height is given as 10 meters. To
determine the greatest specific impact, multiple wind speeds may be required
in conjunction with multiple stability classes.
Stability Class
It is assumed that this simulation is for planning purposes. Therefore,
the stability class is set to "E".
Surface Roughness Length
In keeping with the value suggested for modeling done for planning
purposes, the surface roughness is assumed to be 0.01 m.-
Wind Speed at 10 m Altitude
Since the wind speed at 10 meters height is given, it does not have to
be estimated using stability class and surface roughness.
Ambient Temperature. Relative Humidity, and Pressure
The ambient temperature, relative humidity, and pressure are all assumed
to be available from on-site equipment. See the observable data, Table 5-14.
5-119
-------
5.7.18 Output Definition
The appropriate concentrations to be specified when generating output
are the three ERPG (Emergency Response Planning Guideline) concentrations
given in the ethylene oxide data table in Appendix B. The actual impact
values to be predicted are listed in Section 7.0.
•
5.8 SINGLE-PHASE LIQUID RELEASE (LOW VOLATILITY) EXAMPLE
In this example,'an ambient temperature vessel is used to store
30% weight hydrochloric acid in its liquid phase. The vessel is a 12-foot
diameter tank with the liquid level 9 feet (2.7 m) above grade. A 0.75-inch
tank glass connection is broken at the vessel wall near the bottom of the
vessel, 1 foot (0.3 m) above grade level. The resulting hole is 0.75 inches
(0.0191 m) in diameter. The storage temperature is equal to the ambient
temperature, which is 658F (291.5 °K). The tank is located within a 576 ft2
(53.54 m2) diked area. The dike is designed to contain the entire contents of
the vessel when full.
The 30% weight hydrochloric acid is released as a liquid under its
column head pressure within the vessel. Since the boiling point of the bulk
liquid hydrochloric acid (370.2 °K) is above ambient temperature (291.5 °K),
the liquid forms a pool within the dike from which evaporation occurs. Figure
5-7 provides a schematic of the release.
The meteorological conditions at the time of the release are:
Northeast wind at 5 mph;
Temperature at 65°F;
Relative Humidity of 58%;
Pressure of 1 atm (sea level);
3/8 cloud cover;
Late-afternoon release; and
Measurement height of 35 ft.
The nearest public boundary is 450 feet from the release point.
5-120
-------
Saturated 30% Hydrochloric Acid
at Ambient Temperature
Low Volatility Spill
Contained in Dike
Liquid Spill
Figure 5-7. Single-Phase Liquid Release (Low Volatility)
5-121
-------
Table 5-15 summarizes the model input for this example. The following
sections describe how the input values were developed.
5.8.1 Observable Data
The observable data were presented in the scenario description.
Table 5-16 summarizes these data.
5.8.2 Chemical Data Requirements
The chemical data for a 30% weight hydrochloric acid solution is given
in Appendix B.
5.8.3 Release Class
In this scenario, only one release occurs. The liquid pours out of the
container until the liquid level is at the hole level. The release class is
determined using the flow diagram in Figure 4-1.
The hydrochloric acid solution is stored as a liquid. The flash diagram
in Appendix B indicates that flashing will not occur since the storage
temperature is below the boiling point and the flash diagram curve begins its-
upward branch just at the boiling point.
Since no flash will occur, the next step is to decide if the boiling
point is above or below the ambient temperature. In this example, the boiling
point is 370.2 °K. The ambient temperature is 291.48 "K. Thus, the boiling
point is higher than the ambient temperature. According to the flow chart in
Figure 4-1, this scenario fits the definition of a low volatility liquid
release. Since the boiling point is above the ambient temperature, the
material will evaporate but will not boil. The evaporation rate will be
temperature- and wind-speed-controlled. Use the flow chart in Figure 4-5 to
determine the model input.
5-122
-------
TABLE 5-15. INPUT SUMMARY FOR THE LOW-VOLATILITY LIQUID RELEASE EXAMPLE
Chemical Data (5.8.2)
From Appendix B
Release Type (5.8.3)
low volatility liquid
Continuous or Instantaneous Release Categories (5.8.4)
Continuous
Emission Rate (5.8.7)
0.005365 kg/s (total), 0.004272 kg/s (hydrogen chloride)
Release Temperature (5.8.8)
291.48 °K
Vapor Fraction (5.8.9)
Frel - 0 (flash) All emission is in vapor state
Initial Concentration (5.8.10)
fa - 0.9689
fw - 0.0185
fi - 0.01258
Density (5.8.11)
prel - 1.204 kg/m3
Pm' - 1.203 kg/m3
Release Diameter or Area (5.8.12)
Diameter - 8.256 m
Release Buoyancy (5.8.13)
Cp — 136.7 > 6 (do not use dense gas model)
Release Height (5.8.14)
0.0 m
Ground Surface Temperature (5.8.15)
Same as ambient
Averaging Time (5.8.16)
One hour
5-123
-------
TABLE 5-15. (CONTINUED)
Meteorology (5.8.17)
Wind Speed and Direction
45° <§ 5 mph (2.24 m/s)
Measurement height - 10.7 m
Speed @ 10 m - 2.23 m/s
Stability Class
C
Surface Roughness Length
ZO - 0.01 m
Ambient Temperature, Relative Humidity, and Pressure
Temperature - 65 °F (291.48 °K)
Relative Humidity - 58%
Pressure - 1 atm (101325 Pa)
Output Definition (5.8.18)
ERPG concentrations
Minimum distance of concern - 450 ft (137.2 m)
Observable Data (5.8.1)
See Table 5-16
5-124
-------
TABLE 5-16. OBSERVABLE DATA SUMMARY FOR THE LOW-VOLATILITY LIQUID RELEASE
EXAMPLE
Release Description
Species: - Hydrogen chloride 30% wt solution
Container: Vertical cylinder
Diameter: 12 ft (3.66 m)
Height of liquid: 9 ft (2.74 m)
Total liquid: 63126 Ib (28634 kg)
Temperature: 65°F (291.48 °K)
Pressure:
Bounding dike:
1 atm (absolute) (1.01325 x 105 Pa)
24 x 24 ft, 576 ft2 (53.54 m2)
Hole site:
Inner diameter:
Area:
Height of hole:
0.75 inch (0.0191 m)
3.0680 x 10~3 ft2 (2.85 x 10'* m2)
1 ft (0.30 m)
Nearest boundary: 450 ft (137.2 m)
Meteorology:
Temperature:
Relative Humidity:
Pressure:
Wind Speed:
Wind Direction:
Measurement hgt:
Surface Roughness:
Cloud cover:
Time:
65°F (291.48 °K)
58%
1 atm (101325 Pa)
5 mph (2.24 m/s)
Northeast (45°)
35 ft (10.7 m)
0.01 m .
3/8
Late-afternoon
5-125
-------
5.8.4 Continuous or Instantaneous Release Categories
In this example, the emission rate into the atmosphere is 0.005365 kg/s
(from Section 5.8.8). The duration of the release is how long it takes for
the liquid to escape the container. There is 28634 kg liquid in the
containerat a depth of 2.74 meters. The hole is at a height of 0.3 meters
above the container bottom. This means that the fraction of liquid that
remains in the container after the liquid stops being released is 0.3/2.74.
The duration of the release (td) then is:
t = (1 ~ (0.3m/2.74m)))28634kg
0.005365^
s
= 4752822s
= 1320 hours
There is, however, another emission rate, that of the liquid coming from
the container. Its duration is calculated as:
t = (1 ~ (0.3nV2.74m)))28634kg
1.273^
s
= 20031 s
= 5.56 hours
For a non-source-term model, the value of td must be compared either to
the advection time to a receptor or to the time it takes a concentration of
interest to reach the maximum downwind distance. Using the formula in
Section 4.4 for ttrav, and assuming a wind speed of 5 mph (2.24 m/s) and a
distance of 450 ft (137.2 m), the advection time is calculated at 122 seconds.
Since both emission durations are longer than the time required to reach the
downwind point of interest, the release is considered continuous.
5-126
-------
For a source-term model, the two values of td also are sufficiently long
to treat this as a continuous release for most receptors.
5.8.5 Release-class-specific Calculations
The flow chart showing the calculations of input for a single-phase low
volatility liquid release is presented n Figures 4-5 and 4-6.
5.8.6 Determination of Choked Flow for Gas Releases
This determination is not required since it is a low volatility liquid
release.
5.8.7 Emission Rate
There are two emission rates of importance in a low volatility liquid
release. The first is the emission rate of the liquid as it releases from the
container. The second is the evaporation rate from the pool that forms.
These two emission rates must be compared to determine which one is the
limiting emission.
To calculate the emission rate from the container, the same technique is
followed as for a high volatility spill (Section 4.7.7). First, the storage
pressure (psv) must be calculated using:
= 101325Pa
5-127
-------
p_ = 101325Pa exp
' (2.3549xl06J/kg)(2124kg/kmol)
(8314J/kmol°K)
1 1 "1
37
-------
E = (0.65)(2.85xlO ^m2)
±
[2(993.30kg/m3)((1.2509xl05Pa) - (1.01325xl05Pa))]2
= 1.273 kg/s
Finally, the maximum pool evaporation rate must be calculated and com-
pared to the container emission rate. Whichever emission rate is the smaller
is the limiting emission rate. The pool evaporation rate (Ep00l in Section
4.7.8) is calculated from:
Epooi = 6.94x10 -7(1 + 0.0043(1^ - 273.15)*2)Ura75ApM-!-l
Before E^^ can be calculated, three parameters must be known. One of the
parameters, release temperature (Trel) , is determined in Section 5.8.8 as
291.48 °K. The other two, pv and p^,, need to be calculated here. As
discussed in Section 4.7.8, the emission Ep0ol should be calculated at both
Trel and Ta; the larger value of Ep00l should then be used. In this example,
however, Trel and Ta are the same, so only one calculation is required.
The parameter pv is the vapor pressure of the entire hydrochloric acid
solution. It is the sum of the partial pressures of the water vapor and the
anhydrous hydrogen chloride over the solution. The pool emission rate must be
compared to the liquid emission rate from the container. Since the liquid is
made up of both water and hydrogen chloride, the pool emission rate should
take into account both species. From the data base, at Trel:
pv(Hd) = 1274.2Pa
p^O) = 655.8Pa
pv = 1930.0Pa
If this were not a mixture, the parameter pv is calculated from the equation:
5-129
-------
pv = 101325Pa exp
R
r*
The parameter
calculated from:
is the vapor pressure of hydrazine at Trel. It can be
= exp 76.8580 -
0.00615571
= exp|76.8580 - 7245>2 - 8.221n(291.48°K)
*| 291.48°K
+ 0.0061557(291.48°K)]
= 1270.8Pa
This makes the value of Ep00l:
E, = 6.94xlQ-7(l + 0.0043(291.48°K - 273.15°K)*2)(2.24m/s)1
,0.75
"pool
• VM ^hrltannl^ 193°'OPa
1270.8Pa
(53.54m2)(21.24kg/kmol)
which gives a value of 0.005365 kg/s. Thus, E^,^ is less than E. The pool
size is at its maximum of 53.54 m2. An important point about the emission
rate is that it includes the emissions of both water vapor and hydrogen
chloride. The actual emission rate used in a model should be for the hydrogen
chloride alone. The emission rates of the water and hydrogen chloride vapors
are in the same ratio as their weight per cent ratio. This means that the
hydrogen chloride emission rate (EHCL) can be calculated using:
5-130
-------
= 0.004272kg/s
where: MgC1 - molecular weight of hydrogen chloride (36.46 kg/kmol); and
MT - average molecular weight of the water vapor and hydrogen
chloride mixture.
^HQ +
^^
.0PaJ
+ ^
*
930.0Pa
= 30.19kg/kmol
5.8.8 Release Temperature
The liquid is released at its storage temperature. For a conservative
approach, the release temperature should be assumed to be either the storage
or ambient temperature, whichever is higher. Both of these temperatures are
the same in this example, so the release temperature (Ta) is assumed to be
291.48 °K.
5.8.9 Vapor Fraction
The value of Frel was determined to be zero in the process of
determining the release class (Section 5.8.3). That is, no flashing occurred.
5 . 8 . 10 Initial Concentration
The method for calculating initial concentration is given in
Section 4.10.2. The release in the current example has all three components
5-131
-------
allowed: air, water vapor, and chemical of interest. The molar fractions of
each of these components must be determined. The ft parameter is the ratio of
the vapor pressure of the hydrogen chloride to the ambient pressure. The
partitioning of the emission is of particular importance because, in the air,
the chemical of interest is hydrogen chloride, not water vapor. Therefore,
only the hydrogen chloride vapor pressure is used in determining the initial
concentration of the chemical of interest. This gives:
f _ pv(Hd)
1 P.
1274.2Pa
101325Pa
= 0.01258
The remaining components are just the air and water vapor. Water vapor
comes from both the ambient atmosphere and the source. The easiest approach
is to solve for the molar fraction of air and then determine the water vapor
fraction by subtracting the sum of fi and f a from 1.
The molar fraction in the atmosphere that is not taken up from the
«
emission species is (1 - fj/) where f±' is the molar fraction due to the water
vapor and hydrogen chloride being emitted from the source. The value of ft'
is calculated just as f± was calculated, namely:
f,'-
pv(HCl) * p.CH.Q)
P.
1930.0Pa
101325Pa
= 0.01905
The molar fractions in the ambient air of air and water vapor also need to be
calculated. The water vapor fraction is a function of the temperature and
5-132
-------
relative humidity. The water vapor molar fraction in the ambient air (fw')
is:
'-ir^-
\ /
where: es is calculated from:
2353
Iog10e8 = 6.3994 -
= 6.3994 -
T
2353
291.48°K
which leads to a value of es of 0.0212. With a relative humidity of 58%, this
leads to a value of fw' of 0.0123. The value of molar air fraction in the
atmosphere (fa') is the complement of fw' , or 0.9877. The molar fraction of
air in the emission is:
f. = tfd - ^
= 0.9877(1 - 0.01905)
= 0.9689
and the molar fraction of water vapor is:
= 1 - (0.01258 + 0.9689)
= 0.0185
In summary:
fa - 0.9689
fw - 0.0185
fi - 0.01258
5-133
-------
5.8.11 Density
The density input parameters required are calculated using the formulas
given in Section 4.11. Since the emission has the three components (air,
water vapor, and hydrogen chloride), each component contributes to the density
of the emission. The first step in calculating the overall density is to
calculate the individual densities of the three components. Since only the
gas phase of hydrogen chloride is being emitted, p^ is equal to p&. The
densities of the three components, if they were the only components in the
release, would be given by:
Pi = P
= (101325Pa)(36.47kg/kmol)
(8314J/kmol°K)(291.48°K)
= 1.525kg/m3
P ,
a T?T
K1id
= (10132SPa)(28.9kg/kmol)
(8314J/kmorK)(291.48°K)
= 1.208kg/m3
5-134
-------
and finally,
« P^
Pw =
= (101325Pa)(18.02kg/kmol)
(8314J/kmorK)(291.48°K)
= 0.7534kg/m3
Note that the molecular weight for pure hydrogen chloride gas was used rather
than the mixture molecular weight. The mixture molecular weight is for the
liquid phase. Only the gas phase emissions into the atmosphere are being
considered here. The release density (prel) is then given by as described in
Section 4.11):
„ -
where :
In this case:
MT = (0.9689)(28.9kg/m3) + (0.0185)(18.02kg/m3)
+ (0.01258)(36.47kg/m3)
= 28.79kg/m3
and so ,
5-135
-------
- ( (0.9689)(28.9kg/kmol)
p — I ___^ — -
U28J9kg/kmol)(1.208kg/m3)
+ (0.0185)(18.02kg/kmol)
(28.79kg/kmol)(0.7534kg/m3)
+ (0.01258)(36.47kg/kinol V1
(28 J9kg/kmol)(1.525kg/m3) J
= 1.204kg/m3
The fact that the release comprises three components also complicates
the calculation of the actual ambient air density. In the above calculation,
a value was calculated for pa, but that value was for dry air only. The
ambient air density needs to be calculated in the same manner as the release
density, except that the fa' and fw' are used instead of fa and fw. First, the
apparent molecular weight of the air is given by:
M' =
= (0.9877)(28.9kg/kmol)
= 28.77kg/kmol
(0.0123)(18.Q2kg/kmol)
Then the ambient air density is given by the equation:
p. =
( (0.9877)(28.9kg/kmol)
{ (28.77kg/kmol)(1.208kg/m3)
+ (0.0123)(18.02kg/kmol) ^
+ (28.77kg/kmol)(0.7534kg/m3)>
U03kg/m3
5-136
-------
5.8.12 Release Diameter or Area
Part of this release is from an evaporating pool. The exit velocity of
evaporation is negligible. The release diameter is simply the diameter of the
pool. Assuming that the pool is circular, this leads to:
5.8.13 Release Buoyancy
7C
8.256m
•(53.54m2)
The release buoyancy is used to determine if a dense gas model should be
used, rather than a neutral or positive buoyancy model. The first step is to
compare prel to p&' . If prel is less than or equal to pa, the release is not to
be treated as a dense gas release. In this example, prel is 1.204 kg/m3 and pa
is 1.203 kg/m3. The value for pa is actually the value of pa' , which was
calculated in Section 5.8.11 and represents the moist ambient density. Since
Prei is greater than pa, the next step is to apply the denser-than-air
criterion.
The calculation of the denser-than-air criterion is based on the formula
in Section 4.13.1, since this release is classified as a continuous release
(Section 5.8.4). The criterion (Cp) is given by:
c =
5-137
-------
C = (2.24m/s)
p
8.256m
1.204kg/m3 - U03kg/m3
1.203kg/m3
= 136.7
Since Cp > 6, a dense gas model should not be used. In most low volatility
spills, this will be the case. In the process of arriving at this point,
however, most of the input for a non- dense gas model has also been developed.
5 . 8 . 14 Release Height
The hole in the tank is 0.3 meters above the ground. The liquid spills
out of the hole onto the ground and evaporates. The gas phase of the release
enters the atmosphere at the pool level, which is at ground level. Therefore,
the release height is zero.
5.8.15 Ground Surface Temperature
No information is given on the ground surface temperature . Therefore ,
the ground surface temperature is assumed to be equal to the ambient
temperature, or 291.48 °K.
5.8.16 Averaging Time
The averaging time is specified as 1 hour to make the model -predicted
concentrations comparable to the ERPG (Emergency Response Planning Guideline)
concentrations .
5-138
-------
5.8.17 Meteorology
Wind Speed and Direction
The wind speed and direction are assumed either to be available from
on-site meteorological equipment or are representative of average conditions
at the site. The wind speed in this example is given as 5 mph (2.24 m/s).
The wind direction is northeast (45°). The measurement height is given as
10.7 meters. To determine the greatest specific impact, multiple wind speeds
may be required in conjunction with multiple stability classes.
Stability Class
The stability class is not given explicitly. However, a class can be
estimated from the information provided using the method described in the
Workbook, Section 3.1.2. In this example, there is 3/8 cloud coverage, and
the release is in the late-afternoon, which indicates that the insolation is
slight. As shown in Table 3-2 of the Workbook, the stability class estimated
for a wind speed of 2.24 m/s at 10 meters during a day with slight insolation
is "C."
Surface Rouehness Length
In keeping with the value suggested for modeling done for planning
purposes, the surface roughness is assumed to be 0.01 meters.
Wind Speed at 10 m Altitude
With the surface roughness, stability class, and wind speed at 10.7
meters, the wind speed at 10 meters can be estimated. Using the equation in
4.17.3:
5-139
-------
Zl
= 2.23m/s
This wind speed at 10 meters altitude still indicates a "C" stability class.
Ambient Temperature. Relative Humidity, and Pressure
The ambient temperature, relative humidity, and pressure are all assumed
to be available from on-site equipment. See the observable data, Table 5-16.
5.8.18 Output Definition
The concentration levels to specify when generating output should be the
three ERPG (Emergency Response Planning Guideline) concentrations given in the
hydrogen chloride data table in Appendix B. The actual impact values to be
predicted are listed in Section 7.0.
5-140
-------
SECTION 6
MODEL INPUT USED
This section describes the input used by each model for each release
class. Any differences between input derived in Section 5 for a given release
class and that required by a specific model are noted. Example output from
each of the models is provided in Section 7, as well as how to find impact
estimates.
The computer models discussed in this document operate in one of two
ways. "Batch type" models accept input by reading a file containing the
numerical values of required parameters. "Interactive type" models, accept
input as the user enters parameter values through input screens presented by
the program.
In a batch model, the input must be created before the model is run.
The input is simply a list of numbers arranged in a set order. In an
interactive model, the input is entered while the program is running by
"filling in the blanks" on input screens. Interactive models often allow the
user to enter the input in any order. In this section, the input used by the
batch models is given as lists to resemble what the input would need to look
like for the model to run. The input for the interactive models is listed in
a more descriptive table.
Of the models considered in this document, ADAM and HGSYSTEM require the
most discussion. In the case of ADAM, there are some outside calculations
that may be required for a simulation. The outside calculations are unique to
ADAM and not provided in Sections 4 or 5. The outside calculations are
described in this section. In the case of HGSYSTEM, since it is made up of a
number of models run in series, there can be more than one input stream per
6-1
-------
simulation. Normally, the output from one HGSYSTEM model, which becomes the
input for another HGSYSTEM model, does not require user interaction. However,
since the input may be varied by the user during a specific simulation, all
intermediate input files (called "link" files) are shown.
6.1 ADAM
The ADAM model is an interactive model. The input is entered in a
screen-by-screen method. The input for each scenario modeled by ADAM is given
in Tables 6-1 through 6-6. Only six of the scenarios were simulated. The
Single-Phase Gas Release (Unchoked) was not modeled because ADAM cannot model
stack releases and does not account for non-anhydrous chemical releases. The
Single-Phase Liquid Release (Low Volatility) was not modeled because ADAM only
simulates heavier-than-air releases. Even if the Single-Phase Liquid Release
(Low Volatility) scenario resulted in a heavier-than-air release it still
would not have been modeled because ADAM does not account for non-anhydrous
chemical releases.
Since ADAM was designed and developed specifically to satisfy the needs
of the Air Force, the capabilities of the model are very specific regarding
the scenarios that can be modeled. Certain technical details about the ADAM
model that need to be taken into account when preparing input are:
• All releases are assumed to occur at ground level. Elevated
releases are not accounted for.
• In the case of liquids and liquified gases, storage tanks are
assumed to be cylinders full of liquid with axes normal to the
ground. The bottom of the tank is assumed to be at ground level.
• Tank storage pressure and temperature are assumed to remain
constant throughout a release. Tank blow-down is not accounted
for.
• The physical and thermodynamic state of a chemical is calculated
by ADAM and fixed by specifying the storage temperature and
pressure.
6-2
-------
TABLE 6-1. INPUT PARAMETERS FOR THE TWO-PHASE ETHYLENE OXIDE GAS RELEASE
(CHOKED) (Reference Table 5-2)
Chemical Name:
Run Title:
Data File:
Plot File:
Release:
Release Type:
Calculate Source Terms?
Pipe Diameter:
Fluid Velocity in Pipe:
Hole diameter:
Storage Temperature:
Storage Pressure:
Meteorology:
Temperature:
Relative Humidity:
Pressure:
Wind Speed:
Measurement Height:
Wind Direction:
Surface Roughness:
By-pass Atmospheric calculation?
Wind Speed @ 10 meters:
Stability Class:
Concentration Contour:
Averaging Time:
Include Variable Wind Effects:
Minimum Distance of Interest:
Time & Location Parameters:
Scaling:
Ethylene Oxide (C2H40)
EOX1
EOX1.DAT
EOX1.PLT
Hole in Pipe
Continuous
Yes
0.0762 m (3 in.)
Arbitrary for gas flow
0.01347 m .(-53 in.)
306.67 °K (92.2 °F)
2.02xl05 Pa (2.0 atm abs)
287.52 °K (57.2 °F)
62 %
L01325 Pa (1 atm abs)
5.37 m/s (12.0 mph)
4.57 m (15.0 ft)
Northeast (45°)
0.01 m
Yes
6.04 m/s (13.5 mph)
C (2.5 index)
800 ppm IDLH1
30 minutes
No
200 m (656.2 ft)
N/A
Automatic
Because of the currently set up distance step sizes in ADAM distances
calculated to LEL & UEL tend to be inaccurate
6-3
-------
TABLE 6-2. INPUT PARAMETERS FOR THE TWO-PHASE ETHYLENE OXIDE GAS RELEASE
CHOKED) (Reference Table 5-4)
(UN-
Chemical Name:
Run Title:
Data File:
Plot File:
Release:
Release Type:
Calculate Source Terms?
Pipe Diameter:
Fluid Velocity in Pipe:
Hole diameter:
Storage Temperature:
Storage Pressure:
Meteorology:
Temperature:
Relative Humidity:
Pressure:
Wind Speed:
Measurement Height:
Wind Direction:
Surface Roughness:
By-pass Atmospheric calculation?
Wind Speed @ 10 meters:
Stability Class:
Concentration Contour:
Averaging Time:
Include Variable Wind Effects:
Minimum Distance of Interest:
Time & Location Parameters:
Scaling:
Ethylene Oxide (C2H<,0)
EOX2
EOX2.DAT
EOX2.PLT
Hole in Pipe
Continuous
Yes
0.0762 m (3 in.)
Arbitrary for gas flow
0.0127 m (.5 in.)
298.16 °K (76.9 °F)
151990. Pa (1.5 atm abs)
296.5 "K (74.0 °F)
37 %
101325 Pa (1 atm abs)
2.68 m/s (6.0 mph)
6.09 m (20.0 ft)
Southwest (225°)
0.01 m
Yes
2.89 m/s (6.47 mph)
C (2.5 index)
800 ppm IDLH2
30 minutes
No
300 m (984.3 ft)
N/A
Automatic
2Because of the currently.set up distance step sizes in ADAM distances
calculated to LEL & UEL tend to be inaccurate
6-4
-------
TABLE 6-3. INPUT PARAMETERS FOR THE TWO-PHASE PRESSURIZED LIQUID CHLORINE
RELEASE (Reference Table 5-6)
Chemical Name:
Run Title:
Data File:
Plot File:
Release:
Release Type:
Calculate Source Terms?
Tank Diameter:
Volume of Chemical:
Storage Temperature:
Storage Pressure:
Meteorology:
Temperature:
Relative Humidity:
Pressure:
Wind Speed:
Measurement Height:
Wind Direction:
Surface Roughness:
By-pass Atmospheric calculation?
Wind Speed @ 10 meters:
Stability Class:
Concentration Contour:
Averaging Time:
Include Variable Wind Effects:
Minimum Distance of Interest:
Time & Location Parameters:
Scaling:
Chlorine (C12)
CLX1
CLX1.DAT
CLX1.PLT
Tank
Ins tantane ous
Yes
0.762 m (30.0 in.)
0.152 m3
294.3 °K (70.0 °F)
6.95xl05 Pa (6.86 atm abs)
294.3 °K (70.0 °F)
50 %
101325 Pa (1 atm abs)
4.47 m/s (10.0 mph)
10 m (32.81 ft)
South (180°)
0.01 m
Yes
4.47 m/s (10.0 mph)
C (2.5 index)
1 ppm (STEL)
15 minutes
No
100 m (328.1 ft)
N/A
Automatic
6-5
-------
TABLE 6-4. INPUT PARAMETERS FOR THE TWO-PHASE PRESSURIZED LIQUID SULFUR
DIOXIDE RELEASE (Reference Table 5-8)
Chemical Name:
Run Title:
Data File:
Plot File:
Release:
Release Type:
Calculate Source Terms?
Pipe Diameter:
Hole diameter:
Fluid Velocity:
Storage Temperature:
Storage Pressure:
Meteorology:
Temperature:
Relative Humidity:
Pressure:
Wind Speed:
Measurement Height:
Wind Direction:
Surface Roughness:
By-pass Atmospheric calculation?
Wind Speed @ 10 meters:
Stability Class:
Concentration Contour:
Averaging Time:
Include Variable Wind Effects:
Minimum Distance of Interest:
Time & Location Parameters:
Scaling:
Sulfur Dioxide (S02)
SFD1
SFD1.DAT
SFD1.PLT
Hole in Pipe
Continuous
Yes
0.05 m (2 in.)
0.0127 m (.5 in.)
28.0 m/s (91.9 ft/s)
323.15 °K (122 CF)
1.52xl06 Pa (15 atm abs)
291.15 °K (64.4 °F)
42 %
101325 Pa (1 atm abs)
3.13 m/s (7.0 mph)
6 m (19.7 ft)
Northeast (45°)
0.01 m
Yes
3.38 m/s (7.56 mph)
E (4.5 index)
5 ppm (STEL)
15 minutes
No
80 m (262.5 ft)
N/A
Automatic
6-6
-------
TABLE 6-5. INPUT PARAMETERS FOR THE SINGLE-PHASE ANHYDROUS HYDROGEN FLUORIDE
GAS RELEASE (CHOKED) (Reference Table 5-10)
Chemical Name:
Run Title:
Data File:
Plot File:
Release:
Release Type:
Calculate Source Terms?
Pipe Inner Diameter:
Fluid Velocity in Pipe:
Storage Temperature:
Storage Pressure:
Meteorology:
Temperature:
Relative Humidity:
Pressure:
Wind Speed:
Measurement Height:
Wind Direction:
Surface Roughness:
By-pass Atmospheric calculation?
Wind Speed @ 10 meters:
Stability Class:
Concentration Contour:
Averaging Time:
Include Variable Wind Effects:
Minimum Distance of Interest:
Time & Location Parameters:
Scaling:
Hydrogen Fluoride (HF)
HFX1
HFX1.DAT
HFX1.PLT
Severed Pipe
Continuous
Yes
0.0635 m
Arbitrary for gas flow
365.0 °K (197.0 °F)
2.0397xl05 Pa (2.01 atm abs)
295.15 °K (71.6 °F)
45 %
101325 Pa (1 atm abs)
6.26 m/s (14.0 mph)
6.096 m (20.0 ft)
Northwest (315°)
0.01 m
Yes
6.74 m/s (15.1 mph)
D (3.5 index)
6 ppm (STEL)
15 minutes
No
76.2 m (250 ft)
N/A
Automatic
6-7
-------
TABLE 6-6. INPUT PARAMETERS FOR THE SINGLE-PHASE LIQUIFIED ETHYLENE OXIDE
RELEASE (Reference Table 5-14)
Chemical Name:
Run Title:
Data File:
Plot File:
Release:
Release Type:
Calculate Source Terms?
Container:
Diameter:
Volume of Liquid:
Height of Hole:
Hole diameter:
Storage Temperature:
Storage Pressure:
Meteorology:
Temperature:
Relative Humidity:
Pressure:
Wind Speed:
Measurement Height:
Wind Direction:
Surface Roughness:
By-pass Atmospheric calculation?
Wind Speed @ 10 meters:
Stability Class:
Concentration Contour:
Averaging Time:
Include Variable Wind Effects:
Minimum Distance of Interest:
Time & Location Parameters:
Scaling:
Ethylene Oxide (C2H^O)
EOX3
EOX3.DAT
EOX3.PLT
Tank Release
Continuous
Yes
Vertical Axis Cylinder
3.5 m (11.4 ft)
26.9 m3 (950.0 ft3)
0.5 m (1.64 ft)
0.00635 m (0.25 in )
283.85 °K (51.3 °F)
l.OlxlO5 Pa (1.0 atm abs)
301.15 °K (82.4 °F)
50 %
101325 Pa (1 atm abs)
2.0 m/s (4.47 mph)
10 m (32.81 ft)
North (0°)
0.01 m
Yes
2.0 m/s (4.47 mph)
E (4.5 index)
800 ppm IDLH3
30 minutes
No
100 m (328.1 ft)
N/A
Automatic
3ERPG limits are not available
6-8
-------
Mechanical losses and two-phase flows in pipes are not accounted
for.
The mass flow rate throughout a release is assumed to be constant
and equal to the initial flow rate. All releases occur in the
direction of the wind and chemical dispersion continues until the
concentration of interest is reached. Limited duration releases
are not accounted for, i.e., continuous releases of finite
quantities (in the tank) are not modeled exactly.
• Dispersion of chemicals not in the ADAM list of chemicals is
modeled as a non-reactive chemical release, provided the
thermodynamic property data are included in standard database
format.
Presently, there are eight anhydrous chemicals in the model. Materials
not appearing in the standard ADAM chemical selection may be modeled if:
1) the material is .anhydrous (pure, non-aqueous);
2) the vapors produced after the release of the chemical into the
atmosphere are heavier than air;
3) the material is non-reactive upon release, i.e., it does not
polymerize, undergo spontaneous dissociation, or combine with
ambient moisture to form other species, and;
4) physical and thermodynamic properties of the material are
available and are similar to those of chlorine.
If the above conditions are met, a database of physical properties of
the chemical to be modeled must be compiled, the name of which is assigned the
file name for chlorine (CLXPORP.DAT). The actual chlorine properties file
must then be moved from the ADAM directory and saved by the user, and the new
chemical file substituted for it. It is important that the original
CLXPROP.DAT file be saved so that it can replace the new chemical file in case
chlorine is to be modeled in the future. Modeling the new material simply
requires choosing the chemical "Chlorine" from the ADAM chemicals list. This
method was implemented for the ethylene oxide releases.
Although the method of modeling alternate materials is rather
simplistic, the format of the properties data base is extremely structured and
6-9
-------
specific. Deviation from the file format will cause unpredictable model
operation and results. If the modeling of alternate materials is desired, the
providers of ADAM should be consulted for assistance.
ADAM has the capability to calculate dispersion model input parameters
(source-term mode) based on observable data (storage and atmospheric
conditions) or to provide for the input of these values calculated by other
means (non-source-term mode). To run ADAM in the non-source-term mode
requires that the following parameters be provided:
• Width and depth of the released plume;
• Heat rate to source plume; and
• Mass rate of air entrained initially.
The methods for calculating these parameters have not been provided in
this document. Since these parameters are internally calculated by ADAM from
obsevable data, all the release classes were simulated using the ADAM source-
term mode of operation.
In the title of each of the tables listing the ADAM input is a reference
table. This reference table refers to the table in Section 5 from which the
input was derived. Since all the simulations were done with the source-term
mode of operation, all the tables refered to are the observable data tables
rather than the calculated input tables.
Extra Calculations
Some calculations are needed to create input outside the ADAM model
either because the input is unique to ADAM or because the ADAM model developer
suggests a method of calculation different from that presented in Section 4.
Since these calculations or methods affect more than one release class, they
are presented here as general calculations for ADAM. The input given in
Tables 6-1 through 6-6 is only that which is used after a decision has been
made as to whether the release should be treated as instantaneous or
6-10
-------
continuous. As discussed below, it is suggested that ADAM be run assuming
both conditions and then selecting the appropriate simulation. In the
discussion of each release class, all input requiring calculation for both the
instantaneous and continuous cases is presented.
Stability Class
In most cases it is desirable to over-ride atmospheric calculations and
simply supply to the model the atmospheric stability category from which
dispersion parameters are determined. The stability class can be determined
from cloud cover and insolation information by following the guidance provided
in the EPA Workbook, Section 3.1.2. Instead of the letter classification (A
through F) , ADAM uses a numerical index for stability. Stability class "A" is
represented by an index between 0 and 1. Stability class "B" is represented
by an index between 1 and 2, etc.
When over-riding atmospheric calculations, it is also necessary to input
the wind speed at the standard meteorological reference height of 10 meters.
In some cases, the wind speed may be listed in the scenario description at a
height other than 10 meters. The method described here is what is used
internally in ADAM, which is different from that described in Section 4.17.3.
Assuming a logarithmic velocity distribution, the wind velocity at 10 meters
(U10) may be Calculated by:
where: Uz - wind speed at height z (m/s) ;
Z0 — surface roughness (m) ; and
Z = height at which velocity U2 is measurement (m) .
6-11
-------
The Effect of the Vertical Axis Tank Assumption on Liquid and Liquified
Gases
In ADAM, it is assumed that all tank releases are from cylindrical
containers with their bases on the ground. That is, the cylinder's axis is
vertical. If the tank has liquid in it, it is assumed to be full. Tanks that
have horizontal axes must be simulated as if they were vertical. For a
continuous liquid release from a tank, the rate at which the liquid escapes is
determined by the pressure inside the tank and the column pressure due to the
depth of liquid. To simulate in ADAM the rate at which a liquid releases from
«
a horizontal cylinder, the depth of liquid in the horizontal cylinder must be
equal to the depth of liquid in a vertical cylinder. This is done by
specifying a vertical tank height equal to the depth of liquid (h,) in the
cylinder.
Calculating the depth of liquid in a partially full horizontal cylinder
is not trivial. If the volume of liquid in the horizontal cylinder is known,
hj can be calculated iteratively from:
i . £ Cos-'ci^Mf-
L 4 d 2
where: V, - volume of liquid in tank (m3);
L - tank length (m); and
d - tank diameter (m).
The only unknown value in the equation is h,. Therefore, the equation can be
solved iteratively by solving for one of the hfs values in the equation. To
do this, assume a value of h, (for example, one half of d) and solve for a new
value of h,. Use the new value to calculate another h,. Continue solving for
new values of h, until two consecutive values of h, do not differ
6-12
-------
significantly (perhaps one percent of dr) . If the mass of liquid m, (kg) is
known, the volume of liquid, V, (m3) , may be calculated by:
V = ^
1 p.'
where the liquid density, pt (kg/m3) , is obtained from the chemical data in
•
Appendix B.
The discussion above describes the calculation required for a
continuous liquid release. For a continuous gas release, to simulate a
horizontal cylinder under ADAM, it is only necessary that the vertical
cylinder assumed in the model be of the same volume as the horizontal
cylinder. When simulating an instantaneous release, the volume of liquid or
gas released is required. For instantaneous releases of liquid or liquified
gas from a tank, only the volume of material in the tank is significant. If
an instantaneous release of compressed gas is being modeled, the volume of
material specified should be the volume of gas at atmospheric pressure and the
temperature of the gas released. Because only the volume is required for an
instantaneous release, the shape of the container is not assumed in ADAM.
Pipe Flow Velocity
When using the pipe release option for liquids or liquified gases, the
velocity of fluid in the pipe (which is assumed to be unaffected by friction)
is a necessary input parameter. This variable is unique to ADAM and is not
required by any of the other models in this document. If the mass flow rate
is available for a given release, the liquid velocity, U,, may be determined
by:
u, = Mfl~
where: Mfj^, - mass flow rate in pipe (kg/s);
PI ~ liquid density at chemical temperature (kg/m3); and
6-13
-------
- inner"cross sectional area of pipe (m2).
In the case of a storage tank supplying a liquid or liquified gas pipe
flow, the velocity in the pipe can be set equal to the fluid exit velocity at
the tank/pipe connection. In this case, the tank liquid exit velocity, can be
determined by:
where: Cc - contraction loss coefficient (.61);
Cy - velocity loss coefficient (.98);
**stor ~ storage pressure (Pa) ;
Patm — atmospheric pressure (Pa) ;
pk - liquid density at the storage temperature (kg/m3) ;
h| — liquid height in tank (m) ; and
hb - height of hole above ground level (bottom of tank)
For a partially- full horizontal cylinder tank, the height of liquid, h, ,
should be set equal to a tank height that is calculated based on the tank
diameter and volume of material modeled. This calculation is described in the
preceding section on the "Vertical Axis Tank Assumption."
For gas pipe flows , the velocity of the gas at any point in the flowing
stream is calculated based on the pressure, density, and ratio of specific
heats at constant pressure and constant volume. Although the velocity
parameter is available as input for gas flows in pipes , the value entered can
be anything. ADAM ignores whatever is entered and internally calculates the
velocity.
6-14
-------
Instantaneous vs. Continuous Release
As suggested in Section 4.4, to determine whether a given release should
be modeled as instantaneous or continuous, the release duration needs to be
compared to the dispersion duration based on a continuous release. If the
resulting dispersion duration is less than the release duration, the release
should be treated as continuous. Conversely, if the dispersion duration is
greater than the release duration, the release should be treated as
instantaneous.
An alternative method of determining the release type is to compare the
dispersion distance for both instantaneous and continuous cases. If the
dispersion distance to the concentration of interest is greater for a
continuous release than that for an instantaneous release, the release should
be modeled as instantaneous; otherwise the release is continuous. This method
requires running ADAM twice for each release. The volume of material released
is required for each simulation.
In summary, to determine whether a release is continuous or
instantaneous for a given scenario:
1) Model the scenario as a continuous release, recording the
dispersion distance;
2) Model the scenario as an instantaneous release, recording
the dispersion distance;
3) If the dispersion distance for the continuous case is less
than that for the instantaneous case, the release can be
treated as continuous; otherwise, it is instantaneous.
Two-phase Gas Release (Choked) Example
This scenario may be modeled utilizing the continuous release hole-in-a-
pipe option. Since the scenario description assumes a pipe length of zero, the
assumption of frictionless pipe flow in the ADAM source model is valid. In
6-15
-------
using the pipe option, it is necessary to provide as input the pipe diameter
and the fluid velocity. However, for gas flows, the velocity entered is
arbitrary because it is internally calculated by the model.
ADAM assumes, by default, that at the chemical temperature of 303.67 °K
(86.9 °F) and pressure of 202650 Pa (2.0 atm abs), the ethylene oxide exists
in the pipe as a saturated liquid (ADAM assumes that all materials are in the
liquid phase at the saturation temperature and pressure). To model the
material as a saturated vapor, it is necessary to specify chemical temperature
approximately 3 degrees higher than the saturation temperature. Therefo're, in
modeling the scenario as a continuous release, the temperature of the material
must be specified as 306.67 °K. .
For an instantaneous release, ADAM assumes that the temperature of the
ethylene oxide gas in the atmosphere (right after release) is equal to the
ambient temperature of 287.2 °K. The volume of ethylene oxide at atmospheric
conditions is determined and modeled as equivalent to the instantaneous
volume.
Since ADAM models only ground level releases, it is assumed that the
pipe height relative to the ground is zero (i.e., the height of 12 feet cannot
be accounted for in ADAM). Also, since fixed duration continuous releases are
not accounted for, the plume is assumed to exist for the entire duration over
which the contour of the specified concentration is achieved.
For the given release conditions and a specified concentration of
interest of 30,000 ppm LEL, the mode of output from ADAM suggests that the
source concentration upon mixing with entrained air is at least lower than the
30,000 ppm limit of interest. In this case, dispersion calculations were
aborted and the predicted dispersion distance was output as zero meters. These
results indicate that for a 30,000 ppm concentration (and any concentration
higher), a toxic vapor cloud hazard does not exist. To illustrate the method
by which this scenario is modeled, the IDLH limit of 800 ppm is used.
6-16
-------
Wind speed at 10 meters. Based on a wind speed of 5.37 m/s (12.0
mph) at 4.57 m (15.0 ft), a surface roughness of .01 meters, and a
logarithmic wind velocity profile, the wind speed at 10 meters is:
to-10'
U10m = 5.37 m/s x -^2L». = 6.06 m/s.
to———
Instantaneously released volume. Assuming ideal gas behavior and
that the temperature of the gas is equal to that ambient
temperature, the in air volume of material that would be
instantaneously released, based on 8 minutes and the flow rate
calculated in Section 5.1.7 is:
_ (0.06341^X4808) (-z-jfc) <287.7°K) _ 3
(101325 p,)
This value would be used in modeling the release as instantaneous.
Instantaneous or continuous release. The results of the
simulation for this release indicate that the maximum downwind
distance for a continuous and instantaneous release is 9 m and 248
meters, respectively. Based on the methodology described above,
the release should be modeled as continuous.
Two-phase Gas Release (Unchoked) Example
With the exception of meteorological conditions, chemical temperature
and pressure, and chocked flow, this scenario is essentially identical to the
pipe release in the first release class example. In light of this, the
discussion presented for the two-phase gas release (choked) example is also
applicable here.
As in the two-phase gas release (choked) example, the temperature of the
material must be specified at least 3 degrees higher than the saturation
temperature (298.16 °K instead of 295.16 °K) in order for it to be modeled as
a release of saturated vapor. Also, as in the previous scenario, the UEL and
LEL limits were higher than the concentration at the source because of the
chemical release and air entrainment. Likewise, for this scenario, dispersion
calculations were aborted and the predicted dispersion distance was set to
<*
6-17
-------
zero. To illustrate the method of modeling this release, the IDLH limit of 800
ppm will be used.
• Wind speed at 10 meters. Based on a wind speed of 2.68 m/s (6.0
mph) at 6.09 m (20.0 ft), a surface roughness of .01 meters, and a
logarithmic wind velocity profile, the wind speed at 10 meters is:
fa 10 a
U1Qm = 2.68 m/s x -£-. = 2.89 m/s.
fn P™* •*
.mm
Instantaneous or continuous release. Assuming that the flow rate
through the pipe hole would not be substantially less than that of
the choked flow in the first release class example, and the fact
that the continuous dispersion distance for this release is
minimal (9 meters) , the release in this scenario will be modeled
as continuous.
Two-Phase Pressurized Liquid Example
This release scenario (as described in Section 5)involves mechanical
losses and two-phase flow through a tube, which is not a release class that is
within the capabilities of ADAM. However, if these phenomena are neglected
(equivalent to assuming the pipe length is zero) the scenario can be modeled
as a continuous release based on the severed pipe. In order to use this
option correctly, the pipe diameter specified in ADAM must be set equal to the
severed tube diameter described in the scenario. Since the severed pipe
option requires the input of the liquid velocity in the pipe, the flow rate
from the one-ton container must be calculated and the velocity determined from
the resulting value. ADAM assumes the height of the pipe relative to ground
is zero.
To model the release as instantaneous, it is necessary to input the tank
diameter and-volume of material released. The calculations are:
6-18
-------
Chemical density at 293.4 "K. Using the correlation* A+T*(B+T*C)
where A-2170.0, B--2.6, and C-0.0, the storage density is:
p, (293.4°K) = 2170.0 + 294.3 x (-2.6) = 1405.0 X
m3
Initial volume of chemical. At the storage temperature of 293.4
°K, the total volume of chlorine is:
VA = mgk8 = 0.161 m3.
1405 kg/in3
Similarly, the volume of liquid remaining after the liquid level
falls below the eductor is:
ch = "* *i = 0.009 m3
1405 kg/to3
Therefore, the volume of liquid released is 0.152 m3.
Height of liquid in one ton container. For the one -ton container
described in the scenario description, the liquid height based on
500 Ibs of liquid ethylene oxide (0.161 m3) is 0.145 meters.
Similarly, the height at which 30 Ibs (13.6 m3) of liquid remain
in the container is 0.0234 meters.
Liquid velocity in tube. Disregarding mechanical losses and
transition to two-phase flow and assuming that the velocity
through the tube is equal to the velocity at the tank/eductor
connection, the velocity is:
U, = (.6l)(.99\2° * - 101325 * + 9.81 ^(.145 m - .0234 m))
V 140S kg/in3 s2
= 17.43 m/s
Instantaneous or continuous release. The results of the
simulation indicate that the maximum downwind distance for a
continuous and an instantaneous release is, respectively, 4159 *m
and 2984 m. Based on the methodology described above, the release
should be modeled as instantaneous (the instantaneous release is
based on 470 Ibs of liquid chlorine).
*Cor.relation and coefficients obtained from the United States Coast Guard
Hazard Assessment Computer Systems (HAGS). This correlation is used in ADAM.
6-19
-------
Two -phase Refrigerated Liquid Example
To model this scenario as a continuous release, it must be assumed that
there are negligible losses along the length of pipe through which the liquid
sulfur dioxide flows. If this assumption is made, then the "hole-in-a-pipe"
option may be used to model the scenario. To use this option, it is necessary
to determine the liquid velocity in the pipe and specify it as an input to the
model. To calculate the liquid velocity, it is assumed that the pipe
connection occurs at the bottom of the tank and that the tank is liquid full.
The velocity at the pipe/tank location is then calculated and the fluid
velocity in the pipe set equal to the resulting value. ADAM, by default,
assumes that the height of the pipe relative to the ground is zero.
For an instantaneous release, the volume of material is determined based
on the mass of sulfur dioxide released (6586 kg). The calculations are:
Chemical liquid density at 323.15 °K. Using the ADAM liquid sulfur
dioxide density correlation5 A+T*(B+T*C) , where A-2085.6, B — 2.4,
and C— 0.0, the liquid storage density is:
p, (323.15°K) = 2085.6 + 323.15 x <-2.40) = 1310 X
M
• Initial volume of chemical. At the storage temperature of 332.15
°K, the volume of chemical is:
V = ^^ = 5.03 m3.
1310 kg/m3
Note: although the scenario description specifies the tank volume
as 4.277 m3 (calculated from the specified diameter and length),
the difference between 4.277 m3 and the calculated chemical volume
of 5.03 m3 is probably made up in the semi- spherical ends of the
tank.
Correlation and coefficients obtained from the United States Coast Guard
Hazard Assessment Computer Systems (HAGS). This correlation is used in ADAM.
6-20
-------
Fluid velocity in the pipe. Assuming that the tank is liquid full
(depth of liquid - tank diameter) and the pipe is connected to the
bottom of the tank, the fluid velocity in the pipe is:
U. = C61)(.98)./2(1519875P'-10132SP' + 9.81 ^(1.22 m - 0.0 m))
f V 1310 kgfa3 s2
= 28.0 m/s
Wind speed at 10 meters. The wind speed at 10 meters must be
obtained from the given meteorological conditions. Based on a
wind speed of 3.13 m/s (7.0 mph) at 6 m (19.7 ft), a surface
roughness of .01 meters, and a logarithmic wind velocity profile,
the wind speed at 10 meters is:
to-!°
uiom = 3-13 m/s x -^ = 3-38
• Instantaneous or continuous release. For the scenario being
modeled, the maximum downwind distance for a continuous and
instantaneous release is 16972 meters and 24032 meters,
respectively. Therefore, the release should be modeled as
continuous .
Single-phase Gas Release (Choked) Example
If the effects of friction are neglected, the release may be modeled
using the severed pipe, continuous release option. This option requires the
user to enter the pipe's inner diameter and the fluid velocity. However, for
gas flows in pipe, the fluid velocity is arbitrary. Since ADAM has a specific
algorithm for handling severed pipes the release diameter at choked conditions
at the hole (0.06962 m) will not be used. Only the pipe's inner diameter will
be used.
•
In the case of an instantaneous release, the volume of gaseous hydrogen
fluoride released into the atmosphere must be determined at the release
temperature and ambient pressure. To be consistent with ADAM calculations, the
6-21
-------
release temperature is assumed to be equal to the ambient temperature of
295.2 °K. The calculations are:
Wind speed at 10 meters. For a wind speed of 6.26 m/s at 6 . 10
meters, a surface roughness of .01 meters, and a logarithmic wind
velocity profile , the wind speed at 10 meters is :
10
Uiom = S-26 ^s x - = 6-74
Instantaneous release volume. For 226 . 8 kg of hydrogen fluoride ,
the volume of gas at the release temperature and ambient pressure,
assuming ideal behavior , is :
V = (226* kg) (8314 I/bnoTK) (295^-K) = gn 7 m3
(68.1 kg/kmol) (101325 Pa)
Note: 68.1 kg/kmol is the apparent molecular weight of pure
hydrogen fluoride at 295.2 °K as calculated internally by ADAM
based on the equation (156.67- .3*T) where T is the ambient
temperature for this scenario.
Instantaneous or continuous release. For a concentration of 6
ppm, the maximum downwind distance for a continuous and
instantaneous release is 1407 meters and 4951 meters,
respectively. Based on the methodology described above, the
relea'se should be tre'ated as continuous .
Single-phase Liquid Release (High Volatility) Example
No modifications to the scenario description are necessary. The tank
volume specified in ADAM was determined based on the total mass of ethylene
oxide and the liquid density at the material's normal boiling point (882.67
kg/m3).
•
Because the ethylene oxide is stored at its normal boiling point, the
material is released as a single phase liquid (flash fraction is 0.0 at the
normal boiling point of 283.85 °K) . Also, since the flash fraction is only
0.05 at the ambient temperature of 301.15° K, ADAM models the release as a
high volatility liquid that forms a pool on the ground which subsequently
6-22
-------
boils and evaporates. It is assumed that the release from the tank occurs in
the liquid phase, i.e., the release of ethylene oxide vapor is not accounted
for. The calculations are:
Initial volume of chemical. At the storage temperature of
283.35° K, the volume of liquid ethylene oxide is:
V = ^M"8 = 26.9 m3.
Instantaneous or continuous release. For the scenario being
modeled, the evaporation rate from a slowly forming pool resulting
from the continuous release of ethylene oxide is much less than
that resulting from an instantaneous release and the formation of
a pool of a volume equal to the total volume of material in the
tank (executing the source- term model for both continuous and
instantaneous releases resulted in evaporation rates [chemical
ejection rate to the atmosphere] of 0.020 kg/s and 0.813 kg/s,
respectively) .
In light of the reduced evaporation rate of the continuous release, it
is reasonable to assume that the dispersion distance for the continuous
release would also be less than that for an instantaneous release. Based on
this assumption, the release should be treated as continuous. In fact,
executing the instantaneous release model for decreasing evaporation rates
(characteristic of a smaller size pool) results in decreasing dispersion
distances.
6.2 ALOHA
The ALOHA model falls into the interactive model category. When the
model is run, a summary of the input is provided to the user. Parts of the
text summary are duplicated here to indicate the input used. The input is
given in Tables 6-7 through 6-14. The input is entered into the model through
a series of input screens and menus from which the user selects options and
enters numerical and text values. The input used for each release example
comes from the tables cited in the title for each of the Tables 6-7 through
6-14.
6-23
-------
TABLE 6-7. TEXT SUMMARY FOR TWO-PHASE GAS RELEASE (CHOKED)
(Reference Table 5-2)
SITE DATA INFORMATION:
Location: Newport Beach, California
Building Air Exchanges Per Hour: 0.01 (User Specified)
Date & Time: Using computer's internal clock
CHEMICAL INFORMATION:
Chemical Name: ETHYLENE OXIDE Molecular Weight: 44.05 kg/kmol
TLV-TWA: 1.00 ppm IDLH: 800.00 ppm
Note: Potential or confirmed human carcinogen
Footprint Level of Concern: 800 ppm
Boiling Point: 10.70 °C
Vapor Pressure at Ambient Temperature: greater than 1 atm
Ambient Saturation Concentration: 1,000,000 ppm or 100.0%
ATMOSPHERIC INFORMATION: (MANUAL INPUT OF DATA)
Wind: 12 mph from NE No Inversion Height
Stability Class: C Air Temperature: 57.2 °F
Relative Humidity: 62% Ground Roughness: 1 centimeter
Cloud Cover: 4 tenths
SOURCE STRENGTH INFORMATION:
Gas leak from hole in spherical tank selected
Tank Diameter: 2.3 meters
Tank Volume: 6.37 cubic meters
Internal Temperature: 86.9 °F Internal Press: 2 atmospheres
Chemical Mass in Tank: 23.3 kilograms
Circular Opening Diameter: 0.5 inches
Release Duration: 8 minutes
Max Computed Release Rate: 2.59 kilograms/min
Max Average Sustained Release Rate: 2.45 kilograms/min
(averaged over a minute or more)
Total Amount Released: 11.2 kilograms
6-24
-------
TABLE 6-8. TEXT SUMMARY FOR TWO-PHASE GAS RELEASE (UNCHOKED)
(Reference Table 5-4)
SITE DATA INFORMATION:
Location: Newport Beach, California
Building Air Exchanges Per Hour: 0.01 (User Specified)
Date & Time: Fixed at November 7, 1992 & 0600 hours
CHEMICAL INFORMATION:
Chemical Name: ETHYLENE OXIDE Molecular Weight: 44.05 kg/kmol
TLV-TWA: 1.00 ppm IDLH: 800.00 ppm
Note: Potential or confirmed human carcinogen
Footprint Level of Concern: 800 ppm
Boiling Point: 10.70 °C
Vapor Pressure at Ambient Temperature: greater than 1 atm
Ambient Saturation Concentration: 1,000,000 ppm or 100.0%
ATMOSPHERIC INFORMATION: (MANUAL INPUT OF DATA)
Wind: 6 mph from SW No Inversion Height
Stability Class: C Air Temperature: 57.2 °F
Relative Humidity: 37% Ground Roughness: 1 centimeter
Cloud Cover: 6 tenths
SOURCE STRENGTH INFORMATION:
Gas leak from hole in spherical tank selected
Tank Diameter: 2.4 meters
Tank Volume: 7.24 cubic meters
Internal Temperature: 71.6 °F Internal Press: 1.48 atmospheres
Chemical Mass in Tank: 20.0 kilograms
Circular Opening Diameter: 0.5 inches
Release Duration: 8 minutes
Max Computed Release Rate: 1.88 kilograms/min
Max Average Sustained Release Rate: 1.75 kilograms/min
(averaged over a minute or more)
Total Amount Released: 6.27 kilograms
6-25
-------
TABLE 6-9. TEXT SUMMARY FOR TWO-PHASE PRESSURIZED LIQUID (Reference Table 5-6)
SITE DATA INFORMATION:
Location: Newport Beach, California
Building Air Exchanges Per Hour: 0.01 (User Specified)
Date & Time: Fixed at September 25, 1992 & 1100 hours
CHEMICAL INFORMATION:
Chemical Name: CHLORINE Molecular Weight: 70.90 kg/kmol
TLV-TWA: 0.50 ppm IDLH: 30.00 ppm
Footprint Level of Concern: 30 ppm
Boiling Point: -34.03 °C
Vapor Pressure at Ambient Temperature: greater than 1 atm
Ambient Saturation Concentration: 1,000,000 ppm or 100.0%
ATMOSPHERIC INFORMATION: (MANUAL INPUT OF DATA)
Wind: 10 mph from S No Inversion Height
Stability Class: C Air Temperature: 70.0 °F
Relative Humidity: 50% Ground Roughness: 1 centimeter
Cloud Cover: 5 tenths
SOURCE STRENGTH INFORMATION:
Liquid leak from short pipe or valve in horizontal cylindrical tank
selected
Tank Diameter: 0.762 meters Tank Length: 2.07 meters
Tank Volume: 0.94 cubic meters
Internal Temperature: 70 °F
Chemical Mass in Tank: 500.0 Ibs Tank is 16% full
Circular Opening Diameter: 0.277 inches
Opening is 0 meters from tank bottom
Release Duration: ALOHA limited the duration to 1 hour
Max Computed Release Rate: 22.7 kilograms/min
Max Average Sustained Release Rate: 12.9 kilograms/min
(averaged over a minute or more)
Total Amount Released: 193 kilograms
Note: The release was a two phase flow.
6-26
-------
TABLE 6-10. TEXT SUMMARY FOR TWO-PHASE REFRIGERATED LIQUID
(Reference Tables 5-7 and 5-8)
SITE DATA INFORMATION:
Location: MOBILE, ALABAMA
Building Air Exchanges Per Hour: 0.01 (User Specified)
Date & Time: Fixed at October 27, 1992 & 0600 hours
CHEMICAL INFORMATION:
Chemical Name: SULFUR DIOXIDE Molecular Weight: 64.07 kg/kmol
TLV-TWA: 2.00 ppm IDLH: 100.00 ppm
Footprint Level of Concern: 100 ppm
Boiling Point: -10.02 °C
Vapor Pressure at Ambient Temperature: greater than 1 atm
Ambient Saturation Concentration: 1,000,000 ppm or 100.0%
ATMOSPHERIC INFORMATION: (MANUAL INPUT OF DATA)
Wind: 7 mph from NE No Inversion Height
Stability Class: E Air Temperature: 64.4 °F
Relative Humidity: 42% Ground Roughness: 1 centimeter
Cloud Cover: 5 tenths
SOURCE STRENGTH INFORMATION:
Direct Source: 4.15 kilograms/sec Source Height: 0 m
Release Duration: ALOHA limited the duration to 1 hour
Release Rate: 249 kilograms/min
Total Amount Released: 14954 kilograms
Note: This chemical may flash boil and/or result in two phase flow.
6-27
-------
TABLE 6-11. TEXT SUMMARY FOR SINGLE-PHASE GAS RELEASE (CHOKED)
(Reference Tables 5-9 and 5-10)
SITE DATA INFORMATION:
Location: NEWPORT BEACH, CALIFORNIA
Building Air Exchanges Per Hour: 0.01 (User Specified)
Date & Time: Fixed at June 1, 1992 & 1700 hours
CHEMICAL INFORMATION:
Chemical Name: HYDROGEN FLUORIDE
Apparent Molecular Weight: 35.00 kg/kmol
TLV-TWA: 3.00 ppm IDLH: 30.00 ppm
Footprint Level of Concern: 30 ppm
Boiling Point: 19.52 °C
Vapor Pressure at Ambient Temperature: greater than 1 atm
Ambient Saturation Concentration: 1,000,000 ppm or 100.0%
ATMOSPHERIC INFORMATION: (MANUAL INPUT OF DATA)
Wind: 14 mph from NW No Inversion Height
Stability Class: D Air Temperature: 71.6 °F
Relative Humidity: 45% Ground Roughness: 1 centimeter
Cloud Cover: 3 tenths
SOURCE STRENGTH INFORMATION:
Direct Source: 51 kilograms Source Height: 0 m
Release Duration: 1 minute
Release Rate: 851 grams/sec
Total Amount Released: 51.1 kilograms
Note: This chemical may flash boil and/or result in two phase flow.
6-28
-------
TABLE 6-12. TEXT SUMMARY FOR SINGLE-PHASE GAS RELEASE (UNCHOKED)
(Reference Tables 5-11 and 5-12)
SITE DATA INFORMATION:
Location: NEWPORT, CALIFORNIA
Building Air Exchanges Per Hour: 0.01 (User Specified)
Date & Time: Fixed at November 2, 1992 & 0737 hours
CHEMICAL INFORMATION:
Chemical Name: HYDROCHLORIC ACID ANHYDROUS
Molecular Weight: 36.46 kg/kmol
TLV-TWA: 5.00 ppm IDLH: 100.00 ppm
Footprint Level of Concern: 100 ppm
Boiling Point: -121.00 °F
Vapor Pressure at Ambient Temperature: greater than 1 atm
Ambient Saturation Concentration: 1,000,000 ppm or 100.0%
ATMOSPHERIC INFORMATION: (MANUAL INPUT OF DATA)
Wind: 2.2 mph from E No Inversion Height
Stability Class: B Air Temperature: 64 °F
Relative Humidity: 36% Ground Roughness: 1 centimeter
Cloud Cover: 5 tenths
SOURCE STRENGTH INFORMATION:
Direct Source: 4250 cubic feet/min Source Height: 28 m
Source State: Gas
Source Temperature: 120 °F
Source Pressure: equal to ambient
Release Duration: ALOHA limited the duration to 1 hour
Release Rate: 367 pounds/min
Total Amount Released: 22,046 kilograms
Note: This chemical may flash boil and/or result in two phase flow.
6-29
-------
TABLE 6-13. TEXT SUMMARY FOR SINGLE-PHASE LIQUID RELEASE (HIGH VOLATILITY)
(Reference Tables 5-13 and 5-14)
SITE DATA INFORMATION:
Location: MOBILE, ALABAMA
Building Air Exchanges Per Hour: 0.01 (User Specified)
Date & Time: Fixed at June 26, 1992 & 0600 hours
CHEMICAL INFORMATION:
Chemical Name: ETHYLENE OXIDE Molecular Weight: 44.05 kg/kmol
TLV-TWA: 1.00 ppm IDLH: 800.00 ppm
Footprint Level of Concern: 800 ppm
Boiling Point: 10.70 CC
Vapor Pressure at Ambient Temperature: greater than 1 atm
Ambient Saturation Concentration: 1,000,000 ppm or 100.0%
ATMOSPHERIC INFORMATION: (MANUAL INPUT OF DATA).
Wind: 2 meters/sec from N No Inversion Height
Stability Class: E Air Temperature: 28 °C
Relative Humidity: 50% Ground Roughness: 1 centimeter
Cloud Cover: 5 tenths
SOURCE STRENGTH INFORMATION:
Liquid leak from hole in vertical cylindrical tank selected
Direct Source: 3.5 meters Tank Length: 6 meters
Tank Volume: 57.7 cubic meters
Internal Temperature: 51.3 °F
Chemical Mass in Tank: 23,852 kilograms
Tank is 47% full
Circular Opening Diameter: 0.25 inches
Opening is 0.5 meters from tank bottom
Release Duration: ALOHA limited the duration to 1 hour
Max Computed Release Rate: 6.88 kilograms/min
Max Average Sustained Release Rate: 6.53 kilograms/min
(averaged over a minute or more)
Total Amount Released: 392 kilograms
Note: The release was a two phase flow.
6-30
-------
TABLE 6-14. TEXT SUMMARY FOR SINGLE-PHASE LIQUID RELEASE (LOW VOLATILITY)
(Reference Tables 5-15 and 5-16)
SITE DATA INFORMATION:
Location: NEWPORT, CALIFORNIA
Building Air Exchanges Per Hour: 0.01 (User Specified)
Date & Time: Fixed at November 3, 1992 & 1600 hours
CHEMICAL INFORMATION:
Chemical Name: HYDROCHLORIC ACID, ANHYDROUS
Molecular Weight: 36.46 kg/kmol
TLV-TWA: 5.00 ppm IDLH: 100.00 ppm
Footprint Level of Concern: 100 ppm
Boiling Point: -85.00 °C
Vapor Pressure at Ambient Temperature: greater than 1 atm
Ambient Saturation Concentration: 1,000,000 ppm or 100.0%
ATMOSPHERIC INFORMATION: (MANUAL INPUT OF DATA)
Wind: 5 mph from NE No Inversion Height
Stability Class: C Air Temperature: 65 °F
Relative Humidity: 58% Ground Roughness: 1 centimeter
Cloud Cover: 4 tenths
SOURCE STRENGTH INFORMATION:
Direct Source: 0.005365 kilograms/sec Source Height: 0 m
Release Duration: ALOHA limited the duration to 1 hour
Release Rate: 322 grams/min
Total Amount Released: 19.3 kilograms
Note: This chemical may flash boil and/or result in two phase flow.
6-31
-------
ALOHA has source-term models for puddles, tanks, and pipes. In those
cases the observable data can be supplied to the model and it will internally
compute emission rates, temperatures, etc. The model also allows direct input
of emission rate, etc., if the release is not covered by one of the source
term sub-models. Variations between the supplied input and those used in the
ALOHA simulations are discussed below.
Two-phase Gas Release (Choked and Unchoked) Examples
ALOHA does not treat the case of a step-function emission. The size of
the tank must be assumed to be large enough to supply a steady discharge of
gas for 8 minutes. Larger tanks are immaterial as far as the source strength
is concerned. However, large tanks will overestimate doses because of an
increase in exposure time.
By trial and error, a tank having a 2.3-meter diameter was found to
produce an 8-minute discharge. The volume contained in a 3-inch pipe, limited
to a 200-to-l length-to-diameter ratio by ALOHA, is much too small to supply
an 8-minute discharge.
The assumption of a steady-state emission rate would be more
conservative than simulating an 8-minute step function using a 2.3-meter
diameter tank. However, the difference between a 2.3-meter diameter tank and
a 10-meter diameter tank was about 15 percent. Considering the fact that the
original pipe's length had already been set to zero, the above error is
probably in the model noise level.
In the unchoked release example, ALOHA warned that the supplied storage
pressure was too high for vapor-liquid equilibrium at the supplied
temperature. This is probably due to differences in the vapor pressure curve
data used in ALOHA and the data given in this document.
6-32
-------
Single-phase Gas Release (Choked) Example
ALOHA does not account for the varying molecular weight of HF as its
temperature and concentration levels change. Because of this fact, when ALOHA
is being used to model HF releases, it must be clearly understood that the
results are only an estimation of what may occur and these may be a larger
error than when simulating other compounds.
Single-phase Gas Release (Unchoked) and
Single-phase Liquid Release (Low Volatility) Examples
These examples were modeled as pure hydrogen chloride releases, although
the emissions were actually mixtures. Therefore, in ADAM, the concentrations
reported should be modified by the factor fit which has value of 0.7295 for
the gas release example and 0.01258 for the liquid release example. ALOHA
does not allow specifying an initial concentration. The initial concentration
is assumed to be 100%. It may be more correct to add new pseudo chemicals to
the database. Apparent molecular weights for the gas phase were calculated in
Sections 5.6.11 and 5.8.7. For the gas release example, an apparent gas phase
specific heat was calculated in Section 5.6.2. The gas-phase specific heat is
already supplied in the database in Appendix B for the 30 weight% hydrogen
chloride mixture.
ALOHA can not simulate an elevated release of a heavier-than-air
release. Therefore the standard ALOHA model was used, rather than the ALOHA-
DEGADIS model even though indications are that the release is heavier-than-
air.
6.3 DEGADIS
Table 6-15 lists the specific input required by the DEGADIS model and
gives a brief description of each input. It includes information on the
physical property information on the material released; release rate, height
6-33
-------
TABLE 6-15. DEGADIS MODEL INPUT PARAMETERS
Input Parameter Name
Description of Input
Parameter
Comments
UO
ZO
ZR
INDVEL
ISTAB
RML
TAMB
PAMB
RELHUM
TSURF
GASNAM
GASMW
AVTIME
TEMJET
ambient wind speed (m/s)
at height ZO
reference height (m) of
wind speed UO
surface roughness (m)
Wind velocity profile
and Monin-Obukhov
calculation flag
Pasquill-Gifford
stability category
Monin-Obukhov length (m)
ambient temperature (°K)
ambient pressure (atm)
relative humidity (%)
ground surface
temperature (°K)
three letter designation
of contaminant
contaminant molecular
weight (kg/kmol)
averaging time (s)
temperature of the
released contaminant
4.17.1
4.17.1
4.17.3
1 - DEGADIS will
calculate Monin-Obukhov
length. This was used
in all cases.
1 - A, 2 - B, etc.
Set to 0.0 for all cases
since INDVEL set to 1
4.17.5
4.17.5
4.17.5
4.15
4.2
4.16
4.8
GASUL
GASLL
Input Parameter Name
higher concentration for 4.18
determining extent (mole
fraction)
lower concentration for 4.18
determining extent (mole
fraction)
Description of Input Comments
Parameter
6-34
-------
TABLE 6-15. (CON'T.)
Input Parameter Name
Description of Input
Parameter
Comments
ZLL
INDHT
CPK
GPP
NDEN
F, C, RHO
ERATE
ELEJET
DIAJET
TEND
DISTMX
receptor elevation (m) 4.18
switch to indicate if
atmosphere heat transfer
is to be included
contaminant heat
capacity constant (J/kg
•K)
contaminant heat
capacity power
number of density
triplets
density triplet of
contaminant mole
fraction (F),
contaminant
concentration (C,
kg/m3) , and mixture
density (RHO, kg/m3)
emission rate (kg/s)
jet elevation (m)
jet diameter (m)
duration of release (s)
distance between output
points (m)
If 0, no calculation to
be done. In all cases
here it is set to 0.
Since INDHT is 0, this
value is ignored
Since INDHT is 0, this
value is ignored
Set to 2 for all cases
with aerosol or HF with
changing molecular
weight. Otherwise 0.
4.11
0, 0, Pa (triplet 1)
°. Pr*i> Prei (triplet 2)
Only required if aerosol
present or release is of
HF
4.8
4.14
4.12
0.0 indicates continuous
release
6-35
-------
(jet), and duration; emission densities; concentration averaging time; maximum
downwind distance between output points; meteorological information; and site
roughness characteristics. In the comments column for many of the input
parameters of Table 6-15 the subsection in Section 4 is given which discusses
how that parameter is calculated.
Table 6-16 shows the value of each input parameter used for the first
seven release class examples. The eighth example was not simulated due to the
fact that the release is not heavier-than-air. Standard passive atmospheric
dispersion methods should apply. In the column header for each release class
(RC) example, the table in Section 5 from which the input was taken is
referenced. The tables in Section 5, in turn, each reference the section in
which the calculation was done.
The release class examples are:
Release Class (RC) Title
1 Two-Phase Gas Release (Choked)
2 Two-Phase Gas Release (Unchoked)
3 Two-Phase Pressurized Liquid
4 Two-Phase Refrigerated Liquid
5 Single-Phase Gas Release (Choked)
6 Single-Phase Gas Release (Unchoked)
7 Single-Phase Liquid Release (High Volatility)
8 Single-Phase Liquid Release (Low Volatility) •
Averaging time was taken as the lesser of the release duration or time
requested for averaging. When the averaging time used is less than the
requested averaging time, the output must be converted to represent the
requested averaging time. See Section 7.3 for further discussion on the
output conversion technique.
6-36
-------
TABLE 6-16. DEGADIS INPUT. A () INDICATES REFERENCE TABLE OF INPUT DATA.
Input
Parameter
Name
UO
ZO
ZR
INDVEL
I STAB
RML
TAMB
PAMB
RELHUM
TSURF
GASNAM
GASMW
AVTIME
TEMJET
GASUL
GASLL
ZLL
INDHT
CPK
GPP
NDEN
F, C,
RHO
F, C,
RHO
ERATE
ELEJET
DIAJET
TEND
DISTMX
Input
Values
RC 1
(5-1)
5.37
4.57
0.01
1
3
0.0
287.5
1.0
62
287.5
BOX
44.05
0.0
283.85
1.0
0.03
0.0
0
0.0
0.0
2
0, 0,
1.220
1, 1.916,
1.916
0.0634
3.66
0.01347
0.0
0.1
Input
Values
RC 2
(5-2)
2.68
6.1
0.01
1
3
0.0
296.5
1.0
37
296.5
ETO
44.05
0.0
283.85
1.0
0.03
0.0
0
0.0
0.0
2
0, 0,
1.183
1, 1.934,
1.934
0.0503
3.66
0.0127
0.0
0.1
Input
Values
RC 3
(5-3)
4.47
10
0.01
1
3
0.0
294.3
1.0
50
294.3
CL2
70.91
678.0
239.1
1.0x10''
1.0x10-*
0.0
0
0.0
0.0
2
0, 0.
1.191
1, 20.11,
20.11
0.3170
2.0
0.00704
0.0
50.0
Input
Values
RC 4
(5-4)
3.13
6
0.01
1
5
0.0
291.15
1.0
42
291.15
S02
64.06
900.0
263.13
1.0xlO-s
5.0x10"'
0.0
0
0.0
0.0
2
0, 0,
1.206
1, 13.76,
13.76
4.154
0.3049
0.0127
0.0
50.0
Input
Values
RC 5
(5-5)
6.26
6.1
0.01
1
4
0.0
295.15
1.0
45
295.15
HF
20.0
45.0
313.6
1.2xlO-s
6.0x10-'
0.0
0
0.0
0.0
2
0, 0,
1.188
1, 1.360,
1.360
0.8513
3.66
0.06649
0.0
50.0
Input
Values
RC 6
(5-6)
0.98
10
0.01
1
2
0.0
290.9
1.0
36
290.9
HCL
33.1
720.0
322.0
2.74x10-'
1.37xlO-4
0.0
0
0.0
0.0
0
2.515
8.54
0.86
0.0
50.0
Input
Values
RC 7
(5-7)
2.0
10
0.01
1
5
0.0
301.0
1.0
50
301.0
ETO
44.05
1800.0
283.85
i.eoxio-3
S.OOxlO-4
0.0
0
0.0
0.0
0
.
0 . 1220
0.5
0.1372
0.0
50.0
6-37
-------
In all cases, the release is initially modeled as a vertically pointing
jet. The simulation of a vertical jet should represent the lower bound of
predicted concentrations since such a release represents maximum dilution.
All other jet orientations should result in either the same amount of mixing
or less as long as the jet is not directed into the wind. It is suggested
that an upper limit of the concentration impacts be determined by also
simulating the release as a low-momentum, ground-level source. The source
diameter would need to be estimated by using the low-momentum formula given in
Section 4.12.1.
In all cases, it was initially assumed that the release was continuous
(TEND was set to 0.0). After each simulation, the travel time to the distance
of interest was estimated. If the travel time was longer than the release
duration, a transient simulation was performed. A transient simulation is
made by changing the TEND parameter in the DEGADIS input file to the release
duration. If the simulation can be assumed to be continuous rather than
transient, the interpretation of the output is made simpler. See Section 7.3
for further discussion on the interpretation of the output.
Two-phase Gas Release (Choked and Unchoked) Examples
The concentrations of concern are so high and the dispersion is so rapid
that the jet sub-model needed to be run twice. The first time the distance
between outputs was too large (50 m) to see the distance downwind of a 3%
concentration. Therefore, the model had to be-rerun with a step decrease of
0.1 meters.
6.4 HGSYSTEM
Table 6-17 presents a summary of the models that were used in analyzing
the eight release class examples. The rationale for model selection and
implementation is presented in the following sub-sections. It should be noted
that there may be equally valid alternative methods for modeling the
6-38
-------
TABLE 6-17. SUMMARY OF THE HGSYSTEM MODELS USED
HGSYSTEM Models Used
Release
Class
1
2
3
4
5
6
7
8
Chemical
Ethylene
Oxide •
Ethylene
Oxide
-
Chlorine
Sulfur
Dioxide
Hydrogen
Fluoride
Hydrogen
Chloride
Ethylene
Oxide
30%
Hydrochloric
Acid
Release
Conditions
Two -Phase Gas
Release
(Choked)
Two -Phase Gas
Release
(Unchoked)
Two -Phase
Pressurized
Liquid
Two -Phase
Refrigerated
Liquid
Single -Phase
Gas (Choked)
Single -Phase
Gas (Unchoked)
Single -Phase
Liquid Release
(High
Volatility)
Single -Phase
Liquid (Low
Volatility)
Post
Near Far Field Process -
Field ing
HEGADASS HEGADASS HSPOST-
GET2COL
HEGADASS HEGADASS HSPOST-
GET2COL
HEGADASS HEGADASS HSPOST
PLUME PGPLUME
HEGADAST HEGADAST' HTPOST
PLUME PGPLUME
PLUME HEGADASS GET2COL
HEGADASS HEGADASS
6-39
-------
consequences of these scenarios other than the ones presented in this
analysis.
Averaging Time Issues
HGSYSTEM allows for finite duration releases. Accordingly, the
averaging time selection may affect the accuracy of the model, especially if
the averaging time is larger than the release duration. For scenarios for
which this is true, HGSYSTEM models were applied in the following manner. The
averaging time for ay (AVTIMC) was set equal to the release duration. Thus
the amount of plume meander was restricted to the length of time that a plume
actually existed. That is, if the duration of a release was 5 minutes and we
are interested in a 15-minute average, it is not appropriate to use a ay
averaging time that is longer than the duration of the release (5-minutes not
15-minutes). This approach is taken because there is a 10-minute period in
which the plume does not exist and, therefore, will not undergo any meander.
For situations where the model predicts that a. gas blanket (i.e., a situation
in which the cloud is not advected downwind because of density differences)
exists for a period longer than the actual release duration, the length of
time that the gas blanket persisted should be used to establish the try
averaging time to determine the period over which the gas blanket remained
(this would require iterative model runs).
There is a second aspect to calculating the average concentration for
this situation. It is related to the integration of the plume in the along-
wind (x axis). At a given downwind distance, the plume concentrations must be
integrated to account for the variations in concentration as a function of
time. To illustrate, consider a case in which the plume would persist at a
given location for a period of approximately 5 minutes (the HEGADAST model
will calculate this time precisely) and during the remaining 10-minutes, of
the 15-minute period, no plume would be present. In order to compute an
accurate average concentration, this integration of the time-varying
concentrations needs to be included in the model predictions. This can be
accomplished by using the ax averaging option in the HTPOST post processor
6-40
-------
program for HEGADAST. Thus, in the example given above where the averaging
time is 15-minutes, the
-------
In this simulation the ICNT parameter was set to 1 to enable the model
to provide concentration isopleths for the upper explosive limit (UEL) and
lower flammability limit (LFL). The upper concentration limit (CU) was set
equal to the UEL (100 % or 1.44 kg/m3) and the lower concentration limit (CL)
was set equal to LFL (3% or 5.4X10"2 kg/m3). The calculation output step
(DXFIX) was set to 0.5 meters. This enabled precise calculations in the near
field region so that the distance to the LOG could be identified. The size of
the area source was arbitrarily selected (see POOL DATA in Table 6-18), and a
second modeling run was made to ensure that the distance to the LOG was not
overly sensitive to this assumption. It has been our experience for this
model and this type of release that the results are relatively insensitive to
the specification of area source size.
Two-phase Gas Release (Unchoked) Example
This case described a two-phase release of ethylene oxide under unchoked
conditions. The consequences of this release were simulated using the
HEGADASS model, which was selected for the same reasons as in the first
release example. Table 6-19 presents the input parameters selected for
simulating this analysis. Much of this input was taken from Table 5-3. The
rest of the input is explained in the HGSYSTEM documentation. For this, case,
the model was initialized in an analogous manner to the first release example
using the data provided.
Two-phase Pressurized Liquid Example
This release represents a two-phase pressurized liquid release from a
chlorine cylinder. The release conditions for this case represent a two-phase
horizontal jet release which is directed downwind. To compute more accurate
concentrations for this case, the transient version of HEGADAS should be used.
This approach is preferable because the release duration is less than the
averaging time (the release duration is 11 minutes and the averaging time of
concern is 15 minutes). However, when the HEGADAST model was executed for
6-42
-------
TABLE 6-18. INPUT PARAMETERS USED IN THE HEGADASS MODEL TO SIMULATE RELEASE
CLASS 1 (Reference Table 5-1)
TITLE SCENARIO 1 CHOKED ETHYLENE OXIDE RELEASE
CONTROL
ICNT
ISURF -
AMBIENT
AIRTEMP -
ZAIRTEMP-
RHPERC -
UO
ZO
TGROUND -
DISP
ZR
PQSTAB -
AVTIMC -
CROSSW -
GASDATA
GASFLOW -
TEMPGAS -
CPGAS
MWGAS
WATGAS -
HEATGR -
CLOUD
DXFIX -
NFIX
NFIX*DXFIX
XEND
GAMIN
CU
CL
POOL
PLL
PLHW
1
3
14.4
4.6
62.
5.37
4.6
14.4
0.01
C
10.
2
.0634
10.70
24.49
44.
0.
29.
.5
300
3000.
0.00001
1.44
0.05397
.2
.1
*
*
*
*
*
*
* C
* M
* %
* M/S
* M
* C
*
*
*
* M
*
* S
*
*
•k
*
* KG/S
* C
* J/MOLE/C
* KG/KMOLE
* -
*
*
*
*
* M
*
* M
* KG/M3
* KG/M3
* KG/M3
*
*
*
* M
* M
-> DATA BLOCK: CONTROL PARAMETERS
output code (isocontours, cloud contents)
code for surface heat/water transfer
-> DATA BLOCK: AMBIENT CONDITIONS
air temperature at height z - ZAIRTEMP
height at which AIRTEMP is given
relative humidity
wind speed at height z = ZO
height at which UO is given
earth's surface temperature
-> DATA BLOCK: DISPERSION DATA
.
surface roughness parameter
Pasquill stability class
averaging time for concentration
<<7y>formula (don't normally change)
--> DATA BLOCK: GAS DATA
gas emission rate (excl. water pick-up)
temperature of emitted gas
specific heat of emitted gas
molecular weight of emitted gas
water pick-up by gas (don't norm. change)
gas group for natural -conv. heat flux
-> DATA BLOCK: CLOUD OUTPUT CONTROL
fixed- size output step length
fixed steps up to distance x —
x at which calculations are stopped
CA (cone.) at which calcs . are stopped
upper concentration limit
lower concentration limit
-> DATA BLOCK: POOL DATA
pool length
pool half -width
6-43
-------
TABLE 6-19. INPUT PARAMETERS USED IN THE HEGADASS MODEL TO SIMULATE RELEASE
CLASS 2 (Reference Table 5-2)
TITLE SCENARIO 2 UNCHOKED ETHYLENE OXIDE
PONTROT
Owli 1 DAJl^
ICNT
ISURF -
AMRTFNT
E\I i n j. diHATA RTflPV • PnNTRfYT PAR AMTTTPRC
L/Ain DIAsOR. . OU1N -i-RU-Li ir AJxAnd J. £jl\.o
output code (isocontours, cloud contents)
code for surface heat/water transfer
•> OATA RTDPTf- AMRTFTJT PONDTTTOTJ^
** IsAXA QiAJ\*K^ . AILO J-ul*! J. \s\JIXLtl. X XwlilO
air temperature at height z - ZAIRTEMP
height at which AIRTEMP is given
relative humidity
wind speed at height z - ZO
height at which UO is given
earth's surface temperature
surface roughness parameter
Pasquill stability class
averaging time for concentration
formula (don't normally change)
>T\ATA RT f\f*V ' (7A^ OAT A
1JA.LA Oi^A/wfw . V7Aw 1/AXA
gas emission rate (excl. water
temperature of emitted gas
specific heat of emitted gas
molecular weight of emitted gas
water pick-up by gas (don't norm. change)
gas group for natural - conv . heat flux
"> DATA RTjOfK* fTjOTTT) nTTTPTTT PONTROT
^ Wc\in JJl^/VfCt . wi^UUlS WVJX.LU1. OWli LrvWi^i
fixed- size output step length
fixed steps upto distance x =
x at which calculations are stopped
CA (cone.) at which calcs . are stopped
upper concentration limit
lower concentration limit
- rvATA RT r\CW • POflT DATA
pool length
pool half -width
6-44
-------
this scenario with the predefined meteorological conditions (stability class
"C"), the source blanket was too thin for the model to accurately compute
downwind concentrations. This resulted from the simulation of a small release
rate (0.32 kg/sec) and the convective atmospheric conditions. It was found
that the HEGADAST model would run for this scenario under stable, worst-case,
meteorological conditions. From a pragmatic perspective, the HEGADASS model
runs provide a reasonable estimate of impacts for this case because the
uncertainty in
-------
TABLE 6-20. CONCATENATED INPUT FILE FOR RELEASE CLASS 3 (Reference Table 5-3)
* (case: run ;
TITLE EPA SCENARIO 3 HI
CONTROL
ICNT
ISURF -
AMRTFNT
miijxi-u.i .L
AIRTEMP -
ZAIRTEMP-
RHPERC -
UO
ZO
TGROUND -
nTOp
ZR
PQSTAB -
AVTIMC -
CROSSW -
pAcnA*TA
\s£\& U A J. f\
GASFLOW -
TEMPGAS -
CPGAS
MWGAS
WATGAS -
CLOUD
DXFIX
NFIX
XEND
GAMIN
CU
CL
POOL
PLL
PLHW
0
3
21.2
10.
50.
3
10.
21.2
.01
F
900.
2
.3170
-34.
6.8
70.9
0.
1.
1000.
3000.
0.000001
0.0001
0.00001
1.
".5
a-d standard 2
started at poc
EGADASS RUN Ct
*
*
*
*
^
* C
* M
* %
* M/S
* M
* C
*
* ........
*
* M
*
* S
*
*
.
* KG/S
* C
* J/MOLE/C
* KG/KMOLE
* -
*
^
*
* M
*
* M
* KG/M3
* KG/M3
* KG/M3
*
* M
* M
.nput i lie airuuiJNU.ttai
>1, normal thermodynamics)
ILORINE RELEASE
•> DATA BLOCK: CONTROL PARAMETERS
output code (isocontours, cloud contents)
code for surface heat/water transfer
^ DATA RT.On?' AMRTFMT rONFnTTTON5?
^ i^AlA DLAJ\jf^ . ATLDlHJ.il wwlNLrl J. ILrliO
air temperature at height z - ZAIRTEMP
height at which AIRTEMP is given
relative humidity
wind speed at height z - ZO
height at which UO is given
- earth's surface temperature
•
>T\A'PA DT f\f*V . r\T CPTT1? dTf^W ftATA
UA1A DLAJ^jf**. l/lol: iMxo l\JiN UA1A.
surface roughness parameter
Pasquill stability class
averaging time for concentration
formula (don't normally change)
•> HATA RT/lPlf- f5A<> DATA
^ unm. ijj^jur^ . \jt\& J^^\IA
gas emission rate (excl. water pick-up)
temperature of emitted gas
specific heat of emitted gas
molecular weight of emitted gas
water pick-up by gas (don't norm. change)
-> DATA BLOCK: CLOUD OUTPUT CONTROL
fixed- size output step length
fixed steps upto distance x - NFIX*DXFIX
x at which calculations are stopped
CA (cone.) at which calcs. are stopped
upper concentration limit
lower concentration limit
-> DATA BLOCK: POOL DATA
pool length
pool half-width
6-46
-------
TABLE 6-21. PLUME MODEL INPUT FOR RELEASE CLASS 4 (Reference Table 5-4)
TITLE SCENARIO 4 S02 RELEASE
GASDATA
TEMPGAS
MFGAS
MFH20
MWGAS
CPGAS
PIPE
DMDT
DEXIT
ZEXIT
PHISTK
DURATION
-
-
-
-
-
-
-
-
-
—
-10.0
100 . 00
0.0
64.1
21.0
.4.154
.0127
.3049
90.00
1585
AMBIENT CONDITIONS
ZO
UO
AIRTEMP
AIRPRESS
RHPERC
DISP.
ZR
PQSTAB
TERMINAT
SLST
DLST
ZLST
DXLST
ULST
BETLST
•
MATCH
RULST
UJET/UAMB-1
RELST
RGLST
RNLST
DISPERSION
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
6.0
3.13
18.0
1.00
00.0
0.01
E
500.
-1E6
-.35
-500
-0.1
1E-7
.2
.3
.3
.1
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
PHYSICAL PROPERTIES OF GAS
CELSIUS
PERCENT
PERCENT
g/mol
J/mol/C
TEMPERATURE OF POLLUTANT
MOLE FRACTION OF POLLUTANT
MOLE FRACTION OF WATER
MOLECULAR WEIGHT POLLUTANT
ISOBARIC SPECIFIC HEAT
PIPE EXIT- PLANE (CHOKE- FRONT) CONDITIONS
KG/S
M
M
DEGREES
S
ATMOSPHERIC
M
M/S
CELSIUS
ATMOSPHERES
PERCENT
DISPERSION
M
DISCHARGE RATE
EFFECTIVE ORIFICE DIAMETER
HEIGHT ABOVE (LEVEL) GROUND
RELEASE DISCHARGE ANGLE
RELEASE DURATION (<0 FOR STEADY)
AMBIENT CONDITIONS
REFERENCE HEIGHT
WIND VELOCITY AT HEIGHT ZO
AIR TEMPERATURE
AMBIENT PRESSURE
RELATIVE HUMIDITY
DATA
SURFACE ROUGHNESS PARAMETER
PASQUILL STABILITY CLASS
JET/PLUME DEVELOPMENT TERMINATION CRITERIA
M
M
M
M
M/S
PERCENT
LAST REQD. DOWNWIND DISP.
LAST REQD. PLUME DIAMETER
LAST REQD. PLUME CENTROID RISE HT
.
LAST REQD. HORIZONTAL DISPLACEMENT
LAST REQD. (MEAN) PLUME VELOCITY
LAST REQD. POLLUTANT CONCENTRATION
MATCHING CRITERIA FOR HEGADAS/PGPLUME
LAST REQD. ABS . VALUE OF
LAST REQD. JET/(JET+HEG) ENTRAINM
.
MAX. BUOYANCY EFFECT FOR ADVECTION
MAX. BUOY. EFF. FOR PASS.
6-47
-------
TABLE 6-22. PGPLUME LINK FILE CREATED BY THE PLUME MODEL FOR RELEASE CLASS 4
TITLE SCENARIO 4 S02 RELEASE
GASDATA * released gas composition datablock.
CPGAS- 21.0 * pollutant specific heat (J/mol/C).
MWGAS- 64.1 * pollutant molecular weight (g/mol).
GASFRAC- 1.00 * release mole-fraction pollutant (-).
WATGAS- O.OOOE-01 * release mole-fraction water-vapour (-).
GEOMETRY * plume geometry at matching datablock.
DXPLUME- 161. * matching plane displacement (m).
ZPLUME- 29.0 * centroid height above ground (m).
DPLUME- 41.6 * near-plume (effective) diameter (m).
PHIPLUME- -0.300 * plume axis orientation (degrees).
STATE * plume dynamic/thermodynamic state.
UREL- -0.477 * plume relative velocity (m/s).
CMASS- 7.929E-04 * nearfield mass-concentration (kg/m3).
RREL- 1.390E-03 * plume (mean) excess density (kg/m3).
DURATION- 1.585E+03 * (steady) release duration (s).
AMBIENT * ambient atmosphere datablock.
AIRTEMP- 19.2 * ambient (air) temperature (C).
AIRPRESS- 0.997 * ambient (absolute) pressure (atm).
RHPERC- O.OOOE-01 * ambient (relative) humidity (%).
UATM- 6.02 * wind-speed at centroid height (m/s).
RATM- 1.20 * ambient atmosphere density (kg/m3).
DISP * Pasquill/Gifford dispersion data.
ZR- l.OOOE-02 * ground surface roughness (m).
PQSTAB- E * Pasquill/Gifford stability class (-).
*AVTIMC- 600. * concentration averaging time (s).
TERMINAT * output control datablock.
XFIRST- 161. * first required downwind distance (m).
*STEP= 100. * arithmetic series step-length (m).
*NSTEP= 10 * maximum number of (arithmetic) steps (-).
*FACTOR- 1.20 * scale factor for geometric series (-).
*XLAST- 1.016E+04 * last required downwind distance (m).
*VFLAST- 2.99 * last required mole concentration (ppm) .
Note: Parameters which are marked with a * need to by updated to run PGPLUME
6-48
-------
TABLE 6-23. PGPLUME INPUT FOR RELEASE CLASS 4
* (case of pressurised release with near-source data from HFPLUME/PLUME)
*
* NOTE: HFPLUME/PGPLUME or PLUME/PGPLUME .PGL link file is appended
* at the end of this file]
*
DISP * Pasquill/Gifford dispersion data.
*
AVTIMC - 900. * SECS averaging time for concentrations
*
ZR- l.OOOE-02 * ground surface roughness (m).
PQSTAB- E * Pasquill/Gifford stability class (-).
TERMINAT * output control datablock.
* [XFIRST from HFPLUME/PLUME]
*
STEP - 100.0 * M arithmetic-progression step length
NSTEP - 10 * maximum number of such steps
FACTOR - 2.0 * . distance scale (geometric factor)
XLAST - 10000.0 * M last required cross-section distance
VFLAST - 3.0 * PPM last required centre-line concentration
* [GASDATA, GEOMETRY, STATE and AMBIENT data blocks from HFPLUME/PLUME]
XFIRST- 161. * first required downwind distance (m).
TITLE SCENARIO 4 S02 RELEASE
GASDATA * released gas composition datablock.
CPGAS- 21.0 * pollutant specific heat (J/mol/C).
MWGAS- 64.1 * pollutant molecular weight (g/mol).
GASFRAC- 1.00 * release mole-fraction pollutant (-).
WATGAS- O.OOOE-01 * release mole-fraction water-vapour (-).
GEOMETRY * plume geometry at matching datablock.
DXPLUME- 161. * matching plane displacement (m).
ZPLUME- 29.0 * centroid height above ground (m).
DPLUME- 41.6 * near-plume (effective) diameter (m).
PHIPLUME- -0.300 * plume axis orientation (degrees).
STATE * plume dynamic/thermodynamic state.
UREL- -0.477 * plume relative velocity (m/s).
CMASS- 7.929E-04 * nearfield mass-concentration (kg/m3).
RREL- 1.390E-03 * plume (mean) excess density (kg/m3).
AMBIENT * ambient atmosphere datablock.
AIRTEMP- 19.2 * ambient (air) temperature (C).
AIRPRESS- 0.997 * ambient (absolute) pressure (atm).
RHPERC- O.OOOE-01 * ambient (relative) humidity (%).
UATM- 6.02 * wind-speed at centroid height (m/s).
RATM- 1.20 * ambient atmosphere density (kg/m3).
6-49
-------
The following summarizes the important model input for the PLUME model.
The temperature of the gas emitted (TEMPGAS) was taken from the Phase I
report. The release duration (DURATION) was set to 26.4 minutes. This
parameter was then passed to PGPLUME, which incorporates the effect of release
duration into the dispersion calculations once the release travel time exceeds
the release duration. The release angle was set to vertical (PHISTK 90).
Because this version of PLUME cannot simulate the condensation of water vapor
within the jet, the relative humidity was set to 0 percent.
In the PGPLUME model, the averaging time (AVTIMC) was set to 900
seconds, which corresponds to the averaging time of the STEL. The model
output parameters were selected to provide concentration listings at 100 meter
intervals (STEP) for the first 1000 meters (NSTE-10) and then a geometric
distance step of 1.2 beyond that distance.
Single-phase Gas Release (Choked) Example
This scenario describes an accidental release of anhydrous HF as a
result of a relief valve opening during the filling of a storage vessel.
There are several uncertainties regarding the definition of this scenario.
The first problem is that the storage conditions specified in the resource
document are unrealistic (a temperature of 165 °F and a pressure of 30 psig
results in a non-equilibrium condition). Even the conditions in the exit pipe
(197 °F) are out of equilibrium. In addition, from the data presented, it is
unclear what the release pressure of the relief valve is (does it release at
30 psi or is this the pressure at which it resets). For the conditions
defined, the thermodynamics subroutine in HFPLUME determined that the storage
conditions are above saturation conditions and the model would not execute
because this is a non-equilibrium situation. As a result of the uncertainty
in the storage conditions provided, this release was modeled in a simplified
fashion by making some assumptions regarding storage conditions.
Based on the temperature and pressure given in the definition of this
scenario, or even if more reasonable storage conditions are assumed, this
6-50
-------
release will not result in a plume that is denser than ambient air. This was
demonstrated by running HFPLUME (version 2.1) for more representative storage
conditions with the result that a transition was made from HFPLUME to PGPLUME.
The definition of this scenario is further complicated by the fact that
the release duration is only 45 seconds. Using the HFPLUME and PGPLUME
models, it is not possible to obtain an accurate 15-minute average
concentration for a release of this short duration.
As a result of these issues, this case was modeled as a ground level
area source with HEGADAST and assuming that the release rate and release
temperatures given are correct. Thus, in modeling the impacts of such a
release, we have not accounted for the fact that a release height of
3.7 meters was given. As a result of the modeling approach taken (ground
level area source), we may have overstated near field impacts.
Table 6-24 presents the HEGADAST input parameters used in this analysis.
Much of this input was taken from Table 5-9. The rest of the input is
explained in the HGSYSTEM documentation. The following presents an overview
of parameters that require additional explanation. The ICNT parameter was set;
to 1, which will provide contours of the cloud width and height to the upper
and lower concentration limits. These contours reflect an averaging time
equivalent to the setting of the AVTIMC parameter (for this scenario, it is 45
seconds). For this scenario, the lower concentration limit (CL) was set to
4.088 xlO"6 kg/m3 (6 ppm). The model output provides the cloud width and
height, to the lower concentration limit for each time step and downwind
distance. Thus, one can examine the cloud width as a function of time.
The emission rate in HEGADAST is controlled by the TIMEDATA block.
There are three aspects to this data block: TSTPOOL; TSTEPR; and SOURCE. The
TSTPOOL input reflects the start time of the emission data. For this case it
is set to 0.0, which implies that the emissions are released at the start of
the simulation. This value would be set to a positive time if near field
6-51
-------
TABLE 6-24. HEGADAST INPUT PARAMETERS FOR RELEASE CLASS 5
(Reference Table 5-5)
TITLE SCENARIO 5 HF RELEASE
fiONTROT
\J\J1X X AwX*
ICNT
ISURF -
AMRTFNT
AI ID x Cii.i x
AIRTEMP -
ZAIRTEMP-
RHPERC -
UO
ZO
TGROUND -
ni^p
X/Xwx
ZR
PQSTAB -
AVTIMC -
CROSSW -
HA on ATA
VJAO l/r\ X «.
THERMOD -
CPGAS
MWGAS
TEMPGAS -
HFLIQFR -
PTDTTTJ
VjJJ_/UU
XSTEP
GAMIN
CU
CL
TAT P
uAXrfW
TSTAR
TSTAR
TSTAR
TSTAR
TSTAR
TSTAR
TSTAR
1
3
22.
6.1
45.
6.26
6.1
22.
.01
D
45.
2 , ,
2
29.0
20.01 .
27.00
1.0
50.
5E-5
5E-5
4.088E-6
10
12
15
20
40
60
80
* .... .
*
*
*
*
•# _ _ ..
*
* C
* M
* %
* M/S
* M
* C
*
* . ._..
*
* M
*
* S
*
*
•it .
*
*
* J/MOLE/C
* KG/KMOLE
* C
* -
*
*
*
* M
* KG/M3
* KG/MS
* KG/MS
*
*
*
* S
* S
* S
* S
* S
* S
* S
^ HAT A RT nrtf • rOIJTO HT PAR AMFTFT? DATA RTDCTf- AMRTFNT rfWnTTTOTJ*!
** \Jt\Ln. Q±#J\tc* • fUTLD X £u. ftATA RTjOfK' HT'sPVR^TON DATA
^ UAXA OlAJ\jr* . lformula (don't normally change)
"5> HATA RT/VK- HF tiATA
** UAxA OXA/w^ . Clx X^AX/\
HF thermodynamics model
specific heat of HF
molecular weight of HF
temperature of HF pool (see NOTE 1)
liquid mass fraction of pool (see NOTE1)
> DATA RTOfJf- fT-DITT) OTTTPTTT fONTROT
^ U£\l£\. DLAJ\JK^, VJXA^UX/ WWXXUX OW1 nATA fiT-Oftf- OTITPTTT TTMF9
s* Ut\in OluAJ\jE< . WUXXUX X Xlll.«k>
6-52
-------
TABLE 6-24 (CONTINUED)
TSTAR
TSTAR
TSTAR
TSTAR
TSTAR
TSTAR
TSTAR
TSTAR
TSTAR
TSTAR
TSTAR
-
-
-
_
-
_
-
-
—
-
_
100
120
140
160
180
200
220
240
260
280
300
TIMEDATA
TSTPOOL
TSTEPR
0.
5.
**
*
*
SOURCE-
SOURCE-
SOURCE-
SOURCE-
SOURCE-
SOURCE-
SOURCE-
SOURCE-
SOURCE-
pool :
(m
.1
.1
.1
.1
.1
.1
.1
.1
.1
* s
* s
* s
* s
* s
* s
* s
* s
* s
* s
* s
*
*
*
* s
* s
-> DATA BLOCK; POOL DATA
start time (zero data for time
-------
dispersion were described using another model and HEGADAST was started at some
distance downwind of the release. The TSTEPR parameter is used to set the
time step for the emission data. For this simulation this parameter was set
equal to 5 seconds. This parameter provides a check to ensure that emission
data are consistent with the dispersion calculations. The SOURCE data
provides a description of the size of the area source (pool in the context of
the model) and the emission rate. The size of the area source was set to a
radius of 0.1 meter. (Model predictions are relatively insensitive to the
area source radius value beyond 100 meters.) The number of source steps to be
included depends on the actual release duration and the value"of TSTEP. Thus,
for a 45-second release, 9 SOURCE input was used. The TSTAR parameter
contained in the CALC .data block specifies the times at which dispersion
calculations are performed. The best way to establish the TSTAR value is to
estimate the plume transport time to the distance of concern and then bracket
this time by several other times. By setting TSTAR in this manner, it ensures
that peak concentrations are being calculated. Also, keep in mind that
computer time will increase as the number of TSTARs specified in a simulation
increases. In the CLOUD data block, the parameter XSTEP is used to specify
the downwind distances at which calculations are performed. For this
scenario, an XSTEP of 50 was selected. For a specified TSTAR, the
concentration of the cloud is computed over a distance of XSTEP. For example,
if TSTAR is specified as 60 seconds and XSTEP is 50 meters, then the model
will compute the concentrations at a time of 60 seconds from the time of
release at 50-meter intervals in a downwind direction.
Because the release duration is 45 seconds, a ay averaging time of 45
seconds was used (AVTIMC). This reflects the amount of plume meandering that
would occur while the plume is actually present.
The following summarizes the issues associated with these modeling runs.
First, the modeling results at ground level near the point of release are
conservative because this release is modeled as a ground level area source.
Modeling this release using HFPLUME resulted in predicted concentrations at
the fenceline of approximately 700 ppm versus 984 ppm predicted by HEGADAST.
6-54
-------
Second, there is some confusion regarding the terminology used in the
reference document, for estimated cloud times. For example, the duration of
the maximum concentration at the site boundary will be, by definition, a
15-minute average. From HEGADAST, the peak (45-second) concentration will
persist about 20 seconds. Similarly, the peak concentration will drop to 6
ppm after about 15-20 seconds.
Third, horizontal and vertical concentration isopleths to a
concentration of 6 ppm can be obtained from the HEGADAST report file (YCL and
ZCL). These concentrations, however, reflect a 45-second average and are
presented as a function of time. Thus, the model will provide an estimate of
the area encompassed within the 6 ppm isopleth at various time sequences.
From these data it is possible to determine the area or volume of the 6 ppm
isopleth as a function of time. This area should be contrasted to the area
that is obtained from a steady state model with a finite release duration
correction, such as SLAB or HEGADASS. These models will produce an isopleth
that is independent of time.
Single-phase Gas Release (Unchoked) Example
The initial phase of this release was modeled using the PLUME model. A
copy of the input to this model is presented in Table 6-25. Much of this
input was taken from Table 5-11. The rest of the input is explained in the
HGSYSTEM documentation. The following describes the important input to this
model as it pertains to this example. The molecular weight used for this
simulation was a mean molecular weight based on the mean molar composition
(33.1 g/g-mole). The parameter MFGAS was invoked to account for the fact that
the plume is only 72.9% HCL. The MFGAS parameter computes mean cloud
properties and then, based on the fraction of the pollutant of interest, the
total concentration is multiplied by this fraction to estimate the
concentration for the component of concern. The duration of the release (720
seconds) was input using the DURATION parameter. This value was passed to the
far field dispersion model (HEGADAST or PGPLUME depending on the plume density
characteristics).
6-55
-------
TABLE 6-25. PLUME MODEL INPUT FOR RELEASE CLASS 6 (Reference Table 5-6)
TITLE SCENARIO 6
GASDATA
TEMPGAS -
MFGAS
MFH20
MWGAS
CPGAS
PIPE
DMDT
DEXIT
ZEXIT
PHISTK
DURATION -
49.0
72.95
0.1
33.1
21.0
2.52
0.8636
8.54
90.00
720
AMBIENT CONDITIONS
ZO
UO
AIRTEMP -
AIRPRESS -
RHPERC
DISP
ZR
PQSTAB
TERMINAT
SLST
DLST
ZLST
DXLST
ULST
BETLST
MATCH
RULST
UJET/UAMB-1
RELST
RGLST
RNLST
DISPERSION
10.0
1.0
17.9
1.00
36.0
0.01
B
500.
-1E6
-.35
-500
-0.1
1E-7
.2
.3
.3
.1
* PHYSICAL PROPERTIES OF GAS
*
* CELSIUS TEMPERATURE OF POLLUTANT
* PERCENT MOLE FRACTION OF POLLUTANT
* PERCENT MOLE FRACTION OF WATER
* g/mol MOLECULAR WEIGHT POLLUTANT
* J/mol/C ISOBARIC SPECIFIC HEAT
*
* PIPE EXIT -PLANE (CHOKE -FRONT) CONDITIONS
* KG/S DISCHARGE RATE
* M EFFECTIVE ORIFICE DIAMETER
* M HEIGHT ABOVE (LEVEL) GROUND
* DEGREES RELEASE DISCHARGE ANGLE
* S RELEASE DURATION (<0 FOR STEADY)
* ATMOSPHERIC AMBIENT CONDITIONS
* M REFERENCE HEIGHT
* M/S WIND VELOCITY AT HEIGHT ZO
* CELSIUS AIR TEMPERATURE
* ATMOSPHERES AMBIENT PRESSURE
* PERCENT RELATIVE HUMIDITY
*
* DISPERSION DATA
*
* M SURFACE ROUGHNESS PARAMETER
* PASQUILL STABILITY CLASS
*
* JET/PLUME DEVELOPMENT TERMINATION CRITERIA
*
* M LAST REQD. DOWNWIND DISP.
* M LAST REQD. PLUME DIAMETER
* M LAST REQD. PLUME CENTROID RISE HT.
* M LAST REQD. HORIZONTAL DISPLACEMENT
* M/S LAST REQD. (MEAN) PLUME VELOCITY
* PERCENT LAST REQD. POLLUTANT CONCENTRATION
*
* MATCHING CRITERIA FOR HEGADAS/PGPLUME
*
* LAST REQD. ABS . VALUE OF
* LAST REQD. JET/(JET+HEG) ENTRAINM.
* MAX. BUOYANCY EFFECT FOR ADVECTION
* MAX. BUOY. EFF. FOR PASS.
6-56 -
-------
The PLUME model results indicate that the density of the mixture was not
sufficient to cause the plume to touchdown and result in a link to the HEGADAS
model, but instead a link was provided to the PGPLUME model. It is
interesting to note that if this release was modeled as a pure component, as
opposed to the defined mixture, then the PLUME model would predict that the
plume would impact the ground as a dense plume and a link would have been made
to the HEGADAS model.
Table 6-26 provides a summary of the input used for the PGPLUME model as
provided by the PLUME model. Parameters indicated by the "*" are parameters
that need to be refined by the user. The final input for the PGPLUME model is
provided in Table 6-27. Because the release duration is shorter than the
averaging time of concern, a transient model should be used to accurately
reflect average concentrations. However, this is not possible in the context
of PGPLUME. In order to overcome the issue of using a different averaging
time in the crosswind and along-wind planes, PGPLUME was run twice, once with
the averaging time set equal to the release duration and once with the
averaging time equal to the 30-minutes (IDLH for HCL).
Single-phase Liquid Release (High Volatility') Example
This scenario was modeled using the PLUME model, which provided a
transition into HEGADASS. The transition from PLUME to HEGADASS was made at a
downwind distance of 88.3 meters and thus HEGADASS was started at this
distance using a breakpoint. The modeling was done for a 30-minute average
and compared to the IDLH. Tables 6-28, 6-29 and 6-30 present the model input
for the PLUME model, the PLUME link file for HEGADASS, and the HEGADASS model
input file, respectively. Much of the input in Table 6-28 was taken from
•
Table 5-13. The rest of the input is explained in the HGSYSTEM documentation.
In conducting a modeling analysis of a release that persists for the
extent of time calculated in this case, it is important to keep in mind the
assumptions and limitations of models like those contained in HGSYSTEM. Thus,
if the travel time to the level of concern exceeds several hours, a
6-57
-------
TABLE 6-26. PGPLUME LINK FILE CREATED BY THE PLUME MODEL FOR RELEASE CLASS 6
TITLE SCENARIO 6
GASDATA * released gas composition datablock.
CPGAS- 24.2 * pollutant specific heat (J/mol/C).
MWGAS- 33.1 * pollutant molecular weight (g/mol).
GASFRAC- 0.729 * release mole-fraction pollutant (-).
WATGAS- l.OOOE-03 * release mole-fraction water-vapour (-).
GEOMETRY * plume geometry at matching datablock.
DXPLUME- 18.6 * matching plane displacement (m).
ZPLUME- 13.7 * centroid height above ground (m).
DPLUME- 10.4 * near-plume (effective) diameter (m).
PHIPLUME- 2.89 * plume axis orientation (degrees).
STATE * plume dynamic/thermodynamic state.
UREL- -3.361E-02 * plume relative velocity (m/s).
CMASS- 2.267E-02 * nearfield mass-concentration (kg/m3).
RREL- 2.526E-04 * plume (mean) excess density (kg/m3).
DURATION- 720. * (steady) release duration (s).
AMBIENT * ambient atmosphere datablock.
AIRTEMP- 17.9 * ambient (air) temperature (C).
AIRPRESS- 0.999 * ambient (absolute) pressure (atm).
RHPERC- 36.1 * ambient (relative) humidity (%).
UATM- 1.03 * wind-speed at centroid height (m/s).
RATM- 1.21 * ambient atmosphere density (kg/m3).
DISP * Pasquill/Gifford dispersion data.
ZR- l.OOOE-02 * ground surface roughness (m).
PQSTAB- C * Pasquill/Gifford stability class (-)•
*AVTIMC- 600. * concentration averaging time (s).
TERMINAT * output control datablock.
XFIRST- 18.6 * first required downwind -distance (m).
*STEP- 100. * arithmetic series step-length (m).
*NSTEP= 10 * maximum number of (arithmetic) steps (-).
*FACTOR=« 1.20 * scale factor for geometric series (-).
*XLAST- 1.002E+04 * last required downwind distance (m).
*VFLAST- 165. * last required mole concentration (ppm) .
Note: Parameters which are marked with a * need to by updated to run PGPLUME
6-58
-------
TABLE 6-27. PGPLUME LINK FILE CREATED BY THE PLUME MODEL FOR RELEASE CLASS 6
TITLE SCENARIO 6
GASDATA * released gas composition datablock.
CPGAS- 24.2 * pollutant specific heat (J/mol/C).
MWGAS- 33.1 * pollutant molecular weight (g/mol).
GASFRAC- 0.729 * release mole-fraction pollutant (-).
WATGAS- l.OOOE-03 * release mole-fraction water-vapour (-).
GEOMETRY * plume geometry at matching datablock.
DXPLUME- 18.6 * matching plane displacement (m).
ZPLUME- 13.7 * centroid height above ground (m).
DPLUME- 10.4 * near-plume (effective) diameter (m).
PHIPLUME- 2.89 * plume axis orientation (degrees).
STATE * plume dynamic/thermodynamic state.
UREL- -3.361E-02 * plume relative velocity (m/s).
CMASS- 2.267E-02 * nearfield mass-concentration (kg/m3).
RREL- 2.526E-04 * plume (mean) excess density (kg/m3).
DURATION- 720. * (steady) release duration (s).
AMBIENT * ambient atmosphere datablock.
AIRTEMP- 17.9 * ambient (air) temperature (C).
AIRPRESS- 0.999 * ambient (absolute) pressure (atm).
RHPERC- 36.1 * ambient (relative) humidity (%).
UATM- 1.03 * wind-speed at centroid height (m/s).
RATM— 1.21 * ambient atmosphere density (kg/m3).
DISP * Pasquill/Gifford dispersion data.
ZR- l.OOOE-02 * ground surface roughness (m).
PQSTAB- C * Pasquill/Gifford stability class (-).
AVTIMC- 1800. * concentration averaging time (s).
TERMINAT * output control datablock.
XFIRST- 18.6 * first required downwind distance (m).
STEP- 100. * arithmetic series step-length (m).
NSTEP- 10 * maximum number of (arithmetic) steps (-).
FACTOR- 1.20 * scale factor for geometric series (-).
XLAST- 1.002E+04 * last required downwind distance (m).
VFLAST- 100. * last required mole concentration (ppm).
6-59
-------
TABLE 6-28. PLUME INPUT FOR RELEASE 7 (Reference Table 5-7)
TITLE SCENARIO 7 EHTYLENE OXIDE RELEASE
GASDATA * PHYSICAL PROPERTIES OF GAS
TEMPGAS -
MFGAS
MFH20
MWGAS
CPGAS
PIPE
DMDT
DEXIT
ZEXIT
PHISTK
DURATION =
10.70
100.0
0.0
44.1
24.49
0.1220
6.35E-3
0.50
0.000
-1
AMBIENT CONDITIONS
ZO
UO
AIRTEMP -
AIRPRESS -
RHPERC
DISP
ZR
PQSTAB
TERMINAT
SLST
DLST
ZLST
DXLST
ULST
BETLST
MATCH
RULST
UJET/UAMB-1
RELST
RGLST
RNLST
DISPERSION
10.0
2.0
28.0
1.00
50.0
0.01
E
500.
-1E6
-.35
-500
-0.1
1E-7
.2
.3
.3
.1
*
* CELSIUS TEMPERATURE OF POLLUTANT
* PERCENT MOLE FRACTION OF POLLUTANT
* PERCENT MOLE FRACTION OF WATER
* g/mol MOLECULAR WEIGHT POLLUTANT
* J/mol/C ISOBARIC SPECIFIC HEAT
*
* PIPE EXIT-PLANE (CHOKE-FRONT) CONDITIONS
*
* KG/S . DISCHARGE RATE
* M EFFECTIVE ORIFICE DIAMETER
* M HEIGHT ABOVE (LEVEL) GROUND
* DEGREES RELEASE DISCHARGE ANGLE
* S RELEASE DURATION (<0 FOR STEADY)
* ATMOSPHERIC AMBIENT CONDITIONS
*
* M REFERENCE HEIGHT
* M/S WIND VELOCITY AT HEIGHT ZO
* CELSIUS AIR TEMPERATURE
* ATMOSPHERES AMBIENT PRESSURE
* PERCENT RELATIVE HUMIDITY
*
* DISPERSION DATA
*
* M SURFACE ROUGHNESS PARAMETER
* PASQUILL STABILITY CLASS
*
* JET/PLUME DEVELOPMENT TERMINATION CRITERIA
*
* M LAST REQD. DOWNWIND DISP.
* M LAST REQD. PLUME DIAMETER
* M LAST REQD. PLUME CENTROID RISE HT.
* M - LAST REQD. HORIZONTAL DISPLACEMENT
* M/S LAST REQD. (MEAN) PLUME VELOCITY
* PERCENT LAST REQD. POLLUTANT CONCENTRATION
* MATCHING CRITERIA FOR HEGADAS/PGPLUME
*
* LAST REQD. ABS . VALUE OF
* LAST REQD. JET/(JET+HEG) ENTRAINM.
* MAX. BUOYANCY EFFECT FOR ADVECTION
* MAX. BUOY. EFF. FOR PASS.
6-60
-------
TABLE 6-29. HEGADASS LINK FILE GENERATED BY PLUME FOR RELEASE CLASS 7
TITLE SCENARIO 7 EHTYLENE OXIDE RELEASE
**^*^**4HbHMr*4Hlr*^
* Input file for the (steady -state) heavy gas *
* advection model HEGADAS-S. The file is *
* generated by the near field dispersion model *
* PLUME. It incorporates all the breakpoint *
* data generated by PLUME together with such *
* additional variables and flags needed to *
* ensure physical consistency. In addition, the *
* file contains variables needed to complete a *
* viable input file suitable for submission to *
* .HEGADAS-S; Such additional data are prefixed *
* by an asterisk (*) and should be physically *
* and contextually sensible, but may be changed *
* at the user's discretion. Such data may also be *
* overwritten by the addition of keywords to the *
* HEGADAS-S partial input file under HGSYSTEM. *
CONTROL * HEGADAS Control Flags Datablock.
*ICNT- 0 * flag controlling contour generation ( - ) .
ISURF- 3 * flag indicating plume/ground heat transfer ( - )
AMBIENT * Ambient Atmosphere Datablock.
ZAIRTEMP- 0.500 * height of temperature measurement (m) .
AIRTEMP- 28.0 * ambient (air) temperature (C) .
ZO— 10.0 * height of wind-speed measurement (m) .
UO- 2.00 * ambient wind-speed (m/s).
RHPERC= 50.0 * atmosphere relative humidity (%) .
*TGROUND- 27.6 * ground surface temperature (C) .
DISP * farfield/heavy-gas dispersion Data.
ZR- l.OOOE-02 * surface roughness height (m) .
PQSTAB- E * Pasqill/Gifford stability class.
*AVTIMC- 600. * concentration averaging time (s) .
*CROSSW- 2,, * <<7y> formulation selected (-).
GASDATA * Released gas Datablock.
THERMOD- 1 * thermodynamic model flag ( - ) .
GASFLOW- 0.122 * HF mass flow- rate dry gas (kg/s) .
CPGAS- 24.5 * specific heat of gas (J/mol/C) .
MWGAS- 44.1 * molecular weight of gas (g/mol) .
6-61
-------
TABLE 6-29 (CONTINUED)
GASFRAC- 1.00 * mole fraction pollutant in released gas (-).
HEATGR- 20.0 * ground heat transfer factor (-).
WATGAS- O.OOOE-01 * mole-fraction water in released gas (-).
TEMPGAS- -95.7 * gas temperature immediately upon release (C)
TRANSIT * PLUME/HEGADAS Transition Datablock.
DISTS- 88.3 * downwind distance from release point (m).
CONCS- 1.027E-03 * center-line ground-level molar gas fraction
(-).
WS- -11.3 * heavy-gas plume half-width (signed flag) (m)
*CLOUD * Output control datablock.
*DXFIX- 100. * arithmetic progression step length (m).
*NFIX- 10 * maximum number of (arithmetic) steps (-).
*XGEOM- 1.20 * scale factor for geometric series (-).
*XEND- 1.009E+04 * last required downwind distance (m).
*CU- 1.292E-04 * inner contour concentration (kg/m3).
*CL- 1.292E-05 * outer contour concentration (kg/m3).
*CAMIN- 1.292E-05 * last required gas concentration (kg/m3).
6-62
-------
TABLE 6-30. HEGADASS INPUT FILE FOR RELEASE CLASS 7
TITLE SCENARIO 7 EHTYLENE OXIDE RELEASE
CONTROL * HEGADAS Control Flags Datablock.
ICNT- 1 * flag controlling contour generation (-).
ISURF- 3 * flag indicating plume/ground heat transfer (-)
AMBIENT * Ambient Atmosphere Datablock.
ZAIRTEMP- 0.500 * height of temperature measurement (m).
AIRTEMP- 28.0 * ambient (air) temperature (C).
ZO- 10.0 * height of wind-speed measurement (m).
UO- 2.00 * ambient wind-speed (m/s).
RHPERC- 50.0 * atmosphere relative humidity (%).
TGROUND- 28.6 * ground surface temperature (C).
DISP * farfield/heavy-gas dispersion Data.
ZR- l.OOOE-02 * surface roughness height (m).
PQSTAB- E * Pasqill/Gifford stability class.
AVTIMC- 1800. * concentration averaging time (s).
CROSSW- 2,, * formulation selected (-).
GASDATA * Released gas Datablock.
THERMOD- 1 * thermodynamic model flag (-).
GASFLOW- 0.122 * HF mass flow-rate dry gas (kg/s).
CPGAS- 24.5 * specific heat of gas (J/mol/C).
MWGAS- 44.1 * molecular weight of gas (g/mol).
GASFRAC- 1.00 * mole fraction pollutant in released gas (-).
HEATGR- 20.0 * ground heat transfer factor (-).
WATGAS- O.OOOE-01 * mole-fraction water in released gas (-)•
TEMPGAS- -95.7 * gas temperature immediately upon release (C).
TRANSIT * PLUME/HEGADAS Transition Datablock.
DISTS— 88.3 * downwind distance from release point (m).
CONCS- 1.027E-03 * center-line ground-level molar gas fraction
(-).
WS- -11.3 "* heavy-gas plume half-width (signed flag) (m).
CLOUD * Output control datablock.
DXFIX- 10. * arithmetic progression step length (m).
NFIX- 10 * maximum number of (arithmetic) steps (-).
XGEOM- 1.10 * scale factor for geometric series (-).
XEND- 1.009E+04 * last required downwind distance (m).
CU- .1 * inner contour concentration (kg/m3).
CL- .0014 * outer contour concentration (kg/m3).
GAMIN- 1.292E-05 * last required gas concentration (kg/m3).
6-63
-------
persistence analysis should be conducted to ensure that such long travel times
are plausible.
Single-phase Liquid Release (Low Volatility) Example
This case reflects a spill from a storage tank that contains 30 percent
HCL with the balance being water. This release is assumed to be contained as
a liquid pool within the tank diking.
This case was modeled using the HEGADASS model and the emission rate
calculated in the scenario description. It should be noted that the emission
calculations for this case reflect a very simplified screening approach. For
such a release the emission rate will not be constant over time. Effects such
as cooling of the pool as a result of evaporation will suppress the emission
rate. Because this case is designed to compare impacts against an ERPG (60-
minute average) LOG, the variations in emissions could be quite important.
While the EVAP model contained in HGSYSTEM could not be completely adapted to
this scenario, EVAP could provide a more realistic estimate of emissions from
the pool then the rate provided from the screening calculation.
While this release is not a dense gas release, the HEGADASS model can
accurately simulate both the dense gas and trace gas regions and, therefore,
it is applicable to this release. Because the emissions were not denser than
air, this release was modeled by only examining the HCL portion of the cloud.
Thus an emission rate of 0.00427 kg/s was used. This release also could have
been modeled by invoking the GASFRACT parameter, which would automatically
account for the HCL fraction. This approach was not taken because it would
have made obtaining cloud widths more difficult.
Table 6-31 presents the HEGADASS model input used in modeling this case.
Much of this input was taken from Table 5-15. The rest of the input is
explained in the HGSYSTEM documentation. In modeling this case, the CU
parameter was set to 20 ppm and the CL parameter was set to 3 ppm. Thus from
6-64
-------
TABLE 6-31. HEGADASS INPUT FILE FOR RELEASE CLASS 8 (Reference Table 5-8)
* HEGADAS-S standard input file STPOOLNO.HSI
TITLE
(case: run started at pool, normal thermodynamics)
SCENARIO 8 HCL EVAPORATING POOL
V^WJLl XCVWJU
ICNT
ISURF -
AMBIENT
*Vl f tr A. £*Ll 4.
AIRTEMP -
ZAIRTEMP-
RHPERC -
UO
ZO
TGROUND -
DISP
ZR
PQSTAB -
AVTIMC -
CROSSW -
GASDATA
1
3
18.5
10.7
58.
2.24
10.7
18.5
0.01
C
3600.
2
*
* 01
* C(
*
*
* c
* M
* %
* M/S
* M
* c
*
* •>
*
* M
*
* S
*
*
> DATA BLOCK: CONTROL
output code (isocontours,cloud contents)
code for surface heat/water transfer
DATA
air temperature at height z - ZAIRTEMP
height at which AIRTEMP is given
relative humidity
wind speed at height z - ZO
height at which UO is given
earth's surface temperature
DISPERSION
surface roughness parameter
Pasquill stability class
averaging time for concentration
formula (don't normally change)
GASFLOW -
TEMPGAS -
CPGAS
MUGAS
WATGAS -
HEATGR -
CLOUD
DXFIX
NFIX
NFIX*DXFIX
XEND
GAMIN
CU
CL
POOL
PLL
PLHW
.00427
18.5
32.5
30.2
0.
29.
1.
200
3000.
0.1E-5
2.4718E-5
3.708E-6
7.3
3.66
* KG/S
* C
* J/MOLE/C
* KG/KMOLE
* .
*
*
*
*
* M
*
* M
* KG/M3
* KG/M3
* KG/M3
*
*
*
* M
* M
gas emission rate (excl. water pick-up)
temperature of emitted gas
J/MOLE/C specific heat of emitted gas
KG/KMOLE molecular weight of emitted gas
water pick-up by gas (don't norm.change)
gas group for natural-conv. heat flux
> DATA BLOCK: CLOUD OUTPUT
fixed-size output step length
fixed steps upto distance x -
x at which calculations are stopped
CA (cone.) at which calcs. are stopped
upper concentration limit
lower concentration limit
POOL
pool length
pool half-width
6-65
-------
a model run, model predictions on cloud width could be obtained for these two
ERPG concentration levels. The width of the 100 ppm isopleth could, have been
developed in an analogous manner.
6.5 SLAB
The specific input required by the SLAB model and a brief description is
listed in Table 6-32. It includes information on the type of release;
physical property information on the material released; release rate, height
and duration; fraction of the release that is liquid; concentration averaging
time; maximum downwind distance of interest; height above ground of
concentration calculations; meteorological information; and site roughness
characteristics. In the comments column for many of the input parameters of
Table 6-32 the subsection in Section 4 is given which discusses how that
parameter is calculated.
Table 6-33 shows the value of each input parameter used for the eight
release class examples. In a few cases a value different from that given in
Section 5 was used. Values that are different from those given in Section 5
are highlighted in Table 6-33. Those occurrences are described below. In the
column header for each release class, the table in Section 5 from which the
input was taken is given. The tables in Section 5, in turn, each reference
the section in which the calculation was done.
The release class examples are:
Release Class (RC) Title
1 Two-Phase Gas Release (Choked)
2 Two-Phase Gas Release (Unchoked)
3 Two-Phase Pressurized Liquid
4 Two-Phase Refrigerated Liquid
5 Single-Phase Gas Release (Choked)
6 Single-Phase Gas Release (Unchoked)
7 Single-Phase Liquid Release (High Volatility)
8 Single-Phase Liquid Release (Low Volatility)
6-66
-------
TABLE 6-32. SLAB MODEL INPUT PARAMETERS
Input Parameter Name
Description of Input
Parameter
Comments
idspl
type of release 1
2
3
4
- liquid pool
— horizontal
- vertical
- instantaneous
ncalc
wms
cps
tbp
cmed
dhe
cpsl
rhosl
spb
spc
ts
qs
as
tsd
qtis
hs
tav
xffm
number of calculation
steps
molecular weight of
source gas (kg/g-mole)
heat capacity at
constant pressure (j/kg-
degK)
boiling point
temperature (deg K)
initial liquid mass
fraction
heat of vaporization
UAg)
liquid heat capacity
(jAg-deg K)
liquid density of source
material (kg/m3)
saturation pressure
constant b
saturation pressure
constant c
temperature of source
gas (degK)
mass source rate (kg/s)
source area (m2)
continuous source
duration (s)
instantaneous source
mass (kg)
source height (m)
concentration averaging
time (sec)
far field length (m)
4.2
4.2
4.2
4.9
4.2
4.2
4.2
if unknown, spb — -1
if unknown, spc = 0
4.8
4.7
4.12
4.4
4.4
4.14
4.16
6-67
-------
TABLE 6-32 (CONTINUED)
Input Parameter Name
Description of Input
Parameter
Comments
zp(l),zp(2),zp(3),2p(4)
za
ua
ta
rh
stab
concentration 4.18
measurement heights (m)
surface roughness height 4.17.3
(m)
ambient wind measurement 4.17.1
height (m)
ambient wind speed (m/s) 4.17.1
ambient temperature (deg 4.17.5
K)
relative humidity (%) 4.17.5
stability class 4.17.2
6-68
-------
TABLE 6-33. SLAB MODEL INPUT USED FOR THE EIGHT RELEASE CLASSES (RC). NUMBERS
IN () INDICATE THE TABLE TO REFERENCE FOR SOURCE OF INPUT DATA.
Input
Para-
meter
name
idspl
ncalc
wms
cps
tbp
cmed
dhe
cpsl
rhosl
spb
spc
ts
qs
as
tsd
qtis
hs
tav
xffm
zpd)
zo
za
ua
ta
rh
stab
Input
Values
RC 1
(5-1)
2
1
.044054
1078
283.82
0.0133
568954
1972
882.7
2S07.61
-29.01
283.85
.0634
.0002337
480
0
8
5
1000
0
.01
4.57
5.37
287.52
62
3
Input
Values
RC 2
(5-2)
2
1
.044054
1078
283.85
0.0133
568994
1972
882.7
2507.61
-29.01
283.85
.0503
. 0001882
480
0
tf
5
1000
0
.01
6.10
2.68
296.48
37
3
Input
Values
RC 3
(5-3)
2
1
.070906
481.5
239.15
0.8221
287775
927.3
1562
1978.34
-27.01
239.09
.3170
.003020
672
0
2
900
2000
0
.01
10
4.47
294.38
SO
3
Input
Values
RC 4
(5-4)
3
1
. 064063
623.1
263.13
0.786
388747
1386
1460
2302.35
-35.97
263.13
4.154
.01344
1584
0
.3049
900
10000
0
.01
6
3.13
291.15
42
5
Input
Values
RC 5
(5-5)
2
1
.0350
1456
292.67
0
376440
2560
955
3404.51
15.06
313.6
.8513
0.003167
11
0
3.66
900
1000
0
.01
6.1
6.26
295.15
45
4
Input
Values
RC 6
(5-6)
3
1
.03307
1054
188.15
0
442708
1655.85
1194.2
-1
0
322.04
2.5151
.585754
720
0
8.54
1800
10000
0
.01
10
.98
290.9
36
2
Input
Values
RC 7
(5-7)
2
1
.044054
1078
283.85
1
568994
1972
882.7
2507.61
-29.01
283.85
.1220
.014784
7200
0
.5
1800
10000
0
.01
10
2
301
50
5
Input
Values
RC 8
(5-8)
i
i
.03019
980.98
370.2
1
2354863
3475
993.3
-1
0
291.48
.005365
53.53
7200
0
0
3600
10000
0
.01
10.7
2.24
291.48
58
3
6-69
-------
Two-phase Gas Release (Choked and Unchoked) Examples
The height of release is assumed to be ground level, even though the
release occurred at 3.66 meters above the ground. As shown in Figure 5-1, it
is a downward directed release. Since SLAB is not configured to handle
downward directed releases, a ground level release was assumed.
Single-phase Gas Release (Choked) Example
The text describing this scenario suggested that either an instantaneous
release or a 45-second-long continuous release could be used for this
scenario. Even though an instantaneous release was recommended, the scenario
was modeled as a continuous release with a 45 second duration.
SLAB is capable of simulating a release of any duration. In this case
the scenario was also run as instantaneous, with little difference in the
results. It is suggested that if the release duration is known, the actual
duration of the release should be used rather than simulating the release as
instantaneous. This is not to suggest that the instantaneous release option
should never be used. If the release duration is very much less than the
travel time to a receptor, the instantaneous option can be used.
Single-phase Gas Release (Unchoked) Example
The release is a mixture of HCL, water, and air, but the only chemical
of concern is the HCL. This release was modeled as a mixture with the
predicted output concentrations adjusted to account for only the HCL portion
of the release. This was done by multiplying the predicted concentrations by
the value of fit 0.7295. The apparent molecular weight, which was calculated
in Section 5.6.11, and the mixture gas phase specific heat, calculated in
Section 5.6.2, were used to describe the chemical.
6-70
-------
Single-phase Liquid Release (High Volatility) Example
Because this release is a highly volatile liquid release, there is a
concern that this release may actually cause a pool formation. Section 4.7.7
suggests that the release be assumed to volatilize immediately upon release,
so that is what was done.
Single-phase Liquid Release (Low Volatility) Example
The release in this scenario is a liquid mixture of 30% HCL and 70%
water. It forms a liquid pool, from which HC1 and wter evaporate into the
atmosphere, as described in Section 5.8.7. It was suggested that the release
be modeled as if it were only HCL. However, for the SLAB modeling, this
scenario was modeled as a mixture, with the predicted concentrations adjusted
to account for the HCL-only portion of the release by multiplying the
concentration results by the value of fi( 0.01258.
6-71
-------
6-72
-------
SECTION 7
MODEL OUTPUT
This section gives examples of each model's output and describes how to
use the output to determine the following impacts of concern:
• Maximum off-site concentration;
• Maximum distance downwind of specified concentration;
• Maximum width of specified concentration;
• Maximum time-averaged (e.g. 30 minutes) off-site concentration;
• Maximum time-averaged concentration at a specific point;
• Time at which the maximum concentration is reached at a specific
point;
• Duration above specified concentration at a point;
• Time the specified concentration is reached at a point; and
• Total area impacted by specified concentration.
Not all of the above impacts of concern are available from each model.
Furthermore, for certain models further manipulation of the output may be
required to generate some of the impacts presented here. If so, the method is
described in this section. Examples of available graphical output are
presented.
Rather than discuss all output of each model, in this document only the
procedures required to generate output for the specified impacts of concern
are discussed. For more discussion of the output of a specific model, consult
the user's guide for that model.
7-1
-------
7.1 ADAM
By default, ADAM presents its output as a graphic representation of the
cloud contour for the input concentration or dose of interest. The graphic
shows a concentration (ppm) or dose (ppm-s) isopleth superimposed on an x-y
coordinate system representing the north, south, east and west directions. A
typical output is shown in Figure 7-1. The box to the right of the figure
lists the chemical name, wind speed, source mass (instantaneous release) or
source rate (continuous release), dispersion duration, maximum cloud width,
and maximum centerline downwind distance to the specified concentration or
dose. A hard copy of the output can be obtained by dumping the screen to the
available printer.
Obtaining output on specific parameters is discussed in the following
paragraphs.
Maximum off-site concentration.
Maximum time-averaged off-site concentration.
cimum time-averaged concentration at a specific point.
The above parameters are not directly available from ADAM. However,
these parameters can be estimated as long as the maximum hazard distance
(labeled MAX HAZARD DIST in Figure 7-1) is greater than the distance to the
site boundary or a specific point of interest. The estimates are made by
invoking the re-plot routine, which is available after leaving the graphics
display, and specifying that the cloud contour be re-plotted for a
concentration or dose higher than the original value. If the resulting plot
indicates a corresponding distance greater than the site boundary distance,
the concentration or dose level must be increased and the plot re-displayed.
Likewise, if the resulting distance is less than the site boundary distance,
the concentration or dose should be decreased and the results re-plotted.
Since the re-plotted contours are based on the dispersion calculations
for the original concentration or dose of interest, the exact distance of
7-2
-------
SEAT!
Vapor Dispersion flodel
CONTINUOUS PtUllE OUILIflE
CHEMICAL RELEASED:
EIHOEHE QXIDE
Source BBSS 1.7? k/s
Cone contour 1.8 pp»
Mind speed (IB irt 5.8 n/s
Disper duration 1598.6 s
Max hazard width 1857. n
MAX HAZARD DIST 7938. n
Grid radii = 2888.88 M
1 inch = 3225.98 •
PRESS TO CONTINUE
Figure 7-1. Typical ADAM Model Output
7-3
-------
interest can not be obtained. However, this distance can be estimated by
interpolating between two distances, one on each side of the distance of
interest.
To determine the maximum off-site concentration, the distance to the
site boundary must be used as a specific point. Then treat the maximum
offsite concentration as though it were the maximum concentration at that
point. This approach is only rigorously true if the emission remains at
ground level.
Maximum distance downwind of specified concentration.
This parameter is displayed in the standard ADAM output (see
Figure 7-1). It is labeled "MAX HAZARD DIST."
Maximum width of specified concentration.
This parameter also is displayed the standard ADAM output. It is
labeled "Max hazard width."
Time at which the maximum concentration is reached at a specific point.
Duration above specified concentration at a point.
Time the specified concentration is reached at a point.
The only information available from ADAM for concentration timing is the
location of the leading edge of a specific concentration. That information is
made available by storing the results (called "output DATA" in ADAM) in a
file. An example is given in Table 7-1. For the concentration of the
contour, its distance from source as a function of time is tabulated. There
is no information output describing the timing of the maximum concentration or
duration at specific points. The peak concentration is given as a function of
time, not location.
7-4
-------
TABLE 7-1 EXAMPLE DATA OUTPUT FILE FROM THE ADAM MODEL
DISPERSION CALCULATION RESULTS
99 - # of Points
Contour for Concentration —
5.000000 ppm
Time Since
Release
(s)
•
16.47
16.76
17.25
17.98
19*. 01
20.40
22.24
24.62
27.67
31.54
36.42
Distance
From Source
(m)
27.51
28.00
28.82
30.05
31.77
34.09
37.14
41.10
46.15
52.53
60.56
Velocity of
Plume or Puff
(m/s)
1.6780
1.6774
1.6761
1.6743
1.6720
1.6690
1.6652
1.6607
1.6557
1.6499
1.6438
Peak
Concentration
(ppm)
15817.
15734.
15598.
15400 .
15102.
14414.
12957.
10963.
8901.8
7045 . 6
5486.7
Peak
Dose
(ppm-s)
.0000
.0000
.0000
.0000
.0000 .
.0000
.0000
.0000
.0000
.0000
.0000
Contour
Width
(m)
27.51
29.04
29.39
29.92
30.65
31.63
32.88
34.48
36.47
38.92
41.92
7-5
-------
Total area Impacted by specified concentration
Although not calculated by ADAM, the ground area covered by the cloud
contour can be estimated by assuming that the shape of the contour is
elliptical. The hazard area within the isopleth can then be determined from
the area equation of an ellipse having a major axis equal to the centerline
maximum downwind distance (X,,,,^) and a minor axis equal to the maximum cloud
width (Ymax) . Thus, the ground area covered by the cloud would be:
AREA = (i) x (X^J x (Y.J.
7.2 ALOHA
ALOHA was initially designed to be a first-responders tool rather than a
planning tool. Because of this, the available output are somewhat limited.
The only impact output available from ALOHA is:
• Dispersion footprint;
• Concentration versus time; and
• Dose versus time.
A text summary of input required by ALOHA and examples of output are
provided in Section 6.2. In addition, the output parameter, Source Strength
versus Time, is also available.
The Source Strength does not indicate downwind impacts. Furthermore, it
is recommended that footprints not be used in transient releases, unless it
has been confirmed through point-by-point interrogation (i.e., the use of the
"Concentration" option) that the footprint is not being over-estimated. The
maximum output time is 60 minutes.
Three examples of ALOHA output are presented in Figures 7-2 through 7-4.
The published version of ALOHA does not include tabular output. The
concentration values are assumed to be for five-minute averages; other
7-6
-------
Footprint Window
2804080279
Chemical Name: ETHYLENE OXIDE
Note: Potential or confirmed human carcinogen.
Model Run: Heavy Gas
Wind: 2 meters/sec from N
FOOTPRINT INFORMATION:
Model Run: Heavy Gas
User specified LOG: equals IDLH (800 ppm)
Max Threat Zone for LOG: 50 meters
Note: The Heavy Gas footprint is an initial screening.
For short releases it may be an overestimation.
Be sure to check concentration information at specific locations.
en
^
CD
J_)
s
25
15
5
0
5
15
25
7
10
0
10
20 30
meters
40
50
60
Figure 7-2. Footprint Window
7-7
-------
Concentration Window
2804080354
Chemical Name: ETHYLENE OXIDE
Note: Potential or confirmed human carcinogen.
Model Run: Heavy Gas
Building Air Exchanges Per Hour: 0.01 (User specified)
TIME DEPENDENT INFORMATION:
Concentration/Dose Estimates at the point:
Downwind: 100 meters
Off Centerline: 0 meters
Max Concentration:
Outdoor: 279 ppm
2.69 ppm
Indoor:
Max Dose:
Outdoor:
Indoor:
16,200 (ppm,min)
78.6 (ppn,min)
Note: Indoor graphs are shown with a dotted line.
300-1
•200-
Q,
100-
0
20
40
60
minutes
Figure 7-3. Concentration versus Time Plot
•
7-8
-------
Dose Window
2804080315
Chemical Name: ETHYLENE OXIDE
Note: Potential or confirmed human carcinogen.
Model Run: Heavy Gas
Building Air Exchanges Per Hour: 0.01 (User specified)
TIME DEPENDENT INFORMATION:
Concentration/Dose Estimates at the point:
Downwind: 100 meters
Off Centerline: 0 meters
Max Concentration:
Outdoor: 279 ppm
Indoor: 2.69 ppm
Max Dose:
Outdoor: 16,200 (ppm,min)
Indoor: 78.6 (ppm,min)
Note: Indoor graphs are shown with a dotted line.
20,OOOH
minutes
Figure 7-4. Dose versus Time Plot
7-9-
-------
averaging times are not available. The concentration of interest, however, is
selectable.
Specific parameters are discussed below.
Maximum off-site concentration.
Maximum time-averaged off-site concentration.
Maximum time-averaged concentration at a specific point.
Time at which the maximum concentration is reached at a specific point.
Duration above specified concentration at a point.
Time the specified concentration is reached at a point.
To determine the maximum off-site concentration, the distance to the
site boundary must be used as a specific point. Then treat the maximum off-
site concentration as though it were as the maximum concentration at that
point. This approach is only rigorously true if the emission remains at
ground level.
Figure 7-3 shows a plot reflecting Concentration-versus-Time. The
location of the point, which is selectable, is given as a pair of downwind and
off-centerline distances. The maximum concentration and dosage are given
numerically on the-output. This plot can be used to determine maximum
concentration at a point, as well as duration over some concentration
(graphically). Concentration predictions for both indoor and outdoor
scenarios are available.
The Dose-versus-Time parameter (Figure 7-4) contains the same numerical
data as given in the Concentration-versus-Time output. The dose plot is that
derived from the equation given in Section 4.16. Since the dosage is given in
ppm-min, a new concentration with an averaging time larger than five minutes
can be estimated by dividing the dose by the new time. Dosage predictions for
both indoor and outdoor scenarios are available.
7-10
-------
Maximum distance downwind of specified concentration
This parameter is available from the output displayed in Figure 7-2. It
is labeled "Max Threat Zone for LOG."
Maximum width of specified concentration
This parameter can be approximated from the output displayed in
Figure 7-2. The size of the axis on the left of Figure 7-2 represents the
cross-wind width, which can be used to plot the plume.
Total area impacted by specified concentration
The Footprint output, shown in Figure 7-2, graphically presents the
extent of a specified concentration. For short releases, the footprint may be
over-estimated. The maximum area of impact can only be determined by a trial-
and-error method.
7.3 DEGADIS
• •
The DEGADIS model output is separated into two parts. The first gives
the results from the jet sub-model. This output represents plume
characteristics of the elevated jet portion of a release. An example of the
output is in Table 7-2. The elevation of the jet is calculated as a function
of downwind distance. Once the jet touches ground, DEGADIS switches from the
jet sub-model to the ground level plume dispersion sub-model. An example of
the output from that sub-model of DEGADIS is shown in Table 7-3. The example
outputs given in Tables 7-2.and 7-3 are for the steady-state case. If the
release is transient, the output is very similar, except that a table is
created for varying times after the release. An example and discussion of a
transient table is given below.
7-11
-------
TABLE 7-2. EXAMPLE OUTPUT FROM THE JET SUB-MODEL OF DEGADIS--JETPLU/DEGADIS v2.1
22-DEC-1992 12: 3:48.59
Radian case 3: Chlorine release from a pipe attached to a one-ton cylinder
Two-phase release from pressurized (saturated) liquid
release rate: 0.3139 kg/si Initial density: 19.71 kg/m3
Ambient Meteorological Conditions...
Ambient wlndspeed at reference height:
Reference height:
Surface roughness:
Pasqulll stability class:
Monln-Obukhov length:
Friction velocity:
Ambient temperature:
Ambient pressure:
Ambient humidity:
Relative humidity:
^ Specified averaging time:
, DELTAy:
I-1 BETAy:
10 DELTAz:
BETAz:
GAMMAz:
4.4700 mis
10.000 m
l.OOOOOE-02 m
-30.431
.24814
294.30
1.0000
7.98079E-03
50.000
678.00
.21520
.90000
9.62300E-02
.94770
-2.00000E-03
m/s
K
atm
Contaminant Properties.
Contaminant molecular weight:
Initial temperature:
Upper level of interest:
Lower level of interest:
Average heat capacity:
70.910
238.70 K
l.OOOOOE-05
l.OOOOOE-06
476.44 J/kg K
NDEN flag: 2
Mole fraction
.00000
1.0000
Concentration
(kg/m3)
.00000
19.710
Density
(kg/m3)
1.1936
19.710
ISOFL flag:
Release Properties...
Release rate:
Discharge elevation:
Discharge diameter:
.31390 kg/s
2.0000 m
7.03600E-03 m
-------
TABLE 7-2 (CONTINUED)
Model Parameters...
ALFA1: 2.80000E-02
ALFA2: .37000
DISTMX: 50.000 m
Downwind Elevation Mole Fraction Density
Distance
(m)
3.965E-13
26.7
57.0
84.7
111.
146.
179.
223.
273.
323.
373.
423.
473.
523.
573.
623.
673.
723.
773.
823.
873.
923.
973.
1.023E+03
1.073E+03
1 123E+03
1.173E+03
(m)
2.05
8.63
8.83
8.86
8.86
8.85
8.83
8.81
8.78
8.76
8.73
8.71
8.69
8.67
8.65
8.63
8.62
8.60
8.59
8.57
8.56
8.55
8.54
8.52
8.51
8.50
8.49
Concentration
(kg/m3)
1.00 19.7
4.489E-04 1.319E-03
1.232E-04 3.619E-04
6.224E-05 1.828E-04
4.113E-05 1.208E-04
2.831E-05 8.315E-05
2.161E-05 6.346E-05
1.609E-05 4.726E-05
1.203E-05 3.533E-05
9.305E-06 2.733E-05
7.402E-06 2.174E-05
6.026E-06 1.770E-05
5.001E-06 1.469E-05
4.219E-06 1.239E-05
3.608E-06 1.059E-05
3.122E-06 9.167E-06
2.729E-06 8.014E-06
2.407E-06 7.069E-06
2.140E-06 6.285E-06
1.916E-06 5.626E-06
1.726E-06 5.068E-06
1.563E-06 4.591E-06
1.423E-06 4.179E-06
1.302E-06 3.822E-06
1 195E-06 3.510E-06
1.102E-06 3.235E-06
1.019E-06 2.992E-06
(kg/n>3)
19.7
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
1.19
Temp.
(K)
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
Sigma y Sigma z
(m)
3.880E-03
4.26
8.27
11. B
14.9
19.1
23.0
28.0
33.5
39.0
44.4
49.7
55.0
60.2
65.3
70.4
75.5
80.5
85.5
90.5
95.4
100.
105.
110.
115.
120.
124.
(m)
3.884E-03
2.36
4.45
6.33
8.06
10.4
12.5
15.3
18.4
21.5
24.6
27.6
30.6
33.6
36.5
39.4
42.3
45.2
48.1
51.0
53.8
56.6
59.5
62.3
'65.1
67.8
70.6
Mole Width to molX:
Fraction
®
.000
.000
3.445E-05
4.578E-05
4.125E-OS
3.190E-05
2.461E-05
1.800E-05
1.314E-05
9.971E-06
7.822E-06
6.303E-06
5.190E-06
4.352E-06
3.704E-06
3.194E-06
2.784E-06
2.450E-06
2.173E-06
1.943E-06
1.747E-06
1.581E-06
1.438E-06
1.314E-06
1.205E-06
1.110E-06
1.026E-06
(m)
17.8
25.3
32.1
41.0
49.4
60.1
72.0
83.6
90.0
95.4
99.7
103.
106.
107.
108.
108.
107.
104.
101.
96.0
89.7
81.3
70.2
54.7
28.4
(m)
1
4
13.0 1
20.5 6
25.1 4
29.1 3
30.9 2
30.3 1
24.8 1
9
7
6
5
4
3
3
2
2
2
1
1
1
1
1
1
1
1
Max i.mum
Mole
Fraction
©
1.00
. 489E-04
.232E-04
.235E-05
.235E-05
.190E-05
.461E-05
.800E-05
.314E-05
.971E-06
.822E-06
.303E-06
.190E-06
.352E-06
.704E-06
.194E-06
.784E-06
.450E-06
.173E-06
.943E-06
.747E-06
.581E-06
.438E-06
.314E-06
.205E-06
.110E-06
.026E-06
Elevat Ion
for Max
Mol Frac.
(m)
I
2.05 '
8.63
8.82
8.45
5.94
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
1.198E+03 8.49
9.812E-07 2.881E-06 1.19
294.
127.
72.0
9.879E-07
9.879E-07 .000
-------
TABLE 7-3. EXAMPLE OUTPUT FROM THE GROUND LEVEL DISPERSION SUB-MODEL OF DEGADIS.O
UOA_DEGADIS MODEL OUTPUT
*************** 22-DEC-1992 15:51:18.95
VERSION 2.1
***************
Data input on
Source program run on
0 TITLE BtOCK
22-DEC-1992 12:31:51.90
22-DEC-1992 15:51;18.95
Radian case 3: Chlorine release from a pipe attached to a one-ton cylinder
Two-phase release from pressurized (saturated) liquid
release rate: 0.3139 kg/si Initial density: 19.71 kg/m3
Wind velocity at reference height
Reference height
0 Surface roughness length
0 Pasqulll Stability class
0 Monln-Obukhov length
Gaussian distribution constants
Specified averaging time
Deltay
Betay
0 Wind velocity power law constant Alpha
Friction velocity
0 Ambient Temperature
Ambient Pressure
Ambient Absolute Humidity
Ambient Relative Humidity
4.47
10.00
l.OOOE-02
C
-30.4
678.00
.21520
.90000
.16070
.24814
294.30
1.000
7.981E-03
50.00
mis
Input:
Mole fraction
.00000
1.00000
Specified Gas Properties:
mis
K
atm
kg/kg BDA
X
CONCENTRATION OF C
kg/m**3
.00000
19.71000
Molecular weight:
Release temperature:
Density at release temperature and ambient pressure:
Average heat capacity:
Upper mole fraction contour:
Lower mole fraction contour:
Height for Isopleths:
GAS DENSITY
kg/m**3
1.19357
19.71000
70.910
238.70 K
19.710 kg/m**3
476.44 J/kg K
l.OOOOOE-05
l.OOOOOE-06
.00000 m
-------
TABLE 7-3 (CONTINUED)
Source Input data polite i
Initial (puce contaminant) mass In cloud:
.00000
Time
s
.00000
60230.
60231.
60232.
Contaminant
Mass Rat*.
kg/*
.31390
.31390
.00000
.00000
Source Radius
4.22000E-02
4.22000E-02
.00000
.00000
0 Calculation procedure for ALPHA: 1
0 Entralnment prescription for PHI: 3
0 Layer thickness ratio used for average depth:
0 Air entralnment coefficient usedi .590
0 Gravity slumping velocity coefficient used: 1.150
0 Isothermal calculation '
0 Heat transfer not Included
0 Water transfer not Included
Contaminant
Mass Fraction
kg contain/kg mix
1.0000
1.0000
1.0000
1.0000
2.1500
Temperature
K
294.30
294.30
294.30
294.30
Enthalpy
J/kg
.00000
.00000
.00000
.00000
Ul
-------
TABLE 7-3 (CONTINUE))
CALCULATED SOURCE PARAMETERS
Tim*
ICC
.000000
2.500000E-03
5.000000E-03
7.952711E-03
1.090S42E-02
1.95503SE-02
2.819529E-02
4.490576E-02
6.161624E-02
8.933680E-02
117057
.172885
.228713
.263658
Gas Radius
4
4
4
5
5
7
a.
-------
TABLE 7-3 (CONTINUED)
Tim.
.298603
.399346
.498898
.598449
.759585
.843749
.927914
1.00790
1.16938
1.25088
1.42260
1.59432
1.90508
Gas Radius
n
.419,034
.412110
.403627
.394419
.379492
.372180
.365451
.359705
.350221
.346502
.34068U
.337103
.333952
CALCULATED SOURCE PARAMETERS
C
Height
0
1.285349E-02
1.148123E-02
1.034069E-02
9.363875E-03
8.077288E-03
7.537811E-03
7.082606E-03
6.722174E-03
6.180213E-03
S.984797E-03
5.701425E-03
5.534678E-03
S.393740E-03
Qatar
kg/m**2/s
.585 1,', 5
.622694
.657612
.691420
.743574
.768699
.791831
.811613
.844251
.857008
.876605
.888789
.899504
SZ(x°L/2.) Mole frac
m
4.699319E-02
4.427244E-02
4.189124E-02
3.975436E-02
3.678365E-02
3.S47179E-02
3.433208E-02
3.340676E-02
3.197469E-02
3.144576E-02
3.066765E-02
3.02021SE-02
'2.980437E-02
.764099
.794083
.818850
.840267
.869212
.881694
.892414
.901025
.914154
.918930
.925826
.929921
.933392
Density
kg/ra»*3
7.22069
7.95015
8.64531
9.32936
10.4061
10.9359
11.4284
11.8518
12.5513
12.8235
13.2349
13.4902
13.7133
Rich No.
.756144
.756144
.756144
.756144
.756144
.756144
.756144
.756144
.756144
.756144
.756144
.756144
.756144
-------
TABLE 7-3 (CONTINUED)
CALCULATED SOURCE PARAMETERS
Tim*
sec
2.33720
2.9*799
3.21337
Gas Radius
m
.332610
.332228
.331976
3.39247 .330961
8.79073 .329693
OSourc* strength (kg/i) .
Equivalent Primary scuice length [m] :
Secondary source cone cue ration [kg/m**3] :
Contaminant flux rate; .91811
Height
ID
5.336527E-03
5.323322E-03
S.082750E-03
5.043425E-03
5.002260E-03
Qatar
kg/ra**2/s
.903977
.905023
.911321
.9U620
.918112
SZ(x*L/2. ) Hole frac
m
2.964218E-02
2.960630E-02
2.933861E-02
2.922283E-02
2.910116E-02
.934783
.9)5053
.938231
.939201
.940222
Density
kg/n>»*3
13.8046
13.8225
14.0354
Rich No.
.756144
.756144
.756144
14.1016 .756144
14.1717 .756144
.31390 Equivalent Primary source radius (m) ; 4.22000E-02
8.44000E-02 Equivalent Primary source half-width (m) : 3.31438E-02
13 815
Secondary source SZ [m] i
2.91012E-02
Secondary source mass t (.actions ... contaminant: .974809
Enthalpy: .00000 Density. 14.172
ale: 2.49913E-02
Secondary source length (m) : .65979 Secondary source half-width (ra) : .25910
0 Distance Mole Concentration Density Temperature Half Sz Sy Width at z-.OO m to:
u tyi.aE.«n(;«
i
M (•»)
CO
.330
.380
.430
.563
.696
.845
.994
1.36
1.73
2.83
3.94
6.72
9.50
17.1
24.6
26.7
66.7
147.
207.
267.
327.
nuiv i.uuueiii.1 «i. iuii ueiiaiix *cui|jci.«iui.
Fraction
5
3
1
6
b
4
1
7
4.
3.
©
.940
.931
.926
.903
.870
.829
.776
.628
.494
.247
.142
.571E-02
.024E-02
. 106E-02
.015E-03
. 625E-03
295E-04
482E-04
, 506E-05
550E-05
061E-05
lkg/m**3)
13.8
13.2
12.8
11. A
9.86
6.25
6.71
3.96
2.51
920
475
172
U 116E-02
3 280E-02
1 775E-02
) b60E-02
1 262E-03
L 351E-04
i idO<.E-04
1 336E-04
B 989E-OS
(kg/m**3)
14.2
13.5
13.3
11.9
10.5
8.95
7.50
4.91
3.55
2.06
1.64
1.35
1.28
1.22
1.21
1.21
1.19
1.19
1.19
1.19
1.19
(K)
29<..
291.
294.
294.
294.
294.
294
294.
294.
294.
294
294
294.
294.
294.
2'J4
294.
294
294
294.
294.
a n*ii,i,
Width
(m)
.259
.235
.252
.307
.360
.414
.462
.556
.625
.745
.799
.818
.763
.476
.112
.000
.000
.000
.000
.000
.000
•»* *?/ W J-ULIl ML «t» . UU OI V.U ;
1 . OOOE-04mol«Xl . OOOE-OSmoleX
2.
2.
2.
1.
1.
1.
1.
2.
3.
5.
. 8.
2
4
6
7
9
(m)
910E-02
S29E-02
246E-02
B95E-02
788E-02
790E-02
898E-02
408E-02
080E-02
535E-02
2S2E-02
154
226
418
605
656
.59
.36
.04
.65
.21
(m)
.000
9.419E-02
.139
.232
.312
.394
.471
.648
.811
1.25
1.64
2.49
3.25
5.04
6.62
6.55
17.5
27.7
37.5
47.0
56.3
(m)
.259
.604 .
.795
1.21
1.57
1.93
2.27
2.99
3.62
5.18
6.47
9.08
11.2
15.9
19.6
19.2
43.1
61.9
77.9
91.8
104.
(m)
.259
.575
.753
1.14
1.47
1.81
2.12
2.78
3.36
4.76
5.89
8.16
9.98
13.8
16.9
16.5
34.0
45.5
53.2
57.8
59.5
-------
TABLE 7-3 (CONTINUED)
0 Distance
Mole
Fraction
Concentration Density Temperature Half
Width
( 1.347E+03
v£> 1.407E+03
1.467E+03
. 1.527E+03
1.587E+03
1.647E+03
1.707E+03
1.767E+03
1.827E+03
1.887E+03
1.947E+03
0
2.205E-05
1.667E-05
1.306E-05
1.052E-05
8.665E-06
7.265E-06
6.183E-06
5.329E-06
4.642E-06
4.082E-06
3.619E-06
3.232E-06
2.904E-06
2.624E-06
2.384E-06
2.176E-06
1.994E-06
1.834E-06
1.693Er06
1.568E-06
1.457E-06
1.357E-06
1.267E-06
1.186E-06
1.113E-06
1.046E-06
9.854E-07
6
4
3
3
2,
2
1|
1
1
1.
1.
9.
8,
7.
7,
6.
5,
5.
4.
4.
4,
3,
3
3.
3,
3
2.
. 476E-05
.895E-05
.836E-05
.090E-05
.544E-05
. 133E-05
. 816E-05
565E-05
363E-05
. 199E-05
.063E-05
490E-06
, 528E-06
, 707E-06
. 001E-06
. 389E-06
.855E-06
386E-06
, 972E-06
.605E-06
.278E-06
, 985E-06
, 721E-06
. 483E-06
, 268E-06
.072E-06
894E-06
(kg/m**3) (K)
1
1,
1,
1.
1.
1.
1.
1.
1
1.
1.
1.
1,
1.
1.
1.
1.
1,
1.
1.
1.
1.
1,
1.
1.
1.
1.
.19
.19
.19
.19
.19
.19
19
.19
.19
.19
,19
,19
.19
.19
.19
,19
,19
.19
19
19
.19
,19
.19
.19
,19
,19
,19
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
294.
(m)
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
Sz
(ra)
10.7
12.2
13.7
15.1
16.6
18.0
19.3
20.7
22.1
23. A
24.7
26.1
27.4
28.7
30.0
31.2
32.5
33.8
35.0
36.3
37.5
38.8
40.0
41.2
42.4
43.7
44.9
Sy Width at z-.OO m to:
1.000E-04moleX1.000E-03moleX
(m)
65.4
74.4
83.2
92.0
101.
109.
118.
126.
135.
143.
151.
159.
168.
176.
184.
192.
200.
208.
216.
224.
232.
23,9.
247.
255.
263.
271.
278.
(o)
115.
125.
133.
141.
148.
154.
159.
163.
167.
170.
171.
173.
173.
173.
171.
169.
166.
162.
157.
150.
142.
132.
120.
105.
85,9
57.4
W
58.2
53.2
43.0
20.7
For the UFL of l.OOOOOE-03 mole percent, and the LFL of l.OOOOOE-04 mole percent:
The mass of contaminant between the UFL and LFL Is: 54.399
The mass of contaminant above the LFL Is: 85.514 kg.
kg.
-------
Information on the impact of two concentrations at a single receptor
.height may be requested. Both the elevated jet and ground level sub-models in
DEGAD1S report the widths of the requested concentrations at the receptor
height as a function of downwind distance. The elevated jet sub-model stops
when the plume touches ground or the lowest requested concentration ceases to
exist. If the lowest concentration still exists when the plume reaches the
ground, the ground-level dispersion model begins. The ground-level model
stops when the lowest concentration ceases to exist.
The important tables in the output for determining the impacts are the
Centerline table (labeled 1 in Table 7-2) from the jet sub-model and the table
(labeled 2 in Table 7-3) following the Calculated Source Parameter information
in the ground-level dispersion sub-model. In both tables the width for each
of the two requested concentrations as a function of distance is given. Also
included is the maximum concentration as a function of downwind distance at
the requested altitude. The concentrations are average concentrations for the
averaging time specified. The specified averaging time should be the minimum
of the release duration and the requested averaging time. Some manipulation
of the reported concentrations and distances may be required if the averaging
time is less than the requested averaging time or if the release is simulated
as transient. See below for more discussion on interpreting the output.
Specific parameters are discussed below.
Maximum off-site concentration.
Maximum time-averaged off-site, concentrations.
Maximum time-averaged concentration at a specific point.
•
For the jet sub-model three centerline concentrations are given. One is
the concentration at the vertical center of the plume. The second is given
for the altitude requested. The third is the maximum concentration. The
vertical center mole fraction concentration is given in the column labeled 3
in Table 7-2. The constant altitude concentration is given in the column
7-20
-------
labeled 4 in Table 7-2. The maximum concentration and its altitude are given
in the columns labeled 5 in Table 7-2.
In the ground level sub-model the mole fraction is in the column labeled
6 in Figure 7-3. Since the release is at ground level, this value is also the
maximum concentration.
The maximum concentration at any downwind location can be found by
finding the downwind distance of interest in the first column of either the
jet or ground level sub-model and then reading the maximum concentration from
the appropriate column.
Maximum distance downwind of specified concentration.
Maximum width of specified concentration.
Total area impacted bv specified concentration.
For the jet sub-model, the columns under "Width to mol%" contain the
width information. For the ground level sub-model the table's last two
columns contain the width information. The maximum distance downwind of the
concentration is the last distance with a width for that concentration. The
maximum width of the concentration can be determined by selecting the maximum
width listed. The data provided are sufficient to calculate the plume
dispersion area; however, DEGADIS does not do it. The area of impact at the
specified height is calculated using the width/distance data.
Time at which the maximum concentration is reached at a specific point.
Duration above specified concentration at a point.
Time the specified concentration is reached at a point.
The time that a specific concentration arrives at a specific location is
not explicitly available in the DEGADIS output. However, it can be estimated
by dividing the distance to the location from the release site by the ambient
wind speed. This method, however, becomes less reliable as the location gets
7-21
-------
closer to the source and the higher the exit velocity of the release. The
duration of impact can be estimated as being the same as the release duration.
SPECIAL CONSIDERATIONS
The above discussion of DEGADIS output is based on the assumption that
the release is simulated as a continuous release. When the release duration
is less than the desired averaging time, the averaging time input to DEGADIS
should be the release duration. When the averaging time used is less than the
one desired, the reported average concentrations must be corrected to
represent the averaging time requested. When a release is simulated as
transient, extra effort is necessary to determine concentrations. This
section describes some of the techniques required.
When averaging time used is less than requested
When the averaging time used is less than the requested averaging time,
the output must be converted to represent the requested averaging time. To
extrapolate from the reported average concentration to the requested averaging
time, multiply the reported average concentration by the ratio of the actual
averaging time to the requested averaging time. This means that to find a
concentration averaged over the requested time, a higher concentration may
need to be found in the output. For example, if the release duration is 680 s
and the requested averaging time is 900 s, DEGADIS should be told to average
over 680 s. To convert the concentrations reported in the output to
concentrations for the requested averaging time, each reported concentration
must be multiplied by (680 s/900 s) or 0.756. This means that if a 900 s
average concentration of 1 ppm is of interest, the output will have to be
searched for a concentration of (1 ppm/0.756) or 1.32 ppm.
Transient Simulations
After each simulation, DEGADIS estimates the travel time to the distance
of interest. If the travel time was longer than the release duration, a
7-22
-------
transient simulation was performed. Because DEGADIS does not presently output
time-averaged concentrations directly, additional effort is necessary for
transient release simulations.
For an example, assume that a steady-state ground-level release is being
simulated in which the duration of the release is 45 seconds and an averaging
time of 15 minutes (900 seconds) is requested. A portion of the output from
that simulation is shown in Table 7-4. In this simulation, the travel time to
the location downwind (plant boundary) is greater than the release duration,
so a transient simulation must be made. A portion of the transient simulation
output is shown in Table 7-5. Although not used in this example, it is
presented to indicate the difference between the steady-state output and the
transient output.
The following procedure demonstrates how to determine the maximum
downwind extent to the 6 ppm (15 min) concentration:
• From the steady-state output, the distance to
6 ppm(900 s/45 s)-120 ppm is approximately 700 m. So, use DEG4 to
get the concentration time histories at 700 m and 0.8(700 m)=560 m
(the factor of 0.8 was chosen from experience). DEG4 will create
output similar to that shown in Table 7-6. The mole fraction at
the centerline of the plume at the requested downwind distance is
always presented. The mole fraction at an off-center point at the
same downwind distance is optional. A dosage (D) at the point
downwind can be estimated by summing predicted concentrations (in
the Mole Fraction column) multiplied by he time interval between
the predicted concentrations (in the Time column):
D = £ F4 At
7-23
-------
TABLE 7-4. PARTIAL LISTING OF A STEADY-STATE SIMULATION
0 Distance Mole Concentration Density Temperature
Fraction
(m) (kg/m**3) (kg/m**3) (K)
Ni
1.23
1.28
1.33
2.73
7.06
11.4
20.9
30.5
54.5
114.
174.
234.
294.
354.
414.
474.
534.
594.
654.
714.
774.
834.
894.
954.
.569
.565
.561
.453
.250
.155
.605
.600
.594
.455
.229
.136
7.367E-02 6.265E-02
4.412E-02 3.708E-02
2.090E-02 1.741E-02
4.333E-03 3.585E-03
1.861E-03 1.S38E-03
1.039E-03 8.586E-04
6.658E-04 5.500E-04
4.643E-04 3.B35E-04
3.430E-04 2.833E-04
2.642E-04 2.182E-04
2.100E-04 1.734E-04
1.711E-04 1.413E-04
1.422E-04 1.174E-04
1.202Er04 9.923E-05
1.029E-04 8.501E-05
8.920E-05 7.367E-05
7.809E-05 6.449E-05
6.895E-05 5.695E-05
1.27
1.27
1.26
25
22
21
.20
.20
.19
.19
.19
1.19
19
19
19
1.19
1.19
1.19
19
19
19
19
19
1.19
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
Half
Width
.965
.890
.863
.705
.653
.629
.538
.409
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
Sz
(m)
.263
.263
.263
.281
.371
.476
.712
.943
1.50
3.36
5.03
6.60
8.10
9.55
11.0
12,
13.
15.0
16
17
18.8
20.0
21.3
22.5
Sy
6,
(m)
2.680E-07
9.368E-02
.133
.537
1.17
1.66
2.57
3.37
4.68
8.63
12.4
16.0
19.5
23.0
26.4
29.8
33.1
36.4
39.6
42.8
46.0
49.2
52.3
55.4
Width at z-
, OOOE-04moleX
(m)
.965
1.21
1.32
2.52
4.49
5.94
8.44
10.5
13.4
22.1
29.6
36.3
42.4
48.0
53.1
57.9
62.4
66.5
70.5
74.1
77.6
80.8
83.8
86.6
.00 m to:
1.200E-03moleX
(m)
.965
1.20
1.30
2.46
4.37
5.76
8.15
10.1
12.8
20.9
27.8
33.8
39.1
44.0
48.3
52.3
56.0
59.3
62.3
65.0
67.4
69.6
71.6
73.3
-------
TABLE 7-5. PARTIAL LISTING OF A TRANSIENT SIMULATION
0 Time after beginning of spill
0 Distance
(m)
258.
269.
279.
290.
301.
312.
323.
334.
345.
356.
368.
379.
390.
402.
413.
425.
"^ 436.
ro 448.
cn
460.
471.
483.
495.
507.
519.
531.
543.
555.
567.
579.
592.
Mole
Fraction
Concentration
(kg/m**3)
2.218E-04
3.286E-04
4.311E-04
5.064E-04
5.510E-04
5.565E-04
5.421E-04
5.164E-04
4.876E-04
4.592E-04
4.315E-04
4.070E-04
3.842E-04
3.633E-04
3.442E-04
3.261E-04
3.090E-04
2.929E-04
2.775E-04
2.626E-04
2.475E-04
2.318E-04
2.162E-04
1.998E-04
1.826E-04
1.646E-04
1.459E-04
1 .250E-04
1.056E-04
8.6B4E-05
1.832E-04
2.714E-04
3.561E-04
4.183E-04
4.
4.
4.
4.
4.
3.
3.
3.
3.
3.
2.
2.
2.
2.
2.
2.
2.
1.
1.
1.
1 .
1.
1.
1.
8.
7.
551E-04
597E-04
478E-04
266E-04
028E-04
793E-04
564E-04
362E-04
174E-04
001E-04
843E-04
693E-04
552E-04
419E-04
292E-04
169E-04
044E-04
914E-04
786E-04 '
650E-04
508E-04
359E-04
205E-04
032E-04
723E-05
172E-05
96.00000 sec
Density
(kg/m**3)
1.1904
1.1904
1.1905
1.1905
1.1905
1.1905
1.1905
1.1905
1.1905
1.1905
1.1905
1.1905
1.1905
1.1904
1.1904
1.1904
1.1904.
1.1904
1.1904
1.1904
1.1904
1.1904
1.1904
1.1904
1.1904
1.1904
1.1904
1.1904
1.1904
1.1904
Temperature
(K)
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
Half
Width
(m)
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
Sz
Sy
Width
at z=
6.000E-04moleX
(m)
7.69
7.82
7.77
7.99
8.26
8.52
8.79
9.06
9.32
9.59
9.85
10.1
10.4
10.7
10.9
11.2
11.5
11.7
12.0
12.3
12.5
12.8
13.0
13.3
13.6
13.8
14.1
14.4
14.7
15.2
(m)
17.0
17.8
18.9
19.5
20.2
20.8
21.4
22.1
22.7
23.3
24.0
24.6
25.3
25.9
26.6
27.2
27.9
28.5
29.2
29.8
30.5
31.2
31.8
32.5
33.1
33.8
34.5
35.1
35.7
36.0
(m)
32
35
39
41
42
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
58
59
60
60
61
61
61
61
60
58
.3
.7
.0
.1
.9
.3
.5
.6
.6
.6
.6
.6
.6
.5
.5
.4
.4
.3
.2
.0
.8
.6
.2
.8
.3
.6
.7
.1
.5
.9
.00 m to:
1.200E-03mol<
(m)
29.1
32.4
35.7
37.8
39.4
40.7
41.8
42.8
43.7
44.6
45.4
46.3
47.1
47.9
48.7
49.5
50.3
51.0
51.7
52.4
53.1
53.6
54.1
54.5
54.7
54.7
54.6
53.7
52.7
50.7
For the ULC of 1.20000E-03 mole percent, and the LLC of 6.00000E-04 mole percent:
The mass of contaminant between the ULC and LLC is: .25661 kg.
The mass of contaminant above the LLC Is: 40.109 kg.
-------
where Ft - Mole Fraction; and
At - time interval (1 second in this case).
The dosage can be multiplied by IxlO6 to put the answer into
ppm»s.
• The integrated concentration time histories are 6782. ppm»s
and 4173. ppm«s at 560 m and 700 m, respectively. So, the
15-minute averaged concentrations are
(6782. ppm«s/900 s-)7.54 ppm and
(4173. ppm«s/900 s-)4.64 ppm at 560 m and 700 m,
respectively. Based on log-log interpolation, the distance
to the STEL is approximately 620 m; this could be confirmed
by integrating the concentration time history at 620 m.
Using the same approach, the following procedure can be used to
determine the maximum cloud width to the 6 ppm (15 min) concentration:
• From the steady-state output (Table 7-4), the concentration
at 620 m is approximately 159 ppm, and the maximum width to
(159 ppm/1.1—)140 ppm is about 50 m at a downwind distance
of 350 m (the factor of 1.1 is chosen from experience). So,
use DEG4 to get the concentration time history at x=350 m,
y-25 m, and z-0 m. (A total width of 50 m would be a half-
width of 25 m.) The concentration history is given in
Table 7-6.
• For this point (x-350 m, y-25 m, and z-0 m) , the integrated
concentration time history is 5329 ppm«s which gives a 15-
minute averaged concentration of 5329 ppm«s/900 s=5.9 ppm.
Therefore, a reasonable estimate of the maximum cloud width
is 50 m.
7.4 HGSYSTEM
Multiple models are contained within HGSYSTEM. Therefore, there can be
multiple output. In addition, post-processors are available which read the
model output files and create files that can be imported into other programs
such as Lotus* 1-2-3 or GRAPHER*. These other programs can then be used to
create graphs or figures. However, in this discussion only the standard
tabular output from the HGSYSTEM models will be considered.
7-26
-------
TABLE 7-6. PARTIAL LISTING OF THE OUTPUT FROM DEG4
0 X-Dlrectlon correction was applied.
Coefficient: 4.00000E-02
Power: 1.1400
Minimum Distance: 100.00 in
0 Center line values for the position -->,
x: 350.00 m
0 Time
63
64
65
66
67
68
69
70
71
72
7" 73
NJ 74
75
76
77
78
79.
80.
81.
82.
83.
84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
(s)
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Mole Concentration Density
Fraction
(kg/ra**3)
2
3
3
3
3
4
4
4
4
4.
4.
4
4.
4.
4.
4.
4.
4.
4.
4.
4.
4.
4.
4.
4.
4.
4.
4.
4.
4.
4.
4.
4.
.685E-04
.067E-04
.430E-04
.732E-04
. 968E-04
. 146E-04
. 306E-04
.427E-04
.519E-04
.594E-04
.647E-04
.688E-04
. 714E-04
. 730E-04
742E-04
753E-04
759E-04
764E-04
766E-04
765E-04
767E-04
765E-04
764E-04
765E-04
761E-04
763E-04
762E-04
759E-04
762E-04
760E-04
761E-04
760E-04
755E-04
2
2
2
3
3
3
3
3
3
3
3.
3.
3.
3.
3.
3.
3.
3.
3.
3.
3.
3.
3.
3.
3.
3.
3.
3.
3.
3.
3.
3.
3.
. 218E-04
.533E-04
. 833E-04
. 083E-04
.277E-04
.424E-04
. 557E-04
. 656E-04
. 733E-04
. 795E-04
838E-04
873E-04
894E-04
907E-04
917E-04
926E-04
931E-04
935E-04
937E-04
936E-04
937E-04
936E-04
936E-04
936E-04
933E-04
934E-04
934E-04
931E-04
933E-04
932E-04
932E-04
932E-04
927E-04
**3)
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
.1904
.1904
.1904
.1905
.1905
.1905
.1905
.1905
.1905
.1905
.1905
.1905
.1905
.1905
.1905
.1905
.1905
.1905
.1905
.1905
.1905
.1905
.1905
.1905
.1905
.1905
.1905
.1905
.1905
.1905
.1905
.1905
.1905
Tempera
(K)
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
295.
Half
Width
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
Sz
(m)
9.58
9.45
9.44
9.43
9.43
9.43
9.44
9.44
9.44
9.44
9.44
9.44
9.44
9.44
9.44
9.44
9.44
9.44
9.44
9.44
9.44
44
44
9.44
9.44
9.44
9.44
9.44
9.44
9.44
9.44
9.44
9.44
Sy
(m)
22.8
23.0
23.0
23.0
23.1
23.1
23.0
23.0
23.0
23.0
23.0
23.0
23.0
23.0
23.0
23.0
23.0
23.0
23.0
23.0
23.0
23.0
23.0
23.0
23.0
23.0
23.0
23.0
23.0
23.0
23.0
23.0
23.0
Width at z- .00 m to:
6.000E-04moleX. 1.200E-03moleX
(m) (m)
44.4
45.5
46.2
46.8
47.2
47.4
47.6
47.7
47.8
47.9
48.0
48.0
48.0
48.1
48.1
48.1
48.1
48.1
48.1
48.1
48.1
48.1
48.1
48.1
48
48.
48,
48.
48.
48.
48.
48.
48.1
40.1
41.3
42.1
42.7
43.1
43.4
43.6
43.7
43.8
43.9
44.0
44.0
44.1
44.1
44.1
44.1
44.1
44.1
44.1
44.1
44.1
44.1
44.1
44.1
44.1
44.1
44.1
44.1
44.1
44.1
44.1
44.1
44.1
-------
TABLE 7-6 (CONTINUED)
Time Mole Concentration Density Temperature Half Sz
Fraction Width
(s) (kg/m**3) (kg/m**3) (K) (m) (m)
Sy Width at z= .00 m to:
6.000E-04mole% 1.200E-03moleX
(m) (m) (ra)
96.0 4.753E-04 3.926E-04
97.0 4.738E-04 3.913E-04
98.0 4.713E-04 3.893E-04
99.0 4.658E-04 3.848E-04
100. 4.583E-04 3.786E-04
101. 4.433E-04 3.662E-04
1.1905
1.1905
1.1905
1.1905
1.1905
1.1905
295.
295.
295.
295.
295.
295.
.000
.000
.000
.000
.000
.000
9. 44
9.44
9.44
9.44
9.44
9.44
23.0
23.0
23.0
23.0
23.0
23.0
48.1
48.0
48.0
48.0
47.9
47.7
44.1
44.1
44.0
44.0
43.9
43.7
10
CO
Time
(s)
62.00000
63.00000
64.00000
65.00000
66.00000
67.00000
68.00000
69.00000
70.00000
71.00000
72.00000
73.00000
74.00000
75.00000
76.00000
Mole fraction at:
y> 25.000 n
z- .00000 n
.0000000
8.042S147E-05
9.3628549E-05
1.0491194E-04
1.1464036E-04
1.2258834E-04
1.2793180E-04
1.3265277E-04
1.3612420E-04
1.3881778E-04
1.4098159E-04
1.4258062E-04
1.4385843E-04
1.4461576E-04
1.4511603E-04
Mole fraction at: Mole fraction at:
y- .00000 m y- .00000 m
z- .00000 m z« .00000 m
Mole fraction at:
y= .00000 m
z» .00000 m
77.00000
78.00000
79.00000
1.4547613E-04
1.4580346E-04
1.4599093E-04
80.00000
81.00000
82.00000
83.00000
84.00000
85.00000
1.4611341E-04
1.4617105E-04
1.4615733E-04
1.4617498E-04
1.4613995E-04
1.4609244E-04
86.00000
87.00000
88.00000
89.00000
90.00000
91.00000
1.4610133E-04
1.4599399E-04
1.4601977E-04
1.4599971E-04
1.4589066E-04
1.4595505E-04
-------
TABLE 7-6 (CONTINUED)
Time
(s)
92.00000
93.00000
94.00000
95.00000
96.00000
§7.00000
98.00000
99.00000
100.0000
101.0000
Mole fraction at:
y- 25.000 m
z- .00000 m
1.4591968E-04
1.4590477E-04
1.4589936E-04
1.4570884E-04
1.4563420E-04
1.4518097E-04
1.4440163E-OA
1.4272115E-04
1.4041359E-04
1.3579899E-04
Mole fraction at:
y- .00000 m
X" .00000 m
Mole fraction at:
y— .00000 m
z« .00000 m
Mole
fraction at:
.00000 m
.00000 m
NJ
vo
-------
The HGSYSTEM models are stand-alone computer programs that can be run
separately or linked together to simulate all of the aspects of an atmospheric
accidental release. When models are linked, the first model calculates
parameters required as input by the following model. For example, the
jet/plume model, HFPLUME, will calculate the touch-down of a dense plume but
then, following touch-down, further calculations need to be performed by the
ground-based dispersion model HEGADAS. The HFPLUME model determines when its
calculations should be stopped by checking specific criteria (e.g., jet
velocity). Relevant parameters are then written to an ASCII file using the
same free-forma,t arrangement as files generated by the user. The user is
encouraged to examine the results of the upstream models, at least for a few
cases, before allowing the models to run automatically in sequence.
HEGADASS
An example of the HEGADAS S output is given in Table 7-7. The output
consists of a listing of the input data, some calculated parameters (e.g.
length and width of gas blanket), and a table of plume parameters as a
function of downwind distance. In Table 7-7 only the first page of the output
table is shown. However, the table can be multiple pages. The table length
is dependent on the input parameters describing output step size and extent.
The columns of information most likely to be of interest are:
DISTANCE - Downwind distance in meters;
CONG - Concentration in per cent volume;
YCU - Horizontal extent of the upper concentration, CU, in
meters; and
YCL - Horizontal extent of the lower concentration, CL, in
meters;
7-30
-------
TABLE 7-7. EXAMPLE HEGADASS OUTPUT
HSMAIN
DATE 29/12/92
HEGADAS-S PROGRAM ( VERSION NOV90
STANDARD REPORT FILE
PAGE 0
TIME 11:03
«« SCENARIO 1 CHOKED ETHYLENE OXIDE RELEASE
HECADAS-S INPUT DATA
OUTPUT CODE ICNT =
SURFACE-TRANSFER CODE I SURF -
AIR TEMP. AT HEIGHT ZAIRTEMP AIRTEMP -
REF. HEIGHT FOR AIR TEMP. ZAIRTEMP •>
RELATIVE HUMIDITY RHPERC -
WIND VELOCITY AT HEIGHT ZO UO -
REFERENCE HEIGHT FOR WIND VEL. ZO -
EARTH -S SURFACE TEMPERATURE TGROUND -
SURFACE ROUGHNESS PARAMETER ZR -
PASQUILL STABILITY CLASS PQSTAB -
AVERAG. TIME FOR CONC.MEAS. AVTIMC -
MONIN - OBUKHOV LENGTH OBUKL -
TYPE OF FORMULA FOR SIGMA Y MODSY -
with parameters : DELTA -
BETA -
CONST. IN GRAV. SPREADING LAW CE -
CONST. IN GRAV. SPREADING LAW CD -
EVAPORATION RATE GASFLOW -
EVAPORATION FLUX FLUX -
TEMPERATURE OF EMITTED GAS TEMPGAS -
SPECIFIC HEAT OF EMITTED GAS CPGAS =
MOLECULAR WEIGHT OF EM. GAS MWGAS -
PICKED-DP WATER BY EM. GAS WATGAS -
HEAT GROUP IN HEAT FLUX QH HEATGR -
THERMODYNAMIC MODEL THERMOD -
SOURCE LENGTH PLL -
SOURCE HALF-WIDTH PLHW -
NR, OF SOURCE OUTPUT STEPS NSOURCE -
FIXED OUTPUT STEP DXFIX -
NUMBER OF FIXED STEPS NFIX -
INCREASE FACTOR FOR 'VAR. STEPS XGEOM -
X AT WHICH CALC. IS STOPPED XEND -
CA AT WHICH CALC. IS STOPPED CAMIH =
UPPER CONCENTRATION LIMIT CU -
LOWER CONCENTRATION LIMIT CL -
1
3
30.000
4.6000
62.000
5.3700
4.6000
30.000
l.OOOOOE-02
C
18.750
-30.400
2
5.50000E-02
l.OOOOOE-04
1.1500
5.0000
6.34000E-02
1.5850
10.850
24.490
44.000
O.OOOOOE-01
29.000
1
0.20000
0.10000
4
0.50000
300
1.2000
3000.0
l.OOOOOE-05
1.4400
5.39700E-02
(output of cumulatl'
(only heat transfer
CELSIUS
M
X
HIS
M
CELSIUS
M
SECONDS (value for
M
(Briggs formula)
M**(-l)
KG/S
KG/M2/S
CELSIUS
J/MOLE/CELSIUS
KG/KMOLE
(MOLAR FRACTION)
(normal ,non-reactlvc
M
M
M
M
KG/M3
KG/M3
KG/M3
> CONTROL data block: control parameters
ve cloud data)
, no water vapour)
> AMBIENT data block: ambient data
> DISP data block: dispersion data
> GASDATA data block: gas data
> POOL data block: pool dimensions
> CLOUD data block: control of cloud output
-------
TABLE 7-7 (CONTINUED)
WIND PROFILE EXPONENT ALPHA - 0.18986
FRICTION VELOCITY USTAR - 0.38898 HIS
AIR TEMP. AT GROUND LEVEL TAP - 30.500 CELSIUS
LENGTH GAS BLANKET L - 0.49091 M
WIDTH GAS BLANKET B - 0.24546 M
~-j
to
-------
TABLE 7-7 (CONTINUED)
HSMAIN
DATE 29/12/92
DISTANCE
(M)
0.245
0.123
O.OOOE-01
0.123
0.245
0.500
1.00
1.50
2.00
2.50
3.00
3.50
4.00
4.20
4.50
5.00
5.50
6.00
6.50
7.00
7.50
8.00
8.50
9.00
9.50
10.0
10.5
11.0
11.5
12.0
12.5
13.0
13.5
14.0
14.5
15.0
15.5
16.0
16.5
17.0
17.5
18.0
CONG
(X VOL.
100.
100.
100.
100.
100.
Take-up
65.9
34.3
21.0
14.5
11.0
8.88
7.41
6.34
5.98
5.31
4.40
3.71
3.16
2.73
2.37
2.08
1.84
1.64
1.47
1.32
1.20
1.09
0.991
0.908
0.834
0.769
0.711
0.659
0.613
0.571
0.533
0.499
0.468
0.439
0.413
0.389
0.367
SZ
) (M)
0.00
1.330E-02
2.261E-02
3.058E-02
3.774E-02
««
SY
(M)
0.00
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
HEGADAS-S PROGRAM ( VERSION NOV90 )
STANDARD REPORT FILE
SCENARIO
MIDP
(M)
0.245
0.245
0.245
0.245
0.245
1 CHOKED
YCU
(M)
0.245
0.245
0.245
0.245
0.245
PAGE 1
TIME 11:03
ETHYLENE OXIDE RELEASE »»
ZCU
(M)
0.00
4.446E-03
7.558E-03
1.022E-02
1.262E-02
YCL
(M)
0.245
0.245
0.245
0.245
0.245
ZCL
(M)
0.00
3.862E-02
6.565E-02
8.881E-02
0.110
RIB
0.519
0.881
1.19
1.47
IMP
(DEG.C)
10.8
10.8
10.8
10.8
10.8
CA
(KG/M3)
1.9
1.9
1.9
1.9
1.9
flux = 0.263 KG/M2/S •= 6.346E-02 KG/S
4.652E-02
6.524E-02
8.486E-02
0.113
0.143
0.171
0.198
0.225
0.236
0.252
0.279
0.307
0.334
0.362
0. 389
0.417
0.445
0.472
0.500
0.528
0.556
0.583
0.611
0.639
0.667
0.694
0.722
0.750
0. 778
0.806
0.833
0.861
0.889
0.917
0.945
0.973
1.00
0.110
0.207
0.284
0.351
0.407
0.457
0.502
0.543
0.559
0.582
0.621
0.660
0.699
0.737
0.776
0.815
0.854
0.893
0.932
0.970
1.01
1.05
1.09
1.13
.16
.20
.24
.28
.32
.36
1.40
1.44
1.48
1.51
1.55
1.59
1.63
0.201
0.209
0.220
0.174
0.126
8.435E-02
4.706E-02
1.304E-02
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
0.396
0.532
0.616
0.613
0.588
0.557
0.520
0.478
0.459
0.433
0.376
0.291
0.131
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
0.121
0.138
0.148
0.165
0.177
0.181
0.179
0.173
0.169
0.153
0.120
7.751E-02
2.014E-02
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
1.13
0.782
0.598
0.529
0.482
0.441
0.404
0.369
0.356
0.317
0.258
0.203
0.153
0.106
6.106E-02
1.859E-02
-2.212E-02
-6.131E-02
-9.920E-02
-0.136
-0.172
-0.207
-0.241
-0.275
-0.308
-0.341
-0.374
-0.406
-0.437
-0.469
-0.500
-0.531
-0.562
-0.593
-0.624
-0.655
-0.685
19.5
25.8
27.9
28.8
29.3
29.5
29.7
29.8
29.9
30.0
30.1
30.2
30.2
30.3
30.3
30.3
30.3
30.4
30.4
30.4
30.4
30.4
30.4
30.4
30.4
30.4
30.4
30.4
30.4
30.4
30.5
30.5
30.5
30.5
30.5
30.5
30.5
1.2
0.62
0.37
0.26
0.20
0.16
0.13
0.11
0.11
9.40E-02
7.79E-02
6.56E-02
5.59E-02
4.82E-02
4.20E-02
3.68E-02
3.26E-02
2.90E-02
2.. 59E-02
2.34E-02
2.11E-02
1.92E-02
1.75E-02
1.60E-02
1.47E-02
1.36E-02
1.26E-02
1.17E-02
1.08E-02
1.01E-02
9.42E-03
8.81E-03
8.26E-03
7.76E-03
7.30E-03
6.88E-03
6.49E-03
-------
TABLE 7-7 (CONTINUED)
DISTANCE
(M)
CONC
(X VOL.)
SZ
(M)
SY
(M)
MIDP
(M)
YCU
(M)
ZCU
(M)
YCL
(M)
ZCL
(M)
RIB
IMP CA
(DEG.C) (KG/M3)
18.5 0.347 1.03 1.67 0.OOOE-01 0.OOOE-01 0.OOOE-01 0.OOOE-01 0.OOOE-01 -0.716 30.5 6.UE-03
19.0 0.329 1.06 1.71 0.OOOE-01 0.OOOE-01 0.OOOE-01 0.OOOE-01 0.OOOE-01 -0.746 30.5 5.81E-03
19.5 0.311 1.09 1.75 0.OOOE-01 0.OOOE-01 0.OOOE-01 0.OOOE-01 0.OOOE-01 -0.777 30.5 5.50E-03
20.0 0.296 1.11 1.79 0.OOOE-01 0.OOOE-01 0.OOOE-01 0.OOOE-01 0.OOOE-01 -0.807 30.5 5.22E-03
-^J
£
-------
Specific parameters of interest are discussed below.
Maximum off-site concentration.
Maximum time-averaged off-site concentrations.
Maximum time-averaged concentration at a specific point.
Maximum distance downwind of specified concentration.
Maximum width of specified concentration.
Total area impacted by specified concentration.
Maximum time-averaged concentrations at specific distances downwind and
maximum extent downwind of a specific time-averaged concentration can be
determined from the DISTANCE and CONG columns. Widths and the downwind
distances at which the widths exist can be determined from the DISTANCE and
either YCU or YCL columns. Although not explicitly output, the area impacted
by a concentration can be determined using the width as a function of downwind
distance information. The volume and mass contained within the concentrations
CU and CL can be read from a secondary output of HEGADASS. A portion of that
output is shown in Table 7-8.
Time at which the maximum concentration is reached at a specific point.
Duration above specified concentration at a point.
Time the specified concentration is reached at a point.
Since HEGADASS is a steady-state model, no information on the timing of
an impact can be deduced from the output. However, the amount of time from
the release until a specific downwind location is impacted can be estimated by
assuming that the plume moves at the same speed as the ambient wind speed.
This assumption breaks down, however, if the specific distance is close to the
source or if the release is a jet.
7-35
-------
TABLE 7-8. EXAMPLE OUTPUT OF HEGADASS SHOWING VOLUME OF CONCENTRATIONS
HSMAIN
DATE 29/12/92
DISTANCE
(M)
0. 123
O.OOOE-01
0. 123
0.245
0.500
1.00
1.50
2.00
2.50
3.00
3.50
4.00
4 .20
4.50
5.00
5.50
6.00
6.50
7.00
7. 50
8.00
8.50
9.00
9.50
10.0
10.5
11.0
11.5
12.0
12.5
13.0
13.5
14 .0
14.5
15.0
15.5
16.0
16.5
17.0
17.5
18.0
18.5
19.0
19.5
20.0
VCU
(M)
0.245
0.245
0.245
0.245
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
ZCU
(M)
4.446E-03
7.558E-03
1.022E-02
1.262E-02
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
0, OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
««
YCL
(M)
0.245
0.245
0.245
0.245
0.396
0.532
0.616
0.613
0.588
0.557
0.520
0.478
0.459
0.433
0.376
0.291
0.131
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
3
6
8
0
0
0
0
0
0
0
0
0
0
0
0
7
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
HEGADAS-S PROGRAM ( VERSION NOV90 ) PAGE 1
ISOCONCENTRATION/CLOUD-DATA FILE TIME 11:03
SCENARIO
ZCL
.862E-02
.565E-02
.881E-02
.110
.121
.138
.148
.165
.177
.181
.179
.173
.169
.153
.120
.751E-02
.014E-02
.OOOE-01
.OOOE-01
.OOOE-01
.OOOE-01
.OOOE-01
.OOOE-01
.OOOE-01
.OOOE-01
.OOOE-01
.OOOE-01
.OOOE-01
.OOOE-01
.OOOE-01
.OOOE-01
.OOOE-01
.OOOE-01
.OOOE-01
.OOOE-01
.OOOE-01
.OOOE-01
.OOOE-01
.OOOE-01
.OOOE-01
.OOOE-01
.OOOE-01
.OOOE-01
.OOOE-01
.OOOE-01
1
CHOKED ETHYLENE OXIDE RELEASE »»
CUM. CONTENT (KG)
OCU OCL
4
1
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
.485E-04 1
.211E-03 3
.242E-03 6
.515E-03 1
.515E-03 1
.515E-03 3
.515E-03 4
.515E-03 5
.515E-03 6
.515E-03 6
.515E-03 7
.515E-03 7
.515E-03 B
.515E-03 8
.515E-03 8
.515E-03 8
.515E-03 8
.515E-03 8
.515E-03 8
.515E-03 8
.515E-03 8
.515E-03 8
.515E-03 8
.515E-03 8
.515E-03 8
.515E-03 B
.515E-03 8
.515E-03 B
.515E-03 8
.515E-03 8
.515E-03 8
.515E-03 8
.515E-03 8
.515E-03 8
.515E-03 8
.515E-03 8
.515E-03 8
.515E-03 8
.515E-03 8
.515E-03 8
.515E-03 8
.515E-03 8
.515E-03 8
.515E-03 8
.515E-03 8
.399E-03
. 776E-03
.991E-03
.096E-02
.865E-02
.198E-02
. 366E-02
.353E-02
.181E-02
.867E-02
.424E-02
. 866E-02
.027E-02
.218E-02
.418E-02
.511E-02
.521E-02
.521E-02
.521E-02
.521E-02
.521E-02
.521E-02
.521E-02
.521E-02
.521E-02
.521E-02
.521E-02
.521E-02
.521E-02
.521E-02
.521E-02
.521E-02
.521E-02
.521E-02
.521E-02
.521E-02
.521E-02
.521E-02
.521E-02
.521E-02
.521E-02
.521E-02
.521E-02
.521E-02
.521E-02
CUM. VOLUME (M3)
OCU OCL
2.
7.
1.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
679E-04
233E-04
339E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
099E-03
2.327E-03
6.282E-03
1.163E-02
1.824E-02
3.896E-02
9.884.E-02
0.172
0.252
0.331
0.406
0.474
0.532
0.554
0.582
0.613
0.629
0.631
0.631
0.631
0.631
0.631
0.631
0.631
0.631
0.631
0.631
0.631
0.631
0.631
0.631
0.631
0.631
0.631
0.631
0.631
0.631
0.631
0.631
0.631
0.631
0.631
0.631
0.631
0.631
0.631
-------
HEGADAST
The output from the HEGADAST model is the same as HEGADASS, except that
instead of a steady-state emission assumption, the model assumes a transient
release. Therefore, instead of the output consisting of a single table of
plume parameters as a function of downwind distance (like HEGADASS), HEGADAST
output consists of multiple tables. Each table is for a different time since
the release. Also, notice that each table only displays the downwind
distances which have non-zero concentrations.
Specific parameters of interest are discussed below.
Maximum off-site concentration.
Maximum time-averaged off-site concentrations.
Maximum time-averaged concentration at a specific point.
Maximum distance downwind of specified concentration.
Maximum width of specified concentration.
Total area impacted by specified concentration.
HEGADAST output consists of two types of tables: one shows the plume -
parameters and some concentration parameters, while the other shows the
cumulative volume above specified concentrations. Examples of these two
output are provided in Tables 7-9 and 7-10. To find the parameters above, the
same approach is made from HEGADAST output as with HEGADASS output, except
that all HEGADAST temporal "snapshots" must be used.
Time at which the maximum concentration is reached at a specific point.
Duration above specified concentration at a point.
Time the specified concentration is reached at a point.
A temporal history of a concentration can be constructed from the
collection of snapshots output by HEGADAST. That is, a table showing
7-37
-------
TABLE 7-9. EXAMPLE HEGADAST OUTPUT
HTMAIN
DATE 22/12/92
HEGADAS-T PROGRAM ( VERSION NOV90
STANDARD REPORT FILE
PAGE 0
TIME 14:25
«« SCENARIO 5 USING EPA EMISSION ESTIMATE
HEGADAS-T INPUT DATA
00
OUTPUT CODE ICNT =
SURFACE-TRANSFER CODE ISURF -
GAS-BLANKET FORMULATION BLMODEL •=
CLOUD-SHAPE CORRECTION CODE ICSCOR -
AIR TEMP. AT HEIGHT ZAIRTEMP AIRTEMP •=
REF. HEIGHT FOR AIR TEMP. ZAIRTEMP -
RELATIVE HUMIDITY RHPERC -
WIND VELOCITY AT HEIGHT ZO UO -
REFERENCE HEIGHT FOR WIND VEL. ZO -
EARTH-S SURFACE TEMPERATURE TGROUND -
SURFACE ROUGHNESS PARAMETER 2R -
PASQUILL STABILITY CLASS PQSTAB •
AVERAG. TIME FOR CONC.MEAS. AVTIMC -
MONIN - OBUKHOV LENGTH OBUKL -
TYPE OF FORMULA FOR SIGHA_Y MODSY »
with parameters: DELTA -
BETA -
CONST. IN GRAV. SPREADING LAW CE -
CONST. IN GRAV. SPREADING LAW CD -
TYPE OF FORMULA FOR SIGMA X MODSX «=
vlth parameters: ASIGX ~
BSIGX -
TEMPERATURE OF EMITTED GAS TEMPGAS »
SPECIFIC HEAT OF EMITTED GAS CPGAS =
MOLECULAR WEIGHT OF EM. GAS MWGAS =
PICKED-UP WATER BY EM. GAS WATGAS -
HEAT GROUP IN HEAT FLUX QH HEATGR =
THERMODYNAMIC MODEL THERMOD -
INITIAL LIQUID IN HF HFLIQFR =
TIME-DEPENDENT RECORD DATA: ITYPBR "
- skip Increment for reading INCRT =
- start time for data TSTPOOL =
- time step between read records DT =
- number of records read NTYD -
OUTPUT STEP LENGTH XSTEP =
CA AT WHICH CALC. IS STOPPED CAMIN -
UPPER CONCENTRATION LIMIT CU •
LOWER CONCENTRATION LIMIT CL =
1
3
1
1
22.000
22.000
45.000
6.2600
6.1000
22.000
l.OOOOOE-02
. D
45.000
l.OOOOOE+05
2
4.76543E-02
l.OOOOOE-04
1.1500
5.0000
3
10.000
0.10000
27.000
29.000
20.010
O.OOOOOE-01
24.000
2
1.0000
0
0
O.OOOOOE-01
5.0000
9
50.000
5.00000E-05
5.00000E-05
4.08800E-06
(output of cumulate
(only heat transfer
(new, non-osclllato;
(correction Includec
CELSIUS
M
HIS
M
CELSIUS
M
SECONDS
M
(Brlggs formula)
M**(-l)
(Chatwln/Wllson)
CELSIUS
J/MOLE/CELSIUS
KG/KMOLE
(MOLAR FRACTION)
(hydrogen fluoride)
(MASS FRACTION)
(primary pool data:
S
S
(see list of pool da
M
KG/M3
KG/M3
KG/M3
> CONTROL data block: control parameters
ve cloud data)
, no water vapour)
ry formulation)
d)
> AMBIENT data block: ambient data
> DISP data block: dispersion data
> GASDATA data block: gas data
>TIMEDATA data block: source/breakpoint data
> CLOUD data block: control of cloud output
-------
TABLE 7-9 (CONTINUED)
PHIS. OF TIME FOR CLOUD CALC. TSTAR - 10.000
12.000
15.000
20.000
40.000
60.000
80.000
100.00
.120.00
140.00
160.00
180.00
200.00
220.00
240.00
260.00
280.00
300.00
SEC
SEC
SEC
SEC
SEC
SEC
SEC
SEC
SEC
SEC
SEC
SEC
SEC
SEC
SEC
SEC
SEC
SEC
> CALC data block: control of output times
WIND PROFILE EXPONENT ALPHA = 0.20295
FRICTION VELOCITY USTAR - 0.40006 M/S
AIR TEMP. AT GROUND LEVEL TAP - 22.000 CELSIUS
10
-------
TABLE 7-9 (CONTINUED)
HTMAIN
DATE 22/12/92
HEGADAS-T PROGRAM ( VERSION NOV90 ) PAGE 1
STANDARD REPORT FILE TIME 14:25
«« SCENARIO 5 USING EPA EMISSION ESTIMATE »»
TIME -
TSTPOOL
(S)
0.00
2.50
5.00
7.50
8.91
10.0
12.5
15.0
17.5
20.0
71 22.5
£~ 25.0
O 27.5
30.0
32.5
35.0
37.5
40.0
42.5
45.0
47.5
50.0.
POOL
RADIUS
(M)
0.00
5.000E-02
0.100
0.100
0.100
0.100
0.100
0.100
0.100
0.100
0.100
0.100
0.100
o.ipo
0.100
0.100
0.100
0.100
0.100
0.100
5.000E-02
O.OOOE-01
DATA
POOL
STRENGTH
(KG/S)
0.00
0.437
0.874
0.874
0.874
0.874
0.874
0.874
0.874
0.874
0.874
0.874
0.874
0.874
0.874
0.874
0.874
0.874
0.874
O.B74
0.437
O.OOOE-01
FOR PRIMARY POOL AND SECONDARY VAPOUR BLANKET
BLANKET
RADIUS
(M)
-
0.817
1.50
2.04
2.18
1.70
1.70
1.70
1.70
1.70
1.70
1.70
1.70
1.70
1.70
1.70
1.70
1.70
1.70
1.70
1.10
-
BLANKET
HEIGHT
(M)
_
7.129E-03
5.310E-03
2.078E-03
0.00
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
-
BLANKET
EV.RATE
(KG/S)
_
0.269
0.715
1.17
1.30
0.874
0.874
0.874
0.874
0.874
0'.874
0.874
0.874
0.874
0.874
0.874
0.874
0.874
0.874
0.874
0.437
-
BLANKET
MASS
(KG)
_
0.283
0.711
0.513
0.00 blanket height reduces to zero
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O..OOOE-01
O.OOOE-01
O.OOOE-01
-
Maximum source radius RGMAX - 2.1752 M
Source/blanket data set; total CPU = 7 seconds
OBSERVER
1
9
17
25
33
SOURCE DATA SEEN BY OBSERVERS
START TIME
- TSTPOOL
(S)
-0.4238
1.989
4.402
6.815
9.228
PLACE OF
DOWNWIND
EDGE (M)
0.2774
1.110
1.783
2.145
1.675
LENGTH
(M)
0.4863
1.981
3.235
4.066
3.372
HALF-
WIDTH
(M)
0.1923
0.7840
1,287
1.666
1.343
TAKE-UP
FLUX
(KG/M2/S)
0.1604
0.1187
9.7810E-02
8.8839E-02
9.6330E-02
TAKE-UP
RATE
(KG/S)
2.9997E-02
0.3685
0.8146
1.204
0.8726
-------
TABLE 7-9 (CONTINUED)
OBSERVER START TIME PLACE OF LENGTH HALF-
- TSTPOOL DOWNWIND (M) WIDTH
(S) EDGE (M) (M)
41
49
57
65
73
81
89
97
105
113
121
129
137
145
153
161
11
14
16
18
21
23
26
28
30.
33
35
38.
40.
43.
45.
47
.64
.05
.47
.88
.29
.71
.12
.53
.94
.36
.77
,18 '
.60
.01
.42
.83
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0.
.673
.697
.695
.693
.690
.688
.686
.684
.682
.680
.677
.675
.699
.697
.210
5406
3
3
3
3
3
3
3
3
3
3
3
3
3
3
2
1
.371
.378
.392
.390
.388
.375
.375
.374
.374
.377
.375
.373
.397
.378
.724
.294
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0.
.343
.341
.336
.337
.338
.342
.342
.343
.343
.342
.342
.343
.334
.341
.076
5191
9
9
9
9
9
9
9
9
9
9
9
9
9
9
0
0
. 6330E-02
. 6330E-02
. 6330E-02
.6330E-02
.6330E-02
.6330E-02
.6330E-02
.6330E-02
.6330E-02
.6330E-02
.6330E-02
.6330E-02
.6330E-02
.6330E-02
.1049
.1397
0.8725
0.8728
0.8733
0.8733
0.6732
0.8729
0.8729
0.8728
0.8728
0.8728
0.8727
0.8726
0.8732
0.8728
0.6147
0.1878
Observer-source data sets total CPU - 15 seconds
-------
TABLE 7-9 (CONTINUED)
HTMAIN
DATE 22/12/92
HEGADAS-T PROGRAM ( VERSION NOV90 )
STANDARD REPORT FILE
«« SCENARIO 5 USING EPA EMISSION ESTIMATE »»
PAGE
TIME 14
2
:25
Observer-release frequency - 32i maximum value over all times of [mean error In observer concentration]/[peak conc.J = 0.470
Observer-release frequency • 16| maximum value over all times of [mean error In observer concentration]/[peak cone.] - 0.277
Observer-release frequency - 8| maximum value over all times of [mean error In observer concentration]/[peak cone.] - 6.271E-02
Observer-release frequency - 4j maximum value over all times of (mean error In observer concentration]/[peak cone.) - 3.678E-02
Convergence tolerance OBSEPS » S.OOOE-02 Is satisfied
Cloud shape correction performed
Observer-dispersion data set for 41 observers; total CPU - 983 seconds
DISPERSION DATA AT TIME - 10.00
SEC
DISTANCE
(M)
O.OOOE-01
50.0
CONC
( X VOL . )
97.3
8.149E-05
SZ
(M)
1.501E-02
2.00
SY
(M)
9.116E-02
3.43
MIDP
(M)
1 31
O.OOOE-01
YCU
(M)
1.60
O.OOOE-01
9
0
ZCU
(M)
.947E-02
.OOOE-01
YCL
(M)
1.63
O.OOOE-01
0
0
ZCL
(M)
.120
.OOOE-01
CA
(KG/M3)
0.841
6.764E-07
DISPERSION DATA AT TIME - 12.00
SEC
DISTANCE
(M)
O.OOOE-01
50.0
100.
CONC
( X VOL . )
99.0
1.731E-02
9.192E-22
SZ
(M)
7.465E-03
2.02
3.59
SY
(M)
O.OOOE-01
3.43
6.78
MIDP
(M)
1.34
O.OOOE-01
O.OOOE-01
YCU
(M)
1.34
3.52
O.OOOE-01
ZCU
(M)
4.951E-02
2.11
O.OOOE-01
YCL
(M)
1 34
6.47
O.OOOE-01
ZCL
(M)
5.988E-02
5.80
O.OOOE-01
CA
(KG/M3)
0.846
1.434E-04
7.613E-24
DISTANCE
(M)
CONC
C/. VOL.)
DISPERSION DATA AT TIME = 15.00
SEC
O.OOOE-01 97.3
50.0 0.488
100. 2.309E-10
SZ
(M)
1.427E-02
2.00
3.63
SY
(M)
5.511E-02
3.45
6.76
MIDP
(M)
1.29
O.OOOE-01
O.OOOE-01
YCU
(M)
1.46
•7.24
O.OOOE-01
ECU
(M)
9.455E-02
6.85
O.OOOE-01
YCL
(M)
1.49
9.06
O.OOOE-01
ZCL
(M)
0.114
9.96
O.OOOE-01
CA
(KG/M3)
0.838
4.080E-03
1.913E-12
-------
TABLE 7-9 (CONTINUED)
CO
DISTANCE
(M)
O.OOOE-01
50.0
100.
DISTANCE
(M)
O.OOOE-01
50.0
100.
150.
200.
DISTANCE
(M)
50.0
100.
150.
200.
250.
300.
350.
CONG
U VOL.)
99.2
3.30
6.448E-04
CONC
( X VOL . )
98.6
3.23
0.650
0.179
3.597E-03
CONC
(X VOL.)
2.15
0.639
0.274
0.156
6.708E-02
7.965E-03
1.856E-04
SZ
(M)
O.OOOE-01
1.16
3.71
SZ
(M)
9.841E-03
1.17
2.83
4.36
6.93
SZ
(M)
1.27
2.83
4.18
5.29
6.75
9.93
11.3
DISPERSION
SV
(M)
O.OOOE-01
4.10
6.78
DISPERSION
SV
(M)
6.209E-02
4.13
7.47
10.5
13.4
DISPERSION
SY
(M)
4.07
7.48
10.8
14.1
17.2
20.0
23.3
DATA AT TIME
MIDP
(M)
1.34
O.OOOE-01
O.OOOE-01
DATA AT TIME
MIDP
(M)
1.29
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
DATA AT TIME
MIDP
(M)
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
0 OOOE-01
O.OOOE-01 I
: - 20.00
YCU
(M)
1.34
10.3
O.OOOE-01
- 40.00
YCU
(M)
1.48
10.4
16.2
19.4
O.OOOE-01
- 60.00
YCU
(M)
9.87
16.1
21.1
25.5
26.7
10.6
3. OOOE-01
SEC
ZCU
(M)
O.OOOE-01
5.41
O.OOOE-01
SEC
ZCU
(M)
6.527E-02
5.41
10.2
12.0
O.OOOE-01
SEC
ZCU
(M)
5.51
10.2
12.7
14.1
14.0
3.43
O.OOOE-01
YCL
(M)
1.34
12.2
3.52
YCL
(M)
1.50
12.3
20.0
25.6
18.9
YCL
(M)
11.8
20.0
27.2
33.9
38.1
33.4
O.OOOE-01
ZCL
(M)
O.OOOE-01
7.13
1.25
ZCL
(M)
7.894E-02
7.14
14.5
19.1
12.3
ZCL
(M)
7.41
14.5
19.3
22.7
25.4
23.3
O.OOOErOl
CA
(KG/M3)
0.847
2.856E-02
5.352E-06
CA
(KG/M3)
0.846
2.727E-02
5.375E-03
1.482E-03
2.985E-05
CA
(KG/M3)
1.781E-02
5.290E-03
2.280E-03
1.300E-03
5.571E-04
6.606E-05
1.538E-06
-------
TABLE 7-10. EXAMPLE OUTPUT OF HEGADAST SHOWING VOLUME OF CONCENTRATIONS
ISOCONCENTRATION AND CUMULATIVE CLOUD DATA AT TIME -
DISTANCE
(M)
O.OOOE-01
50.0
YCU
(M)
1.60
O.OOOE-01
ZCU
(M)
9.947E-02
O.OOOE-01
ISOCONCENTRATION AND CUMULATIVE
DISTANCE
(M)
O.OOOE-01
50.0
100.
YCU
(M)
1.34
3.52
O.OOOE-01
ZCU
(M)
YCL
(M)
1.63
O.OOOE-01
ZCL
(M)
0.120
O.OOOE-01
CLOUD DATA AT TIME -
YCL
(M)
4.951E-02 1.34
2.11 6.47
O.OOOE-01 O.OOOE-01
ISOCONCENTRATION AND CUMULATIVE
DISTANCE
(M)
O.OOOE-01
50.0
100.
YCU
(M)
1.46
7.24
O.OOOE-01
ISOCONCENTRATION AND
DISTANCE
(M)
O.OOOE-01
50.0
100.
YCU
(M)
1.34
10.3
O.OOOE-01
ZCU
(M)
ZCL
(M)
5.988E-02
5.80
O.OOOE-01
CLOUD DATA AT TIME -
YCL
(M)
9.455E-02 1.49
6.85 9.06
O.OOOE-01 O.OOOE-01
CUMULATIVE
ZCU
(M)
O.OOOE-01
5.41
O.OOOE-01
CLOUD DATA
YCL
(M)
1.34
12.2
3.52
ZCL
(M)
0.114
9.96
O.OOOE-01
AT TIME -
ZCL
(M)
O.OOOE-01
7.13
1.25
10.00
CUM.CONT.
OCU (KG)
1.65
1.65
12.00
CUM.CONT.
OCU (KG)
0.797
0.840
0.840
15.00
CUM.CONT.
OCU (KG)
1.51
3.79
3.79
20.00
CUM.CONT.
OCU (KG)
O.OOOE-01
11.3
11.3
SEC
CUM.CONT.
OCL (KG)
1.65
1.65
SEC
CUM.CONT.
OCL (KG)
0.797
0.875
0.875
SEC
CUM.CONT.
OCL (KG)
1.51
3.84
3.84
SEC
CUM.CONT.
OCL (KG)
O.OOOE-01
11.4
11.4
CUM. VOL.
OCU (M3)
15.0
15.0
CUM. VOL.
OCU (M3)
6.65
527.
527.
CUM. VOL.
OCU (M3)
13.4
3.485E+03
3.485E+03
CUM. VOL.
OCU (M3)
O.OOOE-01
3.917E+03
3.917E+03
CUM. VOL.
OCL (M3)
18.5
18.5
CUM. VOL.
OCL (M3)
8.04
2.637E+03
2.637E+03
CUM. VOL.
OCL (M3)
16.3
6.338E+03
6.338E+03
CUM. VOL.
OCL (M3)
O.OOOE-01
6.099E+03
6.406E+03
-------
TABLE 7-10 (CONTINUED)
ISOCONCENTRATION AND CUMULATIVE
DISTANCE
(M)
O.OOOE-01
50.0
100.
150.
200.
YCU
(M)
1.48
10.4
16.2
19.4
O.OOOE-01
ZCU
(M)
6.527E-02
5.41
10.2
12.0
O.OOOE-01
ISOCONCENTRATION AND CUMULATIVE
DISTANCE
(M)
50.0
100.
150.
200.
1 250.
300.
i 350.
YCU
(M)
9.87
16.1
21.1
25.5
26.7
10.6
O.OOOE-01
ZCU
(M)
5.51
10.2
12.7
14.1
14.0
3.43
O.OOOE-01 0
CLOUD DATA AT TIME -
YCL
(M)
1.50
12.3
20.0
25.6
18.9
CLOUD DATA
YCL
(M)
11.8
20.0
27.2
33.9
38.1
33.4
ZCL
(M)
7.894E-02
7.14
14.5
19.1
12.3
AT TIME -
ZCL
(M)
7.41
14.5
19.3
22.7
25.4
23.3
.OOOE-01 O.OOOE-01
40.00
CUM.CONT.
OCU (KG)
1.05
12.0
21.3
26.6
26.6
60.00
CUM.CONT.
OCU (KG),
7.60
16.8
25.0
32.5
37.1
37.2
37.2
SEC
CUM.CONT.
OCL (KG)
1.05
12.0
21.5
27.1
27.3
SEC
CUM.CONT.
OCL (KG)
7.64
17.0
25.5
33.5
38.8
39.8
39.8
CUM. VOL.
OCU (M3)
9.28
3.936E+03
1.547E+04
3.1B2E+04
3.182E+04
CUM. VOL.
OCU (M3)
3.813E+03
1.532E+04
3.415E+04
5.935E+04
8.559E+04
8.813E+04
8.813E+04
CUM. VOL.
OCL (M3)
11.3
6.142E+03
2.655E+04
6.070E+04
7.699E+04
CUM. VOL.
OCL (M3)
6.116E+03
2.652E+04
6.334E+04
1.172E+05
1.849E+05
2.393E+05
2.393E+05
-------
Goncentration-versus-Time can be generated for a specific downwind distance.
For example, in Table 7-9 the concentration at 150 meters downwind first
apears at 40 seconds and is 0.179 % volume. The concentration at 150 meters
downwind at 60 seconds is 0.274 % volume. Because there was no reported
concentration at 150 meters downwind at 20 seconds after the release, the
plume must have reached 150 meters downwind between 20 and 40 seconds after
the release.
SPECIAL CONSIDERATIONS
The above discussion of the HEGADST output assumes that the release is
simulated as being continuous. When the release duration is less than the
desired averaging time, the averaging time used should be the release
duration. When the averaging time used is less than the one desired, the
reported average concentrations must be corrected to represent the averaging
time requested.
When averaging time used is less than requested
When the averaging time becomes longer than the release duration,
special steps have to be taken to be sure the correct time-averaged
concentrations are calculated. These special steps were taken in Scenario 5
where the release duration was 45 seconds but the averaging time was 15
minutes. To compute an accurate 15-minute average concentration, the model
predictions had to be integrated at fixed downwind locations to account for
the variations in concentration as a function of time. This was accomplished
using the HTPOST post processing utility. HTPOST is an interactive program
that extracts model results from the binary data file (*.HTM) produced by the
HEGADAST model. The HTPOST utility program can be run in batch mode using the
DOS redirect function (HTPOST
-------
concentrations in terms of sigma x. This concept becomes very important as
the averaging time becomes longer than the release duration and emissions vary
as a function of time.
An example of the output from the PLUME model is shown in Table 7-12.
There may be multiple pages of output in the form presented in Table 7-12.
The number of pages of output depends on the distance downwind that is to be
covered by the output. Note that a definition of each of the column headings
is at the bottom of each table. The columns most likely to be used are x
(horizontal displacement) and beta (pollutant mole-concentration).
Specific parameters of interest are discussed below.
Maximum off-site concentration.
Maximum time-averaged off-site concentrations.
Maximum time-averaged concentration at a specific point.
Maximum distance downwind of specified concentration.
Concentrations listed are given at the plume centerline (z) level. If
some height other than centerline is desired, the concentration must be
extrapolated to the other altitude .
Maximum width of specified concentration.
Total area impacted by specified concentration.
Widths of specified concentrations are not given directly. The area can
not be calculated unless external calculations are done that also use the
parameter D (plume-effective diameter).
7-47
-------
TABLE 7-11. INPUT FILE FOR HTPOST FOR SCENARIO 5
scen5.htm Input File
900 AVERAGING TIME In seconds. USED ONLY ONCE
2 Request a history time-series USED ONLY ONCE
1 Graph, time v. daba
a62.dat Output file name
10 180 Mln and Max TIME
1 ' Plot of X Volume Concentration
62. Distance downwind (m)
1 Graph, time v. data
a76.dat Output file name
10 180 Mln and Max TIME
1 Plot of X Volume Concentration
76 Distance downwind (m)
1 Graph, time v. data
a94.dat Output file name
10 180 Mln and Max TIME
1 Plot of '/. Volume Concentration
94 Distance downwind (m)
1 Graph, time v. data
al25.dat Output file name
10 180 Mln and Max TIME
1 Plot of '/, Volume Concentration
125 Distance downwind (m)
1 Graph, time v. data
al88.dat Output file name
10 180 Mln and Max TIME
~-j i Plot of X Volume Concentration
J^ 188 Distance downwind (m)
Oo 1 Graph, time v. data
a250.dat Output file name
10 180 Mln and Max TIME
1 Plot of X Volume Concentration
250 Distance downwind (m)
1 Graph, time v. data
a313.dat Output file name
10 180 Mln and Max TIME
1 Plot of X Volume Concentration
313 Distance downwind (m)
1 Graph, time v. data
a376.dat Output file name
10 160 Min and Max TIME
1 Plot of X Volume Concentration
376 Distance downwind (m)
1 Graph, time v. data
a438.dat Output file name
10 180 Mln and Max TIME
1 Plot of X Volume Concentration
438 Distance downwind (m)
1 Graph, time v. data
a563.dat Output file name
10 180 Mln and Max TIME
1 Plot of X Volume Concentration
563 Distance downwind (m)
1 Graph, time v. data
a626.dat Output file name
10 180 Min and Max TIME
1 Plot of X Volume Concentration
-------
TABLE 7-11 (CONTINUED)
vO
626 Distance downwind (m)
1 Graph, time v. data
a688.dat Output file name
10 180 Mln and Max TIME
1 Plot of X Volume Concentration
688 Distance downwind (m)
1 Graph, time v. data
a7Sl.dat Output file name
10 180 Mln and Max TIME
1 Plot of X Volume Concentration
751 Distance downwind (m)
1 Graph, time v. data
a876.dat Output file name
10 180 Min and Max TIME
1 Plot of X Volume Concentration
876 Distance downwind (m)
1 Graph, time v. data
a939.dat Output file name
10 180 Mln and Max TIME
1 Plot of X Volume Concentration
939 Distance downwind (m)
1 Graph, time v. data
al001.dat Output file name
10 180 Mln and Max TIME
1 Plot of X Volume Concentration
1001 Distance downwind (m)
0 End
0 End
0 End
-------
TABLE 7-12. EXAMPLE PLUME OUTPUT
Output from PLUME Version NOV90
Orifice conditions:
orifice temperature:-10,OOCi
orifice pressure:67.19atm;
orifice diameter: 0.01m;
orifice height: O.Olmi
orifice mass-flux 5.99kg/si
pollutant mass-flux: 5.99kg/s;
release inclination: 90.00degi
Title:
SCENARIO 4 S02 RELEASE
Atmosphere conditions:
data reference height: 6.00m;
atmosphere temperature: 18.00C:
atmosphere pressure: l.OOatmi
relative humidity: 0.00)11
ambient wind-speed: 3.20m/s;
atmosphere density: 1.22kg/m3i
surface roughness: 0.0100m:
Pasqulll/Gifford class: '£'.
Date: 22/12/92 Time: 08:22
Flash conditions:
flash temperature:******Cj
flash pressure: l.OOatmi
gas-jet veloclty:379.19m/S{
flash density: 5.99kg/m3|
flash diameter: 0.06m;
phi
beta
dm/dt
hmd
2.984E-03
1.681E-02
4.556E-02
9.221E-02
0.160
0.251
0.371
0.523
0.714
•^1 0.950
1 1 .24
t n
o J-59
2.01
2.50
3.07
3.72
4.43
5.19
6.00
6.84
7.71
8.61
9.51
10.4
11.4
12.3
13.3
14.2
15.2
16.1
17.1
18.1
19.1
20.0
21.0
22.0
23.0
24.0
24.9
25.9
1.01
2.01
3.01
4.01
5.01
6.00
6.99
7.98
8.96
9.94
10.9
11.8
12.7
13.6
14.4
15.2
15.9
16.5
17.1
17.7
18.2
18.6
19.0
19.4
19.8
20.1
20.4
20.7
21.0
21.3
21.5
21.8
22.0
22.2
22.4
22.6
22.8
23.0
23.2
23.3
0. 313
0.610
0.931
1.28
1.65
2.06
2.51
3.02
3.60
4.26
5.02
5.89
6.87
7.94
9.07
10.2
11.3
12.3
13.2
14.1
14.9
15.6
16.2
16.9
17.4
18.0
18.5
19.0
19.4
19.8
20.3
20.6
21.0
21.4
21.8
22.1
22.4
22.8
23.1
23.4
138.
75.6
50.6
37.3
29.0
23.4
19.3
16.1
13.6
11.6
9. 89
8.54
7.46
6.62
5.98
5.53
5.21
5.00
4.85
4.76
4.70
4.67
4.65
4.64
4.63
4.64
4.64
4.65
4.66
4.67
4.69
4.70
4.71
4.73
4.74
4.76
4.77
4.79
4.80
4.81
89.6
88.8
87.9
86.8
85.5
84.0
82.2
80.2
77.8
74.9
71.5
67.5
62.9
57.8
52.5
47.3
42.4
38.0
34.1
30.9
28.1
25.7
23.7
21.9
20.4
19.0
17.9
16.8
15.8
15.0
14.2
13.5
12.9
12.3
11.7
11.3
10.8
10.4
9.97
9.60
14.0
18.6
19.0
18.9
18.8
18.7
18.6
18.5
18.5
18.4
18.4
18.4
18.4
18.4
18.4
18.4
18.4
18.4
18.4
18.5
18.5
18.5
18.5
18.5
18.5
18.5
18.5
18.6
18.6
18.6
18.6
18.6
18.6
18.6
18.6
18.6
18.6
18.6
18.6
18.7
20.7
10.2
6.53
4.71
3.62
2.89
2.36
1.95
1.63
1.36
1.15
0.963
0.811
0.684
0.581
0.497
0.431
0.378
0.336
0.302
0.275
0.252
0.233
0.217
0.204
0.192
0.181
0.171
0.163
0.156
0.149
0.143
0.137
0.132
0.128
0.123
0.119
0.116
0.112
0.109
1.54
1.36
1.30
1.28
1.26
1.25
1.24
1.24
1.23
1.23
1.23
1.22
1.22
1.22
1.22
1.22
1.22
1.22
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
0.562
0.271
0.174
0.126
9.654E-02
7.699E-02
6.281E-02
5.198E-02
4.338E-02
3.637E-02
3.056E-02
2.569E-02
2.162E-02
1.824E-02
1.548E-02
1.325E-02
1.148E-02
1.008E-02
8.956E-03
8.055E-03
7.322E-03
6.718E-03
6.213E-03
5.786E-03
5.420E-03
5.101E-03
4.821E-03
4.562E-03
4.341E-03
4.143E-03
3.965E-03
3.803E-03
3.655E-03
3.520E-03
3.395E-03
3.281E-03
3.175E-03
3.077E-03
2.985E-03
2.900E-03
16.4
30.0
44.9
60.9
78.3
97.4
119.
143.
170.
203.
240.
285.
338.
400.
471.
550.
634.
722.
812.
902.
992.
1.081E+03
1.168E+03
1.254E+03
1.338E+03
1.422E+03
1.504E+03
1.589E+03
1.670E+03
1.749E+03
1.828E+03
1.906E+03
1.982E+03
2.058E+03
2.133E+03
2.207E+03
2.281E+03
2.353E+03
2.425E+03
2.497E+03
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
0.365
0.199
0.133
9.825E-02
7.645E-02
6.147E-02
5.046E-02
4.195E-02
3.514E-02
2.956E-02
2.490E-02
2.098E-02
1.769E-02
1.495E-02
1.270E-02
1.089E-02
9.440E-03
8.291E-03
7.374E-03
6.636E-03
6.035E-03
5.539E-03
5.124E-03
4.773E-03
4.473E-03
4.210E-03
3.979E-03
3.767E-03
3.585E-03
3.422E-03
3.275E-03
3.141E-03
3.020E-03
2.908E-03
2.806E-03
2.712E-03
2.624E-03
2.543E-03
2.468E-03
2.398E-03
-------
TABLE 7-12 (CONTINUED)
phi
beta
dm/dt
hmd
26.9
27.9
28.9
29.9
30.9
31.9
32.8
33.8
34.8
35.8
36.8
37.8
38.8
39.8
40.8
41.8
42.8
43.8
44.8
45. B
46.8
47.8
48.8
49.8
50.7
51.7
52.7
53.7
54.7
55.7
56.7
57.7
58.7
59.7
60.7
61.7
62.7
63.7
64.7
65.7
66.7
67.7
68.7
69.7
70.7
71.7
72.7
73.7
74.7
75.7
76.7
77.7
78.7
23.5
23.7
23.8
24.0
24.1
24.2
24.4
24.5
24.6
24.8
24.9
25.0
25.1
25.2
25.3
25.4
25.5
25.6
25.7
25.8
25.9
26.0
26.1
26.1
26.2
26.3
26.4
26.5
26.5
26.6
26.7
26.7
26.8
26.9
26.9
27.0
27.1
27.1
27.2
27.2
27.3
27.4
27.4
27.5
27.5
27.6
27.6
27.7
27.7
27.7
27.8
27.8
27.9
23.7
24.0
24.2
24.5
24.8
25.0
25.3
25.5
25.8
26.0
26.3
26.5
26.7
26.9
27,2
27.4
27.6
27.8
28.0
28.2
28.4
28.6
28.8
29.0
29.2
29.3
29.5
29.7
29.9
30.1
30.2
30.4
30.6
30.7
30.9
31.1
31 .2
31.4
31.5
31.7
31.8
32.0
32.1
32.3
32.4
32.6
32.7
32.9
33 0
33.1
33.3
33.4
33.5
4.83
4.84
4.83
4.87
4.88
4.89
4.90
4.91
4.93
4.94
4.95
4.96
4.97
4.98
4.99
5.00
5.01
5.02
5.03
5.04
5.05
5.06
5.07
5.08
5.08
5.09
5.10
5.11
5.12
5.12
5.13
5.14
5.15
5.15
5.16
5.17
5.18
5.18
5.19
5.20
5.20
5.21
5.22
5.22
5.23
5.23
5.24
5.25
5.25
5.26
5.26
5.27
5.27
9.25
8.93
8.62
8.33
8.06
7.80
7.56
7.33
7.11
6.90
6.70
6.51
6.32
6.15
5.98
5.82
5.66
5.52
5.37
5.24
5.10
4.97
4.85
4.73
4.61
4.50
4.39
4.29
4.19
4.09
3.99
3.90
3.81
3.72
3.63
3.55
3.47
3.39
3.31
3.23
3.16
3.09
3.02
2.95
2.88
2.82
2.75
2.69
2.63
2.57
2.51
2.45
2.39
18.7
18.7
18.7
18.7
18.7
18.7
18.7
18.7
18.7
18.7
18.7
18.7
18.7
18.7
18.7
18.7
18.7
18.7
18.8
18.8
18.8
18.8
18.8
18.8
18.8
18.8
18.8
18.8
18.8
18.8
18.8
18.8
18.8
18.8
18.8
18.8
18.8
18.8
18.8
18.8
18.8
18.8
18.8
18.8
18.8
18.8
18.8
18.8
18.8
18.8
18.8
18.9
18.9
0.106
0.103
0.101
9.805E-02
9.S70E-02
9.348E-02
9.138E-02
8.939E-02
8.749E-02
B.569E-02
8.398E-02
8.234E-02
8.078E-02
7.928E-02
7.785E-02
7.649E-02
7.517E-02
7.391E-02
7.270E-02
7.154E-02
7.042E-02
6.934E-02
6.830E-02
6.729E-02
6.632E-02
6.539E-02
6.449E-02
6.361E-02
6.277E-02
6.195E-02
6.115E-02
6.038E-02
5.964E-02
5.891E-02
5.821E-02
5.753E-02
5.687E-02
5.622E-02
5.559E-02
5.498E-02
5.439E-02
5.381E-02
5.325E-02
5.270E-02
5.217E-02
5.165E-02
5.114E-02
5.064E-02
5.016E-02
4.969E-02
4.923E-02
4.878E-02
4.834E-02
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.21
2.B20E-03
2.745E-03
2.675E-03
2.60BE-03
2.546E-03
2.487E-03
2.431E-03
2.378E-03
2.327E-03
2.279E-03
2.234E-03
2.190E-03
2.148E-03
2.109E-03
2.070E-03
2.034E-03
1.999E-03
1.965E-03
1.933E-03
1.902E-03
1.872E-03
1.844E-03
1.816E-03
1.789E-03
1.763E-03
1.739E-03
1.714E-03
1.691E-03
1.669E-03
1.647E-03
1.626E-03
1.605E-03
1.585E-03
1.566E-03
1.547E-03
1.529E-03
1.512E-03
1.494E-03
1.478E-03
1.462E-03
1.446E-03
1.430E-03
1.415E-03
1.401E-03
1.387E-03
1.373E-03
1.359E-03
1.346E-OS
1.333E-03
1.321E-03
1.308E-03
1.296E-03
1.285E-03
2.567E+03
2.637E+03
2.706E+03
2.775E+03
2.843E+03
2.910E+03
2.977E+03
3.043E+03
3.109E+03
3.174E+03
3.239E+03
3.304E+03
3.367E+03
3.431E+03
3.494E+03
3.556E+03
3.618E+03
3.680E+03
3.741E+03
3.802E+03
3.862E+03
3.922E+03
3.982E+03
4.042E+03
4.100E+03
4.159E+03
4.217E+03
4.275E+03
4.333E+03
4.390E+03
4.447E+03
4.504E+03
4.560E+03
4.616E+03
4.672E+03
4.727E+03
4.782E+03
4.837E+03
4.891E+03
4.945E+03
4.999E+03
5.053E+03
5.106E+03
5.159E+03
5.212E+03
5.265E+03
5.317E+03
5.369E+03
5.421E+03
5.472E+03
5.524E+03
5.575E+03
5.625E+03
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
2.332E-03
2.270E-03
2.212E-03
2.157E-03
2.106E-03
2.057E-03
2.011E-03
1.967E-03
1.925E-03
1.886E-03
1 .848E-03
1.812E-03
1.778E-03
1.745E-03
1.713E-03
1.683E-03
1.654E-03
1.627E-03
1.600E-03
1.574E-03
1.550E-03
1.526E-03
1.503E-03
1.481E-03
1.460E-03
1.439E-03
1.419E-03
1.400E-03
1.382E-03
1.363E-03
1.346E-03
1.329E-03
1.313E-03
1.297E-03
1.281E-03
1.266E-03
1.252E-03
1.238E-03
1.224E-03
1.210E-03
1.197E-03
1.185E-03
1.172E-03
1.160E-03
1.148E-03
1.137E-03
1.126E-03
1.115E-03
1.104E-03
1.094E-03
1.084E-03
1.074E-03
1.064E-03
-------
TABLE 7-12 (CONTINUED)
phi
beta
dm/dt
hmd
79.7
80.7
81.7
82.7
83.7
84.7
85.7
27.9
28.0
28.0
28.0
28.1
28.1
28.1
33.7
33.8
33.9
34.1
34.2
34.3
34.4
5.28
5.28
5.29
5.29
5.30
5.31
5.31
2.34
2.28
2.23
2.17
2.12
2.07
2.02
18.9
18.9
18.9
18.9
18.9
18.9
18.9
4.791E-02
4.749E-02
4.707E-02
4.667E-02
4.628E-02
4.589E-02
4.552E-02
1.21
1.21
1.21
1.21
1.21
1.21
1.21
1.273E-03
1.262E-03
1.251E-03
1.240E-03
1.230E-03
1.220E-03
1.210E-03
5.676E+03
5.726E+03
5.776E+03
5.826E+03
5.875E+03
5.925E+03
5.974E+03
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
1.055E-03
1.045E-03
1.036E-03
1.027E-03
1.019E-03
1.010E-03
1.002E-03
Gas Composition: Interpretation:
mole-fraction water-vapour: O.OOE-OlXi x:
mole-fraction dry-air: O.OOE-OlXi z:
mole-fraction pollutant gas: 1.00E+02Xi D:
exit-plane relative humidity: O.OOE-OlXi u:
pollutant molecular weight: 64. g/moli phi:
pollutant specific heat: 0.33 kJ/kg/Cj T:
horizontal displacement (m)t
plume-axis height (m) t
plume effective diameter (m):
mean plume velocity (m/j):
plume axis inclination (degrees);
mean plume temperature (C)i
Interpretation:
beta: pollutant mole-concentration (X);
r: Jet mean density (kg/o>3);
c: pollutant concentration (kg/m3);
xx: mass-fraction pollutant (-)i
hmd: plume relative humidity (%)j
y H20: plume relative humidity (X)j
Ui
NJ
-------
Time at which the maximum concentration is reached at a specific point.
Duration above specified concentration at a point.
Time the specified concentration is reached at a point.
The release is assumed to be steady-state; thus the timing of the
impacts is not available.
PGPLUME
An example output from PGPLUME is shown in Table 7-13. This example
shows only one page of output from the model; however, there may be multiple
pages of output, depending on the number of output steps requested. Each
output consists of a cross-section of the plume at a distance downwind.
Specific paramters of interest are discussed below.
Maximum off-site concentration.
Maximum time-averaged off-site concentrations.
Maximum time-averaged concentration at a specific point.
Maximum distance downwind of specified concentration.
Each cross-section needs to be examined to determine the maximum
concentration. The maximum concentration can exist anywhere in the plane or
at a specific altitude. For a specific downwind distance, the cross-sections
on either side of that distance can provide bounding values. A concentration
at the desired location can be interpolated using the two concentrations at
two downwind distances and the desired downwind distance.
Maximum width of specified concentration.
Total area impacted by specified concentration.
Widths of the specified concentrations are not given directly. However,
the area can be calculated using external calculations with the cross-
7-53
-------
TABLE 7-13. EXAMPLE PGPLUME OUTPUT
Output from PGPLUME Version Jan92 mo Title: SCENARIO A S02 RELEASE
Date: 01/12/92
Time: 20:57
Cross-Section Data:
downwind displacement: 0.161 kmi
peak mole-concentration gas: 1.20E-02 Xi
plume averaging time: 15. rains:
transverse plume "width": 39. mi
maximum concentration height: 29. mi
Cross-Section Data:
peak excess-velocity: -29. cm/si
peak excess-density: 0.56 g/m3i
peak mass-concentration: 0.32 g/m3|
vertical plume "height": 13. mi
section centrold height: 29. m.
Virtual Source Data:
downwind displacement: -5.24E+02 m;
height above ground: 29. mi
source mass-flux pollutant: 6.0 kg/si
matching achieved: "Perfect" match i
I Table of Mole-Concentration gas (ppm) at several I
(heights (z)(m) above (level) ground, and at several!
I distances (y)(m) measured horizontally off-axis: I
I I
O.OOOE-01
O.OOOE-01 17.7
16.3 73.1
19.5 90.8
22.7 106.
25.8 117.
29.0 120.
32.2 117.
35.3 106.
38.5 90.8
41.7 72.9
!
height (m) above 1
(lever) ground. 1
1
7.82
17.3
71.7
89.0
104.
114.
118.
114.
104.
89.0
71.5
15.6
16.3
67.5
83.9
98.0
106.
111.
108.
98.0
83.8
67.3
23.5 31.3 39.1 46.9 54.7 62.5 70.4
14.7 12.8 10.7 8.59 6.63 4.91 3.49
61.1 53.1 44.4 35. .6 27.5 . 20.3 14.5
75.9 66.0 55.1 44.2 34.1 25.3 18.0
88.7 77.1 64.4 51.7 39.8 29.5 21.0
97.4 84.6 70.7 56.7 43.7 32.4 23.1
100. 87.3 72.9 58.5 45.1 33.4 23.8
97.3 84.6 70.7 56.7 43.7 32.4 23.1
88.6 77.1 64.4 51.7 39.8 29.5 21.0
75.8 65.9 55.1 44.2 34.1 25.2 18.0
60.9 53.0 44.2 35.5 27.4 20.3 14.4
1 horizontal off-axis !
1 displacement (m) . 1
1 1
Near-field Matching Data:
mean plume velocity-excess: -49. cm/si
mean plume density-excess: 1.4 g/n>3j
mean plume concentration gas: 0.79 g/n>3;
effective plume "diameter": 42. mi
plume downwind displacement: 1.61E+02 mi
plume centroid height: 29. mi
plume cross-sectional area: 1.36E+03 m2i
mean plume inclination: -0.30 degrees;
Atmosphere Conditions:
atmosphere density:
atmosphere temperature:
atmosphere pressure:
relative humidity:
ambient wind-speed:
surface roughness:
Pasqulll/Glfford class:
Pasquill/Glfford Matching Data:
1.2 kg/m3j peak excess-velocity: -29. cm/sj
21. Ci peak density-excess: .0.56 g/m3|
1.00 atmi peak concentration gas: 0.32 g/m3;
O.OOE-01 Xi peak mole concentration gas: 1.20E-02 Xi
6.0 m/si peak concentration height: 29. mi
l.OOE-02 mi plume averaging time: 15. mlnsi
E (-)i transverse plume "width": 39. mi
vertical plume "height": 13. m.
-------
sectional data. Each cross-section gives the concentration distribution at a
specific distance downwind. .By combining all the cross-sections, an entire
plume's shape at any altitude could be mathematically created. Furthermore, a
three-dimensional shape of a specific concentration could be deduced, and the
area of that concentration could then be calculated. PGPLUME, however, does
not presently perform such a calculation. Instead, another software package
would need to be created, or an existing software package capable of
mathematical manipulation (e.g. Lotus 1-2-3) would need to be used.
All cross-sections must be used to create a contour of the
concentration. To do this, first the cross-sections are connected
mathematically to form the entire three-dimensional outline of the
concentration. This three-dimensional outline could then be cut,
mathematically, by horizontal planes to determine the area of the
concentration at any altitude.
Time at which the maximum concentration is reached at a specific point.
Duration above specified concentration at a point.
Time the specified concentration is reached at a point.
The release is assumed to be steady-state; thus the timing of the
impacts is not available through PCPLUME.
7.5 SLAB
Except for total area, SLAB provides responses for each of the requested
parameters, with only a single model run required for each of the scenarios.
An example of a SLAB output is presented in Table 7-14. In order to determine
the area of the cloud at a specified concentration, Golden Software's UTIL
program, which is part of their SURFUR system, was used. That software
determines the area inside an isopleth by gridding and interpolating the
predicted concentrations from the model output. Because interpolation is
used, the area of the isopleth represents the cloud as it exists in the raw
model output. However, the inaccuracies caused by the interpolation are not
7-55
-------
TABLE 7-14. EXAMPLE SLAB OUTPUT
SESSION INFORMATION
INPUT DATA FILE NAME :
OUTPUT LIST FILE NAME
epalagl.dta
epalagl.1ST
GRAPHICS DATA FILE NAME
GRAPHICS PLOT FILE NAME
problem input
epalagl.OAT
epalagl.XZ
jdspl =
ncalc =
wins =
cps =
tbp =
cmedO =
dhe =
cpsl =
rhosl =
spb =
spc =
ts
qs
as =
tsd =
qtis =
hs
tav =
xffm =
zpd) =
zp(2) =
zp(3) =
zpd) =
zO
za =
2
1
.044054
1121.00
283.85
.02
579450.
1954.00
882.70
2507.61
-29.01
283.85
.06
.00
480.
.00
.00
5.00
1000.00
.00
3.66
.00
.00
.010000
4.57
-------
TABLE 7-14 (CONTINUED)
ua
ta
rh
stab
5.37
287.52
62.00
3.00
release gas properties
molecular weight of source gas (kg)
vapor heat capacity, const, p. (j/kg-k)
temperature of source gas (k)
density of source gas (kg/m3)
boiling point temperature
^ liquid mass fraction
^ liquid heat capacity (j/kg-k)
"^ heat of vaporization (j/kg)
liquid source density (kg/m3)
saturation pressure constant
saturation pressure constant (k)
saturation pressure constant (k)
spill characteristics
spill type
mass source rate (kg/s)
continuous source duration (s)
continuous source mass (kg)
instantaneous source mass (kg)
source area (m2)
vertical vapor velocity (m/s)
source half width (m)
source height (m)
horizontal vapor velocity (m/s)
wms =
cps =
ts =
rhos =
tbp =
cmedO=
cpsl =
dhe =
rhosl=
spa =
spb =
spc =
idspl =
qs =
tsd =
qtcs =
qtis =
as =
ws
bs =
hs =
us =
4.4054E-02
1.1210E+03
2.8385E+02
1.8914E+00
2.8385E+02
1.7200E-02
1.9540E+03
5.7945E+05
8.8270E+02
9.8399E+00
2.5076E+03
-2.9010E+01
2
6.3400E-02
4 . 8000E+02
3.0432E+01
O.OOOOE+00
2.3260E-04
O.OOOOE+00
7.6256E-03
O.OOOOE+00
1.4164E+02
-------
TABLE 7-14 (CONTINUED)
field parameters
concentration averaging time (s)
mixing layer height (m)
maximum downwind distrace (m)
concentration measurement height (m)
m
CO
- tav =
- hmx =
- xffm =
- zp(l)=
- zp(2)=
- zp(3)=
- zp(4)=
5.0000E*00
2.0800E+03
l.OOOOE+03
O.OOOOE+00
3.6600E+00
O.OOOOE+00
O.OOOOE+00
ambient meteorological properties
molecular weight of ambient air (kg) - wmae = 2.8847E-02
heat capacity of ambient air at const p. (j/kg-k)- cpaa = 1.0113E+03
density of ambient air (kg/m3) - rhoa = 1.2227E+00
ambient measurement height (m) - za = 4.5700E+00
ambient atmospheric pressure (pa=n/m2=j/m3) - pa = 1.0133E+05
ambient wind speed (m/s) - ua = 5.3700E+00
ambient temperature (k) - ta 2.8752E+02
relative humidity (percent) - rh = 6.2000E+01
ambient friction velocity (m/s) - uastr = 3.9267E-01
atmospheric stability class value - stab = 3.0000E+00
Inverse montn-obukhov length (1/m) - ala = -3.2907E-02
surface roughness height (m) - zO = l.OOOOE-02
additional parameters
sub-step multiplier
number of calculational sub-steps
acceleration of gravity (m/s2)
gas constant (j/raol- k)
von karman constant
1
ncalc =
nssm =
grav =
rr =
xk =
1
3
9.8067E+00
8.3143E+00
4.1000E-01
-------
TABLE 7-14 (CONTINUED)
instantaneous spatially averaged cloud parameters
X
l.OOE+00
.02E+00
.04E+00
.06E+00
.08E+00
.12E+00
.15E+00
1.19E+00
1.24E+00
1.30E+00
1.36E+00
1.44E+00
1.53E+00
•^ 1.64E+00
ui 1.76E+00
VO
1.90E+00
2.07E+00
2.26E+00
2.48E+00
2.75E+00
3.05E+00
3.41E+00
3.83E+00
4.31E+00
4.88E+00
5.54E+00
6.31E+00
7.21E+00
8.25E+00
9.47E+00
1.09E+01
1.26E+01
1.45E+01
1.67E+01
zc
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
h
1.08E-02
1.09E-01
1.27E-01
1.46E-01
1.66E-01
1.88E-01
2.11E-01
2.36E-01
2.63E-01
2.93E-01
3.24E-01
3.58E-01
3.94E-01
4.33E-01
4.75E-01
5.20E-01
5.69E-01
6.21E-01
6.77E-01
7.38E-01
8.05E-01
8.79E-01
9.61E-01
1.05E+00
1.15E+00
.27E+00
.40E+00
.55E+00
.72E+00
.91E+00
2.14E+00
2.39E+00
2.68E+00
3.02E+00
bb
1.08E-02
5.44E-02
6.34E-02
7.28E-02
8.29E-02
9.38E-02
1.05E-01
1.18E-01
1.32E-01
1.46E-01
1.62E-01
1.79E-01
1.97E-01
2.17E-01
2.38E-01
2.60E-01
2.84E-01
3.10E-01
3.38E-01
3.69E-01
4.02E-01
.4.39E-01
4.80E-01
5.26E-01
5.77E-01
6.34E-01
7.00E-01
7.74E-01
8.60E-01
9.57E-01
1.07E+00
1.20E+00
1.35E+00
1.52E+00
b
9.71E-03
3.04E-02
3.29E-02
3.52E-02
3.74E-02
3.96E-02
4.18E-02
4.39E-02
4.61E-02
4.82E-02
5.02E-02
5.23E-02
5.43E-02
5.62E-02
5.81E-02
5.99E-02
6.16E-02
6.31E-02
6.45E-02
6.59E-02
6.73E-02
6.86E-02
6.98E-02
7.11E-02
7.20E-02
7.30E-02
7.39E-02
7.48E-02
7.58E-02
7.66E-02
7.74E-02
7.82E-02
7.88E-02
7.92E-02
bbx
O.OOE+00
1.66E-02
3.59E-02
5.85E-02
8.48E-02
1.15E-01
1.51E-01
1.93E-01
2.41E-01
2.98E-01
3.64E-01
4.41E-01
5.31E-01
6.36E-01
7.58E-01
9.00E-01
1.07E+00
1.26E+00
1.48E+00
1.75E+00
2.05E+00
2.41E+00
2.83E+00
3.31E+00
3.88E+00
4.54E+00
5.31E+00
6.21E+00
7.25E+00
8.47E+00
9.90E+00
1.16E+01
1.35E+01
1.57E+01
bx
O.OOE+00
1.66E-02
3.59E-02
5.85E-02
8.48E-02
1.15E-01
1.51E-01
1.93E-01
2.41E-01
2.98E-01
3.64E-01
4.41E-01
5.31E-01
6.35E-01
7.57E-01
9.00E-01
1.07E+00
1.26E+00
1.48E+00
1.75E+00
2.05E+00
2.41E+00
2.83E+00
3.31E+00
3.88E+00
4.54E+00
5.31E+00
6.21E+00
7.25E+00
8.47E+00
9.90E+00
1.16E+01
1.35E+01
1.57E+01
cv
l.OOE+00
1.35E-01
1.16E-01
l.OOE-01
8.74E-02
7.67E-02
6.76E-02
5.97E-02
5.29E-02
4.69E-02
4.16E-02
3.69E-02
3.27E-02
2.90E-02
2.57E-02
2.27E-02
2.01E-02
1.77E-02
1.55E-02
1.35E-02
1.17E-02
1.01E-02
8.67E-03
7.37E-03
6.20E-03
5.17E-03
4.27E-03
3.49E-03
2.82E-03
2.26E-03
1.79E-03
1.41E-03
1.10E-03
8.54E-04
rho
1.92E+00
1.32E+00
1.31E+00
1.29E+00
1.29E+00
1.28E+00
1.27E+00
1.27E+00
1.26E+00
1.26E+00
1.25E+00
1.25E+00
1.25E+00
1.24E+00
1.24E+00
1.24E+00
1.24E+00
1.24E+00
1.23E+00
1.23E+00
.23E+00
.23E+00
.23E+00
.23E+00
.23E+00
.23E+00
.23E+00
1.23E+00
1.22E+00
1.22E+00
1.22E+00
1.22E+00
1.22E+00
1.22E+00
t
2.84E+02
2.85E+02
2.86E+02
2.86E+02
2.86E+02
2.86E+02
2.86E+02
2.87E+02
2.87E+02
2.87E+02
2.87E+02
2.87E+02
2.87E+02
2.87E+02
2.87E+02
2.87E+02
2.87E+02
2.87E+02
2.87E+02
2.87E+02
2.87E+02
2.87E+02
2.87E+02
2.87E+02
2.87E+02
2.87E+02
2.87E+02
2.87E+02
2.87E+02
2.87E+02
2.87E+02
2.87E+02
2.88E+02
2.88E+02
u
1.42E+02
2.10E+01
1.81E+01
1.59E+01
1.40E+01
1.25E+01
1.12E+01
1.02E+01
9.24E+00
6.45E+00
7.76E+00
7.17E+00
6.66E+00
6.22E+00
5.84E+00
5.51E+00
5.23E+00
4.99E+00
4.78E+00
4.61E+00
4.46E+00
4.34E+00
4.24E+00
4.16E+00
4.11E+00
4.07E+00
4.06E+00
4.06E+00
4.07E+00
4.10E+00
4.14E+00
4.19E+00
4.25E+00
4.33E+00
ua
1.89E-01
1.44E+00
1 . 56E+00
1.68E+00
1.79E+00
1.89E+00
1.99E+00
2.09E+00
2.18E+00
2.28E+00
2.37E+00
2.45E+00
2.54E+00
2.62E+00
2.70E+00
2.78E+00
2.86E+00
2.93E+00
3.01E+00
3.08E+00
3.16E+00
3.23E+00
3.31E+00
3.3BE+00
3.46E+00
3.54E+00
3.63E+00
3.71E+00
3.80E+00
3.89E+00
3.98E+00
4.07E+00
4.16E+00
4.26E+00
-------
TABLE 7-14 (CONTINUED)
CT>
O
X
1.94E+01
2.24E+01
2.60E+01
3.02E+01
3.50E+01
4.07E+01
4.73E+01
5.50E+01
6.40E+01
7.44E+01
8.66E+01
1.01E+02
1.17E+02
1.37E+02
1.59E+02
1.86E+02
2.16E+02
2.52E+02
2.94E+02
3.42E+02
3.99E+02
4.65E+02
5.42E+02
6.31E+02
7.36E+02
8.58E+02
l.OOE+03
/(.
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
0:OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
h
3.41E+00
3.86E+00
4.37E-t-00
4.96E+00
5.63E+00
6.41E+00
7.29E+00
8.30E+00
9.42E+00
1.07E+01
1.21E+01
1.38E+01
1.57E+01
1.78E+01
2.03E+01
2.31E+01
2.63E+01
2.99E+01
3.41E+01
3.88E+01
4.41E+01
5.02E+01
5.72E+01
6.51E+01
7.40E+01
8.42E+01
9.57E+01
bb
1.72E+00
1.95E+00
2.21E+00
2.52E+00
2.88E+00
3.29E+00
3.77E+00
4.32E+00
4.99E+00
5.77E+00
6.68E+00
7.74E+00
8.96E+00
1.04E+01
1.20E+01
1.39E+01
1.61E+01
1.87E+01
2.17E+01
2.51E+01
2.90E+01
3.36E+01
3.88E+01
4.49E+01
5.18E+01
5.98E+01
6.88E+01
b
7.94E-02
7.96E-02
7.97E-02
7.97E-02
7.97E-02
7.96E-02
7.94E-02
7.92E-02
7.93E-02
7.96E-02
7.97E-02
7.99E-02
8.00E-02
8.01E-02
8.02E-02
B.03E-02
8.04E-02
8.04E-02
8.05E-02
8.05E-02
8.05E-02
8.06E-02
8.06E-02
8.06E-02
8.06E-02
8.06E-02
8.07E-02
bbx
1.84E+01
2.14E+01
2.50E+01
2.92E+01
3.40E+01
3.97E+01
4.63E+01
5.40E+01
6.30E+01
7.34E+01
8.56E+01
9.98E+01
1.16E+02
1.36E+02
1.58E+02
1.85E+02
2.15E+02
2.51E+02
2.93E+02
3.41E+02
3.98E+02
4.64E+02
5.41E+02
6.30E+02
7.35E+02
8.57E+02
9.99E+02
bx
1.84E+01
2.14E+01
2.50E+01
2.92E+01
3.40E+01
3.97E+01
4.63E+01
5.40E+01
6.30E+01
7.34E+01
8.56E+01
9.98E+01
1.16E+02
1.36E+02
1.58E+02
1.85E+02
2.15E+02
2.51E+02
2.93E+02
3.41E+02
3.98E+02
4.64E+02
5.41E+02
6.30E+02
7.35E+02
8.57E+02
9.99E+02
cv
6.57E-04
5.03E-04
3.84E-04
2.91E-04
2.20E-04
1.66E-04
1.25E-04
9.37E-05
7.02E-05
5.25E-05
3.93E-05
2.93E-05
2.19E-05
1.64E-05
1.22E-05
9.11E-06
6.80E-06
5.08E-06
3.79E-06
2.84E-06
2.12E-06
1.59E-06
1.19E-06
8.93E-07
6.71E-07
5.05E-07
3.81E-07
rno
1.22E+00
L.22E+00
L.22E+00
L.22E+00
1.22E+00
.22E+00
1.22E+00
1.22E+00
1.22E+00
1.22E+00
1.22E+00
1.22E+00
.22E+00
.22E+00
.22E+00
.22E+00
.22E+00
.22E+00
.22E+00
.22E+00
.22E+00
.22E+00
.22E+00
.22E+00
.22E+00
.22E+00
1.22E+00
t
2.88E+02
2.88E+02
2.88E+02
2.88E+02
2.88E+02
2.88E+02
2.88E+02
2.88E+02
2.88E+02
2.88E+02
2.88E+02
2.88E+02
2.88E+02
2.88E+02
2.88E+02
2.88E+02
2.88E+02
2.88E+02
2.88E+02
2.88E+02
2.88E+02
2.88E+02
2.88E+02
2.88E+02
2.88E+02
2.88E+02
2.88E+02
u
4.41E+00
4.49E+00
4.58E+00
4.67E+00
4.76E+00
4.85E+00
4.95E+00
5.05E+00
5.14E+00
5.23E+00
5.33E+00
5.42E+00
5.52E+00
5.61E+00
5.70E+00
5.80E+00
5.89E+00
5.98E+00
6.07E+00
6.15E+00
6.24E+00
6.33E+00
6.42E+00
6.51E+00
6.60E+00
6.68E+00
6.77E+00
ua
4.35E+00
4.45E+00
4.55E+00
4.65E+00
4.75E+00
4.85E+00
4.95E+00
5.05E+00
5.14E+00
5.24E+00
5.33E+00
5.43E+00
5.52E+00
5.61E+00
5.71E+00
5.80E+00
5.89E+00
5.99E+00
6.08E+00
6.17E+00
6.26E+00
6.35E+00
6.44E+00
6.53E+00
6.62E+00
6.71E+00
6.79E+00
-------
TABLE 7-14 (CONTINUED)
X
l.OOE+00
1.02E+00
1.04E+00
1.06E+00
1.08E+00
1.12E+00
1.15E+00
1.19E+00
1.24E+00
1.30E+00
1.36E+00
1.44E+00
1 . 53E+00
1.64E+00
1.76E+00
1.90E+00
2.07E+00
2.26E+00
2.48E+00
2.75E+00
3.05E+00
3.41E+00
3.83E+00
4.31E+00
4.88E+00
5.54E+00
6.31E+00
7.21E+00
8.25E+00
9.47E+00
1.09E+01
1.26E+01
1.45E+01
1.67E+01
1.94E+01
2.24E+01
cm
l.OOE+00
1.93E-01
1.66E-01
1.45E-01
1.28E-01
1.13E-01
9.97E-02
8.84E-02
7.86E-02
6.99E-02
6.21E-02
5.53E-02
4.91E-02
4.37E-02
3.87E-02
3.43E-02
3.03E-02
2.67E-02
2.35E-02
2.05E-02
1.78E-02
1.54E-02
1.32E-02
1.12E-02
9.44E-03
7.88E-03
6.51E-03
5.32E-03
4.30E-03
3.45E-03
2.74E-03
2.15E-03
1.68E-03
1.30E-03
l.OOE-03
7.68E-04
cmv
9.83E-01
1.93E-01
1.66E-01
1.45E-01
1.28E-01
1.13E-01
9.97E-02
8.84E-02
7.86E-02
6.99E-02
6.21E-02
5.53E-02
4.91E-02
4.37E-02
3.87E-02
3.43E-02
3.03E-02
2.67E-02
2.35E-02
2.05E-02
1.78E-02
1.54E-02
1.32E-02
1.12E-02
9.44E-03
7.88E-03
6.51E-03
5.32E-03
4.30E-03
3.45E-03
2.74E-03
2.15E-03
1 . 68E-03
1.30E-03
l.OOE-03
7.68E-04
cmda
O.OOE+00
8.02E-01
8.28E-01
8.49E-01
8.67E-01
8.82E-01
8.95E-01
9.06E-01
9.16E-01
9.24E-01
9.32E-01
9.39E-01
9.45E-01
9.50E-01
9.55E-01
9.59E-01
9.63E-01
9.67E-01
9.70E-01
9.73E-01
9.76E-01
9.78E-01
9.80E-01
9.82E-01
9.84E-01
9.86E-01
9.87E-01
9.88E-01
9.89E-01
9.90E-01
9.91E-01
9.91E-01
9.92E-01
9.92E-01
9.93E-01
9.93E-01
cmw
O.OOE+00
5.19E-03
5.36E-03
5.50E-03
5.61E-03
5.71E-03
5.79E-03
5.86E-03
5.93E-03
5.98E-03
6.03E-03
6.08E-03
6.12E-03
6.15E-03
6.18E-03
6.21E-03
6.24E-03
6.26E-03
6.28E-03
6.30E-03
6.32E-03
6.33E-03
6.35E-03
6.36E-03
6.37E-03
6.38E-03
6.39E-03
6.40E-03
6.40E-03
6.41E-03
6.42E-03
6.42E-03
6.42E-03
6.42E-03
6.43E-03
6.43E-03
cmwv
O.OOE+00
5.19E-03
5.36E-03
5.50E-03
5.61E-03
5.71E-03
5.79E-03
5.86E-03
5.93E-03
5.98E-03
6.03E-03
6.08E-03
6.12E-03
6.15E-03
6.18E-03
6.21E-03
6.24E-03
6.26E-03
6.28E-03
6.30E-03
6.32E-03
6.33E-03
6.35E-03
6.36E-03
6.37E-03
6.38E-03
6.39E-03
6.40E-03
6.40E-03
6.41E-03
6.42E-03
6.42E-03
6.42E-03
6.42E-03
6.43E-03
6.43E-03
we
O.OOE+00
O.OOE+00
O.OOE+00
0 . OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
vg
O.OOE+00
7.25E-04
1.56E-03
2.45E-03
3.43E-03
4.49E-03
5.67E-03
6.98E-03
8.42E-03
l.OOE-02
1.18E-02
1.37E-02
1.58E-02
1.80E-02
2.04E-02
2.29E-02
2.55E-02
2.82E-02
3.08E-02
3.34E-02
3.57E-02
3.79E-02
3.97E-02
4.10E-02
4.18E-02
4.20E-02
4.15E-02
4.05E-02
3.90E-02
3.70E-02
3.46E-02
3.20E-02
2.92E-02
2.65E-02
2.37E-02
2.11E-02
ug
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
w
2.52E+03
9.06E+00
6.86E+00
5.36E+00
4.30E+00
3.50E+00
2.90E+00
2.42E+00
2.05E+00
1.75E+00
1.51E+00
1.31E+00
1.14E+00
1.01E+00
9.01E-01
8.11E-01
7.36E-01
6.76E-01
6.26E-01
5.85E-01
5.52E-01
5.26E-01
5.04E-01
4.88E-01
4.75E-01
4.65E-01
4.57E-01
4.52E-01
4.48E-01
4.45E-01
4.43E-01
4.41E-01
4.40E-01
4.39E-01
4.38E-01
4.37E-01
V
7.82E+00
1.58E+00
1.35E+00
1.17E+00
1.02E+00
8.88E-01
7.79E-01
6.85E-01
6.03E-01
5.32E-01
4.70E-01
4.16E-01
3.69E-01
3.28E-01
2.93E-01
2.63E-01
2.38E-01
2.17E-01
2.00E-01
1.87E-01
1.76E-01
1.69E-01
1.63E-01
1.60E-01
1.59E-01
1.59E-01
1.60E-01
1.62E-01
1.65E-01
1.68E-01
1.71E-01
1.75E-01
1.79E-01
1.83E-01
1.87E-01
1.91E-OI
vx
O.OOE+00
4.35E-01
4.50E-01
4.63E-01
4.73E-01
4.82E-01
4.90E-01
4.97E-01
5.03E-01
5.08E-01
5.13E-01
5.17E-01
5.20E-01
5.23E-01
5.25E-01
5.27E-01
5.29E-01
5.30E-01
5.31E-01
5.32E-01
5.32E-01
5.33E-01
5.33E-01
5.33E-01
5.32E-01
5.32E-01
5.31E-01
5.30E-01
5.29E-01
5.28E-01
5.27E-01
5.25E-01
5.23E-01
5.22E-01
5.20E-01
5.18E-01
-------
TABLE 7-14 (CONTINUED)
X
2.60E+01
3.02E+01
3.50E+01
4.07E+01
4.73E+01
5.50E+01
6.40E+01
7.44E+01
8.66E+01
1.01E+02
1.17E+02
1.37E+02
1.59E+02
1.86E+02
2.16E+02
2.52E+02
2.94E+02
3.42E+02
3.99E+02
4.65E+02
5.42E+02
6.31E+02
7.36E+02
8.58E+02
1 OOE+03
* in
5.86E-04
4.45E-04
3.36E-04
2.53E-04
1.91E-04
1.43E-04
1.07E-04
8.02E-05
6.00E-05
4.48E-05
3.35E-05
2.50E-05
1.86E-05
1.39E-05
1.04E-05
7.75E-06
5.79E-06
4.33E-06
3.24E-06
•2.43E-06
1.82E-06
1.36E-06
1.02E-06
7.71E-07
5.81E-07
cmv
5.86E-04
4.45E-04
3.36E-04
2.53E-04
1.91E-04
1.43E-04
1.07E-04
8.02E-05
6.00E-05
4.48E-05
3.35E-05
2.50E-05
1.86E-05
1.39E-05
1.04E-05
7.75E-06
5.79E-06
4.33E-06
3.24E-06
2.43E-06
1.82E-06
1.36E-06
1.02E-06
7.71E-07
5.81E-07
cmda
9.93E-01
9.93E-01
9.93E-01
9.93E-01
9.93E-01
9.93E-01
9.93E-01
9.93E-01
9.94E-01
9.94E-01
9.94E-01
9.94E-01
9.94E-01
9.94E-01
9.94E-01
9.94E-01
9.94E-01
9.94E-01
9.94E-01
9.94E-01
9.94E-01
9.94E-01
9.94E-01
9.94E-01
9.94E-01
CUM
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
cniwv
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
6.43E-03
we
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
vg
1.86E-02
1.63E-02
1.42E-02
1.23E-02
1.07E-02
9.18E-03
7.88E-03
6.73E-03
5.74E-03
4.88E-03
4.14E-03
3.51E-03
2.97E-03
2.51E-03
2.13E-03
1.79E-03
1.51E-03
1.26E-03
1.06E-03
8.98E-04
7.66E-04
6.46E-04
5.46E-04
4.55E-04
3.75E-04
uy
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
w
4.36E-01
4.35E-01
4.33E-01
4.31E-01
4.29E-01
4.26E-01
4.24E-01
4.21E-01
4.18E-01
4.14E-01
4.10E-01
4.07E-01
4.03E-01
3.98E-01
3.94E-01
3.90E-01
3.85E-01
3.80E-01
3.75E-01
3.71E-01
3.66E-01
3.60E-01
3.55E-01
3.50E-01
3.44E-01
V
1.95E-01
1.99E-01
2.03E-01
2.07E-01
2.11E-01
2.15E-01
2.18E-01
2.22E-01
2.25E-01
2.29E-01
2.32E-01
2.35E-01
2.38E-01
2.41E-01
2.43E-01
2.45E-01
2.47E-01
2.49E-01
2.50E-01
2.51E-01
2.51E-01
2.51E-01
2.50E-01
2.49E-01
2.47E-01
VX
5.16E-01
5.14E-01
5.12E-01
5.11E-01
5.09E-01
5.07E-01
5.06E-01
5.05E-01
5.04E-01
5.03E-01
5.02E-01
5.01E-01
5.01E-01
5.00E-01
5.00E-01
5.00E-01
5.00E-01
5.00E-01
5.00E-01
5.00E-01
5.01E-01
5.01E-01
5.02E-01
5.02E-01
5.03E-01
-------
TABLE 7-14 (CONTINUED)
time averaged (tav = 5. s) volume concentration: concentration contour parameters
c(x,y,z,t) = cc(x) * (erf(xa)-erf(xb)) * (erf(ya)-erf(yb)) * (exp(-za*za)+exp(-zb*zb))
c(x,y,z,t) = concentration (volume fraction) at (x,y,z,t)
x = downwind distance (m)
y = crosswind horizontal distance (m)
z = height (m)
t = time (s)
Oi
erf
xa
xb
ya
yb
exp
za
zb
sr2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
X
.OOE+00
.02E+00
.04E+00
.06E+00
.08E+00
.12E+00
.15E+00
.19E+00
.24E+00
.30E+00
.36E+00
.44E+00
.53E+00
.64E+00
.76E+00
.90E+00
.07E+00
0,
4
3,
3
3
3,
2,
2,
2.
2,
2,
2
2,
1
1
1
1
= error functon
= (x-xc+bx)/{sr2*betax)
= (x-xc-bx)/(sr2*betax).
= (y+b)/(sr2*betac)
= (y-b)/(sr2*betac)
= exponential function
= (z-zc)/(sr2*sig)
= (z+zc)/(sr2*sig)
= sqrt(2.0)
cc(x)
.OOE+00
. 18E-02
.86E-02
.58E-02
.35E-02
,14E-02
.95E-02
,77E-02
,61E-02
,46E-02
.32E-02
. 18E-02
,05E-02
.93E-02
.82E-02
.71E-02
.60E-02
9,
3.
3,
3,
3,
3.
4,
4,
4.
4.
5.
5.
5.
5.
5,
5
6.
b(x)
71E-03
.04E-02
.29E-02
.52E-02
.74E-02
.96E-02
.18E-02
.39E-02
,61E-02
,82E-02
.02E-02
.23E-02
.43E-02
.62E-02
.81E-02
.99E-02
.16E-02
betac(x)
2
2
3
3
4
4
5
6
7
7
8
9
1
1
.71E-03
.61E-02
.13E-02
.68E-02
.27E-02
.91E-02
.59E-02
.32E-02
.11E-02
.97E-02
.89E-02
.88E-02
.09E-01
.21E-01
1.33E-01
1
1
.46E-01
.60E-01
0.
0
0
0
0
zc(x)
OOE+00
.OOE+00
.OOE+00
.OOE+00
.OOE+00
O.OOE+00
0
0
0
0
0
0
0
0
0
0
0
.OOE+00
.OOE+00
.OOE+00
. OOE+00
.OOE+00
.OOE+00
.OOE+00
.OOE+00
.OOE+00
.OOE+00
.OOE+00
6
6
7
8
9
1
1
1
1
1
1
2
2
2
2
3
3
sig(x)
.23E-03
.29E-02
.32E-02
.42E-02
.58E-02
.08E-01
.22E-01
.36E-01
.52E-01
.69E-01
.87E-01
.07E-01
.28E-01
.50E-01
.74E-01
.OOE-01
.28E-01
0.OOE+00
1.34E-03
3.33E-03
6.00E-03
9.53E-03
1.42E-02
2.02E-02
2.80E-02
3.81E-02
5.09E-02
6.72E-02
8.79E-02
1.14E-01
1.46E-01
1.87E-01
2.37E-01
2.99E-01
xc(t)
1.OOE+00
1.02E+00
1.04E+00
1.06E+00
1.08E+00
1.12E+00
1.15E+00
1.19E+00
1.24E+00
1.30E+00
1.36E+00
1.44E+00
1.53E+00
1.64E+00
1.76E+00
1.90E+00
2.07E+00
bx(t)
0.OOE+00
1.66E-02
3.59E-02
5.85E-02
8.48E-02
1.15E-01
1.51E-01
1.93E-01
2.41E-01
2.98E-01
3.64E-01
4.41E-01
5.31E-01
6.35E-01
7.57E-01
9.OOE-01
1.07E+00
betax(t)
0.OOE+00
1.35E-04
2.93E-04
4.78E-04
6.92E-04
9.43E-04
1.23E-03
1.57E-03
1.97E-03
2.43E-03
2.97E-03
3.60E-03
4.34E-03
5.19E-03
6.19E-03
7.35E-03
8.70E-03
-------
TABLE 7-14 (CONTINUED)
X
2.26E+00
2.48E+00
2.75E+00
3.05E+00
3.41E+00
3.83E+00
4.31E+00
4.88E+00
5.54E+00
6.31E+00
7.21E+00
8.25E+00
9.47E+00
1.09E+01
1.26E+01
1.45E+01
1.67E+01
1.94E+01
2.24E+01
2.60E+01
3.02E+01
3.50E+01
4.07E+01
4.73E+01
5.50E+01
6.40E+01
7.44E+01
8.66E+01
1.01E+02
1.17E+02
1.37E+02
1 . 59E+02
1.86E+02
2.16E+02
2.52E+02
1C (X)
1.50E-02
1.40E-02
1.31E-02
1.21E-02
1.12E-02
1.03E-02
9.42E-03
8.59E-03
7.78E-03
6.99E-03
6.25E-03
5.54E-03
4.89E-03
4.30E-03
3.76E-03
3.28E-03
2.85E-03
2.47E-03
2.15E-03
1.86E-03
1.61E-03
1.39E-03
1.20E-03
1.04E-03
8.97E-04
7.75E-04
6.70E-04
5.78E-04
5 OOE-04
4.31E-04
3.73E-04
3.22E-04
2.78E-04
2.40E-04
2.08E-04
l'(x)
6.31E-02
6.45E-02
6.59E-02
6.73E-02
6.86E-02
6.98E-02
7.11E-02
7.20E-02
7.30E-02
7.39E-02
7.48E-02
7.58E-02
7.66E-02
7.74E-02
7.82E-02
7.88E-02
7.92E-02
7.94E-02
7.96E-02
7.97E-02
7.97E-02
7.97E-02
7.96E-02
7.94E-02
7.92E-02
7.93E-02
7.96E-02
7.97E-02
7.99E-02
8.00E-02
8.01E-02
8.02E-02
6.03E-02
8.04E-02
8.04E-02
betac(x)
1.75E-01
1.92E-01
2.10E-01
2.29E-01
2.51E-01
2.74E-01
3.01E-01
3.31E-01
3.64E-01
4.02E-01
4.46E-01
4.95E-01
5.52E-01
6.18E-01
6.94E-01
7.82E-01
8.82E-01
9.98E-01
1.13E+00
1.29E+00
1.47E+00
1.68E+00
1.92E+00
2.21E+00
2.53E+00
2.93E+00
3.39E+00
3.92E+00
4.54E+00
5.27E+00
6.10E+00
7.07E+00
8.19E+00
9.49E+00
1.10E+01
zc(x)
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
.O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
sig(x)
3.58E-01
3.91E-01
4
4
.26E-01
.65E-01
5.08E-01
5
6
6
7
8
8
9
1
1
1
1
1
1
2
2
2.
3.
3.
4.
4.
5.
6.
7.
7.
9.
1.
1.
1.
1.
1.
.55E-01
.07E-01
.66E-01
.33E-01
.08E-01
.94E-01
.92E-01
.10E+00
.23E+00
.38E+00
.55E+00
.75E+00
.97E+00
.23E+00
.52E+00
.86E+00
,25E+00
70E+00
21E+00
79E+00
44E+00
17E+00
01E+00
96E+00
05E+00
03E+01
17E+01
33E+01
52E+01
73E+01
3
4
5
7
8
1
1
1
1
2
2
3
3
4
5
t
.75E-01
.67E-01
.79E-01
.15E-01
.77E-01
.07E+00
.30E+00
.58E+00
.90E+00
.28E+00
.72E+00
.24E+00
.83E+00
.52E+00
.32E+00
6.24E+00
7
8.
9.
1.
1.
1.
1.
2.
2.
2.
3.
3.
4.
4.
5.
6.
7.
8.
.29E+00
.49E+00
,87E+00
14E+01
32E+01
53E+01
77E+01
04E+01
34E+01
70E+01
10E+01
56E+01
09E+01
70E+01
39E+01
19E+01
10E+01
15E+01
9.35E+01
xc(t)
2.26E+00
2.48E+00
2.75E+00
3.05E+00
3.41E+00
3.83E+00
4.31E+00
4.88E+00
5.54E+00
6.31E+00
7.21E+00
8.25E+00
9.47E+00
1.09E+01
1.26E+01
1.45E+01
1.67E+01
1.94E+01
2.24E+01
2.60E+01
3.02E+01
3 . 50E+01
4.07E+01
4.73E+01
5.50E+01
6.40E+01
7.44E+01
8.66E+01
1.01E+02
1.17E+02
1.37E+02
1.59E+02
1.86E+02
2.16E+02
2.52E+02
bx(t)
1.26E+00
1.48E+00
1 . 75E+00
2.05E+00
2.41E+00
2.83E+00
3.31E+00
3.88E+00
4.54E+00
5.31E+00
6.21E+00
7.25E+00
8.47E+00
9.90E+00
1.16E+01
1.35E+01
1.57E+01
1.84E+01
2.14E+01
2.50E+01
2.92E+01
3.40E+01
3.97E+01
4.63E+01
5.40E+01
6.30E+01
7.34E+01
8.56E+01
9.98E+01
1.16E+02
1.36E+02
1.58E+02
1.85E+02
2.15E+02
2.51E+02
betax(t)
1.03E-02
1.21E-02
1.43E-02
1.68E-02
1.97E-02
2.31E-02
2.71E-02
3.17E-02
3.71E-02
4.34E-02
5.07E-02
5.92E-02
6.92E-02
8.08E-02
9.44E-02
1.10E-01
1.29E-01
1.50E-01
1.75E-01
2.04E-01
2.38E-01
2.78E-01
3.24E-01
3.78E-01
4.41E-01
5.14E-01
5.99E-01
6.99E-01
8.15E-01
9.51E-01
1.11E+00
1.29E+00
1.51E+00
1.76E+00
2.05E+00
-------
TABLE 7-14 (CONTINUED)
O-i
ui
X
2.94E+02 1
3.42E+02 1
3.99E+02 1
4.65E+02 1
5.42E+02 1
6.31E+02 8
7.36E+02 7
8.58E+02 6
l.OOE+03 5
time averaged
downwind
distance
, x (m)
l.OOE+00
1.02E+00
1.04E+00
1.06E+00
1 . 08E+00
1.12E+00
1.15E+00
. 1.19E+00
1.24E+00
1.30E+00
1.36E+00
1.44E+00
1.53E+00
1.64E+00
1.76E+00
1.90E+00
2.07E+00
2.26E+00
2.48E+00
2.75E+00
3.05E+00
3.41E+00
cc(x) b(x) betac(x) zc(x) sig(x) t xc(t) bx(t) betax(t)
80E-04 8.05E-02 1.27E+01 O.OOE+00 1.97E+01 1.07E+02 2.94E+02 2.93E+02 2.39E+00
56E-04 8.05E-02 1.48E+01 O.OOE+00 2.24E+01 1.23E+02 3.42E+02 3.41E+02 2.79E+00
35E-04 8.05E-02 1.71E+01 O.OOE+00 2.55E+01 1.42E+02 3.99E+02 3.98E+02 3.25E+00
17E-04 8.06E-02 1.98E+01 O.OOE+00 2.90E+01 1.62E+02 4.65E+02 4.64E+02 3.79E+00
01E-04 8.06E-02 2.29E+01 O.OOE+00 3.30E+01 1.87E+02 5.42E+02 5.41E+02 4.41E+00
76E-05 8.06E-02 2.64E+01 O.OOE+00 3.76E+01 2.14E+02 6.31E+02 6.30E+02 5.15E+00
60E-05 8.06E-02 3.05E+01 O.OOE+00 4.27E+01 2.46E+02 7.36E+02 7.35E+02 6.00E+00
60E-05 8.06E-02 3.52E+01 O.OOE+00 4.86E+01 -2.83E+02 8.58E+02 8.57E+02 7.00E+00
73E-05 8.07E-02 4.05E+01 O.OOE+00 5.53E+01 3.25E+02 l.OOE+03 9.99E+02 8.16E+00
(tav = 5. s) volume concentration: concentration in the z = .00 plane.
time of cloud effective average concentration (volume fraction) at (x,y,z)
max cone
(s)
2.40E+02
2.40E+02
2.40E+02
'2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
duration
(s)
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.60E+02
4.80E+02
4.80E+02
half width
1
5
6
7
B
9
2
2
2
2
3
3
3
4
4
bbc (m)
.08E-02
.44E-02
.34E-02
.28E-02
.29E-02
.38E-02
.05E-01
.18E-01
.32E-01
.46E-01
.62E-01
79E-01
97E-01
17E-01
38E-01
60E-01
84E-01
10E-01
38E-01
69E-01
03E-01
39E-01
y/bbc=
0.0
l.OOE+00
2
2
1.
1.
1.
1.
1.
1.
8.
7.
7.
6.
5.
4.
4.
3.
3.
2.
2.
2.
1.
53E-01
18E-01
89E-01
66E-01
46E-01
29E-01
14E-01
01E-01
94E-02
93E-02
04E-02
25E-02
54E-02
91E-02
34E-02
83E-02
37E-02
96E-02
58E-02
24E-02
93E-02
y/bbc=
0.5
l.OOE+00
1.79E-01
1.53E-01
1.32E-01
1.15E-01
1.01E-01
8.89E-02
7.85E-02
6.95E-02
6.16E-02
5.46E-02
4.84E-02
4.30E-02
3.81E-02
3.37E-02
2.98E-02
2.63E-02
2.32E-02
2.03E-02
1.77E-02
1.54E-02
1.33E-02
y/bbc=
1.0
5.31E-01
5.94E-02
5.04E-02
4.34E-02
3.77E-02
3.30E-02
2.90E-02
2.56E-02
2.26E-02
2.00E-02
1.78E-02
1.58E-02
1.40E-02
1.24E-02
I.10E-02
9.70E-03
8.56E-03
7.53E-03
6.60E-03
5.76E-03
5.00E-03
4.31E-03
y/bbc=
1.5
1.31E-02
8.25E-03
7.21E-03
6.33E-03
5.57E-03
4.92E-03
4.36E-03
3.86E-03
3.43E-03
3.05E-03
2.71E-03
2.40E-03
2.13E-03
1.89E-03
1.68E-03
1.48E-03
1.31E-03
1.15E-03
1.01E-03
8.83E-04
7.67E-04
6.61E-04
y/bbc=
2.0
9.52E-06
4.37E-04
4.15E-04
.3.86E-04
3.54E-04
3.23E-04
2.92E-04
2.63E-04
2.37E-04
2.12E-04
1.90E-04
1.70E-04
1.52E-04
1.35E-04
1.20E-04
1.06E-04
9.42E-05
8.30E-05
7.29E-05
6.37E-05
5.54E-05
4.78E-05
y/bbc=
2.5
O.OOE+00
8.44E-06
9.23E-06
9.48E-06
9.37E-06
9.02E-06
8.52E-06
7.94E-06
7.32E-06
6.70E-06
6.10E-06
5.53E-06
4.98E-06
4.47E-06
4.00E-06
3.57E-06
3.17E-06
2.80E-06
2.47E-06
2.17E-06
1.88E-06
1.63E-06
-------
TABLE 7-14 (CONTINUED)
>IOWI,\,M >l
distance
x (m)
3.83E+00
4.31E+00
4.88E+00
5.54E+00
6.31E+00
7.21E+00
8.25E+00
9.47E+00
1.09E+01
1.26E+01
1.45E+01
1.67E+01
1.94E-MU
2.24E+01
2.60E+01
3.02E+01
3.50E+01
4.07E+01
4.73E+01
5.50E+01
6.40E+01
7.44E+01
8.66E+01
1.01E+02
1.17E+02
1.37E+02
1 . 59E+02
1.86E+02
2.16E+02
2.52E+02
2.94E+02
3.42E+02
3.99E+02
time of
max1 cone
(s)
2.40E+02
2.41E+02
2.41E+02
2.41E+02
2.41E+02
2.41E+02
2.41E+02
2.41E+02
2.42E+02
2.42E+02
2.42E+02
2.43E+02
2.43E+02
2.43E+02
2.44E+02
2.45E+02
2.46E+02
2.46E+02
2.48E+02
2.49E+02
2.50E+02
2.52E+02
2.54E+02
2.56E+02
2.59E+02
2.62E+02
2.66E+02
2.70E+02
2.75E+02
2.81E+02
2.88E+02
2.96E+02
3.05E+02
cloud
duration
(S)
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E*02
4.80E+02
4.80E+02
4.60E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4 . 80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
effective
half width
average concentration (volume fraction) at (x,y
y/bbc= y/bbc= y/bbc= y/bbc= y/bbc=
bbc (m)
4
5
5
6
7
7
8
9
1
2
2
2
3
3
4
5
5
6
7
9
1
1
1
1
1.
2.
2.
2.
.aoE-oi
.26E-01
.77E-01
35E-01
OOE-01
76E-01
61E-01
60E-01
07E+00
20E+00
36E+00
53E+00
73E+00
96E+00
23E+00
55E+00
91E+00
33E+00
82E+00
39E+00
07E+00
87E+00
80E+00
87E+00
12E+00
06E+01
22E+01
42E+01
64E+01
91E+01
21E+01
56E+01
96E+01
1
1
1
9
8
6
5
4
3
2
2
1
1
9
7
5.
4
3.
2.
1.
1.
9.
7.
5.
4.
3.
2.
1.
1.
9.
7.
5.
3.
0.0
66E-02
41E-02
18E-02
87E-03
15E-03
65E-03
38E-03
31E-03
41E-03
68E-03
09E-03
62E-03
25E-03
53E-04
25E-04
50E-04
15E-04
13E-04
35E-04
76E-04
32E-04
87E-05
38E-05
51E-05
HE-OS
07E-05
29E-05
71E-05
27E-05
52E-06
11E-06
31E-06
97E-06
0.5
1.14E-02
9.67E-03
8.14E-03
6.78E-03
5.60E-03
4.57E-03
3.70E-03
2.96E-03
2.35E-03
1.84E-03
1.44E-03
1.11E-03
8.56E-04
6.55E-04
4.99E-04
3.78E-04
2.85E-04
2.15E-04
1.62E-04
1.21E-04
9.07E-05
6.79E-05
5.07E-05
3.79E-05
2.83E-05
2.11E-05
1.57E-05
1.17E-05
8.76E-06
6.54E-06
4.89E-06
3.65E-06
2.73E-06
1.0
3.70E-03
3.14E-03
2.64E-03
2.20E-03
1.82E-03
1.48E-03
1.20E-03
9.61E-04
7.62E-04
5.99E-04
4.67E-04
3.61E-04
2.78E-04
2.13E-04
1.62E-04
1.23E-04
9.27E-05
6.98E-05
5.25E-05
3.93E-05
2.95E-05
2.20E-05
1.65E-05
1.23E-05
9.17E-06
6.84E-06
5.11E-06
3.81E-06
2.84E-06
2.12E-06
1.59E-06
1.19E-06
8.87E-07
1.5
5.67E-04
4.81E-04
4.05E-04
3.38E-04
2.79E-04
2.28E-04
1.84E-04
1.47E-04
1.17E-04
9.18E-05
7.16E-05
5.54E-05
4.26E-05
3.26E-05
2.48E-05
1.88E-05
1.42E-05
1.07E-05
8.04E-06
6.03E-06
4.52E-06
3.38E-06
2.52E-06
1.88E-06
1.41E-06
1.05E-06
7.83E-07
5.84E-07
4.36E-07
3.26E-07
2.43E-07
1.82E-07
1.36E-07
2.0
4.10E-05
3.48E-05
2.93E-05
2.45E-05
2.02E-05
1.65E-05
1.33E-05
1.07E-05
8.46E-06
6.65E-06
5.19E-06
4.01E-06
3.09E-06
2.36E-06
1.80E-06
1.36E-06
1.03E-06
7.76E-07
5.83E-07
4.37E-07
3.27E-07
2.45E-07
1.83E-07
1.37E-07
1.02E-07
7.61E-08
5.67E-08
4.23E-08
3.16E-08
2.36E-08
1.76E-08
1.32E-08
9.86E-09
z)
y/bbc=
2 5
1.40E-06
1.19E-06
l.OOE-06
8.35E-07
6.91E-07
5.64E-07
4.56E-07
3.64E-07
2.90E-07
2.27E-07
1.78E-07
1.38E-07
1.06E-07
8.11E-08
6.14E-08
4.66E-08
3.53E-08
2.66E-08
2.00E-08
1.50E-08
1.13E-08
8.48E-09
6.24E-09
4.56E-09
3.54E-09
2.62E-09
1.96E-09
1.43E-09
1.07E-09
7.78E-10
5.69E-10
4.37E-10
3.15E-10.
-------
TABLE 7-14 (CONTINUED)
O-i
downwind
•distance
4
5
6
7
8
1
time
x (m)
.65E+02
.42E+02
.31E+02
.36E+02
.58E+02
. OOE+03
averaged
downwi nd
distance
1
1
1
1
1
1
1
1
x (m)
.OOE+00
.02E+00
.04E+00
.06E+00
.08E+00
.12E+00
.15E+00
19E+00
1.24E+00
.30E+00
.36E+00
1.
1.
2.
2.
2.
2.
44E+00
53E+00
64E+00
76E+00
90E+00
07E+00
26E+00
48E+00
75E+00
tune of
max cone
(s)
3.15E+02
3.28E+02
3.43E+02
3.60E+02
3.80E+02
4 03E+02
(tav = 5.
time of
max cone
(s)
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
cloud
duration
(s)
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
s) volume
cloud
duration
(s)
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4 . 80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
effective
half width
bbc (in)
3.43E+01
3.96E+01
4.58E+01
5.29E+01
6.10E+01
7.02E+01
concentration:
effective
half width
bbc (m)
1.08E-02
5.44E-02
6.34E-02
7.28E-02
8.29E-02
9.38E-02
1.05E-01
1.18E-01
1.32E-01
1.46E-01
1.62E-01
1.79E-01
1.97E-01
2.17E-01
2.38E-01
2.60E-01
2.84E-01
3.10E-01
3.38E-01
3.69E-01
average concentration (volume fraction) at (x,y,z)
y/bbc= y/bbc= ' y/bbc= y/bbc= y/bbc= y/br
2
2
1
1
9
7
0.0
.97E-06
.23E-06
.67E-06
.26E-06
.46E-07
.13E-07
concentration
0.5
2.04E-06
1.53E-06
1.15E-06
8.64E-07
6.50E-07
4.90E-07
In the z =
1.0
6.64E-07
4.97E-07
3.73E-07
2.80E-07
2.11E-07
1.59E-07
3.66 plane
1.5
1.02E-07
7.63E-08
5.72E-08
4.30E-08
3.24E-08
2.44E-08
2.0
7.36E-09
5.53E-09
• 4.14E-09
3.13E-09
2.34E-09
1.77E-09
average concentration (volume fraction) at (x,y
y/bbc= y/bbc= y/bbc= y/bbc= y/bbc=
0
0
0
0
0
0
0.0
.OOE+00
.OOE+00
. OOE+00
.OOE+00
.OOE+00
OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0
0.
0.
0.
1.
2.
4.
7.
2.
2.
OOE+00
OOE+00
OOE+00
OOE+00
14E-40
50E-34
07E-29
66E-25
74E-21
55E-18
0.5
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
7.83E-41
1.72E-34
2.80E-29
5.26E-25
1.88E-21
1.75E-18
1.0
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
2.54E-41
5.59E-35
9.09E-30
1.71E-25
6.11E-22
5.68E-19
1.5
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
3.89E-42
8.56E-36
1.39E-30
2.62E-26
9.36E-23
6.71E-20
2.0
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
2.79E-43
6.14E-37
l.OOE-31
1.89E-27
6.75E-24
6.29E-21
? '..
2.73E-10
1.89E-10
1.23E-10
1.07E-10
9.25E 11
5.35E-11
y/bbc=
2.5
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
0. OOE+00
9.81E-45
2.06E-38
3.37E-33
6.37E-29
2.28E-25
2.14E-22
-------
TABLE 7-14 (CONTINUED)
00
1 ..,,,,,,..1
listance
x (|n)
3.05E+00
3.41E+00
3.83E+00
4.31E+00
4.88E+00
5.54E+00
6.31E+00
7.21E+00
8.25E+00
9.47E+00
1.09E+01
1.26E+01
1.45E+01
1.67E+01
1.94E+01
2.24E+01
2.60E+01
3.02E+01
3.50E+01
4.07E+01
4.73E+01
5.50E+01
6.40E+01
7.44E+01
8.66E+01
1.01E+02
1.17E+02
1.37E+02
1.59E+02
1.86E+02
2.16E+02
2.52E+02
2.94E+02
3.42E+02
i i
max cone
(s)
2.40E+02
2 40E+02
2.40E+02
2.41E+02
2.41E+02
2.41E+02
2.41E+02
2.41E+02
2.41E+02
2.41E+02
2.42E+02
2.42E+02
2.42E+02
2.43E+02
2.43E+02
2.43E+02
2.44E+02
2.45E+02
2.46E+02
2.46E+02
2.48E+02
2.49E+02
2.50E+02
2.52E+02
2.54E+02
2.56E+02
2.59E+02
2.62E+02
2.66E+02
2.70E+02
2.75E+02
2.81E+02
2.88E+02
2 96E+02
f.loil.l
duration
(s)
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
cl f PC! i vo
half width
4
4
4
5
5
6
7
7
8
9
2
2
2
3
3
4
5
5
6
7
9
1
1
1
1
1
2
2
bbc (m)
.03E-01
.39E-01
.80E-01
.26E-01
.77E-01
35E-01
OOE-01
76E-01
61E-01
60E-01
07E+00
20E+00
36E+00
53E+00
73E+00
96E+00
23E+00
55E+00
91E+00
33E+00
82E+00
39E+00
07E+00
87E+00
80E+00
87E+00
12E+00
06E+01
22E+01
42E+01
64E+01
91E+01
21E+01
56E+01
average concentration (vuiuine f i.-ictirm) ,il (f
y/bbc= y/bbc= y/bbc= y/bbc= .y/blj^-
7
9
5
1
3
3
2
1
5
1
4
8
1
1
2
2
2
2
2
1
1
1
1
8
6
4
3
2
2
1
1
9
6
5
0.0
.89E-16
.91E-14
.85E-12
.83E-10
.34E-09
.78E-08
.88E-07
.52E-06
.98E-06
.78E-05
.19E-05
.02E-05
.29E-04
.80E-04
.22E-04
.47E-04
.53E-04
.43E-04
.21E-04
.92E-04
.61E-04
.32E-04
.05E-04
.28E-05
.44E-05
.96E-05
.79E-05
.88E-05
.18E-05
.64E-05
24E-05
30E-06
99E-06
24E-06
0.5
5.43E-16
6.81E-14
4.02E-12
1.26E-10
2.29E-09
2.60E-08
1.98E-07
1.05E-06
4.11E-06
1.22E-05
2.88E-05
5.51E-05
8.85E-05
1.24E-04
.52E-04
.70E-04
.74E-04
.67E-04
.52E-04
1.32E-04
1.11E-04
9.05E-05
7.24E-05
5.69E-05
4.42E-05
3.41E-05
2.60E-05
1.98E-05
1.50E-05
1.13E-05
8.51E-06
6.40E-06
4.80E-06
3.60E-06
1.0
1.76E-16
2.21E-14
1.31E-12
4.09E-11
7.44E-10
8.44E-09
6.42E-08
3.40E-07
1.33E-06
3.97E-06
9.35E-06
1.79E-05
2.87E-05
4.01E-05
4.95E-05
5.52E-05
5.66E-05
5.42E-05
4.92E-05
4.28E-05
3.59E-05
2.94E-05
2.35E-05
1.85E-05
1.44E-05
1.11E-05
8.45E-06
6.43E-06
4.86E-06
3.67E-06
2.76E-06
2.08E-06
1.56E-06
1.17E-06
1.5
2.70E-17
3.39E-15
2.00E-13
6.27E-12
1.14E-10
1.29E-09
9.85E-09
5.21E-08
2.04E-07
6.09E-07
1.43E-06
2.74E-06
4.41E-06
6.15E-06
7.59E-06
8.46E-06
8.67E-06
8.31E-06
7.55E-06
6.56E-06
5.51E-06
4.51E-06
3.60E-06
2.83E-06
2.20E-06
1.70E-06
1.30E-06
9.85E-07
7.46E-07
5.63E-07
4.24E-07
3.18E-07
2.39E-07
1.79E-07
2.0
1.95E-18
2.45E-16
1.45E-14
4.53E-13
8.26E-12
9.37E-11
7.13E-10
3.77E-09
1.48E-08
4.41E-08
1.04E-07
1.99E-07
3.19E-07
4.46E-07
5.50E-07
6.13E-07
6.28E-07
6.02E-07
5.47E-07
4.75E-07
3.99E-07
3.26E-07
2.61E-07
2.05E-07
1.60E-07
1.23E-07
9.39E-08
7.14E-08
5.40E-08
4.07E-08
3.07E-08
2.31E-08
1.73E-08
1.30E-08
i
y/bl,.
2 5
6.63E-20
8.34E-18
4.94E-16
1.55E-14
2.82E-13
3.20E-12
2.44E-11
1.29E-10
5.07E-10
1.51E-09
3.56E-09
6.80E-09
1.10E-08
1.53E-08
1.89E-08
2.10E-08
2.14E-08
2.06E-08
1.88E-08
1.63E-08
1.37E-08
1.12E-08
9.00E-09
7.11E-09
5.44E-09
4.11E-09
3.26E-09
2.46E-09
1.87E-09
1.38E-09
1.04E-09
7.61E-10
5.79E-10
4.31E-10
-------
TABLE 7-14 (CONTINUED)
downwind
distance
x (m)
3.99E+02
4.65E+02
5.42E+02
6.31E+02
7.36E+02
8.58E+02
l.OOE+03
I line of
max cone
(s)
3.05E+02
3.15E+02
3.28E+02
3.43E+02
3 . 60E+02
3.80E+02
4.03E+02
cloud
duration
(s)
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
effective
half width
bbc (m)
2
3
3.
4
5
6.
7
.96E+01
.43E+01
.96E+01
.58E+01
.29E+01
10E+01
02E+01
average concentration (volume fraction) at (x,y,
y/bbc= y/bbc= y/bbc= y/bbc= y/bbc=
0.0 0.5 1.0 l.S 2.0
3
2
2
1
1
9.
7.
.93E-06
.95E-06
.22E-06
.66E-06
.25E-06
43E-07
11E-07
2.70E-06
2.03E-06
1.52E-06
1.14E-06
8.61E-07
6.48E-07
4.89E-07
8.78E-07
6.59E-07
4.94E-07
3.71E-07
2.79E-07
2.10E-07
1.59E-07
1.35E-07
1.01E-07
7.58E-08
5.70E-08
4.29E-08
3.23E-08
2.43E-08
9.76E-09
7.31E-09
5.50E-09
4.12E-09
3.12E-09
2.34E-09
1.76E-09
z)
y/bbr.-
2.5
3.12E-10
2.71E-10
1.88E-10
1.22E-10
1.06E-10
9.22E-11
5.34E-11
time-averaged (tav = 5. s) volume concentration: maximum concentration (volume fraction) along centerline.
cr>
vo
downwi nd
distance
x (m)
l.OOE+00
1.02E+00
1.04E+00
1.06E+00
1.08E+00
1.12E+00
1.15E+00
1.19E+00
1.24E+00
1.30E+00
1.36E+00
1.44E+00
1.53E+00
1.64E+00
1.76E+00
1.90E+00
2.07E+00
2.26E+00
2.48E+00
height
z (m)
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
maximum
concentration
C(X,0.2)
l.OOE+00
2.53E-01
2.18E-01
1.89E-01
1.66E-01
1.46E-01
1.29E-01
1.14E-01
1.01E-01
8.94E-02
7.93E-02
7.04E-02
6.25E-02
5.54E-02
4.91E-02
4.34E-02
3.83E-02
3.37E-02
2.96E-02
time of
max cone
(s)
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.40E+02
cloud
duration
(s)
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
-------
TABLE 7-14 (CONTINUED)
downwind
distance
x (m)
2.75E+00
3.05E+00
3.41E+00
3.83E+00
4/31E+00
4.88E+00
5.54E+00
6.31E+00
7.21E+00
8.25E+00
9.47E+00
1.09E+01
1.26E+01
1.45E+01
^ 1.67E+01
^ 1.94E+01
0 2.24E+01
2.60E+01
3.02E+01
3.50E+01
4.07E+01
4.73E+01
5.50E+01
6.40E+01
7.44E+01
8.66E+01
1.01E+02
1.17E+02
1.37E+02
1 . 59E+02
1.86E+02
2.16E+02
2.52E+02
2.94E+02
height
.z (m)
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
maximum
concentration
c(x.O.z)
2.58E-02
2.24E-02
1.93E-02
1.66E-02
1.41E-02
1.18E-02
9.87E-03
8.15E-03
6.65E-03
5.38E-03
4.31E-03
3.41E-03
2.68E-03
2.09E-03
1.62E-03
1.25E-03
9.53E-04
7.25E-04
5.50E-04
4.15E-04
3.13E-04
2.35E-04
1.76E-04
1.32E-04
9.87E-05
7.38E-05
5.51E-05
4.11E-05
3.07E-05
2.29E-05
1.71E-05
1.27E-05
9.52E-06
7.11E-06
time of
max cone
(s)
2.40E+02
2.40E+02
2.40E+02
2.40E+02
2.41E+02
2.41E+02
2.41E+02
2.41E+02
2.41E+02
2.41E+02
2.41E+02
2.42E+02
2.42E+02
2.42E+02
2.43E+02
2.43E+02
2.43E+02
2.44E+02
2.45E+02
2.46E+02
2.46E+02
2.48E+02
2.49E+02
2.50E+02
2.52E+02
2.54E+02
2.56E+02
2.59E+02
2.62E+02
2.66E+02
2.70E+02
2.75E+02
2.81E+02
2.88E+02
cloud
duration
(s)
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.BOE+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
-------
TABLE 7-14 (CONTINUED)
downwind
distance
x (m)
3.42E+02
3.99E+02
4.65E+02
5.42E+02
6.31E+02
7.36E+02
8.58E+02
l.OOE+03
height
i (m)
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
maximum
concentration
c(x.O.z)
5.31E-06
3.97E-06
2.97E-06
2.23E-06
1.67E-06
1.26E-06
9.46E-07
7.13E-07
time of
max cone
(s)
2.96E+02
3.05E+02
3.15E+02
3.28E+02
3.43E+02
3.60E+02
3.80E+02
4.03E+02
cloud
Jurat K M
(s)
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4.80E+02
4 . 80E+02
4.80E+02
4.80E+02
-------
believed to be significant. A simpler alternative for those willing to accept
a cruder approximation, is to approximate the enclosed area by assuming it to
be roughly equivalent to the area of an ellipse having the same maximum length
and width.
The output from a SLAB simulation includes:
• The input to the model;
• Instantaneous spatially averaged cloud parameters (two
tables);
• Time-averaged concentration parameters;
• Time-averaged concentrations at specified heights (up to
four tables); and
• Time-averaged concentration along centerline.
Of these, the specified concentrations at specified heights and along the
centerline are most important. Note that each simulation has one averaging
time output. The instantaneous, spatially averaged cloud parameters could be
used in determining other averaging times, but it is more efficient to rerun
the model if a different averaging time is required.
If maximum concentrations and/or maximum downwind extent are desired
regardless of altitude, the centerline concentration table should be used.
The centerline may be elevated. If the maximum values are only wanted for a.
specific altitude, normally surface level, the concentration tables at
specified heights should be used. If the plume is always at ground level, the
column labeled "y/bbc- 0.0" for the "z - .00 plane" table should agree with
the column labeled "maximum concentration" for the centerline table.
Specific parameters of interest are discussed below. In this
discussion, it is assumed that all parameters to be found are for a specified
height. For numerical examples, the "z - .00 plane" table in Table 7-14 is
being used. For maximum distances and concentrations, the columns labeled
7-72
-------
"downwind distance" and "y/bbc - 0.0" are used. The distances are given in
meters, and the concentrations are given as volume fraction. To convert the
concentrations to ppm, multiply the fractions by IxlO6.
Maximum off-site concentration.
Maximum time-averaged off-site concentrations.
•
Maximum time-averaged concentration at a specific point.
Maximum distance downwind of specified concentration.
Finding the concentration at a specific distance downwind (or the
distance downwind of a specific concentration) requires interpolation. For
example, to find the downwind distance of 30,000 ppm, locate two
concentrations in the "y/bbc =0.0" column which are on either side of
30,000 ppm or the fraction 3.00E-02. These two concentrations are 3.37E-02
and 2.96E-02. They correspond to distances 2.26E-KX) and 2.48E+00 meters,
respectively. By performing a linear interpolation between the two
concentrations and distances, a downwind distance of 2.46 meters for
30,000 ppm is calculated. A concentration at a specific distance can be found
by finding the closest entry in the "Downwind Distance" column for the
distance desired and interpolating the concentration.
Maximum width of specified concentration
To find the maximum width of a concentration, a little more effort is
required. The columns labeled "y/bbc" represent the concentrations at the
scaled cross-wind distances. Since the off-centerline distances are scaled
and are not actual distance values, they must be changed to actual values
before interpolation can be done.
Using a concentration of 30,000 ppm again for an example, find the
actual cross-wind distances of the concentration through interpolation for a
number of downwind distances. Then select the maximum calculated half-width
and double it for the total width. For the downwind distance of 1.76 meters,
the half-width of the 3.00E-02 concentration lies between "y/bbc - 0.5" and
7-73
-------
"y/bbc - 1.0". The value of bbc at that distance is 2.38E-01. Interpolating
the value of y/bbc gives a value of 0.581. The actual distance is then 0.138
meters. Repeating these calculations for a number of other distances leads to
the following list:
x(m)
1 . 24E+00
1.30E+00
1 . 36E+00
1.44E+00
1.53E+00
1 . 64E+00
1.76E+00
half- width (m)
1.22E-01
1.28E-01
1.35E-01
1.40E-01
1.43E-01
1.43E-01
1.38E-01
The largest half-width is 0.143 meters so the largest width is 0.29 meters.
Time at which the maximum concentration is reached at a specific point.
Duration above specified concentration at a point.
Time the specified concentration is reached at a point.
The time that a maximum concentration is reached and the duration of a.
concentration is listed in both types of tables as a function of the downwind
distance. Therefore, these values can be interpolated the same way that
concentration is interpolated. Duration of some concentration, other than the
maximum, at a specific point cannot be determined directly from the
concentration output.
7-74
-------
Total area impacted by specified concentration.
As stated above, to calculate the area impacted by a specified
concentration, a third-party software package is required. The concentration
tables would be used as input or, if a cruder estimate is acceptable,
approximating the shape of the contour by assuming it to be an ellipse. In
the latter case, the area enclosed by the contour would be estimated by the
following equation:
AREA = (i) x (X^) x (Y.J.
where:
is the centerline maximum downwind distance and ¥„,„ is the maximum
contour width.
7-75
-------
,v
-------
SECTION 8
DETERMINING INPUT FOR MODELING "WORST CASE" IMPACTS
Accidental releases can occur at any time. For this reason the "worst-
case" impact from a release is sometimes needed. Examples of impacts of
concern are given in Section 7.0. To determine a worst-case impact, one -
should consider input that provides maximum or most-effective release rate and
worst meteorological dispersion conditions. Maximum release rate from a tank
occurs when the tank is assumed the most full and the hole size is the biggest
possible (i.e., a full cylinder breaks in half). For most releases the most-
effective release rate is equal to the maximum release rate. For reasons
discussed below this may not be the case for jets. Worst meteorological
conditions for non-jet ground level releases are those associated with very
stable atmospheric conditions and low wind speeds, conditions normally give
poor dispersion. For other releases multiple stability classes and wind
speeds need to be modeled to determine the worst meteorological conditions.
This section discusses the effects which specific input can have on the
type of models considered in this study. This section also discusses the
mechanics of running a model multiple times and how this document can be used
in an off-site consequence analysis.
8.1 MODEL INPUT
Unfortunately, it is not simple to specify all input conditions for
determining the worst-case since, in source-term models, the same input plays
multiple roles. For example, high wind speeds can lead to high evaporative
rates from surface liquid pools, but are also associated with decreased
atmospheric stability (and enhanced dispersion). On the other hand, the
relationship between ambient temperature and worst-case predicted impact is
8-1
-------
more straightforward--higher ambient temperatures tend to result in higher
emission rates and more tendency for the release to behave as a denser-than-
air release. A denser-than-air release can lead to higher ground level
releases.
Models handle several transport zones: jet zone, gravitational slumping
zone, transition from gravitational spread to atmospheric dispersion zone, and
finally the atmospheric dispersion zone. Figure 8-1 illustrates these zones.
Worst-case input varies by each zone and thus downwind distance. For example,
high wind speed will probably have little effect in the jet zone, but a major
impact in the atmospheric dispersion zone. Stability class will have little
effect in the jet and gravitational slump zones. As the release dispersion
becomes dominated by atmospheric mixing the stability class becomes more
important.
\
Atmospheric
Dispcmm
Transtnn From GrrouaonaJ
Spread To Atmo«ph«rc
Figure 8-1. Dispersion Zone Schematic
The worst-case input is a function of the impact of concern. The input
conditions which lead to the largest distance downwind of a given
concentration may not be the same as those which lead to a maximum area
impact. Moreover, when determining a maximum downwind distance, the worst-
case input for the zone of low concentration may be different from the worst-
case input for a high concentration.
8-2
-------
There is a need for extensive sensitivity analysis studies to determine
what input parameters are most important for a particular model. The user
should assess each model input according to the results of such sensitivity
analyses. In determining worst-case conditions, normally one or a few of the
input are varied so as to bracket the maximum impact. Listed here are some of
the input and internal assumptions that may have an effect on determining
impacts.
8.1.1 Exit Velocity. Emission Rate, and Jets
If a non-source-term model is being used, varying the exit velocity
should be done only when considering whether some other parameter should also
be changed for self consistency. Increasing an exit velocity means that
either the release diameter must be decreased or the total volume (emission
rate) flowing through the hole must be increased. The emission rate is
related to the exit velocity as shown in Section 4.12.
Jets are defined in terms of the comparison of the exit velocity with
the ambient wind speed. At low ambient wind speeds a jet will have
significant mixing. 'As the ambient wind speed approaches the exit velocity,
the transport of high concentrations is enhanced due to decreased mixing from
velocity differences. Thus, the assumption that worst-case conditions occur
at low wind speeds does not necessarily hold true for jet releases. In fact,
for high velocity releases, higher wind speeds may lead to higher impacts.
When a liquid or gas release results in a gas and/or aerosol jet, some
new concerns are encountered. Assuming the emission rate and exit velocity
are varied to insure mass conservation, increasing the emission rate of a jet
can lead to a decrease in concentration impacts. This effect depends on the
concentration level of concern. An increase in emission rate leads to an
increase in exit velocity. This in turn leads to greater turbulence and
greater mixing in the near-field, causing the higher concentrations to be
dissipated faster. The lower concentrations that exist beyond the jet regime
of the release may reach further downwind.
8-3
-------
Jets are only slightly dependent on the stability class. Only after the
jet portion of a release is complete does the plume become dependent on the
stability class. This means that little, if any differences in near-field
impacts of jet releases are due to varying the stability class.
8.1.2 Release Temperature
The colder the release temperature of the vapor, the more likely that
the release will behave as a dense gas. Higher release temperatures for the
vapor lead to a lower density which will decrease the tendency to behave as a
dense gas. If there is uncertainty in the release temperature, the coldest
temperature should be used. In most of the examples in this document the
release temperature is calculated. In cases where the temperature is not
known or calculated, use the lower temperature of boiling point or ambient
temperature.
8.1.3 Release Diameter
In most source-term models, a hole size or release diameter is
requested. As this parameter changes it can affect both the emission rate and
the exit velocity.
8.1.4 Release Height
For dense gases, an elevated release may fall quickly to the ground.
The greater the density difference between the release and the ambient
temperature, the more likely that the release will fall toward the ground.
8.1.5 Ground Temperature
For liquid releases, the higher the temperature of the surface on which
the liquid falls, the greater the rate of evaporation. Higher ground
temperatures should be used for maximum emission rates.
8-4
-------
For gas releases, the ground temperature is not as significant in its
effect as for liquid releases. In gas releases, the ground temperature
affects both the released material and air in the atmosphere similarly.
Because of this, differences between the release gas and air aren't enhanced
or decreased significantly.
8.1.6 Meteorology
Ambient Temperature
If a source-term model is being used, the ambient temperature can have a
large effect if the species is released as a liquid and the temperature is
varied above or below the boiling point of the liquid released. In that case
the temperature will affect the emission rate and the initial concentration of
a release.
Ambient temperature can also affect the dispersion of a release. If the
ambient temperature is higher than the release temperature, the released plume
will tend to be denser than the ambient air (although this is also dependent
on molecular weight) and the plume may descend. Conversely, if th'e release
temperature is higher than the ambient temperature, the plume may tend to
ascend. Moving the vertical location of a plume can have large effects on
impacts, especially in the near-field where gradients are the largest.
Wind Speed
If a source-term model is being used, an increase in the wind speed can
increase the rate of evaporation of a liquid release. However, the.increased
wind speed also increases the rate of dispersion of the release in the
atmosphere.
8-5
-------
Stability Class
Stability class for complex models should be treated in much the same
way as for Gaussian models. However, since many of the releases are from
elevated sources, the most stable conditions will not always lead to the
largest ground-level impacts. As in the Gaussian model used to simulate an
elevated release, an unstable atmosphere can mix the effluent to the ground at
higher concentrations than does a stable atmosphere.
Unlike Gaussian model simulations, there may be releases in complex
model simulations that are almost independent of the stability class--at least
in the near field. Jets and strongly buoyant (positively and negatively)
releases are driven by momentum turbulence rather than atmospheric turbulence.
Once a release is no longer a jet and is not denser-than-air, it is not
different from other plumes that can be simulated by a Gaussian model. If a
release is not a jet and is at ground level, it will likely be influenced by
the same worst-case meteorological conditions as a Gaussian model.
8.2 SIMULATION MECHANICS
The ADAM and ALOHA models are interactive. If multiple simulations are
needed, they must be run one at a time with a user controlling the input as
the simulation is done. DEGADIS, HGSYSTEM, and SLAB are non-interactive
models. Computer files can be set up that will allow these models to run
multiple simulations automatically.
For the purpose of this discussion, it is assumed that worst-case
meteorological conditions are being sought. It is possible that, if a model
is being used for planning purposes, non-meteorological parameters could be
varied. For example, relief valve hole size and storage pressure can be
varied for the purpose of finding the configuration that will minimize
impacts.
8-6
-------
The DEGADIS model simulates one set of conditions (meteorological and
non-meteorological) per run. If multiple cases are to be run, separate input
files must be created for each simulation. Each output must then be manually
scanned for the desired impact of concern.
The SLAB model can simulate multiple sets of meteorological parameters
in a single run. The results of all the meteorological simulations are given
in a single output for one set of non-meteorological parameters. If non-
meteorological parameters are to be varied, separate input files must be
created for each simulation.
The HGSYSTEM models provide a user-efficient method of running multiple
meteorological and non-meteorological conditions. This attribute is
illustrated in the following example using Scenario 3. In this example only
meteorological parameters are being varied. However, other parameters could
be varied as well.
To identify worst-case meteorological conditions for this scenario, the
HEGADASS model was executed for 14 different combinations of wind speed and
atmospheric stability class. A file containing each meteorological condition
was prepared. Table 8-1 presents an example of such a file. Then a standard
HEGADASS input file was prepared, and the meteorological conditions were
commented out by placing a * in column 1 (Table 8-2). The file containing the
meteorological data was then concatenated into the HEGADASS input file using a
DOS batch file. The resulting input file for stability A and 1.0 m/s wind
speed is shown in Table 8-3. By incorporating the CALL command into the DOS
batch file, multiple model runs can be executed in sequence. Table 8-4
presents the batch file used to perform these screening runs. The batch file
can also be expanded to extract model results for direct importation into a
graphical software package.
Table 8-5 summarizes the results of this analysis. This table presents,
for each meteorological condition, the concentration at a specific distance
(100 meters) and the distance to the LOG. These data were extracted from the
8-7
-------
TABLE 8-1. EXAMPLE OF A METEOROLOGICAL DATA FILE TO IDENTIFY WOF ' ' ,SE IMPACTS
AMBIENT * > DATA BLOCK: AMBIENT CONDITIONS
UO =1.0 * M/S wind speed at height z = J
X
DISP * > DATA BLOCK: DISPERSION DATA
*
PQSTAB = a * Pasquill stability das*
8-8
-------
TABLE 8-2. HEGADASS INPUT FILE TO IDENTIFY WORST CASE METEOROLOGICAL
CONDITIONS
HEGADAS-S standard input file STPOOLNO.HSI=
(case: run started at pool, normal thermodynamics)
TITLE EPA SCENARIO 3 HEGADASS RUN CHLORINE RELEASE
CONTROL
ICNT
ISURF =
0
3
I J- E.11 .L
AIRTEMP =
ZAIRTEMP=
RHPERC =
UO
ZO
TGROUND -
'P
ijt
ZR
PQSTAB =
AVTIMC =
CROSSW =
IHATA
iLJn.±£\
GAS FLOW =
TEMPGAS =
CPGAS
MWGAS
WATGAS =
IT in
>\j LJ
DXFIX
NFIX
XEND
CAMIN
CU
CL
)T
'J-i
PLL
PLHW
21.3
10.
50.
4.47
10.
21.3
.01
C
900.
2
.3139
-34.
34.
70.9
0.
1.
1000.
3000.
0.000001
0.0001
0.00001
1.
.5
/% -------->
* c
* M
* %
* M/S
* M
* C
*
* ~i
*
* M
*
* S
* <
*
* KG/S g
* C
* J/MOLE/C
* KG/KMOLE
* - w
*
*
* M
* f
* M
* KG/M3
* KG/M3
* KG/M3
*
•*• -5
*
* M
* M
* > DATA BLOCK: CONTROL PARAMETERS
*
* output code (isocontours,cloud contents)
* code for surface heat/water transfer
*
* > DATA BLOCK: AMBIENT CONDITIONS
air temperature at height z = ZAIRTEMP
height at which AIRTEMP is given
relative humidity
wind speed at height z = ZO
height at which UO is given
earth's surface temperature
DATA BLOCK: DISPERSION DATA
surface roughness parameter
Pasquill stability class
averaging time for concentration
formula (don't normally change)
> DATA BLOCK: GAS DATA
gas emission rate (excl. water pick-up)
temperature of emitted gas
specific heat of emitted gas
molecular weight of emitted gas
water pick-up by gas (don't norm.change)
DATA BLOCK: CLOUD OUTPUT CONTROL
fixed-size output step length
fixed steps upto distance x = NFIX*DXFIX
x at which calculations are stopped
CA (cone.) at which calcs. are stopped
upper concentration limit
lower concentration limit
> DATA BLOCK: POOL DATA
pool length
pool half-width
8-9
-------
TABLE 8-3. CONCATENATED INPUT FILE FOR SCENARIO 3
* =====—=—„—, HEGADAS-S standard input file STPOOLNO.HSI =——.
(case: run started at pool, normal thermodynamics)
TITLE EPA SCENARIO 3 HEGADASS RUN CHLORINE RELEASE
> DATA BLOCK: CONTROL PARAMETERS
output code (isocontours,cloud contents)
code for surface heat/water transfer
> DATA BLOCK: AMBIENT CONDITIONS
air temperature at height z = ZAIRTEMP
height at which AIRTEMP is given
relative humidity
wind speed at height z - ZO
height at which UO is given
earth's surface temperature
DATA BLOCK: DISPERSION DATA
surface roughness parameter
Pasquill stability class
averaging time for concentration
formula (don't normally change)
DATA BLOCK: GAS DATA
gas emission rate (excl. water pick-up)
temperature of emitted gas
specific heat of emitted gas
molecular weight of emitted gas
water pick-up by gas (don't norm.change)
CLOUD OUTPUT CONTROL
fixed-size output step length
fixed steps upto distance x = NFIX*DXFIX
x at which calculations are stopped
CA (cone.) at which calcs. are stopped
upper concentration limit
lower concentration limit
• DATA BLOCK: POOL DATA
pool length
pool half-width
OUniR.U.Li
ICNT
I SURF =
AMRTTTMT
nrLDitUNi
AIRTEMP =
ZAIRTEMP=
RHPERC =
UO
ZO
TGROUND -
HT CT>
Dior
ZR
PQSTAB =
AVTIMC =
CROSSW =
GASFLOW =
TEMPGAS =
CPGAS
MWGAS
WATGAS =
CLOUD
DXFIX
NFIX
XEND
GAMIN
CU
CL
"D/~M"*T
rOUL
PLL
PLHW
0
3
21.3
10.
50.
1.0
10.
21.3
.01
a
900.
2
.3139
-34.
34.
70.9
0.
1.
1000.
3000.
0.000001
0.0001
0.00001
1.
.5
* O!
*
*
* - - -~^
* C
* M
* %
* M/S
* M
* C
*
* ~>
^
*
* M
*
* S
* <
*
* KG/S g
* C
* J/MOLE/C
* KG/KMOLE
* - w
*
*
* M
* f
* M
* KG/M3
* KG/M3
* KG/M3
*
.A. ~s
*
* M
* M
8-10
-------
TABLE 8-4. DOS BATCH FILE TO IDENTIFY WORST CASE METEOROLOGY FOR CASE 3
rem screening run to identify worst case met
rem there are 14 different combinations of ws and stability
rem in files named metl.dat ... metl5.dat. These are then
rem concatenated to the base model input (scen3s.hsi) to form a
rem new input file (scen3s?). This process is then repeated.
rem
metl.dat scen3sl
met2.dat scen3s2
met3.dat scen3s3
met4.dat scen3s4
met5.dat scen3s5
met6.dat scen3s6
met7.dat scen3s7
met8.dat scen3s8
met9.dat scen3s9
metl0.dat scen3slO
metll.dat scen3sll
metl2.dat scen3s!2
metl3.dat scen3s!3
metl4.dat scen3s!4
rem to extract data from hegadass model run for input to a graphical program
call get2col scen3sl.hsr sen3sl.dat 1 2
rem format on program get2col is model output file name, data output file name
rem and columns to be extracted (distance and concentration).
call hegadass scen3s.hsi
call hegadass scen3s.hsi
call hegadass scen3s.hsi
call hegadass scen3s.hsi
call hegadass scen3s.hsi
call hegadass scen3s.hsi
call hegadass scen3s.hsi
call hegadass scen3s.hsi
call hegadass scen3s.hsi
call hegadass scen3s.hsi
call hegadass scen3s.hsi
call hegadass scen3s.hsi
call hegadass scen3s.hsi
call hegadass scen3s.hsi
1-11
-------
HEGADAS model runs by using the interactive utility program HSP""T and by
reviewing HEGADAS output. The finite release duration opt .o i ir, t.-.e HSPOST
utility was selected because this option corrects the predicted concentrations
for release duration.
-------
TABLE 8-5. IDENTIFICATION OF WORST-CASE METEOROLOGY FOR SCENARIO 3
Case
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Pasquill
Stability
Class
A
A
B
B
C
C
D
D
D
D
E
E
F
F
Wind Speed
(m/s)
1
3
1
3
1
3
1
3
5
10
1
3
1
3
Concentration
at 100 m
(ppm)
891
58.4
2100
119.6
3660
319
5000
825
391
175
4280
1670
3480
2150
Downwind
Distance
to LOG
(m)
591
370
792
487
1224
745
3455
1983
1566
1086
7119
6098
8343
8343
8-13
-------
the determination of worst-case conditions being influenced by : he ,-• 1r'ction
of which output parameter is the most important criterion.
8.3 USING THIS DOCUMENT FOR OFF-SITE CONSEQUENCE ANALYSIS
This document provides guidance on the use of dispersion models,
preparing input, and interpreting the output from dispersion models used for
determining off-site impacts of potential accidental releases. Thus, the
guidance provided in this document is only one part of the full analysis
required to determine off-site impacts.
A comprehensive methodology to determine the impact of a release
scenario (worst-case, most-likely, etc.) comprises the following steps:
1. Use of Process Hazard Analysis techniques like WHAT-IF, HAZOP, FMEA, and
Fault-Tree to identify the release scenario and mode of failure.
2. Use of release rate algorithms to calculate the amount released,
evaporation algorithms to calculate the vapor cloud formation and
atmospheric dispersion models to calculate the transport of the vapor
cloud.
3. Use of Consequence Analysis algorithms to predict the number of people
or sensitive natural resources impacted by the potential vapor cloud.
By their nature the steps above provide a methodology that tends to be
site-specific. The Process Hazard Analysis techniques are to be used for
specific plant operations (e.g. transfer from storage tank to process area)
and result in the identification of one or more release scenarios.
Additionally, failure rate data for the various components in the scenario can
be used to estimate the probability of the event.
The second set of algorithms (release rates, evaporation and dispersion)
tend to be more generic (note there may still be some significant site
specific parameters like surface roughness, meteorology, etc.). However for
most applications (i.e. flat terrain) the standard release rate, evaporation,
and dispersion algorithms discussed in this document can be applied.
8-14
-------
The chird set of algorithms also use site-specific information. To
calculate the number of people impacted one has to determine the population
distribution around the plant. Some of the additional factors to be
considered relate to the number of people who are indoors/outdoors, age
distribution of the vulnerable population, seasonal and day/night population
variations. Additionally, frequency data related to meteorology (e.g. STAR
data providing the distribution of wind speed, wind direction, stability
class) can be used to determine the likelihood of a particular impact.
It is important to emphasize that the above outlined scheme is a general
framework which can be applied at various levels of complexity. In it's
simplest form, a WHAT-IF Process Hazard Analysis followed by dispersion
algorithms can be used to define a vulnerability zone (circle with radius
equal to the downwind distance to the concentration level of concern). This
analysis would provide an approximate number of people impacted. On the other
hand, a full-fledged Fault-Tree analysis with failure mode data, followed by a
site specific atmospheric dispersion and consequence analysis (using site-
specific meteorology and population distribution) can be used to generate risk
contours surrounding the plant. The level of detail required to do the
analysis should be dictated by the impact potential of the release scenario
(which is dependent upon the release size, meteorology, toxicity/flammability
characteristics, and population distribution). It is possible to have a large
plant-site with minimum impact potential (e.g. due to rural location or due to
the nature of chemicals handled) while on the other hand it is possible to
have a very small plant operation that could have significant impact (e.g.
plant is located in an urban area or small facility that handles highly
toxic/flammable chemicals) .
J-15
-------
8-16
-------
SECTION 9
REFERENCES
1. U.S. Environmental Protection Agency, 1986. "Guideline on Air Quality
Models (Revised)." EPA-450/4-78-027R.
2. U.S. Environmental Protection Agency, 1992. "Workbook of Screening
Techniques for Assessing Impacts of Toxic Air Pollutants (Revised) ,"
EPA-454/R-92-024.
3. U.S. Environmental Protection Agency, 1990. "User's Guide to TSCREEN, A
Model for Screening Toxic Air Pollutant Concentrations," EPA-450/4-90-
013.
4. U.S. Environmental Protection Agency, 1989. "User's Guide for the
DEGADIS 2.1 Dense Gas Dispersion Model," EPA-450/4-89-019. (NTIS PB 90-
213893).
5. U.S. Environmental Protection Agency, 1991. "Evaluation of Dense Gas
Simulation Models," EPA-450/4-90-018.
6. U.S. Environmental Protection Agency, 1991. "Guidance on The
Application of Refined Dispersion Models for Air Toxics Releases," EPA-
450/4-91-007.
7. U.S. Environmental Protection Agency, 1993. "Contingency Analysis
Modeling for Superfund Sites and Other Sources," EPA-454/R-93-001.
8. U.S. Environmental Protection Agency. "Control of Accidental Releases
of Ammonia," Prevention Reference Manual: Chemical Specific. Vol. 11.
AEERL, August 1987.
9. U.S. Environmental Protection Agency. "Control of Accidental Releases
of Chlorine," Prevention Reference Manual: Chemical Specific. Vol. 9.
AEERL, August 1987.
10. U.S. Environmental Protection Agency. "Control of Accidental Releases
of Sulfur Dioxide," Prevention Reference Manual: Chemical Specific. Vol
12. AEERL, August 1987.
11. The Condensed Chemical Dictionary. Ninth Edition, 1977,Van Norstrand
Reinhold Co.
9-1
-------
12. U.S. Environmental Protection Agency. "Control of Accidental Releases
of Hydrogen Fluoride," Prevention Reference Manual: :hemical Specific.
Vol. 8. AEERL, August 1987.
13. C. Mullett and P. Raj, September 1990. User's Manual for ADAM. Air
Force No. GL-TR-90-0321(II) , AF Geophysics Laboratory, Hanscom AFB, MA
01731.
14. User's Manual for the ALOHA Model, ALOHA 5.0: Areal Locations of
Hazardous Atmospheres. Hazardous Materials Response Branch, National
Oceanic and Atmospheric Administration, Seattle, and Chemical Emergency
Preparedness and Prevention Office U.S. Environmental Protection Agency,
Washington, D.C.
15. K. McFarlane, A. Prothero, J.S. Puttock, P.T. Roberts and H.W.M Witlox,
November 1990 Technical Reference Manual for HGSYSTEM, Development and
Validation of Atmospheric Dispersion Models for Ideal Gases and Hydrogen
Fluoride, Shell Internationale Research Maatschappig B.V. 1990.
16. Ermak, D.L., 1989. User's Manual for the SLAB model, An Atmospheric
Dispersion Model for Denser-than-Air Releases, Draft, Lawrence Livermore
National Laboratory.
17. Beckerdite, J.M., Powell, D.R., and Adams, E.T., 1983. "Self-Association
of Gases. 2. The Association of Hydrogen Fluoride." J. Chem. Eng. Data,
28, 287-293.
18. Chemical Process Quantitative Risk Analysis, CCPS-AIChE Publication,
1989.
19. U.S. Environmental Protection Agency, 1987. "On-site Meteorological
Program Guidance for Regulatory Modeling Applications," EPA-450/4-87-
013.
20. Guinnup, D.E. and Q.T. Nguyen, 1991. "A Sensitivity Study of the
Modeling Results from Three Dense Gas Dispersion Models in the
Simulation of a Release of Liquified Methan: SLAB, HEGADAS, and
DEGADIS." Seventh Joint Conference on Applications of Air Pollution
Meteorology with AWMA. January 14-18, 1991, New Orleans, LA.
21. Britter, R.E. and J. McQuaid, 1988. "Workbook on the Dispersion of
Dense Gases." HSE Contract Research Report No. 17/1988, Health and
Safety Executive, Sheffield, England.
22. Petersen, R.L., 1989. "Surface Roughness Effects on Heavier-Than-Air Gas
Dispersion." Sixth Joint Conference on Applications of Air Pollution
Meteorology. January 30 - February 3, 1989, Anaheim, CA.
9-2
-------
23. Irwin, J. S., 1979. "A Theoretical Variation of the Wind Profile Power-
law Exponent as a Function of Surface Roughness and Stability."
Atmospheric Environment 13, pp 191-194.
24. Perry's Chemical Engineers' Handbook, 6th ed. edited by Robert H. Perry
and Don W. Green, McGraw-Hill, Inc., New York, New York, 1984.
25. The Properties of Gases & Liquids, 4th ed., Reid, Prausnitz, and Poling,
McGraw-Hill, Inc., New York, New York, 1987.
26. Data Compilation Tables of Properties of Pure Compounds, T.E. Daubert
and R.P. Danner, Design Institute for Physical Property Data, American
Institute of Chemical Engineers, New York, New York, 1985.
27. CRC Handbook of Chemistry and Physics, 71st ed. , edited by David R.
Lide, PhD, CRC Press, Inc. Boca Raton, Florida, 1990.
9-3
-------
-------
APPENDIX A
Relationships Between Selected SI Units
-------
-------
This document provides answers to ail calcuations in SI units (Le
Systeme International d'Unites). This Appendix is to provide some
relationships between the base SI units and derived SI units. The base SI
units used in this document are given in Table A-l. The derived units used in
this document are given in Table A-2.
Table A-l. Base SI Units
QUANTITY
Length
Mass
Time
Temperature
Amount of substance
NAME
meter
kilogram
second
kelvin
kilogram mole
SYMBOL
m
kg
s
°K
kmol
Table A-2. Derived SI Units
QUANTITY
Force
Pressure
Energy, work
NAME
newton
pascal
joule
SYMBOL
N (- kg m / s2)
Pa (- N / m2)
J (- N m)
Two examples of calculations which will be seen in the document and how to
derive the final units are given below.
Example Density of Vapor State Calculation
Pa
= P.M,_ (10132SPa)(28.9kg/kmol)
RTa (8314J/kmol°K)(296.48°K)
= 1.188 Pa kg/J
= 1.188 (N/m2)(kg)/(N m)
= 1.188 kg/m3
A-3
-------
Example Emission Rate Calculation
A0F1
= (3.888x10 -5m2)(0.7347)
[(2.878xl05J/kg)(70.91kg/kinol)(6.951xl05Pa)>|
{ (8314J/kmol°K)(294.30K)2
( 294.3°K \\
\927.13J/kg°Kj
0.3170 m2 Pa (kg/J)
0.3170 m2 (N/m2) (kg/N m)
±
= 0.3170 (kg m/s2) (kg s2/kg m2)2
= 0.3170 kg/s
A-4
-------
APPENDIX B
Triple Point Diagrams, Flash Diagrams, and
Physical Property Data Tables for:
Ammonia;
Chlorine;
Ethylene Oxide;
Hydrogen Chloride;
Hydrogen Fluoride; and
Sulfur Dioxide.
-------
B-2
-------
tfl
•i-i
n
120
110
100
!>()
80
70
60
1 50
40
30
20
10
-10
Solid
Critical
(I32.6°C, 111.31
Triple Point Region
-200
-100
Gas
AMMONIA
Detail of Triple Point Iteglon
T
-0.5
Triple Point
(-77 6-C, 6 0x10-1 aim)
L I i
.110 -90 -70
-50
_|
-30 -10
100 200
Toni|>erature (°C)
300
400
500
Triple Point Diagram for Ammonia
-------
u
a
in
!8
50
100 150 200 250 300 350
Storage Temperature (°C)
Flash Diagram for Ammonia
B-4
-------
PHYSICAL PROPERTIES OF AMMONIA
CAS Registry Number
Chemical Formula
Toxicity Threshold Concentrations
Molecular Weight
Normal Boiling Point
Vapor Density
Liquid Density
Liquid Specific Gravity (H20 = L)
Vapor Specific Gravity (air - 1)
Vapor Pressure Equation
Log Pv - A - B/(T+C)
where: Pv - vapor pressure,
mmHg
T - temperature, °C
A - 7.3605, a constant
B - 926.132, a constant
C - 240.17, a constant
Liquid Viscosity
Liquid Surface Tension
Solubility in Water
Heat Capacity at Constant Volume
(Vapor)
Heat Capacity at Constant Pressure
(Vapor)
Heat Capacity at Constant Pressure
(Liquid)
Latent Heat of Vaporization
Net Heating Value
7664-41-7
NH,
Time (min) Level (ppm)
ERPG-1 60
ERPG-2 60
ERPG-3 60
IDLH 30
STEL 15
25
200
1000
500
35
17.030
239.72 °K
0.708 kg/m3 @ 293 °K, 1 atm
681.38 kg/m3 @ 239.72 °K
0.638 @ 273 "K
0.587
6320.9 mm Hg (8.32 atm) @ 20
°C
0.000246 Pa-s @ 239.72 °K
0.0234 N/m @ 284 °K
233.99 kg/m3 @ 293 °K, 1 atm
1589.55 J/kg °K (§ 273 °K
2093.19 JAg °K @ 273 °K
4061.58 J/kg °K @ 240 °K
1367208 J/kg @ 240 °K
18603850 J/kg
B-5
-------
PHYSICAL PROPERTIES OF AMMONIA (Contiiviea)
Enthalpy Equation
H = A + Bt + Ct2
where: H = enthalpy, J/kg
t = temperature (°K); and
A, B, and C are constants
for liquid and vapor
enthalpies as
follows :
ABC
Liquid * 4672.5 -0.625
Vapor * 5138.333 -7.5
* Value of A depends on the selected
reference state for the material.
The numerical value of A does not
affect release calculations since
only enthalpy differences at various
temperatures are applied to a single
phase. To calculate enthaJj-;
differences between phases , 3.
reference state must be chosen to
calculate A for each phase using the
enthalpy equation.
Additional properties useful in determining other properties, from physical-
property correlations:
Critical Temperature
Critical Pressure
Critical Density
Energy of Molecular Interaction
Effective Molecular Diameter
405.65 °K
11278000 Pa
234.99 kg/m3
276 °K
3.45 Angstroms
B-6
-------
to
Critical Point
(144.2°C, 76.89 aim)
Triple Point Region
-200
-100
Gas
CHLOKINU
Detail of Triple Point Region
T
Tflplt Point
(-100.9'C, l.3»10-2 ilm)
-0.5
-ISO
-too
-so
100 200
Temperature (°C)
300
400
500
Triple Point Diagram for Chlorine
-------
S3
-C
E
-50 0 50 100 150 200 250 300 350
Storage Temperature (°C)
Flash Diagram for Chlorine
B-8
-------
PHYSICAL PROPERTIES OF CHLORINE
CAS Registry Number
Chemical Formula
Toxicity Threshold Concentrations
Molecular Weight
Normal Boiling Point
Vapor Density
Liquid Density
Liquid Specific Gravity (H20 - 1)
Vapor Specific Gravity (air =1)
Vapor Pressure Equation
Log Pv = A - B/(T+C)
where: Pv - vapor pressure, mmHg
T - temperature, °C
A - 6.9379, a constant
B - 861.34, a -constant
C - 246.33, a constant
Liquid Viscosity
Liquid Surface Tension
Enthalpy Equation
H - A + Bt + Ct2
where: H - enthalpy, J/kg;
t — temperature (°K); and
A, B, and C are constants for
liquid and vapor enthalpies as
follows:
ABC
Liquid * 821.3762 0.303672
Vapor * 616.4956 -0.43698
Solubility in Water
Heat Capacity at Constant Volume (Vapor)
Heat Capacity at Constant Pressure
(Vapor)
Heat Capacity at Constant Pressure
(Liquid)
7782-50-5
Cl,
Time (min) Level
(ppm)
ERPG-1 60 1
ERPG-2 60 3
ERPG-3 60 20
IDLH 30 30
STEL 15 1
70.914
239.12 "K
2.949 kg/m3 @ 293 °K, 1 atm
1562.19 kg/m3 @ 239 °K
1.47 @ 273 °K
2.45
5055.8 mm Hg (6.65 atm) @ 20 °C
0.000489 Pa-s @ 239.09 °K
0.0254 N/m @ 243 "K
* Value of A depends on the
selected reference state for the
material. The numerical value
of A does not affect release
calculations since only enthalpy
differences at various
temperatures are applied to a
single phase. To calculate
enthalpy differences between
phases, a reference state must
be chosen to calculate A for
each phase using the enthalpy
equation.
3.744 kg/m3 @ 293 °K, 1 atm
355.81 JAg°K @ 288 °K
481.49 J/kg°K @ 288 °K
927.13 J/kg°K @ 293.12 °K
B-9
-------
PHYSICAL PROPERTIES OF CHLORINE (Contir^e
Latent Heat of Vaporization
Net Heating Value
287775 J/kg @ 239
°K
0 cal/gmole
Additional properties useful in determining other properties
property correlations:
Critical Temperature
Critical Pressure
Critical Density
Energy of Molecular Interaction
Effective Molecular Diameter
from physical
417.15 °K
7710800 Pa
572.98 kg/m3
275 °K
4.12 Angstroms
B-10
-------
H)
O
n-a
Pressure (atm)
-o
o
T)
I-*
n>
ft
(B
OQ
- t 1
_^ o
PI
rt
y
re
w
re
O
o.
n
•Si?
• n •»
J<
L
-------
o
u
cs
50 100 150 200 250 300 350
Storage Temperature (°C)
Flash Diagram for Ethylene Oxide
B-12
-------
PHYSICAL PROPERTIES OF ETHYLENE OXIDE
CAS Registry Number
Chemical Formula
Toxicity Threshold Concentrations
Molecular Weight
formal Boiling Foint
Vapor Density
Liquid Density
Liquid Specific Gravity (H20 - 1)
Vapor Specific Gravity (air =1)
Vapor Pressure Equation -
Ln Pv - A + B/T + CLnT + DTE
where: Pv - vapor pressure, Pa
T - temperature, °K
A - 96.820, a constant
B - -5433.0, a constant
C - -12.517, a constant
D - 0.01608, a constant
E - 1.00, a constant
Liquid Viscosity
Liquid Surface Tension
Solubility in Water
Heat Capacity at Constant Volume (Vapor)
Heat Capacity at Constant Pressure (Vapor)
Heat Capacity at Constant Pressure
(Liquid)
Latent Heat of Vaporization
Net Heating Value
75-21-8
C2H40
Time (min) Level
(ppm)
ERPG-1
ERPG-2
ERPG-3
IDLH 30 800
STEL
LEL - 3%
UEL = 100%
44.053
283.85'K
1.832 kg/m3 @ 293°K, 1 atm
882.67 kg/m3 @ 283.85°K
0.898 @ 273°K
1.52
145810 Pa (1.44 atm) @ 293°K
(valid for T - 160.718K to
469.15eK)
0.000269 Pa-s @ 293.15°K
0.0244 N/m
0.1841 kg/m3
888.96 JAg °K @ 293°K
1077.96 JAg °K @ 293°K
1971.56 JAg °K @ 283.85°K
569000 JAg @ 283.85°K
27648022 JAg
B-13
-------
PHYSICAL PROPERTIES OF ETHYLENE OXIDE (Continued)
Enthalpy Equation
H - A + Bt + Ct2
where: H - enthalpy, J/kg
t - temperature ( °K) ; and
A, B, and C are constants for liquid
and vapor enthalpies as follows:
ABC
Liquid* -25.25 4.73445
Vapor* -225.576 3.716791
* Value of A depends on the
selected reference state for the
material. The numerical value of
A does not affect release
calculations since only enthalpy
differences at various
temperatures are applied to a
single phase. To calculate
enthalpy differences between
phases, a reference state must be
chosen to calculate A for each
phase using the enthalpy
equation.
Additional properties useful .in determining other provperties:.-ifrom..-physical::
property correlations: ;
Critical Temperature
Critical Pressure
Critical Density
Energy of Molecular Interaction
Effective Molecular Diameter
469.15 °K
7194100 Pa
313.99 kg/m3
326 °K
4.35 Angstroms
B-14
-------
to
i
M
Ui
90
80
70
60
I S°
£ 40
-------
=
o
u
SJ
in
E
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
Hydrogen Chloride
-200 -100 0 100 200 300 400 500 600
Storage Temperature (°C)
Flash Diagram for Hydrogen Chloride
B-16
-------
PHYSICAL PROPERTIES OF HYDROGEN CHLORIDE
CAS Registry Number
Chemical Formula
Toxicity Threshold Concentrations
lolecular Weight
formal Boiling Point
Vapor Density
Liquid Density
Liquid Specific Gravity (H^D = 1)
Vapor Specific Gravity (air =1)
Vapor Pressure Equation :
LnPv - A + B/T + (CLftT + DTE)
where : Pv = vapor pressure , Pa
T - temperature, °K
A — 105 . 16 , aj constant
B - -3748.4-, a. constant
C - -15.214, a constant
D - 0.031737,.: a constant
E - 1.00, a constant
Liquid Viscosity
Liquid Surface Tension
Solubility in Water
Heat Capacity at Constant Volume
(Vapor)'
Heat Capacity at Constant Pressure
(Vapor)
Heat Capacity at Constant Pressure
(Liquid)
Latent Heat of Vaporization
Net Heating Value
7647-01-0
HC1
Time (rain) Level (ppm)
ERPG-1 60
ERPG-2 60
ERPG-3 60
IDLH 30
STEL
3
20
100
100
36.461
188.15 °K
1.516 kg/m3 @ 293 °K, 1 atm
1194.20 kg/m3 @ 188.15 °K
0.916 @ 273 °K
1.26
4187527 Pa (41.33 atm) @ 293
(valid for T - 158.97 °K to
°K)
°K
324.65
0.000084 Pa-s @ 293.15 °K
0.0041 N/m @ 293 °K
6.70 kg/m3
571.46 JAg °K <§ 273 °K
799.81 J/kg °K @ 273 °K
1655.85 J/kg °K @ 163 °K
442708 J/kg @ 188 °K
784436 JAg
B-17
-------
PHYSICAL PROPERTIES OF HYDROGEN CHLORIDE (Continued)
Enthalpy Equation
H = A + B.. + Ct2
where: H - enthalpy, J/kg;
t = temperature (°K); and
A, B, and C are constants
for liquid and vapor
enthalpies as follows:
ABC
Liquid* 1297.277 2.468391
Vapor* 798.9364 6.1 x 10'8
* Value of A depends on the selected
reference state for the material.
The numerical value of A does not
affect release calculations since
only enthalpy differences at various
temperatures are applied to a single
phase. To calculate enthalpy
differences between phases, a
reference state must be chosen to
calculate A for each phase using the
enthalpy equation.
Additional properties useful in determining other properties from, physical; .:.
property correlations:
Critical Temperature
Critical Pressure
Critical Density
Energy of Molecular Interaction
Effective Molecular Diameter
324.65 °K
8308700 Pa
450.02 kg/m3
216 °K
6.4 Angstroms
B-18
-------
PHYSICAL PROPERTIES OF 30 WT% HYDROCHLORIC ACID
CAS Regis cry Number
Chemical Formula
Toxicity Threshold Concentrations
Molecular Weight
Normal Boiling Point
Vapor Density
Liquid Density
Liquid Specific Gravity (H20 - 1)
Vapor Specific Gravity (air = 1)
Vapor Pressure Equation
(HC1 over 30% Hydrochloric Acid)
Log Pv - A - B/T
where: Pv - vapor pressure, mmHg
T - temperature, °K
A - 9.8763, a constant
B - 2593, a constant
Vapor Pressure Equation
(Water over 30% Hydrochloric Acid)
Log Pv - A - B/T
where: Pv - vapor pressure, mmHg
T - temperature, °K
A - 9. 00117, a constant
B - 2422, a constant
Liquid Viscositv
Liquid Surface Tension
Solubility in Water
Heat Capacity at Constant Volume
(Vapor)
Heat Caoacity at Constant Pressure
(Vapor)"
Heat Capacity at Constant Pressure
(Liquid)
Latent Heat of Vaporization
Net Heating Value
7647-01-0
HC1
Time (rain) Level (ppm)
ERPG-1 60 3
ERPG-2 60 20
ERPG-3 60 100
IDLH 30 100
STEL
21.24
370.2 "K
1.516 ke/mj @ 298 °K, 1 atm
993.3 kg/m3 @ 370.2 °K
0.985 @ 273 °K
1.26
Vapor Pressures from the equation
represent the partial pressures of
anhydrous HC1 over a 30% hydrochloric
acid solution.
Vapor pressures from the equation
represent the partial pressures of
water over a 30% hydrochloric acid
solution.
0.00048 Pa-s
0.0808 N/m @ 370.2 °K
980.98 J/kg °K @ 273 °K
1372.97 J/kg °K @ 273 °K
3475.26 JAg °K @ 273 °K
2354863 JAg @ 370.2 °K
1346578 JAg
B-19
-------
PHYSICAL PROPERTIES OF 30 WT% HYDROCHLORIC ACJ'
'c.neir.iied)
Enthalpy Equation
H = A + B. + Ct2
where: H - enthalpy, J/kg
t =• temperature (°K); and
A, B, and C are constants
for liquid and vapor
enthalpies as follows:
ABC
Liquid* 3475.256 -8.0 x 10'8
Vapor* 798.9364 6.1 x 10 ~8
* Value of A depends on the selected
reference state for the material.
The numerical value of A does not
affect release calculations since
only enthalpy differences at various
temperatures are applied to a single
phase. To calculate enthalpy
differences between phases, a
reference state must bechosen to
calculate A for each phase using the
enthalpy equation.
Additional properties useful in determining other properties from physical. .
property correlations:
Critical Temperature
Critical Pressure
Critical Density
Energy of Molecular Interaction
Effective Molecular Diameter
572.74 °K
20405355 Pa
366.52 kg/m3
433.37 °K
4.367 Angstroms
B-20
-------
to
70
50
40
S 30
20
10
-10
-200
Crilical Point
(188.2°C, 64.02 aim)
Liquid
Triple Point Region
Gas
HYDROGEN FLUORIDE
2.0
Detail of Triple Point Region
r
0.0
-0.5
-ISO -100 -50
-100
100 200
Temperature (°C)
300
400
500
600
Triple Point Diagram for Hydrogen Fluoride
-------
e
o
a
i_
u.
Hydrogen Fluoride
-30
50 90 130
Storage Temperature (°C)
170
210
Flash Diagram for Hydrogen Fluoride
B-22
-------
PHYSICAL PROPERTIES OF HYDROGEN FLUORIDE
CAS Registry Number
Chemical Formula
Toxicity Threshold Concentrations
Molecular Weight
Normal Boiling Point
Vaoor Density
Liquid Density
Liquid Specific Gravity (H20 - 1)
Vapor Specific Gravity (air = 1)
Vapor Pressure Equation
Log Pv - A - B/(T+C)
where: Pv - vapor pressure,
mmHg
T - temperature, " C
A - 7.6810, a constant
B - 1475.60, a constant
C - 287.88, a constant
Liquid Viscosity
Liquid Surface Tension
Solubility in Water
Heat Capacity at Constant Volume
(Vapor)
Heat Capacity at Constant Pressure
(Vapor)
Heat Capacity at Constant Pressure
(Liquid)
Latent Heat of Vaporization
Net Heating Value
7664-39-3
HF
Time (min)
ERPG-1 60
ERPG-2 60
ERPG-3 60
IDLH 30
STEL 15
Level
(ppm)
5
20
50
30
6
20.006
292.67 °K
0.832 kg/m3 <§ 293 °K, 1 atm
955.18 kg/m3 @ 292.67
1.004 @ 273°K
0.69
773.08 mmHg (1.02 atm) @ 20
°C
0.000256 Pa-s @ 273 °K
0.0088 N/m @ 273 °K
Complete
1041.49 JAg °K @ 293 °K
1455.57 J/kg °K <§ 293 °K
2559.79 J/kg °K @ 293 °K
376440 JAg <§ 293 °K
7579626 J/kg
B-23
-------
PHYSICAL PROPERTIES OF HYDROGEN FLUORIDE (Continued)
Enthalpy Equation:
H = A + Bt + Ct2
where: H — enthalpy, J/kg
t - temperature (°K);
and
A , B , and C are
constants for liquid and
vapor enthalpies as
follows :
ABC
Liquid* 1238.668 4.264261
Vapor* 1456.623 -7.1 x 1(T9
* Value of A depends on the
selected reference state for the
material. The numerical value of A
does not affect release calculations
since only enthalpy differences at
various temperatures are applied to
a single phase. To calculate
enthalpy differences between phases,
a reference state must be chosen to
calculate A for each phase using the
enthalpy equation.
Additional properties useful in determining other properties from physical.
property correlations:
Critical Temperature
Critical Pressure
Critical Density
Energy of Molecular Interaction
Effective Molecular Diameter
461.15 °K
6484800 Pa
289.94 kg/m3
337 °K
3.24 Angstroms
B-24
-------
HYDROGEN FLUORIDE APPARENT MOLECULAR WEIGHT AS A FUNCTION OF TEMPERATURE AND MOLAR
FRACTION CONCENTRATION.
Temp(°C)
/ fi
20
30
40
50
60
70
1.0
51.59
36.22
25.26
21.06
20.17
20.01
0.9
48.58
33.45
23.66
20.65
20.10
20.01
0.8
43.78
30.51
22.50
20.53
20.05
20.01
0.7
39.53
27.20
21.51
20.32
20.01
20.01
0.6
35.47
24.65
20.88
20.18
20.01
20.01
0.5
31.56
22.91
20.53
20.09
20.01
20.01
0.4
27.03
21.65
20.36
20.02
20.01
20.01
0.3
23.78
20.68
20.15
20.01
20.01
20.01
0.2
21.53
20.37
20.02
20.01
20.01
20.01
0.1
20.32
20.03
20.01
20.01
20.01
20.01
B-25
-------
to
01
-------
u
a
-50 0 50 100 150 200 250 300 350
Storage Temperature (°C)
Flash Diagram for Sulfur Dioxide
B-27
-------
PHYSICAL PROPERTIES OF SULFUR DIOXIDE
CAS Registry Number
Chemical Formula
Toxicity Threshold Concentrations
Molecular Weight
Normal Boiling Point
Vapor Density
Liquid Density
Liquid Specific Gravity (H20 = 1)
Vapor Specific Gravity (air =1)
Vapor Pressure Equation
Log Pv - A - B/(T+C)
where: Pv = vapor pressure, mmHg
T — temperature , ° C
A - 7.2823, a constant
B - 999.900, a constant
C - 237.190, a constant
Liquid Viscosity
Liquid Surface Tension
Solubility in Water
ieat Capacity at Constant Volume
(Vapor)
tieat Capacity at Constant Pressure
(Vapor)
tteat Capacity at Constant Pressure
(Liquid)
Latent Heat of Vaporization
Net Heating Value
Enthalpy Equation
H - A + B. + Ct2
where: H » enthalpy, J/kg;
t - temperature (°K); and
A, B, and C are constants
for liquid and vapor
enthalpies as follows:
ABC
Liquid* -110.556 3.0556
Vapor* 1566.667 -2.00
7446-09-5
S02
Time ('rain) Level (ppm)
ERPG-1 60 0.3
ERPG-2 60 3
ERPG-3 60 15
IDLH 30 100
STEL 15 5
64.06
263.13 "K
2.664 kg/m3 @ 293 °K, 1 atm
1460.42 kg/m3 @ 263.13 "K
1.436 <§ 273°K
2.21
2480.3 mmHg (3.26 atm) @ 20"C
0.000431 Pa-s @ 263.13 °K, 0.97 atm
0.0286 N/m @ 263 °K
13.875 kg/m3 <§ 293 °K, 1 atm
493.77 J/kg °K @ 298 °K
623.09 JAg °K @ 298 eK
1386.32 JAg °K @ 293 °K
388747 JAg @ 263 °K
0 JAg
* Value of A depends on the selected
reference state for the material.
The numerical value of A does not
affect release calculations since
only enthalpy differences at various
temperatures are applied to a single
phase. To claculate enthalpy
differences between phases , a
reference state must be chosen to
calculate A for each phase using the
enthalpy equation.
B-28
-------
PHYSICAL PROPERTIES OF SULFUR DIOXIDE (Continued)
Additional properties useful in determining other properties from, physical
property correlations:
Critical Temperature
Critical Pressure
Critical Density
Energy of Molecular Interaction
Effective Molecular Diameter
430.75 °K
7884100 Pa
527.07 kg/m3
303 °K
4.29 Angstroms
B-29
-------
B-30
-------
TECHNICAL REPORT DATA
(Please read Instructions on reverse oerore completing)
•L.PORT NO 2
SPA-454/R-93-002
-:TLE AND SUBTITLE
Guidance on the Application of Refined Dispersion
Models for Hazardous/Toxic Air Releases
AUTHOR (S)
PERFORMING ORGANIZATION NAME AND ADDRESS
Radian Corporation
8501 N. Mopac Blvd.
Austin, TX 78759
'. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards, TSD
Research Triangle Park, NC 27711
' j^CIPILNT'S ACCESSION NO
5 4EPORT DATE
May 1993
6 PERFORMING ORGANIZATION CODE
8 PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO
11. CONTRACT/GRANT NO.
EPA Contract No. 68-D00125
WA51
13. TYPE OF REPORT AND PERIOD COVERED
Final Report
14 SPONSORING AGENCY CODE
i SUPPLEMENTARY NOTES
This document revises EPA-450/4-91-007
Technical Representative: Jawad S. Touma
This document provides general guidance on characterizing hazardous air pollutant
•eleases and shows how to apply appropriate dispersion models. The document: 1) helps
letermine likely or reasonable storage conditions for specific conditions for which a
•elease might occur; 2) helps determine release classes; 3) defines the steps to be
aken when determining if a release should be considered a dense gas release; 3)
lefines the methods used to determine the input variables used by refined models in the
mblic domain; 4) points out the implications and effects of various choices for input
.nformation; 5) shows by example, the calculation of the input variables used by the
models; 6) describes the output available from these models; and 7) discusses how to
letermine the input that gives the worst-case impact conditions. Because many
:hemicals may form dense gas clouds upon release, and refined dispersion models that
:an simulate these releases are complex, particular attention is paid to models that
:an address these types of release.
------- |