Mitigating the Impacts of Energy Facilities-1981

y OCM LIBRARY
APR 14
Mitigating the Impg;j jQ#,9nerWrA
Facilities: A Local Air Quality
Program for the
Wilmington, North Carolina Area
Rogers, Golden & Halpern
1427 Vine Street
Philadelphia, PA 19102
Engineers for Energy and the Environment
3 Brower Court
Marlton, NJ 08053
DCM COPY DCM COPY
lease do not remove!!!!!
Division of Coastal Management Copy
CEIP REPORT NO. 6
JANUARY 1981

To order:
Residents of North Carolina may
receive a single copy of a
publication free upon request.
Non-residents may purchase
publications for the prices
listed. Because of the
production costs involved, some
of the publications: carry a
minimal charge regardless of
residency. Prices for these are
indicated in the price list as
being "for all requests".
When ordering publications
please provide the publication
number and title and enclose a
check made payable to DNRCD. For
a complete list of CEIP-
publications - or to place an
order - contact:
Coastal Energy Impact Program
Office of Coastal Management
N.C. Department of Natural
Resources and Community
Development
Box 27687
Raleigh, NC 27611
Series Edited by James F. Smith
Cover Design by Jill Miller

MITIGATING THE IMPACTS OF
BY
Rogers, Golden & Halpern
1427 Vine Street
Philadelphia, PA 19102
Engineers for Energy and the Environment
.3 Brower Court
Marlton, NJ 08053
The preparation of this report was financed through a Coastal Energy
Impact Program grant provided by the North Carolina Coastal Management
Program, through funds provided by the Coastal Zone Management Act of
1972, as amended, which is administered by the Office of Coastal Zone
Management, National Oceanic and Atmospheric Administration. This CEIP
grant was part of NOAA grant NA-79-AA-D-CZ097.
Project No. 79-04
Contract No. C-1173
January 1981

TABLE OF CONTENTS
1. . Summary
2. Criteria for Development of Emissions Inventory
rl
3. Recommended Approach for Initial Compilation of Emissions Inventory 11
4. Regional Air Quality Modeling Recommendation 14
Introduction 14
Model characteristics required for successful application 17
Recent model applications 20
Recommended modeling approach 22
5. Alternative Air Quality Planning Structures for the
Cape Fear Council of Governments 31
The need for air quality planning 31
Three alternatives for air quality planning 32
6. An Estimate of the OCS-related Facilities Likely to Locate
in the Region and Their Requirements 38
The history of OCS activity in the South Atlantic 39
Lease sale 43 39
Lease sale 56 40
Description of the requirements of OCS-related facilities
likely to locate in the region
General characteristics of temporary support bases for
OCS exploration
Supply base experience in the South Atlantic
Bibliography 46
Appendix A. Abstracts of Air Quality Dispersion Models
Appendix B. Descriptions of OCS-related Onshore Facilities
Appendix C. The Status of Emissions Inventories in the Cape Fear
Region
41
42
44

LIST OF TABLES AND FIGURES
TABLES
3.1 Sample of Major Stationary Combustion Source Questionnaire
4.1 Pertinent Characteristics of Recent Regional Model Applications
4.2 Candidate Air Quality Dispersion Models
4.3 Air Quality Model Comparisons
6.1 Land Requirements for a Temporary Supply Base
FIGURES
4.1 Location of Industrial Facilities Recently Applying for PSD Permits
6.1 Geologic Features of the South Atlantic Region
6.2 Lease Sale 43 Leased Tracts and Wells
6.3 Tracts Proposed for Lease in Lease Sale 56

1. Summary
Energy development in the coastal zone of North Carolina could change
dramatically over the next ten years depending on the outcome of offshore oil
exploration. It is already clear that outer continental shelf (OCS) oil and gas
exploration and peat mining and use could dominate regional impacts. At the same
time other industrial and recreational facilities cumulatively can have major
environmental, recreational, health, safety and socioeconomic impacts.
This project is intended to help design a system of air quality analysis and
review which will serve to anticipate and work to mitigate energy facility -related
air quality impacts in the Wilmington Area. It was funded in part by the Coastal
Energy Impact Program (CEIP), which was establish by Congress in 1976 to help
states and local communities deal with the social, economic, and environmental
requirements of coastal energy activity. Specific federal regulations implementing
the program can be found in the Code of Federal Regulations (15 CFR, Part 931).
From a regional point of view, the capacity to plan for air quality is a
valuable ingredient in defining strategies of regional development. A well
understood air emissions inventory and its skillful utilization in air quality modeling
can be applied to the region to identify areas suitable for industrial development.
A synthesis of this information with demographic, economic, and infrastructure
data, can identify the optimum intensity, mix, and location of new industrial
development. Currently, these decisions are made by industrial interests with
public review having little influence.
Although the state DNR&CD reviews permit applications for new or modified
air emissions point sources, its focus is statewide. It is understandable that this
-1-

broad focus may not reflect local, social, economic, or physical conditions at the
level of detail needed to achieve regional planning objectives. Without regional
land use and economic policies based in part on solid air quality planning and
technical support, current trends in industrial development may lead to a lack of
job diversification and a consumption of the regional air resource so as to preclude
future industrial development including onshore facilities needed to support OCS
oil and gas activities.
Rogers, Golden & Halpern and Engineers for Energy and the Environment
have reviewed the status of air quality planning in the Cape Fear region in order to
recommend to the Cape Fear Council of Governments (COG) a planning approach
to air quality management. Ideally, this approach should be able to evaluate
alternative emissions scenarios in order to select the preferred path for economic
development of the region and for the concurrent protection of health and property
from deleterious impacts of air pollution.
The focus of our work has been to determine what could be done by Cape
Fear COG to implement its own air quality management component for regional
planning.
The air quality program would have two purposes. The first would be to apply
air quality modeling techniques, using updated point and area emission inventories,
to the identification of those areas in the Cape Fear region where new industrial
growth could occur with respect to specified emissions, and, correspondingly, to
locate areas already impacted by existing industrial growth which new industrial
growth should avoid. The second purpose of the program is to examine PSD
applications, as they are made to DNR&CD, for modeling assumptions and data
inputs to assure that PSD modeling reflects local conditions as accurately as
possible. This non -regulatory review function could be quite useful to DNR&CD in
its regulatory review and modeling of PSD applications and would be based on a
close interagency relationship at the technical level.
A high level of commitment to air quality planning by Cape Fear COG will
require the agency to hire a meteorologist with air quality monitoring experience
to carry out a four step procedure. Chapters two, three, and four contain our
detailed recommendations concerning this procedure which is summarized below.
-2-

Establish criteria for the form and content of a Cape Fear Air Emissions Inventory
Chapter 2 states the purpose and expected use of an air emissions inventory
for the Cape Fear Region. It discusses the content and spatial resolution that
such an inventory should have. It recommends inventory format compatibil-
ity with inventories used by other agencies, particularly the state Depart-
ment of Natural Resources and Community Development (DNR&CD).
Compile the Cape Fear Air Emissions Inventory.
Chapter 3 describes the planning and execution stages of compiling an air
emissions inventory including recommendations concerning institutional rela-
tionships between the Cape Fear COG and DNR&CD.
Review available air quality models for use in the Cape Fear region.
The first part of Chapter 4 establishes the need for air quality modeling, sets
forth seven critical criteria for the selection of the most appropriate model
and then describes various air quality models currently in use in air emission ,
determination and regulation.
Select an air quality model for use in the Cape Fear region.
The second part of Chapter.4 evaluates 13 air quality models with respect to
the seven criteria established earlier. Of all the models reviewed, the RAM
model is selected as the method of choice. Certain modifications to RAM
are recommended to bring it closer to satisfying all review criteria. The
particular data needs and application of the RAM model are discussed.
In the case that Cape Fear COG does not wish to commit the resources
needed to establish an air quality planning capability in house, there are several
other less expensive options available to it that will provide the technical
capability necessary to protect the region's interests by screening PSD applications
as they are made for modeling assumptions and data inputs and by performing
special industrial planning studies based on air quality modeling. In Chapter 5,
-3-

three alternative approaches to achieving an air quality planning objective are
presented and discussed, including the steps, schedule, and budget needed to
implement each approach. Recent industrial growth in the Cape Fear region is
used to illustrate the need for air quality planning.
The three options for Cape Fear COG to obtain technical input to air quality
planning efforts as identified in Chapter 5 are:
1. Create a Permanent Meteorological Staff
Cost
o salary for meteorologist/air quality specialist:. $28,000-$35,000/
year
0 computer support: $80,000-100,000/year
Advantage
o in-house expertise to create emission inventory, run air quality
models for planning purposes and review PSD applications.
Disadvantage
o technical capacity with no enforcement authority
• expensive
2. Retain an Air Quality Consultant
Cost
o retainer arrangement costs between $10,000 to $20,000/year with
special studies extra
Advantage
0 only a small capital expenditure needed to have' expert services on
call
o retainer would insure that consultant remained current of air quality
situation in region
-4-

Disadvantage
o does not provide in-house capability
3. Obtain Consultant Services When Needed
Cost
o depends on frequency and size of authorized study
Advantage
o no expenditure is necessary unless study is needed
Disadvantages
o quick response to need for study is not possible
o consultant must rely on State data base
The last chapter of this report is concerned with the likely onshore impacts
of outer continental shelf (OCS) oil and gas extraction. Chapter 5 reviews the OCS
oil and gas exploration activity in the South Atlantic and discusses the likelihood
that OCS-related facilities will locate in the Cape Fear region. Temporary OCS
support bases are identified as the only likely facility that may locate in the
region. The land, waterfront, and labor requirements of support bases are reviewed
and the particular support base experiences in the South Atlantic are described.
-5-

2. Criteria for the Development of an Emissions Inventory
The following discussion sets forth criteria for the content and form of an
emissions inventory for the Cape Fear area. The criteria are designed to result in
a data base amenable to the desired end use, which is atmospheric dispersion
modeling. Many of the requirements presented are of a generic nature while others
are only applicable to this specific region.
PURPOSE OF INVENTORY
The purpose of creating and maintaining an emissions inventory is to define,
in detail, the air contaminant emissions within the Cape Fear COG area as well as
those important emissions originating outside the area which may have a signifi-
cant air quality impact within the area.
EXPECTED USE OF THE INVENTORY
The emissions inventory for the Cape Fear area can be used to define
emissions densities . and other characteristics vital to the conduct of proper
planning for siting of new industrial and energy -related sources of air contami-
nants. In order to be used for planning purposes the inventory should satisfy the
following criteria directly related to end -use:
o The emissions inventory must treat the particular pollutants which will
likely be emitted by new energy -related facilities.
IM

o The information obtained must be of sufficient spatial and temporal
resolution to permit siting decisions through use of air quality dispersion
modeling.
o The inventory information must be installed and maintained on automa-
tic data processing equipment because of its projected use as input to
dispersion model(s).
In addition to planning, other uses are normally made of emissions data by
enforcement and/or control agencies (i.e., DNR&CD). These uses include design of
monitoring programs, development of control programs, development of enforce-
ment strategies, and development of strategies for input to the State Implementa-
tion Plan. Since, in the State of North Carolina, these uses are legitimately the
function of the DNR&CD or a local air pollution control board authorized and
approved by the Environmental Management Commission, they will not be specifi-
cally addressed in the design of the emissions inventory for the Cape Fear COG.
Content and Spatial Resolution of Inventory
In light of the fact that the inventory will be used solely to provide input for.
air quality modeling studies of the COG Region, the following general inventory
details for major point sources is envisioned:
1) Name of source
2) Name of source owner (Corporation, Company or Individual)
3) Mailing address of source
4) Name and telephone number of contact at source
5) Year and month of information receipt or update
6) For each emission point at the source:
-7-

a) emission rate for each pollutant
b) release height for each pollutant
c) gas temperature
d) volumetric gas flow rate
e) UTM coordinates of emission point
f) temporal variation in source parameters
Suggested methods for obtaining the above information are discussed in
Chapter 3.
Of course, the large total number of point sources in the region makes it
impractical to obtain detailed information on each. The solution to this problem
will rest with the determination of an emissions size cut-off. Those point sources
whose emissions are below the cut-off size will be combined into several "area
sources." These small point sources will be comprised mainly of space heating
sources of SO2, NO2, and particulates. In most cases, their emissions must be
described through indirect methods such as apportionment of regional fuel con-
sumption on the basis of population density, etc.
Additional area source data will be required to describe emissions from
mobile sources. As with space heating emissions, these emissions must also be
estimated through indirect methods such as statistics on vehicle miles traveled,
vehicle mix, and airport and marine operations. The general content of the
inventory for area sources after apportionment to grid zones is expected to be
similar to the following:
1) Area source identification number;
2) Coordinate boundaries;
3) Year and month of data receipt or update;
4) Annual emissions density - each pollutant;
5) Temporal emissions variation
The characteristics of the emitters included in the area sources as well as the
characteristics of certain of the pollutants generally encountered (especially
hydrocarbons) make generation of an accurate area source inventory quite diffi-
cult. For example, the majority of hydrocarbon emissions cannot be accounted for
-8-

by questionnaire methods. Only a relatively small portion of these emissions are
from stationary, discrete combustion sources. The majority arise from bulk
storage evaporation, motor vehicle hot soak, cold start and running emissions,
filling operations, spray painting and solvent use, natural sources, and numerous
other small diffuse sources.
Development of detailed criteria for determination of which pollutants to
address in the inventory must proceed from both the standpoint of the type of new
industry expected to locate in the region and the standpoint of existing and
anticipated federal regulations in terms of ambient standards. The only pollutants
which should be included are those which are expected to be both emitted in the
Cape Fear area and regulated. These are:
1)
Carbon Monoxide
2)
Sulfur Dioxide
3)
Particulates
4)
Nitrogen Oxides
5)
Hydrocarbons
6)
Lead
The inclusion of pollutants (2) through (6) is for obvious reasons. The
inclusion of carbon monoxide at this time is due to uncertainty regarding the
pending implementation of the Set II PSD regulations. Omission of oxidants from
the list is due to the fact that, while they are a Criteria Pollutant, they are not
emitted per se but are rather formed in the atmosphere in a photochemical
reaction involving the hydrocarbons. Still other pollutants such as trace elements
and radionuclides may well be emitted by the energy industry but these are not
expected to have air quality standards attached for at least several years. In
addition, trace emissions are quite difficult to quantify without testing every major
combustion source and rigorously and repeatedly performing expensive analyses on
coal supplied to coal combustion sources.
Of the pollutants recommended for inclusion, lead will probably be the most
difficult to quantify followed by hydrocarbons and nitrogen oxides. SO2 and CO
emissions will likely be the most readily determined.
10

Emissions Data Base Format
The emissions data base developed from the emissions survey and from
engineering estimates of emissions will contain vast amounts of data. In addition,
the eventual use of the data will be as input to a regional dispersion model designed
to project air quality impacts of various planning alternatives. Both the volume of
data involved as well as the eventual input to a computerized model dictate that
the emissions data base be installed on automatic data processing equipment.
The format of information storage should be designed to facilitate informa-
tion interchange with other agencies. This is required since it is expected that
DNR&CD will provide the major input to the COG inventory. The COG inventory
will also require input from DNR&CD or other agencies regarding major sources
located outside of COG's area of jurisdiction. Exact determination of final format
as well as content, therefore, will depend strongly upon the system employed by
DNR&CD. It is our understanding, at this point, that the DNR&CD system is still
in development but should be compatible with EPA's National Emissions Data
System (NEDS). The air quality model should be designed to use this type of
information.
-10-

3. Recommended Approach for Initial Compilation of an Emissions Inventory
The approach for compilation of the emissions inventory may be conveniently
divided into two phases - Planning and Execution.
PLANNING
The elements of planning which remain before data gathering can begin are
strongly dependent on procuring certain vital pieces of information from DNR&CD.
In light of the content of the general discussions presented in Chapter 2, an
important step for the Cape Fear COG is to establish communications at a
technical level with the DNR&CD Wilmington Regional Office. In order to avoid
massive duplication of effort and concomitant excessive costs to COG, a good
working relationship will have to be developed and maintained with the this agency.
The first COG objective should be to obtain the following information:
1) Current and complete emissions inventory for the Cape Fear Region,
and, for major sources (greater than 100 t/yr emissions), information for
up to 20 miles outside of the geographical boundaries of the region;
2) A detailed description of the policies, procedures and schedules used by
DNR&CD to acquire the current emissions data. This is intended to
insure consistency of methods between the COG and DNR&CD and to
enable evaluation of adequacy of the existing DNR&CD inventory;
-11-

3) Copies of all recent PSD applications within or nearby the region. The
information contained in these applications will include measured ambi-
ent air quality and regional dispersion modeling results.
Once the above -described information is in -hand, initialization and manage-
ment of the data base can be accomplished. This initialization phase will address
items such as which sources will get what type of questionnaire, how will follow-
ups be made, what grid system will be used and in which card column or what card
will a certain piece of information be punched. The culmination of the planning
phase will be the execution of the data collection program.
EXECUTION
Execution of the detailed plan developed as indicated above will probably
involve several distinct steps. It is expected that the steps will approximate the
following:
1) Obtain field data through questionnaires or -engineering surveys;
2) Obtain already -collected data from DNR&CD;
3) Make estimates for mobile sources;
4) Make estimates for small point and area sources;
5) Design, code and implement computer programs to store, retrieve,
analyze, and update the emissions data base; and
6) Transfer field and estimated data from hard copy to the computerized
file system.
The questionnaires used to elicit information from major sources will have to
be carefully designed, striking a balance between information desired and the
willingness of industry to obtain and provide that information. Depending on the
status of the existing DNR&CD inventory, it is possible that no questionnaires will
-12-

ever be mailed by COG. If questionnaires are required, a sample point source
questionnaire is provided in Table 3.1.
THE STATUS OF AIR EMISSIONS INVENTORIES
We have reviewed the air emissions inventory for the Cape Fear Region. The
information reviewed appears to be generally adequate and could be used in
dispersion modeling startup and checkout procedures. However, no area source
emissions data were found in the documents reviewed. Furthermore, there appear
to be some omissions in the point source inventory for the region. A detailed
presentation of findings and recommendations is made in Appendix C.
-13-

TABLE 3.1
SAMPLE OF MAJOR STATIONARY COMBUSTION SOURCE QUESTIONAIRV
Firm Name:
Person to Contact Regarding This Questionnaire: Title: Phone:
Ms(ling Address:
Plant Addras r.
Nature of Eualnes. (Products):
Section 1 • Fuel Use for Generation of Heat, Steam, and Power
Norm.l Operating Schedule: Hours Per Day:
SM.-I and/or Peak Operating Periods (Specify)!
Estimate of Percent of Total Fuel Consumed to Provide Space Heat:
Days Per Month:
Week. Par Year:
A
E
C
D E F G H
I J E L
FUEL IISE DATA
STACK DATA
Source
No.
S(ta of
Unit Input
106 Etu/Yr
Type of
Unit
Installation
Data
Type of
Fuel
Amc nt
Consumed
Per es
Callous
Heatin ue
of
tus
Percent
Sulfur
Percent
Ash
Height'
(ft)
Diameter
(ft)
Temgersture
( F)
Exit Velocit
(ft/see)
\ Sectton 11 • Emissions Information
N N 0 P R
S T I U
I V I W li
ANNUAL EMISSIONS
METHODS OF EMISSIONS
ESTIMATION
source
No,
SO
NO
Part.
Lead
WIG
CD
SOT
Code
NOx
Code
Part.
Code
lead
Code
VOC
Code
CO
Code
*Adapted from EPA APTD 1155 June 1972 '
The above Questionnaire would be transmitted to the Firm under cover of a letter explaining Its purpose, providing detailed Instructions for answering each question
and providing an engineering contact at COC who can lend assistance to the Respondsnt.

4. Regional Air Quality Modeling Recommendation
INTRODUCTION
Background and Need for Modeling
The Clean Air Act as amended, requires that the State assure the attainment
and maintenance of the National Ambient Air Quality Standards (NAAQS) within
its borders. It also requires that "significant deterioration" of air quality from its
present status be prevented. These functions are carried out through a plan
developed by each state and approved by the Environmental Protection Agency
(EPA). The plans are known as State Implementation Plans (SIP).
In the Cape Fear region of North Carolina, state -operated monitoring
networks that measure ambient pollutant concentrations have demonstrated that
all pollutants are either in compliance with the NAAQS or that insufficient data
exist to make a compliance or non-compliance determination. The fact that
regional air quality is relatively good has also been confirmed by a measurements
program operated by private industry and not related to the state -run network.
These measurements were taken by the Brunswick Energy Company. Because of
the attainment status of the region, regulatory emphasis and attention is placed on
preventing significant deterioration rather than on attainment of the NAAQS.
This emphasis is important since it implies a numerical limit on the amount
of pollution ("increment") as measured at ground level which can be generated by
sources built or modified subsequent to a given date ("baseline" date). These
Prevention of Significant Deterioration (PSD) limits on increases in ambient sulfur
dioxide or total suspended particulates concentrations are fairly restrictive
-14-

amounting to less than 40% of the NAAQS for a Class II area such as the Cape Fear
region. Every new pollution source or modification to a source which increases
that source's emissions placed in operation since the baseline date consumes some
portion of this increment. Since the increment is defined in terms of ground -level
concentrations and sources are not homogeneously distibuted over the region, the
amount of increment remaining varies in a spatial sense over the region. Also,
since the increments are regulated for more than one averaging period, the
percentage of increment remaining may be different for the different averaging
periods for the same pollutant. When the available increment has been totally
consumed by new or modified sources, no further development which would
increase regional emissions will be permitted. For this reason, it is necessary and
prudent to effectively manage the remaining increment to assure orderly develop-
ment commensurate with the socioeconomic goals of the region.
Management of this increment as well as assurance of long-term maintenance
of the NAAQS can only be accomplished through use of simulation modeling. It is
only through modeling that the vital questions of air quality impact of as -yet
unbuilt facilities can be addressed. It is also only through modeling that the
desirability of alternative long-term growth scenarios can be assessed. Regional
air quality modeling thus becomes one of the indispensible planning tools required
to resolve the potential conflicts between economic, energy, and clean air needs
and goals of the region.
Model Selection Considerations
Air quality modeling may be defined as the application of a mathematical
description of the physical processes involved in transport, dispersion, and chemical
and physical modifications of air pollutants to a particular region for the purpose
of assessing the ground -level pollutant concentrations to which a population may be
exposed. Numerous sets of equations have been programmed for use in digital
computers --each set normally referred to as a "model." Since there are wide areas
of disagreements among scientists as to the correct mathematical description of
the physical processes involved (and even of the physical processes themselves)
there is, naturally, an abundance of models all purporting to describe the same
physical phenomenon.
-15-

This is compounded by the number of physical and chemical processes to be
described (e.g., plume rise, plume dispersion, material deposition, photochemical
reactions, flow around obstacles, etc.) and the model selection becomes very wide
indeed.
Successful choice of a model for application to a particular situation requires
a detailed knowledge of emissions characteristics, pollutant characteristics, mete-
orological characteristics and the behavior of available models in the defined
situation. Successful application of the chosen model requires experienced
professional judgment confirmed, in the final analysis, by a validation study.
Considering the importance of the modeling results to the region, it is imperative
that appropriate models be chosen, that proper and adequate input data be
prepared, that the model application be properly made, that the results be
interpreted with full knowledge of the limitations inherent to the process, and that
the entire modeling program be overseen by experienced and competent personnel.
Since the scope of the contract under which this report is written provides
only for a recommendation regarding the particular model which is appropriate for
use in the Cape Fear region and not for support of model implementation and
application, it is crucial that proper staffing internal to the COG be provided for
this purpose. As an aid in accomplishing this purpose, Chapter 5 of this report
presents a summary of personnel qualifications for a position within the COG out
of which the modeling work would be performed. (See Option 1 in Chapter 5.)
The remainder of this chapter is devoted to the selection of the particular air
quality model which will meet the anticipated requirements of the Cape Fear COG.
Several models are discussed and evaluated and one is recommended for purchase,
modification and use. It should be noted, however, that dispersion modeling is a
continually evolving field. The normal cycle is that models are developed, come
into general use, and are eventually discarded as "improved" models become
available. The model recommended in this report should be continually reviewed
and updated or replaced when it becomes obsolete. Competent internal staff will
be necessary to accomplish this.
-16-

MODEL -CHARACTERISTICS REQUIRED FOR SUCCESSFUL APPLICATION
Criteria are established for model selection in this section. They are based
upon an understanding that the modeling to be performed by the COG will be in
support of regional scale planning activities. This assumption is important as it
defines the spatial scale of the required model to be regional rather than local.
Even though the basic model will be regional in scope, the possible need for
evaluation of local impacts of sources will be addressed in the next section where
local scale models for vehicular emissions (APRAC) and point sources (ISC) will be
described.
Establishment of Selection Criteria
There exist a number of basic model characteristics which must be selected
or made compatible with the situation at -hand. The desired characteristics are
expressed in this Section as criteria for model selection. It may not be possible to
simultaneously satisfy all criteria with any 'off the shelf" model. For instance,
although not of particular concern in the Cape Fear region, the criteria of EPA
acceptability frequently runs counter to the need to address plume impact on
elevated terrain. Also, the need to use available meteorological data runs counter
to the desirability of addressing spatial changes in the wind field. For these
reasons, the selection of a model always represents a compromise between the
ideal and reality and frequently represents a compromise between equally practical
and realistic, but opposed criteria. Thus, the criteria that follow do not represent
the best that science may have to offer since the best would require very detailed
input data which are not available. Neither do they lead to the selection of a
"cookbook" model since all of the needs of the planning function cannot be satisfied
through use of a single 'off -the -shelf" model. Rather, the attempt is made to
specify an approach which addresses the regional problem at hand, is easily
modified or expanded, will operate within the constraints of existing meteorologi-
cal data, and is practical to apply.
Criteria No. 1: Spatial Characteristics. The modeling package must be cap-
able of addressing the regional scale, on the order to tens of kilometers. It must
allow for the specific geographic location of point and area sources throughout the
-17-

region at arbitrary points in the model coordinate system. It must be capable of
predicting ground -level pollutant concentrations at any desired point within the
region.
Criteria No. 2: Temporal Characteristics. The modeling package must ad-
dress the time scales associated with the averaging periods for the NAAQS. These
are 1 hour, 3 hours, 8 hours, 24 hours, 90 days and the annual period. The model
package must be capable of evaluating, at a minimum, seasonal variations in
emission rates and, preferably, be capable of evaluating weekday -weekend and day -
night changes.
Criteria No. 3: Pollutant Types. There are two basic categories of pollu-
tants --those which can be treated as inert gases and those which are reactive or
which may be depleted by deposition. The ideal, completely non -reactive, non -
depleting pollutant does not exist, but for the space and time scales and purposes
involved in this study, SO2, TSP, CO (carbon monoxide), Pb (lead), and NO2
(nitrogen dioxide) may be considered to have these conservative properties. Ozone
cannot. The modeling package should treat the "inert" pollutants only. Photo-
chemical reactions are not recommended to be treated at this time because of 1)
lack of general scientific agreement on proper and practical methods, 2) lack of an
EPA -approved photochemical model for general use, and 3) the fact that available
oxidant monitoring data shows that ambient levels of this pollutant are well within
attainment levels.
Criteria No. 4: Source Handling Capability. The chosen. model must be cap-
able of addressing the types of sources found in the Cape Fear region. It must be
capable of handling sufficient numbers of non-colocated point and areas sources to
realistically describe the emissions characteristics of the region. Each point
source should be able to be described by its own unique location, height, diameter,
flue gas flow rate or exit velocity, flue gas temperature and emission rate
including temporal variations. Each area source should be described by arbitrary
location, size, emission rate.and effective release height. Without the capability
of handling individual point and area sources in this manner, the realism of the
model may be severely compromised.
-18-

Criteria No. S: Receptors. A sufficient receptor density should be available
to permit "hot spots" to be identified. A sufficient number of receptors should also
be available to permit assignment of receptors to air quality monitoring stations
for model validation purposes. Receptors which may be arbitrarily assigned to the
model coordinate grid are desirable.
Criteria No. 6: Basic Model Type. There are two basic approaches to atmos-
pheric diffusion modeling: gradient transport theory and statistical theory.
Gradient transport theory models are most frequently research tools, are not
widely used in the operational sense and are not recommended for application to
the problem at hand. Statistical models (e.g., "Gaussian" models) are widely used
operationally by state and local air pollution agencies as well as the EPA. There
are two basic subsets of the Gaussian approach. These are the variable trajectory
and constant mean wind approach.
In the variable trajectory approach, spatial variations in the horizontal wind
field are accounted for. The variable trajectory model requires detailed meteoro-
logical measurements which are not currently available in the Cape Fear region.
Given this severe limitation, the variable trajectory model cannot be recommended
even though it would probably represent the land -use interface effects better than
a constant mean direction model.
In the constant mean wind or "straight line" Gaussian model, a single wind
speed and direction is assumed to prevail throughout the region of interest at any
given moment. This model is the type that is generally applied in air pollution
studies and has been the type of model historically approved by the EPA for use in
regional air quality analyses. This model type is recommended for application in
the Cape Fear region.
Criteria No. 7: EPA Status. Even within the generally acceptable class of
straight line Gaussian models, a very large degree of variation is possible.
Literally dozens of basically different models, all belonging to the straight line
Gaussian class exist. In addition, there are almost countless variations possible on
each of the basic models. Recognizing the need for national uniformity in
regulatory practices, Congress required, in the Clean Air Act Amendment of 1977,
that EPA adopt a consistent modeling policy. Regulatory positions (written as well
-19-

as unwritten) have been developed since that time which are fairly rigid in
restricting the amount of modeling latitude allowed. From a regulatory standpoint
this is convenient but from a practical standpoint, it makes proper technical
treatment of certain situations very difficult. No single model is scientifically
applicable everywhere and unique meteorological situations seem to be the rule
rather than the exception.
The criteria, therefore, must be that the model be acceptable to the EPA.
This may be translated to mean that the chosen model must produce impact
predictions which compare favorably with EPA's base model "CRSTER."
RECENT MODEL APPLICATIONS
This section briefly presents and discusses several recent examples of
applications of air quality models specifically to the Cape Fear, N.C. region. Each
of the cases discussed below was involved with Prevention of Significant Deteriora-
tion Permit applications. Although truly regional modeling which would include all
regional emissions was not required for any of these applications, examination of
the type of modeling employed is instructive because it is current, specific to the
region, and provides a good example of modeling which is acceptable to the EPA
regional office. Pertinent characteristics of these applications are presented in
Table 4.1. The location of the applicants' facilities is shown in Fig. 4.1.
DuPont - Cape Fear Plant
Two DuPont appications for this facility were reviewed. One, an application
for a permit to construct and operate a facility for the manufacture of Tereph-
thalic Acid which was dated June 14, 1978, utilized the EPA -developed PTMAX and
CRSTER models for determination of carbon monoxide impacts. The other,
representing a recent (11/26/79) state and federal application for construction of
two 245 x 106 Btu/hr input steam generating boilers utilized four separate models,
PTMAX, CRSTER, PTMTP, and CDMQC. In both of these cases, PTMAX was
applied in a screening procedure to better define the application of the CRSTER
model. Non -DuPont sources were not included in the modeling. CDMQC provided
results which were judged by the applicant to be unrealistic. This was attributed to
-20-

TABLE 4.1
PERTINENT CHARACTERISTICS OF RECENT REGIONAL MODEL APPLICATIONS
DuPont - Cape Fear
State permit to con-
PTMAX (for CO and or-
Internal to Model.
7 stacks.
Not considered.
Plant - Terephthalic
struct and operate
ganic compounds).
Acid Facility.
and PSD Preconstruc-
CRSTER
Annual Cycle - 1964
6 stacks.
Considered only "normal"
tlon Permit.
Wilmington Hatteras.
background.
DuPont - Cape Fear
State permit to con-
PTMAX
Internal to Model.
8 stacks for S02
No non DuPont.
Plant - two 245 HM
struct and operate
9 stacks for TSP N
Btu/hr heat input
and PSD Preconstruc-
CASTER
Annual Cycle - 1964
15 existing stacks for
Sources were modeled.
coal-fired boilers.;
tfon Review.
Wilmington (Permit
S02. 17 existing stacks
Background concentrations
Application states
for TSP, NO..
for NO2 only were dis-
"STAR" data but this
12 future stacks for S02-
cussed and were taken from '
is believed to be in
14 future stacks for
state monitoring data.
-
error).
TSP, NOx,
PTMTP
Worst case indentified
Same as CRSTER "future"
in CRSTER analysis.
case except 2 additional
sources considered since
sources not colocated in
model.
CDMQC
Five year STAR data for
16 point sources of S02,
,
Wilmington.
TSP. 17 point sources
NO2-•
Federal Paper Board
State preconstruction
PTMAX (by State of N.C.)
Internal to model.
Not documented in avail-
Not considered.
Riegelwood - Bark/
review and final
able reports.
Residual Oil Boiler
determination (PSD).
CBS (by EPA)
Not documented.
Not documented in avail-
Not documented in avafl-
Also „ supporting
able reports.
able reports.
study for modifica-
PTMAX (by State of N.C.)
Worst case identified
15 point sources in-
Attempt to handle back-
tfon.
in CASTER analysis.
cluding 2 sources at
ground by modeling other
Wright Chemical,
nearbv sources.
Attempt to handle back -
AQDM (by State of N.C.)
Annual Stability - wind
19 point sources in-
rose data. Source and
cluding sources at
ground problem by model -
period of record not
Kaiser, Wright
ing other nearby
documented.
Chemical, DuPont.
sources.
Not considered.
FTMAX (by TRC
Internal to model.
1 source (#5 boiler).
Consultants
Not considered.
CASTER (by TRC
Annual Cycle - 1964
7 colocated point sources.
Consultants)
Wilmin ton/Charleston,
Not considered.
PTMTP (by TRC
Worst case identified
Existing Case - 16
•
Consultants)
it CRSTER analysis.
sources. Proposed
Case - 15 sources.
CDMQC (by TRC
Wilmington STAR Annual
Existing Case - 27
Annual background con -
Consultants)
Cycles 1966-1970.
sources. Proposed
sidered through modeling
Case - 26 sources.
nearby non -applicant
sources.
Brunswick Energy
PSD Permit
PTMAX
Internal to model.
Not specifically docu-
Not considered.
Company
Application.
mented - probably 4
oint sources.
CRSTER
Annual Cycle - 1964
4 refinery stacks. 8
Measured.
Wilmington/Charleston.
non -refinery increment
,
consuming stacks.
VALLEY (modified).
Five year STAR data - _
4 refinery stacks. 8
Measured.
Wilmington. Mixing
non -refinery increment
height - 500 m.
consuming stacks.
Diamond Shamrock
Chrome Chemicals
Plant
PSD Permit
Application
- _
CRSTER MLTCRS
Annual Cycle - 1964
Wilmington/Charleston.
I
7 Diamond Shamrock
sources (includes sev-
eral combined sources).
Addition of worst case
measured background.
*For description of models applied, see Appendix A.

0 Federal Paper Board Co. at Riegelwood:
Bark/Residual Oil Boiler
®Dupont Co. at Phoenix: facility to manu-
facture teraphthalic acid; steam gener-
ating boilers
Diamond Shamrock Corp. near Castle
Hayne: expansion of chrome chemicals
plant
®Brunswick Energy Co. near Belville: refin-
ery
61
Scale
c 10 M,ie$
10 0 10 Kilometer
FIGURE 4.1 RECENT PSD APPLICATIONS IN THE CAPE FEAR REGION
r :
%. hide H.I1�:�
—\DIY "Id Ft
y✓i
"rf Clty t1.
1�t
Top..e Beach rr r,v
V
IBnN..
G.ollh. Beach
Nllminat.h 0...h
QN
V

the "urban" assumptions in CDMQC being inappropriately applied to a rural
location.
Federal Paper Board - Riegelwood
The North Carolina Department of Natural and Economic Resources review
of Federal Paper Board's No. 5 Bark/Residual Oil Boiler was dated May 1977. In
this pre -construction review, the State agency used the models PTMAX, CRS
(predecessor of CRSTER), PTMTP, AQDM, and CDMQC.
Diamond Shamrock — Castle Hayne
A recent PSD permit application for the proposed expansion of the Chrome
Chemicals Plant utilized CRSTER and MLTCRS. MLTCRS is Diamond Shamrock's
consultant's model patterned closely after CRSTER but able to consider multiple
non-colocated sources. EPA has recently released the MPTER model which is
capable of fulfilling the same need. This application was dated March 4, 1980.
Brunswick Energy Company - Wilmington
Another recent permitting application (May 30, 1980) applied the EPA -
developed models CRSTER, PTMAX, and VALLEY to the problem of predicting
ground level concentrations due to the operation of a proposed oil refinery. Again
in this case, PTMAX was used as a screening model to determine the proper
location of ring distances in the CRSTER model. The application of the VALLEY
model to the flat terrain situation in the Cape Fear region, however, is rather
unique. VALLEY is a model designed to operate in the rugged, mountainous terrain
of the western United States. The rationale for its application to the proposed
Brunswick Energy Company facility was that, for annual average impact predic-
tions, it allowed convenient use of a multi -year data base. The meteorological
data base used in modeling was from the New Hanover County Airport. Mixing
heights were determined from upper air data collected at Charlestown, S.C.
-21-

RECOMMENDED MODELING APPROACH
Several potential model and model package candidates have been evaluated.
Those models evaluated include all models recommended for use by EPA in their
modeling guideline. The models are measured against the requirements set forth
earlier in this chapter and a recommended course of modeling action is developed.
Model Selection
Based on criteria No. 6, the model should be selected from the family of
straight line Gaussian models. Every model reported to be used by every PSD
applicant discussed in the previous section was a straight line Gaussian model. The
successful use of this type of model in the Cape Fear region lends weight to the
desirability of selecting such a model for regional application.
Models Considered. The air quality dispersion models listed in Table 4.2 were
considered as candidates for recommended use by Cape Fear Council of Govern-
ments. Their consideration results either from their recent use in a regional
permitting action or their mention in the EPA modeling guideline. Two additional
models, the Industrial Source Complex Model (ISC) and the MPTER model which
were developed too late to be included in EPA's guidelines of 1978 are also
considered. All models listed in Table 4.2 are described in Appendix A.
Model Evaluation. Table 4.3 presents a summary of several of the most
important characteristics of the fourteen models listed above to facilitate compar-
ison.
-22-

Table 4.2. Candidate Air Quality Dispersion Models
MnA-1 Nnmi�
Air -Quality Display Model (AQDM)
APRAC - lA
Climatological Dispersion Model
(CDM and CDMQC)
RAM (including RAMR)
Single Source Model (CRSTER)
Multiple Source CRSTER (MLTCRS)
Multiple Source CRSTER (MPTER)
Texas Climatological Model (TCM)
Texas Episodic Model (TEM)
Point Maximum Model (PTMAX)
VALLEY
PTMTP
Industrial Source Complex Model (ISC)
Reference
Source
Number*
EPA
(1)
EPA
(2)
EPA
(3)
EPA
(4)
EPA
(5)
Dames & Moore
(6)
EPA
(7)
Texas Air
(8)
Control Board
Texas Air
(9)
Control Board
EPA
(10)
EPA
(11)
EPA
(12)
EPA
(13)
*Brief abstracts as numbered are presented for each of these models in Appendix
A.
Every one of the models listed in Table 4.3 requires a digital computer
facility for implementation. Most are available in the FORTRAN language.
Implementation of the models on a computer system requires someone knowledge-
able in computer programming since system compatibility changes to the codes are
frequently required. Several of the models are relatively large consumers of
computer resources, especially those operating on complete annual periods of
hourly meteorological data. The computation resource requirement is, in most
cases, heavily dependent upon the complexity of the situation being investigated.
Numerous point and area sources coupled with many receptors requires substantial
resources in certain models, notably RAM.
-23-

TABLE 4.3 AIR QUALITY MODEL COMPARISONS
Meteorological.
Horizontal
Model
Terrain
Point Sources
Area Sources
Lime Sources
Data
Averaging Periods
Dispersion
1. AQDM
Not treated.
Arbitrarily
Treated by point
Not treated.
Annual
Annual with ata-
22.50 Uniform
located.
source
Stability
tistical ex-
Distribution
approximation.
Wind Rose.
trapolation to
(Sector Average).
shorter periods.
2. APRAC-lA
Not treated.
Not treated.
Sector area
Aggregates to
Arbitrarily
One hour.
22.51 and 45.0
sources.
area sources.
assignment of
Sector Average. .
Street canyon
hourly values.
sub -model
available.
3. CDM-CDMQC
Not treated.
Arbitrarily
Size must be
Not treated.
Annual Stability
Annual with
22.50 Sector
located.
multiple of
Wind Rose.
statistical
Average.
user unit.
extrapolation
to shorter
,
periods.
4. RAM
Not treated.
Up to 250
Must be squares
Not treated.
Hourly wind
One hour and
Horizontal off-
RAMR
arbitrarily lo-
and multiples
speed, direc-
one additional
centerline calcu-
RAMF
cated point
of basic user
tion,.tempera-
period selected
lation. Turner
RAMFR
sources
unit. 'Up to
ture stability
between 2 and
curves (rural).
CUMF
accommodated.
100 area sources
and mixing
24 hours.
McElroy-Pooler
RAMBLK
are accommodated.
depth. (Same
curves (urban).
RAMQ
data format as
RAMMET
CRSTER).
5. CRSTER
Terrain allowed
Up to 19 co-
Not treated.
Not treated'.
Hourly wind
1, 3, 24 hours
Horizontal off -
to rise to top
located sources
speed, direc-
and annual aver-
centerline calculation
of shortest
at center of
tion, tempera-
age. The 24 hr.
Turner curves.
stack.
radial grid.
ture, stability
averages are
and mixing
non -overlapping
depth. (Same
midnight to
data format as
midnight.
RAM)
-
6. MLTCRS
DOCUMENTATION NOT
AVAILABLE FOR THIS MODEL - IT IS UNDERSTOOD TO BE VIRTUALLY IDENTICAL
TO CRSTER EXCEPT THAT
POINT SOURCES NEED NOT BE COLOCATED.

TABLE 4.3 (Cont'd)
Meteorological
Horizontal
Model
Terrain
Point Sources
Area Sources
Lime Sources
Data
Averaging Periods
Dispersion
DOCUMENTATION NOT
AVAILABLE FOR THIS
-
MODEL IT IS UNDERSTOUL)
TU BE
7. MPTER
VIRTUALLY IDENTICAL TO CASTER EXCEPT THAT POINT SOURCES NEED NOT BE
COLOCATED.
8. TCM
Not treated.
Arbitrarily lo-
Arbitrary.loca-
Not treated.
Annual
Annual but may
22.50. Sector,
cated. Un-
tion and square
stability wind
be run for
Average.
limited number.
shape.
rose.
shorter
periods.
9. TEM
Not treated.
Up to 300
Up to 200
Not treated.
User -supplied
10 minutes, 30
Horizontal off -
arbitrarily
arbitrarily
scenarios of
minutes; 1 hour,
centerline calcu-
located sources.
located sources.
3 hours each.
3 hours.
lation. Turner
curves.
10. PT14AX
Not treated.
Single point
Not treated.
Not treated.
Generated in-
One hour.
Lateral center -
source.
ternally in
line concentra-
the program.
tion.
11. VALLEY
Accommodates
Source located
Up to 50 point
Not treated.
User -input both
24 hour and
22.50 Sector
any terrain.
at center of
and area sources.
short-term and
annual..
Average.
radial grid.
Area sources
annual sta-
,
arbitrarily
bility wind
sized and
roses.
located.
12. PTMTP
Not treated.
Up to 25
Not treated.
Not treated.
User -input for
Average pro-
Turner curves.
arbitrarily
up to 24 hrs.
vided for num-
located point
be; of meteoro- .
sources.
logical hours
entered.
13. ISC
Terrain allowed
Arbitrarily lo-
Arbitrarily lo-
Treated as
Hourly wind
1, 2' 3, 4, 6,
Horizontal off -
to rise to top
cated. Limit
cated. .Must be
volume source.
speed, direc-
8, 12 and 24
centerline compu-
of shortest
depends on com-
square with
tion, tempera-
hrs. (ISCST).
tation. Turner
stack.
puter storage
uniform N-S and
ture, stability
Annual (ISCLT).
curves. Urban
available.
E-W dimensions.
and mixing
option uses de -
depth. Accepts
stabilization.
same.data format
as CRSTER.

TABLE 4.3 (Cont'd)
Emission Rate
Vertical Dispersion
Emission Rate
Temporal Variability
Plume Rise
Applicability
Comments
1. Pasquill-Gifford
Single rate for each
Not allowed.
Briggs Option.
Urban Access Only.
Generally considered
Curves
source.
outdated.
2. Modified McElroy-
Calculated from Daily
Hourly emissions are
Not treated.
Intended for CO pre-
Requires an extensive
Pooler
Traffic Volume.
generated internal
dictions for urban
traffic survey.
to the model from
areas only.
Daily Traffic Volume
3. Turner Curves
Single rate for each
Day/night variation
Not treated for area
Urban areas only.
CDMQC is the version
point and area source.
but same variation
sources. Briggs
containing the sta-
assumed for all
neutral/unstable used
tistical model required,
sources.
for point sources.
to determine 1-24 hour
average concentrations.____
4. Turner Curves
Unique, constant rate
Hourly emissions in-
Briggs for point
Urban or rural areas.
The RAM User's Manual
(rural)
for each point and,
put on an optional
sources. User input
states "Urban planners
McElroy-Pooler
area source.
basis.
effective height for
may use RAM to deter -
Curves (urban)
area sources.
mine the effects of new
source locations and of
control strategies upon
short-term air quality." _
5. Turner Curves
Unique rate for each
Monthly rate varia-
Briggs final rise
Urban or rural areas.
This is EPA's "reference"
of the up to 19
tion input by user.
(no intermediate
model. Current EPA
sources.
rise).
policy dictates that all
other models must success-
fully compare against.
CRSTER.
6 IDENTICAL TO CRSTER
---
7
8. Turner Curves
Unique, constant rate
Not allowed.
Brigge neutral/un-
Basically an urban
Model has been reported
for each point and
stable only for
model.
more accurate when used
area source.
point sources.
for point sources than
area sources.
9. Turner Curves
Unique emission rate
Not allowed.
Briggs plume rise.
Urban or rural.
Model manually applied
for each point source.
to determine "worst
case" short-term
impacts.

TABLE 4.3 (ConVd)
Emission Rate
Vertical Dispersion
Emission Rate
Temporal Variability
Plume Rise
Applicability
Comments
10.
Turner Curves
Single emission rate
Not allowed.
Briggs plume rise.
Rural.
Basically a "screening
for single point
model" used to locate
source.
and bound maxima.
A.
Turner Curves
Single emission rate
Not allowed.
Briggs plume rise or
Urban or rural;
Model not useful for
or area source.
optimal'. assignment
urban lacks terrain.
short-term prediction
of fixed plume rise.
except for the 24-hour
period.
12.
Turner Curves
Single emission rate
Not allowed.
Briggs plume rise.
Predominantly rural
for each source.
areas.
13.
Turner Curves;
Unique rate for each
Emissions from all
Briggs plume rise.
Urban or rural
Mainly intended for
urban option
point, area and
or individual sour-
areas.
"industrial park" scale
uses destabili-
volume source.
ces may be varied
but could be applied on
zation.
as a function of
a larger scale. Numer-
time.
ous options require a
high degree of sophis-
tication in running
model.

The choice of an appropriate model for planning purposes for the Cape Fear
region will be arrived at through elimination of those models which clearly do not
satisfy the established criteria. Those models then remaining will be compared for
their merits.
AQDM should not be considered a candidate model since it is outdated and
CDM performs essentially the same function using more current formulation.
APRAC is eliminated since it is designed and intended for a special purpose use
(vehicular emissions only) and cannot treat point sources. CRSTER, MLTCRS, and
MPTER cannot be considered since they cannot accommodate area sources, even
though CRSTER is EPA's benchmark model for rural applications. PTMTP and
PTMAX, generally considered screening models, are also eliminated since they
cannot consider area sources. Although we would recommend the use of ISC in
evaluating, in detail, single industrial comlexes, should the need ever arise, ISC is
intended for use on a spatial scale considerably smaller than the regional scale
required here. VALLEY is intended for use mainly in rugged terrain.
These eliminations leave only CDM, RAM, TCM and TEM remaining. TCM
and CDM are intended for annual average concentration predictions. RAM and-
TEM are basically intended for shorter term predictions. Since TCM is primarily
designed to be applied to determine "worst case" short-term impacts from a
multitude of point and area sources, its usefulness as a regional model where all
cases are to be predicted (to generate frequency distributions) is limited. A .
further limitation of TEM is that the program is capable of handling only three
days of meteorological data at a time. * As the code is now understood to be
written, 122 separate runs would be required to process a year's worth of
meteorological data. For the short-term averaging periods, then, RAM is the
choice. Regarding annual average concentrations, both CDM and TCM have two
serious drawbacks --they are basically urban models (while the region of interest
encompasses substantial rural acreage) and they are inadequate to realistically
describe variations in source emissions over the annual period. The latter point is
more of a problem with TCM than CDM.
Such variations in emissions should be reported in the emissions inventory and
should be included in the modeling. Considering these drawbacks and the fact that
it would be- both logical and convenient to have both short-term and annual
-24-

predictions made using the same modeling assumptions, it is recommended that
RAM be modified to operate both in a short-term and annual mode. This
modification consists mainly of writing a post -processing program capable of
reading the partial concentration tape output from RAM and averaging over the
appropriate periods. Additional analysis capabilities could be built into this
program to allow impacts from selected sources or groups of sources to be
examined and frequency distributions constructed. Since some substantial effort
will be required to ready RAM to operate in this mode and since RAM is a
prodigious consumer of computer resources, the Cape Fear Council of Governments
may wish to begin air quality studies with the immediate implementation of CDM
to obtain an initial reading on the regional air quality situation. This is further
discussed later in this chapter. TCM could also be used in this application but CDM
is preferred since it is an EPA model.
Recommended Model. It is recommended that the RAM model be chosen as
the basic regional model. It is further recommended that the urban and rural
versions of RAM be properly applied to the urban and rural areas of the region of
interest. The urban version of RAM is the EPA -approved and recommended model
for application in urban areas according to the October, 1980 revision to its air
quality model guidelines. Because of the urban/rural classification process
prescribed by EPA, it is likely that areas in addition to downtown Wilmington will
be classified as urban and appropriately modeled by urban RAM. Rural/urban
classification will .be a judgment for the personnel actually responsible for
implementing the model to make. Surface meteorological data from New Hanover
County Airport and upper air data from Charleston, S.C. or Hatteras, N.C. should
be used with Charleston being the first choice of the two.
The RAM model consists of a package of computer programs. RAMMET, the
meteorological preprocessor, prepares National Weather Service meteorological
data for use by the RAM model. It is identical to the preprocessor for the CRSTER
model. RAMQ, the source preprocessor, prepares the source data for use by RAM.
A post -processing program or series of programs will have to be written by COG
staff to perform the proper averaging on the RAM model output tape to obtain
average concentrations for time periods of interest other than the single period
RAM is capable of averaging. There is an excellent opportunity in preparing this
-25-

post -processor to build in analysis tools to enable contributions from chosen
sources or classes of sources to be investigated individually.
Because RAM does not completely fulfill the requirements of Criteria 2 and
Criteria 7, it is recommended that certain other modifications be made to the
model. A professional meteorologist with knowledge and experience in FORTRAN
can readily make and verify these changes.
1. The model should be modified (post -processor written) to enable annual
averages to be computed using an annual cycle of meteorological data.
2. An option should be created within the model to accommodate emissions
changes throughout the annual period. This will affect both RAM and its
source pre-processor, RAMQ.
3. The vertical wind speed extrapolation powers should be changed to be
identical with the CRSTER model.
4. The portion of the code dealing with the "lidded" case should be modified
to limit dispersion in the vertical when stable conditions exist. This is a
CRSTER compatibility change.
5. The stack tip downwash currently employed by RAM should be elimi-
nated to assure comparability with CRSTER.
6. RAM should be modified to incorporate the CRSTER terrain treatment.
7. The Briggs plume rise calculations should be altered in terms of the
stable rise coefficient and also in terms of plume height above ground
before the point of final rise to be compatible with CRSTER.
This model, as any model, should be validated once it has been applied to the
region. This is accomplished through comparison of predicted and measured
concentrations at the same geographical location. Frequency distributions of both
measured and predicted concentrations are most helpful in this comparison. Any
systematic model errors detected at this time should be corrected. Large random
-26-

departures of predicted from measured values should be investigated to determine
the cause and, if possible, correct it.
Use of the RAM model, modified according to the recommendations made
here, will satisfy the criteria established earlier in this chapter.
Recommendation of one particular model does not imply that that, model
should be used to the exclusion of all others. Screening -type models will be
required for proper placement of RAM receptors and for decisions regarding point.
source significance. Also,' the recommendation of RAM, based on past experience,
will not be valid forever. Dispersion modeling is a. rapidly evolving field. Large
research efforts are underway to resolve some of the current modeling limitations.
Although RAM is relatively modularized and easy to follow so that future changes
required by changes in the state-of-the-art will be relatively easy to implement, at
some future point the model will simply be obsolete and require replacement.
Competent professional judgement is required to continually appraise the scientific
status of dispersion modeling and to continually review the merits of maintaining
the recommended model.
Data Required for Recommended Model
Source Data.
o Area Sources
Effective emission height, side lengths, and emission rate are required
for each area source. It will require very substantial effort to generate
these data for the model. They would include emissions for each
appropriate pollutant due to airport operations, space heating, vehicular
traffic, and industrial sources too small to warrant being treated, as a
point source (each below 100 tons/year of a criteria pollutant).
o Point Sources
For those sources sufficiently significant to treat separately, (over 100
tons/year) stack temperature, exit velocity, exit diameter, and emission
rates for each pollutant. are required. Information on temporal emissions
variations will be required when the model is so modified. The emissions
-27-

data should be the most recent available and, ideally, should be no more
than two years old.
Meteorological Data. Considering the fact that the terrain of the area is flat
and the fact that perturbations on the general flow caused by the proximity of the
coastline will have to be ignored due to the current state-of-the-art of approvable
dispersion models, meteorological data collected at New Hanover County Airport
will suffice for wind, stability and temperature input to the model. Of available
National Weather Service stations, this one' is deemed the most representative.
Data will have to be obtained for that period concurrent with the emissions data as
well as concurrent with the validation air quality data used for model validation.
This may require special preparation of data at the National Climatic Center
(NCC) since observations are required for every hour of the year. Available
information indicates that hourly data may not currently be available for certain
periods for the Wilmington/New Hanover Airport. The data are required in CD-144
format for input to the preprocessor. Since a long lead-time may be required to
,. obtain these data, the order should be placed months in advance of anticipated
need assuming that the required physical tape characteristics and year of validity
of the emissions inventory and air quality data are known.
Mixing depth data should be those obtained through radiosonde observations
at Charlestown, S.C. These data, known as the Holzworth morning and afternoon
mixing depths should be calculated using the New Hanover Airport surface
temperature. Hatteras radiosonde data could be used if Charlestown were not
available. The Holzworth mixing depths are used as input to the meteorological
preprocessor.
Receptor Data. Receptors must be carefully specified. Options are available
within the program to allow the model to generate a "honeycomb" receptor array
or to automatically generate receptors near major point sources. Alternatively,
receptors may be specified by the user. Receptors will have to be judiciously
chosen by the meteorologist applying the model. Sufficient receptors will have to
be used to generate predictions at all locations of interest (including the ambient
air quality monitoring location) and to identify any "hot spots." However, economy
will also have to be exercised since the cost of operating the model is almost
-28-

directly proportional to the number of receptors used. A proper balance between
economy and detail must be struck.
Results Expected from Recommended Model
The recommended model and its post -processor will generate predictions of
ground -level ambient air quality for the inert pollutants. These are particulates,
SO2, and CO (when vehicle emissions are treated as area sources). Predictions of
NO (all oxides of nitrogen) concentrations may also be made and may be
interpreted as indicators of actual NO2 patterns. In a rigorous sense, NO2 cannot
be treated by RAM (or by any other model listed in Table 4-1) since the NO2
present in the atmosphere is a result of both NO2 emissions plus the oxidation of
NO also emitted from combustion processes. The atmospheric chemistry associ-
ated with NO to NO2 conversion is not adequately understood. For this reason, it
is customary (and conservative) to assume that all of the NO emitted by a source
is NO2 and to model accordingly with the understanding that the results may be
overestimated. The only criteria pollutant the model will be totally incapable of
handling will be ozone, as discussed earlier.
Accuracy of dispersion model predictions is generally considered to be within
a factor of two of the actual concentrations. If the model validation does not
confirm that the model is reasonably accurate, a calibration may be successful in
improving the accuracy of the estimates. It should be remembered that the model
results can be no more accurate than the input meteorology and source emissions
data. Sources not included in the inventory which are present, sources included
which are now defunct, or gross errors in emissions or stack data may seriously
degrade the modeling. This is not to say that every number input to the model
must have a +/-1 % accuracy but reasonable care must be taken and emission rates
of large sources must be as accurate as possible.
Once the model and its post -processor have been checked -out and validated
and the current emissions and meteorological data assembled, the model can be
used to generate predictions corresponding to current conditions. Due to the
inability to locate receptors at will, the model output will provide a more detailed
spatial picture of current air quality characteristics. This can be done for all inert
-29-

pollutants for all appropriate averaging times. Graphics, especially isopleths of
annual concentrations, helps to promote an understanding of the spatial variation in
air quality across the region. This current case becomes the base case against
which all proposed or desired emissions changes may be evaluated.
Evaluation of a new major point source may be obtained by simply entering
the estimated emissions, grid coordinates, and stack parameters in the RAMQ
source pre-processor and then running RAM. The amount of increment remaining
may be tracked through modeling of those new sources and emissions changes
which have taken place since the "baseline" date. The evaluation of the ambient
effects of generic changes in emission parameters such as a change in the
permissible sulfur content of fuel oil or the substitution of natural gas for fuel oil
on a large scale may also be readily evaluated. This would likely be accomplished
through alteration of the area source parameters. Almost any conceivable
strategy, as long as it can be translated quantitatively into a change in emissions,
can be evaluated for its impact on regional air quality.
Phased Model Implementation
Implementation of RAM and its associated pre-processors and development of
a suitable post -processor to provide for averaging and analysis is a relatively
complicated task. Since the Council of Governments has not had extensive air
quality modeling experience in the past and since it may require a substantial
amount of schedule time to make the RAM package operational, it is recommended
that the entire UNAMAP modeling series be purchased and some of the simpler
models applied first. Especially recommended for this purpose are PTMAX,
PTMTP and CDMQC. Experience gained in applying PTMAX and PTMTP to sources
in the region will be helpful in the decisions required to locate the RAM receptor
fields. Results from the early application of CDMQC should be a good approxima-
tion to those to be eventually obtained for the urban portions of the area from
RAM. Gross departures of CDMQC results from measured data would indicate that
a careful review of the point and area source inventory is required. In addition to
obtaining an early indication of regional air quality characteristics, the practical
experience gained by the modeling staff will be valuable. This is especially true if
the staff generally lacks experience in regional modeling.
-30-

5. Alternative Air Quality Planning Structures
for the Cape Fear Council of Governments
THE NEED FOR AIR QUALITY PLANNING
There are substantial advantages to a well -understood and utilized air
emissions inventory. From a regional planning point of view, it allows for a careful
and well planned strategy of regional development, whereby industries can be
encouraged or discouraged to locate in an area and hence provides for planning the
intensity, mix and location of industries in the region. Although the state
DNR&CD is responsible for issuing permits and approvals for new or modified
industrial air emission point sources, its focus is at the state level. It is
understandable that its broad focus may not always reflect local social, economic,
and physical conditions with the sensitivity the region needs to achieve its planning
objectives.
Recent PSD applications, reviewed in Chapter 4, point out the need for Cape
Fear COG to implement some form of air quality planning or, at a minimum, the
capacity to give technical scrutiny to PSD applications for point sources in, or
affecting the region. For example, most of these recent PSD applications did not
utilize air quality monitoring data to establish baseline conditions but instead
created this data based on certain assumptions on air quality. Compounding this
possible underestimation of actual conditions is the accepted practice of incorpor-
ating the modeling results previous PSD applications as a substitute for monitoring
actual air quality. Thus, the anticipated concentrations of air pollutants are based
on a pyramid of assumptions as to the actual air quality. This observation is in no
way intended as a criticism of the PSD applicants since their modeling efforts used
approved EPA models and followed recommended procedures which were reviewed
and approved by DNR&CD. The point is that each new applicant is allowed the
-31-

opportunity to make a number of assumptions in its favor which may or may not be
questioned or challenged.
Ultimately, these shortcomings of monitoring and modeling will be corrected.
Until they are, however, the actual capacity of the air to absorb pollutants will
decrease with the following possible consequenses:
o The type, scale, and location of industrial development will have been
planned and executed by industrial interests with local public review
having little influence on any of these important parameters. This could
result in an imbalance in the regional economy. It may result in a lack
of diversification of jobs and the consumption of the regional "air
resource" so as to preclude future industrial development needed to
restore economic balance.
o The depletion of the regional air resource may effectively block the
construction of onshore support facilities of OCS activities, should a
major oil or gas find be made in Lease Sale 56. Although such a find
appears unlikely at the present time (see Chapter 6), the region is well
advised to keep this option open for at least a decade.
THREE ALTERNATIVES FOR AIR QUALITY PLANNING
There are various levels of involvement that Cape Fear COG can consider
with respect to air quality management. The most intense level is for Cape Fear
COG to become the local Air Pollution Control Board.
The North Carolina DNR&CD is empowered to delegate its authority as an
Air Pollution Control Agency to the Cape Fear COG. Such delegation of state
authority to the region would give the COG the power to request, acquire, and
compile data for an official regional atmospheric emissions inventory and to review
all PSD applications within its jurisdictions. This approach is not recommended
from several perspectives: First, such delegation of authority is unlikely from a
historical point of view. Second, the level of funding required for a regional
-32-

program to quality for such authorization is possibly beyond what CFCOG would
want to allocate to such a program.
Even though a designation as a local Air Pollutions Control Board should
probably not be sought, based on the above considerations, it would still be highly
desirable from a regional planning perspective for Cape Fear COG to have the
capabilities to do atmospheric dispersion modeling for planning purposes and to
review data inputs and assumptions used in the required modeling efforts of PSD
applicants. Chapters 2-4 presented the steps that Cape Fear COG should take to
fully implement an in-house capability for regional air quality planning. This
approach will require hiring a full time meteorologist/air quality specialist and an
allocation of upwards of $100,000 per year for computer -related expenses. How-
ever, there are two other approaches requiring a lesser committment of resources
that Cape Fear COG can consider. In the discussion that follows we describe each
of these three options and list their advantages and disadvantages.
OPTION 1 - PERMANENT METEOROLOGICAL STAFF
This option would allow the Cape Fear COG to retain the services of a
competent meteorologist/air quality specialist familiar with atmospheric dispersion
modeling techniques for planning purposes. This individual, due to his professional
background, should rapidly establish channels of communication with personnel at
DNR&CD and obtain access to the latest state's regional air emissions data (also
available to the general public). This individual must then have available to him
automatic data processing equipment, as well as computer facilities and models, to
be able to provide maximum inputs to the planning of the Cape Fear Region.
The Senior Meteorologist/Air Quality Specialist should have as a minimum a
bachelor's degree and three years experience in air quality impact analyses or
preferably a graduate degree with five years experience in this field. This
individual must be able to apply dispersion models to evaluate existing and
proposed fuel -burning facilities, a detailed knowledge of Federal and state regula-
tions, previous participation in meteorology/air quality data collection programs,
and experience in data processing (FORTRAN) analysis. This individual must also
have demonstrated technical verbal and writing skills. It is estimated that a salary
offer with a range of $28,000 - 35,000 per year would attract a number of qualified
-33-

applicants for the position. Automatic data processing equipment, computer time .
and programs for this option are estimated to cost between $80,000 - $100,000 per
year.
Advantages
a) In-house expertise is available to provide a level of scientific credence
to the air quality determinents of Cape Fear COG's regional planning
strategies.
b) The staff meteorologist/air quality specialist can initiate activities
described in Chapters 2-4 towards the updating and compilation of
emissions inventory data, preparation of programs, etc., as soon as he is
hired.
c) CFCOG will have the expertise in-house to adapt and fine tune models
applicable to the Cape Fear region for planning purposes and to review
modeling assumptions and data in -puts used in PSD applications.
-, Disadvantages
a) Since regional modeling and introduction of new industrial sources are
highly politically sensitive issues, and since the findings that the COG's
meteorologist may make do not have enforcement authority, this indi-
vidual has the potential of rapidly becoming dissatisfied with his
position.
b) Provisions must be made to keep the meteorologist/air quality specialist
abreast of technical developments in his field by attendance at scientific
conferences and meetings.
OPTION 2 - OBTAIN THE SERVICES OF A CONSULTANT ON A RETAINER BASIS
This option would allow the Cape Fear to retain the services of a qualified
consultant on a standby basis. Under this arrangement, the consultant would be
-34-

paid a nominal fee, but be expected to perform the necessary predictive atmos-
pheric analyses in a rapid fashion.
This option allows the Cape Fear COG to have expert consulting services
available at a limited cost. If additional work is required above the agreed to
terms of the retainment, the consultant would already be familiar with the
environs and the available data and therefore be able to perform these studies and
analyses promptly.
The typical cost for this type of a retainer arrangement is between $10,000 -
$20,000 per year.
Advantages
a) Expert services are available on -call basis.
b) Only a small capital expenditure is required until a specific study is needed;
then initiation of study requires only a telephone call or letter.
c) Retainer would pay for gathering available inventory data, as well as
developing an initimate familiarity with the local area. Models can be
scrutinized and, on a limited basis, evaluated.
Disadvantages
a) Available expertise is not really in=house.
b) Expenditure of monies is made on an on -going basis.
OPTION 3 -OBTAIN THE SERVICES OF A CONSULTANT WHEN NEEDED.
This option would allow the Cape Fear COG to have qualified inputs to the
region's development only when needed, therefore, costs would be nominal.
However, the study or studies performed would rapidly become outdated and
required periodic updating. Long range planning would be difficult at best: since
-35-

industries locate or leave the area at will, regional planning efforts would be
placed in a post -fact reactive position.
Advantages
a) No expenditures are necessary until study is needed.
b) Selection of a consultant is based on bidding thus assuring qualified,
cost-effective work.
c) There is no potential for local bias, since consultant's work is limited.
Disadvantages
a) The need for a study must be recognized early and quickly defined.
Monies must then be allocated and a specific RFP issued, all of which
delays the quick response that is often necessary.
b) The Consultant would usually require 3-6 weeks to catch up with
available information.
c) Work can only be based on the state's available inventory with little, if
any, fine tuning of the Cape Fear region per se.
Rogers, Golden & Halpern and Engineers for Energy and the Environment
recommend that of the three options presented the second is the one that strikes
the best balance between the needs of the region and the planning resources
available. Although the first option would give Cape Fear a strong technical
support for air quality planning as well as for looking after regional interests in the
PSD application process, the high cost of the in-house capability, its potential for
interagency friction, and the lack of enforcement authority argue against it. The
third option would provide Cape Fear with sound technical advice within the data
limitations of the state's emissions inventory, but would, due to the length of time
required to define study needs and obtain consultant services to address those
-36-

needs, result more in ad hoc reaction to decisions already made by industry and the
state than in air quality planning.
The second option is a reasonable intermediate selection. While it relin-
quishes hands-on control of technical inputs and analyses, it would provide the
Cape Fear COG with expert advice from a consultant paid by retainer to be
familiar with air emissions inventories, air quality models applicable to the region,
and regional meteorology and air quality monitoring efforts. Special planning
studies or quick response review studies using this background of region -specific
experience and knowledge could be initiated immediately by the consultant upon
request by the Cape Fear COG. The consultant could also provide technical
assistance to the COG in meetings or correspondence with DNR&CD.
-37-

6. An Estimate of the OCS-related Facilities Likely to
Locate in the Region and Their Requirements
Based on the assumption of significant commercial finds of oil and gas in the
South Atlantic on the Outer Continental Shelf (OCS) the need for a number of
onshore facilities could be directed at the Cape Fear Region. Beginning with
temporary supply bases during the exploratory phase of development, a significant
commercial find would require some or all of the following facilities, depending on
the size and proximity of discoveries.
o bases supporting development and production operations
o bases supporting platform and pipeline installation
o platform fabrication yards
o pipe coating yards
o pipelines
o gas separation/dehydration plants
o tank farms
o gas processing and treatment plants
o refineries
o marine terminals
Certain of these needs, such as platform fabrication yards, for example, may
be satisfied by existing facilities in the Gulf of Mexico. Summary descriptions of
the above facilities are given in Appendix B.
The reality of hydrocarbon explorations to date in the South Atlantic argues
persuasively against the need for any of these facilities in the Cape Fear Region
over the next 4 to 5 years with the remotely possible exception of a temporary
-38-

supply base to service exploratory drilling operations for Lease Sale 56, which will
be held in August 1981. The circumstances on which this conclusion is based are
discussed in the first part of this chapter. The last half of the chapter will
describe the characteristics of temporary supply bases.
THE HISTORY OF OCS ACTIVITY IN THE SOUTH ATLANTIC
The continental margin of the South Atlantic is considered to be a likely
petroleum province. The areas most promising according to petroleum geologists
are the continental shelf, the Blake Plateau, and the continental slope. The
relationship of these features to the coastal areas of North Carolina, South
Carolina, Georgia, and Florida is shown in Fig. 6.1.
Lease Sale 43
After geophysical surveys were initiated in 1960, a deep Continental Offshore
Stratigraphic Test (COST) well was drilled by a group of 25 oil companies under
USGS permit in 1977. The purpose of this well was to develop preliminary
estimates of the area's petroleum potential in order that the companies could
prepare bids for Lease Sale 43. The tracts off the coast of Georgia and Florida
that were leased on March 28, 1978, as a result of estimates based on the COST
well are shown in Fig. 6.2.
Onshore support for the drilling of exploratory wells for Lease Sale 43 was
provided at Savannah, Brunswick, and St. Simon's Island, Georgia. The Savannah
base was used for the first wells drilled by Tenneco. Later Exxon, Getty, and
Transco used Brunswick as a support base where the city had just completed
construction of a 1.2 million dollar OCS support facility. All companies used St.
Simon's Island as a helicopter base during the drilling of exploratory wells for Lease
Sale 43.
Between May 1979 and January 1980, six exploratory wells were drilled by
four oil companies into Lease Sale 43 tracts. All six wells were classified as dry
holes which so far has discouraged further drilling in Lease Sale 43 tracts.
-39-

FIG. 6.1. Geologic.Features of the South Atlantic Region
(Adapted from Jacobson, 1980, by Rogers, Golden & Halpern: see Jackson, 1980.)

FIG. 6.3. Tracts Proposed for Lease in Lease Sale 56
N O R T H C A R O L I N A
\
\
s \
0 \
C
AROLINA
SOURCE: Jackson, 1980.
a v
(� Cabe feu s k
`= oaFooF

Although the negative results from these test wells are not conclusive for the
entire South Atlantic, initial hydrocarbon estimates for the region have been
lowered. At the time the last exploratory well in Lease Sale 43 was being drilled,
the USGS released the risked estimates of undiscovered, economically recoverable
oil and gas resources for all 43 tracts: 7.9 million barrels of oil and 48 billion cubic
feet of gas. These estimates are below commercially producible amounts based on
geologic information from the 6 exploratory wells and the COST well, current oil
and gas prices, and the expense of constructing an offshore -to -onshore pipeline.
The USGS Outer Continental Shelf Oil and Gas Information. Program con-
cludes:
Reserve estimates approximate the cumulative production that can be
expected from a discovery. For this reason, they provide a foundation for
site -specific onshore planning. The entry for reserves is zero at this time
because no discovery of oil or gas has yet been made in the South Atlantic.
However, not all prospective areas in the South Atlantic have been explored.
If and when any company announces a commercial discovery and a
Development and Production Plan is filed, a revision of the reserve estimate
would be appropriate. But until such time, it must be considered that there
are no reserves of either oil or gas in the South Atlantic Region.
Lease Sale 56
Lease Sale 56 is scheduled for August 1981. The 286 tracts being offered are
shown in Fig. 6.3. About a third of them are located in and around the tracts
leased for Lease Sale 43 offshore from Brunswick, Georgia. Due to the negative
findings from Lease Sale 43, not much interest is expected for Lease Sale 56 in this
area. Other tracts offered in Lease Sale 56 are scattered north of Brunswick in
eight locations. No COST wells have been drilled in these locations. The southern
ones are in less than 100 meters of water, while the others and notably the largest
one, 100 miles ESE of Cape Fear, generally lie in water that is 600 to 2000 meters
deep. There are not yet any production drilling rigs capable of extracting oil at
these depths, although there are exploratory rigs that are. However, the latter are
-40-

in greater demand in areas of the world with more favorable geologic characteris-
tics; a several year wait for rigs to be available is not. unlikely. The oil industry
ranks the Blake Plateau (see Fig. 6.1) as 13th of 22 OCS areas worldwide thought
to have significant oil reserves. The USGS ranks the Southeast Georgia Embank-
ment and Blake Plateau as 15th and 17th respectively out of 22 potential OCS
areas. Lease Sale 56 is not expected to be a particularly successful sale.
The now inactive support bases at Brunswick and St. Simon's Island, Georgia,
could be reactivated to supply Lease Sale 56 activities when and if they occur,
although the long 350-mile 2-day trip.to the Cape Fear cluster of tracts may very
likely argue in favor of establishing a much closer base. If tracts in the largest
contiguous cluster of tracts in Lease Sale 56 are actually leased, then the Cape
Fear or Cape Lookout regions, being the closest landfalls, could plausibly be
selected as the site for supply boat and helicopter service support bases sometime
in the next 2 to 3 years.
DESCRIPTION OF THE REQUIREMENTS OF OCS-RELATED FACILITIES
LIKELY TO LOCATE IN THE REGION
In the previous section the possible need for a temporary supply base to
support OCS activity for Lease Sale 56 was identified. Before such a supply base is
actually built, the following assumptions will have to become reality.
o Lease Sale 56 results in the lease of tracts from among the northern
cluster of tracts offered.
o Exploratory drilling rigs capable of drilling at the depths of the lease
tracts are available within the lease period.
o The leased tracts are located closer to Cape Fear than other potential
support areas.
o Cape Fear has suitable waterfront land available where a supply base
could locate.
-41-

General Characteristics of Temporary Support Bases for OCS Exploration
Support bases for OCS exploratory drilling activities are of two kinds: one as
a staging and storage area for supply boat -transported materials needed for drilling
activities; the other is a base for helicopters that ferry drilling rig crews back and
forth to the rig. These two types of bases need not be located at the same site.
The supply boats carry drill pipe, casing, hardware, drilling mud, cement, food,
fuel, water, and other equipment. The onshore support base for supply boats
requires dock space, warehouse space, land for open storage, and access by road
and rail. In the South Atlantic, supply, boats have been 48 to 60 meters long, 12.5
meters wide and draw up to 5 meters when loaded. A more detailed presentation
of the land, waterfront, and labor requirements and potential air emissions of a
temporary supply base is given below.
Land. A minimum of 2 hectares is required per rig serviced for a support
base including helicopter services, although there is some economy of scale for
space if more than one rig is serviced by a supply base. The land area per rig can
be broken down as shown in Table 6.1. The site must have either road or rail
access, preferably both to allow efficient transmission of materials needed in the
drilling operation including freshwater, drilling mud, cement, tubular goods, fuel
for transportation and drilling, food, tools, and parts.
Waterfront. The supply base must be located on the waterfront in an all
weather harbor. Approximately 60 meters of marginal wharf are required per
drilling rig in order to load and unload supply boats. The water depth in approach
channels and at dockside must be at least 4.6 meters at low tide.
-42-

TABLE 6.1 LAND REQUIREMENTS FOR A TEMPORARY SUPPLY BASEI
Approximate Area per Rig
Function Serviced (Hectares)
Warehousing 0.2
Open storage 0.4
Helipad 0.4
Office and communications 0.02
Parking for employees 0.2
Other land (roads, rail access, 0.8
and function separation)
1 From NERBC Factbook, Nov. 1976, p. 113.
Labor. If available, the local labor force can satisfy about 30 to 35 of the 45
onshore jobs created per drilling rig at a temporary support base. The average
annual salary of these jobs is $17,000. The support base job categories are wharf
and warehouse crew, helicopter crew, and supply boat crew. Most of the offshore
jobs are filled by people brought into the area by the oil companies or their
contractors, although these companies are usually willing to hire locally whenever
possible.
Air Emissions. Support bases are accompanied by air emissions due to the
evaporation of fuel during transfer or storage and to its combustion by machinery
and vehicles operating at the site. Evaporation from fuel storage tanks depends on
the tank type (fixed roof, floating roof, or variable vapor space), its size, color and
condition, and its usage patterns. Fuel vapor losses from floating roof type tanks
eliminate breathing losses although they are subject to losses due to improper
sealing between the floating roof and the tank walls. For fixed roof tanks vapor
recovery methods can reduce evaporative losses by 90 to 95%. Loading and
unloading supply boats and tank cars, and supply boats in transit, are activities that
release significant amounts of unburned hydrocarbons to the atmosphere. Carbon
monoxide, nitrogen oxides and hydrocarbons are released by trucks, boats, trains
and helicopters in and around the supply base as combustion emissions. Other air
F-110!!

emissions can arise from dry -pumped cement and drilling mud during transfer
operations in case there is spillage or the blowout of a hose.
Supply Base Experience -in the South Atlantic
In order to give some idea of what to expect if .Lease Sale 56 onshore
exploratory support were to locate in Cape Fear, it is instructive to describe the
salient characteristics of the support bases in coastal.Georgia that serviced drilling
activities for Lease Sale 43.
Savannah. Saylor Marine Corporation owned a 2.8 hectare property with a
152-meter frontage on the Savannah River and enough dock and warehouse space to
serve as a temporary support base for the early Tenneco OCS exploratory
operations in May 1979. The Saylor marina was .also served by rail and adequate
highway access.
Brunswick. The city of Brunswick, in anticipation of exploratory drilling,
built a 1.2 million dollar support base. Coastal Energy Impact Program (CEIP)
funds guaranteed the financing of construction. This facility was specifically
designed to service the particular access, storage, and transfering needs of the
OCS drilling activities. The facility provides warehousing for dry storage, open
storage for pipe and pipe casing, fuel storage tanks, drilling mud containers, dock
space for loading and unloading, and berthage for work boats. The site also
includes loading cranes and office space for operation management. The city and
local business interests actively sought out oil companies to use this support base.
Getty, Tenneco, and Exxon utilized the Brunswick support base. Exxon leased
additional office space .(350 square meters) offsite in Brunswick. Global Marine,
Inc., the company that leased the drilling rig to .Exxon, also rented office space in
Brunswick.
-44-

St. Simon's Island. All drilling operations for Lease Sale 43 were serviced by
helicopters that used Malcolm McKinnon Airport on St. Simon's Island as a base to
ferry work crews. This existing facility met all the needs of the operations,
including hanger space, warehouse, helicopter pad (less than 0.5 hectares) and
off ice space.
In addition to oil industry use of St. Simon's Island, the USGS established a
temporary field office there as a base from which to monitor OCS drilling
activities.
-45-

BIBLIOGRAPHY
40 CFR Parts 50, 51, and 124; Prevention of Significant Deterioration.
40 CFR Part 51; National Ambient Air Quality Standards.
40 CFR Part 51; Emissions Offset: Interpretive Ruling
40 CFR Part 60; Standards of Performance for New Stationary Sources.
U.S. EPA; Background Information for Proposed SO2 Emissions Standards; EPA-
450/2-78-007a.
U.S. EPA; Ambient Monitoring Guidelines for Prevention of Significant Deteriora-
tion; EPA-450/2-78-019.
U.S. EPA; Guidelines on Air Quality. Models; EPA-450/2-78-027.
U.S. EPA; Guidelines on Air Quality Models; Proposed Revisions, October 1980.
Jackson, J.B. 1980 (July), Outer Continental Shelf. Oil and Gas Activities in the
South Atlantic (U.S.) and their Onshore Impacts: A Summary Report. U.S.
Geological Survey Open File Report 80-626; Outer Continental Shelf Oil and Gas
Information Program.
Jacobson, J.P., 1974; Physical Oceanography, vol. 1 (prepared for DOI/BLM).
New England River Basin Commission, 1976 (Nov.); Factbook: Onshore Facilities
Related to Offshore Oil and Gas Development, NERBC-RALI Project.
North Carolina Environmental Policy Act of 1971; as amended.
North Carolina Air Pollution Control Regulations; Administrative Code Title 15
DNR&CD; as amended.
North Carolina Water and Air Resources Acts.
Federal Paper Board Co., 1977 (May); Prevention of Significant Deterioration
Permit Application for Bark/Residual Oil Boiler -Riegelwood Plant.
Dupont De Nemours & Co., Inc., 1978 (July 14); Application for Air Emissions
Permit - Petrochemicals Expansion - Cape Fear Plant.
DuPont De Nemours & Co., Inc.; 1979 (Nov. 26) Application for Air Emissions
Permit: Coal -Fired Boilers Cape Fear Plant.
Dames and Moore, 1980 (Mar. 4); Prevention of Significant Deterioration Permit
Application; Diamond Shamrock Corporation: Expansion of Chrome Chemicals
Plant at Castle Hayne.
Brunswick Energy Company, 1980 (May 30); Prevention of Significant Deterioration
Permit Application: Refinery
-46-

APPENDIX A. ABSTRACTS OF AIR QUALITY DISPERSION MODELS*
1. AQDM
Abstract: AQDM is a climatological steady state Gaussian plume model that
estimates annual arithmetic average sulfur dioxide and particulate
concentrations at ground level in urban areas. A statistical model
based on Larsen is used to transform the average concentration data
from a limited number of receptors into expected geometric mean
and maximum concentration values for several different averaging
times.
2. APRAC-IA
Abstract: APRAC is a model which computes hourly average carbon monoxide
concentrations for any urban location. The model calculates contri-
butions from dispersion on various scales: extraurban, mainly from
sources upwind of the city of interest; intraurban, from freeway,
arterial, and feeder street sources; and local, from dispersion within
a street canyon.- APRAC requires an extensive traffic inventory for
the city of interest.
3. CDM and CDMQC
Abstract: CDM is a climatological steady-state Gaussian plume model for
determining long-term (seasonal or annual) arithmetic average pol-
lutant concentrations at any ground -level receptor in an urban area.
An expanded version (CDMQC) includes a statistical model based on
Larsen to transform the average concentration data from a limited
number of receptors into expected geometric mean and maximum
concentration values for several different- averaging times.
4. RAM and RAMR
Abstract: RAM is a steady state Gaussian plume model for estimating concen-
trations of relatively stable pollutants for averaging times from an
hour to a day from point and area sources. Level or gently rolling
terrain is assumed. Calculations are performed for each hour. Both
rural and urban versions are available.
*Air quality dispersion models are made available by the Environmental Applications
Branch. In addition to having executable programs on EPA's UNIVAC 1110 at
Research Triangle Park, NC, a tape having FORTRAN source programs has been
placed with NTIS (National Technical Information Service). U.S. Department of
Commerce, Springfield, VA 22161. The current tape is called: UNAMAP (Version 3).
Its accession number is PB-277-193. There are eleven dispersion models contained on
the tape.
FA91

5. Single Source CRSTER
Abstract: CRSTER is a steady state Gaussian plume technique applicable to
both rural and urban areas in uneven terrain. The purpose of the technique is:
(1) to determine the maximum concentrations, for certain averaging times
between 1-hour and 24-hours, over a one year period due to a single point source
of up to 19 stacks, (2) to determine the meteorological conditions which cause
the maximum concentrations, and (3) to store concentration information useful
in calculating frequency distributions for various averaging times. The concen-
tration for each hour of the year is calculated and midnight - to - midnight
averages are determined for each 24-hour period.
6.. Multiple Source CRSTER (MLTCRS)
Similar to 5 above but for multiple sources.
7. Multiple Source CRSTER
Similar to .5 above, but applicable to multiple sources up to 19 colouated
elevated stack emissions.
8. Texas Climatological Model
Abstract: The Texas Episodic Model (TEM) is a short-term (10 minute to 24
hour averaging time) Gaussian Plume Model for prediction of
concentrations of nonreactive pollutants due to up to 3400 elevated
point sources and up to 200 area sources. Concentrations are
calculate for 1 to 24 scenarios of meteorological conditions, averag-
ing time, and mixing height.
10. PTMAX
Abstract: Performs an analysis of the maximum short-term concentrations
from a single point source as a function of stability and wind speed.
The final plume height is used for each computation.
11. Valley
Abstract: This algorithm is a steady-state, univariate Gaussian plume disper-
sion algorithm designed for estimating either 24-hour or annual
concentrations resulting from emissions from up to 50 (total) point
and area sources. Calculations of ground -level pollutant concentra-
tions are made for each frequency designed in an array defined by
six stabilities, 16 wind directions, and six wind speeds for 112
program -designed receptor sites on a radial grid of variable scale.
Empirical dispersion coefficients are used and include adjustments
for plume rise and limited mixing. Plume height is adjusted
according to terrain elevations and stability classes.
A-2

12. PTMTP
Abstract: Estimates for a number of arbitrarily located receptor points at or
above ground -level, the concentration from a number of point
sources. Plume rise is determined for each source. Downwind and
crosswind distances are determined for each source -receptor pair.
Concentrations at a receptor from various .sources are assumed
additive. Hourly meteorological data are used; both hourly
concentrations and averages over any averaging time from one to 24
hours can be obtained.
13. ICS
Abstract: The Industrial Source Complex dispersion models (ISC) are intended
to address complicated air quality impact analysis problems that
cannot. be adequately handled by the existing, generally available
computerized models. The ISC short-term model (ISCST) is an
extended version of the CRSTER model. The ISC long-term model
(ISCLT) is a sector -averaged model that extends and combines basic
features of AQDM and CDM. ISC accepts three source types -.
stack, area, and volume. The steady-state Gaussian plume equation
for a continuous source is used to calculate groud-level concentra-
tions for stack and volume sources. Area source contributions are
computed based on the equation for a continuous and finite cross-
wind lime source. The generalized Briggs plume rise equations,
including momentum, are used. Plume, rise is a function of
downwind distance. Procedures suggested by Huber and Snyder are
used to evaluate the effects of structure - induces aerodynamic
wakes and eddies on plume dispersion. The model has rural and
urban options. Wind speeds are adjusted from measurement height
to emission height. Terrain is accounted for by reducing the plume
centerline height by the elevation difference between source and
receptor. The model is also capable of accounting for the effects of
gravitational settling and dry deposition. The ISC Model computer
programs are written in FORTRAN IV and require approximately
65,000 words of storage.
A-3

APPENDIX B. DESCRIPTIONS OF OCS-RELATED ONSHORE FACILITIES
Bases Supporting Development and Production Operations
Support bases set up during the development or production phases of OCS
activities serve basically the same function as temporary support bases--i.e.
transferring materials and workers between shore and the offshore facilities on a
twenty-four hour seven -day -a -week basis --but on a larger scale. The main
difference is in the physical size of the base andthe intensity of support services.
Facilities required in addition to those found on an exploration support base may
include
o engineering facilities,
o food procurement and cold storage facilities,
o service base store,
o facilities for arrangement of laundry services,
o secretarial facilities,
o conference rooms,
o car rental facilities,
o security service facilities, and
o explosives storage.
Permanent support bases may be set up by, the oil companies involved or by
service companies.. The land will be either purchased or leased on a long-term
basis. In some cases a decision will be made to simply expand .the existing
temporary support base.. The company operating the base may opt to bring in other
companies such as cement companies, caterers, and other specialist suppliers as
tenants, or may act as an agent for ordering and distributing supplies.
Bases Supporting Platform and Pipeline Installation
Very similar to the support bases set up during the exploratory phase are the
bases set up to support installation of platforms and.pipelines for OCS develop-
ment. Unless a large volume of work over an extended period of time is
anticipated, these bases will be set up on a short-term basis. Their main
requirement is waterfront warehouse space and service and maintenance facilities
for vessels and barges. The companies involved in platform or pipeline installation
generally set up their own support bases. One base generally has the capacity to
support several operations at once.
If there is no space available at the oil company's service base, an attempt is
made to locate these bases in the same area or even the same port as the service
base. A site with an area of five acres is generally adequate for installation of a
pipeline or up to four platforms. The pipe for pipeline operations usually is not
stored at the base, but rather is shipped directly from the pipe coating yard to the
offshore site. For both platform and pipeline operations a minimum of 200 feet of
marginal wharf is required. An additional 200 feet is preferable for each "spread"
(fleet of ships) supporting each platform or pipeline installation operation. Road
and/or rail access to the base is essential for transporting materials into the base.
Facility descriptions from New Jersey Energy Development Study (1981) prepared
by Rogers, Golden & Halpern for NJ DEP/Division of Coastal Resources.
B-1

Platform Fabrication Yards
Platform fabrication yards are facilities at which development and produc-
tion platforms are constructed. Platforms are either made of steel and anchored
to the seabed by pilings (or cables), or made of concrete and held in place by their
own weight.
Steel platforms consist of two components —a jacket, or base, made of welded
tubular steel members, and a deck supporting modular wellhead, processing
equipment, and living quarters. After assembly, the jacket is towed to its offshore
destination. The deck is towed out separately from the jacket, either preassembled
or in sections, and installed on the jacket once it is secured in place.
Concrete platforms are hollow concrete bases with from one to four slip -
formed towers on which the deck structure is installed in protected waters near
shore. The assembled unit is towed to the offshore drilling site and lowered into
place. Because of the great depth of waterneeded for concrete platform assembly
(480-800 feet) and towing (128 feet), steel platforms will probably be used in the
Mid -Atlantic region.
Two other kinds of platforms, still in experimental stages, are the tension leg
platform and the guyed tower platform.
Steel platform fabrication yards generally are large sites on level land that
has a good load bearing capacity. For fabricating steel platforms a waterfront
location with marginal wharfs, uncrowded access to the sea, and water depths of at
least 20 to 40 feet, preferably 35 to 50 feet, are required. Steel platform
fabrication yards often occupy sites on the order of 1000 acres; however, a site as
small as 50 acres may be adequate in some instances. The size mainly depends on
the number of platforms constructed annually and the number of platform
components constructed on the site.
About 60% of the land at a steel platform fabrication yard is occupied by
storage and support uses (warehouse buildings, layout buildings, machine shop,
welding areas, sandblasting and parking areas, plate and pipe shops, and administra-
tive offices), and 40% is occupied by fabrication uses (jacket and pile fabrication
areas, deck assembly areas, and areas for fabrication of modules such as living
quarters, drilling rigs, production facilities, wellhead, water injection modules, and
helipads).
Pipe Coating Yards
Pipe coating yards supporting OCS operations are faciities that prepare pipe
joints for transporting oil and gas from offshore. The coatings consist of a mastic
coating to prevent corrosion and a concrete coating to overcome buoyancy and
protect the pipeline from mechanical damage. Pipe is delivered to the yard from
the steel mill by barge or railroad gondola car. The pipe joints are unloaded by
mobile cranes and placed in open storage. Prior to coating, the pipe joints are
cleaned and prepped, which includes shot blasting to stipple the surface to provide
good bonding surface. The pipe isprimed with an asphalt and petroleum thinner,
then coated with a bituminous mastic compound, which is applied with an extrusion
die appicator. In order to shrink the coating and produce a tight seal, the inside of
the pipe is sprayed with water. After cooling is accomplished, a whitewash of
B- 2 .

hydrated lime is applied and allowed to harden. The pipe is trimmed, inspected and
any gaps in the coating are patched. The mastic coated pipe is stacked on two sand
berms for several days, undergoes a final inspection, then goes on to the concrete
coating station. In this process, a coating of high density (140-190 lbs/cu.ft.)
concrete is applied with throwing belts or rollers as the pipes rotate past at a rate
of 6000 feet min. Wire mesh is rolled into the coating simultaneously. After the
concrete coating is applied, each joint is weighed, the ends are cleaned, and the
wire mesh is trimmed. A curing membrane is sprayed over the concrete coating
and the pipe is set out to cure for four days prior to stacking. After this initial
curing period, the coated pipe can be stacked for a 28-day curing period, after
which the pipe is loaded out on supply barges which will transport the pipe to the
lay barges that install the pipelines.
Pipe coating yards may be either portable (temporary) or permanent.
Portable plants, also called "railhead operations," are set up for short-term
contract work that can be accomplished in one season, and generally use rented
equipment and are located on leased land. A temporary facility may be converted
to a permanent facility if the projected volume of work justifies the investment.
The size of pipe coating yards reflects the need to devote large areas to open
storage. Generally pipe coating yards fall in the range of 75-100 acres, although
they may be as large as 500 acres, or as small as 30 acres. About 95% of this land
is in open storage, both for pipe and for stock piling materials such as iron ore,
sand, fiberglass wrapping material, lime, and wire mesh, while the remainder
houses the coating buildings and load out operations. About two acres is given over
to the contractor's inspection personnel and their equipment.
Pipe coating yards must be situated on a waterfront with a marginal wharf in
order to accommodate delivery of materials by barge and loading of lay barges. A
750-foot wharf will allow two barges to be loaded simultaneously. A channel depth
of 20-30 feet is needed to accommodate 30,000-ton ore -delivery barges. Easy
access to a railroad is also desirable for delivery of materials.
Pipelines
The pipelines considered in this study include onland oil and gas pipelines and
offshore oil and gas pipelines to the three-mile limit.
The oil and gas transportation strategy chosen for a given production area
depends on a number of factors such as
o total oil and gas reserves,
o distribution of reserves,
o rate of production,
o distance and route from production area to delivery point,
o water depth,
o topography, on sea or land,
o geology and soils,
o types of crossings (water bodies, road, railroad),
o land use,
o existing rights -of -way,
o environmental concerns,
o capital investment required,
B-3

0 operating costs, and
o value of oil and gas.
Natural gas generally is sold.to the gas transmission company at the platform
and the gas company builds the pipeline from the platform to its transmission line.
Normally, when the oil find is large enough to justify the high cost, a pipeline is
constructed to transport offshore oil to onshore processing and distribution
facilities. During oil pipeline construction or in commercially marginal or inac-
cessible fields, oil may be brought ashore by barges and tankers.
A single main pipeline may directly link the platform and onshore facilities or
smaller gathering lines may transport the products of several platforms to a larger,
centrally located transmission line for transport to shore. When several companies
are operating in the same general area, they are required by Federal regulations to
build a common carrier pipeline that transports their combined products. Careful
metering of the throughput at both ends ensures that each company receives its
share.
A marine pipeline system, whether for oil or gas, consists of a pressure
source (if well pressure is not sufficient), gathering lines (if used), a main pipeline,
intermediate pressure booster stations (if necessary), a landfall, and an onshore
destination.
Marine pipeline routes, landfalls, and onland pipeline routes are surveyed
once a production field is delineated. This survey process may take several years
to accomplish. Usually the marine pipeline is routed to the nearest point of land
because of the huge expense of building marine pipelines. However, other factors,
such as earthquake fault zones, bottom topography and other obstructions, exces-
sive depths, bottom currents, shifting sand dunes, environmental sensitivity, marine
activity in the area, and proximity to existing or potential onshore facility sites
may result in taking a longer route. There are three commonly used. methods of
laying a pipeline:
o lay barge method,
o reel barge method, and
o pull method.
Often marine pipelines are buried in order to provide stability and protect them
from mechanical damage. The U.S. Department of the Interior requires burial of
all pipelines in water less than 200 feet deep. After the pipeline is in place, a bury
barge is used to dig a trench beneath it. High capacity pumps furnish water at high
pressure to a jet sled that straddles the pipeline, flushing out the sea floor beneath
the pipe. The pipeline becomes buried as the mud and sand settle to the bottom or
are washed over the pipe by currents. Three other methods --mechanical cutting,
fluidization, and plowing --also may be used to bury the pipeline.
Once the pipeline location is decided on and arrangements are made for
producing and coating the pipe sections (see Pipe Coating Yards). The pipeline
installation can proceed.
The landfall is the section between the points where lay barges and
conventional onland pipe laying equipment can operate. The selection of a landfall
site involves minimizing the length of the more costly marine section, considering
B-4

the onland pipeline route, and accommodating the siting of associated onshore
facilities such as gas processing plants.
Special pipelaying methods are required in the 9andfall area, the method used
depending on whether the, pipeline crosses a wetland area, a barrier island beach -
dune system, or a beach -upland area. For wetland, crossings either the "push" or
"shove" technique or the flotation method can be used, depending on the ability of
the wetland to support the dredging equipment. In a barrier island beach -dune
system a trench is opened from the shore side out to a water depth in which a lay
barge can operate., In the beach -upland area (high energy, open coast beach areas
where the shoreline is receding) the pipeline must be buried deeply to ensure that
erosion will not uncover it. Sheet piled coffer dams and explosives might be
utilized in conjunction with the same technology used in the barrier island beach -
dune system.
An onland pipeline system consists of a main pipeline, valves, and pressure
booster stations, but may also include branch pipelines, loops, multiple mainlines,
and/or meter stations.
The onland portion of the pipeline utilizes existing rights -of -way as much as
possible. Other factors influencing pipeline route selection include topography,
geology and soils, type and number of crossings (water bodies, road and railroad),
land use, and environmental concerns.
The general procedure for laying an onland pipeline is:
o clearing and grading of right-of-way,
o trenching, generally 12" wider than pipe diameter,
o stringing and bending of pipe,
o welding and coating of pipe,
o padding (of irregular or corrosive trench bottoms with sand, gravel,
crushed rock, or screened soils),
o lowering in of pipeline,
o backfill ing, and
o revegetation.
When the pipeline crosses bodies of water, roads, or railroads, special
pipelaying techniques must be used. Stream crossings .may utilize one or more of
the following methods:
o bottom pull
o floating bridge
o floating barge, or
o directionally controlled horizontal drilling.
Road and railroad crossings may utilize several methods including
0 open cutting of roadways (unpaved or lightly travelled),
o boring under roadways (heavily travelled roads or railroads).
Pipeline right-of-way width depends on such factors as
o construction requirements,
0 operation requirements,
B-5

o maintenance requirements,
o engineering requirements,
o pipe diameter, and
o soil conditions.
The right-of-way during operation of a 24-inch diameter pipeline in loamy or rocky
soil typically will be 75 feet during construction), with a 25-foot width maintained
during operation. A 48-inch diameter pipeline typically would require 100 feet and
41 feet for construction and operation, respectively.
Once a pipeline is laid it must be tested to detect any leaks. The pipeline is
flooded with water at a certain pressure and sealed for a certain period of time,
usually 24 hours for oil pipelines and eight hours for gas pipelines. If there is a
drop in pressure over the test period, a leak must be present. All leaks must be
located and repaired before the pipeline can go into operation. Each repair job is
unique, the choice of repair method depending on a number of factors.
Oil Pipelines and Associated Pumping Stations
Oil pipelines generally are built by the oil companies producing the oil, either
on an individual or cooperative basis. Depending on the distance from shore and
the amount of activity in the production area, gathering lines may be used to
collect oil from a number of production platforms and carry it to a central point
from which a larger pipeline carries it ashore.
A pumping station at the production platform, and possibly additional
pressure boosters on platforms along the pipeline route, is required to drive the oil
onshore. If the pipeline goes an appreciable distance inland, additional pumping
stations may be needed near the landfall or along the onland pipeline route.
The number of pumping stations required will depend on the following:
o length of the pipeline,
o diameter of the pipeline,
o characteristics of fluids being transported (i.e., viscosity, specific gravity,
and whether single or two phase flow), and
o bottom characteristics along the route (i.e., slope, topography, depths).
A pumping station may require as much as 40 acres of land and will include
storage tanks, an office, and the pumping station itself.
The landfall location for oil pipelines often will be associated with other
facilities. If the oil is to be transhipped by tanker, the landfall must be located
near a site suitable for a tanker terminal and tank farm. If the oil is to continue by
onland pipe to a refinery, a pumping station may be the only facility associated
with the landfall.
An onshore oil pipeline may be very inconspicuous after completion, particu-
larly where it does not pass through a wooded area. A 50 to 100 foot right-of-way,
either purchased in fee or as an easement, is required for construction of an oil
pipeline. Revegetation can be hastened by separation of topsoil and subsoil during
B-6

excavation. It is general practice and New Jersey State policy to try to locate the
pipeline right-of-way in or parallel to existing rights -of -way of highways, power
transmission systems, railroads, other pipelines,'or similar facilities.
The diameter of oil pipelines will vary depending on the production rate of
the field, the desired operating pressure, and the bottom conditions along the
route.
Gas Pipelines and Compressor Stations
Natural gas usually is sold on the platform to a gas transmission company,
which builds the pipeline to transport the gas from the platform to its transmission
line. Like oil, gas may be collected from a number of platforms by gathering lines
and transported to a central manifold platform, where compression may be
required before it is sent ashore via a larger trunk pipeline. The onshore pipeline
will connect with the existing gas transmission network. Compression may be
required along the onshore route to maintain the desired operating pressure of the
gas. Onshore compressor stations, fueled by natural gas or refinery gas, require
from 10 to 25 acres of land located along the pipeline corridor.
The landfall for gas pipelines usually is located near an existing or potential
gas processing plant site. Gas processing plants usually are owned by the producer,
who retains the rights to the liquifiable hydrocarbons contained in the gas stream
and recovered by processing (see Gas Processing and Treatment Plants).
A 50 to 100 foot right-of-way, either purchased in fee or as an easement, is
required for construction of a gas pipeline. It is general practice and New Jersey
State policy to try to locate the pipeline right-of-way in or parallel to existing
rights -of -way of highways, power transmission systems, railroads, other pipelines,
or similar facilities.
The diameter of gas pipelines will vary depending on the production rate of
the field, the desired operating pressure, and the bottom conditions along the
route.
Gas Separation/Dehydration Plants
Gas separation and dehydration involves separating the gas, oil and free
water components of the well stream, and dehydrating the liberated gas to remove
water vapor. Free natural gas is obtained from the well stream by a separation
process, either by a two-phase process which separates gas from the rest of the
well stream or by a three-phase process which results in separate gas, oil, and free
water components. Heavy liquid hydrocarbons are settled out in one or a series of
separation vessels, each at successively lower pressures. The liberated gas passes
through a valve at the phase interface and passes on to the dehydration process.
Dehydration, by removing water vapor that remains in the gas stream after
separation, prevents the buildup of solid hydrates along the pipelines and minimizes
corrosion by the acid gases often present in the gas stream.
Most often these processing facilities are located offshore, because smaller,
less expensive pipelines can be used, less energy is required to move a single phase,
and pipeline corrosion problems can be minimized. On the other hand, locating
these facilities offshore requires that larger platforms be built, so that short
B-7

distances to shore may favor a decision to locate the facilities onshore. Both long
distances to shore and shallow water depths favor offshore separation/dehydration
while short distances and deep water favor onshore separation/dehydration. Leased
tracts of interest to New Jersey are generally 50-100 miles offshore in water
depths of 150-650 feet. Water depths less than 150 feet and offshore distances less
than 75 miles or offshore distances greater than 75 miles or offshore distances
greater than 75 miles would favor offshore partial processing (Methodologies for
OCS-Related Facilities Planning).
Oil and gas content, and fluid characteristics of the wellstream also play a
role in the siting decision. Gas separated on the platform may be used to generate
steam and electricity, or may be reinjected to maintain well pressure and increase
the oil yield.
Onshore facilities may be located at the pipeline landfall as a separate
facility, or together with a marine terminal or gas processing plant. Oil companies
generally own and operate these facilities, often as a joint venture between several
companies operating in the same area. -
A rule of thumb for the amount of land taken up by combined gas/oil onshore
partial processing facility is 15 acres for every 100,000 barrels of oil and
associated gas, but this figure is inflated because it includes facilities for partial
processing of the oil stream and treatment of the water derived from the
separation process.
Refineries
Refineries convert crude oil into many petrochemical products ranging from
motor fuels and lubricants to stocks for the chemical industry. Refineries are
usually located close to product markets such as large urban centers rather than
crude oil sources. Refineries will expand existing facilities rather than build new
ones. Recent inflation has a major bearing on this decision. The need for a
refinery near an OCS region can only be determined after well -head production
rates are known. The earliest a refinery would be built after a lease sale is about 8
years, allowing 3-4 years for the exploratory phase to establish the need for a
refinery, a year for the selection of a site, and about 3 years to construct it.
Refiners seek large sites necessarily near a deep water port or a crude oil pipeline.
Per 10,000 bbls/day capacity, roughly 22 acres are needed for refinery buildings
and processing units, 13 acres as a buffer to adjacent land uses, and 140 acres for
future expansion. The developed portion of a typical refinery site will include
processing units, storage tanks, water treatment facilities, offices, machine shop,
outdoor storage and warehouses, electrical substation, firehouse, pumping station,
truck loading areas, pipelines rail spur and parking areas. Refiners seek sites
having low environmental constraints in areas with little or no history of public
protest against facility siting. Local business taxes, state taxes and air quality
limitations may influence siting decisions. Other factors important to refinery
siting are
o access to rail
o access to well -maintained highways
o access to product pipelines
o access to utilities (electricity, water, sewerage)
o presence of support services such as machine shops, value manufacturers,
warehouses and contract maintenance.
B-8

During the three year construction phase, a 250,000, bbl/day refinery will
need an average of 1800-2200 pipefitters, welders, electricians equipment drivers
and laborers, up to 70% of whom will be hired locally if in or near an urban area.
Upon completion, such a refinery will employ 400-900 workers, most of whom will
be hired locally.
Marine Terminals
Marine terminals serve as receiving and shipping areas whenever waterborne
shipments of crude oil are involved. Terminals differ from one another in terms of
function, size, type of loading/unloading facilities and processing equipment. The
functions of a marine terminal may include
o loading crude oil received by pipeline from offshore platforms. onto
tankers for delivery to refineries,
o receiving crude oil from tankers for delivery to refineries by overland
pipelines,
o receiving crude oil from very large carriers or supertankers for delivery
to nearby refineries by pipeline, and
o receiving refined petroleum products from tankers and storing them until
overland delivery to markets.
The basic components of a marine terminal may include
o berthing capacity for vessels,
o loading/unloading equipment,
o storage tanks,
o terminal control & safety equipment,
o harbor and navigation facilities,
o partial processing equipment,
o gas processing plant,
o deballast water storage and treatment facilities,
o bunker fuel storage and loading equipment, and
o railroad facilities.
Pipe -fed Tanker Terminals
This type of marine terminal receives crude oil directly from the production
platforms by pipeline. It may be necessary to separate associated natural gas from
the oil at the terminal. Gas processing facilities may also be located at the
terminal to recover liquifiable petroleum gases from the gas. In this case storage
tanks for recovered products will be required. The natural gas may be used to fuel
the terminal's processing equipment or may be sold locally. Partial processing
facilities also may be used to remove brine water from. the crude oil prior to
storage. Tankers arriving at the terminal will be carrying ballast water which will
need to be unloaded and treated prior to disposal. Storage tanks for this purpose,
as well as oil and water separators and possibly flocculation equipment will be
provided.
Tanker -fed Receiving Terminals
Crude oil arrives at this type of terminal in tankers. The unloading tankers
typically berth at shoreside marine piers and pump their oil into surge tanks, from
B-9

which it is piped overland to the refinery. Ballast treatment facilities may not be
required since tankers will be taking on, rather than discharging, ballast water.
Bunker fuel storage and loading will be provided to fuel the tankers prior to their
departure.
Marginal or shoreside fixed pier. This type of pier is usually found inside a
harbor, and is oriented either parallel or perpendicular to the shoreline. The
orientation depends on the following:
o current in the channel or basin,
o width of harbor channel,
o availability of land, .
o availability of tugs, and
o prevailing wind direction.
The pier consists of loading platform, breasting dolphins and mooring
dolphins, connected to each other by steel truss walkways. It is connected to the
shore by a trestle, which serves as access from the shore to the loading platform
and supports all piping. The loading platform itself is built on pilings and supports
o loading arms,
o piping valves,
o surge relief tanks,
0 operation control building,
o loading arm control room,
o firefighting equipment,
o lighting towers,
o instrumentation, and
0 operating equipment.
The breasting dolphins, the structures against which the vessel berths, are
concrete or steel platforms attached by steel truss walkways to the loading
platform and by pilings to the seabed. The breasting dolphins may be either rigid
and equipped with compressible rubber fenders or other device to absorb impact
energy, or flexible and constructed of vertical flexible steel piles with a steel
platform. The berthing vessel's mooring lines are attached to mooring dolphins,
which are steel pile -supported concrete structures that transmit the tanker
mooring forces resulting from wind, wave, and sea conditions. Steel truss
walkways connect the mooring dolphins to the loading platform.
Other types of mooring facilities are discussed below.
Sea island piers. Sea island piers are fixed structures constructed basically
the same as shoreside or marginal piers, except that they are not connected to the
shore by a trestle. They are generally of concrete deck construction supported on
steel or concrete piles, and their foundations are either drilled and anchored into
rock or driven deep into the sea bottom. Submarine pipelines transport the crude
oil to shore. 'An advantage of the sea island pier design is that both sides of the
pier are available for mooring. These piers must be carefully oriented to the most
favorable wind, wave, and current. As with marginal piers, tug boats are required
for safe berthing.
Single point moorings. There are several types of single point moorings which
may be used in transferring. crude oil.from tankers to pipelines. All include as
B-10

components an anchored floating buoy, floating hoses for connecting the tanker
manifold to the buoy, and undersea hoses connecting the buoy to submarine
pipelines. The oil is transferred through the pipeline to onshore storage tanks. The
most common types of floating moorings are single point moorings. The tanker is
attached to the buoy with bow anchor lines, and is free to weathervane around the
buoy. One type of single point mooring is the single anchor leg mooring (SALM),
attached by a single anchor leg to a base on the sea floor. Universal joints in the
structure permit the hoses and base to swivel with the movement of the ship,
helping to prevent the ship from fouling the lines.
Another floating mooring, the caternary anchor leg mooring (CALM), is
anchored by six to eight pretensioned chains attached to piles driven into the sea
bottom. Vessels are moved to a turntable that allows it to swing freely to orient
into the prevailing wind, current, sea and swell.
A third single point mooring system, but one which is not free floating, is the
single point mooring pier. It consists of a floating boom mounted by a semi -
submersible floating arm to a pylon or tower that is fixed to the sea floor. The
floating arm is attached to the fixed tower by a swivel arm, allowing the moored
tanker to feather into the wind, current, and sea swell. The rigid truss structure
eliminates the problem of hose vulnerability to damage by the tanker.
Deep Water Terminals
Supertankers and very large crude carriers (VLCC's), because of their deep
draft, require corresponding deep water. Few natural harbors can meet their
requirements, so most deep water terminals are constructed offshore. They are
generally in water 100 feet or more deep and may use either a fixed sea island pier,
or a floating mooring system. The crude oil cargo is transferred by pipe to a surge
tank farm. These tank farms generally have a larger storage capacity than in other
types of marine terminals to correspond to the large size of the tankers being
accommodated. Deep water terminals are only considered in this study to the
extent that they are feasible within Raritan or Delaware Bays or the three-mile
lim it.
Product Terminals
Product terminalsreceive waterborne shipments of petroleum products from
refineries for distribution to major petroleum market areas. The vessels used for
transporting these products are smaller than crude oil carriers and have shallower
draft requirements. Storage tanks for these products are generally smaller than
those at a crude oil terminal, but a greater number of tanks will be required in
order to store the greater number of products separately. Tank trucks, rail cars, or
small coastal vessels distribute the products to the market, so the appropriate
transport system must be readily accessible.
Tank Farms
Oil storage tank farms may be located at a number of different facilities,
depending on the oil processing and transportation schemes being used. If oil is
piped to shore, tank farms may be located near the pipeline landfall, associated
either with pumping stations or with marine terminals from which the oil will be
transshipped to.refineries in other regions. In the latter case, storage tanks for
B-11

ballast water from the arriving tankers will also be provided. At marine terminals
that receive crude oil via barge or tanker, the crude is pumped into surge tanks,
from which it will be piped overland to the refinery. The use of surge tanks allows
the tankers to unload as quickly as possible, reducing berthing time. A smaller
crude pipeline can be used and a steadier pumping rate can be maintained when the
oil is pumped from storage tanks. Associated with these crude oil receiving
terminals will be bunker fuel storage and loading facilities.
The oil stored onshore, whether arriving through a pipeline or by tankers, may
undergo partial processing. The partially processed oil is pumped to a tank where
it is stored until being transported to the refinery.
Tank farms also are essential elements of refineries and petrochemical
plants, in the first case storing incoming crude oil for refining or in the latter. case
storing refined petroleum petrochemicals. The refinery products or petro chemi-
cals must be stored prior to distribution by pipeline, rail, ship, barge, or truck.
The storage tanks themselves may be one of three basic types --fixed roof
tanks, floating roof tanks, and variable vapor space tanks.
Fixed Roof Tanks
A fixed roof tank is a cylindrical steel container with a conical roof. It is
equipped with a pressure/vacuum vent to allow excess gases to escape. This is the
least expensive type of storage vessel.
Floating Roof Tanks
A floating roof tank is a cylindrical steel container with a roof that "floats"
up and down with the changing amounts of vapor in the tank. Mechanical seals seal
the space between the roof and the walls, but often a fixed roofis also provided to
further reduce gaseous emissions.
Variable Vapor Space Tanks
This variable vapor space tank is designed to respond to the amount of vapor
present above the liquid surface. Unlike the floating roof design, it is equipped
with a diaphragm or "lifter" roof. It is used for storing petroleum products other
than crude oil.
The size of a tank farm depends on a number of factors, including
o production rate of the field,
o throughput rate of the pipeline serving the terminal,
o size of the tankers using the facility,
o number of berths,
o arrival frequency of the tankers,
o number of days necessary to store the oil should transshipment to tankers
be interrupted, and
o whether or not it is necessary to separate crude produced at different
fields.
The following table indicates the approximate land requirements for tank
farms of various sizes.
B-12

APPROXIMATE LAND REQUIREMENTS FOR TANK FARMS
Tank Farm Capacity Land
barrels (acres)
1,000,000 17
2,000,000 37
3,000,000 50
3,500,000 58
Source: Arthur D. Little, Petroleum Development in New England, 1975, Vol. II,
p. IV-7.
Gas Processing and Treatment Plants
Natural gas as it comes from the ground is often associated with other
substances. Assuming commercial quantities of offshore gas are found, some
separation will probably be done offshore (see Gas Separation/Dehydration Plants).
If gas is associated with oil or water, it will be separated and dehydrated. If it is
not associated with oil or water, it will simply be piped ashore.
When the gas reaches shore, it will be further treated and purified and then
repressurized for transmission to market. If the gas is rich in liquifiable
hydrocarbons, such as propane and butane, it may be economically attractive to
separate them. If the gas arriving from the production platform is "sour" gas, or
has a high sulfur content, then it must be "sweetened," or treated to remove the
sulfur compounds. Hydrogen sulfide, a product of this treatment, may be
converted to elemental sulfur in a sulfur recovery plant and sold commercially. If
only a small amount of hydrogen sulfide is recovered, it may be flared or inciner-
ated.
Generally the processing facility is built by the producer, who retains the
rights to the liquifiable hydrocarbons derived from the processing, while the gas
transmission company to which the gas has been sold constructs the gas pipeline.
Gas processing and treatment facilities are designed for the specific gas
stream being received. Thus, there is a great deal of variation in size and design of
these facilities. The throughput capacity typically ranges from two million
cu.ft./day to two billion cu.ft./day. Three processes commonly used to recover
natural gas liquids are
o lean oil absorption,
o mechanical refrigeration, and
o cryogenic refrigeration.
(These processes are described in the NERBC's Factbook.) The richness, or amount
of liquifiable hydrocarbons in the gas stream, and the desired rate of recovery
determine which process is used.
B-13

While the gas processing facilities in themselves do not require a lot of space,
a large amount of land is required as a buffer for safety reasons. For example a
plant with a billion cu.ft./day throughput may require a 75 acre site, of which only
20 acres is occupied by buildings and structures. Other land uses include areas for
o loading,
o storage,
o offices,
o parking,
o communications,
o compressors, and
o generators.
The site may either be coastal or inland, but must be located between the pipeline
landfall and the commercial transmission lines into which the processed gas will be
delivered.
B-14

APPENDDC C. THE STATUS OF EMISSIONS INVENTORIES
IN THE CAPE FEAR REGION
The following documents relating to air emissions inventories were reviewed for
completeness and consistency:
o A print-out titled "Emissions Report by Plant for North Carolina", consisting
of 20 pages and dated 8/1/80. (Report A)
o A print-out of computerized EPA Region IV EIS data AQDM Format. (Report
B)
o Directory of Manufacturers and Processors in the Greater Wilmington Area,
January 1979.
o Working List of Air Pollution Sources in Brunswick, Columbus, New Hanover,
and Pender County. Provided by DNR&CD (Wakild to Reilly, May 15, 1980).
The computerized emissions inventories provided by the Cape Fear Council of
Governments have been evaluated for adequacy as required input to a regional air
quality modeling program. In general, most major sources, as cross-referenced in the
Directory of Manufacturers and Processors in the Greater Wilmington Area, appear to
be covered when both print-outs are considered. It appears that the information
provided is generally adequate and could.readily be utilized to initiate start-up and
check-out procedures for dispersion modeling.
We would like to point out some difficulties in these data. For instance, Report
A contains information on five criteria pollutants (TSP, SO2, NOx, HC, CO), while
Report B covers only two pollutants (TSP, SO2). Report A contains sources that are
not covered in Report B and vice -versa. This may be due to the time frame in which
these data were collected or updated. It would be quite useful, however, if all the
information could be provided in one single database.
It is possible that some or all of this required information is actually in residence
in DNR&CD's data bank but was simply not included on the print-out we received. It
is also possible, based on earlier telephone conversations with DNR&CD, that the
required information exists in "hard copy" files and has not been entered into the data
bank as of yet.
Based on our review of the above -mentioned documents, we note the following:
o Source sizes listed appear to be those emitting one ton per year or more of
any of the. five major pollutants (TSP, SO2, NOx, CO, HC).
C-1

o In Report A,* there are 42 sources of one ton/year or more emissions listed
for county code 2880 (New Hanover County). Twelve sources are listed for
county 0460 (Brunswick County), fifteen sources for county 0880 (Columbus
County), and six sources for county 3100 (Pender County).
o In Report B,* the following sources of one ton/year or more were found:
Fourteen sources are listed for county code 0460 (Brunswick), ten sources for
county 0880 (Columbus), forty-six sources for county 2880 (New Hanover),
and five sources for county 3100 (Pender).
o Emissions are defined only in terms of mass per year. Shorter term maxima
or seasonal variations are not reported.
o Some information is missing, such as UTM coordinates, flow rates, etc. for
some major sources, e.g., point sources 13-16 for DuPont Cape Fear Plant.
Wright Chemical Corporation in Riegelwood emitting 93 tons of SO2 has no
flow rate or stack height given for its point source A.
Comparison of the above characteristics of the inventory with criteria estab-
lished in Chapter 2 reveals the following findings:
Criteria: The emissions inventory must treat the particular pollutants which will
likely be emitted by new energy -related facilities.
Finding: The inventory as found on the print-out is adequate with respect to the
types of pollutants quantified except for lead. Lead is a criteria pollutant and
may be an important consideration should a major facility, such as a coal-fired
facility, or facilities have to be evaluated.
Criteria: The information obtained must be of sufficient spatial and temporal
resolution to permit siting decisions through use of air quality dispersion
modeling.
Finding: The information is sufficient to perform predictions of annual average
pollutant concentration due to major point sources. However, the information
presented is of insufficient temporal resolution to be of use in determining
shorter term (seasonal, monthly) air quality conditions. Emissions of many
sources vary seasonally and this is not shown.
*Note: The number of sources in Reports A and B are actually the number of
facilities. Each facility may have more than one point source listed. Some
listed facilities have no point sources listed for TSP or SO but they may
have other contaminant emissions which do not appear on2the Region IV
inventory.
When we compared the State inventory print-out (Report A) to the Region
IV inventory (Report B), the latter one lists two additional facilities in
Brunswick County, five less in Columbus County, four more in New
Hanover, and one less in Pender County.
C-2

In addition, the following information should also be made available:
o Name of contact at source and telephone number;
o Year and month of information receipt or update.
Regarding the general content of the inventory, two observations are made.
First, no area source emissions data have been received. These data may not currently
be in computerized form but they must be available if dispersion modeling of the COG
region is to eventually be performed. Second, there appear to be possible omissions in
the point source inventory for the four county area. For example, in Columbus
County, comparison of the listings in the "Directory of Manufacturers and Processors
in the Greater Wilmington Area" (published by the Greater Wilmington Chamber of
Commerce and dated January 1, 1979) with the emissions inventory listing reveals the
following areas of concern:
o Omission of Allied Chemical in Riegelwood
o Omission of Kaiser Agricultural Chemicals in Riegelwood
o. Apparent omissions of relatively large employers, textile industries such as
Blue Jeans Corporation in Whiteville and Ithaca-Chadbourn in Chadbourn,
both in Columbus County. Based on the number of employees, these two
manufacturers may emit more than one ton of SO2 per year for space heating
purposes unless natural gas is used, in which case, NO emissions may exceed
one ton per year.
Other potential omissions include France Neckwear Company in Wilmington (An
employer of between 500 and 1,000 people) and Maritime Lumber Service (kiln drying)
also located in Wilmington.
Recommendations
It is recommended that the possible deficiencies identified in the emissions.
inventory be investigated with the DNR&CD at a later date prior to final use of data
in a regional modeling effort. As mentioned previously, much of the required data
may be available but not included on the print-out.
Area source emissions data must be available for meaningful dispersion model -
ing. -
In the case of the possible omission of sources, it is suggested that the specific
sources mentioned above as not being included in the inventory be surveyed on an
informal basis to determine the propriety of their omission. Given the COG's intimate
knowledge of the local industry, this could be accomplished without excessive effort.
If the omission of any of these sources in this "sample" of potentially omitted sources
is found to be unwarranted, the deficiency should be brought to the knowledge of the
DNR&CD for correction and the inventory should be reviewed in detail for other
obvious omissions.
C-3

CEIP Publications
1. Hauser, E. W., P. D. Cribbins, P. D. Tschetter, and R. D. Latta.
Coastal Energy Transportation Needs to Support Major Energy Projects
in North Carolina's Coastal Zone. CEIP Report #1. September 1981.
$10.
2. P. D. Cribbins. A Study of OCS Onshore Support Bases and Coal Export
Terminals. CEIP Report #2. September 1981. $10.
3. Tschetter, P. D., M. Fisch, and R. D. Latta. An Assessment of
Potential Impacts of Energy -Related Transportation Developments
on North Carolina's Coastal Zone. CEIP Report #3. September 1981.
$10. (Available spring 1982)
4. Cribbins, P. S. An Analysis of State and Federal Policies Affecting
Major Energy Projects in North Carolina's Coastal Zone. CEIP Report #4.
September 1981. $10.
5. Brower, David, W. D. McElyea, D. R. Godschalk, and N. D. Lofaro.
Outer Continental Shelf Development and the North Carolina Coast:
A Guide for Local Planners. CEIP Report #5. August 1981. $10.
6. Rogers, Golden and Halpern, Inc., and Engineers for Energy and the Environment,
Inc. Mitigating the Impacts of Energy Facilities; A Local Air Quality Program
for the Wilmington, N.C. Area. CEIP Report #6. September 1981. $10.
7. Richardson, C. J. (editor). Pocosin Wetlands: an Integrated Analysis
of Coastal Plain Freshwater Bogs in North Carolina. Stroudsburg (Pa):
Hutchinson Ross. 364 pp. $25. Available from School of Forestry,
Duke University, Durham, N. C. 27709. (This proceedings volume is
for a conference partially funded by N. C. CEIP. It replaces the
N. C. Peat Sourcebook in this publication list.)
8.
McDonald, C.
B., and A. M. Ash.
Natural Areas Inventory of Tyrrell
County, N. C.
CEIP Report #8. October 1981. $10 for all requests.
9.
Fussell, J.,
and E. J. Wilson.
Natural Areas Inventory of Carteret
County, N. C.
CEIP Report #9.
October 1981. $10 for all requests.
10.
Nyfong, T. D.
Natural Areas Inventory of Brunswick County, N. C.
CEIP Report #10.
October 1981.
$10 for all requests.
11.
Leonard, S. W., and R. J. Davis.
Natural Areas Inventory for Pender
County, N. C.
CEIP Report #11.
October 1981. $10 for all requests.
NOTE: Please note renumbering of reports 5-10.