-------
o
c
.2
'o
Watering Control
Efficiency Estimates
100%
95%
75% I
LLJ
"5
i.
*-
C
O
O
(0
13
O
O
c
2
c
50% -
25% -
Ratio of Controlled to Uncontrolled
Surface Moisture Contents
Figure 7-5. PM10 control efficiency for watering unpaved surfaces.
7-27
-------
Furthermore, this relationship is applicable to all size ranges con-
sidered:
[75 (M-l) 1 < M < 2 (7_n)
(62 + 6.7 M 2 < M * 5
where c = instantaneous control efficiency (%)
M = ratio of controlled to uncontrolled surface moisture contents
(dimensionless)
Costs for watering programs include the following elements:
Capital: Purchase of truck or other device
O&M: Fuel, water, truck maintenance, operator labor
Reference 2 estimates the following costs (1985 dollars):
Capital: $17,100/truck
O&M: $32,900/truck
The number of trucks required may be estimated by assuming that a single
truck, applying water at 1 L/m2, can treat roughly 4 acres of unpaved
surface every hour.
7.4.5 Paved Surface Cleaning
Other than housekeeping, the only method available to reduce the
surface loading of fine particles on paved surfaces is through some form
of street cleaning practice. The three major methods of street cleaning
are: mechanical (broom) cleaning; vacuum cleaning; and flushing. Broom
sweeping itself does remove some debris from the pavement thus preventing
it from becoming airborne by the action of passing vehicles; but it can
also generate significant amounts of finer particulate by the mechanical
action used to collect the material. Thus, broom sweeping without prior
water flushing is not normally recommended for removal of fine particu-
late from paved surfaces.
Measurement-based efficiency values for paved surface control
methods are presented in Table 7-3.2 Note that all values in this table
are for mitigative measures applied to industrial paved roads and reflect
reductions in PM-15 not PM10. It can be assumed, however, that similar
reductions in gross PM10 may be expected for the surface cleaning methods
listed in Table 7-3.
7-28
-------
TABLE 7-3. MEASURED EFFICIENCY VALUES FOR PAVED SURFACE CONTROLS3
Method
Cited
efficiency
Comments
Vacuum sweeping Q%-58%
Water flushing
Water flushing
followed by
broom sweeping
46%
69-0.231 Vc'd
96-0.263 Vc'd
Field emission measurement (PM-15)
12,000-cfm blower0
Based on field measurement of
30 urn particulate emissions
Field measurement of PM-15 emissions0
Field measurement of PM-15 emissions0
aFrom Reference 2. All results based on measurements of air emissions from
industrial paved roads.
PM-10 control efficiency can be assumed to be the same as that tested.
cWater applied at 0.48 gal/yd2.
Equation yields efficiency in percent, V = number of vehicle passes since
application.
7-29
-------
Cost elements Involved with vacuum sweeping include the following
capital and O&M expenses:
Capital: Purchase of truck or other device
O&M: Fuel, replacement parts, truck maintenance, and operator labor
co%ts
Data presented in Reference 2 provides the following estimates for a
vacuum sweeping program in April 1985 dollars:
Initial capital expense: $36,800/truck
Annual O&M expense: $34,200/truck
Determination of the number of trucks necessary can be made by assuming
that one unit can sweep 6 mi/12 h.2
Finally, cost elements involved with flushing in combination with
broom sweeping include the following capital and O&M expenses:
Capital: Purchase of truck or other device
O&M: Fuel, replacement parts (possibly including brushes), truck
maintenance, operator labor, water
Cost data presented in Reference 2 provides the following estimates
for a flushing program:
Initial capital expense: $18,400/truck
Annual O&M expense: $27,600/truck
All costs are based on April 1985 dollars. Determination of the number
of trucks required can be based on the assumption that 3 to 5 mi can be
flushed or broom swept per unit per 8-h shift, respectively.2
7.5 PROCEDURES FOR COMPLIANCE DETERMINATION
In this section, two basic procedures will be discussed for the
determination of dust control compliance for waste stabilization pro-
cesses. These procedures are: permit systems; and indirect measures of
control performance. Each is addressed below.
7.5.1 Permit Systems
The first approach to compliance determination involves the
implementation and enforcement of a permit program for waste stabiliza-
tion processes. This would generally be accomplished as part of the RCRA
Part B permit for the site issued by the appropriate regulatory agency.
7-30
-------
A permit system would require the site owner or operator to file a spe-
cific dust control plan with the appropriate regulatory agency having
jurisdiction.
As part of the permit application, record keeping should be one of
the main conditions for approval. Records of site activity and control
should be submitted to the regulatory agency on a monthly basis. These
records must be certified by a responsible party as to their completeness
and accuracy. All site records should be permanently maintained by the
agency.
To enforce the dust control plan submitted as part of the permit
application, field audits of key control parameters should be made by
regulatory personnel. The results of these audits would then be compared
to site records for that period to determine compliance with permit con-
ditions. If differences are found between application of the control(s)
observed on-site and those recorded by site operating personnel, this
would constitute a violation and would be grounds for further enforcement
action. This approach is, however, predicated on the fact that strict
implementation of the dust control plan will achieve certain reductions
in PM,0 emissions associated with site operation. General information to
be included in a dust control plan, specific operational records, and
general records to be kept by the source for each of the control measures
discussed in Section 7.4 are provided below.2
7.5.1.1 Wind Fences/Barriers.
General Information
1. Locations of all raw materials storage and handling operations
to be controlled with wind fences referenced on a plot plan available to
the operator and regulatory personnel.
2. Physical dimensions of each source to be controlled and
configuration of each fence or screen to be installed.
3. Physical characteristics of material to be handled or stored for
each operation to be controlled by fence(s) or screen(s).
4. Applicable prevailing meteorological data (e.g., wind speed and
direction) for site on an annual basis.
7-31
-------
Specific Operational Records
1. Date of installation of wind fence or screen and initials of
installer.
2. Location of installation relative to source and prevailing
winds.
3. Type of material being handled and stored and physical
dimensions of source controlled.
4. Date of removal of wind fence or screen and initials of
personnel involved.
General Records
1. Fence or screen maintenance record.
2. Log of meteorological conditions for each day of site operation.
7.5.1.2 Capture/Collection Systems.
General Information
1. Locations of hood/enclosure, ductwork, and dust collector
referenced on a plot plan available to the responsible party and regu-
latory personnel.
2. Plans, drawings, and specifications for storage silo(s),
capture/collection system (including all available test data), and
ancillary equipment.
3. Physical characteristics of all materials to be transferred to
silos and stabilization process.
4. If a scrubber is used, source of water and chemical additives to
be used, if any.
5. Location and type of permanently mounted instrumentation for
monitoring operation of capture/collection system.
Specific Operational Records
1. Date of operation and operator's initials.
2. Start and stop time of dust control equipment.
3. Notations of malfunctions and corrective actions taken including
date and time of each.
4. Type of material and number of loads or weight transferred to
silo(s) and/or process between start and stop time.
5. Notations of any visible emissions from dust collector stack or
vent including date and time of each event.
7-32
-------
General Records
1. Equipment maintenance records.
2. Records of equipment malfunctions and downtime.
3. Purchase orders, etc., of maintenance supplies and replacement
parts (e.g., bags, fan motors, etc.).
7.5.1.3 Met Suppression.
General Information
1. Locations of all stabilized waste storage and handling opera-
tions referenced on plot plan of the site available to the operator and
regulatory personnel.
2. Materials transport flow sheet which indicates the type of
material (if variable), its handling and storage, size and composition of
storage piles, etc.
3. The method and application intensity of water, etc., to be
applied to the various materials and frequency of application, if not
continuous.
4. Dilution ratio for chemicals added to water supply, if any.
5. Complete specifications of equipment used to handle the various
materials and for wet suppression.
6. Source of water and chemical(s), if used.
Specific Operational Records
1. Date of operation and operator's initials.
2. Start and stop time of wet suppression equipment.
3. Location of wet suppression equipment.
4. Number of loads (or other measure of throughput) loaded between
start and stop time.
5. Start and stop times for tank filling.
General Records
1. Equipment maintenance records.
2. Meteorological log of general conditions.
3. Records of equipment malfunctions and downtime.
7-33
-------
7.5.1.4 Watering of Unpaved Surfaces.
General Information
1. All travel routes to be treated references on a plot plan
available to both the site operator and regulatory personnel.
2. Length and area of surfaces to be watered.
3. Application Intensity (L/m*) and frequency (a minimum moisture
content may be specified as an alternative).
4. Type of application vehicle, capacity of tank, and source of
water.
Specific Records to Be Kept by Site Operator
1. Equipment maintenance log.
2. Meteorological log of general conditions (e.g., sunny and warm
vs. cloudy and cold).
3. Records of equipment breakdowns and downtime.
An example permanent record form which may be used to record the
above information is shown in Figure 7-6.
Specific Records to Be Kept by Truck Operator
1. Date and time of treatment.
2. Equipment used (this should be referred back to dust control
plan specifications).
3. Operator's initials (a separate operator's 'log may be kept and
transferred later to permanent records by site operator).
4. Start and stop time, average speed, and number passes.
5. Start and stop time for filling of water tank.
7.5.1.5 Vacuum Surface Cleaning.
General Information
1. All road segments (and parking areas, if applicable) referenced
on a map available to both the responsible party and the regulatory
agency.
2. Length of each road (and area of each parking area).
3. Type of control applied to each road/area and planned frequency
of application.
4. Any provisions for weather (e.g., % in of rainfall will be
substituted for one treatment; etc.).
7-34
-------
S3
7-35
-------
Specific Records for Each Road Segment/Parking Area Treatment
1. Date of treatment.
2. Operator's initials (note that the operator may keep a separate
log whose information is transferred to the environmental staff's data
sheets).
3. Start and stop times on a particular segment (or parking area),
average speed, number of passed.
4. Qualitative description of surface loading before and after
treatment.
5. Any areas of unusually high loadings, from spills, pavement
deterioration, etc.
General Records
1. Equipment maintenance records.
2. Meteorological log (to the extent that weather influences the
control program—see above).
3. Any equipment malfunctions or downtime.
7.5.1.6 Flushing/Broom Surface Cleaning.
General Information
1. All road segments (and parking areas, if applicable) referenced
on a map available to both the responsible party and the regulatory
agency.
2. Length of each road (and area of each parking area).
3. Type of control applied to each road/area and planned frequency
of application.
4. Provisions for weather (e.g., program suspended for periods of
freezing temperatures).
Specific Records for Each Road Segment/Parking Area Treatment
1. Date of treatment.
2. Operator's initials (note that the operator may keep a separate
log whose information is transferred to the environmental staff's data
sheets).
3. Start and stop times on a particular segment (or parking area),
average speed, number of passes.
4. Start and stop times for refilling tanks.
7-36
-------
5. Qualitative description of surface loading before and after
treatment.
6. Any areas of unusually high loadings, from spills, pavement
deterioration, etc.
General Records
1. Equipment maintenance records.
2. Meteorological log (to the extent that weather influences the
control program—see above).
3. Any equipment malfunctions or downtime.
7.5.2 Indirect Measures of Control Performance
The final compliance format to be presented relates to various
indirect measures of control performance. These could be used in con-
junction with or in lieu of permit enforcement discussed above. They
will, however, require more effort and expense to implement but should be
at least somewhat defensible as measures of control efficiency.
The most obvious approach to indirectly measuring control
performance involves the collection and analysis of material samples from
the various dust-emitting sources. For example, if some form of paved
surface cleaning is performed, collection of surface samples for silt
content would indicate the efficacy of control for this particular
source. The silt loadings obtained could be compared with "typical" sur-
face loading values for similar uncontrolled surfaces to determine the
degree of loading (and thus emissions) reductions achieved. This would,
of course, necessitate the availability of a data base of "uncontrolled"
silt loadings for comparison with site-specific data. The silt samples
could also be chemically analyzed to determine the degree of contamina-
tion control being achieved.
Another indirect measure of control compliance is the collection and
analysis of material samples from unpaved surfaces and stabilized waste
storage and handling operations. In this case, analysis of the moisture
content of these samples would indicate the amount of water applied and
thus the degree of control achieved by wet suppression. Appropriate
equations (presented above) would be used to determine control efficiency
7-37
-------
based on the sample data. Appropriate sampling and analysis methods to
implement the above approach are included in Appendix D.
7.6 EXAMPLE CALCULATION
An example application of the emission estimation and control
performance procedures for waste stabilization units is presented
below. Although this example focuses on contaminant-specific (i.e., Cr)
emissions, procedures are also provided for gross PMlo emissions as well.
7.6.1 Process Description
A commercial facility is located in a semiarid climate with 320 dry
days per year. The facility operates 5 days per week (261 d/yr), 8 h/d,
and processes the following listed waste streams likely to contain appre-
ciable concentrations of chromium:
Waste Stream Annual Receipts (Mg)
K001 1,000
F006 5,000
The characteristics of the waste are specified as:
Cr Concentration
Waste Stream (ppm by wt)
K001 200
F006 500
In the stabilization process, power plant fly ash is added to the liquid
waste in sufficient quantities to obtain a 50% (weight) solids content in
the stabilized waste. Fly ash is added to the liquid waste from elevated
silos with the process equipment exposed to ambient winds. The stabi-
lized waste is discharged directly into roll-offs which are carried by
trucks across a 0.25 mile (0.40 km) of unpaved surface to the main plant
road. Liquid waste is also delivered by trucks traveling across this
same unpaved surface. A capture/collection system is used for control of
raw materials transfer and handling and watering is used for control of
vehicle-generated dust.
7-38
-------
Since no site-specific S&A has been conducted, contamination of unpaved
travel surfaces is assumed to be similar to that shown in Table 7-2.
Silt content of the unpaved surface material is assumed to be identical
to that shown in Table 7-1. All calculations are for annual average
emissions rounded to two significant figures.
Based on facility records the following extent values are calculated:
Throughput of raw material:
(1,000 + 5,000) M
Delivery of raw material: Assuming 41 Mg capacity tractor trailers
6-000
Liquid waste delivery: Assuming 10,000 gal (37,850 L) capacity tank
trucks and 8 Ib/gal (0.96 kg/L) specific weight
37,850 ±-3 x 0.96 £3 x . * "9. = 36
load L 1,000 kg load
6,000 M(? waste x yg^ x 0.8 J9D- = 130 km/yr
yr 3o Mg load
Stabilized waste haulage: Assuming 5.5 Mg load capacity
(6,000 fly ash + 6,000 liquid waste) M9 waste x I ]oa.d x 0.8 J2L. = 1700 km/y
yr 3.0 Mg load
Total truck traffic:
(120 + 130 + 1700) km/yr = 1950 km/yr
7.6.2 Raw Material Handling
Assuming: Fly ash is pneumatically conveyed from trucks to silo(s).
Silo and discharge are ventilated to a baghouse dust collector
(99.9% collection efficiency = 0.999)
The capture efficiency of the ventilated discharge is 95%
(0.95).
The emissions are uncontaminated (no RCRA metals).
7-39
-------
UE10 = 0.00085 k9 PM'10 x 6,000 M9 f1v ash x 2 transfers
* "
The uncontrolled emissions are calculated:
PM
* 'y
= 10 kg/yr uncontrolled PM-10 (or 5.1 kg/yr for each operation)
The controlled emissions are calculated:
Load- in— 0.00085 £ x 6,000 ^ x (1-0.999) * 0.0051
v '
PM-10
, . .
Mg yr v ' yr
(assumes 100* capture)
Load-out— [5.1 x (1-0.95)] ^3 + [5.1 x 0.95 x (1-0.999)] ^3 = 0.26 ^
yr v " yr yr
(not captured) (not collected)
Total — (0.26 + 0.0051) k3 PM"10 = 0.26 k3 PM~1Q * CE10
' yr yr 10
7.6.3 Vehicle Traffic on Unpaved Surfaces
Assuming: The gross weight of the delivery trucks is 59 Mg (65 tons).
The gross weight of the disposal trucks is 12 Mg (13 tons).
The vehicle speed is 24 km/h (15 mph).
Watering of unpaved surfaces is used at an application
intensity of 2 L/m2.
The annual vehicle traffic is 3000 vehicles/yr.
The uncontrolled emissions are calculated:
Waste and raw material delivery = 1,950 km/yr (total) - 1,700 km/yr
(stabilized waste)
= 250 km/yr delivery traffic
Using the fractional proportion of waste and raw material traffic to
total traffic (Section 7.6.1):
Average vehicle weight = ^^ x 59 Mg + f4^ x 12 Mg
(weighted based on vol.) iybu iyb0
= 7.6 + 10 = 18 Mg
Average number of wheels = T||§ x 18 + i4^ x 10
(weighted based on vol.) iysu iytju
= 2.3 + 8.7 = 11 wheels
7-40
-------
Therefore:
s /Uv0.7 0.5
UECr * o x 0.612 x -r| x ^! x (=£=) x (J) x ^§|iE x source extent
= .67 x 0<612 x 1| x 24 x / 18\ * x /11\ x |f° x 1,950 veh-km/yr
1(11);* if. «+o \f.tll \ 4/ jOO
= 0.33 *a_£r. (or 4,900 kg total PM-10)
yr v yr '
The controlled emissions are calculated assuming watering is conducted
every 4 h (t = 4):
For e = mean annual pan evaporation (from Figure 7-4) = 90 in
p = potential average hourly daytime evaporation rate (from Eq.
7-7a)
= 0.0049 e = 0.0049 (90) = 0.44 mrn/h
and d = average hourly traffic rate
= 3000 Vehic1es x .LSI- x 14 = 1.4 veh,!c1es
yr 261 d 8 h h
Thus C = average control efficiency = 100 - °*8 P d t (from Eq. 7-6)
= 100 . |r0.8(0.44)g.4)(4)1 = gg% = Oi9g
Therefore:
CECr = 0.33 ^J£ x (1 - 0.99) = 0.0033 ^LJ> or
CElo = 4,900 J«L^ilO x (i _ Q.99) = 49 k^ ^10 total emissions
7-41
-------
7.7 REFERENCES FOR SECTION 7
1. Malone, P. G., L. W. Jones, and R. J. Larson. Guide to the Disposal
of Chemically Stabilized and Solidified Waste. SW-872, U.S.
Environmental Protection Agency. Cincinnati, OH. September 1982.
2. Cowherd, C., G. E. Muleski, and J. S. Klnsey. Control of Open
Fugitive Dust Sources. EPA-450/3-88-008. U.S. Environmental Pro-
tection Agency, Research Triangle Park, NC. September 1988.
3. Englehart, P., and D. Wallace. Assessment of Hazardous Waste TSDF
Particulate Emissions. Final Report. EPA Contract No. 68-02-3891,
Assignments 5 and 13. U.S. Environmental Protection Agency,
Research Triangle Park, NC. October 1986.
4. U.S. Environmental Protection Agency. Compilation of Air Pollution
Emission Factors, AP-42. U.S. Environmental Protection Agency,
Research Triangle Park, NC. 1988.
5. Muleski, G. E., F. J. Pendleton, and W. A. Rugenstein. Measurement
of Fugitive Emissions in a Coal-Fired Power Plant. Proceedings:
Sixth Symposium on the Transfer and Utilization of Particulate Control
Technology, EPRI CS-4918, Electric Power Research Institute. Palo
Alto, CA. November 1986.
6. Cowherd, C., Jr., and J. S. Kinsey. Identification, Assessment, and
Control of Fugitive Particulate Emissions. EPA-600/8-86-023. U.S.
Environmental Protection Agency, Research Triangle Park, NC. August
1986.
7. Studer, B. J. B., and S. P. S. Arya. Windbreak: Effectiveness for
Storage Pile Fugitive Dust Control: A Wind Tunnel Study. Journal of
the Air Pollution Control Association. 38:135-143. 1988.
8. Ohio Environmental Protection Agency. Reasonably Available Control
Measures for Fugitive Dust Sources. Columbus, OH. September 1980.
7-42
-------
APPENDIX A.
OPEN DUST SOURCE EMISSION FACTOR RATING AND CONTROL
EFFICIENCY TERMINOLOGY
-------
APPENDIX A. OPEN DUST SOURCE EMISSION FACTOR RATING AND CONTROL
EFFICIENCY TERMINOLOGY
A.I EMISSION FACTOR RATING TERMINOLOGY
In AP-42, the reliability of emission factors is indicated by an
overall Emission Factor Rating ranging from A (excellent) to E (poor).
These ratings take into account the type and amount of data from which the
factors were calculated. Note that measurements underlying each emission
factor are rated on a similar scale of A to D.
The use of a statistical confidence interval may seem desirable as a
more quantitative measure of the reliability of an emission factor.
Because of the way an emission factor data base is generated, however,
prudent application of statistical procedures precludes the use of
confidence intervals unless the following conditions are met:
• The sample of sources from which the emission factor was
determined is representative of the total population of such
sources.
• The data collected at an individual source are representative of
that source (i.e., no temporal variability resulting from source
operating conditions could have biased the data).
• The method of measurement was properly applied at each source
tested.
Because of the almost impossible task of assigning a meaningful confidence
limit to the above variables and to other industry-specific variables, the
use of a statistical confidence interval for an emission factor is not
practical.
The following emission factor ratings are applied to the emission
factors:
A - Excellent. Developed only from A-rated test data taken from many
randomly chosen facilities in the industry population. The source
category is specific enough to minimize variability within the source
category population.
B - Above average. Developed only from A-rated test data from a
reasonable number of facilities. Although no specific bias is
evident, it is not clear if the facilities tested represent a random
A-l
-------
sample of the industry. As in the A-rating, the source category is
specific enough to minimize variability within the source category
population.
C - Average. Developed only from A- and B-rated data from a
reasonable number of facilities. Although no specific bias is
evident, it is not clear if the facilities tested represent a random
sample of the industry. As in the A rating, the source category is
specific enough to minimize variability within the source category
population.
D - Below average. The emission factor was developed only from A-
and B-rated test data from a small number of facilities, and there
may be reason to suspect that these facilities do not represent a
random sample of the industry. There also may be evidence of
variability within the source category population. Limitations on
the use of the emission factor are footnoted in the emission factor
table.
E - Poor. The emission factor was developed from C- and D-rated test
data, and there may be reason to suspect that the facilities tested
do not represent a random sample of the industry. There may be
evidence of variability within the source category population.
Limitations on the use of these factors are always footnoted.
Because the application of these factors is somwhat subjective, the
reasons for each rating are documented in the background files maintained
by the Office of Air Quality Procedures and Standards (OAQPS).
A.2 CONTROL EFFICIENCY TERMINOLOGY
Some control techniques often used for open dust sources begin to
decay in efficiency almost immediately after implementation. The most
extreme example of this is the watering of unpaved roads where the
efficiency decays from nearly 100 percent to 0 in a matter of hours (or
minutes). The control efficiency for broom sweeping and flushing applied
in combination on a paved road may decay to zero in 1 or 2 days. Chemical
dust suppressants applied to unpaved roads can yield control efficiencies
that will decay to zero in several months. Consequently, a single-valued
control efficiency is usually not adequate to describe the performance of
most intermittent control techniques for open dust sources. The control
A-2
-------
efficiency must be reported along with a time period over which the value
applies. For continuous control systems (e.g., wet suppression for
materials transfer), a single control efficiency is usually appropriate.
Certain terminology has been developed to aid in describing the time
dependence of open dust control efficiency. These terms are:
1. Control lifetime is the time period (or amount of source
activity) required for the efficiency of an open dust control measure to
decay to zero.
2. Instantaneous control efficiency is the efficiency of an open
dust control at a specific point in time.
3. Average control efficiency is the efficiency of an open dust
source control averaged over a given period of time (or number of vehicle
passes).
From the above definitions, it is clear that average control
efficiency is related to instantaneous control efficiency by the following
general equation:
C(X) = x 0. In that case, average control
efficiency is given by
c(X) « a x
A-3
-------
APPENDIX B
ESTIMATION OF CONTROL COSTS AND COST EFFECTIVENESS
B-l
-------
Relative costs of alternative control measures are used in the
development and evaluation of particulate fugitive emissions control
strategies. Cost analyses are used by control agency personnel to
develop overall strategies or to evaluate plant specific control strate-
gies. Industry personnel perform cost analyses to evaluate control
alternatives for a specific source or to develop a iplant-wide emissions
control strategy. Although the specifics of these analyses may vary
depending upon the objective of the analysis and the availability of cost
data, the general format is similar.
The primary goal of any cost analysis is to provide a consistent
comparison of the real costs of alternative control measures. The objec-
tive of this section is to provide the reader with a methodology that
will allow such a comparison. It will describe the overall structure of
a cost analysis and provide the resources for conducting the analyses.
Because cost data are continuously changing, specific cost data are not
provided. However, sources of cost information and mechanisms for cost
updating are provided.
The approach outlined in this section will focus on cost effective-
ness as the primary comparison tool. Cost effectiveness is simply the
ratio of the annualized cost of the emissions control to the amount of
emissions reduction achieved. Mathematically, cost effectiveness is
defined by:
. (B-l)
where: C* = cost effectiveness, $/mass of emissions reduction
Ca = annualized cost of the control measure, $/yr
AR = reduction (mass/yr) in annual emissions
This general methodology was chosen because it is equally applicable to
different controls that achieve equivalent emissions reduction on a
single source and to measures that achieve varied reductions over
multiple sources.
B-2
-------
The discussion is divided into three sections. The first section
describes the general cost analysis methodology, including the various
types of costs that should be considered and presents methods for calcu-
lating those costs. The second identifies the primary cost elements
associated with each of the fugitive emissions control systems. The
final section identifies sources of cost data and discusses methods for
updating cost data to constant dollars, and includes example calculation
cases for estimating costs and cost effectiveness.
B.I GENERAL COST METHODOLOGY
Calculation of cost effectiveness for comparison of control measures
or control strategies can be accomplished in four steps. First, the
alternative control/cost scenarios are selected. Second, the capital
costs of each scenario are calculated. Third, the annualized costs for
each of the alternatives are developed. Finally, the cost effectiveness
is calculated, taking into consideration the level of emissions
reduction.
The general approach for performing each of the above steps is
described below. This approach is intended to provide general guidance
for cost comparison. It should not be viewed as a rigid procedure that
must be followed in detail for all analyses. The reader may choose or
may be forced through resource or informational constraints to omit some
elements of the analysis. However, for comparisons to be valid, cautions
that should be observed are: (1) all control scenarios should be treated
in the same manner, and (2) cost elements that vary radically between
cost scenarios should not be omitted.
B.I.I Select Control/Cost Scenarios
Prior to the cost analysis, general control measures or strategies
will have been identified. These measures or strategies will fall into
one of the major classes of fugitive emission control techniques that
were identified. The first step in the cost analysis is to select a set
of specific control/cost scenarios from the general techniques. The
specific scenarios will include definition of the major cost elements and
identification of specific implementation alternatives for each of the
cost elements.
B-3
-------
Each of the general control techniques Identified in this document
has several major cost elements. The first step in any cost analysis is
definition of these major cost elements. These elements include capital
equipment elements and operation/maintenance elements. Information is
provided in Section B.2 on the major cost elements associated with each
of the general control techniques. For example, the major cost elements
for chemical stabilization of an unpaved road include: (a) chemical
acquisition; (b) chemical storage; (c) road preparation; (d) mixing the
chemical with water; and (e) application of the chemical solution.
For each major cost element, several implementation alternatives can
be chosen. Options within each cost element include such choices as
buying or renting equipment; shipping chemicals by railcar, truck tanker,
or in drums via truck; alternative sources of power or other utilities;
and use of plant personnel or contractors for construction and mainte-
nance. The major cost elements and the implementation alternatives for
each of these elements for the chemical stabilization example described
above are outlined in Table B-l.
Note that the following discussion is applicable only to "in-house"
open fugitive dust control programs. Many industrial facilities prefer
to retain outside contractors to implement paved and unpaved road dust
control programs. Some contractors provide complete services, from
ordering and storing raw materials through control application; others
may only supply the vehicles and drivers required for application. As a
result, the reader may need to consider several implementation alterna-
tives (such as those given in Table B-l) prior to selection of a cost-
efficient dust control program.
B.I.2 Develop Capital Costs
The capital costs of an open dust emissions control program are
those direct and indirect expenses incurred up to the date when the
program is implemented. These capital costs include actual purchase
expenses for capital equipment, labor and utility costs associated with
installation of the control system, and system startup and shakedown
costs. In general, direct capital costs are the costs of control
B-4
-------
TABLE B-l. IMPLEMENTATION ALTERNATIVES FOR STABILIZATION OF AN
UhPAVED ROAD
Cost elements/implementation alternatives
I. Purchase and Ship Chemical
A. Ship in railcar tanker (11,000 to 22,000 gal/tanker)
B. Ship in truck tanker (4,000 to 6,000 gal/tanker)
C. Ship in drums via truck (55 gal/drum)
II. Store Chemical
A. Store on plant property
1. In new storage tank
2. In existing storage tank
a. Needs refurbishing
b. Needs no refurbishing
3. In railcar tanker
a. Own railcar
b. Pay demurrage
4. In truck tanker
a. Own truck
b. Pay demurrage
5. In drums
B. Store in contractor tanks
III. Prepare Road
A. Use plant-owned grader to minimize ruts and low spots
B. Rent contractor grader
C. Perform no road preparation
IV. Mix Chemical and Water in Application Truck
A. Put chemical in spray truck
1. Pump chemical from storage tank or drums into application
truck
1. Pour chemical from drums into application truck, generally
using forklift
B. Put water in application truck
1. Pump from river or lake
2. Take from city water line
V. Apply Chemical Solution via Surface Spraying
A. Use plant owned application truck
B. Rent contractor application truck
B-5
-------
materials and equipment as well as the labor and utilities needed to
install the equipment. Indirect costs are overall costs to the facility
incurred by the system but not directly attributable to specific
equipment items.
Direct costs cover the purchase of equipment and auxiliaries and the
costs of Installation. Capital costs also include any cost of site
development necessitated by the control system. For example, if the
storage tanks for chemical unpaved road dust suppressants require that an
access road be constructed, this access road 1s included as a capital
expense. The types of direct costs for typical fugitive emissions con-
trol programs at TSDFs are associated with items such as vehicles (e.g.,
spray or flusher trucks) and storage tanks, pumps, and piping.
Indirect costs cover the expenses not attributable to specific
equipment items. Items in this category include:
1. Engineering costs—including administrative, project, and
general; labor and other costs associated with specification of the
control program parameters; cost analysis; purchasing and accounting;
consultant services.
2. Field and construction expenses—includes equipment rental,
repair, fuel, or lubricants; costs associated with grading or other road
preparation; field supervision; storage areas.
3. Shakedown/startup—includes costs associated with control
startup and shakedown.
4. Contingency costs—the excess account set up to deal with
uncertainties in the cost estimate, including unforeseen escalation in
prices, malfunctions, equipment design alterations, and similar sources.
The values for these items will vary depending on the specific
operations to be controlled and the types of control systems used. Typi-
cal ranges for indirect costs1 based on the total installed cost of the
capital equipment are shown in Table B-2.
B.I.3 Determine Annualized Costs
The most common basis for comparison of an alternative control
system is that of annualized cost. The annualized cost of a fugitive
emission control system includes operating costs such as labor,
B-6
-------
Cost item
TABLE B-2. TYPICAL VALUES FOR INDIRECT CAPITAL COSTS
Ranges of values
Engineering
Construction and field
expenses
Contractor's fee
Shakedown/startup
Contingency
8 to 20 percent of installed cost. High
value for small projects; low value for
large projects
7 to 70 percent of installed cost
10 to 15 percent of installed cost
1 to 6 percent of installed cost
10 to 30 percent of total direct and indirect
costs dependent upon accuracy of estimate.
Generally, 20 percent is used in a study
estimate
Source: Reference 1.
B-7
-------
materials, utilities, and maintenance items as well as the annualized
cost of the capital equipment. The annualization of capital costs is a
classical engineering economics problem, the solution of which takes into
account the fact that money has time value. These annualized costs are
dependent on the interest rate paid on borrowed money or collectable by
the plant as interest (if available capital is used), the useful life of
the equipment and depreciation rates of the equipment.
The components of the annualized cost of implementing a particular
control technique are depicted graphically in Figure B-l. Purchase and
installation costs include freight, sales tax, and interest on borrowed
money. The operation and maintenance costs reflect increasing frequency
of repair as the equipment ages along with increased costs due to infla-
tion for parts, energy, and labor. On the other hand, costs recovered by
claiming tax credits or deductions are considered as income. Mathe-
matically the annualized costs of control equipment can be calculated
from:
ca ' CRF (CP} + Co + °'5 Co
where: Ca = annualized costs of control equipment, $/yr
CRF = Capital Recovery Factor, 1/yr
Cp - installed capital costs, $
CQ = direct operating costs, $/yr
0.5 = plant overhead factor
The various components of this equation are briefly described below.
The annualized cost of capital equipment is calculated by using a
capital recovery factor (CRF). The CRF combines interest on borrowed
funds and depreciation into a single factor. It is a function of the
interest rate and the overall life of the capital equipment and can be
estimated by the following equation:
B-8
-------
en
O
O
LU
2
O
O
z
o
Equipment. Installation, Freight, Tax, and Interest
Depreciation Tax Deduction
LIFE OF EQUIPMENT
Scrap
Value
Figure B-l. Graphical presentation of fugitive emission control costs,
B-9
-------
CRF . 1(1+1)" (B-3)
(l+i)n-l
where: 1 = annual interest rate expressed as a fraction
n = economic life of the control system (yr)
The other major components of the annualized cost are operation and
maintenance costs (direct operating costs) and associated plant overhead
costs. Operation and maintenance costs generally include labor, raw
materials, utilities, and by-product costs or credits associated with
day-to-day operation of the control system. Elements typically included
in this category are:
1. Utilities—includes water, electricity.
2. Raw materials—includes chemical unpaved dust suppressants,
surfactants, etc.
3. Operating labor—includes supervision, skilled and unskilled
labor required by the control program.
4. Fuel costs—includes fuel required by vehicles and other
equipment used in the control program.
5. Maintenance and repairs—includes the manpower and materials to
keep equipment operating efficiently.
Another component of the operating cost is overhead, which is a
business expense not charged directly to a particular part of the process
but allocated to it. Overhead costs include administrative, safety,
engineering, legal, and medical services; payroll, employee benefits;
recreation; and public relations. As suggested by Eq. B-2, these charges
are estimated to be approximately 50% of direct operating costs.
B.I.4 Calculate Cost Effectiveness
As discussed in the introduction to this section the most informa-
tive method for comparing control measures or control strategies for
particulate fugitive emissions sources is on a cost-effectiveness
basis. Mathematically, cost effectiveness is defined as:
B-10
-------
(B-D
where: C* » cost effectiveness, $/mass of emissions reduction
Ca » annual1zed cost of the control measure, $/yr
AR » reduction (mass/yr) 1n annual emissions
The annual1zed cost of control equipment can be calculated using
Eq. B-2. The annual reduction 1n part1culate emissions can be calculated
from the following equation:
(B-4)
AR = M e c '
where: M * annual source extent
e = uncontrolled emission factor (i.e., mass of uncontrolled
emissions per unit of source extent)
c = average control efficiency expressed as a fraction
B.2 COST ELEMENTS OF FUGITIVE EMISSIONS CONTROL SYSTEMS
The cost methodology outlined in Section B.I requires that the
analyst define and select alternative control/cost scenarios and develop
costs for the major cost elements within these scenarios. The objective
of this subsection is to assist the reader in identifying the implementa-
tion alternatives and major cost elements associated with the emission
reduction techniques. For open dust sources, the control techniques
addressed are: wet dust suppression, surface cleaning, and paving.
B-ll
-------
Implementation alternatives for open dust source emission control
measures are presented in Tables B-3 through B-5. Table B-3 presents
Implementation alternatives for water and chemical dust suppressant
systems. Table B-4 presents alternatives for three types of street
cleaning systems—vacuuming, flushing, and a combination of flushing and
broom sweeping. Table B-5 presents alternatives for streets or parking
lot paving.
After the control scenarios are selected, the analyst must estimate
the capital cost of the Installed system and the operating and mainte-
nance costs. The indirect capital costs elements are common to all
systems and were identified 1n Table B-2. The direct capital cost ele-
ments and direct operation and maintenance cost elements which are unique
to each type of fugitive emission control system are identified in
Tables B-6 through B-9. These costs are provided for dust suppressant
programs for open dust sources in Table B-6, street cleaning programs in
Table B-7, paving in Table B-8, and wet suppression systems for process
sources in Table B-9.
B.3 SOURCES OF COST DATA
Collection of the data to conduct a cost analysis can sometimes be
difficult. If a well defined system is being costed, the best sources of
accurate capital costs are vendor estimates. However, if the system is
not sufficiently defined to develop vendor estimates, published cost data
can be used. Cost data are available for both paved and unpaved roads in
References 2 through 8.
Often published cost estimates are based on different time-valued
dollars. These estimates must be adjusted for inflation so that they
reflect the most probable capital investments for a current time and can
be consistently compared. Capital cost indices are the techniques used
for updating costs. These indices provide a general method for updating
overall costs without having to complete in-depth studies of individual
cost elements. Indices that typically are used for updating control
system costs are the Chemical Engineering Plant Cost; Index (available in
any issue of Chemical Engineering Magazine), the Bureau of Labor
Statistics Metal Fabrication Index, and the Commerce Department Monthly
Labor Review.
B-12
-------
rABLE 8-3. IMPLEMENTATION ALTERNATIVES FOR OUST SUPPRESSANTS APPLIED TO
AN UNPAVED ROAD
Program implementation alternative
Dust
suppressant type
ChemicalsWater
I. Purchase and Ship Dust Suppressant
A. Ship in railcar tanker (11,000 to X
22,000 gal/tanker)
B. Ship in truck tanker (4,000 to 6,000 gal/ X
tanker)
C. Ship in drums via truck (55 gal/drum)
II. Store dust suppressant
A. Store on plant property
1. In new storage tank X
2. In existing storage tank X
a. Needs refurbishing X
b. Needs no refurbishing X
3. In railcar tanker
a. Own railcar X
b. Pay demurrage X
III. Prepare Road
A. Use plant-owned grader to minimize ruts and low X
spots
B. Rent contractor grader X
C. Perform no road preparation X
IV. Mix Dust Suppressant/Water in Application Truck
A. Put suppressant in spray truck
1. Pump suppressant from storage tank or drums X
into application truck
2. Pour suppressant from drums into application X
truck, generally using forklift
B. Put water in application truck
1. Pump from river or lake X
2. Take from city water line X
V. Apply suppressant solution via surface spraying
A. Use plant owned application truck X
B. Rent contractor application truck X
X
X
X
X
X
X
B-13
-------
TABLE B-4. IMPLEMENTATION ALTERNATIVES FOR PAVED ROAD CLEANING
Program implementation alternative
I. Acquire Fl usher and Driver
A. Purchase fl usher/sweeper and
use plant driver
B. Rent f lusher/sweeper and
driver
C. Use existing unpaved road
watering truck
Vacuum
sweeping Flushing
X
X
X
Flushing
and
broom
sweeping
X
X
II. Acquire Vacuum Sweeper and Driver
A. Purchase sweeper and use
plant driver
B. Rent sweeper and driver
X
X
III. Fill Flusher Tank With Water
A. Pump water from river or lake
B. Take water from city line
X
X
X
X
IV. Maintain Purchased Flusher
V. Maintain Purchased Vacuum Sweeper
X
X
B-14
-------
TABLE B-5. IMPLEMENTATION ALTERNATIVES FOR PAVING
Program implementation alternative
I. Excavate Existing Surface to Make Way for Base and
Surface Courses
A. 2-in. depth
B. 4-in. depth
C. 6-in depth
II. Fine Grade and Compact Subgrade
III. Lay and Compact Crushed Stone Base Course
A. 2-in. depth
B. 4-in. depth
C. 6-in depth
IV. Lay and Compact Hot Mix Asphalt (Probably AC120-150)
Surface Course
A. 2-in. depth
B. 4-in. depth
C. 6-in depth
B-15
-------
TABLE B-6. CAPITAL EQUIPMENT AND O&M EXPENDITURE ITEMS FOR
OUST SUPPRESSANT SYSTEMSd
(Open Sources)
Capital equipment
• Storage equipment
Tanks
Railcar
Pumps
Piping
• Application equipment
Trucks
Spray system
Piping (including winterizing)
Q&M expenditures
• Utility or fuel costs
Water
Electricity
Gasoline or diesel fuel
• Supplies
Chemicals
Repair parts
• Labor
Application time
Road conditioning
System maintenance
aNot all items are necessary for all systems. Specific items
are dependent on the control scenario selected.
B-16
-------
TABLE B-7. CAPITAL EQUIPMENT AND O&M EXPENDITURE ITEMS FOR
STREET CLEANING
Capital equipment
• Sweeping
Broom
Vacuum system
• Flushing
Piping.
Flushing truck
Water pumps
O&M expenditures
• Utility and fuel costs
Water
Gasoline or diesel fuel
• Supplies
Replacement brushes
• Labor
Sweeping or flushing operation
Truck maintenance
• Waste disposal
B-17
-------
TABLE B-8. CAPITAL EQUIPMENT AND O&M EXPENDITURES ITEMS FOR
PAVING
Capital equipment
• Operating equipment
Graders
Paving application equipment
Materials
Paving material (asphalt or concrete)
Base material
O&M expenditures
• Supplies
Patching material
• Labor
Surface preparation
Paving
Road maintenance
Equipment maintenance
B-18
-------
TABLE B-9. CAPITAL EQUIPMENT AND O&M EXPENDITURE ITEMS FOR
WET SUPPRESSION SYSTEMS (PROCESS SOURCES)
Capital equipment
• Water spray systems
Supply pumps
Nozzles
Piping (including winterization)
Control system
Filtering units
• Water/surfactant and foam systems only
Air compressor
Mixing tank
Metering or proportioning unit
Surfactant storage area
O&M expenditures
• Utility costs
Water
Electricity
• Supplies
Surfactant
Screens
• Labor
Maintenance
Operation
B-19
-------
Operation and maintenance cost estimates typically are based on
vendor or industry experience with similar systems. In the absence of
such data, rough estimates can be developed from References 4 and 7.
B.4 EXAMPLE CALCULATIONS
Table B-10 lists the steps necessary to calculate the cost effec-
tiveness for two control alternatives for stabilizing unpaved travel
surfaces. An example problem illustrating the calculations is presented
in Table B-ll. Tables B-12 through B-15 are referenced in the
calculations of Tables B-10 and B-ll.
REFERENCES FOR APPENDIX B
1. PEDCo Environmental, Inc. Cost Analysis Manual for Standards
Support Document. U.S. Environmental Protection Agency. November
1978.
2. Cuscino, T., Jr., G. E. Muleski, and C. Cowherd, Jr. Iron and Steel
Plant Open Source Fugitive Emission Control Evaluation. EPA-600/2-
83-110, NTIS No. PB84-110568, U.S. Environmental Protection Agency,
Research Triangle Park, NC. October 1983.
3. Muleski, 6. E., T. Cuscino, Jr., and C. Cowherd, Jr. Extended
Evaluation of Unpaved Road Dust Suppressants in the Iron and Steel
Industry. EPA-600/2-84-027, NTIS No. PB84-154350, U.S. Environ-
mental Protection Agency, Research Triangle Park, NC. February
1984.
4. Cuscino, T., Jr. Cost Estimates for Selected Dust Controls Applied
to Unpaved and Paved Roads in Iron and Steel Plants. EPA Contract
No. 68-01-6314, Task 17, U.S. Environmental Protection Agency,
Region V, Chicago, IL. April 1984.
5. Richardson Engineering Services, Inc. The Richardson Rapid Con-
struction Cost Estimating System: Volume I—Process Plant Construc-
tion Estimating Standards. 1983-84 Edition.
6. Robert Snow Means Company, Inc. Building Construction Cost Data.
1979.
7. Neveril, R. V. Capital and Operating Costs of Selected Air Pollu-
tion Control Systems. EPA-450/5-80-002. GARD, Inc. December 1978.
8. Turner, J. H., et al. Fugitive Particulate Emissions from Hazardous
Waste Sites. Prepared for the U.S. Environmental Protection Agency
under Contract No. 68-03-3149, Cincinnati, Ohio. September 1984.
B-20
-------
TABLE B-10. COST AND COST-EFFECTIVENESS ESTIMATION FOR CHEMICAL STABILIZATION
OF AN UNPAVED ROAD
This table lists the steps necessary to calculate the cost-effectiveness for two control
alternatives for stabilizing unpaved travel surfaces.
Step I—Specify Desired Average Control Efficiency (e.g., 50. 75, or 90 percent)
Step 2—Specify Basic Vehicle, Road and Climatological Parameters for the Particular Road of
Concern
Required vehicle characteristics include:
1. Average Daily Traffic (ADT)—this is the number of vehicles using the road regardless of
direction of travel (e.g., on a two-lane road in an iron and steel plant, 100 vehicles in one
direction, and 100 in the other direction during a single day yields 200 ADT);
2. Average vehicle weight in short tons;
3. Average number of vehicle wheels; and
4. Average vehicle speed in mph.
Required road characteristics include:
1. Actual length of roadway to be controlled in miles;
2. Width of road to be controlled;
3. Silt content (in percent)—for an existing road, these values should be measured;
however, for a proposed plant, average values shown in AP-42 can be used;
4. Surface loading (for paved roads) in Ib/mile—this is the total loading on all traveled
lanes rather than the average lane loading; and
5. Bearing strength of the road—At this time, just a visual estimate of low, moderate, or
high is required.
Required climatological characteristics (applicable only to watering of unpaved roads):
potential evaporation in mm/h—the value depends on both the location and the month of concern.
Control efficiency data in this report for watering unpaved roads assume a location in Detroit,
Michigan, in the summer.
Step 3—Calculate the Uncontrolled Annual Emission Rate as the Product of the Emission
Factor and the Source Extent
The emission factor (E) should be calculated using the equations from AP-42.
The annual source extent (SE) is calculated as 365 x ADT x average one way trip distance.
Step 4—Consult the Appropriate Control Program Design Table to Determine the Time Between
Applications and the Application Intensity
Select the appropriate table
Table
containing
Control technique information
Coherex* applied to unpaved roads Table B-12
Petro Tac applied to unpaved roads Table B-13
(continued)
B-21
-------
TABLE B-10. (continued)
Verify that the vehicle and road characteristics listed in Step 2 are similar to those
listed in the footnotes of the selected table. If they are significantly different, the table
cannot be used.
Step 5—Calculate the Number of Annual Applications Necessary by Dividing 365 by the Days
Between Application (from Step 4)
Step 6—Calculate the Number of Treated Miles Per Year by Multiplying the Actual Miles of
Road to be Controlled (from Step 2) by the Number of Annual Applications (from Step 5)
Step 7—Consult the Appropriate Program Implementation Alternatives Table and Select the
Desired Program Implementation Plan
Table
containing
Control technique information
Coherex* applied to unpaved roads Table B--14
Petro Tac applied to unpaved roads Table B-15
Step 8—Calculate Total Annual Cost by Annualizing Capital Costs and Adding to Annual
Operation and Maintenance Costs
To annualize capital investment, the capital cost is multiplied by a capital recovery factor
which is calculated as follows:
CRF = [i(l + i)n] / [(! + i)n - 1]
where CRF = capital recovery factor
i * annual interest rate fraction
n = number of payment years
Scale total annual cost by ratio of actual road width in feet divided by 40 ft.
Step 9—Calculate Cost-Effectiveness by Dividing Total Annual Costs (from Step 8) by the
Annual Uncontrolled Emission Rate (from Step 3) and by Desired Control Efficiency Fraction (from
Step 1)
B-22
-------
TABLE B-H. EXAMPLE CALCULATION CASE: APPLICATION OF COHEREX* TO AN UNPAVED ROAD
This table is an example cost-effectiveness calculation for controlling PM10 using Coherex*
on an unpaved road in a Detroit, Michigan, plant.
Step 1—Specify Desired Average Control Efficiency
Desired average control efficiency * 90 percent
Step 2—Specify Basic Vehicle. Road, and Cl iinatological Parameters for the Particular Road
of Concern
Required vehicle characteristics:
1. Average daily traffic = 100 vehicles per day;
2. Average vehicle weight = 40 ST;
3. Average number of vehicle wheels = 6; and
4. Average vehicle speed * 20 mph
Required road characteristics:
1. Actual length of roadway to be controlled * 6.3 miles;
2. Width of roadway = 30 ft;
3. Silt content = 9.) percent;
4. Bearing strength of road = moderate
Step 3—Calculate Uncontrolled Annual Emission Rate as the Product of the Emission Factor
and the Source Extent
0.7/M\0.5
/ s\/ S\/W\°-'/w\0'5 /565-o\
E •k 5-9 (liXslXs) 6) (lr)
where E = emission factor
k = 0.36 for PMjg (from Section 2.0 of this manual)
s = 9.1 percent (given in Step 2)
S = 20 mph (given in Step 2)
W = 40 ST (given in Step 2)
w - 6 (given in Step 2)
p = 140 (as per Figure 2-4 for Detroit, Michigan)
E = 4.98 Ib/VMT
SE * 365 x ADT x average one-way trip distance
SE = 365 42*1 x ,00 vehicles x 6'3 miles
year day 2 vehicle
SE = 115,000 VMT/year
Emission rate = E x SE
Emission rate = 4.98 Ib/VMT x 115,000 VMT/year x 1 short ton
Emission rate = 286 tons of PM10 per year
(conti nued)
B-23
-------
TABLE 8-11. (continued)
Step 4—Consult the Appropriate Control Program Design Table to Determine the Times Between
Applications and the Application Intensity
Use Table B-12.
The vehicle and road characteristics listed in Step 2 are similar to those In the footnotes
of Table 2-1.
From Table B-12:
Application intensity * 0.83 gal. of 20 percent solution/yd2
(initial application)
» 1.0 gal. of 12 percent solution/yd2
(reapplicat ion)
Application frequency = once every 47 days
Step 5—Calculate the Number of Annual Applications Necessary by Dividing 365 by the Days
Between Applications (from Step 4)
No. of annual applications - ^ = 77.7 applications
47 year
Step 6—Calculate the Number of Treated Miles Per Year by Multiplying the Actual Miles of
Road to be Controlled (from Step 2) by the Number of Annual Applications (from Step 5)
No. of treated miles per year - 6.3 miles x 7.77 applications
year
* 49 treated miles/year
Step 7—Consult the Appropriate Program Implementation Alternatives Tables and Select the
Desired Program Implementation Plan
From Table B-13, the following implementation plan and associated costs are anticipated:
1.
2.
3.
4.
5.
Selected alternative
Purchase Coherex* and ship in truck tanker
Store in newly purchased storage tank
Prepare road with plant owned grader
Pump water from river or lake
Apply chemical with plant owned application
truck (includes labor to pump water and
Coherex® and apply solution)
Capital
investment, $
30,000
5,000
70,000
105,000
Unit
$/Treated
mi le
4,650
135
17785
cost
$/Actual
mile
630
630
(continued)
B-24
-------
TABLE B-ll. (continued)
Step 8 — Calculate Total Annual Cost by Annual! zing Capital Costs and Adding to Annual
Operation and Maintenance Costs
Calculate annual capital investment (PI) » capital investment x CRF
CRF » [i(1 «• !)"]/[(! * i)n - 1]
CRF = capital recovery factor
i « 0.15
n « 10 years
CRF « 0.199252
PI * 105,000 x 0.199252 * S20,900/year
Calculate annual operation and maintenance cost (MO)
MO * $4, 785 /treated mile x 49 treated miles/year + $630/actual mile x 6.3 actual. mi les
year
= 5238,000/year
Calculate total cost (D) * PI + MO
D = $20,900/year + $238,000/year
= $258,900/year
Scale total cost by actual road width
Actual total cost for a 30- ft wide road = $258,900/yr x 2P_±L
40 ft
= J)94,200/yr
Step 9 — Calculate Cost-Effectiveness Dy Dividing Total Annual Costs (from Step 8) by the
Annual Uncontrolled Emission Rate (from Step 3) and by the Desired Control Efficiency Fraction
(from Step 1)
Cost effectiveness
,Q ft
286 ST/year x 0.9
= $754/short ton of PM10 reduced
B-25
-------
TABLE B-12. ALTERNATIVE CONTROL PROGRAM DESIGN FOR COHEREX*
APPLIED TO TRAVEL SURFACESa B
Average
percent
control
desired
50
75
90
Vehicle
passes between
applications
23,300
11,600
4,650
Days
as
100
233
116
47
between applications
a function of ACT
300
78
39
16
500
47
23
9
Calculated time and vehicle passes between application are
based on the following conditions:
Suppressant application:
• 3.7 L of 20 percent solution/m2 (0.83 gallon of
20 percent solution/yd2) initial application
• 4.5 L of 12 percent solution/m2 (1.0 gal. of 12 percent
solution/yd2); reapplications
Vehicular traffic:
• Average weight—Mg (43 tons)
• Average wheels—6
• Average speed—29 km/h (20 mph)
Road structure: bearing strength—low to moderate
DPM-10 = Particles <10 umA.
For reapplications that span time periods greater than
365 days, the effects of the freeze-thaw cycle are not
incorporated in the reported values.
B-26
-------
TABLE B-13. ALTERNATIVE CONTROL PROGRAM DESIGN FOR PETRO TAG
APPLIED TO TRAVEL SURFACES4 D
Average
percent
control
desired
50
75
90
Vehicle
passes between
applications
92,000
47,800
21,200
Days between applications
as a function of ACT
100
920
478
212
300
307
159
71
500
184
96
42
Calculated time and vehicle passes between application are
based on the following conditions:
Suppressant application: 3.2 L of 20 percent solution/in^
(0.7 gal of 20 percent solution/yd2)? each .application
Vehicular traffic:
• Average weight—Mg (30 tons)
• Average wheels—9.2
• Average speed—22 km/h (15 mph)
Road structure: bearing strength—low to moderate
°PM-10 = particles <10 umA.
cFor reapplications that span time periods greater than
365 days, the effects of the freeze-thaw cycle are not
incorporated in the reported values.
B-27
-------
TABLE B-14. IDENTIFICATION AND COST ESTIMATION OF COHEREX*
CONTROL ALTERNATIVES
Program implementation alternatives
Cost
I. Purchase and ship Coherex®
A. Ship in railcar tanker (11,000-22,000
gal/tanker)
B. Ship in truck tanker (4,000-6,000
gal/tanker)
C. Ship in drums via truck (55 gal/drum)
II. Store Coherex®
A. Store on plant property
1. In new storage tank
2. In existing storage tank
a. Needs refurbishing
b. Needs no refurbishing
3. In railcar tanker
a. Own railcar
b. Pay demurrage
4. In truck tanker
a. Own truck
b. Pay demurrage
5. In drums
B. Store in contractor tanks
III. Prepare road
A. Use plant-owned grader to minimize
ruts and low spots
8. Rent contractor grader
C. Perform no road preparation
IV. Mix Coherex3 = and water in application
truck
A. Load Coherex3= into spray truck
1. Pump Coherex3 » from storage tank
or drums into application truck
2. Pour Coherexs = from drums into
application truck, using fork lift
S4,650/treated mile
54,650/treated mile
$7,040/treated mile
$30,000 capital
$5,400 capital
-0-
-0-
$20, $30, $60/treated
mile
-0-
$70/treated mile
-0-
$140/treated mile
$630/actual mile
$l,200/actual mile
-0-
Tank—0 (included in
price of storage
tank)
Drums—$1,000 capital
51,000/treated mile
(continued)
B-28
-------
TABLE B-14. (continued)
Program implementation alternatives
Cost
B. Load water into application truck
1. Pump from river or lake
2. Take from city water line
V. Apply Coherex® = solution via surface
spraying
A. Use plant owned application truck
B. Rent contractor application truck
$5,000 capital
$40/treated mile
$70,000 capital+5135/
treated mile for
tank or $270/treated
mile for drums
Tank—$500/treated
mile
Drums—$l,000/treated
mile
B-29
-------
TABLE B-15.
IDENTIFICATION AND COST ESTIMATION OF PETRO TAG
CONTROL ALTERNATIVES
Program implementation alternatives
Cost
I. Purchase and ship Petro Tac
A. Ship in truck tanker (4,000-6,000 gal/
tanker)
B. Ship in drums via truck (55 gal/drum)
II. Store Petro Tac
A. Store on plant property
1. In new storage tank
2. In existing storage tank
a. Needs refurbishing
b. Needs no refurbishing
3. In railcar tanker
a. Own railcar
b. Pay demurrage
4. In truck tanker
a. Own truck
b. Pay demurrage
5. In drums
B. Store in contractor tanks
III. Prepare road
A. Use plant owned grader to minimize
ruts and low spots
B. Rent contractor grader
C. Perform no road preparation
(IV. Mix Petro Tac and water in application
truck
A. Load Petro Tac into spray truck
1. Pump Petro Tac from storage tank
or drums into application truck
2. Pour Petro Tac from drums into
application truck, generally using
fork!ift
Load water into application truck
1. Pump from river or lake
2. Take from city water line
$5,400/treated mile
$ll,500/treated mile
$30,000 capital
$5,400 capital
-0-
-0-
$20, $30, $60/treated
mile
-0-
$70/treated mile
-0-
$140/treated mile
S630/actual mile
$l,200/actual mile
-0-
Tank - 0 (included in
price of storage
tank)
Drums—$1,000 capital
51,000/treated mile
$5,000 capital
$40/treated mile
(continued)
B-30
-------
TABLE B-15. (continued)
Program implementation alternatives
Cost
V. Apply Petro Tac solution via surface
spraying
A. Use plant-owned application truck
B. Rent contractor application truck
$70,000 capital+5135/
treated mile for
tank or $270/treated
mile for drums
Tank—$500/treated
mile
Drums—$1,000/treated
mile
B-31
-------
APPENDIX C
SCREENING TECHNIQUES, MODELING INFORMATION, AND
HEALTH RISKS INFORMATION
Screening Analysis for Estimating Maximum Annual Average
Ground-Level Particulate Matter Contaminant Concentrations... C-l
Estimating Health Effects From Fugitive Treatment, Storage,
and Disposal Facilities (TSDF) Air Emissions C-19
-------
Screening Analysis for Estimating Maximum Annual Average Ground-
Level Particulate Matter Contaminant Concentrations
Introduction
The purpose of this section is to present a screening
technique for estimating maximum annual average ground-level
contaminant concentrations due to fugitive particulate matter
emissions from treatment, storage, and disposal facilities
(TSDFs).
The screening technique described in this section is a
simplified procedure sufficiently conservative to allow a
determination whether fugitive particulate matter emissions from
a TSOF would: (1) clearly not result in an air quality threat or
(2) pose a potential threat that should be examined with more
sophisticated modeling techniques or measurements. If the
screening estimates indicate that ground-level concentrations are
not likely to exceed critical health risk levels, further
analysis of the TSDF air quality impact is not necessary. If
these screening estimates demonstrate that concentrations may
exceed critical levels, this should not be construed as
indicating a problem. Rather, it is an indication that more
detailed source impact analyses using refined emissions estimates
and air dispersion models are necessary.
The screening technique described in this section was
developed expressly for TSDF guidance and should not be construed
as EPA policy for any other source type. For each TSDF analysis,
the State or U. S. Environmental Protection Agency (EPA) Regional
Modeling contact should be consulted concerning the proper
application of the screening technique and interpretation of
results. The U. S. EPA Regional Modeling Contacts are identified
in Table 1.
C-l
-------
Table 1
Regional Meteorologists/Modeling Contacts
Susan Kulstad
EPA Region I
J.F.K. Federal Building
Boston, HA 02203-2211
(617) 565-3225
Ray Werner
EPA Region II
26 Federal Plaza
New York NY 10278
(212) 264-2517
Alan Cimorelli
EPA Region III
841 Chestnut Building
Philadelphia, PA 19107
(215) 597-6563
Lewis Nagler
EPA Region IV
345 Courtland Street, N.E,
Atlanta, GA 30365
(404) 347-2864
Michael Koerber
EPA Region V
230 South Dearborn Street
Chicago, IL 60604
(312) 886-6061
James Yarbrough
EPA Region VI
First Interstate Bank
Tower
1445 Ross Avenue
Dallas, TX 75202-2733
(214) 655-7214
Richard Daye
EPA Region VII
726 Minnesota Avenue
Kansas City, KS 66101
(913) 236-2896
John Notar
EPA Region VIII
999 18th Street
Denver Place - Suite 500
Denver, CO 80202-2405
(303) 293-1755
John Vimont
EPA Region IX
215 Fremont Street
San Francisco, CA 94105
(415) 974-8223
Robert Wilson
EPA Region X
Environmental Services
Division
1200 Sixth Avenue
Seattle, WA 98101
(206) 442-1531
C
-
-------
Limitations and Assumptions
This screening technique is designed to provide a simplified
approach to obtain upper-bound concentration estimates for
situations that may represent complex release scenarios and
source configurations. Therefore, the screening methods have
certain limitations such as the following:
* These methods assume that the particulate matter emissions
behave as gaseous pollutants with negligible deposition
velocities. Thus, these methods are not applicable for
particulates with appreciable settling or when deposition is of
concern.
* Steady, continuous emissions for the entire year are
assumed.
* Complex and elevated terrain effects are not considered.
However, for particulate matter plumes resulting from near
ground-1evel, non-buoyant fugitive emission sources such as those
at TSDFs, maximum ground-level concentrations are likely to occur
at or near the property boundary. Thus, the assumption of flat
terrain should have no effect on the magnitude of the maximum
concentrations.
* The screening estimates presume that emissions are
distributed evenly over a square area source representing a
composite of all the fugitive particulate matter emission
sources. This presumption may oversimplify a complex fugitive
emissions source configuration.
C-3
-------
The TSDF permit writer is strongly encouraged to discuss the
appropriate application of this screening technique with the
appropriate State and/or U. S. EPA Regional Modeling Contact.
Overview of Screening Procedures
Fugitive particulate matter emissions from TSDFs generally
originate from surface areas such as the following:
* Roads (paved, unpaved)
* Open waste piles and staging areas
* Dry surface impoundments
* Landfills
* Land treatment
* Waste stabilization basins
Concentration estimates of contaminants contained in the
fugitive particulate matter emissions are needed to estimate
health risk to the surrounding populace. The purpose of a
screening analysis is to provide permit writers an upper-bound .
estimate of contaminant concentration levels (based on fugitive
particulate matter concentration levels) using simplified
procedures.
The purpose of this screening analysis is to obtain an
upper-bound estimate of the maximum annual average contaminant
concentration resulting from fugitive particulate matter
emissions from a TSDF. This screening analysis presumes that a
total, combined annual emission rate is known for all applicable
fugitive emission sources. Annual emission rates can be
estimated using the procedures and emission factors detailed in
this document. As noted above, the screening technique is based
on the assumption of continuous emissions during the entire year.
C-4
-------
Screening Analysis Methodology
This analysis utilizes Figure 1.which relates normalized
annual average concentration estimates, CHI/Q (X/Q), to downwind
distances for six generic area source sizes. The figure contains
a "family-of-curves" - representing the six area source sizes -
which plots normalized concentrations as a function of downwind
distance.*
The X/Q plots shown in Figure 1 were developed by assuming
unit emission rate per unit area per year (kg/m2/yr) for each
generic area source size. The procedures to obtain an upper-
bound estimate of the maximum annual average concentration from a
particular TSDF follow.
Step l: Obtain an estimate of the total, combined fugitive
particulate matter emission area source size. This size can be
obtained by first estimating the total area containing each
active source-type (e.g., the total area comprised of unpaved
roads) and then summing the areas from each source-type to obtain
a total combined area source size.
step 2: Refer to Figure 1. From the total, combined area
source size found in Step 1, determine the square area in Figure
The curves were developed by applying the Industrial
Source Complex (ISC) Long-Term Model based on the meteorological
conditions representative of the general locations of the largest
concentration of TSOF sources in the United States.
Normalized concentration is defined as concentration divided
by emission density or X/Q. Alternatively, the normalized
concentration is the concentration that would be predicted if the
emission density were 1 kg/m2/yr.
C-5
-------
er
0
u. en
UJ
UJ M
0 H-l
z en
1- UJ
en u
Si
O
in en
3<
CC UJ
UJ CC
O X
*^^ !•>{
x en
u
c
E £ E E
c O O O O
E o m o o
O *- r- CM K)
m
X X X X
X
c £ £ E E
E o o o o
o o in o o
in i- T- CM K5
i I
i i
i 1
/
e*
S
/'
^ ^
^"^ ^^^- ".'•-•
i i i i I i i i i I i t i i I i
T o m o
M o r^ m
£
o
o
X
o
o
m
I
!
I
•
/ 1 $
• / ^
I |p
/ | 111
' i ill
i Ml
/i •'*
;J
' : i
/ /;i|
/ / 'it
/ / //a
/ / /^ii
/ ; -: '1
-7 / //if
/ x ;' • f
/ ' / I T
<^yn
^••-•-^U-:-.
\ \ \ \ i- i i i i
A
m | c
^
0)
c_
3
O»
Lu
o
o
CO
0
O
o
o
r—
o
o
0
8
00
o
o
CO
o
o
o
o
CN
^
D
^^
^S.
UJ
o
er
o
4^
LU
CC
U.
0
UJ
a
UJ
o
cc
Lu
UJ
o
en
a
(6_oix)
C-6
-------
1 that best represents the combined area source size. For
example, if the total combined area source size from Step 1 were
approximately 10,000 square meters, the square area found in
Figure 1 that would represent this source would be approximately
100 meters by 100 meters.
Step 3: Based on knowledge of the TSDF property lay-out
(e.g., plot plan), estimate where the square area determined in
Step 2 would be located within the property boundary in relation
to the active sources used to determine the total combined area
source size in Step 1.
/
Step 4! Determine the total annual contaminant emissions
(kilograms) from all fugitive particulate matter emission sources
that have been combined in this area source. Convert the total
annual emissions to emissions (kilograms) per square meter using
the area comprised of the square area source determined in Step
2.
Step 5: Refer to Figure 1. Locate the curve that best
matches the square area source size determined in Step 2. Next,
locate the distance to the nearest property boundary from the
edge of the area source (Refer to Step 3). Locate the X/Q value
for the appropriate downwind distance. Multiply the X/Q value by
the fugitive particulate matter emissions per square meter
determined in Step 4. The resulting concentration represents an
upper-bound estimate of the maximum annual average contaminant
concentration (ug/m3) from the area source. This concentration
is considered an upper-bound estimate because (1) emissions from
all sources occur within the same area (i.e., are collocated),
(2) emissions occur concurrently, (3) the annual X/Q values in
Figure 1 represent the maximum values obtained from modeling a
range of meteorological conditions associated with the locations
of the largest TSDFs, and (4) the annual distribution of
C-7
-------
meteorological conditions is such that the maximum concentrations
are assumed to occur in the direction of the nearest property
boundary from the edge of the fugitive emissions area source.
6; Compare the annual average contaminant concentration
obtained in Step 5 with critical health risk levels. If this
value exceeds the critical annual health risk threshold, more
refined analyses are warranted. Otherwise, the TSDF facility may
be assumed to pose minimal risk to the health threshold.
Example Problem: The following is an example of applying
the screening procedure for a hypothetical TSDF. Consider the
following hypothetical data:
Fugitive Annual Contaminant
Source Emission Rate (Kg) Source Area (m2)
Vehicular Traffic 0.90 400
Open Wastepiles 0.50 400
Dry Surface
Impoundments 0.90 600
Landfills 0.10 300
Waste Stabilization
Basin 0.40 500
Total annual contaminant emissions * 2.80 kg
Total fugitive emissions area source size » 2200 m2
Thus Q - 2.80 kg/2200 m2 - .00127 kg/m2/yr
From Figure 1, the fugitive area source size is best represented
C-3
-------
by the 50m X 50m (2500m2) square area.
Distance from edge of square area source to nearest property
boundary » 100 m.
Using Figure 1 as illustrated, X/Q - 11.8xlO~9 yr/m
Therefore, upper-bound annual average concentration * (X/Q)(Q)
- (11.8xlO"9)(.00127)
- .OlSxio""9 kg/m3 or .015 ug/m3
Summary
The X/Q wfamily-of-curves" shown in Figure 1 were developed
by applying the Industrial Source Complex (ISC) Long-Term Model
for six generic area source sizes. It is possible that using
other less rigorous screening techniques to estimate maximum
annual average concentrations may in some cases yield more
conservative results than those estimated from Figure 1. For
example, the U. S. EPA's SCREEN Model4 may be used to obtain
short-term concentrations for area sources. These concentrations
may then be multiplied by a scaling factor to obtain annual
average concentrations. These annual average concentrations may
be more conservative than those estimated from Figure 1 for some
distances from the edge of the area source. However, at this
time, there are no EPA-approved short-term to long-term scaling
factors that may be used for area source screening.
Additionally, the "family-of-curves" shown in Figure 1 were
developed for square area sources no greater than 500m X 500m.
For larger area sources, it is recommended that the area source
be represented by smaller area sources before screening
techniques are applied. The State or U. S. EPA Regional Modeling
Contact should be consulted for the proper techniques to be used
in estimating annual average concentrations from area sources
larger than 500m X 500m.
C-9
-------
Detailed Facility Modeling Procedures
Introduction
The purpose of this section is to present an overview of
detailed facility modeling procedures for estimating maximum
annual average ground-level concentrations due to fugitive
particulate matter emissions from treatment/ storage/ and
disposal facilities (TSDFs)
If screening results indicate that annual average fugitive
contaminant particulate matter concentrations may exceed critical
levels/ a more detailed modeling analysis of fugitive contaminent
emissions from the TSOF is required. This detailed analysis is
intended to predict maximum concentration levels from the
fugitive particulate matter sources more accurately than the
screening approach.
General guidance for conducting dispersion modeling is
contained in U.S. EPA'S Guideline on Air Quality Models
rRevisedl.2 This Guideline may be obtained from:
U. S. EPA National Technical
Source Receptor Analysis Branch Information Service(NTIS)
Techniques Evaluation Section U.S. Dept. of Commerce
MD-14 or Springfield, VA 22161
Research Triangle Park, NC 27711 (703) 487-4650
(919) 541-5685 NTIS No.: PB86-245248
This section describes a general approach for conducting a
detailed modeling study of fugitive particulate matter emissions
from a TSDF based upon information contained in the Guideline.
Before proceeding with a detailed modeling study/ the TSOF permit
applicant should ensure that the appropriate U. S. EPA Modeling
Contact (Table 1) is informed of all facets of the modeling
C-10
-------
protocol as it is developed. This is to ensure the proper
consideration of the latest modeling techniques, procedures and
requirements.
Model Selection
Fugitive particulate matter emissions from TSDFs are
primarily generated from the following sources:
1. Roads (paved, unpaved)
2. Open waste piles and staging areas
3. Dry surface impoundments
4. Landfills
5. Land treatment
6. Waste stabilization basins
Because of the complexity in characterizing these emission
sources, the most appropriate model to use for detailed modeling
is the Industrial Source Complex (ISC) model.2 Also, because
long-term (i.e., annual average) concentrations are of concern,
the ISC Long-Term (ISCLT) model is recommended for use in the
detailed study.2 The ISC model is available as part of the
User's Network for Applied Modeling of Air Pollution (UNAMAP)
(Version 6). The ISC user's guide3 is available from:
Computer Products
National Technical Information Service (NTIS)
U.S. Department of Commerce
Springfield, VA 22161
(703) 487-4650
NTIS No.: PB88-169487
Additionally, the ISC model is available from various commercial
vendors.
C-ll
-------
General Modeling Considerations
Receptor Sites
Receptor sites for the detailed facility modeling should be
sufficient to estimate the highest annual average concentrations
from the facility at locations to which the public has access.
In designing a receptor network, receptor density and location is
more important than the total number of receptors. The selection
of receptor sites should be determined case-by-case considering
the topography, climatology, monitoring sites and results of the
initial screening. However, for near ground-level, non-buoyant
TSDF fugitive particulate matter sources, the highest
concentrations would likely be at or near the facility property
boundary. Therefore, the greatest density of receptors should be
at or near this boundary. Receptor and source locations are
specified using a consistent set of polar or Cartesian
coordinates. If polar coordinates are used/ the origin is
generally located at the facility's geographical centroid.
The State and U.S. EPA Regional Modeling Contacts should be
consulted concerning the appropriate receptor selection prior to
initiating the modeling study.
Gravitational Settling and Deposition
The ISC model contains settling and deposition algorithms
that are recommended for use when the particulate matter size
distribution can be quantified. Additional information
concerning exercising gravitational settling and deposition
algorithms in the ISC model is described below.
C-12
-------
Urban/Rural Classification
The type of land use in which a TSOF is located is important
to selecting the appropriate dispersion coefficients for use in
the ISCLT model. The land use in the area of the TSDF must be
classified as either urban or rural. The Guideline2 details the
procedures to use for determining the land use classification
surrounding the TSOF.
TSDFs located in an area defined as urban should be modeled
using urban dispersion coefficients, and TSDFs located in a rural
area should use rural dispersion coefficients in the modeling.
ISCLT Program General Description
Cautionary note - because the proper application of many of
the ISCLT model features requires a fundamental knowledge of the
concepts of atmospheric transport and dispersion, the user should
seek expert advice before using any ISCLT model feature not fully
understood.
The ISCLT model uses statistical wind summaries to predict
seasonal (i.e., quarterly) and/or annual ground-level
concentrations at receptor sites specified using polar or
Cartesian coordinates. Additionally, the model is designed to
account for the effects of particulate gravitational settling and
dry deposition on the predicted ambient concentrations. The
annual concentration predictions are made for a finite number of
wind-speed, wind-direction, and atmospheric turbulence
combinations. The predicted concentrations are then weighted
according to the observed frequency of occurrence of each
combination.
C-13
-------
The ISCLT model accepts the following source types: stack
(e.g, point sources such as process vents), area and volume. For
TSDF fugitive particulate matter emission sources, the source
types would likely consist primarily of volume and area sources.
The volume source option may be used to simulate line sources
such as paved or unpaved roads. The area source option will
likely be applied for sources such as landfills, dry surface
impoundments, open waste piles, etc. The ISCLT user should seek
expert advice on the proper configuration of the source types
specific to the TSDF.
The ISCLT model has one rural and three urban modes for
specifying the types of dispersion coefficients to use in the
modeling. The selection of the proper option is dependent on the
land use classification (rural or urban) in the vicinity of the
TSDF. As described above, the Guideline2 details the procedures
to use for determining the land use classification. If the
classification is urban, the Urban Mode 3 option should be
selected in the ISCLT modeling.
Summary of ISCLT Input Data
The input requirements for the ISCLT program consist of four
main categories:
* meteorological data
* source data
* receptor data
* program control parameters
Meteorological data
The ISCLT model utilizes meteorological data consisting of
seasonal or annual statistical tabulations of the joint frequency
of occurrence of wind-speed and wind-direction categories,
C-14
-------
classified according to Pasquill stability categories. These
"STAR" summaries* are based on National Weather Service (NWS)
hourly meteorological observations. The Guideline2 provides
additional specifications concerning the appropriate NWS data to
use for the modeling of specific TSDFs. Typically, the nearest
NWS station that best represents the meteorological conditions at
the TSDF are used., The State or U.S. EPA Regional Modeling
Contact should be consulted concerning the appropriate data to
use.
Source Data
The ISCLT model accepts three source types: stack (or point
sources such as process vents), area, and volume. For each
source, input data requirements include the source location with
respect to a user specified origin, the source elevation (if
terrain effects are included in the model calculations) and
annual average fugitive particulate matter emission rates. For
stack or point-type sources (e.g., process vents) additional
source input requirements include the physical stack height,
inner stack diameter, stack gas exit temperature and stack exit
velocity. Additionally, if the stack or point source is adjacent
to a building and aerodynamic wake effects are a possibility, the
length, width, and height of the building are needed as model
input. The horizontal dimensions and effective emission height
are required for each area and volume source. As noted above, it
is anticipated that most sources within a TSDF may be
characterized as area or volume sources.
The "STAR" summary is a tabulation of the joint frequency
of occurrence of wind-speed and wind-direction categories,
classified according to discrete atmospheric turbulence
categories.
C-15
-------
If gravitational settling and dry deposition are considered
in the modeling, then source inputs must include (1) the mass
fraction of particulates in discrete gravitational settling-
velocity categories, and (2) the surface reflection coefficient
and settling velocity of each settling-velocity category (see
Reference 3, section 2.4.1, pg. 2-25). (For the modeling of PH10
fugitive participate matter emissions, it may not be necessary to
include gravitational settling and dry deposition in the ISCLT
modeling).
Receptor Data
The ISCLT model uses either a polar (r, 9) or Cartesian
(X,Y) coordinate system. The typical polar receptor array
consists of radials (usually one for every 10 degrees azimuth) at
several downwind distances. However, the user is not restricted
to a 10-degree angular separation. For example, for the low-
level sources associated with TSOF fugitive particulate matter
emission sources, there may be a need to locate receptors closer
than 10 degrees azimuth at the property boundary. The polar
receptor array is always centered at X-0, Y=0, (usually at the
facility's geographical centroid). Receptor locations in the
Cartesian system may be input as Universal Tranverse Hercator
(UTM) coordinates or as X (east-west) and Y (north-south)
coordinates with respect to a user-specified origin.
Discrete receptor locations corresponding to other points of
interest (e.g., population centroids) may be used with either
coordinate system.
Program Control Parameters
The ISCLT model allows the user to select from a number of
modeling options. Some of these options include:
C-16
-------
* whether to specify concentration or deposition
estimates at receptor locations,
* selection of polar versus Cartesian receptor
grid system,
* specification of discrete receptor locations
(e.g., population centroids, monitors),
* whether seasonal or annual output listings are
desired,
* whether concentration calculations are based on
rural or urban mode,
* output formats options,
The program control parameters for these and other options
are discussed in detail in the ISC model user's guide3 Section
4.2.3.
The TSDF permit applicant should consult with a modeling
expert familiar with the ISCLT input and program control
requirements prior to initiating a modeling analysis.
C-17
-------
REFERENCES
1. Memorandum from D. Doll to L. Elmore documenting the ISCLT
modeling analyses conducted to prepare Figure 1.
2. U.S. Environmental Protection Agency, Guideline on Air
Quality Models (Revised"! and Supplement A; EPA-450/2-78-027R,
Office of Air Quality Planning and Standards, Research Triangle
Park, NC, July 1986 and July 1987. Available from the National
Technical Information Service (NTIS) as PB86-245248 and PB88-
150958, respectively.
3. U.S. Environmental Protection Agency, Industrial Source
Complex (TSC) Dispersion Model User's Guide — Second Edition
Volume 1 & 2; EPA-450/4-86-005a, Office of Air Quality Planning
and Standards, Research Triangle Park, NC, June, 1986. Available
from the National Technical Information Service as PB88-169487.
4. U.S. Environmental Protection Agency, Screening Procedures
for Estimating the Air Quality Impact of Stationary Sources: EPA—
450/4-88-010, Office of Air Quality Planning and Standards,
Research Triangle Park, NC, August, 1988. Available from the
National Technical Information Service as PB89-159388.
C-18
-------
ESTIMATING HEALTH EFFECTS FROM FUGITIVE TREATMENT, STORAGE,
AND DISPOSAL FACILITIES (TSDF) AIR EMISSIONS
Many adverse health effects can result from exposure to
fugitive air emissions from hazardous waste treatment, storage,
and disposal facilities (TSDF). The major pathways for human
exposure to environmental contaminants are through inhalation,
ingestion, or dermal contact. Airborne contaminants may be toxic
to the sites of immediate exposure, such as the skin, eyes, and
linings of the respiratory tract. Toxicants may also cause a
spectrum of systemic effects following absorption and
distribution to various target sites such as the liver, kidneys,
and central nervous system. Generally, the health effects of
greatest concern following intermittent or continuous long-term
exposures are those that cause either irreversible damage and
serious impairment to the normal functioning of the individual,
such as cancer and organ dysfunctions, or death. Many toxicant
effects can be reversible and disappear with cessation of
exposure. Health effects sometimes associated with exposure to
toxicants include: central nervous system effects such as
headaches, drowsiness, and tremors; skin, eye, and respiratory
tract irritation; nausea; reproductive and developmental ef fectaf) v
and olfactory effects such as awareness of unpleasant or
disagreeable odors.
Exposure to contaminants in air can be acute, subchronic, or
chronic. Acute exposure refers to a very short-term (i.e., <24
h), exposure to a contaminant. Acute exposure to very high
concentrations or to low levels of highly toxic substances can
C-19
-------
2
cause serious and irreversible tissue damage, and even death. A
delayed toxic response may also occur following acute exposure to
certain agents. Chronic exposures are those that occur for long
period of time (from many months to several years). Subchronic
exposure falls between acute and chronic exposure, and usually
involves exposure for a period of weeks or months.
The risk associated with exposure to a toxic agent is a
function of many factors, including the physical and chemical
characteristics of the substance, the nature of the toxic
response and the dose required to produce the effect, the
susceptibility of the exposed individual, and the exposure
situation. In many cases individuals may be concurrently or
sequentially exposed to a mixture of compounds, which may
influence the risk by changing the nature and magnitude of the
toxic response.
1. CANCER.
Calculation of cancer risk is based on the hypothesis that
cancer rates in human populations are associated with
individuals' exposure to pollutants present in the ambient air.
In general, the scientific evidence that cancer risks may be
associated with air pollution or specific pollutants in air is of
three main types: epidemiologic studies of factors associated
with cancer incidence; experimental or laboratory studies of the
carcinogenicity and mutagenicity of substances and mixtures in
the ambient air; and monitoring studies of the presence in the
air of substances known to be carcinogenic.1
C-20
-------
3
Cancer characteristically progresses through stages, each of
which may be initiated or accelerated by a number of different
intrinsic and extrinsic risk factors. Each factor may act at one
or more stages, and different factors may interact in an additive
or a synergistic way. Because of the long latency period between
initial exposure and manifestation of many cancers (20 years or
more), numerous opportunities exist for multiple exposures to
potentially carcinogenic agents.2
a. Estimation of Cancer Potency
Two pieces of information are needed to assess the cancer
risks of exposure to TSDF air emissions: (1) an estimate of the
carcinogenic potency, or unit risk estimate, of the pollutants in
TSDF air emissions; and (2) an estimate of the ambient
concentration of the pollutants from TSDF that an individual or
group of people breathe.
The unit risk estimate (unit risk factor) is used by the
Environmental Protection Agency (EPA) in its analysis of
carcinogens. It is defined as the lifetime cancer risk occurring
in a hypothetical population in which all individuals are exposed
throughout their lifetime (assumed to be 70 years) to an average
concentration of 1 ^g/ra of the pollutant in the air they breathe.
Unit risk estimates can be used for two purposes: (1) to compare
the carcinogenic potency of several agents with one another, and
(2) to give a rough indication of the public health risk that
might be associated with estimated air exposure to these agents.2
C-21
-------
4
Unit risk estimates for many of the TSDF pollutants
covered in this report have been developed and reviewed by EPA.
In the development of unit risk factors, EPA assumes that, if
experimental data show that a substance is carcinogenic in
animals, it may also be carcinogenic in humans. The EPA also
assumes that humans may be as sensitive as the most sensitive
animal species tested and that any exposure to a carcinogenic
substance poses some risk.2
The EPA uses data from animal bioassay and epidemiologic
studies to predict risk to humans at low exposure to the
substance. Both the animal and human studies usually reflect
populations exposed to relatively high levels of a given
substance; however, general population exposure to the substance
is usually low. Because the risks at low-level exposure cannot
be measured directly by animal or epidemiologic studies, a number
of mathematical models have been developed to extrapolate low-
dose risk from the high-dose studies. There is no single model
which is most appropriate for identifying "true" risk estimates.
EPA (1986) noted that in assessments conducted by the Agency, "in
the absence of adequate information to the contrary, the
linearized multistage model will be employed." This model has
been used for many, but not all, of the TSDF chemicals. For more
details of the quantitative assessments the reader is referred to
the Integrated Risk Information System (IRIS)at 513-569-7254.
The Agency has chosen to use nonthreshold models for dose-
response assessment. The basis for nonthreshold models is
C-22
-------
5
primarily theoretical based on how background incidence and
heterogenicity in the population effect the shape of population
dose-response curves. [For more information on this issue the
reader is referred to Hoel et al., 19753; Schneiderman et al.,
19794; EPA, 19845; and Gaylor 19886 .
The unit risk values predicted from human data are
generally maximum likelihood or "best estimates" of risk. The
unit risk values predicted from animal data provide a plausible,
upper bound limit on public risk at lower exposure levels if the
exposure is accurately quantified; i.e., the true risk is
unlikely to be higher than the calculated level and could be
substantially lower.
For animal data, the mathematical formulation chosen to
describe the linear nonthreshold dose-response relationship at
low doses is applied to original unadjusted animal data. Risk
estimates produced by this model from the animal data are then
scaled to a human equivalent estimate of risk. This is done by
multiplying the estimates by several factors to adjust for
experiment duration, species differences, and, if necessary,
route conversion. The conversion factor for species differences
is presently based on models for equitoxic dose.7 The method
that has been used in most of the EPA's quantitative risk
assessments assumes dose equivalence in units of mg/body weight
2/3 for equal tumor response in rats and humans. It is assumed
that metabolic rate is roughly proportional to body surface areas
and that surface area is proportional to 2/3 power of body weight
C-23
-------
6
(as would be the case for a perfect sphere). The estimate is
also adjusted for lifetime exposure to the carcinogen considering
duration of exposure (either occupational exposure duration or
duration of the experiment) and animal lifetime.2
For unit risk estimates for air, studies using exposure
by inhalation are preferred. When extrapolating results from the
inhalation studies from animals to humans, consideration is given
to the following factors:
The deposition of the inhaled compound throughout
the respiratory trace
Retention half-time of the inhaled particles
Metabolism of the inhaled compound
Differences in sites of tumor induction.
Unit risk estimation from human or animal studies is only
an approximate indication of the actual risk in populations
exposed to known concentrations of a carcinogen. Differences
between species (life span, body size, metabolism, immunological
responses, target site susceptibility), as well as differences
within species (genetic variation, disease state, diet), can
cause actual risk to be much different. In human populations,
variations also occur in genetic constitution, diet, living
environment, and activity patterns. Some populations may
demonstrate a higher susceptibility due to certain metabolic or
inherent differences in their response to the effects of
carcinogens. Also, unit risk estimates are based on exposure to
.a referent adult male. Exposed individuals are represented by a
C-24
-------
7
referent male having a standard weight, breathing rate, etc. (No
reference is made to factors such as race or state of health.)
There may be an increased risk with exposure by fetuses,
children, or young adults. Finally, humans are exposed to a
variety of compounds, and the health effects, either synergistic
or antagonistic, of exposure to complex mixtures of chemicals are
not known.2'7
b. EPA Unit Risk Estimates
The EPA has developed unit risk estimates (URE) for about
72 compounds that are either known or suspect carcinogens and
that could be present at a TSDF. The EPA has verified about 35
of these. As shown in Table 1, these URE range in value from 5.8
x 10~7 (ng/m3)'1 for perchloroethylene to 3.3 x 10~5 (pg/m3)"1
for dioxin. In other words, the URS for perchloroethylene is 5.8
cancer cases per every 10 million people exposed to 1 tig/ra3
perchloroethylene. The URE for dioxin is 3.3 cancer cases per
every 100 people exposed to l iig/m3 dioxin. The URE for a given
chemical is multiplied by the estimate of exposure to produce a
risk estimate. For example, if exposure to perchloroethylene was
estimated to be 10 ng/m3 the risk would be
10 ng/m3 - 5.8 x 10~7/^g/m3 » 5.8 x 10"6
(or 5.8 cancer cases for every one million people exposed to 10
jig/m3 perchloroethylene). The URE are revised upon occasion,
therefore the user is advised to confirm that the values are the
most correct for the compounds of concern. To verify, call the
IRIS coordinator at 513-569-7254.
C-25
-------
8
c. Noncancer Health Effects
Although cancer is of great concern as an adverse health
effect associated with exposure to a chemical or a mixture of
chemicals, many other health effects may be associated with such
exposures. These effects may range from subtle biochemical,
physiological, or pathological effects to gross effects such as
death. The effects of greatest concern are the ones that are
irreversible and impair the normal functioning of the individual.
Some of these effects include respiratory toxicity, developmental
and reproductive toxicity, central nervous system effects, and
other systemic effects such as liver and kidney toxicity,
cardiovascular toxicity, and immunotoxicity.
d. Health Benchmark Levels
For chemicals that give rise to toxic endpoints other
than cancer and gene mutations, there is often assumed to be a
level of exposure below which adverse health effects usually do
not occur. This threshold-of-effect assumption maintains that an
organism can tolerate a range of exposures from zero to some
finite value without risk of experiencing a toxic effect. Above
this threshold, toxicity is observed as the organism's
homeostatic, compensating, and adaptive mechanisms are overcome.
To provide protection against adverse health effects in even the
most sensitive individuals in a population, regulatory efforts
are generally made to prevent exposures from exceeding a health
"benchmark" level that is hypothetically below the lowest of the
thresholds of the individuals within a population.
C-26
-------
9
Benchmark levels, termed reference doses (RfD), are
operationally derived from an experimentally obtained no-
observed-effect level or a lowest-observed-effect level by
consistent application of generally order-of-magnitude
uncertainty factors. The uncertainty factors reflect various
types of data used to estimate RfD. The RfD is an estimate (with
uncertainty spanning perhaps an order of magnitude or greater) of
daily exposure to the human population (including sensitive
subpopulations) that is likely to be without an appreciable risk
of deleterious effect. These benchmark levels are compared to
exposure to qualitatively evaluate risk. The greater the
exceedance of the RfD the greater the risk.
The Agency has developed verified oral RfD for a large number
of chemicals, but has only recently established a working group
to begin the process for inhalation RfD. In the absence of such
inhalation values, the noncancer health effects assessment
utilizes an interim approach for deriving chronic exposure limits
that is similar to that used in the Agency's proposed rule on the
burning of hazardous waste in boilers and industrial furnaces (52
FR 16982). These interim values will be used until such time as
Agency-verified inhalation RfD become available.
The Agency is using the following strategy to derive the
interim inhalation benchmark levels:
(1) Where a verified oral RfD has been based on an
inhalation study, the inhalation exposure limit will be
calculated directly from the study.
C-27
-------
10
(2) Where a verified oral RfD has been based on an oral
study, the inhalation exposure limit will be calculated by
converting the daily human oral dose to a corresponding
concentration in air. Route extrapolation will only be performed
after careful evaluation of the confounding factors affecting the
conversion. Some of these factors include: (a) occurrence of
critical toxicological effects at the portal of entry, (b)
differences in systemic effects, (c) first pass effects resulting
in inactivation or activation of the chemical before it reaches
the target organ, (d) effect of route upon dosimetry, and (e)
A
variations in temporal patterns of target organ concentrations.
(3) Where there exist appropriate EPA health documents
containing relevant inhalation toxicity data, those data will be
used in deriving an inhalation exposure limit. Such documents
include Health Assessment Documents, Health Effects Assessments,
and Health Effects and Environmental Profiles. Other health
documents (e.g., National Institute of Occupational Safety and
Health criteria documents) or sources of toxicity information
(e.g., data used to support the development of the American
Conference of Governmental Industrial Hygienists (ACGIH)
threshold limit value) will also be considered. The calculation
of an interim inhalation health benchmark level will be made in
accordance with the Agency's RfD Methodology.
The methodology for converting oral RfDs to interim
inhalation benchmark levels is as follows:
C-28
-------
11
Benchmark level (mg/m3) =
x Bod weight x Correction factor
mj Air breathed/day
where
RfD is the oral reference dose
• Body weight is assumed to be 70 kg for an adult male
• Correction factor for route-to-route extrapolation
(going from the oral route to the inhalation route) is
1.0, unless specific data exist that indicate a more
appropriate value
Volume of air breathed by an adult male is assumed to
be 20 mvday.
Until Agency verified inhalation RfD become available,
alternatives measures might be used. These alternatives might
include each State's own reference doses, that are usually
derived from Threshold Limit Values . The Agency's RfDs can be
obtained from the Integrated Risk Information System (IRIS). The
IRIS user support can be accessed by calling 513-569-7254.
e. Chemicals of Concern
A preliminary list of 179 TSDF chemicals of concern for
the noncancer health assessment is shown in Table 2 along with
the available interim health benchmark levels. Constituents were
drawn from the Agency's final rule on the identification and
listing of hazardous waste (Appendix VIII of Reference 9) and
from several hazardous waste data bases identified by the
Research Triangle Institute.10 To be selected from these
sources, the chemical must have had either an Agency-verified
oral reference dose (as of September 30, 1987 )8 or a Reference
C-29
-------
12
Air Concentration (RAC) found in the Agency's proposed rule on
the burning of hazardous waste in boilers and industrial
furnaces.12 Additional chemicals were added to Table 2 based on
knowledge of a high toxicity associated with that substance.
2. RISK ASSESSMENT
a. Maximum Lifetime Risk
Maximum lifetime risk or individual risk refers to the
person or persons estimated to live in the area of highest
ambient air concentrations of the pollutant(s) as determined by
the detailed facility modeling. This individual is assumed to
reside at the plant fenceline. The maximum lifetime risk
reflects the probability of an individual developing cancer as a
result of continuous exposure to the estimated maximum ambient
air concentration for 70 years. The use of the work "maximum" in
maximum lifetime risk does not mean the greatest possible risk of
cancer to the public. It is based only on the maximum exposure
estimated by the procedure used 10, and it does not incorporate
uncertainties in the exposure estimate or the risk factor.
Maximum lifetime risk is calculated by multiplying the
highest annual average concentration at the fenceline by the unit
risk estimate. The product is the probability of developing
cancer for those individuals assumed to be exposed to this
concentration for their lifetimes. The following example problem
utilizes the upper-bound annual average concentration for
chromium that was determined in the example problem discussed in
the "Screening Techniques for Estimating Maximum Annual Average
C-30
-------
13
Ground-Level Concentrations Due to Fugitive Particulate Emissions
from TSDF" located in this appendix. The unit risk estimate for
chromium was obtained from Table 1 and is 0.012/M-g/m3.
Maximum lifetime risk = (highest annual avg. cone.) (unit
risk estimate)
= (15.0 ng/m3) (0.012/ng/m3)
- 1.8 x 10'1
Thus, anyone residing at this average concentration for 70 years
would have about a 2 in 10 chance of developing the cancer of
concern.
b. Noncancer Health Effects
(1) Chronic Exposures
The assessment of noncancer health effects
associated with chronic exposures to TSDF chemicals of concern is
based on a comparison of the chemical-specific health benchmark
levels (as discussed in Section l.d to estimated ambient
concentrations at various receptor locations around a facility.
Inhalation exposure limit will be compared to the highest annual
average ambient concentrations for each chemical at the selected
facilities. These annual concentrations represent an estimation
of the highest average daily ambient concentration experienced
over a year. Ambient concentrations that are less than the RfD
are not likely to be associated with health risks. The
probability that adverse effects may be observed in a human
population will increase as the frequency of exposures exceeding
the RfD increases and as the size of the excess increases. The
C-31
-------
14
results of this assessment will be available after completion of
an ongoing scientific review of the inhalation benchmark levels
discussed in Section l.d.
(2) Acute Exposures
Assessment of the potential for noncancer health
effects associated with short-term (acute) exposure to TSDF
chemicals of concern at selected facilities will be conducted as
a screening effort to provide additional qualitative support to
the overall noncancer health effects analysis. The extent of
this assessment will be limited by the lack of short-term
exposure limits and by the lack of adequate acute inhalation data
for most of the TSDF chemicals of concern. The assessment will
be conducted by comparing maximum modeled ambient concentrations
for averaging times of 15 minutes, 1 hour, 8 hours, and 24 hours
to available short-term health data matched to the appropriate
averaging time. These data will be obtained from various
sources, including EPA reports an documents, data used to support
occupational exposure recommendations and standards (e.g., ACGIH
Documentation of TLVsl. and other published information.
Determination of the risk of adverse health effects associated
with estimated short-term exposures will be based on a
consideration of the quality of the available health data and the
proximity of the exposure concentration to the health effect
level. This phase of the noncancer health effects assessment is
incomplete at this time.
C-32
-------
15
3. ANALYTICAL UNCERTAINTIES APPLICABLE TO CALCULATIONS OF PUBLIC
HEALTH RISKS IN THIS APPENDIX
a. Unit Risk Estimate
The procedure generally used to develop unit risk
estimates is fully described in Reference 2. Nickel was selected
as an example. The model used and its application to
epidemiological and animal data have been the subjects of
substantial comment by health scientists. The uncertainties are
too complex to be summarized sensibly in this appendix. Readers
who wish to go beyond the information presented in the reference
should see the following FEDERAL REGISTER notices: (1) EPA's
"Guidelines for Carcinogenic Risk Assessment/1 51 FR 33972
(September 24, 19862), and (2) EPA's "Chemical Carcinogens; A
Review of the Science and its Associated Principles," 50 FR 10372
(March 14, 1985), February 1985.
Other significant uncertainties associated with the cancer
unit risk factors include: (1) selection of dose/response
model, (2) selection of study used to estimate the unit risk
estimate, and (3) presence or absence of a threshold.
b. ftiubjent Air Concentrations
The following are relevant to the estimated ambient air
concentrations used in this analysis:
Flat terrain was assumed in the dispersion
model. Concentrations much higher than those
estimated would result if emissions impact on
elevated terrain or tall building near a plant.
C-33
-------
16
The estimated concentrations do not account for the
additive impact of emissions from plants
located close to one another.
The increase in concentrations that could
result from reentrainment of pollutant-
bearing dust from, for example, city streets,
dirt roads, and vacant lots, is not directly
considered.
With few exceptions, the emission rates are based on
assumptions and on limited emission tests.
C-34
-------
17
REFERENCES
1. Clement Associates, Inc. Preliminary Health Risk Evaluation
for Emissions from Hazardous Waste Treatment Storage, and
Disposal Facilities. December 1985.
2. U.S. Environmental Protection Agency. Health Assessment
Document for Nickel and Nickel Compounds. Publication No.
EPA/600/8-83/012FF. Office of Health and Governmental
Assessment, Washington, DC. 1986.
3. Hoel, D.G.; Gaylor, D.W.; Kirschstein, R.L; Saffoptto. U.;
and Schneiderman, M.A. Estimation of Risks of Irreversible,
Delayed Toxicity. Journal of Toxicology and Environmental
Health. 1:133-1511. 1975.
4. Schneiderman, M.A.; Decoufle, P. and Brown, C.C. Thresholds
for Environmental Cancer: Billogic and Statistical Con-
siderations. Annals of New York Academy of Sciences. Pages
92-107, 1979.
5. U.S. Environmental Protection Agency. Background Information
Document for Final Rules for Radonuclides. Volume 1. EPA/
520/1-84-022-1. Office of Air and Radiation Programs,
Washington, D.C. 1984.
6. Gaylor, D.W.; Sheehan, D.M.; Young, J.F.; and Mattison, D.E.
The Threshold Dot Concept in Teratogenesis. Teratology.
38:389-391, 1988.
7. U.S. Environmental Protection Agency. Health Assessment
Document for Carbon Tetrachloride. Publication No. EPA-
600/8-82-001F. Environmental Criteria and Assessment Office,
Cincinnati, OH. 1984.
8. Pepelko, w. E., and J. R. Withey. Methods for route-to-route
extrapolation of dose. Toxicol Ind Health. 1(4):153:175.
1985.
9. U.S. Environmental Protection Agency. Hazardous Waste
Management System; Identification and Listing of Hazardous
Waste; Final rule. 51 FR 28296. 1986.
10. Memorandum from Coy, Dave, RTI, to McDonald, Randy,
EPA/OAQPS. May 2, 1986. Listing of waste constituents
prioritized by quantity.
11. U.S. Environmental Protection Agency. Status Report of the
RfD Work Group. Environmental Criteria and Assessment
Office, Cincinnati, OH. 1987.
C-35
-------
IS
12. U.S. Environmental Protection Agency. Burning of Hazardous
Waste in Boilers and Industrial Furnaces; Preamble
Correction. 52 FR 25612. 1987.
C-36
-------
TABLE 1. TSDF CARCINOGEN LIST
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
Constituent
acetaldehyde
(75-07-0)
acryl amide
(79-06-1)
acrylonitrile
(107-13-1)
aldrin
(309-00-2)
an i 1 i ne
(62-53-3)
arsenic
(7440-38-2)
benz(a)anthracene
(56-55-3)
benzene
(71-43-2)
benzidine
(92-87-5)
benzo(a)p>.-ene
(50-32-8)
beryl 1 ium
(7440-41-7)
bis(chloroethyl)
etner (111-44-4)
bis(chloromethyl
ether (542-88-1)
1,3-butadiene
(106-99-0)
cadmium
(7440-43-9)
Unit risk
estimate.
0*g/nH)-l
2.2x10-6
l.lxlO'4
6.8x10-5
4.9x10-3
7.4x10-6
4.3x10-3
8.9xlO-4
8.0x10-6
6.7x10-2
1.7x10-3
2.4x10-3
3.3x10-4
2.7x10-3
2.8x10-4
1.8x10-3
Basis*
CAG UCR
(class B2)
CAG UCR
(class 62)
CRAVE verified
UCR (class 81)
CRAVE verified
UCR (class 82)
CAG UCR
(class C)
CAG UCR
(class A)
CAG UCR
(class 82)
CAG UCR
(class A)
CRAVE verified
UCR (class A)
CAG UCR
(class 82)
CAG UCR
(class B2)
CRAVE verified
UCR (class 82)
CAG UCR
(class A)
CRAVE verified
UCR (class 82)
CRAVE verified
UCR (class Bl)
(continued)
C-37
-------
TABLE 1 (continued)
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
Constituent
carbon tetra-
chloride (56-23-5)
chlordane
(12789-03-6)
chloroform
(67-66-3)
chloromethane
(74-87-3)
chloromethyl methyl
ether (107-30-2)
chromium VI
(7440-47-3)
DDT
(50-29-3)
dibenz(a.h)
anthracene
(53-70-3)
l,2-dibromo-3-
chloropropane
(96-12-8)
1,2-dichloroethane
(107-06-2)
1, 1-dichloro-
ethylene (75-35-4)
dieldrin
(60-57-1)
diethylstilbestrol
(56-53-1)
2,4-dinitrotoluene
(121-14-2)
Unit risk
estimate.
1.5x10-5
3.7xlO'4
2.3x10-5
3.6x10-6
2.7x10-3
1.2x10-2
3.0x10-3
1.4x10-2
6.3x10-3
2.6x10-5
5.0x10-5
•
4.6x10-3
1.4XLO'1
8.8x10-5
Basis3
CRAVE verified
UCR (class B2)
CRAVE verified
UCR (class 82)
CAG UCR
(class 82)
ECAO UCR
(class C)
CAG UCR
(class A)
CRAVE verified
UCR (class A)
CAG UCR
(class 82)
CAG UCR
(class 82)
CAG UCR
(class 82)
CRAVE verified
UCR (class 82)
CRAVE verified
UCR (class C)
CRAVE verified
UCR (class 82)
CAG UCR
(class A)
CAG UCR
(class 82)
(continued)
C-38
-------
TABLE 1 (continued)
30.
31.
32.
33.
34.
35.
36.
37.
38,
39.
40.
41.
42.
43.
Constituent
1,4-dioxane
(123-91-1)
1 , 2-d i pheny 1 hyd raz i ne
(122-66-7)
epichlorohydrin
(106-89-8)
ethylene dibromide
(106-93-4)
ethylene oxide
(75-21-8)
formaldehyde
(50-00-0)
gasoline
(8006-61-9)
heptachlor
(76-44-8)
heptachlor epoxide
(1024-57-3)
hexachlorobenzene
(118-74-1)
hexachlorobutadiene
(87-68-3)
hexachlorocyclohexane .
(no CAS f)
alpha-hexachloro-
cyclohcxane
(319-84-6)
beta-hexachloro-
cyclohexane
(319-85-7)
Unit risk
estimate.
Ug/mJ)-1
1.4x10-6
2.2xlO-4
1.2x10-6
2.2xlO'4
l.OxlO'4
1.3xlO-5
6.6xlO-7
1.3x10-3
2.6x10-3
4.9xlO'4
2.2x10-5
5.4xlO'4
1.8x10-3
5.3xlO-4
^MHMB^O^MMM^M^M^M^^MH^^^H^BMVM^
Basis3
CAG UCR
(class B2)
CRAVE verified
(class B2)
CRAVE verified
UCR (class B2)
CRAVE verified
UCR (class B2)
CAG UCR
(class B1-B2)
CAG UCR
(class Bl)
CAG UCR
(class B2)
CRAVE verified
UCR (class 82)
CRAVE verified
UCR (class B2)
CAG UCR
(class 82)
CRAVE verified
UCR (class C)
CRAVE verified
UCR (class 82)
CRAVE verified
UCR (class 82)
CRAVE verified
UCR (class 82)
(continued)
C-39
-------
TABLE 1 (continued)
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
***B™"^ •'•^^•M^^MMMMBM^^M"'*1'
Constituent
gamma-hexach 1 oro-
cyclohexane
(lindane) (58-89-9)
hexachlorodibenzo-
p-dioxin,l:2 mixture
(57653-85-7 or
19408-74-3)
hexachloroethane
(67-72-1)
hydrazine
(302-01-2)
3 -methyl chol anthrene
(56-49-5)
4,4'-methylene-bis
(2-chloroaniline)
(101-14-4)
methylene chloride
(75-09-2)
methyl hydrazine
(60-34-4)
nickel refinery
dust (7440-02-0)
nickel subsulfide
(12035-72-2)
2-nitropropane
(79-46-9)
n-nitrosodi-n-
butylamine
(924-16-3)
n-nitroso-
diethylamine
(55-18-5)
Unit risk
estimate.
3.8xlO'4
1.3x10-6
4.0x10-6
2.9x10-3
2.7x10-3
4.7x10-5
4.7x10-7
3.1xlO-4
2.4xlO-4
4.8xlO-4
2.7x10-3
1.6x10-3
4.3x10-2
M^MBMMM^B^^^^BM^^MMMM^^M^
Basis3
CRAVE verified
UCR (class C)
CRAVE verified
UCR (class 82)
CRAVE verified
UCR (class C)
CAG UCR
(class 82)
CAG UCR
(class 82)
CAG UCR
(class 82)
CAG UCR
UCR (class 82)
ECAO UCR
(class 82)
CRAVE verified
UCR (class A)
CRAVE verified
UCR (class 82)
CAG UCR
(class 82)
CRAVE verified
UCR (class 82)
CRAVE verified
UCR (class 82)
(continued)
C-40
-------
TABLE 1 (continued)
Constituent
Unit risk
estimate.
Basis*
57. n-nitroso- 1.4xlO*2
dimethyl amine
(62-75-9)
58. n-nitroso-n- 8.6x10-2
methylurea
(684-93-5)
59. ^n-nitroso- 6.1xlO-4
'pyrrolidine
(930-55-2)
60. pplychlorinated 1.2x10-3
biphenyls
(1336-36-3)
61. pentachlorom'tro- 7.3x10*5
benzene
(82-68-8)
62. pronamide 4.6x10*6
(23950-58-5)
63. reserpine 3.0xl0'3
(50-55-5)
64. 2,3,7,8-tetrachloro- 3.3x10*
dibenzo-p-dioxin
(1746-01-6)
65. 1,1,2,2-tetra- 5.8x10*5
chloroethane
(79-34-5)
CRAVE verified
UCR (class 82)
CA6 UCR
(class 82)
CRAVE verified
UCR (class B2)
CAG UCR
(class 62)
CAG UCR
(class C)
CAG UCR
(class C)
CAG UCR
(class B2)
CAG UCR
(class 82)
CRAVE verified
UCR (class C)
66.
67.
68.
tetrach 1 oroethy 1 ene
(127-18-4)
thiourea
(62-56-6)
toxaphene
(8001-35-2)
5.8xlO-7
S.SxlO'4
3.2x10-3
CRAVE verified
UCR (class C)
CAG UCR
(class B2)
CRAVE verified
UCR (class B2)
(continued)
C-41
-------
TABLE 1 (continued)
Constituent
Unit risk
estimate.
Basis*
69. 1,1,2-trichloro-
ethane
(79-00-5)
70. trichloroethylene
(79-01-6)
71. 2,4,6-trichloro-
phenol
(88-06-2)
1.6xlO-5
1.7xlO"6
5.7x10-6
CRAVE verified
UCR (class C)
CAG UCR
(class B2)
CRAVE verified
UCR (class B2)
72.
vinyl chloride
(75-01-4)
4.1x10-6
CAG UCR
(class A)
aThe inhalation exposure limits are derived from unit cancer
risk (UCR) estimates, which were either (1) verified by the
Carcinogen Risk Assessment Verification Exercise (CRAVE) work
group or (2) established by the Carcinogen Assessment Group
(CAG), but not yet verified by CRAVE. The unit risk estimates
for chloromethane and methyl hydrazine were derived by the
Environmental Criteria and Assessment Office (ECAO).
Note: The constituents on this list and the corresponding unit
risk estimates and exposure limits are subject to change.
C-42
-------
m
V
2
ER HEALTH EFFECTS ASSE
*
5
*
5
^
M
a i
;
S
»-
i
t-
*
L
C ^^
ll
j 1
I"
M
U
^*
k
j
u •
ll
Q ^^
i x •
* "
1
1 ^
01
u
*"
1
J
u
i : i :
-
a
i
S
0)
^ •»
a ^
i
^^ ^j u
1 *-> 1 X*
IA M • •
CM 1 » *»
i e v •
*!-!
l-Z*
9 - 1 f
•3 ? ii
i s • «
o *> -9 —
k 9 • •
x x u u
, , .i
1 1 » •
^
•
! PJ ~* J7
1 • « i
i i «
IA IA S
i * '
•• IA a
I 1 u l» s^
* » ^*
! 0 X X
i • • •
£ -9 _ e
I «D X — O
7< i S
5 ™ *~ ^
5 — c a
O 0| o O
*! •
<- • i
^
i
~~s
ciiMi cyan id* (692-01-6
bon disulfid* (76-16-0
bon onyfluorid*b (363-
— k k
• • •
W U U
s : 5
^^
CM
1
S
<•% 1
f a
s a
1
«••. s?
^^ C '
k CM
*. I ?
TJ *• f^
— X S
k *i P<
0 I ^
— CM
x i u
u — e
X —
V 1 U
1 1 1
1 1 1
IA
1
m
bon t*trachlorid*c (60
ord*n*c (12789-03-6)
orin*b (7782-60-6)
• x x
u u u
1 1 1
1 1 1
^
1
1*1
** T
5 Iff
i <
a u 35
^ • w
^* X 1
— S
«•••««
• *»
u e u
— o •
k r •«
! S
£"
g
oroac*t«ld*hyd*b (107-
hloro-l,3-butadi*n*b
126-99-8)
X 1
U CM
1 1 1
1 1 1
~
1
IA
<-> S
« 1
n s
^•^ p^
CM S U
8~-x"
•
o> "o —
S X k
m o o
•»» o —
u • u
e
k X X
•o _ _
1 . .
1 1 I
1
f^
1
M
orofor*^ (67-66-3)
oro«*than*b>c (74-87-3
h 1 oropr op i on i tr i 1 *b (6
X X 1
u u m
fM 1
1
~
1
m
o- X
i ^
a o
IA —
SO
M
O
9 W
•o —
•~ 1
a i
o x i?
X X I
M?
a o m
e e •
1 8 S
|. : S! : : s ! :
^^
/•* j? i
? j?8 J |
T ^ i j?S ~?
» T 2 *? 5 J? *^
V • ^ *• +* ~* 1 X-
» ^ ^ <^ IA a _§
^ V X 1 X • 1 •* 4
M — •» X • 1 C O
•1 > •>-'««IA^'X
•-— k— e • •« a a
• liooe— oo
xxokkaxxx
uuouuwuuu
1 1 1 M 1 S S 1 1
III 1 CM CM 1 1
1
~ IA
IA IA
V l» r»
CM • 1
^ f» CM 0 Q
S<^ •>—/-. i i ^. 7
f» S 0 CM IA IA
~ i x i » i «
X v •» -•
— «D • » f« i •B a
•0*9 C » *- » — «-» ^
.. ^ • ^ v e
X - *• X • •
ax Cxx<~'uc
o • a e u — u
e — .- o - i f -9 —
.« »^2^55^K
••CCBC9w*MQ
C g g •-• — •• M M
• liwKki.ee
^••••xxxx
!
^
1
nog*n chlorid* (606-77
•
x
w
i
p»
i
^^
^
t
^
f»
^^
uf
—
X
k
X
I 1
1 1
u
loh**anon* (108-94-1)
D (dichloroph*noxy*c*
cid) (94-76-7)
U V 01
x -
W M
1 «
i e
u
S
m
A •
v x
i «*
CM X
IA &
CM ^
^^ X
- x ^
X — 1*
e x i
f «* «
a • i
I f £
— X
1
1
J?
a
CM
S
^•r
U
t
1
1
1
CM
IA
^
•
•
X
E
b
0
u
^
!
X
e
4*
e
w
C-43
-------
(continued)
«M !
UJ I
b*nch*ork
. 3\c
S =
: !
2
•
r
e
"•
y
o
ji
b
a j
2i! "^
i
i
s
1
? ' '
" 1 1
»
•a
^* •
*• •
$ r
1 1
s~
— o
« "
Si
j s
•£ U <-«
** — «
• ^ ^
S S 7
x 1 8
f ^ ^
• *.
SiT1
i"il
i i ^
- | . S «
!j - =» I ? ?
'' J Ij
:'« •!
1 ; b "~
•
1 ' e
< ' M
i
1
1 '
i I
( 1
i
^^
ia
i
p«
t ^*. ^^
SM •"
l 1
— » •
p« h. JZ
** 1 1 .
^ IA '
• CD S
V W ^ L
s s V
-i - • f
•1 * • 5
ui e x e
•i 1 4 * 4-
1 *.'!•!!! 2 .' ! 1 1 I S 9 ! « ;
•» •
XK
^
i
N.
C" *
M «^ | f^
_ i m i>. ||
* f i »* 7 -.
• iftaw^J, C1 ^
£• S »T.T?ffir I
* * Nl^eMi>!.Y9M •*
i • t!«§S?iDT,; 7
nli? in^ifflii i{
iii*»i*M 1 inm
imsgJI iinn::!ll
iillifiMliiHllliii
* *• ' «2 i i a ^•>g^^|||jjjj
5 I S : 2 - : 1 - j j j . , j § , 5 ,
•v •
•^ •— /—>
hk CD V
1 ' — » 1 — . •">•
e M m » «7 w
V (^ I m ^^ . 1
. . ^ ^^ " V
* 1 ^ I «y IA ^
8IA m IA i S "^ «O
** 1 1* •» I J? ^ Wl<-s
-'-wC-Sei ? J?7 AS10
rifci4£ ? i?-— s T
imcu. i^isis^s ?^c^
it'S-S-fSci^-- a I 2 ••
SiSl^S.^.^-^ -^^-^222,'
ebJx^SS-^t1*- J^'-eoiTCT?-:
»o55.eb^-B-^iC ewci o»ia*
> a • • 1 a » ^ 2 ? Z T = -5 «va • S S « .
! 8 ! S 85
<*^ -^
•? T
«; i
a 7
Mrcury (7439-97-0)
•wthacry 1 on 1 ir i 1 *b ( 1 20-1
MothcMiyl (107K2-77-6)
Mthonyclort (72-43-6)
•tothyl broaldo (broiMMMtl
(74-83-0)
cu
• 1 1 1 1 1
W i I i i i
4*
U
*^
1
1
1
1
i
1
!
1 1
« 1
M* S «
*" w ^^ flD
O Ok **. ^. i ^
1- « » •» ~ W 1
O I 1 1 M 1 1
- « CD - l IA
X • ^ ^ r» r*.
u M i i a> —
^* ^ ~ 5 *
? u T 8 ^ 0 '
) U ^% ^« »« • '
e 9 *~ ~ — jt i
> •- e a • i
^
C-44
-------
,
5"
9
3
O
CM
111 I
*\
h-
i
J
j'.
ii
J?
J -
M
8
w
L
j
UX
e >•*
x"*I
X J
± 5
•
1 • —
X •
>
' • •
»• ^
1.
•
*>
e
r f*
i
•
u
j
; w
'
: s
a
•4
1
M 1
^^ ^4
3J
e •
f 5
a b
1 X
MA.
A
!
9
1 10
•" 10
„,* t
! T
o •
"5 5
u Ai
e o
I I,
o —
X •-
X V.
I
•
X
0
•o
i
j
J
°w
•
O
j
"•
1
1
10
*•*
1
o
^
<&
u
e
X
5
f
8
i
•
"""" £•
i
?«
?
« 3
C 3
K S
i —
S !
? _
1 1
15 S
«^ i
*
? £
o
*> u
Jl X
X —
X T)
J 0
X X
X X
1 1
?
CM
1
^
>
—
1
*^
1
•
1
•
5
•
e
X
X
I
I ,
oT
0
t
*«
^ CM
1 A
x a
u •^
>
^ ^
t f*
1 (t
S
1
oT S
i *
m 10
0 N
CM X
s> *
x T
• *>
e e
• ••»
u •
e —
x
x w
X •
I N
,
M
N
1
M
CM
«
S
J
_
o
•
0s
1
1
S
CM
e
0
x
L.
a
X
x
f
i
i
i
m
m
i
<•)
j
~e
u
__
o
1
1
^
m
a
m
i
0
m
u
x
x
L
U
0
JC
u
e
x /«. *» i i CM a *s
S« » i • 0 r« CM 10 u> 10 e
010** i i « » i i •
>^ ^ 0 ij • 0 ot i i a •« w
• o 7 ^" a c S a 0 • 0 •
• e CM e o CM S" 0 >«•(». 0 •••• c
N • «. c x it 2 >^««<-'«^f
Sx ^ • a 0 CM *-• • •—• i
*» xo « «n >-> J • • x
f^SSSS? SSSJJfcl'gif
MT 1 X X X X X -W ^ 0 U • X — • 9 i
1 ^ V U ^ U ^ • ^ Al U U C A • ^^
• W»— •* • « • — — - — — — — — I 5T
tt ft • . • • • tXXXXXXXXX
800 CM 90
CMI i » i iS00miM00i i
0I« II 1 1 •" » M • C CM0II
0
~+ «
IA 0
1 10 10 1 1"» •» 0
CM 1 /-> 1 ^-. O) II I
i 10 i a i i *. m «« 9
9 **CMI«CM^ II 010
^ •» ••09910 10CM iv ^
*.»<-, X i 0 i ~^ 1 00 9 9
ir»a «000 0 CM** 10* I
a<— i^-w^^r-00 10 i -o ^ CM
-i mm— 0 •— • c i N* ^ — 10 —
IUVIX0U • f- ^— J .— 10 C 1 1
r««»ll00CM>UN0 * CM~«0I^
10 • CM a i ^^x « c >-• x *> ^\ i — • x ui 0
i099i'» «x« • • u> « y i -
— '-o— «ixc«x — x-> c y i us • 9>^
9a««*eoocM— «T 1*01.2
x x — •-'C-OXBI.CI e <« m •> 0 a x
• L ^ — — cxwiu>«yi0c>«»
•«• o — x»** — xa — — i00« — CM —
e — •Bccuoooo»»3>-'>-'yii u
• »>—«.-«>kkbk0cu y •>
x«»N9*eooo»«ux • f e • —
y L e c o- «J — — — — v— | • c 3 3 -o c
• 0 X X X X X I C .--.-.-•
— — OXUCuyyU— C— OXABfiC
• •— ow3a***o*xa&n«aa
jt jt L u •- — w^4j*)cj:e««««ca
yy^*>e6cece«atooj^oo
— — — — 1 ••••••Xj.XXXOOkl.
e e c c T o a. a a a a E a a. a. a & a a.
i
i
^»
*
i
•J
M
b
U
I
I
x
1
t
0
7
fNk
^
C
j
X
o
u
a.
e
e
3
|
C-45
-------
f
1
I
.
^^
1
e
g
u
*N 1
UJ
a
^
U
j
^
*;
•2
j
b
e
**
i
^J
j
u
c
•
! •*
5
fj
' •
X
|
»M
L
* ^
e
i
i
U
•<
|
"•
±
^»
•
u
J
<•»
X
«*
1
>»
i
•
*
4
M
«
«
d
1
!
.c
V.
i ^
i «
*^
T ^
flb **
w 1
1 «
8 J
• w
e ^
a
2 4
*•;
* J
X S
u &
i §
• —
- •
M e
» •
^
|
/•K
' x*. m
t OB i
n i a>
•• to »
C4 1
i l CD
!! f^ Q
*•• ^^
^^
si
b 3
X *O
1
1
^
1
P4
«
«
^^
_c
•
^
•
o
~
1
N
•
1
2
!
0
^
^
b
o
x
••
b
+>
1
•
•
1
1
^K
INk
1
s
s
«
***
•
^
1
^
>t
1
1
^
u>
1
8
a
*.
^
•
e
b
o
^
*•
b
i
»
"^
i
i
^,
\
*•
2^
i
p>
U>
^
j
*e
»>
u
e
*
p«
•
X
1
e
3
^
o
e
Q
b
O
X
b
1
1
^^
A
1
•^
<*>
^^
**
4
2
o.
•
&
u
e
^^
i
a
e
t
U)
"
j
^^
7
«
M
M
M
!••
U
• 1
c 1
i
i
,
*
^
^
i
g
i
g
•
X
a.
o
b
o
^
.«
^
«>
i
u>
•
*-
X
e
• <^
o 2
-S
*»oe
U M
e j*
u
b •
— u
e
• 4.
I*
c —
^ ••
• b
b 4»
j|
^
e -9
c
^ £
a
• b
5 J
• e
— •
*
b •
3 t
Ik •
3
• 0
b ^
• U
• M
3 •
^
•o e
e
• a
e
Is
O 3
if
b e
O 0
a.
a> •
e 's
3 b
• •B
b •
j» 2.
0 i.
• a
i — •
9 °
: §.§
' X •»
u
X •
•> b
^ ^
• 0
t u e
x •
• a
C — • o
•:l£ .5
i • • ae u
<*» • b
~*l '
. • £ u
•
a
§
x
u
o
*»
^
0
•«^
^J
•
^
b
U
e
^
A
X
—
^
g
J
b
«»
e
jj
^
•o
e
«i
J
-
n
•^
x
o
•
tu«nt
•~
4^
m
e
o
u
X
bJK
f
I
C-46
-------
APPENDIX D
SAMPLING AND ANALYSIS PROCEDURES
D-l
-------
SECTION 1
INTRODUCTION
As Indicated throughout the preceding text, for a given facility the most
reliable estimates of contaminated part1culate emissions are obtained using
site-specific information as developed through a well-designed S&A program.
This appendix outlines procedures and general Information to be used in devel-
oping an S&A program. The appendix borrows extensively from information
contained in References 1 and 2.
The overall goal of the S&A program 1s to collect and analyze (for
physical and chemical properties) "representative" samples of the soil and
surface material at TSDFs. The Information generated from the exercise, in
turn, feeds directly into the emission factor models discussed in the main
text. The sampling parameters of concern include the following:
Silt content
• Silt loading (for paved surfaces)
• Moisture content
• (a) level of contamination
One of the most critical decisions required in development of the S&A
plan involves specification of a sampling design. In practice, sampling
designs can be divided into two broad categories—those that are statistical
or probability based, and those that are based on professional judgment.
Whenever possible, statistically based schemes are preferred as resulting data
often can be reduced to provide measures of sampling and analysis reliabil-
ity. Probably the most widely applied statistical schemes are: (1) simple
random and (2) systematic random sampling. Both of these schemes proceed with
reference to a sample grid like that shown in Figure 1-1. Under the simple
random design, a predetermined number (n) of individual samples are collected,
one from each of n randomly selected grid cells. Under the systematic design,
the initial sampling grid cell is randomly selected; subsequent samples are
then collected at a regular interval (e.g., every other grid cell) in one or
more directions from the starting point. Detailed discussions of statistical
sampling designs can be found in References 3 and 4.
Implicit in the use of statistically based schemes is a substantial
commitment of resources for both sampling and laboratory analyses. If
resources are very limited, it may be desirable to use judgment sampling to
produce initial data for emissions estimates. In essence, judgment sampling
implies that the person responsible for collection can choose sampling
location(s) that are broadly representative of the source in question.
D-2
-------
CFSRiO
WO TV. j-
PERCENT
LENGTH
T
_ I
7
SOS +•
30S
-f-70?:
ICS
30f.
50*
J
0
i O
*
O
O
*.....
O
O
O
O
O
O
O
O
O
O
0
O
0
c
O
0
O
O
O
-
X X OUTlcS SP'KE « STAKES AND ?LASriC ?H<-.GiU(5
-*- PROCESS BOUNDARY AND FUGS OR STAKES
O SAMPLING LOCATION '
/
Figure 1-1. Sample grid for hypothetical TSDF source.
D-3
-------
SECTION 2*
COMPOUNDS MEASURED AND DETECTION LIMITS OF ANALYTICAL METHODS
The compounds measured in assessing the degree of contamination of the
soil fractions are metals, cyanide, semivolatlle organics, oil and grease
(land treatment samples), and pesticides/PCBs. The list of semivolatile
organics and pesticides for which analyses are conducted and their detection
limits as presented were developed from the Hazardous Substance List (HSL) in
EPA's Contract Laboratory Program (CLP) Statement of Work.s The CLP was
chosen because the large number of samples that have currently been analyzed
through this EPA program provide the maximum opportunity for technology
transfer studies. The CLP also draws heavily from the procedures in EPA
SW-8468 which allows other data to be utilized.
The detection limits for pesticides (Table 2-1) and semivolatile organics
(Table 2-2) are based on extracting 30 g of material as specified by the low-
level extraction procedure in the CLP. If other organic: material is present
in significant quantities, sample cleanup procedures will have to be used.
Ultimately, the laboratory may have to dilute the sample extract to protect
the analytical equipment, and these detection limits may not be achievable.
The quantifiable detection limits listed for the metals (Table 2-3) are
those that can be obtained for the compounds listed using the analytical meth-
ods described in this protocol. In the case of chromium, samples are ini-
tially analyzed for total chromium content. If the results for any sample
show a relatively significant concentration of chromium, then another aliquot
of that sample is analyzed for hexavalent chromium (the most toxic form) using
procedures in SW-846. Cyanide content is determined colorimetrically follow-
ing EPA Method 335.26 with a detection limit of 0.5 yg/g.
If the user of this protocol has knowledge of the compounds present or is
only interested in a few compounds, the quantifiable detection limits may be
improved through the use of more specific analytical techniques and more
sophisticated sample cleanup procedures.
The material in this section is taken directly from Reference 2.
D-4
-------
TABLE 2-1. PESTICIDE DETECTION LIMITS
1 ALDRIN 8
2 Alpha - BHC 8
3 Beta - BHC 8
4 Delta - BHC 8
5 Gamma - BHC 8
6 CHLORDANE 80
7 4,4'-ODD 16
8 4,4'-DDE 16
9 . 4.4'-DDT 16
10 DIELDRIN 16
11 ENDOSULFAN I 8
12 ENDOSULFAN II 16
13 ENDOSULFAN SULFATE 16
14 ENDRIN 16
15 ENDRIN KETONE 16
16 HEPTACHLOR 8
17 HEPTACHLOR EPOXIDE 8
18 TOXAPHENE 160
19 AROCLOR 1016 80
20 AROCLOR 1221 80
•
21 AROCLOR 1232 80
22 AROCLOR 1242 80
23 AROCLOR 1248 80
24 AROCLOR 1254 160
25 AROCLOR 1260 160
D-5
-------
TABLE 2-2. SEMIVOLATILE ORGANIC DETECTION LIMITS
Number
1
2
3
4
S
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
Compounds
ACENAFHTHENE
ACENAPHTHYLENE
ANTHRACENE
BENZO (•) ANTHRACENE
BENZOZC ACID
BENZO («) PYRENE
BENZO (0hl) PERTLfiWE
BENZO (b) rLUORANTHENE
BOKO (k) FLUOSANTHENE
BENZYL ALCOHOL
BZS (2-CHLOROETHOXY) METHANE
BIS (2-CHLQftOETHYL) (LTHUt
BIS (2-CHLOROZSOPROPYL) ETHER
BZS (2-fiTHXU£JUfL) PHTHALATE
4-BROMOPHENYL PHENYL ETHEB
BUTYL BENZYL PHTHALATE
4-CHLOROAMZLZNE
4-CKLORO-3-METHYLPHENOL
2-CHLORONAPHTHALENE
2-CHLOROPHENOL
4-CHLOROPHENYL PHENYL ETHER
CHRYSENE
OZBENZO («.h) ANTHRACENE
DZBENZOFUKAN
1.2 OZCHLOROBENZENE
1.3 DZCRLOROBENZENE
1.4 DICHLOROBENZENE
3.3' -OZCHLOROBENZIOINE
2. 4-OICHLOROPHENOL
DZETHYLPHTHALATE
2 . 4-OZMETHYLPHENOL
OZMETHYL PHTHALATE
DZ-M-BUTYLPHTHALATE
2 . 4-OXNZTROPHENOL
2 . 4-OZNZTROTOLUENE
2 . 6-OiflH'KUI'OLUENE
DZ-M-OCTYL PHTHALATE
FLUORANTHENE
FLUORENE
HEZACHLOROBENZENE
BEZACHLOROBUTAOIENE
HEZACHLOROCYCLOPENTAOZENE
BEZACHLOROETHANE
ZNDENO(1.2.3-ed) PYRENE
ZSOPRORONE
2-METRYL-4 . 6-OZNTTROPHENOL
2-METHYLNAPHTHALENE
2-METRYLPHENOL
4-METHYLPRENOL
NAPHTHALENE
2-NZTROANZLZNE
3-NZTROANZLZNE
4-NZTROANZLINE
NITROBENZENE
2-iiL2KOni£NOL
4-NZTROFHEROL
N-NZTROSO-OZ-if-PROPYLAMZNE
N-N1TROSODZPHENTLAMUJE
PENTACHLQROPHENOL
PHENANTHRENE
PHENOL
PYRENE
1 . 2 . 4-TRZCHLOROBENZENE
2.4. 5-TRZCHLOROPHENOL
2.4. 6-TRZCHLOROPHENOL
' Detection
(ug/kg)
330
330
330
330
1600
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
1600
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
1600
1600
16OO
330
330
1600
330
330
1600
330
330
330
330
1600
330
D-6
-------
TABLE 2-3. METALS, MEASUREMENT METHODS, AND QUANTIFIABLE DETECTION LIMITS
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
Element
Aluminum (Al)
Antimony (Sb)
Arsenic* (As)
Barium* (Ba)
Beryllium (Be)
Bismuth (Bi)
Cadmium* (Cd)
Chromium* (Cr)
Cobalt (Co)
Copper (Cu)
Iron (Fe)
Lead* (Pb)
Manganese (Mn)
Mercury* (Hg)
Molybdenum (Mo)
Nickel (Ni)
Osmium (Os)
Selenium* (Se)
Silver* (Ag)
Thallium (Tl)
Vanadium (V)
Zinc (Zn)
Measurement Method**
ICAP
GFAA
GFAA
ICAP
ICAP
ICAP
ICAP
ICAP*
ICAP
ICAP
ICAP
ICAP
ICAP
Cold Vapor AA
ICAP
ICAP
ICAP
GFAA
ICAP
GFAA
ICAP
ICAP
Quantifiable
Detection Limi
(ve/g)
40
1.0
1.0
0.7
0.1
10.0
0.4
0.7
0.7 , '
7-3
100
10.0
5.9
0.25
9-0
2.2
4.0
1.0
10
1.0
3-9
0.2
Eight RCHA metals
ICAP » Inductively-Coupled Argon Plasmography
GFAA * Graphite Furnace Atomic Absorption
AA » Atomic Absorption
Other methods are used to measure hexavalent chromium (Cr IV) , if appropriate
(see page E-4) .
D-7
-------
SECTION 3
FIELD SAMPLING PROCEDURES
3.1 SAMPLING METHODS
The procedures for sampling the soil/surface material at HW TSDFs can be
divided Into four basic types—sweeping, scooping, coring, and vacuuming.
Sweeping refers to the use of a whisk broom and dust pan to remove loose
surface material from the underlying hard packed surface. The material should
be swept carefully so that the fine dust is not injected into the atmo-
sphere. The hard road base below the loose surface material should not be
abraded so as to generate more fine material than exists on the road in its
natural state.
Scooping refers to the use of a small garden spade or comparable
implement to obtain samples of near surface material from a potential dust
emitting source. Note that Reference 2 calls for the use of small disposable
scoops that may be appropriate depending upon the nature of the material being
sampled.
Coring refers to the use of a simple coring tube to extract samples to
nominal depths of 15 cm (6 in). Two types of coring tubes are employed: one
made of stainless steel (to collect samples for organic analyses) and one made
of PVC plastic (to collect samples for metals analyses). The procedure
involves driving the core tube into the surface (with a mallet or other com-
patible tool), extracting the core tube from the soil, and forcing the soil
core into the sample container by pushing a wooden dowel through the tube.
Vacuuming refers to the use of a common household vacuum cleaner to
remove surface material from a paved roadway. Note that the vacuum bag should
use removable, disposable paper bags.
In the following table the sampling procedures are cross-referenced
against potentially applicable source categories.
D-8
-------
Source Category Sampling Procedure
Paved roads Vacuuming
Unpaved roads Broom sweeping
Waste piles Scooping, coring
Dry surface Impoundments Sweep1nga
Landfills Scooping, sweeping
Land treatment Coring, scooping
Stabilization/solidification Scooping, coring
a Assumes wind erosion 1s the only particulate
generating mechanism.
3.2 UNPAVED ROADS
For unpaved roads, MRI typically has recommended collection of a minimum
gross sample of 23 kg (50 lb) for every 3.8 km (3 mi) of unpaved road. The
incremental samples from unpaved roads should be distributed over the road
segment as shown in Figure 3-1. At least four Incremental samples should be
collected and composited to form the gross sample. Figure 3-2 presents a data
form to be used for the sampling of unpaved roads.
3.3 PAVED ROADS
In industrial facilities, like HW TSDFs, MRI typically has recommended
that one gross sample should be obtained for each road segment in the facil-
ity. The gross sample should consist of at least two separate increments per
travel lane, or each 0.5 mi length should have a separate sample.
Figure 3-3 presents a diagram showing the location of incremental samples
for a four-lane road. Each incremental sample should consist of a lateral
strip 0.3 to 3 m (1 to 10 ft) 1n width across a travel lane. The exact width
is dependent on the amount of loose surface material on the paved roadway.
For a visually dirty road, a width of 0.3 m (1 ft) is sufficient; but for a
visually clean road, a width of 3 m (10 ft) is needed to obtain an adequate
sample.
The above sampling procedure may be considered as the preferred method of
collecting surface dust from paved roadways. In many instances, however, the
collection of eight sample increments may not be feasible due to manpower,
equipment, and traffic/hazard limitations. As an alternative method, samples
can be obtained from a single strip across all the travel lanes. When it is
necessary to resort to this sampling strategy, care must be taken to select
sits that have dust loading and traffic characteristics typical of the entire
roadway segment of interest. In this situation, sampling from a strip 3 to
9m (10 to 30 ft) in width is suggested. From this width, sufficient sample
can be collected, and a step toward representativeness in sample acquisition
will be accomplished.
D-9
-------
•o
10
o
O)
id
C
0
O
1/1
0>
en
o
-3
c
_2
cs
C
u
o
CN
o
c
\
S3
•
II
CU
U
O
rd
U
O
I
m
j-
3
O>
=M
D-10
-------
Scmple
No.
Time
^
Laccfion
Surface
Area
Deoth
Gucnf ify
of Scrr.sie
Use cooe given on plant rr.ap for segment idennficatTcn and indicate sample
location on map.
Figure 3-2. Sampling data form for unpaved roads,
D-ll