TECHNICAL BACKGROUND DOCUMENT
 POPULATION RISKS FROM INDIRECT EXPOSURE PATHWAYS, AND
POPULATION EFFECTS FROM EXPOSURE TO AIRBORNE PARTICLES
            FROM CEMENT KILN DUST WASTE
                 Office of Solid Waste
          U.S. Environmental Protection Agency
         August 1997 Draft, Do Not Cite or Quote

-------
                                  TABLE OF CONTENTS

CHAPTER 1
INTRODUCTION AND OVERALL FINDINGS   	  1-1
       1.1     Introduction 	  1-1
       1.2     Overall Findings	  1-2

CHAPTER 2
POPULATION RISKS FROM INDIRECT EXPOSURE PATHWAYS  	  2-1
       2.1     Background and Scope   	  2-1
              2.1.1   Starting Point of this Analysis	  2-1
              2.1.2   What is Included and Excluded from the Scope of this Analysis	  2-3
       2.2     Summary of Population Risk Assessment Framework  	  2-4
              2.2.1   The Risk Scenario Being Assessed	  2-4
              2.2.2   The General Approach for Assessing the Risk Scenario  	  2-8
       2.3     Summary of Overall Approach	  2-9
              2.3.1   Tier 1:  Screening Analysis	  2-10
              2.3.2   Tier 2:  Detailed Analysis	  2-10
       2.4     Results	'.	  2-16
              2.4.1   Results of the Tier 1 Analysis	  .  2-16
              2.4.2   Results of the Tier 2 Analysis	  2-18
              2.4.3   Results Extrapolated to Full Universe	  2-24
       2.5     Major Limitations and Uncertainties  	  2-27
              2.5.1   Limitations and Uncertainties
                     for  the Indirect Exposures Analysis  	  2-27
              2.5.2   Alternative Approach to Calculate Population
                     Risks and Associated Limitations	  2-31

CHAPTER 3
POPULATION EFFECTS DUE TO PM EXPOSURES	  3-1
       3.1     Background and Scope   	  3-1
              3.1.1   Starting Point of this Analysis	  3-1
              3.1.2   What is Included and Excluded from the Scope of this Analysis	  3-2
       3.2     Summary of Overall Approach	  3-3
              3.2.1   Identify An Appropriate "Risk Descriptor" for PM Exposures	  3-3
              3.2.2   Develop the Overall Modeling Framework  	  3-3
              3.2.3   Select Model(s) for Estimating Emissions and Dispersion	  3-6
              3.2.4   Determine Sources of Emissions  	  3-8
              3.2.5   Identify and Develop  Data for Modeling Inputs	  3-9
              3.2.6   Select Facilities for Modeling  	  3-14
              3.2.7   Define Exposure Points  	  3-16
              3.2.8   Model Emissions and Dispersion
                     and Characterize Populations  Effects  	  3-16
       3.3     Results	  3-17
              3.3.1   Emissions from Selected Facilities	  3-17
              3.3.2   Ambient PM Concentrations and Exposed Populations
                     at the Highest-Emitting Facility   	  3-18
              3.3.3   Exposed Populations at 52 Cement Plants Examined 	  3-20
              3.3.4   Exposed Populations at All Cement Plants  	  3-22
       3.4     Major Limitations and Uncertainties  . . :	  3-37

-------
APPENDICES
       A:     Exposure Parameter Values Used for CKD Risk Assessment	A-1
       B:     Individual Risk Estimates from RTC and NODA	B-l
       C:     Results of the Tier 1 Screening Analysis  	C-l
       D:     Facility-Specific Raw Data and Calculations, Fish Ingestion Pathway  	D-l
       E:     Example of Facility-Level Raw Data, Vegetable Ingestion Pathway   	E-l
       F:     Equations and Background Information from AP-42  	F-l
       G:     Comparison of Previous and Current Modeling of Paniculate
              Matter at Two Cement  Plants	G-l

-------
                                       CHAPTER 1
                 INTRODUCTION AND OVERALL FINDINGS
       1.1     INTRODUCTION

       This Technical Background Document analyzes the extent to which current practices for
managing cement kiln dust (CKD) onsite at cement manufacturing plants pose a health risk to nearby,
offsite populations.  The study focuses on:  (1) population risks from indirect, or foodchain, exposure
pathways; and (2) population effects from exposure to airborne particles.  This work builds on earlier
CKD analyses focusing on the health risks  to maximally exposed individuals, presented in the 1993
Report to Congress (RTC) on CKD1 and supporting documentation,2 the 1994 Notice of Data
Availability (NODA) on CKD,3 and a background document supporting the 1995 CKD Regulatory
Determination.

       The assessment of population risks from indirect exposure, presented in  Chapter 2, estimates
the number of cancer cases and the number of people potentially exposed above noncancer effect
thresholds through the ingestion of vegetables, beef and milk, and fish near cement plants.  This
assessment starts by eliminating from concern those facilities that have negligible potential for
significant population  risk, based on previous estimates of individual risk  at a sample of 82 facilities.
For remaining facilities, population risk for the vegetable ingestion pathway is calculated by combining
prior estimates of individual risk with estimates of nearby farmers and backyard gardeners determined
using census data.  Population  risk for the fish ingestion pathway is estimated using the prior
individual risk estimates along with numbers of recreational fishers that could be exposed, calculated
based on fish yield data from local streams. This chapter also includes a  discussion of the major
uncertainties and limitations associated with the assessment of population risks from indirect exposure.

       The assessment of population effects from exposure to airborne particles, presented in
Chapter 3, estimates the number of people  potentially exposed to fugitive CKD  at levels above the
National Ambient Air Quality Standards (NAAQS) for paniculate matter.  Both the existing NAAQS
for coarse particles and a new NAAQS proposed for fine particles are considered.  New modeling of
CKD emissions and downwind dispersion is performed for selected "high risk" cement plants,
substantially improving on the previous work by using a more sophisticated model, estimating
emissions from all CKD handling stages rather than just final disposal as  modeled previously, and
considering the effect  of terrain, among other refinements.  The concentrations of airborne particles are
then  overlaid on census block grids to estimate populations potentially exposed  above the NAAQS.
This chapter also includes a discussion of the major uncertainties and limitations associated with the
assessment of population  risks due to exposure to airborne particles.
    1 Report to Congress on Cement Kiln Dust. EPA Office of Solid Waste, December 1993.

    • Technical Background Document: Human Health and Environmental Risk Assessment in Support of the
Report to Congress on Cement Kiln Dust Waste.  EPA Office of Solid Waste, December 1993.

     Technical Background Document for the Notice of Data Availability on Cement Kiln  Dust: Human Health
and Environmental Risk Assessment in Support of the Regulatory Determination on Cement Kiln Dust. EPA
Office of Solid Waste, August 31, 1994.

    4 Technical Background Document on Potential Risks of Cement Kiln Dust in Support  of the Cement Kiln
Dust Regulatory Determination.  EPA Office of Solid Waste, January 31, 1995.


                             *** Draft, August 1997, Do Not Cite or Quote ***

-------
                                                1-2
        1.2     OVERALL FLNDINGS

        The overall results of EPA's effort to characterize risks via indirect exposure pathways to
populations living near cement facilities are summarized in Exhibit 1-1.  Results shown are for the
"most reasonable" estimates of risks extrapolated to the entire universe of cement plants; the bounds
on the most reasonable estimates are discussed in Section 2.4.3 of Chapter 2.
        The Agency estimates that exposures via
indirect pathways occurring in populations
within five miles of all cement plants nationwide
potentially result in a total of 0.04 excess cancer
cases over a 70-year period.  That is, exposures
would potentially lead to about 0.009 excess
cancer cases in the subsistence farmer
population, and about 0.03 excess cancer cases
in the "homegrown" population.   (Cancer cases
predicted for the recreational  fisher population
are negligible.)  The total population within five
miles of all cement facilities nationwide  is
approximately 3.4 million.5  Thus, the overall
population cancer risk can be characterized as
follows:   a total of 0.0006 excess cancer cases
per year could potentially occur within this
population of 3.4 million due to  indirect
exposures.
       Population Risk Terminology

EPA uses in this document the terminology
"population cancer risk" and "population
noncancer effects," to be consistent with existing
EPA guidance, in particular the March 21, 1995
memorandum  from the EPA Administrator
entitled Policy for Risk Characterization at the
U.S.  EPA. In  this document EPA uses three
specific terms: (i) "population cancer risk" to
denote "excess cancer incidence," i.e., the
number of excess cancer cases in the exposed
population over a 70-year period; (ii) "population
noncancer effects" to denote the number of
persons exposed to levels above the thresholds
for noncancer  effects; and  (iii) "population risk"
as a loose, collective term  to refer to both
population cancer risk and population noncancer
effects (recognizing that noncancer effects are
not equivalent to probabilistic  risks, per se).
        In terms of population noncancer effects
EPA predicts that, across all populations within
five miles of all cement facilities nationwide, a      ""^^~~11""^^~~""^^~~~~"^^~"^^^^~~~
total of about 1,040 people are potentially exposed via indirect exposure pathways to contaminant
levels above the hazard index.6  That is, about  6 individuals from the homegrown vegetable
population are exposed to contamination exceeding noncancer effects thresholds (i.e., hazard index
greater than  1).  At the same time, about 37 individuals from the subsistence farmer population and
about 1,000  individuals from the recreational fisher population are estimated to be exposed to
contamination exceeding noncancer effects thresholds. The overall population noncancer effects can
be characterized as follows: a total of about 1,040 people,  or less than one-tenth of one percent, from
among the population of 3.4 million within five miles of all cement plants nationwide is likely to be
exposed via indirect exposure pathways to  contamination exceeding noncancer effects thresholds.

       Note that the noncancer population effects estimates should  not be interpreted as "cases;"
instead, the estimates should be viewed simply as  the number of people with exposures above the
noncancer effects thresholds, or hazard index of 1. Unlike the estimates for population cancer risk, the
     This is an estimate based on site-specific data for 61 facilities and extrapolated data for the remaining 47
facilities.

     This total assumes that the three receptor populations, i.e., the subsistence farmer, "homegrown," and
recreational fisher populations, are independent and there is no overlap of exposures among the populations.
                                  Draft, August 1997, Do Not Cite or Quote ***

-------
                                                                      Exhibit 1-1
                                         Estimated Population Risks7 from Indirect Exposures to CKD

No population effects because
facility does not generate net CKD
Negligible" population risks
estimated by Tier 1 screening
Population risks estimated based on
Tier 2 methodology
Population risks estimated based on
extrapolation from "known
universe'"7
TOTAL
No. of Facilities
(percent of all
cement plants
nationwide)
22
(20 percent)
31
(29 percent)
29*
(27 percent)
26
(24 percent)
108
Population Cancer Risks
(i.e., number of excess cancer cases)
. "Homegrown"
Population
0
0
0.02
0.01
0.03
Subsistence
Farmer
Population
0
0
0.006
0.003
0.009
Recreational
Fisher
Population
0
0
0
0
0
Potential Population Noncancer Effects
(i.e., number of people potentially exposed to
levels above the threshold for noncancer
effects)
"Homegrown"
Population
0
0
4
2
6
Subsistence
Farmer
Population
0
0
25
12
37
Recreational
Fisher
Population
0
0
670'
330
1,000
" For Tier 1 screening, negligible risks mean the facilities (i) had facility-specific population cancer risks equal to zero or risks so low that they did not contribute
significantly to the total population cancer risk across facilities in the Tier 1 screening, or (ii) had facility-specific population noncancer effects equal to zero.
b A total of 26 facilities for the subsistence farmer and homegrown" populations and 4 facilities for the recreational fisher population (1 facility is common to
both sets).
f For the fish ingestion pathway, Tier 2 analysis needed to be conducted for only 4 facilities.
d The results presented here represent the "most reasonable" risk estimates based on extrapolation to the full universe of facilities; see Exhibit 2-9 for estimated
ranges.
    7 As explained in Chapter 2, page 2-9, EPA uses in this document the term "population risk" as a loose, collective term to refer to both population
cancer risk and population noncancer effects. Population cancer risk is used to denote "excess cancer incidence," i.e., the number of excess cancer cases in
the exposed population, and population noncancer effects is used to denote the number of persons exposed to levels above the thresholds lor noncancer
effects.
                                                                              1-3

-------
                                               1-4

noncancer population effects estimates are not based on probabilistic individual risk estimates.  One
can predict neither how many  of these individuals would actually have adverse noncancer effects as a
result of these exposures, nor when these effects are likely to occur in relation to the exposure
duration.  Also, one cannot compare directly the estimates for population cancer risks and population
noncancer effects. That is, although the estimates for noncancer population effects are numerically
higher (because all the people  exposed above a certain contaminant level are counted), the actual
number of people exhibiting the effects will most likely be lower, by an unknown amount, than those
counted as being exposed above the effects-based reference level.  The estimate of 1,040 for potential
noncancer population effects, therefore, should not be viewed as necessarily being several orders of
magnitude higher than the estimate of 0.04 excess cancer cases.

        As shown in Exhibit 1-1, all the population cancer risks and population noncancer effects
appear to be due primarily to 55 cement facilities nationwide; the remaining cement facilities (about 50
percent of total) have negligible population risks for the indirect exposure pathways.  Of the 55
facilities contributing to total population risks, 29 had risk estimates derived in the Tier 2 analysis and
the remaining  had risk estimates derived based on extrapolation.  For the 29 facilities evaluated in Tier
2, none individually had population cancer risks equal to or greater than one cancer case over 70
years.  In contrast, a total of 10 facilities had potential population noncancer effects of significance,
i.e., one or more people with exposures above the noncancer effects thresholds.  These 10 facilities
include seven  for the  subsistence farmer and "homegrown" populations, and four for the  recreational
fisher population (with one facility in common between the recreational fisher and the other
populations).

        Because the population risks for the subsistence farmer and homegrown vegetable populations
are estimated initially based on the number of people living within  five miles of the facilities, it is
important to describe in socio-economic terms the exposed populations for these seven facilities.
Census block-level data for socio-economic characteristics were  available  for six of these seven
facilities, and are  summarized  in Exhibit  1-2 (comparisons to county-, state-, and national-level data
are also provided).  The socio-economic findings indicate that there is no  definite trend across all the
facilities with  respect to whether certain types of subpopulations are more at risk than others.
Nevertheless, the  findings do indicate that, for some facilities, the population risks are concentrated
within certain  subpopulations (i.e., low-income or minority populations).   For example, two  facilities
(facilities 29 and 60)  have higher minority percentages for the five-mile radius than do the county or
the state as a whole.  Similarly, for the populations living within five miles of facilities 29, 30, 55,  and
60, a given household earns less on average than the average American household.  Particularly for the
vicinities of facilities  55 and 60, a given  household earns less on average  when compared to average
household in the county and the state as well.  Facility 29 appears to be located in a relatively poor
county, where there is a greater percentage of people living below poverty level for the five miles
surrounding  facility 29 than there is in the state as a whole.

        In terms of population effects due to exposure  to airborne paniculate matter released from
CKD waste management activities, EPA characterized  the numbers of residents around cement plants
who are exposed to ambient PM10 and PM25  concentrations  above the respective NAAQS.  The
Agency estimated that about 18 people living  around 82 cement plants may be exposed to airborne
                                 Draft, August 1997, Do Not Cite or Quote ***

-------
                                                1-5
                                            Exhibit 1-2
             Socio-economic Characteristics of Exposed Populations Within 5 Miles

Facility
I.D.
United States

29


30


55


60


62


66



Location
United States
State
County
5-mile radius
State
County
5-mile radius
State
County
5-mile radius
State
County
5-mile radius
State
County
5-mile radius
State
County
5-mile radius

Percent
minority
19.71
9.91
9.83
10.32
17.87
17.21
15.73
12.33
24.38
5.81
3.37
2.13
3.71
29.02
6.86
3.60
11.46
5.80
2.25

Median annual
household income
$30,056
$27,291
520,864
$24,131
$23,577
$17,945
$25,547
$26,362
$27,853
$26,177
$26,229
$25,116
$24,833
$39,386
$41,382
$36,465
$29,069
$32,890
$39,106
Percent below
the poverty
level
13.12
11.48
15.68
14.98
16.71
21.75
15.04
13.34
13.04
11.48
11.48
8.92
9.49
8.27
4.83
4.67
11.13
7.31
3.12
   Mote: "Percent minority" includes people of Black, American Indian, Eskimo, Aleut, Asian, or Hispanic origin.
PM concentrations in excess of the NAAQS.  It is not known what percentage of the population
exposed above the NAAQS is likely to develop  any morbid effects because the dose-response
relationship for PM exposures is not well defined.  In essence, the population effects results are being
driven by a very small number of facilities because the 18 people are estimated to be those living
within 100 or 200 meters of two cement plants.  All the other facilities in the universe analyzed were
predicted  to have zero population effects either because there are no residences within 100, 200, or
500 meters (44 facilities), CKD is watered  and unlikely to be emitted at levels above the NAAQS
(three facilities), or site-specific modeling and analysis indicate that no people live in areas where the
NAAQS are exceeded (three facilities).  As with the indirect exposures analysis, EPA derived a more
complete  picture of potential population effects due to PM exposures by extrapolating from results
within the known universe to determine the potential population effects for the full universe of cement
    8 The estimate of 18 people is based on an evaluation of 52 of the 82 cement facilities; based on analyses
conducted previously, the remaining 30 facilities were determined to have zero or negligible effects in terms of
PM exposures because they do not manage CKD on-site (see methodology and results presented in Technical
Background Document on Potential Risks of Cement Kiln Dusfin Support of the Cement Kiln Dust Regulatory
Determination, January 31,  1995).
                                  Draft, August 1997, Do Not Cite or Quote

-------
                                                1-6

facilities.  In sum, EPA estimated that, across all 108 facilities, a total of between 18 and 4,118 people
living within 500 meters of the facility boundary may be exposed to airborne PM concentrations in
excess of the NAAQS, with the best estimate being 2,378 people.
                               ** Draft, August 1997, Do Not Cite or Quote ***

-------
                                      CHAPTER 2
   POPULATION RISKS FROM INDIRECT EXPOSURE PATHWAYS
       This chapter describes the Agency's estimates of population risks due to indirect exposures to
contaminants in cement kiln dust. In Section 2.1 EPA provides the background and starting point of
this analysis, discussing what specifically is and is not included in the scope  of the analysis and the
resultant implications. Section 2.2 briefly lays out the conceptual framework of the risk scenario and
assessment process, describing specifically the key sources and pathways of exposure and also
receptors that are included in the assessment.  Section 2.3 presents a summary of the approach used to
estimate population risks. Section 2.4 presents a discussion of the results. Finally, Section 2.5
presents a discussion  of the major limitations and Uncertainties associated with the indirect exposure
analysis.
       2.1    BACKGROUND AND SCOPE

       The objective of this analysis is to characterize risks to populations that live near cement
facilities and are potentially exposed via "indirect exposure pathways."  As explained below, indirect
exposures in this context can potentially occur when populations that live, near cement facilities
consume  vegetables, beef and milk, and fish that have been contaminated by releases from the cement
kiln dust  waste management units (e.g., waste piles).

       2.1.1  Starting Point of this Analysis

       For this analysis the Agency used as a starting point the  risk assessment data and results
generated  previously for the Report to Congress on Cement Kiln Dust (RTC) and Notice of Availability
on Cement Kiln Dust (NODA).  From this previous work, estimates of individual cancer risks and
noncancer hazard indices were available for major pathways for  which "indirect exposures" can occur
for constituents released  from CKD waste piles. The previous work was conducted in two stages,
termed the original and expanded risk analyses.

       Original Individual Risk Analysis

       In its original analysis, the Agency conducted a quantitative assessment of the human health
risks associated with the on-site  disposal of CKD at five selected facilities. The methodology and
results of this analysis are presented in the Technical Background Document for the RTC. The five
facilities were selected to represent the two highest risk sites in each of the three pathways examined
(ground water, surface water, and air)  in a relative risk ranking of 15 case-study cement plants.

       The quantitative  analysis used EPA's MMSOILS, a screening-level contaminant release, fate,
and transport model, to estimate ambient concentrations of constituents of concern in ground water,
air, surface water, soils, and the  foodchain. Details on the transport methodology used  in MMSOILS
for each specific medium can be found in MMSOILS:  Multimedia Contaminant Fate, Transport, and
Exposure Model, Documentation and User's Manual (September 1992).
                                Draft, August 1997, Do Not Cite or Quote **

-------
                                               2-2

       The Agency used site-specific, regional, and national level data to characterize the five model
facilities.  These data represented the best readily available sources for simulating the environmental
characteristics at each facility. The data did not represent the level of detail and accuracy needed to
support a site-specific modeling assessment, but were consistent with the screening-level methodology
used in the Agency's analysis.

       The Agency's original analysis presents potential individual cancer risks and individual
noncancer effects associated with on-site disposal of CKD at the five modeled facilities through the
ground water, air, surface water, soil, and foodchain exposure pathways.  The analysis presents both
"best estimate" and "upper bound" exposure concentrations at each facility based on  best estimate and
more conservative values for key environmental transport parameters contributing most to ambient
concentration estimates.

       In order to examine the upper range of the national risk distribution  from the on-site
management of CKD, the Agency also conducted a sensitivity analysis of six higher  risk scenarios that
combined the five sets of actual facility characteristics with selected potentially high  risk practices or
settings observed by EPA.  Each  of these scenarios was constructed from the baseline case study
facilities and only a few key factors not attributable to the facility were modified to simulate  the
potentially  higher risk factor.  The six higher risk scenarios were  (1) disposal of CKD  with the. highest
levels of 2,3,7,8-substituted dibenzo-p-dioxins (CDDs) and dibenzo-furans (CDFs) measured by EPA;
(2) disposal of CKD with- the upper 95th percentile measured constituent metals concentrations, based
on combined EPA and industry samples from nearly 100 CKD  facilities; (3) simulation of a CKD pile
located directly adjacent to an agricultural field with uncontrolled erosion of CKD to the crops; (4)
simulation  of a CKD pile located directly adjacent to a surface  water body with uncontrolled CKD
erosion directly to the water; (5) simulation of CKD management in the bottom of a quarry that is
covered with water from ground-water seepage; and (6) simulation of exposures related to subsistence
level food consumption by  farmers and fishers.  The results of  these sensitivity analyses are also
presented in the Technical  Background Document for the RTC.

       Expanded Individual Risk Analysis

       EPA conducted further analysis of potential human health and environmental damages
associated with the on-site  management of CKD in order to expand the scope of its original analysis,
and to respond in part to public comments received on the RTC.  The methodology and results of this
expanded analysis are presented in the Technical Background Document  for the NODA. In its
expanded analysis, the Agency estimated order of magnitude risks not likely to be exceeded at an
expanded sample of 82 cement plants.1 First, the Agency evaluated CKD generation and
management practices at each facility to determine the potential for contaminant migration and
exposure via environmental media.  Where a release was found to be possible, the Agency evaluated
environmental factors to determine the potential for releases to  specific environmental pathways.
Facilities with negligible risks for all pathways because of CKD generation  and  onsite management
     At the time of the original analyses, EPA evaluated risks at 83 facilities from a total of 115 cement plants
nation-wide.  Since then, to reflect more current data on cement production in the U.S., EPA has revised the total
number of cement manufacturing facilities in the U.S. to be 108. Of this 108 cement plants, 82 are a part of the
original risk analyses conducted for" the RTC and NODA.


                             *** Draft, August 1997, Do Not Cite or Quote ***

-------
                                              2-3

practices were eliminated from further analyses.  For facilities not screened out, the Agency estimated
the risks associated with each pathway by matching the facility to one of the five originally modeled
cement plants to which it was most similar.  The Agency then qualitatively and quantitatively related
the original modeled risk estimates to risks associated with the facility being examined.  By mapping
each of the expanded sample facilities to one of the five facilities modeled in the RTC, the Agency
estimated a rough level of risk for the facilities.

       To relate the potential risks of the expanded list of sample facilities to the risks of one of the
five modeled facilities, the Agency first identified the RTC modeling scenario, either one of the five
baseline onsite CKD management scenarios (best estimate or upper bound) or one of the  sensitivity
analyses of higher  risk scenarios, that most closely resembled the conditions at each of the sample
facilities.  This selection of a "best match" facility  was  accomplished by comparing the sample facility
to the various RTC modeling scenarios along the dimensions of certain  risk-driving variables, such as
the size of the CKD  management unit, the extent of contaminant  features, the distances to water bodies
and potential receptors, and parameters that affect the mobility and  dilution of chemicals  in the
environment (e.g.,  stream flow and wind speed).  As a result of this step, the Agency  identified the
RTC modeled facility that was a best match  to the sample facility as well as a linear scaling factor that
could be used to quantitatively adjust the risk estimates for the best match RTC modeling scenario to
estimate  risk at the sample facility, accounting  for differences in key parameters at the two sites.  The
second step compared the selected RTC modeling scenario and the  sample facility strictly in terms of
relative chemical concentrations in CKD. The chemicals contributing most to cancer  and noncancer
risk estimates for the selected RTC scenario  were identified,  and  for those chemicals,  a ratio of the
concentration modeled in the best match RTC facility to the concentrations measured  in CKD from the
sample facility was developed.  This chemical concentration  ratio was then used with  the scaling  factor
based on other risk-driving parameters to adjust the selected RTC results to estimate risk at the sample
facility.

       2.1.2 What is Included and Excluded from the Scope of this Analysis

       This analysis includes assessing risks to populations due to indirect exposures. In general,
indirect exposures  can occur when contaminants are transferred from primary contaminated media to
secondary exposure media or pathways with which receptors (or populations of interest) come in
contact.  For example, contaminants released to air may pose risks  not only via inhalation (direct
exposure), but also via surface water, soil, and other media that become contaminated through
deposition (indirect exposure).  Atmospheric deposition can also lead to uptake of toxic contaminants
into aquatic and  terrestrial food webs. For example, herbivores can ingest contaminants deposited onto
plants; these contaminants may biomagnify at higher trophic levels leading to significant human
exposures via the food chain. For the purposes of this  analysis EPA uses the term "indirect exposure"
to describe the scenario when populations around cement facilities are exposed via ingestion of
vegetables, beef  and milk, and fish that have accumulated contaminants originating from the cement
kiln dust being managed  at the facilities.  Exposures via such food-chain pathways can be considered
indirect exposures  because the CKD contaminants are transported by one or more transfer media
before residing in the contaminated media with which the receptors come in contact.

        More direct exposure pathways (such as direct  inhalation of contaminants in the  air, and
dermal contact with  and/or ingestion of contaminated ground water, surface water, and soil) were
                             ***  Draft, August 1997, Do Not Cite or Quote ***

-------
                                               2-4

excluded from this portion of the analysis because they are addressed elsewhere (e.g., in looking at
impacts on populations due to airborne paniculate matter from CKD piles) or were deemed to be less
significant, relative to the indirect pathways, in terms of population  risks.

       This analysis includes assessing risks primarily to potential receptors who live within an "area
of influence" that is defined by a five-mile radius around each cement facility. Populations beyond
this area are excluded from the analysis because the Agency assumes that they are not affected
significantly, relative to populations within the radius, by ingestion of contaminated  foods that
originate from within the area of influence.  That is,  this analysis assumes that vegetables and beef and
milk "grown" in the area are not exported and they do not lead to exposures in areas remote from the
facility.  Exposures due to ingestion of contaminated fish are not limited to the "area of influence,"
however, since  fishermen from anywhere can be exposed if they fish in the contaminated streams near
the facility.  (Note that excluding populations beyond the area of influence is an  issue only for
estimating the noncancer population effects; given the linear, non-threshold calculation procedures
estimation of the magnitude of population cancer risks is not dependent on where the exposed
populations are located.  For comparison, EPA also assessed the feasibility of using an alternative
approach to calculate population risks that was not restricted to populations within the area of
influence; this alternative approach and associated limitations are discussed in Section 2.5.2)

       This analysis includes assessing risks directly at 82 of the total 108 cement  facilities. The
remaining 26 facilities  are excluded because a lack of relevant data  prevented them  from being
assessed  directly. To account for potential risks at these excluded facilities, however, results from the
82 facilities were extrapolated to the 26 facilities to derive  a composite picture of potential risks at the
full universe of cement facilities.
       2.2     SUMMARY OF POPULATION RISK ASSESSMENT FRAMEWORK

       2.2.1 The Risk Scenario Being Assessed

       The "risk scenario" assessed by the Agency for this population risk analysis is summarized in
Exhibit 2.1.

       As shown for a given facility, releases of contaminants occur from the CKD waste pile, which
is the source of contamination, due to  wind-blown erosion of paniculate matter and surface runoff of
dissolved and paniculate matter due to precipitation.

       The contaminants are then transported via (i) atmospheric dispersion and deposition and (ii)
overland runoff onto backyard vegetable gardens, agricultural fields,  and rivers/streams in the close
vicinity of the facility.

       Backyard vegetable gardens, agricultural fields, and rivers/streams in the vicinity become the
sources of the final contaminated media with which receptors come into contact. At the backyard
vegetable  gardens and agricultural  fields, the contaminants accumulate in the roots and leafy parts of
vegetables and  in grass.  For agricultural fields they also accumulate in the tissue and milk of cattle
                                 Draft, August 1997, Do Not Cite or Quote ***

-------
                                             Exhibit 2-1
                                           Risk Scenario
    Source
Transport
                                                                    Consumption
                                                                       Media
                     Receptors
 CKD waste
   pile and
handling train
                          airborne
                         paniculate
                       contaminants
                        backyard
                         gardens
                       agricultural
                         fields
vegetables
 beef and
  milk
backyard
gardener
                                                                  resident
                                                                  farmer
                                                                                         subsistence
                                                                                           farmer
surface
runoff
V ^


surface
waters



fish



recreational
fisher
                                                2-5

-------
                                               2-6

that feed on the contaminated grass and vegetables and incidentally ingest soil. Contaminants that
reach the rivers/streams accumulate up the foodchain eventually reaching the tissue of fish.

        Several different types of receptors are potentially exposed to these contaminated media. Two
different groups of receptors  or "populations of interest" were considered relevant in terms of ingestion
of contaminated vegetables and contaminated beef and milk.

        (i)      "Homegrown" population (i.e., "resident farmers" and "backyard
               gardeners")  within five miles of the facility. The term "resident farmers"
               refers to individuals or families living on farms that have annual sales of
               $10,000 or more. These individuals get some portion (i.e., assumed 25%) of
               their vegetable and  beef and milk diet from their farm.  The term "backyard
               gardeners" refers to individuals with vegetable gardens at their place of
               residence. These individuals are essentially part of the non-urban population
               that does not live on farms (i.e., part of the non-urban, non-resident farmer
               population).  These individuals get some portion (i.e., assumed 25%) of their
               vegetable diet from their garden.   For the vegetable ingestion pathway, both
               of these groups are  assumed to be exposed to the same levels of contamination
               (i.e., ingest the same amount, and therefore have the same level of individual
               risk), and therefore, are treated as a single population of interest, referred to as
               the "homegrown" population.  Note that risks to resident farmers due to
               ingestion of contaminated beef and milk are assessed separately from risks due
               to ingestion of vegetables.

        (ii)     "Subsistence farmers" within five miles of the facility.  The term
               "subsistence farmers" refers  to individuals or families with farms that have
               annual sales of less than  $10,000.  These individuals are assumed to get a
               somewhat larger portion  (i.e., 40%) of their vegetable and beef and milk diet
               from their farm.3  Note that risks to subsistence farmers due to ingestion of
               contaminated beef and milk and to ingestion of vegetables are derived as a
               single numeric estimate.  Individuals or families that have annual farm sales of
               $10,000 or less are  expected to earn $2,000 to $2,500 in actual income per
               year from the sale of agricultural products. Assuming that this cannot  be  the
               only source of income for these individuals or families, it is reasonable to infer
               that some, or perhaps most, of the  individuals/families with  farm sales  of
               $10,000 or less use farming to supplement their income and most likely their
               diets. Thus, they are more likely to eat a  larger portion of "homegrown"
               vegetables and beef and milk than  farmers on large, commercial farms. (One
               limitation of this method is that some farmers who report earning $10,000 or
     The "fraction of food from contaminated source" is taken from the assumptions used in the RTC and
NODA analyses (see also Appendix A). The original reference for the values used for this exposure parameter
is USEPA 1989 Exposure Factors Handbook. Office of Health and Environmental Assessment.  EPA/600/8-
89/043.
   3 Ibid.
                             ***  Draft, August 1997, Do Not Cite or Quote ***

-------
                                               2-7

               less from farm sales may not actually be growing vegetables, but instead, may
               be raising livestock or renting out their farms for that income.)  Note:  In
               consultation with county agricultural extension agents, sources in academia,
               and contacts at the Census of Agriculture, EPA determined that a reasonable
               method for estimating the number of subsistence farms was to  use a proxy for
               subsistence farms based on readily available  farm economics data.  Our
               method assumes that farms that earn less than $10,000 per year from the sale
               of their farm products are subsistence farms.  Data on the number of farms
               with earnings  less than $10,000 per year are  available in the 1992 Census  of
               Agriculture.

        Only one "population  of interest," the recreational fishers,  was considered for ingestion of
contaminated fish.  For the purposes of this analysis, EPA assumed that fish (generally, Trophic Level
3 and 4 fish) from the rivers/streams in the area of influence are  caught and consumed by  fishers  at
"recreational fisher" levels.  That is, the daily fish ingestion rate used in calculating the population of
interest is that for a typical recreational fisher.

        For the RTC and NODA, exposures to contaminated vegetables were estimated for the
agricultural field closest to the cement plant.  (The  level of contamination  in this field was predicted
through fate and transport modeling of the air deposition and overland runoff pathways, not accounting
for any situations in which CKD was applied directly to fields as a substitute for agricultural  lime.)
Thus, the individual risk was calculated for a hypothetical person ingesting contaminated vegetables
"originating" from this field.   In using this individual risk to estimate population risk in the current
analysis, EPA made the conservative assumption that the levels of contamination in vegetables from
any agricultural field or backyard garden within five miles of the facility are the same as those in the
vegetables from the closest agricultural field (i.e., the field used in modeling the individual risks).
Likewise, for the RTC and NODA, exposures to contaminated fish were estimated for the  "fishable"
water body that was closest to the cement plant and could potentially be contaminated.  The level of
contamination in this water body was predicted through fate and  transport modeling of the following
pathways:  air deposition, overland runoff, and where applicable, ground water  discharge to surface
water. The individual risk was calculated for a hypothetical person assumed to be ingesting
contaminated fish  caught from this water body. In using the previously calculated individual risks to
estimate population risk for the current analysis, EPA assumed that (i) contaminants can reach all
rivers, streams, lakes, etc. located within five miles of the facility (the "area of  influence"); (ii) all
relevant  water bodies in the area of influence are contaminated to the same level as that estimated for
the closest water body; and (iii) all large streams and their tributaries can support fish populations that
can be caught and eaten by recreational fishers in the area of influence. The Agency did  not include
very small streams or intermittent streams in this analysis because it is unlikely that they can support
significant fish populations.
    4 Another population of potential interest is subsistence fishers; this category includes populations with fish
consumption rates at the subsistence level.  This population was not included in the current analysis because of
the greater uncertainty, in the absence of detailed site-specific surveys, that would be inherent in developing site-
specific estimates of subsistence fishers populations.


                             *** Draft. August  1997, Do Not Cite or Quote ***

-------
                                              2-8

       2.2.2   The General Approach for Assessing the Risk Scenario

       This analysis was conducted in accordance with EPA's existing guidance on risk
characterization and exposure assessment.  Guidance on risk characterization was provided by EPA's
Risk Assessment Council in a February 26, 1992 memorandum entitled Guidance on Risk
Characterization for Risk Managers and Risk Assessors, and more recently by the EPA Administrator
in a memorandum dated March 21, 1995 entitled Policy for Risk Characterization at the  U.S. EPA.
The Agency also relied on direction provided by EPA's most recent guidelines for exposure
assessment, i.e., Guidelines for Exposure Assessment; Notice, Federal  Register 22888, May 29, 1992.
One key principle provided by these guidance documents  is that consistent risk descriptors (i.e.,
consistent across Agency programs) should be used to represent the range of different exposure
conditions encountered (e.g., central tendency and high-end individual risk, risk to highly-exposed or
sensitive  subpopulations, population risk).

       The general approach for assessing the risk scenario consists of two main components:
characterizing individual risks and characterizing population risks.  All the "standard" steps of the risk
assessment framework, i.e., the  hazard identification, dose-response assessment, exposure assessment,
and risk characterization, were conducted for the first component originally as part of developing
estimates of individual risk for the RTC and NODA.  (The exposure assessment step in this earlier
work  included estimating intakes for the exposed individuals based on pathway-specific  contact rates
and exposure assumptions;  these are summarized in Appendix A of this document.)

       Note that no new estimates of individual risk were developed for the current analysis; all
population risk estimates were derived based on previously developed individual risk estimates.

       The population risk characterization component builds upon the exposure assessment and risk
characterization steps noted above.  For exposure assessment, the additional step involved developing
estimates of the number of people living near cement facilities  who potentially are exposed via the
indirect exposure pathways (e.g., resident farmers, recreational  fishers). For risk characterization, the
additional steps involved first selecting the appropriate population  risk descriptors, then selecting the
appropriate individual risk estimates to represent exposure and  risk levels  in the  populations, and,
finally, estimating the number of people living near cement facilities who potentially will exhibit a
cancer or noncancer effect due to exposure.

       Based on the EPA guidance documents, and given the  information on individual risk already
available in the RTC and NODA, the Agency used the following risk descriptors to characterize
population risks.

•      Population cancer risk. This descriptor corresponds to the "probabilistic number of
       health effect cases" noted in the guidance.  The Agency estimated population cancer
       risk, or the "excess cancer incidence" in the exposed population, by multiplying the
       individual cancer risk (i.e., average lifetime excess cancer risk for an individual) by the
       size of the exposed population of interest. The product was the number of excess
       cancer cases expected in that population  over a specified period of time.
                                 Draft, August 1997, Do Not Cite or Quote ***

-------
                                              2-9

•      Population noncancer effects. This descriptor corresponds to the "number of persons
       above... or within a specified range of some reference level" noted in the guidance.  To
       estimate population noncancer effects, EPA used the individual hazard index (HI)
       calculated previously to represent the individual HI to all individuals in the population
       of interest.  Thus, if the individual HI exceeded 1, indicating that the individual's
       estimated exposure exceeded the reference dose (RfD), EPA counted every member of
       the population of interest as having an HI greater than  1.

Note:  For simplicity, EPA uses in this document the terminology "population cancer risk" and
"population noncancer effects,"  which is in accordance with existing EPA guidance,  in particular the
March 21,  1995 memorandum from the EPA Administrator entitled Policy for Risk Characterization at
the U.S. EPA. Throughout the remainder of this document EPA uses three specific terms:  (i)
"population cancer risk" to denote "excess cancer incidence," i.e., the number of excess cancer cases in
the exposed population (while an estimate of individual cancer risk cannot be greater than  1, because it
refers to a probability, an estimate of population cancer risk can be equal to any number, because it
refers to the number of cases); (ii) "population noncancer effects" to denote the number of persons
exposed to levels  above the thresholds for noncancer effects; and (iii) "population risk" as a loose,
collective  term to refer to both population cancer risk and population noncancer effects (recognizing
that noncaner effects are not equivalent to risks, per se).

       Also  in accordance with the guidance, EPA used  the mean  or central tendency estimates of
excess individual  cancer risk or noncancer HI for the extrapolation  to population-level risk estimates.
Because the variation in risk within the "populations of interest" is  unknown, it was  most appropriate
to use central tendency individual risk estimates for the population  cancer risk and population
noncancer effects calculations.  In short, the high-end individual cancer risk and HI estimates, by
definition, are likely to apply only to a small fraction of the population,  so use of these descriptors
would likely  result in substantial overstatement  of the population risk. Thus, the Agency used the
following  data from the RTC and NODA for the population risk analysis:

       •       Individual risk estimates based on modeling using the median (50th percentile)
               constituent concentration  (rather than the  95th percentile concentration).

       •       Individual risk estimates that were denoted as "best estimate" (rather than those
               denoted as "upper bound").

In some cases, there was a range of risk estimates presented in the  RTC or NODA, often with only  the
upper end of the range noted.  Even  though the ranges do not indicate the most likely central tendency
estimate, EPA chose (as a conservative approach) to use the upper  end value of any range that was
presented  without a central tendency estimate.
        2.3     SUMMARY OF OVERALL APPROACH

        In sum, the overall approach for this analysis was to combine the facility-specific individual
risk estimates with facility-specific data on populations potentially exposed via the indirect pathways to
derive facility-specific population risks. Then, the facility-specific population risks were aggregated to
                             ***  Draft, August 1997, Do Not Cite or Quote •*«

-------
                                               2-10

describe population risk across the full universe of CKD facilities.  The Agency used a two-tiered
approach (followed by a final refinement step for the "homegrown" and subsistence farmer
populations) to systematically screen for those facilities that contribute significantly to population risks
and to iteratively derive refined estimates of the populations exposed.

        2.3.1   Tier 1: Screening Analysis

        The Tier 1 analysis was identical for the vegetable ingestion, beef and milk ingestion, and fish
ingestion pathways.  In Tier  1 EPA conducted a screening analysis to derive population risk estimates
using the following readily available data:  (i) the individual risk estimates that had been derived for
the RTC or NOD A for 82 facilities, and (ii) a conservative assumption that, at each facility, 100% of
the surrounding population within five miles is exposed to levels predicted for the individual risk
estimates (i.e., all the "populations of interest" defined above were set as equal to the entire population
within five miles of the facility).  The individual risk estimates for the relevant pathways and facilities,
derived for the RTC or NODA, are summarized in Appendix B  of this document.

        2.3.2   Tier 2: Detailed Analysis

        The results from Tier 1 screening of the beef and milk ingestion pathway, for resident farmers,
indicate that no facilities are  of concern for further analysis for this pathway. Thus, based on the
results of Tier 1, the Tier 2 analysis was conducted for only the fish  ingestion pathway - recreational
fisher population, vegetable ingestion pathway - "homegrown" population, and the vegetable and beef
and milk ingestion pathway - subsistence farmer population.  The focus of Tier 2, the detailed
analysis, was to develop more refined, facility-specific estimates of the  number of people who are
potentially exposed for the facilities identified as being of concern in Tier 1.  That is, in Tier 2  EPA
developed more accurate estimates of the "populations of interest" that were then used to recalculate
the population risks.  For more details on the data used to derive the populations of interest, refer to
Appendices D and E of this document.

        "Homegrown" and Subsistence Farmer Populations

        The step-wise  approach for developing the facility-specific estimates of the two "populations
of interest" for the vegetable and beef and milk ingestion pathway is described below,  with
calculations for Facility #60  shown as an illustrative example.  (Note that the assumptions  and  sources
of data cited for Facility #60 are the same as those used for all  the other facilities.)

        Step 1. Estimate the total  number of fanners within a five-mile radius of the facility
land area within the five-mile radius
land area within the county
* total farm
population in the
county
= the number of farmers
within five miles of
the facility
               land area within the five-mile radius = 78.5 mi
                             *** Draft, August 1997, Do Not Cite or Quote ***

-------
                                                  2-11

                total area within the county for Facility #60 = 568.4 mi2
                (source:  the USA Counties - (State) Home Page, accessed via the 1992 Census of Agriculture)

                total farm population within  the county for Facility #60 = 2,114 people
                (source:  the 1990 Census of Population and Housing)
                                       78.5 * 2,114 = 292 farmers
                                         568.4
        Step 2.  Estimate the number of subsistence farmers within a five-mile radius of the facility
number of farmers *
within five miles
of the facility
(from Step 1)
number of "subsistence farms" in the county =
total number of farms in the county
the number of
subsistence farmers
within five miles
of the facility
                number of farmers within five miles of Facility #60 = 292
                (from Step 1)

                number of "subsistence farms" in the county for Facility #60 = 123
                total number of farms in the county for Facility #60 = 821
                (source:  the 1992 Census of Agriculture)
                                   292  * 123 = 44 subsistence farmers
                                      821
        Step 3. Estimate the number of backyard gardeners within five miles of the facility
Step 3(a)
total population *
within five miles
of the facility
total urban population within the county =
total population within the county
urban population
within five miles
of the facility
                population within five miles of Facility #60 = 29,085

                total urban population within the county for Facility #60 = 37,181
                (source:  the 1990 Census of Population and Housing)

                total population within the county for Facility #60 = 46,733
                (source:  the 1990 Census of Population and Housing)
                               ***  Draft, August 1997, Do Not Cite or Quote ***

-------
                                               2-12
                                29.085 * 37.181-= 23,140 urban people
                                   46,733
Step 3(b)
       total population    -    urban population within five miles of     =     non-urban population
       within five miles        the facility (from Step 3(a))                    within five miles of
       of the facility                                                       the facility
               total population within five miles of Facility #60 = 29,085

               urban population  within five miles of Facility #60 = 23,140
               (from Step 3(a))
              29,085 - 23,140 = 5,945 people (non-urban) within five miles of the facility

Step 3(c)
non-urban -
population within
five miles of the
facility
(from Step 3(b))
number of farmers within five
miles of the facility
= non-urban, non-farm
population within five
miles of the facility
(i.e., potential backyard
gardeners)
               non-urban population within five miles of Facility #60 = 5,945
               (from Step 3(b))

               farmers within five miles of Facility #60 =  292
               (from Step 1)
                            5,945 - 292 = 5,653 potential backyard gardeners
               The number of backyard gardeners calculated in Step 3(c) refers to those
               individuals who, because they are non-urban, could potentially have backyard
               gardens.  EPA's Exposure Factors Handbook (August 1996 SAB Review
               Draft, Office of Research and Development, EPA/600/P-95/002Ba) notes that,
               based on 1986 data from the National  Gardening Association, 38 percent of
               U.S. households participated in home vegetable gardening. The Agency
               contacted the National Gardening Association and updated this information:
               based on 1995 data, 45 percent of U.S. households that can be classified as
               "rural" participated  in home vegetable gardening.  Thus, for this analysis, EPA
               assumed that 45 percent of the 5,653 potential backyard gardeners are more
               likely to be exposed via contaminated vegetables from home gardens. Data
               from the National Gardening Association also indicated that, on average, 22
               percent of U.S. households that can  be classified as "non-rural" (i.e., city,
                                  Draft, August 1997, Do Not Cite or Quote ***

-------
                                                2-13

               small town, and suburban) participated in home vegetable gardening.  Thus, in
               addition to 45 percent of the potential non-urban backyard gardeners, we
               assumed that 22 percent of the 23,140 urban people are likely to be exposed
               via the contaminated vegetables from home gardens.
        Step 4. Derive the two "populations of interest" within five miles of the facility

(i)  "Homegrown" population
   total fanners
   (from Step  1)

   subsistence
   farmers
   (from Step  2)

   resident farmers
potential backyard
gardeners
*
percentage of rural
U.S. households
participating in
vegetable gardening

backyard gardeners
more likely to have
actual home gardens
urban population
*
percentage of
non-rural U.S.
households
participating in
vegetable
gardening

urban population
likely to have
home gardens
"homegrown
vegetable"
population within
five miles of the
facility
  (292 - 44) + (5,653*0.45) + (23,140*0.22) = 7,833 people potentially exposed via "homegrown vegetables"
  for Facility #60
(ii) "Subsistence farmers"
             Subsistence farmers within five miles of the facility (from Step 2) = 44 people
        Recreational Fisher Population

        The general approach for determining the population of interest for the fish ingestion pathway
is to determine the number of recreational fishers that can be supported by the "standing stock" (i.e.,
the fish yields) of the water bodies in the area of influence based on county-specific steam data, fish
exploitation rates, and ingestion rates.  The step-wise approach for developing the facility-specific
estimates of the population of interest is demonstrated below, with Facility #62 as an example.
                                  Draft, August 1997, Do Not Cite or Quote ***

-------
                                          2-14

Step 1. Calculate the stream acres within a five-mile radius of the facility
            stream length
           within the five-
             mile radius
        stream length within the five-mile radius = (a) tributaries of streams/rivers: 24.5 miles
        (b) streams/rivers: 9 miles
        (determined from USGS topographic maps)

        stream width = (a) tributaries:  1/350 mile (b) major rivers: 1/70 mile
        (determined from USGS topographic maps)
     (a) 24.5 miles
1/350 miles
                              1/70 miles
                                                 127.1 stream acres
                                                within five miles of
                                                    the  facility
Step 2. Estimate the pounds of fish caught (i.e.,"harvested") per year within the area of influence
        standing stock = 89.5 Ibs/acre/year
        (determined from aggregate fish biomass data obtained from the local department of natural
        resources.  For this facility, fish biomass data were averaged for two sampling sites located
        within five miles of this the facility).

        stream acres =  127.1
        (from step 1)

        exploitation rate (i.e., percent of standing stock caught by  fishers) = 20%
        (default value, suggested for use in EPA's Hazard Ranking System (HRS) Final Rule, Federal
        Register Vol. 55, No. 241, December 14, 1990)
                            Draft, August 1997, Do Not Cite or Quote ***

-------
                                               2-15
        Step 3. Estimate the number of recreational fishers that can be supported by the harvest
               Ibs fish caught per year = 2,275.1 Ibs/yr
               (from step 2)

               percent of fish tissue (by weight) that is edible = 35%
               (default value, derived based on conversations with local fisheries authorities and data provided
               in EPA's Exposure Factors Handbook, (August 1996 SAB Review Draft, Office of Research
               and Development, EPA/600/P-95/002Ba)

               Ibs fish ingested per year by a recreational fisher = 5.86
               (derived based on the daily fish ingestion rate used in originally calculating individual risks
               from food chain exposure in the RTC/NODA)
2,275
1 Ibs/yr
                                                   5.86 Ibs/year/person    =
        There are several additional site-specific assumptions that EPA employed in determining fish
yields; these are described below.

        Facility #35

        For this facility, there was only one major river flowing within five miles of the facility.  All
other streams were intermittent; these can not support fish populations, and, thus, were not included in
this analysis.  Other water bodies within the area, such as lakes and reservoirs, were not included.

        Facility #37

        The land area surrounding this facility is primarily swamp which is known to support fish
populations.  From conversations with local authorities, EPA determined that fish population studies
have not been conducted for this swamp  area. Thus, the Agency was restricted to applying  standing
stock data from a river within the county where the facility is located to the area covered  by swamp.
                              ***  Draft, August 1997, Do Not Cite or Quote ***

-------
                                                2-16

 In this case, the number of stream acres for this facility consists of only the streams within five miles
 of the facility that are flooded year-round.

        Facility #62

        For this facility, EPA had standing stock data for two small streams located within five miles
 of the facility. In this case EPA  decided to include these streams in  the estimate of the number of
 streams miles within five miles of the facility.  Thus, the total number of stream miles includes (i) the
 miles of major stream within five miles plus its tributaries within five miles of the facility, and (ii) the
 two smaller streams for which EPA has data. Other smaller streams were not included on the
 assumption that they are too small to support fish populations that would attract recreational  fishers.

        Facility #81
        For this facility, the number of stream
miles within five miles of the facility consists of
the major river and its tributaries within five
miles of the facility.  Smaller streams were
excluded from the  analysis, as were lakes and
reservoirs.
        2.4     RESULTS

        2.4.1    Results of the Tier 1 Analysis

        The results of the  Tier 1  screening
analysis are discussed below.  For a more
detailed description of these results, refer to
Appendix C of this document.  Note that results
from Tier 1 screening of the beef and milk
ingestion pathway, for the "homegrown"
population, indicate that no facilities are of
concern for further analysis for this pathway.
As shown in Appendix C  for the beef and milk
ingestion pathway, for the "homegrown"
population, all facilities had no population
noncancer effects, and population cancer risks
were estimated to be  so low that they do not add
up to a single excess cancer case  across all the facilities.

        "Homegrown " and Subsistence Farmer Populations

        The results of the  Tier 1 screening analysis are shown in Exhibits 2.2 and 2.3 for population
noncancer effects and cancer risks, respectively.  In Exhibit 2.2,  facilities not shown had no population
noncancer effects, predicted based on the estimated individual noncancer effects. Exhibit 2.3 shows
     Population Risks of Significance

In this document EPA uses the terms
"significance" or "significant" in association with
population cancer risks or population noncancer
effects to describe the following:

•   Facility-specific estimates of population
    cancer risk that are relatively large
    enough that, when combined with
    estimates from other facilities analyzed,
    cumulatively contribute to more than 99
    percent of a total population risk
    calculated across facilities (and, in Tier
    1, a total population risk equal to at least
    one excess lifetime cancer case due to
    exposure).

•   Facility-specific estimates of population
    noncancer effects that correspond to at
    least one person exposed to levels
    exceeding the noncancer effects
    thresholds.
                                  Draft, August 1997, Do Not Cite or Quote ***

-------
                      2-17
                  Exhibit 2-2




Population Non-cancer Effects, Across Facilities
"Homegrown"

Plant ID

60






No. of People
Above Hazard
Index = 1
2. 91 E+04






2.91 E+04
"Subsistence Farmer"

Plant ID

55
66
60
30
29
62
72
No. of People
Above Hazard
Index = 1
7.25E+04
4.39E+04
2.91 E+04
2.02E+04
1 .25E+04
9.43E+03
5.84E+02
1 .88E+05
                   Exhibit 2-3



  Population Cancer Risks, Across Facilities
'"Homegrown"
Plant ID
55
66
63
60
25
7
49
29
33
62
57
44
54
22
72
80
42
53
30
46
18
4
67
15
83 .
No. of Cancer
Cases
7.18E-01
4.34E-01
3.71 E-01
2.88E-01
2.06E-01
1 .90E-01
1 .54E-01
1.24E-01
9.55E-02
9.34E-02
8.49E-02
6.55E-02
6.21 E-02
5.94E-02
5.78E-02
5.59E-02
3.21 E-02
2.46E-02
2.02E-02
1 .96E-02
1 .92E-02
1.82E-02
1 .74E-02
1 .58E-02
1.01 E-02
3.24E+00
"Subsistence Farmer"
Plant ID
55
66
62
44
54
72
63
60
30
67
4
61
33












No. of Cancer
Cases
7.18E+00
4.34E+00
9.34E-01
6.48E-01
6.21 E-01
5.78E-01
3.71 E-01
2.88E-01
1.82E-01
1.72E-01
1.39E-01
1 .05E-01
9.55E-02












1.56E+01

-------
                                               2-18

the contribution of each facility (denoted by Plant ID) to the total population cancer risks.  Facilities
not shown had facility-specific population cancer risks estimated to be so low that they did not
contribute significantly to the  total population cancer risk across facilities (only facilities that
cumulatively contributed to more than 99 percent of the total population risk are  shown).  The
screening results indicate that  26  of the 82 facilities  included in the analysis potentially have
population cancer  risks of significance, and 7  among these 26 also potentially have population
noncancer effects of significance.  That is, across these 26 facilities there is at least one excess lifetime
cancer case due to exposure (i.e., population cancer risk > 1 excess lifetime cancer case) and/or at least
one person exposed to levels exceeding the noncancer effects thresholds (i.e., population noncancer
effects > 1 person).

       Recreational Fisher Population

       The screening results indicate that only four facilities have population noncancer effects of
significance,  and no facilities have significant population cancer risks.  The results of this  screening
analysis are shown in Exhibit  2.4 (facilities are denoted by Plant ID).  Facilities not shown had no
population noncancer effects (predicted based on the estimated individual noncancer effects), or had
population cancer  risks so low that, across all facilities, they did not contribute significantly to the
total.

       2.4.2   Results of the Tier 2 Analysis

       The Tier 2 analysis was conducted for facilities having potentially significant population risks
in the Tier 1  screening analysis, i.e., 26 facilities for the "homegrown"  and subsistence farmer
populations, and four facilities for the recreational fisher population.

        "Homegrown" and Subsistence Farmer Populations

       The results of this  analysis (as well as the supporting data) are  presented in Exhibits 2.5 and
2.6.  For a more detailed description of the raw data used in this analysis, refer to Appendix D of this
document. In summary, the population cancer risk characterization shows that, across all  the people
living within five miles of the 82 cement plants examined for the NODA, less than one excess cancer
case can be expected in the exposed population  over a 70-year period.  That is, exposures via
vegetable and beef and milk ingestion or  vegetable ingestion would potentially lead to  about 0.08
excess cancer cases in the subsistence farmer population, or about 0.95 excess cancer cases in the
"homegrown" population, respectively.  In terms of population noncancer effects, across all the people
living within five miles of the 82 facilities, about 7,883 individuals from the "homegrown" population
are exposed via vegetable ingestion to contamination exceeding noncancer effects thresholds (i.e.,
hazard index greater than 1).  Likewise, 625  individuals  from the subsistence farmer population are
estimated to be exposed via vegetable ingestion to contamination exceeding noncancer effects
thresholds. It is unknown how many of these individuals would actually have adverse effects as a
result of these exposures (i.e., the noncancer estimates are not cases, but simply  the  number of
exposures above the RfD.)

       As an additional refinement of the estimate of the "homegrown" and subsistence farmer
populations potentially exposed, the Agency  assessed in more site-specific terms how far  away from a
                                 Draft, August 1997, Do Not Cite or Quote ***

-------
                                              2-19
                                                                 Exhibit 2-4
                                                     Population Noncancer Effects, by Facility
                                                              (from Tier I analysis)
                                                         Plant ID


                                                            62

                                                            35

                                                            81

                                                            37
No. of People Above
 Hazard Index = 1

     9.43E+03

     9.40E+03

     5.44E+03

     4.65E+03
given facility contamination is likely to
travel. For this assessment EPA first
determined,  by examining previous
MMSOILs modeling results, that
contamination at agricultural fields and
backyard gardens is due mostly to surface
runoff and only in small part to atmospheric
deposition.  Then EPA examined the
hydrologic/drainage patterns surrounding the
facilities to determine how far contamination
is likely to travel via surface runoff before
being "captured" by any kind of a channel
(e.g.,  stream).  The Agency examined
topographic  maps first for the  seven
facilities identified in Tier 2 as having
significant population noncancer effects and
determined that contamination from CKD
waste piles can travel between a few hundred feet to a few thousand feet via surface runoff before
being captured.  EPA measured the approximate distance that the contamination can travel via
overland runoff for each of the seven facilities and then recalculated the populations potentially
exposed using the smaller radii (determined from the maps) to define the areas of influence. EPA
initially focused the additional refinement on only facilities with population noncancer effects because
the total population cancer risk estimate for these populations was low, i.e., less than one cancer case
across all facilities. For the remaining facilities, EPA assumed that contamination is likely to travel up
to  a distance of 1.0142 miles,  and accordingly recalculated the populations potentially exposed using
this smaller  radius. (The distance of 1.0142 miles was determined to be the furthest that
contamination is likely to travel, among the seven facilities examined in detail.)  The results of this
analysis are  shown in Exhibit  2.7. Across all 26 facilities, approximately 25 subsistence farmers and
approximately 4 people from the "homegrown" population are expected to be  exposed to
concentrations exceeding the noncancer effects thresholds when  (i.e., hazard index greater than 1)
when topographic factors are considered.  Likewise, exposures via vegetable and beef and milk
ingestion or vegetable ingestion  would potentially lead to about  0.006  excess cancer cases in the
subsistence farmer population, or about 0.02  excess cancer cases in the "homegrown" population,
respectively.

       (Note  that a slightly modified methodology was used for estimating the number of subsistence
farmers within the smaller area of influence.  In cases where a single subsistence farmer was
"calculated" to be present,  the Agency assumed that a subsistence farm family was more likely to be
present.  The number of people in such a family  was estimated to be equal to the average number of
people per household within that county.  The census data indicated that approximately three people
were  present per household for counties representing all the seven facilities of interest; thus, for each
   5 Note that other fate and transport models may or may not support this finding, depending on the
complexity of their atmospheric transport and deposition algorithms.
                                 Draft, August 1997, Do Not Cite or Quote ***

-------
                Exhibit 2-5
Data Used For Estimating the Population Risks
I.O.
4
7
15
18
22
25
29
30
33
42
44
46
49
53
54
55
57
60
61
62
63
66
67
72
80
83
Total
county
farm
population
1 ,296
691
540
2,172
1,610
1,310
1,387
1,143
214
208
1,617
189
836
487
1,761
1,277
457
2,114
1,328
2,950
2,299
1,104
4,806
3,307
1,233
2,026
No. of
farmers
within five
miles of the
facility
178
102
81
137
351
274
169
125
28
14
277
16
114
68
332
166
55
292
232
350
312
232
333
32
190
169
Total no. of
farms In the
county
707
354
391
1,872
625
641
922
1,037
258
504
809
195
379
617
621
757
222
821
849
,346
,203
391
,669
,995
,157
1,521
No. of
subsistence
farms in the
county
333
206
226
831
328
382
523
698
157
442
384
72
245
495
210
489
134
123
552
695
659
176
260
615
927
1,126
No. of
subsistence
farmers within
five miles of the
facility
84
59
47
61
184
163
96
84
17
12
132
6
74
55
112
107
33
44
151
ISO
171
104
52
10
26
125
Total
population
within five
miles of the
facility
10,705
1,922
15,781
191,915
59,376
20,812
12,518
20,233
965
3,240
65,458
1,975
15,559
24,553
6,275
72,527
8,572
29,085
10,583
9,433
37,469
43,851
17,407
584
55,918
10,136
Total
county
population
17.035
13,966
292,594
1,185,394
96,246
87,777
38,816
34,119
5,528
480,577
121,393
118,934
30,605
51,832
20,488
633,232
44,739
46,733
42,836
150,208
61,633
247,105
106,913
543,477
335,749
85,167
Total
county
urban
population
9,488
5,150
250,159
1,118,354
44,157
65,102
22,929
15,820
0
459,439
68,172
112,667
11,354
37,223
2,589
611,229
7,479
37,181
18,486
86,689
43,694
173,033
65,957
455,300
261,024
40,007
Urban
population
within five
miles of the
facility
5,962
709
13,492
181,061
27,241
15,436
7,395
9,381
0
3,097
36,760
1,871
5,772
17,633
793
70,007
1,433
23,140
4,567
5,444
26,563
30,706
10,739
489
43,473
4,761
Non-urban
population
within five
miles of the
facility
4,743
1,213
2,289
10,854
32,135
5,376
5,123
10,852
965
143
28,698
104
9,787
6,920
5,482
2,520
7,139
5,945
6,016
3,989
10,906
13,145
6,668
95
12,445
5,375
No. of
"backyard
gardeners"
within five
miles of the
facility
4,565
1,111
2,208
10,717
31,784
5,102
4,955
10,727
937
129
28,421
88
9,672
6,852
5,150
2,354
7,084
5,653
5,784
3,639
10,594
12,913
6,336
63
12,413
5,205
"Homegrown
vegetable"
population
within five
miles of the
facility
3,460
699
3,996
44,732
20,463
5,803
3,929
6,932
433
741
21,022
461
5,663
6,976
2,712
16,520
3,525
7,883
3,689
3,004
10,752
12,694
5,494
158
15,315
3,434
                     2-20

-------
                    Exhibit 2-6
Population Risks via the Vegetable Ingestion Pathway
Plant
I.D.
4
7
15
18
22
25
29
30
33
42
44
46
49
53
54
55
57
60
61
62
63
66
67
72
80
83
Individual
cancer risk;
"subsistence"
1.30E-05
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
9.90E-05
9.90E-06
9.90E-05
O.OOE+00
9.90E-06
O.OOE+00
O.OOE+00
O.OOE+00
9.90E-05
9.90E-05
O.OOE+00
9.90E-06
9.90E-06
9.90E-05
9.90E-06
9.90E-05
9.90E-06
9.90E-04
O.OOE+00
O.OOE+00
Individual
cancer risk;
"homegrown"
1 .70E-06
9.90E-05
1 .OOE-06
1 .OOE-07
1 .OOE-06
9.90E-06
9.90E-06
1 .OOE-06
9.90E-05
9.90E-06
1 .OOE-06
9.90E-06
9.90E-06
1 .OOE-06
9.90E-06
9.90E-06
9.90E-06
9.90E-06
1. OOE-07
9.90E-06
9.90E-06
9.90E-06
1. OOE-06
9.90E-05
1. OOE-06
1 .OOE-06
Individual
non-cancer
Hazard Index;
"subsistence"

-------
                                                                       Exhibit 2-7
                                              Refined Population Risks via the Vegetable Ingestion Pathway

Plant
I.D.
4
7
15
18
22
25
29
30
33
42
44
46
49
53
54
- 55
57
60
61
62
63
66
67
72
80


Without Refinement Based on Topographic Information"
Distance of
Interest (miles)
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5


No. of
Subsistence
Farmers
84
59
47
61
184
163
96
84
17
12
132
6
74
55
112
107
33
44
151
180
171
104
52
10
26


"Homegrown"
Population
3,460
699
3,9%
44,732
20,463
5,803
3,929
6.932
433
741
21,022
461
5,663
6,976
2,712
16,520
3,525
7,883
3,689
3,004
10,752
12,694
5,494
158
15,315
3,434
Tolal
No. of cancer
cases;
"Subsistence"
1.09E-3
O.OOE+0
O.OOE+0
O.OOE+0
O.OOE+0
O.OOE+0
9.48E-3
8.31E-4
1.66E-3
O.OOE+0
I.30E-3
O.OOE+0
O.OOE+0
O.OOE+0
I.1IE-2
1.06E-2
O.OOE+0
4.33E-4
I.50E-3
1.79E-2
1 69E-3
1 .03E-2
5.13E-4
974E-3
OOOE+0
O.OOE+0

No. of cancer
cases;
"Homegrown"
5.88E-3
6.92E-2
4.00E-3
4.47E-3
2.05E-2
5.74E-2
3.89E-2
6.93E-3
4.28E-2
7.34E-3
2.10E-2
4.57E-3
5.61E-2
6.98E-3
2.68E-2
1.64E-I
3.49E-2
7.80E-2
3.69E-4
2.97E-2
1.06E-I
I.26E-I
5.49E-3
I.56E-2
I.53E-2
3.43E-3

No. of people
exposed to
Hazard Index
greater than or
equal to 1 ;
"subsistence"
0
0
0
0
0
0
96
84
0
0
0
0
0
0
0
107
0
44
0
180
0
104
0
10
0
0

No. of people
exposed to
Hazard Index
greater than'or
equal to 1 ;
"homegrown"
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
7,883
0
0
0
0
0
0
0
0

With Refinement Based on Topographic Information
Distance of
Interest (miles)
1.0142
1.0142
1.0142
1.0142
1.0142
1.0142
0.0282
0.4508
1.0142
1.0142
1.0142
1.0142
1.0142
1.0142
1.0142
0.0282
1.0142
0.1127
1.0142
1.0142
1.0142
0.1127
1.0142
0.4508
1.0142
1.0142
No. of
Subsistence
Farmers

3
3
3
8
7
3
3
3
3
5
3
3
3
5
3
3
3
6
7
7
3
3
3
6
5

"Homegrown"
Population

28
163
1,840
842
239
3
55
16
29
865
18
233
286
112
3
144
4
152
124
442
5
225
3
622
141

No. of cancer
cases;
"Subsistence"

OOOE+00
O.OOE+00
O.OOE+00
O.OOH+00
O.OOE+00
2.97E-04
2.97E-05
2.97E-04
O.OOE+00
5 36E-05
OOOE+00
O.OOE+00
O.OOE+00
4.58E-04
2.97E-04
O.OOE+00
2.97E-05
6.I5E-05
735E-&4
6.96E-05
2.97K-04
2.97E-05
2.97E-03
O.OOF.+OO
O.OOE+00

No. of cancer
cases;
"Homegrown"

2.79E-03
1 63E-04
1.84E-04
8.42E-04
2.36E-03
297E-05
5. 5 IE-OS
I.57E-03
2.9IF.-04
8.65E-04
I.8IE-04
2.3 11- -03
2.86E-04
1.10E-03
2.97E-05
1 42E-03
390E-05
1 52E-05
1 22E-03
4.38F.-03
5.04E-05
2.25E-04
2.97E-04
622F.-04
1 4 1 li-04

No. of people
exposed to
Hazard Index
greater than or
equal to 1;
"Subsistence"

0
0
0
0
0
3
3
0
0
1 o
0
0
0
(1
3
0
3
0
7
0
3
0
3
0
0

No. of people
exposed to
Hazard Index
greater than or
equal to 1 ;
"Homegrown"

0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
4
0
0
0
0
0
0
0
0

"This information is summarized from Exhibil 2-6
                                                                          2-22

-------
                          Exhibit 2-8
         Population Risks via the Fish Ingestion Pathway
(including underlying data used for estimating the population risks)
Facility
number
35
37
62
81
Stream length within
the five-mile radius
(miles)
(a) small streams: 2
(b) large streams: 7.5
15
(a) small streams: 24.5
(b) large streams: 9
(a) small streams: 7
(b) large streams: 7
Average stream width
(miles)
(a) small streams: 1/350
(b) large streams: 1/35
1/350
(a) small streams: 1/350
(b) large streams: 1/70
(a) small streams: 1/350
(b) large streams: 1/70
Stream acres
within five miles
of the facility
(acres)
140.8
27
127.1
76.8
Standing
stock
(Ibs/acre/
year)
82.2
593.2
89.5
219
Exploitation
rate
0.2
0.2
0.2
0.2
Pounds of fish
caught per year
within the area ol
influence
(Ibs/yr)
2,314.75
3,250.74
2,275.09
3,363.84
Percent of fish
tissue that is
edible
0.35
0.35
0.35
0.35
Pounds of fish
ingested per year
per recreational
fisher
5.86
5.86
5.86
5.86

Number of
recreational fishers
that can be supported
by the harvest
138
194
136
201
Total
No. of recreational
fishers exposed to
Hazard Index greater
than or equal to 1
138
194
136
201
669
                            2-23

-------
                                              2-24

facility, a total of three people per household was assumed to be exposed at subsistence levels of
vegetable and beef and milk ingestion.)

       Recreational Fisher Population

       The results of this analysis (as well as the supporting data) are presented in Exhibit 2.8. For a
more detailed description of the raw data used in this analysis, refer to Appendix D of this document.
In summary, for potential population noncancer effects, across all the people who fish and consume
recreationally caught fish within five miles of the 82 facilities, about 670 individuals  are exposed via
fish ingestion to contamination exceeding noncancer effects thresholds (i.e., hazard index  greater
than 1).  The predicted population cancer risk for this pathway is extremely small, ar less than one
excess cancer case.

       2.4.3   Results Extrapolated to Full Universe

       As  noted before, the focus of the Tiers 1 and 2 analyses was on assessing risks at 82 of the
total 108 cement facilities.  (These 82 facilities can  be denoted as the "known universe.")  The
remaining 26 facilities were excluded because a lack of relevant data (e.g., data on constituents in
CKD  wastes or on types of waste management practices) prevented them from being assessed directly
in the original individual risk analyses done  for RTC and NODA.  (These 26 facilities can be denoted
as the "unknown universe.")  This does not mean, however, that there are no potential risks due to the
CKD  being generated and managed at these  26 facilities. To derive a composite picture of potential
population  risks across the  full universe of cement facilities, therefore, EPA estimated the potential
population  risks within the  unknown universe by extrapolation from results within the known universe.

       Given the lack of knowledge regarding the unknown universe, EPA used a two-step process to
extrapolate the  results from the known  to the unknown universe, thereby estimating the potential
population  risks for the full universe.

       Step 1:  To begin,  EPA defined the  "bounds" of the results for the full universe of cement
facilities.  To define a conservative upper  bound measure of the population risks, EPA assumed that
every single facility in the  unknown universe is as "risky" as the highest-risk facility in the known
universe. The corollary is  that to define a lower bound measure of the population risks, it is
reasonable  to assume that every single facility in the unknown universe  is as "risky"  as the lowest-risk
facility in the known universe. The Agency believes that, working from the results of the known
universe, the true results for the full universe are unlikely to be beyond  the range defined by upper and
lower bounds as defined above (the results cannot be lower than the lower bound).  The relevant
calculations are  shown below, using as an example  the potential noncancer population effects due to
fish ingestion.  The same approach was used for cancer risk.
                                 Draft, August 1997, Do Not Cite or Quote

-------
 Upper bound
 measure of population
 risks6 across the
 full universe
                                               2-25
Total population
risks  for the known
universe
Risk from
highest-risk
facility
No. of
facilities in the
unknown
universe
                              669
                            201
                        26
Thus, the upper bound measure of population risks (i.e., the potential noncancer population effects due
to fish ingestion) across the full universe is 5,895.
 Lower bound
 measure of population
 risks  across the full
 universe
Total population
risks  for the known
universe
Risk from
lowest-risk
facility
No. of
facilities in the
unknown
universe
                              669
                            0
                         26
Likewise, the lower bound measure of population risks (i.e., the potential noncancer population effects
due to fish ingestion) across the full universe is 669.

        Step 2:  Having defined the bounds, the next step would be to determine where the true results
for the full universe of cement facilities are likely to fall within the range defined by the upper and
lower bounds.  Given the lack of critical data, EPA believes that a reasonable assumption is that the
distribution  of risks among  facilities within the (smaller) unknown universe is similar to the
distribution  of risks among  facilities within the (larger) known universe.  In such a case, the results
from the known universe can be directly extrapolated to the unknown universe, preferably using one or
more  "weighting factors" that are common to both universes and are expected to be related to the
potential risks.   The only such common factor for which data are available is the "quantity of CKD
waste generated."  Thus, EPA extrapolated the results from the known universe, weighted by the
    6 As explained on page 2-9, EPA uses in this document the term "population risk" as a loose, collective term
to refer to both population cancer risk and population noncancer effects.  Population cancer risk is used to denote
"excess cancer incidence," i.e., the number of excess cancer cases in the exposed population, and population
noncancer effects is used to denote the number of persons exposed to levels above the thresholds for noncancer
effects.
                              *** Draft, August 1997, Do Not Cite or Quote ***

-------
                                                2-26

amount of CKD wasted,  as follows (example shown for potential noncancer population effects due to
fish ingestion):
      Total
    population
   risks across
     the full
     universe
   Total
 population
risks for the
  known
  universe
Quantity of CKD
  wasted in the
 known universe
Quantity of CKD
  wasted in the
unknown universe
   Quantity of CKD wasted in the known
                  universe
      998.8
    669
 2,497,911 tons
  1,231,422 tons
                                                                  2,497,911 tons
The upper bound, lower bound, and most reasonable estimates of the total population risks, for all
relevant pathways, across all  108 facilities in the CKD universe are summarized in Exhibit 2-9.8

       Note that although EPA  chose to use the refirted results for the "homegrown" and subsistence
farmer populations (as discussed in section 2.4.2, EPA derived more refined results for the known
universe for these populations based on site-specific topographic data) for extrapolating results to the
full universe, there is significant uncertainty in this extrapolation.  This is because topography data are
site-specific and  assumptions about topography can not easily be made from some facilities in the
known universe to others in the unknown universe.

        In the individual risk estimates derived previously, the driving constituents for cancer risks and
noncancer effects were identified for each exposure pathway. These constituents were identified as
"producing" the highest  estimated risk.  For this population risk analysis, arsenic, which can cause
both systemic and carcinogenic effects, was the driving constituent for cancer risks via the food chain
pathway across all cement plants analyzed.  Likewise, for noncancer effects, the driving constituent in
the food chain pathway  was one of the following across all cement plants analyzed: thallium,
chromium, cadmium, beryllium,  or barium.
    7 Extrapolation using waste quantity as a weighting factor is less straightforward for potential noncancer
population effects than for population cancer risks.  This is because an increase in waste quantity could lead to
either an increase in the magnitude of HI exceedances at only those  facilities that already have potential
noncancer population effects, or, in addition, an increase in the magnitude of HI exceedances at facilities that
previously did not have potential noncancer population effects.  (In effect, this means that one cannot do a linear
extrapolation on a nonlinear dose-response model.)  Total potential population  noncancer effects  (i.e., number of
people exposed above an HI of 1.0) would remain the same in the former case, and increase in the latter.

    8 Note that- a similar extrapolation based on the ratio of the number of facilities in the full universe to that in
the known universe (i.e., not weighted by waste quantity) would yield relatively similar results (scale-up of 1.3
(i.e., 108/82) versus 1.5).
                                  Draft, August 1997, Do Not Cite or Quote ***

-------
                                              2-27

                                           Exhibit 2-9
             Potential Population Noncancer Effects and Population Cancer Risks,
                       Extrapolated to the Full Universe of Cement Plants
1 Lower Bound
Most Reasonable
Upper Bound
Potential Population Noncancer Effects
(i.e., the number of persons exposed to levels above the thresholds for noncancer effects)
"Homegrown"
Subsistence Farmer
Recreational Fisher
4
25
669
6
37
999
108
207
5,895
Population Cancer Risks
(i.e., the number of excess cancer cases in the exposed population)
"Homegrown"
Subsistence Farmer
0.02
0.006
0.03
0.009
0.13
0.08
       2.5     MAJOR LIMITATIONS AND UNCERTAINTIES

       This study has significantly enhanced EPA's understanding of the extent to which populations
living near cement plants are potentially at risk due to indirect pathways. The Agency recognizes,
however, several limitations and uncertainties inherent in the analysis.  Limitations and uncertainties
associated with the indirect exposures analysis include those that apply across the entire analysis, and
those that apply specifically to either the vegetable ingestion or fish pathways; these are discussed in
Section 2.5.1.  In Section 2.5.2, EPA discusses the feasibility of using an alternative approach to
calculate population risks and its associated limitations.

       2.5.1  Limitations and Uncertainties for the Indirect Exposures Analysis

•      Estimates of the individual cancer risk and noncancer hazard indices used for  both the
       vegetable and fish ingestion pathways to develop population risk estimates were
       derived originally based on methodologies explained in the RTC and NODA technical
       background documents.  There are two major limitations associated with using the
       original individual risk estimates as a starting point for the current analysis.

               First, the original individual risk estimates were not designed specifically to feed into a
               population risk analysis.  That is, they were not designed to reflect the distribution of
               risks in specific exposed populations. For example, the individual risk estimates for
               the fish ingestion pathway do not reflect the spatial variability in the exposure
               concentrations, taking into account that  streams farther from the site will likely  have
                             ***  Draft, August 1997, Do Not Cite or Quote ***

-------
                                       2-28

        lower concentrations than the single closest stream for which the individual risks were
        derived.  Because it was not possible to account for such spatial variability in the
        current analysis, the population risks will tend to be overestimated.

-       Second, all the limitations and uncertainties associated with the original estimates of
        the individual risks carry over into the current analysis as well.  For example, for some
        facilities, the original methodology used a scaling approach to derive estimates of
        individual risks that were presented as being within "order of magnitude" ranges. For
        the current analysis  EPA used the upper end of such ranges to represent individual
        risks.  This would most likely lead to an overestimate of the population risks.
        Furthermore, the individual risk estimates do not include effects of exposure to
        dioxins/furans that may potentially be present in the CKD waste.  This would lead to
        an underestimate of the population risks. Also, it should be noted that there is a
        degree of uncertainty associated with using individual hazard index estimates denoted
        as "best estimates" to calculate population noncancer effects.  The best estimate hazard
        index for a given constituent is based on the best estimate exposure concentration of
        that constituent, which  takes the average of all measured exposure concentrations for
        that constituent.  If the average concentration is less than the RfD, then the hazard
        index will be less than  one.  This hazard index is then applied to all the measured
        concentrations  that were used in calculating the average concentration, even though
        some  of the measured concentrations may actually be greater than the RfD (i.e.,
        resulting in HI >  1). Therefore, in using the best estimate  individual hazard indices,
        the population  with  a hazard index greater than one can be underestimated.

Vegetable Ingestion Pathway

The farm population, the number of subsistence farmers, and the urban population within five
miles of the facility were calculated using the total county farm population, the number of
subsistence farms in the county, and the county urban population, respectively,  and the ratio of
the area within five miles of the facility to the total area of the county.  These calculations
implicitly assume that the three populations are distributed uniformly throughout the county.
This would not be the  case  in reality because these populations would most likely be
concentrated in different areas  of the county.  The populations calculated for the five-mile
radius surrounding the facilities may be overestimated or underestimated depending on whether
the facility is  located in an urban or in a rural  area.  For example,  if the facility is located in
an urban area, the calculated urban population within five miles of the facility will be
underestimated and the calculated farm population and number of subsistence farmers within
five miles of the city will be overestimated.  If the facility is located in  a rural  area, then the
calculated urban population within five miles of the  facility will be overestimated and the
calculated farm population and number of subsistence farmers within five miles of the facility
will be underestimated.

The number of subsistence farmers living within the county was estimated using a proxy for
subsistence farms, which in turn were estimated based on data on the sale of agricultural
products. This method assumes that farms with sales  of less than $10,000 per year are
subsistence farms.  Since $10,000 per year is below the poverty level, these people most likely
                          Draft, August 1997, Do Not Cite or Quote ***

-------
                                       2-29

sell agricultural products in order to supplement their income or grow agricultural products in
order to supplement their diet, thus consuming a greater portion of homegrown vegetables and
beef and milk than the general population.  This  methodology would tend to overestimate the
number of subsistence farms in the county because people who farm part-time or who sell
livestock could be counted as subsistence farmers given the assumptions used in this analysis.
The extent of such "false positives"  may be minor, however, because the data used were
collected by the Census of Agriculture in such a manner that they target primarily farmers
(i.e., the Census solicits information from farmers identified  based  on information from
previous censuses, USDA surveys, and IRS information) and exclude to the extent possible
those people who do not identify farming as their principal occupation.

The "homegrown vegetable" population was calculated by determining the percentage  of non-
urban,  non-farm population and also the percentage of non-rural or urban population that  could
be expected to participate in backyard gardening.  1995 data from  the National Gardening
Association  indicate that 45%  of rural and 22% of non-rural U.S. households participate in
backyard gardening.  Thus, 45% of the calculated non-urban, non-farm population  and 22% of
the calculated urban population within five miles of the facility were considered to be the
population of backyard gardeners that actually participates in home gardening. This assumes
that 45% of households participate in  backyard vegetable gardening in all rural areas
throughout the country, while  in reality this percentage most likely varies from region to
region.  Likewise, the 22% of urban households  that participate in  backyard  vegetable
gardening may not be applicable to  all urban areas.

To derive a composite picture of the potential population risks across the full universe of
cement facilities,  EPA estimated the potential population risks within the unknown universe by
extrapolation from results within the known universe.  As discussed in section 2.4.2, EPA
derived more refined results for the known universe for the vegetable ingestion pathway based
on site-specific topographic data for seven facilities. EPA chose not to use these refined
results for extrapolating results for the full universe, however, because topography data are
site-specific and assumptions about  topography can not reasonably be made  from some
facilities in the known universe to others in the unknown universe and, therefore, it would not
be reasonable to believe that the refined population risk results can be extrapolated to the
unknown universe.  Because EPA used the more conservative (i.e., not refined) population risk
estimates for the extrapolation to the unknown universe, the overall results for the full universe
of cement facilities for the vegetable consumption pathway are likely to be overestimated.

Fish Ingestion Pathway

The  approach used for calculating the populations potentially exposed via fish
ingestion  assumes that atmospheric  deposition and surface runoff from the CKD waste
piles can contribute contaminants to all the streams within a five-mile radius of the
facility (i.e., in addition to the single stream closest to  the facility).  This is  likely  to
overestimate the population risks because site-specific  hydrologic factors may dictate
that fewer streams, i.e., only those close  to the facility, may receive significant levels
of the  contamination, at least through surface runoff.
                      ***  Draft, August 1997, Do Not Cite or Quote **

-------
                                       2-30

The approach used in this analysis for calculating the populations potentially exposed via fish
ingestion involved determining the number of fishers that can be "supported" by the available
fish biomass (i.e.,  standing stock) within the  area of interest.  In most cases, standing stock
data were not available for the streams located closest to the facilities.  Standing stock data
were, therefore, extrapolated to the nearby streams from other streams located within the
counties of interest.  In some cases, standing stock data were available for a large stream and
were extrapolated  to small streams. In other cases, data from a small stream were applied to
larger streams located  within five miles of the facilities.  Because many  chemical  and
biological factors can affect the fish population of a stream,  it is likely that extrapolating
standing stock data from one stream to another either overestimates or underestimates the
actual standing stock of the streams located within five miles of the facilities.-

Standing stock data were often available for  only one sampling site along a stream.  In
calculating the standing stock, it was  necessary therefore to assume that data from one
sampling site  are representative of other locations along the  stream.  In cases where data were
provided for more than one sampling site, an average value of the standing stock  was used in
calculating the standing stock within five miles of the facility.

For one facility, the surrounding land is  primarily swamp that is known  to support fish
populations.  Including in this analysis the entire swamp area within five miles  of the facility
would have lead to an extreme overestimate  of the number of fishers that can be supported by
the available standing  stock. It was necessary to refine the estimate by including  only those
areas of swamp that are known to support fish populations year-round.   Thus, the calculation
for this  facility includes only the stream miles that are flooded at all times of the  year.   This
refinement will tend to underestimate the overall available standing stock,  and, consequently,
the number of fishers that can be supported because it does not include areas of the swamp
that support fish populations at only certain times of the year.

The standing stock data provided by the state and local agencies were sometimes  given  as a
single value, representing all species present at the sampling site. In other cases,  the data were
provided separately for each species present  at the sampling  site.  Thus, for some facilities it
was not possible to eliminate data for species that would not commonly be consumed by a
recreational fisher. To be consistent across all facilities, all  species were included in the
calculation of the  standing stock, including those that are not typically consumed  by
recreational fishers.  Since all species  are included in the  estimate of the available standing
stock, the number of recreational fishers that can be supported by this standing  stock is  likely
to be overestimated. However, the degree of the overestimate may be negligible  since the
species  of fish that are not commonly consumed tend to weigh  very little and, thus, would not
contribute significantly to the total standing stock value.

Some facilities have large lakes or reservoirs located within  a five-mile  radius of  the facility;
these waterbodies  were not included in the analysis  even  though they may support fish
populations.  They were not included because it is expected  that since these lakes are very
large, contaminants would likely be diluted significantly in the water column.  (This analysis  «
does not account for any accumulation of metals in  the sediments within the lakes/reservoirs
and potential subsequent transfer into  the foodchain).  In  addition, it is less appropriate  to
                          Draft, August 1997, Do Not Cite or Quote

-------
                                              2-31

       extrapolate lake standing stock data from one area to another since lake fish populations would
       tend to be concentrated in certain areas of the lake and at certain depths.

       2.5.2  Alternative Approach to Calculate Population Risks and Associated Limitations

       In Section 2.3.2 of this Technical Background Document, EPA provides details on the specific
calculations used to estimate population risks associated with the ingestion of contaminated vegetables.
In brief, the approach used in this analysis for calculating population risks assumes that a certain
proportion of the people living  within five miles of a cement facility is likely to be exposed via
ingestion of contaminated vegetables. This proportion is estimated, for each cement facility analyzed,
based on county-level data on the number of farmers and backyard gardeners, prorated to account for
the relationship between the area around the facility to the total area in the county. Thus,  this
approach is based on the assumption that the potentially exposed population  is located entirely  within
the five-mile radius surrounding the facility, and does not account for persons who live outside of this
boundary and  may be exposed to vegetables "exported"  from the vicinity  of  the facility.

       As part of the current analysis, EPA also reviewed potentially applicable alternative approaches
to calculating population risks.  In  the Addendum to the Methodology for  Assessing Health Risks
Associated with Indirect Exposure to Combustor Emissions (EPA/600/AP-93/003, November 1993
Review Draft) EPA has outlined an approach to calculating population risks  due to ingestion of
contaminated food in general that is based on an estimate of total food production within the area
affected by contamination. The suggested equation for calculating population risk in this approach is
as follows:

                                      q*  x ED x Cx  x FPX
               population risk    =       	
                                            BW xLT
where:
       q       =      cancer slope factor (kg-day/mg)
       ED     =      exposure duration (yr)
       Cx     =      concentration of contaminant in the food from area X (mg/g)
       FPX    =      production of the food in area X (g/d)
       BW    =      body weight (kg)
       LT     =      lifetime (yr)

As can be noticed, the key difference between the calculation used in this November 1993 Addendum
approach and that used in the approach for the current analysis is the use of the "food production"
term. The use of this term leads to the assumption  that the number of people potentially exposed to
contaminated food (vegetables, in the case of the current analysis) depends on the amount of food that
is produced  within the area of interest (i.e., all food that is produced within the area of interest is
consumed by someone, either within this same area, or elsewhere).  In contrast to the calculation used
in the current approach, the potentially exposed population calculated according to the November 1993
Addendum approach could include  people living outside the five-mile radius of the facility.
                                 Draft, August 1997, Do Not Cite or Quote ***

-------
                                              2-32

       EPA considered the pros and cons of the two approaches, and decided against using the
November 1993 Addendum approach, primarily because it is relatively less flexible in terms of its data
needs, would not be applicable to all receptors  of interest, and would not yield results that are any
more accurate and/or certain.  Key points against the use of the November 1993 Addendum approach
include the following:

•      This approach is relevant for only linear, non-threshold effects; thus, it can be used to
       estimate population cancer risks, but not population noncancer effects.

•      This approach will not allow one to differentiate between risks to farmers/home
       gardeners and risks to subsistence farmers.

•      This approach would require the development and use of several "adjustment factors,"
       many of which will introduce greater uncertainty in the final results:

       -      the approach may lead to an overestimate of risks because it assumes that all food
               produced is eaten; one would need to derive and apply factors to account for the
               portion of food that is not eaten due to  "wastage" or loss during preparation and
               processing/handling.

       -      the approach would require  facility-specific food production data, calculated possibly
               using county-level data.  The most readily-available source for these data is the Census
               of Agriculture. Production data available from the  Census are given as "the value of
               agricultural products sold per year" or the "number of acres harvested per year" for
               specific agricultural products.  If the value of agricultural products sold per year were
               used for the calculation,  it would be necessary to use a factor to convert the value per
               year to grams per day with the unit price of each individual vegetable.  If the number
               of acres harvested per year were used in the calculation,  it would be necessary to use a
               conversion factor for the amount of a certain vegetable that  can be grown on an acre
               of land.  Such conversion factors are likely  to be highly variable and not easily
               derived.

•      This approach requires use of individual risk estimates that are slightly different from
       those available for use from the previous CKD-related analyses, because it requires
       that concentrations used are  an average for that area of interest.  So far, data are
       available from previous analyses that provide concentrations of contaminants in
       vegetables from one particular farm or region; these concentrations may often represent
       a maximum for that region, but it is unlikely that they represent an average
       concentration across the entire area  of interest.
                             *** Draft, August 1997, Do Not Cite or Quote ***

-------
                                       CHAPTER 3
             POPULATION EFFECTS DUE TO PM  EXPOSURES
       This chapter describes the Agency's estimates of population effects due to exposures to
airborne paniculate matter (PM) from cement kiln dust waste management units.  In Section 3.1 EPA
provides the background and starting point of this analysis, discussing what specifically is not included
in the scope of the analysis and any resultant implications. Section 3.2 presents a summary of the
approach used to estimate the population effects.  Section 3.3 includes a discussion of the results.
Finally, Section 3.4 presents a discussion of the major limitations and uncertainties associated with the
PM exposures analysis.
       3.1    BACKGROUND AND SCOPE

       The objective of this analysis is to estimate the risks to populations exposed to airborne
paniculate matter released from CKD waste management units at cement facilities.  These risks, more
appropriately termed effects, are characterized in terms of the number of people in the populations
surrounding the cement facilities that are exposed to PM concentrations  above certain thresholds. (As
explained below, the impacts due to exposure to PM cannot be calculated in conventional terms  of
risk, such as the  probability of an individual expressing certain health effects or the number of cases of
certain illnesses occurring within the exposed population.)

       3.1.1   Starting Point of this Analysis

       The Agency first analyzed PM10 (particulates with an aerodynamic diameter less than or equal
to a nominal 10 micrometers) concentrations at  cement plants as part of the Notice of Data Availability
on Cement Kiln Dust (NODA). This analysis expanded the original risk modeling conducted for the
Report to Congress on Cement Kiln Dust (RTC) by determining the concentration of airborne CKD
particulates at a given exposure location (i.e., closest agricultural field, nearest residence, and CKD
pile boundary, as well as residences located at five concentric  rings surrounding the facility extending
to 10 kilometers) for each of five case-study facilities. The methodology and results of this analysis
are  presented in the Technical Background Document for the NODA. Releases from CKD piles were
simulated using the landfill  simulation component of MMSOILS. Each  of the CKD piles at the five
facilities was  simulated as an unlined and uncovered landfill unit.  MMSOILS employs one empirical
model to estimate the annual average release rate of PM10 due to wind  erosion.  Due to the nature  of
this screening analysis,  the Agency did not use a complete set of meteorological data or stability array
in the atmospheric dispersion modeling. Rather, the Agency used a single set of meteorologic
conditions to represent a conservative estimate of annual average ambient concentrations in any
direction surrounding the site. The results of this initial PM10 analysis consisted of annual average
PM10 concentrations, for both the best estimate and upper bound modeling scenarios, for the five case-
study facilities originally modeled by the Agency.

       The Agency later expanded its  PM10 analysis of the five original case-study facilities to the
entire sample- of 82 cement plants  addressed in the NODA.  The methodology and results of this
expanded analysis are presented in the  Technical Background Document on Potential Risks of Cement
                                Draft, August 1997, Do Not Cite or Quote ***

-------
                                              3-2

Kiln Dust in Support of the Cement Kiln Dust Regulatory Determination.  Having determined that pile
size, wind speed, and exposure distances were the parameters that influenced PM10 concentrations the
most using the MMSOILS model, the Agency performed a number of MMSOILS runs to estimate
PM10 concentrations for eight different pile sizes and nine different distances, which were based on the
pile  sizes and exposure distances reported for the sample of cement plants examined.  The Agency
also performed four additional MMSOILS  runs changing only the wind speed to reflect the minimum
reported wind speed, the average reported wind speed, the maximum reported wind speed, and an
additional wind speed between the average and maximum reported wind speeds. The Agency then
developed adjustment factors to account for the differences in the PM,0 concentrations using the
average wind speed and the  three other wind speeds. To address the large variability in pile size, wind
speed, and exposure distance at the  sample facilities, the parameters at the individual sample facilities
were matched with the closest modeled parameters to eliminate having to model each individual
facility. The resulting PM10 concentrations presented in this analysis provide a best estimate of the
PM,0 concentrations at each individual facility.

       3.1.2  What is Included and Excluded from the Scope of this Analysis

       In the expanded analysis that was conducted for the Cement Kiln Dust Regulatory
Determination, MMSOILS (an EPA multimedia fate and transport model) and a simplistic modeling
approach were used to  create a matrix of PM concentrations for various CKD pile sizes and receptor
distances. PM10 concentrations at each of a set of 52 facilities were then estimated - for exposure
points defined by the facility boundary and the closest residence - by selecting the pile size-receptor
distance combination from the table that best represented the facility's actual conditions, and scaling
the table value up or down based on differences between the wind speed at the facility and the wind
speed assumed in the modeling.  For the current modeling exercise, EPA used a more  sophisticated
model and approach  to estimate PM10 as well as PM2 5 concentrations under a broader range of
conditions.  In particular, the current exercise assessed releases of PM from other sources at the
facility in addition to the CKD waste pile.  A detailed comparison of previous and current modeling is
provided in Appendix G.  In brief, EPA used these new results to refine the PM estimates for the 52
facilities considered previously and  to determine the number of nearby residents who are potentially
exposed to ambient PM concentrations of significance.  Furthermore, the modeling approach used in
this analysis estimates the ambient PM10 as well as PM25 concentrations due to only releases from the
CKD waste handling and management. Thus, the approach does not account for background
concentrations of PM10 and  PM25 that may be due to other sources.

       As with the indirect  exposures analysis, a total  of 26 facilities have been excluded from this
PM analysis because they could not be assessed directly given a  lack of data. To  account for these
facilities, however, results from the facilities that were analyzed are extrapolated to the 26 facilities to
derive a composite picture of potential population effects at the full universe of cement facilities (i.e.,
108  facilities). In summary, from a grand total of 108  facilities,  82 can be  viewed as having been
analyzed.  Of these 82 facilities, 30 were deemed to have negligible risks/adverse effects based on
initial screening  and the remaining 52 were analyzed in greater detail to characterize risks/adverse
effects due to PM exposures. Finally, results from these 82 (i.e., 52 plus 30) facilities are extrapolated
to the 26 facilities that were not analyzed due to lack of data.
                                 Draft, August 1997, Do Not Cite or Quote ***

-------
                                               3-3

       3.2     SUMMARY OF OVERALL APPROACH

       3.2.1   Identify An Appropriate "Risk Descriptor" for PM Exposures

       Analyses conducted  for the RTC and NODA have described to a large extent the nature of
adverse effects from PM exposure.  For this analysis EPA determined that the most appropriate risk
descriptor would be one that described the extent of adverse effects in terms of the total number of
people exposed to a specific level of PM. A review of EPA's 1996 staff paper on the airborne
paniculate matter standard points  to two conclusions that are key for this analysis:

       (i)      While coarse and fine particles can increase respiratory symptoms and impair
               breathing,  the staff paper concludes that fine particles are more likely to contribute to
               the health  effects described in a number of recently published studies on paniculate
               matter exposures.

       (ii)     The staff paper recommends that, while  retaining the coarse particles (i.e., PM,0)
               standard, more effective and efficient protection could be provided by establishing a
               separate standard for fine particles (i.e.,  PM2 5).
Thus, for this analysis, the Agency characterized
population effects in terms of exposures to both
PM]0 and PM25. Because there are no widely
accepted dose-response measures for PM
exposures, EPA did not describe the population
effects in conventional terms of number of
excess disease cases.  Instead,  the Agency used
the National Ambient Air Quality Standards
(NAAQS) for PM as thresholds against which
the facility-specific  PM concentrations could be
compared at any given receptor point.  The
Agency used both the annual average and the
peak 24-hour average PM concentrations as the
basis for risk estimation. The  NAAQS used in
this analysis are shown in the text box.

Standar
Annual
average
NAAQS
24-hour
NAAQS
ds Used in the Analysis
PM10
50 ug/m3
150 ug/m3
PM2J
15 ug/m3 *
65 ug/m3 *
*
These represent PM standards announced by EPA on
July 17, 1997.

       3.2.2   Develop the Overall Modeling Framework

       For efficiency, EPA's overall modeling approach consisted of (i) selecting the "highest-risk"
facilities, (ii) modeling the emissions and dispersion at these facilities, (iii) predicting PM
concentrations at exposure points for which population data can be overlaid to predict population
effects at the highest-risk facilities, and (iv) using these results to draw broader conclusions for
    1 Review of the National Ambient Air Quality Standards for Paniculate Matter:  Policy Assessment of
Scientific and Technical Information.  Office of Air Quality Planning and Standards Draft Staff Paper, April
1996.
                                 Draft, August 1997, Do Not Cite or Quote ***

-------
                                                3-4

facilities that were determined to be of "lower-risk."  The Agency used the term "highest-risk facility"
here to mean a facility that, among a group of facilities that would experience relatively similar
dispersion patterns, is the one with the highest emissions from all relevant sources and therefore would
result in  highest ambient concentrations of PM at receptor points.

       To select such highest-risk facilities, EPA first created groups of facilities that are expected to
experience relatively similar atmospheric dispersion patterns, and then identified within each group the
single facility with the highest emissions potential.

       Both emission and dispersion of dust particles are  heavily dependent on the climate and
meteorological conditions at a given facility.  As evidenced in some preliminary modeling of cement
facilities that the Agency had conducted, the factors that most significantly influence the dispersion of
airborne  PM from the source  of their emissions include the following:

       •   wind speed,
       •   deviation in wind speeds,
       •   stability class,
       •   mixing height, and
       •   source/receptor distance.

       All  of these factors or variables are determined by the climatic conditions in the  particular
geographic region where the facility is modeled.  Thus, to account for the influence of climate and
meteorology on dispersion modeling, EPA used an approach that divides the continental  United States
into seven climatic regions. Region numbering starts  at the west coast and ends at the east coast as
shown in Exhibit 3-1.  (This approach of dividing the continental United States into seven climatic
zones is  recommended in Rapid Assessment of Exposure to Paniculate Emission from Surface
Contamination Sites (U.S. EPA, September 1984)).  The underlying assumption for this approach  is
that all facilities that fall within the same region are subject to generally similar dispersion patterns.
The seven climatic regions were used for grouping purposes only - the  highest-risk facility was
selected  from within each of the seven climatic regions and then actual  meteorological data were
collected for the individual facilities modeled.

       EPA's general approach to selecting the highest risk (or high emissions) facilities was, first, to
identify the parameters that in combination have the greatest influence on emissions, and, second, to
compare the actual facility-specific values for those parameters among the facilities in each climatic
region.  The relevant parameters, actual values used, and facilities selected based on such comparisons
are discussed in detail in Section 3.2.5 of this chapter. In general, however, it should be noted that
these parameters are both operational (tons of CKD dust wasted/year) and meteorological (average
wind speed, fastest mile, and  the number of days with > 0.01 inches of rain) for each  facility. Thus,
the framework of dividing the country up into climatic zones also helps to simplify the determination
of high emissions facilities.
     Factor analysis was used to examine interrelationships between wind speed/wind direction, precipitation,
and mixing height data from 59 National Weather Service stations.  The climatic zones were defined based on
the results of the factor analysis combined with examination of other climatological information.


                              ***  Draft, August 1997, Do Ndt Cite or Quote ***

-------
                                            3-5

                                        Exhibit 3-1
                                      Climatic Regions
           \
Source:        Rapid Assessment of Exposure to Paniculate Emission from Surface Contamination
              Sites, U.S. EPA, September 1984.

-------
                                              3-6

       Once the highest-risk facilities were determined, site-specific operational and meteorological
data were used with an air quality dispersion model to predict PM concentrations for a grid of
receptors surrounding the facility. These results were then combined with actual population data to
determine the number of people exposed to levels of particulates below the NAAQS and above the
NAAQS, for the highest-risk facilities.  Because the populations located around the highest-risk
facilities  were expected to be exposed to concentrations resulting from emissions that are much higher
than that emitted at other facilities within the climatic region, and because the facilities  should have
similar dispersion patterns, these results were used  to draw conclusions regarding the lower risk
facilities.

       3.2.3   Select Model(s) for Estimating Emissions and Dispersion

       The Agency evaluated the tradeoffs of several modeling approaches and tools that  can be used
for assessing population exposures to PM, and chose the most tractable approach given  the study
design and objectives.  Compared to the previous modeling effort for the RTC and NODA, PM
modeling for the current analysis was much broader, especially in three areas: (i) including emissions
from other sources in addition to the CKD waste pile, (ii) using a dispersion model that uses more
refined meteorological data and that can possibly account for "terrain effects" (e.g., the  effects of
disposing of CKD  in a quarry) and (iii) predicting exposure concentrations within  a grid that includes
multiple  receptor points around the facility.

       Emissions  Modeling

       Emissions  were estimated using methods and equations from EPA's Compilation of Air
Pollutant Emission Factors,  Volume I: Stationary Point and Area Sources,  Fifth Edition (commonly
referred to as AP-42). The methods presented in AP-42 for estimating fugitive  dust emissions are
principally compiled  from Control of Open Fugitive Dust Sources by Cowherd  et. al. (1988),  which
was used as a supplemental  reference.  AP-42  contains the emission estimation  methods and equations
recommended for use by the EPA Office of Air Quality Planning and Standards.  It is the  best
approach available, short of  conducting new field studies to measure emissions. As noted  in AP-42,
significant  atmospheric dust  arises from the mechanical disturbance of granular material exposed to the
air.  Dust generated from such open sources is termed "fugitive" because it is not  discharged to the
atmosphere from a confined point source (stack).  Common sources of fugitive  dust include wind
erosion from CKD piles, aggregate handling (e.g., loading and unloading), unpaved road travel, and
bulldozing of CKD.  For these sources, the dust-generation process is caused by two basic physical
phenomena:

       (1)     Pulverization and abrasion of surface materials by application of mechanical force
               through implements (wheels, blades, etc.), as usually occurs during aggregate  handling;
               and

       (2)     Entrainment of dust particles by the action of turbulent air currents (from passing
               vehicles or high winds) from exposed, disturbed surfaces.
                                 Draft, August 1997, Do Not Cite or Quote ***

-------
                                             3-7

Emissions due to both mechanical abrasion and wind erosion were modeled by EPA; the specific
sections covering the equations and other background information has been extracted from AP-42 and
included in this document as Appendix F.

       Dispersion Modeling

       Several candidate dispersion models/modeling approaches that EPA evaluated are described
below:
       Option 1:  Use the Fugitive Dust Model (FDM), which is specifically designed to estimate
       concentration and deposition impacts from fugitive dust sources. One of the unique
       characteristics of fugitive dust sources, such as CKD piles, is that emission rates are a function
       of the wind speed.  FDM has the advantage of incorporating hourly wind speeds into
       calculations of both the pile emission rate and subsequent downwind dispersion. Also, FDM
       can accept  hourly meteorological data output from the EPA RAMMET meteorological pre-
       processor program.  If a threshold wind speed is specified for emissions, FDM has the
       capability of examining each hour of meteorological data to determine whether the wind speed
       is above or below the threshold, and then turn the emissions "on" and "off  based on these
       wind speeds.   Furthermore, FDM can relatively easily incorporate emissions from both area,
       volume and line sources, which is particularly important for modeling unpaved/paved road
       sources.  The main  disadvantage of FDM is that it does not handle terrain effects, which could
       have a significant influence on site-specific results.  A  secondary disadvantage is that source-
       specific contributions to predicted concentrations are not available directly from normal model
       operation.

       Option 2:  Use the Industrial Source Complex 3-Short Term (ISC3ST) model. The ISC3ST
       model  is recommended in the EPA Guidelines on Air Quality Models for dispersion modeling
       of complex industrial source facilities. (The model is included  in Appendix A of the
       Guidelines, which describes EPA-preferred air quality models.)  Like the FDM, the ISC3ST
       model  can  accept hourly meteorological data (e.g., stability class, wind direction) to predict
       hourly, 24-hour, or  annual average concentrations.  ISC3ST, however, has the advantage of
       accepting and processing terrain elevations for both emission sources and receptor points.
       ISC3ST, which cannot input line sources, can be used to simulate roadways sources by
       breaking up the roadway into consecutive rectangular area sources.  ISCST  also has the
       advantage of automatically generating data on source contributions.

       Option 3:  Use the ISC3 Long Term (ISC3LT)  model.  As with ISCST, ISCLT is
       recommended in the EPA Guidelines on Air Quality Models, and can incorporate terrain
       elevations and source contributions.  ISCLT, however,  uses a Joint Frequency Distribution
       (JFD) or Star data set of meteorological data, not a complete set of hourly meteorological data.
       The Star data  represent annual meteorological data, and this cannot be used to estimate
       concentrations for periods less than a year (e.g., 24-hour average concentration) without using
       period adjustment factors.

       Option 4:  Use the SCREEN model. Even though the  SCREEN model uses many of the
       same dispersion equations and source representations used in the ISC3ST model, it is limited
                                Draft, August 1997, Do Not Cite or Quote ***

-------
                                              3-8

       to operating for one source at a time and worst-case meteorological data.  Actual data from the
       facility location cannot be incorporated into this modeling approach.

       Given this information, EPA selected the ISC3ST modeling approach because (1) terrain
effects could be considered, (2) ISC3ST could discriminate between airborne paniculate concentrations
that are due to emissions from the pile versus those from the handling train, and (3) a year's worth of
actual hourly meteorological data could be included in the modeling.  Also, as  mentioned above, line
sources, such as unpaved roads, could be modeled  as a string of elongated area sources.  The
advantages gained in terms of handling wind-related emissions sources by using the FDM model are
not significant in  this application due to the low wind speeds needed  to entrain cement kiln dust.   Such
wind speeds are low enough that potential emissions can occur during typical wind speeds.  The
Agency did not use the ISCLT or SCREEN approaches  which are simpler than  FDM or ISC3ST,
because the modeling  will not have  been a significant improvement over what has been done
previously.  Furthermore, more extensive model runs would have been required to estimate
concentrations at  multiple receptor locations.

       It is difficult to state with any certainty how the new modeling results compare with those
generated for the  NOD A using MMSOILS.  Because EPA used a more refined  approach (compared to
a conservative screening approach),  concentrations  of PM10 at the facility boundary and the nearest
resident tended to be lower based on the new modeling, although this result is  not unequivocal. The
new modeling did, however, provide an indication  of how the low  and high concentrations are
distributed spatially around the facility.  A discussion of the current versus the previous modeling
approaches is provided in Appendix G.

       3.2.4   Determine Sources of Emissions

       In this step EPA defined the sources or source types from which emissions were to be
estimated for the  modeled facilities.  The previous  modeling efforts had assumed that all or a great
majority  of the PM emissions result from the CKD waste pile3 at a given site.   Thus, emissions from
only waste piles were included in the previous modeling.  For the current analysis EPA first
determined whether, in addition to the waste pile, significant amounts of PM emissions could result
from the  "handling train" associated with transporting and handling the CKD from the point of
generation to the  point of disposal at the facility.

       The general "handling train" is defined as consisting of the following elements of the baseline
CKD disposal practice: loading, transport, unloading, interim storage pile, and  moving into  the
pile/monofill.  For most of the facilities to be modeled,  there was very little facility-specific
information available on how exactly the CKD is handled between collection at the facility and
disposal at the pile (e.g.,  pelletized prior to transport).  EPA consulted several sources (e.g., site visit
trip report, PCA surveys, data collected  for the NODA) in an effort to determine which elements of
the handling train would  be present at the particular facilities.  Ultimately, instead of using  facility-
specific information on the handling train, which was scarce, EPA assumed that the general handling
    3 EPA uses the term "waste piles" here to generally refer to wastes accumulated in piles, quarries, and
landfills.


                             ***  Draft, August 1997, Do Not Cite or Quote ***

-------
                                               3-9

train scenario was applicable at all facilities.  The Agency did, however, modify the specific elements
constituting the train at each modeled facility based on whatever data were available (e.g., interim
storage pile may not be relevant for all sites to be modeled).

       3.2.5   Identify and Develop Data for Modeling Inputs

       Emissions

       Inputs needed to calculate paniculate emissions from the handling train and the CKD pile
include data on amount of material handled, the CKD transport truck, the road traveled, and inherent
CKD properties.  For each facility, the various parameters and their respective values are summarized
below and presented in table form in Exhibit 3-2 (the  facilities are organized in the table by climatic
region).  Included below is a discussion of the input parameters, their values,  the data sources, and the
role each input plays in the emissions modeling calculations for each source type.

       Emissions - Aggregate CKD Handling

       Emissions are calculated for each of the CKD loading and unloading steps  in the handling
train. Required as input for this equation (see Appendix F for the full equation) is  the moisture
content of the CKD, the mean wind speed, and the amount of material handled. For the highest-risk
facilities  modeled, the facility surveys were consulted and none indicated that the facility adds
moisture  to the CKD prior to transport.  Therefore, the lower limit allowed by the emissions equation
was used (i.e., 0.25  percent moisture).  Mean wind speed and data on the amount of material handled
were determined from facility-specific information.

       Emissions - CKD Transport (on Unpaved Roads)

       During transport, the abrasive nature of tires causes dust to be generated and then emitted from
the roadway shoulders due to turbulence generated by vehicle movement. The necessary input
parameters to model this emissions source include the truck data, roadway information, and the
information regarding the level of activity of the trucks on the roadway, as outlined below.

Data required for the truck include:

       •       the  empty weight of the truck (21 tons),
       •       the  number of wheels on the truck (10),
       •       the  vehicle driving speed (20 miles  per hour), and
       •       the  weight of the truck loaded with CKD (34 tons).

       These truck parameters were estimated from brochures on trucks used in hauling operations
(weights/wheels/capacity) and  site visits (speeds).  The capacity was derived by adjusting the truck
payload capacity  for the difference in density between soil (the normal payload) and CKD (the payload
used in this modeling exercise). The value for the truck  weight when carrying CKD also includes the
assumption that the truck is only filled to  75 percent of total capacity in order to prevent CKD from
spilling during transport.
                                 Draft, August 1997, Do Not Cite or Quote ***

-------
                         Exhibit 3-2
Values for Parameters that Affect Emissions from Cement Facilities

2
3
4
5
6
/
8
9
to
11
12
13
14
15
1b
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

Name
National Lebec
Holnam Tijeras
Ash Grove Foreman
Holnam Morgan UT
Calif Portland
Ash Grove Inkom
Holnam Ada
Holnam Laporte
Southdown Odessa
Southdown Lyons
Texas Industries
Ash Grove Ml City
Capitol Aggregates
Lalarge New Braunfels
Dacotah Rapid City
Lalarge Alpena
Ash Grove Chanute
Latarge Paulding
Lone Star Cape Gir
Medusa Charlevoix
Continental Hannibal
Ash Grove Louisville
Holnam Clarksville
Holnam Florence
Essroc Speed
Lone Star Greencastle
Essroc Logansport
River Festus
Lafarge Sugar Creek
Lehigh Mason City
Heartland Independ
Lone Star Oglesby
Lalarge Fredonia
Medusa Demopolis
Lalarge Buffalo
Holnam Dundee
Holnam Artesia
Lafarge Grand Chain
Lehigh Mitchell
linker Miami
Tarmac Medley
Slue Circle Ravena
Signal Mountain
Lehigh Cementon
-ehigh Union Bridge
Roanoke Cloverdale
Southdown Knoxville
Independ Hagerstown
Holnam Holly Hill
Lone Star Nazareth
_ j 	 j /^_._i.:ti
Filef
92
52
4
55
18
5
53
45
109
105
112
7
21
71
103
67
6
69
81
86
27
8
49
46
30
80
31
98
68
73
41
79
66
63
64
48
50
122
74
97
111
13
102
76
75
100
108
57
54
83

Climatic
Region
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
5
5
6
6
6
6
6
6
6
7
7

Handling/
Controls at pile
compacted
wetted.
compacted
compacted
no into
not compacted
compacted
compacted
compacted
no controls
watered,
bulldozed.
compacted
wetted,
compacted
compacted
spray
compacted
compacted
compacted
compacted
no into
compacted
not compacted
not compacted
no info
pelletized, sifted.
bulldoze over clitl.
compacted
no info
dump at edge.
after 1 week
bulldoze into
quarry, compacted
compacted
compacted
pelletized.
compacted
no into
compacted
compacted
into pond
wetted.
compacted
not compacted
compacted
compacted
quarry lake
no info
compacted
compacted
not compacted
watered
no info
no info
no info
compacted
compacted
no into
not compacted
no info
wetted,
CKD Unit
Exposed
Area (m')
79,000
101.726
77,760
46.452
39.265
557
544,062
121.449
141,264
26.662
25.919
18.581
23,226
23,226
16.729
232.260
205.309
234,803
184,877
50,167
58,557
45,300
42.548
41,822
51,095
45.522
35.302
18,208
51,840
32,342
27.123
16.200
8.733
8.906
5,855
3,716
3,624
3.345
1.487
10,219
6,040
200.671
70,685
83,600
51.022
22.483
19.500
3,716
91.282
80,407

Length
of Side
(»)
199
226
197
152
140
17
522
246
266
115
114
96
108
108
91
341
320
343
304
158
171
150
146
145
160
151
133
95
161
127
116
90
66
67
54
43
43
41
27
71
55
317
188
204
160
106
99
43
214
201

Length
of Road
One
Way
(ml)
0.3
085
057
075
057
0.57
7
057
085
0.75
1
0.76
0.2
15
1.5
095
1.2
03
028
1
075
1
038
076
1.7
028
038
0.3
1.5
1
0.25
0.57
06
05
0.17
025
0.33
0.1
1 25
0.25
0.25
0.38
0.5
0.25
~ 0 76
0.38
095
0.5
0.53
025

Total CKD
Wasted
(mt/yr)
63.490
25,752
32,633
17.318
20,092
454
76,095
77,471
42,541
65,000
115.971
19.047
21,463
9,614
15,591
430,569
60.049
29.024
608
75.467
67.082
89,318
214,505
107.120
28.296
20.226
35.404
53.777
435
8,616
1.868
5.446
67,438
16.741
20.861
5.225
8,367
8.163
22,695
907
8,641
26,657
68.993
32,652
14,734
43.997
4,387
12.954
153.398
25.418

1 Round
Trips pw
Day
15.4
62
79
42
4.9
0.1
184
188
103
15.8
25.0
46
5.2
23
3.8
25.0
14.6
7.0
0.1
18.3
16.3
21.7
25.0
25.0
6.9
4.9
8.6
13.0
0.1
21
0.5
1.3
16.3
4.1
5.1
1.3
20
20
5.5
0.2
2.1
6.5
167
79
36
10.7
1.1
3.1
250
62

» Trips
Over
Sam*
Area
5
3
2
2
1
1
6
6
8
3
5
2
2
1
1
8
5
8
8
7
6
5
2
4
3
2
5
1
2
1
1
1
2
1
1
2





6
2
3
4
1
1
1
8
2

80% of 2x
3180
3155
3608
2242
2438
26.7
8345
394.3
182.1
"4252
1847
1724
154.2
1463
172.4
545.2
5126
2334
2314
2408
253.4
2738
548.2
152.7
2126
255.7
105.7
486.5
2414
2576
2035
186.3
86.6
106.8
144.0
436
69.0
681
654
1144
87.9
300.8
506.8
327.1
169.6
255.6
158.0
690
3418
3208

Percent ol
Total Area
That in
Disturbed
15
8
5
7
5
35
7
15
42
7
26
9
9
8
5
14
9
33
33
28
22
18
4
26
12
8
48
2
6
4
5
5
18
12
6
39
13
14
14
8
10
17
4.
7
19
4
6
14
22
6

Annual Mean
WindSpMd
(m/s)
309
4
353
366
389
4.51
449
581
486
415
3.71
3.69
3.71
3.71
518
4.01
4.31
423
365
388
467
367
467
415
382
43
43
3.65
454
5.12
4.31
429
431
295
452
48
2.63
363
382
458
458
4 19
284
315
402
335
284
3.4
3.83
403










































—





VMTper
Day (mi)
92
10.6
90
63
56
0.1
258.3
21 4
17.5
236
50.0
70
21
70
11.3
47.5
34.9
42
0.1
36.6
244
433
190
380
233
27
6.5
78
03
42
02
1.5
19.6
4.1
1.7
0.6
1.3
04
138
0.1
1.0
49
16.7
40
54
8.1
20
3.1
265
31

Adjusted
CKD Unit
Area (m2)
11,747
5,990
5,407
2.621
2.457
192
36.946
17,772
10^929
10,525
6,986
2.153
1,709
1,327
1.241
32.715
17.910
14.002
13,882
12,514
11.126
10685
9.258
4,777
4,379
4,210
4,148
3,503
2,841
1,855
1,465
1 .342
1.051
1.040
1.037
576
497
490
471
823
633
12,074
7,860
6,214
4.343
" 2.191
1,138
520
20,509
4,744

Total
Population
within the
Modeling
Region
13
1.157
1,741
295
111
1 595 "
16,565
5,182
491
2.985
7.439
38
111.996
11.705
30.969
12,231
10,022
3.406
19,683
4,171
7.145
2,341
622
1.073
8,910
11.566
6,419
4,644
29,965
22,844
1,518
5,390
3.212
7,178
978
26,903
389
1.274
5,729
56,949
31,255
9,513
27,362
3,709
4,509
1,094
21,262
49.560
1.559
22.191

Distance
(m)ol
Closest
Block to
the Plant
3,626
389
1.582
1.199
1,653
"~54
151
1.004
4.717
554
1,296
" 3.473"
609
137
523
1,185
BOO
862
607
1,096
502
564
652
1.220
224
623
1.204
1,276
620
1,223
262
758
403
754
707
"269"
1,239
1,342
421
1,622
" 1.257
166
421
130
303
1.441
118
419
2.168
333
Population
of the
Closest
Block
4
11
5
11
4
3
15
4
182
0
17
4
394
1
1
6
12
0
2
1
3
3
3
9
6
17
20
43
1
29
4
30
22
5
31
85
6
6
32
1.146
4
4
237
1
~13~"
16
15
42
13
20


-------
                                              3-11

The input parameters for the road include:

        •       the silt content of the road,
        •       the distance traveled during each trip between the facility and the waste pile, and
        •       the amount of rainfall per year falling on the road.

        The silt content of the road was assumed to be 20 percent, the upper limit allowed by  the AP-
42 equation, in order to reflect the fact that not only dirt is present on the roadway but also a
significant amount of CKD deposited to on the road from previous trips. As for the length of the
road, the distance from the facility to waste pile was determined by examining facility maps and
choosing the most likely route from facility to pile. Maps provided in response to the PCA facility
survey were used in conjunction with U.S.G.S. maps to determine distance.  Data on the number of
days/year with at least 0.01  inches of precipitation were gathered from climatological data sources.

        Roadway activity information, (the number of trips per day) was calculated as the amount of
CKD wasted each day divided by the CKD capacity of the truck (12.5 tons, for hauling CKD).  The
amount of CKD wasted each day is the annual amount of CKD wasted divided by the number of
working days  per year.  For each facility, the amount of CKD wasted per year and the number of
working days  per year was taken from the respective facility's PCA survey.

        Emissions - CKD Pile Wind Erosion  & Bulldozing

        Emissions from the CKD waste pile include emissions from (1) wind erosion, and (2)
additional disturbance of the surface area of the pile.  Emissions from wind erosion of exposed
surfaces were  estimated using the procedures outlined in AP-42, Section  13.2.5, for the interim and
temporary  storage piles  and the area of the monofill disturbed during  each loadout of CKD.
Meteorological data and  information on the CKD material, as weathered in the piles, and data on the
amount of disturbances of the CKD material were  used to estimate these emissions.

        Erosion potential was calculated based on  the fastest wind speed during the period between
disturbances and the threshold friction velocity of  the material.  The "fastest mile" data from "Extreme
Wind Speed at 129 Stations in the Contiguous United States" was used to calculate emissions  for this
analysis.  This wind speed represents the mean annual  fastest mile4.

        Threshold friction velocity data specific to CKD material was also needed for this study. This
information was not available from literature or site-specific data.5  Consequently, the threshold
   4 Analysis of historical meteorological data to determine mean daily fastest mile values for each facility and
the subsequent use of these values to estimate emissions were beyond the scope of this effort.

   5 It is likely that additional field studies and/or wind tunnel measurements would be required to adequately
quantify this parameter, as previous work in the field has not focused on collecting physical data needed to
characterize air emissions of CKD.
                                 Draft, August 1997, Do Not Cite or Quote ***

-------
                                              3-12

friction velocity was estimated based on a graphical relationship developed by Gillette6 between
threshold friction velocity and the size of the aggregate distribution mode.  This graphical relationship
contains a logarithmic relationship between the two variables (i.e., the log of the aggregate size
distribution mode is proportional to the log of the threshold friction velocity) and can be used to
predict threshold friction velocities for particle size distribution modes between 0.1 and  100
millimeters (mm).  Since CKD is predominantly less than 75 micrometers (um) in diameter, the
threshold friction velocity for the smallest distribution mode included on the graph (0.1  mm or  100
urn) was selected. However, the resulting threshold friction velocity, 0.25 meters per second (m/s),
first appeared to be quite low relative to those reported in AP-42 for use in the emissions equation.
Furthermore, EPA believed that the natural tendency of CKD to crust when exposed to  moisture will
tend to increase the threshold friction velocity for weathered surfaces.  After further investigation into
the properties of soils with physical characteristics similar to those found in CKD piles, a value of
0.75 (based on  data for silt loam soil) was chosen for the threshold friction velocity.

        The pile area disturbed during each trip is calculated as a fraction of the total pile surface area.
(The total pile surface area for each facility was obtained from data collected for the previous RTC
and NODA analyses.)  Note that the total surface area normally  is not disturbed during  day-to-day
operations; that is, the majority of the CKD pile remains undisturbed and forms a crust  which
effectively eliminates PM emissions from that portion of the pile (after the initial erosion potential has
been depleted).  Only the disturbed (i.e., driven on) portion of the pile is a source of continuing PM
emissions.

        No specific data on the disturbed area are available for the facilities modeled in the present
analysis; thus, EPA developed an algorithm to determine for each facility the portion of the total  pile
surface  area that is disturbed on a daily basis. The variables of influence used were the total number
of trips per day (which  is determined by the amount of CKD wasted annually), and the  total surface
area of the CKD pile.  First, for purposes of calculation,  EPA assumed that the total pile at a given
facility is in the shape of a rectangle with one of the sides  equal to "X" and the other equal to  "2X."
Knowing the surface area of the pile, the value for X could then be derived. Next, EPA assumed that
during a single trip of a truck driving on the pile to unload the CKD, the truck would travel up 80
percent of the length 2X, turn around and then drive back. Thus, the area covered during a single trip
would be 2X multiplied by 7.2 meters (which is twice the  breadth of a typical truck used in such
operations).   A further assumption employed  was that no more than a third of the total trips within a
day are over the same area of the pile. For example, if a facility had nine trips per day, three of them
would occur  over a (7.2 x 2X) m2 area, the next three would occur over a fresh (7.2 x 2X) m2 area,
and the remaining over yet another (7.2 x 2X) m2  area.  The sum of all the three (7.2 x 2X) m2 areas
would then be counted as the total area disturbed.

        Emissions for direct disturbance of the material (bulldozing) were also examined for a  few
facilities using equations for bulldozing of overburden from Table 11.9-2 of AP-42.  These equations
require  information on the silt content and moisture content of CKD pile dust.  Silt content was
originally estimated to be 90%, based on the weight fraction of particles less than 75 um (from Dust
    6 As cited in Cowherd, Jr., C, et al., 1988.  Control of Open Fugitive Dust Sources, EPA 450/3-88-008,
U.S. Environmental Protection Agency, Research Triangle Park, NC, September 1988.


                             *** Draft, August 1997, Do Not Cite or Quote ***

-------
                                              3-13

"G" from the article "Cement Kiln Dust Management: Permeability," Todres et al., Portland Cement
Association, 1992).7  However, since the material would have weathered and crusted somewhat,
upon further examination, this value was decreased to 30%.  In addition, since the CKD is allowed to
weather prior to bulldozing, it was assumed that a moisture content of 15% exists in  the large CKD
piles prior to bulldozing.

       Dispersion

       For this modeling application, ISC3ST was set up using default regulatory options,  as
recommended by EPA guidance.  Setting this option  automatically selects the following:

       •       Stack  tip downwash,
       •       Final plume rise,
       •       Buoyancy induced dispersion (BID),
       •       Default vertical potential temperature gradients,
       •       Use of the calm processing routine,
       •       Default wind profile exponents,
       •       The appropriate value for pollutant half-life (if needed), and
       •       The building wake effects algorithm (with building information).

EPA selected the "rural"  or "urban" option for each facility modeled, as appropriate (use of this option
incorporates either rural or urban mixing heights and dispersion factors into the analysis).  For each of
the modeled facilities, the meteorological input data for dispersion modeling came from the weather
station closest to the facility and included the following information:

       •       year, month, day, hour (for each value),
       •       wind speed (m/s),
       •       wind vector (degrees),
       •       temperature (deg. K),
       •       stability category, and
       •       mixing height.

       The anemometer height, longitude and latitude, and stations numbers for both the surface and
upper air stations were required for development of the meteorological data file. Physical data on the
locations and dimensions of the emission sources (e.g. U-T-M coordinates for  the corners of the CKD
pile) and receptors was also input to ensure that the emission sources were modeled  as accurately as
possible.  Many of the CKD piles are located in depressed quarries, are partially blocked by nearby
terrain, or are within an area of elevated terrain.  Therefore terrain heights  were input to ISC3ST for
each of the sources modeled and  for each receptor located around the facility.
    7  Todres, H., et al, 1992.  Cement Kiln Dust Management Permeability.  PCA Research and Development
Bulletin RD103T, Portland Cement Association.


                             ***  Draft, August 1997, Do Not Cite or Quote ***

-------
                                               3-14

        3.2.6   Select Facilities for Modeling

        The values used for the input parameters mentioned in the previous section influence the
 results of the emission and dispersion modeling. Some of the input parameters are used directly, but
 most are combined in order to supply the necessary inputs for the modeling equations. In order to
 limit the PM modeling  runs to those facilities that pose the greatest  risk to populations, the input
 parameters (or combined parameters) that have  the greatest influence on the modeling results were
 identified.  The two parameters (both of which  are combinations of  various inputs) that have the
 greatest influence on emissions are listed below, accompanied by the formulas used to calculate their
 values.
           Parameter                                           Formula
CKD Unit Area Disturbed (m2)      =   CKD Unit Exposed Area (m2)  x % Area Disturbed
Vehicle Miles Traveled per Day      =   # Round Trips per Day  x Length of Road One Way (mi) x 2
 Other parameters, such as the amount of CKD wasted annually and the availability of controls at the
 waste pile, were considered as secondary criteria where relevant.

        Based on the actual values calculated for the parameters listed above, the following cement
 facilities were selected as the highest emissions facility  within each climatic region (no facilities were
 located in Region 1).  For a full list of the values for the variable inputs and the calculated values for
 these two  influential parameters, see Exhibit 3-2.  Listed below are the high-emissions facility selected
 from each of the climatic regions, accompanied by the reasoning behind each selection (note that all
 comparisons are only valid within the given climatic region).

 Climatic Region 2: National Lebec

        CKD Unit Area Disturbed.  The value for National Lebec is slightly higher than the next
        highest value (15,800 m vs. 15,259 m ). Additionally, the second highest value is for a
        facility that pelletizes and wets the CKD, both of which would significantly reduce emissions
        at that facility.

        Vehicle Miles Traveled per Day (VMT).  Although the value for National Lebec is slightly
        lower than the highest value (9.2 mi. vs. 10.6 mi.), the facility with the highest value uses
        pelletization and wetting within the "handling train," thus decreasing the emissions.

        Other Factors.  The surrounding population for  the National Lebec facility is extremely low
        (e.g., a total of 13 people in the modeling region).  To account for any bias that  such a low
        population might have on the  population risk results, another facility from this same climatic
        region (Ash Grove Foreman) with similar expected emissions but much larger surrounding
        population was also modeled.
                                  Draft, August 1997, Do Not Cite or Quote ***

-------
                                              3-15

Climatic Region 3: Holnam Ada

        CKD Unit Area Disturbed.  The value for Holnam Ada is over four times that of the next
        largest facility.

        VMT. The value for Holnam Ada is almost twice that of the next highest facility.

Climatic Region 4: Lafarge Alpena

        CKD Unit Area Disturbed.  The value for Lafarge Alpena is slightly higher than the next
        highest value.

        VMT. The VMT value for Lafarge Alpena is slightly higher than the next highest value.

        Other factors. While the previous two factors may not provide much discrimination between
        Lafarge Alpena  and the next highest facility,  the amount of CKD wasted at Lafarge Alpena is
        clearly much greater (twice as much) than the value for the next highest facility.

Climatic Region 5:  Tarmac Medley

        CKD Unit Area-Disturbed.  The selected facility, Tarmac Medley, only has about 60 percent
        as large of a disturbed area as the highest facility.

        VMT. Tarmac Medley has ten times the VMT as the other high emissions facility in this
        climatic region,  and it is for this reason that Tarmac Medley was selected.

        Other factors. In addition, Tarmac wastes almost ten times as much CKD as the next highest
        facility.

Climatic Region 6: Signal Mountain

        CKD Unit Area Disturbed.  Although the value for Signal Mountain is about half that of the
        highest value, the facility with the highest value uses wetting at the pile (EPA assumed that
        wetting leads to around a 50 percent emissions reduction).

        VMT. The VMT for Signal Mountain is over three times the VMT value of the next highest
        facility.

Climatic Region 7: Holnam Holly Hill

        CKD Unit Area Disturbed.  Holnam Holly Hill's area is  significantly higher (approximately  50
        percent greater)  than the next highest value.

        VMT. The value for Holnam Holly Hill is approximately eight times that of the next highest
        facility.
                                 Draft, August 1997, Do Not Cite or Quote ***

-------
                                               3-16

        3.2.7   Define Exposure Points

        For modeling airborne contamination in general, a reasonable number of exposure points must
be determined for predicting concentrations to reflect spatial variations. The spatial variations are
necessary for more  accurately predicting the number of people exposed to particular concentrations of
PM (i.e., to estimate population risk).

        Based on the modeling framework selected (see Section 3.2.2), EPA used a rectangular
network of receptors which extends out five kilometers  from the facility boundary in all directions.
The grid, which was centered on the source of the emissions,  contained a variable receptor spacing.
(ISC3ST allows for either uniform or non-uniform grid spacing, as well as discrete receptor locations.)
The Agency used a receptor grid with intervals of 500 meters within the first 2,000 meters around the
facility. Outside this "close" grid, the receptor intervals  increased to 1,000 meters.

        3.2.8   Model Emissions and Dispersion and Characterize Populations Effects

        EPA  estimated PM10 and PM2 5 emissions from the various sources at the facilities selected,
using the relevant equations from AP-42 and the input data discussed above.

        The dispersion model (ISC3ST) estimates PM concentrations for every point in the receptor
grid.  Estimates of  the number of people who are potentially exposed to various levels of PM
concentrations can  be obtained by combining predicted concentrations at receptor points with data on
the locations  of U.S. Census blocks and the actual 1990 residential populations associated with these
blocks. The  PM concentration for a given census block that falls within the modeling grid is
estimated by  interpolating concentrations at the receptor points to the block centroids. The
interpolation  scheme is based  on a weighted average of all receptor point concentrations in the
modeling domain.  The weighting factor is the inverse of the  square of the distance between the
location of the  centroid and the surrounding receptor points.   The entire residential population of that
Census block is then assumed to be exposed to the concentration predicted for the  block centroid.

        As is the case with all interpolation schemes, the interpolated concentrations will always be
equal to or less than the maximum projected concentration at any  receptor point.  An alternative
procedure would be to assign  to the block centroid the predicted concentration located at the nearest
modeled receptor point.  The interpolation procedure used for this analysis is likely to result in
somewhat lower maxima than the alternative procedure since  the influence of certain predicted
concentrations could be reduced.  The current approach, however, is more likely to provide better
overall estimates of populations exposed, especially if predicted concentration gradients  are steep
among the receptor points  (since the concentration estimate at the block centroid includes information
about all modeled receptor points).
     Note that, to be consistent with routine PM air modeling practices, EPA chose five kilometers as the
maximum distance defining the area for which the Agency would need to gather data on potentially exposed
populations.  For the previous analyses, data were gathered for total populations within five miles of the CKD
facilities.
                                 Draft, August 1997, Do Not Cite or Quote ***

-------
                                              3-17

       3.3     RESULTS

       Four sets of results are presented below:  (1) estimates of fugitive CKD emissions from the six
high-emitting cement plants selected for modeling; (2) estimates of the downwind PM concentrations
and exposed populations at the highest emitter of these six plants;  (3) estimates of exposed populations
at each of the 52 cement plants examined; and (4) estimates of exposed populations around the
universe of all 108 cement plants.

       3.3.1   Emissions from Selected Facilities

       Exhibit 3-3 presents estimated PM emissions at the highest-emitting facility in each of the six
climatic regions with cement manufacturing plants  (again, no facility is located in Climatic Region 1).
Results are presented for both PM10 and PM25 in grams per second (g/s).  Emissions from all the
possible sources at each facility are shown but totaled in two different ways, first assuming that
bulldozing occurs and second assuming there is no bulldozing.

       Comparing results across emission sources, bulldozing the CKD disposal pile would be the
largest source of emissions by  far,  if it were to actually occur at each facility.  PM10 emissions from
bulldozing range from  1.4 to 6 times higher than the next largest source at each plant.  Bulldozing is
known to occur at a few facilities,  such as at the Lafarge cement plant in Fredonia, Kansas, where
temporary CKD piles are periodically  leveled with a bulldozer.  Such activity  is a large source of
emissions  because it kicks up a lot of dust and it significantly disturbs the CKD pile surface, leaving
the dust more susceptible to wind erosion. EPA, however, has no information on whether any
bulldozing occurs at the selected model facilities, and if so, what that bulldozing actually entails.
Therefore, the total emissions with bulldozing in Exhibit 3-3 can be used as upper-bound estimates,
assuming some form of bulldozing for CKD pile maintenance occurs.  The total emissions without
bulldozing are believed to be more realistic and more representative of typical CKD management
practices.

       When emissions from bulldozing are taken out, the two largest sources of emissions at each
facility are estimated to be (1) CKD pile (monofill) wind erosion, and (2) dust suspended from the
road used  by trucks driving back and forth between the facility and the CKD pile.  Together, these  two
sources comprise from 63 to 90 percent of the total PM10 emissions (without bulldozing) from each
facility. However, one source is not always greater than the other. PM10 emissions from the road are
four to six times greater than pile emissions  at National Lebec and Tarmac Medley, while PM10
emissions  from the pile are three times greater than road emissions at Holnam Ada.  PM10 emissions
from the pile and road are about equal at the other three facilities.  The pattern is similar for PM25.

       Comparing results across facilities, Holnam Ada is estimated to be the largest emitter when no
bulldozing is assumed. The total PM10 emissions without bulldozing at Holnam Ada are  1.4 times (42
percent higher than) the emissions  at the next closest facility, Lafarge  Alpena. The PM2 5 emissions
are  at least 1.7  times (70 percent) higher than elsewhere. The relatively high  emissions from Holnam
Ada are believed to be the result of two factors.  First, the CKD pile at Holnam Ada is very large,
with an adjusted area (exposed area times percent area disturbed)  1.1 to 58 times greater than piles at
the  other five facilities. Monofill wind erosion by itself at Holnam Ada is a much  larger source  than
all the other sources combined at the other facilities, except Lafarge Alpena whose total emissions
                                 Draft, August 1997, Do Not Cite or Quote ***

-------
                                              3-18

without bulldozing (7.1 g/s of PM]0) are roughly the same as the monofill wind erosion at Holnam
Ada (7 g/s of PM10).  Second, the road between the Holnam Ada facility and its CKD pile is seven
miles long, which is seven times longer than the next longest road.  The effect of this long road,
however, is mitigated  somewhat because it is paved and thus has lower emissions per unit length than
the dirt roads that exist at the other facilities examined.

        3.3.2   Ambient PM Concentrations and Exposed Populations at the Highest-Emitting
               Facility

        Based on the emissions estimates above, EPA chose to first model dispersion and downwind
concentrations of PM  at Holnam Ada. This initial  modeling was intended to serve as a screen. If it
showed that dispersion at the highest-emitting facility results in ambient PM concentrations below the
NAAQS at all receptor points, then it might be reasonable to conclude that PM concentrations at the
other facilities, which  emit less, are also below the NAAQS. This conclusion would be valid only if
differences in meteorology at the other facilities influence the PM concentrations less than differences
in emissions (in other words, highly unfavorable meteorologic conditions that result in higher
downwind concentrations despite lower emissions are not likely to occur).

        Initial results for Holnam Ada, using the  large receptor grid with intervals of 500 meters near
the facility, indicated the annual average and 24-hour NAAQS for both PM10 and PM25 would be
exceeded farther than  500 away but not as far as 1,000 meters away from emission sources. EPA then
re-modeled the facility using the closer grid to, in effect, "zoom in" and determine more precisely the
distance at which PM concentrations  fall below the NAAQS.  The PMj0 results using the closer grid
are shown in Exhibits 3-4 and 3-5. The axes in these exhibits define a grid with  100-meter intervals
in both directions.  The numbers in each cell are the PM10 concentrations estimated to be caused by
onsite CKD management, not accounting for ambient background concentrations or particulates
emitted from other nearby sources. Three grades of shading are used to signify predicted
concentrations above the NAAQS: (1) the darkest shading represents areas directly over emission
sources, including the facility at the upper right, the road  passing diagonally down to the left, and the
large CKD pile at the bottom modeled as a rectangle; (2) the medium  shading represents areas on the
plant property; and (3) the lightest shading represents areas offsite.  Concentrations that are not shaded
are below the  NAAQS.

        As shown in Exhibit 3-4, the predicted 24-hour average concentrations  of PM10 exceed the
corresponding NAAQS out to 900 meters from the property line. Exhibit 3-5 shows that the predicted
annual average concentrations of PM10 exceed the  NAAQS as far away as 600 meters offsite.
Although not shown, the pattern of results for PM2 5 is the same.

        The latest census data for the Ada vicinity indicate that no people live in the areas predicted to
have NAAQS exceedances.  Therefore, even though CKD management on the Holnam  property is
predicted to cause NAAQS exceedances as far away as 900 meters offsite, it does not by itself result
in a single resident being exposed above the NAAQS.
                                 Draft, August 1997, Do Not Cite or Quote *'

-------
                       Exhibit 3-3
Emissions Estimates for Selected High-Emitting Facilities
Emissions Source

Material handling
Unpaved/paved road
Entrainment from truck
Temporary Storage wind erosion
Bulldozing
Monofill wind erosion



National
Lebec
(Region 2)
PM10(g/s)

0.09
0.76
0.04
0.05
2.80
0.21


Facilities
Holnam
Ada
(Region 3)
PM10(g/s)
0.00
0.55
2.16
0.29
0.15
10.06
6.97

' 10.12
Lafarge
Alpena
(Region 4)
PM10(g/s)
0.00
0.90
2.46
0.11
0.15
18.98
3.50
26.10
7.12
Tarmac
Medley
(Region 5)
PM10(g/s)
0.00
0.02
0.07
0.01
0.01
0.38
0.01
0.50
0.12
Signal
Mountain
Chattanooga
(Region 6)
PM10(g/s)
0.00
0.09
1.08
0.08
0.11
3.04

5.80
2.76
Holnam
Holly Hill
(Region 7)

0.00
0.30
1.60
0.10
0.13
6.76



Emissions Source

Material handling
Unpaved/paved road
Entrainment from truck
Temporary Storage wind erosion
Bulldozing
Monofill wind erosion

Total without bulldozing

National
Lebec
(Region 2)
PM2.5 (g/s)
0.00
0.03
0.20
0.02
0.02
1.06
0.08


Facilities
Holnam
Ada
(Region 3)
PM2.5 (g/s)
0.00
0.17
0.99
0.12
0.06
• 3.79
2.79
7.92
4.12
Lafarge
Alpena
(Region 4)
PM2.5 (g/s)
0.00
0.28
0.65
0.04
0.06
7.16
1.40
9.59
2.43
Tarmac
Medley
(Region 5)
PM2.5 (g/s)
0.00
0.01
0.02
0.00
0.01
0.14
0.00
0.18
0.04
Signal
Mountain
Chattanooga
(Region 6)
PM2.5 (g/s)
0.00
0.03
0.29
0.03
0.04
1.15
0.56
2.10
0.95
Holnam
Holly Hill
(Region 7)

0.00
0.10
0.42
0.04
0.05
2.55
0.68
3.84
1.29
                       3-19

-------
                                              3-20

       3.3.3   Exposed Populations at 52 Cement Plants Examined

       Because emissions from Holnain Ada are predicted to be much larger than emissions from
other facilities, EPA believes the estimated distance of 900 meters  for NAAQS exceedances at Holnam
Ada is an upper-bound distance (without bulldozing). Notwithstanding differences in results caused by
different  meteorologic conditions and terrain, the lower emissions at other facilities should result in
PM concentrations falling below the NAAQS  at closer distances.

       In an effort to better represent other facilities, EPA modeled emissions and dispersion under
four additional scenarios. Of these four scenarios, the  first two were hypothetical scenarios created by
adjusting the conditions modeled for Holnam Ada.  First, EPA modeled a CKD pile one-half the size
of the pile at Holnam Ada combined  with a one-mile, rather than a seven-mile, paved road.  Second,
EPA modeled a CKD pile one-half the  size of the Holnam Ada pile combined with a one-mile
unpaved road. These two hypothetical  scenarios were  thought to better approximate the emission
conditions that actually exist at some of the remaining  high-emission facilities.  (Note that both of
these scenarios were modeled using the Holnam Ada meteorology  and surrounding topography.) For
the remaining two scenarios, EPA chose to model emissions and dispersion for the Lafarge Alpena and
Signal  Mountain Chattanooga facilities, which are the next two highest-emission facilities after
Holnam Ada (see Exhibit 3-3), and also represent climatic regions  that contain, relatively, a large
number of the 52 facilities examined.

       The estimated PM10 concentrations for the first two hypothetical scenarios are presented in
Exhibits 3-6 through 3-9.  As shown, both scenarios resulted in the 24-hour average NAAQS being
exceeded out to a distance of 500 meters and  the annual average NAAQS being  exceeded out to a
distance of 400 meters.  Again, the pattern of results for PM2 5 is the  same.

       The estimated P* "• Q concentrations for the Signal Mountain Chattanooga facility are presented
in Exhibits 3-10 and 3-11, and the estimated PM10 concentrations  for the Lafarge Alpena facility are
presented in Exhibits 3-13 and 3-14.  As shown, at both these facilities, the dispersion resulted in the
24-hour average NAAQS being exceeded out to a distance of 100  meters and the annual average
NAAQS  being exceeded also out to a distance of 100 meters. As  before, the pattern of results for
PM?5  is  very similar to results for PM,0. The only exceedances of NAAQS by  PM25 particles are for
the estimated 24-hour average  concentration, out to a distance of 100  meters for the Signal Mountain
Chattanooga facility, and 100 meters  also for the Lafarge Alpena facility (see Exhibits 3-12 and 3-15).
Census data for the vicinities of the Signal Mountain Chattanooga  and Lafarge Alpena facilities
indicate that no people live in  the areas predicted to have NAAQS exceedances.  Therefore, even
though CKD management on the properties of these two facilities  is predicted to cause NAAQS
exceedances as  far away as 200 meters offsite, it does  not by itself result in a single resident being
exposed above the NAAQS.

       Recognizing that the two scenarios examined for Holnam  Ada are likely to overestimate the
magnitude and areal extent of  NAAQS exceedances at other cement plants, EPA determined that as a
first step, it would be reasonably conservative to assume that everyone living within 500 meters of the
other facilities is exposed to PM concentrations above  the NAAQS, as long as the facilities do not
manage- CKD in a manner that effectively contains it from fugitive emissions (such as in the form  of a
slurry). Starting from this point, EPA  further determined that the  results from the scenarios at the
                                Draft, August 1997, Do Not Cite or Quote

-------
                                              3-21
Signal Mountain Chattanooga and Lafarge Alpena facilities indicate that NAAQS exceedances are
likely to be at distances much closer to the facility or waste pile than the 500 meters.  Thus, the
Agency reasoned that it would be more realistic to derive "distances for NAAQS exceedances" that are
specific for each facility, or, at least, for a group of facilities within a specific climatic region. These
distances were derived first for the set of "high-emitting" facilities (see Exhibit 3-3) by multiplying the
500 meters by the ratio of each facility's emissions estimate to that at Holnam Ada.  (The minimum
distance or the "floor"  for the "distance for NAAQS exceedances" was  set to be  100 meters.) The
distances derived for each climatic region's representative facility were then used for all the remaining
facilities within that region, all of which are estimated to have emissions lower than the representative
facility. The "distances for NAAQS exceedances" calculated  are as follows:
      Facility             Region

      National Lebec      2
      Holnam Ada        3

      Lafarge Alpena      4

      Tarmac Medley      5
      Signal Mountain     6
      Chart anoooga
      Holnam Holly Hill   7
Distance to
NAAQS Exceedance

100m
500m

100m

100m
100m

200m
Basis for Estimation

ratio of emissions X 500 m
emissions and dispersion
modeling
emissions and dispersion
modeling
ratio of emissions X 500 m
emissions and dispersion
modeling
ratio of emissions X 500 m
       Use of 500 meters - across all facilities - as the "fenceline" for determining the number of
people exposed to NAAQS exceedances yields the results  shown in Exhibit 3-16. These results
indicate that a total of 10 facilities have at least one person living within 500 meters of the plant who
may be exposed to airborne PM concentrations in excess of the NAAQS.  All the other facilities are
likely to have no offsite populations exposed above the NAAQS, either because there  are no
residences within 500 meters (36 facilities), CKD is watered and unlikely to be emitted at levels above
the NAAQS (three facilities), or site-specific  modeling and analysis indicate that no people live in
areas where the NAAQS are exceeded (three facilities, i.e., Holnam Ada, Signal Mountain
Chattanooga, and Lafarge Alpena).  As the next step, for the  10 facilities, EPA used the facility-
specific "distances to NAAQS exceedances" (either 100 or 200 meters, determined using the
representative facilities' emissions ratios) and re-derived the estimates of populations potentially
exposed using USGS topographic maps along with GIS tools to map block-level census data.  This
refinement step indicated that only two of the 10 facilities have populations within 100 or 200 meters,
i.e., Ash Grove in Inkom, Idaho (3 people), and Southdown in Knoxville, Tennessee (15 people). In
sum, therefore, the results indicate that across all facilities, a total of 18 people may be exposed to
airborne PM concentrations in excess of the NAAQS.  (Note, however, that there is more uncertainty
associated with estimating the number of people living within 100/200 meters of the facility compared
to 500 meters of the facility — see discussion in the limitations section.)
                                 Draft, August 1997, Do Not Cite or Quote ***

-------
                                              3-22

       3.3.4   Exposed Populations at All Cement Plants

       As with the indirect exposures  analysis, the focus of the PM analysis was on assessing
exposures at 82 of the total  108 cement facilities.  Again, these 82 facilities can be denoted as the
"known universe."  The remaining 26 facilities were excluded because a lack of relevant data (e.g.,
data on constituents in CKD wastes or on types of waste management practices)  prevented them from
being assessed  directly; as before, these 26 facilities can be  denoted as the "unknown universe."  To
derive a composite picture of potential population effects due to PM exposures across the  full universe
of cement facilities, EPA first estimated the potential population effects within the unknown universe
by extrapolating from results within the known universe, and then estimated the potential population
effects for the full universe of facilities.

       For population effects due to PM exposures, EPA determined that it would be appropriate to
first define the  "bounds" of the results  for the full universe  of cement facilities, and then derive a "best
estimate" with the understanding that the best estimate  will be less certain compared to the bounds.

       To define a lower bound measure of the population  effects, it is reasonable to assume that
every single facility in the unknown universe is as "risky" as the lowest-effects facility in  the known
universe.  Being the lowest-effects facility in the known universe implies that, within 500 meters (and,
therefore, within 200 or 100 meters) of the facility boundary, there are no people exposed to levels
above the NAAQS. Thus, the lower bound measure of the population effects is that, across all 108
facilities, a total of  78 people living within 500 meters of the facility boundary may  be exposed  to
airborne PM concentrations  in excess of the NAAQS.

       To define the upper bound measure of the population effects EPA assumed  that, even for the
remaining 26 facilities, the use of the 500 meters "fenceline" for determining the  number of people
exposed to NAAQS exceedances is more reasonable than using 200 or 100 meters.   This is primarily
because the sizes of piles at these facilities are smaller than that at Holnam Ada,  and therefore,
exceedances beyond 500 meters are unlikely.  At the same  time, EPA did not have  sufficient
information to be able to say that exceedances are likely to be only up to 100 or  200 meters.
Therefore, the Agency estimates that, across all the 26  facilities in the unknown universe, a total of
approximately 4,100 people live within 500 meters of the facility boundary.  (Note that this estimate is
less certain than that derived for the 82 facilities in the known universe, because  of uncertainties in
knowledge of the specific locations of the facilities and waste piles.) Thus, the upper bound measure
of the population effects is that, across all 108 facilities, a total of approximately  4,118 people living
within 500 meters of the facility boundary may be exposed  to airborne PM concentrations in excess of
the NAAQS.

       The method for deriving a best estimate measure for PM exposures is different from that used
in the indirect exposures analysis because, in the latter, deriving a  best estimate meant assuming that
the distribution of risks among the 26 facilities (the smaller, unknown universe) is similar to the
distribution of risks among the 82 facilities (the larger, known universe). This assumption was
necessary because EPA could not derive a facility-specific  risk estimate for any of these 26 facilities.
In such a case, the results from the known universe could be directly extrapolated to the unknown
universe, using one or more "weighting factors" that are common to both universes and are expected
                                 Draft, August 1997, Do Not Cite or Quote ***

-------
                                              3-23

to be related to the potential risks. The most relevant fact known about the 26 "unknown universe"
facilities was the quantity of CKD wasted; thus, waste quantity was used for the weighting.

        In contrast, for the PM exposures analysis, EPA had sufficient information such that the
Agency could derive a facility-specific estimate of population affected for each of these 26 facilities.
There would be several uncertainties, however, associated with these predictions; thus, EPA recognizes
that more refined estimates in all  likelihood would be lower than these predicted values.  One way that
EPA chose to make the best estimate measure more realistic was to determine which among the 26
facilities would have no releases from the waste piles because they were watered/wetted.  Using
information from the updated 1995 CKD Survey,  EPA determined that about 25 of the 70 respondent
facilities use watering/wetting to control releases from waste management units: Thus, EPA
designated (using a random number series) a corresponding proportion of the 26 facilities in the
unknown as having watering/wetting controls, effectively eliminating any releases.  Based on the same
reasons discussed for the upper bound measure, EPA decided that, for the best estimate measure, the
use of  the 500 meters "fenceline" for determining the number of people exposed to NAAQS
exceedances is more reasonable than using 200 or 100 meters. With these assumptions, the Agency
estimates that, across all the 26 facilities in the unknown universe, a total of approximately 2,360
people are likely to be exposed (because they live within 500 meters of the facilities where releases
occur).  The best estimate measure of the population effects, therefore, is as follows: across all 108
facilities, a total of approximately 2,378 people living within 500 meters of the facility boundary may
be exposed to airborne PM concentrations in excess of the NAAQS.
                                 Draft, August 1997, Do Not Cite or Quote ***

-------
                                                                        Exhibit 3-4
                                 24-hour  Average  PM10 Concentrations  at Holnam  Ada (Close  Grid)
                                                                                               ABOVE 150 ug/m3 - INSIDE FACILITY
                                                                                               ABOVE 150 ug/m3 - OUTSIDE FACILITY
(JI1X.FI3 5UX.F8 2,3X,A5,2X,A8,2X.A4.6X.A8)
                                                                                  705550
                                                                                  151.17
                                                                                  178 «
                                                                                   213.1
                                                                                  239.72
                                                                                          705650    705750    705850
14729
131.32
170.43
211.16
227.49
214*4
2*2.42
IS2.II    13871
170.17    158.55    14638
1S3.S7    1I3.OT    171.2*    144 76|   10024
240.79    222.69    211.33    17731    11823
322.66   , 2S153    2(4.24    232.35
751«    723«     «32    56317
734.91    713.31
175.91
203.63
245.09
30919
Jli.74
51719
500.09
                                                                                                           1.M    US.M
                                                                                                          I54.4SJ   153»    14122
                                                                                                          128 II    1296
                                                                             3-24

-------
                                                               Exhibit 3-5
                             Annual Average PM,0 Concentrations at  Holnam Ada (Close  Grid)
                                                                                            m3 - INSIDE FACILITY
              ANNUAL VALUES  FOR
                                                                                   ABOVE 50 ug/m3 - OUTSIDE FACILITY
(3(IX,FI35).IX.F82,2X,A6,2X,A8,2X.I8,2X,A8)
                                                                               4233
                                                                               }J,1»    47.55     373
                                                                               W.*7    *}.«    M.11
                                                                               9943    «.!«    766*
                                                                              264.H   251.06   M9.J5
                                                                              I3I.M   J3J.67
                                                                              MI.32
                                                                  3-25

-------
                                                         Exhibit 3-6
                                 24-hour Average PM,0 Concentrations at Holnam Ada
                                           (Hair-pile and Paved Road Scenario)
                                                                           ABOVE 1 SO ug/m3 - INSIDE FACILITY
                                                                           ABOVE ISO ug/m3 - OUTSIDE FACILITY
(3(1X,F13 51.IX.F8 2.3X.A5,2X,A«,2X,A4,6X.A»)
                                                            3-26

-------
                      Exhibit 3-7
Annual Average PM,0 Concentrations at HoJnam Ada
         (Half-pile and Paved Road Scenario)
                                       SOURCE
                                       ABpVEJOug/mJ - INSIDE FACILITY
                                       ABOVE 50 ug/m) - OUTSIDE FACILITY
                      3-27

-------
                                                                    Exhibit 3-8
                                           24-hour Average PM10 Concentrations at Holnam Ada
                                                    (Half-pile and Unpaved Road Scenario)
                                                                                    ABOVE 150 ug/m3 - OUTSIDE FACILITY
FORMAT (3(1X.FI35),IX.F82.3X,A5.2X,A8.2X.A4.6X.A8)
                                                                   12755   13529
                                                                   160.42   173.77
                                                                   261.02   312.74
                                                                   267.01
                                                                     258   373.15
                                                                   526.77
                                                                      3-28

-------
                                                        Exhibit 3-9
                                 Annual Average PM|0 Concentrations at Holnam Ada
                                         (Half-pile and Unpaved Road  Scenario)


—
	

706450
5.47
692


	
	

706553
558
784


	
	 	

706600
569
864

"~

	
-
706700
564
11 67
                                                                         ABOVE SO ug/m3 . INSIDE FACILITY
                                                                         ABOVE SOug/mJ - OUTSIDE FACILITY
3(IX.F13 5),IX.F8 2.2X.A6,2X,A8,2X,I8,2X.A8)
                                                       3-29

-------
                           Exhibit 3-10
24-hour Average PM,0 Concentrations at Signal Mountain Chattanooga

1K87X1X
IXX71IK
1XX6XO<
188611H
18Xh2IX
IXXftllM
1XXMXM
1XX5VOI
1X8.5X01
18XS7IX
1XX5MH
1X8551K
1885400
3885300
1XX5200
18X5100
IXKttKMI
1884y()0
18X471X1
18X4600
1XX4SOO
1XX44OO
1884KXI
1884200
1XX4IOO
1XX4IXN)
1881X00
188111X1
18X211X1


1 1(7 1 65
II,' 143
i>y8 172
1 14 1 45
1 11 liS4
IS 1 M.
1 68 IS
1 XX 1 60
202 1 X7
I 6V 2 54
1 52 2 68
119 259
1 2X 227
1 27 2 IIS
m IK?
2 267
16f> 2X2
1 24 1 MS
14 2.1ft
1 82 112
211 144
2 If. 1,54
2 IX 448
2X8 471
101 611
4 46 8 51
7H 748
2 1ft 1 51


16
1 66
21y
252
2X2
241
242
249
2 7V
252
102
191
462
47
1V1
144
459
1 25
411
5 74
6 K4
752
x 95
1077
II 76
10 V
4 XX
1 54


2 14
109
.116
1 5V
i 7y
lyX
412
4 46
578
6 ft 7
60V
ft.K
6 81
9 54
12(16
9 21

1215
17 51
16 II
2054
11 66
11)21
11 57
6 5V
2 yx


2 11
1 Oft
4 44
4.57
45
4 45
4X2
5 15
656
641
X 46
81
y:x
xy4
145
14 6X

Ih 15
22 21
2576
21 17
17 52
13 n
16 yx
14 61
71)1
1 5ft

— M — hsmnn — 654 II 1214 12 17
16X 17^ 14 401 5 15 651 7X1 y 78 12 HI (5 5y 17 11 17 5X 165 1481 11 15
•* -y 5 17 561 5 15 ft 7 X 77 ID 77 15 14 2001 22 7X 22 6 21) 25 17 14 91 12 ft
<•*! •<('rt 622 5.85 711 y41 II 65 17 115 22 1 1 2457 2155 2051 171 1458 tl 1
55 5X7 ft 87 661 774 IK 21 1282 IV OX _M 5 2644 2441 2068 171 11 75 10 7
5y? 615 75X 751 861 II 12 1412 215 2724 28 16 25 OX 2062 1642 1222 X ft
6 19 66* It 11 857 97 122 16 11 2445 104 10 2y 25 S| 21)07 1481 10 y y
| 	 6J 7?y y 18 y78 noi 115 i»iy 2x08 1407 12 n 2541 "m4s>| 12? n 112 86
705 y II II M 1-102 1479 1837 25 .01 IX 7y 4157 11 W 2028 1681 1098 1202 I262J
1071 12 h4 1547 3477 ?7 1> IX Xft V. *-l ' '^ '"'BJJJBI " lo " lu ln " 'u y 7 .'ft 74 5 44 4 X6 4 A
7 X7 7 44 7 7V S 75 4 W 4 21

8 87 X 22 7 ft 5 75 5 Oft 4 hi
N 61 7 yy 7 48 6 0.5 5 27 -1 (,

X 45 7 X3 1 21 509 1 
HH3I Above NAAOS OuisiJc hiulily
                             3-30

-------
                             Exhibit 3-11
Annual Average PM|0 Concentrations  at Signal Mountain Chattanooga

3887800
3887300
3886800
3886300
3886200
3886100
3886000
3885900
3885800
3885700
3885600
3885500
3885400
3885300
3885200
3885100
3885000
3884900
3884800
3884700
3884600
3884500
3884400
3884300
3884200
3884100
3884000
3883900
3883800
3883300
3882800
3882300
648500 649000
0119 012
01 01.1
0.09 0 14
0119 0 14
0,09 014
0119 0 14
OOK 014
OOK 0 14
O.OH on
0.09 0. 1 3
0.09 013
01)9 0 11
0.09 014
OOK 0.14
0118 0 14
O.OX 0 13
01)9 0 13
009 014
0.09 0. 1 5
0,1 016
on on
012 02
0.13 0.22
0.14 024
015 025
016 027
017 0.28
0.17 0.29
OIK 0.29
0.19 024
016 02
0.14 0.2
649500
0 19
021
022
025
0,26
0 26
027
027
0.27
0.26
0.26
025
0.25
0.25
026
027
026
0 26
028
03
035
041
046
0.49
051
052
0.51
0.49
0.45
036
033
0.25
650000
0.31
0.37
044
053
054
056
057
058
0.6
06.1
0.66
07
073
0,73
072
071
074
078
079
0.88
II
1.29
1 39
1 36
1.22
107
0.98
0.93
09
061
0.4
0.29
650100
0.14
04
05
0.61
066
069
072
075
0.77
0,79
0X2
087
093
098
1
099
0.99
1 06
108
1.2
1 5.1
IK
1.87
1.7
1 46
11
1.23
1 19
III
0.63
0.41
0.3
650200 650300 650400 650500 650600 650700 650800 650900 651000 651100 651200 651300 651400 651500 651600
037 0.4 043 045 047 0.47 0.47 045 0.4.1 041 0.39 037 035 033 0.31
045 05 0.55 0.59 061 062 061 059 056 052 049 0.46 043 041 039
076 091 107 122 1.14 119 1 38 131 12 1 OK 098 0.89 0.8 0.71 0.61
O.K5 105 1.27 1.49 168 176 175 166 151 135 12 1.06 091 076 063
09 113 1.4 1.67 1,9 202 202 192 174 153 134 1,15 0.96 0.7K 064
0.95 1.22 154 1.89 • 2.18 2.14 2.35 2.25 203 176 15 124 1 08 065
I.01| 1.31 1.71 2.15 253 275 2.K 2.7 2.42 2.05 167 1 12| 101 OKI 066
1116 1.42 1.9 . 2.46 2.97 329 342 3.17 3 2.42 185 1.19 l.06| 081 068
M 153 212 285 355 403 434 442 391 289 203 146 109 087 072
1.15 1.64 237 3.34 433 507 5,78 6.35 5.66 347 219 153 115 0.9.1 0.77
121 173 263 394 518 658 8.34 1096 1062 4.14 241 166 125 099 OX
13 1X4 2,92 471 687 X9I 1 4 09 ^3I04|HB 524 27 1.82 1.34 103 0X2
141 2 3.2.3 572 9.IK 1 \ l\ ^Ultfff III i 1 486 286 194 14 1.07 O.K6
1 45| 237 412 K.55 19. 7 3 ^^^^^32^9 1166 6.24 3 K7 2.6.1 1.93 149 I.IK 097
151 Til 454 10.5 33.92^^^H 25.47 10.33 5.77 372 264 197 1.53 121 0.99
1 5K 251 4 69|__I2J9^I
-------
                             Exhibit 3-12
24-hour Average PM2 5 Concentrations at Signal  Mountain Chattanooga

3887800
3887300
3886800
3886200
3686000
3885900
3885800
3885700
3885500
3885400
3885200
3885100
3885000
3884900
3884800
3884700
3884600
3884400
3884300
3884200

3883900


3882300

648500 649000
0 IK 0 5 1
0 43 0.5
0 16 (1 6H
0 44 0 62
059 0.55
0 (if) Of) 1
0.7 075
0.7 OW
Oftl 1.1)7
OS 1 04
0.5 0 7ft
072 072
II « ] 06
OM> 1.13
052 0 K7
05 072
0 55 O.M2
OKA 1.3
089 137
OH9 1.73

II 244


OK3 1.25

649500
1)6
Ofil
0 7h
1 12
0.97
095
093
1 06
121
1.56
1 88
1.57
1 .12
1.83
1.72
1 2
16
2.61
2K5
2.9

Jill


1 16

650000
(I7K
(IKK
1 Id
1 .16
1.5V
1 72
1.98
201
2.65
2..1I
236
382
4.112
368
4.15
343
475
6.14
7.93
58

47


1 14

650100
UK
101
1 16
1 74
169
IKK
2(16
229
254
336
5.4 1
.157
58
5.87
533
4 69
6.24
996
8 16
6.43

6.59


1 22

650200 650300 650400 650500 650600 650700 650800 650900 651000 651100 651200 651300 651400 651500 651600
0.74 0.76 O.K9 107 12* 1.49 169 1 »7 22 27 319 361 39 4 392
1.02 1192 101 126 151 IK 2.07 2 U 299 366 423 459 467 449 414
137 13* I2K 161 1 K5 326 2 .hi .115 429 513 562 5.65 528 i 72 419
1 114 2.2 2.28 229 2K 329 4 573 723 787 746 649 552 466 377
227 2.46 29 2.76 345 3 88 494 716 8.7V K.9K 79 652 5.24 3.96 2.75
2.37 2.56 3.26 314 3.88 4.25 556 809 973 9.53 8(11 637 478 329 2.K9
2.3| 2,95 3.6A 359 441 4.71 6.33 922 IOK 1005 K.OI 59|| 3.99 3.3(1 276
246 313 40H 413 5(17 531 7.3 1063 1203 11148 764 499 4 08| 3.09 325
3.24 15 481 6.14 7.02 7.95 1(125 ISOI 1548 9,55 6.91 418 446 471 483
3.61 418 5.65 7»9 8.49 9.91 12.72 1912 1846 1079 5.47 554 511 569 5.34
444 5.26 7.0V 1172 13 3K I7K3 i^Tff^/jjf^^^ 6,65 6 K4 6,64 5.82 4.85 45
466 684 9.1.3 13.03 1809 25 K^t^^M 14.77 8.1 8.24 731 596 544 5.35 5.07
l,| 681 1131 1573 2V7llBIB| 3877 1365 1042 8.71 689 658 623 577 5,28
S79 II 35[ 1167 21.25 « >^^^H 32.21 1652 116 II. 1 9.45 7.87 672 583 5.1
696 10.74 2I.57| 1718 I.'I59^^^J 4098 '2116 18.49 1392 10.29 7.67 5.97 4.89 4.15
667 10 16 Hh _~^~^IHIPB^HIIH| 28.25 17.29 11.77 845 641 5.13 4.26 3.61
84 13.9 21.37lBHH|^^^^^^jXmHjH| 2815 20.18 1517 1139 864 666 5.25 4.22
13.24 1364 2216 16.4 3907 45.48 19.95 19.65 15.55 996 7.42 5.51 486 454 446
9 11.68 2187 22.93 2476 2876 16.47 1027 11.37 1122 692 5.58 475 1.56 3.2K
721 1384 1399 13.49 16.93 1885 1208 876 6.32 7,53 828 566 442 3.83 3.37

627 661 5.V2 6.55 8.13 74 8.0V 536 5.72 4.98 3.87 2.89 3.29 3.98 3.69
493 549 50V 546 739 6,67 6.9 467 5... 506 4.2 325 2.43 269

IM 207 2.48 2.78 2.87 2.66 2.28 182 149 1.48 I.V 2.31 264 285 294

651700 651800 651900
37 3 39 3.07
373 336 107
377 341 KM
286 2.01 2
2 52 2 22 2 24
2.47 251 279
284 313 .13
1.53 369 3.7
4.7 4 36 1 92
4.111 4.l»| 3 6(1
432| 408 3.8
471 414 398
4.82 4 39 3 99
4.5 3.99 35(v
1.62 1.22 2.92
1.14 287 266
348 315 29
416 3,75 331
299 2.87 28
2.67 2 39 2 26

2.77 22 198
3.19 2,58 186

288 2.71 245

652000 652 00 652200 652300 652400 652500 6b3000 65350U 654.
279 5K 244 214 225 215 lil'v 075 1
2 K7 /I 2 52 2 29 2 1 f,H (I'M 1 ; i
2 69 21 1 76 It 121 1 .' 1 1 V, 19 1
1 84 Ml 2 (16 225 2 IK 24* 191 11 i
2.5 27 2 K 2 K 271 2 57 1 77 6 1
29» 305 3112 2»9 271 25 IK 155 i
3 15 3 27 309 2K7 261 241 1 K5 1 55 1
3 56 311 3 06 2 79 2 54 2 34 1 K3 16 1
3.5 311 2K3 26 242 229 1 K7 1 62 1
323 2.91 275 2(1 246 231 1 84 155 1
3.51 324 3 278 259 24 168 129 1
366 336 309 2 HJ 262 241 162 1 15 0
361 3,28 29K 271 2 4K 2.27 1 54 1 15 II
267 24K 231 217 205 1 V4 153 127 1
249 235 223 2 12 20.1 1 94 1 61 1 16 1
269 2.52 2.18 226 215 206 17 1 46 1
297 281 267 254 242 231 I K9 It,; !
2 74 26 25 242 2 14 2 26 1 92 1 66 1
225 2.22 22 21 211 2 OK | K6 1 65 1

1 K.I 1 82 .66 4 1 41 1 11 1 4t 1 19 1
1
1.29 156 169 16 147 121 075 069
214 IK 4K 1 IK 091 069 1) 5K 077 0

                                                                                           |Ahuvc NAAQS. Oimidc Haulii)
                              3-32

-------
                      Exhibit 3-13
24-hour Average PM,0 Concentrations at Lafarge Alpena

4996800
4996300
4995800
4995300
4994800
4994300
4994200
4994100
4994000
4993900
4993800
4993700
4993600
4993500
4993400
4993300
4993200
4993100
4993000
4992900
4992800
4992700
4992600
4992500
4992400
4992300
4991800
4991300
4990800
307500 308000 308500 309000 309100 309200 309300 309400 309500 309600 309700
408 423 4% 571 5.58 5.36 504 466 426 465 5
5.46 467 473 579 6.18 654 681 693 6.83 657 616
712 697 621 547 5.56 573 6.02 643 695 752 804
1556 1059 814 855 8.57 853 8.47 832 806 769 722
3614 3456 3109 2524 2376 2219 20.65 18.94 174 1583 1424
3404 3803 42.12 4572 46.29 46.83 4736 47.88 4843 4918 50.26
3085 35.08 3948 4435 4526 46.09 46.82 47.4 4795 4842 4903
2736 3142 3585 4123 424 4358 44.77 45.77 4669 4747 4793
2355 276 316 365 37.7 39.04 40.52 4207 4363 4529 4673
1986 23.57 27.38 31.12 3199 3301 3416 356 3724 3916 4141
1615 1973 2347 2633 2687 2732 2782 2845 29.19 3023 3156
1261 1594 1974 2297 23.34 23.46 2355 2359 2339 2331 2328
9.46 12.26 15.97 202 2087 2146 21.93 2204 21.9 2137 2073
677 886 1208 1674 1787 1899 20.16 21.06 21.95 2231 22.48
465 6.06 8.34 1214 1318 1444 1585 174 1915 21.05 22.9
308 384 509 728 802 8.67 9.71 10.7 11.96 1361 1575
317 397 468 496 492 586 72 883 1067 1258 14.27
351 412 4.45 529 637 762 8.99 1032 1148 1232 1276
3.68 4.03 3.92 673 7.77 876 9.59 10.15 10.37 1046 11.02
368 371 384 764 828 8.73 8.88 8.94 9.39 965 971
3.5 3.22 483 7.7 7.84 781 811 832 834 816 773
316 301 566 705 7.1 7.26 728 712 6.78 621 608
2.72 3.76 6.06 6.42 6.45 633 606 562 501 537 6
2.48 • 4.45 5.94 567 547 513 467 448 4.93 5.33 5.71
306 492 538 472 436 389 4.25 4.66 503 53 608
363 503 4.72 37 3.69 405 4.41 4.75 502 5.6 634
345 283 28 395 41 4.47 514 639 7.9 962 1147
1.81 2.39 3.28 5.28 6.21 7.26 8.44 971 1097 1222 1329
2.09 281 397 7.5 841 935 1029 1117 11.9 1241 1261
309800
5.28
563
841
7.75
1259
51 78
4991
48.32
4804
43.92
33.27
23.38
19.96
21 66
24.6
18.67
15.66
1286
II 36
956
7.02
696
64
669
689
8.37
1332
14 1
12.5
310000 310500 311000 311500 311900
5.4 7.65 1344 24.86 289
615 887 1628 2744 3157
834 11.71 2234 31.02 43.94
875 1405 30.78 49.26 6107
1224 1347 3427 7627 85.32
5601 7266 5776 157 113
5338 81.11 102.7 248.22 144.13
49.16 70.84 11706 46544 2392
49.66 59.54 ^T3A\]jjfff 177.93
5031 71 93lHH|63T85 10161
3854 I2T64^^^B 181.71 85.93
23.92HHH| 777 105.46 63.58
I899^^^H 36.42 70.52 41.33
18.91^^^1 34.59 51.15 23.83
24.51^^^1 3278 39.5 17.84
2699^^^! 3534 31.82 15.66
I7.92^m 3554 26.35 13.92
13.55 17629 3575 2214 1437
11.82 59.27 33.93 1887 15.02
983 32.51 27.27 16.2 15.29
9.3 2178 2244 1404 1527
864 1985 19.62 12.23 15.02
912 20 172 1074 1462
II 16 1994 1485 9.49 14.11
1287 19.58 12.82 868 1355
14.18 1901 1112 846 12%
16.13 1426 6.56 7.86 1006
14.37 9.77 5.6 7.44 776
1132 6.92 5.2 674 6.07
312000 312100 312200 312300 312400 312500 313000 313500 31400O
29.53 29.68 28.67 2648 26.48 29.13 41.84 3567 12.97
31.99 31.85 33.98 3671 39.5 4223 4535 18.82 816
45.02 46.77 48.94 50.93 5302 5383 28.76 92 566
61.95 63.01 63.36 619 56.36 4692 10.94 1168 1708
78.98 6736 4874 2883 2209 194 2483 25.57 2671
79.43 6055 57.25 5933 59.86 5831 43.16 3241 259
^9M9 94.92 89.56 821 74.67 67.98 467 3572 29.12
•MM li89 115.44 99.1 " 87.32 78.31 52.72 4048 329
149.53 128.73 112.7 10008 90.02 82 57.02 44.09 3633
77.87 77.07 76.04 7438 719 6919 5505 4467 37.71
7676 59.95 5337 4866 4954 50.04 4739 4185 36.81
6779 6282 50.71 46.01 4253 3796 3745 3653 3416
52.11 56.31 53.41 45.6 39.24 3817 28.83 303 3005
35.89 4481 48.61 4676 41.42 35.47 268 24.4 2561
23.18 32.24 39.45 42.77 4174 37.85 27.89 19.58 2112
16.74 22.34 29.44 35.32 38.15 37.61 2675 21.23 17.25
14.47 1688 21.54 27.3 3204 3447 2441 22.23 1641
13.64 14.5 1682 2081 2546 29.35 2529 2213 1781
13.06 1347 145 16.79 20.21 2406 2636 21 19 1864
1244 12.75 133 1452 1663 1967 2662 1989 1881
11.69 1206 1248 1317 1448 1649 2586 1922 1848
11.71 1135 11.72 12.21 1302 1434 2424 2033 1776
12.11 1072 11.05 1145 12 1287 21 9 21 II 1681
12.31 1034 10.54 10.85 11.24 11.83 194 2142 158
12.35 10.01 10.17 10.39 10.65 1108 1704 2119 1664
12.26 9.92 984 1004 1025 10.54 15.03 2049 1746
1075 10.33 8.86 678 7.2 776 995 13.7 1722
8.81 9.29 9 799 6.5 485 6.38 9.12 1254
7.11 7.9 8.24 803 729 6.17 161 542 824
                                                                                 Source

                                                                                 Above NAAQS, Outside Facility
                          3-33

-------
                      Exhibit 3-14
Annual Average PM10 Concentrations at Lafarge Alpena
4996800
4996300
499581X1
4995300
4994800
4994300
4994200
4994100
4994000
4993900
4993800
4993700
4993600
4993500
4993400
4993300
4993200
4993100
4993000
4992900
4992800
4992700
4992600
4992500
4992400
4992300
4991800
4991300
4990800
307500 308000 308500 309000 309100 309200 309300 309400 309500" 309600 30970O
017 019 021 0.25 0.26 0.27 0,28 03 0.32 0.34 0.37
018 0.2 023 028 029 03 032 033 035 037 039
019 023 027 031 033 0.34 0.36 038 04 042 045
0.19 023 029 037 0.38 04 042 045 0.47 05 053
0.19 0.23 0.29 038 041 0.43 0.46 0.49 0.53 0.58 0.63
017 022 028 039 0.42 0.45 049 053 0.58 063 0.69
017 021 028 039 042 045 049 054 059 065 071
017 021 028 039 042 0.46 05 055 0.6 067 074
0.17 021 028 039 043 0.46 0.51 0.56 0.62 069 078
016 021 028 039 0.43 0.46 051 056 063 07 0.8
016 02 027 038 042 046 0.5 0.55 062 069 0.79
016 0.2 0.26 0.37 04 044 0.48 053 059 0.67 0.76
015 019 024 034 037 041 045 05 056 064 073
014 017 0.23 032 035 0.38 042 047 053 06 07
013 016 021 03 032 0.36 0.39 0.44 05 0.57 067
013 016 02 028 031 034 037 042 047 054 063
012 015 019 027 029 032 035 039 044 0.5 058
012 014 0.18 026 028 031 034 037 041 046 0.51
Oil 014 0.18 025 027 029 032 035 038 041 046
Oil 013 017 024 026 0.28 0.3 0.32 0.35 0.38 0.42
ON 013 017 023 025 026 028 03 032 035 0.38
O.I 013 017 022 023 024 026 0.28 0.3 0.33 036
O.I 013 017 021 0.22 023 0.24 026 0.28 0.31 034
Oil 013 016 02 0.21 022 023 025 027 03 0.33
Oil 013 0.15 0.19 0.2 0.21 022 024 026 029 031
Oil 013 015 018 019 02 022 024 026 028 0.3
009 O.I 0.13 016 0.17 018 02 021 0.22 024 025
008 O.I 012 015 016 017 018 019 02 021 022
0.08 009 0.11 0.14 0.15 0.16 017 017 018 019 02
30980C
04
041
048
057
069
076
079
0.83
088
091
091
089
085
0.82
08
0.75
067
0.58
0.51
047
0.43
0.4
038
0.36
035
0.33
0.27
0.24
021
310000 310500 311000 311500 31190C
046 053 061 065 0.66
048 0.69 0.79 087 089
0.54 0.85 I.I 1.28 1 33
067 1 1.73 2,19 2.17
085 145 2.66 49 437
096 217 522 2513 1668
1 2.41 5.92 4816 2539
1.06 28 658 1553 334
2'>^^^wBBBI l802 11.67
081 8.32 2.29 194 1.47
069 303 193 164 127
062 182 16 14 II
0.56 131 131 121 0%
052 104 108 105 085
048 0.88 0.91 0.92 075
045 0.76 078 0.8 0.67
0.42 0.68 0.68 0.71 06
0.39 0.62 0.6 062 054
031 0.4 0.39 0.38 033
026 0.27 028 026 023
021 02 0.22 0.2 0.18
	 312000 312100 312200 312300 312400 3IJ500 313000 313500 314000
0.67 067 0.67 066 064 063 063 055 042
09 0.9 0.88 0.86 0 S4 (}H2 076 057 048
1.31 1.27 1 22 IIS 115 IM 085 1 68 059
206 1 94 1 86 1 77 1 66 1 53 1 12 U9 072
396 356 315 283 262 241 1 (, 15 088
12.47 9.29 716 575 474 398 203 29 092
15.8 10.93 8.11 6.29 5.06 418 209 31 095
18.41 12.07 869 664 529 435 2.15 36 097
17.78 11.81 857 6.58 5.26 4.34 216 .37 0.98
14.37 10.54 7.95 621 502 4.17 213 .37 099
10.09 8.45 6.96 5.71 4.73 398 208 (5 09')
6.69 6.23 56 4.94 429 372 203 .33 098
445 4.41 4.26 4 368 334 97 31 097
3.16 3.16 3.18 312 301 285 89 29 096
2.4 2.39 241 2.43 2.41 2.36 .79 26 094
1.93 1.89 1.9 1.93 1.95 194 65 22 0.93
1.62 1.56 1 55 1.57 16 1 61 49 17 091
14 1.33 1.3 1.31 1.33 1 35 13 II 089
1.22 1.16 1.12 1.11 1.12 1 15 17 .05 086
1.07 1.03 0.98 096 096 0.98 .03 097 082
094 0.91 0.88 0.85 0.84 085 091 089 079
0.83 0.81 0.79 076 074 074 0.81 081 075
0.74 0.73 071 0.69 0.67 0.66 0.72 074 071
0.66 0.66 0.64 063 0.61 06 065 0.67 066
0.6 0.59 0.59 057 056 0.54 058 0.61 061
0.54 0.54 0.53 0.53 051 05 052 0.56 0.57
0.33 033 0.34 034 034 034 033 0.37 04
0.23 0.22 0.23 0.23 024 024 0.24 025 028
0.17 017 017 0.17 017 017 0.18 018 02
                                                                               Source

                                                                               Above NAAQS. Outside Kacilily
                           3-34

-------
                       Exhibit 3-15
24-hour Average PM2 5 Concentrations at Lafarge Alpena
307500
4996800 123
4996300 181
4995800 229
4995300 537
4994800 11.75
4994300 10.18
4994200 911
4994100 802
4994000 688
4993900 5.81
4993800 475
4993700 3.74
4993600 284
4993500 2.06
4993400 144
4993300 107
4993200 1.24
4993100 1.38
4993000 145
4992900 1.45
4992800 1.38
4992700 124
4992600 107
4992500 088
4992400 0.97
4992300 1.14
4991800 1 15
4991300 0.61
4990800 0.64
308000
1.3
15
23
372
11.5
II 48
1041
92
8.03
685
5.76
469
365
2.68
1 87
14
1.55
1.62
1 58
146
1 27
104
1 18
1 39
1.54
1.57
1 02
073
0.86
308500 309000
162 199
1.45 19
209 176
2.64 2.86
1061 8.81
1292 1441
1182 1353
1053 12.21
916 1057
7.89 889
6.78 751
5.76 661
472 591
3.63 4.98
255 368
179 226
1.83 1.93
1.74 1.64
1.54 207
1.26 236
15 2.39
1.76 2.43
188 23
185 203
182 169
1.7 1.33
0.85 12
1 1 97
1.52 268
309100
196
2.06
1.72
2.88
833
147
13.88
12.59
10.92
9.13
7.64
67
6.09
531
399
2.48
1.9
1.96
239
256
2.59
2.54
2.3
1.96
1 56
1.17
133
2.28
2.98
309200
1.9
2.22
1.79
2.88
78
1499
1422
12.98
11.31
9.41
775
6.7
624
563
436
2.67
1.86
2.34
269
2.7
2.76
2.59
226
1 83
1.4
1 22
1 54
262
3.28
309300 309400
179 167
2.34 242
1.9 207
2.86 282
728 6.68
15.28 1559
14.54 1484
13.38 1374
11.75 1222
973 1013
787 8.03
67 669
635 6.35
596 62
4.78 5.23
299 3.28
22 269
2.75 315
2.95 3.13.
296 3.18
2.89 295
259 253
2 16 2
167 1.46
1.28 1.4
1.33 1.43
1.9 2.32
3 3.4
3.58 3.86
309500
1 53
241
228
2.73
6 13
15.92
15 15
14 1
12.69
10.59
8.22
66
6.28
6.44
574
366
324
3.51
3.43
333
296
2.41
1.79
1.48
1.51
1.64
282
3.81
409
309600
163
235
2.52
261
5.56
1632
1545
1442
13.2
11.12
85
656
609
6.5
629
4 15
3.81
3.78
3.7
341
2.88
2.21
1.6
163
1 78
1.89
3.37
421
425
309700
1 75
2.22
275
246
498
1683
1581
1468
1365
11 76
886
653
5.87
651
681
4.79
433
405
389
342
2.73
1.92
1 81
1 94
207
2 18
397
456
4.31
309800
1.81
2.05
2.94
2.57
438
17.47
16.27
1494
14.08
1246
9.33
6.56
562
621
7.28
5.67
4.77
4.39
4
3.35
2.47
209
2 14
2.29
2.4
2.91
4.56
4.82
426
310000 310500 311000 311500 311900 312000 312100 312200 312300 312400 312500 313000
198 233 3.88 7.26 891 933 9.63 954 901 806 878 1284
221 278 4.7 805 1006 1051 1042 1038 III 11.89 1273 1473
302 3.73 645 9.6 1376 1388 1428 1488 1555 1638 1697 1005
2.79 45 896 14.98 1905 1923 1969 2021 20(7 1915 II, W t S»
,4.4 4.77 9.83 23.49 27.95 2689 2404 17.98 1081 811 7.05 786
19.09 2462 1913 52.96 4457 3094 2403 19.31 1996 2023 1971 1437
17.74 272 346 91.77 57 6s| 3859 3305 3113 2839 2566 2322 1564
1555 2306 39.69 IS(M3 88 17MHB 49.53 4066 3457 3022 2693 1779
14-71 I8.I4^^^7^^BB 6395HHl 4535 -191» M.72 31.03 2811 1921
1425 20.36J^M(25T51 40.34 29.57 25.4 2519 24.72 2394 2304 1825
I0,78^^35im 7165 32.23 2906 2219 198 17 1534 1567 1524
6.7lHHH| 2526 4153 23.04 2506 2347 1898 1669 155 1384 1154
5.3I^^H 12.21 27.75 1434 1866 2055 19.7 1688 1401 1363 847
5.28^^H 12.85 20.1 8.31 1232 1588 17.55 1705 15.17 13.02 952
7.09^^^H 11.12 15.51 7.13 7.42 1097 1387 15.3 1507 1373 978
816^^H 1219 12.49 6.26 54 715 996 1233 1355 1348 919
548^|B I2J5 l033 549 4J7 503 689 9.19 1112 12.16 857
4.74 53.03 12.39 8.67 575 456 411 505 665 853 1012 906
4.09 17.57 11.78 738 6.01 423 3.72 4.15 507 645 801 91ft
2.91 9.61 9.61 633 6.12 4.08 349 37 419 504 627 9.36
2.82 6.72 796 5.48 6.11 443 334 343 3.69 42 501 898
2.92 687 694 4.77 601 469 31 321 337 367 419 828
3.06 6.97 6.06 4.18 585 485 3.29 3.02 314 3.33 365 732
3.75 6.97 5.23 3.69 565 492 3.58 288 297 309 3.3 631
4.33 6.86 451 3.27 5.42 4.94 38 2.79 284 292 3.06 538
4.78 666 4.23 291 5.18 49 397 275 276 281 29 4.6
5.5 495 2.62 2.55 4.02 4.3 413 3.54 2.71 2.03 218 2.79
492 3.33 1.76 2.39 3.11 353 371 36 32 26 194 179
3.85 2.33 1.6 216 243 2.84 3.16 33 321 292 247 117
313500 31400(
1 1 78 4 '
6 55 2 h
2 99 1 7 i
3 59 5.3'
817 86
10.69 8.5.
11.88 9 hi
13.54 IO'J<
1471 120-
1474 123',
13 54 1 1 y.1
1 15 10 H'-
9.24 9.31
72 7 7(
669 624
741 5.01
7.7 5.67
756 6.1^
711 64
6.54 6 4
6 83 6 -'
7 16 5.87
737 5.4(
7.41 5.54
725 581
6.93 607;
419 57
26 3.84;
1.53 238:
                                                                                  Source

                                                                                  Above NAAQS.OulsideFucilily j
                            3-35

-------
                                  3-36
                              Exhibit 3-16
Estimated Population Exposed Above the NAAQS at 52 Facilities Examined
ASH GROVE CEMENT CO
ASH GROVE CEMENT CO
ASH GROVE CEMENT CO
ASH GROVE CEMENT CO
ASH GROVE CEMENT CO
BLUE CIRCLE INT.
CALIF PORTLAND CEMENT
CAPITOL AGGREGATES INC
CONTINENTAL CEMENT CO INC
DACOTAH CEMENT
ESSROC MATERIALS
ESSROC MATERIALS
GIANT CEMENT COMPANY
HEARTLAND CEMENT COMPANY
HOLNAM INC
HOLNAM INC
HOLNAM INC
HOLNAM INC
HOLNAM INC
HOLNAM INC
HOLNAM INC
HOLNAM INC
HOLNAM INC
INDEPENDENT CEMENT CORP
INDEPENDENT CEMENT CORP
LAFARGE CORPORATION
LAFARGE CORPORATION
LAFARGE CORPORATION
LAFARGE CORPORATION
LAFARGE CORPORATION
LAFARGE CORPORATION
LAFARGE CORPORATION
LEHIGH PORTLAND CEMENT CO
LEHIGH PORTLAND CEMENT CO
LEHIGH PORTLAND CEMENT CO
LEHIGH PORTLAND CEMENT CO
LONE STAR INDUSTRIES
LONE STAR INDUSTRIES
LONE STAR INDUSTRIES
LONE STAR INDUSTRIES
MEDUSSA CEMENT COMPANY
MEDUSSA CEMENT COMPANY
NATIONAL CEMENT CO OF CALIFORNIA
RINKER PORTLAND CEMENT CORP
RIVER CEMENT COMPANY
ROANOKE CEMENT COMPANY
SIGNAL MOUNTAIN CEMENT COMPANY
SOUTHDOWN
SOUTHDOWN
SOUTHDOWN
TARMAC FLORIDA
TEXAS INDUSTRIES
TOTAL
a Although these facilities have people
CHANUTE
FOREMAN
INKOM
LOUISVILLE
MONTANA CITY
RAVENA
MOJAVE
SAN ANTONIO
HANNIBAL
RAPID CITY
LOGANS PORT
SPEED
HARLEYVILLE
INDEPENDENCE
ADA
ARTESIA
CLARKSVILLE
MORGAN
DUNDEE
LAPORTE
HOLLY HILL
FLORENCE
TIJERAS
CATSKILL
HAGERSTOWN
APENA
NEW BRAUNFELS
BUFFALO
FREDONIA
GRAND CHAIN
PAULDING
SUGAR CREEK
CEMENTON
MASON CITY
MITCHELL
UNION BRIDGE
CAPE GIRARDEAU
GREENCASTLE
NAZARETH
OGLESBY
CHARLEVOIX
DEMOPOLIS
LEBEC
MIAMI
FESTUS
CLOVERDALE
CHATTANOOGA
KNOXVILLE
LYONS
ODESSA
MEDLEY
MIDLOTHIAN

living within 500
exposed above the NAAQS is assumed to be zero because the
b Although the NAAQS is predicted to be
property line at these three facilities
exceeded out to a
KS
AR
ID
NE
MT
NY
CA
TX
MO
SD
IN
IN
SC
KS
OK
MS
MO
UT
MI
CO
SC .
CO
NM
NY.
MD
MI
TX
IA
KS
IL
OH
MO
NY
IA
IN
MD
MO
IN
PA
IL
MI
AL
CA
FL
MO
VA
TN
TN
CO
TX
FL
TX

meters ,
CKD is
certain
0
0
364
0
0
Oa
0
0
2
5
0
0
0
0
Ob
0
0
0
0
0
0
0
oa
0
0
ob
1
5
oa
0
0
1
0
0
0
7
0
0
7
0
0
0
0
0
0
0
ob
236
6
0
0
0
634
the population
wetted or watered.
distance from the
, no people live in the affected area.
                     Draft, August 1997, Do Not Cite or Quote

-------
                                              3-37

       3.4     MAJOR LIMITATIONS AND UNCERTAINTIES

       This study has significantly enhanced EPA's understanding of the extent to which populations
living near cement plants are potentially at risk due to exposures to airborne paniculate matter from
CKD waste management activities.  The Agency recognizes, however, several limitations and
uncertainties inherent in the analysis.  Limitations and uncertainties associated with the PM exposures
analysis include those related to the emissions and dispersion modeling, and those related to assessing
population exposures across the facilities.

•      All emission and dispersion/air quality modeling applications are limited by the
       accuracy of the input data and the inherent limitations of the specific models used.
       While site-specific  data were used to the extent  possible to develop emission and air
       quality modeling inputs, data were not available for several inputs and, thus,
       assumptions were developed as needed.  Many of the AP-42 equations used in this
       analysis were developed originally based on empirical data, collected from industries
       using soil, gravel or other material (coal) all of which have properties different from
       CKD.   Since no on-site emission testing was available for CKD facilities, use of these
       AP-42 equations was necessary for this analysis. Using equations to model CKD
       emissions that  are not developed using empirical CKD data (or are at the equation
       limit) can introduce error into the  present analysis.  For example, the value for
       moisture content of the CKD during handling and at the pile used in the present
       analysis is the  lower limit allowed by the equation  rather than a value based on
       empirical analysis of actual CKD.  For more realistic emissions modeling, equations
       and parameters that are based on studies of the actual behavior of CKD are required.

•      Emissions controls for CKD (e.g., pelletized prior to transport) can greatly reduce emissions.
       After consulting the information on hand (i.e., site  visit trip reports, PCA surveys, the NODA),
       EPA determined that there was no facility-specific  information provided on how the CKD
       emissions are controlled between collection at the facility and disposal at the pile for the seven
       high-risk facilities modeled.  Thus, EPA developed a general handling train scenario (and
       associated emissions controls) that was used at all seven facilities.  Where this generic scenario
       does not match the actual CKD handling (and emissions controls) of the actual facility, the
       present analysis will not reflect the actual conditions.

       EPA used the "fastest mile" data from "Extreme Wind Speed at 129 Stations  in the
       Contiguous United States" to calculate emissions for  this analysis.  This wind speed represents
       the mean annual fastest mile.  Analysis of historical meteorological data to determine mean
       daily fastest mile values for each facility and the subsequent use of these values to estimate
       emissions were beyond the scope  of this effort.  Consequently, the emission estimates prepared
       for this analysis were calculated by assuming that the mean annual fastest mile occurs between
       every disturbance (i.e., as frequently a^ once per day) instead of once per year, and thus may
       overstate actual emission  rates.

•      The key assumptions underlying the estimation  of the number of people exposed to
       PM concentrations are (i) individuals are always located at their residence; (ii) outdoor
       concentrations at the block centroid can adequately represent outdoor concentrations
                             *** Draft, August 1997, Do Not Cite or Quote ***

-------
                                       3-38

throughout the block; and (iii) outdoor concentrations can adequately represent
concentrations in all microenvironments. The first assumption may lead to over- or
underestimates of population exposure, depending on whether the number of
individuals typically located in each block is greater or less than the residential
population.  For example, an industrial or commercial  area is  likely to have more
people present on average than is indicated by the residential population.  The second
assumption similarly may lead to either over- or underestimates of population exposure
depending on the spatial distribution of the population  throughout the block, and the
steepness of concentration gradients.  The smaller the spatial extent of the block, the
less potential bias is introduced by this assumption.  Because  census subdivisions are
designed to have approximately equal populations, blocks in urban areas tend to be
significantly smaller than those in rural  areas. Because concentration gradients tend to
be steepest at locations closest to the emission source,  the uncertainty introduced by
this assumption is highest near the facility boundary. The third assumption is likely to
lead to overestimates of  the exposure concentration increment, since indoor
concentrations of outdoor generated particulate matter  are generally lower than
concurrent outdoor concentrations, and individuals typically spend more than 80
percent of their time indoors.  The amount of protection afforded by being indoors
depends  on a number of factors, including the air exchange rate, and the presence or
absence of air filtration equipment.

Extrapolation of the  air quality modeling results completed for the Holnam Ada, Signal
Mountain Chattanooga, and Lafarge Alpena facilities to other facilities involves some
uncertainty because the physical characteristics that exist at these three facilities will  not be
replicated exactly at other cement facilities.   Although  these physical characteristics
(meteorology,  terrain, physical layout of sources) were examined at a basic level to determine
which were the most significant in driving predicted concentrations, extensive sensitivity
analysis was not conducted to fully explore  the effect of such variations at other facilities.

Extrapolation of the  air quality modeling results to receptors located in a complete circle
surrounding the facility is a conservative assumption since the maximum modeled
concentrations at the Holnam Ada, Signal Mountain Chattanooga, and Lafarge Alpena facilities
occurred only at receptors located downwind of the  CKD sources.  Concentrations predicted at
receptors located upwind of the sources were considerably lower.  If meteorological conditions
at a given facility were to vary enough to produce concentrations at equal levels at all
directions surrounding the facility, the final  predicted concentrations at any given receptor
point would be lower compared to that predicted at  the three facilities.

While EPA has established that coarse and fine airborne PM can increase respiratory
symptoms and impair breathing, leading to adverse health effects, there are no widely
accepted dose-response or toxicity measures for PM exposures. Thus, for this PM
analysis, EPA could not  describe the population risk in conventional terms of number
of excess disease cases, and had to, instead, rely  on comparing predicted  airborne
concentrations to air quality standards.  Because the characteristics of the dose-
                      ***  Draft, August 1997, Do Not Cite or Quote ***

-------
                                        3-39

response relationship are not well understood as yet, it is possible that some or all of
the people predicted to be exposed to PM levels exceeding the NAAQS will not
exhibit any adverse effects.

Assuming  that emissions are completely eliminated (100 percent effectiveness) at
facilities that report watering or wetting their waste piles involves significant
uncertainty because there may be portions of the pile or the handling train that still can
be sources of PM releases. Use of this assumption will tend to underestimate the
population effects.

There is significant uncertainty associated with predicting the number of people who
live within 100 or 200 meters  of the facility or waste pile boundary. This is because
the finest resolution of Census data available from electronic databases is at the block
level, and  the method used for predicting the populations is the "buffer method."  This
method calculates the population within a circle (defined by a given radius) around the
facility by tabulating the total  populations of Census blocks whose centroids fall  within
that circle. As the circle gets  smaller, it is less probable that  the centroid of a given
block will fall within the circle, although part of the block would still be in the circle.
                          Draft, August 1997, Do Not Cite or Quote ***

-------
                  Appendix A



Exposure Parameter Values Used For CKD Risk Assessment

-------
EXPOSURE PARAMETER VALUES USED FOR CKD RISK ASSESSMENT
Exposure Pathway1
Ground Water
Ingestion of Contaminated
Drinking Water
Inhalation of Volatile
Contaminants from Indoor Air
Dermal Absorption of
Contaminants from Water while
Showering4
Air
Inhalation of Contaminants
from Air
Surface Water5
Dermal Absorption of
Contaminated Surface Water
while Swimming4
Ingestion of Contaminated
Surface Water while Swimming
Soil
Ingestion of Contaminated Soil
(Adult)
Exposure Rate

1.4 liters/day
0.63 m3/hour
(15 m3/day)
-

0.83 m3/hour
(20 m3/day)
(avg.)3

"
0.05
liters/hour3

100 mg/day3
Exposure Time2

-
•
0.116 hours/day
(7 minutes)
(50%)3iC

24 hours/day

2.6 hours/day
(hours/event)
(avg.)a
2.6 hours/day
(hours/event)
(avg.)3

-
Frequency of
Exposure

350 days/year6
350 days/year"
350 days/year11

350 days/year"

26 days/year
(avg.)3'6
26 days/year
(avg.)3'6

350 days/year11
Exposure
Exposure Duration:
Duration: Non-
Carcinogen carcinogen

9 years 9 years3
(50th%)a
9 years 9 years3
(50th%)3
9 years 9 years3
(50th%)3

9 years 9 years3
(50th%)3

9 years 9 years3
(50th%)3
9 years 9 years3
(50th%)3

9 years 9 years3
(50th%)3
Body
Weight

70kga
(avg.)
70kg3
(avg.)
70kg3
(avg.)

70kga
(avg.)

70kg3
(avg.)
70 kg3
(avg.)

70kga
(avg.)
Averaging
Time:
Carcinogen

25,560 days
(70 years)
25,560 days
25,560 days

25,560 days

25,560 days
25,560 days

25,560 days
Averaging
Time:
Non-
carcinogen

3,285
days3
(9 years)
3,285 days
3,285 days

3,285 days

3,285 days
3,285 days

3,285 days
                     Page A - 2

-------
EXPOSURE PARAMETER VALUES USED FOR CKD RISK ASSESSMENT
Exposure Pathway1
Ingest ion of Contaminated
Soil (Child)
Ingestion of Contaminated
Soil (Pica Child)
Dermal Absorption from Soil
(Adult)7
Dermal Absorption from Soil
. (Child)7
Foodchain
Ingestion of Contaminated
Homegrown Root Vegetables
Ingestion of Contaminated
I lomegrown Leaf Vegetables
Ingestion of Contaminated
Homegrown Beef
Ingestion of Contaminated Dairy
Products
Ingestion of Contaminated
Recreationally Caught Fish
Subsistence Fanner
Ingestion of Contaminated
Homegrown Root Vegetables:
Subsistence Farmer
Ingestion of Contaminated Home-
grown Leaf Vegetables:
Subsistence Farmer
Exposure Rate Exposure Time2
200 mg/day
(avg.)3-0
800 mg/day
(high end)0
-
-

46 g/day
(avg.f8
65 g/day
(avg.f8
44 g/day
(avg.)0-9
160 g/day
(avg.)0-10
7.6 g/day
(50th%)3-e-u

74 g/dayc'8
103 g/day
(avg.)0'8
Frequency of
Exposure
350 days/year11
365 days/year6
350 days^earb
350 days/year6

350 daystyear6
350 days/year11
350 daystyearb
350 days/year6
350 days/year11

365 days/year3
365 daystyear"
Exposure
Duration:
Carcinogen
5 years0
5 years0
9 years
(50th%)a
5 years0

9 years
(50th%)'
9 years
(50th%)a
9 years
(50th%)a
9 years
(50th%)a
9 years
(50th%)"

40 yearsf
40 years'
Exposure
Duration:
Non-
carcinogen
5 years0
5 years0
9 years3
5 years0

9 years8
9 years3
9 years3
9 years3
9 years3

40 years'
40 years'
Body
Weight
16 kg0
(50th%)
16 kg0
(50th%)
70kg3
(avg.)
16 kg0
(50th%)

70kg3
(avg.)
70 kg3
(avg.)
70kga
(avg.)
70kga
(avg.)
70kg3
(avg.)

70kga
(avg.)
70kg3
(avg.)
Averaging
Time:
Carcinogen
25,560 days
25,560 days
25,560 days
25,560 days

25,560 days
25,560 days
25,560 days
25,560 days
25,560 days

25,560 days
25,560 days
Averaging
Time:
Non-
carcinogen
1,825 days
(5 years)
1,825 days
3,285 days
1,825 days

3,285 days
3,285 days
3,285 days
3,285 days
3,285 days

14,600
days
14,600
days
                     Page A - 3

-------
                                       EXPOSURE PARAMETER VALUES USED FOR CKD RISK ASSESSMENT



Exposure Pathway'
Ingestion of Contaminated
Homegrown Beef:
Subsistence Farmer
Ingestion of Contaminated Dairy
Products:
Subsistence Farmer
Subsistence Fisherman
Ingestion of Contaminated
Recreationally Caught Fish:
Subsistence Fisherman



Exposure Rate Exposure Time1
75 g/day0'9


300 g/day
(95th%)C|1°


99 g/day
(95th%fe'11



Frequency of
Exposure
365 days/year3


365 days/year3



365 days/year3


Exposure
Exposure Duration:
Duration: Non-
Carcinogen carcinogen
40 yearsf 40 years'


40 yearsf 40 yearsf



30 years 30 years"
(90th%)a



Body
Weight
70kga
(avg-)

70kg'
(avg-)


70kga
(avg-)

Averaging
Averaging Time:
Time: Non-
Carcinogen carcinogen
25,560 days 14,600
days

25,560 days 14,600
days


25,560 days 10,950
days

SOURCES AND NOTES: (Note that the sources listed below may in turn refer to secondary documents as the original source for some of the parameter values.)

(a)      USEPA 1989.  Risk Assessment Guidance for Superfund Volume I:  Human Health Evaluation Manual (Part A1.  Office of Emergency and Remedial
         Response.  EPA/540/1-89/002.

(b)      USEPA 1991.  Human Health Evaluation Manual, Supplemental Guidance: Standard Default Exposure Factors.  Office of Emergency and Remedial
         Response.  OSWER Directive: 9285.6-03.

(c)      USEPA 1989.  Exposure Factors Handbook. Office of Health and Environmental Assessment.  EPA/600/8-89/043.

(d)      USEPA 1991.  Interim Guidance for Dermal Exposure Assessment.  Office of Research and Development.  EPA/600/8-91/011A.

(e)      USEPA 1991.  Internal EPA Memorandum dated August 19, 1991.

(f)       USEPA 1992.  Internal Communication from Office of Research and Development, undated.

(g)      USEPA 1992.  Dermal Exposure Assessment: Principles and Applications. Office of Health and Environmental Assessment. EPA-600/8-91/011B.

1.        Intake via each exposure pathway is calculated using the general  equation described in Section E.2.4.

2.       The general equation is modified for the inhalation and dermal exposure pathways to include the "Exposure Time."  This parameter (in hours/day) is included
        as a multiplicand in the numerator of the equation.

3.       The averaging time to calculate noncancer hazard is equal to the exposure duration expressed in days.
                                                                    Page A - 4

-------
                                        EXPOSURE PARAMETER VALUES USED FOR CKD RISK ASSESSMENT

4.       The general equation is modified to calculate intake via dermal absorption in the water pathway:

         Intake   =       CW x CF x SA x PC x ET x EF x HP , where factors unique to this pathway include:
                                    BWx  AT

         CF      =       Conversion factor (1 liter/lOOOcm3)
         SA      =       Skin surface area available for contact (cm2/day)
                          (19,400 cm2 for adults (central tendency, source (g)))
         PC      =       Chemical-specific dermal permeability constant
         ET      =       Exposure time (hours/day)

5.       Exposure from surface water as a municipal water supply is evaluated using the same exposure  pathways, routes, and parameter values as for ground water.

6.       The exposure frequency value for dermal absorption and ingestion of contaminated surface water while swimming was provided by EPA's Office of Research
         and  Development.  This assumes that an individual swims 2 times a week during  the summer (3 months) only.

7.       The general equation is modified to calculate intake via dermal absorption in the soil pathway:

         Intake   =       CS x CF x SA x  AF x ABS x EF x ED , where factors unique to  this pathway include:
                                     BWx AT

         CF      =       Conversion factor (10"* mg/kg)
         SA      =       Skin surface area available for contact (cm2/day)
                          (5,000 cm2 for adults (central tendency, source (g)), and 2,500 cm2 for children (default value, source (d)))
         AF      =       Soil-to-skin adherence factor (mg/cm2)
                          (0.2 mg/cm2 (central tendency, source (g)))
         ABS    =       Absorption factor (chemical specific constant)

8.       The general equation is modified for the foodchain pathways to include "fraction  from contaminated source." This parameter is included as a multiplicand in
         the numerator of the equation, and  modifies the exposure rate. Thus, the exposure rate accounts for the fraction of total vegetables consumed that comes
         from the contaminated source.  The proportion of contaminated vegetables is assumed to be 25 percent for the general  population (average, source (c)) and
         40 percent for subsistence farmers (reasonable worst case, source (c)).

9.       Exposure rate accounts for the fraction of total beef consumed that comes from the contaminated source. The proportion of contaminated beef is assumed to
         be 44 percent for the general population (average, source (c))  and  75 percent for subsistence farmers (reasonable worst  case, source (c)).

10.       Exposure rate accounts for the fraction of total dairy products  consumed that comes from the contaminated source.  The proportion of contaminated dairy
         products is assumed to be 40 percent for the general population (average, source (c)) and 75 percent for subsistence farmers (reasonable worst case, source
         (c))-

11.       Exposure rate accounts for the fraction of total fish consumed  that comes from the contaminated source. The proportion of contaminated fish is assumed to
         be 20 percent for the general population (average, source (c))  and  75 percent for subsistence fishermen (reasonable worst case, source (c)).



                                                                      Page A. - 5

-------
             Appendix B



Individual Risk Estimates from RTC and NODA

-------
INDIVIDUAL CANCER RISK ESTIMATES
Plant ID
4
7
8
9
11
15
18
19 I
22
23
25
29
30
32
33
34
35
36
' 37
42
43
44
46
49
50
51
52
53
54
55
, 57
| 60
61
62
63
64
66
67
68
70
72
74
75
76
77
80
81
83
Population
within 5 miles
10,705
1,922
2,140
2,707
425
15,781
191,915
14,629
59,376
17,763
20,812
12,518
20,223
948
965
7,534
9,396
4,198
4,647
3,240
10,627
65,458
1,975
15,559
14,559
2.069
3,254
24,553
6,275
72,527
8,572
29,085
10,583
9,433
37,469
14,752
43,851
17,407
4,202
8,571
584
145,192
15,782
3,494
49,602
55,918
5,437
10,136
Exposed Populations
Resident Farmer
Beef and Milk
Ingestion
9.50E-09










1.00E-06
1.00E-08

9.90E-06
1.00E-08





1.00E-06




1.30E-07

1 .OOE-08
1.00E-06

9.90E-06
1.00E-06
1.00E-06


1.00E-06



9.90E-06
4.00E-10




2.60E-09

Recreational
Fisher
1.20E-08
1. OOE-08
1.00E-09

1.00E-09
1.00E-11

1.00E-12
1.00E-09
1.00E-12
1.00E-09
1.00E-06


1.00E-09


1.00E-10
9.90E-06

1.00E-13
1.00E-10

9.90E-06
1.00E-12
1.00E-12
2.20E-06



1.00E-11

1. OOE-08
1.00E-05





1.00E-13


1.00E-07
1.00E-10
1.00E-07

3.80E-06

Subsistence
Fisher
5.50E-07
1.00E-06
1.00E-07

1.00E-07
1.00E-09

1.00E-10
9.90E-06
1.00E-10
1. OOE-08
9.90E-06

1.00E-12
1. OOE-08

9.90E-04
1.00E-09
9.90E-04

1.00E-11
1.00E-09

9.90E-06
1.00E-11
1.00E-10
1.00E-04



1 .OOE-09
1. OOE-08
1 .OOE-06
9.90E-04




9.90E-06
1.00E-11


1. OOE-06
1. OOE-08
1. OOE-06
1. OOE-09
2.00E-04

Homegrown
Population
1.70E-06
9.90E-05
1.00E-07
1.00E-07
1. OOE-06
1. OOE-06
1.00E-07
9.90E-06
1. OOE-06
1.00E-07
9.90E-06
9.90E-06
1. OOE-06
1.00E-07
9.90E-05
1.00E-07
1. OOE-06


9.90E-06

1. OOE-06
9.90E-06
9.90E-06
1.00E-07

1.20E-06
1. OOE-06
9.90E-06
9.90E-06
9.90E-06
9.90E-06
1.00E-07
9.90E-06
9.90E-06
1.00E-07
9.90E-06
1. OOE-06
1. OOE-08
1.00E-07
9.90E-05
1.80E-08
1.00E-07

1.00E-07
1. OOE-06
9.10E-07
1. OOE-06
Subsistence
Farmer
1.30E-05











9.00E-06

9.90E-05
1. OOE-06





9.90E-06



9.90E-06
1 .OOE-05

9.90E-05
9.90E-05

9.90E-06
9.90E-06
9.90E-05
9.90E-06

9.90E-05
9.90E-06


9.90E-04
1 .40E-07




6.60E-06

            Page B- 2

-------
INDIVIDUAL NONCANCER EFFECTS ESTIMATES
Plant ID
4
7
8
9
11
15
18
19
22
23
25
29
30
32
33
34
35
36
37
42
43
44
46
49
50
51
52
53
54
55
57
60
61
62
63
64
66
67
68
70
72
74
75
76
77
80
81
82
83 I
Population
within 5 miles
10,705
1,922
2,140
2,707
425
15,781
191,915
14,629
59,376
17,763
20,812
12,518
20,223
948
965
7,534
9,396
4,198
4,647
3,240
10,627
65,458
1,975
15,559
14,559
2,069
3,254
24,553
6,275
72,527
8,572
29,085
10,583
9,433
37,469
14,752
43,851
17,407
4,202
8,571
584
145,192
15,782
3,494
49,602
55,918
5,437
131,714
10,136
Exposed Populations
Resident Farmer
Beef and Milk
Ingestion
<1













<1
<1





<1



<1
<1

<1
<1

<1
<1
<1
<1

<1
<1


<1
<1




<1


Recreational
Fisher
<1
<1
<1

<1
<1

<1

<1

<1

<1
<1

9.90E+01
<1
9.90E+00


<1

<1
<1
<1
<1



<1
<1
<1
9.90E+01


<1
<1
<1
<1


<1
<1
<1
<1
4.10E-I-00


Subsistence
Fisher
2.50E+00
<1
<1

<1
<1

<1
<1
<1
<1
<1

<1
<1

9.99E+02
<1
9.90E+01


<1

<1
<1
<1
<1


<1
<1
9.90E+00
9.90E+00
9.99E+02
<1

9.90E+00
<1
<1
<1


<1
<1
<1
<1
5.30E+01


Homegrown
Population
<1
<1
<1
<1
<1

<1
<1
<1
<1
<1


<1
<1
<1
<1


<1

<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
9.90E+00

<1
<1
<1
<1
<1

<1
<1
<1
<1


<1
<1
<1
<1
Subsistence
Farmer
<1










9.90E+00
9.90E+00

<1
<1





<1



<1
<1

<1
9.90E+00

9.90E+00
<1
9.90E+00
<1

9.90E+00
<1


9.90E+00
<1




<1


               Page B- 3

-------
          Appendix C



Results of the Tier 1 Screening Analysis

-------
                         POPULATION CANCER RISKS, BY FACILITY
                                  VEGETABLE INGESTION
Population Population Risk for Cumulative Population Population Risk for Cumulative 1
Plant ID within 5 miles "Homegrown" Percent Plant ID within 5 miles Subsistence Farmer Percent \
55
66
63
60
25
7
49
29
33
62
57
44
54
22
72
80
42
53
30
46
18
4
67
15
83
35
77
81
52
74
23
75
64
50
61
70
34
11
9
8
32
68
42 Facilities
72,527
43,851
37,469
29,085
20.812
1,922
15,559
12,518
965
9,433
8,572
65,458
6,275
59,376
584
55,918
3,240
24,553
20,223
1,975
191,915
10,705
17,407
15,781
10,136
9,396
49,602
5,437
3,254
145,192
17,763
15,782
14,752
14,559
10,583
8,571
7,534
425
2,707
2,140
948
4,202
1,049,106
7.18E-01
4.34E-01
3.71 E-01
2.88E-01
2.06E-01
1.90E-01
1.54E-01
1.24E-01
9.55E-02
9.34E-02
8.49E-02
6.55E-02
6.21 E-02
5.94E-02
5.78E-02
5.59E-02
3.21 E-02
2.46E-02
2.02E-02
1.96E-02
1.92E-02
1.82E-02
1.74E-02
1.58E-02
1.01 E-02
9.40E-03
4.96E-03
4.95E-03
3.90E-03
2.61 E-03
1 .78E-03
1 .58E-03
1 .48E-03
1 .46E-03
1.06E-03
8.57E-04
7.53E-04
4.25E-04
2.71 E-04
2.14E-04
9.48E-05
4.20E-05
3.27E+QO
21.94%
35.20%
46.54%
55.34%
61.63%
67.45%
72.15%
75.94%
78.86%
81.71%
84.31%
86.31%
88.20%
90.02%
91.78%
93.49%
94.47%
95.22%
95.84%
96.44%
97.03%
97.58%
98.11%
98.60%
98.91%
99.19%
99.34%
99.50%
99.61%
99.69%
99.75%
99.80%
99.84%
99.89%
99.92%
99.95%
99.97%
99.98%
99.99%
100.00%
100.00%
100.00%

55
66
62
44
54
72
63
60
30
67
4
61
33
81
52
51
74
34
























18 Facilities
72,527
43,851
9,433
65,458
6,275
584
37,469
29,085
20,223
17,407
10,705
10,583
965
5,437
3,254
2,069
145,192
7,534
























488,051
7.18E+00
4.34E+00
9.34E-01
6.48E-01
6.21 E-01
5.78E-01
3.71 E-01
2.88E-01
1.82E-01
1 .72E-01
1.39E-01
1 .05E-01
9.55E-02
3.59E-02
3.25E-02
2.05E-02
2.03E-02
7.53E-03
























1 .58E+01
45.52%
73.05%
78.97%
83.08%
87.02%
90:68%
93.04%
94.86%
96.01%
97.11%
97.99%
98.65%
99.26%
99.49%
99.69%
99.82%
99.95%
100.00%

























(Note that the "subsistence farmer" is exposed via consumption of vegetables and beef and milk.)
                                         Page C- 2

-------
                    POPULATION NONCANCER EFFECTS, BY FACILITY
                                VEGETABLE INGESTION
Population Population Risk for Cumulative Population Population Risk for Cumulative ' |
Plant ID within 5 miles "Homegrown" Percent Plant ID within 5 miles Subsistence Farmer Percent
60






1 Facilities
29,085






29,085
2.91E+04






2.91 E+04
100.00%







55
66
60
30
29
62
72
7 Facilities
72,527
43,851
29,085
20,223
12,518
9,433
584
188,221
7.25E+04
4.39E+04
2.91E+04
2.02E+04
1.25E+04
9.43E+03
5.84E+02
1 .88E+05
38.53%
61.83%
77.28%
88.03%
94.68%
99.69%
100.00%

(Note that the "subsistence farmer" is exposed via consumption of vegetables and beef and milk.)
                                       Page C - 3

-------
POPULATION CANCER RISKS, BY FACILITY
      BEEF AND MILK INGESTION
Population Population Risk for Cumulative
Plant ID within 5 miles Resident Farmer Percent
60
55
44
66
29
61
33
62
72
52
30
4
34
54
74
81
16 Facilities
29,085
72,527
65,458
43,851
12,518
10,583
965
9,433
584
3,254
20,223
10,705
7,534
6,275
145,192
5,437
443,624
2.88E-01
7.25E-02
6.55E-02
4.39E-02
1.25E-02
1 .06E-02
9.55E-03
9.43E-03
5.78E-03
4.23E-04
2.02E-04
1.02E-04
7.53E-05
6.28E-05
5.81 E-05
1.41E-05
5.19E-01
55.52%
69.51%
82.13%
90.59%
93.00%
95.04%
96.89%
98.70%
99.82%
99.90%
99.94%
99.96%
99.97%
99.99%
100.00%
100.00%

              Page C- 4

-------
POPULATION NONCANCER EFFECTS, BY FACILITY
             FISH INGESTION
Population Population Risk for Cumulative
Plant ID within 5 miles Recreational Fisher Percent
62
35
81
37
4 Facilities
9,433
9,396
5.437
4,647
28,913
9.43E+03
9.40E+03
5. 44 E +03
4.65E+03
2.89E+04
32.63%
65.12%
83.93%
100.00%

                 Page C- 5

-------
POPULATION CANCER RISKS, BY FACILITY
           FISH INGESTION
Population Population Risk for Cumulative
Plant ID within 5 miles Recreational Fisher Percent
49
62
37
81
29
52
77
75
4
61
22
25
7
44
8
33
11
36
76
15
57
23
19
50
51
43
70
27 Facilities
15,559
9,433
4,647
5,437
12,518
3,254
49,602
15,782
10,705
10,583
59,376
20,812
1,922
65,458
2,140
965
425
4,198
3,494
15,781
8,572
17,763
14,629
14,559
2,069
10,627
8,571
388,881
1 .54E-01
9.43E-02
4.60E-02
2.07E-02
1.25E-02
7.16E-03
4.96E-03
1.58E-03
1.28E-04
1.06E-04
5.94E-05
2.08E-05
1.92E-05
6.55E-06
2.14E-06
9.65E-07
4.25E-07
4.20E-07
3.49E-07
1 .58E-07
8.57E-08
1 .78E-08
1.46E-08
1.46E-08
2.07E-09
1 .06E-09
8.57E-10
3.42E-01
45.09%
72.71%
86.18%
92.22%
95.89%
97.98%
99.44%
99.90%
99.94%
99.97%
99.99%
99.99%
100.00%
100.00%
100.00%
100.00%
100.00%
100.00%
100.00%
100.00%
100.00%
100.00%
100.00%
100.00%
100.00%
100.00%
100.00%

              Page C - 6

-------
                                        Appendix D
              Facility-Specific Raw Data and Calculations, Fish Ingestion Pathway

       This appendix outlines the raw data that EPA used in calculating the population risks due to
fish consumption.  These data were provided by relevant state agencies, and represent the best
available standing stock information for waterbodies within Five miles of each facility.

       This appendix also contains other data that the Agency used in the analysis for the fish
ingestion pathway, including the number of stream miles within five miles of each facility and the
average stream width.  EPA determined this information from topographic maps of the areas
surrounding the facilities.  Also included in this appendix are exposure assumptions, such as the
exploitation rate  of 20 percent, which is the default value suggested for use in the EPA's  Hazard
Ranking System (HRS) Final Rule, and the percent of fish  tissue that is edible (35%), derived based on
conversations with local fisheries authorities and data provided in EPA's Exposure Factor's Handbook.
The value for the pounds of fish ingested per year by a recreational fisher (5.86 pounds) is based on
the daily fish ingestion rate used in the RTC and NODA for calculating individual risks from food
chain exposure.

       This data provided in this appendix can also be found in Chapter  2 of this report, but is
outlined here in a step-by-step manner intended to provide the reader with additional detail.

-------
Facility #35
  Stream miles within a
  five-mile radius of the
  facility
  (determined from
  topographic maps)
Small streams: 2 miles
Large streams: 7.5 miles
 Average stream width
 (determined from
 topographic maps)
Small streams: 1/350 mile
Large streams: 1/35 mile
  Stream acres within five
  miles of the facility
Small streams:
2 mi * 1/350 mi * 640 acres/mi2 = 3.657 acres

Large streams:
7.5 mi * 1/35 mi * 640 acres/mi2 = 137.142 acres

total = 140.8 acres
  Raw standing stock data,
  for a sampled stream
  length of 550 ft
  (obtained directly from
  studies provided  by the
  relevant state agencies3)
Site  1:
                            Site 2:
        rainbow trout: 0.70 kg
        brown trout: 2.05 kg
        longnose dace: 1.30 kg
        longnose sucker: 2.98 kg
        white sucker: 2.45 kg
        yellow perch:  0.03 kg
        total: 9.51 kg/550 ft
                                    carp: 25.9 kg
                                    flathead minnow: 0.12 kg
                                    longnose dace: 0.32 kg
                                    creek chub: 0.14 kg
                                    longnose sucker: 4.78 kg
                                    white sucker: 4.61 kg
                                    total: 35.87 kg/550 ft
                            Site 3:
                                    carp: 1.4 kg
                                    sand shiner: 0.48 kg
                                    flathead minnow: 0.91 kg
                                    cheek chub: 0.08 kg
                                    longnose sucker: 0.10 kg
                                    white sucker: 0.95  kg
                                    plains topminnow:  0.09  kg
                                    johnny darter: 0.12 kg
                                    total: 4.13 kg/550  ft

                            Total average = 16.5 kg/550 ft
                                                Page D - 2

-------
Standing stock data
(converted to units of
Ibs/acre/year)
Exploitation rate
Pounds of fish caught
per year within the area
of influence
Pounds of fish ingested
per year per recreational
fisher
Percent of fish tissue
that is edible
Number of recreational
fishers that can be
supported by the harvest
Number of recreational
fishers that are likely to
be exposed
16.5 kg/550 ft * 1/35 ftb * 43,500 ft2/acre * 2.2046 Ibs/kg = 82.2 Ibs/acre/year
0.2
82.2 Ibs/acre/yr * 140.8 acres * 0.2 = 2,314.7 Ibs/yr
5.86
35%
2, 314.7 Ibs/yr * 0.35 * 1/5.86 Ibs/year/person = 138 people
138 people
a Standing stock data for this facility were obtained from the state Division of Wildlife and were available for
three different sites along the widest river located within five miles of the facility.
  The average stream width of the wide streams was used in this calculation  because standing stock data was
available for wide streams ©nly.

Facility #37
  Stream miles within a
  five-mile radius of the
  facility
  (determined from
  topographic maps)
15 miles
  Average stream width
  (determined from
  topographic maps)
1/350 mile
  Stream acres within five
  miles of the facility
15 mi *  1/350 mi * 640 acres/mi2 = 27 acres
  Raw standing stock data
  (obtained directly from
  studies provided by the
  relevant state agencies3)
         140.9 kg/ha (all fish)b
         42.8 kg/ha
         72.2 kg/ha
         57.5 kg/ha  (average)
         163.9 kg/ha
         92.5 kg/ha
         128.2 kg/ha (average)
Total average = 108.9 kg/ha
Site  1:
Site  2:
                             Site 3:
                                                 Page D - 3

-------
Standing stock data
(converted to units of
Ibs/acre/year)
Exploitation rate
Pounds of fish caught
per year within the area
of influence
Pounds of fish ingested
per year per recreational
fisher
Percent of fish tissue
that is edible
Number of recreational
fishers that can be
supported by the harvest
Number of recreational
fishers that are likely to
be exposed
108.9 kg/ha * 2.471 ha/ac * 2.2046 Ibs/kg = 593.2 Ibs/acre/year
0.2
593.2 Ibs/acre/yr * 27 acres * 0.2 = 3,203.3 Ibs/year
5.86
35%
3,203.3 Ibs/yr * 0.35 * 1/5.86 Ibs/year/person = 191 people
191 people
a Standing stock data for this facility were obtained from the state Department of Wildlife and Parks
b Data from this site were not broken down by species; values provided were for all fish.

Facility #62
  Stream miles within a
  five-mile radius of the
  facility
  (determined from
  topographic maps)
Small streams: 24.5 miles
Large streams: 9 miles
  Average stream width
  (determined from
  topographic maps)
Small streams: 1/350 mile
Large streams: 1/70 mile
  Stream acres within five
  miles of the facility
Small streams: 24.5 miles * 1/350 miles * 640 acres/mi2 = 44.8 acres

Large Streams: 9 miles *  1/70 miles * 640 acres/mi2  = 82.3 acres

Total = 127.1 acres
  Raw standing stock data
  (obtained directly from
  studies provided by the
  relevant state agencies3)
Site 1:  899 grams/75 m

Site 2:  5,869 grams/75 m

Total average: 3384 g/75
                                                Page D - 4

-------
Standing stock data
(converted to units of
Ibs/acre/year)
Exploitation rate
Pounds of fish caught
per year within the area
of influence
Pounds of fish ingested
per year per recreational
fisher
Percent of fish tissue
that is edible
Number of recreational
fishers that can be
supported by the harvest
Number of recreational
fishers that are likely to
be exposed
3,384g/75m * l/4.5m * InT/l^lxlO^acres * 2.205xlO'3lbs/g = 89.5 Ibs/ac/yr
0.2
89.5 Ibs/acre/yr * 126.8 acres * 0.2 = 2,269.7 Ibs/yr
5.86
35%
2,269.7 Ibs/yr * 0.35 * 1/5.86 Ibs/year/person = 136 people
136 people
a Standing stock data for this facility were obtained from the state Department of Natural Resources
b The average width of the small streams was used in this calculation because the standing stock data are for a
small stream
Facility #81
  Stream miles within a
  five-mile radius of the
  facility
  (determined from
  topographic maps)
Small streams: 7 miles
Large streams: 7 miles
  Average stream width
  (determined from
  topographic maps)
Small streams:  1/350 mile
Large streams:  1/70 mile
  Stream acres within five
  miles of the facility
Small streams: 7 miles * 1/350 miles * 640 acres/mi2 = 12.7 acres

Large Streams: 7 miles * 1/70 miles * 640 acres/mi2  = 64 acres

Total = 76.7 acres
  Raw standing stock data
  (obtained directly from
  studies provided by the
  relevant state agencies3)
brown trout: 7.931 kg/550 ft
rainbow trout: 77 kg 550 ft
LN sucker: 2.981 kg/550 ft

Total = 87.912 kg/550 ft
                                                Page D - 5

-------
Standing stock data
(converted to units^ of
Ibs/acre/year)
Exploitation rate
Pounds of fish caught
per year within the area
of influence
Pounds of fish ingested
per year per recreational
fisher
Percent of fish tissue
that is edible
Number of recreational
fishers that can be
supported by the harvest
Number of recreational
fishers that are likely to
be exposed
87.912kg/550 ft * l/70ftb * 4.35xl04ft2/acre * 2.20461bs/kg = 219 Ibs/acre/year
0.2
219 Ibs/acre/yr * 76.7 acres * 0.2 = 3,359.5 Ibs/year
5.86
35%
3,359.5 Ibs/yr * 0.35 * 1/5.86 Ibs/year/person = 201 people
201 people
a Standing stock data for this site were obtained from the state Department of Natural Resources
  The average stream width of the small streams was used in this calculation because the standing stock data are
for a small stream
                                                Page D - 6

-------
                                        Appendix E
              Example of Facility-Level Raw Data, Vegetable Ingestion on Pathway

       The data used in calculating the population risks associated with vegetable ingestion were
obtained from the National Agricultural Statistics Service (NASS) via the Internet
(http//www.usda.gov/NASS).  This appendix outlines the various data points used in this analysis, and
provides sample pages of the relevant data.  Note that these sample pages do not represent actual data
used in this analysis; the county displayed in this appendix was chosen randomly from all counties in
the United  States for the sake of providing an example of the relevant data.

       The data points used for this analysis are:

1992 Census of Agriculture

       • Total sales, (farms)
       • Value of sales less than $1,000 (farms)
       • Value of sales $l,000-$2,499 (farms)
       • Value of sales $2,500-$4,999 (farms)
       • Value of sales $5,000-$9,999 (farms)

       These data points were summed to give  the total number of farms with sales less than $10,000.

1990 Census of Population and Housing

       • Social Characteristics Table: County urban population
                                            County farm population
                                            Total county population

USA Counties

       • Land area in square miles, 1990

-------
f
                  Agriculture Census for Salt Lake County, Utah
Table 2. Market Value and Farms by SIC
Enter keyword to search for a report item:
f Agriculture Census - Utah Home Page 1
1992 CENSUS OF AGRICULTURE
     Salt Lake County,
       Utah
                                                                         PAGE  1
TABLE 2.
  MKT VALUE OF AGRI PRODUCTS SOLD & FARMS BY STD INDUSTRIAL CLASSIFICATION

Total sales, (see text) (farms)
Total sales, (see text) ($,1000)
Total sales, average per farm, ($)
Value of sales-Less than $1,000 (farms)
Value of sales-Less than $1,000 ($1,000)
Value of sales $1, 000-$2, 499 (farms)
Value of sales $1, 000-$2 , 499 ($1, 000)
Value of sales $2 , 500-$4, 999 (farms)
Value of sales $2 , 500-$4 , 999 ( $1 , 000)
Value of sales $5, 000-$9, 999 (farms)
Value of sales $5 , 000-$9 , 999 ($1 , 000)
Value of sales $10, 000-$19 , 999 (farms)
Value of sales $10 , 000-$19 , 999 ($1,000)
Value of sales $20, 000-$24 , 999 (farms)
Value of sales $20 , 000-$24 , 999 ($1,000)
Value of sales $25 , 000-$39, 999 (farms)
Value of sales $25 , 000-$39 , 999 ($1,000)
Value of sales $40 , 000-$49 , 999 (farms)
Value of sales $40 , 000-$49 , 999 ($1,000)
Value of sales $50, 000-$99 , 999 (farms)
Value of sales-$50,000-$99,999 ($1,000)
Value of sales-$100, 000-$249,999 (farms)
Value of sales-$100,000-$249,999($l,000)
Value of sales-$250,000-$499,999 (farms)
Value of sales-$250,000-$499,999($l,000)
Value of sales-$500, 000 or more (farms)
Value of sales-$500,000 or more ($1,000)
Crops, incl nurs & grnh crops (farms)
Crops, incl nurs & grnh crops ($1,000)
Grains, (farms)
Grains, ($1,000)
Corn for grain, (farms)
Corn for grain, ($1,000)
Wheat, (farms)
Wheat, ($1,000)
l-LJ^J 	
686
20155
29381
160
30
154
237
112
391
72
516
77
1033
13
281
27
833
13
558
14
981
26
3995
11
3911
7
7390
295
10474
56
1469
3
• (D)
22
974
-- - I J.-70 / J
734
23794
32417
193
39
161
254
126
419
97
637
51
711
7
159
20
617
9
406
31
2314
20
3379
12
3917
7
10942
277
8007
70
1040
7
(D)
25
671

805
26142
32475
164
(D)
183
272
159
557
97
674
58
849
18
395
44
1410
9
402
29
1981
25
3977
8
2871
10
12195
337
9428
129
1525
12
194
49
867
                                  Page E - 2

-------
 Soybeans,  (farms)
 Soybeans,  ($1,000)
 Sorghum for grain,  (farms)
 Sorghum for grain,  ($1,000)
 Barley, (farms)
 Barley, ($1,000)
 Oats,  (farms)
 Oats,  ($1,000)
 Other grains,  (farms)
 Other grains,  ($1,000)
 Cotton and cottonseed,  (farms)
 Cotton and cottonseed,  ($1,000)

1992 CENSUS OF AGRICULTURE
     Salt Lake County,
       Utah
  0
  0
  0
  0
 35
294
 10
 31
  1
(D)
  0
  0
  0
  0
  0
  0
 51
261
  8
(D)
  0
  0
  0
  0
    0
    0
    0
    0
 (NA)
 (NA)
   20
   23
   89
  442
    0
    0

PAGE 2
TABLE 2.
  MKT VALUE OF AGRI PRODUCTS SOLD  &  FARMS  BY STD INDUSTRIAL CLASSIFICATION
Tobacco, (farms)
Tobacco, ($1,000)
Hay, silage, and field seeds, (farms)
Hay, silage, and field seeds, ($1,000)
Vegetables, sweet corn, & melons (farms)
Vegetables, sweet corn, & melons ($1 , 000)
Fruits, nuts, and berries, (farms)
Fruits, nuts, and berries, ($1,000)
Nursery and greenhouse crops, (farms)
Nursery and greenhouse crops, ($1,000)
Other crops, (farms)
Other crops, ($1,000)
Livestock, poultry, products (farms)
Livestock, poultry, products ($1,000)
Poultry and poultry products, (farms)
Poultry and poultry products, ($1,000)
Dairy products, (farms)
Dairy products, ($1,000)
Cattle and calves, (farms)
Cattle and calves, ($1,000)
Hogs and pigs, (farms)
Hogs and pigs, ($1,000)
Sheep, lambs, and wool, (farms)
Sheep, lambs, and wool, ($1,000)
Other livestock. .. (see text) (farms)
Other livestock. .. (see text) ($1,000)
SIC-Cash grains (Oil)
SIC-Field crops except cash grains (013)
SIC-Cotton (0131)
SIC-Tobacco (0132)
SIC-Oth field crop(text) (0133,0134,0139)
SIC-Vegetables and melons (016)
SIC-Fruits and tree nuts (017)
SIC-Horticultural specialties (018)
SIC-General farms, primarily crop (019)
SIC-Livestock, except specialties (021)
SIC-Beef cattle, except feedlots (0212)
SIC-Dairy farms (024)
SIC-Poultry and eggs (025)
SIC-Animal Specialties (027)
SIC-Gen farms, prim Ivestk & spec (029)
	 i uyt j 	
0
0
198
1316
47
1265
28
362
45
6031
10
32
400
9681
24
34
11
3742
188
3419
38
106
95
391
170
1989
25
125
0
0
125
20
20
44
12
208
119
10
7
212
3
- I.L:»O ; j
0
0
185
1001
41
1341
29
109
34
4505
8
11
435
15787
41
(D)
18
2863
198
3419
36
(D)
82
957
196
4084
17
131
0
0
131
23
22
31
15
202
108
14
12
252
15
li^O^J
0
0
221
1107
,51
1113
28
111
36
5525
16
47
518
16714
52
(D)
29
3634
242
3421
52
607
117
559
210
(D)
57
135
0
0
135
22
24
32
33
236
116
15
19
216
16
                                    Page E - 3

-------
     (D)  Withheld to avoid disclosing data for individual farms.
     (X)  Not applicable.
     (Z)  Less than half of the unit shown.
     (NA) Not available.
     ...  Unpublished data.
     [B]  Not available due to brackets.
                  Agriculture Census for Salt Lake County, Utah
Table 2. Market Value and Farms by SIC
Enter keyword to search for a report item:
 Agriculture Census - Utah Home Page 1
Navigation to Other Databases
                                     Page E - 4

-------
 m
USA Counties - Salt Lake County, Utah
Land Area
Enter keyword to search for a report item:
\ USA Counties - Utah Home Page 1
   Salt Lake  County, Utah
        -- Land Area
LAND AREA IN SQUARE MILES 1980
LAND AREA IN SQUARE MILES 1990

[ Document Flaes and Footnotes 1
                         756.3
                         737.4
                  USA Counties - Salt Lake County, Utah
Land Area
Enter keyword to search for a report item:
[USA Counties - Utah Home Page 1
Navigation to Other Databases
                                     Page E - 5

-------
                  Social Characteristics for Salt Lake County, Utah
       .-.;,. Summary
       ..' . Housing
         . Income
         . Labor
         . Social
f Utah Home Page 1 f Search for Citv. Town. CDP 1 \ Search for County 1
1990 Census of  Population  and Housing                                    Page  1
     Utah
       Salt Lake County

URBAN AND RURAL RESIDENCE
     Total population	     725, 956
Urban population	     721, 342
    Percent of  total  population	        99.4
Rural population	       4, 614
    Percent of  total  population	         0.6
  Farm population	          73

SCHOOL ENROLLMENT
     Persons 3  years  and over enrolled in school	     239,033
Preprimary school	      18,212
Elementary or high school	     168,237
  Percent in private  school	         3.8
College	      52, 584

EDUCATIONAL ATTAINMENT
     Persons 3  years  and over enrolled in school	     398,673
Less than 9th grade	      12,137
9th to 12th grade,  no diploma	      46,409
High school graduate	     105,528
Some college, no degree	     109 , 605
Associates degree	      29,967
Bachelor ' s degree	      64 , 799
Graduate or professional degree	      30,228

Percent high school graduate or  higher	        85.3
Percent bachelor ' s degree  or higher	        23.8

RESIDENCE IN 1985
      Persons 5 years and  over	     656,130
Lived in same house	     341, 076
Lived in different house in  U.S	     305,677
  Same State	     240 , 263
    Same county	     207,758
    Different county	      32 , 505
  Different State	      65,414
Lived abroad	       9 , 377

DISABILITY OF CIVILIAN NONINSTITUTIONALIZED  PERSONS
     Persons 16 to 64 years	     430,002
With a mobility or self-care limitation	      12,380
  With a mobility limitation	       6,533
  With a self-care limitation	       8,443
With a work disability	      31, 694


                                     Page E- 6

-------
    In labor force	      15 , 977
  Prevented from working	      12,264
      Persons 65 years and over	      58,880
With a mobility or self-care limitation	      11,008
  With a mobility limitation	       8,765
  With a self-care limitation	       6,105

1990 Census of Population and Housing                                    Page 2
     Utah
       Salt Lake County

CHILDREN EVER BORN PER 1,000 WOMEN
Women 15 to 24 years	         336
Women 25 to 34 years	       1, 741
Women 35 to 44 years	       2,602

VETERAN STATUS
Civilian veterans 16 years and over	      62,602
  65 years and over	      15,092

NATIVITY AND PLACE OF BIRTH
Native population	     692,942
    Percent born in state of residence	        70.0
Foreign-born population	      33,014
  Entered the U.S. 1980 to 1990	      13, 678

LANGUAGE SPOKEN AT HOME
    Persons 5 years and over	     656,130
Speak a language other than English	      58,236
    Do not speak Englisjh  ' very well'	      20, 590
  Speak Spanish	      23,207
    Do not speak English  'very well'	       7,960
  Speak Asian or Pacific  Island language	      13,015
    Do not speak English  ' very well'	       6,409

ANCESTRY
  Total ancestries reported	     950, 646
Arab	       1,547
Austrian	       1,991
Belgian	         525
Canadian	       1,762
Czech	       2,170
Danish	      60, 693
Dutch	      24,340
English	.•	     289,136
Finnish	       1,698
French  {except Basque)	      22,110
French Canadian	       2,849
German	     132, 831
Greek	       6,583
Hungarian	       1,256
Irish	      62,623
Italian	     22,275
Lithuanian	         575
Norwegian	     16,280
Polish	       7,593
Portuguese	        839
Romanian	        392
Russian	       2,700
Scotch-Irish	     11,279

1990 Census of  Population and Housing                                    Page  3
     Utah
        Salt Lake County
                                       Page E- 7

-------
Scottish	
Slovak	
Subsaharan African	
Swedish	
Swiss	
Ukrainian	
United States or American	
Welsh	
West Indian  (excluding Hispanic origin groups)
Yugoslavian	
Other ancestries	
 35,848
  2,009
    442
 46,230
 12,441
    564
 26,237
 18,753
    132
  2,145
131,798
                  Social Characteristics for Salt Lake County, Utah
         . Summary
         . Housing
        T. Income
        ". Labor
 Utah Home Page 1 \ Search for Citv. Town. CDP 1 f Search for County 1
Navigation to Other Databases
                                    Page E - 8

-------
                Appendix F
       (includes Attachments A, B, C, and D)

Equations and Background Information from AP-42

-------
                                   ATTACHMENT A
                      BACKGROUND INFORMATION USED IN
                 ESTIMATING EMISSIONS FROM WIND EROSION*
'U.S. Environmental Protection Agency, 1995.  Compilation of Air Pollutant Emission Factors, Volume I: Stationary and Area
Sources, AP-42 Fifth Edition, U.S. Environmental Protection Agency, Research Triangle Park, NC, Section 13.2.5, pp. 13.2.5-1
to 13.2.5-14.

-------
13.2.5 Industrial Wind Erosion

13.2.5.1  General1'3

       Dust emissions may be generated by wind erosion of open aggregate storage piles and
exposed areas  within an industrial facility.  These sources typically are characterized by
nonhomogeneous surfaces impregnated with nonerodible elements (particles larger than approximately
1 centimeter [cm] in diameter).  Field testing of coal piles and other exposed materials using a
portable wind  tunnel has shown that (a) threshold wind speeds exceed 5 meters per second (m/s)
(11 miles per hour [mph]) at IS cm above the surface or  10 m/s (22 mph) at 7 m above the  surface,
and (b) paniculate emission rates tend to decay rapidly (half-life of a few minutes) during an erosion
event.  In other words, these aggregate material surfaces are characterized by finite availability of
erodible material (mass/area) referred to as the erosion potential. Any natural crusting of the surface
binds the erodible material, thereby reducing the erosion potential.

13.2.5.2 Emissions And Correction Parameters

       If typical values for threshold wind speed at 15 cm are corrected to typical wind sensor height
(7 - 10 m),  the resulting values  exceed the upper extremes of hourly mean  wind speeds observed in
most areas of  the country.  In other words,  mean atmospheric wind speeds are not sufficient to sustain
wind erosion from flat surfaces  of the type tested.  However,  wind gusts may quickly deplete a
substantial portion of the erosion potential.  Because erosion potential has been found to increase
rapidly with increasing wind speed, estimated emissions should be related to the gusts of highest
magnitude.

       The routinely measured meteorological variable that best reflects the magnitude of wind  gusts
is the fastest mile.  This quantity represents the wind speed corresponding  to the whole mile of wind
movement that has passed by the 1 mile contact anemometer in the least amount of time.  Daily
measurements of the fastest mile are  presented in the monthly Local Climatological Data (LCD)
summaries. The duration of the fastest mile, typically about 2 minutes (for a fastest mile of 30 mph),
matches well with the half-life of the erosion process, which ranges between 1 and 4 minutes. It
should be noted, however, that  peak  winds can significantly exceed the daily fastest mile.

       The wind speed profile  in the surface boundary layer  is found to follow a logarithmic
distribution:
                                         0.4    z0
                                                       (z > z0)                             (1)
                                        "•*     Z0


where:
         u  = wind speed, cm/s
        u*  = friction velocity, cm/s
         z  = height above test surface, cm
        z0  = roughness height,  cm
       0.4  = von Karman's constant, dimension!ess
 1/95                                  Miscellaneous Sources                             13.2.5-1

-------
The friction velocity (u*) is a measure of wind shear stress on the erodible surface, as determined
from the slope of the logarithmic velocity profile. The roughness height (z0) is a measure of the
roughness of the exposed surface as determined from the y intercept of the velocity profile, i. e., the
height at which the wind speed is zero.  These parameters are illustrated in Figure 13.2.5-1 for a
roughness height of 0.1 cm.
                                                     NO -Sftmo if  tOm
                   Figure 13.2.5-1. Illustration of logarithmic velocity profile.
        Emissions generated by wind erosion are also dependent on the frequency of disturbance of
the erodible surface because each time that a surface is disturbed, its erosion potential is restored.  A
disturbance is defined as an action that results in the exposure of fresh surface material.  On a storage
pile, this would occur whenever aggregate material  is either added to or removed from the old
surface.  A disturbance of an exposed area may also result from the turning of surface material to a
depth exceeding the size of the largest pieces of material present.

 13.2.5.3 Predictive Emission Factor Equation4

        The emission factor for wind-generated paniculate emissions from mixtures of erodible and
nonerodible surface material subject to disturbance may be expressed in units of grams per square
meter (g/m2) per year as follows:
                                   Emission factor
    N
k  S
                                                       (2)
 13.2.5-2
EMISSION FACTORS
                         (Refonnatted 1/95) 9/90

-------
where:

         k = particle size multiplier
        N = number of disturbances per year
        P; = erosion potential corresponding to the observed (or probable) fastest mile of wind for
             the ith period between disturbances, g/m2

The particle size multiplier (k) for Equation 2 varies with aerodynamic particle size, as follows:
Aerodynamic Particle Size Multipliers For Equation 2
30 jim
1.0
<15/mi
0.6
<10 pm
0.5
< 2.5 /im
0.2
       This distribution of particle size within the under 30 micrometer (pm) fraction is comparable
to the distributions reported for other fugitive dust sources where wind speed is a factor.  This is
illustrated, for example, in the distributions for batch and continuous drop operations encompassing a
number of test aggregate materials (see Section 13.2.4).

       In calculating emission factors, each area of an credible surface that is subject to a different
frequency of disturbance should be treated separately.  For a surface disturbed daily, N = 365 per
year, and for a surface disturbance once every 6  months, N =  2 per year.

       The erosion  potential function for a dry,  exposed surface is:


                                  P = 58  (u * - ut*)2 *25 (u *  - ut*)
                                                                                               (3)
                                  P =  0 for u *  «Sut*
where:
        u* = friction velocity (m/s)
        ^ = threshold friction velocity (m/s)

Because of the nonlinear form of the erosion potential function, each erosion event must be treated
separately.

        Equations 2 and 3 apply only to dry, exposed materials with limited erosion potential.  The
resulting calculation is valid only for a time period as long or longer than the period between
disturbances. Calculated emissions represent intermittent events and should not be input directly into
dispersion models that assume steady-state emission rates.

        For uncrusted surfaces,  the threshold friction velocity is best estimated  from the dry aggregate
structure of the soil.  A simple hand sieving test of surface soil can be used to determine the mode of
the surface aggregate size distribution by inspection of relative sieve catch amounts,  following the
procedure described below.
 1/95                                  Miscellaneous Sources                               13.2.5-3

-------
    FIELD PROCEDURE FOR DETERMINATION OF THRESHOLD FRICTION VELOCITY
                  (from a 1952 laboratory procedure published by W. S.  Chepil):

       1.      Prepare a nest of sieves with the following openings:  4 mm, 2 mm, 1 mm, 0.5 mm,
              and 0.25 mm. Place a collector pan below the bottom (0.25 mm) sieve.

       2.      Collect a sample representing the surface layer of loose particles (approximately 1 cm
              in depth, for an encrusted surface), removing any rocks larger than about 1 cm in
              average physical diameter. The area to be sampled should  be not less than 30 cm by
              30 cm.

       3.      Pour the sample into the top  sieve (4-mm  opening), and place a lid on the top.

       4.      Move the covered sieve/pan unit by hand, using a broad circular arm motion in the
              horizontal plane.  Complete 20 circular movements at a speed just necessary to
              achieve some relative horizontal motion between the sieve and the panicles.

       5.      Inspect the relative quantities of catch within each sieve, and determine where the
              mode in the aggregate size distribution lies,  i. e., between the opening size of the
              sieve with the largest catch and the opening  size of the next largest sieve.

       6.      Determine the threshold friction velocity from Table  13.2.5-1.
The results of the sieving can be interpreted using Table 13.2.5-1. Alternatively, the threshold
friction velocity for erosion can be determined from the mode of the aggregate size distribution using
the graphical relationship described by Gillette.5'6  If the surface material contains nonerodible
elements that are too large to include in the sieving (i. e., greater than about 1 cm in diameter), the
effect of the elements must be taken into account by increasing the threshold friction velocity.10
        Table 13.2.5-1 (Metric Units). FIELD PROCEDURE FOR DETERMINATION OF
                             THRESHOLD FRICTION VELOCITY
Tyler Sieve No.
5
9
16
32
60
Opening (mm)
4
2
1
0.5
0.25
Midpoint (mm)

3
1.5
0.75
0.375
u*(cm/s)

100
76
58
43
       Threshold friction velocities for several surface types have been determined by field
measurements with a portable wind tunnel.  These values are presented in Table 13.2.5-2.
 13.2.5-4
EMISSION FACTORS
1/95

-------
             Table 13.2.5-2 (Metric Units).  THRESHOLD FRICTION VELOCITIES
Material
Overburden*
Scoria (roadbed material)*
Ground coal (surrounding
coal pile)*
Uncrusted coal pile*
Scraper tracks on coal pile*>b
Fine coal dust on concrete padc
Threshold
Friction
Velocity
(m/s)
1.02
1.33
0.55
1.12
0.62
0.54
Roughness
Height (cm)
0.3
0.3
0.01
0.3
0.06
0.2
Threshold Wind Velocity At
10 m (m/s)
z0 = Act
21
27
16
23
15
11
z0 = 0.5 cm
19
25
10
21
12
10
* Western surface coal mine. Reference 2.
b Lightly crusted.
c Eastern power plant.  Reference 3.
       The fastest mile of wind for the periods between disturbances may be obtained from the
monthly LCD summaries for the nearest reporting weather station that is representative of the site in
question.7  These summaries report actual fastest mile values for  each day of a given month.  Because
the erosion potential is a highly nonlinear function of the fastest mile, mean values of the fastest mile
are inappropriate.  The anemometer heights of reporting weather  stations are found in Reference 8,
and should be corrected to a 10-m reference height using Equation 1.

       To convert the fastest mile of wind (u+) from a reference anemometer  height of 10 m to the
equivalent friction velocity (u*), the logarithmic  wind speed  profile may be used to yield the following
equation:
                                        u * = 0.053 Uj*0
                                                                             (4)
where:
         u
          10
friction velocity (m/s)

fastest mile of reference anemometer for period between disturbances (m/s)
        This assumes a typical roughness height of 0.5 cm for open terrain.  Equation 4 is restricted
 to large relatively flat piles or exposed areas with little penetration into the surface wind layer.

        If the pile significantly penetrates the surface wind layer (i. e., with a height-to-base ratio
 exceeding 0.2), it is necessary to divide the pile area into subareas representing different degrees of
 exposure to wind.  The results of physical modeling show that the frontal face of an elevated pile is
 exposed to wind speeds of the same order as the approach wind speed at the top of the pile.
 1/95
                      Miscellaneous Sources
13.2.5-5

-------
       For 2 representative pile shapes (conical and oval with flattop, 37-degree side slope), the
ratios of surface wind speed (u,) to approach wind speed (ur) have been derived from wind tunnel
studies.9 The results are shown in Figure 13.2.5-2 corresponding to an actual pile height of 11 m, a
reference (upwind) anemometer height of 10 ra, and a pile surface roughness height (z0) of 0.5 cm.
The measured surface winds correspond to a height of 25 cm above the  surface.  The area fraction
within each contour pair is specified in Table 13.2.5-3.
              Table 13.2.5-3.  SUBAREA DISTRIBUTION FOR REGIMES OF us/ura
Pile Subarea
0.2a
0.2b
0.2c
0.6a
0.6b
0.9
1.1
Percent Of Pile Surface Area
Pile A
5
35
NA
48
NA
12
NA
Pile Bl Pile
5
B2 Pile B3
3 3
2 28 25
29 NA NA
26 29 28
24 22 26
14 15 14
NA
3 4
  NA = not applicable.
       The profiles of us/ur in Figure  13.2.5-2 can be used to estimate the surface friction velocity
distribution around similarly shaped piles, using the following procedure:
1.
               Correct the fastest mile value (u"1") for the period of interest from the anemometer
               height (z) to a reference height of 10 m u10 using a variation of Equation 1:
                                     u10 =
                                      In (10/0.005)
                                       In (z/0.005)
 (5)
2.
               where a typical roughness height of 0.5 cm (0.005 m) has been assumed.  If a site-
               specific roughness height is available, it should be used.

               Use the appropriate part of Figure 13.2.5-2 based on the pile shape and orientation to
               the fastest mile of wind, to obtain the corresponding surface wind speed distribution
                                          *   (u$)
                                                     '10
                                                                                            (6)
 13.2.5-6
                              EMISSION FACTORS
1/95

-------
    Flow
  Direction
                        Pile A
                                 Plte B1
                          Pile B2
               Figure 13.2.5-2.  Contours of normalized surface windspeeds, us/ur
1/95
Miscellaneous Sources'
                                                                                     13.2.5-7

-------
       3.      For any subarea of the pile surface having a narrow range of surface wind speed, use
               a variation of Equation 1 to calculate the equivalent friction velocity (u*):

                                    .-  .iS-L.0.10.;
From this point on, the procedure is identical to that used for a fiat pile, as described above.

       Implementation of the above procedure is carried out in the following steps:

       1.      Determine threshold friction velocity for erodible material of interest (see
               Table 13.2.5-2 or determine from mode of aggregate size distribution).

       2.      Divide the exposed surface area into subareas of constant frequency of disturbance
               (N).

       3.      Tabulate fastest mile values (u"1") for each frequency of disturbance and correct them
               to 10 m (u£) using Equation 5.5

       4.      Convert fastest mile values (u10) to equivalent friction velocities (u*), taking into
               account (a) the uniform wind exposure of nonelevated surfaces, using Equation 4, or
               (b) the nonuniform wind exposure of elevated surfaces (piles), using Equations 6 and
               7.

       5.      For elevated surfaces (piles), subdivide areas of constant N into subareas of constant
               u* (i. e., within the isopleth values of u$/ur in Figure 13.2.5-2 and Table 13.2.5-3)
               and determine the size of each subarea.

       6.      Treating each subarea (of constant N and u*) as a separate source, calculate the
               erosion potential (Pj) for each period between disturbances using Equation 3 and the
               emission factor using Equation 2.

       7.      Multiply the resulting emission factor  for each subarea by the size of the subarea, and
               add the emission contributions of all subareas.  Note that the highest 24-hour (hr)
               emissions would be expected to occur on the windiest day of the year.  Maximum
               emissions are calculated assuming a single event with the highest fastest mile value for
               the annual period.

        The recommended emission factor equation presented above assumes that all of the erosion
 potential corresponding to the fastest mile of wind is lost during the period between disturbances.
 Because the fastest mile event typically lasts only about 2 minutes, which corresponds roughly to the
 half-life for the decay of actual erosion potential, it could be argued  that the emission factor
 overestimates paniculate emissions.  However,  there are other aspects  of the wind erosion process
 that offset this apparent conservatism:

        1.     The fastest  mile  event contains peak winds that substantially exceed the mean value
               for the event.


 13.2.5-8                              EMISSION FACTORS                                 1/95

-------
       2.      Whenever the fastest mile event occurs, there are usually a number of periods of
               slightly lower mean wind speed that contain peak gusts of the same order as the
               fastest mile wind speed.

       Of greater concern is the likelihood of overprediction of wind erosion emissions in the case of
surfaces disturbed infrequently in comparison to the rate of crust formation.

13.2.5.4 Example 1:  Calculation for wind erosion emissions from conically shaped coal pile

       A coal burning facility maintains a conically shaped surge pile 11 m in height and 29.2 m in
base diameter, containing about 2000 megagrams (Mg) of coal, with a bulk density of 800 kilograms
per cubic meter  (kg/m3) (50 pounds per cubic feet [Ib/ft3]). The total exposed surface area of the pile
is calculated as follows:

                                S = i r (r2 + h2)

                                 = 3.14(14.6) (14.6)2 + (11.0)2

                                 = 838  m2

       Coal is added to the pile by means of a fixed stacker and reclaimed by front-end loaders
operating at the base of the pile on the downwind side. In addition, every 3 days 250 Mg
(12.5 percent of the stored capacity of coal) is added back  to the pile by a topping off operation,
thereby restoring the full capacity of the pile.  It is assumed that (a) the reclaiming operation disturbs
only a limited portion of the surface area where the daily activity is occurring, such that the
remainder of the pile surface remains intact, and (b) the topping off operation creates a  fresh surface
on the entire pile while restoring its original shape in the area depleted by daily reclaiming activity.

        Because of the high frequency of disturbance of the pile, a large number of calculations must
be made to determine each contribution to the total annual wind erosion emissions.  This illustration
will use a  single month as an example.

        Step 1:  In the absence of field data for estimating the threshold friction velocity,  a value of
 1.12 m/s  is obtained from Table 13.2.5-2.

        Step 2:  Except for a small area near the base of the pile (see Figure 13.2.5-3), the entire pile
surface is disturbed every 3 days, corresponding to a value of N =  120 per year.  It will be shown
that the contribution of the area where daily activity  occurs is negligible so that it does  not need to be
treated separately in the calculations.

        Step 3:  The calculation procedure* involves determination of the fastest mile for each period
of disturbance.  Figure 13.2.5^4. shows a representative set of values (for a 1-month  period) that are
 assumed to be applicable to the geographic area of the pile location.  The values have been separated
 into 3-day periods, and the highest value in each period is indicated. In this example, the
 anemometer height is 7 m, so that a height correction to 10 m is needed for the fastest  mile values.
 From Equation 5,
                                    u*, ^  fin (10/0.005)]
                                                 In (7/0.005) J

                                    u^ = 1.05 U7*



 1/95                                  Miscellaneous Sources                              13.2.5-9

-------
       Prevailing
       Wind
       Direction
                                     Circled value*
                                     refer  to
          A portion of C2 is disturbed daily by reclaiming activities.
                    Area
                     ID
                     A

                     B
                     cl + C2
    0.9

    0.6

    0.2
                                                            Pila Surface
z
12
48
40
Area («2)
101
402
335
                                                                 Total  83B
         Figure 13.2.5-3.  Example 1:  Pile surface areas within each wind speed regime.
13.2.5-10
EMISSION FACTORS
                                                                                    1/QS

-------
              Local  Climatoloeical  Data
                         MONTHLY sunn*;-
                              WIND

.
cc
e
z
9
or
13
30
01
10
13
12
20
29
29
22
14
29
1 7
21
10
10
0>
33
27
32
24
22
32
29
07
34
31
30
30
33
34
29

.
x
ir-
iS
I/I V_l
u &
cc «/»
U
5.3
10.5
2.4
l l .0
1 1 .3
11.1
19.6
10.9
3.0
14 .6
22.3
7.9
7.7
4.5
6.7
13.7
11.2
4.3
9.3
7.5
10.3

2\*
5.9
1 1 .3
12.1
8.3
8.2
5.0
3. 1
4.9
FOB THE
30
3.3
— •

a
a
w
cc &
«
15
6.9
10.6
6.0
11.4
11 .9
19.0
19.8
l i .2
8.1
15. i
23.3
13.5
15.5
9.6
8.8
13.8
M.5
5.8
10.2
7.8
10.6
17.3
8.S
8.8
11.7
12.2
8.5
8.3
6.6
£.2
5.5
FASTEST
MILE

o ~"
!_)&.
&*
15
0
G
10
16



1 7
15
33
^y

IB
©
1 3
"IT
TS
-L2.
14
(S
I?
£a
ri
1 5
Q_5
1 6
1 1
fi\
To
_2_
8
X
^
CE
O
17
36
01
02
13
1 I
30
30
30
13
12
29
17
18
13
1 l
36
34
31
35
24
20
32
13
02
32
32
26
32
32
31
25
flONTrt:
11.1
31
29
	 (OiTE: 11
                                              22
                                             ^^^^
                                              i
                                              2
                                              3
                                              4
                                              e

                                              I
                                              7
                                              5
                                              5
                                              Pt
                                              l
                                              2
                                             5
                                             6
                                             7
                                             ie
                                             19
                                             20
                                             21
                                             22
                                             23
                                             24
                                             25
                                             26
                                             27
                                             23
                                             29
                                             30
          Figure 13.2.5-4. Example daily fastest miles wind for periods of interest.
1/95
Miscellaneous Sources
13.2.5-11

-------
       Step 4:  The next step is to convert the fastest mile value for each 3-day period into the
equivalent friction velocities for each surface wind regime (i. e., u,/ur ratio) of the pile, using
Equations 6 and 7. Figure  13.2.5-3 shows the surface wind speed pattern (expressed as a fraction of
the approach wind speed at  a height of 10 m).  The surface areas lying within each wind speed
regime are tabulated below  the figure.

       The calculated friction velocities are presented in Table 13.2.5-4.  As indicated, only 3 of the
periods contain a friction velocity which exceeds the threshold value of 1.12 m/s for an uncrusted
coal pile.  These 3 values all occur within the us/ur = 0.9 regime of the pile surface.
                    Table 13.2.5-4 (Metric And English Units).  EXAMPLE  1:
                         CALCULATION OF FRICTION VELOCITIES
3-Day Period
1
2
3
4
5
6
7
8
9
10
u.
mph
14
29
30
31
22
21
16
25
17
13
7
m/s
6.3
13.0
13.4
13.9
9.8
9.4
7.2
11.2
7.6
5.8
u
mph
15
31
32
33
23
22
17
26
18
14
10
m/s
6.6
13.7
14.1
14.6
10.3
9.9
7.6
11.8
8.0
6.1
u* = O.lu+ (m/s)
us/ur: 0.2
0.13
0.27
0.28
0.29
0.21
0.20
0.15
0.24
0.16
0.12
us/ur: 0.6
0.40
0.82
0.84
0.88
0.62
0.59
0.46
0.71
0.48
0.37
us/Uj.: 0.9
0.59
1.23
1.27
1.31
0.93
0.89
0.68
1.06
0.72
0.55
        Step 5:  This step is not necessary because there is only 1 frequency of disturbance used in
 the calculations.  It is clear that the small area of daily disturbance (which lies entirely within the
 us/ur = 0.2 regime) is never subject to wind speeds exceeding the threshold value.

        Steps 6 and 7: The final set of calculations (shown in Table 13.2.5-5) involves the tabulation
 and summation of emissions for each disturbance period and for the affected subarea.  The erosion
 potential (P) is calculated from Equation 3.

        For example, the calculation for the second 3-day period is:

                             P  - 58(u * - u,*)2 * 25(u * - ut")

                             P2 = 58(1.23 - 1.12)2 * 25(1.23 - 1.12)

                                 = 0.70*2.75 = 3.45 g/ra2
 13.2.5-12
EMISSION FACTORS
1/95

-------
     Table 13.2.5-5 (Metric Units).  EXAMPLE 1: CALCULATION OF PM-10 EMISSIONS*
3-Day Period
2
3
4
TOTAL
u* (m/s)
1.23
1.27
1.31

* *
U -U,
(m/s)
0.11
0.15
0.19

P (g/m2)
3.45
5.06
6.84

ID
A
A
A

Pile Surface
Area
(m2)
101
101
101

kPA
(g)
170
260
350
780
* Where u  =  1.12 m/s for uncrusted coal and k » 0.5 for PM-10.
       The emissions of paniculate matter greater than 10 jim (PM-10) generated by each event are
found as the product of the PM-10 multiplier (k = 0.5), the erosion potential (P), and the affected
area of the pile (A).

       As shown in Table 13.2.5-5, the results of these calculations indicate a monthly PM-10
emission total of 780 g.

13.2.5.5  Example 2:  Calculation for wind erosion from flat area covered with coal dust

       A flat circular area 29.2 m in diameter is covered with coal dust left over from the total
reclaiming of a conical coal pile described in the example above. The total exposed surface area is
calculated as follows:
                            s   »   1  d2 - 0.785 (29.2)2 - 670 m2
                                  4

        This area will remain exposed for a period of 1 month when a new pile will be formed.

        Step 1: In the absence of field data for estimating the threshold friction velocity, a value of
0.54 m/s is obtained from Table  13.2.5-2.

               The entire surface area is exposed for a period of 1 month after removal of a pile and
 N = 1/yr.
              *.  From Figure 13.2.5-4, the highest value of fastest mile for the 30-day period
 (31 mph) occurs on the llth day of the period. In this example, the reference anemometer height is
 7 m, so that a height correction is needed for the fastest mile value.  From Step 3 of the previous
 example, u+0  » 1.05 u^, so that uj^  »  33 mph.

        SJSL&  Equation 4 is used to convert the fastest mile value of 14.6 m/s (33 mph) to an
 equivalent friction velocity of 0.77 m/s.  This value exceeds the threshold friction velocity from
 Step 1 so that erosion does occur.

        Step 5:  This step is not necessary, because there is only 1 frequency of disturbance for the
 entire source area.
 1/95
Miscellaneous Sources
13.2.5-13

-------
       Steps 6 and 7:  the PM-10 emissions generated by the erosion event are calculated as die
product of the PM-10 multiplier (k = 0.5), the erosion potential (P) and the source area (A). The
erosion potential is calculated from Equation 3  as follows:

                           P - 58(u'-  ut")2 + 25(u"- ut*)

                           P - 58(0.77 - 0.54)2 - 25(0.77  - 0.54)

                              = 3.07 * 5.75

                              = 8.82 g/m2

Thus the PM-10 emissions for the 1-month period are found to be:

                            E = (0.5)(8.82 g/m2)(670 m2)

                              = 3.0 kg

References For Section 13.2.5

1.     C. Cowherd, Jr.,  "A New Approach To Estimating Wind Generated Emissions From Coal
       Storage Piles", Presented at the APCA Specialty Conference on Fugitive Dust Issues in the
       Coal Use Cycle, Pittsburgh, PA, April 1983.

2.     K. Axtell and C. Cowherd, Jr., Improved Emission Factors For Fugitive Dust From Surface
       Coal Mining Sources, EPA-600/7-84-048, U. S. Environmental Protection Agency,
       Cincinnati, OH, March 1984.

3.     G. E Muleski, "Coal Yard Wind Erosion Measurement",  Midwest Research Institute, Kansas
       City, MO,  March 1985.

4.     Update Of Fugitive Dust Emissions Factors In AP-42 Section 11.2 —  Wind Erosion, MRI No.
       8985-K, Midwest Research Institute, Kansas City, MO, 1988.

5.     W. S. Chepil, "Improved Rotary Sieve For Measuring State And Stability Of Dry Soil
       Structure", Soil Science Society Of America Proceedings,  16:113-117, 1952.

6.     D. A. Gillette, et at.,  "Threshold Velocities For Input Of Soil Particles Into The Air By
       Desert Soils", Journal Of Geophysical Research, S5(C10):5621-5630.

7.     Local Climatological Data, National Climatic Center, Ashevilie, NC.

8.     M. J. Changery, National Wind Data Index Final Report, HCO/T1041-01 UC-60, National
       Climatic Center, Ashevilie, NC,  December 1978.

9.     B. J. B. Stunder and S. P. S. Arya, "Windbreak Effectiveness For Storage Pile Fugitive Dust
       Control: A Wind Tunnel Study", Journal Of The Air Pollution Control Association,
       35:135-143, 1988.

 10.     C. Cowherd, Jr., et al., Control Of Open Fugitive Dust Sources, EPA 450/3-88-008, U. S.
       Environmental Protection Agency, Research Triangle Park, NC, September 1988.

 13.2.5-14                           EMISSION FACTORS                               1/95

-------
                                   ATTACHMENT B
                      BACKGROUND INFORMATION USED IN
         ESTIMATING EMISSIONS FROM AGGREGATE CKD HANDLING*
'U.S. Environmental Protection Agency, 1995. Compilation of Air Pollutant Emission Factors, Volume I: Stationary and Area
Sources, AP-42 Fifth Edition, U.S. Environmental Protection Agency, Research Triangle Park, NC, Section 13.2.4, pp. 13.2.4-1
to 13.2.4-5.

-------
13.2.4 Aggregate Handling And Storage Piles

13.2.4.1  General

       Inherent in operations that use minerals in aggregate form is the maintenance of outdoor
storage piles.  Storage piles are usually left uncovered, partially because of the need for frequent
material transfer into or out of storage.

       Dust emissions occur at several points in the storage cycle, such as material loading onto the
pile, disturbances by strong wind currents, and loadout from the pile.  The movement of trucks  and
loading equipment in the storage pile area is  also a substantial source of dust.

13.2.4.2  Emissions And Correction Parameters

       The quantity of dust emissions from  aggregate storage operations varies with the volume of
aggregate passing through the storage cycle.  Emissions also depend on 3 parameters of the condition
of a particular storage  pile:  age of the pile, moisture content, and proportion of aggregate fines.

       When freshly processed aggregate is  loaded onto a storage pile, the potential for dust
emissions is at a maximum.  Fines are easily disaggregated and released to the atmosphere upon
exposure to air currents, either from aggregate transfer itself or from high winds.  As the aggregate
pile weathers, however, potential for dust emissions is greatly reduced. Moisture causes aggregation
and cementation of fines to  the surfaces of larger particles.  Any significant rainfall soaks the interior
of the pile, and then the drying process is very slow.

       Silt (particles equal  to or less than  75 micrometers [^mj in diameter) content is determined by
measuring the portion  of dry aggregate material that passes through a 200-mesh screen, using
ASTM-C-136 method.1  Table 13.2.4-1 summarizes measured silt and moisture values for industrial
aggregate materials.

13.2.4.3 Predictive Emission Factor Equations

       Total dust emissions from aggregate  storage piles result from several distinct source activities
within the storage cycle:

        1.   Loading of aggregate onto storage piles (batch or continuous drop operations).
       2.   Equipment traffic in storage area.
        3.   Wind erosion of pile surfaces and ground areas around piles.
       4.   Loadout of aggregate for shipment or for return to the process stream (batch or
            continuous drop operations).

        Either adding  aggregate material to a storage pile or removing it usually involves dropping the
material onto  a receiving surface.  Truck dumping on the pile or loading out from the pile to a truck
with a front-end loader are  examples of batch drop operations.  Adding material to the pile by a
conveyor stacker is  an example of a continuous drop operation.
 1/95                                  Miscellaneous Sources                             13.2.4-1

-------
         Table 13.2.4-1. TYPICAL SILT AND MOISTURE CONTENTS OF MATERIALS AT VARIOUS INDUSTRIES"
Industry
Iron and steel production








Stone quarrying and processing

Taconite mining and processing

Western surface coal mining


Coal-fired power plant
Municipal solid waste landfills






No. Of
Facilities
9








2

1

4


1
4






Material
Pellet ore
Lump ore
Coal
Slag
Flue dust
Coke breeze
Blended ore
Sinter
Limestone
Crushed limestone
Various limestone products
Pellets
Tailings
Coal
Overburden
Exposed ground
Coal (as received)
Sand
Slag
Cover
Clay/dirt mix
Clay
Fly ash
Misc. fill materials
Silt
No. Of
Samples
13
9
12
3
3
2
1
1
3
2
8
9
2
15
IS
3
60
1
2
5
1
2
4
1
Content (%)
Range
1.3- 13
2.8 - 19
2.0 - 7.7
3.0-7.3
2.7 - 23
4.4 - 5.4
—
—
0.4 - 2.3
1.3- 1.9
0.8 - 14
2.2-5.4.
ND
3.4- 16
3.8-15
5.1-21
0.6-4.8
—
3.0-4.7
5.0 - 16
—
4.5-7.4
78-81
—
Mean
4.3
9.5
4.6
5.3
13
4.9
15
0.7
1.0
1.6
3.9
3.4
11
6.2
7.5
15
2.2
2.6
3.8
9.0
9.2
6.0
80
12
Moisture Content (%)
No. Of
Samples
11
6
11
3
1
2
1
0
2
2
8
7
1
7
0
3
59
1
2
5
1
2
4
1
Range
0.64 - 4.0
1.6-8.0
2.8-11
0.25 - 2.0
—
6.4 - 9.2
—
—
ND
0.3- I.I
0.46 - 5.0
0.05 - 2.0
—
2.8 - 20
—
0.8 - 6.4
2.7 - 7.4
—
2.3 - 4.9
8.9 - 16
—
8.9-11
26-29
—
Mean
2.2
5.4
4.8
0.92
7
7.8
6.6
—
0.2
0.7
2.1
0.9
0.4
6.9
—
3.4
4.5
7.4
3.6
12
14
10
27
II
* References 1-10.  ND = no data.

-------
       The quantity of paniculate emissions generated by either type of drop operation, per kilogram
(kg) (ton) of material transferred,  may be estimated, with a rating of A, using the following empirical
expression:
          .11
                         E=k(0.0016)
                         E=k(0.0032)
                                           __
                                           2.2
                                            .Mil*
                                             2
              (kg/megagram  [MgJ)
                                                                                             (1)
              (pound [lb]/ton)
where:

        E =  emission factor
        k =  particle size multiplier (dimensionless)
        U =  mean wind speed, meters per second (m/s) (miles per hour [mph])
        M =  material moisture content (%)

The particle size multiplier in the equation, k, varies with aerodynamic particle size range, as follows:
Aerodynamic Particle Size Multiplier (k) For Equation 1
< 30 Mm
0.74
< 15 /xm
0.48
< 10 Aim
0.35
< 5 pm
0.20
< 2.5 nm
0.11
       The equation retains the assigned quality rating if applied within the ranges of source
conditions that were tested in developing the equation, as follows. Note that silt content is included,
even though silt content does not appear as a correction parameter in the equation. While it is
reasonable to expect that silt content and  emission factors are interrelated, no significant correlation
between the 2 was found during the derivation of the equation, probably because most tests with high
silt contents were conducted under lower winds, and vice versa.   It is recommended that estimates
from the equation be reduced  1 quality rating level if the silt content used in a particular application
falls outside the range given:
Ranges Of Source Conditions For Equation 1
Silt Content
(%)
0.44 - 19
Moisture Content
(%)
0.25 - 4.8
Wind Speed
m/s
0.6 - 6.7
mph
1.3- 15 '
 1/95
Miscellaneous Sources
13.2.4-3

-------
       To retain the quality rating of the equation when it is applied to a specific facility, reliable
correction parameters must be determined for specific sources of interest. The field and laboratory
procedures for aggregate sampling are given in Reference 3.  In the event that site-specific values for
correction parameters cannot be obtained, the appropriate mean from Table 13.2.4-1 may be used,
but the quality rating of the equation is reduced by  1 letter.

       For emissions from equipment traffic (trucks, front-end loaders, dozers, etc.) traveling
between or on piles, it is recommended that the equations for vehicle traffic on unpaved surfaces be
used (see Section 13.2.2). For vehicle travel between storage piles,  the silt value(s) for the  areas
among the piles  (which may differ from the silt values for the stored materials) should be used.

       Worst-case emissions from storage pile areas occur under dry, windy conditions.  Worst-case
emissions from materials-handling operations may be calculated by substituting into the equation
appropriate values for aggregate material moisture content and for anticipated wind speeds during the
worst case averaging period, usually 24 hours. The treatment of dry conditions for Section 13.2.2,
vehicle traffic, "Unpaved Roads", follows the methodology described in that section centering on
parameter p.  A separate set of nonclimatic correction parameters and source extent values
corresponding to higher than normal storage pile activity also may be justified for the worst-case
averaging period.

13.2.4.4  Controls12'13

       Watering and the use of chemical wetting agents are the principal means for control of
aggregate storage pile emissions.  Enclosure or covering of inactive  piles to reduce wind erosion can
also reduce emissions.  Watering is useful mainly to reduce emissions from vehicle traffic in the
storage pile area.  Watering of the storage piles themselves typically has only a very temporary  slight
effect on total emissions. A much more effective technique is to apply chemical agents (such as
surfactants) that permit more extensive wetting.  Continuous chemical  treating of material loaded onto
piles, coupled with watering or treatment of roadways, can reduce total paniculate emissions from
aggregate storage operations by up to 90 percent.12

References For Section 13.2.4

 1.      C. Cowherd, Jr., et ai., Development Of Emission Factors For Fugitive Dust Sources,
        EPA-450/3-74-037, U. S. Environmental Protection Agency, Research Triangle Park, NC,
        June 1974.

2.      R. Bohn, et al., Fugitive Emissions From Integrated Iron And Steel Plants,
        EPA-600/2-78-050, U. S. Environmental Protection Agency, Cincinnati, OH, March  1978.

 3.      C. Cowherd, Jr., et al., Iron And Steel Plant Open Dust Source Fugitive Emission Evaluation,
        EPA-600/2-79-103, U. S. Environmental Protection Agency, Cincinnati, OH, May 1979.

 4.      Evaluation Of Open Dust Sources In The Vicinity Of Buffalo, New York, EPA Contract
        No. 68-02-2545, Midwest Research Institute, Kansas  City, MO, March 1979.

 5.      C. Cowherd, Jr., and T. Cuscino, Jr., Fugitive Emissions Evaluation, MRI-4343-L, Midwest
        Research Institute, Kansas City, MO, February 1977.

 6.      T. Cuscino, Jr., et al., Taconite Mining  Fugitive Emissions  Study, Minnesota Pollution
        Control Agency, Roseville, MN, June 1979.

 13.2.4-4                             EMISSION FACTORS                                 1/95

-------
7.      Improved Emission Factors For Fugitive Dust From Western Surface Coal Mining Sources,
       2 Volumes, EPA Contract No. 68-03-2924, PEDCo Environmental, Kansas City, MO, and
       Midwest Research Institute, Kansas City, MO, July 1981.

8.      Determination Of Fugitive Coal Dust Emissions From Rotary Railcar Dumping, TRC,
       Hartford, CT, May 1984.

9.      PM-10 Emission Inventory Of Landfills In the Lake Calumet Area, EPA Contract
       No. 68-02-3891, Midwest Research Institute, Kansas City, MO, September  1987.

10.    Chicago Area Paniculate Matter Emission Inventory — Sampling And Analysis, EPA Contract
       No. 68-02-4395, Midwest Research Institute, Kansas City, MO, May 1988.

11.    Update Of Fugitive Dust Emission Factors In AP-42 Section 11.2, EPA Contract
       No. 68-02-3891, Midwest Research Institute, Kansas City, MO, July 1987.

12.    G. A. Jutze, et al.. Investigation Of Fugitive Dust Sources Emissions And Control,
       EPA-450/3-74-036a, U. S. Environmental Protection Agency, Research Triangle Park, NC,
       June 1974.

13.    C. Cowherd, Jr., et al., Control Of Open Fugitive Dust Sources, EPA-450/3-88-008,
       U. S. Environmental Protection Agency, Research Triangle Park, NC, September 1988.
 1/95                               Miscellaneous Sources                            13.2.4-5

-------
                                   ATTACHMENT C
                      BACKGROUND INFORMATION USED IN
                ESTIMATING EMISSIONS FROM UNPAVED ROADS*
'U.S. Environmental Protection Agency, 1995. Compilation of Air Pollutant Emission Factors, Volume I: Stationary and Area
Sources, AP-42 Fifth Edition, U.S. Environmental Protection Agency, Research Triangle Park, NC, Section 13.2.2, pp. 13.2.2-1
to 13.2.2-8.

-------
13.2.2 Unpaved Roads

13.2.2.1  General

       Dust plumes trailing behind vehicles traveling on unpaved roads are a familiar sight in rural
areas of the United States.  When a vehicle travels an unpaved road, the force of the wheels on the
road surface causes pulverization of surface material.  Particles are lifted and dropped from the
roiling wheels, and the road surface is exposed to strong air currents in turbulent shear with the
surface. The turbulent wake behind the vehicle continues to act on the road surface after the vehicle
has passed.

13.2.2.2  Emissions Calculation And Correction Parameters

       The quantity of dust emissions from a given segment of unpaved road varies linearly with the
volume of traffic. Field investigations also have shown that emissions depend on correction
parameters (average vehicle speed, average vehicle weight,  average number of wheels per vehicle,
road surface texture, and road surface moisture) that  characterize the condition of a particular road
and the associated vehicle traffic.1"*

       Dust emissions from unpaved roads have been found to vary in direct proportion to the
fraction of silt (particles smaller than 75 micrometers [/im]  in diameter)  in the road surface
materials.1 The silt fraction is determined by measuring the proportion  of loose dry surface dust that
passes a 200-mesh screen, using the ASTM-C-136 method.  Table 13.2.2-1 summarizes measured silt
values for industrial and rural unpaved roads.

       Since the silt content of a rural dirt road will vary with location, it should be measured for
use in projecting emissions. As a conservative approximation, the silt content of the parent soil in the
area can be used. Tests, however, show that road silt content  is normally lower than in the
surrounding parent soil, because the fines are continually removed by the vehicle traffic,  leaving a
higher percentage of coarse particles.

       Unpaved roads have a hard, generally nonporous surface that usually dries  quickly after a
rainfall. The temporary reduction in emissions  caused by precipitation may be accounted for by not
considering emissions on "wet" days (more than 0.254 millimeters [mm] [0.01 inches (in.) ]  of
precipitation).

       The following empirical expression may be used to estimate the quantity of size-specific
paniculate emissions from an unpaved road, per vehicle kilometer traveled  (VKT) or vehicle mile
traveled (VMT):

                                                    0.5  [365-pl    .         r,  i/wp,


                                                                                            (1)


        E=k(5.9)    |JL|   41  141""      "J°5  f4!?l     (pounds pbl/VMT)
 1/95                                  Miscellaneous Sources                              13.2.2-1

-------
       Table 13.2.2-1.  TYPICAL SILT CONTENT VALUES OF SURFACE MATERIAL
                     ON INDUSTRIAL AND RURAL UNPAVED ROADS'
Industry
Copper smelting
Iron and steel production
Sand and gravel processing
Stone quarrying and
processing

Taconite mining and
processing

Western surface coal
mining



Rural roads

Municipal roads
Municipal solid waste
landfills
Road Use Or
Surface Material
Plant road
Plant road
Plant road
Plant road
Haul road
Service road
Haul road
Haul road
Access road
Scraper route
Haul road
(freshly graded)
Gravel/crushed
limestone
Dirt
Unspecified
Disposal routes
Plant
Sites
1
19
1
2
1
1
1
3
2
3
2
3
7
3
4
No. Of
Samples
3
135
3
10
10
8
12
21
2
10
5
9
32
26
20
Silt Content (%)
Range
16-19
0.2 - 19
4.1-6.0
2.4 - 16
5.0 - 15
2.4-7.1
3.9-9.7
2.8 - 18
4.9 - 5.3
7.2-25
18-29
5.0 - 13
1.6-68
0.4 - 13
2.2-21
Mean
17
6.0
4.8
10
9.6
4.3
5.8
8.4
5.1
17
24
8.9
12
5.7
6.4
a References 1,5-16.
where:
        E = emission factor
        k = particle size multiplier (dimensionless)
        s = silt content of road surface material (%)
        S = mean vehicle speed, kilometers per hour (km/hr) (miles per hour [mph])
       W = mean vehicle weight,  megagrams (Mg) (ton)
        w = mean number of wheels
        p = number of days with at least 0.254 mm (0.01 in.) of precipitation per year (see
            discussion below about the effect of precipitation.)
 13.2.2-2
EMISSION FACTORS
1/95

-------
follows:
       The particle size multiplier in the equation, k, varies with aerodynamic particle size range as

<30/mia
1.0
Aerodynamic
<30Mm
0.80
Particle
  0).  The lower rating is applied
because extrapolation to seasonal or annual conditions assumes that emissions occur at the estimated
rate on days without measurable precipitation and, conversely,  are absent on days with measurable
precipitation. Clearly, natural mitigation depends not only on how much precipitation falls, but also
on other factors affecting the evaporation rate, such as ambient air temperature, wind speed, and
humidity.  Persons in dry, arid portions of the country may wish to base p (the number of wet days)
on a greater amount of precipitation than 0.254 mm  (0.01 in.).  In addition, Reference 18 contains
procedures to estimate the emission reduction achieved by the application of water to an unpaved road
surface.

       The equation retains the assigned quality rating^ if applied within the ranges of source
conditions that were tested in developing the equation, as follows:
Ranges Of Source Conditions For Equation
Road Silt Content
(wt %)
4.3 - 20
Mean Vehicle Weight
Mg
2.7 - 142
ton
3- 157
Mean Vehicle Speed
km/hr
21 -64
mph
13-40
Mean No.
Of Wheels
4- 13
Moreover, to retain the quality rating of the equation when addressing a specific unpaved road, it is
necessary that reliable correction parameter values be determined for the road in question.  The field
and laboratory procedures for determining road surface silt content are given in AP-42
Appendices C.I  and C.2. In the event that site-specific values for correction parameters cannot be
obtained, the appropriate mean values from Table 13.2.2-1 may be used, but the quality rating of the
equation is reduced by 1 letter.

        For calculating annual average emissions, the equation is to be multiplied by annual vehicle
distance traveled (VDT).  Annual average values for each of the correction parameters are to be
substituted for the equation.  Worst-case emissions,  corresponding to dry road conditions, may be
calculated by setting p = 0 in the equation (equivalent to dropping the last term from the equation).
A separate set of nonclimatic correction parameters  and a higher than normal  VDT value may also be
justified for the  worst-case average period (usually 24 hours).  Similarly, in using the equation to
 1/95
Miscellaneous Sources
13.2.2-3

-------
 -J
 k)
 m
 §
 00
q
O
in
\O
                             Figure 13.2.2-1.  Mean number of days with 0.01 inch or more of precipitation in United States.

-------
calculate emissions for a 91-day season of the year, replace the term (365-p)/365 with the term
(91-p)/91, and set p equal to the number of wet days in the 91-day period. Use appropriate seasonal
values for the nonclimatic correction parameters  and for VDT.

13.2.2.3 Controls18*21

        Common control techniques for unpaved roads are paving, surface treating with penetration
chemicals, working stabilization chemicals into the roadbed, watering, and traffic control regulations.
Chemical stabilizers work either by binding the surface material or by enhancing moisture retention.
Paving, as a control technique, is often not economically practical.  Surface chemical treatment and
watering can be accomplished at moderate to low costs, but frequent treatments are required.  Traffic
controls, such as speed limits and traffic volume restrictions, provide moderate emission reductions,
but may be difficult to enforce. The control efficiency obtained by speed reduction can be calculated
using the predictive emission factor equation given above.

        The control efficiencies achievable by paving can be estimated by comparing emission factors
for unpaved and paved road  conditions, relative to airborne particle size range of interest.  The
predictive emission factor equation for paved roads,  given in Section 13.2.4, requires estimation of
the silt loading on the traveled portion of the paved surface, which in turn depends on whether the
pavement is periodically cleaned.  Unless curbing is to be installed, the effects of vehicle excursion
onto shoulders  (berms) also must be taken into account in estimating control efficiency.

        The control efficiencies afforded by the periodic use of road stabilization chemicals are much
more difficult to estimate. The application parameters that determine control efficiency include
dilution ratio, application intensity, mass of diluted chemical per road area, and application frequency.
Other factors that affect the performance of chemical stabilizers include vehicle characteristics
(e. g., traffic volume, average weight) and road  characteristics  (e. g., bearing strength).

        Besides water, petroleum resin products  historically have been the dust suppressants most
widely used on industrial  unpaved roads.  Figure 13.2.2-2 presents a method to estimate average
control efficiencies  associated with petroleum resins applied to  unpaved roads.19 Several items  should
be noted:

        1.   The term "ground inventory"  represents the total volume (per unit area) of petroleum
            resin concentrate (not solution) applied since the start of the dust control season.

        2.   Because petroleum resin products must be periodically reapplied to unpaved roads,  the
            use of a time-averaged control efficiency value is appropriate. Figure 13.2.2-2 presents
            control efficiency values averaged over 2 common application intervals, 2 weeks and
            1 month.  Other application intervals will  require interpolation.

        3.   Note that zero efficiency is assigned until the ground inventory reaches  0.2 liter per
            square meter (L/m2)  (0.05 gallon per square yard [gal/yd2]).

        As an example of the application of Figure 13.2.2-2, suppose that the equation was used to
 estimate an emission factor of 2.0 kg/VKT for PM-10 from a particular road. Also, suppose that,
 starting on May 1, the road is treated with 1 L/m2 of a solution (1 part petroleum resin to 5  parts
 water) on the first of each month through September.  Then, the following average  controlled
 emission factors are found:
 1/95                                 Miscellaneous Sources                               13.2.2-5

-------
 >
to






m
—i
0
-d
3
O
73
C/i



^
0^
>
0
LU
O
U.
• i
u_
UJ
	 i
M«l
O
DC
h-
2
O
o
UJ
O
rr
UJ
 o
 J\
        100
            0
0.25
         80
         60
         40
         20
          0
                      ,
  0.5
  T
                         GROUND INVENTORY
                          (liters/square meter)
                     0.75        1    0        0.25
                      i	1—i	r
              Note: Averaging periods (2 weeks or 1 month)
                   refer to time between applications
                             2 weeks
            0
0.05
                                       1 month
                               TOTAL PARTICULATE
1
                   i
                           1
                                                      0.5
                                                      ~T
                                                                                    0.75
                                                                   2 weeks
                                                                                1 month
                                                     PARTICLES  =  10 umA
1
0.1
                 0.15
                                                                         0.1
                    0.2 0.25  0      0.05
                    (gallons/square yard)

                   GROUND INVENTORY

Figure 13.2.2-2. Average control efficiencies over common application intervals.
        0.15
                                                              0.2
                                                                               0.25

-------
Period
May
June
July -
August
September
Ground
Inventory
(L/m2)
0.17
0.33
0.50
0.67
0.83
Average Control
Efficiency*
(%)
0
62
68
74
80
Average Controlled
Emission Factor
(kg/VKT)
2.0
0.76
0.64
0.52
0.40
4 From Figure 13.2.2-2, ^ 10 ^m.  Zero efficiency assigned if ground inventory is less than
  0.2 L/m2 (0.05 gal/yd2).
       Newer dust suppressants are successful in controlling emissions from unpaved roads. Specific
test results for those chemicals, as well as for petroleum resins and watering, are provided in
References 18 through 21.

References For Section 13.2.2

1.     C. Cowherd, Jr., et al., Development Of Emission Factors For Fugitive Dust Sources,
       EPA-450/3-74-037, U. S. Environmental Protection Agency, Research Triangle Park, NC,
       June 1974.

2.     R. J. Dyck and J. J. Stukel, "Fugitive Dust Emissions From Trucks On Unpaved Roads",
       Environmental Science And Technology, 10(10): 1046-1048, October 1976.

3.     R. O. McCaldin and K. J. Heidel, "Paniculate Emissions From Vehicle Travel Over Unpaved
       Roads", Presented at the 71st Annual Meeting of the Air Pollution Control Association,
       Houston, TX,  June 1978.

4.     C. Cowherd, Jr,  et al., Iron And Steel Plant Open Dust  Source Fugitive Emission Evaluation,
       EPA-600/2-79-013, U. S. Environmental Protection Agency, Cincinnati, OH, May  1979.

5.     R. Bonn, et al., Fugitive Emissions From Integrated Iron And Steel Plants,
       EPA-600/2-78-050, U. S. Environmental Protection Agency, Cincinnati, OH, March 1978.

6.     Evaluation Of Open Dust Sources In The Vicinity Of Buffalo, New York,  EPA Contract
       No. 68-02-2545, Midwest Research Institute, Kansas City, MO, March 1979.

7.     C. Cowherd, Jr., and T. Cuscino, Jr., Fugitive Emissions  Evaluation,  MRI-4343-L, Midwest
       Research Institute, Kansas City, MO, February 1977.

8.     T. Cuscino, Jr.,  et al., Taconite Mining Fugitive Emissions Study, Minnesota Pollution
       Control Agency, Roseville, MN, June 1979.

9.     Improved Emission Factors For Fugitive Dust From Western Surface Coal Mining Sources,
       2 Volumes, EPA Contract No. 68-03-2924, PEDCo Environmental and Midwest Research
        Institute, Kansas City, MO, July 1981.
 1/95
Miscellaneous Sources
13.2.2-7

-------
10.     T. Cuscino, Jr., et ai.. Iron And Steel Plant Open Source Fugitive Emission Control
       Evaluation, EPA-600/2-83-110, U. S. Environmental Protection Agency, Cincinnati, OH,
       October 1983.

11.     Size Specific Emission Factors For Uncontrolled Industrial And Rural Roads, EPA Contract
       No. 68-02-3158, Midwest Research Institute, Kansas City, MO, September 1983.

12.     C. Cowherd, Jr., and P. Englehart, Size Specific Paniculate Emission Factors For Industrial
       And Rural Roads, EPA-600/7-85-038, U. S. Environmental Protection Agency, Cincinnati,
       OH, September 1985.

13.     PM-10 Emission Inventory Of Landfills In The Lake Calumet Area, EPA Contract 68-02-3891,
       Work Assignment 30, Midwest Research Institute, Kansas City, MO, September 1987.

14.     Chicago Area Paniculate Matter Emission Inventory — Sampling And Analysis, EPA Contract
       No. 68-02-4395, Work Assignment 1, Midwest Research Institute, Kansas City, MO,
       May 1988.

15.     PM-10 Emissions Inventory Data For The Maricopa And Pima Planning Areas, EPA Contract
       No. 68-02-3888, Engineering-Science, Pasadena, CA, January  1987.

16.     Oregon Fugitive Dust Emission Inventory, EPA Contract 68-DO-0123, Midwest Research
       Institute, Kansas City, MO, January 1992.

17.     Climatic Atlas Of The United States, U. S. Department Of Commerce, Washington, DC,
       June 1968.

18.     C. Cowherd, Jr. et ai., Control Of Open Fugitive Dust Sources, EPA-450/3-88-008,
       U. S. Environmental Protection Agency, Research Triangle Park, NC, September  1988.

19.     G. E. Muleski, et al., Extended Evaluation Of Unpaved Road Dust Suppressants In The Iron
       And Steel Industry, EPA-600/2-84-027, U. S.  Environmental Protection Agency, Cincinnati,
       OH, February 1984.

20.     C. Cowherd, Jr., and J. S. Kinsey, Identification, Assessment And Control Of Fugitive
       Paniculate Emissions, EPA-600/8-86-023, U. S. Environmental Protection Agency,
       Cincinnati, OH, August 1986.

21.     G. E. Muleski and C. Cowherd, Jr., Evaluation Of The Effectiveness Of Chemical Dust
       Suppressants On Unpaved Roads, EPA-600/2-87-102, U. S. Environmental Protection
       Agency, Cincinnati, OH, November 1986.
 13.2.2-8                            EMISSION FACTORS                               1/95

-------
                                   ATTACHMENT D
                      BACKGROUND INFORMATION USED IN
                  ESTIMATING EMISSIONS FROM BULLDOZING*
"U.S. Environmental Protection Agency, 1995. Compilation of Air Pollutant Emission Factors, Volume I: Stationary and Area
Sources, AP-42 Fifth Edition, U.S. Environmental Protection Agency, Research Triangle Park, NC, Section 11.9, pp. 11.9-1 to
11.9-14.

-------
11.9 Western Surface Coal Mining

11.9 General1

       There are 12 major coal fields in the western states (excluding the Pacific Coast and Alaskan
fields), as shown in Figure 11.9-1. Together, they account for more than 64 percent of the surface
minable coal  reserves in the United States.2  The 12 coal fields have varying characteristics that may
influence fugitive dust emission rates from mining operations including overburden and coal seam
thicknesses and structure, mining equipment, operating procedures, terrain, vegetation, precipitation
and surface moisture, wind speeds, and temperatures.  The operations at a typical western surface
mine are shown in Figure 11.9-2.  All operations that involve movement  of soil, coal, or equipment,
or exposure of erodible surfaces, generate some amount of fugitive dust.

       The initial operation is removal of topsoil and subsoil with large scrapers.  The topsoil is
carried by the scrapers to cover a previously mined and regraded area as  part of the reclamation
process or is  placed in temporary stockpiles. The exposed overburden, the earth that is between the
topsoil and the coal seam, is leveled, drilled, and blasted.  Then the overburden material  is removed
down to the coal seam, usually by a dragline or a shovel and truck operation.  It is placed in the
adjacent mined  cut, forming a spoils pile.  The uncovered coal seam is then drilled and blasted. A
shovel  or front  end loader loads the broken coal into haul trucks, and it is taken out of the pit along
graded haul roads to the tipple, or truck dump.  Raw coal sometimes may be dumped onto a
temporary storage pile and later rehandled by a front end loader or bulldozer.

       At the tipple, the coal is dumped into a hopper that feeds the primary crusher, then is
conveyed through additional coal preparation equipment such as secondary crushers and screens to the
storage area.  If the mine has open storage piles, the crushed coal passes  through a coal stacker onto
the pile.  The piles, usually worked by bulldozers,  are subject to wind erosion. From the storage
area, the coal is conveyed to a train loading facility and is put into rail cars. At a captive mine, coal
will go from  the storage pile to the power plant.

       During mine reclamation,  which proceeds continuously throughout the life  of the mine,
overburden spoils piles are smoothed and contoured by bulldozers.  Topsoil is placed on  the graded
spoils, and the land is prepared for revegetation by furrowing, mulching, etc.  From the time an area
is disturbed until the new vegetation emerges, all disturbed areas are subject to wind erosion.

11.9  Emissions

       Predictive emission factor equations for open dust sources at western surface coal mines are
presented in  Tables 11.9-1 and 11.9-2. Each equation is for a single dust-generating activity, such as
vehicle traffic on unpaved roads.  The predictive equation explains much of the observed variance in
emission factors by relating emissions to 3  sets of source parameters: (1) measures of source activity
or energy expended (e. g., speed and weight of a vehicle traveling on an unpaved road);
(2) properties of the material being disturbed (e. g., suspendable fines in the surface material of an
unpaved road); and (3) climate (in this case, mean  wind speed).

        The  equations may be used to estimate paniculate emissions generated per unit of source
extent (e. g., vehicle distance traveled or mass of material transferred).  The equations were
 9/88 (Reformatted 1/95)                 Mineral Products Industry                              11.9-1

-------
          COAl TYPE
          LI SUITE      E83
          SUB8ITUMINOUSC3
          BITUMINOUS   ma
                     9
                    10
                    II
                    12
                                  Coal fi«ld

                             Fort Union
                             Powder River
                             March Ceacr»l
                             Bighorn &e«ln
                             Wind lUver
                             HIM Fork
                             Ulnt»
                             Soucbv«*c«ra Utah
                             S«a Juan Klv«r
Scripp»bl«
    CIO*  COM)

      U.529
      56,?:?
  All uad«r(rouad
  A.L1
       1. 000
         30*
  AJ.1
  All under jround
       2.120
                    Figure 11.9-1.  Coal fields of the western United States.
11.9-2
                                      EMISSION FACTORS
                        (Reformatted 1/95) 9/88

-------
                                                                 *     <
                                                                                                       re
                                                                                                       o
                                                                                                       «j
                                                                                                       u
                                                                                                       CJ
                                                                                                      "^
                                                                                                       >,

                                                                                                       CO


                                                                                                       O
                                                                                                       I
                                                                                                       Ofi
9/88 (Reformatted 1/95)
Mineral Products Industry
11.9-3

-------
C/3
O
in
U4
O
ot,

Si*
O u
CJ Z
Z U
tVO

i
u
1?
1/3 *
|;








u
ha
U
as
Q
.Si
i
F
2
o
<
i
 5» ^ 01 H
">v afl ofl gTft ^ ^ ^ ^ ^ eg
&a 2( ^ ^^ .^ MJ ao »» MI 7^
j<-* jrtp- >o — o r-
— «  QQOO ^.
ior~p-f~;r~ MBZ55* Q
b b b b b bbbb Z


„
r4
^
vo>» -i,»'%» ^« _; vf IK. '?
Q ?ip <£- -S- ^=> '3''^ of* Q
bl=-5£-:S6.S6. ^,2 i €• S
aa o P 29 ®
0 K ° °'
cs



T
C1 o^
m ~ ^ "7 J^
ti sP "^^ l-_ 2^ ^ % 3e :_ =
8>rM3?io*?^2^'§2 ^"?) --2^^
| ol- s~ c,- 8- b 8 '"2
x 8
>O o
ON


c
w c c
ul 1 *
11 - - 1 1
<•>«>• U U ' M
o o o o > > o
U O U O O U


£• -3
3 U C
-o =_ 5
I -H « 1 M si i iii
111 1 H I li 1 111
S H 03 Q («^'O>~' Z<^


i2
i
~
1
II
I


•
•22
—
3
c
a.
•3
=
c. JS
3 ^. S
«3 CS u
•• " -^
2 — «
o _. <—
— . S c
II -2 «
a. « 42

W3 '3
a> S >
^J -M .
s i- ,0
*" «
VN- va Q
^ « Z
* e
i 1 e
vi: > —
= j2 
-------
                                                               Table 11.9-1 (cont.).
              s  = material silt content (%)
             u  = wind speed (m/sec)
             d  = drop height (m)
            W  = mean vehicle weight (Mg)
             S  = mean vehicle speed (kph)
             w  = mean number of wheels
             L  = road surface silt loading (g/m2)
       d Multiply the  £ 15 /im equation by this fraction to determine emissions.
       c Multiply the TSP predictive equation by this fraction to determine emissions in the  £2.5 pm size range.
       f Rating applicable to Mine Types I, II, and IV (see Tables 11.9-5 and 11.9-6).
 I
 EL
I
g.
!

-------
            Table 11.9-2 (English Units). EMISSION FACTOR EQUATIONS FOR UNCONTROLLED OPEN DUST SOURCES

                                      AT WESTERN SURFACE COAL MINES'

Operation
Blasting
Truck loading
Bulldozing


Dracline
Scraper
(travel mode)
Grading
Vehicle traffic
(light/medium duly)
Haul truck
Active storage pile
(wind erosion and
maintenance)


Material
Coal or
overburden
Coal
Coal

Overburden
Overburden







Coal

Emissions By
TSP £ 30 pm
0.0005A1 5
j , \ g
(nft J
78.4 (s^2
(M)13
5.7 (si'"2
00021 (d)11
2.7 x 10s ($)' 3 (W)2 4
0.040 (S)25

5.79
(SP
0.0067 (w)3-4 (L)° 2


1.6 u

Particle Size Range (Aerodymanic
£15 /im £10
ND 0.
0119 ' ' 0.
/j^jjO.9
18.6 fs)15 0.
(M)1-4
1 0 (c)' * 0.
0.002 Ud)07 0.
6.2 x 10 6 (s)1 4 (W)2 5 0.
0.051 (S)20 0.

3.72 0.
0.0051 (w)3 J 0.


Diameter)b>c
urn6 £2.5 jim/TSF*
52e ; ND
75 0.019
75 0.022

75 0.105
75 0.017
60 0.026
60 0.031

60 0.040
60 0.017


ND ND ND



Units
Ib/blasl
Ib/ton
Ib/lon

Ib/ton
Ib/yd3
Ib/VMT
Ib/VMT

Ib/VMT
Ib/VMT


Ib

(acrc)(hr)
EMISSION
RATING
C
B
B

B
B
A
B

B
A

.
C1

• Reference I, except for coal storage pile equation from Reference 4. TSP = total suspended paniculate. VMT = vehicle miles traveled.
ND = no data.


b TSP denotes what is measured by a standard high volume sampler (see Section 13.2).
c Symbols for equations:
A = horizontal area, with blasting depth £70 ft. Not for vertical face of a bench.
M = material moisture content (%)


m
oo
Ul
q
O
OO
oo

-------
                                                        Table 11.9-2 (cont.).
      s  =  material silt content <%)
      u  =  wind speed (m/sec)
      d  =  drop height (ft)
     W  =  mean vehicle weight (tons)
      S  =  mean vehicle speed (mph)
      w  =  mean number of wheels
      L  =  road surface silt loading (g/m2)
d Multiply the £ 15 /im equation by this fraction to determine emissions.
c Multiply the TSP predictive equation by this fraction to determine emissions in the  £2.5 urn size range.
f Rating applicable to Mine Types I, II, and IV (see Tables 11.9-5 and 11.9-6).

-------
developed through field sampling of various western surface mine types and are thus applicable to any
of the surface coal mines located in the western United States.

       In Tables 11.9-1 and 11.9-2, the assigned quality ratings apply within the ranges of source
conditions that were tested in developing the equations given in Table 11.9-3. However, the
equations should be derated 1 letter value (e. g., A to B) if applied to eastern surface coal mines.

       In using the equations to estimate emissions from sources found in a specific western surface
mine, it is necessary that reliable values for correction parameters be determined for the specific
sources of interest if the assigned quality ranges of the equations are to be applicable. For example,
actual silt content of coal or overburden measured at a facility  should be used instead of estimated
values.  In the event that site-specific values for correction parameters cannot be obtained, the
appropriate geometric mean values from Table 11.9-3 may  be used, but the assigned quality rating of
each emission factor equation should be reduced by 1 level  (e.  g., A to B).

       Emission factors for open dust sources not covered in Table 11.9-3  are in Table 11.9-4.
These factors were determined through source testing at various western coal mines.
   Table 11.9-3 (Metric And English Units).  TYPICAL VALUES FOR CORRECTION FACTORS
            APPLICABLE TO THE PREDICTIVE EMISSION FACTOR EQUATIONS*
Source
Coal loading
Bulldozers
Coal

Overburden

Dragline


Scraper


Grader

Light/Medium duty
vehicle
Haul truck


Correction Factor
Moisture

Moisture
Silt
Moisture
Silt
Drop distance
Drop distance
Moisture
Silt
Weight
Weight
Speed
Speed
Moisture
Wheels
Silt loading
Silt loading
Number Of
Test
Samples
7

3
3
8
8
19
19
7
10
15
15
7

7
29
26
26
Range
6.6 - 38

4.0 - 22.0
6.0 - 1 1 .3
2.2 - 16.8
3.8-15.1
1.5 - 30
5- 100
0.2 - 16.3
7.2 - 25.2
33 -64
36-70
8.0 - 19.0
5.0-11.8
0.9- 1.70
6.1 - 10.0
3:8 - 254
34 - 2270
Geometric
Mean
17.8

10.4
8.6
7.9
6.9
8.6
28.1
3.2
16.4
48.8
53.8
11-4
7.1
1.2
8.1
40.8
364
Units
%

%
%
%
%
m
ft
%'
%
Mg
ton
kph
mph
%
number
g/irr
Ib/acre
 a Reference 1.
 11.9-8
EMISSION FACTORS
(Reformatted 1/95) 9/88

-------
oo
oo
Table 11 9-4 (English And Metric Units). UNCONTROLLED PARTICULATE EMISSION FACTORS FOR OPEN DUST

                        SOURCES AT WESTERN SURFACE COAL MINES
Source
Drilling

Topsoil removal by tcraper

Overburden replacement
Truck loading by power shovel (batch drop)6
Train loading (batch or continuous drop)0

Bottom dump truck unloading (batch drop)0



Material
Overburden
Coal
Topsoil

Overburden
Overburden
Coal

Overburden
Coal


Mine
Location*
Any
V
Any
IV
Any
V
Any
III
V
IV
III
11
TSP
Emission
Factor6
1.3
0.59
0.22
0.10
0.058
0.029
0.44
0.22
0.012
0.0060
0.037
0.018
0.028
0.014
0.0002
0.0001
0.002
0.001
0.027
0.014
0.005
0.002
0.020
0.010
Units
Ib/hole
kg/hole
Ib/hole
kg/hole
Ib/ton
kg/Mg
Ib/lon
kg/Mg
Ib/ton
kg/Mg
Ib/ton
kg/Mg
Ib/ton
kg/Mg
Ib/ton
kg/Mg
Ib/ton
kg/ton
Ib/ton
kg/Mg
Ib/ton
kg/Mg
Ib/ton
kg/Mg
EMISSION
FACTOR
RATING
B
B
E
E
E
E
D
D
C
C
C
C
D
D
D
D
E
E
E
E
E
E
E
E
VD

-------
VO
 I
o
                                                         Table 11.9-4 (cont.).
Source



End dump (nick unloading (batch drop)c
Scraper unloading (batch drop)0
Wind erosion of exposed areas

Material



Coal
Topsoil
Seeded land, stripped
overburden, graded overburden

Mine
Location*
I

Any
V
IV
Any

TSP
Emission
Faclorb
0.014
0.0070
0.066
0.033
0.007
0.004
0.04
0.02
0.38
0.85
Units
Ib/T
kg/Mg
Ib/T
kg/Mg
Ib/T
kg/Mg
Ib/T
kg/Mg
T
(acrc)(yr)
Me
(heclareXyr)
EMISSION
FACTOR
RATING
D
D
D
D
E
E
C
C
C
C
m
2
CO
CO
•—•
O
Z
O
50
CO
* Roman numerals I through V refer to specific mine locations for which the corresponding emission factors were developed.
  Tables 11.9-4 and  11.9-5 present characteristics of each of these mines.  See text for correct use of these "mine-specific" emission
  factors. The other factors (from Reference 5 except for overburden drilling from Reference 1) can be applied to any western surface coal
  mine.
b Total suspended paniculate (TSP) denotes what is measured by a standard high volume sampler (see Section  13.2).
c Predictive emission factor equations, which generally provide more accurate estimates of emissions, are presented  is Chapter 13.
oo
oo

-------
       The factors in Table 11.9-4 for mine locations I through V were developed for specific
geographical areas.  Tables 11.9-5 and 11.9-6 present characteristics of each of these mines (areas).
A "mine-specific" emission factor should be used only if the characteristics of the mine for which an
emissions estimate is needed are very similar to those of the mine for which the emission factor was
developed.  The other (nonspecific) emission factors were developed at a variety of mine types and
thus are applicable to any western surface coal mine.

       As an alternative to the  single valued emission factors given in Table 11.9-4 for train or truck
loading and for truck or scraper unloading,  2  empirically derived emission factor equations are
presented in Section 13.2.4 of this document.   Each equation was developed for a source operation
(i. e., batch drop and continuous drop, respectively) comprising a single dust-generating mechanism
that crosses industry lines.

       Because the predictive equations allow emission factor adjustment to specific source
conditions, the equations should be used in  place of the factors in Table 11.9-4 for the sources
identified above if emission estimates for a  specific western surface coal mine are needed.  However,
the generally higher quality ratings assigned to the equations are applicable only if:  (1) reliable
values of correction  parameters  have been determined for the specific sources of interest, and (2) the
correction parameter values lie within the ranges tested in developing the equations.  Table 11.9-3
lists measured  properties of aggregate materials that can be used to estimate correction parameter
values for the predictive emission factor equations in Chapter 13, in the event that site-specific values
are not available.  Use of mean correction parameter values from Table 11.9-3 will reduce the quality
ratings of the emission factor equations in Chapter 13 by  1 level.
 9/88 (Reformatted i/95)                 Mineral Products Industry                              11.9-11

-------
o
                  Table 11.9-5 (Metric And English Units). GENERAL CHARACTERISTICS OF SURFACE COAL MINES
                                             REFERRED TO IN TABLE 11.9-4"
Mine
I
11
III
IV
V
Location
N.W. Colorado
S.W. Wyoming
S.E. Montana
Central North Dakota
N.E. Wyoming
Type Of Coil
Mined
Subbitum.
Subbilum.
Subbitum.
Lignite
Subbilum.
Terrain
Moderately
steep
Semimgged
Gently roiling
(o semirugged
Gently rolling
Flat to gently rolling
Vegetative
Cover
Moderate,
sagebrush
Sparse,
sagebrush
Sparse,
moderate,
prairie
grassland
Moderate,
prairie
grassland
Sparse,
sagebrush
Surface Soil Type
And Erodibility
Index
Clayey loamy (71)
Arid soil with clay
and alkali or
carbonate
accumulation (86)
Shallow clay loamy
deposit* on bedrock
(47)
Loamy, loamy to
sandy (71)
Loamy, sandy,
clayey, and clay
loamy (102)
Mean Wind
Speed
m/s mph
2.3 5.1
6.0 13.4
4.8 10.7
5.0 11.2
6.0 13.4
Mean Annual
Precipitation
cm in.
38 15
36 14
28-41 11-16
•
43 17
36 14
m
S
*—i
in
CO
«••*
O
z
•n
         Reference 4.
§>
00
oo

-------
    Table 11.9-6 (English Units). OPERATING CHARACTERISTICS OF THE COAL MINES
                            REFERRED TO IN TABLE 1
Parameter
Production rate
Coal transport
Strati graphic
data





Coal analysis
data



Surface
disposition





Storage
Blasting



Required Information
Coal mined
Avg. unit train frequency
Overburden thickness
Overburden density
Coal seam thicknesses
Parting thicknesses
Spoils bulking factor
Active pit depth
Moisture
Ash
Sulfur
Heat content
Total disturbed land
Active pit
Spoils
Reclaimed
Barren land
Associated disturbances
Capacity
Frequency, total
Frequency, overburden
Area blasted, coal
Area blasted, overburden
Units
K^ton/yr
per day
ft
Ib/yd3
ft
ft
%
ft
%
%, wet
%, wet
Btu/Ib
acre
acre
acre
acre
acre
acre
ton
per week
per week
ft2
ft2

I
1.13
NA
21
4000
9,35
50
22-
52
10
8
0.46
11000
168
34
57
100
—
12
NA
4
3
16000
20000

11
5.0
NA
80
3705
15,9
15
24
100
18
10
0.59
9632
1030
202
326
221
30
186
NA
4
0.5
40000
—
Mine
ra
9.5
i
90
3000
27
NA
25
114
24
8
0.75
8628
2112
87
144
950
455
476
—
3
3
—
—

TV
3.8
NA
65
—
2,4,8
32,16
20
80
38
7
0.65
8500
1975
—
—
—
—
—
NA
7
NA
30000
NA

V
12.06
2
35
—
70
NA
—
105
30
6
0.48
8020
217
71
100
100
—
46
48000
A
7b
—
—
a Reference 4.
b Estimate.
NA = not applicable.  Dash = no data.
References For Section 11.9

1.      K. Axetell and C. Cowherd, Improved Emission Factors For Fugitive Dust From Western
       Surface Coal Mining Sources, 2 Volumes, EPA Contract No. 68-03-2924, U. S.
       Environmental Protection Agency, Cincinnati, OH, July 1981.
9/88 (Reformatted 1/95)
                   Mineral Products Industry
11.9-13

-------
2.      Reserve Base OfU. S. Coals By Sulfur Content: Pan 2, The Western States, IC8693, Bureau
       Of Mines, U. S. Department Of The Interior, Washington, DC,  1975.

3.      Bituminous Coal And Lignite Production And Mine Operations - 1978, DOE/EIA-0118(78),
       U. S. Department of Energy, Washington, DC, June 1980.

4.      K. Axetell, Survey Of Fugitive Dust From Coal Mines, EPA-908/1-78-003, U. S.
       Environmental Protection Agency, Denver, CO, February 1978.

5.      D. L. Shearer, et al., Coal Mining Emission Factor Development And Modeling Study, Amax
       Coal Company, Carter Mining Company, Sunoco Energy Development Company, Mobil Oil
       Corporation, and Atlantic Richfield Company,  Denver, CO, July 1981.
 11.9-14                           EMISSION FACTORS                 reefo™*^ i/9S> 9/88

-------
              Appendix G

Comparison of Previous and Current Modeling of
    Particulate Matter at Two Cement Plants

-------
                                  APPENDIX G
  Comparison of Previous and Current Modeling of Particulate Matter
                            at Two Cement  Plants
       In support of cement kiln dust (CKD) regulatory activities, EPA has, at two
different times, conducted modeling to predict particulate matter (PM) concentrations at
cement manufacturing plants to assess the risks posed by fine particulates. Both
approaches used fate and transport modeling to quantitatively estimate the potential
effects associated with current CKD management practices on-site at cement plants, but
each followed different methods, used different parameters, and yielded different results.
The most recent round of modeling, completed initially in October 1996, predicted PM
concentrations for two facilities: Rinker (Miami, FL) and Lafarge (Fredonia, KS). The
previous analysis, reported in August, 1994,1 also modeled the PM concentrations for these
two facilities plus three others:  Ash Grove (Chanute, KS), Southdown (Lyons, CO), and
Giant (Harleyville, SC).  Since Rinker and Lafarge have been modeled twice, the purpose of
this appendix is to identify and  explain the differences between the two sets of results and
the different modeling approaches.


Results from the Two Modeling Exercises

       The new modeling exercise provided a more complete profile of PM concentrations
at the two facilities than the previous exercise. The new study predicted concentrations for
two sizes of PM (i.e., PM10 and PM25) over two different averaging periods (i.e., 24-hour and
annual).  The previous effort only predicted PM,0 concentrations for the annual averaging
period. The previous effort did, however, generate both a "best estimate" and an "upper
bound" exposure concentration for each receptor point at each facility, based on best
estimate and more conservative characterizations of the management and environmental
parameters contributing most to the model results. Only "best estimates" were generated
by the  new study. Accordingly,  the results of the new study and the previous study overlap
only for the "best estimate" annual PM10 concentrations. These two sets of results are
presented in Exhibit G-l.

       Before drawing conclusions about which results are higher or lower, it is necessary
to account for differences in (among other factors discussed in the remainder of this
appendix) the modeled distance between the emission source and the exposure point. The
previous study estimated concentrations at the nearest property boundary and the nearest
residence. The new study, in contrast, estimated concentrations at points defined by a
fixed grid around the facilities, which do not necessarily correspond to the points modeled
previously. For the purpose of this comparison, EPA selected results from the new
1 Technical Background Document for the Notice of Data Availability on Cement Kiln Dust: Human
Health and Environmental Risk Assessment in Support of the Regulatory Determination on Cement
Kiln Dust. EPA Office of Solid Waste, August 31, 1994.
                                     Page G-2

-------
modeling that correspond most closely to the distances modeled previously, but there are
differences which contribute to the varied results.
                                   Exhibit G-l
                          Comparison of Model Results
Facility:
Current Model

Previous Model

Annual PM]0 Concentration (p.g/m3)
Rinker (Miami, FL)
0.4-0.5
1-6
Lafarge (Fredonia, KS)
40-43
40-50
      In the case of Rinker, the distances modeled are quite similar: 701 and 1,069 meters
in the old analysis compared to 605 and 927 meters in the new analysis. Based on these
similarities, it is reasonable to compare the results directly. This leads to the overall
conclusion that the new modeling at Rinker predicts lower concentrations (by about an
order of magnitude) than the old modeling.

      In the case of Lafarge, the distances modeled in the new study (880 and 1,145
meters) are approximately twice as far as those modeled in the old study (457 and 488
meters). The predicted concentrations from the two different exercises, however, are
almost equivalent. Because concentrations are expected to be higher at points closer to a
ground-level source, such as a CKD pile, it can be inferred that the new results would be
higher than 40-43 g/m3 - and thus higher than the concentrations predicted previously - if
the new model were re-run for the closer distance modeled previously. This leads to the
overall conclusion that the new modeling at Lafarge predicts higher concentrations than
the old modeling.


Discussion of Main Factors and  Direction of Impact

      To explore why the new results for Rinker are lower and the new results for Lafarge
are higher, the main differences between the two modeling exercises are discussed below.
While the precise magnitude of effect that each of these modeling differences has on the
results is beyond the scope of this brief review, a general estimation of the direction  of
impact can be predicted in most cases (i.e., would the difference tend to make the new
model's results higher or lower than the previous model's results). First, differences in the
basic models used are  discussed. Next, the specific modeling parameters, how they differ
between the two approaches, and the direction of impact of each factor are presented.
                                      Page G-3

-------
       Models

       The two studies used different models.  These differences are described below first
for the emissions models (which estimate the amount of PM that escapes from a source)
and then for the dispersion models (which estimate the change in air concentration from
source to receptor).

       Emissions Models

       The new study estimated emissions using emission factors and equations found in
AP-42.2 The methods presented in AP-42 for estimating fugitive dust emissions are
principally compiled from a 1988 reference.3 This is the approach that EPA's Office of Air
Quality Planning and Standards recommends  today. It is the most up-to-date and best
approach for estimating emissions, short of substantially more complicated modeling or
new field studies.

       In the previous effort, MMSOILS was used to estimate releases to the atmosphere
from CKD piles. MMSOILS is a screening-level multimedia model developed by EPA's
Office of Research and Development (ORD) to  simulate the release of hazardous
constituents from a wide variety of waste management units and their subsequent
transport through different environmental pathways.4  The air emissions component of
MMSOILS consists of equations presented in a 1985 rapid assessment methodology for
estimating potential atmospheric contamination and resulting inhalation exposure of
people living near abandoned hazardous waste sites.5 The particular equation used in the
previous CKD modeling exercise, which was developed  from field measurements of highly
erodible soils, estimates PM10 emissions as a function of certain field and climatic factors,
including the fraction of the surface covered with vegetation and the wind speed.  This
equation is older and different than the equations recommended in AP-42.

       EPA has not systematically compared these two approaches for estimating
emissions or compared results from the two approaches under a common set of
assumptions. Therefore, the Agency recognizes the use of different emission models as a
source of variation between the two sets of results, but  the direction and magnitude of the
impact is presently unknown.
2 Compilation of Air Pollution Emission Factors, AP-42, Fifth Edition. EPA, Research Triangle Park,
North Carolina, January 1995.

3 Control of Open Fugitive Dust Sources, Final Report.  Prepared by C. Cowherd et al. of Midwest
Research Institute for EPA Office of Air Quality Planning and Standards, EPA-450/3-88-008,
September 1988.

4 MMSOILS: Multimedia Contaminant Fate, Transport, and Exposure Model, Documentation and
User's Manual. EPA Office of Research and Development, September 1992 (updated in April 1993).

5 Rapid Assessment of Exposure to Particulate Emissions from Surface Contamination Sites. EPA
Office of Health and Environmental Assessment, EPA/600/8-85/002, February 1985.
                                      Page G-4

-------
      Dispersion Models

      The new study used the Industrial Source, Complex 3-Short Term (ISC3ST)
dispersion model. This model is recommended in the EPA Guidelines on Air Quality
Models for dispersion modeling of complex industrial source facilities.  The ISC3ST model
is a significant improvement over the model used in the previous study.  For example,
ISC3ST can discriminate between airborne particulate concentrations that are due to
emissions from the CKD pile versus those from the CKD handling train. Also, the ISC3ST
model can incorporate hourly wind speeds into calculations of dispersion, uses as input the
full array of meteorological data, and handles terrain effects (e.g., the model accounts for
the difference in elevation between the source and various receptor points).

      The previous study differed in terms of both the model and inputs used. The model
itself, a component of MMSOILS, was a sector-averaged form of a Gaussian plume model.
It did not account for particulate emissions from the handling train, terrain effects, and
other complexities considered in the more recent study. Similarly, rather than run the
model using a complete stability array of meteorological data, a simplified set of
conservative atmospheric transport parameters was assumed based on recommendations
from ORD. These assumptions included a stability class of E, a wind speed of 3
meters/second, and a frequency of wind blowing in any one direction of 30 percent.

      As for the emission models discussed above, the Agency has not systematically
compared these two dispersion model approaches to know which one yields higher or lower
results, and by what margin. However, the new study used a model and set of inputs that
are believed to be more sophisticated and more accurate than the previous study.


      Specific Model Factors

      Emissions and dispersion factors used in the two analyses are discussed below.  By
"factor" EPA means inputs to the models, modeling approaches, or underlying assumptions
of the models.  The primary factors are listed and described below, along with predictions
of the direction of influence of each factor on estimated PM concentrations.

      Emissions Factors

      Source of PM. The new modeling predicted PM emissions from both the CKD pile
(through wind erosion and vehicle disturbance) and the handling train (including initial
loading and unloading, interim storage, and transport to the  CKD pile).  The previous
model was limited to emissions from the pile, but included both wind erosion from the pile
and particulate emissions resulting from vehicular disturbance from trucks delivering
CKD to the pile and from the associated spreading operations.
      Direction:  The new modeling would be expected to estimate greater PM emissions
because it considers the handling train in addition to the other sources evaluated
previously (wind erosion and vehicle disturbance at the pile). In particular, fugitive dust
kicked up from roads used to transport CKD to the disposal pile was found in the new
study to add significantly to the PM emissions. At Lafarge, fugitive dust from the road was
                                      Page G-5

-------
estimated to contribute 88 percent of the emissions, whereas the other CKD handling
stages contributed 2 percent and wind erosion from the CKD pile contributed 10 percent.

      Covering the pile.  For Lafarge, the new study accounted for the fact that only
8,600 m2 (7.5 percent) of the 114,000 m2 pile is uncovered (the rest has been covered with
soil and seeded).  The previous study modeled the entire pile as being exposed and.
contributing to PM emissions. (Both analyses treated the pile at Rinker as completely
uncovered.)
      Direction: The new estimate of PM]0 emissions at Lafarge would be more realistic
but lower than the previous estimate. (Not applicable for Rinker.)

      Active pile surface. The new study estimated continual emissions from only the
active part of the CKD disposal pile, where fresh or disturbed CKD without a surface crust
is highly susceptible to wind erosion. Emissions from inactive areas of the pile were
simulated until erodible particles on the surface were blown off and a surface crust forms
and prevents further emissions. In contrast,  the previous study treated the entire surface
of the pile as an unlimited supply of erodible particles.
      Direction: All other factors being equal, this aspect of the new study results in
lower emission estimates  than the previous study.

      Dust suppression.  For the new study, EPA identified the type of dust suppression
used at each facility and reduced the predicted PM emissions according to the estimated
effectiveness of the suppression technique. In particular, at Lafarge EPA assigned a
50 percent emissions reduction during CKD handling due to water addition (nodulization)
at the plant prior to transport of CKD to the pile.  At Rinker EPA did not identify any
emissions controls, so no credit was given. The previous study did not account for any dust
suppression.
      Direction: Giving credit for emission controls at Lafarge in the new study would
tend to reduce emissions estimates compared to the previous study. No such differences
would be expected for Rinker since no emissions reduction credits were assigned in the new
study or the previous study.

      Daily load of CKD  to the disposal pile. The new modeling computed the daily load
by dividing the quantity of CKD wasted annually by the number of operating days in a
year (for Rinker, 340 working days per year; for Lafarge, 339 working days per year).  The
previous study used the same approach but set the number of working days per year at 300
for all facilities.
      Direction: This factor works in two ways.  First, the larger number of working days
used in the new study results in lower daily loads and, thus, lower daily emissions than the
previous study. This difference, however, does not affect emissions over the course of an
entire year, which is the relevant period for comparison of the results in Exhibit G-l.
Second, more working days  per year results in more frequent trips to the pile and more
frequent disturbances of the pile surface.  This second effect results in the new study
estimating higher annual emissions than before.

      Truck capacity.  The initial new modeling for Rinker and Lafarge assumed a truck
can carry 36 tons of CKD, which is the loaded (truck plus payload) weight of a typical
dump truck used in hauling operations, according to truck brochures.  (Note that this value
                                      Page G-6

-------
was refined in subsequent stages of the new modeling, as described in Section 3.2.5 of the
main body of this report.) The previous study assumed a capacity of 80 tons of CKD per
truckload.
      Direction: The smaller truck capacity in the new study results in more trips
occurring between the plant and the pile and more frequent disturbances of the pile
surface, which result in higher emissions estimates than the previous modeling.

      Particle size. In the new study, EPA used particle size data developed by the
Portland Cement Association for CKD from three different kinds of kilns (long wet rotary,
long dry rotary, and precalciner system).6 The type of kiln actually used at Rinker and
Lafarge was then used to select the most representative particle size distribution.  In the
previous study, engineering judgment was used to  estimate a mean particle size, based on
available data for CKD-like material (fine silt).
      Direction: The new modeling presumably provides more accurate results than the
previous modeling; however, it is unclear in which  direction (i.e., higher or lower
concentration) the results would be influenced.

      Dispersion Factors

      Meteorological parameters.  As mentioned previously, the new modeling used the
full set of stability array data from the meteorological station nearest each facility. A
complete set of stability array data consists of a joint frequency distribution of twelve wind
directions, five wind speed categories, and six Pasquill-Gifford stability classes, resulting
in a matrix of 360 entries with unique frequencies  of occurrence. Due to the nature of the
previous screening analysis, a complete set of meteorological data was not used in the
atmospheric dispersion modeling. Instead, a simplified set of conservative atmospheric
transport parameters was used based on recommendations from EPA's ORD. Key
components of the meteorological array are discussed and compared below.

•  Wind direction: The new dispersion modeling used actual wind direction  data from the
   nearest meteorological station.  The previous modeling assigned a 30 percent
   probability that the wind will be in any given direction.

•  Wind speed: Again, the new modeling used actual wind speed measurements from the
   nearest meteorological station.  The previous modeling assumed a wind speed of
   3 meters/second, which was recommended by EPA's ORD as a reasonably slow wind not
   likely to result in significant dispersion.

•  Stability class:  The new modeling used data from the closest meteorological station to
   determine the stability class (the range of possible classes is A-F). The previous
   modeling conservatively used the "slightly stable" stability class (class E).

      In each of these cases, the site-specific data used in the new modeling result in a
more accurate characterization of dispersion conditions than before.  Because the assumed
6 Cement Kiln Dust Management Permeability.  By H. Todres et al. for the Portland Cement
Association. PCA Research and Development Bulletin RD103T, 1992.
                                      Page G-7

-------
set of conditions used in the previous modeling was purposefully selected to be conservative
(i.e., estimate low dispersion), the values used previously likely result in higher PM
concentrations than the values used in the new modeling.

      Terrain. The new modeling used site-specific terrain features as a model input.
The previous modeling did not include any adjustments for terrain.
      Direction:  The effects of terrain can be complex; the direction of impact on the new
modeling results is unknown.


Conclusion

      Results from the new modeling exercise should be more accurate than the previous
modeling results.  The emissions and dispersion models used in the new study are more
sophisticated, the CKD handling and disposal practices at each site were characterized
more thoroughly,  and site-specific meteorological and terrain data were used as model
inputs instead of conservative default assumptions. All of these changes are improvements
over the earlier approach.

      On a general level, the two exercises yield the same conclusion. The annual PM10
concentrations predicted at the closest modeling points, in both exercises, do not exceed the
corresponding National Ambient Air Quality Standard of 50 ug/m3.

      There are so many differences in approach in the two studies, with each likely to
affect the results to a different degree and in a different direction, that it is difficult to
precisely explain the varied results.  Overall, one might expect the new PM concentrations
to be lower than those estimated previously, using the simpler and generally more
conservative approach. This is the case for the Rinker results. The  new PM concentrations
predicted at Lafarge, however, are higher than before, when differences in the modeling
distance in the two studies are taken into account. This result for Lafarge would appear
inconsistent, especially given the fact that some of the changes in the new study (covering
the pile, active pile surface, dust suppression) would tend to drive the estimated
concentrations down from before. The principal offsetting change, which would tend to
increase predicted concentrations, is the added emissions from the CKD handling train
considered in the  new study. In particular, fugitive dust from the road used to haul CKD
from the plant to the pile at Lafarge was estimated to be a significant source of emissions
in the new study.
                                      Page G-8

-------
These levels are intended to assure that high end risk levels do not exceed 1E-05 or HQs of 1 for
any constituent regardless of the agricultural practices employed. These values are estimated by
assuming all agricultural practice parameters are high end and estimating the concentration
required to reach the desired risk or HQ levels. The estimated values were then inserted into the
deterministic risk analysis with all the agricultural practice parameters assumed to be high end.
All exposure parameters are varied singly or doubly in the remainder of the deterministic
analysis. The resulting risk and HQ values are compared to the target values and the constituent
concentrations adjusted  to determine the limiting values.

       This process is straight forward for the metal constituents.  However, for dioxins it is not
appropriate to set a limiting value for for each congener independently. The risk for dioxin
congeners are determined using the TEF methodology and although congener concentrations and
risk are estimated individually, the risks from individual congeners are summed to produce
single TEF risk value for all dioxins and furans. The limiting concentration of dioxin congeners
is estimated using the congener showing the highest risk (1,2,3,4,7,8-hexchlorodibenzodioxin) to
set the limiting value for the TEF risk in the same way it is done for metals. The ratio of this
estimated concentration  to the highest measured value for that congener was found to be 4.5.
The factor of 4.5 was then applied to the 95th percentile of the measured concentration of all
congeners.  The resulting concentrations were then used as the limiting values in the risk
analysis.

-------