EPA/600/R-18/074 | April 2018
www.epa.gov/homelarid-security-research
United States
Environmental Protection
Agency
&EPA
Exposure Assessment of Livestock
Carcass Management Options During a
Foreign Animal Disease Outbreak
Office of Research and Development
Homeland Security Research Program
-------
This page left intentionally blank
-------
EPA/600/R-18/074
April 2018
Exposure Assessment of
Livestock Carcass Management Options During
a Foreign Animal Disease Outbreak
by
Sandip Chattopadhyay, Ph.D.
Threat and Consequence Assessment Division
National Homeland Security Research Center
Cincinnati, OH 45268
Interagency Agreement HSHQPM13X00157
Contract No. EP-C-14-001 to ICF under WA 24
U.S. Environmental Protection Agency Project Officer
Office of Research and Development
Homeland Security Research Program
Cincinnati, OH 45268
-------
Disclaimer
The U.S. Environmental Protection Agency (EPA) through its Office of Research and
Development), in collaboration with the United States Department of Homeland Security funded
and managed the research described here under Interagency Agreement HSHQPM13X00157 and
contract No. EP-C-14-001 to ICF under WA 24. It has been subjected to the Agency's review
and has been approved for publication. Note that approval does not signify that the contents
necessarily reflect the views of the Agency. Numeric results in this assessment should not be
interpreted as "actual" risks. No official endorsement should be inferred. Any mention of trade
names, products, or services does not imply an endorsement by the U.S. Government or EPA.
The EPA does not endorse any commercial products, services, or enterprises.
Questions concerning this document or its application should be addressed to:
Sandip Chattopadhyay, Ph.D.
National Homeland Security Research Center
Office of Research and Development
U.S. Environmental Protection Agency
26 W. Martin Luther King Drive, MS NG16
Cincinnati, OH 45268
Phone: 513-569-7549
Fax: 513-487-2555
E-mail: chattopadhyay.sandip@epa.gov
-------
Table of Contents
Disclaimer ii
List of Tables v
List of Figures vi
Acronyms and Abbreviations vii
Acknowledgements ix
Executive Summary x
1. Introduction 1
1.1 Purpose and Scope 1
1.2 Report Organization 2
2. Problem Formulation 3
2.1 Foot-and-Mouth Disease Virus 5
2.1.1 Measurement of Viruses and Infective Dose 5
2.1.2 Survival and Biological Decay 7
2.2 Livestock Carcass Management Options and Assumptions 9
2.2.1 Carcass Handling 10
2.2.2 Temporary Carcass Storage 11
2.2.3 Transportation 12
2.3 Exposure Assessment Assumptions 12
2.4 Sources of FMDv Releases and Exposure Pathways 14
2.4.1 Temporary Carcass Storage before Transportation 15
2.4.2 On-site Burial 16
2.4.3 Summary of Exposure Pathways for Livestock 17
3. Exposure Estimation 18
3.1 Estimation of FMDv Releases 18
3.1.1 Air 18
3.1.2 Subsurface Soil 20
3.2 Fate and Exposure Estimation Methods 22
iii
-------
3.2.1 Air Dispersion Modeling 22
3.2.2 Concentrations of FMD in Surface Soil 23
3.2.3 Concentrations of FMD in Feed 24
3.2.4 Concentrations in Ground Water 25
3.2.5 Cattle Exposure Factor Values 34
3.2.6 Exposure Estimation 36
3.3 Vectorborne Transmission 38
4. Results and Discussion 42
4.1 Qualitative Exposure Assessment 42
4.2 Base Case Exposure Assessment for FMD 43
4.2.1 Air 44
4.2.2 Ground Water Ingestion 48
4.3 Uncertainty Analysis 51
4.3.1 Uncertainty Analysis for Air Exposure Pathways 51
4.3.2 Uncertainty Analysis for the Ground Water Exposure Pathway 59
4.4 Uncertainty Summary and Research Needs 69
4.5 Summary of Findings 80
5. Secondary Data 84
6. Literature Cited 84
iv
-------
t of Tables
Table ES- 1. Ranking of Off-site Livestock Carcass Management Options for Microbes xi
Table ES- 2. Ranking of On-site Livestock Carcass Management Options for an FMD Outbreak xi
Table 2-1. Foot and Mouth Disease Minimum Infectious Doses and Mode of Transmission 7
Table 2-2. General Survival of Foot and Mouth Disease Virus (FMDv) 8
Table 2-3. Livestock Carcass Management Options Considered for the Exposure Assessment 9
Table 2-4. Containment of Chemical and Microbial Releases from Management Options 10
Table 2-5. Foreign Animal Disease Outbreak Scenario Assumptions3 13
Table 2-6. Livestock Exposure Pathways for Livestock Carcass Management 17
Table 3-1. Estimated risk of FMDv Breakthrough to Ground Water at Soil Depths of 1-8 m 28
Table 3-2. Average Downward Travel Velocities and Time to 1 m Depth in Unsaturated Soils 31
Table 3-3. Estimated Concentrations of Infective FMDv in the Ground Water Pathway for the Base-case
34
Table 3-4. Summary of Exposure Factor Values for Cattle3 36
Table 4-1. Base-Case Estimates of Inhalation Exposure for Dairy Cattle 47
Table 4-2. Forage and Soil Ingestion Exposure Results 48
Table 4-3. Base Case Estimates of Water Ingestion Exposure 50
Table 4-4. Uncertainty Analyses for Air Exposure Pathways3 53
Table 4-5. Uncertainty Analyses for the Ground Water Exposure Pathway for Temporary Storage Pile3.61
Table 4-6. Dilution Attenuation Factors by Area of Storage Pile and Number of Carcasses 65
Table 4-7. Moderate to High Natural Variation in Parameter Values—Potential Bias from Selected Values
70
Table 4-8. Uncertainty in Parameter Value(s) Selected 74
Table 4-9. Simplifying Assumptions—Effects on Exposure Estimates 76
Table 4-10. Research Needs for Livestock Carcass Management Options and Activities 78
Table 4-11. Ranking of On-site Livestock Carcass Management Options for an FMD Outbreak 82
Table 4-12. Ranking of Off-site Livestock Carcass Management Options for Microbes 82
-------
List of Figures
Figure 2-1. Conceptual model for exposure pathways from temporary carcass storage 15
Figure 2-2. Conceptual model for exposure pathways from on-site burial of livestock carcasses 16
Figure 3-1. Number of Monte Carlo simulations (out of 10 million) that failed to reach attenuation target
at four soil depths 28
Figure 4-1. AERMOD receptor locations and highest 1-hour FMDv concentrations 46
Figure 4-2. Uncertainty analysis for particle emission rates to air, inhalation exposure relative to base
case, for dairy cattle with distance from the storage pile 54
Figure 4-3. Uncertainty analysis for the number of carcasses, inhalation exposure for dairy cattle relative
to the base case, with distance from the storage pile 55
Figure 4-4. Uncertainty analysis for the biological decay rate, inhalation exposure for dairy cattle relative
to the base case, by distance from the storage pile 56
Figure 4-5. Uncertainty analysis for particle emissions to air, ingestion exposure for dairy cattle relative to
the base case at 100 m from the storage pile 57
Figure 4-6. Uncertainty analysis for number of carcasses, ingestion exposure for dairy cattle relative to
the base case at 100 m from the storage pile 58
Figure 4-7. Uncertainty analysis for the biological decay rate, ingestion exposure for dairy cattle relative
to the base case at 100 m from the storage pile 59
Figure 4-8. Uncertainty analysis for the particle release rate, water ingestion exposure for dairy cattle by
depth of sand, relative to base case exposure with 1 m silty loam 62
Figure 4-9. Uncertainty Analysis for the Particle Release Rate, Water Ingestion Exposure for Dairy Cattle
by Depth of Silty Loam, Relative to Base Case Exposure with 1 m Depth 63
Figure 4-10. Uncertainty analysis for the particle release rate, water ingestion exposure for dairy cattle by
depth of clay, relative to base case exposure with 1 m silty loam 64
Figure 4-11. Uncertainty analysis for the number of carcasses, water ingestion exposure for dairy cattle by
soil depth, relative to exposure with 100 carcasses and silty loam 66
Figure 4-12. Uncertainty analysis for the biological decay rate, water ingestion exposure for dairy cattle
by depth of sand, relative to exposure with 1 m silty loam 67
Figure 4-13. Uncertainty analysis for the biological decay rate, water ingestion exposure for dairy cattle
by depth of silty loam, relative to exposure with 1 m silty loam 68
Figure 4-14. Uncertainty analysis for the biological decay rate, water ingestion exposure for dairy cattle
by depth of clay relative to exposure with 1 m silty loam 69
VI
-------
Acronyms and Abbreviations
Acronym/Abbreviation
Stands For (Country or Agency Affiliation)
AERMOD
AMS/USEPA Regulatory Model Improvement Committee Model (air
dispersion model)
AMS
American Meteorological Society
APHIS
Animal and Plant Health Inspection Service (USDA)
°C
degrees Celsius
cm
centimeter(s)
DAF
Dilution Attenuation Factors
DHS
Department of Homeland Security (U.S.)
DW
dry weight
EPACMTP
USEPA Composite Model for Leachate Migration with Transformation
Products
°F
degrees Fahrenheit
FAD
foreign animal disease
ft
foot (feet)
FAD
Foreign Animal Disease
FMD
foot and mouth disease
FMDv
foot and mouth disease virus
g
gram(s)
h
hour(s)
HHRAP
Human Health Risk Assessment Protocol (USEPA)
ID50
infectious dose causing illness in 50 percent of the exposed population
Kd
soil/liquid partition coefficient
kg
kilogram(s)
km
kilometer(s)
L
liter(s)
m
meter(s)
m3
cubic meter(s)
mL
milliliter(s)
NHSRC
National Homeland Security Research Center (USEPA)
PFU
plaque-forming units
PPE
personal protective equipment
SOP
standard operating procedure
TCID50
50 percent tissue-culture infectious-dose
(im
micrometer(s)
UM CAHFS
University of Minnesota's Center for Animal Health and Food Safety
U.S.
United States (adjective)
USD A
United States Department of Agriculture
vii
-------
Acronym/Abbreviation
Stands For (Country or Agency Affiliation)
USEPA
United States Environmental Protection Agency
uv
ultra violet
Vlll
-------
dgements
The following individuals and organization have been acknowledged for their contributions
towards the development and/or review of this document.
United States Environmental Protection Agency (EPA)
Office of Research and Development, National Homeland Security Research Center (NHSRC)
Sandip Chattopadhyay, Ph.D. (Principal Investigator)
Sarah Taft, Ph.D.
Paul Lemieux, Ph.D.
United States Department of Agriculture (USD A)
Animal and Plant Health Inspection Service (APHIS)
Lori P. Miller, P.E.
United States Department of Homeland Security (DHS)
Science and Technology Directorate
Chemical and Biological Defense Division
Michelle M. Colby, D.V.M., M.S.
Acknowledgements are extended to reviewers who provided many helpful comments on the draft
report, including: Dr. Kevin Garrahan, USEPA; and Dr. Mohamed Hantush, USEPA. Marti
Sinclair (CSRA) is acknowledged for technical editing; and quality assurance reviewer Eletha
Brady-Roberts (USEPA) is acknowledged for contributions to this report.
IX
-------
Executive Summary
As a product of the collaborative research between the U.S. Environmental Protection Agency's
(USEPA's) Office of Research and Development and the U.S. Department of Agriculture's
Animal and Plant Health Inspection Service (APHIS), this report evaluates livestock carcass
management options following a foreign animal disease outbreak. This assessment helps to
inform a scientifically-based selection of environmentally protective methods in the event of an
outbreak.
The foreign animal disease selected for this assessment is foot and mouth disease. The foot-and-
mouth disease virus (FMDv) infects and is transmitted by livestock including cattle, swine, and
goats. FMDv does not typically infect humans and is not considered a threat to public health.
Healthy livestock can become infected by inhaling or ingesting infective FMDv released from
live animals or the carcasses of infected animals. The potential for carcasses to release infective
FMDv is greatest in the hours and days following death as the carcasses begin to decompose and
fluids are released. Potential exposures become less likely with time because FMDv does not
replicate outside a living host and is progressively inactivated by biological decay.
If carcasses cannot be managed immediately after death, the temporary carcass storage pile
appears to be the most likely source to possibly expose nearby livestock. This assessment
estimates livestock exposure to FMDv released from a temporary storage pile where carcasses
are placed for 48 hours while further management is prepared. The assessment also considers
seven well-established carcass management options with sufficient capacity for a large-scale
mortality: on-site open burning (pyre), on-site air-curtain burning, on-site unlined burial, on-site
composting, off-site fixed-facility incineration, off-site landfilling, and off-site carcass rendering.
Qualitative rankings of the three off-site options are presented in Table ES-1. Commercial
incinerators would totally inactivate FMDv, and rendering facilities similarly apply sufficient
heat for enough time to inactivate the virus. Viable (i.e., potentially infectious) FMDv in
carcasses placed in landfills could contribute to leachate, however, livestock are not likely to
come in contact with the leachate collected and managed under regulatory requirements. For all
the off-site options, all releases to the environment (e.g., incinerator emissions to air, rendering
facility discharge to surface water) are restricted by, and are assumed to comply with, normally
applicable federal regulations. For these reasons the off-site options are not included in the
quantitative assessment. However, this assessment does not consider facilities operating under
emergency exemptions to environmental law, because, in those cases, normal federal restrictions
on emissions and effluent would not be in force. In such cases, additional assessment of off-site
options would be warranted.
X
-------
Table ES-1. Ranking of Off-Site Livestock Carcass Management Options for Microbes
Rank3
Management
Option
Principal Rationale
H
Off-site Incinerator
Thermal destruction of all microbes occurs. Ash is landfilled.
M
Off-site Rendering
Thermal inactivation of all microbes except prions occurs. Workers are
protected from prion exposure with the use of PPE.
L
Off-site Landfill
Containment, includes liner, leachate collection, and cover material, but
no thermal destruction.
Cattle are not likely to come in contact with landfill leachate collected
and managed under normally applicable regulations.
Abbreviations: H = Highest rank; L = Lowest rank; M = Middle rank; PPE = personal protective equipment
a Relative and absolute risks from microbial pathogens depends on initial concentrations in healthy cattle, which is unknown.
FMDv releases from the four on-site carcass management options (Table ES2) are less controlled
than releases from the off-site options. Both open burning and air-curtain burning thermally
inactivate FMDv particles. Composting also involves partial or complete thermally inactivation.
In addition, large animal composting typically takes enough time for complete biological
inactivation. The containment provided by on-site burial prevents the release of FMDv particles
to air, but leaching from the burial trench has the potential to reach ground water similar to the
temporary storage pile.
Table ES- 2. Ranking of On-site Livestock Carcass Management Options for a Foot and
Mouth Disease Outbreak
Rank
Management Type
Principal Rationale
1
Open Burning and
Air-curtain Burning
Thermal descruction of all FMDv.
2
Composting
Bulking material contains almost all FMDv from releases to air and
soil. Thermal inactivation and biological decay eliminate FMDv before
composting is complete.
3
Burial
Cover soil contains releases to air. If a number of conditions are met,
leaching has the potential to infect cattle that drink water pumped from
a ground water well.
4
Temporary Storage
Cattle can be infected by inhaling or ingesting FMDv emitted to air
from a nearby storage pile. If a number of conditions are met, leaching
has the potential to infect cattle that drink water pumped from a ground
water well.
FMDv, foot and mouth disease virus
Exposures of healthy livestock to releases from an unlined temporary storage pile of 100
carcasses are assessed. Exposure pathways include inhalation of airborne FMDv particles,
ingestion of virus particles that settle on foraged vegetation, and ingestion of well water
containing virus particles leached through soil to ground water. The potential for exposure is
affected by several site-specific factors such as the scale of mortality, distance from the source,
XI
-------
and soil type and depth. An uncertainties assessment evaluates how exposure estimates vary
when these parameter values are changed over several orders of magnitude.
Assuming no preferential pathways in the underlying soil, the assessment finds that with at least
1 meter (m) of soil above the water table, there is a high probability of 99.99% attenuation of
FMDv before reaching ground water. Dilution attenuation and biological decay provide further
reduction of infective FMDv depending on the size of the storage pile, distance to the well, and
rate of ground water flow. Inhalation is the more likely cause of exposure because airborne virus
particles can travel more quickly through air than through the ground water pathway. In
addition, there are fewer barriers to a complete exposure pathway for the air pathway than for the
ground water pathway, which includes such considerations as well depth and placement.
This report provides information to compare options and support decision-making in the event of
an actual foreign animal disease outbreak. Managers can use this report with site-specific
information to identify possible exposure pathways, to determine whether complete exposure
pathways actually exist, and to evaluate which carcass management options are compatible at
their site and which are least likely to expose healthy livestock to FMDv.
Xll
-------
1. Introduction
Established by the Department of Homeland Security (DHS), the National Response Framework
is a single comprehensive approach to domestic incident management.1 The framework provides
a context for DHS and other federal departments and agencies to work with each other and
communities to prevent, prepare for, respond to, and recover from hazards such as natural
disasters, acts of terrorism, and pandemics.
Mass livestock mortalities can result from a natural disaster, a foreign animal disease outbreak, a
chemical or radiological incident, or from other large-scale emergencies. Proper management of
livestock carcasses following large-scale mortalities protects humans, livestock, and wildlife
from chemical and biological hazards; maintains air, water, and soil resources; protects
ecological resources and services; and enhances food and agricultural security. In support of the
National Response Framework, the United States Environmental Protection Agency (USEPA)
Office of Research and Development's Homeland Security Research Program, and the United
States Department of Agriculture's (USDA's) Animal and Plant Health Inspection Service
(APHIS) are collaborating in research to ensure proper management of animal carcasses
following major environmental incidents.
Exposure Assessment Objective
1.1 Purpose and Scope
This report focuses on relative
exposures and hazards for different
livestock carcass management
options in the event of a Foreign
Animal Disease (FAD) outbreak.
Selection of foot and mouth disease
(FMD) virus (FMDv) as the FAD
agent for a hypothetical outbreak is described in Problem Formulation in Section 2. This report is
preceded by Exposure Assessment of Livestock Carcass Management Options During Natural
Disasters (USEPA 2017).
The objective of this assessment is to support
planning for environmentally protective livestock
carcass management methods in times of
emergency by providing scientifically-based
information on potential hazards to human
health, livestock, wildlife, and the environment.
The exposure assessment for FAD virus-infected livestock carcasses builds on earlier research,
peer-reviewed data and existing models involving a variety of carcass management options (e.g.,
pyre construction and fuels), scale of mortality, and site conditions as assumed in the case of
mass livestock mortalities from a natural disaster (USEPA 2017). Additional assumptions
required for FMDv-infected carcasses are described in this report.
FMDv is easily spread and can be transmitted via multiple pathways and exposure routes (USDA
2013a). However, FMDv primarily infects and is transmitted by livestock; the risk to public
health posed by this virus is low (Bauer 1997; Prempeh et al. 2001). Adoption of biosecurity
measures mitigate exposure of other susceptible livestock (and humans) to FMDv via many
1 Information about the National Response Framework is available at https: //www, fema. gov/media-
librarv/assets/documents/117791.
1
-------
pathways (USDA 2014a). Most significantly, USDA recommends immediately identifying
Infected Premises2 and euthanizing in-contact susceptible livestock.
As discussed in Section 2.1, the duration of survival of most FMDv in skeletal muscle of a
livestock carcass is short due to changes in pH that accompany rigor mortis and inactivate the
virus (USDA 2013c). Thus, the highest potential for exposures of other livestock to FMDv will
occur before complete rigor mortis during pre-management activities such as carcass handling,
temporary storage, and transport.
The purpose of this assessment is to provide information about the potential sources of FMDv
exposure to uninfected livestock during management of infected carcasses. In addition, the
assessment can support future carcass management decisions by highlighting parameters (e.g.,
soil type, depth to ground water) that influence chances of the spread of FMDv via specific
pathways (e.g., leaching to ground water that supplies neighboring livestock with drinking
water). The findings also might help to identify the most beneficial mitigation measures for
minimizing potential exposures at actual carcass management sites.
1.2 Report Organization
This report is organized in six sections. Section 2 explains the conclusions of problem
formulation for the assessment, while Section 3 describes the approaches for estimating FMDv
releases from carcasses, FMDv fate and transport, and exposure of other live, susceptible
species. Section 3 also discusses transmission of FMDv by insects and other animals. Section 4
presents the results of the exposure assessment and uncertainty analysis, and discusses how the
findings may be applied to varying site-specific situations. The report concludes with quality
assurance documentation in Section 5 and literature cited in Section 6.
2 Infected Premises are defined as location(s) where presumptive or confirmed positive case(s) were identified based on
laboratory results, compatible clinical signs, FMD case definition, and international standards. The Infected Premises is within
the Infected Zones (USDA 2014a).
2
-------
2. Problem Formulation
Problem formulation for the exposure assessment defines the scope of the assessment and
simplifying assumptions used to allow comparison among the different carcass management
options. Problem formulation for the FAD outbreak scenario builds on a previous assessment of
managing livestock carcasses following a natural disaster (USEPA 2017). This assessment uses
many of the same assumptions related to the site setting, environmental conditions, and the
design and use of the carcass management options described in that report (USEPA 2017). This
assessment starts with a base case similar to the previous case (USEPA 2017), but also considers
several soil types and varying distances to the ground water table. In the development of the
evaluation process, no primary data are gathered and the project relies on secondary data for the
analysis. Given the limited availability of data, the screening process outputs likely exhibit high
levels of uncertainty. Following the base case, uncertainty analyses are conducted for
parameters that are highly variable in the real world (e.g., number of carcasses) or for which best
estimates are highly uncertain (e.g., FMDv release rates from carcasses).
The base case for the assessment assumes 45,360 kilograms (kg) of carcasses for all management
options, as in the natural disaster assessment (USEPA 2017). For cattle, that mass would equal
100 animals if they each weighed 454 kg (USEPA 2017). Though the base number might be
relevant based on the past incidents of catastrophic losses of livestock and their associated large-
scale disposal efforts (NBACC 2004), 100- to 1000-fold increase in base case could require
appropriate scale-up and sensitivity analysis during such catastrophic large-scale event. For the
FMD outbreak scenario, all animals in a single herd of cattle are assumed to be infected with
FMDv, although individual viral loads vary when culling cattle begins. Appropriate authorities
and veterinarians confirm the outbreak and identify the animals to be culled. Animals are
collected as they are euthanized and placed in a temporary storage pile. Receptors of concern are
presumably uninfected cattle in a separate herd pastured near the outbreak location (e.g., a
neighboring farm). The neighboring cattle drink water pumped from a ground water well that is
down-gradient from the carcass management location. The neighboring cattle also graze on
pasture that might be downwind from location(s) where presumptive or confirmed positive
case(s) were identified based on laboratory results, compatible clinical signs, FMD case
definition, and international standards.
No other microbial hazards or chemical hazards are considered in this FAD virus assessment.
Exposures to chemicals or naturally occurring microbes from carcasses managed following a
natural disaster were investigated in the previous report (USEPA 2017) and occur independently
from exposure to an FAD virus.
To prevent spread of FMD, many actions are required to minimize the chance that viable (i.e.,
infective) FMDv reaches susceptible animals at a dose sufficient to cause infection, as described
in USDA/APHIS's l'Oreign Animal Disease Preparedness and Response Plan!National Animal
Health Emergency Management System Guidelines (USD A 2014b). As part of the outbreak
response, some livestock are culled according to USDA/APHIS's policy on stamping-out and
depopulation (USDA 2014a). As part of this policy, USD A/APHIS advises that cattle and other
3
-------
susceptible livestock that meet the FMD presumptive positive case definition3 be culled as soon
as possible, but no later than 24 hours following the index case (USDA 2014a). The guidance
also specifies that all cattle in the Infected Premises be culled.
Beyond the Infected Premises is the Infected Zone4, which includes susceptible animals that
might have been infected via contact with infected animals or contact with people or equipment
or other surfaces with viable FMDv. The perimeter of the Infected Zone is at least 3 km (-1.86
miles) from the site of the index case (USDA 2014a), but depends on the travel patterns for the
livestock herd that includes the positive case. Beyond that, USDA defines a Buffer Zone5 of at
least 7 km (-4.35 miles) beyond the perimeter of the Infected Zone, and specifies that a
Surveillance Zone should be established beyond the Buffer Zone.
Cattle beyond the Infected Premises might or might not be culled, depending on whether
vaccination is used to suppress FMDv replication (USDA 2014a). Quarantine and movement
controls are maintained within the Control Area (Infected plus Buffer Zones) until at least 28
days have elapsed since the decontamination of all confirmed Infected Premises and negative
results are found for all surveillance activities (USDA 2014a). Thus, no other susceptible
livestock will be brought in to repopulate the outbreak farm site for at least 28 days after the
outbreak site is cleared (FMD free).
Also as part of the outbreak response, workers clean and apply disinfectant to equipment,
vehicles, and other potentially contaminated surfaces before movement off-site. In its Foreign
Animal Disease Preparedness and Response Plan SOP [standard operating procedure] for
Cleaning and Disinfection, USDAJAPHIS recommends selecting a disinfectant and disinfection
method(s) based on USEPA-registered labels for antimicrobials (USDA 2017). Thus, all
products must be labeled for FMDv disinfection. There are currently six USEPA-registered
products for FMDv with various active ingredients (USDA 2016). Application of any USEPA-
registered disinfectant should follow label instructions for its use and disposal, with measures in
place to prevent contamination of ground water or surface waters during or after
decontamination activities.
The use of personal protective equipment (PPE) and other biosecurity measures implemented
during an FAD outbreak would minimize human exposures to the microbial agents evaluated in
the previous assessment (USEPA 2017). Thus, the focus of this is assessment is evaluating
potential neighboring livestock exposure to FMDv.
3 A presumptive positive case is an FMD-susceptible animal that has both epidemiological information indicative of FMD and
positive laboratory test results (USDA 2014a).
4 The Infected Zone is the area around the initial presumptive or confirmed positive case (USDA 2014a), generally an area over
which the animal would travel daily (e.g., its barn or other sheltering area to its foraging area or feeding station).
5 The Buffer Zone is the area around the Infected Zone that includes susceptible animals that might have been exposed to FMDV,
either directly or indirectly through exposure to other animals, animal products, fomites, or people from the Infected Zone
(USDA 2014a).
4
-------
2.1 Foot-and-Mouth Disease Virus
FMDv causes a severe, highly contagious disease in cows, pigs, sheep, goats, deer, and other
animals with cloven (also termed divided) hooves. The average incubation period6 for cattle is 2-
14 days, 2 or more days for pigs, and 3-8 days for sheep and goats (Ashford 2015). Infected
animals exhibit a fever and blisters on the tongue, lips, mouth, on the mammary glands, and
around the hooves. The pain and discomfort caused by these blisters can lead to additional
symptoms, including depression, anorexia, excessive salivation, lameness, and reluctance to
move, stand, or eat (USDA 2013a). For many infected animals (including cattle and swine), the
lesions and blisters produced by the virus might be so painful that euthanasia is required for the
animals' welfare (Aftosa 2015). Although FMDv does not typically result in death, restricting
movements of a herd of livestock to contain the disease in a specific paddock can cause severe
distress from lack of food and injury from crowding; in such cases, euthanasia would be more
humane. Animals in contact with a confirmed or suspected case of FMD (e.g., in the same herd)
typically are culled to prevent further spread of the disease (USDA 2103d). Animals in separate
herds (i.e., no contact with animals in the infected herd) likely would be tested for FMDv to help
owners or managers to decide whether additional livestock should be culled to contain the
outbreak.
While FMDv is considered zoonotic, and thus transmissible to humans, human infection is rare.
Globally, only 40 human cases were diagnosed between 1921 and 1997 (Bauer 1997). The
circumstances associated with human infection with FMDv are not well defined; however, all
reported cases had close contact with infected animals. No cases of person-to-person
transmission of the virus have been documented world-wide (Aftosa 2015). There are seven
major viral serotypes: O, A, C, SAT 1, SAT 2, SAT 3 and Asia 1. Serotype O is the most
common serotype worldwide (Prempeh et al. 2001). While most viral strains affect all
susceptible host species, some strains have a more restricted host range (Aftosa 2015). In
humans, the typical incubation period for serotype O of the virus is between 2 and 6 days.
Symptoms in humans are generally mild and self-limiting and include blisters on the hands,
tongue, feet, and mouth as well as fever and sore throat. Patients usually recover a week after the
last blister formation (Prempeh et al. 2001).
Different livestock species vary in their susceptibility to FMDv. Cattle are highly susceptible to
FMDv and have been referred to as "detectors" in some outbreaks (Sakamoto 2011).
The next two subsections describe additional complexities related to evaluating infectivity,
survival, and decay of FMDv as part of this assessment.
2,1,1 Measurement of Viruses and Infective Dose
Measuring concentrations of viable (potentially infectious) virus particles in various materials
requires a method of visualizing virus infections caused by a small amount of material. Methods
generally use dilutions of a virus "stock" (e.g., contaminated medium) applied to cultures of
susceptible cells. The plaque assay inoculates susceptible cell monolayers on petri plates that are
6 Incubation period refers to the time from the moment of exposure to an infectious agent until signs and symptoms of the disease
appear.
5
-------
incubated until cells become visible around the initially infected cell(s). The concentration of the
virus stock then can be calculated in plaque-forming units (PFUs) per mL.
The other common method is to add a specified volume of diluted virus-containing fluid (or
other materials) to host tissues in the laboratory.7 The 50-percent tissue-culture infectious-dose
(TCID50) is a statistical derivative of the PFU assay.8 It is calculated as the dilution at which half
of the replicate solutions contained at least one PFU, making it indicative of cell infection and
damage. FMDv TCID50 values correlate with an infectious dose ID50 values (the dose that would
produce infection in 50% of animals (ID50); however, data required to estimate an ID50 from a
TCID50 value are uncertain. Some have speculated that the number of PFUs should be
approximately 0.5 to 0.7 times the value derived from a TCID50 (ATCC 2012).
Infectious dose. Host animals of the same species can range substantially in their susceptibility
to infection. USD A/APHIS has reported "infectious doses" of FMDv for cattle, sheep and goats,
and pigs in TCID50 equivalents (USDA 2013c; Table 1-2). These infectious doses as well as an
estimate of the corresponding PFUs, and common modes of exposure are summarized in Table
2.1. Reported infectious doses for cattle were only 20 TCID50 units (or 10 to 14 PFUs, bovine
thyroid tissue culture) for inhalation compared to 105 to 106 TCID50 units for ingestion (or
50,000 to 700,000 PFUs, bovine thyroid tissue culture) (Kitching 2002; Kitching and Hughes
2002; Kitching and Alexandersen 2002; Alexandersen et al. 2003). Pigs are similarly less
susceptible via ingestion than inhalation exposure (Alexandersen et al. 2002). Infectious dose
depends on the route of exposure for many viruses and animals (Sakamoto 2011; USDA 2013c).
One question not answered by those doses is what proportion of animals exposed at the dose will
become infected? In general, the concept of "infectious dose" as listed in Table 2.1 is vague and
in fact is not included in medical or veterinarian texts (Johnson 2003). The reason is that with
many factors affecting viability of viruses and each virus particle's chances of reaching the
interior of a cell in which it can replicate, infection becomes a probabilistic process just as the
chance of developing cancer from exposure to a chemical mutagen is probabilistic. Moreover,
individual animals can be more or less susceptible to FMDv, and only a couple viral units might
cause infection in sensitive animals and over 20 PFUs might not cause infection in less sensitive
animals.
7 Typical cell culture tests are conducted using a serial dilution series of doses with typically 10 replicate cell culture wells per
dose. The lowest dose(s) (i.e., most highly diluted samples) should produce no infection. The highest dose(s) (i.e., undiluted
material) should produce 100% infection (i.e., viable virus replication in all test wells at that dose). One or more intermediate
doses should indicate viable virus in only some of the dose replicates (e.g., 30%, 60% for two sequential dilutions). From a
model of dose-response that best fits the data from 0 to 100%) infection, the TCID50 value is calculated.
8 50%o Tissue culture Infective Dose (TCID50) is the measure of infectious virus titer. TCID50 might be more common where the
lethal dose of virus must be determined or if the virus does not form plaques. TCID50 method is a statistical derivative of the
PFU assay. Instead of counting individual plaques, multiple replicates of each virus dilution are made and the TCID50 titer is
calculated from the 50% endpoint where half of the replicates contained at least one PFU.
6
-------
Table 2-1. Foot and Mouth Disease Minimum Infectious Doses and Mode of Transmission
Species
Infectious Dose
Route of
Exposure
Referenceb
Estimated PFU-
Equivalent Dosec
Cattle
As low as 10 to 20
TCID50a inhaled
per cow
Inhalation
UM-CAHFS (2014);
Kitching (2002)
5 to 14
Cattle
0.06 TCID50/m3
Inhalation
Donaldson Al (2001) in
UM CAHFS (2014)
0.03 to 0.04/m3
Cattle
1E+05 to 1E+06
TCID50
Ingestion
Kitching (2002)
5E+04 to 7E+05
Sheep and
Goats
As low as 10 to 20
TCID50
Direct contact
UM-CAHFS (2014);
Kitching and Hughes
(2002)
5 to 14
Pigs
>800 TCID50
Inhalation
Alexandersen et al.
(2002)
>400
Pigs
Approximately 1
E+05 TCID50
Ingestion
Kitching and
Alexandersen (2002)
5E+04 to 7E+04
TCID50 = The quantity of vims (generally in 1 inL of fluid or 1 gram of tissue) added to tissue-culture wells
(using cells of the appropriate animal group) that result in 50% of the culture wells exhibiting active infection. An
infectious dose of 20 TCID50 per mL via inhalation is approximately equal to 10 to 14 plaque-forming units (PFUs)
per mL (ATCC 2012). Each PFU equals one (or more) viable virus particle.
b Complete references are found at the end of the report.
0 Multiplied TCID50 Infectious Dose by 0.5 to 0.7 (see text).
2.1.2 Survival and Biological Decay
Survival refers to the ability of an infectious unit of virus to remain infectious in the environment
over a defined period of time (Embrey et al. 2004). FMDv particles can survive in the
environment for long periods (e.g., weeks) under a wide range of conditions. Table 2.2 provides
an overview of the survival of FMDv associated with changes in temperature and pH. FMDv is
inactivated by high temperatures (<50 degrees Celsius [°C]; 122 degrees Fahrenheit [°F]) and
acidic or alkaline conditions (pH <6.0 or >9.0) (Cottral 1969; OIE 2013; USDA 2014a). Survival
is a function of the medium associated with the virus (e.g., specific tissue, excretions), virus
strain, humidity, exposure to ultraviolet (UV) light, pH, and temperature. As a result, there is
high variability in viral survival across natural environments (Alexandersen et al. 2003).
7
-------
Table 2-2. General Survival of Foot and Mouth Disease Virus (FMDv)
Action
Resistance to Low and High
Temperature or pH
Survival in Biotic and Abiotic
Environmental Media
References8
Temperature
FMDv in animal tissues are:
¦ preserved by refrigeration
(4°C; 40°F) and freezing
(0°C; 32°F);
¦ progressively inactivated by
temperatures above 50°C
(122°F);
¦ inactivated by treatment
with high heat (internal
temperature of 70°C;
158°F) for at least 30
minutes.
¦ Can remain viable in muscle, liver,
bone marrow, lymph nodes, and
blood of slaughtered animals when
temperatures are low (i.e.,
refrigeration, freezing);
¦ Exposure to sunlight has little or no
direct effect on infectivity;
¦ May survive for days to weeks in
organic matter and days to a year in
wool and hides under moist and cool
temperatures.
Cottral (1969);
OIE (2013);
USDA (2014a)
pH
FMDv in animal tissues are
quickly inactivated by pH
<6.0 or >9.0.
¦ Survives in lymph nodes and bone
marrow at neutral pH (6.6-7.3); and
¦ Inactivated in muscle at pH <6.0 (i.e.,
after rigor mortis).
Cottral (1969);
OIE (2013);
USDA (2014a)
Abbreviations and acronyms: °C = degrees Celsius; °F = degrees Fahrenheit; FMDv = foot and mouth disease virus
a Complete references are found at the end of the report.
Even under optimum conditions outside the host animal (e.g., > 70% relative humidity and
temperature between 0 and 50°C), in air FMDv inactivates over time due to biological decay
(with zero replacement by replication). In addition, at temperatures progressively higher than
50°C, thermal inactivation (fraction viable FMDv inactivated) per unit time increases, with 100%
inactivation at 70°C for at least 30 minutes (Cottral 1969; OIE 2013; USDA 2014a). Inactivation
rates have not been reported for FMDv in cattle carcasses, specifically. Donaldson and Ferris
(1978) reported a biological decay rate of 50% per hour for FMDv in bovine fluid medium.
FMDv survival within a carcass is dynamic and tissue-specific. Rigor mortis, the hardening of
body muscles after death, occurs about 6-24 hours after slaughter in beef cattle (Edelstein 2014).
The pace of rigor mortis is influenced by ambient temperature (USDA 2013c): rigor mortis is
slower at lower temperatures. The virus present in muscle tissue is inactivated when rigor mortis
reduces tissue pH to below 6 (USDA 2013c). There are also compartments (e.g., bone marrow,
lymph nodes, offal [e.g., kidney, liver], other organs) in cattle carcasses in which pH does not
change due to rigor mortis that could continue to provide a reservoir of virus for extended
durations, especially under environmental conditions that favor virus survival (e.g., low
temperature) (Alexandersen et al. 2003).
In this assessment, all carcasses are assumed to pass through rigor mortis in the temporary
storage pile, with tissue pH below 6 inactivating all FMDv in muscle tissue and other non-hardy
compartments over the 2 days on the pile. However, some viable FMDv could remain in bone
marrow and lymph nodes after 48 hours.
8
-------
Inactivation of FMDv begins when temperatures are above 50°C (122°F). Heating animal
carcasses to a minimum core temperature of 70°C (158°F) for at least 30 minutes completely
inactivates FMDv (USDA 2014a). For carcass management, temperatures associated with
combustion or where heat is either applied (i.e., rendering) or produced indirectly (i.e., on-site
composting) are described in Section 2.2.
2.2 Livestock Carcass Management Options and Assumptions
The management options considered for the exposure assessment are those with documented use
following FAD outbreaks or that are likely to have sufficient capacity for large-scale carcass
management. These include seven well-established methods, which can be categorized into three
groups as shown in Table 2.3.
The carcass management options can be categorized as on-site or off-site. The on-site
management options (i.e., open burning, air-curtain combustion, burial, and composting)
typically are performed on the livestock owner's property if a suitable location is available.
Table 2-3. Livestock Carcass Management Options Considered for the Exposure
Assessment
Management Type
Specific Management Option
Combustion-based Management
¦ On-site Open Burning (Pyre)
¦ On-site Air-Curtain Burning
¦ Off-site Fixed-facility Incineration
Land-based Management
¦ On-site Unlined Burial
¦ On-site Composting
¦ Off-site Lined Landfill
Materials Processing
¦ Off-site Rendering
Additionally, the carcass management options can be categorized by degree of containment, as
summarized in Table 2.4. Containment options prevent or reduce releases of FMDv into
environmental pathways that may lead to exposure of healthy livestock. Containment options in
the assessment include off-site landfilling, on-site burial, and composting. This assessment does
not consider facilities for containment that operate under emergency exemption to environmental
law, because, in those cases, normal federal restrictions on emissions and effluent would not be
in force. In such cases, additional assessment of off-site options would be warranted.
The containment provided by on-site burial prevents the release of FMDv particles to air, but
leaching from the burial trench has the potential to reach ground water. Large animal composting
typically takes six to eight months (Looper 2001), enough time for complete biological
inactivation. FMDv in carcasses placed in landfills could contribute to leachate, however
livestock are not likely to come in contact with the leachate collected and managed under normal
regulatory requirements.
Composting also can be considered a treatment option because heat generated during composting
can completely or partially inactivate many species of bacteria, viruses, and particularly protozoa
and helminthes (Glanville et al. 2006; Ligocka and Paluszak 2008; Wilkinson 2007 as cited in
9
-------
Gwyther et al. 2011). Core temperatures of the compost windrow should reach approximately
65-71°C for several days or even a few weeks or months depending on the size of the windrow
and other conditions (NABCC 2004; Kalbasi et al. 2005). FMDv is 100% inactivated at 70°C for
at least 30 minutes (Cottral 1969; OIE 2013; USDA 2014a).
Table 2-4. Containment of Chemical and Microbial Releases from Management Options
Combustion
Land-Based
Material Processing
On-Site
Off-Site
On-Site
Off-Site
On-Site
Off-Site
Air Curtain
Composting
Not Evaluated
Open Burning
(Pyre)
Incineration
Burial
Landfill
Rendering
| = Releases restricted by regulation
= Releases partially restricted by physical barriers
1 = No barrier to releases
Four of the management options included in Table 2.4 are either combustion based (i.e., on-site
air curtain burning, on-site open burning, and off-site incineration) or involve processes where
heat is applied (i.e., rendering). Exposures are not estimated for these four carcass management
options because they reach temperature and time criteria for FMDv inactivation (i.e., 70°C for at
least 30 minutes (Cottral 1969; OIE 2013; USDA 2014a):
¦ On-site open burning: 550°C (1,022°F) (Bartok et al. 2003);
¦ On-site air curtain burning: 850°C (1,562°F) (Miller 2015);
¦ Off-site incineration: 600-1,000°C (1,112-1,832°F) (Chen 2003, 2004; NABCC 2004);
¦ Rendering: 115-145°C (240-290°F) for 40 to 90 minutes (Meeker 2006).
All of the carcass management options are preceded by activities with the potential to release
virus particles. Among these are carcass handling, temporary storage before the selected
management option, and transportation of the carcasses from the storage location to the
management location. Each of these is discussed and evaluated in the assessment of livestock
management options for natural disasters (USEPA 2017) and the sections below. In addition, off-
site transportation of carcasses to landfills, commercial incinerators, or rendering facilities offers
a possibility of off-site transport of viable FMDv particles.
2.2.1 Carcass Handling
Moving carcasses to and from a temporary storage pile, loading and unloading vehicles, and
placing the carcasses in a management unit will require some workers to come in contact with
the carcasses. The use of PPE by these workers, in addition to the low risk to public health
associated with FMD, suggests that risks to workers are minimal if they follow protocols.
Flying insects and vertebrate scavengers, such as birds or rodents, could spread the virus to other
susceptible species after contact with FMDv-infected cattle carcasses during various handling
activities (including the loading and unloading of carcasses from heavy equipment or vehicles)
(USDA 2013b, 2013c). Animals most likely to contact carcasses during the handling process are
insects (e.g., flies) that land and feed on animals. For flies to transmit FMDv mechanically, they
would have to settle on neighboring live cattle where the virus might fall off the fly and where
10
-------
the cattle could subsequently lick the area (e.g., nose). In addition, distances traveled by flies are
usually less than 2 miles (3.2 km; a few flies might fly farther; Townsend 1997). Scavenging
wildlife (e.g., fox, crow, feral swine, and rats) are less likely to make contact with carcasses
during daytime handling processes due to their avoidance of active humans. At night, carcasses
would be covered with tarps secured to the ground.
During carcass handling, virus particles could be released to air from external surfaces, including
secretions around the head and rear of carcasses. Using heavy machinery also might puncture
carcasses, releasing fluids faster than with intact carcasses. There are accounts of transport of
virus particles up to 300 km (approximately -186 miles) by the wind that included travel over a
water body (Gloster et al. 1981). Sorensen et al. (2000) modeled a simulated FMDv plume using
the computer model Rimpuff and assuming optimal climatic and topographical conditions. The
authors concluded that a virus plume produced by 1,000 infected pigs on a farm could reach
cattle up to 300 km from the infected pigs. Thus, it is plausible that livestock at farms located
outside the FMD response area could become infected under favorable conditions (e.g., cool
temperature, high relative humidity). That scenario, however, is based on live pigs exhaling virus
for 24 hours, and swine are known to shed FMDv at higher rates than cattle, sheep, or goats
(USD A 2013c).
Estimating FMD exposures resulting from carcass handling requires assumptions about the
nature, frequency, and duration of handling actions and virus particle release rates for these
actions. No data have been found to quantify those parameters. However, the releases are
expected to be similar in nature to releases from carcasses piled for 2 days before further
management. This assessment assumes that both handling and temporary storage can release
FMDv and that handling time is included in the 48-hour period prior to transport of carcasses to
management locations (e.g., trench, pyre, off-site transport to rendering plant). Physical
disturbance of cattle carcasses might release hide-bound virus particles and FMDv from
secretions on the exterior of carcasses; movement with large equipment might puncture carcasses
allowing rapid releases of materials from lungs or bowels.
2,2,2 Temporary Carcass Storage
Temporary on-site storage of carcasses might be necessary while available management options
are identified and evaluated, while on-site management units are constructed, and while awaiting
transportation or completion of other logistical requirements (e.g., obtaining burn permits,
obtaining air-curtain burning equipment from off-site). Many state regulations require carcasses
to be managed within a specified timeframe, usually within 24 to 72 hours (USDA 2015). For the
exposure assessment, on-site storage for 48 hours (2 days) is assumed for all management
options.
The location and design of the temporary carcass storage location(s) can affect potential
exposures. Carcasses could be stored in a pile on the ground in open air, in a refrigerated storage
unit, or in containers (USDA 2015). Carcasses on the ground could be covered with a tarp, soil,
or other material, or left uncovered (USDA 2005). Carcasses might be placed on bare earth or on
an impervious surface with or without leachate collection or other management features.
Assumptions about the carcass pile design, management, and FMDv releases are discussed in
Section 2.3 and Section 2.4.
11
-------
2,2,3 Transportation
A semi-quantitative evaluation of chemical and microbial releases and potential exposures from
transportation is presented in the assessment for natural disasters (USEPA 2017). That evaluation
found a very low (7.1E-05) likelihood of materials in carcasses or carcasses themselves being
released as a result of an accident during transit to an off-site management facility. In addition,
mitigation requirements and standard practices (e.g., the use of truck bed covers and liners)
greatly reduce non-accident releases from trucks in transit. These conclusions are further
supported by an assessment conducted by the University of Minnesota's Center for Animal
Health and Food Safety (UM CAHFS) (2014). In their assessment of risks of transmission of
FMD by moving swine and cattle carcasses from an FMD-infected premises to a disposal site,
UM CAHFS found that risk of infection for susceptible swine and cattle associated with
transport of carcasses is (1) negligible if a standard rendering truck (tailgate sealed and tarp
cover) is used together with a sealable liner to contain carcasses, and (2) negligible to low if (a) a
standard rendering truck is used without a bag or (b) a roll-off/dump truck with a bag are used. If
trucks are uncovered or only a liner is used to minimize leaks from a truck, risks of spreading
infection to other susceptible animals are likely to be moderate to high (UM CAHFS 2014).
Moreover, in the specific case of transporting FMD-infected carcasses, all associated exposures
can be assumed negligible if workers adhere to USDA's (2014a) biosecurity standard operating
procedure (SOP) for FMD response. Also, if carcasses are transported after 48 hours of
temporary storage, the pH reduction associated with rigor mortis would further reduce the
amount of viable virus available for release during transportation. Carcasses are also covered
during transport to reduce the scattering of external virus particles (i.e., particles present on hair,
skin, hooves, and other external surfaces) to the environment.
2.3 Exposure Assessment Assumptions
Where possible, this assessment uses assumptions that are consistent with those used in the
assessment for natural disasters (e.g., design of storage pile and burial trench; USEPA 2017).
This section identifies assumptions used for the FAD assessment that differ from or are in
addition to those for the natural disaster. Table 2.5 summarizes the assumptions for the FAD
outbreak scenario.
12
-------
Table 2-5. Foreign Animal Disease Outbreak Scenario Assumptions3
Issue
Assumptions
Scale of Livestock
Mortality
¦ The quantity of carcasses to be disposed is 45,359 kg (50 U.S. tons).
¦ For cattle, 45,359 kg would equal 100 animals if they each weighed 454 kg.
Livestock Types and
Quantity
¦ Livestock category likely to be impacted by an FMD outbreak in the U.S. -
cattle.
Carcass Management
and Post-Management
Assumptions
¦ Seven carcass management options with documented use following large-scale
livestock mortalities are considered.
¦ The assessment begins with placement of carcasses in an outdoor temporary
storage pile, assuming temperatures between 50 and 90°F before movement to
the management location.
¦ Exposures are assessed for releases of FMDv only from management units and
from post-management processes that contribute to the ultimate fate of the
virus.
¦ Off-site livestock carcass management options operate in compliance with
facility permits designed to limit off-site releases to health-protective levels;
hence exposures to releases from off-site facilities are not evaluated.
¦ At a minimum footwear, clothes, equipment, vehicles, and other objects that
could act as fomites in areas designated as Infected and Contact Premises are
disinfected after depopulation by workers who wear appropriate PPEb prior to
exiting those areas.
Carcass Handling
¦ When handling presumptive FMD-in fected livestock and their carcasses (e.g.,
loading and unloading carcasses from vehicles or into management units),
workers use USDA/APHIS-recommended "Level C"protection (USDA
2014b) and, therefore, are not infected
¦ Workers do not take home any items of clothing used during FMD control and
eradication activities.
¦ Non-workers do not touch or otherwise contact carcasses or equipment used to
transport or handle carcasses, and the public is excluded from work sites.
¦ Biosecurity zones and associated biosecurity practices required for FAD
outbreaks are used according to USDA/APHIS recommendations (USDA
2013b).
Temporary Carcass
Storage
¦ Workers move carcasses from the mortality location to an outdoor pile on bare
earth where they stay for 48 h before transport to management locations. There
is no leachate collection or retention system for the pile.
¦ The outdoor storage pile is covered with a tarp to control contact with insects,
wild birds, and other scavengers. However, there may be times when the tarp
is removed and the pile is left uncovered to allow for more carcasses to be
added to the pile.
¦ Disinfection products are not applied to the storage pile or carcasses (USDA
2013d).
¦ For 100 cattle carcasses, the storage pile has a trapezoidal cross sectional shape
that is 8 ft (2.4 m) wide at the base, 3 ft (0.91 m) wide on top, and 5 ft (1.5 m)
13
-------
Issue
Assumptions
high. With a total volume of 196 yd3 (150 m3), the length of the pile is 132 ft
(40.3 m).
Depopulation
¦ To contain the outbreak, all livestock in the herd on the outbreak farm are
culled.
Repopulation
• The farm will not be repopulated with potentially susceptible livestock until
the area is considered to be "FMD-free" (i.e., infection with FMDv unlikely),
a minimum of 28 days.
Hazard Types
• The concentration of viable FMDv is significantly reduced by rigor mortis
and the associated reduction in pH within cattle musculature. All cattle in the
storage pile pass through rigor mortis within the 48-h holding period. FMDv
present in muscle tissue and other non-hardy compartments is therefore
inactivated. However, some viable FMDv could remain in bone marrow,
lymph nodes, and internal organs.
¦ Carcass management will release chemical and other microbial agents that are
naturally present in healthy animals or products of carcass management
activities (e.g., combustion products). Those releases are assumed to be the
same as estimated for the natural disaster scenario (USEPA 2017).
Geographic and
Spatial Issues
¦ All carcass management options are evaluated with the same spatial,
geographic, meteorological, and other environmental characteristics assumed
for the natural disaster assessment (USEPA 2017) with one exception: virus
leaching to ground water is evaluated in this assessment for three soil types:
sand, silty loam, and clay.
¦ The site location and regional factors do not preclude the availability or
feasibility of any carcass management option.
¦ Uninfected cattle are located on neighboring farms.
Abbreviations and acronyms: APHIS = Animal and Plant Health Inspection Service; FAD = foreign animal disease;
ft = feet; FMD = foot and mouth disease; FMDv = foot and mouth disease virus; h = hour; kg = kilograms; km =
kilometers; m = meter; PPE = personal protective equipment; USDA = U.S. Department of Agriculture; yd = yard.
a Assumptions in bold, italic type are specific to the FMD outbreak assessment; all other assumptions were used in
the natural disaster assessment (USEPA 2017).
b Contact Premises are defined as premises with susceptible animals that could have been exposed to FMD, either
directly or indirectly, including but not limited to exposure to animals, animal products, fomites, or people from
Infected Premises. Areas characterized as Contact Premises will be considered part of the Infected Zone and the
Buffer Zone (USDA 2014a).
Complete references are found at the end of the report.
2.4 Sources of FMDv Releases and Exposure Pathways
This section describes the sources of FMDv released to environmental media (i.e., air, water,
soil) and exposure pathways for uninfected livestock. It also includes conceptual models for each
of the quantitatively assessed management activities and options: temporary carcass storage and
burial.
14
-------
2.4.1 Temporary Carcass Storage before Transportation
For the hypothetical FAD outbreak assessed in this report, the first action after euthanasia is to
pile carcasses in a temporary storage area on the ground with strong tarps covering the pile and
anchored firmly into the ground (USDA 2005). In an actual FMD outbreak, livestock might be
herded to a burial trench or the area surrounding a compost windrow before euthanasia, allowing
immediate placement with minimal handling. For this exposure assessment, however, on-site
storage for 48 hours (2 days) is assumed for all management options. Figure 2.1 summarizes the
conceptual model for the temporary carcass storage pile. It traces exposure pathways from the
storage pile to livestock on farms near the outbreak farm or outside the FMD response area.
Particles
Vectors (e.g., insects)
and vertebrate
scavengers)
Leakage to soil
Particle
Deposition
Terrestrial
Plants
Inhalation
Ingestion
Ingestion
Livestock
Ingestion
Air
Well
Water
Groundwater
Storage Pile
Depopulated Cattle
Figure 2-1. Conceptual model for exposure pathways from temporary carcass storage.
A temporary storage pile is likely to be uncovered for short periods, as when the tarp is removed
or adjusted to accommodate additional carcasses. If the storage pile is uncovered, FMDv
particles on the surfaces of carcasses can be released to air and transported beyond the FMD
response area via wind. Neighboring livestock could either inhale the particles directly or ingest
the particles after they deposit to terrestrial plants and soils, which cattle also ingest while
grazing (Herlin and Andersson 1996; Gloster et al. 1981; Sorensen et al. 2000).
Liquid leaching from the storage pile is assumed to percolate down through soil toward ground
water. Water from a ground water well off-site could be used for providing drinking water for
neighboring livestock that have not been culled as part of the response effort.
15
-------
Finally, vertebrate scavengers (e.g., birds, feral swine, rodents) and flying insects might spread
the virus to other susceptible species after contact with FMDv-infected cattle carcasses in the
temporary storage pile (USDA 2013b, 2013c).
2.4.2 On-site Burial
Figure 2.2 is the conceptual model for the on-site burial of FMD-infected carcasses. In this
management option, livestock carcasses are placed in an unlined, excavated pit or trench in a
suitable location on site. The carcasses are covered with 6 feet (ft; 1.8 m) of clean fill including 3
ft (0.9 m) of soil mounded over the site starting at ground level (USDA 2005). This soil cover
will flatten over time as the carcasses lose fluids and other mass during decomposition. Although
access to the site is not restricted, it will not be used in the relatively near future for crop farming
or raising livestock; it will be seeded over for soil stabilization.
Leakage to soil
Particle
Deposition
Terrestrial
Plants
Vectors (e.g., insects
and vertebrate
scavengers)
Inhalation
Ingestion
Ingestion
Livestock
Air
Well
Water
Groundwater
On-site Burial
Depopulated Cattle
Figure 2-2. Conceptual model for exposure pathways from on-site burial of livestock
carcasses.
As the carcasses decompose, rapidly at first (over months) with the remainder decomposing
more slowly (over years), particles could diffuse upward though the soil cover to aboveground
air. However, with the amount of soil cover placed on the burial trench (6 ft), it is unlikely that
FMDv particles could travel through the soil cover and be released to air. Thus, exposure after
burial via air is considered negligible.
In the burial trench, FMDv particles can leach with carcass fluids and with rainwater permeating
through subsurface soils to ground water. If ground water is used to provide drinking water for
cattle outside the FMD-response area, a complete exposure pathway exists that might not be
16
-------
negligible. As discussed in Section 2.1, changes in pH (i.e., <6.0) associated with rigor mortis
are expected to inactivate virus in muscle tissue while the carcasses are in the storage pile and
before being placed in the burial trench. Viable virus can survive in bone and other tissues.
Therefore, the concentration of infectious virus particles present in leachate released from the
burial trench is expected to be lower than the concentration of infectious particles present in
leachate released from the storage pile. Virus particles in buried carcasses will naturally decay
(i.e., lose integrity and inactivate) over time. Based on the estimated rate of biological decay (see
Section 3.1.2; estimated from data reported in Schijven et al. [2005]) the viable viral load would
decline by at least 95% within a month, and none is expected to survive more than a year.
Therefore, exposures from on-site burial of carcasses are evaluated qualitatively in relation to the
exposures quantified for storage pile leaching.
2.4.3 Summary of Exposure Pathways for Livestock
Table 2.6 summarizes FMD exposure pathways for the temporary storage pile, which precedes
all of the seven livestock management options, and for burial.
Pathways with quantitatively estimated exposures are indicated with bold type and footnote "a"
in Table 2.6. Other pathways that are not assessed quantitatively are indicated by endnotes "b"
and "c" in Table 2.6.
Table 2-6. Livestock Exposure Pathways for Livestock Carcass Management
Exposure Source
Carcass Management Options
Temporary Storage Pile
Burial
Air Inhalation
1) Aira
1) Airb
Direct Ingestion
2) Air —~ Plants3
2) Air —>¦ Plants'3
Incidental Ingestion
3) Air —~ Soil3
3) Air —>¦ Soilb
Incidental Ingestion
4) Air —~ Surface water3
4) Air —~ Surface water3
Ground-water Ingestion
5) Leachate —~ Ground water3
5) Leachate —~ Ground water3
Vectorborne Transmission
6) Airborne vectors —> Livestock0
6) Airborne vectors —>¦ Livestock'10
Acronyms: SW = surface water
Notes: Bold type means quantitative methods will be used for exposure assessment.
On-site means inside the Infected Premise of the hypothetical farm. Off-site means beyond the Infected Premise,
potentially on other farms. "—" means no pathways identified
a Quantitative methods will be used for exposure assessment.
b Exposures assumed to be negligible.
0 Pathway will be described qualitatively; quantitative modelling approaches not available.
17
-------
posure Estimation
Section 3.1 describes the approaches used to estimate releases of FMDv to air and to soil for all
carcass management exposure scenarios, including those scenarios evaluated as part of the
uncertainty analysis. Section 3.2 describes the modeling methods employed for specific
environmental media for these scenarios.
3.1 Estimation of FMDv Releases
As illustrated in the conceptual models in Section 2.4, FMDv particles could be released from
carcasses: (1) to air during handling and temporary storage and (2) to soil via leachate from the
temporary carcass storage pile and from the unlined burial trench. Estimates of these releases are
needed to quantify exposure of susceptible livestock.
In this report, all exposure estimates are reported in units of FMDv TCID50 values to allow
comparison with available infectious dose data (see Section 2.1). The TCID50 is a method of
detection and quantification of the viral loading in a clinical or environmental sample. Much of
the recently published measures (e.g., past two decades) of viral load in environmental or
veterinary samples are reported as TCID50 /mL, For this assessment, data reported in units of
TCID50 per unit volume or mass of material are used if reported; if PFUs are reported, they are
not converted to corresponding TCID50 values.
3.1.1 Air
Releases of FMDv particles to air as aerosols could occur from handling the carcasses and the
temporary storage pile when uncovered to add or to remove carcasses. Aerosols are particles
consisting of aggregated smaller particles (e.g., virus, skin cells, dust) together with some liquid
droplets. Aerosols are small enough to have a high surface area to mass ratio so that they remain
suspended in air for some time before aggregating further and falling to the ground. An aerosol
can be characterized by a distribution of particle sizes and each aerosol particle can contain one
to many viral units.
Virus release rates In aerosol particles
Carcass handling and temporary storage is assumed to take place over the course of 2 days (48
hours total). The temporary storage pile is covered with a tarp for no more than 48 hours (Day 1
and Day 2) before carcasses are moved to their respective management location. The estimated
aerosol with virus release rate covers both carcass handling (e.g., movement to the temporary
storage pile) and storage. Separate release rates are not derived for the two activities because
virus release data are available only for live animals.
Infected carcasses have surfaces that can carry FMDv (i.e., act as fomites) during handling and
storage. Aerosolized viable virus could be released from FMD-infected skin, hooves, lesions, or
hide during carcass handling or placement in the storage pile (Sellers and Parker 1969; Dillon
2011). Identified mechanisms for the release of infectious aerosols from carcasses include
exposure to moving air or mechanical abrasion (Dillon 2011). Neither our 2016 literature review
nor the literature reviewed by Dillon (2011) identified any measured emissions data for FMDv
released from cattle carcasses infected with FMDv. As a result, the release of FMDv from cattle
18
-------
carcasses is estimated for handling and placement on the storage pile using modeled releases of
FMDv from live cattle.
The aerosol release rate is selected from reported outputs of the Sorensen et al. (2000) virus
production model, which described the rate of release of FMDv aerosol for live infected cattle
relative to day of clinical disease. This model was calibrated using reported and modeled data
(Sellers and Parker 1969; Donaldson et al. 1970) for live cattle infected with FMDv. The model
also included an extrapolation component when data were not available for the desired number
of days post-infection. The highest estimated release was 5.1 logio TCID50 (1.26E+05 TCID50)
per cow per 24-hour period (Sorensen et al. 2000), which is used as the base-case aerosol release
rate for FMD-infected cattle carcasses. Aerosol release rates from carcasses should be lower than
from living animals that breathe in and out, exhaling aerosolized FMDv particles with each
breath. Carcasses and living cattle might release FMDv particles from external surfaces (e.g.,
adsorbed to hair and skin, from mucous-covered surfaces).
Variability in the viable FMDv particles in aerosol releases from handling and placement in the
storage pile could be affected by the same elements associated with variability in viable FMDv
particles in aerosol released from live, breathing FMDv-infected animals: FMD strain
differences, host breed, stage of infection, ambient environmental conditions (e.g., relative
humidity) that could favor viability of virus (Sellers and Parker 1969; Sellers et al. 1971;
Alexandersen et al. 2003). Additionally, aerosol release rates could be affected by the carcass
location in the storage pile (e.g., more or less contact with potential air currents or mechanical
abrasion). As noted earlier, the selected aerosol release rate was developed from data on live,
infected, breathing cattle, and therefore, it is likely to overestimate releases from hides or
carcasses in a pile. Because of its uncertainty and likely bias, the estimated aerosol release rate
will be varied to examine the sensitivity of exposure estimates to release rate (see Section 4.3).
Particle size and FMDv distribution for aerosol source release
In addition to aerosol particle release rate (with virus in particles), air dispersion modeling for the
exposure assessment requires information about the size distribution and mass of airborne
particles and the density of viral units in the aerosol. The assessment assumes that FMDv-
containing particles from carcass management activities are a mixture of ambient aerosolized
biological and non-biological matter (e.g., FMDv, feed, dust, skin, feces). No data have been
identified that reported the particle size distribution for exhaled aerosols associated with FMDv-
infected cattle, healthy cattle, or the possible aerosol particle size distribution associated with
cattle carcasses.
The FMDv-containing particles in the released aerosols are assumed to have the same particle-
size distribution as that reported for two live cattle farms in the Netherlands (n = 51 and 104
cattle, respectively; Lai et al. 2014). An aerosol particle mass density of 1 gram (g) per cm3 was
assumed for modeling.
Gloster et al. (2007) found no difference in the airborne particle-size distribution associated with
healthy pigs compared with the airborne particle-size distribution from FMDv-infected pigs in a
confined space. FMDv-containing aerosol particles released from the pigs associated with the
ambient particle load (Gloster et al. 2007). Therefore, the particle-size distribution associated
with aerosols from the surfaces of FMDv-infected cattle carcass is assumed to be the same as the
19
-------
particle-size distribution from non-FMDv-infected live cattle. Gloster et al. (2007) found FMDv
infectivity to be generally evenly distributed in airborne aerosol particles measured in the
confined air-space with the FMDv-infected swine. FMDv infectivity (as measured in TCID50
units) in aerosol particles was evenly spread over particles of diameter < 3 |im, 3 to 6 |im, and >
6 |im. The following particle size and mass fraction distribution is used for air dispersion
modeling in this assessment:
1 0.25 to 1.0 |im: 2.5%
¦ 1.0 to 2.5 |im: 2.5%
¦ 2.5 to 10 |im: 36.25%
¦ 10 to 32 |im: 58.75%
Biological decay
Aerosolized FMDv is assumed to inactivate over time outside of the host body, which is called
biological decay. Donaldson and Ferris (1978) reported a biological decay rate of 50% per hour
for FMDv in bovine fluid medium FMDv. Both Sorensen et al. (2000) and Garner et al. (2006)
used that biological decay rate in their models of risks from wind-borne FMDv. In this
assessment of carcass management options, that decay rate also is used for airborne FMDv.
In air dispersion models, the decay rate is often reported as the rate of particle loss per second.
Thus, 0.5 logio loss of FMDv particles per hour was converted to the percentage of FMDv
particles lost per second. The biological decay rate for FMDv equals 1.90E-4 per second or
0.019%) per second.
Because the rate of biological decay has a high level of uncertainty, it is varied in the uncertainty
analysis conducted as part of this assessment.
3,1,2 Subsurface Soil
A Microsoft® Excel™ workbook was developed to estimate concentrations of FMDv reaching
ground water via leaching from the temporary storage pile. Leaching from the burial trench is
evaluated relative to leaching from the storage pile. Exposure to neighboring cattle occurs when
the cattle receive drinking water from an affected well.
Leachate Volumes
Within hours after death, carcasses can release free fluids (e.g., contents of intestines, urine,
fluids in lungs) and within days additional fluids are released due to decomposition. Fluids that
seep into the ground as a leachate may eventually reach ground water. Some constituents that
remain dissolved in water and that do not adsorb to soil particles (e.g., chloride ions) will reach
ground water as the leachate from reaches ground water. Chemicals and particles in leachate that
have a strong tendency to sorb to soil particles, on the other hand, will be retarded relative to
water transport to ground water owing to adsorption to and desorption from soil particles.
The volume of leachate loaded to the surface layer of soils immediately under a temporary
storage pile is estimated from data reported in Young et al. (2001). Assuming that the quantity of
fluid released over the first week after mortality is released evenly over time (i.e., estimate for
one week is divided by 7 days to estimate daily release), approximately 10.7 liters (L) of leachate
20
-------
would be produced per 454 kg carcass per day.9 For 100 carcasses, the total estimated volume of
leachate from the temporary storage pile is 1,070 L per day. The actual amount of leachate
released per carcass per day will vary depending on the size and condition of carcasses, ambient
temperature, and other factors. Thus, the volume of leachate released per carcass is varied as part
of the uncertainty assessment.
Virus load
Alexandersen et al. (2003) collected FMD strain-specific data on the maximum recorded virus
titer in, and volume or weight of, various secretions and excretions from FMD-infected cattle
including blood or serum, feces, and urine. UM-CAHFS (2014) and Gale (2002) also reviewed
virus titer data from various carcass tissue compartments including skeletal muscle, heart muscle,
skin/hides, lymph nodes, and kidney. Secretions or excretions with relatively high total virus
loadings include feces (approximate loading of 1E+05 TCID50 per gram) and urine (approximate
loading of 1E+05 TCIDso/mL) (Kitching 2002). Major internal tissue compartments include the
blood (approximate total loading of 1E+11 PFU per carcass) and muscle (approximate total
loading of 1E+07 PFU per carcass) (Gale 2002; Alexandersen et al. 2003; UM CAHFS 2014).
These authors reported load in various measures, such as TCID50, PFU, and others reported from
various tissue-culture types. While there is no agreed upon value to relate TCID50 to PFU (Gale
2002), UM CAHFS (2014) assumes that one PFU equals about 1.4 TCID50 for FMDv based on
Alexandersen et al. (2003) and Donaldson and Ferris (1978); 1 TCID50 equals approximately 0.7
PFU. For this assessment, all measurement units are assumed to be equivalent to 1 TCID50 unit.
UM CAHFS (2014) estimated that the total FMDv in one cattle carcass could be 1 E+06
PFU/gram, which would equal 1 E+09 PFU/kg or 4.54 E+l 1 PFU/carcass where one carcass
weighs 454 kg. The number of TCID50 values per carcass would be somewhat higher.
The initial virus load is included in the uncertainty analysis for this assessment because of the
limited information on which to base an estimate and because of the potential for the virus load
to vary substantially among cattle in the same herd at the time the animals are culled.
Biological decay
After the leachate is released to the soil from the carcasses in a temporary storage pile, the
concentration of viable (i.e., infectious) FMDv is continually reduced over time by biological
decay without a living host animal. For this assessment, the biological decay rate estimated for
in-ground decay is applied from the time the leachate is released from the carcasses until it
reaches ground water. The concentration of viable virus particles per liter water when the FMDv
particles start to break through to ground water is calculated.
For burial, biological decay can be assumed to continue after the carcasses are moved from the
storage pile to the burial trench. After burial in the summer, the decay rate might be slower than
when carcasses were above ground due to the cooler and more stable temperatures in the burial
trench. In winter, the temperature of buried carcasses and the surrounding subsurface soils (e.g.,
9 Although not reported by Young et al. (2001), ambient temperatures were likely between 40 and 70°F in the spring, summer,
and fall of 2001 in Wales and England where the outbreaks occurred.
21
-------
11°C or 52°F) might be well above the freezing ambient air temperatures possible across much
of the United States.
No data are available to describe the decay rate of FMDv in cattle carcasses under any defined
conditions. The assumed first order biological decay rate in the source for the base case is 0.12
per day (1.4E-06 per second), which is based on measured biological decay of FMDv in cattle
liquid manure at 17°C of 0.05 logio TCID50 per day (Schijven et al. 2005; Table 1). These rates
represent the fraction of viable virus particles that become unviable per unit time. Biological
decay is included in the uncertainty analysis for this assessment because of the limited
information on which to base an estimate, as well as the potential for actual rates to be affected
by environmental conditions.
Concentration In leachate
The resulting assumed pathogen loss of FMDv to leachate from one carcass in the temporary
storage pile for the base case is 9.4E+08 TCID50 per day. Derivation of this value takes into
account both the release of FMDv to leachate (i.e., loss per day; 0.1) and the biological decay
(i.e., loss per day; 0.12). The concentration of FMDv in leachate produced by carcasses in the
burial trench would be lower than this because all virus in the cattle musculature is inactivated by
rigor mortis during the two-day storage prior to burial. If the carcasses are placed in the trench
immediately after death without temporary storage, the virus loading would be the same as the
storage pile.
3.2 Fate and Exposure Estimation Methods
The methods described in this section simulate processes that occur between FMD source
locations and the locations where live cattle are exposed. These processes include dispersion of
FMDv in air, deposition from air to plants and soil ingested by grazing cattle, and leaching from
the temporary storage pile and burial trench to ground water. Each of these fate and exposure
processes requires some time, and so should include the estimates of biological decay.
3,2,1 Air Dispersion Modeling
Several models have been developed to simulate air-borne spread of FMDv over short and long
distances between the time a herd is infected and the time when it is reported and animals are
culled (Gloster et al. 1981, 1982; Moutou and Durand 1994; Sorensen et al. 2000; Mikkelsen et
al. 2003; Garner et al. 2006). The current assessment differs from those assessments in that only
the infectivity of animals that have just been euthanized (culled) is considered (i.e., the virus is
no longer replicating within animals nor are any more animals in the herd infected after the
outbreak is recognized and each animal is culled).
Dispersion of airborne virus particles is modeled with the American Meteorological Society
(AMS)/USEPA Regulatory Model Improvement Committee Model (AERMOD) (version 14134)
for air dispersion.10 AERMOD calculates air concentrations and rates of wet, dry, and total
deposition to the ground resulting from FMDv released to air from an area source11 with
10 Complete documentation of AERMOD and related tools, including AERMOD, AERMET, and AERSURFACE, is available at
https://www.epa.gov/scram/air-quality-dispersion-modeling-preferred-and-recommended-models.
11 An "area source" is used with AERMOD and other air dispersion models when emissions emanate from an area instead of a
"point source."
22
-------
horizontal dimensions equal to the assumed size of the pile (i.e., 96.7 m2 for 100 cattle). The
assessment assumes emissions originate at the height of the pile. The assessment also assumes
that FMDv is emitted from the storage pile at a continuous rate of 116 TCID50 per second (also
equivalent to counts per second and derived from the 24-hour aerosol release rate of 1.3E+05
TCID50 from Sorensen et al. 2000). Rigor mortis is not relevant to this release rate because all
particles are released from the surfaces of the animals. Once emitted, FMDv inactivates at a rate
of 0.019% per second (estimated from biological decay rate of 0.5 logio FMDv particles per hour
reported by Donaldson and Ferris, 1978, as cited by Garner et al., 2006, and Sorensen et al.,
2000).
AERMOD calculates average hourly air concentrations and deposition rates for each hour during
the full year of meteorological data (described further in USEPA 2017), with the source emitting
continuously at a constant rate. All estimated air concentrations are in units of TCID50 per cubic
meter (m3), and deposition rates are in units of TCID50 per m2 per hour. Concentrations and
deposition rates are calculated at 304 locations on a radial grid centered on the source: each of
the 16 radial lines is separated by 22.5° and includes 19 locations (at 0.1 km intervals from the
source to 1 km, and at 1 km intervals thereafter to 10 km).
Because carcass storage is assumed to last 48 hours, the hourly results are processed to find the
highest 48-hour average air concentrations during the year for each location. For comparison
purposes, all results are also recorded for 24-hour and 1-hour averaging periods. For deposition,
the results are processed to find the highest 48-hour total deposition at each location. Separate
results are obtained for wet, dry, and total deposition at each location.
3,"4 mcentrations of FMD In Surface Soil
The deposition results discussed in the previous section (3.2.1) are used to calculate
concentrations of FMDv in surface soil. Deposition processes in AERMOD include both wet and
dry deposition based on the meteorological data for the 48-hour period resulting in the highest
deposition to ground level. Before the soil concentrations are calculated, a first-order decay
equation is used to estimate the amount of viable virus remaining at the end of the 48-hour
deposition period. Equation 3.1 calculates the amount of viable virus remaining after a specified
number hours of decay.
vDP(t) = vDP(0) * e~XH (Eqn. 3.1)
where:
vDP(t) = Viable virus particle deposition (TCID50 per m2 per hour) at time = t hours
vDP(o) = Viable virus particle deposition from AERMOD (TCID50 per m2 per hour)
at time t = 0
X = Fraction of viable virus particles inactivated per hour (i.e.,
biological decay, 5.0E-03 per hour in air).
t = Time, number of hours of decay (t = 1 to 48), hours
23
-------
Equation 3.1 is used to estimate the amount of viable FMDv present at the end of the 48-hour
deposition event. The decay rate (X) is the rate per second discussed in Section 3.1.2 converted to
an hourly rate. Virus particles deposited during the first hour have 47 hours (t = 47) of decay
before the end of the event. Virus particles deposited during the second hour have 46 hours (t =
46) of decay. These calculations continue for each hour of the event until no decay is applied to
the final hour. The hourly viable virus deposition amounts are totaled, see Equation 3.2, for the
count of viable FMDv particles at the end of the 48-hour event.
ZKoVDP(t) (Eqn.3.2)
In the days following the 48-hour event, cattle continue to graze in the deposition area, and the
amount of viable virus decreases each day with continuing decay. Equation similar to Equations
3.1 and 3.2 are used to estimate the amount of viable FMDv remaining each day for 21 days after
deposition ends. The daily estimates are used to calculate the total FMDv ingestion through the
21st day. Day 21 was chosen as the endpoint of estimating ingestion exposure because by that
time decay has reduced the viable FMDv to the point that daily incremental exposure is less than
1%.
Concentrations of FMDv in soil are calculated using Equation 3.3 (below) based on USEPA's
(2005) Human Health Risk Assessment Protocol for Hazardous Waste Combustion Facilities
(HHRAP). In Equation 3.2, the total 48-hour deposition of viable virus particles is mixed with
the surface soil layer. The resulting estimate, Cs, is the concentration of TCID50 per kilogram
bulk soil at the deposition location.
Cs = \vDp(t)\ / (Zs * BD) (Eqn. 3.3)
where:
Cs
Concentration of viable virus in surface
soil, from deposition, TCIDso/kg
vDp(t) =
Total viable particle deposition over 48
hours, TCIDso/m2
Zs
Soil mixing zone depth (m)
BD =
Soil bulk density, kg/m3
For the base case, soil parameter values are HHRAP default assumptions. Specifically, HHRAP
assumes that deposited particles mix with the top 0.02 m (0.79 inches [in]) soil layer. HHRAP
also provides default assumptions for bulk-soil density at 1,500 kg per m3 (surface soil,
unsaturated).
Losses from soil due to erosion and leaching are assumed to be insignificant during the 48-hour
deposition period.
3,2,3 Concentrations of FMD m Feed
HHRAP (USEPA 2005) provides methods to estimate chemical exposures of livestock farmed
for beef, dairy, poultry, eggs, and pork products. Included in the HHRAP methods are equations
24
-------
to estimate contaminant levels in livestock feed. These are adapted for this project to estimate the
amount of viable virus present in vegetation (assuming grasses and herbs) grazed by cattle.
Equation 3.4 uses the wet and dry deposition rates from AERMOD, corrected to account for
biological decay following deposition as described in Section 3.2.2, to estimate the amount of
viable virus on plant surfaces (i.e., grasses and herbs grazed by cattle, also called "forage") at the
end of the 48-hour deposition period.
Cp = \vDpd + (vDpw * Fw)~\ * Rp* Kp / (kp * Yp) (Eqn. 3.4)
where:
Cp = Infective FMDv on aboveground vegetation due to particle deposition,
TCID50/ kg plant, dry weight (DW)
vDpd = Viable FMD particle dry deposition from AERMOD, TCIDso/m2
vDpw = Viable FMD particle wet deposition from AERMOD, TCIDso/m2
Fw = Fraction of wet deposition that adheres to forage plant surfaces (unitless),
HHRAP default of 0.6
Rp = Interception fraction of the edible portion of plant tissue for the plant type,
unitless, HHRAP default of 0.5 for forage
Kp = Plant surface loss coefficient, 0.1 per event (see below)
Yp = Yield or standing crop biomass of the edible portion of the plant
(productivity) (kg DW/m2), HHRAP default of 0.24 for forage
Default values for the parameters Fw, Rp, and Yp are documented in USEPA (2005). The plant
surface loss coefficient (Kp) accounts for loss of particles from plant surfaces with time due to
wind removal, water removal, and growth dilution. HHRAP recommends a default value of 18
per year. Converted for the 48-hour event period used in this assessment, the Kp value is 0.10 per
exposure event. The HHRAP default is based on half-life data for a variety of contaminants on
plant surfaces reported by Miller and Hoffman (1983). These data are assumed to represent the
half-life of virus particles sticking to plant surfaces.
The amount of viable FMDv remaining on aboveground vegetation is estimated each day for 21
days after the end of the 48-hour deposition event. The daily estimates are used to account for
ongoing decay as cattle graze in the deposition area following the end of the deposition event.
3,2,4 Concentrations In Ground Water
If there is no barrier between the carcasses in the storage pile and the ground below, FMDv
particles in carcass leachate could seep downward through the unsaturated soil zone until
reaching ground water. Exposure could occur if ground water drawn from a well downgradient
from the carcass management location is used to provide drinking water for healthy cattle. The
nearest cattle would be at a distance from the source, because all cattle at the affected farm, and
25
-------
possibly other farms in the response zone, would be culled to contain the outbreak (see Section
2.2).
Two steps are used to evaluate the fate and transport of FMDv particles in the soil-to-ground
water pathway. First, the Virulo model developed by Faulkner et al. (2002a) is used to model the
movement of virus particles downward through the unsaturated soil zone, and the probability that
they reach ground water. Virulo does not estimate the amount or concentration of virus particles
reaching ground water and it does not model the fate of viruses after they reach the ground water.
In addition, it does not include biological decay.
In the second step, simple spreadsheet calculations are performed to estimate FMDv
concentrations in well water with the effects of biological decay and dilution. These calculations
overestimate FMDv in well water because they do not include the complex sorption-desorption
dynamics addressed in the Virulo model. However, comparing the leachate and well water
concentrations provides a conservative estimate of the reduction in concentration and potential
exposure. In addition, the base case results can be compared to results calculated with varied
assumptions (e.g., soil depth, numbers of carcasses) to examine how the varied factors
individually affect potential exposures.
Leaching to Ground Water Analysis with Virulo
Virulo (Faulkner et al. 2002a, b) is a screening model that uses probabilistic Monte Carlo
simulations to predict virus transport and survival through soils. It is based on a conceptual
model that simulates several natural processes and forces that influence water flow and virus
transport in variably saturated soils. The model uses built-in distributions of physical, biological,
and chemical factors and a set of default properties virus and soil types. The documentation for
the Virulo model lays out the limitations of the model, "In instances where the ground-water
system in question is connected to potential virus sources by karst, fractured rock, gravel, or a
soil exhibiting preferential flow, the system will be classified as high risk. In other cases the
assessment process will benefit from prediction by mathematical modeling" (USEPA 2002a).
For those cases where mathematical modeling is appropriate, important assumptions in the
Virulo modeling approach are identified below.
• Water flow is one-dimensional, vertical, and uniform (i.e., the soil is homogenous with
respect to geochemistry and hydraulic properties), although degree of soil saturation
varies.
• Flow has reached steady-state.
• Water moves downward under the force of gravity only (there are no abrupt changes in
capillary pressure in the soil).
• Water content is variable, simulating instantaneous and random recharge from
precipitation (rather than cyclical wetting and drying).
• There are no preferential flow pathways (e.g., root pores).
• Virus transport is simulated by linear absorption-desorption processes typical of
dissolved chemicals rather than by colloidal filtration.
The Virulo model includes "instantaneous" equilibrium, and therefore does not include time or a
biological decay rate.
26
-------
Virulo output is expressed as the probability that virus particle attenuation does not equal or
exceed a target level of attenuation by the time the leaked fluids with virus particles "break-
through" to ground water. The default attenuation target is 99.99% (i.e., fails to achieve a four-
fold log 10 reduction in virus concentration). The user can choose the attenuation target level,
values for various soil and virus properties, and the number of Monte Carlo simulations.
In the Virulo documentation (Faulkner et al. 2002a), USEPA provides example results of a
Monte Carlo simulation for polio virus attenuated by 1 m of soil between the release source
(breach of septic system) and ground water for sand, silt-loam, and clay. The example results use
model default values, including a 4 loglO (i.e., 99.99%) attenuation target. The results are
summarized below for the three soil types.
• Sand — The probability of failure to attain 99.99% reduction attenuation was 22
simulations divided by the number of simulations (5697), or 0.39% failure.
• Silt-loam ~ Six simulations out of 2 million (i.e., 0.0003%) failed to reach 99.99%
attenuation.
• Clay — There were no failures to reach 99.99% attenuation out of 9 million runs
(Faulkner et al. 2002a).
These findings suggest that polio virus has a low probability of reaching ground water at a
minimum depth of 1 m with less than 99.99% reduction, particularly at sites with silty-loam or
clay soils.
For this assessment, Virulo was used to examine attenuation of FMDv attenuation with 1 to 8 m
depths of silty loam. In addition to the default attenuation target (i.e., 4 loglO), simulations were
run for attenuation targets of 5 loglO through 8 loglO at intervals of one order of magnitude. The
more stringent attenuation targets were included because higher viral loading rates at the surface
would require a higher target attenuation to reach a specified concentration (e.g., based on
infectious dose and cattle water ingestion rates) entering ground water.
Default inputs for FMDv are not available in Virulo. As a substitute, simulations were run using
a default soil/liquid partition coefficient (Kd) value for another member of the Picornaviridae
family, Echovirus. Specifically, the estimates were made with the Echovirus-clay Kd value of
4.5E-04 m3/g (or 453.5 L/kg). For comparison, the default Kd values for Echovirus in silty-loam
is similar (442 L/kg), and for Kd for Echovirus in sand is higher (744 L/kg). Figure 3.1
summarizes the results.
Figure 3.1 shows that between 40 and 50 out of 10 million simulations failed to achieve
attenuation of 99.99% (i.e., 4 loglO). More simulations failed to achieve the higher attenuations
targets (as shown in Table 3.1). Notably, the probability (or risk) of failing to reach the
attenuation target is similar with soil depths ranging from 1 to 8 m, and the estimated number of
failures do not necessarily increase with soil depth. This indicates that the estimates may be more
sensitive to parameters varied in the Monte Carlo simulations than to soil depth.
The simulations with Virulo indicate that FMDv released from the temporary carcass storage pile
or burial trench will have a very low probability of reaching ground water when the depth of the
ground water is at least a meter. These results alone overestimate the likelihood of exposure to
27
-------
cattle because Virulo does not address dilution in ground water or biological decay in either the
vadose zone or ground water. These two factors examined in the sections below.
140
120
100
80
~ Q) 60
in 4-i
40
20
I
5 6 7
x-log Attenuation
1.0 meters
2.0 meters
4.0 meters
8.0 meters
Figure 3-1. Number of Monte Carlo simulations (out of 10 million) that failed to reach
attenuation target at four soil depths.
Table 3-1. Estimated risk of FMDv Breakthrough to Ground Water at Soil Depths of 1-8 m
Target Attenuation (%)
Number of 10-fold
Reductions
Low-end Risk of
Failing to Achieve
Target Attenuation
High-end Risk of
Failing to Achieve
Target Attenuation
99.99
4
4.0E-06
5.0E-06
99.999
5
5.8E-06
8.4E-06
99.9999
6
7.2E-06
9.3E-06
99.99999
7
8.0E-06
9.5E-06
99.999999
8
9.8E-06
1.3E-05
28
-------
Dilution and Biological Decay Calculations
This section describes simple calculations to examine the extent to which dilution and biological
decay processes, individually and combined, reduce concentration of FMDv in a downgradient
well used to provide drinking water for cattle. Because the estimates presented in this section do
not include the complex processes simulated by Virulo, the difference in estimated
concentrations at the carcass storage pile and the well represents a conservative level of
reduction in this exposure pathway. In addition, an uncertainty analysis with the base case
estimate will examine how the level of exposure varies by soil type, soil depth, scale of
mortality, and other factors.
Dilution Attenuation
After seeping into the ground beneath the temporary storage pile or burial trench, leachate is
subjected to dilution and other physical, chemical, and biological processes that attenuate
leachate constituents. To support regulatory analyses, USEPA (1996) created the USEPA
Composite Model for Leachate Migration with Transformation Products (EPACMTP) to
simulate dilution attenuation in both the unsaturated and saturate zones. In an application of this
model, USEPA developed a set of dilution attenuation factors (DAFs), ratios of leachate
concentrations at the source to the concentration in water at a downgradient well. With a DAF of
one, a constituent concentration at the well would equal concentrations at the source. DAFs
greater than one indicate dilution and attenuation the constituents before reaching the well.
USEPA developed the DAFs by running Monte Carlo simulations with EPACMTP and
nationwide data sets for waste sites and hydrogeological parameters (USEPA 1996). Simulations
were run with six well-placement scenarios that included well distances of 0 m, 25 m, or 100 m,
or distances randomly selected from a distribution of nationwide data. The well's horizontal
offset distance from the plume center line was randomly selected, either within the plume's
width or half the width. Well depths were randomly selected from nationwide data for most
scenarios.
Because sensitivity analyses determined that soil types and the size of the contaminated area
have the greatest effect on the DAFs, USEPA developed DAFs for a sources ranging in size from
1,000 to 5,000,000 ft2 (93 to 464,515 m2). With further analysis, USEPA prepared a default
nationwide DAF of 20 for sources up to 0.5 acres (0.2 hectares).
For this assessment, the DAFs produced using the EPACMTP Monte Carlo analysis are used to
estimate TCID50 concentrations in well water 100 m downgradient from a temporary storage pile
or burial trench. Because DAF are sensitive to the size of the leachate source, the area of the
storage pile was matched to the distribution of DAF values by size presented by USEPA (1996,
Appendix E). For each source size, USEPA presented DAFs corresponding to the 85111, 90th, and
95th percentile of Monte Carlo simulations. Because USEPA based the default DAF on 90th
percentile results, the DAFs for this assessment were based on the 90th percentiles as well. The
DAF applies to leaching from the temporary storage pile with the base case (i.e., management of
100 carcasses) is 1,675. For comparison, the DAFs for storage piles with 1,000 and 10,000
carcasses are 201 and 24, respectively.
29
-------
The EPACMTP modeling effort described above included simplifying assumptions that make
the estimated DAFs conservative. For example, retardation due to absorption/desorption kinetics
were excluded by assuming that soil and porewater concentrations are at equilibrium. In
addition, chemical and biological degradation processes were not considered (USEPA 1996).
Thus, the modeling approach is likely to overestimate chemical concentrations in ground water.
Biological Decay
The effect of biological decay on FMD exposure depends on the inherent biological decay rate of
the virus, which may vary with changes in temperature, moisture, and other environmental
conditions. As discussed in Section 3.2, leachate modeling this assessment uses an FMDv decay
rate of 0.12 TCID50 per day (1.4E-06 per second), which was measured by Schijven et al. (2005)
in liquid cattle liquid manure.
The amount of decay also depends on the time elapsed between release from the storage pile or
burial trench and ingestion by cattle. The time is, in turn, is determined by a number of site-
specific factors including, (1) soil depth and type, (2) the downward velocity of the leachate in
unsaturated soil before reaching ground water, (3) the horizontal ground water flow rate, and (4)
the distance to the well. The data and assumptions in this assessment for each these four factors
are discussed below.
With the estimated FMDv concentration in leachate, biological decay rate, and an estimated
travel time to the well, the amount of viable FMDv in well water is estimated with formula
similar to Equation 3.1.
Soil Depth and Type
Many states recommend or mandate minimum depths of unsaturated soil beneath carcass burial
pits to protect ground water quality. These distances are as little as 1 ft (-0.3 m), but are more
typically 3 ft (~1 m) or more (NABCC 2004). Based on this information, the default soil depth
for this assessment is 1 m. To examine how soil depth affects exposure estimates, the assessment
also includes depths ranging from 0.5 to 6.
The assessment includes three soil types (sand, silty-loam, and clay) with distinct characteristics
(e.g., porosity). These soil types were included in the Faulkner et al. (2002a) example Virulo
analysis with polio virus, described above, which showed that virus mobility differs by soil type.
Downward Velocity in Soil
"Average" downward water velocities (i.e., discharge velocity or apparent velocity), based on
summaries provided by the USDA Natural Resources Conservation Service (USDANRCS
2008) and the United Nations Food and Agricultural Organization (UN FAO 2006), are listed in
Table 3.2 for sand, loam, and clay soils. To approximate the time of travel for a virus particle
from the ground surface to the ground water, the depth of soil above the water table can be
divided by these downward velocities, as shown in Table 3.2 for the base case depth of 1 m.
These simple calculations overestimates the rate of travel for virus particles because they do not
30
-------
account for absorption-desorption processes, which retard their movement as demonstrated by
the Virulo simulations above.
Table 3-2. Average Downward Travel Velocities and Time to 1 m Depth in Unsaturated
Soils
Soil Type
Average Downward Water Velocity
(cm/day)
Average Time to
Breakthrough (day)
Sand
100a
1.0
Loam
18b
5.6
Clay
2.5°
40
Abbreviations and acronyms: cm = centimeter; g = gram
Complete references are found at the end of the report.
a For sand, USDA NRCS (2008) lists > 49 cm/day and UN FAO (2006) lists 5 cm/hr (120 cm/day), while other
sources suggest 1,000 cm/day possible for very coarse sand. Value of 100 cm/day used.
bFor loam USDA NRCS (2008) states 12 to 24 cm/day. Average of NRCS range used here.
°For clay, USDA NRCS (2008) lists 2.5 to 12 cm/day, while UN FAO (2006) lists 1.2 cm/day for clay; low end of
NRCS range used here.
31
-------
Ground Water Flow Rate
Ground water flow rate is one of the parameters varied in USEPA's (1996) Monte Carlo
modeling to develop DAFs. Specifically, USEPA used a nationwide probability distribution of
ambient ground water velocities, with 15th, 50th, and 85th percentile values of 53.2, 404, and 2883
m/yr, respectively. The 50th percentile (i.e., median) value is selected as the default ground water
flow rate for this analysis.
Well Distance
As discussed above under Dilution Attenuation, the well distance assumed for this assessment is
100 m, which is the value associated with the USEPA DAFs presented in Table 3.2. It also is the
minimum distance included in the AERMOD dispersion modeling. A distance of 100 m to the
nearest well is a conservative assumption for the FAD outbreak scenario. State or local
regulations in some areas will preclude siting a burial trench 100 m from a well. However, the
minimum required distance is less than 100 m in several states (NABCC 2004). In addition, live,
uninfected cattle are very unlikely to be kept as close as 100 m from the carcass management site
(e.g., temporary storage pile). However, the cattle do not need to be at the well location to
receive water from the well.
Using survey data, USEPA (1997) prepared a probability distribution of the nearest well
distances. Only wells within 1 mile (1609 km) from a landfill were included in probability
distribution. Considering only those wells, the chance of the nearest well beinglOO m or less
from a landfill is approximately 10%. The 50th percentile distance is 427 m, and a distance of 1
km corresponds to approximately the 80th percentile. Although the proximity of wells to landfills
is not necessarily representative of well distances to carcass management locations, these data
suggest that the assumption for this assessment is reasonable yet conservative.
It the well is assumed to be 100 m directly downgradient from the temporary storage pile or
burial trench, and virus particles move toward the well at the average annual ground water flow
rate (i.e., 404 m/yr discussed above), the travel time from breakthrough directly beneath the
source to the ground water well is 0.247 yr, or 90 d. This estimate is used for all three soil types;
the DAF is not specific to a soil types because it is based on a nationwide distribution of soil
data.
Estimated Well Water Concentration
Table 3.3 compares estimated FAD virus concentrations (in TCID50/L) in leachate and in well
water with and without the effects of dilution attenuation and biological decay. As discussed
previously, these estimates are provided to show the amount of reduction that can be attributed to
these factors individually and together. For the base case, in which 100 carcasses are placed on
bare earth 100 m upgradient from a water well, dilution attenuation is estimated to reduce the
concentration of FMDv particles in water by about three orders of magnitude (i.e., by a factor of
1675). The DAF that is the basis for this difference was developed by USEPA using nationwide
field data and might under or overestimate dilution attenuation at actual sites.
Biological decay is estimated to have a greater effect than dilution attenuation for the base case.
The reduction from biological decay ranges from approximately 4 orders of magnitude with
32
-------
sandy soil to approximately 6 orders of magnitude with clay. These estimates are based on the
same FMDv decay rate measured in liquid cow manure by Schijven et al. (2005). Differences in
the estimated decay by soil type are caused by different water velocities reported by USD A
NRCS (2004) and UN FAO (2006). Slower velocities are associated with longer travel times and
thus greater amounts of decay. The decay estimates also include decay in the ground water
aquifer, which is calculated with the median ambient ground water velocity in nationwide site
data. The flow velocities might over or underestimate values at actual sites. However, decay is
calculated for a well 100 m distant, which is a conservative assumption. Biological decay would
be greater than estimated if the well is more than 100 m away.
The concentration estimates in Table 3.3 overestimate concentrations at the well because they do
not include sorption-desorption processes that retard the movement of the virus particles, leading
to further biological decay and attenuation. The Virulo analysis above indicates a very low
probability (i.e., approximately 5.0E-06) of less than 99.99% attenuation before virus particles
reach the ground water aquifer.
33
-------
Table 3-3. Estimated Concentrations of Infective FMDv in the Ground Water Pathway for
the Base-case
Concentration Estimate Basis
Average Concentration (TCID50/L)
Sand
Loam
Clay
Leachate at the Storage Pile
8.8E+07
Well Water with Dilution Only
5.3E+04
Well Water with Biological Decay Only
1.5E+03
8.4E+02
1.3E+01
Well Water with Dilution and Biological Decay
8.7E-01
5.0E-01
7.8E-03
TCID50 = 50 percent tissue-culture infectious-dose
3.2.5 Cattle Exposure Factor Values
Dairy and beef cattle differ in the amount of air they inhale and in the quantity of food, soil, and
water ingested.
Inhalation
¦ Respiration Rates: At rest, dairy cattle take 26-50 breaths per minute (Merck 2015), while
beef cattle, which typically weigh substantially more, breath more slowly, approximately
10-30 breaths per minute (Ensminger 1992).
¦ Tidal Volume: The tidal volume of air inhaled for each breath is 7.0-8.0 milliliters (mL)/kg
body weight (UWM RARC, Normative Data for Cattle, undated).
¦ Air inhalation Rate, Dairy cattle: 40 (breaths/min) * 7 (mL) * 600 (kg) * 60 (min) =
10,080,000 mL/h = 11 m3/h.
¦ Air inhalation Rate, Beef cattle: 20 (breaths/min) * 8 (mL) * (1,000) kg * 60 min =
9,600,000 mL/h = 9.6 m3/h.
Incidental Soil Ingestion Rate
HHRAP (USEPA 2005) methods for estimating chemical uptake by livestock include the
incidental ingestion of soil by grazing cattle. Based on a review of available literature, USEPA
(2005) recommends a default incidental soil ingestion rate of 0.5 kg per day by grazing, non-
dairy cattle.
Forage Ingestion Rate
Based on a review of available literature, USEPA (2005) recommended assuming a total daily
feed intake for cattle of 12 kg dry weight (DW) per day. In HHRAP methods, the cattle diet
includes forage, silage, and grain. This exposure assessment assumes a forage-only diet. Both
grain and silage, which is forage that has been stored and fermented, require processing before
use as feed. That means that the grain and silage consumed on the day immediately after the
deposition event would be uncontaminated with FMD from deposition, having been harvested at
an earlier time and likely a different location. Therefore, it is conservative to assume an all-
forage diet for the exposure assessment.
34
-------
Water Ingestion Rate
In general, cattle drink more water in the summer than in the winter because they pant to lose
excess body heat, which also increases loss of water vapor via the lungs. Dairy cattle drink more
water than beef cattle (Agriculture and Agri-Food Canada, undated).
¦ Water Ingestion Rate - Dairy cattle: 95 L/day (summer), 77 L/day (winter).
¦ Water Ingestion Rate - Beef cattle: 86 L/day (summer), 55 L/day (winter).
Because these rates are all within the same order of magnitude, the exposure estimates are
expected to be less sensitive to the water ingestion rate than to other parameters included in the
uncertainty analysis. Therefore, a single ingestion rate assumption, 95 L per day, is used for the
assessment.
Summary of Exposure Factor Values
Exposure factor values for dairy and beef cattle are listed below in Table 3.4. Bold text indicates
the exposure factor values used for the results presented in Section 4.
35
-------
Table 3-4. Summary of Exposure Factor Values for Cattle3
Exposure Factor
Parameter
Symbol
Dairy Cattle
Beef Cattle
Respiration rate (breaths/min)
Resp
26-50 (use 40)
10-30 (use 20)
Tidal volume (mL/breath/kg body weight)
TV
7
8
Inhalation rate (m3/h per animal)
Inh
11
9.6
Soil ingestion rate (kg/day)
Qs
0.5
Forage ingestion rate (kg dry forage/kg fresh
body weight)
Op
12
Summer water ingestion rate (L/day)
SWir
95
86
Winter water ingestion rate (L/day)
WWir
77
55
Abbreviations and acronyms: min = minute; m
L = milliliters
a Exposure factors shown in bold are used in the exposure assessment.
3.2.6 Exposure Estimation
In Section 4, exposure estimates are presented separately for inhalation, ingestion of forage and
soil, and water ingestion. All exposures are evaluated in units of total TCID50 during the
exposure event. For the storage pile, the exposure event for all pathways is 48 hours in duration.
For carcass burial, exposures from inhalation and ingestion of forage and soil are assumed to be
negligible because burial beneath soil prevent suspension of virus particles into air. Water
ingestion exposures from burial theoretically could occur over a period of weeks or months,
however exposures would decline to negligible levels very quickly due to the combined effects
of reduced pH in the carcasses, biological decay of FMDv with time, and attenuation of virus in
daily leachate as the source is depleted.
Inhalation
The total TCID50 inhaled by one animal during the period is calculated by multiplying the hourly
inhalation rate (Table 3.4) by the number of hours of exposure and the 48-hour average air
concentration during exposure (Equation 3.5). The exposure duration (ED) equals the period
over which the air concentration is averaged. Exposures are estimated for each distance from the
source using the highest 48-hour average concentrations (i.e., in any direction or 48-hour period).
Einh = Inh * ED* Ca (Eqn. 3.5)
where:
Einh = Event total inhalation of viable FMDv, TCID50
Inh = Inhalation rate, m3/h
36
-------
ED = Exposure duration, 48 hour/event
Ca = 48-hour average concentration of viable virus in air, TCIDso/m3
Forage and Soil Ingestion
Equation 3.6 shows how estimates of viable FMDv in soil (Section 3.2.2) and forage (Section
3.2.3) are used to calculate the total FMDv ingested daily by cattle after the 48-hour deposition
event. This equation is adapted from HHRAP (USEPA 2005).
Eingps = (Qp * Cp) + (Qs * Cs) (Eqn. 3.6)
where:
Eingps = Event total ingestion of viable FMDv, TCID50
Qp = Quantity of forage eaten by a cow per day, kg DW/d (dry weight per day)
Qs = Quantity of soil eaten by a cow each day, kg DW/d
Cp = Concentration of virus deposited on forage plants, TCIDso/kg DW
Cs = Concentration of virus deposited on surface soil, TCIDso/kg DW
In Equation 3.6, the estimated TCID50 concentrations in soil (Cs) and forage (Cp) are multiplied
by daily ingestion rates (Qs and Qp, respectively). Although deposition ends after 48 hours, the
deposited virus particles remain available for ingestion in the days and weeks afterward.
However, the amount of deposited virus available for ingestion decreases each day due to
biological decay. For a series of days, the biological decay rate (1.4E-06 per second) converted
to a daily rate is applied with an equation similar to Equation 3.1, to calculate the amount
ingested each day. By day 21, biological decay diminishes the daily incremental exposure to 1%,
meaning that longer durations have little effect on the total ingestion. Therefore, cattle are
assumed to graze for 21 days in an area that receives the maximum deposition estimated by
AERMOD. Total ingestion is the sum of the first 21 daily ingestion quantities. In this approach,
the amount of virus available for ingestion each day is not affected by grazing in the area on
prior days.
Water Ingestion
The total viable FMDv ingested in drinking water during the 2-day exposure is shown in
Equation 3.7.
Eingw = Ingw * ED* Cw (Eqn. 3.7)
where:
Eingw = Event total ingestion of viable FMDv, TCID50
Ingw = Water ingestion rate, L/day
37
-------
ED = Exposure duration, 2 days/event
Cw = Concentration of viable virus in water supplied to cattle, TCID50/L
As explained in Section 3.2.4, the concentration of viable virus in water, Cw, is less than the
initial concentration of virus in leachate due to dilution and biological decay of virus between the
storage pile and water well. However, the calculations to estimate concentrations of infective
FMDv in well water do not include sorption-desorption processes that would decrease the
concentrations further. Thus, the estimated concentrations overestimate exposure.
3.3 Vectorborne Transmission
No measured data attributing FMD infection from cattle carcasses via vector-borne transmission
have been identified, and no methods were identified to estimate transmission, including
mechanical transmission, via living vectors to susceptible livestock. Thus, the potential for
transmission where vectors could come in direct contact with carcasses is assessed qualitatively
based on information on the vectors' ability to carry and/or to transmit the disease to susceptible
species at distances of up to 10 km. Insects, birds, and other land animal vectors are discussed
and evaluated separately below.
Insects, birds, and/or scavenging animals could become vectors by contacting the carcasses
during handling, going on or in the storage pile, materials released from the storage pile (e.g.,
leachate), or burrowing into the compost windrow or burial trench. However, by the time
carcasses are place in a burial trench or compost windrow, they have passed through rigor mortis
and viable FMDv would remain only on carcass surfaces and in some deep tissue compartments.
When carcasses are in the storage pile, scavenging animals, birds, and insects might contact
infected carcasses. During handling, only insects might contact infected carcasses, because
handling carcasses involves human actions, which would deter scavenging wildlife from
attempting to reach the carcasses. The storage pile is covered with tarp(s) when humans are not
present; however, the tarp must be strong and anchored to the ground to be secure from the
larger, stronger scavenging mammals (e.g., feral swine) and birds (e.g., ravens, vultures).
Scavenging animals can carry virus particles either externally or internally (e.g., if they fed on
carcasses). Externally, infectious FMDv particles could adhere to wildlife surfaces (e.g., cuticle,
scales, feathers, and fur) at the infected farm and travel with the wildlife to neighboring farms,
where virus particles might drop off and become available to other susceptible species (Cottral
1969). Various species of birds, insects, and scavenging carnivorous mammals, such as coyotes
and wolves, in which FMDv does not survive internally, are considered to be potential fomites
that may contact off-farm livestock (Cottral 1969).
Some species of insects and mammals are natural hosts of FMDv, and can carry viable internal
FMDv, which can also replicate (USDA/APHIS 1994). Deer, moose, and bison are cloven-
hooved, obligate herbivore, ruminant ungulates, which like sheep, goats, and cattle, are
susceptible to falling ill if infected with FMDv (USDA/APHIS 1994). Swine, both domestic and
wild, are also cloven-hooved ungulates which are natural hosts susceptible to FMD, although
they are not ruminants; instead they eat almost anything including animal matter. Some other
mammals, including rats and gray squirrels, also are hosts of viable FMDv; although they
generally do not develop illness (USDA/APHIS 1994). Insects categorized as natural hosts
38
-------
include house and biting flies (USDA/APHIS 1994). A few other invertebrates, including
earthworms and ticks, also are natural hosts (USDA/APHIS 1994).
Natural hosts can be categorized by "carrier length," which is defined as the maximum reported
duration of carrier status (i.e., carrying viable, replicating virus but free from illness) or viral
shedding (USDA/APHIS 1994). Carrier length probably varies among host species, but few data
are available. At least one white-tailed deer remained a carrier of viable FMDv for 11 weeks
after infection (Arambulo and Steele 1977; USD A/APHIS 1994) and a 9-month carrier length
has been reported for sheep (USDA/APHIS 1994). FMDv has been carried on flies for up to 10
weeks and on ticks for 15 to 20 weeks, but the viability of those FMDv particles was not
reported (USDA/APHIS 1994).
The distances travelled by both natural hosts and mechanical vectors determines how far beyond
the site of the FMD outbreak those animals might carry the virus. Based on their home range and
activity patterns, deer and foxes might travel 10 km or more from the outbreak site (Ahlstrom
1983; Kramer 2015). However, not all natural hosts and/or mechanical carriers are capable of
traveling long distances. Rats typically travel only 30 m (100 ft) to 91 m (300 ft) from their nest
in search of food (County of Los Angeles, undated). In studies of house flies, 60-80% of marked
flies were captured within 1.6 km (1 mile) of their release point, while a smaller percentage of
flies were caught 3.2 km (2 miles) from the release site within the first 4 days after they were
released (Townsend 1997). Thus, it is unlikely that the typical housefly could travel to farms
located 10 km beyond the site of the outbreak. Ticks do not fly; they jump, and their horizontal
movement is limited to a few centimeters (University of Rhode Island 2016). However, once on
the skin of a mobile animal, bird or mammal, they could be transported beyond 10 km from a
carcass management location. At a new location, the ticks might drop off, molt, and reattach to a
different animal. The molt, however, would leave external FMDv particles on the ground.
For scavenging wildlife that are not natural hosts to spread FMDv, they would need to pull parts
of carcasses from the field, for example prior to collection for the temporary storage pile or from
the storage pile. In the field, workers should collect carcasses as they are culled, preventing
access for scavenging wildlife. Once carcasses are placed in the temporary storage pile, there
would be human activity in the vicinity whenever the tarp over the pile was opened (e.g., for
adding or removing carcasses), which would deter daylight scavengers (e.g., crows, ravens,
vultures). At night, when there might be no workers present, the tarp, presumably secured to the
ground, would prevent ready access by nocturnal scavengers (e.g., foxes).
If appropriate livestock-raising hygienic measures are used (i.e., relatively clean conditions; rat
control), the on-site temporary storage pile should not be within the normal foraging range of
mammalian scavengers, including rats. If workers discovered disturbances to the covered storage
pile in the morning, presumably they would find ways to re-secure the carcass protection and
establish scavenger control measures (e.g., on- and off-premises bait and capture stations). Even
if one or a few mammalian scavengers such as foxes and coyotes removed parts of carcasses and
carried them off the infected premises, they would likely cache (hide) the parts not eaten near
dens or feed the part to pups in dens, which are unlikely to be located in livestock pastures.
In areas with active wolf packs, which could conceivably tear open a tarp, remove substantial
quantities of carcasses in a single night, and carry them far distances, additional biosecurity
measures might be needed for a temporary storage pile. Similarly, in areas with feral swine, tarps
39
-------
would need to be of strong materials and well secured to the ground (e.g., staked at close
intervals).
FMDv can remain infective in bird feathers for 91 hours and in bird droppings for 26 hours
(Bullough 1942; Svidorov et al. 1974; Canadian Food Inspection Agency 2013), allowing
adequate time for birds to travel far from the site of the outbreak. However, birds must be
heavily contaminated with FMDv particles to transmit the disease as mechanical vectors (Wilson
and Matheson 1952; Canadian Food Inspection Agency 2013). Data reported by Dillon (2011)
indicate that the skin of an infected animal (e.g., cow) is a significant virus reservoir (see Section
3.1.1), and birds such as ravens and crows might come in direct contact with the skin of infected
carcasses during the day; however, they would not generally have sufficient time undisturbed
during the day to penetrate a tarp.
After considering information on the identity, host status, and transmission potential of insects
and wildlife scavengers, the following qualitative conclusions can be drawn about vector
transmission associated with the management options and activities:
¦ Handling: Insect vectors might contact carcasses during handling. Given their short-range
flights, however, mechanical transmission by flies beyond 10 km is unlikely. If all
susceptible livestock within the 10 km response area are not culled, insects pose a greater
threat of spreading the FMD outbreak.
¦ Temporary Storage Pile: Although the temporary carcass storage pile is assumed to be
covered with a tarp, some scavengers might smell the carcasses and attempt to dig through
the tarp when people are not around (e.g., at night for mammals). The tarp is moved for
short periods of time as additional carcasses are added to the pile. During this time, insects
might contact the carcasses; however, few individual insects are likely to travel 10 km or
more to nearby farms with susceptible animals. Scavenging birds and mammals might
penetrate a protective tarp when people are not around (e.g., at night) at the temporary
storage pile. Wildlife scavengers could carry and transmit FMDv farther than insects. In
areas with feral swine or wolves, additional protection could be required, such as
temporarily storing the carcasses in lined, leak-resistant roll-offs with secure tarps across the
top. Tarp integrity is key to preventing large masses of insects and any kind of scavenging
wildlife from contact with the carcasses; monitoring might be required, particularly if not all
livestock have been culled within the 10 km response area. The closer susceptible livestock
are to a temporary storage pile, the higher the risk that a vector that made contact with
infected carcasses could make contact with other susceptible livestock. The number of
vector species/types capable of making contact with other susceptible livestock increases
with decreasing distances between the storage pile and susceptible livestock.
¦ Burial: The bulk of viable FMDv in carcasses has been inactivated by low pH prior to
burial, unless carcasses are placed in the burial trench immediately after euthanasia. In either
case, viable virus could still be adsorbed to external surfaces of the carcasses. The burial
trench might present an opportunity for scavenging wildlife to contact carcasses. However,
the 6 feet of dirt covering the burial trench precludes typical scavenging mammals from
reaching the carcasses. Most burrowing small mammals only feed on live insects and other
invertebrates in the soils. Avian scavengers would not dig in soils; they only consume
carrion that is above ground or floating in water. It is unlikely that the denning activities of
foxes and other scavenging mammals would occur immediately over a burial trench.
40
-------
Exposure of susceptible livestock on farms outside the FMD response area, therefore, is
unlikely.
1 Composting: Again, the bulk of viable FMDv in carcasses has been inactivated by low pH
during the two days of temporary storage prior to composting, unless carcasses are
immediately placed on the compost windrow when culled. In either case, viable virus could
still be adsorbed to external surfaces of the carcasses. Avian scavengers are unlikely to
smell or see carcasses in compost piles. One of the functions of the bulking agent (i.e., wood
chips) over top of the windrow is to retain some chemicals that cause odors that might
attract scavenging mammals to the carcasses. USD A/APHIS recommends that the site be
fenced to preclude entrance of larger mammalian scavengers (USDA 2005). Depending on
the fencing material, it is possible that larger mammals, such as feral swine or wolves could
break through the fence and access the windrow. Monitoring of the integrity of the compost
pile covering and fencing is required to ensure that larger scavengers do not reach
composted carcasses, particularly in the early weeks of composting when a possibly
significant proportion of virus in and on livestock carcasses is still viable.
Overall, the likelihood of mechanical transmission of FMDv via insects and scavengers is low
for carcass handling and burial, and somewhat higher for the temporary carcass storage pile and
for composting. For the latter two cases, monitoring the integrity of protective measures (i.e.,
tarp, windrow covering, fence) is important to minimizing the chance of off-site transmission of
FMD.
41
-------
suits and Discussion
In presenting the exposure assessment results, this section evaluates whether managing FMD-
infected carcass at the outbreak farm site might infect healthy cattle in the surrounding area.
Section 4.1 discusses and compares the relative potential for exposures among all of the
management options (on-site open burning, air-curtain burning, unlined burial, and composting;
off-site fixed-facility incineration, lined landfill, and rendering), including those not
quantitatively assessed. Section 4.2 presents the quantitative exposure assessment using "base-
case" data and assumptions, which most closely resembles the case evaluated for natural
disasters. The base case uses a set of reasonably conservative values based on a review of
available literature and previously developed default assumptions for the hypothetical farm site.
The results presented in Section 4.2 are uncertain owing to several gaps in the available scientific
data on FMDv. In addition, natural variation in important environmental characteristics from one
location to another precludes use of a single scenario to represent possible future events.
Important variables that cannot be predicted prior to an event include the scale of mortality, the
type of soil, distance to nearest uninfected livestock herd, depth to ground water, and distance of
nearest ground water wells that might be used to water livestock. To examine how potential
exposures are affected by such factors, Section 4.3 presents an uncertainty analysis where the
base-case assumptions are systematically varied.
4.1 Qualitative Exposure Assessment
For reasons discussed in Section 2, exposures are not quantitatively assessed for the three off-site
management options and two of the on-site management options. Those options, along with
burial and composting, can be qualitatively evaluated based on the degree of thermal destruction
and containment provided by the carcass management options.
At high temperatures, if performed in accordance with permit requirements and best practices,
burning and incineration options are expected to effectively inactivate the FMDv. If open pyre
burning is not performed correctly, some external viable virus particles might rise with warm air
plumes as fires start and travel downwind to a neighboring farm. Thus, air-curtain burning,
which recirculates most fly ash several times resulting in more complete combustion, is less
likely than open-pyre burning to accidentally release viable virus particles to air.
For composting, a large proportion of viable FMDv that remains in carcasses after passing
through rigor mortis is likely to be inactivated at temperatures typical of carcass compost piles.
In general, temperatures of at least 55°C (131°F) must be reached for three or more days to
inactivate microbial populations (NABCC 2004). Guan et al. (2010) reported that FMD was
inactivated in specimens in compost by day 10 and the viral RNA was degraded in skin and
internal organ tissues by day 21. Compost windrow temperatures had reached 50°C and 70°C by
days 10 and 19, respectively. However, Schwarz and Bonhotal (2015) reported that FMDv
survived considerably longer at lower temperatures in laboratory conditions: as long as 21 days
at 30°C and 93 hours at 40°C. Schwarz and Bonhotal (2015) also found that FMDv was
inactivated in less than one hour at 50°C in their laboratory compared with survival over 34 to
44.5 hours outdoors in sewage sludge, even though the highest temperature was 48°C. They
concluded that survival time depends on physicochemical factors in addition to temperature. Not
all parts of a compost row necessarily reach and maintain the temperatures expected to render
FMDv inviable in some tissue reservoirs (e.g., bone marrow). Moreover, USDA cautions that the
42
-------
composting site should be kept isolated from susceptible animals until such time that laboratory
testing is unable to recover virus and/or sentinel animals have confirmed that finding.
Temperatures reached during rendering processes also are likely to inactivate the virus. The five
management options that include thermal destruction, ranked in order of decreasing temperatures
are: off-site incineration, air-curtain burning, open burning, rendering, and composting.
Containment refers to prevention or reduction (e.g., with physical barriers) of releases to the
environment, while control refers to limiting releases to acceptable levels. Among the seven
carcass management options, the three off-site options provide more containment and control
compared with the four on-site options, because commercial facilities must limit their releases to
the environment to meet state and federal standards and statutes. The off-site options, therefore,
are not ranked relative to each other, although off-site incineration would thoroughly inactivate
the virus.
All four of the on-site options include largely unregulated environmental releases. As concluded
in the assessment of carcass management options for natural disasters (USEPA 2017), the on-site
options can be ranked in order of thermal inactivation of microbes generally: air-curtain burning,
open-burning, and composting, with burial offering no high-temperature inactivation. Although
these rankings are based on an analysis that included microbes typically found in healthy cattle
in the United States, they apply to FMD also. Air-curtain burning recirculates fly ash and
gaseous pollutants to result in more complete combustion and higher burn temperatures than
open pyre burning and is expected to completely inactivate all virus particles. Open pyres, if not
well managed, might result in pockets of uncombusted materials near the edges of the pyre that
might travel in wind off-site. Similarly, compost piles require monitoring and management to
ensure that internal temperatures sufficient to inactivate FMD are reached throughout a windrow.
The identification of pathways and impacts of pollutants from seven carcass management options
are discussed for natural disasters (USEPA 2017). Before any of these seven carcass
management options can be used, it may be necessary to handle, move, or temporarily store the
carcasses (e.g., while procuring fuels, excavating a burial trench, arranging for biosecure
transportation off-site). Based on the discussion of these activities in Section 2.2.1, potential
exposures due to handling and temporary storage in a pile on the ground are assessed.
Considering biological decay and reduced pH in carcasses upon rigor mortis, carcasses in the
temporary storage pile contain more viable virus than carcasses during any subsequent phase of
management.
4.2 Base Case Exposure Assessment for FMD
The quantitative exposure assessment evaluates the potential for neighboring healthy cattle to be
exposed to viable FMDv through two release pathways: to air and to soil, with possible transport
to ground water below. This section presents results for the base case, which serves as a baseline
of comparison for additional results in Section 4.3. The results for air and soil-to-ground water
release pathways are presented in Sections 4.2.1 and 4.2.2, respectively.
The base case exposure estimates presented in this section are compared to TCID50 benchmarks
identified in Section 2, Table 2.1. The TCID50 benchmarks are used as points-of-reference only,
and exposure estimates equal to or greater than a benchmark do not necessarily indicate that
FMD infection is likely. Likewise, infection may be possible when the exposure estimates are
43
-------
below the benchmarks. Exposure estimates might be over- or under-estimated for the base case
scenario and the base case scenario might over- or under- estimate exposures at actual sites for
the following reasons:
¦ Exposures are determined in part by parameters such as carcass number and weight, soil
properties, and meteorology with moderate to high natural variation. Conditions at actual
sites might differ from those used in this assessment.
¦ The exposure assessment includes parameters with high uncertainty due to limitations of
available data or methods. Examples of these parameters include the rate FMDv particles
are released to air from the carcasses and biological decay rates.
¦ The exposure assessment uses simplifying assumptions such as details of the carcass
management options (duration of temporary carcass storage, pile size and placement, well
location) that might differ at actual sites.
The exposure assessment examines how the estimated exposures differ when certain parameters
are varied. For releases to air, base case results are presented at distance intervals from the
source. The base case results for releases to soil and ground water examine difference by soil
type and depth. In Section 4.3, the base case results are used as points-of-reference as factors
such as the scale of mortality and biological decay rates are varied. In section 4.3, exposure
estimates are compared to the base case and not to TCID50 benchmarks.
Because of the uncertainties inherent in the assessment, varying parameter values as described
above does not necessarily answer questions such as how far do healthy cattle need to be from
the source to be "safe" from infection. However, the results are useful for questions such as:
¦ How do exposures compare with different soil types?
¦ Is leaching to ground water affected more by soil type or depth?
¦ How do exposures change with 10 or 100 times as many carcasses?
In addition, the exposure assessment provides information to help site managers identify
potential exposure pathways, to evaluate whether complete pathways are likely to exist given
site-specific conditions, and how best to mitigate potential exposures (e.g., storage pile location,
liners, tarps).
4,2,1 Air
Cattle can be exposed to FMDv released to air in two ways: (1) inhalation of virus particles in
air, and (2) ingestion of virus particles deposited from the air to soil and vegetation. This section
estimates exposures separately for inhalation and ingestion.
Inhalation of FMDv Particles In Air
As described in Section 3.2.1, air transport of FMDv emitted from the temporary storage pile is
modeled to estimate concentrations in air at distance intervals of 100 m up tol km and at 1 km
intervals extending to 10 km as shown in Figure 4.1. If biosecurity procedures to contain the
outbreak are fully implemented, only distances beyond 10 km might include live cattle. Exposure
at closer distances is possible if it is infeasible to cull all cattle within the 10-km FMD response
area.
44
-------
Figure 4.1 also shows that virus concentrations in air are highest at 100 m (i.e., closest to the
source) and decrease gradually with increasing distance from the source. The figure also shows
that there is no strong directional pattern in maximum 1-hour modeled air concentrations.
Table 4.1 shows the estimated inhalation exposure concentrations and total 48-hour exposure
estimates by distance from the storage pile. All concentrations are given in units of TCIDso/m3.
The maximum 48-hour average concentrations identified are presented because this is the
duration of storage and exposure. The concentrations at each distance are the highest 48-hour
average concentrations in any direction.
Because weather conditions play a large role in air dispersion, AERMOD estimated
concentrations for each hour for a full year using hourly meteorological data.12 The maximum
concentrations represent the time periods (e.g., 48 consecutive hours) with the highest average
concentrations. Maximum air concentrations and deposition for one 48-hour period over an
entire year are similar in all 16 wind directions, which implies that any single 2-day maximum is
almost equally likely to occur in any of the 16 directions. Central-tendency (e.g., median, mean)
air concentrations and deposition would be higher in the direction corresponding to the
prevailing wind direction. The assessment included hours when weather conditions, specifically
temperature and relative humidity, would be unfavorable to the survival of the FMDv (i.e.,
temperatures above 75°F (23.9°C) and a relative humidity less than 55%).
12 For further information about the meteorological data used for this assessment, see USEPA (2017).
45
-------
Ma* 1-hr-avg Cone
Count perm1
¦
iOe-005-396-3!M
¦
3 91*-0O4- &&e-3G4
¦
¦
3 ne-3Qg -1 i^e-oca
1 14e.DCE ¦ 1 p'ffrCCZ
1 7fl*-G0S - 2 HfrOOQ
jTOf Dca
Aoie-oaj- flMMxa
¦
8«te-QO?-130MJG1
•
1 33c 001 - A 03e-GC1
0-1 km "Si
¦
i
n_ ®
\ ¦
¦1
.. ." @
¦!
~nc"
" / \
^ ¦ i >
V
; : 1
¦
1
¦
dfife
•Sjff
0 > ) 1 4 KMpmjie*)
• I I I I I I I I
Figure 4-1. AERMOD receptor locations and highest 1-hour FMDv concentrations.
46
-------
Table 4-1. Base-Case Estimates of Inhalation Exposure for Dairy Cattle
Distance from
Source
Highest 48-hour
Concentration (TCIDso/m3)
Total 48-hour
Exposure (TCID50)
Exposure Ratioa
100 m
9.1E-02
4.8E+01
2.4
200 m
2.7E-02
1.4E+01
0.72
300 m
1.3E-02
6.9E+00
0.352
400 m
8.0E-03
4.2E+00
0.212
500 m
5.4E-03
2.9E+00
0.142
600 m
3.9E-03
2.1E+00
0.10
700 m
2.9E-03
1.5E+00
0.077
800 m
2.3E-03
1.2E+00
0.061
900 m
1.8E-03
9.5E-01
0.0487
1 km
1.5E-03
7.9E-01
0.0400
2 km
4.0E-04
2.1E-01
0.0107
3 km
1.8E-04
9.5E-02
0.0047
4 km
9.7E-05
5.1E-02
0.0026
5 km
5.9E-05
3.1E-02
0.0016
6 km
3.9E-05
2.1E-02
0.0010
8 km
2.7E-05
1.4E-02
7.0E-4
9 km
1.9E-05
1.0E-02
5.0E-4
10 km
1.4E-05
7.4E-03
4.0E-4
Abbreviations and acronyms: TCID50 = 50 percent tissue-culture infectious-dose
a The Exposure ratio is the estimated 48-hour inhalation exposure divided by the TCID50 benchmark (i.e., 20
TCID50, see Table 2.1).
The 48-hour total inhalation exposure estimates in Table 4.1 are calculated with Equation 3.5 as
described in Section 3.2.6. The exposure estimates are based on the inhalation rate for dairy
cattle (11 m3/hour), which is slightly higher than the inhalation rate for beef cattle (9.6 m3/hour).
Exposure ratios in Table 4.1 are calculated by dividing the inhalation exposure by the inhalation
benchmark of 20 TCID50 (see Table 2.1). If the specific inhalation rates for dairy cattle and beef
cattle had been used in the calculation, the exposure ratios for dairy and beef cattle 100 m from
the storage pile are 2.4 and 2.1, respectively.
Ingestion of Virus Particles on Soil and Vegetation
Along with air concentrations of FMDv particles, air dispersion modeling provided rates of virus
particle deposition to the ground. Virus particles deposited to vegetation are grazed by cattle, and
soil is incidentally ingested along with vegetation. Methods for estimating concentrations of
FMDv in surface soil and on vegetation and are described in Section 3.2.2 and Section 3.2.3,
respectively, and the method for estimating ingestion exposure is described in Section 3.2.6.
Table 4.2 presents the total ingestion estimate for the base case in units of TCID50 per 48-hour
event. Ingestion exposures are presented for the location with the highest exposure concentration
47
-------
at any of the 304 locations modeled. The highest concentration is 100 m from the source, which
is the closest distance evaluated.
Although the exposure assessment is based on some parameters with considerable uncertainty
(e.g., data to estimate emission rates at the source), the exposure estimates are likely conservative
due to several assumptions used to estimate ingestion. The cattle are assumed to eat a diet
entirely of forage (i.e., not supplemented with uncontaminated grain or silage), which they obtain
by grazing entirely at the locations with the greatest deposition. In addition, the cattle are
assumed to graze immediately following the end of deposition, before further biological decay of
the virus can occur.
The results in Table 4.2 include the 48-hour deposition of FMDv, the estimated concentrations of
FMDv in soil and forage at the end of deposition, the total ingestion of viable virus over the 21-
day exposure duration, and the exposure ratio calculated with the ingestion benchmark, 1E+05
TCIDso.
Table 4-2. Forage and Soil Ingestion Exposure Results
Total 48 h
Deposition, with
Decay
(TCIDso/m2)
FMDv
Concentration in
Soil After 48 h
(TCIDso/kg)
FMDv
Concentration in
Forage After 48
h
(TCID5<,/kg-dw)
Amount of
FMDv Ingested
by Cattle Over
21 Days (TCIDS0)
Exposure Ratio
for 21-day
Exposure
2.8E+02
9.2E+00
2.8E+02
2.7E+04
2.7E-01
Abbreviations and acronyms: dw = dry weight; FMDv = foot and mouth disease virus; TCID50 = 50 percent
tissue-culture infectious-dose
4.2.2 Ground Water Ingestion
Fluids released from carcasses during the early stages of decomposition, while carcasses are in
the temporary storage pile, can seep into the ground, eventually passing through soil to ground
water. As discussed in Section 4.1, the fluid released from the temporary storage pile is expected
contain more viable virus than releases at later steps of carcass management, in part because a
pH decrease during rigor mortis is unfavorable to survival of the virus.
Neighboring live cattle could be exposed to FMDv in ground water if their drinking water comes
from an affected well. For that situation to occur, the well must be located downgradient (i.e., in
the direction of ground water flow) from the source and the aquifer would need to be relatively
shallow. In addition, a sufficiently large load of virus must reach the ground water at the well
fast enough to still contain viable virus particles. The amount of time required for horizontal
transport to the neighboring well depends on the rate of ground water flow and the distance from
the source.
As discussed in Section 3.2.4, simulations with the Virulo model indicate that FMDv released
from the temporary carcass storage pile will have a very low probability of reaching ground
water without at least 99.99% attenuation. The risk that 99.99% attenuation not being met with at
least a meter of soil is approximately 5.0E-06. Virulo estimates attenuation based on
hydrogeological processes and virus particle sorption-desorption in the soil between the source
and the water table. It does not estimate concentrations of virus particles in ground water and
does not address biological decay or attenuation processes in the ground water.
48
-------
To examine the ground water exposure pathway further, a series of calculations were performed
to estimate the FMDv concentration in a well 100 m downgradient from the storage pile. These
calculations include biological decay and dilution attenuation between the storage pile and the
well. These simple calculations overestimate exposure because they do not include the complex
vadose zone processes simulated by Virulo,
Table 4.3 presents the estimated concentrations of viable virus in well water, as well as the one-
day ingestion exposure per cow, and the exposure ratio (i.e., one-day exposure relative to the
ingestion benchmark). Note that a cow provided drinking water from the well for more than one
day will have an increased cumulative exposure and risk of infection, and in a herd of cattle
provided drinking water from the same well there is an increased risk that at least one will be
infected. FMD is highly contagious and is likely to spread through the heard if at least one cow is
infected.
Exposure to FMDv from ground water ingestion decreases with greater soil depths between the
carcass storage pile and ground water. Because clay is less permeable than silty loam or sand,
water and virus particles move more slowly through the vadose zone. With a slower rates of
movement in clay than other soils, more biological decay occurs before the virus particles reach
ground water. Table 4.3 shows that soil depth has a greater effect on exposure with clay soils
than with silty loam or sand.
49
-------
Table 4-3. Base Case Estimates of Water Ingestion Exposure
Soil Depth
Concentration of Viable FMDv in
Ground Water (TCID50/L)
One-day Ingestion Exposure per Cow
(TCIDso/day)
Exposure Ratio
(m)
Sand
Silty Loam
Clay
Sand
Silty Loam
Clay
Sand
Silty Loam
Clay
0.5
9.3E-01
7.0E-01
8.8E-02
8.8E+01
6.7E+01
8.3E+00
8.8E-04
6.7E-04
8.3E-05
1
8.7E-01
5.0E-01
7.8E-03
8.3E+01
4.8E+01
7.4E-01
8.3E-04
4.8E-04
7.4E-06
1.5
8.2E-01
3.6E-01
6.9E-04
7.8E+01
3.4E+01
6.6E-02
7.8E-04
3.4E-04
6.6E-07
2
7.7E-01
2.6E-01
6.2E-05
7.4E+01
2.4E+01
5.9E-03
7.4E-04
2.4E-04
5.9E-08
3
6.9E-01
1.3E-01
4.9E-07
6.5E+01
1.2E+01
4.7E-05
6.5E-04
1.2E-04
4.7E-10
4
6.1E-01
6.7E-02
3.9E-09
5.8E+01
6.4E+00
3.7E-07
5.8E-04
6.4E-05
3.7E-12
5
5.4E-01
3.4E-02
3.1E-11
5.1E+01
3.3E+00
2.9E-09
5.1E-04
3.3E-05
2.9E-14
6
4.8E-01
1.7E-02
2.4E-13
4.5E+01
1.7E+00
2.3E-11
4.5E-04
1.7E-05
2.3E-16
FMDv = Foot and mouth disease vims; TCID50 = 50 percent tissue-culture infectious-dose
50
-------
4.3 Uncertainty Analysis
The findings presented in Sections 4.1 and 4.2 are based on several assumptions for the base case
(e.g., the number of carcasses; environmental conditions; configuration, siting, and management
of the storage pile). These parameter values are likely to vary substantially across locations and
by season, and data by which to estimate releases of FMDv to air and leachate are limited.
Although the assessment approach generally uses conservative values for parameters that vary
substantially in the real world, parameters for which data are limited and the value selected is
highly uncertain (e.g., range and central tendency of values in the real world are unknown),
could result in over- or underestimates of exposure.
This section examines the sensitivity of the base case exposure estimates to the most uncertain
parameter values. The uncertainty analysis uses the same modeling framework developed for the
base case, with the parameters listed below varied one at a time over a range of feasible
conditions.
¦ Scale of mortality - Larger numbers of carcasses would release more FMDv to air and the
ground.
¦ Aerosol release rate - The aerosol release rate from cattle carcasses to air is based on live
animals.
1 Viral load to leachate - The amount of virus released in leachate over the first 48 hours is
unknown.
¦ Soil type and depth to soil - The type of soil and the depth of soil to ground water beneath
the storage pile or burial trench affects the potential for exposure via drinking water. These
parameters were varied for the base case and are included in the uncertainty analysis as well.
¦ Biological decay rate - Exposure concentrations would be larger if the biological decay
rate is slower.
4,3,1 Uncertainty \?vtlysls for Air Exposure Pathways
Table 4.4 shows how selected parameter values are varied for the air exposure pathways. Inputs
to AERMOD are varied to examine the sensitivity of exposure to changes in the particle
emission rate, the number of carcasses, and the biological decay rate. Except for the parameters
listed in Table 4.4, all AERMOD runs are performed with the same data and assumptions as the
base case described in Section 4.2.1.
Inhalation of FMDv Particles In Air
Virus particle emission rates are varied from the base-case estimate of 116 TCID50 per second up
to 1 million TCID50 per second. Figure 4.2 shows how this range of values affects the inhalation
exposure by distance. In the figure, as well as Figures 4.3 through 4.11, the base-case results are
shown as a solid line. All exposure estimates are indexed to the base-case exposure estimate at
100 m from the source. That is, all exposure estimates are divided by the exposure estimated for
the base case at 100 m. A red horizontal line distinguishes estimates greater than or less than the
index estimate, which has a value of 1. The vertical axis is the level of exposure relative to the
base case values, which are provided in Section 4.2.
Figure 4.2 shows that estimated exposures decrease by approximately two orders of magnitude
within the first kilometer from the source and about two additional orders of magnitude between
1 and 10 km. With the AERMOD air dispersion modeling framework, the FMDv concentrations
51
-------
in air are directly proportional to the emission rate at all distances. This is evident because the
curves for the varied particle emission rates are equally spaced by order-of-magnitude intervals.
For example, a 1,000-fold increase in emissions results in a 1,000-fold increase in exposure at all
distances. With a 1,000-fold increase in emissions, exposures are greater than the highest base
case exposure (i.e., at 100 m) to a distance of 4 km, and with one additional 10-fold increase
exposures are greater than the highest base case estimate at all distances within 10 km.
52
-------
Table 4-4. Uncertainty Analyses for Air Exposure Pathways3
Scenario
Number of
Carcasses
Pile Area (m2)
Particle Emission
Rate, Whole Pile
(TCIDso/sec)
Particle Emission
Rate per Unit
Area
(TCIDso/m2)
Biological Decay
Rate (sec1)
Particle Emission Rate
Base case
100
96.72
116
1.2
1.9E-4
Rounded base caseb
100
96.72
100
1
1.9E-4
10 xbase case
100
96.72
1,000
10
1.9E-4
100 xbase case
100
96.72
10,000
100
1.9E-4
1,000 x base case
100
96.72
100,000
1,000
1.9E-4
10,000 x base case
100
96.72
1,000,000
10,000
1.9E-4
Number of Carcasses
Base case
100
96.72
116
1.2
1.9E-4
5 x base case
500
483.6
580
1.2
1.9E-4
10 xbase case
1,000
967.2
1,160
1.2
1.9E-4
50 xbase case
5,000
4,836
5,800
1.2
1.9E-4
100 xbase case
10,000
9,672
11,600
1.2
1.9E-4
Biological Decay Rate
Base scenario
100
96.72
116
1.2
1.9E-4
10 xbase case
100
96.72
116
1.2
1.0E-3
Rounded base case
100
96.72
116
1.2
1.0E-4
1/10 xbase case
100
96.72
116
1.2
1.0E-5
1/100 xbase case
100
96.72
116
1.2
1.0E-6
TCID50= 50 percent tissue-culture infectious-dose
a Parameter values in bold text are varied. All other parameters are held constant.
b Scenario is not modeled because it is close to the base-case value.
-------
l.E+04
l.E+03
l.E+02
O
o
4-. l.E+01
ns
ns
u
01
1/1
ns
CQ
01
_>
_ns
01
0£
01
o
Q.
X
1.E+00
l.E-01
l.E-02
l.E-03
l.E-04
Particle Emission
Rate, Whole Pile
(TCID50/second)
—•—116
-•••1000
-•••10000
-•••100000
-•••1000000
0 1000 2000 3000 4000 5000 6000 7000 8000
Distance from Center of Pile (m)
9000 10000
Figure 4-2. Uncertainty analysis for particle emission rates to air, inhalation exposure
relative to base case, for dairy cattle with distance from the storage pile.
Figure 4.3 shows how inhalation exposure changes with larger numbers of carcasses. As
expected, managing greater numbers of caresses leads to greater levels of exposures. The amount
of increase is approximately proportional to the number of carcasses beyond the first few
hundred meters from the source. At locations close to the storage pile, exposure estimates are
affected by the size and configuration of the pile. Because distance is measured from the center
of the pile, the distance from the nearest edge is not necessarily the same with different
configurations. These differences affect the concentration unequally, particularly at distances
close to the sources.
Air concentrations of FMDv, and exposures, drop off steeply with distance. For example, with
100 times as many carcasses (i.e., 10,000 carcasses), exposure is no greater than the highest
baseline exposure (i.e., at 100 m) at distances beyond about 1200 m.
54
-------
1.E-KX2
1.E-K31
E
IE-OS " " " " " " " " " '
0 1000 20X 3000 4000 5000 6000 7000 aDOO 9000 10003
Di stance from Center of Pil e (m)
H l-E-KX)
a
Q
£ 1H>1
2
l.E-02
1E-04
Number of
Carcasses
-¦-100
*¦•¦'500
-•'1000
¦••5000
¦¦•¦ 10000
Figure 4-3. Uncertainty analysis for the number of carcasses, inhalation exposure for dairy
cattle relative to the base case, with distance from the storage pile.
Inhalation of viable FMDv by neighboring live cattle is insensitive to a varying the virus decay
rate over five orders of magnitude. As shown in Figure 4.4, inhalation exposures are similar to
the base case within approximately the first kilometer for the selected biological decay rates.
Because more time for decay elapses before virus particles reach farther distances, differences in
the decay rate have an increasing effect on exposure with distance. However, exposure estimates
remain fairly insensitive to the decay rate at 10 km where a 1,000-fold difference in the decay
rate results in a less than a 50-fold change in exposure.
55
-------
Biological
Decay Rate
(per second)
•••••1.00E-03
1.00 E-04
•••••1.90E-04
•••"1.00E-05
1.00E-06
1000 2000 3000 4000 5000 6000 7000 8000 9000 10000
Distance from Center of Pile (m)
Figure 4-4. Uncertainty analysis for the biological decay rate, inhalation exposure for dairy
cattle relative to the base case, by distance from the storage pile.
Ingestion of Virus Particles on Soil and Vegetation
Figures 4.5 through 4.7 show how varying parameter values affect ingestion by neighboring
cattle of FMDv particles that settled from the air to forage vegetation and soil. These figures
include results only for the location with the highest total deposition, which is at 100 m from the
storage pile. As described in Sections 3.2.2 and 3.2.3, ingestion exposure by grazing cattle is
estimated as the total TCID50 ingested over the first 21 days after the 48-hour release. The base
case exposure estimate, 2.7E+04 TCID50, is included in Table 4.2. Relative to this baseline,
increasing the virus particle emission rate used in the assessment results in proportional increases
in exposure. This finding is seen in Figure 4.5 and is consistent with the related results for
inhalation exposure shown in Figure 4.2.
56
-------
l.E+04
l.E+03 ¦
_TO
-------
m
1000 2000 3000 4000 5000 6000 7000
Number of Carcasses
8000
9000
10000
Figure 4-6. Uncertainty analysis for number of carcasses, ingestion exposure for dairy
cattle relative to the base case at 100 m from the storage pile.
58
-------
1.1
a;
i/i
Q)
i/l
(13
CO
O
+¦>
Q)
>
~ 1
_ro
O)
cc
QJ
1-
3
i/i
o
a.
1.0E-05 1.0E-04 1.0E-03
Biological Decay Rate (per second)
Figure 4-7. Uncertainty analysis for the biological decay rate, ingestion exposure for dairy
cattle relative to the base case at 100 m from the storage pile.
4.3.2 Uncertainty Analysis for the Ground Water Exposure Pathway
Table 4.5 shows the uncertainty analysis parameter values for the ground water ingestion
pathway. Like the base-case results (Section 4.2.2), the uncertainty analysis includes results for
three soil types (sand, silty loam, and clay) and depth to ground water from 0.5 to 6 m.
It is important to note that the results presented in this section evaluate the effects of biological
decay and dilution attenuation between the storage pile and a water well 100 m downgradient.
These results over-estimate potential exposures because they do not include hydrogeological
processes that tend to retard and attenuate virus particles in the vadose zone. The Virulo analysis
presented in Section 3.2.4 indicates high likelihood that these processes will achieve at least
99.99% attenuation with at least 1 m of sand, silty-loam, or clay.
As context for results presented in this section, many states recommend or mandate minimum
depths of unsaturated soil beneath a carcass burial pit to protect ground water quality. These
distances are as little as 1 ft (-0.3 m), but more typically are between 3 ft (~1 m) and 5 ft (-1.5
m) (NABCC 2004). Distance to ground water is not specified for temporary storage piles, but its
consideration is relevant to ground water protection.
0.9
1.0E-06
59
-------
Particle Release Rate
Figures 4.8 through 4.10 show the sensitivity of water ingestion exposure estimates to virus
particle release rates. The three figures correspond to results for sand, silty loam, and clay,
respectively. All exposure estimates are indexed to the base case exposure estimate with 1 m of
silty loam. The red horizontal line in each figure is the index value of 1. The base case exposure
estimates are presented in Section 4.2.2.
The release rates (in TCI Dso/carcass-day) vary over four orders of magnitude with the base-case
(dark blue solid line for 9.44+08 TCIDso/cow) value in the center of the range.
60
-------
Table 4-5. Uncertainty Analyses for the Ground Water Exposure Pathway for Temporary Storage Pile3
Scenario
Number of
Carcasses
Pile Area (m2)
Leachate per Carcass
per Day (L/day)
Particle Release
Rate, per Carcass
(count/day)
Starting Leachate
Concentration
(count/L)
Biological Decay
Rate (s1)
Particles Released per Carcass per Day
Base case
100
96.72
10.7
9.4E+8
8.8E+07
1.4E-6
1/100 xbase case
100
96.72
10.7
1.0E+7
9.4E+05
1.4E-6
1/10 xbase case
100
96.72
10.7
1.0E+8
9.4E+06
1.4E-6
Rounded base case
100
96.72
10.7
1.0E+9
9.4E+07
1.4E-6
10 xbase case
100
96.72
10.7
1.0E+10
9.4E+08
1.4E-6
100 xbase case
100
96.72
10.7
1.0E+11
9.4E+09
1.4E-6
Number of Carcasses
Base case
100
96.72
10.7
9.4E+8
8.8E+7
1.4E-6
1/10 xbase case
10
9.7
10.7
9.4E+8
8.8E+7
1.4E-6
5 x base case
500
483.6
10.7
9.4E+8
8.8E+7
1.4E-6
10 xbase case
1,000
967.2
10.7
9.4E+8
8.8E+7
1.4E-6
50 xbase case
5,000
4,836
10.7
9.4E+8
8.8E+7
1.4E-6
100 xbase case
10,000
9,672
10.7
9.4E+8
8.8E+7
1.4E-6
Biological Decay Rate
Base case
100
96.72
10.7
9.4E+8
8.8E+07
1.4E-6
10 xbase case
100
96.72
10.7
9.4E+8
8.8E+07
1.0E-05
Rounded base case
100
96.72
10.7
9.4E+8
8.8E+07
1.0E-06
1/10 xbase case
100
96.72
10.7
9.4E+8
8.8E+07
1.0E-07
1/100 xbase case
100
96.72
10.7
9.4E+8
8.8E+07
1.0E-08
a Parameter values in bold text are varied. All other parameters are held constant.
Notes:
• Assumes 75 L leaches from a 454 kg animal in the first week (i.e.. Young et al. (2001) stated 33% released in first 2 months with 16.5% in first week).
Assuming an exponential decrease in leachate release (e.g., half as much released one day to the next), one would expect 38 L on day 1 and 19 L on day for
a total of 57 L (or 28.5 L/day).
• The maximum amount leached assumes the full 16.5% expected over the first week leaches during the first two days alone, or 75 L (divided by 2 = 38
L/day).
61
-------
Sand Depth (m)
Virus particles released per
carcass per day
(TCID50/cow-day)
1.00E+07
•••¦ 1.00E+08
-•-9.44E+08
••• 1.00E+09
1.00E+10
l.OOE+11
Figure 4-8. Uncertainty analysis for the particle release rate, water ingestion exposure for
dairy cattle by depth of sand, relative to base case exposure with 1 m silty loam.
Comparing the slopes of the lines in the three figures shows that sand allows higher permeation
and transport of virus particles than the finer soil types, silty loam (Figure 4.9) and clay (Figure
4.10). For the range of particle release rates included in the uncertainty analysis, increasing the
depth of sand between the storage pile and the water table provides little additional protection
from exposure. Estimated exposures decline rapidly with additional depth of clay soil. For all
soil depths and types the exposure estimates are proportional to the particle release rate
62
-------
l.E+04
Virus particles released
per carcass per day
(TCID50/cow-day)
1.00E+07
1.00E+08
-^-9.44E+08
• ••• 1.00E+09
1.00E+10
l.OOE+11
Silty Loam Depth (m)
Figure 4-9. Uncertainty Analysis for the Particle Release Rate, Water Ingestion Exposure
for Dairy Cattle by Depth of Silty Loam, Relative to Base Case Exposure with 1 m Depth.
63
-------
Virus particles released
per carcass per day
(TCID50/cow-day)
1.00 E+07
•••¦¦ 1.00E+08
-•-9.44E+08
1.00E+09
••• 1.00E+10
1.00E+11
Figure 4-10. Uncertainty analysis for the particle release rate, water ingestion exposure for
dairy cattle by depth of clay, relative to base case exposure with 1 m silty loam.
Number of Carcasses
With more carcasses, the amount of leachate released from the storage pile increases. The
amount of leachate seeping into the ground per unit area (e.g., per m2) will remain the same as
long as long as the pile size grows horizontally proportional to the number of carcasses. With no
change in the amount of leachate per area or in the concentration of infective FMDv in the
leachate, physical, chemical, and biological processes in the soil will result in the same
concentrations of infective virus in leachate when it reaches ground water. When developing
DAFs, as described in Section 3.2.4, USEPA (1996) performed sensitivity analyses for
parameters included it its Monte Carlo modeling approach. USEPA identified that the area of the
leachate source (e.g., landfills) as having a large effect on dilution attenuation. Based on this
finding, USEPA provided DAFs for a range of source areas, and that information is used to
identify the DAFs in Table 4.6. Larger DAF values result in greater dilution between the source
and the downgradient water well.
l.E+04
g l.E+02
1.E+00
l.E-02
l.E-04
l.E-06
l.E-08
Clay Depth (m)
l.E-10
64
-------
Table 4-6. Dilution Attenuation Factors by Area of Storage Pile and Number of Carcasses
Number of Carcasses
Area of Storage Pile (m2)
DAF
10
9.7
13929
100
96.7
1675
500
483.6
381
1,000
967.2
201
5,000
4836
46
10,000
9672
24
DAF = dilution attenuation factors
Figure 4.11 shows the relationships between the number of carcasses, soil type, and the
estimated water ingestion exposure. Exposure estimates are indexed to the base case (i.e., 100
carcasses) value with 1 m of silty loam. For all soil types, greater numbers of carcasses cause
greater exposures. However, the rate of increase in exposure declines when the number of
carcasses is greater than 1,000. The relationship between number of carcasses and estimated
exposure results from the DAFs, which are based on USEPA's (1996) Monte Carlo analysis
using databases of landfill site data.
Biological Decay
The biological decay rate is a measure of the persistence of infective FMDv in environmental
media.13 The FMDv is more persistent in the leaching to ground water pathway than in the air
emission pathways because the viral particles in water are protected from drying. The biological
decay rates used in this assessment for FMDv in air and water are 1.9E-04 per second and 1.4E-
06 per second, respectively. The bases of these values are discussed in Sections 3.1.1 and 3.1.2.
Figures 4.12 through 4.14 present the uncertainty analysis for the biological decay rate in the
drinking water exposure pathway. As seen in the other uncertainty analyses, the potential for
exposure through drinking water is greatest when the storage pile is placed over sand and least
when placed over clay. With the slowest decay rates (i.e., 1E+04 and 1E+05), the exposure
estimates are similar with the three soil types and they are not sensitive to soil depth, presumably
because there is a low amount of decay over the estimated time of travel.
13 Persistence refers to the continued presence of a particular virus type in the environment over a period time (Embrey et al.
2004).
65
-------
l.E+03
l.E+02
l.E+01
l.E+OO
l.E-01
l.E-02
l.E-03
Number of Carcasses
l.E-04
Soil Type
-~-Sand
Silty Loam
-i Clay
2000
4000
6000
8000
10000
Figure 4-11. Uncertainty analysis for the number of carcasses, water ingestion exposure for
dairy cattle by soil depth, relative to exposure with 100 carcasses and silty loam.
Exposures estimated with the two fastest decay rates (i.e., 1E+07 and 1E+08) are not included in
Figures 4.12 through 4.14 because they are negligible (e.g.,
-------
• •
¦
«•
• • .
••
Sand Depth (m)
Biological Decay
Rate(per Sec)
¦¦¦*¦¦¦ 1.0E-04
1.0E-05
-^1.4E-06
1.0E-06
1.0E-07
1.0E-08
Figure 4-12. Uncertainty analysis for the biological decay rate, water ingestion exposure for
dairy cattle by depth of sand, relative to exposure with 1 m silty loam.
67
-------
l.E+06
• »-
¦ • •
Biological Decay
Rate (per Sec)
1.0E-04
•••• 1.0E-05
-•-1.4E-06
• •• 1.0E-06
• 1.0E-07
1.0E-08
Silty Loam Depth (m)
Figure 4-13. Uncertainty analysis for the biological decay rate, water ingestion exposure for
dairy cattle by depth of silty loam, relative to exposure with 1 m silty loam.
68
-------
l.E+06
> l.E+04
cH l.E+02
¥ 1.E+00
™ l.E-02
DQ
¦- l.E-04
01 l.E-06
O
D.
* l.E-08
Clay Depth (m)
Biological Decay
Rate (per Sec)
••••- 1.0E-04
••• 1.0E-05
1.4E-06
• •• 1.0E-06
••• 1.0E-07
¦¦¦•¦ 1.0E-08
l.E-10
Figure 4-14. Uncertainty analysis for the biological decay rate, water ingestion exposure for
dairy cattle by depth of clay relative to exposure with 1 m silty loam.
4.4 Uncertainty Summary and Research Needs
This section discusses how the exposure assessment might over- or underestimate exposures in
the event of an actual FMD outbreak. Tables 4.7 through 4.9 summarize three types of
"uncertainties" in the exposure assessment:
¦ Parameters with Moderate to High Natural Variation (Table 4.7)
¦ Uncertain Parameter Values or Models (Table 4.8)
¦ Simplifying Assumptions (Table 4.9)
This assessment necessarily involves numerous selections of values for a broad array of
biological and environmental parameters, some of which are well characterized but vary
substantially (e.g., by location within the United States), and some of which are unknown and
require estimates from limited data (e.g., rates of FMDv release to air). The conceptual models
for the carcass management options revealed that many direct and indirect multimedia exposure
pathways could exist. To provide some quantitative basis for ranking the management options,
many simplifying assumptions about the natural disaster, the types of and numbers of livestock
killed, site and environmental conditions, and carcass management activities were required.
Table 4.7 describes parameters for which substantial variation exists across the United States,
and a value was selected either to be nationally representative, to be health protective (i.e.,
overestimate exposure), or for another reason. The magnitude (low, medium, high) and direction
(under- or overestimate) of bias in the exposure estimates are listed.
69
-------
Table 4.8 describes parameters for which limited data were available to calculate a central
tendency value or to estimate likely variation across conditions possible in the country.
Uncertainty is characterized as low, medium, or high. By definition, the direction of bias is
unknown.
Finally, Table 4.9 includes "simplifying assumptions" that were required to define the scope of
the assessment and limit it to a reasonable level of effort. As for Table 4.7, the magnitude (low,
medium, or high) and direction (under- or overestimate) of bias introduced by the assumption is
summarized.
Based on the uncertainties in Tables 4.7 through 4.9, as well as information gathering for this
assessment, Table 4.10 identifies research needs for the livestock carcass management options
and associated activities.
Table 4-7. Moderate to High Natural Variation in Parameter Values-
Selected Values
-Potential Bias from
Key Topic Selected Parameter Value
Scale of
Mortality
Base-case culling of 100
cattle at one farm with a
total weight of 50 short tons
to match previous analyses
(APHIS 2015, USEPA^
2017).
Uncertainty analysis
includes up to 10,000 culled
cattle.
Possible
Under-
estimate
Rationale
The base-case number of cattle
carcasses assumed for this
assessment could be considered
""small" because culling hundreds to
tens of thousands of cattle might be
required for a large FMD outbreak.
Larger numbers of carcasses could
involve larger temporary storage
pile(s), increasing the chances of
infecting one or more neighboring
susceptible animals via inhalation.
If ground surface area covered by
the collection of carcasses into
temporary storage pile(s) is
proportional to the number of
carcasses, leaching to ground water
could occur over a larger area, but
that would not increase FMD
concentrations in ground water.
Large scale losses of several
thousand cattle could exceed the
capacity of some management
options (e.g., air-curtain burning).
70
-------
Key Topic Selected Parameter Value
Bias
Rationale
Ground
water
¦ FMDv leached from the
temporary storage pile and
burial trench can reach
ground water. Based on
state regulations, ground
water is assumed to be 1 m
below bottom of buried
materials or 1 m below the
temporary storage pile.
¦ Layout Assumption:
Neighboring cattle are
provided water from a
relatively shallow aquifer
that flows in the direction of
the neighboring well from
the infected premises.
High Over-
estimate
¦ Although providing livestock with
drinking water with shallow ground
water is possible, most wells are
dug to tap into deeper aquifers that
provide adequate water during the
drier seasons as well as rainy
seasons.
¦ Well contamination would require
that the well is located down
gradient (in the direction of ground
water flow) from the source and
that the rate of ground water flow is
fast enough for viable virus
particles to remain.
Meteoro-
logical
Conditions
¦ The assessment uses 1 year
of meteorological data from
a weather station in Iowa,
chosen to represent a
moderate climate in the U.S.
agricultural heartland. The
data are used to model fate
and transport of releases to
air.
Moderate
Over- or
Underestimate
¦ The meteorological data used for
this assessment could over- or
underestimate relevant conditions in
other areas of the country (e.g.,
having stronger or weaker winds,
winds predominantly in one
direction compared with other
patterns).
71
-------
Key Topic Selected Parameter Value
Bias
Rationale
Soil Type
and
Properties
¦ The assessment considers
three soil types: two
extremes (i.e., clay and
sand) and one "middle of
the road," silty loam.
¦ Clay-like soils comprised of
fine particles can hold more
water, but retard downward
flow and adsorb a higher
fraction of virus particles.
¦ Sandy soils allow rapid
leaching of water and virus
particles
¦ The assessment does not
consider accelerated
transport through
macropores.
Moderate to
High Over- or
Underestimate
¦ Sites with different soil types and
conditions could have higher or
lower rates of vertical water
movement and capacity to adsorb
viruses. Although the three soil
types were chosen to represent a
range of conditions, other
conditions are possible and
transport though soil of a single
type can vary due to soil density,
homogeneity, and geohydrological
factors.
¦ The exposure estimates for drinking
water overestimate exposure
because they do not include
adsorption-desorption processes
that retard the movement of the
virus particles in soil. These
processes are included in the Virulo
modeling presented in Section
3.2.4.
¦ The presence of macropores would
cause greater transport of virus to
ground water than estimated. Where
the ground-water system in question
is connected to potential virus
sources by karst, fractured rock,
gravel, or a soil exhibiting
preferential flow, there would be a
high risk for viral transport to
ground water.
Receptors
¦ Exposures are assessed for
uninfected cattle on a
neighboring farm. Inhalation
and forage ingestion
exposure are assessed for
cattle at distances from 100
m to 10 km. No distance is
specified for the ground
water well.
Moderate to
High
Overestimate
¦ While uninfected cattle may be
present within 10 km of the
outbreak location, it is unlikely that
uninfected cattle would be allowed
at distances as close as 100 m.
72
-------
Key Topic
Selected Parameter Value
Bias
Rationale
Exposure
Factors
¦ Exposure factors (e.g.,
forage and incidental
ingestion rates, drinking
water rates) are from
USEPA's HHRAP and other
publications. The values
used in the assessment are
central tendency estimates.
Neutral
¦ Central tendency values are used so
that exposure is not over or
underestimated by this aspect of the
approach.
FMD = foot and mouth disease; HHRAP = Human Health Risk Assessment Protocol
73
-------
Table 4-8. Uncertainty in Parameter Value(s) Selected
Parameter
Description
Uncertainty
Rationale for Uncertainty Category
Releases
Estimates
¦ Each exposure pathway in the assessment begins
with a release of FMDv to an environmental
medium. These include emissions to air from the
temporary storage pile and liquid releases from the
storage pile. Data to characterize amount and rate
of viruses released from the carcasses are very
limited.
High
¦ This is one of the most significant sources of
uncertainty in the exposure assessment. Although
release estimates were based on the best available
information, releases might be over or
underestimated. In addition, actual releases can vary
significantly due to many factors (e.g., unit design,
environmental conditions). The effect of this
uncertainty is evaluated in the uncertainty analysis.
Animal
Vectors
¦ FMDv can be transported by insects, birds, or
mammals that come in contact with carcasses
before or during management. The exposure
assessment discusses but does not quantitatively
evaluate animal vectors.
Moderate
¦ The exclusion of animal vectors from the assessment
causes potential exposures to be underestimated. This
uncertainty impacts the composting option more that
burial or the combustion-based options.
Biological
Decay Rate
¦ FMDv undergo natural decay that decreases the
amount of viable virus overtime. The assessment
uses estimates of virus decay rates in air, soil, and
leachate. Data to develop these estimates for the
assessment are very limited. Moreover, the rate of
decay is affected by a number of highly variable
environmental conditions (e.g., ambient
temperature, relative humidity, pH, ultraviolet
exposure).
High
¦ The assumed decay rates are among the largest
sources of uncertainty in the assessment. The base-
case estimates are based on the best available
information and the effect of this uncertainty is
examined in the uncertainty analysis.
74
-------
Parameter
Description
Uncertainty
Rationale for Uncertainty Category
Models
¦ The assessment uses screening-level models and
calculations to estimate the fate and transport
FMDv through air, water, soil, and vegetation.
High
¦ The uncertainties associated with fate and transport
modeling data and methods can individually
contribute to under-or over-estimation of exposures.
In general, the assessment uses more conservative
assumptions and approaches, which would most
likely result in over-estimates of possible exposures.
¦ Because the approach uses pre-existing models and
methods that were developed for different purposes,
they are likely to differ in their level of sophistication
and uncertainty. This could cause the level of
uncertainty to differ among media pathways.
FMD
Properties
and Other
Inputs
¦ Fate and transport modeling uses various properties
of FMDv and environmental media (e.g., soil bulk
density). Properties for environmental media are
from HHRAP (USEPA 2005) unless otherwise
noted.
Moderate to
High
¦ Uncertainty associated with modeling inputs may
contribute to over- or underestimation of exposure.
This uncertainty is lowest for experimentally derived
chemical properties and greater for more variable
inputs.
FMDv = foot and mouth disease vims; HHRAP = Human Health Risk Assessment Protocol
Complete references are found at the end of this report.
75
-------
Table 4-9. Simplifying Assumptions—Effects on Exposure Estimates
Key Topic
Simplifying Assumption
Effect
Rationale for Effect
Temporary
Carcass Storage
¦ The assessment assumes that carcasses are
placed in temporary pile for 48 hours during
preparation for further management.
Moderate
Overestimate
¦ Temporary storage is not a necessary feature of
carcass management (e.g., if cattle are euthanized
immediately prior to further management). The
storage pile is included so that the assessment
does not overlook releases that could reasonably
be expected from a relatively uncontrolled source
early after death.
Design of On-
site
Management
Units
¦ Basic assumptions about the design of on-site
management options (e.g., burial trench
dimensions, storage pile assumptions) are based
USD A guidance and other relevant sources. For
larger mortalities, the unit design and spatial
pattern could be different.
¦ The assessment assumes that the temporary
storage pile is placed on bare earth.
Moderate Over-
or
Underestimates
¦ Assumptions about many aspects of carcass
management units could lead to over- or
underestimation of exposure.
¦ Exposures from the storage pile are
overestimated if liquids released from the
carcasses are collected and appropriately
managed.
Carcass
Handling Before
Management
¦ Workers who handle livestock carcasses are
assumed to use recommended PPE.
Moderate
Underestimate
¦ Exposure to workers is underestimated if no PPE
is used.
Exposure
Pathways
¦ A goal of this assessment is to assess exposure
for reasonably anticipated exposure pathways
from carcass management. Therefore, the
assessment was intentionally designed to
include feasible complete exposure pathways
that might not exist at some sites.
Moderate
Overestimate
¦ The assessment is likely to overestimate exposure
because the layout assumes a worst-case
exposure for each possible pathway, which is
unlikely at most locations.
76
-------
Key Topic
Simplifying Assumption
Effect
Rationale for Effect
Virus
Inactivation
¦ The assessment assumes that FMDv is not
viable in most compartments of the carcass due
to the low pH conditions that coincide with
rigor mortis. However, some FMDv present in
the bone, lymph nodes, liver, and kidneys could
remain viable even after rigor mortis. Because
remaining viable FMDv is further subjected to
natural biological decay processes, releases
after the 48-hour temporary storage pile are
assumed to contain low concentrations of viable
virus.
Low
Underestimate
¦ Viable virus might persist after carcasses after
they are placed in the burial trench or compost
windrow. However, releases to soil from the
burial trench would be the same or less than
releases from the storage pile during the period of
greatest liquid releases. The windrow provides
greater containment of liquid, and both of these
management options contain air releases more
than the storage pile.
Carcass
Transportation
¦ Based on a semi-quantitative assessment during
the natural disaster scenario assessment,
exposures associated with carcass transportation
are assumed to be insignificant.
Low
Underestimate
¦ Carcass transportation would follow biosecurity
measures under the FMD response plan.
Livestock
Grazing
¦ Uninfected livestock are assumed to graze at the
location of greatest estimated virus deposition
from the air.
High
Overestimate
¦ In the event of an actual FMD outbreak, it is
unlikely that uninfected livestock would be
pastured in close proximity to the outbreak
location.
77
-------
Table 4-10. Research Needs for Livestock Carcass Management Options and Activities
Option or
Activity
Research Needs
On-site
Combustion
¦ Monitoring or analysis to verify complete destruction of FMDv, and other viral agents, in air emissions and ash.
On-site Burial
¦ Systematic study to determine survival of FAD agents such as spore-forming microbes and viruses during carcass
decomposition.
¦ Identification of microbes in leachate from burial of FAD agent-infected carcasses.
¦ Research on distribution of FMD viral load across organs and tissues in infected livestock.
¦ Additional field studies of subsurface movement and survival of FAD agents in various soil types and seasons.
¦ Develop model of time-varying addition of viral particles to surface soil that predicts concentration of viable virus units
likely to reach ground water for specified precipitation events, different soil saturation conditions, and different
temperatures.
On-site
Composting
¦ Studies of pathogenic microbes in finished compost.
¦ Field analysis of the fate and transport of pathogenic microbes during composting and following application of compost to
surface soil.
¦ Overnight surveillance (e.g., motion-activated wildlife cameras) to compile data on nocturnal scavenger activity around
compost piles.
Off-site
Options
¦ Survey facilities to find any that might have accepted FMD-infected cattle during past outbreak. Evaluate information
recorded during the incident.
¦ Design monitoring studies for off-site facilities to implement in the event of an FMD outbreak in the United States.
Carcass
Handling
¦ For a quantitative exposure assessment, data on exposure factors (e.g., frequency and duration of hand contact, area of skin
exposed) for carcass handlers, and the effectiveness PPE or likely compliance with PPE use.
¦ Concentrations of FAD agents on contact surfaces.
¦ Explore the role of common items (e.g., vehicles, worker clothing, etc.) as fomites and best practices for decontamination or
"disposal" of contaminated materials after the response actions.
Temporary
Carcass
Storage
¦ Monitoring air downwind of uncovered storage piles for viable microbes.
¦ Analysis of microbial load on fur or feathers of livestock soon after culling.
¦ Research to better characterize the biological decay of FAD agents in livestock carcasses.
78
-------
¦ Collect leachate from freshly killed carcasses daily over one or two weeks at different temperatures.
¦ Assay leachate for viable microbes over time.
Carcass
Transportation
¦ Further research to measure or estimate microbial releases associated with transporting carcasses to off-site facilities.
FAD = foreign animal disease; FMD = foot and mouth disease; FMDv = foot and mouth disease vims; PPE = personal protective equipment
79
-------
4.5 Summary of Findings
This assessment is meant to support selection of environmentally protective livestock carcass
management methods in the event of an FAD outbreak. This exposure assessment addresses only
one FAD agent, FMDv. For this agent, FMD exposure is most likely to result from activities in
the first hours and days after death because:
¦ FMDv does not replicate outside a living host, and the amount of viable FMDv will decline
after death due to natural decay processes.
¦ Within most compartments of a carcass, the viability of FMDv significantly decreases along
with a decrease in pH that coincides with rigor mortis.
It is important to remember that the findings for FMDv exposures in the assessment are not
necessarily applicable to other FAD agents, particularly non-viral microbes such as bacteria and
protozoa. In addition, this assessment concerns exposures to individual animals from the air they
breathe, water they drink, and the forage they graze. It does not address the spread of infection
among animals within a herd.
FMD exposure are estimated using generally conservative scenarios and assumptions that would
overestimate exposures at most actual carcass management locations. Section 4.4 identifies and
discusses uncertainties and assumptions in the assessment. This information can be used to
evaluate exposure scenarios and the potential for exposure to occur at actual sites.
Key findings of the assessment are presented below, organized by management option and
related activities.
Temporary Carcass Storage
¦ If carcasses cannot be managed immediately after death, the temporary carcass storage pile
appears to be the most likely source to possibly expose nearby livestock.
¦ Inhalation is the most likely cause of exposure because airborne virus particles can be travel
more quickly and with fewer barriers compared to the ground water pathway.
¦ If the storage pile is placed on bare earth, exposure to cattle through drinking water is
possible. However, a number of conditions (e.g., a well is in the direction of ground water
flow) must be met for a complete exposure pathway, which is unlikely at many sites.
¦ If the soil depth to ground water is at least 1 m, there is a high probability that at least
99.99% of FMDv particles attenuation before leachate reaches the water table.
¦ The potential for exposure is affected by several site-specific factors and uncertainties
discussed in Section 4.4. The uncertainty analyses in Section 4.3 examines how the
exposure estimates change with varied virus release rates, biological decay rates, soil types
and depths, and numbers of carcasses.
¦ Exposures through air and ground water can be mitigated with tarps or other barriers
beneath and over the storage pile.
On-site Open Burning and Air-curtain Burning
¦ On-site combustion options effectively destroy FMDv when there is an even burn (i.e., all
soft tissues are burned).
80
-------
¦ Because FMDv is inactivated by combustion, exposures were not estimated for these
management options.
Composting
1 Composting provides thermal treatment and containment. Thermal inactivation and natural
decay essentially eliminate potential exposure from the finished compost. The compost can
be kept in the windrow until infective FMDv is not detected.
¦ Although composting is an effective option for FMDv, this is not necessarily the case for
other FAD agents. Prions and environmentally resistant life stages (e.g., spores of spore-
forming bacteria such as anthrax) might not be completely inactivated by composting.
On-site Burial
1 Burial provides containment only, and FMDv has the potential to leach through soil to
ground water similarly to the temporary storage pile. However, soil at least 1 m deep will
provide a high level of attenuation and several conditions must be met for there to be a
complete exposure pathway.
¦ Unlike the temporary storage pile, there would be no exposure from inhalation or ingestion
of forage and soil.
¦ The potential for exposure through the ground water pathway is reduced if carcasses are
placed in the trench after rigor mortis. However, overall exposure from carcass management
could be greater depending on how the carcasses are managed before rigor mortis.
Carcass Handling and Transportation
¦ Adherence to biosecurity measures (e.g., vehicle decontamination) and the use of PPE
recommended by USD A/APHIS (USDA 2014a), as assumed for this assessment, mitigates
human exposure to FMD and the potential for workers to spread FMDv.
¦ Based on an evaluation in USEPA (2017), exposures during carcass transportation are
assumed to be negligible at locations along the transportation route. Using federal
transportation statistics and a scenario in which eight truckloads of carcasses are transported
100 km, USEPA (2017) estimated risk of an accident with cargo spillage to be 7.1E-05.
Off-site Carcass Management Options
¦ Among the three off-site options, commercial incinerators would totally inactivate FMDv,
and rendering facilities similarly apply sufficient heat for enough time to inactivate the
virus, viable FMDv in carcasses placed in landfills could contribute to leachate, however
livestock are not likely to come in contact with the leachate collected and managed under
regulatory requirements. For all the off-site options, all releases to the environment (e.g.,
incinerator emissions to air, rendering facility discharge to surface water) are restricted by,
and are assumed to comply with, normally applicable federal regulations. For these reasons
the off-site options are not included in the quantitative assessment.
Table 4.11 ranks on-site management options based on the exposure assessment and the degree
to which treatment or containment control releases of FMDv to potential exposure pathways. The
81
-------
temporary carcass storage pile is included in the rankings even though it is not a management
option and can be used before any of the on-site or off-site management options. It is included
because releases from to air or the ground could cause higher potential exposures than any of the
management options.
Off-site management options are not included in Table 4.11 because they are not included in the
quantitative assessment, as discussed in Section 2.2. As part of the exposure assessment for the
natural disaster scenario (USEPA 2017), the off-site options were qualitatively ranked relative to
each other for control of microbes based on their level of thermal destruction. Those rankings are
shown in Table 4.12.
Table 4-11. Ranking of On-site Livestock Carcass Management Options for an FMD
Outbreak
Rank
Management
Type
Principal Rationale
1
Open Burning
and Air-curtain
Burning
Thermal descruction of all FMDv.
2
Composting
Bulking material contains almost all FMDv from releases to air and soil.
Thermal inactivation and biological decay eliminate FMDv before
composting is complete.
3
Burial
Cover soil contains releases to air. If a number of conditions are met,
leaching has the potential to infect cattle that drink water pumped from a
ground water well.
4
Temporary
Storage
Cattle can be infected by inhaling or ingesting FMDv emitted to air from a
nearby storage pile. If a number of conditions are met, leaching has the
potential to infect cattle that drink water pumped from a ground water
well.
FMDv = foot and mouth disease virus
Table 4-12. Ranking of Off-site Livestock Carcass Management Options for Microbes
Rank8
Management
Option
Principal Rationale
H
Off-site
Incinerator
Thermal destruction of all microbes, ash is landfilled
M
Off-site
Rendering
Thermal inactivation of all microbes except prions, workers protected
from prion exposure with the use of PPE
L
Off-site Landfill
Containment, including liner, leachate collection, cover material, but no
thermal destruction; when capacity is reached, landfill is closed and new
ones built
Abbreviations: H = Highest rank; L = Lowest rank; M = Middle rank; PPE = personal protective equipment.
a Relative and absolute risks from microbial pathogens depends on initial concentrations in healthy cattle, which is
unknown.
In the event of an actual FAD outbreak, site managers can use this report with site-specific
information and properties of the FAD agent to identify possible exposure pathways, determine
82
-------
whether complete exposure pathways actually exist, and how exposures can be avoided. The
following information provided in this report can aid such evaluations.
¦ Conceptual models - Conceptual models for the temporary storage pile and on-site burial
are included in Section 2.4. These identify the possible pathways by which cattle might be
exposed to the FAD. Conceptual models for all of the carcass management options and
associated activities (e.g., carcass handling, transportation) are available in the exposure
assessment of livestock carcass management options following natural disasters (USEPA
2017).
¦ Environmental fate concepts - The descriptions of FMDv releases and environmental fate
estimation in Section 3 identify factors (e.g., temperature and humidity aquifer, water well
characteristics) that determine whether a complete exposure pathway actually exists at a
particular site.
¦ Management option assumptions - Sections 3.2 and 3.3 and USEPA (2017) provide
information (e.g., management option specifications) compiled from the literature that may
be useful for site-specific assessments.
¦ Biological decay estimation - The report provides equations to calculate biological decay
and describes how decay relates to the management options.
1 Variability relationships - Section 4.4, as well as topics discussed throughout the report,
describe how exposures might differ at sites where scenarios and assumptions differ from
those assumed for this assessment.
¦ Mitigation - By describing the environmental releases and exposure pathways for the
management options, the report can be used to identify effective mitigation measures to
prevent or reduce radiation exposure.
83
-------
S, Set'.- i ^ < >»v i
This report used scientific information extracted from sources of secondary data including
journal articles, publications in the open literature, and government reports both published and
non-published, including distribution limited reports. Data and information were gathered from
published reports to identify the significant pathways by which pathogens might reach
individuals and estimate how many microorganisms an individual is likely to be exposed to
through each pathway. A targeted literature review was performed to identify the most highly
relevant data to inform an exposure assessment. Scientific and technical information from
various sources were evaluated using the assessment factors below:
Focus: The extent to which the work not only addresses the area of inquiry under
consideration, but also contributes to its understanding; it is germane to the issue at hand.
Verity: The extent to which data are consistent with accepted knowledge in the field, or if
not, the new or varying data are explained within the work. The degree to which data fit
within the context of the literature and are intellectually honest and authentic.
Integrity: The degree to which data are structurally sound and present a cohesive story.
The design or research rationale is logical and appropriate.
Rigor: The extent to which work is important, meaningful, and non-trivial relative to the
field. It exhibits sufficient depth of intellect rather than superficial or simplistic
reasoning.
Soundness: The extent to which the scientific and technical procedures, measures,
methods, or models employed to generate the information is reasonable for, and
consistent with, the intended application.
Applicability and Utility: The extent to which the information is relevant for the intended
use.
Clarity and Completeness: The degree of clarity and completeness with which the data,
assumptions, methods, QA, and analyses employed to generate the information are
documented.
Uncertainty and Variability: The extent to which variability and uncertainty (quantitative
and qualitative) related to results, procedures, measures, methods, or models are
evaluated and characterized.
84
-------
6. Literature Cited
Aftosa F (2015). Foot and mouth disease. Retrieved July 22, 2016 from
http://www.cfsph.iastate.edu/Factsheets/pdfs/foot and mouth disease.pdf.
Agriculture and Agri-Food Canada (undated). Water requirements for pastured livestock.
Retrieved April 28, 2016 from
http://wwwl.agric.gov.ab.ca/$department/deptdocs.nsf/ba3468a2a8681f69872569d60073fdel/42
131 e74693 dcdO 1872572dfl)0629626/$FILE/waterrea .pdf.
Ahlstrom ME (1983). The Foxes. New York: Crestwood House.
Alexandersen S, Brotherhood I, Donaldson AI (2002). Natural aerosol transmission of foot-and-
mouth disease virus to pigs: minimal infectious dose for strain 01 Lausanne. Epidemiol Infect
128(2): 301-312.
Alexandersen S, Zhang Z, Donaldson AI, Garland AJ (2003). The pathogenesis and diagnosis of
foot-and-mouth disease. JComp Pathol 129(1): 1-36.
Arambulo PV, Steele JH (1977). A review of the natural history of foot-and-mouth disease.
Philipp J Vet Med 16: 128-165.
Ashford DA (2015). Overview of foot-and-mouth disease. The Merck Veterinary Manual.
Retrieved July 22, 2016 from
http://www.merckvetmanual.com/mvm/generalized conditions/foot-and-
mouth disease/overview of foot-and-mouth disease.html.
ATCC (2012). Converting TCID[50] to plaque-forming units (PFU). Retrieved July 24, 2017
from
https://www.atcc.org/Global/FAOs/4/8/Converting%20TCID50%20to%20plaque%20forming%
20units%20PFU-124. aspx.
Bartok JW Jr., MacKay S, Baker LD Jr, Lassoie JP (2003). Heating With Wood and Coal. Ithaca,
NY: Natural Resource, Agriculture, and Engineering Service (NRAES), Cooperative Extension.
Report No. NRAES-23, 2003 Revision. Retrieved June 15, 2015 from
http://www2.dnr. cornel!. edu/ext/info/pubs/Harvesting/Heating%20wth%20wood%20and%20coa
l.pdf.
Bauer K (1997). Foot-and-mouth disease as zoonosis. Arch Virol Suppl 13: 95-97.
Bullough WS (1942). The starling (Sturnus vulgaris L.) and foot-and-mouth disease. Proc Roy
SocB 131(862): 1-12.
Canadian Food Inspection Agency (2013). Foot-and-Mouth Disease Hazard Specific Plan.
Retrieved October 21, 2016 from http://www.inspection.gc.ca/animals/terrestrial-
85
-------
animal s/di seases/reportable/foot-and-mouth-
disease/plan/eng/13321743 53793/1332174430101?chap=0.
Chen S-J, Hsieh L-T, Chiu S-C (2003). Emission of polycyclic aromatic hydrocarbons from
animal carcass incinerators. Sci Total Environ 313(1-3): 61-76.
Chen S-J, Hung MC, Huang KL, Hwang WI (2004). Emission of heavy metals from animal
carcass incinerators in Taiwan. Chemosphere 55(9): 1197-1205. Retrieved March 18, 2015 from
http://www.ncbi.nlm.nih.gov/pubmed/15081760.
Cottral GE (1969). Persistence of foot-and-mouth disease virus in animals their products and the
environment. Bull Off Int Epizoot 71(3-4): 549-568.
County of Los Angeles (undated). Are you attracting rats and mice to your yard? Baldwin Park,
CA: Vector Management Program, County of Los Angeles, Department of Public Health.
Retrieved October 21, 2016 from
http://www.publichealth.lacountv.gov/eh/docs/Specialized/Vector Management/attractRatsMice.
pdf.
Dillon MB (2011). Skin as a potential source of infectious foot and mouth disease aerosols. Proc
Biol Sci 278(1713): 1761-1769.
Donaldson AI, Ferris NP (1978). The influence of inhibitors and antibody on the airborne
survival of foot and mouth disease virus. Proceedings of the First International Conference on
Aerobiology, 13-15 August 1978, Munich, Germany. Editor: Federal Environmental Agency.
Erich Schmidt Verlag, Berlin 1980: 263-270.
Donaldson AI, Herniman KA, Parker J, Sellers RF (1970). Further investigations on the airborne
excretion of foot-and-mouth disease virus. JHyg (Lond) 68(4): 557-564.
Edelstein S (2014). Food Science: An Ecological Approach (2nd edition). Burlington, MA: Jones
& Bartlett Learning.
Embrey M, Hunter P, Sellwood J, Wyn-Jones P, Percival SL, Chalmers R (2004). Microbiology
of Waterborne Diseases. San Diego and London: Elseview.
Ensminger ME (1992). The Stockman's Handbook (Edition 7). Danville, IL: Interstate
Publishers. https://www.amazon.com/Stockmans-Handbook-7th-M-Ensminger/dp/0813428955.
Faulkner BR, Lyon WG, Khan FA, Chattopadhyay S (2002a). Predicting Attenuation of Viruses
during Percolation in Soils - 1. Probabilistic Model. Cincinnati, OH: Office of Research and
Development, U.S. Environmental Protection Agency. EPA/600/R-02/051a. Retrieved October
18, 2016 from https://www.epa.gov/water-research/virus-fate-and-transport-virulo-model.
86
-------
Faulkner BR, Lyon WG, Khan FA, Chattopadhyay S, Cruz JB (2002b). Predicting Attenuation
of Viruses during Percolation in Soils - 2. User's Guide to the Virulo 1.0 Computer Model.
Cincinnati, OH: Office of Research and Development, U.S. Environmental Protection Agency.
EPA/600/R-02/051b. Retrieved October 18, 2016 from https://www.epa.gov/water-
research/virus-fate-and-transport-virulo-model.
Gale P (2002). Risk Assessment: Use of Composting and Biogas Treatment to Dispose of
Catering Waste Containing Meat. Final Report to the Department for Environment, Food and
Rural Affairs. Great Britain: Department for Environment, Food and Rural Affairs. Retrieved
July 22, 2016 from http://www.organics-
recycling.org.uk/dmdocuments/Risk assessment 2002.pdf.
Garner MG, Hess GD, Yang X (2006). An integrated modelling approach to assess the risk of
wind-borne spread of foot-and-mouth disease virus from infected premises. Environ Model
Assess 11(3): 195-207.
Glanville TD, Richard TL, Harmon JD, Reynolds DL, Ahn HK (2006). Environmental Impacts
and Biosecurity of Composting for Emergency Disposal of Livestock Mortalities. Des Moines,
IA: Iowa Department of Natural Resources. Retrieved March 18, 2015 from
http://www.abe.iastate.edu/cattlecomposting/files/2013/05/Emergencv-Mortalitv-Composting-
Final-Report-4 04 06-B.pdf
Gloster J, Blackall J, Sellers RF, Donaldson AI (1981). Forecasting the airborne spread of foot-
and-mouth disease. VetRec 108(17): 370-374.
Gloster J, Sellers RF, Donaldson AI (1982). Long distance transport of foot-and-mouth disease
over the sea. Vet Rec 110(3): 47-52.
Gloster J, Williams P, Doel C, Esteves I, Coe H, Valarcher JF (2007). Foot-and-mouth disease -
quantification and size distribution of airborne particles emitted by healthy and infected pigs. Vet
J 174(1): 42-53.
Guan J, Chan M, Grenier C, Brooks BW, Spencer JL, Kranendonk C, Copps J, Clavijo A (2010).
Degradation of foot-and-mouth disease virus during composting of infected pig carcasses. Can J
Vet Res 74(1): 40-44.
Gwyther CL, Williams AP, Golyshin PN, Edwards-Jones G, Jones DL (2011). The
environmental and biosecurity characteristics of livestock carcass disposal methods: A review.
Waste Manag 31(4): 767-778. Retrieved April 29, 2014 from
http://www.ncbi.nlm.nih.gov/pubmed/21216585.
Herlin AH, Andersson I (1996). Soil ingestion in farm animals. Lund, Sweden: Swedish
University of Agricultural Sciences, Department of Agricultural Biosystems and Technology,
87
-------
Report 105. Retrieved September 29, 2016 from http://www.nks.org/download/pdf/NKS-
Pub/SLU -105. PDF.
Johnson, B (2003). OSHA infectious dose white paper. AppliedBiosafety 8(4): 160-165.
Kalbasi A, Mukhtar S, Hawkins SE, Auvermann BW (2005). Carcass composting for
management of farm mortalities: A review. Compost Sci Util 13(3): 180-193.
Kitching RP (2002). Clinical variation in foot-and-mouth disease: cattle. Rev Sci Tech Offlnt
Epizoot 21(3): 499-504.
Kitching RP, Alexandersen S (2002). Clinical variation in foot-and-mouth disease: pigs. Rev Sci
Tech OffInt Epizoot 21(3): 513-518.
Kitching RP, Hughes GJ (2002). Clinical variation in foot-and-mouth disease: Sheep and goats.
Rev Sci Tech OffInt Epizoot 21(3): 505-512.
Kramer D (2015). Deer GPS study shows how whitetail bucks really travel during the rut.
Outdoor Life November 16. Retrieved October 21, 2016 from
http://www.outdoorlife.com/blogs/game-changers/deer-gps-studv-shows-how-whitetail-bucks-
reallv-travel-during-rut.
Lai HTL, Aarnick AJA, Cambra-Lopez M, Huynh TTT, Parmentier HK, Groot Koerkamp PWG
(2014). Size distribution of airborne particles in animal houses. Agric Eng Int: CIGR Journal
16(3): 28-42.
Ligocka A, Paluszak Z (2008). Evaluation of meat waste composting process based on fecal
streptocci survival. Pol J Environ Stud 17(5): 739-744.
Looper M. (2001). Whole animal composting of dairy cattle. Guide D-108. Las Cruces, NM:
Cooperative Extension Service, New Mexico Sate University.
Meeker DL (ed) (2006). Essential Rendering: All About the Animal By-Products Industry.
Alexandria, VA: National Renderers Association.
Merck (2015). The Merck Veterinary Manual: Resting Respiratory Rates. Retrieved October 10,
2016 from
http://www.merckvetmanual.com/mvm/appendixes/reference guides/resting respiratory rates.ht
ml.
Mikkelsen T, Alexandersen S, Astrup P, Chambion HJ, Donaldson AI, Dunkerley FN, Gloster J,
S0rensen JH, Thykier-Nielsen S (2003). Investigations of airborne foot-and-mouth disease virus
transmission during low-wind conditions in the early phase of the UK 2001 epidemic. Atmos
ChemPhys 3(6):2101-2110.
88
-------
Miller CW, Hoffman FO (1983). An examination of the environmental half-time for
radionuclides deposited on vegetation. Health Phys 45(3): 731-744.
Miller, L. (2015). Carcass Management During a Mass Animal Health Emergency - Final
Programmatic Environmental Impact Statement—December 2015. Riverdale, MD: United States
Department of Agriculture, Marketing and Regulatory Programs, Animal and Plant Health
Inspection Service.
Moutou F, Durand B (1994). Modelling the spread of foot-and-mouth disease virus. Vet Res
25(2-3): 270-285.
NABCC (National Agricultural Biosecurity Center Consortium) (2004). Carcass Disposal: A
Comprehensive Review. Report prepared by the NABCC, Carcass Disposal Working Group, For
the USD A Animal & Plant Health Inspection Service, Per Cooperative Agreement 02-1001-
0355-CA. Manhattan, KS: National Agricultural Biosecurity Center, Kansas State University.
Retrieved July 5, 2014 from https://krex.k-state.edu/dspace/handle/2097/662.
OIE (World Organisation for Animal Health) (2013). Technical Disease Card: Foot and Mouth
Disease. Retrieved June 26, 2016 from
http://www.oie.int/fileadmin/Home/eng/Animal Health in the World/docs/pdf/Disease cards/F
OPT AND MOUTH DISEASE.pdf.
Prempeh H, Smith R, Miiller B (2001). Foot and mouth disease: The human consequences. BMJ
322(7286): 565-566.
Sakamoto K (2011). Epidemiological developments and control of foot and mouth disease in
Asia. Conf OIE 1-7. Retrieved on February 5, 2018 from
https://www.oie.int/doc/ged/D12816.PDF
Schijven J, Rijs GB, de Roda Husman AM (2005). Quantitative risk assessment of FMD virus
transmission via water. Risk Anal 25(1): 13-21.
Schwarz M, Bonhotal J (2015). Effectiveness of Composting as a Means of Emergency
Disposal: A Literature Review. Proceedings in the 5th International Symposium on Managing
Animal Mortality, Products, By-Products and Associated Health Risk. September 25-October 1,
2015, Lancaster PA.
Sellers RF, Herniman KA, Donaldson AI (1971). The effects of killing or removal of animals
affected with foot-and-mouth disease on the amounts of airborne virus present in looseboxes. Br
Vet J 127(8): 358-365.
Sellers RF, Parker J (1969). Airborne excretion of foot-and-mouth disease virus. JHyg (Lond)
67(4): 671-677.
89
-------
Sorensen JH, Mackay DK, Jensen CO, Donaldson AI (2000). An integrated model to predict the
atmospheric spread of foot-and-mouth disease virus. Epidemiol Infect 124(3): 577-590.
Svidorov AA, Dzhupina SI, Obidor EL (1974). Transmission of foot-and-mouth disease virus by
starlings, Sturnus vulgaris. Veterinariya, Moscow 1: 49-50.
Townsend L (1997). ENTFACT-506: How Far Can House Flies Fly? Lexington, KY:
Department of Entomology, College of Agriculture, University of Kentucky, Retrieved October
21, 2016 from https://entomology.ca.uky.edu/ef506.
UM-CAHFS (University of Minnesota Center for Animal Health and Food Safety) (2014). Risk
Assessment for the Transmission of Foot and Mouth Disease via Movement of Swine and Cattle
Carcasses from FMD-infected Premises to a Disposal Site. Minneapolis, MN: University of
Minnesota Center for Animal Health and Food Safety. Downloaded January 20, 2015 from
http://deq.ne.gOv/NDEOProg.nsf/xsp/.ibmmodres/domino/OpenAttachment/NDEQProg.nsf/74E
7503FD91DC8D586257E9D004EB43F/Bodv/FMD%20Carcass%20Movement%20RA%20Fina
1%20UMN%20CAHFS%20021814.pdf.
UN FAO (Food and Agricultural Organization) (2006). Soil. Chapter 9, Soil permeability. UN
FAO Training Series, Simple methods for aquaculture. United Nations. Retrieved February 6,
2018 from
http://www.fao.org/tempref/FI/CDrom/FAO Training/FAQ Training/General/x6706e/Index.htm
University of Rhode Island (2016). Tick Encounter Resource Center: Tick Habitat. Kingston, RI:
University of Rhode Island. Retrieved October 21, 2016 from
http://tickencounter.org/faq/tick habitat.
USD A (U.S. Department of Agriculture) (2005). Operational Guidelines: Disposal. National
Animal Health Emergency Management System Guidelines. Riverdale, MD: Veterinary
Services, Animal and Plant Health Inspection Service, USD A. Retrieved July 23, 2014 from
http://www.aphis.usda.gov/emergency response/tools/on-
site/htdocs/images/nahems disposal.pdf.
USDA (2013a). Foot-and-Mouth Disease Fact Sheet. Retrieved June 13, 2016 from
https://www.aphis.usda.gov/publications/animal health/2013/fs fmd general.pdf.
USDA (2013b). Foot-and-Mouth Disease and Classical Swine Fever Standard Operating
Procedures: 9. Biosecurity. Riverdale, MD: USDA, Veterinary Services, Animal and Plant
Health Inspection Service, USDA. Retrieved June 13, 2016 from
https://www.aphis.usda.gov/animal health/emergency management/downloads/sop/sop fmdcsf
biosecuritv.pdf.
USDA (2013c). Foot-and-Mouth Disease Standard Operating Procedures: 1 Overview of
Etiology and Ecology Version 1.0. Riverdale, MD: United States Department of Agriculture
90
-------
Animal and Plant Health Inspection Service Veterinary Services. Retrieved August 10, 2016
from
https://www.aphis.usda.gov/animal health/emergency management/downloads/sop/sop fmd e-
e.pdf.
USDA (2013d). Foot-and-Mouth Disease and Classical Swine Fever Standard Operating
Procedures: 15. Cleaning and Disinfection. Riverdale, MD: United States Department of
Agriculture Animal and Plant Health Inspection Service Veterinary Services. Retrieved
November 20, 2015 from
https://www.aphis.usda.gov/animal health/emergency management/downloads/sop/sop cd.pdf.
USDA (2014a). Foot-and-Mouth Disease Response Plan: The Red Book. Riverdale, MD: United
States Department of Agriculture Animal and Plant Health Inspection Service Veterinary
Services. Retrieved June 13, 2016 from
https://www.aphis.usda.gov/animal health/emergency management/downloads/fmd responsepl
an.pdf.
USDA (2014b). Foot-and-Mouth Disease and Classical Swine Fever Standard Operating
Procedures: 8. Health and Safety and Personal Protective Equipment. Riverdale, MD: United
States Department of Agriculture Animal and Plant Health Inspection Service Veterinary
Services. Retrieved June 13, 2016 from
https://www.aphis.usda.gov/animal health/emergency management/downloads/sop/sop fmdcsf
health safety ppe.pdf.
USDA (2015). HPAI Outbreak 2014-2015, Mortality Composting Protocol for Avian Influenza
Infected Flocks. September 24, 2015. Retrieved September 25, 2015 from
https://www.aphis.usda.gov/animal health/emergency management/downloads/hpai/mortalitvco
mpostingprotocol.pdf.
USDA (2016). Potential Pesticides to Use Against the Causative Agents of Selected Foreign
Animal Diseases in Farm Settings. Retrieved March 27, 2017 from
https://www.aphis.usda.gov/animal health/emergency management/downloads/fad epa disinfe
ctants.pdf.
USDA (2017). Standard Operating Procedures: 15. Cleaning and Disinfection. Riverdale, MD:
Animal and Plant Health Inspection Service, U.S. Department of Agriculture. Retrieved March
27, 2017 from
https://www.aphis.usda.gov/animal health/emergency management/downloads/sop/sop cd.pdf.
USD A/APHIS (United States Department of Agriculture/Animal Plant Health Inspection
Services) (1994). Foot-and-Mouth Disease: Sources of Outbreaks and Hazard Categorization of
Modes of Virus Transmission. Retrieved October 21, 2016 from
http://www.fao.org/fileadmin/user upload/eufmd/USDA 1994.pdf.
91
-------
USDANRCS (U.S. Department of Agriculture, Natural Resources Conservation Service). 2004.
Saturated hydraulic conductivity: water movement concepts and class history. Soil Survey
Technical Note No. 6. Lincoln, Nebraska.
https://www.nrcs.usda.gov/Internet/FSE DOCUMENTS/nrcsl42p2 053102.doc.
USDANRCS 2008. Soil Quality Indicators. Infiltration. Available from:
https://www.nrcs.usda.gov/Internet/FSE DOCUMENTS/nrcsl42p2 053289.pdf.
USEPA (U.S. Environmental Protection Agency) (1996). EPA's Composite Model for Leachate
Migration with Transformation Products, Background Document for EPACMTP: Metals
Transport in the Subsurface, Volume 1: Methodology. Office of Solid Waste, Washington, D.C.
USEPA (1997). Exposure Factors Handbook, Volumes I, II and III. EPA/600/P-95/002F. Office
of Research and Development. U.S. Environmental Protection Agency, Washington, D.C.
USEPA (2005). Human Health Risk Assessment Protocol for Hazardous Waste Combustion
Facilities. Office of Solid Waste and Emergency Response, Report No. EPA530-R-05-006.
Retrieved March 12, 2015 from http://www.epa.gov/osw/hazard/tsd/td/combust/risk.htm.
USEPA (2017). Exposure Assessment of Livestock Carcass Management Options During
Natural Disasters. Cincinnati, OH: Office of Research and Development, National Homeland
Security Research Center, U.S. Environmental Protection Agency. Report No. EPA/600/R-
17/027. February 2017.
UWM RARC (University of Wisconsin Research Animal Resources Center) (no date).
Normative Data: Cattle. Retrieved October 28, 2016 from:
https://www.rarc.wisc.edu/animal health/normative data.html.
Wilkinson KG (2007). The biosecurity of on-farm mortality composting. JApplMicrobiol
102(3): 609-618.
Wilson WW, Matheson RC (1952). Bird migration and foot-and-mouth disease. Vet Rec 64: 541-
548.
Young CP, Marsland PA, Smith JWN (2001). Foot and Mouth Disease Epidemic. Disposal of
culled stock by burial: Guidance and Reference Data for the protection of controlled waters.
R&D Technical Report: Version 7. Solihull, Great Britain: National Groundwater &
Contaminated Land Centre, Environment Agency.
92
-------
This page left intentionally blank
-------
vvEPA
United States
Environmental Protection
Agency
PRESORTED STANDARD
POSTAGE & FEES PAID
EPA
PERMIT NO. G-35
Office of Research and Development (8101R)
Washington, DC 20460
Official Business
Penalty for Private Use
$300
------- |