-------
TABLE 2-2
Operational Status of Various Types of Installations*
Type of Combustor at Facility
Number of
Facilities
Number
Operational {%)
Multiple-hearth furnace
Flu1d1zed-bed furnace
Electric Infrared furnace
Rotary kiln
Cocombustlon with refuse
TOTAL
196
54
12
2
4
268
120 (61)
25 :(46)
8 {67)
1 (50)
2 (50)
156 (58)
*Source: U.S. EPA, 1985a
2-3
-------
Table 2-3 shows the distribution of sludge Incineration systems by plant
size, expressed as flow-treated. While most of the plants are 438 a/sec
(L/S) (10 mgd) or larger, a significant number are In the 43.8-219 L/S (1-5
mgd) range; the ratio of operating plants to total plants goes up as the
plant size Increases. Those facilities noted to be nonoperatlonal are
either no longer 1n service, still 1n construction or startup, being retro-
fitted or used seasonally. U.Su- EPA (1985a) lists, as of early 1984,
locations of 206 existing- sludge Incinerators handling municipal sludge
solids from primary, secondary and tertiary treatment.
2.1. INCINERATION
2.1.1. Uniform Feed. Uniform" feed of sludge 1s critical to the satis-
t \ ' t ' -,' V
factory operation of Incineration systems. Realization of uniform feed
requires good control of sludge thickening, blending, sludge age control 'and
pumping before dewateMng. If these tasks are managed and maintained
properly and the dewaterlng equipment Is operated correctly, the output of
the dewaterlng equipment will be uniform. l ;
If the dewaterlng equipment operates on a batch basis (such as the plate
* .«
and frame press) or Is subject to frequent upsets or outages, a sludge cake
storage and uniform feed device (such as a silo or hopper with a flight
conveyor floor and a weigh feeder) 1s required. -
An Incinerator .must have uniform |ee,d; a variation of .±10% of the
selected feed rate over 8 hours 1s acceptable. Uniform feed assures stable
operation and prevents upsets that could lead to excessive emissions. Each
Incinerator should have a dedicated weigh belt feeder, which Is calibrated
weekly, so that the operator can monitor the sludge feed rate on a strip
chart recorder or computer to record historical trends.
2-4
-------
TABLE 2-3
Distribution of Sludge Combustion Systems by Plant S1zea
Flow (mcrd)b
Systems 1n
operation
Systems not
operating
TOTAL
Percentage 1n
operation
0-1
5
9
14
36
1.1-5
28
22
50
56
5.1-10
18
18
36
50
10.1-25
60
12
72
83
25.1-50
23
5
28
82
50. Hc
17
C
22
77
aSource: U.S. EPA, 1985a
bmgd = 43.8 L/S ,
cln the category of plants larger than 50 mgd, virtually all have
multiple units, so a count based on units Installed would possibly show the
rising trend continuing.
2-5
-------
For large treatment facilities with multiple dewaterlng units and
Incinerators, the materials handling equipment (Including conveyor belts and
screw conveyors) must distribute the total sludge quantity dewatered at a
uniform feed rate to each Incinerator. For small facilities with one or two
dewaterlng units and one Incinerator, all sludge generated 1s usually
dedicated to a particular furnace. These smaller plants present special
challenges to the goal of maintaining uniform feed.
Host operating personnel recognize the need for uniform feed to Inciner-
ators. However, approximately 10-20% of U.S. Incinerators do not receive
uniform feed because of problems with equipment design, poor sludge quality
or Inadequate process control.
Nonunlform feed causes Incinerator .upsets, unstable furnace tempera-
tures and subsequent Increases In stack emissions. Nonunlform feed also
frustrates operating personnel because 1t requires frequent operating
changes and may result 1n flare-ups and Increased Incinerator hearth
maintenance 1n multiple hearth furnaces. ,. ,
The cost of maintaining uniform feed 1s minimal 1f the sludge thicken-
ing, conditioning, dewaterlng, storing and conveying fadHtes are designed
and operated properly. Importantly, these qualities depend upon a visible
commitment of plant management to good uniform feed. In any event, appro-
priate operator training, process control and preventive maintenance are
required.
2.1.2. Burning Methods.
2.1.2.1. FLUIDIZED-BED INCINERATION (FB) The FB Incinerator 1s a
plug-flow device (Figure 2-1). An FB operates with >30% excess air and
normally with a bed temperature of 1400-1600°F. The Importance of this
basic process feature became critical 1n the 1970s when rapidly escalating
2-6
-------
EXHAUST AND ASH
SAND FEED
THERMOCOUPLE
O
SLUDGE
INLET
FLUIDIZING
AIR INLET
PRESSURE TAP
SIGHT GLASS
BURNER
TUYERES
FUEL GUN
PRESSURE TAP
STARTUP PREHEAT
BURNER FOR HOT
WINDBOX
FIGURE 2-1
Cross-Section of a FlulcHzed-Bed Furnace
2-7
-------
supplementary fuel costs made the burning of 15-18% solids sludge cake
prohibitively expensive. The development and validation of the hot wlndbox
design has made FB technology economically competitive, although dewaterlng
to better than 30% solids 1s necessary to avoid excessive fuel use.
However, the United States still has many more MHF systems than FB systems
and has only a few operating hot wlndbox systems.
The FB system provides an excellent environment for the destruction of
organic chemicals because of Its excellent mixing at any cross-section, Us
uniform temperature and Us long residence time. Further, Its Inherent
limitation to working temperatures below 1700°F, and preferably to 1500°F,
limits the sublimation-condensation enrichment of the hard-to-collect fine
partlculate matter of the Important heavy metals mercury, cadmium, lead,
zinc and silver. On the negative side, the FB Inherently carries off 100%
of the ash content of the feed sludge, any unburned combustible matter and
any sand attrition 1n the flue gases exiting the furnace. Thus, the demands
placed on the air pollution control (ARC) system are high; such units
regularly meet the partlculate emission limits 1n the United States and meet
considerably more stringent limits 1n European Installations.
Although no definitive data are available, some vendors claim that
organic emissions are minimized 1f sludge 1s fed to the bed as shown In
Figure 2-1 rather than dropped through the freeboard space. Also, the use
of overflre air jets In the freeboard Improves the burnout of greases and
volatilized organlcs.
At the Hyperion plant 1n Los Angeles, which 1s based on an extensive
piloting program, a multistage FB system 1s used for Incinerating dried
c
sewage sludge. In this facility, staged combustion 1s used to limit the
degree of combustion 1n the first bed to less than sto1ch1ometric. This
2-8
-------
reduces NOX generation significantly, since 1t minimizes the peak tempera-
ture. This also results 1n a reduction 1n metals emission. Presumably, the
final stage will not produce excessive carbon monoxide or unburned hydro-
carbon emissions.
2.1.2.2. MULTIPLE-HEARTH FURNACE (HHF) An HHF 1s capable of Incin-
erating dewatered sewage sludge with a solids content between 16 and 50%
(Figure 2-2). Concentrated skimmings or scum with a solids concentration
between 25 and 60% can be colndnerated 1n furnaces In quantities of 1-3
gallons per minute (GPM).
The excess air typically used for Incinerating sludge 1s usually
75-150%, with a furnace Incinerating drier sludges and a top-hearth exhaust
gas temperature of 700-1600°F (371-871°C). Some additional residence time
should be provided after the point of sludge Introduction to ensure burnout
of organic vapors. The excess air requirement for auxiliary fuel 1s usually
10-20% as with premlx (gas) or good atomlzatlon (oil). Some of the MHFs are
equipped with afterburners for raising off-gas temperature to 1400 or 1500°F
for complete combustion of unburned hydrocarbons. The burning hearth
temperature should be controlled 200-300°F below the ash fusion temperature
to prevent slagging and, thereby, to Improve Incinerator operational avail-
ability.
2.1.2.3. ELECTRIC INFRARED FURNACE The electric Infrared furnace
represents a relatively new technological approach to the problem of sludge
Incineration. The first such unit was put Into operation 1n Richardson,
Texas, In 1975. Since that time, a number of Installations have been
constructed and others are 1n various stages of design. The furnace 1s
horizontally oriented and consists of an Insulated enclosure through which
sludge 1s transported on a continuous, woven wire conveyor belt. The
2-9
-------
COOLING AIR
DISCHARGE
RABBLE ARM
2 OR 4 PER
HEARTH
SLUDGE CAKE,
SCREENINGS,
AND GRIT
BURNERS
SUPPLEMENTAL
FUEL
COMBUSTION AIR
SHAFT COOLING
AIR RETURN
SOLIDS FLOW
DROP HOLES
CLINKER .
BREAKER l *'
P
ASH
DISCHARGE
RABBLE ARM
DRIVE
#£«V SHAFT
^ COOLING
FIGURE 2-2
Cross-Section of a Typical Multiple-Hearth Incinerator
2-10
-------
furnace consists of a steel, factory-lined shell with a thermal shock-
resistant ceramic fiber blanket Insulation system and support rollers for
the conveyor belt.
Sewage sludge Is fed Into the unit through a feed hopper. It drops onto
the conveyor belt and 1s leveled by means of an Internal roller Into a layer
~1 Inch thick, spanning the width of the belt. The sludge layer then moves
under Infrared heating elements, which provide supplementary energy (1f
required) to effect the drying and Incineration processes. The resulting
ash 1s discharged from the end of the furnace Into the ash handling system.
Combustion air 1s Introduced at the discharge end of the belt. It 1s often
preheated with an external recuperating exhaust heat exchanger. The air
picks up heat from the hot burned sludge as the sludge and air travel
countercurrent to each other.
The conveyor belt, a continuously woven steel alloy wire mesh, will
withstand the 1300-1500°F temperature encountered within the furnace (Figure
2-3). The refractory 1s ceramic felt (not brick). It does not have a high
capacity for holding heat and, therefore, the furnace can be started up from
cold condition relatively quickly.
Because the primary heat transfer mechanism utilized In the Infrared
furnace 1s radiant, combustion rates can be achieved without rabbling or
plowing of the sludge layer. Therefore, fly ash generation 1s minimized and
the control of partlculate emissions 1s simplified as compared with
multiple-hearth and fluid-bed furnaces.
Complete combustion can be achieved 1n the Infrared furnace with excess
air levels as low as 10-20%. This process efficiency 1s attributed to
several factors. First, the design of the furnace 1s such that uncontrolled
sources of excess air are eliminated; second, the flow of combustion air 1s
2-11
-------
CO
I
CM
01
=D
C3
0>
-4-1
CO
>>
CO
c
o
O)
o
c
TS
O)
CX)
(_
00
o
I/)
O)
u
o
2-12
-------
closely regulated and directed down the channel formed between the belt and
the heating elements overhead; and third, there are no gaseous products of
combustion of supplementary fuel to dilute the supply of combustion air.
This ability to operate at low excess air levels contributes to a further
reduction 1n the size, complexity and energy requirements of the exhaust gas
scrubbing equipment required for the electric Infrared furnace. Usually
only a low-energy gas scrubber, such as a cyclonic scrubber, 1s required to
elevate exhaust gases.
The system 1s divided Into several temperature control zones for the
Incineration process. These zones are maintained at predetermined tempera-
tures by closed-loop control. Temperatures are sensed by thermocouples,
compared with the setpolnt, and the input power to the Infrared heating
elements adjusted upward or downward accordingly. Control, temperatures
range from 1400°F 1n the drying zones to 1700°F 1n the combustion zones.
The flow of air for sludge combustion 1s also controlled by a closed-loop
process. The residual oxygen content 1n the exhaust stream Is detected
continuously and compared with a setpolnt value. The flow of combustion air
1s then regulated to maintain the setpolnt value. In the event that a high-
energy sludge 1s being processed, additional excess air. may be utilized to
limit exhaust temperatures t,o the 1200-1400°F range. The throughput of the
system can be controlled by adjusting the speed of the Internal conveyor
belt. This adjustment 1s accomplished from the control panel and can be
also used to adjust retention time to compensate for different sludge
feeds. To date, the Infrared furnace has been used 1n smaller applications,
where Its operating flexibility provides an advantage over traditionally
larger MHF or FB systems. Because of Its ceramic fiber blanket Insulation
system, it Is well suited for Intermittent operation. This Insulation
2-13
-------
system 1s not subject to the limitations on thermal cycling that are associ-
ated with the traditional types of solid refractory materials. Startup
times of 1-1.5 hours are normal, and shutdown 1s accomplished by pressing a
single "System Stop" button and leaving the furnace unattended until it
again needs to be operated.
2.1.3. Energy Recovery. Many new sewage sludge Incineration Installa-
tions Include energy recovery equipment. Most of the recently constructed
FB Incinerators are equipped with air preheaters where air for flu1d1zat1on
and combustion 1s preheated to as much as 1000°F. Some recent FB Incinera-
tion systems also Include waste heat recovery boilers with steam generation
for driving rotating ancillary equipment, or for providing building heat or
power generation or both. Energy recovery 1s prevalent 1n European
facilities.
Including energy recovery equipment 1n the sludge Incineration system
reduces the amount of uncontrolled partlculate emission to the scrubbers,
since a small fraction of the partlculates In the energy recovery equipment
settle out 1n the hopper bottoms.
The reduced gas temperature exiting from the boiler permits the use of
temperature-sensitive air pollution control devices, such as fabric filters
or electrostatic predpltators, which could not be used 1f energy recovery
were not used.
2.1.4. Instrumentation and Control. Reliable Instrumentation that 1s
easy to use and maintain can reduce the potential for emission upsets
because the operator can control the Incineration process more efficiently.
At present, many Incinerators are operated without the benefit of complete
Instrumentation. Operation 1s good 1f the Incinerators are receiving
nonunlform sludge feed.
2-14
-------
In the last 10 years, many MHFs have been upgraded with automatic hearth
temperature controls. Similar controls are available for FB units. Incin-
erators burning sludge autogenously (without significant fossil fuel use)
benefit greatly from automatic temperature control because 1t ensures suffi-
cient secondary combustion air (cooling) to maintain proper he.arth tempera-
tures. In addition, operators more proficiently reduce fuel and smoke by
using continuous oxygen and stack opacity analyzers. This Instrumentation
requires management support, commitment to frequent and competent mainte-
nance, and a careful review of the data to ensure proper operation. For
Incinerators built after 1973, there Is a regulatory requirement to Install,
calibrate, maintain and operate a flow measuring device (Federal Register,
1974).
The use of on-line total carbon and hydrocarbon analyzers 1s still 1n
the research and development stage. Such Instruments may Improve an
operator's ability to monitor and control the Incineration process. How-
ever, additional research, development and field evaluations are required to
determine usefulness and reliability of these on-Hne Instruments before
they can be used for operational control.
2.2. AIR POLLUTION CONTROL
Virtually all Incinerators currently operating 1n the United States are
equipped with wet scrubbers for emission control. The wide variety of wet
scrubbers 1n use Includes fixed and variable-throat Venturl Impingement
plates and cyclonic scrubbers. Most sludge Incinerators, particularly those
built In the last 10 years, are equipped with variable-throat Venturl
scrubber units ;and an Impingement subcoollng tray separator. Over 70% of
the 'incinerators Installed since 1978 are equipped with combination Venturl/
Impingement tray scrubbers. No Incinerators In the United States are
2-15
-------
equipped with electrostatic predpltators (ESPs). Several European FB
Incinerators use ESPs and achieve partlculate emissions less than the U.S.
New Source Performance Standards (NSPS) of 0.65 g/kg (1.3 Ib/ton).
Almost all MHFs are equipped with wet scrubbers. The most recent
Installations are equipped with Venturl scrubbers and Impingement plate sub-
coolers (Figure 2-4). Venturl scrubbers are capable of meeting the required
emissions discharge rate of 1.3 Ib part1culates/ton dry sludge Incinerated
at an average pressure drop of 30 Inches wet gas (W.G.) across the scrubber.
Host of the subcoolers are designed for cooling scrubbed gases to 100-120°F.
The pressure drops at which these scrubbers are operated varies widely
among different facilities. Pressure drops In the range of 4 Inches W.G. to
over 40 Inches W.G. have been reported. In general, the operating pressure
drops have Increased since promulgation of the NSPS In 1973. Most Inciner-
ators Installed 1n the past 8 years are equipped with wet scrubbers operat-
ing at pressure drops of about 30 Inches W.G. However, poor correlation
between scrubber pressure drop and partlculate emissions have been found.
The performance of wet scrubbers 1n reducing emissions of solid-phase
constituents Is Influenced primarily by the pressure drop across the central
device. As the pressure drop Increases, more solid-phase particles are
removed from the gas stream. However, a review of U.S. Incinerator emission
data shows no general correlations between scrubber pressure drop and
partlculate emissions, because the particle size distribution and total
loading are so variable between different plants that the patterns 1n the
data become confused. In other words, operating a wet scrubber at a given
pressure drop does not guarantee that any specific emission rate will be
achieved. For an Individual Incinerator, solid-phase emissions will
decrease with Increasing pressure drop. Bui the magnitude of this decrease
2-16
-------
GAS FROM QUENCHER
WATER FROM
TREATMENT
PLANT OUTFLOW
WEIR
BOX
QUENCHER
SECTION
VENTURI
SCRUBBER
FLOODED
ELBOW
MIST ELIMINATOR
WATER FROM
TREATMENT OUTFLOW
FLOODED PERFORATED
IMPINGEMENT TRAYS
FIGURE 2-4
Cross-Sectional View of a VentuM/Impingement-Tray Scrubber
2-17'
-------
will vary widely among different Incinerators and different operating
conditions.
Wet scrubbers do not as effectively reduce emissions of solid particles
with mean diameters of <1 ym and remove grease vapors. The effectiveness
decreases as pressure decreases.
To some extent, wet scrubbers also reduce emissions of gaseous species.
The major operating factor affecting performance 1n reducing gaseous emis-
sions Is the I1qu1d-to-gas ratio. Wet scrubbers typically reduce emissions
of SO- by "50%. Wet scrubbers also reduce .soluble gas emissions such as
hydrogen chloride and hydrogen fluoride, and decrease condensable hydro-
carbon emissions.
2-18
-------
3. IDENTIFICATION OF KEY PATHWAYS
The air emissions, ash residue (wet or dry) and scrubber water represent
the major pathways by which pollutants enter the environment and potentially
affect human health. Figure 3-1 1s a schematic diagram that Identifies
these pathways and their potential routes of human exposure*
3.1. AIR EMISSIONS
One of the pollutant pathways Identified 1n Figure 3-1 begins with the
partlculate and gaseous emissions generated by the combustion process.
These emissions pass through a wet scrubbing air pollution control system
that reduces the partlculate and gaseous pollutant concentrations 1n the
exhaust gas. There are several regulatory programs that exercise control
over air pollutants emitted from sludge Incineration processes.
The U.S. EPA has established ambient air standards for sulfur dioxide
(S02), total suspended partlculates (TSP), carbon monoxide (CO), ozone
(03), nitrogen dioxide (N02) and lead (Pb) as mandated by the Clean Air
Act Amendments of 1970 (P.L. 91-604). Depending on size, level of emis-
sions, and the ambient pollutant concentrations 1n the surrounding commu-
nity, a new facility or a modification to an existing facility must undergo
a New Source Review (NSR) or comply with the Prevention of Significant
Deterioration (PSD) regulations before receiving a permit to construct.
These preconstructlon requirements assure that the ambient air concentra-
tions defined by the National Ambient A1r Quality Standards (NAAQS) will not
exceed levels established to protect public health and welfare and, 1f
applicable, will not exceed the allowable Increment defined 1n the PSD
review process.
3-T
-------
Sludge
Incineration
and
Scrubbing
FIGURE 3-1
Sludge Incineration Pathways
3-2
-------
For mercury, the National Emission Standards for Hazardous Air Pollu-
tants (NESHAPs) establish an emission limit of 3200 g/day. State and local
regulations may Impose additional standards or more stringent requirements
than those defined above.
Sludge Incinerator emission limits ....are 'also established through New
Source Performance Standards (40 CFR Part 60, Subpart 0) that must be met
for partlculates [0.65 mg/kg (1.3 Ib/ton) of partlculate emitted/ton of dry
solid sludges fired]. In addition, any emissions that equal or exceed 20%
opacity are prohibited.
Airborne contaminants can affect several environmental media (see Figure
3-1). The most direct route of human exposure 1s direct Inhalation of
partlculate and gaseous emissions. Deposition of airborne partlculates on
land or 1n water bodies 1s a potential concern. Future work will assess
this route of human exposure.
3.2. ASH RESIDUE AND SCRUBBER WATER
Municipal sludge Incineration 1n well-operated facilities produces an
odorless ash weighing between 30 and 60% of the weight of the original
sludge on a dry ba,s1s. In the FB system, all ash 1s carried out the top of
the chamber with the combustion gases. Most of the ash Is removed from the
combustion gas stream by a scrubber. ..In well-operated facilities, the
resulting ash 1s completely burned out and, therefore, 1s not sooty or
tacky. In an MHF system, most of thetash exits through the bottom of the
furnace. Ash handling 1n this system Is either wet (for use In slurry pipe-
lines) or dry, although the ash 1s generally wetted for dust control before
ultimate disposal.
3-3:,
-------
Municipal wastewater plants 1n the United States generate about 72
million dry tons of sludge each year. Approximately 2 million tons of this
are Incinerated, resulting 1n approximately 700,000 tons of ash that must
finally be disposed of 1n an environmentally compatible manner.
Both dry and wet ash handling systems are currently In use. Dry ash
handling, which applies principally to MHFs, Is best when the ultimate
disposal site Is remote from the plant and when a time lag may occur between
generation and shipping. Wet ash handling Is most likely to be chosen when
a lagoonlng site Is available on or near the plant property. Wet ash han-
dling 1s essentially the only method for an FB system when wet scrubbers are
employed, because all of the ash 1s blown out of the combustor and caught 1n
the scrubber. The result Is a fairly low ash concentration compared with
the slurry from an MHF wet system. The FB ash slurry 1s usually thickened
1n a tank and may also be dewatered on a filter before 1t Is shipped to the
disposal site. Ash can usually be handled by standard earth-moving equip-
ment at the landfill site or In the lagoon cleanout process.
Trace elements found In sludge ash Include silver, cadmium, cobalt,
chromium, copper, mercury, nickel, lead and zinc. To this 11st are often
added the metalloids, arsenic and selenium, and from the alkaline earth
group, barium and beryllium. Using these elements as a reference base, a
search of the literature found various tabulations of bulk trace element
concentrations for different sludge ashes. Extraction procedure (EP)
toxldty test data on sludge Incinerator ash are presented In Table 3-1.
Strict comparisons among the. values listed 1n Table 3-1 are not advis-
able because of the Individual variations In sampling procedures, digestion
techniques and analytical methods.
3-4
-------
co
ra
js
10
c
r~
O
c
1 1
O)
§*
3 .
00 -"v
av
o -
o»
c
p-
4-1
>
^J
p
^
X
o
1
a.
LU
O
ro
Q*
_J
O>
^_)
P"
E
p
0
5
3
£
"O
03
0
r
t=
S
t_
a>
o. cz
1 0
c
t~^
C
o
r-
4->
ra
u
o
_j
V0
0
r
P~-
0
10
0
o
10
CZI
o
00
o
o
=c
II
s:
A
J->
P
o
&_
^-*
a>
C3
OO cr> 00 O P-«
O O i i O O
o o o o o in
jQ
in r 10 vo o o
O"> CO r I CSJ CO O
in o o r-> en in
CO r CO CO
in i0 co co co
"* O O O ,O O
o o o o o in
co in i CM csj
o >* < o o o
O CZ) CD Q O i
cz> i co cvj c;
C3 C? C3 CD C3 O
O o o o o in
I £.-£. i £
. ^ . ..'.'." ' ;, -i-> .
^-
E
..''. ," " .; :Zj
ac . i i .. *j
' -r^ t 1 «'';'
!-; a: s: o
3: QJ i/i .r-
0 «r- «% , X
C P >»... .C . O',-
>> tt> o c: o i
ra "O O* ra &«
QO C ro Q. rs Q.
C g. r5 .? =«= ^
O) tt> -r- (S> J-> «t
o TS ~a i_ t_ a:
t- c c ro o xj
CD t i l i Q. o. OS
»
Q\
in
3E
T3
p
"o
OO
^_
o
a>
r"
<4-
<4»
0
ft
p~
a.
' »r-
CL
cx
o
Q,
4
"5
c
o
4-'
ra o
oo , c
O1 Zl
""" E
« o .
, r«c o
a.
UJ i
c
OO O
=3 (-
0)
E 0.
o
(» E
u o
o «-
*-> (/)
a. -r-
O O)
- «C - ;'=s'
I
0^ >
u
L. V5
3 «
o .c
00 p
ra .0
3-5
-------
Despite these variations, it 1s possible to gain an acceptable evalua-
tion of elemental concentration ranges In sludge ash. Several aspects of
the data are worth expanding upon. Inspection of data on metals In raw and
digested sludge (on a wt/wt basis) Indicates the concentration effect for
all metals measured 1n sludge ash (except mercury) 1s close to a 4-fold
Increase, as would be expected for municipal sludges with -25% fixed solids.
Mercury and selenium compounds are quite volatile, even at relatively low
temperatures.
Wells et al. (1979) studied stack emissions from a sludge Incinerator
and reported the presence of organlcs such as ethyl and methyl ethyl
benzene, toluene, styrene and other aromatic compounds, as well as saturated
and unsaturated hydrocarbons. Organlcs occurring In sludge-Incinerated ash
can also Include chlorinated hydrocarbons and polynuclear aromatic hydro-
carbons .
While Furr et al. (1979) found polychlorinated blphenyls (PCBs) 1n
municipal sewage sludges, neither Furr et al. (1979) nor Farrell and Salotto
(1973) detected PCBs 1n sludge ash. However, Parrel! and Salotto (1973)
proposed that the disappearance of PCBs was not completely due to thermal
decomposition but also to the adsorption of volatilized condensates onto
partlculates that escaped collection and were emitted from the Incinerator
stack.
Based on U.S. EPA-conducted tests of sludge and Incinerator ash from TO
different cities (see Table 3-1), most Incinerator ash should be acceptable
for disposal without requiring special Subtitle C standards under the
Resource Conservation and Recovery Act (RCRA) (U.S. EPA, 1984a). All sludge
Incinerator ashes.passed the regulatory threshold standards specified Th -the
EP toxlclty test In 40 CFR 261.24. ...-; ';-.. v - ' M?
3-6
-------
If a sludge Incinerator ash should fan the U.S. EPA toxldty test, then
the ash would become a hazardous waste. The ash would then be subject to
RCRA manifesting requirements and would be disposed of 1n a permitted secure
landfill.
Scrubber water 1s Influenced by a number of factors Including the
following:
Type of Incineration process;
Temperature of combustion;
Characteristics of the feed sludge; and
Scrubber efficiency.
Incinerator scrubber water generally contains only a small amount of
biochemical oxygen demand (BOD) or chemical oxygen demand (COD). For MHFs,
the concentration of suspended solids In the water depends upon scrubber
efficiency and Incinerator type. For FB, higher concentrations of suspended
solids 1n the water may be found. Scrubber water from both MHF and FB
Incineration units will be high In dissolved CO, low In pH, and will
contain trace elements.
The scrubber water 1s usually treated separately by flocculatlon and
sedimentation to reduce Us solids content. Residue produced during
scrubber water treatment may be dewatered and disposed of with the Inclner-
.ator ash. Treated scrubber water effluent after solids removal 1s almost
universally recycled back through the wastewater treatment plant Influent.
However, problems do occur at the publicly owned treatment works (POTW) that
are due to rapid buildup of total dissolved solids. Thus, scrubber water
effluent 1s assumed to be treated to the point where any remaining
contaminants pose no significant human health risk.
3-7
-------
-------
4. METHODOLOGY FOR AIR EMISSIONS PATHWAY
4.1. OVERVIEW OF METHOD
Selection of critical pathways Involves Identifying the waste streams
that have the potential for creating an adverse Impact on humans. The
potential for an adverse Impact from air emissions, ash residue and scrubber
water depends on the assumptions made about how these waste streams are
handled and how they behave 1n the environment. General practice Involves
treating and recycling the scrubber water back to the wastewater treatment
plant or Incineration process or to a landfill. A groundwater methodology
has been developed for Iandf1ll1ng of sludge that may be used for leaching
of scrubber water (U.S. EPA, 1989).
The ash residue from an incinerator may be either bottom ash or fly ash
or both, depending upon the type of Incinerator. Where both bottom ash and
fly ash are obtained, they are usually mixed to form a composite residue.
If U.S. EPA-EP toxldty testing shows that a particular ash is hazardous,
disposal should be 1n an appropriately permitted landfill.
The greatest concern associated with air emissions from Incinerators 1s
the potential public health risk associated with the Inhalation of airborne
gases and particles. Certain emitted contaminants are suspected human
carcinogens, and others can exert other acute or chronic effects. Thus,
incinerator air emissions arc selected as the critical pathway 1n evaluating
human health risk from the sludge Incineration process.
The methodology requires that the human health and environmental impacts
be defined as follows:
1
Identify the human health and environmental Impacts and
exposures considered acceptable (Chapter 5).
Define the human population exposed to the ground level
ambient concentrations for representative facilities
(Chapter 5).
4-1
-------
3. Define, for the air exposure route, the maximum allowable
dally and annual ground level concentrations of the pollu-
tants of concern that will satisfy the health and exposure
criteria In Step 1 (Chapters 4 and 5).
4. Define the stack modeling characteristics, such as stack
height, exit diameter, gas flow, gas temperature, and meteor-
ology for selected model plants (Chapter 4).
5. Select the appropriate air dispersion model(s) for the
models/plants and calculate the emission rates associated
with the acceptable ambient concentrations (Chapter 4).
6. Calculate allowable sludge concentrations and mass loadings
for the pollutants of concern (Chapter 6).
The major assumptions for these steps are given In Table 4-1.
4.2. FATE AND TRANSPORT
4.2.1. POTW Incinerator Data Base. The U.S. EPA's Office of A1r Quality
Planning and Standards 1n early 1985 performed a telephone survey of 127
Incinerator sites 1n the United States. The survey Included 80% or more of
all operational POTW Incinerators 1n the country. The survey contains,loca-
tions and coordinates of the 127 facilities, as well as data on the capacity
of the Incinerators, the types of Incinerators, building dimensions, the
stack parameters and the type of gas scrubbing equipment.
An analysis of the data base revealed the following Information:
The 127 sites have a total of 227 Incinerators categorized as
follows:
192 multiple hearths
23 fluldlzed beds
5 electric (Infrared) furnaces
7 colnclneratlon with solid waste facilities
The total capacity of all of the Incinerators 1s 8175 dry
metric tons of sludge per day. The distribution of capacity
for the 127 sites Is shown 1n Figure 4-1. The average capacity
1s 35 dry metric tons per day.
4-2
-------
TABLE 4-1 ,, ,:;
Major Assumptions for the Sludge Incineration Methodology
Assumption
Ramification
Modeling was performed using a
unit emission rate of 1 g/sec
for each plant Incinerator stack,
The MEI resides 1n the area of the
maximum ground level concentra-
tion and Is exposed 24 hours/day.
The Incinerator Is assumed to be .
operational for the life of the
Individual (70 years) and on-
stream 100% of the time.
Body weight of adult MEI Is 70 kg
for cancer. Body weight for
systemic toxicants will vary
depending upon the age group of
the MEI.
Risk level of exposure to carcino-
gens 1s set at 10~6 (an upper-
bound excess cancer risk of one
case in one million exposed
Individuals over the background
cancer rate).
Volume of air resplrated on a
dally basis 1s 20 mVday for
a 70-kg adult.
Human exposure 1s assumed to be
100% of the annual average.
Only source of exposure 1s Inha-
lation of ambient air impacted by
sludge Incinerator emissions.
Actual concentrations can be obtained
at each receptor by scaling the 1 g/sec
concentrations by the actual pollutant-
specific emission rate after the model-
Ing has been completed, this was used
because of lack of actual emissions data
for.pollutants of concern.
Overpredlcts.
Wastewater treatment plants .are expected
to be operated Indefinitely Into the
future, ....-..
Consistent with other Agency policies.
Consistent with other Agency studies and
policies.
Consistent with other Agency studies and
policies.
Consistent with U.S. EPA assumptions for
exposures calculated 1n other studies.
Exposure from Incinerator ash will be
regulated under RCRA (Subtitle ,D);
exposure from scrubber water will be
regulated under the CWA using NPDES
permits.*
4-3
-------
TABLE 4-1 (cont.)
Assumption
Ramification
Indoor and outdoor air concentra-
tions are the same.
Model cannot adequately take Into
account the Indoor/outdoor distribution
of air concentrations.
*Terrestr1al deposition model may need to be developed to assess human
exposure.
4-4
-------
I
CO
i
30
60
90 120 150 180
Capacity, Dry Tons/Day
250 330 410
FIGURE 4-1
Distribution of Capacity for 127 Sludge Incinerator Sites
4-5
-------
The distribution of stack heights is presented 1n Figure 4-2.
The distribution of stack exit gas velocities 1s shown 1n
Figure 4-3.
The average operating week 1s 5 days.
The distribution of air pollution control systems by Inciner-
ator type 1s shown In Table 4-2.
4.2-.2. Model Plant Selection Criteria. Based on the U.S. EPA survey,
eight facilities wore selected to represent the different conditions for
sludge Incinerators In the United States. The selection of the model study
plants was made using the following factors:
Capacity - Total Incinerator capacity 1n dry tons per day;
Stack height - Stack height above ground level 1n meters;
Stack exit velocity - Stack exit velocity In meters per second;
Population - Population density around the facility, either'
rural or urban; ;
Terrain - Locations where terrain 1s known to be a potential :
factor 1n causing Increased human exposure;
Meteorology - Locations where meteorologlc factors may cause
Increased human exposures, I.e., frequent atmospheric Invert
slons.
The following ranges were selected for each of the above factors:
Capacity: Based on the distribution of Incineration capacity In
Figure 4-1, the following three ranges were selected:
-- maximum single-site capacity
- 90-100 dry tons per day
less than 10 dry tons per day
Stack height: Based on the distribution of stack heights 1n Figure
4-2, the following ranges were Identified:
5-7 meters
18-20 meters
44-46 meters
4-6
-------
(0
0)
JB
at
0
1
E
3
Z
26-
24-
22-
20-
18-
16-
14-
12-
10-
8-
6-
4-
2-
0
- ' rl -
P 8
s ^
^* s 3
tf s §
11
17
Stack Height/Meters
FIGURE 4-2 " "
Distribution of Stack Heights for Sludge Incinerators
4-7
-------
m m 1
i
10
13 16 19 22
Stack Velocity, Meters/Sec
i
25
i
28
31
FIGURE 4-3
Distribution of Stack Exit Gas Velocities for Sludge Incinerators
4-8
-------
TABLE 4-2
A1r Pollution Control Systems by Furnace Type
Wet scrubbers (unspecified)
Ventur1/1mp1ngers
VentuM/packed tower
Implngers
Wet cyclone
VentuM
Wet cyclone/lmplnger
Spray chamber
Unknown
ESP
Multiple
Hearth
64
53
NA
36
17
25
1
1
3
NA
Flu1d1zed
Bed
6
3
2
3
1
6
NA
2
.-.. NA
NA
Electric
Furnace
NA
2
NA
NA
NA
3
NA
NA
NA
NA
Co1nc1nerat1on
NA
NA
; NA
NA
NA
3
NA
NA
NA
4
NA = Not applicable
4-9
-------
Stack exit velocity: Based on the distribution of exit gas veloci-
ties In Figure 4-3, the following values were
selected:
16 meters per second
10 meters per second
3 meters per second
Population:
Terrain:
Meteorology:
Population near a wastewater treatment plant should
present a fairly high degree of capacity of, the
treatment plant and the Incinerator capacity.
Therefore, the selection of the model plants based
on total Incineration capacity should also reflect
the regional population.
At least 3 of the 10 facilities selected should have
terrain above the elevation at the stack 1n close
proximity to the Incinerator stack.
At least two of the facilities selected should be 1n
locations where Inversions or other unfavorable
meteorologlc conditions exist.
4.2.3. Facilities Selected. The 10 representative facilities selected
are shown for operating conditions and cfor design conditions 1n Table 4-3.
Each facility Is modeled for two different conditions. The operating condi-
tions represent current practice at that facility. The design conditions
represent potential future practice (maximum capacity),. Each of these 10
facilities was picked for one or more of the following reasons:
Facility 1: This facility has a moderate, capacity and a large
number of residences located above the stack level.
Facility 2: This facility Is a small capacity electric furnace
facility ,with a short stack.,
Facility 3: This facility was picked because 1t has a small
capacity flu1d1zed bed Incinerator, and although
located 1n a rural setting, 1s sited In a valley on
the Mississippi R1ver.; -..- ; r
Facility 4: This facility was selected because of Its median
capacity and tall stacks. " :
Facility 5: This facility 1s the largest Incineration facility In
the United States with a capacity of 1080 dry metric
tons per day.
4-10
-------
VI
*4
e
'" ,--.''".
i .
O>
a>
a
c/»
ei
m
2
V) ~
VI
o> c:
*! ;:
VI O
e u
o
f c
vi en
C f
0) VI
j= oi ,
o
5 *
o en
r- c
a >
co to
T3 a>
C CL
(O O
VI i ' i
L. IO
o> =>
,cu u
E 10
m
IO
CL
U
IO
CO
VI
01
IO
c
b.
o
o
>,
^
U -
IO
u_
a>
sr
o> *-»
M- -O
M- *-
LUX
,' j; ; :
I~
VI
c
o .e
v» o> »
4> O) <^-
O
e j=
i- o>
15*" S
"5
CO
"S~
O E
*~ **
,,
-4-»
U VI
«J 1=
.c|£.
2
IO 0>
U "
£§?.
'tis
IO
CO
c.
CD
«- E
0) «
3:
.
O VI
o
Z u-
?%
ig v»
_J 5
^J ^^
o i* e
** «o ^
(Q UJ
U
F QJ
v* .c
I_ O> --> -
tt irt u. E
*ss~
c >
=3 V)
. C
IO CO
L. C
ISI
O
o
. . . =
c u
O */»
_J ..
E
, 0) 01
- "2 "°
^
w
_J
2*
u
«J
u.
CM
LA
CM
en
eo
in
eo co
vi co co
o
CL
O
u. r P»
CLO O
p* , P*
r" r
CM CM
i CM
:' X X '
0) 01
in r i .
u u
4-11
P-
CO
V0
in
in
in
CM
eo
eo
,00'-
CM
I '
o
o
o
CM
in
CM
CO
eo
o
CO,
CO
CM
_
^^
CO'
3
»
CM
r~
in
CO
CO
~ 10
CO
CO
CM
CO
CO
^_
eo
e§
en
io
o
OO
O
00
0
co
0
-in !
CM
inueT
en «»
P> CM
o co*
= r- en
CM 03
*~ ^-
coco
eo'eo
en
o
f
»,
p*
CM
^^
^f
CO
3
V
in
"*
CM
IO
00
in
p» '
o <
S1
oo
p^
CO
SM
1^
*
-r*
10
tO
CO
«o:
lO
CO
10
10
eo
CO
co
o
in in
-^
o -~
coco
>0i
r*.
o en
eo CM
eo eo
CM
in
"'
^
in
in
' __
CM
' r^
3
P»
O
CM
CO
eo
en
10
5
en
in
p*
00
en
in
CO
^
" u*
vt C
X. O
J2 ^
o> *-
.- i 1
(Q U
=j T- C
ft. * ^ O>
L. U l/»
O (U
e ^ -
10 -° ~~
-------
Facility 6:
Facility 7:
Facility 8:
Facility 9:
Facility 10:
This moderately large-capacity facility Is located In
an area with a large number of residences on a hill
adjacent to the Incinerator stacks.
This facility was picked because 1t has a large
capacity and median stack heights.
This facility has a capacity close to the median and
has a very tall stack 1n a rural area.
This facility has only one unit, FB Incinerator with
a very low capacity. ,
,-'?*-,, %, . V"
This facility has only one unit, HHF Incinerator, and
1s located 1n a rural area.
The selection of these 10 facilities and the subsequent, air dispersion
modeling provide a range of conditions that result 1n an evaluation of the
Impact of the major variables of capacity, stack height, terrain and meteor-
ology on allowable emissions.
4.2.4. Modeling Long-Term Average Concentration Patterns. Since the
actual emission rate of, each chemical was unknown, modeling was performed
using a unit emission rate (1 g/sec) for each plant Incinerator stack to
establish long-term average concentration patterns. This was done so that
actual concentrations of each chemical species can be obtained by scaling
the concentration pattern by the actual species-specific emission rate after
the modeling has been completed. This has the same effect as rerunning the
model using the actual emission rate for each emitted pollutant. .
The unit Incinerator emissions were modeled using U.S. EPA-approved
computer codes and methods (U.S. EPA, 1984b). The models used In the study
were selected according to the applicability to urban or rural areas, simple
or complex terrain and available model options (such as building aerodynamic
downwash). The models selected and the rationale for selection are provided
as follows:
4-12
-------
Industrial Source Complex Long-Term (ISCLT) was utilized to assess
urban or rural concentrations when Including the effects of build-
Ing aerodynamic downwash.
LONGZ was used to model for surface concentrations 1n urban complex
terrain.
COMPLEX I was used to compute surface concentrations In rural com-
plex terrain.
4.2.4.1. MODEL SELECTION -- Most of the sewage sludge Incinerators
have short stacks that are not much higher than the buildings In which they
are located. As a result, aerodynamic building downwash was the primary
concern for the sludge Incinerator assessment. Therefore, the ISCLT model
was the primary tool of the assessment, since It 1s the only U.S.
EPA-accepted atmospheric dispersion model that considers the effects of
building aerodynamic downwash 1n computing concentrations, the limitation
of the ISCLT model Is that H Is only Intended to compute concentrations 1n
simple terrain (terrain less than or equal to stack height). Consequently,
In the case where receptors were located In complex terrain (terrain greater
than stack height), a supplemental model was used.
ISCLT model options used 1n the assessment are specified In Table 4*4.
A brief description of the ISCLT model 1s provided below. For more informa-
tion on the ISCLT model, refer to the ISCLT model user's guide (U.S. EPA,
1979).
The ISCLT model contains the following features:
Uses the cllmatologlc form of the steady-state Gaussian plume
equation for a continuous source;
Uses the generalized Briggs (1969, 1971, 1972) plume rise equations
(including momentum terms) to compute plume rise;
Uses the Pasquill-Gifford (P-G) curves (Turner, 1970) to compute
vertical dispersion;
4-13
-------
TABLE 4-4
Options Used in ISCLT Modeling
Model
Option Description
ISCLT Wind direction and speed, and Pasqu1ll-G1fford (P-G) stability data
were Input 1n the form of the STabllity ARray (STAR) data.
Annual mixing heights for each site were obtained from the
Holzworth (1972) mixing height study and were Input by stability
class according to ISCLT guidance.
Annual ambient air temperatures were obtained from local cllmato-
loglc data summaries and were Input by stability class according
to ISCLT guidance.
Wind profile exponents used were the following for P-G
stability classes A through F:
rural applications -- 0.07, 0.07, 0.10, 0.15, 0.35, 0.55
urban applications 0.15, 0.15, 0.20, 0.25, 0.30, 0.30
The midpoint of the first wlndspeed class for the STAR Input data
was specified to be 1.5 meters/second Instead of the ISCLT
default 0.75 meters/second.
Vertical potential temperature gradients and entrapment coeffi-
cients used were ISCLT default values.
The rural mode was used for all rural sites, and the urban mode 3
was used for all urban sites. Rural or urban land uses were
decided according to the U.S. EPA guidance document (U.S. EPA,
1984b).
Plume rise was specified to be distance-dependent.
Wind system measurement height was set to 10 meters.
Terrain effects were used.
Polar coordinate receptors were used to describe worst-case down-
wind concentrations.
Only stack sources were modeled and unit emission rates (e.g., 1
g/second) remained constant in time.
Aerodynamic building downwash analysis was done.
4-14
-------
TABLE 4-4 (cont.)
Model
Option Description
ISCLT , Program control parameters, receptors and source Input data were
(cont.) output.
Annual concentrations were calculated and output.
4-15
-------
Uses sector-averaging Instead of explicit lateral dispersion.
Sector-averaging 1s the assumption that over the long averaging
time Involved, the plume will be found at many azimuths. That Is,
during many hours of "east" winds, the plume at times may be blown
down the center, either edge, or elsewhere within a 22.5° segment
of arc to the west;
Uses the formulas of Huber and Snyder (1976) and Huber (1977) to
compute building downwash;
Uses wind profile exponents to compute wlndspeed variation with
height according to stability for either rural or urban applica-
tions;
Uses meteorologlc data Input as follows: statistical summaries
that categorize winds Into 16 compass-point directions, six wind-
speed classes, and six P-6 stabilities;
Allows receptors to be Input as polar or cartesian coordinates; and
f '
Allows mixing height and temperature to be specified according to
stability.
When the dispersion of Incinerator emissions 1n complex terrain was
assessed for an urban area, LONGZ was used to supplement the ISCLT modeling
results. The LONGZ model options utilized 1n the assessment are Indicated
In Table 4-5. A brief description of LONGZ 1s provided below. For more
Information on the LONGZ model, refer to the LONGZ user's guide (U.S. EPA,
1982).
The LONGZ model differs from the ISCLT model as follows:
LONGZ: Uses different plume rise equations (Bjorklund and Bowers,
1979);
Uses vertical turbulence Intensities to compute vertical
plume dimensions;
Uses "Cramer dispersion coefficients" rather than the P-G
dispersion curves used 1n ISCLT and COMPLEX I; and
Does not treat aerodynamic building downwash as does ISCLT.
4-16
-------
TABLE 4-5
Options Used 1n LON6Z Modeling
Model
Option Description
LON6Z Wind direction, wlndspeed, and P-G stability data were Input 1n the
form of STAR data.
Annual average morning and afternoon mixing height were obtained
from the Holzworth (1972) mixing height study and Input.
Annual concentrations were calculated and output.
Terrain effects were utilized.
The midpoint of the first wlndspeed class for the STAR Input data
was specified to be 1.5 meters/second Instead of the LONGZ
default 0.75 meters/second.
Wlndspeed power law was based on an emission elevation above the
meteorologlc data measurement elevation, I.e., ISW(9) = 0.
Polar coordinates were used.
The model was run 1n the urban mode.
Entrapment coefficients, dispersion parameters and wind profile
exponents were selected to be default values for LONGZ.
4-17
-------
When -Incinerator emissions Impacts were assessed 1n rural complex terrain,
COMPLEX I was used to supplement ISCLT modeling results. COMPLEX I model
options used In the assessment are shown In fable 4-6. A brief description
of COMPLEX I 1s given below. COMPLEX I Is largely based on the MPTER
model. Since there 1s no user's guide for COMPLEX I, for further Informa-
tion on the COMPLEX I model refer to the MPTER model user's guide (U.S. EPA,
1980b). The differences from MPTER; are ^given In,comment statements In the
first few pages of the COMPLEX I source, code (U.S. EPA, 1983);
The COMPLEX I model contains the following features:
i. ...'.
Is a rural complex terrain .model that uses the steady-state
Gaussian plume equation for each hour of a meteorologlc record 1n
sequence;
Does not use statistical wind summaries.,as does ISCLT and LONGZ;
Like ISCLT, uses Brlggs (1969, 1971, 1972) plume rise, P-G disper-
sion curves, and sector-averaging Instead of explicit horizontal
dispersion;
Computes the change 1n wlndspeed with height according to stability
using wind profile exponents; and; :<> >.-.,,..;. -,;,:., , ;,,
Uses meteorologlc data Input In the form of hourly observations of
wlndspeed, wind direction, temperature, stability and mixing height.
The COMPLEX I model. Is not a cl1matolog1c model (although 1t computes
annual concentrations), and 1t does not use statistical, wind summaries as
Input. VALLEY does not use statistical wind summaries. However, VALLEY
lacks two options desired 1n this assessment that are Incorporated 1n
COMPLEX I ~ the computation of the change, of wlndspeed with height using
wind profile exponents, and the adjustments made to the plume height
according to terrain and atmospheric stability; Thus, COMPLEX'''I,' Instead of
VALLEY, was selected for use 1n modeling In complex rural terrain.
4-18
-------
TABLE 4-6
Options Used In the COMPLEX I Modeling
Model
Option Description
COMPLEX I Wind direction, wlndspeed, and P-G stability data were selected
from STAR data and converted Into sequential hourly
observations usable by COMPLEX I. Annual morning and afternoon
mixing heights for each site were obtained from the Holzworth
(1972) mixing height study. Mixing height was computed to vary
with stability according to ISCLT and LONGZ guidance and was
Input with the sequential hourly meteorologlc data.
Ambient air temperature was specified to vary according to
stability class using ISCLT and LONGZ guidance and was Input
with the sequential hourly meteorologlc data.
Wind profile exponents utilized were the following for P-G
stability classes A through F:
0.07,0.07,0.10,0.15,0.35,0.55
Plume rise was specified to be distance-dependent.
Wind system measurement height was set to 10 meters.
Minimum plume helght-to-complex terrain was set to 10 meters.
Terrain effects were utilized and the terrain adjustment factors
for P-G stability classes A through F are as follows:
0.5, 0.5, 0.5, 0.5, 0.0, 0.0
Polar coordinate-configured receptors were converted to carte-
sian coordinates to describe worst-case downwind concentrations.
Only stack sources were modeled and unit emission rates
(e.g., 1 g/seC) remained constant 1n time.
Program control parameters, receptors and source Input data were
output.
Annual concentrations were calculated and output.
4-19
-------
To use COMPLEX I 1n a manner similar to ISCLT and LONGZ, modifications
to the Input data and processing of the receptor concentration results were
required. These modifications are discussed below. "f
i . [ V;-*,'
COMPLEX I can compute concentrations at a limited number of recep-
tors. Modeling the Incinerator emissions required more receptors
than could be modeled 1n a single COMPLEX I run. Thus, multiple
model runs were performed and the receptor concentration results
merged.
The model can use only sequential hourly meteorologlc data as
input. Consequently, to use the same data that were Input inta
ISCLT, a substitute statistical wind summary was needed.
Modeling the emulated statistical wind summary produced receptor-
concentrations for each hour of meteorologlc data. These hourly
receptor concentrations were then multiplied by the appropriate
statistical wind summary frequency factor, and each respective
receptor concentration was summed over all stabilities to obtain
the annual mean concentration for each receptor. The concept
underlying this approach is explained in Section 4.2.4.6.
4.2.4.2. MERGING MODELING RESULTS -- ISCLT wasi'used' to model all
receptor concentrations because it incorporates the effects of building
aerodynamic downwash. To apply ISCLT at complex terrain receptors, the
complex terrain receptor elevations were set equal to stack height. Complex
terrain receptors were also modeled using actual elevations, with the LONGZ
model for urban facilities and the COMPLEX I model for rural facilities (see
Table 4-3 for rural/urban classification of facilities). The ISCLT and
either LONGZ or COMPLEX I'result's were compared to find the higher'concen-
tration for each complex terrain receptor. The higher concentration was
then selected and merged with the ISCLT simple terrain receptor results to
obtain the complete input data file for the Human Exposure Model (HEM).
4.2.4.3. BUILDING AERODYNAMIC DOWNWASH -- The ISCLT model is'recom-
mended for computation of building downwash (U.S. EPA, 1984b). Receptors
and building dimension Input were specified according to U.S. EPA associated
4-20
-------
methods., According to U.S. EPA guidance (U.S. EPA, 1983), addressing down-
wash requires receptors to be located from 1DQ meters to 2 kilometers
downwind of the source at 100-meter Intervals. The ISCLT model requires the
Input of the width of the building adjacent to the stack. If the building
1s not square, the "effective width" of a square building of equal
horizontal area is Input.
The "effective width" for each facility configuration 1s shown 1n
Table 4-3.
4.2.4.4. RECEPTOR GRID SPECIFICATION Receptors are used beginning
at the distance of 100 meters frbm the modeled source to 2000 meters from
the sourceat 100-meter Increments. The receptors were configured 1n polar
coordinates with 16 compass-point radlals (e.g., north, ^north-northeast).
When modeling multiple sources, the grid was centered between the sewage
sludge Incinerator stacks to allow symmetric representation pf concentration
contours about the facility. Downwind rings beyond 2000 meters were speci-
fied at 5,000, 10,000, 20,000 and 50,000 meters.
Additional receptor rings were Included based on the results of Inciner-
ator plume rise screening modeling. Screening modeling of the Incinerators
plume rise was performed using the U.S. EPA-approved model PTPLU. This
model computes the worst-case surface concentrations at downwind distances
according to stability and wlndspeed. The PTPLU-computed downwind distance
of the highest concentration for each stability class was Included 1n the
receptor grid. If the Indicated rings were at a distance less than 2000
meters (where receptors selected for the aerodynamic building downwash were
already .specified at 100-meter Intervals), they were not used. Table 4-7
shows the ring distances utilized In the study according to the facility
modeled.
4-21
-------
TABLE 4-7
Ring Distances According to Facility Modeled
Facility
Receptor Ring Distances (km)
0.104, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0, 1.1,
1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, 2., 2.17, 3.67,
3.99, 5.0, 10.0, 20.0, 50.0
0.105, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0, 1.1,
1.2, 1.3,, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, 2.0, 5.0, 10.0,
20.0, 50.0
0.120, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0, 1.1,
1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, 2.0, 5.0, TO.O,
20.0, 50.0
0.125, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, l.Q, 1.1,
1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, 2.0, 5.0, 10.0,
20.0, 50.0
0.161, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0, 1.1,
1..2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, 2.0, 3.0,
3.318, 5.0, 6.186, 6.323, 10.0, 14.913, 15.0, 20.0, 50.0
0.115, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0, 1.1,
1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, 2.0, 2.198, 5.0,
10.0, 20.0, 50.0
0.122, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0, 1.1,
1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, 2.0, 2.87, 5.0,
8.96, 10.0, 20.0, 50.0
0.116, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0, 1.1,
1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, 2.0, 3.0, 3.949,
5.0, 7.979, 10.0, 20.0, 50.0
0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0, 1.1,
1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, 2.0, 5.0, 10.0,
20.0, 50.0
10
0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0, 1.1,
1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, 2.0, 5.0, 10.0,
20.0, 50.0
4-22
-------
The PTPLU model output also describes the plume height of the worst-case
downwind concentration according to'"'windspeed and stability. This height
can be used to seject receptor elevations that will describe worst-case
complex terraini, .surface concentrations .that are due to plume Impingement.
PTPLU model results of metebroldglc conditions conducive to plume Impinge-'
merit ;on ;complex""terra1n ;(evg.', P-G stability class !"F" at 2.5 meters per
second) were used to find the critical plume height. U.S. Geological Survey
maps wer'e used to select polar "coordina'te receptor ;elevat1ons.
.'--' -- , - - ~ , ,. : . , , f -.
As was Indicated 1n earlier sections, the COMPLEX I receptors needed
special treatment because; COMPLEX^, I- typically1s not used as a cllmatologlc
model. The 16 compass-point radlals normally input into ISCLT and LONGZ
cou;id,;not be used because; the'COMPLEX'I model requires radlals 1n 10-degree
Increments. Therefore, the 16-polnt radlals' and rings were processed to
obtain cartesian,coordinate5 receptors. The cartesian receptors were then
used When 'modeling with COMPLEX I.
4.2>;4,5;/\LAND^ USE!: DESCRIPTION-- Rural; or urban dispersion coeffi-
cients and wind profile coefficients were selected according to U.S. EPA
guidance;; {U.S._EPAv1984b)-. Thls^ was achieved^ by|means of a land-use-type
-. .- - .' '< ,. ; _.- ; *j
scheme (Auer, 1978) to assess land use within a 3-kilometer radius of the
facility., Accprding to,;;the schen)e,;Tf. certain land-use types corresponding
to urban descriptions account "fof ">50% of the area within the 3-kilometer
radius, urban, dispersion/ and wfnd profile "coefficients should be used.
Conversely, 1f rural land-use types account for ^50% of the area within the
3-kilometer raid1,us;,"y;uraT Dispersion1 arid -wind prof fie coefficients should be
used. The results of this assessment are Indicated in Table 4-3.
4.2.4.6. HETEOROLOGIC DATA SELECTION The meteorologic data used by
the ISCLT and LONGZ models were Input In the form of joint frequency
4-23
-------
distributions broken down by wlndspeed, wind direction and Pasqu1Tl-G1fford
(P-G) stability. These data, sometimes referred to as "statistical wind
summaries," are also known as STablHty ARray (STAR) data and are generally
available for major commercial airports. Table 4-8 Indicates the sludge
incineration sites and the nearest available meteorologlc stations for which
STAR data were available. The data sets selected represent the best
available data found 1n the area of each sludge Incineration facility.
Stability class data for each STAR distribution were evaluated for
neutral stability day-night splits and unit total frequency. Day-night
neutral distributions were combined to obtain an overall neutral stability
class. All STAR data base frequency of occurrence totals summed to unity.
The ISCLT and LONGZ models utilize annual mixing heights and tempera-
tures according to stability. Annual mixing height and ambient temperature
data were taken from cl1matolog1c literature (Holzworth, 1972; NOAA, 1978)
and were specified to vary by stability according to ISCLT and LONGZ guide-
lines. Table 4-9 shows the mixing heights and temperatures according to
stability, which were used as input In the modeling assessment for each
facility.
The special treatment required for the COMPLEX ,1 meteorologlc wind data
consisted of developing an appropriate wind summary for the site by using
hourly meteorologlc data. The statistical wind summary was developed by
specifying each hour of meteorologlc data to correspond with each wind
direction, wlndspeed and stability combination of the wind summary. The
basis for this approach is that, in effect, the resultant concentration at a
receptor for 1 modeled hour of a wlndspeed, wind direction and stability is
the same as the annual average concentration at the same receptor for 1 year
(8760 consecutive hours) of the same wlndspeed, wind direction and stability.
4-24
-------
TABLE 4-8
Meteorologlc Input Data for ISCLT and LONGZ Models
Facility
1 '
2
;' 3
4-:- - ; '
-. 5 -
6
7
8
9
10
Period
of
Record
1966-1970
1959-1963
1960-1964
1969-1973
1969-1973 '
1958-1962
1969-1973
1964-1973
1965-1970 ' '
1966-1970
Time Between
Observations
(hours)
3
1
1
3
3
' i ' -
3
3 - ' "
3
1
Summary
Type
annual
annual
annual
annual
annual
annual
annual
annual
annual
annual
4-25
-------
TABlE 4-9' '''
Mixing Height and" Ambient
Stability
Temperature Data' According" to
and Facility,,.,. \,, , H .... .iit . .,,s .,JWi..
'"' ' Pasqui 11 -61 fforcT Stability ' f ;A'" ""'"
Facility
1
2
3
4
5
6
7
8
9
10
Data
temperature
mixing height
temperature
mixing height
temperature
mixing height
temperature
mixing height
temperature
mixing height
temperature
mixing height
temperature
mixing height
temperature
mixing height
temperature
mixing height
temperature
mixing height
A
292
2255
288
1760
288
1938
287
1700
294
2243
285
1760
291
1985
291
2255
289
1352
294
1950
B
292
1503
288
1173
288
1292
287
1113
294
1495
285
1173
291
1323
291
1503
289
901
294
1300
C
292
1503
288
1173
288
1292
287
1113
294
1495
285
1173
291
1323
291
1503
289
901
294
1300
D
287
971
280
794
283
849
281
766
289
910
280
794
286
866
286
971
286
888
288
950
E
282
439
275
415
278
406
276
419*
284
324
275
415 ,
280
408
280
439
282
875
282
600
F
282
439
275
415
278
406
276
419*
284
324
275
415
280
408
280
439
282
875
282
600
*Temperature 1s In degrees Kelvin; mixing height 1s 1n meters
4-26
-------
Each receptor was modeled for all combinations of stability, wlndspeed
and wind direction found 1n the statistical wind summary. These receptor
concentrations were output for each hour of meteorologlc data. The annual
average concentrations were then obtained by processing these hourly concen-
trations with the appropriate wind summary frequency factor and summing each
receptor concentration over all stabilities.
4-27
-------
-------
5. EXPOSURE AND ASSESSMENT OF HEALTH EFFECTS
Two approaches for exposure health assessment may be performed by this
methodology: an aggregate risk and the most exposed Individual (MEI). The
human exposure model (HEM) assesses aggregate risk as described In Section
5.1. The MEI methodology 1s described 1n Section 5.2.
5.1. HUMAN EXPOSURE MODEL (HEM)
The U.S. EPA's Office of Air Quality Planning and Standards' Pollutant
Assessment( Branch has developed a human exposure model. The Impact param-
eters (exposure, hazard and risk) are the basis of the HEM computations of
the Individual and community health effects resulting from the emissions of
chemical species. These concepts are defined as follows:
Population The number of persons In contact with the concentra-
tion.
Exposure The population multiplied by the concentration.
Carcinogenic Potency This parameter Is quantified by the unit
risk factor, the probability of developing cancer due to continuous
exposure to 1 ng/m3 of the species over a 70-year lifetime.
Hazard Concentration multiplied by the unit risk factor.
Risk Exposure multiplied by the unit risk factor.
The human exposure model uses a finely detailed national census data base to
compute the Impact parameters of exposure and dose (Anderson et al., 1981;
Anderson and Lundberg, 1983). The resulting risk patterns are dependent not
only on concentration but also on population patterns.
5.1.1. Exposure. The degree of contaminant exposure to Individuals
residing 1n an area where emissions from a municipal sludge Incinerator
exist depends upon the following:
5-1
-------
the duration of exposure; .
the volume of air Inhaled;
the particle size distribution for Incinerator emissions;
the annual average contaminant concentrations; and
the number of people exposed.
5.1.1.1. EXPOSED POPULATION The population that 1s assumed for the
exposure estimate 1s from the 1980 census population data. The population
Is located within a 50-kilometer radius for each case-study Incinerator.
This approach calculates the annual ground level concentration from the
stack for the area around the Incinerator and then superimposes the location
and numbers of people on a concentration grid. The exposures are calculated
for each section of the grid and then are summed to give an aggregate risk
value for the total population. The assumption of a 70-year exposure 1s
used to calculate the lifetime aggregate cancer risk.
The exposure estimate Includes consideration of the amount of time that
the Individual spends 1n a location exposed to the Incinerator emissions.
Also, the duration of exposure Is affected by the percentage of time on an
annual basis that the Incinerator 1s operational and the expected operating
life of the system.
5.1.2. Inhalation Volume. This refers to the volume of air Inhaled on a
dally basis. A volume of 20 m /day Is assumed, which 1s consistent with
other Agency analysis and the U.S. EPA's Human Health Assessment Group
calculations. ,
5.1.3. Particle Size Distribution. This refers to the distribution of
sizes of the partlculate emitted from the Incinerator. For this analysis,
1t 1s assumed that 100% of the Inhaled partlculate enters and 1s retained 1n
the upper and lower respiratory tract. Any clearance of partlculates from
the upper respiratory tract results in partlculate Ingestlon and absorption.
5-2
-------
5.1.4. Contaminant Concentration. Ambient concentrations are predicted
In terms of an annual average 1n mlcrograms; per cubic meter (yg/m3).
The human exposure 1s assumed to be 100% of the, annual average^ which 1s
consistent with the U.S. EPA assumptions for exposures calculated 1n other
Agency studies.
5.1.5. Model Description, 'fhfe model consists of two major computational
algorithms (a plume dispersion computation algorithm and an exposure/risk
algorithm) and major Input data files, the HEM does not make use of the
plume computation module, since the annual concentration patterns will be
computed using more sophisticated U.S. EPA-approved atmospheric dispersion
models (U.S. 'EPA, 1984b) described 1n Chapter 4. the exposure/risk algo-
rithms Involve 'joint processing of the computed concentration patterns and
human population data files. Population'data resolved at the level of
census enumeration districtsTare "usedf with specif fealty located facility
concentration patterns.
5.1.6. HEM Estimation Scheme. A two-level scheme was adopted to pair up
concentrations and populations before computing exposures and risks. The
two-level approach 1s appropriate because the concentrations are defined on
a radius-azimuth (polar) grid pattern with nonuniform spacing,' because the
fine/coarse relationship varies with radius. At small radii, the grid cells
are much smaller than enumeration district/block groups (ED/BGs); at large
rad11 the grid cells are much larger than ED/BGs. To form the product of
the 'population'times concentration, both factors at the same set of points
are required. Techniques to accomplish this are most appropriately applied
by Interpolating values of the factor*'defined on the coarse network at the
locations of the finer grid, thus maximizing the resolution and minimizing
the uncertainties of Interpolation. "
,5-3
-------
For ED/B6 centrolds located between 0.1 and 2.8 kilometers from the
source, populations will be apportioned among neighboring concentration grid
points. Associated with each of these grid points, at which the concentra-
tion Is known, Is a smaller polar sector bounded by two concentric arcs and
two radial lines. The boundary concentric arcs were the downwind receptor
rings (located at 0.1, 0.2, etc., kilometers from the source), and the
boundary radial lines will be drawn In the middle of two wind directions.
Each of these concentration grid points will be assigned to the nearest
ED/BG centrold Identified In the population data set. .The population at
each centrold will then be apportioned among all concentration grid points
assigned to that centrold. The exact land area within each polar cell will
be considered 1n the apportionment, and the population density will be
assumed to be the same for all grid cells assigned to a single centrold.
Both concentration and population counts will be available for each polar
grid point.
Log-linear Interpolation will be used to estimate the concentration at
each EO/B6 population centrold located between 2.8 and 50 kilometers from
the source. For each ED/BG centrold, four reference points will be located
as the four corners of the polar sector In which the centrold 1s located.
These four reference points would surround the centrold as depicted 1n
Figure 5-1. The linear relationship that Is known to exist between the
logarithm of concentrations and the logarithm of distances for receptors >2
kilometers away from the source would be used to estimate the concentrations
at points E and F (see Figure 5-1). These estimates, together with the
polar angles, will then be used to Interpolate the concentration at the
centrold.
5-4
-------
SOURCE
ED/BG CENTROID
FIGURE 5-1
Reference Points for an Enumeration District/Block Group (ED/BG) Centrold
5-5
-------
5.2. HOST EXPOSED INDIVIDUAL METHODOLOGY
The most exposed Individual (MEI) 1s assumed to reside 1n the area of
the maximum annual ground level concentration and 1s exposed 24 hours a day.
The Inclnerator(s) will be assumed to be operational for the life of the
Individual (70 years) and to be operating 100% of the time. The 70-year
exposure 1s a valid estimate since sludge Incinerators are expected to be
operational Indefinitely. A reference air concentration (RAC) 1s calculated
as the maximum concentration that the MEI will be permitted to be exposed to
for any particular contaminant.
5.2.1. Reference A1r Concentration Derivation. A reference air concen-
tration (RAC, In mg/m3) Is defined as an ambient air concentration used to
evaluate the potential for adverse effects on human health as a result of
sludge Incineration. That 1s, for a given Incinerator site, and given the
practice definitions and assumptions stated previously 1n this methodology,
the criterion Is the sludge contaminant concentration that 1s calculated to
result 1n air concentrations below the RAC at a compliance point downwind
from the site. Exceedance of the RAC would be a basis for concern that
adverse health effects may occur 1n a human population 1n the site vicinity.
RAC 1s determined based upon contaminant toxldty and air Inhalation
rate, from the following generalized equation:
Reference A1r Concentration: RAC (mg/m3) = I /I (5-1)
pa
where I Is the acceptable pollutant Intake rate (1n mg/day) based on the
potential for health effects and I 1s the air Inhalation rate (1n
mVday). This simplified equation assumes that the Inhaled contaminant Is
absorbed Into the body, through the lungs, at the same rate 1n humans as 1n
the experimental species tested, or between routes of exposure (e.g., oral
and Inhalation). Also, this equation assumes that there are no other
5-6
-------
exposures of the contaminant from other sources, natural or mahmade. I
P
win vary according to the pollutant evaluated and according to whether the
pollutant acts according to a threshold or nonthreshold mechanism of
toxlclty as shown below.
5.2.1.1. THRESHOLD-ACTING TOXICANTS Threshold effects are those
for which an acceptable (I.e., subthreshold) level of toxicant exposure can
be estimated. For these toxicants, RAC 1s derived as follows:
Reference Air Concentration: RAC (mg/m3) = Rf° x bw _ JBI * la
RE ",.. (5"2)
where:
RfD = reference dose for Inhalation (mg/kg/day)
bw = human body weight (kg)
TBI = total background Intake rate of pollutant from all other
sources of exposure (mg/day)
Ia = air Inhalation rate (m3/day)
RE = relative effectiveness of Inhalation exposure (unltTess)
The definition and derivation of each of the parameters used to estimate RAC
for threshold-acting toxicants are further discussed below.
f
5.2.1.1.1. Reference Dose (RfD) When toxicant exposure 1s by
1ngest1on, the threshold assumption has traditionally been used to establish
an "acceptable dally Intake," or ADI. The Food and Agricultural Organiza-
tion and the World Health Organization have defined ADI as "the~daily intake
of a chemical which, during an entire lifetime, appears to be without
appreciable risk on the basis of all the known facts at the time. It 1s
expressed 1n milligrams of the chemical per kilogram of body weight (mg/kg)"
(Lu, 1983). Procedures for estimating the ADI from various types of toxico-
logic data are outlined by the U.S. EPA (1980a). More recently the Agency
has preferred the use of a new term, the "reference dose," or RfD, to avoid
the connotation of acceptability, which is often controversial.
5-7
-------
RfD Is defined for the purposes of this document as that dose, In mg/kg/
day, which Is estimated to be without effects 1n sensitive, Individuals
during a lifetime Inhalation exposure. RfD Is estimated from observations
In humans whenever possible. When human data are lacking, observations 1n
animals are used, employing uncertainty factors as specified by existing
Agency methodology.
Values of RfD for noncardnogenlc (or systemic) toxldty have been
derived by several groups within tho Agency. An effort 1s currently under
way to verify these values and to produce a master list of RfDs for use by
the various Agency programs. Most of the noncardnogenlc chemicals that are
currently candidates for sludge criteria for the Incineration pathway are
Included on the Agency's RfD 11st, and thus no new effort w1ll.be required
to establish RfDs for deriving sludge criteria. For any chemicals not so
listed, RfD values should be derived according to established Agency
procedures (U.S. EPA, 1990).
5.2.1.1.2, Human Body Weight (bw) and A1r Inhalation Rate (I,)'
«
The choice of values for use 1n risk assessment depends on the definition of
the Individual at risk, which 1n turn depends on exposure and susceptibility
to adverse effects. The RfD (or ADI) was defined before as the dose on a
body weight basis that could be safely tolerated over a lifetime. An
assumption of 20 m3 Inhalation/day by a 70-kg adult has been widely used
1n Agency risk assessments and will be used 1n this methodology when adults
are Identified as the MEI. Table 5-1 shows values of I for a typical
Q
man, woman, child and Infant with a typical activity schedule, as defined by
the International Commission on Radiological Protection (ICRP, 1975).
Additional values have been derived for an adult with the same activity
schedule, but using upper-limit rather than average assumptions about respi-
ration rates for each activity; and for an adult with normal respiration
5-8
-------
T
S!
£
t:
. '' ' '
VI >
'-' 0) l/l
CD 0)
_J >
CD
>» 1
H-T >|
> *->
S' "» r
u «-
«c -4->
u
"ra .- . ' -
«- re
CL E
>> u.
t o ,
-4-> I
Q C
re
i/i x: , "
re i
3 t-
D CD
i- x: - .
. > O)
5 IE
. c
' o '
t~
re 03
i- 3 ^
,0 r B
Z O E
C
VI >>«-
re o i/i
3 >-> 0)
a re E
>£,=.
H- Q. 0
"O t/) >
c o> *
HH Qi
s r
u E
0* O
0) 1
O) , .
0) S
E
3 .C
iS
^* y^
' .2 r~~
re x!
"S. t. ;
I/I O
CD it-
ce
:>> c:
> ra
ra
a
&
CD
ore ~
C «4J U.
a> c re
u. ra a>
i£ "c: ^
0) 1 1 r
£
S
0) .
U 1*1
c ra t-
CD X» ra
U r C>
0) «- >>
t So
s « -
/'
a>
f- r^
*-> Z)
u ~o
-*^
' ( 3
S. "O
Q|
CL
~J
E
U
C ro
(U C
U CO
0) E
4- O
.4J3
a±
CD
c
1- C
CD re
s
t/1
10 3 1 1 1
O 1 < 1
o
r*"
CM 1 CM :
i i
u
1 1 *
* oo
r
XI XI
o i or
CVJ CM
i r i i
> 0 0
" re re
i- . *> x: x:
> U Ol Ol
i- re «- »-
re -w
re r"~ ^~ ,
w i- re re
x: 0} c: c
01 -o o o
^" E ** ^*
re «
* o. a.
01 01 3 3
c c u u
f- «- U U
.* J< 0 -0
"- 1- C C
o o o o
X * c c
3 333
O O O O
d XT .C JZ
1 111
oo oo oo r~
t/i
t- ' ao
| CO 3 CO CO
r x; co o cs
«- "~
*
... co in
CO 1^ ^}-
1 >
CM « CO O
CO
u ^>
m *p* 01 ^}*
cvj co oo in C3
*
o r~
1 O O «-
r*- r in o
CO 10
i en i co
CM r CO O
CM in
CO
I Ip CO CO
CO CM O O
CM r-r
01
3 cn
o IB
>. >.
>- XJ
>» o o
-4-» -U XI
H ^ °
U Irt Ol 0}
re a> .*: e
fc. ^ 3
> *" >»-«-> O
re i/i t x; >
O) a> T- 01
a CD o
oo "re >> ^
x: x: -»-> -o -»-»
ii o o re
i op ( OQ ce
-
j^
en
*
ra
o
^
XJ
c
re
o
c
ra
^
4-*
I/I
>1
^
O
01
01
re
u
^»
|7;
0
ra
-4-*
cn
O
<4-
," s
ra
E
§
*
o
'c
I
L.
ra
c
o
Q
j_
O
14-
in in
I a>
en 3
ra
a.
at ->->
" 1
u T
U. -£-
1- 01
3 O.
O O.
VI =>
re xt
_^
p
cn
'
w
'~
.c
re
o
o
', °^
^
re
T3
C
re
pi
t/i
J*
i/i
a>
t-
0
Ol
CD
re
u
^>
t
^A
ra
ra
r-
=
i-
o
1"
ra
t-
X!
O
O
14-
-
M
C
ra
E
o
*
&-
o
>>
re
^
x>
i_
o
= v
u-
1/1
CD
3
S
X>
CD
Ol
ra'
i-
CD
>
«c
Eu ^
,
en
c
g
x:
t-
U-
j_.
1/1
- -«-»
ra
1/1
t-
re
cu
^
cn
CO
1
CM
O)
01
re
3
re
i_
o
U-
CD
^J
C
' U '
^
CD
0.
.C
tn
cn
fe"J
a.
a.
=3
O'
'*--,
. -; r
f-
o
VI'
CD
O
Q.
Q.
-.$
^ t
«
'-*)
=1
^_
U.
0
u-
i/i
0)
3
f
,ra
X)
CU
u
c
0)
CD
U-
0)
U
cu
x:
4-*
Q
t-
01
^
L.
ai
X)
S
ai
XI
CD
ra
.c
VI
o
ra
t-
4->
x:
Ol
1
"o
1
o
1
rj
o
nhalatlon v>
o>
5-9
-------
rates, but whose work 1s moderately active and who practices 1 hour of heavy
activity (I.e., strenuous exercise) per day (Fruhman, 1964; Astrand and
Rodahl, 1977). Representative body weights have been assigned to each of
these Individuals to calculate a respiratory volume-to-body weight ratio.
[Note: These ratios have been derived for-Illustrative purposes only.] The
resulting ratio values range from 0.33 to 0.47 m3/kg/day, all of which
exceed the ratio value of 0.29 mVkg/day estimated for the 70-kg adult
Inhaling 20 mVday, used currently by the Agency. Therefore, the
typically assumed values for adults may underestimate actual exposure. In
cases where, effects have a short latency (<10 years) and where children and
Infants are known to be at special risk, 1t may be more appropriate to use
values of bw and I for children or Infants. ;
O
5.2.1.1.3. Total Background Intake Rate of Pollutant (TBI) It Is
Important to recognize that sources of exposure other than those from the
sludge disposal practice may exist, and that the total exposure should be
maintained below the RfD. Other sources of exposure Include background
levels (whether natural or anthropogenic) In drinking water, food or air.
Other types of . exposure, which are due to occupation or habits such as
smoking, might also be Included depending on data availability and
regulatory policy. These exposures are summed to estimate TBI.
Data for estimating background exposure usually are derived from analyt-
ical surveys of surface, ground or tap water, from FDA market basket surveys
and from air-monitoring surveys. These surveys may report means, medians;
percentHes or ranges, as well as detection limits. Estimates of TBI may be
based on values representing central tendency or on upper-bound exposure
situations, depending on regulatory policy. Data chosen to estimate TBI
should be consistent with the value of bw: Where background data are
5-10
-------
,rePorted ,in terms of a; concentration-;1n air .or water, Ingestion or Inhala-
;, tlon ,srates applicable to adults or children can. be used to estimate the
-proper daily background Intake value. Where data, are reported as total
,da1lSf d1.etary intake for adults and similar values ,for children are unavaH-
i;iable, conversion to a.n Intake for .children may be required. Such a conver-
eouTd :,be estimated .pn, the basis of relative .total food intake or
total caloric intake, between adults and children. «; > *
-------
BI = background Intake of pollutant from a given exposure
route, Indicated by subscript (mg/day)
RE = relative effectiveness, with respect to Inhalation
exposure, of the exposure route Indicated by subscript
(unltless)
5.2.1.1.4. Fraction of Inhaled A1r from Contaminated Area It 1s
recognized that an Individual exposed to air emissions from an Incinerator
may not necessarily remain 1n the Incinerator proximity for 24 hours/day.
However, 1f 1t 1s assumed that residential areas may be contaminated, 1t 1s
likely that less mobile Individuals will Include those at greatest risk.
Therefore 1t Is prudent to assume that 100% of the air breathed by the most
exposed Individuals will be from the area of the Incinerator.
5.2.1.1.5. Relative Effectiveness of Exposure (RE) -- RE 1s a unit-
less factor that Indicates the relative toxlcologlc effectiveness of an
exposure by a given route when compared with another route. The value of RE
may reflect observed or estimated differences In absorption between the
Inhalation and Ingestlon routes, which can then significantly Influence the
quantity of a chemical that reaches a particular target tissue, the'length
of time 1t takes to get there, and the degree and duration of the effect.
The RE factor may also reflect differences 1n the occurrence of critical
toxlcologlc effects at the portal of entry. For example, carbon tetra-
chlorlde and chloroform were estimated to be 40 and 65% as effective,
respectively, by Inhalation as by Ingestlon based on high-dose absorption
differences (U.S. EPA, 1984d,e). In addition to route differences, RE can
also reflect differences 1n bloavallablHty due to the exposure matrix. For
example, absorption of nkkel Ingested 1n water has been estimated to,be.5
times that of nickel Ingested 1n food (U.S. EPA, 1985c). The presence of
food 1n the gastrointestinal tract may delay absorption and reduce the
5-12
-------
availability of orally administered compounds, as demonstrated for halo-
carbons (NRC, 1986).
Physiologically based pharmacoklnetlc (PB-PK) models have, evolved Into
particularly useful tools for predicting disposition differences due to
exposure route differences. Their use 1s predicated on the premise that an
effective (target-tissue) dose achieved by one route 1n a particular species
1s expected to be equally effective when achieved by another exposure route
or In some other species. For example, the proper measure of target-tissue
dose for a chemical with pharmacologlc activity would be the tissue concen-
tration divided by some measure of the receptor binding constant for that
chemical. Such models account for fundamental physiologic and biochemical
parameters such as blood flows,, ventilatory parameters, metabolic capacities
and renal .clearance, tailored by the phys1cochem1cal and biochemical prop-
erties' of the agent In question. The behavior of a substance administered
by a different exposure route can be determined by adding equations that
describe the nature of the new Input function. Similarly, since known
physiologic parameters are used, different species (e.g., humans vs. test
species) can be modeled by replacing the appropriate constants. It should
be emphasized that PB-PK models must be used 1n conjunction with toxldty
and mechanistic studies In order to relate the effective dose associated
with a certain level of risk for the test species and conditions to other
scenarios. A detailed approach for the application of PB-PK models for
derivation of the RE factor Is beyond the scope of this document but the
reader Is referred to the comprehensive discussion 1n NRC (1986). Other
useful discussions on considerations necessary when extrapolating route to
route are found 1n Pepelko and Withey (1985) and Clewell and Andersen (1985);
5-13
-------
Since exposure^ for the air pathway Is by Inhalation, the ;RE factors,
applied are ,all with respect to the Inhalation route. Therefore,' the value;
of RE In equation 5-2 gives the relative effectiveness of the exposure route'
and matrix on which the RfD was based when compared with Inhalation of con-:
tamlnated air. Similarly, the RE factors In equation 5-3 show the relative
effectiveness, with, respect to the inhalation route, of each background?
exposure route and matrix. " .'/*f$w
An RE factor should only be applied where well documented/referenced
Information Is available on the contaminant's observed relative 'effective^
ness or Its pharmacok1net1cs. When such Information Is not available, RE 1s:
equal to 1. /;
5.2.1.2. CARCINOGENS -- For carcinogenic chemicals, the Agency
considers the excess risk, of cancer to be linearly related to dose (except
at high dose levels) (U.S. EPA, 1986a). The threshold assumption,
therefore, does not hold, as risk diminishes with dose but does not become
zero or background until dose becomes zero.
The decision whether to treat a 'chemical as a threshold- or non-
threshold-acting (I.e;, carcinogenic)' agent depends oh the weight of the
evidence that It may be carcinogenic ,to humans. Methods for classifying
chemicals as to their weight of evidence have been described by U.S. EPA
(U.S. EPA, 1986a), and most of the chemicals that currently are candidates
for sludge criteria have recently been-classified In Health Assessment
Documents or other reports prepared by the,'U.S.. EPA's Office of Health .and
Environmental Assessment (OHEA), or 1n connection with the development of
recommended maximum contaminant levels (RMCLs) for drinking water contaml-
' * ;-
nants (U.S. EPA, 1985d)., To derive values of RAC, a decision, must be made
as to which classifications constitute sufficient evidence for basing a
5-14
-------
quantitative risk assessment on a presumption of carclnogenlcity. Chemicals
1n classifications A and B, "human carcinogen" and "probable human carcino-
gen, !' respectively, have usually been assessed as carcinogens, whereas those
1n classifications D and E, "not classifiable as to human cardnogenlclty
because of Inadequate human and animal data" and ^evidence of noncardno-
gehicUy for .humans," respectively, have usually been assessed according to
threshold effects. Chemicals classified as C, "possible human carcinogen,"
have received varying treatment. For example, llndane, classified by the
Human Health Assessment Group of the U.S. EPA as "B2-C", or between the
lower range of the B category and category C, has been assessed using both
the linear model for tumorlgenlc effects (U.S. EPA, 1980a) and based on
threshold effects (U.S. EPA 1985d). Table 5-2 gives an Illustration of
these U.S. EPA classifications based on the available weight of evidence.
The use of the we1ght-of-ev1dence classification without noting the
explanatory material for a specific chemical may lead to a flawed conclu-
sion, since some of the classifications are exposure-route-dependent.
Certain compounds (for example,* nickel) have been shown to be carcinogenic
by the Inhalation route but not by 1ngest1on. The Issue of whether or not
to treat an agent as carcinogenic by 1ngest1on remains controversial for
several chemicals.
If a pollutant Is to be assessed according to nonthreshold, carcinogenic
effects, the reference concentration In air, RAC, 1s derived as follows:
Reference A1r Concentration: RAC = (mg/m3)
where:
/RL x bw\
\qi* x RE,/ "
TBI
* la (5-4)
qi* = human cancer potency [(mg/kg/day) a]
RL = risk level (unltless); e.g., 10~5, 10~6
bw = human body weight (kg)
5-15
-------
TABLE 5-2
Illustrative Categorization of Evidence Based on Animal and Human Data*
Animal Evidence "
Human
Evidence
Sufficient
Limited
Inadequate
No data
No evidence
Sufficient
A
Bl
B2
B2
B2
Limited
A
Bl
C
C
G
Inadequate
A
Bl
D
D
D
No Data
r
A
Bl
D
D
D
-. No
Evidence
A -:
fil
D
.. ; E .-
, E -
*The above assignments are presented for Illustrative purposes. There may
be nuances In the classification of both animal and human data Indicating
that different categorizations than those given 1n the table should be
assigned. Furthermore, these assignments are tentative and may be modified
by ancillary evidence. In this regard all relevant Information should be
evaluated to determine 1f the designation of the overall weight of evidence
needs to be modified. Relevant factors to be Included along with the tumor
data from human and animal studies Include structure-activity relationships,
short-term test findings, results of appropriate physiologic, biochemical
and toxlcologlc observations, and comparative metabolism and pharmaco-
klnetlc studies. The nature of these findings may cause an adjustment of
the overall categorization of the weight of evidence.
5-16
-------
RE = relative effectiveness of Inhalation exposure (unltless)
Ia = air Inhalation rate (mVday)
TBI = total background Intake rate of pollutant from all other
sources of exposure (mg/day)
The RAC, 1n the case of carcinogens, 1s thought to be protective since the
q-j* Is typically an upper limit value, I.e., the true potency 1s consid-
ered ...unlikely to be greater and may be less. The definition and derivation
of ,each,of the parameters used, to estimate RAC for carcinogens are discussed
1n later sections. "-. ;
5.2.1.2.1. Human Cancer Potency (q,*) For most carcinogenic
chemicals, the linearized multistage model 1s recommended for estimating
human cancer potency from animal data (U.S. EPA, 1986a). When epldemlo-
loglc data are available, potency Is estimated based on the observed
relative risk 1n exposed vs. nonexposed Individuals, and on the magnitude of
exposure. Guidelines for use of these procedures have been presented 1n the
Federal Register (U.S. EPA, 1980a, 1985d) and 1n each of a series of Health
Assessment Documents prepared by OHEA; one example 1s U.S. EPA (1985b), The
true potency value 1s considered unlikely to be above the upper-boUnd
estimate of the slope of the dose-response curve 1n the low-dose range, and
1t 1s expressed 1n terms of r1sk-per-dose, where dose 1s in units of
mg/kg/day. Thus, q-j* has units of (mg/kg/day)1. the Office of
Health and Environmental Assessment has derived potency estimates for each
of. the potentially carcinogenic chemicals that are currently candidates for
sludge criteria. Therefore, no new effort 1s required to develop potency
estimates to derive sludge criteria.
5.2.1.2.2. Risk Level (RL) Since by definition no "safe" level
exists for exposure to nonthreshold agents, values of RAC are calculated to
reflect various levels of cancer risk. If RL 1s set at zero, then RAC will
5-17
-------
be zero. If RL 1s set at TO"6, RAC will be the concentration that, for
lifetime exposure, Is calculated to have an upper-bound cancer risk of one
case In one million Individuals exposed. This risk level refers to excess
cancer risk, that 1s, over and above the background cancer risk 1n unexposed
Individuals. By varying RL, RAC may be calculated for any level of risk In
the low-dose region, for Instance, RL <10~2. Specification of a given
risk level on which to base regulations 1s a matter of policy. Therefore,
1t 1s common practice to derive criteria representing several levels of risk
without specifying any level as "acceptable."
5.2.1.2.3. Human Body Weight (bw) and A1r Inhalation Rate (I )
a
Considerations for defining bw and I are similar to those stated 1n
a
Section 5.2,1.1.2. The Human Health Assessment Group assumes respective
values of 70 kg and 20 mVday to derive unit risk estimates for air, which
are potency estimates transformed to units of (yg/m3)"1. As Illus-
trated 1n Table 5-1, exposures may be higher In children than 1n adults when
the ratios of Inhalation volumes to body weights are compared. However,
exposure 1s lifelong, and therefore values of bw and I are usually chosen
a
to be representative of adults.
5.2.1.2.4. Relative Effectiveness of Exposure (RE) RE 1s a unit-
less factor that Indicates the relative toxlcologlc effectiveness of an
exposure by a given route when compared with another route. The value of RE
may reflect observed or estimated differences 1n absorption between the
Inhalation and 1ngest1on routes, which can then significantly Influence the
quantity of a chemical that reaches a particular target tissue, the length
of time 1t takes to get there, and the degree and duration of the effect.
The RE factor may also reflect differences In the occurrence of critical
toxlcologlc effects at the portal of entry. For example, carbon tetra-
chlorlde and chloroform were estimated to be 40 and 65% as effective,
5-18
-------
respectively, by Inhalation as by ingestion based on high-dose absorption
differences (U.S. EPA, 1984d,e). In addition to route differences, RE can
also reflect differences In bioavallabnity due to the exposure matrix. For
example, absorption of nickel Ingested 1n water has been estimated to be 5
times that of nickel ingested in food (U.S. EPA, 1985c). The presence of
food in the gastrointestinal tract may delay absorption and reduce the
availability of orally administered compounds, as demonstrated for 'halo-
carbons (NRC, 1986).
Physiologically based pharmacoklnetic (PB-PK) models have evolved into
particularly useful tools for predicting disposition differences due to
exposure route differences. Their use is predicated on the premise that an
effective (target-tissue) dose achieved by one route In-a particular species
is expected to be equally effective when achieved by another exposure route
or in some other species. For example, the proper measure of target-tissue
dose for a chemical with pharmacologlc activity would be the tissue concen-
tration divided by some measure of the receptor binding constant for that
chemical. Such models account for fundamental physiologic and biochemical
parameters such as blood flows, ventllatory parameters, metabolic capacities
and renal clearance, tailored by the physicochemlcal and biochemical prop-
erties of the agent 1n question. The behavior of a substance administered
by a different exposure route can be determined by adding equations that
describe the nature of the new Input function. Similarly, since known
physiologic parameters are used, different species (e.g., humans vs. test
species) can be modeled by replacing the appropriate constants. It should
be emphasized that PB-PK models must be used in conjunction with toxicity
and mechanistic studies in order to relate the effective dose associated
with a certain level of risk for the test species and conditions to other
5-19
-------
scenarios. A detailed approach for the application of PB-PK models for
derivation of the RE factor 1s beyond the scope of this document but the
reader 1s referred to the comprehensive discussion 1n NRC (1986). Other
useful discussions on considerations necessary when extrapolating route to
route are found In Pepelko and Wlthey (1985) and Clewell and Andersen (1985).
Since exposure for the air pathway Is by Inhalation, the RE factors
applied are all with respect to the Inhalation route. Therefore, the value
of RE 1n equation 5-4 gives the relative effectiveness of the exposure route
and matrix on which the q^ was based when compared with Inhalation of
contaminated air. Similarly, the RE factors 1n equation 5-3 show the
relative effectiveness, with respect to the Inhalation route, of each back-
ground exposure route and matrix.
An RE factor should only be applied where well documented/referenced
Information 1s available on the contaminant's observed relative effective-
ness or Its pharmacoklnetlcs. When such Information 1s not available, RE 1s
equal to 1.
5.2.1.2.5. Total Background Intake Rate of Pollutant (TBI) It 1s
Important to recognize that sources of exposure other than the sludge dis-
posal practice may exist, and that the total exposure should be maintained
below the determined cancer risk-specific exposure level. Other sources of
exposure Include background levels (whether natural or anthropogenic) 1n
drinking water, food or air. Other types of exposure, which are due to
occupation or habits such as smoking, might also be Included depending on
data availability and regulatory policy. These exposures are summed to
estimate TBI.
Data for estimating background exposure usually are derived from analyt-
ical surveys of surface, ground or tap water, from FDA market basket
5-20
-------
surveys, and from air monitoring surveys. These surveys may report means,
medians, percentHes or ranges, as well as detection limits. Estimates of
TBI may be based on values representing central tendency or on upper-bound
exposure situations, depending on regulatory policy. Data chosen to
estimate TBI should be consistent with the value of bw. Where background
data are reported 1n terms of a concentration 1n air or water, Ingestlon or
Inhalation rates applicable to adults can be used to estimate the proper
dally background Intake value. For certain compounds (for example, nickel)
that have been shown to be carcinogenic by the Inhalation route, but not by
the Ingestlon route, the TBI should not Include background exposure from the
Ingestlon route. Thus, 1n such a case only background exposures from other
air emission sources should be Included 1n the TBI.
As stated previously, the TBI 1s the summed estimate of all possible
background exposures, except exposures resulting from a sludge disposal
practice. To be more exact, the TBI should be a summed total of all toxlco-
loglcally effective Intakes from all nonsludge exposures. To determine the
effective TBI, background Intake values (BI) for each exposure route must be
divided by that route's particular relative effectiveness (RE) factor.
Thus, the TBI can be mathematically derived, after all the background
exposures have been determined, using the following equation:
BI(food) BI(water) BI(a1r)
RE(food) + RE(water) * RE(a1r)
BI(n)
RE(n)
(5-3)
where:
TBI = total background Intake rate of pollutant from all other
sources of exposure (mg/day)
BI = background Intake of pollutant from a given exposure
route, Indicated by subscript (mg/day)
RE = relative effectiveness, with respect to Inhalation
exposure, of the exposure route Indicated by subscript
(unHless)
5-21
-------
5.2.1.2.6. Fraction of Inhaled A1r from Contaminated Area -- It Is
recognized that an Individual exposed to air emissions.from an Incinerator
may not necessarily remain 1n the Incinerator prox1m1tyv for 24 hours/day.
However, 1f It 1s assumed that residential areas may be contaminated, It Is
likely that less mobile Individuals will Include - .those at greatest risk.
Therefore, 1t 1s reasonable .to assume that 100% pf the air, breathed by the,
most exposed Individuals will be from the .area of th.e Incinerator. ,; ,
5-22
-------
6. EXAMPLE CALCULATION
After modeling 1s completed, the highest annual ground level concen-
trations can be Identified at each of the model sites. By using the maxi-
mum allowable ambient concentrations for the pollutants of concern, calcu-
lated from the Reference A1r Concentration (RAC) as described 1n Chapter 5,
the air modeling results can be used to calculate the stack emission rates
for each pollutant. This Is conveniently done by simple ratios of the
ambient concentrations and emission rates, since the modeling results are
linearly related In the model.
The modeled emission rates for each plant may be compared with the
actual emissions determined by the testing of sludge Incinerators to deter-
mine whether the emissions are acceptable. If the RACs are being exceeded,
some type of criteria or management practice (or both) to reduce emissions
will need to be developed. This process 1s Illustrated 1n Figure 6-1.
6.1. STEP ONE
This example calculation 1s for a carcinogen 1n sludge being Incinerated
at Facility 7. The Facility 7 site Is located 3/4 mile from a river and 1n
an urban, Industrialized area.
Four Incinerator flue stacks (see Table 4-3} at the facility are taller
than the critical height where building aerodynamic downwash would be
expected to Increase near-source surface concentrations. The Incinerator
plant building 1s a complex structure with varying roof height.
Using the air dispersion model, the highest concentration for the facil-
ity (under actual operating conditions) Is located at the first receptor
ring (104 meters) and 1s estimated to be 12.5 yg/mVg/s. Under design
conditions, the highest concentration 1s estimated at 16.1 yg/m3.
6-1
-------
Facility
Design and
Operation
Calculate
Combustion
Parameters
Dispersion
Model
Control
Technology
Allowable
Contaminant
Emission Rate
Emission
Fraction
Allowable
Contaminant
Feed Rate
Sludge
Feed Rate
Allowable
Sludge
Concentration
Allowable
Ground Level
Concentration
(RAC)
Human Health
Threshold and
Nonthreshold
Toxicants
FIGURE 6-1
Criteria Derivation Approaches for Sludge Incineration
6-2
-------
6.2. STEP TWO
The RAC for the carcinogen chlordane 1s derived using equation 5-4:
RAC =
M
/
RL x bw
n* x R
- TBI
The risk level (RL), the body weight (by) and the dally Inhalation volume
(I )'-.-ate set for this example at . 10~«. 70 kg and 20 mVday, respec-
tively. The relative effectiveness factor (RE) 1s set at 1,. The human
cancer potency for chlordane has been determined by the U.S. EPA to be 1.61
(mg/kg/day)"1. Current total background Intake (TBI) of chlordane from
all other sources (I.e., except from Incineration of sludges) has not been
determined for 1986, but for Illustrative purposes TBIs of 0 and 20 ng/day
(20xlO~6 mg/day) are used here to derive example RACs. Determination of
an RAC for a specific Incinerator site should be based on a current local
assessment of TBI.
Example 1
(TBI = 0 mg/day)
RAC =
(
IP"6 x 70 kg
1.61 (mg/kg/day)"1 x
- 0
* 20 mVday
2.17 x 10~6 mg/m3
2.17 x 10~3 yg/m3
SC =
RAC x CFi x CF2
DP x FE x FR
6-3
-------
where:
RAC = Reference A1r Concentration (ng/m3)
CF, = conversion factor (sec/hr)
CFp = conversion factor (mg/g) .
DP = dispersion parameter [yg/m3 (g/s)"1] (operating conditions)
FE = fraction emitted (unltless)
FR = feed rate [kg/hr, dry weight (DW)]
SC = allowable sludge concentration [mg/kg, dry weight (DW)]
_ 2.17xlO"3 ug/m3 x 3600 sec/hr x IP3 mg/g
= 12.5 yg/m3 (g/s)"1 x *0.05 x **2660 kg/hr DW
=4.7 mg/kg DW
*The fraction emitted for this calculation assumes 95% efficiency.
**The feed rate for Facility 1 1s an assumption.
Example 2
(TBI = 20 ng/day)
RAC =
fc
IP"6 x 70 kg
61 (mg/kg/day)"1 x 1,
= 1.17x10~6 mg/m3
= 1.17xlO"3
- 20x10~6 mg/day
* 20 m3/day
SC
1.17xlO~3 ug/m3 x 3600 sec/hr x IP3 mg/g
12.5 yg/m3 (g/s)"1 x 0.05 x 2660 kg/hr DW
2.57 mg/kg DW
6-4
-------
7. REFERENCES
Anderson, G.E. and G.W. Lundberg. 1983. Us'er's manual for SHEAR-A computer
code for modeling human exposure and risk from multiple hazardous air pollu-
tants In selected regions. SYS APP-83/124, Systems Applications, Inc., San
Rafael, CA.
Anderson, G.Ev, C.S. Llu^ H.Y. HbTman'and' J.P. KIlTus. '1981: Human expo-
sure to atmospheric concentration of selected chemicals. EF-156R, Systems
Applications, Inc.,;San Rafael,' CAf;/ ; ;
- '' t :: ' - % " ' , 4 '-.',' '' -- ',. -
fistrand, P.-O. and K. Rodahl. 1977. Textbook of Work Physiology, 2nd ed.
McGraw-Hill, New York, NY. (Cited 1n Flserova-Bergerova, 1983}
Auer, A.H., Jr. 1978. Correlation of land use and 'cover with meteorologi-
cal anomalies. J. Appl. Meteor. 17: 636-643.
Bjorklund, J.R. and J.F. Bowers. 1979. User's Instructions for the SHORTZ
and LONGZ; computer programs. Technical Report TR-79-131 -01, H.E. Cramer
Company, Inc., Salt Lake City, UT.
Brlggs, G.A. 1969. Plume rise. Available as TIO-25075 from Clearinghouse
for Federal Scientific and Technical Information, Springfield, VA.
Brlggs, G.A. 1971. Some recent analysis of plume rise observations. Jji:
Proc. 2nd Int. Clean Air Congress, Academic Press, NY.
7-1
-------
Brlggs, G.A. 1972. Chimney plumes 1n neutral and stable surroundings.
Atmos. Environ. 6: 507-510.
Clewell, H.3. and H.E. Andersen. 1985. Risk assessment extrapolation and
physiological modeling. Toxlcol. Ind. Health. 1(4): 111-134. ;
D1em, K. and C. Lentner, Ed. 1970. Scientific Tables. C1ba-6e1gy, Ltd.,
Basle, Switzerland.
Parrel!, O.B. and B.V. Salotto. 1973; The effect of Incineration on
metals, pesticides and polychloMnated blphenyls In sewage sludge. Proc.
Natl. Symp. on Ultimate Disposal of Wastewaters and Their Residuals.
Sponsored by Research Triangle Universities. ;
Federal Register. 1974. Vol. 39, p. 4308. March 8.
Flserova-Bergerova, V., Ed. 1983. Modeling of Inhalation Exposure to
Vapors: Uptake, Distribution and Elimination. Vol. II. CRC Press, Inc.,
Boca Raton, FL. '
I
Fruhman, 6. 1964. Title not given. Z. Exp. Med. 138: 1. (Cited in D1em
and Lentner, 1970)
Furr, A.K., T.F. Parkinson, T. Wachs et al. 1979. Multielement analysis of
municipal sewage sludge ashes. Absorption of elements by cabbage grown 1n
sludge ash-soil mixture. Environ. Sc1. Techno!. 13: 1503.
7-2
-------
Holzworth, G.C. 1972. Mixing heights, wind speeds and potential for urban
air pollution throughout the contiguous U.S. Publ. No. AP-101. U.S. EPA,
Research Triangle Park, NC.
Huber, A.H. 1977. Incorporating building/terrain wake effects on stack
effluents. Preprint Volume for Joint Conf. on Application of Air Pollution
Meteorology, American Meteorological Society, Boston, MA.
Huber, A.H. and W.H. Snyder. 1976., Building wake effects on short stack
effluents. Preprint Volume for the 3rd Symp. on Atmospheric Diffusion and
A1r Quality, American Meteorological Society, Boston* MA.
ICRP (International Commission on Radiological Protection). 1975. Report
of the Task Group on Reference Man (No. 23). Pergamon Press, Ltd., Oxford.
Lu, F.C. 1983. Tox1colog1cal evaluations of carcinogenic and noncardno-
gens: Pros and cons of different approaches. -Reg. Toxlcol. Pharmac'ol. 3:
121-132.
NOAA (National Oceanic and Atmospheric Administration). 1978. Climates of
the U.S. Book Tower, Detroit, MI.
NRC (National Research Council). 1983.' Risk Assessment 1n the Federal
Government: Managing the Process. National Academy Press, Washington, DC.
NRC (National Research Council).1 1986. Dose-route extrapolations: Using
Inhalation toxlclty data to set drinking water limits. In.: Drinking Water
and Health, Volume 6. National Academy Press, Washington, DC.
7-3
-------
Pepelko, W.E. and J.R. Wlthey. 1985, Method* for route-to-ro,ute extrapftla-
tlon of dose. Toxicol. Ind. Health. 1(4): ;153-175. , ;,. - ,-, .-, ;,';
' ' ' - ' , ; - . ':-,:. :.- ':">'-. . "-.: .. . ;' v- ' -- - , .!/^-s
Turner, D.B. 1970. Workbooks of atmospheric dispersion estimates. PHS
Publ. No. 999-AP-26. U.S. DHEW National A1r Rolliftlon ^Control Adm1ri1straH
tlon, Cincinnati, OH. -. > ; --j'1- j^
U.S. EPA. 1979. Industrial Source Complex (ISC) Dispersion Model User's
Guide, Volumes I and II. Office of -Air. .Quality Planning and Standards,;
Research Triangle Park, NC. EPA 450/4^79-030 .and 450/4-79-031 . ; .
PB80-1 33044 and PB80-1 33051. ..... -.., , - - .' . --
U.S. EPA. 1980a. Water Quality Criteria Documents; Availability. Federal
Register. 45(231): 7931 8-79379. .-. ......,, ,- ....... ,--..:.... ....-.,.. , ,;
U.S. EPA. 1980b. HPTER (Multiple Point Algorithm with. ...Terrain Adj.ustme/ij;),
Model-Simulation. Environmental Sciences Research Laboratory, Meteorology
and Assessment , Division, Research. .Triangle Park,, NC. (Magnetic tape),
EPA/DF-80/004. NTIS PB80-168156. :..----». ' , S-T*,-? , :t'-: ' - : ; > li
U.S. EPA. 1982. User's Instructions for the SHORTZ and LONGZ Computer
Programs, Volumes I and II. .EPA/903/9-82/004A and B., , ; : , ,
U.S. EPA. 1983. User's Network, for Applied Modeling of A1r Pollution
(UNAMAP), Version 5 (computer programs on tape). NTIS PB 83-244368.
7-4
-------
U.S; EPA. 1984a. Project Summary: Analysis and Assessment of Incinerated
Municipal Sludge Ashes and Leachates. Municipal Environmental Research
Lab., Cincinnati, OH. EPA/600/S2-84/038. NTIS PB84-155563/REB.
U.S. EPA. 1984b. Guidelines on A1r Quality Models. Revised Draft. Office
of A1r Quality Planning and Standards, Research Triangle Park, NC.
EPA/450/2-78/027.
U.S. EPA. 1984c. Mercury Health Effects Update. Health Issue Assessment.
Office of Health and Environmental Assessment, Environmental Criteria and
Assessment Office, Research Triangle Park, NC. EPA/600/8-84/019F. NTIS PB
85-123925.
U.S. EPA. 1984d. Health Assessment Document for Carbon Tetrachlorlde.
Office of Health and Environmental Assessment, Environmental Criteria and
Assessment Office, Cincinnati, OH. EPA/600/8-82/001F. NTIS PB 85-124196.
U.S. EPA. 1984e. Health Assessment Document for Chloroform. Office of
Health and Environmental Assessment, Environmental Criteria and Assessment
Office, Research Triangle Park, NC. EPA/600/8-84/004F. NTIS PB 86-105004.
U.S. EPA. 1984f. Guidelines for Deriving Numerical Aquatic Site-Specific
Water Quality Criteria by Modifying National Criteria. Environmental
Research Lab., Duluth, MN. EPA/600/3-84/099. NTIS PB85-121-101.
U.S. EPA. 1985a. Seminar Publication: Municipal Wastewater Sludge Combus-
tion Technology. Center for Environmental Research Information, Cincinnati,
OH. EPA/625/4-85/015.
7-5
-------
IKS. EPA. 1985b. Health Assessment Document for Polychlorlnated Dlbenzo-
£-d1ox1ns. Office of Health and Environmental Assessment, Environmental
Criteria and Assessment Office, Cincinnati, OH. EPA/600/8-84/014F. NTIS PB
86-122546.
U.S. EPA. 1985c. Drinking Water Criteria Document for Nickel. Prepared by
the OfPico of Health and Environmental Assessment, Environmental Criteria
and Assessment Office, Cincinnati, OH for the Office of Drinking Water,
Washington, DC. EPA/600/X-84/193. NTIS PB-117801.
U.S. EPA. 1985d. National Primary Drinking Water Regulations; Synthetic
Organic Chemicals, Inorganic Chemicals and Microorganisms; Proposed Rule.
(40 CFR Part 141) Federal Register. 50(219): 46936-47022.
U.S. EPA. 1986a. Guidelines for Carcinogen Risk Assessment. Federal
Register. 51(185): 33992-34003.
U.S. EPA. 1986b. Guidelines for Health Risk Assessment of Chemical
Mixtures. Federal Register. 51(185): 34014-34025.
U.S. EPA. 1990. Integrated Risk Information System (IRIS). Online.
Office of Health and Environmental Assessment, Environmental Criteria and
Assessment Office, Cincinnati, OH.
7-6
-------
U.S. EPA. 1989. Development of Risk Assessment Methodology for Municipal
Sludge Landfllllng. Prepared by the Office of Health and Environmental
Assessment, Environmental Criteria and Assessment Office, Cincinnati, OH for
the Office of Water Regulations and Standards, Washington, DC.
EPA/600/6-90/008. NTIS PB91-100172/AS.
'"'.':-"';'»' - '
Wells, J.F..F.J. Crehwlng and .G.C. McRonald. 1979. Case histories of
waste activated sludge Incineration. J. Water Pollut. Control fed, 51:
2886.
-------
-------
-------
-------
-------
TJ O
n> £5
03 O
8
m
55
-------