United States
Environmental Protection
Agency
Office of
Research and Development
Washington, DC 20460
EPA/600/6-91/004
October 1990
Development of
Risk Assessment
Methodology for Municipal
Sludge Incineration

-------

-------
                               E3?V600/6-91/004
                               October 1990
DEVELOPMENT  OF RISK ASSESSMENT  METHODOLOGY
     FOR  MUNICIPAL  SLUDGE INCINERATION
   Environmental Criteria and Assessment Office
   Office of Health and Environmental Assessment
        Office of  Research and Development
              Cincinnati,  OH  45268
                                Printed on Recycled Paper

-------
                                  DISCLAIMER

    This  document  has  been  reviewed  1n  accordance with  U.S.  Environmental
Protection  Agency  policy  and approved  for  publication.   Mention  of  trade
names or  commercial  products does not constitute  endorsement  or  recommenda-
tion for use.
                                      11

-------
                                    PREFACE
    This  1s  one  of  a  series  of reports  that  present methodologies  for
assessing the  potential  risks to  humans  or other  organisms  from management
practices for  the  disposal  or  reuse of municipal sewage sludge.  The manage-
ment  practices  addressed by this  series  Include land application practices,
distribution  and  marketing programs,  Iandf1ll1ng,  Incineration and  ocean
disposal.   In   particular,  these  reports  deal  with  methods  for evaluating
potential health and environmental  risks  from  toxic chemicals  that  may be
present  1n  sludge.  This document  addresses risks  from chemicals associated
with municipal sludge Incineration.

    These  proposed  risk assessment   procedures  are  designed  as  tools  to
assist  1n  the  development  of regulations  for  sludge  management practices.
The procedures are structured  to allow calculation  of technical criteria for
sludge  disposal/reuse options  based on the  potential for adverse  health or
environmental  Impacts.   The criteria  may address management  practices (such
as site design  or  process control  specifications),  limits on  sludge disposal
rates or limits on toxic chemical concentrations 1n the sludge.

    The methods for criteria derivation presented 1n this report are Intended
to be used  by  the U.S.   EPA Office  of  Water  Regulations  and  Standards  (OWRS)
to develop  technical  criteria for  toxic  chemicals  1n sludge.   The  present
document  focuses  primarily  on  methods for  the  development  of  nationally
applicable criteria by OWRS.

    This  document was   externally peer  reviewed  and  completed  1n  1986.
Subsequent  to   further  review by   the U.S.  EPA Science  Advisory  Board,  a
revised draft  Incorporating review comments was produced 1n  1987.   Various
scientific and editorial changes, which clarify  but do not alter the  overall
thrust of the document,  have been made since that date.
                                     111

-------
                             DOCUMENT DEVELOPMENT
Document Authors and Contributors

L. Fradkln, Document Manager
A.M. Garabek, Co-Document Manager
W.B. Pelrano
Environmental Criteria and Assessment
  Office
Office of Health and Environmental
  Assessment
U.S. Environmental Protection Agency
Cincinnati, OH  45268

G. Grumpier
Wastewater Solids Criteria Branch
Office of Water Regulations and
  Standards
U.S. Environmental Protection Agency
Washington, DC  20460

T. Braverman
Source-Receptor Analysis Branch
Office of A1r Quality Planning and
  Standards                   .
Research Triangle Park, NC  27711

M. Dusetzlna
Office of A1r Quality Planning and
  Standards
Research Triangle Park, NC  27711

R. McCarthur and G.E. Anderson
Systems Applications Inc.
San Rafael, CA  94903
Editorial Reviewer

Judith A. Olsen
Environmental Criteria and Assessment
  Office
U.S. Environmental Protection Agency
Cincinnati, OH  45268
Document Preparation
External Reviewers

Mr. A. Baturay
Carlson Associates Technical
  Services, Inc.           ''  ;•
Rand Engineering            ,
West Redding, CT  06896

Mr. C.R. Brunner
Incinerator Systems
CH2M H111 Inc.
Reston, VA  22090

Mr. R. Dykes
Radian Corporation
Research Triangle Park, NC  27709

Dr. R.M. Gerstle, V.P.
PEI Associates
Cincinnati, OH  45246

Dr. D. Johnson
Department of Environmental Health
University of Cincinnati Medical
  Center
Cincinnati, OH  45267

Mr. W. Nlessen
Camp Dresser and McKee, Inc.
Boston, MA  01810

Dr. C.V. Pearson
Argonne National Laboratory
Argonne, IL  60439

Dr. M. Radlke
Department of Environmental Health
University of Cincinnati Medical
  Center
Cincinnati, OH  45267

Dr. P.A. Vesiland
Department of Civil Engineering
Duke University
Durham, NC  27706
Bette  L.  Zwayer,  Patricia  A.  Daunt  and  Jacqueline  Bohanon,  Environmental
Criteria and Assessment Office, Cincinnati
                                      1v

-------
                              TABLE OF CONTENTS
1.  INTRODUCTION AND DESCRIPTION OF GENERAL METHODOLOGIC APPROACH
Page

•1-1
    1.1.   PURPOSE AND SCOPE.	   1-1
    1.2.   DEFINITION AND COMPONENTS OF RISK ASSESSMENT  .......   1-2
    1.3.   RISK ASSESSMENT IN THE METHODOLOGY DEVELOPMENT  PROCESS  .  .   1-3

           1.3.1.   Exposure Assessment ...............   1-3
           1.3.2.   Hazard Identification and Dose-Response
                    Assessment.,  .,.	 .-,  ......  .,.,,  .,,.   1-7
 !          1.3.3.   Risk Characterization 	  .......   1-8

    1.4.   POTENTIAL USES OF THE METHODOLOGY IN RISK MANAGEMENT  ...   1-10
    1.5.   LIMITATIONS OF THE METHODOLOGY ....."	  k  .   1-11

2.  DEFINITION OF DISPOSAL PRACTICE .......  	  ...   2-1

    2.1.   INCINERATION ..... 	  .......   2-4

           2.1.1.   Uniform Feed	   2-4
           2.1.2.   Burning Methods .........  .  .  .  .  .  .  .  .   2-6
           2.1.3.   Energy Recovery ............  .  ,  ...   2-14
           2.1.4.   Instrumentation and Control ........  ...   2-14

    2.2.   AIR POLLUTION CONTROL	   2-]5

3.  IDENTIFICATION OF KEY PATHWAYS.	   3-1

    3.1.   AIR EMISSIONS. ..	  .......     3-1
    3.2.   ASH RESIDUE AND SCRUBBER WATER	 ...  .  .  .  .......  , .  3-3

4,  METHODOLOGY FOR AIR EMISSIONS PATHWAY	  .  .  .  ...  .   4-1

    4.1.   OVERVIEW OF METHOD	   4-1
    4.2.   FATE AND TRANSPORT	   4-2

           4.2.1.   POTW Incinerator Data Base	   4-2
           4.2.2.   Model Plant Selection Criteria	   4-6
           4.2.3.   Facilities Selected ...........  .''.,.  .   4-10
           4.2.4.   Modeling Long-Term Average Concentration
                    Patterns. »	  ... .   .  ....  .  ..  .  .  , c,4-12

5?  EXPOSURE AND ASSESSMENT OF HEALTH EFFECTS ............   5-1

    5.1.   HUMAN EXPOSURE MODEL (HEM)  .... 	  .....   5-1

           5.1.1.   Exposure	   5-1
           5,1.2.   Inhalation Volume . 	  .....   5-2
    '       5.1,3.   Particle Size Distribution.  .   . .   .  .  ...  .  .  .  ..  5-2
           5.1.4.   Contaminant Concentration .  .   . .   .  .  ......   5-3
           5.1.5.   Model  Description  	   5-3
           5.1.6.   HEM Estimation Scheme                               5-3

-------
                          TABLE OF CONTENTS (cont.)


    5.2.   HOST EXPOSED INDIVIDUAL METHODOLOGY	  5-6

           5.2.1.   Reference A1r Concentration Derivation.  .  ... .  .  .  5-6

6.  EXAMPLE CALCULATION	. ,	  6-1

    6.1.   STEP ONE	  6^1
    6.2.   STEP TWO	.6-3

7.  REFERENCES	  7-1
                                     v1

-------
LIST OF TABLES
                                        Page
No.                                Title
2-1     Distribution of Sludge Combustion Facilities by
        State and Type/;. >.  . .  . %  .  .  . .  .  .  .  .  ;  .  .  ..•'.  .  .  .  .   2-2
2-2     Operational Status of Various  Types  of Installations.  .  ...   2-3
2-3     Distribution of Sludge Combustion Systems  by Plant Size  .  .  .   2-5
3-1     EP Toxlclty Testing of Sludge  Incinerator  Ash	  .  .   3-5
4-1     Major Assumptions for the Sludge Incineration Methodology  .  .   4-3
4-2     A1r Pollution Control Systems  by Furnace Type  ........   4-9
4-3     Facility Coordinates, Stack Parameters and Building
        Dimensions of the Sewage Sludge Incinerator Plants	4-11
4-4     Options Used 1n the ISCLT Modeling	4-14
4-5     Options Used 1n the LON6Z Modeling	4-17
4-6     Options Used 1n the COMPLEX I Modeling	   4-19
4-7     Ring Distances According to Facility Modeled. .	4-22
4-8     Meteorologlc Input Data for ISCLT and LONGZ Models	4-25
4-9     Mixing Height and Ambient Temperature Data According to
        Stability and Facility	4_26
5-1     Dally Respiratory Volumes for "Reference" Individuals
        (Normal Individuals at Typical Activity Levels) and for
        Adults with Hlgher-than-Normal Respiratory Volume or
        Hlgher-than-Normal Activity Levels	5-9
5-2     Illustrative Categorization of Evidence Based on Animal
        and Human Data	5_16

-------
                               LIST OF  FIGURES
No.                                Title                               Page
1-1     Relationship of Risk Assessment Methodology to Other
        Components of Regulation Development for Sewage Sludge
        Reuse/Disposal Options	T-4
2-1     Cross-Section of a Fluldlzed-Bed Furnace	2-7 s
2-2     Cross-Section of a Typical Multiple-Hearth Incinerator. . .  .  2-10
2-3     Process Flow Diagram Infrared Incineration System 	  2-12
2-4     Cross-Sectional View of a Ventur1/Imp1ngement-Tray
        Scrubber	  2-17
3-1     Sludge Incineration Pathways		  3-2
4-1     Distribution of Capacity for 127 Sludge Incinerator Sites .  .  4-5
4-2     Distribution of Stack  Heights for Sludge Incinerators  ....  4-7
4-3     Distribution of Stack  Exit Gas Velocities for Sludge
        Incinerators	4-8
5-1     Reference Points for an Enumeration District/Block Group
        (ED/B6) Centrold	5-5
6-1     Criteria Derivation Approaches for Sludge Incineration. . .  .  6-2
                                    V111

-------
                             LIST OF ABBREVIATIONS
 ADI
 ARC
 BI
 BOD
 Btu
 bw
 CF
 COD
 DP
 DM
 ED/BG
 EP
 ESP
 FB
 FE
 FR
 GPM
 HEM
 la
 ICRP
 JP
 ISCLT
MEI
MEIM
MHF
 Acceptable dally Intake (mg/kg)
 Air pollution control
 Background Intake (mg/day)
 Biochemical  oxygen demand
 British thermal  unit
 Body weight
 Conversion factor
 Chemical  oxygen  demand
 Dispersion parameter
 Dry weight
 Enumeration  district/block group
 Extraction procedure
 Electrostatic predpUator
 Flu1d1zed  bed
 Fraction emitted  (unltless)
 Feed  rate  (kg/hr)
 Gallons per minute
 Human exposure model
 A1r  Inhalation rate (mVday)
 International Commission on Radiological Protection
 Acceptable pollutant Intake rate (mg/day)
 Industrial Source Complex Long-Term'
Most-exposed Individual
Most-exposed Individual  methodology
Multiple hearth furnace
                                      1x

-------
                         LIST OF ABBREVIATIONS (cont.)
NAAQS
NESHAPS

NSPS
NSR
PAH
PB-PK
PCB
P-G
POTW
PSD
q-,*
RAC
RCRA
RDF
RE
RfD
RL
RHCL
SC
STAR
TBI
TSP
H.G.
National Ambient Air Quality.Standards
National Emission Standards for Hazardous A1r
Pollution
       1    ' f -      '    .     •..'*•     '" '
New Source Performance Standards
New Source Review
 _t i       '      •    ,'  • '         ' -   .-..,<-••
Polynuclear aromatic hydrocarbons
Physiologically based pharmoklnetlc
PolychloMnated blphenyls
      :   •.  ..-.  '•,-:..••      .     - •   t   ':   • .• ••
Pa'squl 11-61 f ford'"
Publicly owned treatment works
Prevention of significant Deterioration
Human cancer potency [(mg/kg/day)"1]
Reference air concentration  (mg/m3)
Resource Conservation and Recovery Act
Refuse-derived fuel
Relative effectiveness (unltless)
Reference dose for  Inhalation  (mg/kg/day)
  ,  " .    .   i      '     •   '    .   ' , i -:      : '   '
Risk level (unltless)
Recommended maximum contaminant  level
Allowable sludge  concentration  (mg/kg)
STabUHy ARray
Total background  Intake  (mg/day)
Total suspended partlculates
Wet gas

-------
        1.   INTRODUCTION AND  DESCRIPTION OF GENERAL METHODOLOGIC APPROACH
 1.1.    PURPOSE AND SCOPE
     This  1s   one  of  a  series  of  reports  that  present  methodologies  for
 assessing the potential  risks  to humans or  other  organisms  from  management
 practices  for  the  disposal  or  reuse  of  municipal  sewage  sludge.    The
 management practices  addressed  by  this  series  Include  land application
 practices, distribution  and  marketing  programs,  landfllllng,  Incineration
 and  ocean  disposal.   In  particular, these  reports  deal  with  methods  for
 evaluating potential  health and  environmental  risks from  toxic   chemicals
 that may be present  1n sludge.  This document addresses risks  from chemicals
 associated with  sludge Incineration practices.
    These proposed  risk  assessment   procedures  are  designed  as   tools  to
 assist  In  the development  of  regulations  for  sludge management  practices.
 The procedures are structured  to  allow calculation of technical criteria  for
 sludge  disposal/reuse  options  based  on  the potential for  adverse  health  or
 environmental  Impacts.  The criteria may address  management practices (such
 as site  design or  process  control  specifications),  limits  on sludge disposal
 rates or  limits on toxic chemical concentrations 1n the sludge.
    The  methods   for  criteria  derivation  presented  1n  this  report  are
 Intended  to be used by  the U.S.  EPA  Office  of Water Regulations  and Stan-
 dards  (OWRS)  to develop  technical  criteria  for  toxic chemicals 1n sludge.
 The present document focuses primarily on methods  to  be  used by OWRS for the
 development  of  nationally  applicable  criteria.   It  1s   suggested that  a
 user-oriented  manual  based on  these  methods  be  developed  for wider  use 1n
 deriving  site-specific  criteria   for  these   sludge  management  practices.
Additional uses  for  the methodology may  exist,  such  as developing guidance
 for local  authorities'  selection  of  sludge  management  options,   but  these
uses  are not the focus  of these documents  and  will  not be  discussed.

                                     1-1

-------
    These documents do  not  address health risks resulting  from  the presence
of  pathogenic  organisms  1n  sludge.  The  U.S.  EPA  will examine  pathogenic
risks  In  a  separate  risk assessment  effort.   These  documents  also  do  not
address potential  risks associated with  the treatment, handling  or  storage ;
of  sludge;  transportation  to the  point  of reuse or disposal;  or  accidental
release.
1.2.  ,:DEFINITION AND COMPONENTS OF RISK ASSESSMENT
    The  National  Research  Council (NRC,  1983) defines  risk assessment  as
"the  characterization  of  the  potential  adverse  health   effects  of  human
exposures to environmental hazards."  In  this document, the NRC's.. definition;
1s  expanded  to  Include  effects  of exposures  of  other  organisms  as well.   By
contrast,  risk   management   Is  defined  as   "the  process   of  evaluating
alternative   regulatory  actions   and   selecting  among    them,"   through
consideration of costs,  available technology and other nonrlsk factors,    ,•-,..
    The  NRC further  defines  four components   of  risk assessment.   Hazard
Identification 1s defined as  "the  process  of.determining whether exposure to
an  agent  can cause  an  Increase  1n the  Incidence  of   a  health  condition."
Dose-response  assessment  1s "the process  of  characterizing  the  relation
between  the  dose of  an agent  ...  and the Incidence of (the) adverse health
effect...."  Exposure assessment  1s "the  process of measuring or  estimating
the  Intensity, frequency and  duration of  ...  exposures to an agent currently
present  or  of  estimating hypothetical exposures  that  might arise	"  Risk
characterization  1s   "performed by  combining the exposure  and dose-response
assessments" to  estimate  the likelihood of an  effect  (NRC, 1983).  The U.S.
EPA  has  broadened the definitions  of  hazard  Identification and dose-response
assessment  to  Include  the  nature and   severity  of  the  toxic   effect .In-.
addition to the  Incidence.
                                      1-2

-------
    Figure T-l shows how  these  components  are Included In the development of
these, risk  assessment methodologies  for  sludge  management practice's".   The
figure further  shows  how each  methodology may be used  to  develop technical
criteria, and how  these criteria  could  be  used  or modified  by  the  risk
manager to develop regulations and permits.
1.3.   RISK ASSESSMENT IN THE METHODOLOGY DEVELOPMENT PROCESS
    As Illustrated 1n  Figure  1-1,  the  methodology development process begins
by  defining   the  management  practice.   Even within  a  given  reuse/disposal
option,  "real world"  practices  are  highly  variable,  and  so  a  tractable
definition must  be  given  as  a  starting  point.   As a  general  rule,  this
definition should Include the types  of  practices  most frequently used.   That
1s, the definition should not be limited  to  Ideal  engineering practice, but
also'need not Include  practices  judged  to  be  poor or  substandard (unless the
latter are  widespread).   This  definition, presented in Chapter  2  of  this
document, helps to determine  the limits of applicability of  the methodology
arid the exposure pathways that  may  be  of concern.  However, as also shown 1ri
Figure  1-1   and  as  discussed  1n  Section 1.4.,  this  definition  could  be
modified as  the methodology  Is  applied,  since  the  methodology  Itself  will
help to define acceptable practice.
1.3.1.   Exposure Assessment.  The  exposure  assessment step begins  with the
Identification  of  pathways  of  potential  exposure.   Exposure  pathways  are
migration routes of chemicals  from (or within) the disposal/reuse  site  to a
target organism.  For  those pathways where humans are the target of concern,
special  consideration  1s  given  to   Individual  attributes  that  Influence
exposure potential.   Individuals differ  widely  in  consumption  and  contact
patterns relative to  contaminated  media and,  therefore, also  vary  widely in
their degree of exposure.                                 .
                                      1-3

-------
III
E
to
CO
LU
CO
CO
CO
f£
2
                                                                                                               o>
                                                                                                               O)
                                                                                                               >
                                                                                                               0)
                                                                                                               O)
                                                                                                               O)
                                                                                                               o>
                                                                                                         I—    °
                                                                                                              o>
                                                                                                              OJ
0)
o>
                                                                                                                 OD

                                                                                                                 CO
                                                                                                              O)
                                                                                                             Di
                                                             1-4

-------
                                         o
                                         u
                                         =3
                                         CD
1-5

-------
     An Ideal way to  assess  human exposure Is .to define the  full  spectrum  of
 potential  levels of  exposure and  the number of  Individuals at each  level,
 thus  quantifying  the  exposure  distribution  profile  for a  given  exposure
 pathway.   The methodologies  described In these reports will  not attempt  to
 define exposure  distributions  1n  most  cases,   for the  following  reasons.
 First, 1t  1s very difficult to estimate  the total distribution  of exposures,
 since  to  do  so  requires  knowledge  of   the  distributions  of  each  of the
 numerous  parameters  Involved  1n  the  exposure  calculations, and requires the
 modeling  of  actual  or  hypothetical  population  distributions and  .habits  In
 the  vicinity  of disposal  sites.   Such   a  task  exceeds   the  scope  of the
 present methodology development effort.
     Second,  while knowledge  of the  total  exposure distribution may be  useful
 for  certain  types  of decision-making,  1t Is  not  necessarily  required for
 establishing criteria  to  protect  human health  and  the environment.    If
 criteria  are  set   so as  to   be  reasonably protective  of all  Individuals,
 Including those at  greatest  risk,- then as long as the risk assessment  proce-
 dures  can reasonably  estimate the risk to these  Individuals,  the quantifica-
 tion of lesser risks  experienced by other Individuals 1s not required.
    The drawback, however, of examining only  a maximal-exposure situation  1s
 that the  true Hkellhopd of  such a situation occurring may  be  quite  small.
The  compounding  of worst-case assumptions may  lead to Improbable  results.
Therefore,  the key to  effective use  of  this methodology 1s a  careful  and
 systematic  examination  of the  effects of varying  each  of the  Input param-
eters,  using  estimates of central tendency and upper-limit values  to gain an
appreciation for the variability of the result.
                                      1-6

-------
     Therefore,  exposure will be  determined  for a "most-exposed  Individual,"
 or.MEI.*   The definition of the MEI -win-vary with each human exposure path-
 way.   Chapter 3  of  this document  will  enumerate the  exposure  pathways and
 will  define  the  MEI  1n qualitative  terms;  for  example,  for  the Inhalation
 pathway,  the MEI  1s  a  person  residing  1n  the vicinity of the  Incinerator.
 The  MEI will  not be  quantitatively defined  1n  this  chapter,  but  relevant
 Information  that allows the  user  to  do so  (such  as available  data  on the
 range  of  ventilation  rates)  will  be  provided 1n  later  chapters.   For
 exposure pathways  concerning organisms other than humans,  the  term  "MEI" Is
 not  applied,  but conservative  assumptions  are  still made  regarding  the
 degree  of  exposure.   The remaining chapters  (Chapters  4-6  1n  this document)
 explain  the  calculation  methods  and  data  requirements  for conducting  the
 risk assessments for each pathway.
 1.3.2.   Hazard  Identification  and  Dose-Response Assessment.    To determine
 the allowable exposure  level  for  a  given contaminant, the  hazard Identifica-
 tion  and  dose-response  assessment  steps  must be  carried  out.  For  human
health effects,  these  procedures  already are fairly  well  established  1n  the
Agency  (although they  still  require  Improvement,  and specific  assessments
for many chemicals  remain  problematic).   Hazard  Identification  1n this  case
consists first  of all  of  determining whether  or not  a  chemical should  be
treated as a  human carcinogen.  Procedures for  weighing evidence of  cardno-
genlclty  have  been   published  1n  the  U.S.   EPA (1986a)  and  are   further
*The definition of  the MEI does  not  Include workers exposed 1n  the  produc-
 tion/ treatment,  handling or transportation of  sludge.  This methodology  1s
 geared toward the protection of  the general public  and  the  environment.   It
 Is assumed that workers can  be  required to use special measures  or  equipment
 to  minimize  their  exposure  to  sludgeborne  contaminants.    Agricultural
 workers,  however, might  best be considered members  of  the general  public,
 since the use of  sludge may  not be  Integral  to their occupation.
                                     1-7

-------
discussed  1n  later sections  of  this document.  If  treated  as carcinogenic,
dose-response  assessment would  then consist  of  the  use  of Agency-accepted
potency  values.    If  none are  available,  cancer risk  estimation procedures
published by the Agency  (U.S. EPA, 1986a) would be used to determine potency.
    If  not  carcinogenic, hazard  Identification  and  dose-response assessment
normally  consist  of Identifying  the critical  systemic effect, which  1s the
adverse  effect  occurring at  the  lowest dose,  and the reference  dose (RfD),
which  1s "the dally  exposure ... that  1s  likely to  be  without  appreciable
risk  of deleterious effects  during  a  lifetime"  (U.S. EPA,  1990).   Further
description  and  procedures   for  deriving  RfDs  are  published  1n  U.S.  EPA
(1990).
    For  certain disposal  options,  effects  on other  organisms are  of concern.
                                  1   -i .' t   '-"•'''•,'':- i  '   "'''..  '    "   " •   ''''•*
In  these cases, existing Agency methodologies  have  been used where  avail-
able.   For  example, existing guidelines for  deriving  ambient  water  quality
criteria  (U.S.  EPA,  1984f)   are  used  to determine  levels  for aquatic  life
protection.  Where  effects  on terrestrial species are  of  concern,  there are
no  existing  Agency guidelines,  but  suggested  procedures   for  identifying
adverse  effects  (hazard   identification) and threshold levels (dose-response
assessment) are provided.
1.3.3.   Risk  Characterization.   Risk  characterization consists  of  combin-
ing the  exposure and dose-response assessment  procedures  to derive  criteria.
Risk  assessments  ordinarily proceed  from  source  to  receptor.  That  1s,' the
source,  or  disposal/reuse practice,  1s  first  characterized and  contaminant
movement  away  from the   source  Is   then  modeled  to estimate the degree  of
exposure  to  the   receptor,   or  MEI.   Health  effects  for  humans  or  other
organisms are  then  predicted  based  on the estimated exposure.  The  calcula-
tion of  criteria,  however,  Involves  a reversal  of  this process.   That is,  an
                                      1-8

-------
 allowable exposure,  or  an  exposure that  1s  not  necessarily allowable  but
 corresponds  to  a given  level  of risk,  1s  defined  based  on health  effects
 data,  as  specified  above.   Based  on  this  exposure  level,  the  transport
 calculations  are  either  operated  1n  reverse  or performed  Heratlvely  to
 determine the corresponding  source  definition.   In this case, the  resulting
 source  definition   1s  a  combination  of  management  practices   and  sludge
 characteristics,  which  together  constitute the  criteria.    These  steps  are
 carried  out  on  a chem1cal-by-chem1cal  basis, and criteria values are  derived
 for  each chemical  assessed and  each exposure pathway.  An example Illustrat-
 ing  how these calculations  may be carried  out  1s  provided  1n this  document
 for  each pathway assessed.  However, as  Indicated  by Figure 1-1, the  compi-
 lation of  data  on specific chemicals to be used as Inputs to  the methodology
 Is a process  separate from methodology  development.  Health effects data  for
 Individual  chemicals must be collected from  the  scientific  literature.  In
 many cases,  the  U.S.  EPA has  already  published approved values  for cancer
 potency  or  RfD.    Data   pertinent   to  a  chemical's   fate  and  transport
 characteristics,  such  as  solubility,  partition coefficient,  bloconcentratlon
 factor   or  environmental   half-life,  must  also  be   selected   from  the
 literature.  In  some cases,  data for  particular  health or  fate  parameters
were gathered  for a  variety of  chemicals  1n  the process of  developing the
methodology.   Where  this  was   done,   the  Information  may  appear  as  an
appendix.  In most  cases, however,  such  Information  does not appear  1n the
methodology document and must be gathered  as a  separate  effort.
    Once these data  have  been selected, even on  a  preliminary basis,  1t may
be useful  to carry  out  a rough  screening exercise, using  these data  plus
Information on occurrence  1n sludges,  to  set priorities  for  risk  character-
ization.    Screening  could reveal that certain  pollutants  are  unlikely  to
                                     1-9

-------
pose any risk, or  that  existing  data  gaps  preclude more detailed characteri-
zation of  risk.   Methods  for  carrying out  such a screening  procedure  will
not be discussed 1n this document.
    Following  chemical-specific   data  selection,  risk  characterization  or
criteria  derivation may  be  conducted.   The  values  derived as  limits  oh
sludge concentration or disposal  rate,  together  with  the management  practice
definitions, will  constitute the  criteria.   When calculating the  numerical
limits,  1t  1s  advisable   to  vary  each of  the  Input  values  used  over  ViV
typical or plausible range  to determine the  sensitivity of  the result to the
value  selected.  Sensitivity  analysis helps  to give a  more  complete  picture
of the potential variability surrounding the result.
1.4.   POTENTIAL USES OF THE METHODOLOGY IN RISK MANAGEMENT
    The results of  the  risk  characterization  step  can  then  be used  as Inputs
for  the  risk  management  process,   as shown  1n  Part  II  of  Figure  1-1.
Although this  document  does not  specify  how risk management  should  be  con-
ducted, some  potential  further  uses  of the  methodology  1n  the  risk  manage-
ment process are briefly  described here.   These optional steps  are  shown as
dashed lines 1n Figure 1-1.
    As suggested by NRC  (1983),  a risk manager may evalute the feasibility
of  a  set  of  criteria  values  based  on   consideration  of   costs,  available
technology and  other  nonrlsk factors.  If 1t 1s  felt  that  certain  chemical
concentrations  specified   by the calculations  would  be  too  difficult  or
costly to  achieve,  the management  practice  definition could  be modified by
Imposing  controls  or  restrictions.  For   example,   reo;u1rement   of  more
stringent  pollution  control  technology  devices  could  result  1n  higher
permissible concentrations  for  some  pollutants.  The same degree of  protec-
tion would still be achieved.
                                     1-10

-------
     Following  promulgation  of  the  criteria,  it  may  also  be  possible to
evaluate  sludge  reuse or disposal practices  on  a site-specific basis, using
locally  applicable data to  rerun  the  criteria calculations.  Criteria could
then  be  varied  to reflect  local  conditions.  Thus, the  methodology can be
used as a tool for the  risk  manager to develop and  fine-tune the criteria.
1.5.   LIMITATIONS OF THE METHODOLOGY
    Limitations  of the  calculation methods  for each pathway are given 1n the
text  and  in  tabular  form  in  the  chapters   where calculation  methods  are
presented.  However,  certain  limitations  common to  all  of the  methods  are
stated here.  .
    Municipal  sludges   are  highly  variable   mixtures   of  residuals   and
by-products  of  the  wastewater  treatment  process.   Chemical  Interactions
could affect  the fate,  transport and toxidty of individual  components,  and
risk from  the whole  mixture may be greater than that from any single compo-
nent.  At  present, these  methodologies  treat each chemical as  though  each
acts in  isolation from all  the others.   It should be noted that  U.S. EPA's
mixture  risk  assessment guidelines  U.S. EPA (1986b) caution  that  a great
deal of dose-response information  is required before  a  risk assessment could
be  quantitatively  modified  to  account . for  toxic  interactions;.   Future
revisions to .these documents  to include consideration of  interactions  will
most likely be limited to qualitative discussion  of such interactions.
    Transformation  of  chemicals  occurring  during  the  disposal  practice
(Including  combustion)   or   following  release  may  result  in  exposure  to
chemicals other  than  those originally found  in  the  sludge.  In  many cases,
these assessment procedures may  not adequately characterize risks  from these
transformation products.
                                     1-11

-------
    In  addition,  these methodologies  compartmentalize  risks  according  to
separate  exposure  pathways.  The  use of an  MEI approach, which  focuses  on
the most  highly  exposed  Individuals  for  each  pathway,  reduces the likelihood
that any  single Individual  would  simultaneously receive  such  exposures  via
more than one pathway, and  therefore the addition of doses  or  risks across
pathways  1s   not  usually  recommended.   However,  1t  1s  possible that  risk
could be underestimated 1n a small  number of Instances.
    Finally,   the  methodologies look at   exposed  organisms  1n  Isolation.
Population-level or  ecosystem-level   effects  that  could  result from  a  reuse
or disposal practice might not be predictable by this  approach.
                                     1-12

-------
                      2.  DEFINITION OF DISPOSAL PRACTICE
    Sludge  Incineration Involves  the combustion  of sludge within  a mechan-
ically  controlled  environment.   Incineration  Is  a  treatment method  used
before  disposal  or  reuse  of  the   ash.   Because   Incineration  drastically
reduces the  volume and mass  of  sludge,  1t has  been traditionally considered
a disposal  method  similar  to landfUUng  and  ocean dumping.   Land appllca-
 .,...;•;  r.:....   •. .   ',,•••;•:•.-;•  :•••-..,,,.-.:»   •...  ;••-.}•   ....'"/•.• ^ ,<•••;.••-..  ..,-;".  „ :.. >~ ~'..•; t
tlons  (Including  distribution and  marketing of  sludge)  are  considered use
 •.->••  •-   ,  •  • '.'•:•'•••- '  ,'    ;''s-,-  " '  ; .V ;  '".,••>'•••-•• ••'•:P -'.--.'•   ^  ',•'•<•   !--r, ••-.'•  -"•
options, since  1n  these cases  beneficial  properties  of  sludge are utilized.
    Currently,  only three  types  of  Incineration  technology are employed for
dedicated sludge combustion 1n the United States:

    1.  Multiple-hearth furnace  (MHF);
    2.  Flu1d1zed-bed (FB)  furnace;  and
                                            •   t
    3.  Infrared electric  furnace.
    In addition,  sludge has been colndnerated  with municipal  solid waste 1n
mass  burning Incinerators.   However,  for the  purposes  of developing stan-
dards  and  management practices,  this methodology will not  consider colncln-
eratlon  of  sludge  with  other  wastes,  but  will  concentrate on  dedicated
sludge Incineration that uses  fossil fuels for  auxiliary firing.
    Table 2-1  presents  the  distribution  of sludge  combustion  types  by state
and  Indicates  the  number   of  operational  facilities.   The   facilities  are
located primarily  on the  east coast  and  In the Midwest.   As can be seen from
Table  2-1,  most of  the  plants are  In nine states and MHFs are the dominant
type (73%).  FB  furnaces represent 20% of the total.
    Table 2-2  shows that  58% of the Incineration facilities are operational.
Of the 156 operating units,  120  or 80% are MHF  and 25 or  16% are FB furnaces.
                                      2-1

-------

























7
CM

IjJ
jo
*~



















































a.
f
re
01
re
CO
4?
«
i-1
u
re

c
0
(A
3
U
at
en


CO
o
c
o
^
1
1.
I/I
























ei

•r- 3
4-1 r—
IA CO
f= i

a
u u-»
0 OJ
u ce.

c
£
>
44
0
o:

o>
^
re
a>
a.
o


^
re
0
H-
4-1
re
e>
o


re
o


_
•o
re
u
u-
c
u
u.
u
e>

01
c
re
a>


re

o
UJ I—




'Q
Gl
CD
1
TJ
•a
3
U.

en
c
*»™
re

m
a.
o


re
4->



5
i_
re
ei

i
0)
"a
3
K

01
n
re

ai
a.
o


*re

o
I—








a>
re
4-1
CO


i— i—








r— O




- " -




i— CM r- . O




CM CM » 1— r—






^ CM r~ ^ ^ CM r~ r~ O r~





i
CM VO «• i— r— CM I— i— CM CM

'

I— OinOOOr-^-r— r-i— CMCMOCOCM9OCM







CMi— Ol— COCM^J-CO^-CMCMCMCMCOinenOCM
1— CM






tA
44

re 3 a>
•r- u : re en re
IACOlC'O3C44
rev)oreu
iAre-^-oci-t.re"--<-»->>vi.cc
re^r— r-COOSr- t3SCC3l_IA(JC
^ treoo^"0)re^~corefliore.re»^*™
«OCJULL-CPrct-»— r— CMi— o
•— CM « — CX






CM CMCOr-OCMCMCMr— P^-COOJ^'r— Or— CMlTJi — tfi
r— CO r— r— r— Ol
r—




re re
a> c c re
w *- re ts «i— *-
*- r— »- C r— CO
.c >> o creo c-r-u
via> i> re r— u a> oo>c-r*
*r-re o.vtj^re re >vireo) re44urerea>o oc>> v> c c > c o
ore^3z>~3>>.c j=ov>a>£a)iAoreo>cT>44cre o>.c 44 u i- «r
vi^is*Si-'r-i— a>co3cxi_viviviai H-
a)a>a)Of^!ua>x:oa>a>T-rea)-<-3 o
s: z z z z z z o o o a. ec to i— i— =» 3 3 3 a- I—




















































re
m
CO
en

-------
                                   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















I—I
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


-------