United States
          Environmental Protection
          Agency
Office of
Toxic Substances
Washington, DC 20460
EPA 560/5-88-002
May, 1988
          Toxic Substances
&EPA  Assessing Asbestos
          Exposure In Public Buildings


                                           •T.


-------
                                               May,  1988
ASSESSING ASBESTOS EXPOSURE IN PUBLIC BUILDINGS


                  Prepared by:
           Battelle Columbus Division
             Washington Operations
              2030 M Street, N.W.
            Washington, D.C.  20036
          EPA Contract No. 68-02-4294
                Price Associates
              1825 K Street, N.W.
            Washington, D.C.  20006
          EPA Contract No. 68-02-4294
       Alliance Technologies Corporation
              213 Burlington Road
         Bedford, Massachusetts  07130
   EPA Contract                   -6«-e2
           R.   J.   Lee  Group,   Inc.
               350 Hochberg Road
        Monroeville,  Pennsylvania  15146
          EPA Contract No. 68-03-3406
           Midwest Research Institute
              425 Volker Boulevard
          Kansas City, Missouri  64110
          EPA Contract No. 68-02-4252
                    for the:
          Exposure Evaluation Division
           Office of Toxic Substances
   Office of Pesticides and Toxic  Substances
      U.S.  Environmental Protection Agency
               401 M Street,  S.W.
            Washington, D.C.  20460

-------
          This  document  has  been  reviewed  and  approved  for
publication  by  the  Office   of  Toxic  Substances,   Office  of
Pesticides  and Toxic Substances,  U.S.  Environmental Protection
Agency.   The  use of trade names or  commercial products does not
constitute Agency endorsement or recommendation for use.
                               11

-------
                    AUTHORS AND CONTRIBUTORS

          This study  pf  asbestos  in public buildings represents
the   combined  efforts   of   several   organizations  and   many
individuals.  The names of the principal authors and contributors
of  the  various   organizations,   along  with  the  role  of  each
organization, are summarized below.

Battelle —  study design, planning,  Quality Assurance Plan (QAP)
preparation,  building and site selection,  external  analyses  of
air samples, two laboratory audits of R. J.  Lee Group, Inc., data
processing  and management,  statistical  analyses,   study  report
preparation.  Key Battelle staff included:

     Jeff Hatfield            Julius Ogden
     Jerry Stockrahm          Barbara Leczynski
     Fred Todt


Price Associates, Inc. — study design, planning,  help with QAP,
placement  of  air  sampling  pumps,   statistical analyses,  study
report preparation.   Key Price Associates staff included:

     Bertram Price            Jean Chesson
     James Russell


Alliance  Technologies Corporation  —  provided the two  "core"
raters, bulk  sample  collection and analyses,  field  work for air
sampling,  provided  portions  of  QAP   and  study  report.    Key
Alliance Technologies Corporation staff included:

     Patrick Ford             James Thomas
     John Fitzgerald          Richard Roat


R.  J.  Lee   Group,   Inc.  —  analyses  of   air  samples  using
transmission  electron microscopy, provided portions of  QAP and
study report.  Key R. J. Lee Group, Inc. staff  included:

     Rich Lee                 Drew Van Orden
     George Dunmyre


Midwest Research  Institute — external  analyses of bulk samples,
5 field audits of Alliance Technologies Corporation.  Key Midwest
Research Institute staff included:

     Paul Constant            James McHugh
                               111

-------
Georgia Institute  of Technology — provided  training  course for
raters.  Key Georgia Institute of Technology staff included:
     Dave Mayer
     Bill Ewing
William Spain
Steve Hays
McCrone   Environmental   Services
           provided  the   building
inspector.  Key McCrone Environmental Services staff included:
     Rich Hatfield
Anthony Claveria
EPA, OTS, Exposure  Evaluation Division — supervised all aspects
of this  study including design, planning, QAP,  analyses of bulk
and air samples, study report.  Key EPA staff were:
     EPA Task Managers:
     EPA Project Officers:
Joan Blake
Elizabeth Dutrow
Brad Schultz

Cindy Stroup
Mary Frankenberry
Joseph J. Breen
                               iv

-------
                        TABLE OF CONTENTS
AUTHORS AND CONTRIBUTORS

ACKNOWLEDGEMENTS

EXECUTIVE SUMMARY
1.0  INTRODUCTION  .....................    1
     1.1  BACKGROUND ....................    1
     1.2  OBJECTIVES ....................    3
     1.3  ORGANIZATION OF REPORT ..............    4

2.0  CONCLUSIONS ......................    5

3.0  QUALITY ASSURANCE ...................   15
     3.1  INTRODUCTION ...................   15
     3.2  BULK SAMPLE AND POLARIZED LIGHT MICROSCOPY
               QUALITY ASSURANCE ..............   15
          3.2.1  Side-by- Side Duplicates ..........   16
          3.2.2  External Analyses ........ * . . .  .   16
          3.2.3  Replicate Analyses  ............   16
     3.3  AIR SAMPLE AND TRANSMISSION ELECTRON MICROSCOPY
               QUALITY ASSURANCE ..............   17
          3.3.1  Production Lot Blanks ...........   17
          3.3.2  Field Blanks  ...............   17
          3.3.3  Field Audits  ...............   18
          3.3.4  Laboratory Audits .............   18
          3.3.5  Replicate and External Analyses  ......   18
          3.3.6  Examination of Additional Grid Openings  .  .   19

4.0  STUDY DESIGN  .....................   21

5.0  BUILDING SELECTION, INSPECTION AND ASSESSMENT
     FIELD METHODS .....................   25
     5.1  BUILDING SELECTION ................   25
     5.2  BUILDING INSPECTION  ...............   29
     5.3  ASSESSMENT ....................   30

6.0  RESULTS OF THE FIELD TEST OF THE ASSESSMENT  METHOD   .  .   35
     6.1  DATA ANALYSIS  ..................   35
     6.2  ASSESSMENT RESULTS ................   36
     6.3  CONCLUSIONS OF TEST OF ASSESSMENT METHOD  .....   39

-------
7.0  BULK  SAMPLE ANALYSIS AND POLARIZED LIGHT MICROSCOPY
     QUALITY ASSURANCE	   45
     7.1   BULK  SAMPLE ANALYSIS	   45
     7.2   BULK  SAMPLE AND POLARIZED LIGHT MICROSCOPY
                QUALITY ASSURANCE  .	   45
           7.2.1 Side-by-Side Duplicates 	   48
           7.2.2 External Analyses  	   48
           7.2.3 Replicate Analyses	   49
     7.3   Building Classification  	   49

8.0  AIR MONITORING	   53
     8.1   FIELD METHODS	   53
     8.2   AIR SAMPLE ANALYSIS	   53
     8.3   AIR SAMPLE AND TRANSMISSION ELECTRON MICROSCOPY
                QUALITY ASSURANCE	   54
           8.3.1 Production Lot Blanks	   55
           8.3.2 Field Blanks	   55
           8.3.3 Flow Rate Calibration	   56
           8.3.4 Field Audits	   56
           8.3.5 Laboratory Audits	   57
           8.3.6 Replicate and External Analyses  	   57
           8.3.7 Examination of Additional Grid Openings  .  .   59
     8.4   ANALYSIS OF AIR MONITORING  DATA	   61
           8.4.1 Methods	   61
           8.4.2 Results	   64

REFERENCES	   68

APPENDIX A RESPONSES OF  INDIVIDUAL RATERS IN EACH ASSESSED
            AREA WITHIN EACH REGION  TO CONDITION, POTENTIAL
            FOR DISTURBANCE, AND  AIR  FLOW FACTORS	   71

APPENDIX B COUNTS  OF THE RESPONSES  OF THE  RATERS  IN EACH
            ASSESSED  AREA WITHIN EACH REGION FOR CONDITION,
            POTENTIAL FOR DISTURBANCE, AND AIR FLOW
            FACTORS	   97

APPENDIX C CLASSIFICATION OF ACM CONDITION	133

APPENDIX D AIR SAMPLING FIELD METHODS 	   137

APPENDIX E AIR SAMPLE   PREPARATION   AND  SUMMARY  OF  TEM
            ANALYTICAL PROTOCOL   	   147

APPENDIX F ANALYSIS OF TEM GRID  OPENING DATA	155

APPENDIX G AIR MONITORING DATA LISTING	161

APPENDIX H GLOSSARY  	   181
                                Vi

-------
                         LIST OF FIGURES

Figure 2-1  Scatter  Plots   and  Medians  of   the   Average
            Airborne Asbestos  structure Concentrations  for
            Each Building Category and Outdoors  	    6

Figure 2-2  The  Average  Agreement  Index,  Also Called  the
            Average A-Value, for Condition,  Disturbance,  and
            Air Flow in Each of the 5 Study Regions   ....    9

Figure 2-3  Average  A-Values for  Condition with  Core  and
            Local Raters Plotted Separately in Each  of the 5
            Study Regions	   10

Figure 2-4  Average A-Values  for Disturbance with Core  and
            Local Raters Plotted Separately in Each  of the 5
            Study Regions	   11

Figure 2-5  Average  A-Values for  Air  Flow with  Core  and
            Local Raters Plotted Separately in Each  of the 5
            Study Regions	   12
     • .y
Figure 5-1  Overview of Field Methods	   26

Figure 5-2  Form  Used While at GSA  Regional  Off-ices  to
            Collect  Information  About  the  ACM   in  GSA
            Buildings	   27

Figure 5-3  Form  Used While at GSA  Regional  Offices  to
            Collect   Information  About   the  ACM   Within
            Specific Areas in GSA Buildings	   28

Figure 5-4  Chain-of-Custody Form for This Study	   31

Figure 5-5  Assessment Form  for Recording  Information About
            the ACM in a Given Area	   33

Figure 6-1  The  Average Agreement  Index,  Also Called  the
            Average A-Value, for Condition, Disturbance, and
            Air Flow in Each of the 5 Study Regions   ....   38

Figure 6-2  Average  A-Values for  Condition  with Core  and
            Local Raters Plotted Separately in Each of the  5
            Study Regions	   40

Figure 6-3  Average A-Values for Disturbance  with  Core and
            Local Raters Plotted Separately in Each of the  5
            Study Regions	   41
                               vii

-------
Figure  6-4  Average  A-Values  for Air  Flow  with  Core  and
            Local  Raters Plotted  Separately in Each of the 5
            Study  Regions	   42

Figure  7-1  Analytical  Data Form for Reporting  Bulk Sample
            Analysis	   46

Figure  8-1  Scatter   Plots  and  Medians  of  the  Average
            Airborne  Asbestos Structure  Concentrations  for
            Each Building  Category and Outdoors   	   65

Figure  D-l  Pump Diagram	141

Figure  D-2  Field  Data Form Used  for Air Monitoring  ....  144



                          LIST OF TABLES

Table 1     Summary Statistics for Average Airborne Asbestos
            Structure Concentrations (s/cc)   	   xv

Table 2-1   Summary Statistics for Average Airborne Asbestos
            Structure Concentrations (s/cc)   	    7

Table 5-1   List  of  Study  Regions,  Week . of  Rating,  and
            Number of Buildings  and Sites Within Buildings
            Rated  in Each  Study Region	   32

Table 6-1   Percentages   of   Sites  Which   Showed   Total
            Agreement  Among Raters  and the  Percentages of
            Sites with Minimum Agreement Among Raters  ...   37

Table 7-1   Final  Classification  of  Buildings in Categories
            1, 2,  and 3 in Each Study Region	   51

Table 8-1   Period of Air  Sampling Within Each Study Region    54

Table 8-2   Comparison  of   Airborne Asbestos  Concentrations
            Estimated  by   the  Original,  Replicate  (Same
            Laboratory), and External (Different  Laboratory)
            TEM Analysis	   58

Table 8-3   Estimated Mean, Variance, and Value of k for the
            Number of Structures  Counted Per Grid Opening
            Based on Examination  of 50 Openings	   61

Table 8-4   Summary Statistics for Average Airborne Asbestos
            Structure Concentrations (s/cc)   	   66
                               viii

-------
Table 8-5   Results  of  Randomization  Test  Indicating  the
            Statistical  P-Values  for  Differences  between
            Median Airborne Asbestos Concentrations  in Each
            of  the Three  Building Categories  and  Outdoor
            Concentrations 	   67

Table A-l   Responses   of   Raters  to  Overall   Condition
            Variable Separated by Region, Building, and
            Area	   73

Table A-2   Responses of Raters to Potential for Disturbance
            Separated by Region, Building,  and Area  ....   81

Table A-3   Responses  of Raters  to  Air Flow Separated  by
            Region, Building,  and Area	   89

Table B-l   Responses   of   Raters  to  Overall   Condition
            Variable,  Separated  by  Region,  Building,  and
            Area	   99

Table B-2   Responses    of    Raters    to   Potential    for
            Disturbance, Separated by  Region,  Building,  and
            Area	114

Table B-3   Responses  of Raters to Air  Flow, Separated  by
            Region, Building,  and Area	123

Table G-l   Air Monitoring Data Listing Showing the Asbestos
            Structure Concentration (s/cc)  at Each Site that
            was Air Sampled	163

-------
                         ACKNOWLEDGEMENTS

          Many  people  are to be acknowledged  for their time and
effort  towards  the successful  completion  of this  study.    In
particular,  from  the   U.S.  Environmental  Protection  Agency  we
thank  Martin  P.  Halper,  Director  of  the Exposure Evaluation
Division, Susan F. Vogt, Deputy Director  of the  Office of Toxic
Substances,  and  David  Kling,   Chief  of  the  Regulatory  and
Technical Assistance Section of the  Chemical  Control Division's
Hazard Abatement Assistance Branch.

          The  study could  not have  been performed  without the
full  cooperation  of the General  Services  Administration (GSA).
At GSA,  we  thank  Robert  J.  DiLuchio,  Assistant Commissioner for
Real  Property  Management  and  Safety,  and  Henry  J.  Singer,
Director  of the  Safety  and Environmental  Management  Division.
Special  thanks  go to  the many regional and local  GSA  staff who
participated in the  study, in a variety of ways.

          We are grateful to  all  the individuals  who  served as
raters in the field  in  each of the study regions.  They are to be
commended for  the many  hours  of service that  each provided for
the assessment  portion  of this study.

          From  Battelle,  we  acknowledge Jeanette Hochstedler for
assistance  in  the building selection  process,  Dean Margeson for
help  with  site selection,  Nick  Sasso  for  set-up  of computer
files, Dennis Haney for field coordination, Jill Daffer for help
with data processing, and Jan Clark and Pat Lyday for data entry.
Finally,  we  thank  Steve  Jones,  Katherine  Raeder,   and  Karen
Krasner   for   their   many   hours   of   word   processing   and
administrative  support.

-------
EXECUTIVE SUMMARY
          The   U.S.   Environmental   Protection   Agency   (EPA)
conducted  a field  study  during 1987  to  address a  number  of
asbestos exposure issues.   The study was designed and conducted
during  a  period when  information needs  concerning asbestos  in
buildings  were  rapidly changing  and expanding.   Late in  1985,
while  administering  the  Asbestos  School  Hazard  Abatement  Act
(ASHAA)  (Public Law 98-377),  EPA was  asked to  conduct  an  air
monitoring  study of  asbestos  levels in buildings.  The request,
submitted to the 99th Congress in a report by the House Committee
on Appropriations  (House  Report  99-212),  appropriated  $500,000
for  air monitoring  studies when  signed  on  November 25,  1985.
These  air  monitoring studies  were  intended  to provide data  on
asbestos   exposure   levels   inside   buildings  with   asbestos-
containing  materials  (ACM)  and ambient  outdoor  levels.    In
response to this  request,  EPA  began planning an  air  monitoring
study to be conducted in the latter part of 1986 and early 1987.

          Simultaneously,  EPA  was engaged  in the  development of
an assessment method for differentiating areas of ACM requiring
immediate  abatement  action from  areas  where  abatement  action
could be deferred.   An  assessment approach  had been proposed and
refined during a two-day workshop by consultants, administrators,
and  others with  asbestos  management  experience  (USEPA  1986a).
Plans were being developed to field test the assessment method.

          Plans for  the two  studies were  reconsidered when,  in
October, 1986, the Asbestos Hazard Emergency Response Act (AHERA)
(Public  Law  99-160) was   signed  into  law.    AHERA  introduced
assessment  and  abatement  concepts  and  terminology  that  were
different   in   some   major  respects  from  the   concepts   and
terminology that were  about to be field  tested.   AHERA required
EPA  to  promulgate  regulations  within   one  year,   addressing
inspections, abatement, and management of ACM in schools.

          In addition  to  regulations  pertaining to  asbestos in
schools, Section 213 of AHERA required EPA to report to Congress
within  one  year on  regulatory  issues for  public  and commercial
buildings.  Specifically, the report to Congress was to:

          •    assess  the  extent  to which  asbestos-containing
               materials  are  present  in  public  and  commercial
               buildings;

          •    assess   the   condition   of  asbestos-containing
               materials   in   commercial  buildings   and   the
               likelihood  that  persons  occupying such buildings,
               including service  and maintenance personnel,  are,
               or may be, exposed to asbestos fibers;
                                XI

-------
           •    consider   and  report   on  whether   public  and
               commercial buildings should be subject to the same
               inspection  and response  action  requirements that
               apply to school buildings;

           •    assess   whether   existing   federal   regulations
               adequately     protect     the    general    public,
               particularly  abatement personnel, from exposure to
               asbestos during  renovation and demolition of such
               buildings; and

           •    include  recommendations  that explicitly address
               whether  there is  a  need to  establish  standards
               for,  and  regulate  asbestos exposure in public and
               commercial buildings.

           Section  213  of  AHERA placed significant new demands on
EPA   to   collect  and  analyze  data  relating  to  asbestos  in
buildings.    Due  to the  short  time  frame to  meet  the  AHERA
requirements,  EPA  had to rely primarily on existing data for the
Section  213  report to Congress.  For example, to gain insight on
the   extent  and  condition   of  ACM  in public  and  commercial
buildings, additional analyses  of data  collected in a  1983-84 EPA
survey of  asbestos - in buildings were conducted  (Rogers 1987).  In
addition to using existing  information,  EPA conducted a field
study which  included air monitoring.  The study was designed with
three objectives:

           •    to   determine if  airborne   asbestos   levels  are
               elevated in buildings that have ACM;

           •    to  field test an  assessment  method  developed to
               facilitate   abatement  decision  making   in  the
               context of an asbestos management program; and

           •    to  gather data to help EPA make recommendations to
               Congress  on   future  regulation  of  public  and
               commercial  buildings (in  order to  meet the AHERA
               Section 213 requirements).

           In   order  to  satisfy  these   objectives,  inspection,
assessment,  and  air monitoring were conducted  in  three types of
buildings:   (1)  buildings  without  ACM;   (2)  buildings with all or
most  of  the  ACM in good condition  allowing for a limited number
of areas of moderate damage;  and (3) buildings which had at least
one  area  of  significantly   damaged  ACM  or numerous  areas  of
moderately   damaged  ACM.     These  three   building  types  are
subsequently referred to as  Categories  1, 2, and 3, respectively.

          Severe resource and time  constraints  precluded doing a
national   survey  of  public  and  commercial   buildings,  and
necessitated  the  identification  of  an appropriate subset  of

                               xii

-------
buildings.   During 1987,  when the  study was being planned,  the
General  Services  Administration  (GSA)  was  in  the  process  of
implementing a national asbestos management  program in federally
owned buildings.   The population of GSA owned buildings provided
an efficient laboratory for  this  study.  GSA buildings  had been
inspected  and  the  ACM  had  been   identified   and   evaluated.
Buildings   were   selected   based  on  prior   evaluation   and
identification with one of the building categories.  Without this
information, an  expensive and  time-consuming preliminary  study
involving  a much  larger number  of  buildings would  have  been
required  to  identify buildings that  contained asbestos.   Also,
building access was assured in GSA buildings, avoiding the access
problems typically encountered in privately-owned buildings.

          Based  on GSA asbestos building  records,  67  buildings
distributed  across  five  study regions were chosen for  study and
initially classified  into the three building  categories defined
above.  These buildings had  been inspected  for ACM previously by
GSA.  The buildings were  reinspected,  bulk  samples collected and
analyzed,  and  ACM condition rated  in  four or  more sites  per
building  by four  inspectors.   The bulk samples and  assessment
data  that were collected by the  field teams were used  to field
test   the   assessment   method  and   to   verify  the   initial
classification of  the building categories.   Forty-nine buildings
were  chosen  for  air monitoring.   Among  the  49 buildings six had
no  ACM (Category  1),  six had  ACM primarily in good condition
(Category  2),  and 37 had  at  least  one  area of  significantly
damaged ACM or numerous areas of moderately damaged ACM  (Category
3).   A  total  of  387  air   samples were collected  from  the 49
buildings (an average of eight per building) including 48 samples
of outdoor air (one per building with one exception where outdoor
sampling was not possible).

          Buildings initially  classified as Category  1 (no ACM)
or  Category 2  (ACM primarily in  good  condition) were entirely
reinspected  by  an  experienced building  inspector to  confirm the
classifications.   During  each  of  these inspections,  bulk samples
were  collected and analyzed  by polarized light microscopy  (PLM).
The ACM  in  buildings in Category 2 and Category 3  was  evaluated
by  an assessment  team consisting  of four  raters.    Two  "core"
raters  evaluated  four or more sites  in each Category 2  and  3
building  in each of the five study regions.  The other two  raters
were  "local" (regional EPA,  GSA,  or local city government  staff)
and  evaluated  buildings  in their  own  region only.    With the
exception of one study region, the raters attended a professional
training  course  before the  field work  began.   The  rating data
were  used to  confirm the  classification  of Category  2  and  3
buildings and to measure the consistency of the rating method.

          Air  monitoring was  conducted in  an average  of  seven
areas  inside each building  and  one area  outside.    Half  of the
inside samples were collected near the  most  damaged  ACM in each

                               xiii

-------
building,  if  damaged material was present.   ("Most damaged" could
mean   ACM  in  good   or   moderately  damaged  condition  if  no
significantly damaged ACM  was  found in  the  building.)    The
remaining samples were collected  in public  areas.   Each sample
was  taken  over  a two-day period,   eight  hours per  day, during
periods  of normal building  activity.   The samples were  analyzed
by  transmission  electron  microscopy (TEM)  using a direct filter
preparation method  to estimate the airborne asbestos structure
concentrations within the  buildings. Asbestos  structures include
asbestos  fibers and  asbestos bundles, clusters,  and matrices.

           A comprehensive  quality assurance (QA) program  governed
all field data collection and laboratory analysis activities.  A
variety  of standard  QA samples were collected and analyzed.  The
results  indicated a high  level  of  precision and accuracy in the
data.


Conclusions;

     Objective  (1):    Determine  if airborne  asbestos structure
     levels  are   elevated  in  buildings  that  have  asbestos-
     containing materials.

           While the  differences in  airborne asbestos levels are
small  in  absolute magnitude, the results of this study indicate a
tendency  for  average  airborne  asbestos  levels  in buildings with
ACM to be  higher than average  levels in  buildings  without ACM
(comparing the  medians  of the building  averages).    Airborne
asbestos  concentrations in Category  1 buildings  (no ACM)  have the
smallest  median  level; the  median  level  in Category 2 buildings
(all or  most  of the  ACM in good condition  allowing for a limited
number of  areas  of  moderate damage)  is higher than the median
level  in  Category 1;  and the median  level in Category 3 buildings
(at least  one area of  significantly  damaged ACM or numerous areas
of  moderately  damaged ACM)  is highest.    The  air monitoring
results  are  summarized  in Table   1.    The difference  between
Category  1  and Category 3  medians is statistically significant at
the  0.02   level.     The   remaining  comparisons  among   building
categories  are  not  statistically significant (each has  p-values
of 0.18 or greater).

           The median of building averages in Category 3 buildings
is higher than the  median ambient   outdoor  level.   The  evidence
for a  significant difference is not strong, but is suggestive of
a trend:   the difference is  statistically significant at  the 0.09
level  of  significance  (i.e,  the p-value is 0.09).   The other two
building  categories  when  compared  to ambient outdoor levels
suggest no difference (that is, they  have  p-values greater than
0.60).   Estimates of indoor  asbestos levels  are more  precise than
estimates  of  outdoor  levels because indoor levels are  based on
several  samples per  building.    Outdoor levels  are based on one

                               xiv

-------
Table 1.  Summary   Statistics   for  Average   Airborne  Asbestos
          Structure Concentrations (s/cc)
Statistic
Standard
  deviation
                                                   ACM
Outdoor
Category I    Category 2    Category 3
Median
Mean
Sample
size
<0. 00001
0.00039
48
(sites)
0.00010
0.00099
6
(buildings)
0.00040
0.00059
6
(buildings)
0.00058
0.00073
37
(buildings)
0.00096
  0.00198
0.00052
0.00072
Notes:

          1.   The data  points used in  the calculation  of  each
statistic  are the average  concentration within a  building  (for
indoor  samples)  or the concentration outside  each  building  (for
outdoor samples).

          2.   The mean for  Category  1  is  heavily  influenced by
one  sample in one building which produced an unexplained large
s/cc value.   The Category  1 mean,  excluding  this  one  value, is
0.00020 s/cc.
sample  per  building.   Thus,  an observed difference  between two
building  categories corresponds  to a  smaller p-value  than the
same  observed   difference  between  a  building  category  and
outdoors.

          The buildings in this  study  were selected  from three
building categories in order to investigate relationships between
building category and  airborne  asbestos  levels.   It is important
to  note that  the  method  of  selection does  not  allow formal,
statistical  projection of  the  total number  of  buildings  which
have characteristics measured in the study to the population of
GSA  buildings,   federally   owned   buildings,  or   public  and
commercial buildings.  However, since the buildings were selected
without prior  knowledge of airborne asbestos levels  or without
prior knowledge of  any  other  variable measured in the study, the
resulting relationships  are suggestive  of true  relationships in
buildings similar to those studied.

                                xv

-------
     Objective  (2);  Field test an assessment method developed to
     facilitate  abatement decision making  in the context  of an
     asbestos management program.

          Using rater consistency as an evaluation criterion, the
definitions of material condition, potential for disturbance, and
air flow used in this study, show promise as assessment tools for
use  in  the  field.    P-values  less  than  0.05  indicated  that
consistency  among raters  was  greater  than expected  if  ratings
were applied at  random.   Two hundred fifty seven areas within 60
buildings in  five study  regions  were  assessed by a  team of two
core raters.   In addition,  in  each of the  five  study regions a
team of  two local raters evaluated the buildings  in their study
region.

          Assessment  of  material condition  was the  factor most
consistently rated across the  five  study  regions.   Assessment of
potential for disturbance was less consistent.  Assessment of air
flow showed  the greatest level of  variability.   Trends observed
in the data suggest that  lack of consistency among raters can be
attributed,  in  part,  to  imprecision  in definitions  and  lack of
training.


     Objective  (3):  Gather data to help EPA make recommendations
     under AHERA to  Congress on  future regulation  of public and
     commercial buildings  (in order to meet the AHERA Section 213
     requirements).

          This objective  was met  indirectly by the total body of
information  collected  in this   study  which  was  combined  with
information  from a  variety  of  sources  and  considered  in the
preparation of  the AHERA Section 213 report  to Congress.  While
the air  monitoring  data collected  to  satisfy this  study's first
objective are not necessarily representative of air levels in all
public   and  commercial  buildings,   they  provide  information
pertinent to  exposure issues raised  in AHERA  Section  213.   The
conclusions  regarding the second  objective,  assessment method
evaluation,  are pertinent to  management  programs in public and
commercial buildings  as  well as the schools  that  are covered by
the AHERA  regulations.    This  information will be  considered in
the preparation  of future guidance  documents, training programs,
and regulations for public and commercial buildings.
                               xvi

-------
1.0  INTRODUCTION

1.1  BACKGROUND

          This study, referred to as the  Public  Buildings Study,
was designed and conducted during a period when information needs
concerning  asbestos  in  buildings  were   rapidly  changing  and
expanding.  Late in 1985, while administering the Asbestos School
Hazard Abatement Act  (ASHAA)  (Public Law 98-377), EPA  was asked
to  conduct  an  air  monitoring  study of  asbestos  levels  in
buildings.   The request,  submitted to the 99th  Congress  in  a
report by the House Committee on Appropriations (House Report 99-
212),  appropriated  $500,000   for  air  monitoring  studies  when
signed on November  25,  1985.   These air monitoring  studies were
intended  to  provide  data  on  asbestos  exposure  levels  inside
buildings with  asbestos-containing materials  (ACM)  and  ambient
outdoor levels.  In response to  this request,  EPA began planning
an  air  monitoring  study to  be conducted in  the latter  part  of
1986 and early 1987.

          Simultaneously, EPA  was  engaged in the  development  of
an  assessment method  for differentiating areas  of ACM  requiring
immediate  abatement  action  from  areas  where  abatement  action
could be deferred.   An  assessment approach  had been  proposed and
refined during a two-day workshop by consultants, administrators,
and  others  with asbestos  management  experience  (USEPA  1986a).
Plans were being developed to field test  the assessment method.

          Plans for  the two studies were reconsidered when,  in
October, 1986, the  Asbestos Hazard Emergency Response Act (AHERA)
(Public  Law  99-160)  was  signed  into law.    AHERA  introduced
assessment  and  abatement  concepts  and  terminology  that  were
different   in  some  major  respects  from  the  concepts  and
terminology that were about  to be field tested.   AHERA required
EPA  to  promulgate   regulations within   one  year,  addressing
inspections, abatement,  and management of ACM in schools.

          In  addition to regulations  pertaining to  asbestos  in
schools, Section 213  of AHERA  required EPA to report to Congress
within one  year on regulatory issues  for  public  and commercial
buildings.  Specifically, the report to Congress was to:

          •    assess  the extent  to  which  asbestos-containing
               materials  are  present  in  public  and  commercial
               buildings;

          •    assess   the  condition  of  asbestos-containing
               materials   in   commercial  buildings   and  the
               likelihood that  persons occupying such buildings,
               including service and maintenance personnel, are,
               or may be, exposed to asbestos fibers;

-------
           •    consider   and  report   on  whether   public   and
               commercial buildings should be subject to the same
               inspection  and response action  requirements  that
               apply to school buildings;

           •    assess   whether   existing   federal   regulations
               adequately     protect    the    general    public,
               particularly  abatement personnel, from exposure to
               asbestos during  renovation and demolition of such
               buildings; and

           •    include  recommendations  that explicitly address
               whether  there is  a  need  to  establish  standards
               for, and  regulate  asbestos exposure in public and
               commercial buildings.

           Section  213  of  AHERA placed significant new demands on
EPA   to   collect  and  analyze  data  relating  to  asbestos  in
buildings.    Due  to  the  short  time  frame to  meet  the  AHERA
requirements,  EPA  had  to  rely primarily on existing data for the
Section  213  report to  Congress.   For example, to gain insight on
the   extent  and  condition   of  ACM  in  public  and  commercial
buildings, additional analyses of data  collected in a  1983-84 EPA
survey of  asbestos in buildings were conducted  (Rogers 1987).  In
addition to using existing  information,  EPA conducted a  field
study which  included air monitoring.

           In order to meet the objectives of the public  buildings
study, which are  formally  stated and  discussed  in  Section 1.2,
inspection,  assessment,  and air monitoring  were conducted  in
three categories  of buildings:   (1) buildings without ACM;  (2)
buildings  with all or  most of the ACM in good condition allowing
for   a  limited  number of   areas  of  moderate damage;  and  (3)
buildings  which  had at  least one area of  significantly damaged
ACM  or numerous areas  of moderately damaged  ACM.    A number of
practical    problems    prevented    EPA   from   constructing   a
comprehensive  list of all  public and  commercial  buildings from
which to draw  a  national probability sample of buildings with the
desired  characteristics.   To develop such a  listing would have
required  more  resources than were  available for this study and
the additional time necessary would have precluded the collection
of information in  time to consider it  in the preparation of the
AHERA Section  213  report, due to Congress in October 1987.

          The  General  Services Administration  (6SA)   was  in  the
process of implementing a national asbestos management program in
federally owned buildings.   The population of GSA owned  buildings
provided an  efficient  laboratory for this  study.   GSA  buildings
had been inspected and the ACM had been identified and evaluated.
Buildings   were    selected   based   on   prior  evaluation   and
identification with one of the building categories.  Without this
information, an  expensive  and  time-consuming  preliminary  study

-------
involving  a much  larger  number  of  buildings would  have  been
required to  identify buildings that  contained asbestos.   Also,
building access was assured in GSA buildings/  avoiding the access
problems  typically  encountered  in  privately-owned  buildings.
Details concerning the study design, including the method used to
select buildings and data collection techniques,  are presented in
subsequent sections of this report.


1.2  OBJECTIVES

          The Public Buildings Study had three objectives:

          •    to  determine  if  airborne  asbestos  levels  are
               elevated in buildings that have ACM;

          •    to  field  test an  assessment method  developed to
               facilitate   abatement   decision  making   in  the
               context of an asbestos management program; and

          •    to gather data to help EPA make recommendations to
               Congress  on   future  regulation   of  public  and
               commercial  buildings (in order to  meet  the AHERA
               Section 213 requirements).

A brief discussion of each objective follows.


     Objective  (1);   Determine  if  airborne asbestos  levels are
     elevated   in   buildings  that    have   asbestos-containing
     materials.

          GSA  buildings  were  selected from three categories of
buildings for air monitoring.  Indoor and ambient  outdoor air was
monitored  for  asbestos  at  each  building.   Airborne  asbestos
levels measured in buildings  with ACM were compared to levels in
buildings  without  ACM.    Indoor  and ambient  outdoor levels were
also  compared.   These  comparisons  were used  to  determine if
airborne asbestos  levels in buildings with ACM are elevated.  The
buildings  in  this  study  were  selected  from   three  building
categories in order to investigate  relationships between building
category and  airborne  asbestos levels.   It is important to note
that the method of selection does  not allow formal, statistical
projection   of  the  total   number  of   buildings  which  have
characteristics measured in  the  study to the population of GSA
buildings,  federally owned buildings,  or public  and commercial
buildings.   However, since  the buildings were  selected without
prior  knowledge of  airborne  asbestos  levels  or  without  prior
knowledge  of  any other  variable  measured  in  the  study, the
resulting  relationships  are  suggestive  of true relationships in
buildings similar  to those studied.

-------
     Objective (2);  Field test an assessment method developed to
     facilitate  abatement decision making  in the context  of an
     asbestos management program.

          The  assessment  factors  described  in the  November 7,
1986 draft  of "Guidance  for  Assessing and Managing  Exposure to
Asbestos  in  Buildings"  (USEPA, 1986a) were tested  in the field.
Consistency  among  different raters assessing the same sites was
evaluated.   The buildings and  areas in buildings  used in the
study were  selected to ensure that a  range of  ACM  materials and
conditions would be rated.  The rating data collected to test and
analyze  consistency  among  raters,   therefore,  do  not  provide
information  about  the   characteristics  of  ACM  that  can  be
projected  to  the   population of  all  federal   buildings,    The
results,  however,   can   be  used  to  suggest   ways  in  which
consistency  among raters can be improved.


     Objective (3);  Gather data to help EPA make recommendations
     under AHERA to Congress on future regulation  of public and
     commercial buildings  (in order to meet the AHERA Section 213
     requirements).

          Section  213  of  AHERA  requires  that information  be
developed to address  the five specific issues previously listed.
Most  of  the  information  required  for  Section 213  has  been
developed in other  studies.   Information collected in the current
study  may   be   used  to  supplement  these  other  sources,  as
appropriate.


1.3  ORGANIZATION OF REPORT

          The report  consists of  eight sections.   This section,
Section 1,  includes the  background and objectives,  which provide
an  introduction to  the  study.    Conclusions are  summarized in
Section 2.   Quality assurance,  the study design,  and the  field
methods are  presented in  Sections  3,  4, and 5, respectively.  The
sample  analysis  methods,  statistical analysis, and  results are
presented in Sections  6,  7 and 8.   Section 6 addresses the  field
test  of  the  assessment  factors,   Section  7  addresses  bulk
sampling, and Section  8  is  the  air  monitoring portion  of the
study.

-------
2.0  CONCLUSIONS

          The  study conclusions are  organized and  discussed in
relationship to  the three study objectives.   During  1987,  when
the study was  conducted,  GSA was in  the  process  of implementing
an asbestos management  program in  buildings.   The  impact of the
management  program,   if  any,   is   not  addressed   in  these
conclusions.   Results  of  statistical tests are indicated by "p-
values."  A p-value is the probability of obtaining a result as
extreme or  more  extreme than the result  observed under the null
hypothesis  of  no difference  or relationship between the factors
being studied.   A small p-value indicates that the magnitude of
the observed  result is unlikely under the null  hypothesis,  and
therefore  lends  support  to  the alternative  hypothesis,  namely
that  the   difference  or   relationship   is   real.     Detailed
presentations of the results  supporting the conclusions are found
in Sections 6, 7, and 8.


     Objective  (1);   Determine if  airborne  asbestos  structure
     levels  are  elevated   in buildings that   have   asbestos-
     containing materials.

          While the differences  in airborne asbestos  levels  are
small in absolute magnitude,  the results of this  study indicate a
tendency for average  airborne asbestos levels in buildings  with
ACM to  be higher than average levels  in buildings without  ACM
(comparing  the  medians  of  the building averages).    Airborne
asbestos concentrations in Category 1 buildings (no ACM) have the
smallest median  level;  the median  level  in Category 2 buildings
(all or most of  the ACM in good condition allowing for a limited
number of  areas  of moderate damage)  is  higher  than  the median
level in Category 1; and the median level in Category 3 buildings
(at least one area  of significantly damaged ACM or numerous areas
of  moderately damaged ACM)   is  highest.   The  air  monitoring
results are presented in Figure 2-1 and summary  statistics  are
given  in  Table  2-1.   The  difference  between   Category 1  and
Category  3 medians is   statistically  significant  at  the  0.02
level.  The remaining comparisons among  building categories are
not  statistically  significant (each has  p-values of  0.18  or
greater).

          The median of building averages in Category 3 buildings
is higher  than the median ambient  outdoor level.   The evidence
for a significant difference  is not  strong, but  is suggestive of
a trend:  the difference is statistically significant at the 0.09
level of significance (i.e.,  the p-value is 0.09).  The other two
building  categories  when compared   to  ambient  outdoor  levels
suggest no  difference  (that  is, they have p-values greater than
0.60).  Estimates of indoor asbestos  levels are more precise than
estimates  of  outdoor  levels  because  indoor  levels  are based on
several samples  per building.  Outdoor levels are  based on one

-------
  0.0054
  0.004
  0.003
S
T
R
U
C
T
U
R
E

C
0
N
C
E
N
T
R
A
T
I
0
N
s/cc
  0.000
  0.002
  0.001
                                      A
                                      C
                                                      A
                                                      B
                    Outdoor
                  (48 sites)

            Figure  2-1.
                                     Building
                                    category 1
                                   (6 buildings)
  Building
 category 2
(6 buildings)
                                            Building
                                           category 3
                                         (37 buildings)
Scatter plots*and medians of the average  airborne
asbestos structure  concentrations  for each building
category and outdoors.
       *The  data  points  for  each scatter  plot  are the  average  concentration
       within  a building  (for indoor samples)  or the concentration  outside each
       building (for  outdoor  samples).-   A=l  data point, B=2  data  points,  ...,
       J=10  data points, and  Z=41  data points.   The diamond  represents the median
       of  the  data points in  each  scatter plot.

-------
Table 2-1.  Summary   Statistics   for  Average   Airborne  Asbestos
            Structure Concentrations (s/cc).
Statistic
Standard
  deviation
                                                   ACM
Outdoor    Category 1    Category 2
0.00096
0.00198
0.00052
                           Category 3
f f ' • •
Median
Mean
Sample
size
<0. 00001
0.00039
48
(sites)
0.00010
0.00099
6
(buildings)
0.00040
0.00059
6
(buildings)
0.00058
0.00073
37
(buildings)
0.00072
Notes:

          1.   The data  points  used in  the calculation  of  each
statistic are  the average  concentration within a building  (for
indoor  samples)  or the concentration outside  each building  (for
outdoor samples).

          2.   The mean for  Category  1  is  heavily influenced by
one  sample  in one building which produced an unexplained large
s/cc value.   The Category  1 mean,  excluding  this one  value,  is
0.00020 s/cc.
sample per  building.   Thus,  an observed difference  between two
building  categories  corresponds  to a  smaller  p-value  than the
same  observed   difference  between  a  building  category  and
outdoors.

          The buildings  in this  study  were selected  from three
building categories in order to investigate relationships between
building category and airborne  asbestos  levels.   It is important
to  note  that  the method  of  selection does  not  allow formal,
statistical  projection  of  the  total number  of  buildings  which
have characteristics measured in the study to  the population of
GSA  buildings,   federally   owned   buildings,   or   public  and
commercial buildings.   However, since the buildings were selected
without prior  knowledge of airborne asbestos levels  or without
prior knowledge of any other  variable measured  in the study, the

-------
resulting relationships  are suggestive of  true  relationships in
buildings similar to those studied.


     Objective (2);  Field test an assessment method developed to
     facilitate  abatement  decision making  in the context  of an
     asbestos management program.

          Using rater consistency as an evaluation criterion, the
definitions of material condition, potential for disturbance, and
air flow used in this study, show promise as assessment tools for
use  in  the  field.    P-values  less  than  0.05  indicated  that
consistency  among raters  was  greater  than expected  if  ratings
were applied  at  random.   Two hundred fifty-seven areas within 60
buildings in  five study regions  were  assessed by a  team of two
core raters.   In addition,  in each of the  five  study regions a
team of  two local raters evaluated the buildings  in their study
region.   The study  regions are  numbered  in the order  in which
they were sampled,  (i.e.,  Study Regions 1  to 5).  This numbering
scheme  is  not  related  to  the  regional  classification  used by
either GSA or EPA.

          Figure 2-2  shows  that assessment of material condition
was  the factor  most consistently  rated  across  the  five  study
regions.    Assessment  of  potential  for   disturbance  was  less
consistent.  Assessment  of  air flow showed the greatest level of
variability.  Trends observed  in Figures 2-3, 2-4 and 2-5 suggest
that lack of consistency among raters can be attributed, in part,
to imprecision in definitions  and lack of training, both of which
can  be remedied.    This conclusion  is based  on the  following
results:

          •    There  is greater consistency  among  raters  when
               assessing condition  than when assessing potential
               for  disturbance.    In  this  study, condition was
               defined in quantitative terms (e.g., terms such as
               greater than 10% damage) whereas  the definitions
               for disturbance were more qualitative.

          •    Consistency  in  air flow ratings varies from study
               region to study  region,  with low  consistency in
               some study regions and high consistency in others.
               The  present  two-part  scale does  not distinguish
               significant air flow from very slight air flow.  A
               three-part  scale  (high,  moderate, and  low/none)
               for air flow may increase consistency.
                                8

-------
                                    All Raters
            Condition
                    Study Region

                   Disturbance
Air Row
Figure 2-2.
The average agreement  index,  also called  the average A-
value,  for condition,  disturbance,  and air  flow in each
of the  5 study  regions.
         *The A-value is an agreement index developed  for this analysis
         to  demonstrate  consistency  in  scoring   of  the  assessment
         factors.   A-values  range from  1 for  maximum agreement  among
         raters  to 0 for minimum  agreement.   The basis  for A-values is
         explained in Section 6.

-------
                                   Condition
                                 Study Region
                    Local raters
•*•   Core raters
Figure 2-3.   Average A-values for condition with core  and local raters
              plotted separately in each of the 5 study regions.

         *The A-value is an agreement index developed for this analysis
         to  demonstrate  consistency  in  scbring  of   the  assessment
         factors.   A-values  range from  1 for maximum agreement  among
         raters  to 0 for minimum  agreement.   The basis  for A-values is
         explained in Section 6.

-------
          1
                                    Disturbance
         0.9 -


         0.8 -
     a>
     o»
    0.7 -

•
§   0.6 -


<   0.5 -


    0.4 -


    0.3 -


    0.2 -


    0.1 -


     0
            1
                                   Study Region
                     Local raters
                                       •*•   Core raters
Figure 2-4.    Average  A-values  for  disturbance  with  core  and  local
               raters plotted  separately in each of the  5  study regions.

          *The A-value is an agreement index developed for this analysis
          to   demonstrate  consistency  in  scoring  of  the  assessment
          factors.    A-values  range from  1  for maximum agreement  among
          raters to 0 for minimum agreement.   The basis for A-values is
          explained in Section 6.

-------
                                        Air Flow
10
                                      Study Region
                         Local raters
Core raters
   Figure 2-5.   Average A-values for  air flow with core and  local raters
                 plotted separately in each of the  5  study  regions.

             *The A-value is an agreement  index developed for this analysis
             to   demonstrate  consistency  in  scoring  of  the  assessment
             factors.   A-values  range  from 1 for  maximum  agreement among
             raters to 0 for minimum agreement.   The basis  for A-values is
             explained in Section 6.

-------
          •    Consistency  between  the  core  raters  is  greater
               than consistency  between the  local  raters.   The
               core raters  had more  experience in applying the
               assessment method.

          •    Region  5  local  raters,  who  did  not  attend the
               training,    showed  the   least   consistency   in
               assessing condition and potential for disturbance.


Objective  (3);    Gather  data  to help  EPA make  recommendations
under  AHERA  to   Congress  on  future reflation  of  public  and
commercial  buildings   (in  order  to  meet  the  AHERA Section 213
requirements).

          This objective was met indirectly by the  total  body of
information  collected in  this  study  which   was  combined  with
information  from a  variety of  sources  and  considered  in  the
preparation of the AHERA Section 213 report to Congress.   While
the air monitoring data  collected to satisfy  this  study's  first
objective are not necessarily representative of air levels in all
public  and  commercial   buildings,   they  provide  information
pertinent to exposure issues raised in AHERA Section 213.

          The  conclusions   regarding  the   second   objective,
assessment  factor  evaluation,   are  pertinent  to   management
programs  in  public   and commercial  buildings as  well  as  the
schools  that  are  covered  by  the  AHERA  regulations.     This
information  will be   considered  in  the  preparation  of  future
guidance documents,  training programs, and regulations for public
and commercial buildings.
                               13

-------
3.0  QUALITY ASSURANCE

3.1  INTRODUCTION

          This section presents an overview of  quality assurance
(QA) procedures followed in  this  study and the results  of  those
procedures.  The details on  the statistical analysis  of  QA data
are presented in Sections 7  and 8.  Guidelines  for  the QA of the
data collected  in  this study were set forth in a  comprehensive
Quality Assurance  Plan (Hatfield et al.  1987).  All  procedures
were employed, and unavoidable deviations were documented.

          Two types of measurements were  collected  in  the field:
bulk samples  and  air samples.   Bulk  samples were  analyzed with
polarized  light microscopy  (PLM)  to  determine  the  percentage of
asbestos  present.    For the purposes   of this  study, only  the
presence  (">1%")   or  absence  ("none  detected" or "trace")  of
asbestos  from  a  given  site was  utilized.    Air  samples  were
analyzed  with  transmission  electron   microscopy  (TEM)  direct
filter   preparation   to   estimate    the   asbestos   structure
concentration for  each site.   QA procedures were  performed for
all aspects of data collection and analysis.

          Chain-of-custody procedures  were  implemented  for  all
samples collected  during the project.    Field custody  procedures
were used to document the location  and handling of each sample
from the   time  of collection  until  received by the  analytical
laboratory.  At this point, internal laboratory records were used
to  document  the chain-of-custody of each sample through to its
final disposition.


3.2  BULK  SAMPLE AND POLARIZED LIGHT MICROSCOPY
     QUALITY ASSURANCE

          As specified  in  the QA plan,  various types  of quality
control (QC) samples were collected and analyzed to determine the
accuracy  and  precision of asbestos content  estimates.   Included
were:  side-by-side duplicates, and external  (referee  laboratory)
analyses  and  replicate analyses.   A side-by-side duplicate is  a
sample collected in the immediate area of the original sample but
handled separately.   The degree of agreement of  the analyses of
the  original  sample  with  its duplicate  indicates  the  level of
precision  in the sample collection and field  handling  procedures.
An  external  analysis is one in which the sample  is  analyzed  a
second  time by another  analytical  laboratory.    This  type of
analysis  is  performed as  a  QC check  on the  performance of the
method by the primary laboratory.  The degree of agreement  of the
original  analysis  with  the  external   analysis   indicates  the
consistency of  the method performance.   A replicate analysis is
one  in which  the  same sample  is  analyzed twice by  the  same
analytical  laboratory.    The degree  of  agreement of  the two

                                15

-------
analyses  indicates  the  level  of  precision  in the  laboratory
analysis procedures.

          All  three of  the QC  procedures described  above  were
used  for  the analyses  of the bulk  samples  in this study.   The
results  of  these  QC  analyses  indicated  a very  high  level  of
precision  and accuracy in  the  bulk  sample  collection  and PLM
analyses.


3.2.1  Side-by-Side Duplicates

          A  total  of   279  bulk  samples  were  collected by the
building inspector  and rating team in the field and analyzed for
asbestos content.  Of  these  279 samples, 20 were collected in the
field as side-by-side  duplicates  (20/259  = 8%).  With respect to
presence  or  absence  of  asbestos,  all 20  of  the side-by-side
duplicates agreed with their original  sample (100%  agreement).


3.2.2  External Analyses

          From the  279 bulk samples, 31 were randomly chosen for
external analysis  by  a second  laboratory (31/279  =  11%).   With
respect  to  the  presence or absence of  asbestos,  30 of  the 31
externals  agreed  with  their  original   samples   (30/31  =  97%
agreement).    The   one  disagreement  did  not  result  in  the
misclassification   of   the  building  category  as  verified  by
additional  bulk samples  collected  at that  area  by  the rating
team.

          The  primary  analytical  laboratory also  performed its
own QC  checks.   The laboratory participated  in the EPA Asbestos
Bulk  Sample Analysis  Quality Assurance  Program.   Sixteen bulk
samples  (for which the "true"  percentage of  asbestos is known)
were  submitted as blind QC samples along with the  field samples.
With  respect to the presence  or absence of asbestos,  all  16 of
these  results  agreed with the  original  determination   (100%
agreement).


3.2.3  Replicate Analyses

          The analytical  laboratory  randomly  chose 33 of the 279
field-collected  samples and resubmitted them  for replicate bulk
sample analysis  (33/279 = 12%).   With respect to the presence or
absence  of  asbestos,  32 of the 33  replicates agreed  with the
original result  (32/33 •  97%  agreement).   The one disagreement
did npt result in the  misclassification of the building category
because that building's classification was based on another area
where  several  samples  were  used  to verify   the presence  of
asbestos.

                                16

-------
3.3  AIR SAMPLE AND TRANSMISSION ELECTRON MICROSCOPY
     QUALITY ASSURANCE

          The  QA procedures  used  to  ensure  the  accuracy  and
precision of the air  sample collection  and  TEM analyses  included
the collection of production lot and  field  blanks,  field audits,
laboratory audits, replicate and  external analyses, and  a  study
to further evaluate the results obtained by  the TEM method.

          Production  lot blanks are filters chosen prior to  the
start  of  field  work.    They   are  analyzed  by  the  analytical
laboratory to  check  for filter contamination.   Field  blanks  are
filters taken  into  the field and handled in  the same manner as
exposed  air sample  filters.    Their  purpose  is  to  check  for
contamination which might occur in  the  field but  not as  a result
of air sampling.   Field audits determine whether the  field team
is following set procedures.   Laboratory  audits  determine  the
same  for the  analytical laboratory  personnel.    Replicate  and
external analyses serve the  same  purpose as discussed above  for
PLM analyses.


3.3.1  Production Lot Blanks

          Blank  filters from  prescreened  production  lots  were
randomly  selected  three times during  the  project:    at  the
beginning   of   field   activities,  in   the  middle,  and   near
completion.    Each  time,  two filter  cassettes  were  randomly
selected from  a  previously  unopened box of 50  filters.   A total
of  26 production lot blanks  were  selected  in  this  way  for
analysis.   The analysis of the production lot  blanks indicated
that   there   was  not  a  problem   with  background   filter
contamination.


3.3.2  Field Blanks

          During the pump set-up,  preloaded filter  cassettes were
selected as field blanks.  These filters were labeled and handled
in an  identical  manner as were the sample filters,  except that
they were  not attached to the sampling pump.   The filters were
capped during  active  sampling  periods  and  open faced  during the
non-run  hours  when the actual sample  cassettes were also open
faced.   Field blanks  were collected  in  30  of  the buildings
sampled.     The  purpose  of  the  field  blanks  was  to  measure
contamination  which  might occur  during  periods  when the  pumps
were not running.

          Of the 30 field blanks collected, 19 were selected for
analysis.   If  a high  level of contamination was  found  from the

                                17

-------
analysis  results  of the 19 blanks, the remaining 11 blanks would
have been analyzed.  The  19 blanks that were analyzed were chosen
at  random  from the  30  blanks  collected.   No  structures  were
detected  in 18  of the  19 field blanks  that  were  analyzed.   A
single  fiber was counted  on the  remaining blank.   This level of
blank contamination corresponds to an airborne asbestos structure
concentration of  0.00015  s/cc when 5,000 L of air is collected, a
very low  level  of contamination.  Thus,  it was not necessary to
analyze the  remaining blanks.


3.3.3  Field Audits

          Five  field  audits were conducted  by an independent
field auditor,  one  audit  in each  of the study regions.  The field
auditor accompanied the field crew during pump set-up in several
buildings per study region.  He checked to be sure that the field
crew  was  following  the  guidelines  set  forth  in the  Quality
Assurance Plan  (Hatfield et  al.  1987),  and  documented  any
violations   in  procedures  so  they  could  be  corrected.    For
example,  an  air hose on one pump  was found to be punctured.  This
was  noted and  immediately  corrected.    The field  auditor  also
measured  216 flow  rates  in  pumps in these  buildings.   This was
done  in  order  to  estimate the  relative  accuracy  of the  flow
rates,  defined  as  [(field  value-audit  value)/(audit  value)]  x
100.  The percentage of flow rates within  + 20% relative accuracy
was 99%.
3.3.4  Laboratory Audits

          To ensure the accuracy of the air sample analyses using
TEM,  two  laboratory  audits  were  performed.    An  independent
laboratory  auditor  visited  the  TEM  analytical laboratory  to
verify  that all  procedures  specified  in the  Quality Assurance
Plan  (Hatfield  et al.  1987)  were  followed.    He  audited  the
analytical laboratory twice,  at the beginning of  the  analyses and
at the end.
3.3.5  Replicate and External Analyses

          Twenty  air   samples  to  be  used  for  replicate  and
external TEM analyses  were chosen at random  from a total of 387
air  samples  (20/387   = 5%).   These  samples  were  receded  and
submitted  to the  original  analytical  laboratory  for replicate
analyses.   They were  then sent  to  a second TEM laboratory for
external analyses.   Thus,  for  these 20  samples,  3 measurements
were .collected for each sample:   the original, the replicate, and
the external.
                                18

-------
          Very few asbestos structures were counted in any of the
original,  replicate,   or  external   analyses.      No  asbestos
structures were detected on 13 of the 20 samples (65%).  A single
structure was  detected  by one or more of the three analyses on
the  remaining  seven samples.    Statistical  analysis  (Section
8.3.6)  of these  data  indicated  that  there  is  no evidence of
inconsistency  between  the  original,  replicate,  and  external
analyses.


3.3.6  Examination of Additional Grid Openings

          In  most of  the  original  387  TEM analyses,  10  grid
openings were counted per sample filter to estimate the number of
asbestos  structures  on  each filter.   These results  were  used to
compute structure concentrations.   To determine whether  10 grid
openings  per  sample  provided a sufficiently precise estimate of
the  number of structures  on  the   filter,  40  additional  grid
openings were  counted on  16 randomly-selected air samples,  for a
total  of 50  grid openings per  filter.   Statistical  analysis
(Section  8.3.7)  indicated  that,  in general,  examination  of 10
grid openings is sufficient.
                                19

-------
4.0  STUDY DESIGN

          The objectives of the Public Buildings Study, which are
formally stated and discussed in Section 1.2,  call for inspection
and air  monitoring in three types  of buildings:   (1)  buildings
without ACM;  (2)  buildings with all  or  most  of the ACM  in good
condition  (allowing  for a  limited number of  areas of  moderate
damage);   and  (3)  buildings   with  at  least  one  area   of
significantly damaged ACM or numerous areas of moderately damaged
ACM.  These are referred to  subsequently as Categories 1, 2,  and
3,  respectively.   The objectives focus  on relationships  between
building categories  and airborne asbestos levels and  ratings of
ACM characteristics in different types of buildings.

          Buildings  were   selected  for  the   study   from  the
population  of federally owned  buildings in  five geographically
dispersed  regions  of the United States  (two  cities on the east
coast, one midwestern city, one western city, and a  west coast
region  consisting of  two cities).    In this  report the  study
regions  are identified as  Regions  1 to  5  according to the order
in which they were sampled.  This numbering scheme is  not related
to  the  regional  classification  used  by  either  the  General
Services Administration (GSA) or EPA.

          A  target  quota  of   20   buildings   was  specified  for
evaluation in each study region:   four  buildings  in  Category 1;
four  buildings in Category  2;  and  12  buildings in  Category 3.
The initial classification of these buildings into categories was
based on inspection and evaluation information available from GSA
records.  The classification of each building was to be confirmed
by the project team, and ten buildings (two in Category 1, two in
Category 2, and six in Category 3) were to be  selected  in each
study region  for the air monitoring portion of the study.

          By pooling data across the study regions, estimates for
Category 3 would be  based  on measurements in  30 buildings  (i.e.,
five  study regions,  six buildings  per  study  region).   Estimates
for the  other categories were to be  based on measurements in 10
buildings  (i.e.,   five study  regions,   two  buildings  per study
region).     When  buildings   are   selected   randomly  and  the
coefficient of variation of  individual measurements is between  1
and 1.25, a range observed in previous studies, the likelihood of
detecting  a  five-fold difference  between Category 3  and one of
the other  categories using Student's  t-test  with a significance
level of 0.05 is at least 0.90 (i.e., the statistical power is at
least  0.90).     Relatively  low airborne asbestos  levels were
anticipated in Categories 1 and 2  (i.e., buildings with no ACM or
buildings  with ACM primarily in good condition), and therefore,
sample sizes  sufficient to detect  a five-fold differential were
considered adequate.
                                21

-------
          The    study   was   intended   to    investigate   two
relationships:   (i)   the  relationship  between airborne asbestos
levels and building category as defined previously; and (ii)  the
relationship  between  assessments conducted by different  raters.
It is  important to determine if these  relationships  can  be used
to differentiate among  extremes — i.e., damaged  ACM and ACM in
good condition,  or damaged ACM  and no ACM.   Therefore, buildings
with these conditions must be included in the  study regardless of
how  frequently  or  infrequently  these conditions  occur  in  the
population of all buildings.

          The  population  of  GSA  owned buildings   provided  an
efficient  laboratory for  this  study.    GSA   buildings had been
inspected  and   the   ACM  had  been  identified  and  evaluated.
Buildings   were   selected  based  on   prior  evaluation  and
identification with one  of the building categories.  Without this
information,  an expensive  and time-consuming preliminary study
involving  a  much larger  number  of  buildings  would  have been
required  to  identify buildings  that contained  asbestos.   Also,
building access  was assured in GSA buildings,  avoiding the  access
problems typically encountered in  privately-owned buildings.

          By  selecting  buildings  in  the manner  described,  the
results regarding the relationships studied apply, from a  formal
statistical perspective, only to the buildings in the study.  The
approach  used to  select the buildings,  however,  is  similar to
many experimental studies where  the  experimental  units selected
satisfy   predetermined   specified   criteria.     Under   these
circumstances,  projecting  the  total  number   of buildings which
have  characteristics measured  in  the  study  to  an  appropriate
target population is  not possible.  However,   since the buildings
were selected without prior knowledge of  airborne asbestos  levels
or  without  prior  knowledge  of  any  other   variable  that  was
measured  and  analyzed in  the  study,  the resulting relationships
are  suggestive  of  true relationships  in  buildings  similar to
those  studied.   [In  concept,  the study circumstances are typical
of   "analytical"   studies,   which   are  differentiated   from
"enuraerative" studies by Deming  (1950).]

          As  discussed  in  detail in  Section  6,  differences in
airborne  asbestos levels among building categories  and outdoors
are  indicated by plots  and tables  of summary  statistics.   The
measured   airborne   asbestos   levels,   and   consequently  the
statistics calculated from them,  are  subject  to various sources
of statistical error  including air sampling and analytical  error.
A statistical test was  applied  to provide a quantitative measure
of  the   strength of  evidence  associated  with  the  observed
differences  (i.e.,  probabilities  that the observed  differences
may have occurred only by chance were estimated).

          For planning  purposes,  specifically the determination
of sample size  discussed  above,  airborne  asbestos  levels were

                                22

-------
assumed to follow a  lognormal distribution  and to be amenable to
standard analysis of variance techniques used in previous studies
(USEPA 1985b, 1986b; Tuckfield et al.  1987).   Once the data were
collected it was apparent  that standard methods of analysis were
not appropriate  for this  study  because of  the large  number of
zero observations.   Therefore a  permutation (also referred to as
randomization) approach was used.  The permutation test, which is
based  on  the  null hypothesis  that   all  measured  levels  are
independent  observations   from  the  same underlying  statistical
distribution, is consistent with the objectives and design of the
study.  Buildings were selected according to condition of ACM and
without any knowledge of airborne asbestos levels.  Therefore, in
the absence  of any  relationship between condition  and airborne
asbestos   levels,   the  measured   values   will   be  equivalent
observations  from  a single  distribution.    To  compare building
categories, a "p-value," the level  of  significance,  is estimated
for each comparison.  The p-value is the probability of obtaining
a  difference as great  or greater  than the difference observed
under the hypothesis that  no true difference exists.   A small p-
value indicates that the magnitude  of  the  observed difference is
unlikely  under  the  hypothesis of  no  true  difference,  and
therefore  lends  support  to the alternative  hypothesis,  namely
that the difference is real.
                                23

-------
5.0  BUILDING SELECTION, INSPECTION AND ASSESSMENT
     FIELD METHODS

          Sixty-seven buildings were selected for this study from
a population  of  several thousand  GSA  buildings.   Selection  was
based on GSA's asbestos building records.   The assessment factors
were tested  in  60 of  the  buildings.   Forty-nine buildings  were
selected from the  original 67  for air monitoring.   This section
describes the  field methods used to  select  and categorize  the
buildings  by  asbestos   condition.     It  also  describes   the
procedures used  to test  the assessment  factors  and to collect
information to select  the  air  monitoring sites.  An  overview of
these methods is provided in Figure 5-1.


5.1  BUILDING SELECTION

          Initial   selection   of  buildings   was  achieved   by
reviewing the  existing asbestos  building records  maintained by
GSA in  each  of  the  five  study  regions.   Buildings  were  chosen
based upon the following criteria:

        • Each building must be GSA-owned to ensure  easy access.

        • Each building must contain occupied office space (e.g.,
          storage sheds were eliminated).

        • Each  building  must   have  adequate  asbestos  building
          records  indicating whether  or  not  an assessment  had
          been  performed  and  whether or not  ACM  was  present.
          Buildings with information on condition of the ACM were
          preferred.

        • Buildings with surfacing ACM were preferred.

        • All  buildings  within  a given  study  region must be
          within  a  small  enough  area   to  facilitate  sampling
          logistics.   The exception to this was  Study Region 4,
          which  consisted of  2 cities,   each  sampled separately
          but counted as one study region.

        • Buildings were  excluded if more than a 3-day security
          clearance was required for the  field personnel prior to
          gaining entry.

          Information  on  each  building  was collected  using the
forms shown  in Figures  5-2 and  5-3.   Selected  portions  of the
asbestos records for each  building were photocopied.

          Only  67  buildings  were found to  satisfy  the above
criteria.  Therefore,  the  target quota of 20 buildings per study
region  could not  be  obtained.   However, there  were sufficient

                                25

-------
i
            I
                                     Review GSA's
                                    building records
                                      Select 67
                                 buildings and classify
                                    as Categories
                                      1. 2. or 3
Building Inspector
inspects 28 buildings
from Categories
1 and 2



                                  Reclassify buildings
                                   if necessary and
                                  select 49 buildings
                                   for air sampling
I
                                                                              Rating team
                                                                           assesses selected
                                                                         material In 60 buildings
                                                                            from Categories
                                                                               2  and 3
i
6 buildings from
Category 1



6 building* from
Category 2
!

37 buildings from
Category 3


                                 Select air monitoring
                                   sites within each
                                    of 49 buildings
                                  — approx 7 inside
                                  — 1  outside
                                         Ajseismant data
                                        from approximately
                                       4 sites within each
                                         of 60 buildings
           Figure  5-1.    Overview of  field  methods
                                           26

-------
     U.S. EPA Study of Airborne Asbestos Levels in Buildings
           Initial Building Survey - Overall Building
                   Identification and Location
Building ID
Building Name
Address 	
City/State/Zip
Contact person
GSA owned?
GSA inspected?
   If yes, date?
                    Eligibility Verification
                   Yes   NO
                            Phone
                                    Yes
                   No
               Bldg type okay?
               Within geog. area?
                   Building and ACM  Information
Group letter
Assessment number
Exposure index 	
               Asbestos-containing surfacing
               material present?
               	Yes  	No  _
               If yes, condition?
             Don't know
Asbestos-containing
TSI present?
   If yes, condition?
            Yes
No
Don't know
Number of ACM areas identified
Bldg height (number of stories)
                           Year built
                            Comments
     Figure 5-2.
Form used while at  GSA regional offices to collect
information about the ACM in GSA buildings.
                                  27

-------
     U.S. EPA Study of Airborne Asbestos Levels in Buildings
          Initial Building Survey - ACM Containing Areas
Building  ID
Building Name
                            Location
Area Number __________
Type of ACM:           Surfacing
Condition:   Surfacing ________
                                          TSI
Other
             TSI
Have abatement procedures been implemented?
Comments:    	        	
                                                   Yes
        No
Area Number ____________
Type of ACM:      	Surfacing
Condition:   Surfacing 	
                            Location
                                          TSI
Other
             TSI
Hava abatement procedures been implemented?
Comments: 	
                                                   Yes
        No
   Figure 5-3.
                Form used while  at GSA regional offices to collect
                information about  the  ACM within specific areas in
                GSA buildings.
                                28

-------
buildings to allow air sampling in  10  buildings  per study region
as planned.  The breakdown of the number of buildings selected in
each study region is as follows:   Study Region 1  — 17 buildings,
Study Region 2  —  10 buildings,  Study Region 3  — 11 buildings,
Study  Region  4 —  13  buildings,  and  Study   Region  5  —  16
buildings.

          Based  on  GSA's  building  records,  each building  was
initially classified as  Category 1, Category 2,  or  Category 3.
Buildings which  contained  ACM,  but had little or  no information
on  the condition  of the  material were  placed  in Category  2.
Based  on  subsequent field  work  performed  in  this  study  the
classification  of  buildings  into  each  category  was  confirmed
prior to air monitoring.   These results are discussed in
Section 7-
5.2  BUILDING INSPECTION

          In order to verify the classification of the Category 1
and  Category  2  buildings,   a  qualified,  independent  building
inspector inspected the buildings which  were  initially placed in
these  categories.   The  verification  of  the classification  of
buildings in Category 3 was accomplished by the assessment team.

          In  each Category . 1 building,  the  building inspector
performed a  one-day inspection and collected bulk  samples from
any  areas  containing  friable  surfacing material  and  thermal
systems  insulation  (TSI).    In each  Category  2 building,  the
inspector bulk  sampled  any  friable  surfacing material and TSI in
areas  that  were  indicated  in  the  GSA  building  records  as
containing damaged ACM.   He  also  searched  for  and  sampled any
other  friable  material he determined  to be  in  worse condition.
Building inspection was performed in each study region during the
weeks  of:   Study Region  1  — March 16,  Study Region  2 — March
23,  Study Region 3 — April  6,  Study Region  4  —  April 20, and
Study Region 5  — May 4.

          Bulk   sampling  was  performed  following  the   random
sampling  procedures  described  in  USEPA  (1985a).   In addition,
convenience    samples    were    collected    in   the   following
circumstances:

          •     Pipe wrap  in fan/boiler room;

          •     Limited  access  to  sprayed-on material  due  to
                piping, ductwork, etc.; and

          •     Specific   requests   by   the  building  escort  or
                management not to collect more  than one  sample per
                site.
                                29

-------
Undamaged TSI, except for exposed material, was not sampled since
this would violate  GSA's  asbestos management program guidelines.
Only  friable  surfacing  materials  and TSI  were  sampled  (e.g.,
ceiling and floor tiles were not sampled).  Detailed notes of the
sampled  areas,  including the  condition of  the  material,  were
maintained by the  building  inspector.  A  chain-of-custody  form
was completed for  all  samples (Figure  5-4).   The  bulk samples
were  mailed  to  the laboratory for analysis  at  the end  of the
inspection period in each study region.

          Resulting from the  building  inspection is  the final
classification  of  each building  in Category  1 and  Category 2.
These results are presented in Section 7.


5.3  ASSESSMENT

          This section provides the methods for the field test of
the assessment method for  asbestos described  in  USEPA (1986a).
The  "field  test"  involved training  individuals  to  apply the
assessment  factors  and  determining   the  degree  of  consistency
among different  individuals assessing  the same material.

          Another   goal,   which   was   achieved   by   using   the
individuals  applying  the  assessment  factors, was  to  further
verify  the   condition   of   ACM  in  the  buildings  which  were
classified   (based   on  GSA's  asbestos   building  records)  as
Category 2 and Category  3.   The assessment team, also called the
rating  team,  collected bulk samples of  damaged friable material
to verify either the presence  or  absence of damaged ACM in these
buildings.   This information  was then used  to reclassify these
buildings, when  necessary,  for air monitoring.   The information
was also  used in the selection of  air monitoring sites for pump
placement.

          The  rating team  consisted  of two  "core"  raters who
visited every  study region,  and two "local" raters in each study
region.   The local  raters were regional  EPA,  GSA, or local  city
government staff.

          A  professional, two-day  training course was  held for
the raters  1$ weeks  before field  work  began.  This  produced a
group  of individuals  with  experience and  training  typical of
those  likely  to be  assessing ACM in  real-world applications.
Study Region  5 local raters did  not  attend  the training course
because  the  decision  to  include  this city  was made  after the
training course had taken place.

          The    rating    team    assessed    approximately   four
predetermined sites in each of between 9 and 15 buildings in each
of the  study regions.   The sites  were selected  to  represent a
variety of conditions,  locations,  and types of ACM.   Table 5-1

                                30

-------
Sample Slle	
Dal* Sampled.
Shipped By	
DaU Shipped .
Carrier 	
                                   ID Number	
                                   Invoice Number
                                   Dale Recelvad_
                                   Received By	
                                   Condition	
                                            . Test Number.
   SAMPLE
NUMBER
SAMPLE
DESCRIPTION
 SHIPPING
CONDITION
RECEIVING
CONDITION
Signature ol Sender .

Signature ol Receiver.
                                                     Dale.

                                                     Dale.
            Figure  5-4.   Chain-of-custody form for  this  study.

-------
Table 5-1.  List of  Study Regions,  Week of Rating, and Number of
            Buildings  and Sites Within Buildings Rated  in Each
            Study Region
Study
region
1
2
3
4
5
Week of
rating
March 23
April 6
April 20
May 4
May 18
Number of
buildings
15
9
11
11
14
Number of
sites3
64
37
47
52
57
aRated by two or more raters
lists  the  study  regions,  week  of  rating,  and  the number  of
buildings and sites.  The study regions are numbered in the order
in which they were sampled.  This numbering scheme is not related
to the  regional classification used by either  GSA or EPA.   Each
rater completed the  form shown in  Figure  5-5  in order to obtain
the  necessary  assessment  factors  described  in  USEPA  (1986a).
This form contains the key  factors  discussed in USEPA (1986a) for
assessing   ACM   (i.e.,    overall   condition,   potential   for
disturbance,  and  air flow) as well  as  other  information.   The
exact definitions  of these factors can be found in the Glossary
(Appendix H).   The rating  team did not discuss the sites before
or  during  rating in order  to  ensure  that  the ratings  were
accomplished independently.

          After all  predetermined sites in a building were rated
by the  team,  additional information was  collected  in  order  to
verify the building category and to collect information to select
air monitoring  sites.  The  location of the area thought to be the
most damaged  ACM in  the building,  based  on GSA building records
or information  collected by the building inspector, was contained
in  a sealed envelope  and opened  by  the  rating  team at  the
conclusion  of  the rating  process.   (Note that  "most  damaged"
could  mean  good  condition  if  there  were  no  moderately  or
significantly damaged areas in the building.)   The  rating team
was instructed  to  bulk sample this  area to verify the presence of
asbestos and any areas which  they had visited and found to be in

                                32

-------
                         Assessment Form
Building ID

Area ID 	
Type of ACM

     Surfacing      Thermal System Insulation       Other.

     Description
Condition
     Percent Damage      _____ %
     Distribution of Damage: Localized 	  Even
     Type of Damage:
      Deterioration       Physical Damage	Water Damage

     Description	
     Overall condition:  Sig Damage     Mod Damage      Good.

Potential for Disturbance

     Accessibility:  High          Low
          Description 	
     Vibration: High     Low      None..
          Source
     Air Erosion: High	   Low	  None_
          Source
     Overall Potential for Disturbance:
                              High	   Moderate	  Low_

Air Flow
     In Air Flow?   Yes	No
     Type of Flow 	
Comments
Signed 	             Date
   Figure 5-5.  Assessment form for  recording information about the
                ACM in a given area.
                                 33

-------
worse  condition.    Occasionally,  the  building escort  suggested
additional areas that were also bulk sampled.

          Bulk  sampling  was   performed  following  the  random
sampling  procedures   in  USEPA  (19853),   although  convenience
samples were collected in the following circumstances:

          •    Pipe wrap in fan/boiler room;

          •    Limited  access  to  sprayed-on  material  due  to
               piping, ductwork, etc.; and

          •    Specific  requests   by   the  building  escort  or
               management not to collect more than one sample per
               site.

Undamaged  TSI,  except for  exposed  material, was  not  sampled
because  that  would violate  GSA's  asbestos  management  program
guidelines.    Only  friable  surfacing  materials   and  TSI  were
sampled  (e.g.,  ceiling and floor  tiles were  not  sampled).   The
rating  team kept detailed  notes of  the areas sampled  and also
sketched  the location of  the material.   The bulk samples were
carried  back to the  laboratory for analysis  at  the end  of the
rating period in each  study region.

          The purpose  of bulk sampling the areas  described above
was to  confirm the initial classification  of  buildings  based on
GSA records  as Category 2  or Category 3.   Depending on whether
the areas rated  contained  asbestos or not,  the  buildings were
reclassified as  necessary  prior to air sampling.   These results
are discussed in Section 7.
                                34

-------
6.0  RESULTS OF THE FIELD TEST OF THE ASSESSMENT METHOD

6.1  DATA ANALYSIS

          ACM condition for a specific area within a building can
take one  of three  possible  values:   good,  moderate  damage,  or
significant  damage.    These  values  are coded  as  1,  2,  and  3,
respectively.   Likewise,  potential  for disturbance is  rated  as
low,  moderate,   or  high.    These  are coded  as  1,   2,   and  3,
respectively.   For  the remainder  of this report,  potential for
disturbance will be referred to as "disturbance."  Air flow takes
the values yes or no, coded as 1 and 0, respectively.

          The data  listing in Appendix A shows  the responses  of
the  individual   core  and  local  raters for  each  of  the three
assessment factors  at  each  site.   Table A-l gives  the condition
of  the sites,  Table A-2  gives  the disturbance  rating  of the
sites,  and  Table  A-3  shows the  air  flow  of  the site.   The
frequency of occurrence of each score is given in Appendix B.

          Consistency of ratings may be measured in a  variety of
ways.   One  simple  measure   is  the  percentage  of sites where
perfect agreement occurs (i.e., all raters give the same rating).
Another  measure  is  the  percentage   of  sites  where  minimum
agreement  occurs  (i.e., at  least  one  rater assigns the maximum
rating and at least one rater assigns the minimum rating).  While
both  measures  have  been  calculated  and  summarized,   neither
distinguishes between  situations  where just one rater disagrees
with the  remaining  raters  from situations where there is little
agreement among any of  the raters.   An "agreement"  index, A, was
developed  to measure  the overall  degree  of  consistency among
raters  taking  into  account  both  the number  and magnitude  of
disagreements.    A  is  defined  as  1  minus  the  sum   of  the
differences  between the raters'  scores,  normalized so  it takes
values between 0 and 1.  It is calculated as:
          A  =  1 -
                         max
where    i  *    n   = number of raters which rated the site;
        %'-. $£• &'                         . V
         «     X;  = response of the i   rater; and

               max = the theoretical maximum that the numerator
                     can take for n raters.

          When  there is  perfect agreement  between raters  at  a
site,  A=l.     A=0   indicates  minimum  agreement.    A   similar
measurement of  rater agreement was used  in  a previous EPA  study
(USEPA 1981).  In that case the measure was divided  by the number

                                35

-------
of  comparisons  rather  than  normalized to  range from  0 to  1.
Although  the  two measures are  not  quantitatively identical,  the
qualitative results of the two  studies can be compared.

          A-values  were calculated for each  site  and  averaged
within  study  regions to provide a summary  of  rater consistency
for each  of the  three assessment factors.

          A statistical  test was developed  to  determine whether
the  assessment  factors  generate  more  agreement than  would  be
expected  if  the  raters  were assigning  ratings  at  random.   The
probability  distribution  of A was  calculated  under  the  null
hypothesis  of  random assessment.   From  this,  the probability
distribution  of A averaged over sites  was  generated by computer
simulation for each combination of number  of  ratings, number of
raters, and number of sites.  If the observed average is found in
the  upper tail of  the distribution,  it is  unlikely  that  the
ratings are being assigned at random.   An observed average equal
to the  (l-a)th percentile,  i.e., a p-value of a, would be called
statistically  significant with  a significance level equal to oc.


6.2  ASSESSMENT RESULTS

          Table 6-1  shows  the percentages  of  sites  with total
agreement among raters  and  the  percentages  of   sites  having
minimum  agreement.   Minimum agreement  occurs  at  a site  if  at
least one rater scores  the lowest value of a factor and at least
one rater scores the highest value of the factor.  Note that all
the percentages are  sensitive  to  the number of  raters  present.
Total agreement is more likely with  fewer  raters,  while minimum
agreement is more likely when there are  more raters.

          Overall,  there  is greater  total  agreement  and  less
minimum  agreement for  condition than  for  disturbance,  although
the  relationship  varies  slightly  from study  region  to  study
region.    The  highest  frequency of minimum agreement occurs  in
Study Region  5.  High total  agreement  and  minimum agreement for
air flow  are  not surprising since  there are only two ratings (0
and 1), and the  2 percentages must  sum to 100.

          In   Figure  6-1,   average   A-values   for  condition,
disturbance, and air  flow  are plotted by study region with study
region numbered in chronological order.   The A-value for a given
site  is  based on the  ratings  of  between  two   and  four raters
depending on  how many raters  were actually present  at  the site
and  whether  all  of the  raters  completed  all  the  assessment
factors at a  site (there were occasional missing entries).  Each
average   measure  of  consistency   shown in  Figure  6-1  has  a
significance  level  less  than  0.05 when  tested against random
ratings.    The  study,   therefore,  indicates   a  tendency  for
consistency among raters.

                                36

-------
Table 6-1.
Percentages of Sites Which Shoved Total Agreement
Among Raters and the Percentages of Sites with
Minimum Agreement Among Raters
Study
region
    Assessment
      factor
 % Total
agreement
% Minimum
agreement
 All
    Condition
    Disturbance
    Air flow

    Condition
    Disturbance
    Air flow

    Condition
    Disturbance
    Air flow

    Condition
    Disturbance
    Air flow

    Condition
    Disturbance
    Air flow

    Condition
    Disturbance
    Air flow
  31.3
  35.9
  55.6

  48.6
  13.5
  29.7

  29.8
  25.5
  72.3

  38.5
  15.4
  75.0

  22.8
  17.5
  66.7

  33.1
  22.6
  61.3
    0
    4.7
   44.4

    0
    8.1
   70.3

    6.4
    4.3
   27.7

    1.9
    7.7
   25.0

   19.3
   19.3
   33.3

    5.8
    8.9
   38.7
Note:  Minimum  agreement  occurs at a site if  at least one rater
scores the lowest value of a factor and at least one rater scores
the highest value of a factor.
                                37

-------
                                          All Ratera
oo
           1
           1
           0
           O»
                  Condition
 Study Region

Disturbance
Air Row
    Figure 6-1.    The average  agreement index,  also called  the average A-
                   value, for  condition, disturbance,  and air  flow in  each
                   of the 5 study regions.

              *The A-value is an agreement  index developed for this analysis
              to  demonstrate  consistency   in  scoring  of  the  assessment
              factors.   A-values range  from  1  for maximum  agreement among
              raters to 0 for minimum agreement.

-------
          The average  A-value for condition  tends to  be higher
than  the  A-value  for disturbance.    This  suggests  that  the
definition of condition  used in this  study can be  applied more
consistently than the definition  of  potential  for  disturbance.
The average  A-value  for  air flow varies  from  study  region  to
study  region.    In the  first two  study regions  there  is  less
agreement on air  flow  than on condition  and disturbance.   In the
last three study regions the opposite is true.

          A  previous   EPA study  on  assessing  ACM  also  showed
greater  consistency when  rating  condition  (USEPA  1981).    The
assessment method used in  that study differs  from the  current
one,  but factors  related to condition  (water  damage,  physical
damage)  showed  greater rating consistency  than  those  related to
disturbance (accessibility, activity).  The 1981 study considered
a factor "air-moving system" which was more narrowly defined than
in the present study and showed high consistency among raters.

          A-values were also calculated for core and local raters
separately (Figures 6-2,  6-3  and 6-4).   The core raters were the
same  two individuals  across  all study regions,  while  the  local
raters were  represented by  different  individuals in  each  study
region.   For condition and  disturbance (Figures  6-2 and  6-3,
respectively),  the amount of agreement  between the core raters
begins at between 0.8  and 0.9 and  remains fairly constant across
study regions (with some  slight improvement over time).  For the
core  raters,  all  average A-values for condition and disturbance
have  p-values less than 0.05.   In Study Region  1,  local raters
have  about the  same   amount  of agreement  as  the core  raters.
Local  raters'  level   of  agreement  tends  to decrease  for  the
remaining study regions.   For condition, the test of consistency
among local raters had a significance level less than 0.05 in all
study regions except  Study Region  5.   Study  Region  5  raters did
not attend  the  training  course.   A-values for  disturbance have
significance levels less than 0.05 only in Study Regions  1 and 4.
For air  flow  (Figure  6-4), the  relationship between the core and
local raters varied from  study  region  to study region.   In Study
Regions  2  and 4,  the core raters  show  less agreement than the
local  raters.   In  Study  Regions  1, 3,  and 5,  the  local raters
show  less  agreement.    All air  flow A-values  have significance
levels less  than 0.05 with  the  exception of Study  Region  2 for
core raters and Study Region  1 for local raters.


6.3  CONCLUSIONS  OF TEST OF ASSESSMENT METHOD

          Based   on  the  criterion  of  rater   consistency,  the
assessment  factors,  as defined in this  study,   show  promise as
assessment tools  for  use in  the field.   Lack of consistency can
be attributed, in part, to imprecision in definitions  and lack of
training,  both  of  which  can be  remedied.   This  conclusion is
based on the following results:

                                39

-------
                                  Condition
                                 Study Region
                   Local raters
Core raters
Figure 6-2.   Average A-values for condition with core and local raters
              plotted separately in each of the  5 study regions.

         *The A-value is an agreement index developed for this analysis
         to  demonstrate  consistency  in  scoring   of  the  assessment
         factors.   A-values  range from  1 for  maximum agreement among
         raters to 0  for minimum agreement.

-------
            1


          0.9-


          0.8-

             I
          0.7-


          0.6 -


          0.5 -


          0.4-


          0.3 -


          0.2 -


          at-I


           0
                                     Disturbance
             1
                                    Study Region
                   •   Local ratera
Core ratera
Figure 6-3.    Average  A-values  for  disturbance  with  core  and  local
               raters plotted  separately in each of the  5  study regions.

          *The A-value is an agreement index developed for this analysis
          to   demonstrate  consistency  in  scoring   of  the  assessment
          factors.   A-values  range from  1 for maximum agreement among
          raters to 0 for minimum agreement.

-------
                                        Air now
to
                              234

                                      Study Region

                         Local raters               +  Core raters
   Figure 6-4.   Average A-values  for air flow with  core and local raters
                 plotted separately in each of the 5 study regions.

            *The A-value is an  agreement index developed for this analysis
            to  demonstrate  consistency  in  scoring  of  the  assessment
            factors.   A-values  range from  1  for maximum  agreement among
            raters to 0 for minimum agreement.

-------
Consistency   between   raters   is   significantly
greater than would be predicted if assessment were
at random.

There  is greater  consistency  among raters  when
assessing condition than  when  assessing potential
for  disturbance.    In  this  study,   condition  was
defined in quantitative terms (e.g., terms such as
greater  than  10%  damage)  whereas the  definitions
for    disturbance    were    more    qualitative.
Consistency in air  flow ratings varies  from study
region to  study region,  with  low  consistency in
some study regions and high consistency in others.
The  present  two-part  scale  does  not  distinguish
significant air flow from very slight air flow.  A
three-part  scale  (high,  moderate,   and  low/none)
for air flow may increase consistency.

Consistency  between the  core  raters  is  greater
than consistency  between  the  local raters.   The
core raters  had more  experience  in applying  the
assessment method.

Region  5 local  raters,  who did not  attend  the
training,   showed   the  least   consistency   in
assessing condition and potential for disturbance.
                 43

-------
7.0  BULK SAMPLE ANALYSIS AND POLARIZED LIGHT MICROSCOPY
     QUALITY ASSURANCE

7.1  BULK SAMPLE ANALYSIS

          All bulk  samples collected  by the building inspector
and rating  team were  analyzed with  polarized light  microscopy
(PLM), as per Appendix A of  USEPA (1982).   For the purposes of
this study, however, only  the information about presence (">1%")
or absence ("none detected" or "trace") of asbestos was used.

          The method described in USEPA  (1982) is  most commonly
used to  identify and quantify asbestos fibers in  bulk samples.
It can  distinguish fibers  of the serpentine group (chrysotile)
from  those   of  the  amphibole   group  (amosite,   crocidolite,
anthophyllite,   tremolite-actinolite),  and   is    sensitive  to
asbestos content as  low as 1%.  Asbestos fibers are reported as
area percentages of the total sample.   This method is  limited to
fiber sizes greater than 1 JOB in length.
          Sample  preparation   involved  gross   examination  to
characterize  the sample and to determine  homogeneity.    If the
sample  was   not homogeneous,   each  phase  was  separated  for
individual  identification.   At least  one  microscope slide was
prepared for  each phase by teasing a  small piece of sample from
the bulk and mounting it on the slide with a refractive index oil
(n  = 1.54)  and  cover slip.    All  gross  examinations and slide
preparations  were  performed  in  the   glovebox   to  protect  the
analyst.

          Sample  examination   involved   analysts   trained  in
classical crystallography  techniques.   The  specific techniques
used depended upon the  nature  of the sample,  but in general, the
    lination utilized magnifications  of 20X  to 400Z,  and at least
four fields were counted.

          The analytical  laboratory used standard  asbestos data
forms (Figure 7-1), bound in  a notebook format for recording all
analytical data.  Data were recorded in duplicate,  with one copy
remaining  in  the  analyst's  notebook,   and  the  other  being
submitted to the data entry technician.


7.2  BULK SAMPLE AHD POLARIZED LIGHT MICROSCOPY
     QUALITY ASSURANCE

          The  collection  of  bulk  samples   by  the  building
inspector  and  rating  team  was  a  QA  activity  because  this
information was used to verify the classification of the building
categories.   The final  classification  of  these buildings   is
discussed  in  Section 7.3.   In addition,  specific  QA procedures


                                45

-------
                          ASBESTOS ANALYSIS WORKSHEET
                            ASBESTOS ANALYSIS  GROUP
PROJECT:                                          CONTRACT i


FIELD ID *                    ID #                  PacUity  ID *


SAMPLE LOCATION:


SAMPLE DESCRIPTION/CONDITION:


ASBESTOS FOUND:      	NONE  DETECTED        TRACE     	>1%

PLEASE COMPLETE FOR ALL ITEMS.  ENTER "NONE DETECTED" IF NONE  FOUND.

         MATERIAL               PERCENT             IF A RANGE, SPECIFY HERE.

ASBESTOS;

CHRYSOTILE                      	             	 - 	
AMOSITE                         	                       " -           "
OTHER ASBESTOS
SPECIFY OTHER TYPE(S)  OF ASBESTOS:

FIBROUS NON-ASBESTOS:
  GLASS WOOL OR MINERAL WOOL
  FIBERGLASS
  CELLULOSE
  OTHER FIBROUS NON-ASBESTOS
    MATERIALS
SPECIFY OTHER TYPE(S) OF FIBROUS NON-ASBESTOS:_

NON-FIBROUS:

  PERLITE                       	
  VERMICULITE                   	
  OTHER NON-FIBROUS
    MATERIALS
SPECIFY OTHER TYPE(S)  OF NON-FIBROUS:
DATE OF ANALYSIS:                               ANALYST:
  Figure 7-1.  Analytical  data  form  for  reporting  bulk   sample
                 analysis.
                                     46

-------
were used for  the  bulk sample collection and  analyses  to ensure
their precision and accuracy.

          Chain-of-custody  procedures  were  implemented  for  all
samples collected  during  the project.   Field  custody procedures
were used to document  the location of  a sample from  the  time of
collection until  its receipt by  the analytical laboratory.   At
this point, internal laboratory records were used to document the
handling of the sample through its final disposition.

          Standard sample custody (traceability)  procedures were
used during this project.   Each  sample was  labeled with a unique
random identification  number immediately after collection.   This
number was recorded in the field logbook along with the following
information:

          •    Name(s) of the sampler;

          •    Date of collection;

          •    Sample location;

          •    Sketch of location (rating team only);  and

          •    Comments.

          A chain-of-custody form was filled out in the field for
all samples.   A  copy of  the form was included with each  shipment
of  samples  to  the  analytical  laboratory.    Figure  5-4  is  a
representative copy  of the  chain-of-custody form used during the
project.

          Upon   receipt   of  the  samples   at   the   analytical
laboratory, the following steps were taken:

          •    Sample   labels   were   cross-checked   with  the
               accompanying custody form.

          •    Samples were logged in a master sample logbook.

          •    Prior to  and after analysis,  samples  were  stored
               in a controlled area.

          •    Samples  handled   by laboratory  personnel  were
               traced by proper log-in/log-out procedures.

          Specific QA  procedures were performed  to estimate the
precision  and  accuracy  of  the  bulk  sample  collection  and
analyses:   side-by-side  duplicates,  and external  and replicate
analyses.  A side-by-side duplicate is a sample collected  in the
immediate  area of  the original  sample but  handled separately.
The degree  of  agreement  of  the  analyses of  the original  sample

                                47

-------
with its duplicate indicates the level of precision in the sample
collection  and field handling procedures.   An external analysis
is one  in  which the sample is analyzed  a  second time by another
analytical  laboratory.   This type of analysis  is  performed as a
quality  control (QC) check on  the performance of  the  method by
the primary laboratory.   The  degree of  agreement of the original
analysis with  the  external analysis indicates the consistency of
the method  performance.  A replicate analysis is one in which the
same sample is analyzed twice by the analytical laboratory.  The
degree  of  agreement of  the  two analyses indicates the level of
precision in the laboratory analysis procedures.

          All  three of  the QA  procedures described  above were
used for the  analyses  of the bulk  samples  in this  study.   The
results  of  these  QA  samples  indicated a  very  high  level  of
precision  and  accuracy in  the  bulk sample collection  and PLM
analyses.


7.2.1  Side-bv-Side  Duplicates

          A total  of   279 bulk  samples were  collected  by the
building inspector and rating team in the field and analyzed for
asbestos content.  Of  these 279  samples, 20 were collected in the
field as side-by-side  duplicates (20/259 = 8%).  With respect to
presence  or  absence  of  asbestos,  all  20  of  the side-by-side
duplicates  agreed  with their  original samples  (100% agreement).


7.2.2  External Analyses

          From the 279 bulk samples, 31 were randomly chosen for
external analysis  by  a  second laboratory (31/279  =  11%).   With
respect  to the presence  or  absence of  asbestos,  30  of  the 31
externals   agreed  with   their   original  samples   (30/31  =  97%
agreement).     The  one  disagreement  did  not  result  in  the
misclassification   of   the  building  category  as  verified  by
additional  bulk samples  collected  at  that  area by  the rating
team.

          The  primary analytical  laboratory also  performed its
own QC  checks.  The laboratory  participated in the EPA Asbestos
Bulk Sample Analysis Quality Assurance  Program.   Sixteen of the
EPA bulk QA samples  (for which the "true" percentage of asbestos
is known) were  submitted as blind QC samples along with the field
samples.   With respect  to the presence or  absence of asbestos,
all  16   of   these  results  agreed   with  the   original  EPA
determination  (100%  agreement).
                                48

-------
7.2.3  Replicate Analyses

          The analytical laboratory randomly chose  33  of the 279
field-collected samples  and  resubmitted them for replicate bulk
sample analysis (33/279  = 12%),  With  respect to the presence or
absence  of  asbestos,  32 of the  33 replicates  agreed with  the
original result  (32/33 - 97%  agreement).   The  one disagreement
did not  result in the  misclassification  of  the  building category
because that building's  classification was  based on another area
where  several  samples  were  used  to  verify  the  presence  of
asbestos.


7.3  Building Classification

          The goal  of  the building inspector was to  verify  the
classification of  the  Category 1  and Category  2  buildings.   A
building's  final  classification as  Category 1  depended on  GSA
records and the bulk sampling  results.   A Category  1 building is
defined as one in which  no  friable  asbestos-containing surfacing
materials or TSI were noted in the GSA records and none was found
during  the building inspection.    Asbestos  was defined  to  be
present  in  a  bulk  sample if the PLM analysis found greater than
1%  asbestos.   Absence of asbestos was defined  as a  result  of
"none detected" or  "trace".  Using  these definitions,  6 of the 7
potential Category 1 buildings remained in that  category. Because
of the presence of asbestos in the remaining building, it was not
included in the study.

          In  Category  2  buildings,   the  building  inspector
searched for  and bulk  sampled  any friable surfacing material and
TSI that were indicated in  the GSA building records  to contain
damaged ACM and any areas in worse  condition.   The damaged areas
found in these buildings were  subsequently  visited by the  rating
team.   The functions  of the  rating  team,  in addition to field
testing   the   assessment  method,  were  to  help  verify  the
classification  of  buildings  in  Category  2 and  3  before  air
sampling.   Based on the assessments  of the rating team and the
bulk sample results, the final classification of Category 2 and  3
buildings was achieved.

          To assess  the  condition of the ACM in a given area, the
rating team used definitions from USEPA (1986a)  for the various
conditions:   good  condition,   moderate  damage,  and  significant
damage.   The  definitions of  these  condition  categories  for  a
given area  are found in  Appendix  C.   Only the assessments  of the
rating team leader  (Core Rater 1,  the most experienced member of
the team) were used for  the classification of building category.
A  building  was classified as  Category 2 if  all or  most of the
areas with friable asbestos-containing surfacing materials  or TSI
were in  good  condition with  perhaps a few areas (less than four)
of  moderate damage.   Recall  that  the  rating  team  visited the

                                49

-------
areas that  were thought to  be the most  damaged  ("most damaged"
can  mean   good  condition   if   there  are  no   moderately  or
significantly damaged  ACM  areas in the building),  and they bulk
sampled those areas  as well  as any areas  in worse condition.  If
an  area  contained greater  than 1% asbestos  in the  bulk sample
results  and the  condition  was  rated as  significantly damaged,
then  the   building   would   be  reclassified  as   Category  3.
Similarly,  a number  of moderately damaged areas  with positive
bulk sample  results  in one of these  buildings  would exclude the
building from Category 2.

          The areas in the Category 3 buildings were chosen based
on the GSA building records.  Since the rating team assessed what
was thought  to  be the  most  damaged ACM area in each building and
bulk  sampled  those  and  any  areas   in   worse  condition,  this
information  was  used  to  verify  the  condition  of  Category  3
buildings for  air sampling.   A building's  final classification
was defined as Category 3 if there was at  least one significantly
damaged area of friable asbestos-containing surfacing material or
TSI, or there were numerous moderately damaged areas.  Thus, most
buildings placed  in  Category 3 were found to have positive  (>1%)
bulk sample  results  in areas  with damaged ACM in the building.
However,  in  4   buildings   (1  in  Study  Region  2,  1 in  Study
Region 3, 2  in  Study Region 4),  positive  confirmation of damaged
ACM within  the  building was not obtained.   These buildings were
classified  as  Category  3  because  GSA   records   indicated  the
presence of damaged ACM within the buildings.

          Forty-nine  of the  initial  67  buildings  were selected
for air monitoring.   Table  7-1 shows  the  final classification of
the 49  buildings by region.   One of  the remaining 18 buildings
was not used for  air sampling because  it was initially classified
as Category  1,  but was found to contain asbestos.  The remaining
17  buildings   were    excluded  from   air  sampling  because  of
geographic location  (related to sampling logistics), uncertainty
about classification as a Category 2 or   3 building,  or,  in the
case of one  building,  because the ACM had been removed since the
GSA records had been reviewed.
                                50

-------
                                                         PLEASE RETURN TO?
                                                      NCIC/OTS CHEMICAL LIBRARY
                                                        401  M ST., S.W., TS-793
                                                        WASHINGTON, D.C. 20460
Table  7-1.  Final Classification of Buildings  in Categories 1, 2,
             and 3 in Each  Study Region
Number of Number of
Study Category 1 Category 2
region buildings buildings
1 2 2
2 1 1
300
4 21
512
Total : "6 ~6
Number of
Category 3
buildings
5
8
10
7
7
17
Note:     Only  the  49  buildings  chosen  for  air   sampling  were
classified.
                                   51

-------
8.0  AIR MONITORING

8.1  FIELD METHODS

          Air  samples  were  collected  in  49  buildings  (9  in
Study Region 1, 10 in each  of  Study  Regions 2  through 5).   Seven
areas were sampled inside each building and one  area was sampled
outside.    (One  building,  Building  44,  could  not  be  sampled
outside and one building,  Building 18,  could be sampled at  only
six indoor sites.)  About half of the sampling pumps were located
near the most damaged ACM  in  the building.   "Most  damaged"  could
mean  ACM  in  good  condition  if  there  were  no  moderately  or
significantly damaged materials  in the  building.   The  rest  were
placed  in  public  areas nearby.   Table  8-1  shows the  five  Study
Regions, period of air  sampling,  and the number  of  buildings and
sites at each  building.  A more  detailed description  of the air
sampling field methods appears in Appendix D.

          The  air  samples were  collected  on  cellulose  ester
(Millipore) filters.  Two side-by-side samples were drawn at each
site:   one of 5,000 L  and one of 2,500 L.   If the  high volume
sample  contained  too  much debris  for  analysis,   then the  low
volume  sample was available for  analysis.  Each  pump ran for two
periods of approximately eight hours during  consecutive weekdays
during  normal building  activity.  The volume of air sampled was
determined from the flow  rate readings  taken at  the beginning,
during, and at the  end of the sampling period.  One field blank
filter  was collected  in each  building to  check for contamination
from sources other than the sampled air.

          An  independent  auditor  accompanied  the   field crew to
each  study region to ensure  that all procedures  were  followed.
After  sampling was completed in  each  region,  the  filters  were
hand  carried  to  the   laboratory and analyzed  by transmission
electron microscopy (TEM).


8.2  AIR SAMPLE ANALYSIS

          The TEM protocol is given in the Quality Assurance Plan
(Hatfield et al.  1987).  A summary appears  in Appendix E of this
report.    Sample  preparation involved  collapsing the filter,
plasma  etching, and directly coating the filter with a thin layer
of carbon  by evaporative  deposition under  vacuum.   The samples
were cleared with acetone,  leaving the  particles attached to the
carbon  film.   The  samples were  analyzed  at a  magnification of
20,OOOX.   A minimum of ten grid  openings with  a  total area of
0.062 eg.  mm  were examined on each  filter.   The total  structure
count   includes  asbestos  fibers  (structures  with  essentially
parallel  sides and an aspect   ratio  of  3:1  or   greater)  and
asbestos bundles, clusters, and  matrices  as defined  in the TEM


                               53

-------
Table 8-1.  Period of Air Sampling Within Each Study Region
Study
region
1
2
3
4
5
Period of
air sampling
April 12-17
May 6-12
May 20-28
June 10-16
June 21-26
Number
Category 1
2(16)
1(8)
0
2(16)
1(8)
of buildinqs
Category 2
2(16)
1(8)
0
1(8)
2(16)
(sites)
Category 3
5(40)
8(63)
10(80)
7(56)
7(55)
Note:   The number  of  sites within buildings,  including outdoor
sites, is given in parentheses.)
protocol.   Two samples from Building 43  were  too heavily loaded
to be analyzed with the direct TEM method.


8.3  AIR SAMPLE AND TRANSMISSION ELECTRON MICROSCOPY
     QUALITY ASSURANCE

          Chain-of-custody  procedures were  implemented  for  all
air  samples   collected   during  the  project.    Field  custody
procedures  were  used  to document the existence  of  a sample from
the  time   of  collection  until  received  by  the  analytical
laboratory.  At this point, internal laboratory records were used
to  document   the  custody   of  the   sample  through  its  final
disposition.

          Standard sample custody  (traceability)  procedures were
used during this project.   Each sample  was labeled with a unique
random identification  number  immediately  after collection.   This
number  was  recorded   on  the  field  data  form  along  with  the
following information:

          •    Name(s) of the sampler;

          •    Date of collection;

          •    Sample location;


                                54

-------
          •    Sketch of location; and

          •    Comments.

          A chain-of-custody form was filled out in the field for
all air  samples.   A copy  of the  form was  included with  each
shipment of samples to the analytical laboratory.   Figure  5-4 is
a representative  copy of 'the  chain-of-custody form  used  during
the project.

          Specific QA procedures used to ensure  the accuracy and
precision of the  air  sample  collection  and  TEM analyses  included
the collection of production lot  and field  blanks,  field audits,
laboratory audits,  replicate  and external  analyses,  and a study
to further evaluate the results obtained by the TEM method.

          Production  lot  blanks  are filters chosen prior  to the
start  of  field   work.    They  are  analyzed  by  the  analytical
laboratory to check for filter contamination.   Field blanks are
filters  taken  into  the field and  handled  in the same manner as
exposed  air sample filters.   Their  purpose  is  to check  for
contamination which might occur in  the  field  but not as  a result
of air sampling.   Field audits determine whether the field team
is  following  set procedures.   Laboratory  audits  determine the
same  for the  analytical  laboratory personnel.    Replicate and
external  analyses  serve  the   same purpose  as  discussed  in
Section 7 for PLM analyses.


8.3.1  Production Lot Blanks

          Blank  filters  from  prescreened  production lots  were
randomly  selected  three  times  during the   project:     at  the
beginning   of   field  activities,   in   the  middle,   and  near
completion.    Each  time,  two  filter  cassettes  were  randomly
selected from  a  previously unopened box of 50  filters.   A total
of  26  production  lot  blanks  were  selected  in  this  way for
analysis.   The analysis  of  the production lot  blanks indicated
that   there   was  not   a   problem   with  background  filter
contamination.


8.3.2  Field Blanks

          During the pump set-up, preloaded filter cassettes were
selected as field blanks.  These filters were labeled and  handled
in an  identical  manner as  were the sample  filters,  except that
they were  not attached to  the  sampling pump.   The filters were
capped during  active  sampling periods  and  open faced during the
non-run  hours  when the actual sample  cassettes were  also open
faced.   Field  blanks  were  collected  in  30  of  the buildings
sampled.    The  purpose  of  the  field blanks  was  to   measure

                                55

-------
contamination  which might  occur during  periods when  the pumps
were not running.

          Of the  30 field blanks collected, 19 were selected for
analysis.   If a  high  level of contamination  was  found from the
analysis results  of the 19  blanks,  the remaining 11 blanks would
have been analyzed.  The  19 blanks that were analyzed were chosen
at  random  from  the  30  blanks  collected.   No  structures  were
detected  in 18  of the  19  field blanks  that  were analyzed.   A
single  fiber was  counted on the  remaining blank.   This level of
blank contamination corresponds to an airborne asbestos structure
concentration  of  0.00015  s/cc when 5,000 L of  air is collected, a
very low  level of contamination.  Thus,  it was  not necessary to
analyze the remaining blanks.


8.3.3  Flow Rate  Calibration

          All  data collected in  the  field were  transcribed from
the field  data sheets onto a computer  disk.   Flow readings were
corrected to standard temperature and pressure (STP) via internal
calculations built into the computer spreadsheet.

          Since the flow  rate was controlled by  limiting orifices
and no  adjustment  could  have  been made to the  diaphragm vacuum
pump, the  equipment limited to  calibration were the rotameters,
barometric gauges,  and thermometers.  Two  rotameters of differing
capacity  were  used  to  measure  the  flow  rate  under  field
conditions.    A  0  to  5  Lpm  rotameter was used to  monitor the
2.5 Lpm limiting orifice side,  and  a 0 to 20 Lpm rotameter was
used  on the   5  Lpm orifice  side.    The  procedure  to calibrate
rotameters  to  STP  used a bubble tube as  a secondary standard.
The procedure  is  described in USEPA  (1977).

          Using  STP,  a calibration curve  was  developed for each
rotameter.   Upon  return  from  the field,  the  recordings  made in
the field were compared to the calibration curve, and a STP flow
was  achieved.    The  STP flow  was  then  recalculated  using the
computer to  finalize the flow-to-field operating conditions.   A
random  set   of   final   flow  volumes   was  recalculated  as  a
confirmation  check.   Final  flows,   their matching  random I.D.
numbers and locations were  tabulated and  matched  by a computer
with the TEM results from the electron microscopy laboratory.


8.3.4  Field Audits

          Five field  audits were  conducted  by an  independent
field auditor, one audit  in each  of the study  regions.  The field
auditor accompanied the  field  crew during pump set-up in several
buildings*per  study region.  He checked to be  sure that the field
crew  was   following the guidelines  set  forth   in the  Quality

                                56

-------
Assurance  Plan  (Hatfield  et  al.  1987),  and  documented  any
violations  in  procedures  so  they could be  corrected.    For
example, an air hose on one pump was found to be punctured.  This
was  noted and  immediately corrected.   The  field auditor  also
measured  216  flow rates in pumps  in these buildings.   This was
done  in  order  to estimate  the relative  accuracy  of the  flow
rates,  defined as  [(field value-audit  value)/(audit value)]  x
100.  The percentage of flow rates  within + 20% relative accuracy
was 99%.


8.3.5  Laboratory Audits

          To ensure the accuracy of the air sample analyses using
TEM,  two  laboratory  audits  were  performed.    An  independent
laboratory  auditor  visited  the  TEM  analytical  laboratory  to
verify  that  all  procedures  specified in  the Quality Assurance
Plan  (Hatfield et  al.  1987)  were  followed.    Two  audits  were
conducted, one  at the beginning of the  analyses and  one  at the
end.  The auditor concluded that:

          •    The  sample  identification was   traceable  from
               sample  acceptance  through  preparation,  analysis
               and reporting.

          •    The  sample  preparation  was  done  according  to
               protocol,  with   the exception   of  the  use  of
               scissors  instead of a scalpel for  the cutting of
               filters.  .

          •    The TEM analysis was done according to protocol.

          •    The  reporting  procedure was  implemented properly
               and was accurate.


8.3.6  Replicate and External Analyses

          Twenty  air  samples, four  from each  study region, were
selected  at  random to investigate within  and between laboratory
performance.    The  samples  were  reanalyzed  by  the  original
laboratory   (replicate  analysis)   and  by a  second laboratory
(external  analysis).    These  40 QC analyses increase the  total
number  of analyses by just over 10%.  Within each study  region,
three samples  were selected at  random from sites within Category
2  and  3  buildings.   The  fourth   sample  was  selected  from  the
outdoor and  Category  I sites.   The samples were receded to  avoid
analyst bias in the replicate analysis.

          Table   8-2   presents   the  results  of  the   original,
replicate, and external analyses.  No  asbestos structures were


                                57

-------
Table 8-2.
Comparison   of   Airborne  Asbestos   Concentrations
Estimated   by    the   Original,    Replicate   (Same
Laboratory), and  External (Different Laboratory)  TEM
Analysis
Original
Number of
structures
0
1
0
0
1
0
0
0
I
0
0
0
0
0
0
0
0
0
0
0
s/cc
0.000
0.001
0.000
0.000
0.003
0.000
0.000
0.000
0.003
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
Replicate
Number of
structures
0
0
0
0
0
0
0
0
1
1
0
0
0
0
0
1
1
0
0
0
s/cc
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.003
0.003
0.000
0.000
0.000
0.000
0.000
0.002
0.003
0.000
0.000
0.000
External
Number of
structures
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
s/cc
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.002
0.000
0.000
                               58

-------
detected on  13 of  the 20  filters.    One or more  of the  three
analyses detected  a  single  structure  on the  remaining  seven
filters.

          There is  no  evidence of inconsistency among the  three
sets of analyses.   A Wilcoxon signed-rank test  (Sokal and  Rohlf
1969)  did  not  detect any  significant  tendency  for  any  one
analysis (original,  replicate, or  external)  to  give higher  or
lower structure counts than any other.

          A confidence interval for the  mean  structure count can
be  constructed by  assuming a statistical  distribution  for  the
counts.  A Poisson distribution gives a  95%  confidence  interval
of  (0,  3.0)  when  zero structures  are counted.   Therefore,  an
observation of zero structures is not inconsistent with a mean of
3  structures,  indicating  that  the  differences  of  a  single
structure between analyses  are not  significant.   This conclusion
is even stronger  if structure counts  follow  a  negative  binomial
distribution.  The confidence interval for a negative binomial is
wider than the corresponding interval for a Poisson distribution.


8.3.7  Examination of Additional Grid Openings

          Sixteen  air  samples  were  selected  for  additional
analysis to  determine if the  10  grid openings specified by the
TEM  protocol  provide estimates of  sufficient precision  for the
purposes of the  study.   The  16 samples  were  selected as follows
to provide a range of structure counts:

          •    4 "indoor" samples which had structure counts of 3
               or more in the first 10 grid openings counted;

          •    8 "indoor" samples which had structure counts of 0
               in the first 10 grid openings;

          •    2 "outdoor" samples;  and

          •    2 field blanks.

          An additional 40  grid openings, giving a  total of 50,
were examined on each sample and the number of structures in each
opening recorded.

          The  precision of  the TEM analysis  was investigated by
fitting  a  negative  binomial  distribution  to  the  number  of
asbestos structures per grid opening.  The negative binomial is  a
discrete distribution which is often  used to describe clumped or
aggregated populations.   Javitz  and Fowler (1981) found that the
negative binomial  was  superior  to  the  Poisson  for describing
asbestos structure  counts obtained by electron microscopy.  The
variance of  the  negative binomial is  m (m +  k)/k where  m  is the

                                59

-------
mean  and k  is a  measure of aggregation.   As  k increases, the
variance  decreases and,  consequently, the precision of estimated
airborne   asbestos  concentrations   increases.     The  Poisson
distribution  is a  limiting case  of  the negative  binomial  as k
becomes very  large.

          The parameter k  was  estimated for each  filter with a
non-zero  structure  count.     Given   k,  the   precision  of  the
structure count can be determined as a function  of the number of
grid  openings counted.   (Details  are  given in Appendix F.)

          Estimates  of  the mean  number  of structures  per grid
opening  (m),   the  variance, and  k for each filter  are given in
Table 8-3.    Estimates  of k equal to infinity  indicate that the
variance  does  not exceed  the  mean  and  that   a  Poisson  or more
uniform   distribution   is  more  appropriate  than  the   negative
binomial.   This  implies  a small variance and hence increased
precision.

          No  asbestos  structures were counted  on eight of the 16
filters.   The eight include  the two  field blanks  and  the two
outdoor samples.   Of the eight  filters  with non-zero  counts, five
have  estimates  of  k  equal to infinity.   The  remaining three
estimates of  k are 0.6,  0.4, and  0.07.

          For k equal  to infinity, i.e., a Poisson distribution,
a  95% confidence  interval  for  the true  structure  count when no
structures are counted  in 10 grid openings is (0, 3.0).   The size
of  the confidence  interval increases slightly to  (0,  3.1)  as k
decreases to  0.4.   Thus,  for values of  k  greater than or  equal to
0.4 the  examination of  10  grid openings  in this study yields an
airborne  asbestos  concentration  that is  sufficiently precise to
distinguish  0 s/cc  from 0.009  s/cc  with high  probability.    (In
this   study    one  structure    corresponds    to  approximately
0.003  s/cc.)

          The data  in  Table  8-3 indicate  that  k  is  usually
greater than  0.4,  but that smaller values,  such as k = 0.07 are
possible.  The standard deviation of this  estimate of k is 0.9.
For k =  0.07, a 95% confidence interval for the true structure
count  when no structures  are  counted  in  10  grid  openings  is
(0, 50).  If  the number of grid openings counted is  increased to
50, the confidence interval shrinks to  (0, 4.7).

          Of  the 16 filters examined,  all  but  one indicate that
examination  of  10  grid  openings is sufficient  to distinguish
0 s/cc from 0.009  s/cc with high probability.   Although, without
additional  data,  it  is  difficult  to  predict how frequently
exceptions will occur,  the results  suggest that examination of
additional grid openings is generally unnecessary unless higher
precision*is  required.


                                60

-------
   Table 8-3.
     Estimated Mean, Variance,  and Value of k  for the
     Number  of  Structures  Counted  Per  Grid  Opening
     Based on Examination of 50 Openings
   Type of filter
                  Openings
                    1-10
                   Mean
Mean
                                                Openings
                                                 1-50
Var
Indoor with 3
or more structures
in openings
1-10
Indoor with
0 structures in
openings 1-10





Outdoor

Field blank

1.1
0.3
0.4
0.3
0
0
0
0
0
0
0
0
0
0
0
0
0.36
0.18
0.12
0.1
0.02
0.02
0
0
0
0
0.02
0.04
0
0
0
0
0.52
0.27
0.11
0.21
0.02
0.02
0
0
0
0
0.02
0.04
0
0
0
0
0.60
0.41
00
0.07
00
00
-
—
—
—
00
00
—
-
^
—
8.4  ANALYSIS OF AIR MONITORING DATA

8.4.1  Methods

          For each of the 387 air samples collected and analyzed,
an estimate of  airborne  asbestos  concentration,  c,  in structures
per cubic centimeter (s/cc) is given by:
where
           c = (ns * A/a)/V


ns - the   number   of   structures  counted  in   the
     microscope;

A = the effective area of the filter;
                                61

-------
          a = the area of  filter examined; and

          V = the volume of air collected  (in cubic centimeters).

The  area of  filter  examined is  calculated by  multiplying the
number  of grid openings by the area  of one grid  opening.   The
analytical  sensitivity is  the  smallest value, other  than zero,
that  c  can  take.   It corresponds  to  the observation of a single
structure  and depends  on  the  values of  A,  a and V.   In this
study,   the   analytical   sensitivity  for   most  samples   is
approximately 0.003 s/cc.

          Individual  buildings are  the basic  statistical units
for comparisons between different  building categories  and between
indoor  and  outdoor levels.   The airborne  asbestos level in each
building  was  estimated  by the arithmetic mean  of  the samples.
The  distribution  of  these  building averages  was plotted  and
summary  statistics   (percentiles,   mean,  standard  deviation)
calculated  for  each  building  category  and  for  the  outdoor
measurements.

          Differences   of   distributions  of  airborne  asbestos
levels  between building categories are indicated  by the plots and
tables  of summary  statistics.  A statistical test was applied to
provide a  quantitative  measure  of  the  strength of  evidence
associated  with observed  differences  (i.e.,  probabilities that
the  observed  differences  may have occurred only by chance were
estimated).   A "p-value,"  the level of  significance,  is reported
for each comparison.  The  p-value  is  the probability of obtaining
a  difference  as  great or greater than  the difference observed
under the hypothesis that  no true difference exists  between the
categories  being compared.   A small  p-value  indicates that the
magnitude  of  the  observed  difference  is  unlikely   under  the
hypothesis of  no  true difference,  and therefore  lends support to
the alternative hypothesis, namely that the  difference is real.

          A permutation (also referred to as randomization) test
(Cox  and  Hinckley  1974,   Section 6.2)   was  used  to  test  for
differences between  building categories and outdoors.    (See also
Edgington 1987 for a discussion of randomization  tests.)  Medians
were  chosen  to represent  the distributions of airborne asbestos
levels.   The  median is  appropriate  for summarizing the location
of this type  of data because it gives equal weight to large and
small data values  and is not unduly influenced by a small number
of  extreme  values.    The   permutation  approach,  rather  than
analysis  of  variance or Student's t-test, was  used because the
data contained a large  number of  zero observations.   In previous
air monitoring  studies, where the majority of measured airborne
asbestos   concentrations   were   greater   than   zero,   a   log
transformation was  used to equalize  variances  prior to analysis
by standard analysis  of variance  techniques (USEPA 1985b, 1986b;
Tuckfield et al. 1987).

                                62

-------
          Under the null hypothesis of the permutation test,  each
of  the  387  airborne  asbestos  measurements  is  an  independent
observation from  the  same probability distribution.   Therefore,
every possible permutation  of  the 387 values is  equally likely.
The distribution  of any  given  statistic  (e.g.,  the  difference
between two building category medians) under  the  null hypothesis
can be determined by  calculating the value of the  statistic for
each  possible  permutation  and  tabulating  the  frequency  of
occurrence of  each value.    The  enumeration of  every  possible
permutation is  impractical for  large data sets.    Instead,  the
distribution of the statistic is based on  a random sample of all
possible permutations.  The precision of  any estimated percentile
is determined by the size of the random sample.

          A  random  sample of   1,000  permutations  of  the  387
airborne   asbestos  measurements  was   used   to   estimate   the
distribution of   the  differences  between  building  category  and
outdoor  medians   under the  null  hypothesis  that   all  airborne
asbestos measurements  are  independent observations  from the same
probability distribution.   For  each permutation,  the difference
between  medians   was   calculated  in  the  same  way as  for  the
original   data,   i.e.,  measurements  were   averaged   within  a
building,  the median  of each building category and outdoors was
determined, and the difference between medians  calculated.  With
1,000  replications,  the  width  of  an  approximate  confidence
interval for the  95th percentile is 0.02.  The p-value for each
observed   difference   is  read   directly  from   the  estimated
distribution.       For   example,   if  an   observed  difference
corresponded to the 97th percentile of the distribution estimated
under the null hypothesis, the p-value would be 0.03.

          The  p-values obtained in this  way do not  take  into
account  the  fact that   several  comparisons   are  being  made
simultaneously.   Although the reported  p-values may  be smaller
than a p-value  obtained from the joint  probability distribution
governing  all  comparisons  simultaneously, the  reported  values
indicate which observed differences  are  most  consistent with the
null hypothesis of no  true difference, and which provide support
for the alternative hypothesis that a true difference exists.

          The  permutation method can also  be  used  to compare
different  building categories using  indoor-outdoor differences.
This can be thought of as adjusting  each  indoor  measurement for
outdoor  levels by subtracting the outdoor measurement.   A test
using  adjusted data  will  be more  powerful  if   the adjustment
reduces  the variance  of the  data.  As  will  be  seen below with
this particular  data  set, all but 7 of the  outdoor values were
zero.  For most of the data points,  adjusting for outdoor levels
has no effect.  For the 7  buildings with non-zero outdoor levels
(one structure  observed), the adjustment results  in a negative
building mean  because  none of  these buildings  has  an  average

                                63

-------
indoor  level  as high as one  structure  per  sample.   The adjusted
data  have  a higher variance  than the unadjusted data.   For this
data  set,  adjusting for outdoor levels conveys  no advantage and
results  in a  less powerful  statistical test.   Therefore,  this
approach was not used.


8.4.2  Results

          The estimated  airborne  asbestos concentration for each
sample  is  given  in  Appendix 6.   No  asbestos  structures  were
detected in 83% of the 387 samples.  For one sample from Building
23, the TEM result,  a  zero  structure  count,  was reported after
the statistical analysis was  completed.  This  388th sample is not
included in any of the following tables or analyses.  The maximum
number  of  structures   counted  on  a  single  sample  was  11,
corresponding to an  airborne asbestos  concentration at that site
of  0.033  s/cc.   This  sample was collected in Building 2, which
was categorized as containing no ACM  (Category  1).   No asbestos
structures  were observed  on five  of  the remaining  six filters
collected   in   the  building;  the remaining  filter  had  one
structure.   Neither  GSA,  nor  the building   inspector  for  this
study,  identified ACM  despite  thorough  inspections  of  the
building  and analyses  of  bulk samples.   The  reason  for  this
unusually  high value  is  unknown.    (The next  highest airborne
asbestos  concentration at  a single  site is  0.013 s/cc.)   The
source of asbestos structures could not be found.

          Figure   8-1  presents  scatter  plots   of  the  average
airborne  asbestos concentration  in each building,  by building
category, and for  the 48 individual outdoor samples.  The medians
of  each category are also shown.   With the  exception of the one
Category  1  building,  the   highest  average  airborne  asbestos
concentrations  occur  in buildings  from Category 3.   However, in
27% of  the buildings in Category  3,  no asbestos structures were
detected  at  the  indoor  sites.    (No  asbestos  structures  were
detected in 16% of the Category 2 buildings,   50% of the Category
1 buildings, and 85% of the  outdoor sites.)

          The   medians   and  arithmetic  means  of  the  average
airborne  asbestos  concentrations for each  building category and
outdoor samples  are reported in Table 8-4.   Outdoor samples have
the  smallest  median,  followed by buildings  from  Category  1,
buildings   from   Category   2,   and    finally,   buildings  from
Category 3.    The  arithmetic  mean  of  Category  1   is  greatly
influenced  by  the one  unusually   high  value  in  that category,
although the absolute magnitude of the mean is still very small.
The  other means  follow  the same  trend  as   the medians.   The
outdoor means by building category, 0.00043, 0.00048, and 0.00036
s/cc  for% Categories 1, 2,  and  3,   respectively,  do  not show any
apparent  trend.    Therefore,  the  trend  in  indoor means  with


                                64

-------
Ui
o.oos
s
T
R
U
0.004
T
U
R
E
0.003
C
0
N
C
E 0002
N
T
R
A
T
j 0001
0
N
8/ec
0.000







A

A
A
A
fl







Z o
1 1
Outdoor
(48 sites)
                                                            A


                                                            A
                                                            ft   *
                                                            A	
                                                                            A
                                                                            A
                                                                            A
A
A
B
e
B
                                          Building         Building         Building
                                         category 1       category 2        category 3
                                        (6 buildings)     (6 buildings)    (37 buildings)
                  Figure 8-1.  Scatter plots*and medians of the average  airborne
                                asbestos structure concentrations for each  building
                                category and outdoors.

             *The  data  points  for each  scatter plot  are the  average  concentration
             within  a building  (for  indoor samples)  or the concentration  outside each
             building (for  outdoor samples).  A=l  data point,  B=2  data  points,  ...,
             j=10 data points,  and Z=41 data points.   The  diamond represents the median
             of the  data points in each scatter plot.

-------
 Table 8-4.   Summary  Statistics  for   Average  Airborne  Asbestos
             Structure Concentrations  (s/cc)
 Statistic
 Standard
  deviation
                                                    ACM
Outdoor
Category 1
Category 2
0.00096
 0.00198
 0.00052
Category 3
Median
Mean
Sample
size
<0. 00001
0.00039
48
(sites)
0.00010
0.00099
6
(buildings)
0.00040
0.00059
6
(buildings)
0.00058
0.00073
37
(buildings)
 0.00072
Notes:

           1.    The data  points  used  in  the calculation  of each
statistic  are  the average  concentration  within a  building (for
indoor  samples)  or  the  concentration  outside each  building (for
outdoor  samples).

           2.   The mean for Category 1 is heavily influenced by one
sample  in one  building  which  produced an  unexplained large s/cc
value.     The  Category  1  mean,   excluding  this  one value,  is
0.00020  s/cc.
building  category cannot  be attributed to  differences in outdoor
airborne  asbestos concentrations.

          The   analytical   sensitivity   of  a   given   mean  is
approximately the analytical sensitivity of  a single sample divided
by  the number of air  samples used to  calculate  that  mean.    Thus,
the  analytical  sensitivity  of  a  building  mean  is  approximately
0.0004  s/cc  (0.003/7).   The analytical sensitivity of  the Category
3 mean is approximately 0.00001  s/cc  (0.003/255).   Similarly, the
analytical sensitivity  for Category 1  and 2 means is approximately
0.00007  s/cc  (0.003/42).    For  the outdoor  mean the analytical
sensitivity is approximately 0.00006 s/cc  (0.003/48).

          The  results  of   the  permutation test  are listed  in
Table 8-5,.    The  difference between  Category  3  buildings  and
Category  1 buildings has the smallest p-value (p < 0.02).  The next
smallest  p-value  is  obtained for  the  comparison between buildings

                                66

-------
Table 8-5.  Results   of   Randomization   Test   Indicating   the
            Statistical P-Values  for Differences between  Median
            Airborne Asbestos Concentrations in Each of the Three
            Building Categories and Outdoor Concentrations
                           Difference between
  Comparison                 medians (s/cc)             P-valuea
Category 1 versus               0.00010                p < 0.96
  outdoor

Category 2 versus               0.00040                p < 0.65
  outdoor

Category 3 versus               0.00058                p < 0.09b
  outdoor

Category 2 versus               0.00030                p < 0.21
  Category 1

Category 3 versus               0.00048                p < 0.02
  Category 1

Category 3 versus               0.00018                p < 0.18
  Category 2


          Probability  of  obtaining a  difference  as great  or
greater  than  the  difference  observed when  there  are  no  real
differences  among  building  categories   or  between   a  building
category and the outdoor air.

           This p-value is based on 2,000 replications to provide
additional precision.
in  Category 3 and  outdoors  (p  < 0.09).   The p-values  for the
remaining comparisons  are 0.18 or greater.   Estimates of indoor
asbestos levels are more precise than estimates of outdoor levels
because indoor levels  are based  on several samples per building.
Thus,  an  observed  difference  between  two  building  category
medians corresponds  to a smaller p-value  than  the same observed
difference  between  a  building category median and  the  outdoor
median.
                                67

-------
REFERENCES

Bishop,   YMM,   Fienberg  SB,  Holland   PW.     1980.    Discrete
raultivariate analysis:  Theory and practice.  Cambridge, MA:  MIT
Press.

Cox  DR,  Hinckley  DV.   1974.   Theoretical  Statistics.   London:
Chapman and Hall.

Deming WE.    1950.   Some  Theory of  Sampling.   New  York:   John
Wiley and Sons.

Edgington,  ES.   1987.   Randomization Tests  (2nd  edition).   New
York:  Marcel Dekker,  Inc.

Javitz HS, Fowler  DP.   1981.  Statistical analysis of microscopic
counting  data,  in  "Electron  Microscopy  and X-ray Applications,"
Russell PA  (ed.),  Ann  Arbor  Science.

Hatfield  J,  Leczynski B,  Chesson  J et  al.   1987.   Battelle
Columbus  Division.    Public buildings  study  quality  assurance
plan.     Final   report.    Washington,  DC:    Office  of  Toxic
Substances,  U.S.  Environmental Protection  Agency.   Contract No.
68-02-4243.

Miller,  R6.    1981.    Simultaneous  statistical  inference  (2nd
edition).  New York:   Springer-Verlag.

Rogers,  J.    1987.   Westat, Inc.   Additional  analysis  of data
collected  in  the  asbestos  in  buildings  survey.    Draft  final
report.   Washington,   D.C.:   Office  of Toxic  Substances,  U.S.
Environmental Protection Agency.  Contract  No.  68-02-4243.

Sokal, RR,  Rohlf,  FJ.   1969.   Biometry.    San  Francisco:   W.H.
Freeman.

Tuckfield  RC,  Chesson J,  Tsay  Y-L,  et  al.   1987.   Battelle
Columbus  Division.  Evaluation  of  asbestos abatement techniques
phase  III:    removal.    Draft final  report.   Washington,  D.C.:
Office of Toxic Substances, U.S.  Environmental Protection Agency.
Contract No. 68-02-4243.

USEPA.   1977.   U.S.  Environmental  Protection Agency.   Quality
assurance handbook for air pollution measurement systems, volume
II -  ambient air  specific  methods.   Washington, DC:   Office of
Toxic  Substances,  U.S. Environmental  Protection  Agency.    EPA
600/4-77-027a.

USEPA.  1981.  U.S. Environmental Protection Agency.  Asbestos in
schools..   Washington,  DC:  Office   of  Toxic  Substances,  U.S.
Environmental Protection Agency.  EPA 560/5-81^002.


                                68

-------
USEPA.   1982.   U.S. Environmental  Protection Agency.   Friable
asbestos-containing  materials in  schools:    identification  and
modifications.  Washington, DC:  Office of Toxic Substances, U.S.
Enviornmental Protection Agency.   40 CFR Part 763.

USEPA.   1985a.   U.S. Environmental  Protection Agency.   Asbestos
in buildings:   simplified sampling  scheme  for friable surfacing
materials.   Washington,  DC:   Office  of Toxic  Substances,  U.S.
Environmental Protection Agency.   EPA 560/5-85-030a.

USEPA.  1985b.  U.S. Environmental Protection Agency.  Evaluation
of asbestos abatement techniques phase  I:   removal.  Washington,
DC:   Office  of  Toxic  Substances,  U.S.  Enviornmental Protection
Agency.  EPA 560/5-85-019.

USEPA.   1986a.   U.S. Environmental  Protection Agency.   Guidance
for  assessing and  managing  exposure to asbestos  in buildings.
Draft  Report.    Washington,  D.C.:    Office of  Toxic Substances,
U.S. Environmental Protection Agency.  Contract No. 68-02-4243.

USEPA.  1986b.  U.S. Environmental Protection Agency.  Evaluation
of asbestos  abatement techniques  phase II:   encapsulation with
latex paint.   Washington, DC:  office  of  Toxic substances, U.S.
Environmental Protection Agency.   EPA 560/5-86-016.

Yamate G,  Agarwal SC,  Gibbons RD.   1984.   Methodology  for the
measurement of  airborne asbestos by electron  microscopy.   Draft
report.    Washington,   DC:    Office  of Toxic  Substances,  U.S.
Environmental Protection Agency.   Contract No.  68-02-3266.
                                69

-------
                          APPENDIX A

     RESPONSES OF INDIVIDUAL RATERS IN EACH ASSESSED AREA
WITHIN EACH REGION TO CONDITION, POTENTIAL FOR DISTURBANCE,
                     AND AIR FLOW FACTORS

-------
TABLE A-l. RESPONSE OF RATERS TO OVERALL CONDITION VARIABLE
           SEPARATED BY REGION, BUILDING. AND AREA
           1=GOOD 2=MODERATE DAMAGE 3=SIGNIFICANT DAMAGE

                           REGIONe1
BUILDING

13
13
13
13
13
14
14
14
14
14
IE
16
IS
IE
16
16
16
16
16
17
' 17
17
17
17
Be
50
60
50
61
61
61
61
62
62
52
62
53
53
63
53
64
54
54
54
54
55
56
55
56
56
56
AREA

1
2
4
6
7
1
2
3
4
7
1
2
3
4
1
3
4
7
8
1
2
3
4
5
1
2 .
3
4
1
2
3
4
1
2
3
4
2
3
4
8
1
3
4
5
8
1
2
3
4
1
2
CORE RATER
ONE
1
3
1
3
3
3
1
1
2
3
2
2
3
3
2
3
2
3
2
2
2
2
2
1
2
2
2
2
2
2
3
1
2
2
3
2
2
2
2
2
2
3
2
2
3
1
2
2
3
2
2
CORE RATER
TWO
1
2
1
3
3
3
1
1
2
2
3
2
2
3
2
2
3
2
2
2
2
1
3
2
2
2
2
3
3
1
3
2
2
2
t
3
1
2
2
3
t
•
•
1
2
3
3
.
t
LOCAL RATER
ONE
2
3
2
3
3
3
1
2
2
3
3
3
3
3
2
2
3
2
3
2
2
,
2
2
2
2
2
3
3
1
3
3
3
1
3
3
2
2
2
3
3
3
;
,
.
,


LOCAL RATER
TWO
2
3
2
3
3
3
1
1
3




3 '
3
2
3
2
2
3
3
•
3
3
3
3
3
3
3
1
3
3
3
1
3
3
2
3
3
3
3
3
i
2
3
3
3
3

-------
BUILDING

0
56
66
67
67
67
67
7
7
7
7
7
7
8
8
8
8
AREA


3
4
1
2
4
6
1
2
3
4
7
8
1
2
3
4
CORE RATER
ONE

2
2
2
3
1
2
2
2
1
3
3
3
2
1
2
2
TABLE A-l. RESPONSE OF RATERS TO OVERALL CONDITION VARIABLE
           SEPARATED BY REGION, BUILDING, AND AREA
           IsGOOD 2eMODERATE DAMAGE ^SIGNIFICANT DAMAGE

                           REGIONS1
ISl —
CORE
TWO
















RATER

2
2
3
3
1
1
1
2
1
3
3
3
2
1
1
1
LOCAL RATER
ONE

,
3
2
2
2
2
2
1
3
.
.
2
2
2
2
LOCAL RATER
TWO
3
3
,
.
,
.
.
.
.
.
.
.
.
.
.
.

-------
TABLE A-l. RESPONSE OF RATERS TO OVERALL CONDITION VARIABLE
           SEPARATED BY REGION, BUILDING, AND AREA
           1=GOOD 2=MODERATE DAMAGE 3=SIGNIFICANT DAMAGE
BUILDING

18
18
16
18
18
19
19
19
19
19
19
19
20
20
20
20
20
If
15
21
21
22
22
22
22
23
23
23
23
24
24
24
24
25
25
25
25
9
9
9
9
AREA

1
2
3
4
8
1
2
3
4
7
8
9
1
2
3
4
5
8
]
2
3
4
1
2
3
4
1
2
3
4
1
2
4
5
1
2
3
4
1
2
3
4
CORE RATER
ONE
1
2
2
1
2
1
3
3
2
.
.
.
2
2
1
2
•
2
3
2
2
2
3
1
2
1
2
2
3
2
2
3
2
1
2
1
2
2
1
2
2
CORE RATER
TWO
1
3
2
1
3
2
3
3
2
3
3
2
3
•
3
2
2
2
3
1
2
1
2
2
3
3
(
3
2
1
(
2
3
2
1
2
3
LOCAL RATER
ONE

3
2
2
3
1
3
3
2
3
3
1
2
2
3
3
2
3
3
1
2
1
3
2
3
3
3
3
2
1
2
1
3
2
1
.
3
LOCAL RATER
TWO
1
2
2
1
2
2
3
2
2
3
3
1
2
i
3
2
2
2
3
1
2
1
2
2
3
3
3
3
2
1
2
1
2
1
1
2
3

-------
BUILDING

26
26
26
26
27
27
27
27
27
28
28
28
28
28
28
29
29
29
^ 29
II
30
30
30
30
31
31
31
31
32
32
32
32
32
33
33
33
33
34
34
34
34
35
35
35
35
58
58
58
58
AREA

1
2
3
4
1
2
3
4
8
1
2
3
4
5
6
1
2
3
4
1
3
4
6
8
1
2
3
4
1
2
4
6
9
1
4
5
9
1
2
3
4
1
2
3
4
1
2
3
4
CORE RATER
ONE
3
2
2
2
3
2
2
3
2
2
2
1
1
.
•
2
2
2
2
2
3
2
2
2
3
2
2
2
3
2
2
2
3
3
2
3
2
3
2
2
2
2
1
2
2
2
2
3
2
TABLE A-l. RESPONSE OF RATERS TO OVERALL CONDITION VARIABLE
           SEPARATED BY REGION, BUILDING, AND AREA
           IsGOOD 2=MODERATE DAMAGE 3=SIGNIFICANT DAMAGE

                           REGION=3
CORE RATER
TWO
2
2
2
2
3
1
2
3
3
2
2
2
2
•
2
2
2
1
3
3
2
2
3
3
2
3
3
3
1
2
2
3
3
2
3
2
3
1
2
2
1
1
2
3
2
3
3
2
LOCAL RATER
ONE
2
2
2
1
2
3
2
3
2
1
1
1
1
3
2
2
2
2
,
2
3
2
2
2
2
3
3
2
2
1
2
2
2
2
2
2
1
3
2
1
1
2
2
2
3
3
2
LOCAL RATER
TWO

.
.
,
3
1
2
3
2
3
3
2
1
3
3
3
2
3
.
2
3
3
3
3
3
3













2
1
.
3
3
3
3
3

-------
BUILDING

10
10
10
10
36
36
36
36
36
37
37
37
37
37
38
38
38
38
-J 39
-J 39
39
39
39
40
40
40
40
40
40
41
41
41
41
41
41
41
41
42
42
42
42
59
59
59
59
59
60
60
60
60
61
AREA

1
2
3
6
1
2
3
4
7
1
2
3
5
6
1
2
3
4
1
2
3
4
6
1
2
3
4
7
13
1
2
3
4
5
6
7
8
1
2
3
4
1
2
3
4
6
1
2
3
4
2
CORE RATER
ONE
1
1
1
1
1
2
2
2
3
3
2
1
1
2
1
2
2
1
2
2
1
1
2
2
1
1
1
1
2
2
3
2
2
2
1
2
2
1
2
2
2
2
2
1
2
2
2
1
1
2
1
TABLE A-l. RESPONSE OF RATERS TO OVERALL CONDITION VARIABLE
           SEPARATED BY REGION, BUILDING, AND AREA
           1=GOOD 2=MODERATE DAMAGE SeSIGNIFICANT DAMAGE

                           REGION=4
CORE RATER
TWO
1
1
1
1
1
2
2
2
3
3
2
1
•
2
2
2
2
1
2
2
3
1
2
2
2
1
1
1
2
2
3
2
2
2
1
2
2
1
2
2
2
1
2
1
1
2
2
1
2
2
LOCAL RATER
ONE
1
2
2
2
2
2
2
2
.
3
2
2
3
2
2
,
2
2
•
3
1
2
2
2
1
1
1
•
2
3
2
*
2
2
2
2
1
3
2
2
2
2
2
2
3
2
2
2
2
LOCAL RATER
TWO
1
1
1
1
2
3
2
2
2
3
.
•
2
2
1
2
1
2
2
3
1
1
2
2
1
1
1
3
3
3
2
3
2
2
3
2
1
•
3
3
1
3
1
1
3
1
2
2
1

-------
                              TABLE A-l. RESPONSE OP  RATERS  TO  OVERALL  CONDITION VARIABLE
                                         SEPARATED BY REGION, BUILDING,  AND  AREA
                                         IsGOOD 2=MODERATE DAMAGE  3=SIGNIFICANT DAMAGE
                                                          REGION**
                 BUILDING             AREA       CORE  RATER      CORE RATER     LOCAL RATER     LOCAL RATER
                                                ONE              TWO             ONE             TWO

                 61                      31221
                 61                      42222
                 61                      61121
00

-------
                               TABLE A-I.  RESPONSE OF RATERS TO OVERALL CONDITION VARIABLE
                                          SEPARATED BY REGION,  BUILDING, AND AREA
                                          1=GOOD 2&MODERATE DAMAGE ^SIGNIFICANT DAMAGE

                                                          REGION=6
sO
BUILDING

11
11
11
11
12
12
12
43
A3
43
43
44
44
44
44
46
46
46
46
46
46
46
46
46
47
47
47
47
48
48
48
48
49
49
49
49
62
62
62
63
63
63
63
64
64
64
64
66
66
66
66
AREA

1
2
3
4
1
2
3
1
2
3
6
1
2
3
4
2
3
4
7
2
2
4
6
6
1
2
3
4
1
2
3
4
1
2
3
4
1
7
8
1
2
3
8
2
3
4
7
1
2
6
6
CORE RATER
ONE
1
1
1
2
2
1
1
2
3
1
1
2
3
2
2
2
2
1
2
1
1
2
2
2
2
2
2
3
2
2
3
2
2
1
3
2
2
1
2
2
2
2
2
2
1
2
1
2
2
2
2
CORE RATER
TWO
1
1
1
2
2
1
1
3
2
1
1
2
3
2
2
1
2
2
2
2
2
2
2
2
2
2
1
3
2
2
3
2
2
1
3
3
2
2
1
2
2
2
2
2
1
2
1
2
2
2
1
LOCAL RATER
ONE
1
1
1
2
2
1
2
2
2
1
2

2
2
2
1
1
1
2
1
1
1
2
1
1
1
1
1
1
1
1
1
1
1
2
1
2
2
1
2
1
2
2
2
1
1
1
1
1
1
1
LOCAL RATER
TWO
1
2
1
1
•
,
,
2
3
1
2
.
,
,
,
3
3
2
2
2
2
2
3
3
3
2
3
3
3
2
3
2
3
1
3
3
3
3
2
3
2
3
3
,
,
.
,
2
2
2
1

-------
                              TABLE A-l.  RESPONSE OF RATERS TO OVERALL CONDITION VARIABLE
                                         SEPARATED BY REGION, BUILDING, AND AREA
                                         1=GOOD 2=MODERATE DAMAGE 3=SIGNIFICANT DAMAGE
                                                         REGIONsB
                 BUILDING            AREA      CORE RATER      CORE RATER     LOCAL RATER     LOCAL RATER
                                               ONE             TWO            ONE             TWO

                 66                     11112
                 66                     21112
                 66                     31112
                 66                     42112
                 66                     51112
                 66                     61112
CX)
O

-------
BUILDING

13
13
13
13
13
14
14
14
14
14
16
IB
16
16
16
16
16
16
16
00 17
~ 1?
17
17
17
60
60
60
60
61
61
61
61
62
E2
62
62
63
63
63
63
64
54
64
64
54
55
65
55
55
66
56
AREA

1
2
4
6
7
1
2
3
4
7
1
2
3
4
1
3
4
7
8
1
2
3
4
5
1
2
3
4
1
2
3
4
1
2
3
4
2
3
4
6
1
a
4
5
8
1
2
3
4
1
2
CORE RATER
ONE
1
3
2
3
3
2
2
2
2
2
2
2
3
2
2
3
3
3
3
2
2
3
31
2
3
2
2
2
2
2
2
1
2
2
2
1
3
3
3
2
2
3
3
3
3
2
3
3
2
2
2
TABLE A-2. RESPONSE OF RATERS TO POTENTIAL FOR DISTURBANCE
           SEPARATED BY REGION, BUILDING, AND AREA
           1=LOW 2=MODERATE 3=HIGH POTENTIAL

                           REGION=1
CORE RATER
TWO
2
2
3
2
3
2
1
1
1
3
2
3
1
3
2
2
3
2
2
2
2
2
3
2
2
2
2
3
2
2
3
2
2
2
3
3
2
2
2
3
3
3

2
3
2
2
2
2
LOCAL RATER
ONE
2
3
2
3
3
3
1
2
2
3
3
3
2
3
2
2
3
2
2
3
2
,
3
2
2
3
2
3
3
1
2
3
3
1
3
3
2
2
2
3
3
3

f

t
,
,

LOCAL RATER
TWO
3
3
3
3
3
3
1
)
3

.
.
.
3
3
3
3
3
3
3
3
,
3
3
3
3
3
3
3
3
3
3
3
2
3
3
3
3
3
3
3
3
1
2
3
3
3
3
3

-------
BUILDING

» 56
66
67
67
67
67
7
7
7
7
7
7
8
8
8
8
AREA

3
A
I
2
4
6
1
2
3
4
7
8
1
2
3
4
CORE RATER
ONE
2
2
2
2
3
2
2
3
2
3
3
3
2
2
2
2'
                               TABLE A-2.  RESPONSE OF RATERS  TO POTENTIAL FOR DISTURBANCE
                                          SEPARATED BY REGION,  BUILDING,  AND AREA
                                          1=LOW 2=MODERATE 3=HIGH POTENTIAL

                                                          REGIONsl
CORE
TWO
















RATER

2
3
3
3
2
2
2
3
2
3
3
3
2
2
2
2
LOCAL RATER
ONE

.
3
2
2
2
2
3
2
3
.
,
2
2
2
2
LOCAL RATER
TWO
3
3
,
.
.
.
,
.
,
.
.
.
.
,
.
•
ffi

-------
BUILDING

18
18
18
18
18
19
19
19
19
19
19
19
20
20
20
20
20
20
21
~r 21
21
21
22
22
22
22
23
23
23
23
24
24
24
24
26
2E
26
26
9
9
9
9
AREA

1
2
3
4
8
1
2
3
4
7
8
9
1
2
3
4
6
8
1
2
3
4
1
2
3
4
1
2
3
4
1
2
4
6
1
2
3
4
1
2
3
4
CORE RATER
ONE
2
2
1
2
2
1
3
2
2
,
,
,
2
2
3
2
.
.
2
3
3
3
2
3
1
1
2
1
3
3
3
3
3
3
2
1
1
2
3
2
2
2
TABLE A-2. RESPONSE OF RATERS TO POTENTIAL FOR DISTURBANCE
           SEPARATED BY REGION,  BUILDING, AND AREA
           1=LOW 2=MODERATE 3=HIGH POTENTIAL

                           REGION=2
CORE RATER
TWO
2
3
2
3
3
1
3
2
2
•
•
3
3
3
2
2
3
2
3
2
3
1
1
2
1
3
3
2
3
2
2
2
2
3
2
3
2
3
2
LOCAL RATER
ONE
3
3
2
3
3
1
3
•3
3
3
2
3
3
3
3
3
2
3
2
)
2
2
2
3
3
3
3
3
3
1
1
1
1
2
3
3
3
LOCAL RATER
TWO
2
2
1
1
2
1
2
2
2
3
2
3
3
2
2
2
2
2
2
1
1
1
1
3
3
2
2
2
2
2
1
2
2
3
1
3
3

-------
             TABLE A-2. RESPONSE OF RATERS TO POTENTIAL FOR DISTURBANCE
                        SEPARATED BY REGION, BUILDING, AND AREA
                        1=LOW 2=MODERATE 3=HIGH POTENTIAL
                                        REGION&3

BUILDING
26
26
26
26
27
27
27
27
27
28
28
28
28
28
28
29
29
29
29
30
30
30
30
30
31
31
31
31
32
32
32
32
32
33
33
33
33
34
34
34
34
36
35
35
36
58
58
 58
 58
AREA
   1
   2
   3
   4
   1
   2
   3
   4
   8
   1
   2
   3
   4
   6
   6
   1
   2
   3
   4
   1
   3
   4
   6
   8
   1
   2
   3
   4
   1
   2
   4
   6
   9
   1
   4
   5
   9
    1
   2
    3
   4
    1
    2
    3
    4
    1
    2
    3
    4
CORE RATER
ONE
         2
         2
         2
         1
         1
         3
         2
         2
         2
         2
         2
         2
         2
         2
         2
         2
         3
         2
         2
         2
         2
         2
         3
         2
         2
         2
         2
         2
         3
         3
         2
         2
         2
         2
         2
         3
         2
         2
         2
         2
         2
         2
         2
         2
         2
IB a — — — —
CORE RATER
TWO
3
3
2
2
2
1
2
3
2
2
3
1
1
2
2
2
2
3
3
2
2
2
3
2
3
2
3
2
2
2
3
2
2
2
2
2
2
3
2
2
2
2
3
.
2
3
3
LOCAL RATER
ONE
3
3
1
1
2
1
1
2
2
1
2
2
3
2
1
2
1
2
.
1
2
1
2
2
2
2
3
1
1
1
3
2
2
2
2
2
3
3
3
3
2
1
2
1
1
2
2
LOCAL RATER
TWO
.
.
.
.
3
1
2
3
3
2
3
3
3
2
2
2
2
2
.
2
3
2
2
3
3
2













2
3
2
3
2
2
3
3

-------
BUILDING

10
10
10
10
36
36
36
36
36
37
37
37
37
37
38
36
38
38
39
00 39
in 39
39
39
40
40
40
40
40
40
41
41
41
41
41
41
41
41
42
42
42
42
59
59
59
59
59 ... - -
60
60
60
60
61
AREA

1
2
3
E
1
2
3
4
7
1
2
3
E
6
1
2
3
4
1
2
3
4
E
1
2
3
4
7
13
1
2
3
4
E
6
7
8
1
2
3
4
1
2
3
4
6
1
2
3
A
2
CORE RATER
ONE
2
3
2
2
3
2
3
3
2
2
2
2
2
1
2
1
2
1
3
3
2
2
2
2
2
1
1
1
3
2
2
2
3
3
2
2
2
2
2
2
2
2
2
2
1
2
2
2
2
2
1
TABLE A-2. RESPONSE OF RATERS TO POTENTIAL FOR DISTURBANCE
           SEPARATED BY REGION, BUILDING, AND AREA
           1=LOW 2=MODERATE 3=HIGH POTENTIAL

                           REGION=4
CORE RATER
TWO
2
3
2
1
2
3
2
3
3
2
2
2
2
2
2
2
2
3
3
3
2
3
3
2
1
1
1
2
2
2
3
3
3
2
2
2
2
3
2
2
2
2
2
1
3
2
3
2
1
LOCAL RATER
ONE

3
2
3
2
2
2
3
3
3
2
2
2
2
2
3
2
2
3
3
1
2
2
3
1
2
2
3
2
3
2
3
3
2
3
3
2
3
2
2
2
2
2
2
3
2
3
2
3
LOCAL RATER
TWO
3
,
3
2
3
3
3
3
2
2
2
2
i
3
1 .
1
1
3
3
3
1
3
3
3
2
1
1
3
3
3
3
3
3
3
3
3
3
3
3
3
2
3
3
2
3
3
3
3
1

-------
                              TABLE A-2. RESPONSE OF RATERS TO POTENTIAL FOR DISTURBANCE
                                         SEPARATED BY REGION, BUILDING, AND AREA
                                         1=LOW 2=MODERATE 3*HIGH POTENTIAL
                                                         REGION**
                 BUILDING            AREA      CORE RATER      CORE RATER      LOCAL  RATER      LOCAL  RATER
                                               ONE             TWO             ONE              TWO

                 61                     31131
                 61                     42232
                 61                     62232
00
ON

-------
TABLE A-2. RESPONSE OF RATERS TO POTENTIAL FOR DISTURBANCE
           SEPARATED BY REGION, BUILDING, AND AREA
           1=LOW 2=MODERATE 3=HIQH POTENTIAL
BUILDING

11
11
11
11
12
12
12
43
43
43
43
44
44
44
44
45
46
4B
46
°° 46
^ 46
46
46
46
47
47
47
47
48
48
48
48
49
49
49
49
62
62
62
63
63
63
63
64
64
64
64
66
66
65
65
AREA

1
2
3
4
1
2
3
1
2
3
5
1
2
3
4
2
3
4
7
2
2
4
6
6
1
2
3
4
1
2
3
4
1
2
3
4
1
7
8
1
2
3
8
2
3
4
7
1
2
5
6
CORE RATER
OKIE
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1
2
2
2
3
3
1
2
3
2
2
3
2
2
2
3
2
2
2
2
1
1
2
2
2
2
                           REGIONsS
                                 CORE RATER
                                 TWO

                                          2
                                          2
                                          2
                                          3
                                          2
                                          1
                                          1
                                          2

                                          2
                                          2
                                          2
                                          2
                                          2
                                          2
                                          2
                                          2
                                          2
                                          2
                                          2
                                          2
                                          2
                                          2
                                          2
                                          2
                                          2
                                          2
                                          2
                                          2
                                          3
                                          3
                                          1
                                          2
                                          2

                                          3
                                          3
                                          2
                                          2
                                          2
                                          2
                                          2
                                          2
                                          2
                                          2
                                          2
                                          2
                                          2
                                          2
                                          2
                                          2
LOCAL RATER
ONE

          1
          1
          1
          2
          1
          1
          1
          1
          1
          1
          2
          1
          2
          2
          1
          1
          1
          1
          3
          2
          1
          2
          1
          1
          1
          1
          1
          1
          1
          1
          1
          1
          1
LOCAL RATER
TWO

          1
          1
          1
          2
          2
          2
          1
          1
          2
          3
          2
          3
          2
          2
          2
          2
          3
          2
          2
          3
          3
          2
          2
          2
          2
          3
          2

          3
          3
          2
          2
          3
          1
          2
          2
          2
          2
          2
          1

-------
                              TABLE A-2. RESPONSE OF RATERS TO POTENTIAL FOR DISTURBANCE
                                         SEPARATED BY REGION, BUILDING, AND AREA
                                         1=LOW 2=MODERATE 3=HIGH POTENTIAL
BUILDING

* 66
66
66
66
66
66
AREA

1
2
3
4
B
6
CORE RATER
ONE
1
1
1
1
1
1
00
00
                                                         REGION=5
                                                               CORE RATER     LOCAL  RATER     LOCAL  RATER
                                                               TWO            ONE             TWO

                                                                         1                1                1
                                                                         1                1                1
                                                                         1                1
                                                                         1                1                1
                                                                         1                1                1
                                                                         1                1                1

-------
TABLE A-3.RESPONSE OF RATERS TO AIR FLOW
          SEPARATED BY REGION, BUILDING. AND AREA
          0=NO AIR FLOW 1=AIR FLOW

                           REGIONsl
BUILDING

13
13
13
13
13
14
14
14
14
14
16
16
16
16
16
16
16
16
~ 16
00 17
vo 1?
17
17
17
60
60
60
60
61
61
61
61
62
62
62
62
63
63
63
63
64
64
64
64
64
66
66
66
66
66
66
AREA

1
2
4
6
7
1
2
3
4
7
1
2
3
4
1
3
4
7
8
1
2
3
4
6
1
2
3
4
1
2
3
4
1
2
3
4
2
3
4
8
1
3
4
6
8
1
2
3
4
1
2
CORE RATER
ONE
0
1
0
0
0
0
0
0
0
0
0
0
1
0
0
1
1
1
0
0
1
1
1
1
1
1
3
1
0
0
0
0
0
0
0
0
1
1
0
0
1
1
1
1
0
0
0
0
1
0
e>
CORE RATER
TWO
0
0
1
0
0
0
0
0
•
0
0
1
0
0
1
1
1
•
1
1
1
0
0
0
0
0
^
^
0
0
0
,
,
1
1
1
0
1
0
0
0
0
0
0
1
0
0
LOCAL RATER
ONE
0
1
0
0
0
0
0
0
0
0
0
1
0
0
1
1
0
0
1
0
6
0
0
0
0
0
1
0
0
0
0
0
0
1
1
1
0
0
0
0
0
;
.
,

.
.
LOCAL RATER
TWO
0
1
1
0
0
0
0
0
0
;
.
.
B
0
i
i
i
•
0
i
i
i
*
i
i
0
i
0
0
0
1
1
1
1
0
1
1
1
0
1
1
1
1
0
0
1
1
e
e

-------
                              TABLE A-3.RESPONSE OF RATERS TO AIR FLOW
                                        SEPARATED BY REGION, BUILDING, AND AREA
                                        0eNO AIR FLOW 1=AIR FLOW

                                                         REGION*!
BUILDING

» 56
66
57
57
57
67
7
7
7
7
7
7
8
8
8
8
AREA

3
A
1
2
4
6
1
2
3
4
7
e
i
2
3
4
CORE RATER
ONE
1
0
0
0
0
0
1
0
0
0
1
0
1
0
0
0
CORE RATER LOCAL F
TWO ONE
1
0
0
0
1
0
1
0
0
1
1
1
t
0
1
0
*ATER LOCAL RATEF
TWO
:
:
0
0
i
0
0
0
1
0
.
,
0
0
0
0
{

I















vo
O

-------
TABLE A-3.RESPONSE OF RATERS TO AIR FLOW
          SEPARATED BY REGION, BUILDING, AND AREA
          0=NO AIR FLOW 1=AIR FLOW

                           REGION=2
BUILDING

18
18
18
18
18
19
19
19
19
19
19
19
20
20
20
20
20
20
21
vO 21
" 21
21
22
22
22
22
23
23
23
23
24
24
24
24
26
25
26
26
9
9
9
9
AREA

1
2
3
4
8
1
2
3
4
7
8
9
1
2
3
4
6
8
1
2
3
4
1
2
3
4
1
2
3
4
1
2
4
6
1
2
3
4
1
2
3
4
CORE RATER
ONE
1
1
0
0
0
1
0
0
0
.
,
.
1
0
1
0
.
.
0
0
0
0
0
0
0
0
0
0
1
1
0
0
0
1
1
0
1
1
1
0
0
0
CORE RATER
TWO
0
1
1
0
1
1
1
1
1
•
i
i
i
0
i
i
i
i
0
0
0
0
0
0
1
1
1
1
0
1
0
1
0
1
1
1
0
0
LOCAL RATER
ONE
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
1
1
LOCAL RATER
TWO
1
0
0
0
0
0
0
0
0
1
0
1
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
1
0
0
1

-------
TABLE A-3.RESPONSE OF RATERS TO AIR FLOW
          SEPARATED BY REGION, BUILDING, AND AREA
          0=NO AIR FLOW IsAIR FLOW

                           REGION=3
BUILDING

. 26
* 26
26
26
27
27
27
27
27
28
28
28
28
28
28
29
29
29
29
•° 30
10 11
30
30
30
31
31
31
31
32
32
32
32
32
33
33
33
33
34
34
34
34
35
35
35
35
58
56
56
58
AREA

1
2
3
4
1
2
3
4
8
1
2
3
4
5
6
1
2
3
4
1
3
4
6
8
1
2
3
4
1
2
4
6
9
1
4
5
9
1
2
3
4
1
2
3
4
1
2
3
4
CORE RATER
ONE
1
1
0
0
0
0
0
0
0
0
0
0
1
.
,
0
1
0
0
0
1
1
0
1
0
0
1
0'
1
0
0
0
0
.
0
0
0
0
0
.
0
1
1
0
1
0
0
0
0
CORE RATER
TWO
1
1
0
1
0
1
0
1
0
0
1
0
0
0
0
0
0
0
1
1
0
1
0
0
1
0
1
0
0
0
0
0
0
0
0
1
1
0
1
1
0
0
0
0
0
0
LOCAL RATER
ONE
1
1
0
0
0
1
0
0
0
0
1
0
1
•
0
0
0
0
,
0
0
0
0
0
1
0
1
1
0
0
0
0
0
0
0
1
1
0
1
0
0
1
0
0
0
0
LOCAL RATER
TWO
.
.
.
.
0
0
0
0
0
0
0
0
1
•
0
0
0
0
0
.
1
e>
i
0
0
1
0












i
0
0
0
0
i
0
0

-------
TABLE A-3.RESPONSE OF RATERS TO AIR FLOW
          SEPARATED BY REGION, BUILDING, AND AREA
          0=NO AIR FLOW 1=AIR FLOW

                           REGIONS
BUILDING

10
10
10
10
36
36
36
36
36
37
37
37
37
37
38
38
38
38
39
39
vO 39
W 39
39
40
40
40
40
40
40
41
41
41
41
41
41
41
41
42
42
42
42
59
59
59
69
59
60
60
60
60
61
AREA

1
2
3
5
1
2
3
4
7
1
2
3
5
6
1
2
3
4
1
2
3
4
5
1
2
3
4
7
13
1
2
3
4
5
6
7
8
1
2
3
4
1
2
3
4
6
1
2
3
4
2
CORE RATER
ONE
0
1
0
0
0
0
0
1
0
0
1
1
0
0
0
0
0
0
1
0
0
0
0
1
1
0
.
0
0
1
1
1
1
1
0
0
0
0
1
1
1
0
0
0
1
0
0
1
0
0
0
CORE RATER
TWO
0
0
0
0
0
0
0
i
0
0
0
0
0
0
0
0
0
•
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
0
0
0
0
1
1
1
0
0
0
0
0
0
0
0
0
LOCAL RATER
ONE
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
1
,
0
0
0
0
0
0
0
0
0
0
0
0
1
1
0
0
0
1
1
1
1
0
0
0
1
0
0
e
e
e
LOCAL RATER
TWO
0
0
0
0
0
0
0
1
0
0
0
0
•
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
0
0
0
0
0
1
1
0
0
0
0
0
0
0
0
e

-------
                             TABLE A-3.RESPONSE  OF  RATERS  TO  AIR  FLOW
                                       SEPARATED BY REGION, BUILDING,  AND  AREA
                                       B=NO  AIR  FLOW 1=AIR FLOW
                                                         REGIONS
                BUILDING             AREA      CORE RATER      CORE RATER     LOCAL RATER     LOCAL RATER
                                               ONE             TWO            ONE             TWO

                61                      30000
                61                      41010
                61                      60000
\O

-------
TABLE A-3.RESPONSE OF RATERS TO AIR FLOW
          SEPARATED BY REGION, BUILDING, AND AREA
          0sNO AIR FLOW 1=AIR FLOW

                           REGIONS
BUILDING

11
11
11
11
12
12
12
43
43
43
43
44
44
44
44
46
46
46
46
vO 46
U) 46
46
46
46
47
47
47
47
48
48
48
48
49
49
49
49
62
62
62
63
63
63
63
64
64
64
64
66
66
66
66
AREA

1
2
3
4
1
2
3
1
2
3
6
1
2
3
4
2
3
4
7
2
2
4
6
6
1
2
3
4
1
2
3
4
1
2
3
4
1
7
8
1
2
3
8
2
3
4
7
1
2
6
6
CORE RATER
ONE
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
1
1
1
0
1
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
e
CORE RATER
TWO
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
(
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
LOCAL RATER
ONE
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
LOCAL RATER
TWO
0
0
0
1
,
,
f
0
0
0
0
,
,
,
,
1
1
1
J
0
0
1
0
0
1
1
1
1 .
1
1
1
.
0
0
0
0
0
1
0
0
0
0
0
.
.
.
.
0
0
0
0

-------
                               TABLE A-3.RESPONSE OF RATERS TO AIR FLOW
                                         SEPARATED BY REGION, BUILDING, AND AREA
                                         0=NO AIR FLOW 1=AIR FLOW
                                                          REGIONeB
                  BUILDING            AREA      CORE RATER      CORE RATER      LOCAL  RATER      LOCAL  RATER
                                                ONE        '     TWO             ONE              TWO

                  66                     10000
                  66                     2000^
                  66                     30000
                  66                     40000
                  66                     60000
                  66                     60000
vo
ON

-------
                      APPENDIX B
                                                      •

COUNTS OF THE RESPONSES OF THE RATERS IN EACH ASSESSED
   AREA WITHIN EACH REGION FOR CONDITION, POTENTIAL
         FOR DISTURBANCE, AND AIR FLOW FACTORS

-------
     TABLE B-l. RESPONSES OF RATERS TO OVERALL CONDITION VARIABLE,
                SEPARATED BY REGION, BUILDING AND AREA.
REGION 1

BUILDING (AREA
13
14

16
17
1
2
4
6
7
1
2
CONDITION
GOOD |
COUNT |
.1
• 1
21
• 1

• 1
41
3 | 3|
4 -1
7
0
2
3
• 1
•1
•1
.1
4 .1
1
• 1
3 | -I
4 I -I
7
8
1
•1
• 1
• 1
2 J
MODERATE | SIGNIFICANT
COUNT |
4
11
2|
•1
•1
•1
•1
H
3|
•1
2|
11
11

11
2|
4|
•1

4|
3|
COUNT

3
.
4
4
4
•
•
1
1
1
2
2
2
3
2
•
4
•
-
1

-------
                              TABLE B-l. RESPONSES OF RATERS TO OVERALL CONDITION VARIABLE,
                                         SEPARATED BY REGION, BUILDING AND AREA.
                         REGION 1
O
O



BUILDING |
17


50

\
u--mi,i
_„ 	 __ITJ
61 • •-- •'"'



62



63



64




AREA
3
4
6

2
3
4


2
3
4
1
2
3
4
2 1
3
4
8
1
13
CONDITION
GOOD | MODERATE
COUNT | COUNT
1
3

2|
• 1 2
•1 3
•1 3
•1 3

•1 3
•1 1

41
•1 1

•1 1
2| 2

•1 1
11 3
•1 3
•1 3
• 1

SIGNIFICANT
COUNT

1
1

2
1
1
1

1
3
4
•
3
2
3
1
2
3
•
1 1
1 1
4

-------
     TABLE B-l. RESPONSES OF RATERS TO OVERALL CONDITION  VARIABLE,
                SEPARATED BY REGION, BUILDING AND AREA.
REGION 1



BUILDING
64


56



56



57



7








AREA
4
6
8
1
2
3
4
1
2
3
4
1
2
A
5
1
2
3
4
7
8
CONDITION
GOOD MODERATE
COUNT COUNT

1
1

3|
-1 3
	 ,„.:! 	 I
•1
•1 1
1
-1 21
2|
1
1
2| 1|
1 2|
1 2|
•1 3|
3| -1
•1 -1
• •
.1

| SIGNIFICANT
COUNT

2
1 2
1
•

2
3
1
1
1
1
2
2
•
•
•
•
•
3
2
2

-------
     TABLE B-l. RESPONSES OF RATERS TO OVERALL CONDITION VARIABLE,
                SEPARATED BY REGION, BUILDING AND AREA.
REGION 1



BUILDING
8






AREA
1
2
3
4

GOOD
COUNT

•
2
1
I
CONDITION
MODERATE
COUNT

3
1
2
2

SIGNIFICANT
COUNT

•

•


-------
                              TABLE B-l. RESPONSES  OF  RATERS  TO OVERALL CONDITION VARIABLE,
                                         SEPARATED  BY  REGION,  BUILDING AND AREA.
                         REGION 2
O
OJ



BUILDING
IB




19






20





21





AREA
1
2
3
4
6
1
2
3
A
7
B
9
1
2
3
4
6
8
1
2
3
CONDITION
- GOOD 1 MODERATE
COUNT | COUNT

3
2
1 - 4
3| 1
2
2 2|
•1
1
4
1
•
•1
11
1|
3| 1|
3|
•1
•1
1 2|
•1
31

| SIGNIFICANT
COUNT

•
2


2
-
A
. 3
•
•
1
1
3
3
•
1
1
1
'
4
1

-------
     TABLE B-l. RESPONSES OF RATERS TO OVERALL CONDITION VARIABLE,
                SEPARATED BY REGION, BUILDING AND AREA.
REGION 2



BUILDING |
21 |
22



23



24



26



9



CONDITION
GOOD MODERATE | SIGNIFICANT
COUNT | COUNT COUNT
AREA
4 4
1 .31
2 .|.4
3 4|
4 .|4.
1 4|
2 .( 3 1
3 .| 4
4 .|.4
1 .| 1 3
2 > . 1| 2
4 4
E .| 4|
1 4
2 . 3| .
3 3| 1 .
4 . 2| 2
1 1| 3|
2 4
3 . 3| .
4 . 1| 3

-------
                              TABLE B-l.  RESPONSES OF RATERS TO OVERALL CONDITION VARIABLE.
                                         SEPARATED BY REGION, BUILDING AND AREA.
                         REGION 3
O
Ln



BUILDING
26



27




28





29



30




(AREA
1
2
3
4
I
Z
3
4
8
1
2
3
4
5
6
1 1
2
3
4
1
3
CONDITION
GOOD MODERATE
COUNT COUNT

2
3
.| ; 3
1 2
1
2| 1|
4
• •
3|
1 2|
H 21
2| 2|
3| 1|
1 -1
H
2|
3|
3|
H 3|
2|
•1

(SIGNIFICANT
COUNT

1

•
.
3
1
.
A
1
1
1
•
•
•
•
2
1
1
•
2
2

-------
     TABLE B-l. RESPONSES OF RATERS TO OVERALL CONDITION VARIABLE,
                SEPARATED BY REGION, BUILDING AND AREA.
REGION 3



BUILDING |
30


31



32




33



34



35



AREA
4
6
6
1
2
3
4
1
2
4
6
9
1
4
6
9
1
2
3
4
11
CONDITION
GOOD MODERATE
COUNT \ COUNT

4
2
2
1
3
2
1

1 2
3
1 2
1
1
3
1
3
1
2| 1
2
•1 3
2| 2

SIGNIFICANT
COUNT

•
2
2
3
1
2
3
3
•
•

2
2
.
2

2

1 1
1


-------
     TABLE B-l. RESPONSES OF RATERS TO OVERALL CONDITION VARIABLE,
                SEPARATED BY REGION, BUILDING AND AREA.
REGION 3



BUILDING
35


68



. ,...,.


AREA
2
3
4
1
2
3
4
CONDITION
GOOD | MODERATE
COUNT | COUNT
.*"-: |
«l
•1 3
•1 2
•1 3
•1 1
•1
.1 3|

SIGNIFICANT
COUNT

•

2
1
3
4
I

-------
                             TABLE B-l. RESPONSES  OF  RATERS  TO  OVERALL CONDITION VARIABLE,
                                        SEPARATED  BY  REGION,  BUILDING AND AREA.
                        REGION A
O
00



BUILDING |
10



36




37




38



39



GOOD
COUNT
AREA
1
2
3
6
1 1
2 I
3 I
< 1
7 |
1 1
2
3 I
5 I
6
1 1
2 I
3
4
1 1
2 1
3 I
CONDITION
MODERATE (SIGNIFICANT
COUNT COUNT

< .
3 1
3 1
3 1
2 2|
3 1
4
•1 «
.| 1 2
4
•1 3
21 1
11
3 1
1 3
1 3
3
3 1
4
.| 3|
1 . 3

-------
     TABLE B-l. RESPONSES OF RATERS TO OVERALL CONDITION VARIABLE,
                SEPARATED BY REGION, BUILDING AND AREA.
REGION 4

BUILDING (AREA
39
40
41
42
59
4
CONDITION
GOOD MODERATE | SIGNIFICANT
COUNT COUNT |
.1
COUNT

5 I r si
1 . 4|
2 1 3|
3
.
.
4 .1
4 1 4 .|
7
13
1
2
3
4
« -1
2|
3|
•1
4|
.21
B | • 4|
6 | 2| 2|
7
8
1
2
3|
•
.
1
1
4
•
1
•
•
1
.| 4|
4| .|
•1 21
3 I .1 3|
4 i .i r\
1
21 2|
1
1
1
.

-------
     TABLE B-l. RESPONSES OF RATERS TO OVERALL CONDITION VARIABLE,
                SEPARATED BY REGION, BUILDING AND AREA.
REGION 4



BUILDING |
59



60



61






AREA
2
3
4
6
1
2
3
4
2
3
4
6
CONDITION
GOOD MODERATE
COUNT COUNT

3
3 1
2| 2
•1 1
2
2 2
2| 2
4
1 21 2
2| 2
4
3| 1

SIGNIFICANT
COUNT

1
•
•

2
•







-------
     TABLE B-l. RESPONSES OF RATERS TO OVERALL CONDITION VARIABLE,
                SEPARATED BY REGION, BUILDING AND AREA.
REGION 6



BUILDING
11



12


43



44



46



46


GOOD
COUNT
(AREA
1 4
2 3
3 4
< I 1
1 1
2 3
3 | 2
1
2
3 I 4
6 2|
1 -I
2 -I
3 .|
4 | . I
2 | 2|
3 | 1|
4 | 2|
7 1 -1
2 I 3|
4 | 11
CONDITION
MODERATE (SIGNIFICANT
COUNT COUNT


11

3|
3|
•1
1
3| 1
2 2

21
2|
1| 2
3|
3|
H 1
2| 1
2|

-------
     TABLE B-l. RESPONSES OF RATERS TO OVERALL CONDITION VARIABLE,
                SEPARATED BY REGION, BUILDING AND AREA.
REGION 5




BUILDING (AREA
46
47
48
49
62
63
5
6
1
2
3

1
2
3
4
1
2
3
4

7
8 I
1
2
3
8

GOOD
COUNT

1
1

2|
1
1
1
1
1
1
'4



1
2
•
1
•

CONDITION

MODERATE (SIGNIFICANT
COUNT |

2|
2|
3|

• 1
2|
3|
•1
3|
2|
• 1
11

3|
2|
2|
3|
3|
J 3|
3|
COUNT
1
1
1
•
1
3
1
•
3
•
1
•
3
2
1
1
•
1
•
1
1

-------
                              TABLE B-l.  RESPONSES  OF  RATERS  TO  OVERALL CONDITION VARIABLE,
                                          SEPARATED  BY  REGION,  BUILDING AND AREA.
                         REGION 5
u>



BUILDING
64



66



66








AREA
2
3
4
7
I
2
S
6
1
2
3
4
5
6

GOOD
COUNT

•
3
I
3
1
1
1
3
3
3
3
2|
3
3|
CONDITION
MODERATE | SIGNIFICANT
| COUNT COUNT

3
• 1
2|
.
3|
3|
3|
1
1
1
11
2|
1
1

-------
 TABLE B-2. RESPONSES OF RATERS TO POTENTIAL FOR DISTURBANCE,
            SEPARATED BY REGION, BUILDING AND AREA.

REGION 1



BUILDING |
13




14




16



16




17

DISTURBANCE
LOW | MODERATE) HIGH
COUNT | COUNT | COUNT
AREA 1 I I
1 1 l| 2|
2 .| H
4 .| 2|
e -I H
' 1 -1 -I
1 | .| 2|
2 | 3| 1|
3 2| 2|
4 | 1| 2|
7 | .| 1|
1 .1 H
2 | .| 2|
3 I .| .1
4 | 1| 2|
1 | .| 1|
3 .| 2|
4 | .) 2|
7 .| .1
e 1 .1 -I
1 .1 3|
2 .| 3|




1
3
2
3
4
2
.
•
1
•
2
1
3
•
3
2
2
4
1
1
1
TABLE B-2. RESPONSES OF RATERS TO POTENTIAL FOR DISTURBANCE,
           SEPARATED BY REGION, BUILDING AND AREA.
                                                                       REGION 1



BUILDING
17


60



61



62



63



64




(AREA
3
4 |
6
1
2
3 I
4 |
1 1
2
3 I
4
1
2
3
4
2
3 I
« 1
e |
i 1
3 I
DISTURBANCE
LOW (MODERATE) HIGH
COUNT COUNT | COUNT

1| 3
2| 2
2|
•1 ^
3| 1
• 1 3| 1
2| 2
3| 1
• 1 H 3
2| 2
2| H 1
.| 2| 2
.1 2| 2
• 1 2| 2
2| 2|
• 1 .1 <
• 1 4
2| 2
3| 1
3| 1
.1 -I 4

-------
 TABLE B-2. RESPONSES OF RATERS TO POTENTIAL FOR DISTURBANCE,
            SEPARATED BY REGION, BUILDING AND AREA.

REGION 1



BUILDING
64


66



69



67



7






1
C(
(AREA
4
6 1
8
1 1
2 1
3 1
4 1
1 1
2 1
3
4 |
1 1
2
4
6 i
1 1
2
3
4 I
7 1
6 1
DISTURBANCE
LOW | MODERATE)
9UNT | COUNT |
I
• 1
•1 -1
•1 -I
•1 3|
•1 -I
H
•1 2|
•1 2|
•1 21
2|
11
•1 H
•1 2|
2|
3|
•1 31
•1 -I
•1 3|
•1 -I
.1 -I
• 1 ' -1

HIGH
COUNT

4
4
1
•
3
2
1
1
1
1
2
2
1
1
•
•
3
•
3
2
2
 TABLE B-2. RESPONSES OF RATERS TO POTENTIAL FOR DISTURBANCE,
            SEPARATED BY REGION,  BUILDING AND AREA.

REGION 1



BUILDING
6






IAREA
"i 	
2
3 I
4
DISTURBANCE
LOW | MODERATE) HIGH
COUNT | COUNT | COUNT

•1 3|
•1 3|
•1 3|
•1 3|

-------
 TABLE B-2. EESPONSES OF RATERS TO POTENTIAL FOR DISTURBANCE,
            SEPARATED BY REGION, BUILDING AND AREA.

REGION 2
 TABLE B-2. RESPONSES OF RATERS TO POTENTIAL FOR DISTURBANCE,
            SEPARATED BY REGION, BUILDING AND AREA.

REGION 2


L
CO
BUILDING (AREA
IB
19
20
21
1
2 1
3
4
e
i 1
2 1
3 1
« 1
7
B
9
1 1
2
3 1
« 1
5
8
1 1
2 1
3 I
DISTURBANCE
OW | MODERATE)
LJNT | COUNT |
3|
2)
2| 2|
»l H
2|
« -I
H
.1 3|
•1 3)
H .1
1 .1
n
11
•1 3)
.1
•1 2|
•1 H
•1 H
3|
•1 M
•1 2)
HIGH
COUNT
1
2
.
2
2
.
3
1
1
•
•
.
3
1
4
2
•
•
1
3
2
DISTURBANCE
LOW | MODERATE) HIGH

COUNT
BUILDING (AREA
21 |4
22
23
24
26
9
COUNT | COUNT
2 2
1 .) 3) 1
2 | .| 2) 2
3 | 4 . .
< ) aj 1) .
1 | 1 3)
2 ) 3) 1)
3 I .) . 4
4 .| .) 4
1 1
2) 2
2 .13
4 2) 2
6
2) 2
1 1 H 3|
2 | 3) 1 .
3 | 2) 1 1
4 I H 3)
1 .13
2 1) 2) 1
3
H 3
4 | .| 2| 2

-------
 TABLE B-2. RESPONSES OF RATERS TO POTENTIAL FOR DISTURBANCE,
            SEPARATED BY REGION, BUILDING AND AREA.
REGION 3



BUILDING |
28



2T




28





29



30

DISTURBANCE
LOW | MODERATE |
COUNT | COUNT |
AREA 1 1 I
1 1 .1 l|
2 i .1 .1
3 I If 2?
4 I ll 2l
i r .1 3i
2 1 4| .|
3 21 21
4 | -1 H
e .1 3|
1 1 11 3|
2 | .| 2|
3 | 1| 2|
4 I 11 H
6 I H -1
8 1 -I H
1 1 -1 «l
2 H 31
3 1 -1 
-------
                                      TABLE  B-2.  RESPONSES  OF  RATERS TO POTENTIAL FOR DISTURBANCE,
                                                 SEPARATED  BY  REGION,  BUILDING AND AREA.
                                     REGION 3
oc

BUILDING (AREA
36
68
2
3
4
1
2
3
4
DISTURBANCE
LOW | MODERATE | HIGH
COUNT

1

1
1
*
•
COUNT
3
COUNT
1
3|
2| 2
21
31
2| 2
2| 2

-------
       TABLE B-2.  RESPONSES OF RATERS TO POTENTIAL FOR DISTURBANCE,
                   SEPARATED BY REGION, BUILDING AND AREA.

       REGION 4
 TABLE B-2. RESPONSES OF RATERS TO POTENTIAL FOR DISTURBANCE,
            SEPARATED BY REGION, BUILDING AND AREA.

REGION 4
vO



BUILDING
10



36




37




38



39


DISTURBANCE
LOW (MODERATE)
COUNT | COUNT |
AREA I I
•I 2|
2 .1 .1
3 -1 31
6 | 1| 2|
I 1 -I 2|
2 | .) 2|
3 | .| 2|
4 1 .1 .1
7 -1 21
1 -1 31
2 -1 
-------
                                          TABLE B-2. RESPONSES OF RATERS TO POTENTIAL FOR DISTURBANCE,
                                                     SEPARATED BY REGION, BUILDING AND AREA.
                                         REGION 4
NJ
o



BUILDING
69



60



61






AREA
;
3
*
6
1
2
3
4
2
3
<
6
DISTURBANCE
LOW (MODERATE
COUNT COUNT

3
3
2| 2
1
1
3|
1
•1 3
31
3|
•1 3
•1 3

| HIGH
COUNT

1
1
•
•
3
1
3
1
1
1
»
»

-------
 TABLE B-2. RESPONSES OF RATERS TO POTENTIAL FOR DISTURBANCE,
            SEPARATED BY REGION, BUILDING AND AREA.
REGION 6



BUILDING |
11



12


43



44



46



46

DISTURBANCE
LOW (MODERATE I
COUNT | COUNT |
AREA 1 1
1 2| 2|
2 | 2| 2|
3 | 2| 2|
4 1 1 3|
1 H 2|
2 2| 1|
3 2| 1|
1 1 H 3)
2 | 1| 2|
3 | 2| 2|
6 1 H 3|
'l * ll «I
2 -I 3|
3 1 -1 »l
4 | 1| 2|
'2 * ll 3|
3 | 1| 2|
4 | 1| 3|
7 H 2|
2 ll «I
4 H 31

HIGH
COUNT

•
•
•
1
•
•
•
•
•
•
•
•
•
•
*
•
1
•
1
•
•
 TABLE B-2. RESPONSES OF RATERS TO POTENTIAL FOR DISTURBANCE,
            SEPARATED BY REGION, BUILDING AND AREA.

REGION 6
DISTURBANCE
LOW | MODERATE) HIGH

BUILDING (AREA
46
47
48
49
62
63
6
6
1
2 I
3 I
« I
1 1
2
3 1
4
1
2 I
3
* I
1 1
1 1
B I
1 1
2 I
3 I
8 I
COUNT |
ij
H
H
2|
1|
H
H
11
H
3|
H
H
11
n
.1
.1
11
.1
2|
n
n
COUNT | COUNT
3
2| 1
3|
2|
2| 1
2| 1
3|
H 2
H 2
1
2| 1
2| 1
n
H 2
•1 4

-------
                                           TABLE B-2.  RESPONSES  OF  RATERS TO POTENTIAL FOR DISTURBANCE,
                                                      SEPARATED  BY  REGION,  BUILDING AND AREA.
                                          REGION 6
to



BUILDING |
64



66



66








AREA
.
3
«
7
1
2
6
6
1
2
3
4
6
6
DISTURBING!
LOW (MODERATE
COUNT | COUNT

J 2
1| 2
2| 1
2| 1
1| 3
H 3
1| 3
1 21 2


3|
1 «l

-------
           TABLE B-3.  RESPONSES OF RATERS TO AIR FLOW,
                      SEPARATED BY REGION, BUILDING AND AREA.

            REGION 1
TABLE B-3. RESPONSES OF RATERS TO AIR FLOW,
           SEPARATED BY REGION, BUILDING AND AREA,

 REGION 1
to
co

I
1
Cl
BUILDING (AREA
13
14
16
16
17
1
2 1
* |
« 1
7 1
1 1
2
3 1
4 |
7 |
1
2
3
4 |
1
3
4 I
7 I
B I
1 1
2 1
MR FLOW
40 | YES
1UNT| COUNT
4
H 3
2| 2
<|
•«!
4

-------
           TABLE B-3. RESPONSES OF RATERS TO AIR FLOW,
                      SEPARATED BY REGION, BUILDING AND AREA.

            REGION 1
N>



BUILDING |
64


66



66



67



7








AREA
4
6
8
1
2
3
4

2
3
4
1
2
4
6
1
2
3
4
7
8
AIR FLOW
NO | YES
COUNT (COUNT

2 2
2| 2
1
31
3|
2| 1
•1 3
1 3| .
3|
1 • 3
1 21 1
1 3|
31
1 M 2
3|
1| 2
1 31
2| 1
| 2| 1
1 -1 2
1 1
TABLE B-3. RESPONSES OF RATERS TO AIR FLOW,
           SEPARATED BY REGION, BUILDING AND AREA.

 REGION 1



BUILDING
e






AREA
7 1
2 1
3
< 1
AIR FLOW
NO | YES
COUNT | COUNT

ll I
3|
2| 1
3|

-------
           TABLE B-3. RESPONSES OF RATERS TO AIR FLOW,
                      SEPARATED BY REGION, BUILDING AND AREA.

            REGION 2
TABLE B-3. RESPONSES OF RATERS TO AIR FLOW,
           SEPARATED BY REGION, BUILDING AND AREA.

 REGION 2
10



BUILDING
18




19






20





21





IAREA
i
2
3
4
8
1
2
3
4
7
B
9 1
1
2
3 I
4 I
6
B
1
2
3 1
AIR FLOW
NO | YES
COUNT (COUNT

2| 2
1 21 2
1 3| 1
«l
1 3| 1
2| 2
3| 1
3| 1
3| 1
H
H
H
1| 3
3| 1
•1 «
«|
•1 1
•1 1
3| 1
3| 1
31 1


1
Cl
BUILDING (AREA
21 M
22
23
24
26
8
1 1
2
3 I
4
1 1
2 I
3
« 1
1 1
2 I
4 I
E
1
2 I
3 I
« 1
» 1
2 I
3 1
« 1
MR FLOW
W YES
JUNTI COUNT
3 1
4|
*|
4
4|
4
4
2| 2
1 3
3| 1
3| 1
4|
2| 2
3| 1
3| 1
3| 1
2| 2
•1 4
3| 1
3| 1
2| 2

-------
            TABLE B-l.  RESPONSES OF RATERS TO AIR PLOW,
                       SEPARATED BY REGION, BUILDING AND AREA.

             REGION I
TABLE B-l. RESPONSES OF RATERS TO AIR FLOW,
           SEPARATED BY REGION, BUILDING AND AREA.

 REGION 8
r\)



BUILDING |
36



27




21





29



IB




AREA
1
2
I
4
1
2
1
4
8
1
2
1
4
E
6
1
2
1
4
1
1
AIR FLOW
NO | YES
COUNT (COUNT

.| B
.1 >
>l
2| 1
4|
2| 2
4|
• 1 1
4|
4|
2| 2
4|
1 H •
11
H
II
II 1
4|
41
1 4|
.1 2



BUILDING
se


si



12




II



14



IB



AREA
; 	
e
t
i
2
»
4
1
2
4
e
9
1
4
B
9
1
8
1
4
1
AIR FLOW
NO | YES
COUNT (COUNT

l| »
4|
H >
4|
4|
•1 4
4|
•1 »
2| 1
»l
•1
•1
21
»l
S|
•I
•1
11 2
•1 2
II
•1 4

-------
                                          TABLE B-S. RESPONSES OP RATERS TO AIR PLOW,
                                                     SEPARATED BY REGION, BUILDING AND AREA.

                                           REGION I
to



BUILDING
IS


SI






AREA
1 	
1
4
1
2
I
4
AIR PLOW
NO | YES
COUNT (COUNT

2| 2
«l
2| 9
*\
*\ 1
<|
<|

-------
           TABLE B-3. RESPONSES OF RATERS  TO  AIR  FLOW,
                •*     SEPARATED BY REGION,  BUILDING  AND  AREA.

            REGION 4
TABLE 6-3. RESPONSES OF RATERS TO AIR FLOW,
           SEPARATED BY REGION, BUILDING AND AREA.

 REGION 4
to
00



BUILDING |
10



36




37




38



39





AREA
1
2
3
6
1
2
3
4
7
1
2
3
B
8
1
2
3
4
1
2
3
AIR FLOW
NO | YES
COUNT (COUNT

4|
3| 1




4|
•1 4
4|

3| 1
3| 1
1 H
4|


4|
1 4|
1 -1 3
1 3|
4|



BUILDING
39

40





41







42



69



|AREA
4
6
1
2
3
4
7
13
1
2
3
4
6
6
7
B
1
2
3
4
1
AIR FLOW
NO | YES
COUNT (COUNT

*\
«l
3| 1
3| 1
«l
3|
«l
«l
3| 1
3| 1
3| 1
•I <
•1 <

-------
                                        TABLE B-3. RESPONSES OF RATERS TO AIR FLOW,
                                                   SEPARATED BY REGION, BUILDING AND AREA.

                                         REGION 4
NJ



BUILDING
60



60



61






AREA
>.
3
4
6 1
1 1
2 1
a 1
4 I
2 1
a
< 1
6
AIR FLOW
NO | YES
COUNT (COUNT

4
4
2| 2
11
4
3| 1

-------
         TABLE B-3.  RESPONSES OF RATERS TO AIR FLOW,
               '    SEPARATED BY REGION, BUILDING AND AREA.

          REGION S
TABLE B-3. RESPONSES OF RATERS TO AIR FLOW,
           SEPARATED BY REGION, BUILDING AND AREA.

 REGION 6
10
O



BUILDING |
11



12


43



44



46



46




AREA
1
2
3
<
1
2
3
1
2
3
6
1
2
3
4
2
3
4
7
2
4
AIR FLOW
NO | YES
COUNT (COUNT

3| 1


-------
TABLE B-3. RESPONSES OF RATERS TO AIR FLOW,
           SEPARATED BY REGION, BUILDING AND AREA.
 REGION 6



BUILDING
64



66



66








AREA
>.
3
4
7
1
2
6 1
6 }
I
2
3 1
4 1
6
6 I
AIR FLOW
NO | YES
COUNT) COUNT
|
3| .
31
3|
3|
«l

-------
          APPENDIX C

CLASSIFICATION OF ACM CONDITION
         (USEPA 1986a)

-------
SURFACING MATERIAL

Significant  Damage  —  ACM with  one  or  more  of the  following
characteristics:    the  surface crumbling  or  blistered over  at
least one tenth  of  the  area if the damage  is evenly distributed,
or  at  least one quarter if the damage  is  localized;  large areas
.of  material hanging from the  surface,  delaminated, or  showing
adhesive failure; at least one tenth of the surface water-stained
or  heavily  gouged,  marred or abraded  (or  one  quarter  if  the
damage  is  localized);  large  accumulation of  powder,  dust,  or
debris on surfaces beneath the ceiling or wall.

Moderate  Damage  —  ACM  with one or  more  of  the  following
characteristics:   up to one tenth of the  surface (if the damage
is  evenly  distributed)  or up  to  one quarter of  the  surface (if
the damage  is  localized)  blistered,  crumbling,  water-stained,  or
gouged, marred or abraded; some accumulation of  powder,  dust  or
debris on surfaces beneath the ceiling or wall.

Good^Condition — ACM with no visible damage or deterioration,  or
showing only very limited damage or deterioration.


THERMAL SYSTEM INSULATION

Significant  Damage  —  ACM with  one  or   more  of the  following
characteristics:   mostly  missing  jackets;  water-damaged, crushed
or  heavily  gouged or punctured insulation on  at  least one tenth
of  pipe  runs/risers if the  damage is evenly  distributed, or at
least one  quarter if the damage  is  localized;  powder,  dust, and
debris on surfaces beneath pipe/boilers/tanks, etc.

Moderate  Damage  — ACM  with one  or more   of the  following
characteristics:    a few  water  stains or  sections  of  missing
jackets;  crushed insulation or water  stains,  gouges,  punctures,
or  mars  on  up to one tenth of the insulation if  the damage is
evenly  distributed,  or  up to  one  quarter  if  the  damage  is
localized;  some  accumulation .of powder, dust, debris on  surfaces
beneath pipes/boilers/tanks, etc.

Good Condition — ACM with  no  visible damage or deterioration, or
showing only very  limited  damage  or deterioration.
                                135

-------
        APPENDIX D



AIR SAMPLING FIELD METHODS

-------
           Asbestos  air  samples were collected  on 0.45  /tra pore
size, cellulose acetate membrane filters enclosed in preassembled
37 minicassettes.  Two side-by-side samples were collected at each
location, each at a different flow rate.  Volumes were controlled
through the  use  of two limiting orifices, 5.0 1pm,  and 2.5 1pm,
with air flow being drawn across the filter by a diaphragm vacuum
pump.   Collection of side-by-side samples at  two  different flow
rates allowed  for a backup  filter,  should the higher  flow rate
sample  become  overloaded.   Samples were logged in  the  field and
hand carried to the analytical laboratory.

A.  Site Selection
           Sampling  locations in each of  the monitored buildings
were based  on  the representativeness of  the location,  proximity
to the ACM, accessibility, potential for vandalism, and access to
power.   In  general,  eight  sampling  locations were  selected in
each of the  monitored  buildings:   seven indoor locations and one
outdoor ambient location.

           Pump placement involved locating two pumps directly in
the  area with  the  most  damaged ACM,  2  in the  nearest public
access  area,  2 in other  assessed  areas,  one in a public access
area  adjacent  to another   area;  and  one  pump  at  an  outdoor
location.   Local circumstances may have  required  pump placement
at other sites in a few instances.

           In general, a sampling survey in each study region was •
split  into  two  segments,   with  each  segment  involving sample
collection  in 5  buildings.   The  exception to  this  was   Study
Region  4,  consisting  of  2 cities,  in which  5  buildings were
sampled in each  city.   Each sampling period contained an initial
set-up  day  for  pump  placement,  a 2-day  sampling  period, and
approximately one day  for breakdown and filter delivery.


B.  Sampling Equipment

           The sampling  system used during the project consisted
of the  following:

        •  Two open-faced 37 mm cassettes, each containing  a
           0.4 fim. cellulose  acetate membrane filter;

        •  Two  flow control orifices;   one  at  2.5  1pm and the
           other at  5  1pm;

        •  A pump with a muffler;

        •  Associated  plumbing and  stand;
                                139

-------
         •  A 7-day timer; and

         •  A clock to record elapsed time.

           The  sampler  setup it  represented in Figure  D-l with
two modifications.   The 36-inch rod used  to hold  the filters in
place was  attached to a separate  laboratory stand  and not to the
pump base.   This modification served to  minimize  the effects of
pump vibrations  on  the  filter.   The second modification was the
use of  a T-fitting with  two  orifices  (5.0  1pm  and  2.5  1pm) and
two separate  filter  cassettes.   This allowed for  the collection
of two simultaneous samples, each at a different flow rate.


C.  Sample Collection Procedures

           Sample  collection  in  each  building  was  conducted
during periods of maximum activity  (daylight hours)  over a 2-day
period.   Generally,  sample collection  hours were  between 7:00
a.m. and 5:00 p.m.  on  each  day of  the  2-day  sampling period.
Exceptions occurred when timer  malfunctions required alterations
of the sampling period.

           Pumps were set up  one day prior to  the actual sample
collection and set  to activate the morning of  the following day
via an  in-line,  7-day  timer.   Advance  set-up was  necessary to
ensure that  all  samples  started  at approximately the  same time
each day,  since  geographic  locations were dispersed and building
access in the early morning hours uncertain.  The following
details the sampling procedures followed during the program.


           Sampling Protocol

           1.   Visually  inspect  preloaded  filter  cassettes for
                damage.   Label filter cassette with random I.D.;

           2.   Place  filter  in  cassette  holder,  clamp  into
                position, and  attach pump tubing.   Ensure that
                filter  holder  (ring stand)  is  placed in  such a
                way as to minimize or eliminate vibration effects
                caused by the pump;

           3.   Rotate  filter  holders   to  a  vertical  position
                (perpendicular to the ground);

           4.   Check plumbing for any leaks;

           5.   Check flow rates with a flowmeter;

           6.   Set automatic timer to correct date and time;


                               140

-------
                                                              Hoidrt
           Oflfict
           DotlH
                                         1 4" CODOtr
                                         facing Wound
                                         4
J»oot 14' i 318'
         I-2-MMC4T
 NOU Connwior to Mm
     M'  Mill PiD« 10
 1 *" l.O
             me
    Mofl«i 107CA1I
                              *«•<    w 6" i iv Mm SUM)
         Mi't Hoow 90*
' I M»it 0iO» "n'MO«C to
1 4 TuB»
       ••200;.
                         Figure  0-1.    Pump  diagram.
                                    141

-------
            7.    Make  appropriate  logbook entries;

            8.    Conduct  sampling;

            9.    After sampling, check flows  (leave pump running);
                 and

           10.    Stop  pump and remove filters.


            Filter  Handling Procedures

            1.    Use   preloaded   filter  cassettes   to  minimize
                 contamination;

            2.    After sampling,  place the  cover over the filter
                 holder,  maintaining exposed side up  during the
                 handling, and transport to  the  laboratory;

            3.    Hand  deliver  all  samples   at  the  end  of  each
                 sampling  period  to  the   electron  microscopy
                 laboratory;  and

            4.    Maintain the  filter  in  a  horizontal  position
                 during handling,  transport  and storage.   Handle
                 in  such  a  way  as  to   minimize  dislodging
                 structures from the filter  surface.


            Post-Sampling Procedures

            1.    Measure  the  flow;

            2.    Check filter condition and  location of sampler;

            3.    Record time  position of timer  clock and elapsed
                 time;

            4.    Record  the  relative  humidity and  temperature
                 inside and outside the building; and

            5.    Complete  chain-of-custody  record   prior   to
                 packaging and shipment to the laboratory.


            Logbook/Data  Form Entries

           An  important part  of  any  field  program  are  the
observations  and  accurate  records  of  the field  team.    The
following  information was recorded  in  the  logbook and data forms
for each sampling  location:


                               142

-------
           1.    Name of field program;
           2.    Date of record;
           3.    Site number and location;
           4.    Tag numbers of pump and timer;
           5.    Relative  humidity  and  temperature  inside  and
                outside the building;
           6.    Position of sampler within the  site;
           7.    Brief site description;
           8.    Corresponding filter number;
           9.    Sample flow rate at the start of sampling;
           10.  Settings of timer clock;
           11.  Sample flow rate at end of sampling period;
           12.  Comments; and
           13.  Name(s) of samplers.
           A  copy  of the data  form used during  this  program is
included as Figure D-2.
           Field Flow Measurement
           At a minimum,  flow rate measurements were taken twice
during a  sample  run:  during sampler  set up prior to initiation
of the  run,  and at  the  completion of  sampling.   If possible,  a
mid-point  flow  measurement  also  was  taken.     The  following
describes the procedures used to determine  sample flow rates in
the field.
           1.   Turn on the sampling pump;
           2.   Set  up the sampling  system as  shown  below  with
                both  rotameters in  line  between  the  filter  and
                the orifice;
                               143

-------
 Project  No.
 Building  I.D.
 Location
                          Field  Data  Sheet
                            Start Date.
                       Stop Date.
 Pump  I.D. No.



 Filter Lot »..



 Box *.	



 Day 1:



 Start   	



 Day 2:



 Start   	
Flow Checks:



Date _
Tine
Flow Rate (st?).
Date.
Time.
F'.ow Rate (sr,p)



Date
Time
Flow Rate (st?)



Picture  Roll -



Comments :
                           Flow Control Device.



                                    Type	
                                    Random  I.D.
Stop
                             Stop
                    Temp..
                  Rctaseter reading.
                  Rotameter neading.
                    Temp
                  Rotameter Reading.
                                                        Elapsed
                          Elapsed
           B.P.
V.?
                Rctameter No..
                                                        v.?.
                Rotameter No.
                           V.P.
                                    Frame *
                Rotameter No.
   Figure D-2.  Field data form  used for air monitoring.
                                 144

-------
FILTER 1
FILTERS
         fl
•o
ORIFICE
         ROTA«TER
         BOTAMETER
         ft—<>
       PUMP
       WITH
       MUFFLER
                                         TIMER
                               ELECTRICAL
                               POWER
                               SOURCE
     3.   Record the rotameter readings  in  the notebook;

     4.   Turn off the pump and remove the  rotameters;

     5.   Reconnect  all  tubing and turn the  filters to  a
          horizontal position;

     6.   Repeat procedures  1 through 5 at the end  of  the
          sampling period; and

     7.   Calculate the flow  as follows:

          a.   Using   the   calibration  curve   for   the
               rotameter, determine the  flow rates for each
               rotameter reading  and record these values on
               the data sheet.

          b.   Calculate  the average   flow  rate  for  the
               sampling    period   using    the   following
               equation:


 Average flow rate   =  (initial  flow rate^ final flow rate)


          c.   Calculate   the  actual   volume   of  sample
               collected  by multiplying the average sample
               rate  by the sampling time.
                          145

-------
                APPENDIX B

AIR SAMPLE PREPARATION AND SUMMARY OF TEM
            ANALYTICAL PROTOCOL

-------
•AMfLl HUBtAlATIQM

        Tht  following  it an  abbrtviattd vtrtion  of tht  aamplt
preparation procedure utiliiod in thii projtot,  A mort detailed
vtraion of  tho procedure, along with  tne referenoti rtlattd  to
tht development of  thii  procedure,  ia  contained in  Itotion  7  of
Yamati tt al.  (1914).

        Tho Low Ttmporaturt Aahor  (LTA)  uiod in thoin  toata wai
calibrated  through  a aerioa  of  ttata to determine tht  etching
rate of  a mixtd colluloat tattr  filttr,  It  WAI  tht  inttnt  of
thii projtot to ttoh approximattly  1 0m  of tht  turfact  to rtvtal
itruoturt dttail that may havt bttn hiddtn in tht rtplicatt,


rrootdurti

1.   A atotion of tht mtmbrtnt filttr  !• out with  a ioalptl, and
     plaotd  on a eltan  mioroioopt ilidt with  tht  lampltd  ildt
     laoing up.

a.   Tht out Motion  ia  faattntd  on all  eidtt  to  tht  ilidt  with
     narrow atripa of tranapartnt tapt.

I.   Tht ilidt, with tht out itotion,  it  txpoitd to aettont vapor
     (not  liquid)   for  approximattly  10 minutti.    Tht  aottont
     vapor  oolltpata tht atruoturt of  tht filttr  and  product• •
     fuitd,  rtlativtly  amooth-aurfaotd  film.    Tht  aiit  of tht
     aottont vapor bath and timt of filttr rtaponat to tht vapori
     art oritioal in obtaining tht dtairtd amooth,  fuatd iurfaot.

4.   laeh oollapatd filttr  atgmtnt with   a known dtpoait  arta  io
     oartfully plaotd in  a oltan ttat  tubt  (II  mm  x 10 mm) uaing
     a oltan twttitr.

B.   With  forotpa.  tht  tubta  containing tht  atmplt,  and  1 lab
     blank  (unuitd  filttr atgmtnt of  tht aamt  aiit and  typt  of
     filttr material  aa  tht aamplt) art plaotd Itngthwiit,  iidt
     by aidt in tht ohambtr,  with  tht  moutha of tht tubta facing
     tnt  optn  tnd  (door) of  tht aahtr  ohambtr.  Tht  tubta art
     laid In tht  otnttr  of tht ohambtr  within  tht rtgion of tht
     eoila  aurrounding tht ohambtr.  Up  to  four aamplt tubta and
     1 blank can bt laid  likt loga inaidt tht ohambtr•

I.   Tht  powtr  ia   •lowly  and  oartfully  incrtaatd to   prtvtnt
     "flaaning"  of  tht  filttr,  which would  rtault in  loaa  of
     aamplt.

7.   Tht  filttr mtmbrant ia  ttohtd for approximattly 30  atoonda.
     Tht chamber ia •lowly allowed to  rtaon  ambient  preaaurt.


                               149

-------
8.   The  etched-collapsed  filter   section   is   placed  on  the
     rotating stage of the vacuum evaporator for carbon-coating.

9.   A  3-nun-diameter  portion  of  the  carbon-coated  filter  is
     transferred to an EM grid in the modified Jaffe wick washer.

10.  Acetone is used in dissolving the fused membrane filter.

11.  Transfer to the properly labeled grid storage container.


SUMMARY OF TEM ANALYTICAL PROTOCOL

        The  following is  a  summary  of  the  Analytical Protocol
utilized  in  this project.   A  more  detailed description  of the
protocol is contained in Appendix B to the Quality Assurance Plan
(Hatfield et al. 1987) for this project.

Procedure:

1.   Start a new Count  Sheet  for  each  sample  to  be analyzed.
     Record  on  that  sheet:    Client Name;  Project or  Job No.;
     Sample No.; Volume  of Air Analyzed  (from  TEM Working Log);
     Microscope; Magnification for  Analysis;  Filter  type,  and
     Diameter.

2.   Start with the grid in capsule labelled No. 1 located in the
     Specimen Box.

3.   Determine Suitability of Grid

     A. Look at grid  in Low  Mag  mode  (100X)  to  determine its
        suitability for detailed study at higher mags.

     B. Reject grid if:

        i.   Replica  does  not  cover  at  least  15  full  grid
             openings  with  0% holes  in  any  grid,  and  <  15%
             coverage maximum.  Discount any grid opening that is
             doubled or folded for counting.

        ii.   Specimen is  too  dark due to  incomplete dissolution
             of the filter.

        iii.  The average particulate loading exceeds 15%.

     C. If grid is  rejected,  load grid from capsule No. 2,  etc.

     D. If grid is  acceptable, continue on to next step.
                               150

-------
4.   Scan the Grid

     A. Set the magnification to 19,OOOX

     B. Scan grid as follows:

        i.   At the  appropriate  magnification,  make a  series  of
             parallel   traverses  across   the   grid   opening.
             Traverse the  grid  opening  (also referred  to as  a
             field), starting  at 1 corner  (upper left  or upper
             right)  and using the  area defined by  the  small
             square  of  the  fluorescent  screen  (area  of  screen
             that   lifts   up  for  photograph  purposes)   as   a
             "window".

        ii.  On reaching the end of 1 traverse, move the image 1
             "window" width, and reverse the  traverse.   A slight
             overlap should be used so as not to miss any part of
             the opening.

        iii. Make  parallel  traverses  until  the  entire  grid
             opening has been scanned.

     C. Ten  good  fields or grid openings or  100  structures need
        to be counted (whichever comes first).


5.   Identify  each structure  morphologically and analyze  as  it
     enters the "window".


6.   For morphology:  appearance and size

     A. Determine  morphologically  if  the  structure is  a "fiber",
        "bundle",  "cluster", or  "matrix".

     B. If  record "bundle",  "cluster",  or  "matrix", then record
        also  how  many  figures  are  involved;  i.e.,   Bundle  7,
        Bundle > 50, etc.

     C. Size each structure using the  calibrated 20 mm rule on
        the screen.


7.   Selected area electron diffraction pattern (SAED)

     A. Center structure, focus and obtain SAED pattern

     B. From  a visual  examination of  the  electron diffraction
        pattern (camera length (CL)  of  22;  through  binoculars on
        small   screen),  classify  the  observed   structure  as


                               151

-------
        belonging to one of the following categories by comparing
        it to known patterns:

        i.   Chrysotile:   The  chrysotile  asbestos pattern  has
             characteristic streaks on layer lines other than the
             central line  and some  streaking also on the central
             alternate  lines  (2nd,  4th,  etc.)*    The  repeat
             distance between layer lines is about 0.53 mm.

        ii.  Amphibole  Group   (includes  amosite,  crocidolite,
             anthophyllite, tremolite and actinolite):  Amphibole
             asbestos  structure  patterns who  layer  lines  formed
             by very closely spaced dots, and the repeat distance
             between   layer   lines  is   also  about   0.53   mm.
             Streaking in  layer lines is occasionally present due
             to structure  defects.

        iii. Ambiguous (incomplete spot patterns).

        iv.  N, if there is  no  pattern present.   (This should go
             under SAED column).

     C. If  the  pattern is a suspected  chrysotile  or amphibole,
        then take picture  of diffraction pattern as needed.

8.   X-ray Analysis (EDS).

     A. For  each  structure  that  chemistry  is   necessary,  take
        chemistry with EDS system.

     B. If EDS signal  is weak,  take another spectrum, being sure
        that spot is still on structure.

     C. If EDS is used for confirmation, record structure
        identification.    Record a  check mark or an  "X"  in EDS
        column when chemistry is checked but not saved.

     D. If  EDS   is   used  in  case   of  unknown  or  ambiguous
        structures, categorizing amphiboles or showing
        representative  structures   on  particular  field,  save
        spectra to disk and record Disk No. and File No. on Count
        Sheet under EDS Column.

9.   After all necessary analyses of structure, continue scanning
     until all structures  are identified, measured, analyzed, and
     categorized in the grid opening.

10.  Select additional grid  openings at  low mag,  scan at 19,OOOX
     and analyze  until the  total  number of  asbestos structures
     exceeds  100,  or  a  minimum of  10  grid openings  have  been
     examined, whichever comes first.


                               152

-------
11.  Carefully  record  all  data  as  it  is  being collected,  and
     check for accuracy.


12.  After  finishing  with  grid,   remove   from  microscope,and
     replace in appropriate polyethylene capsule.
                              153

-------
           APPENDIX F



ANALYSIS OF TEM GRID OPENING DATA

-------
BACKGROUND

          When air samples are analyzed for asbestos by TEM using
a  direct  preparation  technique  the  spatial  distribution  of
asbestos structures on the electron microscope grid is similar to
their  distribution on  the  filter at  the time  of  collection.
(Some changes may take place during transport  to the laboratory.)
Concerns  have been  raised over  the  uniformity  of the  spatial
distribution.   Since only  a small proportion  of the  filter is
examined, a  highly clumped  or  non-uniform  distribution may yield
low structure counts by chance, even though the average structure
density  is  high.   Conversely,  if the area  of  filter examined
happens  to   include  an  aggregation  of asbestos  structures  the
airborne asbestos concentration will be overestimated.

          No asbestos structures were detected on over 80% of the
samples  collected  in  this study.    This  prompted  additional
analyses  to  determine   if  the  low  structure  counts  reflected
actual low airborne asbestos concentrations or could be explained
by  a  non-uniform distribution  of asbestos  structures  on  the
surface of the filter.  The objective of the investigation was to
characterize the  spatial distribution of asbestos structures and
determine the effect  of the  distribution on  the precision of
structure counts.
STUDY DESIGN

          Sixteen  air  samples  were   selected   for  additional
analysis to  determine if the  10  grid openings specified  by the
TEM protocol  provide estimates of  sufficient precision for the
purposes of the  study.   The 16 samples were  selected as follows
to provide a range of structure counts:

     •    4 "indoor"  samples which  had structure counts of  3 or
          more in the first 10 grid openings counted;

     •    8 "indoor"  samples which  had structure counts of  0 in
          the first 10 grid openings;

     •    2 "outdoor" samples;  and

     •    2 field blanks.

Samples were selected at random within each category.

          An additional  40  grid openings, giving a total  of 50,
were examined on each sample and the number of structures  in  each
opening recorded.
                               157

-------
 STATISTICAL MODEL

           The  negative binomial is a discrete distribution which
 is  often  used to  describe  clumped or  aggregated populations.
 Javitz  and  Fowler  (1981) found  that the  negative binomial was
 superior  to the Poisson for describing asbestos structure counts
 obtained  by electron  microscopy.   The variance  of the negative
 binomial  is  m(m+k)/k where m  is  the mean and k is a measure of
 aggregation.     As  k  increases   the   variance  decreases  and
 consequently  the  precision   of   estimated  airborne  asbestos
 concentrations increases.  The  Poisson distribution is a limiting
 case  of the negative binomial as k becomes very large.

           If  the number  of  asbestos  structures  in one   grid
 opening  is  assumed  to  be distributed  according  to  a negative
 binomial  distribution  with parameters m and k, then the number of
 structures in  n grid openings  is  distributed  according  to  a
 negative  binomial with parameters  nm and  nk.   This  result assumes
 that  the  number of structures in a grid opening is  independent of
 the number in any other  grid opening.   The assumption will  hold
 if  grid  openings are  selected  at random.   If grid openings are
 not  selected  at  random  then  one must  assume that there  is no
 spatial correlation between the number of structures in different
 grid  openings.


 PARAMETER ESTIMATES

           The  maximum  likelihood estimate of m is the  sample  mean
 x.  The  maximum  likelihood estimate of  k is obtained by solving
 the equation
                   x      °°
           nlog(l  + P )  =  Z f^l/k +l/(k+l)+..-+l/(k+j-l)),
                         j=l   -1
 where fj  is  the number of  grid  openings with j structures (Bishop
 et al 1975).   Estimates  of m and k were obtained for each sample.
 Large  values  of  k  mean  higher   precision,  and  hence  narrower
 confidence intervals for the  total structure count on a filter.
CONFIDENCE INTERVALS

          Exact  100(l-a)%  confidence  intervals,   (HIL^U)'  were
obtained for x> 0 by finding mL and my  such that

          F(x,mu)=
-------
«)% confidence  interval is  given by  (Q,mq)  where nty satisfies
F(x,mu)=a.

          A negative binomial  with parameters nra and  nk  is used
to obtain  confidence  intervals for the  total count on a filter
when  n  grid  openings  are  examined.    Increasing  k,   and/or
increasing n,  increases the precision of  the count and  reduces
the width of the confidence interval.
RESULTS

          No asbestos structures were counted  on  eight of the 16
filters.   The  eight  include the  two field blanks and  the  two
outdoor samples.  Of the eight filters with non-zero counts, five
have  estimates of  k equal  to  infinity.    The  remaining  three
estimates of k are 0.6, 0.4, and 0.07.

          For k=», i.e., a Poisson distribution, a 95% confidence
interval  for the  true structure  count when  no  structures  are
counted  in  10  grid  openings  is  (0,3.0).    The  size  of  the
confidence interval increases slightly  to  (0,3.1) as k decreases
to 0.4.   Thus,  for values of k greater than or equal to 0.4 the
examination of  10  grid  openings  in this study  yields an airborne
asbestos   concentration   that   is  sufficiently  precise   to
distinguish  0  s/cc from 0.009  s/cc with high  probability.   (In
this  study  one  structure  corresponds  to approximately  0.003
s/cc.)

          The data indicate that k is  usually greater than 0.4,
but  that  smaller values,  such as k=0.07  are  possible.    The
standard  deviation of  this estimate of k is 0.9.   For k-0.07, a
95%  confidence interval  for the  true  structure count  when no
structures are  counted in  10  grid openings is  (0,50).   If the
number  of   grid   openings  counted  is  increased  to  50,  the
confidence interval shrinks to (0,4.7).

          Of the  16 filters examined, all  but one indicate that
examination  of  10 grid openings  is sufficient to distinguish 0
s/cc  from 0.009 s/cc with high probability.   Although, without
additional  data,  it  is   difficult  to predict   how  frequently
exceptions will occur, the  results suggest that examination of
additional grid openings  is generally  unnecessary unless higher
precision is required.
                               159

-------
        APPENDIX G



AIR MONITORING DATA LISTING

-------
                                           TABLE G-l. AIR MONITORING DATA LISTING SHOWING THE ASBESTOS
                                                      STRUCTURE CONCENTRATION (S/CC) AT EACH SITE THAT WAS
                                                      AIR SAMPLED.  THE "0" SITE AT EACH BUILDING*IS
                                                      ALWAYS THE OUTDOOR LOCATION.  SITES 1-7 DO NOT
                                                      CORRESPOND TO THE SITE NUMBERING USED IN APPENDICES
                                                      A AND B.
                                                	 BUILDING NUMBER =1 BUILDING CATEGORY
                                                                    SITE    STRUCTURE
                                                                            CONCENTRATION

                                                                       0            0.000
                                                                       1            0.001
                                                                       2            0.000
                                                                       3            0.000
                                                                       4            0.000
                                                                       6            0.000
                                                                       6            0.000
                                                                       7            0.000
                                                       BUILDING NUMBER =2 BUILDING CATEGORY =1
u>
                                                                    SITE    STRUCTURE
                                                                            CONCENTRATION
1
2
3
4
5
6
7
                                                                                      002
                                                                                      000
                                                                                      033
                                                       BUILDING NUMBER =3 BUILDING CATEGORY el
                                                                    SITE
                                                                       0
                                                                       1
                                                                       2
                                                                       3
                                                                       4
                                                                       6
                                                                       6
                                                                       7
     STRUCTURE
     CONCENTRATION
               .000
               006
               000
               000
               000
               000
             0.000

-------
                                          TABLE G-l. AIR MONITORING DATA  LISTING  SHOWING THE  ASBESTOS
                                                     STRUCTURE CONCENTRATION  (S/CC)  AT  EACH SITE  THAT WAS
                                                     AIR SAMPLED.  THE  "0"  SITE AT EACH BUILDING  IS
                                                     ALWAYS THE  OUTDOOR LOCATION.  SITES 1-7  DO NOT
                                                     CORRESPOND  TO THE  SITE NUMBERING USED IN APPENDICES
                                                     A AND B.
                                                      BUILDING NUMBER  =4  BUILDING  CATEGORY
                                                                    SITE    STRUCTURE
                                                                            CONCENTRATION

                                                                       0            0.000
                                                                       1            0.000
                                                                       2            0.000
                                                                       3            0.000
                                                                       4            0.000
                                                                       5            0 •000
                                                                       6            0.000
                                                                       7            0.000
                                                       BUILDING NUMBER =5 BUILDING CATEGORY =1
ON
SITE    STRUCTURE
        CONCENTRATION

   0            0.003
   1            0.000
   2            0.000
   3            0.000
   4            0.000
   6            0.000
   6            0.000
   7            0.000
                                                       BUILDING NUMBER =6 BUILDING CATEGORY si
                                                                    SITE    STRUCTURE
                                                                            CONCENTRATION

                                                                       0            0.000
                                                                       1            0.000
                                                                       2            0.000
                                                                       3            0.000
                                                                       4            0.000
                                                                       5            0.000
                                                                       6            0.000
                                                                       7            0.000

-------
                                           TABLE G-l.  AIR MONITORING DATA LISTING SHOWING THE ASBESTOS
                                                      STRUCTURE CONCENTRATION (S/CC) AT EACH SITE THAT WAS
                                                      AIR SAMPLED.   THE "0" SITE AT EACH BUILDING IS
                                                      ALWAYS THE OUTDOOR LOCATION.  SITES 1-7 DO NOT
                                                      CORRESPOND TO THE SITE NUMBERING USED IN APPENDICES
                                                      A AND B.
        	 BUILDING NUMBER =7 BUILDING CATEGORY *2


                                                                    SITE    STRUCTURE
                                                                            CONCENTRATION

                                                                       0            0.000
                                                                       1            0.001
                                                                       2            0.000
                                                                       3            0.000
                                                                       4            0.003
                                                                       6            0.000
                                                                       6            0.003
                                                                       7            0.000


        	 BUILDING NUMBER =8 BUILDING CATEGORY =2


                                                                    SITE    STRUCTURE
                                                                            CONCENTRATION

,_,                                                                     0            0.003
0%                                                                     1            0.004
Ui                                                                     2            0.000
                                                                       3            0.003
                                                                       4            0.003
                                                                       5            0.000
                                                                       6            0.000
                                                                       7            0.000
                                                       BUILDING NUMBER =9 BUILDING CATEGORY =2
                                                                    SITE    STRUCTURE
                                                                            CONCENTRATION

                                                                       0            0.000
                                                                       1            0.003
                                                                       2            0.000
                                                                       3            0.000
                                                                       4            0.000
                                                                       5            0.000
                                                                       6            0.000
                                                                       7            0.000

-------
TABLE G-l. AIR MONITORING DATA LISTING SHOWING THE ASBESTOS
           STRUCTURE CONCENTRATION (S/CC) AT EACH SITE THAT WAS
           AIR SAMPLED.  THE "0" SITE AT EACH BUILDING IS
           ALWAYS THE OUTDOOR LOCATION.  SITES 1-7 DO NOT
           CORRESPOND TO THE SITE NUMBERING USED IN APPENDICES
           A AND B.
            BUILDING NUMBER =10 BUILDING CATEGORY =2
                         SITE    STRUCTURE
                                 CONCENTRATION

                            0            0.000
                            1            0.000
                            2            0.000
                            3            0.000
                            4            0.000
                            5            0.000
                            6            0.000
                            7            0.000
            BUILDING NUMBER =11 BUILDING CATEGORY =2
                         SITE    STRUCTURE
                                 CONCENTRATION

                            0            0.000
                            1            0.000
                            2            0.002
                            3            0.000
                            4            0.000
                            E            0.000
                            6            0.000
                            7            0.000
         — BUILDING NUMBER =12 BUILDING CATEGORY =2
                         SITE    STRUCTURE
                                 CONCENTRATION

                            0            0.000
                            1            0.000
                            2            0.002
                            3            0.000
                            4            0.000
                            6            0.000
                            6            0.000
                            7            0.000

-------
TABLE G-l. AIR MONITORING DATA LISTING SHOWING THE ASBESTOS
           STRUCTURE CONCENTRATION  (S/CC) AT EACH SITE THAT WAS
           AIR SAMPLED.  THE "0" SITE AT EACH BUILDING IS
           ALWAYS THE OUTDOOR LOCATION.  SITES 1-7 DO NOT
           CORRESPOND TO THE SITE NUMBERING USED IN APPENDICES
           A AND B.
            BUILDING NUMBER =13 BUILDING CATEGORY =3
                         SITE    STRUCTURE
                                 CONCENTRATION

                            0            0.000
                            1            0.000
                            2            0.006
                            3            0.004
                            A            0.002
                            6            0.000
                            6            0.006
                            7            0.000
            BUILDING NUMBER =14 BUILDING CATEGORY =3
                         SITE    STRUCTURE
                                 CONCENTRATION

                            0            0.000
                            1            0.004
                            2            0.000
                            3            0.000
                            4            0.000
                            6            0.000
                            6            0.000
                            7            0.000
            BUILDING NUMBER =16 BUILDING CATEGORY =3
                         SITE    STRUCTURE
                                 CONCENTRATION

                            0            0.000
                            1
                            2
                            3
                            4
                            6
                            6
.001
.003
.000
.001
.000
                            7            0.000

-------
                                           TABLE G-l.  AIR  MONITORING DATA LISTING SHOWING THE ASBESTOS
                                                      STRUCTURE CONCENTRATION (S/CC)  AT EACH SITE THAT WAS
                                                      AIR  SAMPLED.   THE "0"  SITE AT EACH BUILDING IS
                                                      ALWAYS  THE OUTDOOR LOCATION.  SITES 1-7 DO NOT
                                                      CORRESPOND TO THE SITE NUMBERING USED IN APPENDICES
                                                      A AND B.
                                                       BUILDING NUMBER =16 BUILDING CATEGORY =3
                                                                    SITE    STRUCTURE
                                                                            CONCENTRATION

                                                                       0            0.000
                                                                       1            0.001
                                                                       2            0.000
                                                                       3            0.002
                                                                       4            0.000
                                                                       5            0.000
                                                                       6            0.000
                                                                       7            0.003
                                                       BUILDING NUMBER =17 BUILDING CATEGORY =3
ON
00
                                                                    SITE
0
1
2
3
4
5
6
7
     STRUCTURE
     CONCENTRATION
                                                                                     ,000
                                                                                     ,000
                                                                                     ,000
                                                                                     ,000
                                                                                     ,000
                                                                                     ,000
                                                       BUILDING NUMBER =18 BUILDING CATEGORY =3
                                                                    SITE    STRUCTURE
                                                                            CONCENTRATION

                                                                       0            0 «000
                                                                       1            0.000
                                                                       2            0.000
                                                                       3            0.000
                                                                       4            0.000
                                                                       5            0.003
                                                                       6            0.003

-------
                                           TABLE G-l. AIR MONITORING DATA LISTING SHOWING THE ASBESTOS
                                                      STRUCTURE CONCENTRATION (S/CC) AT EACH SITE THAT WAS
                                                      AIR SAMPLED.  THE ••• SITE AT EACH BUILDING IS
                                                      ALWAYS THE OUTDOOR LOCATION.  SITES 1-7 DO NOT
                                                      CORRESPOND TO THE SITE NUMBERING USED IN APPENDICES
                                                      A AND B.
                                                       BUILDING NUMBER =19 BUILDING CATEGORY =3


                                                                    SITE    STRUCTURE
                                                                            CONCENTRATION

                                                                       0            0.000
                                                                       1            0.000
                                                                       2            0.003
                                                                       3            0 «000
                                                                       4            0.000
                                                                       o            0«000
                                                                       6            0.000
                                                                       7            0.000


                                                       BUILDING NUMBER =20 BUILDING CATEGORY =3


                                                                    SITE    STRUCTURE
                                                                            CONCENTRATION

_                                                                     0            0.000
£                                                                     1            0.000
Co                                                                     2            0.000
                                                                       3            0.003
                                                                       4            0.000
                                                                       6            0.000
                                                                       6            0.000
                                                                       7            0.000
                                                       BUILDING NUMBER =21 BUILDING CATEGORY =3
                                                                    SITE    STRUCTURE
                                                                            CONCENTRATION

                                                                       0            0.000
                                                                       1            0.000
                                                                       2            0.000
                                                                       3            0•000
                                                                       4            0.000
                                                                       5            0 »000
                                                                       6            0.000
                                                                       7            0.000

-------
TABLE G-l. AIR MONITORING DATA LISTING SHOWING THE ASBESTOS
           STRUCTURE CONCENTRATION (S/CC) AT EACH SITE THAT WAS
           AIR SAMPLED.  THE "0" SITE AT EACH BUILDING IS
           ALWAYS THE OUTDOOR LOCATION.  SITES 1-7 DO NOT
           CORRESPOND TO THE SITE NUMBERING USED IN APPENDICES
           A AND B.
            BUILDING NUMBER =22 BUILDING CATEGORY =3
                         SITE    STRUCTURE
                                 CONCENTRATION

                            0            0.000
                            1            0.000
                            2            0 • 000
                            3            0.000
                            4            0.000
                            E            0.000
                            6            0.000
                            7            0.000
            BUILDING NUMBER =23 BUILDING CATEGORY =3
                         SITE    STRUCTURE
                                 CONCENTRATION

                            0            0 • 000
                            1            0.003
                            2            0.013
                            3            0.000
                            4            0.000
                            6            0.003
                            6
                            7            0.000
          - BUILDING NUMBER =24 BUILDING CATEGORY =3
                         SITE    STRUCTURE
                                 CONCENTRATION

                            0            0.000
                            1            0.000
                            2            0.000
                            3            0.000
                            4            0.000
                            6            0.000
                            6            0.000
                            7            0.000

-------
TABLE G-l. AIR MONITORING DATA LISTING SHOWING THE ASBESTOS
           STRUCTURE CONCENTRATION (S/CC) AT EACH SITE THAT WAS
           AIR SAMPLED.  THE "0" SITE AT EACH BUILDING IS
           ALWAYS THE OUTDOOR LOCATION.  SITES 1-7 DO NOT
           CORRESPOND TO THE SITE NUMBERING USED IN APPENDICES
           A AND B.
            BUILDING NUMBER =25 BUILDING CATEGORY =3
                         SITE
                            0
                            1
                            2
                            3
                            4
                            6
                            6
                            7
STRUCTURE
CONCENTRATION
         .000
         .000
         .000
         .013
         .000
         .000
        0.000
            BUILDING NUMBER =26 BUILDING CATEGORY =3 	
                         SITE    STRUCTURE
                                 CONCENTRATION

                            0            0 *000
                            1            0.003
                            2            0.000
                            3            0.000
                            4            0.000
                            O            0•000
                            6            0.000
                            7            0.000
            BUILDING NUMBER =27 BUILDING CATEGORY =3
                         SITE    STRUCTURE
                                 CONCENTRATION

                            0             .000
                            1             .000
                            2             .000
                            3             .000
                            4             .000
                            5             • 00o
                            0             • 000
                            7             .000

-------
TABLE G-l. AIR MONITORING DATA LISTING SHOWING THE ASBESTOS
           STRUCTURE CONCENTRATION (S/CC) AT EACH SITE THAT WAS
           AIR SAMPLED.  THE "0" SITE AT EACH BUILDING IS
           ALWAYS THE OUTDOOR LOCATION.  SITES 1-7 DO NOT
           CORRESPOND TO THE SITE NUMBERING USED IN APPENDICES
           A AND B.
   	 BUILDING NUMBER =28 BUILDING CATEGORY =3 	
                         SITE    STRUCTURE
                                 CONCENTRATION

                            0            0 .000
                            1            0.000
                            2            0.000
                            3            0 *000
                            A            0.000
                            6            0.008
6                                         A AAA
                                         0 • 0190
                            7            0.000
            BUILDING NUMBER =29 BUILDING CATEGORY «3 ----
                         SITE    STRUCTURE
                                 CONCENTRATION

                            0            0.000
                            1            0.003
                            2            0.000
                            3            0.000
                            4            0.000
                            B            0.000
                            6            0.003
                            7            0.000
      	 BUILDING NUMBER =30 BUILDING CATEGORY =3 	
                         SITE    STRUCTURE
                                 CONCENTRATION

                            0            0.000
                            1            0.007
                            2            0.000
                            3            0.000
                            4            0.000
                            5            0.000
                            6            0.000
                            7            0.000

-------
                                          TABLE G-l. AIR MONITORING DATA LISTING SHOWING THE ASBESTOS
                                                     STRUCTURE CONCENTRATION  (S/CC) AT EACH SITE THAT WAS
                                                     AIR SAMPLED.  THE "0" SITE AT EACH BUILDING IS
                                                     ALWAYS THE OUTDOOR LOCATION.  SITES 1-7 DO NOT
                                                     CORRESPOND TO THE SITE NUMBERING USED IN APPENDICES
                                                     A AND B.
                                                      BUILDING NUMBER =31 BUILDING CATEGORY =3 —


                                                                   SITE    STRUCTURE
                                                                           CONCENTRATION

                                                                      0            0.000
                                                                      1            0.009
                                                                      2            0.000
                                                                      3            0 • 00D
                                                                      4            0.000
                                                                      o            0.000
                                                                      6            0 • 000
                                                                      7            0.000


                                                      BUILDING NUMBER =32 BUILDING CATEGORY =3 	


                                                                   SITE    STRUCTURE
                                                                           CONCENTRATION

                                                                      0            0.000
h->                                                                    1            0.000
-J                                                                    2            0.000
1°                                                                    3            0.000
                                                                      4            0.000
                                                                      5            0.000
                                                                      6            0.000
                                                                      7            0.
                                                      BUILDING NUMBER =33 BUILDING CATEGORY =3 —
                                                                   SITE    STRUCTURE
                                                                           CONCENTRATION

                                                                      0            0.003
                                                                      1            0.000
                                                                      2            0.000
                                                                      3            0.000
                                                                      4            0.003
                                                                      5            0 • 000
                                                                      6            0.003
                                                                      7            0.000

-------
TABLE G-l. AIR MONITORING DATA LISTING SHOWING THE ASBESTOS
           STRUCTURE CONCENTRATION  (S/CC) AT EACH SITE THAT WAS
           AIR SAMPLED.  THE "0" SITE AT EACH BUILDING IS
           ALWAYS THE OUTDOOR LOCATION.  SITES 1-7 DO NOT
           CORRESPOND TO THE SITE NUMBERING USED IN  APPENDICES
           A AND B.
 	 BUILDING NUMBER =34 BUILDING CATEGORY =3
                         SITE    STRUCTURE
                                 CONCENTRATION

                            0            0.000
                            1            0.000
                            2            0.000
                            3            0.000
                            4            0.000
                            6            0.000
                            6            0 *000
                            7            0.000
            BUILDING NUMBER =35 BUILDING  CATEGORY  =3
                         SITE    STRUCTURE
                                 CONCENTRATION

                            0            0.000
                            1            0.000
                            2            0.000
                            3            0.000
                            4            0.000
                            E            0.000
                            6            0.003
                            7            0.000
            BUILDING NUMBER =36  BUILDING CATEGORY =3 	
                         SITE    STRUCTURE
                                  CONCENTRATION

                             0            0.000
                             1            0.000
                             2            0.000
                             3            0.000
                             4            0.000
                             6            0.000
                             6            0.000
                             7            0.003

-------
                                          TABLE 6-1. AIR MONITORING DATA LISTING SHOWING THE ASBESTOS
                                                     STRUCTURE CONCENTRATION (S/CC) AT EACH SITE THAT WAS
                                                     AIR SAMPLED.  THE "0" SITE AT EACH BUILDING IS
                                                     ALWAYS THE OUTDOOR LOCATION.  SITES 1-7 DO NOT
                                                     CORRESPOND TO THE SITE NUMBERING USED IN APPENDICES
                                                     A AND B.
                                      	 BUILDING NUMBER =37 BUILDING CATEGORY =3 —
                                                                   SITE    STRUCTURE
                                                                           CONCENTRATION

                                                                      0            0.000
                                                                      1            0.000
                                                                      2            0.006
                                                                      3            0.005
                                                                      4            0.000
                                                                      5            0.000
                                                                      6            0.000
                                                                      7            0.000
        	 BUILDING NUMBER =3B BUILDING CATEGORY =3


                                                                   SITE    STRUCTURE
                                                                           CONCENTRATION

                                                                      0            0.002
,_                                                                    1            0.000
Cj                                                                    2            0.004
Ui                                                                    3            0.002
                                                                      4            0.002
                                                                      6            0.000
                                                                      6            0.003
                                                                      7            0.003
                                                      BUILDING NUMBER =39 BUILDING CATEGORY =3
                                                                   SITE    STRUCTURE
                                                                           CONCENTRATION

                                                                      0            0.003
                                                                      1            0 •000
                                                                      2            0.000
                                                                      3            0.000
                                                                      4            0.002
                                                                      5            0.000
                                                                      6            0.003
                                                                      7            0.000

-------
                                          TABLE G-l. AIR MONITORING DATA LISTING SHOWING THE ASBESTOS
                                                     STRUCTURE CONCENTRATION  (S/CC) AT EACH SITE THAT WAS
                                                     AIR SAMPLED.  THE "0" SITE AT EACH BUILDING IS
                                                     ALWAYS THE OUTDOOR LOCATION.  SITES 1-7 DO NOT
                                                     CORRESPOND TO THE SITE NUMBERING USED IN APPENDICES
                                                     A AND B.
                                                   -- BUILDING NUMBER =40 BUILDING CATEGORY =3
                                                                   SITE    STRUCTURE
                                                                           CONCENTRATION

                                                                      0            0.000
                                                                      1            0.000
                                                                      2            0.000
                                                                      3            0.000
                                                                      4            0.000
                                                                      5            0.002
                                                                      6            0.000
                                                                      7            0.000
                                                      BUILDING NUMBER =41 BUILDING CATEGORY =3
ON
SITE    STRUCTURE
        CONCENTRATION

   0            0.000
   1            0.000
   2            0.000
   3            0.000
   4            0.000
   6            0.000
   6            0.000
   7            0.000
                                                       BUILDING NUMBER  =42  BUILDING CATEGORY  =3
                                                                    SITE
                                                                       0
                                                                       1
                                                                       2
                                                                       3
                                                                       4
                                                                       5
                                                                       6
                                                                       7
        STRUCTURE
        CONCENTRATION

                0.000
                0.000
                0.000
                0.000
                0.000
                0.000
                0.000

-------
TABLE G-l. AIR MONITORING DATA LISTING SHOWING THE ASBESTOS
           STRUCTURE CONCENTRATION fS/CC) AT EACH SITE THAT WAS
           AIR SAMPLED.  THE "0" SITE AT EACH BUILDING IS
           ALWAYS THE OUTDOOR LOCATION.  SITES 1-7 DO NOT
           CORRESPOND TO THE SITE NUMBERING USED IN APPENDICES
           A AND B.
            BUILDING NUMBER =43 BUILDING CATEGORY *3
                         SITE    STRUCTURE
                                 CONCENTRATION

                            0            0.002
                            1
                            2
                            3            0.003
                            4            0.000
                            D            0 »000
                            6            0.000
                            7            0.000
            BUILDING NUMBER =44 BUILDING CATEGORY »3 --------
                         SITE    STRUCTURE
                                 CONCENTRATION
                            i            0.000
                            2            0.000
                            3            0.000
                            4            0.000
                            5            0.000
                            6            0.000
                            7            0.000
            BUILDING NUMBER =46 BUILDING CATEGORY =3 	
                         SITE    STRUCTURE
                                 CONCENTRATION
                            0            0.
                            1            0.000
                            2            0.000
                            3            0.003
                            4            0.000
                            5            0.003
                            6            0.000
                            7            0.000

-------
TABLE G-l. AIR MONITORING DATA LISTING SHOWING THE ASBESTOS
           STRUCTURE CONCENTRATION  (S/CC) AT EACH SITE THAT WAS
           AIR SAMPLED.  THE "0" SITE AT EACH BUILDING IS
           ALWAYS THE OUTDOOR LOCATION.  SITES 1-7 DO NOT
           CORRESPOND TO THE SITE NUMBERING USED IN APPENDICES
           A AND B.
            BUILDING NUMBER =46 BUILDING CATEGORY =3
                         SITE    STRUCTURE
                                 CONCENTRATION

                            0            0.002
                            1            0.000
                            2            0.000
                            3            0.000
                            4            0.000
                            5            0.000
                            6            0.000
                            7            0.005
            BUILDING NUMBER =47 BUILDING CATEGORY =3
                         SITE    STRUCTURE
                                 CONCENTRATION

                            0            0.000
                            1            0.000
                            2            0.006
                            3            0.003
                            4            0.000
                            6            0.000
                            6            0.000
                            7            0.003
            BUILDING NUMBER =48 BUILDING CATEGORY =3
                         SITE    STRUCTURE
                                 CONCENTRATION

                            0            0.000
                            1            0.000
                            2            0.000
                            3            0.000
                            4            0.000
                            6            0.000
                            6            0.000
                            7            0.000

-------
                                          TABLE G-l.  AIR MONITORING DATA LISTING SHOWING THE ASBESTOS
                                                     STRUCTURE CONCENTRATION (S/CC)  AT EACH SITE THAT WAS
                                                     AIR SAMPLED.   THE "0«  SITE AT EACH BUILDING IS
                                                     ALWAYS THE OUTDOOR LOCATION.  SITES 1-7 DO NOT
                                                     CORRESPOND TO THE SITE NUMBERING USED IN APPENDICES
                                                     A AND B.
                                                  — BUILDING NUMBER =49 BUILDING CATEGORY =3
                                                                   SITE    STRUCTURE
                                                                           CONCENTRATION

                                                                      0            0 • 000
                                                                      1            0.002
                                                                      2            0.000
                                                                      3          ,  0.000
                                                                      4            0.000
                                                                      5            0.000
                                                                      D            0 .000
                                                                      7            0.000
vO

-------
APPENDIX H



 GLOSSARY

-------
ACM:  asbestos-containing material.

Air  sample:   a  filter  through which a  known volume of  air has
passed in  order to measure the  asbestos structure concentration
in the air during the period of sampling.

Air  flow:    an  air flow  transports fibers  from  the  point  of
release  from  the  ACM  to  other  areas  in  the building.    Air
plenums, air shafts and elevator shafts represent different types
of air flow.

Air  monitoring:   the  process  of  collecting air  samples  in  a
building.

Asbestos:  a group of  naturally occurring minerals that separate
into fibers.   There are six asbestos minerals  used commercially
(chrysotile, amosite, crocidolite,  anthophyllite,  tremolite, and
actinolite).

Bulk sample:  a portion of friable material collected in order to
measure the asbestos content of the material.

Categories of buildings:

     Category   1—a  building   in  which  no   friable  asbestos-
     containing  surfacing materials  or TSI were noted in the GSA
     records and none was found during the building inspection.

     Category  2—a building in  which all  or most of  the areas
     with  friable  asbestos-containing surfacing materials or TSI
     were  in  good condition  allowing  for a limited  number  of
     areas of moderate  damage.

     Category  3—a building which had at least one significantly
     damaged   area  of  friable   asbestos-containing  surfacing
     material  or TSI,  or there  were numerous moderately damaged
     areas.

Condition:  See Appendix C  for definitions  of ACM condition.

Disturbance:   (classifications revised March  19, 1987)
     High potential for disturbance—ACM which has two  or more of
     the  three  factors (accessibility,  vibration,  air erosion)
     rated  "high,"  or  one factor  sufficiently  high  that the
     material   is   almost  certainly  going   to  be   disturbed.
     Examples  are  (1) acoustic plaster on  a low ceiling in a high
     school band room;  (2)  thermal system  insulation  on air  ducts
     connected to  ventilation  fans and  readily  accessible to
     workers  conducting maintenance on  the  ventilation  system;
     and  (3)  fireproof ing  on  low beams  in a  work room  located
     just downstream from an air  vent.


                                183

-------
      Moderate  potential  for disturbance—ACM which is accessible,
      subject to vibration,  or subject to air erosion, but has no
      more  than one factor rated as high.  ACM on corridor walls,
      on  a  ceiling underneath  a  gymnasium, or in an elevator shaft
      are  examples  of  material with a  moderate potential  for
      disturbance.

      Low  potential   for    disturbance—ACM   which   has   low
      accessibility,  is  not  subject  to  vibration,  and  is  not
      subject to air erosion.

 External analysis:  an  analysis in  which  a sample is analyzed a
 second  time by  another analytical  laboratory.    This  type of
 analysis is  performed as a  QC  check on the  performance of the
 method by  the  primary  laboratory.  The degree  of  agreement of the
 original  analysis with the  external  analysis indicates  the
 consistency of the method performance.

 Field blank:   a filter taken into the field,  handled in  the  same
 manner   as  exposed   air  sample   filters,   and  analyzed  for
 contamination  which might occur in the field but not as  a result
 of air sampling.

 Friable:   capable of  being  crumbled, pulverized,  or reduced to
 powder by  hand pressure.

 Production lot blank:   a filter chosen prior to field work and
 analyzed by the laboratory to check  for  filter contamination.

 PLM:  polarized light  microscopy.

 Replicate  analysis:   an analysis in which a sample is analyzed a
 second  time  by the  same analytical laboratory.    The  degree of
 agreement  of   the  original  analysis  with  the  replicate  analysis
 indicates  the level  of precision   in  the  laboratory  analysis
 procedures.

 Side-by-side duplicate:   a sample  collected in the immediate  area
 of the  original  sample  but  handled separately.    The  degree of
 agreement  of   the  analyses  of  the  original  sample  with  its
 duplicate  indicates   the  level  of  precision  in  the  sample
 collection and field handling procedures.

 Structure:  An asbestos  fiber,  bundle, cluster, or matrix.

 Surfacing:     ACM  sprayed  or  troweled  on   surfaces,   such  as
 acoustical  plaster  on  ceilings  and fireproofing  material  on
 structural members.

TEM:   transmission electron microscopy.
                               184

-------
Thermal  systems  insulation:    ACM  applied to  pipes,  boilers,
tanks,  ducts,  etc.  to  prevent  heat  loss  or  gain  or  water
condensation.
 	w
condensation.


TSI:  thermal systems insulation.
                               185

-------
 50272-101
REPORT DOCUMENTATION
 	_J>AGE	
4. Title and Subtitle
                      1. REPORT NO.
                        EPA 560/5-88-002
2.
              3. Recipient's Accession No.
 Assessing Asbestos Exposure in Public Buildings
                                                              -.j	
                                                                5. Report Date

                                                                 May. 1988
 7. Author**)
        Hatfield, J. et al
 9. Performing Organization Name and Address                  '	"	
  Battelle  Columbus Division,  Washington Operations, 2P30 M St.  NW
  Washington, D.C. 20036; Price Associates,Inc.,  1825 K St, NW,
  Wash,  DC,  20006; Alliance Technologies Corp,  213 Burlington Rd,
  Bedford,  MA 07130; R. J. Lee Group, Inc., 350 Hochberg Rd, Mon-
  roeville,  PA  15146; Midwest  Research Inst., Kansas City, MO 641K>(G)
 12. Sponsoring Organization Name and Address
  U.S.  Environmental Protection Agency
  Office  of Toxic Substances
  Exposure Evaluation Division (TS-798)
  401 M St.,  S.W.  Washington, D.C.  20460
 15. Supplementary Notes
                                                               8. Performing Organization Rept. No.
                                                               10. Project/Task/Work Unit No.


                                                               "* B^-O^-^?^""^' N°'
                                                               (C) 68-02-3997
                                                                 '68-03-3406,68-02-4252
                                                               13. Type of Report & Period Covered

                                                               Peer-reviewed report
                                                               14.
 is. Abstract (Limit: 200 words) Airborne  asFes"fos  levels  were measured "by "dl'recrfc  l/ransmissio
 electron microscopy in 49 public buildings from  three categories:  (1)
 buildings  without  asbestos-containing material  (ACM); (2) buildings with all
 or  most of-the ACM in good' condition allowing for  a limited number of  areas
 of  moderate damage; and  (3) buildings which had  at least  one area of
 significantly damaged- ACM or numerous areas of moderate  damage.  Approximatel
 seven areas were monitored inside  and one area  outside each building.
 Although the absolute airborne asbestos  levels were very  low,  Category (3)
 had the highest median levels followed by Category (2),  Category (1),  and
 outdoors.   Category (3)  levels were significantly  higher  than  Category (1);

 Another objective  was to  field test an assessment  method  for ACM developed  t
 facilitate abatement decision making in  the context of an asbestos managemen
 program.   Material condition, potential  for disturbance,  and air flow  were
 assessed by trained raters in 257  areas  in 60 public buildings.   Using rater(
 consistency as an  evaluation criterion,  the three  factors showed promise as j
 assessment tools for use- in the  field.   Each factor showed statistically     .
 significant consistency  among raters.  A further observation associates      |
 disagreement among raters with  imprecision in definitions and  the absence  ofi
 proper training.
 17. Document Analysis  m. Descriptors


    Asbestos, asbestos exposure, asbestos air monitoring, asbestos assessment, TEM,
    transmission electron microscopy
   b. Identifiers/Open.Ended Terms
   c. COSATI Field/Group
 18. Availability Statement
                                                   19. Security Class (This Report)

                                                    unclassified
                                                  20. Security Class (This Page)
                                                    unclassified
                                                                         21. No. of Pages
                                                                           202
                                                                         22. Price
(See ANSI-Z39.18)
                                    See Instructions on Reverse
                                                                        OPTIONAL FORM 272 (4-77
                                                                        (Formerly NTIS-35)
                                                                                • C  ,:-,erce

-------