PUBLIC COMMENTS
Received on External Review Draft entitled, Draft Proposed Sampling Program to
Determine Extent of World Trade Center Impacts to the Indoor Environment
(EPA/600/R-04/169A)
January 2005
Background
In March 2004, EPA convened an expert technical review panel to address issues
of concern relating to the collapse of the World Trade Center (WTC) Towers. The panel
is comprised of representatives from the federal agencies directly involved in the air
quality monitoring and response and New York City Department of Health and Mental
Hygiene, the New York City Department of Environmental Protection, and outside
experts. The individual panel members have been tasked to assist the EPA in studying
two main issues: evaluating health effects attributed to exposure to WTC contaminants
through the use of health registries and related mechanisms, and implementation of a
sampling plan to determine the extent of WTC-related contamination in the indoor
environment and to assess the need for further remediation. A draft proposed sampling
plan for this latter study entitled, Draft Sampling Program to Determine Extent of World
Trade Center Impacts to the Indoor Environment was released for public comment on
October 21, 2004. The original 30-day deadline for comments was extended twice at the
request of members of the WTC community-labor coalition, and the public comment
period formally closed on January 18, 2005. To date, a total of 13 comments have been
provided. This document includes all comments received to date.
Several broad-ranging issues were raised in the comments. These issues generally
fall into one of the following categories: 1) study title and objectives; 2) geographic
extent; 3) statistical procedure for building selection; 4) proposed contaminants for
sampling; 5) sampling and analysis plans for unit sampling; 6) HVAC sampling; 7)
decision criteria for unit cleanup and building cleanup; 8) WTC signature study; 9) WTC
background study, and 10) the need for a quality assurance/quality control plan.
Three attachments included here are: 1) a listing of the individuals and
organizations who submitted comments; 2) a synopsis of key issues from these comments
as developed by EPA; and 3) the full content of all comments received.
-------
ATTACHMENT 1
LIST OF COMMENTORS
1. Theresa Perils, Ph.D, Chemical Dependency Institute of Beth Israel Medical Center,
New York, NY.
2. E-Docket: Anonymous comment focusing on the need to "document the potential
contamination in an indoor environment before we have the ability to effect a change for
the better."
3. E-Docket: Dr. Georgi Popov
4. E-Docket: Anonymous comment questioning if the USEPA believes it will get
enough buildings tested in order to get a statistically valid, geographically disbursed
sample.
5. E-Docket: Anonymous comment asking how the USEPA plans to identify and
exclude buildings that have been substantially renovated.
6. E-Docket: Handwritten note from Steve Hopkins, New Rochelle, NY, supporting the
notion of a scientifically valid study.
7. E-Docket: Comment suggesting the use of "polarized light microscopy (PLM), X-ray
fluorescence (XRF), transmission electron microscopy (EM), and scanning electron
microscopy (SEM) to identify if the WTC signature is present" submitted by Dr. Richard
Lee, President, RJ Lee Group.
8. Testimony (Sept. 13, 2004) to WTC Expert Panel from Marjorie J. Clarke, Ph.D. on
WTC contamination.
9. E-Docket: Comments submitted by The New York Environmental Law and Justice.
10. E-Docket: Comments submitted by Catherine McVay Hughes and Micki Siegel de
Hernandez on behalf of the World Trade Center Community-Labor Coalition.
11. Synthesis Report from the Community-Based Participatory Research Expert Advisory
Committee to the WTC-Community-Labor Coalition.
Note: This document was also an appendix to the WTC Community-Labor Coalition e-
docket submission.
12. E-Docket: Julie M. Panko, CIH, Managing Health Scientist, ChtmRisk, Inc.
13. E-Docket: Comment submitted by Sherrie R. Savett, Jeanne A. Markey, Michael T.
Fantini, Berger & (and) Montague, P.C. and Bert A. Blitz, Esquire, Shandell, Blitz, Blitz
-------
and Bookson LLP.
-------
ATTACHMENT 2
SYNOPSIS OF COMMENTS
EPA has developed summary statements on the primary issues raised in the public
comments. In addition to these summary statements, provided below is an example of
the actual text from received comments. These example passages demonstrate the depth
of the actual comments provided.
1. Study Title and Study Objectives
ISSUE: The title is inaccurate and inadequate. It should be specific with regard to
the geographic extent, the fact that analysis is included as well as sampling, and that it
should include a statement concerning adequacy of the cleanup for the safety of
building occupants.
Example Text:
"The proposed plan does not purport to attempt to determine the full extent of
contamination, either in terms of geographic distribution or of different types of WTC
contaminants, so in that respect the title is inaccurate. It should be revised to state the
".. .Local Geographic Extent of World Trade Center Impacts of Five Selected
Contaminants of Potential Concern (COPC)..." The words "and Analyses" should be
added to the title immediately following the word "Sampling." The proposed program is
not simply a sampling program."
ISSUE: The study objectives should clearly identify the goal of cleanup where
warranted.
Example Text:
The stated objectives of the proposal are incomplete. The primary objective of the
sampling program should be to identify habitable spaces with ongoing World Trade
Center (WTC) contamination and provide cleanup where warranted.
ISSUE: There are problems with the objectives as currently crafted. They presume the
dominance of geography while other factors may be critical, and they are also
contingent on the success of the signature study, which is doubtful.
Example Text:
The first objective may place too much emphasis on characterization of the geographic
extent of contamination, and too little emphasis on non-geographic factors such as
cleaning history. It may be more sensible to structure the proposal and objectives around
the identification and characterization of all factors that are predictive of contamination,
rather than presupposing the dominance of geography.
-------
2. Geographic Extent
ISSUE: The sampling protocol is not extensive enough to cover all areas likely
affected by the building collapse and the ensuing fires. Specifically, Brooklyn should
be included, as well as Chinatown, areas impacted by the transport of waste, and other
areas. Expanding the area of sampling would obviate the need for a Phase II.
Example Text:
The proposal does not describe the rationale for excluding Brooklyn or limiting the study
area to lower Manhattan. Sampling of buildings should be much broader than planned
and should be based on the extent of the plume as determined either by the NASA photos
or other appropriate methods. There does not appear to be adequate testing of Brooklyn
to rule out contamination, and there should be identifying and sampling upwind locations
for inclusion in establishment of background levels. On perhaps the day of the most
intense emissions, September 11th, the plume can be clearly seen moving east to
southeast over Brooklyn. Newsday reported that the National Weather Service Data
indicated that the plume was over Brooklyn eighty percent of the time. It is recognized
that Manhattan was contaminated from the WTC to the East River. The East River
provides a sink for some of the dust traveling close to the surface, but the plume from the
fires easily transported to Brooklyn. So, given the distance, there may have been less of
the larger heavier particles depositing in Brooklyn than Manhattan, but the smaller fibers
are readily suspended and can travel to Brooklyn. Given the frequent wind direction to
the east and southeast, Brooklyn should be tested in Phase I.
-------
3. Statistical Procedure for Building Selection
ISSUE: Voluntary participation will likely result in a non-representative sampling,
and more importantly, may result in selecting volunteers who are more likely to have
taken preventative or remedial action (e.g., professional cleaning) already. Alternately,
buildings should be selected on a statistical random basis, to sample from a stratified
population, and then the participation of selected building should be sought.
Stratification could include building cleaning history, some modeling (e.g., plume
reconstruction) or other measure of likely contamination status. If EPA is to retain the
volunteer approach, then efforts should be made to understand the possible bias
introduced, possibly by comparing characteristics of volunteered versus non-
volunteered buildings.
Example Text:
The proposed study design collects no information on buildings that were not
volunteered, making it impossible to determine the extent of participation bias. An
alternative sampling approach is to first determine the location of each type of eligible
building in the study region and select a sample of those buildings, and then contact
building owners to request study participation. Variables that can be obtained without
access to buildings (e.g. building type, location, type of ventilation system and cleaning
history) can then be compared for volunteered and non-volunteered buildings in order to
determine whether or not volunteered buildings are likely to be representative of all
eligible buildings. This approach would also allow investigators to calculate a
participation rate, and under certain assumptions to adjust for selection bias using missing
data techniques such as the EM algorithm or multiple imputation.
ISSUE: The proposal for use of a spatially balanced sampling design is vague and
appears to be flawed. The procedure does not consider 3-dimensional space — air
intake locations on buildings are critical-and also does not consider a variety of non-
geographic factors that affect building contamination.
Example Text:
Altitude of air intake(s) may also be an important geographic factor in building
contamination, but published GRTS techniques do not consider 3-dimensional space. At
any given distance, elevation and orientation to WTC toxic sources of exposure, there are
a wide variety of non-geographic factors that will affect a building unit's accumulation
and retention of WTC toxics. Buildings, and units within buildings, are expected to vary
in their accumulation and retention of WTC toxics depending on many factors: distance,
altitude, cardinal orientation, penetration rates (i.e., how easy it is for outside toxics to
penetrate a building through closed windows, ventilation intakes, tracking in, etc.),
window usage, type of ventilation system and cleaning history.
-------
ISSUE: The spatially balanced approach will lead to a characterization of the average
extent of contamination with regard to distance and orientation to the source, but EPA
has not explained why spatial balance may be desirable in this situation. Such an
approach could avoid clustering, and if the actual building contamination occurs in
clusters in certain locations, than spatially balanced sampling would not identify such
clusters.
Example Text:
The proposed sampling plan relies on GRTS design, a sampling technique developed for
spatially balanced sampling of natural resources. GRTS combines elements of systematic
and random sampling in order to achieve a statistically efficient sample that is evenly
distributed across a 2-dimensional region. The proposal does not explain why the GRTS
technique will be used, or why spatial balance might be desirable in this
situation.. .Although spatial balance might be helpful for assessing average patterns over
a large geographic area, a non-spatially balanced approach may be more informative for
meeting other study objectives. For example, some degree of spatial clustering of
sampled buildings would provide better estimates of within-neighborhood contaminants
variability which would be useful for planning Phase II sampling and cleanup. GRTS
and other systematic sampling approaches deliberately reduce the likelihood of spatially
clustered samples and may therefore work against some study goals. Depending on
which objectives are most important, a cluster sample or a simpler stratified random
sample may be more appropriate.
-------
4. Proposed Contaminants for Sampling
ISSUE: EPA used an inadequate set of criteria for selection of contaminants of
potential concern (COPCs). It was based on frequency of detection or exceedance of a
criteria in outdoor sampling, and this would be inappropriate for indoor conditions.
Importantly, contaminants on smaller particles would likely impact the indoor
environment, and this was not considered when choosing COPCs. Shorter fibers —
fibers less than 5 um - should be counted along with longer fibers. Also, dioxin and
mercury should be included as COPCs.
Example Text:
EPA also eliminated potential COPCs if they were below a benchmark based on
proportionate mass of the sampled dust. This criterion eliminated many possible COPCs
that may be in hazardous concentrations in indoor environments since the samples
evaluated were typically from outdoor settled dust dominated by large, heavy mineral
fibers and particles (e.g., from cement and gypsum). These larger particles become
separated from the finer particulates in the indoor environment. Particulate penetration
rates of buildings are higher for small particulates than larger particulates. During typical
indoor cleaning, many of the larger particles are removed, leaving behind the smaller
airborne particulates to resettle, or adhere to surfaces. These two factors result in a
higher concentration of the smaller particulates indoors as compared to outdoors, hence
increasing the relative concentration of trace contaminants that are found on smaller
particulates owing in part to the larger surface areas characteristic of smaller particles in
contrast to equal weights of larger sized particles. Unfortunately, many of the settled dust
samples were collected by brushing or scooping up the dust, which results in the loss of
many of the finer invisible particulates since they become airborne by the process.
Alternatively, Micro Vac methods were used with a large pore size (e.g., EPA's method
used a filter >1.1 ji) that did not collect the very fine particulates, which were found by
others to be in extraordinary high concentrations.
-------
5. Sampling and Analysis Plans for Unit Sampling
ISSUE: Hard and soft surfaces should both be sampled for all contaminants. EPA
should be specific about what hard and soft surfaces to sample, and not leave it up to a
field decision. "Inaccessible" areas should be sampled, as they represent reservoirs of
contamination which may become resuspended during renovation or cleaning.
Example Text:
Particle associated lead and PAHs present in soft surfaces should then be sampled in an
identical fashion to what is proposed for asbestos, silica and MMVF. In the present
version of the proposed Plan this sampling method is a HEPA vacuum technique. XRF is
a useful method for determining metal levels in soil, and perhaps could be applied to
carpets and textiles, although at least in the case of mercury the Practical Quantification
Limits for the Niton Instruments XRF is 5-10 times greater than typical laboratory
detection limits, and therefore may not have adequate sensitivity.
ISSUE: The HEPA method is a cause for concern, mainly because it does not collect
small enough particles.
Example Text:
The HEPA method for asbestos is a cause for concern. The 2003 Background Study
seems to have used the Micro Vac method for sampling of surfaces. The HEPA method
will result in collection of excessive amounts of organic and inorganic material which
may obscure detection of short chrysotile fibers. For asbestos fiber analyses, a cleanup of
the sample by ashing followed by analyses of samples by the 'indirect method' for TEM
should be used. No details of this are given in the documents reviewed. The Micro Vac
proposed has an efficiency that will not collect particulates less than 1.1 n, and this is a
cause for concern since a great deal of the particulates are less than this size (Horgan,
unpublished observations, see answer to question 18).
The proposed use of the HEPA vacuum technique is appropriate for this application,
provided limitations inherent in the method are understood.
ISSUE: TEM should be used as an analysis method, since it has the ability to
characterize the smallest sized fibers.
Example Text: PCM should not be used (for asbestos, MMVF & silica). It is too crude
of a method to measure the thin chrystotile fibers and the shorter pulverized fibers. TEM
is a better alternative. It is essential to see the smallest of fibers. They should report all
fibers counted. ASTM has the ASTM D57656-02 Standard Test Method for
Microvacuum Sampling and Indirect Analysis of Dust by Transmission Electron
Microscopy for asbestos mass concentration, and another method for determining
asbestos structure. ASTM methods are used when available for the other COPCs and
also should be used here.
-------
6. HVAC Sampling
ISSUE: HVACs are critical reservoirs for dispersion of contaminants; the sampling
plan needs to more fully realize this by placing a high priority on HVAC sampling.
The HVAC sampling plan should be more detailed (e.g., specify what specific parts of
HVACs) and more uniform between buildings (e.g., sampling at uniform distances
within HVACs in all buildings).
Example Text:
Both hard and soft surfaces will be encountered (sic, in HVACs), so both HEPA and wipe
samples would appear appropriate. The proposal does not address the different kinds of
duct interiors that will be encountered. Some will be interior lined and some will be
exterior lined or unlined, resulting in the same hard vs. soft surface problems. In addition
some interior lined HVAC systems have tar-like waterproofing, which will likely
contribute significant background concentrations of PAHs at these locations. More
attention needs to be paid to documenting the type of system sampled.
The sample plan should designate what parts (intake, blowers, ducts, corners, splits,
diffusers, etc.) of the HVAC system should be sampled and what minimum number of
samples per sq foot need to be collected. This will also lead to a better correlation when
comparing different buildings. It would not be appropriate, for example, to compare
results for a building which had 1 sample per 50 foot of ductwork and to results for a
building which had one sample per 1,000 sq foot of ductwork.
-------
7. Decision Criteria for Unit Cleanup and Building Cleanup
ISSUE: Cleanup should proceed as soon as possible, and should be based on
measurement of COPCs, and even in the absence of a signature.
Example Text:
As a general principle, if dusts collected in an individual residence/apartment or
workplace are found to contain COPC/target parameters above threshold levels, that
residence or workplace should be cleaned. This should be the practice despite the
outcome of statistical analyses done on all samples collected in that building. These
locations could be considered "hot spots." All hot spots should be remediated. This, too,
is a common practice in the remediation of hazardous waste sites, which EPA draws
reference to for guidance in several places in the proposed sampling plan. In many types
of adverse environmental exposure, it is the people that fall into the upper tail of
probability that are exposed. This exposure is real, and so the contamination needs to be
remediated. The reason for the high level of contamination in a particular area needs to
be addressed.
ISSUE: Results from "inaccessible" areas should also be used as a basis for cleanup
decisions.
Example Text:
Since it is likely that most of the particles are of a very small size, most homeowner
vacuums are going to suck it up off the floor or out of the couch and blow it right out the
back of the vacuum. It then floats around until it settles in an inaccessible location
(where it accumulates) or an accessible location (where it is once again sucked up by the
vacuum cleaner). The result is less contaminated accessible locations, while a reservoir
accumulates in infrequently cleaned and/or inaccessible locations. Consequently results
from none of the sampling should be excluded from cleanup decision-making criteria. If
it is determined based upon program objectives and statistically based sampling design
that samples are to be collected from a particular location then the results should be
considered part of the cleanup criteria. Dust samples present in inaccessible locations,
like those found in HVAC ducts or ceiling plenums, represent the most significant
reservoirs of contaminated dusts available for introduction into residential living space
and work space alike. This applies as well to living and work space cleaned previously
and viewed as free of dust contamination. These reservoirs must receive the highest
priority in the sampling program design.
-------
ISSUE: EPA should provide a detailed rationale for all cleanup criteria. The criteria
as laid out currently are flawed. The 3X background criteria is not adequately justified
and may be inappropriate. EPA should consider the cumulative effects of being
exposed to more than one World Trade Center contaminant.
Example Text:
No consideration has been given in the sampling plan to the cumulative effects of the
COPCs or signature compounds when individual contaminants are found below
published health effects thresholds. More importantly, no consideration has been given
to the complex chemical universe present on these dusts. Many of these compounds are
not currently regulated in any fashion and the vast majority has not been adequately
assessed for health effects. Some of the organic compounds known to be associated with
the dusts (based upon published analytical data) have not been reported previously in the
environment. The WTC disaster and ensuing fire was a unique event and accordingly the
combustion chemistry in many respects was also unique. The issue of chemical mixtures
is particularly important when dealing with both carcinogenic and non-carcinogenic
substances. For example, the effects of asbestos exposure and smoking are known to be
more than additive (synergistic). The carcinogenic substances in cigarette smoke include
PAHs, which are major WTC contaminants of concern. Lead, mercury, PCBs and
dioxins are all neurobehavioral toxicants and, at present, we do not know whether their
effects are additive or synergistic. Co-planer PCBs act via the same mechanism as
PCDD/Fs that were not measured adequately in the EPA studies to determine their
combined effects. Brominated PBBs and PBDD/Fs were likely to be created in the fires
in high quantities (primarily due to PBDE fire retardants) and act like the coplanar PCBs
and PCDD/Fs, but were not measured. As a result of the failure to consider mixtures,
specific health effects are likely to be underestimated by the benchmark of one set of
contaminants. As an indication of how important the federal government considers the
issue of chemical mixtures, the Agency for Toxic Substances and Disease Registry has
released a series of draft "Interaction Profiles" as a part of their Toxicological Profiles in
2002.
ISSUE: The use of the upper confidence limit on the mean contaminant level in a
building as a cleanup criteria is not justified. Building-specific factors need to be
considered in building cleanup decisions.
Example Text:
The use of an upper confidence limit (UCL) on the mean contaminant level in a building
is not justified, and has odd implications which have probably not occurred to the
proposal authors. The use of UCLs for EPA hazardous site assessments is based on the
assumption that individuals exposed to hazardous substances at those sites are equally
likely to encounter any sampled location, so that their long term average exposures will
be well represented by averaging the available measurements. This assumption is
probably not true of most buildings in Manhattan, where individuals consistently live or
work in the same unit or on the same floor and may never visit most units in that
building. If there is any true variability in contamination across units within a building,
-------
the UCL will reflect an averaging of exposure across individuals rather than an averaging
of concentrations to which any one person might be exposed. To understand the
unintended implication of this plan, imagine one small and one large apartment building
that have identical distributions of contaminants across units. In this case, the large
building has more individuals at risk, but the smaller building is more likely to be
selected for cleanup due to a lower sample size producing a large UCL. Clearly the UCL
is not a defensible criterion in this context.
ISSUE: In addition to unit cleanup, decisions need to be made on testing other units
within a building, cleaning an entire building based on results of testing in the selected
units of the building, sampling and possibly remediating other buildings in the
neighborhood, and expanding beyond the borders of Phase I further out as part of a
Phase II testing program. EPA has not provided adequate discussions of these other
decision endpoints.
Example Text:
If the presence of contaminants has been detected in samples taken from a given building
under this program, three important decisions must be made: whether or not further
testing should be done of that building; whether or not further testing should be done in
buildings in the surrounding area; and if the site is located near the border of the Phase I
testing zone, whether or not testing should be expanded beyond that border. The EPA
proposal does not provide satisfactory answers to these questions. The answer is
relatively straightforward with regard to individual buildings. The (CBPR) Expert
Advisory Committee recommends that if units within a building tend to have similar
levels of contaminants, then the entire building should be cleaned. It notes, in addition,
that in some instances it may be more practical to clean an entire building ventilation
system regardless of variation in contamination of units in the building. The questions of
expanded neighborhood testing and expansion of testing zones are more challenging —
and yet critical to answer. The goal of this project, after all, should be to identify and
clean up all contaminated indoor spaces that threaten human health. Because the
sampling plan is not designed to promote collection of samples from multiple buildings
in the same neighborhood, the Expert Advisory Committee suggests conducting multiple
building sampling in some neighborhoods and plume corridors as a means to assess
whether data from one building predict those in neighboring buildings, and as a step
toward evaluating what factors predict area-wide contamination. It notes that EPA
probably will need to evaluate a variety of geographic and a non-geographic factor to
determine what best predicts contamination of untested buildings. We strongly urge that
EPA provide a clear plan for identifying the "next step" expanded cleaning needs.
-------
8. Signature Study
ISSUE: The "signature" is a work in progress that may or may not come to fruition.
It could differ as a function of distance from Ground Zero, particle size, dispersion
patterns, indoor versus outdoor, original source (collapse, fires, site work, etc.), and
other factors. There are several other contaminants not considered by EPA, such as
metals as promoted by R.J. Lee, or other organic compounds including PCB
congeners, PCNs, or PBDEs. Also, there could be problems with the contaminants
EPA has focused on - PAHs are also associated with transportation and other sources.
EPA has not presented anything quantitative and may not be able to in a reasonable
time frame to conduct their study. They have not provided specific details, such as the
criteria with which to evaluate the validity of a signature. Certainly a signature study
would need to be peer reviewed, further delaying its use in this program.
Example Text:
Matching of the WTC source signature to chemical signatures found in actual
environmental samples is limited by the sample types collected during the WTC disaster
and available for use in development of the source signature. For example, if only size
fractionated bulk particle samples (e.g., < 10 ji or < 2.5 ji) are available for use in
development of the WTC source signature then only samples with identical size
fractionation can now be used to develop the signature of dust samples found in living
and work spaces within affected buildings. Many chemical compounds likely to
comprise the WTC source signature (especially combustion by-products formed during
the post-9/11 fires) will not be equally distributed amongst all particle sizes. The
concentrations of chemicals (weight or mass basis) found on various particle size
fractions will vary. This is true for both WTC emissions as well as dusts now residing in
living and work spaces over three years after the 9/11 event A successful WTC
chemical signature will actually be two (or more) chemical signatures: one associated
with building collapse and a second associated with WTC fires. What constitutes a
signature may also vary with distance from the site. The best chemical signature for the
WTC fires will be comprised of a chemical compound or more likely a series of
compounds (likely combustion by-products) that can be determined with a high degree of
certainty to be unique to emissions from the WTC. Finding this signature may require
extensive chemical analyses of the WTC source sample set currently archived. Further,
the analytical procedures needed to accurately measure these compound(s) in dust
samples may not be readily available (EPA sanctioned reference methods not available)
or may be time-consuming and costly. For example, brominated aromatics may represent
one such class of compounds that apparently EPA has already taken into consideration.
-------
9. Background Study
ISSUE: Background sampling is critical. The EPA program should extend beyond
the impacted areas into background areas, and should include descriptions of how
background locations are selected. The sampling methods used in background
sampling should be identical to those used in the impacted area sampling.
Example Text:
More importantly, the EPA program design places a great deal of significance on
concentrations of COPC/signature compounds found in background buildings in
Manhattan. The EPA plan suggests a "trigger" of 3X background in affected buildings as
the basis for cleanup. As a result, it is imperative that the background determination
phase of the program results in measured concentrations of the COPC/signature
parameters (see also answer to question 4). Otherwise, the concept of measured
concentrations above threshold, when threshold is "Non-Detect" has no meaning.
-------
10. Need for a QA/QC Plan
ISSUE: There is a need for a QA/QC Plan.
Example Text:
The analytical methods are not adequately described as would be the case in a typical
EPA-sanctioned QAPP prepared for a program of this nature. For example, there are a
number of conflicts in the HEPA Vacuuming Method appended to the Plan as
Attachment 1... .The plan does not address any of the normal QA/QC issues such as
blanks, duplicates, replicates and spikes. Things like this will help to determine how well
the sample was collected from both a methods and personnel approach.
-------
ATTACHMENT 3
FULL TEXT OF COMMENTS
1. Email from Theresa Perils, Chemical Dependency Institute of Beth Israel Medical
Center, sent to EPA Region 2 (on December 29, 2004):
Subject: Draft Plan - Extent of WTC Impacts to Indoor Environment
To Members of EPA WTC Expert Technical Review Panel:
I have reviewed your Draft Proposed Sampling Plan to Determine Extent of WTC
Impacts to the Indoor Environment rather briefly, but as a Statistician I immediately
focused on the methodology of the sampling and statistical analysis. I have a few
comments as follows:
Although you have gone into great detail regarding the spatially balanced sampling
methodology for sampling of buildings, the proposal provides virtually no details about
the second stage of sampling - that of units within buildings. In the section entitled
Approach to Building Characterization you state "an appropriate number of units will be
sampled based on ". Exactly how is this "appropriate" number determined? What is the
minimum and maximum number of units per building? What is the sampling procedure
for the units? What happens if one of the selected units subsequently refuses to
participate? Note that confidence interval estimates cannot be legitimately used with
very small samples - so what do you plan to do about buildings with few units?
Your paragraph on the decision-making process for building cleanup is unclear and many
of the statements are not really accurate. One statement in particular is "The 95% UCL
defines a value that will be greater than or equal to the true mean approximately 95% of
the time in repeated sampling". In fact, the statement should say 97.5% of the time in
repeated sampling. Anyway, I have re-written the entire paragraph for you. Although I
did keep some of your phrases, I'm not really convinced that you need to include an
explanation about confidence intervals - it seems out of place here. The "Mods"
document attached shows the original paragraph and the changes, whereas the "Modsa"
document shows just the new version.
In the Decision Tree diagram some of the box titles are misleading or unclear, and there
are some inaccuracies.. This is a great shame because a really good diagram would be
very useful. I am assuming that the decision box "Building Sampling Completed" refers
to sampling of units within a single building (i.e. have all the units to be included in the
building sample actually been inspected), and that the process box "Sample Unit Area"
refers to collecting dust samples inside the unit. Based on those assumptions, at least two
flowchart paths appear to be incorrect:
-------
i) If there is no evidence of COPC exposure within the unit the flowchart branches back
to decision box "Building Sampling Completed". However, if there is evidence of COPC
exposure within the unit and WTC signature is present, the unit is cleaned and then the
flowchart branches back to collecting more dust samples in the same unit! This cannot be
right.
ii) If the building sampling is completed (i.e., all designated building units have had dust
samples collected and examined) then you compare the 95% UCL of the mean of all unit
samples with the benchmark value. If the 95% UCL is below the benchmark value you
clear the building (I suggest use of a word other than "clear" which looks too much like
"clean" on the document) and go back to selecting the next building. However, if the
95% UCL is above the benchmark value you clean the building and then proceed to
selecting more units within the same building! Surely selection and inspection of units
should be completed before a decision is made whether or not to clean the entire building.
Sincerely
Theresa Perils, Ph.D.
New York, NY 10007
See Mods.doc and Modsa.doc for edited paragraphs.
-------
2. Public E-Docket submission - Anonymous comment focusing on the need to
"document the potential contamination in an indoor environment before we have the
ability to effect a change for the better" (dated October 29, 2004):
I need help understanding why, 3 plus years later, we are going to perform this testing.
As an environmentalist who works for a very risk-averse corporation, I have always
questioned why we would document the potential contamination in an indoor
environment before we had the ability to effect a change for the better.
-------
3. Public E-Docket submission from Dr. Georgi Popov.
Comments:
1. Sample collection p. 30 The description of the method is not very clear. Why not
utilize the "collection of micro vacuum samples" from 10cm x 10 cm and 0.45 [j, m
cassette? Also, the Alsock capturing efficiency is 1.1 micrometers (not "microns" -
reference: International System of Units/ Systeme International d'Unites), and it is not
very clear what will happen to the particles that are less than 1.1 micrometers in size.
2. p. 20 Table 1 Proposed Sampling and Analytical Methods for the Building Sampling
Program.
"Settled Dust Porous Soft Surface" - Analytical method TBD: PLM/TEM method if the
sample collection is done as described on p. 28 "Interim Final WTC Residential
Confirmation Cleaning Study".
If the collection area is 10 cm x 10 cm the results could be reported as number of fibers
per mg of settled dust. A pre-weighted cassette filters could be used and from the weight
of the dust some quantities per square foot could be calculated. Therefore, number of
fibers in the sampled area. Also, particle identification could be done at the same time.
3. The standards are based on 8 hr exposure. Could that be extrapolated to 24 hr
exposure? My concern is that some of the building occupants might have been in the
building 24 hr a day. Further more; the infants breathe more air per body weight. Is there
any way to calculate their exposure?
4. On p. 28 quote "A diagram of the Nilfisk GS-80 vacuum cleaner is presented in Figure
1". - Figure 1 in this proposed document is:" Figure 1. Display of boundaries of expected
deposition based on analysis conducted by EPA's Environmental Photographic
Interpretation Center (EPIC, 2004)".
-------
4. Public E-docket submission - Anonymous comment questioning if the USEPA
believes it will get enough buildings tested in order to get a statistically valid,
geographically disbursed sample (dated November 14, 2004):
Three years after 9/11, even the most concerned downtown residents have long ago put
indoor air issues behind them - or moved - so which buildings does EPA think are going
to volunteer for testing? Not to mention downtown employers and landlords, who have
no desire to re-open this economically dangerous issue. Has EPA really thought the
dynamics of this through? A bunch of physical scientists and statisticans certainly won't
have a clue. And does EPA really believe it will get enough buildings to volunteer that it
will get a statistically valid, geographically disbursed, sample? Look at the problems the
WTC Health Care Registry had just getting folks to be interviewed despite massive
outreach. How is EPA factoring in tenant turnover - many current residents/occupants
may have no idea how things were cleaned. The lack of a commitment to clean or any
timeline to do so means virtually no one will care. So EPA will be able to say that it has
done the study the advocates demanded but that the results were inconclusive.
-------
5. Public E-Docket Submission - Anonymous comment asking how the USEPA plans to
identify and exclude buildings that have been substantially renovated (dated November
14, 2004):
One more thought: How does EPA plan to identify - and exclude - buildings that have
been substantially renovated or reconstructed? While it is unlikely that such buildings
will volunteer, given the vast amount of all types of construction activity downtown, how
is EPA going to be sure that this does not bias the sample?
-------
6. Public E-Docket submission - Handwritten note from Steve Hopkins, New Rochelle,
"76
•4]
9 IHJML
CXM^.
Ttk)
M
xWM^Ji, -^AMW-M
NY, supporting the notion of a scientifically valid study.
-------
7. Public E-Docket submission from Dr. Richard Lee, President, RJ Lee Group(dated
November 18,2004):
Comment suggesting the use of "polarized light microscopy (PLM), X-ray fluorescence
(XRF), transmission electron microscopy (EM), and scanning electron microscopy
(SEM) to identify if the if the WTC signature is present."
-------
Please feel free to let us know how we can help you to identify the WTC Dusl Markers and to
assess the geographic extent of WTC contaminants in Manhattan.
Yours very truly,
Dr. Richard Lee
President, RJ Lee Group, Inc.
End.
3SOHochbergRoad,Monroeville,PA 15146 • 724-325-1776 • 724-733-l799fax
-------
RJLeefiroup, Inc.
The Malerlals Cliaraclwlzalton Specialise
350 HdcllbcTK Rund
MiuinittViHV, PA 1 51
Tel ("24)J2M~6
FH (724) 733-
R.I Lee Group Recommendations
Summary: RJ Lee Group would like to propose a modified sample analysis plan for
determining the geographic extent of \VTG dust in Manhattan. Hie goal of the analyses
is to determine the extent and distribution of contamination by two different sources of
dust: pulverization of the building material and contents, and combustion products
produced by the fires. While PAH analyses provide a good manner of detection of
combustion products, we recommend another approach to identify the dust that was
created during the building collapses.
RJIjG has collected approximately 150,000 samples of WTC' dust and hazardous
substances from over 30 buildings in Manhattan since September 11,2001. Results of
our sample analysis have been used to determine the geographic distribution of \VTC
Dust (identified through scanning electron microscope dust characterization) and WTC
combustion products (PNAs) as illustrated in Figure 1 and Figure 2.
wSasf^^asftSWWB rC*"'" \ ^fefeKfi
. -
Figure 1. Percentage of samples testing Figure 2, Average asbestos concentrations.
positive for WTC Dust'.
'Samples positive for WTC Dust ai i diTim-d litre as samples (luit contain both mineral wool and gypsum,
and at kasi one of the following particle types: spherical iron, vesicular ctirbonaceousand chrysolite
asbestos.
Privileged and Confidential
-------
RJ LeeGroup, Inc.
The Materials Characterfxatton Specialist*
RJLG recommends a four tiered analytical plan to positively identify the presence of
WTC Dust Markers on non-porous surfaces. The four analytical procedures to use are:
1) polarized light microscopy (PLM), 2) X-ray fluorescence (XRF), 3) transmission
electron microscopy (TKM), and 4) scanning electron microscopy (SEM). The four tiered
plan requires WTC* signature dust markers to be observed before proceeding to the next
tier. The analytical procedures are conducted from lowest to highest cost; therefore
higher cost analytical procedures will not be conducted at locations in which WTC dust
markers are not observed with lower cost techniques.
WTC Dust Geographic Extent Plan: The plan is based on observations of WTC' dust
markers observed in samples collected at numerous locations throughout Manhattan,
Figure 1. These WTC signature markers are listed along with RJLG's recommended
analytical techniques in 'fable 1. Based on the large number of potential sites in the
proposed EPA Geographic Extent Survey, it would be most efficient to screen sites first
with lower cost analytical techniques before proceeding to the full suite if samples and
analyses proposed by the EPA.
Table 1: Proposed Analytical Methods for WTC' Signature Dust.
Analytical
Method
Polarized Light
Microscopy
X-Rav
Fluorescence
X-Ray
Fluorescence
Electron
Microscopv
Electron
Microscopy
W 1C Signature Dust „ . _
. , , Sample I vpe
Marker ' • '
Mineral Wool and Gypsum SEM Lifl
Principal Metals: I.,ead or Microvacvvith
Mercury 0.4 u,m PC" Filter
Other Metals: Cadmium, x ,. . ,
,., . . , Microvac with
Iron, Barium, Maneanese,
,,, - ... \r , , 0.4 um PC Filter
C hromium. /.me or Nickel ^
Asbestos Wipe
Mineral Wrool &Gypsum
Spherical Iron, Chrvsolile „ , .„
, i . ... . .,, ' . SEM Lift
Asbestos, High leinperature
Alumino-silicates, Vesicular
Carbonaceous Particles
WTC Signature
Dust C'riteria for
Positive
Both Positive,
Appendix 1
Lead or Mercurv
Positive, Appendix 2
At least three of
seven Metals
Positive. Appendix 2
Asbestos Positive,
NY FLAP 198.4
Both Positive,
Appendix 3.
At least one of four
phases Positive,
Appendix 3
Privileged and Confidential
-------
RJ LeeGroup, Inc.
The MaU'riaJs Characterization Spedalkls
Proposed WTC Gt»* Analytical Methods: As proposed by the EPA, sampling in b«low ceiling
'inaccessible areas' will be used define the extent to WTC dust distribution. The RJ Lee Group
analytical techniques are illustrated in a flow chart,
Figure 3, and are designed to precede and guide the more extensive sampling and
analytical program used to identify buildings and residences that should be cleaned.
Optical Microscopy; Polarized light microscopy (PLM)wtll he used to identify
both mineral wool and gypsum, common constituents in WTC dust, on hi\ samples
collected from non-porous surfaces. The PI,M procedure for conducting this analysis is
detailed in Appendix 1. Tile positive identification of mineral wool and gypsum on the
lift samples would necessitate the next analytical procedure. X-ray fluorescence.
X-Rav Fluorescence: XKF techniques will be used to identify heavy metals, which
are common constituents in Wit.' dust, on PC filters containing dust that was collected
using a microvac. The XRF procedure for conducting this analysis is detailed in
Appendix 2. "Hie WTC dust metal markers include lead and mercury as principle
constituents, as well as cadmium, iron, barium, manganese, chromium, zinc and nickel.
A positive WTC metals signature would include at least one of the principal metals, lead
or mercury, and at least three of the other metals cadmium, iron, barium, manganese
chromium, zinc and nickel, A positive identification ol the deimed WTC metal markers
would necessitate the next analytical procedure, transmission electron microscopy.
Transmission Electron Microscouv: A modified ASTNI D-6480 TKM analysis will
be used to determine the presence of either chrysolite or asbestiform amphibolcs in
surface dust wipe samples. The TEM procedure for conducting this analysis is New
York Environmental Laboratory Approval Program (NY F.LAP) 198.4. The presence of
any asbestos would be interpreted as a positive \VTC asbestos signature and this would
necessitate analysis, scanning electron microscopy.
Scannine Electron Microscopy (5EM): A manual 100 particle count SK.M
characterization will be performed on dust hit samples. This analysis will permit
validation of the WTC Dust Markers by identifying mineral wool, gypsum, spherical
iron, chrysotile asbestos, and vesicular carbonaceous particles. SRM analysis can also be
used to identify presence of 4 out of the 5 KPA COPCs; asbestos, man-made vitreous
fibers (MMV1-'), silica, and lead, including lead paint chips. "Hie SliM procedure for
conducting this analysis is detailed in Appendix 3.
To accurately document all collected samples and sample locations, all samples should
be documented on site using a Personal Data Assistant (PDA). Each sample will be
given a unique identification number using adhesive bar code labels that are affixed to
the sample container. Information regarding the sample including location, area
sampled, component sampled, matrix, and visual observations of the area are recorded
and stored along with the sample number. The stored information is automatically
updated into a database using an XML format.
The presence of WTC Dust markers observed by SEM, the final tier a s shown in
Privileged and Confidential
-------
"BJLeeGroup, Inc.
Th* Materials ChanicfertiatioM Specialists
Figure 3, would confirm the presence of WTC Dust in the specific building or residence.
This would initiate the proposed EPA sampling procedure in accessible areas to
determine whether the specific location should be cleaned; the absence of WTC Dust
would indicate that no additional testing or cleaning would be required at that location.
EPA Sampling Plan: The EPA proposal requires individual wipes for PAII and lead,
and complex HEPA vacuum sample collection, screen analysis and sample preparation
and analysis for asbestos, MMVF and silica. The PAH and lead analyses of wipe
samples can be evaluated with respect to health risk-based benchmarks, but they do not
by themselves, identify the presence of WTC dust. The EPA approach requires the
complete suite of samples merely to survey the extent of WTC Dust distribution.
Without first screening sites for the presence of WTC? dust, this will be a costly and
expensive approach due to the numerous 'non-deteets' that will be found near the
perimeter of the WTC Dust distribution. Screening for WTC dust will quickly identify
locations that are positive for W1X" dust and outline the geographic extent of WTC dust
in Manhattan in a less costlv manner.
Privileged and Confidential
-------
RJ LeeGrouo. Inc.
The Materials CharaclerbaOoil Specialists
EPAAnalytical Plan for WTC Dust
Not WTC Dust
Pohrized Light Microscopy
Mineral Woo! and
X-Ray Fluorescence
Isaa a Mercuy AND a ie»9 Knee ot Ihe loloraig
mei^te Cadmium, don. Banwi.
are or Nckel
Trim mission Electron Microscopy
Aebestos
Scanning Electron Microscopy
MineralWoot and GypsumiJfi8"«*I oneoline
lotomng ptases Spheritjl tan, O»ysrtile, *(h-
Temperature Mumtntm Silicatts, arid ^BSictJ^r
Carbonaceous Pedicles
WTC Dust Continued
Figures. Proposed Analytical ProcessPlowChart
Privileged and Confidential
-------
RJ Leedroup, Inc.
The Mstftrljih CdanicterlzaUon Special! sis
Appendix 1 Polarized Light Microscopy
Dust Charauteri/ation of Lift Samples by PLM
Dust mounted on a sampling strip can he optically analyzed using a Polarized Light
Microscope (PI AI \ While particles can be een using a stereo microscope, the resolution
is not high enough to facilitate accurate characterization (Figure 4) •
Figure 4, Dust by stereomicroscope. Note thin straight fibers that can either be mineral wool
or Fibrous glass. Also, smaller participate is difficult to identify.
When anaty/.ing a dust lift sample by PLM, a small piece ol the lift can he removed Irom
the sample and placed on a glass slide, submerged in 1.550 index oil and covered with a
glass cover slip. The lift itself has a refractive index of 1.550, and therefore, becomes
invisible when submerged m the 1.550 index oil. Analy/ing the ltf\ sample in a higher or
lower index oil would make the lift visible and could interfere with the analyses.
By PLM, mineral wool and fibrous glass can be readily distinguished from one another
because fibrous glass has a refractive index that is close to 1.550 (will appear taint or
invisible in 1.550 index oil) while mineral wool has a higher index of 1.6 to 1 .X(will be
visible with dark edges). Additionally, the morphology of the two types of fibers can be
observed and aid in the fiber eharacleri/.alion as fiber glass is straight with a constant
diameter and mineral wool can be bent or curved and can occur in a range of diameters
(Figure 5).
Privileged and Confidential
-------
RJ LeeGroup, Inc.
The Materials Charaetc-risiitton Special! *t
Figure 5. Mineral woo) and glass fibers on dust lift submerged in 1.550 index oil, Ixft image
shows dislinet mineral wool fibers with dark edges and curved appearance. Right image is the
same field of view with the amount of transmitted light reduced. Fibrous glass can now he-
observed,(field of view = 1.0mm)
In addition to mineral wool, other WTC Markers can be identified by PLM. Gypsum can
he identified by its angular crystal morphology or as very small particles due to
pulverization or the processing that it underwent during the manufacturing of the
source building product (Figure 6). Also, while the chemical composition can not be
identified, opaque spherical particles can also be observed (Figure 7).
F'igureS. Gypsum particles coatinga mineral wool fiber as seen in 1.550 index oil
(Fieldof view = 0.5mm).
Privileged and Confidential
-------
RJ LeeGroup, Inc.
The Miiierisils Chaiuccerizatkm Specialists
Figure 7. Spherical opaque particle observed by PLM in 1.550 index oil
(Field rf view = 0,25mm).
In contrast to lift samples containing WTC Dust, samples that are negative for WTC
Dust are generally absent of visible WTC Dust Markers, such as mineral wool, gypsum
and spherical particles (Figure 8)
*
Figure 8. Lift sample in 1.550 index oil. Sample contains fine paniculate and cellulose fibers and
does not contain mineral wool or other WTC' Markers (Field of view = 1,5mm).
While PLM can be a useful tool in rapid dust characterization of lift samples, there are a
few drawbacks to the technique. First, the lift itself has a 1.550 index if other index oils
are desired to identify different particle types, the lift would interfere with the analysis.
Also, particle identification can be hindered by the optical properties of the lift when the
stage is rotated, as the light travels through the lift in different manners depending on its
orientation.
Privileged and Confidential
-------
KJLeeGroup, Inc.
The Materials Characterization Specialists
Appendix 2; X-ray Fluorescence
Metals Analysis by X-ra> Fluorescence
A rapid procedure for the presence of potential WTC dust can be determined by
Wavelength Dispersive X-ray Fluorescence Spectroscopy (WDXRF). 'Hie procedure is
two phase. The first is to determine the presence or absence ci lead (Pb) or mercury
(Hg). W a sample is positive for lead or mercury, the sample is then scanned for
chromium (Cr), manganese (Mn), iron (Fe). nickel (Ni), zinc (Zn). cadmium (Cd), and
barium (Ba). If a sample is positive for at least three of the seven elements the sample is
considered positive for WTC dust. If a sample is negative for lead and mercury, then the
sample is not considered WTC' dust.
Potential WTC dust is collected by microvac or some other standard bulk sampling
technique. The dust is placed into a 31 mm polyethylene X-ray cell. A Mylar or other X--
ray transparent film is placed over the opening and sealed with a snap-ring. The X-ray
cell is inverted and placed into a sample cup for analysis.
'Hie samples are run on a Brukcr AXS S4 Explorer WDXRF. The qualitative elemental
analysis is done automatically, but each scan is also visually analyzed to verify the
elemental components. All results and raw data is saved electronically. Known
reference materials are run along with the samples for quality control.
Privileged and Confidential
-------
"RJLeeCHroup, Inc.
The Mulrrfiiis diarsicli-rlzaliou SpcrUIUls
Appendix 3: Scanning Electron Microscopy
DusC Characterization of i ill Samples by SEM
The presence or absence of WTC Dust markers can he determined using a scanning
electron microscope (SEM) with energy dispersive spectroscopy (EDS) using a two
phase analysis procedure. First, 100 particles are identified using chemical and
morphological characteristics to determine the volume percentages of the particle types
present in the sample. Second, the sample is scanned for the presence d" WTC Dust
Markers that may occur in amounts less ""iyrL 1% on the sample. Particle counts and
example images are recorded in a directory for each sample analyzed.
For a sample to be considered WTC* Dust positive, it must contain gypsum, mineral
wool (Figure 9 and Figure 10), and one of the following: chrysolite asbestos, spherical
iron particle, and vesicular carbonaceous particle, that have been exposed to high
temperatures (Figure 11 to Figure 13). The following provides an example image of each
of the WTC Dust markers as they appear by SEM/EDS.
SEM/EDS image of a mini-nil wool
fiber.
Figure 10. SEM/EDS image of a gypsum
particle.
Privileged and Confidential
10
-------
RJ LeeGroup, Inc.
The MatctLib Characterization Specialists
Figure 11. SEM/EDS image Ufa chrjsolilc I
-------
8. Comments submitted by Marjorie J. Clarke, Ph.D.:
Testimony to WTC Expert Panel on WTC contamination on September 13, 2004:
The signature - a diversion
The EPA WTC advisory panel has had a fixation on developing a signature for WTC dust
from the beginning. Almost every panel meeting has had presentations in one way or
another to settle on one or two pollutants, that if present in particular quantity, signifies
WTC dust. This is a diversion from the real task at hand. EPA as the ultimate protector
of the environment of the country, and the agency with the most resources to do so,
should have and should be moving expeditiously to identify indoor locations in which
there are hazardous levels of contaminants, and to clean them up. It would be a huge
mistake that if EPA's sampling program finds toxic dust, that is, dust containing
exceedances of health-based benchmarks for one or more contaminants of potential
concern as defined in a prior EPA report (COPC), including mercury, EPA does not
clean up if their limited signature pollutants (let's say just vitreous fibers and PAHs) are
not present in sufficient quantities. There could be large amounts of lead or asbestos,
cadmium or dioxin, but if the signature pollutants are below EPA's trigger, it looks like
the location would be certified as clean. This is unacceptable. It would also be a great
quote for a newspaper: "EPA won't clean up toxic dust." Why wouldn't the human
health impact of whatever dust is found be grounds for a remediation? Even basic
criteria for cleanup on exceedance by any single COPC including mercury, ignores any
synergy or multiplied impact of these complex dusts on health, so I argue that a safety
factor needs to be included in any benchmarks used for defining these cleanup criteria.
Not only can a signature be misused to preclude cleanup of toxic dust, the use of a
signature can be misunderstood by the public as lacking in COPCs for which a signature
was lacking (e.g., lead, mercury, vitreous fibers). The objective should be to clean up,
identify failures in emergency response, write up and publicize 'Lessons Learned' so that
in the event of future environmental disaster, we don't make the same mistakes again. I
recommend that EPA focus on measuring the pollutants we can measure, see if COPCs
are in hazardous concentrations, and if so, schedule a thorough remediation.
How do we determine if there is a hazardous concentration of a COPC including mercury
present in a dust? Despite the fact that EPA has yet to set standards for contaminants in
settled dust, except for one, EPA does have other means of determining when soils are
contaminated (e.g. to target Superfund sites), and when incinerator ash is hazardous.
EPA should make use of all its own resources to establish benchmarks for toxicity of the
dust, and apply a safety factor since these many hazardous air pollutants (the COPCs plus
mercury) undoubtedly work together in synergy along with pH and other factors of the
dust to produce much greater health impacts than any one of them would alone.
Sampling Universe
Another serious issue is how to sample - which buildings, which units within buildings,
and which locations within units are to be sampled? On p.4. of the Proposed Monitoring
Program to Determine Extent of WTC Impact, September 1, 2004, the criteria for
building selection are discussed. But the list of buildings eligible for sampling, which is
-------
already being generated assuming that there will be 8-hour, modified aggressive air
sampling, will be a smaller list than if EPA indicates that it's only planning to gather dust.
By the end of the September 13 meeting, it appeared as if the panel supported only the
gathering of dust to determine if a building needs remediation. It is imperative to stop
this identification - exclusion - of buildings until it is clear what type of testing will be
done. It is also imperative that EPA include all buildings in its sampling universe that
may have WTC dust. This would include but not be limited to buildings under the plume
as seen in aerial photographs.
There is also a serious problem with relying on a self-selected sample of volunteer
buildings. For one, those buildings that were never cleaned would be undersampled
because their landlords had been asked to certify to the NYC DEP that their buildings had
been remediated. Such landlords would not want to be caught in a lie. Also, in p.6
paragraph 2 of the proposed monitoring program it is stated that dust samples taken from
inaccessible areas will be caveated. Since I think I heard that samples will be taken from
accumulations of undisturbed dust, where possible, does that mean that EPA will say that
data collected in this study is of little value, and therefore an excuse not to do cleaning?
Second, in this paragraph, results of sampling will only be shared with building owners.
And if there is an exceedance, recommendations for cleaning again is given only to
owners, and an offer to clean is only given to owners. I believe it is unethical to withhold
health information from tenants (the residents and workers). Owners have liabilities as I
described, and have little motivation, indeed a conflict of interest, to share results,
exceedances or offers of cleaning with tenants, so even if there is a major sampling done,
and toxic dust is found, getting the cleanup done can be thwarted by the study design's
fixating on inflated importance of a signature, bias in building sampling due to its
volunteer nature, and allowing building owners total control over a decision to remediate.
I hope that these flaws in the sampling and remediation program will be remediated.
What should the objectives of this exercise be? It is essential that once completed, there
should be no toxic dust left in interiors of buildings impacted by the World Trade Center
collapses and fires. Therefore, the sampling protocol should be designed to inform where
toxic dust remediation should take place.
In addition, the results of this effort should be to inform studies and cleanups to be
performed Immediately After FUTURE environmental disasters (e.g., building collapses
and fires - due to terrorism, or even earthquakes). If there are no lessons learned from
9/11, it would be shameful and reprehensible.
-------
9. Public E-docket comment submitted by The New York Environmental Law and
Justice Project (dated January 19, 2005):
COMMENTS TO THE DRAFT PROPOSED SAMPLING PROGRAM TO
DETERMINE EXTENT OF WORLD TRADE CENTER IMPACTS TO THE
INDOOR ENVIRONMENT
The New York Environmental Law and Justice Project support the recommendations by
the Technical Experts retained by the WTC community and labor coalition. We share
the sentiments raised by the coalition itself. However, we feel that it is necessary to
emphasize certain concerns that we have regarding the draft sampling plan.
1. Need to have a plan that actually helps
Many agency plans to sample and cleanup often suffer criticism that the plans
actually are designed to find nothing and placate the public that there is no
problem. Whether such a problem arises from inherent bias, scientific flaw, or lack
of enforcement, such a problem seriously endangers a relationship between the
agency and the public, in addition a failure to protect the public health. It would be
prudent for the EPA to avoid any such potential criticism from the public by starting
to address these areas of concern appropriately.
a. include Cleanup as part of the plan in the title and objective : plan to sample
without a guarantee of cleanup is not going to induce participation.
b. Enforcement / Quality control: the previous indoor residence cleanup program
suffered much criticism that the sampling / cleanup activity was not adequately
supervised and was performed by untrained, unmonitored and insufficiently protected
workers. Review procedure for the actual sampling activity and cleanup activity (we will
assume that cleanup activity indeed will be the main component of this study) must be
"set in stone" in a thorough manner in order to assure the public that the study is
adequately monitored, (see OSHA and DOL standards for proper asbestos abatement)
c access/ sample size issues / voluntary bias issues: IT IS VERY UNCERTAIN
THAT THE EPA WILL OBTAIN ADEQUATE SAMPLES TO REALLY FIND
ANYTHING USEFUL. The way the plan is written currently, it never specifies what
would be an adequate sample size. The plan lacks in discussion of how EPA will make
the initiative to solicit volunteers/negotiate for access. This bias created by voluntary
sampling will probably be discussed in depth in other comments. We just want to note
that EPA needs to take account of the bias and deal with them accordingly with
appropriate statistical tools, but not at all to exclude an opportunity to sample and
cleanup.
d. signature study: comments from the community coalition and the community
technical experts will have explored this section in depth. We share the opinion that the
current signature study is not going to be successful due to the simple fact that there is
such a wide variety of WTC dust chemical compositions depending on buildings and in
geographic areas. We would like to see sampling and cleanup plan that is less dependent
on the signature study.
-------
e. meaningful background levels: background studies mentioned should be carefully
reviewed. Background samples should have been collected in the same methodology this
draft plan is using for the sampling.
2. Need to assume that there is contamination.
We suggest an approach to assume that there is WTC contamination unless disproved by
the sampling than the other way around. This is not a new approach. In the DEP housing
survey, the DEP indicated to the landlords to assume that there is WTC contamination in
the building unless the landlords prove otherwise. The letter from OSHA's HENSHAW
(DOL Assistant Secretary) stated likewise. By changing our assumption, we shift the
burden of proof to the EPA and landlords to prove that there is no contamination in that
particular building. If the lack of sampling size and other technical problems fail to
adequately detect the levels of WTC contaminants in a particular building sampled, EPA
should clean up that building.
3. Getting the public buildings tested
Many of the sample size issues and access issues can be solved if the EPA just makes the
effort to actively include the public buildings - owned or rented by the public agencies.
Public buildings should be role models for other building sampling and cleanup cases.
Postal offices / police precincts and firehouses serve as an excellent geographical
sampling points. This idea is not new. This concept has been appearing in the WTC
panel for months. The EPA needs to at least make a written request letter to enlist the
government buildings to be tested. The EPA should sample its own headquarters and
make the results known. If the letter does not initially succeed in enlisting the
government authorization, then the community can help, in addition to legal means.
Enlisting help of OSHA/NYSDOL may be crucial to get access to workplaces. Even
if there is little history of EPA/OSHA cooperation, this may be a good time to start.
The Law Project would like to finish the comment by noting that all the comments, all
the changes to the plans must be made public. All the sampling results and cleanup
results must be made public (unless otherwise provided). We realize, with the
understanding of the community and labor coalition as well, that this study plan, even at
its maximum potential, will not be able solve "all" the WTC contamination issues -
especially when Brooklyn has still been left out.
-------
10. Public E-docket comment submitted by Catherine McVay Hughes, Community
Liaison, and Micki Siegel de Hernandez, Alternate Community Liaison, on behalf of the
World Trade Center Community-Labor Coalition (dated January 18, 2005):
-------
WTC COMMUNITY-LABOR COALITION
COMMENTS ON EPA'S
"DRAFT PROPOSED SAMPLING PROGRAM TO
DETERMINE EXTENT OF WORLD TRADE CENTER
IMPACTS TO THE INDOOR ENVIRONMENT"
January 18,2005
Respectfully submitted on behalf of the WTC Community-Labor Coalition
by:
Catherine McVay Hughes, Community Liaison and Micki Siegel de
Hernandez, Alternate Community Liaison to the EPA WTC Expert
Technical Review Panel
-------
Table of Contents
Section I: Executive Summary
Section II: Recommendations, Comments, and Questions
Objectives of the Sampling Plan Proposed List of Contaminants to be Sampled
Proposed Geographic Scope of Sampling Proposed "Trigger Factors" to Determine
the Need for Cleanup Proposed Method to Determine Cleanup Responsibility (The
"WTC Signature" Issue) Proposed "Trigger Factors" for Expanded Testing Proposed
Sampling Design Sample Collection and Analysis Excessive Reliance on Proposed
Signature Study Need for Quality Assurance and Quality Control and Disclosure of
Test Results Conclusion Appendix I: Community Requests of EPA that Remain
Unmet Appendix II: 7 Principles Letter and Supporting Groups
Section III: Synthesis Report of the Expert Advisory Committee
Executive Summary
We are community, environmental, labor, tenant, religious, disaster recovery, small
business, and social service organizations, residents, school parents, workers, property
and small business owners in areas affected by World Trade Center pollution, who have
been concerned about unaddressed environmental and public health issues since
September 11, 2001. We have diligently participated in the World Trade Center Expert
Technical Review Panel process since it began in March 2004. Our work as community
and labor representatives in that process has been acknowledged on a formal basis by the
Environmental Protection Agency (EPA) through the Community Based Participatory
Research (CBPR) process.
On October 21, 2004, EPA published a revised version of the agency's, "Draft Proposed
Sampling Program to Determine Extent of World Trade Center Impacts to the Indoor
Environment" (hereafter, "Draft Proposed Sampling Program") in the Federal Register.
In response to community and panel member concerns, the Draft Proposed Sampling
Program extends the geographic boundaries for sampling to Houston Street, includes
workplaces and public buildings, includes an expanded list of contaminants to be
sampled, and eliminates air sampling (in favor of dust sampling). The improvements are
an important step towards addressing the as yet unresolved problem of indoor toxic
contamination caused by the September 11th attack on the World Trade Center and the
aftermath of that attack. The WTC Community-Labor Coalition is also appreciative of
the efforts of the EPA World Trade Center Expert Technical Review Panel for its role in
helping to spur this progress. However, the Draft Proposed Sampling Program in its
current form contains serious scientific and programmatic flaws which must be rectified
in order for the plan to be scientifically valid, effective and credible.
-------
Funded through EPA's CBPR process, we assembled a committee of highly qualified
scientists and practitioners with expertise in environmental health, epidemiology,
toxicology, industrial hygiene, statistical analysis, chemistry, and atmospheric transport
and modeling. The WTC Community-Labor Coalition's CBPR Expert Advisory
Committee (hereafter, "Expert Advisory Committee") has provided an independent
assessment of EPA's Draft Proposed Sampling Program. The WTC Community-Labor
Coalition supports the findings of the Expert Advisory Committee. The findings of the
Expert Advisory Committee are included in their entirety in Section 3 of this document,
are referenced throughout this document, and additionally, have been submitted to EPA
under separate cover.
We urge all members of the EPA Expert Technical Review Panel to give serious
consideration to the findings of the Expert Advisory Committee.
We call upon the EPA to reformulate the Draft Proposed Sampling Program based
upon the findings of the Expert Advisory Committee and the WTC Community-
Labor Coalition's comments.
To this end, we submit the following comments on EPA's Draft Proposed Sampling
Program in our own names, and in the names of the thousands of workers and residents
whom we represent.
The summary of the WTC Community-Labor Coalition's recommendations is as follows:
• The EPA must sample for and clean up toxic indoor WTC contamination as
quickly as practicable. The Draft Proposed Sampling Program does not contain a clear
commitment to clean up contaminants when found and the decision-making criteria for
cleanup are vague. The Expert Advisory Committee states that the primary objective of
the Draft Proposed Sampling Program "must be to identify habitable spaces with ongoing
World Trade Center (WTC) contamination and provide cleanup where warranted." The
Expert Advisory Committee further recommends, "There is an urgent need to quickly
identify indoor spaces where WTC toxics pose a threat to human health and to clean
these spaces immediately.";
• The EPA must not wait for discovery and validation of a "WTC chemical
signature" (which may never be identified) before beginning a sampling and cleanup
program. The sampling and cleanup program must proceed independently of the
signature research, which will likely take years to complete and ultimately, may not be
successful. In contrast, the Draft Proposed Sampling Program is contingent upon the
discovery of a WTC signature. The Expert Advisory Committee states, "Whether or not a
signature is found, it is essential to clean up any contamination resulting from the WTC
event.";
• Affected neighborhoods in Brooklyn and other areas likely to have been
impacted by the disaster must be included in the initial sampling and cleanup
program. The EPA's proposed geographic zone for initial sampling fails to include
these areas, relegating them to a "Phase II" which may never materialize. The Expert
-------
Advisory Committee states, "The proposal does not describe the rationale for excluding
Brooklyn or limiting the study area to lower Manhattan. Sampling of buildings should be
much broader than planned and should be based on the extent of the plume as determined
either by the NASA photos or other appropriate methods." The Expert Advisory
Committee further recommends, "Consideration must also be given to areas from which
residents have been reported to exhibit adverse respiratory health effects...";
• The results of all samples taken of toxic substances must be included in the
cleanup decision-making criteria. The Draft Proposed Sampling Program wrongly
discounts contamination of infrequently cleaned or low contact and "inaccessible" indoor
areas. The Expert Advisory Committee states, "The dust samples present in inaccessible
locations, like those found in HVAC ducts or ceiling plenums, represent the most
significant reservoirs of contaminated dusts available for introduction into residential
living space and work space alike. These reservoirs must receive the highest priority in
the sampling program design.";
• Small asbestos fibers (i.e., less than 5 microns in length) must be included in the
sampling results and considered in assessments as to whether or not cleaning is
warranted. The proposed sampling methods are inadequate for identifying very small
fibers and particles that may pose significant health threats. The Expert Advisory
Committee states, "Short fibers should be sampled and reported. Any assumption that
short fibers, less than 5 microns in length, are not hazardous cannot be justified based on
the available science.";
• The list of proposed contaminants for sampling (asbestos, man-made
vitreous fibers, crystalline silica, polycyclic aromatic hydrocarbons, and lead) must
be expanded to include particulate mercury and dioxin. The Expert Advisory
Committee noted that many of the sampling and analytical methods used by the EPA to
restrict the contaminants of potential concern (COPCs) to be tested were "inadequate"
and resulted "in the elimination of many substances found frequently at hazardous levels
in many buildings";
• The process for selection of buildings to be tested must be elucidated and
should incorporate additional factors that can, according to the Expert Advisory
Committee, "affect a building unit's accumulation and retention of WTC toxics."
Information must also be collected for buildings that are not included in the sampling
program to assess whether or not the buildings sampled are truly representative, and if
not, to determine how the results may be biased. The Expert Advisory Committee states,
"The proposal for selection of buildings is vague and appears to be flawed.";
A detailed rationale must be provided for the "triggers" for cleanup, (i.e.,
benchmarks used to determine whether cleanup is conducted). The EPA plan
proposes an arbitrary "trigger" of "3X background" for certain contaminants without
describing how the background levels will be determined and without providing a
rationale as to why the "3X background level" was chosen for each of the contaminants
tested. The EPA plan fails to consider the potential health consequences of chemical
-------
mixtures. The Expert Advisory Committee states, "As a result of the failure to consider
mixtures, specific health effects are likely to be underestimated by the benchmark of one
set of contaminants.";
The Draft Proposed Sampling Plan must contain a quality assurance/quality
control (QA/QC) protocol to make sure that testing and analysis will be carried out
properly. The Draft Proposed Sampling Program barely addresses the issue of quality
assurance/quality control. The Expert Advisory Committee states, "The proposed
sampling program must represent a state-of-the-art sampling and analysis effort with
adequate QA/QC employed such that the data are fully defensible."
The ultimate success of this endeavor will depend upon public confidence that the
sampling program is designed to find - rather than avoid finding - any remaining toxic
indoor hazards from 9/11. A necessary component of any large-scale sampling
program, and one which is omitted from EPA's proposal, is an aggressive, well-
developed and well-funded outreach program, designed with close involvement of the
affected communities and incorporating appropriate incentives to encourage
participation. This can best be achieved by a clear and unequivocal public commitment
from the federal government to conduct cleanup of hazardous WTC contaminants when
they are discovered.
SECTION II: RECOMMENDATIONS, COMMENTS, QUESTIONS
Objectives of the Sampling Plan
It is our position that the goal of this sampling program should be to find whatever
World Trade Center pollution is out there in the home or workplace environment that
may still present a risk to human health, and to remove it.
The Expert Advisory Committee has come to a similar conclusion, stating, "It is
imperative that indoor spaces be cleaned of WTC toxics whether or not a signature is
found. Since the studies needed to determine whether or not a signature exists are likely
to take years, the cleanup process must not be held hostage to the development of a
signature. The stated objectives need to be restated with this recognized. In addition, to
assure participation, cleanup and health liabilities need to be addressed."
Proposed List of Contaminants to Be Sampled
The sampling proposal identifies five substances to be sampled - asbestos, man-made
vitreous fibers (MMVF), crystalline silica, polycyclic aromatic hydrocarbons (PAHs),
and lead. The EPA proposal does not call for testing of short, very thin chrysotile
asbestos fibers, or for the testing of mercury, nor does it address the potential presence of
toxic halogenated organic chemicals, such as dioxins.
The WTC Community calls on EPA to expand its list to include, at a minimum,
particulate mercury and dioxin, and also to count and report short asbestos fibers.
-------
The Expert Advisory Committee noted that many of the sampling and analytical methods
used by the EPA to restrict the COPCs to be tested were "inadequate," and resulted "in
the elimination of many substances found frequently at hazardous levels in many
buildings"
Non-governmental testing found mercury at levels of concern at 90 Church Street, 130
Liberty Street, 4 Albany Street, and 30 West Broadway. The Expert Advisory
Committee observes that EPA may have missed the presence of mercury in its testing
because it tested for mercury in vapor form, not particulate form, which is the more
likely form in which mercury would persist in the indoor environment.
The Expert Advisory Committee notes that dioxin has been found at levels above
health-based benchmarks in some commercial and government buildings. In fact, with
regard to dioxin, EPA itself has stated "It would be reasonable to conclude that the
concentrations to which individuals could be potentially exposed ... within and near the
WTC site through the latter part of November are likely the highest ambient
concentrations that have ever been reported." (Exposure And Human Health Evaluation
Of Airborne Pollution From The World Trade Center Disaster, Oct. 2002) While
acknowledging the cost of dioxin testing, the Expert Advisory Committee nevertheless
urges that sampling be conducted for dioxins, and suggests that EPA consider using the
CALUX assay for dioxin-like activity.
The Expert Advisory Committee states that "any assumption that short (asbestos) fibers,
less than 5 a [microns] in length, are not hazardous cannot be justified based on the
available science." The WTC Community agrees with the Expert Advisory Committee's
recommendation that any sampling plan report and take into account sampling results for
short asbestos fibers. The Expert Advisory Committee provides numerous citations in the
scientific literature to support its concern about the toxicity of such short fibers. It further
observes that fiber alteration may be a variable in ability to cause disease, and that fibers
may have been altered as a result of the WTC event.
Proposed Geographic Scope of Sampling
Affected neighborhoods in Brooklyn and other areas likely to have been impacted by the
disaster should be included in the initial sampling and cleanup program, not relegated to a
"Phase II" which may never materialize.
The Expert Advisory Committee states, "The proposal does not describe the rationale for
excluding Brooklyn or limiting the study area to lower Manhattan. Sampling of buildings
should be much broader than planned and should be based on the extent of the plume as
determined either by the NASA photos or other appropriate methods." Their report notes
that the contaminants that traveled across the river are likely to include more of the
smaller, combustion-related particles and observes that, "there may have been less of the
larger, heavier particles depositing in Brooklyn than Manhattan, but the smaller fibers are
readily suspended and can travel to Brooklyn. Given the frequent wind direction to the
east and southeast, Brooklyn should be tested in Phase 1."
-------
The Expert Advisory Committee is in agreement with the Community's position as stated
in the "Seven Principles" adopted by more than 50 community and labor organizations
and all 3 Lower Manhattan community boards (included as Appendix II of this report).
The Expert Advisory Committee recommends further that the sampling plan consider
"areas from which residents have been reported to exhibit adverse respiratory health
effects." EPA should, in particular, consider the medical evidence provided by the
Stonybrook University study of asthma impacts from WTC pollution in determining the
geographic scope of sampling. The scientists who conducted that study determined that
asthmatic children who lived within 5 miles of Ground Zero suffered a deterioration in
their health and had to visit their doctors more often for treatment and take more
medicine for asthma during the year after the 9/11 attack than the children at a
comparison clinic in Queens. The Expert Advisory Committee's recommendation against
delay in testing, that no area at risk should be relegated to a so-called "Phase II" for initial
testing, applies to such neighborhoods as well.
-------
Anthony Szema, et a/., "Clinical Deterioration in Pediatric Asthmatic Patients After
September 11, 2001," J. Allergy CM. Immunol. 113(3):420-426 (2004).
Proposed "Trigger Factors" to Determine the Need for Cleanup
EPA's proposed decision-making criteria to determine whether or not cleanup is needed
fails to protect public health because it: (1) inappropriately discounts the presence and
availability of contamination in low contact and/or less frequently cleaned and/or
inaccessible areas; (2) proposes vague, arbitrary, and unsubstantiated guidelines for the
contamination level(s) that would trigger a cleanup; and (3) relies on inappropriate
averaging of test results that would significantly underestimate levels of contamination.
Less frequently cleaned and inaccessible areas: We oppose EPA's plan to disregard test
results from "inaccessible" areas in the decision-making for cleanup. The two examples
that EPA cites as "inaccessible" areas are "behind or on top of cabinets." In reality, such
locations are not inaccessible, but rather low contact or less frequently cleaned locations.
Contamination that may accumulate in these areas can be disturbed by activities such as
cleaning, moving furniture, removing items from tops of cabinets, and/or
remodeling/renovations.
The Expert Advisory Committee further warns that exposures from inaccessible areas -
such as HVAC systems and ceiling plenums - can occur without direct human contact
with the area itself. The Expert Advisory Committee states "it is not appropriate to fail to
utilize contaminant levels from inaccessible areas as a consideration for cleanup," such as
HVAC systems or ceiling plenums, because they represent "reservoirs" that could cause
the release of pollutants to recontaminate an indoor area.
The Executive Summary of the Expert Advisory Committee report states, "Inaccessible
areas are restrictive to human activity, not to airborne toxics that circulate in all air spaces
to some degree. These represent reservoirs of contamination or "hot spots" and should be
prioritized for examination and cleanup if warranted. There is an urgent need to quickly
identify indoor spaces where WTC toxics pose a threat to human health and to clean
these spaces immediately."
We concur with the Expert Advisory Committee that "none of the sampling should be
excluded from cleanup decision-making criteria" and that "cleaning should target the
reservoirs of the toxics."
Arbitrary guideline for clean-up: In the absence of available health-based benchmarks
for dust samples of asbestos, MMVF, and crystalline silica, the current sampling plan
proposes a 3X background level as the trigger for cleanup. However, the EPA does not
define background or how background will be determined, nor is a plausible rationale
provided for using the 3X background benchmark. The Expert Advisory Committee
states that "further justification of the 3X background criteria is needed" We agree with
the Expert Advisory Committee that there must be "reasonable assurances that 3X
background is safe." It is also inappropriate to use a 'one size fits all' approach for all of
the contaminants. The Expert Advisory Committee notes, for example, that asbestos
-------
would be of much greater concern at three times background than man-made vitreous
fibers. Further, valid comparison of background and sampled levels for any given
substance requires that collection and analysis methods be identical. The sampling
proposal gives no consideration to this basic concept.
The 3X background trigger also fails to take into consideration the potential health
effects of contaminant mixtures. The Expert Advisory Committee notes that, "No
consideration has been given in the sampling plan to the cumulative effects of the
COPCs or signature compounds when individual contaminants are found below
published health effects thresholds. More importantly, no consideration has been given
to the complex chemical universe present on these dusts. Many of these compounds are
not regulated in any fashion and the vast majority has not been adequately assessed for
health effects."
Inappropriate averaging of test results: EPA proposes using a 95% upper confidence
limit standard on the mean contaminant level in a building to determine whether the
building should be cleaned. The Expert Advisory Committee warns that the 95% upper
confidence limit is "not a defensible criterion" in this context due to variability in
contamination as well as to building characteristics such as size, etc. Under the EPA
proposal, the potential impact of any reservoirs of contamination will be misleadingly
underestimated by averaging test results of these areas with results from frequently
cleaned areas. The Expert Advisory Committee states, "Hard accessible surfaces should
not be included in the overall mean."
We agree with the Expert Advisory Committee statement that, "As a general principle, if
dusts collected in an individual residence/apartment or workplace are found to contain
COPC/target parameters above threshold levels, that residence or workplace should be
cleaned. This should be the practice despite the outcome of statistical analyses done on
all samples collected in that building. These locations should be considered 'hot spots'.
All hot spots should be remediated."
Proposed Method to Determine Cleanup Responsibility (The "WTC Signature"
Issue)
The proposed sampling plan states that "the existence of a reliable signature for WTC
dust and/or combustion products" is a "cornerstone of this proposal" (Sampling Proposal,
p. 2). However, the proposal's virtually exclusive reliance on possible identification of
one or more signatures raises 2 types of issues - scientific validity and ramifications for
the sampling process.
As a scientific matter, discovering and validating an actual WTC signature is difficult, or
even improbable, for a number of reasons. The current proposal states that the signature
study must be "fully successful in identifying a signature in indoor dust that can be
reliably tied to the building collapse," yet states no objective criteria by which to assess
the validity of a signature.
-------
In the search for a signature, the proposal assumes that dispersion of contamination from
the events of 9/11 and thereafter derived solely from 2 events - the collapse of the towers
and 4 months of ensuing combustion. However, primary dust contamination resulted
from the collapse and secondary contamination from the transport of debris by truck
through the streets to the waste transfer stations. Further, dust originating in all collapsed
buildings should not be presumed to be identical, due to differing structural components,
furnishings, and materials and substances present (e.g., freon, diesel fuel, electrical
transformers, jet fuel, battery acid, etc.). The Expert Advisory Committee points out that
there were three types of fires - the fire before the collapse, high- temperature fires, and
low- temperature smoldering fires. The products of these fires would be different as well.
The amorphous, heterogeneous nature of the WTC dust, both in the original polluting
event and through the course of possible re-suspension over time increases the difficulty
of signature validation. The current sampling plan does not consider that the content of
WTC dust varied based on deposition distance, since different substances and
differently sized particles have varying abilities to be transported over distance. For
instance, a signature derived solely from samples taken close to the WTC site is likely to
be inappropriate for assessing Brooklyn sites. The Expert Advisory Committee
emphasizes that the deposition patterns of the different materials varied. It states,
"WTC dust and combustion products will become fractionalized by distance,
orientation, elevation, building characteristics, interior surfaces and cleaning history."
The current proposal does not adequately acknowledge that the sampling program
occurs more than three years after the event. (This delay is not the fault of the
community. We have been requesting proper testing throughout this entire period.) The
Expert Advisory Committee points out that the passage of time "means that there will be
an overlay of new substances that can obscure a WTC pattern." We must assume that
some mixing of dust will occur in some, if not many, instances. If developed at all, any
signature must be defined with ample flexibility to consider the likely mixing of
ordinary dust with WTC dust. Indeed, the Expert Advisory Committee, taking account
of all these complexities, observed that "multiple signatures may need to be developed
to account for location-specific factors."
Whether or not one or more signatures does in fact exist and will be validated is
unknown at this time. The length of time that may be necessary to identify and validate
a signature, or to fail to do so, is likewise unknown. In addition, signature validation
should be subject to independent peer review, adding further delay to the process of
sampling and possible cleanup.
The Expert Advisory Committee states, "Awaiting availability of valid WTC chemical
signatures, applicable to building collapse and emissions form the ensuing fires,
respectively, will likely delay cleanup, if cleanup takes place at all. It is essential that
cleanup of WTC toxics occurs as soon as practical even in the absence of a valid
signature."
The current sampling proposal is almost entirely dependent upon one or more
signatures, yet to be confirmed. There is no "Plan B," that is, the proposal does not posit
-------
a way to move forward with sampling and cleanup if a signature cannot be validated and
does not take into account that delays in signature validation effectively will result in
delays in sampling. A consequence of reliance on yet to be determined signatures for
cleanup decisions is that commitment to cleanup cannot be made, with probable
negative ramifications for voluntary enrollment in the sampling program.
It continues to be the WTC Community's position that the EPA must not wait for
discovery and validation of a "WTC chemical signature" (which may never be
identified) before beginning a sampling and cleanup program. That program must
proceed independently of the signature research, which, in the opinion of the Expert
Advisory Committee, may take a lengthy period of time to complete and ultimately may
not be successful.
It may be more realistic to expect that WTC dust can only be identified on a "more likely
than not" basis, as indicated by the presence of one or more substances from a list of
likely contaminants. Decisions about further testing in a neighborhood or beyond the
Phase I test zone could be based reasonably on evidence that falls short of a definitive
"WTC chemical signature." The Expert Advisory Panel affirms that cleanup is the
priority, not the "signature": "Whether or not a signature is found, it is essential to clean
up any contamination resulting from the WTC event."
While the discussion of the so-called "signature" for identifying WTC dust has been
posed as a scientific issue, EPA has directly linked it to the policy issue of who cleans
up contamination when it is found. Under the current proposal, if testing a space reveals
exceedences of contaminants known to be components of WTC dust-even those
identified as COPCs in EPA's own 2003 guidance document, "World Trade Center
Indoor Air Assessment—but the precise WTC signature is not found, EPA refuses to
provide a cleanup. As a pragmatic matter, EPA must not be allowed to overly limit any
such signature. Any signature, if used at all, should not be so rigidly defined as to
exclude genuine cases of WTC contamination and unfairly burden individual owners
with a cleanup problem that rightly should be remedied by the federal government.
Too, inappropriately limiting EPA cleanup would tend to discourage participation in the
program by people who, in the event that the so-called WTC signature is not found, could
not afford to clean up the dust in a coop apartment or home that they own, or in their
small business. This policy would also present a problem for tenants who cannot afford to
battle with their landlords over cleanup and would have little or no practical remedy if
contaminants are discovered The Expert Advisory Committee argues that a viable plan
can be developed in the absence of a validated signature, noting that many buildings in
the WTC vicinity have already been sampled, cleaned and remediated without it. Our
Advisory Committee proposes that if the contaminants of potential concern exceed health
criteria or another acceptable benchmark, "then that space should be identified for
cleanup." It argues that if a chemical such as lead can be demonstrated to have come
from a non-WTC source, then government agencies should cooperate to identify the
responsible party.
-------
The standard for cleanup of contaminated dust should not - and cannot - be absolute
"certainty" of WTC origin. It would be more appropriate to structure the program so that
where exceedences of the tested Contaminants of Potential Concern occur, EPA conducts
cleanup unless EPA demonstrates clearly that another source resulted in the
contamination and another identifiable party is responsible for cleaning it up.
The program will need to be designed to attract participation by residential and
commercial tenants, owner-occupied residences and large building owners. Getting a
home or building tested involves inconvenience and sometimes disturbance of personal
belongings. For owner-occupied residences where the owner has volunteered to
participate, special care must be taken to ensure that there is no disincentive to
participate. The Expert Advisory Committee observes, "If unit cleanup costs are
relatively inexpensive compared to the costs of conducting the research, some researchers
consider it an ethical obligation to pay for the cleanup of contaminated units at the end of
the study," which, the Expert Advisory Committee notes, provides an incentive for
participation and thus improves the overall study design.
As noted above, any "signature," if developed at all, must be defined with ample
flexibility to consider the likely mixing of ordinary dust with WTC dust, the
heterogeneous nature of the dust itself, and the uneven deposition of the dust in the
environment. The standard should not - and probably cannot - be absolute "certainty."
The cleanup trigger must be designed to protect the public from further exposure to
WTC dust. That must be the primary goal - not absolute certainty of source more than
three years after an event.
Proposed "Trigger Factors" For Expanded Testing
If the presence of contaminants has been detected in samples taken from a given
building under this program, three important decisions must be made:
whether or not further testing should be done of that building;
whether or not further testing should be done in buildings in the surrounding
area and
if the site is located near the border of the Phase I testing zone, whether
or not testing should be expanded beyond that border.
The EPA proposal does not provide satisfactory answers to these questions.
The answer is relatively straightforward with regard to individual buildings. The
Expert Advisory Committee recommends that if units within a building tend to have
similar levels of contaminants, then the entire building should be cleaned. It notes, in
addition, that in some instances it may be more practical to clean an entire building
ventilation system regardless of variation in contamination of units in the building.
The questions of expanded neighborhood testing and expansion of testing zones are more
challenging — and yet critical to answer. The goal of this project, after all, should be to
identify and clean up all contaminated indoor spaces that threaten human health. Because
-------
the sampling plan is not designed to promote collection of samples from multiple
buildings in the same neighborhood, the Expert Advisory Committee suggests conducting
multiple building sampling in some neighborhoods and plume corridors as a means to
assess whether data from one building predict those in neighboring buildings, and as a
step toward evaluating what factors predict area-wide contamination. It notes that EPA
probably will need to evaluate a variety of geographic and non-geographic factors to
determine what best predicts contamination of untested buildings.
We strongly urge that EPA provide a clear plan for identifying the "next step" expanded
cleaning needs.
Proposed Sampling Design
Sampling plan design as currently proposed is seriously flawed in several regards. It is
almost entirely reliant on self-enrollment of participants. As a result, due to potential
liability concerns, the proposed sampling design creates a disincentive to enrollment by
landlords or employers who did not engage in post 9/11 cleanup. Consequently, indoor
spaces that have been remediated are likely to be over-represented in sampling results,
while spaces that have not been cleaned up are likely to be under-represented. The Expert
Advisory Committee notes, "Self-selection prior to sampling will bias the results
significantly by underestimating the degree of contamination. Volunteers are more likely
to be aware of the WTC risks, and are more likely to have taken preventative and
remedial action (e.g., professional cleaning and remediation). Building owners and
employers who fear health and cleanup liability are less likely to volunteer."
Perhaps the most serious disincentive to enrollment in the sampling plan as currently
proposed is the omission of a clear and unambiguous government commitment to cleanup
where warranted. Given the history of the past 3 years and the levels of mistrust that have
ensued, it is likely that many downtown property owners, employers, and residents will
be reluctant to participate in a government program that contains no clear promise to
address contaminants, if found. The Expert Advisory Committee notes that,
"Government-funded cleanup of toxics is essential for gaining participation", and
recommends making a structured random sample of buildings, and then approaching the
selecting building owners and tenants. The Committee further suggests that building
variables (such as building type, location, type of ventilation system, and cleaning
history) be compared for volunteered and non-volunteered buildings in order to
determine whether or not volunteered buildings are likely to be representative of all
eligible buildings .
Since the proposed sampling plan contains no provision for assessing prior sampling
results or cleanup histories of buildings that do not self-enroll, the actual extent of
sampling bias will not be known. In addition, although the plan allows for individual
residential tenants to self-enroll, access to common spaces and to mechanical
ventilation systems is to be controlled by the landlord, and access to workplaces is to
be controlled by the employer. This arrangement effectively disenfranchises the
overwhelming majority of downtown residents and workers by preventing them from
requesting sampling, and possibly cleanup, of their indoor spaces.
-------
The sampling proposal relies on a two-dimensional grid process to ensure that selected
buildings are geographically representative. The current proposal, however, is not
designed to factor in non-spatial criteria which may ultimately influence the how
representative the sample will be. According to the Expert Advisory Committee,
"buildings, and units within buildings, are expected to vary in their accumulation and
retention of WTC toxics depending upon many factors: distance, altitude, cardinal
orientation, penetration rates (i.e., how easy it is for outside toxics to penetrate a building
through closed windows, ventilation intakes, tracking in, etc.), window usage, type of
ventilation system, and cleaning history." 3,3 Three-dimensional factors such as height
above the ground of ventilation intakes cannot be factored in to the proposal's 2-
dimensional model.
Building selection must also take into account the dispersion and deposition processes
of the various types, and sources, of contaminants. But according to the Expert
Advisory Committee, "A sufficient sample size will be necessary to be able to
characterize the range of penetration possibilities within and between buildings. The
proposal is not sufficiently detailed to demonstrate that this will be accomplished."
Sample Collection and Analysis
Given the high proportion of very small particles and fibers deposited inside homes and
work spaces by the World Trade Center Collapse and subsequent fires, we believe it is
imperative that sampling and analysis methods utilized be capable of capture and
detection of very small particles. The HEPA vacuum method proposed in the sampling
plan will not capture these very small asbestos and man made vitreous fibers.
Consequently, the Expert Advisory Committee recommends that wipe sampling, as well
as HEPA sampling, be used to test hard surfaces for those analytes: "Smoke residues and,
in particular, contaminants associated with smaller yet inhalable particles, will not be
sufficiently removed by the proposed HEPA collection technique. The wipe sample will
ensure these contaminants are included in the total concentrations of target
compounds/COPC present on the sampling surface."
We are very concerned that sampling proposal is seriously deficient in its failure to
address sampling for lead and PAHs on soft surfaces such as carpets and upholstered
furniture. According to the Expert Advisory Committee, "Soft surfaces such as fabrics
are ideal deposition surfaces for particulates and serve as reservoirs for the
contaminants", and recommends: "Particle associated lead and PAHs present in soft
surfaces should then be sampled in an identical fashion to what is proposed for asbestos,
silica, and MMVF."
Within HVAC systems, particulate deposition is most likely to occur in low velocity
areas in duct work and at bends in high velocity areas. Although the proposed sampling
plan calls for monitoring at various locations within HVAC systems, it does not address
"dead spots" which are the areas most likely to have become reservoirs for WTC
contaminants and potential sources for their resuspension and entrainment. The Expert
Advisory Committee recommends that the sampling plan be revised to more specifically
designate what parts of the HVAC system should be sampled and the minimum number
-------
of samples per square foot to be collected
Analytical methods for asbestos, MMVF, and silica are not provided in EPA's draft
sampling proposal. The proposal is not complete without this information and
therefore cannot be adequately evaluated in this regard.
Need for Quality Assurance and Quality Control and Disclosure of Test Results
EPA's current Draft Proposed Sampling Plan addresses the need for quality assurance
and quality control (QA/QC) only in passing. The Expert Advisory Panel has stated that
the proposal should contain a fully developed quality assurance/quality control (QA/QC)
protocol to ensure that testing and analysis is carried out properly.
First, such a protocol must include a system for independent monitoring (including actual
"spot checking") of sampling and analysis. EPA and the WTC Expert Technical Review
Panel have heard extensive testimony on numerous testing and cleanup protocol
violations and other improper work practices that occurred during the 2002 EPA
Residential Cleanup Program, supervised - or perhaps unsupervised - by EPA. For
example, the testers did not run a fan during air testing, and the cleaners did not cover the
intake/discharge registers with plastic. Cleanup workers were also observed working
without personal protective equipment. Additional flaws and failures in EPA's 2002
indoor cleanup program are documented in the Sierra Club's 2004 report, Pollution and
Deception at Ground Zero.
Second, disclosure of sampling results and monitoring data in a timely fashion to building
owners and to all building occupants - including workers - is critical. In testimony at
Panel meetings about this issue, affected tenants or building occupants have voiced
concern because of the difficulties in obtaining the results of EPA sampling in a timely
way.
Finally, quality assurance and quality control will be key to building public
confidence in a future WTC contamination cleanup project.
The Expert Advisory Committee states, "The proposed sampling program must represent
a state-of-the-art sampling and analysis effort with adequate QA/QC employed such that
the data are fully defensible."
Conclusion
The Expert Advisory Committee rightly observes that this "proposed program is
not simply a sampling program," and that the title should "include a statement concerning
adequacy of the cleanup for the safety of the occupants." This, ultimately, is the goal of
our endeavors. We urge EPA to adopt the recommendations set out in these comments
and in the Expert Advisory Committee's review of the sampling program, and we urge
the WTC Expert Technical Review Panel to join in this effort to obtain a truly effective,
scientifically valid, credible program for sampling and cleanup of WTC pollution.
-------
Appendix I: Community Requests of EPA that Remain Unmet
We urge EPA to respond to these five unmet requests regarding
information and procedure to our CBPR process:
1. Declare a solid commitment to conduct prompt cleanup. Members of the
community have emphasized that the sampling plan will fail - because people will not
volunteer their homes or offices for testing - unless EPA makes a strong commitment to
conduct cleanup of any hazards likely to have come from WTC pollution.
2. Assert regulatory authority over the imminent or ongoing demolition of the
buildings highly contaminated by WTC dust. The highly contaminated high-rise
buildings such as the 40-story former Deutsche Bank Building (130 Liberty Street) and
Fiterman Hall (30 West Broadway) are slated for demolition, while 4 Albany Street
currently is already underway. Other contaminated buildings such as 130 Cedar Street
may be demolished or undergo substantial renovation soon as well. We have urged that
EPA supervise safety during the demolition of such buildings.
3. Provide a legal memorandum describing the powers of various agencies to
gain access to buildings for environmental testing and how those powers might be
combined to help effectuate this sampling and cleanup project. While a resident can
request testing of his or her own apartment, one cannot have testing of a building's
common areas or ventilation system where the owner opposes it without power of access.
Similarly, employees cannot obtain testing of their workplace, where an employer and
building owner oppose it, unless an agency can override such opposition. Although this
issue has been repeatedly requested at Panel meetings, to date no such legal
memorandum has surfaced.
4 Establish a systematic effort to collect and to establish a central database of
all 9/11-related indoor environmental sampling data results from public and private
sources. This should include government agencies, academic centers and independent
sources. Although there have been repeated requests, there appears to have been no
movement on this issue.
5. Produce verbatim transcripts of EPA Panel meetings. This is important both
as a matter of public record and to ensure the integrity of the panel process. Community
members have pointed out anomalies and errors in EPA summaries of the meetings in the
past.
-------
Appendix II: 7 Principles Letter
October 26, 2004 (signatories updated 12/17/04)
Michael O. Leavitt Administrator
U.S. Environmental Protection Agency Ariel Rios Bldg. 1200 Pennsylvania Ave NW
Washington, DC 20460
Dear Administrator Leavitt:
We are community, tenant, religious, disaster recovery, social service, environmental
and labor organizations, and residents, workers, and small business owners in the
affected areas, who have been concerned by unaddressed environmental and public
health issues since Sept. 11, 2001. Many of us have diligently participated in the EPA
World Trade Center Expert Technical Review Panel process that began in March 2004,
and our work as community representatives in that process was recently placed on a
formal basis by EPA.
In our own names, and in the names of the thousands of workers and residents
whom we represent, we make the following statement and request:
The lower Manhattan and Brooklyn communities, both residents and workers, have, for
three years, called on EPA to clean up the contaminants left behind by the terrorist
attacks of September 11, 2001. For three years, EPA has been unresponsive to the
appeals of our communities, our elected representatives, and EPA's own Inspector
General. For the last eight months, lower Manhattan and Brooklyn residents and workers
have worked, in good faith, as closely with the EPA WTC Technical Expert Review
Panel as we have been permitted to do. We appreciate the efforts of panel members and
we hope to be able to continue working with the panel.
Nevertheless, eight months after this panel began its work, no additional environmental
testing or clean-up has been conducted. Our children, our neighbors, our co-workers, and
our firefighters continue to live with the uncertainty of possible exposure and
unnecessary risk. After three years of delay by EPA and eight months of work by this
panel, EPA has yet to make a public commitment to testing and decontamination.
We therefore call upon EPA, by the end of October 2004, to publicly commit itself in a
written statement released at a press conference presided over by an official EPA
spokesperson to the following seven principles:
1. EPA will conduct, with appropriate input from the community, comprehensive
indoor environmental testing for multiple contaminants. The testing will occur as
promptly as possible.
-------
2. EPA will expand the geographic range of the testing from its original boundaries
to include, at a minimum, additional southern Manhattan communities, including all of
Chinatown, the Lower East Side, and also the neighborhoods in Brooklyn affected by
World Trade Center dust.
3. EPA will test both residences and workplaces. Landlords, residents, employers,
and employees will all be given the option of volunteering to have their respective
buildings, residences, and workplaces tested.
4. EPA testing will include mechanical ventilation systems.
5. Where test results warrant, EPA will decontaminate not only the tested buildings
but the neighborhoods affected by 9/11 contaminants. The clean-up clearance criterion
for each identified contaminant will be based upon consideration of health-based
benchmarks and background levels, utilizing the criterion that is more protective.
6. EPA will, with appropriate community input, take the lead role in supervising the
environmental safety of all 9/11-related clean-up, demolition, and reconstruction
activities.
7. As EPA evaluates unmet health needs resulting from the attacks, it will support all
necessary national and local efforts to ensure public health education, outreach, and long-
term medical follow-up for affected communities and to ensure medical care for affected
individuals.
This statement of principles is endorsed by the following community, residential, tenant,
religious, disaster recovery, social service, environmental, small business and labor
organizations and businesses:
Manhattan Community Board No. 1 (by resolution)
Manhattan Community Board No. 2 (by resolution)
Manhattan Community Board No. 3 (by resolution)
9/11 Environmental Action (residents and school parents organization)
Asian American Legal Defense and Education Fund (AALDEF)
Association of Legal Aid Attorneys, UAW 2325, AFL-CIO
Battery Park City United
Candy World (small business)
Chinese Progressive Association
Citizens Environmental Coalition (CEC)
Civil Service Employees Association (CSEA)
Communications Workers of America (CWA), District 1
Communications Workers of America (CWA), Local 1180
District Council 37, AFSCME
Duane Street Block Association
Essex World Cafe (small business)
Family Association of Tribeca East (FATE)
-------
Fiscal Policy Institute
Good Jobs New York
Good Old Lower East Side (GOLES)
Greater NY Labor and Religion Coalition
Independence Plaza North Tenants Association (IPNTA)
Investor Data Services (small business)
Little Italy Neighbors Association (LINA)
Manhattan Trustee Rudy Sanfilippo, Uniformed Firefighters Association
Met Council on Housing
National Postal Mail Handlers Union, Local 300
National Treasury Employees Union, Chapter 293
New Jersey Work Environment Council
New York City Coalition to End Lead Poisoning (NYCCELP)
New York Committee for Occupational Safety and Health (NYCOSH)
New York Disaster Interfaith Services (NYDIS)
New York Environmental Law & Justice Project (NYELJP)
New York From the Ground Up (represents 600 small businesses in the WTC area)
New York State Public Employees Federation (PEF)
Organization of Staff Analysts (OSA)
Parents Association of Stuyvesant High School
Physicians for Social Responsibility - New York City
Pop Filter Music (small business)
Professional Staff Congress (PSC)
Puerto Rican Legal Defense and Education Fund (PRLDEF)
Rebuild with a Spotlight on the Poor Coalition (represents 20 community-based
organizations)
Residents of 125 Cedar Street
Sierra Club
Sierra Club - Fairfield County Group (Connecticut)
Tenants and Neighbors
The 2M Corporation (small business)
Transport Workers Union (TWU), Local 100
Uniformed EMTs & Paramedics - FDNY (EMTs & paramedics from Fire Dept)
Uniformed Fire Officers Association
United Federation of Teachers
University Settlement
Worthy Eyes (small business)
WTC Residents Coalition (represents 30,000 Battery Park City residents)
-------
11. Community-Based Participatory Research (CBPR) Expert Advisory Committee
Synthesis Report, Review of the Document Entitled, "Draft Proposed Sampling Program
to Determine Extent of World Trade Center Impacts to the Indoor Environment."
Note: This report was also an appendix to WTC Community-Labor Coalition's E-docket
submission.
Expert Advisory Committee Synthesis Report
CBPR EXPERT ADVISORY COMMITTEE REVIEW OF THE DOCUMENT
ENTITLED,
"Draft Proposed Sampling Program to Determine
Extent of World Trade Center Impacts to the
Indoor Environment"
David O. Carpenter, M.D., University at Albany, Chair Scott M. Bartell, Ph.D., Emory
University Paul W. Bartlett, B.E.S., M.A., City University of New York (on leave) John
Dement, Ph.D, CIH, Duke University Liam O. Horgan, CIH, Assessment Resources &
Technologies, Inc. Gary T. Hunt, M.S., QEP, TRC Companies, Inc. Richard A. Lemen,
Ph.D., Ass't Surgeon General, US Public Health Service (retired)
-------
EXECUTIVE SUMMARY
Implementation of the sampling plan as written will be problematic. The Phase I
and Phase II sampling presupposes the identification of a WTC chemical
"signature". Development and validation of this signature is a work in progress at
present. Awaiting availability of a valid WTC chemical signature applicable to
both building collapse and emissions from the ensuing fires, respectively, will
likely delay cleanup, if cleanup takes place at all. It is essential that cleanup of
WTC toxics occurs as soon as practical even in the absence of a valid signature.
The sampling proposed is not extensive enough to cover all areas likely affected
by the building collapse and approximately 100 day period of fires that ensued.
Furthermore the sampling should extend beyond the affected areas in order to
obtain information on background levels utilizing identical types of sampling
locations within buildings, sample collection methods, analytical procedures and
quality assurance/quality control (QA/QC) methods. The plan for selecting
buildings and building units for contaminants sampling is not described in
sufficient detail. The proposal to sample only buildings that volunteer to
participate will likely seriously bias the results toward not finding representative
levels of contamination.
The sampling procedures are not adequate to collect small fibers and particles,
which may pose significant health threats and be more concentrated in the indoor
environment. There is no developed QA/QC protocol in the plan. Given the
period of time that has passed since 9/11, it is not appropriate to fail to utilize
contaminant levels from inaccessible areas as a consideration for cleanup.
Inaccessible areas are restrictive to human activity, not to airborne toxics that
circulate in all air spaces to some degree. These represent reservoirs of
contamination or "hot spots" and should be prioritized for examination and
cleanup if warranted. There is an urgent need to quickly identify indoor spaces
where WTC toxics pose a threat to human health and to clean these spaces
immediately.
-------
PROPOSED OBJECTIVES AND TITLE
1. Are the stated objectives of the Sampling Program [1] appropriate and
complete? Should the stated objectives include cleanup of WTC
contaminants, where found?
The stated objectives of the proposal are incomplete. The primary objective of
the Sampling Program should be to identify habitable spaces with ongoing World
Trade Center (WTC) contamination and provide cleanup where warranted. The
intent of the two parallel programs being described (geographic extent and
signature study) may be to provide data needed to determine the need for
cleanup, but the statement of the objectives omits mention of cleanup and
cleanup benchmarks. The first objective may place too much emphasis on
characterization of the geographic extent of contamination, and too little
emphasis on non-geographic factors such as cleaning history. It may be more
sensible to structure the proposal and objectives around the identification and
characterization of all factors that are predictive of contamination, rather than
presupposing the dominance of geography.
It is essential that if WTC contaminants are found that they be cleaned up.
However, meeting the objectives of the program as stated is contingent upon
objective #3, the successful validation of a chemical signature for dusts and/or
combustion products. As a result, the three stated objectives cannot be met
simultaneously. Objectives # 1 (the proposed sampling program) and #2 (the
Phase 2 sampling program), as stated, can only be met if #3 is accomplished
first. It is not clear that such a signature exists, and indeed this committee feels
it is unlikely that one (or more) will be found this long after the WTC collapse. It
is imperative that indoor spaces be cleaned of WTC toxics whether or not a
"signature" is found. Since the studies needed to determine whether or not a
signature exists are likely to take years, the cleanup process must not be held
hostage to the development of a signature. The stated objectives need to be
restated with this recognized. In addition, to assure participation, cleanup and
health liabilities need to be addressed.
-------
2. Does the title, "Draft Proposed Sampling Program to Determine Extent of
World Trade Center Impacts to the Indoor Environment" accurately
characterize the purpose and limits of the proposed plan?
The proposed plan does not purport to attempt to determine the full extent of
contamination, either in terms of geographic distribution or of different types of WTC
contaminants (see answer to question 4), so in that respect the title is inaccurate. It should
be revised to state the ".. .Local Geographic Extent of World Trade Center Impacts of
Five Selected Contaminants of Potential Concern (COPC)..." The words "and Analyses"
should be added to the title immediately following the word "Sampling". The proposed
program is not simply a sampling program. To fulfill this revised title, there needs to be
a clearer understanding of the criteria to expand the boundaries of the study and
assurances that the sampling is sufficient to be truly representative. The title should be
more inclusive to include a statement concerning adequacy of the cleanup for the safety
of the occupants.
PROPOSED GEOGRAPHIC SCOPE OF SAMPLING
3. Will the proposed process for selection of buildings to be sampled
provide an adequately representative sample with regard to distance from
Ground Zero, building type, building occupancy, building ventilation
systems, and building cleaning history?
The proposal for selection of buildings is vague and appears to be flawed. The
plan states that the generalized random-tessellation stratified (CRTS) spatially
balanced sampling design described by Stevens and Olsen [2] is to be used, but
does not commit to details of how it is to be employed, especially in regards to
non-spatial factors. A spatially diverse sample is required to characterize the
geographic extent of WTC contaminants. Although no experience using CRTS to
characterize pollution from a point or a limited area source is reported, the
technique will provide a sample that is evenly distributed across 2-dimensional
map space, and should allow characterization of the average extent of
contamination in regard to distance and orientation to the source. Altitude of air
intake(s) may also be an important geographic factor in building contamination,
but published CRTS techniques do not consider 3-dimensional space.
At any given distance, elevation and orientation to WTC toxic sources of
exposure, there are a wide variety of non-geographic factors that will affect a
building unit's accumulation and retention of WTC toxics. Buildings, and units
within buildings, are expected to vary in their accumulation and retention of
WTC toxics depending on many factors: distance, altitude, cardinal orientation,
penetration rates (i.e., how easy it is for outside toxics to penetrate a building
-------
through closed windows, ventilation intakes, tracking in, etc.), window usage,
type of ventilation system, and cleaning history. The building selection process
should rely solely on criteria (1) and (3) stated on pages 4 and 5 of the Plan and
other statistical criteria as appropriate. There should be a clear decision as to
how many of which type of building will be sampled (apartments, public
buildings, businesses), and a plan for the numbers of each type of building at
varying distances from Ground Zero.
The proposal suggests drawing a sample from a list of volunteered buildings.
Voluntary participation will likely result in non-representative sampling as stated
on page 4 of the Plan, potentially biasing the study results through self-selection.
Buildings would ideally be selected in a totally random fashion, without regard to
whether or not they have been "volunteered". However, building access may
ultimately require the consent of building owners, making it difficult to entirely
avoid participation bias. The proposed study design collects no information on
buildings that were not volunteered, making it impossible to determine the extent
of participation bias. An alternative sampling approach is to first determine the
location of each type of eligible building in the study region and select a sample
of those buildings, and then contact building owners to request study
participation. Variables that can be obtained without access to buildings (e.g.
building type, location, type of ventilation system, and cleaning history) can then
be compared for volunteered and non-volunteered buildings in order to determine
whether or not volunteered buildings are likely to be representative of all eligible
buildings. This approach would also allow investigators to calculate a
participation rate, and under certain assumptions to adjust for selection bias
using missing data techniques such as the EM algorithm [3] or multiple
imputation [4].
4. Given that EPA EPIC results identify probable deposition of WTC dust in
Brooklyn and that NASA photos of the plume show it crossing Brooklyn on
September 11, is the exclusion of Brooklyn from phase 1 sampling
appropriate? Are there other considerations that should be a part of the
sampling plan, such as obtaining background information and considering
published health effect studies?
The proposal does not describe the rationale for excluding Brooklyn or limiting
the study area to lower Manhattan. Sampling of buildings should be much
broader than planned and should be based on the extent of the plume as
determined either by the NASA photos or other appropriate methods. There
does not appear to be adequate testing of Brooklyn to rule out contamination,
and there should be identifying and sampling upwind locations for inclusion in
establishment of background levels. On perhaps the day of the most intense
emissions, September 11 , the plume can be clearly seen moving east to
southeast over Brooklyn. Newsday reported that the National Weather Service
Data indicated that the plume was over Brooklyn eighty percent of the time [5]. It
-------
is recognized that Manhattan was contaminated from the WTC to the East River.
The East River provides a sink for some of the dust traveling close to the surface,
but the plume from the fires easily transported to Brooklyn. So, given the
distance, there may have been less of the larger heavier particles depositing in
Brooklyn than Manhattan, but the smaller fibers are readily suspended and can
travel to Brooklyn. Given the frequent wind direction to the east and southeast,
Brooklyn should be tested in Phase 1.
We recommend extending the geographic area to be sampled in Phase I to
include all impacted areas, obviating the need for a Phase II. Consideration
should also be given to areas from which residents have been reported to exhibit
adverse respiratory health effects, such as the Chinatown clinic study by Szema
et al. [6] that found increased respiratory effects in children living within five miles
of Ground Zero, and Reibman et al.,
[7] who investigated respiratory disease in residents within one mile of
Ground Zero. The more time that passes the more difficult it will be to
determine the source of contaminants, so all sampling should be done as
soon as possible.
Another important consideration in defining the scope of the sampling is the
necessity of obtaining background levels of the contaminants of potential concern
(COPC) using precisely identical sampling and analytical procedures (methods,
types of sampling locations, etc.) to those to be used in the Plan. It is not
acceptable to use previously obtained background data with differing procedures,
and it is not clear from the available documents that the sampling and analytical
procedures used in previous studies were identical to those to be employed here.
Unless absolutely identical procedures were used in whatever previous
information is proposed for determination of background levels of COPCs, the
background sampling program should extend to areas within New York City
known with a great deal of certainty not to have been affected by the WTC
collapse and ensuing fires. For example, determination of background levels
should include samples from upwind and downwind (out of the anticipated impact
area) of the fall/winter seasonal prevailing winds.
PROPOSED SAMPLING DESIGN
5. Does the proposal adequately address the potential for heterogeneous
dispersion of particulates and combustion byproducts? Does it adequately
consider primary sources of contamination from collapse and combustion
as well as secondary sources of contamination from rescue, recovery, and
trucking and waste transfer operations?
The plan as stated does not adequately recognize the reality of the situation
likely to exist in these buildings. More specifically, dusts found will be a
heterogeneous mixture of the following types/dust sources: 1) those present
-------
historically prior to 9/11; 2) dusts from building collapse on 9/11; 3) dusts and
smoke emitted from WTC property for approximately 100 days after 9/11 and; 4)
dusts unrelated to 9/11 and WTC cleanup from normal activity and use such as
tobacco smoke, wood smoke, construction etc. in the time period 2002-2005.
The COPCs include fibers, which due to their relatively large surface area to
mass ratio tend to stay suspended. The classic case (as sited in asbestos
abatement classes) being an asbestos fiber, which with no other air currents
present takes 88 hours to settle a distance of 12 feet. These fibers should be
capable of migrating the greatest distance. Smaller and denser particles, of the
same mass as asbestos, [polycyclic aromatic hydrocarbons (PAHs), lead] should
migrate a much shorter distance. Therefore, there is almost certainly
segregation of contaminants by distance. This suggests that potential "signature
compounds" will vary by distance, elevation, degrees of particulate size and
degree of resuspension. Therefore, the "signature compounds" may also change
over time. Other metals, such as particulate mercury (mercury was known to be
present in the millions of fluorescent lights, which were crushed), or perhaps
ratios of two or more metals, may be better indicators. The sampling largely
addresses dust dispersion. Secondary dispersion is not directly considered. The
dispersion and deposition processes will vary by type of source (building
collapse, combustion, and waste transfer) and type of particulate. Particulate
dispersion, degradation and deposition vary by particulate size, shape and
chemical properties. Particulates from the pulverization of the building tend to be
larger in size and prone to rapid gravitational settling. The largest particles tend
to settle outside close to the emitting site. They can "grasshopper" further
distances when re-suspended by heavy winds. Combustion byproducts have a
higher distribution of finer particulates, which can be suspended in the air and
travel airborne over great distances, subject to horizontal and upward vertical
movement of turbulence and wind. Cahill, et al. [8] documented extraordinary
concentrations of very fine toxic particulates on top (50 meters) of a building a
mile away, but did not detect significant amounts of large fibers. The finer
particulates are deposited chiefly by coming into contact with surfaces from air
movements, wind and turbulence and not gravitational settling. Generally, hard
surfaces are less subject to deposition, soft surfaces more so. Semi-volatile
organic substances favor organic surfaces (organic films can form on hard
surfaces, such as glass, providing a favorable surface for semi-volatile organic
toxics). Porous fabric surfaces are ideal fine particulate deposition surfaces, so
much so that they are used in particulate pollution control devices. Essentially,
the original mixture of WTC contaminants will become fractionated with distance,
elevation, orientation, building characteristics, interior unit surfaces and cleaning
history. Complicating this process, the source emissions of some contaminants
are more episodic than others so they may not be deposited uniformly in all
directions or elevations. For example, PCBs were most likely emitted in greatest
amounts when the electrical substation below WTC 7 was burning. Deposition
will be greatest for those buildings and units in the path of the plume on those
days. Deposition to interiors will depend upon the physical and chemical
properties of the particle, building characteristics and interior surfaces. A
-------
sufficient sample size will be necessary to be able to characterize the range of
penetration possibilities within and between buildings. The proposal is not
sufficiently detailed to demonstrate that this will be accomplished.
In summary, one cannot expect a uniform mixture of WTC particulate fallout.
This means that some COPCs may be present in some locations, and not other
COPCs, yet originate from the WTC. Similarly, potential signature compounds
from the pulverized buildings may not always be accompanied by potential
signature compounds from the WTC fires, and vice versa.
6. Does the proposal adequately consider the limitations of using
current sampling results to characterize the nature and extent of
contamination that occurred three years earlier?
The current plan does not address the limitations of sampling done three years
ago. The earlier sampling was not well coordinated to evaluate the areas now
under consideration for sampling. The sampling took into account the massive
debris from the plumes immediately after the attacks, but did not follow the
affected areas adequately and did not utilize adequately sensitive sampling
devices to determine the true extent of risk (i.e., short asbestos fibers and other
fibers). Residues left from transport of waste should be studied by following their
trail to the waste disposal sites and need to be assessed to determine further
spread of the contamination to areas not thought to be directly affected by the
collapse and combustion of the impact areas of the attack.
Current sampling results can provide evidence to the low end of exposure of
persistent toxics that are not easily cleaned, but not the extent of initial exposure
nor exposure to substances that are more subject to degradation and/or re-
suspension and escape (e.g., volatiles). Mercury, for instance, was likely to be
emitted in the vapor and particulate form. Particulate mercury will persist over a
greater length of time. If a valid signature cannot be demonstrated (as we
suspect) all of the sources listed in response to Question #5 above will likely
contribute to levels found. Without a valid WTC signature historical perspective
or the time line of contamination will be lost.
Matching of the WTC source signature to chemical signatures found in actual
environmental samples is limited by the sample types collected during the WTC
disaster and available for use in development of the source signature. For
example, if only size fractionated bulk particle samples (e.g., < 10 u or < 2.5 u)
are available for use in development of the WTC source signature then only
samples with identical size fractionation can now be used to develop the
signature of dust samples found in living and work spaces within affected
buildings. Many chemical compounds likely to comprise the WTC source
signature (especially combustion by-products formed during the post-9/11 fires)
will not be equally distributed amongst all particle sizes. The concentrations of
chemicals (weight or mass basis) found on various particle size fractions will
-------
vary. This is true for both WTC emissions as well as dusts now residing in living
and work spaces over three years after the 9/11 event.
7. What limitations and ramifications, if any, will self-selection and the
voluntary nature of participation place on the likelihood of numerically
adequate and representative enrollment?
Self-selection prior to sampling will bias the results significantly by
underestimating the degree of contamination. Volunteers are more likely to be
aware of the WTC risks, and are more likely to have taken preventative and
remedial action (e.g., professional cleaning and remediation). Building owners
and employers who fear health and cleanup liability are less likely to volunteer. It
would be preferable to make a structured random sample of buildings, and then
approach selected building owners and tenants. Government-funded cleanup of
toxics found is essential for gaining participation. For certain classes of buildings,
where a high degree of participation may be possible (e.g., firehouses, schools),
it still would make a better sample to approach the school and firehouse after
being selected by a statistical method than to base selection on a subset of the
class that volunteered. If insurmountable problems compel the original study
approach of sampling from among volunteers only, external survey data should
be collected to determine the potential extent of self-selection bias.
8. Will over- or under- enrollment of presumed clean or presumed dirty
buildings skew results and affect the ability to draw valid conclusions
from sampling data results?
Yes, unless the extent of over- or under-enrollment is known and adjusted for. This is
one of the likely outcomes if the self-selection or voluntary participation process does
not result in a representative population of buildings and no information is available on
buildings that were not volunteered. Given the knowledge of contaminant plume
locations and the availability of some indoor sampling results from previous years, it
may be possible to stratify the sampling design between presumed cleaned and
contaminated buildings. If there is an under-enrollment of one or the other type,
provisions should be made to expand that sample set, consistent with other criteria. It
is essential that the cleaning history of the building be factored into the consideration of
how representative the sample is.
-------
9. Will the "spatially balanced sampling" statistical approach proposed
adequately captures the diversity of possible contamination scenarios?
The proposed sampling plan relies on CRTS design, a sampling technique
developed for spatially balanced sampling of natural resources [2]. CRTS
combines elements of systematic and random sampling in order to achieve a
statistically efficient sample that is evenly distributed across a 2-dimensional
region. The proposal does not explain why the CRTS technique will be used, or
why spatial balance might be desirable in this situation. Although spatial balance
might be helpful for assessing average patterns over a large geographic area, a
non-spatially balanced approach may be more informative for meeting other
study objectives. For example, some degree of spatial clustering of sampled
buildings would provide better estimates of within-neighborhood contaminants
variability which would be useful for planning Phase II sampling and cleanup.
CRTS and other systematic sampling approaches deliberately reduce the
likelihood of spatially clustered samples and may therefore work against some
study goals. Depending on which objectives are most important, a cluster
sample or a simpler stratified random sample [10] may be more appropriate.
Stevens and Olsen's [2] CRTS approach is described to use a two-dimensional
spatial grid. In the EPA proposal, stratification by distance from Ground Zero is
indicated, presumably with two-dimensional hierarchical spatial grids randomly
laid out within each stratum (although the proposal does not describe the latter
step). Because the plume and dust had episodic movements of varying
durations by elevation and cardinal direction, it may be preferable to use a radial
grid that isolates regions of similar cardinal orientation and along urban canyons
(wind/plume corridors), or an approach that defines spatially defined regions in
regards to proximity to the multiple WTC contamination sources. (This is
particularly important for the truck routes used to transfer the WTC waste
materials.)
The most difficult task is to assure that each geographic region defined by
distance and orientation has an adequate representative sample of building and
unit characteristics relevant to penetration, deposition and retention of WTC
toxics. This is not addressed in the current proposal, but might be aided by
developing a list of all eligible buildings and their known characteristics prior to
sampling.
-------
10. Does the proposed spatially balanced sampling frame adequately
specify the decision-making criteria, conditions, and methodology to be
utilized in determining whether to extend the area for sampling?
The decision-making criteria, conditions and methodology appear to be totally
dependent on identification of a "WTC signature", which may or may not be
found. Whether or not a signature is found, it is essential to cleanup any
unacceptable contamination resulting from the WTC event. As stated in the
answer to question # 4, it would be vastly preferable to sample not only all areas
impacted by the plume in Phase I, but also to extend the sampling into areas not
impacted by the WTC collapse in order to obtain comparable background
information. Page 11 of the EPA sampling plan provides only a general
statement with regard to data analyses and decision criteria.
11. If significant WTC contamination is found in one part of the building
meriting cleanup, should there be sampling and/or cleanup of the rest of
the building? Should testing and cleanup be extended to nearby buildings
and if large parts of the geographic grids are found to warrant cleanup,
should more extensive testing be conducted in adjacent grids?
It is difficult to address this question without data on the extent and variability of
WTC contaminants within buildings, neighborhoods, and larger geographic
regions. If units within buildings tend to have similar levels of contaminants,
then it is sensible to remediate entire buildings based on measurements from a
sample of units. The proposed study design will collect measurements from
multiple units per building, and will, therefore, generate data that can be used
to assess the extent of variability within buildings. However, some cleanup
efforts involving ventilation systems may be most practical to implement at the
building level, regardless of variation in contamination among units in that
building.
If buildings in the same neighborhoods, plume corridor or region are found to
have consistent levels of contamination, it may be reasonable to select entire
neighborhoods for cleanup or sampling on the basis of sampling results in a few
buildings in any neighborhood. The EPA proposal implicitly presupposes that
large scale geography will be a strong predictor of contamination, and employs a
sampling design that reduces the potential for collection of multiple samples in
the same neighborhood. In the absence of any previous evidence that
neighborhood contamination levels are well predicted by single buildings in each
neighborhood, it might be wise to choose a multilevel sampling design to obtain
samples from multiple buildings in some neighborhoods and plume corridors and
to specifically address the predictive capability for neighboring buildings in the
data analysis.
-------
Ultimately, the question of predicting contamination in buildings that have not been
sampled suggests the development of a new study design focused on determining what
measurable factors are predictive of current contamination, rather than focusing primarily
on the large scale spatial distribution. It is likely that the best predictions of additional
buildings for sampling or cleanup would be made by a variety of geographic and non-
geographic factors after a Phase I study designed to determine the effects of all relevant
measurable factors.
12. Will absence of a clear commitment in advance to clean indoor spaces
found to be contaminated adversely impact rates of participation in a
sampling study?
Yes. Owners and employers of buildings and units within buildings that suspect
their space to be contaminated will have fears of liability for cleanup and health
impacts. This almost certainly will discourage them from participating. Owners
and employers that have had professional cleaning and have good insurance are
more likely to volunteer to gain assurances of the safety of the building, but not
be representative of other buildings.
PROPOSED CONTAMINANTS FOR SAMPLING
13. The proposal identifies five substances to be sampled (asbestos, man-
made vitreous fibers, crystalline silica, PAHs, and lead). Is testing for
these substances sufficient or are there additional substances for which it
would be appropriate to test?
EPA's selection of COPCs was not meant to be a complete set of hazardous
substances, but they were selected to serve as a set of health-based indicator
chemicals for use in identifying buildings and space within these buildings
warranting cleanup [11 ]. EPA used a set of criteria to restrict the COPCs
tested. One such criterion was the frequency of detection. Unfortunately, many
of the sampling and analytical methods were inadequate. Areas unlikely to be
contaminated, and areas incomparable to inside conditions (e.g., outside ambient
air during episodes where the plume was going in the opposite direction)
produced an excess of non-detects, resulting in the elimination of many
substances found frequently at hazardous levels in many buildings. EPA also
eliminated potential COPCs if they were below a benchmark based on
proportionate mass of the sampled dust. This criterion eliminated many possible
COPCs that may be in hazardous concentrations in indoor environments since
the samples [11-13] evaluated were typically from outdoor settled dust dominated
by large, heavy mineral fibers and particles (e.g., from cement and gypsum).
These larger particles become separated from the finer particulates in the indoor
environment. Particulate penetration rates of buildings are higher for small
particulates than larger particulates. During typical indoor cleaning, many of the
larger particles are removed, leaving behind the smaller airborne particulates to
-------
resettle, or adhere to surfaces. These two factors result in a higher concentration
of the smaller particulates indoors as compared to outdoors, hence increasing
the relative concentration of trace contaminants that are found on smaller
particulates owing in part to the larger surface areas characteristic of smaller
particles in contrast to equal weights of larger sized particles. Unfortunately,
many of the settled dust samples were collected by brushing or scooping up the
dust [12,13], which results in the loss of many of the finer invisible particulates
since they become airborne by the process. Alternatively, Micro Vac methods
were used with a large pore size (e.g., EPA's method used a filter >1.1 u) that did
not collect the very fine particulates, which were found by others to be in
extraordinary high concentrations [8,9].
EPA's elimination of dioxin as a COPC to be sampled also eliminated an
indicator of toxic halogenated organic chemicals. Dioxin was identified by the
EPA as a WTC COPC. Chlorinated dioxin is semi-volatile, as are PAHs, but is
distinguishable as an indicator of combustion with chlorine (e.g., polyvinyl
chloride). If dioxin is present in extraordinary amounts, then other toxic
organochlorines (many of which are unregulated and do not yet have their
toxicity quantified) are likely to be present as well. Cahill [8,9] found a
proportionately high amount of chlorine in the fine particulates. Dioxin has been
found at levels above health based benchmarks at commercial and government
buildings (e.g., 130 Liberty St., 90 Church St., 30 West Broadway). Dioxins
should be tested, but the cost of the test is an issue. One possibility would be to
use the CALUX assay for dioxin-like activity.
14. Given the preponderance of short, very thin chrysotile fibers in WTC
dust, should fibers
-------
did not demonstrate biological activity. It must be kept in mind that Dr. Stanton
has never said long fibers are bad and short fibers are good. In fact, he
appreciated that a large number of short fibers, individually of low tumorogenic
probability, might be more hazardous than fewer long fibers, individually of high
probability [18].
Studies have also found that the majority of asbestos fibers in lung and
mesothelial tissues were shorter than 5 um in length, thus indicating the ability of
the shorter fibers to reach the tumor site and remain there. Therefore, their role
in the etiology of disease is implicated [19]. In typical occupational environments
fibers shorter than 5 um in length outnumber longer fibers by a factor of 10 or
more [20]. Shorter fibers must be studied in more depth and they should not be
disregarded especially when clearance is retarded [21]. Dement and Brown [22]
have reviewed the evidence that chrysotile fibers tend to split longitudinally as
well as partially dissolve, resulting in shorter fibers within the lung.
In humans the majority of asbestos fibers in mesothelial tissues are shorter than
5 um in length, thus indicating the ability of the shorter fibers to reach the tumor
site and remain there [23]. Fubini [24] argues that, because all asbestos appear
nearly equally potent, length and fiber form do not appear influential on the
outcome of disease. Fubini makes this conclusion based on work of Boffenta et
al. [25] which concludes that the specific type of asbestos is not correlated with
lung cancer risk but that industry-specific exposure appears to fit the linear slope
best, a finding also supported by Dement and Brown [26]. For mesothelioma,
induction was related to the time since first exposure and potency with both
industry type and asbestos type [25]. These findings would indicate that fiber
alteration in the manufacturing process is an important variable in ability to cause
disease. The same may be true for those fibers altered as a result of the WTC
event.
The fact that short fibers (< 5 u in length) have been shown to produce toxic
effects in macrophages in vitro and to be fibrogenic and tumorigenic in animals
in vivo [27]; and that they reach the site of mesothelioma development [28]
support the inappropriateness of discounting their role in asbestos-related
diseases as has been done by the EPA contractors Berman and Crump in their
risk assessment index [29]. The data, to date, strengthen the role of short
fibers in the etiology of asbestos-related diseases. The indoor sampling plan
should utilize an analytical methodology with the sensitivity to include short
fibers in addition to those greater than 5 u in length.
-------
15. Environmental sampling of commercial and government buildings at
130 Liberty Street, 30 West Broadway, and 90 Church Street identified
extensive and significantly elevated indoor concentrations of dioxins,
mercury, heavy metals, and other contaminants. How can the apparent
disparities between these findings in commercial and government
buildings and EPA's findings in residential buildings be reconciled? What
impact, if any, should these findings have on the sampling proposal?
130 Liberty Street, 30 West Broadway, and 90 Church Street were heavily
impacted buildings. They are very close to the site and suffered physical damage
leaving the buildings directly exposed. The toxic compounds measured in
extraordinarily high amounts are likely to have impacted residences tested by
EPA, but to a lesser degree. In part this may be because commercial and
government buildings and some residence have HVAC systems, whereas some
residences have windows with individual AC systems. In addition, most of the
EPA residential sampling took place post-cleanup, not pre-cleanup, so these
results are likely biased towards the lower levels. The residences that elected
testing, and not cleanup, were likely to have been cleaner units.
There is also a difference between the concentration of toxics in samples
measured in 130 Liberty Street, 30 West Broadway, and 90 Church Street, and
the samples measured by Lioy, et al [12] and Offenberg et al [13] that EPA
reference. Since the commercial buildings have a much larger sample set [30],
the discrepancy may indicate that the Lioy et al. [12] and Offenberg et al. [13]
samples are not representative. The Lioy and Offenberg samples were primarily
outdoor samples. The bulk samples collected by Lioy and Offenberg were not
collected in such a way as to capture all the finer particulates (they were
collected by brushing and scooping up the samples, resulting in loss of the finer
particulates). The EPA Micro Vac method used a filter that had an efficiency
rating above 1.1 u, which misses a great deal of the smaller particulates, found to
be of great importance by other research [8,9] and commercial testing (Morgan,
see question 18 below). There are also many differences in sampling and
analytical methods used by EPA and their contractors. A great deal of the
sampling and analytical methods used in data reported by EPA were not
sufficiently sensitive to detect contaminants at background levels.
Mercury was frequently found in high concentrations in commercial buildings,
but not in EPA measurements. Some of the discrepancy may be due to EPA
testing of mercury in vapor form rather than particulate form. Mercury is more
likely to persist in an indoor environment in the particulate form. Mercury was
likely produced in the fires in oxide and chloride compounds in the particulate
form. In addition commercial buildings are more likely to have fluorescent lights,
which would be a source of mercury if the fluorescent tube is broken.
-------
The proposed sampling program must represent a state-of-the-art sampling and
analysis effort with adequate QA/QC employed such that the data are fully
defensible. The data resulting from this program can be expected to be
subjected to a great deal of public attention and must withstand scrutiny applied
in a court of law, if necessary.
More importantly, the EPA program design places a great deal of significance
on concentrations of COPC/signature compounds found in background
buildings in Manhattan. The EPA plan suggests a "trigger" of 3X background in
affected buildings as the basis for cleanup. As a result, it is imperative that the
background determination phase of the program results in measured
concentrations of the COPC/signature parameters (see also answer to question
4). Otherwise, the concept of measured concentrations above threshold, when
threshold is "Non-Detect" has no meaning.
16. Has there been adequate attention to the problems of chemical
mixtures?
No consideration has been given in the sampling plan to the cumulative effects
of the COPCs or signature compounds when individual contaminants are found
below published health effects thresholds. More importantly, no consideration
has been given to the complex chemical universe present on these dusts. Many
of these compounds are not currently regulated in any fashion and the vast
majority have not been adequately assessed for health effects. Some of the
organic compounds known to be associated with the dusts (based upon
published analytical data) have not been reported previously in the environment.
The WTC disaster and ensuing fire was a unique event and accordingly the
combustion chemistry in many respects was also unique.
The issue of chemical mixtures is particularly important when dealing with both
carcinogenic and non-carcinogenic substances [see 31]. For example, the
effects of asbestos exposure and smoking are known to be more than additive
(synergistic) [32]. The carcinogenic substances in cigarette smoke include
PAHs, which are major WTC contaminants of concern. Lead, mercury, PCBs
and dioxins are all neurobehavioral toxicants and, at present, we do not know
whether their effects are additive or synergistic. Co-planer PCBs act via the same
mechanism as PCDD/Fs, that were not measured adequately in the EPA studies
to determine their combined effects. Brominated PBBs and PBDD/Fs were likely
to be created in the fires in high quantities (primarily due to PBDE fire
retardants) and act like the coplanar PCBs and PCDD/Fs, but were not
measured. As a result of the failure to consider mixtures, specific health effects
are likely to be underestimated by the benchmark of one set of contaminants. As
an indication of how important the federal government considers the issue of
chemical mixtures, the Agency for Toxic Substances and Disease Registry has
released a series of draft "Interaction Profiles" as a part of their Toxicological
Profiles in 2002.
-------
PROPOSED METHODOLOGIES FOR COLLECTION AND ANALYSIS
17. The draft sampling plan proposes a HEPA vacuuming method for
sampling asbestos, silica, and MMVF on hard and soft surfaces in
residences and workspaces. For sampling those substances in HVAC
systems, wipe samples as well as bulk samples are proposed. (The
proposed HEPA vacuuming method is described in Attachment 1, which
begins on page 22 of the sampling plan, and is then amended on page 30 of
the sampling plan.) Is sufficient attention paid in the sampling proposal to
obtaining data for both hard surfaces and porous surfaces?
Hard and soft surfaces should be sampled for all target parameters and, in the
event a valid signature (or signatures) can be determined, all WTC signature
chemicals. The major problem is sampling of surfaces that have been used for
three-plus years. This is not going to adequately determine the building
contamination. The HEPA method for asbestos is a cause for concern. The 2003
Background study seems to have used the Micro Vac method for sampling of
surfaces. The HEPA method will result in collection of excessive amounts of
organic and inorganic material which may obscure detection of short chrysotile
fibers. For asbestos fiber analyses, a cleanup of the sample by ashing followed
by analyses of samples by the 'indirect method' for TEM should be used. No
details of this are given in the documents reviewed. The Micro Vac proposed
has an efficiency that will not collect particulates less than 1.1 u, and this is a
cause for concern since a great deal of the particulates are less than this size
(Horgan, unpublished observations, see answer to question 18).
The plan should identify what hard or soft surfaces will be sampled. If where to
sample is left to a field decision, it may skew the results. Sampling categories
should include high contact locations and low contact locations (which are most
likely to contain the reservoir of materials that may cause recontamination). The
analytical methods are not adequately described as would be the case in a
typical EPA-sanctioned QAPP prepared for a program of this nature. For
example, there are a number of conflicts in the HEPA Vacuuming Method
appended to the Plan as Attachment 1.
18. Is the proposed method the best method for collection of contaminants
that may remain in carpets, fabric furniture, or drapery more than three
years after the collapse of the World Trade Center? If not, can you suggest
another method?
The proposed use of the HEPA vacuum technique is appropriate for this
application, provided limitations inherent in the method are understood.
Regardless of which type of vacuum device is used for dust collection (HEPA or
Micro Vac), the resulting samples will not be size fractionated. For comparison to
the WTC source sample chemical signature, size fractionation of samples
collected in building spaces will likely be needed. This will not be required,
however, if an identical WTC chemical signature is determined to exist in equal
-------
concentrations in all sizes of particles. This outcome is not likely.
Further, both the HEPA and Micro Vac sample collection techniques will not be
effective for collection of very small sized particles (<1 u). These size particles
will likely pass through the filtration media and reenter the room air via the
vacuum exhaust stream. There is reason to believe that a large portion of the
particles are extremely small in diameter. Several days after the WTC disaster,
there were questions about turning on the HVAC System in various facilities.
The question was will the normal HVAC filters prevent contamination of the
facility? Ambient air samples were collected for total suspended particulate
matter (TSP) (NIOSH Method 0500). After the samples were analyzed for TSP,
an optical size distribution (OSD) (method in development) was performed to get
a rough idea of the size distribution. While the OSD method is not a
standardized method, the heavy skewing to the very small particles diameters
(Morgan, unpublished) was sufficiently compelling to postpone reactivating the
HVAC systems.
TSP
mg/m3
0.05
0.07
0.02
0.13
0.08
0.05
0.04
0.35
0.08
0.08
0.11
0.06
0.08
0.04
0.04
Optical Size Distribution
(microns)
% <2.5
87
83
86
85
80
79
72
97
86
87
86
88
86
94
95
% 2.5-10
13
17
14
14
19
21
28
7
12
13
14
12
14
6
5
%>10
0
1
0
1
1
0
0
0
2
0
0
0
0
0
0
Because of this issue, the ASTM has withdrawn its two guidelines for Vac sampling for
dust. Please note that ASTM has kept its procedures for Vac sampling for asbestos fibers
(see answer to question 24). There is inadequate information on the health effects of
small fibers, but concern because of the fact that they will enter deeper into the lungs than
longer or thicker fibers. Ellouk and Jaurand [33] review information that concludes that
for glass fibers there is greater toxicity of large fibers on a per fiber basis, but equal
toxicity of large and small fibers on a per weight basis when tested against growth and
viability of cultured tumor cells, while thin fibers were more cytotoxic and transformant
than thick fibers on a per weight basis in embryo cells.
-------
Cahill [8,9] found an extraordinary amount of particulates in the range of 0.26-0.09 |i.
However, the Amendment to the Vacuum Sampling Method states that there is a 97%
capture of particles >1.1 jim. That leaves a lot of particles escaping out the back of the
sample collection device. Based upon the current objectives of the plan and end use of
the data, it is not clear whether contaminant concentrations by particle size is needed, but
this may be a serious problem since these small particles, if inhaled, will enter deep into
the lung. The efficacy of the proposed modified HEP A/Micro Vac method could be
compared with ultrasonication, to determine whether the proposed method is adequate as
an indicator of the degree the soft surfaces serve as reservoirs for exposure.
Other sample collection techniques would involve direct removal of soft surfaces from
the building premises for off-site analytical preparation employing perhaps sonication or
direct extraction techniques. These methods are probably too aggressive for the needs of
the current sampling program but selective use could provide very valuable information
on the collection efficiency of the Micro Vac for smaller particles. Also, it is imperative
that particulate-associated heavy metal contamination on porous surfaces be assessed as
part of any indoor contamination assessment.
The plan does not address any of the normal QA/QC issues such as blanks, duplicates,
replicates and spikes. Things like this will help to determine how well the sample was
collected from both a methods and personnel approach.
19. Is the proposed method the best method for the detection of
contaminants on hard surfaces in residences and work spaces? Would
another vacuum method be better? Or, given the amount of time elapsed,
would wipe samples be a preferable method of collection?
The HEPA vacuum sampling technique is appropriate for this application, but
only if fine particulates are not target substances. Both hard and soft surfaces
should be sampled for all target parameters. For hard surfaces both vacuum and
wipe samples should be used. Smoke residues and, in particular, contaminants
associated with smaller yet inhaleable particles, will not be sufficiently removed
by the proposed HEPA collection technique. The wipe sample will insure these
contaminants are included in the total concentrations of target
compounds/COPCs present on the sampling surface.
Methods for this study must be related to what is normal background in New
York City using the same methods. There also must be sampling in
'inaccessible areas'. Contamination in these areas represents potential for
future airborne contamination during cleaning, renovation, etc., as well as
providing a better indication of the original level of contamination.
20. The draft sampling plan does not specify a method for collecting
samples of lead and PAHs from soft surfaces. Should soft surfaces be
sampled for these analytes? If yes, what would be the best method(s) to
-------
use? What would be the best methods to use for analyzing such
samples?
Soft surfaces such as fabrics are ideal deposition surfaces for particulates and
serve as reservoirs for the contaminants. However, the degree to which they
constitute a reservoir is unknown, and soft surfaces and carpets are difficult to
sample in a quantitative way. Metals and persistent organics can be stable for
many years. The finer particulates, particularly if they have organic constituents,
may adsorb and/or absorb to fabric materials. Semi-volatiles, such as PAH's
and dioxins would reemit, due to equilibrium partitioning between the solid and
vapor phase. The degree of exposure from this process is not well reported, but
could be tested and estimated under a variety of conditions (cleaning, moisture,
temperature). Particle associated lead and PAHs present in soft surfaces
should then be sampled in an identical fashion to what is proposed for asbestos,
silica and MMVF. In the present version of the proposed Plan this sampling
method is a HEPA vacuum technique. XRF is a useful method for determining
metal levels in soil, and perhaps could be applied to carpets and textiles,
although at least in the case of mercury the Practical Quantification Limits for
the Niton Instruments XRF is 5-10 times greater than typical laboratory
detection limits, and therefore may not have adequate sensitivity.
21. Should the proposal's methodology for sampling in HVAC systems
be amended to include sampling for deposition in low velocity areas in
duct work and at bends in high velocity areas in duct work?
Yes. Low velocity areas such as plenums upstream of heating/cooling coils are good
locations for collection of settled particles. The sites are reservoirs for dispersion of
contaminants, and the contaminants can be dislodged during irregular disturbances.
HVAC systems and in particular the ductwork represent significant "sinks" of dusts and
associated contamination. The sampling program employed should place a high priority
on this reality.
22. Are the proposed methods for sample collection in HVAC systems the
best methods to use? Would other methods be more appropriate?
These are not provided in detail in the draft document. Both hard and soft
surfaces will be encountered, so both HEPA and wipe samples would appear
appropriate. The proposal does not address the different kinds of duct interiors
that will be encountered. Some will be interior lined and some will be exterior
lined or unlined, resulting in the same hard vs. soft surface problems. In
addition some interior lined HVAC systems have tar-like waterproofing, which
will likely contribute significant background concentrations of PAHs at these
locations. More attention needs to be paid to documenting the type of system
sampled.
-------
The sample plan should designate what parts (intake, blowers, ducts, corners,
splits, diffusers, etc.) of the HVAC system should be sampled and what minimum
number of samples per sq foot need to be collected. This will also lead to a better
correlation when comparing different buildings. It would not be appropriate, for
example, to compare results for a building which had 1 sample per 50 foot of
ductwork and to results for a building which had one sample per 1,000 sq foot of
ductwork.
23. Sampling is proposed on accessible horizontal surfaces such as floors
and table tops that would be most likely to result in dust-related exposures
from residual WTC dust. However, these are also the surfaces most likely
to have been disturbed and/or cleaned since 9/11. Sampling is also
proposed for inaccessible locations such as behind or on top of cabinets.
Residual WTC dust in these areas is less likely to have been disturbed or
cleaned up. However, samples from these inaccessible areas are excluded
from cleanup decision-making criteria? Is this exclusion appropriate?
Absolutely not. Since it is likely that most of the particles are of a very small size,
most homeowner vacuums are going to suck it up off the floor or out of the couch
and blow it right out the back of the vacuum. It then floats around until it settles
in an inaccessible location (where it accumulates) or an accessible location
(where it is once again sucked up by the vacuum cleaner). The result is less
contaminated accessible locations, while a reservoir accumulates in infrequently
cleaned and/or inaccessible locations. Consequently results from none of the
sampling should be excluded from cleanup decision-making criteria. If it is
determined based upon program objectives and statistically based sampling
design that samples are to be collected from a particular location then the results
should be considered part of the cleanup criteria. Dust samples present in
inaccessible locations, like those found in HVAC ducts or ceiling plenums,
represent the most significant reservoirs of contaminated dusts available for
introduction into residential living space and work space alike. This applies as
well to living and work space cleaned previously and viewed as free of dust
contamination. These reservoirs must receive the highest priority in the sampling
program design. Regularly cleaned hard surfaces will give an indication of
regular exposure to dust, but is not the same as a direct measurement of the
reservoirs of the sources. If inaccessible areas and/or soft surfaces are found to
be contaminated enough to be a source of exposure, they should be cleaned, or,
in the case of soft surfaces, removed. Cleaning should target the reservoirs of
the toxics. Also samples collected from 'inaccessible areas' are a good indicator
or surrogate of past contamination.
24. The analytical methods for asbestos, MMVF and silica collected from
hard and soft surfaces by HEPA vacuum have not been defined. What
would be the best analytical method to use for such samples when
collected from soft surfaces? From hard surfaces?
-------
PCM should not be used. It is too crude of a method to measure the thin
chrystotile fibers and the shorter pulverized fibers. TEM is a better alternative.
It is essential to see the smallest of fibers. They should report all fibers counted.
ASTM has the ASTM D57656-02 Standard Test Method for Microvacuum
Sampling and Indirect Analysis of Dust by Transmission Electron Microscopy for
asbestos mass concentration, and another method for determining asbestos
structure. ASTM methods are used when available for the other COPCs, and
also should be used here.
PROPOSAL FOR SIGNATURE STUDY
25. Can a WTC signature be scientifically validated?
If a WTC signature does exist (and it may not) it can be scientifically validated.
However there are several problems. It may not be practical to achieve the
identification of a signature in the required time frame, even if one exists. Explicit
quantitative criteria constituting "validation" must be defined prior to collecting
data for validation. The primary difficulty is the heterogeneous nature of the
contaminants, which results in different transport characteristics in the natural
and built environment. This was discussed in detail in the answer to Question 5.
The phenomenon also applies to prospective signature compounds: WTC dust
and combustion products will become fractionalized by distance, orientation,
elevation, building characteristics, interior surfaces and cleaning history. The
passage of over three years since the event means that there will be an overlay
of new substances that can obscure a WTC pattern.
A successful WTC chemical signature will actually be two (or more) chemical
signatures: one associated with building collapse and a second associated with
WTC fires. What constitutes a signature may also vary with distance from the
site. The best chemical signature for the WTC fires will be comprised of a
chemical compound or more likely a series of compounds (likely combustion by-
products) that can be determined with a high degree of certainty to be unique to
emissions from the WTC. Finding this signature may require extensive chemical
analyses of the WTC source sample set currently archived. Further, the
analytical procedures needed to accurately measure these compound(s) in dust
samples may not be readily available (EPA sanctioned reference methods not
available) or may be time-consuming and costly. For example, brominated
aromatics may represent one such class of compounds that apparently EPA has
already taken into consideration.
Regarding a chemical signature for emissions during the building collapse, the
RJ Lee
[30] approach represents a viable option. He has found a large suite of
compounds frequently found at the close-by, heavily damaged 130 Liberty Street
Deutsche building. However it is unlikely that all of these compounds will be
found in distant locations. Cahill [8.9] discovered a pattern in very fine airborne
suspended particulates one mile away at 50 meters height. While we know that
-------
very fine participates are easily resuspended through human activities, especially
sweeping, we do not know much about their cycle in interior spaces, particularly
retention in soft surfaces and their ability to serve as a reservoir for future
exposure.
EPA is proposing a PAH signature for emissions associated with WTC fires.
This may not be valid for buildings exposed to PAHs on a regular basis, for
instance buildings near the Holland tunnel are exposed to PAHs from traffic that
may overwhelm a PAH signature. Alternatively, areas subject to bright light
containing ultraviolet radiation may enhance the degradation of some PAHs,
resulting in a "weathered" pattern that could render WTC source PAH signature
identification problematic.
Given these difficulties, multiple signatures may need to be developed to account for
location-specific factors.
26. Clifford Weisel has written that for a contaminant to be validated as a
surrogate for the possible presence of other contaminants, it must be
shown to be a component of all dispersed materials, it must not become
separated from other substances during dispersion; there must be
consistency in the ratio of the proposed surrogate to other contaminants;
and cleaning must be as effective for all contaminants as for the surrogate.
Are Weisel's criteria applicable to the signature process?
In the absence of meeting the criteria provided in response to Question #13
these are good criteria. Contaminants have different dispersion and deposition
characteristics. Therefore, their relative rates of contamination will vary by
distance, elevation, surface type, cleaning, building penetration rates, etc. (See
answer to question 5). Other criteria should also be considered in the selection
process such as those indicated in response to Question # 13. Criteria stated by
EPA in the Sampling Plan (top page 9 and middle page 12) should also be
considered. These candidate criteria should be reexamined and the ones
substantiated by EPA test data selected for use in the development of the
signature for use in the Sampling Plan. There is currently some redundancy and
contradiction in the compiled listing of criteria available for consideration from the
above sources.
27. Does the proposal for a signature study adequately specify objective
criteria for validating or invalidating a signature? Should such objective
criteria be determined and stated before or after collection and analysis
of data in the signature study? Does the proposal adequately address the
issues of specificity and sensitivity for any proposed signature(s)?
The signature study proposed by EPA in its current form lacks specificity and in
particular the quantitative criteria needed for a WTC signature to be applied
successfully on dusts present in contaminated buildings. At its present stage of
development and documentation it is impossible at present to reach any
determination of its ability to be conclusive one way or another.
-------
The effort to determine the signature of the WTC fires solely on the PAH profile is
risky (destruction by ultraviolet light, many other sources of PAHs). It may require
more expensive testing to establish the geographic extent of contamination. The
fires had a rich source of halogenated compounds: chlorine (plastics), bromine
(flame retardants) and possibly fluorine (Freon). There is evidence that there
were three types of fires, the fire before the collapse, high temperature fires and
low temperature smoldering fires. The Canadian study [34] identified a possible
signature in the predominance of PCB 126, which is produced by combustion
and not evaporative sources, as a possible signature. Their team also found
evidence of a signature with PCNs and PAHs. The use of brominated
compounds as a signature, especially polybrominated diphenyl ethers (PBDEs),
should be further explored. There has been little reported on the progress of this
effort. Analysis of lead speciation and isotope ratios could result in a signature
for lead. For instance, lead monoxide is used in great amounts in computer
monitor screens. It is possible that there is a sufficiently unique isotope ratio for
this source. Other metals may be found in the very fine particulates, but EPA has
not reported sampling and analysis of particles in this size range.
The Canadian study of organic films on windows [34] provided a uniform
approach and gave guidance to the degree the contamination decreased with
distance. It will be difficult to get as good results this long after the event, but it
may provide a supplemental approach.
The signature validation study must be viewed as a work in progress at
present. At present EPA does not have the data that can provide the
qualitative or quantitative specificity needed to define a successful signature.
These will have to be developed and incorporated for use in the sampling plan
when available. If a chemical signature can be found then one would certainly
want all of those compounds to be a part of the sampling protocol. The best
results can be expected when you have a chemical signature of compounds
found in the source that are not present in the environment or ubiquitous in the
environment under study and not present in the source.
The list of signature compounds suggested also appears to be limited to
compounds with known health effects and/or historical regulatory compliance
significance. The best signature compounds for WTC dust and smoke may
actually be chemicals/elements with little or no health effects, no known health
effects or of no prior regulatory interest.
The list of candidate chemical signature compounds should be expanded to
include compounds unique to WTC dusts or smoke and not commonly found in
the environment. Additional candidates might include brominated semi-volatile
organics (already suggested in EPA document undated), phthalates [see 35],
strontium perhaps as Celestine [strontium sulfate; see 6], particulate mercury
(from the millions of fluorescent lights), cadmium, chromium and/or PCBs. The
WTC had large amounts of PBDEs, present as flame retardants hence providing
feedstock for production of a wide variety of brominated combustion by-products.
According to some documents, the EPA WTC signature task force is
investigating brominated compounds as an indicator of WTC contamination. If so,
-------
it may also be an indicator of the presence of chlorinated dioxins, which are likely
to have also been combustion by-products formed in the 100 days of WTC fires.
28. Can a viable sampling and cleanup plan be developed in the
absence of a validated in a unique signature?
Yes. Many buildings in the vicinity of the WTC have already been sampled,
cleaned and remediated without a validated signature. A number of these
buildings have employed an approach similar to what is presented in the
proposed EPA Sampling Plan without the signature qualification feature. EPA
has already proposed in the current Sampling Plan that COPC levels found in
dust samples will be compared to existing health effects thresholds (if available)
and/or compared to pre-established concentrations of COPCs found in New York
City background samples. If levels of COPCs found in samples in living/work
spaces exceed these established criteria and/or are found to be greater than 3x
background levels (or another acceptable factor) then that space should be
identified for cleanup. This represents a viable cleanup goal that is already
contained in the EPA Sampling Plan.
A good sampling plan that is capable of measuring down to background levels,
and is well structured to include inaccessible areas, building type, elevation,
cleaning history, interior surface type should reveal a spatial pattern of
contamination and what other factors are important. This may require using
multi-variate statistical techniques to make clear the relevant factors that have
resulted in persistent contamination.
29. If one or more valid signatures are found, and elevated levels of
contaminants are found in multiple residences or workplaces within a
building without the presence of the WTC signature, what action, if any,
should be taken?
If no evidence of a valid WTC signature exists and the levels of a contaminant
such as lead exceed the clean-up criteria, for public health reasons the dusts
should still be removed. The problem is deciding who the responsible party is.
There is clear evidence of lead in the dust and smoke aerosol that settled east
of the site [12, 35, 38], so ruling out the WTC as a source is not easy. The unit
nevertheless, out of health concerns, needs to be de-contaminated. If this is
indeed a non-WTC source, City, State and Federal agencies should cooperate
to identify the responsible party.
30. If no signature is validated and elevated levels of COPCs are found
in multiple residences or workplaces within in a building what actions
should be taken?
If the elevated levels area health concern, a plan for cleaning and remediation of
-------
the building needs to be developed regardless of the source and regardless of
the existence or lack thereof of a validated signature. If levels of COPCs found in
samples in living/work spaces exceed established cleanup criteria then that
space should be identified for cleanup. This should be the course of action
regardless of the existence of a valid WTC source signature. The question of
responsibility is more difficult than that for the previous question, though, as the
COPCs in this case may have originated from the WTC, from another source(s),
or from both the WTC collapse and another source(s). In this case there may be
a strong presumption of the WTC as the source if Phase I produces evidence the
COPCs in question are strongly associated with factors related to contaminant
spread from WTC sources. However such evidence will only prove an
association, not a definitive attribution. The strength of association of COPCs
with distance and other factors, as well as ruling out other potential sources, may
help in judging whether or not WTC is likely to have been the source.
Responsible parties for all affected space should be identified and directed to
clean the property, even if the contamination cannot be conclusively linked to
WTC emissions. This is the responsible course of action for protection of the
health of New York City residents and workers.
If unit cleanup costs are relatively inexpensive compared to the costs of
conducting the research, some researchers consider it an ethical obligation to
pay for cleanup of contaminated units at the end of the study. It is worth noting
that this would also provide an incentive for participation, thereby improving the
overall study design by increasing participation and reducing the potential for
selection bias.
31. Is it appropriate to base the whole proposal on an as yet
unvalidated "signature"?
No, it is not appropriate to delay cleanup based on the development of a signature. It is
appropriate to proceed with the sampling plan but only if there is a commitment to
cleanup indoor spaces that are found to have levels of COPCs that exceed those in the
background areas even in the eventual absence of a definitive WTC signature.
PROPOSED CRITERIA FOR CLEANUP
32. In the absence of health-based benchmarks for settled dust, EPA has
proposed 3X background as the trigger for cleanup of asbestos, MMVF,
and silica. Is 3X background an appropriate benchmark for these
contaminants in the indoor environment? In HVAC systems?
By there own admission in the Sampling Plan EPA states (pg 10) that this
approach of factors above an established background does not "allow for
comparison to health-based benchmarks". Ideally the levels of contaminants
measured in dust samples should be evaluated by comparison to health based
levels. In the absence of health based benchmarks, as is the case for the three
parameters above, further justification of the 3X above background criteria is
needed. For example, asbestos at 3X background would be of much greater
concern than MMVF. Why should one value fit all? Further, it is not clear in the
-------
Plan as to what the actual criteria to trigger cleanup are. Will cleanup take place
if only one parameter is above its corresponding health based threshold or 3X
above background (as appropriate) or is it necessary that all parameters meet
these criteria
The crucial issue is how the background levels are determined and reasonable
assurances that 3X background is safe. The NCEA review [37] was defective by
ascribing upper ranges of measured toxic levels in urban areas as reference
background levels, when they were orders of magnitude above the mean. Upper
ranges of measured levels in urban environments are usually cases of proximity
to local unknown sources. There is no information in the plan as to how
background locations were to be collected, nor whether there was to be one
background for the whole city, or by boroughs, or by neighborhood (see answer
to question 4).
Lastly, there is no consideration given in the plan to cumulative effects if all
parameters are below their individual health effects. More importantly no
consideration is given to any of the compounds present in the dusts and not
measured as part of the proposed sampling program. Health effects levels and
acceptable backgrounds even in an urban setting like New York City likely do
not exist for the majority of the compounds (especially semi-volatile organics)
known to be associated with WTC dusts and smoke, [see 12 and 35].
33. Is the proposed 95% upper confidence limit standard on the mean
contaminant level in a building an appropriate methodology to use in
determining whether a cleanup of that building's spaces should take
place? What, if anything, should be done with regard to cleanup of
residences or workplaces in a situation where one or more residences or
workplaces in a building are found to exceed WTC contaminant
benchmarks but the mean contaminant level for the building as a whole
does not satisfy the UCL? Are there other evaluative methods that should
be employed in the determination of whether a cleanup is to occur?
The use of an upper confidence limit (UCL) on the mean contaminant level in a
building is not justified, and has odd implications which have probably not
occurred to the proposal authors. The use of UCLs for EPA hazardous site
assessments is based on the assumption that individuals exposed to hazardous
substances at those sites are equally likely to encounter any sampled location,
so that their long term average exposures will be well represented by averaging
the available measurements. This assumption is probably not true of most
buildings in Manhattan, where individuals consistently live or work in the same
unit or on the same floor and may never visit most units in that building. If there
is any true variability in contamination across units within a building, the UCI will
reflect an averaging of exposure across individuals rather than an averaging of
concentrations to which any one person might be exposed. To understand the
unintended implication of this plan, imagine one small and one large apartment
-------
building that have identical distributions of contaminants across units. In this
case, the large building has more individuals at risk, but the smaller building is
more likely to be selected for cleanup due to a lower sample size producing a
large UCL. Clearly the UCL is not a defensible criterion in this context.
Hard accessible surfaces should not be included in the overall mean. They are
expected to be fairly clean at this point. The critical question is whether there is
a reservoir that will recontaminate an area. For example, if there is a high lead
loading in the dust in a ceiling space, and this is averaged out by no lead on the
floor and counter top, does this really warrant not cleaning it up? If the ceiling
space is a plenum, then cleanup can be justified, if this information is available.
Therefore the different sample locations and collection methods should be
weighted.
The statistical use of the 95% confidence level is only valid if the sampling was
representative and not biased against more highly contaminated areas in the
building. However, it is not clear what statistical measure for 'background' this will
be compared to? Will the study samples be compared to the mean, median, LCL
or UCL of the background samples? The distribution of contaminants in a given
building also should be reviewed carefully. For example, it is entirely possible
that there are physical reasons for some units or areas receiving more
contamination. When a benchmark is exceeded, there should be an investigation
for the possible reservoir sources of the contamination, and a review of the
adequacy of the sample areas tested. Further testing may be called for. A
protocol needs to be developed for this situation.
As a general principle, if dusts collected in an individual residence/apartment or
workplace are found to contain COPC/target parameters above threshold levels,
that residence or workplace should be cleaned. This should be the practice
despite the outcome of statistical analyses done on all samples collected in that
building. These locations could be considered "hot spots". All hot spots should be
remediated. This, too, is a common practice in the remediation of hazardous
waste sites, which EPA draws reference to for guidance in several places in the
proposed sampling plan In many types of adverse environmental exposure, it is
the people that fall into the upper tail of probability that are exposed. This
exposure is real, and so the contamination needs to be remediated. The reason
for the high level of contamination in a particular area needs to be addressed.
-------
REFERENCES
[1] EPA/600/R-04/169A(2004). Draft Proposed Sampling Program to
Determine Extent of World Trade Center Impacts to the Indoor
Environment.
[2] Stevens DL Jr, Olsen AR (2004). Spatially-balanced sampling of natural
resources. J Am Stat Assoc 99(465): 262-278.
[3] Dempster AP, Laird NM and Rubin DB (1977). Maximum likelihood from
incomplete data via the EM algorithm. J Royal Stat Soc 39: 1-38.
[4] Rubin DB (1996). Multiple imputation after 18+ years. J Am Stat Assoc 91:
473-479.
[5] Garrett L (2002). Ill Winds of 9/11: Little scrutiny for Brooklyn - where
attack's toxic smoke drifted. Newsday August 23.
http://www.newsday.com/news/local/newyork/ny-
usplum232835314aug23.story?coll=ny-nynews-headlines NASA photo:
http://www.newsdav.com/news/local/newvork/nv-plumegallerv 0823,0,2622915.
photogallery?coll=ny-nynews-headlines.
[6] Szema AM, Khedkar M, Maloney PF, Takach PA, Nickels MS, Patel H,
Modugno F, Tso AY and Lin DH (2004). Clinical deterioration in pediatric
asthmatic patients after September 11, 2001. J Allergy Clin Immunol 113: 420-
426.
[7] Reibman J, Lin S, Hwang SA, Gulati M, Bowers JA, Rogers L, Berger Kl,
Hoerning A, Gomez M and Fitzgerald EF (2004). The World Trade Center
residents' respiratory health study: New onset respiratory symptoms and
pulmonary function. Environ Health Perspect. On line 20 December 2004.
[8] Cahill TA, Cliff SS, Perry KD, Jimenez-Cruz M, Bench G, Grant PG, Ueda D,
Shackelford JF, Dunlap M, Meier M, Kelly PB, Riddle S, Selco J and Leifer R
(2003-4). Analysis of Aerosols from the World Trade Center Collapse Site, New
York, October 2 to October 30, 2001, Aerosol Science and Technology, 38: 165-
183, 2004. (p. 165), American Chemical Society Meeting 2003
http://delta. ucdavis. edu/WTC. htm
[9] Dalton L. (2003). Chemical Analysis Of A Disaster, Scientists struggle to
understand the complex mixture of aerosols released during and after the
destruction of the World Trade Center, C&EN Washington, October 20, 2003,
Volume 81, Number 42, CENEAR 81 42 pp. 26-30
http://pubs.acs.org/cen/NCW/8142aerosols.html
-------
[10] Cochran WG (1977). Sampling Techniques, 3 Ed. John Wiley & Sons,
New York.
[11] The EPA, in response to peer review stated that "The primary reason for
this decision was to eliminate the time-consuming process of initiating a
remedial investigation (i.e., developing a sampling and analysis plan,
conducting representative sampling of residential dwellings, analyzing a large
number of samples, and finally interpreting results) at a time when re-habitation
of residential dwellings in Lower Manhattan was nearly complete. As a result of
this decision, the COPC selection process associated with the WTC Cleanup
Program assumed a somewhat modified purpose. ... the intent of the COPC
selection process was to identify risk-driving chemicals and to establish specific
health-based benchmarks for the WTC Cleanup Program." (World Trade
Center Indoor Environment Assessment, Response to Peer Review Comments
on the Report for Selecting Contaminants of Potential Concern and Setting
Health-Based Benchmarks, May, 2003
http://www.epa.gov/wtc/response peer review.pdf)
[12] Lioy PJ, Weisel CP, Millette JR, Eisenreich S, Vallero D, Offenberg J,
Buckley B, Turpin B, Zhong M, Cohen MD, Prophete C, Yang I, Stiles R, Chee
G, Johnson W, Porcja R, Alimokhtari S, Hale RC, Weschler C and Chen LC
(2002). Characterization of the dust/smoke aerosol that settled east of the
World Trade Center (WTC) in lower Manhattan after the collapse of the WTC 11
September 2001. Environ Health Perspect 110: 703-714.
[13] Offenberg JH, Eisenreich SJ, Chen LC, Cohen MD, Chee G, Prophete C,
Weisel C and Lioy PJ (2003). Persistent organic pollutants in the dusts that
settled across lower Manhattan after September 11, 2001. Environ Sci Technol
37: 502-508.
[14] Dodson RF, Atkinson MAL and Levin JL (2003). Asbestos fiber length as
related to potential pathogenicity: A critical review. Am J Ind Med 44: 291-297.
[15] "The first decision made concerned that part of the dust spectrum which
should be counted and it was agreed that only fibers or fiber bundles having a
minimum length of 5 u and a maximum of 100 u should be counted, the definition
of a fiber being arbitrarily taken as a particle whose length was at least three
times it diameter. This decision was taken in the light of evidence to the effect
that the particle size distribution or spectrum of an asbestos dust cloud was
reasonably constant over a wide range of textile processes, although later work
has suggested that this might not be strictly true." This decision represents the
conclusions made for use of the Thermal Precipitator Method in collecting
asbestos-containing dust and when the Membrane Filter Technique came into
use, the basis for the method referred to as the PCM method, it was determined
that the 5 u in length would remain the standard as "The filter on the other hand,
having a pore size in the region of 0.45 u, would appear to be quite adequate for
-------
trapping fibers in the length range 5-100 |j." While it was thought the Membrane
Filter Technique would be more representative in assessing the "true health
hazard to which an operative is subjected" it did not rely upon knowledge that
fibers less than 5 u in length had been shown harmless. Holmes S, 1965.
Developments in dust sampling and counting techniques in the asbestos
industry. Ann NYA Sciences: 132(1); 288-297.
[16] Stanton MF and Wrench C (1972). Mechanisms of mesothelioma
induction with asbestos and fibrous glass. J Natl Cancer Inst 48: 797.
[17] Stanton, MF, Laynard, M, Tegeris, A, et al. (1981). Relation of particle
dimension to carcinogenicity in amphibole asbestoses and other fibrous minerals.
JNCI 67(5): 965.
[18] Greenberg M (1984). S Fibers. Am J Indust Med 5: 421-422 &
Personal correspondence from Dr. Morris Greenberg, 23 May 2003.
[19] Suzuki Y, Yuen S (2002). Asbestos fibers contributing to the induction of
human malignant mesothelioma. Ann NY Acad Sci 982: 160-176 & Dodson
RF, O'Sullivan MF, Brooks DR, Bruce, JR (2001) Asbestos content of omentum
and mesentery in nonoccupationally exposed individuals. Tox Indust Health 17:
138.
[20] Dement JM, Wallingford KM (1990). Comparison of phase contrast and
electron microscopic methods for evaluation of occupational asbestos exposures.
Applied Occ Env Hyg 5: 242.
[21] Oberdorster G (2001). Fiber characteristics, environmental and host factors
as determinants of asbestos toxicity. 2001 Asbestos Health Effects Conference,
May 2425, Oakland, CA, US Environmental Protection Agency.
[22] Dement JM and Brown DP (1993). Cohort mortality and case-control
studies of white male chrysotile asbestos textile workers. J Occup Med Toxic
2(4): 355.
[23] Suzuki Y, Yuen SR (2002). Asbestos fibers contributing to the induction of
human malignant mesothelioma. Ann NY Acad Sci 982:160-176; Dodson RF,
O'Sullivan MF, Brooks DR, Bruce JR (2001). Asbestos content of omentum and
mesentery in nonoccupationally exposed individuals. Toxicol Ind Health
17(4): 138-43.
[24] Fubini B (2001). The physical and chemical properties of asbestos fibers
which contribute to biological activity. 2001 Asbestos Health Effects Conference,
May 24-25, Oakland, CA, U. S. Environmental Protection Agency.
[25] Boffetta P (1998). Health effects of asbestos exposure in humans: a
quantitative assessment. Med Lav 89(6): 4714.
-------
[26] Dement JM, Brown DP (1993). Cohort mortality and case-control studies of
white male chrysotile asbestos textile workers. J Occup Med Toxic 2 (4): 355.
[27] Yeager H Jr, Russo DA, Yanez M, Gerardi D, Nolan RP, Kagan E,
Langer AM (1983). Cytotoxicity of a short-fiber chrysotile asbestos for human
alveolar macrophages: preliminary observations. Environ Res 30(1): 224-32.
[28] Suzuki Y, Yuen SR (2002). Asbestos fibers contributing to the induction of
human malignant mesothelioma. Ann NY Acad Sci 982: 160-176.
[29] ERG (2003). Report on the peer consultation workshop to discuss a
proposed protocol to assess asbestos-related risk. Eastern Research Group,
Inc. Prepared for: USEPA, Contract No. 68-C-98-148, work assignment 2003-
05. Final Report, May 30.
[30] RJ Lee Group, Inc. (2004). Review of "Draft Proposal Sampling Program
to Determine Extent of World Trade Center Impacts to the Indoor
Environment", EPA, External Review Draft, from October 15, 2004.
[31] Carpenter DO, Arcaro K and Spink DC (2002). Understanding the human
health effects of chemical mixtures. Environ Health Perspect 110 (Suppl 1)
25-42.
[32] Erren TC, Jacobson M, Peikarski C (1999). Synergy between
asbestos and smoking on lung cancer risks. Epidemiology 10: 405-411.
[33] Ellouk SA and Jaurand MC (1994). Review of animal/In Vitro data on
biological effects of man-made fibers. Environ Health Perspect 102 (Suppl
2): 47-63.
[34] Butt CM, Diamond MLD, Truong J, Ikonomou MG, Helm PA, Stern GA
(2004). Semivolatile organic compounds in window films from lower
Manhattan after the September 11 World Trade Center attacks. Environ
Sci Technol 38: 3514-3524.
[35] Yiin LM, Millette JR, Vette A, llacqua V, Quan C, Gorczynski J, Kendall M,
Chen LC, Weisel CP, Buckley B, Yang I, Lioy PJ. (2004). Comparisons of the
dust/smoke particulate that settled inside the surrounding buildings and
outside on the streets of southern New York City after the collapse of the
World Trade Center, September 11, 2001. J Air Waste Manag Assoc 54(5):
515-528.
[36] Meeker, G, Sutley SJ, Brownfield IK, Lowers HA, Bern AM, Swayze GA,
Hoefen TM, Plumlee GS, Clark RN and Gent CA (2005). Materials
characterization of dusts generated by the collapse of the World Trade Center.
Submitted to the American Chemical Society for publication.
-------
[37] Exposure and Human Health Evaluation of Airborne Pollution from the World
Trade Center Disaster, October 2002, EPA/600/P-2/002A
http://www.epa.gov/ncea).
[38] US EPA. 2003a. World Trade Center Indoor Air Assessment: Selecting
Contaminants of Potential Concern and Setting Health-Based Benchmarks. Prepared
by the US EPA Region 2, New York Response and Recovery Operations. April, 2003
-------
Biographies of CBPR Expert Advisory Committee Members:
Dr. David O. Carpenter is a Research Physician who has held previous
positions as Director of the Wadsworth Laboratories of the New York State
Department of Health and then as Dean of the School of Public Health of the
University at Albany. He is currently Director of the Institute for Health and the
Environment at the University at Albany and Professor of Environmental Health
and Toxicology within the School of Public Health. His research interests are
focused on environmental causes of human disease, where he has worked
extensively on health effects of polychlorinated biphenyls, dioxins, persistent
pesticides and neurotoxic metals. He has research studies of human health
ongoing in Vietnam, Alaska, Alabama and New York, in addition to which he
directs an animal research laboratory that performs investigations on effects of
environmental agents on the nervous and immune systems. He has over 250
peer reviewed publications on various aspects of neuroscience and
environmental health.
Dr. Scott M. Bartell is Assistant Professor of Environmental and Occupational
Health for the Rollins School of Public Health at Emory University. He has
conducted environmental health risk assessment research for most of the last
decade, during which time he also earned graduate degrees in statistics,
environmental health and epidemiology. His research activities include the
development and application of statistical models for exposure assessment,
risk assessment and environmental epidemiology.
Dr. John Dement is a Professor in the Division of Occupational and
Environmental Medicine, Duke University Medical Center. Dr. Dement has
conducted research concerning exposures and health effects of asbestos and
other fibers for over 30 years. Prior to joining the Duke University faculty in 1993,
Dr. Dement served in the U.S. Public Health Service for 22 years where he was
employed in various research and management positions by the National
Institute for Occupational Safety and Health (NIOSH) and the National Institute of
Environmental Health Sciences (NIEHS). He has authored more than 50 peer
reviewed publications concerned with asbestos or man-made fibers. Dr. Dement
is Certified in the Comprehensive Practice of Industrial Hygiene and holds a B.S.
in Mechanical Engineering, a M.S. in Industrial Hygiene, and a Ph.D. in Industrial
Hygiene/Epidemiology.
Mr. Liam Morgan is a Certified Industrial Hygienistwith seventeen years
experience in the industrial hygiene and environmental engineering fields. He
has been responsible for the development, implementation and management of a
wide variety of projects with emphasis in the hazardous materials field. Mr.
Horgan has been involved in the management of investigations and remediations
of over 200 hazardous waste sites with responsibilities ranging from project
safety officer to project manager.
-------
Mr. Gary Hunt is a Vice President of Air Toxics Programs and Director of Air
Toxics Monitoring within TRC in their Lowell, MA office. He works principally
in the toxic air pollutant area and, in particular, the characterization,
quantification and control of toxic air pollutant emissions from stationary and
fugitive sources, as well as their distribution, occurrences, transport and fate
in the atmosphere. Mr. Hunt, who holds a B.S. in chemistry from Villanova
University and an M.S. in Environmental Sciences from Rutgers University,
has more than 27 years of experience in air quality consulting. He is an
internationally recognized expert in the field of toxic air pollutants. Mr. Hunt is
a Qualified Environmental Professional (QEP) and Fellow Member of the Air
& Waste Management Association. He is also a member of the American
Chemical Society, Sigma XI, the Water Environment Federation, and the
American Society of Mechanical Engineers. Mr. Hunt has authored more
than 100 journal manuscripts and symposia presentations on primarily air
quality related topics.
Dr. Richard Lemen is a Consultant in Occupational Safety and Health who is a
retired Assistant Surgeon General in the US Public Health Service. He has
served in senior positions at NIOSH, including Deputy Director and Acting
Director. He holds a MSPH degree from the University of Missouri in
epidemiology and a PhD degree in epidemiology from the University of
Cincinnati. He has been involved in epidemiology research on a variety of
occupationally-related diseases for over 30 years. He was primary author of the
International Agency for Research on Cancer monograph on Asbestos- Vol. 14.
He has published exclusively on the subject of occupational disease and injury
and is co-editor of the book Dust and Disease. Dr. Lemen, an Adjunct
Professor, has taught International Aspects of Occupational and Environmental
Health at Emory University.
Mr. Paul W. Bartlett is a specialist in atmospheric transport and deposition
modeling, monitoring and measurements of trace organic contaminants. For over
nine years, as a research associate at the Center for the Biology of Natural
Systems, Queens College, City University of New York, Mr. Bartlett participated
in path-breaking research in pollution prevention, emission inventories,
atmospheric pollutant modeling, and environmental measurements. Mr. Bartlett
was the principal modeler for a study on the long-range air transport of dioxin
from North American sources to ecologically vulnerable receptors in Nunavut,
which was useful for the development of the Stockholm Convention on POPs and
heavy metals. A collaborative publication of a dioxin study on the Great Lakes
recently won an outstanding scientific award from the Office of Oceanic and
Atmospheric Research at NOAA. Mr. Bartlett has also adapted NOAA's
HYSPLIT model to a regional scale for Florida (atrazine) and the Hudson River
(PCB). Prior to CBNS, Mr. Bartlett worked for the New York City Department of
City Planning on the environmental problems of industrial uses (Environmental
Impact Statements, policy and legislation). He studied chemistry at the
University of Minnesota (B.E.S.) and modeling at the Graduate Faculty, New
-------
School (M.A., ABD). Presently, Mr. Bartlett is on a writing sabbatical leave, but
continues to collaborate with the POPs international inter-comparison modeling
program under the UN Convention on Long-range Trans-boundary Air Pollution.
-------
12. Public E-docket comment submitted by Julie M. Panko, C.I.H., Managing Health
Scientist, ChemRisk, Inc. (dated January 18, 2005):
Risk'
January 18,2005
United States Environmental Protection Agency Office of Environmental Information
Docket EPA Docket Center EPA West Building, Room B102 1301 Constitution Avenue,
N.W. Washington, D.C. 20460
RE: COMMENTS FORE-DOCKET ID No. ORD-2004-0003
Dear Sir/Madam:
This letter responds to EPA's notice published in the Federal Register (69FR 61838)
requesting comments on a document entitled Draft Proposed Sampling Program to
Determine Extent of World Trade Center Impacts to the Indoor Environment
(EPA/600/R-04/169A). ChemRisk has been involved in the evaluation of several
buildings impacted by the World Trade Center collapse. As such, we have reviewed the
above mentioned document with great interest and offer these comments for your
consideration.
1) Usefulness of a WTC Dust Signature: ChemRisk agrees with EPA that a
signature for WTC dust is necessary for making determinations of areas and/or
buildings which have been impacted by WTC dust. Additionally, we agree that
there is not likely to be a single marker that will identify WTC dust from non-
WTC dust, but that a combination of several dust components will be necessary to
identify a signature for WTC dust.
2) Available information from known WTC Dust Samples: In addition to the data
collected by EPA at ground zero and the surrounding areas, three published or
publicly available studies provide a wide variety of analytical information
regarding the chemical constituents of known WTC dust and debris. (USGS,
2001; Chatfield and Kominsky, 2001; Lioy et al., 2002). The WTC Expert
Technical Review Panel is proposing that the signature include various man-made
vitreous fibers (MMVF) such as slag wool, mineral wool and soda-lime glass
possibly in combination with concrete or gypsum as one part of the signature.
ChemRisk agrees that these materials could derived from WTC building related
materials; however, because they are present in many building materials, it is
likely that they may be present in non-WTC dust generated during normal
building renovation or maintenance. Therefore, it would be difficult to discern
the source of these materials unless the sample analysis included a morphological
-------
characteristic that would indicate WTC dust as the source (such as calcined
mineral wool; e.g., that exposed to very high temperatures).
One group of chemical constituents not discussed in the WTC Signature Dust Study is
metals; however, a good deal of information is available regarding the metals
concentrations in known WTC dust samples (USGS, 2001; Chatfield and Kominsky,
2001; Lioy et al., 2002). We have evaluated the metals data provided by these
researchers and we believe that metals may serve as a valid signature for WTC dust.
While the researchers evaluated a different suite of metals in their analyses, there are 10
metals that were common to all three data sets. Additionally, the EPA's National Human
Exposure Assessment Survey (NHEXAS) data (EPA, 2001) provides information on
metals concentrations in normal house dust, including the 10 that were common to the
WTC Dust samples. A summary of these data sets is provided in Table 1 below.
Table 1 Availability of Metals Data from WTC Dust Studies and Indoor House
Dust Sampling *
Metal
Arsenic
Barium
Cadmium
Chromium
Copper
Lead
Manganese
Nickel
Vanadium
Zinc
Lioy et al.,
2002 (N=3)
X
X
X
X
X
X
X
X
X
X
USGS, 2001
(N=12)
X
X
X
X
X
X
X
X
X
X
Chatfield &
Kominsky, 2001
(N=2)
X
X
X
X
X
X
X
X
X
X
NHEXAS -
Arizona (N=26)
X
X
X
X
X
X
X
X
X
X
* These data sets had metals data in mass per mass units (i.e., |ig/g)
Using a statistical fingerprinting method called discriminant analysis; one can distinguish
between samples of known sources based on characteristics of the data sets. The
discriminant analysis establishes a statistical test that allows one to take samples for a
known source and attempt to classify the samples based on certain characteristics.
ChemRisk has conducted a discriminant analysis of the metals data sets for the known
WTC dust samples and that of indoor house dust from NHEXAS and we were able to
predict nearly perfectly which samples came from the WTC Dust data set and which ones
came from the NHEXAS data set based on their metals data. Table 2 is the classification
table that resulted from our discriminant analysis.
-------
Table 2 Metals Classification Table from Discriminant Analysis of WTC Dust and Indoor
House Dust
Actual Data Set
NHEXAS-Arizona
WTC Dust
Total
Predicted Data Set
NHEXAS - AZ
26
1
27
WTC Dust
0
16
16
% Correct
100 (26 of 26)
94 (16 of 17)
98 (42 of 43)
3)
As can be seen from Table 2, only one sample in the WTC Dust data set misclassified as
NHEXAS data. This particular sample was taken from a table top indoors that was
presumed to have WTC dust on it. Therefore, its misclassification may indicate that it is
not WTC dust.
The use of statistical techniques to "fingerprint" sources of environmental contamination
is a common practice. In the case of fingerprinting dioxin contamination, scientists have
used principal component analysis (PCA), polytopic vector analysis (PVA) as well as
discriminant analysis to derive dioxin/furan congener profiles for various emission
sources. Therefore, the use of these techniques would be a valid way to determine the
WTC dust signature.
Although we have reviewed the EPA's World Trade Center Background Study Report
(EPA, 2003), the metals data sets consist of only one metal - lead. Thus, background
metals concentrations in New York City indoor dust are not currently known. If any of
the previously collected background samples have been retained, they could (and should)
be re-analyzed for additional metals and the discriminant analysis re-run using the NYC
data instead of the NHEXAS data.
Signature Criteria: The Expert Panel has proposed five criteria that need to be met in
order to successfully define the WTC dust signature. We believe that none of the
constituents currently proposed by EPA meet Criteria 1: "unique to WTC dusts."
However, we also believe that while none of the proposed constituents of the signature
are distinct from indoor urban dusts, the relationships between the constituents or the
morphological features of some of them are unique. Hence, while metals are found
normally in the indoor dust (as evidenced by the NHEXAS data), the relationships
between the metals in WTC dust are clearly different and distinguishable from indoor
house dust as demonstrated by the discriminant analysis.
With regards to the other four signature criteria proposed by the Panel, ChemRisk
believes that the metals meet all of them. It should be noted that of the various metals
analyzed in the WTC dust samples, mercury was not detected in either the Chatfield and
Kominsky or the Lioy et al. samples, and the USGS did not provide analytical data for
mercury. Thus, inclusion of mercury as part of the signature may not be supported.
-------
4) Lead as a component of the signature: ChemRisk agrees with the EPA, that use of lead
as a component of the WTC dust signature is problematic because it is difficult to
determine whether the source of lead indoors is from lead-based paint or from WTC dust.
Further we agree that additional information should be collected during EPA's sampling
program to determine the likelihood that any lead detected above the established
benchmark is from lead-based paint.
5) We appreciate the opportunity to comment on the proposed WTC Dust Signature Study
and look forward to further scientific dialogue on the topic.
In the way of disclosure, you should be aware that Dennis Paustenbach, the President of
our firm, was on one of the EPA expert panels in 2002 which addressed WTC dust.
Also, like many other consulting firms, we have been retained by a firm to study the
composition of WTC dust.
Sincerely,
Julie M. Panko, CJH Managing Health
Scientist ChemRisk, Inc.
Dennis J. Paustenbach, PhD., CIH, CSP, DABT President and Owner ChemRisk, Inc.
-------
References:
Chatfield, Eric and Kominsky, John. (2001). Summary Report: Characterization of
Particulate Found in Apartments After Destruction of the World Trade Center, October
12, 2001. Requested by Ground Zero Elected Officials Task Force.
(http ://911 di gital archive. org/webcontent/nycosh/WTCcatasSeptNov/WTC_cleanup_testi
mon_y.html) Accessed on December 18, 2005.
EPA (2001). NHEXAS Phase I Arizona Study-Metals in Dust Analytical Results.
Office of Research and Development.
(http://oaspub.epa.gov/heds/study_dir_frame?st_id=23159). Accessed December 18,
2005.
EPA (2003). World Trade Center Background Study Report. Interim Final. April
2003. United States Environmental Protection Agency, Region 2.
(http://www.epa.gov/wtc/backround_study.htm) Accessed on December 18, 2005.
Lioy, P.J.; Weisel, C,P.; Millette, J.R.; et al. (2002). Characterization of the
dust/smoke aerosol that settled east of the World Trade Center (WTC) in lower
Manhattan after the collapse of the WTC 11 September 2001. Environmental Health
Perspectives. Vol. 110. Number 7. July 2002.
United States Geological Service (USGS). (2001). Environmental Studies of the World
Trade Center area after the September 11, 2001 attack. U. S. Geological Survey, Open
File Report OFR-01-0429. Version 1.1 Published November 27, 2001.
(http://pubs.usgs.gov/of/2001/ofir-01-0429) Accessed December 18, 2005.
-------
13. Public E-Docket submission from Sherrie R. Savett, Jeanne A. Markey, Michael T.
Fantini, Berger & (and) Montague, P.C. and Bert A. Blitz, Esquire, Shandell, Blitz, Blitz
and Bookson, LLP (dated January 18, 2005):
-------
01/18/05 14:43 FAX 215 875 3053
BEJRGER & MONTAGUE
12)002
1622 IOCU5T STREET | PHILADELPHIA, PA l»103-6S05 I phone 21 5 / 87 5 -300 0 I fax 21 S / I 7 5-1604 I www.bergermontague.coin
Berger£Montague,P.C.
ATTORNEYS AT
SHERR1E I;MKiN SAVETT
WSI'ES'S DIRECT DIAL
WATER'S DIRECT FAX
WftlTI I'S DIRECT E-MAIS,
215/875-1071
215/875-5715
January 18,2005
VIA FIRST CLASS MAIL
and FACSIMILE
(202) 566-1744
Office of Environmental Information Docket
EPA Docket Center
EPA West Building, Room B102
1301 Constitution Avenue, NW,
Washington, D.C. 20460
Re: Docket Identification Number. QRD-2004-QQ3
Dear Sir/Madam:
This correspondence is written in reference to ORD-2:X)4-003, "Draft Proposed
Sampling Program to Determine Extent of World Trade Center Impacts o: the Indoor Environment",
hereinafter the "Draft Sampling Program". It is submitted by the undersigned as counsel to the
representative plaintiffs, and the putative class said plaintiffs seek to represent, in the case captioned
Benzman et al v. Whitman, et al, 1:04-CV-01888 filed in U.S. District Court for the Southern
District of New York on or about March 10,2004. The representative j i laintiffs and putative class
are herein collectively referred to as "Plaintiffs". Specifically, this lette i enumerates certain issues
pertaining to the adequacy of the Draft Sampling Program which co.insel for Plaintiffs hereby
request be addressed prior to adoption of the Final Sampling Plan. The <. omments contained herein
should not be considered exhaustive and Plaintiffs and their counsel reserve their right to raise
additional issues pertaining to the Draft Sampling Plan and any subsec ient drafts of said plan, or
the Final Sampling Plan, at a later date. Moreover, submission of this cc > rrespondence should by no
means be construed as a waiver, and the Plaintiffs do not waive, the spe :ific relief demanded in the
above-captioned case.
-------
01/18/05 14:44 FAI 21S 875 3053 BERGER & MONTAGUE g,OQ3
Berger£Montague,P.C
»TTO«NIYS AT LAW
January 18,2005
Page-2-
1. Voluntary Nature of the Draft Sampling Plan - Publi; and private buildings
participating in the sampling to determine the geographic extent of the World Trade
Center ("WTC") collapse, and fire plume residues, will do so on a voluntary basis.
This would most likely lead landlords who believe their bi i ilding may contain WTC
contaminants to not volunteer their buildings to participale in the sampling so that
they may avoid potential liability or cleanup costs. The I .'raft Sampling Plan itself
states with respect to the voluntary nature of building aarticipation that, "Self-
selection could result in a non-representative sampling."
2. Consider requiring federal government buildings to paiticipate in the sampling
program.
3. Consider extending the "EPIC" analysis to include dates after September 13,2001
to account for subsequent changes in the direction of the WTC plume.
4. Consider adding buildings in Brooklyn to Phase 1 of the i kaft Sampling Plan since
WTC dust settled there as well.
5. Two dust samples are to be taken within each unit: one: in locations "where dust-
related exposures are likely to occur" such as tabletops iiad a second sample, from
locations where WTC dust may have accumulated but no; necessarily been cleaned,
such as on top of cabinets. But, these later samples will not be used to determine
whether a clean up of the unit will be offered. Whynof' Further, that apparently
means that under the Draft Sampling Plan the areas to I 5 tested are limited to just
those areas which have likely been subjected to multiple Meanings since 9/11. This
would not appear to result in an accurate assessment c.: the presence of COPC's
(Contaminants of Potential Concern) in a given unit or b: lilding but instead create a
distinct bias in favor of finding no need for a cleaning. : 'or each unit and building,
from how many different locations where "dust-related ex | x>sures are likely to occur"
will samples be taken?
6. Units with central HVAC systems will be prioritized ove: those units with individual
air conditioning units, which seems to assume that r,contamination by HVAC
systems would be higher than with individual air conditi< i aer units. What is the basis
for that assumption?
7. Consider the reasonableness of excluding dioxin as a COPC in view of the test
results indicating high levels of contamination with diox i n at 130 Liberty Street, and
in view of the European dioxin standard for emissions Irom municipal incinerators
at stack height of 0.1 ng TEQ/m3.
-------
01/18/05 14:44 FAX. 215 875 3033
BERGER & MONTAGUE
Berger£Montague,P.C.
January 18,2005
Page -3-
8. Consider the reasonableness of excluding mercury as a CO. 'C in view of the findings
of Uday Singh, a certified industrial hygienist who has tinted for the presence of
mercury in Lower Manhattan post-9/11.
9. A health-based standard for lead of 25 ug/ft2 is being used. Consider whether this
is sufficiently conservative in light of the serious learning < I isabilities and behavioral
problems lead can cause in children and the likelihood that children will touch
contaminated surfaces, particularly with wet hands.
10. Consider what support exists for a determination that using the benchmark for
identifying an "observed release" (three times above bad,ground concentrations, as
set forth in the December 14,1990 Federal Register Notice Hazard Ranking System;
Final Rule 55 FR 51532) for asbestos, MMVF and silicon as the benchmark for
determining whether a cleanup of these COPC's will occur, will be sufficiently
protective of human health. To put the concept of an "ob • erved release" in context,
pursuant to Section 105(a)(8)(A) of CERCLA, the EPA a< i opted the Hazard Ranking
System ("HRS") which is a "scoring system used to issess the relative threat
associated with actual or potential releases of substance: at sites" and the primary
way of determining whether a given site will be included on the National Priorities
List for Superfund. gee, Part I of December 14, 199(1 Federal Register Notice,
Hazard Ranking System; Final Rule (55 FR 51532). Purs 'imt to the scoring system,
numerous factors are evaluated and number values are assigned to each with the
ultimate goal of quantifying the relative degree of risl to human health and the
environment nosed at each site by actual or potential releases of hazardous
substances. The "observed release" benchmark in the H azardous Ranking System
is just a subset of the "likelihood of release" factor to n'hich a numerical value is
assigned. The "likelihood of release" factor is, in turn,, ast one component among
many to be evaluated and assigned a numerical value in : rder to ultimately arrive at
a numerical quantification of the health risk associated i"ith a given site based on a
host of factors. Accordingly, even if there is not a "observed release" of a
hazardous substance at a given site, that site could nevertheless qualify as a
Superfund site and, moreover, be subject to remediatior based upon the magnitude
of risk to human health and the environment posed by tf at hazardous substance. In
contrast, using the "observed'release" benchmark in iso. ition, as you propose to do
here, would mean there will never be a cleanup for asbef: tos, MMVFs or silica if the
three times above background level is not found in tiie unit sample, despite the
presence of a risk to human health and the environmen
-------
01/18/05 14:44 PAI 215 875 3053 BERGER & MONTAGUE
— —— —— —• — ——
Berger£Montague,P.C.
ATTORNEYS AT CAW
January 18,2005
Page-4-
Notably, according to Part I of 55 FR 51532, even the entire HRS - and thus
consideration of all its components—is not required to be "i i n accurate determination
of the full nature and extent of contamination at sites cr the projected levels of
exposure such as might be done during remedial inveitigations and feasibility
studies."
In summary then, the "three times background concentra lions" used to identify an
"observed release" is a benchmark which in isolation app ;ars to bear no reasonable
relationship to human safety and thus may well prove to b i; inadequately protective.
11. Consider the impact of synergy between COPC's.
12. Due to the potential cost andburdensomeness, enlisting no ;i-public buildings eligible
for sampling should not be left up solely to volunteers w ithin the community.
13. The Draft Sampling Plan is unclear as to when a whole building, as opposed to a
single unit in which testing has taken place, will be ; leaned. In particular, it
ambiguously states:
The UCL [Upper Confidence Level] will be used in the < 1 ecision process as
follows: If the 95% UCL for the estimated building mean exceeds the
benchmark value for COPC, and concurrently, there is e\ dence of the WTC
signature in the sampled dust in the building, then this m :y be considered to
provide support for the decision to clean the building. («.nphasis added.)
14. To whom will test results be supplied? Tenants and net just owners or landlords
should be notified. And who determines whether a cleaning, if offered, actually
takes place - owners and landlords or tenants?
RespectfuEy submitted,
//
BERGm&MONTAf
ySnerrieR.Savett'
/ Jeanne A. Markey
Michael T. Fantini
1622 Locust Street
Philadelphia, PA 19103
(215) 875-3000
- and-
-------
01/18/05 14:44 FAX 215 875 3053 BERGER & MONTAGUE
m 006
Berger£Montague,P.C.
ATTORNEYS AT LAW
January 18, 2005
Page-5-
Bert A, Blitz, Esquire
SHANDELL, BLITZ, BLITZ & BOOKSON, LLP
150 Broadway, 14* Floor
New York, NY 10038-4498
(212)513-1300
On Behalf of Plaintiffs
390118
-------
------- |