Revisions
to
Ambient
Air
Monitoring
Regulations
Docket
ID
EPA­
HQ­
OAR­
2004­
0018
George
Allen
<
gallen@
nescaum.
org>

05/
08/
2006
12:
13
PM
To
Tim
Hanley/
RTP/
USEPA/
US@
EPA
cc
Phil
Lorang/
RTP/
USEPA/
US@
EPA,
Dick
Valentinetti
<
dick.
valentinetti@
state.
vt.
us>,
Eric
Stevenson
<
EStevenson@
baaqmd.
gov>,
Mary
Stewart
Douglas
<
mstewartdouglas@
4cleanair.
org>,
Dirk
Felton
<
hdfelton@
gw.
dec.
state.
ny.
us>,
Charlie
Pietarinen
<
Charles.
Pietarinen@
dep.
state.
nj.
us>
Subject:
A
followup
on
continuous
pm­
coarse
methods
from
last
week's
Boston
Monitoring
Steering
Comm.
Mtg
Hi
Tim;
time
didn't
permit
me
to
respond
to
your
comment
in
our
meeting
last
week
that
daily
PM­
coarse
will
be
required
because
we
expect
viable
continuous
methods
to
be
available
over
the
next
2­
3
years,
and
that
continuous
pm­
coarse
will
be
easier
than
continuous
pm2.5
because
there
isn't
much
of
an
issue
with
semivolatile
aerosols.
The
semivolatile
part
is
mostly
true,
but
there
are
other
issues
with
PM­
coarse
continuous
methods.

I
don't
agree
with
your
comment
that
a
solid
[
robust]
method
for
continuous
pm­
coarse
will
be
substantially
easier
than
for
PM2.5.
For
non­
compliance
use
[
the
majority
of
the
pm­
c
data
use
in
the
eastern
US],
continuous
PM­
coarse
may
turn
out
to
be
just
as
complex
[
and
have
substantial
data
quality
issues]
as
PM2.5
has
been
over
the
last
few
years.

I
won't
spend
time
here
dealing
with
the
PM­
c
NAAQS
compliance
application.
Why
the
concern
about
levels
well
below
compliance?
We
talked
about
putting
this
method
initially
at
many
NCORE
sites
in
the
east
where
we
don't
expect
compliance
issues
even
at
mid­
scale
sites.
This
means
pm­
c
levels
will
often
be
pretty
low
­­
commonly
into
the
single
digits,
since
means
for
many
non­
megacities
are
under
10
ug/
m3
at
the
neighborhood
scale
and
at
inlet
heights
high
enough
to
reflect
that
scale.
But
we
still
want
these
data
to
have
"
useful"
sub­
daily
precision
for
the
other
[
non­
NAAQS]
uses
of
the
data
­
primarily
for
health
studies.

The
problems
start
with
those
lower
levels
­
typically
1/
2
of
what
we
see
for
PM2.5
for
the
eastern
US
situation.
Right
there,
precision
starts
to
suffer,
especially
on
a
sub­
daily
basis.
I'm
also
assuming
continuous
PM­
coarse
methods
will
be
based
on
the
dichot
approach
[
difference
would
be
truly
ugly].
Virtual
impactors
are
not
perfect,
and
require
maintenance
[
cleaning];
the
cleaning
of
the
PM­
10
inlet
may
also
become
more
important
[
relative
to
what
is
needed
for
the
PM2.5
FRM
inlet].
You
are
also
relying
on
2
continuous
detectors,
not
1,
for
a
dichot
method
to
function
properly.
All
this
adds
to
the
complexity,
maintenance
needs,
and
failure
rates.
What
I'd
like
to
be
sure
happens
is
that
PM­
coarse
continuous
methods
are
tested
not
just
for
NAAQS
compliance
scenarios,
but
also
in
the
situations
I
describe
above.
That's
critical
to
a
successful
deployment
of
these
methods
for
much
of
the
country.
Otherwise,
we're
probably
going
to
fall
back
on
manual
filter
methods
in
the
Northeast,
and
lose
the
sub­
daily
time
resolution
that
we
all
agree
is
also
important.

Finally,
as
stated
in
our
Nescaum
docket
comments,
I
don't
think
you
can
rationally
defend
requiring
a
daily
pm­
coarse
sample
requirement
and
a
3rd­
day
pm2.5
requirement.
In
many
areas,
but
especially
in
the
Northeast
assuming
the
proposed
PM
NAAQS
are
implemented,
the
daily
pm2.5
standard
will
be
the
controlling
NAAQS,
not
the
annual.
With
that
situation,
the
difference
in
required
sampling
schedules
makes
no
sense...
Please
don't
use
a
daily
sampling
requirement
to
drive
the
need
for
continuous
monitoring
[
as
you
implied
in
our
meeting
last
week]!
There
are
better
ways
to
get
there.
­­
George
___________________________
George
Allen,
NESCAUM
101
Merrimac
Street
­
10th
floor
Boston,
MA
02114
Tel.
617­
259­
2035
[
direct]
gallen@
nescaum.
org
Fax:
617­
742­
9162
www.
nescaum.
org
