CRWI's
HANDOUTS
OF
OMB
MEETING
OF
8­
26­
2005
Summary
We
agree
with
the
way
EPA
is
handling
the
following
issues.

1.
The
proposed
methods
for
developing
the
standards.
2.
The
use
of
the
Maximum
Deviation
concept.
3.
CRWI
supports
a
flexible
approach
for
setting
standards
for
liquid
fuel­
fired
boilers.
4.
CRWI
supports
the
use
of
chlorine
as
a
surrogate
for
metals
in
the
HAF
subcategory.
5.
CRWI
supports
EPA's
decision
not
to
go
beyond­
thefloor
for
the
chlorine
standard
for
solid
fuel­
fired
boilers.
6.
CRWI
supports
the
use
of
a
health­
based
alternative
standard
for
chlorine
and
hydrochloric
acid.

We
have
concerns
about
how
EPA
is
handling
the
following
issues.

7.
CRWI
believes
that
the
health­
based
alternative
standard
for
chlorine
(
Cl2)
and
hydrochloric
acid
(
HCl)
should
be
self­
implementing.
8.
Unless
at
least
one
source
is
able
to
meet
all
new
source
standards,
the
new
source
standards
are
not
achievable.
9.
The
top
performers
used
to
determine
the
new
source
standards
must
be
similar
to
the
other
sources
within
the
category.
10.
EPA
should
not
incorporate
data
from
facilities
that
have
already
upgraded
to
meet
the
interim
standards
into
the
calculations
for
the
permanent
replacement
standards.
11.
EPA
should
make
sure
that
data
used
to
develop
the
standards,
especially
for
new
sources,
meet
the
requirements
for
data
quality.
12.
EPA
needs
to
ensure
that
the
current
EPA
approved
methods
can
be
used
to
show
compliance
with
the
proposed
standards.
13.
EPA
should
ensure,
especially
for
new
sources,
that
all
aspects
of
data
uncertainty
are
considered.
14.
EPA
needs
to
justify
any
beyond­
the­
floor
standards.
15.
EPA
should
not
require
site­
specific
risk
assessments
(
SSRA)
as
a
part
of
the
permanent
replacement
standards
rulemaking.
OMB
paper
­
2
­
8/
26/
05
­
2
­
Detailed
discussion
Issues
where
we
agree
with
EPA
1.
The
proposed
methods
for
developing
the
standards.

 
The
Hazardous
Waste
Combustion
(
HWC)
source
category
is
comprised
of
a
diverse
mix
of
industrial
activities,
ranging
from
commercial
and
non­
commercial
incinerators
utilizing
thermal
treatment
to
reduce
the
volume
and
toxicity
of
hazardous
waste,
cement
kilns,
lightweight
aggregate
kilns,
industrial
boilers,
and
halogen
acid
recovery
furnaces
combusting
hazardous
waste
primarily
for
fuel
value,
and
the
specialty
facilities
for
munitions
and
chemical
weapons
destruction.
 
Individual
categories
of
HAPs
 
dioxins
and
furans
(
D/
F),
semi­
volatile
(
SVM)
and
low­
volatile
metals
(
LVM),
particulate
matter
(
PM),
chlorine
and
hydrochloric
acid
(
Cl2/
HCl))
 
require
different
control
technologies.
There
is
no
one­
size­
fits­
all
methodology
for
judging
the
performance
of
the
HWC
source
category.
 
EPA
proposed
to
use
straight
emissions
to
develop
D/
F
emission
standards,
a
technology
approach
to
develop
PM
emission
standards,
and
a
SRE/
Feed
approach
to
develop
most
of
the
rest
of
the
emission
standards.
 
CRWI
continues
to
support
these
proposed
methods
for
developing
the
HWC
MACT
standards.

2.
The
use
of
the
Maximum
Deviation
concept.

 
We
agree
that,
in
setting
emission
standards,
using
the
detection
limit
as
the
default
emissions
rate
for
non­
detects
underestimates
run­
to­
run
variability.
It
appears
that
EPA
has
found
a
reasonable
solution
in
the
Maximum
Deviation
concept
(
Max
Dev)
that
can
be
applied
uniformly
to
all
cases
where
non­
detect
data
are
used
to
provide
a
consistent
method
of
estimating
variability.
 
Our
most
immediate
concern
with
the
concept
is
our
inability
to
independently
verify
this
concept
on
the
actual
data
used
to
develop
the
standards.
These
data
will
not
be
released
until
the
final
rule
is
published.
When
the
method
is
applied
to
the
selection
of
the
top
performers
and
to
development
of
the
floor
values
for
each
HAP
in
each
sub­
category,
problems
may
be
encountered
that
could
not
be
anticipated
from
a
theoretical
evaluation.
 
We
also
have
concerns
about
using
non­
detect
data
to
develop
emission
standards.
When
the
agency
is
forced
to
use
non­
detect
data
to
develop
OMB
paper
­
3
­
8/
26/
05
­
3
­
standards,
we
believe
that
EPA
should
follow
their
own
guidance
and
use
the
reliable
detection
limit
(
RDL)
rather
than
the
method
detection
limit
(
MDL)
 
(
RDL=
2.623
x
MDL).
See
Developments
of
Compliance
Levels
from
Analytical
Detection
and
Quantification
Levels,
USEPA,
NTIS,
PB95­
216321.

3.
CRWI
supports
a
flexible
approach
for
setting
standards
for
liquid
fuelfired
boilers.

 
In
our
comments,
CRWI
suggested
finalizing
the
liquid
fuel­
fired
boiler
standards
as
either
a
thermal
based
emission
limit
or
as
a
concentration
based
emission
limit.
 
EPA
has
indicated
that
the
final
rule
will
allow
flexibility
in
how
the
heatrecovery
units
meet
their
standards.
We
support
the
use
of
flexibility
but
cannot
specifically
endorse
this
new
concept
without
having
seen
it.

4.
CRWI
supports
the
use
of
chlorine/
hydrochloric
acid
as
a
surrogate
for
metals
in
the
HAF
subcategory.

 
CRWI
supports
the
proposal
that
total
chlorine
(
Cl2
+
HCl)
can
be
used
as
a
surrogate
for
particulate
matter,
mercury,
semi­
volatile
metals,
and
low­
volatile
metals
standards
for
the
halogen
acid
recovery
furnaces
subcategory
for
both
existing
and
new
sources.
 
Most
hydrochloric
acid
production
furnaces
use
wet
scrubbers
to
recover
HCl
as
a
product.
These
scrubbers
will
also
remove
the
small
amounts
of
metals
and
particulate
that
may
be
present
in
the
waste
feeds.
 
We
agree
that
controlling
CL2/
HCl
emissions
will
also
ensure
control
of
metal
HAPs
and
particulates.

5.
CRWI
supports
EPA's
decision
not
to
go
beyond­
the­
floor
for
the
chlorine
standard
for
solid
fuel­
fired
boilers.

 
CRWI
supports
EPA's
decision
not
to
go
beyond­
the
floor
for
Cl2
and
HCl
for
solid
fuel­
fired
boilers
6.
CRWI
supports
the
use
of
a
health­
based
alternative
standard
for
chlorine
(
Cl2)
and
hydrochloric
acid
(
HCl).

 
Congress
gave
EPA
the
authority
in
Section
112(
d)(
4)
to
establish
health­
based
alternative
standards
for
HAPs
with
established
healthbased
thresholds.
In
the
report
language,
Congress
encouraged
EPA
to
use
its
discretion
to
avoid
compliance
costs
that
secure
no
public
health
or
environmental
benefit.
OMB
paper
­
4
­
8/
26/
05
­
4
­
 
HCl
is
a
threshold
pollutant.
The
reference
dose
(
RfC)
is
listed
in
the
IRIS
database.
 
The
RfC
for
HCl
has
an
uncertainty
factor
of
300,
thus
already
meeting
the
ample
margin
of
safety
requirement
for
112(
d)(
4).
 
EPA
properly
noticed
and
should
promulgate
the
health­
based
alternative
for
chlorine
and
HCl
as
proposed
with
one
exception
as
discussed
below.

Issues
of
concern
7.
CRWI
believes
that
the
health­
based
alternative
standard
for
chlorine
(
Cl2)
and
hydrochloric
acid
(
HCl)
should
be
self­
implementing.

What
are
our
concerns?

 
As
proposed,
EPA
would
require
agency
approval
of
a
facility's
eligibility
demonstration
prior
to
the
compliance
date.
Should
the
Agency
fail
to
approve
the
demonstration
prior
to
the
compliance
date,
the
facility
is
required
to
comply
with
the
technology­
based
standard.
 
The
Agency's
past
record
for
other
HWC
MACT
approvals,
such
as
CPT
plans,
alternative
monitoring
applications,
etc.,
has
not
given
us
much
faith
that
these
eligibility
demonstrations
will
be
promptly
acted
upon.
Timeliness
of
agency
review
is
further
complicated
by
state/
regional
interactions
and
by
what
authority
will
or
will
not
have
been
delegated
by
the
time
approvals
are
needed.

Why
is
this
a
problem?

 
We
do
not
object
to
submitting
the
eligibility
determination
to
the
Agency
for
approval.
 
Our
only
concern
is
that
the
permitting
authority
can
effectively
remove
this
option
by
failing
to
approve
the
eligibility
demonstration
in
a
timely
manner.
While
the
proposed
rule
allows
the
Agency
6
months
to
make
a
decision,
there
are
no
provisions
for
any
extensions
nor
are
there
any
penalties
on
the
permitting
agency
for
a
failure
to
act.
This
could
force
eligible
facilities
to
expend
resources
to
reduce
emissions
that
would
have
no
benefit
to
human
health
and
the
environment
simply
because
the
Agency
failed
to
act
on
a
timely
basis.
 
We
do
not
believe
that
facilities
should
be
punished
because
the
permitting
agencies
fail
to
approve
these
demonstrations
in
a
timely
manner.
This
is
contrary
to
Congress'
intent
when
allowing
these
types
of
standards.
OMB
paper
­
5
­
8/
26/
05
­
5
­
What
do
we
want
OMB
to
do?

 
If
the
health­
based
Cl2/
HCl
standard
is
self­
implementing
in
the
final
rule,
no
action
by
OMB
is
needed
 
If
the
health­
based
Cl2/
HCl
eligibility
demonstration
requires
prior
approval
in
the
final
rule,
we
suggest
that
EPA
develop
a
method
that
allows
for
extensions
of
the
compliance
date
for
facilities
that
have
submitted
complete
eligibility
demonstrations
but
are
waiting
on
agency
approval.

8.
Unless
at
least
one
existing
source
is
able
to
meet
all
new
source
standards,
the
new
source
standards
are
not
achievable.

What
are
our
concerns?

 
The
"
A"
in
MACT
stands
for
"
achievable."
When
Congress
imposed
the
technology­
based
MACT
regimen
in
1990,
the
first
step
for
dealing
with
HAPs
was
to
determine
what
HAP­
emitting
source
categories
were
actually
achieving,
then
setting
emission
standards
based
on
the
best
level
of
industry
performance.
 
In
making
this
determination
for
the
HWC
MACT,
EPA
has
proceeded
on
a
HAP­
by­
HAP
basis,
calculating
an
average
for
the
best
performing
existing
sources.
In
the
proposed
rule,
EPA
stated
that
at
least
6%
of
existing
facilities
could
simultaneously
meet
all
of
the
proposed
standards
for
existing
sources.
 
In
setting
the
significantly
more
stringent
MACT
standards
for
new
sources,
however,
EPA
did
not
consider
whether
any
single
existing
source
can
actually
meet
the
new
source
standards
for
each
HAP.
 
In
our
comments,
we
showed
that
no
one
source
could
meet
all
the
new
source
standards.
 
We
consider
that
this
renders
the
proposed
new
source
standards
unachievable.

Why
is
this
a
problem?

 
EPA
has
chosen
to
use
a
HAP­
by­
HAP
process
to
develop
existing
standards.
 
This
works
for
existing
sources
because
the
method
averages
emissions
for
the
top
performers
and
many
of
the
top
performers
in
the
HAP
categories
overlap.
 
When
EPA
chooses
the
best
top
performer
for
each
HAP
category
to
set
the
new
source
standards,
each
standard
comes
from
a
different
facility.
OMB
paper
­
6
­
8/
26/
05
­
6
­
For
example,
the
chlorine
standard
is
from
source
349,
the
mercury
standard
is
from
source
3019,
the
LVM
standard
is
from
source
341,
the
SVM
standard
is
from
source
810,
the
PM
standard
is
from
source
3011,
and
the
dioxin/
furan
standard
is
from
source
222.
 
As
a
result,
no
single
source
can
simultaneously
meet
all
the
new
source
standards.
 
In
addition,
technologies
are
incompatible.
For
example
the
best
technology
to
control
PM,
SVM,
and
LVM
is
fabric
filters,
the
best
technology
to
control
mercury
and
dioxin/
furan
is
activated
carbon,
and
the
best
technology
to
control
chlorine
is
wet
scrubbing.
Fabric
filters
and
activated
carbon
can
work
well
together.
However,
neither
fabric
filters
nor
activated
carbon
works
well
with
wet
scrubbing
systems
without
some
method
to
remove
the
water
in
the
gas
stream
(
not
impossible
but
not
easy
and
requires
additional
energy
input).
 
Congress
anticipated
this
problem
and
addressed
it
during
a
colloquy
between
Senators
Dole
and
Durenberger
during
the
Senate
debate
on
the
Clean
Air
Act
Amendments
of
1990
conference
report
(
Leg
History
Vol.
1:
1129).
See
text
of
exchange
below.

"
Mr.
DOLE.
This
section
also
requires
the
development
of
standards
for
a
variety
of
pollutants.
It
is
entirely
possible
that
different
technologies
may
reduce
one
pollutant
better
than
another.
For
example,
technology
A
may
reduce
heavy
metals
better
than
technology
B
while
technology
B
may
reduce
particulates
better
than
technology
A;
yet,
one
would
not
be
compatible
with
the
other.
I
would
assume
that
EPA
would
have
adequate
discretion
to
balance
environmental
benefits
to
determine
which
technology
on
the
whole
represents
a
better
MACT.
I
would
appreciate
some
discussion
on
this
point
as
well
from
my
distinguished
colleague
from
Minnesota.

Mr.
DURENBERGER.
The
Senator
is
correct.
Where
differing
air
pollution
control
technologies
result
in
one
technology
producing
better
control
of
some
pollutants
and
another
producing
better
control
of
different
pollutants
but
it
is
technically
infeasible
according
to
the
MACT
definition
to
use
both,
EPA
should
judge
MACT
to
be
the
technology
which
best
benefits
human
health
and
the
environment
on
the
whole."

What
do
we
want
OMB
to
do?

 
To
make
the
standards
for
new
sources
"
achievable,"
at
least
one
existing
facility
has
to
be
able
to
meet
all
standards
simultaneously.
If
OMB
paper
­
7
­
8/
26/
05
­
7
­
this
is
not
done,
the
standards
for
new
sources
are
not
achievable
and
are
vulnerable
to
challenge.
 
Direct
EPA
to
address
the
statutory
requirement
that
new
source
standards
be
based
on
actual
performance
of
the
best
performing
single
source.
To
do
this,
EPA
has
to
show
that
at
least
one
existing
facility
can
meet
the
all
new
source
standards
simultaneously.

9.
The
top
performers
used
to
determine
the
new
source
standards
must
be
similar
to
the
other
sources
within
the
category.

What
are
our
concerns?

 
Based
on
the
database
used
for
the
proposed
permanent
replacement
standards
for
PM,
the
database
contains
38
rotary
kilns,
39
liquid
injection
units,
6
fixed
hearths,
5
fluidized
beds,
3
rotary
hearths,
2
roller
hearths,
2
controlled
air
units,
and
1
moving
hearth.
 
The
two
most
common
configurations
are
rotary
kilns
and
liquid
injection
units.
A
rotary
kiln
will
typically
be
40­
60
feet
long,
12­
16
feet
in
diameter,
feed
10,000­
15,000
pounds
per
hour,
and
have
a
thermal
rating
of
60­
120
MM
BTU/
hr.
 
The
top
performers
are
as
follows:


Chlorine
 
Unit
349,
rotary
kiln,
5
feet
inside
diameter,
thermal
rating
of
4.9
MM
BTU/
hr,
reported
Cl2/
HCl
emissions
of
0.18
ppmv.


Mercury
 
Unit
3019,
vertical
fired,
liquid
injection,
reported
mercury
emissions
of
8.1
ug/
dscm.


LVM
 
Unit
341,
fixed
hearth,
feed
900
pounds
per
hour,
thermal
rating
of
6.4
MM
BTU/
hr,
reported
LVM
emissions
of
8.9
ug/
dscm.


SVM
 
Unit
810,
liquid
injection,
reported
SVM
emissions
of
6.5
ug/
dscm.


PM
 
Unit
3011,
rotary
hearth,
small
arms
munitions
furnace,
2080
pounds
per
hour
feed
rate,
reported
PM
emissions
of
0.0007
gr/
dscf
(<
2
mg/
dscm).

Why
is
this
a
problem?

 
Section
112(
d)(
3)
of
the
Clean
Air
Act
requires
that
the
standards
for
new
sources
be
developed
based
on
"
the
emissions
control
that
is
achieved
in
practice
by
the
best
control
similar
source "
(
emphasis
added).
 
When
setting
MACT
standards
for
existing
sources,
the
average
of
several
sources
is
used,
thus
diffusing
the
effect
of
one
anomalous
facility
on
the
eventual
standard.
When
the
regulatory
task
is
selecting
one
facility
to
use
as
the
benchmark
to
set
new
source
MACT
standards,
OMB
paper
­
8
­
8/
26/
05
­
8
­
however,
special
attention
must
be
paid
to
how
representative
that
"
top
performer"
is
of
the
overall
source
category.
 
For
the
following
reasons,
CRWI
submits
that
the
"
top
performers"
used
as
the
models
for
EPA's
new
source
MACT
standards
are
not
representative
of
the
overall
HWC
source
category.


The
chlorine
standard
for
new
sources
is
based
on
a
facility
that
is
much
smaller
than
the
typical
hazardous
waste
combustor.
It
is
less
than
half
the
diameter
(
5
feet)
of
a
typical
rotary
kiln
(
12­
16
feet)
and
has
a
significantly
lower
thermal
rating
(
4.9
MM
BTU/
hr
vs.
60
MM
BTU/
hr).


The
mercury
standard
is
based
on
what
could
be
considered
a
representative
unit
 
however,
the
unit
did
not
feed
significant
amounts
of
mercury
based
on
a
strategy
to
develop
an
annualized
permit
limit,
and
a
need
to
show
only
40%
SRE.
Furthermore,
the
testing
protocol
employed
a
non­
standard
sampling
rate
with
the
BIF
metals
sampling
method
in
Part
266
Appendix
IX,
which
has
a
potential
low­
bias
for
mercury
if
precipitate
forms
in
the
KMnO4
impingers
(
Method
29
is
required
to
show
compliance
under
the
HWC
MACT
and
has
procedures
to
avoid
the
bias.).


The
LVM
standard
is
based
on
a
small,
fixed
hearth
unit,
atypical
of
common
HWC
technology.
In
addition,
this
unit
feeds
significantly
less
waste
(
900
lb/
hr
vs.
10,000
lbs/
hr)
and
has
a
much
lower
thermal
rating
(
6.4
MM
BTU/
hr
vs.
60
MM
BTU/
hr)
than
does
typical
rotary
kilns.
Furthermore,
seven
of
the
nine
measurements
in
the
database
were
non­
detects,
and
80%
of
the
waste
profile
is
nonhazardous
(
Note:
CRWI
commented
that
there
were
several
errors
in
the
test
report
and
reporting
methodology
for
this
facility,
making
it
questionable
for
use).


The
PM
standard
is
based
on
a
small
rotary
hearth
unit,
also
atypical
of
common
HWC
technology.
In
addition,
the
feed
rate
of
this
unit
is
significantly
less
than
typical
rotary
kilns
(
2080
lb/
hr
vs.
10,000
lb/
hr)
and
its
feed
type
(
small
arms
munitions)
is
not
similar
to
typical
rotary
kilns
combusting
organic
waste.


The
SVM
standard
is
based
on
what
could
be
considered
a
typical
unit
 
however,
this
unit
had
already
upgraded
to
meet
the
interim
standard
prior
to
testing.

What
do
we
want
OMB
to
do?

 
EPA's
methodology
for
setting
HWC
MACT
standards
for
new
sources
does
not
meaningfully
address
the
statutory
requirement
that
benchmark
performers
must
be
similar
to
the
overall
source
category
and
likely
proposed
new
facilities.
Ask
EPA
to
modify
their
technique
for
setting
OMB
paper
­
9
­
8/
26/
05
­
9
­
new
standards
to
include
some
mechanism
to
ensure
that
the
best
performing
source
for
each
HAP
is
actually
similar
to
the
units
in
each
source
category.

10.
EPA
should
not
incorporate
data
from
facilities
that
have
already
upgraded
to
meet
the
interim
standards
into
the
calculations
for
the
permanent
replacement
standards.

What
are
the
concerns
with
the
revised
database?

 
After
the
court
vacated
the
1999
HWC
MACT
standards,
EPA
initiated
a
data
gathering
effort
to
update
the
database.
 
EPA
appropriately
removed
facilities
that
have
shut
down
or
no
longer
burn
hazardous
waste.
 
In
addition,
EPA
added
data
that
was
gathered
from
facilities
after
they
had
upgraded
to
meet
the
interim
standards.
We
believe
that
including
data
from
upgraded
sources
is
not
appropriate.

Why
is
this
a
problem?

 
We
believe
that
including
facilities
that
have
already
upgraded
to
meet
the
interim
standards
no
longer
measures
the
state­
of­
the­
art
as
it
existed
at
the
time
mandated
by
Congress,
but
rather
the
performance
of
a
source
category
already
affected
by
the
MACT
regulatory
initiative.
The
result
of
such
a
modified
MACT
pool
is
an
inevitable
downward
ratchet
of
emission
requirements
not
anticipated
or
intended
by
Congress.
The
shorthand
description
for
this
unique
regulatory
phenomenon
is
"
MACT
of
MACT,"
a
MACT
standard
derived
from
performance
data
of
facilities
that
have
already
modified
their
operations
to
meet
the
MACT
standards.
 
EPA
has
partially
acknowledged
this
and
has
already
removed
data
from
a
new
cement
kiln
because
the
new
kiln
was
designed
to
meet
the
interim
standards.

Would
removing
data
from
upgraded
facilities
make
any
difference
when
developing
the
standards?

 
This
will
only
matter
if
one
of
the
facilities
that
has
upgraded
or
completed
its
comprehensive
performance
test
is
included
in
the
top
performers
(
EPA's
way
of
designating
the
top
12%
as
required
by
Section
112
of
the
Clean
Air
Act).
 
The
top
performing
facility
for
SVM
(
810
 
Eastman
Chemical,
Kingsport,
TN)
had
already
upgraded
to
meet
the
interim
standards.
OMB
paper
­
10
­
8/
26/
05
­
10
­
 
Not
only
will
this
have
an
impact
on
the
existing
source
SVM
standards,
it
will
set
the
new
source
standard
for
SVM.
Thus,
the
new
source
standard
will
be
set
by
a
facility
that
has
already
upgraded
to
meet
the
interim
standard.

What
do
we
want
OMB
to
do?

 
If
data
from
upgraded
facilities
is
included
in
the
top
performers,
we
believe
that
data
should
be
removed
and
the
data
from
the
facility
that
is
ranked
next
should
be
substituted.
This
would
not
require
EPA
to
completely
redo
the
analysis
 
only
that
they
substitute
one
or
more
facilities
and
recalculate
the
second
step
of
the
standards
setting
process.

11.
EPA
should
make
sure
that
data
used
to
develop
the
standards,
especially
for
new
sources,
meets
the
requirements
for
data
quality.

What
are
our
concerns?

 
The
data
used
to
develop
the
standards
was
developed
for
a
totally
different
purpose.
 
There
is
nothing
wrong
with
using
data
developed
for
another
purpose.
However,
to
do
this,
the
suitability
of
the
data
should
be
examined
in
light
of
the
new
use
for
the
data.
For
example,
when
a
facility
is
testing
to
show
compliance
with
a
0.08
gr/
dscf
PM
limit
under
RCRA,
there
is
no
reason
for
the
permitting
agency
to
question
data
quality
of
a
test
that
showed
0.008
gr/
dscf
in
PM
emissions,
simply
because
the
test
results
were
so
far
below
the
required
limit.
When
this
data
is
used
for
a
different
purpose
(
to
set
emissions
standards
under
MACT),
additional
scrutiny
of
this
data
is
appropriate.

Why
is
this
a
problem?

 
One
way
to
show
this
is
to
look
at
the
top
performers
for
PM.
 
EPA's
Emissions
Measurement
Center
developed
a
report
in
1996
that
concluded
that
the
minimum
catch
for
a
±
10%
accuracy
for
PM
would
be
7.2
mg.
 
Based
on
the
data
from
the
2002
database,
a
member
company
(
Lilly)
estimated
the
catch
from
the
top
performers
for
PM.
The
catch
ranged
from
0.9
mg
to
1.9
mg,
far
below
the
minimum
catch
determined
by
EPA
to
achieve
reasonable
accuracy.
Again,
for
the
purpose
of
showing
compliance
with
a
RCRA
limit
of
0.08
gr/
dscf,
the
quality
of
the
data
was
more
than
adequate
since
resulted
in
PM
emissions
ranging
from
0.0007
OMB
paper
­
11
­
8/
26/
05
­
11
­
to
0.0012
gr/
dscf.
However,
data
quality
objectives
did
not
exist
to
validate
the
result
at
the
much
lower
concentrations
(
two
orders
of
magnitude
below
the
0.08
gr/
dscf
limit).
Without
an
adequate
data
quality
determination,
these
results
should
not
be
considered
as
adequate
for
setting
standards
in
the
0.0007
range.
These
analyses
were
provided
to
EPA
during
the
comment
period.
 
From
this,
one
could
conclude
that
the
proposed
PM
standard
for
new
sources
is
based
on
data
of
questionable
quality.

What
do
we
want
OMB
to
do?

 
OMB
should
require
EPA
to
reanalyze
the
database
in
the
following
manner.


Select
representative
sources.


Exclude
facilities
that
have
already
upgraded
to
meet
MACT.


Follow
the
data
quality
program
recommended
by
EPA's
Office
of
Environmental
Information.

12.
EPA
needs
to
ensure
that
the
current
EPA
approved
methods
can
be
used
to
show
compliance
with
the
proposed
standards.

What
are
our
concerns?

 
EPA
committed
in
a
response
to
previous
comments
that
"
MACT
standards
will
not
be
set
below
levels
which
are
not
consistently
or
accurately
achievable
using
established
(
and
required)
sampling
methods."
 
Most
of
the
methods
used
to
show
compliance
have
never
been
evaluated
at
the
levels
being
considered
for
some
existing
and
all
new
source
standards.
 
For
example,
Section
1.2
of
SW­
846
Method
0050
states
that
"
This
method
is
not
acceptable
for
demonstrating
compliance
with
HCl
emissions
standards
less
than
20
ppm,"
but
both
the
proposed
existing
and
new
source
standards
are
below
20
ppm.

Why
is
this
a
problem?

 
While
the
laboratory
techniques
themselves
can
be
very
accurate,
the
accuracy
and
precision
of
the
sampling
train
may
not
be
capable
of
consistently
delivering
results
to
show
that
new
sources
are
in
compliance
with
the
standards.
 
When
standards
were
fairly
high,
the
variability
relative
to
the
actual
standard
was
not
a
significant
issue.
As
standards
get
more
stringent,
OMB
paper
­
12
­
8/
26/
05
­
12
­
the
variability
becomes
relatively
higher
and
designing
a
test
to
show
compliance
with
these
more
stringent
standards
becomes
more
complicated
(
e.
g.,
target
measurement
levels
must
be
set
further
below
the
emission
standards).
 
For
example,
Method
5i
for
PM
states
a
practical
quantification
limit
of
3
mg.
DOE
funded
research
has
shown
that
the
95%
confidence
interval
for
a
3
mg/
dscm
PM
measurement
is
bounced
by
2
and
4
mg/
dscm.
A
99%
confidence
interval
(
one
chance
in
a
hundred
for
random
failure)
would
be
even
broader.
The
new
source
PM
standard
is
less
than
2
mg/
dscm.

What
do
we
want
OMB
to
do?

 
Require
EPA
to
document
method
performance
as
necessary
to
demonstrate
compliance.

13.
EPA
should
ensure,
especially
for
new
sources,
that
all
aspects
of
data
uncertainty
are
considered.

What
are
our
concerns?

 
EPA
has
not
considered
all
the
uncertainties
associated
with
the
feed
MTEC
calculation
that
it
uses
in
its
rankings.
EPA
is
only
accounting
for
run­
to­
run
variability
which
is
an
assessment
of
precision
only,
not
accuracy.
 
Even
in
accounting
for
run­
to­
run
variability
in
the
feed
MTEC
calculation,
EPA
has
apparently
excluded
the
uncertainties
associated
with
stack
flow
measurements
and
stack
oxygen
measurements,
both
of
which
are
part
of
the
MTEC
calculation.
 
Trial
Burn
quality
assurance
plans
typically
specify
an
acceptable
accuracy
for
metals
in
waste
feeds
as
+/­
35%.
Accuracy
of
metals
assays
in
organic
matrices
is
typically
poor,
but
precision
may
be
good.

Why
is
this
a
problem?

 
For
example,
take
source
341
which
is
reportedly
the
best
performer
for
LVM.
Its
feed
MTEC
ranking
is
11th.
Is
the
feed
MTEC
ranking
of
the
13th
ranked
source
statistically
different?
The
13th
ranked
source's
feed
MTEC
(
source
3016)
is
only
11%
higher
which
is
certainly
within
the
typical
+/­
35%.
Furthermore,
the
measured
average
MTEC
for
the
13th
ranked
source
(
without
the
run­
to­
run
variability
factor)
is
actually
less
than
the
11th
ranked
source.
How
can
the
performance
of
these
sources
be
considered
different?
OMB
paper
­
13
­
8/
26/
05
­
13
­
 
The
calculation
of
the
feed
rate
MTEC
is
a
function
of
the
metals
being
fed,
the
stack
flow,
and
the
oxygen
concentration
(
the
later
because
the
MTEC
is
corrected
to
7%
oxygen).
There
are
several
measurements
of
stack
flow
and
oxygen
taken
during
a
trial
burn,
but
the
EPA's
calculation
apparently
uses
a
constant
value
for
both
when
calculating
feed
MTEC.
This
artificially
dampens
the
apparent
run­
to­
run
variation.
What
is
the
potential
affect?
Recalculating
the
feed
rate
MTEC
for
source
341
using
the
highest
variation
of
stack
flow
and
oxygen
resulted
in
a
31
%
increase
in
feed
rate
MTEC,
moving
its
ranking
two
places
down
on
the
list.
 
A
more
comprehensive
assessment
of
uncertainty,
using
one
or
both
of
the
examples
above,
would
change
source
341'
s
overall
performance
ranking
from
uniquely
#
1
to
tied
for
#
1
with
the
next
two
sources.
Accordingly,
the
new
source
LVM
standard
would
potentially
change
from
8.9
ug/
dscm
to
something
close
to
the
interim
standard
of
97
ug/
dscm.

What
do
we
want
OMB
to
do?

 
OMB
should
require
EPA
to
assess
all
aspects
of
uncertainty
in
its
calculations
and
rankings,
and
not
just
run­
to­
run
precision.

14.
EPA
needs
to
justify
any
beyond­
the­
floor
standards.

What
are
the
concerns?

 
We
have
concerns
about
a
dioxin/
furan
beyond­
the­
floor
standard
for
halogen
acid
furnaces

In
the
proposed
rule,
EPA
justified
a
beyond­
the­
floor
standard
for
this
sub­
category
based
on
reducing
dioxin/
furan
emissions
by
2.3
grams
per
year.


About
a
third
of
the
units
in
the
database
used
to
propose
the
standard
were
either
misclassified
or
have
ceased
operations.
CRWI
or
member
companies
have
pointed
this
out
to
the
Agency
during
the
comment
period.


When
the
database
is
updated,
the
dioxin/
furan
emissions
reduction
will
be
reduced
to
1.3
grams
per
year.

Why
is
this
a
problem?

 
It
seems
like
a
lot
of
effort
and
expense
for
a
small
amount
of
reduction
in
dioxin/
furan
emissions.
OMB
paper
­
14
­
8/
26/
05
­
14
­
What
do
we
want
OMB
to
do?

 
Ask
EPA
to
re­
examine
this
beyond­
the­
floor
decision
to
determine
if
the
calculated
reduction
in
dioxin/
furan
emissions
is
worth
considering.

15.
EPA
should
not
require
site­
specific
risk
assessments
(
SSRA)
as
a
part
of
the
permanent
replacement
standards
rulemaking.

What
is
the
history
of
SSRAs?

 
EPA's
Hazardous
Waste
Minimization
and
Combustion
Strategy
(
1994)
recommended
that
site­
specific
risk
assessments
(
SSRA)
become
a
part
of
the
RCRA
permitting
process
for
hazardous
waste
combustors
where
necessary
to
protect
human
health
and
the
environment.
 
In
1998,
EPA
released
draft
guidance
on
conducting
SSRAs.
 
The
preamble
of
the
1999
HWC
MACT
rule
(
see
the
discussion
beginning
at
64
FR
52840)
recommended
that
for
facilities
subject
to
the
new
standards,
permit
writers
evaluate
the
need
for
a
SSRA
on
a
caseby
case
basis.
EPA
went
on
to
state
that
"
SSRAs
are
not
anticipated
to
be
needed
for
every
facility,
but
should
be
conducted
for
facilities
where
there
is
reason
to
believe
that
operations
in
accordance
with
the
MACT
standards
alone
may
not
be
protective
of
human
health
and
the
environment."
 
These
instructions
left
considerable
room
for
interpretation
by
the
states
and
the
regions.
Some
states
and
regions
were
using
the
policy
statements
made
in
1994
as
the
justification
for
requiring
SSRAs.
 
CKRC
petitioned
EPA
to
withdraw
the
SSRA
guidance,
suggesting
that
it
was
a
rule
in
the
form
of
guidance
and
suggesting
that
if
EPA
believes
that
SSRAs
are
necessary,
they
should
initiate
the
rulemaking
process
to
make
them
mandatory.
 
In
partial
response
to
the
CKRC
petition,
a
memo
from
the
Office
of
Solid
Waste
and
Emergency
Response
Assistant
Administrator
Marianne
Horinko
was
sent
to
the
Regions
on
April
10,
2003
(
copy
attached).
This
memo
made
it
clear
that
the
only
authority
to
require
a
SSRA
is
the
RCRA
omnibus
authority
which
requires
a
fully
documented
finding
in
the
administrative
record.
 
In
the
2004
Federal
Register
notice
rule,
EPA
proposed
to
add
two
new
paragraphs
(
270.10(
l)
and
270.32(
b)(
3))
in
response
to
the
Cement
Kiln
Recycling
Coalition's
petition.
 
EPA
has
stated
at
public
meetings
that
these
two
paragraphs
do
not
make
any
changes
in
their
authority
to
require
site­
specific
risk
assessments
and
has
reiterated
that
these
provisions
do
not
change
current
EPA
policy.
OMB
paper
­
15
­
8/
26/
05
­
15
­
What
do
we
want
OMB
to
do?

 
Since,
in
the
Agency's
opinion,
these
two
paragraphs
do
not
add
to
their
authority
or
make
any
changes
in
current
policy,
we
see
no
real
reason
to
keep
them
in
the
final
rule.
OMB
should
suggest
that
EPA
remove
these
two
paragraphs
from
the
final
rule.
 
If
the
Agency
believes
that
additional
explicit
language
must
be
added
to
make
it
clear
that
permit
writers
have
the
authority
to
require
site­
specific
risk
assessments,
then
OMB
should
request
that
the
following
language
is
necessary
to
define
how
this
process
will
proceed.

"
The
factual
and
technical
basis
for
any
decision
under
this
paragraph
should
be
included
in
the
administrative
record
for
the
facility
according
to
the
requirements
in
40
CFR
124."

 
This
change
is
required
to
be
consistent
with
the
Horinko
memo,
dated
April
10,
2003,
which
defines
current
EPA
policy
on
when
and
how
to
require
risk
assessments.
OMB
paper
­
16
­
8/
26/
05
­
16
­
SLIDE
PRESENTATION
SLIDE
#
_
1_

The
Hazardous
Waste
Combustion
(
HWC)
MACT
Rule
Continuing
Concerns
of
the
Coalition
for
Responsible
Waste
Incineration
August
26,
2005
SLIDE
#
2__

Who
We
Are

The
Coalition
for
Responsible
Waste
Incineration
(
CRWI)
is
a
trade
organization
representing
virtually
all
facets
of
the
Hazardous
Waste
Combustion
(
HWC)
source
category
affected
by
this
proposed
MACT
Rule.


CRWI
members
include
companies
engaged
in
commercial
and
captive
hazardous
waste
incinerators,
cement
kilns,
halogen
acid
recovery
furnaces,
and
chemical
weapon
disposal
facilities,
as
well
as
consulting
experts
and
academic
programs
supporting
hazardous
waste
combustion
as
a
responsible
waste
disposal
alternative.


The
membership
of
CRWI
is
currently
implementing
the
interim
MACT
standards
and
will
be
impacted
by
this
proposed
rule.

SLIDE
#
3__

Why
We
Are
Here

The
HWC
MACT
source
category
includes
diverse
industrial
activities
affecting
CRWI
members,
all
addressed
in
previous
dialogue
with
EPA.


EPA's
currently
proposed
HWC
standards
leave
some
of
these
issues
unresolved,
particularly
as
they
will
impact
new
sources.


The
Comparison
Table
attached
illustrates
the
history
of
the
HWC
MACT
standards
and
how
EPA's
proposed
regulatory
approach
to
this
Rule
will
impact
the
MACT
emission
standards
for
the
subcategory
of
hazardous
waste
incinerators.
OMB
paper
­
17
­
8/
26/
05
­
17
­
SLIDE
#
_
4_

Comparison
Tables
Original
MACT
Standards
Interim
MACT
Standards
Proposed
MACT
Standards
HAPs
Existing
New
Existing
New
Existing
New
Dioxins/
Furans
(
expressed
in
TEQ)
0.20
ng/
dscm
or
0.40
ng/
dscm
0.20
ng/
dscm
0.20
ng/
dscm
or
0.40
ng/
dscm
0.20
ng/
dscm
0.28
ng/
dscm
or
0.40
ng/
dscm
0.11
ng/
dscm
or
0.2
ng/
dscm
Particulate
Matter
34
mg/
dscm
34
mg/
dscm
34
mg/
dscm
34
mg/
dscm
0.015
gr/
dscf
0.00070
gr/
dscf
Semivolatile
Metals
240
µ
g/
dscm
24
µ
g/
dscm
240
µ
g/
dscm
120
µ
g/
dscm
59
µ
g/
dscm
6.5
µ
g/
dscm
Low
Volatile
Metals
97
µ
g/
Dscm
97
µ
g/
Dscm
97
µ
g/
dscm
97
µ
g/
dscm
84
µ
g/
dscm
8.9
µ
g/
dscm
OMB
paper
­
18
­
8/
26/
05
­
18
­
SLIDE
#
5__

Comparison
Tables
(
cont.)

Original
MACT
Standards
Interim
MACT
Standards
Proposed
MACT
Standards
HAPs
Existing
New
Existing
New
Existing
Mercury
130
µ
g/
dscm
45
µ
g/
dscm
130
µ
g/
dsc
m
45
µ
g/
dsc
m
130
µ
g/
dsc
m
HCl/
Cl
77
ppmv
21
ppmv
77
ppmv
21
ppmv
1.5
ppmv
Hydrocarbons
10
ppmv
or
100
ppmv
CO
10
ppmv
or
100
ppmv
CO
10
ppmv
or
100
ppmv
CO
10
ppmv
or
100
ppmv
CO
10
ppmv
or
100
ppmv
CO
99.99%
99.99%
DRE
99.99%
99.99%
99.99%
OMB
paper
­
19
­
8/
26/
05
­
19
­
SLIDE
#
6__

Concurrence:
Methods
for
Deriving
Standards

There
is
no
one­
size­
fits­
all
methodology
for
judging
the
performance
of
the
HWC
source
category.
Individual
categories
of
HAPs
­­
dioxins
and
furans
(
D/
F),
Semi­
volatile
(
SVM)
and
Low­
volatile
Metals
(
LVM),
particulate
(
PM),
Total
Chlorine
and
Hydrochloric
Acid
(
CL/
HCl))
 
require
different
control
technologies.


CRWI
concurs
in
EPA's
chosen
methodologies
for
setting
technology­
based
MACT
standards.

SLIDE
#
_
7_

Concurrence:
The
Maximum
Deviation
Approach

The
database
used
to
set
the
HWC
MACT
standards
includes
many
instances
of
"
non­
detect"
of
a
particular
pollutant.


EPA
acknowledges
that
using
an
analytical
detection
limit
in
lieu
of
actual
emission
data
tends
to
bias
the
statistical
variability
values
used
to
set
MACT
standards.


The
Maximum
Deviation
concept
is
a
reasonable
approach
for
more
accurately
estimating
variability.

SLIDE
#
8__
A
Caveat
to
The
Maximum
Deviation
Approach

CRWI
has
not
been
able
to
test
this
approach
on
the
actual
data
being
used.


CRWI
continues
to
have
reservations
about
using
"
non­
detect"
data
points
for
setting
emission
standards.
If
non­
detect
data
are
used,
the
Agency
should
follow
their
own
guidelines
and
use
the
Reliable
Detection
Limit
(
RDL)
rather
than
the
Method
Detection
Limit
(
MDL).

SLIDE
#
9__
Concurrence:
Flexibility
for
HWC
Subcategories

CRWI
supports
the
options
for
setting
standards
for
liquid
fuel­
fired
boilers.


CRWI
supports
using
chlorine
as
a
surrogate
for
metals
in
the
halogen
acid
furnace
subcategory.


CRWI
supports
EPA's
decision
to
not
develop
beyond­
the­
floor
chlorine
standards
for
solid
fuel­
fired
boilers.
OMB
paper
­
20
­
8/
26/
05
­
20
­
SLIDE
#
10__

Concurrence:
The
Health­
Based
Chlorine/
HCl
Standard
Alternative

Section
112(
d)(
4)
authorizes
EPA
to
set
HAP
emission
standards
based
on
established
health
thresholds.


HCl
is
a
threshold
pollutant
with
an
established
reference
dose
that
incorporates
a
conservative
uncertainty
factor,
satisfying
the
"
ample
margin
of
safety"
requirement
of
the
CAA.


CRWI
supports
EPA's
proposed
Chlorine/
HCl
standard
as
reasonable
and
consistent
with
the
CAA,
with
one
exception.

SLIDE
#
11__

Concern:
Implementation
of
the
Health­
Based
Alternative
Standard

As
proposed,
the
permitting
agency
is
given
6
months
to
review
an
eligibility
determination.
Such
eligibility
applications
are
necessarily
time­
sensitive
because
of
imminent
MACT
compliance
dates.
If
approval
is
not
timely,
the
facilities
are
obliged
to
implement
costly
technology­
based
controls
of
little
discernible
benefit
to
public
health
or
the
environment.


Six
months
should
be
enough
time
for
an
agency
to
determine
whether
an
eligibility
application
is
submitted
in
good
faith.
At
that
time,
the
burden
should
shift
to
the
agency
for
a
prompt,
decisive,
and
reasonable
action
on
the
eligibility
application.

SLIDE
#
12__

Implementation
of
the
Health­
Based
Alternative
Standard:
Proposed
Resolution

If
EPA's
final
HWC
MACT
Rule
does
not
require
prior
agency
approval
of
an
eligibility
demonstration
before
of
a
health­
based
alternative
standard
can
be
implemented,
no
action
by
OMB
is
necessary.


If
the
final
Rule
requires
agency
approval
before
health­
based
MACT
standards
can
be
implemented,
CRWI
suggests
the
following:


Provisions
for
a
compliance
date
extension;
or

Default
approval
after
agency
review
for
six
months
with
no
action.
OMB
paper
­
21
­
8/
26/
05
­
21
­
SLIDE
#
13
__

Concern:
Achievability
of
New
Source
Standards

Section
112(
d)(
3)
of
the
CAA
states
that
new
source
MACT
standards
should
be
set
based
on
"
the
emission
control
that
is
achieved
in
practice
by
the
best
controlled
similar
source,
as
determined
by
the
Administrator"
(
emphasis
added).


Note,
in
the
Table
provided,
how
many
of
the
new
source
standards
are
much
more
stringent
than
existing
source
standards.
This
stems
from
the
HAP­
by­
HAP
approach
used
by
EPA
in
setting
the
new
source
standards.

SLIDE
#
_
14_

Concern:
Achievability
of
New
Source
Standards

The
HAP­
by­
HAP
approach
works
for
existing
sources
in
the
HWC
source
category
because
the
top
performers
are
averaged.
In
setting
the
new
source
standards,
EPA
chooses
the
top
performer
for
each
individual
HAP
as
the
standard
setter;
in
the
HWC
source
category,
however,
no
one
facility
is
the
top
performer
for
each
HAP.


Emission
performance
of
existing
facilities:
Of
the
six
HWC
MACT
HAPs,
the
best
any
facility
can
demonstrate
is
compliance
with
four
new
source
standards
(
see
provided
spreadsheet).

SLIDE
#
15_

Concern:
Achievability
of
New
Source
Standards

Congress
expressly
addressed
this
issue
in
a
colloquy
between
Senators
Dole
and
Durenberger
during
the
Senate
Conference
Report
on
the
1990
CAA
Amendments.
Senator
Durenberger
stated
that
if
it
is
technically
infeasible
to
combine
incompatible
technologies
for
HAP
controls,
"
EPA
should
judge
MACT
to
be
the
technology
which
best
benefits
human
health
and
the
environment
on
the
whole."


CRWI
respectfully
submits
that
EPA
has
skipped
this
step
in
setting
new
source
MACT
standards
for
the
HWC
source
category.
OMB
paper
­
22
­
8/
26/
05
­
22
­
SLIDE
#
_
16_

Achievability
of
New
Source
Standards:
Proposed
Resolution

In
order
to
meet
the
statutory
mandate
of
the
CAA,
at
least
one
existing
HWC
facility
must
be
able
to
meet
EPA's
proposed
new
source
standards.
To
assure
that
this
mandate
is
honored,
CRWI
suggests
OMB:


Require
EPA
to
show
that
at
least
one
existing
facility
can
meet
all
new
source
standards
simultaneously.


Further
require
EPA
to
assure
that
the
benchmark
"
top
performer"
is
comparable
to
the
new
HWC
facility
(
see
next
concern
below).

SLIDE
#
17
Concern:
Comparability
of
"
Top
Performers"

 
The
accompanying
CRWI
Position
Paper
identifies
the"
top
performers"
for
five
of
the
six
HAPs
regulated
by
the
MACT
Rule
(
Position
Paper
at
page
8).

 
None
of
the
five
"
top
performers"
are
representative
of
the
typical
hazardous
waste
combustors,
based
on
differences
in
physical
configuration,
feed
rate,
or
pre­
MACT
performance.

SLIDE
#
_
18
Comparability
of
"
Top
Performers:"
Proposed
Resolution

Ask
EPA
to
clarify
their
techniques
for
setting
new
source
standards
to
include
some
means
to
assure
that
the
benchmark
existing
facility
is
actually
similar
as
required
by
Section
112(
d)(
3)
of
the
CAA.

SLIDE
#
19
Concern:
"
MACT
of
MACT"


"
MACT
of
MACT"
is
shorthand
for
a
concern
arising
from
the
legal
history
of
this
particular
MACT
Rule.


The
original
HWC
MACT
Rule
was
proposed
in
1996,
promulgated
in
1999,
and
judicially
vacated
in
2001.
Interim
Standards
were
put
into
place
in
2002.


In
2002,
EPA
reopened
the
"
MACT
Pool"
as
a
prelude
to
setting
new
replacement
standards.
This
updated
"
Pool"
reflects
the
following
new
performance
data:


Facilities
that
had
ceased
combustion
of
hazardous
waste
were
removed
(
which
CRWI
supported);
and

Facilities
that
had
upgraded
to
meet
the
Interim
Standards
were
included
(
which
CRWI
strenuously
opposed).
OMB
paper
­
23
­
8/
26/
05
­
23
­
SLIDE
#
20_
Concern:
"
MACT
of
MACT"


CAA
Section
112(
e)
of
the
CAA
clearly
expresses
Congressional
intent
to
set
technology­
based
performance
standards
by
a
date
certain.


As
a
result
of
the
EPA's
decision
to
re­
open
the
HWC
MACT
Pool
in
2002,
a
proactive
facility
modifying
its
technology
to
meet
the
Interim
Standards
became
a
HWC
"
top
performer,"
ratcheting
existing
and
new
source
HWC
MACT
standards
downward.


This
more
stringent
standard
was
derived,
not
from
the
state­
of­
the­
art
technology
at
the
time
Congress
intended,
but
rather
a
facility's
good
faith
effort
to
comply
with
post­
MACT
standards
(
i.
e.,
"
MACT
of
MACT").

SLIDE
#
_
21_

"
MACT
of
MACT:"
Proposed
Resolution

Suggest
to
EPA
that
any
facility
that
has
upgraded
to
meet
MACT
standards
should
not
be
included
in
the
MACT
Pool
for
setting
technology
standards.

SLIDE
#
22_

Concern:
Data
Quality
Issues

As
emission
standards
become
more
stringent,
data
quality
becomes
suspect.
For
instance,
the
PM
standards
may
be
set
at
lower
than
acknowledge
accuracy
levels.


Also,
it
will
be
more
difficult
for
facilities
to
demonstrate
compliance.
The
proposed
standard
for
chlorine
is
set
lower
than
the
confidence
level
of
established
test
methods.

SLIDE
#
23__

Data
Quality
Issues:
Proposed
Resolution

EPA
should
re­
analyze
the
MACT
database
to
assure
that
data
quality
guidelines
are
followed.


EPA
should
assure
that
compliance
testing
methods
that
yield
consistent
and
accurate
data
are
available.
OMB
paper
­
24
­
8/
26/
05
­
24
­
SLIDE
#
24
__

Concern:
Beyond­
the­
Floor
Standards

EPA
has
proposed
a
beyond­
the­
floor
standard
for
dioxins/
furans
for
the
halogen
acid
furnace
subcategory.
This
will
result
in
added
compliance
costs
to
achieve
a
reduction
of
only
1.3
grams
per
year
of
dioxin/
furan
emissions.


Proposed
Resolution:


Ask
EPA
to
re­
examine
the
cost­
effectiveness
of
this
beyond­
the­
floor
decision.

SLIDE
#
25__

Concern:
Site
Specific
Risk
Assessments
(
SSRAs)


In
response
to
a
petition
by
the
Cement
Kiln
Recycling
Coalition,
EPA
is
proposing
amendments
to
RCRA
regulations
clarifying
its
authority
to
require
SSRAs.


CRWI
does
not
believe
that
risk
assessment
guidance
should
be
promulgated
as
a
rule.


CRWI
does
believe
that
any
clarification
of
SSRA
authority
should
also
include
strictures
on
when
risk
assessments
should
be
performed.

SLIDE
#
26__

SSRAs:
Proposed
Resolution

Either
delete
the
amendments
regarding
SSRA
authority;
or

Add
specific
language
making
the
amendments
consistent
with
the
April
10,
2003,
policy
memo
from
Marianne
Horinko.
