Subject:
Extension
of
Alternative
Compliance
Periods
Under
the
Anti­
Dumping
Program:
Seasonal
NOx
Emissions
Determination
From:
Christine
Brunner,
Chemical
Engineer
Assessment
and
Standards
Division
To:
Docket
A­
2002­
14
Background
This
rule,
Extension
of
Alternative
Compliance
Periods
Under
the
Anti­
Dumping
Program,
would
extend
the
amount
of
time
available
to
an
approved
refinery
to
comply
with
its
anti­
dumping
NOx
requirement.
Under
the
original
rule
wherein
a
refiner
could
petition
and
be
approved
for
an
Alternative
Anti­
Dumping
Compliance
Period,
EPA
required
a
refiner
to
purchase
NOx
credits
when,
on
a
quarterly
basis,
its
average
NOx
emissions
exceeded
the
annual
statutory
NOx
value.
As
stated
in
the
September
8,
2000,
rule,
"...
credits
function
as
collateral
against
any
NOx
deficiency
that
the
refiner
creates
to
minimize
the
possibility
of
environmental
harm
in
the
event
the
refinery
does
not
fulfill
its
obligation
under
the
other
requirements
of
the
rule."

Direct
Final
Rule/
Proposed
Rule
("
today's
action")

EPA
has
determined
that
the
current
method
of
calculating
NOx
emissions
deficits
and
benefits
at
80.101(
k)(
3)(
ii)
may
have
the
unintended
effect
of
creating
an
unnecessary
pattern
of
buying
and
selling
of
credits
by
an
approved
refiner
because
of
the
combined
effects
of
seasonal
gasoline
quality
and
the
determination
of
emissions
in
the
Complex
Model.
EPA
believes
that
the
same
environmental
outcome
will
be
achieved
by
the
changes
contained
in
today's
action
while
not
resulting
in
the
unnecessary
outlay
of
capital
by
the
refiner
at
a
time
when
the
refiner
is
operating
under
an
alternative
compliance
period
specifically
to
reduce
the
impact
of
financial
constraints.

On
a
quarterly
basis,
gasoline
production
during
the
first
and
fourth
quarters
of
the
calendar
year
is
almost
all,
if
not
entirely,
winter
gasoline
for
most
refiners.
Winter
gasoline
is
not
only
gasoline
with
different
properties
(
like
higher
RVP)
than
summer
gasoline,
but
it
must
also
be
evaluated
using
the
winter
Complex
Model.
For
a
given
fuel
composition
(
based
on
fuel
properties
used
in
the
Complex
Model),
the
winter
Complex
Model
yields
higher
emissions
than
the
summer
Complex
Model.
Winter
emissions
determined
using
the
Complex
Model
will
always
be
higher
than
annual
emissions
determined
by
combining
winter
and
summer
results
from
the
Complex
1The
annual
statutory
baseline
value
for
NOx
emissions
is
1461
milligrams
per
mile
(
mg/
mile).
[
See
80.91(
c)(
5)(
iv).]
The
summer
and
winter
statutory
baseline
values
for
NOx
are
1340
and
1540
mg/
mile,
respectively.
These
were
determined
by
inputting
the
summer
and
winter
statutory
baseline
fuel
parameters
into
the
summer
or
winter
Complex
Model,
respectively.
[
See
80.45(
b).]
The
annual
statutory
NOx
baseline
emissions
value
was
determined
from
the
weighting
of
the
summer
and
winter
baseline
emissions
using
a
0.396
fraction
of
summer
and
a
0.604
fraction
of
winter
gasoline.

2
While
we
believe
that
this
approach
is
appropriate
for
the
limited
purpose
of
calculated
NOx
credit
purchase
obligations,
as
described
herein,
we
do
not
believe
that
such
an
approach
would
be
appropriate
for
purposes
of
demonstrating
compliance
with
an
annual
performance
requirement.
Model1.
Thus,
under
the
current
equation
for
calculating
NOx
deficit
and
benefit
under
80.101(
k)(
3),
even
a
refiner
that
is
producing
winter
gasoline
that
would
assure
annual
compliance
(
once
averaged
with
its
summer
gasoline
production)
would
in
many
instances
have
to
purchase
credits
for
the
first
and
fourth
quarters.
A
refiner
could
produce
extremely
clean
winter
gasoline
in
these
quarters
in
order
to
not
have
to
purchase
credits
under
80.101,
but
it
is
unlikely
that
a
refiner
in
start­
up
mode
would
be
able
to
produce
such
clean
gasoline.

Conversely,
the
gasoline
that
a
refiner
produces
in
the
second
and
third
quarters
(
which
includes
relatively
little
winter
gasoline
 
roughly
20­
30
percent)
would
perform
significantly
better
than
the
annual
statutory
NOx
value.
In
fact,
under
the
current
80.101(
k)(
3)(
ii),
this
overcompliance
in
the
second
and
third
quarters
(
resulting
in
quarterly
NOx
benefits)
would
likely
more
than
exceed
the
deficit
created
in
the
first
and
fourth
quarters,
allowing
those
credits
to
be
sold.

Thus,
as
a
result
of
the
combined
effects
of
seasonal
gasoline
quality
and
the
provisions
of
the
Complex
Model,
the
current
manner
in
which
NOx
benefit
and
deficit
are
determined
might
unnecessarily
require
approved
refiners
to
buy
and
sell
NOx
credits
each
year.
This
would
create
an
economic
burden
without
any
related
environmental
benefits.
Accordingly,
we
believe
that
the
NOx
credit
provisions
of
80.101(
k)(
3)
should
reflect
the
fact
that
the
NOx
performance
of
gasoline
produced
during
different
times
of
the
year
is
expected
to
vary
relative
to
the
annual
average
statutory
baseline.
In
quarters
with
a
greater
percentage
of
winter
gasoline
production
NOx
performance
may
be
less
stringent
than
the
annual
average,
and
in
quarters
with
little
winter
gasoline
production
NOx
performance
should
be
better
than
the
annual
average.
2
The
quarterly
NOx
emissions
values
contained
in
this
action
reflect
these
seasonal
differences,
e.
g.,
the
second
and
third
quarter
NOx
values
are
more
stringent,
and
the
first
and
fourth
quarter
less
stringent,
than
the
annual
average.

This
action
would
modify
the
equations
at
80.101(
k)(
3)
used
to
calculate
the
quarterly
NOx
deficit
or
benefit,
but
does
not
change
the
environmental
protection
aspects
intended
under
80.101(
k).
In
fact,
today's
action
may
result
in
the
continued
and
accelerated
start­
up
of
an
approved
refinery,
enabling
the
refinery
to
not
require
the
full
use
of
its
alternative
averaging
period
and
getting
cleaner
gasoline
to
the
market
sooner.
Specifically,
for
the
limited
application
of
determining
an
approved
refiner's
NOx
deficit
or
benefit
in
a
quarter,
today's
action
requires
that
the
refiner's
quarterly
NOx
emissions
will
now
be
compared
to
a
quarterly
NOx
emissions
3
U.
S.
Department
of
Energy,
Energy
Information
Administration
value
rather
than
to
the
annual
statutory
baseline
NOx
emissions
value.
The
net
determination
of
NOx
deficit
or
benefit
is
unaffected.

Equation
Modification
The
quarterly
statutory
NOx
values
were
determined
as
follows.
First,
the
number
of
summer
and
winter
days
of
gasoline
production
was
determined
for
each
quarter.
In
the
first
quarter,
January
through
March,
there
are
a
total
of
90
days
(
excluding
leap
years).
All
of
these
days
are
winter
days.
In
the
second
quarter,
April
through
June,
there
are
91
days.
Because
summer
gasoline
must
be
at
terminals
on
May
1
and
at
the
pump
on
June
1,
both
of
those
months
are
considered
summer
months.
Because
of
these
deadlines,
refiners
generally
will
start
to
produce
summer
gasoline
some
time
in
April.
To
account
for
this,
approximately
10
days
in
April
were
allocated
as
summer,
leaving
20
winter
days.
The
approximation
of
10
days
of
summer
gasoline
in
April
will
be
discussed
below.
In
the
third
quarter,
July
and
August
are
considered
summer
months.
Although
the
high
ozone,
or
summer,
season,
does
not
end
until
after
September
15,
refiners
will
generally
begin
reducing
production
of
summer
gasoline
during
early
September.
To
account
for
this,
only
approximately
10
days
in
September
were
allocated
as
summer
days,
leaving
about
20
days
as
winter
gasoline.
As
for
April,
this
allocation
will
be
discussed
further
below.
There
are
92
days
in
the
fourth
quarter,
and
all
gasoline
produced
in
that
quarter
is
winter
gasoline.
See
Attachment
1.

Part
of
the
consideration
in
approximating
10
days
of
summer
gasoline
in
both
April
and
September,
in
addition
to
the
reality
of
early
production
and
early
stoppage
of
production
of
summer
gasoline
in
these
months,
respectively,
was
maintaining
the
statutory
summer
and
winter
fractions,
0.396
and
0.604,
respectively.
These
fractions,
along
with
seasonal
statutory
emissions
values,
determine
the
annual
statutory
emissions
value,
such
as
the
Phase
II
NOx
standard
for
conventional
gasoline.

Using
weekly
average
finished
gasoline
production
rates
from
1990
(
see
Attachment
2)
3,
average
quarterly,
seasonal
production
rates
were
determined.
Thus,
in
the
first
and
fourth
quarters,
a
single
average
production
rate
was
determined;
in
the
second
and
third
quarters,
part
of
the
data
was
allocated
as
summer,
part
as
winter,
based
on
a
pre­
estimation
of
the
starting
and
stopping
production
dates
for
summer
gasoline.
The
average
seasonal
volumes
were
determined
by
multiplying
the
number
of
days
of
seasonal
gasoline
by
the
seasonal
gasoline
production
rate.
NOx
quarterly
emissions
were
determined
by
multiplying
the
seasonal
volumes
of
each
quarter
by
the
corresponding
seasonal
statutory
NOx
value
and
dividing
the
result
by
the
total
volume
for
the
quarter.
The
number
of
summer
and
winter
days
in
the
second
and
third
quarters
were
iterated
until
1)
the
statutory
seasonal
ratios
were
achieved;
2)
the
average
of
the
quarterly
NOx
emissions
determined
from
gasoline
production
equaled
the
annual
statutory
NOx
emissions
value
of
1461
mg/
mile;
and
3)
the
quarterly
NOx
values
determined
from
gasoline
production
and
solely
from
calendar
days
were
about
the
same.

The
quarterly
NOx
values
in
the
first,
third
and
fourth
quarters
were
identical
using
either
method;
the
second
quarter
NOx
value
determined
using
gasoline
production
was
slightly
more
stringent
than
that
determined
using
calendar
days
alone.
To
further
ensure
the
environmental
protection
aspects
of
this
action,
we
are
using
the
slightly
more
stringent
value
as
the
second
quarter
comparison.
