►
From YouTube: CESM Workshop: Fires in CESM Cross Working Group
Description
The 26th Annual CESM Workshop will be a virtual workshop with a modified schedule on its already scheduled date. Specifically, the virtual Workshop will begin with a full-day schedule on 14 June 2021 with presentations on the state of the CESM; by the award recipients; and three invited speakers in the morning, followed by order 15-minute highlight and progress presentations from each of the CESM Working Groups (WG) in the afternoon.
On 15-17 June 2021, working groups and cross working groups have half-day sessions, some with presentations and some that are discussion only.
A
A
E
B
A
Okay,
I'm
gonna
go
ahead
and
get
started
with
a
brief
introduction.
I
know
that
we'll
have
more
people
filtering
in
my
name
is
jackie
schumann.
This
is
the
fire
cross
working
group
session,
and
so
oh
geez
there
we
go.
A
It's
good
that
I
figured
this
out
now.
So
we'll
have
six
talks,
five
minutes
each
followed
by
two
minutes
for
questions,
and
then
we
have
two
posters
that
will
give
a
little
preview
talk
and
then
we
will
have
a
discussion
at
the
end,
and
so
oh
my
gosh.
This
is
going
to
drive
me
crazy.
Okay,
so
we've
had
historic
fires
globally.
I
just
want
to
put
us
back
into
context
depending
on
where
you're
coming
from
atmosphere
land.
You
know
surface,
you
think
about
a
different
thing.
A
A
These
are
fires
that
overwinter
and
then
once
the
fuel
is
dry
enough,
they'll
re-emerge
with
burning,
but
there
was
a
recent
paper
about
how
to
detect
this
from
remote
sensing
and
it
seems
to
be
it's
been
around
for
a
while,
but
it's
a
larger
problem
than
previously
was
thought
and
then
here
in
the
united
states
you
know,
drier
fuels
are
connected
to
larger
fires,
and
so
the
figure
on
the
left
is
a
study
that
was
published
in
2016
and
for
the
recent
period
2000
to
2015.
A
You
can
see
the
fuels
are
drier
and
they
have
a
much
larger
burned
area.
But
if
you
were
to
do
this
research
again,
the
trend
continues
within
california.
Here
in
colorado,
we've
been
breaking
records
in
terms
of
area
burned
and
then
just
because
you
don't
have
a
fire
occurring
where
you
are
locally,
you
still
experience
those
events.
This
is
a
picture
I
took
from
my
backyard,
but-
and
this
was
probably
a
local
fire,
colorado
fire,
but
we
have
fires
from
you
know
california,
that
travel
here
to
colorado
the
australian
smoke
traveled
much
further
than
australia.
A
So
it's
it's
a
big.
It's
a
big
issue
that
we
need
to
think
about
on
a
lot
of
scales,
but
in
good
news
governments
are
taking
action,
joe
biden
just
released
his
20
22
budget
and
it
makes
significant
investments
to
wildfire
risk
and
to
address
wildfire
risk
and
climate
change,
80
million
dollars
to
mitigation,
20
million
dollars
to
rehabilitation
of
lands,
and
then
five
million
dollars
to
the
joint
fire
science
program.
A
E
Okay,
great
so
I'm
actually
going
to
present
on
behalf
of
fong
li,
who
is
the
primary
author
of
the
fire
code
that
exists
in
clm5,
and
she
gave
me
most
of
these
slides
to
present.
And
so
you
know,
as
always,
there's
lots
of
people
who
are
involved
in
this
project.
But
I've
been
the
main
facilitator
over
the
last
few
years
of
helping
fong
get
the
fire
code
into
clm.
C
E
F
E
There
are
deforestation
fires
which
are
fires
that,
where,
especially
in
the
tropics,
where
close
chocolate
forest
where
people
are,
are
using
fire
to
do
deforestation,
and
then
there
can
be
fire
that
sort
of
moves
beyond
the
area
of
the
deforestation,
so
they're
called
escape
fires,
and
then
there's
also
peat
fires
in
places
like
indonesia,
where
those
can
have
big
emissions
as
well
so
and
each
one
of
these
fires
has
a
different
kind
of
way.
It's
ignited
and
controlled.
E
If
you
look
at
crop
fires
that
sort
of
timing
of
of
what's
going
on
with
the
crop
cycle,
so
where
people
are
maybe
clearing
land
you
know
to
before
they
before
they
plan
to
do
planting
it's
dependent
on
socioeconomic
factors
like
gdp
and
the
bot,
and
that
and
the
population
density
will
affect
agricultural
fires
in
the
natural
fires
region.
Fires
are
ignited
by
lightning
or
humans
and
then
suppressed
by
by
humans
and
that's
controlled
again
by
population
density
or
gdp.
E
Whether
or
not
a
fire
can
get
ignited,
there
needs
to
be
vegetation
there
and
then
also
the
weather
and
climate
conditions
need
to
be
suitable
and
in
in
clm5.
That's
it's.
The
air
humidity
and
the
soil.
Moisture
which
are
the
are
the
triggers
deforestation.
Fires
are
triggered
by
a
land,
use,
change
event
in
the
model,
and
then
peat
fires
are
triggered
by
weather
and
climate
conditions
that
are
suitable
for
peat
fires.
So
once.
E
Fire
and
I'll
talk
about
very
briefly
once
you
know
the
area,
then
you
can
calculate
the
carbon
consequences
and
the
fire
carbon
emissions
from
that
area
has
been
burned.
Based
on
the
vegetation
carbon
distributions
in
that
grid
cell,
there
are
emissions
factors
associated
with
each
plant,
functional
type.
That
will
then
also
determine
other
trace
gases.
E
E
So
if
you
do,
you
know,
determine
fire
currents.
You
use
this
equation
here
so
that
every
time
step,
you
would
calculate
this
and
find
out.
The
number
of
fire
counts
within
a
grid
cell,
which
is
based
on
the
emission
counts,
which
is
ni,
which
is
lightning
plus
human
ignitions.
E
We
take
lightning
from
from
offline
database,
climatology
fuel
availability.
There
needs
to
be
carbon
there
to
burn
fuel
combustibility,
that's
the
climate
factor
and
then
there's
this
non-suppression
rate.
So
the
extent
that
humans
are
not
suppressing
fire,
so
that
number
would
be
one
if
there's
no
suppression
zero,
if
there's
strong
suppression-
and
that
tends
to
have
this
sort
of
distribution
where
it
peaks,
you
know
sort
of
there's
very
little
suppression
of
very
low
densities.
E
It
comes
up
at
higher
population
densities
and
then
comes
to
very
low
fire
occurrence
at
high
densities.
When
gdp
is
high
enough,
so
people
can
can
stop
fire
and
then
the
climate
consequences
here
or
climate
factors
fm
relative,
humidity
assault
when
it's
both
have
these
curves
that
determine
the
likelihood
that
an
initiation
will
occur
next
slide.
E
In
a
grid
cell,
then
you
need
to
check
out
the
area.
I
won't
go
through
the
details
in
basically
the
up.
There
is
a
key
value
in
the
middle
of
a
slide
there,
which
is
the
fuel
wetness
times
the
wind
speed
and
those
are
the
two
factors
which
will
control
or
determine
how
how
much
the
fire
was
spread
and
the
average
then
area
that
you
calculate
from
that
is
and
then
down
regulated
by
the
by
some
suppression,
efforts
by
population
or
or
gdp
next
slide.
E
I
won't
go
into
this
one
in
the
interest
of
time,
but
there's
different
controlling
factors
for
agricultural
fires,
deforestation,
degradation,
fires
and
peat
fires
next
slide,
and
so
just
you
know
how
well
does
the
model
do?
I
realize
I'm
at
five
minutes
comparison
g
fed
g
fed
four
or
g
five
forest,
which
includes
more
small
fires.
You
know
it's
in
the
right
ballpark.
This
is
the
clm45.
Cm5
is
even
a
little
bit
better
and
csm2
is
roughly
equivalent
to
clm5
so
compared
to
other
global
fire
models.
E
I
think
the
clm5
fire
model
is
comparing
okay,
you
can
certainly
find
deficiencies
and,
if
you,
you
know,
went
into
a
specific
region,
I'm
co-authoring
a
paper
right
now,
that's
looking
at
the
western
u.s
and
we're
overestimating
fire
in
the
western
west
by
almost
a
factor
of
two.
But
you
know
globally.
The
map
looks
about
right
next
slide
and
this.
E
A
Thanks
dave
will,
are
there?
Is
there
time
for
questions.
C
E
Yeah,
it's
kim
my
understanding
is
that
cam
can
calculate.
E
Although
it's
not
a
regular
part
of
a
cam
calculation,
and
so
currently
we're
prescribing
it
from
an
offline
file,
but
that
was
one
of
our
discussion
topics
later
would
be
a
nice
way
to
connect
cam
more
to
clm.
Is
we
actually
took
the
prognostic
lightning
from
cam
to
the
extent
that
they
can
do
that?
So
that
is
a
hopeful
future
endeavor,
no
matter
which
way
we
go
in
terms
of
developing
the
fire
model.
A
Okay,
thank
you
so
I'll
be
talking
about
opportunities,
additional
opportunities
for
fire
research
when
we
include
dynamic
fuels
and
size,
structured,
vegetation.
A
So
fates
is
the
functionally
assembled
terrestrial
ecosystem.
Simulator,
it's
a
module
that
runs
within
ctsm.
It
replaces
the
traditional
big
leaf
vegetation
shown
on
the
top
left
with
more
realistic
size,
structured
vegetation.
So
by
doing
this
we
can
capture
dynamic
vegetation,
competition,
ecosystem
assembly
and
vegetation
distribution.
Through
time
on
a
heterogeneous,
we
can
have
grid
cell
heterogeneity
with
different
patches
and
so
spitfire
uses
the
same
rothermell
fire
behavior
equations
that
fong
uses
and
it
brings
in
the
ignition
data
set
as
well
as
weather.
A
The
the
fire
weather
is
evaluated
in
conjunction
with
the
fuel
state
and
what
how
much
of
that
fuel
will
combust
from
that?
We
calculate
fire,
behavior,
total
area,
intensity
and
duration
duration
for
a
specific
fire.
You
then
evaluate
vegetation
mortality
based
on
cambial
damage,
so
damage
to
the
bark
as
well
as
crowns
to
scorch
damage
to
the
foliage.
A
A
We
also
have
collaborators
at
los
alamos
are
leading
an
effort
incorporating
live
fuel.
Moisture
live
fuels
are
very
dynamic
through
time,
much
more
dynamic
than
dead
fuels,
which
change
very
slowly
using
fates,
hydro.
We
can
capture
water
content
at
sub
daily
scales
and
look
at
changes
in
live
fuel
moisture,
and
so
the
figure
on
the
right
shows
the
number
the
increasing
number
of
days
with
live
fuel
below
a
critical
79
threshold
and
that
number
of
days
and
subsequent
fire
risk
is
increasing.
A
As
you
move
out
into
the
future
for
an
rcp
4-5
scenario
active
crown
fire
where
we
have
a
procedure
implemented
where
you
can
track
the
vertical
propagation
of
fire
through
the
crown,
depending
on
the
fire
intensity
and
the
crown
fuel
density
and
moisture,
we're
looking
at
topography
and
hydrology
interactions
using
the
ctsm
hill
slope,
hydrology,
remapping,
and
so
do
these
does
the
do
heterogeneous
hill
slopes
and
their
different
water
tables
correspond
carry
on
to
fire
behavior
management
and
mitigation
fires
within
the
wildland
urban
interface
are
increasingly
problematic,
and
so
fuel
management
are
inherently
size,
structured
and
so
using
the
fates
logging
module.
A
Colleagues
at
lawrence
berkeley
are
going
to
be
looking
at
whether
or
not
we
can
capture
changes
in
fire.
Behavior
based
on
management,
prescribed
management
treatments
and
then
also
degradation
associated
with
fires.
Degradation
shown
at
the
top
is
a
figure
for
the
amazon,
with
areas
of
deforestation
highlighted
in
red
and
so
for
this
particular
region.
Deforestation
is
more
than
40
of
all
the
fires.
A
This
type
of
degradation
increases
forest
edges,
so
you
get
small
patches
of
forest
and
larger
edges
are
associated
with
increased
heat,
percolation
and
then
increased
fires,
and
so
the
figure
on
the
bottom
is
a
long-term
burning
experiment,
and
so
you
can
see
different
disturbed
canopies.
These
disturbed
canopies
have
less
humidity
inside
the
canopy
and
drier
fuels
on
the
ground,
making
them
more
susceptible
to
fires.
A
We're
integrating
land
covered
land
use
change.
Data
sets
into
fates,
but
including
this
type
of
degradation
is
an
important
feedback
onto
fire
behavior,
and
so
with
that,
I
just
want
to
highlight
fate
spitfire
as
a
tool
for
action,
and
so
we
need
to
consider
the
consequences
of
our
choices
in
terms
of
management,
and
so
this
is
a
decision
framework
put
out
for
stakeholders
and
managers
of
systems.
A
C
A
So
the
fundamental
fire
behavior
equations
are
actually
the
same.
They
both
use
a
rothermell
approach,
which
is
a
traditional
rothermel
van
wagner,
equations
of
fire
behavior,
and
so
it
has
to
do
with.
You
know,
empirically
derived
equations
of
how
the
fire
acts
in
terms
of
fuel,
moisture
and
subsequent
fire
area
and
behavior.
G
All
right,
great
thanks,
so
this
is
a
quick
summary
of
two
studies
that
have
been
involved
with
over
the
past
year
that
have
really
changed
the
way
that
I
look
at
wildfire
as
an
interactive
component
of
the
climate
system,
so
rather
than
just
a
passive
kind
of
responder
to
the
climate,
to
suggest
that
wildfires
are
really
an
essential
part
of
the
component.
Sorry,
the
couple's
climate
system
and
the
picture
that
you
see
here
is
a
good
example
of
that.
G
So
this
is
the
australian
wildfires
from
early
january
and
2020.,
the
red
dots
has
the
individual
fires
and
you
see
the
emissions
over
the
continent
being
injected
over
the
southern
oceans
and
starting
to
interact
with
the
cloud
fields,
and
I
think
this
is
a
really
a
good
picture
to
emphasize
the
cross
working
group
aspect
of
this
discussion,
because
those
are
injected
over
the
southern
ocean.
They
interact
with
clouds
on
a
microphysical
level.
They
have
direct
radiative
effects
as
well
and.
F
G
As
I'll
show,
they
actually
have
response
in
the
ocean
and
so
you're
tying
together.
The
atmosphere,
the
land
surface
and
the
ocean
was
just
kind
of
one
topic
of
case.
Study
go
ahead
and
skip.
G
So
some
of
the
background-
wildfires,
you
know
if
you
look
through
the
literature,
the
wildfires
have
traditionally
been
viewed
as
kind
of
this
passive
component.
That's
responding
to
the
climate
system
and
really
what
we
see
now
in
cesm2
is
that
they
play
an
active
role
in
driving
climate
responses,
and
this
speaks
to
the
whole
manner
in
which
wildfires
are
prescribed,
for
example,
in
the
cmf
configurations,
and
so
we
really
need
to
rethink
in
recent
decades.
G
Is
you
know
what
are
the
climate
responses
to
recent
major
wildfire
events
in
cesm2
and
what
does
that
suggest
about
how
we
actually
go
about
representing
fire
in
climate
predictions
and
also
climate
projections?
And
I'm
arguing
here
that
fire
is
both
an
important
feedback
within
the
climate
system,
but
also
a
source
of
predictability.
And
these
two
case
studies
kind
of
illustrate
that
by
the
next
slide.
G
So
the
first
case
is
something
that
was
discussed
briefly
yesterday
morning.
Perhaps
you
saw
it.
It
speaks
to
how
the
theme
of
six
biomass
emissions
are
prescribed,
and
so
there
was
a
decision
made
when
putting
together
the
emissions
data
sets
that
the
variability
in
the
satellite
era
that
you
see
in
the
red
kind
of
on
the
right
hand,
side
of
that
graph
would
be
preserved.
G
And
so
you
have
this
large
variability
that
spliced
onto
a
record
with
almost
none
and
so
there's
the
step,
change,
kind
of
spurious
transition
and
variability,
and
the
question
we
had
is:
what's
the
climate
effect
of
that
step,
change
and
so
in
the
red.
You
have
the
cmf6
projections,
and
that
applies
to
the
time
series
on
the
right
as
well
and
then
in
the
black
we've
homogenized.
G
These
emissions,
where
we've
just
prescribed
a
mean
annual
cycle-
and
this
is
just
from
40
to
70
north
and
what
we
find
is
that
there's
a
large
climate
response,
just
this
narrow
band
of
emissions
and
the
variability
in
that
band.
So
we
have
20
ensemble
members
in
the
black
and
11
in
the
red,
and
what
you
see
is
that
north
of
around
30
degrees,
north.
You
have
this
spurious
warming
that
is
driven
by
this
spurious
transition
of
variability
and
it's
not
small.
G
It's
about
half
a
degree
c
over
much
of
the
northern
hemisphere
and
at
its
core.
It's
driven
by
the
fact
that
the
large
variability
in
the
emissions
is
thin
in
the
cloud
field,
and
so
that's
allowing
more
short
wave
radiation
to
be
absorbed
by
the
system,
both
at
the
top
of
the
atmosphere
and
at
the
surface
and
in
turn,
that's
probably
triggering
a
number
of
feedbacks
in
the
cryosphere,
with
snowmelt
at
the
surface
and
sea
ice.
G
And
so,
interestingly,
even
though
we
have
the
same
net
amount
of
emissions,
the
structure
of
those
emissions
and
associated
non-linearity
seems
to
be
driving
this
robust
climate
response
and
it's
not
a
small
effect.
As
I
say,
if
you
look
at
the
difference
between
cesm1
in
the
blue
and
cesm2
and
the
black,
you
know
the
majority
of
that
difference
and
in
the
theme
of
six
forces
in
the
red
is
explainable
by
the
actual
biomass
emissions.
You
know,
otherwise
they
look
like
very
much
the
same
model.
G
So
it's
not
about
climate
sensitivity,
it's
really
about
the
forcing
and
what
they're
doing
within
the
model.
If
you
want
more
details
on
this
by
the
way
I
have
a
poster
later
in
the
poster
section,
and
I
can
go
into
the
nitty-gritty
next
slide.
Please.
G
And
so
the
second
case
is
trying
to
simulate
the
climate
response
to
this
2019-2020
australian
bushfire
season,
and
what
we've
done
is:
we've
actually
prescribed
the
emissions
in
the
model
and
looked
at
the
coupled
climate
response
using
50
ensemble
members.
G
Okay,
I'll
make
it
quick,
so
the
the
the
effects
of
these
aerosols
in
the
clear
sky
is
actually
quite
small.
So
it's
about
two
tenths
of
a
watt
per
meter
squared,
but
when
you
look
at
the
all
sky
effects,
they're
actually
very
large,
around
two
watts
per
meter
squared
and
that's
on
par
with
volcanic
eruption,
and
the
reason
is
is
because
the
aerosols
are
brightening,
the
cloud
fields
and
they're
doing
so
across
the
southern
hemisphere.
G
So
it
gives
you
this
large
intra-hemispheric
energy
imbalance
and
in
turn,
the
model
simulates
a
disposition
towards
la
nina
in
2021,
which
is
actually
what
we
saw
and,
like
I
said,
if
you'd
like
more
information,
see
the
poster
session
and
then
you
can
just
skip
to
the
next
slide.
For
summary
points
and
I'll
take
any
questions.
C
C
E
G
Right
yeah,
I
think
the
latitude
is
a
key
part
of
the
effect
I
mean,
so
one
subtle
aspect
of
what
I
just
presented
is
the
fact
that
we
have
an
opposite
response
to
the
wildfire
variability
in
the
northern
hemisphere
from
40
to
70
north
and
the
australian
wildfires,
one
of
them
cools
the
climate
system
and
won't
actually
warm
the
climate
system
right,
and
so
why
is
that?
And
I
think
the
way
that
there
are
many
ways
that
the
aerosols
can
interact
with
clouds.
G
They
can
thin
the
cloud
field
or
they
can
brighten
the
clouds,
and
there
are
probably
many
more
ways
that
they
can
interact,
and
so
you
know
trying
to
work
through
and
then
evaluate,
which
is
right.
Whether
either
is
right
with
observations
is
it's
work
that
we're
doing,
but
certainly
the
latitude
is
part
of
the
effect
and
cryospheric
feedbacks
are
part
of
the
effect.
I
Okay,
so
I'll
talk
about
future
projections
of
welfire
under
different
scenarios
in
csm2
next
slide,
please.
I
So
we
looked
at
future
projections
under
seven
scenarios
from
the
year
2015
to
the
year
2100
and
namely
five
ssp
scenarios
and
the
tool
geo
engineering
scenarios,
and
these
ssp
scenarios
range
from
the
sustainable
development
scenario,
which
is
ssp1
2.6
to
the
unmitigated
baseline
scenario
and
which
is
ssp
5
8.5
and
the
ssp
5
3.4.
Os
is
an
overshoot
scenario.
I
It
follows
ssp
5,
8.5
through
2040
and
then
negative
emissions
is
undertaken
to
rapidly
rapidly
reduce
the
emissions
to
zero
by
about
2070.
I
I
This
slide
shows
the
time
series
of
global
total
burn
area
under
those
seven
scenarios
and
the
global
total
welfare
burn
area
is
projected
to
decrease
on
those
virtual
engineering
and
overshoot
scenarios,
but
increase
under
all
the
other
scenarios
and
the
largest
increase
in
the
global
burn
area
is
found
under
the
ssp5
8.5
scenario,
which
is
around
20
and
the
largest
decrease
is
found
in
zero.
G6
sell
for
scenario,
which
is
around
negative
eleven
percent,
and
the
next
slide
is.
I
This
slide
shows
the
horizontal
change
of
an
area
in
the
period
of
2091
to
2100
relative
to
the
period
of
2021
to
2030,
and,
as
we
can
see
in
that
circle,
the
40
degree
north
to
70
degree.
North
latitude
is
the
only
latitude
binder
in
which
the
burn
area
consistently
increase
under
other
scenarios
and
even
the
geoengineering
scenario
scenarios.
And
if
you
go
to
next
slide,
if
we
look
at
the
regions,
the
strongest
relative
increase
is
found
over
the
four
year
north
america,
under
the
ssp
5
8.5
and
ssp
3
7.0
scenarios.
I
The
burn
area
over
boreal
north
america
will
be
about
three
times
higher
by
the
end
of
the
century,
compared
to
the
period
of
2021
to
2030
and
over
the
temperature.
North
america
ssp
3
7.0
leads
to
the
strongest
increase
in
the
by
the
end
of
the
century,
which
is
about
a
hundred
percent
increase
and
followed
by
ssp58.5
scenario,
which
is
50
and
the
next
slide.
I
I
This
figure
shows
the
correlations
of
bone
area
with
the
driving
factors
over
both
border
north
america
and
the
temperate
north
america,
as
we
can
see,
relative
humidity
and
the
soil
moisture
are
the
most
important
driving
factors
for
the
trend,
which
is
consistent
with
previous
studies,
and
it
is
also
worth
pointing
out
that,
under
different
scenarios,
the
impact
of
these
driving
factors
can
be
different
and
that's
all
thank
you.
C
C
Alan
also
had
a
question
in
the
chat
that
I
might
go
ahead
and
answer.
The
question
is
their
fires
and
their
smoke
chemistry
connected
to
climate
and
at
least
the
chemistry
is
not,
emissions
aren't
yeah
yeah.
The
emissions
are
not.
B
Yeah,
so
this
is
not
an
interactive
fire
run.
What
it
is
is
really
just
looking
at
the
area
burned,
which
really
nicely
leads.
To
my
talk,
I
guess
because
we
don't
have
at
the
moment
modern
simulation
that
are
fully
coupled
with
the
climate
and
run
interactively,
and
that's
really
because
we
haven't
really
managed
to
produce
a
model
version
that
realistically
reproduces
the
historical
trend
of
temperatures.
E
You
that
the
carbon
emissions
are
accounted
for,
so
the
carbon
consequences
on
land
are
part
of
the
main
carbon
cycle
and,
if
you're
doing
a
fully
coupled
bgc
simulation,
that
would
go
into
the
atmospheres
you
somewhat
going
down
through
co2
and
contribute
to
the
overall
co2
budget.
C
Yep
I'll
reset
the
timer
go
ahead.
B
Okay
yeah,
so
this
is
really
about
the
interactive
fire
emissions
in
csm.
I
really
is
motivated
by
the
talks
we
had
before
that
many
of
the
simulations
we
are
doing
are
really
was
prescribed
area
with
prescribed
emissions,
for
example
from
cmp6
the
biomass
burning,
but
it
is
really
important
to
have
interactive
fire
emissions
to
get
the
right
variability
to
have
the
connection
to
the
meteorology
because
prescribed
fire
emissions
are
not
responding
to
any
of
those
climate
situations.
B
However,
we
do
have,
and
that's
what,
when
food
study
is
about
really
we
are,
the
land
model
is
describing
all
this
area
burn.
So
we
have
the
information
and
what
we
already
have
also
is
the
coupling
to
the
atmosphere.
We
have
the
capability
to
couple
and
we
have
done
previous
model
simulations
where
actually
those
fire
emissions
are
getting
into
the
atmosphere.
B
It
was
even
cooling
and
when
fu
actually
just
spoke
had
done
a
lot
of
work
on
it
before
and
identified
some
some
reasons
for
four
problems,
and-
and
one
thing
is
here,
for
example,
when
you
look
on
the
left
plot,
you
see
that
in
gray,
this
is
the
cmip6
fire
emissions
at
the
fire,
carbon
emissions
and-
and
what
we
can
see
is
that
it
has
a
trend
which
is
going
from
two
to
almost
three
and
then
going
down
and
then
up
again
and
usually
fire
emissions
or
area
burned
is
evaluated,
as
we
have
seen
through
the
present
day
period,
where
we
have
observations,
but
it
hasn't
been
much
evaluated
in
the
past.
B
Besides,
there
were
now
some
recent
studies
like
hamilton
at
all
2018,
showing
that
we
actually
probably
assuming
rate
to
low
fire
emissions
during
pre-industrial
control,
and
so
what
we
have
done
is
we
took
the
c
mid-five
model
and
the
cmap6
model,
the
basic
csm
2.2
setup
and
started
from
the
pre-industrial
simulations
in
1850
and
run
a
historical
simulation.
B
But
what
happens
was
that
we
saw
a
sudden
increase
in
the
fire
carbon
emissions,
as
you
can
see
in
that
plot,
and
then
it
would
kind
of
stay
the
same
and
then
go
a
little
bit
down.
But
the
problem
is
in
our
pre-industrial
control
simulations.
We
are
running
without
the
land
use,
and
so
there
is
an
accumulation
of
unburned
carbon
and
once
we
start,
then
the
fires
and
in
in
the
historical
simulation
we
get
the
sudden
increase
of
carbon
emissions.
B
And
obviously,
if
you
go
to
the
next
slide,
that's
impacting
the
climate,
and
you
see
on
the
right
that
the
temperature
would
suddenly
drop
and
then
we,
while
the
model,
then
itself
would
show
a
very
similar
trend.
This
kind
of
situation
was
something:
obviously
we
cannot
keep,
and
so
in
the
next
slide,
we
had
put
now
some
effort,
together
with
some
people
like
dave
and
john
marika,
holland
and
others
to
really
set
up
a
new
pre-industrial
control
simulation.
If
you
go
to
the
next
slide,
please
yeah.
B
So
that's
a
new
csm2
setup
instead
of
starting
from
the
pre-industrial
controller
in
1850
and
run
it
without
the
lane
use.
We
actually
start
in
1750
and
run
with
interactive
land
and
in
order
to
kind
of
spin
up
the
model
as
a
pre-industrial
control,
but
have
have
the
land
evolving.
So
we
can
get
the
interactive
fire
emissions
at
the
same
state
as
you
can
see
on
the
left
top
plot
there,
and
so
we
also
noticed,
though,
that
after
we
first
started
there
were
some
imbalances
of
the
model.
B
We
had
a
new
setup.
We
have
the
latest
land
version
we
used
updated,
but
scavenging.
We
were
a
bit
out
of
balance.
We
were
positive
in
in
the
top
of
the
atmosphere
imbalance,
so
we
also
added
some
oasis
or,
as
is
ocean
emissions
to
this,
which
which
changed
the
balance,
and
then
we
performed
several
rounds
of
tuning
which
finally
got
us
to
what
you're
seeing
here,
which
is
a
pre-industrial
control
run
or
with
interactive
land.
B
So
only
the
land
is
evolving
from
7
70
50,
but
the
whole
rest
of
the
climate
model
is
an
1850
control
simulation
and
in
that
way
now
you
see
that,
besides
having
the
higher
carbon
emissions,
what
we
want
we
do
have
about
reasonable
surface
temperature,
and
then
we
were
a
bit
concerned
about
the
sea
ice
as
well,
which
is
a
little
bit
low
still.
Nevertheless,
since
we
are
expecting
that
the
trend
may
be
different
than
the
fire
emissions
in
their
stored,
iran,
we
decided
to
start
a
historical
simulation
and
so
in
the
next
slide.
B
This
is,
if
you
go
to
the
next
slide,
please
you
see
just
the
start.
B
So
this
is
the
last
slide
and
it
really
is
just
started.
I
didn't
have
much
time.
There
was
some
other
mistake
in
the
run,
so
I
had
to
run
over,
but
nevertheless
what
we
are
seeing
is
so
far
that
you
do
have
higher
carbon
emissions,
while
the
temperature
and
the
sea
ice
is
looking
so
far
reasonable.
So,
let's
we
will
go
from
here.
We
we
continue
this
run
and
hopefully
we
can
produce
a
simulation
that
eventually
can
be
run
fully
coupled.
B
C
Thanks
just
about
a
minute,
if
anyone
has
a
question
for.
B
B
B
B
I
think
we
have
this
community
computer
time
project.
You
know
some
computer
time
with
our
community
project.
That
includes
also
other
model
simulations,
but
this
is
one
that
we
can
actually
do
some
of
the
historical
simulations
and
do
some
ensemble
members.
C
J
Okay,
so
I'm
going
to
talk
more
about
the
atmospheric
composition,
impacts
of
fires
on
the
next
slide.
There's
already
been
a
lot
of
motivation
for
the
importance
of
fires,
but
I'm
going
to
sort
of
outline
some
of
the
big
uncertainties
related
to
the
emissions,
the
chemistry
and
the
transport
of
the
plumes,
so
80
to
90
percent
of
the
emissions
into
the
atmosphere
are
co2
and
a
fraction,
and
then
of
that
non-co2
portion,
which
you
know
seems
like
relatively
small
rco
and
both
volatile
organic
compounds
and
primary
particles.
J
But
they
really
play
a
very
important
part
in
determining
the
atmospheric
composition
impact
of
fires
on
and
on
air
quality
pollutants,
such
as
pm
2.5,
and
which
includes
secondary
aerosols
and
ozone,
which
depends
on
the
chemistry.
So,
as
you
can
see,
for
example,
on
the
september
4th
the
fires
in
the
west,
western
u.s.
J
J
So
I've
mostly
been
working
with
offline
emissions,
inventories
which
have
different
sorts
of
uncertainties.
J
Some
ways
are
simpler
than
what's
calculated
on
the
land,
but
they
also
have
other
inherent
uncertainties,
so
they're,
based
on
the
actual
detection
on
the
location
of
the
fire
from
satellite
from
modis
and
beers,
as
shown
on
the
right
here
and
then
the
emissions
are
calculated
based
on
the
area
burned
and
s
which
has
to
be
estimated.
J
The
biomass
density
and
the
fraction
of
the
biomass,
that's
actually
consumed,
and
then
the
emission
factor
that
tells
you
how
much
of
any
given
compound
was
emitted
per
amount
of
fuel
consumed
so
for
offline
emissions,
inventories,
there's
the
uncertainty
of
actually
detecting
the
fire
and
getting
a
fire
count
for
the
fires
and
then
matching
that
up
with
a
vegetation
map
to
determine
the
type
of
fuel
burned
and
estimating
how
much
was
consumed.
J
J
So
the
uncertainty
in
just
the
emission
factors
of
things
like
no
and
vocs
will
result
in
the
in
uncertainty
in
the
ozone
and
secondary
aerosols
that
are
produced
and
there's
also
uncertainty
in
the
representation
of
all
the
chemistry
in
the
model.
We
have
to
simplify
the
amount
of
chemistry
we
have
in
the
model,
and
so
there
are
many
more
vocs
that
are
observed
and
fire
plumes
that
are
rarely
included
in
3d
chemical
schemes
and
then,
of
course,
the
model
resolution
has
a
big
impact,
both
horizontal
and
vertical,
on
the
chemical
regimes.
J
So
this
plot
shows
the
observed
budget
of
nitrogen
species
in
fire
plumes
during
the
weekend,
experiment
in
2018
and
how
it
changes
downwind.
So
it's
including
both
chemical
processing
and
dilution
of
the
plume,
and
so
you
can
see
like
in
the
light
blue,
a
secondary
product
of
pan
increases
as
you
go
downwind
of
the
plume,
but
the
primary,
no
and
no2
in
in
green
decrease
next
slide.
J
J
What
happens
in
the
boundary
layer
for
the
mixing
detrainment
from
the
mixing
from
the
boundary
layer
or
entrainment
of
remote
smoke
back
into
the
boundary
layer
and
then,
of
course,
getting
the
long-range
transport
because
we
know
smoke
can
travel
across
continents
and
around
the
hemispheres?
J
And
finally
that's
fine.
We
started
doing
some
work
on
using
the
regionally
refined
version
of
camcam
with
14
kilometer
resolution
over
the
u.s
analyzing.
The
two
recent
aircraft
campaigns
focused
on
fires.
J
We
can
and
fire
x,
and
so
it's
just
an
example
of
the
output
from
for
one
day
with
the
conus
grid
and
when
foods
made
a
nice
plot
showing
that
each
the
co
concentration
for
at
the
surface
for
each
grid
box
for
a
particular
fire
that
the
dc-8
sampled
and
showing
the
the
flight
tracks-
and
this
really
indicates
that
perhaps
this
resolution
is
sufficient
for
analyzing
these
campaigns
and
and
critically
evaluating
the
model.
E
So
have
you
tried
to
look
at
like
a
even
lower
resolution?
Version
of
the
model
see
how
it's?
Because,
as
we
go
into
this
fully
coupled
one
degree,
first
of
all,
I
guess
there's
a
question
about:
are
those
emissions
being
processed.
F
J
Yeah,
we've
done
lots
of
runs
at
one
degree,
and
it
depends
on
the
scale
that
you're
looking
at.
So
I
think
you
know
away
from
the
fires,
often
the
the
errors
in
chemical
regimes
average
out,
but
if
we
want
to
really
evaluate
the
say,
the
emissions,
we
want
something
at
a
finer
scale
and
maybe
even
you
know
finer
than
14
kilometers,
so
still
for
something
for
huge
fires,
like
the
australia
fires.
Probably
one
degree
is
fine
where
you
have
fires
over
many
grid
cells,
even
at
that
scale.
J
C
A
Yeah,
thank
you
everybody.
These
are
it's
a
really
nice
motivating
talks
for
our
discussion
and
so
now
we're
going
to
switch
over
and
we'll
have
a
short
poster
presentation
like
sort
of
a
come
see
my
poster,
but
please
do
come
to
the
poster
session.
We
have
some
great
posters,
let
me
and
so
douglas
you'll
be
up.
First.
A
K
Great
thanks,
and
so
just
like
to
start
by
thanking
my
co-authors
and
the
organizers
for
allowing
me
to
come
and
talk
today,
and
also
thanks
to
simone
for
kind
of
highlighting
some
of
what
I
want
to
just
point
out
in
this
short
talk
here.
So
aerosols
from
fires
can
impact
the
climate
and
earth
system.
In
many
ways
my
poster
will
be
focusing
on
two
particular
aspects:
the
aerosol
radiative
forcing
and
also
the
role
of
fires
in
providing
nutrients
to
remote
ecosystems.
K
K
And
if
we
look
at
the
numbers
at
the
bottom
of
the
plot
here,
we've
got
four
four
different
estimates
of
pre-industrial
fire
emissions
and
also
the
present
day
carbonaceous
aerosol
emissions.
For
comparison,
though,
it
can't
go
into
all
of
the
details
and
in
the
differences
between
these
four
different
estimates,
they
can
broadly
be
summarized
that
the
two
fire
models
on
the
left
hand
side
with
the
higher
fire
emission
estimates
specifically
represent
land
fragmentation
and
the
impacts
that
has
on
fire
suppression.
K
So
things
like
increasing
agricultural
or
pastoral
land
use,
whereas
the
two
estimates
on
the
left
right
hand,
side.
The
lower
estimates
have
either
the
minimal
representations
of
how
land-use
land
cover
change,
impacts,
fire
emissions
or
in
the
case
of
ericom.
None
at
all,
and
this
produces
a
wide
range,
as
we
can
see
here
in
the
aerosol
emissions
and
but
when
we
investigate
paleo,
paleo
environmental
archives
of
the
change
in
fire
emissions,
for
example,
things
like
ice
cores
or
the
charcoal
record
or
dendrinology
for
firing
stars.
K
They
all
generally
seem
to
point
to
higher
fire
emissions
in
the
pre-industrial
as
compared
to
the
present
day.
So
how
that
uncertainty
in
those
different
fire
emission
estimates
and
pre-industrial
propagates
into
the
range
of
uncertainty
in
and
the
pre-industrial
to
present-day
aerosol
radiator,
forcing
is
what's
shown
on
the
plot
on
the
top.
K
With
a
higher
fire
emission
in
the
pre-industrial,
we
end
up
with
a
smaller
difference
in
the
aerosol
burden
between
pre-industrial
and
present
day
and
that
diminishes
the
aerosol
radiator
forcing
over
the
same
period,
and
it
was
a
small,
aerosol
radiator,
forcing
in
comparison
with
a
lower
aerosol
emission
than
pre-industrial.
A
more
pristine,
aerosol
state
has
been
assumed,
you'll
end
up
with
a
higher
aerosol
radiator,
forcing
over
that
period.
K
We've
run
this
experiment
twice
now
and
most
recently
on
the
orange
dots
that
was
with
the
csm2
and
we
can
see
here,
is
in
comparison
to
our
previous
experiment
in
2018.
This
year
ii
has
a
higher
sensitivity
to
this
change
in
aerosol
and
in
the
pre-industrial
than
the
uk
model
that
we
used
before
for
comparison
and
in
the
gray
region.
K
What
I'm
showing
there
is
the
full
range
across
the
aerosol,
forcing
from
ar5
from
all
the
models
that
partook
in
the
cmic
5.,
and
we
can
see
that
that
actually
mirrors
and
the
range
that
we're
predicting
here
from
just
the
uncertainty
in
pre-industrial
fire
emissions.
K
So
that
is
suggesting
that
fires
could
represent
the
largest
uncertainty
in
predicting
human
climate
change.
K
Fires
are
a
core
component
of
the
coupled
earth
climate
system
and,
if
you
just
like
to
go
on
yeah,
lovely
thanks
and
and
also
within
the
poster-
and
I
want
to
just
highlight
that-
we've
been
working
on
developing
a
module
for
mam4,
which
can
predict
both
aerosol
iron
and
phosphorus,
which
further
links
the
land,
atmosphere
and
ocean
components.
And
if
you're
interested
in
knowing
a
little
bit
more
about
that
and
how
fires
and
impact
marine
ecosystems.
Then
I
welcome
you
to
come
along
and
have
a
chat
with
me
later
or
about
the
aerosol
radiator.
K
A
Thanks
douglas
sorry,
I
wasn't
sure
if
I
was
supposed
to
advance
until.
E
E
On
the
right
hand,
side
to
work
with
with
the
with
ma'am
yeah.
J
K
Separate
okay,
yeah.
This
is
separate,
although
it's
it's
available
on
a
branch
in
cam6,
and
if
anybody
is
interested
in
running
iron
within
the
atmosphere,
then
you
know
we
can
help
support
that
happy
to
collaborate
with
anyone,
but
it's
not
currently
in
the
project
line
for
for
cam6.
As
far
as
I'm.
A
L
L
You
can
start
on
the
next
slide.
Okay,
go
ahead,
so
this
work
is
motivated
by
recent
debris
flows
and
flash
floods
that
have
occurred
after
fires
here
in
southern
california,
and
also
across
the
western
u.s,
and
so
next
slide,
please,
and
so
using
the
cesm
large
ensemble,
I'm
a
little
behind
you
guys,
I'm
still
using
cesm1.
L
L
So
we
wanted
to
use
the
c
e
some
large
ensemble
to
see.
If
we
can
capture
can
capture
the
changes
in
these
rare
events
and
kind
of
their
aiming
and
see
if
they're
occurring
kind
of
consequentively
to
create
conditions
that
could
lead
to
debris,
flows
and
flash
floods,
and
what
we
find
is
that
indeed
they
can.
L
L
And
we
see
that
this
change,
can
you
advance
one
more?
I
think
I
have
an
animation
in
there
yeah,
so
so
this
change
in
the
likelihood
is
partly
driven
by
an
expansion
and
amplification
of
wet
seasons
and
fire
season.
So
this
is
kind
of
the
seasonal
changes
in
california,
and
we
see
that
there's
kind
of
this,
both
amplification
and
expansion,
of
both
the
fire
season
and
the
wet
seasons
in
the
future,
and
I
think
I
have
one
more
slide.
L
That's
an
advertisement
yeah
so
come
to
my
poster
to
see
more
details
and
methods
and
data,
more
seasonal
and
regional
variations,
and
these
types
of
compounding
extremes
and
much
more.
So
thanks
for
having
me
here,
it's
been
a
great
session
so
far.
E
L
Yeah,
so
we
look
at
both
within
one
year,
two
years,
five
years,
and
then
we
also
look
at
like
sub-annual
to
kind
of
capture,
these
more
type
of
extreme
events,
extreme
conditions
that
could
more
likely
lead
to
things
like
debris,
flow
and
flash
floods.
So
that's
kind
of
so.
L
Yeah,
so
that's
yeah,
we
actually
try
only
use
the
99.9
percentile
of
rainfall
because
we
want
so
there's
this
kind
of
issue
with
the
csm
large
ensemble
simulations,
where
we,
the
smallest
kind
of
temporal
scale
we
have
is
a
six
hourly
but
to
really
capture
things
like
debris,
flows
and
flash
floods,
you
need
more
of
a
sub
hourly
estimate
of
rainfall
intensity.
L
L
L
A
C
A
You
to
all
of
the
speakers
and
thank
you
to
everyone
who's
here
for
the
session.
Let's
take
a
short
break
and
come
back
at
five
minutes
past
the
hour
to
have
a
discussion.
A
A
Okay,
welcome
back,
I'm
gonna
put
up
a
slide
with
some.
A
But
in
the
meantime,
if
anyone
wants
to
well,
I'm
looking
for
this
discussion
point
slide.
If
anyone
wants
to
start
us
off,
that
would
be.
C
A
So
we
really
just
wanted
to
open
this
up
to
the
community
so
that
we
could
come
together
and
talk
about
how
do
we
handle
fire
in
a
coupled
fashion?
What
are
our
priorities
and
what
are
the
obvious
challenges,
and
so
I
know
that
there
are
a
lot
of
communities
that
are
interested
in
fire.
It
affects
a
lot
of
different
parts
of
the
system,
and
so
that's
part
of
the
reason
why
we
wanted
to
come
together
as
a
group
to
talk
about
this.
Are
there
opportunities
that
we
have
to
look
at
things
together?
A
It's
really
it's
really
open,
but
one
of
the
things
that
came
up
in
the
chat
after
the
talks
was
the
connection
of
lightning
and
so
at
the
moment,
lightning
is
considered
as
a
part
of
a
static
data
set,
but
it
a
data
set
that
varies
spatially
as
well
as
temporally,
but
there
are
there's
definitely
an
opportunity
to
include
lightning
as
a
as
connected
to
cam.
D
A
And
so
your
question
is
like
an
operational
scale
model
for
like
firefighters,
on.
A
I
think
it's
a
good
point.
The
opportunity
is
certainly
there
to
connect
with
folks
at
rel
who
are
doing
the
sort
of
operational
scale.
Fire
forecasting,
the
the
lee
fire
model
and
spitfire
are
both
more
at
the
fire
regime
scale
and
so
they're
much
it's
a
larger
spatial
scale
as
as
well
as
a
longer
time
scale,
but
something
that
you
can
get
from
these
models
connecting
to
an
operational
model
is
a
description
of
how
the
fuels
change
in
time,
and
so
with
a
lot
of
these
operational
mile
models.
They
have.
E
C
I
was
going
to
change
topics
a
little
bit
as
well,
but
it's
it's
semi-related.
I
was
just
going
to
make
the
comment
that
it
might
be
interesting
to
be
able
to
use
fire
counts
from
satellites
to
put
into
the
model.
To
then
use
those
to
drive
everything
else
interactively,
and
then
you
can
test
real
fires
and
the
result
from
real
fires
in
the
interactive
model.
C
Could
I
jump
in
this
is
sam
levis.
Let
me
just
turn
on
my
video
as
well
in
response
to
what
rebecca
just
said,
some
of
the
work
that
I
did
with
lara
a
year
and
a
half
ago,
or
so
was
to
expand
that
capability
of
ignitions
to
be
on
just
lightning
actually
even
observed
ignitions,
which
may
include
ignitions
because
of
lightning,
but
also
ignitions
because
of
other
human
activity,
so
so
that
capability,
I
believe,
is
now
within
the
the
main
trunk
of
the
ctsm.
C
A
C
D
Yeah
just
to
clarify,
so
what
can
be
done
is
that
the
lightning
and
the
population
density
can
be
used
by
either
sorry
a
little
dog
barking.
But
the
lightning
ignition
can
only
be
used
by
fate.
A
And
by
that
you
mean
the
like:
a
a
user
prescribed
ignitions
data
set.
D
F
Fates
has
the
ability
in
the
parameter
file
you
set,
which,
which
ignition
data
set
you
you
want
to
use.
So
it's
could
be
lightning,
it
could
be
observed,
ignitions
or
it
could
be
some
some
other
anthropogenic
type,
but
we
we
still
haven't
implemented
sort
of
more
anthropogenic
admissions
in
in
faiths.
Yet,
but
it's
a
it's
a
selection.
D
Oh
thanks,
so
I
wanted
to
follow
up
on
rebecca's
point
about
the
satellite
validation
because,
especially
in
the
high
latitudes,
I'm,
I
was
kind
of
nervous
about
the
tuning
that
we
did
for
the
pre-industrial
run
and-
and
I
don't
really
have
a
good
sense
of
you-
know
how
realistic
are
the
fires
that
you're
getting
in
the
pre-industrial,
and
you
know
so
once
it's
through
the
20th
century
it'll
be
really
interesting
to
see
how
it
compares
you
know,
present
day
and
so
forth,
because
yeah
those
sea
ice,
albedo
values
that
simone
presented
are
definitely
at
the
high
end
of
what
I
feel
comfortable.
A
A
Okay
douglas,
do
you
want
to
follow
on
that.
K
Yeah,
maybe
just
to
advocate
that
when
we're
looking
at
these
kind
of
fires
in
the
past,
it's
maybe
not
a
good
idea
to
overly
constrain
with
present
day
satellite
information,
but
rather
to
use
the
information
that
we
understand
about
the
past.
And
otherwise
it's
taking
the
assumption
that
the
way
that
fires
are
impacting
various
different
climate
regimes
or
being
impacted
by
climate
regimes
or
impacted
by
human
activity
in
the
present
day,
that
that
is
exactly
the
same
in
the
past
and
it's
most
likely
quite
different.
K
And
so
we
can
there's
a
wealth
of
paleo
proxy
data
kind
of
sitting
around
some
of
it
has
been
analyzed.
There's
been
a
brand
new
paper
out
and
just
in
the
last
couple
weeks
in
science,
for
example,
from
the
harvard
group
where
they've
been
looking
at
black
carbon,
any
ice
cores
in
antarctica.
I
looked
at
some
of
the
black
carbon
in
the
ice
cores
in
the
northern
hemisphere
and
I
think
that
there
is
more
that
we
can
do
there
to
constrain
what's
happening
in
the
past.
A
E
So
that's
a
good
point.
I
think
this
whole
validation
thing
is
is
quite
challenging,
because
I
want
to
step
back
and
just
muster
how
how
much
it's
clear
that
this
new
simulation
that
we're
doing
with
interactive
fire
as
far
as
I'm
aware,
might
be
unprecedented
in
terms
of
a
persistent
model.
Someone
can
correct
me
if
I'm
wrong
if
another,
but
there'd
be
very
few
groups
around
the
world.
Who
could
even
try
to
do
this,
and
so
we
kind
of
have
to
expect
that
things
are
going
to
kind
of
go
wrong
when
you.
E
So
we're
going
to
see
what
those
are
over
the
next
several
months.
I
think
about
what
what
goes
wrong,
but
one
you
know.
One
thing
that
occurs
to
me
is
that
one
possibility
is
that
you
know
we're
going
to
be.
E
A
fire
almost
every
day
in
the
western
us,
but
just
a
small
fire
that
could
be
what
it
looks
like,
which
is
maybe
not
that
unrealistic
these
days.
But
you
know
so
one
question:
I'm
gonna
I
I'm
gonna
be
looking
for
as
we
go.
G
E
A
Yeah,
that's
a
really
good
point
will
do
you
want
to
follow
on
that.
C
Yeah,
I
guess
I
was
curious
since
building
on
dave's
comment
that
this
is
maybe
the
first
time
this
has
really
been
done
in
a
coupled
sense.
If
there
were
contributions
from
the
paleo
group
to
the
to
this
kind
of
fire
working
group
session
kind
of
speaking
to
douglas's
point
about
using
that
record.
In
addition
to
the
satellite
record
and
then
the
other
question
I
had.
C
I
guess
this
is
more
for
louisa,
but
in
the
chemistry,
is
there
any
effort
to
try
to
keep
the
vegetation
that
the
that's
generating
the
aerosols
consistent
with
the
vegetation
that
clm
thinks?
Is
there
it's
probably
easier
for
the
big
leaf
model
than
it
is
for
fates,
but
for
these
runs
we're
prescribing
vegetation?
Are
you
guys
using
the
same
vegetation
that
the
land
thinks
is
there.
J
I'm
not
entirely
sure
what
you're
asking,
but
when
I
do
runs
like
with
the
thin
emissions
inventory
or
q-fed
or
whatever
they
use
their
own
vegetation
map.
We
could
certainly
modify
thin
to
read
the
a
clm
vegetation
map,
but
it's
thin
currently
used
as
a
modis
file.
C
A
B
Yeah
I
was
going
to
comment
first
on
the
under
chemistry.
Too,
I
mean
if
you
run
interactive.
Of
course,
we
use
the
same
lent
types
right
that
that
wasn't
your
question
or
was
it
the
interactive
fire
model
and
the
emissions
are
obviously
consistent
with
the
land.
B
Yeah,
if
it
has
the
chemistry
turned
on
it,
will
take
the
emission
directly
from
what
the
land
is
producing.
Okay,
so
there's
no
other
land
map
that
the
chemist
would
use.
I
think
that
was
also
the
press,
or
maybe
that
was
one
of
the
thoughts
yeah.
So
that's
consistent
if
you
run
interactively
the
the
point
that
dave
made
on
in
regard
to
the
variability,
I
mean:
that's
really
what
john's
studies
showed,
and
maybe
john
you.
You
wanted
to
comment
on
that
right.
B
If
you
have
the
large
fire,
the
large
variability,
you
suddenly
get
a
very
different
climate,
forcing
different
aerosols,
and
so
that's
that's,
probably
something
we
can
learn
and-
and
the
last
point
I
wanted
to
really
make-
is
that
the
problem
is
with
all
these
fire
models
are
compared,
as
we
already
said
to
the
satellite
era,
and
so
the
how
much
we
have
really
in
present
day
is
how
we
tune
those
models
which
doesn't
mean
that
it's
the
correct
way
and
that's
a
really
good
point
what
douglas
made.
So,
how
do
we?
B
How
do
we
actually
tune
the
model?
Can?
How
do
we
use
the
black
carbon
from
the
sea
ice
to
coarse?
You
know
to
compare
to
how
much
black
carbon
was
in
the
atmosphere.
I
think
that
would
be
an
interesting
thing
to
to
think
about
for
pre-industrial
times.
A
Yeah
good
points,
john,
do
you
want
to
follow
up.
G
Actually,
I
had
a
question
and
that
is
from
somebody
who's,
not
a
land
modeler
and
is
new
to
this.
How
will
we
know
if
the
model
is
wrong?
So
you
know
black
carbon
burdens
and
aerosol
in
ice
course
will
tell
you
some
aspects,
but
we
really
want
to
know
more
than
just
the
mean
deposition.
We
want
to
know
the
full
pdf
of
fires
and
those
will
have
a
climate
effect,
and
so
what
are
the
ways
that
we'll
be
able
to
tell
whether
our
simulation
is
reasonable
or
not?
G
E
I
I
can
try
to
answer
that
a
little
bit.
I
think
it's
I
think
it's
hard.
Maybe
jackie
has
a
better
sense,
but
you
know
what
fong
has
done
in
the
past
is
is
do
things
like
compare
to
the
g,
fed
satellite
for
fire
counts
or
fire
or
burned
area,
and
so
that's
one
measure,
but
that's
pretty
gross,
and
you
can
also
look
at
internal
variability
in
burned
area
and
fire
count
and
then
I'm
not
sure
which
status
she
uses,
but
you
know
total
fire
carbon
emissions.
There
are
estimates
of
that
at
present.
F
G
Yes,
back
in
my
mind,
was
the
caution
that
doug
just
made
that
we
shouldn't
be
using
the
satellite
error
to
extrapolate
back
in
time
on
what's
reasonable
right,
so
that
throws
both
of
those
you
know,
sources
out
the
window
and
so
what's
left
you
can
look
at
the
temperature
record
or
other.
You
know
very
indirect
indicators,
but
that's
not
ideal
either.
So.
A
I
think
so
adriana
foster
put
something
in
the
in
the
comments
to
look
at
above
and
below
ground
combustion,
post
fire
dynamics
and
fire
intensity,
and
then
I
want
to
also
loop
in
the
connection
to
the
chemistry
and
so
these
emission
factors,
a
lot
of
them
are
biome
level
emissions
factors,
and
so,
if
we
can
get
more
data
on
emissions
connected
with,
you
know
genus
or
more
specific
vegetation
types
that
would
allow
us
to
connect
a
specific
type
of
vegetation
burning
to
types
of
emissions,
but
but
you're
right.
A
I
Yeah,
I
just
want
to
follow
up
dave's
question
before
so
is
it
the
same
when
you
have
a
a
big
large
fire
or
a
bunch
of
small
fires
emitting
the
same
amount
of
aerosols
to
the
atmosphere?
I
guess
the
question
is
also.
I
Besides
the
total
amount
of
aerosols
get
into
the
atmosphere.
The
the
location
where
they
get
into
is
also
important
that
when
you
have
a
many
small
files,
their
emissions
will
be
most
likely
within
the
boundary
layer.
But
if
you
have
a
huge
large
files,
the
the
emissions
will
get
into
the
future
atmosphere
and
even
the
stratosphere,
so
that
will
have
different
impact.
Even
they
have,
even
if
they
have
the
same
total
amount
of
emissions.
I
E
E
E
B
I
Now
I
have
a
question
for
jackie,
so
you
you
plan
to
you
plan
to
implement
the
file
phase
file
to
crm
is
that
I
might
understand
correctly.
A
Yes,
yes,
I'm
I'm
implementing
it
now
and
I've
been
testing
it
and
then
yeah,
but
we
have
parallel
efforts
and
so
I've
been
testing
it
in
the
tropics
and
then
we
have
folks
testing
it
in
california
and
I'm
testing
it
in
boreal
canada.
And
so
so.
Yes,.
I
I
have
a
question
about
the
resolution,
for
example,
if
in
the
future
the
resolution
gets
higher,
how
do
you
deal
with
a
single
the
column,
not
communicating
with
other
core.
C
C
I
Yeah
like
when
the
model
resolution
is,
gets
really
high,
like,
for
example,
the
current
regional
refund.
We
are
using
14
kilometer
if
we
want
to
couple
that
to
the
land
and
some
files
maybe
have
larger,
like
light
files
with
size,
larger
than
14
kilometer
could
occur
in
reality,.
A
A
Yeah
but
then
the
other
point
with
this
is
even
if
you
have
grid
cells
that
have
similar
fuels,
you're
going.
If
you
have
similar
conditions,
you
will
get
ignition
on
co-located
grid
cells,
and
so
it's
not
propagation,
and
so
we
we
do
need.
Excuse
me.
We
need
to
look
at
this
issue
of
propagation
absolutely,
and
so
it
depends
very
much
on
the
resolution
that
you're
simulating
at
and
and
so
like.
No
fates
is
not
doing
propagation.
At
this
point
and
as
eric
said,
we're
not
ctsm
isn't
set
up
to
do
that.
Yet.
M
But
yeah
I
brought
up
propagation
because
peter
had
asked
about
operational
use
and
I
think
you
know
the
firefighters
they
act.
They
really
need
it
to
be
spatially
propagating
because
they
they
need
to
know
like.
Where
are
we
gonna
go
tomorrow
or
like?
Where
do
we
need
to
put
our
or
like
in
the
next
hour?
H
E
A
And
so
in
the
chat
lara
also
brought
up
the
point
of
multi-day
burning
and
so
within
a
lot
of
areas.
Globally.
Multi-Day
burning
is
a
consistent
behavior
within
the
spitfire
model.
There
is
a
limit
on
the
length
of
time.
Fires
can
burn.
I
don't
recall
if
that's
true
with
the
lee
fire
model,
if
if
there
is
multi-day
burning
and
so
it
I've,
we've
talked
about
some
solutions
for
that,
but
it's
something
to
keep
in
mind
as
you're
looking
at
these
models,
because
it's
definitely
unrealistic
to
have
a
cap
on
burning.
B
Yeah
and
I
think
that's
the
connection
to
the
forecasting
right
so
far,
how
do
you
simulate
10
days
ahead
if
you
want
to
do
a
forecast
and
the
fire
is
continuously
burning
right
now?
What
we
do
is
we
just
continue
our
fire
emissions
as
they
are
when
we
stop.
You
know,
at
the
day
of
present
day,
we
just
let
the
fire
burn,
but
how
can
this
be
integrated
better,
that
you
can
do
better.
A
Forecasts
well-
and
I
I
think
the
before
it
depends
on
what
time
scale
you're
talking
about
with
forecasting,
because
you
know
like
operational
fire
models
are
at
the
level
of
the
fire
event
and
that's
like
well
outside
the
abilities
for
something
like
spitfire.
You
know,
but
if
you're
talking
about
projections,
where
you're
going
to
look
at
management
for
your
community
and
look
at
you
know,
how
am
I
going
to
handle
fires
inside
the
wildland
urban
interface
having
a
cross-scale
inter
integration
of
you
know,
these
fine
scale
operational
models
as
well
as
something
coarse
scale.
A
C
B
B
B
A
Yeah,
I
think
I
think,
with
these
models,
there's
a
lot
of
uncertainty
at
the
daily
scale,
but
I
don't
know,
maybe
dave
you
feel
differently.
M
I
guess
this
is
shifting
gears
a
little
bit
to
in
like
how
we
can
validate
like
fire
intensity,
because
that's
one
of
the
main
things
that
we
calculate
in
spitfire
model
from
raw
thermal,
and
I'm
wondering
you
know
that
modis
has
the
fire
radiative
power.
But
I
have
had
a
lot
of
trouble
sort
of
getting
the
fire
radiative
power
to
be.
K
Douglas
so
I
I
don't
know
how
you
could
validate
it,
but
for
the
last
point
about
whether
you
need
to
and
the
intensity
of
the
fire
will
predict
the
ratio
of
black
carbon
to
organic
carbon,
which
is
released.
So
in
terms
of
the
climate
response
for
black
carbon
being
warming
and
organic
carbon
generally
being
scattering
and
cooling.
K
F
K
Yeah,
I
don't
know
about
at
the
global
scale,
but
certainly
when
they
do
prescribe
burns
you
can
you
can
see
it
quite
clearly.
You
know
in
the
at
the
fire
front,
it's
all
black
smoke.
You
know
that's
kind
of
coming
out
where
it's
hot,
so
the
very
high
black
carbon
and
then
in
the
smoldering
phase,
which
lasts
for
a
long
time
afterwards.
You
know
it
is
much
more
organic
that
white
smoke,
but
at
the
moment,
the
way
that
most
models
do
it
is
they
just
take.
K
You
know
one
emission
factor
for
one
particular
fire,
regardless
of
its.
You
know
the
power
or
the
intensity
of
that
fire.
So
you
know
the
emission
factors
themselves
would
likely
need
to
be
retuned.
I
guess
or
rethought
about
how
they
how
they
interact
in
terms
of
the
not
just
the
pft,
but
how?
How
complete
that
combustion
was.
I
guess
of
the
pft.
G
A
John,
thank
you
for
that
point
and
I
dave
do
you
want
to
make
a
comment,
but
then
I
would
like
to
shift
to
talk
about.
You
know
characterization
of
smoke
and
plume
dynamics.
If
we
can
dave.
E
Well,
that's
actually,
maybe
where
my
question
is
leading,
I'm
wondering
if
you
know
we're
doing
these
simulations
as
a
new
baseline,
whatever
simulation
of
climate
change
with
with
interactive
fire
and
whether
we
need
to
design
some
some
sensitivity
experiments
around
this,
they
would
probe
some
of
these
questions
about
you
know.
Even
if
we
don't
know
the
answers
you
know,
we
could
obviously
make
decisions
about
injection
height.
E
We
could
make
or
simple
primarization
of
injection
height
based
on
a
you
know,
simple
fire
risk
factor,
or
you
could
do
something
simple
about
black
carbon
versus
brown
carbon.
You
know
just
see
what
the
climate
response
is
and
get
a
sense
of.
You
know
which
one
of
these
things
is
the
most
important
thing
to
to
work
on.
A
I
We
just
started
a
study
trying
to
improve
the
representation
of
files
in
the
chem
cam,
so
we
tested
adding
file
diagonal
cycles
and
also
add
improvements,
but
those
are
offline.
So
what
we
did
was
what
we
did
was
we
added
those
at
the
diagonal
cycle
and
plume
inject
height
to
the
offline
emissions
and
put
that
as
input
use,
that
as
input
to
the
model
to
chem
cam,
because
that's
what
we
have
done
so
far,
we
do.
We
do
find
that
adding
dino
cycle
and
plume
rise.
I
I
And
I
have
a
presentation
tomorrow
if
anyone
is
interested
in
the
chemistry
on
climate
session
in
the
morning.
A
Yeah
yeah
they're.
Definitely
thanks
for
bringing
that
up
one
food
that
sounds
really
good.
There
are
definitely
relevant
fire
talks
happening
in
other
sessions.
A
A
It
needs
to
be
a
component
of
you
know,
development
moving
forward,
and
so
you
know
I
I
talked
about
we're
thinking
about
that
on
the
fate
side
and
then
adriana
said,
lower
branch
pruning,
and
so
how
can
you
modify
your
forest
to
adapt
to
these
fire
regimes?
E
Well,
I
know
fungus
is
intending
to
work
on
it
more,
but
I
think
she's
not
looking
at
from
the
forest
structure
point
of
view.
You
know:
she's
got
this
economic
population
trajectory
kind
of
overarching
perspective
which
which
probably
needs
to
factor
in
right
you
it's
those
it's
the
rich,
rich
countries
that
can
actually
manage
their
forests
or
try
to
manage
their
force.
E
So
I
think,
with
fates
is
where
that
you
know
it's
gonna
be
pretty
interesting
to
try
to
merge
that,
but
I
do
feel
like
that.
You
know
it'd
be
good
to
start
bringing
in
at
least
the
fong's
inputs.
You
know
to
help
drive.
You
know
as
a
forcing
factor
in
terms
of
determining
where,
where
fire,
where
the
force
are
being
managed,
you
know
gdp
or
population
or
other
socio-economic
factors.
A
A
A
lot
of
the
applications
at
this
stage
are
regional,
and
so
there
is
a
lot
of
interest
in
targeted
management
that
would
be
specific
to
an
area
like
california,
and
so
it's
a
specific
approach
and
so
there's
a
balance
of
approaches
within
fates
and
then
danica
points
out
in
the
chat
that
we
need
a
better
representation
of
agricultural
fires.
A
And
so-
and
I
absolutely
agree,
one
thing
to
note
with
agricultural
fires
is,
for
the
satellite
record.
Satellites
definitely
have
a
blind
spot
in
terms
of
detecting
these
fires,
a
lot
of
the
areas
they.
They
know
that
the
satellites
are
passing
over
and
use
these
detect
detection,
algorithms
to
capture
fires,
and
so
they
do
their
burning
at
night.
And
so
there
are
incidences
of
fires
where
they
burn
their
crops.
A
But
you
don't
necessarily
know
about
it,
and
so
it's
it's
sort
of
a
missed
connection
and
then
lara
points
out
that
fuel
wood
gathering
is
important,
so
yeah,
and
we
don't
we're
talking
about
this
within
fates.
But,
as
I
said,
this
is
regional.
A
Specific
sunlin,
do
you?
Would
you
like
to
comment
you've
been
giving
us
a
lot
of
information
in
the
chat?
Do
you
want
to
comment
on
warfire
or
some
of
the
other
things
that
you've
brought
up.
H
Yeah
yeah
sure
so
so
earlier,
I
put
something
in
the
chat
regarding
both
the
emissions
and
the
polymerizes.
So
oh,
some
of
the
studies
already
used.
The
quantitative
connections
are
relationships
between
the
bc
versus
oc
emission
ratios
to
the
combustion
efficiency,
and
some
people
use
that
in
the
global
simulations
of
the
fire
emitted
aerosol
effects
for
the
plumerizing
injection
head.
H
I
know
there
are
already
some
parameterizations
used
in
world
fire
for
the
interactive
simulations
and
the
last
comment
or
questions
I
put
in
the
chat
is
that
do
we
have
the
observed
or
sort
of
analysis
fire
data
where
the
different
file
types,
for
example,
the
natural
versus
anthropogenic,
are
separated?
H
Otherwise,
though,
we
could
not
constrain
that,
I
know
some
of
the
fire
datasets
like
the
ifc
national
inter-agency
fire
center
or
the
mtbs
the
monitoring
trends
in
burn
severity.
Maybe
some
of
these
data
sets
have
a
differentiation
in
different
file
types,
but
but
I
I
don't
know
if
anyone
knows
any
like
comprehensive
data
set.
People
already
put
it
together
for
us
to
validating
or
constrain
the
model
file,
input.
A
Sure
I'll
give
it
a
minute.
If
anyone
wants
to
step
in
and
comment
on
that,
I
don't
know
lara
did
you
want
to
oh,
she
dropped
away
before
there.
A
The
data
sets
that
you
highlighted
and
then
danielle
tuma
highlighted
cal
fire,
and
so
I
know
that
there's
been
work
done
within
california,
and
so
it
you're
going
to
be
limited
based
on
the
regions
for
that
separation
of
natural
versus
anthropogenic
there's
also
been
some
work
in
the
north
american
boreal
separating
those
it,
but
it
really
does
depend
on
where
you're
talking
about
and
so
finding
a
global
validation
is
going
to
be
a
challenge
laura.
Did
you
want
to
comment
on
natural
versus
anthropogenic
it's
possible
she's?
Not
there.
F
Yeah,
no
I'm
here
yeah.
I
was
just
chatting
about
something
else
in
the
chat.
The
frap,
the
cal
fire
frap
data
set
does
does
do
that
separation
but
but,
like
you
just
said,
that's
for
that's
for.
C
I
guess
I
do
wonder
if
someone's
suggestions
about
wharf-
that's
something
that
can
come
into
ctsm,
some
of
that
plume
height
and
emission
factors
are
already
in
a
you
know.
In
a
model,
can
we
can
we
bring
those
into.
E
Ctsm,
it
seems
like
it's
worth
some
exploration
that
there's
there's
the
broader
question
about:
how
long
are
we
gonna
be
maintaining
war
fire
with
this
unification
effort
across
ncar?
And
so
at
some
point,
if
you
know
at
some
point
wharf
is
not
no
longer
be
supported.
It
might
be
many
years
down
the
road,
but
if
war
fire
is
not
supported
anymore,
they
have
to
have
a
replacement
for
it
could
be
and
pass
ctsm
right,
someday,
ctsm
fire.
H
Yeah,
so
so
I
want
to
follow
up
with
dave's
comments,
so
so,
regarding
that,
I
think
there
are.
There
is
also
a
potential
issue
when
we
look
at
the
ctsm
file
of
phase
file
and
the
word
file,
because
the
warfare
model,
as
far
as
I
know
it
is
more
suitable
for
the
fire
event
simulations
where
you
have
a
very
detailed
fire
fire
spread
and
fire
related
heat
flux
and
the
interactions
with
the
weather
and
for
the
ctsm
fire
or
even
the
fade
spark.
H
It
is
probably
more
based
on
the
parameterization
suitable
for
a
climate
simulations
or
like
relevant
scale
studies.
I'm
not
sure
if
I
understand
correctly,
but
if
that
is
the
case.
So
when
we
apply
these
different
models
to
the
specific
applications
we
we
may
need
to
keep
in
mind
if
that
is
a
suitable
tool
to
use,
especially
when
we
do
the
unifications,
because
some
people
raise
the
question
that
part
of
the
parameterizations
used
in
the
current
crm.
H
Fire
is
actually
based
on
the
statistic
relationships
which
is
more
in
the
climate
sense,
instead
of
maybe
a
weather
scare
sense.
So
it's
just
a
a
caution.
I
guess
yeah
absolutely.
E
E
That
these
these
are
very
different
beasts,
so
this
is,
this
is
maybe
a
different
problem
outside
of
the
problem
we're
discussing
here,
which
is
more
climate
related
which
how
is
anchor
going
to
manage
these
different
fire
models
going
forward?
It's
a
different
question.
I
should
leave
that
for
another
time.
C
Yeah
just
have
a
quick
comment
on
the
natural
versus
human
fires.
I
would
just
have
to
find
the
reference,
but
jennifer
bulch
recently
published
a
paper
looking
at
the
whole
of
the
u.s
and
lightning
ignition
versus
human
ignition
and
how
humans
expand
the
season
of
the
fires,
and
so
she
used
the
u.s
forest
service
fire
program
analysis
by
a
current
database,
and
that
has
information
about
how
the
fire
started
as
well.
As
I
think
these
are
only
relating
to
fires
that
I
think
needed
to
be
suppressed.
A
A
So
we
only
have
a
few
more
minutes
before
the
end
of
the
session
and
then
the
poster
session
starts
at
3
20..
So
there
will
be
a
break
after
our
discussion
and
then
I
do
encourage
folks
to
check
out
the
poster
sessions
you'll
be
able
to
visit
individual
posters
and
interact
with
the
authors
and
so
in
the
chat.
M
Sure
I
mean
I
guess
we
don't
maybe
know
the
importance
of
them,
yet
I
think
that
most
disturbances
are
predicted
to
increase
in
extent
and
severity
and
the
area
like
sort
of
expanding
into
new
areas,
and
I'm
just
you
know,
I
think
that
the
potential
for
disturbances
to
interact
is
going
to
increase,
and
you
know
I
think,
there's
been
a
lot
of
studies
looking
individually
at
that
and
then
studies
saying
like
oh,
this
is
really
important.
M
We
need
to
look
at
it
and
I
don't
know
if
it's
really
been
put
into
a
model
yet,
and
I
think
it's
an
important
step
forward
and
I
think
if
we
can
get
the
fuels
and
interactivity
right
for
the
fire,
it
will
set
the
stage
to
then
add
in
you
know
like
what
insects
or
like
thermo
karst
and
speaking
from
from
like
a
borealis
perspective,
but
yeah.
A
Yeah,
I
I
agree:
there's
there
are
a
lot
of
uncertainties
and
a
lot
of
opportunities
and
so
improving
our
ability
to
capture
how
how
all
of
these
pieces
work
together
has
the
capacity
to
improve
all
of
these
fire
models.
There
are,
you
know,
known
uncertainties
with
emissions
factors
and
so
using
capitalizing
on
the
measurements,
from
the
campaigns
getting
a
better
characterization
of
emissions
where
we
can
attach
that
to
changing
fuels
on
the
ground
would
improve
our
emissions.
A
In
that
sense,
thinking
about
the
plume
model
and
and
yeah
following
up
on
all
of
these
connections,
I
think
will
be
important.
Moving.
A
Forward
and
sunderland
made
a
point
in
the
chat
about
drought,
induced
vegetation,
soil,
moisture
and
fire
ignitions,
and
so
I
brought
it
up
in
in
my
talk
briefly,
but
the
importance
of
live
fuels
and
the
transitions
of
live
fuels
to
becoming
an
important
fuel
source,
I
think
is,
is
a
critical
thing
as
we
move
into
this
more
drought
prone
climate,
especially
here
in
the
united
states,
but
it's
valid
in
other
systems
systems
that
have
you
know
a
dominance
of
shrubs
or
grasses.
A
J
A
Well-
and
I
think
this
comes
back
to
the
scale
of
where
the
data
is,
and
so
if
you
know
a
lot
of
this
data
is
point
measurement
or
a
field
campaign,
and
so
it's
not
necessarily
global.
But
how
can
we
take
those
field
campaigns
or
these
site
level
measurements
and
use
that
to
inform
things
in
a
mechanistic
way
so
that
we
can
handle
it
more
at
at
a
larger
scale?
J
Yeah
yeah,
I
think
there
could
be
a
lot
of
benefit
in
using
this
fate
capability
of
you
know.
Reading
in
existing
fire
counts
fire,
detections
from
satellite
and
and
then
using
all
of
the
information
in
the
land
model
to
calculate
the
emissions.
I
I'd
really
like
to
pursue
investigating
that
using
that
and
then
comparing
to
directly
to
some
field
campaigns.
J
I
think
that
that
could
be
very
beneficial
to
help
understand
all
the
driving
variables
and
and
where
some
of
the
errors
are.
A
Absolutely
well,
if
I
think,
I
think,
the
methods
of
how
we're
going
to
tackle
this
is
really
important
and
we're
just
going
to
have
to
keep
talking
with
one
another
and
staying
in
touch
on
it.
A
One
of
the
other
opportunities
that
I
see
with
with
with
fates
and
then
with
consideration
for
this
interactive
fire
is
the
importance
of
the
paleo
world
and
so
looking
at
how
things
have
transitioned
over
time,
and
so
I
know
that
I've
talked
in
the
past
with
the
paleo
community
about
just
looking
at
dynamic
vegetation
and
how
fire
has
changed
the
distribution
of
vegetation,
and
so
it
would
be-
and
I
haven't
thought
about
that,
how
you
take
how
you
go
from
paleo
to
neo
like
how
you
go
from
paleo
to
modern.
A
I
recognize
that
that's
a
it's
a
big
challenge,
and
so,
but
it
would
be
good
to
explore
that,
especially
with
large-scale
fire
histories
coming
on.
A
A
I
we're
definitely
not
going
to
solve
this
as
a
party
of
one,
and
so
please
do
get
in
touch
if
you
have
further
ideas
or
further
opportunities
that
you
want
to
bring
to
the
group,
and
so
please
do
come
and
visit
the
posters
and
yeah.
Let
me
know
if
you
have
any
follow-up
points,
I'll
save
the
chat
from
the
discussion.
You
can
also
save
the
chat
by
clicking
on
the
little
dots
down
there
and
save
it
for
yourself.
Thank
you
to
everyone
who
put
information
and
references
into
the
chat
and
for
sharing
your
ideas.