►
From YouTube: CESM Workshop: Alternative Earths
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
Okay,
great
well
welcome
everyone
to
this
cross-working
group
meeting
on
alternative
earths
and
I'm
very
excited
to
see
this
great
diversity
of
presentations
have
been
submitted
and
I'm
hoping
for
an
exciting
discussion
after
and
I
think
eddie,
and
I
thought
that
this
would
be
a
really
nice
session
to
have,
because
we
can
see
the
use
of
csm
for
non-traditional,
let's
say,
non-traditional
or
or
at
least
pushing
the
model
out
of
its
normal
comfort
range.
A
I
would
say
where
we're
asking
to
do
things
like
run
high
co2
atmospheres
or
changing
the
orbital
dynamics
and
there's
been
a
great
advancement,
I
think,
over
the
last
couple
of
years
on
making
the
model
more
transparent
in
terms
of
how
you
change
things
and
also
that
the
move
to
github
and
the
idea
that
you
can
fork
the
model
relatively
easily.
A
That
now,
as
I
think,
maybe
opened
up
the
model
to
collaboration
and
co-development
more,
and
so
I
think,
while
this
may
not
be
what
you
could
consider
to
be
under
encore's
purview
model
development,
I
think
that
the
whole
community
is
probably
could
benefit
from
maybe
a
bit
of
coordination.
In
terms
of
co-development.
A
Of
of
these
alternative
earth
applications
for
the
model,
so
looking
forward
to
the
session
with
that
I'll
just
turn
over
to
betty
who's,
going
to
introduce
the
speakers
and
keep
everyone
on
time
and
and
then
we'll
have
hopefully
some
discussion
for
the
last
half
hour
of
you,
buddy.
B
Well,
thanks
everyone
for
being
here.
It's
I'm
really
excited
that
there's
so
much
interest,
so
many
talks
that
were
sent
in
to
speak
today,
so
we're
gonna
spend
about
the
first
hour
we
have
or
hour
and
a
half
with
eight
talks
and
then
we'll
have
a
discussion
as
dan
said
to
just
kind
of
think
what
we
could
do
together.
B
What
should
be
done,
how
we
can
co-develop
this
effort
both
for
kind
of
early
earth
and
for
other
planets
or
other
moons
of
planets,
so
I
will
be
the
person
that
keeps
time.
I
will
give
you
a
maybe
a
two
minute
warning
before
the
end
of
your
allotted
time.
Just
so,
you
can
think
how
to
wrap
up
what
to
finish
with
we'll
see.
B
If
we
have
time
for
questions
my
experience
this
week
is,
it
was
often
challenging,
but
questions
can
be
put
in
the
chat
box,
and
we
ask
the
speakers
to
look
at
that.
The
chat
will
also
be
saved
and
then
we,
as
dan
said
we
may
have
a
little
bit
of
time,
also
to
discuss
some
of
the
science
during
our
discussion
section.
B
One
other
thing
is
I'm
pretty
awful
at
pronouncing
names.
So
if
I
mispronounce
your
name,
please
forgive
me
and
feel
free
to
correct
me.
If
I
do
poorly
on
your
name,
so
our
first
speaker,
an
invited
speaker,
is
colin
goldblatt
and
he's
a
professor
at
the
university
of
victoria,
so
colin
I'll,
give
you
the
floor.
C
Well,
thanks
everyone:
it's
a
real
pleasure
to
be
opening
this
meeting,
I'm
just
going
to
see
if
I
can
get
this
onto
full
screen.
First,
that's
right!
Oh
we're!
Not!
Even
on
the
first
side,
this
ain't,
so
I'd
like
to
present
some
work
today,
which
is
really
my
first
experiment
in
using
the
community
earth
system
model.
So
I'm
very
grateful
to
a
couple
of
incredibly
capable
collaborators
and
co-authors
I've
got
victoria
mcdonald,
who
was
the
student
who
did
the
model
run
who's
now
university
of
washington
in
grad
school?
C
She
was
a
uvac
undergrad
then,
and
kelly
mccasker,
who
guided
us
through
learning
how
to
to
use
the
model
and
she's
a
rhodium
group.
C
C
So
if
we
go
back
to
the
neoarchaea
and
say
2.8
billion
years
ago,
and
that's
the
most
that's
the
earliest
time,
we've
really
got
any
constraint
on
the
amount
of
co2
or,
to
be
honest,
anything
in
the
atmosphere
and
we've
need
evaded
a
fourth
thing
of
about
50
watts
per
square
meter
to
keep
the
earth
warm
in
the
abs.
C
You
know,
with
the
sun
being
dimmer
at
the
time
now
that
would
need
about
a
hundred
thousand
parts
per
million
of
co2,
we're
just
using
co2,
but
that's
more
than
the
geochemical
constraints
indicate
that
we're
allowed
so
a
little
bit
more
background.
Here's
an
artist's
infection,
I'm
going
to
call
it
of
the
evolution
of
earth's
atmosphere
and
key
things
to
point
out
is
oxygen.
We
do
know
the
history
of
gets
really
reactive.
C
It
leaves
a
great
geochemical
record
that
grows
from
part
per
million
to
percent
levels
around
two
and
a
half
billion
years
ago.
Our
other
major
constituent
nitrogen.
We
have
very
little
constraint
on
the
history
of
it.
It
was
probably
about
what
it
is
now
to
within
a
factor
of
two
or
three.
Probably
water
is,
of
course
determined,
physically
and
and
co2
is
determined
by
to
a
large
extent
the
carbonate
silicate
weathering
cycle,
so
that
probably
decreased
over
time.
C
C
I
change
low
clouds
on
the
bottom
row,
middle
clouds
on
the
middle
and
high
clouds
on
the
top,
and
we've
got
the
shortwave
long
wave
and
net
forcing
so
you
can
see
if
we
just
look
at
the
bottom
row
for
the
moment
that
at
the
moment,
low
clouds
give
us
about
30
watts
per
square
meter
negative
forcing
so
if
we
took
low
clouds
out
that
could
give
us
a
lot
of
the
forcing
that
we
need
to
resolve
the
same
young
sun.
Equally.
C
If
we
go
up
to
top
row
and
the
middle
we
can
see,
we
get.
You
know
some
forcing
from
high
clouds
and
if
we
were
to
increase
the
amount
of
high
clouds,
we
could
again
contribute
to
the
faint
young
sun
and
we're
going
to
focus
here
in
this
talk
on
the
low
clouds.
C
So
here
are
some
low
clouds
and
we're
going
to
be
talking
about.
You
know
marine
strata
stratocumulus.
These
are
you
know
just
hanging
out
in
the
boundary
layer,
so
not
very
high,
don't
have
much
with
greenhouse
effect,
but
they
really
are
very
reflective,
so
they're
good
at
reflecting
sunlight
away
that
if
we
had
less
of
this
type
of
cloud,
would
that
make
a
contribution
to
keeping
early
earth
warm
despite
the
dimmer
sun.
C
Let's
look
a
little
bit
at
the
physics
of
them
these.
These
clouds
form
up.
You
know
in
the
descending
limb
of
the
hadley
cell.
You've
got
warm
dry
air
descending
aloft
and
then
you've
got
a
a
cooler
ocean
surface,
shallow
convection
and
they're.
Also
in
the
you
know
quite
near
the
equator,
these
these
belts
of
clouds-
that's
actually
where
it
is
sunny.
You
know
up
here
in
in
the
north.
C
It
doesn't
really
matter
how
much
clouds
we've
got
because
there's
not
much
sunlight
to
reflect,
except
today
in
victoria,
it's
a
beautiful
sunny
day
here
in
paradise.
So
this
again
is
the
physics
that
we're
looking
at.
So
it's
these
clouds
here
that
we're
going
to
be
seeing
whether
we
can
get
rid
of
some
of
and
just
one
more
schematic
about
the
physics
here.
C
This
format
below
the
inversion
and
a
stronger
inversion
is
really
good
for
these
clouds,
because
it
prevents
the
entertainment
of
dry
air
from
the
free
troposphere
into
the
boundary
layer
which
would
dry
out
the
cloud
and
the
shallow
convection
is
driven
by
cloud
top
rated
of
cooling.
So
we
can
think
about
that.
If
well,
maybe,
if
we
had
more
co2
we'd
inhibit
that
cloud
top
radiative
cooling
and
that
would
be
bad
for
the
clouds
they.
We
wouldn't
drive
that
convection
as
well,
that
shallow
convection
then
one
more
schematic.
C
This
is
that
of
a
resident
paper.
I
think
and
there's
a
couple
of
metrics
that
are
used
to
to
measure
the
strength
of
the
inversion
at
the
planetary
boundary
layer.
One
is
the
lower
tropospheric
stability
and
that's
the
temperature
difference,
so
the
potential
temperature
difference
between
the
surface
and
700
hectare
pascals
and
the
other
is
the
estimated
inversion
strength
which
is
the
difference
in
wet,
bulb
potential
temperature
between
the
lifting
condensation
level
and
the
and
the
three
troposphere
you
could
measure
that
at
700
pascals.
C
So
now,
let's
get
on
to
the
modeling.
Here
we
were
interested
in
atmospheric
processors
and
you
know
had
somewhat
limited
computer
time,
so
we
were
running
with
a
slab
ocean
and
we
really
wanted
to
do
a
vey,
controlled
experiment
here.
So
all
we
changed
was
co2
and
solar
constant
and
we
changed
them.
Pairwise
such
that
we
got
the
same
mean
surface
temperature,
so
we're
trying
to
use
the
model
in
a
place
where
we
can
really
trust
that
the
model
should
be
working
well.
At
a
mean
surface
temperature
of
288
289
kelvin.
C
You
know
the
model
really
should
work,
so
we
turn
down
the
solar,
constant
and
turn
up
co2.
Just
until
it's
balanced-
and
we
did
this
both
in
cam,
4
and
cam
5.
We
were
successful
in
getting
cam
4
to
run
as
low
as
0.7
and
as
high
as
1.1
solar,
constant
relative
to
the
modern,
and
we
did
runs
with
cam
5
in
a
in
a
narrower
range
where
we
could
successfully
get
the
model
to
work.
C
So
this
is
a
controlled
experiment.
One
of
the
things
to
emphasize
is
a
lot
of
the
critical
parameterizations
change
between
cam
4
and
can
5.
the
boundary
layer,
the
sun,
the
cloud
parameterization,
especially
for
low
cloud,
the
radiation
as
well.
So
some
of
the
things
that
really
matter
are
different
between
these
two
model
versions,
and
we
did
check
on
the
the
radiation
scheme
and
co2
is
moderately
well
behaved
up
to
about
30
000
ppm
in
the
out
of
the
box
model.
You
know
maybe
even
a
little
higher.
C
C
So
I'm
going
to
show
you
a
bunch
of
figures
that
look
like
this
and
on
the
left
hand
panel.
We
have
the
modern
case
and
on
the
right
middle
panel,
we
have
the
0.8
solar,
constant
case
on
the
right
hand,
side
we
have
a
distance,
and
this
is
for
this
is
for
the
this
is
for
point
eight
solar,
constant
with
cam
four,
so
we're
just
looking
at
a
couple
of
bits
of
model
climatology
surface
temperature.
You
can
actually
see
a
difference
without
looking
at
the
difference
field.
C
You've
got
warmer
poles
slightly
cooler,
equator,
but
the
same
means
surface
temperature,
and
that's
of
course,
because
we've
got
a
stronger
greenhouse,
that's
good
at
keeping
the
poles
warm
but
weaker,
solar,
constant.
So
that's
heating,
the
equator,
less
and
then
later
heat
flux,
we're
we're
seeing
less
latent
heat
flux.
C
That's
because
we're
we're
warming
up
the
equatorial
ocean
less
and
then,
let's
go
on
to
these
stability
metrics,
which
are
telling
us
about
how
strong
our
bowed
miller
is,
and
we
see
globally
much
weaker,
lower
tropics,
develop,
lower
tropospheric
stability,
much
weaker
estimated
inversion
spends
and
if
it's
projecting
well
enough
on
zoom
in
the
stratocumulus
region.
So
here
of
california,
here
of
peru
of
namibia,
you
can
actually
see
the
biggest
differences
are
in
the
stratocumulus
region.
So
that's
that's
quite
exciting
about
getting
some
potentially
interesting
results.
C
So
here's
a
little
bit
more
of
the
climatology
here
are
some
zonal
means.
We
can
see
those
big
differences
in
the
equator,
the
pole
temperature
gradient
again,
and
then
you
know
what
else
would
I
want
to
draw
your
attention
to
some
of
the
temperature
structure
here
and
also
some
of
the
drying
of
the
equatorial
atmosphere
here
and
and
a
little
drag
of
the
three
troposphere
in
the
in
the
sub-topics
of
mid-latitudes?
C
Okay?
So
let's
go
into
our
low
cloud
changes
and
you
know
again,
this
is
just
we
were
so
excited
when
we
saw
this
result
because
it's
something
that
you
can
see
without
plotting
up
the
distance
field,
which
is
for
something
where
we're
looking
at
early
earth.
That's
just
so
great
because
it's
such
a
strong
signal.
We
might
actually
trust
it.
You
can
see
that
we've
annihilated
our
low
clouds,
the
difference
between
modern
here
and
point
eight
solar,
constant
in
the
different
field.
C
So
have
a
super
exciting
result
to
see-
and
this
is
looking
for
a
time
flight
from
0.7
solar,
constant
at
the
bottom
right
to
1.1
at
the
top
left-
and
we
see
this
just
really
well
behaved
signal
throughout
our
transect
of
solar,
constant
co2
space.
C
The
talk
very
briefly
about
high
clouds
is,
we
do
actually
see
more
high
cloud,
but
those
high
clouds
are
thinner.
So,
overall
those
two
effects
seem
to
cancel
out.
So
the
there's
not
much
difference
in
the
long
wave
cloud
forcing
that
we,
you
know
dominated
by
those
high
clouds
and
again,
if
we
look
at
that
transect,
we
really
don't
see
much
difference
in
the
in
the
high
clouds.
So
this
is
everything
all
together.
C
If
we
look
at
the
top
left
plot,
this
is
how
much
co2
we
needed
to
maintain
the
same
surface
temperature
and
if
you
can
see
on
zoom
the
gray
line,
it's
about
a
factor
of
three
less
than
we'd
expect
if
we'd
run
a
1d
model
with
with
fixed
clouds
and
to
give
credit
supply
of
art.
I
think
we've
seen
results
like
this
and
some
of
eric's
wolf
works
from
a
few
years
ago,
so
that
wasn't
entirely
a
surprise,
but
let's
really
get
at
least
these
systematic
changes
to
clouds.
C
So
if
we
go
to
the
bottom
panels
on
the
left,
we've
got
this
really
systematic
change
in
low
cloud
fraction.
The
bottom
middle,
changing
low
cloud
water
path
in
these
dark
blue
lines
and
then
bottom
right.
B
C
Minutes,
thank
you
and
then
you
know
the
physics
driving
this
is
you
know
it's
a
lot
of
it
is
this
difference
in
the
inversion
strength,
so
we've
got
a
we've
got
a
stronger
greenhouse,
so
the
clouds
come
coolers
cool
as
well,
so
that
drives
a
shallow
convection
less
and
we've
got
this
weaker
inversion.
So
it's
easier
to
get
exchange
with
the
three
troposphere
and
that's
really
important
in
drying
these
clouds
out
and
killing
them
off
early
in
earth
history.
C
So
on
time
here
are
my
conclusions.
I
think
I've
told
you
about
the
first
few,
so
some
of
my
musings
here
is,
with
this
smaller
equated
pole
temperature
gradient
quite
systematically.
That's
going
to
help
protect
us
from
glaciation.
That's
a
big
deal.
The
equitable
climate
means
that
our.
C
If
you
move
bits
of
weatherable
land
around
in
early
earth,
we
might
be
less
susceptible
to
changes
in
weathering.
Due
to
that
relative
to,
for
example,
the
cenozoic
I've
got
a
student
keegan
gamma
working
on
that
right
now
and
the
clouds
are
all
gonna
really
depend
on
the
the
base
climate
state.
So
how
much
co2
do
we
have?
You
know
what
solar
constant?
C
B
Perfect,
so
thank
you
fascinating,
so
I
think
we
should
move
on.
So
we
have
plenty
of
time
for
discussion,
but
please
put
some
questions
for
colin
in
the
chat
and
chat
colin.
If
you
could
kind
of
keep
an
eye
on
that
and
then
we
can
also
come
back
to
discussions
or
questions
during
our
discussion.
B
So
the
next
speaker
is
my
professor
shield,
a
professor
of
the
university
of
california
in
irvine.
You
pronounce
it
for
me
and
introducing
them.
D
Yes,
omaha
hi,
everybody,
okay,
let
me
go
ahead
and
share
my.
D
D
Okay,
hopefully
you're
all
seeing
that
I'm
excited
to
be
here
today.
I
want
to
thank
the
organizers
for
the
invitation
to
share
with
you
some
of
the
ways
that
my
group
has
been
using
cesm
to
explore
the
possible
climates
of
planets,
orbiting
stars
other
than
the
sun,
and
I
want
to
make
sure
to
thank
cc
bits.
My
phd
co-advisor,
who
long
ago
introduced
an
astronomer
by
training
to
the
capabilities
of
csm
as
a
powerful
tool
for
this
work.
D
Of
course,
the
only
planet
that
we
absolutely
know
of
where
life
exists
is
the
earth.
However,
we
now
have
a
growing
number
of
planets
that
may
possess
the
conditions
amenable
to
surface
liquid
water,
which
we
know
all
life
on
earth
needs
and
is
sort
of
our
criterion
or
major
criterion
for
surface
habitability.
D
As
we
look
beyond
the
solar
system
for
where
else
life
might
might
exist
and
among
the
most
exciting
prospects
include
several
planets
orbiting,
the
star
of
the
ultra
cool
star,
trappist-1
proxima,
centauri
b,
the
nearest
potentially
habitable
planet
to
us,
lhs
1140b,
for
which
we
have
both
mass
and
radius
information,
allowing
us
to
constrain
its
density
as
rocky.
D
So
then
making
the
possibility
of
oceans
more
likely-
and
we
know
of
these-
this
large
number
of
exoplanets,
largely
due
to
nasa's
kepler
mission,
which
stared
at
a
particular
patch
of
the
sky
and
took
snapshots
looking
for
planets
that
transited
in
front
of
their
stars.
D
And
even
more
recently,
we
have
a
new
planet,
toi
700d,
very
nice,
romantic
name,
which
is
the
first
earth-sized
planet
found
by
tess
in
the
habitable
zone,
which
I'll
talk
about
briefly
shortly
and
tess
is
looking
at
the
brightest
stars
in
the
solar
neighborhood.
D
There
are
30
to
100
times
brighter
than
those
sampled
by
kepler
and
that's
great
because
that
means
these
planets
are
going
to
be
able
to
be
followed
up
by
next
generation
space
based
missions.
A
few
of
them
are
shown
here,
both
missions
that
are
upcoming
and
current
mission
concepts,
which
we
hope
will
eventually
fly
in
concert
with
dedicated
efforts
on
the
ground
by
extremely
large
telescopes.
D
That
is
a
technical
term
elts,
so
that
that
dual
based
approach
of
space
and
ground-based
imaging
and
spectroscopy
is
hopefully
going
to
allow
us
to
actually
determine
whether
these
planets,
in
fact
could
host
life
and
do
host
life.
D
So
we're
entering
an
era
where
we
have
a
quality
problem.
We
aren't
going
to
have
infinite
telescope
time
to
follow
up
on
every
single,
potentially
habitable
world,
to
look
for
signs
of
life,
so
we'll
need
a
way
to
prioritize
those
planets
to
follow
up
on
and
that's
where
the
field
of
exoplanet
climatology
enters
in.
D
Historically,
what
observers
do
is
find
planets
that
orbit
their
stars
at
a
particular
region
in
space
around
the
star
which
we
call
the
habitable
zone
to
close
in
the
planet
is
in
danger
of
losing
its
entire
water
inventory
to
space
in
a
runaway
greenhouse
state.
That's
what
we
think
the
fate
of
venus
was
too
far
away
and
the
maximum
co2
greenhouse
limit
is
reached.
D
That
point
beyond
which
she
can't
continue
to
increase
co2
and
keep
warming
temperatures
to
keep
temperatures
warm
enough
for
surface
liquid
water,
but
we
only
have
to
look
at
our
own
earth.
As
an
example
of
the
fact
that
there
are
many
factors
and
capabilities,
many
factors
and
processes
that
affect
planetary,
habitability
beyond
orbital
distance,
our
planet
likely
experienced
snowball
episodes
several
times
in
its
history,
while
it
lay
very
comfortably
in
its
host
star
habitable
zone.
D
So
the
long-term
presence
of
surface
liquid
water
is
influenced
by
a
number
of
different
factors,
including
stellar
effects,
planetary
system
effects
and
properties.
The
list
is
quite
complicated
and
my
team
has
been
slowly
chipping
away
at
this
list
for
several
years
now,
with
a
particular
focus
on
surface
composition.
D
So
let's
talk
about
starlight,
I'm
showing
you
here
the
spectra
of
several
different
stars,
the.
Hopefully
you
can
see
my
cursor
here.
The
sun
is
shown
in
yellow
and
we've
got
a
star
a
little
cooler
than
the
sun,
which
we
call
a
k
dwarf.
So
the
sun
is
a
g
dwarf
a
k
dwarf
here,
an
orange
and
a
star
a
little
brighter
than
the
sun
that
f,
dwarf
and
blue.
D
D
What
we
do
is
go
into
source
mods,
src
cam,
and
we
can
actually
change
the
percentage
of
the
flux
in
the
different
wave
bands.
They're
input
to
cam
and
that's
how
we
can
change
the
host
star
from
the
sun
to
a
different
star,
a
different
type
of
star.
D
D
So
it's
a
two-band
albedo
parameterization,
where
you
split
it
at
0.7,
microns
for
visible
and
and
ir,
and
if
we
look
at
em,
dwarf
g-dwarf
and
f-dwarf
planets.
What
we
did
was
created
these
hysteresis
curves
and
the
ammo
dwarf
planet
had
a
smaller
climate
hysteresis,
so
in
lay
person's
terms,
it's
harder
to
freeze
it's
easier
to
thaw
and
when
it
thawed
it
thawed
with
a
shorter
jump
in
iceline
latitude,
where
an
ice
line
where
the
hotter
and
brighter
the
star,
the
greater
the
jump
in
iceline
and
potentially
making
endor
planets
better
abodes
for
life.
D
We
also
found
we
looked
at
the
meridianal
stream
function,
a
squatter
and
weaker
hadley
cell
on
the
m
dwarf
planet,
so
co2
and
water
vapor,
which
are
present
in
all
planet
surfaces.
All
planets
atmospheres
absorb
more
strongly
in
the
near
ir,
and
so
we
have
a
lot
of
shortwave
absorption
happening
in
the
troposphere
and
that
reduces
shortwave
reaching
the
surface.
D
That's
needed
to
drive
convection
and
we
ended
up
seeing
that
more
heats
retained
in
the
tropics,
less
distributed
to
highest
higher
latitudes
and
that
aids
m,
dwarf
planets
and
thawing
more
easily
out
of
global
ice
cover.
D
And
in
later
work
with
eric
wolf
and
others,
we
were
able
to
see
this
trend
of
solar,
stellar,
driven
glaciation,
being
more
effective
for
planets,
orbiting,
redder
stars
across
a
range
of
climate
regimes,
from
snowball
through
moist
greenhouse,
so
to
change
the
ice
and
snow,
two
band
albedos.
We
use
the
sea
ice
or
size
name
list
and
we're
able
to
change
the
ir
and
visible
albedos
for
both
ice
and
snow.
D
They
are
the
spectral
wavelength.
Dependence
is
already
present
by
default
in
sea,
ice
or
seis,
but
that's
weighted
by
this,
the
sun's
the
solar
spectrum
and
so
outside
of
csm.
I
have
a
code
that
awaits
the
two
band
albedos
by
the
particular
spectrum
of
the
host
star.
D
More
recently,
we've
done
trend
birth
diagrams
for
exoplanets.
This
is
the
first
time
that
this
has
ever
been
done,
because
we
knew
we
knew
that
an
ammo
dwarf
planet
is
warmer
than
a
planet
orbiting
a
hotter
brighter
star,
ironically,
at
an
equivalent
stellar
flux
distance.
We
knew
this
was
because
of
this
ice
albedo
effect,
as
well
as
increased
atmospheric
absorption
on
amdwor
planets,
but
we
didn't
actually
know
the
specifics
of
how
this
was
occurring,
so
how
the
energy
budgets
or
shape
different
between
these
planets,
and
that
was
important
to
quantify.
D
So
what
we
did
was
carry
out
a
range
of
simulations
for
f,
dwarf
g,
dwarf
and
m,
dwarf
planets
and
calculated
the
percentages
so
how
the
energy
the
downwelling
radiation
from
each
star
was
distributed,
was
budgeted
throughout
the
planet's
atmospheric
surface
and
back
out
again
and
for
example,
looking
at
the
f
dwarf
planet,
much
more
radiation
incoming
radiation
is
reflected
by
the
atmosphere
and
the
surface
compared
to
the
m
dwarf
planet,
and
so
we
have
this
table
in
this
paper:
energy
budgets
of
terrestrial
exoplanets
that
contains
all
of
these
percentages
for
for
different
host
star
types.
D
We've
also
looked
at
non-um
characteristic
surface
types
like
hydro
halite,
which
can
precipitate
and
bare
sea
ice
at
low
temperatures
and
created
our
own
albedo
parameterizations
for
the
formation
of
that
surface
type
in
our
gcm
studies,
and
so
hydro
halide,
as
you
can
see,
is
quite
reflective
compared
to
salt
free
water
ice,
and
this
we
we
discovered
has
implications
for
amador
planets
and
so
where
we
were
able
to
create
our
own
just
add
a
few
lines
of
text
into
the
ice
shortwave
file
that
allows
for
this
albedo
parameterization
and
when
we
do
that,
we
see
quite
a
difference
here.
D
We're
looking
at
the
control
case,
subtracted
from
having
an
albedo
parameterization
for
hydro
halide,
and
we
see
quite
a
difference,
especially
or
even
for
planets
that
are
in
the
habitable
zone
of
their
stars.
So
what
this
means
is
that
an
accurate
assessment
of
tele
of
surface
temperatures
for
exoplanets
requires
parameterizations
for
this
surface
type
into
gcm
studies,
and
we
see
this
even
stronger
this
climate
sensitivity.
If
the
planet
is
synchronously
rotating,
where
it's
got
the
same
side
facing
the
star
at
all
times,
it's
always
daytime
and
the
other
side.
D
It's
always
night,
and
we
can
do
that
in
csm
by
changing
the
orbital
parameters,
the
rotation
rate
and
these
source
mods
sr
sheets,
src,
share
and
src
cam
directories,
and
I'm
happy
to
talk
more
about
these
modifications
separately
with
anyone
who'd
like
to
start
running
their
own
simulations
of
exoplanet
climates.
D
So
I'm
going
to
stop
there.
We've
done
some
studies
with
land
as
well,
both
with
eric
and
marisa
who's
here
as
well,
and
this
work
was
spearheaded
by
my
postdoc
andrew
rushby
and
we're
able
to
actually
use
clm
to
look
at
the
effects
of
different
land,
surface
types
on
trappist-1
planets
and
the
the
short
end
of
the
take-home
point.
There
is
that
it
matters.
D
The
difference
is
50
kelvin
across
the
lowest
and
highest
albedo
and
surface
types
so
and
it
affects
cross-equatorial
energy
transport,
where
we
end
up
seeing
much
higher
surface
temperatures
on
the
planets,
with
the
lowest
albedo
surface
type,
both
due
to
atmosphere
and
surface
effects,
making
trappist-1d
the
most
likely
candidate
for
habitable
conditions
in
this
system.
D
For
the
most
part,
we
tend
to
keep
our
intrinsic
parameters
similar
to
earth,
earth
radius,
earth's
mass
surface
gravity,
and
we
change
what
what
we
can
without
really
going
too
far
into
the
hard
coding
there,
and
it's
helped
us
to
highlight
a
very
powerful
effect
on
climate,
which
is
this
effective,
stellar
spectrum
and
surface
composition.
D
So
for
more
to
see
more
about
our
work,
please
visit
the
our
publications
page
and
thank
you
to
my
funding
sources
and
all
the
collaborators
of
all
the
work
that
we've
done
over
the
past
decade
or
so
and
I'll.
Take
any
questions.
Thank
you.
B
Well,
thank
you
and
yeah.
I'm
gonna
go
to
the
publication
site.
I
know
that
and
start
reading,
so
we
should
move
on,
but
I
already
see
one
question
or
two,
maybe
in
the
chat
and
so
thank
you
and
we'll
move
on
to
their
next
talk.
Greg
cook
who's,
a
phd
student
at
the
university
of
leeds.
E
Yeah,
okay,
so
you
can
see
the
main
powerpoint
with
the
the
arrow
okay.
So
yes,
hello,
everyone
and
good
evening
from
the
uk,
I'm
very
very
pleased
to
be
able
to
give
this
talk.
It's
a
great
opportunity
to
present
and
discuss
some
of
the
work
I've
been
leading.
E
E
So
let's
dive
right
in
this
figure
is
from
the
lions,
ital
2014
paper,
and
it
shows
the
estimated
oxygenation
history
of
earth
for
the
past
four
billion
years.
We
know
that
life
has
been
present
in
at
least
three
geological
eons,
the
arcane,
the
protozoa
and
the
phanerozoic
and
oxygen
was
very
low.
E
During
the
rkm,
but
then
we
had
the
great
oxidation
event
about
2.4
billion
years
ago
and
oxygen
rose
to
near-modern
levels,
but
then
we
had
this
dip
for
quite
a
long
time
where
oxygen
levels
were
between
10
to
the
-4
and
10
to
the
minus
one.
E
The
present
atmospheric
level
before
rising
again
during
the
neo-protest
oxidation
event
and
then
during
the
phanerozoic
levels,
have
remained
relatively
constant,
although
they
have
fluctuated
so
by
extension,
we
can
think
that
terrestrial
exoplanets
rocky
planets
around
other
stars
will
also
go
through
geological
evolution,
just
like
the
solar
system,
terrestrial
planets
have
so.
The
motivation
behind
our
simulations
is
to
better
understand
earth's
history
and
the
potential
state
of
the
terrestrial
exoplanets
that
we
may
characterize
in
the
future.
E
E
Several
other
species
were
held
at
a
constant
mixing
ratio
at
the
surface,
but
here
we've
just
changed
oxygen,
so
we
start
with
the
pre-industrial
atmosphere
where
oxygen
is
21
by
volume
at
the
surface,
but
this
decreases
at
the
mesopause
and
then
we
change
oxygen
to
the
values
shown
here,
which
is
150
percent.
The
present
atmospheric
level,
a
possible
high
during
earth,
history
all
the
way
down
to
a
thousand
or
0.1
percent
the
present
atmospheric
level.
But
we
we
really
wanted
to
so
better
simulate
protozoa
conditions
too,
after
the
rise
in
oxygen.
E
So
we
have
variations
on
the
one
percent
pl
simulation.
We
have
two
methane
flux
simulations
where,
instead
of
fixing
the
lower
boundary,
we
had
a
flux.
E
So
each
of
these
simulations
was
run
out
until
roughly
into
annual
equilibrium
was
reached
for
the
last
eight
years
of
of
the
simulation,
and
then
we
actually
I'm
going
to
present
time
average
means
for
the
last
four
years
of
each
simulation.
E
So
here
we
have
the
oxygen
ozone
and
temperature
profiles
on
the
x-axis,
with
atmospheric
pressure
on
the
y-axis,
the
oxygen
profile
here,
with
all
the
colours
now
from
black
and
then
integrate
to
to
go
to
the
150
percent.
Then
the
greens
50
to
5
and
the
blue
is
one
to
0.1
the
present
atmosphere,
level
of
o2,
so
yeah.
E
This
oxygen
profile
basically
affects
the
ozone
profile,
because
oxygen
has
a
large
influence
on
the
production
of
ozone,
with
oxygen,
photosys
being
the
main
source
of
odd
oxygen
and
then,
as
ozone,
absorbs
ultraviolet
radiation
and
heats
the
stratosphere.
We
see
that
as
oxygen
and
thus
ozone
reduces
we
actually
have
a
much
colder
stratosphere
and
essentially,
when
we
get
to
one
percent,
the
present
atmospheric
level.
E
The
stratosphere
basically
disappears-
and
this
is
good
news
because
it's
been
seen
in
previous
1d
modeling,
so
the
lack
of
ozone
and
oxygen
allows
increased
uv
to
reach
lower
in
the
atmosphere
all
the
way
to
the
surface,
and
this
results
in
increased
photolysis
of
important
greenhouse
gases
such
as
water,
vapor,
methane,
carbon
dioxide
and
nitrous
oxide.
So
methane
in
particular,
is
rapidly
destroyed.
E
First,
at
the
surface
because
of
increased,
oh
and
then
in
the
trip
just
above
the
tropopause
because
of
increased
photolysis
as
oxygen
and
ozone
reduce
the
water
vapor
content
entering
the
stratosphere
is
much
lower
because
of
a
colder
trip
pause.
So
this
middle
atmospheric
loss
of
greenhouse
gases
is
really
important
and
it
has
implications
for
detecting
these
molecules
in
exoplanet
atmospheres,
particularly
in
transits
as
I'll.
E
So
how
protective
are
the
ozone
states
that
we
predict
here?
We
have
the
total
ozone
column
plotted
on
earth
for
each
oxygen
case.
The
ozone
column
is
simply
the
total
number
of
ozone
molecules
above
our
heads
per
unit
area.
Note
the
difference
in
the
scales,
so
we
have
a
peak
of
375
dobson
units
in
the
pre-industrial
atmosphere,
and
that
goes
all
the
way
down
to
a
peak
of
24
dobson
units
in
the
0.1
pl
atmosphere.
E
So,
in
this
pre-industrial
case
we
can
see
that
the
minimum
ozone
column
is
actually
found
across
the
equator,
as
the
brew
dobson
circulations
distributes
ozone
to
high
latitudes,
but
this
changes
as
oxygen
is
reduced
and
essentially
the
distribution
is
no
longer
the
same.
We
don't
have
quite
the
whole
equatorial
minimum
across
this
whole
band
here
that
breaks
down,
and
there
is
no
longer
this
distribution
of
ozone
to
high
latitudes
and
also
notice
that
the
magnitude
of
ozone
does
change
quite
drastically
as
we
reduce
oxygen.
E
But
how
does
this
compare
with
previous
work?
Well,
this
plot
shows
how
the
o2
concentration
varies
with
the
ozone
column
it
produces.
So,
as
oxygen
increases,
the
ozone
column
generally
increases,
including
for
most
previous
work
too.
So
the
previous
work
is
generally
1d
modeling.
Apart
from,
we
have
one
3d
model
here,
which
is
the
wear
tile,
2017
rocky
3d
model
and
from
their
work,
and
that's
this
dotted
line
here.
E
Biological
and
astrobiological
research
has
generally
accepted
that
when
we
get
to
one
percent,
the
present
atmospheric
level
of
o2,
we
have
a
substantial
ozone
layer
which
is
able
to
shield
the
life
and
the
biosphere
from
uv
radiation,
and
this
is
indicated
by
this
orange
shaded
box.
E
Here,
however,
our
results
suggest
that
this
assumption
is
problematic,
for
both
historical
habitability
and
for
exoplanet
habitability,
and
this
is
because
we
need,
between
five
and
ten
times
the
amount
of
oxygen
to
get
to
the
same
uv
protective
state
as
estimated
by
previous
research
with
the
amount
of
ozone
that
is
needed.
So
we
think
that
the
uv
levels
in
the
literature
may
have
been
underestimated
for
much
of
the
past
2.4
billion
years
and
that
could
have
implications
for
complex
animal
life,
particularly
when
the
ozone
layer
is
significantly
perturbed
by
volcanic
eruptions
or
flares.
E
E
Oh
okay,
so
between
1.8
billion
years
ago
and
0.8
billion
years
ago,
there's
no
evidence
of
widespread
glaciation,
but
the
sun
was
between
13
and
6
fainter
as
colin
discussed
earlier.
So
how
is
this
possible?
Well,
previous
research
has
invoked
high
methane
and
carbon
dioxide
to
warm
the
earth
under
a
faintest
sun,
but
is
increased
methane
during
the
mesoprotozoic
possible.
Well,
we
don't
think
so.
E
Recent
research
has
estimated
that
during
the
protozoa
mean
influxes
might
have
been
a
lot
lower
than
present
day,
perhaps
even
lower
than
10
of
the
pre-industrial
emissions,
and
here
we
have
a
figure
that
shows
the
method,
mixing
ratio
and
lifetime
at
these
fluxes.
So
the
lifetime
is
substantially
reduced
here
and
then
at
the
fluxes.
E
We
also
see
that
the
methane
mixing
ratio
is
really
really
reduced
down
to
very
small
values,
so
the
protozoa
fitness
and
paradox
does
in
this
sense
remain
very
much
alive
and
these
results
make
it
more
difficult
to
solve
I'll
just
quickly
skip
ahead
to
some
transmission
spectral
results,
so
as
oxygen
reduces,
we
have
less
ozone,
and
this
basically
means
that
all
the
molecules
across
the
board
here
indicated
from
0.1
microns
to
20.
E
E
If
we
focus
on
this
ozone
feature
here
and
we
go
and
look
at
the
next
generation
telescope
luvoir,
which
has
been
proposed
and
we
place
a
planet
30
parsecs
away,
we
do
25
transits,
then
we
find
that
we
can
detect
ozone
in
the
pre-industrial
atmosphere,
but
not
in
the
one
percent
pl
atmosphere
and
we
need
a
much
much
bigger
telescope,
maybe
even
100
meters
in
diameter
to
all.
E
We
might
need
to
observe
for
many
many
more
years
so
I'll,
just
script
to
future
work,
so
observational
predictions
and
recommendations
for
next
generation
telescopes.
That's
what
we
want
to
do
with
all
this
work,
not
just
lava
but
the
european
extremely
large
telescope
and
some
of
the
other
elts,
and
we
also
want
to
make
wacom
6
much
more
flexible,
and
this
will
involve
big
jobs
like
modifying
the
chemistry
and
rotative
transfer
in
the
model
so
yeah.
E
I
have
skipped
some
things
out
there,
but
if
you'd
like
to
see
extra
details
of
the
work,
then
we
have
an
archive
preprint,
which
is
due
to
be
updated,
otherwise,
yeah
send
me
a
question
in
the
chat
or
send
me
an
email.
Thank
you
very
much.
B
Oh,
thank
you
greg.
So
again
put
your
questions
in
the
chat
and
we'll
move
on
here
to
the
next
talk,
which
is.
G
G
Sorry
deep
night
trying
to
I
just
like
get
up
watch
just
about
me
to
go
okay.
Could
you
hear
me
and
let's
say
that
again.
G
Okay,
okay,
hi
everyone,
I'm
jin
yang
from
pika
university.
Today
I
will
show
using
cam
system
3
csm
for
exo
planet
climate
simulations.
G
G
Because
the
mass
of
m-style
is
much
lower
than
j-star
and
the
temperature
of
m-star
is
much
lower,
so
they
have
the
zone
of
m.
Star
is
much
more
close
to
the
whole
star,
so
planets
in
the
habitable
zone
of
m
star
will
have
will
have
much
more
higher
tide
force.
G
So
the
tide
force
can
push
the
system
in
tight,
lock,
orbits
such
as
one
to
one
dialogue
orbits.
It
means
that
the
rotation
period
is
equal
to
the
orbital
period,
so
the
day's
side
is
permanent
and
also
the
night
side
is
also
permanent.
So
only
one's
hemisphere
of
the
planet
can
receive
standard
energy
from
the
host
star
for
simulation,
the
climate
of
tide,
local,
rocky
planets.
G
The
process
is
similar
to
climate
stimulation
of
earth's
winners
or
mass,
but
for
different
atmosphere.
Mass
atmospheric
compositions
clouds
connections,
ocean
depths
and
so
on.
G
G
This
is
this
is
one
example
of
the
simulator
surface
temperature
of
thailand
rocky
planet.
G
There
is
at
the
most
heat
transport
from
the
day
side
to
a
nice
side
of
from
the
sea
surface
temperature
from
the
surface
air
temperature.
We
can
know
that,
because
led
side
is
warmer
than
the
night
side,
so
the
surface
air
pressure
on
this
side
is
lower
than
that
on
the
nice
side.
So
this
surface
air
pressure
contrast,
can
join
strong
convergence
from
the
night
side
to
the
day
side
and
the
meanwhile,
because
the
sub-cell
region
is
warm,
so
it
can
promote
strong
connection.
G
So
the
most
vasculation
is
that
over
the
substrate
region
get
there.
There
is
young
convergence
and
connection,
so
clouds
can
form
over
the
substrate
region.
G
When
the
air
flows
close
to
the
tropos,
it's
they
flow
to
the
night
side
or
the
polar
region
of
the
day
side
and
then
downward
on
the
nice
side
and
flow
back
to
the
the
substrate
region
near
the
surface,
because
the
clouds
are
right
at
the
region
where
the
of
flux
is
the
maximum,
so
the
clouds
can
effectively
cool
the
service
and
stabilize
the
climate.
G
So
all
this
simulation
supports
that
on
dialog
planning,
the
subset
region
should
be
called
by
already
sing
clouds,
but
because
different
models
use
different
cloud
schemes
and
different
connection
schemes.
So
the
simulated
cloud,
water
pad
water
path
and
cloud
fraction
have
large
difference
between
different
models.
G
Besides,
the
effect
of
clouds
ocean
can
also
influence
the
climate,
at
least
so.
These
patterns
show
the
surface
air
temperature
and
the
ocean
surface
currents,
as
you
can
see
that
the
the
hot
spot
horizontal
spot
is
not
at
the
substrate
in
the
center
of
the
panel,
but
on
the
east
and
the
south
north
of
the
sub-region.
G
This
is
due
to
that
in
the
ocean.
There
are
raspberries
and
the
cow
waves,
the
raspberry
will
transport
heat
to
the
west
and
the
polar
region
of
the
south
sub-zero
point
and
the
carbon
will
transfer
heat
to
the
east
of
the
substrate
region.
So
the
spatial
pattern
is
like
a
monsoon
gear
mode.
G
G
Meanwhile,
because
the
ocean
can
transport
heat
from
the
substrate
region
to
the
nice
side,
the
ice
only
in
several
meters
or
tens
of
meters,
this
synchronous
is
close
to
the
singleness
of
sea
ice
in
the
polar
region
of
of
our
earth,
so
the
ocean
hitch
transport
limited
the
growth
of
say,
ice
to
much
zinc.
G
Besides
of
the
the
waves,
there
are
als,
also
a
strong
current
in
the
deep
tropical
region
of
dialogue
planets
the
we
call
it
oceanic
super
rotation,
because
the
ocean
currents
is
from
west
to
east
and
it
can
reach
zero
meters,
so
it
is
actually
rotate
faster
than
the
solid
earth,
a
solid
plan
part
of
the
planet,
so
it
can
be
called
oceanic
shoe
rotation
in
both
system,
three
and
ces,
and
we
can
found
a
strong
current
in
the
equator
region.
G
At
least
the
metallics
of
this
oceanic
subrotation
is
through
raspberry
wave,
which
transports
moment
from
polar
region
to
the
deep
tropical
region.
Besides
of
the
ocean
sea
ice,
did
you
hear
two
minutes?
Okay,
okay
and
besides
of
the
ocean
sea
ice
tubes
can
also
influence
the
climate.
G
In
this
slide,
I
want
to
show
how
ci
chief
influence
the
climate
of
thailand
planets.
We
want
to
know
how
sea
ice
chips
can
push
the
system
from
an
eyeball-like
state
to
a
snowball-like
state
and
the
list
video
show
the
results.
G
The
left
panel
is
for
without
ci
strip,
and
the
right
panel
is
for
with
cis
drift.
You
can
see
that
once
the
egypt
is
included
in
assimilation,
the
planet
and
into
a
snowball
state,
all
the
surfaces
are
covered
by
ice
and
snow
for
future
work.
I
think
there
are
two
parts
are
very
important.
G
One
is
about
radio
transfer,
because
the
atmospheric
competition
temperature
and
pressure
on
exoplanets
are
very
different
from
us,
so
the
radio
transfer
should
be
extend,
expand.
These
two
panels
show
the
compilation
of
radio
transfer
model
between
g7
and
lamellar
models.
You
can
see
that
in
both
short
wave
and
long
wave,
there
are
very
large
difference.
G
I
think
eric
wolf
will
talk
more
about
this.
The
second
part,
I
think
the
most
important
part,
is
about
the
convection
cloud
schemes,
because
there
are
no
observation
yet
so
cloud
resource
model
will
be
very
useful.
This
is
example
group
that
shows
the
near
global
cloud
resource
simulation
on
thai
local
planets.
G
You
can
see
that
the
model
shows
that
in
a
subset
region,
there
are
another
cloud
forms
people
convention
clouds,
so
these
confirm
that
gcm
results
and
also
provide
a
benchmark
for
future
gcm
cloud
scheme
development.
G
Okay,
this
is
the
end.
The
summary-
and
this
is
the
end
of
my
talk,
thank
you
for
your
listening,
so
any
question.
B
So
please
put
your
questions
in
the
chat
again
to
leave
time
for
the
discussion
at
the
end,
but
thank
you.
It's
fascinating!
So
we'll
move
on
to
our
next
talk,
which
is
a
wang,
yin
kang,
a
postdoc
at
mit.
H
H
Yes,
all
good
yeah,
okay,
perfect!
So
thank
you
so
much
for
having
me
here.
It's
a
great
opportunity.
H
I
would
like
to
share
some
of
my
work
on
high
obliquity
planets,
so
it's
a
different
type
of
planets
than
what
june
young
just
talk
about
it
have,
and
so
the
obliquity
is
what
brings
us
the
season
so
for,
for
example,
on
earth
we
have
an
obliquity
of
23.5
degree,
and
that
brings
us
a
very
modest
system
of
variability,
but
on
higher
higher
obligated
planets,
the
rotating
axis
of
the
planet
is
almost
the
flipped
over
and
that
will
lead
to
extreme
climate,
seasonal
seasonal
cycle
and
also
making
the
polar
regions
warmer
than
the
guitar
region.
H
So
that's
the
key
feature
of
these
type
of
planets.
So
I'm
sorry,
I
myself
can't
see
the
title
now.
Also.
Let
me
move
this
thing
around
a
little
bit.
Okay.
So
the
reason
why
we
are
interested
in
these
high
obligation
plan
is
partially
because
that
there's
a
lot
of
them
in
the
universe
and
partially
because
it's
connected
with
paleoclimate
so
shown
here
is
the
delta
o18
signal.
H
After
doing
a
fourier,
transform
for
and
and
and
and
this
period
of
time
is
speaking
from
the
early,
it's
picked
from
the
early
plasticine.
So
what
we
find
is
that
the
object
and
the
frequency
of
the
obliquity
variations
stands
out
from
over
all
the
other
variabilities,
and
that
indicates
that
obliquity
could
be
a
very
strong
factor.
That
controls
our
climate
and
also
very
interesting.
H
Experiments
that
have
done
by
jenkins
and
other
people
early
on
is
that,
at
the
same
keeping
all
the
other
parameters
exactly
the
same,
but
just
changing
the
obliquity.
The
overall
climate
can
change
significantly
and
it
turns
out
that
high
obliquity
planets
with
the
same
insulation
can
be
a
tens
of
degree
warmer
than
the
low
obligation
equivalents,
and
that
has
been
proposed
to
be
an
potential
solution
to
the
phanyonsome
paradox.
H
Whether
or
not
this
is
relevant
to
the
over
earth's
history.
It's
not
what
I
want
to
focus
on
today
and
but
what
instead,
what
I
am
I
want
to
talk
about
is
why,
and
these
high
ability
planets
are
generally
water
than
the
low
obligatory
landers.
So
one
mechanism
that
have
been
proposed
before
is
the
ice
arbiter
feedback.
H
They
have
because
the
high
up
liquid
planets
have
strong
seasonal
cycles,
so
all
latitudes
receive
strong
and
direct
radiation
from
their
star
during
some
period
of
the
of
the
year
and
therefore
it's
hard
to
form
a
permanent
astro
and
that
will
lead
to
low
ice
coverage,
a
lower,
arbitral
and
therefore
generally
warmer
climate.
So
we
will
see
like
carefully
examine
whether
that
is
indeed
what
was
happening.
H
So
what
we
do
here
is
we
take
axocam
thanks
eric
for
building
this
wonderful
model
up,
and
we
we
put
some
water
on
the
ground
and
we
put
some.
We
fill
the
atmosphere
with
nitrogen
and
then
we
gradually
increase
the
insulation
for
a
low
obliquity
scenario
and
for
a
higher
liquid
scenario
so
shown
on
the
right.
H
Is
the
temperature
showing
the
dash
curves
as
a
function
of
insulation
for
high
abdicity
in
red
and
low
objective,
the
temperature
difference
between
the
high
and
low,
because
it's
shown
by
the
solid
curve
with
y-axis
on
the
left
and
as
you
can
clearly
see,
no
matter
what
insulation
and
it
is
and
what
how
much
ice
is
is
on
the
surface.
H
The
higher
obliquity
planets
is
always
warmer
than
the
law
of
the
equivalent,
and
that
is
kind
of
a
general
feature,
even
even
when
there's
no
eyes
at
all.
So
what
so?
So
then
our
question
is
what
causes
that
so
we
will
take
one
specific
insulation
and
study
that
case
and
and
trying
to
figure
out
what
makes
the
high
objective
warmer
so
shown.
H
Here
is
the
the
anime,
the
enemy
temp
surface,
temperature
for
high
and
low
obliquity
as
a
function
of
latitude
and
mark
the
global
mean
temperature
using
the
in
by
using
the
dashed
lines.
You
can
see
that
I'm
sorry
so
so
so
the
before.
So,
if
you
go
back,
we
will
find
that
the
high
of
liquidity
is
warmer
than
the
log
equivalent
by
15
kelvin.
So
now
here
the
first
thing
we
examine
is
the
is
our
b
of
feedback.
H
So
we
turn
off
by
setting
that
our
eyes
might
be
able
to
be
the
same
as
the
ocean,
and
what
we
find
here
is
that
the
higher
ability
is
still
warmer
than
the
low
of
the
equivalence
by
90
degree.
This
is
smaller
than
before,
but
still
most
of
the
temperature
difference
still
there.
So
there
must
be
some
other
reasons
makes
the
high
obligated
planets
keeps
the
hybrid
planets
warmer
warm.
So
what
is
that?
So?
The
next
thing
we
examine
is
the
cloud
derivative
and
effect.
H
We
turn
that
off
by
overriding
the
cloud
relief
forcing
and
what
we
find
is
that
this
high
ability
and
low
ability
temperature
difference
disappears,
and
that
tells
us
this
cloud
play
a
role
in
this
warmness
and
what
happened.
So
we
examined
the
cloud
coverage
and
everything
in
these
by
mounts
by
mounts
from
the
law
between
the
highway
experiment,
and
what
we
find
is
that
in
the
lower
particular
situation,
there's
no
small
cycle
and
therefore
the
star
is
always
above
the
equator,
and
that
is
where
the
temperature
is
warmest.
H
In
the
surface
and
there's
very
strong,
convections
and
then
dcd,
that
is
locating
right
at
the
equator.
That
forms
the
cloud
which
will
reflect
all
the
sunlight
direct
back
to
the
space,
but
on
a
high,
absolutely
situation,
especially
during
the
may
season,
where
the
opposite
decision
for
the
southern
hemisphere,
one
is
when
the
sun
is
moved
to
the
around
60
degree.
H
North
does
the
ocean
surface,
there
is
still
very
cold
and
therefore
the
overall
atmosphere
circulation
is
subsiding
there
and
that
will
clear
up
the
sky,
allowing
this
this
sunlight
to
reach
the
surface,
and
that
would
lead
to
a
so.
I
summarize
here
in
the
high
ability
situation
there,
the
the
there's
more
radiation,
which
is
the
surface
because
the
because
of
the
low
cloud
albedo.
H
So
we
test
this
hypothesis
by
just
instead
of
turning
off
the
cloud
radio
effects
completely.
We
just
shut
off
the
seasonal
cycle,
and
what
we
find
is
that
the
temperature
difference
between
high
and
low
obliquity
planet
is
small,
and
also
this
even
becomes
opposite
than
before,
so
that
now
the
law
objective
is
warmer,
and
that
indicates
that
is
indeed
this
seasonal
cycle
and
coupled
with
the
cloud
really
fat
cause
leads
to
this.
The
warmness
on
pipeline
replanners.
H
The
next
question
that
so
given
that
the
higher
liquidity
is
warmer
than
the
law
of
pt
equivalents,
the
next
question
that
interests
us
is
does
that
have
an
impact
on
the
habitability,
whether
that
will
make
the
high
liquidity
planets
more
vulnerable
to
high
insulation.
So
the
and
the
question
to
understand
this
question.
We
we
need
to
see
that
we
need
to
study
the
stratosphere,
because
that
determines
how
quickly
the
water
vapor
will
water
from
that
planet
will
be
will
escape
to
the
space.
H
And
if
that
happened
too
quickly,
the
surface
won't
have
any
ocean
labs
and
that
and
and
alternate
and
alter
and
eventually
will
becomes
inhabitable
so
shown
here
is
a
specific
humidity
at
10
millibars
from
the
high
and
the
low
obliquity
experiments
rather
than
the
black
respectively,
and,
as
you
can
see,
they
follow
com,
very
different
paths
and
in
most
of
the
installations
the
higher
obliquity
planets
strongest
stratosphere
is
about
three
orders.
Two
or
three
orders
magnitude,
more
moist,
moisture
than
the
law
of
the
equivalent
and.
I
J
H
At
the
spatial
pattern
on
the
right
we
will.
We
can
see
that
the
high
liquidity
is
very
moist
and
highlighted
and
low
up,
because
they
have
the
opposite
kind
of
pattern.
So
we
so
in
this
experiment.
We
can
turn
on
and
off
different
feedbacks
and
surface
and
control
the
seasonal
cycle,
etc.
To
figure
out
why
why
this
difference
happened
and
we
figure
out
there
are
three
mechanisms
causing
causing
this
difference.
H
The
first
is
that
the
high
obliquity
planet
is
generally
warmer
than
low,
objective
equivalence
and
that
increase
the
saturated
water,
water
vapor
pressure,
and
the
second
is
that,
because
there
is
a
seasonal,
yes
one
minute:
okay,
because.
B
H
The
season
okay,
so
because
there
is
a
seasonal
cycle,
the
because
there
is
a
seasonal
cycle,
so
the
part
the
polar
regions
on
the
on
the
higher
liquid
planets
can
reach
extremely
high
temperature,
which
cannot
be
achieved
in
the
law
of
the
equivalence,
and
that
is
the
second
contributor.
H
And
the
third
reason
is
that
the
is
because
of
some
dynamic
configuration
so
shown
here
is
on.
The
left.
Is
the
temperature
profile
for
the
higher
obliquity
on
the
top
and
low
of
degree
on
the
bottom
and
what
happens
that
in
the
low
optical
situation?
Because
the
production
circulation
is
pumping
the
water
vapor
into
the
strongest
sphere?
So
you
have
to
go
through
these
cold
trap
regions
above
the
equator
and
that
and
in
that
process,
all
the
water
vapor
will
be
freezed
out.
H
You
know,
whatever
laughs
in
the
stratosphere
is
very
dry,
but
in
the
higher
potential
situation.
First
of
all,
the
low
the
coal
trap
is
not
as
strong,
but
also
it's
still
located
at
the
equator.
So
when
the
when
the
water
vapor
inject
enters
the
stratosphere,
it
doesn't
necessarily
need
to
go
through
this
coal
trap
filter
and
that
will
that
allows
the
water
vapor
to
be
high
in
the
high
ability
situation.
H
So
clearly,
I
don't
have
the
time
to
go
through
the
dynamics,
and
so
I
will
just
highlight
some
some
take
homes
from
from
this
part.
So
we
also
study
the
general
circulation
and
dynamics
on
how
the
planet,
because
of
the
reverse,
the
meridian
temporary
gradient.
H
There
are
a
few
consequences,
the
first
of
all
the
way
that
the
roughly
wave
dynamics
becomes
more
concentrated
and
constrained
near
the
surface,
and
and
also
that,
makes
that
that
changes,
the
hotly
cell
structure
so-
and
it's
very
interesting
that
we
find
when
the
reverb,
when
the
temperature
grid
is
reversed.
H
Instead
of
having
atmosphere
sinking
at
the
equator,
we
still
have
upward
motions,
and
that
is
have
to
do
with
the
wave
pumping
momentum
from
the
equator
to
the
to
the
to
the
midlight
use
and
the
the
circulation
compared
to
the
normal
situation
is
much
weaker
and
also
we
find
that,
as
we
gradually
increase
the
insulation
these
over.
H
This
highly
circulation
and
other
overturning
cells
in
the
atmosphere
changes
its
characteristics
so
in
the
low
low
insulation
scenario
and
the
the
overturning
circulation
is
and
is
pumping
at
air
upwards
in
the
equator,
but
in
a
very
high,
very
high
insulation,
when
the
climate
is
very
warm
and
the
thermal
component
dominance
and
that
you
have
sinking
motion
at
the
equator.
H
So
these
can
you
see
this
really
changing?
Have
some
observation
impacts?
Okay?
So
that's
all
thanks.
Thank
you.
B
Obvious
we
need
another
meeting.
We
have
so
much
interesting
things,
so
our
next
talk
is
thomas
oushey
who's,
a
planetary
scientist
at
nasa
goddard
space
flight
center.
K
K
Okay,
so
hi
everyone
yeah,
I'm
thomas,
I'm
a
planetary
scientist
at
godard
and
today,
I'm
going
to
discuss
about
all
model
intercomparison
ray
will
be
very
useful
for
exoplanets,
and
so
first
of
all,
you
may
have
noticed
that
every
time
we
discover
a
new
planet,
a
rock
exoplanet
inside
the
habitable
zone,
it's
really
a
little
bit
the
rush
from
the
community
to
be
the
first
to
simulate
the
potential
climate
and
habitability
of
this
planet.
So
it's
completely
fine,
it's
really
great.
K
There
is
here
also,
the
k3d
is
showing
a
dynamic
ocean
which
is
different
from
the
other
model,
but
anyway
we
have
a
lot
of
parametrization
that
can
again
provide
different
results,
and
so
the
question
is
like:
how
do
we
know?
Which
model
is
closer
to
the
truth
without
having
data?
Yet
because
we
need
to
to
probably
wait
for
the
james
webb
space
telescope
to
give
us
some
information
about
those
planets
and
so,
but
before
we
count.
K
But
what
we
can
do
right
now
is
to
compare
the
model
between
each
other
to
quantify
differences
between
them
and
so
to
bracket
the
prediction
between
mean
and
max
value
between
various
gcms
and
therefore
providing
to
observer,
more
robust
and
less
model
dependent
prediction
and
data
interpretation,
and
so
that
kind
of
approach
has
been
done,
of
course,
or
for
many
decades
by
the
earth
science
community,
and
I
can
highlight
the
the
the
semi
the
couple
model
intercomparison
project.
K
That
is,
of
course,
very,
very
useful,
and
here
I'm
just
showing
that
we
have
a
lot
of
different
models
that
are
on
board
semi.
This
is
just
showing
the
equilibrium
climate
sensitivity
on
the
y-axis.
It's
not
really
important.
K
I
just
wanted
to
show
that
we
have
many
models
and
well
the
the
the
output
slightly
different
as
well,
and
so,
of
course,
semit
was
also
used
and
still
used
to
predict
the
future
climate
of
earth
and
what
is
very
interesting
and
when
we
have
mini
models
that
can
have
a
model
range
where
we
can
again
bracket
the
the
prediction
and
be
less
model
dependent,
and
so
we
had
interconversion
for
earth,
but
also
in
the
solar
system.
K
We
have
several
photographing
mars
and
also
venus,
but
one
I
wanted
to
highlight
here
is
a
very
interesting
one
for
titan,
which
has
been
done
by
reitel
in
2019,
and
the
objective
was
to
prepare
the
the
nasa
dragonfly
mission,
and
so
three
gcm
were
on
board
the
epsil
titan,
the
colon
titan
and
the
titan
atmospheric
model,
and
so
from
this
intercomparison.
There
are
few
key
results
like
in
general.
They
have
good
agreements
on
atmospheric
structure
and
circulation
that
are
created
by
the
those
three
models
and
also
for
the
methane
content
at
low
altitude.
K
But
there
are
some
disagreements
versus
the
data
on
the
winds
over
the
boundary
layer.
I
mean
the
winds
are
quite
difficult,
sometimes
to
simulate
and
also
have
different
boundary
layer
parametrization
between
the
model,
but
overall
they
give
similar
simulation
environment
that
observe
with
the
wigan's
mission,
except
for
the
methane
cycle,
which
is
highlighted
by
this
plot.
Here,
where
we
see
the
methane
precipitation
and
especially
the
epsl
model
here,
show
much
more
precipitation
of
net
and
any
other
model
and
also
at
different
different
places.
K
So
at
the
dawn
of
jwst,
we
need
to
start
inter
model
comparison
for
exoplanets
as
well,
and
so
the
first,
inter
comparison
that
has
been
done
were
by
a
junion
that
just
presented
a
few
minutes
before
me
and
the
first
one
was
about
a
1d
model
and
a
difference
in
the
weather.
Vapor
relative
translator.
K
K
But
if
you
put
a
planet
around
a
star
which
is
much
colder
much
more
red,
then
you
need
to
better
resolve
the
infrared
water
absorption
bands,
and
that
has
been
an
important
finding
for
this
inter
comparison
and
then
the
model
has
been
adapted
to
that
then
june
and
colleague
follow
up
on
2019
with
a
gcm
intercomparison
this
time
for
a
teddy
lock
planet
and
they
found
that
between
the
different
gcms.
K
There
are
discrepancies
between
20
to
30
kelvin
of
surface
temperature.
That
was
mostly
due
to
differences
in
atmospheric,
dynamic
clouds
and
relative
transfer.
It
was
mostly
clouds
because
they
are
quite
difficult
to
pharmacize
in
ngcs,
so
in
our
case
algoda
we
had
and
godard
elsewhere.
We
had
the
thai
the
trappist
habitable
atmosphere
in
comparison,
and
the
objective
here
was
really
to
focus
on
on
few
cases
and
see
how
different
gcms
would
predict
alien,
atmospheric
outputs.
That
may
create
different
observations
for
fortune
threads.
K
We
had
one
planet
which
is
trappist-1
e,
because
it's
probably
the
best
target
for
jwst
atmospheric
characterization
and
four
gcm
onboard
rocky
3d
from
nasa
geese
of
exocar
by
rick
wolf.
So
it's
a
csm
1.2.1
adapted
for
exoplanets
by
eric
the
unified
model
in
the
met
office
and
the
lmd
generic
model
in
france,
and
we
have
four
different
cases
on
the
left.
K
We
have
two
benchmark
cases
where
we
have
no
water
on
the
surface
at
all,
we
have
a
modern
atmospheric
composition
and
a
one
bar
co2,
and
we
look
at
those
cases
in
order
to
check
differences
in
relative
transfer,
planetary,
binary
layer
and
dynamical
core,
and
then
we
move
on
on
the
same
atmospheric
composition
on
the
right.
But
this
time
we
fully
cover
our
planet
with
an
ocean,
and
so
we
look
at
the
moist
convection
and
clouds
and
all
those
differences.
K
And
so
we
have
published
in
2020
the
protocol
for
this
inter
comparison,
and
now
we
are
about
to
write
a
trilogy
or
three
papers
that
will
really
debrief
this
into
comparison
and
actually
more
to
come
so
to
quickly
show
some
some
of
the
results
here.
What
you
can
see
are
surface
temperature
maps
that
are
centered
at
the
substellar
point
for
each
of
the
gcm,
so
you
have
lmdg
on
the
top
left
and
exocam
rocky3d,
and
so
above
the
temperature
map.
K
So
what
we
can
see
quickly
there
that
well,
they
agree
roughly
between
each
other,
but
you
can
see,
for
example,
it's
much
colder
at
some
location
there
and
there,
and
so
the
the
temperature
distribution
is
different.
B
K
Yeah
and
so
now,
when
we
had
all
the
satellite
output
for
the
modern
earth,
we
say:
okay,
if
we
feed
them
to
a
spectrum
generator
to
predict
what
james
well
could
see
from
this
planet.
What
are
we
going
to
get?
We
got
that
here?
What
we
are
putting
is
the
atmospheric
transit
depth
on
the
left
as
a
function
of
the
wavelength
for
each
of
the
atmospheric
output
provided
by
this
gcm.
K
As
you
can
see,
we
have
a
spectrum
with
different
abstraction
lines
and
what
is
the
most
different
between
each
of
these
gcm
is
actually
the
continuum
level,
because
the
continuum
level
for
a
transmission
spectroscopy
is
going
to
be
driven
by
the
altitude
of
the
cloud
deck.
So
if
you
have
a
model
that
predicts
higher
clouds
and
thick,
then
you're
going
to
have
a
higher
continuum.
K
It's
the
case
here
for
exocarp,
for
instance,
and
so
it's
going
to
reduce
your
absorption
line
from
the
bottom,
and
because
you
reduce
the
amplitude
of
the
line,
then
it
will
reduce
its
detectability
at
the
end.
If
you
try
to
predict
how
many
transit
you
will
need
by
james
webb
to
detect
this
atmosphere
looking
at
co2?
Well,
you
will
need
20,
for
example,
for
example,
for
exocarbon
mdg,
you
will
need
16
for
rocky
sweden
that
have
lower
clouds.
So
it's
a
20
different,
it's
not
dramatic,
but
still
it
can.
K
We
are
still
looking
with
the
food
acronym
climates,
using
iterative
suits
of
inter
comparison
nested
for
exoplanets
studies,
cuisines,
where
here
the
objective
is
actually
to
ask
many
more
inter-comparisons
like
tai
for
exoplanet
and
so
cuisine
will
be
a
meta
framework,
so
framework
hosting
frameworks
to
compare
benchmark
and
validate
exoplanet
models.
So
it
could
be
gcm,
they
could
be
energy
balance
model.
Relative
transfer,
1d
photochemical
models,
anything,
but
we
want
to
help
the
community
to
benchmark
them
and
to
provide
those
benchmarks
for
a
long
long
time
for
future
models
to
come
as
well.
K
And
so,
if
you
are
interesting,
we
are
going
to
have
it's
not
yet
announced,
but
the
workshop
on
this
cuisine
project
is
going
to
be
held
virtually
in
september.
K
We
are
going
to
announce
it
very
soon,
but
if
you're
very
interesting
also,
please
send
me
an
email
and
yeah.
I
will
take
any
question.
B
Well,
thanks
thomas,
so
we'll
move
on
to
the
last
two
talks.
The
next
one
is
howard,
chen,
a
phd
student
at
northwest
university.
B
I
had
to
cancel
here
right
at
the
last
second,
but
eric
wolf
agreed
to
give
us
talk
as
far
as
I
know
so,
eric
if
you
could
give
howard's
talk
and
then,
following
that
eric
wolf
who's,
a
research
scientist
at
university
of
colorado
boulder
will
give
his
talk.
L
Hello,
sorry
about
that,
can
you
see
the
screen?
Yes,
okay,
great
so
hi
everyone,
I'm
eric
wolf,
so
howard,
unfortunately
couldn't
make
it
today
he's
traveling.
He
smartly
spent
the
pandemic
at
in
his
native
taiwan
with
his
family,
but
it's
traveling
back
stateside.
So
we
couldn't
give
this
talk
today,
so
I'm
going
to
give
it
on
his
behalf,
so
characterizing
atmospheres
tomorrow.
L
Sorry
about
that,
I
just
hit
the
wrong
damn
button.
Oh,
hopefully
we're
almost
done
with
these
virtual,
these
virtual
conferences,
my
my
my
mistake.
L
View
slideshow
all
right
we're
we
should
be
back
so
tomorrow
talked
a
little
bit
about
atmospheric
characterization
through
transit
spectroscopy
from
the
thai
project,
and
it
really
combines
three
different
aspects:
the
composition
of
the
atmosphere,
which
has
chemistry,
the
chemistry
of
the
atmosphere,
the
chemical
composition
and
also
clouds
which
play
an
outsides
role,
which
tomorrow,
we'll
talk,
talked
a
lot
about
and
will
be
in
the
papers
about
the
thai
project.
L
So
traditionally
such
transmission,
spectroscopy
type
works
and
thinking
about
the
chemistry
of
exoplanet
atmospheres,
photochemistry
affected
by
the
different
stellar
types
has
been:
the
purview
of
1d
photochemical
models
which
are
fast
and
have
well
defined
chemistry
and
flexible
chemistry,
chemistries
for
a
variety
of
atmospheric
types,
and
then
3d
models
have
been
used
for
modeling,
the
climate
and
clouds
of
extrasolar
planets
shown
here
are
some
results
of
tidally
locked
planets.
L
Where
you
have
that
substellar
point
that
has
developed
a
thick
cloud
deck,
which
june
has
talked
about
in
a
warm
spot
right
at
the
middle
of
the
planet.
However,
we
really
need
to
start
combining
these
two
aspects
together,
but
it's
easier
said
than
done
so
howard
has
done
his
phd
work
at
northwestern
university,
using
the
wacom
model
to
study
the
chemistry
and
climate
of
tidally,
locked
exoplanets
around
m
dwarf
stars.
So
combining
these
two
elements,
so
the
atmospheric
circulation
on
titally,
lock
planets,
is
very
different
from
that
on
earth.
L
So
the
advection
of
species,
as
well
as
the
different
stellar
spectra
coming
in,
can
both
can
affect
your
atmosphere,
chemistry
and
also
gradients
of
where
you
see
atmospheric
species
existing.
So
on
the
left,
we
have
an
earth-like
planet
with
your
three
weaker
meridianal
cells
of
circulation
and
then
for
a
tightly
locked
planet.
L
You
tend
to
see
this
one
big,
strong,
meridianal
circulation
and
higher
gradients
between
day
to
night,
so
howard
has
used
wacom,
which
wacom
obviously
has
been
developed
for
the
modern
earth,
and
is
you
know,
includes
chemistry
and
karma.
Can
is
well
linked
to
it
for
different
hazes
for
different
types
of
aerosols.
It
has
a
high
model
top.
L
I
guess
one
of
the
problems
with
exoplanet
modeling
is
the
more
complex.
The
model
gets
the
harder.
It
is
to
adapt
and
modify
it
to
suit
different
exoplanets
with
different
stellar
spectra
and
different
atmospheric
compositions
gets
to
be
particularly
difficult
for
for
say,
with
wacom
and
mozart,
trying
to
force
atmospheres
that
have
wildly
different
co2
than
the
modern
day.
L
For
instance,
so
howard
has
published
a
series
of
three
papers:
the
first
one
in
appj
letters
using
camchem,
which
was
a
kind
of
a
proof
of
concept
that,
on
these
tidally
locked
planets,
you
do
end
up
with
day
to
night
asymmetries
and
important
biopos
bio
signatures
for
terrestrial
exosolar
planets.
The
second
paper
used
wacom,
and
that
was
in
appj,
and
it
examined
the
combination
of
atmospheric
dynamics
and
different
spin
rights
with
photochemistry
from
different
stars.
L
Looking
at
moist
greenhouse
atmospheres.
Moist
greenhouse
atmospheres
are
those
it's
a
term,
an
ancient
term
divided
by
jim
casting
of
when
you
get
a
certain
amount
of
water
vapor
in
the
stratosphere.
Typically
above
about
a
thousand
ppm,
I
think
it
is.
You
can
start
photolyzing
your
water
and
then
losing
and
leaking
your
hydrogen
to
space
at
at
rates,
which
are
geologically
significant
that
you
can
actually
leak
away.
L
A
whole
earth
ocean
of
water
within
several
billion
years,
so
howard,
was
looking
at
these
moist
greenhouse
atmospheres
with
wacom,
where
you
can
actually
model
the
upper
atmosphere
and
the
chemistry
quite
well,
and
the
race
at
which
h20
was
being
dissociated
and
then
finally
published
in
nature.
Astronomy.
L
Just
recently,
howard
used
wacom
using
its
features
for
simulating
flares
and
including
solar,
energetic
protons
for
exoplanets,
around
g
k
and
m
dwarf
stars
using
flaring
rates
and
strengths
derived
from
observations
using
the
test
satellite,
so
to
kind
of
show
some
of
the
key
plots
here.
This
is
from
howard's
first
paper,
looking
at
patterns
of
dimethyl
sulfide
emissions
compared
to
an
earth
like
case
with
global
dimethyl,
dimethyl
sulfide
and
compared
to
a
totally
locked
planet
where
you
may
only
have
a
mission
on
the
day
side.
You
start
seeing
these.
L
You
know
definite
the
change
in
the
gradients
from
equator
to
pole,
from
day
side
to
night
side
same
an
ozone
as
well.
You
can
see
that
on
the
left-hand
panel
for
a
low-mass
star
and
tidally
locked
planet,
you
see
this
minimum
and
ozone
right
on
the
where
the
sun
is
shining
and
ozone
tends
to
collect
on
the
night
side
of
the
planet.
L
So
I'll
have
to
refer
you
to
howard's
papers
for
the
much
more
in-depth
discussion
of
these
topics,
I'm
kind
of
giving
a
cursory
overview
of
his
entire
thesis
work
at
the
moment
and
in
his
2019
paper.
He
looked
at
this
casting
moist
greenhouse
effect,
where,
instead
of
in
the
casting
mode
when
water
gets
above
mixing
ratio
of
10
to
the
negative
three
in
the
stratosphere,
you
kind
of
hand
wave
and
say
that
water
loss
could
be
significant.
L
But,
however,
with
wacom,
you
can
actually
predict
calculate
how
much
water
is
being
fatalized,
how
much
hydrogen
is
being
produced,
because
hydrogen
is
what's
leaking
away
to
space,
not
water.
So
in
these
experiments,
using
stellar
spectra
from
a
variety
of
stars,
howard
found
that
it's
really
the
stellar
activity
that
drives
the
photolysis
of
water
vapor.
So
it's
active
stars.
L
You
can
have
water
loss
to
space
significant
at
significant
rates
and
dry
out
planets.
However,
if
you
have
a
quiet
star
that
water
ocean
ocean
water
and
water
in
the
atmosphere
will
stick
around
for
quite
a
long
time,
so
the
encasing
you
know
hand
waving
limit
of
the
moist
greenhouse
is
really
appropriate.
Thinking
about
active
stars
and
not
quiet
stars,
two
minutes.
B
L
Okay,
I
will
kind
of
jump
ahead
a
little
bit
so
howard
in
collaboration
with
zuk
chong
zhan
at
mit,
calculated,
a
bunch
of
transmission
spectra
and
emission
spectra
which
I'm
not
going
to
go
into.
I'm
going
to.
I
guess,
talk
a
little
bit
briefly
about
howard's
latest
paper
in
nature:
astronomy,
where
he
used
wacom's
flare
capabilities
to
simulate
flares
onto
earth-like
earth
composition,
planets
around
different
stars,
including
wacom,
has
the
feature
where
you
can
control,
where
the
latitudes
of
which
the
solar,
energetic
protons
can
impinge
on
the
planet.
L
So
with
a
magnetic
field,
these
energetic
protons
are
kept
towards
the
poles.
However,
it's
thought
that
some
of
these
totally
locked
planets
may
have
limited
magnetic
fields,
so
the
solar,
wind
and
the
charged
particles
may
just
slam
right
into
the
planet
unprotected
instead
of
funneling.
Those
particles
to
the
poles
so-
and
this
figure
kind
of
shows
shows
the
idea
there
so
for
a
g-star.
This
is.
I
should
also
mention
that
flares
on
m
dwarf
stars
can
be
much
much
more
powerful
than
from
the
sun.
L
So
first
for
a
magnetized
planer
around
a
g-star.
This
is
nitrous
oxide.
At
the
flare
peak
you
can
see
pre-flare
flare
peak
and
then
the
post
flare
and
the
300-day
mean
magnetized.
You
see
most
of
the
action
happening
at
the
polls
here,
but
for
an
unmagnetized
planet
around
em,
dwarf
star.
L
You
can
see
that
the
whole
planet,
the
whole
day
side
of
the
planet
is
affected
and
you
can
actually
cause
a
state
change
that
doesn't
go
away
where
on
earth
the
earth
sun
system,
you
may
have
one
strong
flare
that
occurs,
and
then
that
hits
the
earth
and
then
a
lot
of
a
long
time
period
with
nothing.
So
the
chemistry
can
damp
back
down.
L
But
on
for
these
m
stars,
these
big
flares
can
hit
at
high
frequencies
enough
that
you
actually
pump
the
atmosphere
into
a
different
state,
comparing
pre-flare
to
post
layer
than
300
amine,
because
these
larger
flares
are
actually
hitting.
L
You
know
every
you
know
I
don't
know
the
number
off
the
top
of
my
head,
but
on
a
frequencies
of
less
than
300
days,
you're
continually
pumping
up
the
chemistry
so
changing
the
state
of
the
atmosphere,
the
atmospheric
composition
based
on
flares
and
strong
flares
at
high
rates
of
impinging.
L
So
howard
is
going
to
be:
he
accepted
a
postdoc
at
the
goddard
space
flight
center,
the
npp
postdoc,
and
he
so
he's
going
to
be
continuing
to
work
with
myself
and
then
also
dan
marsh
to
do
different
projects
extending
the
capabilities
of
wacom
for
studying,
extrasolar
planets.
These
include
trying
to
put
in
anoxic
chemistry
into
mozart
with
the
preprocessor,
so
this
will
be
a
really
big.
L
You
know
a
big
positive
step
to
allow
us
to
model
different,
the
chemistry
of
different
types
of
planets,
not
just
oxygenated
oxygenated
ones,
and
he's
also
working
on
implementation
of
prognostic,
high-energy
ion
reactions
with
vladimir
patien
and
hanley
liu.
So
I
think
I
think
that's
the
end
of
howard's
talk.
So
hey
howard
is
a
very
talented
student
soon
to
be
great
phd
graduate
I'm
sorry.
He
couldn't
be
here
today
with
us,
but
he
is
a
definitely
going
to
be
a
major
contributor
to
cam
cesm
and
wacom.
B
Okay,
sorry
he's
not
here
today,
yeah,
so
we'll
go
to
the
last
talk
here
and
okay
and
just
okay
again
and
I'll.
Give
you
a
two.
L
L
Okay,
great,
can
you
see
my
screen
good?
L
L
You
probably
heard
the
name
exocam
bantered
about
a
few
times
earlier
in
the
day
today,
which
I'm
happy
about,
so
I'm
at
cu,
boulder
and
last,
but
I
also
have
the
nasa
logo
up
here,
because,
despite
cam
being
an
ncar
nsf
branch
of
science,
funding
nasa
has
been
has
funded
the
development
work
for
exocam
over
the
last
decade
or
decade
or
so.
So
what
is
exocam?
L
So
it's
a
model
extension
to
end
car
csm
1.2.1
that
facilitates
a
variety
of
exoplanet
climate
modeling
applications,
while
most
of
the
applications
tend
to
be
in
these
idealized
regimes,
aqua
planets,
land
planets.
You
know
simple,
atmospheric
compositions,
you
know
nitrogen
and
water
vapor
or
just
co2
and
water
vapor.
It's
can
be
surprisingly
difficult,
especially
for
people
new
to
the
field
and
students
to
try
to
take
a
major
climate
model
from
a
national
lab
and
just
say.
Okay,
I
want
to
strip
away
all
the
complications.
L
I
just
want
to
do
something
really
simple
in
the
context
of
an
idealized
exoplanet
world.
So
that's
where
exocam
kind
of
comes
in
so
it's
easily
installable
on
top
of
csm
1.2.1.
It
contains
source
code,
modifications,
initial
conditions,
files,
nameless
csm,
util,
scripts,
initial
conditions
and
analysis
scripts.
Things
like
that,
so
exocam
grew
out
of
my
graduate
work
in
early
postdoc
post
phd
projects
so
and
here's
just
a
gratuitous
video
of
a
k218b
simulated
with
40
h2
in
its
atmosphere,
water
vapor
at
the
200
millibar
level.
L
So
so
what
can
exocam
do
currently
out
of
the
box?
It
supports
compositions
of
nitrogen
water,
vapor
co2,
methane
and
hydrogen
for
surface
pressures
of
0.25
up
to
10
bars.
I
think
actually
recently,
one
of
the
omaha
students
has
been
working
on
has
gotten
the
model
to
run
down
to
0.1
bar
as
well,
so
that
we're
kind
of
increasing
the
balance
there.
Temperatures
between
100
and
500
k,
that's
in
the
rate
of
transfer
limits,
and
if
you
use
a
dry
planet,
you
can
tend
to
get
hot
hot
climates.
L
But
if
you
have
an
aquaplanet,
you
tend
to
run
into
numerical
instabilities
and
run
away
greenhouses
and
things
like
that
surface
types
generally
we're
thinking
about
idealized.
We
don't
know
much
about
exoplanet
surfaces,
so
we
tend
to
think
about
idealized
configurations,
aquaplanet,
slab,
oceans,
there's
a
q,
flux,
ocean
land
planet
with
earth
continents.
L
I've
also
I've
hooked
up
exocam
to
the
dynamic
ocean
model
before
as
well,
but
I
have
not
had
the
patience
to
work
through
the
paleoclimate
toolkit
to
actually
change
the
boundary
conditions,
because
the
paleoclimate
toolkit
for
those
familiar
works
in
binary
files,
which
is,
I
just
lost
patience,
but
you
can
run
it
with
the
the
point
is
exocamp
can
be
run
with
a
dynamic
ocean
model
as
well.
Exocam
provides
a
number
of
stellar
spectra
between
2600
and
6600k.
L
It
collects
in
one
module.
It
collects
orbital
plant
parameters
and
planet
parameters
that
are
all
in
one
place
and
you
can
configure
xocam
in
different
ways.
Typically,
I
run
it
with
cam4
and
the
finite
volume
scheme,
but
I've
run
it
with
the
spectral
element
scheme.
You
can
leverage
some
more
senegalmen
cloud
physics
and
karma
and
different
things
like
that
as
well.
L
So
so,
how
is
it
structured,
so
exocam
actually
is
split
into
two
different
github
distributions:
first,
there's
xo
cam
and
second
there's
xort,
so
I'll
start
with
xor
t.
First,
it's
the
radiation
scheme
and
I
decided
to
split
it
off
from
exocam
a
number
of
years
back
and
into
its
own
version,
so
that
it
facilitates
development.
L
Having
a
flexible
and
easy
to
work
with
radiation
scheme
is
critical
for
exoplanet
modeling,
where
things
like
the
dynamical
core
and
surface
boundary
conditions
and
tend
to
get
in
cloud
physics,
which
is
also
a
very
important
thing
tend
to.
We
tend
to
reuse
them
from
the
earth
model.
L
To
start
getting
to
weirder
atmospheres,
you
start
needing
to
expand
on
your
radiation
scheme.
If
you
want
to
do
stuff
like
a
pure
co2
atmosphere-
or
you
know,
a
sub-neptune
atmosphere
with
fractions
of
hydrogen
and
things
like
that.
So
and
it's
so
then
on
the
xocam
side.
So
exocam
contains
the
standard,
csm
machinery,
source,
mods,
nameless,
etc
and
then
its
functions
are.
It
has
a
various.
It
has
a
collection
of
smaller
fixes.
L
I've
acquired
over
the
years
in
different
places
in
the
code,
which
I
won't
spend
my
minutes
describing
at
the
moment,
and
then
it
also
collects
into
a
single
build
time
module
many
of
these
important
geophysical
and
atmospheric
parameters
that
you're
interested
in.
So
I
know
there's
numerous
ways
you
could
set
things
by
name
lists
and
different
parts
of
the
code.
L
This
just
kind
of
makes
a
collects
everything
into
one
place
that
if
you're,
okay,
you
just
want
to
model
an
exoplanet,
just
go
to
the
one
file
and
change
your
plant
radius
and
its
spin
rate
and
the
amount
of
co2
all
in
one
place,
and
it
overrides
everything
else.
So
with
xocam
you
copy
all
the
files
to
the
right
places
after
you
already
have
csm
1.2.1
and
then
for
xort.
L
Typically,
I
include
it
via
the
user
underscore
src
option.
However,
you
could
just
copy
all
the
the
files
into
the
source,
mods
src.cam
file,
with
a
cam
directory
directory.
If
you
wanted
to
so
where
to
get
it,
you
got
to
get
csm
1.2
first
and
then
on
my
github
site.
You
can
get
both
exocam
and
xort,
so
these
were
first
made
available
to
the
public
on
github
in
november
2018,
and
I
think
it's
important
to
know
that.
L
Originally
I
put
these
on
github
simply
to
keep
myself
organized
as
starting
to
splinter
into
different
projects,
and
lo
and
behold,
other
people
started
to
find
it
useful
and
started
to
use
it.
So
to
date
there
are
32
papers,
including
what
I'll
call
some
proto
exocam
papers.
I
wrote
as
a
grad
student
by
17
different
lead
authors.
L
So
there's
definitely
been
you
know,
interest
in
the
community
and
you
know
an
interesting
value
found
in
it
in
the
community
so
and
okay,
two
minutes
good,
so
the
latest.
So
recently
I
just
did
a
big
overhaul.
The
radio
transfer
scheme
I've
improved
it
over
the
years.
L
L
Absorption
coefficient
tables-
you
can
just
do
one
gas
at
a
time
and
the
dominant
gas
is
dynamically
selected
so
and
then
for
correlated
k,
I've
been
using
the
helios
k,
gpu
driven
sorter
from
kevin
hanger's
group
in
burn,
which
is
just
incredibly
just
blisteringly
fast.
So
combined
between
the
helios
k
and
the
extinction
equivalent
extinction,
modification
of
the
radio
transfer
now
can
be
pretty
fast.
L
You
still
have
to
unlike
a
model
like
socrates,
which
handles
all
the
back
end
stuff
for
you,
you,
if
you
want
to
add
a
gas,
you
have
to
get
your
hands
dirty
a
little
bit,
but
making
the
k
coefficients
with
helios
k
can
take
a
matter
of
minutes
to
hours
and
you
can
really
get
things
done
fast.
So
I'm
excited
about
that
update
and
here's
just
you
know
proof
I
did
something.
This
is
for
a
pure
co2
atmosphere.
L
Two
bars,
the
outgoing
long
wave
radiation,
comparing
smart
to
socrates
to
n68
equip
is
just
the
version,
tag
and
xor
t
now,
and
you
can
see
that
you
can
see
they
do
pretty
well.
So
the
new
rt
and
xo
cam
and
socrates
actually
get
almost
the
same.
Exact
answer
a
little
bit
different
than
smart,
which
is
line
by
line.
I
think
I
surmise
the
differences.
It
involves
the
line
wing
treatments
between
the
two,
but
nonetheless,
you
can
see
going
to
say
a
two
bar
pure
atmosphere.
L
You
need
to
start
expanding
the
flexibility
of
your
radio
transfer
code,
so
the
future.
So
I
have
a
couple
of
funded
projects,
an
early
which
will
hopefully
dovetail
back
into
exocam
stuff,
which
will
be
made
public.
So
I
have
a
funded
early
mars
project.
There
already
is
a
mars
scam,
but
I'm
going
to
kind
of
integrate
some
of
the
functionality
back
into
exocam,
including
co2,
surface
and
cloud
compensation,
karma
cloud
and
dust
applications
and
some
low
pressure,
atmospheric
bug
fixes
and
files
that
I've
already
run
across
the
funded
h2
rich
planets
project.
L
These
are
the
type
planets
jwst
is
going
to
see
so
dovetail
development,
stuff,
we're
talking.
Expansion
of
the
k
coefficient
sets
the
higher
pressures
and
temperatures
some
new
gases,
helium
ammonia
and
whatever
high
pressure,
atmosphere,
bugs
and
stuff
and
files.
I
run
across
I've
run
into
along
the
way,
finally
funded
toi,
700d
simulations,
which
will
incorporate
more
work
with
fractal
hazes
with
karma
and
then
so.
L
Finally,
to
finish
so,
I
I
think
it's
important
to
note
that
you
know
exocam
is
merely,
I
would
say,
a
convenient
hack
based
on
my
personal
work,
which
I
put
on
github,
which
has
fortunately
been
useful
for
other
people.
I
don't
proclaim
that
it's
the
greatest
the
best
way
to
do
things,
and
perhaps
probably
not
with
cesm,
especially
now
as
dan
mentioned
at
the
beginning.
L
So
csm2
is
now
forkable
on
github,
unlike
when
I
was
working
with
csm1,
1.2
and
csm,
is
the
community
or
system
model,
and
so
right
now,
exocam
is
kind
of
my
my
little
toy
model
that
I
post
on
github,
but
I
think
the
whole
all
of
us
would
be
benefited,
perhaps
by
moving
past
exocam
and
pulling
together
towards
like
an
official
planet,
csm
branch,
perhaps
forked
off
a
cesm2.
I
know
easier
said
than
done:
it's
a
ton
of
work
and
structure
and
needs
to
be
done.
L
But
having
you
know,
multiple
people
and
all
the
interest
today
of
having
multiple
people
feeding
into
and
developing
the
branch,
I
think,
would
really
start
to
get
gain
us
more
traction
as
opposed
to
you
know.
I
love
exocam.
It's
taken
me
this
far
in
my
career,
but
I've
kind
of
reached
that
wall
of
productivity
and
tasks
and
things
to
do
that,
it's
hard
to
keep
it
hard
to
keep
it
floating.
L
I
mean
I
was
the
only
contributor
at
the
official
contributor
at
the
moment,
so
I
think
that
it
would
be
really
cool
if
we
could,
you
know,
make
something
bigger
with
the
community
with
you
know
everyone
contributing.
So
that
finishes
my
talk.
So
thank
you.
Thank
you
all
for
listening
to
my
talk
and
also
my
talk
on
behalf
of
howard,
who
put
together
those
slides.
So
thank
you.
B
I
just
want
to
thank
all
the
speakers
a
lot
of
exciting
talks.
I
learned
a
lot
spurred
my
imagination,
remind
you
that
these
are
all
recorded
on
youtube,
so
you
can
go
back
and
look
at
them
and
I'm
going
to
turn
it
over
to
dan.
Now,
for
the
discussion
we've
gone
a
bit
over
quite
a
bit
over.
I
did
look
at
the
schedule
and
there's
nothing
after
this
in
this
esm
workshop.
A
All
right,
hopefully,
you
can
see
that
you
can
hear
me
all
right.
I
have
a
few
notes
here
that
I
thought
might
it's
for
some
discussion.
I
first
of
all
want
to
thank
all
the
speakers
for
some
really
interesting
talks
and
the
fact
that
we
went
over,
I
think
just
indicates
that
we
need
a.
We
need,
a
larger
venue.
We
need
more
time
to
talk
about
all
the
exciting
work
that's
going
on
with
with
csm
and
in
this
area.
A
So
I
just
thought
that
perhaps
we
could
have
have
some
structured
discussion
in
two
areas.
One
is
sort
of
this
idea
about
co-development,
about,
as
eric
pointed
out,
maybe
thinking
about
how
all
of
this
work
could
be
better
or
coordinated,
and
what
would
be
projects
that
we
could
work
together
on
within
the
sort
of
csm
development
paradigm
of
forking
and
pushing
things
back
and
and
moving
everything
along.
A
So
I
have
you
know
a
big
list
here
of
opera,
not
only
just
things
to
do,
but
perhaps
new
opportunities
that
have
come
along
since
csm
1.2
and
then
the
second
slide.
I
thought
we
could
have
a
discussion
a
little
bit
about
how
how
to
proceed,
but
I
just
want
to
point
out
that
I
think
there's
some
really
nice
stuff-
that's
happening
within
csm2.
A
A
You
know
those
sort
of
things
and-
and
the
fact
is
that
we're
talking
about
containerizing
these
models
means
that
you
may
be
able
or
hopefully
be
able
to
take
some
of
these
simpler
models
grab
a
container
and
have
it
running
on
your
laptop
within
minutes,
and
then
you
can
play
and
create
your
own
worlds.
So
that's
sort
of
exciting
there's
been
a
lot
of
talk
at
this
meeting
about
things
that
are
hopefully
making
the
physics
interface
easier,
using
such
things
as
ccpp
and
in
the
future.
A
Maybe
it
might
be
easier
to
bring
in
different
types
of
radio
transfer
code,
but
though
I
think
it's
something
along
the
lines
of
this
model
into
comparison
project,
I
think
understanding
structural
uncertainty
in
in
sort
of
climate
prediction
can
be
applied
to
exoplanets,
so
we
just
saw
quite
an
interesting
difference
there
between
two
climate
models
in
that
tie
into
comparison
that
both
run
the
same
radiative
transfer
code
but
have
very
different
surface
temperatures.
I
didn't
so
that
that
was
an
interesting
thing,
so
why?
A
Why
are
they
different?
Is
it
to
do
with
a
dynamical
core?
We
don't
it
does
it's
not
clear,
and
then
I
just
think
that
there's
some
work
in
the
payload
community-
this
isn't
just
all
about
exoplanets
right.
This
is
this
is
about
pushing
the
model
into
regimes
that
are
are
not
typical
or
tuned
for
in
cam
and
csm,
and
so
there's
some
discussions
about.
A
I
know
that
some
development
in
the
paleo
group
about
thinking
about
simplified
land
models,
so
I
as
them
as
the
standard
climate
model,
runs
off
and
starts
to
have
things
like
interactive
crops
and
those
sort
of
things.
We
need
to
keep
some
sort
of
simpler
model,
simpler
land
components
or
a
way
to
take
that
model
and
apply
it
to
say,
ice
covered
a
rocky
planet,
and
so
so
there's
that
idea.
There
was
some
discussion
about
atmospheric
loss
process
and
I
think
around
active
stars.
A
We
really
need
to
start
leveraging
some
of
the
work
that's
been
talked
about
in
terms
of
coupling
wacom,
say
to
magnetospheric
models
and
looking
at
atmospheric
loss
processes
around
active
stars,
so
maybe
there's
an
overlap
there
and
some
some
opportunities
and
then
just
just
as
we've
seen
a
few
examples
of
taking
the
output
from
the
model
and
and
and
using
that
to
generate
spectra
that
can
help
in
the
design
of
new
instruments.
A
So
so
those
are
some
ideas.
I
maybe
have
a
discussion
about
that
now
for
what
five
or
ten
minutes,
and
then
we
could
talk
about
next
steps.
After
that,
anyone
want
to.
L
B
L
Oh
sorry,
I've
got
kind
of
half
trying
to
answer
the
chats
and
I
think
I
spammed
you
down
on
direct
messages
and
I
need
to
get
copy
them
back
to
everybody.
But,
as
I
kind
of
finished
with
at
the
end
of
my
talk,
I'd
love
to
see
a
you
know
a
broader
community
involvement.
But
I
know
it's
not
easy.
It's
just.
It's
not
easy
funding
wise!
It's
not
easy
organization,
wise
and
not
having
really
a
native
home.
L
As
I
said
that
you
know
my
exocam
work
has
been
nasa
funded,
not
anchor
nsf
funded.
So
it's
you
know
I
you
know,
I
don't
really
know.
I
would
love
to
see
us
have
a
forked
official
csm2
planet
branch
that
we
can
collect
all
the
work
that
we're
doing
together.
But
as
far
as
the
organizational
structure,
I
almost
would
dare
to
say
it's
sort
of
above
my
above
my
pay
grade
about
well
who's,
leading
this
thing
who's
making
the
key
decisions.
It's
I
mean,
I
guess
that's.
A
Yeah,
I
think
that
moving
csm
to
github
and
really
sort
of
embracing,
open
source
and
co-development
is
means
that
we
aren't
necessarily.
A
We
don't
need
the
blessing
of
nsf
who
may
raise
an
eyebrow
about
kencar,
doing
a
lot
of
exoplanet
work.
But
but
the
infrastructure
is
there
and
we
can
and
we're.
We
can
leverage
that
and
and
embrace
the
open
source
development,
and
they
will
see
the
benefit.
I
think
when
we
start
bringing
in
these
new
codes
that
perhaps
allow
us
to
understand
things
like
structural
uncertainty
and
climate
prediction.
So
I
think
it's
you
know,
I
think
you've
done
an
amazing
job
of
of
of
with
exocam.
A
But
I
I'd
like
to
maybe
share
that
that
burden
among
the
community
and
trying
to
have
everyone
community
and
contribute
together.
I
I
guess
I
just
throwing
in
one
more
vote
for
if
we
could
bring
exocam
into
a
sort
of
branched
off
cesm2
state
on
github.
That
would
allow
us
to
leverage
a
lot
of
other
development.
That's
going
on.
So
there
is
an
idealized
land
surface
model.
It's
called
slim
that
runs
in
place
of
clm
and
we
at
one
point
actually
with
andrew
rushby,
tried
to
couple
it
to
exocam.
I
But
it's
a
bit
of
a
their
their
all
of
the
coupling
infrastructure
has
changed
so
much
in
the
last
like
five
years
that
it
it
was
a
bit
of
a
nightmare,
but
but
that
would
let
you
know
have
having
it
on
the
the
cesm2
branch.
Would
let
us
borrow
things
like
that
that
are
already
there.
L
Yeah,
I
love
to
do
that.
I
just
I
things
have
just
become
time,
limited
time
limiting
as
far
as
actually
getting
getting
to
it.
I've
been
meaning
to
it,
took
me
forever
to
jump
from
cam
3
to
csm
one
and
now
get
you
if
they're
going
to
be
on
to
csm3.
By
the
time
I
get
to
cesm2.
L
L
A
And
perhaps
but
that's
the
nice
thing
about
github
is,
I
think
we
can
keep
them
in
track
and
just
keep
pushing
and
updating
the
model
as
we
go
along
and
and
not
have
them
diverged
so
far,
but
just
sort
of
keep
them
in
parallel.
A
Brian,
you
had
a
question
point.
J
Question
I
just
wanted
to
also
agree
with
you
dan
that
I
think
that
we
can
bring
a
lot
of
this
into
the
simpler
models
approach.
I
mean
we've
done
a
lot
to
make
it
easier.
I
think
to
do
a
lot
of
these
changes,
and
I
think
peter
mentioned
in
the
chat
that
we
have
changes
on
the
dynamic
side
too.
That
should
make
it
easier
to
make
these
things,
so
it
might
not
even
be
a
fork
of
cesm.
J
A
Yeah
I
mean
most
most
of
these
changes
in
gravity
rotation
period.
Radius
are
now
just
nameless
variables.
So,
and
hopefully
I
mean
this
is
always
the
difficult
part
is
that
those
dependencies
are
now
put
into
one
place
in
the
model,
and
so
you
don't
have
to
find
everywhere
where
the
earth's
gas
constant
has
been
defined,
or
something
like
that,
so
so
yeah.
It
could
be
that
there's
at
least
something
simpler.
That
represents
a
mars
which
is,
which
is
a
essentially
just
nameless
selection.
M
Yeah
hi,
I
want
to
respond
to
eric
wolf,
who
said
that
he
was
looking
for
the
blessing
of
nsf
and
I
work
at
nsf
and
I'm
happy
to
bless
this
on
behalf
of
nsf.
But
my
sense
of
it
is
that
you
don't
actually
need
nobody
needs
nsf's
blessing.
What
you
need
is
nss
money.
I
mean
it
is
a
funding
agency.
After
all,
and
I
think
the
the
key
to
nsf's
money,
I
mean
there's
a
few
things
here.
M
The
issue
is
that
the
software
engineering
actually
does
have
to
be
done
at
ncar,
because
that
way
it'll
be
done
in
a
way,
that's
compatible
with
everything
else,
and
the
other
thing
is
that
you
know
nsf
sort
of
somehow
prides
itself
on
or
claims
to
be
community
driven,
and
so
I
think
the
the
key
to
nsf's
money
is
to
sort
of
identify
how
this
fits
in
to
some
concept
of
the
community
that
nsf
funds,
because
you
know
traditionally,
I
think
nsf
has
taken
the
view
that
exoplanets
are
the
purview
of
nasa,
and
I
don't
know
why.
M
That's
not
true.
I
don't
know
why.
I
don't
know
what
aspects
of
exoplanet
science
are
not
being
covered
by
nasa,
and
so
I
think
you
know
a
getting
clarity
on
where
nsf
really
fits
into
into
a
picture
that,
I
think,
ought
to
be
to
some
extent
driven
by
nasa
and,
secondly,
kind
of
defining.
You
know
what
the
community
is.
Those
would
be
things
that
that
we
would
want
to
talk
about.
L
Thanks
eric,
if
I
could
just
comment
so
I
think
that
so
you're
right
nasa
does
have
a
large
and
ever
increasing
funding
pie
for
exoplanet
science
in
a
variety
of
facets
and
they
have
their
own
3d
model
for
exoplanets
based
off
the
model
e2
that
they're
putting
money
towards.
But
amongst
this
bias
crowd.
So
I
get
a
little
bit
of
money
from
that
that
team,
as
well
amongst
us,
biased
crowd.
I
could
say
honestly
without
having
to
hide
my
true
feelings.
L
L
Model
the
way
it
you
know
and
now
that
it's
on
github,
I
think
it's
even
better.
So
while
nasa
does
have
the
exoplanet
thing
covered,
you
know,
csm
provides
another.
You
know
you
can
never
have
too
many
models
which
all
disagree
with
each
other,
especially
when
we
get
into
planets
and
exoplanets
and
things
diverge,
and
I
think
csm
is
a.
You
know,
a
very
powerful
tool,
a
more
powerful
tool
than
than
the
model
e2
version
that
nasa
is
funding
at
the
moment.
So
it.
M
May
also
be
that
there
are
interesting
sort
of
you
know:
science
taxonomy
ways
of
of
getting
at
this
questioning
one
of
the
things
that
we
do
sometimes
in
my
program,
although
it's
not
really
our
thing
is
snowball
earth
projects.
M
You
know
people
taking
cesm
or
something
similar
and
seeing
what
happens
when
you
cover
the
earth
with
snow
and
and
ice
and
all
that-
and
there
are
these
interesting
projects
about
refugia
for
life
in
in
you
know,
in
the
snowball
time,
and
so
on
other
things
that
you
know
we
have
certain
things
under
one
roof
that
nasa
may
not
the
business
of
putting
flaring
into
wacom
x.
M
For
instance,
you
know
it's
easy
for
us
to
do,
because
the
geospace
part
of
nsf
sits
right
in
the
same
section
as
the
sort
of
regular
atmospheric
part
of
nsf,
and
so
that's
kind
of
makes
life
easier
for
kind
of
mixing
and
matching
some
things.
I
don't
know
that
you
would
ever
have
the
same
sophistication
for
an
exoplanet
model
that
wacom
x
gives
you
in
looking
at
the
upper
atmosphere.
M
You
know
so
so
there
are
those
kinds
of
considerations
that
go
along
with
it.
Another
one
is
that
a
lot
of
people
that
we
fund
to
do
regular
climate
dynamics
dabble
in
exoplanets.
M
So
it
is
kind
of
it's
partly
a
technical
question
about
you
know
who
has
the
better
model
and
who
has
the
better
modeling
framework,
but
there's
other
aspects
of
how
we
conceive
of
the
science
and
how
we
sort
of
approach
it.
That
might
be
worth
considering.
F
Yeah,
I
just
wanted
to
echo
a
lot
that's
been
set,
but
just
pointing
out
from
like
I'm
a
cam
model
developer,
and
I
really
see
huge
benefits
in
improving
our
codes
by
you
know,
bringing
like
supporting
exoplanets
and
so
on.
We
need
to
generalize
our
codes
more
and
that
that
really
improves
our
code,
so
so
it's
in
some
way
can
in
a
more
official
way,
support
exoplanets
within
csm.
I
think
it's
a
it's
really
a
win-win
situation.
There's
of
course,
the
aspect
of
the
software
engineering
support
from
the
ncaa
side.
F
That
obviously
needs
to
be
to
be
funded
and
then
also
from
the
science
side.
I
was
looked
into
simulation
of
super
rotation
on
planets,
where
they
found
really
interesting
sensitivity
to
the
numerics
where
they
showed.
If
your
dicor
cannot
conserve
angular
momentum,
you
cannot
simulate
super
rotation
and
that's
really
exciting
science.
That
can
be
done.
So
I
really
hope
that
this
can.
Actually,
this
can
happen,
and
if
the
code
does
so
that
end
car
you
know,
then
then
it's
easier
to
move
the
code
along
with
whatever
is
happening
with
cesm.
F
A
Colin
has
questions
and
then
goku
for
her
point
to
make
yeah.
Please.
C
Yeah,
I
think
this
is
a
this
was
a
an
encouraging
point
to
my
kids
if,
as
well
as
exoplanets,
which
are
like
way
away
from
the
core
business
of
contemporary
climate
change,
I
think
some
of
the
modifications
we
need
for
higher
greenhouse
gases
warmer
colder
climates
apply
to
earth's
paleoclimate.
Is
that
it's
the
same
physics
on
the
same
planet,
and
if
the
model
can
get
that
right,
then
we
might
actually
believe
model
results
for
future
climate,
which
there's
no
reason
to
believe
at
the
moment.
C
I
know
the
paleo
climate
working
group-
you
know
work
a
lot
on
that,
but
I
think
we
can
go
to
some
deeper
paleo
climate
cases
do
some
weirder
stuff
with
different
climate
states,
and
I
think
that
really
should
feed
back
into
the
core
mission
of
the
of
the
model.
A
Yeah
and
from
a
selfish
viewpoint,
the
more
people
working
on
adding
these-
you
know
things
like
the
radiative
transfer
that
is
capable
of
simulating
very
different
atmospheres,
the
the
the
more
I'm
allowed
to
use
the
model
for
for
the
for
earth's
atmosphere.
So
so
I
think
it
could
be
a
two-way
street
here
in
which
it's
not
simply
that
we're
trying
to
serve
the
astronomy
community
here
that
we
benefit
both
ways.
As
you
put
peter
and
colin
point
out,
go
con,
you
have
a
question.
N
Well,
he's
just
jumping
from
the
other
crossworking
group
and
it's
interesting
that
the
discussion
is
going
the
same
way
and
they
were
saying
that
encar
and
cesm
and
nsf
should
be
essentially
be
the
hub
for
machine
learning
and
we
should
be
supporting
the
community
at
really
high
levels.
So
I'm
glad
that
the
same
conclusion
is
coming
essentially
so
we'll
be
eric.
You
are
listening.
M
Right,
yes,
I
hear
you,
you
know,
and
should
should
money
fall
from
the
sky,
it's
good
to
have
a
basket
to
collect
it
in
right
I
mean
we
don't
know
how
much
longer
biden
is
going
to
rule
the
roost
before
the
house
and
senate
flip
and
we
go
back
to
sort
of
government
shutdowns
and
things
like
that.
But
in
the
event
that
there's
interest
in
you
know
increasing
the
the
the
funding
to
you
know
to
climate
science
of
all
kinds.
I
think
it
is.
M
You
know
it
is
worth
considering
what
would
what
would
be
the
appropriate
things
for
for
ncar
to
provide
in
that
scenario-
and
you
know
the
problem
with
this
kind
of
stuff-
is
that
you
can
work
all
night.
You
know
dreaming
and
scheming
and
can
talk
concocting
ways,
and
then
the
money
doesn't
materialize
for
one
reason
or
another:
that's
that's
life
in
the
federal
government,
of
course,
but
yeah
I
mean
I,
I
think
the
you
know
the
issue
of
at
some
point.
This
becomes
a
political
process.
M
Right
I
mean
you
can
only
fit
so
many
people
in
ncar.
If
you
want
to
have
different
model
configurations,
they
all
basically
compete
for
the
same
machine
cycles
and
software
engineering
time,
and
things
like
that,
so
you
know
there
are.
M
There
are
cautionary
sort
of
notes
to
sound
in
terms
of
wanting
ncar
to
be
the
nexus
of
of
of
all
things,
but
it
is
worth
this
is
the
time
to
be
thinking
about
that
stuff.
A
So
so
maybe
that's
a
good
way
to
tie
into
how
we
actually
move
this
forward.
A
So
let
me
just
put
up
a
slide
here
that
I
thought
might
be
some
ideas
here
about
how
to
how
to
move
this
infrastructure
for
alternative
earths
forward,
and
so
one
one
idea
would
be
to
just
create
a
community
faq
around
adapting
csm2.
A
I
think
everyone
probably
goes
to
christine
shields
faq
when
they
want
to
do
something
a
little
extreme
with
with
in
paleo,
and
she
has
a
great
thing
there
and
if
they
can't
figure,
then
they
go
on
and
they
start
posting
questions
on
the
bulletin
board,
and
then
this
idea
that
maybe
we
can
start
thinking
about
adding
extensions
to
the
simpler
models
to
the
repository.
A
If
some
this
doesn't
require
necessarily
any
kind
of
blessing
from
from
csm,
I
mean
it's
a
it's
something
that
it's
a
community
model.
Anyone
can
contribute
to
it.
It's
not
the
one
we'll
be
running
for
sigma
7,
but
it
doesn't
mean
people
can't
contribute.
We
we
may
given
the
excitement
of
it
and
and
the
fact
we've
ran
out
of
time.
Everyone
was
even
just
given
ten
minutes
and
we
easily
filled
two
hours
here.
A
I
think
maybe
thinking
about
how
we
could
reach
out
and
have
something
to
do
with
csm
at
planet
workshops,
and
then,
of
course
you
know,
money
does
help.
So
we
should
think
about
coordinating
proposals,
and
I
I
don't
know
I'm
not
big
on
formal
structures.
I
think
you
usually,
if
you
have
a
group
of
except
people
that
are
excited
about
something
they
get
together
and
they
start
working
on
projects.
But
at
some
point
maybe
this
grows
into
something
where
we
do
need
more
formal
structure.
A
I
don't
know,
and
you
know
the
bigger
it
gets.
Yes,
we
may
be
imposing
on
limited
resources
and
cars,
so
we
have
to
think
about
that.
So
any
thoughts
on
where
we
go
next.
A
D
I
have
to
off
for
another
meeting,
but
I
I
would
be
very
much
in
favor
of
the
community
fact
and
possibly
a
tutorial.
I
mean
I
know
that
that
could,
as
you
say,
infringe
upon
existing
in
car
resources,
but
I
know,
as
I
think
it
was
either
a
late
stage,
post-doc
or
early
career
faculty
member.
D
I
attended
the
csm
tutorial
and
I
had
already
been
using
csm,
but
there
were
so
many
things
that
I
didn't
know
that
I
could
do
with
it
that
I
learned
to
do
in
the
tutorial,
but
I
knew
it
was
very
much.
It
was
very
much
geared
towards
how
to
do
earth
simulations
and
I
was
like
the
one
exoplanet
person
in
the
tutorial
saying,
but.
N
D
How
can
we
do
this?
How
can
we
do
that
so
being
able
to
to
have
some
kind
of
a
dedicated,
dedicated
tutorial
or
offshoot
session
at
a
workshop
where
we
learn
how
to
adapt
csm2
for
these
alternative
earth
scenarios
or
exoplanet
scenarios?
D
I'd
love
that
and-
and
then
maybe
in
the
meantime,
this
community
fact
where
you
can
see
like
what
you
can
do,
because
it
sounds
like
from
what
I
see
in
the
chat.
There
may
not
be
always
the
necessity
to
go
to
something
else
like
exocam
or
e2.
You
know
like
that
that
there's
already
some
modifications
in
place
beyond
the
ones
that
I
may
be
using
that
that
I'd
love
to
know
about.
A
B
So
I'll
just
follow
up
on
those
comments.
Maybe
we
could
consider
something
like
an
asp
workshop
in
the
summer
where
we
bring
in
speakers,
because
I
learned
a
lot
here,
there's
a
lot
of
climate
dynamics,
which
is
my
background
and
a
lot
of
these
applications.
B
But
you
could
add
a
tutorial
in
the
middle
you
bring
in
students,
you
bring
in
speakers
and
I
think
it'd
be
exciting,
whether
we
need
another
working
group.
Yet
probably
not
I've
been
involved
co-chairing
the
paleoclimate
working
group
for
years
I
stepped
down.
B
Colin's
talk
was
actually
submitted
to
our
working
group
this
time,
but
I
convinced
them
to
speak
today
and
I
kind
of
coined
the
word
alternative
earth
because
it
really
we
are
doing
things
even
beyond
what
we've
done
in
paleoclimate
go
into
with
wacom
and
wacom
x
and
exoplanets
and
thinking
of
chemistry,
so
there's
so
much
exciting.
B
So
in
some
ways
it's
an
extension,
but
not
really.
I
think
let's
not
do
a
working
group
quite
yet,
unless
gokan
really
wants
it
but
yeah.
I,
like
the
idea
of
because
just
some
of
the
mods
you
showed
amaya
were
very,
I
didn't
know
we
could
do
some
of
that
with
cesm
already
so
I
met
enter.
A
I
think
I
think,
first
of
all,
we
probably
need
to
put
together
a
mailing
list
to
try
and
at
least
think
about
how
we
could
yeah.
I
mean
we,
it's
probably
not
that
hard
to
create
an
aq,
somewhere
and
and
and
we'll
have
to
look
into
that,
but
yeah.
Let's,
let's
try
and
build
a
little
bit
on
this
momentum
that
I've
seen
today.
A
I
think
it's
amazing
and
and
and
so
what's
come
to
us
just
we
have
a
list
of
people
that
attended
but
but
send
me
an
email
and
I'll
try
and
collect
some
put
an
email
list
together.
I
know
eric
had
a
a
list
of
people
that
he
sort
of
solicited
to
attend
in
this
meeting
and
we
can
build
on
that
and
and
maybe
start
putting
together
this
faq
and
then,
let's
see
where
it
goes
from
that.
B
And
colin
just
mentioned,
he
has
an
agu
session
on
earth
system,
evolution.
C
A
Great
and
then
there
was
a
great
amount
of
discussion.
I
hope
that
you
know
in
during
this
and
unfortunately
we
didn't
have
time
to
to
have
that.
You
know
post
talk,
but
I
I
think
hoping
that
that
triggers
at
least
a
few
emails
going
back
and
forth
and
follow-ups.
A
So
at
the
end,
I
think
we've
we're
already
20
minutes
over,
so
we
should
probably
call
it
a
call
it
a
day
and
thanks
everyone
for
their
talks.
B
B
Well
thanks
everyone
and
have
a
good
rest
of
the
day
or
maybe
it's
morning
for
some
of
you,
so
thank
you.