►
From YouTube: 2021 CESM Tutorial: Day 4
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
A
Otherwise
you,
then,
if
it
has
less
than
six
people.
That
means
it's
available
and
you
can
sign
up.
A
A
reminder
that
we
have
informal
chat
on
the
simplified
models
today,
and
this
is
with
dr
eila
simpson,
and
it
goes
from
11
a.m,
to
12,
20
p.m.
Mountain
time,
everyone
is
welcome
to
attend
an
optional
event,
and
if
you,
if
you
are
not
interested
about,
if
you're,
not
working
directly
with
the
five
models,
I
still
encourage
you
to
attend
and
learn
about
it,
because
it's
a
fantastic
tool
that
we
ability
here
with
csm.
A
A
A
reminder
of
the
of
tomorrow's
q
a
panel
session
that
the
poll
for
submitting
questions
beforehand
will
close
today
at
5
pm
mountain
time.
So
please
submit
your
questions
beforehand.
Otherwise
you
have
the
option
to
send
them
tomorrow,
either
via
the
chat
or
live
speaking
up
and
tomorrow
morning.
We'll
have
the
group
photo
at
8.
A
We
can
address
that
afterwards,
but
if
there's
anything
related
to
the
content,
please
let
us
know
harry.
Please
go
ahead
and
mute
yourself,
yeah.
B
My
question
is
actually
in
regards
to
tomorrow's
practicals:
all
the
components
will
they
all
be
happening
like
simultaneously
or
like
like?
How
will
we
be
able
to
like
kind
of
go
to
different
ones
if
we.
A
A
C
Yes,
so
I'm
christine
childs,
I
am
a
project
scientist
in
the
climate
and
global
dynamics
division.
I
focus
a
lot
on
hydrological
cycle
things:
atmospheric
grippers,
monsoon
stuff,
like
that.
I
am
your
moderator
for
today
and
I
let
me
I'm
going
to
share
that
my
screen
so.
C
Anyway,
so
I'm
sharing
my
screen
and
oops
here
we
go.
Let's
put
it
into
present,
though
so
you
can
see
it
full
screen.
Here
are
our
panelists.
I
put
a
little
brief
bio
on
everyone,
so
if
all
of
my
panelists
can
have
their
videos
on
so
our
first
panelist
is
she's
a
project
scientist
also
in
the
climate
and
global
dynamics,
division,
her
expertise
or
is
climate
extremes
and
tropical
cyclones
way.
Do
you
wanna,
say
something
and
wave.
D
C
Thank
you.
Our
our
next
panelist
is
flavio
laner.
He
is
an
associate
assistant
professor
at
cornell,
and
he's
also
connected
to
us
here
at
cgd.
He
used
to
be
a
project
scientist
here
and
still
works
with
us.
His
expertise
is
climate
variability,
natural,
enforced
and
large
ensembles
flavio.
Do
you
wanna,
show
yourself
and
wave
and
say
something.
E
Yeah
thanks
christine
good
morning,
everyone
and
yeah,
thanks
for
having
me
yeah,
as
christine
said,
I
I'm
still
working
with
people
at
anchor
and
used
to
for
many
years.
That's
always
exciting.
I
the
only
thing
I
would
add
besides
yeah.
The
great
description
of
my
interests
is
that
I
also
tend
to
sometimes
delve
into
regional
questions
of
like
drought,
wildfire
and
extremes,
but
the
sort
of
my
general
background
is
really
in
large
scale,
circulation
and
global
dynamics.
C
F
F
C
Thanks
aishu
and
finally,
last
but
definitely
not
least,
is
gerrymail.
He's
a
senior
scientist
here
in
clem
cgd
he's
our
fearless
leader
in
the
section
that
we
don't
that
all
the
panda
panelists
are
a
part
of,
and
that's
the
section's
actually
called
climate
change,
research,
jerry's
expertise
or
modes
of
variability
and
future
climate
change
and
decadal
prediction
jerry.
Do
you
want
to
say
something.
G
Yeah
thanks
christine.
Yes,
I'm
interested
in
been
working
a
lot
on
deca,
the
digital
prediction
problem
and,
in
particular,
trying
to
figure
out
the
processes
and
mechanisms
that
are
occurring,
addiction,
time
scales
that
are
internally
generated
by
the
system
that
we
could
try
to
actually
predict,
which
is
a
fairly
big
challenge
for
a
lot
of
different
reasons.
But
that's
kind
of
the
basis
of
the
premise
that
we
can
actually
predict
something
on
decadal
time,
skills
that
there's
actually
something
in
the
system.
G
We
can
predict
that's
being
internally
generated,
also
interested
in
you
know
like
it
says
there
are
modes
of
variability
climb,
variable
and
climate
change
and
looking
at
how
extremes
kind
of
fit
into
that
picture.
C
Thanks,
jerry,
okay,
so
I'm
actually
going
to
go
out
of
present
mode
here,
might
have
to
stop
sheriff
for
a
sec
and
sort
of,
I
think,
the
sort
of
describe
how
we're
I
think
we're
going
to
go
through
this.
C
We
want
to
actually
make
this
a
little
interactive
so
rather
than
us
sort
of
talking
at
you.
You
know.
If
you
have
like
you
know,
we
can
just
sort
of
expand
upon
topics
and
sort
of
see
where
see
where
it
all
takes
us.
So
I
guess
just
to
start
off.
Well
we'll
just
I'm
gonna
go
over
one
of
the
pre
submitted
questions.
I'm
gonna
actually
share
my
screen
again,
so
we
can
all
see
them,
but
I'm
not
necessarily
going
to
oops
oops.
Where
are
we
hold
on
a
second.
C
C
There
we
go
all
right:
let's
try,
this
again
share
a
screen,
and
then
here
we
are,
I'm
gonna
just
sort
of
have
this.
Have
the
pre-submitted
questions
here,
not
in
not
in
sort
of
full
screen
mode.
Just
so
I
can
sort
of
see
everyone
a
little
bit
better.
C
So
the
first
question
is:
how
does
climate
change
and
variability
affect
extreme
events,
so
this
was
submitted
yesterday?
Does
one
of
the
panelists
want
to
to
tackle
this
question.
D
I
think
I
can
say
something
about
this
question
so,
first
of
all,
I
think
this
is
a
very
broad
and
definitely
very
important,
interesting
question,
so
maybe,
as
a
start,
we
can
think
about
what
do
you
think
are
extremes
right.
So
there
are
different
kinds
of
extremes.
You
think
about
precipitation
or
temperature
extremes.
That's
one
kind
more
like
a
climate
extreme
or
you
can
think
of
other.
Like
events
in
terms
of
extreme,
like
a
tropical
cyclone
or
a
tornado,
you
know
those
events,
dreaming
extremes.
D
So
to
understand
all
these
kind
of
extremes
can
it
can
have
large
scale
controls.
So
how
do
large
scale
control
this
extreme
is
by
modulating
the
environmental
factors
that
can
lead
to
these
events
from
deceivers
to
happen.
For
example,
if
you
think
about
a
hurricane,
a
tropical
cyclone,
there
are
some
important
environmental
factors
that
can
contribute
to
tropical
cyclone
genesis
as
well
as
their
intensity
and
precipitation.
D
So
some
important
factors
include
the
sea,
surface
temperature,
ocean
operation,
heat
content,
as
well
as
upper
level,
wind,
shear,
etc,
and
climate
variability
such
as
mjo
or
enso,
can
all
modulate
these
environmental
factors.
If
we
think
about
climate
change
as
the
climate
becomes
warmer,
the
atmosphere
can
hold
more
water,
and
in
this
way
recent
studies
have
shown
that
there
is
a
high
probability
that
tropical
cyclone
precipitation
will
increase.
D
G
So
maybe
I
could
say
something
about
this,
so
one
of
the
things
that
people
always
ask
is
you
know
how
can
a
really
tiny
increase
in
global
mean
temperature
haven't
make
any
difference,
and
the
answer
of
course
is
always.
G
You
can
really
see
a
small
change
in
average
temperature
by
a
much
bigger
amplification
of
extremes,
and
one
example
of
this
is
that
if
you
just
look
at
record
high
maximum
temperatures
daily
record
high
maximum
temperatures
and
daily
record
low
minimum
temperatures,
if
the
background
climate
wasn't
changing,
you'd
expect
there
would
be
about
an
even
chance
on
any
given
day
of
setting
a
record
high
or
record
low
for
temperature.
But
what's
been
happening,
is
this?
G
The
ratio
of
recordized
record
lows
has
been
shifting
so
in
the
first
decade
of
the
21st
century.
That
ratio
of
recognized
lows
is
about
two
to
one.
So
for
every
two
record,
high
temperatures
daily
high
temperatures
across
the
us
there's
only
one
record
low
temperature
set,
and
since
then
it's
it's
now
above
two
and
it's
still
increasing.
So
that's
something
that
people
can
actually
see
that
you
can
actually
see
that.
G
There's
a
shift
in
the
odds
towards
more
extreme
heat
and
less
extreme
cold,
and
then
the
flip
side
of
that
is
that
you
still
get
extreme
cold.
So
in
the
winter
time,
every
time
we
get
a
record
low
minimum
temperature
in
in
denver,
we
start
getting
calls
to
end
cars
saying
what
happened
to
global
warming.
You
say:
well,
you
know
it
still
gets
cold,
but
it
just
doesn't
get
cold
as
often
or
as
it
doesn't
get
as
severely
cold.
So
that's
kind
of
one
aspect
of
mean:
climate
change
affecting
extremes.
G
Another
example
of
internal
variability
was
there
was
something
called
the
warming
hole
in
the
in
the
southeastern
us
that
was
real
popular
in
the
90s
and
early
2000s,
and
there
was
people
that
were
saying.
Well,
you
know
that
there's
no
global
warming,
because
it's
not
getting
warm
in
atlanta,
and
it
was
that's
true
because
the
trend
over
the
20
years
of
the
last
dec
two
decades
of
the
20th
century
was
atlanta
and
other
cities
and
areas
around
the
southeastern
u.s.
G
The
temperature
trend
was
more
or
less
flat,
but
of
course
it
was
warming
up
a
lot
in
the
western
us,
and
so
we
could
trace
that
back
to
what
was
going
on
in
the
tropical
pacific
at
least
one
contributor,
and
that
the
interdict
pacific
oscillation
was
in
its
positive
phase
during
that
time
period,
and
so
that
meant
that
the
on
average,
the
tropical
pacific
seas,
temperatures
were
a
little
bit
warmer
than
normal.
G
Even
though
the
el
nino
la
nina
events
were
still
superimposed
on
that,
but
that
basically
set
up
some
convective
heating
anomalies,
which
were
forcing
anomalous
rostry
waves.
So
over
time
you
had
this
pattern
of
anomalous
ridge
over
the
western
u.s,
an
anomalous
trial
for
the
eastern
u.s,
meaning
it
was
warmer
in
the
west
and
somewhat
cooler
in
the
east.
G
But
then,
when
the
ipo
shifted
the
phase
from
positive
to
negative
around
2000,
the
warming
hole
actually
kind
of
went
away
for
a
while,
and
you
had
much
more
even
warming
across
the
us.
But
then
the
ipo
has
kind
of
shifted
back
to
positive
phase
around
2014
or
2015.
G
And
since
then,
we've
been
seeing
a
really
big
ramp
up
of
temperatures
in
the
western
us
has
kind
of
manifested
by
the
heat
waves
in
the
last
few
years
in
the
western
u.s
one
going
on
right
now,
actually
in
the
pacific
northwest
and
and
it's
and
you
get
this,
these
kind
of
regional
patterns
that
are
associated
with
internal
variability.
But
superimposed
on
that
is
this
kind
of
warming
trend.
That's
going
on
from
the
external,
forcing.
C
Thanks
jerry,
so
does
anyone
have
any
follow-up
questions
on
this
climate
change
and
extreme
events?
It
was
very
broad,
but
it
may
be
a
follow-up
question.
That's
related.
We
can.
We
can
ask
it
now
or
later.
I
just
want
people
to
give
up,
give
people
opportunity
to
follow
up
questions
and
unmute
yourself
and
go
ahead
and
ask
or
raise
your
hand.
C
C
I
don't
see
any
raised
hands,
so
maybe
what
we'll
do
is
what
we
meant.
Oh
flavio's,
raising
his
hand.
E
Well,
no
because
I
I
feel
like
this
is
such
an
interesting
topic
right
now
with,
as
sherry
said
with
with
the
western
us,
with
these
extreme
events
and
yeah.
Some
of
us
got
a
lot
of
questions
from
journalists
in
the
last
few
weeks
about
these
extreme
events,
like
predominantly
the
pacific
northwest
heat
wave,
but
like
also
other
extreme
events
around
the
world,
and
this
question,
like
oh,
has
there
been
sort
of
a
step
change
in
climate,
so
the
questions
you
sometimes
get
from
journalists
are
pretty
sort
of
broad
right.
E
They
they
see
an
extreme
event
in
one
location,
that
is
very
surprising
for
the
local
population
and
then
they
asked
was
there
like
a
step
change
in
climate
and
this
kind
of
ties
back
to
to
the
things
that
the
panel,
the
two
panelists
said,
that
you
can
have
a
gradual
climate
change
globally.
E
But
originally
there
is
a
strong
modulation
of
these
extreme
events
by
internal
variability,
and
you
can
indeed
go
for
yeah
decades
without
a
new
heat
record,
and
suddenly
you
smash
your
decade-old
heat
record
a
little
bit
like
when
somebody
suddenly
sets
a
new
olympic
record
out
of
out
of
nowhere,
for
I
don't
know
100
meters
or
something,
and
so
that's
really
like
a
challenging
thing
to
communicate
because
like
on
the
other
hand,
I
feel
like
we
also
get
a
lot
of
questions
like
when.
When
is
it?
E
Is
it's
like
one
and
a
half
degrees
of
global
warming?
Is
that
the
end
of
the
world
kind
of
thing?
And
we
always
have
to
sort
of
say
there
that
no
most
climate
impacts
sort
of
gradually
scale?
If
you
look
at
it
globally,
but
then
at
the
same
time,
we
have
these
kind
of
events
and
we
have
to
explain
the
role
of
internal
variability
and
that's
that's
often
confusing
and
you'll.
E
You'll
probably
find
yourself
in
that
situation
eventually
talking
to
the
public
and
journalists
and
trying
to
trying
to
explain
that
so
there's
like
poses
an
interesting
scientific
problem
there,
but
also
a
communication
challenge.
I
think.
C
So
we
had
a
question
in
the
chat,
love
you,
you
sort
of
addressed
this
a
little
bit
about
what
is
the
difference
between
climate
change
and
climate,
variability
and
term.
I
don't
know
if
you
want
to
expand
on
that,
a
little
bit
more
like
explicit.
You
can,
if
you
want
to
comment
on
that.
E
Yeah,
that's
a
that's,
a
very
good
question
and
I
think
it's
there's
probably
not
a
clear-cut
definition
but
like
in
in
I
think
in
the
science
community.
We
think
of
these
two
terms:
climate
change
and
climate
variability
as
somewhat
different,
but
part
of
the
same
thing
like
climate
change
is
probably
the
gradual,
long-term
change.
E
E
So
like
the
variability
that
happens
regardless
of
which
climate
state
we're
in
whether
we're
in
a
changing
climate
or
a
stable
climate,
then
that
like
helps
you
distinguish
so
the
long-term
trend,
the
gradual
trend,
the
change
in
the
base
state
versus
this
variability
that
happens
on
a
day-to-day
year-to-year
scale,
but
like
yeah,
again
like
in
the
public
discourse.
In
particular.
These
two
terms
get
sort
of
thrown
around
without
a
clear
definition,
and
I
mean
we
have
yeah.
E
We
have
a
a
working
group
and
and
sections
at
encar
that
are
called
climate,
variability
and
change,
and
so
we're
looking
at
all
of
these
things
together
there
but
yeah
you
can.
You
should
just
try
to
define
it
as
clearly
as
possible
when
you
talk
about
it
again,
especially
in
the
context
of
public
communication,
yeah.
G
Maybe
just
just
to
add
a
little
bit
from
the
csm
angle.
You
know
when
you're
looking
at
observations,
everything's
mixed
together
right
so
you've
got
the
increasing
grenache
gases.
You
got
volcanoes
going
off.
You've
got
increasing
air
pollution
from
sulfur
dioxide
and
sulfate
aerosols
that
have
time
varying
and
spatial
varying
kinds
of
things,
and
so
you're.
Looking
at
the
observations
over
say
about
120
or
maybe
150
years
at
the
most
and.
G
Mixed
together,
so
it's
really
hard
to
separate.
What's
what's
responding
to
what
you
know
to
attribute,
what's
internally
generated
variability
that
would
happen
be
happening
anyway,
which
is
what
flavio
was
talking
about,
versus,
what's
being
forced
by
all
kinds
of
things,
natural
or
anthropogenic.
G
These
so-called
pre-industrial
control
runs
are
great
for
this,
because
that's
one
way
that
you
can
get
a
handle
on
what
the
internally
generated
part
of
the
variability
is
in
the
climate
system.
Of
course,
people
can
always
say
you're
in
model
world
and
that's
true,
but
that's
about
the
only
way
we
have
of
really
getting
at
this
problem.
What
could
be
the
internal
generator
part,
and
then
you
can
compare
that
to
the
x
the
we
got
all
the
forcings
happening,
all
at
once
and
say:
okay.
Well,
it
looks
like
during
this
time
period.
G
Maybe
you
have
a
big
contribution
from
internal
variability
during
another
time
period
we
could
have
a
lot
of
external
forcing
and
then
you
can
run
the
model
over
and
over
again
just
with
the
single
forcings.
In
other
words,
you
run
it
for
the
20th
and
early
21st
century,
just
with
the
changes
of
solar
and
volcanoes,
or
just
with
the
changes
of
anthropogenic
aerosols
or
just
with
the
changes
in
greenhouse
gases,
and
then
you
can
actually
look
to
see
which
forcing
is
causing
what
in
the
observational
record
and
see
that's
one
way.
We
have.
G
It's
really
nice,
actually,
a
really
really
nice
use
of
the
model
to
kind
of
deconstruct.
What
we
see
in
the
observations
and
really
try
to
understand
what
is
causing
what,
during
which
time
period
in
the
observational
record,.
C
Thanks
jerry
so
tim,
why
don't
you
unmute
yourself
and
ask
a
question,
then
we'll
go
to
the
next
question
in
the
chat
after
that.
B
Yeah
sure
so
I'm
I'm
I'm
going
to
talk
a
little
bit
more
about
this
communication
fingers
where
the
slavia
just
touched
upon.
How
would
you
then
say
like
if
I
was
a
journalist
and
ask
you
like
okay,
what
what
is
it?
What's
the
linkage
between
like
what's
happening
in
portland
or
in
italy
or
whatever
you
like,
and
right
at
the
moment,
to
climate
change?
How
would
you
and
I'm
I'm
a
journalist
or
your
mother-
or
I
don't
know-
I
I
don't
know
anything
about
about
what
is
happening
at
the
moment?
E
Yeah,
I
don't
know
exactly
what
I
said,
but
it's
tricky
yeah.
I,
like
one
analogy
that
I
saw
posted
on
twitter
is
not
mine.
That
I
thought
is
is
useful.
I
can
try
to
see
if
this
resonates
with
you
was
like
with
the
heat
wave
in
the
pacific.
Northwest
like
this
was
like
just
a
very
strong
heat
wave,
and
the
question
is
not
really
is
the
heat
wave
caused
by
climate
change
because
heat
waves
happen.
Naturally
they
happen
in
any
climate.
E
The
question
is
just
how
much
if
and
if?
Yes,
how
much
stronger
was
the
heat
wave
because
of
climate
change,
and
so
even
just
disentangling?
This
question
is,
I
think,
quite
useful,
because
you
see
a
lot
of
headlines.
Climate
change,
cost,
pacific,
northwest
heatwave
and
that's
not
quite
correct
right.
The
heatwave
just
happened.
E
The
question
is
like:
was
it
made
stronger
and
yes
by
how
much
and
so
that
I
spent
a
lot
of
time
already
explaining
that
and
like
one
analogy
was
like,
especially
if
you
have
a
really
extreme
event
like
that,
it's
like
how.
How
can
you
put
that
in
context
with
the
role
of
climate
changes,
and
so
the
analogy
was
like
you
have
four
dices
and
you
roll
them
and
how
are
the
odds
that
you
get
four
sixes
they're,
very
small?
E
We
all
know
that,
but
it
can
happen
without
any
other
influence,
and
then
the
additional
question
is
like.
Okay,
we
have
these
very
high
temperatures
like
what's
the
role
of
climate
change
in
that,
and
we
can
think
of
that
as
just
adding
one
number
to
the
total
number
of
that
we
get
from
rolling
four
dices,
so
the
maximum
number,
instead
of
24,
would
suddenly
be
25.,
it's
impossible
to
get
to
25
without
this
additional
number,
because
you
only
have
four
dices
maximum
is
24..
E
So
that's
this
one
is
climate
change,
so
it's
actually
it
could
be
very
small,
but
it's
the
thing
that
makes
a
difference
between
what
you
maximally
can
get
in
a
claim
climate
without
climate
change
and
one
with
climate
change.
I
think
that
really
helps,
at
least
for
me,
to
to
contextualize
that
the
heat
wave
most
of
it
was
natural,
but
it
was
amplified
by
climate
change
by
a
little
bit
pushing
it
into
these
temperatures
that
we
have
not
seen
before.
G
Yeah,
maybe
I
could
just
add
the
another
analogy:
it's
a
certain
analogy
world
so
one
and
we're
in
baseball
season
right.
So,
oh.
B
E
G
The
baseball
steroids
is
one
is
one
of
my
favorites
that
I
use
with
journalists
a
lot
and
I
don't
know
every
it's
great
you've
seen
the
links.
I've
already
seen
it.
So
that's
that's
one
of
my
favorites.
The
idea.
G
Is
that
you
know
a
baseball
players,
keep
perfectly
capable
hitting
home
runs
just
like
the
climate
system
is
perfectly
capable
of
having
heat
waves,
but
if
the
players
on
steroids,
you
have
a
much
better
chance
of
hitting
home
runs
and
if
the
climate
system
has
increased
greenhouse
gases,
the
base
states
warmer
and
you
have
a
much
better
chance
of
having
heat
waves.
So
that's
another
way
of
doing
it
and
the
good
thing
about
that
with
performance-enhancing
drugs,
you
can
apply
it
to
bicycle
racing
or
almost
any
any
sport.
G
C
Hasn't
had
a
chance
to
watch
that
video.
You
should
at
some
point,
so
we
do
have
another
question
in
the
chat
and
it's
from
shangho.
C
How
reliable
are
the
future
projections
at
different
temporal
resolutions
in
terms
of
climate
extreme,
so
shanghai,
you
may
want
to
unmute
yourself
and
sort
of
explain
what
you
mean
by
temporal
resolutions.
Do
you
mean
the
difference
between,
like
you
know,
daily
precipitation
or
day-to-day
weather
or
versus
decadal
climability?
Is
that
what
you're,
referring
to
in
terms
of
temporal
resolution.
B
Thank
you.
I
actually
mean
the
difference
between
different
tempo
resolutions
like
three
hourly
output,
six
hourly
output
or
daily
maximum
minimum,
or
should
we
just
look
at
the
monthly
extreme
values
to
to
access
the
future
change
in
climate
extremes?
G
D
Yeah,
so
for
tropical
cyclones,
if
you
want
to
know
how
it
changes
we'll
have
to
first
of
all
detect
it
in
the
model.
So
in
order
to
detect
the
storms
in
the
model,
you
really
have
to
have
at
least
a
daily
resolution.
C
Yeah
and
in
terms
of
reliability,
I
mean
it's
the
it's
the
same
model
simulation,
so
you
know
the
veracity
or
the
the
correctness
of
the
simulation
is
going
to
be.
You
know
consistent,
no
matter
what
your
temporal
resolution
is
in
terms
of
your
output.
You
know,
yeah.
C
D
Oh,
can
I
also
add
one
more
thing,
so,
even
with
one
kind
of
extreme,
there
are
different
aspects.
You
wanna,
you
wanna
differentiate.
Take
the
example
of
hurricanes.
Again,
if
you
wanna
know
the
frequency
of
the
change
versus
intensity
of
the
change,
that's
a
different
story
as
well.
Not
only
temporal
resolution
matters.
The
spatial
resolution
also
matters,
and
if
you
want
to
know
about
the
intensity.
F
I
can
comment
on
something
for
the
grammy
change
I
see
at
the
very
beginning
where
we
could
start
defending
and
we
just
use
a
1d
model
like
a
twist
earth
at
a
point.
Basically,
we
just
get
the
deviation
balance
by
another
side.
We
can
know
how
much
the
first,
the
one
giving
away
right
so
giving
like
specific
right
to
still
and
then
afterwards
we
are
going
to.
F
We
not
only
want
to
know
whether
the
time
will
change
or
not.
We
also
want
to
know
about
their
climate
like
phenomena
such
as
in
groups,
then
1d
model
does
not
work.
So
I
have
to
use
two
people
for
3d
models.
We
can.
We
will
need
to
increase
the
horizontal
or
wait
for
resolution.
So
that's.
I
think
this
is
very
important
to
see
the
phenomena
you
want
to
study
is
the
model
can
result
right.
F
C
Thanks,
aisha,
so
harry,
I
see
you
have
your
hand
raised.
Do
you
want
to
unmute
and
ask
your
question.
B
Yeah
hi,
how
are
you
I
mainly?
This
is
like
pertaining
to
the
discussion
of
resolutions
for
extreme
events,
so
I
know
like
in
one
of
the
first
talks
that
was
given
on
the
atmosphere
or
the
on
the
cam
component,
that
we
have
to
be
mindful
of
like
the
acoustical
or
sorry
the
acoustic
and,
like
the
gravity
waves,
defining
them
within
and
like
that,
also
determines
like
the
how
long
our
time
step
can
be
and
how
long
like
the
the
the
model
is
stable.
B
So
I
was
just
wondering
like
in
regards
to
like,
if
you
want
to
like
set
a
very
short
time
step.
If
you're,
like
wanting
to
look
at
tropical
cyclones
like
would
be
like
a
good
like
kind
of
like
trick
or
like
role
in
regards
to
saying,
like
your
horizontal
resolution,
just
to
make
sure
that
your
model
doesn't
blow
up
while
it's
running.
D
Yeah,
so
actually,
if
you're
running
climate
models
at
csm,
that's
actually
another
thing
I
was
worried
about
because
I
never
go
to
a
resolution.
That's
so
high
because
because
in
climate
models,
the
the
highest
I've
used
is
actually
a
quarter
of
a
degree,
it
was
about
25
kilometers.
D
So
at
that
resolution
you
can
have
the
model
to
generate
its
own
tc's.
It's
I
can
see.
I
can
only
see
tc
like
vertices,
that
have
good
climatology
and
good
wind
structure
and
intensity,
but
it's
not
a
very
well
resolved
tc.
That's
why,
even
at
a
quarter
of
a
degree
atmosphere,
we
cannot
fully
simulate
the
observed
tropical
cyclone
intensity,
so
yeah
for
climate
models
because
of
the
computational
cost.
D
Yeah,
that's
the
excellent
question,
so
actually
the
track
tcs
are
kind
of
sensitive
to
the
tracking
tracking
algorithms
that
you
use.
So
before
I
came
to
uncar,
I
used
to
use
this
tracking
mechanism.
That
kind
of
depends
on
layers
of
layers
of
threshold,
for
example.
I
would
you
know,
try
to
find
a
local
minimum
in
the
pressure
first
and
to
search
for
a
warm
core
near
it
and
then
to
see
always
the
maximum
wind
near
this
tc
center
right
and
that's
how
I
tracked
it
and
I
after
I
come
to
one
car.
D
I
start
to
use
this
software.
That's
called
temp
tempest
extreme.
That
is
very
useful.
I
I
see
that
you've
probably
already
heard
about
it.
Yeah,
it's
really
useful.
You
can
take
different
parameters
and
it
can.
It
can
not
only
be
used
to
track
tropical
cyclones
but
other
like
blobs
like
mcs,
and
things
like
that
as
well.
So
I
highly
recommend
that
software
gotcha
thank
you.
C
No
problem
so
I'll,
just
sort
of
add
to
what
we
said
in
terms
of
the
the
higher
resolutions
that
we
can
go
to
there.
Csm
actually
is
developing
something
called
a
variable
resolution
where
it
globally.
The
resolution
would
be,
like,
maybe
one
degree
and
then
at
a
specific
region.
You
actually
can
drill
down
into
higher
resolutions,
and
I
don't
know
off
the
top
of
my
head,
but
I
think
it
I
think
we
can
get
down
to
something
like
10
kilometers
with
the
variable
resolution,
but
it's
it.
We
can
go
below
25
kilometers.
I
do.
C
I
do
know
that,
but
but
yeah,
but
that
that's
just
sort
of
starting
at
this
point
we
do
have
some
simulations
that
people
are
actively
working
on
that.
So
that's
sort
of
down
the
road
and
in
terms
of
like
figuring
out
what
the
best
time
step
is
often
it's
trial
and
error,
an
iterative
type
of
process.
So
it's
you
know
you
can
use
your
your
fluid
dynamics,
sort
of
theory
to
try
to
figure
it
out
in
general,
but
a
lot
of
times
it's
just
sort
of
iterative.
C
So,
okay,
were
there
any
other
questions
or
follow-ups
on
what
we've
discussed
so
far.
C
I
don't
see
any
hand
raise
or
new
chats,
so
I'm
going
to
just
sort
of
verbally
read
out
one
of
our
pre-submitted
questions
from
one
of
you
guys.
This
actually
is
going
to
be
directed
to
jerry.
In
regards
to
your
talk
that
that
was
posted
on
the
website,
do
you
think
csn
2,
cam
cam6
would
produce
a
different
outcome
in
terms
of
the
role
of
aerosols
and
the
relationship
between
a
and
b
and
pdb.
G
We
look
at
these
days
in
terms
of
climate
sensitivity
in
terms
of
cloud
feedbacks
in
terms
of
climate
system
response
to
things,
looking
historically
the
role
of
aerosols
and
producing
decade
old
timescale
variability
on
their
own,
because
you
can
have
time
varying
aerosol
emissions
or
precursor
emissions
coming
from
various
locations,
and
so
aerosols
are
really
hard
to
observe,
they're
really
hard
to
model,
and
it's
hard
to
figure
out
if
your
aerosol
scheme
is
actually
doing
the
right
thing,
because
a
lot
of
times
you
don't
even
know
exactly
what
the
right
thing
is
that's
happening
in
the
real
system.
G
There's
been
a
lot
of
aerosol
observation,
programs,
aircraft,
observations
and
things
over
big
metropolitan
areas.
Where
there's
a
lot
of
a
lot
of
aerosol
pollution.
There's
some
satellite
observations
we
can
kind
of
look
at,
but
those
don't
really
get
at
the
processes.
Very
much
aerosols
can
affect
clouds,
and
so
that
means
you
can
affect
cloud
brightness
when
aerosols
are
present
and
then
that
gets
into
the
whole
cloud
feedback
kind
of
issue.
G
So
aerosols,
I
think,
are
one
of
the
biggest
challenges
we
have
right
now
in
in
climate
modeling,
both
from
the
observational
point
of
view
and
from
the
modeling
point
of
view,
because
you.
G
It's
really
hard
to
model
something
you
don't
understand
how
they're
working
in
the
real
systems.
This
is
an
ongoing
area
of
active
research,
not
only
in
terms
of
observational
campaigns,
satellite
observations
and
also
modeling.
So,
having
said
all
that,
when
we
get
into
the
new
generation
of
models,
the
the
cmp
six
generation
models
that
we're
in
now
the
csm2
generation,
a
lot
of
these
models
now
have
interactive
aerosol
schemes
and
they
also
have
the
direct
effect
and
indirect
effects,
meaning
indirect
effects.
G
The
model
was
tracking
the
20th
century,
global
temperatures
pretty
well
and
then
about
oh
just
19,
70
years
or
just
fit
1950
or
so
all
of
a
sudden.
The
model
would
cool
off.
It
would
stay
cool
until
about
1980
and
then
suddenly
warm
up,
and
so
what
was
with
that,
and
so
that
behavior
had
really
not
been
seen
before,
and
a
lot
of
modeling
groups
saw
this
behavior,
which
actually
gave
a
worse
simulation
20th
century
climate
than
the
previous.
G
Of
models
gave
that
didn't
have
these
complicated
aerosol
schemes.
So
then
immediately
people
people
thought
it
was.
The
aerosols
were
causing
the
problem.
So
there's
a
lot
of
of
course
leeway.
G
You
can
say
when
you're
parameterizing
things
in
in
a
space
where
you
don't
quite
know
exactly
what
the
right
answer
is,
and
so,
with
some
adjusting
of
some
of
these
parameters,
the
csm2
ended
up
getting
a
fairly
good
20th
century
simulation
and
if
you
looked
across
different
models,
in
fact,
everybody
thought
initially
that
the
new
aerosol
emissions
that
were
being
provided
for
cm6
were
wrong,
because
so
many
model
groups
were
getting
this
big
cooling
that
was
going
on
post
world
war
ii,
but
it
turned
out
that
it
was
probably
a
combination
of
things.
G
It
may
have
been
a
combination
of
the
new
aerosol
emissions
and
those
were
adjusted
a
little
bit.
There
was
a
little
bit
of
an
issue
with
how
the
aerosols
were
affecting
clouds
and
cloud
feedbacks,
and
that
was
you
can
adjust
that
a
little
bit
so
in
the
end,
most
groups
got
something
that
looked
fairly
close
to
the
20th
century
climate.
A
couple
of
models
left
this
in
like
the
e3sm,
which
is
the
dewey
model,
their
version
one
model
they
they
left
that
cooling
in
and
so
that's
a
good
example.
G
If
you
look
at
the
time
series
from
e3sm
for
20th
century
climate,
you'll
see
that
behavior
that
a
lot
of
modern
groups
were
getting
that
they
ended
up
ironing
out
some
other
groups
did
so
that's
kind
of
just
pref
preparatory
remarks
to
how
this
can
affect
dedical
variability,
because
one
of
the
big
arguments
that
goes
on
is
how
much
of
atlantic
declarability,
particularly
with
the
atlantic
meridional
or
the
outlanding
multi-vehicular
oscillation,
the
amo.
G
How
much
of
that
is
due
to
decadal
variations
of
aerosol
emissions
and
aerosol
forcing
and
how
much
of
it
is
internally
generated
and
there's
pretty
good
evidence
that
a
lot
of
it
is
probably
externally
forced
because
you
get
this
really.
When
you
look
at
the
aerosol
only
simulations,
and
you
see
that
you
can
get
kind
of
an
ammo-like
time,
evolution
in
the
atlantic
just
from
the
variations
of
aerosol
forcing.
G
But
when
you
run
the
model
as
a
pre-industrial
control
run,
you
can
see
that
the
internal
there's
internal
processes
involving
the
amok,
the
atlantic
brownial
restraining
circulation
that
are,
can
produce
decadal
time
scale,
variability
of
atlantic
sea
surface
temperatures
so
like
everything
in
the
climate
system,
there's
more
than
one
thing
going
on.
So
in
our
paper
that
we
in
that
talk
that
I
gave
this
nature
geoscience
paper,
we
we're
trying
to
address
this
in
two
ways.
G
We
did
these
pacemaker
experiments
where
we
were
specifying
c-service
temperatures
in
one
region
of
the
world
oceans
and
then
watched
how
the
rest.
The
couple
climate
system
responded
that
and
we
had
this
these
pacemakers
in
two
different
configurations.
One
was
so
called
idealized
pacemakers
where
we
had
this
pattern
of
what
we
thought
was
the
internally
generated
variability
coming
out
of
the
pacific
and
then
out
of
the
atlantic
and
just
put
that
into
the
homologous
ran
tenure
simulations.
So
there
was
no
changes
in
external.
G
Forcing
all
you
were
getting
was
the
model
response
to
that
pattern
of
internally
generated
c-service
temperatures.
We
could
study
the
interactions
between
the
ocean
basins.
That
way,
then
we
ran
these
so-called
time-evolving
pacemakers,
where
we
had
the
actual
time
evolution
of
the
external
forcings.
G
G
Things
in
these
connections
between
the
ocean
basins,
we
saw
in
the
idealized
pacemakers
and
that's
where
we
started
seeing
these
complications
with
the
aerosols,
especially
in
the
atlantic,
and
especially
understanding
how
this
this
relationship
between
the
internally
generated
part
and
the
externally
forced
part
could
actually
be
manifested
and
really
complicates
the
whole
picture.
When
you're
trying
to
understand
these
ocean-based
interactions.
So
I
think,
with
the
current
state
of
the
csm2
and
the
aerosol
scheme-
that's
that's
now
active.
G
I
can
see
he's
just
anxious
to
say
something
about
externally
forced
responses
to
the
ammo
in
the
atlantic,
but
I
think
you'd
still
see
a
similar
kind
of
thing
going
on
you'd
have
this
mixture
of
internally
generated
variable
and
external,
forcing
for
the
ammo
and
then
how
that
interacts
with
the
pacific
would
kind
of
follow
from
what's
going
on
in
the
atlantic
and
also
then,
what's
going
on
in
the
pacific,
because
there's
probably
an
externally
forced
part,
there's
some
papers
that
show
that
asian
aerosols
can
actually
affect
tropical
pacific
sea
source
temperatures
indicative
time
scales.
G
So
that's
a
little
bit
of
it,
but
there's
probably
a
big
internally
generated
part
of
decay
variability
in
the
pacific.
That's
also
going
on
and,
of
course,
superimposing
all
that's
the
increase
in
greenhouse
gas
forcing
so
I
guess
the
the
short
answer
to
a
very
long.
A
long
build
up.
Is
that
probably,
would
we
probably
see
a
similar
kind
of
response
in
the
csm2
that
we
saw
in
the
csm1,
but
I
think
issue.
Maybe
you
could
say
something
more
about
this
external:
forcing
versus
infinite
variability
for
the
amo
in
the
atlantic.
F
F
One
big
difference
between
these
two
model
views
in
their
language.
We
see
a
time.
Evolution
of
the
amo
is
very
like
very
much
like
the
operation,
but
in
ccsm4
it
does
not,
and
one
difference
between
these
two
models.
I
guess
is
the
rsl
forcing
and
then
actually
at
that
time
here
we
also
look
at
the
aerosol,
only
simulation
and
in
front
of
ammo
time
revolution
in
that
one
is
also
very
similar
to
the
sony
century
solution.
F
That
means
that
at
least
from
csm1
we
can
save
ourselves,
say,
asymmetry,
but
not
in
c74,
and
that's
also
one
reason
at
that
time.
We
didn't
really
write
a
paper
on
it,
but
we
noticed
this
result
and
if
we
just
look
at
the
the
control
graph,
no
matter
c
s
and
4
say
that
c,
as
in
line,
we
can
clearly
see
the
relationship
between
the
a
and
o
and
the
amount
they
are
likely
to
young.
F
I
think
now
this
is
still
they
use
the
regression
and
how
much
and
it's
a
yeah
I
said
here
is
that
how
much
the
amo
is
generated
by
the
internal
protein
and
how
much
is
related
to
the
internal
ability
is
still.
I
actually
missed
the
question
now,
if
anybody's
interviewing,
it
definitely
can
chat
with
you.
C
Thanks
ashoo,
so
I
hope
that
sufficiently
answered
your
question,
whoever,
whoever
posted
that
thank
you
for
posting,
were
there
any
follow-ups,
any
questions
to
to
anything
that
jerry
raichu
said.
H
Yeah
thanks
can
I
ask
a
question:
pulling
up
the
pacemaker
simulations
introduced
in
jerry's
lecture
yeah,
I'm
very
interested
yeah
in
that
realistic
time
involving
type
of
pacemaker
simulations
within
the
specific
regions.
So
I
wonder
at
end
cards
they
have
an
ensemble
of
tropical
west
pacific,
only
pacemaker
simulations
because
well
in
jerry's
lecture.
If
I
understand
correctly,
the
worker
circulation
is
key
to
the
teleconnections
between
the
tropical
pacific
and
the
tropical
atlantic,
but
I
think
often
in
literature.
H
We
also
see
the
discussion
about
this
tyler
connections
from
the
tropical
western
pacific
and
within
the
tropical
pacific.
Everything
is
so
coupled
like
wind
assets
or
circulation.
So
I
wanted
to
have
this
ensemble
of
a
with
a
specified
assisting
anomalies
within
the
tropical
west
pacific.
Only
and
if
so,
do
we
expect
to
see
similar
teleconnections
to
the
tropical
atlantic,
or
at
least
the
the
cooling
trend
in
the
east
pacific.
G
That's
a
really
interesting
question,
because
one
of
the
things
that
we
noticed
we
when
we
started
doing
these,
what
we
call
the
time
series
pacemakers
where
we're
putting
in
the
time
evolution
of
the
basically
the
kind
of
nino
3.4
area,
it's
a
little
bit
bigger
area
than
that.
But
we
were
following
the
original
kasaka
and
xi
methodology
and
they
were
they.
They
put
this
in
in
this
region
and
then
other
people,
and
we
thought
the
community
thought
well.
This
is
interesting.
G
This
may
be
a
good
way
of
trying
to
figure
out
isolating
influences
of
one
basin
on
others,
because
you
let
the
model
be
fully
coupled
and
it's
just
responding
to
something
in
in
one
location.
So
we
put
that
in
there
in
that
same
area
and
those
we
run
all
these
pacemaker
experiments.
We
also
did
some
indian
ocean
pacemakers,
where
we
just
had
the
tropical
indian
ocean
sea
surface
temperatures
specified
and
then.
H
G
G
But
then
one
of
the
things
that
we
noticed
was
that
the
one
of
the
systematic
errors
in
in
most
models,
including
the
csm,
is
that
the
pacific
coal
tongue
is
overextensive.
It
goes
too
far
to
the
west
and
it
goes
into
the
western
pacific,
warm
pool.
And
so,
when
you
put
in
if
you're
nudging
sea,
surface
temperatures
in
the
kind
of
the
eastern,
equatorial
or
eastern
tropical
pacific
to
be
out
like
the
observations
you
get,
this
kind
of
discontinuity.
G
Because
then
you
won't
have
you'll
have
these
kind
of
cooler
than
observed
sst's
in
the
western
pacific.
And
you
get
this
kind
of
weird
that
could
it
could,
in
theory
affect
the
teleconnections,
like
you
say,
so,
we're
running
another
set
of
pacemakers
with
csm2
right
now,
and
I
don't
know
if
they're
done
yet
christine.
G
I
don't
know
if
we
know
I
know
we're
running
it
and
we
decided
to
extend
the
nudged
area
over
into
the
western
pacific,
and
so
now
it's
going
kind
of
more
or
less
all
the
way,
not
quite
all
the
way
across
the
pacific,
because
what
kind
of
narrows
in
the
western
pacific
and
comes
down
to
kind
of
a
triangular
shape
in
the
western
pacific,
but
as
to
take
out
the
effects
of
the
anomalous
over
extensive
coal
tongue
and
so
we'll
analyze.
Those
to
see.
G
If
we
see
much
difference
in
these
teleconnections,
I
think
the
connections,
the
atlantic
and
there's
some
there's
at
least
one
or
one
paper
that
was
looking
at
the
at
least
on
the
el
nino
time,
scale
that
the
walk
circulation.
The
connection
between
the
pacific
atlantic
is
coming
mostly
from
the
eastern
tropical
pacific.
Not
so
much
from
the
western
tropics
that
direct
water
circulation
connection
between
the
tropical
atlantic
and
tropical
pacific
is
is
kind
of
more
in
the
eastern
pacific.
G
G
When
you
get
the
precipitation
anomalies,
where
you're
putting
in
the
where
you're
nudging,
the
sea,
surface
temperatures,
you
get
these
anomalous
rossby
waves,
the
pna
pattern
over
the
northern
hemisphere,
psa
pattern
in
the
southern
hemisphere,
and
then
you
get
these
anomalous
loads
in
the
subtropical
atlantic
and
that
ended
up
being
important
for
how
the
atlantic
was
responding
different
from
the
pacific
that
that
would
be
interesting
to
look
at
that
to
see
how
that's
affected,
if
you
actually
can
get
a
better
specification
of
c-service
temperatures
going
further
across
the
tropical
pacific,
but
we
don't-
I
don't
know
if
anybody's
run.
G
Actually
a
western
pacific
pacemaker
just
isolated
on
that
it's
itself.
But
that
may
be
an
interesting
thing
to
do
so.
Maybe
you
could
configure
the
model
and
do
some
runs
like
that.
That
may
be
interesting.
G
C
C
Okay,
so
I
don't
see
anything
else
in
the
chat.
Does
anyone
have
a
new
topic
they
want
to
bring
up?
I
I
do
have
we
do
have
in
the
pre-submitted
question
some
questions
about
large
ensembles.
C
One
of
them
is
what
is
the
best
way
to
initialize
large
ensembles,
so
we
can
start
talking
about
large
ensembles.
If
you
know
that
that
works
for
people
so
flavio,
do
you
wanna
tell
me
how
to
initialize
large
ensembles.
E
I'm
not
sure
I
can.
I
can
tell
you,
I
mean,
there's
different
recipes
and
it
depends
a
little
bit
on
what
you
want
to
achieve,
as,
as
is
often
the
case
with
running
the
model
it
you
start
out
with
a
question
and
the
problem
a
little
bit
with
large
ensembles.
Is
that
especially
the
ones
coming
out
of
ncar?
They
are
always
a
great
community
resource.
E
So
there's
long
discussions
going
on
usually
before
starting
it
in
terms
of
what
is
the
setup
that
is
well
certainly
makes
the
most
sense
scientifically,
but
also
like
is
he
is
useful
for
a
lot
of
different
people
and
same
goes
for
like
which
variables
to
save.
If
we
have
to
make
any
any
any
cuts
there,
then
you
have
to
discuss
for
a
long
time
which
ones
to
save
and
at
which
resolution
etc.
E
I
think
that
the
yeah
so
to
the
extent
that
you're
familiar
with
large
ensembles
like
there
was
the
I
mean-
that's
been
a
thing
for
yeah
over
10
years,
but,
like
I
think,
one
of
the
really
breakthrough
large
ensembles
was
the
csm1
large
ensemble,
because
it
was
a
yeah,
a
big
one.
At
one
degree
resolution
it
was
made
publicly
available,
so
it
was
really
useful
for
a
lot
of
people
outside
of
anchor
as
well
at
universities
and
that
one
was
initialized
from
effectively
the
same
ocean
state,
sea,
ice,
state,
etc.
E
E
It
dominates
sort
of
decadal
variability
initially
before
the
all
these
ensemble
members
start
to
diverge
and
so
for
the
new,
the
csm2
large
ensemble
that
is
now
online.
E
We
did
a
little
bit
of
a
different
setup
where
we
had
different
ocean
initial
states
taken
from
a
long
pre-industrial
control
run
so
that
we
have
different
ocean
initial
conditions,
for
example
the
amok
and-
and
I
should
say
a
little
bit
more
about
that
and
then
for
each
of
those
we
have
small
ensembles
of
like
20
members
that
are
then
again
only
just
perturbed
in
the
atmosphere
so
you're
like
now.
E
You
have
like
a
mix
of
these
different
initial
conditions
and
that
allows
you
to
look
at
different
questions
like
a
terminology
that
is
being
used
for
like
these
initialization
ideas
is
micro
and
macro
micro
would
be,
if
you
just
perturb
the
atmosphere
like
you
make
a
small
perturbation.
If
you
think
about
the
climate
system
as
a
whole
and
then
a
macro
initialization
would
be
yeah
like
very
completely
different.
E
Initial
conditions
like
a
different
ocean
state
and
a
different,
for
example,
a
different
amok
state
at
the
start
of
your
ensembles,
and
these
these
different
setups
allow
you
to
answer
different
questions.
F
Okay,
I
think
one
thing
we
can
think
about
is
the
for
the
foundationally
large
example
and
using
different
whole
things
new
transition.
At
least
we
can
say
for
the
first,
maybe
30,
to
50
years
that
different
ultimate
condition.
It
can
generate
a
bit
larger
screen
for
the
model
result
and
beyond
50
years,
and
it
seems
like
the
variability
becomes
maybe
similar
and
from
my
previous
study
shows
when
we
do
it
at
the
35
to
50
times
here
we
do
see
the
different
different
effects
for
from
ocean
medial
state.
F
That
means
for
me,
especially
if
look
at
the
signal
change
I
can
see
the
see
I
would
change
is
the
viability
I
can
see
larger,
but
the
chain
itself.
I
didn't
really
look
at
the
chain
itself,
but
it
is
pretty
similar
because
and
although
the
unique
ocean
state
is
a
little
bit
different,
but
they
still
belong
to
the
same
climbing
we
start
from
like
the
action
program,
so
the
concurrent
slam
is
spring
are
the
same.
B
Yeah,
I
have
a
question
to
follow
up
to
flavio's
detailed
explanation.
I
would
also
turn
on
my
camera,
but
I'm
still
in
my
pajamas.
B
A
B
Watching
ces
on
videos
yeah,
so
my
question
for
flavio
is:
you
gave
some
exam.
You
mentioned
that
you
know
for
the
macro
large
ensembles
perturbing,
something
very
specific
in
in
the
system.
Let's
say
like
amok
or
or
an
ocean
state,
and
then
you
you
mentioned
that
there
are
different
questions
that
could
be
answered.
Could
you
provide
maybe
like
one
or
two
example,
questions
that
would
be
asked
under
a.
E
E
Yeah,
no,
no,
of
course,
so
I
mean
it
goes
back
to
some
of
the
things
also
that
jerry
talked
about.
So,
if
you're
thinking
about
macro
initialization,
you
you're
starting
from
different
ocean
states,
and
so
you
could
think
of
a
state
where
the
amoc
or
also
the
pacific,
the
cable
variability
or
the
ipo
are,
in
a
particular,
let's
say
in
a
warm
state
or
in
a
high
state,
a
strong
state
and
then
over
the
first
two
decades.
E
This
whatever
this
decadal
variability
index,
is
that
you're
focusing
on
goes
down,
then
that
will
be
imprinted
in
in
in
your
climate
in
general.
But
then,
on
top
of
that,
you
can,
if
you
take
one
of
these
macro,
initialization
states
so
like
so,
you
have
different
macro
ones
like
you
might
have
one
we're
starting
in
the
high
one
and
then
one
starting
in
a
low
one
and
ramping
up
or
one
starting
in
the
middle
and
not
doing
a
lot
over
the
first
couple
of
decades.
E
So
that
gives
you
these
large
differences,
but
then
for
each
of
those
you
could
start
a
micro
ensemble
where
you
just
perturb
the
atmosphere,
and
so,
if
you
think
about
the
first
example
where
you
start
at
the
strong
amox
state
that
might
go
down
over
the
first
decades,
but
then
you
have
10
or
20
members
of
those.
E
E
Then
you,
you
could
look
at
the
predictability
of
like,
for
example,
teleconnections
to
north
america
in
terms
of
precipitation
from
those
ocean
states
or,
for
example,
if
you
start
at
a
warm
ocean
state
and
you
go
to
a
cold
one
in
all
of
these,
then
how
is
this
modulated
by
slightly
different
initializations
of
the
atmosphere?
Do
you
do
all
of
them
show
let's
say
a
drying
over
the
southwestern
u.s
or
not,
so
the
predictability
that
you
get
from
ocean
initial
states
can
then
be
tested
by
having
these
different
atmospheric
initial
conditions.
E
E
B
Awesome
and
quick
follow-up.
While
I
have
you,
I
guess
you
could
also
think
about
like
limits
of
predictability
in
that
way,
maybe.
C
C
Five,
more
minutes.
Okay,
so
I
elizabeth
do
you
want
to
go
ahead
and
launch
that
poll?
We
have
sort
of
a
quiz
for
you
so
and
these
are
what
goes
over
some
like
large
ensemble
questions
and
pacemaker
questions,
and
maybe
you
know
we
can,
in
the
last
five
minutes
sort
of
go
over
what
these
questions
are.
So
I
don't
know
if
it.
G
C
Be
that's
not
enough
time
elizabeth!
Is
that
not
enough
time
to
launch
the
poll
think
she's
trying
to
think
about
it?
Well,
while
elizabeth
is
launching
the
poll
in
the
last
five
minutes
flavio,
do
you
want
to
go
ahead
and
sort
of
explain
the
tension
between
choosing
how
many
ensemble
members
you
you
like
the
having
as
many
ensemble
members
as
possible
versus
what
higher
higher
resolution.
E
Yeah,
well,
it's
just
a
as
you
say
it's
and
like
once
again
it
depends
on
the
research
question
that
you're
asking
so
thinking
back
of
the
tropical
cyclones.
You
probably
do
want
to
invest
some
of
your
resources
into
resolution,
but
then
you,
you
run
into
the
problem
that
your
model
simulation
will
only
have
a
certain
number
of
tropical
cyclones,
because
it's
maybe
only
50
or
100
years
long
and
not
thousands
of
years
long
so
it'll
become
it
becomes
a
bit
harder
to
do
robust
statistics
on
it
and
vice
versa.
E
If
you
have
a
large
ensemble,
you
can
simulate
many
precipitation
extremes,
but
they
are
not
at
a
very
high
resolution.
E
So
might
not
capture
all
these
processes,
and
so
once
again,
it's
like
and
you'll
find
yourself
in
that
situation.
In
your
career,
where
you,
you
have
a
limited
amount
of
computing
resources
and
you
have
to
make
the
decision
that
makes
most
sense
for
your
for
your
project
and
it
takes
a
lot
of
a
lot
of
thinking
and
strategizing
and
it's
usually
a
good
motivation
to
yeah
lobby
for
more
funding
support
for
for
climate
science.
So
we
don't
have
to
make
these
trade-offs,
as
often
as
we
do.
C
You
know
between
what
people
are
saying
in
the
poll,
but
I
think
we've
got.
Let's
see,
I'm
not
sure
we'll
just
go
over
question
one
well
what
we
have
so
the
question
was:
is
climate
models
such
as
csm
can
be
used
to
create
large
ensembles
for
the
scientific
community?
C
What
types
of
analysis
can
be
performed
using
large
ensembles,
so
there's
a
couple
different
choices,
and
I
actually
just
composed
this
question
from
one
of
flavio's
presentations.
So
vladio
should
know
the
answer
to
this
question,
but
the
answer
actually
is
all
of
the
above.
You
can
use
all
of
these
methods
to
all
of
these
things
would
or
large.
Ensembles
are
gonna
help
you
to
answer.
C
C
Continue
to
go
over
it,
although
is
there
any?
Were
there
any
follow-up
questions
to
you
know
what
we
talked
about
in
terms
of
you
know
what
flavia
talked
about
in
terms
of
the
large
ensemble,
a
number
of
ensemble
members
versus
high
resolution,
or
this
first
poll
question
just
go
ahead,
and
you
can
also
just
submit
yourself
and
ask
or
type
in
a
chat.
C
So
the
second
question
I
see
most
people
are
getting
this
right.
Yes,
the
second
question
is
the
correct
it's
b
and
c,
and
it
is
this-
is
on
a
decadal
prediction.
This
was
based
on
one
of
steve
yeager's
presentations
that
you
guys
watched,
which
was,
and
the
question
is
the.
G
C
Climber
prediction
is
not
the
same
thing
is
predicting
the
weather,
10
or
20
years
from
now.
We
are
not
forecasting
individual
events,
we
are
predicting
statistics
of
weather,
for
example
like
computing,
probabilities,
oops
and
so
oops.
I
think
maybe
some
of
the
answers.
Some
of
the
question
was
cut
off,
so
it
was
supposed
to
be.
Please
select
which
best
example,
which
example
demonstrates
a
decadal
prediction
application
and
the
the
best
examples
were
b
and
c
which
are
performing
hindcast
to
assess
whether
or
not
climate
models
can
skillfully
reproduce
climate
change.
C
Statistics
of
the
past
is
one
one
way
and
then
the
other
is
anything
that
has
to
do
with
predicting
skill.
That's
that
site
that
the
sea
was
having
to
do
with
net
primary
production.
For
those
of
you
who
are
land,
people
and
biogeochemistry
for
like
erc,
fluxes
and
stuff
so
and
then
here
so
are
there
any
questions
on
that.
C
A
We
have
a
is
a
break.
It's
like
we
have
a
break
and
shield
tank.
So
if
you
just
want
to
go
and
quickly
play
in
four
and
then
we
can
let
the
branches
go
and
stay
for
question.
C
Okay,
so
so
the
third
one
is
was
a
pacemaker
question
and
the
answer
to
the
question
was:
the
pacemaker.
Type
of
experiment
is
designed
to
determine
effects
of
ssd
forcing
from
one
ocean
region
on
the
global
climate
system.
And
how
do
you
set
this
up
and
the
answer
is
in
a
fully
coupled
climate
system,
specify
your
ssts
in
one
region
and
then
the
rest
of
the
model
goes
fully
unforced
with
a
in
a
coupled
sort
of
context.
The
fourth
question
is:
there
are
many
ways
to
create
ensembles.
C
What
is
the
quickest
way?
So
a
lot
of
these
are
correct,
but
you
I'm
asking
you:
what
is
the
quickest?
The
quickest
way
in
the
csm
system
and
the
quickest
way
is
actually
in
the
nameless
parameter
in
the
cam.
Nameless
there's,
actually
something
called
purtland,
which
is
a
perturbation,
so
you're,
just
adding
a
quick
perturbation,
and
this
would
be
the
micro
change
that
flavia
was
talking
about
in
terms
of
micro
versus
macro
and
then.
Finally,
the
last
question
is
how
many
ensemble
members
are
needed
to
evaluate
natural
variability.
C
C
Launched
this
poll
a
little
bit
earlier,
but
I
thought
we
were
having
a
nice
discussion.
So
I
sort
of
want
to
give
everyone
an
opportunity
to
to
ask
questions
and
thank
you
all
for
participating.
A
Thank
you
guys
for
for
volunteering.
Thank
you.
Thank
you
all
the
panelists
and
in
your
christine
and
we'll
come
back
here
at
10
a.m,
with
the
midas.