►
From YouTube: 7th PAWS Webinar
Description
The seventh webinar from the Paleoclimate Advances Webinar Series (PAWS) which took place on December 2nd 2022.
David McGee discussed "What ITCZ shifts? Exploring controls on tropical Indian Ocean rainfall using stalagmite records from Madagascar" and Alyssa Atwood discussed "Is the concept of a zonal mean ITCZ shift useful?"
For more information and to signup for the PAWS Google Group visit:
https://www.cesm.ucar.edu/events/webinars/paws/
A
So
by
joining
this
meeting,
you
agree
to
follow
the
code
of
conduct
of
ncar
and
new
card,
so
you
will
offer
constructive
feedback
and
consider
new
ideas.
Show
appreciation,
encourage
Innovation,
acknowledge
teamwork
and
share
the
air
with
everyone.
So
the
format
of
this
webinar
is
that
we
have
two
talks.
Each
consists:
20
minutes,
plus
five
minutes
of
questions
and
answers,
and
hopefully,
by
the
end
of
the
of
the
two
talks,
we'll
have
10
minutes
for
General
discussions
with
all
of
the
audience
and
I
will
set
two
minutes.
A
So
today
will
help
us
to
moderate
the
question
and
answer
session,
and,
as
always,
we
welcome
nominations
of
future
speakers.
You
can
use
this
link
to
nominate
future
speakers
and
you
can
also
contact
anyone
of
our
organizing
committee
members,
including
Deputy,
run
brown,
one
myself,
clay
and
John,
and
we're
happy
to
hear
from
you.
A
So
it's
my
great
pleasure
to
introduce
our
two
speakers
for
today's
session.
That's
focused
on
the
itcz
shifts.
The
first
speaker
is
Dr.
David
McGee
from
the
Massachusetts
Institute
of
Technology
Dr
David
McGee,
is
is
an
associate
professor
in
the
department
of
Earth
atmospheric
and
planetary
Sciences
at
the
Massachusetts
Institute
of
Technology,
his
research
group
reconstructs
past
climate
change
using
natural
archives
such
as
the
lag
mites,
Lake
deposits
and
deep
sea
sediments,
with
a
focus
on
understanding
how
and
why
rainfall
patterns
have
varied
over
the
Earth's
history.
A
Dr
Edwards
research
utilizes
a
wide
variety
of
geochemical
and
computational
tools
to
investigate
how
the
short
instrumental
record
can
be
enhanced,
with
paleo
data
data
to
inform
dynamical,
Theory,
establish
test
beds
for
climate
models
and
ultimately
improve
projections
of
future
climate
change
and
I.
Guess
I
forgot
to
introduce
the
title
of
each
speakers,
so
the
topic
of
Dr
David,
McGee's
presentation
today
is
what
itcz
shifts
exploring
controls
on
tropical
Indian.
A
B
B
B
B
B
The
dust
from
the
dusty
Northern
Hemisphere
and
producing
this
strong
gradient
and
so
found
some
cores
from
the
incredible
resource
that
the
ocean
drilling
program
is
that
that
formed
a
transect
across
that
gradient
and
then
used
that,
with
after
a
little
Interruption
from
Hurricane
Katrina
use
that
to
look
at
where
that
gradient
fell
during
the
late
Holocene
and
during
the
last
glacial
maximum
and
I.
Just
continued
thinking
about.
You
know
this,
this
itcc
literature
and
as
I
you
know
continued
through
my
PhD
and
into
my
postdoc
and
and
started
here
at
MIT.
B
There
were
a
number
of
really
neat
studies
that
were
coming
out
that
were
trying
to
put
numbers
on
these
hypothesized
shifts
sort
of
estimating
how
far
the
itcz
had
moved
in
different
regions
of
the
tropics
throughout
the
throughout
the
past
in
different
climates.
So
this
paper
in
2013
suggested
a
seven
degree.
Southward
shift
of
the
Atlantic
itcz
at
the
last
glacial
maximum
paper
from
from
2009
from
the
University
of
Washington
Group,
suggested
a
five
degree
Southward
shift
of
the
Central
Pacific
itcz
during
the
little
Ice
Age.
B
So
pretty
big
shifts
not
quite
as
big
as
were
suggested
by
some
studies
of
of
deeper
time.
So
this
study,
drawing
upon
data
from
the
early
2000s
from
Mitch,
Lyle
and
others,
suggesting
that
the
Central
Pacific
itcz
was
somewhere
up
near
25
degrees
north
in
the
early
Maya
scene,
shown
in
the
black
circles
here
and
so
I
I
was
just
getting
confused
and
I
kind
of
wanted
to
understand.
B
You
know
what
how
far
sorry
I
sort
of
just
wanted
to
know
how
far
the
itcz
could
move
and
so
I
went
and
talked
to
some
of
the
dynamicists
in
our
department,
principally
John
Marshall
and
Aaron
Donahoe.
A
postdoc
working
with
him
and
I
was
expecting
that
they
would
have
some
constraints
based
upon.
You
know
something
dynamical
like
you
know
something
about
Earth's
rotation
or
wind,
stress,
curl
or
something.
Instead,
you
know
Aaron
had
been
heavily
influenced
by
the
work
of
Sarah,
Kang
and
others.
B
And
this
isn't
this
extra
energy.
That's
in
the
northern
hemisphere,
isn't
because
the
Albedo
is
different.
It
isn't
because
the
northern
hemisphere
is
emitting
less
radiation
into
space.
It's
because
of
cross-equatorial
heat
transport
from
the
ocean
and
just
to
put
numbers
on
that.
You
know
roughly
0.4
petawatts
of
cross-secretorial
heat
transport
by
the
ocean
heat
transport,
mostly
in
the
Atlantic,
which
is
compensated
a
bit
by
just
more
radiation
from
a
warmer
surface
in
the
Northern
Hemisphere
and
a
bit
by
cross-equatorial
heat
transport
associated
with
the
tropical
atmospheric
circulation.
B
B
Of
course,
you
know
when
we
were
writing
this
up.
The
thing
that
we
wanted
to
acknowledge
is
that
no
one
experiences
the
zonal
and
annual
mean
itcc,
so
the
the
the
thing
we
we
tried
to
say
was
that
this
suggests
that
probably
there
was
a
lot
more
going
on
at
local
scales
and
seasonal
scales
and
that
this
made
our
work
as
paleoclimatologists
all
the
more
important,
because
probably
the
zonal
and
annual
metric
wasn't
all
that
indicative
of
what
was
going
on
at
any
particular
season
or
any
particular
place
on
the
world
surface.
B
And
so
there
were
real
limits
to
this
energetic
argument,
and
so
having
seen
these
limitations
of
kind
of
trying
to
think
at
a
global
scale,
I
was
really
excited
when
Stephen
Burns
approached
me
about
joining
in
this
collaboration
between
he
and
colleagues
at
UMass,
Amherst
and
colleagues
at
the
University
of
Montana
Revo.
To
look
at
past
rainfall
changes
in
Northwest
Madagascar
using
speleothems
and
I'm,
going
to
very
briefly
describe
one
of
the
key
results
of
our
work.
B
So
part
of
the
the
excitement
here,
of
course,
is
that
Northwest
Madagascar,
where,
where
these
caves
are
lies
directly
under
the
austral,
Summer,
Rain
belts,
and
so
it's
a
great
place
for
for
trying
to
to
reconstruct
southern
hemisphere.
Tropical
rainfall
in
a
part
of
the
part
of
the
world.
That's
not
doesn't
have
an
abundance
of
paleo
records
so
that
the
cave
where
a
lot
of
the
our
work
is
focused
is.
D
B
Bay
cave
in
Northwestern
Madagascar,
this
is
a
region
of
the
island
that
dominantly
gets
its
rainfall
in
austral
summer
shown
in
the
blue
bars
here.
So
a
very
clean
one
season
of
precipitation.
When
winds
reverse
and
draw
draw
moisture
in
from
the
Northwest,
it's
also
a
place
where
we
think
Delta
o18
of
rainfall
is
a
pretty
good
proxy
for
local
rainfall,
though
we've
tried
to
very
verify
that
with
Trace
element
and
Delta
C13
measurements
in
the
same
stalagmites.
B
So
this
is
the
correlation
between
local
precipitation
amount
and
the
simulated
oxygen,
isotope
composition
of
precipitation
using
an
isotope-enabled
reanalysis
product
I,
don't
want
it
all
a
member
of
a
different
group
studying
the
same
region,
and
so
this
is
the
the
stalagmite
that
that
I'll
be
talking
about
today
and
that
Ben
has
been
studying.
It's
called
ab12.
It
was
found
broken
in
the
cave.
B
This
is
Stephen
Burns
standing
over
the
sample
when
it
was
collected,
a
very
heavy
thing
to
carry
out
of
a
cave,
so
I
appreciate
all
the
work
that
went
into
getting
this
thing
out
of
the
cave
and
into
our
into
our
samples
is
about
1.6
meters
long.
So
when
Ben
arrived
at
MIT
after
we
got
out
of
covet
related
lab
lockdowns,
he
set
about
dating
the
sample
and
showed
that
it
formed
between
about
26,
000
and
14
000
years
ago.
B
So
spanning
spanning
Heinrich
Stadium
won
this
period
of
really
the
most
pronounced
perturbations
of
hemispheric
energy
budgets
of
the
last.
You
know
100
000
years
and
some
of
the
biggest
precipitation
anomalies
in
the
tropics
really
around
the
world.
So
we
were
really
excited
to
start
to
develop
the
proxy
record
from
this
stalagmite
and
as
we
were
developing
it,
of
course,
the
thing
that
was
in
our
mind
were
these
records
from
other
parts
of
the
tropics
and
this
in
this
dominant
Paradigm
of
this
anti-phase
Behavior
between
the
northern
and
southern
hemisphere.
B
Of
course,
this
is
a
region
where,
although
on
the
seasonal
cycle,
north-south
rain
belt
shifts-
or
you
know,
Drive
the
seasonal
cycle
in
terms
of
inter-annual
and
longer
climate
variability,
it's
really
East-West
sea
surface
temperature,
gradients
that
are
the
primary
driver
of
of
variability
and
so
on
inter-annual
time
scales.
We
see
that
with
the
the
sea
surface
temperature
variability,
it's
characterized
as
the
Indian
Ocean
dipole,
with.
C
B
Times
of
warmer
than
normal
sea
surface
temperatures
in
the
Western
Indian
Ocean
shown
on
the
right
here,
corresponding
to
higher
rainfall
throughout
the
the
Western
Indian
Ocean
Tropics,
including
over
Northwestern
Madagascar,
the.
So
we
we
were
wondering
well
is
this
drying
during
Heinrich
Stadium,
one
related
to
changes
in
the
East-West
sea.
Surface
temperature
gradient
across
this
time.
B
B
B
You
know
pronounced
changes
in
the
in
in
the
relative
heating
of
the
two
hemispheres
atmospheres,
and
so
now
the
question
that
that
we're
interested
in
looking
into
and
I
think
zhajin
do
at
rice
is
making
some
real
progress
and
thinking
about
is
what
is
changing
this.
What
is
the
mechanism
linking
these
changes
in
the
relative
heating
of
the
hemispheres
and
kind
of
a
meridional
pattern?
B
Speleothem
records
and
I
didn't
show
them,
but
also
some
Lake
records.
For
example,
pollen
records
suggest
that
Heinrich
Stadium
one
was
dry,
Northwestern
Madagascar
and
that
and
that
this
drawing
was
related
to
changing
pronounce
changes
in
the
east
west
Sea
Service
Center
your
gradient
that
appear
to
hold
at
much
longer
time
scales
than
we've
been
able
to
directly
observe
in
the
observational
record.
E
D
D
You
know,
there's
this
IOD
imprint
in
there,
but
it
does
have
this
link
to
kind
of
Rapid
climate
change
event
that
largely
thought
to
be
kind
of
driven
out
of
high
Northern
latitudes.
So
I
was
wondering
what
what
do
you
think
in
in
the
small
number
of
words?
Are
the
kind
of
mechanistic
links
in
there.
B
F
B
Great
I'm
so
glad
you're,
here
and
and
I
should
say,
I
think
just
following
up
on
that
really
quickly.
I
know
we
should
get
to
Alyssa's
talk,
but
it
from
work
done
by
another
group
in
the
area
does
look
like
an
opposite
response
is
observed
during
the
8.2
event,
that
is
Northwest
Madagascar
gets
wetter,
and
so
that
that
again
points
to
what
you
were
just
saying.
B
H
Okay,
how
does
that
look?
Lovely?
Okay?
Well,
yes,
thank
you
so
much
for
having
me
as
well.
It
was
really
a
pleasure
to
be
here-
and
this
seems
like
such
a
fantastic
seminar
series
one
that
I
want
to
start
tuning
into.
So
thank
you
for
organizing
and
thank
you
for
having
me
okay.
So,
yes,
this
is
a
little
bit
of
a
straw.
Man
question,
but
you
know
throw
it
out.
There
is
the
concept
of
a
zonal
mean
itcc
shift,
useful
I,
think!
H
I
am
going
to
show
yeah
I'm
going
to
show
a
few
different
pieces
here.
This
is
kind
of
work
that
bits
and
pieces
of
work.
That's
done
with
different
collaborators,
so
a
lot
of
this
was
done
with
Aaron
Donahoe
and
David
batisti,
and
some
of
the
simulations
on
the
show
are
from
Francesco
posadas
work.
So
this
follows
on
really
nicely
from
David's
talk.
H
I'm
very
happy
to
follow
up.
Some
of
it
also
is
not
published
work
so
not
to
all
of
it
are
in
these
two
okay,
so
Jing
suggested
that
part
of
the
seminar
is
also
just
about
talking
about
kind
of
your
approach
to
science
and
and
a
little
bit
about
you
and
then
providing
some
of
the
early
career
folks
with
thoughts
on
tools
and
and
thoughts
on
sort
of
where
the
field
is
going
and
so
I
just
wanted
to
put
a
couple
slides
on
that.
H
H
Don't
do
so
much
of
that
these
days
sort
of
moves
from
organic
geochemistry
to
inorganic
geochemistry
and
but
yeah,
trying
to
focus
on
kind
of
developing
high
resolution,
paleo
climate
records
from
regions
of
the
tropical
oceans,
where
we
don't
have
a
lot
of
data,
largely
the
Pacific
cold
tongue
region,
so
yeah
I
work
on
the
geochemistry
side
of
things.
I
also
integrate
those
reconstructions
with
models
of
various
complexity
and
dynamical
Theory
to
try
and
inform
these
reconstructions.
So
I
do
some
modeling
I.
H
Do
some
geochemistry
I
do
try
to
do
a
little
bit
of
everything:
okay,
yeah,
Swiss,
Army
knife,
all
right
so
in
terms
of
you
know,
thoughts
about
like
where
the
the
feel
I've
seen
the
field
developing,
and
you
know
things
you
might
want
to
think
about
getting
into
or
or
get
developing
that
understanding
of
for
The,
Graduate
students
and
postdocs
out
there.
You
know
in
the
last
decade,
I
think.
H
H
Data
assimilation
product,
which
is
a
really
interesting,
take
on
on
trying
to
integrate
specifically
hydroclimate
records
with
other
temperature,
sensitive
records
and
then
just
Tierney
and
others
have
been
doing
a
lot
of
really
interesting.
Paleo
da
work
on
things
and
and
tripty
is
here
just
a
lot
of
interesting
stuff
here
on
paleo
D
reconstructions
on
things
like
the
lgm
and
and
deeper
time
scales.
H
So
you
know
there's
a
lot
of
interesting
things
with
paleo
data
assimilation.
You
can
create
these
Global
gridded.
You
know
multi
multivariate
reconstructions
to
really
dive
into
the
Dynamics
behind
behind
paleo
climate
reconstructions,
but
largely
they've
been
focused
on
temperature,
reconstructions
and
temperature
records,
and
there's
there's
sort
of
good
reason
for
that,
because
there's
there's,
there's
been
yeah.
Far
less
attention,
I'd
say
paid
paid
to
paleo
Hydro
climate
records
because
paleo
hydroclimate
reconstructions
are
complicated.
You
know,
precipitation
anomalies
are
far
more
localized
than
temperature
anomalies.
H
So
this
is
just
a
a
figure
we've
put
together
from
this
is
based
on
the
iso
2K
Bronwyn
conic
is
depiction
of
hydroclimate
data
and
then
putting
together
all
these
different
sort
of
proxies
that
inform
upon
different
components
of
hydroclimate
variables,
but
we
can
think
about
kind
of
categorizing
hydroclimate
into
three
major
categories:
that
different
proxy
records
capture,
net,
moisture
balance
or
precipitation.
Minus
evaporation,
the
isotopic
composition
of
precipitation
ocean
salinity
and
for
each
of
these
hydroclimate
variables
are
a
huge
number
of
different
proxies
for
each
of
those.
H
So
I
think
it's
a
very
exciting
way
that
this
field
is
going
to
be
working
on
and
moving
towards
and
all
it
also.
You
know
lots
of
important
discoveries
to
be
made
in
terms
of
reconstructing
hydroclimate
changes
in
the
past
and
how
we
do
that
with
large
networks
of
hydroclinic
data,
and
so
just
a
quick
plug
for
project.
That's
just
being
launched.
H
So,
if
you're
interested
in
getting
involved
and
giving
a
talk
on
it
at
agu
this
year
in
the
water
isotope
systematic
session
information
on
the
bottom
right
here,
we're
also
holding
a
topical
science
meeting
at
the
pages
with
a
Pages
climate
cdos
climate
variability
across
scales
working
group
in
March
of
next
year.
So
there
are
both
in
person
and
virtual
attendance
options.
If
you
want
to
get
involved
and
that
the
link
and
more
for
more
information,
is
there?
H
H
It's
this
really
wonderful,
very
robust,
Theory
linking
zonal
mean
energy
fluxes
in
the
atmosphere
and
its
relationship
to
energy
fluxes
in
the
ocean
to
the
zonal
mean
tropical
rainfall
shifts
and
it's
very
robust
across
you
know:
observational
data
looking
at
the
seasonal
cycle,
as
well
as
in
climate
model,
simulations
of
various
complexities
from
aquaplanet
simulations
to
fully
coupled
gcms
this.
This
linkage
between
zonal
mean
it
seesaceous
and
zonal,
mean
energy
transports
across
the
equator
is
really
robust.
H
David
talked
a
little
bit
about
the
seasonal
Cycles
role
in
all
of
this,
but
you
know
just
to
to
as
a
reminder
of
the
annual
mean
Hadley
circulation.
It's
really
just
a
small
residual
with
these
two
competing
solstitial
cells
with
a
winter
cell
dominating
an
H3
region.
That's
shown
on
the
bottom
plots
here,
and
so,
but
observational
data
tells
us
that
we
can
see
the
position
of
the
zonal
mean
itcc,
it's
linear,
related
to
cross-equatorial
heat
transport
in
the
atmosphere.
H
And
then,
if
we
look
at
that,
if
we
translate
that
relationship
or
we
compare
that
relationship
from
the
seasonal
cycle
to
pass
to
to
this
is
now
for
annual
mean
changes.
But
from
climate
model
simulations
we
see
a
similar
relationship
and
again
I.
Don't
think
I
need
to
go
through
this
in
details
because,
because
David
showed
this,
but
you
know,
if
we
look
at
now,
the
annual
mean
zonal
mean
itcz
shift
in
paleoclimate
simulations.
H
So
that's
interesting
and
again,
you
know,
speaks
to
the
power
of
this
framework
and
when
we're
discussing
zonal,
mean
energy
transports
and
zonal
mean
precipitation
patterns.
But
you
know
we're
really
interested
in
moving
beyond
this
zonal
main
framework
and
exploring
how
the
tropical
response
varies
as
a
function
of
longitude
under
I'm,
going
to
look
at
this
under
a
variety
variety
of
different
climate,
forcings
and
so
I
would
argue,
unless
you're
strictly
interested
in
sort
of
Highly
idealized
representations
of
the
climate
system.
H
It's
the
regional
patterns
of
precipitation
that
are
far
more
interesting
to
study.
They're
really
the
thing
that
matters
when
we're
interested
in
the
societal
or
ecological
impacts
of
changing
rainfall
patterns
when
we're
working
or
when
we're
working
with
paleo
hydroclimate
proxy
records,
which
are
inherently
contained,
Regional
or
local
signals.
H
It's
also
a
really
important
to
consider
these
Regional
changes
when
we're
interested
in
the
dynamical
impacts
of
these
of
changing
rainfall
patterns
and,
if
I
have
time
and
then
touch
on
this
very
briefly
in
the
in
my
last
slide
or
two
we'll
see
if
I
have
time,
but
on
the
sensitivity
of
enso
to
changes
in
the
Eastern
Pacific
rain
band
under
Mill
water,
forcing
so
hopefully
I'll
plan
for
that.
But
I'm
going
to
draw
on
a
large
suite
of
future
and
paleo
climate
model.
H
Simulations
now
to
show
that
the
zonal
mean
framework
is
generally
not
useful
for
characterizing
shifts
of
the
tropical
rain
belt
at
Regional
scales,
regardless
of
the
characteristics
of
the
forcing.
So
that's
that's
my
plan,
so,
firstly,
I'm
going
to
show
you
a
lot
of
maps
showing
changes
in
the
precipitation
centroid
and
before
I.
Do
so
I
just
want
to
give
you
a
sense
of
what
that
looks
like
when
we're
talking
about
how
how
precipitation
centroid
corresponds
to
the
full
2D
precipitation
fields.
H
So
this
figures
comparing
those
two
things-
the
colored
Contours
here-
show
the
mean
annual
precipitation
field
from
observations
and
the
dark,
green
and
dark
black
lines
here
show
the
precipitation
centroid
in
green.
This
is
in
observations
and
in
Black
this
is
a
multi-model
Ensemble
from
cement.
5
pre-industrial
control
runs
so
interestingly,
despite
you
know,
really
well-known
double
itcz,
biases
and
large
tropical
preset
biases,
the
precipitation
centroid
itself
is
fairly
well
captured
in
models.
H
H
The
the
centroid
sits
farther
north
in
the
Eastern
Pacific
and
that's
because
tropical
precept
borns
are
really
tight
band,
just
north
of
the
equator
and
there's
no
southern
hemisphere
counterpart
to
sort
of
drag
it
South,
whereas
in
the
Western
Pacific
the
centroid
lies
in
the
Southern
Hemisphere
and
that's
due
to
the
dominant
presence
of
the
South
Pacific
Convergence
Zone,
so
anyways
looking
across
longitudes.
We
see
this
really
rich
zonal
structure
of
the
precipitation,
centroid
okay.
H
So
this
whole
talk
is
going
to
be
focused
on
shifts
of
this
centroid
there's
a
lot
of
different
ways.
We
can
characterize
Regional
changing
Regional
rainfall
patterns,
but
for
the
purposes
of
this
talk,
I'm
just
going
to
be
focused
on
on
North
South
shifts
of
the
precipitation
centroid,
but
now
at
a
regional
scale,
instead
of
instead
of
averaging
overall
longitudes.
H
So
the
first
of
these
four
things
are
Heinrich
events
or
sort
of
an
idealized
representation
of
Heinrich
events.
You
dump
a
bunch
of
fresh
water
into
the
North
Atlantic
and
it
shuts
down
amok,
induces
this
large
hemispheric
asymmetry
and
heat
transport,
and-
and
we
can
look
at
the
response-
we
expect
a
large
zonal
mean
shift.
But
the
question
is:
to
what
degree
do
we
see
Regional
changes
in
that
shift?
H
Okay.
Okay.
So
let
me
start
with
these.
These
melt
water
simulations
okay,
so
most
of
what
well
I'll
show
you
a
few
yeah
I'll
show
you
all
the
all
the
simulations
that
we
compared,
but
most
of
them
are
the
simulation
that
I
did
with
the
community
or
System
model
added,
either
0.1
sphere
drop
or
one
spear
drip
of
fresh
water
across
the
entire
North
Atlantic
for
100
years.
H
This
is,
you
know,
an
idealized
representation
of
the
meltwater
event,
basically
just
a
way
to
force
the
model
to
shut
down
amok,
see
what
happens
so
in
response
to
this
freshwater,
forcing
amok
weakens
by
about
85
in
20
years,
and
so
we're
looking
at
the
the
rainfall
response
over
the
last
80
years.
These
simulations
and
comparing
them
to
their
respective
control
runs
okay.
H
H
So
this
is
showing
the
change
in
the
precipitation
of
centroid
is
a
function
of
longitude,
and
here
it's
been
multiplied
by
a
factor
of
two
just
for
visual
Clarity
from
these
Maps.
We
can
get
a
sense
for
how
the
tropical
rain
rain
belt
shifts
north
or
south
in
different
regions
of
the
tropics,
so
under
melt
water.
H
Forcing
we
see
large
longitudinal
variations
in
the
rain
belt
shifts
with
pretty
different
responses
across
different
configurations
of
the
model
I'll
get
into
that
a
little
bit
later,
but
in
most
locations
we
see
a
Southward
shift
of
the
rainbow
General.
Generally,
we
see
the
largest
Southward
shifts
in
the
eastern
and
Central
Pacific,
as
well
as
in
the
Atlantic,
but
the
location
and
the
longitudinal
extent
of
the
rain
band
shifts
vary
widely
across
models
or
and
across
different
configurations
of
the
model
and
in
the
central
in
the
bias,
correct.
H
H
Okay,
yeah
I
think
that's
fine,
so
the
I'm
going
to
distill
this
information
into
the
plot
on
the
left-
and
this
is
the
scatter
plot-
is
showing
the
relationship
between
the
zonal
mean
itcz
shift
on
the
y-axis
and
the
standard
deviation
of
the
shift
with
respect
to
longitude
on
the
x-axis.
So
points
inside
the
blue
triangle
indicate
that
the
zonal
variations
in
the
shift
are
larger
than
the
zonal
mean
and
points
outside.
H
So
in
the
case
of
the
Melt
water
simulations,
we
see
that
the
points
sort
of
these
different
models
follow
the
one-to-one
lines
so
that
showing
that
the
the
the
the
The
Zone
the
magnitude
of
the
zonal
mean
shift
is
is
about
as
large
as
the
zonal
variation
of
that
shift,
and-
and
this
is
about
looking
at
all
the
simulations.
This
is
about
as
large
as
the
zonal.
H
This
is
definitely
as
large
as
a
zonal
mean
shift
again
under
no
forcing
is
the
zonal
mean
shift
substantially
larger
than
the
zomal
variation
in
the
shift,
and
so
I'll
just
I'm
not
going
to
go
through
and
detail
these
other
types
of
forcings,
but
just
see
just
to
show
you
look
at
the
relationship
of
these
points
again
now
with
respect
to
this
one-to-one
line.
So
this
is
for
volcanic,
forcing
Northern,
Hemisphere
and
southern
hemisphere.
Volcanoes
follows
sort
of
a
similar
pattern.
H
It's
a
more
muted
response
than
the
Melt
water
forcing,
but
it's
still
the
zonal
variation
is
as
large
as
the
zonal
mean
shift
and
now
I'm
just
going
to
quickly
go
through
for
the
lgm
simulations.
Now
these
for
all
other
simulations,
the
zonal
variation
is
much
larger
than
the
zonal
mean
shift.
So
for
lgm.
That's
the
case
for
the
mid
Holocene,
that's
the
case
and
for
abrupt
CO2
quadrupling.
G
H
We
can
also
we
just
look
at
these
all
together,
I'll
just
end
here,
but
this
is.
We
can
see
a
few
things
out
outside
of
just
looking
at
the.
If
we
look
at
the
sort
of
zonal
structure,
we
see
a
few,
a
few
things:
the
the
in
the
Western,
Pacific
and
sort
of
the
maritime
continent.
The
response
tends
to
be
much
more
muted.
In
the
you
often
see
the
largest
changes
in
the
Central
and
Eastern
Pacific.
H
In
almost
in
no
simulation
is
the
itcz
shipped,
the
Unicorn,
even
across
the
Pacific
basin,
and
in
some
cases
it's
wildly
different,
and
so
yes,
this
is
just
a
a
representation
of
of
of
paying
attention
to
sort
of
this
Rich
zonal
structure
and
then
I'll
just
say
that
the
it
turns
out
I'll
just
give
away
the
the
lead
of
these
differences
across
in
CSM.
With
a
different
bias,
Corrections
in
the
model,
I'll
just
end
with
this,
the
largest
response
in
the
itcz
shift
occurs
without
bias.
H
Corrections,
we
added
various
bias
corrections
to
CSM
and
what
ends
up
being
really
important
is
actually
aurography
over
Central
America,
and
this
was
in
one
of
those
papers
that
I
showed
at
the
Top
If
You
Want
For
more
information
about
that.
But
it's
really
important
in
in
models,
even
in
in
one
by
one
degree,
resolution
atmosphere,
land
models,
orography
over
Central,
America,
there's
a
really
fine
topography
there,
and
so
how
that.
H
Between
changes
in
the
Atlantic
and
the
Pacific
is
there's
there's
too
much
communication
between
those
specific
basins
and
it
affects
the
the
it
makes
the
Pacific
response,
the
tropical
Pacific
more
responsive
to
changes
in
the
Atlantic
than
it
than
it
should
be,
and
that
has
a
variety
of
implications,
including
10,
so
it
turns
out
anyways,
okay,
I
will
leave
it.
There.
E
J
Can
I
ask
why
all
right
so
question
for
both
speakers?
First.
Thank
you
very
much
for
the
nice
talks
and
my
question
is
about
the
potential
narrowing
and
broadening
of
the
itcz
in
past
and
future
climates.
So
I
would
like
to
get
your
thoughts
on
whether
paleo
can
offer
some
like
information
or
constraint
on
this
aspect
of
ITC.
They
change.
Thank
you.
H
Sure
I
mean
I,
certainly
think
that
it
has
the
potential
to
and
again
I
would
come
back
to.
This
idea
of
you
know
with
what
we
could
do
with
large
networks
of
hydroclimate
data,
but
certainly
I
think
with
sufficient
resolution
spatial
you
know
resolution
that
you
should
be
able
to
capture
narrowing
expansions
at
the
tropical
rain
belt
and,
yes,
that's
that's
proving
to
be
to
really
come
out
as
a
line
of
an
important
mode
of
hydroclimate
variability
yeah.
B
And
I'll
just
say
that
I
I
would
love
to
see
some
close
work
between
modelers
and
and
and
observationalists
around,
maybe
using
proxy
system
models.
Just
you
know
in
in
simulations
that
show
a
broadening
or
a
contraction.
What
do
those
simulations
predict
at
important
paleoclimate
sites
for
what
we,
what
we
might
expect
to
see,
because
right
now,
I
think
the
the
use
of
paleo
data
to
infer
past,
broadening
and
and
or
contraction
is,
is
fairly
simple.
B
E
H
Yeah,
that's
a
really
interesting
question.
Ram!
Thank
you
for
that
yeah.
So,
with
respect
to
the
high
resolution,
hosing
experiments
I
would
expect
more
realistic
response
to
at
least
in
the
Pacific
and
the
Atlantic,
because
it
turns
out
that
when
you
raise
topography,
you're
muting
the
response
in
the
Pacific
but
you're
actually
amplifying
the
Atlantic,
because
the
the
wind
pad
it
all
comes
down
to
like
the
wind
patterns
and
the
yeah,
the
the
changes
and
surface
temperature
and
then
precipitation
associated
with
that.
H
So
it's
really
important
to
both
the
Atlantic
and
the
Pacific.
So
I
would
expect
to
see
really
different
responses
and
high
resolution.
Hosing,
simulations
and
I
think
we
should
probably
trust
those
more
and
then
in
terms
of
the
elevation
of
the
Andes.
That's
a
really
interesting
question,
so
in
the
in
Jane
Baldwin
and
and
my
paper
that
I
quickly
flashed
by
we
actually.
So
she
did
look
at
that.
We
looked
at
this
influence
of
raised
orography
in
CSM
I.
H
H
You
know
the
Pacific's,
probably,
as
you
suggested,
sensitive
to
the
Andes
in
addition
to
Central
America
more
than
mountains
anywhere
else,
but
it
turns
out
that
Central
America
seems
to
be
the
most
important
that
Andes
do
also
have
a
a
a
notable
impact
on
the
climatology.
We
didn't
look
at
hosing
in
the
gfdl
model,
though
that
was
only,
and
these
I
only
did
that
in
these
CSM
simulations.
E
H
Yes,
so
it
definitely
would
so
a
few
things,
I
think
what
the
you
know.
What
some
of
my
hosing
simulations
with
and
without
the
flux
Corrections
show
is
that
the
response
is
sensitive
to
mean
State,
biases
and
so
I
did
this
orography
experiment
that
I
also
added
a
surface
heat
flux
correction
in
a
separate
set
of
model
in
a
separate
set
of
simulations
instead
of
raising
orography
just
to
correct
the
SST
biases,
which
are
a
big
part
of
what
gives
rise
to
the
double
itczv
bias?
H
And
you
also
see
big
differences,
not
as
large
as
with
the
raised
neurography.
But
if
you
just
sort
of
fix
SST
as
to
what
we
have
in
observations,
you
also
see
a
sensitivity
in
the
tropical
Pacific
rainfall
response
to
that,
so
that
that
would
suggest
that
this
that
getting
the
climatology
right
is
important
to
the
response.
It
probably
doesn't
shock
anyway,
one
but
yes,
so
and
and
and
the
double
itcd
bias
is
a
big
part
of
that.
E
K
Yeah
thanks
both
of
you
for
such
awesome
talks.
I
guess
I
have
like
a
really
big
picture
question,
which
is
that
so
one
of
the
reasons
that
we're
motivated
in
the
Paleo
world
to
study
rainfall
changes
that
might
tell
us
about
the
idcc
is
that,
ultimately,
we
want
to
understand
like
the
response
of
the
Hadley
circulation
to
four
things
and
I
guess
I'm,
wondering
I
mean
that's
not
the
only
reason,
but
it's
one
of
the
reasons
and
I
guess
I'm
wondering
what
you
would
each
say
about.
I.
K
Does
your
interpretation
of
what's
happening
with
with
this
with?
Maybe
not
the
zonal
mean
rain
belt,
but
with
the
rain
belt
in
different
parts
of
the
world?
Does
that
tell
us
anything
about
how
predictable
these
responses
are
in
the
Hadley
circulation?
Or
is
this
I
mean
it
seems
like
the
centroid
of
precipitation,
should
theoretically
tell
us
something
about
about
what's
happening
with
the
Hadley
cells,
but
is
that
maybe
not
the
case
or
I
guess
I'm,
just
wondering
sort
of
what
does
this
mean
big
picture
in
terms
of
Hadley.
H
H
I
think
that
would
be
a
really
interesting
question
to
probe
and
sort
of
a
you
know
some
yeah
some
some
data
model
integration,
experiments
in
the
so
in
the
simulations
that
we've
done,
the
zonal
mean
usually
follows
what
happens
in
the
eastern
and
Central
Pacific,
because
it's
a
big
Basin
and
because
the
zonal
lean
shifts
are
really
low.
I
mean
the
regional
shifts.
There
are
usually
really
large,
and
then
the
eastern
and
Central
Pacific
usually
follow
what's
happening
in
the
Atlantic,
so
that
those
would
be
really
important
places
to
go.
I
think.
B
And
we
looked
at
this
a
little
bit,
it's
a
great
question
and
and
yeah
it
it's
it's
a
tough
one
right
with
with
paleo
data.
We
looked
at
this
a
little
bit
in
models
and
data
looking
at
Tradewind
proxies,
and
you
know,
as
sort
of
the
surface
expression
of
of
the
the
local
Hadley
circulation,
I
guess
and
in
a
2018
paper
and
found
that
you
know,
as
you
might
expect,
you
do
get
these
opposite
responses
of
the
trade
winds
in
the
two
hemispheres
across
most
of
the
globe.
E
H
I
Yeah
sorry
I
was
just
thinking.
I
was
just
looking
at.
You
know
your
your
Regional,
your
zonal
changes
in
the
itcz.
You
know
when
you
do
sort
of
like
the
hosing
experiment.
It
seems
like
the
models
agree
generally
on
like
directionality
at
least
whereas
for
like
the
lgm
data
you
showed,
you
know
it
seemed
like
some
models.
You
know
a
Zone
will
it'll
shift
up
or
down.
You
know,
there's
a
lot
more
there's
a
lot
more
disagreement.
H
You
know
if
you're,
comparing
the
zonal
variation
to
the
zonal
mean,
but
that
was
there,
and
that
is
what
we
would
expect
if
you
put
large
ice
sheets
I,
think
in
the
northern
hemisphere
and
then
the
same
was
actually
true
of
the
mid
Holocene.
That
again,
the
I
the
zonal
mean
itcc
shift
is
very
small,
but
most
models.
Eight
out
of
ten
models
in
that
case
show
a
slight
northward
shift.
H
So
in
terms
of
why
there
are
why
there's
variation
across
models
and
an
example
of
that
would
be
in
the
CO2
quadrupling.
There's
no
agreement
across
models-
and
you
know
there
are
really
interesting.
Regional
changes,
it's
just
not
there's,
there's
yeah,
there's,
there's
nothing
that
is
pushing
the
the
inner
you
know.
I
would
still
think
about
it.
From
a
from
the
energetic
perspective,
there's
nothing!
That's
really
imposing
a
large
inner
hemispheric
asymmetry
there,
at
least
as
the
models
are
seeing.
E
It
the
next
question,
is
from
Tiffany
Napier,
so
this
is
actually
for
both
speakers,
so
she's
wondering
about
seasonality
and
the
monsoons
and
how
monsoons
relate
to
the
itcz
interpretation.
So
could
you
speak
a
bit
about
how
we
can
think
of
itcz
results
in
terms
of
the
influence
of
various
Monsoon
circulations?
And
obviously
this
is
one
of
my
Topic
near
and
dear
to
my
heart,
so
I'm
curious
to
hear.
B
B
B
H
Yeah
and
I
would
just
completely
agree.
I
think
we
only
get
so
far
in
thinking
about
itcz
shifts
in
general,
like
I've
talked
this
whole
time
about
regional
itcz
shifts
I.
Don't
even
think
that
really
makes
sense.
We
should
just
break
this
down
into
you
know
we
should,
if
you're
interested
in
the
regional
patterns,
you
should
look
for
at
the
regional
patterns
themselves,
and
so
there's
really
interesting,
zonal
and
meridianal
structure
in
there,
and
you
know
we
know
that
there
are
these
really
unique
characteristics
of
tropical
rainfall
patterns
for
very
good
reasons.
B
Yeah
here
I
need
to
out
myself
as
a
paleoclimatologist
who
flips
axes
axes
at
will,
so
that
was
actually
a
positive
shift
in
Delta
18
during
100
Studio
One,
the
I
guess.
The
other
thing
that
I'll
say
is
that
certainly
we
always
have
to
think
about
changes
in
the
isotopic
composition
of
the
vapor.
That's
coming
to
the
site
and
I
think
that's
that'd
be
really
neat
to
to
probe
into
without
trace
or
with
other
with
other
tools,
as
I
mentioned,
but
did
not
show
the
the
trace
element.
B
Data
also
suggests
that
there
is
a
real
reduction
in
moisture
availability
at
the
site
at
this
time
and
pollen
data.
What
else
from
a
lake
in
Northern
Madagascar
would
also
suggest
drying
at
this
time
sort
of
shift
from
more
humid
to
more
vegetation
at
this
time.
So
I
think
there
really
is
a
drying
at
this
time
and
it'll
be
interesting
to
approach
both
the
isotopic
signal
and
and
and
and
the
moisture
signal
and
the
trace
elements.