►
Description
2020 CESM Workshop Day 3 - Wednesday, 17 June 2020 - 8:30-12:00 - Polar Climate Working Group, Paleoclimate Working Group & Land Ice Working Group
The 25th 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 15 June 2020 with presentations on the state of the CESM. On 16-17 June 2020, working groups and cross working groups have half-day sessions, some with presentations and some that are discussion only.
More information about the CESM Polar Climate Working Group:
http://www.cesm.ucar.edu/working_groups/Polar/
More information about the CESM Paleoclimate Working Group:
http://www.cesm.ucar.edu/working_groups/Paleo/
More information about the CESM Land Ice Working Group:
http://www.cesm.ucar.edu/working_groups/Land+Ice/
More information about the workshop:
http://www.cesm.ucar.edu/events/workshops/2020/
Talks included:
Increases in Northern Hemisphere vegetation cover as a potential solution to the Holocene Temperature Conundrum
Moist climate across the North Africa and subtropical Eurasia in mid-Pliocene models and proxies
Can we directly estimate ECS using LGM climate reconstructions?
Progress on variable resolution for polar science
Evolution of the Indian monsoon since the last glacial maximum: forcing vs feedbacks
Speleothems of South American and Asian Monsoons Influenced by a Green Sahara
Dynamic Effect of Last Glacial Maximum Ice Sheet Topography on the East Asian Summer Monsoon
Hydroclimate Footprint Accompanying Pan-Asian Monsoon Water Isotope during the Last Deglaciation
MJO-like Variability in the Eastern Pacific during the Pliocene
Glacial AMOC: Strength versus Depth
Collapse of the Deep-Sea Circulation during the PETM with implications for the extinction of benthic foraminifera
Land Ice, Polar Climate, Paleo Climate Working Groups Discussion
Arctic Freshwater Storage and Export in CMIP6 Models
Opposite responses of the dry and moist eddy heat transport into the Arctic in the PAMIP experiments
Arctic Amplification as a Rapid Adjustment to Increased CO2
Polynyas and Antarctic Slope Current in iHESP-CESM simulation
Integrating proxy and model estimates of Antarctic climate variability over the last millennium
Forced Changes in the Arctic Freshwater Budget Emerge in the Early 21st Century
Detecting Arctic sea ice melt onset with a satellite simulator
The influence of sea ice thickness on transient climate change: a case for tuning from CESM2
Large Arctic Warming from Ozone-Depleting Substances in the Latter Half of the 20th Century
Multi-scale Arctic sea ice modeling with a grid hierarchy in CESM
Atmospheric teleconnection processes linking winter air stagnation and haze extremes in China with regional Arctic sea ice decline
A
C
D
B
E
B
B
D
B
D
A
A
I
I
B
D
C
A
J
F
D
D
A
D
D
A
chat
bubble
at
10:00,
okay,
okay,
Annabel
at
12
and
a
half
minutes,
and
then
yeah
all
right,
it's
8:30!
So
let's
get
started
I'd
like
to
welcome
everyone
to
the
joint
session
of
the
land,
ice,
Paleo,
climate
and
polar
climate
working
group.
On
behalf
of
all
the
other
co-chairs,
my
name
is
Esther.
Brady
I'll
be
moderating
the
first
speakers
just
a
few
things.
To
get
going
to
encourage
a
welcoming
environment.
Please
abide
by
our
code
of
conduct.
I
will
be
recording
this
and
posting
later
online.
D
Please
use
the
chat
for
questions
to
all
of
the
speakers.
If
we
have
a
minute
or
two
during
the
longer
talks,
we
may
be
able
to
read
a
question
from
the
chat,
but
all
of
the
speakers
should
try
to
follow
up
with
the
questioners
in
the
chat
discussion
and
please
use
the
chat
discussions
because
it
can
be
saved
by
everyone.
D
Please
unshare
your
screen
at
the
end
of
your
talk
during
the
talks,
please
mute
your
microphone
and
turn
off
your
video
just
for
more
less
distraction.
If
you
wish
also,
we
have
a
very
tight
schedule,
so
we
are
going
to
stick
to
the
strict
time
limits
for
the
talks
that
we've.
Given
you
the
15
minute
blocks
of
time
you'll.
D
D
K
Going
into
presenter
mode
alright
good
morning,
everyone
I'm
pleased
to
be
the
first
speaker
of
this
session.
I
think
it'll
be
a
great
morning.
So,
like
Esther
said,
my
name
is
Alex
Thompson
I'm,
a
PhD
candidate
at
the
University
of
Michigan
and
I'm,
going
to
talk
about
a
project
I've
been
working
on
that
looks
at
Northern
Hemisphere
vegetation
cover
as
a
potential
solution
to
the
Holocene
temperature.
Conundrum.
I
am
working
on
this
with
collaborators
Jung
Joo
who's
going
to
present
on
something
else.
A
little
later
and
my
advisor
at
Michigan,
Chris
Wilson.
K
So
updating
the
Holocene
temperature
proxies.
A
new
paper
came
out
last
month
with
the
temperature
12
K
temperature
records,
and
this
is
important
because
it
basically
is
showing
the
same
signal
of
the
temperature
proxies.
So
here
we
can
see
from
these
different
records
that
we
have
warmer
temperatures
in
the
early
and
middle
Holocene
and
then
trending
towards
cooler
temperatures
to
the
pre-industrial.
So
this
is
just
showing
that,
with
updated
data,
we
still
have
this
Holocene
temperature
conundrum.
K
So
when
we
look
at
the
proxies,
we
can
see
that
they're
broadly
resembling
the
northern
hemisphere
insulation
signal
where
insulation
is
higher
during
the
early
and
middle
Holocene.
We
can
see
that
with
both
marcotte
at
all
2013,
this
blue
line
and
kauffman
at
all
2020,
but
then
lower
insulation
and
cooler
temperatures
towards
the
pre-industrial.
K
We
also
see
that
the
largest
temperature
response
occurs
in
northern
hemisphere,
high
latitudes,
so
that
is
shown
by
this
top
panel,
where
the
northern
hemisphere
of
30
degrees,
north
to
90
degrees,
north
shows
large
increases
in
temperature
that
we
don't
see
in
either
of
the
tropics
or
the
southern
hemisphere,
as
shown
by
the
other
two
panels.
So
this
is
just
telling
us
that
the
global
temperature
during
the
Holocene
is
largely
driven
by
the
temperature
change
in
the
northern
hemisphere.
K
So,
on
the
flip
side,
when
we're
looking
at
the
models,
these
are
the
transient
models
that
were
used
in
Lou
at
all
2014
to
define
the
Holocene
temperature
conundrum.
And
here
we
can
see
that
they
are
sensitive
to
greenhouse
gas,
orbital
and
ice
sheet
dynamics,
and
we
can
see
in
the
black
lines
for
each
of
these
model
simulations
the
warming
trend
through
the
Holocene
and
largely
these
are
because
of
increases
in
greenhouse
gases
shown
by
the
orange
lines
and
also,
to
a
little
bit
lesser
degree,
the
melting
of
ice
sheets
shown
by
the
green.
K
K
K
Over
1300
individual
records,
so
this
is
a
very
robust
proxy
record.
We
also
see
that
it's
showing
a
consistent
result
from
this
market
for
2013
on
the
left,
and
also
this
logical
pattern
and
somewhat
following
the
northern
hemisphere
insulation
signal.
Now
this
is
not
to
say
that
proxies
don't
have
their
problems
in
themselves.
K
There
is
possible
bias
in
a
proxy
record
towards
the
summer
season,
based
on
the
record
locations,
which
are
only
about
16%
of
the
proxy
records,
do
come
from
the
southern
hemisphere,
so
most
of
them
are
in
the
northern
hemisphere,
but
through
knowledge
of
the
different
types
of
proxies.
There
are
ways
to
calibrate
for
these
biases,
which
Coffman
2020
does
so
that
leads
to
looking
at
the
models
where
the
the
problem
more
likely
exists,
and
mostly
this
is
because
the
models
may
be
missing.
A
mechanism
for
warming,
as
I
showed
a
couple
slides
ago.
K
We
see
that
the
orbital
forcing
temperature
change
is
fairly
minimal
in
these
simulations.
So
if
the
models
may
be
missing
a
mechanism
that
would
be
able
to
amplify
this
insulation
signal
from
the
northern
hemisphere,
so
using
this
concept,
a
couple
studies
have
tested
this
one
is
Lu
at
all
in
2018,
which
completely
removed
global
dust
during
the
early
and
mid
Holocene,
and
found
that
this
did
resolve
the
holocene
temperature
conundrum,
which
was
a
great
step
in
the
first,
a
great
first
step
in
the
right
direction.
K
However,
in
the
early
middle
Holocene
dust
was
more
like
40%
of
the
pre-industrial,
so
this
definitely
represents
an
overestimation
of
the
temperature
signal.
But
a
great
first
step,
Park
at
all
in
2019,
also
showed
that
in
the
middle
all
seen
Arctic
amplification
correlated
to
increases
in
global
temperature.
However,
they
used
an
ensemble
of
simulations
and
even
the
warmest
simulation
that
they
used
was
still
a
little
bit
too
weak
to
fully
resolve
the
temperature
conundrum.
So
these
are
great
first
steps,
but
we
still
conclude
that
we
need
to
find
the
correct
mechanism.
K
We
know
that
vegetation
can
warm
the
climate
through
changes
in
albedo,
as
lowering
albedo
by
adding
vegetation
in
the
northern
hemisphere
has
been
shown
to
shift
the
energy
flux
equator,
as
shown
in
this
figure
on
the
right,
as
well
as
warm
the
temperature
warm
the
climate
through
transpiration
effects
and
vegetation
checks
off
the
key
part
that
northern
hemisphere
vegetation
can
amplify
that
insulation
signal.
So
vegetation
is
promising
as
well
as
we
know
from
pollen
records.
K
The
increased
vegetation
cover
occurred
in
the
early
and
mid
Holocene
in
both
the
African
Sahara,
largely
known
as
the
green
Sahara
and
northern
hemisphere.
High
latitudes
increases
in
boreal
forest
cover,
so
we
know
that
this
occurred
in
the
early
and
middle
Holocene.
So
this
is
a
good
mechanism
to
test.
K
So
with
that,
in
order
to
test
this
mechanism,
we
use
fully
coupled
ICS
m1
x
ice
simulations,
so
this
shows
the
different
time
places
that
we're
running
at
from
12
ka
to
the
pre-industrial
at
3
K
time
slices
and
at
each
of
these
time
slices.
We
run
a
base
simulation
for
900
years
with
the
model
specs
you
can
see
on
the
screen
as
well
as
we
correct
this
for
the
calendar
effect.
K
K
However,
we
now
prescribed
vegetation
throughout
these
types
of
simulations
to
change
the
vegetation
combinations
so
at
each
of
those
time
slices
that
I
showed
before
we
have
a
simulation
that
includes
pre-industrial
conditions.
So
we
have
a
desert
sahara,
like
we
see
on
the
left
here
with
the
bare
P
of
T
and
Arctic
tundra,
as
we
see
on
the
right
with
Arctic
sea
free
grass.
K
However,
for
each
of
the
early
and
middle
Holocene
simulations,
we
run
sensitivity,
experiments
that
show
the
green
Sahara
and
Arctic
boreal
forest
expansion,
green
Sahara,
shown
on
the
top
and
boreal
tree
being
set
to
where
Arctic
c3
grass
was
set
in
the
pre-industrial
on
the
bottom.
So
this
shows
the
sensitivity
experiments
were
at
each
time
slice.
We
have
a
pre-industrial
vegetation
case,
but
then
in
the
sixth
of
9ka
we
added
green
Sahara
and
then
further
add
a
green
Sahara
and
a
boreal
forest
simulation.
K
So
to
look
at
our
results.
I
just
want
to
mention
that
the
data
that
I'm
comparing
to
for
this
model
proxy
comparison
is
from
Coffman
a
tall
2020,
and
these
lines
represent
a
calibrated
annual
line
in
the
black
which
use
the
temperature
proxies
that
have
been
confirmed
to
be
showing
an
annual
temperature
signal
and
we're
calibrating
to
a
temperature
anomalies
here
and
then
in
the
purple.
We
have
calibrated
temperature,
but
these
are
some
of
the
records
that
are
not
necessarily
confirmed
to
be
an
annual
signal.
K
So
when
we
first
look
at
the
no
vegetation
change,
so
these
are
all
the
pre-industrial
vegetation
cases.
We
can
confirm
the
Holocene
temperature
conundrum,
because
we
see
that
the
proxies
are
showing
this
cooling
trend,
but
the
models
are
showing
a
warming
trend.
But
when
we
add
a
green
Sahara
here,
we
can
see
that
in
the
6k
simulation
the
Green
Sahara,
warms
global,
mean
temperatures
to
match
very
well
with
the
calibrated
annual
curve
from
the
proxies
and
mine
ka.
K
The
addition
of
green
Sahara
goes
to
with
an
error
of
the
proxy
record,
but
is
still
a
little
bit
cooler,
so
these
do
represent
an
improvement
of
the
match
with
the
proxy
record
and
then,
when
we
add
boreal
forest
to
those
simulations.
So
now
this
is
including
a
green
Sahara
and
a
boreal
forest.
We
can
see
that
the
9k
a
simulation
matches
very
well
with
the
calibrated
annual
signal,
as
well
as
the
6k
is
a
little
bit
warmer,
but
is
still
within
error
of
that
calibrated
annual
signal.
K
So
we
see
that
the
temperature
proxies
and
the
models
match
much
better
now
through
these
changes
in
northern
hemisphere.
Vegetation.
So,
just
broadly,
to
conclude,
we
see
that
increased
vegetation
cover
is
promising
as
a
missing
mechanism
and
the
models,
and
we
can
see
that
this
may
help
resolve
the
Holocene
temperature
conundrum
and
then
just
broadly.
D
K
K
If,
by
new
land
models,
you
mean
some
of
the
more
recent
updates
like
the
fates
models,
I
would
think
so.
I'm
still
learning
what
the
capabilities
of
these
models
are,
but
by
being
able
to
batter
or
simulate
these
changes
in
vegetation,
especially
in
making
these
dynamic
vegetation
models
that
may
better
capture
some
of
these
changes
in
vegetation.
So
we
don't
need
to
prescribe
to
match
with
proxy
records.
I
think
this
would
be
a
great
improvement
on
some
of
these
cases,
like
the
Holocene
temperature
conundrum.
Definitely.
I
D
H
All
right
good
morning,
everyone
and
tonight
today
I,
would
like
to
share
with
you
some
of
the
new
results
we
gathered
from
analyzing
bit,
pricing
simulations
and
proxy
records,
as
well
as
some
of
the
new
simulations
we
have
performed
so
for
those
who
aren't
familiar
with
mid
pricing.
This
is
a
period
in
our
history
that
we
last
saw
a
pro
long
time
with
400
ppm
co2.
H
So
there
is
a
lot
of
international
interest
in
studying
this
period
to
learn
things
about
model
skills
in
reproducing
such
a
long
period
as
well
as
some
of
the
climate
changes
that
could
have
implications
to
future
climate.
So
this
is
a
work
that
I'm
working
on
in
collaboration
with
trip
T,
but
I
tried
at
Syracuse,
and
some
of
those
simulations
are
being
performed
in
collaboration
with
Betty
and
Esther,
and
a
lot
of
simulation
set
and
you
I
am
using
are
from
the
apply
me
to
work
group
right.
H
In
RCP
8.5
scenario,
the
precipitation,
evaporation,
the
quantity
that
marries
the
net
or
the
flux
to
the
surface
changes
very
differently
between
across
the
ocean
and
on
land,
so
across
the
ocean.
The
change
follows
the
classical.
What
gets
wetter
dry
gets
drier
paradigm
well
on
land,
more
complex,
the
regions
that,
in
close
to
subtropics,
actually
get
wetter,
and
there
are
definitely
regions.
H
Asia
is
broadly
consistent
with
proxy
hydro.
Climate
indicators
will
also
calculate
the
pattern
correlation
between
the
proxy
hydro
climate
indicators
and
endemic
individual
climate.
We
model
simulations,
as
well
as
the
multi
model
Ming,
showing
this
plot
here.
The
metric
will
use
this
Gratz
ac4
model,
simple
proxy
hydro,
climate
indicators,
they're,
typically
categorized
the
hydro
climate
change
into
three
categories,
weather
dryer
or
no
change,
so
I
have
to
do
the
same
for
model
simulations.
H
However,
for
model
simulations,
we
have
to
come
up
with
a
classically
as
a
classification
scheme
to
categorize
the
model,
simulated
changes,
that's
wet
or
dry
and
no
change
and
the
way
we're
doing
it.
This
is
by
using
a
threshold
so
whether
it
would
be
assimilated,
P
minus,
u
changes
in
certain
percentage
of
pre-industrial
and
dryer
would
be
the
but
percentage
that
lower
the
the
potential
for
industrial
and
no
change
would
be
somewhere
in
between,
and,
we
can
calculate
glad
say,
C
as
a
function
of
the
threshold
for
individual
models.
I
was
take-home
message
here.
H
Is
that
for
most
of
the
climate
models,
as
well
as
the
climate,
a
multi
model
average
for
a
range
of
threshold,
the
model
does
seem
to
match
the
proxy
data.
So
this
pattern
of
whitening
in
the
North
Africa
and
the
East
Asian
monsoon
regions
is
robust
across
the
model
and
data,
and
this
calculation
is
done
by
trip
T
at
Circus.
H
There
are
couple
hypotheses
that
have
been
raised
during
the
pack
during
the
past
few
years
to
explain
the
this
wedding
of
continent,
especially
across
the
sub
tropics
during
past
warm
periods
and
then
first
one
of
them
is
this
potential
weakening
of
the
Halle
circulation
in
a
lot
of
past
warm
periods,
including
the
mid
pricing.
What
I'm
saying
is
this
very
strong
amplification
of
warming
in
the
polar
regions?
This
result
in
a
very
much
relaxed
sea,
surface
temperature
gradient
and
under
this
very
much
relaxed
this
of
his
temperature
gradient.
H
H
I
want
me
to
like
you
and
lastly,
more
recently
changes
in
the
land
warming
pattern
is
also
being
noticed
to
explain
a
lot
of
hydro
climate
variability
across
the
land
in
future
climate
simulations
in
general.
The
difference
in
warming
between
the
land
and
ocean
and
the
difference
between
the
continent,
material
and
the
coast
can
actually
drive
some
of
the
variability
in
the
simulated
by
continental
hydro,
climate
response.
H
Alright,
so
in
order
to
investigate
or
quantify
some
of
those
mechanisms,
how
important
some
of
those
mechanisms
in
driving
the
simulated,
P
E
changes
in
our
simulations,
we
further
developed
the
moisture
budget
analysis
following
Seaver
and
Henderson
2013
so
started.
We
start
from
this
general
P
E
budget,
so
P,
minus
e,
equals
of
the
song
hope,
temporal
variability
of
column,
integrated,
moisture,
plus
the
divergence
of
integrated,
moisture
transport.
H
H
So
this
is
a
final
equation.
I
will
walk
you
through
what
all
those
terms
mean
from
the
left
to
the
right.
The
first
term
here
is
changes
in
seasonal
cycle.
We're
applying
this
mostly
to
focus
on
summary
changes.
So
here's
the
first
change
a
single
cycle
and,
as
it
turned
out
this
term,
this
change
is
very
small,
so
we're
not
going
to
show
here
and
the
second
term
is
the
domain
moisture
content.
H
Certain
one
is
changing,
as
you
know,
variability
of
moisture,
followed
by
change
ourselves
in
the
main
circulation
here,
changes
in
stationary
wave
kinetics
and
covering
changes
in
moisture
content
and
horizontal
wind.
And
lastly,
there
are
two
residual
terms.
One
describes
changes
in
surface
pressure,
the
other
one
describes
changes
in
added,
kinetics
and
adding
moisture
transport,
and
those
terms
are
calculated
as
residuals,
given
that
we
don't
have
enough
available
data
and
I
will
show
you
in
a
few
minutes
that
there
are
important
anyway.
H
So
we're
going
to
focus
on
the
CSM
experiments
to
understand
the
drivers
of
increasing
P
minus
e
in
those
monsoon
regions
of
meet
pricing.
We're
going
to
look
at
those
four
sets
of
experiment.
The
first
set
is
a
milk
pricing
simulations
and
a
pre
industrial
control,
and
those
two
simulations
have
been
uploaded
to
see
me
up
six
database.
Those
people
are
interested.
You
can
go
to
ESG
and
download
them
and
I
like
them,
for
your
own
research
purpose.
H
So
for
the
mid-price
and
simulations
the
model
featured
a
400
ppm,
co2
mean
Pliocene
photography
and
the
symmetry
vegetation
I
she
followed
supplied
to
protocol
those
three
additional
sets
of
experiments
are
new
and
are
ongoing.
Hopefully
we
can
get
it
right
up
and
published
and
upload
the
data
to
the
arm
database
as
well
for
the
co2
I
experiment.
We
only
change
the
co2
in
our
experiment
where
everything
else
is
set
to
pre-industrial
for
the
for
the
geography
and
topography
change
experiment,
we
change
the
topography
and
the
symmetry
to
minimizing.
H
All
right
right
anyway,
so
for
those
two
sets
of
experiment,
will
only
change
to
the
topography,
geography
and
or
the
vegetation
okay.
This
is
last
important
size,
and
so
what
we
find
is
for
full
mid
pricing
simulations.
It
looks
like
the
most
changes
in
p,
minus
e
can
be
explained
by
the
vegetation
and
the
ice
cover
changes
in
the
experiment.
H
Co2
actually
accounts
for
very
small
and
change
alright.
So
in
conclusion,
well,
I'm
skipping
the
moisture
budget
analysis
like
if
you're
interested
check
out
the
slides
so
just
to
conclude
in
general
changes
in
similarity
amid
pricing.
P-
is
primarily
driven
by
clinical
greening
and
lots
of
ice
sheet
and
the
probe.
The
mechanism
is
through
enhancing
the
moisture
content
of
the
troposphere
and
altering
the
stationary
wave
kinetics,
and
will
also
find
uncoiling
role
of
Halle
circulation
changes
and
El
Nino
like
SSPs,
sorry,
oh
and
over.
H
D
Ok,
thanks
Ren
I.
Don't
think
we
have
time
for
questions,
but
we
certainly
can.
You
can
certainly
follow
up
in
the
chat
with
some
questions
which
I
actually
don't
see
at
the
moment,
but
we
need
to
get
started
with
our
next
speaker.
All
right,
look.
You
can
unshare
and
our
next
speaker
is
Jean
Joo
who
is
speaking
on.
J
Okay,
good
morning,
everyone
and
John
drew
a
new
project,
scientists
added
a
paleo
and
the
polar
climate
section.
So
today,
I'm
trying
to
answer
a
very
simple
question:
can
we
directly
estimate
the
equilibrium
current
medicine
cetera
T,
using
the
constructions
of
the
last
glacial
Maxima?
So
the
equilibrium
climate
sensitivity
is
a
metric
for
the
global
warming
in
response
to
a
doubling
of
atmospheric
co2
concentration
after
accounting
for
feedback
process,
for
example,
changes
in
Klaus
changes
in
water,
vapor,
exact
ECS
is
an
uncertain
number
and
the
Paleo
climate
constraints
can
be
used
to
inform
ezs.
J
One
major
advantage
of
paleo
climate
constraints
is
that
they
represented
real
world
data
with
not
forcing
and
in
response.
However,
Paleo
temperatures,
your
the
integrates
effects
across
fast
and
the
slow
time
scales,
so
this
temp
skill
mismatch
be
making
it
challenging
to
directly
using
this
Paleo
approach.
As
a
result,
the
pillows
and
project
member.
They
proposed
this
new
approach
in
which
slow
processes
are
treated
as
forces
in
the
calculation.
For
example,
the
changes
in
carbon
cycle
and
changing
eland
algid
are
treated
as
a
forcing
and
you
quantify
the
forcing
with
the
temperature
response.
J
This
approach
has
been
widely
applied
to
different
climates,
including
the
last
glacial
maximum.
A
good
calculation
relies
on
a
good
knowledge
of
the
magnitude
of
their
GM,
cooling
and
also
knowledge
of
the
full
seems
this
regarded
non.
Greenhouse
gases
are
especially
challenging
to
quantify,
which
I
will
talk
about
later
and
also
there's
a
implicit
assumption
that
all
the
different
forcing
agents
they
have
the
same
efficiency
in
changing
surface
temperature.
Another
assumption
is
recalculation,
is
not
fast.
Feedbacks
are
independent
and
slow
processes.
That's
why
you
can
do
this
timescale.
J
So
regarding
the
inland
artstreet,
forcing
the
I
have
a
very
simple
sketch,
showing
the
figure
on
the
right,
so
that's
the
land
ocean
and
a
lower
LGM
sea
level
and
the
presence
of
algae,
my
sheets.
So
traditionally
people
have
quantified
as
a
short
wave,
forcing
associated
with
the
albedo
change
that
is
due
to
the
presence
of
the
iron
sheets
and
the
Shelf
exposure
from
this
simple
sketch.
J
You
can
also
tell
that
for
LGM,
the
long
wave
emission
occurs
at
higher
elevation
and
a
decoder
temperature,
which
means
a
long
wave
radiative,
forcing
in
addition
to
the
radiative
pathways.
They
are
also
non
radiative
pathways
that
I'll
states
can
change
surface
temperature.
It
has
been
long
known
that
the
ice
sheet
can
change.
Surface
will
pet,
which
could
potentially
impact
surface
temperature
and
also
from
the
simple
sketch
you
can
tell
that,
comparing
the
two
configurations,
the
definition
of
surface
differs,
which
means
that
your
surface
temperature
will
be
different
without
involving
changing
radiation.
J
So,
in
light
of
these
complications,
there
is
actually
a
solution.
That
is
that
we
use
the
concept
of
effective
radiative
forcing
and
we
add
another
term,
the
efficacy
of
the
4c,
trying
to
measure
the
efficiency
of
different,
forcing
agents
in
changing
surface
temperature
so
to
calculate
the
effective
radiative,
forcing
eye
contact
Adams
here
only
simulation
with
fixed
pre-industrial
SLT.
Of
course,
I
added
the
LGN
and
acid,
forcing
in
the
simulation.
So
by
definition,
the
top
of
Adams
via
radiation
in
balance
is
a
effective
radiative
forcing
in
this
fixed
as
a
cheek
of
a
collision.
J
So
here
I
got
a
number
of
negative
1
point:
9
watts
per
meter
square,
which
is
a
result
of
the
large
magnitude
of
shortwave
cooling
above
arteries
and
over
places
with
shelf
exposure
and
a
positive,
long
wave
forces
above
ice
sheets.
So
with
running
the
atoms
here
only
model
for
30
years.
We
have
this
nice
spatial
distribution
of
the
40
and
a
nice
uncertainty.
J
Importantly,
using
the
same
model
same
for
C
and
the
units
instead
in
a
slab
ocean
configuration,
so
you
you
can
calculate
the
crime
is
acitivity
using
this
equation
here
and
you
plug
the
lambda.
Here
you
can
correct,
get
a
corrected,
effective,
radiative.
Forcing
the
nice
thing
about
this
approach
is
that
you
can
also
easily
calculate
the
efficacy
of
net
force.
It
will
as
the
expense
of
to
attentional
simulation
with
Dublin
co2
for
C.
So,
basically
the
idea
is
that
you'll
scale
the
sensitivity
to
learn.
J
Here
are
some
results
for
the
length
for
the
LGM
lanighta
enforcing
the
effective
radiative,
forcing
is
quantified
to
be
negative
3
upon
2
once
per
meter
square
and
as
a
efficiency
is
one
point
one
so
it's
higher
and
for
the
LGM
greenhouse
gases,
the
forcing
is
negative.
Two
point:
eight
was
per
meter
square
and
the
efficacy
is
point
nine.
So
it's
lower.
J
A
quick
comment
here
is
that,
with
six
simulations
with
a
total
of
270
years,
we
can
get
a
complete
quantification
of
the
forcing
and
the
efficacy
of
the
algae,
enforcing
and
I
think
this
is
really
cost-efficient.
Very
importantly,
these
results
are
consistent
with
a
concept
of
ecs
which
only
contains
fast
feedback
processes,
and
these
readers
meets
the
requirements
of
forcing
feedback
framework.
Basically,
is
a
POA
energy
imbalance
scales
with
the
surface
temperature
change
so
before
using
these
numbers
to
calculate
easily?
Is
s
let's
quickly,
try
to
understand
why
the
efficacy
is
not
one?
J
So
your
lower?
The
LG
is
the
co
tool
because
I
our
house
gas
concentrations
to
the
AO
GM
level.
You
end
up
with
a
cooling
for
C
and
lower
he
faked
assay
in
order
to
understand
that
I
performed
our
quantified
of
the
climate
feedbacks
and
you
can
see
that
most
of
the
difference
comes
from
the
short
wave
cloud
feedback,
which
means
that,
in
response
to
the
cooling,
forcing
the
cloud
feedback
is
less
efficient
in
changing
surface
temperature
in
amplifying
this
surface
temperature.
J
J
Next,
why?
The
efficacy
from
the
land
I
Street
is
higher
than
one.
This
is
complicated
and
it's
very
interesting
and
I
have
talked
about
before
that
in
response
to
the
land
I
stood
for
see.
We
have
this
big
adjustment
over
the
over
the
ice
sheets,
so
contributing
to
a
global
mean
of
negative
one
point
three.
So
these
adjustments
that's
a
characteristics
of
associated
with
this
for
C.
So
that's
part
of
the
explanation.
J
Basically,
a
larger
adjustments
in
surface
temperature
and
also
more
interestingly,
you
can
say
that
they
are
also
large
magnitude
Cooley
over
the
North
Pacific
and
North
Atlantic
again,
that
shading
is
the
temperature
in
the
Slovakian
simulation
and
on
top
of
the
SST
I
plot,
the
wind
anomalies.
Importantly,
this
is
diagnosed
from
the
atoms
they're
only
simulation
with
the
land
I
stood
for
C.
So
in
a
couple
mode,
this
is
a
couple
system,
the
actually
the
for
C,
so
that
is
associated
with
the
topographic
effects
of
the
land
I
streets.
J
The
nice
respondents
between
the
two
suggest
some
causality
here.
The
idea
is
that
it
is
when
the
pattern
can
change
the
surface.
Temperatures
through
cupholder
feedback
for
examples
of
when
the
evaporation
as
as
T
feedback
and
the
important
to
list
this
as
a
sheepherding
favours,
are
stronger,
close
feedback,
that's
ensuring
in
a
figure
on
the
bottom.
You
have
a
much
higher
cloudy
feedback
over
these
regions
all
together.
J
It
gives
you
a
larger
efficiency
in
changing
surface
temperature,
with
that,
let's
do
some
simple
calculations
to
see
whether
we
can
directly
calculate
Hangul
in
EECS,
using
the
knowledge
current
knowledge
we
have.
So
we
assume
CES
and
1.2
is
a
perfect
model
and
I
make
use
of
the
full
a
couple,
our
GM
simulation
against
in
models
in
forcing
and
has
been
run
to
equilibrium,
and
we
know
because
of
this
simulation.
J
We
know
that's
the
truth,
seeing
LG
m
Cooley
in
say
yes
and
one
point
two
is
six
point:
eight
degrees
C
and
by
conducting
turbine
co2
experiment
in
slab
ocean
configuration
I,
also
know
that
the
truth
is
as
its
three
point.
Six,
the
figure
on
the
right
shows
the
direct
calculation
consuming
TIF
in
the
efficacy
in
the
forcings.
You
can
see
that
no
matter
will
you
to.
You
will
always
overestimate
the
truth
by
close
to
one
degrees
C.
What
is
going
on.
It
turns
out
that
amazing
piece
is
in
the
ocean
dynamic
effect.
J
The
top
line
figure
shows
the
temperature
difference
in
the
between
the
fully
cappadocia
and
the
production.
The
bottom
figure
shows
that
in
the
slab
ocean
configuration,
so
there
is
1.5
degrees.
A
difference
between
the
two,
the
top
right
figure
shows
the
difference
between
the
two.
So
you
can
see
that
by
using
slab
ultra
model
using
print
astral,
Oshin
dynamic
effects
for
industrial
kill
flags,
you
cannot
reproduce
a
large
magnitude
of
over
the
southern
ocean
over
the
east,
no
equatorial
Pacific
and
over
the
sub
tropics.
J
There
are
also
differences
over
no
snow,
no
hemisphere
in
a
poem
figure
I
instead
of
using
pre-dispute,
kill
flags.
I
put
the
cue
flags
from
the
fully
coupled
LGM
simulation.
Now
you
can
see
that
you
can
largely
remove
the
difference
between
the
fully
coupled
and
slab
ocean
configurations.
So
the
main
message
here
is
an
slow
ocean
dynamics
can
amplify
our
GM
cooling
likely
through
impacting
fast
feedbacks,
for
example,
sea
ice
and
cloud
feedbacks.
So
next,
let's
do
another
calculation
in
this
one.
J
We
subtract
the
ocean
dynamic
effects
which
is
1.5
degree
C
from
the
total
area.
Muli
then
I
get
the
rightmost
box
here,
which
agrees
much
better
with
truth
in
the
model.
So
a
quick
summary
panel
cram
it
contains
in
variable
information,
but
we
need
model
to
provide
a
qualification
of
the
forcing
and
efficacy
and
to
explore
the
potential
state
dependence
of
fast
feedbacks
and
slow
processes.
D
D
I'm
sorry,
I
I,
don't
see
a
question:
oh
wait
from
gen
K,
it's
so
cool
how
you're,
bringing
in
understanding
of
The,
Fast
and
Slow
feedbacks,
Thanks,
yeah
and
I.
Just
have
a
quick
question
that
maybe
we
don't
have
time
for,
but
I
wondered
if
I
know,
there's
a
lot
of
different
ice
sheet
reconstructions.
D
J
D
C
You
see
my
point,
sir,
yes
great
well,
everybody
so
I'm
gonna
be
primarily
talking
about
these
Arctic
in
arctic
risk
grids
that
I
generated
that
are
now
part
of
CSM
2.2.
So
anybody
can
run
them.
It's
a
quarter
degree
you
Fineman
over
the
broader
Arctic
and
then
this
Grist
grid
just
throws
an
eighth
degree
or
find
a
patch
over
Greenland.
C
C
So
I
were
in
20
year
historical
simulations
with
six
grits
in
CSM
that
are
available
in
CSM
t2.
We
got
the
final
volume
2
degree,
fine,
dividing
one
degree
km/sec
slam,
which
is
denoted
by
any
30
PG
three
at
one
degree
and
then
km/sec
slam,
one
degree
with
coarser
resolution
physics,
which
is
denoted
as
excuse
me
any
30
PG
and
then
obviously
the
two
Arctic
grids
gonna
be
running
it
for
20
years
from
1979
to
1998,
time-varying,
ssth,
teas
and
aerosols.
This
is
the
F
hist
concept
that
runs
out
of
the
box.
C
Clm
is
satellite
phonology,
most
there's
no
carbon
and
nitrogen
cycling.
Sysm
is
in
no
evolved
mode,
which
means
that
comm
still
uses
computes
the
surface
mass
balance
at
these
multiple
elevation
classes
and
is
used
down
scale
the
surface
mass
balance
to
the
four
kilometer
sysm
grid,
but
it
does
not
drive
a
dynamic
ice
sheet
and
for
the
verbal
resolution
grids,
I
wanted
to
spin
up
the
snow
depths
and
the
surface
mass
balance
from
scratch.
C
C
Another
thing
that
I
think
is
relevant
for
cryosphere
studies
is
prior
to
see
us
I'm
2.2
clm
would
interpolate
a
finite
volume
to
degree
initial
condition.
Data
set
to
your
target
grid
at
initialization,
and
that
can
be
problematic
if
you're
running
it
higher
than
2
degree
resolution.
This
is
a
global
quarter,
degree
simulation
and
it's
the
the
monthly
mean
snow
depths
for
the
first
month
of
the
simulation.
You
can
clearly
see
the
aliasing
of
the
two
great
here,
and
so
what
I
did
is
I
just
ran
out
a
global
quarter.
C
C
Topography
and
ice
mass,
so
in
order
to
compute
consistent
variable
resolution
topography,
you
define
a
refinement
factor
so
in
the
arctic
risk
grid.
It's
one
out
here
and
it's
the
factor
eight
in
here
and
that
dictates
the
smoothing
radius
that
you
apply
to
the
raw
data
set
that
you
eventually
mapped
your
target
grid,
such
that
in
the
Arctic
period.
You
resolve
very
steep
features
and
you
have
a
nice-looking
Greenland
ice
sheet
or
is,
if
you
use
the
smoothing
radius
in
the
coarse
region
over
Greenland,
you
would
have
some
energy
that
looks
like
this.
C
C
C
You're
starting
to
resolve,
obviously
is
orographic
precipitation
along
with
southeast
coast
and
your
ablation
zones,
and
then
here's
where
aquifer
references
so
we're
starting
to
do
pretty
good
you're
getting
some
orographic
precip
over
here
in
the
apprentice
region
and
the
hooli
region.
Up
here.
This
is
sort
of
the
base
of
the
talk.
Now
is
going
to
be
about
how
the
downscaling
seems
to
impact
the
surface
mass
balance,
and
so
the
downscaled
sysm
surface
mass
balance
on
the
bottom
rows
and
for
the
coarser
resolution
grids.
C
It
does
a
good
job
of
recreating
a
high-resolution
surface
mass
balance,
but
for
high
resolution,
we're
really
starting
to
resolve
the
surface
mass
balance.
It
turns
out
that
the
down
scaling
degrades
the
skill
of
the
surface
mass
balance
field.
It
systematically
lowers
the
ablation
rates
and
also
lowers
the
accumulation
rates
that
you
benefited
from
the
high
resolution.
C
So
we'll
look
at
this
a
little
closer.
These
are.
This
is
the
look
at
some
Institute
transects.
The
K
transect
is
a
the
most
popular
transect
in
Greenland,
and
then
this
is
the
lesser
known,
I
think
e
transect
here,
and
so
this
is
just
the
elevation
profiles
in
the
different
grids
where
the
black
is
the
raw
cam
topography
data
set,
and
then
the
different
grid.
F
C
But
it's
just
a
matter
of
the
number
of
grid
cells
used
to
capture
that
steepness,
and
you
can
clearly
see
the
Arctic
and
Arctic
crisp
grids
are
starting
to
capture
that
parabolic
nature
of
the
ice
sheet,
similar
story
with
the
B
transect,
and
so
the
SNB
versus
elevation
on
these
transects.
The
tops
to
be
transect.
The
bottoms
decay
transect
left
panels
of
this
SNB
on
clm
and
then
the
right
panel
is
the
SMB
in
sysm
and
the
b
transect
ends.
It
doesn't
really
go
through
an
accumulation
zone.
C
C
It
starts
increasing
cumulation
rates
and
then
it
gets
out
here
and
it
goes
down
a
little
bit.
So
the
dots
are
the
observations
and
then
that
final
leg
down
to
the
steep
margins
there
you
start
to
get
these
large
accumulation
rates
associated
with
that
orographic
precipitation
and
the
pink
and
the
purple
are
the
smbs
and
the
Arctic
anarchic
Chris
grids
and
you're,
actually
resolving
some
of
that
orographic
precipitation.
You
get
in
close
to
observations
where
your
coarser
resolution
grids
are
not
down
here
on
the
K
transect.
C
C
It
actually
tends
to
increase
the
accumulation
rates
in
this
region
here,
where,
before
it
was
a
little
bit
too
low,
so
just
a
bit
a
metric
go
sort
of
the
strength
of
the
downscaling.
Is
you
take
the
sysm
topography,
the
cam
topography
in
these
different
transects
and
basically,
what
it's
doing
in
the
B
transect?
C
Is
it's
going
to
use
those
higher
elevation
classes
to
do
the
correction
and
in
this
case
it
ends
up
helping
bring
about
a
better
circus
nice
balance
in
both
the
PP
transect
and
in
the
K
transect,
the
nd
30
PG,
3
kind
of
gets
a
double
bonus
because
you're
both
using
lower
elevation
classes
down
here
to
get
you
higher,
ablation
and
then
higher
elevation
classes
over
here
to
get
you
better
accumulation,
and
if
you
recall
from
here
you
anomaly
get
this
any
30.
P3
does
really
well,
and
that
seems
to
be
that
reason.
C
It's
there's
a
large
correction
to
the
down
scaling.
This
is
a
problem
that
is
my
fault.
The
topography
used
in
cam
jam
Ted
is
systematically
lower
than
the
topography
in
sysm.
So
that's
ISM.
The
GM
Ted
is
this
black
line
and
you
notice
as
a
positive
bias,
pretty
much
everywhere
throughout
the
transect.
And
so
even
if
we
had
the
highest
resolution,
cam
model
you're
still
going
to
activate
the
downscaling
because
of
this
difference,
and
so
that
needs
to
be
fixed.
C
I,
don't
think
I
have
much
time,
but
these
are
by
violin
plots
after
leo
and
is
just
showing
the
PDFs
of
the
differences
flipped
on
their
side
of
the
nearest
grid.
Point
to
an
Institute
observation
in
the
Lieb
tiki
toolkit
and
we
break
it
up
by
the
accumulation
zone,
blue
dots
and
the
ablation
done
moves
dots
and
then
the
white
radar
paths
IceBridge
in
the
accumulation
zone.
C
So
that
gives
you
three
different
plots
and
it's
just
showing
that
as
you
go
to
higher
resolution
across
here,
you
tend
to
getting
smaller
biases
and
the
PDFs
are
thinning
out
and
getting
closer
to
observations.
This
horizontal
bar
is
the
median,
and
so
in
this
case-
and
you
can
Institute
accumulation
zone
you're,
getting
that
median
rate
at
the
zero
point
in
the
ablation
zones.
The
biases
are
always
going
to
be
larger.
You
have
larger
melt
rates,
the
accumulation
rates
assume
that's
my
one
or
two
minutes
a
warning.
You
go
to
higher
resolution.
C
You
tend
to
do
better
with
your
accumulation
rates,
PG
2,
PG
3,
and
maybe,
but
definitely
when
you
go
to
Arctic
in
Arctic
Gris.
You
really
start
to
resolve
the
mass
balance
and
here's
what
it
looks
like
on
the
sysm
grid
and
I'll
just
point
out
that
you're
losing
that
skill
that
you
really
gained
from
resolving
the
ablation
zone.
When
you
go
into
the
system,
it's
asthmatic
and
pushes
it
up,
whereas
it
really
does
seem
to
help
these
lower
resolution
grids.
And
lastly,
I
just
want
to
talk
about
this
high
mountain
asian
project.
C
For
one
minute,
we
want
to
see
whether
sixteenth
degree
refinement
can
simulate
the
surface
mass
balance
of
high
mountain
nation.
What
high
mountain
Asia
glaciers.
So
this
is
what
the
grid
looks
like.
They're
fine
effect
are
sixteen
in
here
and
then
I
just
want
to
point
out.
I've
ran
global
quarter,
degree
simulations
and
there
is
no
perennial
snow
and
how
high
mountain
Asia.
C
So
this
is
monthly
means
snow
depths
for
each
month
of
the
year
and
you
can
see,
there's
no
snow
depths
that
survive
the
summer
and
so
we're
hoping
that
resolving
the
topography
is
going
to
help
with
that.
But
more
on
that
later,
this
is
with
Rin
angling
guard
and
so
I'm
going
to
stop
there
because
I
see
them
over.
If
you
have
any
questions,
please
let
me
know.
D
Thank
you,
Adam
I
am
looking
in
the
chat.
I
do
not.
Let's
see
super
exciting
from
Jana
Lin
arts,
fantastic
work,
Adam
and
Marika
says
nice
to
see
good
improvement
in
SMB.
Do
you
think
this
improvement
will
also
lead
to
improved
SMB
changes
over
time
and
we
need
to
wrap
up
in
a
just
a
couple
minutes.
I
think
this
is
our
only
question.
C
D
D
An
I'll
just
tell
people
what's
happening
yeah,
so
we
have
a
series
of
five-minute
talks.
Yon
will
move
the
slides,
some
of
them
will
be
recorded
and
we
won't
have
any
time
for
questions
so,
as
Yan
gets
the
first
one
going,
questions
are
going
to
be
answered,
asked
and
answered
in
the
chat.
Only
thank
you.
M
Okay,
everyone,
so
we
are
speeding
up
a
little
bit,
and
these
this
will
be
a
series
of
seven
lightning
talks,
each
of
five
minutes
and
again,
as
Esther
mentioned
this
is
this
is
going
to
be
pretty
fast.
Todd
will
also
use
the
one-minute
reminder.
There
are
a
few
recorded
talks,
including
the
first
one,
so
I
suggest
we
go.
We
get
started
with
Chetan
Kumar
Charlie.
How
and
he
is
going
to
talk
about
his
work
on
the
monsoon
in
the
LGN.
A
On
one
truth
in
this
work,
the
the
role
of
impacts
in
amplifying
the
orbital
for
objective
is
to
identify
the
feedback
and
then
quantify
the
relative
of
host
exact
effects
to
the
changes
I
want
to.
In
this
simulation,
we
have
used
energetics
ebook.
This
framework
renders
imminency
precipitation,
evaporation
or
muslin
to
me
as
a
function
of
two
variables,
students
and
email
Judith
is
the
dead
energy
present
into
a
smoke
view
from
the
top
and
bottom,
whereas
GM
is
it
across,
but
who
will
90
deg
contribution
from
surface
any
differences?
Businesses
and
therefore
cube
is
primary.
A
The
imprint
like
information
and
Club
3d
feedback
stability
is
essentially
a
function
of
vertical
profile
of
the
ones
that
energy
and
vertical
profile
of
Omega
GMS
can
be
a
success
as
a
unique
function
of
water
vapor.
Therefore,
we
now
have
given
AC
as
a
function
of
students
and
model
shown
in
red
is
the
evolution
of
humanity
over
India
over
the
last
22
years.
It
starts
with
variable
ending
of
the
Botswana
mccurdy
more
depth.
Ripples
by
the
end
of
deviation
could
be
in-house
mm
per
day
and
that
gradually
decreases
and
quality
come.
A
You
think
this
equation
we
have
identified,
which
one
of
these
to
believe
is
to
dip
or
water
vapor.
But
don't
let
duty
do
you
serious,
II
keep
water
in
for
X
in
the
technological
patient
at
a
Nevada
to
do
it
already.
This
gives
the
green
clothes
shown
here.
The
suggest
that
Judith
travel
giving
quality
the
net
energy
flux
as
the
developers
must
use,
decreased
killing
per
se
because
of
profession,
and
producers
must
be
enough.
A
Here
is
the
schematic
intervention
services
through
two
different
paths.
We
do
integration,
it
travels
through
water
vapor
and
do
interval
it.
It
travels
through
energy
of
a
different
energy
available.
A
set
of
segments
using
website
router
Egypt
is
back.
We
could
also
see
it
as
insulation.
The
net
of
insulation,
one
one
come
in,
can
suffice.
Buddy
dr.
de
deeper
reporter
specters
is
the
influence
of
genome
Cassadine.
He,
which
tend
to
be
very
large
during
the
team
issues.
A
M
I
I
So
Alex
already
did
a
good
job
of
presenting
the
mid
Holocene,
so
I'll
skip
that.
But
we
want
to
see
if
a
vegetated
Sahara
and
the
mid
Holocene
could
improve
proxy
agreement,
particularly
in
the
low
latitude
monsoon
regions,
where
the
model
tends
to
underestimate
the
isotopic
change.
So
to
do
this,
we
use
the
isotope
enabled
version
of
the
community
or
system
model.
I
We
ran
three
experiments
of
Korea
industrial
control,
a
six
ka
with
greenhouse
gas
and
orbital
change,
only
which
is
known
as
mid
Holocene
desert
and
a
six
ka
with
greenhouse
gas
orbital
and
a
vegetated
Sahara,
as
suggested
by
reconstructions
known
as
mid
Holocene
veg,
and
here
is
just
plot,
showing
the
difference
in
the
vegetation
between
those
experiments.
Alright
go
ahead
and
advance
the
slide
so
just
jumping
into
the
results
here.
In
the
first
column,
I
have
the
mid
Holocene
desert
pre-industrial
change.
I
In
Delta
80,
no
precipitation
for
the
South,
America
and
Asia
regions,
each
one
of
the
circles
is
the
change
in
the
speed
'litham,
measured
delta
18o
between
the
mid
policy
and
the
pre-industrial
as
well
and
in
the
right
column.
I
have
the
same.
Experiments
are
the
same
plots
with
the
mid
Holocene
vegetated,
pre-industrial,
Delta,
Chino
and
the
same
speeds
and
values
on
top.
I
So
we
can
look
at
this
with
a
simple
scatter
plot
as
well
on
the
y
axis.
Here
we
have
the
model
change
in
the
delta
18o
between
the
mid
Holocene
desert
and
pre-industrial,
and
the
mid
Holocene
vegetated
and
pre-industrial
and
blue
and
orange
respectively,
and
that's
compared
against
this
field
and
change
in
Delta
H.
I
You
know
at
the
sites
I
showed
before
and
so
overall,
as
you
can
see
in
the
orange,
the
model
with
the
veg
suhara
is
significantly
improving
these
relationships
getting
closer
to
the
one-to-one
line
overall,
and
this
improvement
is
due
to
a
combination
of
both
improved
LQK,
no
precipitation,
as
well
as
the
change
in
temperature,
which
also
affects
the
Dell
KACE.
You
know
fractionation
I'm,
at
the
speed
at
them.
A
I
Can
go
ahead
and
guys
slide,
so
don't
really
have
time
to
talk
much
about
the
mechanisms
but
very
quickly.
Here's
precipitation
changes
and
the
top
plot
is
the
change
in
precipitation
between
mid
Holocene
desert
and
pre-industrial
for
Austral
summer,
and
you
have
a
southward
shift
of
the
Atlantic
Intertropical
Convergence
zone
with
the
mid
Holocene
orbital
and
greenhouse
gas
changes.
I
However,
in
the
bottom
plot,
with
the
addition
of
a
vegetated
Sahara,
the
Austral
summer,
itcz
in
the
Atlantic,
actually
shifts
northward
due
to
changes
in
inner
hemispheric
energy
balance,
changes
that
leads
to
less
efficient
precipitation
in
eastern
South
America,
leading
to
an
enrichment
signal
there.
Alright,
you
can
advance
the
slide.
I
So,
looking
at
Northern
Hemisphere
summer
now
and
trying
to
understand
the
Asian
monsoon
response,
we
get
a
northward
shift
to
the
Intertropical
Convergence
zone
with
a
desert
sahara
due
to
orbital
changes
in
the
mid
Holocene.
But
with
a
vegetated
sahara.
We
get
a
even
further
advancement
of
the
ITCZ,
stronger
monsoon
in
Africa
and
India,
and
that
leads
to
more
depleted
water
vapor
getting
into
Asia
and
eventually
leading
into
Southeast
Asia
as
well.
Alright,
so
I'm
at
a
time,
I
think
so
that's
all
I
have
happy
to
answer
any
questions
in
the
chat
or
by
email.
N
O
O
You
know,
observational
studies,
the
impact
of
ice
bottom
on
monsoon
has
often
been
thought
conceptually
from
a
single
commodity,
McNew
and
I
stood
between
in
the
northern
hemisphere
reduces
surface
albedo
and
therefore
these
two
warming
over
the
northern
hemisphere,
which
will
be
hence
the
lens
in
thermal
contrast
and
in
turn
the
same
emotion.
Our
model
results
showed
receding.
Ice
sheets
result
in
a
weakened
ESN,
with
the
reduced
ice
thickness
playing
a
major
role,
which
is
just
opposite
to
what
expected
from
the
simple
thermodynamic
you
whiter.
This
difference
occur.
O
First,
let's
see
the
answer
impact
on
es
than
in
the
model.
Citizen,
380,
31
resolution
simulates
the
years
are
reasonably
well
with
the
ESM.
Characterized
final
sound
isn't
easily
penetrated
their
internal
in
China,
which,
through
the
year
as
some
index
at
the
summer,
family
alumnus
at
850
minute
by
averaging,
is
China
exact.
What
year
the
ESM
index
is
calculated
in
experiments
in
a
hierarchy
of
model
considerations,
the
red
line
indicates
that
the
EAS
awakens
in
response
to
retreating
ice
sheets,
which
contains
the
effect
the
postman,
snoozed
albedo
and
reduce
ashes
thickness.
O
The
black
circles
indicate
that,
when
considering
the
effect
of
reduced
ice
sheet
degrees,
only
the
strengths
of
esm
shows
a
weakening
trend
similar
to
that
in
the
red
line,
suggesting
the
dominant
role
of
high-street
ography.
On
a
year,
sir,
in
response
to
the
removal
of
northern
hemisphere,
I
see
a
topography,
we
can
see
reduced
subtlest
and
hit
813
maneuver
and
reduce
precipitation
over
the
easm
region,
two
mechanisms
in
the
connection
between
the
ice
sheet
and
easm.
O
First,
in
the
judge
shifting
mechanism,
the
Lord
I
stood
over
him
because
in
the
ASM,
through
shifting
the
position
of
the
main
Latino
judge,
controls
show
the
westerly
jet
and
other
and
lower
levels.
Jennings
shows
on
a
wing
anomalous.
The
removal
of
northern
hemisphere
ice
sheet
20
leads
to
a
northward
sheet
of
westerly
jet
and
decreased
the
nominal
weight
over
the
subtropics.
We
now
focus
on
the
mechanism
of
weakening
easm
from
the
moist
static
energy
perspective.
O
Positive
vertical
advection
of
ms
e
correspond
to
vertical
ascending
motion,
in
vice
versa,
because
the
vertical
or
integral
of
the
MS
in
stratification,
culture,
education,
people,
a
travesty,
is
Nexus
host
a
present,
almost
insulting
infection,
hi
mark
Troy
and
so
convection
plays
an
essential
role
in
sustaining
the
main
ring.
Let's
come
back
to
the
JLT
medicine,
the
reduces,
although
new
buddy
jack
in
subtropical
region,
leads
to
reduction
of
the
dryness
of
the
infection
into
east
asia
and
in
turn
a
reduction
of
the
year
as
a
precipitation.
O
Second,
in
the
stationary
wave
mechanism,
the
Lord
I
skipped
over
the
weakest,
the
ESM
through
altering
northern
hemisphere,
stationary
waves,
shadings
our
energy
or
potential
height
anomalies,
are
covered
and
lower
levels.
When
I
Street
ography
is
removed,
the
upper
tropospheric
circulation
is
generated
in
is
less
memory
and
an
anomaly
of
stationary
wave
trains
generated
in
the
mid
latitudes
at
200,
mini
par.
The
way
between
propagates
from
nas
mark
all
the
way
to
East
Asia
along
the
west.
He
gently
guides
the
wave
propagation
causes
an
equivalent
for
chocolate.
O
No
response
over
East
Asia
accompanied
by
reduced
lower
troposphere
externalist,
which
in
turn
reduces
the
adventure
of
dry
enthalpy
and
hence
yes,
M
rain
for
our
Caesars
m3
model
status,
Johnston
were
sitting
ashes,
weakened
the
ESM
with
ice
cuticle
working
playing
a
major
role.
Two
mechanisms
are
implicated
in
the
connection
between
ice
sheet
and
ESF,
one
in
the
judge,
shifting
mechanism
and
the
other
is
the
stationary
wave
mechanism.
Thanks
for
listening.
M
D
D
M
M
M
Show
and
please,
if
you
have
any
questions
for
the
speakers,
it
also
those
speakers
that
have
recorded
their
presentation.
Please
add
them
to
the
chat
it
will
be
stored,
so
we
can
also
reach
out
to
the
speakers.
If
they're
not
presently,
you
know
taking
part
in
this
workshop
right
now,
which
we
can
reach
out
to
them
with
the
questions.
So
let's
go
ahead
and
listen
to
the
next
talk.
Hi.
A
Good
morning
to
us,
my
name
is
Margaret
Chilcotin
from
an
8
year-old
Kayla,
India
and
reversible
climate
change
enter
inside
University.
Today,
I'm
going
to
talk
about
Emma's
over
like
variability
in
the
eastern
Pacific
during
the
players,
micro
producer,
producer,
Sunil
and
and
process
their
hand,
King
and
Joe
is
important,
oscillation
in
the
tropics,
and
it
interacts
mainly
better
systems
globally.
The
many
researchers
found
that
increasing
global
warming
has
included
the
face
plate
and
amplitude
of
the
M
0.
A
However,
variability
of
ENSO
in
the
bombing
environment
largely
unknown
in
this
study,
we
folks
don't
enjoy
availability
during
Pliocene,
which
is
analog
of
near
future.
Climate.
Local
temperatures
are
nearly
3
to
4
degrees
warmer
than
the
perfect
time
I
met
and
for
me,
a
link
to
a
computer,
500
ppm.
So.
A
The
results
are
discussing
now
what
a
peak
in
your
season,
so
this
is
the
global-
mean
surface
temperature
comparisons
between
the
players
in
and
control
simulation.
This
is
the
mean
SST
doing
close
and
an
air
control
and
a
difference.
We
can
see
that
during
the
Pliocene
tropical
eastern
pacific
is
much
warmer
than
the
period
pre-industrial
and
jor-el
radiant
is
very
less
single
plate.
Temperature
anomalies
can
be
found
in
the
surface
temperature.
Also.
A
So
painful
difference
to
scalation
the
most
of
rainfall
is
concentrated
in
players
in
north
of
the
Equator.
This
figure
shows
the
standard
deviation
of
man,
posts,
filtered
rainfall
anomalies.
You
can
see
that
most
of
the
rainfall
is
very
less
during
the
players
in
the
tropics
in
the
Indian
Ocean
and
western
Pacific,
so
similar
features
can
observe
in
the
variants
of
train
for
including
the
winter
in
the
filter
and
unfiltered
rainfall.
Also,
this
figure
shows
the
Mirador
monitor,
creating
difference
between
the
two
simulations.
A
So
we
can
observe
that
in
the
eastern
Pacific
the
gradient
is
very
high.
During
the
Pliocene,
however,
crispa
to
water
is
more
during
the
Pliocene
compared
to
the
pre
initially
because
of
warm
temperatures
during
Pliocene.
This
figure
shows
the
eastward
propagation
of
the
observable
rainfall
anomalous
during
closing,
almost
stationary
in
the
Indian
Ocean
compared
to
the
pre-industrial.
So
in
the
pre-initiation
showing
that
clear
eastward
propagation,
this
figure
shows
the
eastward
propagation
of
the
enjoy
rainfall
anomalous
in
the
Pacific
Ocean.
A
This
figure
shows
the
wave
number
we
can
see
power
spectra,
so
most
of
the
power
is
concentrated
and
you
replace
in
one
to
two
wave
number
one
to
two
so,
whereas
in
the
pre-industrial
it's
spread
to
one
to
five,
this
table
shows
the
base
period
wave
number
comparison
steering
pliers
in
a
pre-industrial
control.
This
field
shows
the
normal
symmetry
component
of
evening
but
retains
the
power
spectrum.
A
This
figure
shows
the
composite
of
band,
pass,
filter,
painful
and
I
always
believe
in
Java
like
so
it
gives
one
phase
1
to
8.
So
during
Pliocene
this
is
very
active
over
the
eastern
Pacific
and
which
was
optioned
in
the
pre-industrial,
so
in
the
pre-industrial
it
is
initiating
over
the
Indian
Ocean
and
it
is
a
western
Indian
Ocean.
So,
whereas
the
quiet
during
cleansing
so
brain
fun,
animal
is
very
active
over
the
central
and
eastern
Pacific.
This
figure
shows
the
MJ
amplitude
in
this
study.
A
Encho
is
active
over
central
and
eastern
Pacific
significantly
and
whereas
the
diffused
over
Indian
Ocean
in
the
stratosphere,
so
the
degrees
internal
temperature
gradient,
plays
a
crucial
role
in
modulating
that
MJ
availability
over
eastern
Pacific.
We
are
mostly
confined
to
the
north
of
the
Equator.
Thank
you.
D
M
G
M
G
So
in
this
study,
which
we've
just
submitted,
we
are
looking
at
the
effect
of
a
mock
strength
versus
depth
on
carbon
isotopes.
And
for
this
we
used
the
Ubik
model,
but
it
is
related
to
recent
study
with
CCS
m3
published
by
Google
and
this
VP
SL
paper
here
on
the
left
and
what
we
were
interested
in.
Initially,
we
had
published
series
of
simulations
that
you
can
see
here
in
the
diagonal
in
this
a
mock
depth
versus
a
mock
strength
diagrams.
G
So
these
this
left
diagonal
here
were
the
experiments
we've
published
before,
where
we
have
looked
at
radiocarbon
on
the
left
and
c-13
on
the
right
and
compared
the
model
simulations
with
sediment
data
from
the
last
glacial
maximum,
and
our
conclusion
in
that
paper
was
that
the
weak
and
shallow
a
mock
fits
the
data
best.
That
would
be
those
simulations
here
around
2400
meters,
a
Monday
and
around
nine
three
drops
of
a
mock
strength.
M
Okay,
that's
that's
too
bad,
so
maybe
let's
wait
a
few
more
seconds
and
if
Andrea
doesn't
come
back
right
now,
I
suggest
we
move
to
our
end
and
maybe
if
we
have
time
we
move
back
to
under
yes.
So
let's
do
that
because
it
doesn't
seem
this
back
so
I'm
not.
Are
you
online
I'm
online?
Can
you
hear
me
yes
perfectly?
So
if
you
are
ready,
we
already.
Let
me
know
if
you
went
to
the
dance
lights
and
okay.
A
Let's
get
started
and
the
present
paper
I'd
like
to
address
the
environmental
parameter
that
could
have
contributed
to
the
extinction
of
benthic,
foraminifera
Roger
and
said
paleocene-eocene
thermal
maximum
also
called
PTM
about
55
to
56
million
years
ago
with
deep
time.
Gsm
simulation
one
degree
ocean
resolution
as
part
of
the
Deep
Web
project.
I
want
to
acknowledge
my
co-authors
and
funding
agencies.
A
The
microhabitat
a
formally
fira
here
after
name
forums,
depend
on
a
critical
food
supply
by
rain
of
organic
material,
available
oxygen
concentration
and
ocean
acidification
shown
on
the
figure,
are
the
extinction
levels
of
panting
forums
at
various
IO
DP
sides
with
the
largest
extinction
level
occurring
in
tropical
and
subtropical
South
Atlantic
Ocean
during
the
PTM
next
slide.
Please.
A
Here
shown
are
the
Global
Emerging
overturning
circulation,
with
the
dominant
dominant
formation
of
seven
component
water
masses
due
to
closure
of
the
Drake
Passage
opening
of
the
Central
American
shear,
a
and
lack
of
ice
masses
in
Greenland
and
Antarctica.
These
funding
are
consistent
with
earlier
studies
from
Ovie
mughelli
vision,
ants
my
Reimer
published
in
1993.
A
vegan
shallow
ocean
circulation
exists
in
the
North
Atlantic
Ocean,
with
increase
in
co2
radiative,
forcing
to
a
PTM
condition
at
six
times
C
or
to
relative
to
pre-industrial.
A
Ever
the
ocean
circulation
mode
shifts
from
a
thermal
mode,
this
polar
deep
water
formation
to
a
line
mode.
The
soft
rubber
fell
in
deep
water
formation,
eventually
a
collapse
of
the
Mach
at
12
times
co2
Levitz
next
slide.
Please
of
importance
for
the
forum.
Any
fair
extension
is
a
stress
due
to
lack
of
food
here
shown
is
a
primary
productivity
of
the
PTM
bit
slower
than
present
productivity,
in
particular
in
the
subtropics,
but
also
in
the
equatorial
upwelling
zones.
A
The
difference
between
the
six
times
and
three
times
scenarios
acquainted
changes
between
the
PTM
and
pre
PGM
conditions
such
a
substantial
reduction
to
productivity.
That
could
have
been
caused
by
reduced
nutrient
supply
to
the
surface
due
to
reduction
of
Ekman
and
youth
upwelling
and
decreased
overturning
next
slide.
Please
next
slide.
A
And
PPG
m83
time,
co2
levels
in
the
South
Atlantic
Ocean
is
about
12
to
13
degrees
warmer
compared
to
present-day
button
water.
In
this
six
times
in
co2
summer,
life
simulation
previous
slide.
Please,
the
water
masses,
the
water
mass
temperature
increased
by
four
to
five
degrees,
Celsius
relative
to
the
switch
and
co2
simulation
agreement
with
value
proxies,
the
left,
oxygen
decline
about
20
to
25
percent,
due
to
reduced
solubility
and
ventilation
of
the
deep
sea,
and
then
the
final
slide
next
slide.
Please.
A
Carthoris
forum,
shirts
all
shall
be
affected
by
the
predicted
increase
in
ocean
acidification
due
to
the
increase
in
pco2
by
about
800
saudi
ppm,
be
the
simulated
increase
is
consistent
to
applause
rise
in
their
lives
of
kine
inferred
by
Jim's
arrows
our
lower
left
figure
and
estimate
by
John
Penman
at
a
lower
right
seeker.
In
summary,
as
a
decline
of
forum,
diversity
at
the
PGM
could
have
been
caused
by
warming
stratification,
a
reduced
food
supply,
deoxygenation
and
increased
acidification,
I'm
happy
to
take
questions
as
a
chat.
M
G
M
G
I'm
not
sure,
really
how
much
you
could
understand
what
I
was
talking
earlier.
But
so
we
essentially
we
had
these
correlated
experiments
in
this
diagonal
here
and
then
we
were
where
we
evaluated
the
circulation
in
terms
of
c-13
on
the
ride
and
radiocarbon
the
left,
and
we
found
as
the
circulation
with
a
cello
and
we
came
up,
would
fit
that
the
LGM
data
the
best,
but
we
could
not
distinguish
having
only
this
diagonal
between
a
mock
depth
and
amok
strengths.
G
So
that
is
kind
of
the
main
takeaway
from
the
from
this
study,
and-
and
it
is
consistent
with
this
recently
published
paper
using
CC
803
by
goo
era,
where
they
had,
they
used
three
simulations
with
CCS
m3,
that
I've
indicated
here.
The
pre-industrial
control,
LG
m
and
LJ
and
hosing
run
where
they
also
found
that
the
both
LD
m
runs,
which
had
similar
aim
of
depth,
but
different
strength
had
very
similar
c-13
and
radiocarbon
distributions
and
therefore
it
was
difficult
to
distinguish
between
those.
G
And
so
we
have
now
filled
in
that
face
place
a
little
bit
more
and
we
are
coming
to
essentially
the
same
conclusion
and
in
the
paper
we
go
on
and
can
I
have
the
next
slide.
Please
and
look
a
little
bit
more
into
the
reasons
for
that.
So
this
one
here
in
the
top
shows
the
the
effect
of
a
mock
strengthening.
G
So
we
pick
two
experiments
from
the
previous
to
face
space
that
have
a
similar
aim
of
depth:
1.9
kilometers
but
different
strength,
14
verses,
7,
shredder,
ups
and
and
there
you
can
see
in
the
in
the
left,
the
c-13
difference
in
the
Atlantic
and
in
the
Pacific.
So
in
this
the
see
here
there's
increases.
So
an
a
mock
strengthening
increases
this
southward
erection
of
high
c-13
at
that
depth
of
North
Atlantic,
deep
water,
which
in
this
case
is
1.9
kilometers.
G
G
M
Maybe
you
go
to
stop
here.
Okay,
yeah
I
think
that
okay
super
thanks
very
much
every
one
participating
in
this
session.
What
we
will
do
now
in
the
schedule
is
a
small
break.
We
are
a
little
bit
delayed,
but
not
much
so
we
will
start
again
at
15
past
the
hour,
so
in
approximately
eight
minutes.
So
please
join
us
again
for
the
discussion,
which
will
be
led
by
winter
leggy
and
Marika
Holland.
Thank
you
very
much.
G
F
F
So,
and
so
welcome
back
after
that
very
short
break
and
a
lot
of
great
talks
this
morning.
So
we
are
gonna,
now
have
a
brief
discussion
and
just
to
to
get
started.
I
just
wanted
to
let
everyone
know
how
we're
gonna
do
this
so
in
terms
of
participating
in
the
discussion,
because
there's
quite
a
few
of
us
on
the
call,
if
you
could
raise
your
hand
and
then
we
will
call
on
you,
so
this
shows
how
to
raise
your
hand.
F
So,
basically,
from
the
zoom
window,
you
can
click
on
participants
which
should
be
at
the
bottom
of
your
screen
and
then
from
the
participant
window.
There's
a
raise
hand
feature,
and
so
you
should
be
able
to
just
click
on
that.
Your
hand
will
show
up
and
we
should
be
able
to
see
you,
and
so
this
is
so.
I
was
actually
planning
to
do
the
checking
of
hands.
But
it's
a
little
hard
for
me,
because
I've
got
my
slide
sharing
so
Gunter.
You
might
have
to
look
at
that.
F
I
can
help
you
okay
and
Dave
thanks.
So
just
again
our
code
of
conduct-
and
you
know
everybody-
be
constructive
and
considerate
in
their
feedback
in
their
discussion
items
and
and
so
Gunther
was
going
to
lead
a
discussion
on
this
CES
I'm
tutorial
and
so
I
will
leave
it
to
Gunther
de
to
take
this
yeah.
N
Maybe
could
also
be
a
good
time
to
rethink
the
way.
We
actually
use
a
tutorial
and
we
would
really
like
to
have
the
input
of
the
community
on
what
on
how
the
tutorial
could
evolve
from
now,
and
so
there
are
five
questions
listed
here,
which
I
would
love
that
integer.
The
entire
tutorial
community
would
love
to
get
your
feedback
on
and
I'm
just
going
to
read
them
out
loud
and
then
wait
for
some
of
your
comments.
N
So
what
should
the
purpose
of
the
tutorial
be
so
typically
the
lectures
and
then
particular
session
with
labs
and
interaction
with
scientists
and
participants
and
different
working
groups?
Should
there
be
early
career,
focused
activities
during
the
annual
the
annual
CSM
workshops,
including
tutorial
activities?
N
If
the
timing
of
the
Tauri
adequate,
it's
always
a
first
full
week
of
August
or
for
other
time,
be
considered
pranks,
for
example,
there
was
a
workshop
to
boost
overall
attendance
function,
for
example,
there's
a
current
format
of
CSM
tutorial
work
that
changes
the
community
would
like
to
see
implemented
and
what
is
target
audience
for
the
tutorial.
I
have
an
international
national,
a
different
type
of
institution
and
so
on.
B
We
raised
the
same
question
in
the
ocean
model
working
group
yesterday
and
there
was
a
woman
on
there,
Lily,
Jones
I
believe
and
she
she
was
from
South
Dakota
and
worked
very
closely
with
the
Native
Americans
there
and
educational
programs
there,
and
so
one
of
her
comments
and
I
don't
see
Lily
on
here
see
even
if
she
was
here.
She
could
correct
me,
but
but
I
mean
I.
A
L
N
B
A
P
Basically,
every
young
scientist
I
work
with
who
wants
to
start
running
CSM
I
recommend
that
they
apply
it
to
the
workshop
and
I'll
continue
doing
so,
basically
forever
I
think
it's
really
important
to
keep
offering
it
at
that
level
for
students
who
are
ready
to
sink
their
teeth
in
and
don't
and
you
know,
but
need
kind
of
a
primer
on
on
how
some
of
the
different
components
of
the
model
works.
I
mean
I.
Think
it's
just
so
useful.
P
The
other
thing,
if
there
are
changes
to
the
format
that
I
would
really
recommend
trying
to
keep
them
as
well
as
possible,
is
the
networking
aspect
which
I
know
is
hard
to
do
virtually,
but
I
just
I
met
so
many
mentors
and
friends
and
colleagues
at
the
tutorial
years
ago
and
I
they're
still
part
of
my
network
today
and
I,
don't
know
where
I
would
be
without
that.
So
definitely,
if
it's
possible
to
keep
that
moving
forward.
I
think
that's
really
important.
N
P
H
H
I'm
happy
to
give
feedback,
if
it's
helpful
know
where
to
direct
that
and
we'll
see
how
that
goes.
But
it
is
really
great
to
have
that
distance
learning
module
not
to
discount
what
Bronwyn
said
at
all
I
think
it
would
be
really
great
for
them
if
there
was
some
kind
of
way
to
network
as
well,
while
they're
doing
this
remotely
during
this
time.
So
a
plus
one
for
both
the
distance
learning
and
having
everything
available
and
for
figuring
out
a
way
to
keep
the
networking
going.
B
Yeah
and
definitely
once
your
students
have
had
a
chance
to
go
through
that
comment
module.
We
very
much
welcome
feedback
on
that,
because
you
know
we
want
to
understand.
Is
it
the
right
level
you
know:
can
they
do
it
at
their
own
pace
for
they're
happy
but
I
think
you're
right.
The
missing
piece
is
that
interaction
with
the
individuals
as
a
part
of
that
yeah
well,.
H
But
also
the
mentoring,
if
they
get
stuck
like
like
this
I,
think
both
of
those
are
are
really
important.
Pieces
of
this
I
mean
I'm
connected
enough.
That
I
can
get
some
help
on.
We
can
figure
things
out,
but
I
don't
know
it,
but
for
people
without
that
connection,
it
would
be
really
difficult
right.
P
Thanks
for
all
the
feedback,
everyone,
so
it
sounds
like
networking-
is
really
important,
which
is
something
we
suspected
what
it
then
makes
sense
to.
Would
people
like
the
idea
of
potentially
saying
like
before
you
come?
We
expect
you
to
watch
the
lectures
online
and
then
the
in-person
part
would
be
more
like
focused
on
you
know
the
actual
labs
and
networking
and
building
connections.
Or
do
you
think
that
having
the
lectures
as
part
of
that
week
is
an
important
part
of.
H
I'll
just
say
I'll
just
say,
having
sat
in
on
the
CSM
tutorial
in
the
lectures.
One
thing
that
really
struck
me
is
during
those
lectures
the
questions
of
the
students
asked
and
how
just
like
bleeding-edge
like
great
ideas,
they
were
and
I
think
like
you're,
missing
that
interaction
between
the
scientists
and
the
students
as
well.
N
N
Right,
I'm
going
to
follow
up
on
this
and
and
yeah
we
what
we
start
was
this
year.
We
thought
it
would
be
tedious
to
have
students,
it's
who
all
the
lecturers
remotely.
That's
why
we
asking
them
to
watch
them
online.
But
what
we're
going
to
do
we're
going
to
have
a
question
and
answer
session
for
about
40
minutes
with
a
panel
of
scientists
where
that
could
be.
N
They
could
ask
whatever
question
they
want
on
the
div
on
the
specific
topic
that
the
panel
is
expert
in,
and
so
this
might
be
something
we
could
adopt
and
so
that
the
student
would
not
sit
through
like
forth
of
lectures
every
day
doing
a
week.
I
don't
know
how
students
feel
about
all
those
lectures
intensive
lectures
during
the
entire
week.
B
They
they
feel
like
it's
just
so
much
material
and
and
they
would
rather
be
doing
the
hands-on
stuff
a
lot
of
the
time
and
I
think
the
problem
is
shockingly
that
there
are
people
who
are
interested
in
the
atmosphere
who
might
not
want
to
pay
attention
during
the
sea-ice
lecture.
You
know
so
so
you
know
so
I
think
it's
it's
a
nice
idea.
If
they
can
sort
of
do
the
lectures
they
want
at
their
own
pace.
Maybe.
F
I'll
say,
though,
that
I
also
think
there's
value
in
getting
that
whole
system.
I
mean
CSM
is
very
complicated,
but
during
the
tutorial
you
get
a
sense
of
them.
It
is
you
know
you
get
a
little
bit.
Even
if
your
expertise
is
in
sea
ice,
you
have
some
sense
of
what
the
atmosphere
is
doing,
those
animals
doing
etc.
So,
there's
there's
some
balance
there
that
it's
hard
to
meet
everybody's
needs,
but
I
think
there's
value
in
having
that
very
comprehensive
overview
of
the
entire
model.
F
So
I
wanted
to
bring
up
another
question,
which
is
it's
not
exactly
on
the
tutorial,
but
it's
related.
You
know
we
have
these.
We've
had
additional
tutorials,
so
there's
the
main
CSM
tutorial,
but
then
there's
then
polar
modeling
workshops.
There's
miss
the
LM
tutorial
and
the
component
model
tutorials
and
I
wanted
to
get
a
sense
from
the
participants
on
the
line
about
the
value
of
those
additional
activities,
because
it's
hard
to
manage
all
of
these
different
things.
F
B
Yeah
I'm,
seeing
in
the
chat
when
there's
not
much
we're,
not
really
advertising
the
comment
site
very
well,
apparently
so
so
definitely
more
work
there,
but
yeah
I,
don't
know,
as
Marika
said,
though,
is
there
other
thoughts
about
these
special
certain,
more
focused
workshops
and
things
like
that?
Like
these
polar
workshops,
we
had
a
site
tutorial
earlier
this
year
as
well.
A
J
O
I
mean
I,
haven't,
attended
them,
but
I
think
those
specialized
ones
are
probably
even
at
least
equally
or
more
important
for
networking
opportunities
because
it
really
brings
together
people
who
are
like
getting
started
and
thus
in
this
in
the
same
general
field
and
I,
think
they're
really
useful
from
what
I've
heard
from
students
who
have
attended
them
and
so
I
think
they're
really
valuable.
They
should
yeah.
It's
not
a
substitute
for
the
other
CSM
workshop,
but
I
think
what's
worked
well
in
the
past,
is
having
that
like
the
polar
one
right
after
the
CSM
tutorial.
O
So
you
don't
have
to
duplicate
the
things
like
okay.
How
do
we
run
the
model,
but
can
then
focus
in
a
little
bit
more,
and
maybe
that
can
be
couple
was
doing
it.
Online
I
haven't
looked
at
the
comment
stuff,
but
that
could
be
it
to.
Maybe
maybe
that's
that's
a
good
starting
point
to
for
people
to
do
that
before
they
come
for
these
specialized
workshops,
and
they
have
that
in
person
like
maybe
a
catch-up
day
where
they
can
ask
other
questions
where
they
got
stuck
or
that
kind
of
yeah
I
think.
Q
B
F
I
think
we
didn't
really
have
enough
time
for
that
during
this
discussion.
If
you're
interested
in
some
joint
experiments
across
the
land
ice,
the
Paleo
climate
and
the
polar
climate
working
groups,
please
send
an
email
to
one
of
the
co-chairs
because
we're
thinking,
we
probably
all
have
some
more
discussion
of
that
later
this
summer
and
we'd
love
to
involve
the
community
in
that
so
send
an
email
to
one
of
us
and
we'll
put
you
on
the
list
for
further
discussion,
I'm
somewhere,
you
know
a
month
or
so
off.
F
F
I
think
that
you
can
share
your
slides
and
unmute
yourself,
and
so
everyone
knows
how
this
is
going
to
work
I'm
at
ten
minutes
into
your
talk,
I'm
going
to
say
that
you
have
two
minutes
left
so
I'm,
just
gonna
talk
over
you
and
and
say
two
minutes
and
and
then
you
should,
you
know,
spend
a
couple
minutes
finishing
up
so
that
we
have
some
times
for
questions
and
and
and
to
transition
to
the
next
speaker.
So
Hanna
I
will
let
you
take
it
away.
I
think
we
can
all
see
your
slide.
So,
okay.
E
E
All
right
so
today,
I'm
gonna
be
talking
to
you
guys
about
Arctic,
fresh
water
storage
and
export
and
seemeth
six
models.
This
is
work,
I've
been
doing
with
Marika,
Holland
and
car
and
then
Alexia
at
CU
as
part
of
a
install
as
well.
So,
oh
goodness,
okay,
so
just
to
start
out,
if
you
don't
know,
the
Arctic
is
one
of
the
fastest
changing
regions
on
the
planet.
I
have
a
couple
figures
here
on
the
Left
showing
a
couple
of
ways
in
which
the
Arctic
is
changing.
E
It's
by
no
means
an
exhaustive
list,
but
the
easiest
one
to
kind
of
focus
on
is
the
figure
up
here
at
the
top,
which
is
the
decline
in
Arctic
sea
ice
extent,
from
1982
to
a
sort
of
2019
I.
Think
so
you
see
this
sort
of
very,
very
strong
decline
there.
Another
way
in
which
the
Arctic
is
changing
is
River
discharge.
E
So
you've
got
things
like
ice
transport
coming
in
and
out
of
some
of
these
straits.
You've
got
ocean
transport
as
well.
There's
fear
transport
in
and
out
of
the
Arctic
precipitation,
evaporation
and,
of
course,
this
riverine
inputs
as
well.
So
it's
actually
a
really
incredibly
complex
system
and,
of
course,
it's
not
actually
just
detached
from
the
rest
of
the
world.
If
an
email,
we
know
that
it's
very
affordable
impacts.
So
that's
why
we
want
to
study
the
freshwater
in
this
region.
So
for
me,
what
I'm,
trying
to
figure
out
here?
E
I
really
would
just
want
to
understand
what
the
state
of
Arctic,
fresh,
water,
storage
and
fluxes
are
in
the
seam
of
six
models
and
how
this
is
going
to
change
in
the
future.
So
the
goal
here
is
to
characterize
seam
in
six
model:
Arctic
fresh
water
storage
and
export
during
the
end
of
the
20th
and
21st
centuries
mm,
and
to
do
that.
E
I'm
gonna,
compare
all
models
from
the
cement,
six
suite
of
models
that
have
at
least
ten
ensemble
members
for
their
historical
simulation
and
that
actually
provide
me
the
relevant
variables
on
ESG,
F
and
believe
it
or
not.
That
is
a
larger
constraint
than
you,
maybe
would
think
so
for
the
cmips
six
simulations
I'll
be
looking
at
the
historical
period.
E
Technically,
this
goes
from
1850
to
2015,
but
when
I'm
looking
at
average
as
I'll
be
looking
at
1980
to
2000
and
then
some
of
these
future
scenarios,
which
are
defined
by
the
shared
socioeconomic
pathways,
I'll
just
be
looking
at
two
of
these
ssp
one.
Two
point:
six
and
five
eight
point:
five.
The
number
here
that
you
maybe
want
to
focus
on
is
is
the
the
number
to
the
right
of
the,
which
is
just
the
radiative,
forcing
at
the
end
of
the
21st
century.
E
So
two
point,
six
watts
per
meter
squared
and
eight
point
five
watts
per
meter
squared
so
to
calculate
storage.
We
one
of
the
ways
you
can
do,
that
is
to
kind
of
cast
it
in
term
or
fresh
water
storages
to
cast
it
in
terms
of
basically
the
amount
of
salinity,
that's
in
that's
in
the
Arctic.
So
what
would
I
do
here
is
I
pick
a
reference
salinity
in
this
case
of
34.8.
This
is
often
what's
used
for
observations.
E
You
calculate
the
amount
of
you.
You
just
take
the
actual
salinity
that
you
have
and
you
look
at
what
the
salinity
is
relative
to
that
so
anything's
sort
of
above
or
below.
That
is
going
to
give
you
fresh
water
or
not
fresh
water,
and
so
in
order
to
make
sure
that
this
term
always
stays
positive,
so
that
your
volumes
are
positive.
You
only
integrate
this
down
to
the
depth
of
this
thirty-four.
I
sell,
Highland
Halen.
Now
export
is
sort
of
similar.
This
is
import
or
export.
It's
just
volume
fluxes
in
and
out
of
the
arctic's.
E
So
moving
on
I'm
going
to
start
by
showing
you
sort
of
what
the
model
fresh
water
storage
looks
like
over
the
historical
period
and
seven
of
these
different
models
that
actually
have
all
of
the
SIRT
of
all
the
constraints
that
I
needed.
So
for
everybody
for
reference,
if
you're
looking
at
CSM,
it's
always
going
to
be,
these
are
always
gonna,
be
an
alphabetical
order.
E
So
four
figures
like
this:
it's
going
to
be
second
from
the
bottom
or
a
second
from
the
left,
so
these
are
actually
box
and
whisker
plots
that
I'm,
showing
here
they're
solid
storage
in
the
liquid
storage.
This
vertical
magenta
line
here
is
the
observations,
so
this
is
averaged
over
1980-2000
you're.
Looking
at
all
these
different
ensemble
members,
which
are
this
is
a
number
of
ensemble
than
the
rest
at
each
of
these
models
have
for
the
historical
period.
So
what
you're
kind
of
seeing
here
is
that
we're
you
know
kind
of
all
over
the
map.
E
Sometimes
somebody
gets
the
storage,
the
solid
or
the
liquid
storage.
Pretty
well,
but
again,
not
surprising
they're,
quite
variable
across
models.
Now
you
can
also
look
at
how
these
the
storage
then
changes
in
the
future.
What
we
end
up
seeing
here
on
the
left
is
ensemble.
This
is
ensemble
mean
freshwater
storage.
E
So
this
is
the
mean
of
all
the
ensemble
members
for
you.they
simulations,
historical
and
gray,
Charon,
socio-economic
pathway,
one
2.6
and
the
Navy
and
585
in
the
sort
of
yellow
color
here,
and
so
this
is
kind
of
how
these
across
models,
how
these
different
stories
terms
change.
So
we
have
stolen
fresh
water
storage,
liquid
and
then
the
total
and
what
you
see
for
the
solid
storage,
of
course
across
all
these
models,
it
declines
and
then
the
liquids
freshwater
storage
typically
increases.
Now,
how
do
we
actually
kind
of
get
there?
E
You
can
kind
of
look
at
the
time
evolution
and
this
figure
on
the
right.
So
this
is
now
multi
mean
fresh
water
storage,
I'm,
just
taking
one
ensemble
member
from
each
of
the
models
in
this
figure
and
I'm
averaging
it.
So
this
is
an
average
of
seven
different
models
to
get
these
lines
here.
So
the
gray
is
just
that
historical
period
and
then
the
SS
peas
again
are
2.6,
then
in
blue
and
then
85
in
in
the
yellow,
color
and
that's
consistent
across
this
talk
here.
E
Some
of
you
see
is
this
very
strong
decline
and
solid
storage,
and
then
corn
have
been
increase
in
liquid
storage
as
well,
and
the
envelopes
are
just
the
maximum
or
minimum
value
than
any
of
the
ensemble
members
attained
over
that
period.
So
you
can
also
look
at
historical,
solid,
freshwater
fluxes,
and
these
are
now
the
fluxes
through
these
straight.
So
for
reference,
there's
always
a
map
down
here
for
you,
the
sign
convention
is
such
that
positive
zones
into
the
Arctic
for
flux
is
negative
out
of
the
Arctic
again.
E
What
I
have
the
vertical
magenta
lines
are
compared
to
observations
over
the
1980-2000
period?
And
now
these
gray
sort
of
shading
regions
are
when
the
observations
actually
had
an
error
associated
with
them,
so
it's
plus
or
minus
that
error,
and
so
again,
you've
seen
this
kind
of
really
large
spread
in
the
models.
Do
you
note
that
some
of
the
size
of
the
actual
spreads
themselves
are
related
to
how
many
ensemble
members
are
within
each
model?
E
They
try
to
get
closer
observations,
but
again
we're
not
they're
kind
of
all
over
the
place,
and
so,
if
you
again
want
to
look
at
what
happens
in
the
future,
so
now
I'm
showing
these
fluxes
through
all
these
different
streets,
and
then
the
total
flux
is
down
here
across
by
adding
up
all
these
5k
plates
together
again,
this
is
across
models.
No
so
CSM
to
here
is
always
in
a
second
from
the
right.
E
You
get
this
really
strong
decline
into
solid
freshwater
fluxes,
and
so
that's
kind
of
consistent
with
what
we
saw
in
the
decrease
in
storage.
The
flux
out
of
these
gateways
also
decline,
and
it's
much
strong,
typically
stronger
for
SP
585
than
it
is
for
one
2.6
same
sort
of
thing.
You
can
plot
this
as
a
time
series
instead,
so
this
is
now
the
multi
model
means
solid
freshwater
flux
through
all
of
these
different
streets.
E
Over
these
periods-
and
again
all
of
these
are
sloped
slowly
trending
toward
zero.
Basically,
so
that's
that
very
strong
shift-
and
so
it's
kind
of
nice
that
the
models
at
least
all
agree
on
that
historical,
liquid
freshwater
fluxes
also
really
very
widely
across
the
models.
Once
again,
we're
going
back
to
this
historical
period
in
1982
2000
through
these
various
streets
doing
a
little
bit
better
in
the
Bering
Strait.
E
Although
I
think
that
maybe
because
I
think
a
lot
of
these,
these
models
may
be
tuned,
but
overall
I'd
say
the
models
seem
to
beginning
at
least
closer
to
doing
the
liquid
fluxes
through
these
streets
compared
to
observations,
for
example,
though,
in
Barrow
Strait,
everybody
underestimates
the
odds,
but
I
PSL
seems
to
at
least
cross
the
observations
in
at
least
four
of
these
regions.
So
that's
that's
doing
pretty
well.
E
If
you
want
to
look
then,
and
the
future,
of
course,
these
liquid
fluxes
show
increased
freshwater
export,
basically
everywhere,
except
for
Bering
Strait
and
the
Barents
Sea,
so
Bering
Strait
kind
of
remains
fairly
stable.
There
are
some
increases,
depending
on
the
model
that
you're
looking
at,
do
note
again,
that
these
axes
are
quite
different,
so
largely
everything's,
driven
by
Fram
Strait
to
some
degree.
And
then,
when
you
look
at
the
barents
sea,
you
actually
see
this
interesting
sort
of
reversal
in
sign
of
the
flux.
E
Almost
there
all
right,
so
future
liquid
flux
is
of
course
showed
this
increased
export
like
I'm,
saying,
Bering
Strait
remains
relatively
stable
over
this
period,
but
then
you
have
this
sort
of
transition
to
the
barren
sea
in
in
sort
of
a
change
in
sign
of
that
flux,
and
so
what
I
wanted
to
do
was
figure
out
what
act
is
actually
happening
there
in
the
barren
sea.
So
I've
done
a
multi
model
mean
liquid,
fresh
water
flux,
decomposition
and
that's
where
you
just
kind
of
decompose,
the
anomalous
fluxes
into
their
volume
and
their
salinity
components.
E
So
the
darker
colors
here
are
for
one
2.6,
lighter
colors
are
for
585,
and
purple
is
for
volume
and
green
is
for
solidity,
and
when
you
look
at
the
barren
see
when
you
end
up
seeing
is
that
what's
driving
that
entire
change
in
the
fluxes?
This
is
entirely
due
to
salinity
rather
than
changes
in
the
volume.
So
again,
you're
not
getting
a
reversal.
L
E
Current
you're
actually
just
getting
a
change
in
the
salinity
and
a
very
strong
freshening,
which
leads
to
sort
of
import
into
the
Arctic
overall,
and
then
you
can
also
see
why
Bering
remains
stable,
largely
because
these
are
compensating
errors.
So
I'm
going
to
just
leave
it
with
conclusions
here
and
if
I'll
just
take
questions
since
I
think
I'm
running
out
of
time.
F
E
Do
freshwater
fluxes
from
Greenland,
particularly
well
yawn?
So
that's
a
little
bit
hard
for
me
to
say
one
of
the
things
I
am
gonna
be
doing
or
I'm
in
the
process
of
doing
is
calculating
fluxes
out
of
Davis
Strait,
and
so
in
that
case,
I
think
we're
gonna
see
probably
a
pretty
strong
impact
on
on
Greenland
in
these,
but
again,
I
think
with
a
lot
of
these
models.
E
F
E
F
Alright,
so
I
think
we
need
to
follow
on
a
follow
up
to
the
next
talk,
and
so,
if
you
have
more
questions
for
Hanna,
please
put
them
in
the
chat
and-
and
we
can
do
it
that
way-
we've
had
a
really
great
chat
discussion,
I
have
to
say
so.
We
will
move
on
to
the
next
talk,
which
is
from
Alexandria
depth
from
the
University
of
Tarte
Toronto.
Talking
about
I'm,
some
results
from
the
PA
myth
experiment.
So
great.
We
can
see
your
slides
and
take
it
away
as
long
as
you're
unmuted,
okay,.
R
R
So,
as
I
said,
I
was
I'm
using
the
EMF
experiments
from
six
different
models
that
were
able
to
provide
variables
that
needed
on
the
adequate
frequency
and
I
use
three
experiments
from
the
PMF
protocol.
A
year
2000,
control
and
future
Arctic
sea
ice
conditions
and
future
sea
surface
temperature
conditions,
and
all
these
experiments
are
emit
simulations.
R
So
no
coupling
with
the
ocean
or
sea
ice.
It's
just
prescribed
sea
ice
and
prescribed
sea
surface
temperature.
So,
let's
take
into
a
surface
response
to
those
PMF
for
seams.
So
I'm,
looking
at
the
moist
static
energy
response
and
dry
psychic
energy
response
as
well
to
see
what
is
the
effect
of
moisture
as
well
in
the
quarter,
heat
transport
response.
R
So
in
response
to
sea
ice
loss
here
and
that'll
be
what
you
can
see
is
in
red
the
more
static
energy
response
and
the
drastic
energy
response
in
blue,
with
the
intermodal
spread
shown
here
in
the
shading
as
you
can.
What
we
see
here
is
basically
poor
amplification
and
it
is
mostly
due
to
an
increase
in
temperature
as
moisture
doesn't
play
a
large
role
here
in
the
Arctic,
but
in
response
to
sea
surface
temperature
increase.
R
We
have
a
tropical
amplified
response
due
to
the
sensitivity
of
moisture
of
to
temperature
increase
in
the
tropics,
and
one
thing
that
we
didn't
see
is
that
the
agreement
between
the
models
is
much
better
response
to
sea
surface
warming
compared
to
sea
ice
loss.
There's
much
more
variability
here
in
response
the
ice
loss.
R
Then,
if
you
look
at
the
MSC
and
DSC
gradient,
which
are
linked
to
the
EDG
flops
with
what
we
have.
Basically
it's
the
gradient
of
panel
B
and
C
shown
in
that
E
and
F,
so
we've
got
a
weakening
of
the
planet
lost
or
gradient,
which
we
could
expect
from
that.
A
weakening
of
the
ad
board
heat
transport.
Previous
studies
have
associated
the
weakening
in
the
MSC
gradient
to
weakening
in
polar
heat
transport
and
then
responds
to
su
surface
temperature.
R
R
So,
basically,
we've
got
a
reduction
in
the
Eddie
polar
heat
transport
that
is
mostly
coming
from
a
reduction
in
dry,
static,
energy
transport
because,
again,
moisture
doesn't
play
a
large
role
at
these
levels
and
their
response
to
sea
as
temperature
we've
got
this
increase
here
of
Eddie
Porter
heat
transport
for
MSC
and
a
decrease
in
the
around
30
degrees
north
and
for
the
DSC
transport.
So
moisture
plays
a
very
large
role
in
these
latitudes
for
the
heat
transport.
R
Another
thing
to
note
is
that
the
spread
between
the
models
is
now
smaller
in
response
to
sea
ice
loss
compared
to
your
response
to
surface
warming,
so
that
that
leads
us
to
ask
the
questions
about
like
what
is
exactly
the
role
of
the
and
how?
How
is
this
weakening
and
poor
heat
transport
accomplished
by
the
weakening
of
the
gradient
that
we
saw
here?
Is
it?
Is
that
linked
very
as
strong?
And
what
is
the
mechanism
between
that?
R
So
to
do
that,
we
use
a
different
approach
compared
to
what
we
did
before
as
just
integrating
the
Eddy
flux.
What
we'll
do
is
for
every
vertical
level
longitude
and
time
step.
We
will
take
a
dry
sight,
energy
pair
with
the
middle
wind,
as
is
shown
here
in
this
panel.
So
with
pressure
and
dry
static
energy
on
the
x-axis
we've
got
a
middle
win
shown
in
the
shading
red
is
positive,
so
northward
and
blue
is
negative
software
and
we
can
do
that
as
well.
We
can,
we
say.
R
Energy
here
is
just
the
SE
first,
for
example,
and
what
is
in
between
those
two,
this
those
two
regions,
the
red
and
blue
region-
is
the
zone,
mean
dry
static
energy.
So,
if
you
are
warmer
than
these
all
mean,
try
static
energy
you're
going
forward,
and
if
you
are
cooler,
you're
doing
equator
work,
and
this
may
don't
win
on
pressure
levels
gives
us
in
mass
flux
that
we
can
then
integrate
into
a
stream
function,
and
this
stream
function
gives
us
the
thermally
direct
circulation
when
we
use
my
static
energy
or
theta
here.
R
So
what
we
have
is
two
counter-rotating
cells
in
one
in
each
hemisphere.
So
basically,
any
heating
that
you
see
here
is
diabetic
heating
and
diabetic
cooling
and
what
you
have
is
to
counter
with
a
cell
clockwise
in
the
northern
hemisphere
and
counterclockwise
in
the
southern
hemisphere.
This
includes
the
effect
of
Eddie's
and
moisture,
there's
also
a
parameterization
of
this
circulation,
that
is,
that
was
provided
by
Paul.
R
We
at
all
in
nineteen
thousand
eighteen
thousand
eleven
in
this
or
2010
I'll
have
would
have
to
check,
but
to
have
a
to
be
able
to
compute
this
a
stream
function
with
not
having
to
do
the
bidding
process
that
I
had
to
do
for
this
block
here.
So
it's
not.
It
is
much
faster
and
also
allows
to
an
explicit
separation
between
the
mean
and
the
eddy
circulation.
So
we
can
only
have
so.
R
But
what
this
allows
us
to
do
is
to
separate
this
heat
flux
into
a
pull
word
and
equate
word
heat
transport
by
just
defining
two
functions
that
are
positive
or
zero
when
it's
forward
or
equate
word.
So
we
can
get
a
pull
word.
An
equator
word,
mass
transport
F
by
just
injuring
M
times,
D
and
D
functions
and
doing
the
same
for
the
heat
transport.
So
we've
got
forward
and
equator
work
heat
transport.
So
we
can
separate
the
two,
the
two
branches
of
this
isentropic
circulation
and
because
of
mass
conservation.
R
R
This
is
a
reduction,
important
heat
transport
that
I
showed
before,
with
the
other
form,
as
with
the
other
diagnostic
of
pH
D,
and
then
by
decomposing
it
into
the
dynamics
and
thermodynamics
part.
So
the
dynamics
here
is
a
dash
dotted
line
here.
It
doesn't
play
a
large
role
in
response
to
sea
ice
loss.
It
is
a
thermodynamics,
the
black
dotted
line
that
plays
a
large
role
and
in
fact
it's
the
warming
of
the
equator
word
branch.
That
is
the
dominant
factor
here.
R
So
what
it
means
is
that,
in
response
to
sea
ice
loss,
it's
not
a
weakening
of
the
eddies
that
creates
this.
This
reduction
in
eddy
polar
heat
transport,
but
it
is
a
warming
of
the
equatorward
branch.
So
a
warming
of
the
air
that
is
coming
from
from
the
polls
that
increases
the
equatorward
heat
transport
and
actually
not
it's
not
a
decrease
in
Porter
heat
transport.
It's
just
in
the
net
that
we
see
it
like
that.
F
R
Thank
you
and
in
response
to
Sisyphus
temperature
increase,
we've
got
a
thermodynamics
that
play
a
larger
role.
I
south
of
four
degrees,
north
and
north
of
that
we've
got
dynamics
that
they
drop,
that
play
a
larger
role.
I
didn't
show
the
individual
branches
here
because
they
are
there's
a
large
translation
between
both
and
the
response
is
just
interesting
when
we
look
at
take
the
net
for
a
therm.
That
I
mean
scared.
R
So
in
summary,
peer
experiments
allow
us
to
you
keenly
separate
of
the
effect
of
sea,
ice
loss
and
sea
surface
warming
on
the
atmospheric
response
and
sea
ice
loss
and
seas
for
swarming
have
a
very
opposite
effect
on
the
surface,
energetic
response,
one
of
them.
So
the
warming
is
at
opposite
ends
of
the
hemisphere
and
the
agreement
between
the
models
on
the
surface,
energetic
conditions
doesn't
seem
to
have
an
effect
on
the
agreement
of
different
models
on
the
eddie
pollard
heat
transport.
R
So
the
agreement
is,
is
poor
response
loss
for
the
surface
MSC
and
DSC,
but
it's
good
for
the
transport
of
those
twenties
or
is
better
for
a
transport
of
those
twenties
and
then,
as
I
showed
so,
the
response
to
the
poor
heat
transport
reduction
that
we
see
is
actually
a
an
increase
in
equatorward
heat
transport
due
to
that
is
thermodynamically
driven.
So
the
air
returning
from
the
poles
is
warmer
than
when
we
lose
us
yes
and
then
responds
to
cease
force
warming,
the
thermodynamics
dominate
at
lower
latitude
and
high
latitude.
R
F
F
R
F
R
K
F
If
there's
you
know,
if
it's
all
about
the
ice,
concentration
or
thickness
is
playing
some
role,
yeah
also,
all
right
I
think
we
should
move
on
so
again
put
questions
into
the
chat.
So
thanks
for
the
great
talk-
and
so
our
next
speaker
is
Tyler
Janoski
from
Columbia
and
Tyler.
If
you
can
share
your
screen.
Q
All
right,
my
name
is
Tyler
Janaki
I
am
a
PhD
soon
at
Columbia.
This
work
was
done
with
my
advisers
mikrobitti
and
learn
zouk
Levani,
along
with
Gabriel
Kyoto
and
Karen
Smith
and
Wendy,
showing
how
Arctic
amplification
can
be
considered.
A
rapid
response
to
radiative
forcing
so
Arctic
amplification
can
be
simply
defined
as
enhanced
Arctic,
surface
warming.
So,
in
this
plot
is
the
change
in
surface
air
temperature
averaged
over
the
last
30
years
of
seemeth
5/4
times
co2
simulations.
Q
You
can
see
that
the
highest
latitudes
of
the
Arctic
were
more
than
14
degrees
and
that
warming
is
considerably
less
in
mid-latitudes
or
tropics.
You
can
also
see
that
the
lachen
gradient
is
only
apparent
in
the
northern
hemisphere
and
doesn't
appear
in
the
southern
hemisphere
indicating
that
there's
different
physical
processes
at
work
in
the
Arctic
versus
the
Antarctic,
so
Arctic
amplification
is
thought
to
be
caused
by
feedbacks
and
changes
in
heat
transport
on
the
diagram
on
the
left.
It
just
shows
various
feedbacks
and
transports
as
an
implication
and
either
contributing
to
or
doing
or
amplification.
Q
In
this
study
we
specifically
look
at
surface
albedo
cloud
feedback,
water,
vapor
and
temperature
feedbacks.
As
well
as
heat
transport
by
the
ocean
in
the
atmosphere
and
while
they're
allowed
to
be
as
to
what
feedback
is
or
transport,
it
contributes,
the
most
to
Arctic,
amplification
I
think
it's
important
to
note
that
these
physical
processes
are
operating
on
very
different
timescales.
Q
You
know
how
fast
does
Arctic
amplification
occur
following
four
times
co2
and
there's
a
relative
importance
of
the
Arctic
amplification
mechanism
vary
with
time.
You
know,
is
it
possible
that
there
that
was
causes
Arctic
amplification
in
the
initial
response
is
different
than
you
know,
once
we
once
we're
in
the
quasi
equilibrium
situation.
So
to
do
this,
we
use
simulations
from
twenty
one
scene
with
five
models.
We
use
the
fully
coupled
pre-industrial
control
and
the
oppressed
for
time.
Co2
runs
so
in
those
runs,
co2
is
instantaneously.
Quadruple
and
they're
spun
off
from
the
grains
austrial
runs.
Q
We
quantify
the
feedbacks
using
the
radial
kernel
method.
A
radiative
kernel
is
the
sensitivity
of
the
top
of
atmosphere.
Radiation
to
a
change
in
some
atmospheric
feel
like
see
I
like
surface
albedo
or
specific
humidity,
and
the
kernels
that
we
use
in
this
study
are
from
CSM
cam
five.
We
defined
the
Arctic
as
being
from
7b
north
to
90
north.
This
choice
was
convenient
because
this
has
about
the
same
fractional
area
of
land
as
the
global
average.
Q
We
also
calculated
the
atmospheric
and
ocean
mix
Lurky
convergence
into
the
artist
from
changes
in
calm
energy
content,
as
well
as
radiative
flux.
Is
that
a
better
output
from
the
model-
and
you
know
just
to
know
we're
specifically
looking
at
the
mixer
heat
convergence
here,
because
ultimately
Arctic
amplification
is
surface
warming
and
make
of
the
next
layer
will
be
it's
the
most
influential
part
on
on
surface
warming.
Q
So
we
found
that
Arctic
amplification
appears
very
quickly
following
four
times
see
you
to
you
know
within
within
askew
bond.
So
in
these
plots,
I
just
show
the
change
in
Arctic
temperature
and
change
in
global
average
temperature.
The
solid
line
is
the
multi-model
mean
of
the
twenty
one
model,
and
the
shading
indicates
one
standard
deviation
you
can
see
by
the
third
month.
The
Arctic
temperature
change
has
already
begun
to
separate
from
the
global
average
temperature
change.
Q
You
can
also
see
the
seasonality
of
arts
cancelled
vacation,
take
shape
within
the
first
year
with
a
maximum
in
amplification
in
the
cold
season
and
minimum
in
the
warm
season
on
the
right.
Its
annual
average
change
in
temperature
for
the
150
years
of
the
above
four
times
due
to
run
you
can
see
for
the
first
decade
or
so.
Temperature
rises
quite
rapidly
for
both
the
Arctic
and
global
average
and
for
the
remaining
140
years
temperature
rises
at
a
slower
rate
as
components
like
the
ice
in
the
ocean
adjust.
Q
We
can
actually
quantify
the
amount
of
Arctic
amplification
using
the
polar
amplification
factor
or
the
pif,
and
that's
just
the
fines
as
the
ratio
that
is
changing
polar
temperature
to
the
change
in
global
average
temperature,
and
you
know
on
this
slide
just
shows
the
annual
T
is
450
years,
and
you
can
see
that
in
the
first
year
we
already
have
a
pH
of
about
2.5,
indicating
that
the
Arctic
has
warmed.
You
know
two
and
a
half
times
more
than
the
global
average.
Q
Q
We
also
looked
at
hours
over
the
first
year
just
so
we
capture
an
entire
annual
cycle
as
well
as
the
last
30
years,
which
is
our
you
know,
quasi-equilibrium
response
and
that's
comparable
to
what
some
other
studies
have
looked
at
in
our
amplification.
So
you
can
see
what
was
in
the
first
three
months
that
the
latitudinal
gradient
and
the
northern
hemisphere
here
starts
to
form
and
it's
very
much
constrained
to
the
surface.
Q
You
know
that,
because
this
is
during
polar
night,
the
arts
economist
here
is
very
stable,
so
that
really
traps
the
warming
near
the
surface.
When
we
look
at
the
first
year,
we
can
see
the
center
of
warming
move
to
North
to
be
right.
Over
the
pole,
it's
a
little
bit
less
well
constrained
to
the
surface,
but
the
latitudinal
gradient
temperature
strengthens,
and
by
the
time
we
look
at
the
last
30
years.
Q
So
when
you
just
look
at
the
radiative
forcing
from
quadrupling
see
you
too,
that
actually
opposes
Arctic
amplification.
So
this
plot
just
shows
that
the
annual
mean
radium
forcing
from
beforetime
co2,
because
the
co2
is
going
to
absorb
and
emit
more
long
wave
radiation
when
over
places
that
are
hotter
because
they
have
more
outgoing
long-wave
radiation.
Q
You
know
the
go
through
the
x-axis
P
Prime
indicates
the
local
deviation
of
the
planks
feedback,
which
is
basically
a
measure
of
how
efficiently
an
area
is
cooling
itself
compared
to
a
global
average.
That's
followed
by
the
lapse
rate,
feedback,
surface
albedo
feedback,
water,
vapor
feedback,
the
total
cloud
feedback
and
the
atmospheric
and
oceanic
heat
transport.
You
can
see
in
the
first
three
months.
The
only
significant
contribution
to
Arctic
amplification
is
from
the
lapse
rate
feedback.
Q
Any
plots
the
thoughts
indicates,
a
multi-model
mean,
and
the
error
bars
are
a
95%
confidence
interval
from
the
twenty
one
models,
and
you
can
see
that
you
know
the
lapse
rate
feedback
is
been
training
about
half
a
degree,
more
warming
to
the
Arctic
than
the
global
average.
You
can
also
see
that
there's
a
small
opposing
contribution
from
the
cloud
feedback,
which
is
selectively
worn
in
the
global
average
over
the
Arctic.
Q
Now
things
look
a
little
bit
different
when
we
consider
the
first
year
at
this
point,
the
Planetree
back
and
albedo
feedbacks
become
important
and
in
fact
the
albedo
feedback
becomes
the
dominant
feedback
producing
this
Arctic
amplification
that
we
see
in
the
first
year.
At
this
point,
the
cloud
feedback
remains
in
opposition
to
Arctic
amplification
and
there's
also
a
contribution
to
our
camp
location
from
the
ocean
heat
transport.
Q
In
the
last
30
years,
the
same
three
feedbacks
are
contributing
to
Arctic
amplification
and
it's
mostly
the
albedo,
followed
by
lapse
rate
and
then
followed
by
Planck.
We
see
that's
a
water,
vapor
and
cloud
feedbacks
are
opposing
Arctic
amplification.
The
atmosphere
key
transport
actually
is
opposing.
You
know
it's
by
changing
the
meridian
o
gradient
in
temperature
that
can
produce
a
decrease
in
the
atmosphere,
key
convergence
into
the
Arctic,
and
you
know
we
find
at
this
point
that
there
is
not
a
significant
contribution
from
the
mix.
Lurky
convergence
in
the
Arctic.
Q
So,
in
conclusion,
you
know
we
found
the
arts
complication
occurs
rapidly,
like
within
a
few
months
when
we
quadruple
co2
and
because
of
this
we
did
it's
likely
to
be
driven
largely
by
atmospheric
processes,
and
you
know,
specifically,
we
found
that
this
response
in
the
first
three
months
is
driven
largely
by
the
lapse
rate
feedback.
You
know
this
is
polar
night.
Albedo
there
isn't
incoming
sunlight.
The
albedo
feedback
isn't
important
at
this
point.
F
Q
Yeah
definitely
you
know,
one
of
the
takeaways
from
the
study
is
that
you
know
the
last
three
is
probably
significant
in
the
first
three
months
because
we're
you
know
quadrupling
co2
during
polar
night,
and
so
you
know
we're
actually
thinking
you
know
how
that
might
be
different.
If
we
quadruple
co2
in
the
other
month,
you
know
I
think
it's
inherently
a
fast
response,
but
I'm
not
sure
if
the
feedbacks
would
stay
the
same.
F
Q
F
F
But
there's
other
questions
in
the
chat,
so
I
encourage
you
to
look
at
those
and
we
will
move
on
to
the
last
of
our
15-minute
talks.
Just
so
Jillian
gal,
who
is
gonna,
share
some
work
from
the
I
house
simulations
these
high-resolution
simulations
on
Antarctic
conditions.
So
if
you
can
share
your
screen
and
unmute
and
we
can
see
your
screen.
S
So
the
model
we
used,
it's
the
CSM
1.3,
so
the
land
and
the
atmosphere
resolution
or
in
the
25
degree
the
ocean
and
sea
ice
part
is
the
point
one
degree
with
three
in
dark
field
and
greenhouse
can
exposing.
This
is
the
high
resolution
fully
couple
and
planet
who
run
about
500
years,
so
this
preliminary
result
is
based
on
the
first
three
hundred
and
twenty
years
data.
S
So
some
background
information,
so
in
the
widow
in
the
widow
chair
in
the
windowsill
between
the
most
dominant
circulation,
in
the
way,
no
joy,
so
there
there
is
a
high
ideal
mixing
at
the
Eastern
Eastern
are
no
of
the
dryer
when
the
flow
go
back
to
west
world.
This
is
relatively
one
flow
you
can
see.
It
can
has
both
no.
L
S
South
of
the
motorized
Sigma-
and
here
we
focus
on
to
a
new
linear,
the
motorized
Polonia
will
form
exactly
about
the
moderate
Sima.
This
is
the
satellite
observation
of
the
motorized
colonias
and
sometimes
the
motorized
linear
scan
expanded
westward
and
on
this
huge
weather
solonius.
This
is
the
satellites
observation
of
the
weather
chip
brownie,
you
know
yeah
simulation,
sometimes
the
motorized
Polonia
can
be
similar
interval.
It's
exactly
about
the
motorized
Seema.
S
S
From
the
observation,
so
there
is
a
high
temperature
pen
to
run
in
the
motorized
seamount,
so
this
is
also
true
yeah
model
for
both
cases,
and
this
is
due
to
their
telecom
effect.
Still,
the
one
water
is
westward
flow,
so
the
teller
can
about
the
motorized
amount,
preventing
the
flow
from
going
over
it,
so
it
must
go.
S
S
S
M
S
S
So
these
are-
and
this
figure
shows
the
hemispheres
of
the
sea-ice-
come
nutrition
and
mix
mix
layer
tab.
First,
we
consider
C
as
a
nutrition
match
the
very
well
with
mix
the
layer
tab.
It
didn't
mean
the
main
ponys
occurred.
The
tip
connection
even
though
follows,
and
you
can
see
there
is
clearly
a
West
westward
drug
pattern
or
the
pro
nears.
It
first
occurred
at
the
street
week
region
and
westward
to
the
motor
right,
SEMA
and
aplomb.
A
motorized
clonal
occur
and
it
became
stronger
and
we
always
were
dragged
to
the
central
mweadows
imaging
time.
S
The
the
weather
see
Polonius,
who
will
occur
and
another
feature
is
the
colonial
event.
Is
a
period
airy
occur
through
time.
The
time
series-
and
this
is
the
only
did
how
we
have
now
for
the
500
years
due
to
the
Kobe
1930
attachments.
We
have
some
problem,
but
from
this
data
we
can
see
after
300
years
the
Loney
event
still
occur
period.
So
this
is
a
very
good
data
for
us
to
investigate
the
reasons
for
the
intermittent
occurrence
of
the
flow
near
you.
S
And
this
is
the
sea-ice
were
tasty,
so
from
observation:
there's
a
hydrating
pang
along
the
north
and
western
bank
of
the
motor
right
Sigma.
So
you
know
model.
This
feature
is
also
very
clearly
represent.
The
rear,
wheel
and
another
feature
is,
there
seems
like
Eddy
transport
pathway
through
the
high
dev
activity
at
the
eastern
end
of
the
weather,
dryer
to
the
Polonia
event
location.
So
this
may
be
Eddy
transport
pathway
in
in
this
region.
So
we
need
to
continue
to
explore
here.
S
This
is
this:
is
the
tuna
velocity
along
the
motorized
section,
so
about
is
the
left?
One
is
a
folder
summer
season
and
right
away,
I
need
for
the
winter
season,
so
we
can
say
I
do
not
plan
of
the
motorized
there's
a
strong
westward
flow
and
about
it
here
there
is
a
counterpart.
Easter
will
flow
about
the
motorized
Zimmy
and
it
is
stronger
in
the
summer
and
a
little
weaker
as
winter.
S
This
figure
shows
the
relationship
between
the
ocean,
open
ocean
Polonius
and
the
wind
strength
when
related
variables.
So
the
color
band
is
the
Polonia
event
years
and
we
can
say
when
Vanya
has
occurred,
the
Sam
index
would
be
him.
The
value
would
be
him
or
bigger
when
the
time
when
Polonius
occur
another,
why
bubbles
of
entry
switch,
the
wench
Resco
and
when
stress
curve
will
became
more
negative
when
the
Lonnie
event
occurs,
notice
that
the
value
is
the
worst
here.
So
this
is
the
firm
of
division
and
in
our
simulation,
this
is
also
true.
So.
S
S
And
finally,
we
will
compare
the
Antarctic
sloop
current
or
three
cases.
The
left
one
is
the
fresh
air.
In
this
case,
the
entire
loop
currently
is
a
very
strong,
so
the
enthalpy
line
will
Bend
downward,
so
the
one
motor
cannot
go
on
to
the
shelf.
So
in
this
model,
I'm
repressing
these
Keys
very
well,
and
the
second
case
is
the
tanks
trail
testers
intention
of
the
most
dominant
feature
in
the
way
structure.
S
The
structure
here
and
the
one
water
flow
can
go
on
to
transport
to
onto
the
shell
due
to
some
Eddy
activity
here
in
this
model,
and
this
process
can
also
be
represented
very
well.
The
restructure,
I
saw
villain
and
some
one
motor
will
be
transferred
onto
the
shelf,
and
the
third
case
is
the
one
shelf
in
the
one
shelf
the
one
water
can
go
to
from
co2
onto
the
shelf
region,
so
young
model.
This
is
also
the
case.
The
one
motor
transport
onto
the
shelf.
S
S
S
F
You
that
was
really
interesting
and-
and
we
have
several
questions
in
the
chat
but
I-
think
we're
running
a
little
bit
late
so
did
in
order
to
keep
on
time
and
I'd,
encourage
you
to
answer
the
questions
within
the
chat
because
I
see
at
least
four
of
them
may
be
fine
with
them
and
there's
definitely
a
lot
of
interest
in
this
work.
It's
great
to
see
these
results
coming
from
the
I
have
sent
simulations.
B
Okay,
so
I'm
sharing
my
screen
I'll
just
say
for
the
lightning
talks.
What
I'm
gonna
do
is
I'm
just
gonna
call
on
you
by
your
given
name
and
then
and
there's
a
one
pre-recorded
one
here,
I
think,
and
then
we
will
be
giving
you
a
the
4-minute
mark
heads-up
that
you've
got
a
minute
left
to
wrap
up
okay,
so
first
up
is
Olivia.
P
Will
do
so
hello.
My
name
is
Olivia
Truex
and
I'm,
a
PhD
student
at
the
University
of
Otago
here
to
talk
briefly
about
an
ongoing
project
comparing
proxy
records
of
Antarctic
climate
variability
with
simulations
from
the
CSM
last
millennium,
ensemble
I'd
like
to
start
by
thanking
Betty,
Otto,
Blaser
and
Esther
Brady
for
working
with
me
on
this
project
and
virtually
hosting
me
at
anchor
these
past
few
months.
P
So
my
research
aims
to
identify
the
drivers
of
last
millennium
cooling
in
the
southern
high
latitudes
by
quantitative,
qualitatively,
comparing
full
and
single,
forcing
simulations
from
the
last
Moni
ensemble
to
proxy
reconstructions
from
Antarctica
to
the
left.
I've
shown
a
figure
from
Otto,
please
not
at
all
2016
outlining
the
initial
states
and
simulation
links
of
the
LME
control
and
four
strands
and
to
the
right
I've
included
a
recent
compilation
of
temperature
reconstructions
from
Antarctica
ice
cores
by
steady
at
all
2017.
P
These
reconstructions
show
between
0.2
and
point
5
degrees
of
cooling
across
continental
Antarctica
over
the
last
millennium.
Now
previous
research
has
attributed
this
cooling
to
increased
volcanism,
solar
minima
and
or
multi
centennial
trends
and
modes
of
atmospheric
variability.
But
as
of
yet
there's
not
a
widely
agreed-upon
driver
for
last
millennium
cooling
in
the
Antarctic
within
the
Paleo
climate
community,
so
the
most
basic
question
is:
does
Antarctica
get
colder
in
the
LME
next
slide?
Please
and
broadly,
the
answer
is
yes.
P
I've
caught
to
the
left,
I've
plotted
the
full
forcing
ensemble
mean
in
50-year
average
anomalies
from
the
thousand-year
mean,
and
we
can
see
a
warm
period
from
850
seee
to
1250
seee,
which
is
followed
by
about
half
a
degree
of
cooling
and
coldest
temperatures
after
1600
on
the
right,
I've
plotted
annual
average
temperature
between
60
and
90
degrees,
south
the
full,
forcing
volcanic,
only
soap
and
solar.
Only
simulations
are
in
gray,
with
the
ensemble
mean
in
black.
P
The
timing
of
the
10
largest
volcanic
eruptions
is
denoted
by
the
black
bars,
and
so
in
the
LME
simulations.
They
indicate
that
large
volcanic
eruptions
may
cool
the
southern
high
latitudes,
strongly
enough
to
force
the
Antarctic
into
a
colder
climate
that
persists
after
direct
radiative,
forcing
from
volcanic
aerosols
dissipates.
However
other
for
things
such
as
solar
minima
that
are
necessary
to
produce
the
coldest
temperature
is
seen
after
1600
next
slide.
P
Please,
and
as
we
would
expect
cooling
his
greatest
over
areas
of
the
southern
ocean
with
largest
increases
in
sea
ice
area,
indicating
the
ice
ocean
feedbacks
may
have
played
an
important
role
in
sustaining
a
colder
climate
in
the
Antarctic
in
the
decades
after
major
volcanic
eruptions.
Next
slide,
please.
P
P
Inferred
from
a
compilation
of
proxy
work,
instructions
shown
on
the
right
in
a
figure
from
leading
at
all
2019
ice
cores
indicate
that
the
eastern
drazi
booked
the
last
alone
IAM
cooling
trend
to
warm
over
the
last
millennium.
Next
slide.
We
want
to
suggest
that
the
homogeneous
cooling
in
the
LME
may
be
due
to
the
simulated
dynamics
of
the
southern
annular
mode.
P
20Th
century
analysis
indicates
that
a
positive
sam
cools,
the
eastern
LC,
well
a
negative
sam,
warms
the
eastern
rossi,
a
proxy
based
Simek
in
structured
by
detweiler
at
all
2018
shown
in
the
bottom
left,
shows
a
multi
centennial
trend
to
a
more
negative
sam
between
1350
and
1800
c
e,
which
could
have
acted
to
warm
eastern
LC
against
the
backdrop
of
a
continent-wide
cooling
trend.
By
contrast,
in
the
last
millennium
ensemble,
the
Sam
Index
remains
semi,
constant
until
a
positive
trend
during
the
industrial
period.
This
phenomenon
has
also
been
noted
in
other
studies.
P
Global
climate
models
typically
do
not
simulate
the
Centennial
trends
in
atmosphere,
dynamics
inferred
from
proxy
records.
So,
to
summarize,
the
LMA
simulates
last
millennium
cooling
in
the
Antarctic,
with
all
forcings
and
only
volcanic
forcing,
but
with
the
reduced
magnitude
for
only
volcanic.
However,
the
LME
does
not
simulate
the
sam,
like
regional
heterogeneity,
observed
in
proxy
records
and
I'm
in
the
beginning
stages
of
this
work
and
my
PhD
so
I'd
welcome
any
feedback
or
questions
you
have
to
offer
via
the
chat
and
I'll
include
my
email
in
the
chat.
Thank
you.
A
B
O
O
The
work
is
motivated
by
observations
of
large
changes
in
the
Arctic
freshwater
storage.
Both
the
Arctic
liquid
and
solid,
fresh
for
storage,
the
liquid
fresh
restore,
has
been
increasing
since
the
mid
1990s,
which
is
shown
in
the
figure
here
where
the
green
stars,
and
this
increase
is
relatively
large,
makes
about
an
increase
of
10%
in
the
link
refresh
or
storage
or
10,000
cubic
kilometers.
O
The
solid
freshwater
storage
has
decreased
over
the
same
time
as
the
sea
s
volume
has
decreased,
but
this
increase
in
the
liquid
fresh
water
storage
is
more
than
just
the
melting
sea-ice,
as
the
increase
in
the
liquid
storage
is
larger
than
the
decrease
in
the
solid
storage.
The
oceanic
gateway
flexes
have
not
shown
any
trends
over
the
period
that
they've
been
monitored,
which,
since
the
late
1990s
for
some
or
in
the
2000s
for
others.
O
So
the
question
that's
been
in
the
community
is
when
we
can
actually
expect
to
see
a
force
signal
emerge
in
the
different
parts
of
our
for
a
budget
in
particular,
because
we
know
that
there's
large
multiplicative
variability
in
terms
of
the
Arctic
freshwater
budget
so
to
disentangle
the
force
response
from
this
multi
decade
of
variability,
we're
using
the
CSM
large
ensemble,
which
is
forced
by
the
RCP
8.5
into
the
21st
century,
and
the
CSM
low
warming
ensemble,
which
uses
forcing
that
limits.
Warming
to
two
degrees
by
2100.
O
To
answer
this
question:
we're
going
to
focus
on
two
different
terms
to
detect
a
forced
response
in
the
system.
The
first
one
is
shift,
which
signifies
will
be
the
first
time
that
a
freshwater
term
is
outside
the
pre-industrial
background
variability
as
sampled
in
the
1800
years.
Own
control
simulation
from
the
CSM
water
comes
something
stronger
criteria
which
really
signifies
the
emergence
into
a
completely
different
regime
and
is
called
emergence,
which
is
when
the
simulated,
fresh,
Florida
term
stays
outside
the
background.
Very
quality
you
till
2100,
so
the
fluxes
or
variability
in
the
storage
is
cause.
O
No
overlap
was
the
pre-industrial
variability.
So,
using
these
two
definitions,
we
can
compile
shift
and
emergence
timing
in
different
ensemble
members
for
all
the
different
terms
of
the
Arctic
freshmen
budget
was
just
shown
in
the
summary
figure
here.
The
thin
bars
signified
the
shift.
The
thicker
bars
the
emergence,
which
is
what
we're
going
to
focus
on
right
now
and
then
the
two
colors
show
the
different
scenarios
to
see
if
there's
a
scenario
impact.
O
If
we
focus
on
the
earliest
complete
emergence
and
all
ensemble
members,
we
find
that
that
actually
occurs
in
the
solid
and
liquid
fresh,
more
storage
terms,
and
so
that
indicates
that
the
observed
increase
in
the
liquid
fresh
water
storage
actually
may
already
be
a
force
changes
in
the
CSM.
It's
indicated
that
that
emergence
is
complete
before
the
year
2020
and
we
already
know
that.
There's
a
force
component
for
a
solid
French
were
storage,
but
for
the
liquid
fresh
first
story
so
far
that
has
been
unclear.
O
The
earliest
flush
reflects
terms
that
shall
complete
emergence
and
all
ensemble
members
is
the
liquid
there's
straight
fresh
water
export,
followed
by
the
liquid
Davis
straight
fresh
for
an
export
about
a
decade
later,
and
then
those
two
have
some
overlaps
as
to
see
there's
straights
downstream
off
their
Street,
and
that's
really
interesting
because
they
restraint
has
been
monitored.
There's
been
a
gap
of
a
couple
years
note,
but
the
monitoring
is
for
starting
this
year.
So
if
this
is
current,
you
may
actually
be
able
to
observe
this
in
the
future.
O
In
terms
of
scenario
impact,
we
see
that
emergence
is
only
affected
for
two
terms:
the
very
runoff
and
the
solid
from
straight
fresh
for
export.
It
shows
emergence
under
the
high
warming
scenario,
but
not
over
water.
So
this
is
the
accelerate
key
point
that
I
just
showed
you
more
hopefully
soon
in
the
paper.
Thank
you
very
much.
B
P
So
why
do
we
care
about
the
melt
on
set
and
melt
on
set
and
other
seasonal
transition?
Dates
can
be
used
to
evaluate
how
well
kind
of
models
represent
sea
ice.
So
what
I'm,
showing
here
is
the
average
melt
on
set
day
over
the
satellite
period
for
satellite
data,
the
CES
end
mark
ensemble
and
some
seeming
six
models,
and
we
see
that
the
model
is
seeing
six
models
included,
show
a
realistic
but
varying
melt
on
set
dates
and
that
these
model
differences
in
not
onset
are
unlikely
due
to
internal
variability
along.
P
This
paper
is
under
review
in
the
cryosphere
discussions.
If
you
want
more
information
on
how
we
come
to
those
conclusions
and
we
can
use
these
differences
to
investigate
other
modeled
sea
ice
characteristics
such
as
the
ice
area
and
the
picnics.
So
it's
a
very
useful
metric.
However,
comparisons
between
climate
models
and
satellite
observations
are
indirect
and
I'm.
Gonna
talk
a
little
bit
about
how
next
slide
please
so
right
now,
Mountain
step
products
from
satellite
observations
are
taken
from
observed.
Brightness
temperatures
and
brightness
temperatures
aren't
currently
simulated
in
clamp
models,
including
in
the
CES
m2.
P
So
if
we
want
to
get
melt
onset
from
a
climate
model,
we
have
to
use
standard
model
output
and
to
find
it
differently
so,
for
example,
with
surface
temperature.
Next,
please
so
we're
actually
doing
here
is
using
a
satellite
simulator
to
produce
simulated
brightness
temperatures,
and
this
opens
up
a
whole
new
set
of
comparisons
between
the
model
and
observations.
The
first
comparison
would
be
between
these
new
simulated
brightness
temperatures
and
the
observed
brightness
temperatures,
and
then
we
can
also
use
the
simulated
brightness
temperatures
to
create
new
simulated
melt
on
set
dates.
P
That
should,
in
theory,
be
more
comparable
to
the
observed
melt
on
set
products,
and
so
that's
another
new
comparison.
And
then
we
can
also
compare
the
simulated
sea
ice
melt
on
set
dates
from
the
brightness
temperatures
to
those
melt
on
set
dates
that
we
get
from
the
standard
model
output
and
next,
please,
today,
I'll
just
show
some
early
results
of
the
these
two
comparisons
circled.
P
Next,
please
so
on
the
Left
I'm,
showing
the
average
brightness
temperatures
for
June
2003
in
the
CES
m2
and
in
satellite
data,
and
we
see,
from
this
kind
of
first
cut,
that
the
simulator
can
produce
realistic,
brightness
temperatures
for
CES
m2.
So
this
is
good
and
on
the
right
I'm,
showing
the
CES
M
to
melt
on
set
dates
for
2003,
using
a
kind
of
remedial
definition
to
melt
on
set
from
the
brightness
temperatures
and
as
well
as
the
adult
onset
definition
from
surface
temperature.
P
P
A
P
So
I
am
going
to
present
actually
on
behalf
of
a
larger
effort
going
on
in
the
polar
climate
working
group
to
look
at
the
influence
of
CSLB
dough
on
transient
climate
change,
and
here
we're
trying
to
argue
potentially
that
there's
a
case
for
tuning
the
sea
ice
albedo.
So
there's
a
number
of
folks
but
I
put
together
the
slides.
So
all
credit
goes
to
them
and
all
errors
go
to
me
next
slide.
P
P
Here's
the
CES
on
to
these
are
new
results
from
the
sea
dsm2
large
ensemble,
but
this
is
CES
m2,
cam,
six
and
basically,
what
we
see
is
that
the
observed
sea
ice
is
has
greater
extent
than
CES
on
to
over
the
historical
period
and
that
there's
actually
greater
loss
in
the
CES
on
to
than
there
is
in
the
observed
record-
and
we
know
this
from
special
issue
papers
that
have
already
been
put
out
about
CES
on
to
camp
six
in
particular
Patricius.
But
here
we
really
start
to
see
the
value
of
the
large
ensemble.
P
F
P
So
if
you
go
to
the
next
slide,
what
what
really
we're
doing
here
is
a
street
tune,
so
a
straight
tune
is
turning
the
car
in
its
actual
environment
and
during
actual
driving
conditions.
So
the
analogy
here
would
be
that
the
CES
m2
has
already
been
released.
People
are
using
it
they're
driving
it
around,
and
we
just
want
to
come
in
and
see
if
we
can
do
a
street
tune
to
make.
The
C
is
a
little
bit
better
and
the
Arctic,
so
Dave
Bailey
I
was
really
the
architect
of
these
changes
and
David.
P
If
you
could
just
advance
one
more
forward,
these
experiments
are
really
quite
simple.
They
have
two
changes
to
the
sea
ice,
one
to
the
snow,
green
radius
and
the
other
to
the
melt,
onset
temperature,
but
no
other
tuning
was
done
on
these
runs.
So
this
is
basically
a
straight
tune
of
CES
m2,
with
a
couple
of
small
changes
to
the
sea
ice
albedo.
P
So
next
slide.
So
here's
kind
of
the
punchline
of
what
this
street
tune
does.
So
we
have
again
the
CES
m2
large
ensemble
in
gray
the
observations
in
red,
and
now
we
have
three
new
lines
on
this
plot
and
then
they're
in
the
light
blue
and
then
in
the
dark.
Blue
they're
ensemble
mean,
and
so
what
we
can
see
is
that
this
new
tuned
ice,
small
ensemble,
just
three-
has
much
more
realistic,
areal
extent
of
the
September
sea
ice
and
the
sea
ice
loss
is
now
within
the
observed
sea
ice
loss.
P
Go
to
the
next
slide.
Dave
I
just
want
to
highlight
that
one
of
the
things
that
we
think
is
really
playing
a
role
here
is
the
sea
ice
thickness,
and
so
this
is
just
a
nice
figure
from
Patricia
just
saying
that
the
tuned
runs
have
thicker
sea
ice
and
but
the
mean
state
is
only
one
component
of
Arctic
sea
ice
loss.
An
Arctic
warming,
so
I
think
there's
a
lot
more
to
look
into
here.
So
last
slide
the
last
one.
P
It's
greater
than
I'm
circuits
and
I
also
encourage
you
to
look
at
Patricia's
paper.
We
know
C
s
means
state
is
one
of
many
factors
that
can
affect
sea
ice
lost
race,
so
we
did
some
Street
tuning
and
so
far
we're
seeing
improved
simulations
in
the
Arctic.
There
are
impacts
outside
of
the
Arctic
in
particular.
The
Antarctic
sea
ice
also
looks
quite
different,
but
I
also
want
to
highlight.
The
initial
results
really
here
show
a
modest
non
polar
impacts.
P
B
T
T
So
those
of
you
who
unfamiliar
ods
is
halogen
gases
containing
chlorine
and
or
bromine
with
the
potential
to
break
down.
I
was
only
in
the
stratosphere
and
we
normally
think
of
them
used
for
refrigerants
air
conditioners
stuff,
like
that
and
over
the
second
half
of
the
20th
century,
their
emissions
rapidly
increased
until
the
introduction
of
the
Montreal
Protocol
in
1987
and
although
they're
much
less
abundant
than
co2
on
a
molecule
by
molecule
basis,
they're,
really
a
potent
greenhouse
gas
and
so
on.
T
The
right
here
is
a
figure
of
the
radiative
forcing
and
the
red
is
OD
acid.
So
you
can
see
that
over
this
period
they
are
the
second
most
important
greenhouse
gas
after
co2.
Next
slide,
please
so
to
explore
the
the
impacts
of
ozone
depleting
substances
on
the
Arctic.
We
performed
a
set
of
simulations
10
ensemble
members,
which
are
companion,
runs
to
the
CSM
large
ensemble,
where
we
fixed
the
esses
and
ocean
concentrations
at
1955
values.
T
So
the
main
result
here
is
that
on
the
left
is
the
forced
historical
warming
over
this
time
period
in
the
10
members
of
the
CSM,
large
ensemble
and
there's
about
a
1.6
degree
warming,
and
then,
when
we
keep
fixed
weode
ozone
depleting
substances
and
ozone,
this
forced
warming
reduces
by
about
50
percent.
It
goes
down
to
about
0.8
in
the
Arctic,
so
these
are
really
large
effects.
We're
seeing
here.
So
one
question
you
can
have
is
is
how
do
we
know?
T
This
is
due
to
the
radiative
effects
of
OD
s's
and
not
the
effects
on
ozone.
So
next
slide,
please.
So
we
performed
another
10.
Member
ensemble,
where
we
only
fixed
the
ozone
depleting
substances
and
we
allowed
for
ozone
to
to
evolve
as
it
does
in
the
CSM
large
ensemble,
and
we
got
statistically
kind
of
the
same
results
as
in
panel
B,
where
it's
about
reduced
warming
of
about
50%
compared
to
the
historical
roots.
T
Next
slide,
please!
So,
then,
along
with
this
reduced
warming
due
to
ODS,
is
being
held
fixed.
We
also
see
the
sea
ice
loss
being
reduced
by
about
a
half
as
well.
So
this
is
September
sea
ice
loss
between
1955
and
2005,
and
you
can
see
the
reduction
is
about
50%
as
well.
So
these
are
really
large
changes
due
to
ozone
depleting
substances.
T
So
in
these
runs
in
which
we
held
ODS
is
fixed
at
1955
values,
the
forced
Arctic
warming
and
sea
ice
loss
were
reduced
by
about
a
half,
and
we
showed
that
this
was
due
to
the
radiative
effects
of
those
own
depleting
substances,
not
through
their
effects
on
ozone,
and
we
all
know
that
internal
variability
can
be
really
large
in
the
Arctic.
But
this
signal
is
is
much
larger,
and
so
actually
we
could
have
done
it
with
probably
much
less
than
than
10
members.
T
What
I
haven't
shown
you
here
is:
we
did
this
with
two
different
climate
models,
so
we
did
also
with
CSM
welcome
and
you
can
see
it
in
the
paper
for
the
further
results.
Fair
and
one
nuance,
which
I
really
haven't
mentioned
by
think
it's
important
to
take
away.
One
minute
is
ozone.
Depleting
substances
can
have
an
outsized
effect
due
to
a
large
cancellation
over
this
period
between
cooling
from
aerosols
and
warming
from
co2
and
there's
the
reference
for
that
paper.
Thank
you
very
much.
No.
B
M
L
L
L
My
University
in
Beijing
today,
I'm
going
to
talk
about
average
signing
I,
knew
with
hurricane
for
alternates
as
models
and
the
impending
CSM.
So
the
new
growth
market
consists
of
three
popular
grids
that
are
the
open,
45
0.15
and
a
little
on
205
or
122
degree.
So
the
target
at
the
clam
is
nation
all
the
way
to
submit
a
scale
simulations
in
the
Arctic,
which
is
the
focus
our
work
here.
The
model
grades
are
22,
kilometers,
7.3,
kilometers
and
2.4
kilometers.
L
So
the
great
system
is
openly
available
for
the
community
to
use
and
also
we
have.
They
are
going
to
San
shaken
model
development
paper
which
are
I
saw
on
the
bottom
right
here
photo
size
kinematics.
We
have
picked
a
two
days
of
different
Arctic
Oscillation
index
on
20th
December
20th,
which
high
so
on
the
right.
L
The
deformation
rates
fields
is
correspond
to
a
negative
index
and
we
can
show
that
we
can
see
that
the
information
that
fields
are
much
finer
and
much
richer
in
structure
for
the
fibers
in
advance
and
which
are
ugly
but
I
go
to
be
better
as
well,
because
we
have
three
resolutions
and
three
consistent
climatology.
We
are
able
to
study
the
scaling
from
tomorrow
unless
we
study
the
cumulative
distribution
of
the
of
the
total
defamation
which
so
that's
the
by
using
the
high
resolution
ones
as
a
reference.
L
We
are
able
to
say,
evaluates
a
low
resolution
price
for
their
impact
resolution
and
we
are,
we
conclude
that's
the
year.
The
effective
tomorrow
is
about
six
to
seven
times,
coarser
and
equitable
resolution
for
simulating
extreme
or
very
large
defamation
events,
and
also
comparing
the
two
days
we
showed
as
the
spatial
scaling
and
structural
function.
Although
they
are
both
they're
both
days.
They
are
multifractal
and
consistent
without
the
lesions.
L
They
are
drastic
differences
between
the
two,
so
they
are
the
oscillation
or
the
forcing
independent
scaling
as
a
is
witnessed
for
the
EVP
convergence,
because
we
have
a
traditional
EVP
implementation.
Our
model,
we
tend
to
see
fun
simulation,
we
tend
to
see
more
subtitling
of
the
EVP
are
needed
for
the
higher
resolution
runs
to
attend
a
commitment
or
a
Kennison
talk,
asymptotic
behavior.
L
This
applies
to
the
termination
PDF
on
the
right
and
also
as
well
as
at
the
definition
instantaneous
or
daily
definition
on
the
left
in
which
the
non
convergent
solutions
are
much
noisier,
and
this
doesn't
really
systematically
shifts
the
cam
tally,
but
first
four
regions
such
as
a
compatible
due
to
the
LAN
past.
The
dominance
I
promise
here
in
the
winter
will
show
that
the
non
convergent,
the
convergence
systematically
shifts
the
ice
thickness
or
ice
pollen
here.
L
The
non
convergent
solutions,
simulates
thicker
ice
because
of
excessive
noisy
kinematic
events,
as
well
as
a
more
dynamic
response
to
it.
So
inside
we
highlights
that's
the
Eevee.
Convergence
can
have
a
big
impact
on
the
and
well,
as
the
experiment
lead
to
a
length
of
ice,
so
the
sea
ice.
So
the
whole
system
is
openly
available
for
the
public
and
we
want
to
hear
from
you
the
concern
comments,
especially
for
the
for
the
ongoing
discussion
paper.
Ok,.
J
Q
A
You
hear
me
yes,
ok,
thank
you!
Thank
you!
David
and
agrimony
everyone
that
surgery
fee
from
canal,
so
today,
I
want
to
give
you
a
quick
overview
of
our
late
histories
and
study
about
10
o'clock,
show
relationship
between
winter,
air
stagnation
and
history,
extremes
in
China,
with
regional
Arctic
sea
ice
going.
This
work
was
founded
by
the
USDA
NSF
and
I.
Want
to
thank
all
my
week
is
the
cleverest
for
their
contributions
to
this
work.
So
next
slide,
please!
A
So,
as
we
know,
Easter
China,
Plains
area
or
the
ECP
area
is
one
of
the
hot
spell
regions
with
severe
air
pollution,
especially
in
winter,
and
this
heavy
air
provision
can
be
attributed
to
three
major
factors
versus
massive
anthropogenic
emissions.
The
rapid
chemical
formation
and
also
a
favorable
majority
of
conditions
here,
were
mainly
focused
on
a
third
part
by
investigating
the
possible
relationship
between
original
air
stagnation,
ways,
global
climate
change,
or
specifically,
the
Arctic
change,
to
do
that.
We
defined
a
so-called
pollution
potential
index
or
PPI
by
weighted
average
in
the
near
surface.
A
Wind
speed
and
air
temperature
gradient
to
describe
the
horizontal
and
vertical
airport
and
dispersion
capacity,
and
those
numbers
in
a
table
on
on
the
lower
right
corner
shows
a
good
collision
between
PPI
and
the
major
air
pollutant
air
pollution.
Observations
so
just
appear
as
a
good
proxy
to
describe
original
air
ventilation
and
regional
air
stagnation,
and
the
please
note
that
higher
values
in
PPI
suggest
weekend
air
stag
weekend,
air
ventilation
or
stronger
air
stagnation.
A
So
next
slide
please
so
several
recent
studies
have
suggested
a
potential
linkage
between
high
latitude
in
bomber
changed
with
Madonna
to
the
weather
extremes,
and
our
previous
study
have
also
has
also
reviewed
a
close
relationship
between
PPR
variations
with
Arctic
sea
ice
change.
So
to
understand
the
underlying
dynamic
processes.
Here
we
design
a
series
of
climate
sensitivity,
experiments
by
using
a
high
table
model
and
Ernest
ASM
framework.
A
An
election
pattern
resembling
an
active
phase
of
eurasia
pattern,
as
shown
in
those
contour
lines
in
sub
applause
C
due
to
a
transient
a
devotees
enforcing
and
which
strong,
significantly
suppressed,
managed
to
wish
a
winter
muscle
and
the
enhancing
original
air
stagnation
expand
extremes.
So
please
stay
more.
Did
here
the
dynamic
diagnosis
in
your
manuscript.
So
next
slide.
Please
and
besides
our
our
own
climate
sensitivity,
experiments.
A
This
is
more
evident
in
the
city
at
curves,
in
post,
p2
and
p3
time
periods
and
I'm
also
showing
you
those
model
specific
results
in
the
right
paper
here
and
though
those
numbers
very
kissable
would
consider
opening
and
most
no
models
are
great
on
a
positive
shift
in
the
PPI
positive
extremes.
So
next
slide,
please
just
stop
here
with
a
schematic
diagram
and
please
go
to
check
our
minutes.
Query
publishing
is
AP
for
more
technical
details
and
thank
you
for
attention.
B
Alright
well,
thank
you
for
such
a
wonderful
session.
We
have
gone
two
minutes
over
I
apologize
for
that.
That's
mostly
me
for
the
movie,
but
now
we're
scheduled
for
like
a
10-minute
break
and
then
we're
due
to
come
back
at
we're
due
to
come
back
at
12:20
for
the
individual
working
group,
lunch
sessions
and
so
I'm
just
trying
to
find.