►
Description
2020 CESM Workshop Day 2 - Tuesday, 16 June 2020 - 13:00-16:30 - Ocean Model 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 Ocean Model Working Group:
http://www.cesm.ucar.edu/working_groups/Ocean/
More information about the workshop:
http://www.cesm.ucar.edu/events/workshops/2020/
A
A
A
A
A
D
B
A
B
B
B
B
B
B
The
first
item
that
I
think
I
would
like
to
hear
your
thoughts
on
is
later
this
summer.
We
will
need
to
be
writing
our
CSL
proposal
for
computing
resources
for
the
following
two-year
period.
I
believe
the
CSL
allocation
period
runs
November
to
October,
so
this
would
be
resources
starting
1,
November
2021.
B
B
B
Based
on
our
experience
in
the
current
year,
our
working
group
does
not
have
a
whole
lot
of
production
work
going
on
we're,
not
running
coordinated
experiments
much
in
our
working
group,
because
most
of
our
focus
has
been
on
development
and
I
kind
of
anticipate.
That
will
be
the
case
at
least
through
much
of
the
next
year,
and
so
my
inclination
would
be
to
structure
our
request
with
a
heavy
bias
towards
the
development
side
of
the
allocation
which,
in
the
current
year,
is
like
twelve
point,
eight
million
hours
in
development
versus
three
point:
six
in
production.
F
Okay,
yeah
I
finally
found
them.
Ooh
I
think
something
that
that
we
need
to
get
some
kind
of
agreement
on
is
the
time
scale,
because
I
think
if
the
development,
when
is
CES
m3
what
is
mom,
six
etc
because
I
think
you're
right
as
long
as
those
aren't
in
the
next.
You
know
a
couple
of
years
out
at
least
then
the
next
year
will
be
mostly
development
for
CES
m3
in
the
ocean
model.
In
three
and
mom
six
I,
don't
know
what
that
timescale
now
is.
F
B
I
think
this
has
probably
been
discussed
in
the
co-chair
meeting
and
I'm,
not
remembering
precise
dates,
but
one
of
the
you
know
I
think
one
of
the
activities
that
will
be
going
on
in
the
next
say
at
least
six
months,
probably
a
little
more
is
the
kind
of
the
dye
course
do
doubt
for
the
atmosphere
under
the
SEMA
framework
and
they
were
hoping
to
be
able
to
do
fully
coupled
runs
with
Mom
six
for
at
least
part
of
that.
So
so
that's
sort
of
you
know
sort
of
pushing
the
development
cycle.
B
I
think
well
into
twenty
one.
I
think
it
would
be
optimistic
to
think
that
there
might
be
full
CES,
m3
prototypes
in
maybe
a
year
from
now.
But
of
course
you
know
our
experience
with
CES
m2
was
that
you
know
what
we
thought
was.
You
know
our
target
date
was
missed
by
almost
two
years
and
300
experiments
trying
to
get
it
to
work.
So
it's
really
difficult
to
say:
oh
yeah,
we're
gonna
have
C
sm-3,
and
you
know
summer
21
or
something
I
think
we
should.
B
You
know-
and
you
know
where
we
are
now-
you
mean
you've-
been
involved
in
these
discussions
bill.
You
know
we're
plotting
our
way
through
the
biases
and
you
know
well,
we
can't
can't
say
for
sure
when
we'll
have
them
all
mitigated
to
the
point
where
we're
happy
to
have
it
being
used
in
a
quasi
production
way.
So.
A
B
Yeah,
but
it
is
also
the
case-
I
mean
if
there
is
a
strong
interest
in
the
working
group
in
some
particular
working
group.
Wide
production
integrations
with
pop.
That's
still
a
completely,
you
know,
legitimate
request
that
we
can
put
in.
We
did
have
production
time
for
some
pop
experimentation
in
the
current
proposal,
and
there
has
been
very
low
enthusiasm
for
doing
the
analysis
of
those
runs.
B
It's
also
the
case,
I
think
you
know
a
lot
of
a
lot
of
the
work
that
you
know,
I
think
we
participate
in
is
production,
it's
kind
of
led
by
other
working
groups.
You
know
either
BGC
or
you
know,
I'm,
anticipating
the
our
system.
Prediction
working
group
will
be
wanting
to
a
lot
of
production
work,
both
with
pop
and
hopefully
with
mom
and
they're,
not
too
distant
future.
B
D
So
you
know,
mom
will
be
available
in
a
CSM
tag
as
a
to
to
my
goal
is
to
have
size
six
available
in
something
like
two
three
so
later
this
year.
So
but
I
think
that
it'd
be
great
if
part
of
your
allocation
and
part
of
ours
and
the
polar
climate
working
group
to
would
include
room
for
coupled
ice
ocean
simulations
with
size,
six
and
mom.
G
A
B
D
H
B
A
Yeah
we
just
had
in
the
chat
a
comment
from
Chuck
he's
talking
about
in
the
slide
here.
The
third
big
bullet
point
within
that
the
third
smaller
bullet
point
is
higher
resolutions
and
right.
He
says
how
much
of
a
higher
resolution
is
desired.
How
is
this
determined?
How
much
additional
computer
time
would
be
required
to
run
the
model
with
this
higher
resolution
right.
B
You
know
there
there's
a
lot
of
new
science,
that's
facilitated
by
that
and
from
the
perspective
arm
of
our
working
group,
I
think
the
this.d,
at
least.
In
my
view,
the
strongest
motivation
is
that
those
are
simulations
that
capture
the
intrinsic
variability
of
the
oceans,
not
just
all
forced
variability,
and
that
opens
up
a
whole
new
range
of
physics
that
make
for
interesting
climate
problems.
B
I
think
there's
interest
in
that
atmospheric
model
working
group,
as
well
as
to
moving
to
higher
resolution,
so
I
think
we
should
be
looking
to
have
ocean
configurations
or
commensurate
with
the
resolutions
that
the
atmosphere
is
pushing
towards
in
terms
of
cost
and
mom.
You
know
I
think
with
Mom
6.
B
It's
going
to
go
basically
scaling
with
a
standard,
CFL
kind
of
scaling
on
the
horizontal
resolution,
so
factors
of
2
increase
in
resolution
and
the
horizontal
result
in
factor
of
8
increase
in
the
cost
of
the
model,
so
I
just
was
doing
a
little
back
in
the
envelope
calculation
and
with
our
current
allocation.
If
I
did
this
right,
we
would
have
with
16
and
1/2
million
hours.
That
would
give
us
something
like
18
thousand
years
of
simulation
with
the
forced
ocean
ice
at
in
the
in
this
current
2/3
degree
version.
C
B
Date,
so
that's
for
those
of
you
are
new
to
the
encar
system.
Typically,
the
first
couple
months,
when
the
new
system
arrives,
they
open
it
up
to
a
very
small
number
of
projects,
with
large
numbers
of
hours
to
basically
beat
on
the
machine
as
friendly
users
to
shake
it
down.
So
you
have
to
be
prepared
to
be
up
and
running
when
they
turn
the
machine
on.
B
B
I
A
H
I'm
wondering
if
we
can
learn
something
from
I
think
dfdl
ran
a
quarter
degree
omit
run.
Maybe
I
can
be
corrected
if
I'm
wrong,
but
it'd
be
nice
to
kind
of
look
a
bit
more
into
that
simulation
and
maybe
before
we
restart
because
I
very
much
supportive
of
the
ad
permitting
regime
and
it'd
be
nice
to
learn
from
what
they've
done.
H
F
Well,
well,
for
the
record:
I'm,
not
okay,
I
mean
and
the
reason
they
went
to
a
quarter
of
degree.
Is
that
completely
based
on
computing
budgets
not
to
do
with
physics
whatsoever.
However,
there
is
a
lot
of
reason
to
go
to
these
higher
resolutions
to
resolve
channels
to
resolve
the
RET
reflection
that
you
pointed
out
Justin,
rather
than
just
oh
I've
got
some
eddies
I.
F
F
A
It's
just
gonna
comment:
it
you're
a
back-of-the-envelope
estimate
of
number
of
model
years.
We
can
run
I,
don't
think,
took
into
account
running
with
BGC.
That
is
correct
and
I.
Don't
think
the
project
of
coupling
BGC
into
mom
is
ongoing
and
I.
Think
it's
too
early
to
know.
Yet
what
the
impact
of
extra
chasers
on
the
cost
of
the
model
is.
You
know
we're
not
even
computing
sourcing
terms
yet
so
we
don't
know
yet
what
the
impact
of
bgp
on
model
cost
is
going
to
be.
B
Well,
there's
well
may
this
is
a
question
for
you.
Kees
is
in
order
to
prepare
a
well
supported
proposal.
What
do
you
think
this
is
to
anybody?
What
do
you
think
we
might
need
to
do
between
now
and
say
August
when
we
need
to
be
putting
this
together
to
collect
the
data
required
to
get
metrics
on
how
much
various
configurations
are
going
to
cost?
What
should
we
be
putting
together,
at
least
in
a
prototype
way,
to
met
some
measurements?
B
Should
we
be
setting
up
these
higher
resolution
grids
and
at
least
running
you
know
a
year
or
something
to
measure
them?
Should
we
be
putting
in
enough
of
the
BGC
infrastructure
to
get
some
measurements,
even
if
we
know
that
the
solutions
aren't
going
to
be
scientifically
useful?
Where
should
we
be
putting
our
effort
in
the
next
couple
months.
A
I
think
we
do
want
to
have
at
least
a
cost
estimate
for
our
primary
configurations,
and
one
of
those
is
going
to
be
BGC.
We
also
might
want
to
come
up
with
a
cost
estimate
for
a
lower
res
model-
a
two-degree
model.
Yes,
if
we
think
we're
going
to
be
doing
a
substantial
chunk
of
development
at
that
resolution,
as
a
testbed.
C
C
B
B
B
B
G
Yeah
I
think
what
we're
thinking
now
is
to
keep
moving
ahead
with
one
degree
or
a
kind
of
equivalent
workhorse
set
up.
I
think
you
know
Fred's
work
with
the
point.
One
degree
pop
has
shown
that
you
know
using
some
of
these
other,
like
lighter-weight
da
tools,
could
be
a
good
way
forward
for
exploring
some
of
that
work,
but
I
think
you
know
we're
kind
of
exploring
two
prongs,
one
of
which
is
to
use
the
Foley
nkf
with
a
sort
of
relatively
low
resolution.
G
You
don't
say
one
degree
ocean
coupled
model
and
there's
a
lot
of
interest
in
seeing
what
we
can
do
with
lower
resolution
models
and
thinking
about
you
know,
state
estimates
over
long
timescales,
and
that
is
something
you
know,
maybe
towards
the
end
of
this
cycle.
That
you
know
would
be
interesting
to
explore.
So
definitely
a
vote
for
lower
resolution
applications
in
da.
A
B
Know
that
there
has
been
interest
from
other
working
groups
as
well.
In
addition
to
B
G
C's
Keith
mentioned,
the
paleo
group
is
very
interested
in
a
lower
resolution
version,
as
I
mentioned
in
the
the
overview.
My
shorter
view
talked
the
other
day.
You
know,
that's,
certainly
something
we
can
work
on.
What
would
be
very
helpful
to
move
that
work
forward
would
be
to
have
someone
or
several
people
from
those
other
working
groups
working
with
us
and
configuring
and
evaluating
those
setups
going
forward
we're
you
know
this
is
actually
a
pretty
small
time
effort.
B
There's
three
or
four
people
actively
working
on
mom
six
in
terms
of
run,
writing
code
and
running
the
code.
You
know
and
another.
You
know
no
more
than
half
a
dozen
people
advising
those
three
or
four
people,
so
we
really
need
some
more
participation
from
the
working
group
and,
in
you
know,
doing
hit
the
hands-on
work
of
running
an
experiment,
running
the
analysis
and
you
know,
debugging,
etc.
B
To
move
these
things
forward
more
rapidly,
I
think,
as
you
know,
as
we
are
getting
a
version
out
into
the
to
point
to
release
and
the
sort
of
the
same
infrastructure
is
all
working,
you
can
do
a
create
new
case
and
get
a
mom
six
version
of
the
model
running
with
relatively
low
with
you
know,
without
a
whole
lot
of
pain,
I'm
hoping
that
we'll
we'll
start
to
see
more
participation.
But
that's
it's
really
going
to
take
a
broader
effort
to
move
this
in
all
these
different
directions.
At
the
same
time,.
A
B
B
B
You
know
my
my
experience
may
not
be
other
people's
experience
is
that
tuning
the
low
resolution
models
is
much
much
more
difficult
than
tuning
the
high
resolution
models.
There
are
many
more
ad-hoc
assumptions
and
approximations
and
tweaks
that
you
have
to
make
to
get
an
earth-like
simulation
out
of
these
very
low
resolutions
of
ocean
dynamics.
B
B
That
is
something
I
think
we
can
do
something
fairly
close
to
what
we've
implemented
and
pop,
with
relatively
small,
with
relatively
small
effort
in
mom
six,
the
overflows
and
gravity
currents,
the
price
in
yang
based
entrainment
scheme
we
have
in
pop,
which
I'm
not
as
familiar
with
maybe
gokon
or
bill,
can
comment
on
this.
But
my
understanding
of
the
way
it's
implemented
in
pop
is
sort
of
fairly
closely
tied
to
the
representation
of
the
topography
in
pop,
which
is
very
different
than
in
mom.
B
Six
and
I,
don't
see
you
know
a
clear
path
for
a
simple
port
over
of
the
the
the
way
that
it's
implemented
in
pop.
You
know,
it
may
well
be
the
case
that
a
price
in
yang
stream
to
kind
of
entrainment
model
can
be
implemented
in
an
ale
coordinate
model.
But
it's
not
obvious
to
me
what
the
path
to
that
is.
C
B
F
Well,
I'd:
it's
not
the
stream
tube
model,
that's
trouble,
it's
the
it's,
the
coupling
of
it
yeah,
it's
the
boundary
conditions.
That's
let's
get
that
clear
and
that
that's
really
what
it
is
and
at
least-
and
we
were
trying
to
do
in
the
other
model-
really
the
far-field
stuff.
So
a
lot
of
things
could
be
simplified,
but
it
is
it
isn't
the
stream
tube
model
itself.
That's
no!.
F
Pretty
pretty
drawing
good,
it's
just
how
to
how
to
interface
it
with
these
different
coordinates,
especially
when
they
move
around
so
it'll.
Take
some
thinking,
but
I
do
believe.
The
the
manual
actually
has
the
prescription
that
Bruce
brigly
brought
out.
I've
never
really
studied
it,
but
it's
not
trivial.
That's
for
sure,
yeah.
B
F
A
C
F
On
the
other
hand,
that
the
problems
are
actually
trying
to
resolve
these
things
out,
some
other
way
are
not
trivial
either.
So
it
is.
There
was
a
reason
to
do
that
and
go
through
the
troubles
because
of
even
very
very
high
resolution.
Two
kilometers
wasn't
getting
the
right
entrainment,
so
it
is,
it
is
not
a
trivial
exercise
either
way
to
get
those
things
working
right,
but.
C
The
my
what
I'm
worried
about
is
essentially
Frankie.
You
indicated
also
that
I
mean
I
agree.
Gfdl
has
issues
as
well
it
with
the
representation
of
bottom
flows
the
if
they
are
going
to.
If
we
are
seriously
thinking
about
implementing
something
like
overflows,
we
need
to
figure
out
how
much
or
what
it's
going
to
take
us
to
do
that
in
terms
of
release
or
else,
and
maybe
you
can
look
for
funding
for
that
purpose,
so
that,
like
another
path
to
go.
B
Well,
there
may
be,
you
know,
you
know
we're
learning
the
model,
so
we're
not
really
sure
what
what
the
best
pass
forward
would
be.
But
you
know
the
ail
coordinate
does
give
you
tremendous
flexibility
in
the
treatment
of
you
know
things
like
a
bottom
boundary
layer.
So
you
know
we
could
conceivably
do
something
like
you
know,
a
two-dimensional
bottom
boundary
layer
rather
than
a
stream
tube.
B
B
B
Mom
six
is
a
structured
grid
model.
We
have
some
limited
ability
to
do
regional
refinement
within
the
context
of
a
general
orthogonal
grid.
Doing
regional
refinement
globally
is
quite
challenging
and
would
require
tremendous
amount
of
work
and
basically
redesigning
the
basic
data
structures
and
mom.
Six
we
do
have
I
haven't
put
anything
on
this
list.
B
B
They
were
still
you
know,
hammering
out
some
of
the
final
issues
and
quirks
and
that,
but
it
does
look
to
be
working
quite
well
and
because
we
are
able
to
run
it.
You
know
through
our
regular
coupler
with
our
regular
cap,
it
it's
a
much
less
complicated
regional
model
to
get
working
in
CSM
than
going
to
a
non-native
model
like
roms,
so
that
that
is
an
option
for
doing
more,
focused
regional
work,
but
not
as
straightforward
to
couple
into
a
in
a
fully
global
simulation.
B
A
C
Just
speak
to
that
maybe
Scott
can
see
me
a
few
things.
We
didn't
the
simple
modeling
framework.
With
this
NSF
proposal.
We
were
thinking
of
essentially
contacting
Allister
in
more
detail
with
respect
to
what
needs
to
be
done
in
within
mom
six.
For
that,
because
it'll
be
good
for
our
plan
of
applications.
Do
you
have
anything
to
add
to
that
Scott
yeah.
A
Actually
I
emailed
with
Allister
this
morning
and
he
said
that
it's
it's
not
trivial.
He
said
that
there
needs
to
be
some
changes
within
MMS
and
then
there
would
have
to
be
some
changes
to
the
way:
mom
six
interfaces
with
MMS
and
probably
be
two
or
three
months
of
work,
but
I'm,
trying
to
figure
out
whether
we
can
go
that
direction
or
if
there's
a
way
to
do
it,
where
everyone
wins.
B
So
there
are
some
alternatives
there
beyond
non
spherical,
I,
don't
know
if
one
or
more
of
those
might
be
more
easily
implemented
in
FMS
or
whether
they,
what
their
numerical
properties
are
I'm,
really
not
that
familiar
with
it.
It's
my
understanding
that
the
Tsinghua
University
group,
which
runs
a
version
of
CSM,
is
using
a
one
of
these
Schwarz
Christoffel
grids
in
their
fully
coupled
systems
with
pop.
B
B
So
if
some
you
know
an
applied,
math
student
wants
to
help
us
on
that.
That
would
be
awesome.
Joning
had
a
question
or
a
comment:
she's
just
pointing
out.
There
will
be
more
on
the
ocean
side
for
the
two-degree
Ocwen
Ridge
planets
in
an
upcoming
mam.
6
webinar,
that's
correct,
so
that
sort
of
goes
to
the
second-to-last
point
here
about
getting
I.
Don't
want
we'll
come
back
to
things.
I
just
wanted
to
respond
to
zoning,
but
remind
people
that
we've
got
four
more
webinars
on
the
schedule
through
the
summer.
B
The
next
one
in
next
Monday
is
on
regional
modeling
they're,
each
sort
of
thematic.
The
following
one
will
be
on
the
idealized
models.
Then
we
have
data
assimilation
day
and
then
the
final
one
in
early
August
will
be
on
tracers
in
mom.
Six,
so
and
reminder
that
all
of
the
webinars
have
been
video
recorded
and
the
slides
are
online.
So
you
can
watch
the
seminar
and
have
the
slides
available
for
reference
and
I
think
they're
they're
quite
helpful
in
learning.
B
F
I
know
after
Hallberg
seminar,
the
webinar
I
called
him
and
again
he
also
thinks
it
might
be
time
to
move
away
from
the
constraint
on
the
coupling
that
ain't,
an
ocean
model
or
an
atmospheric
model
has
to
keep
be
given
a
a
certain
forcing
and
keep
it
through
its
whole
coupling
interval.
Even
though
the
it
states
changing
and
the
fluxes
would
change.
That's
one
thing
to
get
away
from
that
restriction
too.
F
Is
work
really
hard
at
just
some
details
so
that
we
can
really
basically
get
high
vertical
resolution
down
to
order
a
metre
or
less
at
the
surface
so
that
the
whole
diurnal
cycle
doesn't
have
to
be
guessed
at
and
the
third
thing
probably
that's
coming
out
and
it
came
out
of
ching
lee's
wonderful,
expose
on
how
different
models
behave
in
these
boundary
layer
schemes
when
it's
transient
forcing.
I
think
we
should
really
consider
moving
away
from
this
idea
that
there's
some
kind
of
quasi
equilibrium
on
a
time
step
and
there's
ways
to
do
that.
F
So
exactly
what
everybody
thinks
and
and
and
who
might
take
a
cut
at
this
in
the
next
year
or
so
and
start
putting
it
forward.
Some
of
these
things
may
not
work
out,
but
if
one
of
them
does
it
would
really
be
a
quantum
change
in
what
we
do
for
CSM
3,
and
it
would
be
nice
to
have
some
really
different
things
in
see
M
3
than
we've
done
in
in
the
past,
and
some
of
these
things
go
back
to
CES,
M,
1
and
B,
and
before
and
it's
time
to
at
least
revisit
them.
B
F
F
B
Yep
I
know
someone
who
wants
to
work
on
that.
If
we
give
them
the
time
to
do
it,
where's
a
smiling
Keith
Moore.
Do
you
do
you
want
to
say
something?
I
I
really
appreciate
your
point.
I
think
you
know
we
really
so
Keith
makes
the
point
in
the
chat
in
terms
of
development.
What
is
our
goal
for
CES,
m3
I.
Think
that's
a
really
good
question.
B
H
I,
do
you
know
so
it
you
know,
I,
guess
it
really
does
make
a
big
difference.
What
our
goal
is
to
do
short
term
climate,
we
don't
necessarily
have
to
be
getting
the
deep
water
formation
right,
but
to
get
longer-term
climate
right.
That's
the
absolutely
absolute
critical
thing
to
get
right,
and
so
you
know
I
would
argue
that
we
should
be
looking
for
these
longer
timescales
and
prioritize
that,
but
maybe
it's
something
we
should
discuss
and
see
with
the
group
how
we,
how
we
feel
about
it.
H
I'm
sure
you
guys
are
way
more
frustrated
than
me
that
were
not
running
our
ocean
models
and
Eddy
resolving
resolutions
for
climate
studies.
I
think
we
really
need
to
and
I
think
as
a
working
group,
we
should
be
pushing
for
the
resources
to
do
that
and
I
think
you
know
over
the
last
10
years
the
atmosphere
has
gotten
a
lot
more
resource
priority
and
CPU
time
and
we've
kind
of
been
trapped
at
our
resolution
by
computational
constraints.
B
H
B
B
C
C
Computers
and
it'll
be
delivered
up
around
it's
still
on
schedule
by
the
end
of
this
year
at
crinoline,
and
they
are
actually
preparing
a
cloud
resolving
OSHA's
for
cadres
of
emotion
what
colors
are
coming
atmospheric
model
at
five
kilometer
resolution
with
a
sea
dike
or
they
are
converting
that
model
to
run
on
sun
way
and
a
system,
and
apparently
they
already
have
a
five
kilometer
version
of
pop
running
in
beijing
on
some
machine.
So
the
their
goal
is
the
couple
that
five
kilometer
ocean
to
five
kilometer
atmosphere
and
run
it
for
some
time.
C
C
H
B
C
B
B
Let's
see,
no
anyone
else
have
a
subject
or
topic.
I
guess,
on
this
question
of
future
meetings,
I
ran
through
what
we're
doing
on
webinars
I
think
when
we're
done
with
this
series,
we're
gonna
want
to
take
a
break,
but
I
guess
I'd
like
to
hear
whether
people
outside
of
our
little
group
feel
that
they've
been
helpful.
Whether
this
is
something
they
would
like
to
see
happen
again
in
the
future.
B
You
know
we
probably
wouldn't
do
the
whole
mom
six
focus
again,
but
other
topics
might
be.
Oh,
you
know
we
might
sort
of.
Look
to.
You
know
a
series
of
webinars
in
a
particular
area.
Is
this
something
again?
This
would
be
something
that
you
know
it
would
happen
if
someone
pitches
in
and
helps
organize
it
and
and.
B
F
It's
been
very
useful
and
informative
for
one
thing,
so
so
that
that
has
been
really
helpful
and
you
see
what
other
people
are
doing
and
how
well
they're
doing
it.
But
it's
also
spawned
some
conversations,
as
I
said,
with
Alburgh
and
whatever,
and
it
gives
you
the
impetus
and
things
that
have
been
on
the
backburner
in
the
back.
Your
mind
come
to
the
fore
so
know
that
this
has
been
and
the
participation
has
been
very
high
and
I.
Think
you
should
take
that
as
a
yes,
okay,.
B
C
D
A
I
To
get
some
of
these
things
done,
we
need
to
have
people
start
using
so
then
things
will
be
identified
and
improved,
so
I
just
want
to
psych
on
there
like
right
now,
like
you
know,
we
have
a
summation
three
people
actually
working
with
Mon
six
in
the
ocean
section
and
then
seaneen
came
along
and
is
helping
a
lot.
So
I
think
you
move
way.
Faster
was
more
people
start
yeah.
G
A
B
B
H
B
Base
but
the
you
know,
the
scope
I'm
pointed
out
the
the
tenth
degree
ocean
model
configuration
is
one
we
put
together,
I
think
in
2012,
so
there
are
a
number
of
things
that
we've
developed
that
would
be
applicable
since
then,
things
like
tidal,
mixing
and
others.
You
know
other
the
the
EBM
is
not
fully
implemented
or
hasn't
been
fully
implemented.
In
the
tenth
degree.
C
B
But
all
of
those
things
would
require
some
human
resource
effort
and
that
would
draw
resources
away
from
our
mom
six
development
unless
some
other
people
stepped
in
to
take
take
the
lead
on
those
things.
So
I
guess
that's.
The
question
is:
are
there
people
in
the
OM
WG
who
would
be
willing
to
take
the
lead
on
bringing
you
know,
whatever
new
physics
were
deemed
appropriate
into
the
pop
pop
code?
We've
snow,
you
know
it's
something,
we've
sort
of
said:
we're
really
not
gonna.
F
F
Fine,
so
I
mean
one
of
the
things
that
has
been
a
big
problem
for
a
long
time.
Is
that
all
the
model
components
go
off
and
create
their
versions
and
then
plug
them
all
back
together
in
a
couple
version,
and
then
it's
sometimes
years
as
we
heard
trying
to
get
them
to
play
nice
right.
So
from
an
ocean
point
of
view,
this
is
a
chance
to
change
that
too,
or
an
asynchronous
development
and
say
look.
C
Know
I
mean
that's
the
thing,
because
there's
the
vertical
great
vertical
resolution
issue
the
dynamical
course
election.
If
you
do
not
at
this
point,
creates
and
that's
what
I
feel
like
if
we
do
not
at
this
point
create
a
new
high-res
version
of
CES
M,
it
will
not
happen
for
another
five
for
six
years.
F
Yeah,
but
what
component?
What
is
the
component
strategy?
Well
together?
We're
all
gonna,
go
I
res
and
then
bung
them
all
together
again,
not
necessarily
yeah.
Well,
that's
what
I'm
saying
think
about
a
whole
evolution
of
maybe
more
of
an
asynchronous
strategy
to
get
to
where
you
want
to
go,
I
mean
the
thing
is
that
a.
C
Another
issue
is
that
certain
components
will
not
work
on
older,
with
older
versions
of
the
models
that'll
take
a
long
time
to
do
also
yeah
so
I
mean
land
model
is
a
prime
example.
We
can't
really
run
the
clm
for
in
the
newer
version,
essentially
without
a
lot
of
work,
and
if
we
continue
with
our
current
high
resolution
configuration
then
model
people
are
not
really
interested
in
looking
at
the
resolution,
a
high
resolution
with
CLM,
for
they
are
saying
that
we
passed
that
stage.
C
F
Mean
high
res
atmosphere
ocean.
What
do
you
mean?
What
components,
components
right
then
model
yeah?
Well,
the
land
model
might
have
to
be
one
of
the
early
developers,
but
not
all
the
components
necessarily
is
all
I'm
suggesting,
but
that
that's
why
you're
in
the
chair
and
you
can
work
that
out.
Yeah
well.
B
I
guess,
there's
sort
of
two
levels
of
questions
he's
addressed:
one
is
whether
there
are
sort
of
coding
compatibilities,
with
versions
of
the
model
that
preclude
it
being
used.
You
know
in
a
newer
version,
even
if
we
don't
change
any
of
the
physics
I.
Don't
think
that's
the
case
with
pop,
because
we've
already
got
it
running
with
new
op,
C
and
so
I
think
we're
good,
but
you
know
yeah
so
and
then.
B
If
this
is
a
super
high
priority
for
the
CSM
project,
you
know
we
would
have
to
consider
how
we
would
accommodate
those
needs.
But
you
know
I
I'm
kind
of
with
Bill
that
you
know
if,
unless
there's
something
that
prevents
the
model
from
running
with
the
version
we've
got
now,
you
know,
let's,
let's
see
if
that
is
sufficient.
I
do
I.
B
Do
think
that
it
is
useful
for
us
to
sort
of
play
in
that
game
to
some
degree,
because
this
kind
of
relates
to
some
of
Justin's
work
on
coastal
SST
biases
that
the
biases
in
certain
aspects
of
the
atmospheric
forcing
emerge
or
manifests
themselves
differently
at
different
ocean
model
resolutions,
and
it
can
be
useful
to
see
that
diversity
in
in
our
own
development
cycle.
So
so
I
don't
know.
Maybe
you
know
this
summer.
Is
there
doing
the
SEMA
exercise?
B
C
A
And
that
the
conversation
started
off
with
we're
using
this
configuration
that
steady
resolving
that
we've
had
the
same
thing
for
10
years.
Well,
why
not
switch
to
CES
m2?
It
includes
Ocean
BGC,
if
you're,
if
you're
saying
that,
there's
a
complaint
that
it's
the
sable
sable
then
do
something
really
different.
Well,.
C
A
It
seems
to
me
that
there's
just
simply
not
sufficient
resources
to
invest
a
lot
of
time
in
upgrading
up
with
a
parameterization.
So
if
the
coupling
effort
is
to
be
undertaken
with
minimal
effort,
a
couple
kind
of
you
know
a
minimally
upgraded
version
of
pop,
then
that
that
should
proceed,
but
otherwise
it's
gonna,
be
it's
gonna,
be
hard
cuz
we're
just.
We
can't
do
everything.
D
A
B
D
So
so
this
came
up
in
the
context
of
CIA's,
because
you
know
we've
been
stuck
on
the
old
sites
for
physics
for
high
resolution.
As
you
say,
ten
years,
and
it's
been
something
I've
been
really
wanting
to
do-
is
to
move
to
the
CSM
to
physics
that
we
use
for
the
low-resolution
sea
ice.
But
we
haven't
had
the
opportunity
to
really
vet
that
and
test
that
out
yet
so
so
that
that's
fine,
that's
on
me
to
test
and
I'm,
not
suggesting
that
should
be
added
to
your
list
of
things.
D
But
then
my
concern
that
I'm
hearing
from
Baylor
too,
is
about
the
wave
model
and
you
know:
are
there
issues
if
you
start
running
high
resolution
pop
and
sea
ice?
You
have
to
also
do
increase
resolution
in
the
wave
model
too,
like
how
do
you?
How
do
you
do
that
coupling
so
I
can
definitely
see
that
there
are
some
potential
barriers
here
to
getting
everything
upgraded,
but.
C
I
guess
the
question
that
I'm
coming
from
repeating
in
a
sense
what
I
said
already
the
current
high
res
version
that
we
have
is
order
10
years
old
and
if
we
essentially,
if
the
working
groups
do
not
want
to
essentially
spend
the
time
to
upgrade
that
version
of
the
model,
to
CSM
to
something
and
that's
exactly
what
I'm
saying
then
they
likely.
We
will
not
create
an
high
resolution
CSM
to
version
and
we
have
to
wait
likely
till
CCS
see
sm-3
is
created,
and
that
is
probably
five
to
six
years
ahead.
C
F
A
A
C
A
But
it's
not
it's
not
our
job
to
tune.
The
acid
I
called
you
run
with
the
ocean.
I
mean
we,
our
the
the
CSM
version
of
the
model,
like
the
part
model,
is,
is
available
and
we
have
simulation
with
that.
So
need
know
if
you
Eve,
so
that's
why
I'm
confused
like
do
we
do?
We
need
to
also
a
grade
like
to
the
latest
and
greatest
physics.
Then
that's
well.
C
B
A
C
I'm
not
talking
about
the
case
I'm
talking
about
essentially
a
fully
coupled
version
of
the
high-res
version
with
physics
that
we
would
like
to
see.
Some
is
a
scale
mixing
Frank
messaging,
as
jury-box
model
their
interactions.
They
mentioned
sea
ice
issues
and
all
that,
so
those
are
those
that's
the
thing
that
I'm
talking
about
I.
D
Think
Keith
Moore
typed
into
the
chat
and
I
agree
with
this
that
it
just
it
seems
at
this
point.
If
you're
already
going
wholesale
to
mom
you're
changing
the
motto
completely,
your
target
should
rather
be
see
sm-3
for
a
new
high-res
version,
because
there's
going
to
be
a
new
ocean
component
sea
ice
component,
you
know
I
think
that
that's
that's
a
much
better
target
to
test
out
all
the
new
physics
in
rather
than
trying
to
retrofit
pop
with
physics,
changes
and
things
like
that.
I
I
I.
D
Personally,
don't
think
it's
and
I'm
you
know
and
I'm
thinking.
You
know,
I'm
not
going
to
use
I'm,
not
gonna
update
size,
five,
which
is
in
the
CSM
to
code
base.
You
know
so
you
know
I
move
into
size,
six
in
the
near
future,
so
I
think
personally,
as
Keith
Moore
was
saying
that
we
should
maybe
just
target
CSM
three
for
this.
A
F
That
does
not
mean
when
you
couple
it
back,
it's
gonna
actually
give
you
a
better
solution
in
any
regard,
and
that's
the
thing
I
think
we're
really
scared
about,
because
we
know
sometimes
reconciliation
of
these
kind
of
changes
can
take
a
year
or
so
so
I
think
you've
got
a
lot
of
reluctance
here
to
do
anything
more
on
the
high
resolution
Park
and
for
good
reason,
I
mean
it's
not
just
saying
no
I
think
there
are
really
good
reasons.
No.
C
I
mean
that's
so
that's
fine
I
mean
that's
similar
message
that
I
got
from
the
atmosphere
while
working
as
well,
even
though
they
wanted
a
feed
in
high
res
model.
They
don't
want
to
put
in
the
time
essentially
to
create
that
or
because
they
think
that
it's
going
to
be
the
next
version,
and
we
need
just
a
committed
message
that
the
next
resume
high-res
version
of
the
CSM
may
not
be
available
until
CES
m3,
and
you
just
make
that
message
clear.
Like.
D
Phil's
point
about
staggering
things
to
that
that
you
know
we
can
work
on
a
couple
dice
ocean
high
res
configuration
over
the
next
couple
of
years
before
maybe
the
cam
folks
are
ready
and
do
some
independent
testing
of
this
with
JRA
forcing
and
then
sometime
later
in
the
future,
we
can
aspire
to
that
fully.
Coupled
version.
B
I
I
B
I
I
I
I
So
the
first
point
here
is
like
the
purpose
of
the
tutorial
so
like
it
hasn't
change
and
I
know
this
is
the
second
year
and
participating
they
need
the
more
I
heard.
It
hasn't
changed
in
a
long
time
and
the
format
and
the
lectures
get
adjusted,
but
it's
basically
the
same
as
when
you
was
created.
So
maybe
it's
time
to
rethink,
including
what's
the
purpose
of
the
tutorial
and
I
shouldn't
be
focused
more
like
on
early
career
and
should
it
be
merged
with
the
CSN
workshop.
A
Opinion
is
that
it
should
not
knowledge
with
the
workshop
I
think
it's
completely
different
purpose
I've
participated
in
many
years
and
in
particular,
the
interaction
with
the
young
people,
who
are
there
the
personal
interaction
with
both
the
scientists
and
with
the
people
who
are
helping
them
to
actually
run
the
code.
I
think
is
essential
to
make
it
work
well
this
year
we
won't
be
able
to
do
that,
but
I
think
it
will
not
work
nearly
as
well
without
the
personal
interactions
that
are
created
in
this
in
this
format.
A
I
think
in
particular
that
the
tutorial
is
more
important
to
have
one-on-one
interactions
than
there
may
be
this.
This
annual
meeting
I
think
it
really
essential
that
those
occur
now
I,
don't
know
what
criticism
has
been
of
the
present.
You
know
the
the
present
format
that's
been
over
the
last
few
years,
because
in
most
of
those
times,
I've
heard
that
the
feedback
has
been
that
it's
really
been
pretty
successful.
A
I
One
of
the
criticism
is,
for
example,
participation
from
minority
institutions,
so
this
year
in
particular,
we
had
extra
funding
and
going
to
you
can
confirm
that
and
we
reached
out
to
those
institutions,
but
we
didn't
get
any
response
so,
like
you
know,
when
we
talk
about
like
including
minorities
and
making
science
more
available
to
those
like
if
the
entry
point
for
them,
it's
not
being
effective
at
bringing
them
up.
It's
like
it's,
that's
a
my
view
of
a
big,
the
negative
of.
C
The
tutorial
is
right
now
is
that
we
don't
have
miner
to
serving
institution
representation,
a
lot
and
also
other
represented
groups,
and
we
try
to
reach
actively
to
some
of
those
institutions.
And
we
are
bringing
too
many
people
just
like
us
from
the
universities
that
are
known
to
us,
and
that
is
a
major
criticism
of
the.
F
Okay,
well,
let's
fix
that
rather
than
the
format,
because
Peter
I
think
I,
think
you're,
right
and
and
there
there's
been
questionnaires
after
every
tutorial
now
clearly
they're,
the
ones
that
are
here
so
I
think
read.
If
you
have
to
fix
a
problem
fix
the
problem,
but
you
don't
have
to
sink
the
ship
to
do
it
right
right.
D
So
one
of
the
one
of
the
issues
from
my
experience
on
the
tutorial
committee
is
that
a
lot
of
these
minority
serving
institutions
and
historically
black
colleges,
often
don't
have
strong
graduate
student
programs
related
to
what
we
do
and
so
then
the
issue
is,
you
know
we
we
try,
we
try
to
go
out
to
them,
but
they're
not
applying
because
they
don't
feel
qualified.
They
don't
feel
like
there
ready
for
this
sort
of
thing,
so
is
one
thing
that
we
could
do
is
somehow
dumb
it
down.
D
Well,
that's
sorry,
that's
not
really
the
right
phrase,
but
but
I'm
just
saying
lower
the
level
to
an
undergraduate
level
rather
than
targeting
graduate
level
and
above
you
know,
I
mean
I
I,
just
I.
Just
think
that
the
problem
is,
you
know
we
we
wouldn't
accept
undergraduates
previously
and
and
maybe
there's
an
opportunity
there.
If
we,
if
we
went
to
undergraduate
level,
it's
just.
A
A
J
C
G
Yeah
I
would
agree
that
this
seems
to
be
a
years
long
process
to
try
and
train
and
involve
people
I
wonder
to
what
extent
there
are
sort
of
complementary
or
parallel
resources
online,
where
people
could
kind
of
get
a
foot
in
the
door
gain
competence.
They
know
what
this
is
about
and
then
they
might
be
excited
about
going
learning
more
and
kind
of
networking
you
know,
can
I
go
on
YouTube
or
something
and
take
the
CES
m-class
or
you
know
some
sort
of
like
equivalent
of
MIT
OpenCourseWare
or
something
does
that
exist.
A
Most
important
things
for
for
shepherding,
you
know,
mine
already
groups
into
the
into
the
process
is
mentorship,
and
so
perhaps
there's
an
opportunity
to
pair.
You
know
establish
a
program
where
people
who
are
already
in
the
CSM
community
can
either
volunteer
or
evolve
to
to
serve
as
mentors
and
then
make
it
advertise
kind
of
a
companion
program
where
you
know
you're
you're
you
arrive
and
you're
supported
automatically
by
by
having
a
sort
of
mentor
or
an
advisor.
A
J
J
I
think
you
cheer
with
with
NCAR
because
of
that
tutorial,
and
so
if
there
was
some
kind
of
an
effort
to
reach
out
to
tribal
colleges,
I
can
tell
you,
as
a
graduate
and
as
currently
faculty
that
there
aren't
really
tribal
colleges.
There
are
you
know
over
a
hundred
tribal
colleges
in
the
country,
but
they
don't
have
graduate
programs.
J
They
have
undergraduate
programs
and
I
know
that
there
are
students
that
would
be
interested
in
a
CSM
tutorial,
but
it
seems
like
something
that
is
just
so
far
beyond
their
reach
right
now,
and
so,
if
there
was
some
kind
of
a
way
to
make
it
more
accessible
to
an
interested
undergraduate
program,
students
I
think
that
that
would
be.
You
know
a
way
forward.
In
terms
of
you
know,
I
can't
speak
to
it
to
historically
black
institutions,
but
it's
from
the
tribal
college
perspective.
B
J
J
J
You
know,
like
I,
said:
Katie
dagon
from
the
clm
came
and
she
did
a
guest
lecturer
and
my
students
were
really
engaged
and
interested
in
the
work
that
cars
doing
but
to
them
it
seems
like
you
know
most,
my
students
have
grown
up
the
reservation
they've
not
been
to
endcard.
They
don't
really
have
you
know
that
sort
of
I
don't
know
how
to
put
it
that
sort
of
interaction
with
the
level
of
science
happening
it
and
car.
J
You
know
a
lot
of
our
students
will
go
on
to
the
jobs
in
tribal
offices
and
and
things
like
that
and
so
yeah
I
think
there
there's
definitely
interest.
But
there's
it's
like
the
barrier
is,
you
know,
partially
yeah
yeah
the
source
program,
yep
and
I.
Had
you
know
some
students?
This
that
applied
to
the
the
program
out
of
Haskell
this
summer,
so
those
are
definite
bridges
and
maybe
yeah
strengthening
those
bridges
too.
So
sorry,
I.
F
Have
a
question
because,
based
on
what
what
people
in
Sizzlin
rich
loft
tried
really
hard
to
do
that
with
a
lot
of
success,
and
one
thing
he
did
is
he
took
things
out
to
these
institutions
rather
than
have
them
first
come
here
and
show
it
goes
on
to
the
guest
lecturer,
but
you
could
consider
a
Primmer
tutorial.
That's
a
one-person
show
that
several
people
could
take
out
perhaps
and
at
least
get
the
introduction
and
that
think
about
it.
That's
what
he
found
was
extremely
important
was
not
to
assume
people
can
travel
like.
We
do
absolutely.
D
D
And
and
Pallas
you
know
tell
us
like
what
sort
of
tweaks
you
could
imagine
that
might
help.
You
know,
because
you're
right
traditionally,
we've
favored
graduate
level
and
above
because
we
always
felt
like
the
material
was
at
a
level
that
would
fit
that.
But
we've
had
a
couple
of
really
strong
undergraduate
supply
in
the
past
and
we've
had
to
turn
them
down
and
and
I've
always
felt
bad
about
that,
and
so
I
think
you
know
if
there's
some
things
that
we
could
do
to
make
it
more
accessible.
D
J
That
builds
trust
and
that
you
know
kind
of
breaks
down.
The
barrier
of
you
know
in
car
is
something
so
far
of
so
far.
Apart
from
what
we
do,
you
know
on
the
Pine
Ridge
Reservation
that
there's
almost
like
there's
a
mental
barrier
to
even
you
know,
have
a
thought
for
any
student
to
apply
to
go
to
a
tutorial
at
anchor,
so
I
think
yeah
having
scientists
be
more
accessible.
Maybe
do
some
outreach.
You
know
time
permitting
who
has
time
for
anything
right
well,.
B
J
J
J
J
E
E
E
Heiko
is
really
some
of
the
points
that
even
when,
when
I
first
thought,
I
was
heard
about
the
CSF
workshop.
I
was
already
in
grad
school
looking
at
oceanography
and
still
it
seems
to
me
to
be
a
beautifully
both
and
a
bit
too
fancy
and
like
I,
wasn't
sure
you
face
the
really
accessible
for
me.
But
what
time.
E
E
Just
to
make
a
comment
on,
what's
being
shown
by
Kurosawa
here,
appoint
a
few
ports
of
discussion,
particularly
for
this
year.
I
heard
that
he
might,
in
my
experience,
when
I,
when
I
did
it
two
years
ago
and
the
word
and
and
also
sitting
there
last
year,
I
I,
think
in
terms
of
interacting
for
the
students,
the
the
concurrent
or
the
conventional
format,
it's
pretty
effectively
in
combining
the
the
theory
and
the
practice,
and
especially
its
interaction
with
the
scientists.
E
What's
what
this
is
like,
and
and
even
though,
even
though
the
in-person
component
cannot
be
replaced,
I
all
say
that
and
still
is
going
out
in
the
world
right
now,
everybody
tries
to
establish
these
connections
through
the
virtual
format
and
but
recently
seeing
sizzle
and
other
organizations
replacing
their
entire
internship
program
in
the
online
format.
So
I
think
there
are
there
definitely
losses
of
not
being
physically
present,
but
additionally,
we
can
can
also
use
some
of
the
new
opportunities
being
opened
in
particular
this
year
because
of
our
for
school.
E
A
I
D
For
the
first
you
know,
and
and
as
he's
seen,
you
know
we
have
to
sort
of
try
it
out
first
and
see
how
it's
gonna
work
out
with
you
know
so
many
students
as
it
is
but
I
agree,
Peter
I
think
that
future
workshops
could
be
a
combination
of
virtual,
an
in-person
potentially,
and
maybe
that
would
allow
you
know
more
undergraduates,
especially
those
that
couldn't
travel
to
to
attend
and
just
sort
of
see
it,
and
they
don't
get
that
in
person,
interaction
necessarily
as
xiangning
missing.
But
it's
still,
I
think,
maybe.
B
So
let
me
just
sum
this
up.
I
think
there's
been
a
great
conversation,
but
maybe
let
me
see
if
I've
captured
this,
that
people
think
the
current
format
is
very
effective
for
the
community
we've
been
targeting,
but
that
perhaps
we
need
to
think
of
a
new
format
and
other
opportunities
to
bring
in
communities
that
we
have
not
reached
thus
far.
A
B
B
So
I
don't
know
that
we
want
to
mess
around
too
much
with
the
current
tutorial
because
it's
obviously
under
it
is
there's
a
huge
demand
for
it.
We're
serving
it's
not
like
we're
serving
a
small
community
here.
There's
a
lot
of
people
want
to
come
and
do
this
tutorial,
but
we
there
we
should
look
for
opportunities
for
how
to
expand
our
community
so.
B
B
D
I'm,
not
I,
guess
I'm,
not
quite
sure
what
you're
getting
at
Joanie,
but
the
the
main
restriction
for
limiting
the
participants
in
the
past
has
been
just
space
on
the
Mesa.
You
know
we
just
simply
could
not
accommodate
more
than
80
people
on
the
Mesa
for
the
further
tutorial,
but
I
think-
and
you
know,
I-
think,
there's
generally
enough
volunteers
and
help
to
assist
with
potentially
more
but
but
it's
just
as
far
as
I
can
assess
I
mean.
Maybe
Gustavo
would
like
to
comment
separately
there.
I
A
D
A
D
To
the
LS
fq,
and
so
that
helps
naturally
soar
to
stagger
things
anyway,
so
so
I
think
Joe
need
to
to
sort
of
get
back
to
your
original
question.
I
think
that
we
could
potentially
expand
online
presence
broadcast
for
these
things,
and
then
it's
just
that
some
of
the
online
participants
might
not
get
that
one-on-one
help
as
much
as
those
that
are
physically
here.
So.