►
From YouTube: SimPEG Meeting June 11th
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
B
Yeah
I've
got
something
so
I
guess
just
looking
for
a
review
on
the
fourth
level
or
discretised
tutorials,
there's
also
a
pull
request
for
models
and
mapping
for
simpang
tutorials
with
like
the
tiniest
refactor.
So
if
those
could
get
in
that
would
be
great,
it
seems
like
pretty
low
hanging
fruit
and
that
the
review
should
take
too
long.
But
those
are
things
that
we
could
get
in
fast
and
through
the
website.
B
E
On
that
note
would
like
I'm
happy
to
try
and
jump
in
and
review.
I
honestly
will
not
be
able
to
get
to
it
until
this
weekend,
but
I
actually
think
if
somebody
who's,
if
perhaps
a
bit
newer
to
simpe,
is
willing
to
actually
go
in
and
give
that
a
try
read
through
post
questions
and
things
like
that.
That
would
be
extremely
valuable.
Is
there
anybody
who
would
be
willing
to
actually
like
just
take
a
crack?
It
could
take
a
crack
at.
B
C
E
E
On
dashboarding,
so,
basically
the
idea
being.
How
do
you
creates
interactive
interfaces
from
jupiter
notebooks
that
are
a
bit
richer
or
in
some
like
actually
simpler,
so
not
necessarily
showing
any
of
the
code?
And
so
I
can
show
you
a
quick
demo
if
that's
of
interest
set
of
interest,
yeah,
okay,
it's
not
he's
interested,
so
I
will
do
so.
E
There
is
panel
which
integrates
with
bouquet
tasks
a
lot
of
the
like
more
anaconda
developed
ecosystem,
and
then
there
is
wallah,
which
I
think
is
quite
natural
as
a
connection
to
to
the
jupiter
ecosystem
and
it's
fairly
general.
So
what
I'll
show
you
is
just
a
quick
demo
of
a
notebook
I
put
together
actually
with
the
ground
water
project
in
mind,
so
the
geoscientists
without
borders.
So
I
can
show
you
where
this
is.
E
You
see
that
yeah
I
mean
yes,
okay,
excellent!
So
here,
if
you
go
right
now,
this
is
just
sitting
on
my
github,
because
I
was
just
playing
around
but
I'm
happy
to
move
it
over
to
some
of
its
impact.
Related
projects
and
right
now,
I
mean
I'll.
Add
a
readme
and
stuff
like
that
before
I.
Do
it
but
I'll
put
this
link
in
the
notes.
It's
under
my
github,
it's
GWB,
/
DC,
slash
dashboard,
so
I'll
put
that
here.
E
So
what
it
looks
like
if
you
come
there,
there's
actually
just
two
two
notebooks
right
now:
there's
the
DC
for
word
simulation
actually
I
think
that's,
probably
the
only
notebook
that's
in
there.
So
what
I've
done
is
basically
set
up.
It's
running
a
2d
DC
simulation,
but
it's
just
it's.
Basically
we're
gonna
like
the
goal,
is
to
think
through
a
1d
inversion
down
the
road
you
set
it
up,
so
we
can
create
a
layered
earth
model.
E
E
So
that's
cool,
so
that's
just
a
widget
that
I've
created
in
the
notebook,
but
then
what
we
can
actually
do
is
we
can
run
this
same
thing
with
voila,
and
so
what
that
then
looks
like
is
this,
and
so
what
you
actually
see
is
just
a
single.
It's
a
webpage
right
it
just
you,
don't
actually
even
see
that
there's
just
like
you,
don't
see
anything
related
to
Jupiter,
but
here
we
have
the
same
the
same
functionality,
so
I
can
go
in
and
change
change.
E
So
what
I'm
doing
to
run
this
is
instead
of
running
Jupiter.
Notebook
I
just
run
voila
in
this
in
this
repository,
and
then
it
gives
you
a
very
simple
tree
view
on
the
repository.
So
this
is
what
happens
if
you
run
at
the
repository
level,
and
then
we
have
the
same
ford
simulation
notebook
they'll
render
as
just
a
web
page.
E
So
here
yeah
we
again
have
sort
of
the
same
functionality.
You
can
go
in
and
change
change
your
model
and
look
at
the
currents
and
potentials,
and
things
like
that,
and
so
it's
kind
of
cool
about
this
is
it's
actually
the
the
way
that
it
is
serving
the
notebook
is
different
or
the
way
that
it's
serving
the
content
is
actually
very
different
than
the
notebook.
E
This
actually
was
set
up
with
the
use
case
in
mind
that
you
don't
want
users
to
arbitrarily
execute
code,
and
so,
in
this
case,
basically,
all
that
the
user
can
do
is
interact
with
the
interface
that
I
gave
them
and
see
inputs
and
outputs
from
there,
whereas
in
the
notebook
you
know,
users
can
actually
go
in
and
execute
whatever
code
that
they
want.
So
if
you
want
to
sort
of
create
it
like
a
self-contained
app
follow
is
a
really
cool
way
to
do
that.
E
H
I
just
will
say
this:
this
looks
great
Lindsey,
I'm,
sorry,
I
missed
about
the
last
minute
or
so
I'm,
not
sure
what
happened
but
I'll
check
I'll
check
it
out.
Okay,.
E
I
E
This
is
like
the
updated
version,
basically
so
Jupiter
dashboard
I
think
like
there's
sort
of
thoughts
that
might
go
into
what
that
will
look
like
down
the
line.
But
voila
is
like
it's
flexible
enough
that,
even
if
there
are
changes
are
like
look
even
if
there
is
like
a
dashboarding
mode
within
jupiter
library
can
lay
stuff
out.
Specifically
so
they've
used
that
idea
with
him
for
a
lot
to
serve
like
ask
everyone
out.
J
E
E
We
can
do
the
exact
same
thing
on
binder,
just
as
you
would
with
a
notebook,
but
then
the
other
thing
that
we
chatted
a
bit
about
is
the
ability
to
basically
wrap
it
up
into
a
QT,
app
and
so
like
I
can
basically
define
any
standalone
application
that
runs.
This
runs
this
notebook
and
I'm
gonna,
guys.
E
Yeah,
okay,
so
you
can.
You
can
set
this
to
me
a
basically
like
set
it
up,
so
that
it
is
a
standalone,
installable
Apple
if
so
desired,
and
so
that's
not
like
hey.
It's
not
a
streamline
trajectory
yet,
but
I
did
work
with
one
of
the
developers
who
was
attending
a
workshop
as
well.
To
actually
do
that
for
this
app,
so
I've
gone
through
the
workflow
once
and
we
can
basically
deploy
this
as
like
a
standalone,
you
double
click
installer
and
go
which
is
pretty
exciting.
E
E
And
then
the
other
quick
update
on
the
GWB
project
I
was
working
a
bit
with
UV
he's
one
of
the
core
jupiter
developers
here
and
we
were
figuring
out
basically
well,
he
was
figuring
out.
I
was
just
telling
him
what
we
need
and
he's
figuring
out,
which
is
awesome,
but
it's
expressing
some
of
the
constraints
around
not
knowing
what
kind
of
computers
people
are
gonna
have
not
wanting
to
actually
have
to
really
deal
with
the
installation
and
all
that
sort
of
stuff.
E
And
so
what
he's
been
working
on
is
basically
setting
up
effectively
like
the
idea
of
a
repo,
a
repository
to
a
USB
stick.
So
what
we
can
do
is
basically
set
up
an
Ubuntu
or
like
a
Linux
machine
that
actually
runs
off
of
a
USB
stick
that
has
all
of
the
software
pre-installed
and
then
all
of
you,
as
a
user
have
to
do,
is
restart
your
computer
and
instead
point
it
to
point
it
to
the
Linux
version.
E
Running
off
of
this
USB
stick
rather
than
your
root
environment,
and
we
can
have
basically
complete
control
over
that
software
environment.
So
people
can
get
you
don't
have
to
worry
about.
Trying
to
deal
with
their
crazy
Windows
systems,
so
we
still
need
to
test
this
out,
and
it's
gonna
take
some
yeah
some
trial
and
error
to
get
it
all
going,
but
I
think
it's
an
exciting
trajectory
forward
for
getting
software
to
people
do.
J
E
E
E
The
other
thing
is
like
the
next
time
around.
Well,
we
could
talk
to
basically
see
if
they
can
get
their
IT
to
have
some
of
like
some
stuff
in
place,
so
that
we
can
actually
work
with
anaconda.
If
not,
then
we
can
probably
set
up
our
own
installer,
but
I.
Don't
know
enough
about
like
corporate
firewalls,
to
know
how
how
to
get
this
set
up.
C
Maybe
a
small
update,
yeah
well
I,
went
into
which
regard
to
this
thing
for
kids
around
trying
to
create
you
with
a
few
people
and
I
went
to
the
science
world
here
to
speak
to
the
people
that
do
outreach
and
work
with
teachers
and
stuff
like
that
and
there's
just
something
quite
interesting,
actually
is
to
try
to
work
on
something
with
scratch.
Do
you
know
this
language
scratch
it's
like
according
to
for
kids?
C
So
it's
something
like
Lego
who
actually
you
get
blocks
that
contain
like
if
statements
or
else
statement,
sir,
you
can
actually
put
anything
with
them,
but
it's
more
like
you
have
to
place.
It's
visually
answered
north,
and
so
you
can
create
from
little
stories
with
animated
figures
up
till
we
full-fledged
applications.
C
That
also
learned
going
actually
while
learning
something
but
you're
first
thinking
if
I
still
find
the
time
or
afterwards
should
be
when
I'm
still
here
to
be
good
to
pick.
Some
may
actually
from
the
same
peg
website
likes
all
the
education
that
you
have
made
simple
things
and
try
to
reconstruct
that
with
this
similar
language,
maybe
a
different
way
with,
so
that
we
can
also
refer
them
to
the
more
complex
code
and
so
on
later
on,
but
that
the
simple
people's
witness
you
know,
we've
been
actually
nice
and
I,
think
it
was
a
nice
feedback.
C
We
got
from
science
world,
that's
actually
when
they
teach
teachers
to
to
start
working
on
this
computational
thinking
gets
contemn
yeah
like
making
its
thing
like
a
program.
That's
actually
the
idea.
Yeah.
Do
you
have
any
thoughts
about
an
example
like
if
we
write
some
scripts
on
content
and,
for
example,
we
make
one
on
a
compass?
C
Well,
maybe
just
a
seismograph,
but
it
would
actually
be
nice
to
code
some
of
the
examples
or
the
experiments
that
we
put
there,
for
example,
that
they
can
actually
create
their
own
compound
compass
and
that,
for
example,
we
in
the
code,
you
could
have
an
objects
that
is
magnetic
and
if
you
move
it
around
in
the
image
it
affects
your
compass
or
something
like
that.
I'm
just
doing
something
playful,
but
yeah
and
he's
done
a
related
to
the
jello
content.
We
have
them
the
it's
still
I.
C
Think
if,
like
in
the
course
of
developing
this,
it
would
be
nice
to
have
multimedia
content,
which
is
related
to
each
other
and
I.
Think.
That's
that's
a
possible
way
to
go
because
I
think
the
Jupiter
notebooks
is
really
12
asked
for
younger
kids
could
be
maybe
in
the
last
year's
of
secondary
school
with
them.
That's
all,
whereas
with
a
small
scratch
thing,
there's
even
a
scratch
Junior,
which
is
for
kids
for
4
to
5
years
old.
So
that's
really,
but.
C
Well,
not
inside
because
they
were
actually
that
was
quite
funny.
They
were
actually
like,
also
saying:
how
are
we
done
to
funding
for
these,
and
if
you
create
fountains,
we
will
help
you
put
it
on
our
website
and
getting
it
to
other
people.
So
it's
not
that
they
have
so
much
more
than
you.
They
have
a
whole
set
of
experiments
already
on
their
website,
but
not
so
much
in
scratch.
It's
what
are
they
health
teachers
to
get
to
know
the
tool
and
to
develop
their
own
lesson
plans.
So
are
these
done.
C
Is
something
which
is
developed
by
MIT
and
on
that
website?
You
have
a
lot
of
examples
of
projects
by
people
all
around
the
world.
All
users
can
put
their
stuff
there
and,
for
example,
I
found
something
on
play
technology
like
some
application,
where
you
can
click
and
explain
different
stuff
and
they
kicked
on
mixer
yeah.
But
if
you
know
examples
that
are
interesting
in
some
that
are
they're
all
very
relevant
for
us,
but
you
can
create
anything
with
that.
You
can
make
an
animation
just
a
story.
Yeah.
G
E
Just
posted
a
link
because
the
the
example
that
I
was
thinking
of
when
you
you
mentioned
seismic,
was
the
one
that
we
did
for
ray
tracing
so
basically
showing
reflection,
paths,
refraction
paths
in
the
direct
ray
and
those
might
be
nice
for
us,
just
like
a
single
layer
layer
over
a
half
space
model.
Just
because
it's
it
like
the
physics
is
pretty
simple
and
very
visual.
So
Rowan
has
a
nice
page
where
there's
like
a
whole
bunch
of
like
small
samples,
sort
of
explained.
You
might
be
less
about
race.
C
So
the
good
thing
is
that
they
were
saying:
actually,
teachers
are
asking
for
applications
like
they
often
don't
have
to
the
time
to
develop
a
new
project
play
scratch.
They
then
your
other
use
existing
project.
So
if
we
would
develop
some
button
that
goes
through
the
curriculum
here,
it
will
really
be
useful
for
a
large
community
to
yeah.
O
O
O
O
O
O
O
361
no
money,
the
process
that
we
remove
the
one
being
put
in
there,
our
money.
Here
you
see
there,
we
go
differently.
O
E
E
E
O
After
this
upward
continuation,
we
have
a
rough
estimate
for
the
first
residual
normally
say
between
500
to
1000.
Then
we
calculated
a
power
spectral
density
to
see
the
alignment
of
the
anomalies.
The
symmetrical
aspects
of
all
of
them
are
just
for
one
degree
one
and
say
that
it's
a
symmetric
weather
size
of
animal
is
different
shape
of
anomalies,
and
we
have
two
alignments.
O
The
electricity
of
the
anomalies-
it
is
for
the
degree
one
that
is
not
so
nice,
and
this
is
the
residual
anomaly
for
degree
true,
so
we
use
degree.
This
is
some
local
anomalies
here
more
than
first-degree,
it's
absolutely
different
from
first
degree.
This
is
the
problem,
sometimes
how
or
which
degree
we
should
use
one
two
three
maybe
give
some
different
reasons.
Sweetie,
and
you
see
that
you
see
it
here.
O
O
Section
we
just
looking
for
the
age
of
two
anomalies
as
closely
as
possible.
It
means
that
for
this
part
for
this
red
part,
you
see
for
this
red
part.
You
have
a
sharp
border
here,
or
this
part
also
here
what
is
fun
here,
showing
the
borders
often
on
money,
and
then
we
have
regional
animal,
it's
very
difficult
to
find
some
short
boundaries.
You
say
that
the
edge
detection
is
somehow
some
kind
of
and
safe
or
suitable
for
local
animal
is
not
right.
O
O
We
plotted
these
anomalies
were
on
the
topography.
We
had
magnetic
also
survey
that
we
didn't
to
it,
but
I
just
got
the
data
and
plotted.
We
are
the
first
ballot.
The
first
bullet
is
topography
from
SRTM
that
I
use
just
a
window
here.
This
is
the
topo,
then
we
have
magnetic,
and
the
top
is
their
gravity
first
degree.
We
see
we
see
that
the
location
of
the
animal
is
very
consistent
with
magnetic
this
one
magnetic
and.
O
O
This
is
for
they
agree
to
a
part:
continuation
for
degree
two,
because
in
the
ability
we
have
more
local
animals
in
our
continuation,
100
meters
to
talk
to
thousand
method
and
offered
way.
Considering
these
figures,
we
have
a
rough
estimate
between
say
that
the
market
on
this
is
under
500
meters
that
actually
Oh
nuts.
It's
not
very
deep
anomalies.
O
Again,
the
power
spectral
density
of
degree,
two
is
even
more
symmetrical
than
one
degree
various
sizes,
and
we
just
see
a
very
weak
alignment
in
these
two
directions.
So
we
can
say-
and
you
can
come
come
to
this
conclusion
from
this
PhD-
that
we
do
not
have
a
sharp
or
separated
anomalies
here
and
the
grid
and
the
size
of
the
animal
are
quite
different
from
each
other.
This
power,
spectral
density.
O
Of
course,
we
can
calculate
the
depth
from
this
power
spectral
densities
by
radial
power
spectrum
and,
as
I
tried
to
do
it,
I
didn't
get
such
good
results
for
the
depths
using
power
spectral.
Our
is
the
radiant
power,
spectral
density,
so
yeah.
Now
we
go
to
the
into
the
inversion
step.
This
is
the
tree
mesh
that
we
use
for
D.
Everyone
just
I
wanted
to
use
to
show
you.
O
O
These
are
the
results
of
the
inversion
for
the
degree
one,
and
we
see
that
we
see
some
anomalies
up
to
say:
hi
800,
900
meters,
that
considering
the
height
of
the
topography,
something
less
than
700
800.
Something
like
that.
The
EPS
is
not
too
high,
even
for
the
first
degree
that
we
expected
some
deeper
anomalies.
I
show
here,
six
slices.
O
Using
this
first
sticker,
they
actually
made
to
cross
section.
That
is
not
so
very
nice
because
we
see
the
extension
of
this
positive
and
negative
anomalies
here,
and
it's
not
so
satisfying
in
this
modeling
for
decorate
to
again
we
select
at
some
point
but
still
get
it
free
inside
pie,
and
this
is
for
decorate
to
that
show
this
consequence
of
the
locating
these
anomalies
on
each
other.
It's
absolutely
nice-
and
this
is
these-
are
the
results
for
degree
to
which
a
slices
say:
mushroom
maximum
height
is
less
than
1000.
O
They
say
about
1000
matter
from
this
height
to
this
one
that
is
showing
here
and
again
for
their
great
food.
The
modeling
result
of
not
so
satisfying,
because
you
see
here.
The
extension
of
this
animal
is
not
such
good
results
for
modeling
by
these
animals.
No,
we
go
to
the
second
step.
Second,
approach.
That
is
more
important
for
us
because
from
the
first
three
are
normally
to
the
inversion.
O
Just
we
use
the
Python
numpy
points
intact
without
using
any
other
software
like
Joseph,
or
something
like
that,
so
input
or
input
to
all
cost
is
free
air
on
normally
that
actually,
the
raw
gravity
data
that
made
some
collection
like
say
to
his
collection
and
free
air
correction,
which
can
be
done
in
Excel,
is
very
easy.
Again.
We
generated
the
salty
and
one
oxygen
like
before:
PO
correction,
booger,
anomaly,
residual
damage,
all
of
them
have
been
done
by
using
the
Integra
gravity,
modeling
force
in
synthetic
and
the
other
steps.
O
O
This
is
a
very
free
erin
ovalle
on
trying
correction
extracting
furnace
archaea.
This
fear
anomaly
forward
modeling
and
complete
anomaly.
The
compiler
Pagar
normal
is
one
102
50.
That
is
absolutely
acceptable.
Reasonable
forward,
try
model
training
has
been
done
by
computing.
The
model
from
reference
surface
I
ellipsoid
to
the
ground
surfaced.
So
we
computed
the
whole
gravity
effect
that
this
area
and
subtracted
from
the
free
air
animal
itself
is
reached
to
the
component
to
Ghana
Mali.
O
This
is
the
residual
anomaly
we
use
the
two
teacher
and
integrity
to
by
protocol
skirt
in
Python.
So
we
see
here
is
the
upcam
area
full
of
hot
air.
Spinning
activities
we
have
negative,
as
is
expected.
Nice
here
is
a
village
say
jorge
village.
That
is
also
an
region
full
of
these
activities.
A
lot
of
sproings
are
here.
This
is
a
neo
annum
that
we
detected
here.
It's
not
been
reported
after
now.
Maybe
it
can
be
a
fault.
O
Jokerman,
but
in
this
small
area
is
difficult
to
speak
about
the
sauce.
Just
we
have
some
traces
of
it
here.
This
is
again
the
80
detection.
You
should
see
the
edge
detection
ISM
very
nicely
detected
here.
This
is
the
negative
negative
region,
negative
region.
This
is
the
positive
region
sphere.
The
borders
are,
the
string
latest
can
be
seen
here
also.
O
Again
now
we
have
some
kind
of
shift
between
all
gravity
residual
energy
with
the
magnetic
desert,
but
I
have
some
interpretation,
so
logical
interpretation
for
this
shift,
because
the
depths
of
all
animal
here
is
much
is
say
about
two
thousand
one
thousand
five
hundred
that
the
magnetic
the
magnetic
depth
is
less
than
1000.
So
this
magnetic
field,
this
magnetic
anomaly-
does
not
belong
to
the
source
of
the
jolted.
No,
it
can
be
an
intrusion
or
some
particular
like
that,
but
the
chamber
or
the
say
the
magma
chamber
is
not
here.
O
Some
more
investigations,
and
also
we
will
probe
it
in
our
complex
region
surrounding
this
central
part
for
this.
For
approving
this
anomaly,
the
source
of
the
geothermal.
Here
again,
we
used
the
out
for
continuation
method
for
the
estimating
the
maximum
depth
see
here
about
two
thousand
is
still
the
pictures
are
different,
I,
somehow
area,
so
we
say
that
it
can
be
some
kind
of
say
more
than
one
thousand
is
or
it's
difficult
to
say,
1,500
1,400.
By
of
what
it's
just
a
rough
estimate,
this
is.
O
The
PSD
is
very
interesting
that
this
PSD
power
spectra
to
DB
PST
is
very
symmetric,
smooth
and
vertical
trend
that
actually
was
expected
from
or
George
pal
evidences
or
previous
surveys,
and
we
have
Northwest
salt
is
dominant,
is
for
alignment
and
the
weak
alignment
here
that
this
is
the
direction
that
we
were
waiting
for
for
much
better.
This
Pierce
is
much
better
than
or
former
result
that
I
showed
you.
O
O
The
result
are
much
better,
and
so
we
can
see
a
very
positive
on
Mali
with
a
positive
area
here
in
the
middle,
and
it
continues
up
to
75
health.
It
means
that
about
1000,
500,
600
700
that
it
absolutely
man
is
much
if
deeper
than
they
say
the
one
did
or
two
degree
extracting
from
Joe
soft.
So
somehow
we
are
close
to
find
the
source
of
the
geothermal,
it
sure
be
some
hole
between
two
about
2000
between
five
thousand
five
hundred
to
two
thousand
and
our
modeling
result
is
very
close
to
this
value.
O
I
Vahid
he
said
there
was
like
a
processing
step
where
you
mentioned
each
fending
yeah.
It's
like
a
regional
removal.
Is
that
correct?
Yes,
yes,
okay,
I,
just
kind
of
in
general.
What
are
the
state
of
art
in
in
terms
of
regional
removal?
So
your
it
seems
like
what
you're
doing
you
use
some
sort
of
function
that
you
can
assume
smooth
and
like
basically
fit
that
and
then
subtract
it
is
that
what
you're
doing
yeah
yeah
so
like
is
that
the
state
of
art
and
what's.
O
Yeah
there
are
many
different
methods
to
delete
the
original
effects
and
gravity
one
of
the
most
popular
ones.
It
is
fitting
the
polynomials
one
for
me,
polynomial,
say
one
degree
or
two
DV
or
three
degrees
to
the
b'ougar
anomaly
and
actually
finding
the
best
shape
and
subtracting
from
the
Google
element.
Q
O
Is
this
is
very
popular,
but
there
are
many
others
methods,
for
example,
which
of
our
continuing
and
subtracting
from
Google
anomaly,
with
getting
the
riveted
subtracting
extracting
one
area
from
the
idea
that
dog
and
his
colleague
published
a
paper
in
magnetic
I
tried
it,
but
it
somehow.
You
should
have
a
concentrated
on
normally
to
extract
from
the
great
that
this
is
not
all
case.
We
have
various
scattered
animals
in
different
parts
of
Libya.
If
you
want
to
extract
these
troops,
they're
all
absolutely
extract,
so
we.
O
I
O
I
Is
that,
like
a
it's
always
kind
of
a
little
bit
scary
cuz
that
regional,
sometimes
the
value
of
the
region,
was
really
back
right
compared
to
your
normally
like
it
could
be
right,
sometimes
I'm,
not
sure,
maybe
not
that's
not
always
case.
Suppose,
like
that's
the
case
and
simply
subtracting
it
I
yeah,
I,
wasn't
kind
of
it
was
like
kind
of
always
bugging
me.
Is
that
really,
okay,
or
not
just
a.
O
Common
yeah
I
saw
I,
saw
this
side.
Actually
them
came
back
at
me.
This
link
I
even
downloaded
one
of
130
of
his
picture
by
Charlie
and.
O
I
I
can
send
you
later
I
think
that's
when
we
were
in
Argentina
one
of
the
electro
named
Alfonso
Alfonso.
He
was
like
he's
working
on
a
tectonic
scale,
so
yeah
he
actually
did
the
like
a
earth,
scale,
gravity,
inversion,
I,
think
yeah.
He
was
assuming
like
a
three
layers
or
something
like
that
and
then
inverting
for
actually
the
earth,
so
one
of
their
motivation
was
actually
having
a
model
and
then
that
could
one
of
the
application
they
were
having
in
mind
using
like
some
sort
of
regional
removal
for
like
a
local
scale.
M
O
O
Not
some
kind
of
like
waves
in
ocean
I,
don't
know
I,
don't
know
I
can
I
can
charge
it.
I
just
should
put
more
on
see
the
results
on
that
resummon
taneous
area
on
the
Alborz
and
see.
What's
because
this
area
is
not
a
very
rough
training
area
and
you
see
the
results.
No,
we
should
see
there
on
the
mountains.
How
can
they
be
so
at
the
moment?
I
forgot
it
to
use
it
as
a
train
correction,
because
they
are
much
more
say.
O
In
my
opinion,
more
accurate
results
for
train
corrections,
as
is
global
models.
Maybe
it
can
be,
it
can
use
for
some
very
deep
anomalies
and
very
coarse
grid.
Spacing
Oh
like
this
but
home.
You
know
for
mineral
exploration
or
near
surface
targets,
a
less
than
one
kilometer,
two
kilometers.
We
should
apply
some
more
accurate
methods
for
trying
correction.
M
O
O
O
P
M
M
O
I
showed
you
to
two
different
apples
in
one
approach.
I
used
the
jaws
of
Troy
collection,
the
second
approach.
We,
you
model
the
train
between
reference,
surface
I,
think
I
literally
computed
the
gravity
effect
of
this
bone
from
surface
ground
surface
to
delivery,
and
they
are
all
areas,
so
it
actually
includes
play
cook
anomaly
answering
and
strum
them
both
of
them
in
a
second.
In
your
second
approach,.
P
O
O
O
O
J
F
J
O
O
M
O
C
M
G
J
O
R
K
R
O
J
R
O
M
J
J
H
J
O
M
J
M
You
something
you
never
know
you
know,
what's
their
take
to
take
those
say,
take
those
same
data
locations
same
day,
the
locations
put
in
something
we
thank
you
that
you
don't
so
it's
it's
on
the
topic
of
a
surface
and
do
put
a
block
to
a
terrain
correction,
see
if
you
get
what
you'd
expect
just
from
from
the
block.
So
that's
already
telling
you
something,
and
then
you
just
go
ahead
and
do
those
data
you
go
without
interpolating,
then
you
do
the
same
thing.