►
From YouTube: The value of weather observations - Stavros Keppas
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
Hi
all
thank
you
very
much
for
having
us
in
this
beautiful
and
quite
inspiring.
I
would
say
event.
My
name
is
tavros
campos
and
I
am
the
lead
meteorologist
of
weather
exam.
A
A
So
yeah,
I'm
starting
my
presentation
with
this
simple
or
not
question,
which
is
what,
if
we
had
fewer
weather
observations.
What
would
happen
if
we
had
pure
weather
observations
on
on
earth's
surface
and
in
the
atmosphere
and
I'm
gonna
answer
this
question
using
a
real
life
and
up-to-date
example.
A
So
in
early
2020,
due
to
the
pandemic,
there
were
there
were
lots
of
restrictions.
We
experienced
lots
of
this
a
lot
of
restrictions
and
one
of
these
was
the
reduction
in
the
number
of
daily
flights
by
50
to
75
on
a
global
scale,
and
so
you
can
see
this
sudden
drop
actually
in
in
the
number
of
the
flights
in
march
and
april,
2020.
A
This
is
actually
another
graph
that
shows
the
number
of
flights
on
the
same
day
in
2019
in
april
and
2020,
we
can
see
this
huge
reduction
of
num
of
the
number
of
the
flights
and
what?
Why
am
I
saying
this?
How
that
how
this
is
related
to
weather?
A
So
these
lack
of
observations
reduce
the
six
hour
forecasting
ability
by
up
to
15
percent
and,
of
course,
this
this
fact
triggered
further
resets,
and
it
was
found
that
if
we
removed
aircraft
observations
by
100,
so
we
didn't
have
any
any
observations
by
aircrafts.
This
would
decrease
the
forecast
accuracy
by
up
to
50
percent,
especially
during
the
winter.
A
So,
okay,
we
talked
about
weather
observations
and
how
they
are
used
in
the
forecast
process,
but
this
is
only
one
or
one
of
the
aspects
we
use
weather
observations
in
order
to
build
also
a
a
climatic
time
series
of
data.
So
what
is
the
difference,
though,
of
between
weather
and
climate?
This
is
another
question.
So
let's
clarify
this.
A
With
the
word
weather,
we
describe
the
current
situation
of
the
atmosphere.
We
describe
the
weather
conditions
that
we
have
right
now
or
in
a
short
period
of
time,
so
we
can
say,
for
example,
that
the
weather
outside
now
is
partly
cloudy.
The
temperature
is
around
24
degrees
celsius
and
the
relative
humidity
is
about
50
percent,
or
we
can
say
also
that
the
weather
during
the
last
hours,
the
last
the
days
or
even
weeks,
was
mostly
rainy
and
the
temperature
was
around
the
climatic
levels.
A
So
we
need
to
know
the
climate
of
the
region
and
in
order
to
know
the
climate
of
region,
we
need
to
have
a
large
archive
of
data,
so
we
need
weather
observations
for
a
long
period
of
time
in
order
to
be
able
to
say
that,
in
a
certain
period
of
time,
at
a
certain
place,
the
average
weather
is
around.
That
is
hot
and
dry.
A
A
It
is
an
automatic
station
and
the
advantage
here
is
that,
firstly,
there
is
no
need
for
actually
we
need
limited
supervision,
so
there's
no
need
someone
to
go
there
and
manually
record
the
the
weather
observations
every
hour
or
every
minute,
and
the
second
thing
is.
The
second
advantage
is
that
we
don't
need
a
large
area
of
deployment.
So
it's
just
a
metallic
mask
and
we
place
on
the
top
this
set
of
sensors.
A
A
Okay,
I'm
a
weather
geek.
So
I
really
not.
I
really
need
to
know
what
is
the
weather
outside
my
house
just
because,
because
I
want
to
know
that
if
there
is
a
thunderstorm
around,
I
want
to
go
and
chase
it.
But
this
is
just
a
small
proportion
of
people
so
yeah
there
are
people
like
me
that
are
interested
in
this
in
weather,
but
there
are
people
like
you
that
want
to
invest
and
this
investment
is
affected
by
weather
and
climate.
A
So
these
are
some
some
of
the
industries
and
sections
of
economy
that
are
interested
and
are
affected
by
weather
climate.
So
agriculture,
energy
transportation,
which
includes
aviation
and
shipping
sports
sport
activities
insurance,
which
is
an
industry
that
is
affected,
is,
is
really
interested
in
the
in
in
the
risk
of
the
extreme
weather
in
the
future,
for
example,
tourism,
which
is
a
quite
important
sector
of
economy
for
some
countries
and
property
management.
A
So
there
are
people
that
they
need
to
know
remotely
if
their
property
is
in
danger
because
of
extreme
weather
events,
and
of
course
the
list
goes
on
and
on
and
on,
and
there
are
even
small
skill
activities
that
we
do
every
day
in
our
life
and
they
are
affected
by
weather.
So
we
need
to
know
what
is
the
weather
now
and
what
the
weather
will
be
in
the
future.
A
A
For
example,
we
need
to
know
the
climate
of
an
area
in
order
to
plant
a
vineyard,
for
example.
So
we
need
to
know
if
the
the
climate
is
ideal,
for
example,
or
we
need
to
know
if
this
is
the
ideal
place
to
construct
a
solar
farm,
and
then
we
need
weather
the
current
weather
and
the
weather
in
the
future
to
protect,
to
increase
the
functionality
and
to
optimize.
Finally,
eventually
our
our
activities,
our
our
ventures.
A
A
These
are
the
words
of
a
meteorologist
wilhelm
bjerknes
in
1904,
but
he
said
he
said
that
later
atmospheric
conditions
are
linked
with
previous
ones
through
the
physical
laws
so
have
the
past.
We
have
the
current
situation
or
the
future,
and
there
are
some
equations
somewhere
here
that
make
simulations
about
the
all
these
atmospherical
processes,
other
processes
that
occur
in
the
atmosphere.
So
in
order
to
forecast
what
is
what
are
the
two,
the
two
keys,
the
two
key
ingredients
for
a
successful
forecast.
The
first
one
is
the
accurate
knowledge
of
the
physical
laws.
A
So
we
need
to
know
what
are
all
these
processes
in
terms
of
math
and
physics
and
that
occur
in
the
atmosphere.
Okay,
all
these
processes
are
quite
imperfect,
imperfect
right
now
and,
of
course,
there's
a
lot
of
research
on
this
and
will
be
in
the
future
as
well.
But
the
other
thing
that
we
don't
have
is
the
accurate
knowledge
of
the
current
state
of
the
atmosphere.
A
So
we
need
to
know
what
is
the
weather
now
in
order
to
feed
the
models
which
will
calculate
the
difficult,
the
sorry,
the
different
processes
and
finally
will
produce
weather
forecasts,
and
I
will
try
to
explain
simple
way
in
simple
words:
what
is
the
process
in
order
to
produce
forecast
weather
forecast?
A
So
the
first
step
is
the
data
simulation
is
really
important.
This
this
step
is
really
important,
so
we
actually
do
a
combination
of
observations,
weather
observations
and
the
latest
produced
forecast.
So
we
have
a
simulation
on
the
one
hand,
and
we
have
weather
observations
on
the
other
hand
and
we
try
to
combine
them
in
order
to
understand
the
in
order
to
get
the
best
representation
of
the
current
state
of
the
atmosphere.
A
A
So
after
we
produce
the
the
current
state
of
the
atmosphere,
this
good
representation-
and
we
have
this
data
set.
We
feed
the
models,
so
a
model
actually
divides
the
the
earth
surface
in
the
atmosphere
into
cells
with
a
certain
size,
and
this
size
is,
is
determined
by
the
spatial
resolution
of
a
model.
So
if
we
say,
for
example,
that
the
spatial
resolution
of
a
model
is
10
kilometers,
that
means
that
the
this
cell,
every
single
cell,
on
the
surface
and
in
the
atmosphere
has
dimensions.
A
It
is
actually
a
cube
with
dimensions
of
10
kilometers.
So
we
feed
every
single
cell
with
the
product
of
data
simulation
and
then
the
model
simulates
all
these
physical
processes
in
the
atmosphere.
In
order,
finally,
to
produce
the
forecast
to
produce
to
show
us
how
all
the
meteorological
variables
change
in
the
future
in
time
and
in
space-
and
this
is
the
forecast
actually.
A
So,
okay,
we've
got
a
model,
we
simulate
this
the
situation
and
we
go
forecast,
but
is
that
enough?
So
now
do
we
know
the
weather
in
the
future.
The
answer
is
probably
no
so
different
models
use
different
approaches,
but
this
is
not
all
the
only
case
here.
We
can
see.
Another
example
is
actually
another
process
that
we
follow.
Sometimes
it's
the
assemble
forecasting,
so
we
have
multiple
models,
sometimes,
as
in
this
case,
with
the
same
approach,
let's
say
scientifically,
and
what
we
change
is
the
initial
conditions.
A
A
We
can
see
that
after
the
third
day
increases-
and
finally,
we
can
see
that
this
is
actually
magnified
in
a
chaotic
way
in
the
future.
So
we
can
see
that
10
days
after
we
would
expect
from
a
real
winter
to
something
like
spring.
A
A
So
we
try
to
develop
a
global
weather
station
network
in
order
to
get
all
this
data
to
get
well
distributed
data
across
the
globe
and
to
get
good
quality
data,
and
in
order
to
do
this,
we
we
use
the
web3
technology
in
order
to
get
to
to
give
an
attractive
incentive
to
people
to
deploy
and
also
to
maintain
their
weather
station.
This
is
really
important.
A
Appropriately,
so
we
designed
the
hardware,
the
software
in
the
infrastructure
on
web3
technology,
and
we
will
use
our
token
our
company's
token,
in
order
to
reward
stations
station
owners
for
the
good
quality
data
that
they
will
share
with
us,
and
so
we
develop
all
these
mechanisms
to
ensure
the
quality
of
data
provided
by
by
the
users,
and
so
we
in
order
to
do
this,
we
cleanse
we
supervise
and
we
reward
not
only
the
collection
but
also
the
distribution
of
the
weather
data
in
the
the
spatial
distribution
of
the
weather
data
in
order
to
make
finally
better
weather
related
decisions
and
and
of
course,
to
provide
a
better
forecast
for
each
region
in
the
world.
A
So
closing
this
presentation,
I
would
like
to
say
that
weather
exam
is
here
to
help
the
planet
make
a
better
weather,
related
decision,
better
weather
related
decisions,
because
we
actually
work
on
one
of
the
key
ingredients
for
a
successful
forecast,
which
is
the
the
weather
that
we
measure
collect
and
share,
and
we
do
this
in
a
fair
way
using
the
web3
technology.
Thank
you
very
much
for
for
for
for
your
attention.
Please
join
us.
We
are,
this
project
is
quite
new,
and
so
we
are
open
to
discussions.
A
We
need
to
find
solutions
to
our
problems
and
we're
open
to
discuss
this
with
you.
There
are
also
of.