>> From the Library of
Congress in Washington, DC.
>> Stephanie Marcus: Welcome,
this is the final lecture
of our 2017 NASA Goddard series
and we already have eight
planned for next year,
which will be our 12th year.
We don't have the date set yet,
but we've got some exciting
things on the horizon.
But today I want to welcome you
and I'm glad that you didn't go
to the Zumba party you came
here to expand your minds.
I'm Stephanie Marcus from
the Science, Technology
and Business Division
and I guess I already
said what I wanted to say.
Anyway, I was thinking
back to how
when I was a little kid we used
to drive around in December
and see all the holiday
light displays.
And a lot of this was before
NASA was even created,
but I never imagined that I'd
be standing next to a scientist
who can see all of the holiday
lights or any surges in lighting
across the globe from
space, so it's really --
we've come an awful long way.
Today our speaker is
Miguel Roman and he was born
in Puerto Rico, he's
from a very large family,
and his uncle was an
electrical engineer there
and that inspired him
to do that instead
of whatever else he might
be able to do on the island.
But when he was still in college
he got a fellowship to NASA
and was assigned to
an earth scientist
and was absolutely hooked
on science and NASA.
So, he's ended up there in 2009,
before that he went to Cornell
and got a master's in systems
engineering and then a PhD
at Boston University
in remote sensing.
His team some of them are
here today, so that's great
to have a support group.
And he will be telling
us how they go
about creating these
maps, you've got a map
out there that's hot off the
press, I guess he picked them
up Monday so it's
the latest images.
And he will tell
us how they do that
and I think it's not all holiday
things because there are a lot
of places where the lights
have dimmed or gone out due
to poverty, war and
hurricanes recently.
So, please help me welcome
Miguel Roman to the Library.
Thank you.
>> Dr. Miguel Roman: The work
that we do at NASA as defined
in the National Space
Act Agreement is
to explore the earth and
understand the earth as a system
and that provides NASA
with a lot of flexibility
to understand a lot of
different components
that gives us a wholistic view
of the earth from the space.
And the idea here is that we
want to use space as a mean
to understand all these
different components, the land,
the ocean, the atmosphere,
the ice,
and how we as humans are
also making an impact
and human activity
is [inaudible] an art
of the earth's system.
The challenge with our system
science is that we always try
to find what happens when
a small disturbance in any
of these subsystems can affect
the entire planet as a whole.
Think of it for instance
what happens
if you know we release
too much CO2
in the atmosphere the
planet begins to heat,
the ocean becomes acidic.
You know these are questions
that are really difficult
to answer with just a few
measurements here and there.
NASA for the last 45
years or so we've tried
to tackle the problem of
earth's system understanding
by doing the following, which
is we launch satellites.
We use these engineering
marbles,
these 5,000-pound school
bus-sized satellites
and in them we carry
these instruments and each
of these instruments have a
purpose, they have a mission
to understand the
different components.
One of them could measure
the earth's radiation budget,
the other one could
measure aerosols,
the other one could
measure images
which you can then
directly relate
to some activity patterns
taking place at the surface.
Our emphasis is to use the
unique vantage point of space
to understand how the
earth is changing.
This image is known
as the blue marble,
today is the 45th anniversary
from the time the Apollo 17
astronaut took this picture
14,000 miles away on their
way to going to the moon,
December 8, 1970
something, sorry.
I can't do that math,
there you go 70.
And so, how do you become
an [inaudible] scientist,
it's not like you know
you grow up in Puerto Rico
and you're saying [inaudible],
I want to be an [inaudible]
scientist, no it's not.
Well to answer that question I
need to take you on a quick tour
over the land where I grew
up, which is Puerto Rico.
As Steph was saying, I coming
from a very, very large family.
This is a picture from our
grandparents' 50th anniversary
back in 1988, the family has
grown exponentially since then.
Now we have 21 cousins
and most of them and a lot
of extended family they
all live in Puerto Rico.
And here we used to go on
trips, we don't have Six Flags
or Disney in Puerto Rico was
have the Arecibo Observatory,
we could go out cruising
around La Parguera here.
This is my here right me
and this is my brother
Jonathan who's now a doctor,
he's an internist in
Bayamon and that's my mom.
And another thing that
you're always being told is
to appreciate nature because
really about all we have
on the island and it's free,
it's just like the Library
of Congress it's free.
And that you also
need to respect nature
because nature has a
way to correct itself
and from the maladies of man.
This is La Parguera and
you can see the docks here,
this is where my mom is
standing right there.
And this is La Parguera
after Maria just 60 days ago.
And I'm going to be talking
about Maria later in my talk,
but I just wanted to give
you a sense of what drove me
to announce that which is that
you know when you witness change
and it impacts your communities
that's quite a telling story.
This is Puerto Rican astronaut
Joe Acaba who's currently
already in the International
Space Station
and took this image
of Puerto Rico,
although it's a little bright
here you can start seeing a lot
of the defoliation particularly
around flooded areas,
all the rivers, all
those areas in brown.
Another thing that we
are really well-known
in Puerto Rico is our
hot blood, our music,
the fact that we see ourselves
as a community not in terms
of material aspects, but
we see ourselves as part
of a larger culture that
contributes a lot to the world
in many different ways.
And that includes things
like music, our food,
the way we celebrate Christmas.
We celebrate Christmas like the
longest Christmas in duration.
Folks here finish up their
holidays on January 1st I was
like what, but where's
the Three Kings festival
and the Octavitas, and the San
Sebastian festival that we go
through up to January 20,
the lights remain
on until January 20.
And what I try to
impress upon you is
that all these things
really matter in terms
of understanding how humanity
interacts with the environment
around us because every time we
do this we're consuming energy.
And so, that kind
of lead to my trying
to really tackle what are the
biggest challenges that we have
at NASA in the scope of
understanding humanity's role
in the earth's system
and that started with,
I'm very data oriented
and you'll see
that very clearly in my slides.
But the idea of tackling the
massive amount of information,
this may sound familiar
and volumes of information
that we got and try to make
sense of it that's the challenge
that we are going through today.
And so, every time there's
an opportunity for us to talk
and say look, at least
here's a snapshot
of what we've learned I
think that's important.
And so, I went to take
you back 45 years ago
of what we have done.
This is an image of
Las Vegas in 1972
when the first land imaging
satellite was launched by NASA.
And for 45 years we've been
collecting images every 16 days
over the entire world
and its land system.
I want you to notice the
variation in water in Lake Mead
because as a city
becomes more intense
and more urbanized you need more
water, you need more resources,
you are replacing the natural
ecosystem with urban buildup.
And you can definitely see the
effects that humanity has had
by just looking at these
long-term measurements
from space.
And that's one of the values
of earth system science
from the vantage view of space
is just that you collect data
for very long periods of
time that is as valuable
as having an ice score and
understanding past warming
because it gives you a really
good picture of what's going on.
Humanity's footprint
in the earth's system can also
be seen in the atmosphere.
This is an image composite
of the concentration
of nitrogen dioxide that's taken
from the ozone mapping
instrument, an instrument
that is well-known for
helping us track the reduction
of ozone depletion
in the South Pole.
We've managed through the mantra
protocol to reduce the ozone,
meaning that humans can actually
alter the earth for the benefit
of future generations.
This is a different type
of gas, this is an ozone,
this is nitrogen dioxide is
a short-lived gas which means
that it quickly combines with
other molecules like water
to form acid rain, to form
the kind of pollutants
that impact air quality
and result
in large cases of
childhood asthma.
EPA has a full monitoring
program and on [inaudible]
and all different
types of molecules.
And what you'll see here is
that the concentration really
trails the weekly pattern
that humans take in terms
of routines within the city.
You can see that by Sunday
when everybody's at church
and they're at Wegman's you know
having brunch with their kids
because you don't
want to have breakfast
and you just go to
Wegman's, sorry.
You see that the
concentration goes down and now
if you did om a weekly basis and
on a recurring basis you begin
to see patterns in the data that
can give you some information
about the extent of
contamination, but also how
that contamination
has directly led
to an actual pattern
of energy use.
As you'll see, this
is the introduction
to what we call the black marble
and the posters that you'll find
at the front these are the first
ever, this is the first release
of those posters and
[inaudible] one of them here.
And you might have seen
these posters already,
we've been generating
them on an annual basis
for almost 20 years now.
The importance -- what I'm
going to do now is I'm going
to quickly go through how
we actually extract and map
of annual nightlights.
So, you get a sense of not just
the massive amount of data,
but have a context of
what you're seeing here,
what does it means that I
see a bright pixel here,
is that my house,
is that a city.
Well let's go through
that quickly.
This is the city of
example in Barcelona
and as you can see here
the entire area is bright.
And if you can catch at least
one little [inaudible] pixel,
then really in terms of
size what we're talking
about is a 700-square
meter area,
so it's roughly a three
block by three block area,
every pixel that you
see is about that size.
That doesn't mean that we can't
detect a small light emission
within that area, in fact
we've conducted experiments
in Puerto Rico and we've
used flashlights like these
and we go all the way to
a very dark spot here,
this is the [inaudible]
farmland site in Cabo Rojo
and we brighten up this
[inaudible] wit 12,000 lumens
of light and we can
see it from space.
So, it means that you
have this what we call
in geography the support,
but inside that support there is
a target and we can detect it,
just the same way we can
detect fires from space.
It doesn't mean that I know
the location of a fire I know
that the fire is inside
that 1-kilometer area.
The holy grail remote-sensing,
the study of satellite images
from space is to be able to
detect the location of that lamp
and the intensity of
that energy output,
that's really the
technical hurdle.
And it's quite a hurdle because
the data that we get directly
from the satellite is
not a pretty picture,
you just don't turn on a camera
and you get lights
directly there are a lot
of things happening
in that image.
This is an actual scan from the
visible imaging radiometer suite
VIIRS which has this
nighttime instrument known
as the day-night band.
And the image is 3,000
kilometers along the scan
direction and this image is a
granule of roughly 5 minutes.
And this is of course
an image of India
and you can see the
border between India
and Pakistan right here.
And roughly about 2.2%
of the earth's landmass
is classified as urban.
That means is that we have using
land sat remember the image
of Las Vegas, if you count
the number of buildings
and infrastructure about roughly
only 2% of the total landmass,
all deserts, you
know trees is urban.
But this is not urbanization
this is energy
and it's showing you settlements
outside of urban areas just
so you know not everybody
lives inside a building
in this planet, some people
live in rural settlements,
some people live in internally
displaced settlements.
And the value of the black
marble as it allows you
to really measure human
presence not directly related
to infrastructure, but
related to humanity.
So, if you take an image like
this, which is one single tile
of one orbit and you combine
5,000 plus orbits you collect
enough data to create the
black marble on an annual basis
and that's roughly 42 terabytes.
If you were doing this
back in the 70's you'd have
to have the entire
holdings in the Library
of Congress times four
just to make this map,
that's how much data
we're talking about.
And the challenge doesn't stop
there, the NASA program wants
to use these measurements
on a daily basis
and at very fine resolution,
that's the challenge.
So how do we go about
doing that?
Well we need to contend with
all the issues surrounding the
collection of images at
night and that's a fact
that there are many, many,
many different other sources
of light emissions that
also impact our ability
to estimate lights
coming from cities.
And if you're an atmospheric
scientist or you want
to study the [inaudible] this
is really, really awesome
because now this is a new source
so that you can detect
the aurora,
you can measure concentration
of particles like nightglow.
But if you're a scientist just
studying human settlements all
this is noise.
And so, I'm going to go
through a few of these.
The biggest source
of contamination
in nighttime data is the moon.
You can see here that
the moon can reflect
from the earth's surface and
the surface is very variable
like in this case desert is very
stable, so it stays very stable.
You can constrain the amount
of light coming from the moon
versus light coming
from cities very easily.
But as soon as you go to another
place like here, this is,
you know, this is vegetation
around the [inaudible]
of Africa that is changing.
So, there are many
dynamics, there's a confluence
of issues associated with
estimating nightlights
from space being that we have
multiple emission sources,
we have the moon and we have
other sources as I'll show you.
We have snow, snow is
very bright and if you go,
like if you look at cities
like Toronto or Montreal
and Quebec it looks as
bigger as New York City,
but it's not really, the extent
is not real it's just a physical
property known as snow
multiple scattering.
And by the way, I'm putting all
these books here describing the
principles behind how we map all
these different characteristics
from space.
And it's very important for us
to characterize snow because 25%
of human settlements
exhibit some kind
of snowfall throughout the year.
And so, if you really
want to get a clean map
on a daily basis you've
got to measure the snow.
You also have to measure this
thing known as the atmosphere
which is in the middle
between you and the city.
And especially if the atmosphere
becomes disturbed as a result
of natural variations
like fires.
So, these are fires taking
place in Siberia as measured
from the mode sensor
on [inaudible]
and you can see them
in here red dots.
And then here fires at night,
it's almost like we
have an x-ray vision.
And we can actually see
the burn scar of the fires.
And of course, fires are in
places where people live,
so you have to also
separate fires
between cities and other areas.
Here's another emission
source, airglow and Aurora.
Let's see if this one works.
The last issue I want to
talk about is vegetation.
If you study cities you have
to study vegetation not only
because lights have to
penetrate vegetation in order
to be detected, but also
vegetation is important.
It helps mediate ecosystem
services and function
and unless you're like you known
hanging out with [inaudible]
on the First Moon of Endor
most human settlements are
in the ground.
And so, we have to
develop models
that describe how light passes
through a forest canopy.
In other words, it's important
to study urban infrastructure,
but you also have to study
forest urban structure.
All right, so you
do all these things
and the takeaway being you have
to combine all these
measurements,
it's not enough just to take
nighttime images you need
to understand what's
going on during daytime
so that you can characterize
the vegetation,
[inaudible] of activity,
etcetera.
And you can do things like
this, you see the raw image,
this is Paris at night and
this is a French countryside,
you can detect where
the clouds are,
where the aerosol is,
where the snow is.
This is the alps,
this is [inaudible],
this is how many
reflections we are to expect.
And these areas in
green here are areas
where we're seeing foliage
and there's [inaudible]
below the foliage.
And what happens is
basically if you don't correct
for this the entire French
countryside disappears during
the summer.
And I know part of that is
because French people take
longer vacations than Americans,
but it's also because
there's vegetation structures
that you need to account for.
All right, so here's the final
image and so there you go.
From a daytime image raw to
a daytime black marble image.
So, the [inaudible] you see
in the front are
annual, this is daytime.
Now I can show you all
the lights from space
because now we can, we
break through that issue.
And so, here's an
image of the DC area
and now we have these daily
images we can start slicing them
by season.
And what you see here
is a seasonal change
in nighttime light,
so that means
that we're comparing
the amount of lighting
that we see during the
holiday period relative
to the rest of the year.
Anywhere you see green you're
seeing an increasing lighting
of roughly 20 to 30%.
Notice that the entire
area around congress is red
because they're out of session,
you know hopefully they pass
their tax plan before Christmas,
they'll also be able
to go back [inaudible].
But notice that pretty
much all of Fairfax County
or Arlington County, all
Howard County in green.
And everywhere you see
green there you see increase
from the lights from space.
And so, and I hope that
I've impressed upon you
and I know it became a little
you known technical at the front
that I showed you that
lamp in Puerto Rico,
but I hope that you believe me
that if we can detect a lamp
like that that we can
detect Christmas lights.
But the question is
what does it mean.
Here's Baltimore area
you'll notice that unlike DC
which has a much more
commercial and federal workforce
that disappears during a holiday
you know we have a lot more
green inside the densely
populated residential areas.
There's also an issue of space,
if in the suburbs
people have larger areas
to provision lights, whereas
in the more densely popular
areas the amount of light
that you can put is just so much
because there's not
enough space.
And so, the reason why
this is important it's
because of the topic of energy.
These measurements are allowing
us to map cities not just
as a function of their
size, their density,
but as a function
of energy behaviors.
Behaviors that are
driven by a status,
behaviors that are driven
by the values of a culture,
we can see culture from space.
And in the case of the United
States we can see Christmas
lights everywhere, meaning that
whether you're rich or poor,
religious or not everybody
celebrates the holidays.
And what we know is that
the process is short-term
and it's shifted,
meaning that people leave,
people leave their offices
and they go and stay
in their homes with
their families.
So, there's an energy increase,
but it's also shifts
in location.
And when we did research
across other cultures we found
that the energy signature
is very different
from that at Christmas.
Like if we go here to
this is the Middle East,
this is near Mecca and I'm going
to concentrate on this rock
from here, you can see here.
And again, we can measure
small pixels on a daily basis
so you can get images
like these.
So, this is raw data
monthly and daily products
and they're all following
a similar trend,
but notice the reoccurring
pattern of the holiday.
And of course, the holiday
that we're seeing here is not
Christmas but it's the Hajj.
This is Mount Arafat
and in Mount Arafat Muslims
come together roughly about,
up to 10 million pilgrims
come and pray in Mount Arafat
and they stay here in these
camps, this is known as the city
of air-conditioned tents,
this is the city of Mina.
And every one of these tents
has an air-conditioned unit
that is not your grandpa's way
of doing you know the Hajj,
this is the new way
of doing the Hajjaj.
Hag. . And we also the
same effect during Ramadan,
but remember the change in
energy use is different,
people are not moving during
these periods, they are staying
where they are and
they're moving
and they're shifting all
their energy use at night.
And that's a different
sociocultural construct
because what we've learned
is that there's a portion
of the energy that
humans use that is related
to individual behavior
and decision-making.
What kind of car do you
drive, how big your house is,
you know what you
define as comfort,
is it a 2000 square foot
house or is it 4,000,
but there's also a portion
of the energy that we use
that is related to the
energy required to live
within a certain
cultural context.
When and how much people
use energy is dictated
by the society in which we live
and the routines
of that society.
And this is really
important in the context
of global energy policy
because there's a lot of savings
that we can gain not just by
looking at individual incentives
like how much tax I'm
going to put on carbon
or you know how much I'm
going to tax for an SUV,
but looking at community
wide energy use.
All right, so this is another
study where we conducted,
where we looked at
the clustering
of different voting districts
in Egypt during Ramadan.
And as you can see here,
this is before Ramadan,
during Ramadan it peaks
during the [inaudible],
the end of Ramadan
and again goes down.
The difference between
these three groupings is
that these are rich people,
these are moderate you
know income people,
and then these are
poor districts.
So even within the same religion
energy is being used differently
because some people have more
access to energy and services,
this is an issue of inequality.
This is also an issue of
religious backing because folks
that are rich also
tend to be less pious,
you know they don't follow
the full teachings of Islam.
But what is common
across all of them is
that everybody celebrates,
everybody must and is obligated
to celebrate during
the [inaudible].,
So, this is showing us
that we're really looking
at cultural boundaries even
within the same culture.
So that pretty well goes into
the holiday light feature
and the last, thig
was two years ago.
The last two years we've
been exploring other methods
in which we can use
these measurements
to also study humanity in new
ways, especially in the context
of the United Nations
agenda of 2030
for sustainable development
goals.
And I'm going to go through
three different examples
where we are using these
measurements to assess some
of these sustainable development
goals, as we know them SDGs.
So, this is goal seven, which is
to ensure access to affordable,
reliable, sustainable and
modern energy for all,
which is bring electricity to
Africa, that's what it means.
And what it means is also
that by bringing electricity
we're also helping address some
of the issues that we have
in terms of air quality,
in terms of energy access
for children who can't study
at night because they don't
have access to electricity,
but especially on how we examine
routines for shifting energies
to a more sustainable fashion.
Because the way cooking is done
in Africa really often times
relies on really dirty sources
of fuel like wood and carbon
and dung and you know byproducts
from farmland and of course
that causes [inaudible]
quality issues.
And so, the point of the
goals is to be able to tackle
and trace progress along
a number of indicators
and in this case
looking at night lights
to address how that
is being done.
And so, here's the other issue
with energy is also we're trying
to figure out how to
electrify an entire continent
and we haven't yet to
decide what's the proper mix
between a centralized electrical
grid where power is coming
in from a large turbine
and into small sectors.
But in the case of
Africa powering
that last mile is really
hard and you really have
to be off the grid and
use distributed systems.
And so, here's an example of
how we can do, this is the city
of Korhogo in Ivory
Coast and you can see
that on a daily basis
we know the exact timing
of where the electrification
took place.
And we compare these with
records from the World Bank
and the energy ministries
and we managed
to find objectively
whether the programs
that are electrifying these
communities are working.
If you're spending $400 million
a year on electrification most
of it hopefully will
go to address the needs
of these communities, some
of them will be wasted.
And it's going to be in the
interest of everybody to ensure
that we can effectively
assess progress.
The same goes with what
happens when you're not looking
at sustainability, but
also looking at resilience.
Building a power
system that lasts
through extreme weather
events and that's the case
of Puerto Rico where we've
had, we've released maps
that also show us the impacts
immediately following Maria.
So, these are maps that we
released and we've been working
with FEMA, the Army National
Guard and the Coast Guard,
we show the aftermath of Maria
and you'll see the slider coming
in now and this was
immediately three days after.
And because we have these
long-term measurements we can't,
we don't stop just at three days
after we can keep looking
further and further
into time 13 days
after, 17 days after.
And because we're looking
at lighting we're looking
at a different metric compared
to what is being reported
by the government, which is
total power amount that's coming
in from the transmission lines,
which does not give
you any information
about who actually has
electricity right now.
And so, what we've been able to
do is to study outdoor lighting
versus the percent number
of houses with electricity
which again, it's only
the count of total amount
of megawatt hours that
are being produced,
whether they are making it back
to the endpoint we don't know.
And what we argue
is that the residual
between these tell
you those folks
who have diesel powered
generated
because they're off the grid
and so you can start looking
at energy access in a much
more broader wholistic way,
not just looking at one system,
but looking at the
entire system as a whole.
The other thing that we also
can trace is inequality,
especially in the context
of the modernization
of the energy sector globally.
Even after let's say like we
decide to spend $1 trillion
in solar panels tomorrow we're
still going to have issues
because we still have a path
dependence of communities
which for ages, for generations
have only done this, mined coal
and you don't have to go to
China you can just go two hours
from here and you know
what I'm talking about.
And the issue there
is that what happens
when there's a policy enacted
to lay off 1.2 million
Chinese workers in the district
of Shenmu, well this
is what happens.
It looks like a disaster, it
looks just like Puerto Rico.
And there wasn't a disaster,
the disaster really
is an economic trend
of the urbanization
associated with a sector
that has been wiped
out from the map.
And humanity has gone
through these upswings many,
many different times.
We didn't you know evolve from
the Stone Age because we ran
out of rocks we evolved
because --
I know, it's because
we evolved from coal
to more modernized means.
So, we have to see this
as a positive but also,
we need to see the energy
policy in a community context
because there's people who
are being impacted as a result
of these changes
in modernization.
And finally, the other
thing I want to talk
about is what happens you get
1.2 million coalminers showing
up in all your cities looking
for jobs, that's what's going
to happen, massive
migration of workers.
And there is one place
where that happens
and then suddenly
you get [inaudible],
which in disaster
reduction we call, you know,
you want to prevent
regional instability
because that is going to
affect the central cities
that drive the economy
of the country.
And if there's one place
where that has already
happened it's Syria.
The issue of Syria has a lot,
I mean it's a very complicated
story, but the main issue
that we see is that when the
energy sector when bust in 2008
and all the fields, all
the winter wheat fields
that were dependent on water
and pumps started to crash,
you started seeing all these
farmers coming into the cities
and that just created the right
recipe for ISIS to take over.
And so, here's another,
we have video that we did
with our colleagues at the BBC.
[ Music ]
[ Foreign Language ]
I'll show you a snapshot of
this one district in Aleppo
and this is how it looks.
And I've shown this to
economists and it's like you're
like monitoring the
EKG of a city,
that's really what you're
doing you're monitoring,
I mean the buildings are
still there and destroyed.
But if the people aren't
there the next question is
where are they going and
when are they coming back.
And I can tell you where some
of them are going, they're going
around different countries
and they're setting up shops
in all these internally
displaced settlements.
This is the largest
settlement in Jordan,
the Zaatari Refugee Camp and
as you can see right in 2012,
you know this town of
20,000 people needed
to accommodate 200,000 refugees.
And these are really tough
camps, there's no electricity
and it's just tents,
thousands of them.
And we can see that in the data
because all these swings are
related to weather patterns
where there's flooding,
constant flooding taking place
and the relief workers
have a hard time bringing
in more people because they have
to you know clean
up and bring back.
The United Arab Emirates
contributed a new camp just 40
kilometers away it's
called the Emirati camp
and you can see these
are in containers,
but they can only
accommodate up to 5,000 people.
This is how it looks
from Google Earth.
And of course, people are
coming back, this is the town
of Jarabulus which was
taken over by El Shabab
and the free Syrian
army was backed
by the Turkish government
to take over in 2016.
And this is what happens
when 20,000 people come back,
so you can see this is the
Euphrates, so it's a pretty --
it's right in the
corner where [inaudible],
an area where ISIS cannot
really stay for a long time.
All right, so this
is where we are
with the black marble right now,
we have a massive amount of data
and it's giving us
really rich information
about the human condition.
And it's helping us define our
communities not just as places
that you put on a
map, but as a process.
And it's allowing us
to study the confluence
of the many different
challenges that humanity faces.
When you look at climate change
it's not just about climate it's
about urbanization, urban
development, it's about energy.
And to be able to ensure that
every community has a chance
to improve their cities
in one way or another,
no city in this country is
sustainable, is resilient,
is equitable and safe.
They may be three out of
four, they may be three,
Vancouver is very pretty, but
it's still not sustainable
because a lot of the materials
that they bring they bring
from China, so they need
to account for materials
and everything like that.
So, what's NASA's role?
Well we don't save
lives, if I wanted to do
that I would be joining the army
[inaudible] in Puerto Rico now.
Our job is to provide everyone
with the data and the tools
that they need in order
to tackle these problems,
in order to push the
technology in a way
that we can really turn this
into societal benefit
for everyone.
And that's it, that's my
talk, thank you so much.
[ Applause ]
>> Stephanie Marcus: Thank you,
we can have our questions now
and if you would
repeat the question
so everybody can hear it.
>> Dr. Miguel Roman: Yes.
[ Inaudible Comment ]
Yeah, go ahead [inaudible].
>> All right, so now that
you have the data created
or you're creating this
data who do you see
as being the major
consumers of this data?
>> Dr. Miguel Roman:
There are three consumers
that we're targeting and we're
already collaborating with.
One of them are economists and
demographers who are trying
to study, who are trying to come
up with economics statistics
in areas where there's
absolutely no data.
Think, I mean we have so
much data on like labor rates
and everything in the US
that adding lights it
doesn't add more value,
it doesn't improve
your statistics.
But when you go into a place
like Ivory Coast, you know,
having this information
really adds value in terms
of understanding trends,
understanding demographic
changes, and doing it
in a way that's new
not like doing
like a census map every 10
years, but monitoring humanity
in real time using social media
and using a combination
of datasets.
The second area is
disaster response,
FEMA already is obsessed
with this data.
And they see it as a new
changing game technology nothing
that they are using
operationally yet, but they want
to get this in a more mature way
so that they can get
measurements from anywhere
in the United States, anywhere
in the world within three hours,
so that's a separate community.
And the third community of
course is the energy community.
We want to be able to trust
and verify any efforts
to restore power let's say
in Puerto Rico and to be able
to say wall there's
only three, you know,
75% of people should have
power, so it should [inaudible],
so there's half actually.
And to be able to
provide an objective means
to track progress is
also very powerful.
So those are the
three main ones.
>> My question was if
you could please speak
to your team how big it is and
the different kind of areas
of expertise for
example on your team?
>> Dr. Miguel Roman: All
right, so let me introduce you
to my team, [inaudible] can you,
that's our team,
it's him and me.
So, I'm the guy that
does all the advertising
and he's the mastermind behind
the black marble he does all
the coding.
And there's Jessica [inaudible]
who's our coordinator
for disaster.
So, we also engage with all
different parts of members
of the headquarters program
across different missions
and we're hopefully
recruiting a new programmer,
that's it that's our team.
We have to do the science,
the data crunching,
and we need to tell you a
story of why it matters.
And we're lean and mean at NASA
because it's very competitive.
All right, any other questions?
>> Will this data
ever get compared
with you know the other searches
for setting intelligent life?
What you see in terms of
lights here ever be analyzed
in a way that might be used?
>> Dr. Miguel Roman: So, to
analyze let's say there's a --
let's ponder if there's
going to be like a UFO
and it's just going around
at night, can we detect that?
Probably not because if
it's moving too fast,
we actually had an
inactivation [inaudible] to see,
to answer the question,
can you capture MH370,
the Malaysian airliner
that was lost and was found
in [inaudible] and
we were like honing
over the entire Indian Ocean
trying to find the smallest area
and we found a tiny little spot
it was like this must be it.
But then our colleagues
have now confirmed
that it was the condensation
of the trail of a meteor,
I was like oh okay well at least
we can see a meteor that's fun.
We can also see other satellites
going through because we're
up at 833 kilometers, so once
in a while there's a satellite
going behind us, on top of us
and we're seeing a
satellite and it's like get
out of the way I'm
measuring the earth.
So, we can see things moving,
well if that's alien
life I don't know.
But yeah, we -- it was,
this sensor was not designed
to measure holiday
lights from space,
it's just that American
engineering is that good
that we managed to get
that much sensitivity.
But the program didn't
design this
to do this you know that's how
NASA works it's like whoops,
oh yeah you can measure
[inaudible] okay cool.
But it's really, this is
a [inaudible] instrument
that was designed to measure
extreme weather events in Alaska
because we have geosynchronous
satellites and Alaska's right
in the corner and they
don't have really good maps,
so we came up with this night
vision map so that we can look
at you know sea ice and
all these other aspects.
And by accident NASA is also
having fun with the data
and doing all this science,
but that's how it works.
[ Inaudible Comment ]
So, if you see a lightning
strike we basically say we
couldn't see anything that day
because a lightning strike would
saturate that image to the point
that it cannot be corrected.
However, if it's aerosol we
can can use radiated transfer
techniques, we can study,
looked at book of aerosols
and study how much light
interacts with aerosols
and we can really pin down the
uncertainty and the error there.
If it's clouds probably
we're going
to say we didn't detect
anything that day.
And so, what we do is we go back
to the clearest best data
possible and we report
as this is not today, but
this is what we got yesterday.
If that works for you
know for you that's fine.
But if you're detecting
[inaudible] you want
to have a clear shot, but that's
really hard over the Caribbean
where there's so many clouds.
So, you have to really
build your confidence
and that's the uniqueness of
having a polar orbiting sensor
where you're visiting
Puerto Rico every night
and you can really build a large
record over time to really ramp
down the uncertainties.
So, depending on the
phenomenon sometimes we give up
and sometimes we say actually
we can, let's use these models
to then describe what is
actually being seen that's only
human driven.
[ Inaudible Comment ]
Yes.
[ Inaudible Comment ]
Yes, roughly 40 to 60% of
light will actually pass
through aerosols
depending on the sphericity
and the size distribution
of that aerosol.
It could be dust from the
Sahara, it could be fire
and smoke and because we
have daytime measurements
that really accurately measure
those particles we make some
assumptions of what happens
between the day and night,
about the photochemistry
because everything is changing
at night when the sun is out.
And [inaudible] and
I have built models
to really constrain
the errors there.
It's a big problem and
we're really starting
to really explore aerosol
retrievals at night,
so yeah that's for like
the next generation.
>> Does the Department of
Defense have any interest
in looking at say North
Korea [inaudible]?
>> Dr. Miguel Roman:
The Department
of Defense likely has
an even much broader
and more refined
measurement than these
where they can even get an
even finer resolution directly.
I think the Air Force and
Weather Agency also has a lot
of interest because this
phenomenon where you have
to secure an accurate
forecast when troops are going
through oceans and everything.
So yeah there -- and in fact
the first-generation nighttime
sensor was from the DOD, so
this is a DOD heritage sensor.
And I think you had one more.
>> In terms of energy
consumption,
will there come a time when
the increasing use of LEDs
or other much more
energy efficient sources
of light impact on how you
have to deal with this data
and you know in the near future
is that like [inaudible]?
>> Dr. Miguel Roman: So, there's
two issues with LEDs and one
of them is that the eyes of the
sensor that measures the earth
at night they can't see blue and
LEDs have a lot of blue on them.
You may not see it,
but they combine a lot
of blue [inaudible] because
then you'll saturate.
And so, if you look
at the wavelength
in the [inaudible] spectrum
blue is high at 480 nanometers,
this sensor's shortest
wavelength is 500
which is green, meaning
that we can't detect LEDs.
But LEDs also have a lot of
information in the green,
in the red, but most -- there's
a huge [inaudible] in the blue
that we cannot detect.
So, meaning that
from year-to-year
if someone switches lights using
this sensor we cannot detect an
LED change.
What we can detect
is improved methods
for abatement of
light pollution.
So, if you use a full cutoff
light instead of putting
like contaminating the night
sky which you know messes
up all the astronomy and
messes up like bird habitats,
then we can detect that change
because now you're concentrating
the light to the surface
and so you're going to
see a downward trend.
But unless you give me $200
million and I put another sensor
in the international
space station,
which we could do we
can't detect the LEDs,
it's a really small signal.
[ Inaudible Comment ]
In fact, yes in fact, if you
use the only and the records
from the early 1990's
we're already seeing
which is good news that
an overall negative trend
in short-lived gases
across the United States,
especially in California a lot
of smog has finally started
to ramp down, but it's also
an issue of mountain metrology
and all these other
issues associated with it.
And so yeah, we can
detect long-term trends
and as we add more
instruments that measure ozone,
that measure plumes we can also
start characterizing aerosols
for very long periods of time.
The value that nighttime images
provide in that context is
that we can measure
optical properties at night,
which no other sensor can do.
And oftentimes you just rely on
the thermal data to detect fires
and clouds, but finally we can
also detect optical properties
and that means that we can
build a more complete record
because oftentimes
you get clouds
and then suddenly you need
to look at the interaction
between clouds and aerosols
and that becomes difficult.
And so, and this is very
early work that's started
at NASA [inaudible].
And usually you need
like a 30, 40-year record
to really start seeing a trend
and this is just five years,
so I'll make that clear.
But you can start seeing
a lot of typologies,
you start seeing cities
grow and everything.
All right.
>> Stephanie Marcus: Thank you
and thank you all for coming.
>> This has been a
presentation of the Library
of Congress, visit
us at LOC.gov.