Satellites for everyone
[Music]
as most of us know
our planet is going through a lot of
changes
climate change biodiversity declines
polluted oceans intensive land use and
so on
but earth is a big place
and there is one tool at our disposal
that provides us with the big picture
about the health of our planet
and that tool is earth observing
satellites
and we’re in a revolution of sorts when
it comes to earth observation that’s
been a long time in the making
in 1957 when the soviet union launched
sputnik 1
the first artificial satellite it opened
up
the era of earth observation
and as you would expect from the height
of the cold war
a lot of these satellites had secret
military objectives
and their data were classified and not
available to the public
in 1972 nasa launched landsat 1
which was the first civilian satellite
to provide
global coverage of the earth’s land
surface
since then satellites have really
transformed
how we see and what we know about our
planet
in particular satellites have
really shown us how our planet is
changing
i think we’re living in a golden age
when it comes to earth observation and i
say this for four main reasons
the first reason is that over the past
10 years
more satellites more civilian satellites
have been launched into orbit
than any decade prior and it’s getting
increasingly more cost effective
to build launch and deploy satellites
for example cubesats like the one you
see that’s being built in this image
are bigger are about as big as a loaf of
bread
and are made of commercial grade
uh electronic components that are
relatively inexpensive
the second reason is free and open
access to satellite data
in 2008 nasa
opened up its landsat archive going back
to 1972 that launch you saw earlier
and every landsat data
since then will be free to the public
similarly the european commission’s
copernicus program
which has several pro several satellites
in orbit at the moment and several more
are planned in the future
is also based on a open data policy
other countries that have national space
programs like japan
for example also provide some of their
satellite data available to the public
what all this means is increased
accessibility
to satellite data and data products for
the public
the third reason is the growth
of parallel computing and increasing
processing power
think about it it was only 15 years ago
that intel launched the pentium d which
was its first
consumer grade dual core processor today
you can get a processor with roughly
eight times the number of cores for
about the same price as in 2005.
on top of that we have tech behemoths
like google
and amazon and microsoft that provide
a cloud computing infrastructure where
you can get near unlimited uh
space for uh for satellite data as well
as insane computing power
for relatively low price
the fourth and final reason i think
we’re in a golden age
is the increasing popularity of machine
learning
and shared knowledge repositories the
thing about machine learning
is that it is capable of extracting
information
and patterns from satellite data with
increasing levels of accuracy and access
to machine learning has never been
easier than it is today uh there are
tons of online tutorials
and shared knowledge repositories like
github
where people post post their code for
others to use and build upon
and there are also uh structured online
programs
like udemy or coursera for anyone who
wants to delve deeper into machine
learning
all of this has been great for science
we’ve been able to
map and quantify and measure aspects of
our planet that we never thought were
possible
20-30 years ago for example in the last
couple of years
we’ve been able to map
whales in the world’s oceans trees in
the deserts
individual elephants in complex savannah
landscapes
and even albatrosses which are large sea
birds on remote islands in the antarctic
in a way satellites have become sort of
a macroscope
on our planet engaging itself
but you’re probably wondering what does
this mean what does all this mean what
does it mean for you to be in this
golden age of earth observation
it means a real democratization of
data and science it means that anybody
with a computer and an internet
connection can
download satellite data process it using
open source tools
and extract information for their own
use
it’s that simple and is right at our
fingertips
to give you an example of what this
combination can do
these 13 lines of code that you see here
produce an index that can tell you how
green
any given location on earth is
this is from google’s online platform
called earth engine
where with just a google account users
can manipulate
large amounts of satellite data at
global scales
you don’t even have to write a single
line of code
because lots of examples like this are
already provided and the data is there
as well
this one uses data from the landsat 8
satellite
so you’re probably wondering like what
did this index of greenness
look like right it looks like this
where the green areas show you healthy
vegetated areas
the brown and the yellow are either
stressed or no vegetation
and because it’s an image from a single
season the white areas are
are places that are covered by clouds
and in a lot of ways a few lines of code
is really all you need
to extract information like this from
satellite data
and you can do it globally and has real
practical applications for landscape
planning or for conservation
you can even use it to assess the
quality of life
in cities as research has shown that
greener neighborhoods have higher
quality of life
other examples include merging different
kinds of satellite data
to map different aspects of the land
surface for example here
where users can merge optical satellite
data
with data on the elevation of the land
surface
combine it in a ready-made ready-to-go
algorithm called principal component
analysis to extract
geologic maps for uh exploration of
minerals for example
you can also go back in time because we
do have that archive going back 50 years
and see the development of road networks
you can go online download data from the
80s the 90s the 2000s
and see how the urbanization took place
in a city
for example like dubai over here
or maybe you live in a rural setting and
you’re interested to find out the
distribution of crops
in your area so you go
to the fields you take a few samples of
the crop types
sunflowers here beets there cereals
there you combine that information with
satellite data
in the machine learning framework and
you map the crop distribution
where you live
let’s be honest we don’t know our planet
as much as we think we do and because of
this
the possibilities of what can be done
with satellite data
are truly exciting with all this
data and tools readily available we’re
really accelerating the rate at which
scientific discoveries are made
so the two things i want you to
get from this presentation the two
takeaways are that satellite data are
available
and they’re accessible to you thank you
very much