A Planetary Perspective: With Landsat and Google Earth Engine

Uploaded by EarthOutreach on 23.07.2012


TOM LOVELAND: Landsat has become one of the more common
sets of remote sensing images.

It was launched in July of 1972.
That started the process of continuous observation of the
land surface of the globe.
It represents the longest running record of the
landscape of the planet that's ever been recorded.

BRUCE PENGRA: The Landsat satellites pass over the same
place on the earth every 16 days.
Their orbit goes from pole to pole, and they capture a new
place, and then every 16 days they start to repeat that.
And it's been doing so for 40 years now, continuously
orbiting, collecting data.
TOM LOVELAND: NASA takes the lead on the development of all
of the space hardware.
They build satellites and they put them in orbit.

Once they're in orbit, we collect and distribute the
imagery, and ensure that the data are preserved
over the long term.
MATT HANSEN: Working with satellite data is important
because we can monitor the environment, how it changes
over time in a very consistent way.
We can quantify the dynamics.
Whether it's related to agriculture, expansion, fire
regimes, deforestation, urbanization, and have a
record that allows us to relate those
dynamics to the drivers.
Population or economic policy, what have you.
BRUCE PENGRA: It's really kind of an underpinning of what we
need to know to understand the health of the planet.

TOM LOVELAND: One of the really big stories of
environmental change is the deforestation in the Amazon.
This whole area here, if you go back to 1999, the vast
majority of it is intact forest.
And you can just see the herringbone roads come out
through the forest, and then roads come out from, and the
land clearing and the forest disappearing.

BRUCE PENGRA: The value of Landsat is in
the use of the data.
That means that the data have to be in the hands of users.
In December of 2008, we completely rewrote our data
policy to one in which all Landsat data were available to
anybody without restriction at no cost.
MATT HANSEN: So they open up this archive, this treasure
trove of data going back to the early '70s.
We used to say we could only use the data we could afford,
not the data we needed.
Well now, we can have all the data.
BRUCE PENGRA: The challenge, of course, isn't collecting
the imagery as much as it is interpreting and applying
those to problems.
And that's simply a massive job.
MATT HANSEN: And this is where the Google Earth
Engine comes in.
We have to have something like cloud computing to be able to
analyze this data and quantify the changes.
And Google provides the computing to match up and mind
the archive.

An example application with the Google Earth Engine was
done by us in December.
We accessed over 50,000 images.
We brought them in, we processed them with the cloud.
If we had one CPU working on that, it would have taken over
three years.
We did it in about 15 hours.
BRUCE PENGRA: I think the key to this partnership is that
our data, being applied very creatively by the Earth Engine
activity, is really pushing Landsat to that level that
we've all dreamed about over the years of its history.

Landsat data are now in far more hands than ever before.
When we start having simple to use tools that are proven,
tested, and can be run by people with less training,
then we'll make more progress.
The Earth Engine promises to be that simple tool
that many can use.