Jane Burston- Understanding Climate Change with in-orbit satellite calibration

1953, a storm surge covered 160,000 acres of land
near the mouth of the River Thames in London, killing
more than 300 people. If it had reached the low-lying,
densely populated areas of London, where
we’re standing now, it would have been
even more horrific. The government
reacted by agreeing to build a Thames
barrier– essentially a gate across the
river, which can be raised when needed to
protect both against storm surges coming in from
the sea, but also flood water coming
down the river. It was built in 1982. And in its first
decade of existence, it was lifted 10 times. In the last decade, it’s been
lifted more than 80 times. And the kinds of
extreme weather events that have seen it be
lifted so many more times than was
originally forecast are exactly the kind of
extreme weather events that it predicted to become
more numerous and more intense as a result of climate change. Now, as a consequence,
people have started to ask the question, for
how long will this $1.5 billion piece of infrastructure
remain fit for purpose? Now, climate models can help
give us some of the answer. But their results vary. And there’s uncertainties
around those results. The innovation I want
to talk about today will allow us to collect
data about the climate at 10 times the accuracy
that we’re currently able to, constraining those
uncertainties in the models, and helping us to determine
which climate models to base our investment decisions on
much more quickly than we’re currently able to. But back to the problem. This chart is from the
Intergovernmental Panel on Climate Change. And it shows two
different scenarios for what the future
temperature rise will be by the end of the century,
given two different input emissions. The blue line at the bottom
is the likely temperature rise if we immediately start to
reduce emissions and carry on doing so quite sharply. The red line at the top is if
we carry on emitting as now. And the fuzzy bounds
around both of these two are the uncertainties. So the different
answers to the question, what would happen with this
level of input emissions? As you can see,
the uncertainties are unpredictably and
unacceptably large. Just to take the red band, for
example, what that’s saying is if we carry on emitting
at our current rate, the answer to the question,
what will the temperature be by the end of
the century, is it will be somewhere between
3 degrees more than now and 5 and 1/2 degrees more
than now, which corresponds to a massive range in
the potential impact that we would see,
and a huge range in the amount of
investment that we would need to be able to adapt. So why can’t we get
the kind of accuracies that we need in the measurements
to predict the level of climate change more accurately? Well, the answer is
that we’re trying to detect very small trends
against a very large background of natural variability,
something like 0.2 of a degree of temperature
rise, for example, against natural variability
that sees the temperature vary 10, 15, 20 degrees on
even on any given day, let alone around the globe. So the measurements have
to be very accurate. The problem is calibration. I work for the National
Physical Laboratory in the UK with the National
Measurement Institute. And we were set up in 1900 to
provide traceable measurements to science and industry. Our lab and many like
it around the world will calibrate satellite
images using instruments like this cryogenic solar
absolute radiometer. It measures optical parallel
light at very high accuracies. And the way that it works is
it’s a very cold, black cavity on the inside that heats up
when you shine light on it. When the light is
off, you can obtain that same temperature rise
using electrical power, and then you can choose
the optical power from the electrical power. So we can calibrate the
instruments very accurately in the lab using
this kind of kit. But then the rigors of the
launch and the harsh space environment mean that
calibration is lost in orbit. Instruments will drift. This chart shows the drift
and instrument called VIIRS. It was launched by NASA. And the lines are
different spectral bounds. You’ll see over a
period of 10 months some of the spectral bounds
are out now by 15% or 30%. And that’s over 10 months. Obviously, we can
calibrate the instruments once the satellite is in orbit. And there’s two main
ways of doing that. But both have their limitations. The first way is by
very accurately mapping a piece of ground. In this case, the
Antarctic snow field, whilst some locals
are investigating. And once we very accurately
map the reflectance of this piece of ground,
you can compare that to what the satellite
sees what it views the same piece of ground
and see if the satellite’s getting it right. The limitations of this
are both our ability to very accurately
characterize the site, but also, the fact that the
satellite is seeing that through the atmosphere. So we have to correct for that. The second way of calibrating
satellites in orbit is on orbit collaboration. And it works by having
a source of light mounted on a
satellite, like a lamp. Or actually, in
some cases, the sun is used, shining onto a
diffuser into the image. And when you know the brightness
of the lamp or the sun, you can compare that to
what the imager is seeing. The limitations of
this are in fact, this collaboration
system in the lamp, plus diffuser, or even
the sun, plus diffuser, can degrade itself in orbit. So although we know how
much the energy degraded by, we don’t know how much the
calibration system degraded by. And we can’t correct the bat. So the upshot of all of this
is that currently, our best systems are able to collect
data about the climate only within 2% to
3% uncertainty. And what that
means is it’s going to take us between 30 to 50
years, potentially even longer for some parameters to detect
the trends in the climate to know which climate
models are getting it right and which to base our
investment decisions on. If you could imagine you’re
responsible for building the next Thames barrier,
and I say, oh, absolutely. I can tell you when
you need to build it and how high it needs to be. Just give me 30 to 50 years,
and I’ll be right back to you. Fortunately, we have Nigel Fox. Nigel Fox runs our Earth
Observation Lab at NPL. And he thinks about this kind
of problem all of the time. Nigel was thinking, how
do we correct for this? How can we get a more accurate
collaboration in orbit? Wouldn’t it be great if
I could launch myself onto a satellite
with the [INAUDIBLE] that we use in the
lab and calibrate it once it’s up there? Now, the kind of satellites
that we’re dealing with don’t take that kind of payload. Sorry Nigel. But he came up with a way of
militarizing it and mounting it onto a satellite so that it
could calibrate satellites whilst they’re in orbit. Like existing systems, it
uses the lamp plus diffuser. But unlike existing systems,
it doesn’t stop there. It can detect how
much that calibration system has degraded. So we can account for that
and correct for that, as well. It worked by shining a
laser into this black cavity and then onto the diffuser. So we know by how much
that system degraded. Now, you might be sitting
there thinking, great. So we get that you’ve
taken this calibration chain one step further back. But how do we know how much
the [? saw ?] graded by? Isn’t that going to hamper us? And the answer is no. The [? saw ?] is
coated on the inside with the blackest
material on earth. It lets out less than 0.002%
of the light that enters it, which is a UK first. It’s carbon nano tubes. This cavity could degrade
by a factor of 100 and still give us the accuracies
that we need for climate. So across all of
that, we can now, if we can launch this
satellite, detect climate data at the kind of accuracies that
are 10 times better than we can at the minute– 0.3% uncertainty
rather than the 2% to 3% that we can currently get, which
means that we can bring forward the time that we know which
climate models to depend on by a factor of two
or three times. The US economist
working with NASA has calculated that that
would save about $5 trillion. In fact, between $5
and $30 trillion, which is a large rate. But both very big numbers. As a result of the better
and more timely investment decisions that you would make
about adapting to climate change. Now, there’s a
second thing that’s important about truth, which
is because it has a truth polar orbit, it could calibrate
other satellites in orbit by passing over the
same piece of ground, taking a picture of the
same piece of ground at about the same time. And then you can use
the image from one to calibrate the other. So you don’t just get one
super accurate satellite. You get to upgrade the whole
earth observing system. Imagine the kind of
difference that that could make to existing satellite
services like precision farming, or mapping forest
cover, pollution monitoring. So I wanted to
finish by letting you know where we’re up
to with this project. And we’ve been funded,
along with our partners, by the UK Space Agency to
develop the mission concept. And most recently, to
develop a detailed technical specification and costing. The next stage is
prototyping for space, which will cost just
over a million pounds. And we’re already
progressing quite well to obtaining that funding. After that, the
full cost of launch is between 80 to
100 million pounds. 80 to 100 million pounds to
say the least, $5 trillion, which is a good business
case in anybody’s currency. But this isn’t just
good value for money. This is great science. [APPLAUSE] FEMALE SPEAKER: Thank you Jane. And similar, what are
the big challenges that you’re currently facing? And where could we pick
these fantastic brains and get some ideas? JANE BURSTON: Cheers. Well, we obviously
want to launch it. And to launch it, we need
all of the components. So we do need the
full 100 million, even though the
novel component of it would probably cost somewhere
between 20 to 25 million. So I think there’s three main
ways that it might be launched. The first is full funding
from one government. For example, the UK government. And what they have
asked us to do is come up with specific
UK businesses or academics and how they would
use this data that’s 10 times more accurate, what the
value of that would be to them. So if you run a business
that uses satellite data or if you run a business that
has infrastructure and could benefit from this knowledge
about when you need to upgrade it and how, we’d
like to know that. I suppose a second way
of getting it funded is in collaboration
with another country. So if you have any links to
satellite companies or space agencies in other countries,
that would be useful. And then finally, we’ve just
seen about the crowd funding at the lunar landing. Is this the kind of thing
that could get crowd funded? How would that work? Does anyone have a spare
satellite we can borrow? Any [INAUDIBLE]. FEMALE SPEAKER: Brilliant. JANE BURSTON: Thank you. FEMALE SPEAKER: Thank
you so much, Jane. [APPLAUSE]

Tags:, ,

Add a Comment

Your email address will not be published. Required fields are marked *