*34C3 preroll music*
Herald: Hello everybody to the next talk,
here at stage Clarke. The next talk will
be held in English. And here is a quick
announcement in German for the
translation. Der nächste Vortrag wird in
Englisch sein. Und wir haben eine deutsche
Übersetzung unter streaming.c3lingo.org.
Und wir haben das auch auf einer Folie.
Und es wird auch eine französische
Übersetzung geben für diesen Vortrag.
There will also be a french translation,
as well as an German translation for the
next talk. And you can find everything
under streaming.c3lingo.org. And, I hope,
displayed behind me. The next talk is
called "Watching the changing Earth".
Satellite data and change in the
gravitational field of the earth can tell
us a lot, especially when there's so much
public domain satelite data coming in from
different projects or maybe CC-BY
satellite data. And how this is done, this
new knowledge finding out of this big
heap of data, this will be explained by
Manuel in the talk. He dropped stuff to
see if gravity still works, or, in fancy
words, he does gravimetric methods and
sensory in geodesy. Is that pronounced
right? I'm not sure, but give a big hand
and a round of applause for our speaker
Manuel.
*Applause*
*No Audio*
Manuel: Geiler Scheiß. Oh, das war Sound.
So again, so hello and welcome to my
presentation on watching the changing
earth. This year's call for papers for the
Congress offered me the opportunity to
talk about my work in the related fields,
which is gravity. As far as Congress is
concerned, a misunderstood force of
nature. So in the following couple of
minutes, I want to talk about gravity,
gravitation, about the GRACE satellite
mission, which maps the earth gravity
field every month, about the gravity
fields, and I will show good results and
then we will go forward into the future.
That's nice. So it's actually called,
actually called geodesy. Let me give you a
short introduction on geodesy. Friedrich
Robert Helmert defined it in 1880 at as
the science of mapping and measuring the
earth on its surface, and this still holds
up today. It depends on your methods and
applications, but he was correct. The most
known profession is probably land
surveying, people with colorful
instruments and traffic cones. You find
them on construction sites, on the side of
the road, but we actually have a lot of
applications not only in geodesy but in
related fields like geophysics,
fundamental physics, if you want to build
an autonomous car you need geodesists,
metrology. This talk is specifically about
physical geodesy, which is the mapping of
the gravitational field of the earth, and
in this case specifically with satellites.
So I drop stuff on the earth, which is
terrestrial gravimetry, this talk is about
satellite gravimetry. Now gravity and
gravitation, we usually talk about
gravitational potential. This is a scalar
field. Gravitational acceleration is the
gradient of the gravitational potential
and when we talk about gravity in geodesy,
it's usually the combination of attraction
of the masses, gravitation, and the
centrifugal acceleration, but here we talk
mostly about gravitation. And the
potential can easily be calculated, at
least according to this very short
equation. We have G, which is the
gravitational constant of the earth, or
other planets if you want to do. We have
an ugly triple integral about the whole
earth, and this is basically what breaks
the neck. We have to integrate about the
whole mass of the earth, we divide up into
small parts and we need to know the
density of these parts. So, density times
small volume, you have the mass of the
earth if you integrate over it. So what,
the density of the whole earth is not
known. So if you want to calculate the
potential sufficiently, you would need the
density of a penguin on the other side of
the world. We don't know that. So, what
do you do if you cannot calculate the
quantity? You write a proposal and get all
the funding. This is what happened about,
let's say, twenty years ago, and the
result was the gravity recovery and
climate experiment, or GRACE for short. In
this talk, we will only cover gravity
recovery, so gravity field of the Earth.
As we can see, these are two satellites.
They are flying in the same orbit, and the
main instrument is distance measurement
between these two satellites, Here we see
the two satellites prior to its launch in
2002, and this K-Band Microwave ranging,
which is the instrument, gives us a high
resolution gravity field of the Earth.
This is spatial resolution of around 200
kilometers (km). You might think 200 km is
not really high resolution, but we have it
for the whole planet and not, let's say,
for Germany. And also we got the temporal
variations. So for 15 years now, we have
each month, with only a few exceptions, a
picture of the gravitational field of the
earth. The satellites fly in height of
about 450 km, 220 km apart, and we see
here the orbits of a single day. So 15
orbits per day, and we take one month of
data to generate one gravity field. The
working principle is quite simple: The
distance between the two satellites is
affected by gravity, so we measure the
distance and then we calculate gravity. In
a homogenous gravity field, this is quite
simple: Let's say we take a spherical
earth, it has only a single density, the
satellites fly along, and the distance
between the two satellites does not
change. There is nothing to pull one or
another, they just move along, not
changing the distance. Now we introduce a
mass, let's say a mountain, this can be
any mass change or density change
somewhere inside the earth, and the
leading satellite experiences a
gravitational pull by this mass. And as
gravitation falls off with distance, it is
a stronger than the pull experienced by
the trailing satellite. So the distance
between the two satellites increases. Now,
the satellite, the trailing the leading
satellite has passed the mass, and it is
still feeling its gravitational pull, but
now it is being decelerated because the
mass is behind. And the trailing satellite
is still being accelerated towards the
mass. This means the distance between the
satellites decreases. And finally, the
second satellite passes the mass and it
now also feels the gravitational pull
decelerating the satellite. The leading
satellite is feeling less and less
gravitational pull and once both
satellites left the gravitational
influence of this mass, we will have the
same distance as prior to encountering the
mass. So the gravitational acceleration is
a zero sum at this point. So of course,
the Earth is a little more complex than a
single mountain or a single density
anomaly in the ground, but this is the
basic concept. Now, how do we come from
these measurements to the actual
potential? The formula is basically the
same as a couple of slides earlier. We are
still calculating the potential. It looks
more complicated, but we don't have triple
integrals anymore, and all these
quantities in here are basically easily
calculated. We start with the
gravitational constant and the mass of the
earth, which we can get from a physics
book, if we like. And then we have a
couple of geometric quantities, a and r
are basically the size of my earth
ellipsoid, the major axes and r is the
distance from a calculating point, let's
say this podium, for which I want to know
the potential value to the center of the
ellipsoid. And then we have lambda and
theta at the end, these are the
geographical coordinates of this podium. P
is short for the associated Legendre
functions, also depending solely on
geometry, not on the mass of the earth,
depending on the software where you want
to implement this formula, it probably has
already a function to calculate this, and
if not, it is easily done by yourself as
the formulas look very long, but they are
quite simple. The interesting part are the
two parameters C and S, these are
spherical harmonic coefficients. They
include all the information about the
mass of the earth, as measured by the
satellites. So we have the satellites in
space, and the user gets just the C and S
coefficients, which are a couple of
thousand for the gravity field. Implements
this formula and has a potential value.
So, these spherical harmonic coefficients
are calculated from the GRACE Level 1B
products. These are the actual
measurements done by the satellites. This
is the ranging information, the distance
between satellites, satellite orbits, star
camera data, and so on. You add a couple
of additional models for earth's gravity,
which you do not want to include in your
satellite gravity field, and then you do
your processing. This is done by a couple
of different groups JPL and GFZ, which is
a German research center for the
geosciences. CSR is the center for space
research at university Austin. These three
institutes also provides these GRACE Level
1B data. So they take the raw satellite
data, process it to theGRACE Level 1B
products, which are accessible for all
users, and then calculate further these
coefficients, C and S. But there are also
additional groups who provide gravity
fields who calculate these coefficients,
for example, Institute for Geodesy of the
University of Graz, or the Astronomical
Institute of the University of Bern. They
all have slightly different approaches to
topic and come to more or less the same
conclusions. There are countless papers,
comparing these different gravity fields
with each other, but the user usually
starts with the coefficients C and S, and
then it takes a formula like the one on
top of this slide and calculates your
gravity value or whatever you want. Now,
I'm talking about potential, I'm talking
about accelaration. These are not really
useful quantities in day to day life. If
someone told to you in Greenland gravity
decrease by 50 microGal, you have two
choices, you can say "wow, awesome" or you
can say "oh no, we're all gonna die" It's
a 50:50 chance you'd say the correct
thing. So we are looking for a more useful
representation of the changes in
gravity field. Now gravity field reflects
mass redistributions and the most dynamic
redistribution we have is water storage,
summer/winter, more snow, more rain, less
water in summer, so we express our gravity
change in a unit called equivalent water
height. This is the layer of water on the
surface with a thickness, equivalent to
the mass change measured with the
satellites. This is also easily
calculated. This is my last equation, I
promise, but this looks familiar. The
second half of this equation, is basically
the same we saw one slide prior and the
parameters in front of the sum is the
average density of earth, which is around
5500 kg/m^3. We need the density or water,
let's say it 1000 kg/m^3. And in this
fraction in the middle, we need to
parameter K, which are the so-called Love
numbers. Now, this is not a numerical
representation of mutual attraction, but
was put forward by, I think, Albert Love
in 1911, and they are parameters
concerning the elastic response of the
earth to forces. So, if you put a lot of
weight on a part of the earth, the earth
deforms and these parameters, describe the
elastic response of the earth to such
loading. Now we have calculated our
equivalent water height, let's say for two
months, let's say, in May 2002 and 15
years later in May 2017 and we just
subtract these two gravity fields, these
two equivalent waterheights, from these
two epoches. What we have left is the
change in gravity between these 2 epochs,
15 years apart, expressed in water layer
equivalent to the change in gravity
measured. And we can see a couple of
features here. There should not be any
seasonal variations because it's the same
month, just 15 years apart. So we see long
term gravity change between these two
epochs. And what we see is, for example,
mass loss in the northern and southern ice
shields, and we see two red blobs, one in
northern canada and one in northern
europe, which are geophysical processes.
So this is glacial isostastic adjustment
and during the last ice age the ice
shields deformed the earth downward.
The material in the "Mantel" had to flow
aside, and now that the ice is gone, the
lead is uplifting and the material in the
"Mantel" is flowing back. So it's flowing
back and the earth is uplifting. This
process has been going on for 10000 years
and will probably a couple of years
longer. Now how do you get your data?
Everyone can get the GRACE Level 1B data,
which are the observations by the
satellite, like again, ranging information
between the satelite, orbits,
accelerometer data, star camera data and
so on. You can get them without hurdles at
the ISDC, which is the information system,
a data center at the Geoforschungszentrum
Potsdam, or at the Physical Oceanography
Distributed Active Archive Center run by
JPL. And if you'd like, you can calculate
your own spherical harmonic coefficients
for gravity fields. Or you can compare for
example, satelite orbits they give you
with one you integrated yourself using
your own gravity field, to see if they fit
together or not. You can get gravity field
models, if you'd like. A large collection
is at the International Centre for Global
Earth Models. They have recent and
historic gravity models all in the same
data format. So you only need to implement
your software once from the 1970s to
today. They also have the proper
references, the papers you want to read to
work with them. These are so-called Level
2 Products. So, you can take a gravity
field from there, use the equation, I
showed you earlier and calculate your
equivalent water height, if you'd like.
If you don't want to do this, there is
someone to help you, a service called
"TELLUS", which is a play on words I
don't want to go into detail about. They
offer equivalent water heights calculated
for each monthly solution from the GRACE
satellites. This tells us a lot about the
earth, if you look closer into it. In the
following, I will use the monthly
solutions from the ITSG-GRACE 2016,
provided by Institute for Geodesy at
University of Graz. The previous graph I
showed you was also created with that
gravity model. I will not go into detail
about further processing like filtering
and gravity reductions done to this, not
enough time. So here are some results,
let's start with the most obvious one, the
greenland ice shield, which has, as we saw
earlier, the greatest loss of mass
according to the gravity field and we see
here, a water layer on the whole landmass,
describing the loss of mass expressed as a
water layer of a certain thickness.
So let's say in the southern tip, you have
one meter water layer. This would be
equvalent in gravity to the actual mass
lost in Greenland. But we also see, that
the signal is not very localized. So it's
not bound to the land mass. It's also in
the ocean. This effect is called leakage.
If you do signal processing you will know
this. There are methods to reduce leakage.
My next slide will show such a result, but
I have done no reduction to this. So if
you use my formula I showed you, you will
pretty much get a result like this. This
gives you a trend of around 280 gigatons
per year in mass loss over the whole land
mass of greenland. And now gigatons is
also not very useful an expression. One
cubic meter of water has a weight of a
1000 kilos; one tonne, 1 gigatonne is
10^9 tonne, if you are familiar with ball
sports, 1 soccer field with the 140 km
high water column has the weight of
1 gigatonne, or if you are not fan of sports
ball, if you're more of a plane guy or girl
the A380-800 has a maximum takeoff weight
of 575 tonne, so we need 1.7 mio of these
airplanes for one gigatonne. So this is a more
beautiful representation of the process in
greenland, done by NASA JPL. If you go to
the website of the GRACE project, they
have a couple of these illustrations, they
obviously worked hard on the leakage.
You can see localized where most of the
gravity, most of the mass is lost on the
left and on the right you see accumulated
over time, the mass which is lost, and
which trend it gives you. Also, if look
closely in the center of greenland, you
see black lines, these are the ice flow,
as determined by radar interferometry.
So now pretty much know where ice is lost,
where mass is lost. This goes into the
ocean, and this would be a good idea to
see, to check our GRACE results, the mass
we find missing on earth, so the melted
ice, and the additional mass in the ocean,
does this agree with other methods who
determine the sea level rise. One of these
methods is satellite radar altimetry, that
started in the 70's, but since 1991, we
have lots of dedicated satellite missions,
which only job is basically mapping the
global sea surface. So, they send down a
radar pulse, which is reflected at the sea
surface. They measure the run time and
then they have a geometric representation
of the global sea surface. Now, if we
compare this with the mass we calculated
or we got from the GRACE result, calculate
a sea level rise rise from this additional
mass in the ocean than these two systems
would not add up. The geometric sea level
rise is higher than just the additional
mass. So there is the second process which
is thermal expansion of the water. If
water gets warm it needs more space.
In 2000 the deployment of so-called ARGO
floats started. These are free-floating
devices in the ocean. Currently, there are
over 3000 and they measure temperature and
salinity between sea surface and a depth
of 2000 meters. These are globally
distributed. So, we have at least for the
upper layer of the ocean, how much thermal
expansion there is. And what we want to
see is, do these components of additional
mass in the ocean as determined by GRACE
and thermal expansion of the upper ocean
layer come to the same result as
geometrical measurements done by satellite
altimetry? On the left we see an image
taken from the last IPCC report on climate
change from 2013. In green we see the
sealevel rise as measured with satellite
altimetry in the time span 2005 to 2012
and in orange we see the combination of
additional mass, as measured by GRACE, and
thermal extension as determined with ARGO
inside the ocean. And these 2 graphs follow
each other quite well. On the right. We
see a recent publication by Chen, Wilson
and Tapley, the latter one being one of
the PIs of the GRACE mission, who
accumulated the data from 2005 to 2011. We
basically come to the same conclusion. So
now if you really don't want to do the
math, there are online services who make
the graphs for you. One of those is
EGSIEM European Gravity Service for
Improved Emergency Control. If we can
measure how much water is stored in a certain
area, we know that this amount of water
has sooner or later to be removed from
this area. This can be a flood, for
example, and with a mission like GRACE, we
can determine how much mass, how much
water is there and are the rivers large
enough to allow for this water to be
flowing away. That was the intention
behind this service. Oops, no, this is not
the future. So, I wanted to do the life
demo but. So, yeah, the live demo did not
work as expected. So, you will be greeted
with this graphic. You can plot for all
areas in the world. The first thing you
have to do is you change your gravity
functional, we want water heights. This is
what I talked about in this talk. Then you
want to look at the data set and at the
bottom you see a large list of GRACE
gravity fields. These are different
groups, I mentioned, providing these
monthly solutions. And so we choose one of
these groups. Then we choose an area which
we are interested in. You can freely
choose one area like here Finno-Scandia,
or you can use pre-determined areas, for
example, the Amazon river basin or Elbe
river or something like that. These areas
all over the world and you can see the
gravity change in this area. So let's look
here at Finno-Scandia, and then you are
greeted with a plot like this. This is
equivalent water height, even though this
is a geophysical process. So we see here
the layer of water, which would have been
added to the region as selected, and we
see a clear trend upward. Again, this is a
geophysical process. This is not
additional ice or water or anything. Can I
return to my...? No, I cannot. So, yeah,
live demo did not work. If you want to do
this yourself. I have uploaded to the
Fahrplan all my resources, all my links.
And the EGSIEM page also includes the
description of what is done in the backend
and were the data comes from and what you
can see in the various fields. Now I want
to give a last impression on the future,
because unfortunately while I was
preparing my abstract for this conference,
one of the GRACE satellites was turned off
due to age. It was launched in 2002,
planned for a five mission year; it
survived 15 years, which is quite good,
but now we have no more ranging
information between these satellites. We
had ranging information in micrometer
accuracy, a couple of micrometer, and now
we cannot rely on these information
anymore. And this means mo more gravity
fields with high spatial resolution, and
I'm not sure about the temporal
resolution. So, the current work which is
done is taking all satellites which are in
the low-enough orbits and calculate the
gravity field from their positions,
because everything which is in low-earth
orbit is affected by the Earth's gravity
field. So, if I take the satellite orbits,
look "how does this orbit change" and the
reason is gravity, then I can calculate
the gravity field. Unfortunately, not in
this higher resolution we are used to.
And... But fortunately, there already is a
next-generation gravity field mission on
its way. It arrived last week in the US,
where it will be launched in late March,
early April by SpaceX. You might look at
this image and think, "I just saw this
earlier" and you are quite correct: The
mission called "Grace Follow On" is a copy
of Grace, which improved components, of
course, and now with lasers. We see not
only the microwave ranging between the two
satellites, but additionally a laser
interferometer. So, from micrometer
accuracy in the distance measurements we
go to nanometer accuracy, hopefully. But
the main instrument will be the
microwave ranging. So, in conclusion,
I hope I showed you that the gravity field
can show mass transport on the surface and
inside the Earth; that this offers, in
combination with other methods, new
insights and also some kind of mutual
verification. If several different types
of observations coming to the same
conclusion, none of them can be awfully
wrong; and that the access to these
methods are relatively easy: the data is
available, all the methods are described
in geodesy textbooks and the technical
documentation; and there are other
applications, other than, let's say,
climate change; you can look into drought
and flood prediction; the El Niño–Southern
Oscillation you can predict from Grace's
gravity field data. So, lot's of work to do.
So, this would be the end for my talk.
I thank you for your interest in the topic.
*applause*
Herald: Thank you, Manuel, for the talk.
And I think we have time for one or two,
maybe two very short questions. Please be
seated during the Q&A session. Is there
some questions? Okay, microphone 3,
please.
Mic 3: Yeah, hi. *In a quiet voice*
Hi, hello? Can you hear me? *Now loud*
Herald: Yeah.
Mic 3: Okay. Hey. So, my question is
regarding acceleration. What's the
influence of Earth atmosphere and all the
planetary bodies, like the moon, and does
it need to be accounted for?
Manuel: The external gravity needs to be
accounted for, so the tidal effects of sun
and moon would be one of those additional
models you put into the processing of the
satellite data. The Earth's atmosphere has
an effect on the satellites themselves,
which is measured onboard by
accelerometers and then reduced. And the
gravitational effect of the atmosphere:
Part of this is averaged out, because we
take a month of time series, and the rest
are also inclu... provide as extra
products; at least by the Institute for
Geodesy in Graz. So atmosphere... the mass
of the atmosphere is... has to be
accounted for, yes.
Herald: Okay. Microphone 2 has vanished
all of a sudden. Then, microphone 1,
please.
Mic 1: Hi. Is it possible to measure
changes in the temperature of the oceans
or of the ocean streams, like... Can you
see if El Niño is active by just measuring
the gravity... change in gravity fields?
Manuel: As a precursor tool, El Niño, as I
understand it... certain regions of the
ocean get warmer; it's a density change;
and, of course, this would be measured as
part of ARGO and it's also in the GRACE
gravity field. There are probably papers
on it. So, the last... the extend of the
last El Niño was predicted by GRACE. I
don't know to what extend this was
correct, but...
Mic 1: Okay, then.
Herald: Good. Then, that's all the time we
have. A big round of applause for Manuel
and his talk, please.
*Applause*
*34C3 Music*
subtitles created by c3subtitles.de
in the year 2020. Join, and help us!