Mapping Earth's surface in 3D

Iceland (DLR) TanDEM's view of Iceland: The country was beyond the sight of the shuttle topography mission in 2000

The German satellite radar twins - TanDEM-X and TerraSAR-X - are a year through their quest to make the most precise, seamless map of varying height on Earth.

They've now acquired data across the entire globe at least once. However, some tricky sampling areas, such as tall mountains and thick forests, will require several passes and so we don't expect to see a fully finished product before 2014.

The Digital Elevation Model, or DEM, has become one of those must-have technical tools.

Planners use them to work out where best to put new buildings, roads and railways; airline pilots need the information to fly planes safely; telecos look at DEMs before siting their transmitters; and generals feed the data into autonomous weapons. Archaeologists, volcanologists, glaciologists, hydrologists, ecologists - they all want to know the shape of the Earth's surface. Even a games developer wanting to produce ever more realistic landscapes will have reason to resort to a DEM.

Space shuttle Endeavour famously made a world DEM in 2000. It hung a radar system out of its payload bay and mapped about 80% of the Earth's land surface.

I've heard it said that this is the most used Earth observation dataset in history. Someone can put us right on that if it's wrong, but the popularity of the shuttle's 3D map of the Earth doesn't surprise me.

And it explains why the German space agency (DLR) and Europe's biggest space company, Astrium, are now trying to produce a more detailed and a more extensive version a decade on.

The Shuttle Radar Topography Mission (SRTM) product has a best spatial resolution of 30m by 30m, and a vertical resolution that varies from 16m to 10m.

Artist's impression of TanDEM-X and TerraSAR-X (DLR) The satellites trace a helix across the sky as they move just 150m apart

The intention of the TanDEM mission is to go down to a spatial resolution of 12m by 12m and a vertical resolution of two metres.

"The two metres is what we call the relative accuracy," explained Dr Manfred Zink, the project manager for the TanDEM-X ground segment at DLR. "It means if you take an area one degree by one degree, latitude by longitude, which at the equator is roughly 100km by 100km, you can take any two points in the DEM and the error between the two points is less than two metres," he told me. "In absolute terms, each and every single point has to be better than 10m."

Airbone lidars can achieve much better precision, but these maps are necessarily regional in extent - they will cover only relatively small areas. The purpose of TanDEM is to build a world DEM that is single-source and has "no joins".

Of course, what I find fascinating about the TanDEM mission is the way that data is being acquired.

The pair's radars work by constantly bouncing microwave pulses off the ground and sea surface. By timing how long the signal takes to make the return trip, the instruments can determine differences in height. As they circle the Earth, so they build up their DEM. That much is obvious.

But TanDEM-X and TerraSAR-X do something a little more interesting.

Their slightly offset orbits bring them extremely close to each other - to a separation of just 150m, as of last week.

Think about that for a moment - 514km up, moving at 7km/s with a gap of 150m. It leaves simply no room for error.

This compact orbital dance gives the pair "stereo vision", by enabling them to operate an interferometric mode in which one spacecraft acts as a transmitter/receiver and the other as a second receiver. Varying the geometry allows the pair to discern better the signal they see from different types of surface.

Error map in data acquistion (DLR) Earth is being mapped in strips. Some hard to see areas will need perhaps three or four more acquisitions

If you look at the map of acquisition, the colours tell you how well the TanDEM mission is progressing.

The green areas are where the two-metre requirement has already been met.

Yellow denotes those areas that will need at least a second acquisition. If you know your geography, you can see these include forested and desert locations like the Sahara which are not easy surfaces to read for radar.

Red areas are in most need of further acquisitions. You'll see these include the big mountain ranges.

"We always image a swath at an off-nadir angle of 30-40 degrees. In other words, we don't look straight down," said Dr Zink.

"So, if you see a tall mountain and you image it only from one side, there is going to be a slope on the farside that you cannot see. This applies to places such as the Alps, the Andes and the Himalayas where after one acquisition, we still do not yet have enough data. These are the gaps we have at the moment."

If you are wondering, the grey describes those areas where the data has been acquired but not yet processed.

Salar de Uyuni Salar de Uyuni: The largest salt flats (blue) in the world covering 10,000 sq km, located in southwest Bolivia

This is a monumental job. SRTM collected 8.6 terrabytes of data. TanDEM will collect 350 terrabytes of data in mapping the Earth's roughly 150 million square kilometres.

All the radar information has to be downlinked to ground stations, and in the case of the remote O'Higgins facility in Antarctica, the data is even put on tapes and physically carried back to Germany.

"The first coverage of the Earth is done and was very successful," says Dr Vark Helfritz from Astrium Geoinformation Services.

"The test data is really very promising and we believe we could potentially outperform the original specification.

"We need double coverage for the easier parts of the Earth's surface, and triple - and maybe even quadruple - for the more difficult parts.

"The mountains in northwest America and Norway - its water fjords - have been particularly challenging. But the hardware is working really well.

"I'm really looking forward to the first dual coverage because I think when you see that, you'll really notice a big difference."

The TanDEM-X Digital Elevation Model of Mount Etna, Sicily, Italy The TanDEM Digital Elevation Model of Mount Etna, Sicily, Italy
Jonathan Amos Article written by Jonathan Amos Jonathan Amos Science correspondent

Lisa Pathfinder: 'Exquisite' gravity probe leaves UK

UK industry completes construction of the modules that make up the Lisa Pathfinder satellite - a remarkable probe that will test the key technologies needed to detect gravitational waves in space.

Read full article

More on This Story

Related Stories

The BBC is not responsible for the content of external Internet sites


This entry is now closed for comments

Jump to comments pagination
  • rate this

    Comment number 36.

    Robert Lucien wrote:

    "we really cant leave it to the Americans, I just don't think they have the imagination any more for space."

    That's funny considering that most of the current significant space exploration achievements/most adavanced technologies are American. Would you like to me list them?

    NASA also doesn't exist to serve other countries and their space ambitions and obsessions.

  • rate this

    Comment number 35.

    For me the answer is a larger bigger and better manned program. And we really cant leave it to the Americans, I just don't think they have the imagination any more for space.
    I just wish I could tell the space industry the simple truth that the world needs to make a manned program work - bigger is cheaper. Bigger gives you better fuel loads, larger cargo spaces, much cheaper costs per passenger.

  • rate this

    Comment number 34.

    Space is still dying on its feet, and it isn't lack of money that is the problem. Since space technology tends to be mostly capital type programs it still works in a time of recession. It pays back dividends in jobs and in tech base and the pay-offs often outweigh the initial costs. And that's before we even consider the impetus and inspiration it can give to younger generations.

  • rate this

    Comment number 33.

    In the past copying the Americans was no bad thing, but today there is little doubt that NASA and US science in general have kind of lost their way. There just seems to be a general lack of imagination, a lack of fore-planning and an inability to push the administration for proper funding of more ambitious programs. Despite all that is happening space is still dying on its feet. ...

  • rate this

    Comment number 32.

    Ian Brown wrote:

    "Envisat, SMOS, Cryosat-2 are all unique, important and innovative. Europe (DLR, CNES, ASI etc.) is a leader in ..."

    There is no "Europe" as a country or organization. The fact also remains that such so-called "European" organizations routinely copy the same types of missions that America has done simply for the sake of trying to show they can do it better. It's quite obvious.

  • rate this

    Comment number 31.

    #29. The European Space Agency has been responsible for some of the most innovative satellite missions ever. ERS-1 was ground breaking in 1991. Envisat, SMOS, Cryosat-2 are all unique, important and innovative. Europe (DLR, CNES, ASI etc.) is a leader in radar remote sensing, antenna engineering, small satellites, altimetry and much more.

  • rate this

    Comment number 30.

    #26 curiousman. It is in fact 2 m. This is an average and is due to the spatial averaging (over 12x12 m areas), vegetation, and quality differences due to orbital effects. It is worth remembering that ERS-1 and -2 acquired a large number of repeat observations to obtain sub centimetre accuracy.

  • rate this

    Comment number 29.

    "Endeavour famously made a world DEM in 2000"

    "the German space agency (DLR) and Europe's biggest space company, Astrium, are now trying to produce a more detailed and a more extensive version"

    Does any space agency in Europe do anything original? Sad that they often feel they have to copy America so as to try and prove they can do better.

  • rate this

    Comment number 28.

    Wil it map rocky oceanic outposts near tourist attractions ?

  • rate this

    Comment number 27.

    It could not possibly reach me on a "Monday morning"

  • rate this

    Comment number 26.

    Sporpo you are right : 2 metres is very poor. ERS-1 had an altimeter which measured to an accuracy of 10 cm from a height of 777 km. My guess is someone, somewhere has got a decimal point wrong. Perhaps they mean plus/minus two centimetres.

  • rate this

    Comment number 25.

    23. stevepilley

    I wasn't going to sign in but had to just to down-rate your comment.

  • rate this

    Comment number 24.

    It is worth considering the engineering challenges you need to overcome to achieve these accuracies with a satellite based interferometer. For example: the relative position of the satellites need to be known to less than a mm and their orientation to less than an arc second. Their absolute position needs to be known to less than a m. Clearly not impossible, but very difficult.

  • rate this

    Comment number 23.

    "Think about that for a moment - 514km up, moving at 7km/s with a gap of 150m."

    It'd be a lot easier to think about if you'd give the measurements in English. Lots of us still have no instinctive feel for these foreign weights and measures that have been dumped on us.

  • rate this

    Comment number 22.

    I've seen this pair of satellites flying across the night sky - they are amazing, especially when they catch the sunlight and flare up like the
    Iridium satellites do. They are only one-fiftieth of a second apart, a separation that has to be maintained by some incredible orbital precision.

  • rate this

    Comment number 21.

    17. Actually it's kind of the opposite the mm accuracy is from a large expanse, the footprint of the altimeter is km over the open ocean and most oceanographers average waveforms and so have less resolution.
    Over land, you do get a signal from all types of land with the altimeter but it is less reflective therefore the useful footprint shirks to 10s of metres.

  • rate this

    Comment number 20.

    "Airbone lidars can achieve much better precision, but these maps are necessarily regional in extent - they will cover only relatively small areas."

    sure, but could map the whole of the globe a few times over from planes for the price of building and launching the two satellites.

  • rate this

    Comment number 19.

    I would wish the radar measurements were more accurate. However people could put in selected more accurate surveys where needed. Then perhaps the stress in rocks and sub-strata could be measured as the tectonic plates move. When the stress builds up to breaking point the "noise" of the break up is an Earth-quake. As for sat-nav accuracy, the more you pay the better the measurement $$$$$$$ !!

  • rate this

    Comment number 18.

    Waiting to get on the coach and watch any square of Earth in stereo 3d!

  • rate this

    Comment number 17.

    Thanks Richyburger. Having read about the level of lakes varying from end to end due to wind direction and being aware of the influence of the moon and of currents on variance of water levels in the sea the notion of levelness is a difficult one to apply; any body / area of water consists of slopes. I assume the mm accuracy is limited to small chunks of area and is very time specific.


Page 1 of 2



Copyright © 2015 BBC. The BBC is not responsible for the content of external sites. Read more.

This page is best viewed in an up-to-date web browser with style sheets (CSS) enabled. While you will be able to view the content of this page in your current browser, you will not be able to get the full visual experience. Please consider upgrading your browser software or enabling style sheets (CSS) if you are able to do so.