Witnessing the Big Melt

Initially the Grace satellite data were a bit disappointing. The gravity recovery and climate experiment satellites became operational in March 2002, revolving around the Earth in a north-south direction, sensing the gravitational pull of the Earth by measuring the variation of the distance between the satellite pair. The distance between them was typically a few hundred kilometres, the variation only a few micrometers – hundred thousand times million times smaller.  But when results of subsequent passes were compiled to arrive at a gravity atlas of the world, broad noisy bands spoiled the picture.
“It was the result of changed circumstances at the Earth surface – like for example the tide had changed”, explains dr Bert Wouters, who defended his thesis last Tuesday at the faculty of Aerospace Engineering (AE). The different amounts of water meant a variation of the local mass, and hence another gravity value. “Of course people at Nasa and DLR [German Aerospace, ed.] have foreseen the effect, but the correction they applied couldn’t take the noise bands away”, Wouters says.

So the Delft engineer set out to design a filtering algorithm himself, called EOF or ‘empirical orthogonal function’, making use of the by then large existing dataset consisting of six years worth of measurements. This allowed Wouters to filter out measurements with the same periodicity in time. The shortest periods are those of the ocean tides, while seasonal variations are longer and ocean currents have a longer periodicity still. Selecting measurements with the same periodicity and rejecting others allowed him to get rid of ‘non-coherent’ signals – another word for ‘noise’. Needless to say this involved massive amounts of number crunching, putting elaborate datasets through impressive mathematical formulas. The work even led to a proposition in his thesis: “Gathering scientific observations is a sleep-inducing and time-consuming activity, what is flagrantly ignored by the end-user of the data.” Indeed, the resulting gravity atlas looks disturbingly normal.

The filtered data allowed the melting of Greenland’s land ice to be measured directly for the first time. Wouters published his findings, together with dr. Ernst Schrama (AE), in the October 2008 issue of Geophysical Research Letters. They showed that between 2003 and 2008 Greenland had lost on average 195 cubic kilometres of land ice per year – enough to raise the oceans worldwide by 0.5 millimetres. Their publication drew a lot of attention. ‘Greenland ice loss behind a sixth of sea level rise,’ wrote New Scientist.
Until the late 1990s, sea level rise was solely caused by the expansion of warming seawater. It was only in the early 2000s that melting land ice started to contribute as well, and now both causes are equally large. Melting ice from Greenland, Antarctica and glaciers adds 1.5 millimetres per year to the sea levels. Another 1.5 millimetres comes from the thermal expansion.
‘The synergy between Grace data (…) and regional climate models (…) promises a leap forward in our understanding of the mass balance of the ice sheets’, Wouters writes in his thesis. He refers to the collaboration he has set up with climate modellers from the institute of marine and atmospheric research Utrecht (IMAU), which is part of Utrecht University. IMAU has developed a model for the Greenland ice sheet based on the temperature data of the last fifty years. The predictions it has produced for ice loss in recent years were within range of the measurements from GRACE data. “This means that their model is basically correct, and they can start predicting ice sheet loss for the next 20-30 years”,  Wouters explains.

The ability to monitor ice sheets from space is not guaranteed, however. The Grace mission is still active but long overdue. The Goce satellite, aimed at producing a more detailed gravity map, is expected to have a much shorter life span because of its reduced altitude. 

Bert Wouters: Identification and Modelling of Sea Level Change Contributors, On Grace satellite gravity data and their applications to climate monitoring. January 19, 2010.

De 186 studentenwoningen zijn vorig jaar in tien maanden tijd neergezet. Dat is meer dan twee keer zo snel als normaal, zeer nodig in tijden van stijgende studentenaantallen.

Ze konden zo snel worden gebouwd, doordat de bouwers werkten met stapelbare units. In september kon de eerste toren al worden bewoond, in november en januari volgden de tweede en de derde.

In de torens zitten zelfstandige studentenwoningen, groepswoningen voor drie en kamers waarin twee studenten kunnen wonen. Allemaal zijn ze bedoeld voor internationale studenten.

Op de begane grond van iedere toren is een gemeenschappelijke ruimte, een fietsenstalling en een paar bedrijfsruimtes voor startende ondernemers. In de middelste toren bevindt zich sinds drie maanden een huisartsenpost voor studenten.

Volgens beheerder Duwo trekken de woontorens veel belangstellenden in de snelle bouwmethode.

De opening begint om 16.15 uur.