To stop rising seas from flooding coastlines, inundating cities and creating millions of refugees in countries like Bangladesh and Vietnam, scientists in Germany have looked into a novel solution: pump seawater up onto the Antarctic ice sheet, where it will freeze and turn into a salty mountain of snow and ice.
Of course, extracting some fraction of the oceans as if the planet were a flooded bathtub and dumping it in Antarctica like it’s a big bucket is an extreme idea. But given the collective failure to dramatically reduce carbon emissions so far – which is the sensible way to address the risks of climate change and slow the rising seas – we are now moving into an era when we need to take such ideas seriously.
Unlike other consequences of a warming planet, sea level rise is driven by simple but unforgiving physics: as the temperatures rise, ice melts and flows into the oceans and water expands. Even if everyone who owned a car traded it in for a skateboard and every coal plant were shut tomorrow, the planet’s temperature would continue to rise for centuries, due to the climate system’s inertia.
Advertisement
Sea level rise locked in
Each 1˚C rise in temperature above pre-industrial averages bakes in another 2 metres of sea level rise over the next two millennia. So that means no matter what we do, we’re committed to this rise as we’re already beyond 1˚C. And that’s a conservative estimate – studies of past climates suggest the rise could be much greater.
Right now, a lot of money and brainpower is being applied to figuring out how to protect cities like New York and London from rising seas. And as the authors of the Antarctica study – led by the Potsdam Institute for Climate Impact Research in Germany – point out, many valuable assets can be protected with walls and dykes.
But you can’t wall off entire ecosystems. And cities such as Miami, which is built on porous limestone, are in real trouble since the water will seep under any barriers, while nations like Bangladesh don’t have the money to protect the millions of people who live along its shoreline.
Bail out the oceans
So what do you do? The simplest response, of course, is to retreat from the shoreline and watch the old world go under. Another possibility – nearly as outlandish as pumping seawater onto Antarctica – is to inject tiny particles into the stratosphere, which would reflect away a small amount of sunlight, lowering the Earth’s temperature and, in theory, slowing the melting of the ice sheets.
The idea of bailing out the oceans may sound crazy, but it’s not unthinkable. The authors calculate that to suck out 3 millimetres per year of water (about the current rate of sea level rise) would require 90 pumps the size of the biggest pumps that protect New Orleans.
Because the Antarctic ice sheet itself is slowly sliding into the sea, the pumped water would have to be moved far enough inland – 700 kilometres, according to the study – to avoid speeding up coastal ice loss. Once there, after 100 years of pumping, it would raise the elevation of the ice sheet by about 25 metres.
Antarctic industrial zone
Powering the machinery to do this would require about 1275 gigawatts, which, thanks to the steady winds in Antarctica, could be provided by 850,000 1.5-megawatt wind turbines.
Granted, this would transform the last pristine continent into an industrial zone. The study foresees pumping for 100 years, so the project as imagined would only buy us a century of grace to prepare for the inevitability of the seas starting to rise again.
Such geoengineering is unlikely to ever happen. But there’s virtue in exploring deliberately provocative ideas like this, because it reminds us of what we’re up against. Earth is warming, and the seas are rising – fast. Millions of lives are at stake and the shape of the known world is being transformed. The question is, what are we going to do about it?
Journal:
Read more:
Latest numbers show at least 5 metres sea-level rise locked in
Geoengineering our way out of the climate mess
Paris climate deal is agreed – but is it really good enough?
