A chunk of ice the size of downtown Manhattan fell off the Jakobshavn Glacier in Greenland on August 16, the fastest moving ice sheet in the world at present. As it melts, the glacier calves off icebergs and dumps freshwater into the North Atlantic at a rapid clip, a clip that has doubled in recent years.

Though this iceberg may be one of the biggest ever calved from Jakobshavn, the Greenland glacier is not unique in melting down. Arctic sea ice dwindles and glaciers atop mountains in more temperate and even tropical lands retreat. The meltdown of west Antarctica could raise sea levels around the world by more than three meters.

Some recent papers have argued that such meltdowns could happen by the end of this century, if not in the next few decades, as a result of the greenhouse gases piling up in the atmosphere. Rising levels of carbon dioxide are how ice ages are brought to a close and Earth may have entered a new geologic epoch, provisionally dubbed the Anthropocene, which looks likely to prove too hot for permanent icepacks. Given the risks of rising seas, thawing permafrost and lost glacial water reserves, some scientists have begun to wonder if aggressive intervention may prove necessary—the suite of techniques commonly known as geoengineering.

Already, Swiss ski resorts are covering retreating glaciers in a cloak of thin, reflective foil to stave off meltdown. Residents in Peru have used sawdust or whitewash to similar effect. But some scientists want to go further, given the absence of action to restrain greenhouse gas pollution. They want to change how much sunlight the planet reflects in a bid to save glaciers, cool the planet as a whole, or both. "These strategies range from lighter colored roofs or road surfaces” that could reflect sunlight, cooling cities, “to something as controversial as putting particles high in the stratosphere," explains Ken Caldeira, a climate modeler at the Carnegie Institution's Department of Global Ecology at Stanford University. Particles aloft would reflect some sunlight back into space, keeping Earth cooler.

Another alternative has been suggested by glaciologist Slawek Tulaczyk of the University of California, Santa Cruz, who studies west Antarctica's Whillans Ice Sheet, among other glaciers. A hundred kilometers wide, this ice sheet, unlike most of its peers, is actually growing instead of melting, because it has slowed its flow toward the sea in recent decades. Tulaczyk is not entirely sure why, but a better understanding of what slowed Whillans down might suggest ways to put the brakes on other polar glaciers. That might include draining away the water that lubricates the bottom of an ice sheet, speeding its progress to the sea, or installing barriers to prevent warming ocean waters from hitting the bottom of such glaciers and hastening meltdown.

Other experts, such as atmospheric scientist and astrobiologist Jacob Haqq-Misra of the Blue Marble Space Institute, say geoengineering could be used to return to an icier age [pdf]. Polar ice reflects more sunlight back into space than polar oceans or continents, helping to cool the overall climate. To grow the ice caps, Haqq-Misra suggests making marine clouds whiter or mimicking volcanoes by spreading a solar shield of aerosols in the stratosphere. Such sulfuric acid aerosols are already responsible for the bulk of nacreous clouds that form in the polar stratosphere; added particles would just amp up the natural process (although it might also amp up damage the ozone layer).

But the long-term side effects may be too much to bear, even if a larger ice cap could provide long-term insurance of a steadier climate as the sun brightens over hundreds of millions of years . Growing the polar ice caps, if they crept southward enough, might drive people out of northern Europe, Asia and North America, Haqq-Misra notes. The advancing ice could also bury some agricultural lands and make the planet an overall colder place, "likely reducing the total amount of habitable area on Earth," Haqq-Misra wrote in a paper laying out the thought experiment. "Yet if the objective is to ensure a stable climate so that civilization can exist for millions of years into the future, then perhaps such changes may be a small price to pay."

The price is not small, however, if the scheme doesn't work. Modeling studies on geoengineering to reflect sunlight away from the Earth suggest that modifying the planet's reflectivity could slow the meltdown of the Greenland ice sheet in the short term, but not stop it entirely, and could still allow an eventual total meltdown in the next millennia or so. Even if the sky is blanketed with sulfuric clouds, the models suggest, it would be difficult to regrow the Greenland ice sheet once it starts to melt—which it has.

Seeding the sky with particles might also not affect west Antarctica at all, as Southern Ocean waters continue to warm and lap up against the ice sheets, speeding melting even under a solar sunshade. And albedo modification to grow and thicken Arctic sea ice still lets Siberian and North American permafrost thaw, releasing greenhouse gases that would enhance global warming, computer modeling suggests.

Worse could be the impact on human psychology. "There is a potential risk that if you cool the planet by albedo modification, it could provide less incentive to reduce reliance on fossil fuels," says Marcia McNutt, a geophysicist, current editor-in-chief of Science and chair of a committee that evaluated climate intervention techniques for the U.S. National Academy of Sciences. "It's a therapy for the problem but it doesn't cure the problem."

"If somebody deployed a sulfur layer in the atmosphere, that could have catastrophic outcomes," like changing rainfall patterns over India that farmers rely on, adds Caldeira, who also served on the committee. "Anything big enough to have substantial climate impacts is going to be big enough to have climate consequences throughout the northern hemisphere and probably throughout the world."

A fleet of stratospheric jets spewing sulfuric acid to mimic volcanoes and reflect sunlight is not enough to save the ice sheets, but it may prove a tempting choice for a world faced by climate catastrophe. But the cure may prove worse than the disease. "We would dearly love to have found that we can go on emitting greenhouse gases and magically either clean them up after the fact or magically reflect sunlight to cool the Earth," McNutt says of her committee's analyses. "There is no silver bullet here. We need to mitigate and adapt. They are still the cheapest and lowest risk solutions to our climate trajectory and the best approaches we have at hand."