Antarctica's Ice Attacked from Above and Below

One of the fastest-warming places on the planet, the Antarctic Peninsula, has lost two massive ice shelves in the past 20 years: the Larsen A and the Larsen B. Each floating tongue of ice disappeared in a matter of weeks.

No one knows for sure why the ice shelves collapsed, but a new study finds the remaining Larsen C ice shelf is melting from more than just toasty air. This ice shelf is also disappearing due to warming in the ocean, scientists reported May 12 in the journal The Cryosphere.

"We know there is strong atmospheric warming in this region," said lead study author Paul Holland, a climate scientist with the British Antarctic Survey. For instance, this summer, the Antarctic Peninsula set new heat records. But until now, no one thought the ocean played a role in the collapsing ice shelves, Holland added. "On the Larsen C ice shelf, we see 4 meters [13 feet] of ice loss that can't be explained solely by atmospheric warming," Holland told Live Science.

The research team monitored thickness changes in the ice shelf between 1998 and 2012, using satellites and ground-based radar. The scientists saw Larsen C lose thickness in two ways during this period.

First, air escaped from the old snow layer on top of the ice. This old snow, called firn, has tiny, permeable air pockets that help it insulate the ice underneath like a down comforter. In summer, meltwater can trickle down through the snow like syrup through a frozen treat. The firn loses thickness when it melts down, and the air is lost. If the firn freezes into ice, meltwater ponds on the surface and absorbs sunlight and heat, driving further melting.

The Larsen B had completely lost the air in its firn layer before the shelf's catastrophic collapse in 2002, Holland said.

Second, the researchers think that Larsen C is also melting from below. During the 14-year study, the ice shelf surface lowered by an average of 3 feet (1 m). The surface lowering could simply be from the compacting snow, but the ground-based radar can reveal how much air is in the firn, Holland said. By measuring the air levels, the scientists determined the ice itself thinned by an average of 13 feet (4 m) between 1998 and 2012.

"Larsen C is getting lower in the water due to loss of ice," Holland said.

However, glaciologists are concerned that the Larsen C ice shelf could crack and retreat by detaching from a small island called Bawden Ice Rise. A crack is forming near the northern edge of the shelf.

"When Larsen A and B were lost, the glaciers behind them accelerated, and they are now contributing a significant fraction of the sea-level rise from the whole of Antarctica," David Vaughan, a glaciologist at the British Antarctic Survey, said in a statement. Vaughan was not involved in the study. "Larsen C is bigger, and if it were to be lost in the next few decades, then it would actually add to the projections of sea-level rise by 2100."

Average ocean temperatures in Antarctica have been rising since the 1990s, especially around the peninsula where Larsen C is located. Scientists reported in 2015 that Larsen C was riding lower in the water than it had previously and had lost 13 feet (4 meters) of ice that could not be attributed entirely to warming air temperatures.

The first signs of a northward-extending crack in Larsen C appeared in 2010 and progressed in 2014, according to a study published in 2015 in the journal The Cryosphere.

The electromagnetic spectrum covers a wavelength range of about 24 orders of magnitude, from the shortest, most energetic, high frequency gamma rays to the longest, very low energy, low frequency radio waves). The range of interest for climate science and atmospheric warming is approximately in the middle of this spectrum from about 0.1 to 100 μm (10–1 to 102 μm). This includes the visible region of the spectrum (0.4 to 0.7 μm) and the adjacent higher energy (shorter wavelength) near ultraviolet (UV) and lower energy (longer wavelength) infrared (IR), including much of the “thermal infrared”. Incoming solar radiation is in the shorter wavelength (higher energy) part of this range from UV through near IR, between about 0.1 to 4 μm. Radiation from the warmed Earth is mainly in the thermal IR region between 4 and 30 μm.

Molecular vibrations and some energetic rotations have energy level spacings that correspond to energies in the IR region of the electromagnetic spectrum (most rotations are in the microwave range which runs between thermal IR and radio wavelengths). Thus IR radiation absorbed by molecules causes increased vibration. Collisions between these energized molecules and others in the sample transfer energy among all the molecules, which increases the average thermal energy and, hence, raises the temperature. Conversely, molecules that emit IR radiation lose their vibrational energy and their collisions with other molecules decrease the average thermal energy and lower the temperature.

The atmosphere is, of course, actually “wet” and may contain several percent water vapor, as well as liquid and solid water in clouds, from the natural water cycle (evaporation-condensation-precipitation). Water vapor also has strong absorptions in the IR, as shown in the schematic diagram. The majority of atmospheric warming is due to these absorptions by water vapor that occur at both ends of the thermal IR region.

Note that there is a “window” in the water vapor spectrum from about 8 to 15 μm where there is little IR absorption and hence little contribution to atmospheric warming. The strong absorption by CO2 at the long wavelength end of this region narrows this window a bit and adds to the warming effect. Radiation from the Earth that is in this window region passes through the atmosphere with little absorption and contributes little to atmospheric warming. Other gases that absorb in this window region or in other narrower window regions of the thermal IR (where water vapor and CO2 do not absorb appreciably) can make significant relative contributions to atmospheric warming by absorbing energy that would otherwise be lost to space.

GWPs for water vapor and tropospheric O3 are not calculated because their atmospheric lifetimes are only days long and their concentrations highly variable. More to the point for water vapor is that human activities have almost no direct influence on its tropospheric concentration, which is controlled by the temperature of the atmosphere and the liquid water from which it evaporates (or ice from which it sublimes). Rising planetary temperature increases the amount of water vapor in the atmosphere, which increases its warming effect. This is a feedback mechanism that adds substantially to the radiative forcings of the other non-condensable greenhouse gases.