the released fresh water and air bubbles form a turbulent boundary layer along both ice and seawater and rise through a series of thermal terraces. These carved subsurface facets expose yet more ice surfaces and encourage still faster melting. If sufficient meltwaters accumulate on the top, they may flow down the sides to give the berg a fluted appearance. But the roughened bottom, gouged by large crevasses, disintegrates most rapidly. The fissures widen and rise; the berg thins, making it more susceptible to wave-induced flexing; and the berg itself calves.
Of the two processes, melting ultimately triumphs over mechanical disintegration. Melting prepares the ice mass for ruptures, large and small, and unlike breakage it reduces the volume of the ice. Melting is the final solution: ice is no longer reformed into ice but transformed into water, a change of state that will remove it from the ice field completely. Yet the mechanical processes assist melting by increasing the proportion of the total ice that is exposed to thermal activity. Some ice spalls off the sides. Some is mechanically eroded by waves—melting into exposed pinnacles that quickly rot and into terraces that, as overhanging cliffs, soon fail and drop. The thinning of the berg encourages rupture by allowing the ice mass to flex amidst long ocean swells. Some disintegration follows from simple collision, especially where a grounded berg is struck by a free-floating one. Grounded bergs, in fact, are a prominent source of brash ice. Differential heating—the sharp contrast between cold ice and warmer sea—can lead to thermal spalling, with chunks and slabs of ice breaking free like exfoliating granite and sandstone. The permeability boundary between firn and glacial ice, a zone of potential penetration by brine, may lead to the large-scale slumping of firn, a process of mass wasting.
The marks of the strain that produced calving will persist for a time, although fissures will slowly heal shut, and some of the fractures may become zones of weakness for further mechanical disintegration as the berg experiences a new set of stresses. The intensity of this activity will vary with the size of the berg, and it will be reflected in the berg’s shape. Gigantic bergs—with dimensions measured by tens of kilometers—will not undergo much internal change. Only the edges of the berg will be affected. Smaller bergs, with higher proportions of newly exposed edges, will show proportionately greater change. No longer subjected to a high confining pressure and no longer protected by an enveloping shield of ice, the sides of the ice mass will ablate rapidly.
The berg’s motions, too, are a curious amalgamation of its history and its present state. All the movements that have characterized the ice mass on its journey are present. Past motions are preserved in the internal ice fabric. Current motions are revealed by the gross movement of the berg, as many former stresses vanish and new stresses appear. Its free-floating motions give the berg many of its distinctive characteristics. Unlike other ice masses, the berg will not merely flow internally, within the confines of the rigid ice field, but will respond more or less freely to its new environment of fluids, the sea and the air. The berg will drift in ocean currents and wind fields. It will bob, rock, and spin. It will tilt or even overturn as erosion modifies its density profile. From the simplest of motions, that which governs the settling and compression of snow, the iceberg has acquired an almost limitless mobility. The price paid for this mobility is disintegration. Time itself accelerates; events crowd one upon the other; the more rapidly the berg moves, the more swiftly it decays. The ice began in a nearly timeless state, 15 prolonged over centuries, even millennia, because the extreme cold of the source region slowed movement to a vanishing point. But as the ice acquires composition, shape, movement, variety on its journey outward, it correspondingly wastes away. The smaller the berg, the greater its mobility and the faster its disintegration.
Very large tabular bergs are the least mobile and show the greatest persistence. They cling to the shore, grounded or entrapped in pack ice. The heavy concentration of large bergs near the coast contributes to the preservation of a wide belt of fast ice and shore ice even during the summer. Only when they break up into smaller units and proceed through the pack do bergs respond freely to wind and wave. In broad terms, their drift is set by ocean currents. Because of its large draught, its keel, the berg behaves as a current integrator, sailing roughly eastward with the nearshore flow of the Antarctic circumpolar current and ultimately breaking free of shore influences to enter the broad west wind drift out to the Antarctic convergence. There are considerable variations, however.
The act of breaking loose from the coast is typically not a single event. Several years may pass before the berg actually becomes a free-floating vessel. Frequently, the berg grounds as it attempts to sail from the continental shelf or to pass around peninsulas of ice or land, or as it encounters eddies within the current that trap it temporarily in a frozen gyre of bergs and pack ice. From its first calving the berg may be recaptured several times by the ice field—to be refrozen, reincorporated, then liberated again. The vast majority of large bergs hug the coast, and most of the bergs that survive many years do so because they move ponderously out from the ice shore. The average longevity of an Antarctic iceberg is four to six years. The colossal Trolltunga berg, however, was tracked for over eleven years, slowly creeping along the frozen shoreline of Queen Maud Land before it was absorbed within the Weddell gyre, where it spent more than two years before being catapulted into the South Atlantic. Once it enters more or less open ocean, a berg advances an average of 8–13 kilometers a day.
Other influences shape the actual drift track of the berg. The Coriolis force, pronounced at high latitudes, gives a small northward component to the berg’s circumpolar drift. Tides work to lift grounded bergs and to move others, while winds shape considerably the local pattern of drift. How influential winds become depends on the strength of the wind, the size of the berg, and the proportion of sail to keel. High winds (over 50 kilometers per hour), small bergs, and large sails make for responsive ice. Even large bergs will be sensitive to very high winds and the waves they generate. But the storm cells that orbit the continent follow the general trend of the Antarctic circumpolar current. While, in the short term, wind and current may compete, in the larger scale of things they complement each other. The bergs circle the continent like ice debris in the rings of Saturn.
The perimeter of ice varies with the ebb and flow of the circumpolar current and the general climate. The pack ice remains frozen much of the year, retarding the movement of trapped bergs. At other times, the perimeter of the circumpolar current swells outward and icebergs range widely over the Southern Ocean. Occasionally, very large bergs encounter favorable circumstances that fling them outward well beyond the Antarctic convergence altogether. Bergs have been sighted from South Africa after being hurled out of the Weddell Sea gyre, as well as off the coast of Peru, embedded in the cold Humboldt current.
Ocean swells cause the berg to oscillate (if the ice is rigid) or vibrate (if the ice is elastic). Some of the movement is linear, making the berg bob up and down like a glacial cork. Some is angular, encouraging the berg to rock back and forth. Both oscillations and vibrations, however, set up stress fields within the ice. How the berg responds depends on the wavelength of the swell, the thickness of the berg, and the presence of internal fissures. Very large, thick bergs absorb the vibrations with little effect; the entire berg may slowly bob or rock but there will be little internal deformation. Thinner bergs or ice shelves may simply bend elastically without rupture. But under the proper circumstances—if the wavelength of the swell and the ratio of thickness to width in the berg are in the right proportion—rupture may occur. Should the berg vibrate near its natural frequency, it will shake into large pieces. Flexure and fatigue failure seem to be important processes in the calving of bergs from shelves. For large bergs, especially those with residual cracks that may propagate under the proper stresses, the process of stress failure continues to operate.
The berg even rotates. One rotation is slight but constant, the product of a sheath of meltwater that surrounds the berg as it ablates. This liberated freshwater has a lower density than the surrounding seawater and accordingly it rises. At lower levels, the pressure differential is sufficient to have an impact. Seawater enters the sheath, but under the impress of the Coriolis force, the flow deflects to the left; the berg spins. More dramatic are those rotations from top to bottom of the berg. The density profile of an Antarctic iceberg shows a lighter top of snow and firn riding above a much denser ice substratum. But as the top erodes, as side disintegration results from thermal ablation and the mechanical response to the new stress field, and as the bottom reshapes from crevasse erosion and thermal convection along the sides,