Recent changes in high-latitude glaciers, ice caps and ice sheets

Transcription

1 Recent changes in high-latitude glaciers, ice caps and ice sheets Waleed Abdalati NASA/Goddard Space Flight Center, Greenbelt, Maryland The glaciers and ice sheets of the world contain enough ice to raise sea-level by approximately 70 m if they were to disappear entirely, and most of this ice is located in the climatically sensitive polar regions. Fortunately changes of this magnitude would probably take many thousands of years to occur, but recent discoveries indicate that these ice masses are responding to changes in today's climate more rapidly than previously thought. These responses are likely to be of great significance, primarily in terms of their implications for sea-level, but also in terms of how their discharge of freshwater, through melting or calving, may impact ocean circulation. For millions of years, oceans have risen and fallen as the Earth has warmed and cooled and ice on land has shrunk and grown. Today is no different in that respect, as sea-levels have been rising at a rate of nearly 2 mm yr 1 during the last century (Miller and Douglas 2004), and 3 mm yr 1 in the last 12 years (Leuliette et al. 2004). What is different today, however, is that tens perhaps hundreds of millions of people live in coastal areas that are vulnerable to changes in sea-level. Rising seas erode beaches, increase flood potential and reduce the ability of barrier islands and coastal wetlands to mitigate the effects of major storms and hurricanes. The costs associated with a onemetre rise in sea-level are estimated to be in the hundreds of billions of dollars in the United States alone. The worldwide costs in human terms would be far greater as some vulnerable low-lying coastal regions would become inundated, especially in poorer nations that do not have the resources to deal with such changes. Such considerations are particularly important in light of the fact that a one-metre rise in sea-level is not significantly outside the 0.09 to 0.88 m range of predictions for this century (IPCC 2001) and rises of this magnitude have occurred in the past in as little as 20 years (Fairbanks 1989). While the expansion of the warming oceans is estimated to have contributed about a third of the observed twentiethcentury sea-level rise (Miller and Douglas 2004), the greatest potential for significantly increasing sea-level lies in the Greenland and Antarctic Ice Sheets. For different reasons, each exhibits characteristics that suggest they are potentially unstable. In West Antarctica, most of the ice cover rests on a soft bed that lies below sea-level, making it vulnerable to runaway retreat. The Greenland Ice Sheet experiences considerable melt, which has the potential to accelerate rapidly the flow of ice toward the sea. While smaller ice masses, such as the Alaskan Glaciers and the Patagonian Ice Fields, do not have anywhere near the same potential to impact sea-level as the vast ice sheets do, many are shrinking rapidly (e.g. Arendt et al. 2002; Rignot et al. 2003) posing a significant near-term threat. To understand how ice sheets and glaciers may contribute to sea-level in the next 100 years, it is important to understand what controls the balance of mass of a glacier or ice sheet. Mass balance refers to the difference between those factors that add mass mainly snow accumulation and those that remove mass, such as surface melt, melting beneath floating ice, calving and sublimation. These processes are shown schematically for an ice sheet in Fig. 1. As the climate warms, all of the components of mass loss tend to increase, contributing to a shrinking of glaciers. At the same time however, the amount of snow that falls on the surface also increases, contributing to their growth. For the large ice sheets, it has not been obvious which of these processes will dominate throughout the next century or by how much. As a result there has been considerable interest in understanding the state of balance of the Earth's great ice sheets, as well as that of high-latitude glaciers and ice caps, and their associated contributions to sea-level rise. The last decade has seen tremendous advances in our ability to observe changes in the Earth s ice cover with deployment of new spaceborne and airborne instruments and the advent of key satellite datasets, such as surface elevation changes, melt extent, surface velocities, surface temperatures, etc. These observations, combined with complementary field activities and modelling efforts, continue to provide important new insights into the mechanisms that control mass balance, and they provide some indication of how these processes may behave in the future. Fig. 1 Schematic diagram of an ice sheet with its individual mass balance components 95

2 Recent changes in high-latitude glaciers The Antarctic Ice Sheets In the case of Antarctica (Fig. 2), which contains the ice equivalent of 61 m of sea-level, the West Antarctic Ice Sheet (WAIS) rests on a soft deformable bed that lies below sealevel. As a result, a significant retreat would allow water to fill the ice-bed interface, reducing friction and increasing the tendency of the WAIS (with up to 6 m sea-level equivalent) to discharge its ice to the surrounding seas. For this reason, it is much less stable than its East Antarctic counterpart. More than 4 km thick in some areas, some of this ice is still responding to climate changes of thousands of years ago, and until very recently, it was the prevailing view that the only way ice sheets respond to changes in climate is slowly. However, while changes Fig. 2 RADARSAT mosaic of Antarctica ( Canadian Space Agency; created by K. C. Jezek and others at The Ohio State University s Byrd Polar Research Center). The box and the arrow on the Antarctic Peninsula respectively identify the area and upward orientation of the image shown in Fig. 3. at the ice sheet interior may occur over millennia, near the edges the story is different. Much of Antarctica s perimeter is covered with floating ice shelves or ice tongues that are susceptible to the effects of a warming atmosphere and ocean, which as a result have important implications further into the ice. This vulnerability was made dramatically apparent in 2002, when in a matter of weeks, much of the Larsen B Ice Shelf collapsed (Fig. 3). Prior to its collapse, Larsen B was hundreds of metres thick, and is believed to have existed for more than years, yet it disappeared in what can be considered in glaciological terms to be an instant. The collapse of this ice shelf alone is not of consequence to sea-level, since this was ice that had been floating, but it offers a dramatic example of the sensitivity of this ice cover to its changing environment. What occurred in the months and year that followed the break-up, however, was similarly dramatic and of direct consequence to sea-level. Scambos et al. (2004) reported that all of the glaciers that flowed into what was the Larsen B Ice Shelf, had accelerated, some by as much as a factor of 8. As a result, these glaciers thinned rapidly one by 38 m in six months as they increased their discharge to the sea. At the same time, glaciers that feed the remnants of the Larsen that did not collapse showed no sign of acceleration. This experiment of nature helped confirm a disputed hypothesis from several decades earlier that ice shelves act to buttress a glacier and restrain its flow (Mercer 1978). Once that barrier is removed, as was the case with the Larsen B, the buttressing effect is diminished and the ice accelerates its flow to the (a) (b) Fig. 3 Images of the Larson B Ice Shelf (a) from 31 January 2002 prior to its collapse and (b) from 7 March 2002, after its collapse. The images are from the MODIS instrument on NASA s Aqua satellite. 96

3 sea. This faster flow is not necessarily sustained, however, as over time these glaciers adjust to their new boundary conditions and approach a new state of equilibrium. The location of the Larsen B Ice Shelf on the Antarctic Peninsula provides an extreme example, because the peninsula has warmed several degrees Celsius in the latter part of the twentieth century (Vaughan et al. 2004), and many of its glaciers and ice shelves are retreating (Cook et al. 2005). This warming is greater than that over the rest of the continent, most of which has experienced cooling in recent decades. Still, we see signs of this behaviour of retreat followed by acceleration and thinning elsewhere on the ice sheet, most notably in the areas around Thwaites and Pine Island Glaciers that drain into the Amundsen Sea (Thomas et al. 2004). Often referred to as the weak underbelly of the WAIS for its potential instability, this area has been estimated to contribute as much as 0.13 to 0.24 mm yr 1 to sea-level (Shepherd et al and Thomas et al respectively), which is as much as 10% of the total estimated rate of sea-level rise. The reason has been an acceleration of the glaciers that flow into the Amundsen Sea, most likely in response to a weakening (thinning and retreating) of the ice shelf caused by warm ocean currents that are about 0.5 C above freezing (Shepherd et al. 2004). What is happening in this region is not as dramatic as what was observed in the vicinity of the Larsen B Ice Shelf, but it is of greater consequence, given the magnitude of the sealevel contribution and potential instability of the region. If the ice shelves disappear and glacier retreat proceeds further into the basin, these glaciers could continue to accelerate, further increasing their sea-level contribution. Not all of the West Antarctic outlet glaciers and ice streams are accelerating however. Some have slowed and even stopped, most notably the Kamb Ice Stream (formerly Ice Stream C) that feeds into the Ross Ice Shelf. As a result, some areas of the WAIS are growing, and actually slowing the current rate of sea-level rise (Table 1). At the equivalent of about 0.07 mm yr 1 of sea-level rise (SLR) (Joughin and Tulaczyk 2002), however, the net growth of ice streams like the Kamb is not enough to offset the losses observed elsewhere on the WAIS. Zwally et al. (2005) conducted a comprehensive analysis of satellite radar altimetry measurements for the period from the European Remote Sensing Satellites (ERS-1 and ERS-2) to determine the total mass balance of both the West and East Antarctic Ice Sheets. On the WAIS, they report an average surface lowering of 2.9 cm yr 1 dominated by the significant thinning in the Thwaites/Pine Island region. After adjusting for crustal uplift and firn Table 1 Contributions of ice sheets and some recently surveyed glaciers and ice caps to sea-level rise based on observations of the last decade. Where estimates exist for more than one time period, the earlier is shown on the left and the most recent is on the right. Ice covered region Sea-level input (mm yr 1 ) Ross Ice Shelf Ice Streams (1997) Amundsen Embayment ( ) ( ) West Antarctica ( ) East Antarctica ( ) East Antarctica ( ) All of Antarctica 3* ( ) All of Antarctica 5* +0.4 ( ) Greenland (1993/4 1998/9) ( ) Greenland (1996) (2000) (2005) Greenland ( ) Greenland Average Alaskan Glaciers (mid 50s mid 90s) ( ) Canadian Ice Caps ( ) Patagonian Ice Fields (1968/ ) ( ) 1 Joughin and Tulaczyk (2002); 2 Shepherd et al. (2004) and Thomas et al. (2004) for and respectively; 3 Zwally et al. (2005); 4 Davis et al. (2005); 5 Velicogna and Wahr (2006); 6 Krabill et al. (2004); 7 Rignot and Kanagaratnam (2006); 8 Arendt et al. (2002); 9 Abdalati et al. (2004); 10 Rignot et al. (2003). * The differences between the earlier and later values do not necessarily indicate an acceleration of mass loss, but rather are a result of both the differences in the time period covered, and the differences in the techniques used. compaction, they estimate the WAIS mass balance for the 9-year period to be 51 km 3 yr 1 of ice (0.13 mm yr 1 SLR). In the case of the vast East Antarctic Ice Sheet, the Zwally et al. (2005) analysis reveals an increase in surface height of 0.8 cm yr 1, which corresponds to a gain in ice of 18 km 3 yr 1 of ice ( mm yr 1 SLR). A similar analysis was conducted by Davis et al. (2005) using the ERS-1 and ERS-2 dataset for They estimated the East Antarctic Ice Sheet growth exclusive of the margins to be 49 km 3 yr 1 of ice ( 0.12 mm yr 1 SLR). While Davis et al. attribute this growth to recent accumulation anomalies, there is evidence to suggest that some of this growth may be a continued response to increases in accumulation that began in the early Holocene (Joughin and Bamber 2005). Despite the difference in magnitude between the two estimates, they are both consistent in finding growth in the East Antarctic that offsets some of the losses in the WAIS. Taking Zwally s numbers for East and West Antarctica together yields an estimated net loss for Antarctica of 33 kg 3 yr 1 (0.084 mm yr 1 SLR). Most recently, Velicogna and Wahr (2006) used time-variable gravity measurements from the Gravity Recovery and Climate Experiment (GRACE) for the period from April 2002 through August They estimate a rate of mass loss from summer 2002 to summer 2005 of 152 km 3 yr 1 of ice. However, with uncertainties in excess of 50% and a very wide range of sub-annual and sub-seasonal variability, great caution needs to be exercised when interpreting these three-year results in the context of longer-term changes. The Greenland Ice Sheet With a total volume of 2.25 km 3 and an average thickness of 1.6 km, the Greenland Ice Sheet is roughly one tenth the size of the Antarctic Ice Sheet, and it contains the equivalent of seven metres of sea-level. Unlike Antarctica, however, much of this ice sheet experiences melt. It is this melting of floating ice shelves and ice tongues around the perimeter and of snow and ice in the interior that can create a potentially unstable state of balance in Greenland. This melting has been increasing. From 1979 through 2005, the period for which we have the satellite data that form the basis for determining ice sheet melt extent (Fig. 4), the average melt area during the April October time period, increased by 31% (Steffen and Huff 2005). This strongly positive 26-year melt trend is consistent with the overall temperature trend that suggests the Arctic has warmed at twice the rate of the Earth as a whole in the past few decades (ACIA 2004). These higher temperatures contribute to the melt increases, and because melting darkens the surface, a positive feedback mechanism is established, whereby the melt process is self-compounding. The satellite-derived melt results are also consistent with modelled surface mass balance estimates of Hanna et al. Recent changes in high-latitude glaciers 97

4 Recent changes in high-latitude glaciers 98 (a) (b) Fig. 4 (a) Average April October melt extent on the Greenland ice sheet showing a melt trend of 1.2% per year. (b) Maximum spatial extent of satellite derived melt during 2005 (Both Figs. (a) and (b) are from Steffen and Huff 2005). Pixels shown in red are pixels that did not exhibit melt during the satellite melt record. Also identified in the figure is the Jakobshavn Isbrae area shown in more detail in Fig. 6. (2005). Their comparisons between the periods and indicate that melt rates and associated runoff did increase in the later years, as did precipitation. The net effect was that the sea-level contribution from Greenland s surface mass balance was 0.15 mm yr 1 greater than that of the earlier period. Apart from contributing directly to the loss of ice through the runoff of melt water, this melt also has the potential to contribute to ice loss through other, perhaps more substantial, means. Data from the equilibrium line on the western flank of the Greenland Ice Sheet (the area where net losses from melt and flow are equal to gains from accumulation) indicate that surface melt water penetrates through crevasses or moulins (Fig. 5(a) front cover, Fig. 5(b)) through over a kilometre of ice, where it lubricates the interface between the ice and the bedrock, causing a summertime acceleration (Zwally et al. 2002). This summer acceleration hastens the flow of ice toward the edges of the ice sheet, where it either is discharged to the sea or melts. Over time, this phenomenon can change the shape of an ice sheet, causing a lowering of these regions making them more susceptible to future melt (Parizek and Alley 2004). Many of Greenland s outlet glaciers are tidewater glaciers that flow directly into the sea, where they terminate with floating ice tongues that may be up to tens of kilometres in length. These ice tongues experience melt at the surface during the spring and summer months, and from underneath as a result of their contact with relatively warm seawater. As with Antarctica, when these floating tongues retreat or collapse, the associated elimination of their buttressing effect allows the glaciers that feed them to accelerate, causing these glaciers to thin very rapidly in some cases. One place in Greenland in which this has been vividly observed is on the Jakobshavn Isbrae. While observations of many of Greenland s outlet glaciers through the midlate 1990s showed general thinning, the Jakobshavn Ice Stream, which at about 7kmyr 1 had been one of the fastest glaciers in the world, actually thickened slightly. This thickening appears to have been a direct result of a slowing down of the ice stream in the mid-1990s (Joughin et al. 2004). Since 1997 however, the floating ice tongue began to thin substantially, as air and water temperatures in the vicinity increased (Thomas et al. 2003). This thinning weakened the ice tongue, which, after remaining in a stable location for roughly 50 years, began retreating in 2001 at several kilometres each year until reaching its current 2005 location (Fig 6). The thinning and retreat were accompanied by successive accelerations of the ice stream such that by 2003, the flow rate had nearly doubled, reaching 12.6 km yr 1 (Joughin et al. 2004). Associated with these velocity increases, have been rapid ice stream thinning rates of about 15 m yr 1 (Krabill et al. 2004) This retreat has continued through 2005 and the full implications are not yet known, as there are limits to how fast the ice stream can flow. However, the rapid thinning that began in 1997 has propagated inward, as the ice stream continues to lose mass and draw down ice from further upstream and the surrounding vicinity (Thomas et al. 2003). On the east coast, the flow rates on the Kangerdlugssuaq Gletscher, which thinned in the mid-to-late 1990s by more than 10 m yr 1 (Thomas et al. 2000), have increased by as much as 210% between 2000 and This acceleration suggests a thinning of the glacier by up to 50 m yr 1 (Rignot and Kanagaratnam 2006). While Jakobshavn and Kangerdlugssuaq are among the more extreme cases, there is evidence that many of Greenland s outlet glaciers were shrinking in the 1990s as a result of flow rates that exceed that which is needed to balance the difference between surface accumulation and ablation (Abdalati et al. 2001). The situation has become more extreme as outlet glaciers throughout much of southern Greenland, particularly in the east, have shown remarkably widespread and very dramatic acceleration from 1996 to 2000 to 2005 (Rignot and Kanagaratnam 2006). These dynamic effects are causing the ice sheet margins to shrink significantly.

5 Fig. 5 (b) Schematic of pathways that transport water from the surface to the bed (from Zwally et al., 2002). GPS measurements at the surface by Zwally et al. revealed a seasonal increase in velocity (V) during summer. increases in accumulation (i.e. approximately in the last decade or two) rather than a longer-standing imbalance, then the Zwally et al. estimate becomes close to zero or slightly negative. A third approach, using airborne laser altimetry measurements from flights made over parts of the ice sheet during the spring of most years between 1993 and 2005, yields estimates in between of 59 km 3 yr 1 (0.15 mm yr 1 SLR) of ice during the mid to late 1990s, and 80 km 3 yr 1 (0.2 mm yr 1 SLR) between 1997 and 2003 (Krabill et al. 2004). These seemingly large discrepancies result from: a) differences in the time periods examined, b) differences in the approaches used (each with its strengths and limitations), and c) the underlying assumptions about accumulation, ablation, firn compaction and dynamics employed in each method. The differences between these three approaches offer important insights into various aspects of the ice sheet behaviour and collectively are painting a picture of the Greenland Ice Sheet mass balance. That picture is one of an ice sheet that is behaving as would be expected in a warming climate. The high and cold interior is growing as warm temperatures enhance precipitation potential and the margins are shrinking substantially as melt and flow rates increase. While there may be disagreement on the relative magnitudes of these competing phenomena, the magnitude and scale of the recent outlet glacier accelerations, coupled with the increasing melt and its associated effects on flow rates, indicate that the Greenland Ice Sheet is likely to be a major contributor to near-term sea-level rise. Recent changes in high-latitude glaciers Fig. 6 Landsat image of the Jakobshavn Isbrae and Fjord showing calving front locations since Locations from 1953 and earlier were adapted from Weidick (1995). Others were derived from satellite imagery. At the same time, however, the interior of the ice sheet has grown thicker by a few cm yr 1 (Johannessen et al. 2005; Zwally et al. 2005). While this seems very small compared to the tens-of-metre rates of thinning at the margins, it occurs over a very large area, and significantly offsets the loss of ice due to melt and calving/discharge. These competing effects have led to a wide range of mass balance estimates from the Greenland Ice Sheet (Table 1). The lowest (most negative) of these is 224 km 3 yr 1 of ice (0.57 mm yr 1 SLR) in 2005, which was derived from comparing estimates of the net surface mass input (difference between accumulation and surface ablation) to the estimated discharge calculated by satellite observations of outlet glacier velocities and airborne measurements of their thicknesses (Rignot and Kanagaratnam 2006). Rignot and Kanagaratnam also estimated values of 91 km 3 yr 1 (0.23 mm yr 1 SLR) and 138 km 3 yr 1 (0.35 mm yr 1 SLR) for 1995 and 2000 respectively, suggesting that there has been a dramatic increase in net ice loss due to enhanced discharge over the last decade. The highest estimates are those derived using ERS-1 and ERS-2 radar altimetry measurements of ice sheet surface elevation changes combined with aircraft laser altimetry observations near the ice sheet margins, where conventional radar altimetry does not work. Using this approach Zwally et al. (2005) estimate a slight growth of the Greenland ice sheet between 1992 and 2002 of +12 km 3 yr 1 of ice ( 0.03 mm yr 1 SLR) (Table 1). If the interior growth is considered to be a result of only very recent High-latitude glaciers and ice caps The world s glaciers and ice caps are estimated to contain the equivalent of 0.5 m of sea-level (IPCC 2001), far less than the colossal Greenland and Antarctic Ice Sheets, making them less of a threat in the longterm. However, because glaciers and ice caps are smaller and in many cases warmer than ice sheets, their responses to a changing climate may be more rapid. In that sense, their potential to impact sea-level in the near term may be more significant than that of the large ice sheets. Quantitative estimates of the mass change of these ice masses are limited, but there have been some elevation and in situ measurements that provide important information about their current state of balance in high latitudes. In the late 1990s, airborne surveys were carried out on many of Alaska s glaciers, and these data were compared to ground surveys from the mid- 1950s (Arendt et al. 2002). The results showed that in the four decades leading up 99

6 Recent changes in high-latitude glaciers 100 to 1995, the glaciers lost 57 km 3 of ice per year (0.14 mm yr 1 SLR), but between 1995 and 2000/01, that rate of ice loss nearly doubled to 105 km 3 yr 1 (0.27 mm yr 1 SLR), making Alaskan glaciers one of the most significant contributors to current sea-level rise (Arendt et al. 2002). In the summer of 2005, repeats of the original airborne surveys were conducted, which will provide important information in the near future as to whether the rate of ice loss has slowed, accelerated or remained the same. Similar surveys of the Canadian Ice Caps indicate that they too have shrunk in volume such that their contribution to sealevel was mm yr 1 during the late 1990s. This is only one quarter that of Alaska, but like Alaska, the rate appears to have accelerated in the late 1990s (Abdalati et al. 2004). As with Greenland, the diminishing ice is generally found in the low-lying regions that experience melt. The higher elevation accumulation zones show little or no change during the survey period. The largest ice losses in Canada, however, appear to be from the ice caps on Baffin Island, where the temperatures were not especially warm during the 1990s. Thus, these changes are likely a continued response to the end of the Little Ice Age, which ended during the nineteenth century. Additional repeat surveys of some of these ice caps were also conducted in spring 2005, with more planned for The time period was significantly warmer around Baffin Bay than the time period, so we expect ice losses to have increased in this area. Mass balance measurements are very sparse in the Russian Arctic, but the few that do exist suggest slightly negative mass balance (Dowdeswell and Hagen 2004). The notion of net ice loss in these areas is further supported by observations of ice front retreat in the Russian Arctic Islands (Glasovskiy 1996). In eastern Svalbard, there has been noticeable growth on Austfonna, the largest ice cap in the Eurasian Arctic. This growth is likely a result of increased accumulation caused by shrinking sea ice cover in the region (Bamber et al. 2004). On the whole, the ice caps and glaciers in the northern high latitudes are clearly showing a net loss of ice that is quite substantial (Dyurgerov and Meier 2005). While in parts of Alaska and the Canadian Arctic, the rate of loss increased in the mid-to-late 1990s, a number of locations in the Arctic show little change in the magnitude of negative balance and even hint at a slight trend toward less negative balance, though this 'trend' is not statistically significant (Dowdeswell and Hagen 2004). As a result, there is no clear evidence of an Arctic-wide acceleration of wastage (Braithwaite, 2006), as Arctic warming tends not only to increase surface melt, but also precipitation. In the southern hemisphere, the major high-latitude ice cover outside of Antarctica is the Patagonian Ice Fields. Comparisons between surface field measurements and data from NASA s Shuttle Radar Topography Mission (SRTM) by Rignot et al. (2003) show that accelerating high-latitude ice loss is not only limited to some locations in the northern hemisphere. From 1968/75 (the periods of surface measurements) to 2000, the ice fields lost the sea-level equivalent of mm yr 1, but from 1995 to 2000, that amount increased to mm yr 1. Their wastage is not of the same magnitude as the Alaskan glaciers, but relative to their size, they are shrinking much more rapidly. Summary and conclusion Clearly the Earth s high-latitude glaciers, ice caps, and ice sheets are changing significantly. The satellite, aircraft, and in situ data from the last decade have revealed that the vast ice sheets respond very rapidly to changes in climate. These are evidenced by a) the acceleration and shrinking of glaciers that fed the collapsed Larsen B Ice Shelf, b) the recent widespread acceleration and thinning of many outlet glaciers in southern Greenland, some of which is quite dramatic, and c) the seasonal acceleration in areas of melt on Greenland that can only be explained by the penetration of melt water. The associated sea-level contributions are compounded by significant shrinking of Alaskan Glaciers, Canadian Ice Caps, and the Patagonian Ice Fields, and these rates of change appear to have accelerated in each case at the end of the last century. At the same time, however, there has been growth in parts of the ice sheets, most notably east Antarctica and central Greenland, as a result of relatively high accumulation rates. The slowing of some ice streams in West Antarctica has also contributed to ice sheet growth. The question remains: which effects those related to ice loss or those related to gain dominate on the ice sheets? In Greenland, most, but not all estimates suggest that the ice sheet is shrinking, and if the recent acceleration of outlet glaciers is sustained or expands, ice loss in Greenland will probably become quite significant. In Antarctica, losses in the west more than offset gains in the east resulting in an overall ice loss. Moreover, if Antarctica s climatically sensitive fringes experience behaviour similar to that of Greenland s outlet glaciers, as has already been observed in the Peninsula and the Thwaites/ Pine Island region, it too will likely increase its contributions to sea-level rise. It is essential to recognise that the processes that govern growth are generally slow: accumulation of relatively low-density snow in these vast polar deserts, where the annual snowfall is in many places less than a few tenths of a metre of water equivalent (much less in East Antarctica), and the slowing down of ice streams. The processes that cause the ice to shrink can happen much more rapidly, in particular accelerating flow rates and to a lesser extent enhanced surface melt with its positive albedo feedbacks. As a result, contributions of ice sheets to sea-level rise are potentially more dramatic than their potential to slow or reverse that rise. In view of the societal vulnerability and the sensitivity of polar regions to climate change, understanding how these high-latitude glaciers and ice sheets are changing is of critical importance. Acknowledgements I would like to thank Edward Hanna and one anonymous reviewer for their constructive and insightful comments. I would also like to thank H. J. Zwally for providing an advanced copy of his Zwally et al. (2005) manuscript and participating in lengthy discussions about its content. Various figures were graciously provided by K. Farness, R. Huff, K. Jezek, K. Steffen and H. J. Zwally. Support for this work was provided by NASA s Cryospheric Sciences Program and Goddard Space Flight Center. References Abdalati, W., Krabill, W. B., Frederick, E. B., Manizade, S., Martin, S. C., Sonntag, J., Swift, R., Thomas, R. H., Yungel, J. and Koerner, R. (2004) Elevation changes of ice caps in the Canadian archipelago. J. Geophys. Res. Earth Surface, 109, No. F4, F00407 Abdalati, W., Krabill, W. B., Frederick, E., Manizade, S., Martin, C., Sonntag, J., Swift, R., Thomas, R., Wright, W. and Yungel, J. (2001) Near-coastal thinning of the Greenland ice sheet. J. Geophys. Res. Atmos., 106, No. D24, pp. 33, , 742 ACIA (2004) Impacts of a warming Arctic. Cambridge University Press Arendt, A. A., Echelmeyer, K. A., Harrison, W. D., Lingle, C. S. and Valentine, V. B. (2002) Rapid wastage of Alaska glaciers and their contribution to rising sea-level. Science, 297 (5580), pp Bamber, J., Krabill, W., Raper, V. and Dowdeswell, J. (2004), Anomalous recent growth of part of a large Arctic ice cap: Austfonna Svalbard, Geophys. Res. Lett., 31, (12), L12402 Braithwaite, R. J. (2006) Mass balance characteristics of Arctic glaciers, Annals of Glaciology, 42 (In Press) Cook, A. J., Fox, A. J., Vaughan, D. G., and Ferrigno, J. G. (2005) Retreating glacier fronts on the Antarctic Peninsula over the past half-century. Science, 308, (5721), pp Davis, C. H., Li, Y., McConnell, J. R., Frey, M. M. and Hanna, E. (2005) Snowfalldriven growth in East Antarctic ice sheet mitigates recent sea-level rise. Science, 308, (5730), pp

7 Dowdeswell, J. A. and Hagen, J. O. (2004) Arctic ice caps and glaciers. In: Bamber, J. L. and Payne, A. J. (Eds.) Mass Balance of the Cryosphere, Cambridge University Press Dyurgerov, M. B. and Meier, M. F. (2005) Glaciers and the Changing Earth System: a 2004 Snapshot. INSTAAR Occasional Paper 58 Fairbanks, R. A. (1989) A year glacio-eustatic sea-level record Influence of glacial melting rates on the Younger Dryas event and deep-ocean circulation. Nature, 342, (6250), pp Glasovskiy, A. F. (1996). Russian Arctic. In: Jania, J. and Hagen, J. O. (Eds.) Mass Balance of Arctic Glaciers. International Arctic Science Committee Report No. 5 Hanna, E., Huybrechts, P., Janssens, I., Cappelen, J., Steffen, K. and Stephens, A. (2005) Runoff and mass balance of the Greenland ice sheet: , J. of Geophys. Res. Atmos., 110, (D13), D13108 IPCC (2001) Climate change 2001: The Scientific Basis, Cambridge University Press Johannessen, O. M., Khvorostovsky, K., Miles, M. W. and Bobylev, L. P. (2005) Recent ice sheet growth in the interior of Greenland, Science, 310, (5750), pp Joughin I. and Tulaczyk, S. (2002) Positive mass balance of the Ross Ice Streams West Antarctica. Science, 295, (5554), pp Joughin, I., Abdalati, W. and Fahnestock, M. (2004) Large fluctuations in speed on Greenland s Jakobshavn Isbrae glacier. Nature, 432, pp Joughin, I. and Bamber, J. L. (2005) Thickening of the ice stream catchments feeding the Filchner-Ronne Ice Shelf, Antarctica. Geophys. Res. Lett., 32, L17503 Krabill, W., Hanna, E, Huybrechts, P., Abdalati, W., Cappelen, J., Csatho, B., Frederick, E., Manizade, S., Martin, C., Sonntag, J., Swift, R., Thomas, R. and Yungel, J. (2004) Greenland Ice Sheet: Increased coastal thinning. Geophys. Res. Lett., 31, L24402 Leuliette, E. W., Nerem, R. S. and Mitchum, G. T. (2004) Results of TOPEX/Poseidon and Jason calibration to Construct a Continuous Record of Mean Sea-Level. Marine Geodesy, 27, No. 1 2, pp Mercer, J. H. (1978) West Antarctic ice sheet and CO2 Greenhouse effect; a threat of disaster, Nature, 271, pp Miller, L. L. and Douglas, B. C. (2004) Mass volume contributions to twentiethcentury global sea-level rise. Nature, 428, (6981), pp Parizek, B. R. and Alley, R. B. (2004) Implications of increased Greenland surface melt under global-warming scenarios: ice-sheet simulations. Quaternary Sci. Rev., 23, (9-10), pp Rignot, E. and Kanagaratnam, P. (2006) Changes in the velocity structure of the Greenland ice sheet. Science, 311, (5763), pp Rignot, E., Rivera, A. and Casassa, G. (2003) Contribution of the Patagonia Icefields of South America to sea-level rise. Science, 302, (5644) pp Scambos, T. A., Bohlander, J. A., Shuman, C. A., and Skvarca, P. (2004) Glacier acceleration and thinning after the ice shelf collapse in the Larsen B embayment, Antarctica, Geophys. Res. Lett., 31, L18402 Shepherd, A., Wingham, D., and Rignot, E. (2004) Warm ocean is eroding the West Antarctic Ice Sheet, Geophys. Res. Lett., 31, L23402 Steffen, K., and Huff, R. (2005) cires.colorado.edu/science/groups/ steffen/greenland/melt2005/ Thomas, R. H., Abdalati, W., Akins, T., Csatho, B., Frederick, E., Gogineni, S., Krabill, W. B., Manizade, S. and Rignot, E. (2000) Substantial thinning of a major east Greenland outlet glacier. Geophys. Res. Lett., 27, No. 9, pp Thomas, R. H., Abdalati, W., Krabill, W. B., Manizade, S. and Steffen, K. (2003) Investigation of surface melting and dynamic thinning of Jakobshavn Isbrae, Greenland. J. Glaciol., 49, No. 165, pp Thomas, R., Rignot, E., Cassassa, G., Kanagaratnam, P., Acuña, C. Akins, T., Brecher, H., Frederick, E., Gogineni, P., Krabill, W., Manizade, S., Ramamoorthy, H., Rivera, A., Russel, R., Sonntag, J., Swift, R., Yungel J. and Zwally, J. (2004) Accelerated sea-level rise form West Antarctica, Science, 306, (5964) pp Vaughan, D. G., Marshall, G. J., Connolley W. M., King, J. C., and Mulvaney, R. (2004) Climate change Devil in the detail. Science, 293, (5536), pp Weidick, A. (1995) Satellite Image Atlas of Glaciers of the World: Greenland, United States Geological Survey, professional paper 1386-C, (Eds.) R.S. Williams, J. Ferrigno, U.S. Government Printing Office, Washington, DC Velicogna, I. and Wahr, J. (2006) Measurements of time variable gravity show mass loss in Antarctica. Science, DOI: 10/1126/Science (online through Science Express) Zwally, H. J., Abdalati, W., Herring, T., Larson, K., Saba, J. and Steffen, K. (2002) Surface melt acceleration of Greenland ice sheet flow. Science, 297, No. 5579, pp Zwally, H. J. Giovinetto, M. B., Li, J. Cornejo, H. G., Beckley, M. A., Brenner, A. C., Saba, J. L., and Yi, D. (2005) Mass changes of the Greenland and Antarctic ice sheets and shelves and contributions to sea-level rise J. Glaciol., 51, (175) pp Correspondence to: Dr Waleed Abdalati, Cryospheric Sciences Program, NASA/Goddard Space Flight Center, Greenbelt, MD waleed.abdalati@nasa.gov Royal Meteorological Society, 2006 doi: /wea Recent changes in high-latitude glaciers Lion s head in snow Glaciated highlands with low cloud filling the valleys, South Georgia, January 2004 Jeff Brown 101