INTRODUCTION:

Dr. Jack Kaye
Director of the Research Division, Office of Earth Science, National Aeronautics and Space Administration (NASA), Washington, DC

SPEAKERS:

Dr. Claire L. Parkinson
National Aeronautics and Space Administration, Goddard Space Flight Center, Greenbelt, MD

Dr. D. Andrew Rothrock
Applied Physics Laboratory, School of Oceanography, University of Washington, Seattle, WA

Dr. Ted Scambos
National Snow and Ice Data Center, University of Colorado, Boulder, CO

Wintertime sea ice covers 15 million square kilometers of the north polar region, an area exceeding one and a half times the area of the U.S. Even at the end of the summer melt season, sea ice still covers 7 million square kilometers. This vast ice cover is an integral component of the climate system, being moved around by winds and waves, restricting heat and other exchanges between the ocean and atmosphere, reflecting most of the solar radiation incident on it, transporting cold, relatively fresh water equatorward, and affecting the overturning of ocean waters underneath, with impacts that can be felt worldwide. Sea ice also has a major influence on Arctic ecosystems, affecting life forms ranging from minute organisms living within the ice to large marine mammals like walruses that rely on sea ice as a platform for resting, foraging, social interaction and breeding.

Since 1978, satellite technology has allowed the monitoring of the vast Arctic sea ice cover on a routine basis. The satellite observations reveal that overall, the areal extent of Arctic sea ice has been decreasing since 1978, at an average rate of 2.7% per decade. Through 1998, the greatest rates of decrease occurred in the Seas of Okhotsk and Japan, and the Kara and Barents Seas, with most other regions of the Arctic also experiencing decreases in the extent of sea ice. The two regions experiencing increases in ice extent over this time period were the Bering Sea and the Gulf of St. Lawrence. Furthermore, the satellite data reveal that the sea ice season shortened by over 25 days per decade in the central Sea of Okhotsk and the eastern Barents Sea, and by lesser amounts throughout much of the rest of the Arctic seasonal sea ice region, although not in the Bering Sea or the Gulf of St. Lawrence. Concern has been raised that if the trends toward shortened sea ice seasons and lesser sea ice coverage continue, this could entail major consequences to the polar and perhaps global climate, and to the lifestyles and survivability of selected Arctic plant and animal species.

At present, it is uncertain whether such decreases in Arctic sea ice will continue. If the basic, underlying cause is global warming, and global warming continues, then further retreat of the Arctic ice cover would be expected. If, however, the basic cause lies elsewhere, then the current trends might well reverse. A possibility along those lines relates to a large-scale atmospheric pressure oscillation termed the North Atlantic Oscillation (NAO). The NAO was in its positive phase, with a strong Icelandic low-pressure system, for much of the late 1980s and early 1990s. This phasing of the NAO can probably explain some of the observed spatial contrasts in the sea ice cover changes. Should the NAO shift to its negative phase for an extended period, one consequence may be shifts in the Arctic sea ice trends as well. A third possibility is that the observed decrease in sea ice reflects a more complex intertwining of causes, including long-term global warming and natural variability operating on shorter timescales.

The sea ice cover in the Arctic Ocean has decreased in thickness over the last several decades, from a little over three meters to under two meters. The ice cover has decreased somewhat more in the eastern portion of the Arctic Ocean, in the direction of Eurasia, than in the western portion, in the direction of Alaska and the Canadian archipelago.

The data that lead to this conclusion came from a number of U.S. Navy nuclear submarines. These submarines carry upward-looking sonar to determine their distance from the underside of floating sea-ice, and pressure sensors to determine the submarine's depth below the base of the ice. The combination of these measurements, in turn, leads to a determination of "ice draft" (the thickness of ice below the water surface) which can then be used to calculate sea-ice thickness. Ice floats mostly submerged and thus, ice thickness is approximately 1.12 times ice draft. The observed losses in ice thickness were recorded in two intervals: 1958 to 1976, and 1993 to 1997, the latter includes data from three SCICEX (Scientific Ice Expedition) cruises. SCICEX was a program in which the U.S. Navy and several non-defense agencies supported six Arctic cruises for civilian science, between 1993 and 1999. At present, there are not enough publicly available submarine data to resolve the temporal record of sea-ice thickness on an annual basis. However, plans are in motion to have more defense-related data made generally available.

The uncertainty in these measurements is about 0.15 to 0.3 m. Thus, the observed decrease in sea-ice thickness is well established. Given that sea-ice thickness has a substantial seasonal cycle (a change in ice thickness from winter to summer of about 1 meter), in order to avoid masking any inter-annual changes by seasonal changes in ice thickness, only data from summers (between July and October) were used. These data, in turn, were slightly adjusted to account for the fact that not all observations were made on the same day of the year. Without such adjustments, the observed decrease in sea-ice thickness would appear even greater. There are several climate-related changes in the Arctic we might look for to explain the change in ice thickness: for instance, changes in atmospheric temperatures and cloudiness and in the heat supplied to the ice by the ocean. The supporting evidence so far lies in a warming of winter surface temperatures and a depletion of riverine water in the eastern latitudes of the Arctic Ocean, and in a changed ice circulation pattern that transports ice out of the Arctic Ocean earlier in its development. Whether a change of the observed magnitude in ice thickness over several decades is part of a natural cycle of climate variability, or represents a mix of human influences on the climate system and natural climate variability, is presently unknown from the perspective of these observations alone.

The last five decades have witnessed a distinct warming trend in the mean annual and summertime temperatures in the Antarctic Peninsula, and it has begun to have a significant impact on the ice in that area. The current climate is approximately 2.5 ^oC (4 ^oF) warmer than the earliest station records, which date back to the 1940s. Sea ice to the West of the peninsula, which both caps relatively warm ocean water and reflects incoming radiation back to space, has decreased by about 20% between 1973 and 1993, and continues to decline in this part of the Antarctic as temperatures have risen. In response to these and perhaps other changes in regional climate, ice shelves in the northern portion of the peninsula have begun to disintegrate.

Ice shelves are plates of floating ice hundreds of meters thick that are attached to land and fed by glaciers. They are good indicators of regional climate trends because they appear to respond to climate change on a decadal scale, rather than fluctuating from year to year. From their flow rates and surface features, one can establish that the ice shelves in the peninsula have existed for several centuries prior to the last two decades. The total area of ice lost to date is about 8000 km^-2, an area larger than Delaware. The retreat of the Larsen B shelf continues, with the loss of an additional roughly 300 km² this season. The causes of the warming, or more specifically, the possible links between this regional warming and the rest of the climate-ocean system, are unclear.

Far from being a slow melting process, ice shelf disintegration is rapid, occurring within a few years to decades, once mean summer temperatures reach 0 ^oC, the melting point. Although this limiting temperature for ice stability was predicted based upon previous studies of ice shelves, the speed of the response to warming was unanticipated. Melting at the ice shelf surface, specifically the accumulation of water in ponds, apparently determines how fast ice shelves crumble. Meltwater, which is denser than the surrounding ice, seems to act like a wedge that seeps into and forces surface cracks deeper, penetrating the full ice thickness and weakening the shelf. Once the shelf is shattered by cracks full of water, wave and wind action disintegrate it into relatively small iceberg slivers.

The consequences of the current breakups in terms of sea-level are negligible, but that is not the case for all ice shelves. Ice shelves float and thus, their breakup does not contribute to sea level rise directly. Instead, ice shelves serve to restrain the flow of ice from the glaciers that feed them. Since the glaciers that feed the peninsula ice shelves are small, a speed-up in flow resulting from removal of these shelves will not significantly affect sea level. However, much larger shelves, such as the Ross and Ronne ice shelves in West Antarctica, are fed by much larger glaciers. As such, these shelves play a much more important role in restraining the flow of these very large glaciers (ice streams). Thus, if a warming trend were to take hold in the vicinity of these shelves, to the point that melt ponds began to form on their surfaces, one might expect that within a few years the shelves would begin to retreat. If the shelf were completely removed (a process that would probably require several decades for shelves as large as the Ross and Ronne), the speedup in flow from the newly-exposed ice streams could lead to a significant increase in the rate of sea level rise (currently about 2 mm per year).

Dr. Claire L. Parkinson is a climatologist at NASA's Goddard Space Flight Center in Greenbelt, Maryland, where she has worked since July 1978. For much of that time, her research has centered on satellite data analysis of sea ice and the role of sea ice in the global climate system. She has also numerically modeled sea ice and has done field work in both the Arctic and Antarctic, most recently as Chief Scientist on an expedition to the North Pole in April 1999. She is lead author of an atlas of Arctic sea ice from satellite data and co-author of two other atlases of sea ice. Beyond her sea ice work, Dr. Parkinson has co-authored a textbook on climate modeling and co-edited a book on satellite observations related to global change. She has also written an introductory text on satellite imagery and a book on the history of western science from 1202 to 1930.

Since 1993, Dr. Parkinson has been Project Scientist for the Earth Observing System's Aqua mission (formerly named EOS PM), scheduled for launch in December 2000; and since 1996, she has been a member of the EOS Science Executive Committee. Dr. Parkinson has a B.A. degree in mathematics from Wellesley College, MA, and a Ph.D. in climatology from Ohio State University.

Dr. D. Andrew Rothrock is a Principal Research Scientist at the Applied Physics Laboratory, and a Research Associate Professor in the School of Oceanography, at the University of Washington. Dr. Rothrock investigates properties of sea ice that affect its balance of mass, heat, and momentum, including ice motion, ice thickness, and floe and lead geometries, focussing primarily on the Arctic Ocean. He has investigated measuring these properties from satellites. He is also interested in estimating geophysical variables by assimilating observations into models. Most recently, Dr. Rothrock has focused his attention on the assessment of sea-ice state and interdecadal change from submarine sonar observations of sea-ice thickness. He has served in an editorial capacity for journals such as the Journal of Geophysical Research and Annals of Glaciology, as chair of several NASA Panels and Working Groups, and as chair of his department.

He received his B.S.E. degree from Princeton University in 1964, and his Ph.D. from the University of Cambridge, UK in 1968.

Dr. Ted Scambos is a Research Scientist for the National Snow and Ice Data Center (NSIDC) at the University of Colorado, where he has worked since 1993. Prior to his arrival at NSIDC, he spent three years at NASA's Goddard Space Flight Center. His research interests include Antarctic glaciology and the application of satellite data to problems in polar science. Dr. Scambos has spent three field seasons in the Antarctic, using ground-based methods to evaluate the recent history of ice streams and the Ross Ice Shelf. More recently, he has used remote sensing techniques to study the links between climate warming and ice shelf breakup, and the relationship of the atmosphere and the snow surface in East Antarctica.

Dr. Scambos received a B.S. in Earth and Space Science from the SUNY (State University of New York) at Stony Brook in 1977, and an M.S. degree in Geology from Virginia Tech in 1980. He received his Ph.D. in Geology from the University of Colorado in 1991, earning the Harry Waldrop graduate student award.