Earth Observations from Space: The First 50 Years of Scientific Achievements (2008)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

7 Cryosphere No other single technological development has revolu- cover (see Chapter 6), detection of earlier spring thaw and tionized cryosphere research as much as satellite observa- associated lengthening of the growing season (Chapter 9, tions. Most of Earth’s frozen regions are remote and access Box 9.4), the new perspective of the dynamic ice streams by land or seas involves often great risks; therefore, conduct- in Antarctica, the decrease in sea ice in the Arctic, and the ing in situ observations is logistically difficult and expensive. change in glacier extent. The synoptic view from satellites increases the data coverage by multiple orders of magnitude, and access is no longer Nonuniform and Dynamic Ice Streams in restricted by seasons. Antarctica Understanding changes to ice sheets, sea ice, ice caps, and glaciers is important for understanding global climate Field exploration of the Antarctic ice sheet is time change and predicting its effects. In particular, “shrinking ice consuming, logistically intensive, costly, and sometimes sheets” and their contribution to sea-level rise were identified dangerous. Prior to satellite observations, spatial coverage as the third most significant “Breakthrough of the Year” for was very sparse: information on the Antarctic ice sheet was 2006 according to Science magazine: acquired slowly over the years by numerous surface traverses (Figure 7.1). Glaciologists nailed down an unsettling observation this year: The world’s two great ice sheets—covering In 1997, Radarsat data were used to create the first Greenland and Antarctica—are indeed losing ice to the complete radar-based map of Antarctica (Figure 7.1). oceans, and losing it at an accelerating pace. Researchers Analyses of radar images from various sensors through don’t understand why the massive ice sheets are proving the years have enabled detailed measurements of surface so sensitive to an as-yet-modest warming of air and ocean velocity, and in turn these measurements have enabled water. The future of the ice sheets is still rife with uncer- calculation of strain rates and basal shear at the bed. tainty, but if the unexpectedly rapid shrinkage continues, For the first time, satellite data revealed the extent of low-lying coasts around the world—including New the ice stream network, leading to the discovery of new Orleans, South Florida, and much of Bangladesh—could ice streams and the ice stream tributaries (Joughin et al. face inundation within a couple of centuries rather than 1999). For example, satellite-based measurements of sur- millennia. face velocity within Antarctic ice streams reveal a com- —Science (2006) plex pattern of flow not apparent from previous measure- ments (Bindschadler et al. 1996). Furthermore, satellite This breakthrough is one of the examples of major observations led to the discovery that ice streams move at accomplishments in cryosphere science presented in this variable speeds, resulting in a more dynamic picture than chapter. Other examples include the change in seasonal snow the previously held view that ice sheets move at a constant velocity (Figure 7.2; Bindschadler and Vonberger 1998).  The first-ranking breakthrough of the year was the proof of the Poincaré Satellites provide improved data collection methods to conjecture, a long-standing problem in mathematics. The second-ranking increase data density and to improve velocity estimates breakthrough was in the area of paleogenomics: the sequencing of Neander- substantially. thal DNA proves that Neanderthal evolution diverged from modern humans at least 450,000 years ago. 58 

CRYOSPHERE 59 a b c FIGURE 7.1  Spatial coverage of data from Antarctica. (a) Surface transverses since the 1957-1958 International Geophysical Year. SOURCE: National Snow and Ice Data Center, University of Colorado. (b) Airborne surveys, over snow radio-echo sounding (RES), seismic surveys, gravimetric surveys, and ice-core missions since the 1957-1958 International Geophysical Year. SOURCE: BEDMAP consortium. (c) Satellite coverage. SOURCE: John Crawford, Canadian Space Agency, National Aeronautics and Space Administration, Jet Propulstion Laboratory. 7-1 a,b,c Accelerating Ice Sheet Flow in Antarctica thought to be determined by the difference between melting and Greenland and precipitation. Satellite observations have revolutionized this thinking One of the central questions in climate change and cryo- by allowing scientists to monitor precise ice sheet elevation, sphere research is how the warming climate will affect the ice velocity, and overall mass. Satellite images revealed that sheets because the amount of continental ice and melt water in fact the overall mass is declining (Luthcke et al. 2006). entering the ocean strongly contributes to the change in sea In addition to observing great variability in the ice stream level. Glaciologists and climatologists have long been debat- velocity over time and space, satellite images revealed that ing whether a warming climate would decrease ice mass. the overall velocities of the ice streams in Antarctica and However, early research focused on how increased melting Greenland have increased during the past decade, resulting would be offset by increased precipitation. The ice mass bal- in more ice flow into the ocean (Bindschadler and Vonberger ance and thus its contribution to sea-level rise was originally 1998, Joughin et al. 2001). 

60 EARTH OBSERVATIONS FROM SPACE: THE FIRST 50 YEARS OF SCIENTIFIC ACHIEVEMENTS 100 km haps from increased basal lubrication by meltwater pen- etrating from the surface. These new discoveries indicate that ice stream dynamics (the balance between the forcing, such as ice thickness and surface slope, and the resistance, such as internal stiffness) are the primary drivers of rapid sea-level change instead of the balance between melting and precipitation. D5 The ability to estimate the overall mass of ice sheets is a remarkable accomplishment of satellite observations. E3 Numerous techniques, including radar images, measurements E2 D4 of surface elevation from laser altimeters, and GRACE’s gravity data, now show that both Greenland and Antarctica D3 have been losing ice over the past 5 to 10 years. From 2003 E1 to 2005, Greenland lost more than 155 gigatons per year at lower elevations and gained about 54 gigatons per year D2 at higher elevations, with most of the losses occurring dur- m/a 700 ing summer (Chen et al. 2006b, Luthcke et al. 2006, Rignot D1 M2 + + DDE 600 and Kanagaratnam 2006, Wahr et al. 2006). In Antarctica 500 the gravity data show mass losses of 70-200 km3 per year DDE + M3 + L4 400 (60-160 gigatons per year). Most of the loss is from West + K3 300 Antarctica, with East Antarctica in approximate balance K4 + 200 (Figure 7.3). 100 0 Declining Arctic Summer Sea Ice FIGURE 7.2  Velocity variations in Antarctica ice streams. Just as miners once had canaries to warn of rising con- SOURCE: Binschadler et al. (1996). Reprinted from the Annals centrations of noxious gases, researchers working on of Glaciology with permission of the International Glaciological climate change rely on arctic sea ice as an early warning Society, copyright 1996. 7-2 system. one column —Arctic Climate Impacts Assessment (2004) For many reasons, observing trends in sea ice reliably has been possible only with the advent of satellite observa- The discoveries of accelerating ice loss from Antarctica tions. Navigating the remote and frozen seas off Antarctica and Greenland and the importance of ice sheet dynamics in or in the Arctic to obtain in situ measurements of sea ice their mass balances rest on measurements by a suite of satellite extent is treacherous, and sea ice extent is highly variable and airborne sensors using novel techniques (Bindschadler in time and space due to wind advection and localized melt- et al. 1996, Chen et al. 2006a, Kerr 2006, Luthcke et al. ing. Before satellite observations became available, spatial 2006, Rignot and Kanagaratnam 2006). These discoveries coverage of sea ice was monitored by tracking the location are possible because of decades of optical and radar images, of the ice edge from ships. Because the ice edge is moving laser and radar altimeters, and more recently the National with winds and ocean currents, it is not a robust indicator of Aeronautics and Space Administration’s (NASA) Gravity basin-scale sea ice extent. Thus, accurate and quantitative Recovery and Climate Experiment (GRACE) mission, which interannual comparisons of the basin-scale ice coverage measures ice mass directly through its gravitational pull. became only possible with the availability of the synoptic In addition, airborne laser altimeter data show thinning of view from satellites. ice near the coastline, radar data show faster flow, Landsat Sea ice has been monitored continuously with pas- data show retreat of the grounding line, and the Moderate sive microwave sensors (Electrically Scanning Microwave Resolution Imaging Spectroradiometer (MODIS) data show R ­ adiometer [ESMR], Scanning Multichannel Microwave calving of large icebergs. Warming ocean waters seem to Radiometer [SMMR], Special Sensor Microwave/Imager have increased calving of the ice shelves, thereby allowing [SSM/I], and Advanced Microwave Scanning ­ Radiometer- the ice sheet’s outlet glaciers to flow more quickly (Box 7.1). Earth Observing System [AMSR-E]) since 1979. Not limited Glaciers in Greenland have also increased in velocity, per- by weather conditions or light levels, they are particularly The location along the coast where ice is no longer supported by the  ground and where it begins to float.  1 gigaton = 1 billion metric tons. 

CRYOSPHERE 61 BOX 7.1 Ice Shelf Collapse The observation of the collapse of the Larsen B Ice Shelf was astonishing in the sheer dimension and abruptness of change observed via satellite, and it alerted the Earth science community due to its potential implications for sea- level rise (Figure 7.3). The dynamics contributing to the collapse were documented by various satellites: the thinning of the ice shelf toward the coast by satellite altimetry, the accelerated flow by the interferometric synthetic aperture radar (InSAR), the retreat of the grounding line by Landsat, and the calving of the icebergs by MODIS. a 31 Jan 2002 b 17 Feb 2002 c 23 Feb 2002 d 05 Mar 2002 FIGURE 7.3  Collapse of the Larson B Ice Shelf in western Antarctica, January-March 2002. Two thousand square kilometers of the Larsen Ice Shelf disintegrated in just 2 days. SOURCE: National Snow and Ice Data Center, University of Colorado. 7-3 a,b,c,d 

62 EARTH OBSERVATIONS FROM SPACE: THE FIRST 50 YEARS OF SCIENTIFIC ACHIEVEMENTS well suited for monitoring sea ice because of the strong warming associated with a thinner, less extensive ice cover. contrast in microwave emission between open and ice-cov- These observations of shrinking Arctic sea ice are consistent ered ocean. The long-term 35-year data set from the passive with climate model predictions of enhanced high-latitude microwave sensors has enabled us to produce trend analyses warming, which in turn are driven in significant part by beyond the strong interannual variability of sea ice. Recent ice-albedo feedback (Holland and Bitz 2003). In contrast estimates indicate that Arctic sea ice extent decreased by to the Arctic, no clear trend in the extent of Antarctic sea ice approximately 7.4 percent from 1978 through 2003, while coverage has been detected. multiyear ice area has decreased by approximately 7.0-11.0 Over the past few years, there have been a growing percent per decade (Comiso 2002, Johannessen et al. 2004; number of reports forecasting sea ice conditions, and these Figure 7.4). The past several years have been nothing short of reports are based entirely or mostly on data from satellites. extraordinary (NRC 2007c). Since 2000, record summer ice For example, the Arctic Climate Impact Assessment (ACIA minima have been observed during 4 out of the past 6 years 2005) concluded that continued reductions in Arctic sea ice in the Arctic (Stroeve et al. 2005). Moreover, most recent might soon lead to a seasonally ice-free Arctic and increased indications are that winter ice extent is now also starting maritime traffic because shipping routes through the Arctic to retreat at a faster rate, possibly as a result of the oceanic Ocean are much shorter than routes through the Panama or Suez Canals. However, there is some evidence that a reduc- tion in the ice cover will be accompanied by greater interan- nual variability, at least in certain regions (Atkinson et al. 2006); the potential combination of increased maritime traf- fic, high interannual variability in the ice cover, and regional variations will require improved regional sea ice forecasts for maritime operators. Glacier Extent and Position of Equilibrium Line The study of glacier regimes worldwide reveals wide- spread wastage since the late 1970s, with a marked accelera- tion in the late 1980s. Remote sensing is used to document changes in glacier extent (the size of the glacier) and the position of the equilibrium line (the elevation on the glacier where winter accumulation is balanced by summer melt; König et al. 2001). Since 1972, satellites have provided opti- cal imagery of glacier extent. The synthetic aperture radar (SAR) is used to study zones of glacial snow accumulation and ice melt to determine climate forcing, and laser altim- etry is used as well to measure change in glacier elevation. For example, a study in the Ak-shirak Range of the central Tien Shan plateau used aerial photographs in the 1970s, along with the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) imagery from 2001, to document a reduction in glacier area of 20 percent between FIGURE 7.4  Deviations in monthly sea ice extent for the northern 1977 and 2001 (Figure 7.5; Khromova et al. 2003). and southern hemispheres from November 1978 through December Because glaciers respond to past and current climatic 2004, derived from satellite passive-microwave observations. The changes, a complete global glacier inventory is being Arctic sea-ice decreases are statistically significant, with a trendline developed to keep track of the current extent as well as the slope of −38,200 ± 2,000 km2/year, and have contributed to much rates of change of the world’s glaciers. Coordinated by the concern about the warming Arctic climate and the potential effects National Snow and Ice Data Center, the Global Land Ice on the Arctic ecosystem. The Antarctic sea ice increases are also Measurements from Space project is using data from ASTER statistically significant, although at a much lower rate of +13,600 and the Landsat Enhanced Thematic Mapper to inventory ± 2,900 km2/year. The northern hemisphere plot is extended from about 160,000 glaciers worldwide. This effort will likely P ­ arkinson et al. (1999), and the Southern Hemisphere plot is e ­ xtended from Zwally et al. (2002). SOURCE: Courtesy of Claire P ­ arkinson and Donald Cavalieri, NASA Goddard Space Flight  Ice-albedo feedback is a positive feedback loop whereby melting sea ice Center, as updated from Parkinson et al. (1999) and Zwally et al. exposes more seawater (of lower albedo, or less reflective), which in turn (2002). absorbs heat and causes more sea ice to melt. 

CRYOSPHERE 63 FIGURE 7.5  Location and changes of the Ak-shirak glacier system, 1943-2001. (a) ASTER image for September 14, 2001. (b) Location map. In the other insets the green lines indicate glacier outlines in 1943: (c) decrease in glacier size through climate change and direct anthropogenic impact, (d) decrease in size of a surging glacier and appearance of new glaciers, (e) increase in area of outcrops and in the perimeters of water divides between glaciers, and (f) disappearance of former small glaciers. SOURCE: Khromova et al. (2003). Reprinted with permission from the American Geophysical Union, copyright 2003. 7-5 result in major scientific advances in the near future with tant indicators of climate change and exemplify the value important ramifications for climate research. As for the other and importance of long-term data sets for understanding the examples of accomplishments listed in this chapter, these complex climate system. measurements and the resulting trend analyses are impor-