Changing Climate: Report of the Carbon Dioxide Assessment Committee (1983)

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.

Probable Future Changes in Sea Level Resulting from Increased o Atmospheric Carbon Dioxide Roger R. Revelle Many processes can cause an apparent change in sea level at any par- ticular location. They include local or regional uplift or subsidence of the land; changes of atmospheric pressure, winds, or ocean currents; changes in the volume of the ocean basins owing to volcanic activity, marine sediment deposition, isostatic adjustment of the Earth's crust under the sea or changes in the rate of seafloor spreading; changes in the mass of ocean water brought about by melting or accumulation of ice in ice sheets and alpine glaciers; and thermal expansion or contraction of ocean waters when these become warmer or colder. Only the last two processes are of primary interest in considering worldwide changes in sea level resulting from climatic change, such as the warming that may be induced by increasing atmospheric carbon dioxide. (Melting or formation of sea ice and floating ice shelves have no effect on sea level—a glass of ice water filled to the brim does not overflow while the ice melts.) But the other processes contribute to the "noise" that afflicts all sea-level records and may make their interpretation over periods of a few decades difficult or impossible. For orientation, it is useful to keep in mind that sea level has risen l50 m in the l50 centuries since the peak of the last glacial period. Hence, the present rate of l0-20 cm per century is small compared with the average rate of l m per century over the past l5 millennia and very much smaller than the inferred maximum rise of perhaps 5 m per century immediately following the glacial period. Indeed, judging by Figure 8.l, the present is a time of quiet sea level compared with the violent oscillations that occurred during most of the last l00,000 years. 8.l THE OBSERVED RISE IN SEA LEVEL DURING PAST DECADES Barnett (l983d) examined all available records of annual averages of relative sea level. These were obtained from tide gauges around most of the world's continental margins and on a few islands. Barnett found l55 stations that had 30 or more years of record. Averaging over small areas of unusually high station density reduced this number to 82 usable data series. The longest covers the period from l88l to l980, and nearly all series extend from l930 to l980. The data are very unevenly 433

434 LU LU _1 < U1 1A K 1U C AGE(Kyr) < > 20 40 60 80 100 120 140 160 +10 -— | I 1 I 1 1 1 I 1 I 1 I 1 1 I 1 b VII I i "^ - N % % Jv \ - \ « 1 V1 ' • I V f « } • ** - 1 -20 ~ | •« n / 1 1 i H1 ,' N I I - I I ii > . ( l ,• '• / i ; » .-N r\ r\ 1 \ / ! ' -40 - .11 I ' U -60 - » : 1 : , I : '- -80 1 l i • i -100 - I 1 i 1 1 -120 — j i 1 *^1 -140 L FIGURE 8.l High stands of sea level during the past l50,000 years. Note 6-m terrace above present sea level l20,000-l25,000 years ago, and 2-m terrace l35,000 years ago. (Source: Moore, l982.) distributed; there are only six stations in the southern hemisphere and only two (Hawaii and Bermuda) in the central ocean gyres. To obtain some semblance of a global average, Barnett divided the stations, using empirical orthogonal function analysis (Barnett, l978), into six oceanic regions in which there was strong coherence among different stations. For the l00 years from l88l to l980, the average sea level for all regions combined, weighted by the area of each, can be fitted by a linear trend of rising sea level of l4.3 ± l.4 cm per century. During the 5l years from l930 to l980, a linear trend of 22.7 i 2.3 cm per century was observed. The trend for the last l00 years represents a compromise between a time of little change (l88l-l920) and a time of steady increase (l920-l980). Stations from Alaska and Scandinavia, where isostatic rebound is known to cause strong uplift of land, were omitted from the averages. Other recent estimates of the rate of global sea-level rise, using geographically distributed selected stations or different methods of combining groups of stations, have been made by Fairbridge and Krebs (l962), Lisitzin (l974), Gornitz et al. (l982), Barnett (l983a), and Emery (l980). Their estimated rates were, respectively, l2, ll.2, l2, l5.l, and 30 cm per century. These different results reinforce

435 Barnett's conclusion (l983d) that estimates of global sea-level change with data available at present can vary substantially, depending solely on the method of analysis. He believes that the data do not support Emery's contention that the rate of rise has accelerated during recent decades. Barnett (l983a,b,c) investigated the probable causes of the sea-level rise. He concluded from analysis of changes of dynamic heights in the upper water layers measured by repeated hydrographic casts, together with measurements, corrected for instrumental bias, showing an average rise in ocean surface temperature of about 0.4°C in the northern hemi- sphere since the beginning of the twentieth century, that warming and expansion of the upper water layers can account for less than 5 cm of the estimated rise in sea level. Gornitz et al. (l982) agree that only a small part of this observed rise is due to thermal expansion of the upper ocean waters. Barnett tentatively inferred that perhaps three fourths of the rise represents water added to the ocean from ablation or melting of the Greenland and Antarctic ice caps and alpine glaciers. Observed changes in the length of the day since l900 and a displace- ment by 730 cm toward 70° W in the position of Earth's mean pole of rotation are consistent with this explanation, which would mean that between 30,000 and 40,000 km3 of ice (less than 0.2% of the land-ice volume) has melted, mainly from Greenland and Antarctica. Similar conclusions with respect to the length of the day were reached by Etkins and Epstein (l982). Estimates of mass balance of the Antarctic Ice Sheet (Bentley, l983) suggest that the mass is stable and perhaps even increasing, but the noise level of the estimates is so high that a small net loss corresponding to a rise in sea level of 0.5 mm per year is not forbidden. Perhaps two thirds of the ice loss, or about 390 km3 per year (l mm/year of sea-level rise) may have come from Green- land. This idea is supported by the observations of Brewer et al. (l983) and Swift (l983) that the salinity of the deep water north of 50° in the North Atlantic has diminished over the past 20 years or so, as would be expected if freshwater from a melting ice cap were being added to the northern ocean. 8.2 THE FUTURE RISE IN SEA LEVEL The projected climatic warming from increasing atmospheric carbon dioxide will lead to an increased transfer of water mass to the sea from continental and alpine glaciers. As shown below, the resulting rise in sea level could be about 40 cm over the next century. Increased downward infrared radiation will also lead to a warming and, therefore, expansion of the upper ocean waters, which can contribute another 30 cm for a total of 70 cm. Assuming the correctness of the figure of 4 W m 2 for the increased downward infrared flux with a doubling of carbon dioxide (Ramanathan et al., l979),* both the estimates for ice *Higher concentrations of other infrared gases will further increase the downward infrared flux (see Machta, this volume, Section 4.3).

436 melting and ocean thermal expansion still have large error bars—at least +25%. These are due to our uncertainty over the causes of the present rise in sea level, our inability to predict whether changes in atmospheric circulation will cause more or less snow to fall on the ice caps, our ignorance of the conditions for advance or retreat of alpine glaciers, and our lack of understanding of the physical processes asso- ciated with the flux of heat to the ocean. Of even greater uncertainty is the potential disintegration of the West Antarctic Ice Sheet, most of which now rests on bedrock below sea level. This could cause a further sea-level rise of 5 to 6 m in the next several hundred years. We shall discuss each of these sources in turn. 8.2.l Melting of Greenland Ice Cap and Alpine Glaciers Ambach (l980) has investigated the effects of changes in atmospheric temperature on the Greenland ice cap. He shows that the altitude at which ice ablation occurs increases by about l00 m for each 0.6°C rise in air temperature, and he calculates that the area of ice ablation will increase by 340,000 km2 for a 3°C rise in temperature. There would be an equal diminution in the area of ice accumulation, with the result that the ice volume would decrease by the equivalent of 480 km3 each year, equal to an annual rise in sea level of l.2 mm (l2 cm per century). For a temperature rise of 6°C or more, which the general circulation models of Manabe and Wetherald (l975) indicate at the latitude of Greenland for a C02 doubling, the rate of rise would be at least 24 cm per century. The average rate over the next 80-l00 years would be half of this, or l2 cm per century. We are unable to calculate the rise in sea level due to a possible retreat of alpine glaciers. According to Barnett (l983a), glaciers contain about 240,000 km3 of ice, and a 20% reduction in their volume during the next l00 years would bring about a rise of l2 cm in sea level. Adding these figures for accelerated ice loss from Greenland and alpine glaciers to the estimated rate of loss from ice melting over the past 50 years of l7 cm per century, we arrive at a probable rise in sea level resulting from an increase in the mass of ocean water during the next century of 4l cm. The volume of ice lost from ice caps and alpine glaciers over the century would be approximately l60,000 km3, between 5 and l0% of the Greenland ice and around 0.5% of the total ice above sea level. The average latent heat of melting would be l.2 x l020 cal yr"l or l.5 x lOl^ W. This is about l% of the calculated increased downward flux to the ocean surface of infrared radiation of 4 W m-2 when atmospheric carbon dioxide has doubled and other greenhouse gases have increased by the expected amounts. Clearly, a major fraction of the downward infrared flux should be available to heat the upper ocean waters. _J

437 8.2.2 Heating of the Upper Oceans This effect is strongly focused in the upper oceans for two reasons: it is where the heating is greatest and where the coefficient of thermal expansion is largest. For simplicity, the ocean can be divided into two layers: a mixed surface layer with an average thickness of 70 m in which the temperature change is uniform throughout and equal to the change in atmospheric temperature at the surface and the remainder of the ocean in which heat is transported by advection and turbulent diffusion. Unfortunately, there is no good understanding of what happens in the deeper layer; major ongoing research efforts such as the projected World Ocean Circulation Experiment of the World Climate Research Program are directed at remedying this unsatisfactory situation. The evidence points toward the importance of heat transport along isopycnal (quasi- horizontal constant-density) surfaces from surface regions in higher latitudes toward the midlatitude interior. In comparison, crosspycnal (quasi-vertical) mixing processes, such as double diffusion (i.e., diffusion of heat and salinity at different molecular diffusion rates) and intermittent internal wave breaking, appear to be of lesser impor- tance. This means that the heat transport needs to be considered as at least a two-dimensional (latitude-depth) problem and probably as a three-dimensional one. Nevertheless, one-dimensional mixing models persist, if for no other reason than that they can be solved. Here we sketch a one-dimensional treatment in the hope that, with the tradi- tional empirically based choice of mixing parameters, we may arrive at some feeling as to the magnitude of the effect. Let 6 (z,t) represent the temperature change at different times and depths below the mixed layer relative to an assumed steady state in the year l860. Then, in a simple two-layer system, — = ^1 + — dt dz2 dz where k is the coefficient of vertical eddy diffusion and w is the vertical component of advective velocity taken positive downward. Both k and w are assumed to be constant with depth down to l000 m and zero below l000 m. Cess and Goldenberg (l98l) found that the time rate of change in temperature resulting from radiative heating of the atmosphere caused by observed and projected increases in atmospheric CO2 is proportional to the temperature change, divided by an appropriate time constant, TC de 6 dT = —' (2) c and hence the solution of Equation (l) is 6 (z,t) = 6 (0,t)exp "(Z " m* > (3)

438 where m = 70 m and I „ w_ + l w! H 2k 2 k2 Using the past and projected future changes in atmospheric CO2 since l860, estimated by Machta and Telegadas (l974) and Keeling (l977), and the radiative heating at the surface computed by Ramanathan et al. (l979) for atmospheric CO2 concentrations of 320, 426, 534, and 640 ppm, Cess and Goldenberg computed a value of 33 years for TC. If, as seems likely, the future proportional rate of increase of atmospheric CO2 is lower than in the past, TC will be longer than 33 years, and the temperature changes at different depths below the mixed layer will be a larger fraction of the temperature changes at the surface. But the uncertainties in the calculation are so large that it is not very useful to attempt a more precise estimate of TC. By using a value of 33 years, we are probably computing a minimal rise in sea level from ocean warming resulting from a doubling of atmospheric carbon dioxide. For the vertical component of advective velocity and the vertical- eddy diffusion coefficient in latitudes between 30° N and 30° S, we use Munk's (l966) suggested values, w = l.4 x l0-5 cm sec-l r k < 30° = l.3 cm2 sec"l • The density stratification of the subsurface waters at higher latitudes is weaker than in the tropics, and consequently we assume that k north and south of the parallels of 30° N and 30° S, respectively, has twice its value in low latitudes, that is, k > 30° » 2.6 cm2 sec"l • Calculations of the increases in water temperature at different depths down to l000 m were carried out for l0° latitude bands from 60° N to 80° S. Initial average surface ocean temperatures within each latitude band were taken from Sverdrup et al. (l942, p. l27). These were added to the change in average surface temperature for each latitude band projected by Flohn (l982) for the middle of the twenty-first century, assuming a doubling of atmospheric carbon dioxide and the probable increase in other greenhouse gases during the next 70-80 years.* Cess and Goldenberg show that a time delay of at least 20 years can be expected before the upper ocean waters come near to equilibrium with these projected values. Thus, our calculation gives the rise in sea level about l00 years from now. *Flohn's projected global temperature rise is 4.2°C, somewhat higher than the value of 3°C used in the earlier discussion.

439 80°S .§ X Q- LU Q FIGURE 8.2 Present temperatures in the top l000 m of the World Ocean, estimated from Atlantic and Pacific N-S sections (Figures 2l0 and 2l2, Sverdrup et al., l942). The average increases in temperature at depths between 0-l00 m, l00-200 m, 200-500 m, and 500-l000 m were calculated from Equation (3) and added to the present values at these depth intervals. The latter were estimated from Figures 2l0 and 2l2 of Sverdrup et al. (l942, pp. 748 and 753). These figures give the distribution of temperature with depth for north-south sections in the Western Atlantic and Pacific Oceans, respectively. (See Figures 8.2 and 8.3.) The increase in specific volume of seawater resulting from a given increase in temperature will vary markedly with temperature and depth. This quantity was calculated using the coefficients of thermal expansion at various temperatures for a salinity of 35°/oo and depths of zero and 2000 m shown in Table 9 of Sverdrup et al. (l942, p. 60). The total change in specific volume (and thus in sea level) between zero and l000 m for each l0° latitude band was multiplied by the percentage of ocean area in that band. We thus arrive at a rise in sea level about l00 years from now of at least 30 cm, resulting from ocean warming. A

440 LATITUDE 60 N 40 N 20 N 20° S 40° S 60° S 80° S 200 -g 400 jT a. UJ Q 600- 800 1000 1 7 J s J 3 5 5.8 4.2 4.2 2.7 1.6 2.0 2.7 3.9 3.5 3.4 * .4 2.9 2.5 1.7 r~ I 4.0 I 2.8 2.3 1.6 0.9 1.1 1.6 2.2 r"i 2.3 -2.0-1 1.7 J "I 1 1.1 1.8 1.3 1.3 0.5 0.3 0.4 0.5 0.7 0.6 1.0 1 JO 0.9 C ).8 0.5 FIGURE 8.3 Computed near-equilibrium changes in ocean temperature for a doubling of atmospheric carbon dioxide and probable increase in other greenhouse gases, about 2080. The surface temperature increase is based on Flohn's (l982) prognosis. similar calculation has been made by Gornitz et al. (l982); they estimate a rise of 20-30 cm during the next 70 years. Adding this estimate for ocean warming to our estimates for melting in Greenland and Antarctica and in alpine glaciers, we arrive at a probable rise in sea level during the next l00 years of about 70 cm. But a much larger rate of rise is not unlikely during the following several centuries because of events in Antarctica.

44l 8.2.3 Possible Disintegration of the West Antarctic Ice Sheet West of the Transantarctic Mountains (approximately from the Meridian of Greenwich across the Antarctic Peninsula to l80° W), most of the Antarctic Ice Sheet rests on bedrock below sea level, some of it more than l000 m beneath the sea surface. In its present configuration, this "marine ice sheet" is believed to be inherently unstable; it may be subject to rapid shrinkage and disintegration under the impact of a CO2-induced climatic change (Mercer, l978). (See Figure 8.4.) A collapse of the West Antarctic Ice Sheet would release about 2 million knr of ice before the remaining half of the ice sheet began 90- W -90-E \ AMUNDSEN SEA \ EDGE Of FLOATiNG iCE GROUNDiNG LiNE^X • MOUNTAINS ^™ -^ •'/Mti BEDROCK ABOVE SEA iKX//. LEVEL SUPPORTING ICE • BEDROCK MORE THAN ONE KILOMETER ~ANTAnrVTr~riHcCE BELOW SEA LEVEL SUPPORTiNG ICE ISO- FIGURE 8.4 The West Antarctic Ice Sheet (WAIS) lies north and west of the Transantarctic Mountains (shown in black). It is believed to be unstable because most of it lies on rock below sea level. Disappear- ance of the ice above sea level would raise the world ocean by 5 to 6 m. At present, the ice sheet is held back by the Ross and Filchner- Ronne ice shelves, which, though mostly floating, are "pinned" by high places on the seafloor. From "Carbon Dioxide and World Climate" by Roger R. Revelle. Copyright* l982 by Scientific American, Inc.

442 to float (Bentley, l983). The resulting worldwide rise in sea level would be between 5 and 6 m. The oceans would flood all existing port facilities and other low-lying coastal structures, extensive sections of the heavily farmed and densely populated river deltas of the world, major portions of the state of Florida and Louisiana, and large areas of many of the world's major cities. Table 8.l shows the percentage of the area of each coastal state of the United States that would be flooded by a 4.5-m rise and by a 7-m rise (Schneider and Chen, l980). Well-preserved fossil corals in late Pleistocene wave-cut or depositional terraces at 2l localities throughout the world have been dated by several groups of workers using measurements of the ratios of uranium and thorium isotopes. The results have been evaluated by Moore (l982) of the University of South Carolina. He shows that at almost all localities, a high sea level—perhaps 7 m above the present— occurred about l25,000 years ago. A somewhat lower high stand of the sea—about 2 m above present sea level—appears to have occurred around l35,000 years ago. This lower stand was accompanied by a major melt- water event in the Gulf of Mexico, as shown by oxygen/isotope ratios in deep-sea sediments. The influx of freshwater and the 2-m rise in sea level may have been caused by partial melting of the northern hemisphere ice sheet, while the higher stand l25,000 years ago could have resulted from the temporary disappearance of the West Antarctic ice cap. If this happened quickly, the surface layer of low-salinity water might have persisted long enough to leave a measurable trace in the oxygen isotope ratios in the oldest coral fossils on the terraces. A rapid rise might also be indicated if coral and algal species that grow best a few meters below the surface were found near the bottom of the terraces underneath remains of shallower water organisms (Figure 8.l). In contrast to the conditions of l25,000 years ago, at the end of the last glacial period l7,000 to l2,000 years ago when sea level was as much as l50 m lower than at present, the West Antarctic Ice Sheet extended over the continental shelves of the Ross, Amundsen, Bellings- hausen, and Weddell Seas, and the interior ice elevation was 500 to 2000 m higher than today (Hughes, l983). Thus, perhaps two thirds of the ice has disappeared during the last l7,000 years. Evidence from radar soundings of flow lines extending across the Ross ice shelf indicates that the remaining West Antarctic Ice Sheet has been relatively stable for the last l000-2000 years. Indeed, various lines of evidence suggest that the mass balance of the entire Antarctic Ice Sheet may be positive—i.e., ice may be accumulating. The uncertainty of the measurements is such, however, that a diminution of as much as l80 km3 per year is not unlikely, corresponding to a rise in sea level of 0.5 mm per year, as we have suggested in discussing the present rate of sea-level rise. The rate at which the remaining West Antarctic Ice Sheet could disappear under the impact of a CO2-induced warming has recently been examined by Charles Bentley of the University of Wisconsin-Madison. He concludes that the mechanism for a relatively rapid disappearance would involve two processes occurring in sequence: accelerated melting on the underside of the Ross and Filchner-Ronne ice shelves by compara- tively warm seawater (temperature between 2 and 3.5°C) circulating

443 m 10 X O <w 41 SJ ° ** a*3 < CO X O 41 kl IW 4J dP 03 Is t° o e o. — x o v4 IH X O 41 u iu .U C -~ 0 in -H C 41 O 10 -H rH rH 3 rH Q,.r( b ~ eg in i co m m 1N i m vo 10 vo en Q in VO 1 rH in 1 rH rH CM fM r- ol CM o ro ^ rH VO OV « CM rH in in a f IN i vorn^TrHOoi n i 01^vorovoeorHH ivororM in 1 rH CN rH 1N CN | rH rH I rHrHOOCMOOOrHOOOinOOCOOOOOlOlrHrHOOin in m CO 01 M inrHvoo01oooorooororHCM0ooio*«^r'<roir~'«rH rH -» 1N M1 iN U1 in 00 M i• r- w • • CM O CD CM rH O CO CN o o CM rH rH 000 vo r* CO O rH ro ao r- 1o r* 01-aIt)I OT)lT)I o o-ai-al H m O n o * iM 41 rH EH C9 a 10 a C -H 10 CO -H 0] 01 -H '3 ffl u -H 4J gg OO u k l 1010 'OlM4l COk i id e jj >. coi 10U CO 41 -O 4J t, 4J C 4J .r. 41 >, J<-r< u*' o o ra rHio l'>3 MJ3 o Qi 3-5 ia< rH1 <'a -M X inrHj3»c io»-H a c D' s g^ vT O ** C C -it C n 41 o n 41 a 4J rH C rH O < u -O JS ? 41 0 4 o O 00 4J en ig c•° 41 ki M U! u 10 41 k l -O 'O C0 -H 1U 4J 5 i § • H 10 - U a ^S ^T 41 rH 4J • 10 O ffc Si5 tj^ ." a n 41 41 "^ S- t

444 under the shelves, which could bring about a thinning of the shelves of one to several meters per year, followed by relatively rapid movement of ice streams and calving of icebergs along the ice front. Thinning of ice shelves would result in their becoming "unpinned" from most high points of the seafloor and in a rapid retreat of the "grounding lines" (the boundaries between the floating ice shelves and the grounded inland ice), followed by still more rapid acceleration of the ice shelf thinning and grounding line retreat. This process of ice-shelf removal would not be necessary in the Pine Island Bay-Thwaites Glacier area of the Amundsen Sea because no ice shelves exist there. After the removal of the ice shelves, "ice streams" in the inland ice could discharge directly into the ocean (Figure 8.5). According to Bentley, the width of the ice streams, including Thwaites and Pine Island Glaciers, would be about half the width of the new coastline, or 500-600 km. Removal of 2 x l06 km3 of ice in 200 years would require a speed of discharge of 20 km/year, about 20 times faster than the present speed of movement. This is nearly 3 times faster than the fastest known ice stream/outlet glacier speed of 7 km/year in the Jakobshaven Glacier of Greenland. Bentley concludes that the ice streams cannot move at 20 km/year in their unbroken form but that this could occur by another process: calving and removal of icebergs, with the rate of the ice loss being equal to the width of the ice streams times their average thickness of l km, times the sum of the glacier speeds and the speed of retreat of the grounding lines. The limiting factor here would be the rate of removal of the icebergs by winds and ocean currents. Bentley calculates that the component of ocean current velocity at right angles to the glacial shoreline would have to be 2 cm sec"l and that the current velocities in a series of gyres in the Ross, Amundsen, and Weddell Seas would have to be at least l0 cm sec"l. He concludes that such rates of discharge and removal of icebergs "might barely be possible, although unlikely, and that the ice sheet could dis- appear in 200 years, but only after removal of the ice shelves. ..." Bentley considers the suggestion by some workers that all the ice could be discharged into the Amundsen Sea through Thwaites and Pine Island Glaciers in 200 years would require "unreasonably high" glacier speeds of at least 50 km yr"l and ocean current speeds to carry away calved icebergs of 50 cm sec"l. However, he suggests that one quarter of the West Antarctic ice could be discharged through Pine Island Bay in 200 years and half in 400 years. During this period the Ross and Filchner-Ronne ice shelves could have disappeared, and all the ice could be discharged within 500 years. If the time required for the ice shelves to disappear is l00 years, Bentley's analysis would not be incompatible with a minimum time of 300 years for disintegration of the West Antarctic Ice Sheet. The corre- sponding average rate of rise of sea level would be slightly less than 2 m/l00 years, beginning about the middle of the next century. Bentley's "preferred" minimum time of about 500 years would give a rate of sea- level rise of l.l m/l00 years, which, as we have pointed out, is about the mean rate for the last l5,000 years. To either of these figures we must add a rise of 70 +_ l8 cm between l980 and 2080, which we have shown is likely to result from ocean warming and ice ablation in

445

446 Greenland and Antarctica, plus a possible retreat of alpine glaciers. These processes may well continue in later centuries. Disintegration of the West Antarctic Ice Sheet would have such far-reaching consequences that both the possibility of its occurrence and the rate at which disintegration might proceed should be carefully researched. We have already suggested studies of the coral structures in the 5-6-m terraces that occur throughout tropical seas as a means of estimating the minimum time for the rise in sea level that created the terraces. Other studies and monitoring programs should be undertaken (American Association for the Advancement of Science, l980). Among these, five problems deserve special emphasis: possible change in the mass balance of the Antarctic Ice Sheet; interaction between the Ross and Filchner-Ronne ice shelves and adjacent ocean waters; ice stream velocities and mass transport into the Amundsen Sea from Pine Island and Thwaites Glaciers; modeling of the ice sheet response to C02-induced climate change; and deep coring of the ice sheet to learn whether it in fact disappeared l25,000 years ago. Ground-based and satellite monitoring of possible changes in the topography of the ice sheet will help to reveal changes in the mass balance of the ice and their contribution to observed changes in sea level. It may be necessary to develop satellite instrumentation for this purpose. Bottom melting of the ice shelves may be brought about by the penetration of relatively warm seawater under the shelves. It might be supposed that the effect would be inhibited by the development of a relatively stable layer of cold, low-salinity water just under the ice. However, double diffusion may play a determining role. The dynamics of ocean boundary currents near the shelves also need to be studied in order to estimate northward iceberg transport. The existence and rate of bottom melting can be monitored by radar profiling of the shelves. The Pine Island and Thwaites Glaciers are the principal region of West Antarctica in which the ice sheet is not held back by an ice shelf. What are the actual and potential ice stream velocities and the rate of mass transport in this region? Physically based diagnostic and prognostic models of the probable response of the ice sheet to a carbon dioxide-induced climate change should be improved, both as a guide for monitoring programs and to enable better estimates of the rates of possible disappearance of the West Antarctic Ice Sheet. In both Greenland and Antarctica, ice cores obtained by drilling have given valuable information on atmospheric carbon dioxide content and temperature over the past 20,000-30,000 years. Equipment now exists for deep drilling into the ice sheet, which could extend the paleocli- matic record back to l00,000-200,000 years. Such deep drilling in West Antarctica might give positive evidence as to whether the West Antarctic Ice Sheet disappeared during the last interglacial period. 8.4 ACKNOWLEDGMENTS As so often in the past, I am indebted to Walter Munk for many improvements in the manuscript. I thank also Tim Barnett for helping

447 me understand the uncertainties in existing data related to sea level and Jesse Ausubel for careful editing. The discussion of the West Antarctic Ice Sheet is partly based on an informal conference at the Scripps Institution of Oceanography attended by C. Bentley, A. L. Bloom, W. S. Broecker, E. S. Epstein, R. Etkins, V. Gornitz, T. Gross, D. Katcher, W. Nierenberg, J. S. Perry, R. Wetherald, and the author. REFERENCES Ambach, W. (translated by G. P. Weidhaas) (l980). Increased C02 concentration in the atmosphere and climate change: potential effects on the Greenland ice sheet. Wetter und Leben 32:l35-l42, Vienna. (Available as Lawrence Livermore National Laboratory Report UCRL-TRANS-ll767, April l982.) American Association for the Advancement of Science (l980). Response of the WAIS to CO? Warming. Report of a meeting, Orono, Maine, April l980 (mimeo.). (Made available by C. Bentley, University of Wisconsin-Madison.) Barnett, T. P. (l978). Estimating variability of surface air temperature in the northern hemisphere. Mon. Wea. Rev. l06: l353-l367. Barnett, T. P. (l983a). Recent changes in sea level and their possible causes. Climatic Change 5:l5-38. Barnett, T. P. (l983b). Long-term trends in surface temperature over the oceans. Mon. Wea. Rev., in press. Barnett, T. P. (l983c). Long-term changes in dynamic height. J^. Geophys. Res., in press. Barnett, T. P. (l983d). Some problems associated with the estimation of "global" sea level change. Climate Research Group, Scripps Institution of Oceanography, U. of Calif., San Diego, 40 pp. (mimeo). Bentley, C. R. (l983). The West Antarctic Ice Sheet: diagnosis and prognosis. Proceedings, Carbon Dioxide Research Conference, Carbon Dioxide, Science and Consensus, Berkeley Springs, W. Va., September l982. CONF-820970, NTIS, Springfield, Va. Brewer, P. G., W. S. Broecker, W. J. Jenkins, P. B. Rhines, C. G. Rooth, J. H. Swift, T. Takahashi, and R. T. Williams (l983). A climatic freshening of the deep North Atlantic (north of 50° N) over the past 20 years. Science, to be submitted. Cess, R. D., and S. D. Goldenberg (l98l). The effect of ocean heat capacity upon global warming due to increasing atmospheric carbon dioxide. J. Geophys. Res. 86:498-502. Emery, K. 0. (l980). Relative sea levels from tide gauge records. Proc. Nat. Acad. Sci. 77:6968-6972. Etkins, R., and E. S. Epstein (l982). The rise of global mean sea level as an indication of climate change. Science 2l5:287-289. Fairbridge, R., and 0. Krebs, Jr. (l962). Sea level and the southern oscillation. Geophys. J. Roy. Astron. Soc. 6:532-545. Flohn, H. (l982). Climate change and an ice-free Arctic Ocean. In Carbon Dioxide Review: l982, W. C. Clark, ed. Oxford U. Press, New York, pp. l45-l99.

448 Gornitz, V., S. Lebedeff, and J. Hansen (l982). Global sea level trend in the past century. Science 2l5:l6ll-l6l4. Hughes, T. (l982). The stability of the West Antarctic Ice Sheet: what has happened and what will happen. Proceedings, Carbon Dioxide Research Conference, Carbon Dioxide, Science and Consensus, Berkeley Springs, W. Va., September l982. CONF-820970, NTIS, Springfield, Va. Keeling, C. D. (l972). Impact of industrial gases on climate. In Energy and Climate. Geophysics Study Committee, National Research Council, National Academy of Sciences, Washington, D.C., pp. 72-95. Lisitzin, E. (l974). Sea-Level Changes, Elsevier Oceanography Series, 8, Elsevier, New York. Machta, L., and K. Telegadas (l974). Inadvertent large-scale weather modification. In Weather and Climate Modification, W. Hess, ed. Wiley, New York. Manabe, S., and R. T. Wetherald (l975). The effects of doubling the C02 concentration on the climate of a general circulation model. J. Atmos. Sci. 32:3-l5. Mercer, J. H. (l978). West Antarctic Ice Sheet and CO2 greenhouse effect: a threat of disaster. Nature 27l:32l-325. Moore, W. S. (l982). Late pleistocene sea-level history. In Uranium Series Disequilibrium: Applications to Environmental Problems, M. Ivanovich and R. S. Harmon, eds. Clarendon, Oxford, p. 4l. Munk, W. H. (l966). Abyssal recipes. Deep Sea Res. l1:707-730. Ramanathan, V. M., M. S. Lian, and R. D. Cess (l979). Increased atmospheric CO2: zonal and seasonal estimates of the effect on the radiation energy balance and surface temperature. J. Geophys. Res. 84:4949-4958. Revelle, R. (l982). Carbon dioxide and world climate. Sci. Am. 247.: 36-43. Schneider, S. H., and R. S. Chen (l980). Carbon dioxide warming and coastline flooding: physical factors and climatic impact. Ann. Rev. Energy 5:l07-l40. Sverdrup, H. U., M. W. Johnson, and R. H. Fleming (l942). The Oceans. Prentice-Hall, New York. Swift, J. H. (l983). A recent 0-S shift in the deep water of the northern Atlantic. MLRG, Scripps Institution of Oceanography, La Jolla, Calif. Submitted for Proc. Vol. 4th Biennial Ewing Symposium.